Elucidation of the pathways responsible for the biosynthesis of
UDP-N,N'-diacetylbacillosamine in bacterial pathogens
by
Michael James Morrison
B.A. Chemistry
Wesleyan University, 1999
M.A. Chemistry
Wesleyan University, 2000
Submitted to the Department of Chemistry
in Partial Fulfillment of the Requirements for the Degree of
ARGNVEz
MASSACHUSETTS iNSiWtEi
" TECHNOLOGY
Doctor of Philosophy
BAR1 20
E4
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LIBRARIES
February 2014
0 2013 Massachusetts Institute of Technology
All rights reserved
Signature of Author
Department of Chemistry
October 16, 2013
Certified by
Barbara Imperiali
Class of 1922 Professor of Biology and Professor of Chemistry
Thesis Supervisor
Accepted by
Robert W. Field
Haslam and Dewey Professor of Chemistry
Committee on Graduate Students
Departmental
Chairman,
This doctoral thesis has been examined by a committee of the Department of Chemistry as
follows:
Professor Catherine L. Drennan
Committee Chair
Professor of Chemistry and Biology
Howard Hughes Medical Institute Investigator and Professor
Professor Barbara Imperiali
Thesis Supervisor
Class of 1922 Professor of Biology and Professor of Chemistry
Professor Robert T. Sauer Salvador E. Luria Professor of Biology
2
Elucidation of the pathways responsible for the biosynthesis of UDP-N,N'diacetylbacillosamine in bacterial pathogens
by
Michael James Morrison
Submitted to the Department of Chemistry
on October 30, 2013 in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ABSTRACT
The highly-modified, bacterial sugar N,N'-diacetylbacillosamine (diNAcBac) has been
implicated in the pathogenicity of certain microbes through its incorporation onto various protein
virulence factors. In particular, diNAcBac is found at the reducing end of glycans in both
asparagine (N-linked) and serine/threonine (0-linked) protein glycosylation pathways. The
second and third chapters examine the O-linked protein glycosylation pathway responsible for
the biosynthesis of the UDP-diNAcBac nucleotide sugar in Neisseria gonorrhoeae and
Acinetobacter baumannii. UDP-diNAcBac is biosynthesized from UDP-N-acetylglucosamine
through the action of a dehydratase, aminotransferase, and acetyltransferase. Specifically, these
enzymes are purified, biochemically characterized, and compared to the N-linked pathway
proteins from Campylobacterjejuni. Furthermore, the substrate specificity of the A. baumannii
phosphoglycosyltransferase that catalyzes the transfer of UDP-diNAcBac onto undecaprenylphosphate is determined.
The fourth chapter explores the structural characterization of the acetyltransferases from
the O-linked protein glycosylation pathways in N. gonorrhoeae (PglB-ATD) and A. baumannii
(Weel). These enzymes are members of the left-handed P-helix family and are responsible for
the acetylation of UDP-2-acetamido-4-amino-2,4,6-trideoxy-a-D-glucose (UDP-4-amino) to
produce UDP-diNAcBac. Based upon these structures, a series of active site mutations are
generated and kinetically characterized for both the AcCoA and UDP-4-amino substrates. These
results suggest that although each enzyme catalyzes the acetyltransferase reaction with identical
substrates, key residues within the binding pockets can lead to a diverse set of catalytic
efficiencies.
The final three chapters investigate the inhibition of UDP-diNAcBac pathway enzymes in
C. jejuni, N. gonorrhoeae, and A. baumannii. The fifth chapter explores a fragment-based
approach to identify small molecules that inhibit the aminotransferase in C. jejuni. To this end, a
crystal structure of this protein is solved in complex with a fragment molecule and analogs of
this compound synthesized. The sixth chapter identifies small molecule acetyltransferase
inhibitors through a high-throughput screening effort in collaboration with the Broad Institute.
Lastly, the seventh chapter describes a fragment-based approach to establish small molecule
inhibitors for the acetyltransferase from N gonorrhoeae.
Thesis Supervisor: Barbara Imperiali
Title: Class of 1922 Professor of Biology and Professor of Chemistry
3
Acknowledgments
First and foremost, I would like to thank my advisor Barbara Imperiali for five amazing
years in the world of bacterial protein glycosylation. I owe you a debt of gratitude for allowing
me to be a part of such a great lab with exciting research opportunities. The amount of scientific
rigor and training received from you is more than I could have ever envisioned. Thank you for
making me the scientist I am today. I would also like to thank Professor Cathy Drennan for all
the guidance you have provided me throughout my life as a graduate student. It was such a
pleasure to be a 5.111 teaching assistant for you during my first year at MIT. Lastly, I am
extremely grateful to Professor Bob Sauer for taking the time to teach me the finer points of
crystal structure refinement. Without your help, none of the structures in this thesis would have
been possible.
I would like to thank all of the wonderful members of the Imperiali lab I have had the
pleasure of working with; you have made graduate school a truly enjoyable experience. I
consider myself extremely fortunate to work alongside such great minds. I wish to thank
Angelyn, Meredith, James, and Jay for sharing all of your knowledge on the field of
glycosylation with me. You are great scientists and I have always strived to maintain the level of
excellence each of you has achieved. I wish to thank Austin for our many scientific discussions
and being such a great collaborator. It was a pleasure working with you and I am sure you will
achieve great things in the future. Andrew, thank you for sharing a lab bench and great ideas
over the past few years, it was a pleasure to work beside you. I also wish to acknowledge the
rest of the glyco team (Michelle, Vinita, Garrett, and Marcie) for all of the support and advice
throughout the years. I wish you all the best on making new, exciting discoveries in the lab!
Lastly, I would like to thank Elizabeth Fong for being a great reference on all things MIT.
To Professor Rex Pratt at Wesleyan University, thank you for encouraging me to
continue on the scientific path. You have been a source of inspiration throughout these many
years. I also wish to thank Dr. Dan Treiber for his endless scientific vigor and support in my
returning to graduate school. You have been a great scientific mentor and friend.
I was extremely lucky to have such great crystallography resources while at MIT.
Special thanks to Dr. Robert Grant for all the time you spent with me explaining the exciting
world of crystallography. You were a tremendous help in all the structures presented here. I am
grateful to Jeremy Setser for all the support with crystallography; thank you for passing on your
expertise to me. To Dr. Weslee Glenn, thank you for all of your support and being such a great
friend. To the rest of my classmates, I wish you all continued success in whatever you may
pursue. I'll miss our yearly Thanksgiving get-togethers!
Finally I would like thank my wife Alyssa for her endless support during graduate school.
She is my foundation and the one constant in my life; without her, none of this would be
possible. I would also like to thank my son Colin, whose smile and infectious laughter can light
up a room and make even the toughest days seem easy. To my sister Melissa and your family
(Dan, Brandon, and Lucas), thank you for always being there for me; I am proud to call you my
friend and sister. Lastly, I would like to thank my grandfather Stanley Miody for being such a
source of inspiration and positive influence on my life. I still miss our golfing days together; I
wish we could play one more round together.
I would like to dedicate this thesis to all of the family members I've lost while in
graduate school: Claire Miody, Lois Morrison, Roy and Lu Spalthoff, and Frank and Bette
Henderson. I'll forever keep you alive in my memory.
4
Table of Contents
Abstract
.............................................................
Acknow ledgm ents..........................................................................................................................
3
4
Table of Contents...........................................................................................................................5
List of Figures ................................................................................................................................
9
List of Tables ...............................................................................................................................
12
List of Schem es............................................................................................................................
13
List of A bbreviations ...................................................................................................................
14
Chapter 1. The Renaissance of Bacillosamine and Its Derivatives: Pathway
Characterization and Implications in Pathogenicity...........................................................
16
Introduction..................................................................................................................................
17
N,N'-Diacetylbacillosamine ........................
Discovery and Characterization................................................................................................
Biosynthesis in Bacterial Pathogens ........................................................................................
Connection to Pathogenicity.....................................................................................................
20
20
21
25
Derivatives
y
.............................................................................
Legionam inic Acid.......................................................................................................................
Pseudam inic Acid ........................................................................................................................
28
28
32
Beyond
36
y
........................................................................................
Conclusions..................................................................................................................................
39
Acknow ledgm ents........................................................................................................................
42
References....................................................................................................................................
42
Chapter 2. Biochemical Characterization of the O-linked Glycosylation Pathway in
Neisseria gonorrhoeae Responsible for Biosynthesis of Protein Glycans Containing N,N'Diacetylbacillosam ine ................................................................................................................
50
Introduction..................................................................................................................................
51
Results and D iscussion ................................................................................................................
Determ ination of UDP-DATDH Stereochem istry by NMR ....................................................
Functional Characterization of PglB-ATD ..............................................................................
55
55
59
5
Kinetic Characterization of PglC and PglB-ATD...................................................................
Functional Characterization of the Glycosyltransferases........................................................
UDP-Saccharide Specificity of Glycosyltransferases..............................................................
60
62
64
Undecaprenyl Diphosphate Disaccharide Specificity of PglE.................................................
66
Characterization of PglH, an Alternative Glycosyltransferase ................................................
Functional Characterization of Oligosaccharyltransferase, PglO ............................................
67
68
Glycan Donor Specificity of PglO ............................................................................................
70
Conclusions..................................................................................................................................73
Acknowledgm ents........................................................................................................................74
Experim ental Procedures .............................................................................................................
74
References....................................................................................................................................86
Chapter 3. Biosynthesis of UDP-N,N'-Diacetylbacillosamine in Acinetobacter baumannii:
Biochemical Characterization and Correlation to Existing Pathways .............................
90
Introduction..................................................................................................................................
91
Results and Discussion ................................................................................................................
94
Expression and Purification of W eeK, WeeJ, Wee!, and W eeH ............................................
Functional and Kinetic Characterization of the Dehydratase WeeK ........................................
Functional and Kinetic Characterization of the Aminotransferase W eeJ................................
Functional and Kinetic Characterization of the Acetyltransferase W eeI...................................
The A. baumannii enzymes W eeK, J, and I produce UDP-diNAcBac ......................................
Substrate Specificity of the Phosphoglycosyltransferase WeeH ...............................................
Active Site Comparison Between 0- and N-linked UDP-diNAcBac Pathway Proteins...........
UDP-diNAcBac enzyme diversity in N- and O-linked glycosylation .......................................
94
95
98
100
101
102
103
111
Enzymatic flux through the UDP-diNAcBac pathway ..............................................................
113
Conclusions................................................................................................................................
114
Acknowledgm ents......................................................................................................................
114
Experim ental Procedures ...........................................................................................................
115
References..................................................................................................................................
120
Chapter 4. Biochemical Analysis and Structure Determination of Bacterial
Acetyltransferases Responsible for the Biosynthesis of
UDP-N,N '-Diacetylbacillosam ine ...........................................................................................
123
Introduction................................................................................................................................
124
Results and Discussion ..............................................................................................................
127
Structure of the N. gonorrhoeaeAcetyltransferase PglB-ATD .................................................
127
Structure of the N. gonorrhoeaeAcetyltransferase PglB-ATD Bound to AcCoA .................... 130
6
Structure of the A. baumanniiAcetyltransferase Weel
............................
135
Analysis of Acetyltransferase Active-Site M utants ...................................................................
138
M utagenesis of the UDP-4-Amino Binding Pocket Reveals Kinetic Diversity......................... 142
Dichotomy Among N- and O-Linked Acetyltransferase AcCoA Binding Pockets...................143
Phylogenetic Analysis of Bacterial Acetyltransferases .............................................................
145
Conclusions................................................................................................................................
148
Acknowledgm ents......................................................................................................................149
Experim ental Procedures...........................................................................................................
149
References..................................................................................................................................
155
Chapter 5. Biochemical Characterization and Fragment-Based Inhibition of the
Campylobacterjejuni Am inotransferase PglE .......................................................................
158
Introduction................................................................................................................................
159
Results and Discussion..............................................................................................................
164
Expression and Purification of PglE ..........................................................................................
PglE Enzyme Characterization and Assay Development ..........................................................
PgIE Fragment Screening Results..............................................................................................
Small Molecule Fragment Inhibition of PglE Activity
............................
PglE Capillary Electrophoresis Assay Development.................................................................
Crystallization of Pg lE..................................................173
164
165
169
170
PgIE-M B730 Crystal Structure..................................................................................................
178
Second Generation M B730 Analogs..........................................................................................
180
Conclusions................................................................................................................................
171
185
Acknowledgm ents......................................................................................................................186
Experim ental Procedures...........................................................................................................
186
References..................................................................................................................................
194
Chapter 6. The Development of Inhibitors for the C.jejuni Acetyltransferase PglD
Utilizing a H igh-Throughput Screening Approach ..............................................................
197
Introduction................................................................................................................................
198
Results and Discussion..............................................................................................................202
Expression and Purification of PglD..........................................................................................202
Assay Development of PgD......................................................................................................202
Large-Scale Biosynthesis of UDP-4-Amino..............................................................................205
Broad HTS Screening Campaign ..........................................................................................
208
Synthesis of Thienopyrimidine Analogs.................................................................................215
Selectivity Screening with Homologous Acetyltransferases
7
........................
219
Discovery of the W eel Inhibitor 6010833 .................................................................................
222
Conclusions................................................................................................................................224
A cknow ledgm ents......................................................................................................................225
Experim ental Procedures ...........................................................................................................
225
References..................................................................................................................................233
Chapter 7. Biochemical Characterization and Fragment-Based Inhibition of the Neisseria
236
gonorrhoeae A cetyltransferase PglB-A TD ............................................................................
Introduction................................................................................................................................237
243
Results and D iscussion ..............................................................................................................
243
Expression and Purification of PglB-ATD ................................................................................
PglB-ATD Enzyme Characterization and Assay Development.................................................244
246
PglB-ATD Fragment Screening Results ....................................................................................
Second Generation Fragment Inhibition of PglB-ATD Activity ............................................... 247
Pg1B-ATD-Bound jma65 Crystal Structure.............................................................................251
Conclusions................................................................................................................................254
A cknow ledgm ents......................................................................................................................255
Experim ental Procedures ...........................................................................................................
References..................................................................................................................................260
8
255
List of Figures
Chapter 1
Figure 1-1.
Figure 1-2.
Figure 1-3.
Figure 1-4.
Figure 1-5.
Figure 1-6.
Figure 1-7.
Figure 1-8.
Figure 1-9.
The N- and O-linked protein glycosylation pathways ..........................................
Structures of bacterial carbohydrates ...................................................................
The UDP-diNAcBac biosynthetic pathway ........................................................
PglE and PglD crystal structures from C. jejuni
........................
The legionaminic acid biosynthetic pathway ......................................................
The GDP-GlcNAc biosynthetic pathway .............................................................
18
19
22
24
30
31
The pseudam inic acid pathway ............................................................................
34
The operon containing the pgl genes for production of Und-PP-diNAcBac .....
37
Phylogenetic tree comparing the genera of Campylobacterand Neisseria ......... 39
Chapter 2
Figure 2-1. Biosynthetic pathway of the pilin glycan in N gonorrhoeae..............................
52
Figure 2-2. Schematic representations of bacterial protein glycosylation pathways..............53
Figure 2-3. SDS-PAGE gel and Western blot of N. gonorrhoeaePgl proteins.......................55
Figure 2-4. 'H NMR spectrum of UDP-diNAcBac .................................................................
57
Figure 2-5. Kinetic analysis of PglB-ATD and PglC.............................................................60
Figure 2-6. Normal phase HPLC with fluorescence detection of 2-AB labeled glycans .....
63
Figure 2-7. Specificity analyses of PglB, PglA, and PglE ......................................................
65
Figure 2-8. peifcity of polyprenyldiphosphate-linked substrates of PgiE ......................... 66
Figure 2-9. PglO reaction turnover following incubation with Und-PP-diNAcBac-[ 3H]Gal.....69
Figure 2-10. PglO reaction turnover following incubation with pilin protein........................70
Figure 2-11. PglB reaction turnover following incubation with pilin protein ........................ 72
Chapter 3
Figure 3-1. The UDP-diNAcBac biosynthetic pathway in A. baumannii................................
Figure 3-2. SDS-PAGE gel of A. baumannii Wee proteins....................................................
Figure 3-3. Electropherogram trace of WeeK, WeeJ, and Weel reactions.............................
Figure 3-4. Michaelis-Menten binding curves for WeeK......................................................
Figure 3-5. Michaelis-Menten binding curves for WeeJ ........................................................
Figure 3-6. Michaelis-Menten binding curves for Weel......................................................
Figure 3-7. Substrate specificity of W eeH ................................................................................
Figure 3-8. Surface representation of the C. jejuni PglE binding pocket .................................
Figure 3-9. Aminotransferase primary sequence alignment .....................................................
Figure 3-10. Illustration of the aminotransferase binding pocket........................................
Figure 3-11. Surface representation of the C. jejuni PglD binding pocket...........................
Figure 3-12. Acetyltransferase primary sequence alignment................................................
Figure 3-13. Illustration of the acetyltransferase binding pocket .............................................
93
95
96
97
99
101
103
105
106
107
109
109
110
Chapter 4
Figure 4-1. Glycosylation pathways that utilize diNAcBac .................................................
Figure 4-2. The N. gonorrhoeaeapo PglB-ATD crystal structure ...........................................
9
125
129
Figure
Figure
Figure
Figure
Figure
Figure
4-3.
4-4.
4-5.
4-6.
4-7.
4-8.
Composite omit map of AcCoA electron density in PglB-ATD............................
AcCoA binding pockets in PglB-ATD and PglD...................................................
AcCoA binding pocket comparison between PglB-ATD structures......................
The A. baumannii apo Weel crystal structure ........................................................
Phylogenetic tree comparing bacterial acetyltransferases......................................
SDS-PAGE gel of acetyltransferase mutants.........................................................
132
134
135
136
147
152
Chapter 5
Figure 5-1. The C. jejuni N-linked protein glycosylation pathway ..........................................
Figure 5-2. Proposed aminotransferase mechanism of PglE ....................................................
Figure 5-3. Fragment-based approach for development of PglE inhibitors..............................
Figure 5-4. SD S-PA GE gel of PglE ..........................................................................................
Figure 5-5. UV trace of PglE protein purification ....................................................................
Figure 5-6. DTNB coupled enzymatic activity assay ...............................................................
Figure 5-7. Michaelis-Menten binding curves for PglE ...........................................................
Figure 5-8. PglE enzyme activity for DMSO and freeze thaws................................................
Figure 5-9. Enzym e titration of PglE ........................................................................................
Figure 5-10. PglE fragment IC 5 0 results....................................................................................
Figure 5-11. PglE follow-up fragment IC 50 results...................................................................
Figure 5-12. Electropherogram trace of PglE activity ........................................................
Figure 5-13. Electropherogram trace of MB730 PglE inhibition .........................
Figure 5-14. PglE sitting drop crystals......................................................................................
Figure 5-15. PglE crystals from streak seeding ....................................................................
Figure 5-16. PglE crystals from seed beads........................................................................
Figure 5-17. PglE asymmetric unit ...........................................................................................
Figure 5-18. PLP electron density from the PglE crystal structure .......................
Figure 5-19. PglE crystal structure with MB730 bound ...........................................................
Figure 5-20. Interactions between PglE and M B730................................................................
Figure 5-21. Second generation MB730 PglE inhibitors..........................................................
Figure 5-22. PglE MB730 analog compounds with their respective IC 50 values .....................
Figure 5-23. Final 'H-NMR for MB730 derivatives.................................................................
Figure 5-24. Final 'H-NMR for MB730 derivatives.................................................................
Figure 5-25. MB730-bound PglE crystal structure...................................................................
160
161
163
164
165
166
167
168
168
169
171
172
173
174
175
176
177
177
179
180
181
182
183
184
185
Chapter 6
Figure 6-1. The C. jejuni N-linked protein glycosylation pathway .......................................... 199
Figure 6-2. The C. jejuni PglD acetyltransferase crystal structure ........................................... 200
Figure 6-3. SDS-PAGE gel of purified PglD protein ...............................................................
202
Figure 6-4. PglD activity at varying MgCl 2 concentrations......................................................203
Figure 6-5. Michaelis-Menten binding curves for PglD ...........................................................
204
Figure 6-6. Electropherogram trace of UDP-4-amino biosynthesis ......................................... 207
Figure 6-7. Comparison of UDP-4-amino biosynthetic methods ............................................. 207
Figure 6-8. PglD and WeeI enzyme titration ............................................................................
208
Figure 6-9. IC 50 comparison of known acetyltransferase inhibitors ......................................... 209
Figure 6-10. PglD and WeeI maximum diversity screen..........................................................210
Figure 6-11. PglD HTS screen of the DOS compound collection............................................211
10
Figure 6-12.
Figure 6-13.
Figure 6-14.
Figure 6-15.
Figure 6-16.
Figure 6-17.
Figure 6-18.
Figure 6-19.
Figure 6-20.
Figure 6-21.
Figure 6-22.
Figure 6-23.
Figure 6-24.
PglD HTS screen of the MLPCN compound collection ...................................... 212
Compound hits from the MLPCN screen.............................................................212
Compound analogs from the MLPCN screen ......................................................
213
IC 50 values for the indolinone compound across EDTA concentrations ............. 214
Kinetic determination of the binding mechanism for BRD-K3819 ..................... 215
PglD-bound structure with MM-I........................................................................
215
Final 'H-NMR for thienopyrimidine derivatives .................................................
217
Final 1H-NMR for thienopyrimidine derivatives .................................................
218
Thienopyrimidine analogs with PglD IC 50 values................................................219
IC 50 selectivity results for Weel and PglB-ATD..................................................221
Additional analogs synthesized by the Broad Institute ........................................ 221
The isoxazole class of WeeI inhibitors ................................................................
222
Final 'H-NMR for 5906862 derivatives...............................................................223
Chapter 7
Figure 7-1. The N. gonorrhoeae Type IV pili...........................................................................238
Figure 7-2. The N- and 0-linked protein glycosylation pathways ........................................... 240
Figure 7-3. Biosynthesis of Und-PP-diNAcBac in N. gonorrhoeae......................................... 240
Figure 7-4. Fragment-based approach for development of PglB-ATD inhibitors....................242
Figure 7-5. SDS-PAGE gel of PglB-ATD ................................................................................
244
Figure 7-6. PglB-ATD activity in the presence of MgCl2 ............................... ................. .. ... ... 245
Figure 7-7. Michaelis-Menten binding curves for PglB-ATD..................................................245
Figure 7-8. PglB-ATD fragment melting and IC50 results ........................................................
247
Figure 7-9. Inhibition of PglB-ATD activity with MB211 analogs..........................................248
Figure 7-10. Inhibition of PglB-ATD activity with jm a48 .....................................................
249
Figure 7-11. IC 5 o analysis for second generation analogs of MB211.......................................250
Figure 7-12. SAR of MB211 second generation analogs with PglB-ATD............................... 250
Figure 7-13. Representative PglB-ATD crystals .................................................................
252
Figure 7-14. The PglB-ATD crystal structure with jma65 bound .......................................... 253
Figure 7-15. PglD-MB21 1, PglB-ATD-jm a65, and PglB-ATD-AcCoA structures...............253
Figure 7-16. Structural comparison of AcCoA and jm-a65 .....................................................
254
11
List of Tables
Chapter 2
Table 2-1. 1H chemical shift and coupling constant assignments for UDP-diNAcBac .......... 58
Table 2-2. Percent sequence identity for Pgl proteins.............................................................59
61
Table 2-3. Steady-state kinetic parameters for PglC and PglB-ATD .....................................
Chapter 3
Table 3-1.
Table 3-2.
Table 3-3.
Table 3-4.
Table 3-5.
Table 3-6.
Kinetic parameters for dehydratase enzyme..........................................................98
Kinetic parameters for aminotransferase enzymes ..................................................
Kinetic parameters for acetyltransferase enzymes...................................................
Sequence identity for aminotransferase enzymes....................................................
Sequence identity for acetyltransferase enzymes ....................................................
Constructs, accession numbers, and oligonucleotides .............................................
Chapter
Table 4-1.
Table 4-2.
Table 4-3.
4
Chapter
Table 5-1.
Table 5-2.
Table 5-3.
5
100
101
106
108
117
Kinetic parameters for the UDP-4-amino acetyltransferase substrate..................... 139
Kinetic parameters for the AcCoA acetyltransferase substrate ............................... 139
Data collection and refinement statistics for PglB-ATD and Weel......................... 153
Kinetic parameters for aminotransferase enzymes ..................................................
Pg1E data collection and refinement statistics .........................................................
PglE optimized protein geometry from MolProbity ................................................
168
178
178
Chapter 6
Table 6-1. Kinetic parameters for the C. jejuni acetyltransferase PglD....................................204
Chapter 7
Table 7-1. Kinetic parameters for acetyltransferase enzymes...................................................246
12
List of Schemes
Chapter 5
Scheme 5-1. Synthetic route for MB730 analogs utilizing tetrakis........................................... 182
Scheme 5-2. Synthetic route for MB730 analogs utilizing silica-bound DPP-Pd..................... 182
Chapter 6
Scheme 6-1.
Scheme 6-2.
Scheme 6-3.
Scheme 6-4.
Synthetic
Synthetic
Synthetic
Synthetic
route of the phosphonate ethyl ester.....................................................216
route of the thienopyrimidine ethyl ester product ................................ 216
route for the final thienopyrimidine product ........................................ 217
route to obtain isoxazole analogs .........................................................
223
13
List of Abbreviations
2-AB
Ab
AcCoA
AUC
BIS-TRIS
BSA
CE
CEF
CHAPS
C]
CMP
CoASH
Da
DDM
diNAcBac
DMSO
DOS
DTNB
EDTA
Gal
GalNAc
GDP
Glc
GlcNAc
GST
HEPES
HMQC
HR-MAS NMR
HTS
IC 50
IPTG
c-KG
L-Glu
LB
LE
Legionaminic acid
MALDI MS
MLPCN
MPD
MWCO
N-linked
NAD
2-aminobenzamide
Acinetobacter baumannii
acetyl coenzyme A
analytical ultracentrifugation
2,2-bis(hydroxymethyl)-2,2',2"-nitrilotriethanol
bovine serum albumin
capillary electrophoresis
cell envelope fraction
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate
Campylobacterjejuni
cytidine monophosphate
coenzyme A
dalton
n-dodecyl-p-D-maltopyranoside
N,N'-diacetylbacillosamine or 2,4-diacetamido-2,4,6-trideoxy-L-Dglucose
dimethyl sulfoxide
Diversity-Orientated Synthesis
5,5'-dithio-bis-(2-nitrobenzoic acid) or Ellman's reagent
ethylenediaminetetraacetic acid
galactose
N-acetylgalactosamine
guanosine diphosphate
glucose
N-acetylglucosamine
glutathione S-transferase
4-(2-hydroxyethyl)piperazine- 1 -ethanesulfonic acid
heteronuclear multiple quantum coherence
high-resolution magic angle spinning nuclear magnetic resonance
high-throughput screening
half maximal inhibitory concentration
iso-p-D-thiogalactosylpyranoside
cc-ketoglutarate
L-glutamate
lysogeny broth or Luria-Bertani broth
ligand efficiency
5,7-diacetamido-3,5,7,9-tetradeoxy-D-glycero-D-galacto-nonulosonic acid
matrix-assisted laser desorption ionization mass spectrometry
Molecular Libraries Probe Production Centers Network
2-methyl-2,4-pentanediol
molecular weight cutoff
asparagine-linked
nicotinamide adenine dinucleotide
14
NDP
Ng
Ni-NTA
NMR
nOe
O-linked
OMV
OTase
PDB
PEG
Pgl
PglB-ATD
PglB-PGTD
PLP
PMP
POC
Pseudaminic acid
PSUP
r.m.s.d.
SAR
SDS-PAGE
TEV
TFSS
TMHMM
UDP
UDP-4-amino
UDP-4-keto
UDP-DATDH
Und-P
Und-PP
nucleotide diphosphate
Neisseriagonorrhoeae
nickel-nitrilotriacetic acid
nuclear magnetic resonance
nuclear Overhauser effect
serine- or threonine-linked
outer membrane vesicles
oligosaccharyltransferase
Protein Data Bank
polyethylene glycol
protein glycosylation
acetyltransferase domain of PglB
phosphoglycosyltransferase domain of PglB
pyridoxal 5'-phosphate
pyridoxamine 5'-phosphate
percent of control
5,7-diacetamido-3,5,7,9-tetradeoxy-L-glycero-L-manno-nonulosonic acid
pure solvent upper phase
root mean square deviation
structure activity relationship
sodium dodecyl sulfate polyacrylamide gel electrophoresis
tobacco etch virus
type IV secretion system
tied mixture hidden markov model
uridine diphosphate
UDP-2-acetamido-4-amino-2,4,6-trideoxy-a-D-glucose
UDP-2-acetamido-4-keto-2,4,6-trideoxy-a-D -glucose
UDP- 2,4-diacetamido-2,4,6-trideoxy-ax-D-hexose
undecaprenyl phosphate
undecaprenyl diphosphate
15
Chapter 1. The Renaissance of Bacillosamine and Its Derivatives:
Characterization and Implications in Pathogenicity
16
Pathway
Introduction:
Glycosylation is the one of the most abundant protein modifications in nature and
regulates a variety of cellular processes including protein stability and folding, cell-cell
interactions, cell signaling, and the host immune response (1-3).
It is now recognized that
bacteria possess the machinery necessary to glycosylate proteins and that this modification may
play a role in its fitness and pathogenicity (4). In some bacterial protein glycosylation pathways,
the glycan is first assembled in a step-wise fashion onto a polyprenyl-diphosphate-linked carrier
on the inner membrane prior to being translocated into the periplasm for transfer onto an
acceptor protein.
In this case, attachment of bacterial glycans is accomplished by
oligosaccharyltransferase-mediated
en
bloc
transfer
onto
asparagine
(N-linked)
or
serine/threonine (0-linked) residues. Glycosylation of specific protein residues can also occur in
a sequential manner with nucleotide-activated sugars by Leloir glycosyltransferases.
Highly-
2 ,4 -diacetamido-2,4,6-trideoxy-D-glucose
(NN'-
modified,
bacterial
sugars,
including
diacetylbacillosamine or diNAcBac) are known to be incorporated into many proteins and in
some cases the presence of such sugars has been related to pathogenicity.
In particular,
diNAcBac is found at the reducing end of glycans in N- and O-linked protein glycosylation
pathways. The N-linked protein glycosylation (Pgl) system that produces a heptasaccharide in
Campylobacter jejuni is the most well-characterized pathway to date (Figure 1-1A).
This
modification is found on over 65 proteins in C. jejuni (5). The analogous O-linked pathway in
Neisseria gonorrhoeae generates a trisaccharide (Figure 1-1B) that, to date, has been identified
on 19 glycoproteins including the pilin protein PilE (6).
In each case, diNAcBac is first
biosynthesized as a UDP-sugar from the UDP-N-acetylglucosamine
(GlcNAc).
Further
modification of diNAcBac through a series of two enzymes results in legionaminic acid, a
17
molecular mimic of sialic acid (N-acetylneuraminic acid) (Figure 1-2). An analogous pathway
that utilizes an isomer of diNAcBac (2,4-diacetamido-2,4,6-trideoxy-L-altropyranose) produces
pseudaminic acid, another sialic acid-like sugar (Figure 1-2). In contrast to the N- and O-linked
glycosylation pathways that implicate diNAcBac directly, these elaborated diNAcBac derivatives
are integrated into O-linked glycoproteins via sequential addition to proteins. Legionaminic and
pseudaminic acids are essential for flagellar assembly in Campylobacter spp., Legionella
pneumophila, and Helicobacterpylori (7).
The biosynthetic pathways responsible for these
unique sugars have recently been linked to bacterial pathogenesis (8-10) and therefore represent
a novel target in the fight against microbial resistance.
A
*1M
Udip-G
PuDCfATO~
UD P SMP 'D '
UD P 4
* V-Acctylgiueossinc
PilE
phak
P
I
Qc.ainsos
ff.
dophphazcPH11
Figure 1-1. (A) The N-linked protein glycosylation pathway from C jejuni showing the
heptasaccharide glycan attached to the PEB3 protein. (B) The O-linked protein glycosylation
pathway from N. gonorrhoeaeshowing the trisaccharide glycan attached to the PilE protein.
Both pathways utilize the unique, bacterial sugar diNAcBac at the reducing end of the glycan.
ATD, acetyltransferase domain; PGTD, phosphoglycosyltransferase domain.
18
NHAc
AOHH
Ha
AcHHAj\
H
HO
H
HO0
NHAc
O-UDP
AcHN-i:
HO
0-UDP
UDP-diNAcBac
OCMP
O
UDP-2,4-diacstamIdo-2,4,6-
CMP-sIalic acid
trideoxy-p-L-altropyranose
OH
0OCMP
A
O-CMP
pH
COOH
HO
0
AcHN--*MH
AcHN HOAH
COOH
HO
NHAc
CMP-legionaminic acid
CMP-pseudaminic acid
Figure 1-2. Structural comparison of carbohydrates found in bacteria that are discussed in this
review.
The main focus of this review is the bacterial sugar diNAcBac, which is biosynthesized
from a series of three conserved enzymes in all pathways identified thus far. The assembly of
diNAcBac is composed of a dehydratase, aminotransferase, and acetyltransferase that utilize
UDP-GlcNAc as the initial substrate.
These enzymes have been extensively studied in the
Gram-negative bacterium C. jejuni (26-28). Subsequent to the work on this N-linked protein
glycosylation pathway, diNAcBac was discovered in glycans, which modify serine and threonine
residues (0-linked) in other pathogenic bacteria including N. gonorrhoeae and Acinetobacter
baumannii (11-12).
Importantly, diNAcBac and a stereoisomer of this sugar serve as a starting
point for the biosynthesis of legionaminic and pseudaminic acid. In this review, the pathway
enzymes responsible for the biosynthesis of these unique bacterial carbohydrates will be
explored in detail.
Our current understanding of the biosynthesis and incorporation of these
19
highly-modified sugars onto protein virulence factors provides the necessary motivation to
investigate their biological relevance regarding bacterial pathogenicity.
N,N'-Diacetylbacillosamine
Discovery and Characterization
The serendipitous discovery of bacillosamine occurred in 1957 by Nathan Sharon while
exploring polypeptide synthesis in Bacillus licheniformis, a Gram-positive bacterium usually
found in soil (13).
Following purification of an uncharacterized polysaccharide from B.
licheniformis, an unknown amino sugar was detected by paper chromatography. Elemental and
chemical analysis of this sugar revealed the presence of two nitrogen atoms at the C-2 and C-4
positions, with the latter site acetylated. The final structure of this carbohydrate was assigned as
4-acetamido-2-amino-2,4,6-trideoxyhexose (4-N-acetylbacillosamine) based upon these initial
experiments (14-15). Confirmation of this structure occurred 10 years later through a 12-step
chemical synthetic approach utilizing glucosamine as the starting material (16). More recently, a
chemical synthesis has afforded the undecaprenyl pyrophosphate-linked bacillosamine (17) as
well as bacillosamine-containing disaccharides (18). Since its discovery, bacillosamine and the
corresponding N-acetylated derivatives have been found in a variety of pathogenic bacteria. For
example, it is found as the reducing-end sugar in N-linked glycoproteins (C. jejuni) and O-linked
glycoproteins (Neisseriaspp.). Additionally, bacillosamine has been identified in the O-antigen
of Pseudomonas reactans(19) and Vibrio cholera (20), the core region of the lipopolysaccharide
(LPS) in Francisellanovicida (21), and the capsular polysaccharide (CPS) from Alteromonas sp.
CMM155 (22). The fundamental question as to why bacteria utilize bacillosamine is currently
20
unanswered and remains an important area of research although some hypotheses suggest that
this sugar is not recognized by mammalian hosts and therefore may serve as a decoy to host
immune systems and glycan degrading enzymes.
Biosynthesis in BacterialPathogens
Although the biosynthetic route to diNAcBac was first suggested by Sharon in 1964 (23),
it took over 40 years to verify the initial proposal. Following genome sequencing of C. jejuni
(24), a gene locus distinct from the lipooligosaccharide cluster was identified that shared
significant homology to previously characterized protein glycosylation genes (25).
These
encoded proteins were ultimately identified through biochemical characterization and found to
be responsible for the biosynthesis of diNAcBac from the UDP-activated form of GlcNAc.
Biochemical analysis of Cj 1 120c, later renamed PglF, resulted in the identification of the
first enzyme in this pathway, a membrane-bound NAD+-dependent dehydratase (26).
PglF
catalyzes the NAD+ dependent C4 oxidation of UDP-GlcNAc, which promotes elimination of
water across the C5-C6 carbons of the pyran ring. Reduction of the resultant cP-unsaturated
system at C6 produces the UDP-4-keto sugar and regenerates NADH back to its oxidized state
(Figure 1-3). One- and two-dimensional NMR experiments confirmed the stereochemistry of
this product to be UDP-2-acetamido-4-keto-2,4,6-trideoxy-a-D-glucose
(26).
Unlike the
pseudaminic acid dehydratase (Cj 1293/PseB) also found in C. jejuni (see below), PglF does not
contain C5 epimerase activity. Kinetic characterization of PglF resulted in a kcat/Km of 17 M- s-1
for UDP-GlcNAc, making the dehydratase the least catalytically efficient enzyme on the
diNAcBac pathway and thus the rate limiting step (27).
21
Further characterization of PglF
homologs in N. gonorrhoeae (PglD) and A. baumannii (WeeK) have resulted in similar kinetic
parameters, lending support to the proposal that the dehydratase plays the role of "gatekeeper" in
this pathway (11-12).
The diNAcBac dehydratase enzymes have yet to be structurally
characterized probably due to the challenges associated with membrane protein crystallization.
OH
HO
HO
0
dehydratase
-X
NAD* NADH
AcHN
O-UDP
UDP-GIcNAc
aminotransferase
0
HO
AcHNO-UDP
H2 0
UDP-4-keto
/
PMP
PLP
H2 N
HO
0
AcHN
/
_N
AcCoA
CoA
AcHN
H
0
A
N
0-UDP
><O-UDP
L-Glu aKG
acetyltransferase
UDP-4-amino
UDP-diNAcBac
Figure 1-3. The biosynthetic pathway in pathogenic bacteria that produces the nucleotideactivated UDP-diNAcBac sugar.
The adjacent gene to PglF in the pgl (protein glycosylation) locus (Cj 1121c/PglE) was
defined as a pyridoxal 5'-phosphate(PLP)-dependent aminotransferase that catalyzes the transfer
of the amino group from L-glutamate to the C4 position of UDP-4-keto in two distinct steps that
cycle between the PLP and PMP forms of this cofactor (26,28).
Catalysis is initiated by the
formation of an imine involving the UDP-4-keto sugar and pyridoxamine 5'-phosphate (PMP).
Following the conversion to the external aldimine, the UDP-4-amino product is released via
transimination of the catalytic lysine residue in the active site. The internal aldimine resulting
from this reaction results in the recycling of PMP through the conversion of L-glutamate to aketoglutarate. Although the amino-group donor was determined to be glutamate, PglE has also
been shown to exhibit moderate activity with methionine, glutamine, alanine, and cysteine (28).
The UDP-4-amino product of this reaction was again confirmed as UDP-2-acetamido-4-amino2,4,6-trideoxy-a-D-glucose based upon NMR experiments including the nuclear Overhauser
effect (nOe) peak pattern and the J-coupling constants (28). Kinetically, PglE is a more efficient
22
enzyme with respect to PglF when comparing UDP-sugar substrates (kcat/Km = 6600 M~1 s-1)
(12). However, L-glutamate is a poor substrate for this reaction (Km = 11 mM), which is the
result of the high intracellular concentration of this amino acid (29).
Studies with the
aminotransferase homologs in N. gonorrhoeae (PglC) and A. baumannii (WeeJ) confirmed the
low binding affinity to L-glutamate.
With respect to UDP-4-keto turnover, both of these
enzymes were catalytically less active relative to PglE. Bacterial aminotransferases such as PglE
have been shown to form homodimers in solution following previous work with PseC and WbpE
(30-31). Furthermore, the crystal structure of PglE supported solution state studies revealing that
the enzyme exists as a dimer in the asymmetric unit (32) (Figure 1-4A). The two active sites are
formed on opposite faces of the dimer interface and are separated by a ~30 A distance. At the
bottom of each binding pocket resides the PLP cofactor necessary for the transamination
reaction. Structures of the apo and PLP-bound forms have been solved, however attempts to
crystallize this protein in the presence of the UDP-sugar substrate or product have not yet proven
successful, but it is presumed that substrate binding generally mimics what has been described
from previous aminotransferases (30). However, as these proteins seem to be highly stereospecific for a particular UDP-sugar, the questions remains as to what confers this selectivity.
23
A
B
Figure 1-4. (A) The C. jejuni PglE aminotransferase crystal structure (PDB code 1061) bound to
PLP depicted in cartoon (left) and space-filling (right) format. The dimer is the biological unit
and each protomer has been individually colored for clarity. (B) The composite C. jejuni PglD
acetyltransferase crystal structure constructed from the UDP-4-amino (PDB code 3BSS)
(depicted in brown) and AcCoA (PDB code 3BSY) (depicted in gray) bound structures. For the
purpose of clarity, the 2 additional binding pocket substrates have been removed and the
protomers individually colored. The biological unit is a trimer illustrated in cartoon (left) and
space-filling (right) form.
The final step of diNAcBac biosynthesis relies upon PglD (CjI 123c) to acetylate the C4
position on the UDP-4-amino sugar in an acetyl coenzyme A (AcCoA)-dependent reaction. This
reaction is catalyzed by an active site histidine that acts as a general base to abstract a proton
from the C4 amine promoting nucleophilic attack on the thioester of AcCoA.
Utilizing a
combination of radiolabel transfer with [3 H] AcCoA, ESI-MS, and NMR, this sugar was
unequivocally shown to be UDP-diNAcBac (UDP-2,4-diacetamido-2,4,6-trideoxy-x-D-glucose)
(27).
This acetyltransferase also exhibited the greatest catalytic efficiency among pathway
enzymes for its UDP-sugar (kcat/Km = 4.0 x 107 M-1 s-1) and AcCoA (kcat/Km = 5.5 x 107 M-1 s-1)
24
substrates (12). Similar Michaelis-Menten catalytic efficiencies were again obtained for the N.
gonorrhoeae(PglB) and A. baumannii (Weel) diNAcBac acetyltransferases. This high degree of
enzyme efficiency creates a pathway flux where rapid consumption of the UDP-4-amino sugar
drives the rate-limiting step of UDP-4-keto conversion by PglF. Interestingly, PglD contains a
relaxed substrate specificity based upon its ability to acetylate UDP-4-amino-4,6-dideoxy-P-LAltNAc, an intermediate along the pseudaminic pathway (33) that may allow for cross-talk
between these two pathways.
However, this acetyltransferase is specific only for sugar-
nucleotide substrates as it is unable to acetylate aminoglycosides. PglD forms a homotrimer in
solution based upon sedimentation velocity analytical ultracentrifugation (AUC) experiments
and a protein crystal structure (34-35). The C-terminal left-handed P-helix domain of adjacent
protomers forms the AcCoA binding pocket, whereas the N-terminal domain contains a
-a-p-a-p-
Rossmann fold motif to accommodate the UDP-4-amino substrate (Figure 1-
4B).
Connection to Pathogenicity
Due to the ever-increasing resistance to bacteriocidal antibiotics from selective pressure,
developing therapies that target pathways related to pathogenicity (such as glycosylation) have
become an important strategy (36). This approach is an attractive option because strategies that
target pathogenicity would not affect bacterial survival and therefore would potentially
circumvent
selective pressures associated with current antibiotics. In the context of
pathogenicity, the N-linked protein glycosylation pathway in C. jejuni is a significant area of
interest. This bacterial pathogen is the leading cause of gastroenteritis and may result in the
25
development of Guillain-Barre Syndrome (37-39).
This pathway produces a heptasaccharide
containing diNAcBac that modifies a variety of proteins associated with virulence (40). The
enzymes responsible for the biosynthesis of diNAcBac are appealing antibacterial targets since
they are specific only to prokaryotes. Additionally, C. jejuni has exhibited increased resistance
towards front-line antibiotics including the macrolides and fluoroquinolones (41), which inhibit
protein synthesis and DNA unwinding, respectively.
Previous studies have examined the importance of global N-linked protein glycosylation
by disrupting the genes responsible for diNAcBac biosynthesis (pglF, pglE, pglD) (42).
Utilizing high-resolution magic angle spinning nuclear magnetic resonance (HR-MAS NMR)
and whole-cell lysate reactivity to SBA lectin, the authors concluded that loss of these genes
resulted in the inability to produce the heptasaccharide in C. jejuni. Additionally, the ApglD and
ApglE strains were examined for their ability to colonize 1-day-old chicks (42). In both cases, no
colonization was detected following inoculation due to inactivation of this glycosylation pathway
validating these two genes as targets in pathogenicity. Transposon mutagenesis of C. jejuni
verified these results by identifying pglF and pglE as essential genes for colonization of the
chick gastrointestinal tract (43). In a related study, the C. jejuni pglE mutant impaired the
invasion of intestinal epithelial cells and colonization of intestinal tracts in mice (44).
Confirmation of the relationship between pathogenicity and diNAcBac biosynthesis has
increased the focus on identifying the individual glycoproteins responsible for cell invasion and
colonization. The glycoprotein VirB 10, a structural component to the type IV secretion system
(TFSS), was previously identified in C. jejuni (45). Disruption of the pglE gene resulted in the
conclusion that the Pgl system glycosylates VirBlO at two sites, N32 and N97. Removal of the
N97 glycosylation site produced a 10-fold decrease in natural competency that could be rescued
26
by complementing with wild-type VirB 10. This was the first example of N-linked glycosylation
being attributed to stability and function of a known virulence factor. Recently, 16 N-linked
glycoproteins were identified and found to be associated with C. jejuni outer membrane vesicles
(OMVs) including the known antigenic PEB3 adhesin (46). Pathogens employ OMVs to deliver
bacterial proteins into host cells, making this an important finding in the relationship between
immunogenic glycoproteins located in the periplasm.
Similar to the work exploring the connection between N-linked protein glycosylation and
bacterial pathogenicity, recent studies have focused on a comparable role for O-linked
glycosylation.
Specifically, the association between O-linked protein glycosylation and
pathogenicity has been examined in N. gonorrhoeae. Studies have identified the PilE protein in
Type IV pilin to be glycosylated at a single site (Ser-63) with diNAcBac at the reducing end of
the trisaccharide (47). Further experiments in Neisseria spp. with the PilE glycoprotein have
shown that it is both immunogenic and antigenic (48). Mass spectrometry analysis following 2D
gel electrophoresis and immunoblotting identified additional periplasmic glycoproteins that are
implicated in protein folding, solute uptake, and respiration (49).
Strains of N. gonorrhoeae
deficient in the ability to biosynthesize diNAcBac through disruption of the dehydratase gene
(pglD) exhibited decreased adherence and invasion to primary human cervical epithelial (pex)
cells (50). Similar to N. gonorrhoeae,Neisseria meningitidis contains a homologous O-linked
protein glycosylation pathway that can modify PilE with the same trisaccharide (51-53). Recent
studies have indicated that this pilin-linked glycan is essential for the adherence of N.
meningitidis to human bronchial epithelial cells (54). Further work in an in vivo system is
necessary to identify a link between pathogenicity and glycoproteins biosynthesized from this
pathway, but this is an exciting and active area of research.
27
Derivatives of N,N'-Diacetylbacillosamine
LegionaminicAcid
Sialic acid is a 9-carbon a-keto sugar that is expressed on mammalian glycoproteins
found on the cell surface and is responsible for cell-cell communications (Figure 1-2) (55).
Bacteria have also demonstrated the ability to display sialic acid and the nonulosonic derivatives
(legionaminic and pseudaminic acid) on their outer surface. Bacteria produce legionaminic acid
(5,7-diacetamido-3,5,7,9-tetradeoxy-D-glycero-D-galacto-nonulosonic
acid) that retains the
exact stereochemistry of sialic acid as determined by total synthesis of this sugar (56). It has
been hypothesized that bacterial pathogens utilize legionaminic acid as a molecular mimic of
sialic acid, which is prominently presented on mammalian cells and is an important factor in
This sugar was first identified as a repeating
immune system regulation and adhesion (56).
homopolymer in the 0-polysaccharide of LPS in L. pneumophila, which is the causative agent of
Legionnaires' disease (57). Recently, legionaminic acid has been found in a variety of other
pathogenic bacteria associated with the flagella of Campylobacter coli (58), the O-antigen of A.
baumannii (59-60), as well as Cronobacterturicensis (61) and E. coli (62). In fact, over 20% of
the 1000 microbial genomes examined to date contain the putative biosynthetic genes for this
pathway, making this sugar far more widespread than originally believed (63). Work with the 0antigen of Vibrio fischeri has illustrated the importance of legionaminic acid in colonization of
the natural host of this bacterium (64). The disruption of this O-antigen through a gene knockout of waaL, the ligase responsible for O-antigen assembly onto the LPS, resulted in a motility
defect. Further studies indicate that this O-antigen null strain has a significantly reduced ability
to colonize its natural host organism and cannot compete with wild type V. fischeri in cocolonization assays (64). Although legionaminic acid is located on known bacterial virulence
28
factors, the relationship between this sugar and host cell interactions remain poorly understood.
Further investigation is therefore warranted to determine whether disruption of the biosynthesis
of legionaminic acid has an effect on bacterial pathogenicity.
The CMP activated form of legionaminic acid was originally shown to be biosynthesized
from UDP-diNAcBac by a series of 3 enzymes in L. pneumophila (Figure 1-5) (65-66). These
enzymes show homology to those in the sialic acid biosynthetic pathway (NeuC,B,A) and are
necessary for assembly of functional flagella. Interestingly, a phylogenetic analysis has recently
reported that the enzymes responsible for legionaminic acid biosynthesis were most likely
adapted to produce sialic acid in bacteria (63). Previous hypotheses suggested that pathogens
acquired the sialic acid pathway through horizontal gene transfer or convergent evolution. The
first enzyme of this pathway, a NeuC homolog, creates 2,4-diacetamido-2,4,6-trideoxymannose
by hydrolyzing the giycosidic-UDP linkage and inverting the stereochemistry of UDP-diNAcBac
at the C2 position. A two-dimensional heteronuclear NMR experiment (HMQC) was utilized to
show that the a-anomer is initially formed with retention of stereochemistry at Cl. The authors
concluded that the mechanism for NeuC proceeds through anti elimination of UDP and syn
hydration of the glycal double bond in a similar fashion to the homologous enzyme in the sialic
biosynthetic pathway (65, 67-68). The kinetic parameters for this reaction (kcat/Km = 1.6
x
106
M- s-) suggest that UDP-diNAcBac is the physiological substrate. The inability of NeuC to
turnover UDP-GlcNAc, the natural substrate for the sialic and pseudaminic acid pathways,
further supports these findings.
The NeuB homolog, then utilizes phosphoenolpyruvate (PEP)
to condense a 3-carbon unit onto 2,4-diacetamido-2,4,6-trideoxymannose
diacetyllegionaminic acid.
to yield N,N -
This reaction presumably proceeds through an oxocarbenium
intermediate following attack on the open chain aldehyde form of 2,4-diacetamido-2,4,629
trideoxymannose by C3 of PEP. Addition of water to this intermediate results in displacement of
phosphate and the formation of the a-keto acid. NeuB had a surprisingly low level of activity
that may be in part due to the conditions in the assay format used in this study (65). The natural
sialic acid substrate ManNAc exhibited no activity with NeuB confirming that this enzyme is not
involved in this biosynthetic pathway. The final enzymatic step in the process, accomplished by
a NeuA homolog, activates the a-keto
sugar with CTP to yield the CMP-N,N'-
diacetyllegionaminic acid donor. Due to low amounts of substrate obtained from the previous
step, mass spectrometry was utilized to follow generation of the CMP activated sugar product.
LegG
OH
hydrolase/
AcHN S
2-epimerase
o-UDP
(GDP)
UDP(GDP)-diNAcBac
H20
\NF
AHH
L
synthase
PEP
UDP
(GOP)
H
AcHN
Pi
0
COOH
HO
Legionaminic Acid
2,4-diacetamido-2,4,6-
LegF0-M
synthetase
CTP
ACHN
0
COOH
PPiH
CMP-Legionaminic Acid
trideoxymannose
Figure 1-5. The legionaminic acid pathway that utilizes diNAcBac in its UDP or GDP
nucleotide-activated form.
Utilizing a targeted metabolomics approach with mass spectrometry and NMR, McNally
et al. (58) identified a series of genes responsible for the biosynthesis of legionaminic acid in C.
coli VC 167 that are distinct from the Neu homologs. Interestingly, inactivation of the diNAcBac
pathway enzyme PglE did not have an effect on the production of CMP-legionaminic acid in C.
coli. It is apparent then, that biosynthesis of legionaminic acid involves a distinct pathway that
does not utilize diNAcBac as a substrate in C. coli. Further work expanded the knowledge of
this pathway in C. jejuni where it was demonstrated through a bioinformatic and metabolomic
approach that the biosynthesis of legionaminic acid involved GDP-linked intermediates rather
30
than UDP-activated sugars (69).
This pathway was shown to proceed through a series of 5
enzymes that converts fructose-6-phosphate to GDP-GlcNAc with the nucleotidyltransferase
PtmE activating glucosamine-l-phosphate (Gln-l-P) with GDP followed by acetylation by
GlmU (Figure 1-6). Although GlmU was utilized in this study, the acetylation reaction did not
result in complete conversion and the putative N-acetyltransferase for this reaction has not been
identified.
Conversely, the formation of UDP-GlcNAc relies upon the bifunctional GlmU
enzyme that first acetylates Gln-1-P followed by uridylation (70).
From this activated GDP
sugar, a dehydratase (LegB), an aminotransferase (LegC), and an acetyltransferase (LegH)
produce GDP-diNAcBac in a similar fashion to the Pgl pathway enzymes that biosynthesize
UDP-diNAcBac. The final three enzymes on this pathway resemble the enzymatic function of
the Neu homologs discussed above (Figure 1-5). LegG is a GDP-sugar hydrolase/2-epimerase
that catalyzes the conversion of GDP-diNAcBac to 2,4-diacetamido-2,4,6-trideoxymannose, the
same product from the NeuC reaction. Condensation of PEP with this sugar proceeds through
the synthase LegI and activation of this sugar with CTP is accomplished by LegF to yield the
final activated sugar CMP-legionaminic acid. These enzymes were unable to efficiently use the
UDP-sugar precursors from the UDP-diNAcBac pathway, confirming the specificity for GDPactivated sugars. Since C. jejuni utilizes UDP-linked sugars for flagellar O-linked glycosylation
(pseudaminic acid) and N-linked protein glycosylation (diNAcBac), an alternative GDP-linked
sugar pathway provides another method for cellular regulation and control.
PtmA
O' O
0H
HO
H
OH
Fru-6-P
ON
/
PtmF
L-Gin
OH
m_
3
PgmL
H
HO
L-Giu
OH
HONO
opo3
H
GIcN-6-P
OH
PtmE
GIcN-1-P
GTP
OH
P
OH
O
H_0
N O-GDP
GDP-GIcN
GImU?
/I
_"
AcCoA
CoA
OH
HO
0
H A
O-GDP
Ac.HNHfN
GDP-GIcNac
Figure 1-6. The initial steps of the legionaminic acid biosynthetic pathway that forms the GDPGlcNAc starting substrate for the remaining portion of the reaction.
31
PseudaminicAcid
Similar to legionaminic acid, pseudaminic acid is a 9-carbon sialic acid analog that is
found on flagellin proteins in C. jejuni and H. pylori (Figure 1-2) (7,71). These glycoproteins
are absolutely essential for proper assembly of functional flagella and bacterial motility making
Pseudaminic acid (5,7-diacetamido-3,5,7,9-
this an important virulence target (72-73).
tetradeoxy-L-glycero-L-manno-nonulosonic acid) is an isomer of legionaminic acid that is
biosynthesized
from UDP-GlcNAc
with a stereoisomer of UDP-diNAcBac
(inverted
stereochemistry at C4 and C5) as an intermediate resulting in the CMP-activated sugar (Figure
1-7).
The first enzyme, PseB, exhibits C6 dehydratase and C5 epimerase activity (7,26) to
generate UDP-2-acetamido-2,6-dideoxy-p-L-arabino-4-hexulose.
It is interesting to note that
PseB utilizes NADP+ to oxidize UDP-GlcNAc at C4 rather than NAD+, the cofactor in the PglF
dehydratase reaction on the diNAcBac pathway (26,74-75). PseB is able to bind UDP-GlcNAc
with a much greater affinity (140-fold) with respect to the diNAcBac dehydratase (PglF)
resulting in a higher catalytic efficiency (31-fold) (76). Upon accumulation of this 4-keto sugar,
PseB is able to catalyze an additional C5 epimerization to regenerate UDP-2-acetamido-4-keto2,4,6-trideoxy-a-D-glucose, the UDP-4-keto sugar utilized in the diNAcBac pathway (77). This
additional reaction allows for cross-talk between the two pathways and potentially establishes
another level of control for the production these sugars in relation to pathogenicity in the
bacterial cell. The keto sugar generated from PseB is utilized by the aminotransferase PseC in a
PLP-dependent fashion to form the 4-amino product (26,30,78).
