Metabolic Engineering Of Central Metabolic Pathway And Comparative Genomics Of Riboflavin-produced Bacillus Subtilis

The whole research can be subdivided into the following several parts. The first part is enhancement of riboflavin production with Bacillus subtilis by metabolic engineering of central metabolic pathway including pentose phosphate pathway (PPP) and gluconeogenic pathway. And the second part is genome sequencing of riboflavin-produced B.subtilis 24 and B.subtilis RH33 using Roche 454 GS FLX. Based on the whole sequenced genome, comparative genomics was conducted to find mutations related to biosynthesis of riboflavin with B.subtilis. Finally, a marker-free genetic manipulation system was constructed, which can be used to accomplish mutation analysis and reengineer a B.subtilis riboflavin producer.Zwf and gnd genes from Corynebacterium glutamicum were firstly cloned, and then site-directed mutagenesis was successfully introduced to remove allosteric inhibition by intracellular metabolites. Expression of the mutant zwf243 and gnd361 in B.subtilis RH33 resulted in significant enhancement of riboflavin productivity, while the specific growth rate decreased slightly and the specific glucose uptake rate was unchanged. Introduction of the mutant zwf243 and gnd361 led to approximately 18% and 22% increased riboflavin production respectively. An improvement by 31% and 39% of the riboflavin production was obtained by co-expression of the mutated dehydrogenases in shaker flask and fed-batch cultivation. Intracellular metabolites analysis based on LC-MS indicated that metabolites such as AIR,DRL and Ru5P detected in pentose phosphate pathway or riboflavin synthesis pathway of engineered strains showed higher concentration (increased 15%-46%) , while TCA cycle and glycolysis metabolites detected were lower abundance than that of parent strain B.subtilis RH33.Integrative vectors were constructed to over-express genes involved in gluconeogenic pathway, and the constructed vectors were denominated as pUC18-TPA, pUC18-NPB, pUC18-SPF, pUC18-PFB and pUC18-PFBA for over-expression of pckA, gapB, fbp, co-expression of fbp and gapB and co-expression of fbp, gapB and pckA, respectively. The integrative vectors pUC18-TPA, pUC18-NPB, pUC18-SPF, pUC18-PFB and pUC18-PFBA were integrated into the B. subtilis RH33 chromosome by single crossover and the transformants were selected by the corresponding antibiotics. Over-expression of fbp in B.subtilis RH33 resulted in significant enhancement of riboflavin productivity, which led to approximately 18% increased riboflavin production with glucose as carbon source in shaker flask. Effect of co-expression of genes on riboflavin production indicated that over-expression of fbp and gapB had largest positive influence, resulted in 22% increased riboflavin production, about 4.89g/L.The next generation 454 GS FLX system was used to sequence riboflavin-produced B.subtilis 24 and B.subtilis RH33 genomes. We generated～195 megabases (Mb) that resulted in～48-fold genome coverage for 24 and～201 megabases (Mb) of sequence data or～46-fold coverage for RH33, and the average read length was 388bp for 24 and 394bp for RH33. With combination of Sanger sequencing for gap filling, the whole genome sequences of B.subtilis 24 and B.subtilis RH33 were obtained. The genome of B.subtilis 24 consists of a 4194978bp single circular chromosome and has an average G+C content of 43.55%, and the B.subtilis RH33 genome is 4195221bp in size, contains 43.55% G+C.The GS Mapper application in the 454 GS FLX software package 1.1.03 (454 Life Sciences) was used to map the pyrosequencing reads onto the B.subtilis 168 reference genome (GenBank accession: AL009126) . Mutations with lower mutation rate (<80%) in high confidence differences (HCDiffs) were indentified by Sanger sequencing to eliminate false mutation. By combining of HCDiffs obtained from GS Mapper application with mutations identified from Sanger sequencing, we indentified all 512 mutations in B.subtilis 24 genome, including 29 insertion mutations, 20 deletion mutations, and 463 substitution mutations. There are 549 mutations indentifided in B.subtilis RH33, including 31 insertion mutations, 24 deletion mutations, and 494 substitution mutations. Comparative genomics of B.subtilis 24 and B.subtilis RH33 indicated that 468 mutations of above-mentioned are common to both genomes, while 44 mutations are peculiar to B.subtilis 24 and 81 mutations are unique to B.subtilis RH33.Manual annotation of mutations was accomplished with Matlab tool and Kegg database, and the mutations were further classified according to their function in metabolic network, and the principle of mutation analysis by“wet”experiment was established. In order to identify the function of mutations, we constructed a marker-free genetic manipulation system with upp as negative selection marker.