SEARCH

SEARCH BY CITATION

Keywords:

  • 6dEB;
  • acyl-CoA;
  • erythromycin;
  • Escherichia coli;
  • LC-MS/MS;
  • PrpE

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Aims:  This paper utilized quantitative LC-MS/MS to profile the short-chain acyl-CoA levels of several strains of Escherichia coli engineered for heterologous polyketide production. To further compare and potentially expand the levels of available acyl-CoA molecules, a propionyl-CoA synthetase gene from Ralstonia solanacearum (prpE-RS) was synthesized and expressed in the engineered strain BAP1.

Methods and Results:  Upon feeding propionate, the engineered E. coli strains had increased the levels of both propionyl- and methylmalonyl-CoA of 6- to 30-fold and 3·7- to 6·8-fold, respectively. Expression of prpE-RS resulted in no significant increases in acetyl-, butyryl- and propionyl-CoA when fed the corresponding substrates (sodium acetate, butyrate or propionate). More interesting, however, were the results from strain BAP1 engineered for native prpE overexpression, which indicated increases in the same range of acyl-CoA formation.

Conclusions:  The increased acyl-CoA levels across the strains profiled in this study reflect the genetic modifications implemented for improved polyketide production and also indicate flexibility of the native PrpE.

Significance and Impact of the Study:  The results provide direct evidence of enhanced acyl-CoA levels correlating to those strains engineered for polyketide biosynthesis. This information and the inherent flexibility of the native PrpE enzyme support future efforts to characterize, engineer and extend acyl-CoA precursor supply for additional heterologous biosynthetic attempts.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Metabolite profiling poses a challenging analytical problem because of the vast array of chemically diverse metabolites found in cellular systems. Additionally, the rapid turnover rate and low concentration of particular metabolite pools require rigorous methods for extraction, quenching and analysis to obtain a more complete and accurate snapshot of intracellular dynamics (Villas-Boas et al 2005a; Dettmer et al. 2007). Of the estimated 300 mmol l−1 metabolite concentration within Escherichia coli, a large portion is composed of only a small number of metabolites, leaving many classes representing <1% of the total pool (Buccholz et al. 2001; Bennett et al. 2009). Included in the <1% are acyl-CoA metabolites used in a variety of metabolic processes and as building blocks for interesting polyketide natural products. The analysis of these metabolites serves as an important complement to the metabolic engineering attempts to improve complex polyketide formation, especially because success (or failure) is often measured by final product titres without any knowledge of the metabolic processes that support production. Additionally, knowledge of intracellular metabolite concentrations may help direct future metabolic engineering approaches.

As such, targeted metabolite analysis provides an apt platform for testing hypothesis-driven studies that focus on specific classes of important metabolites and can generate a more practical and directed assessment of isolated metabolic perturbations or genetic modifications. Much work is still being conducted to optimize and individualize extraction and analytical protocols for each metabolite class, but methods have been developed for a broad range of metabolites with each seeking to allow rapid, repeatable and accurate snapshots of metabolic flux (Villas-Boas et al. 2005b; Dunn et al. 2005; Yang et al. 2009; Bennett et al. 2009; Yanes et al. 2011). Using a targeted metabolite approach, several studies have sought to quantify both short- and long-chain acyl-CoA molecules to better understand the effects of various genetic or nutrient perturbations, such as the dynamics of varying carbon sources (Yang et al. 2009), enzyme regulation (Boynton et al. 1994), or for the production of a variety of interesting products including biopolymers (Valentin et al. 2000) and complex natural products (Park et al. 2007; Se Jong et al. 2008).

