Complete assembly of the Leishmania donovani (HU3 strain) genome and transcriptome annotation

Abstract

Leishmania donovani is a unicellular parasite that causes visceral leishmaniasis, a fatal disease in humans. In this study, a complete assembly of the genome of L. donovani is provided. Apart from being the first published genome of this strain (HU3), this constitutes the best assembly for an L. donovani genome attained to date. The use of a combination of sequencing platforms enabled to assemble, without any sequence gap, the 36 chromosomes for this species. Additionally, based on this assembly and using RNA-seq reads derived from poly-A + RNA, the transcriptome for this species, not yet available, was delineated. Alternative SL addition sites and heterogeneity in the poly-A addition sites were commonly observed for most of the genes. After a complete annotation of the transcriptome, 2,410 novel transcripts were defined. Additionally, the relative expression for all transcripts present in the promastigote stage was determined. Events of cis-splicing have been documented to occur during the maturation of the transcripts derived from genes LDHU3_07.0430 and LDHU3_29.3990. The complete genome assembly and the availability of the gene models (including annotation of untranslated regions) are important pieces to understand how differential gene expression occurs in this pathogen, and to decipher phenotypic peculiarities like tissue tropism, clinical disease, and drug susceptibility.

Figure 1

Read-depth analyses along the chromosomes generated by the fusion of two PacBio-assembled contigs (panels A–E). Coverage (log2 scale) was determined by sliding window analysis (bin 200 bp) with either Illumina (in blue) or PacBio (in red) reads, along chromosomes 10, 12, 16, 27 and 35. The size and position of the contigs used are shown by lines with arrow-heads. Panel F, relative somy of the L. donovani (HU3) chromosomes. The somy estimation was performed using a 2-loop method. The median coverage of the genome is shown by a solid line, and it was assigned as 2, taking into account that diploid is considered the major ploidy status in Leishmania. The dotted lines indicated the estimated values for other somies. Graphs were generated from the median coverage values for each chromosome using the barplot function of R package (https://cran.r-project.org).

Figure 2

Comparison of tandemly repeated genes existing in the L. donovani (LdHU3) and the L. major (LmjF) genomes. Panel A, tandemly repeated genes in both species distributed according to the similarities/differences in the number of copies. Panel B, schematic representation of the loci coding for the elongation factor 1-alpha genes in the L. donovani and L. major genomes. The repeated copies are shown as green boxes in both species. The flanking ORFs are shown in red (ADP-ribosylation factor-like protein 1) and dark blue (receptor-type adenylate cyclase A). Panel C, genes encoding for a hypothetical protein (with a leucine-rich repeat domain) are shown as green boxes in both species. The flanking ORFs are shown in red (VIT family, putative) and dark blue (C-8 sterol isomerase-like protein).

Figure 3

Illustration of the process followed for correcting the automatic annotation of CDS based on the transcript definition. Panel A, mapping of RNA-seq reads in the genomic region expanding the LDHU3_05.0940 gene. Panel B, mapping of SL-containing RNA-seq reads (two SL-addition sites (SASs) were mapped, the main site was covered by 72 reads and the secondary –small arrow- by only 2). Panel C, LDHU3_05.T0940 transcript annotation based on the position of the main SAS and the polyadenylation site (not shown). Panel D, automatic CDS annotation generated by Companion. Panel E, manually corrected CDS annotation, after delimiting the transcript. Panel F, proposed gene model for the LDHU3_05.0940 gene.

Figure 4

Processing by cis-splicing of the transcript encoding for the poly-A polymerase in L. donovani. Panel A, gene model for LDHU3_29.3990; E1 and E2, exons; I, intron. The red arrow indicates the position of the main SAS and the alternative SASs are indicated by black arrows (the number of RNA-seq mapped to each SAS is shown in parentheses). Blue arrows point to the poly-A addition sites. SASs mapped in the intron sequence are indicated by arrows above it. The position of the primers used for the PCR amplification are shown (maroon arrows; forward: 5′-GCGAGTTTCT GAAGTGCTGC-3′; reverse: 5′-TTCAGCACTG GGAACAGGTC-3′). The distribution (coverage) of Illumina reads along the region in study obtained after mapping of either RNA-derived reads (coverage RNA-seq) or DNA-derived reads (coverage DNA-seq) are also shown. Panel B, electrophoresis of PCR products on a 1% agarose gel; lanes 1 and 3, PCR amplification using cDNA derived from L. donovani total RNA and using for retrotranscription either SuperScript III (lane 1) or ThermoScript (lane 3) retrotranscriptases; lanes 2 and 4, PCR amplification from L. donovani total RNA (without previous retrotranscription step); lane 5, PCR amplification from L. donovani total DNA. Relative migration and size of molecular weight markers (Φ29 DNA digested with HindIII) are shown on the left. Uncropped gel shown in Supplementary Information Fig. S12. Panel C, schematic representation of the exon-intron junctions as determined after sequencing of the RT-PCR amplicon. Conserved nucleotides (upper case) in the equivalent intron existing in the gene coding for poly-A polymerase in T. brucei. The positions of 5′ and 3′ splice sites (5′ ss and 3′ ss, respectively) are indicated.