Exome sequencing identifies SLCO2A1 mutations as a cause of primary hypertrophic osteoarthropathy

Abstract

By using whole-exome sequencing, we identified a homozygous guanine-to-adenine transition at the invariant -1 position of the acceptor site of intron 1 (c.97-1G>A) in solute carrier organic anion transporter family member 2A1 (SLCO2A1), which encodes a prostaglandin transporter protein, as the causative mutation in a single individual with primary hypertrophic osteoarthropathy (PHO) from a consanguineous family. In two other affected individuals with PHO from two unrelated nonconsanguineous families, we identified two different compound heterozygous mutations by using Sanger sequencing. These findings confirm that SLCO2A1 mutations inactivate prostaglandin E(2) (PGE(2)) transport, and they indicate that mutations in SLCO2A1 are the pathogenic cause of PHO. Moreover, this study might also help to explain the cause of secondary hypertrophic osteoarthropathy.

Figure 1

The Pedigrees of the Chinese Families Affected by PHO Black symbols represent the affected individuals, open symbols represent the unaffected individuals, and the half-blackened symbols represent asymptomatic mutation carriers. Circles and squares indicate females and males, respectively. The arrows identify the probands in the families. Double lines indicate consanguineous marriages. All living individuals were the individuals available for genotyping in families 1, 2, and 3.

Figure 2

Clinical Images of the Affected Individual, Family1-P1 The images show the thickening and furrowing of facial skin (A) and the clubbing of fingernails and toenails (B and C). Hand radiographs show a loss of the normal tabulation of metacarpals and phalanges and cortical thickening of the metacarpals and the proximal and middle phalanges (D and E). A radiograph of the feet shows cortical thickening and acroosteolysis (F). A radiograph of a knee shows periosteal hyperostosis of the knee region and shows patellae sclerosis and sclerosis of both the distal femur and tibiofibula (G). All images are published with permission from the affected individual.

Figure 3

Locating and Sequencing the SLCO2A1 Mutations (A) The location of the mutations identified in the SLCO2A1 gene. Family1-P1 (P1) had a homozygous mutation, and the father (F) and mother (M) were both heterozygous carriers. Family2-P2 (P2) and family3-P3 (P3) had compound heterozygous mutations, and their fathers and mothers were heterozygous carriers. (B) Both the p.Gly222Arg and p.Gly255Glu heterozygous missense mutations occur at a highly conserved position in SLCO2A1, as shown by a comparison of the corresponding sequences of ten vertebrates. The bases that are identical to those in Homo sapiens are highlighted in blue. Abbreviations are as follows: Homo, Homo sapiens; Xenopus, Xenopus laevis; Macaca, Macaca mulatta; Canis, Canis lupus familiaris; Sus, Sus scrofa; Bos, Bos taurus; Rattus, Rattus norvegicus; Mus, Mus musculus; Gallus, Gallus gallus; and Danio, Danio rerio. (C) Two missense mutations (p.Gly222Arg and p.Gly255Glu) in the transmembrane model of SLCO2A1. The mutations are indicated by arrows.

Figure 4

Protein Modeling of the SLCO2A1 p.Gly222Arg Mutation (A) The wild-type structure of SLCO2A1. H1, H5, and H6 represent three of the twelve helices in the protein structure. Two amino acids of interest, Gly222 and Gly255, are labeled and shown in the Corey, Pauling, Koltun (CPK) model. The mesh grid surface and the CPK molecule represent docked PGE₂ in the middle of the structure. (B and C) The effect of the p.Gly222Arg mutant on the interaction between H1 and H5. Gly222 of H5 and the neighboring portion of H1 are shown in green and blue, respectively. The shortest distances between these two parts of H1 and H5 are 1.74 Å and 2.82 Å at the labeled positions in the p.Gly222Arg mutant and wild-type proteins, respectively. The symbols Φ and ψ represent the dihedral angles of Tyr48 in H1.