ACETYL-COA-DERIVED BIOSYNTHESIS

Abstract

A method to improve the production of acetyl-CoA-derived biochemicals by overexpression of an acetyl-coenzyme A synthetase or acetate-CoA ligase from naturally acetate-utilizing organisms with or without an added acetate transporter. The production of free fatty acid and its derivatives from renewable carbon source was used as a non-limiting example. Using this approach, the production of free fatty acids with yield close to the maximum theoretical yield at high titer can be achieved. As such, this invention will provide the necessary framework to produce many other products sharing or branching out from the fatty acid synthesis pathway economically. These products include hydrocarbons, fatty alcohols, hydroxy fatty acids, dicarboxylic acids, fatty acid esters, etc.

Claims

1) A recombinant microbe, said microbe overexpressing i) an acetyl-coenzyme A synthetase or an acetate-CoA ligase from a natural acetate-utilizing organism, and ii) an acetate transporter, said microbe having a higher rate of acetyl-coA synthesis than a comparable microbe without both i and ii.

2) The recombinant microbe of claim 1, said naturally acetate-utilizing organism selected from Methanosaeta thermophila PT, Methanosaeta harundinacea, Methanosaeta concilii, Methanosaeta sp. ASM2, Methanosaeta sp. NSP1, Methanosaeta sp. NSM2, Methanobacteriaceae archaeon 41_258, Methanothermobacter sp. MT-2, Methanothermobacter sp. CaT2, Methanothermobacter marburgensis, Methanothermobacter thermautotrophicus, Methanobacterium, Methanobacterium congolense, Methanobacterium formicicum, Methanobacterium formicicum DSM 3637, Methanobacterium lacus, Methanobacterium paludis, Methanobacterium sp. Maddingley MBC34, Methanobacterium sp. SMA-27, and Methanobacterium sp. 42_16.

3) The recombinant microbe of claim 1, wherein said acetyl-coenzyme A synthetase or acetate-CoA ligase is selected from those listed in Table 1.

4) The recombinant microbe of claim 1, wherein said acetate transporter is selected from ActP and SatP.

5) The recombinant microbe of claim 1, wherein said acetyl-coenzyme A synthetase or said acetate-CoA ligase is from Methanosaeta and said acetate transporter is selected from ActP and SatP from Escherichia.

6) The recombinant microbe of claim 1, said microbe having a genotype of i) acs1Mst.sup.++ or acs2Ea.sup.++ or acs3Mm.sup.++ and ii) actP.sup.++.

7) The recombinant microbe of claim 1, said microbe capable of producing at least 50% more of a product requiring acetyl-coenzyme A than a comparable microbe lacking i) and ii).

8) The recombinant microbe of claim 7, wherein said naturally acetate-utilizing organism is selected from Methanosaeta thermophila PT, Methanosaeta harundinacea, Methanosaeta sp. ASM2, Methanosaeta sp. NSP1, Methanosaeta sp. NSM2, Methanobacteriaceae archaeon 41_258, Methanothermobacter sp. MT-2, Methanothermobacter sp. CaT2, Methanothermobacter marburgensis, Methanothermobacter thermautotrophicus, Methanobacterium, Methanobacterium congolense, Methanobacterium formicicum, Methanobacterium formicicum DSM 3637, Methanobacterium lacus, Methanobacterium paludis, Methanobacterium sp. Maddingley MBC34, Methanobacterium sp. SMA-27, and Methanobacterium sp. 42_16.

9) A recombinant microbe, said microbe having i) an overexpressed acetyl-coenzyme A synthetase or an acetate-CoA ligase from a naturally acetate-utilizing organism, plus optionally ii) overexpression of an acetate transporter, said naturally acetate-utilizing organism selected from Methanosaeta thermophila PT, Methanosaeta harundinacea, Methanosaeta sp. ASM2, Methanosaeta sp. NSP1, Methanosaeta sp. NSM2, Methanobacteriaceae archaeon 41_258, Methanothermobacter sp. MT-2, Methanothermobacter sp. CaT2, Methanothermobacter marburgensis, Methanothermobacter thermautotrophicus, Methanobacterium, Methanobacterium congolense, Methanobacterium formicicum, Methanobacterium formicicum DSM 3637, Methanobacterium lacus, Methanobacterium paludis, Methanobacterium sp. Maddingley MBC34, Methanobacterium sp. SMA-27, and Methanobacterium sp. 4216.

10) (canceled)

11) A recombinant microbe, said microbe being E. coli having a genotype of i) acs1Mst.sup.++ or acs2Ea.sup.++ or acs3Mm.sup.++ and ii) actP.sup.++ and optionally iii) acyl-ACP thioesterase.sup.+.

12) A method of producing a product, comprising: a) inoculating a microbe of claim 1 into a nutrient broth containing a carbon source; b) growing said microbe in said nutrient broth for a time sufficient to overexpress said i) and ii)(if present) in an amount sufficient to make acetyl-coA and convert said acetyl-coA into an acetyl-co-A derived product, and c) isolating said acetyl-co-A derived product from said microbe, said nutrient broth, or both.

