Metabolic bottlenecks of Pseudomonas taiwanensis VLB120 during growth on d-xylose via the Weimberg pathway

Abstract

The microbial production of value-added chemicals from renewable feedstocks is an important step towards a sustainable, bio-based economy. Therefore, microbes need to efficiently utilize lignocellulosic biomass and its dominant constituents, such as d-xylose. Pseudomonas taiwanensis VLB120 assimilates d-xylose via the five-step Weimberg pathway. However, the knowledge about the metabolic constraints of the Weimberg pathway, i.e., its regulation, dynamics, and metabolite fluxes, is limited, which hampers the optimization and implementation of this pathway for bioprocesses. We characterized the Weimberg pathway activity of P. taiwanensis VLB120 in terms of biomass growth and the dynamics of pathway intermediates. In batch cultivations, we found excessive accumulation of the intermediates d-xylonolactone and d-xylonate, indicating bottlenecks in d-xylonolactone hydrolysis and d-xylonate uptake. Moreover, the intermediate accumulation was highly dependent on the concentration of d-xylose and the extracellular pH. To encounter the apparent bottlenecks, we identified and overexpressed two genes coding for putative endogenous xylonolactonases PVLB_05820 and PVLB_12345. Compared to the control strain, the overexpression of PVLB_12345 resulted in an increased growth rate and biomass generation of up to 30 % and 100 %, respectively. Next, d-xylonate accumulation was decreased by overexpressing two newly identified d-xylonate transporter genes, PVLB_18545 and gntP (PVLB_13665). Finally, we combined xylonolactonase overexpression with enhanced uptake of d-xylonate by knocking out the gntP repressor gene gntR (PVLB_13655) and increased the growth rate and biomass yield by 50 % and 24 % in stirred-tank bioreactors, respectively. Our study contributes to the fundamental knowledge of the Weimberg pathway in pseudomonads and demonstrates how to encounter the metabolic bottlenecks of the Weimberg pathway to advance strain developments and cell factory design for bioprocesses on renewable feedstocks.

Keywords: Pseudomonas taiwanensis VLB120; Renewable feedstocks; Weimberg pathway; Xylonate transport; Xylonolactonase; Xylose utilization.

Fig. 1

Schematic representation of the Weimberg pathway from P. taiwanensis VLB120. New insights from this publication are highlighted in orange. d-Xylose is hypothesized to be taken up into the periplasm by porin B (OprB). The periplasmic pyrroloquinoline quinone-dependent glucose dehydrogenase (Gcd, PVLB_05240) performs the conversion to d-xylonolactone. Electrons are transferred to ubiquinone (UQ), which is reduced to ubiquinol (UQH₂). D-xylonolactone is hydrolyzed spontaneously or by a xylonolactonase (XLA, PVLB_05820/PVLB_12345) to d-xylonate. d-Xylonate is taken up into the cell by the two transporters GntP (PVLB_13665) and PVLB_18545. Xylonate dehydratase (XAD, PVLB_18565) performs the conversion to 2-keto-3-deoxy-d-xylonate and 2-keto-3-deoxy-d-xylonate dehydratase (KDXD, PVLB_18560) to α-ketoglutaric semialdehyde. Conversion to α-ketoglutarate is performed by α-ketoglutarate semialdehyde dehydrogenase (KGSADH, PVLB_11380/PVLB_18510/PVLB_18550). OM (outer membrane), PP (periplasm), IM (inner membrane). The numbers in the figure represent locus tags without PVLB prefix. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

Growth of P. taiwanensis VLB120ΔC at differentd-xylose and ammonium chloride (NH₄Cl) concentrations. Growth rates during exponential growth (A) were determined from the backscatter signal of the cultures. Final biomass concentrations (B) and biomass yields (C) were calculated from OD₄₅₀ measurements at the end of the cultivations. Final pH values (D) were measured at the end of the cultivations. Cultivations were performed in a BioLector I in modified M9 medium at 1 mL scale, 1200 rpm, 30 °C for 72 h. The mean values and error bars (standard deviations) were calculated from two biologically independent cultivations.

