Overexpressions of xylA and xylB in Klebsiella pneumoniae Lead to Enhanced 1,3-Propanediol Production by Cofermentation of Glycerol and Xylose

Abstract

1,3-Propanediol (1,3-PD) is a valuable platform compound. Many studies have shown that the supplement of NADH plays a key role in the bioproduction of 1,3-PD from Klebsiella pneumoniae. In this study, the xylA and xylB genes from Escherichia coli were overexpressed individually or simultaneously in K. pneumoniae to improve the production of 1,3-PD by cofermentation of glycerol and xylose. Compared with the parent strain, the xylose consumption was significantly increased by the introduction of these two genes. The 1,3-PD titers were raised from 17.9 g/l to 23.5, 23.9, and 24.4 g/l, respectively, by the overexpression of xylA and xylB as well as their coexpression. The glycerol conversion rate (mol/mol) was enhanced from 54.1% to 73.8%. The concentration of 2,3-butanediol was increased by 50% at the middle stage but drastically decreased after that. The NADH and NADH/NAD⁺ ratio were improved. This report suggests that overexpression of xylA or xylB is an effective strategy to improve the xylose assimilation rate to provide abundant reducing power for the biosynthesis of 1,3-PD in K. pneumoniae.

Introduction

1,3-Propanediol (1,3-PD), a versatile platform chemical, is one of the most valuable products selected by the US Department of Energy [3]. I t can be used as an important monomer in the synthesis of polytrimethylene terephthalate, which has considerable commercial value due to its superior stretch and unmatched recovery [11]. The conventional chemical synthesis of 1,3-PD requires rigorous conditions and generates multiple byproducts. These environmentally unfriendly chemical routes hinder their commercial applications in the production of 1,3-PD.

There are some routes in different microorganisms for the bioproduction of 1,3-PD from renewable resources [1,2,13,14,17]. Among these strains, Klebsiella is the most efficient cell factory. In Klebsiella, glycerol can be assimilated by the oxidative and reductive pathways. In the oxidative pathway, glycerol is oxidized to dihydroxyacetone and pyruvate successively and NADH is generated. In the reductive route, glycerol is dehydrated by a coenzyme B12-dependent glycerol dehydratase (DhaB) to 3-hydroxypropionaldehyde (3-HPA). 3-HPA can be further reduced to 1,3-PD by 1,3-propanediol oxidoreductase (DhaT) with the consumption of NADH, which is generated by the glycerol oxidative pathway, to maintain intracellular redox balance. Previous attempts to increase the production of 1,3-PD in K. pneumoniae by overexpression of DhaB and DhaT demonstrated that the catalytic activities of these two enzymes were not the limiting factors in the synthesis of 1,3-PD [20]. Apart from the enzymes involved in the 1,3-PD biosynthesis pathway, some reports showed that the intracellular redox balance might be critical for the synthesis of 1,3-PD from glycerol [18,19].

The generation of intracellular reducing power mainly depends on the oxidation of carbohydrate. Generally, bacteria prefer to utilize glucose as the sole carbon source. However, glucose is also the inhibitor for the utilization of glycerol, impeding the synthesis of 1,3-PD from glycerol. Xylose is another abundant monosaccharide after glucose. It has been reported that xylose could be used as a cosubstrate to increase the production of 1,3-PD in K. pneumoniae from glycerol [8]. In Klebsiella, xylose is converted into xylulose by xylose isomerase (XylA) and then catalyzed into xylulose 5-phosphate by xylulokinase (XylB). The xylulose 5-phosphate is further metabolized to generate equivalent power and energy for cell growth. The addition of xylose as a cosubstrate also improved the production of 1,3-PD by about 60% in a recombinant Escherichia coli with introduced 1,3-PD synthesis pathway [15]. As further described by Jin et al. [7], the production of 1,3-PD could also be enhanced by the addition of hemicellulosic hydrolysates rich in xylose. These studies have suggested the feasibility of using xylose as a cosubstrate to improve the synthesis of 1,3-PD. Zhou et al. [21] described that overexpression of xylA was beneficial for the xylose utilization and ethanol production in Saccharomyces cerevisiae. Similarly, the overexpressed xylB also improved the alcoholic production of xylose in yeast [4]. These reports indicate that improving xylose consumption may be a feasible strategy to provide abundant reducing power and energy for the biosynthesis of 1,3-PD.

