Examining the genetic basis for visceral leishmaniasis disease: Comparison of distinct Leishmania donovani isolates that cause cutaneous and visceral leishmaniasis in Sri Lanka

Among the most central questions in Leishmania research is why some species remain in the skin dermis at the site of infection by the sand fly vector whereas other species migrate to visceral organs where they cause fatal visceral leishmaniasis. Although L. donovani is the species typically responsible for visceral leishmaniasis, an atypical L. donovani strain is the etiologic agent for cutaneous leishmaniasis in Sri Lanka. To identify molecular determinants for visceral disease, we have analysed the phenotype and genotype of two L. donovani clinical isolates from Sri Lanka where one isolate was derived from a cutaneous leishmaniasis patient (CL) and the other from a visceral leishmaniasis patient (VL). These isolates cause dramatically different pathology when introduced into mice; notably the CL isolate has lost the ability to survive in visceral organs while the VL isolate was highly virulent in visceral organs of BALB/c mice. Whole genome sequencing of the CL and VL isolates revealed that these genomes were very similar as there were no gene deletions and few individual gene amplifications. Indels resulting in frame shifts and loss/gain of stop codons resulted in 13 distinct pseudogenes present in each of the CL and VL isolates. There were 154 non-synonymous SNPs specific to the CL isolate and 193 non-synonymous SNPs specific to the VL isolate. Genome wide gene expression analysis revealed several transcript level differences, including the A2 virulence gene resulting in higher expression of A2 proteins in the VL isolate than in the CL isolate. Genotypic variations relevant to pathology and tropism in Leishmania can be interrogated by reverse genetics. Experimentally increasing A2 expression in the CL isolate through gene transfer significantly increased it’s ability to survive in the spleen of BALB/c mice and conversely, down-regulating A2 expression in the VL isolate abrogated attenuated its survival in BALB/c mice. These observations reveal that there are relatively few genetic differences between the CL and VL isolates apart from the A2 genes, but collectively these have profound effects on human disease and experimentally infected mice. 6 Samples in total, 3 each from VL and CL causing isolates were analyzed by Splice Leader RNASeq. These three samples from each of the isolates were grown to form one of the following three lifestages, Promastigotes, Macrophage derived Amastigotes, Axenic Amastigotes.