Microbial Production of Fatty Acid Via Metabolic Engineering and Synthetic Biology
Microbial Production of Fatty Acid Via Metabolic Engineering and Synthetic Biology
REVIEW PAPER
Fig. 1. FAB and FAD in E. coli. The pathways of FAB, FAD and membrane lipids metabolism are shown with the respective enzymes
(blue). ATP and reduction equivalents are colored red (consumed) and green (gained). The figure was reproduced from Janssen and
Steinbuchel [10].
tance in the context of sustainable chemical production. results in cell lysis, induces stress responses, reduces cell
viability, and impairs basic physiological process [19-21].
1.2. The difficulties of FFAs production in E. coli Therefore, the overproduction of FAs and FA-derived
The microbial synthesis of FAs is a promising strategy to chemicals in microorganisms needs extensive engineering
produce valuable biofuels from renewable feedstocks [18]. of cellular metabolism. In this review, we mainly focus on
FA metabolism includes both the biosynthesis (FAB) and the regulation of FA biosynthesis, and on the metabolic
degradation (FAD) of FAs in addition to membrane lipid engineering and synthetic biology approaches to develop
biosynthesis (Fig. 1). FA metabolism is reported as it is potent E. coli strains for the production of FFAs at high
tightly regulated at transcriptional and post-transcriptional yield and productivity for commercial exploitation.
levels as well as regulated by intermediate/product feedback
(Fig. 2). The FAB pathway employs iterative condensation
reactions to extend the acyl chain from building blocks, 2. Fatty Acid Biosynthesis in E. coli
including acetyl-CoA, to produce the desired length of
FAs. To synthesize one mole of FA, lots of enzymatic Numerous studies have explained the FAB mechanism in
reactions, categorized into initiation, condensation, elongation, E. coli, so the process is well understood [1,10,15,22,23].
and termination steps and intracellular molecules, are Prokaryotic microorganisms, including E. coli, harbor
involved [1,10] (Fig. 1). Rerouting lipid synthesis into type II FA synthase (FAS II), which produces cellular
FFAs production negatively impacts membrane integrity membrane lipids. The genes encoding FAS II are composed
due to alterations of the membrane lipid composition, which of discrete and mono-functional enzymes. Thus, the
Microbial Production of Fatty Acid via Metabolic Engineering and Synthetic Biology 3
Fig. 2. Transcriptional and post-transcriptional regulations of fatty acid metabolism in E. coli. FAB and FAD are directly or indirectly
regulated by various transcription regulatory proteins such as FadR, FabR, oxygen responsive ArcBA, external osmotic pressure
responsive OmpR, and regulatory molecules such as CRP-cAMP and ppGPP. ArcBA controls FAD in anaerobic condition. The level of
ppGPP, an alarmone, is increased during stringent response and it suppress the target gene expression in association with its transcription
factor, DksA. Critical FAB enzymes are product-feedback-inhibited. Red lines with T-bar heads refer gene inactivation/repression by the
respective regulators. Green arrows refer gene activation by the respective regulators. FAD gene names are given in shaded ovals. FAB
gene names are given in white ovals. Regulatory proteins/molecules are given bold letters in boxes.
intermediates of the type II system could be accessible and by the NADPH-dependent β-hydroxyacyl-ACP reductase
converted into a wide range of FA products [1]. (FabG) in the first elongation cycle to produce β-
E. coli can produce both saturated FAs (SFAs) and hydroxyacyl-ACP that is dehydrated by the β-hydroxyacyl-
unsaturated FAs (UFAs) for membrane phospholipids. The ACP dehydratase (FabZ) to yield enoyl-ACP, which is
first step of bacterial FAS II is the conversion of acetyl- finally reduced by enoyl-ACP reductase (FabI) [27]. The
CoA into malonyl-CoA by the action of the acetyl-CoA resulting saturated acyl-ACP can be further elongated by
carboxylase (ACC) [24]. The ACC is composed of four condensing it with malonyl-ACP by β-ketoacyl-ACP
subunits, the acetyl-CoA carboxyltransferase (heterotetramer, synthase I (FabB) and/or II (FabF) to elongate the chain by
consisting of two subunits of AccA and two subunits of two carbons. The elongation steps could be further catalyzed
AccD), the biotin carboxyl carrier protein (AccB) and, the until C16 saturated FA is produced. To synthesize UFAs, a
homodimer of AccC (biotin carboxylase) [25]. The bifunctional enzyme, β-hydroxydecanoyl-acyl carrier protein
malonyl moiety is transferred to the acyl carrier protein (ACP) dehydratase/isomerase (FabA) introduces double
(ACP) by the malonyl-CoA:ACP transacylase (FabD) to bond at the C10 chain length [28]. Thus, the FabA
yield malonyl-ACP, which participates in subsequent dehydrates β-hydroxydecanoyl-ACP and isomerizes the
condensation and elongation. The next step is the initial product, trans-2-decenoyl-ACP, into the cis-3 isomer, cis-3-
condensation of the acetyl-CoA and malonyl-ACP to decenoyl-ACP [29]. The FabB enzyme also condenses cis-
synthesize β-ketoacyl-ACP by β-ketoacyl-ACP synthase 3-decenoyl-ACP with malonyl-ACP to produce the cis-5-
III (FabH) [26]. The resulting β-ketoacyl-ACP is reduced ene-3-ketododecenoyl-ACP, which can undergo all the
4 Biotechnology and Bioprocess Engineering
subsequent reactions catalyzed by the FabG, FabZ and FabI known to repress expression of the FabA and FabB that are
[10,28,30]. The elongation cycle continues until the long- involved in synthesis of UFAs [44]. The deletion of the
chain acyl-ACPs are incorporated into the phospholipids fabR significantly increased expression level of the fabA
by acyltransferases (PLS) [1,10] (Fig. 1). and fabB, resulting in a high proportion of UFA in lipid
membrane [44]. All the long-chain acyl-ACP or acyl-CoA
could interact with the FabR [45]. However, the FabR-
3. The Regulation of Fatty Acid Metabolism DNA complex could be formed only in the presence of the
unsaturated acyl-ACP or acyl-CoA, indicating that the
FA metabolism is regulated at transcriptional and post- FabR detects the intracellular ratio of the unsaturated form
transcriptional levels by various regulatory proteins and of the acyl group and regulates the expression of the genes
molecules (Fig. 2). The main regulator protein FadR is a [45]. Thus, the FabR is one of regulators in maintaining
multifunctional dual regulator that negatively (repressor) membrane homeostasis. Since the two genes, fabA and
regulates the genes of the FAD pathway (β-oxidation) [31] fabB, are controlled by both the FabR and FadR, the
and positively (activator) regulates the genes of the FAB balanced activity of the regulators is required to maintain
pathway [32]. The FadR specifically binds to the palindromic lipid homeostasis [44,46].
consensus DNA sequence, located in the upstream of the The regulation of ACC in the FAB has been robustly
gene coding region [33]. The DNA-binding activity of the studied in the past [25]. The transcript levels of each
FadR is antagonized by the long-chain fatty acyl- subunit of the ACC complex are correlated with the rates
CoAs. Binding of the acyl-CoAs to the FadR leads to a of cellular growth and FAB [47]. Furthermore, the
conformational change in the FadR and releases it from its expression of accBC has been regulated by the binding of
DNA sequence [34-36]. Since the interaction of the FadR AccB to its promoter [48]. The autoregulation was also
and long chain acyl-CoA tunes the synthesis or degradation found in the translation of accAD mRNA [49]. Thus, the
of the FA, the presence of the long-chain acyl-CoA has a ACC has been considered as a gatekeeper in the FA
determinant role in FA production. The high expression of synthetic pathway. Nutrient availability also regulates the
FadR in E. coli strains with deficient β-oxidation rate of FA synthesis [50]. The activity of the ACC is also
considerably improves FFA production with change in the regulated by nitrogen metabolism, which responds to the
central carbon metabolism, cofactor metabolism, and intracellular level of α-ketoglutarate [51,52].
energy metabolism [35,37]. These approaches have been The fabG is co-transcribed with fabH and fabD, as they
successful in terms of dual regulation as an expression of are encoded in fabHDG operon [53]. The deletion of fabH
FadR enhanced carbon flux toward FFA synthesis as well results in growth retardation but not lethality, implying that
as inhibited β-oxidation. However, an excess level of FadR other enzymes must be involved in the initial condensation
was found to be disadvantageous, as it blocks the alteration reaction [50]. The transcription of the fabHDG genes was
of membrane composition through β-oxidation, resulting in found to start from the promoter of its upstream gene,
a retarded growth rate [37,38]. In addition, the FadR rpmF, encoding the ribosomal protein L32, suggesting that
activates the expression of a regulatory protein IclR, which the synthesis of the ribosome and membrane is co-regulated
downregulates the glyoxylate (GLX) bypass operon [54]. The promoter activity is significantly reduced in
(encoded by aceBAK) [39]. The FadR appears to prevent response to amino acid starvation or in the stationary phase
an increase in IclR during the growth on the FAs and [53,55] to block unnecessary anabolic metabolisms in
results in a GLX bypass, suggesting an essential regulation unfavorable environments such as starvation and nutrient
to prevent the carbon loss as CO2 in the tricarboxylic acid limitation. However, the expression of genes involved in
(TCA) cycle during growth on the FAs [39,40]. Thus, as the FAB such as accBACD, fabHDG, fabF, fabBA, fabZ,
reported by Iram and Cronan [41], the FadR regulates both and fadR, generally decreases during the stringent response
the conversion of FAs to acetyl-CoA and the utilization of mediated by the elevated concentration of the modified
acetyl-CoA by the TCA. Moreover, it is noteworthy that nucleotide guanosine 3’-diphosphate 5’-diphosphate (ppGpp)
anaerobic growth on the FAs is not strongly regulated by and in the presence of the RNAP cofactor DksA [53,56].