The diNAcBac C. jejuni
aminotransferase (PglE) produces an isomer of this sugar that varies in its stereochemistry at the
C4 and C5 positions.
Interestingly, PglE is a catalytically more efficient enzyme (44-fold
increase in kcat/Km) with respect to PseC.
The aminotransferases from the pseudaminic and
32
diNAcBac pathways display no cross-talk with their respective substrates demonstrating the
stereospecificity of each enzyme. The PseC crystal structure with PMP bound to the UDP-sugar
product has not been able to address how these aminotransferases can differentiate the UDP-4keto substrates that vary in stereochemistry only at the C5 position. Clearly, a PglE crystal
structure bound to either the substrate or product UDP-sugar would finally answer how these
enzymes accomplish complete substrate specificity. PseH then acetylates the 4-amino sugar in
an AcCoA-dependent manner forming the UDP-diNAcBac isomer UDP-2,4-diacetamido-2,4,6trideoxy-p-L-altropyranose (79). The fourth step of the pathway relies on UDP hydrolysis by
PseG resulting in 2,4-diacetamido-2,4,6-trideoxy-L-altropyranose.
Mechanistic studies of this
enzyme determined that the hydrolysis of UDP proceeded in a concerted fashion with attack by a
water molecule at C-I and cleavage of the C-O anomeric bond (80). Additionally, apo and UDPbound PseG crystal structures allowed for the identification of His 17 as the general base utilized
for activating the nucleophilic water molecule (81).
The Psel synthase catalyzes the
condensation of PEP with 2,4-diacetamido-2,4,6-trideoxy-L-altrose generating pseudaminic acid
in a similar fashion as the NeuB homolog in the legionaminic acid pathway (82). Analysis of
this enzyme revealed the requirement of a divalent metal ion for catalysis and that the formation
of pseudaminic acid proceeds through a tetrahedral intermediate after attack of C-3 from PEP to
the open chain aldehyde sugar. Following collapse of this intermediate, inorganic phosphate is
released followed by cyclization to the pyranose form of pseudaminic acid. The final step in the
pathway forms the CMP-activated pseudaminic acid that relies upon the enzyme PseF and CTP
(79).
This reaction was found to be dependent upon alkaline pH and Mg2 , and that CMP-
pseudaminic acid inhibited the formation of pseudaminic acid in a "one-pot" reaction. Further
studies determined that CMP-pseudaminic acid was a potent inhibitor of the first enzyme (PseB)
33
in the pathway with a Ki(app) of 18.7 pM, allowing for control over the biosynthesis of this
product (77).
A metabolic approach utilizing gene knockouts and detection of nucleotide
intermediates by capillary electrophoresis-electrospray mass spectrometry confirmed the direct
involvement of the Pse proteins in pseudaminic acid biosynthesis (83). Recently, pseudaminic
acid was chemically synthesized from GlcNAc allowing for the ability to conduct large-scale
studies to better understand the relationship this sugar has with bacterial O-linked glycosylation
(84). Whereas some of the pathway enzymes described above have been examined in an in vitro
biochemical setting, a comprehensive kinetic analysis is still lacking. Such studies would aid in
the understanding of the interplay between substrates and enzymes and are essential for
developing these enzymes as antibacterial targets.
OH
HO4H
PseC
PseB
0
HO
HO
O-UDP
UDP-GIcNAc
NADP
AcHNI
O-UDP
NADPH
H2 0
L-Glu
UDP-4-keto(Pse)
PLP
H2N
a-KG
NHAc
AcCo
-D
__OH
PseH
-UP
__OH
PMP
CoACHN
UDP-4-amino(Pse)
NHAC
UDP-dINAcBac(Pse)
I
H 20
PseG
UDP
OH
P-H
O-CMP
OH
PseF
0
AcHN
COOH
HO
NHAc
Psel
/--'
COOH
PP
AcHN
CTP
CMP-Pseudaminic Acid
HO
NHAc
Pi
Pseudaminic Acid
0 f
H
AcHN
PEP
NHAc
dINAcBac(Pse)
Figure 1-7. The pseudaminic acid pathway that produces an isomer of diNAcBac that results in
the formation of the CMP-activated form of pseudaminic acid.
The two most studied bacterial pathogens in terms of pseudaminic acid flagellar
glycosylation are C. jejuni and H. pylori, however other species such as Clostridium botulinum
and Aeromonas caviae are also currently under investigation (85-91).
34
The main C. jejuni
flagellin protein FlaA is glycosylated at 19 serine/threonine residues, with 8 of these sites
contributing to motility and autoagglutination of the bacteria (71,92).
Disruption of the
pseudaminic acid biosynthesis genes pseB and pseC resulted in non-glycosylated flagellin
confirming their importance in this bacterium. Further mutational studies were also conducted in
C. jejuni with pseF,pseG, and pseH to verify their role in pseudaminic acid production (73). In
each case, disruption of these genes resulted in a non-motile phenotype that lacked flagella
filaments and hook structures. These mutational strains also exhibited a reduction in adherence
and invasion of intestinal epithelial cells.
Recent studies have demonstrated that flagellin
proteins can undergo spontaneous antigenic variation through dimethylglyceric derivatives of
pseudaminic acid (93-94).
Although the homopolymeric-tract-containing Cj 1295 gene is
responsible for this modification, the question of why this occurs remains unanswered. This type
of structural diversity, which is observed in variant pilin glycoproteins in N. gonorrhoeae (9596), may be important for evading the host immune response during colonization. H pylori has
also been extensively studied resulting in the identification of the FlaA and FlaB flagellin
proteins that are modified with pseudaminic acid (97-98). It was previously found that flagellar
motility is a requirement for colonization in both in vitro and in vivo model systems (99-100).
Insertional mutagenesis led to the discovery of 3 genes (HP0326A, HP0326B, HP0178) that are
directly involved with the biosynthesis of pseudaminic acid (72).
Disruption of each gene
resulted in a non-motile phenotype, decreased flagellin protein, and accumulation of UDP-sugar
nucleotide precursors. H pylori is also able regulate pathogen motility through deglycosylation
of pseudaminic acid on FlaA through the HP0518 protein (101). The HP0518 knockout mutant
exhibited hyper-motility and a superior ability to colonize C57BL/6 mice in vivo confirming the
role H pylori flagella play in pathogen-host interactions. A concerted effort has led to the
35
identification of two flagellin proteins (FlaA and FlaB) that play a role in pathogenesis, however
the glycome of H pylori is still poorly understood. Glycan metabolic labeling coupled with
mass spectrometry analysis has resulted in the identification of 125 O-linked glycoproteins,
many linked to pathogenesis (102). Work still remains on the characterization of these putative
glycoproteins and what role glycosylation plays in H pylori pathogenesis.
Beyond N,N'-Diacetylbacillosamine
An interesting facet of the diNAcBac pathway is the observation that in some pathogenic
bacteria, the acetyltransferase and phosphoglycosyltransferase enzymes are joined as a single
polypeptide chain with two distinct domains.
The phosphoglycosyltransferase enzyme is
responsible for catalyzing the reaction between UDP-diNAcBac and undecaprenyl phosphate
resulting in the formation of the polyprenyldiphosphate-monosaccharide product (Und-PPdiNAcBac).
Membrane localization of diNAcBac through attachment to undecaprenyl
phosphate increases the local minimum concentration of this lipid-linked sugar allowing for
efficient synthesis of the final glycan (103).
The question still remains as to whether these
enzymes first existed as separate enzymes and were later integrated into one functional unit or
vice versa. A bioinformatic and phylogenetic approach were utilized in an attempt to provide a
potential answer to this question. The genes responsible for the biosynthesis of UDP-diNAcBac
and transfer to the lipid carrier are all located consecutively on the same operon (Figure 1-8). It
has previously been hypothesized that these enzymes work in concert to provide the final lipidlinked monosaccharide through a channeling mechanism (27,76).
Since these enzymes are
expressed collectively, one can envision the sequential transfer of products from one enzyme to
the next to allow for segregation of the UDP-sugar intermediates from other closely related
36
pathways. This type of mechanism is employed in the biosynthesis of sialic acid in C. jejuni
(104). An increase in pathway efficiency would then be achieved by creating a bifunctional unit
containing two enzymes from this pathway. Neisseria spp. is the only bacterial pathogen that
has so far exhibited this type of fusion between the two enzymes. Interestingly, several species
from Neisseriathat biosynthesize diNAcBac do not contain a bifunctional entity and are instead
comprised of separate acetyltransferase and phosphoglycosyltransferase enzymes.
Campylobacter spp.
Neisseria spp.
PGT
PGT
ACT
AMT
DHT
ACT
AMT
DHT
Figure 1-8. The operon containing the pgl genes necessary for the production of Und-PPdiNAcBac as indicated by arrows. The phosphoglycosyltransferase (PGT) and acetyltransferase
(ACT) genes are separate in Campylobacterspp. (top) and one bifunctional gene in Neisseria
spp. (bottom). AMT, aminotransferase; DHT, dehydratase.
To gain an understanding on the possible evolutionary aspects of the fusion or separation
of these enzyme domains, a phylogenetic approach was undertaken comparing their respective
dendrograms (Figure 1-9).
For comparison purposes, the N-linked protein glycosylation
enzymes from the Campylobacter genus, which include no bifunctional enzymes, were included
in the analysis.
Interestingly, the neighbor-joining dendrograms of acetyltransferase and
phosphoglycosyltransferase enzymes have notable differences (Figure 1-9). Importantly, the
Neisseria species containing separate enzymes fall between the bifunctional Neisseria enzymes
37
and the Campylobactergenus, which are comprised of unconnected enzymes. Furthermore, the
sequence identity between C. jejuni (detached) and N. gonorrhoeae (bifunctional) enzymes are
also divergent.
The phosphoglycosyltransferase homologs are 52% identical, whereas the
acetyltransferase homologs are 34% identical. If these two domains were originally attached in
the evolutionary ancestors, it would be unusual to observe such dichotomy between the present
day bifunctional enzymes. These results would then suggest that the enzymes were previously
separate entities that evolved into a fused bifunctional protein that may have resulted in an
increase in the efficiency of the diNAcBac pathway. Interestingly, the area between the domains
in the bifunctional enzymes is completely conserved whereas the C-terminus of the
phosphoglycosyltransferase and the N-terminus of the acetyltransferase-detached enzymes
contains a great degree of variability. In three of the four cases the disconnected Neisseria spp.
enzymes can be joined together to form the consensus sequence of the bifunctional unit. Since
these enzymes are adjacent to one another, a deletion mutation could have resulted in the
removal of the stop codon and generation of a bifunctional protein. However, caution must be
taken in the interpretation of these results due to the small subset of enzymes examined. Further
work to identity additional enzymes and biochemically characterize them in the context of
diNAcBac biosynthesis is necessary.
38
A
B
0.1
0.05
Figure 1-9. Phylogenetic trees constructed with the neighbor-joining method from the
Campylobacter genus (blue), Neisseria genus containing detached enzymes (green), and
Neisseria genus composed of bifunctional enzymes (red). Comparison of the acetyltransferase
(A) and phosphoglycosyltransferase (B) enzymes found within each genus. The evolutionary
distances were computed using the Poisson correction method (105) and are in the units of the
number of amino acid substitutions per site. The scale bar indicates substitutions per site.
Evolutionary analyses were performed in MEGA 5.2 (106).
Conclusions
The ever-increasing resistance towards present-day antibiotics has resulted in the search
for novel antibacterial targets to circumvent this challenge. Bacterial protein glycosylation as it
relates to pathogenicity has recently become an increased area of focus to combat this evergrowing problem. This is mainly due to the absence of these unique biosynthetic enzymes in
eukaryotes and the decrease in pathogenicity when glycosylation is disrupted. It has been less
than 15 years since the discovery that N-linked protein glycosylation, once thought to be
exclusive to eukaryotes, also occurs in bacteria.
39
The biochemical characterization of the C.
jejuni Pgl N-linked pathway has increased our understanding of bacterial glycosylation at the
molecular level. The biosynthetic glycosylation machinery is also present as an O-linked system
in Neisseria spp. and A. baumannii, which highlights the ubiquity of the UDP-diNAcBac
pathway in pathogenic bacteria. Although each of these pathways glycosylate diverse proteins
with different glycans, the one remaining constant is the reducing-end sugar diNAcBac. This
highly-modified bacterial sugar is biosynthesized by a series of three enzymes that are
exclusively conserved in these pathogens.
The biosynthesis of diNAcBac has been shown to be
of utmost importance to the formation of the complete oligosaccharide as disruption of these
genes results in the complete loss of protein glycosylation. Importantly, the protein targets of
this pathway are known virulence factors that play a key role in pathogenicity. For example,
disrupting glycosylation of VirB 10 from the type IV secretion system in C. jejuni and PilE from
type IV pilin in Neisseria spp. led to decreases in competency and cell adherence, respectively.
What is still unknown is why bacteria utilize diNAcBac in N- and O-linked glycosylation as well
as in the O-antigen and core region of LPS. Eukaryotes use GlcNAc as the reducing-end sugar,
which is the starting substrate for the diNAcBac pathway. One possibility is that diNAcBac acts
through molecular mimicry to avoid detection in the eukaryotic host cell.
Clearly, further
investigation into understanding the precise role of diNAcBac is warranted.
The importance of diNAcBac in bacterial pathogenicity has been further validated from
the characterization of O-linked legionaminic and pseudaminic acid biosynthetic pathways.
These 9-carbon a-keto sugars are molecular mimics of sialic acid, a carbohydrate found
predominately on the exterior of mammalian cells that is essential for cell-cell communication
and adhesion.
Legionaminic acid is produced from nucleotide-activated (UDP or GDP)
diNAcBac by the actions of two enzymes that hydrolyze the nucleotide-diphosphate, invert the
40
stereochemistry at the C2 position, and condense a 3-carbon unit from PEP onto the sugar. A
final enzymatic step activates legionaminic acid through CTP, forming the CMP-sugar.
Although experimental evidence has pointed to the presence of legionaminic acid in flagellae
and the O-antigen of LPS, in vivo validation is still lacking.
Future studies are needed to
elucidate the role legionaminic acid plays in the assembly of these virulence factors and in
bacterial pathogenicity. Pseudaminic acid is produced from a series of enzymes homologous to
the legionaminic acid pathway that utilizes the diNAcBac isomer 2,4-diacetamido-2,4,6trideoxy-L-altropyranose as an intermediate. Pseudaminic acid is found on multiple sites in
flagellar proteins and disruption of the genes responsible for this sugar has adverse effects on
motility, adherence, and invasion. Further investigation into the glycome of these pathogenic
bacteria in the context of pseudaminic acid glycosylation is still necessary.
Whereas each of the diNAcBac-related pathways utilize similar starting substrates, the
question remains as to how pathogenic bacteria elicit control over these systems and under what
circumstances.
Similarly, there appears to be a high-level of cross-talk between some of the
homologous enzymes that act on different substrates responsible for pathogenicity. Is this an
evolutionary remnant that is controlled by segregation of pathways, or is this an additional
mechanism whereby bacteria can elicit a level of control based upon selective pressures in the
environment?
Lastly,
it
is
fascinating
that
bacterial
enzymes
such
as
the
phosphoglycosyltransferase and acetyltransferase are expressed as a single unit in certain
pathogens. Were the ancestral enzymes detached and later fused together through removal of the
intermediary stop codon or conversely did they start out as a bifunctional unit? Furthermore,
why did this event occur and what affect does this have on the efficiency of the overall pathway?
In conclusion, great strides have been made to understand the biosynthesis of unique, bacterial
41
sugars in the context of pathogenicity. A tremendous amount of work still remains to validate
the pathway enzymes of bacterial sugars as true antibacterial targets, however we are ever closer
to understanding the biological context of protein glycosylation in relationship to pathogenicity.
Acknowledgments
I would like to thank Dr. Angelyn Larkin, Dr. Meredith Hartley, and Austin Travis for critical
reading of this chapter.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
Varki, A. (1993) Biological roles of oligosaccharides - all of the theories are correct.
Glycobiology, 3, 97-130.
Mitra, N., Sinha, S., Ramya, T. N., and Surolia, A. (2006) N-Linked oligosaccharides as
outfitters for glycoprotein folding, form and function. Trends Biochem. Sci. 31, 156-163.
Rudd, P. M., Elliott, T., Cresswell, P., Wilson, I. A., and Dwek, R. A. (2001) Glycosylation
and the immune system. Science 291, 2370-2376.
Nothaft, H., and Szymanski, C. M. (2013) Bacterial protein N-glycosylation: new
perspectives and applications. J. Biol. Chem. 288, 6912-6920.
Scott, N. E., Parker, B. L., Connolly, A. M., Paulech, J., Edwards, A. V., Crossett, B.,
Falconer, L., Kolarich, D., Djordjevic, S. P., Hojrup, P., Packer, N. H., Larsen, M. R., and
Cordwell, S. J. (2011) Simultaneous glycanpeptide characterization using hydrophilic
interaction chromatography and parallel fragmentation by CID, higher energy collisional
dissociation, and electron transfer dissociation MS applied to the N-linked glycoproteome of
Campylobacter jejuni. Mol. Cell. Proteomics 10, M000031-MCP201.
Anonsen, J. H., Vik, A., Egge-Jacobsen, W., and Koomey, M. (2012) An extended spectrum
of target proteins and modification sites in the general 0-linked protein glycosylation system
in Neisseria gonorrhoeae. J. Proteome Res. 11, 5781-5793.
Goon, S., Kelly, J. F., Logan, S. M., Ewing, C. P., and Guerry, P. (2003) Pseudaminic acid,
the major modification on Campylobacter flagellin, is synthesized via the Cj 1293 gene. Mol.
Microbiol. 50, 659-671.
Szymanski, C. M., Yao, R., Ewing, C. P., Trust, T. J., and Guerry, P. (1999) Evidence for a
system of general protein glycosylation in Campylobacter jejuni. Mol. Microbiol. 32, 10221030.
Szymanski, C. M., Burr, D. H., and Guerry, P. (2002) Campylobacter protein glycosylation
affects host cell interactions. Infect. Immun. 70, 2242-2244.
42
10. Craig L., Pique M.E., and Tainer J.A. (2004) Type IV pilus structure and bacterial
pathogenicity. Nat. Rev. Microbiol. 2, 363-378.
11. Hartley, M.D., Morrison, M.J., Aas, F.E., Borud, B., Koomey, M., and Imperiali, B. (2011)
Biochemical characterization of the 0-linked glycosylation pathway in Neisseria
gonorrhoeae responsible for biosynthesis of protein glycans containing N,N'diacetylbacillosamine. Biochemistry 50, 4936-4948.
12. Morrison, M. J., and Imperiali, B. (2013) Biosynthesis of UDP-N,N'-diacetylbacillosamine
in Acinetobacter baumannii: biochemical characterization and correlation to existing
pathways. Arch. Biochem. Biophys. 536, 72-80.
13. Sharon, N. (2007) Celebrating the golden anniversary of the discovery of bacillosamine, the
diamino sugar of a Bacillus. Glycobiology 17, 1150-1155.
14. Sharon, N., and Jeanloz, R.W. (1959) The isolation of an aminohexose from Bacillus subtilis.
Biochim. Biochim. Acta. 31, 277-278.
15. Sharon, N., and Jeanloz, R.W. (1960) The diaminohexose component of a polysaccharide
isolated from Bacillus subtilis. J. Biol. Chem. 235, 1-5.
16. Liav, A., Hildesheim, J., Zehavi, U., and Sharon, N. (1974) Synthesis of 2-acetamido-2,6dideoxy-D-glucose (N-acetyl-D-quinovosamine), 2-acetamido-2,6-dideoxy-D-galactose (Nacetyl-D-fucosamine), and 2,4-diacetamido-2,4,6-trideoxy-D-glucose from 2-acetamido-2deoxy-D-glucose. Carbohydr. Res. 33, 217-227.
17. Weerapana, E., Glover, K.J., Chen, M.M., and Imperiali, B. (2005) Investigating bacterial Nlinked glycosylation: synthesis and glycosyl acceptor Activity of the undecaprenyl
pyrophosphate-linked bacillosamine. J. Am. Chem. Soc. 127, 13766-13767.
18. Bedini, E., Esposito, D., and Parrilli M. (2006) A versatile strategy for the synthesis of Nacetyl-bacillosamine-containing disaccharide building blocks related to bacterial O-antigens.
Synlett 6, 825-830.
19. Molinaro, A., Evidente, A., lacobellis, N.S., Lanzetta, R., Cantore, P.L., Mancino, A., and
Parrilli, M. (2002) O-Specific chain structure from the lipopolysaccharide fraction of
Pseudomonas reactans: a pathogen of the cultivated mushrooms. Carbohydr. Res. 337, 467471.
20. Chowdhury, T.A., Jansson, P.E., Lindberg, B., Lindberg, J., Gustafsson, B., and Holme, T.
(1991) Structural studies of the Vibrio cholerae 0:3 O-antigen poly-saccharide. Carbohydr.
Res. 215, 303-314.
21. Vinogradov, E. and Perry, M.B. (2004) Characterisation of the core part of the
lipopolysaccharide O-antigen of Francisella novicida (Ul 12). Carbohydr. Res. 339, 16431648.
22. Zubkov, V.A., Nazarenko, E.L., Gorshkova, R.P., Ivanova, E.P., Shashkov, A.S., Knirel,
Y.A., Paramonov, N.A., and Ovodov, Y.S. (1995) Structure of the capsular polysaccharide
from Alteromonas sp. CMM 155. Carbohydr. Res. 275, 147-154.
23. Sharon, N., Shiff, I., and Zehavi, U. (1964) The isolation of D-fucosamine (2-amino-2,6dideoxy-D-galactose) from polysaccharides of Bacillus spp. Biochem. J. 93, 210-214.
24. Parkhill, J., Wren, B.W., Mungall, K., Ketley, J.M., Churcher, C., Basham, D.,
Chillingworth, T., Davies, R.M., Feltwell, T., Holroyd, S., Jagels, K., Karlyshev, A.V.,
Moule, S., Pallen, M.J., Penn, C.W., Quail, M.A., Rajandream, M.A., Rutherford, K.M., van
Vliet, C.W., Whitehead, S., and Barrell, B.G. (2000) The genome sequence of the food-borne
pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665-668.
43
25. Szymanski, C.M., Yao ,R., Ewing, C.P., Turst, T.J., and Guerry, P. (1999) Evidence for a
system of general protein glycosylation in Campylobacter jejuni. Mol. Microbiol. 32, 10221030.
26. Schoenhofen, I.C., McNally, D.J., Vinogradov, E., Whitfield, D., Young, N.M., Dick, S.,
Wakarchuk, W.W., Brisson, J.R., and Logan, S.M. (2005) Functional characterization of
dehydratase/aminotransferase pairs from Helicobacter and Campylobacter. J. Biol. Chem.
281, 723-732.
27. Olivier, N.B., Chen, M.M., Behr, J.R., and Imperiali, B. (2006) In vitro biosynthesis of UDPN,N'-diacetylbacillosamine by enzymes of the Campylobacter jejuni general protein
glycosylation system. Biochemistry 45, 13659-13669.
28. Vijaykumar, S., Merkx-Jacques, A., Ratnayake, D.B., Gryski, I., Obhi, R.K., Houle, S.,
Dozois, C.M., and Creuzenet, C. (2006) Cj 1121c, a novel UDP-4-keto-6-deoxy-GlcNAc C-4
aminotransferase essential for protein glycosylation and virulence in Campylobacter jejuni. J.
Biol. Chem. 281, 27733-27743.
29. Bennett, B.D., Kimball, E.H., Gao, M., Osterhout, R., Van Dien, S.J., and Rabinowitz, J.D.
(2009) Absolute metabolite concentrations and implied enzyme active site occupancy in
Escherichia coli. Nat. Chem. Biol. 5, 593-599.
30. Schoenhofen, I.C., Lunin, V.V., Julien, J.P., Li Y., Ajamian, E., Matte, A., Cygler, M.,
Brisson, J.R., Aubry, A., Logan, S.M., Bhatia, S., Wakarchuk, W.W., and Young, N.M.
(2006) Structural and functional characterization of PseC, an aminotransferase involved in
the biosynthesis of pseudaminic acid, an essential flagellar modification in Helicobacter
pylori. J. Biol. Chem. 281, 8907-8916.
31. Larkin, A., Olivier, N.B., and Imperiali, B. (2010) Structural analysis of WbpE from
Pseudomonas aeruginosa PAOl: a nucleotide sugar aminotransferase involved in O-antigen
assembly. Biochemistry 49, 7227-7237.
32. Badger, J., Sauder, J.M., Adams, J.M., Antonysamy, S., Bain, K., Bergseid, M.G., Buchanan,
S.G., Buchanan, M.D., Batiyenko, Y., Christopher, J.A., Emtage, S., Eroshkina, A., Feil, I.,
Furlong, E.B., Gajiwala, K.S., Gao, X., He, D., Hendle, J., Huber, A., Hoda, K., Kearins, P.,
Kissinger, C., Laubert, B., Lewis, H.A., Lin, J., Loomis, K., Lorimer, D., Louie, G., Maletic,
M., Marsh, C.D., Miller, I., Molinari, J., Muller-Dieckmann, H.J., Newman, J.M., Noland,
B.W., Pagarigan, B., Park, F., Peat, T.S., Post, K.W., Radojicic, S., Ramos, A., Romero, R.,
Rutter, M.E., Sanderson, W.E., Schwinn, K.D., Tresser, J., Winhoven, J., Wright, T.A., Wu,
L., Xu, J. and Harris, T.J.R. (2005) Structural analysis of a set of proteins resulting from a
bacterial genomics project. Proteins, 60, 787-796.
33. Demendi, M. and Creuzenet, C. (2009) Cj 1 123c (PglD), a multifaceted acetyltransferase
from Campylobacterjejuni. Biochem. Cell Biol. 87, 469-483.
34. Olivier, N.B. and Imperiali, B. (2008) Crystal structure and catalytic mechanism of PglD
from Campylobacterjejuni. J. Biol. Chem. 283, 27937-27946.
35. Rangarajan, E.S., Ruane, K.M., Sulea, T., Watson, D.C., Proteau, A., Leclerc, S., Cygler, M.,
Matte, A., and Young, N.M. (2007) Structure and active site residues of PglD, an Nacetyltransferase from the bacillosamine synthetic pathway required for N-glycan synthesis
in Campylobacter jejuni. Biochemistry 47, 1827-1836.
36. Clatworthy, A.E., Pierson, E., and Hung, D.T. (2007) Targeting virulence: a new paradigm
for antimicrobial therapy. Nat. Chem. Biol. 3, 541-548.
37. Hughes, R. (2004) Campylobacterjejuni in Guillain-Barre syndrome. Lancet Neurol. 3, 644.
44
38. Komagamine, T., and Yuki, N. (2006) Ganglioside mimicry as a cause of Guillain-Barre
syndrome. CNS Neurol. Disord. Drug Targets 5, 391-400.
39. Yu, R. K., Usuki, S., and Ariga, T. (2006) Ganglioside molecular mimicry and its
pathological roles in Guillain-Barre syndrome and related diseases. Infect. Immun. 74,
6517-6527.
40. Nothaft, H., and Szymanski, C.M. (2010) Protein glycosylation in bacteria: sweeter than
ever. Nat. Rev. Microbiol. 8, 765-778.
41. Luangtongkum, T., Jeon, B., Han, J., Plummer, P., Logue, C.M., and Zhang, Q. (2009)
Antibiotic resistance in Campylobacter: emergence, transmission and persistence. Future
Microbiol. 4, 189-200.
42. Kelly, J., Jarrell, H., Millar, L., Tessier, L., Fiori, L.M., Lau, P.C., Allan, B., and Szymanski,
C.M. (2006) Biosynthesis of the N-Linked glycan in Campylobacter jejuni and addition onto
protein through block transfer. J. Bacteriol. 188, 2427-2434.
43. Hendrixson, D.R., and DiRita, V.J. (2004) Identification of Campylobacter jejuni genes
involved in commensal colonization of the chick gastrointestinal tract. Mol. Microbiol. 52,
471-484.
44. Szymanski, C.M., Burr, D.H., and Guerry, P. (2001) Campylobacter protein glycosylation
affects host cell interactions. Infect. Immun. 70, 2242-2244.
45. Larsen, J.C., Szymanski, C.M., and Guerry, P. (2004) N-Linked protein glycosylation Is
required for full competence in Campylobacter jejuni 81-176. J. Bacteriol. 186, 6508-6514.
46. Elmi, A., Watson, E., Sandu, P., Gundogdu, 0., Mills, D.C., Inglis, N.F., Manson, E., Imrie,
L., Bajaj-Elliott, M., Wren, B.W., Smith, D.G.E., and Dorrell, N. (2012) Campylobacter
jejuni outer membrane vesicles play an important role in bacterial interactions with human
intestinal epithelial cells. Infect. Immun. 80, 4089-4098.
47. Aas, F.E., Egge-Jacobsen, W., Winther-Larsen, H.C., Lovold, C., Hitchen, P.G., Dell, A.,
and Koomey, M. (2006) Neisseria gonorrhoeae Type IV pili undergo multisite, hierarchical
modifications with phosphoethanolamine and phosphocholine requiring an enzyme
structurally related to lipopolysaccharide phosphoethanolamine transferases. J. Biol. Chem.
281, 27712-27723.
48. Borud, B., Aas, F.E., Vik, A., Winther-Larsen, H.C., Egge-Jacobson, W., and Koomey, M.
(2010) Genetic, structural, and antigenic analyses of glycan diversity in the O-linked protein
glycosylation systems of human Neisseria species. J. Bacteriol. 192, 2816-2829.
49. Vik, A., Aas, F.E., Anonsen, J.H., Biosborough, S., Schneider, A., Edde-Jacobsen, W., and
Koomey, M. (2009) Broad spectrum O-linked protein glycosylation in the human pathogen
Neisseria gonorrhoeae. Proc. Natl. Acad. Sci. USA 106, 4447-4452.
50. Jennings, M.P., Jen, F.E.-C., Roddam, L.F., Apicella, M.A., and Edwards, J.L. (2011)
Neisseria gonorrhoeae pilin glycan contributes to CR3 activation during challenge of primary
cervical epithelial cells. Cell. Microbiol. 13, 885-896.
51. Stimson, E., Virji, M., Makepeace, K., Dell, A., Morris, H.R., Payne, G., Saunders, J.R.,
Jennings, M.P., Barker, S., Panico, M., Blench, I., and Moxon, E.R. (1995) Meningococcal
pilin: a glycoprotein substituted with digalactosyl 2,4-diacetamido-2,4,6-trideoxyhexose.
Mol. Microbiol. 17, 1201-1214.
52. Jennings, M.P., Virji, M., Evans, D., Foster, V., Srikhanta, Y.N., Steeghs, L., van der Ley, P.,
and Moxon, E.R. (1998) Identification of a novel gene involved in pilin glycosylation in
Neisseria meningitidis. Mol. Microbiol. 29, 975-984.
45
53. Power, P.M., Roddam, L.F., Rutter, K., Fitzpatrick, S.Z., Srikhanta, Y.N., and Jennings, M.P.
(2003) Genetic characterization of pilin glycosylation and phase variation in Neisseria
meningitidis. Mol. Microbiol. 49, 833-847.
54. Jen, F.E.-C., Warren, M.J., Schulz, B.L., Power, P.M., Swords, W.E., Weiser, J.N., Apicella,
M.A., Edwards, J.L., and Jennings, M.P. (2013) Dual pili post-translational modifications
synergize to mediate meningococcal adherence to platelet activating factor receptor on
human airway cells. PLoS Pathog. 9, e1003377.
55. Schauer R. (2009) Sialic acids as regulators of molecular and cellular interactions. Curr.
Opin. Struct. Biol. 19, 507-514.
56. Tsvetkov Y.E., Shashkov A.S., Knirel Y.A., and Zahringer U. (2001) Synthesis and
identification in bacterial lipopolysaccharides of 5,7-diacetamido-3,5,7,9-tetradeoxy-Dglycero-D-galacto- and -D-glycero-D-talo-non-2-ulosonic acids. Carbohydr. Res. 331, 233237.
57. Knirel Y.A., Rietschel E.T., Marre R., and Zahringer U. (1994) The structure of the 0specific chain of Legionella pneumophila serogroup 1 lipopolysaccharide. Eur. J. Biochem.
221, 239-245.
58. McNally D.J., Aubry A.J., Hui J.P., Khieu N.H., Whitfield D., Ewing C.P., Guerry P.,
Brisson J.R., Logan S.M., and Soo E.C. (2007) Targeted metabolomics analysis of
Campylobacter coli VC167 reveals legionaminic acid derivatives as novel flagellar glycans.
J. Biol. Chem. 282, 14463-14475.
59. Haseley, S. R. and Wilkinson, S. G. (1997) Structural studies of the putative O-specific
polysaccharide of Acinetobacter baumannii 024 containing 5,7-diamino-3,5,7,9-tetradeoxyL-glycero-D-galactononulosonic acid. Eur. J. Biochem. 250, 617-623.
60. Hu, D., Liu, B., Dijkshoom, L., Wang, L., and Reeves, P.R. (2013) Diversity in the major
polysaccharide antigen of Acinetobacter baumannii Assessed by DNA sequencing, and
Development of a molecular serotyping scheme. PLoS One 8, e70329.
61. MacLean, L.L., Vinogradov, E., Pagotto, F., and Perry, M.B. (2011) Characterization of the
lipopolysaccharide O-antigen of Cronobacter turicensis HPB3287 as a polysaccharide
containing a 5,7-diacetamido-3,5,7,9-tetradeoxy-D-glycero-D-galacto-non-2-ulosonic acid
(legionaminic acid) residue. Carbohydr. Res. 346, 2589-2594.
62. Li, X., Perepelov, A.V., Wang, Q., Senchenkova, S.N., Liu, B., Shevelev, S.D., Guo, X.,
Shashkov, A.S., Chen, W., Wang, L., and Knirel, Y.A. (2010) Structural and genetic
characterization of the O-antigen of Escherichia coliO 161 containing a derivative of a higher
acidic diamino sugar, legionaminic acid. Carbohydr. Res. 345, 1581-1587.
63. Lewis, A.L., Desa, N., Hansen, E.E., Knirel, Y.A., Gordon, J.I., Gagneux, P., Nizet, V., and
Varki, A. (2009) Innovations in host and microbial sialic acid biosynthesis revealed by
phylogenomic prediction of nonulosonic acid structure. Proc. Natl. Acad. Sci. USA 106,
13552-13557.
64. Post, D.M.B., Yu, L., Krasity, B.C., Choudhury, B., Mandel, M.J., Brennan, C.A., Ruby,
E.G., McFall-Ngai, M.J., Gibson, B.W., and Apicella, M.A. (2012) O-antigen and core
carbohydrate of Vibrio fischeri lipopolysaccharide: composition and analysis of their role in
euprymna scolopes light organ colonization. J. Biol. Chem. 287, 8515-8530.
65. Glaze, P.A., Watson, D.C., Young, N.M., and Tanner, M.E. (2008) Biosynthesis of CMPN,N'-diacetyllegionaminic acid from UDP-N,N'-diacetylbacillosamine in Legionella
pneumophila. Biochemistry 47, 3272-3282.
46
66. Lfineberg, E., Zetzmann, N., Alber, D., Knirel, Y. A., Kooistra, 0., Zihringer, U., and
Frosch, M. (2000) Cloning and functional characterization of a 30 kb gene locus required for
lipopolysaccharide biosynthesis in Legionella pneumophila. Int. J. Med. Microbiol. 290, 3739.
67. Murkin, A. S., Chou, W. K., Wakarchuk, W. W., and Tanner, M. E. (2004) Identification and
mechanism of a bacterial hydrolyzing UDP-N-acetylglucosamine 2-epimerase. Biochemistry
43, 14290-14298.
68. Chou, W. K., Hinderlich, S., Reutter, W., and Tanner, M. E. (2003) Sialic acid biosynthesis:
Stereochemistry and mechanism of the reaction catalyzed by the mammalian UDP-Nacetylglucosamine 2-epimerase. J. Am. Chem. Soc. 125, 2455-2461.
69. Schoenhofen, I.C., Vinogradov, E., Whitfield, D.M., Brisson, J.R., and Logan, S.M. (2009)
The CMP-legionaminic acid pathway in Campylobacter: biosynthesis involving novel GDPlinked precursors. Glycobiology 19, 715-725.
70. Mengin-Lecreulx, D., and van Heijenoort, J. (1994) Copurification of glucosamine-1phosphate acetyltransferase and N-acetylglucosamine-1-phosphate
uridyltransferase
activities of Escherichia coli: Characterization of the glmU gene product as a bifunctional
enzyme catalyzing two subsequent steps in the pathway for UDP-N-acetylglucosamine
synthesis. J Bacteriol. 176, 5788-5795.
71. Thibault, P., Logan, S.M., Kelly, J.F., Brisson, J.R., Ewing, C.P., Trust, T.J., and Guerry, P.
(2001) Identification of the carbohydrate moieties and glycosylation motifs in Campylobacter
jejuni flagellin. J. Biol. Chem. 276, 34862-34870.
72. Schirm, M., Soo, E.C., Aubry, A.J., Austin, J., Thibault, P., Logan, S.M. (2003) Structural,
genetic and functional characterization of the flagellin glycosylation process in Helicobacter
pylori. Mol. Microbiol. 48, 1579-1592.
73. Guerry, P., Ewing, C.P., Schirm, M., Lorenzo, M., Kelly, J.F., Pattarini, D., Majam, G.,
Thibault, P., Logan, S.M. (2006) Changes in flagellin glycosylation affect Campylobacter
autoagglutination and virulence. Mol. Microbiol. 60, 299-311.
74. Ishiyama, N., Creuzenet, C., Miller, W.L., Demendi, M., Anderson, E.M., Harauz, G., Lam,
J.S., and Berghuis, A.M. (2006) Structural studies of FlaA1 from Helicobacter pylori reveal
the mechanism for inverting 4,6-dehydratase activity. J. Biol. Chem. 281, 24489-24495.
75. Morrison, J.P., Schoenhofen, I.C., and Tanner, M.E. (2008) Mechanistic studies on PseB of
pseudaminic acid biosynthesis: a UDP-N-acetylglucosamine 5-inverting 4,6-dehydratase.
Bioorg. Chem. 36, 312-320.
76. Creuzenet, C. (2004) Characterization of CJ1293, a new UDP-GlcNAc C6 dehydratase from
Campylobacter jejuni. FEBS Lett. 559, 136-140.
77. McNally, D.J., Schoenhofen, I.C., Houliston, R.S., Khieu, N.H., Whitfield, D.M., Logan,
S.M., Jarrell, H.C., and Brisson, J.R. (2008) CMP-pseudaminic acid is a natural potent
inhibitor of PseB, the first Enzyme of the pseudaminic acid pathway in Campylobacter jejuni
and Helicobacter pylori. ChemMedChem 3, 55-59.
78. Obhi, R.K. and Creuzenet, C. (2005) Biochemical characterization of the Campylobacter
jejuni Cj1294, a novel UDP-4-keto-6-deoxy-GlcNAc aminotransferase that generates UDP4-amino-4,6-dideoxy-GalNAc. J. BioL. Chem. 280, 20902-20908.
79. Schoenhofen, I.C., McNally, D.J., Brisson, J.R., and Logan, S.M. (2006) Elucidation of the
CMP-pseudaminic acid pathway in Helicobacter pylori: synthesis from UDP-Nacetylglucosamine by a single enzymatic reaction. Glycobiology 16, 8C-14C.
47
80. Liu, F. and Tanner M.E. (2006) PseG of pseudaminic acid biosynthesis: a UDP-sugar
hydrolase as a masked glycosyltransferase. J. Biol. Chem. 281, 20902-20909.
81. Rangarajan, E.S., Proteau, A., Cui, Q., Logan, S.M., Potetinova, Z., Whitfield, D., Purisima,
E.O., Cygler, M., Matte, A., Sulea, T., and Schoenhofen, I.C. (2009) Structural and
functional analysis of Campylobacter jejuni PseG: a UDP-sugar hydrolase from the
pseudaminic acid biosynthetic pathway. J. Biol. Chem. 284, 20989-21000.
82. Chou, W.K., Dick, S., Wakarchuk, W.W., and Tanner M.E. (2005) Identification and
characterization of NeuB3 from Campylobacter jejuni as a pseudaminic acid synthase. J.
Biol. Chem. 280, 35922-35928.
83. McNally, D.J., Hui, J.P.M., Aubry, A.J., Mui, K.K.K., Guerry, P., Brisson, J.R., Logan,
S.M., and Soo, E.C. (2006) Functional characterization of the flagellar glycosylation locus in
Campylobacter jejuni 81-176 using a focused metabolomics approach. J. Biol. Chem. 281,
18489-18498.
84. Lee, Y.J., Kubota, A., Ishiwata, A., and Ito, Y. (2011) Synthesis of pseudaminic acid, a
unique nonulopyranoside derived from pathogenic bacteria through 6-deoxy-AltdiNAc.
Tetrahedron Lett. 52, 418-421.
85. Carter, A.T., Paul, C.J., Mason, D.R., Twine, S.M., Alston, M., Logan, S.M., Austin, J.W.,
and Peck, M.W. (2009) Independent evolution of neurotoxin and flagellar genetic loci in
proteolytic Clostridium botulinum. BMC Genomics 10, 115.
86. Parker, J.L., Day-Williams, M.J., Tomas, J.M., Stafford, G.P., and Shaw, J.G. (2012)
Identification of a putative glycosyltransferase responsible for the transfer of pseudaminic
acid onto the polar flagellin of Aeromonas caviae Sch3N. Microbiologyopen 1, 149-160.
87. Kaakoush, N.O., Deshpande, N.P., Wilkins, M.R., Raftery, M.J., Janitz, K., and Mitchell, H.
(2011) Comparative analyses of Campylobacter concisus strains reveal the genome of the
reference strain BAA-1457 is not representative of the species. Gut Pathog. 3, 15.
88. Kandiba, L., and Eichler, J. (2013) Analysis of putative nonulosonic acid biosynthesis
pathways in Archaea reveals a complex evolutionary history. FEMS Microbiol. Lett. 345,
110-120.
89. Vinogradov, E., Frimmelova, M., and Toman, R. Chemical structure of the carbohydrate
backbone of the lipopolysaccharide from Piscirickettsia salmonis. Carbohydr. Res. 378, 108113.
90. Logan, S.M., Kelly, J.F., Thibault, P., Ewing, C.P., and Guerry, P. (2002) Structural
heterogeneity of carbohydrate modifications affects serospecificity of Campylobacter
flagellins. Mol. Microbiol. 46, 587-597.
91. Josenhans, C., Ferrero, R., Labigne, A., and Suerbaum, S. (1999) Cloning and allelic
exchange mutagenesis of two flagellin genes of Helicobacter felis. Mol. Microbiol. 33, 350362.
92. Ewing, C.P., Ekaterina, A., and Guerry, P. (2009) Functional characterization of flagellin
glycosylation in Campylobacterjejuni 81-176. J. Bacteriol. 191, 7086-7093.
93. Hitchen, P., Brzostek, J., Panico, M., Butler, J.A., Morris, H.R., Dell, A. and Linton D.
(2010) Modification of the Campylobacter jejuni flagellin glycan by the product of the
Cj 1295 homopolymeric-tract-containing gene. Microbiology 156, 1953-1962.
94. Zampronio, C.G., Blackwell, G., Penn, C.W., and Cooper, H.J. (2011) Novel glycosylation
sites localized in Campylobacter jejuni flagellin FlaA by liquid chromatography electron
capture dissociation tandem mass spectrometry. J. Proteome Res. 10, 1238-1245.
48
95. Johannessen, C. Koomey, M., and Borud, B. (2012) Hypomorphic glycosyltransferase alleles
and recoding at contingency loci influence glycan microheterogeneity in the protein
glycosylation system of Neisseria species. J. Bacteriol. 194, 5034-5043.
96. Borud, B, Aas, F.E., Vik, A., Winther-Larsen, H.C., Egge-Jacobsen, W., and Koomey, M.
(2010) Genetic, structural, and antigenic analyses of glycan diversity in the O-linked protein
glycosylation systems of human Neisseria species. J. Bacteriol. 192, 2816-2829.
97. Kostrzynska, M., Betts, J.D., Austin, J.W., and Trust, T.J. (1991) Identification,
characterization and spatial localization of two flagellin species in Helicobacter pylori
flagella. J. Bacteriol. 173, 937-946.
98. Suerbaum, S., Josenhans, C., and Labigne, A. (1993) Cloning and genetic characterization of
the Helicobacter pylori and Helicobacter mustelae flaB flagellin genes and construction of H.
pylori flaA- and flaB-negative mutants by electroporation-mediated allelic exchange. J.
Bacteriol. 175, 3278-3288.
99. Eaton, K.A., Suerbaum, S., Josenhans, C., and Krakowka, S. (1996) Colonization of
gnotobiotic piglets by Helicobacter pylori deficient in two flagellin genes. Infect. Immun. 64,
2445-2448.
100. Ottemann, K.M., and Lowenthal, A.C. (2002) Helicobacter pylori uses motility for initial
colonization and to attain robust infection. Infect. Immun. 70, 1984-1990.
101. Asakura, H., Churin, Y., Bauer, B., Boettcher, J.P., Bartfield, S., Hashii, N., Kawasaki, N.,
Mollenkopf, H.J., Jungblut, P.R., Brinkmann, V., and Meyer, T.F. (2010) Helicobacter
pylori HP0518 affects flagellin glycosylation to alter bacterial motility. Mol. Microbiol. 78,
1130-1144.
102. Champasa, K., Longwell, S.A., Eldridge, A.M., Stemmler, E.A., and Dube, D.H. (2013)
Targeted identification of glycosylated proteins in the gastric pathogen Helicobacter pylori
(Hp). Mol. Cell. Proteomics 12, 2568-2586.
103. Glover, K.J., Weerapana, E., Chen, M.M., and Imperiali, B. (2006) Direct biochemical
evidence for the utilization of UDP-bacillosamine by PglC, an essential glycosyl-1phosphate transferase in the Campylobacter jejuni N-linked glycosylation pathway.
Biochemistry 45, 5343-5350.
104. Power, P.M., and Jennings, M.P. (2003) The genetics of glycosylation in Gram-negative
bacteria. FEMS Microbiol. Lett. 218, 211-222.
105. Zuckerkandl, E. and Pauling, L. (1965) Evolutionary divergence and convergence in
proteins. Edited in Evolving Genes and Proteins by V. Bryson and H.J. Vogel, pp. 97-166.
Academic Press, New York.
106. Tamura K., Peterson D., Peterson N., Stecher G., Nei M., and Kumar S. (2011) MEGA5:
Molecular Evolutionary Genetics Analysis using Maximum Likelihood, Evolutionary
Distance, and Maximum Parsimony Methods. Molecular Biology and Evolution 28, 27312739.
49
Chapter 2.
Biochemical Characterization of the O-Linked Glycosylation
Pathway in Neisseria gonorrhoeae Responsible for Biosynthesis of Protein
Glycans Containing N,N'-Diacetylbacillosamine
A significant portion of this chapter has been published in the following reference:
Hartley, M. D., Morrison, M. J., Aas, F. E., Borud, B., Koomey, M., and Imperiali, B. (2011)
Biochemical characterization of the O-linked glycosylation pathway in Neisseria gonorrhoeae
responsible for biosynthesis
of protein glycans
containing N,N -diacetylbacillosamine.
Biochemistry 50, 4936-4948.
Dr. Meredith Hartley purified and biochemically characterized PglA, PglE, PglH, and PglO from
this work.
50
Introduction
In Neisseriagonorrhoeae, individual pilin subunits rapidly assemble and disassemble to
form the flagellar-like Type IV pili, which mediate essential interactions with host cells and
affect many aspects of pathogenicity including surface motility, bacteria-host communication,
cell signaling, bacterial dissemination and biofilm formation (1-4).
Recently, the gonococcal
pilin glycosylation system was shown to be a general O-linked system in which many
structurally distinct periplasmic proteins undergo glycosylation (5).
Glycan-modifications on
pili, flagella and other extracellular proteins have been implicated in bacterial pathogenicity,
which has led to increased interest in bacterial glycosylation pathways as potential antibacterial
targets (2,6-10). The focus of this study is the protein glycosylation (pgl) locus identified in N.
gonorrhoeae, which is responsible for glycan addition to distinct serine residues (11-12). The
protein glycan modifications present in N. gonorrhoeaepgl-null strains have been analyzed by
top-down mass spectrometry (MS) (5,11) and the following model of the protein glycosylation
pathway has been developed (Figure 2-1). The core pgl locus contains four genes, three of
which (pglD, pgC, and pgB) are required for the synthesis of undecaprenyl diphosphate 2,4diacetamido-2,4,6-trideoxyhexose (Und-PP-DATDH) (13).
PglD and PglC perform NADH-
dependent dehydratase and aminotransferase reactions, respectively, to convert UDP-HexNAc
(UDP-GlcNAc or UDP-GalNAc) to UDP-2-acetamido-4-amino-2,4,6-trideoxyhexose (UDP-4amino). PglB is a bifunctional enzyme, which catalyzes the amino acetylation of UDP-4-amino
to form UDP-DATDH and the transfer of phospho-DATDH to undecaprenyl phosphate (Und-P).
The fourth gene, pglF, shares homology with ABC transporter-type flippases and is putatively
involved in the translocation of the undecaprenyl-linked glycan across the periplasmic
51
membrane. Although the function of this gene has not been demonstrated, the pglF-null strain
does exhibit diminished glycosylation (11-12).
UDP-GIcNAc
OH 0Pg1D
Hq
UDP-4-keto
H
UDP-4-amino
0
PgC
/- ql;r/-7
NAD+ NADH HO
0
AcH
PLP
ACHN OUDP
L-GIu
Ouop
Und-PP-dlNAcBac-(Ga
PMP
0
O
H2N
HO
UDP-dINAcBac
PgB-ATD
ACHN-/- ';
HO
AcCoA CoASH
O
\
AcHN OUDP
AcHN OUDP
a-KG
2
OHOH
OH
OH
HO
PgIE
PgIA
0
0
UDP UDP-GaI
Und-PP-dIMAcBac
UDP UDP-Gal
HO
OH
AcHN A 0
AHN'
0
Cytoplasm
PgIB-GTD
Und-PP-dIMAcBac-Gal
OHOH
0
HO
OH
OCH
AcHN
0
0
0
AcH
AcH
0
PgIF?
I
Periplasm
C
PgIO
NHAc
.
pilin
HO
0A.OOH
OHH~
O
OH
HO
HHHO
HO O
OH OH
OH
H
O
HO OH
O
H
NHAc
Pilin O-linked with diNAcBac-(Gei 2
Figure 2-1. Biosynthetic pathway of the pilin glycan in Neisseriagonorrhoeae.
Interestingly, the other genes involved in pilin glycosylation are not linked to the core pgl
locus. The products of the pglA and pglE genes further elaborate the polyprenyl-linked DATDH
with the transfer of two sequential galactose units (Figure 2-1) (11). The pglA and pglE genes
undergo phase variation in which the genes are alternately turned on and off. Phase-variant pglA
alleles have been proposed to be associated with more virulent strains of N. gonorrhoeae,
although these studies have been disputed (7-8,11). In addition to PglA and PglE, an alternate
52
glycosyltransferase PglH adds a Glc unit instead of Gal to Und-PP-DATDH (Figure 2-2) (14).
Finally, a gene has been identified, pglO, which shares homology with the 0-antigen ligase
(WaaL) family and was required for formation of the protein-glycan linkage (Figure 2-1) (5).
Neisseria
UDP-Q
Campylobacterjejun!
2
UDP-a
CL
a.
gI
PgIE
PglD
C
0..
'0.91
UDPE]
0.
UDP4I
-N-T-S-A-G-
Neisseria (alternative)
UDP-
PgID PgIC
-
0.
*L
Gal
UDP-aZ
4.,-
gI-
-N-T
A0.
0.N~
""P
UdP
0
,___________________________
T
gIO0
Figure 2-2. Schematic representations of bacterial protein glycosylation pathways: (top, left) 0linked pathway in Neisseria gonorrhoeae and Neisseria meningifidis; (bottom, left) alternative
O-linked pathway in Neisseria species; (right, top) N-linked pathway in Campylobacterjejuni.
Recent studies have focused on the highly homologous pilin glycosylation pathway in the
related species Neisseria meningitidis. The proposed model of the N. meningitidis pathway is
similar to the N. gonorrhoeae pathway (Figure 2-2) and was also developed by bioinformatic
analysis and experiments with pgl-null strains (15-20).
The oligosaccharyltransferase from N
meningitidis, PglL(Nm), has been purified to homogeneity and shown to glycosylate pilin using a
53
farnesyl substrate analog (21). In addition, upon heterologous expression in E. coli, PglL(Nm)
was shown to transfer a variety of complex glycan substrates to pilin (21,22).
The Pgl pathways from N. gonorrhoeaeand N. meningitidisrepresent the first example of
O-linked protein glycosylation from polyprenyl-linked intermediates; all other identified 0linked pathways glycosylate protein substrates by sequential transfer of individual saccharide
units from nucleotide or polyprenyl-phosphate activated glycan donors. Another intriguing facet
of the O-linked pilin pathway is that the first three enzymes (PglD, C, and B) share homology
with the first four enzymes in the N-linked protein glycosylation (also designated Pgl) pathway
in Campylobacter jejuni (23) with the exception that the C. jejuni locus encodes separate
enzymes for the sequential acetyltransferase and phospho-glycosyltransferase reactions. Both
the N. gonorrhoeaeand C. jejuni pathways produce an initial Und-PP-DATDH intermediate, but
this intermediate is elaborated in distinct ways (Figure 2-2).
The N. gonorrhoeae pathway
produces a serine-linked mono-, di- or trisaccharide (13) and the C. jejuni pathway generates an
asparagine-linked heptasaccharide (24).
The C. jejuni glycosylation pathway serves as an
important model for the N. gonorrhoeaesystem and previous work has resulted in the complete
biochemical characterization of the C. jejuni Pgl pathway enzymes except for the flippase (PglK)
(25-28).
In this chapter, the biochemical functions of the proteins PglD, C, B, A, E, and 0 from N.
gonorrhoeae are characterized for the first time through in vitro biochemical analysis.
Importantly, the previously undefined stereochemical assignment of the UDP-DATDH produced
by PglD, C, and B is unequivocally shown to be UDP-2,4-diacetylbacillosamine, which is also
the identity of the first sugar added in the C. jejuni N-linked glycosylation pathway. In vitro
assays demonstrate that the phospho-glycosyltransferase (PglB) and two glycosyltransferases
54
(PglA and PglE) build the glycan on an undecaprenyl-diphosphate linker prior to en bloc transfer
to protein and that these enzymes display strict specificity for the UDP-saccharide donor.
Finally, glycan substrate specificity analyses suggest that the O-linked OTase is highly selective
for native N. gonorrhoeaeglycan substrates in vitro.
Results and Discussion
Determinationof UDP-DA TDH Stereochemistry by NMR
The biosynthesis of UDP-DATDH from UDP-GlcNAc was carried out in the presence of
purified
dehydratase
(PglD),
aminotransferase
phosphoglycosyltransferase (PglB).
(PglC),
and
acetyltransferase-
PglC, a soluble protein, was purified to homogeneity
(Figure 2-3, lane 2).
kDa
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
98
62
49
~
38
28
14
Figure 2-3. SDS-PAGE (left) and anti-His 4 Western blot (right) analysis of nine Pgl proteins
purified for this study. MBP-PglH (lane 8) does not contain a His6 tag. The identity of PglH was
verified through anti-MBP Western blot analysis (data not shown). The SeeBlue@&Plus2 Prestained Standard (Invitrogen) was used as a standard in the first lane. Lane 1, PglD; lane 2, PglC;
lane 3, PglB-ATD; lane 4, PglB CEF; lane 5, PglA; lane 6, PglE CEF; lane 7, PglO;
lane 8,
MBP-PglH; lane 9, pilin.
TMHMM, a transmembrane prediction program, (29) predicts that PglD and PglB
contain four and one transmembrane helices, respectively. SDS-PAGE analysis of purified PglD
55
demonstrated that the desired protein product is the dominant component (Figure 2-3, lane 1).
PglB was purified for this experiment, but the enzyme was used as a partially purified CEF
(Figure 2-3, lane 4) in all other assays to avoid problems with protein stability. The anti-His4
Western Blot analysis revealed that both purified PglD and PglB CEF contained His6 -tagged
truncation products that formed during protein expression (Figure 2-3, lanes 1 and 4).
The biosynthesis of UDP-diNAcBac by the action of PglD, PglC, and PglB was followed
by capillary electrophoresis to ensure complete turnover of the substrates. This method also
verified that the HexNAc substrate of PglD is UDP-GlcNAc, and not UDP-GalNAc (data not
shown).
Purification by reverse phase-HPLC removed unreacted substrates and cofactors
leading to a final UDP-DATDH purity of > 95%. To determine the final stereochemistry of the
sugar, 'H NMR (Figure 2-4) was employed to compare the chemical shifts and coupling
constants with UDP-diNAcBac from the C. jejuni pathway (Table 2-1).
