1

1

Systems-level metabolic flux profiling elucidates a complete, bifurcated TCA 2

cycle in Clostridium acetobutylicum 3

Daniel Amador-Noguez1, Xiao-Jiang Feng

2, Jing Fan

1,2, Nathaniel Roquet

1,2, Herschel Rabitz

2 and 4

Joshua D. Rabinowitz1,2,

* 5

1 Lewis Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA,

2 6

Department of Chemistry, Princeton University, Princeton, NJ, USA 7

8

Running title: A bifurcated TCA cycle in C. acetobutylicum 9

10

Key words: Clostridium acetobutylicum, metabolomics, metabolic networks, anaerobic 11

metabolism, mass spectrometry 12

13

* Corresponding author. Department of Chemistry and Lewis Sigler Institute for Integrative 14

Genomics, Princeton University, Princeton, NJ 08544, USA. 15

Tel.: +1 609 258 8985; Fax: +1 609 258 3565; E-mail: joshr@princeton.edu 16

17

18

Copyright © 2010, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.J. Bacteriol. doi:10.1128/JB.00490-10 JB Accepts, published online ahead of print on 9 July 2010

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

2

19

Abstract 20

Obligatory anaerobic bacteria are major contributors to the overall metabolism of soil and the 21

human gut. The metabolic pathways of these bacteria remain, however, poorly understood. 22

Here we directly map, using isotope tracers, mass spectrometry, and quantitative flux modeling, 23

the metabolic pathways of Clostridium acetobutylicum, a soil bacterium whose major 24

fermentation products include the biofuels butanol and hydrogen. While genome annotation 25

suggested the absence of most TCA cycle enzymes, our results demonstrate that this bacterium 26

has a complete, albeit bifurcated, TCA cycle: oxaloacetate flows to succinate both through 27

citrate/α-ketoglutarate and via malate/fumarate. Our investigations also yielded insights into the 28

pathways utilized for glucose catabolism and amino acid biosynthesis and revealed that the 29

organism’s one-carbon metabolism is distinct from that of model microbes, involving reversible 30

pyruvate decarboxylation and use of pyruvate as the one-carbon donor for biosynthetic reactions. 31

This study represents the first in vivo characterization of the TCA cycle and central metabolism 32

of C. acetobutylicum. Our results establish a role for the full TCA cycle in an obligatory 33

anaerobic organism, and demonstrate the importance of complementing genome annotation with 34

isotope tracer studies for determining the metabolic pathways of diverse microbes. 35

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

3

Introduction 36

In soil ecology, obligatory anaerobic bacteria are key contributors to the putrification of dead 37

organic matter (18). In the human intestine, they are the dominant flora, playing a central role in 38

metabolism, immunity, and disease (16, 24, 30). Obligatory anaerobes also encompass some of 39

the most promising bioenergy organisms. The soil bacterium Clostridium acetobutylicum is 40

capable of fermenting carbohydrates into hydrogen gas and solvents (acetone, butanol and 41

ethanol). During World War I, it was used to develop an industrial starch-based process for the 42

production of acetone and butanol that remained the major production route for these solvents 43

during the first half of the last century (5). Since then, and particularly during the last few 44

decades, an active research area has developed to understand and manipulate the metabolism of 45

this organism with the goal of improving hydrogen and solvent production (5, 15). Despite this 46

long history, there are still key pathways of primary metabolism that remain unresolved in C. 47

acetobutylicum. In particular, as is common for most anaerobic bacteria, the TCA cycle remains 48

ill defined (14, 23, 28). 49

C. acetobutylicum is capable of growing on minimal media (i.e. using glucose as sole carbon 50

source) (20) and it therefore must be able to synthesize α-ketoglutarate, the carbon skeleton of 51

the glutamate family of amino acids. Its genome sequence, however, lacks obvious homologues 52

of many of the enzymes of the TCA cycle, including citrate synthase, α-ketoglutarate 53

dehydrogenase, succinyl-CoA synthetase, and fumarate reductase/succinate dehydrogenase (23). 54

The apparent lack of these genes precludes the production of α-ketoglutarate by running the TCA 55

cycle in either the oxidative or reductive direction. 56

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

4

Two recent attempts at reconstructing a genome-scale model of C. acetobutylicum metabolism 57

have encountered this problem. In one case, it was proposed that the TCA cycle functions in the 58

reductive (counterclockwise) direction to produce α-ketoglutarate (14). In the second, it was 59

hypothesized that glutamate is synthesized from ornithine by running the arginine biosynthesis 60

pathway in reverse, bypassing the need for the TCA cycle (28). With the exception of the TCA 61

cycle, the other core metabolic pathways, e.g., of glucose catabolism and amino acid and 62

nucleotide biosynthesis, appear based on sequence homology to be complete and analogous to 63

those in more well-studied bacteria such as E. coli (23). 64

Here, we use 13

C-labeled nutrients as isotopic tracers to follow directly in live cells the operation 65

of C. acetobutylicum’s TCA cycle and other primary metabolic pathways. In contrast to the 66

previously proposed hypotheses, we find a complete, albeit bifurcated, TCA cycle. α-67

Ketoglutarate is produced in the oxidative direction from oxaloacetate and acetyl-CoA via 68

citrate. Succinate can be produced in both the reductive direction via malate and fumarate and 69

the oxidative direction via α-ketoglutarate. We also observe that C. acetobutylicum’s one carbon 70

metabolism is distinct from that of more well-studied bacteria: the carboxyl group of pyruvate 71

undergoes reversible exchange with free carbon dioxide, and the one-carbon units required for 72

methionine, purine, and pyrimidine biosynthesis are primarily derived from the carboxyl group 73

of pyruvate with minimal contribution from serine or glycine. 74

To obtain a quantitative understanding of the newly proposed metabolic network, we formulated 75

an ordinary differential equation (ODE) model that allowed us to calculate the metabolic fluxes 76

through glycolysis, the Entner-Doudoroff pathway (which was inactive), the non-oxidative 77

pentose phosphate pathway (there is no oxidative PPP pathway), the TCA cycle and adjacent 78

amino acid biosynthesis pathways. Beyond providing a quantitative description of metabolic 79

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

5

flux, this model was useful for unraveling ambiguities in the network structure that were not 80

readily distinguished based on qualitative labeling patterns alone. This study represents the first 81

in vivo experimental characterization of the TCA cycle and central metabolic pathways in a 82

Clostridium species and demonstrates the importance of complementing genome annotation with 83

isotope tracer studies in the construction of genome-scale metabolic networks. 84

85

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

6

86

Materials and Methods 87

Media, culture conditions and metabolite extraction. 88

C. acetobutylicum ATCC 824 was grown anaerobically at 37ºC inside an environmental chamber 89

(Bactron IV SHEL LAB Anaerobic Chamber) with an atmosphere of 90% nitrogen, 5% 90

hydrogen and 5% carbon dioxide. The minimal media formulation used in both liquid and filter 91

cultures was 2 g/L KH2PO4, 2 g/L K2HPO4, 0.2 g/L MgSO4·7H2O, 1.5 g/L NH4Cl, 0.13 mg/L 92

biotin, 32 mg/L FeSO4·7H2O, 0.16 mg/L 4-aminobenzoic acid and 10 g/L glucose (20). In the 93

pertinent experiments, acetate, glutamate, aspartate or ornithine were added at a concentration of 94

2 g/L. In addition to the appropriate minimal media, the plates used in filter cultures contained 95

1.5% ultrapure agarose. 96

Detailed protocols for preparing filter cultures and extracting metabolites in Escherichia coli 97

have been published (2, 31), these methods were adapted for their use in C. acetobutylicum. 98

Briefly, for the preparation of filter cultures single colonies were picked from agar-solidified 99

reinforced clostridial medium (RCM; Difco), resuspended in liquid RCM, heat-treated at 80ºC 100

for 20 minutes and grown to saturation overnight. This overnight culture was then used to 101

inoculate a liquid minimal media culture to an initial OD600 (optical density at 600 nm) of 0.03. 102

When this liquid culture reached an OD600 of ~ 0.1, 1.6 ml aliquots were taken and passed 103

through 47 mm diameter round hydrophilic nylon filters (Millipore, HNWP04700) which were 104

then placed on top of agarose plates with the appropriate minimal media. Cellular metabolism 105

was quenched and metabolites were extracted by submerging the filters into 0.8 ml of -20 °C 106

40:40:20 acetonitrile/methanol/water (25). The filters were then washed with the extraction 107

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

7

solvent and the cellular extractions were transferred and centrifuged in eppendorf tubes, the 108

supernatant was collected and stored at -20 °C until analysis. To measure growth, filters from 109

parallel cultures were washed thoroughly with 1.6 ml of fresh media and absorbance at 600 nm 110

determined. 111

112

Metabolite and flux measurement 113

Cell extracts were analyzed by reversed-phase, ion-pairing liquid chromatography coupled by 114

electrospray ionization (ESI, negative mode) to a high resolution, high mass accuracy mass 115

spectrometer (“Exactive”, ThermoFisher) operated in full scan mode for the detection of targeted 116

compounds based on their accurate masses. This analysis was complemented with liquid 117

chromatography coupled by ESI (positive and negative mode) to Thermo TSQ Quantum triple 118

quadrupole mass spectrometers operating in selected reaction monitoring mode (1, 19). 119