13) The method of claim 12, said nutrient broth supplemented with about 10 mg Mg.

14) The method of claim 12, further comprising feeding additional carbon source to said microbes when a pH of said nutrient broth becomes higher than 7.6.

15) The method of claim 12, wherein said acetyl-coA derived product is selected from free fatty acid, hydrocarbons, fatty alcohols, hydroxy fatty acids, dicarboxylic acids, and fatty acid esters.

16) The method of claim 12, wherein said carbon source is selected from glucose, sucrose, xylose, arabinose, galactose, mannose, acetate, glycerol, sugar mixtures, and hydrolysate with mixed sugars.

17) The method of claim 12, wherein said carbon source is acetate.

18) The method of claim 12, said nutrient broth comprising glucose, said acetyl-co-A derived product being free fatty acids, and said microbe producing 0.3 g of free fatty acid per gram of glucose.

19) The method of claim 12, wherein an induction level of said i) and ii) expression is optimized to not be so high as to slow cell growth by >10% or so low as to produce <90% of a theoretical maximum of product.

20) The method of claim 12, said acetyl-coenzyme A synthetase or acetate-CoA ligase and said acetate transporter being inducible with about 100 M of IPTG in said nutrient broth.

21) A method of producing a product, comprising: a) inoculating a microbe of claim 9 into a nutrient broth containing a carbon source; b) growing said microbe in said nutrient broth for a time sufficient to overexpress said i) and ii) (if present) in an amount sufficient to make acetyl-coA and convert said acetyl-coA into an acetyl-co-A derived product, and c) isolating said acetyl-co-A derived product from said microbe, said nutrient broth, or both.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0062] FIG. 1. Fatty acid synthesis pathway.

[0063] FIG. 2. Sequences.

DESCRIPTION OF EMBODIMENTS OF THE INVENTION

[0064] We have shown that the acetyl-coenzyme A synthetase from M. thermophila performs much better than the native E. coli acetyl-coenzyme A synthetase using the production of fatty acid as an example. Furthermore, we have shown that co-expression of an acetate transporter further improves its performance, as measured by ratio of grams of fatty acids produced to grams of glucose used.

[0065] A comparison of the protein sequence between the M. thermophile Acs and E. coli Acs is shown in FIG. 2.

TABLE-US-00004 Strains table Strains Genotype Referance ML103 MG1655 fadD (Li et al., 2012)(Zhang et al., 2011) ML212 ML103 sucC fabR (Li et al., 2017)

TABLE-US-00005 Plasmids table Plasmids Description Referance pWL1T Modified pTrc99a carrying a long chain acyl-ACP thioesterase (Li et al., 2017) (TE) under a constitutive promoter pWL1TZ pWL1T overexpression of a -hydroxyacyl-ACP dehydratase (Li et al., 2017) (fabZ) under a constitutive promoter pWL1TZR pWL1TZ overexpression of a transcriptional dual regulator This invention (fadR) under a normal Ptrc promoter pWL8 Replace araBAD promoter of pBAD33 using lad promoter This invention pWL8-acsEc pWL8 carrying an acetyl-CoA synthetase (acs) from E. coli This invention pWL8-acs1Mst pWL8 carrying a codon optimized acs from Methanosaeta strain This invention pWL8-actP-acsEc pWL8-acsEc carrying an acetate transporter (actP) This invention pWL8-actP-acs1Mst pWL8-acs1Mst carrying an actP This invention pWL8-actP-acs2Ea Replace acs1Mst of pWL8-actP-acs1Mst using acs2Ea from This invention Euryarchaeota archaeon ex4484_162 pWL8-actP-acs3Mm Replace acs1Mst of pWL8-actP-acs1Mst using acs3Mm from This invention Methanothermobacter sp. MT-2

[0066] The metabolically engineered strains were studied in shake flasks as well as controlled bioreactor systems. For shake flask experiments, the strains were grown in 250 mL flasks, with 40 mL Luria-Bertani (LB) broth medium supplemented with glucose, appropriate IPTG, and 100 g/mL ampicillin and/or 35 g/mL chloramphenicol. The inoculation size is 1% (v/v). The cultures were grown in a rotary shaker at 250 rpm.

[0067] Samples of the media were taken at specific time after inoculation. The fatty acids were analyzed and quantified by GC/MS and GC/FID after extraction. Odd number saturated straight chain fatty acids, such as C13, C15 and/or C17 carbon chain length fatty acid, were used as the internal standard. The results shown in following tables are the sum of all major free fatty acids in the sample. The data shown are means for triplicate experiments.

[0068] Periodic-fed batch bioreactor experiments were performed on a 1-L bioreactor (BioFlo 110, New Brunswick Scientific Edison, N.J.) at 30 C. with 1% inoculation size in 600 mL LB broth supplied with glucose (and then fed at specific times), 100 g/mL ampicillin and/or 35 g/mL chloramphenicol, appropriate IPTG. The initial pH was adjusted to 7.3 with 2 N NaOH, the aeration was maintained at 1.0 vvm with filtered air. Continuous-fed bioreactor experiments were performed at the same conditions except using super broth as medium and continuous feeding glucose when pH in the medium was higher than 7.60.