The values for the
UDP-DATDH sugar from N. gonorrhoeaeexactly match the values of UDP-diNAcBac from C.
jejuni (28). Further confirmation was provided by the
31
P,
13 C,
and 'H- 1H COSY NMR spectra.
Therefore, the stereochemistry of the DATDH sugar in the N. gonorrhoeaepathway is confirmed
as diNAcBac.
56
46
AcHN
520
HO
* AcHN
OUDP
H-6
H-2
H-5
IniN
8.5
8.0
75
7.0
6.5
6.0
5.5
5.0
4.5
IIC
4.0
Ppm
3.5
3,0
2.5
20
1.5
1.0
0.5
0.0
Figure 2-4: 'H NMR spectrum of UDP-diNAcBac.
The extreme diversity of bacterial glycans is highly significant since these glycans
typically decorate the bacterial cell surface facilitating interactions with host cells (6,9) and
potentially confounding the immune response (7,30-33). Bacillosamine was originally identified
as an unusual 2,4-diamino-2,4,6-trideoxy-a-D-glucose in Bacillus subtilis (34), but it also
appears in the mono- and di-aminoacetylated
forms in a large variety of bacterial
glycoconjugates (35). It has been found frequently in the O-antigen and capsular polysaccharide
of Gram-negative bacteria, but has also been identified in the S-layer of Gram-positive bacteria
57
and as the UDP-activated donor in cellular extracts (36). DiNAcBac was initially discovered in
N-linked glycans in C. jejuni (24), but more recently, a second route to diNAcBac was
biochemically characterized in C. jejuni, in which GDP-diNAcBac is an intermediate in the
CMP-legionaminic acid biosynthetic pathway (37). The pilin oligosaccharide in N. gonorrhoeae
was thought originally to comprise Gal-a-(l,3)-GlcNAc-P-Ser63 (38). Mass spectrometry and
bioinformatic analysis suggested that the linking sugar unit was DATDH instead of GlcNAc
(11). Herein, we confirm the stereochemical assignment of this sugar for the first time showing
that the DATDH sugar in N. gonorrhoeae is diNAcBac (Table 2-1). This adds to the growing
number of oligosaccharides identified in bacteria that contain forms of bacillosamine.
Table 2-1. Comparison of C. jejuni and N. gonorrhoeaeUDP-diNAcBac IH chemical shift and
Please refer to Figure 2-4 for UDP-diNAcBac proton
coupling constant assignments.
assignment.
Moiety
C.jejuni (28)
dH (ppm)
HI
5.48 (dd)
H2
4.02 (m)
H3
3.79 (at)
H4
H5
H6
3.69 (at)
4.05 (m)
1.19 (d)
JI,2= 3.2 Hz
Jlp = 6.9 Hz
2,3 = 10.2 Hz
J3,= 10.2 Hz
J4,5 = 10.2 Hz
J5, 6= 6.2 Hz
N. gonorrhoeae
dH (ppm)
JI,2= 3.2 Hz
5.46 (dd)
Jip = 6.9 Hz
4.03 (m)
2,3 = 10.1 Hz
3.76 (at)
J3,4 = 10.2 Hz
J4,5= 10.1 Hz
3.67 (at)
4.03 (m)
1.17 (d)
J5, 6= 6.2 Hz
PglD, PglC and PglB produce Und-PP-diNAcBac in N. gonorrhoeae; these three
enzymes have functional homology to PglF(Cj), PglE(Cj), PglD(Cj), and PglC(Cj) in C. jejuni,
which produce the same polyprenyl-linked intermediate (Figure 2-2). Even though the early
enzymes in these two pathways carry out identical functions, the sequence identity is relatively
low (25-30%), except for the phospho-glycosyltransferase domain of PglB, which has 52%
58
identity with PglC(Cj) (Table 2-2).
These numbers starkly contrast the sequence identity
observed between N. gonorrhoeaeand N. meningitidis, which indicate much closer homologies
(> 84%, Table 2-2). These numbers imply that C. jejuni and N. gonorrhoeaepathways are only
distantly related from an evolutionary standpoint.
Table 2-2. Percent sequence identity (%) between N. gonorrhoeae(Ng) and C. jejuni (C) or N.
meningitidis (Nm) proteins.
Function
dehydratase
aminotransferase
acetyltransferase
P-glycosyltransferase
glycosyltransferase
glycosyltransferase
OTase
C jejuni
PgIF
PglE
PgLD
PglC
PglA
PglJ
Pg1B
%
29.8
21.2
29.7
52.3
21.3
12.0
11.8
N gonorrhoeae
Pg1D
PglC
PglB-ATD
PglB-PGTD
PglA
PglE
PglO
%
92.5
92.8
84.9
90.3
95.5
93.0
95.0
N. meningitidis
PglD
PglC
PglB-ATD
PglB-PGTD
PglA
PglE
PglL
FunctionalCharacterizationofPglB-ATD
The similarity of the N. gonorrhoeaeprotein glycosylation pathway to the pathway in C.
jejuni suggests that the acetyltransferase domain of PglB acts first on UDP-4-amino to generate
UDP-diNAcBac, which is then utilized as a substrate by the phosphoglycosyltransferase domain
of PglB (PglB-PGTD, Figure 2-1). The C-terminal acetyltransferase domain of full-length PglB
(PglB-ATD, based upon a ClustalW alignment with PglD(C)) was purified (Figure 2-3, lane 3).
This provided a suitable amount of well-behaved, soluble protein in the absence of the Nterminal phospho-glycosyltransferase domain, which is predicted by TMHMM to contain a
single transmembrane domain (29).
Functional analysis of PglB-ATD described below
confirmed definitively that this domain acetylates UDP-4-amino to produce UDP-diNAcBac,
which is a substrate for PglB-PGTD mediated transfer of P-diNAcBac to Und-P.
59
Kinetic Characterizationof PglC and PglB-A TD
Both the aminotransferase (PglC) and acetyltransferase (PglB-ATD) reactions exhibited
typical Michaelis-Menten kinetics over a wide range of substrate concentrations (Figure 2-5).
Initial velocity data were used to calculate kinetic parameters of L-glutamate and UDP-4-keto for
PglC and AcCoA and UDP-4-amino for PglB-ATD. Each reaction was run in duplicate and the
initial velocities fit to equation 1 to yield the kinetic parameters in Table 2-3.
(A)
0.01
0.016
0.014
0.008
-00.012
0,006
S0.01
0.008
0,004
. 0.006
0.004
OM2
0.002
0.002
0
0
200
400
600
800
1000
0
1200
200
[UDP-4amino] (yM)
400
600
[Ac-CoAl (pM)
800
1000
(B)
0,8
0.6
-C*
0.6
-
I.
09
0.4
0t
0.4
.8
0.2
0.2
A
-
0
200
400
600
800 1000
[UDP-4keto ([,M)
1200
0
I
I
20
I
I
40
60
[L-gtutamate] (mM)
m
I
80
Figure 2-5: Kinetic analysis of (A) PglB-ATD with UDP-4-amino and AcCoA and (B) PglC
with UDP-4-keto and L-glutamate.
60
Table 2-3. Steady-state kinetic parameters for PglC and PglB-ATD.
Enzyme
substrate
KW (lAM)
PglC
PglC
PglB-ATD
PglB-ATD
L-glutamate
UDP-4-keto
AcCoA
UDP-4-amino
4900 900
233 ± 35
456 ± 34
122 ± 17
k
kcat(s)k,/Kni (M"' S-1)p
0.025 0.001
5.1
0.039 . 0.002
167
0.928 0.032
2035
0.416 0.016
3410
Work is ongoing to understand the low level of homology between the C. jejuni and N.
gonorrhoeae diNAcBac biosynthetic enzymes. As a first step, this study describes the kinetic
parameters of both the aminotransferase (PglC) and the acetyltransferase (PglB-ATD) (Table 23).
The apparent Km of the UDP-4-k do sugar fo r PglC (2 3 3 gM) was comparable to the
PglE(Cj) homologue (48 pM (39) and 610 gM (40)) and well within the typical binding
efficiency for this type of substrate. Likewise, Km values of the UDP-4-amino substrate for the
acetyltransferases PglB-ATD (122 gM) and PglD(Cj) (410 pM (28)) lead to a similar conclusion.
However, the N. gonorrhoeaeenzymes presented here are catalytically much less efficient (kcat is
10-100 fold less for PglC and 1000-fold less for PglB-ATD) than their C. jejuni counterparts
with respect to the UDP-sugar. This observation is reflected in the differences between their
specificity constants (kcat/Km) (Table 2-3). The high acetyltransferase activity in the C. jejuni
pathway is used to drive the biosynthesis of the UDP-diNAcBac sugar (28).
A similar
phenomenon is observed in the N. gonorrhoeae pathway, with a 20-fold enhancement in kcat/Km
of PglB-ATD with respect to aminotransferase activity.
For PglB-ATD, one cannot rule out interplay between the acetyltransferase domain and
the missing C-terminal phospho-glycosyltransferase domain. Therefore, care must be taken in
interpreting the reduced N. gonorrhoeae acetyltransferase efficiency as compared to PglD(C).
Further work will be necessary to clarify how domain interactions affect kinetic parameters. In
61
addition, the low sequence homology between the C. jejuni and N. gonorrhoeaeUDP-diNAcBac
pathway enzymes (Table 2-2) could contribute to the differences in catalytic efficiency observed
here.
FunctionalCharacterizationof the Glycosyltransferases
As mentioned previously, TMHMM (29) predicts that PglB has a single N-terminal
transmembrane helix. In addition, PglE is predicted to contain two C-terminal transmembrane
helices. Purification of these proteins by detergent solubilization and extraction resulted in low
yields and loss of activity; to avoid these problems, both PglB and PglE were purified as crude
CEFs for the glycosyltransferase assays. SDS-PAGE and Western blot analysis showed that
PglB and PglE are the predominant bands present in the respective CEFs (Figure 2-3, lanes 4
and 6).
In all assays involving PglB and PglE, negative controls with CEFs lacking
overexpressed PglB or PglE showed no glycosyltransferase activity (data not shown). PglA is
predicted to be soluble and was purified to homogeneity (Figure 2-3, lane 5).
Tritium-labeled products of PglB, PglA, and PglE were analyzed by NP-HPLC. Und-PP[3H]diNAcBac, Und-PP-diNAcBac-[ 3H]Gal and Und-PP-diNAcBac-Gal-[3H]Gal were retained
on the column with retention times consistent with glycan size. Each product was analyzed
separately in order to confirm the identity of the peaks. In addition, the glycosyltransferase
products were characterized by a standard 2-AB fluorescence-labeling protocol as previously
described (41). The 2-AB labeled disaccharide and trisaccharide were purified and MALDI MS
was used to verify the masses of the products (Figure 2-6). These studies definitively annotate
the biochemical functions of PglB, PglA, and PglE as the phospho-glycosyltransferase and the
two glycosyltransferases that produce Und-PP-linked mono-, di- and trisaccharides, respectively.
62
2-AB labeled dlNAcBac-(Gal) 2
200000
i
OHOH
HO
150000
0
O OH
OH
100000
0
HO
50000
NH
AcH
AAK
S
-
-
0
HN
-AcHN
10
20
30
40
Time (minutes)
2-AB labeled dlNAcBac-(Gal) 2
mass = 691.7 g
Figure 2-6. Normal phase HPLC with fluorescence detection of 2-AB labeled glycans.
DiNAcBac-(Gal) 2 the product of PglB, PglA, and PglE elutes at 29 minutes. The product peak is
marked with an arrow. MALDI MS confirmed the identity of the separated fluorescent product.
The same experiment was performed with diNAcBac-Gal (product of PglB and PglA, data not
shown).
The C. jejuni and N. gonorrhoeae pathways diverge after the synthesis of UndPPdiNAcBac. The C. jejuni pathway continues to N-linked glycan assembly with the successive
addition of five a-(1,4)-linked GalNAc units and a branching Glc unit. However, whereas the C.
jejuni N-linked heptasaccharide is highly conserved, N. gonorrhoeae strains display high 0linked glycan diversity.
disaccharide
Strains have been identified which contain not only 0-linked
(Gal-a-(1,3)-diNAcBac) and trisaccharide (Gal-fP-(l,4)-Gal-a-(l,3)-diNAcBac)
produced by PglA and PglE, respectively, but also an alternate disaccharide (Glc-a-(1,3)diNAcBac) produced by PglH (14). Further glycan modification occurs from the addition of 0acetyl groups by PglI (11).
In addition, an alternate allele (pglB2) has been identified in N.
meningitidis that contains a domain proposed to transfer a glyceroyl moiety instead of an acetyl
group to produce 4-glyceramido-2-acetamido-2,4,6-trideoxy-a-D-hexose (GATDH) (20).
63
This
combination of biosynthetic enzymes allows Neisserialstrains to display a glycan repertoire with
at least 13 identified glycan permutations (14). Additional glycan variation can occur within a
single strain as phase variation of the genes encoding for PglA, PglE and PglH acts as another
mode of glycan regulation (11).
UDP-SaccharideSpecificity of Glycosyltransferases
The substrate specificities of PglB, PglA and PglE were explored through the use of
radioactivity-based assays. Organic extraction of the hydrophobic undecaprenyl-linked product
allowed for quantification of the amount of radiolabeled sugar transferred to the undecaprenyl
substrate similar to previously described assays (26, 42). The isoprenyl-linked substrates for the
assays (Und-PP-diNAcBac for PglA and Und-PP-diNAcBac-Gal for PglE) were produced
enzymatically and purified by NP-HPLC. The undecaprenyl phosphate required for the PglB
reaction was generated in situ from undecaprenol and ATP with S. mutans undecaprenol kinase
as previously described (43). The activities of the three enzymes were screened with UDP-Glc,
UDP-Gal, UDP-GlcNAc, UDP-GalNAc and in the case of PglB, UDP-diNAcBac (Figure 2-7).
In addition, the ability of PglB and PglA to distinguish between UDP-4-amino and UDPdiNAcBac was evaluated through a coupled assay. In all cases, the enzymes were highly
specific for the corresponding predicted sugar substrate; PglB exclusively transferred phosphodiNAcBac, whereas PglA and PglE transferred only Gal (Figure 2-7).
64
A
C 120
40
20
35
35
30
15
~25
100
20
80
15
C
20
f
15!
10
10
40
15
AL
'? 10
5
o0
0
0
50
Time (seconds)
5
20
0
0
100
2
6
4
8
10
12
Time (minutes)
B
60
30
50
25
~40
""UDP-Gic
**
UDP-GicNAc
.*." UDP-Gat
UDP-GaINAc
UDP-diNAcac
2
E
--
30
'5
20
10
10
5
0
0
0
2
4
6
Time (minutes)
Figure 2-7. Specificity analyses of Pg1B (A), the phospho-glycosyltransferase, and PglA (B)
and PglE (C), the galactosyltransferases, in the presence of a panel of UDP-sugar substrates. The
reactions were carried out in a volume of 100 pL. The assays were performed in triplicate and
the error bars indicate standard deviation.
In light of the significant amount of protein glycan variation present within strains of N.
gonorrhoeae, it is surprising that the glycosyltransferases display such strict specificity (Figure
2-7).
We have demonstrated that PglB, PglA and PglE are specific for the native substrate
(UDP-diNAcBac or UDP-Gal) and will not accept any other form of the nucleotide-activated
sugars commonly found in vivo, even though one of the alternate substrates contains only a
65
single stereochemical change (UDP-Glc vs. UDP-Gal) and another contains only an additional
acetamido group (UDP-GalNAc vs. UDP-Gal). Parallel work has shown similar strict specificity
of the fourth glycosyltransferase, PglH (14).
These results suggest that glycan identity is
regulated at the level of biosynthesis and that these enzymes have evolved to electively catalyze
reactions in the milieu of intracellular NDP-sugars.
UndecaprenylDiphosphateDisaccharideSpecificity of PglE
Facile enzymatic synthesis of Und-PP-diNAcBac-Gal and Und-PP-diNAcBac-GalNAc
produced by the Neisseria and C. jejuni pathways, respectively, allowed for examination of the
substrate specificity of PglE for the acceptor oligosaccharide. Somewhat surprisingly, PglE is
able to add a Gal residue to both the native substrate, Und-PP-diNAcBac-Gal, and to the C.
jejuni substrate Und-PP-diNAcBac-GalNAc (Figure 2-8).
This confirms in vivo studies in
which the C. jejuni PglA was expressed in N. gonorrhoeae and the resultant trisaccharide
(diNAcBac-GalNAc-Gal) was observed as a covalent pilin modification (11).
120
100
E
3
20
60
40
-
a.2
.--
a.
+'n.
20 -5
0
0
0
2
6
4
8
10
12
Time (minutes)
Figure 2-8. Determination of the polyprenyldiphosphate-linked substrate preferences of PglE in
the presence of Und-PP-diNAcBac-Glc (dashed line), Und-PP-diNAcBac-Gal (solid line), or
Und-PP-diNAcBac-GalNAc (dotted line). The reactions were carried out in a volume of 100 RL.
The assays were performed in triplicate and the error bars indicate standard deviation.
66
Interestingly, PglE transferred a Gal unit onto native Und-PP-diNAcBac-Gal, but it
showed little activity with Und-PP-diNAcBac-Glc, the alternate PglH disaccharide (Figure 2-8).
PglE has evolved to detect the stereochemical difference between Glc and Gal, which is
consistent with the model that PglA and PglE have evolved in tandem to produce a trisaccharide
that is structurally distinct from the disaccharide produced by PglH (14).
In contrast, it is
surprising that PglE would recognize the C. jejuni substrate Und-PPdiNAcBac-GalNAc (Figure
2-8), but it is consistent with the hypothesis that the glycosyltransferases exhibit specificity
relative to other substrates present in the organism. PglE may not have developed selectivity
against the additional acetamido group in the C. jejuni disaccharide, because it is not found in the
native N. gonorrhoeaeglycome.
Characterizationof PglH,an Alternative Glycosyltransferase
Recently, an alternative glycosyltransferase in N. gonorrhoeae, PglH, was identified and
shown to transfer a Glc unit to Und-PP-diNAcBac (14) (Figure 2-2). PglH was shown to be
responsible for the specific addition of Glc to Und-PP-diNAcBac by a variety of in vivo and in
vitro methods. In vitro radiolabeled assays demonstrated that PglH transfers Glc from UDP-Glc
to Und-PP-diNAcBac and does not transfer Man, Gal, GlcNAc or GalNAc; MALDI MS of the
2-AB-labeled diNAcBac-Glc confirmed the identity of the PglH product (14).
Herein, we further characterize PglH and compare its function to the other N.
gonorrhoeae glycosyltransferases (Figure 2-3, lane 8).
Analysis of the radiolabeled PglH
product, Und-PP-diNAcBac-[ 3H]Glc, by NP-HPLC revealed that the retention time (30 minutes)
was very similar to the Und-PP-diNAcBac-[ 3 H]Gal retention time (29-30 minutes). In addition,
in vivo evidence suggested that unlike the C. jejuni disaccharide, the PglH product was not
67
further modified by the third glycosyltransferase PglE (14). This result was validated by the in
vitro specificity assay described above for PglE and established that PglE was unable to transfer
Gal to Und-PP-diNAcBac-Glc (Figure 2-8).
FunctionalCharacterizationof Oligosaccharyltransferase,PglO
PglO and pilin are integral membrane proteins and were expressed in E. coli, extracted
from the CEF with Triton X-100 and purified to homogeneity (Figure 2-3, lanes 7 and 9). To
assay for OTase activity, purified PglO was incubated with pilin and radiolabeled UndPPdiNAcBac-[ 3H]Gal glycan donor. After overnight incubation, the reaction mixture was bound
to Ni-NTA resin and washed thoroughly to remove most (>99%) of the unreacted Und-PPdiNAcBac'-[3H]Gal donor. The pilin protein was then eluted with imidazole and the radioactivity
associated with the wash and elution fractions was determined by scintillation counting. Under
these assay conditions, in which pilin protein is in excess over Und-PP-diNAcBac-[ 3 H]Gal, PglO
transferred ~60% of the sugar substrate to pilin (Figure 2-9). Interestingly, unlike the OTases in
the bacterial N-linked glycosylation pathways (27, 44), PglO does not readily glycosylate a short
peptide based on the pilin glycosylation sequence.
These results were further verified via Western blot analysis utilizing a monoclonal
antibody recognizing a diNAcBac-associated epitope (Figure 2-9).
For the Western blot
analysis, the pilin glycosylation reaction was performed with equimolar amounts of protein
substrate and Und-PP-diNAcBac donor and under these conditions -13% of the pilin protein was
associated with glycan. The antibody raised against diNAcBac showed strong staining with the
glycosylated pilin and was unreactive with the unmodified pilin (45).
68
70
anti-His4
anti-diNAcBac
60
so
40
30
~20
10
0
pilin
no pilin Glycan
-
+
-
+
Figure 2-9. PglO reaction turnover after overnight incubation with Und-PP-diNAcBac-[ 3 H]Gal
in the presence of pilin or a negative control with no protein substrate (left). Western blot
analysis (right) of unmodified pilin and pilin glycosylated by PglO using His4 antibody (left blot,
positive control) and a diNAcBac-epitope recognizing monoclonal antibody termed npgl (right
blot, specific for glycan).
The C. jejuni and N. gonorrhoeaepathways culminate in transfer of the oligosaccharide
to protein; in the bacterial N-linked glycosylation pathway, PglB(Cj) transfers a heptasaccharide
en bloc to the amide side chain of asparagine residues. Herein we demonstrate that PglO acts in
a similar en bloc manner to transfer mono-, di- and trisaccharides to hydroxyl side chains of
serine or threonine residues (Figures 2-9, 2-10). PglO was originally identified as the enzyme
responsible for glycosylation of N. gonorrhoeae Type IV pili, but has since been shown to
glycosylate a wide variety of periplasmic and extracellular lipoproteins. In all examples of nonpilin glycosylation substrates, the acceptor serine or threonine residues are present in loop
regions predicted to have undefined structures rich in Ala, Ser, and Pro residues (5).
PglO
glycosylates a wide range of periplasmic proteins containing serine and threonine residues in
vivo, but it is unclear what binding determinants affect this reaction. Further biochemical and
structural analyses are needed to understand how PglO recognizes pilin and non-pilin protein
substrates.
69
80
PgIO substrate specificity
70
$
60
50
40
S30
7
0
10
0
Figure 2-10. PglO reaction turnover after overnight incubation with pilin protein in the presence
of a panel of Und-PP-linked substrates from N. gonorrhoeae (diNAcBac-Gal, diNAcBac-(Gal) 2,
diNAcBac-Glc), C. jejuni (diNAcBac-GalNAc, diNAcBac-(GalNAc) 2) or both (diNAcBac). A
negative control is shown in which the assay is performed in the presence of Und-PP-diNAcBacGal and the absence of PglO. The assays were performed in triplicate and the error bars indicate
standard deviation.
Glycan Donor Specificity of PglO
To further characterize PglO, a screen of various glycan donors was performed. Previous
studies by Feldman and coworkers (21, 22) on PglL, the homologous oligosaccharyltransferase
found in Neisseria meningitidis, have suggested that these enzymes exhibit relaxed substrate
specificity in vivo and can transfer oligosaccharides composed of different sugars, linkages and
lengths. Thus, PglO was assayed with the four native substrates (the products of PglB, PglA,
PglE and PglH) and with two substrates from the C. jejuni pathway: Und-PP-diNAcBac
modified with one or two GalNAc residues.
All glycosyl donors were prepared with high
specific activity and purified via NP-HPLC. Surprisingly, and in contrast to the in vivo studies
70
with PglL(Nm) (21, 22), PglO was only able to transfer the four native substrates; the two C.
jejuni substrates had < 3% turnover (Figure 2-10).
To verify that the C. jejuni substrates were functional as glycan donors, all six substrates
were assayed with PglB(C).
Pilin was used as the protein substrate in these assays as well,
because it contains an N-linked glycosylation sequon ( 59ENNTS 63 ) adjacent to the site of 0-
linked glycosylation (46-47). As seen in Figure 2-11, PglB(C) transferred the native C. jejuni
substrates in addition to the diNAcBac-Gal disaccharide from N. gonorrhoeae; it showed low
reactivity with diNAcBac-Glc and diNAcBac-(Gal) 2.
Since pilin is glycosylated by both
PglB(Cj) and PglO(Ng), it was important to confirm the identity of the glycosylated residues.
Pilin variants were prepared with alanine mutations at the expected sites of glycosylation for
PglB(Cj) and PglO, Asn6l and Ser63, respectively. PglB(Cj) was unable to glycosylate pilinN61A, validating this residue as the N-glycan acceptor site. However, PglO showed ~85% of
normal activity with the pilin S63A mutant suggesting that another site, potentially Thr62, is a
glycosyl acceptor site in the absence of Ser63. Further mutational analysis confirmed this
hypothesis; pilin-T62A exhibited normal glycosylation, whereas glycosylation was greatly
reduced in the pilin double mutant (T62A/S63A).
71
PgIB(C) substrate specificity
35
30
0
E
S25
20
10
5
0
Figure 2-11. PglB reaction turnover after overnight incubation with pilin protein in the presence
of a panel of Und-PP-linked substrates from N. gonorrhoeae (diNAcBac-Gal, diNAcBac-(Gal) 2 ,
diNAcBac-Glc), C. jejuni (diNAcBac-GalNAc, diNAcBac-(GalNAc) 2 ) or both (diNAcBac). A
negative control is shown in which the assay is performed in the presence of Und-PP-diNAcBacGalNAc and the absence of PglB. The assays were performed in triplicate and the error bars
indicate standard deviation.
Recent in vivo analyses revealed that both PglL(Nm), which has 95% sequence identity
with PglO, and the N-linked OTase PglB(Cj) were promiscuous enzymes that transferred a
variety of Und-PP-linked O-antigen substrates to serine or asparagine residues in proteins,
respectively (21-22).
These experiments were performed by co-expressing PglL(Nm) or
PglB(Cj) heterologously in E. coli with the glycan-acceptor protein and a locus encoding for the
biosynthesis of an Und-PP-linked substrate.
In addition, previous studies showed that N.
gonorrhoeae strains with heterologously expressed PglA(C) contained proteins modified by the
C. jejuni disaccharide, diNAcBac-GalNAc, implying that PglO can recognize this glycan in vivo
(11). However, in this study, we have found that the OTases do not show a comparable substrate
72
promiscuity in vitro; in fact, it appears that PglO and PglB(C) are both specific for their native
substrates in vitro (Figures 2-10, 2-11).
The previous in vivo studies on PglO, PglL(Nm) and PglB(C) were performed in the
absence of native substrate and under conditions in which the non-native undecaprenyl-linked
glycan accumulated in the membrane (21-22). Thus, the local concentration of the substrate
within the two-dimensional plane of the membrane was likely to be much higher than in the in
vitro assay, which would promote reaction with PglO. Additionally, in the context of a lipid
bilayer, the role of the membrane-bound undecaprenyl moiety may play a greater role in enzyme
recognition of the substrate. In our assay, the concentration of undecaprenyl substrate (10-20
nM) was well below 2.7 gM, the apparent Km of PglB(Cj) for Und-PP-disaccharide (46), and
thus specificity differences between native and nonnative substrates were easily distinguished.
In addition, it should be noted that the OTases exhibit substrate specificity in native cellular
contexts. In the native bacteria, PglO, PglL(Nm), and PglB(C) selectively transfer the correct
oligosaccharide to pilin in the presence of other undecaprenyl-linked substrates, including those
involved in capsular polysaccharide biosynthesis in N. meningitidis (48) and C. jejuni (49) and
the peptidoglycan subunits in the cellular membranes of all three species (50).
Conclusions
In conclusion, this work represents the first complete biochemical characterization of the
unusual O-linked glycosylation pathway in N. gonorrhoeae.The stereochemistry of the DATDH
sugar has been identified as diNAcBac. In addition, the substrate preferences of the
glycosyltransferases have been characterized and in general these enzymes are shown to be
specific for their native substrates. Finally, in vitro characterization of the OTases from N.
73
gonorrhoeae and C. jejuni has suggested that these enzymes prefer their respective native
glycans to closely related oligosaccharides. The O-linked pathways found in N. gonorrhoeaeand
N. meningitidis are interesting hybrids of O-linked and N-linked glycosylation pathways.
Although the role of the O-linked glycans in Neisseria pathogenicity is not yet understood, the
Pgl enzymes may represent unique virulence targets and this initial study provides the foundation
for further investigations into the biochemistry of the enzymes.
Acknowledgments
Dr. Meredith Hartley was an amazing collaborator and it was a pleasure working with her on the
work presented in this chapter. I would like to thank Dr. Matthieu Sainlos, Dr. Langdon Martin,
and Dr. Cliff Stains for obtaining the MALDI MS data. In addition, I am grateful for the
assistance of Dr. Jeff Simpson of the MIT Department of Chemistry Instrumentation Facility in
the NMR characterization of UDP-diNAcBac. Finally, I am grateful for the many productive
discussions of this manuscript with members of the Imperiali lab including Dr. Jerry Troutman
and Marcie Jaffee.
Experimental Procedures
Common Materials
All radioactive materials and undecaprenol were obtained from American Radiolabeled
Chemicals. UDP-4-amino and UDP-diNAcBac were prepared as previously described (28). All
other chemicals were obtained from Sigma-Aldrich unless stated otherwise. Radioactivity was
determined using a LS6500 Beckman Scintillation Counter; organic samples were dried and
74
resuspended in 200 ptL SolvableTM (Perkin-Elmer) and 5 mL of scintillation fluid (Opti-Fluor,
Perkin-Elmer).
Aqueous samples were mixed with 5 mL of Ecolite(+)TM (MP Biomedicals)
prior to scintillation counting.
Preparationof genetic constructs
The genes pgD, pgC, pgB, pgA, pgO, and pilE were PCR amplified from the N.
gonorrhoeae strain MS 11(7,11-12), whereas pg/E was amplified from N. gonorrhoeaestrain FA
1090 and pg/H was amplified from the N. meningitidis strain Z249 1. The PCR products ofpgD,
pgC, pgB, pgA, pgO, and pg/E were cloned into BamH I/Xho I in the pET-24a(+) vector
(Novagen). The pilE and Pg/H genes were cloned into Nde I/Xho I in the pET-24a(+) vector
(Novagen). The Xho I site was inserted prior to the stop codon to encode for a His 6 tag at the Cterminal end of each protein.
The acetyltransferase domain of PglB (PglB-ATD) was identified through sequence
homology with the related C. jejuni protein, PglD(Cj).
amplified
from
the
full-length
gene
The gene encoding the domain was
using
the
forward
primer
5'-
CGCGGATCCATGGCGGGGAATCGCAAACTCG-3'
and
GCAACCCGGCAAAGCCCCTTTAGCTCGAGCGG-3'
to generate a gene encoding the
the
reverse
primer
5'-
acetyltransferase domain. The gene was inserted into BamHI/XhoI in a modified pET-30b(+)
vector which contains an N-terminal octa-histidine tag followed by a Tobacco Etch Virus (TEV)
protease site. Also, pg/H was amplified by PCR and inserted into BamH I/Xho I in the pMALc2X vector. This construct encoded for the addition of an N-terminal maltose binding protein
(MBP).
75
Expression ofproteins
In general, all proteins (PglD, PglC, PglB, PglB-ATD, PglA, PglE, PglO, Pg1H, PilE
(pilin), PglB(Cj) (27) and S. mutans undecaprenol kinase (43)) were expressed heterologously in
E. coli BL21 cells (Agilent). PglD, PglC and PglB-ATD were expressed in the BL21(DE3)
pLysS strain; all other proteins were expressed in the BL2 1-Gold (DE3) strain.
A typical
expression protocol involved preparation of an overnight culture of cells (5 mL), which was used
to inoculate 1 L of LB media with shaking at 37 0C. After the cells reached an optical density of
~0.8 absorbance units, the temperature was lowered to 16 C and the cells were induced with 0.5
mM iso-p-D-thiogalactosylpyranoside (IPTG). After 16-18 hours of incubation, the cells were
harvested and the pellets were stored at -80 C.
Proteinpurification
In general, all steps of protein purification were carried out at 4
C.
Protein
concentrations were determined with the appropriate extinction coefficients at a UV absorbance
of 280 nm, with the exception of PglO and pilin, which were quantified with the MicroBCA
Assay (Pierce) due to the presence of the highly absorbent detergent Triton X- 100.
The cell pellets generated from the expression of the soluble proteins, PglC and PglBATD, were resuspended in 50 mL of ice-cold 50 mM HEPES (pH 7.4) and 100 mM NaCl
(Buffer A), supplemented with 30 mM imidazole and lysed by sonication. In the case of PglC,
200 pM pyridoxal-5'-phosphate was also added to the buffer.
The lysate was cleared by
centrifugation (145,000 x g) for 45 min. Cleared lysate was mixed with 2 mL of Ninitrilotriacetic acid (Ni-NTA) resin (Qiagen), tumbled for 4 hours, and then packed into a fritted
PolyPrep column (Biorad). Using gravity flow, the resin-bound protein was washed with 10
76
column volumes of Buffer A containing 30 mM imidazole. The resin was further washed with
20 column volumes of Buffer A supplemented with 40 mM imidazole and then 10 column
volumes of Buffer A containing 60 mM imidazole.
The protein was eluted in Buffer A
supplemented with 250 mM imidazole and 1 mL fractions were collected. Fractions containing
purified material were assessed by SDS-PAGE (12%) and Western blot analysis probing for the
His 6 tag. Pooled fractions of PglC and PglB-ATD were dialyzed against Buffer A, concentrated,
supplemented with a final glycerol concentration of 15% and frozen at -80 C (Figure 2-3, lanes
2 and 3).
Purification of the glycosyltransferase PglA was similar to PglC and PglB-ATD with a
few exceptions. A buffer containing 50 mM Tris (pH 8.0) and 150 mM NaCl (Buffer B) was
used instead of Buffer A and the cells were incubated with 1% Triton X-100 for 20 minutes
immediately following lysis and prior to centrifugation. In addition, 5% glycerol was added to
all buffers. Following elution, the most concentrated 1.5 mL fraction (~10 pM) as determined by
SDS-PAGE was desalted using a Hi-Trap desalting cartridge (GE Healthcare) with Buffer B and
stored at -20 C in 30% glycerol (Figure 2-3, lane 5).
To purify the membrane-associated proteins (PglD, PglB, PglE, PglO, and pilin), a cell
envelope fraction (CEF) was first prepared. The cells were thawed in 40 mL of buffer per L of
cell culture and lysed by sonication. PBS supplemented with 200 pM NAD+ was used for PglD
and Buffer B with 1 mg/mL lysozyme was used for PglB, PglE, PglO and pilin. Cellular debris
was cleared by centrifugation at 9000 x g for 45 minutes.
The resulting supernatant was
transferred to a clean centrifuge tube and subjected to centrifugation at 145,000 x g for 65 min to
pellet the CEF. In the case of PglB and PglE, the CEF was resuspended in half the volume of the
77
unlysed cell pellet weight (i.e. 1.5 mL was used for 3 g cell pellet). The CEF was aliquoted and
stored at -80 'C (Figure 2-3, lanes 4 and 6).
PglD, PglO and pilin were further purified from the CEF. The CEF was homogenized in
10 mL of buffer containing 1% Triton X-100 (PBS with 200 pM NAD+ for PglD and Buffer B
for PglO and pilin) per liter of cell culture. Each CEF was incubated with detergent for several
hours and then centrifuged again (145,000 x g) to remove insoluble material.
The resultant
supernatants were incubated with 0.5-2 mL of Ni-NTA resin for 1-2 hours; the resins were
washed as previously described with the addition of 0.1% Triton X-100 to the wash and elution
buffers. The proteins were eluted from the resin in 1 mL fractions. Pooled fractions of PglD
were dialyzed against PBS containing 200 jM NAD+ and 0.1% Triton X-100, supplemented
with a final glycerol concentration of 30%, and frozen at -80 C (Figure 2-3, lane 1). The most
concentrated fractions of PglO and pilin (5 [tM and 40 jM, respectively) were desalted as
described above for PglA and stored at -80 C (Figure 2-3, lanes 7 and 9).
PglH was expressed as an MBP-fusion protein and purified as described elsewhere (14)
(Figure 2-3, lane 8).
In addition, PglB(Cj) and undecaprenol kinase from S. mutans were
expressed and purified as cell envelope fractions as described previously (27,43).
Acetyltransferase (PglB-A TD) Activity Assay
Determination of the kinetic constants for PglB-ATD were carried out using Ellman's
reagent, 5,5'-dithio-bis-(2-nitrobenzoic acid), in a continuous fashion. Ellman's reagent was
utilized to quantify substrate turnover as monitored by measuring conversion of AcCoA to
CoASH using the released TNB chromophore (kmax = 412 nm,
Xmax
= 14,150 M- cm 1 ). The in
vitro assay contained 50 mM HEPES (pH 7.4), 1 mM EDTA, 1 mM DTNB, and 25 nM PglB-
78
ATD in a quartz cuvette. The substrate concentrations of AcCoA and UDP-4-amino were varied
separately to determine kinetic parameters using initial velocity measurements while keeping the
other substrate at saturation. The reaction was initiated with the UDP-4-amino substrate and
took place at room temperature over a 200 second time period. The absorbance change at 412
nm was measured.
A blank reaction lacking UDP-4-amino was prepared as a background
control. Steady-state rate parameters were calculated from equation 1 using the program GraFit
6.0.12 (Erithacus Software).
V = Vmax[S]/(Km + [S])
(1)
Aminotransferase (Pg/C) Activity Assay
The aminotransferase reaction was assayed by coupling generation of the UDP-4-amino
product from the PglC reaction to the acetyltransferase activity of PglD from C. jejuni (QY)
producing CoASH, which was detected by Ellman's reagent in a similar fashion to the PglBATD assay. In a flat bottom 96-well plate (Nunc) 50 mM HEPES (pH 7.4), 1 pM PglD(C), 400
pM AcCoA, and 400 nM PglC were added. Since PglC activity was coupled to the turnover of
the acetyltransferase PglD(Cj), addition of an excess amount of PglD(Cj) ensured that the initial
velocity measurements were dependent only upon PglC activity.
The concentrations of L-
glutamate and UDP-4-keto were varied separately to determine kinetic parameters using initial
velocity measurements while keeping the other substrate at saturation. Interference of Ellman's
reagent with PglC activity required the implementation of a discontinuous assay in which
reactions were initiated with L-glutamate and quenched with 20% 1 -propanol, 2 mM DTNB, and
1 mM EDTA over a 30 minute time period. The absorbance at 415 nm was followed on an
79
Ultramark EX microplate imaging system (BioRad). A blank reaction without L-glutamate was
set up as a background control.
Biosynthesis and stereochemicalassignmentof UDP-DA TDH
In order to biosynthesize UDP-DATDH, 0.3 mg of PglD (bound to Ni-NTA resin) was
added to 15 mL buffer containing 50 mM HEPES (pH 7.4), 100 mM NaCl, 200 mM NAD+, and
30 mg UDP-GlcNAc. The reaction was carried out at room temperature for 12 hours with gentle
rocking.
Once conversion to the UDP-4-keto sugar was complete, as verified by capillary
electrophoresis as described previously (28), the reaction was filtered and the flow-through
collected. The filtrate containing the UDP-4-keto sugar was supplemented with 15 mg PglC
(bound to Ni-NTA resin), 20 mM L-glutamate, and 200 mM pyridoxal-5'-phosphate.
This
reaction was filtered after rocking for 18 hours at room temperature and reaching 80%
conversion to the UDP-4-amino sugar. The filtrate was supplemented with 0.2 mg full-length
PglB (bound to Ni-NTA resin) and 1.2 mM AcCoA and allowed to react at room temperature
with rocking for 12 hours. The slurry was filtered and the flow-through containing the UDPDATDH sugar was collected. Purification and NMR characterization of the final UDP-DATDH
product was completed as previously described (28).
Preparationof radiolabeledUnd-PP-linkedsubstrates
In general, radiolabeled Und-PP-substrates were prepared at two different specific
activity levels: a higher specific activity for the OTase assay and analysis by normal phase-high
performance liquid chromatography (NP-HPLC) and a lower specific activity for the
glycosyltransferase assays.
80
Und-PP-[ 3 H]diNAcBac was enzymatically synthesized using the S. mutans undecaprenol
kinase described previously (51) and PglB. An undecaprenol kinase from N. gonorrhoeae has
not been characterized, and thus the undecaprenol kinase from S. mutans (43) was used a tool to
affect the undecaprenol phosphorylation in situ. A typical reaction contained 3% DMSO, 1%
Triton X-100, 50 mM MgCl 2 , 30 mM Tris-Acetate (pH 8.0), 500 pM undecaprenol, 1 mM ATP,
500 pM UDP-4-amino, 500 pM [3 H]acetyl-CoA (20 mCi/mmol), 15-20 piL of undecaprenol
kinase CEF, 15-20 pL PglB CEF and water to a final volume of 100 ptL. The reaction was
modified to prepare Und-PP-diNAcBac with high specific activity by adjusting the undecaprenol
and UDP-4-amino concentrations to 100 jM and the [3H]acetyl-CoA concentration to 4.5 pM
(20 Ci/mmol). After incubation at room temperature for 2 hours, the reactions were quenched
into 1 mL 2:1 CHCl 3 :MeOH and extracted three times with 400 pL of an aqueous extract
prepared by dissolving 1.83 g of potassium chloride in 235 mL water, 240 mL chloroform, and
15 mL methanol. The organic layer containing the Und-PP-diNAcBac product was dried down
and purified using NP-HPLC as described below.
Und-PP-diNAcBac-[3H]Gal was prepared in a similar manner to Und-PP-diNAcBac. The
reaction components are as described above for Und-PP-diNAcBac with the following
exceptions; 500 jM UDP-diNAcBac was added instead of UDP-4-amino and AcCoA, and 2 pM
PglA and 500 pM UDP-[3H]Gal (20 mCi/mmol) were added to affect the transfer of the
galactosyl unit, which is the second sugar in the glycan. To prepare Und-PP-diNAcBac-[ 3 H]Gal
with higher specific activity, undecaprenol and UDP-diNAcBac concentrations were lowered to
100 pM and the UDP-[3H]Gal concentration was adjusted to 4.5 pM (20 Ci/mmol). The
reactions were quenched after two hours and extracted as described above.
81
The synthesis of Und-PP-diNAcBac-Gal-[3H]Gal utilized unlabeled Und-PP-diNAcBacGal, which was prepared as described above with the exception that the UDP-Gal added was not
radioactive. A typical biosynthesis reaction contained 3% DMSO, 0.05% Triton X-100, 50 mM
MnCl 2, 50 mM HEPES (pH 7.5), 20 pM Und-PP-diNAcBac-Gal, 20 gM UDP-[ 3H]Gal (20
mCi/mmol), 20 gL PglE CEF and water to a final volume of 100 pL. To prepare the substrate
with higher specific activity, the UDP-Gal concentration was lowered to 4.5 pM UDP-[3H]Gal
(20 Ci/mmol). The reactions were quenched after two hours and extracted as described above.
Und-PP-diNAcBac-[3H]Glc was prepared from unlabeled Und-PP-diNAcBac.
The
reaction contained 3% DMSO, 0.1% DDM, 50 mM MgCl 2, 30 mM Tris (pH 8.0), 20 RM UndPP-diNAcBac, 20 pM UDP-[3H]Glc (20 mCi/mmol), 10 pM PglH and water to a final volume of
100 piL. The substrate was also prepared with a higher specific activity by lowering the UDP[3 H]Glc concentration to 4.5 pM (20 Ci/mmol). The C. jejuni substrates (Und-PP-diNAcBac[3H]GalNAc
and Und-PP-diNAcBac-GaNAc-[3H]GalNAc)
for the OTase reactions were
prepared as previously described (25-27,29,31) with similar specific activities to the N.
gonorrhoeaeOTase substrates.
Normalphase HPLCpurificationof Und-PP-linkedsubstrates
The dried Und-PP-linked substrates were purified via NP-HPLC with a Varian Microsorb
column using the previously described gradient (42). The substrates were resuspended in 100 piL
of 4:1 CHCl 3 :MeOH for injection onto the column. Fractions of 1 mL were collected and 10 piL
of each fraction was solubilized in 200 piL SolvableTM for detection of radioactivity. The
fractions containing substrate were combined, aliquoted and stored at -20 C.
82
To obtain the NP-HPLC analytical traces, Und-PP-linked glycan fractions were resolubilized in 4:1 CHCl 3 :MeOH andlOO pL of the appropriate sample was injected onto the
column.
The 1 mL elution fractions were dried completely and resuspended in 200 pL
SolvableTM for scintillation counting.
Preparationand analysis of 2-AB labeled oligosaccharides
Unlabeled versions of Und-PP-diNAcBac, Und-PP-diNAcBac-Gal, Und-PP-diNAcBac(Gal) 2 , and Und-PP-diNAcBac-Glc were prepared in an identical manner to the preparation of
the radiolabeled substrates, except that unlabeled substrates were used in all reactions. The
oligosaccharides were labeled with 2-aminobenzamide as previously described (26,52) and
purified using the GlykoNSep column (Prozyme). The appropriate peaks were collected and
matrix-assisted laser desorption ionization mass spectrometry (MALDI MS) was used to
determine the mass of the 2-AB labeled glycans.
Glycosyltransferasesubstrate specificity assays
To determine the UDP-sugar specificity of PglB, PglA, and PglE, radioactivity-based
were performed on a variety of UDP-linked sugar substrates. The ability of PglB to transfer
UDP-diNAcBac, UDP-Glc, UDP-Gal, UDP-GlcNAc, and UDP-GalNAc was analyzed.
The
activity of PglB was coupled to the action of S. mutans undecaprenol kinase in order to provide
the undecaprenyl phosphate in situ (43). The specificity assays included 3% DMSO, 1% Triton
X-100, 50 mM MgCl 2 , 30 mM Tris-Acetate (pH 8.0), 20 gM undecaprenol, 1 mM ATP, 2 pM
UDP-[3H]sugar (20 mCi/mmol), 20 pL of undecaprenol kinase CEF, 5 piL PglB CEF and water
to a final volume of 100 ptL. In the case of UDP-diNAcBac transfer, 2 pM UDP-4-amino and 2
83
pM AcCoA (20 mCi/mmol) were included to assay both activities of the bifunctional PglB,
which carries out transfer of the acetyl group to UDP-4-amino and transfer of phosphodiNAcBac to undecaprenyl phosphate. The reactions were initiated with a mixture of ATP and
PglB and were monitored by quenching 15 gL aliquots at 20, 40, 60, 80, and 100 seconds. The
radioactivity present in the organic and aqueous layers was determined as described above.
The ability of PglA and PglE to transfer UDP-Glc, UDP-Gal, UDP-GlcNAc, and UDPGalNAc was also analyzed. In the case of PglA, the assays contained 3% DMSO, 0.05% Triton
X-100, 50 mM MgCl 2 , 30 mM Tris-Acetate (pH 8.0), 10 pM Und-PP-diNAcBac, 2 pM UDPsugar (20 mCi/mmol), 0.1 pM PglA and water to a final volume of 100 ptL. The reaction was
monitored by quenching 15 pL at 1, 2, 3, 4, and 5 minutes. For PglE, the assays contained 3%
DMSO, 0.05% Triton X-100, 50 mM MnCL2, 50 mM HEPES (pH 7.5), 2 gM Und-PPdiNAcBac-Gal, 5 pM UDP-sugar (20 mCi/mmol), 5 ptL PglE CEF and water to a final volume of
100 pL. The reaction was monitored by quenching 15 pL at 2, 4, 6, 8, and 10 minutes. For both
assays, the reactions were initiated by addition of enzyme.
In addition, to test the ability of PglB and PglA to distinguish between UDP-4-amino and
UDP-diNAcBac, a coupled reaction was performed.
The reaction components contained 3%
DMSO, 1% Triton X-100, 40 mM MgCl 2 , 30 mM Tris-Acetate (pH 8.0), 50 jM undecaprenol, 1
mM ATP, 500 pM UDP-4-amino, 500 pM AcCoA, 90 nM UDP-Gal (20 Ci/mmol), 5 piL
undecaprenol kinase CEF, 10 piL PglB CEF and 1 pM PglA with a final volume of 100 pL. The
reaction was initiated with a mixture of ATP and PglB. The extent of the reaction was monitored
by quenching 15 pL aliquots at 1, 3, 5, 7, and 9 minutes.
To test if PglB and PglA could
recognize UDP-4-amino, a second reaction was prepared in which the AcCoA was omitted. The
84
quenched aliquots were extracted as described above and the radioactivity present in the organic
fraction was determined by scintillation counting.
Oligosaccharyltransferaseassays
The OTase reactions were performed with a variety of Und-PP-linked glycosyl donors.
In general, the reactions contained 5% DMSO, 0.7% Triton X-100, 50 mM MnCl 2, 25 mM
HEPES (pH 7.5), 70 mM sucrose, 10-20 nM Und-PP-substrate (20 Ci/mmol), 8 pM pilin and 1
pM PglO in 100 pL reaction volume.
The reactions were incubated overnight at room
temperature with shaking. The glycosylated pilin protein was isolated via Ni-NTA purification.
Briefly, the reaction was incubated with 15 pL of Ni-NTA resin for several hours in a 1.5 mL
eppendorf tube. The tube was briefly centrifuged and the supernatant was removed. The resin
was then washed five times with 500 ptL Buffer A containing 30 mM imidazole and 0.1% Triton
X-100.
For each wash, the buffer was added to the eppendorf tube, the resin was mixed
thoroughly with buffer, and the supernatant was removed following a brief centrifugation. The
protein was eluted in three fractions of 500 ptL of Buffer A containing 300 mM imidazole and
0.1% Triton X- 100. Scintillation fluid (Ecolite(+)TM, MP Biomedicals) was added to all flowthrough, wash and elution fractions and the radioactivity of each sample was determined.
Glycosylated protein samples for MALDI MS and Western blot analysis were prepared and
purified in the same manner, except that unlabeled versions of Und-PP-diNAcBac or Und-PPdiNAcBac-Gal substrate were used at concentrations of 10-24 pM.
Parallel reactions with
radioactive substrates were performed at identical concentrations in order to determine reaction
yields. The Western blot analysis was performed following standard protocols. An antibody
specific for His 4 (Qiagen) was used as a positive control for the purified proteins and a
85
diNAcBac-epitope monoclonal antibody termed npg 1, which was previously described (45), was
used to detect diNAcBac-modified protein.
References
1. Craig L, Pique ME, Tainer JA. Type IV pilus structure and bacterial pathogenicity. Nat. Rev.
Microbiol. 2004; 2:363-378.
2. Bieber D, Ramer SW, Wu CY, Murray WJ, Tobe T, Fernandez R, Schoolnik GK. Type IV
pili, transient bacterial aggregates, and virulence of enteropathogenic. Escherichia coli,
Science. 1998; 280:2114-2118.
3. Hansen JK, Forest KT. Type IV pilin structures: insights on shared architecture, fiber
assembly, receptor binding and type II secretion. J. Mol. Microbiol. Biotechnol. 2006;
11:192-207.
4. Chamot-Rooke J,Mikaty G, Malosse C, Soyer M, Dumont A, Gault J, Imhaus AF, Martin P,
Trellet M, Clary G, Chafey P, Camoin L, Nilges M, Nassif X, Dumenil G. Posttranslational
modification of pili upon cell contact triggers N. meningitidis dissemination. Science. 2011;
331:778-782.
5. Vik A, Aas FE, Anonsen JH, Bilsborough S, Schneider A, Egge-Jacobsen W, Koomey M.
Broad spectrum O-linked protein glycosylation in the human pathogen Neisseria
gonorrhoeae.Proc. Natl. Acad. Sci. U S A. 2009; 106:4447-4452.
6. Karlyshev AV, Everest P, Linton D, Cawthraw S, Newell DG, Wren BW. The
Campylobacterjejuni general glycosylation system is important for attachment to human
epithelial cells and in the colonization of chicks. Microbiology. 2004; 150:1957-1964.
7. Banerjee A, Wang R, Supernavage SL, Ghosh SK, Parker J, Ganesh NF, Wang PG, Gulati S,
Rice PA. Implications of phase variation of a gene (pgtA) encoding a pilin galactosyl
transferase in gonococcal pathogenesis. J. Exp. Med. 2002; 196:147-162.
8. Power PM, Ku SC, Rutter K, Warren MJ, Limnios EA, Tapsall JW, Jennings MP. The phase
variable allele of the pilus glycosylation gene pglA is not strongly associated with strains of
Neisseriagonorrhoeaeisolated from patients with disseminated gonococcal infection. Infect.
Immun. 2007; 75:3202-3204.
9. Szymanski CM, Burr DH, Guerry P. Campylobacterprotein glycosylation affects host cell
interactions. Infect. Immun. 2002; 70:2242-2244.
10. Banerjee A, Ghosh SK. The role of pilin glycan in neisserial pathogenesis. Mol. Cell.
Biochem. 2003; 253:179-190.
11. Aas FE, Vik A, Vedde J, Koomey M, Egge-Jacobsen W. Neisseria gonorrhoeae O-linked
pilin glycosylation: functional analyses define both the biosynthetic pathway and glycan
structure. Mol. Microbiol. 2007; 65:607-624.
12. Hegge FT, Hitchen PG, Aas FE, Kristiansen H, Lovold C, Egge-Jacobsen W, Panico M,
Leong WY, Bull V, Virji M, Morris HR, Dell A, Koomey M. Unique modifications with
phosphocholine and phosphoethanolamine define alternate antigenic forms of Neisseria
gonorrhoeaetype IV pili. Proc. Natl. Acad. Sci. U S A. 2004; 101:10798-10803.
13. Stimson E,Virji M, Makepeace K, Dell A, Morris HR, Payne G, Saunders JR, Jennings MP,
86
Barker S, Panico M, et al. Meningococcal pilin: a glycoprotein substituted with digalactosyl
2 ,4 -diacetamido-2,4,6-trideoxyhexose.
Mol. Microbiol. 1995; 17:1201-1214.
14. Borud B, Viburiene R, Hartley MD, Paulsen BS, Egge-Jacobsen W, Imperiali B, Koomey M.
A novel glucosyltransferase, PglH, further expands the high glycan diversity in 0-linked
protein glycosylation of Neisseria species. Proc. Natl. Acad. Sci. U S A. 2011; 108, 96439648.
15. Jennings MP, Virji M, Evans D, Foster V, Srikhanta YN, Steeghs L, van der Ley P, Moxon
ER. Identification of a novel gene involved in pilin glycosylation in Neisseria meningitidis.
Mol. Microbiol. 1998; 29:975-984.
16. Power PM, Roddam LF, Dieckelmann M, Srikhanta YN, Tan YC, Berrington AW, Jennings
MP. Genetic characterization of pilin glycosylation in Neisseria meningitidis. Microbiology.
2000; 146(Pt 4):967-979.
17. Power PM, Roddam LF, Rutter K, Fitzpatrick SZ, Srikhanta YN, Jennings MP. Genetic
characterization of pilin glycosylation and phase variation in Neisseria meningitidis.Mol.
Microbiol. 2003; 49:833-847.
18. Power PM, Seib KL, Jennings MP. Pilin glycosylation in Neisseriameningitidis occurs by a
similar pathway to wzy-dependent O-antigen biosynthesis in Escherichia coli. Biochem.
Biophys. Res. Commun. 2006; 347:904-908.
19. Warren MJ, Roddam LF, Power PM, Terry TD, Jennings MP. Analysis of the role of pglI in
pilin glycosylation of Neisseria meningitidis. FEMS Immunol. Med. Microbiol. 2004; 41:4350.
20. Chamot-Rooke J, Rousseau B, Lanternier F, Mikaty G, Mairey E, Malosse C, Bouchoux G,
Pelicic V, Camoin L, Nassif X, Dumenil G. Alternative Neisseria spp. type IV pilin
glycosylation with a glyceramido acetamido trideoxyhexose residue. Proc. Natl. Acad. Sci. U
S A. 2007; 104:14783-14788.
21. Faridmoayer A, Fentabil MA, Haurat MF, Yi W, Woodward R, Wang PG, Feldman MF.
Extreme substrate promiscuity of the Neisseria oligosaccharyl transferase involved in protein
Oglycosylation. J. Biol. Chem. 2008; 283:34596-34604.
22. Faridmoayer A, Fentabil MA, Mills DC, Klassen JS, Feldman MF. Functional
characterization of bacterial oligosaccharyltransferases involved in O-linked protein
glycosylation. J. Bacteriol. 2007; 189:8088-8098.
23. Szymanski CM, Yao R, Ewing CP, Trust TJ, Guerry P. Evidence for a system of general
protein glycosylation in Campylobacterjejuni.Mol. Microbiol. 1999; 32:1022-1030.
24. Young NM, Brisson JR, Kelly J, Watson DC, Tessier L, Lanthier PH, Jarrell HC, Cadotte N,
St Michael F, Aberg E, Szymanski CM. Structure of the N-linked glycan present on multiple
glycoproteins in the Gram-negative bacterium, Campylobacterjejuni.J. Biol. Chem. 2002;
277:42530-42539.
25. Glover KJ, Weerapana E, Chen MM, Imperiali B. Direct biochemical evidence for the
utilization of UDP-bacillosamine by PglC, an essential glycosyl-1-phosphate transferase in
the Campylobacterjejuni N-linked glycosylation pathway. Biochemistry. 2006; 45:53435350.
