Extensive variation in the intelectin gene family in laboratory and wild mouse strains

Abstract

Intelectins are a family of multimeric secreted proteins that bind microbe-specific glycans. Both genetic and functional studies have suggested that intelectins have an important role in innate immunity and are involved in the etiology of various human diseases, including inflammatory bowel disease. Experiments investigating the role of intelectins in human disease using mouse models are limited by the fact that there is not a clear one-to-one relationship between intelectin genes in humans and mice, and that the number of intelectin genes varies between different mouse strains. In this study we show by gene sequence and gene expression analysis that human intelectin-1 (ITLN1) has multiple orthologues in mice, including a functional homologue Itln1; however, human intelectin-2 has no such orthologue or homologue. We confirm that all sub-strains of the C57 mouse strain have a large deletion resulting in retention of only one intelectin gene, Itln1. The majority of laboratory strains have a full complement of six intelectin genes, except CAST, SPRET, SKIVE, MOLF and PANCEVO strains, which are derived from different mouse species/subspecies and encode different complements of intelectin genes. In wild mice, intelectin deletions are polymorphic in Mus musculus castaneus and Mus musculus domesticus. Further sequence analysis shows that Itln3 and Itln5 are polymorphic pseudogenes due to premature truncating mutations, and that mouse Itln1 has undergone recent adaptive evolution. Taken together, our study shows extensive diversity in intelectin genes in both laboratory and wild-mice, suggesting a pattern of birth-and-death evolution. In addition, our data provide a foundation for further experimental investigation of the role of intelectins in disease.

Figure 1

Intelectin genes in rodents and primates. (a) A phylogenetic tree based on amino acid sequences is drawn to scale, with branch lengths measured in the number of substitutions per site. Numbers at each node indicate bootstrap support for that node. (b) Amino acid sequence alignment of the N-terminal region of intelectin.

Figure 2

Analysis of intelectin expression in mice. mRNA expression levels measured by RT-qPCR are shown across tissues for (a) Itln1 in C57BL/6NCRL and (b) all intelectins in 129S2/SvPasCRL. Absolute quantification of mouse Itln mRNA transcript counts from tissue samples (two technical duplicates for each of four mice) determined from standard curves using a sequence specific plasmid and presented as transcripts per 10 ng total RNA. Note that qPCR primers were designed to amplify all six intelectin transcripts from 129S2/SvPasCRL with equal efficiency. Error bars represent standard error of the mean of results from four mice.

Figure 3

Analysis of intelectin expression in ileum and colon across different mouse strains. Quantification of mouse Itln mRNA levels in small intestinal ileum and colon tissue of mouse strains determined by RT-qPCR (two technical duplicates for each of four mice). Note that qPCR primers correspond to targets of identical sequence in Itln1 in C57BL/6NCRL and all six intelectin mRNA in 129S2/SvPasCRL. Data are normalised to Actb transcript levels. Error bars represent standard error of the mean (n = 4 mice). Because of high cost of some strains, specimens from a single mouse were analysed and data presented without error bars.

Figure 4

Paralog ratio tests (PRTs) to measure relative copy number of mouse intelectin genes. Four assays (A–D) designed to amplify different sized PCR products (green) using a single primer pair in each assay. Predicted amplicon lengths from the contig of the region previously generated from the 129S7 mouse strain.

Figure 5

Optical mapping of the intelectin region in laboratory strain mice. Optical mapping results from four strains (PWD, PWK, C57BL/6NCrl, CAST/EiJ) have been aligned against the reference genome (green bar, reference assembly, blue bar new strain assembly with depth of coverage indicated by the dark blue). The yellow and blue vertical lines in these bars represent NtBspQI restriction enzyme cutting sites or DLE-1 binding sites (CTTAAG motif). A blue vertical line indicates sites that are matching between the new assembly and the reference genome, yellow indicates that the site is there in the reference or the new assembly, but are considered non-matching. Regions showing structural variation, with respect to the C57BL6/J reference genome, are annotated as SV, green lines showing putative insertions and red lines showing putative deletions. An extra ~ 400 kb of sequence is in both PWK and PWD strains in the Itln1 region, consistent with the presence of the full complement of intelectin genes. Gene annotations are at the top and are taken from the UCSC Genome Browser.

Figure 6

Deletions found by sequence analysis of laboratory strain mice. The region covered by the original 129S7/Sv contig is shown with genes annotated. Extent of deletions found by sequence read depth analysis in laboratory strain mice shown below the size scale.

Figure 7

Comparison of the calcium- and glycan-binding regions of mouse and human intelectins. The aligned partial amino acid sequences of human and mice intelectins are shown (amino acid residue numbering from human ITLN1). Amino acids previously shown to mediate calcium coordination are shown in blue and those mediating glycan binding highlighted with yellow. Residues identical to those previously shown to bind carbohydrate in human ITLN1 are shown in black within the yellow background and the polymorphic amino acids are shown in red.