Overexpression of ACS or ACS and actP

[0069] Shake flask experiments were performed to demonstrate the effect of overexpression of acs1Mst and actP on the production of fatty acids as a non-limiting example. However, any other A-coA derived product could have been used for proof of concept.

[0070] The strain ML103 carrying the acyl-ACP thioesterase plasmid pWL1T produced moderate quantity of free fatty acids, about 1.52 g/L with a yield of 0.101 g/g on 15 g/L glucose. However, the strain ML103(pWL1T) bearing the acetyl-coenzyme A synthetase only produced a little fatty acid. When combining overexpression of acetate transporter ActP and acetyl-coenzyme A synthetase, the strains produced much higher fatty acid with more than 75% improvement compared with ML103(pWL1T+pWL8) on 15 g/L glucose. Furthermore, the corresponding improvement increased with the increase concentration of glucose. Moreover, the strains bearing an acetyl-coenzyme A synthetase from M. thermophila (Acs1Mst) always showed better performance than the strains bearing an acetyl-coenzyme A synthetase from E. coli (AcsEc).

TABLE-US-00006 ~15 g/L glucose ~30 g/L glucose ~40 g/L glucose Titer Yield Improvement.sup.# Titer Yield Improvement.sup.# Titer Yield Improvement.sup.# ML103 carrying Relevant genotype (g/L) (g/g) (%) (g/L) (g/g) (%) (g/L) (g/g) (%) pWL1T + pWL8 fadD, acyl-ACP 1.52 0.101 1.27 0.044 0.08 0.002 thioesterase.sup.+, cloning vector pWL8 pWL8-acsEc fadD, acyl-ACP 0.07 0.005 95.39 0.09 0.003 92.91 0.12 0.003 50.00 thioesterase.sup.+, acsEc.sup.++ pWL8-acs1Mst fadD, acyl-ACP 0.08 0.005 94.74 0.19 0.007 85.04 0.27 0.007 237.50 thioesterase.sup.+, acs1Mst.sup.++ pWL8-actP-acsEc fadD, acyl-ACP 2.66 0.177 75.00 4.13 0.145 225.20 3.97 0.106 4862.50 thioesterase.sup.+, actP.sup.++, acsEc.sup.++ pWL8-actP-acs1Mst fadD, acyl-ACP 2.71 0.181 78.29 4.64 0.162 265.35 5.10 0.136 6275.00 thioesterase.sup.+, actP.sup.++, acs1Mst.sup.++ pWL8-actP-acs2Ea fadD, acyl-ACP 2.58 0.172 70.13 thioesterase.sup.+, actP.sup.++, acs2Ea.sup.++ pWL8-actP-acs3Mm fadD, acyl-ACP 2.59 0.173 70.51 thioesterase.sup.+, actP.sup.++, acs3Mm.sup.++ acsEc.sup.++ = overexpression of E. coli Acs; acs1Mst.sup.++ = overexpression of M. thermophila Acs; acs2Ea.sup.++ = overexpression of E. archaeon Acs; acs3Mm.sup.++ = overexpression of Methanothermobacter sp. Acs; actP.sup.++ = overexpression of ActP; .sup.#Percentage improvement based on fatty acid produced by ML103(pWL1T + pWL8)

[0071] The strain ML103 carrying the acyl-ACP thioesterase and -hydroxyacyl-ACP dehydratase (FabZ) plasmid pWL1TZ produced 3.28 g/L fatty acid on 15 g/L glucose, while strain ML103(pWL1TZ) bearing the acetyl-coenzyme A synthetase and acetate transporter ActP also got higher fatty acid under the same concentration of glucose. 7.73 g/L fatty acid was obtained from 40 g/L glucose when the strain carrying acetyl-coenzyme A synthetase Acs1Mst and acetate transporter ActP.

TABLE-US-00007 ~15 g/L glucose ~30 g/L glucose ~40 g/L glucose Titer Yield Improvement.sup.# Titer Yield Improvement.sup.# Titer Yield Improvement.sup.# ML103 carrying Relevant genotype (g/L) (g/g) (%) (g/L) (g/g) (%) (g/L) (g/g) (%) pWL1TZ + pWL8 fadD, fabZ.sup.++, acyl- 3.28 0.219 4.23 0.141 0.22 0.005 ACP thioesterase.sup.+, cloning vector pWL8 pWL8- fadD, fabZ.sup.++, acyl- 0.07 0.005 97.87 0.06 0.002 98.58 0.05 0.001 77.27 acsEc ACP thioesterase.sup.+, acsEc.sup.++ pWL8- fadD, fabZ.sup.++, acyl- 0.07 0.005 97.87 0.12 0.004 97.16 0.17 0.004 22.73 acs1Mst ACP thioesterase.sup.+, acs1Mst.sup.++ 3.35 0.223 2.13 4.81 0.16 13.71 6.36 0.159 2790.91 pWL8-actP- fadD, fabZ.sup.++, acyl- acsEc ACP thioesterase.sup.+, actP.sup.++, acsEc.sup.++ 3.6 0.24 9.76 5.7 0.19 34.75 7.73 0.193 3413.64 pWL8-actP- fadD, fabZ.sup.++, acyl- acs1Mst ACP thioesterase.sup.+, actP.sup.++, acs1Mst.sup.++ acsEc.sup.++ = overexpression of E. coli Acs; acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ; .sup.#Percentage improvement based on fatty acid produced by ML103(pWL1TZ + pWL8)