26. Glover KJ, Weerapana E, Imperiali B. In vitro assembly of the undecaprenylpyrophosphatelinked heptasaccharide for prokaryotic N-linked glycosylation. Proc. Natl. Acad. U.S.A.
2005; 102:14255-14259.
87
27. Glover KJ, Weerapana E, Numao S, Imperiali B. Chemoenzymatic synthesis of
glycopeptides with PglB, a bacterial oligosaccharyl transferase from Campylobacterjejuni.
Chem. Biol. 2005; 12:1311-1315.
28. Olivier NB, Chen MM, Behr JR, Imperiali B. In vitro biosynthesis of UDP-N,N'diacetylbacillosamine by enzymes of the Campylobacterjejunigeneral protein glycosylation
system. Biochemistry. 2006; 45:13659-13669.
29. Krogh A, Larsson B, von Heijne G, Sonnhammer EL. Predicting transmembrane protein
topology with a hidden Markov model: application to complete genomes. J. Mol. Biol. 2001;
305:567-580.
30. Takeuchi K, Taguchi F, Inagaki Y, Toyoda K, Shiraishi T, Ichinose Y. Flagellin
glycosylation island in Pseudomonas syringae pv. glycinea and its role in host specificity. J.
Bacteriol. 2003; 185:6658-6665.
31. Horzempa J, Carlson PE Jr, O'Dee DM, Shanks RM, Nau GJ. Global transcriptional response
to mammalian temperature provides new insight into Francisellatularensis pathogenesis.
BMC Microbiol. 2008; 8:172.
32. van Sorge NM, Bleumink NM, van Vliet SJ, Saeland E, van der Pol WL, van Kooyk Y, van
Putten JP. N-glycosylated proteins and distinct lipooligosaccharide glycoforms of
Campylobacterjejuni target the human C-type lectin receptor MGL. Cell. Microbiol. 2009;
11:1768-1781.
33. Verma A, Arora SK, Kuravi SK, Ramphal R. Roles of specific amino acids in the N terminus
of Pseudomonas aeruginosa flagellin and of flagellin glycosylation in the innate immune
response. Infect. Immun. 2005; 73:8237-8246.
34. Sharon N, Jeanloz RW. The diaminohexose component of a polysaccharide isolated from
Bacillus subtilis. J. Biol. Chem. 1960; 235:1-5.
35. Sharon N. Celebrating the golden anniversary of the discovery of bacillosamine, the diamino
sugar of a Bacillus. Glycobiology. 2007; 17:1150-1155.
36. Schaffer C, Scherf T, Christian R, Kosma P, Zayni S, Messner P, Sharon N. Purification and
structure elucidation of the N-acetylbacillosamine-containing polysaccharide from Bacillus
licheniformis ATCC 9945. Eur. J. Biochem. 2001; 268:857-864.
37. Schoenhofen IC, Vinogradov E, Whitfield DM, Brisson JR, Logan SM. The CMPlegionaminic acid pathway in Campylobacter: biosynthesis involving novel GDP-linked
precursors. Glycobiology. 2009; 19:715-725.
38. Parge HE, Forest KT, Hickey MJ, Christensen DA, Getzoff ED, Tainer JA. Structure of the
fibre-forming protein pilin at 2.6 A resolution. Nature. 1995; 378:32-38.
39. Schoenhofen IC, McNally DJ, Vinogradov E, Whitfield D, Young NM, Dick S, Wakarchuk
WW, Brisson JR, Logan SM. Functional characterization of dehydratase/aminotransferase
pairs from Helicobacter and Campylobacter: enzymes distinguishing the pseudaminic acid
and bacillosamine biosynthetic pathways. J. Biol. Chem. 2006; 281:723-732.
40. Vijayakumar S, Merkx-Jacques A, Ratnayake DB, Gryski I, Obhi RK, Houle S, Dozois
CMCreuzenet C. Cj1121c, a novel UDP-4-keto-6-deoxy-GlcNAc C-4 aminotransferase
essential for protein glycosylation and virulence in Campylobacterjejuni. J. Biol. Chem.
2006; 281:27733-27743.
41. Bigge JC, Patel TP, Bruce JA, Goulding PN, Charles SM, Parekh RB. Nonselective and
efficient fluorescent labeling of glycans using 2-amino benzamide and anthranilic acid. Anal
Biochem. 1995; 230:229-238.
42. Troutman JM, Imperiali B. CampylobacterjejuniPglH is a single active site processive
88
polymerase that utilizes product inhibition to limit sequential glycosyl transfer reactions.
Biochemistry. 2009; 48:2807-2816.
43. Hartley MD, Larkin A, Imperiali B. Chemoenzymatic synthesis of polyprenyl phosphates.
Bioorg. Med. Chem. 2008; 16:5149-5156.
44. Imperiali B, Shannon KL. Differences between Asn-Xaa-Thr-containing peptides: a
comparison of solution conformation and substrate behavior with oligosaccharyltransferase.
Biochemistry. 1991; 30:4374-4380.
45. Borud B, Aas FE, Vik A, Winther-Larsen HC, Egge-Jacobsen W, Koomey M. Genetic,
structural, and antigenic analyses of glycan diversity in the O-linked protein glycosylation
systems of human Neisseria species. J. Bacteriol. 2010; 192:2816-2829.
46. Chen MM, Weerapana E, Ciepichal E, Stupak J, Reid CW, Swiezewska E, Imperiali B.
Polyisoprenol specificity in the CampylobacterjejuniN-linked glycosylation pathway.
Biochemistry. 2007; 46:14342-14348.
47. Kowarik M, Young NM, Numao S, Schulz BL, Hug I, Callewaert N, Mills DC, Watson DC,
Hernandez M, Kelly JF, Wacker M, Aebi M. Definition of the bacterial N-glycosylation site
consensus sequence. EMBO J. 2006; 25:1957-1966.
48. Masson L, Holbein BE. Role of lipid intermediate(s) in the synthesis of serogroup B
Neisseriameningitidis capsular polysaccharide. J. Bacteriol. 1985; 161:861-867.
49. St Michael F, Szymanski CM, Li J, Chan KH, Khieu NH, Larocque S, Wakarchuk WW,
Brisson JR, Monteiro MA. The structures of the lipooligosaccharide and capsule
polysaccharide of Campylobacter jejuni genome sequenced strain NCTC 11168. Eur J
Biochem. 2002; 269:5119-5136.
50. Hill DJ, Griffiths NJ, Borodina E, Virji M. Cellular and molecular biology of Neisseria
meningitidis colonization and invasive disease. Clin. Sci. (Lond.). 2010; 118:547-564.
51. Lis M, Kuramitsu HK. The stress-responsive dgk gene from Streptococcus mutans encodes a
putative undecaprenol kinase activity. Infect. Immun. 2003; 71:1938-1943.
52. O'Reilly MK, Zhang G, Imperiali B. In vitro evidence for the dual function of Alg2 and
Algi 1: essential mannosyltransferases in N-linked glycoprotein biosynthesis. Biochemistry.
2006; 45:9593-9603.
89
Chapter 3. Biosynthesis of UDP-N,N'-Diacetylbacillosamine in Acinetobacter
baumannii:
Biochemical
Characterization and Correlation to Existing
Pathways
A significant portion of this chapter has been published in the following reference:
Morrison, M. J., and Imperiali, B. (2013) Biosynthesis of UDP-NN'-Diacetylbacillosamine in
Acinetobacter baumannii: Biochemical Characterization and Correlation to Existing Pathways.
Arch. Biochem. Biophys. 536, 72-80.
90
Introduction
An alarming trend in antibiotic resistance continues to escalate among human pathogens.
A prime example is Acinetobacter baumannii, which has garnered a great deal of attention from
the medical community stemming from its capacity to resist a majority of antimicrobial therapies
(1). A. baumannii is a Gram-negative, aerobic, non-motile opportunistic pathogen that affects
immunocompromised patients in a hospital setting. Although much effort has been invested in
uncovering the mechanism of action of antibiotic resistance (2,3), little has been accomplished in
the understanding of pathogenicity. The AYE strain of A. baumannii was originally isolated
from the 2001 epidemic outbreak in France resulting in a 26% mortality rate among infected
individuals (4,5). Disturbingly, this strain contains an 86-kb genomic island that encodes for 45
of its 52 resistance genes (6). This resistance island, the largest identified to date, is responsible
for the inactivation of P-lactams, aminoglycosides, chloramphenicol, rifampin, and tetracycline
(7).
Extensive work has corroborated the link between virulence and bacterial glycosylation
in the model system C. jejuni (8). An interesting characteristic of virulence in this pathogen is
the biosynthesis of the unique, bacterial UDP-diNAcBac sugar and its incorporation into
complex glycoconjugates.
Importantly, the disruption of the enzymes responsible for its
production results not only in the diminished ability of C. jejuni to adhere to and invade human
epithelial cells, but also a reduction in chick and mouse colonization (9-10).
Three distinct
enzymes are employed in the biosynthesis of UDP-diNAcBac. First, a dehydratase catalyzes the
NAD* dependent C4 oxidation, which promotes elimination of water across the C5-C6 glycosyl
bond. This is followed by re-reduction of the ccP-unsaturated system at C6 to generate the UDP4-keto sugar (11).
Subsequently, an aminotransferase catalyzes the transfer of the amino group
91
from L-glutamate to the C4 position of UDP-4-keto in a pyridoxal-dependent manner to generate
the
UDP-4-amino
sugar (11).
Lastly,
an acetyl-coenzyme
A
(AcCoA)-dependent
acetyltransferase generates the UDP-diNAcBac sugar nucleotide (12). Phospho-diNAcBac is
then enzymatically transferred to undecaprenyl phosphate and serves as the starting membranebound monosaccharide building block for the assembly of more complex oligosaccharides. In C.
jejuni, the pathway that utilizes UDP-diNAcBac culminates in the transfer of a heptasaccharide
onto the side-chain amide nitrogen of asparagine (N-linked), whereas the system in N.
gonorrhoeae transfers a trisaccharide onto a serine or threonine residue (0-linked) (13).
Importantly, the first sugar in glycan biosynthesis for this O-linked system has been confirmed as
UDP-diNAcBac (14).
Bioinformatic analysis of the C. jejuni and N. gonorrhoeae UDP-diNAcBac systems
resulted in the identification of a series of enzymes that catalyze the biosynthesis of UDPdiNAcBac in the AYE strain of A. baumannii. The ultimate protein glycosylation steps in these
pathways can be classified by the distinct oligosaccharyltransferases; PglB in C. jejuni (Nlinked) and PglO in N. gonorrhoeae (0-linked).
Comparative assessment of their respective
oligosaccharyltransferases supported the hypothesis that the AYE strain of A. baumanniiwas an
O-linked system as it bears a close resemblance to PglO in N. gonorrhoeae. Furthermore,
genetic analyses based upon sequence homology to their respective C. jejuni and N. gonorrhoeae
enzymes were consistent with a series of analogous proteins (WeeK, WeeJ, and Weel)
responsible
for
UDP-diNAcBac
biosynthesis
(Figure
3-1).
Additionally,
a
phosphoglycosyltransferase (WeeH) that catalyzes the transfer of phospho-diNAcBac to an
undecaprenol phosphate (Und-P) polyisoprenyl carrier was also identified.
Given that less
virulent strains of A. baumannii exist in nature, these genomes were searched for the existence of
92
this biosynthetic pathway to determine its prevalence. Whereas multiple strains contained this
particular pathway, the antibiotic susceptible A. baumannii strain (ATCC 17978) did not.
Instead, this strain contains a distinct 0-linked glycosylation system with a core GalNAc sugar
anchoring a branched pentasaccharide (15).
(GlcNAc3NAcAOAc)
The terminal 0-acetylated glucuronic acid sugar
shares homology to a similar pathway in the PAO1
strain of
Pseudomonas aeruginosa (16) however these enzymes are absent in the AYE strain of A.
baumannii.
UDP-GlcNAc
UDP-4-keto
UDP-4-amino
UDP-diNAcBac
Und-PP-d.NAcBac
OH
HOO
AHH
N
7
NAD+ NADH
O-UDP
N
2
~
AcHN
ACHN
O-UDP
AcCoA
AH~CoA AcH
O-UDP
NI
UndP
UMP
O-UDP
O-PP-Und
H20
A. baumannil
C. jejuni
N. gonorrhoeae
W
PgIF
-AP
PgID
WeeJ
Weel
WeeH
PgJE
PgID
PgIC
PgIC
Pgl8-ATD
PgIB-PGTD
bifunctional enzyme
Figure 3-1. The UDP-diNAcBac biosynthetic pathway in the AYE strain of A. baumannii. C.
jejuni and N. gonorrhoeaepathways are shown for comparison purposes.
This chapter presents the expression, purification, and kinetic characterization of the
three enzymes (WeeK, WeeJ, and WeeI) responsible for the biosynthesis of UDP-diNAcBac in
the AYE strain of A. baumannii. We also determined the substrate specificity of the
phosphoglycosyltransferase (WeeH) that catalyzes the transfer of the UDP-activated bacterial
sugar onto an Und-P lipid carrier. Furthermore, the active site homology between 0- and Nlinked UDP-diNAcBac pathway proteins in the context of binding and catalysis was examined.
This work establishes the presence of the biosynthetic machinery necessary for the production of
the UDP-diNAcBac nucleotide sugar in A. baumannii. Biochemical characterization of the
93
UDP-diNAcBac biosynthesis pathway in A. baumannii is significant in the context of its
potential relationship to the more pathogenic and antibiotic resistant strains of this serious human
pathogen.
Results and Discussion
Expression and Purificationof WeeK, WeeJ, Weel, and WeeH
Full-length WeeK, WeeJ, Weel, and WeeH were cloned from the AYE genomic DNA
and ligated into the pET-24a(+) vector. Each protein contained an N-terminal T7 tag and a Cterminal His 6 tag for purification purposes. Following overexpression in E. coli BL21 RIL cells
and purification with Ni-NTA resin, multi-mg quantities were achieved for each protein from 1 L
of culture:
WeeK (3 mg), WeeJ (62 mg), Weel (81 mg), WeeH (2.9 mg).
WeeJ was
preincubated with excess PLP throughout the entire purification procedure. The stoichiometry of
bound PLP to WeeJ ratio was established as 0.9:1 based upon the extinction coefficient of the
cofactor (6600 M- cm'1) at an absorbance of 390 nm in 0.1 M NaOH. Membrane proteins
WeeK and WeeH were solubilized from the lipid membrane with Triton X- 100 detergent. Purity
for each protein was assessed by SDS-PAGE to be >95% (Figure 3-2). Full-length constructs
were confirmed through Western blot analysis probing with antibody for the T7 and His 6 tags.
Storage of proteins for >6 months at -80 C had no affect on enzyme activity.
94
kDa
781
55 1
1
2
3
4
5
6
Figure 3-2. SDS-PAGE gradient gel (4-20%) of a MW standard (lane 1), WeeK (lane 2),
WeeK 140 (lane 3), WeeJ (lane 4), Weel (lane 5), and WeeH (lane 6).
Functionaland Kinetic Characterizationof the Dehydratase WeeK
Activity of the full-length WeeK was first investigated with the previously characterized
C. jejuni dehydratase substrates NAD+ and UDP-GlcNAc. Substrate turnover was followed by
capillary electrophoresis utilizing the UV absorbance of uridine at 254 nm (Figure 3-3A). After
an overnight reaction at 1 pM of WeeK, < 5% of the UDP-4-keto product was observed. A
control reaction containing the UDP-4-keto sugar product in the same reaction buffer was run in
parallel to ensure that this component was stable for the duration of the assay. Steps were taken
to understand the very poor of activity for this enzyme. The presence of divalent metals (MgCl ,
2
MnC12, ZnC12, and CaCl2) yielded no improvement in product formation. Furthermore, varying
amounts of salt (NaCl and KCl) produced a similar result. WeeK was also unable to catalyze the
reaction containing the substrate pairs UDP-GaNAc/NAD+ or UDP-GlcNAc/NADP+.
Two
additional detergents were utilized for purification in the anticipation of stabilizing a soluble,
active
protein.
In both
cases,
n-dodecyl-p-D-maltopyranoside
cholamidopropyl)dimethylammonio]-1-propanesulfonate
95
(DDM)
and
3-[(3-
(CHAPS) resulted in less substrate
turnover than the aforementioned Triton X-100 purified material. Lastly, buffer and pH were
examined for their effect on dehydratase activity in an in vitro activity assay. Although alternate
buffer solutions yielded no improvement in activity, pH had a substantial effect. Increasing the
pH from 7.4 to 8.5 resulted in a 10-fold increase of the UDP-4-keto product formation. All
further experiments utilized the Triton X- 100 detergent purified protein at a buffer pH of 8.5.
,2
A)
C)
0
.
(C
21
23
25
27
29
31
time (minutes)
Figure 3-3. Electropherogram trace representing the WeeK (A), WeeJ (B), and WeeI reactions
(C). Each numbered peak corresponds to a specific analyte: (1) NAD+; (2) UDP-GlcNAc; (3)
UDP-4-keto; (4) UDP-4-amino; (5) UDP-diNAcBac; (6) AcCoA; (7) CoA.
WeeK is annotated in the NCBI protein database as a UDP-glucose 4-epimerase, which
would reversibly convert UDP-glucose into UDP-galactose. To test this, WeeK was incubated
with 1 mM of either UDP-glucose or UDP-galactose for 18 hours at room temperature in the
presence of 1 mM NAD+. Following analysis of these reactions by CE, it was concluded that
WeeK does not convert UDP-glucose/UDP-galactose to their respective C4 epimers.
Based
upon experimental evidence, we conclude that WeeK is not an epimerase and instead exhibits
only NAD+ dependent dehydratase activity
96
With a reliable and robust WeeK CE assay in place, steps were taken to measure the
kinetic rate constants of UDP-GlcNAc.
The UDP-GlcNAc was varied over a substrate
concentration of 200 - 0.8 pM at 1 mM NAD+.
UDP-4-keto formation was quantified by
integrating the area under the product peak from the CE electropherogram trace. The reaction
velocities generated from product formation were an average from two separate experiments.
These data were plotted versus substrate concentration with equation 1 (Figure 3-4A) to yield
the kinetic parameters in Table 3-1. For comparison, the previously calculated C. jejuni NAD+dependent dehydratase (PglF) values are included in the table. Inhibition of WeeK activity was
observed at high concentrations of substrate (> 500 pM) and therefore not included in the final
analysis.
(A)
(B)
0.18
0.24
0.16
0.22
0.2
0.14
-
2 0.12
0
0 0.1
-E
0
0
0.16
0
Q 0.14
o.0.08
D
00.18
0
a
o
0.06
0.12
.
01
D 0.08
1 0.06
0.04
0.04
0.02
0
0
0
40
80
120
160
[UDP-GlcNAc] (,M)
200
-
0
500
1000
1500
2000
[UDP-GIcNAc] (pM)
Figure 3-4. Representative Michaelis-Menten binding curves for the full-length WeeK
dehydratase (A) and N-terminal truncated construct (WeeK140) (B). Kinetic characterization
was conducted at 50 mM TRIS-HCl pH 8.5, 1 mM NAD+, 0.005 % Triton X-100, and varying
amounts of UDP-GlcNAc.
97
Table 3-1. Steady-state kinetic parameters for A. baumanniiand C. jejuni dehydratase enzymes.
dehydratase
substrate
Km (jAM)
kc (s)
WeeK
WeeK140
PgIF0
UDP-GIcNAc
UDP-GIcNAc
UDP-GIcNAc
5.8± 1.2
23 ± 4.5
7000
2.7 x 10- ± 1.2 x 10
4.7 x 10' ± 1.7 x 10~'
0.12
kcat/Km (M- 1 s1)
466
20
17
a Kinetic parameters published in reference 12.
Based upon previous studies with the C. jejuni PglF (11-12), the transmembrane domain
was removed through cloning to yield the soluble domain of WeeK.
This soluble construct
improves upon the low yield from full-length construct purification and allows for a facile way
to biosynthesize large amounts of UDP-4-keto. To define the optimal truncation site, a sequence
alignment with PglF 1o and examination of a hydropathy plot with TMHMM (17) resulted in the
removal of 140 amino acids from the N-terminus (WeeK 14 0). Expression and purification in the
pET-24a(+) vector resulted in a 28 mg/L of culture yield at > 95% purity by SDS-PAGE This
construct catalyzed the conversion of UDP-GlcNAc to UDP-4-keto, however at a reduced rate
with respect to the full-length protein. Kinetic characterization of UDP-GlcNAc for WeeK 14 0
was carried out (Figure 3-4B) and compared directly to the full-length construct in Table 3-1.
Functionaland Kinetic Characterizationof the Aminotransferase WeeJ
Capillary electrophoresis was initially utilized to confirm the conversion of UDP-4-keto
to UDP-4-amino by WeeJ (Figure 3-3B).
Since this readout allows for direct comparison of
substrates and products, standards of each UDP-sugar were run in parallel with this reaction.
UDP-4-keto sugar biosynthesized from the C. jejuni and N. gonorrhoeaepathways resulted in
the production of UDP-4-amino by WeeJ in both cases.
These results confirm that the A.
baumannii enzyme exhibits aminotransferase activity with same stereospecificity observed
previously from the C. jejuni (PglE) and N. gonorrhoeae(PglC) aminotransferases and confirms
98
our analysis that WeeK does not show epimerase activity. To determine the kinetic constants for
the substrates L-glutamate and UDP-4-keto, an in vitro assay coupling the production of UDP-4amino to the C. jejuni acetyltransferase Pg1D was developed.
reagent, generation of the TNB2- chromophore
(X4l2nm
=
In the presence of Ellman's
14,150 M-1 cm-') indicates that
acetylation of the UDP-4-amino sugar has transpired. Kinetic characterization of each substrate
occurred at saturating levels (10
x
Kn) of the other substrate to ensure the rate of reaction was
dependent only upon the concentration of varying substrate. Typical Michaelis-Menten kinetics
were observed for all concentrations of L-glutamate (200 - 1.6 mM) and UDP-4-keto sugar (4 0.03 mM) (Figure 3-5). Initial velocity measurements were averaged between two separate runs
and plotted to yield the final kinetic parameters in Table 3-2.
(A)
(B)
2.2
2
1.8
.
1.6
1.4
1
1.2
1
1.4
1.2
~
o
-
0.6
0.8
--
0.6
-
0.4
0.8
=L
_
0.4
OA2
0.2
-
0.2
~0
0
0
40
80
120
160
200
[L-Glu] (mM)
0
1000
2000
3000
4000
5000
[UDP-4-keto] (pM)
Figure 3-5. Representative Michaelis-Menten binding curves for the A. baumannii
aminotransferase WeeJ. Kinetic characterization was conducted at 50 mM HEPES pH 7.4, 0.05
% BSA, 0.00 1 % Triton X-100, 1 pM C. jejuni PglD, 400 gM AcCoA, 2 mM DTNB, and 400
nM WeeJ. L-glutamate (A) was varied in the presence of 10 x Km of UDP-4-keto. UDP-4-keto
(B) was varied in the presence of 10 x Km of L-glutamate.
99
Table 3-2. Steady-state kinetic parameters for A. baumannii,C. jejuni, and N. gonorrhoeae
aminotransferase enzymes.
aminotransferase
substrate
Km (JM)
kc (s)
kcat/Km (M -1 s1)
WeeJ
WeeJ
PgIE
PgIE
PgICa
PgICG
UDP-4-keto
L-glutamate
UDP-4-keto
L-glutamate
UDP-4-keto
L-glutamate
1003 110
25,000 1900
366 57
11,000 ± 340
233 ±35
4900 ± 900
0.030 ± 0.002
0.17 ± 0.004
2.4 0.1
0.028 ± 0.0003
0.038 0.001
0.025 ± 0.01
30
6.7
6600
2.6
164
5.1
a Kinetic parameters published in reference 14.
Functionaland Kinetic Characterizationof the Acetyltransferase WeeI
Similar to the characterization of WeeJ, a capillary electrophoresis assay confirmed that
Weel produced UDP-diNAcBac from the UDP-4-amino substrate generated by the C. jejuni and
N. gonorrhoeaepathways (Figure 3-3C). A continuous, in vitro assay relying on the generation
of the TNB2- chromophore from Ellman's reagent was again employed.
Initial attempts to
determine kinetic parameters for UDP-4-amino resulted in an atypical sigmoidal binding curve
suggestive of positive cooperativity (Hill coefficient = 2). Further experiments were applied to
establish that WeeI activity was dependent upon MgCl 2.
In the presence of EDTA, Weel
retained the ability for substrate turnover establishing that MgCl 2 is not essential for catalytic
function of this enzyme.
Comparison of the reaction rates indicated a 6.4-fold increase in
activity with the addition of 5 mM divalent magnesium. The presence of MgCl 2 resulted in
typical Michaelis-Menten kinetics over a range of AcCoA (3 - 0.02 mM) and UDP-4-amino (10
- 0.08 mM) concentrations (Figure 3-6). As a result of the poor affinity of UDP-4-amino to
WeeL, the apparent AcCoA Km was determined at a UDP-4-amino concentration of 4 mM.
Kinetic parameters listed in Table 3-3 are the outcome of initial velocity measurements repeated
in duplicate.
100
(A)
00
.2
12
(B)
220
200
180
160
140
120
100
500
400
S
80
.2
E
60
40
300
200
100
20
0
0
0
400
800
1200
1600
2000
0
2000
[AcCoA] (pM)
4000
6000 8000
[UDP-4-amino] (1,M)
10000
Figure 3-6. Representative Michaelis-Menten binding curves for the A. baumannii
acetyltransferase Weel. Kinetic characterization was conducted at 50 mM HEPES pH 7.4, 5
mM MgC 2 , 0.05 % BSA, 0.00 1 % Triton X-100, 1 mM DTNB, and 1 nM Weel. AcCoA (A)
was varied in the presence of 1 x Km of UDP-4-amino. UDP-4-amino (B) was varied in the
presence of 10 x Km of AcCoA.
Table 3-3. Steady-state kinetic parameters for A. baumannii, C. jejuni, and N. gonorrhoeae
acetyltransferase enzymes.
acetyltransfe rase
substrate
KM (yM)
Weel
Weel
PgID
PgID
PgIB-ATD
PgIB-ATD
UDP-4-amino
AcCoA
UDP-4-amino
AcCoA
UDP-4-amino
AcCoA
2520± 540
78.9 28
274 6.4
295 2.8
99.0 ± 7.1
286 ± 35
kt (s-)
5.1 ± 0.08 x 10 5
1.3 ±0.06 x 10 5
8.0 ± 1.6 x 10s
6.1 ± 1.0 x 10 5
7.2 ± 0.8 x 104
5.0 ± 0.7 x 104
kat/Km (M4 s4 )
2.0 x 108
1.6 x 10 9
2.9 x 10 9
2.1 x 10 9
7.3 x 108
1.7 x 108
The A. baumannii enzymes WeeK, J, andI produce UDP-diNAcBac
Previous studies have focused on characterizing the UDP-diNAcBac glycosylation
pathway enzymes in N-linked (C. jejuni) and 0-linked (N. gonorrhoeae) systems (12,14). The
finding that the AYE strain of A. baumannii contains an 0-linked bacillosamine biosynthesis
pathway further confirms the connection between the sugar and glycoconjugates that may be
101
involved in pathogenicity while adding to the growing number of bacteria with this system. This
strain of A. baumannii is of particular interest due to its extreme antibiotic resistance.
Understanding the virulence factors associated with A. baumannii is of great importance to the
medical community particularly as a potential new target in efforts to address the ever-growing
resistance to current antibiotics.
The A. baumannii dehydratase (WeeK), aminotransferase
(WeeJ), and acetyltransferase (WeeI) were individually investigated for their ability to catalyze
their anticipated substrates and characterized kinetically.
Activity assays with UDP-sugar
substrates generated from the C. jejuni and N. gonorrhoeae glycosylation pathways confirmed
that the A. baumannii enzymes utilize the same general mechanism to produce UDP-diNAcBac.
From a kinetic viewpoint, it appears that this pathway uses a similar overall strategy employed
by homologous enzymes in C. jejuni and N. gonorrhoeae. Specifically, the committed step in
UDP-diNAcBac biosynthesis is controlled by the first and rate-limiting enzyme on this pathway
(WeeK), whereas the final acetylation of UDP-4-amino by Weel represents the most catalytically
efficient reaction. Collectively, these enzymes are responsible for the biosynthesis of the highlymodified bacterial NDP sugar, UDP-diNAcBac. Although the main focus of this chapter has
been on the characterization of the early pathway enzymes responsible for UDP-diNAcBac
biosynthesis, composition of the final protein-linked oligosaccharide is still unknown.
Substrate Specificity of the PhosphoglycosyltransferaseWeeH
Substrate specificity of WeeH was determined using a previously established
radioactivity-based assay (14,18).
This method relied on the transfer of a tritium-labeled
phospho-sugar (from the UDP-activated substrate) to a hydrophobic undecaprenyl phosphate.
The polyprenyldiphosphate-monosaccharide product (Und-PP-diNAcBac) is extracted into the
102
organic phase separating it from the aqueous soluble unreacted radioactive UDP-diNAcBac. In
total, five UDP-sugars were analyzed for their ability to act as a substrate for WeeH. This
phosphoglycosyltransferase exhibited clear selectivity for UDP-diNAcBac over all other UDPsugars (UDP-GlcNAc, UDP-GalNAc, UDP-Glc, UDP-Gal) (Figure 3-7).
18
16
14
12
10
UDP-diNAc8ac
--
4-- UDP-GicNAc
8 --
UDP-GaINAc
-1-
UDP-GIc
---4-
- UDP-Gal
2
0
2................
0
10
20
..
30
40
50
60
time (minutes)
Figure 3-7. Specificity of the A. baumannii phosphoglycosyltransferase WeeH in the presence
of an assortment of UDP-sugars. Aliquots of the reaction containing 2 nmol Und-P, 1% Triton
X-100, 3% DMSO, 30 mM TRIS-acetate pH 8.0, 50 mM MgCl 2, 40 pM UDP-sugar, 20 pM
UDP-GalNAc (20 mCi/mmol), 4.5 pM C. jejuni PglA, and 200 nM WeeH were taken over a 30
minute time course. Error bars represent standard deviation from triplicate measurements.
Active Site Comparison Between 0- and N-linked UDP-diNAcBac Pathway Proteins
To better understand the relationship between UDP-diNAcBac pathways in the 0- and Nlinked protein glycosylation systems, active site sequence homology was investigated with an
emphasis on the aminotransferase and acetyltransferase
gonorrhoeae, and A. baumannii pathways.
enzymes in the C. jejuni, N.
The NAD+-dependent dehydratase enzymes were
103
excluded from this comparative analysis since structures of these enzymes have not yet been
determined.
To define the binding pocket for the aminotransferase PglE from the N-linked
glycosylation pathway, the structural analysis of a homologous enzyme from Helicobacterpylori
(PseC) was employed (19).
A crystal structure of PseC with the bound external aldimine
(2FNU) was aligned with the PLP bound PglE (1061) crystal structure (20) to define the active
site residues. PseC catalyzes the transamination reaction at the C4 position of a UDP-activated
sugar similar to the PglE UDP-4-keto substrate.
In this case, the only variation is the
stereochemistry of the methyl group at the C5 position (P-L-arabino-hexulose as opposed to CD-xylo-hexulose). Alignment of the PseC and PglE structures resulted in a root mean square
(rms) value of 1.1 A for the monomer and a rms value of 0.6 A for the active site residues. A
second aminotransferase structure from Pseudomonas aeruginosa (WbpE) containing the bound
external aldimine was used to provide further support for the PseC findings (21). The residues
defining the PglE active site were identical in both examples.
The aminotransferase binding
pocket was defined to a 5 A distance from the external aldimine-bound molecule in the structural
visualization program PyMOL (Figure 3-8) (22). Sequence alignment of the N. gonorrhoeae
(PglC) and A. baumannii (WeeJ) aminotransferases to PglE was accomplished using Clustal
Omega. The final alignment among the three aminotransferases was visually represented by the
program Jalview (Figure 3-9) (23). Although the overall sequence identity between 0-linked
and N-linked aminotransferases is relatively low, the enzymes from the 0-linked glycosylation
pathway (PglC and WeeJ) exhibit exceptionally high homology (67% sequence identity) (Table
3-4). This observation is even more apparent when comparing the residues within the active site.
Not surprisingly, the catalytic lysine residue responsible for product formation is completely
104
conserved among the C. jejuni (K184), N. gonorrhoeae (K185), and A. baumannii (K185)
aminotransferases (Figure 3-10). Six of the ten PseC-binding residues interact with PLP and
homologous amino acids can be accounted for in the PglE structure. Of note, the PglE residues
D155, S179, N227 and T57 are completely conserved among the three aminotransferases and
associate directly with PLP. Although E158 and N181 are not identical in the N. gonorrhoeae
and A. baumannii model, a similar role can be hypothesized by glutamine at both positions.
Only Y316 in the PseC structure has direct interaction with the sugar moiety. No obvious
counterpart can be identified in the sequence alignment with PglE, PglC, and WeeJ.
Figure 3-8. Surface representation of the C. jejuni PglE binding pocket. The crystal structure of
the homologous aminotransferase PseC bound to the PMP-UDP-L-AltNAc external aldimine
(2FNU) was utilized to define the active site. Following alignment with the PLP-bound PglE
crystal structure (1061), amino acids within 5 A of the external aldimine were identified as
binding-pocket residues.
105
9
.
1
100IW"
19
4
MVatUALI040
3
V1
1
3'
£1T* v
3
.v
1
3 d
3 ...
h.
k' k1a1
T A0
M
-*4-*
-------i4 VV
tfta-es
1.
-9
----fM6 -P -G
GV W-- E i
09=110
34 131333031411
SM
9
AU'04:1# a16-.u
4
..
1
.3l
Q:X
tIllxlI*
rl' A066a~i
....
..'..9.
3334*fV
01wx*91411,
a
S
------ - -----l-------
310
..
~gaA G
t.g y ge t e
..
.
--
--.
t Qi
---
.......
f
inska.-sra
...
.. .....
--
.3401,441314
4143'
3 mkth
...
C
......
.
3
4...
Figure 3-9. C. ]ejuni, N. gonorrhoeae,and A. baumannii aminotransferase alignment. Enzymes
were aligned with the program Clustal Omega and visually represented by Jalview. Active site
residues (highlighted in red) were assigned utilizing the PseC external aldimine (2FNU) and
PLP-bound PglE (1061) crystal structures.
Table 3-4. Sequence identity for A. baumannii (A b), C. ]ejuni (C]j), and N. gonorrhoeae (Ng)
aminotransferase enzymes.
.wminotransfras
pIr
PgIE(Cj)/PgIC(Ng)
PgIE(Cf)/WeeJ(Ab)
WeeJ(Ab)/PgIC(Ng)
totat prt4i
%
external aldimne active slte %)
38
38
90
22
18
67
1
106
Asn 239 (Ab)
Thr 0 (Ab)
Asn239(Ng)
Asn22.7
Thr 0 (Ng)
Thr 57
Gln183(Ab)
Gln 183(Ng)
Asn,81
0
Asp,5 6(Ab)
Asp1,5 (Ng)
H2N
OH
H2N
' .Ser
O
17g
,;-
P0
0
oLys
Aspiss
5
18 (Ab)
Lys1 85,(Ng) 0
HN~
O)
6 H3N 0
NH
Lys1 4
0 -N
0
Glu, 58
Gin159(Ng)
Gln1 59(Ab)
Seri80 (Ab)
Ser180(Ng)
OH
NH
H
-0-
0
O
HO
OH
OH
/
Phe 82
Phe8 s(Ng)
Phess(Ab)
NH
Trp332
Trp345 (Ng)
Trp345(Ab)
Ng: N. gonorrhoeae
Ab: A. baumannii
Tyr219
Trp231(Ng)
Trp231(Ab)
Figure 3-10. Illustration of the relevant amino acids responsible for the aminotransferase
binding pocket in PglE(C]). Residues labeled as Ng and Ab represent analogous positions in
PglC(Ng) and WeeJ(Ab), respectively.
In order to establish the acetyltransferase binding pockets of Weel and PglB-ATD, the C.
jejuni PglD crystal structures containing bound UDP-4-amino (3BSS) and AcCoA (3BSY) were
utilized (25). Each binding site was defined as for the aminotransferases, with a 5 A distance
surrounding the respective substrate (Figure 3-11). Sequence alignment and visualization were
again accomplished utilizing Clustal Omega and Jalview (Figure 3-12).
Interestingly, the
sequence identity of both active sites varies highly depending upon the specific substrate-binding
pocket and acetyltransferase pair being examined (Table 3-5). The AcCoA binding site exhibits
more homology between the 0-linked (WeeI/PglB-ATD) pathway enzymes. The majority of
interactions between AcCoA and protein side chains occur at the carbonyl oxygen of the
107
thioester (Figure 3-13A).
The nucleotide and pantetheine moieties of AcCoA are held in the
substrate-binding site by a network of water molecules and backbone interactions from 1155 and
G173, which are represented by threonine and glycine in the O-linked acetyltransferase enzyme
sequences. Unexpectedly, the UDP-4-amino binding pocket shares more similarity between Nlinked PglD and O-linked PglB-ATD. Interactions between the pyranose C4 amine (H125), the
ribosyl 3' hydroxyl group (D35), and the uridine imide (D36) are exclusively conserved (Figure
3-13B). The uracil of the NDP sugar is stabilized by a similar hydrophobic pocket (YlO, 155,
160, and 164) in PglB-ATD and WeeI model structures. There are two major differences in the
proposed Weel UDP-4-amino binding pocket with respect to PglD. The asparagine at position
162 that interacts with the carbonyl oxygen of the pyranose C2 acetyl group is replaced with a
glycine (G175). Importantly, the H15 residue that interacts with the sugar substrate P-phosphate
moiety and inserts into the pocket to accommodate UDP-4-amino is replaced with a
phenylalanine (F13).
Table 3-5. Sequence identity for A. baumannii (Ab), C. jejuni (C), and N. gonorrhoeae(Ng)
acetyltransferase enzymes.
PgID(C])/PgB-ATD(Ng)
34
37
64
PgID(Cj)/Wee(Ab)
26
34
48
WeeI(Ab)/PgB-ATD(Ng)
26
56
42
108
Figure 3-11. Surface representation of the C. jejuni Pg1D binding pocket. Amino acid residues
within 5 A of the UDP-4-amino (top left) and AcCoA (bottom right) substrates were classified as
contributing to the acetyltransferase active site. Binding pocket identification relied on the UDP4-amino (3BSS) and AcCoA (3BSY) bound PglD crystal structures.
PP'D(PY.495
PgIO
)ATWID6-413
WMW(A
210
6
IMARTEIY I
I96AGNRKLAV
MM IGV
LVCEVAKNM-G...YKECIFf
KVVAELAAAL-Ga-TYGE
KEVMPLVRQFPTLSKEQFAF
MK------.-
--..
FESTLPKYOFFI
FPVGTTLLLENSLSPEFDIT
TTLNGYPVLSYLDFIS--KPADHKAVTI
iVF
SVN
RO
EK
YQKISENGFKIVNLINKSALISPSAIVEENAG9g
ENAAALGFKLPVLIHPDATVSPSAIGQG--304
SLLEKDGVQHLAVOSTNTVILDE--VEGEGIOO
Caeon
CGeusy
--. KLIUMt
r
MA-T-KI -VYGASGHGKVV.-LA .
PgID(Cy-195
PMg)OJATW96-413
W0.)M
I
21
97
305
110
SLLCPFTCL TSNI 8KFF
5 ii:
.a-s.~
-
57
"465
36~
1
IL M
SL-MP--V--A--KI-
G. --
SSVi
1LIMPYVVINAKAKIEKGV
SVVMAKAVVQAGSVLKDG
AATV
IYSYV
51
i
_IM1
Y-E
ECV
-FLDD--
IGEF
DCLLDAFY
OCV
NGDY
547
-G
'K
AKC
IHIEDHA
548562
I
NG-PV.
AKC
AIL
ma
M
t_
-
.0 4645.
0.......-
A
-
A
8.07
PDKPLGKGA-
5
---
5
450
'7O
len
mM
-
-TTVGSGV
w-
GVI-Nt-S-V-HDCVIG-FVH-SPGAK
-.--
SL---.-L8LADDS-
RIGEEN
17
dI
D-TIAIGNN-IR-KI
..
,.-
eiei
-ON--0-.---IGTGAC--
--..
-
. L--G-G--
kuin
a
GFK-
-
LINI
-A-ISPSAIVE-G-G
KNODEK
P
K
--.
--19U
TGKNPKTGTA413
KSVPAG
R
ERK-..--..216
56457
,
S7
.....
aass
at
it ML
1,k~
-G-GAV-VK--P.G-TVVGNPAK-L.-K
Figure 3-12. C. jejuni, N. gonorrhoeae,and A. baumanniiacetyltransferase alignment.
Enzymes were aligned with the program Clustal Omega and visually represented by Jalview.
The AcCoA binding pocket is highlighted in red, the UDP-4-amino active site
is denoted in blue,
and shared residues between both sites are highlighted in purple. The AcCoA and UDP-4-amino
binding pockets were assigned utilizing the PglD crystal structures 3BSY and 3BSS respectively.
109
(A)
H2 0
H20
e H20
I/
NH 2 ,
H
H
Asnils
16
Asn 32(Ng)
OH,
+Oc
0
N
N~
N
His342 (Ng)
Thr 147(Ab)
3-120
N
N
NH
101s%N
Thra36(Ng)
Thr 1, 6 (Ab)
HIS134
HH
HO'
H
N
N
20
20
-6
\
0-
Asn 131(Ab)
'H
/
0
N
Ng: N. gnorrhoeae
H20
Gly17a
Ala381(Ng)
Met 1g1(Ab)
(B)
H20
iAb: A. baumanniLJ
ValIS(A b)
1e 7a(Ab)
Val 2 s(Ng)
HN
116270(p n)
11H60
HN
0
HN
8[4
His 125
His 3 3n(Ng)
O
His 13a(Ab)
H2 p
0
HO
NH--'O-P-0-P-O HHO
0
N
01
O
NH
HO
N
Gly 175(Ab)
H
NN
K
H
N,
0
e
00
HN-<
Asp36
Asp232(Ng)
AsP3 8(Ab)
Ser13
His1 s
Gly14
Phe13(Ab)
Gly 20g(Ng)
GIy 12(Ab)
His 21 (Ng)
1e 20 (Ng)
Tyr10
HO
NH
NH 2 -
Tyr8 (Ab)
N
H
HN/N
Asn162
Gtn 3 ?o(Ng)
Leug(Ab)
11e2 74(Ng)
G1y 204(Ng)
Ser 1 (Ab)
Asps
Asp 23(Ng)
Asp 37 (Ab)
Figure 3-13. Illustration of the relevant amino acids responsible for the AcCoA (A) and UDP-4amino (B) binding pockets in PgLD(C). Residues labeled as Ng and Ab represent analogous
positions in PglB(Ng)ATD and WeeI(Ab), respectively.
110
UDP-diNAcBac enzyme diversity in N- and O-linked glycosylation
Previous kinetic characterization of the C. jejuni dehydratase Pg1F resulted in a Km of 7
mM and a kcat of 0.12 s~1 (12). Surprisingly, the A. baumannii WeeK binds to UDP-GlcNAc with
a significantly higher affinity (1000-fold) however catalyzes this reaction at an appreciably
reduced rate (44-fold). As a result, WeeK is a catalytically more efficient enzyme (kcat/Km = 466
M-1 s-1) relative to PglF (kcat/Km = 17 M-1 s-). The two NAD+-dependent dehydratases have a
sequence identity (31%) that is similar to other homologous proteins on this pathway yet exhibit
contrasting kinetic parameters. It is interesting that WeeK binds UDP-GlcNAc with such high
affinity since this substrate is utilized for many other pathways within the cell including biofilm
formation (25), lipooligosaccharide (26), and various cell envelope components.
To define the WeeJ aminotransferase UDP-sugar binding pocket, a homologous structure
from Helicobacterpylori (PseC) was employed (Figure 3-8). When comparing the sequence
identities of the three aminotransferases (overall and active site), a trend emerges (Table 3-4).
Although the N-linked C. jejuni aminotransferase catalyzes the same reaction, it shares little in
identity to its O-linked relatives.
This observation however is not reflected in the catalytic
efficiency for the substrate L-glutamate (Table 3-2) as all three aminotransferases share
comparable kinetic parameters.
When evaluating catalytic efficiency for the UDP-4-keto
substrate in relation to its sequence, a different story unfolds.
PglE has an elevated rate of
turnover in comparison to the O-linked pathway proteins that result in a 39-fold (PglC) and 217fold (WeeJ) increase in catalytic efficiency. In contrast, PglC and WeeJ share a similar albeit
lower efficiency for UDP-4-keto catalysis.
The final step in the biosynthesis of UDP-diNAcBac is catalyzed by the acetyltransferase
Weel. The AcCoA and UDP-4-amino binding pockets have been well established through C.
111
jejuni PglD crystallographic analysis (Figure 3-11) (24). Whereas a clear trend was established
when comparing 0-linked versus N-linked aminotransferases, a different picture emerges with
the acetyltransferase enzymes. With respect to sequence identities, PglD and PglB-ATD from N.
gonorrhoeaeshare greater homology in all aspects apart from the AcCoA binding pocket (Table
3-5). This is a surprising observation when relating these results with homology between both
the dehydratases and aminotransferases. It is apparent that the gene products of UDP-diNAcBac
biosynthesis in A. baumannii were acquired collectively as they are located consecutively in the
same operon. It is not currently understood why changes to the acetyltransferase binding pocket
may have occurred with respect to the substrate affinity and fitness of this particular strain of the
bacterium. The similarity in AcCoA kinetic parameters (Table 3-3) is directly reflected in the
conserved residues for cofactor binding. Unexpectedly, PglD and PglB-ATD share a higher
homology in their UDP-4-amino binding sites relative to Weel (Table 3-5).
residues
that
interact
acetyltransferases.
with the UDP-sugar
are
strictly
conserved
Many of the
across
all three
Surprisingly, Weel exhibits poor affinity for UDP-4-amino with respect to
the other acetyltransferases (>10-fold). Only two major differences are observed in the sugarbinding pocket (Figure 3-13B). First, an asparagine side chain that interacts with the carbonyl
oxygen of the pyranose C2 acetyl group in the PglD structure is replaced with glycine. However,
an adjacent glutamine in WeeI may serve as a hydrogen-bond donor depending upon its location
within the tertiary structure.
Second and seemingly more noteworthy is the replacement of
histidine (H15) with phenylalanine in Weel (F13). This residue interacts with the sugar substrate
P-phosphate and repositions in the pocket to accommodate UDP-4-amino. Therefore, this amino
acid is positioned to act as a gatekeeper for sugar substrate binding. The hydrophobicity and size
112
of this residue with respect to histidine may partially explain the poor affinity of this substrate
towards Weel (Table 3-3).
Enzymatic flux through the UDP-diNAcBacpathway
To eliminate adverse byproducts and off-pathway reactions, nature often exploits
substrate channeling, wherein intermediates are shuttled to successive enzymes to increase the
efficiency of a particular pathway. In the case of the UDP-diNAcBac biosynthesis, the initial
NAD+-dependent dehydratase is catalytically inefficient with respect to the subsequent enzymes
in the pathway (12,27).
Nevertheless, interpretation of these assay results must be viewed with
some caution since the in vitro analysis conditions may not reflect the true kinetic potential of the
membrane-bound dehydratase in its natural cellular environment. WeeK appears to function as a
gatekeeper to UDP-diNAcBac production as the formation of the UDP-4-keto sugar is the ratelimiting step.
In order to drive this pathway forward, downstream enzymes appear to be tuned
to increase their catalytic efficiency. In all three acetyltransferases examined in this chapter, the
catalysis of UDP-4-amino to UDP-diNAcBac is significantly elevated with respect to the earlier
pathway enzymes. The high catalytic efficiency of the acetyltransferase drives the production of
UDP-diNAcBac by rapidly consuming UDP-4-amino and in turn promotes the conversion of
more UDP-GlcNAc to UDP-4-keto.
A similar effect can be observed in the biosynthesis of
UDP-ManNAc(3NAc)A in P. aeruginosa (16); additionally, this type of flux is prevalent in
metabolic pathways. For example, glycolytic flux in bacteria utilizes a feed-forward loop where
high levels of fructose- 1,6-bisphosphate signal for increased activity of glyceraldehyde 3phosphate dehydrogenase (GAPDH) (28-29). Pyruvate kinase (PK) is also part of this loop and
its activity is responsible for flux into the lower half of glycolysis. Metabolic flux control is also
113
elicited through the pyruvate node during anaerobic growth in Escherichiacoli to maintain redox
balance in the cell (30). UDP-diNAcBac biosynthesis is yet another example of flux created by a
highly active enzyme at the terminal end of the pathway that can overcome the deficient catalytic
efficiency of preceding enzymes.
Conclusions
In conclusion, these studies establish details of the characterization of the early UDPdiNAcBac pathway proteins WeeK, WeeJ, Weel, and WeeH. Comparison to the analogous C.
jejuni and N. gonorrhoeae systems has resulted in an understanding of the similarities and
differences between N- and O-linked glycosylation pathway enzymes.
Although a direct
correlation between pathogenicity and O-linked glycosylation in the AYE strain of A. baumannii
remains to be elucidated, this work highlights an analogous pathway previously shown to
diminish infectivity when disrupted.
Future work focusing on inhibiting the A. baumannii
enzymes responsible for UDP-diNAcBac biosynthesis will strengthen the correlation between
pathogenicity and bacterial glycosylation. The rise of this multi-drug resistance strain in the
hospital environment is cause for alarm and makes the search for novel virulence targets all that
more important. The enzymes responsible for UDP-diNAcBac biosynthesis may well represent
novel targets in the struggle against A. baumannii resistance.
Acknowledgments
I wish to thank Dr. Angelyn Larkin, Dr. Meredith Hartley, and Austin Travis for critical reading
of this chapter and useful discussions regarding experimental data.
114
Experimental Procedures
Common Materials
All chemicals were purchased from Sigma-Aldrich unless otherwise stated. The UDP-4keto, UDP-4-amino, and UDP-diNAcBac sugars were biosynthesized as described previously
from the C. jejuni enzymes PglF, PglE, and PglD (12).
Cloning, Expression, and Purification
The WeeK, WeeJ, WeeI and WeeH genes were amplified via the polymerase chain
reaction (PCR) from the genomic DNA of the AYE strain of Acinetobacter baumannii (ATCC)
using the primers found in Table 3-6 (4). BamHI and XhoI restriction sites were engineered to
facilitate cloning of each construct into the pET-24a(+) vector (Novagen). Amplifications were
accomplished with the PfuTurbo DNA Polymerase (Stratagene) as described by the
manufacturer. Amplicons were purified and double-digested with BamHI and XhoI restriction
enzymes (NE Biolabs). Digested inserts and linearized vectors were fractionated by agarose gel
electrophoresis and purified with the Wizard SV Gel and PCR Cleanup Kit (Promega).
Ligations were conducted with the T4 DNA ligase kit (Promega) using a 15 minute incubation at
room temperature.
Sequencing by Genewiz (Cambridge, MA) confirmed the presence of all
gene products.
The pET24a(+) plasmids containing each gene were used to transform
BL21(DE3) RIL competent cells (Stratagene).
E. coli
1 L of LB media containing 50 pg/mL
kanamycin and 30 pg/mL chloramphenicol was inoculated with 8 mL of an overnight culture of
cells. The cells were then allowed to grow at 37 C while shaking until they reached an optical
115
density of -0.8 (X = 600 nm). The culture was cooled to 16 0C and induced with 0.5 mM iso-pD-thiogalactosylpyranoside (IPTG). After incubating for 18 hours with shaking, the cells were
harvested (2600 x g) and stored at -80 0C until needed.
Each protein purification step was carried out at 4 C.
The cell pellet (~3 g) was
resuspended in 40 mL of 50 mM HEPES pH 7.4/100 mM NaCl/30 mM imidazole (Buffer A)
and then lysed by sonication. WeeK resuspension buffer was supplemented with 200 pM NAD+
and WeeJ with 200 pM pyridoxal 5'-phosphate. For WeeJ and Weel, the lysate was cleared by
centrifugation (145000 x g, 60 min) and added to 2 mL of Ni-NTA resin (Qiagen). The slurry
was allowed to tumble for 3 hours and then packed into a fritted PolyPrep column (Biorad). The
resin was washed with 20 column volumes of Buffer A and then eluted with a buffer containing
50 mM HEPES pH 7.4/100 mM NaCl/ 300 mM imidazole (Buffer B). Fractions containing the
purified protein by SDS-PAGE were pooled, dialyzed against 50 mM HEPES pH 7.4/100 mM
NaCl (Buffer C) to remove the imidazole, and then supplemented with 15% glycerol. Protein
concentrations were calculated based upon the predicted extinction coefficients at X = 280 nm.
Aliquots of the protein were stored at -80 C until needed.
The membrane-associated proteins WeeK and WeeH required additional purification
steps to those of the soluble proteins presented above. Following sonication, cellular debris was
cleared at 8000 x g (45 min) and the supernatant was transferred to a clean centrifuge tube.
Further centrifugation took place at 145000 x g (60 min) to pellet the cell envelope fraction
(CEF). The CEF was homogenized in 5 mL of Buffer A supplemented with 1% Triton X-100.
This solution was allowed to tumble overnight and then subjected to centrifugation 145000 x g
(60 min) to remove any unsolubilized material. The supernatant was combined with 2 mL of NiNTA resin and tumbled for 3 hours. The slurry was added to a PolyPrep column and washed
116
with 20 column volumes of Buffer A supplemented with 0.1% Triton X-100. The protein was
eluted from the resin with Buffer B containing 0.1% Triton X- 100.
through dialysis with Buffer C/0.1% Triton X-100.
conditions resulted protein precipitation.
Imidazole was removed
In the case of WeeH, these dialysis
This issue was remediated by increasing the NaCl
concentration in the dialysis buffer to 350 mM. Due to the addition of a UV-active detergent,
protein concentrations were calculated with the DC Protein Assay Kit (Biorad). Purified protein
was supplemented with 15% glycerol and stored at -80 C.
Table 3-6. Constructs, accession numbers, and oligonucleotides used for this study.
construct
vector
accession number
WeeK
pET-24a(+)
YP_001715522.1
pET-24a(+)
YP_001715522.1
WeeJ
pET-24a(+)
YP_001715523.1
Weel
pET-24a(+)
YP_001715524.1
WeeH
pET-24a(+)
YP_001715525.1
WeeK1
40
primer (5'
fwd:
rev:
fwd:
rev:
fwd:
rev:
fwd:
rev:
fwd:
rev:
> 3')
CGCGGATCCGTGAAAAAGATTATTTATC
GTAAAGATATTATGGTTAATCTCGAGCGG
GCGGATCCATGTTACAGACTGGTGAAGAG
TAGTAAAGATATTATGGTTAATCTCGAGCGG
CGCGGATCCATGTTAAACACTGCATTTG
GTTAATAGAGCTTTACAATCACTCGAGCGG
CGCGGATCCATGACAATGATTATTGG
CAAGAATTTTAGAAAGAAAGCTCGAGCGG
CGCGGATCCGTGTTAAAACGTTTACTTG
GAAGGAAATAGAGAAAAAACTCGAGCGG
Dehydratase (WeeK) Activity Assay
Kinetic characterization of WeeK utilized capillary electrophoresis (CE) to directly
determine UDP-4-keto product formation.
The assay contained 50 mM Tris-HCl pH 8.5,
0.005% Triton X-100, 1 mM NAD+, and varying amounts of UDP-GlcNAc. The reaction was
initiated with 0.5 pM WeeK and time points were taken over a time span of 180 minutes at 25
C. The reaction was stopped by filtration through a 10K MWCO membrane (Millipore) to
remove the enzyme and the filtrate was injected for 15 s at 30 mbar on a P/ACE MDQ system
(Beckman Coulter). Separation of analytes occurred at 20 kV over a 45 minute time period on a
117
bare silica capillary (75 pm x 80 cm) with a 25 mM sodium tetraborate (pH 9.3) running buffer
and monitored at a ) = 254 nm. Substrate and product peaks were manually integrated utilizing
the Beckman 32 Karat software suite.
Steady-state rate parameters were calculated from
equation 1 using the program GraFit 6.0.12 (Erithacus Software). The kinetic parameters are a
result of duplicate measurements at each substrate concentration.
V = Vmax[S]/(Km + [S])
(1)
Aminotransferase (WeeJ) Activity Assay
Calculation of kinetic constants was carried out as described previously (3). Briefly, the
generation of the UDP-4-amino product from the WeeJ reaction was coupled to an excess of the
C. jejuni acetyltransferase PglD and the activity of WeeJ was determined by following the
production of CoASH at 25 0C. In a flat, clear bottom 96-well plate (Falcon) was added 50 mM
HEPES pH 7.4, 0.05% BSA, 0.001% Triton X-100, 1 pM PglD, 400 pM AcCoA, and 400 nM
WeeJ. The substrate concentrations of L-glutamate and UDP-4-keto were varied separately to
determine kinetic parameters utilizing initial velocity measurements while keeping the other
substrate at saturation (10
x
Km). Reactions were initiated with the L-glutamate substrate and
quenched with 20% n-propanol, 2 mM DTNB (5,5'-dithio-bis-(2-nitrobenzoic acid), and 1 mM
EDTA over a 30 minute time period, which provides a spectroscopic readout for the production
of CoA.
