Date Published: August 8, 2017
Publisher: Springer Berlin Heidelberg
Author(s): Lamis Gomaa, Michael E. Loscar, Haggag S. Zein, Nahed Abdel-Ghaffar, Abdelhadi A. Abdelhadi, Ali S. Abdelaal, Naglaa A. Abdallah.
Isoprene represents a key building block for the production of valuable materials such as latex, synthetic rubber or pharmaceutical precursors and serves as basis for advanced biofuel production. To enhance the production of the volatile natural hydrocarbon isoprene, released by plants, animals and bacteria, the Kudzu isoprene synthase (kIspS) gene has been heterologously expressed in Bacillus subtilis DSM 402 and Bacillus licheniformis DSM 13 using the pHT01 vector. As control, the heterologous expression of KIspS in E. coli BL21 (DE3) with the pET28b vector was used. Isoprene production was analyzed using Gas Chromatography Flame Ionization Detector. The highest isoprene production was observed by recombinant B. subtilis harboring the pHT01-kIspS plasmid which produced 1434.3 μg/L (1275 µg/L/OD) isoprene. This is threefold higher than the wild type which produced 388 μg/L (370 μg/L/OD) isoprene, when both incubated at 30 °C for 48 h and induced with 0.1 mM IPTG. Additionally, recombinant B. subtilis produced fivefold higher than the recombinant B. licheniformis, which produced 437.2 μg/L (249 μg/L/OD) isoprene when incubated at 37 °C for 48 h induced with 0.1 mM IPTG. This is the first report of optimized isoprene production in B. licheniformis. However, recombinant B. licheniformis showed less isoprene production. Therefore, recombinant B. subtilis is considered as a versatile host for heterologous production of isoprene.
Isoprene is a small volatile hydrophobic molecule containing five carbon atoms and is also known as 2-methyl-1,3-butadiene. It is a colorless organic compound that is produced by animals, plants and bacteria. It has a low solubility in water as well as a low boiling point of 34 °C which enables withdrawal from the upper gas phase of a bioreactor when produced via biotechnological processes (Xue and Ahring 2011). This aspect turns it valuable for downstream chemical products. Isoprene as biofuel contains more energy, is not miscible in water and does not show corrosive effects compared to ethanol (Atsumi and Liao 2008; Lindberg et al. 2010). Companies develop bioisoprene production such as Genencor and Goodyear, published their efforts to develop a gas-phase bioprocess for production of isoprene (Whited et al. 2010). Their work involved metabolic engineering of E. coli capable of producing high yields of isoprene, in addition to developing a large-scale fermentation process with high rates of the isoprene recovered from the off-gas. They reported a titer of over 60 g/L, a yield of 11% isoprene from glucose and a volumetric productivity of 2 g/L/h (Chandran et al. 2011). Several researchers reported using isoprenol as anti-knock agent, in which branched C5 alcohols store more energy than ethanol and high octane numbers (RON, or research octane number, of 92–102), that helps their use as gasoline alternatives and as anti-knock additives (Cann and Liao 2010; Mack et al. 2014). In addition, they have been verified in various engine types and they proved to have better gasoline-like properties than ethanol (Yang et al. 2010). All isoprenoids are known to be derived from the two universal five-carbon (C5) building blocks, isopentenyl diphosphate (IPP) and its isomer dimethylallyl diphosphate (DMAPP). These universal precursors are known to be produced by either one of the three identified pathways: the mevalonate (MVA) pathway and/or the pathway, which is also 1-deoxy-d-xylulose-5-phosphate (DXP) pathway. Additionally, an alternative MVA-independent pathway has been identified for the biosynthesis of IPP and DMAPP in bacteria, algae and plants which is named the methylerythritol 4-phosphate (MEP) pathway and lacks the first two steps of the DXP pathway (Rodríguez-Concepción and Boronat 2012). Bacillus subtilis uses the DXP pathway and was found to be the best naturally isoprene producing bacteria (Kuzma et al. 1995). It is known that the isoprene synthase utilize dimethylallyl diphosphate (DMAPP) as substrate (Withers et al. 2007). The isoprene synthase gene was not identified in bacteria yet, however it has been characterized from many plants such as Populus species, e.g. aspen, Poplar Alba (Beatty et al. 2014; Fortunati et al. 2008; Miller et al. 2001; Sasaki et al. 2005; Sharkey