In this study, the xylA and xylB genes from E. coli were overexpressed in K. pneumoniae. With xylose as a cosubstrate, the titer of 1,3-PD in the recombinant strains was greatly enhanced. The influences of the overexpressed enzymes on cell metabolism and the potential mechanisms involved were also analyzed.

 

Materials and Methods

Strains, Plasmids, and Chemicals

The K. pneumoniae CICIM B0057 isolated and stored in our laboratory was used as the parent strain [22]. E. coli JM109 (Invitrogen) and pEtac [22] were utilized for plasmid construction. 1,3-PD was purchased from Sigma-Aldrich (Germany). The DNA polymerase, restriction endonucleases, ligase (solution I), Bacterial Genomic DNA Purification Kit, and Gel DNA Purification Kit were purchased from Takara (China). The Amplite Fluorimetric NAD/NADH Ratio Assay Kit was purchased from AAT Bioquest (USA). Tryptone and yeast extract were supplied by Oxoid (UK). NAD+, NADH, isopropyl β-ᴅ-1-thiogalactopyranoside (IPTG), and kanamycin were bought from Sangon (China). All primers were synthesized by Sangon (China). All other compounds were of reagent grade or higher quality.

DNA Isolation and Manipulation

The genomic DNA of E. coli BL21 (DE3) was extracted with the Bacterial Genomic DNA Purification Kit and used as the template for the cloning of the xylA and xylB genes. The primers were designed based upon the genomic sequence of E. coli BL21 (DE3) (NCBI No. NC_012967.1). After the xylA gene was cloned by primers P1 (5’-GGC GGA ATT CAT GCA AGC CTA TTT TGA CCA GC-3’) and P2 (5’-TAT AAG CTT CGG GCC AAC GGA CTG CAC AG-3’), it was digested and inserted into the EcoRI and HindIII sites of pEtac to generate the plasmid pEtac/xylA. The primers P3 (5’-GCG GAA TTC ATG TAT ATC GGG ATA GAT CT-3’) and P4 (5’-TAG CTC GAG TTA CGC CAT TAA TGG CAG AA-3’) were used to clone the xylB gene into the EcoRI and XhoI sites of pEtac, generating the plasmid pEtac/xylB. The xylB containing the tac promoter from the plasmid pEtac/xylB was amplified by the primers P5 (5’-TAA AAG CTT GGA GCT TAT CGA CTG CAC G -3’) and P4, and the fragment obtained was subsequently inserted into the HindIII and XhoI sites of pEtac/xylA to generate the plasmid pEtac/xylA-xylB. The PCR amplification conditions were as follows: initial denaturation at 94℃ for 10 min followed by 30 cycles each consisting of 98℃ for 15 sec, 60℃ for 15 sec, and 72℃ for 2 min. The recombinant plasmids obtained and the plasmid pEtac were transformed into K. pneumoniae CICIM B0057, resulting in four recombinant strains named as K. pneumoniae CICIM B0057 (pEtac/xylA), K. pneumoniae CICIM B0057 (pEtac/xylB), K. pneumoniae CICIM B0057 (pEtac/xylA-xylB), and K. pneumoniae CICIM B0057 (pEtac), respectively.

