The Establishment And Applications Of The Pathway Assembly Method Based On Standard Biological Elements In Saccharomyces Cerevisiae

In metabolic engineering,construction along with optimization of the exogenous pathways and modification of the endogenous metabolic network are necessary for the transformation of Saccharomyces cerevisiae into a cell factory.However,yeasts have evolved an intricate metabolic regulatory network,making the exogenous pathways face many difficulties such as deficiency of precursors and low expression of enzymes.Therefore,optimizing exogenous metabolic pathways in yeasts is challenging.At present,development of rapid construction and integration of exogenous metabolic pathways,and the introduction of mathematical models to guide the optimization process,are two important directions of efforts.In this study,by rational design in yeasts,we developed a method named YeastFab(Yeast Fabrication)for the rapid construction,assembly and integration of multiple-gene complex exogenous metabolic pathways.Meanwhile,we built an artificial neural network(ANN)ensemble to guide the optimization of exogenous metabolic pathways.After being trained and verified experimentally with the data generated by YeastFab,ANN ensemble was employed to precisely tune the expression level of each gene in pathways for the efficient synthesis of metabolites in cell factories.The YeastFab developed in this study has many advantages,such as a high-throughput manner,easy-operation,and stable integration of the entire metabolic pathway in yeasts with high efficiency.Also,the entire process from cloning to testing could be fulfilled within one week.Firstly,standard biological parts,which were the promoters(PRO),open reading frames(ORF)and terminators(TER),were cloned into standard accepting vectors,respectively.Secondly,any three of each biological part could be assembled into the transcription units(TUs)with the adoption of the Golden-Gate assembly method.Lastly,according to the number of TUs in the metabolic pathway,multiple TUs were assembled and integrated into a target locus in yeasts in different ways in one or several times.In addition,the introduction of double-stranded breaks at the site of integration further improved the efficiency of the integration.Also,the ANN ensemble was established,in combination with YeastFab,to guide the optimization of metabolic pathways.First,by YeastFab,a certain number of metabolic pathways with different promoters were constructed,of which the data was used to train the model.Then,in order to achieve the optimal expression of metabolic pathways,the model predicting the optimal promoter strength of each TU was further verified experimentally and was revised to improve the predictive ability.The?-carotene pathway was taken as an example.A starting library(~20),which is about 2%of the total library(10~3),was used to train the model.And the yield of the predicted best?-carotene-producing strain increased by 65%over the best one in the starting library,indicating the simulation and prediction of the model were excellent.In summary,in this study,the YeastFab was developed to rapidly construct and integrate exogenous metabolic pathways into yeasts,and to optimize them.In addition,the ANN ensemble was adopted to predict the exogenous metabolic pathways,and to guide their optimization to improve the production efficiency.All in all,the combination of model guidance and experimental techniques was proved an excellent potential of application in cell factories.