This study focused on the short-chain acyl-CoA molecules utilized as building blocks for the production of polyketide products. More specifically, the analysis focused on those hosts and enzymes designed for the production of 6-deoxyerythronolide B (6dEB), the macrocyclic precursor to the antibiotic erythromycin. The biosynthesis of 6dEB requires one molecule of propionyl-CoA and six molecules of (2S)-methylmalonyl-CoA and is catalysed by an enzyme complex termed the deoxyerythronolide B synthase. Further processing of this polyketide molecule by glycosylation, hydroxylation and methylation generates the antibiotic erythromycin A. For reasons of technical and scientific convenience, production of 6dEB was established using E. coli as a heterologous host (Pfeifer et al. 2001). During heterologous biosynthesis, the intracellular supply of propionyl-CoA and (2S)-methylmalonyl-CoA is generated from the exogenous supplementation of propionate. As such, propionate feeding studies were conducted here with a targeted metabolite analysis of intracellular acyl-CoA across a series of E. coli strains engineered to support complex polyketide formation.

In addition, the analytical method developed for specific acyl-CoA detection was directed towards a comparison between native and heterologous versions of the propionyl-CoA synthetase (PrpE). Escherichia coli polyketide production was first engineered by overexpressing the native prpE gene to convert exogenously fed propionate to propionyl-CoA. In an attempt to expand the acyl-CoA substrate pools available for complex natural product formation, a heterologous propionyl-CoA synthetase was introduced to E. coli strain BAP1. PrpE enzymes from Ralstonia solanacearum and Salmonella choleraesuis previously demonstrated substrate flexibility, generating acetyl-, propionyl-, butyryl- and acrylyl-CoA molecules in vitro (Rajashekhara and Watanabe 2004). Here, we sought to utilize the R.  solanacearum PrpE (PrpE-RS) for in vivo substrate variation. In doing so, we compared the intracellular acyl-CoA formation capabilities of the native and heterologous PrpE enzymes.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Chemicals and supplies

Acetyl-, butyryl-, glutaryl-, malonyl-, methylmalonyl-, propionyl- and succinyl-CoA were purchased from Sigma-Aldrich (St Louis, MO, USA) and stored in either ddH2O or 20 mmol l−1 ammonium acetate at −20°C for the duration of the study. HPLC-grade methanol and acetonitrile were also purchased from Sigma-Aldrich as were sodium acetate, sodium propionate and sodium butyrate. Ammonium acetate, ammonium formate, glacial acetic acid, chloramphenicol, kanamycin, carbenicillin and IPTG were purchased from Fisher Scientific (Pittsburgh, PA, USA).

Bacterial strains and plasmids

Table 1 presents the bacterial strains and plasmids used in this study. The pJexpress411-prpE-RS plasmid was synthesized by DNA 2·0 (Menlo Park, CA, USA) based upon the NCBI reference sequence number NP_5196391. Plasmid pJWA1 was constructed by transferring the prpE-RS insert from pJexpress411-prpE-RS into the MCS2 of pACYCDuet-1 (Novagen, Madison, WI) using NdeI and XhoI. Transformations, SDS-PAGE analysis and other standard molecular biology techniques were carried out as described by Sambrook et al. (1989).

Table 1.   Strains and plasmids used in the study
StrainSourceModification
BAP1(Pfeifer et al. 2001)BL21(DE3); ΔprpRBCD: T7prom-sfp-T7prom-prpE
BAB2(Boghigian et al. 2011a,b)BAP1; Δsbm-ygfDGH::FRT
TB3(Zhang et al. 2010)BAP1; ΔygfH::kan
YW22(Boghigian et al. 2011a,b)JM109(DE3); araA::T7prom-dxs-T7term-T7prom-idi-T7term-T7prom-ispDF-T7term
BL21(DE3)NovagenF ompT hsdSB (rB, mB) gal dcm λ(DE3)
Plasmid
 pJexpress411-prpE-RS(DNA 2·0)pJexpress411 [Kan-resistant]; T7prom-prpE-RS-T7term
 pJWA1(this study)pACYCDuet-1; T7prom-prpE-RS-T7term
 pACYCDuet-1(Novagen)[Cm-resistant]

Culture conditions

Strains were made electrocompetent and transformed with pJexpress411-prpE-RS, pJWA1 or pACYCDuet-1; resulting transformants were stored as glycerol stocks at −80°C. Analytical experiments began with 3 ml cultures inoculated from glycerol stocks and grown overnight at 37°C in LB medium with appropriate antibiotics (included at 34 [chloramphenicol], 50 [kanamycin] and 100 [carbenicillin] μg ml−1). The overnight cultures were used to inoculate 15 ml LB medium in baffled Erlenmeyer flasks supplemented with 20 mmol l−1 sodium acetate, sodium propionate or sodium butyrate and induced with 100 μmol l−1 IPTG as indicated. Cultures were inoculated to an OD600 nm = 0·1 in triplicate. The production cultures were allowed to incubate at 22°C for 24 h before extraction and subsequent LC-MS/MS analysis. Cell density was measured spectrophotometrically at 600 nm prior to extraction.

