Transcriptional Profiling and Flux Measurements of
Polyhydroxybutyrate Production in Synechocystis
by
Saliya Sudharshana Silva
Master of Science in Chemical Engineering Practice
Massachusetts Institute of Technology, 2002
Master of Engineering Science
University of Oxford, 1998
Submitted to the Department of Chemical Engineering
in partial fulfillment of the requirements for the degree of
Master of Science in Chemical Engineering
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
September 2004
( 2004 Massachusetts Institute of Technology. All rights reserved.
Author
..............
.................................... .. .... ............
Departmentof Whmic
1
/
Certified by ................................................
ginering
June 7, 2004
.... .
rego'Steph'hopoulos
Gre'
Bayer Professor of Chemical Engineering
Thesis Supervisor
Accepted by ....................
MASSA.,HUSEfTTS INSTU1
O' rzTECHNOLOY
I
E
....................
Daniel Blankschtein
Professor of Chemical Engineering
Chairman, Committee for Graduate Students
JUN 1 6 2004
.
m
,..i
A
F
.
LlbKA ~ltt=
j
I
ARCHIVES
Transcriptional Profiling and Flux Measurements of
Polyhydroxybutyrate Production in Synechocystis
by
Saliya Sudharshana Silva
Submitted to the Department of Chemical Engineering
on June 7, 2004 in Partial Fulfillment of the
Requirements for the Degree of Master of Science in
Chemical Engineering
Abstract
The metabolism of Synechocystis PCC6803 cells has been investigated using fullgenome DNA micro-arrays and C14 tracer techniques. Full-genome (3169 genes) DNA microarrays were used to probe transcript levels of Synechocystis cells grown under a variety of
medium conditions. Canonical discriminant analysis was used to identify transcript levels that
allowed discrimination between growth media conditions, and allowed predictions of
polyhydroxybutyrate (PHB) levels. Phosphate-related genes were found to alter in response to
phosphate limitation and were found to include differentially regulated multi-gene families.
Nitrogen-related genes were not found to be substantially reflective of nitrogen limitation under
the conditions studied. Finally, transcription of PHA biosynthetic pathway genes were found to
reflect the media conditions of greatest PHB accumulation, suggesting that constitutive overexpression of the PHA biosynthetic genes may lead to greater PHB accumulation levels.
A methodology using C14 tracers was developed for the accurate measurement of CO2
uptake rates and the partitioning of the fixed carbon into different biosynthetic fractions. These
techniques were applied to the characterization of WT Synechocystis cells in late exponential
phase. A stoichiometric model of Synechocystis metabolism was used to determine constraints
between the measurements. A balance on C14 counts was obtained and significant levels of
secreted compounds were not detected. The measured carbon fixation rates were found to be
consistent with the observed growth rates, but inconsistent with measurements of oxygen
evolution in the light and uptake in the dark made using a Clarke Electrode apparatus. These
techniques may be used in future studies to determine the CO 2 uptake rates and PHB production
rates of strains engineered for enhanced CO2 fixation and PHB production respectively.
Thesis Supervisor: Gregory Stephanopoulos
Title: Bayer Professor of Chemical Engineering
2
Table of Contents
ABSTRACT ..................................................................................................................................................................
......................................
2
TABLE OF CONTENTS...................................................................................
3
LIST OF FIGURES ...................................................................................
4
LIST OF TABLES ...................................................................................
4
1. INTRODUCTION ...................................................................................................................................................
5
1.1
1.2
1.3
1.4
SYNECHOCYSTIS & CYANOBACTERIA AS PRODUCTION VEHICLES FOR COMPOUNDS .........................................
SYNECHOCYSTIS METABOLISM AND CARBON CONCENTRATING MECHANISMS .................................................
POLYHYDROXYBUTYRATE AND POLYHYDROXYALKANOATES............................................................................
THESIS OVERVIEW ...................................................................................
2. METHODS & TECHNIQUES ...................................................................................
11
2.1 CULTURE CONDITIONS AND GROWTH MEDIA ........................................
...........................................................
2.2 TRANSCRIPTIONAL PROFILING USING DNA MICRO-ARRAYS ............................................................................
2.3 ' 4C-BICARBONATE LABELING AND BIOMASS FRACTIONATIONS ......................................................................
3. TRANSCRIPTIONAL
5
5
7
10
11
11
13
PROFILING OF PHB SYNTHESIS BY WILD TYPE SYNECHOCYSTIS ............ 17
3.1 GROWTH PHASE AND MEDIA CONDITIONS MICRO-ARRAY ANALYSIS .............................................................
3.2 FISHER DISCRIMINANT ANALYSIS TO DISTINGUISH BETWEEN PHYSIOLOGICAL STATES ..................................
17
18
3.3 GENESPREDICTIVEOF PHB ACCUMULATION
........................................
...........................................
20
3.4 CHANGES IN PHOSPHATE AND NITROGEN RELATED GENES ..............................................................................
22
3.5 CHANGESIN PHB-RELATED GENES...................................................................................
3.6 SUMMARYANDCONCLUSIONS........................................
..........................................
24
25
4. CHARACTERIZATION
OF LATE EXPONENTIAL
PHASE SYNECHOCYSTIS
METABOLISM USING C14 TRACERS ........................................
.....................................
...........................................
27
27
4.1 DESIGN OF CO 2 -LABELING SYSTEM .......................................
............................................
4.2 LABELING OF WILD TYPE LATE EXPONENTIAL PHASE CELLS AT 5% CO 2.
....................................
27
28
4.3 RESULTSOF FRACTIONATION
ASSAYS........................................
30
...........................................
4.3 FORMULATION OF STOICHIOMETRIC MODEL FOR SYNECHOCYSTIS METABOLISM
..............................
4.4 RECONCILIATION OF EXPERIMENTAL DATA AGAINST STOICHIOMETRIC MODEL AND DETECTION OF GROSS
MEASUREMENT
ERRORS........................................
...........................................
4.5 SUMMARY AND REMAINING ISSUES FOR IMPROVING FLUX QUANTITATION .....................................................
5. FUTURE RESEARCH DIRECTIONS.....................................................................................
5.1
5.2
5.3
5.4
5.5
CHARACTERIZATION OF WILD TYPE CARBON FLUX PARTITIONING .................................................................
INVESTIGATION OF THE ROLE OF ACETATE IN STIMULATING PHB PRODUCTION .............................................
INVESTIGATION OF THE PHB REMOBILIZATION PATHWAY ................................................................................
DIRECTED MUTANTS TO INCREASE THE FIXATION RATE OF CO 2........................................................................
DIRECTED MUTANTS TO INCREASE THE PRODUCTION RATE OF PHB .................................................................
31
32
33
35
35
36
37
38
38
REFERENCES........................................
...........................................................................................................................................................39
APPENDIX A ......................................................................................
43
APPENDIX B .....................................................................................
55
APPENDIX C ......................................................................................
57
3
List of Figures
FIGURE 1-1. SIMPLIFIED REPRESENTATION OF SYNECHOCYSTIS METABOLIC PATHWAYS..............................................
FIGURE 1-2. INORGANIC CARBON SPECIES INTERCONVERSIONS AND CYANOBACTERIAL CARBON CONCENTRATING
MECHANISMS.........................................................................................................................................................
........................................
FIGURE 1-3. MAJOR BACTERIAL POLYHYDROXYALKANOATE MONOMER UNITS .............................................................
FIGURE 1-4. PATHWAY OF PHB BIOSYNTHESIS AND KNOWN REGULATION MECHANISMS ............................................
FIGURE 3-1. DISCRIMINATION OF CELLS GROWN UNDER DIFFERENT MEDIA CONDITIONS ............................................
FIGURE 3-2. CORRELATION OF CV2 AGAINST THE PHB CONTENT OF THE CELLS (% DRY CELL WEIGHT) ....................
FIGURE 3-3. EXPRESSION LEVELS OF PHOSPHATE-RELATED GENES .............................................................................
FIGURE 3-4. EXPRESSION LEVELS OF NITROGEN-RELATED GENES ..............................................................................
FIGURE 3-5. EXPRESSION LEVELS OF PHB BIOSYNTHESIS-RELATED GENES AND PHB ACCUMULATION LEVELS.........
FIGURE 4-1. EXPERIMENTAL SETUP TO ENSURE LABELING AT CONSTANT 14C-BICARBONATE SPECIFIC ACTIVITY .......
FIGURE 4-2. COUNTS TAKEN UP BY THE WT LATE EXPONENTIAL PHASE CELLS ..........................................................
FIGURE 4-3. THE BEST CONSERVATION OF COUNTS OBSERVED IN EACH OF THE CELLULAR FRACTIONS ......................
FIGURE 5-1. GROWTH CURVE, GLYCOGEN CONTENT (% CDW) AND CO 2 UPTAKE RATES (G/G CDW/HR) ......................
FIGURE 5-2. ROLE OF ACETATE IN ENHANCING PHB ACCUMULATION IN SYNECHOCYSTIS..........................................
FIGURE 5-3. MODEL FOR THE PHB REMOBILIZATION PATHWAY .................................................................................
6
7
8
9
20
21
23
24
25
29
30
31
35
36
37
List of Tables
TABLE 3-1. Summary of growth conditions and PHB accumulation..................................................
17
TABLE 3-2. Top 30 discriminatory genes: gene name, family, proposed role for PHB predictive
genes.........................
.......................
.......................................................................
TABLE 4-1. Equilibrium inorganic carbon pool sizes for BG11 medium...............................................
TABLE 4-2. C
2
22
28
uptake rate, PHB & glycogen contents & production rates for WT late exponential phase
cells..................................
..........................................................................................
4
30
1.
Introduction
1.1 Synechocystis& Cyanobacteriaas ProductionVehiclesfor Compounds
Concern over the use of non-renewable resources to synthesize bulk chemicals has lead
to interest in their production by photosynthetic organisms using carbon dioxide and light.
Cyanobacteria are a group of photosynthetic prokaryotes belonging to the subclass of gramnegative eubacteria, whose photosynthetic apparatus bears remarkable similarity to eukaryotic
chloroplasts [1].
Species in the genus Synechocystis are unsheathed unicellular spherical
cyanobacteria which divide successively by binary fission in more than one plane [2].
Synechocystis PCC6803 is a model unicellular cyanobacterium for two reasons: first, its
extremely
efficient
natural genetic transformation
capability; second,
its ability to grow
photoheterotrophically on glucose, a characteristic that is necessary in order to study strains or
mutants deficient in photosynthesis-related functions [3, 4]. Few other cyanobacteria share these
characteristics. Finally, the sequencing of the entire genome of Synechocystis PCC6803 [5] has
opened new possibilities for the elucidation of pathways and the metabolic steps therein.
Production of polyhydroxybutyrate (PHB), a polyoxoester of interest as a biodegradable
polymer, has been investigated in the unicellular freshwater cyanobacterium Synechocystis as a
case study of chemical production in a photosynthetic organism.
1.2 Synechocystis Metabolism and Carbon Concentrating Mechanisms
Figure 1-1 shows the principal cyanobacterial metabolic pathways of concern, pooling
metabolites to show the branch points in the network. The major pathway of photosynthetic
carbon fixation in cyanobacteria is via the Calvin cycle [6].
Ribulose-1,5-bisphosphate
carboxylase-oxygenase (RUBISCO) carboxylates ribulose-1,5-bisphosphate to two molecules of
3-phosphoglycerate.
Intermediates of varying carbon chain length derived from 3-
phosphoglycerate are then diverted for synthesis of amino acids as shown (lipid synthesis is not
shown here). Secondary pathways of carbon fixation via phosphoenolpyruvate carboxylase have
been reported to account for up to 20% of the total inorganic carbon fixed by cyanobacteria [7].
5
I
lycogen
CO 2
_2'
ar IY
Jeflb
%lUf
t
,,
-
Y
KW
'
v
--
^
v
·-
:
e
-.-
^1·
11r
I-·
a
1
1
He
ILMS,&I=j
Il:
serint
gychw,
pkelnyalanine
ahnine
Valine
tzyptepkan
leucite
aspartate
PEP/PYR
metkionine
tkneonte
isoaucine
lhtvste
'
AcCoA
a
t
tarate
ghtamate
OAA-.
Figure 1-1. Simplified representation of Synechocystis metabolic pathways. The main CO 2fixing reactions are via RUBISCO (R5P -> 3PG) and PEP carboxylase (PEP -) OAA). The
main sinks for fixed carbon dioxide are glycogen, PHB and other biomass constituents.
Cyanobacteria are thought to have evolved at a time where atmospheric carbon dioxide
concentrations were significantly higher than at present [8]. Figure 1-2 shows a model for the
carbon-concentrating mechanism thought to raise the local CO2 concentration at the site of CO2fixation. Bicarbonate transported across the cell membrane diffuses to the carboxysome, where
it is converted to CO2 by the action of carbonic anhydrase and fixed into phosphoglycerate by the
action of RUBISCO. CO2 leakage from the cell is minimized by a scavenging mechanism
utilizing carbonic anhydrase activity at the cell membrane.
6
H2C03 (aq)
3.0x10-' s-/
11.9 's4.7x
10
lM1 1Iss-N
8x10
6
s-1
CO3 2- (aq) + 2H+
(
Figure
1-2.
Kinetics
of aqueous
inorganic
carbon
species
interconversions,
model of
bicarbonate uptake and carbon concentrating-mechanisms present in cyanobacteria.
1.3 Polyhydroxybutyrate and Polyhydroxyalkanoates
A diverse range of bacterial strains synthesize PHAs as a means for storing carbon and
energy, the balance between synthesis and depolymerisation of PHAs being tightly regulated by
the physiological state of the cell. Figure 1-3 depicts the chemical structure of the major
polyhydroxyalkanoate monomer units found in bacteria [9]. Poly-3-hydroxybutyrate (PHB) is a
straight chain homopolymer of D(-)-3-hydroxybutyrate, found as a reserve compound in a wide
variety of micro-organisms [10]. A few of the bacterial strains that have been employed for the
production of PHB in particular include Ralstonia eutropha, Alcaligenes latus, Azotobacter
vinelandii, several strains of methylotrophs, and recombinant Escherichia coli [11]. The carbon
to nitrogen (C/N) ratio is a critical factor triggering PHA accumulation under excess carbon
conditions, or utilization of the polymer in the opposite case. Given the widespread production
of PHAs as storage polymers by bacteria, many microbes have evolved the ability to
depolymerize PHAs in the environment [12]. Thus, PHAs have received much attention in the
past several years regarding their use as biodegradable plastics [13].
