Engineering E. Coli For Environmentally-Friendly Petrol Production From Lignocellulosic Feedstock

You are a process engineer working in a major multinational company. There is within the company a desire to make petrol as a product in the most environmentally-friendly way possible, using a biological process. In addition, part of your company generates lignocellulosic hydrolysate in large quantities as a by-product of a process. You need to use this as a feedstock for your process.

Your company has a background in biological processes using the bacterium Escherichia coli. Your process must use this.

In addition, the organism that you engineer must be used in an immobilized form to retain biomass in the bioreactor.

Your company has a large bioprocess / fermentation development division, so you do not need to discuss this. You need to focus on how you will engineer the organism (E. coli) to make your product from the stated feedstock.

Metabolic Engineering

From biotechnology viewpoint, fatty acids are energy sufficient and are thus integrated into intracellular lipid. Lately, fatty acid metabolism has got a considerable focus as a pathway for generating high-density transference gasses.  Lignocellulosic constituents comprise mostly of three polymers: hemicellulose, cellulose and lignin (Kim, Ximenes, Mosier and Ladisch 2011). The Lignocellulosic feedstock requires forceful pretreatment to produce a substrate simply hydrolysable by viable cellulolytic enzymes or by enzyme generating microbes, to release sugar required for fermentation (Rumbold et al. 2010).  Cellulose is a key constituent of the biomass which is a polymer of β-D-glucopyranose bits connected through β-(1, 4) glycosidic bonds with known polymorphs.  Hemicellulose is the second plentiful polymer comprising approximately 20-50% of the lignocellulose (Kim et al. 2011). The hemicellulose has sort sideways chains comprising of various types of sugars. These monosaccharides comprise the pentose, hexoses, and uronic acids.  Lignin is the third abundant polymers. It is a sophisticated, huge molecular assembly cross-connected polymer of phenolic monomers (Kim, Block and Mills 2010). 

Successful use of microorganism for biomass catalysis relies on the creature’s capability to create biofuels in industrialized scale at low cost. Numerous microbes have intrinsic biochemical paths that change biomass into yields that match required biofuels (Xu et al. 2013). But, commercial-scale overproduction of biofuels from the above-secluded microbes frequently require genetic variation and genetic factor import to fine-tune the multiphase biological routine leading to biofuels. E.coli has distinctive benefits of a well-researched model creature in regard to genetic material expression and regulation, and with a broadest molecular approach accessible for genomic engineering (Mazumdar, Bang and Oh 2014). E .coli strains can certainly use a range of carbon bases such as sugars and sugar alcohols under aerobic and anaerobic settings and are best fit for numerous industrial produces (Kim, Block and Mills 2010).

Current signs of progress in metabolic engineering, synthetic and systems biology have played a key part in creating attention in the commercial manufacture of biofuels from microbes counting E.coli (Bokinsky et al. 2011). The above advances facilitated progress in natural paths, to build novel biosynthetic ways for the optimum generation of the preferred biofuel yields. Similarly, the growth of novel sequencing expertise supported the recognition of the hereditary changes, comprehension range, and description of the genetic character of the organism, which could take part in creating new groups of biofuels (Mazzoli, Lamberti and Pessione 2012).

Omega-3 fatty acids are polyunsaturated fatty acids which have binary ends, methyl end which is reflected as tail of the sequence and carboxylic acid terminal which is deliberated at the start of the sequence.  Three classes of omega-3-fatty acids intricate in human makeup are eicosapentaenoic acid, alpha-linolenic acid and docosahexaenoic acid.  An omega 3-fatty acid has multiple double bonds, where the main double link is between the 3^rd and 4^th carbon bits from the terminal of the carbon atom sequence (Zhang, Carothers and Keasling 2012). Short series has a set of 18 carbon atoms or less while long chain has a sequence of 20 carbon atoms or more.  The above three polyunsaturates have 3, 5 or 6 double links in a carbon sequence of 18, 20, or 22 carbon atoms. As with most-generated fatty acids, all double links are in the cis-conformation (Zhang et al. 2012).