the FadR [42]. It has been reported that the anaerobic FAD The regulation of the FabI, the last enzyme responsible
is mediated by FadK, FadI and FadJ, whose transcriptional for FA elongation, is important to block the high energy
regulation is independent of FadR control [42]. Thus, the consuming pathway, which determines the rate of the acyl-
anaerobic FAD is regulated by the aerobic respiration ACP formation, as the preceding reaction is thermody-
control protein Arc through an oxygen-responsive two- namically unfavorable [57]. Since the FabB and FabF are
component (ArcBA) regulatory system [43]. responsible for chain elongation, they are considered as
The FA biosynthesis repressor protein (FabR) has been activators of FAS [46,58,59]. The FabF also has similar
Microbial Production of Fatty Acid via Metabolic Engineering and Synthetic Biology 5
substrate specificity as FabB, but has higher activity 4.1. Engineering the rate-limiting steps
towardsaturated (C6:0- to C16:0-) acyl-ACP and C16:1- It is a well-known that increasing precursor availability
ACP; it has been shown to catalyze the synthesis of long- generally leads to the enhanced synthesis of desired
chain-UFA [60,61]. As the long-chain-UFAs in membranes product(s). The simplest method to increase the pathway
were significantly decreased by heat shock [62], the precursors is by overexpressing the desired pathway(s) or
activity of FabF might be decreased in response to high by blocking the competing pathway(s). The starting
temperature (thermal regulation) [61]. Furthermore, it was molecule in FAB, the acetyl-CoA, should be converted into
shown that the FA productivity at 20°C was 2.7-fold higher malonyl-CoA through ACC activity. As a gatekeeping
than that at 37°C in an E. coli strain overexpressing the enzyme, the expression and activity of the ACC is tightly
FabF, thus exhibiting thermal regulation, since enzyme regulated, resulting in a low concentration of malonyl-
activity was increased at lower temperatures [58]. CoA. This limitation has led to several studies aiming to
Two signal molecules, the acyl-CoA and acyl-ACP, increase the intracellular concentration of malonyl-CoA
influence the FA synthesis by regulating the function of through metabolic engineering. The overexpression of the
transcriptional factors as well as by directly controlling the ACC has shown to significantly increase FA production by
enzymatic activities in the FAB. A high concentration of 6-fold over a short period of time; however, the fold-
acyl-ACP changes the activities of several FAB enzymes, increase in FA production decreases as incubation time
resulting in the conversion of malonyl-ACP into acetyl- increases [69]. The overexpression of the E. coli ACC is
CoA [63]. The activity of the ACC and FabH is feedback found to be detrimental to the cell viability [69]. Thus, high
inhibited by the acyl-ACP [64], with larger chain lengths ACC activity is not an efficient approach to improve the
having a greater effect [65]. In the case of the FabD, the FFA [70-73]. To address this issue, a heterologous ACC
kinetic analysis revealed that the acetyl-CoA can act as a has been employed, and it has increased the level of
competitive inhibitor to the FabD [66,67]. The direction of malonyl-CoA and its derivatives without a detrimental
reaction could be modulated toward malonyl-CoA synthesis impact on cell growth [74-76]. Bypassing the ACC-
in the presence of an equal concentration of substrate and mediated malonyl-CoA synthesis has also been carried out
product [66]. Therefore, enhancing downstream enzyme by expressing methylmalonyl-CoA carboxyltransferase
reactions would be important to prevent a reversal reaction and phosphoenolpyruvate carboxylase, and a 2-fold increase
of the FabD. The activity of the FabI is inhibited by the in the FFA titer was achieved [73].
long-chain acyl-ACP and acyl-CoA [27,57,68]. The acyl- As an elongation unit provider, the high expression of
CoA reduces FabI activity up to 50%, thereby preventing the FabD produces around 10% higher FFA [77]. The
the FA elongation [27]. The complicated interactions between episomal expression of the FabH in E. coli leads to 20%
the two regulatory metabolites and their associated proteins higher lipid production with an altered FA composition
indicate the intricate regulation that E. coli has evolved to toward shorter chain length [78,79]. It is reported that
elaborately control the FA synthesis as well as the ratio of feedback inhibition is conserved in the FAS II and that
unsaturated to saturated FAs, which is necessary for proper mutations on the regulatory site in the plant FabH led to
membrane function under varying conditions. feedback resistance with improved acyl-ACP concentration
in several plant seeds [80]. Thus, employing the mutations
in the E. coli FabH might be one approach to improve FFA
4. The Metabolic Engineering Strategies for production. Previously, several studies revealed that the
Enhancing Fatty Acid Production FabH is an important factor for determining the type of
FAs to be produced. For example, the odd-chain-FA, which
The various significant synthetic biology and metabolic normally does not existin the wild-type E. coli, could be
engineering approaches applied in the last decade to synthesized [81,82] because the FabH showed high substrate
enhance FA production in various E. coli strains are listed specificity to propionyl-CoA as much as to acetyl-CoA
in Table 1. Metabolic pathways of E. coli including [65]. Furthermore, the FabH expression from the Bacillus
biosynthesis and degradation of FAs, (Embden-Meyerhof- subtilis in E. coli produced branched FA [83-85]. Thus, the
Parnas pathway [EMP], pentose phosphate pathway [PPP] FabH can be manipulated for a wide range of chemical
and Entner-Doudoroff pathway [EDP]), and glycerol production.
catabolic pathways along with the major catalytic reactions, In E. coli, there are two β-hydroxyacyl-ACP dehydra-
which are either directly or indirectly connected to the tases, FabZ and FabA, which have broad substrate
regulation of FA metabolism and the overproduction of specificity with different activities on the chain length and
FAs are shown in Fig. 3. The detailed mechanisms of saturation [86]. The FabZ has only dehydratase activity and
applied strategies are described in this section. is encoded in the lipid A cluster. Given that the two
6
Table 1. Various significant engineering strategies to enhance the production of fatty acids in various E. coli strains
E. coli Genetic modifications Thioesterase Media Titer (g/L) FA composition Time (h) Reference
BL21 ΔfadD, ACC↑ TesA’+ CcTEa LB 0.44 C12, C12:1, C14, C14:1, C16:1, C18:0, 17p.i. [71]α
C18:1
BL21 ΔfadD, ACC↑ TesA’+ CcTE M9 with gly. 2.5 (FB) C12, C12:1, C14, C14:1, C16:1, C18:0, 17p.i. [71]
C18:1
BL21 ΔfadD, ACC↑ TesA’+ CcTEb LB 0.94 C12:0, C12:1, C14, C14:1, C14:3-OH, 24 [163]
C16, C16:1, C18:1
BL21 ΔfadD, ACC↑ TesA’+ CcTE M9 with gly. 4.5 (FB) C12:0, C12:1, C14, C14:1, C14:3-OH, 24 [163]
C16, C16:1, C18:1
DH1 ΔfadD TesA’ M9 medium with 2% glu. ~0.7 C8 - C18 48 [96]
DH1 ΔfadE TesA’ M9 medium with 2% glu. ~1.1 C8 - C18 48 [96]
MG1655 ΔfadD, ΔaraBAD UcTE LB with 0.4% gly. 0.70 C12 and C14 (SFA and UFA) 29p.i. [70]
MG1655 ΔfadD, ΔaraBAD, ACC↑ UcTE LB with 0.4% gly. 0.81 C12 and C14 (SFA and UFA) 29p.i. [70]
MG1655 ΔfadD DbTE LB with 1.5% glu. 0.5 C14, C16, C16:1 24 [99]
MG1655 ΔfadD GhTE LB with 1.5% glu. 0.8 C14, C16, C16:1 48 [99]
MG1655 ΔfadD RcTE or JcTE LB with 1.5% glu. 2.1 C14, C16, C16:1 36 [99]
MG1655 ΔfadD UcTE LB with 0.4% gly. 0.77 C16,16:1, C18, C18:1 24 [117]
MG1655 ΔfadBA,ΔfadD, ΔyqhD, Δpta, FadM Minimal with 3% glu. 7.0 (RBO;BR) C16 - C18 60 [120]
ΔadhE, ΔfrdA, ΔfucO, ΔarcA, crp*,
fadR*, atoC*
BL21 -- AcTE M9 with 0.5% tryptone 3.6 (FB) C6 - C18 (SFA and UFA) 48 [164]
BL21 ΔfadL TesA’ Minimal with 2% glu. 4.8 (FB) C12, C12:1, C14, C14:1, C16, C16:1, 38 [104]
C18, C18:1
MG1655 ΔfadD, FabD↑ RcTE LB with 1.5% glu. ~1.3 C14, C16, C16:1 24 [77]
MG1655 ΔfadD, SaFabD↑ RcTE LB with 1.5% glu. ~1.4 C14, C16, C16:1 24 [77]
MG1655 ΔfadD, ScFabD↑ RcTE LB with 1.5% glu. ~1.4 C14, C16, C16:1 24 [77]
MG1655 PaAccA↑, PaFabD↑ SpTE M9 with 0.05% YE ~0.24 C12, C14, C16, C16:1, C18, C18:1, 24 [165]
C18:2
MG1655 ΔfadD, ΔsucC RcTE M9 with 1.5% glu. 1.3 C14, C16, C16:1, C18:1 48 [102]β
MG1655 ΔfadD, FabZ↑ RcTE M9 with 1.5% glu. 1.7 C14, C16, C16:1 48 [102]
DH1 ΔfadE TesA’ Minimal with 2% glu. 3.8 C12 to C20 (SFA and UFA) 72 [166]
DH1 ΔfadE, FadR↑ TesA’ Minimal with 2% glu. 5.2c C12 to C20 (SFA and UFA) 72 [35]
MG1655 ΔfadD, ΔsucC, FabZ↑ RcTE LB with 1.5% glu. 5.7 C14, C16, C16:1, C18: l NA [130]
W3110 ΔfadD TesA’ MR with 1% glu., 0.3% YE 0.31 C8, C10, C12, C14, C14:1, C16, C16:1 NA [3]
W3110 ΔfadD TesB MR with 1% glu. ~0.18 C8, C10, C12, C14, C14:1, C16, C16:1 NA [3]
W3110 ΔfadD UcTE MR with 1% glu. ~0.12 C8, C10, C12, C14, C14:1, C16, C16:1 NA [3]
BL21 ΔfadD and modular optimization of CnFatB2 MK with 2% glu., 1% YE 8.6 (FB) C14, C16, C16:1, C18, C18:1 70 [122]
multiple genesγ
BL21 ΔfadD, Acc↕, FabADGI↕ TesA’ MK with 2% glu., 1% YE 3.9 (FB) C14, C16, C16:1, C18, C18:1 44 [127]
MG1655 ΔfadD, ΔfabR, FabZ↑, NadK↑ RcTE LB with 1.5% gly. 4.5 C14, C16, C16:1, C18:1 72 [145]
Biotechnology and Bioprocess Engineering 24: 000-000 (2019)
Table 1. Continued
E. coli Genetic modifications Thioesterase Media Titer (g/L) FA composition Time (h) Reference
MG1655 ΔfadD, ΔfabR, FabZ↑, NadK↑, RcTE LB with 1.5% gly. 4.8 C14, C16, C16:1, C18:1 72 [145]
PntAB↑
MG1655 ΔfadD, ΔfabR, FabZ↑, NadK↑, RcTE LB with 1.5% c.gly. 3.5 C14, C16, C16:1, C18:1 72 [145]
PntAB↑
MG1655 ΔfadD, Δaas, FadR↑ UcTE TB with 1.5% gly. 0.9 C12, C12:1, C14, C14:1 48 [20]
MG1655 ΔfadD, Δaas, FadR↑ UcTE TB with 1.5% gly. with 1.4 C12, C12:1, C14, C14:1 48 [20]
dodecane overlay
MG1655 ΔldhA, ΔpoxB, Δpta, ΔadhE, Ydil LB with 2% gly. 1.1 C6 - C10 48p.i. [4]