The absorbance at 415 nm was monitored on an Ultramark EX microplate imaging
system (BioRad). Reactions were performed in duplicate and a blank reaction without WeeJ was
set up as a background control and subtracted from the final observed reaction rate.
118
Acetyltransferase (WeeI) Activity Assay
Kinetic characterization of Weel was carried out using a previously modified procedure
(14). CoASH generation resulting from the acetyltransferase reaction carried out by Weel was
monitored in the presence of Ellman's reagent (DTNB) through the generation of the TNB2
chromophore in a continuous fashion. To a flat, clear bottom 96-well plate (Falcon) was added
50 mM HEPES pH 7.4, 5 mM MgCl 2 , 0.05% BSA, 0.001% Triton X-100, 1 mM DTNB, and 1
nM Weel. Reactions were completed in duplicate and initial rates were measured in the linear
portion of the reaction curve over a 5 minute time period at 25 C. The substrate concentrations
of AcCoA and UDP-4-amino were varied separately to determine kinetic parameters using initial
velocity measurements while holding the other substrate at a saturating level.
Due to the
solubility and poor binding of UDP-4-amino to Weel, the AcCoA Km was determined at Km of
the sugar substrate. A background control in the absence of UDP-4-amino was subtracted from
each reaction rate.
Phosphoglycosyltransferase(WeeH) Activity Assay
A radioactive assay (31) was utilized to establish the UDP-sugar specificity of WeeH. In
a 1.5 mL eppendorf tube, 2 nmol of undecaprenyl phosphate was solubilized in 3% DMSO and
1% Triton X-100 by sonication. To this solution was added 30 mM Tris-acetate pH 8.0, 50 mM
MgCl 2 , and 20 pM UDP-sugar (20 mCi/mmol), in a final volume of 100 pLL. The reaction was
initiated with 200 nM WeeH and time points taken over a 60 minute duration at 25 C. Aliquots
of 15 ptL were quenched in 1 mL of 2:1 chloroform:methanol mixture and extracted 3 times
using 400 ptL PSUP (3% chloroform, 49% methanol, 48% water). Following extraction, 5 mL
Ecolite(+) scintillation fluid (MP Biomedicals) and 5 mL OptiFluor (Perkin Elmer) were added
119
to the aqueous and organic layers respectively. A Beckman scintillation counter (LS6500) was
employed to determine the radioactivity in each sample.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Dijkshoorn, L., Nemec, A., and Seifert, H. (2007) An increasing threat in hospitals:
multidrug-resistant Acinetobacterbaumannii. Nat. Rev. Microbiol. 5, 939-951.
Perez, F., Huher, A.M., Hujer, K.M., Decker, B.K., Rather, P.N., and Bonomo, R.A.
(2007) Global challenge of multidrug-resistant Acinetobacter baumannii. Antimicrob.
Agents Chemother. 51, 3471-3484.
Peleg, A.Y., Seifert, H., and Paterson, D.L. (2008) Acinetobacter baumannii: emergence
of a successful pathogen. Clin. Microbiol.Rev. 21, 538-582.
Vallenet, D., Nordmann, P., Barbe, V., Poirel, L., Mangenot, S., Bataille, E., Dossat, C.,
Gas, S., Kreimeyer, A., Lenoble, P., Oztas, S., Poulain, J., Sigurens, B., Robert, C.,
Abergel C., Claverie, J.M., Raoult, D., Medigue, C., Weissenbach, J., and Cruveiller, S.
(2008) Comparative analysis of Acinetobacters: three genomes for three lifestyles. PloS
One 3, e1805.
Poirel L., Menuteau, 0., Agoli, N., Cattoen, C., and Nordmann, P. (2003) Outbreak of
extended-spectrum beta-lactamase VEB- 1-producing isolates of Acinetobacter baumannii
in a French hospital. J. Clin. Microbiol. 41, 3542-3547.
Fournier, P.E., Vallenet, D., Barbe, V., Audic, S., Ogata, H., Poirel, L., Richet, H.,
Robert, C., Mangenot, S., Abergel, C., Nordmann, P., Weissenbach, J., Raoult, D., and
Claverie, J.M. (2006) Comparative genomics of multidrug resistance in Acinetobacter
baumannii. PLoS Genetics 2, e7.
Adams, M.D., Goglin, K., Molyneaux, N., Hujer, K.M., Lavender, H., Jamison, J.J.,
MacDonald, I.J., Martin, K.M., Russo, T., Campagnari, A.A., Hujer, A.M., Bonomo,
R.A., and Gill, S.R. (2008) Comparative genome sequence analysis of multi-drug
resistant Acinetobacterbaumannii.J.Bacteriol. 190, 8053-8064.
Nothaft, J. and Szymanski, C.M. (2010) Protein glycosylation in bacteria: sweeter than
ever. Nat. Rev. Microbiol. 8, 765-778.
Kelly J., Jarrell H., Millar L., Tessier L., Fiori, L.M., Lau, P.C., Allan, B., and
Szymanski, C.M. (2006) Biosynthesis of the N-linked glycan in Campylobacter jejuni
and addition onto protein through block transfer. J Bacteriol.188, 2427-2434.
Karlyshev, A.V., Everest, P., Linton, D., Cawthraw, S., Newell, D.G., and Wren, B.W.
(2004) The Campylobacter jejuni general glycosylation system is important for the
attachment to human epithelial cells and in the colonization of chicks. Microbiology 150,
1957-1964.
Schoenhofen, I.C., McNally, D.J., Vinogradov, E., Whitfield, D., Young, N.M., Dick, S.,
Wakarchuck, W.W., Brisson, J.R., and Logan, S.M. (2006) Functional characterization of
dehydratase/aminotransferase pairs from Helicobacter and Campylobacter: enzymes
distinguishing the pseudaminic acid and bacillosamine biosynthetic pathways. J. Biol.
Chem. 281, 723-732.
120
12. Olivier, N.B., Chen, M.M., Behr, J.R., and Imperiali, B. (2006) In vitro biosynthesis of
UDP-N,N'-diacetylbacillosamine by enzymes of the Campylobacter jejuni general
protein glycosylation system. Biochemistry 45, 13659-13669.
13. Vik, A., Aas, F. E., Anonsen, J. H., Bilsborough, S., Schneider, A., Egge-Jacobsen, W.,
and Koomey, M. (2009) Broad spectrum 0-linked protein glycosylation in the human
pathogen Neisseriagonorrhoeae.Proc.Natl. Acad Sci. US.A. 106, 4447-4452.
14. Hartley, M.D., Morrison, M.J., Aas, F.E., Borud, B., Koomey, M., and Imperiali, B.
(2011) Biochemical characterization of the O-linked glycosylation pathway in Neisseria
gonorrhoeae responsible for the biosynthesis of protein glycans containing NN'diacetylbacillosamine. Biochemistry 50, 4936-4948.
15. Iwashkiw, J.A., Seper, A., Weber, B.S., Scott, N.E., Vinogradov, E., Stratilo, C., Reiz,
B., Cordwell, S.J., Whittal, R., Schild, S., and Feldman, M.F. (2012) Identification of a
general O-linked protein glycosylation system in Acinetobacter baumanniiand its role in
virulence and biofilm formation. PLoS Pathogens8, e1002758.
16. Larkin A. and Imperiali, B. (2009) Biosynthesis of UDP-GlcNAc(3NAc)A by WbpB,
WbpE, and WbpD: enzymes in the Wbp pathway responsible for 0-antigen assembly in
PseudomonasaeruginosaPAO 1. Biochemistry 48, 5446-5455.
17. Krogh, A., Larsson, B., von Heijne, G., and Sonnhammer, E. L. (2001) Predicting
transmembrane protein topology with a hidden Markov model: Application to complete
genomes. J Mol. Biol. 305, 567-580.
18. Glover, K.J., Weerapana, E., Chen, M.M., and Imperiali, B. (2006) Direct biochemical
evidence for the utilization of UDP-bacillosamine by PglC, an essential glycosyl-1phosphate transferase in the Campylobacter jejuni N-linked glycosylation pathway.
Biochemistry 45, 5343-5350.
19. Schoenhofen, I.C., Lunin, V.V., Julien, J.P., Li, Y., Ajamian, E., Matte, A., Cygler, M.,
Brisson, J.R., Aubry, A., Logan, S.M., Bhatia, S., Wakarchuk, W.W., and Young, N.M.
(2006) Structural and functional characterization of PseC, an aminotransferase involved
in the biosynthesis of pseudaminic acid, an essential flagellar modification in
Helicobacterpylori. J Biol. Chem. 281, 8907-8916.
20. Badger, J., Sauder, J.M., Adams, J.M., Antonysamy, S., Bain, K., Bergseid, M.G.,
Buchanan, S.G., Buchanan, M.D., Batiyenko, Y., Christopher, J.A., Emtage, S.,
Eroshkina, A., Feil, I., Furlong, E.B., Gajiwala, K.S., Gao, X., He, D., Hendle, J., Huber,
A., Hoda, K., Kearins, P., Kissinger, C., Laubert, B., Lewis, H.A., Lin, J., Loomis, K.,
Lorimer, D., Louie, G., Maletic, M., Marsh, C.D., Miller, I., Molinari, J., MullerDieckmann, H.J., Newman, J.M., Noland, B.W., Pagarigan, B., Park, F., Peat, T.S., Post,
K.W., Radojicic, S., Ramos, A., Romero, R., Rutter, M.E., Sanderson, W.E., Schwinn,
K.D., Tresser, J., Winhoven, J., Wright, T.A., Wu, L., Xu, J., Harris, T.J. (2005)
Structural analysis of a set of proteins resulting from a bacterial genome project. Proteins
60, 787-796.
21. Larkin, A., Olivier, N.B., and Imperiali, B. (2010) Structural analysis of WbpE from
Pseudomonas aeruginosa PAO1: a nucleotide sugar aminotransferase involved in 0antigen assembly. Biochemistry 49, 7227-7237.
22. The PyMOL Molecular Graphics System, Version 1.5.0.4 Schr6dinger, LLC.
23. Waterhouse, A.M., Procter, J.B., Martin, D.M.A., Clamp, M., and Barton, G.J. (2009)
Jalview version 2 - a multiple sequence alignment editor and analysis workbench.
Bioinformatics25, 1189-1191.
121
24. Olivier, N.B. and Imperiali, B. (2008) Crystal structure and catalytic mechanism of PglD
from Campylobacterjejuni. J.Biol. Chem. 283, 27937-27946.
25. Choi, A.H.K., Slamti, L., Avci, F.Y., Pier, G.B., and Maira-Litran, T. (2009) The
pgaABCD locus of Acinetobacter baumannii encodes the production of poly-[I~~1-6-Nacetylglucosamine, which is critical for biofilm formation. J Bacteriol. 191, 5953-5963.
26. St Michael, F., Szymanski, C. M., Li, J., Chan, K. H., Khieu, N.H., Larocque, S.,
Wakarchuk, W. W., Brisson, J. R., and Monteiro, M. A. (2002) The structures of the
lipooligosaccharide and capsule polysaccharide of Campylobacter jejuni genome
sequenced strain NCTC 11168, Eur. J.Biochem. 269, 5119-5136.
27. Creuzenet, C. (2004) Characterization of Cj 1293, a new UDP-GlcNAc C6 dehydratase
from Campylobacterjejuni, FEBS Lett. 559, 136-140.
28. Hynne, F., Dano, S., and Sorensen, P.G. (2001) Full-scale model of glycolysis in
Saccharomyces cerevisiae. Biophys. Chem. 94, 121-163.
29. Teusink, B., Bachmann, H., and Molenaar, D. (2011) Systems biology of lactic acid
bacteria: a critical review. Microbial Cell Factories1 O(Suppl 1), S11.
30. Wang, Q., Ou, M.S., Kim, Y., Ingram, L.O., and Shanmugam, K.T. (2010) Metabolic
flux control at the pyruvate node in an anaerobic Escherichia coli strain with an active
pyruvate dehydrogenase. Appl. Environ. Microbiol. 76, 2107-2114.
31. Chen, M.M., Weerapana, E., Ciepichal, E., Stupak, J., Reid, C.W., Swiezewska, E., and
Imperiali, B. (2007) Polyisoprenol specificity in the Campylobacter jejuni N-linked
glycosylation pathway. Biochemistry 46, 14342-14348.
122
Chapter 4. Biochemical Analysis and Structure Determination of Bacterial
Acetyltransferases
Responsible
for
the
Biosynthesis
of
UDP-N,N'-
Diacetylbacillosamine
A significant portion of this chapter has been published in the following reference:
Morrison, M. J., and Imperiali, B. (2013) Biochemical Analysis and Structure Determination of
Bacterial
Acetyltransferases
Responsible
for
Diacetylbacillosamine. J. Biol. Chem. (in press).
123
the
Biosynthesis
of
UDP-N,N'-
Introduction
The unique, bacterial sugar NN'-diacetylbacillosamine (diNAcBac) has recently attracted
attention due to its role in bacterial pathogenesis in Campylobacterjejuni (1-3). Importantly, the
enzymes responsible for the biosynthesis of this sugar have also been found in other human
pathogens including selected strains of Acinetobacter baumannii (4) and Neisseria gonorrhoeae
(5).
In all three bacteria, UDP-diNAcBac is biosynthesized from UDP-N-acetylglucosamine
(UDP-GlcNAc) by a series of three enzymes.
The first two enzymes, an NAD+-dependent
dehydratase and a pyridoxal-5'-phosphate-dependent aminotransferase, form the UDP-4-amino
sugar that acts as a substrate for the final step in diNAcBac biosynthesis. Acetylation of the C4
amine on this sugar is accomplished by an acetyl coenzyme A (AcCoA)-dependent
acetyltransferase to generate UDP-diNAcBac (Figure 4-1).
This reaction is catalyzed by an
active site histidine that acts as a general base to abstract a proton from the C4 amine resulting in
nucleophilic attack on the thioester of AcCoA. The biosynthetic machinery necessary for UDPdiNAcBac production has been found in both asparagine (N-linked) and serine/threonine (0linked) protein glycosylation pathways with this sugar acting as the anchor point for further
carbohydrate elaboration. Both glycosylation pathways rely on the sequential build-up of sugars
on a polyprenyl-diphosphate-linked isoprene lipid carrier and transfer of the oligosaccharide en
bloc onto an acceptor protein. For the N-linked glycosylation pathway in C. jejuni the ultimate
glycan is a heptasaccharide consisting of GalNAc-ca-1,4-GalNAc-a-1,4-(Glc-p-1,3)-GalNAc-L1,4-GalNAc-a-1,4-GalNAc-a-1,4-GalNAc-ct-1,3-diNAcBac
(6,7).
Conversely,
the
N.
gonorrhoeae O-linked glycosylation pathway utilizes the Gal-p-1,4-Gal-x-1,3-diNAcBac
trisaccharide (5,8). The final oligosaccharide for the O-linked pathway in the AYE strain of A.
124
baumannii is still unknown, however the glycan in a less pathogenic strain (ATCC 17978) of A.
baumanniiwas recently characterized (9), and found not to include diNAcBac.
C. Jejun
/dINAcBac
Pathway
HO
_____0_OH
HO0
H
HIDID
OH
0H
O-UDP
ACOA
COASH
AHO
O-UDP
H
HO
UOP-4-amino
.
AcCoA
H
O
HO
AcNH
O
O
0
HO
HO OH
HHO
O-UDP
AcNH
OH
UDP-diNAcBac
gonotrhoeaa diNAcBac Pathway
HOAcNHI
O
7
COASH
UDP-4-amino
AcHN 0
AcNH
0
NH
HO
OH
AcNH
O-UDPAcHN
0
0
UDP-d!NAcBac
ONH
AcNH
Figure 4-1. The C. jejuni (top) and N. gonorrhoeae (bottom) glycosylation pathways that utilize
diNAcBac as the reducing end sugar.
Previous structural characterization of the diNAcBac biosynthetic pathway has focused
on the acetyltransferase PglD, an N-linked glycosylation pathway enzyme from C. jejuni (10,11).
Additionally, genetic studies have shown that deletion of the pglD gene in C. jejuni results in the
loss of the final heptasaccharide and dramatic reduction of colonization in a chick animal model,
however a low level of glycosylation was still detected by lectin blotting and mass spectrometry
(12). PglD is a member of the left-handed P-helix family and consists of two separate domains.
The N-terminal domain contains a P-o-p-a-p-o Rossmann fold motif to accommodate UDP4-amino sugar binding.
A hexapeptide repeat motif defines the C-terminal domain that is
responsible for the left-handed P-helix and AcCoA binding.
The oligomeric state of PglD
consists of a homotrimer that utilizes the left-handed P-helix motif of two protomers to form the
cleft for AcCoA binding. Structures of other bacterial N-acetyltransferases have recently been
reported (13-15), although they are distant homologues of PglD based upon their divergent sugar
125
substrates.
However, the sugar acetyltransferases maintain the same overall protein fold by
forming a trimer as the biological unit. In addition, they utilize the same left-handed P-helix
motif from adjacent protomers to form the AcCoA binding pocket. Structures of mammalian
acetyltransferases such as HATI (16) belonging to the GCN5-related N-acetyltransferase
(GNAT) superfamily bear no resemblance to their bacterial counterparts. This is most likely due
to the considerable difference between their respective acyl acceptor substrates; histone H4
(HAT1) and UDP-4-amino (PglD). Interestingly, AcCoA has been shown to adopt one of two
distinct conformations, either bent or curved, depending upon the specific acetyltransferase in
question (17).
Similar to citrate synthase (18), AcCoA bound to PglD adopts a compact
conformation with a bend at the pyrophosphate moiety.
To further our understanding of acetyltransferases from the different UDP-diNAcBac
biosynthetic pathways and to gain insight into the divergent nature of N- and O-linked protein
glycosylation in prokaryotes, acetyltransferases from N. gonorrhoeae (PglB-ATD) and A.
baumannii (Weel) were investigated. To this effect, these enzymes were purified, crystallized,
and the structures solved to high resolution. In addition, a co-crystal structure of PglB-ATD
bound to AcCoA was determined. In this context, a comparison between these structures and the
previously solved C. jejuni acetyltransferase (PglD) crystal structures (10) was explored.
Interestingly, the assumption that these bacterial acetyltransferases should closely resemble each
other since they catalyze the identical reaction is not founded. Surprisingly, the substrate binding
pockets for each of these enzymes vary considerably. Based upon this structural comparison, a
series of active site mutations were carried out on all three acetyltransferases and the enzymes
were characterized kinetically for both AcCoA and UDP-4-amino substrates to gain insight into
the catalytic mechanism.
These studies suggest that whereas each enzyme catalyzes the
126
acetyltransferase reaction with identical substrates, key residues within the binding pockets lead
to a diverse set of catalytic efficiencies. Lastly, a phylogenetic analysis of acetyltransferases that
catalyze the conversion to UDP-diNAcBac in N- and O-linked glycosylation pathways is
examined. The three acetyltransferases presented exhibit a high level of evolutionary diversity
despite their ability to generate the identical final UDP-diNAcBac sugar. Unexpectedly, PglBATD from the O-linked glycosylation pathway shares a more common ancestral lineage with the
PglD (N-linked) when compared to Weel (0-linked).
Results and Discussion
Structure of the N. gonorrhoeaeAcetyltransferase PglB-ATD
PglB from N. gonorrhoeae is a bifunctional enzyme containing an N-terminal
phosphoglycosyltransferase domain (PGTD) and a C-terminal acetyltransferase domain (ATD)
that are homologous to the C. jejuni enzymes PglC and PglD, respectively (5).
For
crystallographic studies, the membrane-bound phosphoglycosyltransferase domain was removed
based upon a Clustal Omega alignment with PglD, thus leaving behind the acetyltransferase
domain referred to herein as PglB-ATD. The structure of apo form of PglB-ATD was solved by
molecular replacement utilizing the previously solved acetyltransferase PglD (sequence identity
= 34%) (10). Difficulties in crystallization of this protein were addressed by removing the final
ten amino acid residues from the C-terminal tail based upon a sequence alignment with PglD.
The removal of these PglB-ATD residues, which are not present in corresponding PglD
sequence, results in a comparable C-terminal tail between the two constructs. PglB-ATD was
crystallized in the cubic space group P2 13 with a single protomer in the asymmetric unit.
Previous work has indicated that bacterial acetyltransferases trimerize in solution (10,19).
127
Although the structure of PglB-ATD shows a single molecule in the asymmetric unit, the
homotrimer can be observed through crystallographic symmetry centered on a 3-fold axis
(Figure 4-2A). This acetyltransferase contains two distinct domains that are responsible for the
catalysis of UDP-4-amino to UDP-diNAcBac using the AcCoA cosubstrate.
The N-terminal
section (N199 - L285) comprises a binding pocket for the UDP-4-amino sugar substrate through
a P-c-P-c-P-a Rossmann fold motif. The C-terminus (P286 - L403) is composed of a lefthanded P-helix motif that, in conjunction with an adjacent PglB-ATD protomer in the trimeric
state, forms an extended cleft that is utilized for AcCoA binding.
Although the N. gonorrhoeae acetyltransferase catalyzes the same reaction as PglD from
C. jejuni and has the same general fold (r.m.s.d. = 0.79 A), there are a few notable differences in
the structures.
The PglB-ATD structure contains a flexible loop (R233 - T246) that is not
observed in PglD (Figure 4-2B).
This loop is tucked in between o-helices 1 and 2 in the N-
terminal sugar-binding domain and makes numerous backbone interactions to the second P-sheet
(F229/D231/D232).
For instance, the side chain amide nitrogen of N239 has a hydrogen-
bonding interaction with the E216 acid moiety on helix cl.
Similarly, the backbone amide
nitrogen of L248 and L249 on helix a2 interacts with the hydroxyl and carbonyl moiety from the
T246 loop residue, respectively. In the apo state, PglD contains a cofactor gate, comprising the
final 10 C-terminal residues that interacts with the adjacent, active-site protomer (10).
To
accommodate AcCoA binding, this gate undergoes a conformation change such that an
interaction is formed with the cognate protomer in a coiled motif. Surprisingly, the apo structure
of PglB-ATD reveals that no such cofactor gate is evident (Figure 4-2B). Instead, the apo state
structure exists as the coiled motif resembling the AcCoA-bound structure of PglD (3BSY).
128
Additional structures of the apo form of PglB-ATD were solved under distinct crystallization
conditions that further supported the absence of the cofactor gate.
A
B
E21.6
N239
L249
L248
Figure 4-2. (A) The N. gonorrhoeae apo PglB-ATD crystal structure depicted in cartoon and
space-filling form. The biological assembly is a homotrimer, individually colored for clarity.
(B) Top-down view of the PglB-ATD homotrimer. The boxed region highlights the additional
loop (represented in orange for clarity, R233 - T246) between helices al and a2 that is absent in
the C. jejuni PglD structure. Intramolecular hydrogen bonds within this loop are depicted as
dashed lines.
Multiple attempts to crystallize PglB-ATD in the presence of UDP-4-amino were
unsuccessful.
Therefore, a structural alignment of apo PglB-ATD and UDP-4-amino bound
PglD (3BSS) was explored due to the minimal changes within the N-terminal domain upon sugar
binding in the PglD structures (r.m.s.d
=
0.70). PglD residues D35 (D231 in PglB-ATD), D36
129
(D232), and H125(H333), which accept hydrogen bonds from the ribosyl 3'-hydroxyl group,
uracil imide, and pyranose C4-amine, respectively are strictly conserved between the two
structures. Only two notable changes between the structures are observed. The N162 amino
acid in PglD, which interacts with the carbonyl oxygen of the pyranose C2-acetyl group, is
modified to the homologous Q370 residue in the PglB-ATD structure. Interestingly, S13 in the
PglD structure, which plays a significant role in the sugar binding pocket by hydrogen bonding
to the a-phosphate of UDP-4-amino and NE of K38, is replaced by G208 in PglB-ATD.
One
cannot rule out the significance of the aforementioned loop in PglB-ATD with respect to sugar
binding. Upon UDP-4-amino binding to Pg1D, an unwinding of helix a2 (M40 - T45) to
accommodate sugar binding and allow for optimal interactions is apparent in the crystal
structure. The PglB-ATD flexible loop is located adjacent to this helix (Figure 4-2B) and upon
sugar binding, could elicit a conformational change in this enzyme to mimic missing interactions
within this site. Clearly, a PglB-ATD UDP-4-amino-bound structure would be necessary to
confirm this hypothesis.
Structure of the N. gonorrhoeaeAcetyltransferase PglB-ATD Bound to AcCoA
The AcCoA-bound PglB-ATD structure was solved by molecular replacement using the
apo PglB-ATD structure. This protein was crystallized in the tetragonal space group P4 32 12 with
three PglB-ATD protomers in the asymmetric unit. Three AcCoA molecules were observed
between the clefts formed by adjacent left-handed P-helices in a compact conformation with a
bend at the pyrophosphate moiety (Figure 4-3). AcCoA binds to PglB-ATD in a similar fashion
with respect to PglD, however there are noticeable differences between coenzyme and binding
pocket residues. Notably, PglB-ATD utilizes a series of seven residues from both protomers to
130
bind AcCoA. In particular, S350 and the backbone amide nitrogen of G351 form a hydrogen
bond to the carbonyl oxygen of the thioester (Figure 4-4A). This is in stark contrast to PglD
where the acetyl group is rotated 1800 and forms hydrogen-bonding interactions with H134 and
N 118 (Figure 4-4B). Although the contacts between protomer B in Pg1D and AcCoA are mainly
hydrophobic, this protomer plays a much larger role in PglB-ATD. Both hydrogen-bonding
interactions of the thioester carbonyl originate from this protomer.
Likewise, the water
hydrogen-bonding network binding the pyrophosphate moiety and the 3'phosphate is replaced by
R368 and K401, respectively. Only two water molecules are observed binding to AcCoA in the
PglB-ATD structure, whereas seven water molecules are contributing factors in the PglD
structure. In fact, there are no conserved residues responsible for binding to AcCoA observed
when comparing the two structures.
However, backbone interactions between A381 (PglB-
ATD) and G173 (PglD) serve a similar purpose by hydrogen bonding to a carbonyl oxygen in the
pantetheine moiety and the C6 amine on the adenine ring.
Further hydrogen-bonding
interactions in the PglB-ATD structure can be observed from T363 (protomer A) and Q369
(protomer B) on the amide adjacent to the thioester in the pantetheine moiety.
131
Figure 4-3. Composite omit map depicting the 2(F, - F) electron density of AcCoA contoured
at 1.5a in PglB-ATD. Protomer A is represented in green, protomer B in magenta, AcCoA in
grey, and water molecules as red spheres. Hydrogen-bond interactions are represented as dashed
lines.
Whereas PglD undergoes a conformational change in the C-terminal tail upon AcCoA
binding, no evidence of this change is observed in the PglB-ATD structure. In fact, since the apo
state of PglB-ATD is already in the coiled motif as observed in the AcCoA-PglD structure, no
other conformational change is necessary to accommodate AcCoA binding. This lack of change
between the AcCoA-bound and apo state of PglB-ATD is reflected in the minor change in
r.m.s.d between the two structures (0.30 A). However care must be taken in interpreting these
results since a small change in r.m.s.d. may be biased since the AcCoA-bound structure was
solved by molecular replacement with the apo PglB-ATD structure. There are only two key
conformational changes in the active-site cleft necessary for AcCoA binding. Most importantly,
R368 in the apo PglB-ATD structure serves to block access to the channel prior to AcCoA
132
binding (Figure 4-5). Upon binding, R368 rotates out of the cleft to allow AcCoA access to the
binding site. This residue is also essential for the binding of AcCoA as it has a total of four
hydrogen-bonding interactions with the coenzyme (Figure 4-4A). Whereas no such C-terminal
cofactor gate exists in PglB-ATD, R368 may play a similar role to allow for AcCoA binding.
Glutamine 369 also plays a role in coenzyme binding by rotating 900 out of the pocket to form
part of the pantetheine binding site and picks up a favorable hydrogen-bonding interaction with
AcCoA. Analogous residues are not apparent in a structural alignment between PglB-ATD and
PglD, adding to the dichotomous nature of these two proteins.
133
A
Arg368
GIn369
0
Lys401
Gly35
1
NH2,,
NH2
V-4b
H20
Ser350
H3N@
0
0H OHI
OH% O-P-O-
H
"I
H2 0
OH%
H20,
H2N10
~NH 2
HN
NAN
,-
O
N
H
-y N
Thr3563
Arg360
H2N
I
O
H2 N
0
A a381
Asn398
N-N
H 20
B
H20
20
H2 0
0
0 OHO
Asn118
NH2
H
N"
H
N1
OH
O 11
1
O- N
'-
H20
N/ NH
2
NN00
N"
:N,/
His134
0
e155
H
1
N
N
O
H20
H2 0
Gly173'~
0
Figure 4-4. Disparity in AcCoA binding to the acetyltransferases PglB-ATD (N. gonorrhoeae)
and PglD (C. jejuni). Representation of the AcCoA binding pocket in (A) PglB-ATD and (B)
PglD. Amino acids responsible for coenzyme interactions are depicted in red with key water
molecules in blue. Hydrogen-bond interactions are illustrated as dashed lines.
134
(A)I(B)
Figure 4-5. Comparison of the AcCoA binding pocket in the PglB(Ng)ATD apo (A) and
AcCoA-bound (B) structures. The amino acids Q369 and R368 act as gatekeeper residues to
accommodate AcCoA binding. These residues are also instrumental in AcCoA substrate
binding.
Structure of the A. baumanniiAcetyltransferase WeeI
The Weel structure was solved by molecular replacement using the previously solved apo
PglB-ATD structure (sequence identity = 26%).
This acetyltransferase crystallized in the
hexagonal space group P3 121 and contained six protomers in the asymmetric unit forming a
dimer of the biological trimer assembly. Optimization of the crystals was a necessity due to the
poor diffraction quality of the original conditions. In particular, the addition of 0.7% 1-butanol
to the crystallization buffer improved resolution by 0.8 A (Hampton Additive Screen). Similar to
PglB-ATD and PglD, WeeI is composed of N- (Ml - H90) and C-terminal (L91 - L213)
domains that are each responsible for binding to UDP-4-amino and AcCoA, respectively (Figure
4-6A). A cleft is formed between two adjacent protomers from the C-terminal left-handed 0helix domain that accommodates AcCoA binding as previously observed in the PglB-ATD
crystal structure. Unfortunately, multiple screening attempts to solve the WeeI structure bound
to the UDP-4-amino sugar and AcCoA proved unsuccessful.
135
A
B
Figure 4-6. (A) The A. baumanniiapo WeeI crystal structure depicted in cartoon and spacefilling form. The biological assembly is a homotrimer, individually colored for clarity. (B) Topdown view of the Weel homotrimer. The boxed region highlights the additional loop
(represented in orange for clarity, Q174 - P 180) that forms the UDP-4-amino binding pocket in
close proximity to the pyranose moiety. Intramolecular hydrogen bonds are depicted as dashed
lines.
Similar to the apo structure of PglB-ATD, Weel contains a flexible loop between helices
1l and a2 in the N-terminal sugar-binding domain. As previously observed with PglB-ATD,
this loop has numerous intramolecular interactions with the protein backbone. Interestingly, the
conserved residue N46 (N239 in PglB-ATD) exhibits a similar hydrogen-bonding interaction
with N121 on an adjacent protomer. Of note, P49 (P242 in PglB-ATD) is also conserved in this
region and serves to stabilize this loop through hydrogen bonding of the backbone carbonyl to
the conserved F35 (F229 in PglB-ATD) amide nitrogen. The essential sugar binding residues
136
observed in the PglD/UDP-4-amino structure are strictly conserved in Weel including S13 (S 11
in PglD) that is conspicuously absent in the PglB-ATD structure. Residues D35, D36, and H125
in PgiD, which contribute hydrogen-bonding interactions with UDP-4-amino are conserved in
Weel (D37, D38, H138). The only exception in this binding pocket is the PglD N162 (Q370 in
PgiB-ATD) residue.
In Weel, the pyranose moiety of the UDP-4-amino binding pocket is
formed by a seven amino acid loop (Q174 - P180) from the adjacent protomer. This loop is not
observed in the two other acetyltransferase structures (Figure 4-6B) and contains two residues
(Q174 and T176) in the vicinity of hydrogen bonding to the carbonyl oxygen of the pyranose C2acetyl group.
Alanine mutagenesis was performed on these two sites to ascertain their
relationship to UDP-4-amino binding (see below).
In PglD, a conformational change in H15 is
observed to accommodate sugar substrate binding. In the apo structure, this residue occludes the
UDP-4-amino pocket. However upon substrate binding, this residue tucks into the pocket and
interacts with the p-phosphate moiety of the sugar. Although this residue is conserved in PgiBATD (H21i0), the more bulky, hydrophobic phenylalanine residue is found in WeeL. This small
change could have a deleterious binding effect on the UDP-4-amino substrate (see below).
The Weel AcCoA binding pocket exhibits a stronger homology to the PglB-ATD site
(56% sequence identity) when compared with PgiD (34% sequence identity). Not surprisingly,
this can also be observed when comparing the crystal structures. Similar to PglB-ATD, Weel
does not appear to utilize a cofactor gate for AcCoA binding (Figure 4-6B).
From the apo
structure, the C-terminal tail is in a coiled motif that resembles the AcCoA-bound PglD structure.
Weel also contains an analogous residue to R368 (PglB-ATD) that may act as a gate to AcCoA
binding. Lysine 173 is positioned in a similar fashion to R368 and obstructs the binding cleft in
the apo state. Although no structure of AcCoA bound to Weel exists, one can hypothesize that
137
this residue plays an analogous role in coenzyme binding.
Key residues that interact with
AcCoA in the PglB-ATD crystal structure are mostly conserved in Weel. PglB-ATD residues
G351 (G156 in Weel), Q369 (Q174), and T363 (T168) are strictly conserved. Substitutions at
K401 (R21 1) and S350 (N155) are complementary in nature and a similar role can be envisioned
at these positions. Of note, the
124
EHE (PglD) and
332
DHD (PglB-ATD) motifs that are critical
for catalysis are slightly modified in Weel (137AHD). The carboxylate moiety of PglD (E126),
PglB-ATD (D334), and Weel (D139) is hydrogen-bonded to the imidazole ring of histidine
increasing its basicity. This enhancement allows for the NE2 nitrogen of histidine to act as a
general base in catalysis by de-protonating the C4 amine on the UDP-4-amino sugar. Although
the carboxylate moiety in PglD (E124) and PglB-ATD (D332) may serve to recycle histidine
back to its pre-catalytic state by abstracting a proton from NE2 following substrate turnover (10),
this cannot be the case in WeeI due to the alanine moiety at this position.
Analysis ofAcetyltransferase Active-Site Mutants
To better understand the contributions of particular residues in binding and catalysis, a
series of mutations were created in the active sites of PglD, PglB-ATD, and Weel based upon
their crystal structures. While holding one substrate at saturating levels for PglD and PglB-ATD,
the other was varied to determine kinetic parameters through initial velocity measurements.
Ellman's reagent (5,5'-dithiobis(2-nitrobenzoic acid)) was utilized to monitor AcCoA conversion
to CoASH through generation of the TNB2- chromophore (42 nm
=
14,150 M-1 cm'). Due to the
poor binding of UDP-4-amino to Weel, the AcCoA kinetic parameters were determined at the
Km of the UDP-sugar. Typical Michaelis-Menten kinetics were observed for all concentrations
of UDP-4-amino and AcCoA.
Initial velocity measurements were averaged between two
138
duplicate experiments. UDP-4-amino and AcCoA kinetic parameters for the acetyltransferase
mutants are listed in Table 4-1 and Table 4-2, respectively.
Table 4-1. Steady-state kinetic parameters for the UDP-4-amino acetyltransferase substrate.
AcetyttrMaqfe ase;<
2<a/o(
K
PgID WT
PgID H15F
PgID E124A
PgIB-ATD WT
PgIB-ATD H21OF
PgIB-ATD D332A
PgIB-ATD Q369A
274 ±6.4
2780 ±450
1562 ± 370
99.0 ±7.1
320 ± 20
560 ± 140
96.7 ± 17
8.0 ± 1.6 x 10s
9.6 ±3.2 x 104
1.9 0.08 x 105
7.2 0.8 x 104
1.2 0.2 x 103
3.6 0.2 x 10 3
5.6 0.4 x 10 3
PgIB-ATD Q370A
89.9 ± 11
Weel WT
Weel F13A
Weel Q174A
Weel T176A
2520± 540
4510 ±71
2420 ±430
4600 ±480
3.0 0.6 x 10 4
5.1 0.08 x 105
8.2 0.2 x 104
1.9
0.9 x 103
3.3
0.3 x 10 5
2.9 x 10 9
3.4 x 107
1.2 x 108
7.3 x 10 8
3.7 x 106
6.5 x 106
5.8 x 107
3.4 x 10 8
2.0 x 108
1.8 x 10
8.0 x 10s
7.1 x 107
Table 4-2. Steady-state kinetic parameters for the AcCoA acetyltransferase substrate.
Acetyltra nsfe rase
PgID WT
PgID H15F
PgID E124A
PgIB-ATD WT
PgIB-ATD H21OF
PgIB-ATD D332A
PgIB-ATD Q369A
K,
ft A4
295
35.0
104
286
600 ±
503
k.t(S1
6.1 ±1.0 x 10 5
3.0 ± 1.9 x 10 4
1.6 ±0.008 x 10 5
5.0 0.7 x 10 4
1.7 0.07 x 10 3
3.9 ± 0.2 x 10 3
2.8
0.6
7.6
35
130
35
kcatl.4d(M.1W.1
2.1 x
8.7 x
1.5 x
1.7 x
2.8 x
7.8 x
10 9
108
109
108
106
106
PgIB-ATD Q370A
716 23
480 ± 110
8.2 1.1 x 10 3
4.9 ± 0.2 x 10 4
1.1 x 107
1.0 x 108
Weel WT
Weel F13A
Weel Q174A
Weel T176A
78.9 28
84.9 9.5
196 53
118 59
1.3 0.06 x 105
4.3 1.6 x 10 4
5.1 0.9 x 102
6.9 0.4 x 10 4
1.6 x 109
5.0 x 10
2.6 x 106
5.9 x 108
When comparing UDP-4-amino affinity to other bacterial acetyltransferases, WeeI
exhibits poor binding (Km is increased approximately 10-fold). From a structural alignment
standpoint, the phenylalanine at position 13 may contribute to such a poor Km in Wee.
139
This
observation is based upon the absence of changes in UDP-4-amino binding pocket residues with
respect to PglD and PglB-ATD. The aforementioned H15 residue in PglD can be classified as a
type of gatekeeper moiety due to its ability to tuck into the pocket to accommodate UDP-4amino binding and interact directly with this substrate. The histidine is conserved in PglB-ATD
resulting in a similar UDP-4-amino Kin, however this site is a phenylalanine in Weel. This
change in steric bulk, hydrophobicity, and loss of hydrogen bonding at this key position may
result in reduced binding affinity. Therefore, a mutation in PglD (H1i5F), PglB-ATD (H210F),
and Weel (F13A) was explored. This mutation had a deleterious effect on both catalysis and
binding on both PglD and PglB-ATD, whereas the F13A Weel mutation mainly affected
turnover (Table 4-1). In order to ascertain if these mutations have any effect on the adjacent
AcCoA binding pocket, kinetic parameters were explored for this substrate.
Surprisingly, the
PglD H15F mutation resulted in a 10-fold increase in binding affinity to AcCoA while
decreasing kcat by 20-fold (Table 4-2). Mutation of H21OF (PglB-ATD) and F13A (Weel)
resulted in no change in binding affinity, however kcat decreased considerably. This particular
site in the UDP-4-amino binding pocket contributes significantly to binding and catalysis in
these acetyltransferases.
The inability to crystallize UDP-4-amino with either PglB-ATD or Weel prompted a
series of alanine mutations to determine specific sites within each binding pocket that contribute
to binding and catalysis. Based upon an alignment with the UDP-4-amino PglD structure, PglBATD (Q369A and Q370A) and Weel (Q174A and T176A) mutants were created. In both cases,
these changes are within the vicinity of the PglD residue N162, which interacts with the carbonyl
oxygen of the pyranose C2-acetyl group. Although neither PglB-ATD mutation resulted in a
change in UDP-4-amino binding, Q369A had a significant effect on turnover (13-fold decrease).
140
Likewise in the Weel mutations, only T174A resulted in a considerable (270-fold) loss in kcat.
Due to the proximity of these mutations to the AcCoA binding site, kinetic parameters were also
established for this substrate. Mirroring the UDP-4-amino results, a decrease in kcat was only
observed for the PglB-ATD Q369A mutant (6-fold) and Weel Q I74A (260-fold). It is apparent
that the Weel Q I74A mutant plays an extremely important role in catalysis of this reaction.
There is still an ongoing discussion over the acetyltransferase catalytic mechanism and in
particular the protonation state of UDP-4-amino substrate (10,11,15). Although this study does
not address this question specifically, the role that E124 (PglD) plays in catalysis was explored.
This position has been implicated in returning the catalytic histidine (11125) back to its preturnover state by transferring the proton on the imidazole moiety to the thiolate on CoAS~ (10).
Most homologous acetyltransferases incorporate either a glutamate or aspartate at this position,
however this site is occupied by an alanine in Weel. Interestingly, the catalytic efficiency of
Weel is comparable to PglD and PglB-ATD.
To better understand the catalysis and binding at
this site, mutant variants of PglD (E124A) and PglB-ATD (D332A) were prepared. In both
cases, kcat was reduced for both substrates (Tables 4-1 and 4-2), however the loss was more
significant in PglB-ATD (20-fold). Mutation to alanine in PglD and PglB-ATD then has a
detrimental effect on catalysis, yet the wild-type Weel is still a competent enzyme with alanine at
this position. Therefore, recycling of the active site must be accomplished in another manner.
The closest amino acid site that can act as a general base is K15, however that residue is over 6
A away from the catalytic histidine. The most straightforward solution would rely on the thiolate
from CoAS- (following acetylation of the UDP-4-amino sugar) to act as a base to directly
remove the proton from the catalytic histidine. This would regenerate the active site to its precatalytic state and explain the absence of a general base adjacent to H138 in Weel. In lieu of
141
these results, the glutamate/aspartate moiety in PglD (E124) and PglB-ATD (D332) appears to
be a non-absolute requirement for catalysis and its essentiality may have previously been
overstated.
Mutagenesis of the UDP-4-Amino Binding Pocket Reveals Kinetic Diversity
The structures of PglB-ATD and Weel add to the growing number of acetyltransferases
that are associated with UDP-diNAcBac biosynthesis. Importantly, these structures represent the
first O-linked glycosylation pathway enzymes that result in the production of this bacterial sugar.
Although the overall architecture of these proteins is similar with respect to PglD, there are
notable differences that contribute to their contrasting kinetic parameters. In particular, Weel
binds to UDP-4-amino with a significantly lower affinity (10-fold) in comparison to PglD and
PglB-ATD (Table 4-2). From a structural alignment standpoint, Weel contains one key residue
(F 13) that may be responsible for this dramatic Km shift. In PglD, this position (H15) undergoes
a conformational change to accommodate UDP-4-amino binding and interacts with sugar Pphosphate. Site-directed mutagenesis of this position (H15F) resulted in a 10-fold loss in affinity
for UDP-4-amino binding with PglD.
However, the same mutation in PglB-ATD (H210F)
produced a more modest loss in binding (3-fold). This position is extremely important for
acetyltransferase activity since there is a 100-fold decrease in catalytic efficiency (kcat/Km) when
mutating this residue to a phenylalanine in both PglD and PglB-ATD. Despite the poor binding
affinity of UDP-4-amino, Weel must contain a compensatory effect since this enzyme retains a
similar efficiency with respect to PglD and PglB-ATD.
Weel contains an additional loop (Q174 - P180) that forms the UDP-4-amino binding
pocket near the pyranose moiety. Residue Q174 seems to be critical for catalysis since an
142
alanine mutation results in a 270-fold loss in kcat while maintaining its affinity for UDP-4-amino.
When aligned to the PglD UDP-4-amino structure, this position is analogous to N162 that
interacts with the carbonyl oxygen of the pyranose C2-acetyl group. In the Weel apo structure,
Q174 is within 5 A of the catalytic base (H 138) and 3.6
A to the AcCoA thioester
when aligned
to the PglB-ATD AcCoA structure. Likewise, the Q174A mutation has a detrimental effect on
AcCoA catalysis with a 260-fold loss in turnover. Clearly, this residue plays a key role in the
overall function of Weel.
Dichotomy Among N- and O-Linked AcetyltransferaseAcCoA Binding Pockets
A general theme of binding and catalysis among homologous proteins is conservation of
key amino acids that result in the comparable activity between enzymes. Although the AcCoA
binding parameters of PglD and PglB-ATD are extremely similar (Table 4-2), the way in which
the enzymes bind the coenzyme are distinct (Figure 4-4).
AcCoA is mainly held into the
binding pocket of PglD by hydrophobic interactions and a network of water molecules. In fact,
only two side chains (N118, H134) contribute to the overall affinity of this substrate in PglD.
Surprisingly, these analogous residues in PglB-ATD play no role in binding to AcCoA. Instead,
the coenzyme is held in the binding site by a series of hydrogen-bonding interactions from a total
of seven residues. Interactions between the phosphate moieties in PglB-ATD have replaced the
water molecules in PglD with R368 and K401.
In addition to this major change in binding-site functionality, PglB-ATD does not appear
to utilize a C-terminal cofactor gate for AcCoA binding. Upon AcCoA binding in the PglD
structure, the C-terminal tail undergoes a conformational change to accommodate the coenzyme
in the form of a coiled motif. This coiled motif is already apparent in the apo structure of PglB-
143
ATD (Figure 4-2), however the removal of the final ten amino acids from the C-terminal tail for
crystallization purposes could have elicited this result.
However this is unlikely since this
deletion does not remove the residues responsible for this conformational change as observed in
PglD. Furthermore, the apo Weel structure is of the full-length protein and also does not exhibit
a conformational change in this cofactor gate (Figure 4-6). However, one cannot rule out the
possibility of a crystallographic artifact when discussing these types of small conformational
changes between enzymes.
Although both O-linked glycosylation acetyltransferases do not
contain this cofactor gate, comparisons between the apo and AcCoA-bound structures of PglBATD resulted in the discovery of a residue that may have analogous function.
In the apo
structure, R368 can be observed blocking the AcCoA binding cleft (Figure 4-5).
Upon
coenzyme binding, this residue rotates out of the pocket and interacts with the phosphate and
pantetheine hydroxyl moieties. When comparing the coenzyme binding pockets in the apo and
AcCoA-bound structures, no other large conformational changes are detected. Similar to this
observation, WeeI contains K173 at this position when aligned with the PglB-ATD AcCoA
structure. This residue seems to function in a similar fashion to R368 in PglB-ATD, as the lysine
side chain is also observed blocking access to the AcCoA binding channel in the apo state.
These changes are not surprising in the context of homology between N- and O-linked
glycosylation pathways as the O-linked acetyltransferases retain a high degree of sequence
identity in the AcCoA binding pocket. PglB-ATD and Weel bear a stronger resemblance in their
structural homology when compared with PglD. This observation is also evident in the sequence
homology between their respective aminotransferase active sites (4).
144
Phylogenetic Analysis of BacterialAcetyltransferases
Whereas the three acetyltransferases presented here carry out the same reaction and
display the same general protein fold, homology within the substrate binding sites is quite
divergent. To further our understanding on the evolutionary aspect of acetyltransferases within
the diNAcBac pathway, a phylogenetic analysis was carried out.
diNAcBac pathway were
oligosaccharyltransferases
first identified by having a > 35%
Bacteria containing the
homology to known
(OTases) from C. jejuni, N. gonorrhoeae, and A. baumannii.
Comparative assessment of these enzymes allowed for classification of PglD as an N-linked
glycosylation system and PglB-ATD/WeeI as O-linked systems. Acetyltransferases were further
classified similarly to the OTase analysis above and sequentially aligned with the software
program MUSCLE. Interestingly, the neighbor-joining dendrogram (Figure 4-7) is broken up
into multiple clades and exhibits evolutionary diversity, which is observed within the
acetyltransferase binding pockets. This is somewhat surprising since the acetyltransferases from
C. jejuni, N. gonorrhoeae, and A. baumannii carry out the identical reaction to produce the
diNAcBac
sugar.
Similar results were previously observed
using dehydratase
and
aminotransferase homologs from Campylobacter and Neisseria (20). Of note, homologous A.
baumannii acetyltransferases are evolutionarily more distant with respect to C. jejuni and N.
gonorrhoeae.
Glycosylation is a ubiquitous post-translational modification and is known for
modulating cellular processes such as protein folding, stability, and cell signaling (21-22).
Significantly, bacteria also utilize protein glycosylation for purposes of mediating colonization,
adhesion, and invasion of eukaryotic cells (1,12,23).
In fact, recent work on the ATCC 17978
strain of A. baumannii has demonstrated a link between pathogenicity and protein glycosylation
145
(9,24). To better understand the module responsible for the biosynthesis of UDP-diNAcBac,
research has focused on the specific enzymes that lead to the synthesis of this unusual sugar.
Bacterial glycosylation can be classified as N- (asparagine-linked glycan) and 0-linked
(serine/threonine-linked glycan).
Both modifications, in the context of UDP-diNAcBac
biosynthesis, have been studied extensively in C. jejuni (N-linked) (7,10,25) and to a lesser
extent in N. gonorrhoeae and A. baumannii (0-linked) (4,5,8).
Although the enzymes
responsible for the biosynthesis of this unique, nucleotide sugar are present in these bacteria,
they are evolutionarily divergent with regard to their acetyltransferases (Figure 4-7). Homologs
of these enzymes from their respective organisms are separated into multiple clades within the
dendrogram. There are two explanations to account for this observation. First, these enzymes
could have covergently evolved by acquiring the biosynthetic enzymes necessary for the
production of UDP-diNAcBac. Secondly, these enzymes could have evolved from a common
ancestor and diverged over an extended period of time. This is the simpler explanation and
could account for the varying degrees of identity observed within the AcCoA and UDP-4-amino
binding pockets (4). For instance, the C. jejuni PglD UDP-4-amino binding pocket shares a
higher sequence identity with PglB-ATD from N. gonorrhoeae. Conversely, WeeI from A.
baumannii shares a higher homology with its O-linked counterpart, PglB-ATD, in the AcCoA
binding pocket. In either case, it is interesting that the acetyltransferases from two O-linked
pathogens (N. gonorrhoeaeand A. baumannii) are evolutionarily more divergent with respect to
the N-linked C. jejuni enzyme. It is currently unknown as to whether A. baumannii acquired this
enzyme from an N- or O-linked pathway. Whereas the true significance of UDP-diNAcBac is
presently unclear, it is important to recognize its ubiquitous nature in pathogenic bacteria. Why
specific bacteria acquired the UDP-diNAcBac biosynthetic pathway remains a mystery.
146
Additionally, questions surrounding the motility of the UDP-diNAcBac module between bacteria
in lieu of the entire glycosylation pathway remain unanswered. Further work is warranted to
address these questions in the context of bacterial fitness and pathogenicity.
0
30031IH210
*770
Cwv~bakecwco 37-4
N
N.
N
I
I
1
~1
Figure 4-7. Phylogenetic tree constructed with the neighbor-joining method from the
Campylobacter genus (green clade), the Neisseria genus (red clade), and the Acinetobacter genus
(blue clade) acetyltransferases. The three acetyltransferases utilized for this analysis are
indicated with an arrow. The evolutionary distances were computed using the Poisson correction
method (39) and are in the units of the number of amino acid substitutions per site. The scale bar
indicates substitutions per site. Evolutionary analyses were performed in MEGA 5.2.
147
Conclusions
In conclusion, the structures of the O-linked glycosylation pathway acetyltransferases
PglB-ATD and Weel bring us closer to understanding the intricacies of UDP-diNAcBac
biosynthesis. Importantly, these structures establish the divergent nature of the UDP-4-amino
and AcCoA binding pockets in contrast to the N-linked acetyltransferase PglD. Although these
three enzymes catalyze the same reaction, minor modifications of each binding site can have
large ramifications on binding and catalysis. These results provide insight into the surprising
structural diversity among bacterial acetyltransferases that catalyze the same reaction with
similar efficiencies. C. jejuni, N. gonorrhoeae, and A. baumannii occupy specific and different
environments within their host organisms. For instance, C. jejuni is an enteric pathogen and
resides in the digestive tract; A. baumannii colonizes the respiratory tract that leads to hospitalacquired pneumonia.
The changes outlined in this study may reflect the adaptability of the
components in the UDP-diNAcBac pathway to their respective environments. Due to the high
catalytic efficiency of these acetyltransferases, pathway flux may be attenuated through these
enzymes (4,25). Depending upon the environment in which the bacterial pathogen resides,
virulence factors that rely upon diNAcBac glycosylation may need to be tuned in a positive or
negative fashion. Therefore, changes within the acetyltransferase binding pockets may be the
result of these circumstances. Additional research is necessary to provide further evidence for
this hypothesis.
The structural and mutagenesis work presented here strengthens our
understanding of bacterial glycosylation in relation to N- and O-linked glycosylation pathways
from significant pathogenic bacteria.
148
Acknowledgments
I am extremely grateful to Professor Robert Sauer, Dr. Robert Grant, and Jeremy Setser for
assistance with data refinement and technical advice on crystallography. I would like to thank
Dr. Nina Leksa for Weel data collection, Professor Michael Laub for advice on phylogenetic
analysis, and Dr. Angelyn Larkin for critical reading of this chapter. Lastly, we would like to
thank Austin Travis for PglB-ATD AcCoA data collection and critical reading of this chapter.
Experimental Procedures
Common materials
All chemicals were purchased from Sigma-Aldrich unless otherwise stated. The UDP-4amino sugar was biosynthesized as described previously from the C. jejuni enzymes PglF and
PglE (25).
Molecular biology
The acetyltransferase domain (ATD) of the pglB gene from N. gonorrhoeaeFA1090 was
identified through a Clustal Omega alignment (26) with the C. jejuni acetyltransferase (PglD).
The gene encoding this domain was amplified via the polymerase chain reaction (PCR) with the
forward primer 5'-CGCGGATCCATGGCGGGGAATCGCAAACTCG-3'
primer 5'-GCAACCCGGCAAAGCCCCTTTAGCTCGAGCGG-3'
and the reverse
from the N. gonorrhoeae
FA1090 strain (8). The weeI gene was amplified via PCR from the genomic DNA from the A.
baumannii AYE strain (ATCC BAA-1710) (27).
BamHI and XhoI restriction sites were
engineered to facilitate cloning of each construct into a modified pET30b(+) vector (Novagen)
149
containing an N-terminal His8 tag followed by a tobacco etch virus (TEV) protease site prior to
the BamHI site. Amplifications were accomplished with the PfuTurbo DNA Polymerase
(Stratagene) as described by the manufacturer. Amplicons were purified and double-digested
with BamHI and XhoI restriction enzymes (NE Biolabs). Digested inserts and linearized vectors
were fractionated by agarose gel electrophoresis and purified with the Wizard SV Gel and PCR
Cleanup Kit (Promega). Ligations were conducted with the T4 DNA ligase kit (Promega) using a
15 min incubation at room temperature. Sequencing by Genewiz (Cambridge, MA) confirmed
the presence of all gene products.
Site-directed mutagenesis was accomplished utilizing the
QuikChange protocol (Stratagene) with pglD(C)-pET24a(+), pglB-ATD(Ng)-pET24a(+), and
weeI(Ab)-pET24a(+) as the template plasmids from previous studies (4,5,25).
Protein Expression
The modified pET30b(+) plasmid containing each gene was used to transform
Escherichia coli BL21 (DE3) RIL competent cells (Stratagene). One liter of LB media containing
50 pg/mL kanamycin and 30 ptg/mL chloramphenicol was inoculated with 8 mL of an overnight
culture of cells. The cells were then allowed to grow at 37 C while shaking until an optical
density of-0.8 (k = 600 nm) was reached. The culture was cooled to 16 C and induced with 0.5
mM iso-p-D-thiogalactosylpyranoside (IPTG). After incubating for 18 h with shaking at 16 C,
the cells were harvested by centrifugation (2600g, 30 min) and stored at -80 0C until needed.
Protein Purification
Each protein purification step was carried out at 4 C. For crystallization experiments,
the cell pellet (~3 g) was resuspended in 40 mL of 50 mM HEPES pH 7.4/100 mM NaCl/30 mM
150
imidazole (Buffer A) and then lysed by sonication. The lysate was then cleared by centrifugation
(145,000g, 60 min) and added to 2 mL of Ni-NTA resin (Qiagen). The slurry was allowed to
tumble for 3 h and then packed into a fritted PolyPrep column (Biorad). The resin was washed
with 20 column volumes of Buffer A and then eluted with a buffer containing 50 mM HEPES
pH 7.4/100 mM NaCl/ 300 mM imidazole. Fractions containing the purified protein as analyzed
by SDS-PAGE were pooled and dialyzed against 50 mM TRIS pH 8.0/5 mM EDTA/5 mM
p-
mercaptoethanol in the presence of 6 pM TEV protease for 24 h to remove the His8 tag.