7
R
0
i
V
R A
R =
R
R
AR
AR
R =
R =
R *
R
yogen
Ohy
eyl
propyl
'Buty
penty
hexyl
heptyl
ocy
nonyl
3-hydroxypropale
3.hydroxybutyrate
3hydroxyvaerat
3-hydroxycaproate
3-hydtoxypanoale
3-hy
yoctanoate
3-hydroxynonanoate
3-hydroxydecnoate
3-hdroxyurnecanoae
3-hdroxydodo
aes
[-o
n
n
I
- C4 -CH2 -C-q
3
4
(3HP)
(3H8)
(3HV)
(3HC)
(3HH)
(3HO)
(3HN)
(3HD)
(3HUD)
(3H)
FC,4-)C-]
4-hydroxybutyrae
5-hydroxyvalerale
(4HB)
[SHV)
Figure 1-3. Chemical structure of the major bacterial polyhydroxyalkanoatemonomer units 191
Production of these polyesters through bacterial fermentation is not currently costcompetitive with petrochemical-based commodity polymers such as polyethylene or
polypropylene [13]. The cost of the carbon source and, for copolymer production, cosubstrates
have been identified as contributing significantly to the costs of production [11]. Furthermore,
life-cycle analysis comparisons of oil-derived polymers and PHA production from corn-derived
fermentations have found that the fermentation processes compare unfavorably in terms of total
environmental impact [14].
As a result their production by photosynthetic carbon dioxide
fixation in plant systems has been investigated [15].
However, since cyanobacteria are
indigenously the sole organisms exhibiting PHA accumulation by oxygenic photosynthesis [16],
their use offers several advantages for PHA production over that in plants.
Namely,
cyanobacteria exhibit significantly higher growth rates and CO2 fixation rates than higher
photosynthetic plants, and are easier to work with in terms of their genetic and phenotypic
plasticity.
Figure 1-4 depicts a model for the cyanobacterial PHB biosynthetic and regulatory
pathway based upon studies in Alcaligenes eutrophus [17-19]. PHB synthesis is catalyzed by the
sequential action of three enzymes. 3-ketothiolase catalyzes the reversible condensation of two
acetyl-CoA moieties to form acetoacetyl-CoA.
This step is inhibited by CoA [20, 21]. NADPH-
dependent acetoacetyl-CoA reductase then reduces the acetoacetyl-CoA to R-3-hydroxybutyryl8
+
CoA. This step is thought to be regulated by the cellular NADPH/NADP
ratio [22, 23]. The R-
3-hydroxybutyryl-CoA is then polymerized by PHA synthase to form PHB. The synthase step is
thought to be activated post-translationally by phosphopantetheine [24]. Acetyl phosphate has
been
hypothesized
[16] to activate
the native
PHB
synthase
in the cyanobacterium
Synechococcus MA19. The enzymes in this pathway exhibit broad specificity, such that higher
This is utilized in the
chain length monomeric units can also be incorporated into the PHA.
commercial production of BiopolTM, a random copolymer of 3-hydroxybutyrate and 3hydroxyvalerate produced by fermentation of Alcaligenes eutrophus on glucose and propionic
acid.
ilvNBG
Pyrwate
I
A.
.a
M
T
cetohctate
~1
Acetyl- CoA
rrr
Extnrceuar
OtS
>
CoA
Aeetae
P.A __
Acetate
- ---- -__
,ewacel
NADPH,
eA.
Pheqphate
e
phaB
-.
Hltmxybutyryl-Co
- --
phaIC
t iLtyl
.
acdyl.iCoAsyrtas e
A
~S~ia8 i ,,-,,,,,IFPH
acwtolactdaantie
(--_ 1___
prz2
phosphdrcaoa
acetdae kiras
phaA
3-ketolthlm e
paB
ac0
Wax
-CAredictas e
PHA syniha
P~PPPPP~2C~~PPPPP~~~~~
Figure
1-4. Pathway ofPHB biosynthesis
Figure 1-4. Pathway of PHB biosynthesis and known regulation mechanisms.
The native PHA synthase of Synechocystis PCC6803 is a two-component protein, similar
to that of anoxygenic purple sulphur bacteria [25]. While the properties of this synthase have not
been investigated, homologous synthases from other organisms preferentially polymerize
coenzyme A thioesters of short chain length (3-5 carbon atoms, [26]).
9
1.4 Thesis Overview
The physiology of wild-type Synechocystis has been investigated using full-genome
microarrays and 14 C-bicarbonate tracers.
Chapter 3 describes the results of transcriptionally
profiling cells grown under a variety of medium conditions, in an attempt to identify genes that
correlate with PHB levels.
The important parameters
for commercial
PHB production in
Synechocystis are the CO2 fixation rate and PHB yield. Chapter 4 describes the use of 14Cbicarbonate tracers to characterize the autotrophic metabolism of
exponential-phase
Synechocystis, with special emphasis on measuring the CO2 uptake rate and determining the
partitioning of the fixed carbon into PHB and other cellular constituents. Chapter 5 describes a
set of environmental perturbations and genetic transformants that may be characterized using
these developed methods.
10
2. Methods & Techniques
2.1 CultureConditionsand GrowthMedia
Growth on Plates. Cells were plated onto solid media (1:1 ratio of 2% agar and 2x BG11),
supplemented either with 5 mM glucose or with 20 mM HEPES and 23 mM NaHCO3 and grown
in an environment with 5% CO2.
Growth in Liquid Media. Batch cultures of Synechocystis PCC6803 (WT) (Pasteur Culture
Collection) were maintained at 30 °C in BG11 medium (SIGMA, St. Louis, MO.). Throughout
each experiment, continuous irradiance of ca. 250 mol photons m '2 s-1 was provided by cool
white fluorescent bulbs. All growth experiments were performed in shake flask cultures in a
light-tight incubator (Percival, Perry, IA). Full BGl1 medium was diluted to 0.3% (0.3N) and
10% (1ON) of the full BG 1 nitrogen level or 10% of the full BG 11 phosphate levels (OP).
In
addition, identical full and diluted cultures were supplemented with 10 mM Acetate (0.3NA,
O1NA, O1PA).Cultures grown at 5% CO2 were further supplemented with 20 mM HEPES and
23 mM NaHCO 3 .
2.2 TranscriptionalProfilingusing DNA Micro-arrays
DNA Microarrays.
Full-length PCR amplified gene products for nearly every gene in the
Synechocystis genome were provided by Dupont Co. PCR products were resuspended in 50%
DMSO, spotted using a BioRobotics quill pin micro-arrayer onto Coming GAPS slides (Acton,
MA), cross-linked using a UV stratalinker, and stored in the dark until use.
RNA Purification. RNA was purified using Qiagen Mini, Midi, and Maxi kits. Immediately
after transfer from the growth culture into 50 ml polypropylene centrifuge tubes, cells were
placed into liquid N2 and chilled to <5 °C within 20 seconds.
Chilled cells were immediately
centrifuged at 4000 x g for 3 minutes in a pre-cooled centrifuge (4
C), supernatant was
discarded, and cell pellets were immediately frozen in liquid N2 prior to permanent storage at 20 °C. Cell pellets were resuspended in buffer RLT (Qiagen) and an equal volume of 0.1 mm
glass beads (B. Braun Biotech, Inc., Allentown, PA.) and ground in a bead-mill (B. Braun) for
four cycles of 1 minute grinding and 1 minute on ice. All grinding was performed in a walk-in 4
11
°C cold room. Lysed cells were then purified following the exact protocols of the Qiagen RNA
purification kit. To remove carbohydrates and chromosomal DNA, a final precipitation using
4M LiCI was performed.
This modified RNA purification procedure produced results
comparable to those previously described for light-dark studies of Synechocystis [27].
cDNA creation and Labeling. RNA were reverse transcribed to fluorescently labeled cDNA
using 15 U Superscript II/ug RNA, 1X superscript buffer, 1X DTT, 0.5 mM dCTP, dATP, dGTP,
0.2 mM dTTP, and 0.1 mM of Cy-dUTP (Amersham-Pharmacia Biotech, Sweden). Reverse
transcription was performed for 2 hours at 42 °C. 1.5 g 1 N NaOH was added and RNA
template was degraded at 65 °C for 10 minutes followed by neutralization with 1 N HCL. Cy3
and Cy5 labeled sample and control cDNA were mixed and ETOH precipitated. Precipitated
cDNA was resuspended in 16 ul (32 ul for whole-genome arrays) of pre-warmed 65°C Clontech
hybridization buffer and denatured for 10 minutes at 95 °Cprior to applying to the micro-arrays.
Hybridization and Scanning. Micro-arrays were denatured for 2 minutes in 95 °C H20 and
flash-cooled in -20 °C ETOH. After heat denaturation, labeled cDNA was flash cooled in an
ice-slurry and briefly spun at 7000 x g to collect evaporated liquid. A small aliquot (1 ul) of
cDNA was removed for spectrophotometric diagnostics and the remaining was carefully pipetted
over the micro-array. A Clontech glass slide was placed over the hybridization solution and
hybridization was performed in a water bath overnight at 50°C in water-tight humidified
hybridization chambers (Coming, Acton, MA). Arrays were washed in an excess of XSSC +
0.1%SDS for 5 minutes, 0.2XSSC for 3 minutes, and 0.1XSSC for 5 minutes. Cleaned arrays
were briefly washed with 1M Ammonium Acetate and immediately spun at 500 x g for 4
minutes to remove all salt deposits prior to scanning. Clean slides were scanned using the Axon
Instruments GenePix 4000B (Foster City, CA) for all full-genome arrays.
Micro-Array Data Acquisition, Filtering, and Analysis. Micro-arrays were quantified using
the GenePix Pro software from Axon Instruments. Erroneous spots were manually flagged and
removed from the final data set. All micro-array results were filtered to remove any spots in
which at least 60% of the signal pixels were not greater than one standard deviation of the local
background value for both lasers (532 nm, 632 nm). The median pixel ratio of the filtered data
12
for each spot was used for all subsequent analysis. This ratio was adjusted by the median
Cy3/Cy5 ratio automatically generated by the GenePix Pro software.
2.3 14C-Bicarbonate Labelingand BiomassFractionations
Inorganic Carbon Concentration Measurements. 0.875g NaHCO3 and 1.1025g Na2 CO3 were
dissolved in 250 ml freshly autoclaved milliQ H2 0 to obtain a 1000 mg C/L inorganic carbon
standard solution. Dilutions in the range 100-500 mg C/L were prepared with autoclaved milliQ
H2 0, and samples were run on the inorganic carbon analysis program of a Shimadzu TOC-5000
total carbon analyzer (Kyoto, Japan).
50 ml cell culture samples were cooled rapidly using
liquid nitrogen, centrifuged at 5000g for 5 minutes to precipitate the cells, and the supernatant
stored in tightly capped 15 ml tubes at 40 C prior to analysis.
Biomass Fractionations
adapted from [2811. Small molecule extracts were obtained by
extracting the pelleted cells thrice with 10 ml of 80% methanol for 1 hour, once with 50%
methanol at 60°C for 10 minutes, and 4 ml H2 0 at room temperature for 10 minutes. Small
molecule extracts were dried overnight in a Meyer N-Evap Analytical Evaporator
(Organomation
Associates
Inc, Berlin, MA) and counted in 15 ml Scintisafe plus 50%
scintillation cocktail (Fisher Chemical, Fair Lawn, NJ). Glass microfibre filters (Whatman,
Clifton, NJ) were used in all subsequent filtrations. Lipid extracts were obtained by suspending
the small-molecule extracted pellets in 10 ml ice-cold 10% TCA, filtering with 2 ml ice-cold
10% TCA, collecting the filtrate from further washes with 10 ml 70% ethanol at 45°C and 5 ml
of 1:1 ethanol-diethyl ether at 45°C. Lipid extracts were dried and counted with scintillation
cocktail as described previously. RNA extracts were obtained by suspending the small-molecule
extracted pellets in 10 ml ice-cold 10% TCA, filtering with 2 ml ice-cold 10% TCA, air-drying
the filter in a scintillation vial for 10 minutes, hydrolyzing the filter with 4 ml of 0.5 N NaOH at
37°C for 40 minutes, incubating with 1.2 ml ice-cold 50% TCA at 4°C for 30 minutes, and
collecting and counting the supernatant. DNA extracts were obtained by suspending the smallmolecule extracted pellets in 10 ml ice-cold 10% TCA, filtering, hydrolyzing the filter with 4 ml
0.5 N NaOH at 37°C for 90 minutes, incubating with 1.2 ml ice-cold 50% TCA at 4C for 30
minutes, filtering the liquid onto a new filter, hydrolyzing both filters in 7 ml 5% TCA at 80°C
for 30 minutes, adding 0.1ml of 1.0 mg/ml BSA immediately prior to incubating at 4°C for 30
13
minutes, filtering and counting the filtrate. Counts in proteins were obtained by suspending
small-molecule extracted pellets in 10 ml 5% TCA, heating at 80°C for 30 minutes to hydrolyze
RNA & DNA, cooling on ice for 30 minutes, collecting precipitated proteins on a filter, washing
the filter once with 3 ml ice-cold 70% ethanol, twice with 5 ml 70% ethanol at 45°C, twice with
5 ml ethanol-diethyl ether (1:1) at 45°C, once with 5 ml diethyl ether, and air-drying the filter
prior to counting in Scintisafe plus 50% scintillation cocktail.
