Genetic analysis of Leishmania donovani tropism using a naturally attenuated cutaneous strain

Abstract

A central question in Leishmania research is why most species cause cutaneous infections but others cause fatal visceral disease. Interestingly, L. donovani causes both visceral and cutaneous leishmaniasis in Sri Lanka. L. donovani clinical isolates were therefore obtained from cutaneous leishmaniasis (CL-SL) and visceral leishmaniasis (VL-SL) patients from Sri Lanka. The CL-SL isolate was severely attenuated compared to the VL-SL isolate for survival in visceral organs in BALB/c mice. Genomic and transcriptomic analysis argue that gene deletions or pseudogenes specific to CL-SL are not responsible for the difference in disease tropism and that single nucleotide polymorphisms (SNPs) and/or gene copy number variations play a major role in altered pathology. This is illustrated through the observations within showing that a decreased copy number of the A2 gene family and a mutation in the ras-like RagC GTPase enzyme in the mTOR pathway contribute to the attenuation of the CL-SL strain in visceral infection. Overall, this research provides a unique perspective on genetic differences associated with diverse pathologies caused by Leishmania infection.

Figure 1. The L. donovani CL-SL (CL) and VL-SL (VL) strains cause different pathology in BALB/c mice.

For visceral infection, BALB/c mice (5 mice/group) were infected in the tail vein with 5×10⁷ stationary phase promastigotes and 4 weeks after infection, mice were examined for spleen weight and parasite burden in the liver (Leishman Donovan Units: LDU) and spleen (parasite numbers) as determined by limiting dilution of spleen homogenates. The VL-SL strain (VL) infection in mice resulted in splenomegaly (Panel A) and high levels of infection in the liver (Panel B) and spleen (Panel C) compared to the CL-SL strain (CL) that caused negligible infection levels in the liver and spleen. Values plus standard error are shown. Panel D, cutaneous infections, BALB/c mice (5 mice/group) were injected subcutaneously in the rear footpad with 5×10⁶ stationary phase promastigotes and footpad lesion development was measured over 11 weeks.

Figure 2. Expression of VL-SL Rag C gene (366149) in the CL-SL isolate increased its survival in the spleen.

Genes from the VL-SL isolate were expressed in the CL-SL isolate, and spleen infection levels assessed in mice following intravenous infection. The VL-SL genes included are: LdBPK171200.1, steroid dehydrogenase; LdBPK220120, phosphoinositide phosphatase; LdBPK252290, DnaJ protein; LdBPK320370, hypothetical protein; LdBPK322160, rab-2a; LdBPK341900, NAD dependent deacetylase; LdBPK366149, Rag C (ras like GTPase C). Panel A. RT-PCR shows that all the corresponding VL-SL genes were transcribed in the transfected CL-SL L. donovani cells. Panel B. Infection levels in the spleen 4 weeks after infection. Values are mean plus standard error. Panel C. Alignments show that amino acid R231 in Rag C proteins is highly conserved in Leishmania and Trypanosomes.

Figure 3. Chromosome copy number comparison between the VL-SL and CL-SL strains (Panel A).

Read depth was scaled to give a value of 2 for disomic chromosomes. Comparison of chromosome transcript levels in the VL-SL and CL-SL strains (Panel B). Total RNA was prepared from promastigotes, axenic amastigotes and macrophages infected with the VL-SL and CL-SL strains. The cDNA libraries prepared from the RNAs were subjected to SL-sequencing and the cDNA coverage from each chromosome was normalized to the entire genome. Pro, axenic promastigotes; Ax, axenic amastigotes; Am, infected macrophage-derived amastigotes.

Figure 4. Involvement of A2 in visceral infection. Panel A.

Amplification in the region of chromosome 22 containing the A2 gene family. Normalized read coverage for each base position was obtained by dividing its raw read coverage by genome level median coverage (Grey line). Note that there are more A2 gene reads in the VL-SL strain (Red) than from the CL-SL strain (Green). Panel B. Comparison of A2 protein expression in the VL-SL and CL-SL strains. A2 expression was induced by 8 h heat shock at 40°C, and A2 proteins (Top panels) and tubulin (Bottom panels) were detected by Western blot. Panel C. Ectopically increased expression of A2 in CL-SL. The CL-SL strain was transfected with the pKSneo vector encoding an extra copy of the A2 gene (CL+A2) or the control empty pKSneo vector (CL). A2 proteins (upper panels) and tubulin loading control (lower panels) were detected by Western blot after 4 h at 40°C to induce A2 expression (transfected A2 indicated with an arrow). Panel D. Virulence of CL-SL with transfected A2 (CL+A2) was assessed four weeks following intravenous injection (1×10⁸ stationary phase promastigotes) in the tail vein of BALB/c mice (5 mice per group) and spleen parasite burden was determined by limiting dilution of spleen homogenates. Values plus standard error are displayed.