Hydrophilic interaction chromatography was used for positive-mode ESI and ion-pairing 120

reversed phase chromatography for negative-mode. Amino acids were derivatized with benzyl 121

chloroformate before their quantitation by negative-mode LC-ESI-MS/MS (13). The 122

determination of absolute intracellular metabolite concentrations was performed using an isotope 123

ratio-based approached previously described (2). Briefly, C. acetobutylicum was grown in U-13

C-124

glucose medium to near complete isotopic enrichment and then extracted with quenching solvent 125

containing known concentrations of unlabeled internal standards. The concentrations of 126

metabolites in the cells can be then calculated using the ratio of labeled endogenous metabolite 127

to non-labeled internal standard. 128

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

8

We used kinetic flux profiling (KFP) for measuring metabolite fluxes and elucidating the 129

metabolic network structure (31). Filter cultures were grown on minimal media plates to an 130

OD600 of 0.35 and then transferred to minimal media plates containing uniformly 13

C-labeled 131

glucose as the sole carbon source. At defined time points after the transfer (e.g., 1, 2, 4, 7, 10, 15, 132

30 and 60 min) metabolism was quenched and cell extracts were prepared and analyzed. The 133

multiple isotopomers produced by the 13

C-labeling were monitored simultaneously using LC-134

MS. Metabolic fluxes were calculated based on the kinetics of the replacement of the unlabeled 135

metabolites by the labeled ones. Similarly, he KFP experiments with uniformly 13

C-labeled 136

acetate were performed by transferring the cells to glucose minimal media plates with added U-137

13C-acetate. 138

For the long-term labeling experiments using 3-13

C-glucose, 4-13

C-glucose, 1,2-13

C-glucose, 13

C-139

glutamate, 13

C-ornithine or 13

C-aspartate, the cells were extracted 2 hours after transferring them 140

to the plates containing each labeled substrate. In all the experiments with 13

C-labeled amino 141

acids, the usual concentration of non-labeled glucose was maintained. The 13

CO2 labeling 142

experiments were performed by adding increasing concentrations of NaH13

CO3 into 143

exponentially growing liquid cultures (OD600=0.35). After 1 hour the cells were quickly filtered 144

and extracted using -20 °C 40:40:20 acetonitrile/methanol/water. 145

The labeling of the C1 units pool was determined from the labeling patterns of various 146

intermediates in nucleotide biosynthetic pathways that incorporate C1 units. We used 5’-147

Phosphoribosyl-N-formylglynamide and Inosine 5’-monophosphate, which incorporate C1 units 148

from 10-Formyl-tetrahydrofolate as well as dTMP, which incorporates a C1 unit from 5,10-149

Methylene-trahydrofolate. In addition, methionine, which incorporates a C1 unit from 5-methyl-150

tetrahydrofolate, was used as corroborating data in some instances. 151

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

9

We used metabolic flux profiling to complement the information obtained from KFP and for the 152

determination of flux ratios in various pathways. We followed the general approach described in 153

(9), with the difference that instead of inferring labeling patterns from proteinogenic amino acids 154

we quantify them directly for most metabolites. In all experiments, the labeling data was 155

corrected for natural abundance of 13

C in non-labeled substrates and for the 12

C impurity present 156

in 13

C-labeled substrates in a similar fashion as reported previously (2, 31). 157

158

ODE modeling and parameter identification 159

We constructed an ODE model for the metabolic network shown in Figure 4 and Supplementary 160

Figure 7, and then identified model parameters (fluxes and unmeasured pool sizes) that 161

reproduce the laboratory data. The procedure was based on methods previously developed (8, 162

22). The ODEs described the rates of loss of unlabeled forms of metabolites (and the creation of 163

particular labeled forms) after feeding of U-13C-glucose. The equations are based on flux 164

balance of metabolites and take the form of 165

166

∑=

−=

N

i tot

tot

tot

i

iB

BF

A

AF

dt

dB

1

, 167

168

where a metabolite B, which can be in labeled or unlabeled forms, is downstream of other 169

metabolites Ai. The outflux Ftot of balances the sum of N influxes Fi from Ai. Atot and Btot and 170

total pool sizes of the corresponding metabolite (sum of labeled and unlabled forms). 171

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

10

The unknown model parameters were identified by a genetic algorithm that minimizes a cost 172

function (7, 8). The cost function quantifies the difference between the computational results and 173

the laboratory measurements for the labeling dynamics, together with the additional constraints 174

indicated in Supplementary Table 1. One thousand sets of model parameters that can reproduce 175

the laboratory data were identified, forming a distribution for each parameter (Supplementary 176

Figure 9). The median value and the breadth of the distribution then provide a representation of 177

the fluxes consistent with the laboratory data. The C/C++ program used for the modeling and 178

parameter identification are available upon request. 179

180

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

11

181

Results 182

Glucose catabolism to pyruvate 183

To probe metabolic flux in growing C. acetobutylicum, we monitored the dynamic (time-184

dependent) or long-term (steady-state) incorporation of 13

C-labeled nutrients (glucose, acetate, 185

CO2 and selected amino acids) into downstream metabolites in glycolysis, the pentose phosphate 186

pathway, the TCA cycle and amino acid and nucleotide biosynthetic pathways. 187

C. acetobutylicum can potentially metabolize glucose to trioses via three different pathways: 188

glycolysis (the Embden-Meyerhof pathway), the Entner-Doudoroff pathway, and the pentose 189

phosphate pathway. Homologues of enzymes of each of the above pathways, with the exception 190

of the oxidative pentose phosphate pathway, are present in the C. acetobutylicum genome (23). 191

The relative contribution of glycolysis versus the Entner-Doudoroff pathway to pyruvate 192

synthesis can be determined from cells grown in 1,2-13

C-glucose (carbons #1 and #2 are 13

C-193

labeled) or 3-13

C-glucose (carbon #3 is labeled) since each pathway yields distinct positional 194

labeled forms of pyruvate, which can be distinguished by tandem mass spectrometry (MS/MS). 195

For C. acetobutylicum growing exponentially on glucose as the carbon source, all pyruvate 196

appeared to be derived from glycolysis, with no detectable Entner-Doudoroff pathway flux 197

(Supplementary Figure 1). 198

The pentose phosphate pathway provides essential precursors (ribose-5P and erythrose-4P) for 199

nucleotide and amino acid biosynthesis. Ribose-5P molecules can be produced either by the 200

oxidative pentose pathway from glucose-6P or by the non-oxidative pentose phosphate pathway 201

via the transketolase reaction or by the combined activity of transaldolase and transketolase. 202

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

12

Consistent with lack of oxidative pentose phosphate pathway enzyme homologues in the C. 203

acetobutylicum genome, feeding of 1,2-13

C-glucose resulted in no detectable production of 204

ribose-5P containing a single 13

C atom, the hallmark of oxidative pentose phosphate production. 205

Pentoses were instead produced via transketolase (~80%) and via transaldolase-transketolase 206

(~20%) (Supplementary Figure 2). 207

The above experiments suggested normal catabolism of glucose into pyruvate via glycolysis, and 208

consistent with this, feeding of U-13

C-glucose as the sole carbon source resulted in rapid and 209

complete labeling of glycolysis intermediates through phosphoenolpyruvate. Pyruvate, however, 210

appeared in roughly equimolar amounts in its fully labeled form and in an unexpected form with 211

two 13

C-carbons (Figure 1B). Glycolysis splits glucose down the middle, converting carbon 212

positions 1, 2, and 3 (and 6, 5, and 4) into the methyl, carbonyl, and carboxyl carbons of 213

pyruvate respectively. As shown in Figure 1C, growth of C. acetobutylicum in 1,2-13

C-glucose 214

(100%) resulted, as expected, in ~50% of phosphoenolpyruvate and pyruvate each containing 215

two 13

C-carbons. In contrast, feeding of 3-13

C-glucose (100%) or 4-13

C-glucose (100%) resulted 216

in 50% labeling of phosphoenolpyruvate (and upstream trioses) but only ~25% labeling of 217

pyruvate. This suggested that 13

C-label was specifically being lost from the carboxyl carbon of 218

pyruvate, presumably in an exchange reaction with environmental carbon dioxide (CO2), which 219

comprised 5% of the anaerobic gaseous environment and is ~99% non-labeled. Consistent with 220

exchange of the carboxyl carbon of pyruvate with carbon dioxide, growth of cells in the presence 221

of NaH13

CO2 resulted in formation of 1-13

C-pyruvate, with the fraction of labeling increasing 222

with increasing concentrations of NaH13

CO2 (Figure 1D). There was minimal or no labeling of 223

upstream metabolites. This confirms the rapid interchange between the carboxylic acid group in 224

pyruvate and environmental CO2. 225

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

13

In anaerobic organisms, the oxidative decarboxylation of pyruvate to produce acetyl-CoA and 226

