1. Coupling Transcrip1on and DNA Repair with a dsDNA-­‐
Tracking Motor 2. Post-­‐Transla1onal Modifica1on of Tubulin by Tubulin Tyrosine Ligase (TTL) Miller Spread Courtesy of Dr. Sarah French
g
l pCdo i n
Lodish et al., Molecular Cell Biology
u p c i pu A
u
po
i i ng N
Miller Spread Courtesy of Dr. Sarah French
A Stalled RNA Polymerase Can Cause a Molecular Pile-up
AFM of transcription
complexes on
irradiated template
Figure 2. ppGpp Reduces Arrays of RNAP
(A) Effect of ppGpp on wild-type RNAP. Lane 1, lcro +
lanes 2–6 and 7–11, time courses showing complexe
15, and 20 min in the absence or presence of ppGpp
(B) Effect of UV on promoter activity. Reactions were
Trautinger et al., (2005) Molecular Cell 19, 247-254
Adapted from: http://www.cawlocal584.com/humour.html
•  3000 RNA Polymerase molecules/cell
•  10-20 DNA Polymerase III molecules/cell
“head-to-tail”
fast
“head-on”
slow
RNAP
DNAP
DNAP
RNAP
Double strand breaks
Lodish et al., Molecular Cell Biology
Miller Spread Courtesy of Dr. Sarah French
pc p ng
CCl i s ugg wgi
. o nt
1u a
d
l pCdo i n ng CC1u A
•
•
yl c p me R bl o l NeNl oe -­‐ NrR rR m y Na l o p RNo mc
p
No rR rmo e mN ermo (yml ee rRml h R
u
Miller Spread Courtesy of Dr. Sarah French
RNAP Stalling Leads to Genomic Instability
•
•
•
•
•
RNAP/DNAP collisions result in double-­‐strand breaks DNA repair shapes the muta6onal landscape of cancer cells DNA repair is the main mechanism underlying the development of cancer in all living organisms Also essen6al for the mechanism of chemotherapeu6c agents (irofulven) Essen6al for the e6ology of accelerated-­‐aging diseases (Cockayne Syndrome) Deaconescu et al., (2012) Trends Biochem, Sci.37(12) 543-­‐552 Nudler, E. (2012) Cell 149(7) 1438-­‐1445 Nik-­‐Zainal et al., (2012) Cell 149(5) 994-­‐1007 kpi
Ml i pg 1u
Mfd (mutation frequency decline)
•  mutant is UV sensi6ve •  3-­‐fold lower rate of excision of pyrimidine dimers Witkin EM (1966) Science 152, 1345-­‐1353 Mfd is the TRCF (transcription-repair coupling factor)
Proc. Nati. Acad. Sci. USA
Vol. 88, pp. 11574-11578, December 1991
Biochemistry
Escherichia coli mfd mutant deficient in "mutation frequency
decline" lacks strand-specific repair: In vitro complementation
with purified coupling factor
(trnsciption-repair coupllng/UV mutgenes/SOS rse/nonsense suppreors)
CHRISTOPHER P. SELBY*, EVELYN M. WITKINt,
AND
AZIZ SANCAR*
*Department of Biochemistry and Biophysics, University of North Carolina School of Medicine, Chapel Hill, NC 27599; and tWaksman Institute of
Microbiology, Rutgers, The State University of New Jersey, P.O. Box 759, Piscataway, NJ 08854
Contributed by Evelyn M. Witkin, September 26, 1991
ABSTRACT
Mutation frequency decline (NWD) is the
rapid decrease in the frequency of certain induced nonsense
suppressor mutations occtnng when protein sythesis
translendy inhibited Immediately fter irrtin. MD is
abolihed by mutations inthe uvrA, -B. or -C genes, which
prevent excision repair, or by a mfd mutation, which reduces
the rate of excision but does not affect survival. Using an in vitro
repair synthesis assay we found that although wild-type ells
ibed (template) strand frentially, mfd
repair the
cells are incapable of d-pcfic repar. The iciency in
strand-selective repair of ,*d- cell extract was cor d by
adding highiy purified "transcription-repair coupling fator"
We1conclude
that mfd is, most likely,
Selby et toathe
l., (reaction
1991) Pmixt.
NAS (88) 1574-­‐11578 the gene encoding the tnscription-repai coupling factor.