et al. 2005; Silver and Fall 1995; Chotani et al. 2013; Vickers et al. 2010), Pueraria Montana (kudzu) and/or Pueraria lobata (Beatty et al. 2014; Hayashi et al. 2015; Sharkey et al. 2005). The isoprene synthase gene from poplar was successfully isolated and heterologously expressed in E. coli. Moreover, isoprene synthase cDNA was isolated from Populus alba (PaIspS) and expressed in E. coli for enzymatic characterization (Sasaki et al. 2005). Previous studies failed to isolate the isoprene synthase from bacteria (Julsing et al. 2007; Sivy et al. 2002), thus the Kudzu isoprene synthase gene (kIspS) was codon optimized and heterologously expressed in E. coli (Zurbriggen et al. 2012). Previously, the codon optimized Mucuna bracteata IspS was engineered in S. cerevisiae and it only produced 16.1 μg/L (Hayashi et al. 2015). Additionally, when the codon optimized M. bracteata IspS was engineered in Pantoea ananatis, it produced 63 μg/L (Hayashi et al. 2015). Recent studies involved in overexpression of codon optimized kudzu IspS (kIspS) in E. coli using different constructs (Cervin et al. 2016). The E. coli best isoprene production yield was 10 μg/L. In addition, the codon optimized kudzu and poplar IspS genes were expressed in Yarrowia lipolytica using different methods; in which the isoprene yield was 0.5–1.0 μg/L from the headspace culture (Cervin et al. 2016). This study aimed to develop recombinant Bacillus strains (B. subtilis and B. licheniformis) with high level of isoprene production using the Kudzu isoprene synthase.
In this study, we developed recombinant Bacillus strains (B. subtilis DSM 402 and B. licheniformis DSM 13) in an attempt to enhance isoprene production using the kudzu isoprene synthase. Interestingly, B. subtilis harboring the pHT01-kIspS plasmid showed a higher production of isoprene than B. licheniformis harboring the same plasmid. Recombinant B. subtilis produced 1434.3 μg/L (1275 μg/L/OD isoprene), which is threefold higher than the wild type that produced 388 μg/L (370 μg/L/OD) isoprene, when both incubated at 30 °C for 48 h by 0.1 mM IPTG induction. Our results are in accordance with a recent study for expression of kIspS in B. subtilis, in which the isoprene production levels were increased also with threefold in comparison to the wild type, from 400 μg/L to 1.2 mg/L in batch culture (Vickers and Sabri 2015). To the best of our knowledge no previous work was done for enhancing isoprene production in B. licheniformis, where this is the first report of optimized isoprene production in B. licheniformis. Since recombinant B. subtilis harboring pHT01-kIspS produce a fivefold higher isoprene production than recombinant B. licheniformis harboring pHT01-kIspS at 48 h incubation with induction by 0.1 mM IPTG. Thus, B. licheniformis seems to be not of significant importance for further studies on isoprene production. Additionally, multiple sequence alignment results showed differences in the codon usage of kIspS optimization for B. subtilis and B. licheniformis, which might be the reason that there is difference in isoprene production for both recombinant bacteria. For isoprene production optimization in our study, on one hand we found that induction by different IPTG concentrations (0.1, 0.5, 1 and 2 mM) did not change the level of isoprene production in B. subtilis and B. licheniformis. Recombinant E. coli BL21 (DE3) harboring pET28b-kIspS showed higher isoprene production (76 µg/L/OD) at 37 °C for 4 h incubation when induced by 0.5 mM IPTG. On the other hand for the effect of NaCl on isoprene production, our results demonstrated that 0.3 M NaCl did not enhance the isoprene production for all strains under study except the wild type B. licheniformis. However, it was revealed that NaCl and heat can induce isoprene production, (Xue and Ahring 2011), in which isoprene increases at temperature ranging between 25 and 45 °C then decreases until it reaches 0 at 65 °C in another study, optimum bacterial isoprene production was obtained at 45 °C (Kuzma et al. 1995). Moreover, when utilizing extra carbon sources to the media, i.e. glucose and/or glycerol, highest isoprene production was observed for the recombinant B. subtilis at 5 g/L glucose as an extra substrate. This result is in contrary to the previous observation (Zurbriggen et al. 2012) for E. coli transformed with kIspS, in which glycerol provided higher yields of isoprene compared to glucose, fructose, xylose, or LB media. In