Removal of this tag was monitored by Western blot analysis using an anti-His 4 antibody
(Qiagen). The reaction was diluted 10-fold in 25 mM HEPES pH 7.6 and excess TEV was then
removed with a HiTrap
Q HP
Sepharose anion exchange column (GE Healthcare) utilizing a
linear NaCl gradient. Fractions containing the protein were pooled and dialyzed for 24 h in 50
mM HEPES pH 8.0/150 mM NaCl (SEC buffer). After concentrating to a volume of 1.5 mL
using a 10K Da MWCO Amicon Ultra-15 centrifugal filter unit (Millipore), the protein was
loaded onto a Superdex 200 16/60 column (GE Healthcare) and subjected to size-exclusion
chromatography in SEC buffer. Fractions containing the monodispersed protein were pooled,
concentrated, and used within 24 hours for crystallization experiments. Protein concentrations
were calculated based upon the predicted extinction coefficients at k = 280 nm.
Proteins subjected to mutagenesis were purified using 2 mL Ni-NTA as above.
Following elution from the resin, fractions containing the pure protein were dialyzed in a 4 L
volume against 50 mM HEPES pH 7.4/100 mM NaCl for 24 h to remove the imidazole. Purity
for each protein was assessed by SDS-PAGE to be > 95% (Figure 4-8). Full-length constructs
were confirmed through Western blot analysis probing with antibody for the T7 and His tags.
151
This solution was concentrated as described above to -10 mg/mL and supplemented with 15%
glycerol. Aliquots of each protein were stored at -80 0C until needed.
kDa
I
105
78
45
Figure 4-8. SDS-PAGE gradient gel (4-20%) of acetyltransferase mutants. Lane 1: MW
standard, Lane 2: Pg1D H15F, Lane 3: PglD E124A, Lane 4: PglB-ATD H21OF, Lane 5: PglBATD D332A, Lane 6: PglB-ATD Q369A, Lane 7: PglB-ATD Q370A, Lane 8: Weel F13A, Lane
9: Weel Q174A, Lane 10: Weel T176A, Lane 11: MW standard.
Crystallizationand Data Collection
All crystals were grown as hanging drops by combining 1.5 IL of a 10 mg/mL protein
solution in SEC buffer with 1.5 pL of reservoir solution at 25 C. Each well contained a final
volume of 500 iL of reservoir solution. For the cocrystallization of PglB-ATD with AcCoA, the
substrate was added to the protein so that the final concentration was 10 mM and incubated for
45 min at 25 "C. The reservoir solution for apo PglB-ATD contained 0.1 M sodium acetate pH
4.6, 0.02 M calcium chloride, and 30% 2-methyl-2,4-pentanediol (MPD). The AcCoA-bound
Pg1B-ATD reservoir solution contained 0.1 M BIS-TRIS pH 5.5, 3.0 M NaCl. For apo-Weel,
the well solution contained 0.1 M sodium acetate trihydrate pH 4.5, 3.0 M NaCl, 0.7% 1 -butanol.
After the crystals were fully grown (~24 h), they were cryoprotected in reservoir solution
152
containing 20% glycerol. For AcCoA-bound PglB-ATD, this solution was also supplemented
with 10 mM of substrate. Diffraction data was collected on beamline X25 (National Synchrotron
Light Source, Brookhaven National Laboratory, Upton, NY) at 100 K using a Pilatus 6M
detector. Data sets were processed using HKL2000 (28), MOSFLM (29), TRUNCATE (30-31),
and SCALA (30). Parameters from the data collection are listed in Table 4-3.
Table 4-3. Data Collection and Refinement Statistics.
Weel
PglB-ATD
PglB-ATD(AcCoA)
P2 13
86.22, 86.22, 86.22
43.2-1.7
25105
7.8 (53.7)
23.2 (4.2)
100
10.7 (10.8)
P4 32 12
97.70, 97.70, 173.95
87.0 -2.6
26704
10.5 (50.8)
19.1 (5.2)
100
23.2 (22.9)
P3 121
148.29, 148.29, 182.41
48.5-2.1
180938
7.5 (47.5)
22.4 (4.0)
99.6
5.4 (4.9)
43.2-1.7
16.7/18.8
1543
1434
109
0
64.2-2.6
19.4/23.7
4619
4326
139
154
48.5-2.1
19.1/22.6
10138
9498
640
0
22.4
21.8
29.8
34.6
34.4
37.7
98.5/1.5/0
48.3
47.6
44.9
71.5
96.1/3.6/0.3
0.006
1.13
4M98
0.006
1.16
4M99
0.007
1.10
4M9C
Data Collection
space group
unit cell dimension (a, b, c) (A)
resolution (A)
no. of observed reflections
Rsym (%)a b
I/ -i
completeness (%)
redundancy
Refinement
Resolution
Rwork/Rfar (%)C
total no. of atoms
protein
water
ligands
B factors
(A2)
overall
protein
water
ligand
Ramachandran plot (%)d
r.m.s.d.
bond lengths (A)
bond angles (0)
PDB code
96.3/3.2/0.5
'Statistics for the highest resolution bin are in parentheses. bR,. = YJI- / I,where I is the intensity of a reflection and Iis the
mean intensity of a group of equivalent reflections. CRwork = &DIF(h)bS - F(h)cajI/YhF(h)bI. Rfee was calculated for 5% of the
reflections randomly excluded from the refinement. dRamachandran plot statistics are given as core/allowed/generously allowed
and are for all chains.
153
Structure Determinationand Refinement
Preliminary electron density maps for the PglB-ATD, PglB-ATD-AcCoA, and Weel
structures were generated in PHASER (32) utilizing the previously solved PglD structure
(3BSW) (10) as the molecular replacement search model. Refinement and model building of
each structure was accomplished with COOT (33) and PHENIX (34). Water molecules were
added using COOT and the AcCoA ligand was modeled into PglB-ATD after the Rfree value was
< 30%. Refined structures were validated using MolProbity (35). Composite omit maps for the
AcCoA-bound PglB-ATD structure were generated with PHENIX.
The final refinement
statistics are listed in Table 4-3. Omit maps were generated with PHENIX to check for model
bias.
AcetyltransferaseActivity Assay
Enzyme mutants were analyzed for activity utilizing a DTNB spectrophotometric assay
as described previously (4). Briefly, each assay was carried out at 50 mM HEPES pH 7.4, 2 mM
MgCl 2 , 0.05% BSA, 0.001% Triton X-100, and 1 mM DTNB. The substrate concentrations of
AcCoA and UDP-4-amino were varied separately while holding the other substrate at 2 mM.
Reactions were completed in duplicate and initial rates were measured in the linear portion of the
reaction curve over a 5 min time period at 25 'C.
PhylogeneticAnalysis of UDP-diNAcBac Acetyltransferases
Bacterial organisms containing the UDP-diNAcBac pathway were identified using the
respective oligosaccharyltransferases
from C. jejuni (YP_002344519.1), N. gonorrhoeae
(YP_207345.1), and A. baumannii (YP_002324267.1). Further selection of the relevant
154
acetyltransferases relied on a > 35% sequence identity cutoff in BLASTP (36) with PglD (C.
jejuni; YP_002344516.1), PglB-ATD (N. gonorrhoeae;YP_207258.1) and WeeL (A. baumannii;
YP_001715524.1). Acetyltransferase sequences were aligned simultaneously with the software
program MUSCLE (37) using a gap-opening penalty of -2.9, a gap extend penalty of 0, and a
hydrophobicity multiplier of 1.2. Phylogenetic trees were constructed utilizing the neighborjoining method (38) and Poisson model (39) with MEGA 5.2 (40). The confidence level of this
process was estimated using a bootstrap analysis with 1000 replicate data sets.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
Szymanski, C. M., Burr, D. H., and Guerry, P. (2002) Campylobacter protein glycosylation
affects host cell interactions. Infect. Immun. 70, 2242-2244.
Hendrixson, D. R., and DiRita, V. J. (2004) Identification of Campylobacterjejuni genes
involved in commensal colonization of the chick gastrointestinal tract. Mol. Microbiol. 52,
471-484.
Szymanski, C. M., and Wren, B. W. (2005) Protein glycosylation in bacterial mucosal
pathogens. Nat. Rev. Microbiol. 3, 225-237.
Morrison, M. J., and Imperiali, B. (2013) Biosynthesis of UDP-N,N'-diacetylbacillosamine
in Acinetobacter baumannii: biochemical characterization and correlation to existing
pathways. Arch. Biochem. Biophys. 536, 72-80.
Hartley, M. D., Morrison, M. J., Aas, F. E., Borud, B., Koomey, M., and Imperiali, B. (2011)
Biochemical Characterization of the O-Linked Glycosylation Pathway in Neisseria
gonorrhoeae Responsible for Biosynthesis of Protein Glycans Contdining NN Diacetylbacillosamine. Biochemistry 50, 4936-4948.
Young, N. M., Brisson, J. R., Kelly, J., Watson, D. C., Tessier, L., Lanthier, P. H., Jarrell, H.
C., Cadotte, N., St Michael, F., Aberg, E., and Szymanski, C. M. (2002) Structure of the Nlinked glycan present on multiple glycoproteins in the Gram-negative bacterium,
Campylobacterjejuni.J. Biol. Chem. 277, 42530-42539.
Glover, K. J., Weerapana, E., and Imperiali, B. (2005) In vitro assembly of the
undecaprenylpyrophosphate-linked heptasaccharide for prokaryotic N-linked glycosylation.
Proc. Natl. Acad. Sci. USA 102, 14255-14259.
Aas, F. E., Vik, A., Vedde, J., Koomey, M., and Egge-Jacobsen, W (2007) Neisseria
gonorrhoeae O-linked pilin glycosylation: functional analyses define both the biosynthetic
pathway and glycan structure. Mol. Microbiol. 65, 607-624.
Iwashkiw, J. A., Seper, A., Weber, B. S., Scott, N. E., Vinogradov, E., Stratilo, C., Reiz, B.,
Cordwell, S. J., Whittal, R., Schild, S., and Feldman, M. F. (2012) Identification of a general
O-linked protein glycosylation system in Acinetobacter baumannii and its role in virulence
and biofilm formation. PLoS Pathog. 8, e1002758.
155
10. Olivier, N. B., and Imperiali, B. (2008) Crystal structure and catalytic mechanism of PglD
from Campylobacterjejuni.J. Biol. Chem. 283, 27937-27946.
11. Rangarajan, E. S., Ruane, K. M., Sulea, T., Watson, D. C., Proteau, A., Leclerc, S., Cygler,
M., Matte, A., and Young, N. M. (2008) Structure and active site residues of PglD, an Nacetyltransferase from the bacillosamine synthetic pathway required for N-glycan synthesis
in Campylobacter jejuni. Biochemistry 47, 1827-1836.
12. Kelly, J., Jarrell, H., Millar, L., Tessier, L., Fiori, L. M., Lau, P. C., Allan, B., and
Szymanski, C. M. (2006) Biosynthesis of the N-linked glycan in Campylobacterjejuni and
addition onto protein through block transfer. J. Bacteriol. 188, 2427-2434.
13. Thoden, J. B., Cook, P. D., Schaffer, C., Messner, P., and Holden, H. M. (2009) Structural
and functional studies of QdtC: An N-acetyltransferase required for the biosynthesis of
dTDP-3-acetamido-3,6-dideoxy-a-D-glucose. Biochemistry 48, 2699-2709.
14. Thoden, J. B., and Holden, H. M. (2010) Molecular structure of WlbB, a bacterial Nacetyltransferase involved in the biosynthesis of 2,3-diacetamido-2,3-dideoxy-D-mannuronic
acid. Biochemistry 49, 4644-4653.
15. Thoden, J. B., Reinhardt, L. A., Cook, P. D., Menden, P., Cleland, W. W., and Holden, H. M.
(2012) Catalytic mechanism of perosamine N-acetyltransferase revealed by high-resolution
x-ray crystallographic studies and kinetic analyses. Biochemistry 51, 3433-3444.
16. Dutnall, R. N., Tafrov, S. T., Sternglanz, R., and Ramakrishnan, V. (1998) Structure of the
histone acetyltransferase Hatl: a paradigm for the GCN5-related N-acetyltransferase
superfamily. Cell 94, 427-438.
17. Engel, C., and Wierenga, R. (1996) The diverse world of coenzyme A binding proteins. Curr.
Opin. Struct. Biol. 6, 790-797.
18. Remington, S., Wiegand, G., and Huber, R. (1982) Crystallographic refinement and atomic
models of two different forms of citrate synthase at 2.7 and 1.7 A resolution. J. Mol. Biol.
158, 111-152.
19. Raetz, C. R., and Roderick, S. L. (1995) A left-handed parallel beta helix in the structure of
UDP-N-acetylglucosamine acyltransferase. Science 270, 997-1000.
20. Nothaft, H., and Szymankski, C. M. (2013) Bacterial protein N-glycosylation: new
perspectives and applications. J. Biol. Chem. 288, 6912-6920.
21. Helenius, A., and Aebi, M. (2004) Roles of N-linked glycans in the endoplasmic reticulum.
Annu. Rev. Biochem. 73, 1019-1049.
22. Mitra, N., Sinha, S., Ramya, T.N., and Surolia, A. (2006) N-linked oligosaccharides as
outfitters for glycoprotein folding, form and function. Trends Biochem. Sci. 31, 156-163.
23. Nothaft, H., and Szymanski, C.M. (2010) Protein glycosylation in bacteria: sweeter than
ever. Nat. Rev. Microbiol. 8, 765-778.
24. Lees-Miller, R. G., Iwashkiw, J. A., Scott, N. E., Seper, A., Vinogradov, E., Schild, S., and
Feldman, M. F. (2013) A common pathway for 0-linked protein-glycosylation and synthesis
of capsule in A cinetobacter baumannii.Mol. Microbiol. doi: 10.111 1/mmi.12300.
25. Olivier, N. B., Chen, M. M., Behr, J. R., and Imperiali, B. (2006) In vitro biosynthesis of
UDP-N,N'-diacetylbacillosamine by enzymes of the Campylobacterjejuni general protein
glycosylation system. Biochemistry 45, 13659-13669.
26. Sievers F., Wilm A., Dineen D. G., Gibson T. J., Karplus K., Li W., Lopez R., McWilliam
H., Remmert M., Sbding J., Thompson J. D., Higgins D. G. (2011). Fast, scalable generation
of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol.
7:539 doi:10.1038/msb.2011.75.
156
27. Vallenet, D., Nordmann, P., Barbe, V,, Poirel, L., Mangenot, S., Bataille, E., Dossat, C., Gas,
S., Kreimeyer, A., Lenoble, P., Oztas, S., Poulain, J., Sigurens, B, Robert, C., Abergel, C.,
Claverie, J. M., Raoult, D., Medigue, C., Weissenbach, J., Cruveiller, S. (2008) Comparative
analysis of Acinetobacters: three genomes for three lifestyles. PLoS One 3 e1805.
28. Otwinowski, Z., and Minor, W. (1997) Processing of X-ray Diffraction Data Collected in
Oscillation Mode. Methods Enzymol. 276, 307-326.
29. Leslie, A. G. W., and Powell, H. R. (2007) Processing diffraction data with mosfim.
Evolving Methods for Macromolecular Crystallography. 245, 41-51.
30. Winn M. D., Ballard C. C., Cowtan K. D., Dodson E. J., Emsley P., Evans, P. R., Keegan, R.
M., Krissinel, E. B., Leslie, A. G., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu,
N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A., and Wilson, K. S. (2011)
Overview of the CCP4 suite and current developments. Acta. Crystallogr. D Biol.
Crystallogr. 67, 235-242.
31. French, S., and Wilson, K. (1978) On the treatment of negative intensity observations. Acta
Crystallogr. A34, 517-525.
32. McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., and
Read, R. J. (2007). Phaser crystallographic software. J. Appl. Cryst. 40, 658-674.
33. Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K. (2010) Features and development of
Coot. Acta Cryst. D66, 486-501.
34. Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J.
J., Hung, L. W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W.,
Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T.C., and Zwart, P.
H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure
solution. Acta Cryst. D66, 213-221.
35. Chen, V. B., Arendall, W. B. 3rd, Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G.
J., Murray, L. W., Richardson, J. S., and Richardson, D. C. (2010) MolProbity: all-atom
structure validation for macromolecular crystallography. Acta. Cryst. D66, 12-21.
36. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990) Basic local
alignment search tool. J. Mol. Biol. 215, 403-410.
37. Edgar, R. C. (2004) MUSCLE: multiple sequence alignment with high accuracy and high
throughput. Nucleic Acids Research 32, 1792-1797.
38. Saitou, N. and Nei, M. (1987) The neighbor-joining method: A new method for
reconstructing phylogenetic trees. Molecular Biology and Evolution 4, 406-425.
39. Zuckerkandl, E. and Pauling, L. (1965) Evolutionary divergence and convergence in
proteins. Edited in Evolving Genes and Proteins by V. Bryson and H.J. Vogel, pp. 97-166.
Academic Press, New York.
40. Tamura K., Peterson D., Peterson N., Stecher G., Nei M., and Kumar S. (2011) MEGA5:
Molecular Evolutionary Genetics Analysis using Maximum Likelihood, Evolutionary
Distance, and Maximum Parsimony Methods. Molecular Biology and Evolution 28, 27312739.
157
Chapter 5: Biochemical Characterization and Fragment-Based Inhibition of
the CampylobacterjejuniAminotransferase PglE.
158
Introduction
The first evidence that bacteria utilize an N-linked protein glycosylation system similar to
eukaryotic organisms was discovered in 1999 with the Gram-negative, epsilonproteobacterium
Campylobacterjejuni (1).
This pathogenic bacterium is the leading cause of gastroenteritis
worldwide and can result in Guillain-Barre syndrome following C. jejuni infection.
The
heptasaccharide resulting from this N-linked glycosylation pathway contains the highlymodified, bacterial sugar UDP-NN'-diacetylbacillosamine (UDP-diNAcBac) at the reducing
end. This nucleotide-activated monosaccharide is biosynthesized in C. jejuni from a dehydratase
(PglF), aminotransferase (PglE), and acetyltransferase (PglD) from UDP-N-acetylglucosamine
(UDP-GlcNAc) (2-4). Phospho-diNAcBac is first appended onto an undecaprenyl-phosphate
(Und-P) carrier by a phosphoglycosyltransferase, then an additional six sugars are appended to
form the final heptasaccharide (Figure 5-1).
The final destination of this glycan is to an
asparagine residue with the consensus sequence Asp/Glu-X1 -Asn-X 2-Ser/Thr (where X, and X2
represent any amino acid except Pro) on proteins located in the periplasm (5). To date, over 60
proteins have been identified that contain this heptasaccharide modification.
159
PgIF
OH
Ho%
PgIE
"0
ACNAD*NI
O-UDP
UDP-GIcNAc
HO
H20
P
CH
6
H
AO"APH
O-UDP
A
a-KG L GIu
UDP4-keto
PgID
CKS
ACHN
O-UDP
AcCoA CoA
A1:N
UDP-diNAcBac
UDP4-amino
Campylob acterjejuni
UDP-
UDP4
CyP9!I1m
-
0..H P.
P11
1
'
I'
t
;4-f-I-
renpasm
*N-acetylglucosamine
W NN-diacetylbacillosmmine
* Glucose
N-acetylgalactosamine
Undecaprenyldiphosphate
PEB3
Figure 5-1. C. jejuni N-linked protein glycosylation pathway.
The C. jejuni aminotransferase PglE catalyzes the transfer of an amino group from L-
glutamate to the C4 position of the UDP-keto (UDP-2-acetamido-4-keto-2,4,6-trideoxy-a-Dglucose) sugar in a pyridoxal-5'-phosphate (PLP)-dependent manner to form UDP-4-amino
(UDP-2-acetamido-4-amino-2,4,6-trideoxy-a-D-glucose) (Figure 5-2). Catalysis is initiated by
the formation of an imine involving the UDP-4-keto sugar and pyridoxamine 5'-phosphate
(PMP). Following the formation of the external aldimine, the UDP-4-amino product is released
by transimination with the catalytic lysine residue in the active site.
The internal aldimine
resulting from this reaction is converted back to PMP through the conversion of L-glutamate to
a-ketoglutarate. Although the amino donor was determined to be glutamate, PglE also exhibited
moderate activity with methionine, glutamine, alanine, and cysteine (6).
160
H
yN
e
OPMP
HaN
1410
NH 2
0O
SIP
0
H
H2N
HNG
O2C
I
NHO
0
PMP
Lysl84
1Tautomnerzation
L-glutamate
0
O2HC
a
yCO?
NH2
H
1
Regeneration
H
G
9
Internal Aldimine
0
-
NH2
H2 N
O-UDP
C
Imine intermediat
0
R0
HN
Product Release
9
0-P
Lys184
1
-
0
H2 N
HN0
O-UDP
Lys184
mr-Pzt
HN
_______
HO
Lys184
-CO?
0
Z
O-UDP
UDP-4-keto
a-ketoglutarate
00-4
0
HN
.UDP
Lys184
T
II
Tautomnerizatlon
UDP4-amino
2O-UDP
Lysl84
External Aldimine
Figure 5-2. Proposed aminotransferase mechanism of PglE from C. jejuni.
Pathways related to pathogenicity (such as glycosylation) have become attractive antimicrobial targets due to the ever-increasing resistance to bactericidal antibiotics (7).
In the
context of pathogenicity, the N-linked protein glycosylation pathway in C. jejuni is a significant
area of interest. Importantly, this pathway produces a heptasaccharide containing diNAcBac that
targets a variety of proteins associated with virulence. In particular, the disruption of the pglE
gene has been previously examined for its effect on global N-linked protein glycosylation (8).
The loss of this gene resulted in a lack of detectable heptasaccharide in C. jejuni as determined
by NMR and whole-cell lysate reactivity to SBA lectin. Additionally, the pglE knockout strain
exhibited the inability to colonize intestinal tracts of 1-day-old chicks, thus validating this gene
as a possible pathogenicity target (8). In an analogous study in mice, the C. jejuni pglE mutant
impaired the ability of the bacteria to invade intestinal epithelial cells and to colonize intestinal
161
tracts (9). Transposon mutagenesis of C. jejuni verified these results by identifying pglE as an
essential gene for colonization of the chick gastrointestinal tract (10).
The inadequacies of screening libraries to find novel antibacterial inhibitors for
dangerous pathogens coupled with increasing drug resistance have resulted in the need for new
approaches in anti-microbial drug discovery (11).
One such approach is based upon the
screening and identification of small molecule compounds (MW < 250 Da) that can act as a
starting point for further elaboration. This fragment-based approach relies on the identification
of weak affinity inhibitors (Kd > 0.1 mM) and the ability to extend the molecule into proximal
sites to increase overall potency (12).
Despite their weak affinities, fragments possess the
necessary binding energy to overcome the significant barrier caused by the loss of rigid body
entropy upon protein binding (13).
This important phenomenon is monitored throughout the
fragment optimization process by the calculation of ligand efficiency (LE). Ligand efficiency
can be defined as LE = -AG/HA where AG is the free energy of binding of a compound
(estimated from its IC 50 ) and HA is the number of heavy (non-hydrogen) atoms in the compound
(14).
In order to obtain a small molecule (MW < 500 Da) with nanomolar potency, ligand
efficiency must be > 0.30 kcal/mol. This fragment-based methodology has previously proven to
be successful in the generation of novel, potent inhibitors to targets of therapeutic importance
(15-16).
A fragment approach involves three main steps:
an initial biophysical screen to
establish lead compounds that bind to an active site of interest, an in vitro activity assay to
determine compound potency, and X-ray crystallography to aid in the further optimization of
fragments into adjacent binding pockets (Figure 5-3).
Iterative rounds of synthesis, potency
validation, and visualization of active site binding through X-ray crystallography are necessary
to improve inhibitor efficacy. To test the hypothesis that PglE is a true pathogenic target in C.
162
jejuni, a small-molecule inhibitor is essential. Towards this goal, a fragment-based screening
effort was initiated to develop an inhibitor of PgLE.
Fluorescence-based thermal shift screen
Does the fragment stabilize PgIE?
NMR spectroscopy
Does the ligand displace the fragment?
In vitro activity assay
Does the fragment inhibit PgIE?
X-ray crystallography
Does the fragment bind in a druggable site?
Syn'tes
Figure 5-3. Fragment based approach for the development of novel, potent inhibitors to targets
of therapeutic importance.
This chapter details the structural and biochemical characterization of PglE with the goal
of identifying small molecule inhibitors using a fragment-based screening approach.
This
enzyme was screened with a series of small-molecule fragments to ascertain their ability to
inhibit the aminotransferase reaction. Following the initial fragment hits, a series of secondgeneration analogs to the lead inhibitor were prepared and tested for activity in an in vitro assay.
Furthermore, this protein was crystallized and the structure solved for purposes of a fragmentbased inhibitor approach. This work establishes the ability for a small molecule to inhibit PglE
163
by binding into an allosteric pocket adjacent to PLP and lays the groundwork for future
generations of inhibitors for this protein.
Results and Discussion
Expression and Purificationof PglE
Full-length PglE was cloned from the NCTC 11168 genomic DNA of C. Jejuni and
ligated into a modified version of the pET-30b(+) vector. Each protein contained an N-terminal
His8 tag followed by a tobacco etch virus (TEV) protease site. Following overexpression in E.
coli BL21 RIL cells and purification with Ni-NTA resin, a 10 mg quantity was achieved from 1
L of culture. Purity for the PglE protein was assessed by SDS-PAGE to be > 95% (Figure 5-4).
Removal of the His8 tag was accomplished by the addition of 6 pM TEV protease for 24 hours at
4 0C. Anion exchange chromatography (Figure 5-5A) allowed for the removal of excess TEV
enzyme and the His 8 -TEV tag. Final purification via a sizing column (Figure 5-5B) resulted in
the removal of aggregated protein to yield 8 mg of the mono-dispersed PglE protein.
55 kDa
4mo45 kDa
Figure 5-4. 12% SDS-PAGE gel of PglE. Expected molecular weight of PgIE is 44 kDa.
164
A. -
B3.
I--------------
Figure 5-5. (A) PglE anion exchange column UV trace. (B) PglE size-exclusion column UV
trace.
PglE Enzyme CharacterizationandAssay Development
Multiple assay formats exist to monitor the transfer of the amine from L-glutamate to the
UDP-4-keto sugar. Capillary electrophoresis can be utilized to monitor turnover from UDP-4keto to UDP-4-amino. However, this method is low-throughput and reagent consuming due to
its discontinuous nature. The production of UDP-4-amino from the UDP-4-keto substrate can
also be coupled to the C. jejuni acetyltransferase
PglD and continuously monitored
spectrophotometrically at X = 412 nm through the turnover of Ellman's reagent to the TNB
chromophore (s = 14,150 M- cm-1) (Figure 5-6).
This method was previously utilized in
Chapter 2 to assay the acetyltransferases PglD, Weel and PglB-ATD and proved to be a robust
165
method for determining enzyme activity. The acetyltransferase PglD was kept in excess (1 pM)
to ensure that any UDP-4-amino generated from PglE was immediately converted to the UDPdiNAcBac. Excess acetyltransferase ensures that this part of the coupled reaction will not have
an effect on the rate of PglE reaction. Caution must be taken with this enzyme assay due to the
fact that inhibitor binding to PglD would still be able to decrease the overall rate of reaction.
PglE inhibitors were cross-examined using the PglD assay and found not to inhibit
acetyltransferase activity.
UDP-4-keto
L-Glu
PgIE
UDP-4-amino
AcCoAI PgtD
02 N Q
S-S
02N -
N02 + -SCoA
COO-
_OOC
_00C
DTNB
/S-S-CoA
+ -S
/NO
2
C00TNB2X = 412 nm
t = 14,150 M- cm'
Figure 5-6. Coupled enzymatic activity assay monitoring the turnover of UDP-4-keto with the
C. jejuni aminotransferase PglE.
Increasing amounts of PglE with 1 mM DTNB, 200 uM UDP-4-keto, 1 pM PglD, 5 mM
L-glutamate and 400 pM AcCoA resulted in the conclusion that Ellman's reagent was a viable
assay over a linear time range of 5 minutes with 100 nM of enzyme. The removal of PglE or the
UDP-4-keto substrate resulted in no detectable activity over the 10 minute assay duration.
Further work concentrated on the Michaelis-Menten binding constants for UDP-4-keto and Lglutamate. These experiments explored various concentrations of one substrate while holding
the other at saturating conditions (10 x Kn). Results illustrated that the Km for UDP-4-keto was
166
366 uM while that of L-glutamate was 10.7 mM (Figure 5-7). These values correlate well with
other diNAcBac aminotransferase enzymes (PglC and WeeJ) exhibiting the high level of
conservation of this reaction (Table 5-1). Further kinetic studies with PglE were completed at
the Km for both substrates. DMSO tolerance was tested against PglE to ensure that this solvent
did not have a deleterious effect on enzyme activity when measuring compound inhibition
(Figure 5-8A). The stability of PglE was also tested to ensure that the enzyme was stable after
multiple freeze thaw cycles (Figure 5-8B). DMSO (< 10%) did not have an inhibitory effect on
PglE. A final titration of enzyme yielded a linear response of product turnover at 100 nM PglE
over a time span of 5 minutes (Figure 5-9).
Further work with small molecule inhibitors were
explored at a concentration of 100 nM PglE with substrates fixed at Km.
0.12
-.
1
0.08
0.06
-
0.0
0.1
-
0.06
0.04
00.02
0i
0.02
0.02
0
400
800
1200
[UDP-4-keto] (pM)
1600
0
10
20
30
IL-Gul (mM)
40
50
Figure 5-7. Enzyme velocity as a function of substrate concentration fit to the equation rate =
Vmax[S]/[S] + Km. (A) Varying concentrations of UDP-4-keto with saturating levels of Lglutamate (100 mM). (B) Varying concentrations of L-glutamate with saturating levels of UDP4-keto (4 mM).
167
Table 5-1. Steady-state kinetic parameters for C. jejuni (C]), A. baumannii (Ab), and N.
gonorrhoeae(Ng) aminotransferase enzymes (values from Chapter 3).
kca/Km (M- s)
Km (pM)
kca (s_)
aminotransferase
substrate
UDP-4-keto
L-glutamate
UDP-4-keto
L-glutamate
UDP-4-keto
L-glutamate
PgIE (Cj)
PgIE (Cj)
WeeJ (Ab)
WeeJ (Ab)
PgIC (Ng)
PgIC (Ng)
6600
2.6
30
6.7
164
5.1
2.4 0.1
0.028 ± 0.0003
0.030 t 0.002
0.17 0.004
0.038 t 0.001
0.025 t 0.01
366 ± 57
11,000 ± 340
1003 ± 110
25,000 1900
233 35
4900 ± 900
(B) 0.o0
(A) 0.03
0.07
0.07
0.06
0.06
0.05
0.04
0.04
0.03
0.03
0.02
0.02
0.01
0.01
2.5
0
5
10
%OtMSO
OxFT
1xFT
2xFT
3x FT
Figure 5-8. (A) Enzyme activity over varying concentrations of DMSO. (B) Enzyme activity
after multiple freeze thaws (FT) of PglE.
1.4 .
1.2
1
(PgIE]
0.8
+ 400nM
N200nM
0.6
A lOOnM
x\SOWl
0.4
OnM
0.2J
n go
-
-
----------
-* --
ir i"
M, Ot
4
5
6
time (mI)
Figure 5-9. Final enzyme titration of PglE at the Km for the substrates. Varying concentrations
of protein monitored with Ellman's reagent over time.
168
PglE FragmentScreening Results
Through a collaboration with Professor Alessio Ciulli at the University of Cambridge
UK, a fluorescence based thermal shift assay was utilized for the screening of 660 fragment
molecules against PglE. Samples of fragment (4 mM) and protein (4.5 pM) were incubated with
Sypro Orange, which is a dye that fluoresces upon binding unfolded protein. The samples were
heated in a thermal cycler and the change in fluorescence was monitored as a function of
temperature.
Fragments that bind to protein have a stabilizing effect on protein melting
temperature leading to a greater ATm. Seven fragment hits (Figure 5-10) were further validated
by employing 'H WaterLOGSY and saturation transfer difference (STD) NMR spectroscopy.
These biophysical techniques are utilized as a secondary measurement of fragment binding.
I0
0
\/
o
/
OH
HN-N
\
HO
OH
-
-
H
HO
MB352
MB730
0
ATm= 2.16'C
ICS( = 5.86 mM
L.E. = 0.20
ATm = L37 C
1CS0 = 1.5 mM
LE. = 0.28
MB649
AT. 0.46*C
IC 0 => 2 mM
L.E. = NA
MB143
AT,= .31 OC
1C5 0 = 3.4 mM
LE. 0.28
ATm =0.97 "C
IC= 2.7 mM
L.E. = 027
<
/0
Q-SN~0
0
MB212
HO
0
MB048
ATm= 0 OC
IC50 =12 1DM
L.E. = 0.22
HO
\
MB744
AT,= 1.82 0 C
IC 50 =6.2 mM
L.E. = 0.22
Figure 5-10. PglE fragment IC 50 results. NA = Not Available due to compound solubility issues
and the inability to measure an IC 50 .
169
Small Molecule FragmentInhibition of PglEActivity
The use of Ellman's reagent, 5,5'-dithio-bis-(2-nitrobenzoic
previously utilized for PglE assay development.
analysis in the PglE DTNB assay (Figure 5-10).
acid) (DTNB)
was
The top seven hits were chosen for further
A 6-point, 3-fold compound dilution was
performed to measure PglE activity over time. A positive (DMSO only) and negative (no PglE)
controls were run in conjunction with each assay. Enzyme activity was monitored over time as a
function of inhibitor concentration. IC 50 values from the DTNB PglE assay are provided in
Figure 5-10. Based upon these initial findings, an additional seven fragments that resembled the
two lead compounds MB730 and MB143 were chosen for IC 50 follow-up.
However, these
compounds proved to be weak inhibitors of PglE with respect to previously tested compounds
(Figure 5-11). Interestingly, the furan analog of MB730 was 4-fold less active with respect to
the thiophene core. Although most of these molecules exhibited poor inhibition towards PglE
activity, MB730 demonstrated decent potency for such a small fragment and was therefore
chosen for further elaboration
170
H2N
N
NH 2
N
O
N
OH
MB603
MB718
MB040
IC 5 0=>2mM
IC 50 = >6 mM
1CjO=>6 mM
L.E. = NA
L.E. = NA
L.E. = NA
AT0544
IC 50 => 6 mM
L.E. = NA
0
\I
HO
IC
Figure 5-11.
MM-1
= 6.2 mM
L.E. = 0.22
50
I
HO
MM-2
IC 50 = 2.2 mM
L.E. = 0.26
MM-3
1C50=> 6 mM
L.E. = NA
PglE follow-up fragment IC 50 results. NA = Not Available due to compound
solubility issues and the inability to measure an IC 50 .
PgJE CapillaryElectrophoresisAssay Development
A more direct assay was pursued to ensure that the lead compound MB730 was a true
inhibitor of PglE and not an artifact of the DTNB assay. Capillary electrophoresis is a direct,
diagnostic method to monitor UDP-4-keto turnover and can easily separate the analytes of the
reaction.
Therefore, PglE aminotransferase activity was monitored by detection of the UDP
moiety on the sugar substrate and product at k = 254 nm. The same conditions from the DTNB
assay were replicated and percent substrate turnover was calculated by integrating the area under
the curve for the UDP-4-amino product. These values were plotted over time at an enzyme
concentration of 200 and 100 nM (Figure 5-12).
171
(A)
5 min
10 min
15 min
30 min
(B)
70~
80~
[P8IEJ
so
CL40
*ZoonM
10
0
0
40
60
$
time (min)
Figure 5-12. (A) Electropherogram trace of the turnover of UDP-4-keto to UDP-4-amino in the
presence of Pg1E. (B) PglE enzyme activity as a function of product turnover from (A).
Further experiments with capillary electrophoresis focused on a PglE concentration of
100 nM over a time span of 15 minutes. A 6-point IC 50 curve was initiated at a top concentration
of 6 mM MB730 with a three-fold compound dilution. Product turnover was monitored as a
function of inhibitor concentration (Figure 5-13).
The Pg1E capillary electrophoresis assay
resulted in an IC 50 of 520 pM for MB730, confirming the previous DTNB coupled assay result.
172
(A)
2mM
6mM
(B)a(I~~~
0.67 mM
0.22 mM
24'\
0.074 mM
0.025 mM
0
20
0 1H
MB730
IC5 0 (DTNB)= 1.5 mM
ICso (C.E.)= 0.52 mM
[M87301 (mM)
Figure 5-13. (A) Electropherogram trace of the turnover of UDP-4-amino as a function of
MB730 concentration. (B) MB730 IC 50 curve plotted as a function of product turnover from
(A).
Crystallizationof PglE
Previously, the PglE protein crystal structure was solved by Structural GenomiX Inc.
(1061, 1062, 1069) (17). The PglE construct (NCTC 11168) used for this structure differed
slightly (7 amino acid residues) from the protein characterized in this chapter and the His6 tag
was not removed.
Not surprisingly, crystals were not apparent when using the provided
crystallization conditions. Therefore, an attempt was made to crystallize various concentrations
(15, 10, and 5 mg/mL) of purified, monodispersed PglE. Three Hampton Research screens
(Index HT, Crystal Screen I/II, PEG/Ion) were prepared following size exclusion purification.
The only successful crystallization conditions were from the Hampton Research Index HTTM
screen that resulted in 2D needles. Multiple conditions from the Index screen produced varying
degrees of success, however several trends were observed.
173
Multiple salts (e.g., ammonium
sulfate and sodium chloride) were successful in crystallization of PglE. As the pH was raised
from 5.5 to 7.5, a decrease in crystallization transpired that culminated in the absence of crystals
at pH 8.5. Lower concentrations of PglE led to larger 2D needles with the absence of these
needles at higher concentrations (Figure 5-14).
15 mg/mL
10 mg/mL
5 mg/mL
Figure 5-14. PglE sitting drop crystals utilizing the Hampton Research Index HT Screen with
0.2 M ammonium sulfate, 0.1 M BIS-TRIS pH 6.5, and 25% w/v PEG 3350.
Further optimization using the hanging drop method with 10 and 5 mg/mL of PglE
resulted in similar 2D needles.
Optimization of the original condition that produced the 2D
needles (0.2 M ammonium sulfate, 0.1 M BIS-TRIS pH 6.5, and 25% w/v PEG 3350) was
accomplished with the Hampton Research Additive Screen. Unfortunately, this screen did not
produce any viable changes in crystal formation. Due to the presence of 2D needles as well as
the failure of multiple crystallization screens, streak seeding was explored as an alternative
crystallization technique. Streak seeding relies upon protein microcrystal to act as "seeds" for
crystallization of the target enzyme. 2D needle crystals were utilized as a seeding agent to form
larger crystals in a solution containing either 5 or 10 mg/mL PglE and crystallization buffer (0.2
M ammonium sulfate, 0.1 M BIS-TRIS pH 6.5, and 25% w/v PEG 3350). This technique proved
to be a viable format to crystal growth (Figure 5-15).
174
Figure 5-15. Representative PglE crystals obtained from the streak seeding technique.
As an alternative to streak seeding, seed beads (Hampton Research) were also explored
concurrently as an alternative crystallization technique.
Seed beads rely upon protein
microcrystal to act as "seeds" for crystallization of the target enzyme. Previously grown 2D
needles were utilized as a seeding agent to form larger crystals in a solution containing 10
mg/mL PglE and crystallization buffer (0.2 M ammonium sulfate, 0.1 M BIS-TRIS pH 6.5, and
25% w/v PEG 3350).
This technique proved to be a viable format to crystal growth that
resembled crystals from streak seeding (Figure 5-16).
These crystals were subjected to
cryoprotection conditions yielding an upper tolerance of 20% glycerol.
175
Figure 5-16. Representative apo PglE crystals obtained from the seed bead technique (10
mg/mL PglE, 0.1 M BIS-TRIS pH 6.5, 0.2 M (NH4 )2 SO 4, 25% w/v PEG 3350.
A dataset was collected at the National Synchrotron Light Source at Brookhaven National
Labs (beamline X6A) with the apo crystal form of PglE generated from seed beads (10 mg/mL
protein, 0.1 M BIS-TRIS pH 6.5, 0.2 M (NH 4 )2 SO 4 , 25% w/v PEG 3350).
Datasets were
processed using HKL2000 (18), MOSFLM (19), TRUNCATE (20-21), and SCALA (20).
Molecular replacement was then carried out using PHASER (22) with the C. jejuni PglE crystal
structure as the search model (98% sequence identity, 1069).
PglE was crystallized in the
monoclinic space group C2 with 16 protomers in the asymmetric unit to a resolution of 1.8 A
(Figure 5-17). This high number of protomers is due to the lack of apparent symmetry in this
crystal form. The aminotransferase cofactor PLP, was also evident in this map density adjacent
to lysine 184 (Figure 5-18). Refinement of the Pg1E structure was completed using PHENIX
(23) and COOT (24). Final data collection parameters can be found in Table 5-2. Protein
geometry as analyzed by MolProbity (35) has been refined to optimal values (Table 5-3).
176
Figure 5-17. Asymmetric unit containing eight PglE dimers.
Figure 5-18. 2(F, - F) electron density map of PLP in the PglE crystal structure at 1.8
resolution contoured to 1.5a.
177
A
Table 5-2. Apo PglE data collection and refinement statistics.
Data collection
Space group
Cell dimensions, a, b, c (A)
Resolution (A)
No. of observed reflections
Rsym (%)a
C2
192,192, 195
50-1.8
541469
8.5(31.6)
I/al
17.3 (3.7)
97.4 (93.6)
4.1 (4.0)
Completeness (%)
Redundancy
Refinement
Resolution (A)
42.4 - 1.8
16.2/19.2
Rwork/Rfree (%)
2
B-factors (A
)
Overall
PLP
Ramachandran plot (%)
r.m.s.d.
Bond lengths (A)
Bond angles (0)
14.1
16.5
98.7/1.3/0
0.003
0.720
a The number in parentheses represents the highest resolution bin.
Table 5-3. PglE optimized protein geometry from MolProbity.
Goal: <1%
Porrotamers
Ramachdran
Protein
Geometry
rs
.....
.Goal
<0.2%
R amachandiran favored
Goal: >98%
ICP deviations >0.25A
Goat 0
Residues with bad bonds:
Goat 0%
Resid
Goat<0.1%
s w-it bad mages:
PglE-MB730 Crystal Structure
PglE crystals were soaked with the inhibitor MB730 under the same conditions used to
solve the apo structure (10 mg/mL PglE, 0.1 M BIS-TRIS pH 6.5, 0.2 M (NH 4 )2SO 4 , 25% w/v
PEG 3350). MB730 was incubated at room temperature for one hour at 5 mM with PglE before
the addition of the seed bead micro-crystals. Crystals were observed that were similar in size
and shape as those without ligand. These crystals were soaked in 20% glycerol, looped, frozen,
178
and a dataset was collected at the Brookhaven National Laboratories (X6A beamline). Datasets
were processed as previously with the apo PglE structure. This protein crystallized into the C2
space group with a resolution of 2.2
A with
a total of 36 protomers in the asymmetric unit.
Surprisingly, the apo crystal structure of PglE without the addition of MB730 contained 16
protomers in the asymmetric unit. Even more surprising is the lack of electron density for the
aminotransferase cofactor PLP cofactor (Figure 5-19). Although not apparent in every PglE
copy in the crystal structure, electron density is present in an adjacent pocket to the PLP binding
site. Displacement of PLP may occur due to the binding of MB730 into this allosteric pocket.
The presence of 36 protomers in the asymmetric unit allowed for averaging of the MB730
density for placement of this ligand.
Figure 5-19. 2(F. - Fc) electron density map of PglE with bound MB730. The red circle
denotes the area where PLP density should be expected. The red arrow denotes Lys 184, the
amino acid residue responsible for binding to PLP. In this case, Lys184 picks up hydrogen
bonding interactions to the carbonyl oxygen of Asn1 83 in the absence of PLP.
179
MB730 is bound to PglE via a hydrogen-bonding network through the carboxylate
moiety of MB730 to the NH backbone of Glyl9l and Asnl8l as well as the carbonyl backbone
of Asn54 and hydroxyl group of Ser179 (Figure 5-20). A cavity (5 A distance to the back of the
pocket) beyond the phenyl moiety of MB730 may allow for further modification of this portion
of the molecule to further improve potency.
The back of this pocket contains Glul58 and
Asp155 that can be used to further extend the hydrogen bond network of MB730 analogs.
0
Asn181
0w
N
H
NH 2
3.!
2.8A
-
'HO'\
Ser179
S
2.9A
NH--------H
Gly191
---'
MB730
2.9 A
0
0
N - Asn54
H
H2 N
0
Figure 5-20. Major hydrogen-bond interactions between PglE and the MB730 fragment.
Second GenerationMB730 Analogs
Based on the crystal structure, several analogs of MB730 were designed with chemical
modifications at the carboxylic portion of the molecule.
The following three analogs were
synthesized: a methyl ester (MM-4), a methyl amide (MM-5), and primary alcohol (MM-6) of
MB730 (Figure 5-21). These compounds proved to be weak inhibitors of PglE with respect to
the original MB730 compound (IC 50 = 1.5 mM). Reliable IC 50 values were difficult to obtain
180
due to the insolubility of these analogs. The crystal structure of MB730 with PglE explains the
loss in potency of these analogs due to the derivatization of the carboxylate portion of this
molecule and disruption of the hydrogen-bonding network. The potency of the methyl amide
was similar to MB730 due to the fact that this analog does not disrupt the important interactions
with PglE.
0
0
S,
N
HO
('
Sj
S1
H
MM-4
IC
50 > 2 mM
MM-5
MM-6
IC50 = 1.8mM
ICs = 6mM
Figure 5-21. Second generation MB730 PglE inhibitors.
To develop more potent PglE inhibitors, further modifications were made on the phenyl
moiety of MB730 (Figure 5-22). To this effect, a route utilizing a biaryl Suzuki coupling was
undertaken to achieve the final compounds (Scheme 5-1, 5-2). Two separate catalysts were
utilized based upon the inability of tetrakis (Pd(PPh3)4 ) to catalyze certain reactions. The various
analogs of MB730 were synthesized and characterized by NMR (Figures 5-23, 5-24).
To
measure potency, compounds were diluted 3-fold with a starting concentration of 6 mM resulting
in an 8-point IC 5 0 curve. The MB730 analog compounds exhibited minor changes in potency
with modifications on the phenyl moiety of the molecule.
Most of the analogs had
modifications at the 3 and 4 position, which resulted in a decrease in potency. In contrast, the
addition of 2,6-dimethyl would force the molecule to adapt a conformation similar to the
MB730-bound PglE crystal structure (Figure 5-25).
When tested in the assay, this analog
exhibited a -40-fold increase in compound potency as compared to MB730.
181
0
OH
S\/
00 = 1500 uM
JCO
\/0
/
IC5 0=1260 uM
[C,,=840uM
-~0
OH
OH
IC50 = 2270 uM
HO
\/OH
S
\/
OH
IC,50 = 2600 uM
S
OH
IC5 0= 170 uM
5
IC50 = 1500 uM
H
6
IC50 = 1660 uM
ON
= OH
1C5 0= >6000 uM
=
\/-
OH
IC 5 = 1974 uM
0
H2NOH
IC50 = 3900 uM
IC
50=
2440 uM
Figure 5-22. PglE MB730 analog compounds with their respective IC 50 values.
Scheme 5-1. Synthetic route utilizing a Suzuki coupling to synthesize MB730 analogs with
tetrakis(triphenylphosphine)palladium(O).
HO
0
H0'
OH
0
Ar
ArBr )
I
Pd(PPh 3)4
OH1-
Scheme 5-2. Route utilizing a Suzuki coupling to synthesize MB730 analogs with silica-bound
DPP-Pd (SiliaCat DPP-Pd).
HO
HO(
0
I
ArBr
OH
b.A-S
SiliaCat DPP-Pd
i
DPP-Pd
-S
182
A
0
OH
(A)
(B)
0 2N
0
S
j.
\/
0
L
OH
0
/
.-
H.
.
..
..
\/
OH
..... .........
(C)
(D)
s
\ /
s0
\ /
L
.~
\ /
OH
OH
.............................
.. ....
.......
(E)0
s
0
. ..... .
,
.
.
.
.
.
- ..........
F
.
..
.
\ S
0
\I/
OH
........
...
----
0O
OHH
I.-
.......
........
-
I
..
. ...... .....
.
Figure 5-23. Final 1H-NMR for MB730 derivatives. (A) 4-nitro (24% Yield) (B) 4acetophenone (61% yield) (C) 4-methoxy (71% yield) (D) 4-methyl (75% yield) (E) 2,6dimethyl (52% yield) (F) 4-fluoro (48% yield).
183
(A)
(B)
0
N
\/
H2N
OH
0
S
OH
II~I
(C)
(D)
N
\/1
--
0
OH
S
\/
III
-
~~
OH
AI
(E)
0
HO
\/
OH
I.--
. . .............
.
Figure 5-24. Final 'H-NMR for MB730 derivatives. (A) 4-pyridinyl (28% yield) (B) 4-amino
(33% yield) (C) 3-pyridinyl (48% yield) (D) 3-methyl (78% yield) (E) 3-hydroxyl (52%
yield).
184
0
IC_5= 150 uM
\/
OH
Figure 5-25. MB730-bound PglE crystal structure. The phenyl moiety is rotated out of plane
with respect to the rest of the molecule. Synthesis of the 2,6-dimethyl analog resulted in a 10fold increase in potency. Distances to the catalytic lysine (K184) from the thiophene ring and
carboxylate are show in angstroms.
Conclusions
In conclusion, PglE was expressed, purified, and crystallized for the purposes of utilizing
a fragment-based approach to develop aminotransferase inhibitors. An assay was developed
utilizing a coupled DTNB spectrophotometric readout for measuring fragment inhibition of the
PglE aminotransferase reaction. A discontinuous, capillary electrophoresis assay that directly
measures product turnover was also developed to confirm the results of the DTNB assay. The
lead compound, MB730, was pursued via a synthetic approach to develop structure-activity
relationship (SAR) and to increase potency against its intended target, PglE.
To aid in this
pursuit, a crystal structure of MB730 bound to PglE was solved to 2.2 A resolution. Based upon
185
this structure, an analog of MB730 was synthesized that exhibited a ~10-fold increase in
potency.
Future experiments will focus on obtaining a structure of this compound and
developing more potent inhibitors based upon that crystal structure. Future PglE inhibitors will
be tested for their ability to inhibit glycosylation in C. jejuni. These inhibitors will be utilized
valuable tools for understanding the relationship between bacterial pathogenicity and
glycosylation.
Acknowledgments
I am extremely grateful to Professor Robert Sauer and Dr. Robert Grant for assistance with data
refinement and technical advice on crystallography. I would also like to thank Professor Alessio
Ciulli for the generation of the initial PglE fragment leads. I would like to thank Dr. Meredith
Hartley for critical reading of this chapter and Austin Travis for advice on chemical synthesis
and critical reading of this chapter.
Experimental Procedures
Common materials
All chemicals were purchased from Sigma-Aldrich unless otherwise stated. The UDP-4keto sugar was biosynthesized as described previously from the C. jejuni enzyme PglF130 (4).
PglB-A TD Molecular Biology
The acetyltransferase domain (ATD) of the pglB gene from N. gonorrhoeaeFA1090 was
identified through a Clustal Omega alignment (26) with the C. jejuni acetyltransferase (PglD).
The gene encoding this domain was amplified via the polymerase chain reaction (PCR) with the
186
forward primer 5'-CGCGGATCCATGGCGGGGAATCGCAAACTCG-3'
primer 5'-GCAACCCGGCAAAGCCCCTTTAGCTCGAGCGG-3'
and the reverse
from the N. gonorrhoeae
FA1090 strain (8).
The modified pET30b(+) plasmid containing the pg/E gene was used to transform
Escherichiacoli BL2 1 (DE3) RIL competent cells (Stratagene). One liter of LB media containing
50 pg/mL kanamycin and 30 ptg/mL chloramphenicol was inoculated with 8 mL of an overnight
culture of cells. The cells were then allowed to grow at 37 0C while shaking until an optical
density of-0.8 (k = 600 nm) was reached. The culture was cooled to 16 0C and induced with 0.5
mM iso-p-D-thiogalactosylpyranoside (IPTG). After incubating for 18 h with shaking at 16 C,
the cells were harvested by centrifugation (2600g, 30 min) and stored at -80 C until needed.
PgE Purification
Each protein purification step was carried out at 4 0C.
The cell pellet (~3 g) was
resuspended in 40 mL of 50 mM HEPES pH 7.4/100 mM NaCl/30 mM imidazole (Buffer A)
and then lysed by sonication. The lysate was then cleared by centrifugation (145,000g, 60 min)
and added to 2 mL of Ni-NTA resin (Qiagen). The slurry was allowed to tumble for 3 h and then
packed into a fritted PolyPrep column (Biorad). The resin was washed with 20 column volumes
of Buffer A and then eluted with a buffer containing 50 mM HEPES pH 7.4/100 mM NaCl/ 300
mM imidazole. Fractions containing the purified protein as analyzed by SDS-PAGE were
pooled and dialyzed against 50 mM TRIS pH 8.0/5 mM EDTA/5 mM P-mercaptoethanol in the
presence of 6 pM TEV protease for 24 h to remove the His8 tag. Removal of this tag was
monitored by Western blot analysis using an anti-His 4 antibody (Qiagen). The reaction was
diluted 10-fold in 25 mM HEPES pH 7.6 and excess TEV was then removed with a HiTrap
187
Q
HP Sepharose anion exchange column (GE Healthcare) utilizing a linear NaCl gradient.
Fractions containing the protein were pooled and dialyzed for 24 h in 50 mM HEPES pH 8.0/150
mM NaCl (SEC buffer). After concentrating to a volume of 1.5 mL using a 10K Da MWCO
Amicon Ultra-15 centrifugal filter unit (Millipore), the protein was loaded onto a Superdex 200
16/60 column (GE Healthcare) and subjected to size-exclusion chromatography in SEC buffer.
Fractions
containing
the monodispersed
crystallography experiments.
protein
were
pooled
and concentrated
for
Protein concentration were calculated based upon the predicted
extinction coefficients at k = 280 nm.
Crystallizationand Data Collection
All crystals were grown as hanging drops by combining 1.5 p.L of a 10 mg/mL PglE
solution in SEC buffer with 1.5 ptL of reservoir solution at 25 C. Each well contained a final
volume of 500 ptL of reservoir solution. The reservoir solution for apo PglE contained 0.1 M
BIS-TRIS pH 6.5, 0. 2 M ammonium sulfate, and 25% PEG 3350. After 24 h, a 2D crystal
needle was removed from the cover slip and placed into a microcentrifuge tube containing 50
pL of reservoir solution and a TeflonTM coated seed bead (Hampton Research). This solution
was vortexed for 90 s and then diluted 5-fold in reservoir solution. Hanging drops were then set
up containing 1.5 pL of a 10 mg/mL protein solution and 1.5 piL of diluted micro-crystals from
the seed bead treatment. After the crystals were fully grown (24 h), they were cryoprotected in
reservoir solution containing 20% glycerol. Diffraction data was collected on beamline X6A
(National Synchrotron Light Source, Brookhaven National Laboratory, Upton, NY) at 100 K.
Data sets were processed using HKL2000 (18), MOSFLM (19), TRUNCATE (20-21), and
SCALA (20). Parameters from the data collection are listed in Table 5-2.
188
Structure Determinationand Refinement
Preliminary electron density maps for the PglE structure were generated in PHASER (22)
utilizing the previously solved PglE structure (1061) (17) as the molecular replacement search
model. Refinement and model building of each structure was accomplished with PHENIX (23)
and COOT (24). Water molecules were added using COOT and the PLP cofactor was modeled
into PglE after the Rfree value was < 30%. Refined structures were validated using MolProbity
(25). The final refinement statistics are listed in Table 5-2 and 5-3.
PglE-MB730 Crystal Structure
The MB730-PglE bound crystal structure was accomplished utilizing the same procedure
as the apo structure. The only change to the procedure was the addition of 5 mM MB730 to 10
mg/mL PglE at 25 C for 1 h prior to mixing with the seed bead micro-crystals. Map averaging
was accomplished with the ccp4i software suite (20).
PglE DTNB Assay Development
The aminotransferase reaction was assayed by coupling generation of the UDP-4-amino
product from the PglE reaction to the acetyltransferase activity of PglD from C. jejuni. Ellman's
reagent (DTNB) was utilized to quantify substrate turnover as monitored by measuring
conversion of AcCoA to CoASH using the released TNB chromophore (Xmax = 412 nm, Xmax =
14,150 M~ cm'). In a black-clear 96-well plate (Costar) 50 mM HEPES (pH 7.4), 1 pM PglD,
400 pM AcCoA, 1 mM DTNB, and 100 nM PglE were added. Since PglE activity was coupled
to the turnover of the acetyltransferase PglD, addition of an excess amount of PglD ensured that
189
the initial velocity measurements were dependent only upon PglE activity. The concentrations of
L-glutamate and UDP-4-keto were varied separately to determine kinetic parameters using initial
velocity measurements while keeping the other substrate at saturation as calculated from
equation 1 using the program GraFit 6.0.12 (Erithacus Software). The absorbance at 415 nm was
followed on an Ultramark EX microplate imaging system (BioRad) continuously for a time
period of 10 minutes. A blank reaction without L-glutamate was set up as a background control.