Fatty Acid Biosynthesis in E. coli

Fatty acid biosynthesis and deprivation are well known in E. coli where ages of investigation have been appropriately outlined in the review (Mazumdar, Bang and Oh 2014).  Concisely, fatty acids are generated through an iterative decrease series that functions on acyl transporter protein (ACP) thioesters. In all iteration, binary carbon atoms are added from malonyl-ACP to a developing acyl-sequence and the subsequent keto set is lessened to a saturated methylene (Zhang, Agrawal and San 2012). The routine lasts until elongated-chain acyl-ACPs are combined onto phospholipids by acyltransferase or changed to other metabolites (Wahl, My, Dumoulin, Sturgis and Bouveret 2011). E. coli can also pursue FFAs and utilise them as a single carbon basis by creating acetyl-CoA through the β-oxidation pathways. This series function alike to FAB, but in an opposite, eradicating single acetyl-CoA per cycle. On the other hand, β-oxidation functions on acyl-CoA thioesters intermediates rather than acyl-ACP thioesters. The initial phase of β-oxidation is FFAs activation to acyl-CoA by acyl-CoA synthetase, for instance, FadD (Feng and Cronan 2009). Also, Acyl-CoA is an initial stage for the formation of numerous attractive oleochemicals products such as ketones, fatty alcohols, bioplastics, alkanes and olefins.

When heterologous articulated, numerous additional acyl-ACP thioesterase creates similar effects.  In E. coli, the primary controlling indication for regulating FAB was extended sequence acyl-ACP. The first link came from a reflection that cultured starved of glycerol showed a reduced degree of acyl-ACP production (Wahl et al. 2011). Concurrently, flux over FAB was established to be augmented by cytosolic overexpression of E-coli thioesterase 1, which hydrolyzes acyl-ACPs and CoA to form FFAs and subsequently reduces elongated chain acyl-AC (My et al. 2013).  In vitro researches later established that acyl-ACP directly constrain acetyl-CoA carboxylase and to a slighter extent b-ketoacyl-ACP synthase III (Fabh) and enoyl-ACP reductase (FabI) (My et al. 2013).

Pinpointing the rate-imitating phase in FAB would be worth to manufacturing strains for FFA assembly (Laluce, Schenberg, Gallardo, Coradello and Pombeiro-Sponchiado 2012). Unluckily, little kinetic factors have been established as an effect to the masses of acyl-ACP intermediates and the hardship in measuring the protein-bound products and substrates. Moreover, the complete range of acyl-ACP-mediated restriction of all enzymes in FAB has yet to be established (Nawabi, Bauer, Kyrpides and Lykidis 2011).

The translational and transcriptional parameter of FAB is not fully comprehended (My et al. 2013).  Binary transcriptional controllers, FadR and FabR, are intricate in regulating unsaturated B-oxidation and FAB; however, a transcriptional aspect has not been linked with regulating genes expression coding all FAB enzymes (Feng and Cronan 2012). This is in disparity where other microbes such as Streptococcus pneumonia and Bacillus subtilis where transcription elements have more complete regulator over the FAB genes appearance.

All the biofuels resultants from E. coli are derivatives from the central carbon catabolism variation and the routine comprises the transformation of pentose or hexose sugar molecules to C2 for the microbial production of biofuels (Kim, Block and Mills 2010).

In E.coli, the genetic factor coding ACC are positioned in three distal operons (accD, accBC, accA) (Sabri, Nielsen and Vickers 2012). Planes of all fours ACC subdivision transcriptions have been established to compare with development proportion, identical to the general speed of fab (Bokinsky et al. 2011).  It is recognised that AccB auto-controls transcript of accBC, most probably by DNA attachment in its developer zone. Moreover, accD and accA translation is auto-controlled by linking of RNA to the AccAD compound, thus offering response regulation of translation (My et al. 2013). But, past these networks, the control of ACC is ill understood. Assumed the primary function of malonyl-CoA creation, comprehending the control of ACC, and in what way it coordinates with the FAB remnants, safeguard further examination (Zhang, Agrawal and San 2012).

Rate-Limiting Factor in Fatty Acid Biosynthesis

FEEDBACK ON METABOLIC ENGINEERING : I have read your body of work and I can’t seem to follow through to arrive at the answer to any of the questions asked in the guidelines. If you are sure the answers are there. Please kindly highlight the part where the answers to the questions are in the following colours :

A favourable addition to the overexpression of genetic factor that is intricate in fatty acid biosynthesis provides the regulatory mutants application (Wang et al. 2013). A probable target to advance the FFA products is the carbon-storage controller, which comprises of the CsrA protein and the non-coding RNAs CsrB and CsrC (Ogasawara, Shinohara, Yamamoto and Ishihama 2012).  A more apparent contender to change the controlling system is the fatty acid deprivation repressor, FadR. In an E. coli strain with a fadE removal and with tesA overexpression, coexpression of FadR occasioned in a more than 7-fold improved FFA creation. As a result to fabA and fabB, the FadR coexpression causes to an upsurge of the unsaturated fatty acid content ranging from 13 to 43% in the strain creation (Feng and Cronan 2009).

