Fig 1.
An assembly of genomic sequencing reads from a red colobus blood sample contained significant amounts of sequence from the parasite Hepatocystis spp.
(A) Contigs from the red colobus (Piliocolobus tephrosceles) assembly had a bimodal distribution of AT-content and sequence similarity to Plasmodium spp. (B). A phylogenetic tree of cytochrome b indicated that the closest match for the apicomplexan parasite sequenced from red colobus blood is a Hepatocystis isolate from a monkey host. Parasite cytochrome b sequences derived from RNA-seq assemblies from red colobus blood samples are almost entirely identical to the cytochrome b sequence assembled from Hepatocystis DNA reads from a single monkey. Branches of the tree have been coloured by bootstrap support values from 15 (red) to 100 (green). Some bootstrap support values are also shown next to the nodes as text. Red arrows highlight the Hepatocystis samples from the current study. Blue place names indicate the African continent, green Australia, orange Asia.
Table 1.
Features of the Hepatocystis sp. ex. Piliocolobus tephrosceles assembly compared to P. falciparum 3D7, P. vivax P01 and P. berghei ANKA.
Fig 2.
Whole genome phylogeny and key features of the Hepatocystis genome.
A whole-genome phylogenetic tree is combined with a graphical overview of key features of Hepatocystis and Plasmodium species (genome versions from August 2019). The maximum likelihood phylogenetic tree of Hepatocystis, Plasmodium and Haemoproteus species is based on an amino acid alignment of 2673 single copy orthologs encoded by the nuclear genome. Bootstrap support values of all nodes were 100, except for one node where the value was 79. The rooting of the tree at Hae. tartakovskyi is based on previously published Plasmodium phylogenetic trees [11,45]. TRAP—thrombospondin-related anonymous protein. RBP protein—reticulocyte binding protein.
Fig 3.
Hepatocystis sp. in vivo RNA-seq data supports a lack of erythrocytic schizogony and a variable sex ratio.
(A) Distributions of SNPs per 100 kb in each Hepatocystis sp. RNA-seq sample, relative to the genome assembly reference, highlight consistently low genetic diversity. Samples SAMN07757853, SAMN07757863, SAMN07757870 and SAMN07757873 have been excluded from the figure due to their low expression of Hepatocystis genes. (B) Deconvolution of RNA-seq samples to identify parasite stage composition shows no evidence for blood schizonts. Ring and trophozoite cells are assumed to relate to early stages of gametocyte development, which are not distinguishable from asexual rings and trophozoites. (C) Proportions of early blood stages (ring and trophozoite) are negatively correlated with mature female gametocytes, however male and female gametocyte ratios are poorly correlated, suggesting that sex ratios vary among samples.
Table 2.
Size of known and novel gene families in Hepatocystis sp. in comparison to its relatives.
Fig 4.
Hepatocystis sp. orthologues of genes which are highly expressed in late schizogony in P. berghei tend to be missing from the genome.
P. berghei gene expression clusters from the Malaria Cell Atlas were used to determine whether orthologous genes, conserved with P. vivax and P. ovale, but absent in Hepatocystis, tended to be expressed in particular parts of the life cycle. The only significant cluster was cluster 10, which includes genes most highly expressed in late schizont stages (highlighted by a red box). The top panels show the log2 observed/expected ratios for orthologous genes in each cluster which are shared between P. berghei and Hepatocystis and the same ratio for those which are shared between P. berghei, P. vivax and P. ovale, but not Hepatocystis sp. Cluster 1, 18, 19 and 20 were not tested because they contained fewer than 2 expected counts for either ratio. The asterisk indicates a Fisher’s exact test false discovery rate of < = 0.05.
Table 3.
Top 15 genes with functional annotations ranked by Hepatocystis sp. dN in comparison of Hepatocystis sp., P. berghei ANKA and P. ovale curtisi.
Genes with completely unknown function and genes with very little information on their possible functions have been left out from this table. The rank column indicates the Hepatocystis sp. dN rank of each gene in the complete table (with 4009 genes) that includes genes with unknown function (S4 Table).