Media and Cultivation

Luria-Bertani medium (1% tryptone, 0.5% yeast extract, and 1% sodium chloride) was used for the cultivation of K. pneumoniae and E. coli. The fermentation medium of K. pneumoniae contained (g/l) glycerol, 40; glucose, 5; xylose, 8; K2HPO4, 7.5; (NH4)2SO4, 2; MgSO4·7H2O, 2; FeSO4·7H2O, 0.005; yeast extract, 7; vitamin B12, 0.015; and 1 ml of trace element solution. The composition of the trace element solution was (g/l) ZnCl2, 0.7; MnCl2·4H2O, 1; H3BO3, 0.6; CoCl2·6H2O, 2; CuCl2, 0.2; NiCl2·6H2O, 0.25; and Na2MoO4·2H2O, 0.35. Kanamycin was added at a final concentration of 100 μg/ml if necessary. The shake-flask fermentations of K. pneumoniae were performed in a 250ml gauze and paper-covered conical flask containing 50 ml of medium at 37℃ in a rotary shaker incubator at 100 rpm.

Protein Assay

The recombinant K. pneumoniae cells were collected by centrifugation at 10,000 ×g for 10 min after 12 h cultivation. The precipitated cells were washed twice and suspended in 100 mM potassium phosphate buffer (pH 7.0). After the cells were disrupted by sonication and centrifuged at 10,000 ×g for 10 min, the supernatant was collected as the crude protein extract. All the procedures were performed at 4℃. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed on a 12% running gel [12] and resolved proteins were visualized by staining with Coomassie Brilliant Blue G250.

Analytical Methods

The biomass was determined by the turbidity of the cell culture at 600 nm. The cell dry weight (CDW) was calculated by the following equation: CDW (g/l) = 0.36OD600. Glycerol, 1,3-PD, as well as other metabolites were analyzed by high-performance liquid chromatography (HPLC) with a refractive index detector and a Bio-Rad Aminex organic acids HPX-87H column at 60℃, with 5 mM H2SO4 as the mobile phase at 0.6 ml/min [6]. The concentration of internal NADH and NAD+ levels were measured with the Amplite Fluorimetric NAD/NADH Ratio Assay Kit.

 

Results and Discussion

Effect of XylA on Cell Growth and 1,3-PD Production of Recombinant K. pneumoniae CICIM B0057

It has been reported that the activity of XylA might be the limiting factor for xylose assimilation [10]. To improve xylose utilization, the xylA gene was cloned from E. coli and overexpressed in K. pneumoniae CICIM B0057 as described in the Materials and Methods, generating the recombinant K. pneumoniae CICIM B0057 (pEtac/xylA). The xylA gene was induced by the addition of IPTG to the final concentration of 1 mM and its expression level was analyzed by SDS-PAGE. As shown in Fig. 1, a significant amount of 44 kDa proteins, similar to the theoretical molecular mass of XylA, were detected in the soluble fractions of K. pneumoniae CICIM B0057 (pEtac/xylA), indicating the soluble expression of XylA. The effect of XylA on the 1,3-PD production of K. pneumoniae CICIM B0057 was analyzed with the addition of 8 g/l xylose. The 1,3-PD titer of K. pneumoniae CICIM B0057 (pEtac/xylA) was increased f rom 18.1 to 23.5 g/l and the b iomass was increased by 11.1% with the addition of 0.2 mM IPTG (Fig. 2). However, further increased IPTG concentrations showed no influence on the titer of 1,3-PD in K. pneumoniae CICIM B0057 (pEtac/xylA) (Fig. 2) and K. pneumoniae CICIM B0057 (pEtac) (data not shown). Thus, 0.2 mM IPTG was added to induce the expression of xylA in further study.

Fig. 1.SDS-PAGE analysis of the overexpressed xylA and xylB. Lane M, molecular weight marker; lanes 1, 4, and 6, the intracellular soluble proteins of K. pneumoniae CICIM B0057 (pEtac/xylA), K. pneumoniae CICIM B0057 (pEtac/ xylB), and K. pneumoniae CICIM B0057 (pEtac/xylA-xylB), respectively; lanes 2, 3, and 5, the intracellular soluble proteins of K. pneumoniae CICIM B0057 (pEtac). The recombinant proteins are indicated by an arrow.