Even though the fed-batch fermentation has benefits over batch cultures, steady fermentation provides an even greater prospect, since the cells can be reserved under optimum situations and in the most appropriate development stage (Laluce et al. 2012). With the purpose of FFA generation, steady E.coli cultivations with a substitution of fadAB, fadD and fadE, each by one replica of the thioesterase genetic material from Umbellularia californica have been done (Laluce et al. 2012).

The applications of E. coli for the generation of FFA was established by the discovery that TesA deregulates the firm invention inhibit of fatty acid production when stated as a cytosolic enzyme. As fatty acid is extremely energy-compact, generated in comparatively huge quantity and in all microbes, they denote an appropriate aim for the growth of lone-cells oils. Additionally, the application of substitute carbon bases has been illustrated in many organisms. With a growing focus towards the study of the renewable energy sources, numerous researches have been done in the past one decade and a half with the motive of using fatty acid biosynthesis for biofuel generation.  But due to the stern control of this path, the much fundamental study is required to advance the products of FFA (Xu et al. 2013).

To begin, by application of various thioesterase, the products can be significantly changed with respect to output, a degree of saturation and fatty acid chain length.  But, the expression levels of every thioesterase ought to be fine-tuned appropriately, as already low intensities upsurge of fatty acid titer considerably, and too robust a thioesterase action has been illustrated to damage FFA product, in vivo and in vitro trials.  Additionally, a great titer of FFA culture medium can also result in dire faults in the cellular feasibility.   With regard to the physiological defects of FFA overproduction, it has been illustrated that overexpression of thioesterase can change the saturation extent of the membrane lipids of E. coli.  Of the significance for the microscopic generation of biofuels are approaches to improve the E. coli resistance to carbon-based solvents. The FadR deletions ensued an improved portion of saturated fatty acids in the E. coli sheath, has been noted in the earlier investigation (Feng and Cronan 2009).  The genetic factor products generate an efflux structure for biological liquids and therefore progress the E. coli existence in existence of a high concentration of biological diluents.  By the removal of marR, the multidrug resistance of E .coli as permanently induced. A blend of marR and FadR deletion causes an even higher organic solvent tolerance, matched to the lone deletions. But, it is not ideal for the FadR deletion, if one focus on the creation of fatty acids (Feng and Cronan 2009). Thus, a linkage of marR omission and the advanced saturated fatty acids synthesis looks to be promising.

Transcriptional and Translational Regulation of FAB

To avoid product dilapidation, numerous investigations have been done in a strain that was repressed in fatty acid b-oxidation (Laluce et al. 2012). Even though many researchers established FFA intensities improved upon limited obliteration of the B-oxidation pathways or did not regulate the victory of this removal. Liu et al. (2012) did not notice an affirmative impact when thioesterase overexpression was blended with the obliteration of fadL, fadD and fade.  In the above studies, it was proposed that the B-oxidation pathways have no capability to cope with the robust FFA production (Xu et al. 2013).

To advance FFA manufacture on a comprehensive scale, computational replicas of the E. coli metabolism has been applied, and numerous removals in the glycolysis or Tricarboxylic acid series has been examined alongside with genetic factor of overexpression of fatty acid biosynthesis. Gene’s obliteration accountable for acetate creation has been tried to advance malonyl-CoA titers or FFA output. This approach obviously minimised acetate development.  But, in the two latter researchers of acetate generation reduction did not improve the FFA products. In its place, Zhang et al (2012) state that acetate creation is already reduced in active FFA makers. This also fascinating with regard to the medium pH, as  E. coli generation strains incline to slight grow the pH, rather than reducing it as a wild-type cell.  A comprehensive research in what way to advance the malonyl-CoA strengths in the cytosol has been done by Zhang, Agrawal and San (2012).

Consolidate bioprocessing (CBP) can be described as a single-step procedure in which feedstock is openly changed into a preferred produce by a particular microscopic group without the need of feedstock pre-treatment.  The word can be used to any raw material and any produce but is normally linked with Lignocellulosic biomass. The hardest task with CBP is designing of a correct microbial grouping that must express suitable hydrolytic enzyme matching the Lignocellulosic feedstock (Rumbold et al. 2010). Firmly, the raw material for CBP ought not to need any distinctive physical; chemical or enzyme pre-treatment and unit size decrease should be adequate.

