Lactic Acid Bacteria as a new platform for sustainable production of fuels and chemicals

Abstract

The diminishing natural resources and environmental issues lead us to consider other ways of producing materials, chemicals and energy to satisfy the ever-increasing needs of our society. Lignocellulosic biomass is the most abundant type of substrate in the world; it is also cheap and renewable which makes it a perfect candidate substrate for production of value added products. The second generation biorefineries, employing microorganisms for conversion of lignocellulosic feedstocks into value added products, are not yet employed commercially in a large scale. To increase the economic feasibility of the process, robust microbial catalysts are necessary, both having a broad substrate utilization range and being tolerant to the common inhibitors generated during the lignocellulose pretreatment. Even though many microorganisms are already well characterized and commercially employed in 1st generation biorefineries, the conversion of lignocellulose is a more complex process; thus, the pursue for a suitable microbe continues. In this PhD study, a wide collection of Lactic Acid Bacteria was systematically screened for the strains’ tolerance levels towards various inhibitors coming from the pretreatment of lignocellulosic biomass, as well as for their capabilities to utilize various sugar substrates, including both pentoses and hexoses. Almost 300 strains were tested, including 141 different isolates of Lactobacillus plantarum, L. paraplantarum, L. pentosus, L. brevis, L. buchneri and L. paracasei, and all available Lactobacillus and Pediococcus type strains. Five most promising strains were subjected to further studies; these included L. pentosus LMG 17672, L. pentosus LMG 17673, L. pentosus 10-16, P. pentosaceous ATCC 25745 and P. acidilactici DSM 20284. The strains were tested in growth experiments with increased concentrations of the key inhibitors, such as furfural and HMF, as well as with the presence of the most common combinations of inhibitors, mimicking real-life lignocellulosic feedstocks: sugarcane bagasse, wheat straw and soft wood. The two most promising strains were selected; these were L. pentosus LMG 17673 and P. acidilactici DSM 20284. They were not only found highly resistant to the key inhibitors, but they were also demonstrated to utilize pentoses, xylose and arabinose. For one of the selected most promising strains, P. acidilactici DSM 20284, a chemically defined medium was developed and optimized. The resulting Pediococcus Defined Medium (PDM) proved to support the growth of a variety of other species as well, including all Pediococcus species and several fastidious Lactobacilli. Thus, the PDM medium appears to be superior to the previously published defined media, and can therefore be suitable for physiological, biochemical or nutritional investigations in other LAB species. An efficient transformation procedure is necessary for strain’s rational genetic engineering. To ease strategies for further strain improvement, a transformation procedure was developed and optimized for P. acidilactici DSM 20284, increasing the transformation efficiency by 2 log units. An optimized method allows for the transformation with an efficiency of 2.8·103 transformants per μg DNA, permitting the genetic modification of this strain. In order to even further enhance the P. acidilactici DSM 20284 tolerance to furfural, an adaptation experiment was performed by continuous serial-transfer method. After 408 generations, an adapted strain A28 was isolated and showed an increased growth rate on the rich MRS medium with addition of furfural; yet, it also demonstrated a 27% better growth in MRS medium alone. A whole genome resequencing analysis revealed 62 mutations in the genome of the adapted strain compared to the wild-type. The mutations were mainly single nucleotide polymorphisms, but there were also 12 single insertions identified. More than half of the mutations were non-synonymous substitutions, leading to an amino acid change. Two transcriptional regulators, HrcA and CtsR, were affected by non-synonymous substitutions within the protein or the Shine-Dalgarno sequence, respectively. Several membrane proteins as well as proteins involved in the cell redox homeostasis were also mutated. Purine biosynthesis, salvage and transport related genes were also affected by mutations, likely having an influence on the intracellular nucleotide pool sizes, thereby allowing for an increased growth rate. The analysis of the transcriptomic profiles of the wild-type P. acidilactici DSM 20284 and the adapted strain A28 revealed that the applied furfural concentration did not induce the stress response neither in the wildtype nor in the adapted strain. This finding indicates that both strains are already well adapted to furfural; thereby during the adaptive laboratory evolution experiment the strain adapted towards a faster and more efficient growth on the medium rather than towards furfural resistance. However, several genes related to exopolysaccharide biosynthesis or encoding membrane proteins were induced in the adapted strain, indicating that the cell wall structure might be important for the cell’s protection against furfural. The higher growth rate on the other hand, occurred to be enabled by an optimization of the purine and pyrimidine biosynthesis and salvage pathways, up-regulation of the folic acid biosynthesis as well as several enzymes involved in glycolysis. Finally, the study confirmed the remarkable potential of LAB for their use as microbial cell factories for conversion of lignocellulosic substrates into value-added products.