Synthesis Of Polyhydroxyalkanoates Containing Medium-chain-length Monomers Via The Reversed Fatty Acid β-oxidation

Because of having many excellent properties as biological materials, polyhydroxyalkanoates (PHA) have been used in many areas such as medicine. It is a kind of carbon and energy storage reserves in the presence of the excess carbon source and other large number of elements (N, O, P, S) depleted, can be synthesized by a large class of different types of microorganisms. PHA is composed of monomers containing different carbon chain-length 3-hydroxyalkanoate (3 HA).Pursuing the current trend, the "green-polymers", polyhydroxyalkanoates (PHAs) which are degradable and made from renewable sources have been a potential substitute for synthetic plastics. Due to the increasing concern towards escalating crude oil price, depleting petroleum resource and environmental damages done by plastics, PHAs have gained more and more attractions, both from industry and research. So far, there are more than 150 kinds of monomers that have been reported. Escherichia coli as a model organism, has been widely used to produce a variety of high value-added products for human beings, such as biofuels, biomaterials, bulk chemicals. Although under natural conditions, E. coli can not synthesize PHA, they are still considered as an ideal host for PHA biosynthesis. In general, structurally related carbon source for the synthesis of PHA, i.e., fatty acids, are expensive, toxic to cells, and insoluble in water. Therefore, many researchers are beginning to integrate different microbial genes and to discover the novel metabolic pathways for PHA synthesis from unrelated carbon sources.According to the different monomer composition, PHAs can be divided into three main types:short-chain-length PHAs (scl-PHAs) which contain 3-5 carbon atoms, medium-chain-length PHAs (mcl-PHAs) which contain 6-14 carbon atoms, and scl-mcl PHAs which contain 3-14 carbons in length.Due to its properties, PHAs have high potential in medical field’s applications, such as orthopaedy (screws, scaffolds for cartilage engineering, bone graft substitutes), for cardiovascular system devices (blood vessel replacements, heart valves, cardiovascular stents, atrial and pericardial septal repair patches), wound management (skin substitutes, sutures, dressings, dusting powders), urological stents, nano-and microspheres for controlled drug delivery and others.Since PHA has so many extensive and important applications, we have to build a new platform for efficient synthesis of PHA and make it utilize cheap carbon sources to yield PHA with various chain-length and different type of monomers in order to satisfy different applications.Traditional scl-PHA production such as PHB undergoing a three-step method, gained the very high yield and has been used for large-scale industrial production. The mcl-PHA production was primarily by fatty acid de novo synthesis and fatty acid 0-oxidation cycle. Both pathways have their own synthetic defects. First, for the fatty acid P-oxidation pathway, it utilizes fatty acid as a carbon source to synthesize PHA. Fatty acids are relatively expensive carbon sources and have toxic effects on the microbial cells, so the use of the traditional PHA synthetic pathway has a lot of restrictions. Although the fatty acid de novo synthesis uses carbohydrates as carbon sources for PHA production, the intracellular accumulation of PHA was still low. For these reasons, the urgent need was to build an efficient platform for PHA production from cheap and unrelated carbon sources, thereby providing sufficient (R)-3-hydroxyacyl-CoA precursors.For solving the above problems, enzymes in a reversed fatty acid β-oxidation cycle were overexpressed and we transformed the plasmids into the recombinant E. coli, and then tried to directly use glucose as a carbon source to synthesize mcl-PHA copolymers with different monomers. When adding 30 g l-1 glucose into fermentation medium, all three thioesterase gene-knochout strains could synthesize mcl-PHA copolymers, the highest content of mcl-PHA synthesis was in the tesB gene knockout strain which accounted for about 4.01% of dry cell weight, followed by yciA gene knockout strain, about 2.04% of the dry weight of the cells, the tesA gene knockout strain synthesized the lowest mcl-PHA. At the same time, we try to make use of other low-cost, nonrelated carbon sources such as xylose for the fermentation tests. The results show that when 30 g l-1 xylose as sole carbon source added into the medium, three single thioesterase gene knockout strains harboring the plasmid pQQ05 can synthesize mcl-PHA, the highest yield was 2.33 wt%, but lower than that of the corresponding strain with glucose as carbon source. Furthermore, on the basis of the original strain, we knocked out the ptsG gene encoding the key enzyme of E. coli glucose-specific PTS system component-enzyme IIBC and released the carbon metabolite repression phenomenon of E. coli in order to use glucose and xylose simutaneously. Therefore, we added 15 g l-1 glucose and 15 g l-1 xylose as carbon sources to the medium, three single thioesterase gene knockout strains containing the plasmid pQQ05 were fermented in shake flasks at the same time. The results showed that three single thioesterase gene knockout strains can accumulate mcl-PHA polymers, and the yield is higher than that of the strains with xylose as the single carbon source for mcl-PHA synthesis.In order to further increase mcl-PHA production, double thioesterase genes were knocked out in recombinant E. coli in this study, and double-knockout strains were obtained. The strain LZ05(△tesB△yciA) harboring the plasmid pQQ05 accumulated 6.62 wt% mcl-PHA, LZ06(△tesB△tesA) harboring the same plasmid produced 6.28 wt% of mcl-PHA, while LZ07(AtesAAyciA) containing pQQ05 had the lowest accumulation of mcl-PHA about 5.07% of cell dry weight. These three strains grew well, and achieved the cell dry weight