CO2 is catalyzed by pyruvate-ferredoxin oxidoreductase (PFOR, also known as pyruvate 227

synthase). The use of ferredoxin, whose redox potential is close to that of pyruvate, as the 228

oxidant has the potential to make the overall reaction reversible (10, 26). When 13

C-labeled 229

acetate was added into the media, however, there was no detectable labeling of pyruvate, even 230

when a significant fraction of the acetyl-CoA pool was labeled (Figure 1E). In the proposed 231

mechanism of acetyl-CoA synthesis by PFOR, pyruvate is first decarboxylated to form the 232

intermediate hydroxyethyl-TPP (TPP, thiamine pyrophosphate, is the prosthetic group in PFOR). 233

This intermediate then reacts with CoA (coenzyme A) to produce acetyl-CoA (26). Our data 234

indicates that the carboxylic group in pyruvate interchanges rapidly with CO2 but that the overall 235

PFOR reaction is essentially irreversible. We accordingly propose that the interchange results 236

from the reversibility of the decarboxylation step of the PFOR reaction. 237

The interchange between the carboxyl group in pyruvate and atmospheric CO2 could also be 238

explained by reverse flux through pyruvate dehydrogenase or pyruvate formate lyase. However, 239

the presence of a pyruvate dehydrogenase complex has never been reported in C. acetobutylicum 240

or in any other clostridium species (5). Additionally, isotopic tracer experiments in aerobically 241

and anaerobically grown E. coli indicate that neither pyruvate dehydrogenase nor pyruvate 242

formate lyase is capable of producing the exchange between the carboxyl group in pyruvate and 243

CO2 that we observe in C. acetobutylicum (Supplementary Figure 3). 244

245

246

247

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

14

A complete, bifurcated TCA cycle 248

After gaining an understanding of the pathways that catabolize glucose to trioses, we examined 249

the TCA cycle (Figure 2). Feeding of U-13

C-glucose (100%) resulted in labeling patterns of 250

oxaloacetate, malate and fumarate which matched closely the labeling pattern of pyruvate, with 251

the appearance of close to equimolar amounts of isotopomers with two or three 13

C-carbons 252

(Figure 2B). This observation is consistent with the synthesis of oxaloacetate from pyruvate and 253

atmospheric CO2 (which is non-labeled) and with the production of malate and fumarate from 254

oxaloacetate by running the TCA cycle in the reductive (counterclockwise) direction. Succinate’s 255

labeling pattern, however, did not fully agree with pyruvate’s. Although succinate showed the 256

same predominant labeled forms, their ratios were different, with nearly twice as much succinate 257

containing three versus two 13

C-carbons (Figure 2B). This suggested that although succinate may 258

be produced from fumarate, there must also be another source to account for the enhanced triple 259

13C-labeling. 260

The labeling pattern for α-ketoglutarate differed from that of pyruvate or succinate. If, as 261

previously hypothesized (14, 23), α-ketoglutarate is produced from succinate and CO2 by 262

running the TCA cycle reductively, α-ketoglutarate containing two and three 13

C-carbons should 263

have appeared. However, the predominant form of α-ketoglutarate had four 13

C-carbons. The 264

actual route of α-ketoglutarate production was revealed by the labeling patterns in citrate (Figure 265

2B). Despite the putative lack of citrate synthase, there was a measurable intracellular pool of 266

citrate that labeled rapidly after feeding U-13

C-glucose. Citrate (a molecule with six carbons) was 267

produced in two major isotopic forms, containing either four or five 13

C-carbons. This labeling 268

pattern is consistent with its production from oxaloacetate (with 2 or 3 13

C-carbons) and acetyl-269

CoA (where the 2-carbon acetyl moiety is fully 13

C-labeled). The labeling pattern of α-270

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

15

ketoglutarate was then readily explained based on its production via citrate. Citrate containing 271

either four or five 13

C-carbons produces α-ketoglutarate with four 13

C-carbons because the 272

additional 13

C-carbon in citrate corresponds to the carboxyl group that is lost during oxidative 273

decarboxylation of isocitrate to α-ketoglutarate. 274

The labeling of glutamate matched α-ketoglutarate, consistent with its formation by reductive 275

amination of α-ketoglutarate driven by either ammonia or glutamine. To rule out the previous 276

hypothesis that glutamate could be synthesized from ornithine by running the arginine 277

biosynthesis pathway in reverse (28), we added U-13

C-ornithine into the media. While arginine 278

pathway compounds downstream of ornithine became labeled, we observed no production of 279

labeled glutamate (Supplementary Figure 4). 280

The lack of production of succinate containing four 13

C-carbons initially suggested that there was 281

no production of succinate via α-ketoglutarate. However, experiments with additional 13

C-282

labeled nutrients proved that this does occur. When cells were grown in unlabeled glucose plus 283

U-13

C-acetate, 13

C was assimilated into acetyl-CoA. Consistent with turning of the TCA cycle in 284

the oxidative direction: citrate, α-ketoglutarate and glutamate with two 13

C-carbons carbons were 285

produced; the cycle was, however, incomplete: there was no detectable labeling in oxaloacetate, 286

malate or fumarate. Interestingly, however, we observed the production of succinate with one 287

13C-carbon (Figure 2C). 288

This labeling of succinate is consistent with its production from α-ketoglutarate, but with the 289

stereo-specificity of citrate synthase being opposite of that found in common bacterial model 290

organisms and eukaryotes. This unusual (Re)-stereospecificity of citrate synthase was confirmed 291

by examining the position of 13

C-carbons within glutamate and proline by MS/MS analysis 292

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

16

(Supplementary Figure 5). Production of succinate from α-ketoglutarate explains the succinate 293

labeling patterns in the U-13

C glucose labeling experiments: succinate is synthesized both via 294

fumarate (producing succinate containing two and three 13

C-carbons) and via α-ketoglutarate 295

(producing succinate containing three 13

C-carbons). Growing on glucose as the sole carbon 296

source, the relative contributions from each route to succinate production are ~60% and 40%, 297

respectively. 298

Aspartate and glutamate can be deaminated to oxaloacetate and α-ketoglutarate, respectively, to 299

enter the TCA cycle. When U-13

C -aspartate is added to the media (in the presence of unlabeled 300

glucose), a large fraction (>80%) of the malate, fumarate, succinate and citrate pools become 301

quadrupoly 13

C-labeled. α-Ketoglutarate and glutamate become triply 13

C-labeled (Figure 2D). 302

When cells are grown in the presence of U-13

C -glutamate plus unlabeled glucose, both α-303

ketoglutarate and succinate become fully 13

C-labeled. In this case oxaloacetate, malate and 304

fumarate are not 13

C-labeled (Figure 2E). These observations corroborate the existence of a 305

complete, bifurcated the TCA cycle in this organism. 306

307

Amino acid biosynthetic pathways and C1 metabolism 308

By analyzing the labeling patterns of amino acids and key biosynthetic intermediates we were 309

able to resolve most of the amino acid biosynthesis pathways in C. acetobutylicum. The observed 310

labeling patterns were consistent with those expected based on canonical amino acid biosynthesis 311

pathways, with the exception of isoleucine and glycine production (Supplementary Table 2). 312

The labeling patterns in isoleucine indicated that it is not synthesized by the canonical pathway 313

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

17

via threonine but are instead consistent with its production from acetyl-CoA and pyruvate via the 314

citramalate pathway (Supplementary Table 2). 315

For glycine, there are two alternative pathways (Figure 3A). The more common involves 316

synthesis of glycine from serine by the enzyme serine hydroxymethyltransferase, which transfers 317

the methanol group from serine to tetrahydrofolate (THF). The resulting methyl-folate species 318

provide C1 units for the biosynthesis of purines, thymidine, and methionine. Alternatively, in 319

yeast and some bacteria, glycine can be synthesized by degradation of threonine, e.g., into 320

acetaldehyde and glycine (12, 21). When cells were grown on U-13

C-glucose, serine (synthesized 321

via 3-phosphoglycerate) became fully labeled, whereas threonine (synthesized via pyruvate) was 322

~50% triply 13

C-labeled and ~50% doubly 13

C-labeled. Consistent with its predominant 323

formation from threonine but not from serine, glycine was ~50% fully labeled and ~50% singly 324

13C-labeled (Figure 3B). The synthesis of glycine from threonine was further corroborated by 325

growing cells in U-13

C glucose plus non-labeled aspartate and observing that both the threonine 326

and glycine pools are mostly non-labeled, while serine was largely fully labeled (Figure 3C). 327

Since glycine is synthesized primarily from threonine, C1 units must be obtained from a 328

precursor other than serine. Glycine is also commonly used as a precursor of C1 units but we 329

also found that this route is nearly inactive in C. acetobutylicum. When cells were grown in 1,2-330