Outmoded enzyme immobilization is centred on a connection in or on solid fragments or enzyme cross-linking.  To assemble an immobilization scheme, the complexity and nature of raw lignocellulosic solid ought to be reflected, and even though pretreated, it is still a suspension of insoluble material that excludes the application of orthodox separation procedures such as centrifugation and filtration (Kim et al. 2011).  The key cell immobilization techniques includes entrapment in a polymer medium, adsorption onto a solid carrier, covalent connecting to a solid support, affinity interactions and cross-linking of cell aggregation.    Cross-linking of cell aggregates, using a bifunctional reagent, is utilized to prepare carrier less macro units. The usage of a carrier certainly causes a dilution of action, owing to the introduction of a huge share of non-catalytic ballast, ranging from 90 to >99, which results in lower space-period output. This is not eased by utilisation of higher cell holdings as this leads to loss of activity due to challenges of availability of some of the cell particles when they comprises of many layers on the surface of the carriers. The optimal situation, from a particular action viewpoint, is a monolayer of cells molecules adsorbed on the carrier surface. Subsequently there is an upsurge interest in carrier-free immobilized cells, such as cross-linked cells crystals.  This technique provides an apparent advantages; high stability and concentrated cell activity in the catalyst and low production rates owing to the exclusion of an extra carrier (Laluce et al. 2012).  .

Biofuels Production in E. coli

Immobilization can cause beneficial variations in the physiognomies of an enzyme, for instance, augmented stability is frequently stated together with variations in pH optimum and thermal characteristics and as well as selectivity changes.

Unluckily, in general scenarios, a decrease in definite enzyme action happens after immobilization to an insoluble transporter. A decline in an action of an immobilized enzyme could be instigated by folding of cross-linked assembly, diffusional concerns of the big substrate, changing in enzyme conformation together with variations in the catalytic domain accessibility.  The gain of the immobilized enzymes is their capability to be reused and recovered, however, from the testified info; it is obvious that enzyme action of free and immobilized biocatalyst more or less reduces with a growing interval of saccharification, temperature and the cycle numbers.

Apparently, cell mobilization offers more motivating results than celluloses immobilization; while cell immobilization causes mostly to advancement in fatty acid generation over lethal complexes resistance in lignocellulosic hydrolyzates, the reverse, a reduction in enzymatic action, happens after cellulose immobilization, together with concerns in split-up from the lignocellulosic substances (Kim et al. 2011). In both scenarios, the enzymes and cells, the proposed configuration profitability ought to be evaluated not merely on the product yield but also energy and time consumptions linked with immobilization, separation and handling as well as extra amenities and transporter expenses (My et al. 2013).

The initial phase in making the conversion from concentration to the number of ligand molecules per bead is to examine the number of beads per milliter of packed resin. The approach is to calculate the average bead volume and then apply sphere-packing theory as the basis for estimating the number of those beads in one milliter of resin (Conway and Sloane 2013).

For instance, take the agarose beads with diameter of 45 and 165 um

Therefore, the volume of single bead at diameter of 45um

The radium=0.0225

Volume of sphere=4/3* π *r³ (Valera, Morales, Vanmaercke, Morfa, Cortés and Casañas 2015).

  4/3* π *0.0225³

For radium 0.00225 cm, the single bead volume =4.77*10^-7mL

 For really tiny pipette, the volume is 0.477nL!

For the  diameter 165um

Radium 82.5=0.0825

Volume of sphere=4/3*π*r³

4/3* π *0.0825³

=2.352*10^-3 ml

=2352.07nL!

For the complete bead populace, thus, the individual bead volume range from 0.477nL! (Diameter of 45) to 2352.07nL! (165 um)

Applying the sphere-packing theory which indicates that spheres packs with 65-74% efficiency (Conway and Sloane 2013). Therefore, for instance, beaded agarose is a bead sizes and t they are neither hard nor perfectly spherical. Therefore, the sphere-packing theory might offer only an approximate estimate of the true value for the particle numbers in a specific volume (Conway and Sloane 2013).

Assuming 75% packing efficiency, 0.75 mL of actual bead volume will settle to generate nearly 1mL of bed volume (Conway and Sloane 2013).