Glycogen Analysis [adapted from 28, 29]. 25-50 ml pelleted culture was extracted three times
with 5 ml boiling 80% ethanol for 3 minutes in 18x150 mm glass tubes (VWR International,
West Chester, PA). The extracted pellets were transferred to Eppendorf tubes, dried overnight in
a Meyer N-Evap Analytical Evaporator (Organomation Associates Inc, Berlin, MA), and then
hydrolyzed in 500 pl 50% (w/v) KOH in a water bath at 100°C for 1 hr. 300 1l supernatant was
boiled with 700 Cplethanol in a 100°C water bath for 2 minutes, placed on ice for a further 2
minutes, centrifuged at 4°C and 12,000g for 5 minutes, and allowed to air dry. The pellet was
dissolved by heating at 100°C in 50
itlH2 0 for 2 minutes and centrifuged to pellet cell debris
carried over from the previous step. Incorporated 14C counts were determined by counting 10 il
in 15 ml Scintisafe plus 50% scintillation cocktail (Fisher Chemical, Fair Lawn, NJ)). Total
glycogen levels were assayed by adding 10 pl to 90 ll amyloglucosidase buffer (50 mM acetic
acid, 50 mM NaAcetate, pH 4.7), 0.5 ill amyloglucosidase enzyme (-3.5 U/ml), and incubating
at 57°C for 40 minutes.
ml of glucose assay buffer (85 mM Tris base, 15 mM Tris-HCI, 5 mM
MgCI 2, 0.1 mM dithiothreitol, 150 ItM ATP, 40 giM NAD+), 1 pI of G6P dehydrogenase (-0.2
U/ml) and 1 pl of hexokinase (-0.7 U/ml) were added, and the mixture incubated at room
temperature for 10 minutes, and at 30°C for a further 10 minutes. The NADPH concentrations in
200 gl resulting from the following set of enzymatic reactions:
Glycogen (Glc)
amYno-a-4--l ,6-guosidase
Glucose + ATP
Glucose-6-P + NADP +
hexokinase
>
glucose-6-P + ADP
glucose-6-Pdehydrogemase
>
14
> nglucose
6-P-gluconolactone + NADPH + H +
were measured using a plate reader (excitation 335 nm, excitation 460 nm). Glycogen standards
of concentration 2, 1, 0.5, 0.25, 0.1, 0.05, 0.025, 0.010 and 0.005 g/l were run in parallel to the
samples.
PHB Quantification [adapted from 12911. Between 50-250 ml of cultures in stationary phase
were collected by centrifugation (10 minutes at 3200 x g, 4 °C). The resulting pellet was washed
once in dH 2O and dried overnight at 85 °C. The dry pellets were boiled in 1 ml. of concentrated
H2SO4 for 1 hr., diluted with 4 ml of 0.014M H2SO4 and filtered through a PVDF filter
(Acrodisc LC13 PVDF, Pall Gelman Laboratory, Ann Arbor, MI).
100 gl samples were
analyzed on an Aminex® HPX-87H ion exclusion organic acid analysis column (Biorad®,
Hercules, CA) under running conditions 0.7 ml/min 14 mM H2 S0 4 buffer, 50°C, monitoring UV
absorbance at 210 nm. The PHB peak fraction (23-26 min) was collected in scintillation vials,
dried in a Meyer N-Evap Analytical Evaporator (Organomation Associates Inc, Berlin, MA), and
counted with 15 ml of Scintisafe plus 50% cocktail (Fisher Chemical, Fair Lawn, NJ).
Scintillation Counting. All samples for scintillation counting prepared as described above were
counted for 5 minutes in a Packard 1500 Tri-Carb Liquid Scintillation Analyzer (Downers
Grove, IL). Quenched C14 standards (Packard, Downers Grove, IL) run in parallel were used to
correct for quenching by SIS (Spectral Index of Sample) number.
15
PAGES (S) MISSING FROM ORIGINAL
PAGE 16 MISSING
3. Transcriptional Profiling of PHB synthesis by Wild Type
Synechocystis
(in collaboration with Ryan Gill and Jatin Misra)
3.1 GrowthPhase and Media ConditionsMicro-arrayAnalysis
Transcriptional profiling of cells grown under a variety of different media conditions was
performed in order to identify transcript accumulation levels that discriminate between and may
be predictive of different states of PHB accumulation. Table 3-1 lists the growth conditions and
resultant PHB accumulation levels that were investigated.
Cells were grown to late stationary
phase in full BGII media, media containing 10% nitrogen levels, and media containing 10%
phosphate levels. Cells were also grown to late stationary phase under these media conditions
with an additional 10 mM sodium acetate. Further, cells were grown to late exponential phase in
media containing 0.3% of full media nitrogen levels (with supplementary acetate).
Condition
% DCW
BG11
0.4 +/-9%
BG I + Acetate (BGA)
1.0 +/- 3%
10% Nitrogen (ION)
1.7 +/- 7%
10% Nitrogen + Acetate (10NA)
1.7 +/- 7%
10% Phosphate (OP)
2.6 +/- 28%
10% P + Acetate (10PA)
4.1 +/- 19%
0.3% N + Acetate (0.3NA)
0.8 +/- 46%
Table 3-1. Summary of growth conditions and PHB accumulation
Cells grown in full BG11 media doubled every 40 hrs, grew to a final cell density of
approximately 1.5 x 108 cells/ml, and accumulated PHB to about 0.4% of dry cell weight (DCW)
at early stationary phase. When full BGI11 was supplemented with an additional carbon source,
10 mM Acetate, the growth rate and final cell density were not altered significantly. However,
PHB accumulation levels increased approximately to 1% of DCW. Limitation in nitrogen and
phosphate yielded further increases in PHB accumulation that were enhanced in the presence of
17
acetate. More severe limitations in nitrogen (0.3%) resulted in a dramatic reduction in growth
rate and final density, resulting in exponential phase cells containing a significantly higher PHB
content (0.8%) than exponential phase cells grown in full BG11 medium. RNA yields for 0.3%
nitrogen-limited cells in stationary phase were too poor to run good quality micro-arrays.
3.2 Fisher Discriminant Analysis to Distinguish between Physiological States
Fisher Discriminant Analysis was performed in MATLAB as follows. FDA defines a
projection from the original to a reduced space that maximizes the ratio of the variance-betweengroups to the variance-within-groups. This is mathematically equivalent to maximizing the
mean separation between the various groups or classes in the reduced dimensional space. If
there are c classes in the data, the within-group-variance W and the between-group-variance B
are defined as:
W= (Xk -xk )y(Xk
Jxk
y(3-1)
)
k=l
and
where T is the total variation.
B T-W=
(X - 1))(X
- I)-
W
(3-2)
Xk and X are data matrices for samples in class k and the entire
expression set, respectively. These matrices are organized such that X(i,j) is the expression of
gene j in sample i. xk is the group mean (1 xg) for class k, while x is the mean for all the data.
It can be proved that the separation between pre-defined groups in a reduced dimensional space
is maximized when the space is defined by the eigenvectors of the matrix W-1B [30].
Mathematically, the eigenvalue decomposition of the matrix is given by:
W-'BL = LA
(3-3)
The eigenvector matrix (L) defines the dimensions of the reduced space. Each column of
L defines an axis or Discriminant Function (DF) of the FDA space. The diagonal entries of the
eigenvalue matrix (A) represents the discriminant powers of each corresponding DF. The entries
in L contain the discriminant weight for each gene. The discriminant weight determines the
18
contribution of each gene in defining the DF. Finally, the projections of the individual samples
onto each DF, or the discriminant score, is calculated by:
xiL
Yj=xLj=
(3-4)
i=l
where yj is the discriminant score of the actual sample x on the jth DF. In our analysis, we chose
Wilks' lambda defined as the ratio of the determinant of the between-group variance matrix W to
the determinant of the total variance matrix T for each gene to obtain an initial set of
discriminatory genes. Wilks' lambda can be transformed into an F-distribution, which allows the
selection of discriminatory genes with an appropriate confidence level [30]. These selected
genes were ranked by their F value, and the 30 most discriminating genes were chosen for each
case.
The groups were defined to be those cells grown in identical media conditions, therefore,
a total of seven groups corresponding to full BG 11, 10% Nitrogen (1 ON), 10% Phosphate (1OP),
0.3% Nitrogen + Acetate (0.3NA), 10% Nitrogen + acetate (IONA), 10% Phosphate + Acetate
(1OPA), and BG 11 + Acetate (BGA). A total of 26 arrays were run including parallel flasks and
replicates of the same RNA.
Figure demonstrates that Fisher Discriminant Analysis allows good separation of cell
states corresponding to different nutrient limitations. CV1 and CV2 represent linear weighted
combinations of the 30 genes that were identified to alter most consistently over the eight growth
conditions evaluated. Cells in which PHB had accumulated to the highest level, 10% Phosphate
+ Acetate, are separated by negative values of CV1 and high values of CV2. Similarly, cells
limited in phosphate but not grown in the presence of acetate have negative CVI values but CV2
values lower than 10
OPA grown cells. Each individual point on these figures represents a unique
culture grown under the conditions indicated.
19
·
20
1OP/
15
1ON
10
BGA
10NA
5
>
(
0
-
-5
I
-10
-15
-20
-25
I
I
RBGll
I=
I___ _ ___I
i
-20
-10
0
10
20
I
30
CV1
Figure 3-1. Discrimination of cells grown to late stationary phase in full media (BG11);
media containing 10% phosphate levels (10P); media containing 10% nitrogen levels (10N);
or to late exponential phase with 0.3% nitrogen levels (0.3N). Medium conditions containing
an additional 10 mM sodium acetate are designated (A). Each individual point on the figure
represents a unique culture grown under the conditions indicated.
3.3 Genes Predictive of PHB Accumulation
Figure indicates that CV2 has a statistically significant correlation with the PHB content
of the cells. Table 3-2 shows a selection of the top 30 genes associated with CV2. These include
genes such as cheY, a chemotaxis protein that would not be expected to cause increased PHB
synthesis when over-expressed, but also transcriptional regulators such as hypF that would be
implicated in global cellular stress responses, one of which would be increased PHB synthesis.
Importantly, it can be seen that a specific combination of just thirty transcript accumulation
levels can be used to accurately forecast PHB accumulation levels.
20
5
4
3
e
2
+
l
RL· =
~~~~~~6
0·57
I
0
-1
-10
-20
0
10
20
CV2
Figure 3-2. Correlation of CV2 against the PHB content of the cells (% dry cell weight)
Gene
Unique ID
Function-Gene Category
None
s110008
None
None
s10010
None
cheY
s110039
Chemotaxis protein Y
hypF
s110322
Transcriptional regulator
None
s110361
None
proA
s110373
Amino Acid Biosynthesis
None
s110374
Branched chain AA transporter, braG, livF, livG
None
s110379
Cell envelope, surface polysaccharides
None
s110385
Transport and binding proteins
None
s110396
Regulatory components of sensory transduction
uvrB
s110459
DNA modification and repair, cell stress
prsA
s110469
Ribose-phosphate pyrophosphokinase
None
s110477
j,,~~~~~~~~
.
.....
None
....................
21
None
s110486
None
None
s110550
Hypothetical flavoprotein
None
s110558
None
None
s110703
None
napC
s110873
Carboxynorspermidine
petF
sil 1317
Apocytochrome f, Photosynthesis
None
s11l1376
None
None
s11l
1473
None
None
s111504
None
hspl7
s111514
Heat shock chaperone, Cell stress
None
s111611
None
None
s111623
Transport and binding proteins
None
s111l
630
None
None
s111632
None
None
sil 1702
Hypothetical
truA
sll 11820
Pseudouridine synthase 1, Translation
nth
slrl 822
Endonuclease 11, DNA repair and modification
decarboxylase
Table 3-2. Top 30 discriminatory genes: gene name, family, proposed role for PHB
predictive genes
3.4 Changes in Phosphate and Nitrogen Related Genes.
Synechocystis contains a number of multi-gene families related to phosphate and
nitrogen use, and the change in transcription levels of these specific genes were examined.
Two
different multi-gene families of phosphate transport genes pstABCS are present. The first family
is composed of s110680 (pstS), s110681 (pstC), s110682 (pstA), s110683 (pstB) and s110684 (pstB).
The second family includes slr1247 (pstS), slr1248 (pstC), slr1249 (pstA) and slrl250 (pstB). In
this study only the second family (slrl247-slr1250)
appeared to accumulate preferentially in
phosphate limited conditions.
Figure displays the values for the second family slr1247-slr1250.
These results indicate
that the s110680-0684 phosphate transport system is not active under phosphate limitation in early
to mid stationary
phase.
The phosphate
starvation inducible protein phoH significantly
accumulated in the phosphate limited condition when compared to the other conditions studied.
22
However, the other members of the pho regulon phoU, phoP, and phoA did not demonstrate any
clear trend across the conditions studied.
Interestingly, the phosphotransacetylase
gene pta did show clear accumulation in the
phosphate limited conditions. Phosphotransacetylase has been reported to be involved in the
activation
of the PHA synthase
gene, and this result suggests at least some level of
transcriptional control of this gene. The fact that the highest levels of PHB were observed in
phosphate limited conditions provides further evidence that transcriptional control plays an
important role in PHB accumulation in Synechocystis.
4fl
I ..
1
c
8
0
PA
i NA
zre
EP
z
iN
04
BGA
E
BG
0.2
0
phoP
phoH
pta
pstS
pstC
pstB
pstA
Figure 3-3. Expression levels of phosphate-related genes phoP, plhoH, pta and the phosphate
transport family slr1247-1250 (pstABCS).
Of the thirteen nitrogen-related genes examined only nrtA, nrtB and ntcB had any clear
discriminatory power for nitrogen limited conditions. Figure demonstrates that even these genes
were only significantly altered in the 10% NA conditions but not in the 10% N conditions.
Similar to the phosphate transport systems, there are two nitrogen transport systems nrtABCD
located at either s111450-1453or slr0040-0044. Neither of these families showed any clear
23
accumulation specifically in nitrogen limited conditions.
NarL (nitrate/nitrite response
regulator), narB (nitrate reductase), ntcA (global nitrogen regulator) and ntcB (transcriptional
activator for ntcA) genes also did not show any differential accumulation.
These results suggest
that the majority of control of the nitrogen response is not occurring at the transcriptional level or
that the conditions studied were not sufficiently limiting to initiate regulatory programs at the
gene level.
.
I
1.4
1.2
1
z
PA
B NA
0.8
08
EnP
raN
EO-
; BGA
BG
0.4
0.2
nrtA
nrtB
nrtC
nrtD
ntcA
ntcB
narL
narB
Figure 3-4. Expression levels of nitrogen-related genes nrtA, nrtB, nrtC, nrtD, ntcA, ntcB,
narL and narB.
3.5 Changes in PHB-Related Genes
Figure depicts changes in transcript levels of PHB biosynthesis-related genes. The
phaAB (slrl993 and slrl994) genes code for the PHA specific -ketothiolase and acetoacetylCoA reductase involved in the first steps of the PHA biosynthetic pathway. The phaEC gene
products comprise the PHA synthase which catalyzes the polymerization of hydroxybutyryl-CoA
to form PHB. Transcripts from these genes accumulated preferentially in the conditions of
24
maximum PHB accumulation, 10% PA and 10% P, suggested at least some level of control at the
transcriptional level.