Extraction procedure

A sample (3 ml) of the final culture was centrifuged for 10 min at 1700 g and 4°C. Pellets were resuspended in 1 ml ddH2O and centrifuged for 1 min at 9300 g and 4°C. The resulting cell pellet was subjected to a cell lysis extraction protocol adapted from Bennett et al. (2009). Pellets were rapidly quenched and extracted to a final volume of 130 μl with 45 : 45 : 10 acetonitrile/methanol/H2O + 0·1% glacial acetic acid at −20°C and spiked with 10 μmol l−1 glutaryl-CoA as an internal standard. The resuspended extract was incubated on ice with intermittent vortexing for 15 min. An equal molar volume of ammonium hydroxide was added postincubation to neutralize the acetic acid, and each extract was centrifuged for 3 min at 15 700 g and 4°C, transferred to a new 1·5 ml microfuge tube and centrifuged for 5 min at 15 700 g and 4°C. The clarified extract (10 μl) was injected for HPLC-MS/MS analysis.

LC-MS/MS development and quantification

MS and MS/MS conditions (declustering potential, entrance potential, collision energy and product ions) were optimized in positive ion mode (5·5 kV) for each acyl-CoA by direct injection of 100 μmol l−1 standards diluted in 50 : 50 H2O/MeOH + 5 mmol l−1 ammonium formate to an Applied Biosystems 3200 Q-Trap triple quadrupole (AbSciEx, Foster City, CA) at a flow rate of 5 μl min−1 with a nebulizer gas pressure of 83 kPa, auxiliary gas pressure of 35 kPa, source temperature of 100°C and curtain gas pressure of 69 kPa. The optimal parent/product ion pairs (m/z) in Table 2 were used for further study and were determined from the loss of 507 Da from the positive ion precursor as previously described (Gao et al. 2007). Dry nitrogen was used as desolvation gas.

Table 2.   Parent and product ion pairs used in LC-MS/MS analysis for acyl-CoAs in positive ion mode [M+ - H]
Acyl-CoAParent ionProduct ion
Acetyl-CoA810303
Butyryl-CoA838331
Malonyl-CoA854347
Methylmalonyl-CoA868361
Propionyl-CoA824317
Succinyl-CoA868361
Glutaryl-CoA882375

Chromatographic elution was optimized on a GL Sciences Inertsil ODS-3 C18 analytical column (4·6 × 150 mm, 3 μm; Torrance, CA, USA). The extracted cultures and standards were eluted at a flow rate of 400 μl min−1 with a gradient of 10–22·5% B for 30 min, to 100% B in 5 min, and re-equilibrated and washed at 10% B for 20 min at 500 μl min−1. Buffer A was composed of 20 mmol l−1 ammonium acetate in water (pH 7·4), and Buffer B was composed of 20 mmol l−1 ammonium acetate in HPLC-grade methanol. Gas and temperature conditions were further optimized for LC-MS/MS to obtain maximum signal intensity at curtain gas pressure of 207 kPa, nebulizer gas pressure of 345 kPa, auxiliary gas pressure of 207 kPa and source temperature of 350°C. Acyl-CoAs were eluted in the following order: malonyl-CoA (21·5 min), succinyl-CoA (26·0 min), methylmalonyl-CoA (28·0 min), glutaryl-CoA (31·0 min), acetyl-CoA (39·0 min), propionyl-CoA (40·5 min) and butyryl-CoA (41·5 min) (Fig. 1a).

image

Figure 1.  LC chromatogram of (a) acyl-CoA standards and (b) BAP1/pJWA1 induced with 100 μmol l−1 IPTG and fed 20 mmol l−1 sodium propionate.

