Abstract The global database on donation and transplantation reported 129,681 worldwide organ transplants in 2020, with kidneys and livers being the most commonly transplanted organs. Rejection of the transplanted organ by the host's immune system is a common complication and a major reason for graft failure and patient mortality. To address these limitations, we developed genome-scale metabolic models of liver and CD4+ T cells using the pipeline for MetAbolic Drug Repurposing IDentification-MADRID and integrating our data analysis results from RNA-seq, proteomics, and microarray human sample datasets (step 1). Drug targets obtained from the ConnectivityMap database are mapped to metabolic genes in the model where systematic knockouts are performed on each gene which leads to identifying differential fluxes (step 2). Our human-cell-specific models identified novel immunosuppressive drug targets by integrating differentially expressed genes in post-liver transplanted non-tolerant versus healthy control individuals and differential fluxes (step 3). We identified 193, 217, and 196 potential metabolic gene targets for T-helper 1, 2, and 17, respectively, against patients with a non-tolerant phenotype after liver transplantation. This genome-scale metabolic modeling approach for liver transplantation can assist in personalized treatment by using any individual’s sample data at step 1 and 2. It can also save time and resources in preclinical drug development studies by predicting more precise drug targets and drug repurposing based on patients’ omics data. By using this approach, we aim to improve the success rates of liver transplantations and reduce the complications associated with organ rejection. The work was supported by University of Nebraska collaboration initiatives.
Read full abstract