ΔfrdA, ΔfadA, ΔtesB, RtBktB↑, (RBO;MCFA)
FadB↑, EgTER↑
MG1655 ΔldhA, ΔpoxB, Δpta, ΔadhE, ΔfrdA, TesA’ LB with 2% gly. 1.3 C6 - C10 48p.i. [167]
ΔfadE, Δ6TEs, RtBktB↑, FadB↑, (RBO;MCFA)
EgTER↑
MG1655 -- mTesA’ M9 with 3% glu. 0.1% YE 1.1 C12, C14, C16, C18:1 48p.i. [114]
MG1655 PPC↑, PfMMC↑ TesA’ M9 with 2% glu. 0.1% YE 1.5 C12, C14, C16, C18:1 48p.i. [73]
MG1655 PfMMC↑ TesA’ M9 with 2% glu., 0.1% YE, 1.8 C12, C14, C16, C18:1 48p.i. [73]
5 mM aspartate
BL21 RtBktB↑, FadB↑,EgTER↑, multiple YdiI Minimal with 1.5% glu. 3.8 C6 - C10 48p.i. [101]
geneδ repressions by sgRNAs (RBO;MCFA)
(CRISPRi)
BL21 RtBktB↑, FadB↑, EgTER↑, Acs↑, YdiI Minimal with 1.5% glu. 4.7ε C6 - C10 48p.i. [148]
CbFdh↑, BsPck↑ 0.3% sodium acetate, 0.1% (RBO;MCFA)
sodium formate and 1%
NaHCO3
MG1655 ΔompF RcTE Minimal with 2% glu. 1.7 C14, C16, C16:1, C18:1 72 [157]
Microbial Production of Fatty Acid Via Metabolic Engineering and Synthetic Biology
MG1655 FadL↑ RcTE Minimal with 2% glu. 1.8 C14, C16, C16:1, C18:1 72 [157]
MG1655 ΔompF, FadL↑ RcTE Minimal with 2% glu. 2.3 C14, C16, C16:1, C18:1 72 [157]
MG1655 ∆fadE, ∆fumAC, ∆pta, FadZ↑ AtTE M9 with 1.5% glucose 1.0ζ (FB) C8 100 [168]
MG1655 ∆fadE, ∆fumAC, ∆pta, FadZ↑ AtTE M9 with 1.5% glucose 1.4 (FB) C6, C8, C8-OH, C10, C10-OH, C12, 100 [168]
C14, C16:1, C18, C18:1
Significant synthetic biology and metabolic engineering approaches applied in the last decade to enhance the FA production in various E. coli strains are listed in the table in ascending order of pub-
lication date. At, Anaerococcus tetradius; Ac, Acinetobacter baylyi; Bs, Bacillus subtilis; Cb, Candida boidinii; Cc, Cinnamomum camphorum; Cn, Cocos nucifera;Db, Diploknema butyracea; Eg,
Euglena gracilis; Gh, Gossypium hirsutum; Jc,Jatropha curcus; Pa, Pseudomonas aeruginosa; Pf, Propionibacterium freudenreichii; Rc, Ricinus communis; Rt, Ralstonia eutropha ; Sp, Streptococ-
cus pyogenes; Sc, S. coelicolor; Sa, S. avermitilis; Uc, Umbellularia californica; TE, thioesterase; TesA’, truncated cytosolic E. coli TE 1; mTesA’, mutant TesA ; FB, fed batch fermentation; BR,
C46R
batch fermentation in bioreactor; RBO, Reversal of beta-oxidation pathway; CRISPRi, clustered regularly interspaced short palindromic repeats interference; sgRNA, single guide RNA to imple-
ment the CRISPRi system; MCFA, medium-chain length fatty acids, C6-C10 carbons; 6TEs, E. coli TEs - yciA, ybgC, ydiI, tesA, fadM and tesB; LB, Luria Bertani broth; TB, terrific broth; YE,
yeast extract; glu, glucose; gly, glycerol; c.gly, crude glycerol; p.i., post-induction; SFA, saturated fatty acids; UFA, unsaturated fatty acids; Δ, deletion; ↑, overexpression; ↕, expression of genes con-
trolled by the intracellular level of malonyl-CoA through Bacillus subtilis FapR; , TesA’ and CcTE were expressed separately on two plasmids under the control of pPBAD and T7 promoter,
a
respectively. , TesA’ and CcTE were co-expressed on single plasmid under the control of pPBAD promoter; , for the expression of FadR under the control of pBAD promoter in a biobrick plas-
b c
mid, 1% arabinose was used. , first report showing that E. coli can overproduce FAs by metabolic engineering of the FAB pathway; , OptForce procedure was used for prioritizing genetic manipu-
α β
lations and suggesting that no universal engineering strategy was predicted for overproduction of all fatty acids, indicating the chain length specificity of each of the strategies; , pgk, gapA, aceE,
γ
aceF and lpdA on medium copy number plasmid, fabD*, accA, accB, accC and accD on low copy number plasmid, CnfatB2, fabA, fabH, fabG and fabI on high copy number plasmid; , δ
pta, adhE, frdA and ldhA; , the highest titers of C6-C10 chain length distribution reported to date; , the highest titer of C8 FA to date; *, mutants; NA, Not available; Gene/protein without any suf-
ε ζ
Fig. 3. Schematic representation of successful synthetic biology and metabolic engineering strategies to overproduce FFAs. It includes i)
deletion or downregulation of FPW_genes channelizing the metabolism away from native fermentation products to generate acetyl-CoA;
ii) overexpression or optimized expression levels of FAB_genes including acc and fabD with upstream and downstream flux; iii)
deletion of FAD_genes including fadL and fadD to avoid FA degradation; iv) expression of chain length specific acyl-ACP_TEs or acyl-
CoA_TEs to produce desired FFA(s); (v) dynamic control by transcription regulators, FadR and FabR; (vi) enhancing acetyl-CoA supply
or energy efficient malonyl-coA accumulation by PfMMC; (vii) over expression of FA exporter; (viii) increasing the membrane rigidity
or tolerance to toxic FAs; (ix) reversing β-oxidation pathway with changing genetic background; (x) balancing and regenerating co-
factors. To enhance the production of FFAs, expression level of genes colored green are increased or optimized and also genes colored
red are deleted or downregulated. EMP, Embden-Meyerhof-Parnas pathway; PPP, pentose phosphate pathway; EDP, Entner-Doudoroff
pathway; FPW, fermentative pathway; TCA, tricarboxylic acid cycle; ■ , gene targets (e.g. sucC, fumAC, frdA) to downregulate TCA
cycle; ●, repressed by FadR through IclR. TA and TK in PPP refer the transaldolase and transketolase reactions. PDHc refers pyruvate
dehydrogenase complex which consists of multimers of three distinct subunits (encoded by aceE, aceF and lpdA); E1, thiolase (e.g.,
RtBktB, yqeF, atoB, fadA); E2, hydroxyacyl-CoA dehydrogenase (e.g., fadB); E3, enoyl-CoA hydratase (e.g., fadB); E4, trans-enoyl-
CoA reductase (e.g., EgTER, ydiO); G6P, D-glucose-6-phosphate; F6P, fructose-6-phosphate; GAP, glyceraldehyde-3-phosphate; PEP,
phosphoenolpyruvate; Gly3P, glycerol-3-phosphate; DHAP, dihydroxyacetone phosphate; 6PGNL, 6-phospho D-glucono-1,5-lactone;
6PGNT, D-gluconate 6-phosphate; KDPG, 2-keto-3-deoxy-6-phosphogluconate. The readers are referred to the text for more details.
Microbial Production of Fatty Acid via Metabolic Engineering and Synthetic Biology 9
pathways, FA synthesis and lipid A synthesis, share the FA (MCFA) [4,100,101]. Wu and co-workers [101] obtained
same substrate, β-hydroxymyristoyl-ACP, the expression a 36% increase in the MCFA titer by repressing the
of the genes in the lipid A cluster is tightly regulated to fermentative pathways in E. coli. Another approach to
balance between FA and lipid A synthesis [87]. Thus, high increase the acetyl-CoA was conducted by reducing the
levels of FabZ improved the content of palmitic acid and activity of the TCA cycle, one of the acetyl-CoA consuming
stearic acid [59]. The FabZ prefers short- and long-chain pathways, which results in an improved FFA titer [102].
substrates, whereas the FabA has a high specificity toward However, in many cases, the acetyl-CoA competing
medium chain substrates [86]. The FabA is essential to pathways encode essential enzymes involved in biomass
synthesize UFAs, and it was found that the inactivation of formation or energy production. For this reason, manipulating
the FabA could survive only when the UFA is supplemented such pathways for rerouting carbon flux should be carefully
[29,88-90]. The isomerization reaction catalyzed by the carried out. Therefore, additional genetic strategies might
FabA is readily reversible in a high level of product be required in regulating acetyl-CoA availability, such as a
because of its non-negative Gibbs free energy [10,22]. toggle switch system [103].
Thus, the enzyme involved in the next step, β-ketoacyl- The inactivation of β-oxidation significantly improves FA
ACP synthase I (FabB), should be co-expressed to increase production in several studies [71,96]. However, contrary
the ratio of UFAs [91]. The FabB has broad substrate results were also reported: inactivation of β-oxidation does
specificity from C2-ACP to C14-ACP, but insufficient not show a positive effect on FA production [98,104]. The
activity to longer substrates [60,92]. The deletion of the different achievements might be because β-oxidation is
fabB cannot be compensated by the action of other β- subjected to carbon catabolite repression in the culture
ketoacyl-ACP synthases and leads to UFA auxotroph [93]. conditions containing glucose [10]. The degradation of the
Thus, the expression of both enzymes, FabA and FabB, FAs is prevented by the elimination of the beta-oxidation
should be optimized to synthesize UFAs [91], but it should pathway genes fadD (acyl-CoA synthase) or fadE (acyl-
also be carefully regulated because an increased ratio of CoA dehydrogenase) [70,96]. Furthermore, the deletion of
UFAs in the membrane can cause the loss of cell viability the fadL (outer membrane transport protein) has been
[19]. reported to improve FA production by inhibiting the re-
It was reported that the overexpression of the FabI does uptake of the extracellular FAs [104].
not show any significant increase in the FA titer. This To synthesize the FFAs, thioesterase is required that
might be due to the increased ratio of NAD+/NADH, catalyzes the hydrolysis of thioester bond in fatty acyl-
causing cofactor imbalance [59,94]. Thus, FabI expression ACP or fatty acyl-CoA to produce FFA. In E. coli, the
might require cofactor regeneration through additional acyl-CoA thioesterase I (TesA, encoded by tesA) has been
engineering such as expression of NAD+-dependent malic frequently employed to increase FFA production, as its
enzyme or erythrose-4-phosphate dehydrogenase, as they characteristics are well-studied. The catalytic mechanism,
significantly improve intercellular NADH concentration substrate specificity, and crystal structure of the TesA were
[95]. well-identified by several studies [105-109]. The TesA
The elimination of competing pathways that either degrade prefers long chain acyl-CoAs or acyl-ACPs (> C12)
the product (β-oxidation) or degrade the intermediates and [110,111], and it is located in the periplasmic space [112].
