High subtelomeric GC content in the genome of a zoonotic Cryptosporidium species

Abstract

Cryptosporidium canis is a zoonotic species causing cryptosporidiosis in humans in addition to its natural hosts dogs and other fur animals. To understand the genetic basis for host adaptation, we sequenced the genomes of C. canis from dogs, minks, and foxes and conducted a comparative genomics analysis. While the genomes of C. canis have similar gene contents and organisations, they (~41.0 %) and C. felis (39.6 %) have GC content much higher than other Cryptosporidium spp. (24.3-32.9 %) sequenced to date. The high GC content is mostly restricted to subtelomeric regions of the eight chromosomes. Most of these GC-balanced genes encode Cryptosporidium-specific proteins that have intrinsically disordered regions and are involved in host-parasite interactions. Natural selection appears to play a more important role in the evolution of codon usage in GC-balanced C. canis, and most of the GC-balanced genes have undergone positive selection. While the identity in whole genome sequences between the mink- and dog-derived isolates is 99.9 % (9365 SNVs), it is only 96.0 % (362 894 SNVs) between them and the fox-derived isolate. In agreement with this, the fox-derived isolate possesses more subtelomeric genes encoding invasion-related protein families. Therefore, the change in subtelomeric GC content appears to be responsible for the more GC-balanced C. canis genomes, and the fox-derived isolate could represent a new Cryptosporidium species.

Keywords: Cryptosporidium canis; GC content; Whole genome sequencing; codon usage; host-adaptive evolution.

Fig. 1.

Shared orthologous genes and genome organisations among Cryptosporidium species. (a) Syntenic relationship of genomes among the intestinal species C. canis, C. parvum, and the gastric species C. andersoni. Syntenic blocks with orthologous genes between different genomes are linked with lines. (b) Visualization of orthologous genes and species-specific genes among five Cryptosporidium species by a UpSetR plot. The black dots connected by vertical lines in the UpSetR plot represent the orthologous groups shared by corresponding Cryptosporidium species, while the single black dots indicate the species-specific genes. The vertical bar charts in the plot indicate the number of each orthologous group, while the horizontal bar charts indicate the number of genes encoded by the genome of the corresponding Cryptosporidium species. The observed orthologs are subject to change as the genome sequences of the Cryptosporidium species are at different levels of completeness and annotations. (c) Protein-encoding genes and sequencing depth in the subtelomeric regions of chromosome 8 of the three Cryptosporidium canis genomes. The grey dotted rectangles from the alignment results of Mauve indicate the 5′ and 3′ end subtelomeric regions of chromosome 8. The fold coverages of reads mapped to the subtelomeric regions are shown by blue histograms at the top of the graphs. Each rectangle box at the bottom of each chromosome represents a subtelomeric gene. The SKSR-encoding gene (ortholog of cgd8_30 in C. parvum) present in the chromosome 8 of the fox-derived isolate but absent in the dog- and mink-derived isolates is shown in pink.

Fig. 2.

Distribution of GC content and usage of codons in the genomes of Cryptosporidium spp. (a) Heatmap of the genome-wide GC landscapes of Cryptosporidium canis and other species. The genomic GC content of C. canis and other Cryptosporidium species in a 2 kb sliding window is exhibited in a heatmap in the order of eight chromosomes (Chrs). (b) Distribution of GC content in the subtelomeric and internal genes of Cryptosporidium genomes showed by violin plots. Double asterisks (**) indicate significant differences at P<0.01. (c) Usages of A/T- and G/C-ending codons in Cryptosporidium species. A preferred codon is recognized when its RSCU value is greater than 1.0. Trinucleotides at the top of the species represent the preferred G/C-ending codons. (d) Distribution of Cryptosporidium species based on the principal component analysis of RSCU data. Cpar: C. parvum; Ctyz: C. tyzzeri; Chom: C. hominis; Cmel: C. meleagridis; Cchi: Cryptosporidium chipmunk genotype I; Cubi: C. ubiquitum; Ccan D45460: dog-derived C. canis D45460; Ccan M25894: mink-derived C. canis M25894; Ccan F33844: fox-derived C. canis F33844; Cfel: C. felis; Cbai: C. baileyi; Cbov: C. bovis; Crya: C. ryanae; Cmur, C. muris; Cand: C. andersoni.

Fig. 3.

GO-term enrichment and selective pressure of genes in three classes of GC content in Cryptosporidium spp. (a) Histogram of GO-term enrichment for the proteins encoded by genes in class 1 (GC %<30), class 2 (GC %=30–40), and class 3 (GC %>40). (b) Functional distribution of proteins encoded by genes in GO-term class 3 (GC %>40) of C. canis and C. felis. (c) Selective pressure (dN/dS) of genes in GO-term classes 1–3 (GC %>40) between C. canis or C. felis and C. parvum. A: Cryptosporidium parvum; B: C. tyzzeri; C: C. hominis; D: C. meleagridis; E: Cryptosporidium chipmunk genotype I; F: C. ubiquitum; G: dog-derived C. canis D45460; H: mink-derived C. canis M25894; I: fox-derived C. canis F33844; J: C. felis; K: C. baileyi; L: C. bovis; M: C. ryanae; N: C. muris; O: C. andersoni.

Fig. 4.

Usage of amino acids and intrinsically disordered proteins in Cryptosporidium spp. (a) Energy cost of protein biosynthesis in Cryptosporidium spp. The numbers on the colour scale represent the average consumption of ATP or GTP for amino acid biosynthesis of each protein in different Cryptosporidium species. (b) Amino acid frequency in Cryptosporidium spp. Trinucleotides at the top of the amino acid residues represent the synonymous codons for the corresponding amino acids. (c) Percentage of intrinsically disordered proteins in Cryptosporidium species with different genomic GC contents. (d) Boxplot of GC content in genes encoding proteins with and without intrinsically disordered regions in Cryptosporidium spp. Double asterisks (**) indicate significant differences at P<0.01. Cpar: Cryptosporidium parvum; Ctyz: C. tyzzeri; Chom: C. hominis; Cmel: C. meleagridis; Cchi: Cryptosporidium chipmunk genotype I; Cubi: C. ubiquitum; Ccan D45460: dog-derived C. canis D45460; Ccan M25894: mink-derived C. canis M25894; Ccan F33844: fox-derived C. canis F33844; Cfel: C. felis; Cbai: C. baileyi; Cbov: C. bovis; Crya: C. ryanae; Cmur, C. muris; Cand: C. andersoni.