higher than 6 g l-1.To get scl-mcl PHA containing both short-chain-length and medium-chain-length monomers, we used a low substrate specificity PHA polymerase-PhaC2ps from Pseudomonas stutzeri 1317 to replace the PHA polymerase-PhaC2pa from Pseudomonas aeruginosa PAO1, and construct the plasmid pQQ06 for mcl-PHA synthesis. After the plasmid pQQ06 transforming into the strain LZ05, in the case of addition of 30 g l-1 glucose, it can synthesize about 12.10% scl-mcl PHA of cell dry weight, which accounted for 21.18 mol% 3HB and 78.82 mol% medium-chain-length monomers. This approach can be used to produce scl-mcl PHA in industrial scale after optimization.All of the above synthesized PHA only contained even-numbered mcl-PHA monomers, and have shown desirable physical and mechanical properties. If the odd-numbered monomers were incorporated into mcl-PHA, they may endow the biological materials higher strength and flexibility, making it more innovative and practical. In view of this, the urgent need was to build an efficient PHA biosynthetic pathway, so that the desired mcl-PHA containing odd-numbered monomers can be synthesized from corresponding odd-chain (R)-3-hydroxyacyl-CoA precursors. To date, there has been no report of mcl-PHA production from glucose and propionate in E coli containing different odd-numbered monomers. For this reason, the functional fatty acid β-oxidation reversal was utilized to supply two-carbon extending acyl-CoA molecules from unrelated and cheap carbon source as precursors to synthesize different odd-numbered (R)-3-hydroxyacyl-CoA instead of adding only related carbon sources-fatty acids. To synthesize mcl-PHA that contained odd-chain monomers from the reversed fatty acid cycle, the starting precursor propionyl-CoA must be provided.In the present study, the metabolic pathway for PHA production consists of two parallel pathways leading to the production of the even-chain monomers and the odd-chain monomers. This is the first case revealing that engineered E. coli can produce mcl-PHA consisting of the highest amount of odd-chain ranging from C7 to C13 from glucose with addition of propionate.In order to supply odd-chain monomer precursors, we cloned genes prpP, prpE, pct from R. eutropha H16 and acs from E. coli and then constructed other four plasmids, namely pZQ01, pZQ02, pZQ03 and pZQ04. in the engineered pathway of odd-chain precursor supply, propionate uptake can be promoted by propionate permease (PrpP) according to the previous results. Meanwhile, propionate is also directly activated by propionyl-CoA synthetase (PrpE/Acs) or propionate CoA-transferase (Pct) to incorporate a 3-hydroxyvaleryl-CoA unit into the extended polymer chain by the PHA synthase encoded by phaC2pa from Pseudomonas aeruginosa PAO1.To analyze the effect of genes prpP, acs, prpE and pct on the odd-chain fraction in mcl-PHA production, we co-transformed the plasmids pQQ05 and pZQ01, pQQ05 and pZQ02, pQQ05 and pZQ03, pQQ05 and pZQ04, respectively, into the engineered E. coli strain LZ05 which yielded the highest content of even-chain mcl-PHA. After the shake flask study at 30℃ and 250 rpm with addition of 30 gl-1 glucose and 1.5 g l-1 propionate, the recombinant strain LZ05 harboring the above combination of plasmids were all able to accumulate even- and odd-numbered mcl-PHA, among which LZ05 containing pQQ05 and pZQ03 accumulated the highest content of mcl-PHA compared with the other three combination of plasmids. In addition, the engineered strain LZ05 harboring plasmids pQQ05 and pZQ01 produced mcl-PHA approximately 3.21% of CDW with the highest amount of odd-numbered mcl-PHA monomers up to 11.24 mol%, among which the 3HHp fraction in mcl-PHA was maintained at about 6.46 mol%. These results suggested that a relatively higher metabolic flux was directed to produce propionyl-CoA when overexpressing prpP in the cells. These results showed that prpP, acs, prpE and pct had different impact on mcl-PHA production. According to these experimental results, the strains still did not have sufficient molecules for initiation the formation of odd-chain monomers.To increase the mcl-PHA production and possess more odd-numbered monomers in recombinant E. coli, we rebuilt the biosynthetic pathway by construction of the plasmids pZQ05, pZQ06 and pZQ07. In the above plasmids, prpP and acs, prpP and prpE, and prpP and pct were simultaneously overexpressed separately, due to the highest gain of odd-numbered monomers when overexpression of prpP. Then, they were all co-transformed with pQQ05 into the strain LZ05 to form LZ05(pQQ05, pZQ05), LZ05(pQQ05, pZQ06), and LZ05(pQQ05, pZQ07). Cultivation of these strains showed that they exhibited diverse growth phenomena and different accumulation of mcl-PHA with odd-numbered monomers. The CDW of LZ05 harboring double plasmids pQQ05 and pZQ05 reached up to 6.90 g l-1, higher than that of the strain with single overexpression of prpP or acs. However, the cellular dry weight of LZ05 harboring double plasmids pQQ05 and pZQ06 (or pZQ07) only reached 6.69 g l-1 (or 6.31 g l-1). These results indicated that overexpression of prpP and acs restored the cell growth. With regard to the mcl-PHA content, they had different reflection in different strains. LZ05(pQQ05, pZQ06) accumulated 4.96 wt% mcl-PHA polymers which is the highest among the aforementioned three strains. Nevertheless, the molecular content of odd-numbered fractions in LZ05(pQQ05, pZQ06) of about 10.98 mol% was lower than LZ05(pQQ05, pZQ01).In order to further improve the content of PHA, poxB and pflB genes were deleted from the strain LZ05 and constructed the strain LZ08. Later, pQQ05 and pZQ06 were transformed into LZ08. After the shake flask study,6.23 wt% mcl-PHA harboring odd-numbered monomers about 20.03 mol% in recombinant E. coli was synthesized from 30 g l-1 glucose and 2 gl-1 propionate by GC analysis. This is by far the first report of mcl-PHA production both with even- and odd-numbered monomers.