precursors (fermentative pathways) is one of the con- The cytosolic form of TesA (‘TesA) by removing leader
ventional approaches to enhance the production of the sequence significantly increases FFA production, as it
desired products [71,74,96]. Several competing pathways relieves the feedback inhibition resulting from the
at the branch point in pyruvate to acetyl-CoA were accumulation of the acyl-ACP [113]. Further engineering
inactivated to a provide high level of acetyl-CoA, leading to ‘TesA to increase FFAs production or alter its composition
an improved titer of the FFA products [74,97]. To investigate results in considerable successes in E. coli [3,109,114]. A
the effect of the acetyl-CoA consuming pathway on FFA number of thioesterases have been identified from various
production, the acetate formation pathway was inactivated organisms and expressed in E. coli to produce a variety of
[98]. Although the engineering strategy significantly reduced products, such as SCFA, MCFA, or UFA [91,99,115,116].
acetate formation, it did not improve the FFA yield, Thus, the use of different thioesterase not only accelerates
indicating that acetyl-CoA consumption is not a limiting FFA production but also alters the type of FFA produced.
factor [98]. Moreover, it was reported that acetate formation Although the expression of thioesterase significantly
is already reduced in efficient FFA producers [99]. On the improves FFA production, it causes several negative impacts
other hand, the deletion or down regulation of several genes on the physiology, as a high level of FFA has antibacterial
involved in the formation of the fermentative products effects, such as membrane instability, disruption of oxidative
increased the short-chain FA (SCFA) or the medium-chain phosphorylation, and interference with the electron transport
10 Biotechnology and Bioprocess Engineering
chain. For example, an excess expression of medium yield) under controlled conditions in the E. coli. Previously,
chain-specific thioesterase drains the acyl-ACP destined to seeking the optimal molar ratio of each component in the
become a membrane lipid, resulting in an impediment in FA synthase in vitro could construct a highly efficient
cellular growth [70,117]. Moreover, the resorption of the strain, producing around 50% more FA than a control
medium chain products destabilizes the membrane [117]. strain that expresses only one component of the FA
These toxic effects should restrain for further increase in synthase [125]. Moreover, the optimal ratio found in the
the production of the FFAs and their derivatives. Therefore, study was similar with the molar ratio in a high FA-
additional engineering should be accompanied, such as the producing strain from other recent research [126]. The
expression of a transporter, the block of resorption, and the results from in vivo and in vitro analysis of the protein ratio
recovery of membrane saturation [19,20,104,118,119]. suggest that the balanced activity of the FA synthase is
In the FAS II route, one costly malonyl-CoA is required required to maximize FFA production.
to extend the acyl-CoA chain by two carbons at every A genetically encoded metabolic switch, with rewiring of
elongation cycle. The biosynthesis of malonyl-CoA from the transcriptional regulator FabR, dynamically regulates
acetyl-CoA requires an ATP. On the other hand, reverse the pathway expression and compensates the metabolic
beta-oxidation (RBO) is considered as an energy efficient activity of critical enzymes based on the intercellular level
route, as it directly uses the acetyl-coA for acyl-chain of the malonyl-CoA, which results in the balanced
elongation instead of the malonyl-CoA. By employing metabolism between cell growth and product formation
variants of thiolases, thioesterases, and native dehydro- and significantly improves FAs production [127]. McKee
genases, alcohols and carboxylic acids with various chain and co-workers [128] showed that the neutralization of the
lengths and functionalities are produced through RBO carbon-storage regulator CsrA (RNA-binding protein) by
route in E. coli [4,120]. Dellomonaco and co-workers overexpressing its non-coding RNA antagonist, CsrB, can
[120] produced around 7 g/L of FAs from glucose in enhance the FA productivity by upregulating the glycolysis/
E. coli through the engineered RBO pathway. gluconeogenesis, the TCA cycle, and the PPP.
The accumulation of the acyl-ACP inhibits the activity
4.2. Controlling the metabolic flux of the ACC, FabI, and FabH through feedback inhibition,
FA synthesis from glucose involves multigene pathways. thereby reducing the activity of the FA biosynthesis. Thus
Just increasing the concentration of any one of the by eliminating the accumulation of the acyl-ACP through
precursors of the pathway does not result in the high the expression of thioesterases, it catalyzes the terminal
production of the desired compound. Precursor flux reaction to produce free FAs and it is also crucial in
improvement may not be accommodated by downstream controlling the metabolic flux toward FA synthesis
pathways (imbalanced flux), or accumulated and depleted [70,96,129]. The reported dual advantages of thioesterase
intermediates may stimulate a stress response that com- expression include providing a product sink and depleting
promises cell viability and pathway productivity [121,122]. a key regulatory signal, which increases the flux toward
For instance, increased FabF activity stops cell growth with oleochemicals [23]. However, the expression level of the
the high accumulation of malonyl-CoA, indicating that it thioesterase must be tuned carefully with the downstream
inhibits the process associated with converting malonyl-CoA pathway flux, as it has been shown that too much
to malonyl-ACP [74,123]. Moreover, ACP overexpression thioesterase activity could impair the FFA by ultimately
has led to an inefficient post-translational modification and destabilizing the membrane and impeding growth [10,22,
accumulation of the apo-form of the ACP [124]. The apo- 70,71,125].
ACP acts as a growth inhibitor by interrupting the activity The redirection of the TCA cycle flux by deleting sucC,
of the glycerol-3-phosphate acyltransferase (encoded by fumAC, and gltA toward FA production improves the
plsB). Therefore, elaborate manipulations are required for MCFA production [130] and other malonyl-CoA- derived
the construction of an efficient FFA-producing E. coli. products [102], but this strategy reduces energy generation
Xu and co-workers [122] optimized (modular pathway and reduces the rate and yield of aerobic biomass
optimization) the expression of the different pathway genes generation [23]. The Homologous overexpression of the
involved in the production of FAs from glucose through fabZ results in about a 2-fold increase in FA concentrations
the acyl-CoA by altering the plasmid copy number leading by pulling the carbon flux toward the FA elongation cycle
to a balance in the supply and consumption of the FA [130,131], and the highest yield is achieved when FabZ is
intermediates (acetyl-CoA and malonyl-ACP). Furthermore, at least 10-fold more abundant than other Fab enzymes
the translational efficiency was improved by customizing [125]. Recent reports show the abundance of FabZ in the
the ribosomal binding sites of the FA pathway modules, high-yield strain, 4.5-fold higher than in the background
thus producing altogether 9 g/L FAs (22% theoretical strain; this implies that FabZ is one of the most important
Microbial Production of Fatty Acid via Metabolic Engineering and Synthetic Biology 11
enzymes to improve FA production [125,126]. needed to maintain the redox balance (by expressing
Another way to enhance the FA flux is the overexpression transhydrogenase or altering cofactor specificity of the
of the transcription factor FadR, which regulates FA NADH-consuming enzyme) or to employ genetically
metabolism, catabolism and transport. It upregulates fabA, encoded biosensor that dynamically regulates the synthesis
fabB, and accA leading to the synthesis of FAs and of the NADPH depending on its cellular demands [140-
downregulates the FA degradation genes fadE, fadBA, 143]. Another strategy is utilizing transhydrogenase, inter-
fadH, and fadIJ [30]. The overexpression of any single converting the NADH and NADPH. During the aerobic
gene such as fabA (1.1-fold), fabB (2.3-fold), fadBA (1.7- growth of E. coli in a glucose medium, 35-45% of the
fold), fabF (3.0-fold), fabZ (0 to 3-fold) and accABCD NADPH is synthesized by PntAB (membrane-bound
(1.3-fold), does not show a high yield comparable to the transhydrogenase, encoded by pntAB) [144]. However, the
fadR (7.4-fold) expression (5.2 g/L FAs, reaching 73% of improvement of FFA is modest, around 7.5%, in the strain
theoretical yield), indicating that FadR acted on a global overexpressing the PntAB [145]. Interestingly, the over-
scale, by tuning the expression levels of the FA-pathway expression of NadK (NAD kinase, encoded by nadK)
genes to optimal levels for the production of FAs [35, 71, shows around 20% higher FFA than the strain lacking the
125]. The overexpression of the FadR causes the induction overexpression of NadK. The overexpression of the NadK
of fabA and fabB, leading to an increase of the UFA increases the intracellular concentration of both the
content from 13% to 43% in E. coli strains [35]. A recent NADPH and NADP+ with a 2-fold higher expression of the
report shows that the FA over-producing strain is sensitive PntAB and UdhA (soluble transhydrogenase, encoded by
to metabolic burden and oxygen influx [132]. Energy udhA) [146]. The results indicate that the disturbance of the
metabolism becomes a rate-limiting factor for FA over- intracellular redox metabolism caused by the NADPH
production in E. coli strains. A membrane protein facilitating overproduction should be restored, since the high expression
O2 transport, Vitreoscilla hemoglobin, was introduced into of transhydrogenases with NadK expression only increases
a fatty-acid-producing strain to promote oxygen supply and FFA production. Instead of balancing the NADPH supply,
energy metabolism thus enhancing FA production by 70% Javidpour and co-workers [147] overexpressed the NADH-
[132]. dependent fabG homolog from Cupriavidus taiwanensis in
a FA overproducing E. coli strain, leading to a 60% higher
4.3. Balancing cellular redox metabolism FFA titer under anaerobic culture conditions. Recently,
The balance of cofactors and their regeneration is often RBO-mediated synthesis of the MCFA (C6-C10) achieved
required, as the production of desired chemicals is mediated 4.7 g/L of MCFA, the highest titer reported to date, by
by biological enzymes. Improving, balancing, and rege- balancing the NADH requirement along with the increasing
nerating the cofactors have been carried out through availabilities of acetyl-CoA and ATP through the over
metabolic engineering, such as the redirection of the expression of genes (acs, acetyl-CoA synthetase; fdh, formate
metabolic flux toward PPP [133,134], the replacement of dehydrogenase; pck, phosphoenolpyruvate carboxykinase)
central metabolic enzyme generating other cofactors [135], in the MCFA-producing E. coli strain [148].
