Reconstruction And In Silico Analysis Of Genome-scale Metabolic Model For Resistant Penicillium Digitatum

Citrus fruit decay after harvest is the main reason for the losses of citrus industry,and citrus green mold caused by Penicillium digitatum contributes the most serious loss.Chemical pesticide is one of the most important method for controlling P.digitatum,while the emergence of large amount drug-resistant strains reduced their controlling efficiency.Studies showed that there are many kinds of drug resistance mechanisms in P.digitatum.To tolerance 14?-demethylase inhibitors(DMIs),the strain could generate mutations in genes encoding sterol 14?-demethylase(CYP51),leading to the decrease of affinity between CYP51 and the DMIs.Besides,P.digitatum can overexpress CYP51 to compensate CYP51-catalyzed activity suppressed by DMIs.Moreover,overexpression of drug transport proteins including ABC superfamily proteins,MFS superfamily proteins and MATE proteins to quickly discharge the DMIs from the cells is also an important resistance mechanism in P.digitatum.In order to study the resistance mechanism at a systematic level,a dynamic genome-scale metabolic model(GSMM)was constructed and constrained,which can provide new insight in the analysis of P.digitatum resistance mechanism.In this paper,we formulated a GSMM model(named iPD1512)for P.digitatum using the Merlin automatic model construction software,which was then constrained by context-specific RNA-Seq data and experimentally identified metabolic flux data.The model covers 1006 genes and contains five compartments including mitochondrial,peroxisome,cytosol,extracellular and boundary.It contains 1176 reactions in total including 1043 biochemical reactions,128 transport reactions and 5 virtual response.In comparison,genes data from the transcriptome was used in our model reconstruction,which can directly reflect the characteristics of the organization or conditional characteristics of the model.In addition,we integrated the interaction process of drug and pathogen through specific biochemical reactions into the model.This can be used to quantitatively describe the specific role of each gene for the P.digitatum drug resistance.Further more,we used transcriptome data to analyze the changes of the expression level of P.digitatum in different states based on the gene expression-metabolic data integration method,adjusting the corresponding metabolic response rate,and then calculating the growth status of P.digitatum under different conditions use flow balance analysis.Through the integration of expression data with the metabolic model,the mechanism of bio-resistance was explored and the genes related to bio-resistance were sequenced.The simulation results of iPD1512 model show that the model can reflect the effect of cyp51 gene in the process of P.digitatum resistant drug.By adjusting the flow rate of CYP51 catalyzed reaction,the simulated growth conditions were in accordance with the experimental results under 6mg/L prochloraz,and the simulated growth rate was 94%of the experimental value.Using the same method,the growth status of the sreA gene knockout mutant of HS-F6 strain was simulated and it is consistent with the experimental results under the condition of 0mg/L prochloraz and 6mg/L prochloraz.The simulated growth rate was 87%of the measured values.In the iPD1512 model,the simulation of pdmfs2 gene knockout mutant showed a large deviation,and the simulation results were significantly lower than the experimental results.Through further analysis,we deduced that the main reason for the prediction bias is that there may be a mechanism of expression regulation between transporters which leads to the elimination of a transporter,and the other transporters are compensated by increasing the expression level.In summary,combining GSMM with omics data to study P.digitatum resistance mechanism is superior to the traditional statistical method based on omics data analysis.Simulation studies based on the pathogenic microbial metabolic model effectively reveal the intrinsic dynamic relationship between genes.It provides a theoretical basis for optimizing the experimental program and developing fungal disease prevention and control strategies.