To determine IC 50 values for fragment compounds, the substrate concentrations were fixed at Km
for both UDP-4-keto (366 pM) and L-glutamate (10.7 pM). IC 5 0 values were calculated from
equation 2 using the program GraFit 6.0.12 (Erithacus Software) and measured in duplicate.
V
y
Vmax[S]/(Km
=
+ [S])
100%/(1 + (x/IC 5o)')
(1)
(2)
PglE CapillaryElectrophoresisAssay Development
Capillary electrophoresis analysis was performed using a P/ACE MDQ system (Beckman
Coulter) with UV detection. The capillary was conditioned before each run successively with
0.4 M NaOH, water, and a 25 mM sodium tetraborate (pH 9.3) running buffer for 2 minutes
each. The reaction mixture contained 50 mM HEPES pH 7.4, 10.7 mM L-glutamate, 100 nM
PglE, and 366 pM UDP-4-keto. The reaction was initiated with enzyme and then quenched by
adding 16.7 pL reaction mix to 33.3 ptL H2 0 and boiling for 2 minutes. Each sample was then
prepared by filtration with a 10K MWCO membrane. The filtrate was injected for 15 s at 30
mbar and the analytes separated at 20 kV over a 45 minute time period on a bare silica capillary
190
(75 pm x 80 cm) with a 25 mM sodium tetraborate (pH 9.3) running buffer and monitored at a X
= 254 nm. Substrate and product peaks were manually integrated utilizing the Beckman 32
Karat software suite.
Synthesis of PglE MB730 Analogs
0
MeOH
H
S0
0
Ao-
4
MB730
MM-4
MB730-methyl ester (MM-4). To a solution of concentrated H2SO 4 (0.1 mL) in methanol (5
mL) was added MB730 (50 mg, 0.245 mmol), and the mixture was stirred under reflux for 2
hours. Methanol was then removed under reduced pressure. The residue was dissolved in ethyl
acetate, washed successively with saturated NaHCO 3, water, and brine, and then dried over
anhydrous MgSO 4 . The ethyl acetate solution was concentrated under reduced pressure to give a
residue, which was purified by silica gel column chromatography (5% ethyl acetate, 95%
hexanes) to give a purified solid (64% yield). 'H NMR (CDCl 3) (300 MHz) 6: 3.90 (s, 3H), 7.29
(d, lH), 7.40 (m, 3H), 7.65 (dd, 2H), 7.78 (d, IH).
NH 2CH 3
0
OH
S
0
CHHO
MB730
MM-5
MB730-methyl amide (MM-5). MB730 (100 mg, 0.49 mmol) was dissolved by stirring in
DMSO (50 mL). To this solution, 0.1 g (1.5 mmol) hydrochloride of methylamine, 0.15 mL (1.1
mmol) ethyldiisopropylamine, and 0.3 g (0.6 mmol) PyBOP were added.
The mixture was
stirred for 30 minutes and quenched with water (150 mL). Compound did not precipitate out of
191
solution, so the DMSO/H 2 0 mixture was extracted with ethyl acetate.
Ethyl acetate was
removed under reduced pressure to yield the reaction mixture in DMSO.
Flash column
chromatography using ethyl acetate:hexanes (1:1) gave the final compound (51% yield).
'H
NMR (CDCl 3) (300 MHz) 5: 5.91 (s, 2H), 7.29 (d, lH), 7.40 (m, 3H), 7.65 (dd, 2H), 7.78 (d,
1H).
BH3
/
S
OH
H3C ' CH 3
\I
OH
THF
MB730
S
/
\/
OH
MM-6
MB730-alcohol (MM-6). MB730 (100 mg, 0.49 mmol) was dissolved in dry tetrahydrofuran
(1.5 mL) under nitrogen and cooled to -10 C. Borane dimethylsulfide complex (0.367 mL, 0.74
mmol) was added dropwise and the solution was allowed to warm to room temperature over 12
hours. The reaction was quenched by the addition of methanol (2 mL) and the trimethylborate
azeotropically removed under reduced pressure. The resultant residue was taken up in methanol
(2 mL) and the solvent removed under reduced pressure. This process was repeated three times.
Flash column chromatography using 5% ethyl acetate, 95% hexanes gave the final compound
(68% yield). 'H NMR (CDCl 3) (300 MHz) 6: 3.05 (d, 3H), 7.29 (d, lH), 7.40 (in, 3H), 7.65 (dd,
2H), 7.78 (d, 1H).
192
Br
THF/H 2O
Na0H
HO
OH
H
+
Pd(PPh) 4
5-phenylthiophene-2-carboxylic
acid.
To
OH
a solution containing
2 mmol of the 5-
carboxythiophene-2-boronic acid (Combi-Blocks) was added 2.4 mmol of the aryl bromide
(bromobenzene - Sigma) in 30 mL THF, 24 mL H 20, and 6 mL 1 M NaOH. The resulting
solution was flushed under argon for 10 minutes and 5 mol % (0.1 mmol) of Pd(PPh 3) 4 was
added.
The solution was sealed with a rubber septum, heated to 50'C, and stirred overnight.
After the reaction was complete, 30 mL of 1 M NaOH was added to the solution. This solution
was extracted 3x with dichloromethane (15 mL) to remove unreacted aryl bromide. The aqueous
layer was acidified using 6 M HCl until the final product precipitated out of solution. The solid
was filtered and washed with H 20 and hexanes. The solid was dried in vacuo overnight to afford
the final product. Final yield
=
78% 'H-NMR (300 MHz, CDCl 3): 6 7.88 (d, 1H), 7.65-7.68 (dd,
2H), 7.33-7.43 (in, 3H), 7.33 (d, 1H).
The above experimental procedure was utilized to
synthesize the 4-fluoro (48% yield), 4-acetophenone (61% yield), 4-nitro (24% yield), 4-methyl
(75% yield), 4-methoxy (71% yield) MB730 analogs.
HO
HO
%+
\/
0
00
-OH
+-
+
-i
ODPP-Pd
MeOH
K2C0 3
s/
O
OH
.n
5-(2,6-dimethylphenyl)thiophene-2-carboxylic
acid. To a solution containing 2.0 mmol of the
5-carboxythiophene-2-boronic acid (Combi-Blocks) was added 2.4 mmol of the iodo xylene
(Alfa Aesar) and 3 mmol K2CO 3 in 20 mL MeOH (0.1 M). The resulting solution was heated to
65 "C and refluxed for 10 minutes. Once the solution was homogeneous, 2 mol % (0.04 mmol -
193
154 mgs) of SiliaCat DPP-Pd (Silicycle) was added and refluxed overnight (17 h). After the
reaction was complete, the solution was extracted 3 times with dichloromethane (15 mL) to
remove unreacted aryl iodide. The aqueous layer was acidified using 6 M HCl until the final
product precipitated out of solution.
The aqueous solution was again extracted with
dichloromethane to solubilize the product into the organic layer. The solvent was removed and
the solid was dried in vacuo overnight to afford the final product. Final yield = 52% 'H-NMR
(300 MHz, d6-DMSO): 6 13.13 (s, 1H), 7.77 (d, 1H), 7.13-7.20 (in, 3H), 6.96 (d, 1H), 2.08 (s,
6H). The above experimental procedure was utilized to synthesize the 4-pyridinyl (28% yield),
4-amino (33% yield), 3-pyridinyl (48% yield), 3-methyl (78% yield), 3-hydroxyl (52% yield)
MB730 analogs.
References
1.
2.
3.
4.
5.
6.
Szymanski, C. M., Yao, R., Ewing, C. P., Trust, T. J., and Guerry, P. (1999) Evidence for a
system of general protein glycosylation in Campylobacter jejuni. Mol. Microbiol. 32, 10221030.
Schoenhofen, I.C., McNally, D.J., Vinogradov, E., Whitfield, D., Young, N.M., Dick, S.,
Wakarchuk, W.W., Brisson, J.R., and Logan, S.M. (2005) Functional characterization of
dehydratase/aminotransferase pairs from Helicobacter and Campylobacter. J. Biol. Chem.
281, 723-732.
Vijaykumar, S., Merkx-Jacques, A., Ratnayake, D.B., Gryski, I., Obhi, R.K., Houle, S.,
Dozois, C.M., and Creuzenet, C. (2006) Cj 1121c, a novel UDP-4-keto-6-deoxy-GlcNAc C-4
aminotransferase essential for protein glycosylation and virulence in Campylobacterjejuni. J.
Biol. Chem. 281, 27733-27743.
Olivier, N.B., Chen, M.M., Behr, J.R., and Imperiali, B. (2006) In vitro biosynthesis of UDPN,N'-diacetylbacillosamine by enzymes of the Campylobacter jejuni general protein
glycosylation system. Biochemistry 45, 13659-13669.
Kowarik, M., Young, N.M., Numao, S., Schulz, B.L., Hug, I., Callewaert, N., Mills, D.C.,
Watson, D.C., Hernandez, M., Kelly, J.F., Wacker, M., and Aebi, M. (2006) Definition of the
bacterial N-glycosylation site consensus sequence. EMBO J. 25, 1957-1966.
Vijaykumar, S., Merkx-Jacques, A., Ratnayake, D.B., Gryski, I., Obhi, R.K., Houle, S.,
Dozois, C.M., and Creuzenet, C. (2006) Cjl 121c, a novel UDP-4-keto-6-deoxy-GlcNAc C-4
aminotransferase essential for protein glycosylation and virulence in Campylobacterjejuni. J.
Biol. Chem. 281, 27733-27743.
194
7. Clatworthy, A.E., Pierson, E., and Hung, D.T. (2007) Targeting virulence: a new paradigm
for antimicrobial therapy. Nat. Chem. Biol. 3, 541-548.
8. Kelly, J., Jarrell, H., Millar, L., Tessier, L., Fiori, L.M., Lau, P.C., Allan, B., and Szymanski,
C.M. (2006) Biosynthesis of the N-Linked glycan in Campylobacter jejuni and addition onto
protein through block transfer. J. Bacteriol. 188, 2427-2434.
9. Szymanski, C.M., Burr, D.H., and Guerry, P. (2001) Campylobacter protein glycosylation
affects host cell interactions. Infect. Immun. 70, 2242-2244.
10. Hendrixson, D.R., and DiRita, V.J. (2004) Identification of Campylobacter jejuni genes
involved in commensal colonization of the chick gastrointestinal tract. Mol. Microbiol. 52,
471-484.
11. Payne, D.J., Gwynn, M.N., Holmes, D.J., and Pompliano, D.L. (2007) Drugs for bad bugs:
confronting the challenges of antibacterial discovery. Nat. Rev. Drug Discov. 6, 29-40.
12. Ciulli A., and Abell, C. (2007) Fragment-based approaches to enzyme inhibition. Curr.
Opin. Biotechnol. 18, 489-496.
13. Page, M.I., and Jencks, W.P. (1971) Entropic contributions to rate accelerations in enzymic
and intramolecular reactions and the chelate effect. Proc. Natl Acad. Sci. 68, 1678-1683.
14. Hopkins, A., Groom, C.R., and Alex, A. (2004) Ligand efficiency: a useful metric for lead
selection. Drug Discovery Today 9, 430-431.
15. Saxty, G., Woodhead, S.J., Berdini, V., Davies, T.G., Verdonk, M.L., Wyatt, P.G., Boyle,
R.G., Barford, D., Downham, R., and Garrett, M.D. (2007) Identification of inhibitors of
protein kinase B using fragment-based lead discovery. J. Med. Chem. 50, 2293-2296.
16. Howard, N., Abell, C., Blakemore, W., Chessari, G., Congreve, M.S., Howard, S., Jhoti, H.,
Murray, C.W., Seavers, L.C.A., and van Montfort, R.L.M. (2006) Application of fragment
screening and fragment linking to the discovery of novel thrombin inhibitors. J. Med. Chem.
49, 1346-1355.
17. Badger, J., Sauder, J.M., Adams, J.M., Antonysamy, S., Bain, K., Bergseid, M.G., Buchanan,
S.G., Buchanan, M.D., Batiyenko, Y., Christopher, J.A., Emtage, S., Eroshkina, A., Feil, I.,
Furlong, E.B., Gajiwala, K.S., Gao, X., He, D., Hendle, J., Huber, A., Hoda, K., Kearins, P.,
Kissinger, C., Laubert, B., Lewis, H.A., Lin, J., Loomis, K., Lorimer, D., Louie, G., Maletic,
M., Marsh, C.D., Miller, I., Molinari, J., Muller-Dieckmann, H.J., Newman, J.M., Noland,
B.W., Pagarigan, B., Park, F., Peat, T.S., Post, K.W., Radojicic, S., Ramos, A., Romero, R.,
Rutter, M.E., Sanderson, W.E., Schwinn, K.D., Tresser, J., Winhoven, J., Wright, T.A., Wu,
L., Xu, J. and Harris, T.J.R. (2005) Structural analysis of a set of proteins resulting from a
bacterial genomics project. Proteins, 60, 787-796.
18. Otwinowski, Z., and Minor, W. (1997) Processing of X-ray Diffraction Data Collected in
Oscillation Mode. Methods Enzymol. 276, 307-326.
19. Leslie, A. G. W., and Powell, H. R. (2007) Processing diffraction data with mosfim.
Evolving Methods for Macromolecular Crystallography. 245, 41-51.
20. Winn M. D., Ballard C. C., Cowtan K. D., Dodson E. J., Emsley P., Evans, P. R., Keegan, R.
M., Krissinel, E. B., Leslie, A. G., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu,
N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A., and Wilson, K. S. (2011)
Overview of the CCP4 suite and current developments. Acta. Crystallogr. D Biol.
Crystallogr. 67, 235-242.
21. French, S., and Wilson, K. (1978) On the treatment of negative intensity observations. Acta
Crystallogr. A34, 517-525.
195
22. McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., and
Read, R. J. (2007). Phaser crystallographic software. J. Appl. Cryst. 40, 658-674.
23. Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J.
J., Hung, L. W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W.,
Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T.C., and Zwart, P.
H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure
solution. Acta Cryst. D66, 213-221.
24. Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K. (2010) Features and development of
Coot. Acta Cryst. D66, 486-501.
25. Chen, V. B., Arendall, W. B. 3rd, Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G.
J., Murray, L. W., Richardson, J. S., and Richardson, D. C. (2010) MolProbity: all-atom
structure validation for macromolecular crystallography. Acta. Cryst. D66, 12-21.
26. Sievers F., Wilm A., Dineen D. G., Gibson T. J., Karplus K., Li W., Lopez R., McWilliam
H., Remmert M., S6ding J., Thompson J. D., Higgins D. G. (2011). Fast, scalable generation
of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol.
7:539 doi:10.1038/msb.2011.75.
196
Chapter 6: The Development of Inhibitors for the C. jejuni Acetyltransferase
PglD Utilizing a High-Throughput Screening Approach.
197
Introduction
Antimicrobial resistance was first discovered in 1947 just four years after the introduction
of penicillin to the general public (1).
Due to selective pressure on bacterial growth, the
emergence of new antibiotics has resulted in similar outcomes.
In order to circumvent the
evolutionary aspect of antibiotic resistance, a novel strategy for bacterial cessation must be
implemented.
One such strategy is the targeting of bacterial pathogenicity instead of more
traditional targets like the bacterial cell wall (penicillin) and protein production (erythromycin).
Virulent bacteria rely upon the ability to adhere, colonize, and invade the cell surface of a host in
order to generate a disease state.
The inhibition of these virulence factors would lead to
quiescence, limiting bacterial infection. Since the consequence of this inhibition is not bacterial
cell death, no selective pressure is applied to the system and resistance can be averted. Previous
work has focused on the characterization of the N-linked glycosylation pathway in the Gramnegative bacterium Campylobacterjejuni (2-8). C. jejuni is responsible for gastroenteritis in
humans that may result in the development of Guillain-Barre syndrome and owes its
pathogenicity to the formation of glycans on periplasmic and cell-surface proteins (9-11)
Assembly of this glycan occurs in a stepwise fashion on an undecaprenyl-phosphate isoprene
carrier following the biosynthesis of diNAcBac (PglF, E, D) resulting in the formation of the
heptasaccharide (Figure 6-1). The glycan product is then transported across the inner membrane
by the flippase PglK and transferred to either a nascent or fully-folded protein by the
oligosaccharyltransferase PglB.
Bacterial N-linked systems such as that in C. jejuni target
proteins located in the periplasm for glycosylation by the Asp/Glu-XI-Asn-X 2 -Ser/Thr sequon
(where X, and X2 represent any amino acid except Pro) (12). To date, over 60 proteins (many of
198
which are virulence targets) have been identified in C. jejuni that contain this heptasaccharide
modification (13).
ON
HOx
AcN
Pg1F
O-UDP
UDP-GIcNAc
PgIE
C
NAD*
AcH
H20
C
A
1
O-UDP
UDP4-keto
AH
,
*vT
Periplasm
W
0N-aylglucosamine
V
ACOA ODA
CH
-UDP
UDP-dINAcBac
acterjejuni
LCampylob
UDP4
PgIC~~~~IV
PgA~~~IH
lasm
,J
O-UDP
CH
-
I'
UDP4-amlno
o-KG L-Ok
S
UDP
PgD
4r
7#
-
V.Pid
I
II
PgIK
75-Wr 07 ''1
-- ----
- ----- --
--W
W4W ---- N
A Sa.0-0WOVI
EN-acetylgalactosamine
N-diacetylbacillosamine
* Glucose
Undecwprnyldiphosphate
PEB3
Figure 6-1. C. jejuni N-linked protein glycosylation pathway highlighting the acetyltransferase
PglD. PEB3 is a known virulence factor targeted in the periplasm by this glycosylation system.
The first two steps of the UDP-diNAcBac biosynthetic pathway utilize an NAD+dependent dehydratase (PglF) and PLP-dependent aminotransferase (PglE) to produce the UDP4-amino (UDP-2-acetamido-4-amino-2,4,6-trideoxy-a-D-glucose)
sugar.
The final step of
diNAcBac biosynthesis relies upon PglD to acetylate the C4 position on the UDP-4-amino sugar
in an acetyl coenzyme A (AcCoA)-dependent reaction. This reaction is catalyzed by an active
site histidine (His 125) that acts as a general base to abstract a proton from the C4 amine resulting
in nucleophilic attack on the thioester of AcCoA.
The carboxylate moiety of the adjacent
glutamate residue (Glul26) is hydrogen-bonded to the imidazole ring of histidine, increasing the
basicity of the Nc2 nitrogen (14).
PglD forms a homotrimer in solution based upon
sedimentation velocity analytical ultracentrifugation (AUC) experiments and protein crystal
199
structures (14-15). The C-terminal left-handed P-helix domain of adjacent protomers forms the
AcCoA binding pocket, while the N-terminal domain contains a P--a-p-a--p-a Rossmann fold
motif to accommodate the UDP-4-amino substrate (Figure 6-2).
Figure 6-2. The composite C. jejuni PglD acetyltransferase crystal structure constructed from the
UDP-4-amino (PDB code 3BSS) (depicted in brown) and AcCoA (PDB code 3BSY) (depicted
in gray) bound structures. For the purpose of clarity, the 2 additional binding pocket substrates
have been removed and the protomers individually colored. The biological unit is a trimer
illustrated in cartoon with substrates as spheres (left) and space-filling with substrates as sticks
(right) form.
Due to the increase in resistance to bactericidal antibiotics from selective pressure,
developing therapies that target pathways related to pathogenicity (such as glycosylation) have
become an important strategy (16).
In the context of pathogenicity, the N-linked protein
glycosylation pathway in C. jejuni is a significant area of interest. This pathway produces a
heptasaccharide containing diNAcBac that modifies a variety of proteins associated with
virulence, such as PEB3 and VirB10. Previous studies have examined the importance of global
N-linked protein glycosylation by disrupting the pglD gene responsible for the final step in
diNAcBac biosynthesis (17). Utilizing high-resolution magic angle spinning nuclear magnetic
resonance (HR-MAS NMR) and whole-cell lysate reactivity to SBA lectin, the authors
200
concluded that loss of the pglD gene resulted in the significant reduction of glycosylation in C.
jejuni. Additionally, the ApglD strain of C. jejuni was unable to colonize the intestinal tracts of
1-day-old chicks.
This result validates the hypothesis that inactivation of this glycosylation
pathway through the knockout of diNAcBac biosynthetic genes attenuates pathogenicity of this
bacterium.
To better understand the effects that UDP-diNAcBac biosynthesis has on glycosylation in
relation to pathogenicity, an inhibitor of the PglD acetyltransferase of C. jejuni was sought after.
Towards this goal, a high-throughput screening (HTS) campaign was initiated with the Broad
Institute screening library containing a total of 359,000 small molecules.
Previous efforts
utilizing a fragment-based approach yielded small molecules bound in both the UDP-4-amino
and AcCoA binding pockets that inhibited PglD acetyltransferase activity to single-digit
millimolar.
Second generation analogs that were based on PglD-fragment crystal structures
produced a more potent series of small molecules (IC 50 ~ 300 pIM).
The goal of the HTS
screening project was to obtain a more potent inhibitor scaffold that could be utilized in
conjunction with the fragment-screening results to generate novel lead compounds.
To
accomplish this objective, a combination of in vitro activity assays, protein crystallography and
chemical synthesis was employed to create potent acetyltransferase inhibitors to study the
relationship between glycosylation and pathogenicity in C. jejuni.
201
Results and Discussion
Expression and Purificationof PglD
In preparation for the Broad screening campaign, full-length PglD from the NTCT 1118
strain of C. jejuni was ligated into the pET-24a(+) vector, which contained an N-terminal T7 tag
and a C-terminal His 6 tag for purification purposes. Following overexpression in E. coli BL21 (DE3)-RIL cells and purification with Ni-NTA resin, 28 mg of protein was obtained from 1 L of
culture. Purity for PglD was assessed by SDS-PAGE to be > 95% (Figure 6-3).
kDa
36
22
'f4
Figure 6-3. 12% SDS-PAGE gel of purified PglD protein for the Broad screen. The final protein
yield is 28 mg/L. The expected molecular weight of PglD is 24 kDa.
Assay Development of PglD
A continuous, in vitro assay relying on the generation of the TNB 2 chromophore from
Ellman's reagent (DTNB) was employed to measure acetyltransferase activity.
Previous
experiments showed that the Acinetobacter baumannii acetyltransferase WeeI relied upon MgCl 2
202
for optimal activity and binding of the UDP-4-amino substrate to the active site (Chapter 3).
Therefore, PglD activity was explored in the presence of divalent metals (Figure 6-4). These
results suggested that incorporation of MgCl 2 into the assay increased the basal level of
acetyltransferase activity over 3-fold.
The presence of 2 mM MgCl 2 resulted in typical
Michaelis-Menten kinetics over a range of AcCoA (1 - 0.02 mM) and UDP-4-amino (2 - 0.03
mM) concentrations while holding the other substrate at saturating conditions (10 x Kn) (Figure
6-5, Table 6-1).
Similar to Weel, PglD exhibited increased binding to UDP-4-amino in the
presence of MgCl 2 (4-fold decrease in Kin), whereas this divalent metal had no effect on AcCoA
binding. A final titration of enzyme at the respective substrate Km values resulted in a linear
assay at 0.2 nM of PglD for five minutes.
1.4
1.2
a MgC12
0.6
4 MnC12
3
CaCI2
0.4
0.2
10
5
2.5
1.25
0.625
ImetaImM)
0.313
0.16
0
Figure 6-4. PglD activity with respect to varying concentrations of MgCl 2 , MnCl 2 , and CaCl2.
Higher concentrations of MgCl 2 , and CaCl 2 led to a substantial increase in enzymatic turnover.
203
(B)
20000
16000
16000
CC
t12000
-12000
8000
-.
8000
-8000
4000
-4000
0
' ' '
0
' 'II' ' 'II''
400
' '
1
'.
800
1200
1600
[UDP-4-amino (pM)
2000
0
1'1 .'r'
400
.
800
1200
[AcCoA] (pM)
1600
2000
Figure 6-5. Kinetic characterization of the PglD substrates UDP-4-amino (A) and AcCoA (B).
Table 6-1. Steady-state kinetic parameters for the C. jejuni acetyltransferase PglD.
Acetyltransferases from A. baumannii (Weel) and N. gonorrhoeae(PglB-ATD) have been added
for comparison purposes.
acetyltransferase
substrate
PglD
PglD
Weel
Weel
PglB-ATD
PglB-ATD
UDP-4-amino
AcCoA
UDP-4-amino
AcCoA
UDP-4-amino
AcCoA
Kip
kca (S)
274 ± 6.4
295 ± 2.8
2520 540
78.9 28
99.0 7.1
286 35
8.0 ± 1.6 x I0
6.1 ± 1.0 x 10'
5.1 ± 0.08 x 10 5
1.3 0.06 x10 5
7.2 ± 0.8 x 104
5.0 ±0.7 x 10 4
kca/Km (M-' s_')
2.9
2.1
2.0
1.6
7.3
1.7
x 10 9
x 10 9
x 10"
x 10 9
x 108
x 10"
Concurrently, a capillary electrophoresis (CE) assay was established to confirm any highthroughput screening hits. This assay format relies on the separation of substrates (UDP-4amino and AcCoA) from products (UDP-diNAcBac and CoA) through their ionic charge and
hydrodynamic radius. Although this discontinuous assay is much lower throughput (1 hour per
sample), it allows for direct comparison of substrate turnover by integrating the area under each
electropherogram peak. Therefore, this assay is useful as a confirmatory tool for the indirect
DTNB assay. A PglD CE assay was established under identical substrate concentrations (at Km)
relative to the DTNB assay. Enzyme activity was linear over a 10 minute time course at 0.25 nM
of PglD with substrate turnover remaining under 20%.
204
Large-Scale Biosynthesis of UDP-4-Amino
The UDP-4-amino substrate for the Broad high-throughput screen is not commercially
available; therefore sufficient UDP-sugar (-600 mg) had to be biosynthesized to provide the
necessary material to accomplish this screen. A previous in-house laboratory procedure utilized
the C. jejuni enzymes PglF (dehydratase) and PglE (aminotransferase) to produce the desired
product. Given that the PglF enzyme contains 3 transmembrane helices as determined from a
hydropathy plot with TMHMM (18), protein purification is low-yielding (~300 pg/L).
Therefore, the transmembrane domain consisting of the first 130 amino acids was removed and
replaced with a GST-tag giving the soluble dehydratase domain (PglF
13 0).
This construct allows
for a greater protein yield following purification (30 mg/L) and retains the dehydratase activity
necessary for the generation of UDP-2-acetamido-4-keto-2.4.6-trideoxy-a-D-glucose
(UDP-4-
keto). The aminotransferase PglE is then utilized for the production of the final UDP-4-amino
sugar (UDP-2-acetamido-4-amino-2,4,6-trideoxy-a-D-glucose).
This final step is unable to
catalyze the complete conversion of the reaction leading to a 3:1 mixture of UDP-4-amino:UDP4-keto based upon CE (Figure 6-6).
Taken together with the fact that additional cofactors
(NAD+ and PLP) are necessary for full enzyme activity, a final HPLC purification step is
necessary to remove impurities from the reaction. Previous endeavors using this procedure were
completed on such a small scale (10 mg final yield) that further modification would be necessary
to obtain such a large amount of UDP-4-amino sugar for the HTS screen (Figure 6-7).
The
PglF130 reaction was examined for its ability to maintain activity without the addition of NAD+
during turnover of UDP-GlcNAc. Addition of 500 ptM NAD+ during each purification step and
extensive washes when bound to resin to remove any excess cofactor proved successful in
205
maintaining full dehydratase activity.
production of UDP-4-keto.
PglF130 remained on resin for the duration of the
This allowed for facile purification of the UDP-sugar from the
enzyme by filtering and washing to remove any of the product, which was non-specifically
bound to the protein purification resin. To obtain the upper limit of UDP-4-keto production,
increasing amounts of UDP-GlcNAc were added to the resin and the reaction followed by CE.
From this experiment, it was determined that a maximal loading of 100 mg of UDP-GlcNAc
would result in the complete conversion to UDP-4-keto from 1 L of purified PglF130. Due to the
inefficiency of the second step with PglE, aminotransferase enzymes from other species were
explored. Interestingly, the aminotransferase from N. gonorrhoeae(PglC) is able to completely
convert UDP-4-keto to UDP-4-amino (Figure 6-6). Similar to PglF 130, excess cofactor (PLP =
500 pM) was added throughout the purification of PglC and enzyme was bound to resin for the
duration of the reaction with UDP-4-keto. An upper limit of 50 mg of UDP-4-keto could be
added to 1 L of purified PglC enzyme for complete conversion to UDP-4-amino.
The only
additional substrate that was essential for the aminotransferase reaction was L-glutamate. To
circumvent a purification step, it was necessary to show that this substrate and its product (cLketoglutarate) did not have any deleterious effects on acetyltransferase activity for the HTS
screen. Both molecules did not exhibit any inhibition towards PglD at high concentrations of
these molecules (30 mM). The upper concentration limit of L-glutamate and c-ketoglutarate in
the assay using this purification process would be 1 mM. Finally, a comparison of UDP-4-amino
purified from the original method was compared to the new procedure presented here.
Acetyltransferase activity utilizing the DTNB assay was identical with both UDP-4-amino
substrates. Therefore, this modified procedure provides a facile route to the generation of large
quantities of UDP-4-amino substrate for HTS screening purposes.
206
0
HO-
aminotransferase
C
/-
H2N
\-,
HV
PMP PLP
>--<
A
O-UDP
UDP-4-keto
L-Glu
AcHN
O-UDP
UDP-4-amino
a-KG
0.14
-UUDP
- 20 hrs
P9g E(
01*11
PgIC(Ng)- 12 hrs
am.
P IC(N)-12 hrs
em-1
UDP-4ketospike
602
4i
-
$
-
7's
1$
12.
Wn
4's
200
V2*
M
2i.4
200
lio
200
"A
Figure 6-6. Electropherogram trace of UDP-4-amino biosynthesis from the aminotransferases
Pg1E(Cj) and PglC(Ng). To ensure the complete turnover of UDP-4-keto from the PglC(Ng)
reaction, purified UDP-4-keto was added following the CE run and reanalyzed.
Old Method:
UDP-GIcNAc
UDP-4-amino
UDP-4-keto
OH
HO
PgIF 130
O
HO4
AdHN
O-UDP
NAD*
NADH
H20
PgIE(Cj)
HO
AcHN
O-UDP
H2N
H
/\
PMP
PLP
6
AcHN
O-UDP
HPLC
H
10 mg UDP-4-amino
L-Glu Q-KG
100%
66%
New Method:
UDP-GIcNAc
UDP4-amino
UDP4-keto
OH
HO
0
-o PgIC(Ng)
PgIF130
OACHN NAD* NADH
0-UDP
AcHN
O-UDP
PMP
PLP
H2N
AcHN
O-UDP
50 mg UDP-4-amino
No purification necessary
L-Glu a-KG
H20
100%
100%
Figure 6-7. Comparison of the two methods utilized to prepare mg quantities of the UDP-4amino sugar.
207
Broad HTS Screening Campaign
As an alternative approach to fragment-based drug discovery, a collaboration with the
Broad Institute Screening Facility was initiated to develop a discontinuous, high-throughput
screening assay for PglD and WeeI from A. baumannii (selectivity screen).
To test the
transferability of the 96-well DTNB assay, PglD and Weel were titrated in a 384-well plate and
tested for activity over time. Both assays exhibited robust activity using nanomolar enzyme
concentrations over a 30 minute time period (Figure 6-8). Known acetyltransferase inhibitors
from the PglD fragment-based inhibitor project were also examined to test the viability of this
high-throughput assay. In all cases, the ICso values of these small molecules corresponded to
previously measured values (Figure 6-9). As a proof of concept, both enzymes in the 384-well
format were screened against a Broad maximum diversity library consisting of 2400 small
molecule compounds in duplicate.
0.9
0.5
0.458
0,45
0.7
0,40,6
0 20
03
0.3
[PID
A4
2030400
0.25
5
V
5
20
30
25
X10
0
0
t
20.2
.........
20
30
40
0
*Won)
5
10
is
20
2$
3
35
Wt*n1lMn)
Figure 6-8. (A) PglD enzyme titration in a 384-well discontinuous DTNB assay format. (B)
WeeI enzyme titration in a 384-well discontinuous DTNB assay format.
208
PgID
Weel
1000
N
o
200
460
100
lowo
10
100
/\NH2
96-well format
LK'o
NP2
IC5 0 = 400 pM
N
COOH
NN
Oj
384-well format
lm_a39
/
OH
jma65
IC 5 0 = 294 pM
96-well fonnat
C 5 0 = 40 gM
384-well fornat
IC50 = 92 ptM
Figure 6-9. IC 50 comparison of known acetyltransferase inhibitors in 96- and 384-well format.
Inhibition curves were generated in the 384-well format.
The purpose of the maximum diversity screen was to determine if this assay was a viable
format for the larger 359,000 compound set. Signal-to-background and the rate of false positives
between two separate screens of the same compounds are measurements utilized to assess the
quality of data. Another measurement known as a Z-factor (Z-factor = 1 - 3(cp + an)/p
-
ptl ),
is a statistical value representing the mean signal (pt) of positive (p) and negative (n) controls in
relation to the standard deviation (a) of these controls. A high-throughput screen with a Z-factor
that approaches 1 is ideal and is less likely to include a large amount of false positive and
negative results. An assay with a Z-factor > 0.5 is considered high quality and is robust enough
for high-throughput screening. Screening results from the maximum diversity screening set for
PglD and Weel confirmed that these assays were a viable screening format in terms of signal-tobackground (-8), Z factor (0.65), and a low false positive rate (Figure 6-10). Interestingly, the
209
maximum diversity set was able to uncover one small molecule inhibitor from the WeeL screen
(see below for results).
*/
POC run2.
POC-rtun2
Figure 6-10. PglD and WeeI 384-well format screening results from the maximum diversity set.
(A) PgID validation screen comparison between duplicate screens (runI vs. run.2). (B) WeeI
Each dot represents one well
validation screen comparison between duplicate screens.
containing a compound (blue), DMSO (grey), or no enzyme (red). POC is "percent of control"
with 0 representing no inhibition and -100 full inhibition. One apparent WeeI hit (6010833) was
discovered from this screen. Final compound concentration is 3 ptM.
Following validation of PgID in a high-throughput format, this enzyme was screened in
duplicate at 10 ptM against the Diversity-Orientated Synthesis (DOS) compound collection
containing a set of 83,000 molecules that are underrepresented in pharmaceutical libraries
(Figure 6-11).
There were 16 active compounds that exhibited > 15% inhibition (0.02% hit
rate). However upon the retesting of these compounds, they were all determined to be false
positives.
The NIH's Molecular Libraries Probe Production Centers Network (MLPCN)
collection consisting of 276,000 small molecules was screened in singlicate at 10 ptM against
210
PglD activity (Figure 6-12).
From this screen, 325 small molecules were identified that
inhibited PglD activity > 20% (0.12% hit rate).
Since there was such a low hit rate, the
definition of what qualifies as an "active" compound was determined at > 20% inhibition.
Follow-up confirmation with these compounds resulted in a total of five compounds that
repeated as inhibiting PglD activity (Figure 6-13). By defining a hit at > 20% inhibition, many
of these compounds were within the noise (error) of the assay, therefore a high false positive rate
was to be expected.
a
4U
POC-nni2
Figure 6-11. High-throughput screen consisting of the Broad DOS compound collection
completed in duplicate (runl vs. run2) at a concentration of 10 pM. Each dot represents an
individual well containing compound (red), DMSO (green), or no enzyme (blue). Full inhibition
of PgID is represented by -100 POC.
211
*
100
*. $* .*.
% di
S
S
f
~
*~
a Ccr
so
0
0
40
'100
W
60000
100000
W I
2C
160000
Compound ID
Figure 6-12. High-throughput screen consisting of the Broad MLPCN compound collection
completed in singlicate at a concentration of 10 pM. Each well is represented by a dot that
contains compound (red), DMSO (green), or no enzyme (blue). Full inhibition of PglD
corresponds to -100 POC. A compound was classified as a hit with a POC less than -20.
0
HO
S
NH
N
0
H
OH
S,
S
OH
N_
OH
BRD-K89423819
IC$O = 2 pM
0
OH
BRD-K99639744
IC50a -17 A M
BRD-A95794315
IC 50 = 13 pM
OH
01
N
H2N
N
0
HN
H
N
0
s
NIr~
0
S
\
0
N
/
0
0
HO
HO
0
BRD-K34145310 (Cefixime)
ICO> 50 PM
Br
BRD-K87816786
IC5o > 50 pM
Figure 6-13. The five compounds from the MLPCN screen that exhibited inhibition of PglD
activity. Although inhibition was apparent for BRD-K34145310 and BRD-K87816786, a
reliable IC 5o could not be established due to weak inhibition at high concentrations of the
compound.
212
Analogs of BRD-K89423819 and BRD-A95794315 were selected from the compound
library to determine potency and develop early structure-activity relationship (SAR) between the
two classes of inhibitors. Surprisingly, these analogs exhibited modest potency towards PglD
activity and can be classified as false negatives from the original screen (Figure 6-14). The
indolinone series of compounds (BRD-A95794315) exhibited atypical behavior with high Hill
coefficients (> 2) and partial inhibition at high concentrations (< 75%). Chelation of MgCl 2 in
the assay with ethylenediaminetetraacetic acid (EDTA) abrogated all inhibition with this class of
compounds. Varying concentrations of EDTA were tested with respect to the indolinone BRDA67338652 inhibition. Surprisingly, even small amounts of EDTA resulted in complete loss of
potency of this compound (Figure 6-15), which was confirmed by capillary electrophoresis.
Further development of this compound series was abandoned due to their inability to fully inhibit
PglD and the requirement of MgCl 2 for binding.
O
/r
N
3S
c
OH
OH
OH
OH
CH
N
C1
N
OH
BRD-K38148053
ICs = 122 pM
0
0
N-/O
OH
AMS0001731
ICOa 860 pM
S
NH
ORD-K61068896
IC5 = 8 pM
0
NH
OH
OH
BRD-K22519821
IC50 = 38 pM
SRD-A67338652
IC50 = 17 M
BRD-A27978556
IC5 = 20 IN
Figure 6-14. Analogs of the thienopyrimidine and indolinone series of compounds from the
Broad MLPCN screening library.
213
0.1 mM EDTA
0 iML EDTA
1 mM EDTA
1*
14
14
12
1
B
m
7ss
1.2
00
'14
0.2
k
0.2
0
0
0
I
0.2
A ,A&"
0
A
100
00110
=C36-5gM
JC5 0 17g
1100
I
> 600M
Figure 6-15. IC50 values for the indolinone compound BRD-A67338652 across varying EDTA
concentrations. The addition of any concentration of EDTA led to complete loss in compound
potency.
With the loss of the indolinone scaffold as a viable inhibitor class, focus was directed
towards the thienopyrimidine series. The lead compound from this scaffold (BRD-K89423819)
was investigated for its mechanism of action by increasing the concentration (5 x Km) of each
substrate separately and determining an IC 50. The increase in concentration of UDP-4-amino had
no effect on potency of this compound. However, a 5-fold increase in AcCoA resulted in a 4fold decrease in IC 50 (Figure 6-16). Further validation that this compound class binds in the
AcCoA binding pocket was established by a graduate student in the Imperiali lab, Austin Travis.
He was able to solve a structure with a close analog of BRD-K89423819 (MM-1, see below)
with PglD (Figure 6-17).
Therefore, this compound series can be classified as AcCoA
competitive.
214
100
I.M
AcCoA (5 x K )
1.5 mM UDP-4-amino (5 x K.)
100
so
60
IS
00
HOOH
40
OH
20
BRD-K89423819
I
60
40
2D
ICw= 2p~M
0
IMD= 381 pM)
0
100
0.1
1
10
100
IC 50= 8 g.M
IC50 = 2 pM
Figure 6-16. Kinetic determination of the binding mechanism for the compound BRDK89423819. Increasing the concentration of AcCoA had a direct effect on inhibitor binding
making this class of compounds competitive with respect to this substrate.
(A)
(B)
Figure 6-17. (A) PglD-bound structure with the thienopyrimidine compound MM-1 in the
AcCoA binding pocket. (B) Overlay of MM-I (grey) and AcCoA (pink) PglD structures (PDB
code 3BSY). Water molecules are shown as red spheres. The PglD-MM-1 structure is courtesy
of Austin Travis.
Synthesis of ThienopyrimidineAnalogs
To investigate the thienopyrimidine series of compounds, a series of analogs exploring
the left-hand aryl moiety were prepared. Established literature procedures were developed that
215
allowed for the synthesis of these molecules (19). The first reaction utilizing the alkyl halide
starting material (I) with triethyl phosphite yielded the phosphonate ethyl ester product (II)
(Scheme 6-1).
The second reaction utilizing a Homer-Wadsworth-Emmons reaction was
employed under basic conditions at room temperature with a specific aldehyde to form the
desired ethyl ester product (III) (Scheme 6-2). A final hydrolysis step was carried out to yield
the final free acid product (IV) (Scheme 6-3).
'H-NMR and LC/MS confirmed the desired
product (Figure 6-18, 6-19). With a reliable synthetic procedure, a series of analogs were then
synthesized investigating various left-hand moieties to develop SAR and increase potency of this
compound class (Figure 6-20).
All synthesized molecules were tested for inhibition in the
DTNB PglD assay at the Km of their substrates. These experiments culminated in the discovery
of a more potent compound (MM-7), through replacement of the phenol moiety with an indole
substituent. Inhibition of PglD by these compounds was corroborated by CE, validating these
molecules as true inhibitors of acetyltransferase activity.
Scheme 6-1. Synthetic route of the phosphonate ethyl ester.
N
C, _
OEt
EtO IOEt
150 0 C
0
S
N
OH
EtO,
4h
0
S
N
OEt N
o
O
OH
OH
1
11
70% Yield
Scheme 6-2. Synthetic route of the thienopyrimidine ethyl ester product.
O
EIO..",,
0
N
6E
jO/
OH
N
H
Et
+
S
LiOI
RT, 17 h
H
H.
s
N
OH
OH
11
I+
70% Yield
95% Yield
216
0
0
OEt
Scheme 6-3. Synthetic route for the final thienopyrimidine product.
HO~
N
0
S
HO
HO
Na0H
Et
IH
MeOH/H 20
RT, 4 h
OH
OH
OH
IV
111
80% Yield
(A)
(B)
NN
N
S
)Hr
/
NOH
OH
OH
I
~1
a
idi
~1
...
. ........
(C)
...........
(D)
N
S
0
HOOC
N
NS
-
0
OH
OH
OH
OH
___________________________________________
111
a
..........
Figure 6-18.
Final
1
H-NMR for thienopyrimidine derivatives.
t
rM
:
.....
Overall yields for each
compound are denoted in parentheses. (A) BRD-K89423819 (80%) (B) MM-1 (64%)
(C)
MM-2 (67%) (D) MM-3 (72%).
217
(A)
K:
(B)
N
N-
/
S0
N- /
NOH
OH
.r.
OH
OH
i,~.
.11.
r
ITI
(C)
(D)
H
N
N.
S.
No
\
~
N
OH
S
0
OH
OH
OH
I
Wb
I~iT
Figure 6-19. Final 1H-NMR for thienopyrimidine derivatives. Overall yields for each
compound are denoted in parentheses. (A) MM-4 (52%) (B) MM-5 (49%) (C) MM-6 (62%)
(D) MM-7 (57%).
218
N,H
N
N
OH
N~
0H
HOOC
H
OH
/
N;
OH
OH
OH
OH
MM-1
IC5 =6.8 pM
MM-2
ICO= 130 pM
MM-3
IC5= 5.7pM
N_OH
OH
MM-4
ICSO= 22
pM
H
,-
N
S
0
Figure 6-20.
\N
S
N
H
H
/
NNS
0
OH
N
OH
OH
OH
MM-5
ICSO = 6.0 pM
MM-6
IC50 = 22 pM
IC5 0 = 1.1 pM
MM-7
Analogs of the original thienopyrimidine compound from the Broad MLPCN
screen with PglD IC 5 0 values.
Selectivity Screening with Homologous Acetyltransferases
To determine the selectivity of the thienopyrimidine class of compounds, assays were
established for the A. baumannii and N. gonorrhoeae acetyltransferases Weel and PglB-ATD,
respectively. These two enzymes represent homologous enzymes that are responsible for the
acetylation of UDP-4-amino in an O-linked protein glycosylation pathway. Previous structural
characterization of the N-linked PglD and O-linked PglB-ATD AcCoA binding pockets revealed
a high level of dichotomy between these sites (Chapter 4).
A primary sequence alignment
between the three acetyltransferases (PglD, Weel, and PglB-ATD) confirmed that the binding
pocket residues for PglD are divergent with respect to Weel and PglB-ATD. A continuous, in
vitro assay relying on the generation of the TNB 2- chromophore from Ellman's reagent was
employed to measure acetyltransferase activity.
Concentrations were held at Km for all
substrates excluding UDP-4-amino with Weel; due to the high Km for this substrate, the
concentration was decreased 3-fold. Since this compound class is competitive with respect to
219
AcCoA, a direct comparison of the IC 50 values between enzymes can still be employed.
Whereas the N- and O-linked acetyltransferases catalyze the identical reaction with the same
substrates at comparable Km values, the binding pockets exhibit a great deal of divergence. This
observation is further validated by the IC 50 values with thienopyrimidine compound series
(Figure 6-21). Certain compounds are highly selective for the N-linked acetyltransferase PglD
(BRD-K89423819, MM-1, MM-2, and MM-3), which each contain a mono-aryl substituent.
However when a biaryl ring system is introduced at this position (MM-5, MM-6, MM-7), these
compounds become much more promiscuous. Not surprisingly, IC 50 values between the two 0linked acetyltransferases Weel and PglB-ATD are equivalent. This is most likely due to the
higher homology of their respective binding pockets in relation to the N-linked PglD
acetyltransferase.
In collaboration with the Broad Institute, further thienopyrimidine analogs
were synthesized to increase inhibitor potency (Figure 6-22). All inhibitors were synthesized by
Dr. Jackie Wurst and Dr. Jun Pu (Broad) except the 3-pyridine molecule (AT-1, Austin Travis).
These compounds were cross-screened against Weel and PglB-ATD for selectivity purposes.
PglD IC 50 data was provided by Austin Travis and Dr. James Spoonamore (Broad).
Interestingly, the amino-pyridine compound (BRD-K52239388) exhibited selectivity towards the
O-linked acetyltransferases, the first molecule to exhibit this type of behavior. However, these
compounds exhibited no improvement over the previous lead compound MM-7. Experiments
are ongoing to synthesize additional analogs based upon these results and to co-crystallize these
inhibitors in the presence of PglD, Weel, and PglB-ATD.
220
N
S
N
0
OH
OH
0
/
N
N,r--
N
pM
IC5
=
/
S
=
6.8
M,>200 pM, >200 pM
N
0
HOOC
N
N
OH
OH
MM-2
ICSO
ICS = 130 AM, 530 pM, 830 pM
0
NI
ICS = 22 pM, 80 pM, 87 pM
s
OH
OH
MM-5
= 6.0 pM, 27 AM, 40 pM
IC5
NT
OH
OH
MM.4
pM
H
N,S
/
S
MM-3
5.7 pM, >600 pM, >600
6H0
N-
pM
OH
N
SN
OH
BRD-K61068896
OH
MM-1
IC5
S/
IC 50 = 7.8 AM, 51 pM, 37
>600 pM, 660 pM
N
OH
- .N
pM,
N-
OH
BRD-K89423819
ICS = 2.0 pM, >200 pM, 370 M
NN
OH
C
BRD-K22519821
ICS .38
N
N
0
S
C11
OH
OH
BRD-K38148053
120 pM, 700 pM, 860 pM
OH
N
N
OH
OH
AMS0001731
ICS =860 AM, 3300 pM, 2600
HO
S
N,
MM-C
ICS = 22 pM, 90 pM, 110 pM
IC50
MM-7
= 1.1 pM, 14 pM, 25 pM
IC 50 = PgID, Weet, PgIB-ATD
Figure 6-21. IC 50 selectivity results from the 0-linked acetyltransferases Weel and PglB-ATD.
N
N
N
S
N,-/
0
NN
OH
OH
50 = 24
NN
Y N
S
0
Nf
/
OH
-H
OH
M, 57 pM, 340 pM
ICS
H
= 20 MM, 58 MM, 87 M
ICS0
OH
OH
OH
BRD-K58489728
AT-1
IC
H
N
BRD-K61442517
BRD-K75650841
4.3MM, 110 MM, 70 pM
IC 50 = 4.7 pM, 36 gM, 57 pM
HO
0 2N
N
0H
H
OH
OH
BRD-K61510404
MM, 120 pM, 120 pM
IC50= >100
r' a
;i Y N S 0
/
OH
NH 2NN
N,--
S/
0
N
OHO
OH
BRD-K52239388
IC 50 = >100 MM, 22 pM, 51 pM
BRD-K67085234
IC 50 = >100 MM, 230 MM, 150 pM
i
O
OH
BRD-K76433418
IC
50
H
>100 AM, >500 M, >500 M
IC 50 = PgID, Weel, PgIB-ATD
Figure 6-22. Additional analogs of the thienopyrimidine series synthesized by the Broad
Institute (BRD) and Austin Travis (AT). PglD IC 50 data was provided by Dr. James Spoonamore
and Austin Travis.
221
Discovery of the WeeI Inhibitor 6010833
As previously mentioned, the Broad maximum diversity screen with Weel identified a
single hit that was identified as 6010833, containing an isoxazole core that was commercially
available (Figure 6-23).
Two analogs of this compound, 5908682 and 5904993, were also
available commercially.
These small molecules displayed potency in the micromolar range
against Weel in the DTNB activity assay. Capillary electrophoresis follow-up with the most
potent lead (5906862) was established to ensure that this compound did not interfere with the
assay detection reagents. Following CoASH turnover with varying concentrations of 5906862
resulted in an IC 50 of 7 piM, which agreed with previous inhibition values from the DTNB assay.
CI
IN
IN
N N.
0
5906862-F
IC50 = 30 p1
5904993
=85 pM
5906862
IC5=14sM
I1C.
6010833
IC.5= 30 pM
F
N
N
N
6N
0 2N
5906862-NO 2
IC 50=10 pM
0
N.N
&
5906862-OMe
IC50 = 40 MN1
Figure 6-23. The isoxazole class of Weel inhibitors discovered from the Broad maximum
diversity screening library.
Analogs of these isoxazole inhibitors were synthesized to better understand the structure
activity relationship (SAR) of this molecule as it relates to potency. To this effect, a one-step
222
synthetic route utilizing cinnamaldehyde derivatives was undertaken to achieve the final
compounds (Scheme 6-4).
Efforts to synthesize 4-methoxy, 4-fluoro, and 4-nitro products
through published procedures proved successful (20) (Figure 6-24).
Scheme 6-4. Synthetic route to obtain isoxazole analogs.
N-O
+
4
0
toluene
H
acetic acid
N
-
0reflux
RR
47 - 81% Yield
(A)
(B)
N
N
'0
N
(/
/
0
of
/
/0
N
Figure 6-24. Final 'H-NMR for 5906862 derivatives.
Overall yields for each compound are
denoted in parentheses. (A) 4-fluoro (47%) (B) 4-nitro (65%) (C) 4-methoxy (81 %).
223
These analogs were then subjected to the DTNB continuous Weel activity assay. Initial
velocities of enzyme activity were plotted against inhibitor concentrations to establish an IC 50 for
each compound (Figure 6-23). Substituents added to the 4 position of the phenyl ring either had
a deleterious effect (4-F and 4-OMe) or had no effect (4-NO 2) on potency. Further work with
CE confirmed the DTNB IC 50 results for 5906862-NO 2 (IC 50
=
9 pM). These compounds were
also found to bind to the AcCoA binding pocket by measuring the potency of 5906862 at varying
concentrations of substrates (UDP-4-amino and AcCoA). Further exploration of this scaffold
was put on hold following the discovery of the thienopyrimidine
series as potent
acetyltransferase inhibitors.
Conclusions
To conclude, a high-throughput screening effort with the Broad Institute was successful
in identifying a low micromolar lead compound for inhibiting PglD acetyltransferase activity.
This thienopyrimidine series exhibited tractable inhibition by binding to the AcCoA binding
pocket as determined through structural and biochemical experiments.
Furthermore, a facile
route to synthesize additional analogs of the original lead molecule proved successful, with the
generation of 15 new inhibitors. Interestingly, biaryl ring substituents in this scaffold exhibit
activity towards the O-linked acetyltransferases Weel and PglB-ATD.
However, selectivity
towards the intended target PglD can be achieved through a mono-aryl substituent at this
position. These results confirm the original finding in Chapter 4 that the AcCoA binding pockets
in N- and O-linked acetyltransferases are divergent based upon AcCoA-bound PglD and PglBATD crystal structures.
This outcome is surprising given that these enzymes catalyze the
224
identical reaction and have similar binding coefficients. Further work will focus on synthesizing
more potent analogs of the lead indole compound through the available SAR. These molecules
are currently being utilized by Austin Travis in the Imperiali lab to link with fragment
compounds that bind in the UDP-4-amino binding pocket.
These fragment molecules were
previously identified through a fragment-based approach to inhibit PglD by Dr. James Morrison
(Imperiali lab). Lastly, these compounds will be used as molecular probes in an in vivo setting
to understand the relationship between protein glycosylation and pathogenicity in C. jejuni.
Acknowledgments
I am grateful to Dr. James Spoonamore for the Broad primary screening data with PglD and
providing IC 50 data. I am also grateful to Austin Travis for PglD IC 50 data, the structure of PglD
bound to MM-1, advice on chemical synthesis, and critical reading of this chapter. I would like
to thank Michelle Chang for critical reading of this chapter. I would also like to thank Dr. Jackie
Wurst and Dr. Jun Pu for providing the thienopyrimidine analogs. Lastly, I would like to thank
Dr. James Morrison for providing the initial PglD fragment inhibitors for control compounds
during develop of the HTS assay.
Experimental Procedures
Common Materials
All chemicals were purchased from Sigma-Aldrich unless otherwise stated.
225
PglD Expression
The pET24a(+) plasmid containing the pglD gene was used to transform Escherichiacoli
BL21(DE3) RIL competent cells (Stratagene). One liter of LB media containing 50 pg/mL
kanamycin and 30 ptg/mL chloramphenicol was inoculated with 8 mL of an overnight culture of
cells. The cells were then allowed to grow at 37 0C while shaking until an optical density of -0.8
(X = 600 nm) was reached. The culture was cooled to 16 C and induced with 0.5 mM iso-p-Dthiogalactosylpyranoside (IPTG). After incubating for 18 h with shaking at 16 C, the cells were
harvested by centrifugation (2600g, 30 min) and stored at -80 'C until needed.
PgJD Purification
Each protein purification step was carried out at 4 *C. The cell pellet (-3 g) was
resuspended in 40 mL of 50 mM HEPES pH 7.4/100 mM NaCl/30 mM imidazole (Buffer A)
and then lysed by sonication. The lysate was then cleared by centrifugation (145,000g, 60 min)
and added to 2 mL of Ni-NTA resin (Qiagen). The slurry was allowed to tumble for 3 h and then
packed into a fritted PolyPrep column (Biorad). The resin was washed with 20 column volumes
of Buffer A and then eluted with a buffer containing 50 mM HEPES pH 7.4/100 mM NaCl/ 300
mM imidazole. Fractions containing the purified protein as analyzed by SDS-PAGE were
pooled and dialyzed against 50 mM HEPES pH 7.4/100 mM NaCl for 24 h. Purified PglD
protein was then concentrated to ~10 mg/mL using a 10 kDa MWCO Amicon Ultra-15
centrifugal filter unit (Millipore).
Protein concentration were calculated based upon the
predicted extinction coefficients at X = 280 nm.
226
PglD AcetyltransferaseActivity Assay
Kinetic characterization of PglD was carried out using a previously modified procedure
(21). CoASH generation resulting from the acetyltransferase reaction carried out by PglD was
monitored in the presence of Ellman's reagent (DTNB) through the generation of the TNB2chromophore in a continuous fashion. To a flat, black-clear bottom 96-well plate (Falcon) was
added 50 mM HEPES pH 7.4, 2 mM MgCl 2 , 0.05% BSA, 0.001% Triton X-100, 1 mM DTNB,
and 0.2 nM PglD. Reactions were completed in duplicate and initial rates were measured in the
linear portion of the reaction curve over a 5 minute time period at 25 0C.
The substrate
concentrations of AcCoA and UDP-4-amino were varied separately to determine kinetic
parameters using initial velocity measurements while holding the other substrate at a saturating
level as calculated with equation 1 using the program GraFit 6.0.12 (Erithacus Software).