the interconversion of NADH and NADPH [136], and the The ACP is also a cofactor in FFA production and one
engineering of proteins to change cofactor specificity [137]. of the most abundant soluble proteins in E. coli, accounting
FA formation is an energy-intensive process requiring eight for 0.25% of the total soluble protein [149]. The over-
moles of ATP, NADH, and NADPH each to produce one expression of the ACP (encoded by acpP) increases the
mole of palmitic acid. Moreover, cofactors are also required intracellular production of apo-ACP; however, it significantly
in terminal reductions to convert acids into aldehydes, inhibits cell growth (around 500-fold less cell viability
alcohols, olefins, and alkanes [23]. Therefore, it is important compared with the strain-lacking ACP overexpression)
to consider the availability of the cofactors for enhancing [124]. Interestingly, the low level of ACP overproduction
FFA production. (around 8-fold) with thioesterase expression (E. coli ʹTesA)
A strategy to increase the intracellular concentration of increases the FA synthetic rate by 5- to 6-fold, compared
the NADPH is to redirect carbon flux into the PPP. The with the strain-expressing ʹTesA alone. Further studies
deletion of the pgi encoding glucose-6-phosphate isomerase revealed that the apo-ACP inhibits the activity of glycerol-
can improve the NADPH/NADP+ ratio up to 3- and 2-fold 3-phosphate acyltransferase (encoded by the plsB and
from the theoretical prediction and experimental data, as catalyzing the first committed step in phospholipid bio-
compared with the wild-type E. coli [134,138]. However, synthesis) [124,150]. This indicates that the accumulated
an excess level of the NADPH reduces the expression of acyl-ACP through a reduction in glycerol-3-phosphate
citrate synthase in the TCA cycle, glucose uptake, and cell acyltransferase activity might be converted into FFA and
growth [139,140]. Therefore, additional engineering is might increase the FA synthetic rate in strains co-
12 Biotechnology and Bioprocess Engineering
overexpressing ʹTesA and ACP. Increasing the ACP MCFAs. The overexpressing of acrE, mdtE, and mdtC
concentration in vitro of reconstructed FA synthase also together with a deleting cmr increases MCFA production
increases the FA synthetic rate, but a decline in the by more than 2-fold [156].
synthetic rate is observed over a certain level of the ACP Reducing the stress resulting from FFA accumulation is
[125]. a productive strategy to construct the industrially relevant
strain. One of the strategies that decrease FFA toxicity is
4.4. Increasing the stress tolerance overexpression of a native FFA exporter or a heterologous
Modulating the stress or tolerance of FAs generally exporter. The strategy of engineering two transporters
influences the FA or other bio-product titers. Among the (deletion of ompF and increased expression of fadL)
many stress factors, membrane damage is a frequently increases membrane integrity and produces around a 53%
encountered factor in stress response [151]. The FAs inside higher FA titer (2.3 g/L) [157]. Another study found that
the cell may result in intracellular acidification and cell PssA (phosphatidylserine synthase) overexpression increases
membrane damage, thus causing low productivity [152]. the abundance of one of the three different phospholipids,
Increasing the tolerance to the FAs is essential for the phosphatidylethanolamine (PE), by 7%, which could improve
industrial production strains. The survival rate of the E. coli membrane integrity, membrane electrochemical potential,
BL21 strain, overexpressing dsrA and rcsB, against the and cellular hydrophilicity, thus resulting in both increased
heptanoic acid stress is 2-fold higher than that of the wild tolerance and production of the octanoic acid. The specific
type strain. The proteins, DsrA and RcsB, are involved in growth rate of a pssA overexpressing strain increases by
the glutamic acid-dependent acid resistance [152,153]. A 29% over the control strain in the presence of 20 mM of
mutant E. coli strain obtained by adaptive laboratory octanoic acid [158].
evolution for octanoic acid tolerance improves the octanoic Another approach to avoid FA toxicity is culturing with
acid and total FAs titer by 5-fold as compared to the parent an organic layer (two-liquid phase cultivation system),
strain when expressing the Anaerococcus tetradius thio- which increases the production and extracellular ratio of
esterase [154]. The AcrB mutants obtained by directed the FFA or FFA derivatives [20,159]. The secretion of the
evolution have been shown to overcome toxicity problems FFA relieves toxicity caused by the FFA accumulation in
and to improve product production. The mutants show cytoplasm, resulting in the reduction of acid stress and
improved efflux efficiency for n-octane and α-pinene by up oxidative stress, as well as a reduction in FFA lipotoxicity
to 47% and 400%, respectively [155]. Modulating the acyl- lesions. These studies suggest that rational metabolic
ACP pools has been shown to influence the cell membrane engineering strategies are needed, especially in membrane
integrity. The expression of an unsaturated acyl-ACP engineering to increase tolerance and production of FAs or
targeting thioesterase GeoTE from Geobacillus sp. reduces other FA-derived products.
membrane UFAs, thus increasing cell viability and FFA
productivity [19]. The AcrAB-TolC efflux pump is
responsible for most of the FFA efflux in E. coli [19]. 5. Future Perspectives and Conclusions
The deletion of two regulators, FadR and MarR (multiple
antibiotic resistance repressor, encoded by marR), exhibits The redesigning E. coli through metabolic engineering and
increased solvent tolerance by maintaining the high ratio of synthetic biology has led to increased FFA production by
SFA to UFA and by tightening the cell membrane through identifying rate-limiting enzymes, optimizing metabolic
changing the FA composition and increasing the activity of pathways, introducing mutations in enzymes, and
the AcrB-TolC efflux pump [119]. The deletion of the aas constructing alternative pathways. This combination of
gene, which encodes the 2-acyl-GPE acyltransferase/acyl- strategies has achieved high yields and provided the base
ACP synthase, increases the FFA yield by reducing the for further engineering. Significant improvements have
toxicity and incorporation of MCFA into the membrane been achieved in the microbial production of FFA through
phospholipids. With co-expressing the E. coli FadR and metabolic engineering and synthetic biology with the
Umbellularia californica acyl-ACP thioesterase (BTE) in a highest yield reaching 73% of the theoretical yield in
ΔfadD background, the aas knock-out strain exhibits E. coli [35]. To be industrially relevant strain, the TRY
increased tolerance of MCFA toxicity, partially restored (titer, rate, and yield) of FFA production should be
membrane composition, and around a 126% higher improved by additional metabolic or protein engineering
MCFFA yield than the control strain [20]. Recently, it was that could lead to stable production in the stationary phase,
reported that transporters such as AcrE, MdtC, and MdtE increased tolerance to various stresses and efficient/optimal
play an important role in the efflux of MCFAs, whereas expression of the pathway genes.
Cmr, FadL, MarA, and SoxS are involved in the influx of The transition of cell growth from the exponential phase
Microbial Production of Fatty Acid via Metabolic Engineering and Synthetic Biology 13
to the stationary phase leads to a lot of physiological coli to synthesize free fatty acids. Trends Biotechnol. 30: 659-
changes in the uptake of nutrients, central metabolism, and 667.
2. Cao, Y. X., W. H. Xiao, J. L. Zhang, Z. X. Xie, M. Z. Ding, and
metabolite synthesis. As primary metabolites, FA and Y.J. Yuan (2016) Heterologous biosynthesis and manipulation
phospholipid synthesis are correlated with the cell growth of alkanes in Escherichia coli. Metab. Eng. 38: 19-28.
rate [160]. During the stationary phase, mostly pathways 3. Choi, Y. J. and S. Y. Lee (2013) Microbial production of short-
related to the stress associated with metabolites are tuned chain alkanes. Nature 502: 571-574.
4. Clomburg, J. M., M. D. Blankschien, J. E. Vick, A. Chou,
on, thus decreasing or inhibiting the desired pathway flux S. Kim, and R. Gonzalez (2015) Integrated engineering of
[53,161]. Since the production of the desired compounds β-oxidation reversal and ω-oxidation pathways for the synthesis
during the stationary phase might eliminate the costly of medium chain ω-functionalized carboxylic acids. Metab.
Eng. 28: 202-212.
trade-off between biomass production and product synthesis 5. Sudheer, P. D. V. N., J. Yun, S. Chauhan, J. Kang, and K. Y.
[162], further host engineering, such as increasing the Choi (2017) Screening, expression, and characterization of
uptake of a carbon source or an efficient FFA synthesis, is Baeyer-Villiger monooxygenases for the production of 9-
needed to be performed to design stable metabolism so that (nonanoyloxy)nonanoic acid from oleic acid. Biotechnol.
Bioprocess Eng. 22: 717-724.
carbon is solely directed into FFA production in the 6. Lennen, R. M. and B. F. Pfleger (2013) Microbial production of
stationary phase. fatty acid-derived fuels and chemicals. Curr. Opin. Biotechnol.
Protein engineering of enzymes in the FA synthetic 24: 1044-1053.
pathway is also an attractive strategy to develop high- 7. Do, K. H., H. M. Park, S. K. Kim, and H. S. Yun (2018)
Production of cis-vaccenic acid-oriented unsaturated fatty acid
yielding pathways, for example, by reducing allosteric in Escherichia coli. Biotechnol. Bioprocess Eng. 23: 100-107.
regulation or by altering cofactor requirements. The 8. Sherkhanov, S., T. P. Korman, S. G. Clarke, and J. U. Bowie
expression of long-chain-specific thioesterases such as (2016) Production of FAME biodiesel in E. coli by direct
methylation with an insect enzyme. Sci. Rep. 6: 24239. doi:
E. coli ʹTesA, could relieve the feedback inhibition caused 10.1038/srep24239
by the acyl-ACP accumulation. The inhibition could be 9. Demirbas, A (2009) Progress and recent trends in biodiesel
problematic in the production of short- or medium-chain fuels. Energy Convers. Manag. 50: 14-34.
FA with short- or medium-chain specific thioesterases, 10. Janßen, H. and A. Steinbüchel (2014) Fatty acid synthesis in
Escherichia coli and its applications towards the production of
since the accumulation of long-chain acyl-ACP inhibits the fatty acid based biofuels. Biotechnol. Biofuels 7: 7. doi:
activity of the FA synthase [1]. Indeed, the co-expression 10.1186/1754-6834-7-7
of the medium- and long-chain specific thioesterases in 11. Wu, L. F., P. C. Chen, A. P. Huang, and C. M. Lee (2012) The
E. coli can further increase medium-chain FA production, feasibility of biodiesel production by microalgae using industrial
wastewater. Bioresour. Technol. 113: 14-18.
indicating that allosteric regulation in several enzymes 12. Zhang, F., L. H. Cheng, W. L. Gao, X. H. Xu, L. Zhang, and
needs to be reduced in the production of medium- or short- H. L. Chen (2011) Mechanism of lipid extraction from
chain FA. Botryococcus braunii FACHB 357 in a biphasic bioreactor. J.
Biotechnol. 154: 281-284.
To conclude, the genomic level optimization of FFA
13. Almeida, J. R. M., L. C. L. Fávaro, and B. F. Quirino (2012)
production is further beneficial because of the various Biodiesel biorefinery: opportunities and challenges for microbial
disadvantages of plasmid level metabolic engineering. production of fuels and chemicals from glycerol waste.