as reversion to prototrophy. Nonsense suppressor mutations
account for nearly all ofthe UV-induced reversions, and only
the suppressors exhibit MFD (12-14). Bockrath and colleagues (15-18) conducted a series of elegant genetic experiments on the MFD effect and based on the results of these
experiments concluded that "MFD is a unique process
involving excision repair of premutational lesions located
only in the transcribed strand of DNA" (16). The apparent
strand specificity of MFD led us to consider whether the mnfd
gene might encode or control the synthesis of the TRCF we
detected in our in vitro assay. Therefore, cell-free extracts
from' E. coli B/r and its mfd derivative were tested for
strand-specific repair in vitro. We found that E. coli B/r, like
E. coli K-12, was capable of strand-specific repair. In contrast, E. coli B/r mfd- extract was totally deficient in
Main Players
ATP
TRCF
ADP + Pi
mo e mNyal obm y Nm l hy No
(
u
l c p me
RNAP
momc l o a l o
l p y n (
u
N Rr m yNmyml r Noe
o
( dm g dm u
RNAP
TRCF
UvrB B
A UvrA
lo eh
lo eh
(/ ) P/ u
t (/ ) ) Su
(
kS(P/ uvf kbvv/
D PS(Pu0EbP) /
rl m
u
TRCF Mediates Transcription-Coupled Repair (TCR)
TRCF
RNAP
ATP
ADP + Pi
TRCF
RNAP
UvrB B
A UvrA
TRCF
UvrC, UvrD,
DNA Pol I, DNA ligase
lo eh
lo eh
ct D Do
(/ ) P/ u
t (/ ) ) Su
o mt D(P00ku
ea
vf kbvv/
DPS(Pu0EbP) /
t / E) (vP) f u6vkb2
Ques1on How are the different TRCF ac1vi1es integrated? Multiple Activities Are Integrated in a Multi-Domain Protein
ATPase motifs
D2/D7 clamp
•
•
kD/ m el hal o
g m 3 / kDf g/ 0Dv 4
ATP binding site
l o e h r D
(/ ) ) EuP/ f (ku6v) Sbv/ )
TRCF Induces Transcription Bubble Collapse
D2/D7 Clamp
si
u C k (M
Translocation
translocation Domains
domains
D5/D6
RID (RNAP
Interacting
wedge
domain
po pCp d
p aC
Domain)
-­‐ l u ug
gi unCp n k yh L)
RNAP
TRCF!
RNAP
RNAP
RNAP
RNAP
Park et al. (2002) Cell 109, 757!
RNAP binding domain
The D2-D7 Clamp Is Inhibitory to Translocation on
Naked DNA
ATPase motifs
TRCF!
f
lo eh
(/ ) P/ u
lo eh
(/ ) ) Su
hmyRc
(/ ) ) 0u
lo eh
(/ ) ) Eu
ea
vf kbvv/
DPS(Pu0EbP) /
kSt P2
P/ f (ku6v) Sbv/ )
UvrA Recruitment Requires Opening of the D2/D7 Clamp
UvrA Binding
B. caldotenax UvrB (PDB ID 1T5L)
E. coli TRCF (PDB ID 2EYQ)
TRCF-D2: TRCF-D7
TRCF-D2: Truncated UvrA
SAXS Probes Structure in Solution
Adapted from Petoukhov, M., EMBL lecture
•  Measure isotropic intensity distribution and radially average I(q)
•  Extract shape parameters:
Rg (low q data)
Dmax (maximum intramolecular distance)
a C pT
-­‐ i
gg
n o C ng
4πssin / Deaconescu et al., PNAS (2012) 109 (9): 3353-3358
i dng C gi a gai
-­‐ pau
g g 1n 1g u
(Å-1)
•  Reconstruct 3D-envelope using ab initio algorithms starting with gas phase of
“dummy atoms/residues” in a spherical search volume followed by minimization
and fit against the scattering curve
•  SYSTEM IS UNDERDETERMINED, MULTIPLE SOLUTIONS ARE POSSIBLE
The D2/D7 Clamp is Maintained in Solution in the
Nucleotide-Free State
a C pT -­‐ i
-­‐ ( bP c 1g p i dng Cngi a gai
( l ANo -­‐ e y ml mp
heNo o nRh ead e m R ermao
ml p mo l p l o hma l ou
The Catalytic Cycle Reorganizes Interdomain Contacts
Deaconescu et al., PNAS (2012) 109 (9): 3353-3358
The Catalytic Cycle Reorganizes Interdomain Contacts
No DNA binding
Dmax = 124 ± 5Å
Rg =37.6 ± 0.2 Å
DNA binding
Dmax = 140 ± 5Å
Rg =38.2 ± 0.1 Å
No DNA binding
Dmax = 126 ± 5Å
Rg =36.8 ± 0.1 Å
Deaconescu et al., PNAS (2012) 109 (9): 3353-3358
ATPTRCF Simulations are Robust and Reproducible
•
Use the PDB of apo-TRCF as starting model (input) to seed
the simulation process
Ab initio simulation
PDB-seeded simulation
Probing Clamp Opening Using Interdomain Disulfide
Engineering
-­‐ f R (
fiL(
fiPyfi
fiPbfi
yf (
-­‐ ( R
TRCF Variant
Crosslinking w/ ATP
Crosslinking w/ UvrA
TRCF-D7:RID
(open, more extended)
+++
+++
TRCF-D2:D7
(closed, more compact)
---
---
The D2/D7 Clamp Is Ideally Positioned to Restrain the
TRCF Motor and Provide Spatiotemporal Control
•  Prevents UvrA Binding
•  Compromises DNA binding
•  Impairs ATP turnover (in the
absence of RNAP)