Final calculation

Number of beads per mL in 45 um diameter

=0.75(1bead/4.77*10^-7mL)

=1.572*10⁶ beads/mL of bed

=approximately 1.6 million beads per milliliter of resin

Number of beads per mL in 165 um diameter

=0.75(1bead/2.352*10^-3 ml)

=318.876 beads/mL of bed

=approximately 319 beads per milliliter of resin

Therefore, the small diameter of the carrier will offer a greater level of immobilized cells than large diameter since the porosity will be low.

References

Bokinsky, G., Peralta-Yahya, P.P., George, A., Holmes, B.M., Steen, E.J., Dietrich, J., Lee, T.S., Tullman-Ercek, D., Voigt, C.A., Simmons, B.A. and Keasling, J.D., 2011. Synthesis of three advanced biofuels from ionic liquid-pretreated switchgrass using engineered Escherichia coli. Proceedings of the National Academy of Sciences, 108(50), pp.19949-19954. [Online]. Retrieved at: https://www.pnas.org/content/108/50/19949.short , [Accessed on 11 December 2018].

Conway, J.H. and Sloane, N.J.A., 2013. Sphere packings, lattices and groups (Vol. 290). Springer Science & Business Media. [Online]. Retrieved at: https://books.google.com/books?hl=en&lr=&id=hoTjBwAAQBAJ&oi=fnd&pg=PR5&dq=the+sphere-packing+theory+&ots=_zhHhPh_wL&sig=6GOtwUHfHJL2xEPgMl57OfPMQbg, [Accessed on 11 December 2018].

Feng, Y. and Cronan, J.E., 2009. A new member of the Escherichia coli fad regulon: transcriptional regulation of fadM (ybaW). Journal of bacteriology, 191(20), pp.6320-6328. [Online]. Retrieved at: https://jb.asm.org/content/191/20/6320.short, [Accessed on 11 December 2018].

Feng, Y. and Cronan, J.E., 2012. Crosstalk of Escherichia coli FadR with global regulators in expression of fatty acid transport genes. PloS one, 7(9), p.e46275. [Online]. Retrieved at: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0046275, [Accessed on 11 December 2018].

Kim, J.H., Block, D.E. and Mills, D.A., 2010. Simultaneous consumption of pentose and hexose sugars: an optimal microbial phenotype for efficient fermentation of lignocellulosic biomass. Applied microbiology and biotechnology, 88(5), pp.1077-1085. [Online]. Retrieved at: https://link.springer.com/article/10.1007/s00253-010-2839-1, [Accessed on 11 December 2018].

Kim, Y., Ximenes, E., Mosier, N.S. and Ladisch, M.R., 2011. Soluble inhibitors/deactivators of cellulase enzymes from lignocellulosic biomass. Enzyme and Microbial Technology, 48(4-5), pp.408-415. [Online]. Retrieved at:  https://www.sciencedirect.com/science/article/pii/S0141022911000251, [Accessed on 11 December 2018].

Laluce, C., Schenberg, A.C.G., Gallardo, J.C.M., Coradello, L.F.C. and Pombeiro-Sponchiado, S.R., 2012. Advances and developments in strategies to improve strains of Saccharomyces cerevisiae and processes to obtain the lignocellulosic ethanol− a review. Applied biochemistry and biotechnology, 166(8), pp.1908-1926. [Online]. Retrieved at: https://link.springer.com/article/10.1007/s12010-012-9619-6, [Accessed on 11 December 2018].

Liu, H., Yu, C., Feng, D., Cheng, T., Meng, X., Liu, W., Zou, H. and Xian, M., 2012. Production of extracellular fatty acid using engineered Escherichia coli. Microbial cell factories, 11(1), p.41. [Online]. Retrieved at: https://microbialcellfactories.biomedcentral.com/articles/10.1186/1475-2859-11-41, [Accessed on 11 December 2018].

Mazumdar, S., Bang, J. and Oh, M.K., 2014. L-Lactate production from seaweed hydrolysate of Laminaria japonica using metabolically engineered Escherichia coli. Applied biochemistry and biotechnology, 172(4), pp.1938-1952. [Online]. Retrieved at: https://link.springer.com/article/10.1007/s12010-013-0653-9, [Accessed on 11 December 2018].

Mazzoli, R., Lamberti, C. and Pessione, E., 2012. Engineering new metabolic capabilities in bacteria: lessons from recombinant cellulolytic strategies. Trends in biotechnology, 30(2), pp.111-119. [Online]. Retrieved at: https://www.sciencedirect.com/science/article/pii/S0167779911001521, [Accessed on 11 December 2018].