13C glucose, less than 5% of C1 units were

13C-labeled, even though the methylene group in 331

glycine (the precursor of C1 units) was ~50% labeled (Figure 3D). Conversely, when cells were 332

grown in either 3-13

C-glucose or 4-13

C-glucose, the methylene group in glycine was non-labeled 333

but ~25% of the C1 unit pool was 13

C-labeled (Figure 3E). The possibility that C1 units could be 334

synthesized from the carboxyl group of glycine via some non-canonical pathway was ruled out 335

by the observation that the percentage of 13

C-labeled C1 units is essentially unchanged between 336

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

18

cells grown in U-13

C-glucose versus cells grown in U-13

C-glucose plus non-labeled aspartate 337

(Figure 3E), even though the 13

C-label in the carboxyl group of glycine decreases from ~50% to 338

~10% (Figure 3B and 3C). These experiments show that there is minimal production of C1 units 339

via serine or glycine. We found, however, a strong correlation between the labeled fraction of the 340

carboxylic group of pyruvate and the labeled fraction of C1 units across all these labeling 341

experiments (Figure 3E). In addition, when NaH13

CO2 was added into the media, 13

CO2 was 342

incorporated into C1 units. The fraction of labeled C1 units did not correspond directly to the 343

fraction of labeled CO2 but to the fraction of labeled CO2 that was incorporated into pyruvate 344

(Figure 3F). Our data therefore indicate that in C. acetobutylicum, C1 units are derived primarily 345

(>90%) from the carboxylic group of pyruvate, likely through the combined action of pyruvate 346

formate lyase and formate-tetrahydrofolate ligase. 347

348

Metabolic flux quantitation 349

Among the most important characteristics of a biochemical network are the in vivo reaction rates. 350

To achieve a quantitative understanding of the fluxes in the newly proposed metabolic network, 351

we developed an ordinary differential equation (ODE) model that describes the isotope labeling 352

kinetics of metabolites following the addition of universally labeled 13

C-glucose (see Materials 353

and Methods and Supplementary Figure 7). Given the model equations, we employed a nonlinear 354

global search algorithm to identify the fluxes that can quantitatively reproduce the laboratory 355

data (8). In addition to the labeling kinetics, inputs to the model included intracellular 356

concentrations of glycolysis and TCA cycle intermediates and amino acids (Supplementary 357

Table 3), nutrient uptake rates, excretion rates (Supplementary Figure 6), and specific flux 358

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

19

branch point data obtained from the steady state labeling experiments discussed previously. The 359

details of the cost function used for model fitting are presented in Supplementary Table 1. To 360

avoid overfitting the data, any simulations which fell within the 95% confidence limits of the 361

laboratory data were all considered acceptable; only more severe misfits were penalized during 362

the search. A total of 1,000 well-fitting sets of fluxes were identified and used to estimate flux 363

confidence intervals. 364

Figure 4 shows representative results for the ODE model fitting and a map of the identified 365

median flux values in central metabolism. The complete results are presented in Supplementary 366

Figures 8 and 9. The ODE model fits all of the observed data. Most of the identified fluxes, with 367

the exception of several exchange fluxes, were tightly constrained, indicating that they are 368

reliably defined by the available laboratory data. The results show that glycolytic flux 369

predominates and is directed primarily towards acid production. Other significant fluxes include 370

aspartate production via pyruvate/oxaloacetate, fatty acids production from dihydroxyacetone 371

phosphate, and ribose-phosphate production from glycolytic intermediates. Within the TCA, the 372

flux through the oxidative branch is slightly larger than through the reductive branch. The 373

production of succinate from succinyl-CoA can occur via succinyl-CoA synthetase but is also 374

expected to occur via the canonical methionine and lysine pathways. The computational results 375

indicate that the median contribution of the succinyl-CoA synthetase flux to the total succinyl-376

CoA to succinate flux is about 25%, while the methionine and lysine pathways combined 377

contribute to ~75% of the total flux (Supplementary Figure 9). 378

In addition to providing quantitative flux values, the ODE model also helped to resolve an 379

ambiguity in the network structure that was not adequately addressed by qualitative analysis of 380

the isotope labeling patterns alone. Malate and fumarate production can occur directly via the 381

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

20

reductive TCA cycle, or alternatively from passage of carbon from aspartate to fumarate, which 382

would then be oxidized to malate (see alternative pathway in Supplementary Figure 7). Both 383

pathways result in qualitatively indistinguishable labeling patterns. To distinguish them, we 384

constructed ODE models for the two alternative pathways and performed flux identification 385

using the procedure described above. Both models were able to describe the quantitative 386

dynamic data following U-13

C-glucose labeling. However, in the second model, because 387

fumarate is partly used for production of malate, the contribution of fumarate to succinate 388

production is lower than that in the first model. Quantitatively, the percentage of succinate 389

produced from fumarate is ~54% for the first model and ~6% for the second model. Compared 390

with the experimentally measured value of ~60%, the computational results indicate that the 391

second model is inaccurate and the first one is correct, meaning that malate is produced primarily 392

reductively from oxaloacetate, rather than oxidatively from fumarate.393

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

21

Discussion 394

Comparative genome sequence analysis has become the predominant tool for the genome-scale 395

reconstruction of the metabolic network of microorganisms. Frequently, however, due to 396

incomplete annotation or non-documented functional genes, there are gaps and uncertainties 397

within the metabolic network that need to be resolved experimentally. These limitations get in 398

the way of a comprehensive understanding of their metabolism and interfere with the creation of 399

quantitative genome-scale models of metabolism. This hinders the ability to rationally modulate 400

metabolism for biotechnological or medical purposes. 401

Here, we used 13

C-labeled tracer experiments to elucidate the in vivo function of the TCA cycle 402

and other primary metabolic pathways in C. acetobutylicum. In contrast to the prevailing 403

hypothesis, we found that this organism has a complete, albeit bifurcated, TCA cycle: 404

oxaloacetate flows to succinate both through citrate/α-ketoglutarate and via malate/fumarate. 405

Although there is currently no gene annotated as citrate synthase in C. acetobutylicum, our data 406

revealed the presence of a citrate synthase with Re-stereospecificity. An Re-citrate synthase has 407

been recently identified in C. kluyveri as the product of a gene predicted to encode 408

isopropylmalate synthase (17). The corresponding protein in C. acetobutylicum, CAC0970, has a 409

64% amino acid sequence identity and is one candidate for the Re-citrate synthase in this 410

organism. While aconitase and isocitrate dehydrogenase were not annotated when the genome 411

sequence of C. acetobutylicum was first released (23), the genes CAC0971 and CAC0971 are 412

now annotated as such in KEGG. The products of these genes, however, have not been yet 413

characterized in C. acetobutylicum or in any other clostridia. The α-ketoglutarate dehydrogenase 414

complex is missing in the C. acetobutylicum genome but it has been hypothesized that a putative 415

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

22

2-oxoacid ferredoxin oxidoreductase (CAC2458) could catalyze succinyl-CoA formation from α-416

ketoglutarate (23). There are still no candidate genes encoding fumarate reductase/succinate 417

dehydrogenase or succinyl-CoA synthetase. 418

Initially defined by a set of broad phenotypic characteristics such as rod-like morphology, gram-419

positive cell walls, endospore formation and strict anaerobic metabolism, Clostridium is one of 420

the most heterogeneous bacterial genera (5). In a sequence-based species tree, there are a number 421

of independent and deeply branching sublines within the Clostridium subdivision, which also 422

includes many non-clostridial species (4). Among the clostridia, C. kluyveri shows a unique 423

metabolism; it grows anaerobically on ethanol and acetate as sole energy sources (27). Only 424

about half of the genes in C. kluyveri show a greater than 60% similarity in C. acetobutylicum 425

(3). The similarities that we observe between C. acetobutylicum and C. kluyveri regarding the 426

oxidative production of α-ketoglutarate and one carbon metabolism (as discussed further below) 427

are therefore noteworthy. 428

In both the initial genome sequencing paper and in a subsequent genome-scale reconstruction of 429

the C. acetobutylicum metabolic network, it was proposed that α-ketoglutarate is synthesized 430

from oxaloacetate by running the TCA cycle reductively. It was argued that a reductive TCA 431

cycle would be favored given the low redox potential of the internal anaerobic environment of C. 432

acetobutylicum. It is therefore intriguing that C. acetobutylicum synthesizes α-ketoglutarate 433

exclusively oxidatively. The reasons for this remain unclear, but the conversion of α-434

ketoglutarate into succinyl-CoA appears to be irreversible in this organism: although succinate is 435

readily synthesized via α-ketoglutarate, there is no back-flux from succinate to α-ketoglutarate, 436

even under conditions in which there is ample production of succinate by the reductive TCA 437

cycle (as when cells are grown in the presence of aspartate). The irreversibility is expected if this 438

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

23

reaction is catalyzed by a yet-to-be-identified α-ketoglutarate dehydrogenase but not if it is 439

catalyzed, as previously proposed, by a reversible 2-oxoacid ferredoxin oxidoreductase. 440