Download figure to PowerPoint

Acyl-CoAs were detected in the multiple reaction monitoring (MRM) mode with the m/z parent > m/z daughter (pairs listed in Table 2) in positive ion mode. A dwell time of 75 ms was used for each MRM transition for the 45-min elution. LC-ESI-MS/MS chromatograms were analysed using analyst software ver. 1·5·1 (AbsciEx) and manually integrated after 2× Gaussian smoothing. Acyl-CoA areas were normalized to the glutaryl-CoA internal standard area, and concentrations were determined from an intra-run generated calibration curve of 0·5, 0·75, 1, 2·5, 5, 7·5 and 10 μmol l−1 standard acyl-CoA concentrations with 10 μmol l−1 used for glutaryl-CoA. Concentration of a given acyl-CoA was calculated by correlating the area/area internal standard to intra-run calibration curves. The total nmols of a given acyl-CoA was further normalized to mg DCW by first calculating total acyl-CoA amount (using the final extraction volume) and then using the previously determined correlation (data not shown) of 1 OD600 nm = 0·52 g dry cell weight (DCW) l−1.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Acyl-CoA strain profiling

After establishing an intracellular LC-MS/MS method, five strains with varying genetic modifications (Table 1) were tested to profile and compare the level of five acyl-CoA molecules under different induction and feed conditions. Four conditions were tested: (i) 20 mmol l−1 propionate and 100 μmol l−1 IPTG (ii) propionate only (iii) IPTG only and (iv) no propionate or IPTG. Upon feeding with propionate and IPTG, it was observed that BAB2, BAP1 and TB3 showed fold increases between 6 and 30 for propionyl-CoA, c. 3·7–6·8 for methylmalonyl-CoA and 4–6·8 for butyryl-CoA with respect to the un-engineered controls YW22 (a K12 strain of E. coli) and BL21(DE3) (the parent strain of BAB2, BAP1 and TB3). Acetyl-CoA levels were comparatively lower for BAB2, BAP1 and TB3 strains with the addition of propionate, while malonyl-CoA showed no significant differences across all conditions. No significant differences in propionyl- or methylmalonyl-CoA levels were observed with only propionate or only IPTG conditions or in the case of no propionate and IPTG (Fig. 2).

image

Figure 2.  Acyl-CoA profiles for Escherichia coli strains under the following conditions: (a) 20 mmol l−1 sodium propionate and 100 μmol l−1 IPTG; (b) 20 mmol l−1 sodium propionate; (c) 100 μmol l−1 IPTG; and (d) no sodium propionate and no IPTG. BAB2 inline image, BAP1 inline image, TB3 inline image, YW22 inline image, BL21(DE3) inline image.

Download figure to PowerPoint

Effect of PrpE-RS and native E. coli PrpE

SDS-PAGE was first used to confirm production of PrpE-RS from pJexpress411-prpE-RS and pJWA1 (Fig. 3). BAP1/pJWA1 showed statistically significant fold increases of 1·6, 15·0 and 8·7 of acetyl-, butyryl- and propionyl-CoA, respectively, when fed with the corresponding substrates at 20 mmol l−1 and induced with 100 μmol l−1 IPTG when compared to the uninduced control (Figs 1b and 4a). Significant increases in the remaining acyl-CoA molecules were either not observed or not detected.

image

Figure 3.  SDS-PAGE of BAP1 containing induced (+) and uninduced (−) pJexpress411-prpE-RS (BAP1/pJex) or pJWA1 (BAP1/pJWA1). PrpE-RS (68·7 kDa) is indicated.

Download figure to PowerPoint

image

Figure 4.  Fold increases comparing (a) induced (with 100 μmol l−1 IPTG) and uninduced BAP1/pJWA1 with respective acyl-CoA precursors fed at 20 mmol l−1 and (b) BAP1/pJWA1 and BAP1/pACYCDuet-1 with respective acyl-CoA precursors fed at 20 mmol l−1 and induced with 100 μmol l−1 IPTG. Acetyl-CoA; Acetate Feed inline image, Butyryl-CoA, Butyrate Feed inline image, Propionyl-CoA; Propionate Feed inline image.