Fig. 3

Bioreactor cultivations of P. taiwanensis VLB120ΔC with and without pH control. Uncontrolled cultivation with initial pH of 7.4 (A) and pH-controlled cultivations at pH 5.0 (B), pH 5.6 (C), pH 6.2 (D), pH 6.8 (E), pH 7.4 (F) were performed in 200 mL M9 medium (20 g L⁻¹d-xylose and 2 g L⁻¹ NH₄Cl) in DASbox bioreactors. The batch bioreactors were operated at 30 °C, 1000 rpm, and aeration of 0.25 vvm. The pH was controlled with 1 M NaOH and 1 M H₃PO₄.

Fig. 4

Cultivations of P. taiwanensis VLB120ΔC with plasmid-based overexpression of putative lactonase genes. Growth rates during exponential growth (A) were determined from the backscatter signal. Final biomass concentrations (B) were calculated from OD₄₅₀ measurements at the end of the cultivations. Final pH values (C) and final metabolite concentrations of d-xylonolactone (Xla) and d-xylonate (Xlt) (D) were measured at the end of the cultivations. Cultivations of the empty vector control (EV) and the strains overexpressing PVLB_05820 and PVLB_12345 were performed in a BioLector I in M9 medium (varying concentrations of d-xylose, 2 g L⁻¹ NH₄Cl) at 1 mL scale, 1200 rpm, 30 °C for 72 h. The corresponding growth curves are depicted in Fig. S5. The mean values and error bars (standard deviations) were calculated from at least two biologically independent cultivations. One-way ANOVA tests including post hoc analyses (Bonferroni) were performed for the mean values. Asterisks denote statistically significant difference between two mean values: p ≤ 0.05 (*), p ≤ 0.01 (**), p ≤ 0.001 (***). The level of significance between the engineered strains and the empty vector control is depicted. The detailed statistical report of the one-way ANOVA analysis is shown in the supplementary material (Table S6 to Table S9).

Fig. 5

Complementation assay of genes potentially relevant ford-xylonate transport in P. taiwanensis VLB120. Growth experiments with P. taiwanensis VLB120ΔC (reference strain) and the knockout strains P. taiwanensis VLB120ΔCΔgntP, P. taiwanensis VLB120ΔCΔPVLB_18545 and P. taiwanensis VLB120ΔCΔgntPΔPVLB_18545 (A). The double-knockout strain P. taiwanensis VLB120ΔCΔgntPΔPVLB_18545 was complemented with plasmid-based overexpression of the transporter genes gntP, PVLB_18545, and kguT(B). All cultivations were performed in M9 medium (20 g L⁻¹d-xylose and 2 g L⁻¹ NH₄Cl) at 25 mL scale, 30 °C and 200 rpm. The mean values and error bars (standard deviations) were calculated from two biologically independent cultivations.

Fig. 6

Transporter gene overexpression in P. taiwanensis VLB120ΔC. Panels depict the initial growth rates of the exponential growth phases (A), final biomass concentrations (B), final d-xylonolactone concentrations (C) and final d-xylonate concentrations (D). Cultivations of the empty vector control (EV) and the strains overexpressing gntP, PVLB_18545, and kguT were performed in a BioLector I in M9 medium (20 g L⁻¹d-xylose and 2 g L⁻¹ NH₄Cl) at 1 mL scale, 1200 rpm, 30 °C for 96 h. The growth rate was calculated from the backscatter signal. For the final biomass concentration, the OD₄₅₀ was measured. The mean values and error bars (standard deviations) were calculated from two biologically independent cultivations. One-way ANOVA tests including post hoc analyses (Bonferroni) were performed for the mean values. Asterisks denote statistically significant difference between two mean values: p ≤ 0.05 (*), p ≤ 0.01 (**), p ≤ 0.001 (***). The level of significance between the engineered strains and the empty vector control is depicted. The detailed statistical report of the ANOVA analysis is shown in the supplementary material (Table S10).

Fig. 7

Bioreactor cultivations of single and combinatorial optimization approaches.P. taiwanensis VLB120ΔC containing pCom10Syn35T (empty vector control) (A), P. taiwanensis VLB120ΔC pCom10Syn35T_PVLB12345 (B), P. taiwanensis VLB120ΔCΔgntR pCom10Syn35T (C) and P. taiwanensis VLB120ΔCΔgntR pCom10Syn35T_PVLB_12345 (D) were cultivated in 200 mL M9 medium (20 g L⁻¹d-xylose and 2 g L⁻¹ NH₄Cl). Cultivations were performed in the DASbox over 75 h at 30 °C at 1000 rpm and an aeration of 3 L h⁻¹.