Further experiments measuring inhibitor potency were carried out at the Km for each substrate
(UDP-4-amino = 274 pM, AcCoA = 295 pM). A Weel activity assay (at 1 nM enzyme) was
carried out in a similar fashion with the following substrate concentrations (80 pM AcCoA, 800
pM UDP-4-amino). PglB-ATD had the following assay concentrations (1 nM PglB-ATD, 100
pM AcCoA, 286 pM UDP-4-amino). IC50 values were calculated from equation 2 using the
program GraFit 6.0.12 (Erithacus Software) and measured in duplicate.
v = Vmax[S]/(Km + [S])
y
=
100%/(1 + (x/IC50 )s)
(1)
(2)
227
UDP-4-amino Large Scale Biosynthesis
During purification, PglF130 was supplemented with 500 pM NAD+. Approximately 30
mg (381 nmol) of the GST-PglF130 construct (from the pGEX plasmid) was bound to 4 mL of
glutathione sepharose 4 fast flow resin (GE Life Sciences) for 4 hours at 4 0C and then washed
with 30 mL of lx PBS. 25 mM HEPES pH7.4 was added to the slurry such that the total volume
measured 5 mL; 100 mg (165 pmol) of UDP-GlcNAc was added and the reaction mixture
incubated at room temperature with rocking overnight for 16 hours. A small sample of the
reaction mixture was analyzed by CE in order to ensure that the reaction had gone to completion.
The reaction mixture was resuspended in a slurry and poured into a fritted PolyPrep column
(BioRad) and washed with 5 mL H2 0 to remove any UDP-4-keto bound to enzyme/resin. The
flow-through was collected and used in the subsequent reaction. For the second half of the
reaction, the aminotransferase from N. gonorrhoeae(PglC) was supplemented with 500 pM PLP
during purification. Approximately 60 mg (1360 nmol) of the Hiss-TEV-PglC construct (from
the modified pET30b(+) plasmid) was bound to 4 mL of Ni-NTA resin (Qiagen) and washed
with 50 mL of resuspension buffer (50 mM HEPES pH 7.4, 30 mM imidazole, 100 mM NaCl).
10 mL of UDP-4-keto (50 mg) from the previous reaction was supplemented with 20 mM Lglutamate and added to the PglC bound to Ni-NTA resin. This slurry was then allowed to
incubate overnight at room temperature with rocking.
Importantly, this aminotransferase
reaction has an upper limit of 50 mg of UDP-4-keto for 100% turnover. A small sample of the
reaction mixture was analyzed by CE in order to ensure that the reaction had gone to completion.
The reaction mixture was resuspended in a slurry and poured into a fritted PolyPrep column
(BioRad) and washed with 10 mL H2 0 to remove any UDP-4-amino bound to enzyme/resin.
The flow-through was collected and lyophilized to obtain the UDP-4-amino sugar. UDP-4-
228
amino was re-dissolved in water and its concentration was calculated based upon the extinction
coefficient of uracil (& 260 = 9900 M1 cm-1). The final yield of purified UDP-4-amino was 48 mg
(from 50 mg UDP-GlcNAc starting material).
High-throughputscreening- BroadInstitute
To a Nunc 384-well black-clear bottom plate (Thermo Scientific) was added 200 nL of
compound in neat DMSO (final assay concentration = 10 pM) and 20 piL of enzyme solution
(final assay concentrations: 6 nM PglD, 50 mM HEPES pH 7.4, 0.001% Triton, 0.05% BSA,
100 tM Ac-CoA). Solution dispensing was accomplished utilizing a Multidrop Combi (Thermo
Scientific).
minutes.
This mixture was allowed to incubate at room temperature for a duration of 45
The reaction was started by the addition of 10 piL reaction solution (final assay
concentrations:
50 mM HEPES pH7
A 1)A
TMI
A
--
)D The reaction was quenched
after 30 minutes with the addition of 30 tL stop solution (final assay concentrations:
DTNB, 1 mM EDTA, 20% 1-propanol).
2 mM
The plates were allowed to develop for 5 minutes
before reading at a X = 405 nm. The Weel screen was carried out in a similar fashion with the
following changes in concentration: 15 nM Weel, 200 ptM UDP-4-amino, 400 pM AcCoA.
Synthesis of Thienopyrimidine Compounds
To 5 mL (30 mmol) triethylphosphite was added 500 mg (2 mmol) of the alkyl halide (I,
Scheme 6-1). The material was refluxed for 4 h at 150 0C. The solution was cooled to room
temperature, the resulting white solid filtered, and washed with hexanes.
Silica column
chromatography (96% dichloromethane, 4% methanol) afforded the phosphonate product (II,
Scheme 6-1) in 70% yield. A solution of 0.13 mmol (50 mg) phosphonate (II), 0.16 mmol
229
aldehyde (1.2 equiv), and 0.65 mmol (5 equiv) of LiOH-H20 in 2 mL THF and 1 mL EtOH was
stirred under N2 for 17 h at room temperature. Following this reaction, the solvent was removed
under reduced pressure. The resulting ethyl ester product (III, Scheme 6-2) was dissolved in 2
mL MeOH and 2 mL H20. To this solution was added 0.5 mmol (2 equiv) NaOH and the
mixture stirred at room temperature for 4 h. The solvents were removed under reduced pressure
and the resulting solid was dissolved in 4 mL H2 0. This aqueous layer was acidified using 6 M
HCl and a yellow solid precipitated out of solution. This solid was filtered , washed with ether,
and lyophilized overnight to yield the desired product (IV, Scheme 6-3).
HO
H
OH
BRD-K89423819
BRD-K89423819. 'H NMR (300 MHz, DMSO-d 6 ) 6 12.6 (s, 1H), 7.90 (d, lH), 7.25 (d, 1H),
7.09-6.87 (m, 4H), 2.81 (s, 3H). MS calcd for C16H12N20 4S [M]+
=
328.3, found [M + H]+
=
329.3.
NNS
N
0
OH
OH
MM-I
MM-1. 'H NMR (300 MHz, DMSO-d6 ) 6 12.6 (s, 1H), 7.94 (d, 1H), 7.65 (dd, 2H), 7.44-7.46 (m,
3H), 7.05 (d, lH), 2.81 (s, 3H). MS calcd for C16H12N20 3S [M]+
313.7.
230
=
312.3, found [M + H]=
N
S
0
OH
H
OH
MM-2
MM-2. 'H NMR (300 MHz, DMSO-d 6 ) 6 12.3 (s, 1H), 6.71-6.62 (m, 1H), 6.35 (d, 1H), 2.78 (s,
3H), 1.69 (m, 1H), 0.96-0.70 (m, 4H).
HO
HOOCO
I
N.NS
-- ''
0
HjIN
OH
MM-3
MM-3. 'H NMR (300 MHz, DMSO-d 6 ) 6 12.6 (s, 1H), 8.21 (s, 1H), 8.06-7.89 (m, 3H), 7.61 (t,
1H), 7.11 (d, 1H), 2.79 (s, 3H). MS calcd for C17H12N2 0 5 S [M]+
=
356.1, found [M + H]+
=
357.2.
N-
OH
OH
MM-4
MM-4. 'H NMR (300 MHz, DMSO-d 6 ) 6 12.6 (s, 1H), 7.79-7.74 (m, 1H), 7.63-7.61 (m, 1H),
7.40-7.33 (m, 4H), 7.21-7.14 (m, 2H), 6.54 (d, 1H), 2.79 (s, 3H).
231
NNS
0
N
OH
OH
MM-5
MM-5. 'H NMR (300 MHz, DMSO-d 6) 6 12.6 (s, 1H), 8.18-8.11 (m, 2H), 8.00-7.95 (i, 3H),
7.81 (dd, 1H), 7.59-7.56 (m, 2H), 7.17 (d, 1H), 2.82 (s, 3H).
NN
OH
OH
MM-6
MM-6. 'H NMR (300 MHz, DMSO-d6 ) 6 12.97 (s, 1H), 9.19 (d, 1H), 8.77-8.64 (m, 2H), 8.297.89 (m, 4H), 7.51 (d, 1H), 2.84 (s, 3H).
/
OH
OH
MM-7
MM-7. lH NMR (300 MHz, DMSO-d 6 ) 6 12.51 (s, 1H), 11.41, (s, 1H), 8.11 (d, 1H), 7.86 (s,
1H), 7.49-7.40 (m, 3H), 6.94 (d, 1H), 6.52 (s, 1H), 2.80 (s, 3H).
232
Synthesis of 5906862 Analogs
N-O
Htoluene0
/
acetic acid
N
b
reflux
F
47% Yield
4-Fluorocinnamylidene-isoxazolin-5-one.To a solution of 5-isoxazolinone (10 mmol) in 27 mL
warm (~40 'C) toluene was added 4-fluorocinnamaldehyde (10 mmol). Glacial acetic acid (0.54
mL) was added and the resulting solution was refluxed for 1 hour. Hot hexane (~60 OC, 8.6 mL)
was added to the refluxed solution. After cooling the solution to 4 -C, crystals formed and were
filtered off. Recrystallization from toluene:hexane (2:1) produced yellow crystals.
=
Final yield
47%. 'H-NMR (300 MHz, CDCl 3): 6 8.35 (dd, 1H), 7.40-7.65 (in, 8H), 7.32-7.23 (in, 1H),
7.2 I-7.13 (In, 21). All other 5906862 analogs were synthesized in the same fashion.
4-Nitrocinnamylidene-isoxazolin-5-one. 'H-NMR (300 MHz, CDCl 3): 8 8.35 (dd, 1H), 7.407.65 (in, 8H), 7.30-7.24 (in, lH), 7.20-7.11 (in, 2H).
4-Methoxycinnamylidene-isoxazolin-5-one. IH-NMR (300 MHz, CDCl 3): 6 8.35 (dd, lH), 7.407.65 (in, 8H), 7.28 (t, 1H), 6.89 (d, 2H), 3.85 (s, 3H).
References
1.
Barber, M. (1947) Staphylococci infection due to penicillin-resistant strains. Br. Med. J. 2,
863-865.
2. Schoenhofen, I.C., McNally, D.J., Vinogradov, E., Whitfield, D., Young, N.M., Dick, S.,
Wakarchuk, W.W., Brisson, J.R., and Logan, S.M. (2005) Functional characterization of
dehydratase/aminotransferase pairs from Helicobacter and Campylobacter. J. Biol. Chem.
281, 723-732.
3. Vijaykumar, S., Merkx-Jacques, A., Ratnayake, D.B., Gryski, I., Obhi, R.K., Houle, S.,
Dozois, C.M., and Creuzenet, C. (2006) Cj 1121c, a novel UDP-4-keto-6-deoxy-GlcNAc C-4
233
aminotransferase essential for protein glycosylation and virulence in Campylobacterjejuni. J.
Biol. Chem. 281, 27733-27743.
4. Olivier, N.B., Chen, M.M., Behr, J.R., and Imperiali, B. (2006) In vitro biosynthesis of UDPN,N'-diacetylbacillosamine by enzymes of the Campylobacter jejuni general protein
glycosylation system. Biochemistry 45, 13659-13669.
5. Glover, K. J., Weerapana, E., Chen, M. M. and Imperiali, B. (2006) Direct biochemical
evidence for the utilization of UDP-bacillosamine by PglC, an essential glycosyl- I-phosphate
transferase in the Campylobacter jejuni N-linked glycosylation pathway. Biochemistry 45,
5343-5350.
6. Weerapana, E., Glover, K. J., Chen, M. M. and Imperiali, B. (2005) Investigating bacterial
N-linked glycosylation: synthesis and glycosyl acceptor activity of the undecaprenyl
pyrophosphate-linked bacillosamine. J. Am. Chem. Soc. 127, 13766-13767.
7. Troutman, J. M. and Imperiali, B. (2009) Campylobacter jejuni PglH is a single active site
processive polymerase that utilizes product inhibition to limit sequential glycosyl transfer
reactions. Biochemistry 48, 2807-2816.
8. Glover, K. J., Weerapana, E. and Imperiali, B. (2005) In vitro assembly of the
undecaprenylpyrophosphate-linked heptasaccharide for prokaryotic N-linked glycosylation.
Proc. Natl Acad. Sci. USA 102, 14255-14259.
9. Hughes, R. (2004) Campylobacterjejuni in Guillain-Barre syndrome. Lancet Neurol. 3, 644.
10. Komagamine, T., and Yuki, N. (2006) Ganglioside mimicry as a cause of Guillain-Barre
syndrome. CNS Neurol. Disord. Drug Targets 5, 391-400.
11. Yu, R. K., Usuki, S., and Ariga, T. (2006) Ganglioside molecular mimicry and its
pathological roles in Guillain-Barre syndrome and related diseases. Infect. Immun. 74,
6517-6527.
12. Kowarik, M., Young, N.M., Numao, S., Schulz, B.L., Hug, I., Callewaert, N., Mills, D.C.,
Watson, D.C., Hernandez, M., Kelly, J.F., Wacker, M., and Aebi, M. (2006) Definition of the
bacterial N-glycosylation site consensus sequence. EMBO J. 25, 1957-1966.
13. Scott, N. E., Parker, B. L., Connolly, A. M., Paulech, J., Edwards, A. V., Crossett, B.,
Falconer, L., Kolarich, D., Djordjevic, S. P., Hojrup, P., Packer, N. H., Larsen, M. R., and
Cordwell, S. J. (2011) Simultaneous glycanpeptide characterization using hydrophilic
interaction chromatography and parallel fragmentation by CID, higher energy collisional
dissociation, and electron transfer dissociation MS applied to the N-linked glycoproteome of
Campylobacterjejuni. Mol. Cell. Proteomics 10, M000031-MCP201.
14. Olivier, N. B., and Imperiali, B. (2008) Crystal structure and catalytic mechanism of PglD
from Campylobacterjejuni.J. Biol. Chem. 283, 27937-27946.
15. Rangarajan, E.S., Ruane, K.M., Sulea, T., Watson, D.C., Proteau, A., Leclerc, S., Cygler, M.,
Matte, A., and Young, N.M. (2007) Structure and active site residues of PglD, an Nacetyltransferase from the bacillosamine synthetic pathway required for N-glycan synthesis
in Campylobacterjejuni. Biochemistry 47, 1827-1836.
16. Clatworthy, A.E., Pierson, E., and Hung, D.T. (2007) Targeting virulence: a new paradigm
for antimicrobial therapy. Nat. Chem. Biol. 3, 541-548.
17. Kelly, J., Jarrell, H., Millar, L., Tessier, L., Fiori, L.M., Lau, P.C., Allan, B., and Szymanski,
C.M. (2006) Biosynthesis of the N-Linked glycan in Campylobacter jejuni and addition onto
protein through block transfer. J. Bacteriol. 188, 2427-2434.
234
18. Krogh, A., Larsson, B., von Heijne, G., and Sonnhammer, E. L. (2001) Predicting
transmembrane protein topology with a hidden Markov model: Application to complete
genomes. J.Mol. Biol. 305, 567-580.
19. Shishoo, C.J., Devani, M.B., Ullas, G.V., Ananthan, S., and Bhadti, V.S. (1985) Studies on
the
synthesis
of
2-(2-arylvinyl)thieno[2,3-d]pyrimidines
and
5-(2arylvinyl)triazolothieno[3,2-e]pyrimidines. J. Heterocyclic Chem. 22, 825-830.
20. Francis, C.V., Cross, E.M., Korkowski, P.F., Leung, P.C., Macomber, D.W., Tiers, G.V., and
Trend, J.E. Functionalized merocyanine dyes and polymers incorporating such dyes. Patent
WO 93/15596. August 5, 1993.
21. Hartley, M.D., Morrison, M.J., Aas, F.E., Borud, B., Koomey, M., and Imperiali, B. (2011)
Biochemical characterization of the O-linked glycosylation pathway in Neisseria
gonorrhoeae responsible for the biosynthesis of protein glycans containing N,N'diacetylbacillosamine. Biochemistry 50, 4936-4948.
235
Chapter 7: Biochemical Characterization and Fragment-Based Inhibition of
the NeisseriagonorrhoeaeAcetyltransferase PglB-ATD.
236
Introduction
The rise in antimicrobial resistance has been the result of misuse and selective pressure
from bactericidal antibiotics (1).
Classical targets such as protein synthesis, DNA replication,
and cell-wall synthesis have resulted in an effective form of treatment, however the penalty of
inhibiting bacterial growth is evident. The Center for Disease Control (CDC) has recently
reported that 2 million people in the United States contract an antibiotic-resistant bacterial
infection each year, which results in the death of at least 23,000 people.
The CDC has also
singled out the most urgent bacterial-resistant threats: Neisseria gonorrhoeae, carbapenemresistant Enterobacteriaceae(CRE), and Clostridium difficile. N. gonorrhoeae is the second
most commonly reported infection in the United States, and the treatment of this Gram-negative
bacterium has shifted to third-generation cephalosporins following the loss of efficacy of the
penicillins and quinolones (2-3). This bacterial infection can lead to pelvic inflammatory disease
resulting in infertility as well as increased vulnerability to HIV infection (4). Clearly, a novel
strategy for targeting bacterial infectivity while limiting resistance must be explored. Pathogenic
bacteria rely upon the ability to adhere, colonize, and invade the cell surface of a host to generate
a disease state.
The inhibition of these virulence factors would lead to quiescence, limiting
bacterial infection. Since the result of this inhibition has no effect on bacterial growth, selective
pressure would not be applied to the system and resistance averted.
N. gonorrhoeae relies upon Type IV pili (TFP) for adhesion, motility, colonization, and
transformation to generate a disease state (5). TFP is a membrane-anchored multimeric complex
composed of the pilin protein PilE that protrudes from the bacterial surface (6) (Figure 7-1).
Interestingly, PilE is post-translationally modified at a single serine residue (Ser63) with the
trisaccharide Gal-fP-1,4-Gal-a-1,3-NN'-diacetylbacillosamine (diNAcBac) (7-8). Considered to
237
be the most common type of post-translational modification, glycosylation plays an important
role in recognition and signaling events such as immune response and protein folding (9). The
integration of glycoproteins such as PilE onto the cell-surface of Gram-negative bacteria
contributes to its overall pathogenicity. Therefore, further exploration into the glycosylation
biosynthetic machinery of N. gonorrhoeae would provide a basis for understanding
pathogenicity and bacterial resistance.
PiHE pilin
N. gonorrhoeae
Type IV pilus
monomer
Figure 7-1. N. gonorrhoeaeType IV pili reconstructed from the cryo-EM of the filament (11.5
A) and crystal structure of the PilE monomer (2.3 A) (6).
Previous work has focused on the characterization of the N-linked glycosylation pathway
in the Gram-negative bacterium Campylobacter jejuni (10-12).
C. jejuni is responsible for
gastroenteritis in humans and owes its pathogenicity to the formation of N-linked glycans on
periplasmic and cell-surface proteins. Assembly of this glycan occurs in a stepwise fashion on
an
undecaprenyl
carrier through the biosynthesis
of diNAcBac
(PglF,
E, D) and
glycosyltransferases (PglC, A, J, H, I) resulting in the formation of the heptasaccharide (Figure
7-2A). The glycan product is then transported across the inner membrane by the flippase PglK
238
and transferred to either a nascent or fully-folded protein by the oligosaccharyltransferase PglB.
Whereas bacterial N-linked systems such as C. jejuni are targeted for glycosylation by the
consensus sequence Asp/Glu-X 1 -Asn-X2-Ser/Thr (where X, and X2 represent any amino acid
except Pro), characterization of similar sites for O-linked structures have proven to be more
challenging (13).
However, domains rich in serine, threonine, and proline that are localized to
turn regions in proteins have previously exhibited the ability to accept glycan moieties (14). 0linked protein glycosylation of the N. gonorrhoeae system has recently been shown to act on
periplasmic proteins responsible for protein folding, disulfide bond formation, and solute uptake
(15). These findings illustrate the otherwise unknown importance of O-linked glycosylation to
the overall fitness of this bacterium. The pathogen N. gonorrhoeaeutilizes a biosynthetic system
analogous to C. jejuni to synthesize the O-linked trisaccharide (Figure 7-2B).
In N.
gonorrhoeae, assembly of this glycan occurs in a stepwise fashion on an undecaprenyldiphosphate carrier through the biosynthesis of UDP-diNAcBac
glycosyltransferases (PglB
3).
GTD,
(PglD, C, BATD)
and
A, E) resulting in the formation of the trisaccharide (Figure 7-
PglB is a bifunctional enzyme consisting of an N-terminal phosphoglycosyltransferase
domain (PGTD) and a C-terminal acetyltransferase domain (ATD). The glycan product is then
transported across the inner membrane by the flippase PglF and transferred to a protein by the
oligosaccharyltransferase PglO (7,16).
239
,w I *
(A)
PglFE.DLUD
UDP
SdC0.J
Cvtol
R1.
i.o
PglJ
fr46W
PglH.
0
0.
;'-$
;. ,
k- 1117-1
"I
I
Pepiplasm
V NN-diacetylbacillosamine
4 Glucose
Undecaprenyldiphosphate
PEB3
(B)
UD
PgDCBATD
U PgID,C,B UDP4
P03 PgIA
lanY__
C
PglE,
_F
PerasmrPgO
0
N-acetytglucosamine
if
Galactose
N,N-diacetylbacillosamine
Undecaprenyldiphosphate
PiE
Figure 7-2. (A) C. jejuni N-linked protein glycosylation pathway. (B) N. gonorrhoeae0-linked
protein glycosylation pathway.
OHck
PgID
HO
HO
-
4
O-UDP NAD
HO
AD
NAD
/
O-UDP
NADH
UDP-GIcNAc
HO
HOA H
-U
UDP-4-keto
PMP
PLP
L-Glu
a-KG
PgBATD
N
O-UDP
AcHN
HOU
Ac-CoA CoA
UDP4amino
O-UDP
UDP-dINAcBac
OH
OH0
HO
OOH
OH
HO
0OH
OH
ACHN
NH
NH
H
PgIA,E,F,O
JtN
AcHN
Ser63
AcHN
0
PgBPGTD
AcHN
Pg'\
O-PPUnd UMP UndP
Und-PP-diNAcBac
0-linked glycan In N. gonorrhoea
Figure 7-3. Biosynthesis of UDP-diNAcBac and Und-PP-diNAcBac in N. gonorrhoeae. ATD,
acetyltransferase domain; PGTD, phosphoglycosyltransferase domain
240
The inadequacies of screening libraries to find novel antibacterial inhibitors for serious
pathogens have resulted in the need for new approaches in antimicrobial drug discovery (17).
One such approach is based upon the screening and identification of small molecule compounds
(MW < 250 Da) that can act as a starting point for further elaboration. This fragment-based
approach relies on the identification of weak-binding inhibitors (Kd> 0.1 mM) and the ability to
extend the molecule into proximal sites to increase overall potency (18).
Despite their weak
affinities, fragments possess the necessary binding energy to overcome the significant barrier due
to the loss of rigid body entropy upon protein binding (19).
This important phenomenon is
monitored throughout the fragment optimization process by the calculation of ligand efficiency
(LE). Ligand efficiency can be defined as LE = -AG/HA where AG is the free energy of binding
of a compound (calculated from its IC 50) and HA is the number of heavy (non-hydrogen) atoms
in the compound (20). In order to obtain a small molecule (MW < 500 Da) with nanomolar
potency, ligand efficiency must remain above 0.30 kcal/mol. This fragment-based methodology
has previously proven to be successful in the generation of novel, potent inhibitors to targets of
therapeutic importance (21-22).
A fragment approach involves three main steps:
an initial
biophysical screen to establish lead compounds, an in vitro activity assay to determine compound
potency, and X-ray crystallography to aid in the further optimization of fragments into adjacent
binding pockets (Figure 7-4). Iterative rounds of synthesis, potency validation, and visualization
of active site binding through X-ray crystallography are necessary to improve inhibitor efficacy.
241
Fluorescence-based thermal shift screen
Does the fragment stabilize PglB-ATD?
NMR spectroscopy
Does AcCoA displace the fragment?
In vitro activity assay
Does the fragment inhibit PglB-ATD?
X-ray crystallography
Does the fragment bind in a druggable site?
ESynthesis.
Figure 7-4. Fragment-based approach for the development of novel, potent inhibitors to targets
of therapeutic importance.
Recent work has focused on the development of small molecule inhibitors of the C. jejuni
acetyltransferase PglD aided by the previously solved crystal structures (23). Fragments were
identified with millimolar potency bound in the acetyl coenzyme A (AcCoA) and UDP-4-amino
binding pockets. With this paradigm in place, a fragment-based approach was initiated with the
homologous N. gonorrhoeae acetyltransferase PglB-ATD located on the C-terminus of the
bifunctional enzyme PglB. This chapter details the structural and biochemical characterization
of PglB-ATD with the goal of identifying small molecule inhibitors using a fragment-based
screening approach. This enzyme was screened with a series of small-molecule fragments to
ascertain their ability to inhibit acetyltransferase activity.
Furthermore, this protein was
crystallized and the structure solved for purposes of a fragment-based inhibitor approach.
A
crystal structure of PglB-ATD in the presence of a second generation inhibitor from the PglD
fragment project was also solved. This work establishes the inhibition of PglB-ATD by small
242
molecules utilizing a fragment-based approach and lays the groundwork for future generations of
inhibitors for this protein.
Results and Discussion
Expression and Purificationof PgB-A TD
PglB from N. gonorrhoeae is a bifunctional enzyme containing an N-terminal
phosphoglycosyltransferase domain (PGTD) and a C-terminal acetyltransferase domain (ATD)
that are homologous to the C. jejuni enzymes PglC and PglD, respectively (7). For biochemical
and crystallographic studies, the membrane-bound phosphoglycosyltransferase domain was
removed based upon a Clustal Omega alignment with PglD, thus leaving behind the
acetyltransferase domain referred to herein as PglB-ATD. For biochemical studies, PglB-ATD
was cloned from the FA1090 strain of N. gonorrhoeae and ligated into the pET-24a(+) vector
containing an N-terminal T7 tag and a C-terminal His6 tag for characterization and purification
purposes. For crystallographic experiments, a modified pET-30b(+) vector was utilized that
contained an N-terminal His8 tag followed by a tobacco etch virus (TEV) protease site.
Following overexpression in E. coli BL21 RIL cells and purification with Ni-NTA resin, 15 mg
of pure protein was isolated from 1 L of culture. Purity for the PglB-ATD protein was assessed
by SDS-PAGE to be > 95% (Figure 7-5). This provided a suitable amount of well-behaved,
soluble protein in the absence of the N-terminal phosphoglycosyltransferase domain, which is
predicted by TMHMM to contain a single transmembrane domain (24). Functional analysis of
PglB-ATD definitively confirmed that this protein acetylates UDP-4-amino to produce UDPdiNAcBac, which is a substrate for PglB-PGTD mediated transfer of P-diNAcBac to Und-P.
243
34 kDa
17 kDa
Figure 7-5. 12% SDS-PAGE gel of the acetyltransferase domain of PglB. Expected molecular
weight of PglB-ATD is 24 kDa.
PglB-A TD Enzyme Characterizationand Assay Development
A continuous, in vitro assay relying on the generation of the TNB2- chromophore from
Ellman's reagent (DTNB) was employed to measure acetyltransferase activity. This method was
previously utilized for developing assays for PglD (Chapter 6) and Weel (Chapter 2). Based
upon previous experiments that showed a dependence
on MgCl 2 concentration for
acetyltransferase activity (PglD and Weel), PglB-ATD enzyme activity was measure in the
presence of varying amounts of the divalent metal. Surprisingly, MgCl 2 had a negligible effect
on enzyme activity when compared with Pg1D and WeeI (Chapter 6) (Figure 7-6). Binding
coefficients for PglB-ATD were then determined at 2 mM MgCl 2 , however there was no
substantial change in Km for either substrate when adding MgCl 2 . This could be attributed to the
fact that only minor changes in activity were observed with the addition of this metal ion.
Typical Michaelis-Menten kinetics were observed over a range of AcCoA (2 - 0.02 mM) and
UDP-4-amino (2 - 0.02 mM) concentrations at saturating concentrations of the other substrate
(Figure 7-7).
Kinetic parameters listed in Table 7-1 are the outcome of initial velocity
244
measurements repeated in duplicate. A final titration of enzyme at the respective substrate Km
values resulted in a linear assay at 1 nM of PglB-ATD over two minutes.
4.5
4
3.5
2.5
S 21
0.5
0
5
10
2.5
0.625
1.25
0.3125 0.15625
0
[MgCI2 ) (mM)
PglB-ATD activity with respect to varying concentrations of MgCl 2 . Higher
Figure 7-6.
concentrations of metal resulted in a negligible increase in enzyme activity.
(A)
(B)
........
4000
4000
3000
E
2
2000
0
I2000
1000
0
400
800
1200
1600
[UDP-4-amino] (pM)
2000
0
400
800
1200
[AcCoA] (pM)
1600
Figure 7-7. Representative Michaelis-Menten binding curves for the N. gonorrhoeae
acetyltransferase PglB-ATD. Kinetic characterization was conducted at 50 mM HEPES pH 7.4,
2 mM MgCl 2 , 0.05 % BSA, 0.00 1 % Triton X-100, 1 mM DTNB, and 1 nM PglB-ATD. UDP4-amino (A) was varied in the presence of 10 x Km of AcCoA. AcCoA (B) was varied in the
presence of 10 x Km of UDP-4-amino.
245
Table 7-1. Steady-state kinetic parameters for N. gonorrhoeae,A. baumannii,and C. jejuni
acetyltransferase enzymes (from Chapter 3).
acetyltransferase
PgIB-ATD
substrate
UDP-4-amino
AcCoA
Km f.(LM)
99.0 ± 7.1
286 ±35
kc (s-')
7.2 0.8 x 104
5.0 0.7 x 104
kcst/Km (M-1 s1 )
7.3 x 108
1.7 x 108
Weel
UDP-4-amino
2520± 540
5
5.1 ± 0.08 x 10
2.0 x 108
PgID
UDP-4-amino
AcCoA
AcCoA
78.9
28
1.3
0.06 x 10 5
1.6 x 10 9
274
295
6.4
2.8
8.0
6.1
1.6 x 10s
1.0 x 10 5
2.9 x 10'
2.1 x 109
PglB-A TD FragmentScreening Results
Through a collaboration with Professor Alessio Ciulli at the University of Cambridge
UK, a fluorescence-based thermal shift assay was utilized for the screening of 400 fragment
molecules against PglB-ATD.
Samples of fragment (4 mM) and protein (4.5 pM) were
incubated with Sypro Orange, which is a solvatochromic dye that fluoresces upon binding to
unfolded protein. The samples were heated in a thermal cycler and the change in fluorescence
was monitored as a function of temperature. Fragments that bind to protein have a stabilizing
effect on protein melting temperature leading to a greater ATm. Six fragment hits (Figure 7-8)
were further validated by employing 1H WaterLOGSY and saturation transfer difference (STD)
NMR spectroscopy for their ability to displace the substrate AcCoA.
These biophysical
techniques are utilized as a secondary measurement for fragment binding.
Following this
validation step, these molecules were further validated in the PglB-ATD DTNB in vitro activity
assay. These compounds inhibited acetyltransferase activity in the single-digit millimolar range
(Figure 7-8). Not surprisingly, a majority of these molecules were also identified in the original
PglD C. jejuni acetyltransferase fragment screen. Although MB272 was not detected in the PglD
screen, a close analog containing a furan substituent (MB2 11) in place of the thiophene moiety
246
was identified. MB211 exhibits weak binding to PglD (IC 50 = 8 mM) and binds into the AcCoA
binding site as observed from a PglD-MB211 crystal structure (Dr. James Morrison, unpublished
results).
When tested in the PglB-ATD assay, MB211 demonstrated an increased ability to
inhibit this enzyme (IC 50 = 1.6 mM) that was similar in potency to the thiophene analog MB272.
These results are similar to the differences observed in the thienopyrimidine compounds
(Chapter 6) further validating the diversity within N- and 0-linked AcCoA binding pockets.
0
0
N
HO ~
0OH
H
0
N-NH
MB272
ATM = 2.05 *C
IC50 = 1.4 mM
L.E. =0.30
N
MB129
ATm= 0 *C
MB143
AT. = 2.73 'C
IC 50 > 5 mM
L.E.= NA
IC 50 = 2.3 mM
L.E. = 0.26
0
OH
OH
HO
MB254
ATm 1.36 OC
Figure 7-8.
0
O
-
IC 50 = 3.1 mM
MB048
ATm 1.37 0C
1C 50 =4.4 mM
L.E = 0.23
L.E. =0.27
PglB-ATD fragment melting and IC 50 results.
MB201
AT 1 1= 5.00 C
IC 50 = 6.0 mM
L.E. = 0.22
NA
=
Not Available due to
compound solubility issues and the inability to measure an IC 50 .
Second GenerationFragmentInhibition of PglB-A TD Activity
Small molecule inhibitors based upon the MB211 core scaffold were previously
synthesized for the PglD fragment project by Dr. James Morrison (all compounds are properly
characterized and in his final report). These 6 compounds were screened for activity against
PglB-ATD.
One molecule exhibited 80% inhibition of PglB-ATD activity at 2 mM. This
247
compound, jma35 and its regioisomer jm_a34, were chosen for IC 50 determination based upon
this result (Figure 7-9).
Surprisingly, jm-a35 displays weak inhibition towards PglD
acetyltransferase activity (IC 50 = 3.9 mM) with respect to PglB-ATD (IC 50
=
0.74 mM).
Interestingly, the phenyl substituent on the pyrazole ring has a substantial effect on inhibitor
potency (Figure 7-9).
No such relationship is observed with inhibition of PglD as both
compounds are equipotent.
28
26
24
~22
~20
0,02
0,01
0
'~
S16
14
12
0
10
0.01
01
1
00 I
1
N'N/
OH
OH
jma35
jma34
IC 50 = 0.74 mM (PgIB-ATD)
ICO = 3.9 mM (PgID)
ICSO = 5.73 mM (PgIB-ATD)
IC 50 = 3.6 mM (PgID)
Figure 7-9. Inhibition of PglB-ATD activity with MB211 analogs from the PglD fragmentbased inhibitor project.
Analogs of jma35 were further pursued due to the ability of this compound to inhibit
PglB-ATD activity. In particular, the addition of a 3-nitro substituent on the phenyl moiety
(jma48) resulted in a 15-fold increase in potency (IC 50 = 48.6 pM) (Figure 7-10A). To verify
this inhibition, a capillary electrophoresis (CE) assay was established that followed the turnover
of CoASH (< 20%) over a 15 minute time period. Titration of jma48 in this assay format
resulted in an IC 50 value of 20.7 pM, confirming the original finding in the DTNB assay (Figure
248
7-10B). Corroborating previous results between the dichotomy of PglD and PglB-ATD AcCoA
binding pockets, jm-a48 exhibited poor inhibition with respect to PglD (IC 50 = 5 mM).
Reduction of jma48 resulted in the meta-aniline compound jma65, which maintained the
potency of jm a48 (Figure 7-11). Other molecules explored different substitutions in the meta
position on the phenyl ring as well as para substituted versions of these compounds.
Unfortunately, substitution at the para position (excluding jm a74) or different substitutions at
the meta position resulted in a loss of inhibitor activity (Figure 7-11, 7-12).
(A)
0.1
(B) 10
012
14
0'1
12
0
~004
4
~0
i
I
10
100
12
F
...
~d
10
1
bma48 (A)
IC
DTNB Assay
50= 48.6 pM
100
Dm.a48 (UM)
0.-.
OH
IC
CE Assay
50 = 20.7 pM
jma48
Figure 7-10. Inhibition of PglB-ATD activity with jma48 in the DTNB (A) and capillary
electrophoresis (B) assay format.
249
NO 2
/
\NO
/
2
_NH2
0
/
O
OH
IC50
\
OH
jm a48
5.0 mM
= 49 pM,
0
OH
jm_a65
IC50 = 76 pM, 600 pM
jm a67
IC =120 pM, 3.2 mM
NH 2
/ \
CF3
NPNH
N-N
OH
OH
jma70
IC 50 = 190 pM, > 2mM
OH
jm_a71
jm a74
ICO = 58 pM, 280 g.M
IC 50= 270 pM, > 2mM
IC50 = PgIB-ATD, PgID
Figure 7-11. IC 5 o analysis for second generation analogs of MB211 with PglB-ATD and PglD.
-H
NO 2
R
0
N-N
NH12
1600
0.30
740
0.22
49
0.27
7028
.,NO2
HO
0
1
120
0.24
190
0.22
270
0.21
58
0.29
H
Figure 7-12.
ATD.
Structure-activity relationship of MB211 second generation analogs with PglB-
250
PglB-A TD-Boundjm a65 Crystal Structure
The PglB-ATD crystal structure was previously solved to 1.7 A resolution in the cubic
space group P2 13 with a single protomer in the asymmetric unit (Chapter 4). Difficulties in
crystallization of this protein were addressed by removing the final ten amino acid residues from
the C-terminal tail based upon a sequence alignment with PglD. The removal of these PglBATD residues, which are not present in corresponding PglD sequence, results in a comparable Cterminal tail between the two constructs. These crystals (Figure 7-13) were robust enough to
withstand DMSO concentrations up to 3% for up to 14 days, making this an ideal form for PglBATD soaking experiments.
The second generation acetyltransferase inhibitor jm_a65 was
soaked for 6 days at a 1.5 mM concentration. Longer soak times resulted in extremely fragile
crystals that were prone to cracking and deterioration.
A dataset was collected to 2.1 A at
Boston University on a Bruker AXS Proteum-R instrument with a Platinum 135 CCD area
detector.
Additional electron density in the AcCoA binding pocket was accounted for by
jm-a65. The main hydrogen-bonding interactions between PglB-ATD and compound occur at
the carboxylate moiety of jma65 (Figure 7-14).
Only two residues are involved in these
interactions, the side chain from Arg360 and the backbone amide nitrogen from A381. From a
structural alignment standpoint, the corresponding Arg360 residue in PglD is Phe 152, which may
partially be responsible for such poor binding affinity of jma65 to PglD. Two water molecules
are also present in the active site, one of which has a hydrogen-bonding interaction with the
carboxylate of jma65. Interestingly, jma65 is located further down the AcCoA binding pocket
with respect to the MB21 1-PglD structure (Figure 7-15A). Based upon an alignment between
the AcCoA- and jm-a65-bound PglB-ATD structures, the aniline moiety of jma65 is
superimposed onto the adenine base in AcCoA (Figure 7-15B). Therefore, jm a65 may serve as
251
a mimic for AcCoA due to their similarities in structure (Figure 7-16). However, this structure
does not explain the gain in potency by placing an amino or nitro group at the 3-position. In the
jm-a65 structure, the amino group has no interaction with PglB-ATD, which is a surprising
result in light of a 10-fold increase in potency from the addition of moiety in relation to jm-a35.
This may be due to soaking this compound into preformed crystals, as a conformational change
may be necessary to realize the full potency of this molecule. Further co-crystallization studies
with these PglB-ATD inhibitors will be necessary to understand how these molecules bind into
the AcCoA binding pocket to elicit such a change in potency.
Figure 7-13. Representative PglB-ATD crystals utilized for inhibitor soaking experiments.
252
GIn369
H2N
Ar360
3.0
h
H
NH.3
3
NH
:28
H2N'
'0
27 '
AWa381
NH
NH
3.1
2.8
SN
NH2
jm-a65
Figure 7-14. The PglB-ATD crystal structure with jm a65 bound into the AcCoA binding
pocket (left) and an illustration of the major interactions between protein and compound (right).
Distances between interactions are given in angstroms.
(A)
(B)
Figure 7-15. (A) Overlay of the PglD-MB211 and PglB-ATD-jm a65 crystal structures. The
second generation MB2 11 analog jm_a65 binds in a different position with respect to MB211.
For purposes of clarity, MB21 1 is colored in gray and jm a65 in brown. (B) Overlay of the
PglB-ATD-AcCoA and PglB-ATD-jm_a65 crystal structures. The aniline moiety ofjma65
overlaps with adenine moiety from AcCoA. For purposes of clarity, AcCoA is colored in gray
and jm_a65 in brown.
253
NH 2
'H
HO,
0
0
H
O
N
o
HOAH
HO H
HO -
N
N
NH 2
0
N
HO
N
O'H
AcCoA
jm a65
Figure 7-16. Structural comparison of the Pg1B-ATD substrate AcCoA and the inhibitor
jm-a65.
Conclusions
In conclusion, PglB-ATD was biochemically and structurally characterized for the
purposes of finding small molecules that inhibit acetyltransferase activity. Previously
synthesized compounds from the C. jejuni PglD fragment-based inhibitor project were utilized to
examine efficacy on PglB-ATD activity. Surprisingly, these second generation analogs of
MB21 1 exhibited an increased level of potency when compared to PglD. These results confirm
previous observations that N- and O-linked acetyltransferase AcCoA binding pockets exhibit a
high level of diversity, which is reflected by the lack of promiscuity in small molecule inhibitors.
A crystal structure of jma65 bound to PglB-ATD was solved to high resolution, revealing how
these molecules bind into the AcCoA binding site. Surprisingly, these compounds seem to
mimic the adenine moiety of this cosubstrate and bind in a different location than the original
MB2 11 fragment as observed in PglD. Future work will focus on how certain amino and nitro
substituents in the 3-position affect potency to such a high degree. A PglD crystal structure
bound to an MB211 analog would also aid in the understanding of how acetyltransferases that
254
catalyze identical reactions can exhibit such dichotomous behavior in their respective binding
pockets.
Acknowledgments
I am extremely grateful to Dr. James Morrison for the PglD-MB211 structure and synthesis of all
the second generation MB21 lanalogs. I would like to thank Professor Alessio Ciulli for the
generation of the initial PglB-ATD fragment leads. I would also like to thank Andrew Lynch
and Dr. Jeffrey Bacon (Boston University) for help with data collection of the jma65-bound
PglB-ATD structure. Lastly I would like to thank Austin Travis and Vinita Lukose for critical
reading of this chapter.
Experimental Procedures
Common materials
All chemicals were purchased from Sigma-Aldrich unless otherwise stated. The UDP-4amino sugar was biosynthesized as described previously from the C. jejuni enzymes PglF13 o and
PglE (11).
PglB-A TD MolecularBiology
The acetyltransferase domain (ATD) of the pgB gene from N. gonorrhoeaeFA1090 was
identified through a Clustal Omega alignment (25) with the C. jejuni acetyltransferase (PglD).
The gene encoding this domain was amplified via the polymerase chain reaction (PCR) with the
forward primer 5'-CGCGGATCCATGGCGGGGAATCGCAAACTCG-3'
primer 5'-GCAACCCGGCAAAGCCCCTTTAGCTCGAGCGG-3'
255
and the reverse
from the N. gonorrhoeae
FA1090 strain (16).
BamHI and XhoI restriction sites were engineered to facilitate cloning of
each construct into a modified pET30b(+) vector (Novagen) containing an N-terminal His8 tag
followed by a tobacco etch virus (TEV) protease site prior to the BamHI site. Amplifications
were accomplished with the PfuTurbo DNA Polymerase (Stratagene) as described by the
manufacturer. Amplicons were purified and double-digested with BamHI and XhoI restriction
enzymes (NE Biolabs). Digested inserts and linearized vectors were fractionated by agarose gel
electrophoresis and purified with the Wizard SV Gel and PCR Cleanup Kit (Promega). Ligations
were conducted with the T4 DNA ligase kit (Promega) using a 15 min incubation at room
temperature. Sequencing by Genewiz (Cambridge, MA) confirmed the presence of all gene
products.
PglB-A TD Expression and Purification
The modified pET30b(+) plasmid containing the pglB-ATD gene was used to transform
Escherichiacoli BL21 (DE3) RIL competent cells (Stratagene). One liter of LB media containing
50 pig/mL kanamycin and 30 pg/mL chloramphenicol was inoculated with 8 mL of an overnight
culture of cells. The cells were then allowed to grow at 37 OC while shaking until an optical
density of ~0.8 (k = 600 nm) was reached. The culture was cooled to 16 C and induced with 0.5
mM iso-p-D-thiogalactosylpyranoside (IPTG). After incubating for 18 h with shaking at 16 C,
the cells were harvested by centrifugation (2600g, 30 min) and stored at -80 0C until needed.
Each protein purification step was carried out at 4 C.
The cell pellet (-3 g) was
resuspended in 40 mL of 50 mM HEPES pH 7.4/100 mM NaCl/30 mM imidazole (Buffer A)
and then lysed by sonication. The lysate was then cleared by centrifugation (145,000g, 60 min)
and added to 2 mL of Ni-NTA resin (Qiagen). The slurry was allowed to tumble for 3 h and then
256
packed into a fritted PolyPrep column (Biorad). The resin was washed with 20 column volumes
of Buffer A and then eluted with a buffer containing 50 mM HEPES pH 7.4, 100 mM NaCl, 300
mM imidazole. Fractions containing the purified protein as analyzed by SDS-PAGE were
pooled and dialyzed against 50 mM TRIS pH 8.0, 5 mM EDTA, 5 mM P-mercaptoethanol in the
presence of 6 pM TEV protease for 24 h to remove the His8 tag. Removal of this tag was
monitored by Western blot analysis using an anti-His 4 antibody (Qiagen). The reaction was
diluted 10-fold in 25 mM HEPES pH 7.6 and excess TEV was then removed with a HiTrap
Q
HP Sepharose anion exchange column (GE Healthcare) utilizing a linear NaCl gradient.
Fractions containing the protein were pooled and dialyzed for 24 h in 50 mM HEPES pH 8.0,
150 mM NaCl (SEC buffer). After concentrating to a volume of 1.5 mL using a 10K Da MWCO
Amicon Ultra-15 centrifugal filter unit (Millipore), the protein was loaded onto a Superdex 200
16/60 column (GE Healthcare) and subjected to size-exclusion chromatography in SEC buffer.
Fractions
containing
the monodispersed
protein
were
pooled
and
concentrated
for
crystallography experiments. Protein concentration were calculated based upon the predicted
extinction coefficients at k = 280 nm.
Crystallizationand Data Collection
All crystals were grown as hanging drops by combining 1.5 ptL of a 10 mg/mL protein
solution in SEC buffer with 1.5 ptL of reservoir solution at 25 C. Each well contained a final
volume of 500 pL of reservoir solution. The reservoir solution for PglB-ATD contained 0.1 M
sodium acetate pH 4.6, 0.02 M calcium chloride, and 30% 2-methyl-2,4-pentanediol (MPD). For
the crystallization soaking experiments with PglB-ATD and jm a65, the compound was added to
preformed crystals at 1.5 mM and allowed to incubate at 25 0C for 6 days.
257
After soaking, the
crystals were cryoprotected in reservoir solution containing 20% glycerol and 1.5 mM jm-a65.
Diffraction data was collected on the Boson University home source with a Bruker AXS
Proteum-R instrument and a Platinum 135 CCD area detector. Data sets were processed using
HKL2000 (26), MOSFLM (27), TRUNCATE (29-30), and SCALA (29).
Structure Determinationand Refinement
Preliminary electron density maps for the PglB-ATD-jm-a65 structure were generated in
PHASER (31) utilizing the previously solved PglB-ATD structure (4M98) as the molecular
replacement search model. Refinement and model building of each structure was accomplished
with COOT (32) and PHENIX (33). Water molecules were added using COOT and jm a65 was
modeled into PglB-ATD after the Rfree value was < 30%. Jm_a65 ligand and geometry restraints
were generated using eLBOW.
Refined structures were validated using MolProbity (34).
Composite omit maps for the jm a65-bound PglB-ATD structure were generated with PHENIX.
PglB-A TD DTNB Activity Assay
Kinetic characterization of PglB-ATD was carried out using a previously modified
procedure (7).
CoASH generation resulting from the acetyltransferase reaction carried out by
PglB-ATD was monitored in the presence of Ellman's reagent (DTNB) through the generation of
the TNB2- chromophore in a continuous fashion. To a black-clear bottom 96-well plate (Falcon)
was added 50 mM HEPES pH 7.4, 2 mM MgCl 2 , 0.05% BSA, 0.001% Triton X-100, 1 mM
DTNB, and 1 nM PglB-ATD.
Reactions were completed in duplicate and initial rates were
measured in the linear portion of the reaction curve over a 2 minute time period at 25 C. The
substrate concentrations of AcCoA and UDP-4-amino were varied separately to determine
258
kinetic parameters using initial velocity measurements while holding the other substrate at a
saturating level. Steady-state rate parameters were calculated from equation 1 using the program
GraFit 6.0.12 (Erithacus Software).
The kinetic parameters are a result of duplicate
measurements at each substrate concentration.
IC 50 values were calculated from equation 2
using the program GraFit 6.0.12 (Erithacus Software) and measured in duplicate.
V = Vmax[S]/(Km + [S])
(1)
y = 100%/(1 + (X/IC 50 )')
(2)
PglB-ATD CapillaryElectrophoresisActivity Assay
Capillary electrophoresis analysis was performed using a P/ACE MDQ system (Beckman
Coulter) with UV detection. The capillary was conditioned before each run successively with
0.4 M NaOH, water, and a 25 mM sodium tetraborate (pH 9.3) running buffer for 2 minutes
each. The reaction mixture contained 50 mM HEPES pH 7.4, 50 nM PglB-ATD, 99 ptM UDP-4amino, and 286 pM AcCoA.
The reaction was initiated with enzyme and incubated for 10
minutes. The reaction was stopped by filtering off the enzyme with a 10K MWCO spin filter
membrane. The filtrate was injected for 15 s at 30 mbar and the analytes separated at 20 kV over
a 45 minute time period on a bare silica capillary (75 ptm x 80 cm) with a 25 mM sodium
tetraborate (pH 9.3) running buffer and monitored at a X = 254 nm. Substrate and product peaks
were manually integrated utilizing the Beckman 32 Karat software suite.
259
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Barber, M. (1947) Staphylococci infection due to penicillin-resistant strains. BMJ 2, 863865.
Tapsall J.W. (2009) Neisseria gonorrhoeae and emerging resistance to extended spectrum
cephalosporins. Curr. Opin. Infect. Dis. 22,87-91.
Workowski K.A., Berman S.M., and Douglas J.M. (2008) Emerging antimicrobial resistance
in Neisseria gonorrhoeae: urgent need to strengthen prevention strategies. Ann. Intern. Med.
148,606-613.
Rottingen J.A., Cameron D.W., and Garnett G.P. (2001) A systematic review of the
epidemiologic interactions between classic sexually transmitted diseases and HIV: how much
is really known? Sex. Transm. Dis. 28,579-597.
Craig L., Pique M.E., and Tainer J.A. (2004) Type IV pilus structure and bacterial
pathogenicity. Nat. Rev. Microbiol. 2, 363-378.
Craig L., Volkmann N., Arvai A.S., Pique M.E., Yeager M., Egelman E.H., and Tainer J.A.
(2006) Type IV pilus structure by cryo-electron microscopy and crystallography:
implications for pilus assembly and functions. Mol. Cell. 23, 651-662.
Hartley, M. D., Morrison, M. J., Aas, F. E., Borud, B., Koomey, M., and Imperiali, B. (2011)
Biochemical Characterization of the O-Linked Glycosylation Pathway in Neisseria
gonorrhoeae Responsible for Biosynthesis of Protein Glycans Containing N,N'Diacetylbacillosamine. Biochemistry 50, 4936-4948.
Stimson E., Virji M., Makepeace K., Dell A., Morris H.R., and Payne G. (1995)
Meningococcal pilin: a glycoprotein substituted with digalactosyl 2,4-diacetamido-2,4,6trideoxyhexose. Mol. Microbiol. 17, 1201-1214.
Schoenhofen, I.C., McNally, D.J., Vinogradov, E., Whitfield, D., Young, N.M., Dick, S.,
Wakarchuk, W.W., Brisson, J.R., and Logan, S.M. (2005) Functional characterization of
dehydratase/aminotransferase pairs from Helicobacter and Campylobacter. J. Biol. Chem.
281, 723-732.
Vijaykumar, S., Merkx-Jacques, A., Ratnayake, D.B., Gryski, I., Obhi, R.K., Houle, S.,
Dozois, C.M., and Creuzenet, C. (2006) Cj 1121c, a novel UDP-4-keto-6-deoxy-GlcNAc C-4
aminotransferase essential for protein glycosylation and virulence in Campylobacterjejuni. J.
Biol. Chem. 281, 27733-27743.
Olivier, N.B., Chen, M.M., Behr, J.R., and Imperiali, B. (2006) In vitro biosynthesis of UDPN,N'-diacetylbacillosamine by enzymes of the Campylobacter jejuni general protein
glycosylation system. Biochemistry 45, 13659-13669.
Varki A. (1993) Biological role of oligosaccharides: all of the theories are correct.
Glycobiology 3, 97-130.
Grogan M.J., Pratt M.R., Marcaurelle L.A., and Bertozzi C.R. (2002) Homogeneous
glycopeptides and glycoproteins for biological investigation. Annu. Rev. Biochem. 71, 593634.
Julenius K., Molgaard A., Gupta R., and Brunak S. (2005) Prediction, conservation analysis,
and structural characterization of mammalian mucin-type O-glycosylation sites.
Glycobiology 15, 153-164.
Vik A., Aas F.E., Anonsen J.H., Bilsborough S., Schneider A., Egge-Jacobsen W., and
Koomey M. (2009) Broad spectrum O-linked protein glycosylation in the human pathogen
Neisseria gonorrhoeae. Proc. Natl. Acad. Sci. 106, 4447-4452.
260
16. Aas F.E., Vik A., Vedde J., Koomey M., and Egge-Jacobsen W. (2007) Neisseria
gonorrhoeae O-linked pilin glycosylation functional analyses define both the biosynthetic
pathway and glycan structure. Mol. Microbiol. 65, 607-624.
17. Payne D.J., Gwynn M.N., Holmes D.J., and Pompliano D.L. (2007) Drugs for bad bugs:
confronting the challenges of antibacterial discovery. Nat. Rev. Drug Discov. 6, 29-40.
18. Ciulli A., and Abell C. (2007) Fragment-based approaches to enzyme inhibition. Curr. Opin.
Biotechnol. 18, 489-496.
19. Page M.I., and Jencks W.P. (1971) Entropic contributions to rate accelerations in enzymic
and intramolecular reactions and the chelate effect. Proc. Natl Acad. Sci. 68, 1678-1683.
20. Hopkins A., Groom C.R., and Alex A. (2004) Ligand efficiency: a useful metric for lead
selection. Drug Discovery Today 9, 430-431.
21. Saxty G., Woodhead S.J., Berdini V., Davies T.G., Verdonk M.L., Wyatt P.G., Boyle R.G.,
Barford D., Downham R., and Garrett M.D. (2007) Identification of inhibitors of protein
kinase B using fragment-based lead discovery. J. Med. Chem. 50, 2293-2296.
22. Howard N., Abell C., Blakemore W., Chessari G., Congreve M.S., Howard S., Jhoti H.,
Murray C.W., Seavers L.C.A., and van Montfort R.L.M. (2006) Application of fragment
screening and fragment linking to the discovery of novel thrombin inhibitors. J. Med. Chem.
49, 1346-1355.
23. Olivier, N. B., and Imperiali, B. (2008) Crystal structure and catalytic mechanism of PglD
from Campylobacterjejuni. J. Biol. Chem. 283, 27937-27946.
24. Krogh A, Larsson B, von Heijne G, Sonnhammer EL. Predicting transmembrane protein
topology with a hidden Markov model: application to complete genomes. J. Mol. Biol. 2001;
305:567-580.
25. Sievers F., Wilm A., Dineen D. G., Gibson T. J., Karplus K., Li W., Lopez R., McWilliam
H., Remmert M., S~ding J., Thompson J. D., Higgins D. G. (2011). Fast, scalable generation
of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol.
7:539 doi:10.1038/msb.2011.75.
26. Otwinowski, Z., and Minor, W. (1997) Processing of X-ray Diffraction Data Collected in
Oscillation Mode. Methods Enzymol. 276, 307-326.
27. Leslie, A. G. W., and Powell, H. R. (2007) Processing diffraction data with mosfim.
Evolving Methods for Macromolecular Crystallography. 245, 41-5 1.
28. Winn M. D., Ballard C. C., Cowtan K. D., Dodson E. J., Emsley P., Evans, P. R., Keegan, R.
M., Krissinel, E. B., Leslie, A. G., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu,
N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A., and Wilson, K. S. (2011)
Overview of the CCP4 suite and current developments. Acta. Crystallogr. D Biol.
Crystallogr. 67, 235-242.
29. French, S., and Wilson, K. (1978) On the treatment of negative intensity observations. Acta
Crystallogr. A34, 517-525.
30. McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., and
Read, R. J. (2007). Phaser crystallographic software. J. Appl. Cryst. 40, 658-674.
31. Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K. (2010) Features and development of
Coot. Acta Cryst. D66, 486-501.
32. Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J.
J., Hung, L. W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W.,
Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T.C., and Zwart, P.
261
H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure
solution. Acta Cryst. D66, 213-221.
33. Chen, V. B., Arendall, W. B. 3rd, Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G.
J., Murray, L. W., Richardson, J. S., and Richardson, D. C. (2010) MolProbity: all-atom
structure validation for macromolecular crystallography. Acta. Cryst. D66, 12-21.
262