Download figure to PowerPoint

When compared to the BAP1/pACYCDuet-1 empty vector control, BAP1/pJWA1 had fold increases of 2·1, 1·2 and 1·3 for acetyl-, butyryl and propionyl-CoA, respectively, when the corresponding substrates were fed as earlier (Fig. 4b). Butyryl- and propionyl-CoA fold increases were statistically significant but not reflective of the fold increases observed between induced and uninduced BAP1/pJWA1.

To further probe the fold increases originally observed, a third control was tested where pJWA1 and pACYC were introduced to BL21(DE3) and compared to BAP1 with no additional plasmids. In this comparison, BAP1 showed fold increases in butyryl-CoA of 7·8 and 6·0, while increases of 13·7 and 2·2 for propionyl-CoA were observed. Comparisons were made with respect to BL21(DE3)/pJWA1 and BL21(DE3)/pACYC, respectively, when the corresponding substrates were fed (Fig. 5). No fold increases in acetyl-CoA levels in BAP1 compared to BL21(DE3)/pACYC or BL21(DE3)/pJWA1 were observed in the acetate, propionate or butyrate feed conditions. Increases in methylmalonyl-CoA were observed in both the butyrate and propionate feed conditions (1·4–8·5 fold); in addition, fold increases in propionyl-CoA were observed in all feed conditions ranging from 1·6- to 13·7-fold.

image

Figure 5.  Acyl-CoA profiles comparing BL21(DE3)/pJWA1, BL21(DE3)/pACYCDuet-1 and BAP1 all induced with 100 μmol l−1 IPTG and fed with (a) 20 mmol l−1 sodium acetate, (b) 20 mmol l−1 sodium propionate and (c) 20 mmol l−1 sodium butyrate. BL21(DE3)/pJWA1 inline image, BL21(DE3)/pACYCDuet-1 inline image, BAP1 inline image.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The acyl-CoA profiles observed in this study suggest that the modifications present in the engineered strains BAB2, BAP1 and TB3 provide better support for polyketide production as increased levels of both propionyl- and methylmalonyl-CoA were observed upon induction and propionate feeding (Fig. 2). By improving substrate availability for polyketide production, the potentially rate-limiting step of substrate loading may be alleviated, and the observed increases in acyl-CoA concentrations further confirm the previous metabolic engineering approaches for increased 6dEB production including overexpression of the native prpE in BAP1 (and BAB2 and TB3), the sbm-ygfDGH deletion in BAB2 and the ygfH deletion in TB3 (Pfeifer et al. 2001; Zhang et al. 2010; Boghigian et al. 2011a). Each of these modifications was designed to increase the intracellular levels of biosynthetic substrates. PrpE activity allows exogenous propionate to be converted to propionyl-CoA; whereas strains BAB2 and TB3 feature gene deletions predicted to eliminate competing side reactions for intracellular propionyl-CoA and methylmalonyl-CoA. The engineered strains revealed increased levels in required substrates for 6dEB production, which matched the previously observed improvements in 6dEB titres. Propionate addition alone resulted in a small increase in propionyl- and methylmalonyl-CoA levels across all strains. This observation suggests lax regulation of the T7 expression machinery or possibly innate metabolic capabilities of the given cell lines.

Additionally, the acetyl-CoA levels for the engineered strains were lower when propionate was fed under both induced and uninduced conditions compared to nonengineered controls (i.e. YW22 and BL21(DE3)). The decrease was not apparent when no propionate was fed under both induced and uninduced conditions (Fig. 2c,d). The most plausible explanation for this result is the propionate- and engineered prpE-dependent coupling of free CoA molecules to alternative acyl units. The observation of this trend without the addition of IPTG again suggests leaky expression of the native prpE gene, supported by the reduced differences between acetyl-CoA levels (Fig. 2b) when compared to the condition with both propionate and IPTG present (Fig. 2a). It is also interesting to note the increase in acetyl-CoA levels within the engineered strains with only the addition of IPTG, suggesting an additional capability of the native PrpE in the absence of its preferred propionate substrate.