[0072] The strain ML103 carrying the acyl-ACP thioesterase, FabZ and transcriptional dual regulator (FadR) plasmid pWL1TZR produced higher free fatty acids. However, the product yield decreases with increasing glucose concentrations. The strain with overexpression of E. coli acs and actP improved the fatty acid titer by 9.5% to 4.15 g/L while the strain with overexpression of M. thermophila acs and actP improved the fatty acid titer by more than 22% to 4.64 g/L with 15 g/L of glucose.

[0073] The strain with overexpression of M. thermophila acs only could produce 4.17 g/L fatty acid on 15 g/L glucose, a similar level that of the strain with overexpression of both E. coli acs and actP. Furthermore, the strain with overexpression of M. thermophila acs and actP maintained higher free fatty acid production with high yield even at higher glucose concentrations of 30 g/L and 40 g/L. This strain produced 12.04 g/L of free fatty acids with a yield of 0.301 g/g from 40 g/L of glucose. This high yield of 0.30 g/g is close to 90% of the maximum theoretical yield (e.g. the maximum theoretical yield of palmitic acid is 0.34 g/g glucose).

[0074] While the strains with overexpression of E. archaeon acs or Methanothermobacter sp.acs were shown to have lower the fatty acid titers than the strain with overexpression of M. thermophila acs at 15 g/L, 30 g/L and 40 g/L glucose, they did show a better improvement over the E. coli Acs at 15 g/L of glucose. These observations showed that Acs from different natural acetate using organisms can be used to improve acetate utilization but to various degrees.

TABLE-US-00008 ~15 g/L glucose ~30 g/L glucose ~40 g/L glucose Titer Yield Improvement.sup.# Titer Yield Improvement.sup.# Titer Yield Improvement.sup.# ML103 carrying Relevant genotype (g/L) (g/g) (%) (g/L) (g/g) (%) (g/L) (g/g) (%) pWL1TZR + pWL8 fadD, fabZ.sup.++, fadR.sup.++, 3.79 0.253 6.15 0.205 4.98 0.1245 acyl-ACP thioesterase.sup.+, cloning vector pWL8 pWL8-acsEc fadD, fabZ.sup.++, fadR.sup.++, 0.35 0.023 90.77 0.46 0.015 92.52 0.17 0.004 96.59 acyl-ACP thioesterase.sup.+, acsEc.sup.++ pWL8- fadD, fabZ.sup.++, fadR.sup.++, 4.17 0.278 10.03 0.79 0.026 87.15 0.19 0.00475 96.18 acs1Mst acyl-ACP thioesterase.sup.+, acs1Mst.sup.++ pWL8-actP- fadD, fabZ.sup.++, fadR.sup.++, 4.15 0.277 9.50 8.12 0.271 32.03 10.73 0.268 115.46 acsEc acyl-ACP thioesterase.sup.+, actP.sup.++, acsEC.sup.++ pWL8-actP- fadD, fabZ.sup.++, fadR.sup.++, 4.64 0.309 22.43 8.99 0.300 46.18 12.04 0.301 141.77 acs1Mst acyl-ACP thioesterase.sup.+, actP.sup.++, acs1Mst.sup.++ pWL8-actP- fadD, fabZ.sup.++, fadR.sup.++, 4.47 0.298 17.87 8.10 0.270 31.63 8.50 0.212 70.72 acs2Ea acyl-ACP thioesterase.sup.+, actP.sup.++, acs2Ea.sup.++ pWL8-actP- fadD, fabZ.sup.++, fadR.sup.++, 4.41 0.294 16.30 8.28 0.276 34.66 8.66 0.216 73.91 acs3Mm acyl-ACP thioesterase.sup.+, actP.sup.++, acs3Mm.sup.++ acsEc.sup.++ = overexpression of E. coli Acs; acs1Mst.sup.++ = overexpression of M. thermophila Acs; acs2Ea.sup.++ = overexpression of E. archaeon Acs; acs3Mm.sup.++ = overexpression of Methanothermobacter sp. Acs; actP.sup.++ = overexpression of ActP; fadR.sup.++ = overexpression of FadR; fabZ.sup.++ = overexpression of FabZ; .sup.#Percentage improvement based on ML103(pWL1TZR + pWL8)

[0075] In summary, co-overexpression of M. thermophila acs and actP allows high fatty acid production with high yield. This strain can also maintain high titer and high yield even at high glucose concentrations.

