Molecular profiles of allograft rejection following inhibition of CD40 ligand costimulation differentiated by cluster analysis

Abstract

Recent technological advances in biomedical research, such as genomesequences and DNA microarrays, have dramatically increased the size of relevant databases. A major challenge is the extraction of a limitednumber of parameters from these databases that can differentiate anddiagnose complex biological states. In a model of cardiactransplantation investigating immunosuppression by inhibition of CD40ligand costimulation, we have applied a combination of clusteralgorithms and self-organizing maps to analyze a panel of 60 candidategenes. Dendrograms generated by cluster analysis distinguisheddifferent molecular bases of rejection. Using self-organizing maps, weidentified nine genes (CD4, CCR3, CCR5, LTβ, MIP-1α, MIP-2, CD8α,IP-10, and RANTES), each with a unique profile of transcriptionalexpression, that reproduce the differentiation of states of rejectionin dendrograms. Using histology and immunohistochemistry, we correlateddifferential regulation of CD4 and CD8 at the levels of mRNA andprotein. Our strategy of data reduction successfully decreased thenumber of genes to nine, which are sufficient to differentiate distinctstates of rejection in our experimental protocol.