Genome-wide Determination Of Gene Essentiality And Screening Of Special Functional Genes By Transposon Mutagenesis In Pichia Pastoris

The methylotrophic yeast Pichia pastoris is the most frequently used eukaryotic system for recombinant protein expression.Besides,because of its special ability to utilize methanol as sole carbon source,P.pastoris is also a common model to study methanol assimilation and peroxisome synthesis pathways.Although the complete genome sequence of P.pastoris has already been published in 2009,the functions of most genes have not been identified,and over 30% of the total genes are still annotated as hypothetical genes with unknown functions.In order to accelerate the functional identification of unknown genes of P.pastoris,a highly efficient transposon mutagenesis system based on piggyBac?PB?transposon was developed in P.pastoris for forward genetic screening in this work.In addition,a combinational transposon-based mutagenesis system based on TcBuster?TcB?and Sleeping beauty?SB?transposons was developed in P.pastoris to identify the essential genes and to screen the functional genes.PB transposon has a very high transposition efficiency and precise site-cutting ability,which is one of the most widely used transposons in eukaryotes.The original transposase PBase has a transposition frequency of 0.03,and the hyper active transposase hyPBase can achieve the frequency of 0.43,which means that there almost is a transposon mutant in every two cells.In this study,the PB-based transposon mutant system was used to screen three carbon catabolite repression genes,PAS_chrl-1₀300,PAS_chr4₀334 and PAS_chr4₀769.The first two of them involved in the function of glucose-blocking methanol was reported in the literature,and the latter was the novel gene identified in P.pastoris.Large-scale classification between essential and non-essential genes could help to elucidate the molecular underpinnings of many biological processes.Some specific areas such as metabolic research and synthetic biology benefit a lot from essential gene identification in different organisms.Traditionally it is a challenge to gather essential gene sets since knockouts are lethal and labor-consuming.The combination of transposon mutagenesis and high-throughput sequencing has greatly accelerated the identification of essential genes.For essential gene identification,so far most studies have been done in bacteria with the mature mariner derived or Tn5 derived transposon mediated random insertion system.Transposon based random insertion and subsequent Tn-seq has rarely been tried in eukaryotic systems to classify essential genes.Here,five different transposons were tested in P.pastoris and successfully detected transposition events after the TcB or SB transposon system was introduced,with a net integration frequency-1.2×10-2 and-2.1×10-3,respectively.A "liquid enrichment" approach was also developed to increase the density of transposon mutants and save labor.It was shown that TcB and SB had a certain degree of complementarity in insertional bias in P.pastoris,so the combination of two transposon insertion datasets might be helpful for the analysis of gene essentiality.By merging the TcB and SB insertion libraries and performing Tn-seq,a total of 202,858 unique insertions were identified under glucose supported growth condition,with an average density of 22 insertions per kilobase in genome.Then a machine learning method with logistic regression algorithm was developed to classify the 5040 annotated genes into putatively essential,putatively non-essential,ambigl and ambig2 groups.The putatively essential set possessed 1086 genes,of which 491 genes were also essential in both S.cerevisiae and in S.pombe.These "general" essential genes might reflect the most basic processes required for yeast viability and growth,such as gene transcription,proteins translation and primary metabolism.In contrast to "general" essential genes,orthologs of some essential genes in P.pastoris were not essential in S.cerevisiae and in S.pombe.The genes were likely specific P.pastoris essential genes and play a necessary role in methylotrophic yeast.The accuracy of this classification model was further validated by testing previously reported genes and making new knockouts of predicted essential/non-essential genes by our model.Besides,Tn-seq was also performed under methanol supported growth condition and methanol specific essential genes were identified.The comparison of essential genes between glucose and methanol supported growth conditions helped to reveal potential novel targets involved in methanol metabolism and signaling.As one member of the "non-conventional" yeasts,the number of research groups working on P.pastoris is much smaller than that working on S.cerevisiae and S.pombe.Usually genome-wide knockout and engineering are rarely done in non-conventional yeasts because of limitations in funding resources and labor.Here,we developed a simplified way to classify P.pastoris genes with much less labor consumption.Hopefully this method could also be applied in other non-conventional yeasts.