The absence of C-5 DNA methylation in Leishmania donovani allows DNA enrichment from complex samples

Abstract

Cytosine C5 methylation is an important epigenetic control mechanism in a wide array of Eukaryotic organisms and generally carried out by proteins of C-5 DNA methyltransferase family (DNMTs). In several protozoans the status of this mechanism remains elusive, such as in Leishmania, the causative agent of the disease leishmaniasis in humans and a wide array of vertebrate animals. In this work, we show that the Leishmania donovani genome contains a C-5 DNA methyltransferase (DNMT) from the DNMT6 subfamily, of which the function is still unclear, and verified its expression at RNA level. We created viable overexpressor and knock-out lines of this enzyme and characterised their genome-wide methylation patterns using whole-genome bisulfite sequencing, together with promastigote and amastigote control lines. Interestingly, despite DNMT6 presence, we found that methylation levels were equal to or lower than 0.0003% at CpG sites, 0.0005% at CHG sites and 0.0126% at CHH sites at genome scale. As none of the methylated sites were retained after manual verification, we conclude that there is no evidence for DNA methylation in this species. We demonstrate that this difference in DNA methylation between the parasite (no detectable DNA methylation) and the vertebrate host (DNA methylation), allows enrichment of parasite versus host DNA using Methyl-CpG-binding domain columns, readily available in commercial kits. As such, we depleted methylated DNA from mixes of Leishmania promastigote and amastigote DNA with human DNA, resulting in average Leishmania:human enrichments from 62x up to 263x. These results open a promising avenue for unmethylated DNA enrichment as a pre-enrichment step before sequencing Leishmania clinical samples.

Genomes of Leishmania parasites directly sequenced from patients with visceral leishmaniasis in the Indian subcontinent

“…Although this approach of enrichment is promising, it requires large amounts of total DNA (1–2μg) and high content of pathogen DNA (10% for Plasmodium containing samples), precluding its application on many clinical samples. Moreover, at the beginning of our project, the methylation status of Leishmania DNA was unknown, therefore we did not consider this enrichment method: noteworthy, we recently preprinted a study showing the low methylation status of Leishmania [25], paving the way for exploring enrichment methods based on removal of methylated host DNA. The second alternative enrichment method, selective whole genome amplification (SWGA) [14], is based on amplification with phi29 using short oligonucleotides that preferentially bind to several sequences across the whole genome of a parasite [14, 26, 27].…”

Selective whole-genome amplification reveals population genetics of Leishmania braziliensis directly from patient skin biopsies

“…One open question is how the efficiency of the SWGA method can be improved so that a higher number of patient samples yield parasite genomes. Host-specific restriction enzymes [ 20 , 49 ] may offer one appealing solution for Leishmania , particularly since L . donovani reportedly lacks C-5 DNA methylation, potentially opening the doors to using methylation-sensitive restriction enzymes to preferentially degrade host DNA [ 50 ].…”

The model diatom Phaeodactylum tricornutum provides insights into the diversity and function of microeukaryotic DNA methyltransferases

“…As reported in trypanosomes 10 , we suggest that DNMT6 likely emerged prior to the Chromalveolata radiation. In trypanosomes, its presence in several lineages does not predict DNA methylation per se and must be further investigated 46 .…”

“…As it is reported in trypanosomes 10 , we suggest that DNMT6 likely emerged prior to the Chromalveolata radiation. It is the only known DNMT enzyme in Trypanosomes that does not show extensive DNA methylation pattern in any context 43 . DNMT6 is not active in Leishmania species 43 .…”

“…It is the only known DNMT enzyme in Trypanosomes that does not show extensive DNA methylation pattern in any context 43 . DNMT6 is not active in Leishmania species 43 . Its presence in several lineages, hence, does not predict DNA methylation per se and must be further investigated.…”