Comparative genomics of Giardia duodenalis sub-assemblage AI beaver (Be-2) and human (WB-C6) strains show remarkable homozygosity, sequence similarity, and conservation of VSP genes

Abstract

Giardia duodenalis, a major cause of waterborne infection, infects a wide range of mammalian hosts and is subdivided into eight genetically well-defined assemblages named A through H. However, fragmented genomes and a lack of comparative analysis within and between the assemblages render unclear the molecular mechanisms controlling host specificity and differential disease outcomes. To address this, we generated a near-complete de novo genome of AI assemblage using the Oxford Nanopore platform by sequencing the Be-2 genome. We generated 148,144 long-reads with quality scores of > 7. The final genome assembly consists of only nine contigs with an N50 of 3,045,186 bp. This assembly agrees closely with the assembly of another strain in the AI assemblage (WB-C6). However, a critical difference is that a region previously placed in the five-prime region of Chr5 belongs to Chr4 of Be-2. We find a high degree of conservation in the ploidy, homozygosity, and the presence of cysteine-rich variant-specific surface proteins (VSPs) within the AI assemblage. Our assembly provides a nearly complete genome of a member of the AI assemblage of G. duodenalis, aiding population genomic studies capable of elucidating Giardia transmission, host range, and pathogenicity.

Keywords: Giardia; Annotation; Genome assembly; Long-read sequencing; Ploidy; Synteny.

Figure 1

Composition of the G. duodenalis AI Genotype. Circos representation of the five chromosomes of G. duodenalis AI genotype based on strain Be-2. The outermost track indicates the chromosomal sequences. UNSG represents the 4 unassigned scaffolds containing 24,631 bp. Inner tracks show GC content and chromosomal position of variant-specific surface proteins (VSPs) and tRNAs. The annotation was made in two steps using Liftoff v.1.6.3 and AUGUSTUS as mentioned in the Method section. The Circos plot was drawn with R library circlize.

Figure 2

Synteny map of G. duodenalis AI isolates. (A) Circular plot comparing the levels of synteny among the Giardia strains Be-2 (right) and WB-C6 (left). The outermost grey circle shows the chromosomal organization of the Be-2 genome and compares with annotated chromosomes of WB-C6 with colored bands and lines linking connected syntenic blocks. The dark pink color in WB-C6 Chr5 represents the gap in the assembly of the reference genome of G. intestinalis WB-C6. Each chromosome is labeled with a strain name followed by chromosome number. The Circos plot was drawn with R library circlize. (B) Long read alignment at Chr5 of G. intestinalis WB. White spaces at the five-prime region of the chr5 represent gaps with no physical evidence from long reads. Blue bar plots indicate coverage of ONT long-reads across the genome. Chromosomes are scaled in kilobase pairs. (C) Multiple whole-genome alignments were constructed using the Mauve plot of five long-read assembled Giardia genomes^,. Each horizontal line with homologous segments outlined as colored rectangles represents each genome separately. Inverted segments are set below those that match the forward orientation. Each connecting line represents the aligned blocks between genomes. White areas possibly contain genome-specific sequence elements and those genomic positions that did not adequately align between the selected genomes.

Figure 3

Comparative genomics of G. duodenalis AI isolates. (A) Maximum likelihood phylogenetic tree was conducted using all core accessory genes annotated in all analyzed isolates. The upper clade represents the assemblage A isolates, whereas the bottom cluster includes the assemblage B and E isolates. Assemblages are shown in brackets on the basis of previous designation. Assemblage (B) Comparative analysis of somy between Be-2 and WB-C6 indicates the 4N genome. Somies were calculated based on the average of block coverages which is scaled to the ploidy of the strains using AGELESS software (http://ageless.sourceforge.net/) in a rolling window of 2000 bp and averaging the coverage within each window. Somies were plotted using ggplot in R packages (v.4.1.0, URL: http://www.R-project.org). (C) Genome-wide zygosity profiles in Giardia strains Be-2 and WB-C6 identified long stretches of homozygosity. The total number of SNPs was calculated using a rolling window of 2000 bp. The blue, red, and yellow colors represent the homozygous SNPs (> 90%), heterozygous SNPs (> 90%), and both homozygous and heterozygous SNPs (50% each) respectively. (D) Bottlebrush plots showing the allelic inheritance pattern and loss of heterozygosity in the assemblage A isolates Be-2 and WB-C6. The X-axis represents the size of chromosomes (Mb), and the Y-axis shows allelic inheritance. Red and Green bars represent allele frequencies of Be-2 and WB-C6 respectively.

Figure 4

(A) The Venn diagram illustrates the number of shared and specific genes between Be-2 and WB-C6 based on clusters of orthologs. Horizontal bar plots illustrate the comparative analysis of protein sequences on the 43 and 18 orthogroup clusters representing WB-C6 and Be-2 unique orthogroups. The green color represents WB-C6 whereas the blue color represents Be-2. (B) VSP gene distribution on the assembled genome of Be-2 and WB. Red lines represent the VSPs present in Be-2 whereas blue lines indicate VSPs in WB. Chromosomes are scaled in kilobase pairs.