Time Profile of ACS, ACS, actP

[0076] Experiments were performed to track the time profiles of two strains and to compare their fatty acids production and the glucose utilization at 12 hour intervals. Although the glucose utilization time profiles were very similar among these two strains, the free fatty acids production were very different, with the strain carrying M. thermophila acs and E. coli actP producing the most fatty acids. Moreover, the production rate reached about 0.25 g/L/h between 12 and 24 hr even with these low cell density cultures with shake flask experiments.

TABLE-US-00009 Relevant Time (h) Strains geneotype 0 12 24 36 48 ML103(pWL1TZR + fadD, fabZ.sup.++, Fatty acid (g/L) 0 0.83 3.80 4.61 4.63 pWL8-actP-acs1Mst) fadR.sup.++, Residual 15 8.57 1.71 0 0 acyl-ACP glucose (g/L) thioesterase.sup.+, Acetate (g/L) 0 0.23 0.05 0 0 acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZR + fadD, fabZ.sup.++, Fatty acid (g/L) 0 1.45 3.30 3.80 3.79 pWL8) fadR.sup.++, Residual 15 7.80 1.32 0 0 acyl-ACP glucose (g/L) thioesterase.sup.+, Acetate (g/L) 0 0.27 0.14 0.08 0.06 cloning vector pWL8 acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fadR.sup.++ = overexpression of FadR; fabZ.sup.++ = overexpression of FabZ;

Importance of ACS Expression Level

[0077] Experiments were performed to examine the expression level effect on strain performance. The control strain, ML103(pWL1TZ+pWL8), does not show any independent of the induction level. For this control strain, the fatty acid production is around 3.3 g/L. However, the acs1Mst.sup.++, actP.sup.++ carrying strain showed a strong dependence on the inducer, IPTG, concentrations. The performance dropped significantly at high induction level, probably due to over-burden by high levels of protein expression. The optimal for this construct is around 100 M of IPTG.

TABLE-US-00010 IPTG (M) Strains Relevant Genotype 0 50 80 100 120 150 180 200 ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 3.32 2.67 3.30 3.60 3.56 0.34 0.24 0.19 pWL8-actP- acyl-ACP Residual glucose 0 0 0 0 0 8.34 8.70 8.90 acs1Mst) thioesterase.sup.+, (g/L) acs1Mst.sup.++, actP.sup.++ Acetate (g/L) 0 0 0 0 0 2.69 2.66 2.63 ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 3.37 3.26 3.25 3.28 3.32 3.35 3.28 3.30 pWL8) acyl-ACP Residual glucose 0 0 0 0 0 0 0 0 thioesterase.sup.+, cloning (g/L) vector pWL8 Acetate (g/L) 0 0 0 0 0 0 0 0 acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ.

[0078] In summary, an appropriate expression level of acetyl-coenzyme A synthetase and acetate transporter is needed for optimal strain performance. Too much expression strains the cell, and reduces production. We have found an optimal level to be about 80-120 M or 100 M IPTG to be effective, and titrations with other inducers can be performed as needed to determine optimal levels.

Effect of Induction Timing

[0079] Experiments were performed to examine the effect of timing of induction of gene expression on strain performance. The control strain, ML103(pWL1TZ+pWL8), does not show any independent of the induction timing, as expected. For this control strain, the fatty acid production is around 3.3 g/L. However, the acs1Mst.sup.++, actP.sup.++ carrying strain showed little dependence on the induction time less than 3 hours after inoculation. However, the performance dropped significantly when the culture was induced at 5 hours after inoculation. These observations indicate that sufficient time should be allowed for cells to synthesize the enzyme(s) in order to have good performance.

TABLE-US-00011 Induction time of 100 M IPTG (h) Strains Relevant phenotype no IPTG 0 1 2 3 4 5 ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 3.32 3.60 3.53 3.53 3.55 3.34 0.21 pWL8-actP-acs1Mst) acyl-ACP thioesterase.sup.+, acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 3.37 3.28 3.35 3.36 3.30 3.29 3.34 pWL8) acyl-ACP thioesterase.sup.+, cloning vector pWL8 acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ;

[0080] In summary, induction at or near the beginning of the experiments (2-4 hrs) yields the best performance. This finding will facilitate future strain design and bioprocess operation as the results suggest it is possible to use constitutive promoter systems for the overexpression of M. thermophila Acs.

Repeat Feed in Shake Flasks

[0081] Repeated feed experiments in shake flasks were performed to examine the ability of the strains to produce fatty acid in a sustained manner. In these experiments, 15 g/L glucose was added to the culture at every 24 h interval and the fatty acids in the culture were monitored. The control strain, ML103(pWL1TZ+pWL8), showed poor fatty acid production performance after the second sugar addition. However, the strain ML103(pWL1TZ+pWL8-actP-acs1Mst) obviously showed significant improvement on fatty acid production, glucose utilization and also acetate reduction. As such, this M. thermophila Acs carrying strain has potential for sustained fatty acid production.