These results also suggest that over-expressing the native PHB
biosynthesis genes constitutively may be expected to result in increased PHB levels.
A
11
I.L
1
PA
0.8
B NA
BP
E 0.6
oN
z
] BGA
0.4
BG
0.2
0
phaA
phaB
phaE
phaC
PHB (% cdw)
Figure 3-5. Expression levels of PHB biosynthesis-related genes and PHB accumulation
levels. PhaA (P-ketothiolase), phaB (acetoacetyl-CoA reductase), phaEC (PHB synthase).
3.6 Summary and Conclusions
PHB accumulation differences for wild-type Synechocystis were monitored by fullgenome transcriptional profiling. Canonical discriminant analysis was used to identify transcript
levels that allowed discrimination between growth media conditions, and allowed predictions of
PHB levels. Phosphate-related genes were found to alter in response to phosphate limitation and
were found to include differentially regulated multi-gene families. Nitrogen-related genes were
not found to be substantially reflective of nitrogen limitation under the conditions studied.
Finally, PHA biosynthetic pathway genes were found to reflect the media conditions in which
25
PHB had most substantially accumulated, suggesting that constitutive over-expression of the
PHA biosynthetic genes may lead to greater PHB accumulation levels.
26
4. Characterization of Late Exponential Phase Synechocystis
Metabolism using C14 Tracers
4.1 Design of C0 2 -Labeling System
Flux analysis in cyanobacteria is less straightforward than for heterotrophic bacteria for a
number of reasons. Fewer studies have been conducted on cyanobacteria, intracellular storage
compounds are the major carbon sinks (metabolites are not secreted into the medium in
significant quantities), and the gas phase being the source of substrate carbon complicates tracer
delivery.
Though accurate flux analyses have been performed for growth on 13C-labeled
glucose, the CO2 uptake rate in these mixotrophic cultures was observed to be only twice that of
the glucose uptake [31], indicating that the observed flux distributions were likely to be
significantly different than for a purely autotrophic growth mode.
Methods have been developed using 4 C-bicarbonate for determining the CO2 uptake rate
and partitioning of the fixed carbon into different biosynthetic fractions for purely autotrophic
cultures.
Performing labeling studies with 14C-bicarbonate requires a sufficiently high
bicarbonate specific activity and long enough incubation times to enable sufficient counts to
enter each of the cell fractions to be assayed. The system must also be designed such that the
total inorganic carbon levels presented to the cells do not change significantly during the
incubations, so that the physiology of the cells does not change.
Table 4-1 indicates the results of equilibrium calculations for the total inorganic carbon
species contents of the gas and liquid phases for a shake flask containing 500 ml of BG 1 media
and 500 ml of headspace, for incubation with air and with 5% CO2 . The inorganic carbon pool
sizes in the gas and liquid phases for growth in air are such that the cells would rapidly deplete
the available CO2 upon sealing the flasks from the atmosphere following addition of the
radiolabel. Upon growth at 5% CO2,the inorganic carbon pool sizes are such that only -10% of
the total available carbon would be consumed during a 2 hour incubation. It was decided to
incubate each flask with 0.45 mCi 14 C-bicarbonate, since uptake of 10% of the label by a 500 ml
culture would result in at least 2x106 dpm per 50 ml culture aliquot (sufficient for detection in
cell fractionation assays).
27
Air in Headspace
Liquid
Gas
Uptake rates
- 8x10' 5 moles CO2
6
- 7xl 0-moles
CO 2
4x 10-6 moles CO2 min 1
5% CO2 in Headspace
Liquid
Gas
Uptake rates
Table 4-1.
-
x10'2 moles CO2
110
x- '3 moles CO2
7xl0 -6 moles C02 min-1
Liquid and gas equilibrium inorganic carbon pool sizes for a shake flask
containing 500 ml liquid media and 500 ml headspace. Uptake rates are measured values for
500 ml of late exponential phase culture.
The molar rate of formation Ji of an intracellular compound Mi synthesized from 14 Cbicarbonate taken up by the cells can then be calculated by assuming rapid equilibration of
labeled intracellular metabolite pools, and that the specific activity of the precursor metabolites
leading to Mi are the same as that of the bicarbonate in the medium.
The synthesis rates Ji can
then be calculated by measuring the increase in counts in Mi:
1
dM
Ji= sin---dt
i
i
(4.1)
where ni is the number of carbon atoms in Mi, and si is the specific activity of labeled
bicarbonate in the system (dpm/mole).
4.2 Labeling of Wild Type Late Exponential Phase Cells at 5% CO2
A labeling study of the wild type strain in late exponential phase was performed under a
5% CO 2 incubation system, under conditions of constant 14 C-bicarbonate tracer specific activity.
Figure depicts the experimental setup. A 'test' set of flasks were pre-equilibrated with 5% CO2
and spiked with 0.45 mCi
14C-bicarbonate.
One set of flasks was reserved for measuring TIC
(total inorganic carbon levels) and no 14C-bicarbonate was added to these. Cells grown to late
28
exponential phase under 5% C02 in a second set of flasks were harvested by centrifugation and
transferred to the 'test' flasks. These were allowed to sit for incubation times from 20 min to 2
hr, and fractionation assays performed as described in chapter 2. Figure demonstrates that a
linear uptake of
14 C-bicarbonate was
to ensure a constant
14C
observed, in accordance with the design of the experiment
The percentage of counts taken up
inorganic carbon specific activity.
during the incubation time was also small (-4% of the total counts), indicating that the cells did
not significantly deplete the inorganic carbon levels available to them during the incubations, as
per the experimental design.
Cap flasks, pre-equilibrate media with 5% CO2 , spike with C14 if required
Il-1
,l
GI71
71ir71
to appropriate
point
i iEi-WT1 Ia
ii growth
i iI'l
curve) in second set of
centrifugation, resusl end in pre-equilibrated C14-sp
for varying times
R -1 ri 1 r -1 i 11 F 1
2 i II
/
i
/
20 min
a
i
/mm
-
20 min
40 min
40min
lhr
a
hr
1.5 hr
2hr
Oj.3
1.5 hr
2 hr
TIC
Figure 4-1. Experimental setup to ensure labeling at constant 14 C-bicarbonate specific
activity under 5% CO2 for the WT strain. TIC represents Total Inorganic Carbon.
29
i
·
*
I 141
gIls
i
l*
0
-E
1tie imcbd wlmC4l (min)
Figure 4-2.
ohe
d wb C14(ango)
Counts taken up by the WT late exponential phase cells during the labeling
experiment, depicted either as a percentage of the total initially available counts or as the
total counts taken up by a 25 ml assay sample.
4.3 Results of Fractionation Assays
Fractionations for PHB, glycogen, protein, lipid, RNA and DNA were performed on the
cell pellets to determine the rates of synthesis of these different cellular carbon sinks. These
counts were also observed to be linear, indicating that the intracellular metabolite pool
equilibrated with the external labeled bicarbonate on a timescale shorter than 20 minutes (the
shortest incubation). The specific activity measured at the end of the experiment corresponded
to within experimental error to that at the beginning of the experiment, indicating that dilution of
the bicarbonate label by intracellular unlabeled carbon species was negligible.
Table 4-2 gives
calculated CO2 uptake rates, and molar production rates and total quantities of glycogen and
PHB.
C02 uptake rate
6.0x10 -8 - 2x10 -9 mole/min
PHB production rate
3.3x10-" ± 6x10-'2 mole monomer/min
PHB content
1.8x10'8 ±t 1.6x10-'o moles monomer
Glycogen production rate
7.3xl0-1' ± 7x10'-12moles monomer/min
Glycogen content
1.4x 10- 7 ± 2x 10-8 moles monomer
Table 4-2. Values for CO2 uptake rate, PHB & glycogen production rates (on a monomer
basis), PHB & glycogen contents for AM5 on a per 10ml culture basis.
30
The measured CO2 uptake rates are consistent with the observed cell growth rates. The
cultures were assayed at a later point along the growth curve than intended, and hence the
percentage of total counts taken up by the cells was lower than intended (4% vs 10%). The
fractionation assays had to be modified to increase the detection limits to account for this.
Figure depicts the best results for the fractionation of counts that was observed. An attempt to
further evaluate the consistency of the derived production/consumption rates was performed
using a model of Synechocystis metabolism described below.
rin
14U
120
= 100
[] DNA
0
X
x
=
[ RNA
80
PHB
$
oo
o Lipid
60
SME
arGlycogen
60
i 'Protein'
40
20
0
20
40
60
90
120
incubation time (min)
Figure 4-3. The best conservation of counts observed in each of the cellular fractions versus
the total incorporated into cells. 'SME' refers to 'small molecule extract'. The SME
extraction for these results was performed using methanol. Comparison to extractions
performed with perchloric acid indicate that a large proportion of the SME counts here are
chlorophyll. This has not been confirmed by chromatography.
4.3 Formulation of Stoichiometric Model for Synechocystis Metabolism
A stoichiometric model of Synechocystis metabolism was assembled using the KEGG
database.
Synechocystis PCC6803 and cyanobacteria in general were thought to have an
31
incomplete
tricarboxylic
acid cycle due to the absence
of a detectable
ca-ketoglutarate
dehydrogenase enzyme activity. This prevents the formation of succinate from ct-ketoglutarate
through the action of succinate dehydrogenase, and succinate therefore was presumed to be
synthesized either through a fumarate reductase activity or by reversal of the succinate
dehydrogenase. A model based on this assumption faced the problem of producing reduced
FAD, which is required as a cofactor in this reaction.
In investigating these possibilities, new data supporting the completion of the TCA cycle
were uncovered.
The ORFs s111625 and s110823 were identified as having significant sequence
homology to the FeS subunits of succinate dehydrogenase/fumurate reductase [32]. A double
knockout mutant resulted in a strain that had 17% of the wild-type succinate levels, but only 12% of the wild-type fumarate levels. This indicated that the succinate dehydrogenase was acting
to convert succinate to fumarate - the opposite direction to that expected if the TCA cycle was
incomplete. Addition of a-ketoglutarate to the medium of this mutant resulted in succinate
accumulation.
ca-ketoglutarate could therefore be converted into succinate in vivo, despite the
lack of a traditional a-ketoglutarate dehydrogenase.
Appendix A contains a list of the reactions assumed in this model. The model has as
inputs: S0 4 2-; Mg2+; N0 3 -; P0 3 4-; C0 2; 02; and as outputs: glycogen; PHB; the RNA bases ATP,
GTP, UTP, CTP; the DNA bases dATP, dGTP, dTTP, dCTP; the lipid classes MGDG
(monogalactosyl diglyceride), DGDG (digalactosyl diglyceride), SQDG (sulphoquinovosyl
diglyceride),
PG (phosphatidyl
glycerol); the acyl classes [16:0], [16:1], [18:0] and [18:1];
chlorophyll; and the amino acids serine, aspartate, glutamate, glutamine, arginine, glycine,
threonine, cysteine, methionine, tyrosine, phenylalanine, tryptophan, lysine, leucine, valine,
isoleucine, histidine, proline, asparagine and alanine.
The network was represented in the in-
house software BioSystAnse, and the Metran software [33] was used to derive the independent
pathways listed in Appendix B.
4.4 Reconciliationof ExperimentalData against StoichiometricModel and Detection of Gross
MeasurementErrors
The experimental
data for rate of label incorporation into DNA, RNA, PHB, lipids,
glycogen and proteins was converted into molar rates of production of the model input and
output variables through the following assumptions.
32
DNA and RNA compositions were
assumed to be 29.5% dGTP/GTP, 23.3% dATP/ATP, 24.8% dCTP/CTP and 22.3% dTTP/UTP
[34]. Total counts in the lipid and SME fractions were distributed as 24% chlorophyll, 30%
MGDG, 12% DGDG, 11% SQDG and 9% PG [35]. Fatty acid compositions were assumed to be
47% [16:0], 38.8% [16:1], 1.4% [18:0] and 10% [18:1].
Counts in the protein fraction were
distributed assuming an average protein composition 8.3% Ala, 5.7% Arg, 4.4% Asn, 5.3% Asp,
1.7% Cys, 6.2% Glu, 4.0% Gln, 7.2% Gly, 2.2% His, 5.2% Ile, 9.0% Leu, 5.7% Lys, 2.4% Met,
3.9% Phe, 5.1% Pro, 6.9% Ser, 5.8% Thr, 1.3% Trp, 3.2% Tyr and 6.6% Val. This collection of
42 derivative measurements was then examined for consistency against the 38 derived network
constraints listed in Appendix B. Solving this overdetermined system of equations gave values
for the rates of the independent pathways that resulted in large residuals, and therefore the
experimental measurements were examined for gross errors by the method of measurement
elimination [36].
Appendix C depicts both the initial measurements and those recalculated from the
overdetermined equation system after withholding the oxygen data. The oxygen evolution and
uptake measurements were determined to be an order of magnitude too great compared to the 14C
measurements. The oxygen measurements were also an order of magnitude greater than those
observed in old fermentation data. The likely explanation is that when cells are taken out of the
flasks and put into the Clarke oxygen electrode setup, the increased light flux causes greater
oxygen evolution than in the flasks. The respiration data were obtained by covering the Clarke
electrode incubation cell with foil, which would not have excluded all the light.
4.5 Summaryand RemainingIssuesfor ImprovingFlux Quantitation
A methodology using C14 tracers was developed for the accurate measurement of CO2
uptake rates and the partitioning of the fixed carbon into different biosynthetic fractions. These
techniques were applied to the characterization of WT Synechocystis cells in late exponential
phase. A stoichiometric model of Synechocystis metabolism was used to determine constraints
between the measurements. A balance on C14 counts was obtained and significant levels of
secreted compounds were not detected. The measured carbon fixation rates were found to be
consistent with the observed growth rates, but inconsistent with measurements of oxygen
evolution in the light and uptake in the dark made using a Clarke Electrode apparatus.
33
The most outstanding issue for completing the flux measurements is generating reliable
oxygen evolution and uptake data, to act as a check on the NADPH balance of the cells.