My, L., Rekoske, B., Lemke, J.J., Viala, J.P., Gourse, R.L. and Bouveret, E., 2013. Transcription of the Escherichia coli fatty acid synthesis operon fabHDG is directly activated by FadR and inhibited by ppGpp. Journal of bacteriology, pp.JB-00384. [Online]. Retrieved at: https://jb.asm.org/content/early/2013/06/10/JB.00384-13.short, [Accessed on 11 December 2018].

Nawabi, P., Bauer, S., Kyrpides, N. and Lykidis, A., 2011. Engineering E. coli for biodiesel production utilizing a bacterial fatty acid methyltransferase. Applied and environmental microbiology, pp.AEM-05046. [Online]. Retrieved at: https://aem.asm.org/content/early/2011/09/16/AEM.05046-11.short, [Accessed on 11 December 2018].

Ogasawara, H., Shinohara, S., Yamamoto, K. and Ishihama, A., 2012. Novel regulation targets of the metal-response BasS–BasR two-component system of Escherichia coli. Microbiology, 158(6), pp.1482-1492. [Online]. Retrieved at: https://mic.microbiologyresearch.org/content/journal/micro/10.1099/mic.0.057745-0, [Accessed on 11 December 2018].

Rumbold, K., van Buijsen, H.J., Gray, V.M., van Groenestijn, J.W., Overkamp, K.M., Slomp, R.S., van der Werf, M.J. and Punt, P.J., 2010. Microbial renewable feedstock utilization: a substrate-oriented approach. Bioengineered bugs, 1(5), pp.359-366. [Online]. Retrieved at: https://www.tandfonline.com/doi/abs/10.4161/bbug.1.5.12389, [Accessed on 11 December 2018].

Sabri, S., Nielsen, L.K. and Vickers, C.E., 2012. Molecular control of sucrose utilization in Escherichia coli W, an efficient sucrose-utilizing strain. Applied and environmental microbiology, pp.AEM-02544. [Online]. Retrieved at: https://aem.asm.org/content/early/2012/10/30/AEM.02544-12.short, [Accessed on 11 December 2018].

Valera, R.R., Morales, I.P., Vanmaercke, S., Morfa, C.R., Cortés, L.A. and Casañas, H.D.G., 2015. Modified algorithm for generating high volume fraction sphere packings. Computational Particle Mechanics, 2(2), pp.161-172. [Online]. Retrieved at:  https://link.springer.com/article/10.1007/s40571-015-0045-8, [Accessed on 11 December 2018].

Wahl, A., My, L., Dumoulin, R., Sturgis, J.N. and Bouveret, E., 2011. Antagonistic regulation of dgkA and plsB genes of phospholipid synthesis by multiple stress responses in Escherichia coli. Molecular microbiology, 80(5), pp.1260-1275. [Online]. Retrieved at:  https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-2958.2011.07641.x, [Accessed on 11 December 2018].

Wang, X., Yomano, L.P., Lee, J.Y., York, S.W., Zheng, H., Mullinnix, M.T., Shanmugam, K.T. and Ingram, L.O., 2013. Engineering furfural tolerance in Escherichia coli improves the fermentation of lignocellulosic sugars into renewable chemicals. Proceedings of the National Academy of Sciences, p.201217958. [Online]. Retrieved at: https://www.pnas.org/content/early/2013/02/15/1217958110.short, [Accessed on 11 December 2018].

Xu, P., Gu, Q., Wang, W., Wong, L., Bower, A.G., Collins, C.H. and Koffas, M.A., 2013. Modular optimization of multi-gene pathways for fatty acids production in E. coli. Nature communications, 4, p.1409. [Online]. Retrieved at:  https://www.nature.com/articles/ncomms2425, [Accessed on 11 December 2018].

Zhang, F., Carothers, J.M. and Keasling, J.D., 2012. Design of a dynamic sensor-regulator system for production of chemicals and fuels derived from fatty acids. Nature biotechnology, 30(4), p.354. [Online]. Retrieved at: https://www.nature.com/nbt/journal/v30/n4/abs/nbt.2149.html, [Accessed on 11 December 2018].

Zhang, X., Agrawal, A. and San, K.Y., 2012. Improving fatty acid production in escherichia coli through the overexpression of malonyl coA?Acyl carrier protein transacylase. Biotechnology progress, 28(1), pp.60-65. [Online]. Retrieved at: https://onlinelibrary.wiley.com/doi/abs/10.1002/btpr.716, [Accessed on 11 December 2018].