TABLE-US-00012 Time (h) strains Relevant genotype 24 48 96 120 144 ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 2.73 5.66 6.11 7.30 8.24 pWL8-actP-acs1Mst) acyl-ACP Residual glucose (g/L) 1.83 3.63 0.16 9.98 11.46 thioesterase.sup.+, Acetate (g/L) 0.04 0.03 0.00 1.00 1.03 acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 2.39 4.17 5.21 5.27 5.99 pWL8) acyl-ACP Residual glucose (g/L) 2.18 6.88 14.90 28.18 53.08 thioesterase.sup.+, Cloning Acetate (g/L) 0.05 0.02 0.38 1.29 2.67 vector pWL8 acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ;

[0082] Previously we had shown the Mg.sup.++ transporter or added MgCO.sub.3 could improve the stability of fatty acid-producing strains (San et al., 2015). Here we examined repeated-batch with 10 g/L MgCO.sub.3 supplement to raise the productivity. In these experiments, seed cultures were concentrated and re-suspended in LB to OD600 of around 20. After that, 15 g/L glucose was added at every 12 h interval. The strain ML103(pWL1TZ+pWL8-actP-acs1Mst) produced 14.72 g/L fatty acid after 4 batches or within 48 h, the yield and productivity of fatty acid reached to 0.245 g/g and 0.307 g/L/h, while the control strain only produced 7.41 g/L fatty acid with formation of 17.23 g/L acetate.

TABLE-US-00013 Time (h) strains Relevant genotype 12 24 36 48 ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 3.55 7.25 10.57 14.72 pWL8-actP-acs1Mst) acyl-ACP Residual glucose 0.12 0.07 0.07 0.02 thioesterase.sup.+, (g/L) acs1Mst.sup.++, actP.sup.++ Acetate (g/L) 2.78 1.1 1.47 0 ML103(pWL1TZ + fadD, fabZ.sup.++, Fatty acid (g/L) 2.71 4.76 6.71 7.41 pWL8) acyl-ACP Residual glucose 0.03 0.04 0.03 0.26 thioesterase.sup.+, Cloning (g/L) vector pWL8 Acetate (g/L) 3.46 3.95 7.87 17.23 acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ;

Performance with Other Host Strains

[0083] Experiments were conducted to examine the fatty acid production by other strains carrying the M. thermophile acetyl-coenzyme A synthetase. The host strain ML212, which is a sucC and fabR derivative of the strain ML103, was used as an example. The data in the following table showed that the strain ML212 can improve fatty acid production equally well. Thus, the host strain is not critical.

TABLE-US-00014 Fatty Strains Relevant Genotype acid (g/L) ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl-ACP thioesterase.sup.+, 3.60 pWL8-actP-acs1Mst)# acs1Mst.sup.++, actP.sup.++ ML212 (pWL1TZ+ fadD, fabZ.sup.++, acyl-ACP thioesterase.sup.+, 3.84 pWL8-actP-acs1Mst) acs1Mst.sup.++, actP.sup.++ acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ; #Data from Experiment 3 above for comparison

Periodic-Fed Batch Culture in Bioreactor

[0084] Experiments were performed on bioreactors to examine the periodic-fed batch fermentation ability. Strain carrying acs1Mst.sup.++, actP.sup.++ can maintain sustained fatty acid with additional batches of sugar to obtain higher fatty acid titer, yield and productivity than control strain. After 7 batches of 15 g/L glucose addition, 21.45 g/L of fatty acid was produced by strain carrying acs1Mst.sup.++, actP.sup.++. While the control strain, ML03(pWL1TZR+pWL8), only could feed 5 batches owing to slow utilization rate of glucose and produced 4.81 g/L fatty acid after 5 batches. These observations show that the strain carrying acs1Mst.sup.++, actP.sup.++ is more robust and can handle any adverse effect die to acetate accumulation (or prevent high level of acetate accumulation leading to any potential adverse effect for cell growth or product formation).

TABLE-US-00015 Glucose Fatty acids Maximum Relevant Control consumed Titer Yield Productivity acetate strains genotype strategy (g/L) (g/L) (g/g) (g/L/h) (g/L) ML103(pWL1TZR + fadD, fabZ.sup.++, 10 g/L 105 21.45 0.204 0.134 1.23 pWL8-actP-acs1Mst) fadR.sup.++, acyl- MgCO.sub.3 add ACP at 21 h. thioesterase.sup.+, feed 15 g/L acs1Mst.sup.++, glucose at actP.sup.++ 21, 40, 60, 76, 90 and 112 h, total 7 batches. ML103(pWL1TZR + fadD, fabZ.sup.++, 10 g/L 75 4.81 0.064 0.030 2.26 pWL8) fadR.sup.++, acyl- MgCO.sub.3 add ACP at 21 h. thioesterase.sup.+, feed 15 g/L cloning vector glucose at pWL8 21, 40, 60 and 76 h, total 5 batches. acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fadR.sup.++ = overexpression of FadR; fabZ.sup.++ = overexpression of FabZ;

Continuous-Fed Culture

[0085] Experiments were performed on bioreactors to examine the continuous-fed fermentation ability. The strain carrying acs1Mst.sup.++, actP.sup.++ produced 56.48 g/L fatty acids after 65 h when super broth was used as the starting medium, with continuous feeding of glucose when pH in the medium reached higher than 7.60, and a recycling container to recycle trapped cells was used. The corresponding yield and overall production rate were 0.339 g/g and 0.869 g/L/h, respectively. The fatty acid productivity reached to a high rate of 1.177 g of fatty acids/L/h during the production phase. To our best knowledge, these are the highest reported values for fatty acid titer, yield and productivity by an E. coli culture.