Respiration rates (oxygen uptake in the dark) can most likely be performed in the Clarke oxygen
electrode provided this is maintained in a completely dark environment (such as in a cabinet
drawer or an opaque box). Oxygen evolution rates in the light can be performed in situ in a flask
through inserting the electrode probe into the liquid medium. Upon sealing the flask, net oxygen
production would result in an increase in the liquid and gas phase oxygen concentrations. The
total amount of oxygen production could be calculated from the Henry's law constant for oxygen
and the volumes of liquid and gas in the sealed flask (determined by weight). The approximate
expected change in 02 concentrations that would have to be detected over the 2 hr period would
be on the order of 16.0 % to 16.4 %. Polestar technologies (Cambridge, MA) has developed a
non-invasive fluorometric oxygen probe that may be suitable for this purpose.
A large number of counts observed in the fractionation are in the 'Small Molecule
Extract' pool. Comparison of data between methanol-extracted and perchloric acid-extracted
small molecule extracts indicates that a large proportion of this pool is chlorophyll. This could
be confirmed by HPLC [37].
A relatively minor consideration is measurement of the specific activity of the system.
The measured specific activity is determined by dividing the total inorganic carbon counts
present in a given volume of medium, by the total inorganic carbon species present in the same
volume of sample as determined using a Shimadzu TIC analyzer (courtesy of the Gschwend lab).
Due to the constraints in not being able to run radioactive samples through this machine, the
inorganic carbon concentrations were determined by performing the same procedures described
above with flasks containing no C14 label. Evaluation of the standard deviations for both
measurements leads to a specific activity for the experiment of 7.2x1010°
1.8x109 dpm/mole
carbon. The measured value for the inorganic carbon concentrations is in agreement with
theoretical calculations for carbonate equilibria under 5% CO2. While this does not preclude the
existence of a systematic measurement error through subtle differences between the handling of
samples from radioactive and non-radioactive flasks, the standard error of the specific activity
appears to be acceptable.
34
5. Future Research Directions
5.1 Characterizationof Wild Type CarbonFlux Partitioning
Figure depicts the increase in glycogen contents and decrease in CO2 uptake rates upon
the transition from exponential phase to stationary phase, consistent with a transition from a
regime of growth to a regime of storing fixed carbon for future use. The bicarbonate labeling
techniques described previously can be used to characterize the change in partitioning of fixed
carbon from growth towards the storage compounds glycogen and PHB, and to demonstrate the
ability to measure C0 2-fixation rates over a wide range of values using '4 C-bicarbonate
techniques.
.-
1:-
.
C.
.
.
30
·
l:
l·
m !2
Tite
·
TiO1
l
so
h)
6
ZZ
0C ,
S02
0 _
Time
h)
tire
h'
Figure 5-1. Growth curve, glycogen content (% cdw) and CO2 uptake rates (g/g cdw/hr) for
air-grown WT Synechocystis culture.
35
5.2 Investigationof the Role ofAcetate in StimulatingPHB Production
Figure demonstrates that stationary phase cells grown in the presence of 10 mM sodium
acetate accumulate higher levels of PHB than cells grown without acetate.
Studies in
Synechococcus [38] have supported the notion that acetyl-phosphate plays a role in activating the
PHB synthase under nitrogen starvation conditions. Two potential roles for the action of acetate
then are either in providing acetyl-CoA for PHB synthesis, or in serving as a trigger for PHB
synthesis through raising the levels of intracellular acetyl-phosphate.
5.0
4.5
4,0
35
D 3.0
T
025
2.5
O.
X
5
1 .0
0.5
+-
+
0.0
gBO
BO.+A
Or
10%N
10%N
A
10%P
10%P +A
Figure 5-2. Role of acetate in enhancing PHB accumulation in Synechocystis. BG represents
BG11 cyanobacterial media. A represents 10 mM acetate. 10% N represents media with
10% of the nitrogen content of BG1 1. 10% P represents media with 10% of the phosphate
content of BG11.
The current developed measurement techniques may be used to quantify the contribution
of exogenous acetate to stationary phase PHB accumulation. We postulate that acetate serves as
the major carbon source for stationary phase PHB accumulation.
Previous cultivation
experiments have shown that Synechocystis continues to fix CO2 in the stationary phase, and that
36
the cell dry weight increases continuously as a result of glycogen accumulation. We postulate
that CO2 serves as the major carbon source for glycogen accumulation. This may be evaluated
by culturing cells in the presence of 14 C-labeled acetate, and measuring the specific activity of
both the glycogen and PHB.
5.3 Investigation of the PHB remobilization pathway
The rates of P(3HB) synthesis during the presence of a carbon source and degradation
during carbon starvation have been measured in Alcaligenes eutrophus during growth on fructose
and butyric acid [39], and have demonstrated simultaneous PHB synthesis and remobilization
during the stationary phase. Figure represents a model for the PHB remobilization pathway
based on analogies to other eubacteria.
Pyruvate
Acetyi-CoA
[ct
dEpta
Acetyl
Acetoacety-C
Ite
(R)--
Acdetacetate
......-
|i, (-Ncon ) .3-h.r
PHB
Idepolymerase
j
Figure 5-3. Model for the PHB remobilization
yre
Pe)--HydrzYebutyrate
pathway based on studies in Alcaligenes
eutrophus and other eubacteria .
37
Intracellular PHB depolymerases that have been identified to date do not share significant
homology with each other or to extracellular depolymerases. Candidate ORFs homologous to an
extracellular depolymerase from Burkholderia cepacia [40] (slr0314, s111129, slr932
slr1916), an oligomer hydrolase from Acidovorax
and
sp. SA1 [41] (slr1916 and slr1235), 3-
hydroxybutyrate dehydrogenase (slrO315, slr208 and slr2124) and acetoacetyl-CoA synthetase
(s110542)have been identified. Mutants deleted for these ORFs have been produced, and may be
tested for remobilization of accumulated PHB.
5.4 Directed mutants to increase thefixation rate of CO2
Tobacco plants transformed with a cyanobacterial fructose-1,6-/sedoheptulose-1,7bisphosphatase were larger and grew at a faster rate than wild-type plants [42]. This enzyme has
also been found to have significant flux control over photosynthetic CO 2 fixation in modeling
studies of photosynthesis in sunflowers and other plants.
Phosphoenolpyruvate carboxylase is a key anapleurotic enzyme that has been shown to
increase the growth rate of Corynebacterium glutamicum [43].
We postulate that the
overexpression of either of these genes individually or in concert will increase CO2 fixation
rates.
5.5 Directed mutantsto increasetheproductionrate of PHB
A deletion mutant for glycogen pyrophosphorylase, the first step in the committed
synthesis of glycogen, has been shown to accumulate PHB levels of 20% cdw [44]. We
postulate that over expressing the entire PHB synthesis operon in this background strain will
further increase PHB contents.
38
References
1.
Stanier, R.Y. and G. Cohen-Bazire, Phototrophic prokaryotes: the cyanobacteria. Annu.
2.
Rev. Microbiol., 1977. 31: p. 225-274.
Rippka, R., et al., Journal of General Microbiology, 1979. 110(2).
3.
Vermaas, W.F.J., Molecular-biological approaches to analyze photosystem II structure
andfunction. Annu. Rev. Plant Physiol. Plant Mol. Biol., 1993. 44: p. 457-481.
4.
Williams, J.G.K., Construction of specific mutations in photosystem II photosynthetic
reaction center by genetic engineering methods in Synechocystis 6803. Meth. Enzymol.,
5.
Kaneko, T., et al., Sequence analysis of the genome of the unicellular cyanobacterium
Synechocystis sp. strain PCC6803. 2. Sequence determination of the entire genome and
1988. 167: p. 766-778.
assignment ofpotentialprotein-coding
6.
7.
regions. DNA Res., 1996. 3: p. 109-136.
Pelroy, R.A. and J.A. Bassham, Photosynthetic and dark carbon metabolism in
unicellular blue-green algae. Arch. Mikrobiol., 1972. 86: p. 25-38.
Tabita, F.R., Carbon dioxide fixation and its regulation in cyanobacteria, in The
Cyanobacteria, P. Fay and C. van Baalen, Editors. 1987, Elsevier: Amsterdam. p. 95117.
8.
9.
Badger, M.R. and M.H. Spalding, Ch. 16 CO2 Acquisition, Concentration and Fixation
in Cyanobacteria and Algae, in Photosynthesis: Physiology and Metabolism, R.C.
Leegood, T.D. Sharkey, and S. von Caemmerer, Editors. 2000, Kluwer Academic
Publishers.
Poirier, Y., C. Nawrath, and C. Somerville, Production of Polyhydroxyalkanoates, a
Family of Biodegradable Plastics and Elastomers, in Bacteria and Plants.
Biotechnology, 1995. 13.
10.
11.
12.
13.
14.
Rose, A.H. and D.W. Tempest, Advances in Microbial Physiology. Vol. 10. 1973:
Academic Press.
Lee, S.Y., J. Choi, and H.H. Wong, Recent advances in polyhydroxyalkanoate production
by bacterial fermentation: mini-review. International Journal of Biological
Macromolecules, 1999. 25: p. 31-36.
Jendrossek, D., A. Schirmer, and H.G. Schlegel, Biodegradation ofpolyhydroxyalkanoic
acids. Appl. Microbiol. Biotechnol., 1996. 46: p. 451-463.
Choi, J. and S.Y. Lee, Factors affecting the economics of polyhydroxyalkanoate
production by bacterialfermentation. Appl. Microbiol. Biotechnol., 1999. 51: p. 13-21.
Gerngross, T.U., Can biotechnology move us toward a sustainable society? Nature
Biotechnology, 1999. 17.
15.
Valentin, H.E., et al., PHA production, from bacteria to plants. International Journal of
Biological Macromolecules,
16.
Asada,
1999. 25: p. 303-306.
Y., et al., Photosynthetic accumulation of poly-(hydroxybutyrate) by
cyanobacteria- the metabolismandpotentialfor C02 recycling.InternationalJournalof
Biological Macromolecules,
17.
1999. 25: p. 37-42.
Shi, H., J. Nikawa, and K. Shimizu, Effect of Modifing Metabolic Network on Poly-3Hydroxybutyrate Biosynthesis in Recombinant Escherichia coli. Journal of Bioscience
and Bioengineering, 1999. 87(5): p. 666-677.
39
18.
Leaf, T.A. and F. Srienc, Metabolic Modeling of Polyhydroxybutyrate
19.
Biotechnology and Bioengineering, 1997. 57(5): p. 557-570.
Steinbiichel, A. and H.G. Schlegel, Physiology and molecular genetics of poly(flhydroxy-alkanoic acid) synthesis in Alcaligenes eutrophus. Molecular Microbiology,
20.
1991. 5(3): p. 535-542.
Oeding, V. and H.G. Schlegel, fl-Ketothiolasefrom
Biosynthesis.
Hydrogenomonas eutropha H16 and
its significance in the regulation of poly-J-hydroxybutyrate metabolism. Biochem. J.,
21.
22.
23.
1973. 134: p. 239-248.
Senior, P.J. and E.A. Dawes, The regulation of poly-fl-hydroxybutyrate
Azotobacter beijerinckii. Biochem. J., 1973. 134: p. 225-238.
metabolism in
Lee, I.Y., et al., Regulation of poly-/-hydroxybutyrate biosynthesis by nicotinamide
nucleotide in Alcaligenes eutrophus. FEMS Microbiol. Lett., 1995. 131: p. 35-39.
Mansfield, D.A., A.J. Anderson, and L.A. Naylor, Regulation of PHB metabolism in
Alcaligenes eutrophus. Can. J. Microbiol., 1995. 41: p. 44-49.
24.
Gerngross, T.U., et al., Overexpression and purification of the soluble
polyhydroxyalkanoate synthase from Alcaligenes eutrophus: Evidence for a Required
Posttranslational Modification for Catalytic Activity. Biochemistry, 1994. 33: p. 93119320.
25.
Hein, S., H. Tran, and A. SteinbUichel, Synechocystis
sp. PCC6803 possesses a two-
component polyhydroxyalkanoic acid synthase similar to that of anoxygenic purple sulfur
bacteria. Arch. Microbiol., 1998. 170: p. 162-170.
26.
Rehm, B.H.A. and A. Steinbfichel, Biochemical and genetic analysis of PHA synthases
and other proteins required for PHA synthesis. International Journal of Biological
Macromolecules, 1999. 25: p. 3-19.
27.
Mohamed, A. and C. Jansson, Influence of Light on Accumulation of PhotosynthesisSpecific Transcripts in the Cyanobacterium Synechocystis-6803. Plant Molecular
Biology, 1989. 13(6): p. 693-700.
28.
Daniels, L., R.S. Hanson, and J.A. Phillips, Chapter 22: Chemical Analysis, in Methods
for General and Molecular Bacteriology, P. Gerhardt, et al., Editors. 1994, American
29.
Society for Microbiology: Washington, D.C. p. 512-554.
Karr, D., J. Waters, and D. Emerich, Analysis of Poly-Beta-Hydroxybutyrate
30.
Rhizobium-Japonicum bacteriods by ion-exclusion high-pressure liquid-chromatography
and UVdetection. Applied and Environmental Microbiology, 1983. 46(6): p. 1339-1344.
Dillon, W.R., and Goldstein, M., Multivariate Analysis: Methods and Applications. 1984:
in
John Wiley and Sons. 587.
31.
Yang, C., Q. Hua, and K. Shimizu, Metabolic Flux Analysis in Synechocystis Using
Isotope Distribution from 13C-Labeled Glucose. Metabolic Engineering, 2002. 4(3): p.
202-216.
32.
33.
34.
35.
Cooley, J.W. and W.F.J. Vermaas, Succinate Dehydrogenase and Other Respiratory
Pathways in Thylakoid Membranes of Synechocystis sp. Strain PCC 6803: Capacity
Comparisons and Physiological Function. J. Bacteriol., 2001. 183(14): p. 4251-4258.
Antoniewicz, M., Metran Software Personal Communication. 2003.
Leach, C.K. and M. Herdman, Chapter 9: Structure and Function of Nucleic Acids, in
The Biology of Blue-Green Algae, N.G. Carr and B.A. Whitton, Editors. 1973.
Nichols, B.W., Chapter 7: Lipid Composition and Metabolism, in The Biology of the
Blue-Green Algae, N.G. Carr and B.A. Whitton, Editors. 1973.
40
36.
37.
38.
Stephanopoulos, G., A.A. Aristidou, and J. Nielsen, Metabolic Engineering: Principles
and Methodologies. 1998, San Diego: Academic Press.