Given that α-ketoglutarate is synthesized solely via citrate, succinate becomes a metabolite of 441

apparently limited biosynthetic value. The benefit of maintaining two different routes for its 442

production is therefore unclear. A possibility is that a bifurcated TCA cycle ending in succinate 443

plays a role in cellular redox balance. However, the rate of succinate excretion (~4µmol/hr/ 444

gCDW) is very small as compared to other fermentation products such as acetate and butyrate 445

(~4mMol/hr/gCDW) (Supplementary Figure 6). Another possibility is that this particular 446

arrangement of the TCA cycle facilitates the utilization of certain amino acids as nitrogen 447

sources. For example, when growing in glutamate or aspartate as sole nitrogen sources, large 448

amounts of α-ketoglutarate or oxaloacetate are produced during deamination. While a fraction of 449

these carbon skeletons may be used for biosynthetic purposes, most must be discarded. Their 450

conversion to succinate, and subsequent excretion, provides a short and rapid route. These 451

hypotheses are consistent with the data obtained from our experiments with 13

C-glutamate and 452

13C-aspartate. 453

In most organisms glycine is synthesized from serine, producing a C1 unit during the process. 454

Glycine, in turn, can also be used to produce a C1 unit. In contrast, in C. acetobutylicum the 455

major route (~90%) for the production of glycine is via threonine. This necessitates C1 unit 456

production from a precursor other than serine, and we found that C1 units are predominantly 457

(~90%) derived from the carboxyl group of pyruvate. A related situation has been observed in 458

C. kluyveri, in which 67% of glycine is formed from threonine and 33% from serine, and about 459

25% of C1 units are synthesized from serine and 75% from CO2 (11). The production of C1 units 460

from the carboxyl group of pyruvate (oxidation state +3) can be viewed as reductive pathway 461

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

24

while its production from the methylene group of serine or glycine (oxidation state -1) can be 462

considered as an oxidative pathway. For example, using serine as the source for C1 units, the 463

production of 10-formyl-tetrahydrofolate (used in purine biosynthesis) is accompanied by the 464

production of one NADH; when using glycine two NADHs are produced. However, no NADH is 465

produced when pyruvate is used as the source of C1 units for the production of 10-formyl-466

tetrahydrofolate. Therefore, for an anaerobic bacterium such as C. acetobutylicum, it makes 467

sense to derive C1 one units from the carboxyl group of pyruvate. Also, the capacity to produce 468

C1 units both reductively and oxidatively suggests that the relative utilization of these pathways 469

may be yet another way to control cellular redox balance. 470

Our observations strengthen the notion that pyruvate constitutes a pivotal metabolic crossroad in 471

C. acetobutylicum, linking the TCA cycle, amino acid biosynthesis pathways, one carbon 472

metabolism and acid/solvent producing pathways. It therefore represents a control point that 473

could be exploited to improve biofuel production. For example, decreasing the activity of 474

pyruvate carboxylase should decrease the flux of pyruvate into the TCA cycle and associated 475

amino acid biosynthesis pathways and increase pyruvate flux into acetyl-CoA and solvent 476

production. 477

The dynamic isotope labeling approach (kinetic flux profiling) that we use here is different to the 478

steady-state isotopic approach (metabolic flux analysis) recently used in similar contexts (6, 29). 479

One major advantage of our approach is that it provides absolute fluxes throughout the network 480

instead of just ratios of fluxes at branch points. Additional advantages include easy data 481

deconvolution and short labeling time. The quantitative modeling technique used in this study is 482

generally applicable for the identification of metabolic fluxes from dynamic isotopic tracer 483

experiments (22). In addition to providing quantitative understanding of the target metabolic 484

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

25

networks, we have shown its ability to discriminate among competing network structures that 485

produce qualitatively indistinguishable labeling patterns. Moreover, given appropriate input data, 486

the general nonlinear identification strategy can also be employed for construction of dynamic 487

models that reflect the regulation of metabolic fluxes (8, 32). Such dynamic models can enable a 488

more comprehensive understanding and rational engineering of metabolic networks. In the case 489

of C. acetobutylicum, for example, a model of dynamic regulation could be used to design 490

genetic and nutrient perturbations that enhance solvent and/or biohydrogen production. 491

This study represents the first in vivo experimental characterization of the TCA cycle and central 492

metabolism in C. acetobutylicum and exemplifies the potential of dynamic isotope tracer studies 493

and quantitative flux modeling in complementing genome-based metabolic network 494

reconstruction. 495

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

26

496

References 497

498

1. Bajad, S. U., W. Lu, E. H. Kimball, J. Yuan, C. Peterson, and J. D. Rabinowitz. 2006. Separation 499

and quantitation of water soluble cellular metabolites by hydrophilic interaction 500

chromatography-tandem mass spectrometry. J Chromatogr A 1125:76-88. 501

2. Bennett, B. D., J. Yuan, E. H. Kimball, and J. D. Rabinowitz. 2008. Absolute quantitation of 502

intracellular metabolite concentrations by an isotope ratio-based approach. Nat Protoc 3:1299-503

311. 504

3. Brinkac, L. M., T. Davidsen, E. Beck, A. Ganapathy, E. Caler, R. J. Dodson, A. S. Durkin, D. M. 505

Harkins, H. Lorenzi, R. Madupu, Y. Sebastian, S. Shrivastava, M. Thiagarajan, J. Orvis, J. P. 506

Sundaram, J. Crabtree, K. Galens, Y. Zhao, J. M. Inman, R. Montgomery, S. Schobel, K. 507

Galinsky, D. M. Tanenbaum, A. Resnick, N. Zafar, O. White, and G. Sutton. Pathema: a clade-508

specific bioinformatics resource center for pathogen research. Nucleic Acids Res 38:D408-14. 509

4. Dehal, P. S., M. P. Joachimiak, M. N. Price, J. T. Bates, J. K. Baumohl, D. Chivian, G. D. 510

Friedland, K. H. Huang, K. Keller, P. S. Novichkov, I. L. Dubchak, E. J. Alm, and A. P. Arkin. 511

MicrobesOnline: an integrated portal for comparative and functional genomics. Nucleic Acids 512

Res 38:D396-400. 513

5. Dürre, P. 2005. Handbook on clostridia. Taylor & Francis, Boca Raton. 514

6. Feng, X., H. Mouttaki, L. Lin, R. Huang, B. Wu, C. L. Hemme, Z. He, B. Zhang, L. M. Hicks, J. Xu, 515

J. Zhou, and Y. J. Tang. 2009. Characterization of the central metabolic pathways in 516

Thermoanaerobacter sp. strain X514 via isotopomer-assisted metabolite analysis. Appl Environ 517

Microbiol 75:5001-8. 518

7. Feng, X. J., S. Hooshangi, D. Chen, G. Li, R. Weiss, and H. Rabitz. 2004. Optimizing genetic 519

circuits by global sensitivity analysis. Biophys J 87:2195-202. 520

8. Feng, X. J., and H. Rabitz. 2004. Optimal identification of biochemical reaction networks. 521

Biophys J 86:1270-81. 522

9. Fischer, E., and U. Sauer. 2003. Metabolic flux profiling of Escherichia coli mutants in central 523

carbon metabolism using GC-MS. Eur J Biochem 270:880-91. 524

10. Furdui, C., and S. W. Ragsdale. 2000. The role of pyruvate ferredoxin oxidoreductase in 525

pyruvate synthesis during autotrophic growth by the Wood-Ljungdahl pathway. J Biol Chem 526

275:28494-9. 527

11. Jungermann, K. A., W. Schmidt, F. H. Kirchniawy, E. H. Rupprecht, and R. K. Thauer. 1970. 528

Glycine formation via threonine and serine aldolase. Its interrelation with the pyruvate formate 529

lyase pathway of one-carbon unit synthesis in Clostridium kluyveri. Eur J Biochem 16:424-9. 530

12. Kataoka, M., M. Ikemi, T. Morikawa, T. Miyoshi, K. Nishi, M. Wada, H. Yamada, and S. Shimizu. 531

1997. Isolation and characterization of D-threonine aldolase, a pyridoxal-5'-phosphate-532

dependent enzyme from Arthrobacter sp. DK-38. Eur J Biochem 248:385-93. 533

13. Kraml, C. M., D. Zhou, N. Byrne, and O. McConnell. 2005. Enhanced chromatographic 534

resolution of amine enantiomers as carbobenzyloxy derivatives in high-performance liquid 535

chromatography and supercritical fluid chromatography. J Chromatogr A 1100:108-15. 536

14. Lee, J., H. Yun, A. M. Feist, B. O. Palsson, and S. Y. Lee. 2008. Genome-scale reconstruction and 537

in silico analysis of the Clostridium acetobutylicum ATCC 824 metabolic network. Appl Microbiol 538