Engineering microbial strains to develop microbial platforms Biotechnol. Biofuels 5: 48. doi: 10.1186/1754-6834-5-48
14. Ye, J., Y. Sha, Y. Zhang, Y. Yuan, and H. Wu (2011) Glycerol
based on various successful strategies and also further extracting dealcoholization for the biodiesel separation process.
tuning with future proficient studies could pave a way Bioresour. Technol. 102: 4759-4765.
toward the sustainable production of microbial FA from 15. Rahman, Z., N. Rashid, J. Nawab, M. Ilyas, B. H. Sung, and
inexpensive and renewable biomass sources at an industrial S. C. Kim (2016) Escherichia coli as a fatty acid and biodiesel
factory: current challenges and future directions. Environ. Sci.
scale. Pollut. Res. 23: 12007-12018.
16. Seo, E.-J., Y. J. Yeon, J.-H. Seo, J.-H. Lee, J. P. Boñgol, Y. Oh,
J. M. Park, S.-M. Lim, C.-G. Lee, and J.-B. Park (2018)
Enzyme/whole-cell biotransformation of plant oils, yeast
Acknowledgement derived oils, and microalgae fatty acid methyl esters into n-
nonanoic acid, 9-hydroxynonanoic acid, and 1,9-nonanedioic
This work was supported by the Next-Generation BioGreen acid. Bioresour. Technol. 251: 288-294.
21 Program funded by the Ministry of Agriculture, Food 17. Jeon, E.-Y., J.-H. Seo, W.-R. Kang, M.-J. Kim, J.-H. Lee,
D.-K. Oh, and J.-B. Park (2016) Simultaneous enzyme/whole-
and Rural Affairs (SSAC, Grant#: PJ013457). cell biotransformation of plant oils into C9 carboxylic acids.
ACS Catal. 6: 7547-7553.
18. Ghim, C. M., T. Kim, R. J. Mitchell, and S. K. Lee (2010)
References Synthetic biology for biofuels: building designer microbes from
the scratch. Biotechnol. Bioprocess Eng. 15: 11-21.
19. Lennen, R. M. and B. F. Pfleger (2013) Modulating membrane
1. Lennen, R. M. and B. F. Pfleger (2012) Engineering Escherichia composition alters free fatty acid tolerance in Escherichia coli.
14 Biotechnology and Bioprocess Engineering
PLoS One 8: e54031. doi: 10.1371/journal.pone.0054031 expression of a repressor protein: FadR activates iclR. J.
20. Sherkhanov, S., T. P. Korman, and J. U. Bowie (2014) Improving Bacteriol. 178: 4704-4709.
the tolerance of Escherichia coli to medium-chain fatty acid 40. Kornberg, H. L. (1966) The role and control of the glyoxylate
production. Metab. Eng. 25: 1-7. cycle in Escherichia coli. Biochem. J. 99: 1-11.
21. Schumacher, S. D. and J. Jose (2012) Expression of active 41. Iram, S. H., and J. E. Cronan (2006) The β-oxidation systems of
human P450 3A4 on the cell surface of Escherichia coli by Escherichia coli and Salmonella enterica are not functionally
autodisplay. J. Biotechnol. 161: 113-120. equivalent. J. Bacteriol. 188: 599-608.
22. Handke, P., S. A. Lynch, and R. T. Gill (2011) Application and 42. Campbell, J. W., R. M. Morgan-Kiss, and J. E. Cronan (2003) A
engineering of fatty acid biosynthesis in Escherichia coli for new Escherichia coli metabolic competency: growth on fatty
advanced fuels and chemicals. Metab. Eng. 13: 28-37. doi: acids by a novel anaerobic β-oxidation pathway. Mol. Microbiol.
10.1016/j.ymben.2010.10.007 47: 793-805.
23. Pfleger, B. F., M. Gossing, and J. Nielsen (2015) Metabolic 43. Cho, B. K., E. M. Knight, and B. Ø. Palsson (2006)
engineering strategies for microbial synthesis of oleochemicals. Transcriptional regulation of the fad regulon genes of
Metab. Eng. 29: 1-11. Escherichia coli by ArcA. Microbiology 152: 2207-2219.
24. Guchhait R. B., S. E. Polakis, P. Dimroth, E. Stoll, and J. L. M. 44. Zhang, Y. M., H. Marrakchi, and C. O. Rock (2002) The FabR
Moss (1974) Acetyl coenzyme A carboxylase system of (YijC) Transcription factor regulates unsaturated fatty acid
Escherichia coli: purification and properties of the biotin biosynthesis in Escherichia coli. J. Biol. Chem. 277: 15558-
carboxylase, carboxyltransferase, and carboxyl carrier protein 15565.
components. J. Biol. Chem 249: 6633-6645. 45. Zhu, K., Y. M. Zhang, and C. O. Rock (2009) Transcriptional
25. Cronan, J. E. and G. L. Waldrop (2002) Multi-subunit acetyl- regulation of membrane lipid homeostasis in Escherichia coli.
CoA carboxylases. Prog. Lipid Res. 41: 407-435. J. Biol. Chem. 284: 34880-34888.
26. Lai, C. Y. and J. E. Cronan (2003) β-ketoacyl-acyl carrier 46. Feng, Y., and J. E. Cronan (2011) Complex binding of the FabR
protein synthase III (FabH) is essential for bacterial fatty acid repressor of bacterial unsaturated fatty acid biosynthesis to its
synthesis. J. Biol. Chem. 278: 51494-51503. cognate promoters. Mol. Microbiol. 80: 195-218.
27. Bergler, H., S. Fuchsbichler, G. Högenauer, and F. Turnowsky 47. Li, S. J. and J. E. Cronan (1993) Growth rate regulation of
(1996) The enoyl-[acyl-carrier-protein] reductase (FabI) of Escherichia coli acetyl coenzyme A carboxylase, which
Escherichia coli, which catalyzes a key regulatory step in fatty catalyzes the first committed step of lipid biosynthesis. J.
acid biosynthesis, accepts NADH and NADPH as cofactors and Bacteriol. 175: 332-340.
is inhibited by palmitoyl-CoA. Eur. J. Biochem. 242: 689-694. 48. James, E. S. and J. E. Cronan (2004) Expression of two
28. Feng, Y. and J. E. Cronan (2009) Escherichia coli unsaturated Escherichia coli acetyl-CoA carboxylase subunits is autoregulated.
fatty acid synthesis. J. Biol. Chem. 284: 29526-29535. J. Biol. Chem. 279: 2520-2527.
29. Kass, L. R. and K. Bloch (1967) On the enzymatic synthesis of 49. Meades, G., B. K. Benson, A. Grove, and G. L. Waldrop (2010)
unsaturated fatty acids in Escherichia coli. Proc. Natl. Acad. A tale of two functions: enzymatic activity and translational
Sci. U. S. A. 58: 1168-1173. repression by carboxyltransferase. Nucleic Acids Res. 38: 1217-
30. Fujita, Y., H. Matsuoka, and K. Hirooka (2007) Regulation of 1227.
fatty acid metabolism in bacteria. Mol. Microbiol. 66: 829-839. 50. Yao, Z., R. M. Davis, R. Kishony, D. Kahne, and N. Ruiz (2012)
31. Simons, R. W., P. A. Egan, H. T. Chute, and W. D. Nunn (1980) Regulation of cell size in response to nutrient availability by
Regulation of fatty acid degradation in Escherichia coli: fatty acid biosynthesis in Escherichia coli. Proc. Natl. Acad.
isolation and characterization of strains bearing insertion and Sci. 109: E2561-E2568. doi: 10.1073/pnas.1209742109
temperature-sensitive mutations in gene fadR. J. Bacteriol. 142: 51. Gerhardt, E. C. M., T. E. Rodrigues, M. Müller-Santos, F. O.
621-632. Pedrosa, E. M. Souza, K. Forchhammer, and L. F. Huergo
32. My, L., N. G. Achkar, J. P. Viala, and E. Bouveret (2015) (2015) The Bacterial signal transduction protein GlnB regulates
Reassessment of the genetic regulation of fatty acid synthesis in the committed step in fatty acid biosynthesis by acting as a
Escherichia coli: global positive control by the dual functional dissociable regulatory subunit of acetyl-CoA carboxylase. Mol.
regulator FadR. J. Bacteriol. 197: 1862-1872. Microbiol. 95: 1025-1035.
33. Xu, Y., R. J. Heath, Z. Li, C. O. Rock, and S. W. White (2001) 52. van Heeswijk, W. C., H. V. Westerhoff, and F. C. Boogerd
The FadR·DNA complex. J. Biol. Chem. 276: 17373-17379. (2013) Nitrogen assimilation in Escherichia coli: putting
34. Cronan, J. E. (1997) In vivo evidence that acyl coenzyme A molecular data into a systems perspective. Microbiol. Mol. Biol.
regulates DNA binding by the Escherichia coli FadR global Rev. 77: 628-695.
transcription factor. J. Bacteriol. 179: 1819-1823. 53. My, L., B. Rekoske, J. J. Lemke, J. P. Viala, R. L. Gourse, and
35. Zhang, F., M. Ouellet, T. S. Batth, P. D. Adams, C. J. Petzold, E. Bouveret (2013) Transcription of the Escherichia coli fatty
A. Mukhopadhyay, and J. D. Keasling (2012) Enhancing fatty acid synthesis operon fabHDG is directly activated by FadR and
acid production by the expression of the regulatory transcription inhibited by ppGpp. J. Bacteriol. 195: 3784-3795.
factor FadR. Metab. Eng. 14: 653-660. 54. Podkovyrov, S. and T. J. Larson (1995) Lipid biosynthetic genes
36. Iram, S. H. and J. E. Cronan (2005) Unexpected functional and a ribosomal protein gene are cotranscribed. FEBS Lett. 368:
diversity among FadR fatty acid transcriptional regulatory 429-431.
proteins. J. Biol. Chem. 280: 32148-32156. 55. Podkovyrov, S. M. and T. J. Larson (1996) Identification of
37. He, L., Y. Xiao, N. Gebreselassie, F. Zhang, M. R. Antoniewiez, promoter and stringent regulation of transcription of the fabH,
Y. J. Tang, and L. Peng (2014) Central metabolic responses to fabD and fabG genes encoding fatty acid biosynthetic enzymes
the overproduction of fatty acids in Escherichia coli based on of Escherichia coli. Nucleic Acids Res. 24: 1747-1752.
13
C-metabolic flux analysis. Biotechnol. Bioeng. 111: 575-585. 56. Durfee, T., A. M. Hansen, H. Zhi, F. R. Blattner, and D. J. Jin
38. Farewell, A., A. A. Diez, C. C. DiRusso, and T. Nyström (1996) (2008) Transcription profiling of the stringent response in
Role of the Escherichia coli FadR regulator in stasis survival Escherichia coli. J. Bacteriol. 190: 1084-1096.
and growth phase-dependent expression of the uspA, fad, and 57. Heath, R. J., and C. O. Rock (1995) Enoyl-acyl carrier protein
fab genes. J. Bacteriol. 178: 6443-6450. reductase (fabI) plays a determinant role in completing cycles of
39. Gui, L., A. Sunnarborg, and D. C. Laporte (1996) Regulated fatty acid elongation in Escherichia coli. J. Biol. Chem. 270:
Microbial Production of Fatty Acid via Metabolic Engineering and Synthetic Biology 15
26538-26542. 504.