Thayer, S.S. and 0. Bj6rkman, Leaf Xanthophyll content and composition in sun and
shade determined by HPLC. Photosynthesis Research, 1990. 23: p. 331-343.
Miyake, M., et al., Control of poly-beta-hydroxybutyrate synthase mediated by acetyl
phosphate in cyanobacteria. J Bacteriol, 1997. 179(16): p. 5009-13.
39.
40.
Kawaguchi, Y. and Y. Doi, Kinetics and Mechanism of Synthesis and Degradation of
Poly(3-hydroxybutyrate) in Alcaligenes eutrophus. Macromolecules, 1992. 25: p. 23242329.
Jaeger, K.E., A. Steinbuchel, and D. Jendrossek, Substrate specificities of bacterial
polyhydroxyalkanoate depolymerases and lipases: bacterial lipases hydrolyze
poly(omega-hydroxyalkanoates).
41.
42.
Appl Environ Microbiol., 1995. 61(8): p. 3113-3118.
Sugiyama, A., et al., Cloning and Sequencing of an Intracellular D(-)-3-Hydroxybutyrate
Oligomer Hydrolase from Acidovorax sp. Strain SAI and Purification of the Enzyme.
Current Microbiology, 2002. 45: p. 123-127.
Miyagawa, Y., M. Tamoi, and S. Shigeoka, Overexpression of a cyanobacterialfructose-
1,6-/sedoheptulose-1,7-bisphosphatase in tobacco enhances photosynthesis and growth.
Nature Biotechnology, 2001. 19: p. 965-969.
43.
44.
Koffas, M.A.G., G.Y. Jung, and G. Stephanopoulos, Engineering metabolism and
product formation in Corynebacterium glutamicum by coordinated gene over-expression.
Metabolic Engineering (submitted), 2002.
Wu, G.F., Z.Y. Shen, and Q.Y. Wu, Modification of carbon partitioning to enhance PHB
production in Synechocystis p. PCC6803. Enzyme and Microbial Technology, 2002.
30(6): p. 710-715.
41
PAGES (S) MISSING FROM ORIGINAL
PAGE 42 MISSING
o
SH
0
a)
0~~~
E
~ a)
-
C~~~~~
a)
EE a)EEE 701
1a E
cow
E
*T
U)
0 :
E
0)
0C
-3
(DEl
cL
a)
I
A
+
++
0O au
c.
C)
Cl E
rJ2
0
.2 +
C)
0
a)
+C
'6
[ .-
= -C
+
II
±0
a
_
+ Co
.
+ +E=
3)
0
;
0
;
o
a)
c
a,
+ -at
=
H-O
n
co
- CO O
0)
(n (
,
0o 0C
++
o
a>
0C
a
a.
A
II
V
v
C)
CX
.5 =
0-
t1O
<0
<+ <0
a
oCF-
<a
V)=
Cl
r-
C)
a,
1
CL
30
6.
+ a~
I-t:
0
C)
ct C's
u
+
+
< w
E
.r,
l4t
-
Cn
Co
LO
o)
rl-
=
(n
--2
IC)
C\1
C)
Ozcv
Z< <
(n
0
U:
)
C)
=n
=F
O
E
,'
C)
Ct
C,
o
v)
cd
s
a)
0
E
V)
a)
C
Ec
.C
C)
.i
(.,
cd
U
In
a)
a,
6
C)
7u
o
C)
a)
I'll
0
o
-t::
00
0,
o0
ci)
.
.c0
z
E
0
w
0
0
0
0)
0
.E
a-
.l
C
V)
a)
r.
EC
=,
Iz
+ Cn
H
2
V
Uo0 u
cn
O
C N
=
S
ct
o
,+,;
CZ
1 +
0
S
4-(
a
-o
,
I +a.+
;^
Ci
±
a.
II<
a-
w
0
-Z
C)
0
II
r
co
a)
0< <
0
A U
III
0)
+ A +tOV-1
fVII a,
0
a
+
aC
CL
a
I
o0
o
C
V
O0
O
_
. 0'A
+
o
t
~- o_,
1a)~
> -- a
a)
.<
.ie
+0
ctE
0< a - -aej
a 6
a.
0O
('.
+
C)
E
E
VA
A 0
,
.
=
· .;_> ~~-~
~~~~~~~~~~~~~~~~~~~~~~~~~~~1~,E
a
(
,1 *
0 0CZ
v
o>
, [r x'oo
ucc1
(ED
t0.
w
Au
+
r.
co
0< . -'
11
II
a)3
od1
;la)~t:'',:~lu
~ ~1°
~-171
)
CL
0
c
v5
3
U
E
E
a)
C0
a) U
0
C
(n
0
a)
In-
a
C)
._
o
O
U)
C)
IU
E
U
.th)
c
oN
c)
C)
c
'In
V
E:
sQ
a)
0
S
ci)
0
v0
0,
u
a)
a.)
x
C
12
a:
.4
au
0
u
N
a)
,F
L.
a
E
E
o0
C
.o
-0
Z
W
10
C.
(
cd
(d
(6
Lq
1-
.
C
6
C
(
0)
(4
o
-0
U)
U
U)
U)
a(Da
I
:o
c
c
E
u)
aw
.c
a)
°:
c
.2
c
0
=
E,
C
EI
C: cn
, U), (D
V00
~ 0 oU3
&0 >n0
U 0o c0 o
> >> oo
ci=
-Co
XUE
0 5-C.))-=
CD,
.
=
o o
;-x
0
.c
C4
a
I
)
, +
A
c
II a
.
+
V
II
O
d
1
0
_ 0
OA~
d
0 ao
O
0
0.
a
6
O
r0.
0, .0
£12
0
£1
o_
C v
Q s
_
C
0
i
7nD
C) 0~L
I
t'-.
(0
(N
(O
o o= o I.
on
co
F,~
cn
o
0u
0
D
2
a)
+
+
+
+
+
+
E
+ E+
o
+C
u)
+
.
o
0
0E
o
V
0d,,
c
'C I!w
+
0~
0.
a t~~~
·-
+
r-. co
ra t co
, oo
oCD
C
=-C\J
r-
"=
o
,=
.~vi5~o o~i~
£
I
'
0n
0
0
ZC
A
C
.
0I
(N
*Ca4
0'l
0
C)
C
=~~~~~~ =,
co
Cl
co)
'C
O)
CN
(t'C
En
O
' In
O
C)
CU
0
0.
0
ac
Crx
°
V£U
Cu2
D
rt
o
t: *a
'0'04-S
0+
C)
s0
0
0
0. 0
.6
(NJ
.I4
0
cn
1
*v
ri
A
+
>11OQ
On
V C3
V
o
+.
a
9cu0!
A
C)
~1~~~~C
-
C-
o,- e
-
a
3
p
o a)Ct~ . E" C)+£e.
..
CU+ ,~0 . I
~312.0C)0%sta
0.<.
P
£1
L:
'
"t00 t )
'
0
0.*
c
EQ
O 0
ci
'I4T
*~4
L0a
mo
0 0
C
qda
a~~
O cn
O
ai a
cC)CU
E
'
O
Cci)
.
1~1~1
Z
2 +
caCi2af
C ==
O~da°U= -O
scn
£12
N
~
0I0
o
E?
._
C.
4
X
oC0
E
..0
C
li
3
*L
CaU
<
c
-
ci)
m
. _
0
+ Ce>
3
cn
o
'C
.c8
+
rq
0
,O
EC
Co
QEr ++
0 . +C
+
ai)
.!
.
V
C
'cEa)C,
+
11 +
E
0
IA0
4.
.0
p
*l
4-
04c
'jO
N
J
Nl- coc 0t LO
CD
I
cn
CU
E
0
C)
C
a-
-E
C
E
L
ID
o
00
O
3
'
O
C
O
Q
E
o
o
.0
a0
0
to
0c
j.
LO
0
EC)
el
6
0)
0
0O
.o
L.
.
I3.
0
°
.00.
4
u
En
00.
£12
.0
00.
t-
?0.0
-a
W
-U
~
C(I
.I'l:
L
(N
O
4
.
C
C
C)
f
L()
LO
'
U)
CU
S
Q)
X
C
O
O
E
r0
C)
(N
N
)
0o c0
012
0.
(0j
S-
0.
-rI1cyci
~
z
C)
0
CU
-~o
>
.CC)
Cu
*
0D
C)
CU
C)
c.C)
=
o
Y
-a
0
E
C.)
C
S..
o
E
C4
.L
CL
E
C)
av
._,=:
>4
-E
0
1
L()
C)
c)ou C)
4
c£C
£
>4
-a
C
U1)I
i;1
0
w
aL)
C
L)
a,'-
a) 0
-
-
o
a,
C-)
>a
cU)I
>
i°
._4°
*.-
+_~
~o
CD
> I'l
+_
0
C
A
7-
>
oCY
a)
0
v:
a)
Uoa)
._
r:+
Cl
C)
C)
Co
0o c
'
X
X
E,
Fg
-
· +
I
O
II
tl
++S
-oa
I0
+
+
V
VO
E" C)
n3
t,,b
II
oU)
ed
C
Cl
~
'
)
O
X
A
LO
-0
~~~~~s
~
._
jj> ' + O'Q
s
-
/1
*
S
eIm
Q.
+,
CY.
11
3
-.
CD
QD
-
0I-
d
"
u)
£2x
C)
0V
a-)o0z
._ (d
a)
>
)
£
Ct
0
U
O
+X
)
A
C)
=O
m
+ °
+
(.)n
_
C_
or
I
C)
o0.n.)
£a
U)-
5-
(t
-
-a
,
v
CS
o0 E +
: ::a
~
0)
oo
d
o
co
O
£
n
)< -enc
o
+
E
L
"+
7
cC)00
0 ~0C)0D
toO
x
;>-
X
V
O
x
+
+
A
COODo
o Q
110+
C)
u
o
V I
1
_
=
C
A w
112 W
CM
C
4
v 0
+
+:
o
-
I-
V
O
0
0
0
0
O(O CCuQ
(3C
l0
1 fl
_ ·,
)
a)
~1888 °o g0
cn
0
-5
-
o
a)
C
o>
aj
co
01
£2
UI
a)
II o
)0:a.
CNl
0)ll
01C
DO
00 0
5-
5-
011
Vto)
i
3iI uI
LIL
mlcT
v)
V)
-o
£2
-a
,
I.
n
C.
Cs
a
C)
aC1)
'd
>1
5-
C).
uC)3
CC
0
0d
LO
Cl
Cl
Vn
)
a
C
Q
u
C)
oC)
0
~
'_
oO
=)
_
s
CZ-1
-ov
£2
M
C)
CC.
E
m
C.
0
ctr
-I
V)
0
C,
C)
v
C.
co
0)
C)
0
C)
C)
£o
;
C
C'
7
Cl)
C.
a
.5,
aC
v
Ie
C'
.
C)
C
O
J-
.
-
C,)
.O
)
-
:
CO
w-
_.3::
,
0)
r
-
O
o)
CO
K
nA
)0L
o=nQ)0
co)
(n
a)
-0
-o CU
U)
0
0
.: .n
CU)
-0
a)
.U)
a_
-I
>5Cy"o+
c
7-
+
0 +
U
zCxl
Ca
0
0
"1
-0
7tII
Cl3
d.) 0-
02C
.f- _
a
cl
+
0
V
U
o0
ct
.cq
C)
-a
0
o C
II
11
A
II
+0
-V
V
CD
-
0)
n Q
O+
I a
0) 0
a
+0
-O
gO
1111
U)
Cl
o1
0
C)0
F0
(N
C)
-
I
o 0 a 0)
= U2 a a
.I
~
-
0 O
0
z
<<
rv
a
a
0)
+
A
+ >
<ao
>
o+
m
'r
-
o
0tf
cL
+ 3)0
°a
00CL°
C
4
0~
a
=t
Cl)
C)
0= b~ =
0)0dv
7+ -~
OF < ZO)
<
+a
(N
(N
)
0
0s
.t
(n
+n.
00
r 0) A
V
I
° )
o~~~~~0(D
o~~~~II ° Oa
C,.'0f
o
o3 's4?
,
In
0 0 ,,=:
2
0)
0z
cd
0)
w
=m
J FOZcO
Z<
zz nn
,U Z <
'T
.o
C)+
W
0
N-
0)
N-
n0
-+ v
Q 0.V1
0
Oa
0
w
-E
U)
(I) EF
+
C + I)
o
E
U)
_0 . LO
(U
av
+
Cl
0
10 U "
o0
0
+0_
O CL
0
+
> w.__
c0.0.)
j
A
cq
0
aD
O,.
-6 Z
0)
-0
cy, +
0
o'-
a)101
> 1
> L? 45 I
20
a
c
0C
+
o 0
+ a
-o
C
S-
7:
0
0
zp
+0) C)
+
Q)
C4 zr
C
cUA 0
V +
0_ v0
O~
->
=0
0
VI
Ct
0)._
A
II
V
A
0)
-C
C)
-,
c)
Cl
A
a )
+
0
U
E C)
C, ±
I
+ a.
_
0
z
0
C) C:
-)
0
C.)
0)
(n
) E _
E
c0-
0)
a E :3
E 0
1o ooP- W-
U-
,
(IE
E)
o
a)
cn
.
ac
C0 )
0
Q
00Q
h
U)
(n
LL O+
ao o
a)
~u
(ne-n
0)
C.'
c~
U
I-
00
a
.)
a
0.
2
a,..
-o
0
C,.'
>.
n
1-0
0
;>1
0
0.'
0
._-
o0
C/
,.
oO.
0a
0
U)
N
~C
av0
T
C4
clco
~.
-
C1
st
.
U0
CO
L6
c .o
L6
-
0
a
5::
0.
-Ca
61
6
~
c5
0
,.'
,c~
O
'A
Sa.'
00
0
C
0:
0
6E
0(N
oO. C,)
o0ad
J=
o
0
a
0
3
U)
0)
0)
0)
0
4 .:
-S
0
0
aJ
"c;
0)
CZ
0
CZ
0
.E,,.,
:
aS;
CZ.
-a
._
CO.
-
c)
.'
.0E
M
0)
-o
_
_0
m
0c~
oIn
a
0)
'l~o
0
In
0,
0
C)
-
0
ava)
_
Cd.'
aaL
0
a_ 0)
a
O
-
0
C-O
0
a:
n
z
~
C,'
-
aO
O
3
a
00
c
6
CY.