In an attempt to further improve the availability and diversity of the acyl-CoA pool in engineered E. coli, the flexible PrpE from R. solanacearum (PrpE-RS) was expressed in BAP1. PrpE-RS had previously shown broad in vitro flexibility by converting acetate, propionate and butyrate into the corresponding CoA thioesters with variations in specificity and efficiency for each substrate (Rajashekhara and Watanabe 2004). The preferences reported indicated strength of activity as acetate>propionate>butyrate but greater kcat/Km values for propionate>acetate>butyrate. When tested in BAP1, improvements in all three intracellular substrates were observed ranging from 1·5- to 15-fold, with the greatest increase in butyryl-, followed by propionyl- and then acetyl-CoA. However, these improvements could not be confirmed in the context of a key control. The same level of intracellular activity could not be repeated for PrpE-RS when compared to an empty pACYCDuet-1 vector in BAP1 (Fig. 4). Induction of BAP1/pJWA1 vs BAP1/pACYCDuet-1 only revealed small fold increases in acyl-CoA compounds between 1·2 and 2·0. This result suggests that PrpE-RS may either not have full activity in BAP1 or, more interestingly, the native PrpE in E. coli may be demonstrating substrate flexibility.

To further confirm this hypothesis, pJWA1 and pACYCDuet-1 were transformed into BL21(DE3) and compared to BAP1 alone. As above, acetate, butyrate and propionate were fed, and the induced acyl-CoA profiles were compared between each strain (Fig. 5). BAP1 showed increased levels of butyryl- and propionyl-CoA, while BL21(DE3)/pJWA1 did not show increased levels with respect to the control BL21(DE3)/pACYCDuet-1. Additionally, increases in methylmalonyl-CoA were observed for BAP1 when propionate was fed. The increase in methylmalonyl-CoA in the absence of a dedicated propionyl-CoA carboxylase points to native metabolism capable of producing this compound, as observed previously in the context of 6dEB formation (Pistorino and Pfeifer 2009, Zhang et al. 2010).

