Reconstruction And Application Of Pichia Pastoris Genome-scale Metabolic Model

Pichia pastoris(P.pastoris)is one of the most important cell factories for production of industrial enzymes and foreign proteins.High quality genome-scale metabolic models(GSMMs)are crucial for comprehensively understanding P.pastoris metabolism,for optimizing protein production processes and for systematically engineering novel strains with high performance.In this paper,in order to release the limits of currently existing P.pastoris GSMMs,based on the combination of latest genome annotations and literature data,we firstly updated P.pastoris GSMM,then comprehensively evaluated the model performance and initially tested the model applicability.The new P.pastoris GSMM,iRY1243,was reconstructed using the latest genome annotations in three databases(KEGG,IMG and UniProtKB).Compared with the recently published model iMT1026,the unique ORF number in iRY1243 was increased from 1026 to 1243,the reaction number from 2035 to 2407,and the metabolite number from 1018 to 1094.Especially,to improve the model coverage to the P.pastoris metabolism,the biosynthesis pathways of vitamins and cofactors were carefully checked and added into iRY 1243.As the sensitivity analysis indicated that the biomass composition has obvious effects on the model validation and the strain improvement,we updated the cell composition and non-growth associated ATP maintenance of P.pastoris.The performance of the new model iRY1243 was evaluated using transcriptome data,physiological data,the cell growth capability on different carbon and nitrogen sources,and the metabolic flux distribution when using 13C labeled glucose as the sole carbon source.Transcriptome data showed that 79.37%single-gene reactions and 91.93%multi-gene reactions were detected,indicating that the gene annotation of the model was accurate.With maximization of cell growth rate as the objective function,FBA analysis showed that iRY1243 predicted the specific cell growth rate with 9%average error,the specific oxygen consumption rate with 8.2%average error,and the specific carbon dioxide production rate with 11.1%average error,respectively.The in silico cell growth of iRY1243 on 30 carbon and 21 nitrogen sources were well predicted.The fluxes simulated by FBA was compared with 13C fluxes,and the results showed a good consistency(R2=0.88).These results indicated that iRY1243 can describe the P.pastoris metabolism with satisfied precision.iRY1243 was used to predict the essential genes for cell growth and the potential gene targets to increase the production of beta-galactosidase and S-Adenosyl-L-methionine.On synthetic medium,123 essential genes were found,which were related to energy metabolism,TCA cycle,amino biosynthesis,etc.The simulation data showed that the beta-galactosidase production can be improved from overexpression of genes in PPP pathway or deletion of genes in pathways producing by-products(acetic acid,ethanol and glycerol,etc.).The insertion of vgb gene,the deletion of spe2,aoxl and cys4 genes,and the overexpression of zwf1 and sam2 genes were also benefit to the enhancement of S-Adenosyl-L-methionine production.These predicted results were accordant to the experimental data,indicating that iRY1243 has potential applicability.In summary,iRY1243 has significantly improved the description capability of P.pastoris GSMMs,which lays the foundation for optimizing protein production processes using P.pastoris and for systematically engineering novel P.pastoris strains with high performance.