6
1
n.
6
6
.-
c
0)
CY)
.
~ ~ ~~~
N-
coN
r-
OD
rN:
c(
o00)
E0 E E
a
"I
oW
0M
o0
0
Ea
cc
(U
(/)
n
a)
E
a,
w
a)
E E
E
a)
a)
0
,,
C-
61
+
a.
.Csw
v
0. -
C-
C
0+
; +
H,,:
+
+
aa+0
+)
C-
<<
0+
a
C)I
._
O<
H D <<
H H( 0 o
+ 4- + + +
O U a~
c
C)
II
+
<<<
o
A
V 11
V
+
a.
c
-.
io
<b e
U
N
+
a.
H
QA
6
< 11
A
a
A
o
V V
ac-C
C)
+ +
o -o
1'
o v
03
CL
a
U
U
C
V) -O
V
C
:D
p_
¢
U
C)e
<
+
+
<2 v
c0 n v
+++
<'
C E
o
i
+
+
H
H
+
+
¢ <a
H
C 'I
-,1
71 1L1
_
_
0)
-
in
0,)
0,CD
U1
0
CN
10
C:
0,LL
E1
I)U )
l 0U
1
1 751T5 75 7
M
C0
=:
J::
Y
i
w
0O
n
O>
(7)
CD4
O ,
C°
1
C)
C)
-
0-,
+ + s
-
H0 da~Ei.U)H
LO
N)
C-
a
0
ti
U C
V)
O -o
.
0
.)
0
-
+
11
< < '*
C)
0
+
+
'a
CL.0
o
0
H H
)
U)W
A
o
Q
Cc- +
a
C
:o
M
0
C
O
Moo
2H
c
V
e
C
a)
E
0C
ED
D
oa
+
A
'Ec -
0 -I 0 a
0 C
a:5 a .,:5 a fElf >-1 Dl
..:_
E
.D
0
-o
a)
(U
U) -o
=3
z1.
a
-O
0 0 0
-o
a)
0
-5
w
a)
EI
oo
EI E E
a)
0
M
3C
D-
>
0
u)
E
E E
E
E
U
) .Ln
'-O
0
.n
0 (U -5
0
U)
a)
a)i
I--Z
Cn
C
r:
C/)
-z
0)
(A
0
0)
-o
:
0
.
0
0)
)
o~. *
0)
.0>.0_
o .0.C.
C
.
-o
C.- ._$O-o
Q
_
0
V
CL
0 .0
~ CL
_
_
Q
O
0)
a
a
C
0)
-m
c
0t
-:
C)
0
0
0
0)
)
o
0)
0)
000
0 *0
U
._
72
'3
U
NNC(N
(O
O
O
N-"T'N-N-T
r .N _
N
N-
(N
CD
N
c(
0
(CO(D
r-
0
.
U)
C-
2C
-0
0E=
U:
bo
5
C)
0aa
Sd
0
C)
Q
.0
a,)
U)
0)
(A
v,
M
C
0
-,
C.
2L
;>1
1.
Q
Uo
't
-
CN
- .
0
c9
-3
c
cN N
0.)
(A
0
.4
r)
0,
0)0
t)
N(N
0)
0
wD
_ . C:5
_ C3
C
0)0)0)
(O
C
0
v
(A
C3
)
Q)
0) 0)
C'4
3
C
NC
N-
o
o
a>
0)
.
-S
0)
.2
O
O
o
co
C
N-
C-
N
st
V
.
C
(A
C:
o 0o
-
V
_ a)-
a
.6
i
c)
V
C
a
co
a,
OC
coil
WO
=
fla,
C (D
>C
(!)I
co
-C
C
4
.U) .=
Cr a-
-D
U0
oa)
.j1c
0000>
a
aO
II
Qjo
o
c C-U1
C~ .D00O CU
> -
00
a
7o o
c
.,) '
7501c
,
O
-)
._.1°1
1 Ca
El
I cu
C1
~1
1
'C-
a
0)
+
_
+
A
Ca.o 0
m
Z
o'0
o
+ ~ 0'-"s
Cu
=._
·
v
En
+ o
cu
0
00.
It+
v+
CN
c=
C-+
C
U+Cu
0'
00 O-a
0.
0
0
0
r-
0)
Zo
) a
Co < 0~
U0)
hC ~~~
o-c5:,
0)
0-a-
'tt
0o
cn
to
= =
0
0
0
ou
o
o
0
n
cd
Uo
YC C
od
V.
cd
Ec
ccE
m
0o
ao
.5
.5
.5
Uo
a
cd
cE
.b
.S
la
0
0
co
O
C
CN
Ci
'.0Eo
0
c
r-0
U)
Cu
=
>1
v)
C
N
Ci
CD
C4
I
Cu
°
U)
0
Cu
-0
-D
U,
I.
0
-C
(0
OCi
c
0SC
.0
E
0
E
Cc
a:
In
rz)
O
.C
u
,-
"c
Cu
6)
(N
t~-
ulil
Uz
O
. II I + +
0 C
E
Cu4
0
CO)
o
v
i+
ON
+o 0
x
-0
+
,Cu
00
II
tw
+
C",
,
~
+~-~.~
+
:
m:
0
cu
CCY)
0
C
II
:c
-
+-
z0+E I +n vo
0
AO
zo0
cn
a)
E
+
A
0
c+
_, c
0u
O
0.
(n
D_
E+
6 J
0.
_
Qo
rK2
ii
CC
'Ll(Cco.-.
1
1 .
~ c j o
2E ) C.Z -: o )E
1
E
E
a co
o+
A
II
-Q
_
~C co
i<z
,
Al
<
C
co
cn
,2
CZ
0.
a,~
a
U1
El
co0a- a) l-~l W'
co
v
+C
).
co
2c
-c I
+
,
VI
ECO
+r) na)
C.).
C.)
C)o C.)
Cu Cu
d
0
aC a)
C c: E-
,
V
U)
0
0n
CJ
Ct
I1-
co
t9
C
0.
N
o
CNoI
L
Cii
U
CI
U)
U) ''
C(i
C*
i
00oo
C
(1
(I
-C
C
(V
0
. T.2w
0 ) 06-
o
- E _
M0 M
.s ,vD
E
E
ena
Q).U)IOJ
.0 C
U1)
M
A3I
-c
(VJc
w M
-
-6o
.) _
.S
a.C
M~
U)
CuI.Q
cu
M 0 -i
1 C1
m Ifl
:l
C
ci,
=1 Q)
C cI
C
ai .Ca
U)
I
)a)
-01M
0. .C~-C
.0 ai IL)o a
> 4)
0
cu
U,
I'
E :2E
0
2,
-.
cd
C
a,
CD
Q
Cg
-
8
+
+
-
II
a
0
U
En4
O0
0U
0
C,
V
<
*<U =
', to
0,D;
uZc
Co
CzD
z<
-0
-05
u
.0i
0
>3
*
0
ct
0
CU
4-
cn
g
Ca
3 0rs5-
C
Cu
s
C.
.0:
U]
..5-
000v
o
0
qo
ar,
to
o
to
S
Co
hi)
It
Q.
0.
Q,
0
.0
0.
0
00
>3
M~0
00
(D
4 -a
Uu~
S- 0
'St
(N1i
Clj
7S
cli
co
clV
-
C!
clu
C
0.
I>U3.
o
L..
'
C/)
CUr
'r0~
o4X
'--
~.
Ei
x
O
0
0
CU
;>
0
0
cU
C
s
C.
!O
0.
CU
OU,
=Q
I
0
CZ
0)
C
Ci
U)
67
cl
E
$S
'a
Co
6
Cl
C
1
Cl
cl C
I
.
Ut
U)
C
C
U,
0
CA
6
C
C-i
0
CX
c
0v
ov
-a
-
0
0
= M0
C
ux
0-4
cu
t, >CZ
co
-
0
0
0
.0
O,
.
M.
C 0 o
C.
410
S
0.
Us,
0
0
X
.
I
C
qs
E
-CU
, rCE
-fs,vC
,
x
CZ
0
a
0
s
0
zill
H ai
<
CMC0
i
Co
0
C
cn
+±
aC>
~°a ' X.
O
0
C)
>3
0
-
o0 >1
00c
.-
(U
P,
In
raU)
0
.i1
0
CU
sCU
._
+ .
co
E-.I
0
Cd,
'o
C
>.I3
~ I~
I~
: o I>1~4jCU00
00.0
-S
a
)u)
0
-0
a.
C-
00
(NJIce)
0oico
0H zI---
0D
-n
'P.
C.):
0
CX
I:E
t-,
*S-T
0')
0
In
V}
P
5-Xo
0
co
3:
C _i_
a)
U)
>-+
(U
aI x0 x =-
0
CU
-o
x
0.
U C
U
-
=
0
0)
Nrj
C
c,
0
co
U
0
-F-z
C
co
0
0
0
Co
'V-
,. -
fc
0
w
Or
A
+
U
I
0.
oZ
0
cn
ccO
0ZC~P~(
ai)
0
-a
+ .
M
O> C
C+
I
OU
U
I
0
C
o
0
U+
C
-
,o
cJ 0
+0
I
0 ;
o
v
VI
m
E
<I
o
OsE
11
M O O++
+
A
11
V
o
I
II
ci
CO)
0
U)
a
-
C,,
En
+
0
+
i)0 sC01
U
L .0
+
.L L
2co
V
C
c
H-
a)
M
-
C
Co
0
0
0
O
CM
CU
0In
U,
0
U,
0e
-5
0
0
Ca
CU
"I
5-
CU
CU
U
CU
C
(0
co
CN
Cli
Cl
Cl
Cli
cli
Cl~
Ci
cli
'0
C
_0
C
E
a)
C
E a
.0
.5
U)
a)
E
(U
U) ch
¢-. C =
.
00 01 -Q1 1C Ul
0 (U
QO ) E mUO I5
E
U)
r
C 75
aU)
El O
.0.
A
II
V
C)
as
0
bb
I'U)
a)
DW)
0
CQ )
.)
C)
-
6
0-
+n
.
II
_
, *-
(I-
(/)6.+
(1)
C.
M
.L
+
-
O=
"S c
n ) LO
v
N
+I~
C O
-C Ct
E
0
.0
22
O
a:
oo+
- C
Cl
+
C.
V
,
C)
+
Q
+
Z
o.o
.EU
_ IIA
5 _0
Ct
00
=
t
I -1
-0
ZOO
On
O
N
, -)
n
U)
0)
0
co
CO
-0
C
~-
(0)
0)
N-
U)
U)
-
0
(n
0
l
W) C)
C)
0
-C)
z C)
+Z
25
4+
z
+
+
n
o
C-4
Z °
a) s
+
.0.
C)
C)
o
sI
~>z0
I--
U
Cl
-
C (
I
0r
Z LU.<
::
+ +C)
0
>
.0
80 _
Ct
o
'IZ
X
I, C)
w
0
o)
<: w
0
(C)
zz n
0 -Oc,
n
-
C0
Os~
(U+
I
w
00)
v
+ +
II
_d 43
c+~
=
0.
ct
C
A
11
V
Q-
A
>o
-
z
_rC
C-
o0-
._.
A XI
a>
N~C)
0
<0+
o 0.
_
U
(EIU
a)
(+
o0
Itf-0
+
= Q LL
N0)
0 (U'E E.(U :~C
A
11
V
-
+
v
u, S ¢< n o doO
v
0p 75
W- C)
Z
8<~+
+ + +o-
.
+ It o
I~~~
> AV-C
C.)
v I C.
<
'A
*~ "
0+
-0
. _
0=.
00
C-0
(90 C
L
C,
CUCU
1..
-0 0
-s-.
.0
0 >
0
U
,0
hO
6
+Av -Z
C)
C)
-O
I
Q+
_n
X
U1)
Q-,Q
c
C. -. C
'2
v
0
(U
C
I
n0 C
a)l
C
?C)
.)
O +
C
+
C)
CU
C1j
Al
V +
r-C:o
C.
C
+
(U
+
+
C)
0
U)
~
a °1
EC 1 oE
10
U (U a) I *LŽ
a) L
05 (U O
0
a.)
E
L-
0
.0
-- IQ
U2
-0
0
0
.0
a
U/)
+
+
Q)
E
.0 (
E
ol
C -E5
0
*F3
) >
-a
0
U)
Un
a)
.0
E
Z < <:
M'2
Z00
F-° U)
(0
(0)
2f)
o1-
0
)
C
CU
C)
C)
.o_
ocn
C)
C_.s
C)
-0
v
Q
-0
C.)
C)
C)
C.
C)
0
E
0
Cl)
U
C,.'
CU
CU
'A
0co
I-
I0
C)
E
C
5
,~
1C
&
o
os_
0<
C'
0.
Y
O
I
CU
C,''
U
1.
-
C
,-
0
0._
O
0C
0
E
C).0
:
C)
C1)
-0
a)
~0.
I.0
0.
.>
V).
Ct
CZ
o0
C
CU
0
E
O
:
0
C)
.0
s2
0
'IO
o
00
C)4
~
o
~~~~U
o.
t.
.-
0
0
C
7
4-
n
0.
C
-
C
0
L0
co
CI
cli
o0
O
1-
r
.
(0
(0
(0
'
V-I?
-
co
N
vE
Q.
(NCl4
o|
V)
co
CC)
0CCU
s-
0
03
t0
T
(L
r
0CIC)
X
0o.'
.05,
tz
0
C
U.
'C
-
C)
)
._U
cn
CO
CC
C)
>.
V.
CU
_'
U
0
-0
0
a)
V)
-
C)
CU
C
U
0
,,~~~~~~~,~~~
-0
E
(1)1
E
.0 CIU)I
U))
00
a)
+
Al.=
+
V O+
+ 6
+ U
("l
o
=
"a
6 = z Z
0
II'U
+
2o
0+
o
0,
II
V
v
Z+
+
+
a6
+ + o 2
+
N
U')
+
a
0
0
v
X
:L
CU
0
A
'-
V
<
M0
N-
z-
?o
CdC
00
X
* +
Oc
+o Q
,. >,
0100
wO
x
O'P
D~~o
n
[_
>
r
t:
-
2<(_E
la
Ma
V
X
+ ~m-2o
00a._a,C
N
+
.
Ia)
0 '
V
*.