Biotechnol 80:849-62. 539

15. Lee, S. Y., J. H. Park, S. H. Jang, L. K. Nielsen, J. Kim, and K. S. Jung. 2008. Fermentative butanol 540

production by Clostridia. Biotechnol Bioeng 101:209-28. 541

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

27

16. Ley, R. E., M. Hamady, C. Lozupone, P. J. Turnbaugh, R. R. Ramey, J. S. Bircher, M. L. Schlegel, 542

T. A. Tucker, M. D. Schrenzel, R. Knight, and J. I. Gordon. 2008. Evolution of mammals and their 543

gut microbes. Science 320:1647-51. 544

17. Li, F., C. H. Hagemeier, H. Seedorf, G. Gottschalk, and R. K. Thauer. 2007. Re-citrate synthase 545

from Clostridium kluyveri is phylogenetically related to homocitrate synthase and 546

isopropylmalate synthase rather than to Si-citrate synthase. J Bacteriol 189:4299-304. 547

18. Ljungdahl, L. G. 2003. Biochemistry and physiology of anaerobic bacteria. Springer, New York. 548

19. Lu, W., B. D. Bennett, and J. D. Rabinowitz. 2008. Analytical strategies for LC-MS-based 549

targeted metabolomics. J Chromatogr B Analyt Technol Biomed Life Sci 871:236-42. 550

20. Monot, F., J. R. Martin, H. Petitdemange, and R. Gay. 1982. Acetone and Butanol Production by 551

Clostridium acetobutylicum in a Synthetic Medium. Appl Environ Microbiol 44:1318-1324. 552

21. Monschau, N., K. P. Stahmann, H. Sahm, J. B. McNeil, and A. L. Bognar. 1997. Identification of 553

Saccharomyces cerevisiae GLY1 as a threonine aldolase: a key enzyme in glycine biosynthesis. 554

FEMS Microbiol Lett 150:55-60. 555

22. Munger, J., B. D. Bennett, A. Parikh, X. J. Feng, J. McArdle, H. A. Rabitz, T. Shenk, and J. D. 556

Rabinowitz. 2008. Systems-level metabolic flux profiling identifies fatty acid synthesis as a target 557

for antiviral therapy. Nat Biotechnol 26:1179-86. 558

23. Nolling, J., G. Breton, M. V. Omelchenko, K. S. Makarova, Q. Zeng, R. Gibson, H. M. Lee, J. 559

Dubois, D. Qiu, J. Hitti, Y. I. Wolf, R. L. Tatusov, F. Sabathe, L. Doucette-Stamm, P. Soucaille, M. 560

J. Daly, G. N. Bennett, E. V. Koonin, and D. R. Smith. 2001. Genome sequence and comparative 561

analysis of the solvent-producing bacterium Clostridium acetobutylicum. J Bacteriol 183:4823-562

38. 563

24. Qin, J., R. Li, J. Raes, M. Arumugam, K. S. Burgdorf, C. Manichanh, T. Nielsen, N. Pons, F. 564

Levenez, T. Yamada, D. R. Mende, J. Li, J. Xu, S. Li, D. Li, J. Cao, B. Wang, H. Liang, H. Zheng, Y. 565

Xie, J. Tap, P. Lepage, M. Bertalan, J. M. Batto, T. Hansen, D. Le Paslier, A. Linneberg, H. B. 566

Nielsen, E. Pelletier, P. Renault, T. Sicheritz-Ponten, K. Turner, H. Zhu, C. Yu, M. Jian, Y. Zhou, 567

Y. Li, X. Zhang, N. Qin, H. Yang, J. Wang, S. Brunak, J. Dore, F. Guarner, K. Kristiansen, O. 568

Pedersen, J. Parkhill, J. Weissenbach, P. Bork, and S. D. Ehrlich. A human gut microbial gene 569

catalogue established by metagenomic sequencing. Nature 464:59-65. 570

25. Rabinowitz, J. D., and E. Kimball. 2007. Acidic acetonitrile for cellular metabolome extraction 571

from Escherichia coli. Anal Chem 79:6167-73. 572

26. Ragsdale, S. W. 2003. Pyruvate ferredoxin oxidoreductase and its radical intermediate. Chem 573

Rev 103:2333-46. 574

27. Seedorf, H., W. F. Fricke, B. Veith, H. Bruggemann, H. Liesegang, A. Strittmatter, M. Miethke, 575

W. Buckel, J. Hinderberger, F. Li, C. Hagemeier, R. K. Thauer, and G. Gottschalk. 2008. The 576

genome of Clostridium kluyveri, a strict anaerobe with unique metabolic features. Proc Natl 577

Acad Sci U S A 105:2128-33. 578

28. Senger, R. S., and E. T. Papoutsakis. 2008. Genome-scale model for Clostridium acetobutylicum: 579

Part I. Metabolic network resolution and analysis. Biotechnol Bioeng 101:1036-52. 580

29. Tang, Y. J., S. Yi, W. Q. Zhuang, S. H. Zinder, J. D. Keasling, and L. Alvarez-Cohen. 2009. 581

Investigation of carbon metabolism in "Dehalococcoides ethenogenes" strain 195 by use of 582

isotopomer and transcriptomic analyses. J Bacteriol 191:5224-31. 583

30. Turnbaugh, P. J., and J. I. Gordon. 2009. The core gut microbiome, energy balance and obesity. J 584

Physiol 587:4153-8. 585

31. Yuan, J., B. D. Bennett, and J. D. Rabinowitz. 2008. Kinetic flux profiling for quantitation of 586

cellular metabolic fluxes. Nat Protoc 3:1328-40. 587

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

28

32. Yuan, J., C. D. Doucette, W. U. Fowler, X. J. Feng, M. Piazza, H. A. Rabitz, N. S. Wingreen, and J. 588

D. Rabinowitz. 2009. Metabolomics-driven quantitative analysis of ammonia assimilation in E. 589

coli. Mol Syst Biol 5:302. 590

591

592

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

29

593

List of Figures and Tables: 594

Figure 1: Glycolysis and the rapid interchange between the carboxyl group of pyruvate and CO2. 595

Figure 2: A complete, bifurcated TCA cycle in C. acetobutylicum 596

Figure 3: One carbon metabolism in C. acetobutylicum 597

Figure 4: Quantitation of fluxes in central metabolism. 598

599

Supplementary Figure 1: The Entner-Doudoroff pathway is inactive in C. acetobutylicum 600

Supplementary Figure 2: The pentose phosphate pathway in C. acetobutylicum 601

Supplementary Figure 3: Aerobically or anaerobically grown Escherichia coli does not show 602

the interchange between the carboxyl group in pyruvate and atmospheric CO2 603

Supplementary Figure 4: The arginine biosynthesis pathway does not operate in reverse to 604

produce glutamate 605

Supplementary Figure 5: C. acetobutylicum shows Re-citrate synthase activity 606

Supplementary Figure 6: C. acetobutylicum fermentation profiles during exponential phase. 607

Supplementary Figure 7: Reactions of the ordinary differential equation model. 608

Supplementary Figure 8: Model fitting to the U-13

C –glucose dynamic labeling data 609

Supplementary Figure 9: The distribution of the one thousand sets of model parameters that 610

that quantitatively describe the laboratory data. 611

612

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

31

616

Figure Legends 617

Figure 1: Glycolysis and the rapid interchange between the carboxyl group of pyruvate and CO2. 618

(A) Overview of active and inactive pathways. Glycolysis operates normally through 619

phosphoenolpyruvate with the Entner-Doudoroff pathway inactive (see Supplementary Figure 1). 620

The reaction catalyzed by pyruvate ferredoxin oxidoreductase (PFOR) is partially, but not fully, 621

reversible. GAP, glyceraldehyde-3-phosphate; DHAP, dihydroxyacetone-phosphate. 622

(B) Dynamic incorporation of uniformly 13

C-labeled glucose (100%) into glycolysis 623

intermediates. Glycolysis intermediates through phosphoenolpyruvate labeled rapidly and 624

completely. Pyruvate, however, appeared in roughly equimolar amounts in its fully labeled form 625

and in an unexpected form with two 13

C-carbons. Environmental CO2 was ~99% non-labeled. 626

The x axis represents minutes after the switch from unlabeled to U-13

C-glucose media and the y 627

axis represents the fraction of the observed compound of the indicated isotopic form. 628

(C) Steady state labeling patterns of phosphoenolpyruvate and pyruvate obtained from cells 629

grown in 3-13

C-glucose (100%), 4-13

C-glucose (100%) or 1,2-13

C-glucose (100%). In 3- or 4-630

13C-glucose, about half of phosphoenolpyruvate was labeled but only about a quarter of pyruvate 631

was labeled. In contrast, growth in 1,2-13

C-glucose resulted in identical labeling of 632

phosphoenolpyruvate and pyruvate. Environmental CO2 was ~99% non-labeled. These results 633

indicate that 13

C-label in pyruvate is specifically lost from the carboxyl carbon. 634

(D) The fraction of 1-13

C-pyruvate increased with increasing amounts of NaH13

CO3 added into 635

the media. Cells were fed unlabeled glucose throughout and labeling of upstream glycolysis 636