TABLE-US-00016 Glucose Fatty acids Maximum Relevant Control consumed Titer Yield Productivity acetate strains genotype strategy (g/L) (g/L) (g/g) (g/L/h) (g/L) ML103(pWL1TZR + fadD, fabZ.sup.++, 10 g/L 167 56.48 0.339 0.869 0.60 pWL8-actP-acs1Mst) fadR.sup.++, acyl- MgCO.sub.3 add ACP at 17 h. thioesterase.sup.+, continuous acs1Mst.sup.++, feed 500 actP.sup.++ g/L glucose using pH as a indictor (pH 7.60, start feed) acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fadR.sup.++ = overexpression of FadR; fabZ.sup.++ = overexpression of FabZ;

Various Carbon Sources

[0086] Experiments were performed to examine repeated batch fermentation using acetate as the sole carbon source. The seed cultures were concentrated and re-suspended in LB to OD600 of around 20, 10 g/L MgCO.sub.3 was added at the beginning, and 5 g/L acetate was fed at every 12 h interval. The strain carrying acs1Mst.sup.++, actP.sup.++ produced 3.45 g/L fatty acids after 4 batches or 48 h while the control strain just produced 2.77 g/L fatty acid with 8.44 g/L acetate unused.

TABLE-US-00017 Time (h) strains Relevant genotype 12 24 36 48 ML103(pWL1TZ + fadD, fabZ.sup.++, acyl- Fatty acid (g/L) 1.71 2.49 3.04 3.45 pWL8-actP-acs1Mst) ACP thioesterase.sup.+, Residual acetate (g/L) 0 0.52 0.71 1.05 acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ + fadD, fabZ.sup.++, acyl- Fatty acid (g/L) 1.37 2.13 2.67 2.77 pWL8) ACP thioesterase.sup.+, Residual acetate (g/L) 0.04 1.02 4.55 8.44 Cloning vector pWL8 acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ;

[0087] Experiments were performed to examine the repeated batch fermentation using acetate mixed with glucose or glycerol as the carbon source. The seed cultures were concentrated and re-suspended in LB to OD600 of around 20, 10 g/L MgCO.sub.3 was added at the beginning, and some amount of acetate mixed with glucose or glycerol was fed at every 12 h interval. After 3 batches or 36 h, the strain carrying acs1Mst.sup.++, actP.sup.++ produced 14.06 g/L fatty acids with a yield of 0.234 g/g when 15 g/L glucose and 5 g/L acetate was used as the carbon source. Feeding with 7.5 g/L glucose and 5 g/L acetate could improve the yield to 0.271 g/g. Moreover, the yield could be further improved when feeding carbon source containing glycerol, and reached to more than 0.27 g/g from 3 batches of 15 g/L glycerol and 0-4 g/L acetate.

TABLE-US-00018 Fatty acid Titer Yield Strains Relevant genotype Feeding carbon source (g/L) (g/g) ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (15 g/L glucose) 6.71 0.149 pWL8) ACP thioesterase+, # Cloning vector pWL8 ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (15 g/L glucose) 10.57 0.235 pWL8- ACP thioesterase.sup.+ actP-acs1Mst) # acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (15 g/L glucose 14.06 0.234 pWL8- ACP thioesterase.sup.+ +5 g/L acetate) actP-acs1Mst) acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (7.5 g/L glucose 10.17 0.271 pWL8- ACP thioesterase.sup.+ +5 g/L acetate) actP-acs1Mst) acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (15 g/L glycerol) 12.45 0.277 pWL8- ACP thioesterase.sup.+ actP-acs1Mst) acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (15 g/L glycerol + 14.41 0.283 pWL8- ACP thioesterase.sup.+ 2 g/L acetate) actP-acs1Mst) acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (15 g/L glycerol + 15.00 0.278 pWL8- ACP thioesterase.sup.+ 3 g/L acetate) actP-acs1Mst) acs1Mst.sup.++, actP.sup.++ ML103(pWL1TZ+ fadD, fabZ.sup.++, acyl- 3 batches (15 g/L glycerol + 15.48 0.272 pWL8- ACP thioesterase.sup.+ 4 g/L acetate) actP-acs1Mst) acs1Mst.sup.++, actP.sup.++ acs1Mst.sup.++ = overexpression of M. thermophila Acs; actP.sup.++ = overexpression of ActP; fabZ.sup.++ = overexpression of FabZ; # Data from Experiment 5b above for comparison

[0088] Experiments were performed to examine fermentation on mix sugar using 7.5 g/L glucose and 7.5 g/L galactose. After 48 h, the strain carrying acs1Mst.sup.++, actP.sup.++ produced 4.22 g/L fatty acid while control strain just produced 3.38 g/L fatty acid.