O
: la
C)
a)Z
Z+
+
+
:c
oQ a+ _
.Pa aOa
a0.
3="
Mo
76Co
RMC>
U) I'_
00
C C
1.+
0
0+ o
0o
+n
Z
M
-i
+ ~'
0
+
V
4
C
A <
A
II
V
a)+C
+ C
<
L
(1)(U
.La
11
-
-n
2
M
n
c5z
*_to
i -
"- x
01
CY)
Co
N
:x
0+
+,
L-U
-o
6
+aX
0-
-00
.c
:3
-C
o t
-E.
-0
> ()
Q (
(D 6 O l=
Q.
C.)
E
0
1)
+
0(N
Ig
V
a)
00
c
.L.
0
O
o
E
a)
0
EoEO a E o E ( (O
IIiQ
>
U)
C
2
UFna
B - 0 0
=) -0D
0n =3
- x>U -7-2
0) (UO
0o
- oo
M.
.a
C
1
_
C
(U
C:
cn (U'-i
:E
tv
co0 0 _0 o0o )co 4D.1
00il
(U~E~-t~~
-.,. 'I.#" 0M
l
0
0
l(
0
(Ua 0 0 0
-C)0CC
L.
C
M
0)
14
co
O ,
co
Z CO
N-o) k
z
U_ U_
<
0
0
0
-o
U,
I
a)
o01
3
01
a
Q)
?)
ao
0.
C
I>
C)
C)
o0
C)
C
z
w
00
I.
~0
w
0
- .-
Oa)
C
U)
o
(0
CO,
(0
CY)
CY)
(0
)
a)
0
a)
0
o
o
0
3
0
C.)
3
0
0
0
0
a)
a)
CA Mn M
a)
Cd
U
a.
a)
a)
"a
oX
-o
~
ao
a
0
EE
-C
a)
-o
0
C:
a
.
7
0
0
a)
a)
"
-
_
01
t:
01
CL
0.
0S.
O
- E,
0.
a.
<
o
a '-a
0
EnE
C
0
0.2 to
;>.
30
C) 0
3 I- M4
=
0
I,
"
0
0.
U0,
0
C.)
,
6
X
aO
,..
-
01
0.~
-z
ot _;76
cOi
.
4
a
o
01 O
C)
cli
a
0
cN
.qJ
Ij
( )
U)IT
rr
a)
n
~c0
0
o
*a
1
MVa)t
(
+
0
0
a
cn
=*
a)
In
a00)
a)
z
.:
6
0
a)
"a
0.
U,
C
In
U)
a0
.0
av
a)
0
C.)
0.
a)
In
)
*4,
a)
C
a.
01
-0
;>1
a)
cn
O
'.
-
'
-
LO
'T
)
(D
o
c
U)
-c
0a
cu
Q
0
Ca
o
o1
a)
a_
A
+
v
0b
o
V)o
l
V
+
A
II
>,
3
+
=+
cj
O
,~~
-c
uu
+C's0
=
rX2o
Ml
Q
-o
Q=
U+
[
a+
v l.
5
- :
C
.
E
n
u... nu n1-1u
r-
+
+
<
+
Z
+
z
ZE
-~
-
.c
+ <r
+
O+
>+
a
U)
+
,+
o
+
0. o+
a)
g
3d
.~E
II
a)
C)
0<
0°
rn -3
7zt<
Z
5In
n :
( 1. oI
O
D
0
-
Co
a
7
-t
+ + C+c
+ +
++
+ a
G
a 0Ir
I -t
+
+Z o
II
=)
Z
+
18~
o
+
a
V
>
I E (9?
+
II
00
. n
1.0a)
V +
+.
OC
00
a+
4)
o
I
i
A
II
+a
0
V0
¢
+
75
a) Ln
ea
Q
.-
z
0
II
V
E
U,
v
0
+
zot
A
C)
O
+ U
+
+
i fa0
II
V
(lil
ofMEM
~-
0
E .
0 0 a) 0 fl
0 cu- j
)6
I)
)
> MR
-
a)
)
a
U,
0c
0
C
C)a)
aa)
0)
0-
C
(u
a)
o.
U1)
(I)
C
-.
a)
-z
Icl
z
><Z'7
0
+
z.~+
'.q7
i
I~-~4
L0.-C(C
Uo
ZD Q
C)
aD
Cub
<
.
z
<_
.i00
cn
a)
CO o 'n
o C
c
fi!~1~_1
Icu~~°
n Z
Dl
Z _
Jzlu
C:
10)
U)
1'
n
0
vl
C0-
( ( I
5Iv5
Ti5 ~10c
-~
0
,
4)
4)
o,
,-
4)
a
0
a.
¢0
u
C)
-'
,0
-0
a)
ct
0.
o
oQ
4)
7-
0
o
ao
.V
\~
v0
C
10
v0
C.'
O4
£
C
M
0
4
o
0o
0
0o
0
ce
-v
0
ZO
o
.0a)
._a)
C
C
C')
- U)ci
4
C>
-v
_
'M
£
4)
0
.
c)
o
0
t
a)
ob
o >
c°·
o
¢0
a)
)
C'
k
C
M
I
-c
0
E
0
0
w
<
z
0
Z°._o
<z o
o
'dt
0(
C')
t
-
C)
c(
C)
·
.
C
.
.
17
M- 1
(9
(9
C
w
Z
Z
0
Z
w
0
Z
Z
Z
Z
Z
Z
z
z
z
w
w
W
0
LLI
z
*0
U)
U)
(U
c
(U
a)
M
(
E
U)
C
U)
aCD
C
(A
(U
= E
LL 0n
±+
a
+
a.
+4
7
z
¢g
+
0
0
0 -0
C0
OJO0
7: a) 0 Ea
0
n
Cl
IE
~
EI
o0
z+~0CI z
~Cl 007
+
00
i V 00 X
P-HC
04
c7
+
I6
=
70C
Q o +
Z
U)
U)M
ai
(-
1
a
z
C+
(N
I
t. 0
z0
+
Cl+
04
+
Z+
Lb
(D
E1E
I+
r0
+
a)
<
+
+
+
o
o
II C
II
+a)-Q,
\O
0
a)
a)
Zo
a)
o
~cl
v
a
b
+
;
+
+ +
c2
"t
zzzz
C
+ + ¢ C)
U
oZ
+Ho<Z
+c
+ CO II
<
a+a
a +0
-a~
+
+
a.
0o0
(Ulol
a)Lnu
cog
+
a.
II
n
0
cu
.0 E.c75>
a -a
C4+
+
C:1
C
(U
Co
O
Qu
E
+
o
o
E
11
+
+ <Om
0_+
. +
-D
0.
+ c
11
00
I
+a
u)
a-oEa. E
0
I
-
0
I0
00
Z +Z
7+
c
+
66
C
+_
o
E
I=
*J1~
= C- =
SC Q
+
c
C
VII-
NZ
0
Z<0ClO
0<
o0 A
s
+ °
O
o tAZ
O
co
n Ccen
L-)
co
< m
w
I
0)
o+
o
C
Uo
~n
H
1.
0
Oz
w
or w
ll
m
II
0J
zZ
0
Z
z
w w
zZ
00
· 3
-
a)
LL
4I
LL
no
0 0UJ
_
z
z00z
ZZ
ZZ
o
-< <
U
U
O
.
a)
a)
E
wLw
L
L
LLJ w
w
Z
4:Z
Z
H
0
Z
ZF<:
Z
H
0Z
Z
Z Z
1ZZ
i:Z
0000zZH0Z 0HZZ
Z Z
LLU
w
wLU
z6
zw
z4: z6
zLL zw
P--
0J
0
r:
Zw
w
0
I
m
>zz
(d
0Oo
-oI
I_-
Co
A
,,o
o>- w
~~
Z
c:
-v
a)
o
a)
U
2·
;g
$
O
-r O-H
0 ZZ
000
0:: O
>w
Uo
n
aL
I:
CD
I0)
CD
L
L()
C-
,)
-o
0
ea
,
0
C-
C,
r
>i a)
cn 0
0)
co
a
.oC
O
c0
0
a
-Q
(u
0
a)
0
e
-ol
Ou
a-
+ O aC. +
L6
+
11
w Q
U)
L.
O
C ,O
t. +
-o<
z a
°
+
O)
'
)cn
+
O
>-
O
C
C U
a) I
O O
-LL
>0,
~rn a.
Q
6
O
(\J U)
oo
<
c
0
a)
0
,
0
)
C
O
.m
N
c
a)
C
0
,t
r
Appendix B
Set of independent pathways derived from stoichiometric model of Synechocystis metabolism.
These act as constraints
on the experimental rate measurements
'glycogen' and 'PHB' represent the biosynthesis of the monomers.
1. 12 CO2 +2 S042 + 4 NO3-> 2 Gly + 2 Met + 34 0 2
2. 114 C02 + 2 Mg2+ + 10 N03- - 2 Gly + 2 Chl a + 181 02
3. 24 C0 2 + 6 N03- + 6 P0 34 - 2 Gly + 2 dTTP + 46 02
4. 12 C0 2 + 6 N03-+ 3 P034--Gly
5. 18 CO2 + 6 N0 3-+ 6 P0 34 -
+ dATP + 33.5 02
2 dCTP + 40 02
6. 8 CO2 + 4 NO3 -> Gly + His + 22.5 02
7. 12 CO2 + 6 N03 + 3 P0 34
-
8. 12 CO2 + 6 N03 + 3 P034' -
Gly + dGTP + 33 02
Gly + ATP + 33 02
9. 6 C0 2 + NO3 -> Ile + 11.5 02
10. 12 C0 2 + 6 N0 3 + 3 P0 3 4--> Gly + GTP + 32.5 02
11. 34 C0 2 + 6 N0
3 +
P0
4
3 - ->
PG + 2 Try + 2 Ala + 60 02
12. 6 C0 2 + 4 N03- - Arg + 21.5 0 2
13. 18 C02 + 6 N03-+ 6 P0 34' -
2 CTP + 39 0 2
14. 18 C0 2 + 4 N03-+ 6 P0 3 4 -
2 UTP + 31 02
15. 5 C0 2 + 2 N0 3 --> Gn + 12.5 02
16.4 CO2 + N03 -
Thr + 8 02
17. 6 C0 2 + N03'
Leu + 11.502
18.5 C0 2 + N03 -
Pro + 9.5 02
19. 5 C02 + N03- - Glu + 8.5 0 2
20. 6 C0 2 + 2 N03- - Lys + 1502
21. 5 C0 2 + N03- ->Val + 10 02
22. 9 CO2 + S0 42 --
SQDG + 10 02
23. 9 C02 + N03- - Phe + 14 02
55
made.
The equations for
24. 9 CO2 + N03 -->Tyr + 13.5 02
25. 3 CO2 + N03- + S0 42 -> Cys + 8.5 02
26. 4 CO2 + 2 N03- -->Asn + 11 02
27. 3 C0 2 + N03- -> Ala + 7 02
28. 15 C02 -> DGDG + 15.5 02
29. 4 C0 2 + N03- -->Asp + 7 02
30. 18 C0 2 -> [18:1] + 25.5 02
31. 9 C0 2 -->MGDG + 9.5 02
32. 18 C0 2 -
[18:0] + 26 0 2
33. 16 CO2 -
[16:1] + 22.5 02
34. 16 CO 2 -> [16:0] + 23
02
35. 4 CO2 -> PHB + 4.5 02
36. 3 CO 2 + N03- -
37. 6 C
2+
Ser + 6.5 02
P034 -> PG + 7 02
38. 6 C02 -> glycogen + 6 0 2
56
Appendix C
Experimentally measured data and values recalculated through applying network constraints
following the withholding of the oxygen rate data. All measurements are in moles/min/10 ml
culture for wild type late exponential phase cells described in Chapter 4.
Measured Data
Recalculated Data
C02 Uptake Rate
6.0x109
6.0x10 -8
Glycogen Synthesis Rate
PHB Synthesis Rate
7.3x10-'1
3.3x101"
4.7x10- "
4.7x10-
5.7x10 10
5.1x10O-'
Amino Acid Synthesis Rates
Ala
Arg
10
2.8x1 0-°
-
4.0x10-
10
Asn
3.1x10 '0
2.3x10
Asp
3.7x10-1 °
2.9x10-' 0
Cys
1
1.2x10-
' 1
6.0x10-
Glu
4.3x10-
°
3.3x10-'0
Gin
2.8x10 -1 0
1.8x10 10
Gly
5.0x10-1 0
3.5x10-°0
His
1.5x10-
lie
3.7x101- 0
2.5x10'
Leu
6.2x10-1 0
5.0x10 ' 0°
Lys
4.0x10 0
2.8x10' 0
Met
1.7x10-
Phe
2.7x10-'1
10
0
1.4x10-1
10
0
1.8x10-1
0
9.0x10-11
1- 0
0' °
Pro
3.7x10
2.7x1
Ser
Thr
4.8x10-'1
' °
4.0x10-
4.2x10 10
3.2x10-01
Trp
9.0x10-11
1.3x101 0
Tyr
2.2x10-'
Val
4.5x10-10
3.5x10 - 0°
Chi a
6.7x10-1 0
3.1x10-' 0
MGDG
1.0x109
8.2x10-'
DGDG
3.3x10
°
4.0x10-'1
Lipid Synthesis Rates
SQDG
' °
3.1x10'-"
0
1.7x10-'
3.5x10-'
57
°
°
PG
3.1x10 -1 °
1.9x10 -' 0
[16:0]
2.0x10 - 9
1.7x10 - 9
[16:1]
1.6x10 -9
1.3x10 -9
[18:0]
3.0x10 1 0
5.8x10 11
4.1x10
- 10
dGTP
1.9x101
-0
1.0x10
dATP
1.5x10
-10
6.3x101
-1
1.6x10
1- 0
1.9x10
-1 1
1.4x10
1- 0
5.3x10 -11
GTP
1.6x10
- 10
ATP
1.3x10 -10
4.3x10' 11
CTP
1.4x10
- 0°
3.9x10-11
UTP
1.2x10-'
[18:1]
5.1x10 -1 1
DNA Synthesis Rates
dCTP
dTTP
1- 0
RNA Synthesis Rates
°
7.3x10
-1 1
1
5.6x10-
1
Oxygen Rate Data
Evolution in light
7.2x10 -6
Uptake in dark
6.2x10
58
-8
1.0x10 -7
1.3x10 -8