Fig. 2.Effects of IPTG concentrations on the cell growth and 1,3-PD production with the addition of 8 g/l xylose. (A) K. pneumoniae CICIM B0057 (pEtac/xylA); (B) K. pneumoniae CICIM B0057 (pEtac/xylB); (C) K. pneumoniae CICIM B0057 (pEtac/xylA-xylB).

The overexpression of XylA had different effects on the consumption of different carbon sources. When glucose was added to improve the biomass of the recombinant bacteria, it was exhausted at about 4 h (data no shown). The introduced xylA had no effect on the consumption of glycerol. However, the consumption of xylose was increased from 50% to 77.5% and the biomass was improved from 2.94 to 3.42 g/l by the overexpression of xylA (Fig. 3), suggesting significantly improved xylose catabolism capability and cell growth. The enhanced utilization of xylose provided more reducing equivalent than the parent strain, leading to the improved 1,3-PD titer (23.5 g/l), which was nearly 1.3-fold of that produced by K. pneumoniae CICIM B0057 (pEtac) (17.9 g/l). The 2,3-butanediol (BDO) concentration of K. pneumoniae CICIM B0057 (pEtac) and K. pneumoniae CICIM B0057 (pEtac/xylA) peaked at 8.0 and 12 g/l at about 35 h (Fig. 3), respectively, suggesting increased BDO accumulation at the middle stage of fermentation by the overexpression of xylA. Since xylose can be directly transformed into BDO in Klebsiella, the enhanced xylose consumption of K. pneumoniae CICIM B0057 (pEtac/xylA) led to an increasing accumulation of BDO. Besides this, the synthesis of BDO also benefited from the improved NADH level. It was noteworthy that the overexpressed xylA also resulted in the rapidly reducing concentration of BDO at the late stage of fermentation (Fig. 3). The accumulated high level of BDO might be used by the reversible BDO synthesis pathway [5] to maintain cell metabolism balance when the carbon sources are mostly consumed. The regenerated NADH from the degraded BDO might further promote the synthesis of 1,3-PD, resulting in the more efficient production of 1,3-PD in the recombinant harboring xylA after 40 h. Even though overexpression of xylA improved xylose consumption, more carbon source was redistributed to 1,3-PD and biomass, leading to decreased acetic acid accumulation by 25% (Fig. 3).

Fig. 3.Time course of the fermentation of recombinants under optimum conditions. -■-, K. pneumoniae CICIM B0057 (pEtac); -○-, K. pneumoniae CICIM B0057 (pEtac/xylA); -△-, K. pneumoniae CICIM B0057 (pEtac/xylB); -◇-, K. pneumoniae CICIM B0057 (pEtac/ xylA-xylB).

Effect of XylB on Cell Growth and 1,3-PD Production of K. pneumoniae CICIM B0057

The xylB gene was introduced into K. pneumoniae CICIM B0057, resulting in the recombinant K. pneumoniae CICIM B0057 (pEtac/xylB). The soluble expression of xylB induced by 1.0 mM IPTG was confirmed by SDS-PAGE analysis (Fig. 1). Notably, a high concentration of IPTG significantly inhibited cell growth and 1,3-PD synthesis with the addition of 8 g/l xylose (Fig. 2), indicating the negative effects of the high-level expression of xylB on the cell growth and 1,3-PD production. In the xylose metabolic pathway, xylulose is converted to xylulose 5-phosphate by XylB with the consumption of ATP. Thus, the higher XylB activity may lead to a lower ATP level and the accumulation of toxic xylulose 5-phosphate, which are adverse to cell growth [9,16]. As shown in Fig. 2, when the concentration of IPTG was 0.2 mM, the biomass and 1,3-PD titer reached peaks, but they reduced gradually along with the increasing of the concentration of the inducer.

The performance of K. pneumoniae CICIM B0057 (pEtac/xylB) under the optimized conditions (0.2 mM IPTG) is shown in Fig. 3. In comparison with K. pneumoniae CICIM B0057 (pEtac/xylA), the xylose assimilation rate of K. pneumoniae CICIM B0057 (pEtac/xylB) was further increased by about 10%, suggesting that the activity of XylB might be more important to xylose assimilation. The 1,3-PD titer of K. pneumoniae CICIM B0057 (pEtac/xylB) reached its maximum of 23.9 g/l. The concentrations of other compounds and the glycerol consumption of K. pneumoniae CICIM B0057 (pEtac/xylB) were similar to those of K. pneumoniae CICIM B0057 (pEtac/xylA) (Fig. 3).