58. Lee, S., S. Lee, Y. J. Yoon, and J. Lee (2013) Enhancement of 76. Leonard, E., K. H. Lim, P. N. Saw, and M. A. G. Koffas (2007)
long-chain fatty acid production in Escherichia coli by Engineering central metabolic pathways for high-level flavonoid
coexpressing genes, including fabF, involved in the elongation production in Escherichia coli. Appl. Environ. Microbiol. 73:
cycle of fatty acid biosynthesis. Appl. Biochem. Biotechnol. 3877-3886.
169: 462-476. 77. Zhang, X., A. Agrawal, and K. Y. San (2012) Improving fatty
59. Jeon, E., S. Lee, S. Lee, S. O. Han, Y. J. Yoon, and J. Lee (2012) acid production in Escherichia coli through the overexpression
Improved production of long-chain fatty acid in Escherichia of malonyl-CoA-Acyl carrier protein transacylase. Biotechnol.
coli by an engineering elongation cycle during fatty acid Prog. 28: 60-65.
synthesis (FAS) through genetic manipulation. J. Microbiol. 78. Lee, S., E. Jeon, H. S. Yun, and J. Lee (2011) Improvement of
Biotechnol. 22: 990-999. fatty acid biosynthesis by engineered recombinant Escherichia
60. Edwards, P., J. S. Nelsen, J. G. Metz, and K. Dehesh (1997) coli. Biotechnol. Bioprocess Eng. 16: 706-713.
Cloning of the fabF gene in an expression vector and in vitro 79. Tsay, J. T., W. Oh, T. J. Larson, S. Jackowski, and C. O. Rock
characterization of recombinant fabF and fabB encoded (1992) Isolation and characterization of the beta-ketoacyl-acyl
enzymes from Escherichia coli. FEBS Lett. 402: 62-66. carrier protein synthase III gene (fabH) from Escherichia coli
61. Garwin, J. L., A. L. Klages, and J. E. Cronan (1980) Beta- K-12. J. Biol. Chem. 267: 6807-6814.
ketoacyl-acyl carrier protein synthase II of Escherichia coli. 80. Abbadi, A., M. Brummel, and F. Spener (2000) Knockout of the
Evidence for function in the thermal regulation of fatty acid regulatory site of 3-ketoacyl-ACP synthase III enhances short-
synthesis. J. Biol. Chem. 255: 3263-3265. and medium-chain acyl-ACP synthesis. Plant J. 24: 1-9.
62. Mejia, R., M. C. Gómez-Eichelmann, and M. S. Fernandez 81. Tseng, H. C. and K. L. J. Prather (2012) Controlled biosynthesis
(1999) Fatty acid profile of Escherichia coli during the heat- of odd-chain fuels and chemicals via engineered modular
shock response. Biochem. Mol. Biol. Int. 47: 835-844. metabolic pathways. Proc. Natl. Acad. Sci. 109: 17925-17930.
63. Heath, R. J. and C. O. Rock (1995) Regulation of malonyl-CoA 82. Torella, J. P., T. J. Ford, S. N. Kim, A. M. Chen, J. C. Way, and
metabolism by acyl-acyl carrier protein and beta-ketoacyl-acyl P. A. Silver (2013) Tailored fatty acid synthesis via dynamic
carrier protein synthases in Escherichia coli. J. Biol. Chem. 270: control of fatty acid elongation. Proc. Natl. Acad. Sci. 110:
15531-15538. 11290-11295.
64. Davis, M. S. and J. E. Cronan (2001) Inhibition of Escherichia 83. Choi, K. H., R. J. Heath, and C. O. Rock (2000) Beta-ketoacyl-
coli acetyl coenzyme A carboxylase by acyl-acyl carrier protein. acyl carrier protein synthase III (FabH) is a determining factor
J. Bacteriol. 183: 1499-1503. in branched-chain fatty acid biosynthesis. J. Bacteriol. 182:
65. Heath, R. J. and C. O. Rock (1996) Inhibition of beta-ketoacyl- 365-370.
acyl carrier protein synthase III (FabH) by acyl-acyl carrier 84. Haushalter, R. W., W. Kim, T. A. Chavkin, L. The, M. E. Garber,
protein in Escherichia coli. J. Biol. Chem. 271: 10996-11000. M. Nhan, P. D. Adams, C. J. Petzold, L. Katz, and J. D. Keasling
66. Joshi, V. C. and S. J. Wakil (1971) Studies on the mechanism of (2014) Production of anteiso-branched fatty acids in Escherichia
fatty acid synthesis: XXVI. Purification and properties of coli:next generation biofuels with improved cold-flow properties.
malonyl-coenzyme A—Acyl carrier protein transacylase of Metab. Eng. 26: 111-118.
Escherichia coli. Arch. Biochem. Biophys. 143: 493-505. 85. Jiang, W., Y. Jiang, G. J. Bentley, D. Liu, Y. Xiao, and F. Zhang
67. Joshi, V. C. (1972) Mechanism of malonyl-coenzyme A-acyl- (2015) Enhanced production of branched-chain fatty acids by
carrier protein transacylase. Biochem. J. 128: 43-44. replacing β-ketoacyl-(acyl-carrier-protein) synthase III (FabH).
68. Heath, R. J. and C. O. Rock (1996) Regulation of fatty acid Biotechnol. Bioeng. 112: 1613-1622.
elongation and initiation by acyl-acyl carrier protein in 86. Heath, R. J. and C. O. Rock (1996) Roles of the FabA and FabZ
Escherichia coli. J. Biol. Chem. 271: 1833-1836. beta-hydroxyacyl-acyl carrier protein dehydratases in Escherichia
69. Davis, M. S., J. Solbiati, and J. E. Cronan (2000) Overproduction coli fatty acid biosynthesis. J. Biol. Chem. 271: 27795-27801.
of acetyl-CoA carboxylase activity increases the rate of fatty 87. Zeng, D., J. Zhao, H. S. Chung, Z. Guan, C. R. H. Raetz, and
acid biosynthesis in Escherichia coli. J. Biol. Chem. 275: P. Zhou (2013) Mutants resistant to LpxC inhibitors by
28593-28598. rebalancing cellular homeostasis. J. Biol. Chem. 288: 5475-
70. Lennen, R. M., D. J. Braden, R. M. West, J. A. Dumesic, and 5486.
B. F. Pfleger (2010) A process for microbial hydrocarbon 88. Silbert, D. F. and P. R. Vagelos (1967) Fatty acid mutant of
synthesis: overproduction of fatty acids in Escherichia coli and E. coli lacking a beta-hydroxydecanoyl thioester dehydrase.
catalytic conversion to alkanes. Biotechnol. Bioeng. 106: 193- Proc. Natl. Acad. Sci. USA 58: 1579-1586.
202. 89. Baba, T., T. Ara, M. Hasegawa, Y. Takai, Y. Okumura, M. Baba,
71. Lu, X., H. Vora, and C. Khosla (2008) Overproduction of free K. A. Datsenko, M. Tomita, B. L. Wanner, and H. Mori (2006)
fatty acids in E. coli: implications for biodiesel production. Construction of Escherichia coli K-12 in-frame, single-gene
Metab. Eng. 10: 333-339. knockout mutants: the Keio collection. Mol. Syst. Biol. 2:
72. Liu, D., Y. Xiao, B. S. Evans, and F. Zhang (2013) Negative 2006.0008. doi: 10.1038/msb4100050
feedback regulation of fatty acid production based on a 90. Cronan, J. E., D. F. Silbert, D. L. Wulff, and D. L. Wulff (1972)
malonyl-CoA sensor-actuator. ACS Synth. Biol. 4: 132-140. Mapping of the fabA locus for unsaturated fatty acid
73. Shin, K. S. and S. K. Lee (2017) Introduction of an acetyl-CoA biosynthesis in Escherichia coli. J. Bacteriol. 112: 206-211
carboxylation bypass into Escherichia coli for enhanced free 91. Cao, Y., J. Yang, M. Xian, X. Xu, and W. Liu (2010) Increasing
fatty acid production. Bioresour. Technol. 245: 1627-1633. unsaturated fatty acid contents in Escherichia coli by coex-
74. Zha, W., S. B. Rubin-Pitel, Z. Shao, and H. Zhao (2009) pression of three different genes. Appl. Microbiol. Biotechnol.
Improving cellular malonyl-CoA level in Escherichia coli via 87: 271-280.
metabolic engineering. Metab. Eng. 11: 192-198. 92. D’Agnolo, G., I. S. Rosenfeld, and P. R. Vagelos (1975)
75. Miyahisa, I., M. Kaneko, N. Funa, H. Kawasaki, H. Kojima, Multiple forms of beta-ketoacyl-acyl carrier protein synthetase
Y. Ohnishi, and S. Horinouchi (2005) Efficient production of in Escherichia coli. J. Biol. Chem. 250: 5289-5294.
(2S)-flavanones by Escherichia coli containing an artificial 93. Cronan, J. E., C. H. Birge, and P. R. Vagelos (1969) Evidence
biosynthetic gene cluster. Appl. Microbiol. Biotechnol. 68: 498- for two genes specifically involved in unsaturated fatty acid
16 Biotechnology and Bioprocess Engineering
biosynthesis in Escherichia coli. J. Bacteriol. 100: 601-604. lysophospholipase L1 are governed by its switch loop movement.
94. Lee, S., Y. Jung, S. Lee, and J. Lee (2013) Correlations between Biochemistry 44: 1971-1979.
FAS elongation cycle genes expression and fatty acid 110. Barnes, E. M. and S. J. Wakil (1968) Studies on the mechanism
production for improvement of long-chain fatty acids in of fatty acid synthesis. XIX. Preparation and general properties
Escherichia coli. Appl. Biochem. Biotechnol. 169: 1606-1619. of palmityl thioesterase. J. Biol. Chem. 243: 2955-2962.
95. Kim, S., C. H. Lee, S. W. Nam, and P. Kim (2011) Alteration of 111. Barnes, E. M. (1975) Long-chain fatty acyl thioesterases I and II
reducing powers in an isogenic phosphoglucose isomerase from Escherichia coli. Methods Enzymol. 35: 102-109.