•  Still releases RNAP off templates
due to stimulation by the RNAP
elongation complex 
e
llA
c
la
N e
Nn
E. coli TRCF
la
UvrA may be recruited after release of RNAP/transcript from
the elongation complex
u1no p
i un i 1l Tpu-­‐ pal C
l y oNo l rR
/ g S p y
dm / / m mhNrp or
n NeNl o
l 1i
i
m e
1 Tpu
er No
Rc ml c eNe
l m-­‐ m
rmo e l a l o
Arm ANo
No No
Post-translational modification of tubulin by a tubulin
tyrosine ligase (TTL)
(a collaboration with the Roll-Mecak Lab, NIH)
http://schulmanlab.jhu.edu/research.html
http://micro.magnet.fsu.edu/cells/microtubules/
microtubules.html
Microtubules are Dynamic
Conde & Caceres (2009) Nat Rev Neuroscience
MiMicrotubules are Heterogeneous
Tyrosination
Carsten Janke & Jeannette Chloë Bulinski
Nature Reviews Molecular Cell Biology (2010) 12, 773-786
Tubulin Tyrosine Ligase (TTL) Modifies the
α-Tubulin Tail
subsequent addi≠
ulin [Barra et al.,
conserved penul≠
aturle≠ Lafanechere
[Lí Hernault and
senbaum, 1985],
in [Edde et al.,
er et al., 1992;
a≠ and b≠ tubulin
of b≠ tubulin [Eip≠
lin [Caron, 1997;
001]. Acetylation
ently [Chu et al.,
als that we do not
bulin post≠ transla≠
uliní s post≠ transla≠
Fig. 1. Tubulin post≠ translational modifications. Ribbons
reased metazoan
representation of the tubulin dimer (PDB 1TUB) [Nogales
d variety of post≠
et al., 1998] with a≠ and b≠ tubulin colored green and blue,
Garnham & Roll-Mecak.
specially complex
respectively. Cytoskeleton
The unstructured C≠ terminal tails were modeled to
69, Issuetheir
7, pages
span 442-463
and are colored red. The location and type
or cilia. (2012) Volume illustrate
of known post≠ translational modifications is indicated on the
e is evident both
N ature Structural & Molecular Biology doi:10.1038/nsmb.2148
A Putative Mechanism for Tyrosination
1. Phosphorylation of the α-carboxylate
2. Nucleophilic attack on the
phosphorylated carboxylate
TTL-How Does It Work, and Why Is It Important?
•
•
•
•
Important for organization of neural networks
Microtubules in dendrites are enriched in Tyr tubulin
TTL suppression leads to cancer
TTL prefers to tyrosinate free tubulin heterodimer over tubulin that
has been incorporated into a lattice (e.g. microtubules)
•  TTL inhibits tubulin polymerization
Szyk et al., (2012) NSMB
TTL Binds Tubulin Not only Through Its Tail
Kd (tubulin) ~ 1 µM
Kd (peptide) ~ 144 µM
a
b
Asp165
1.5
Glu168
Pro102∆
∆Asn260
Arg73
∆Asp126
Lys70
Arg46
Arg44
Arg66
Arg32
Arg29
Normalized activity with peptide
Tyr228∆
1.0
0.5
His54
X. tropicalis TTL (Monomer)
Szyk et al., (2012) NSMB
reserved.
0
WT
K70A
R73A
R73E
R4
R4
Lys31
Figure 3 Molecular determinants of tubulin tyrosination. (a) TTL molecular surface
stick atomic representation. (b) Normalized relative tyrosination activity with A-tubu
TTL mutants (n q 3). Error bars indicate s.e.m.
What is the basis for this substrate specificity?
α
β
-
longitudinal
Heterodimer
versus
α
β
Polymer
α
β
α
β
+
(lateral, between protofilaments)
Hypothesis
TTL binds to a surface that is buried in the polymeric tubulin form
The TTL-Tubulin Complex is Elongated
TTL
α
Szyk et al., (2012) NSMB
?
or
TTL
α
The TTL Contacts Primarily α-Tubulin
Is the tubulin heterodimer flipped?
-  AUC confirmed complex formation and 1:1 stoichiometry
-  From frictional ratio aspect ratio is 3.8 (versus 2.2 for tubulin) denoting
an elongated structure
Szyk et al., (2012) NSMB
Antibodies against α-Tubulin Inhibit Tyrosination
Szyk et al., (2012) NSMB
TTL May Modulate the Partitioning of
Heterdimeric versus Polymeric Tubulin and
Tune Motors and End-Binding Proteins
Szyk et al., (2012) NSMB
The Grigorieff Lab (Brandeis University)
Dr. Niko Grigorieff
Drs. J. Gelles and J. Kondev
Collaborators
Artsimovitch Lab (Ohio State University)
Dr. Irina Artsimovitch
Dr. Anastasia Sevostyanova
Roll-Mecak Laboratory
(NIH-NINDS)
Dr. Antonina Roll-Mecak
Dr. Aga Szyk
The Nicastro, Miller, Cohen, Gelles and Petsko-Ringe Labs (Brandeis University)
The Nogales Lab (UC Berkeley)
Staff at Sibyls Beamline (ALS)
Staff at Beamline 8.2.1 (ALS)