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

32

intermediates was minimal or nonexistent (not shown). This experiment was performed in liquid 637

closed-vessel cultures. The data, in conjunction with (B) and (C), indicates exchange of the 638

carboxyl carbon of pyruvate with carbon dioxide. 639

(E) U-13

C-acetate was assimilated and incorporated into acetyl-phosphate and acetyl-CoA. 640

Pyruvate, however, remained unlabeled. The data suggests that the reaction catalyzed by PFOR 641

is not fully reversible. The error bars in (B) through (E) show ± s.d. (n = 2 to 4 independent 642

experiments). 643

644

Figure 2: A complete, bifurcated TCA cycle in C. acetobutylicum 645

(A) The diagram represents the proposed bifurcated TCA cycle in C. acetobutylicum. α-646

Ketoglutarate is produced from oxaloacetate and acetyl-CoA via citrate. Succinate can be 647

produced reductively from fumarate or oxidatively from α-ketoglutarate. Gray boxes show the 648

fate of the carbons in the incoming acetyl group from acetyl-CoA, and dotted boxes from the 649

carboxyl group of pyruvate. The unusual stereospecificity of citrate synthesis was confirmed by 650

MS/MS analysis (Supplementary Figure 5). 651

(B) and (C) show the dynamic incorporation of U-13

C-glucose (100%) and U-13

C-acetate (in the 652

presence of non-labeled glucose) into TCA metabolites and glutamate. The x axis represents 653

minutes after the switch from unlabeled to 13

C-labeled media and the y axis represents the 654

fraction of the observed compound of the indicated isotopic form. There was no detectable 655

labeling of oxaloacetate, malate or fumarate in the U-13

C-acetate experiments (not shown). 656

These results are consistent with a bifurcated TCA cycle in which oxaloacetate flows to 657

succinate both through citrate/α-ketoglutarate and via malate/fumarate as shown in (A). 658

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

33

(D) and (E) show the long-term labeling patterns of TCA metabolites when cells are grown in 659

glucose minimal media supplemented with U-13

C-aspartate or with U-13

C-glutamate. This data 660

corroborates the results obtained in (B) and (C) and the existence of a bifurcated TCA cycle. 661

In all experiments, the environmental CO2 comprised 5% of the anaerobic gaseous environment 662

and was ~99% non-labeled. In (B) through (D), the error bars show ± s.d. (n = 2 to 4 independent 663

experiments). 664

665

Figure 3: One carbon metabolism in C. acetobutylicum 666

(A) Proposed network of one carbon metabolism in C. acetobutylicum. Blue arrows highlight the 667

major production route for glycine, serine, and one-carbon units (C1-folates). The fate of the 668

carbons originating from pyruvate is highlighted by gray and dotted boxes. 669

(B) Dynamic incorporation of U-13

C-glucose (100%) into the aminoacids serine, glycine and 670

threonine. The labeling patterns observed in glycine indicate that its primary route of production 671

is via threonine and not serine. 672

(C) The synthesis of glycine from threonine was confirmed by growing cells on U-13

C glucose 673

plus non-labeled aspartate and observing that both the threonine and glycine pools are largely 674

non-labeled, while the serine pool remains largely labeled. 675

(D) Cells grown in 1,2-13

C glucose (100%) showed less than 5% label in C1 units, even though 676

the precursor carbon in glycine (methylene group, highlighted in gray in (A)) is ~50% labeled. 677

(E) Correlation between the labeled fractions of the carboxylic acid carbon of pyruvate and the 678

labeled fractions of C1 units across diverse labeling experiments. The addition of unlabeled 679

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

34

aspartate to cells growing in U-13

C-glucose (100%), which results in production of unlabeled 680

glycine (C), does not affect labeling of C1 units. 681

(F) Cells grown with increasing concentrations of NaH13

CO3 showed increasing labeling of C1 682

units that followed closely the labeling in the carboxyl group of pyruvate but not the labeling of 683

CO2 present in the media. Fraction of media CO2 labeled was determined based on labeling of 684

CO2 assimilated into pyrimidines. 685

In (B) through (E), the environmental CO2 comprised 5% of the anaerobic gaseous environment 686

and was ~99% non-labeled. In (F), the experiments were performed in liquid closed-vessel 687

cultures to minimize interchange between atmospheric 12

CO2 and NaH13

CO3. 688

689

Figure 4: Quantitation of fluxes in central metabolism. 690

(A) Ordinary differential equation (ODE) model fitting (lines) to the U-13

C-glucose dynamic 691

labeling data (error bars) for three representative metabolites. Complete results are in 692

Supplementary Figure 8. 693

(B) Metabolic fluxes identified from the ODE model. Arrow sizes indicate absolute values (in 694

logarithmic scale) of net fluxes. Fluxes shown are median values of one thousand sets of 695

identified fluxes, whose distributions are plotted in Supplementary Figure 9. The flux from 696

succinyl-CoA into succinate is a combination of the flux through succinyl-CoA synthetase 697

(~25% median contribution) and the fluxes through the methionine and lysine biosynthesis 698

pathways that are coupled to the conversion of succinyl-CoA into succinate (~75% median 699

contribution). Hexose-P, combined pools of glucose-1-phosphate, glucose-6-phosphate and 700

fructose-6-phosphate; FBP, fructose-1,6-bisphophate; DHAP, combined pools of 701

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

35

dihydroxyacetone-phosphate and glyceraldehyde-3-phosphate; 3PG, combined pools of 702

glycerate-3-phosphate and glycerate-2-phosphate; PEP, phosphoenolpyruvate; Pentose-P, 703

combined pools of ribose-5-phosphate, xylulose-5-phosphate and ribulose-5-phosphate; OAA, 704

oxaloacetate; αKG, α-ketoglutarate; SucCoA, succinyl-CoA; Asp, aspartate; Glu, glutamate; 705

Gln, glutamine. 706

707

Supplementary Figure 1: The Entner-Doudoroff pathway is inactive in C. acetobutylicum 708

The Entner-Doudoroff pathway encompasses an alternative set of reactions for the catabolism of 709

glucose to pyruvate. The relative contribution to pyruvate synthesis from the Entner-Doudoroff 710

pathway vs. glycolysis was determined by growing cells in (A) 3-13

C-glucose (100%) or (B) 1,2-711

13C-glucose (100%). The characteristic labeled forms of pyruvate expected from glucose 712

catabolism by the Entner-Doudoroff pathway: 3-13

C-pyruvate when grown on 3-13

C-glucose and 713

1,2-13

C pyruvate when grown on 1,2-13

C-glucose, were not observed. In (A), feeding 3-13

C-714

glucose results in only ~20% of 1-13

C-pyruvate (as opposed to 50%) because of the interchange 715

between the carboxyl group of pyruvate and atmospheric non-labeled CO2 (see Figure 1). The 716

error bars show ± s.d. (n = 3 independent experiments). 717

718

Supplementary Figure 2: The pentose phosphate pathway in C. acetobutylicum 719

(A) Proposed network of pentose phosphate pathway (PPP) reactions in C. acetobutylicum. Gray 720

or dotted boxes are used to highlight the fate of carbons #1 and #2 of glucose. 721

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

U � 1 3 C � g l u c o s e u p t a k e

N a H 1 3 C O 3 ( m M )D Eﾉ;HWﾉWS"aヴ;Iピﾗﾐ 1 � 1 3 C � p y r u v a t e U � 1 3 C � a c e t a t e u p t a k et w o 1 3 C � c a r b o n s

AT w oN o n 7 l a b e l e d 1 3 C 7 c a r b o nt h r e e 1 3 C 7 c a r b o n s S i x 1 3 C 7 c a r b o n syW

ﾉ;ピ┗W"aヴ;Iピﾗﾐ 0 2 0 6 000 . 20 . 40 . 60 . 81 G l u c ﾗs e U 6 P D H A P

~ｷﾏW"ふﾏｷﾐぶ

00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0G l y c e r a t e U 3 P 00 . 20 . 81 0 2 0 4 0 6 0P h ﾗs p h ﾗe ﾐﾗl p y r u ┗a t e 00 . 20 . 8 1 0 2 0 4 0 6 0P y r u ┗a t e~ｷﾏW"ふﾏｷﾐぶ ~ｷﾏW"ふﾏｷﾐぶ

BEヴ┌IデﾗゲWどヱがヶv

A c e t y l i C o AOCH 3 C C o AOCH 2 C C O O HP O 3 H 2P h o s p h o e n o l p y r u v a t eOCH 3 C C O O HP y r u v a t e P F O R C O 2+OCH 3 C P F O RA D PA T PC o A2 e �

G l u c o s e

P F O R

G l u c o s e i 6 PF r u c t o s e i 1 , 6 P3 i P h o s p h o g l y c e r a t e D H A PG A PE ﾐ

t ﾐer ºD ﾗ

ud ﾗ

r ﾗdpath way CyWﾉ;ピ┗W"aヴ;Iピﾗﾐ 00 . 20 .40 . 60 .8 1 3 Ë 1 3 C 4 Ë 1 3 C 1

が2 Ë 1 3 CP h

ﾗs p hﾗe ﾐﾗl p y r u ┗a t e 00 . 20 .40 . 60 .8 1 3 Ë 1 3 C 4 Ë 1 3 C 1

が2 Ë 1 3 CP y r u ┗a t e N o n Ü l a b e l e do n e 1 3 C Ü c a r b o nt w o 1 3 C Ü c a r b o n s

