2,3-Butanediol (2,3-BD) with low toxicity to microbes, could be a promising alternative for biofuel production. However, most of the 2,3-BD producers are opportunistic pathogens that are not suitable for industrial-scale fermentation. In our previous study, wild-type Bacillus subtilis 168, as a class I microorganism, was first found to generate only d-(−)-2,3-BD (purity >99 %) under low oxygen conditions. In this work, B. subtilis was engineered to produce chiral pure meso-2,3-BD. First, d-(−)-2,3-BD production was abolished by deleting d-(−)-2,3-BD dehydrogenase coding gene bdhA, and acoA gene was knocked out to prevent the degradation of acetoin (AC), the immediate precursor of 2,3-BD. Next, both pta and ldh gene were deleted to decrease the accumulation of the byproducts, acetate and l-lactate. We further introduced the meso-2,3-BD dehydrogenase coding gene budC from Klebsiella pneumoniae CICC10011, as well as overexpressed alsSD in the tetra-mutant (ΔacoAΔbdhAΔptaΔldh) to achieve the efficient production of chiral meso-2,3-BD. Finally, the pool of NADH availability was further increased to facilitate the conversion of meso-2,3-BD from AC by overexpressing udhA gene (coding a soluble transhydrogenase) and low dissolved oxygen control during the cultivation. Under microaerobic oxygen conditions, the best strain BSF9 produced 103.7 g/L meso-2,3-BD with a yield of 0.487 g/g glucose in the 5-L batch fermenter, and the titer of the main byproduct AC was no more than 1.1 g/L. This work offered a novel strategy for the production of chiral pure meso-2,3-BD in B. subtilis. To our knowledge, this is the first report indicating that metabolic engineered B. subtilis could produce chiral meso-2,3-BD with high purity under limited oxygen conditions. These results further demonstrated that B. subtilis as a class I microorganism is a competitive industrial-level meso-2,3-BD producer.

Succinate has been recognized as one of the most important bio-based building block chemicals due to its numerous potential applications. However, efficient methods for the production of succinate from lignocellulosic feedstock were rarely reported. Nevertheless, Corynebacterium glutamicum was engineered to efficiently produce succinate from glucose in our previous study. In this work, C. glutamicum was engineered for efficient succinate production from lignocellulosic hydrolysate. First, xylose utilization of C. glutamicum was optimized by heterologous expression of xylA and xylB genes from different sources. Next, xylA and xylB from Xanthomonas campestris were selected among four candidates to accelerate xylose consumption and cell growth. Subsequently, the optimal xylA and xylB were co-expressed in C. glutamicum strain SAZ3 (ΔldhAΔptaΔpqoΔcatPsod-ppcPsod-pyc) along with genes encoding pyruvate carboxylase, citrate synthase, and a succinate exporter to achieve succinate production from xylose in a two-stage fermentation process. Xylose utilization and succinate production were further improved by overexpressing the endogenous tkt and tal genes and introducing araE from Bacillus subtilis. The final strain C. glutamicum CGS5 showed an excellent ability to produce succinate in two-stage fermentations by co-utilizing a glucose–xylose mixture under anaerobic conditions. A succinate titer of 98.6 g L−1 was produced from corn stalk hydrolysate with a yield of 0.87 g/g total substrates and a productivity of 4.29 g L−1 h−1 during the anaerobic stage. This work introduces an efficient process for the bioconversion of biomass into succinate using a thoroughly engineered strain of C. glutamicum. To the best of our knowledge, this is the highest titer of succinate produced from non-food lignocellulosic feedstock, which highlights that the biosafety level 1 microorganism C. glutamicum is a promising platform for the envisioned lignocellulosic biorefinery.

Lactoyl-CoA is critical for the biosynthesis of biodegradable and biocompatible lactate-based copolymers, which have wide applications. However, reports on acetyl-CoA: lactate CoA-transferases (ALCTs) are rare. To exploit novel ALCTs, amino acid sequence similarity searches based on the CoA-transferases from Clostridium propionicum and Megasphaera elsdenii were conducted. Two known and three novel enzymes were expressed, purified and characterized. Three novel ALCTs were identified, one each from Megasphaera sp. DISK 18, Clostridium lactatifermentans An75 and Firmicutes bacterium CAG: 466. ME-PCT from Megasphaera elsdenii had the highest catalytic efficiency for both acetyl-CoA (264.22 s−1 mM−1) and d-lactate (84.18 s−1 mM−1) with a broad temperature range for activity and good stability. This study, therefore, offers novel and efficient enzymes for lactoyl-CoA generation. To our best knowledge, this is the first report on the systematic mining of ALCTs, which offers valuable new tools for the engineering of pathways that rely on these enzymes.