This result demonstrates the flexibility associated with the native E. coli PrpE enzyme. Such information may then be used directly by applying protein engineering to expand or improve upon these native capabilities. In addition, the analytical method described here will be used to profile future metabolic engineering attempts to improve the availability of acyl-CoA molecules for heterologous biosynthetic attempts or to probe the introduction of new pathways to further diversify this important pool of metabolites.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors thank Ming Jiang for technical assistance throughout the study and Christopher Haynes for guidance during early-stage method development. This work was supported by the NSF (award nos. 0712019 and 0924699).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Bennett, B.D., Kimball, E.H., Gao, M., Osterhout, R., van Dien, S.J. and Rabinowitz, J.D. (2009) Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat Chem Biol 5, 593599.
  • Boghigian, B.A., Zhang, H. and Pfeifer, B.A. (2011a) Multi-factorial engineering of heterologous polyketide production in Escherichia coli reveals complex pathway interactions. Biotechnol Bioeng 108, 13601371.
  • Boghigian, B.A., Salas, D., Ajikumar, P.K., Stephanopoulos, G. and Pfeifer, B.A. (2011b) Analysis of heterologous taxadiene production in K- and B-derived Escherichia coli. Appl Microbiol Biotechnol ????, ????????.(Epub August 2011)
  • Boynton, Z.L., Bennett, G.N. and Rudolph, F.B. (1994) Intracellular concentrations of coenzyme A and its derivatives from Clostridium acetobutylicum ATCC 824 and their roles in enzyme regulation. Appl Environ Microbiol 60, 3944.
  • Buccholz, A., Takors, R. and Wandrey, C. (2001) Quantification of intracellular metabolites in Escherichia coli K12 using liquid chromatographic-electrospray ionization tandem mass spectrometric techniques. Anal Biochem 295, 129137.
  • Dettmer, K., Aronov, P.A. and Hammock, B.D. (2007) Mass spectrometry-based metabolomics. Mass Spectrom Rev 26, 5178.
  • Dunn, W.B., Bailey, N.J. and Johnson, H.E. (2005) Measuring the metabolome: current analytical technologies. Analyst 130, 606625.
  • Gao, L., Chiou, W., Tang, H., Cheng, X., Camp, H.S. and Burns, D.J. (2007) Simultaneous quantification of malonyl-CoA and several other short-chain acyl-CoAs in animal tissues by ion-pairing reversed phase HPLC/MS. J Chromatogr B Analyt Technol Biomed Life Sci 853, 303313.
  • Haynes, C.A., Allegood, J.C., Sims, K., Wang, E.W., Sullards, M.C. and Merrill, A.H. Jr (2008) Quantitation of fatty acyl-coenzyme As in mammalian cells by liquid chromatography-electrospray ionization tandem mass spectrometry. J Lipid Res 49, 11131125.
  • Park, J.W., Jung, W.S., Park, S.R., Park, B.C. and Yoon, Y.J. (2007) Analysis of intracellular short organic acid-coenzyme A esters from actinomycetes using liquid chromatography-electrospray ionization-mass spectrometry. J Mass Spectrom 42, 11361147.
  • Pfeifer, B.A., Admiraal, S.J., Gramajo, H., Cane, D.E. and Khosla, C. (2001) Biosynthesis of complex polyketides in a metabolically engineered strain of E. coli. Science 291, 17901792.
  • Pistorino, M. and Pfeifer, B.A. (2009) Efficient experimental design and micro-scale medium enhancement of 6-deoxyerythronolide B production through Escherichia coli. Biotechnol Prog 25, 13641371.
  • Rajashekhara, E. and Watanabe, K. (2004) Propionyl-coenzyme A synthetases of Ralstonia solanacearum and Salmonella chloraesius display atypical kinetics. FEBS Lett 556, 143147.
  • Sambrook, J., Fritach, E.F. and Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual. New York: Cold Spring Harbor Laboratory.
  • Se Jong, H., Woo Park, S., Woo Kim, B. and Jun Sim, S. (2008) Selective production of epothilone B by heterologous expression of propionyl-CoA synthetase in Sorangium cellulosum. J Microbiol Biotechnol 18, 135137.
  • Valentin, H.E., Mitsky, T.A., Mahadeo, D.A., Tran, M. and Gruys, K.J. (2000) Application of a propionyl coenzyme A synthetase for poly(3-hydroxypropionate-co-3-hydroxybutyrate) accumulation in recombinant Escherichia coli. Appl Environ Microbiol 66, 52535258.
  • Villas-Boas, S.G., Hojer-Pedersen, J., Akesson, M., Smedsgaard, J. and Nielsen, J. (2005a) Global metabolite analysis of yeast: evaluation of sample preparation methods. Yeast 22, 11551169.
  • Villas-Boas, S.G., Mas, S., Akesson, M., Smedsgaard, J. and Nielsen, J. (2005b) Mass spectrometry in metabolome analysis. Mass Spectrom Rev 24, 613646.
  • Yanes, O., Tautenhahn, R., Patti, G.J. and Siuzdak, G. (2011) Expanding coverage of the metabolome for global metabolite profiling. Anal Chem 83, 21522161.
  • Yang, S., Sadilek, M., Synovec, R.E. and Lidstrom, M.E. (2009) Liquid chromatography-tandem quadrupole mass spectrometry and comprehensive two-dimensional gas chromatography-time-of-flight mass spectrometry measurement of targeted metabolites of Methylobacterium extroquens AM1 grown on two different carbon sources. J Chromatogr A 1216, 32803289.
  • Zha, W., Rubin-Pitel, S.B., Shao, Z. and Zhao, H. (2009) Improving cellular malonyl-CoA level in Escherichia coli via metabolic engineering. Metab Eng 11, 192198.
  • Zhang, H., Boghigian, B.A. and Pfeifer, B.A. (2010) Investigating the role of native propionyl-CoA and methylmalonyl-CoA metabolism on heterologous polyketide production in Escherichia coli. Biotechnol Bioeng 105, 567573.