00 . 20 .40 . 60 .8 1 A c e t y lð P A c e t y lð C o A P y r u v a t e

00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0

00 . 20 .40 . 60 .8 0 2 5 5 0 7 5 1 0 0 1 2 5 ﾉ;HWﾉWS"aヴ;Iピﾗﾐ

Figure 1

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

C O 2CH 2 C O O HCH 2 C O O H CH 2 C O O HC H 2C SO C o AS u c c i n y l � C o ACH 2 C O O HC C O O HOH CH 2 C O O HC i t r a t e

CH 2 C O O HC H 2C H C O O HNH 2 G l u t a m a t eC O 2CH 2 C O O HCH C O O HC H C O O HOH I s o c i t r a t eOCH 3 C C o AA c e t y l � C o AOCH 3 C C O O HP y r u v a t e

CH 2 C O O HCH 2 C O O HCH C O O HCH C O O HF u m a r a t eCH 2 C O O HC H C O O HOH M a l a t e OCH 2 C C O O HC O O HO x a l o a c e t a t eS u c c i n a t e

CH 2 C H C O O HC O O HNH 2 A s p a r t a t e G l u c o s eA c e t a t eC O 2CH 2 C O O HC H 2C C O O HO

ü� k e t o g l u t a r a t e

B 00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0P y r u v a t e 00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0O x a l o a c e t a t e 00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0M a l a t e 00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0F u m a r a t e00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0S u c c i n a t e 00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0ü� k e t o g l u t a r a t e00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0C i t r a t e 00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0G l u t a m a t eyW

ﾉ;ピ

┗W"aヴ

;Iピ

ﾗﾐ T i m e ( m i n )U ¤ 1 3 C ¤ g l u c o s e u p t a k e

00 . 0 50 . 10 . 1 50 . 20 . 2 5 0 2 0 4 0 6 0C i t r a t e 00 . 0 50 . 10 . 1 50 . 20 . 2 5 0 2 0 4 0 6 0ü � k e t o g l u t a r a t e 00 . 0 50 . 10 . 1 50 . 20 . 2 5 0 2 0 4 0 6 0G l u t a m a t e 00 . 0 50 . 10 . 1 50 . 20 . 2 5 0 2 0 4 0 6 0S u c c i n a t eT i m e ( m i n )yWﾉ;

ピ┗W

"aヴ;I

ピﾗﾐ

C U ¤ 1 3 C ¤ a c e t a t e u p t a k eyW

ﾉ;ピ

┗W"aヴ

;Iピ

ﾗﾐ

D E U ¤ 1 3 C ¤ g l u t a m a t e u p t a k eU ¤ 1 3 C ¤ a s p a r t a t e u p t a k e N o n ¼ l a b e l e d o n e 1 3 C ¼ c a r b o n t w o 1 3 C ¼ c a r b o n st h r e e 1 3 C ¼ c a r b o n s f o u r 1 3 C ¼ c a r b o n sgv e 1 3 C ¼ c a r b o n s00 . 20 . 40 . 60 . 81 00 . 20 . 40 . 60 . 81

A Figure 2

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

00 . 20 . 40 . 60 . 81 0 2 0 4 0 6 0S e r i n e 0 2 0 4 0 6 0t h r e o n i n eyWﾉ;ピ┗W"aヴ;Iピﾗﾐ

~ｷﾏW"ふﾏｷﾐぶ

U � 1 3 C � g l u c o s e u p t a k eyWﾉ;ピ┗W"aヴ;Iピﾗﾐ

BC D N o n % l a b e l e d o n e 1 3 C % c a r b o nt w o 1 3 C % c a r b o n s t h r e e 1 3 C % c a r b o n s00 . 10 . 20 . 30 . 40 . 50 . 60 . 7 U 1 3 C"

G l uIo s e U 1 3 C

"G l u

Io s e "+ "A s p a r t a t e 1 , 2"

1 3 C"G l u

Io s e 3 O 1 3 C"

G l uIo s e 4 O 1 3 C

"G l u

Io s eﾉ;HWﾉWS"aヴ;Iピﾗﾐ U R13 C Rglucose 1 ,2 R13 C Rglucose 3 R13 C Rglucose 4 R13 C RglucoseU R13 C Rglucose+ Aspartate C a r b o x y l g r o u p o fP y r u v a t eC 1 u n i t s

U � 1 3 C � g l u c o s e +n o n � l a b e l e d a s p a r t a t eyWﾉ;ピ┗W"aヴ;IピﾗﾐE 1 , 2 { 1 3 C { g l u c o s e u p t a k eA 00 . 20 . 40 . 60 . 8 1 0 2 0 4 0 6 0G ly I i n e

00 . 20 . 40 . 60 . 81 S e r i n e " G ly I i n e " T h r e o n i n e F N a H 1 3 C O 3"ふm M ぶ

00 . 10 . 20 . 30 . 40 . 50 . 60 . 7 0 5 1 0 1 5 2 0 2 5C a r b o x y l g r o u p o f p y r u v a t eC 1 u n it sC O 200 . 20 . 40 . 60 . 81 C 1"u n i t s S e r i n e G l y Ii n eH 3N H 2C H C H C O O HO HCH 3 T h r e o n i n eN H 2CH 2 C O O HG ly c i n e C O 2 NT H FN H 2CH 2 C H C O O HO H S e r i n e 5 , 1 0 ó m e t h y le n e ó T H F1 0 ó f o r m y ló T H F T H FOCH 3 C C o AA c e t y ló C o A

A c e t a l d e h y d eA s p a r t a t e +3 ó p h o s p h o g ly c e r a t eO HC H 2 C H C O O HH 2 P O 3G l u c o s e C O 2B i o s y n t h e

ピ cm e t a b o l is mOCH 3 C C O O HP y r u v a t e

OCH 3 C C O O HP y r u v a t e 5 , 1 0 ó m e t h y le n e ó T H Fﾉ;HWﾉWS"aヴ;Iピﾗﾐ

Figure 3

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

B concentration(mM) F r u c t o s e � 1 , 6 P α� K e t o g l u t a r a t e t i m e ( m i n )t i m e ( m i n )t i m e ( m i n ) S u c c i n a t eA 051 01 52 02 53 03 54 04 5 2 01 010 00 . 511 . 522 . 53 2 01 010 00 . 10 . 20 . 30 . 40 . 50 . 60 . 7 2 01 010n o n 7 l a b e l e d o n e 1 3 C 7 c a r b o n t w o 1 3 C 7 c a r b o n st h r e e 1 3 C 7 c a r b o n s f o u r 1 3 C 7 c a r b o n s s i x 1 3 C 7 c a r b o n sE x t r a c e l l u l a r g l u c o s eIn t r a c e l l u l a r g l u c o s eH e x o s e Y PF B PD H A P3 P GP E PP y r u v a t eA c e t y l Y C o A A l a n i n eP e n t o s e Y PC i t r a t e

gK GM a l a t eO A A

G l y c o s y l a t i o n o rp o l y m e r i z a t i o nF a t t y a c i d sA s pO t h e r a m i n o a c i d s a n dp y r u v a t e e x c r e t i o n G l uA c i d p r o d u c t i o nS u c C o AS u c c i n a t eF u m a r a t e G l nO r n i t h i n eC i t r u l l i n eA r g i n i n eM e t h i o n i n e L y s i n eO t h e r a m i n o a c i d sa n d p y r i m i d i n e s

M e t a b o l i t e f l u x :< 1 m M / m i n1 0 m M / m i n1 0 0 m M / m i nP r o t e i n s

Figure 4

S u c c i n a t e e x c r e t i o n

on May 22, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

JOURNAL OF BACTERIOLOGY, Dec. 2011, p. 6805 Vol. 193, No. 230021-9193/11/$12.00 doi:10.1128/JB.06216-11Copyright © 2011, American Society for Microbiology. All Rights Reserved.

AUTHOR’S CORRECTION

Systems-Level Metabolic Flux Profiling Elucidates a Complete, BifurcatedTricarboxylic Acid Cycle in Clostridium acetobutylicum

Daniel Amador-Noguez, Xiao-Jiang Feng, Jing Fan, Nathaniel Roquet,Herschel Rabitz, and Joshua D. Rabinowitz

Lewis Sigler Institute for Integrative Genomics, Princeton University, Princeton, New Jersey, and Department of Chemistry,Princeton University, Princeton, New Jersey

Volume 192, no. 17, p. 4452–4461, 2010. Page 4453: The following sentence should appear at the end of paragraph 6 in Materialsand Methods. “All 13C-labeling patterns reported for oxaloacetate were inferred from LC-MS analysis of aspartate.”

6805