(pgi)-disrupted Escherichia coli expressing NAD(P)-dependent 112. Cho, H. and J. E. Cronan (1993) Escherichia coli thioesterase I,
malic enzymes and NADP-dependent glyceraldehyde 3- molecular cloning and sequencing of the structural gene and
phosphate dehydrogenase. Lett. Appl. Microbiol. 52: 433-440. identification as a periplasmic enzyme. J. Biol. Chem. 268:
96. Steen, E. J., Y. Kang, G. Bokinsky, Z. Hu, A. Schirmer, A. 9238-9245.
McClure, S. B. Del Cardayre, and J. D. Keasling (2010) 113. Cho, H. and J. E. Cronan (1995) Defective export of a
Microbial production of fatty-acid-derived fuels and chemicals periplasmic enzyme disrupts regulation of fatty acid synthesis.
from plant biomass. Nature 463: 559-562. J. Biol. Chem. 270: 4216-4219.
97. Mazumdar, S., M. D. Blankschien, J. M. Clomburg, and 114. Shin, K. S., S. Kim, and S. K. Lee (2016) Improvement of free
R. Gonzalez (2013) Efficient synthesis of L-lactic acid from fatty acid production using a mutant acyl-CoA thioesterase I
glycerol by metabolically engineered Escherichia coli. Microb. with high specific activity in Escherichia coli. Biotechnol.
Cell Fact. 12: 7. doi: 10.1186/1475-2859-12-7 Biofuels 9: 208. doi: 10.1186/s13068-016-0622-y
98. Li, M., X. Zhang, A. Agrawal, and K. Y. San (2012) Effect of 115. Ohlrogge, J., L. Savage, J. Jaworski, T. Voelker, and D. Post-
acetate formation pathway and long chain fatty acid CoA-ligase Beittenmiller (1995) Alteration of acyl-acyl carrier protein
on the free fatty acid production in E. coli expressing acy-ACP pools and acetyl-CoA carboxylase expression in Escherichia
thioesterase from Ricinus communis. Metab. Eng. 14: 380-387. coli by a plant medium chain acyl-acyl carrier protein
99. Zhang, X., M. Li, A. Agrawal, and K.-Y. San (2011) Efficient thioesterase. Arch. Biochem. Biophys. 317: 185-190
free fatty acid production in Escherichia coli using plant acyl- 116. Lee, S., S. Park, C. Park, S. P. Pack, and J. Lee (2014) Enhanced
ACP thioesterases. Metab. Eng. 13: 713-722. free fatty acid production by codon-optimized Lactococcus
100. Clomburg, J. M., J. E. Vick, M. D. Blankschien, M. Rodríguez- lactis acyl-ACP thioesterase gene expression in Escherichia
Moyá, and R. Gonzalez (2012) A synthetic biology approach to coli using crude glycerol. Enzyme Microb. Technol. 67: 8-16.
engineer a functional reversal of the β-oxidation cycle. ACS 117. Lennen, R. M., M. A. Kruziki, K. Kumar, R. A. Zinkel,
Synth. Biol. 1: 541-554. K. E. Burnum, M. S. Lipton, S. W. Hoover, D. R. Ranatunga,
101. Wu, J., X. Zhang, X. Xia, and M. Dong (2017) A systematic T. M. Wittkopp, W. D. Marner, et al (2011) Membrane stresses
optimization of medium chain fatty acid biosynthesis via the induced by overproduction of free fatty acids in Escherichia
reverse beta-oxidation cycle in Escherichia coli. Metab. Eng. coli. Appl. Environ. Microbiol. 77: 8114-8128.
41: 115-124. 118. Lennen, R. M., M. G. Politz, M. A. Kruziki, and B. F. Pfleger
102. Ranganathan, S., T. W. Tee, A. Chowdhury, A. R. Zomorrodi, (2013) Identification of transport proteins involved in free fatty
J. M. Yoon, Y. Fu, J. V. Shanks, and C. D. Maranas (2012) An acid efflux in Escherichia coli. J. Bacteriol. 195: 135-144.
integrated computational and experimental study for over- 119. Oh, H. Y., J. O. Lee, and O. B. Kim (2012) Increase of organic
producing fatty acids in Escherichia coli. Metab. Eng. 14: 687- solvent tolerance of Escherichia coli by the deletion of two
704. regulator genes, fadR and marR. Appl. Microbiol. Biotechnol.
103. Soma, Y., K. Tsuruno, M. Wada, A. Yokota, and T. Hanai 96: 1619-1627.
(2014) Metabolic flux redirection from a central metabolic 120. Dellomonaco, C., J. M. Clomburg, E. N. Miller, and R. Gonzalez
pathway toward a synthetic pathway using a metabolic toggle (2011) Engineered reversal of the β-oxidation cycle for the
switch. Metab. Eng. 23: 175-184. synthesis of fuels and chemicals. Nature 476: 355-359.
104. Liu, H., C. Yu, D. Feng, T. Cheng, X. Meng, W. Liu, H. Zou, 121. Leonard, E., P. K. Ajikumar, K. Thayer, W. H. Xiao, J. D. Mo,
and M. Xian (2012) Production of extracellular fatty acid using B. Tidor, G. Stephanopoulos, and K. L. J. Prather (2010)
engineered Escherichia coli. Microb. Cell Fact. 11: 41. doi: Combining metabolic and protein engineering of a terpenoid
10.1186/1475-2859-11-41 biosynthetic pathway for overproduction and selectivity control.
105. Huang, Y. T., Y. C. Liaw, V. Y. Gorbatyuk, and T. H. Huang Proc. Natl. Acad. Sci. 107: 13654-13659.
(2001) Backbone dynamics of Escherichia coli thioesterase/ 122. Xu, P., Q. Gu, W. Wang, L. Wong, A. G. W. Bower, C. H. Collins,
protease I: evidence of a flexible active-site environment for a and M. A. G. Koffas (2013) Modular optimization of multi-gene
serine protease1. J. Mol. Biol. 307: 1075-1090. pathways for fatty acids production in E. coli. Nat. Commun. 4:
106. Lo, Y. C., S. C. Lin, J. F. Shaw, and Y. C. Liaw (2003) Crystal 1409. doi: 10.1038/ncomms2425
structure of Escherichia coli thioesterase I/protease I/ 123. Subrahmanyam, S. and J. E. Cronan (1998) Overproduction of a
lysophospholipase L1: consensus sequence blocks constitute the functional fatty acid biosynthetic enzyme blocks fatty acid
catalytic center of SGNH-hydrolases through a conserved synthesis in Escherichia coli. J. Bacteriol. 180: 4596-4602.
hydrogen bond network. J. Mol. Biol. 330: 539-551. 124. Keating, D. H., M. R. Carey, and J. E. Cronan (1995) The
107. Lee, L. C., Y. L. Chou, H. H. Chen, Y. L. Lee, and J. F. Shaw unmodified (apo) form of Escherichia coli acyl carrier protein is
(2009) Functional role of a non-active site residue Trp23 on the a potent inhibitor of cell growth. J. Biol. Chem. 270: 22229-
enzyme activity of Escherichia coli thioesterase I/protease 22235.
I/lysophospholipase L1. Biochim. Biophys. Acta - Proteins 125. Yu, X., T. Liu, F. Zhu, and C. Khosla (2011) In vitro
Proteomics 1794: 1467-1473. reconstitution and steady-state analysis of the fatty acid
108. Lee, L. C., Y. L. Lee, R. J. Leu, and J. F. Shaw (2006) synthase from Escherichia coli. Proc. Natl. Acad. Sci. USA 108:
Functional role of catalytic triad and oxyanion hole-forming 18643-18648.
residues on enzyme activity of Escherichia coli thioesterase 126. Tao, H., Y. Zhang, X. Cao, Z. Deng, and T. Liu (2016) Absolute
I/protease I/phospholipase L1. Biochem. J. 397: 69-76. quantification of proteins in the fatty acid biosynthetic pathway
109. Lo, Y. C., S. C. Lin, J. F. Shaw, and Y. C. Liaw (2005) Substrate using protein standard absolute quantification. Synth. Syst.
specificities of Escherichia coli thioesterase I/protease I/ Biotechnol. 1: 150-157.
Microbial Production of Fatty Acid via Metabolic Engineering and Synthetic Biology 17
J. V. Shanks, J. B. Klauda, and L. R. Jarboe (2017) Engineering engineering of fatty acid biosynthesis in E. coli. Metab. Eng. 12:
Escherichia coli membrane phospholipid head distribution 378-386.
improves tolerance and production of biorenewables. Metab. 164. Zheng, Y., L. Li, Q. Liu, J. Yang, Y. Cao, X. Jiang, G. Zhao,
Eng. 44: 1-12. M. Xian, and M. Xian (2012) Boosting the free fatty acid
159. Youngquist, J. T., M. H. Schumacher, J. P. Rose, T. C. Raines, synthesis of Escherichia coli by expression of a cytosolic
M. C. Politz, M. F. Copeland, and B. F. Pfleger (2013) Production Acinetobacter baylyi thioesterase. Biotechnol. Biofuels 5: 76.
of medium chain length fatty alcohols from glucose in doi: 10.1186/1754-6834-5-76
Escherichia coli. Metab. Eng. 20: 177-186. 165. Lee, S., E. Jeon, Y. Jung, and J. Lee (2012) Heterologous co-
160. Bremer, H. and P. P. Dennis (1996). Modulation of chemical expression of accA, fabD, and thioesterase genes for improving
composition and other parameters of the cell by growth rate. pp. long-chain fatty acid production in Pseudomonas aeruginosa
1553-1569. In: Neidhardt, F. C., R. Curtiss III, J. L. Ingraham, and Escherichia coli. Appl. Biochem. Biotechnol. 167: 24-38.
E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. 166. Zhang, F., J. M. Carothers, and J. D. Keasling (2012) Design of
Riley, M. Schaechter, and H. E. Umbarger (eds.), Escherichia a dynamic sensor-regulator system for production of chemicals
coli and Salmonella typhimurium: cellular and molecular and fuels derived from fatty acids. Nat. Biotechnol. 30: 354-359.
biology. American Society for Microbiology Press, Washington, 167. Kim, S., J. M. Clomburg, and R. Gonzalez (2015) Synthesis
D.C. of medium-chain length (C6-C10) fuels and chemicals via
161. Chubukov, V. and U. Sauer (2014) Environmental dependence of β-oxidation reversal in Escherichia coli. J. Ind. Microbiol.
stationary-phase metabolism in Bacillus subtilis and Escherichia Biotechnol. 42: 465-475.
coli. Appl. Environ. Microbiol. 80: 2901-2909. 168. Tan, Z., J. M. Yoon, A. Chowdhury, K. Burdick, L. R. Jarboe,
162. Peralta-Yahya, P. P., F. Zhang, S. B. del Cardayre, and C. D. Maranas, and J. V. Shanks (2018) Engineering of E. coli
J. D. Keasling (2012) Microbial engineering for the production inherent fatty acid biosynthesis capacity to increase octanoic
of advanced biofuels. Nature 488: 320-328. acid production. Biotechnol. Biofuels 11: 87. doi: 10.1186/
163. Liu, T., H. Vora, and C. Khosla (2010) Quantitative analysis and s13068-018-1078-z