Evolution of the CD163 family and its relationship to the bovine gamma delta T cell co-receptor WC1

Abstract

Background: The scavenger receptor cysteine rich (SRCR) domain is an ancient and conserved protein domain. CD163 and WC1 molecules are classed together as group B SRCR superfamily members, along with Spalpha, CD5 and CD6, all of which are expressed by immune system cells. There are three known types of CD163 molecules in mammals, CD163A (M130, coded for by CD163), CD163b (M160, coded for by CD163L1) and CD163c-alpha (CD163L1 or SCART), while their nearest relative, WC1, is encoded by a multigene family so far identified in the artiodactyl species of cattle, sheep, and pigs.

Results: We annotated the bovine genome and identified genes coding for bovine CD163A and CD163c-alpha but found no evidence for CD163b. Bovine CD163A is widely expressed in immune cells, whereas CD163c-alpha transcripts are enriched in the WC1+ gammadelta T cell population. Phylogenetic analyses of the CD163 family genes and WC1 showed that CD163c-alpha is most closely related to WC1 and that chicken and platypus have WC1 orthologous genes, previously classified as among their CD163 genes.

Conclusion: Since it has been shown that WC1 plays an important role in the regulation of gammadelta T cell responses in cattle, which, like chickens, have a high percentage of gammadelta T cells in their peripheral blood, CD163c-alpha may play a similar role, especially in species lacking WC1 genes. Our results suggest that gene duplications resulted in the expansion of CD163c-alpha-like and WC1-like molecules. This expanded repertoire was retained by species known as "gammadelta T cell high", but homologous SRCR molecules were maintained by all mammals.

Figure 1

Schematic representation of bovine CD163 exon-intron structure. Exon-intron structures of (A) CD163A and (B) CD163c-α were determined based on annotation of the CD163 genes in the bovine genome assembly Btau_3.1. Proposed placement of the exon encoding CD163c-α transmembrane region is shown in grey and could not be confirmed due to a gap in the genomic sequence assembly. SRCR domain numbers are indicated by roman numerals. Scale is shown in base pair increments beneath the schematic. Abbreviations are as follows: ID, interdomain sequence; TM, transmembrane region; ICD, intracytoplasmic domain.

Figure 2

Amino acid sequences of bovine CD163. Translated amino acid sequence, based on cDNA sequences for (A) CD163A and based on annotated and determined cDNA (shaded) sequences for (B) CD163c-α. Individual SRCR domains, identified by comparison to consensus group B SRCR domain sequence, are underlined, domain numbers are indicated by roman numerals, and cysteines are boxed. Putative transmembrane regions (TM) were determined by the DAS transmembrane prediction server and are also underlined [70].

Figure 3

Expression profiles of bovine CD163 genes. (A) Bovine ex vivo PBMC (lane 1), ConA-activated PBMC (lane 2), mesenteric lymph node (MLN, lane 3), lung (lane 4) and intestinal epithelial lymphocytes (IELs, lane 5) were evaluated for expression of CD163A and CD163c-α by RT-PCR. GAPDH is shown for comparison. These results are representative of reactions (n = 3) performed for at least two animals. (B) Bovine total PBMC, γδ TCR⁺ cells, γδ TCR ^- cells, WC1⁺ cells, and WC1^- cells were evaluated for CD163A, CD163c-α, and GAPDH expression by RT-PCR. Templates were serially diluted by a factor of three. These results are representative of two reactions.

Figure 4

CD163 family member SRCR domain organization and cytoplasmic tyrosine-based motifs. >(A) CD163 family members from multiple species were identified by BLAST searches with bovine WC1, CD163A and CD163c-α SRCR and cytoplasmic domains. The evolutionary history of 242 taxa was inferred, using MrBayes3.2 to approximate the posterior probabilities of trees, shown at branch nodes [65]. SRCR domains clustering in a clade with pre-defined SRCR domains (asterisks) were identified and SRCR domains clustered together, but not in association with known SRCR domains, were assigned a new letter designation [1]. Rattus norvegicus (Rn) CD163c-α, Pan troglodytes (Pt) CD163c-α, Canis lupus familiaris (Clf) CD163c_1 and ClfCD163c_2, Gallus gallus (Gg) CD163_1 through GgCD163_18, and Ornithorhynchus anatinus (Ora) CD163_1 through OraCD163_10 are predicted from genomic sequence; Bos taurus (Bt) WC1, Sus scrofa (Ss) WC1, Homo sapiens (Hs) CD163c-α (partial cDNA, corrected from genomic sequence), Mus musculus (Mm) SCART1, MmSCART2, HsCD163A and HsCD163b cDNA have been isolated. (B) Summary of CD163 family member SRCR domain organization and cytoplasmic tyrosine based motifs. SRCR domain assignments were made as shown in Fig. 4A, with confirmation by analysis utilizing the Neighbor-Joining algorithm with the JTT model (data not shown). CD163 family members with transmembrane domains are shown with the plasma membrane and location of the transmembrane domains represented by the horizontal line. Tyrosine-based signalling motifs similar to that found in BtWC1 are shown [35]. CD163 family members without transmembrane domains are shown horizontally, with the N-terminus on the left.

Figure 5

WC1 and CD163cα tyrosine-based motifs in the cytoplasmic domain. (A) Cytoplasmic domains were determined by ascertaining the transmembrane domain, using the DAS transmembrane server [70]. Proteins are labelled as described in Fig. 4, with the addition of bovine WC1 cytoplasmic domains [31]. The evolutionary history of 34 taxa was inferred using Bayesian analysis in MrBayes3.2 [65]. Markov chain Monte Carlo analysis was performed for 830,000 cycles, using 2 runs of 4 chains each, a temperature setting of 0.2, and an amino acid mixed model to approximate the posterior probabilities of trees, shown at branch nodes. The average standard deviation of split frequencies was 0.01, which was diagnostic of convergence at < 0.05. (B) Multiple alignment of the amino acids surrounding the tyrosine motif phosphorylated in WC1 (bold) with similar amino acid sequences from CD163 family members [35].

Figure 6

CD163 family members in multiple species. Numbers of group B SRCR CD163 family members in a representative of the sauropsid lineage, a prototherian mammal, and eutherian mammals are shown. The number of group A SRCR domain molecules in the echinoderm Strongylocentrotus purpuratus are shown [71]. Echinoderms diverged from chordates 595 million years ago; other divergence times are described on the Tree of Life web project http://tolweb.org[72].