addition, isoprene production assays demonstrated that kIspS expression in E. coli best activity were obtained at 37 °C for 6 h when induced by 0.1 mM IPTG (Zurbriggen et al. 2012). Previous studies demonstrated that Synechocystis PCC6803 and E. coli are responsive strains for heterologous transformation by the IspS gene, in which they express and store the isoprene protein into their cytosol (Lindberg et al. 2010). Recent studies involved in overexpression of codon optimized kudzu IspS (kIspS) in E. coli using different constructs (Cervin et al. 2016). In This study, the E. coli best isoprene production yield was 10 μg/L. Moreover, the codon optimized kudzu and poplar IspS genes were expressed in Y. lipolytica using different methods; in which the isoprene yield was 0.5–1.0 μg/L from the headspace culture (Cervin et al. 2016). Previously, the codon optimized M. bracteata IspS was engineered in Corynebacterium glutamicum and produced 24.2 μg/L, additionally it was also engineered in Enterobacter aerogenes and produced 316 μg/L (Hayashi et al. 2015). Moreover, the kIspS was expressed in Trichoderma reesei; in which it yields 0.5 μg/L isoprene. Also, the synthetic tagged kIspS gene was introduced into Synechocystis sp. PC6803 (Lindberg et al. 2010). Results revealed that low levels of IspS protein were detected, in addition to the codon optimization that significantly enhanced protein production, in which latter strain produced 50 μg isoprene per gram dry cell weight per day, which is equivalent to 4 μg isoprene/L culture/h−1 (Hong et al. 2012). The same kIspS construct was used with a glucose-sensitive version of Synechocystis sp. PC6803; which produced isoprene that peaked at 100–130 μg/L culture (Bentley et al. 2014; Bentley and Melis 2012). Additionally, the isoprene production was slightly improved to 300 μg/L culture by introducing a heterologous MVA pathway (Bentley et al. 2014). There are previous studies on expression of more than one isoprene synthase, which shows are highly production of isoprene. In which the expression of ten isoprene synthase genes from Arachishypogaea together with the MVA pathway in E. coli resulted in the production of up to 35 mg/L/h/OD of isoprene (Beatty et al. 2014). Also, it was shown that heterologous expression of P. alba IspS and S. cerevisiae MVA pathway in E. coli, yield 532 mg/L isoprene in a fed-batch fermentation (Yang et al. 2012). Previous study on enhancing isoprene production through heterologous expression of B. subtilis DXS and DXR yield 314 mg/L isoprene, while over expression endogenous DXS and DXR in E. coli harboring P. nigra IspS gene enhanced isoprene production from 94 to 160 mg/L (Zhao et al. 2011). Additionally, by introducing RBS and nucleotide spacers provide the maximum isoprene expression in E. coli batch cultures from 0.4 mg/L isoprene of the control culture to 5 mg/L isoprene per of MEP super-operon transformants culture and up to 320 mg/L isoprene of MVA super-operon transformants culture (Zurbriggen et al. 2012). It was demonstrated that B. subtilis bears an isoprene synthase activity which utilizes the dimethylallyl diphosphate (DMAPP) as a substrate for isoprene production. Additionally, the isoprene synthase activity was optimal at pH 6.2 as well as it requires low levels of divalent ions and it was found to be separated from the chloroplast isoprene synthase (Sivy et al. 2002). Recently, 1-deoxy-d-xylulose-5-phosphate synthase (Dxs) and 1-deoxy-d-xylulose-5-phosphate reductoisomerase (Dxr) were overexpressed separately in B. subtilis DSM 10 strain. Over expression of Dxs increased the yield of isoprene by 40%. While over expression of Dxr had no change on the level of isoprene production (Xue and Ahring 2011). Concerning the control we have successful results in transformation of recombinant plasmid pET28b-kIspS-C-term in BL21 cells and the highest isoprene production for the recombinant BL21 cells harboring pET28b-kIspS-C-term was 70 μg/L/OD when
incubated at 37 °C for 24 h induced by 0.1 mM IPTG. Previous assays for kIspS expression in E. coli, revealed that the best isoprene production activity were obtained at 37 °C for 6 h with 0.1 mM IPTG induction (Zurbriggen et al. 2012). Moreover, heterologous expression of the codon optimized kIspS in E. coli has been carried out and results showed that there is no significant difference of kIspS gene expression in recombinant and non-recombinant E. coli (Zurbriggen et al. 2012).