TABLE-US-00019 Fatty acid Strains Relevant Genotype (g/L) ML103(pWL1TZR+ fadD, fabZ++, fadR.sup.++, acyl-ACP 4.22 pWL8-actP-acs1Mst) thioesterase+, acs1Mst++, actP++ ML103(pWL1TZR+ fadD, fabZ++, fadR.sup.++, acyl-ACP 3.38 pWL8) thioesterase+, cloning vector pWL8 acs1Mst++ = overexpression of M. thermophila Acs; actP++ = overexpression of ActP; fadR++ = overexpression of FadR; fabZ++ = overexpression of FabZ;

[0089] In summary, these results showed the strain carrying acs1Mst.sup.++, actP.sup.++ could ferment efficiently various carbon sources such as sugar, acetate, glycerol and their mixture, and even could use acetate as sole carbon source. Further, with optimized culture conditions, yields of 0.177 g of fatty acids/L/h during the production phase.

[0090] The following publications are incorporated by reference in their entirety for all purposes herein. [0091] US20040199941 INCREASED BACTERIAL COA AND ACETYL-COA POOLS [0092] US20100248233 ACETYL-COA PRODUCING ENZYMES IN YEAST [0093] US20170159075 MICROORGANISMS AND METHODS FOR IMPROVING PRODUCT YIELDS ON METHANOL USING ACETYL-COA SYNTHESIS [0094] U.S. Pat. No. 9,598,696 BACTERIA AND METHOD FOR SYNTHESIZING FATTY ACIDS [0095] WO2017189421 REDOX BALANCING IN YEAST. [0096] Aristidou, A A, San, K. Y., Bennett, G. N., 1995. Metabolic engineering of Escherichia coli to enhance recombinant protein production through acetate reduction. Biotechnol. Prog. 11, 475-478. doi:10.1021/bp00034a019 [0097] Chang, D., Shin, S., Rhee, J., 1999. Acetate Metabolism in a pta Mutant of Escherichia coli W3110: Importance of Maintaining Acetyl Coenzyme A Flux for Growth and Survival Acetate Metabolism in a pta Mutant of Escherichia coli W3110: Importance of Maintaining Acetyl Coenzyme A Flux for Gro. J. Bacteriol. 181, 6656-6663. [0098] Eiteman, M. A., Altman, E., 2006. Overcoming acetate in Escherichia coli recombinant protein fermentations. Trends Biotechnol. doi:10.1016/j.tibtech.2006.09.001 [0099] Farmer, W. R., Liao, J. C., 1997. Reduction of aerobic acetate production by Escherichia coli. Appl. Environ. Microbiol. 63, 3205-3210. [0100] Fujita, Y., Matsuoka, H., Hirooka, K., 2007. Regulation of fatty acid metabolism in bacteria. Mol. Microbiol. doi:10.1111/j.1365-2958.2007.05947.x [0101] Lee, S. Y., 1996. High cell-density culture of Escherichia coli. Trends Biotechnol. 14, 98-105. doi:10.1016/0167-7799(96)80930-9 [0102] Li, M., Zhang, X., Agrawal, A., San, K. Y., 2012. Effect of acetate formation pathway and long chain fatty acid CoA-ligase on the free fatty acid production in E. coli expressing acy-ACP thioesterase from Ricinus communis. Metab. Eng. 14, 380-387. doi:10.1016/j.ymben.2012.03.007 [0103] Li, W., Wu, H., Li, M., San, K.-Y., 2017. Effect of NADPH availability on free fatty acid production in E. coli. Biotechnol. Bioeng. doi:10.1002/bit.26464 [0104] Lin, H., Castro, N. M., Bennett, G. N., San, K.-Y. Y., 2006. Acetyl-CoA synthetase overexpression in Escherichia coli demonstrates more efficient acetate assimilation and lower acetate accumulation: A potential tool in metabolic engineering. Appl. Microbiol. Biotechnol. 71, 870-874. doi:10.1007/s00253-005-0230-4 [0105] Luo, Q., Shen, Y.-L., Wei, D.-Z., Cao, W., 2006. Optimization of culture on the overproduction of TRAIL in high-cell-density culture by recombinant Escherichia coli. Appl. Microbiol. Biotechnol. 71, 184-191. doi:10.1007/s00253-005-0131-6 [0106] San, K.-Y., Han, S., Li, W., Li, M., Li, Z., 2015. Fatty Acids With Mg Transporter and Mg. [0107] THIEL, et al., The Effect Of Enhanced Acetate Influx On Synechocystis Sp. Pcc 6803 Metabolism, 16:21 (2017). [0108] VEERAVALLI, et al., Strain Engineering To Reduce Acetate Accumulation During Microaerobic Growth Conditions In Escherichia coli, Biotechnol Prog (2017). [0109] WOLFE, The Acetate Switch. Microbiol Mol Bio Rev, 69:12-50 (2005). [0110] Zhang, X., Li, M., Agrawal, A., San, K.-Y., 2011. Efficient free fatty acid production in Escherichia coli using plant acyl-ACP thioesterases. Metab. Eng. 13, 713-722. doi:10.1016/j.ymben.2011.09.007