Effect of Co-Expression of xylA and xylB on Cell Growth and 1,3-PD Production of K. pneumoniae CICIM B0057

To further improve the synthesis of 1,3-PD, the xylA and xylB were simultaneously overexpressed in K. pneumoniae CICIM B0057, resulting in the recombinant K. pneumoniae CICIM B0057 (pEtac/xylA-xylB). The soluble expression levels of xylA and xylB induced by 1.0 mM IPTG were also detected by SDS-PAGE analysis (Fig. 1).

Considering the potential toxic effect of the overexpressed xylB on cell growth, the concentration of IPTG was also optimized. As shown in Fig. 2, the cell growth and 1,3-PD production decreased with the increasing IPTG concentration and the peaks were also obtained at the concentration of 0.2 mM. The cell growth and 1,3-PD production of K. pneumoniae CICIM B0057 (pEtac/xylA-xylB) were higher than those of K. pneumoniae CICIM B0057 (pEtac/xylB) at each IPTG concentration tested, suggesting that the xylose assimilation capacity was further improved by the coproduction strategy. The 1,3-PD titer of K. pneumoniae CICIM B0057 (pEtac/xylA-xylB) reached the peak at 24.4 g/l which was slightly improved than that of the strains harboring the xylA or the xylB gene. The glycerol conversion rate (mol/mol) of K. pneumoniae CICIM B0057 (pEtac/xylA-xylB) reached 73.8%, which was 1.36-fold that of K. pneumoniae CICIM B0057 (pEtac) (54.1%). The concentrations of other compounds and the glycerol consumption were also similar to those of K. pneumoniae CICIM B0057 (pEtac/xylA) (Fig. 3). Although three gene expression strategies were adopted here, there were no significant differences detected among these recombinant strains, suggesting that the upstream pathway was no longer the limiting factor for xylose utilization. Further studies should pay attention to the genetic modification of the downstream of the xylose utilization pathway or the combination with other strategies, such as strengthening the pyruvate-CO2 pathway, to improve the biosynthesis of 1,3-PD.

Effects of the Overexpressed Genes on Internal Reducing Equivalent Balance

At the beginning of fermentation, glucose was utilized as the main carbon source. Owing to the unmodified native EMP pathway, all the recombinants had a similar glucose assimilation rate, resulting in similar NADH and NAD+ levels (Fig. 4). In all strains tested, the levels of NADH were raised with the consumption of carbon sources but decreased after 40 h. Compared with the parent strain, the recombinants harboring the introduced genes had a more efficient xylose catabolism pathway, leading to a higher NADH pool. The NAD+ levels of all strains tested were reduced at the prophase of fermentation but increased after 24 h. The ratios of NADH to NAD+ were significantly improved with the overexpressed genes and kept at a higher level after 12 h (Fig. 4). These results suggested that the overexpressed genes increased the NADH/NAD+ ratio, leading to the improved 1,3-PD titer.

Fig. 4.Effects of the overexpressed genes on the internal reducing equivalent at optimum condition. -■-, K. pneumoniae CICIM B0057 (pEtac); -□-, K. pneumoniae CICIM B0057 (pEtac/xylA); -△-, K. pneumoniae CICIM B0057 (pEtac/xylB); -○-, K. pneumoniae CICIM B0057 (pEtac/xylA-xylB).

In summary, we have strengthened the xylose consumption efficiency by overexpression of XylA and XylB in K. pneumoniae. The recombinant strains showed improved xylose assimilation rates and 1,3-PD titers by cofermentation of glycerol and xylose. This study provided a novel strategy to improve the biosynthesis of 1,3-PD and realize its commercial application in future.