High prevalence of SLC6A8 deficiency in X-linked mental retardation

Abstract

A novel X-linked mental retardation (XLMR) syndrome was recently identified, resulting from creatine deficiency in the brain caused by mutations in the creatine transporter gene, SLC6A8. We have studied the prevalence of SLC6A8 mutations in a panel of 290 patients with nonsyndromic XLMR archived by the European XLMR Consortium. The full-length open reading frame and splice sites of the SLC6A8 gene were investigated by DNA sequence analysis. Six pathogenic mutations, of which five were novel, were identified in a total of 288 patients with XLMR, showing a prevalence of at least 2.1% (6/288). The novel pathogenic mutations are a nonsense mutation (p.Y317X) and four missense mutations. Three missense mutations (p.G87R, p.P390L, and p.P554L) were concluded to be pathogenic on the basis of conservation, segregation, chemical properties of the residues involved, as well as the absence of these and any other missense mutation in 276 controls. For the p.C337W mutation, additional material was available to biochemically prove (i.e., by increased urinary creatine : creatinine ratio) pathogenicity. In addition, we found nine novel polymorphisms (IVS1+26G-->A, IVS7+37G-->A, IVS7+87A-->G, IVS7-35G-->A, IVS12-3C-->T, IVS2+88G-->C, IVS9-36G-->A, IVS12-82G-->C, and p.Y498) that were present in the XLMR panel and/or in the control panel. Two missense variants (p.V629I and p.M560V) that were not highly conserved and were not associated with increased creatine : creatinine ratio, one translational silent variant (p.L472), and 10 intervening sequence variants or untranslated region variants (IVS6+9C-->T, IVS7-151_152delGA, IVS7-99C-->A, IVS8-35G-->A, IVS8+28C-->T, IVS10-18C-->T, IVS11+21G-->A, IVS12+15C-->T, *207G-->C, IVS12+32C-->A) were found only in the XLMR panel but should be considered as unclassified variants or as a polymorphism (p.M560V). Our data indicate that the frequency of SLC6A8 mutations in the XLMR population is close to that of CGG expansions in FMR1, the gene responsible for fragile-X syndrome.

Figure 1

Mutations and polymorphisms in patients with XLMR. Primer sequences were designed specifically to amplify all exons of the SLC6A8 gene (and not the SLC6A10 gene, a presumed creatine transporter pseudogene mapped at chromosome 16 [Eichler et al. 1996]), including short fragments of the flanking intronic sequences. By comparing the SLC6A8 and SLC6A10 sequences, we have found that at least two nucleotide variations were present in all amplicons, confirming selective amplification of the SLC6A8 sequences (data not shown). PCR reactions were performed with HotStar Taq (Qiagen) in a PE Applied Biosystems model 9700. Direct sequence analysis was performed on purified PCR products (Millipore vacufold) by use of BigDye v3.1 terminators and an ABI 3100 sequence machine (PE Applied Biosystems). The obtained electropherograms were assembled and analyzed to identify potential genomic alterations by use of the Mutation Surveyor software package (SoftGenetics). Sequence variants were annotated according the guidelines of den Dunnen and Antonarakis (2001). On the basis of impaired uptake in fibroblasts, five alterations (asterisks [*]) have been proven elsewhere to be mutations. p.F107del was also found in our XLMR cohort. One novel nonsense mutation (p.Y317X) is strongly predictive of impaired creatine uptake. p.M560V is a rare polymorphism, and p.V629I is an unclassified variant. The implications of the translational silent variant, the IVS variant, and the 3′ UTR variant could not yet be investigated, and these variants are therefore assigned as “unclassified.” IVS variants in introns 1, 2, 3, and 4 may have been missed, since only small exon-flanking parts were included.

Figure 2

Multiple-sequence alignment with hierarchical clustering among the SLC6A8 proteins of different species and the superfamily of neurotransmitter transporters. The SLC6A8 protein sequence is shown for the following species: Homo sapiens (A8), Rattus norvegicus (RN), Mus musculus (MM), Bos taurus (BT), Oryctolagus cuniculus (OC), and Torpedo marmorata (TM). SLC6A8 genes were identified by a protein blast (BLASTP). Alignment was determined by the ClustalW program; the SLC6A8 proteins that were used for this analysis were the ones identified by the BLASTP search to be most related to the Homo sapiens SLC6A8 protein. All were saved in FASTA format. The BOXSHADE program was used to visualize identical amino acids (highlighted in black) and functionally conserved amino acids (gray). Functionally conserved amino acids are classified as follows: V, I, L, and M; D, E, Q, and N; F, Y, and W; G, S, T, P, and A; and K, R, and H. The codes A1–A14 represent neurotransmitter transporters SLC6A1–SLC6A14 (A1 = GABA1, A2 = noradrenaline, A3 = dopamine, A4 = serotonine, A5 = glycine2, A6 = taurine, A7 = proline, A9 = glycine 1, A11 = GABA3, A12 = betaine/GABA, A13 = GABA2, and A14 = ATB⁰⁺). In the protein sequence of A5, certain amino acids (20–46, 51–93, 116–127, 143–152, and 168–184) were cut out, because these stretches occur only in SLC6A5 and would make the figure disorganized. TMI–TMXII = putative transmembrane domain, predicted by ExPASy (Swiss-Prot S6A8_Human P48029); ♦ = putative cAMP-PK phosphorylation site (2×); ▴ = putative N-glycosylation sites (3×); ▪ = putative Leu zipper motif (4×).

Figure 2

Multiple-sequence alignment with hierarchical clustering among the SLC6A8 proteins of different species and the superfamily of neurotransmitter transporters. The SLC6A8 protein sequence is shown for the following species: Homo sapiens (A8), Rattus norvegicus (RN), Mus musculus (MM), Bos taurus (BT), Oryctolagus cuniculus (OC), and Torpedo marmorata (TM). SLC6A8 genes were identified by a protein blast (BLASTP). Alignment was determined by the ClustalW program; the SLC6A8 proteins that were used for this analysis were the ones identified by the BLASTP search to be most related to the Homo sapiens SLC6A8 protein. All were saved in FASTA format. The BOXSHADE program was used to visualize identical amino acids (highlighted in black) and functionally conserved amino acids (gray). Functionally conserved amino acids are classified as follows: V, I, L, and M; D, E, Q, and N; F, Y, and W; G, S, T, P, and A; and K, R, and H. The codes A1–A14 represent neurotransmitter transporters SLC6A1–SLC6A14 (A1 = GABA1, A2 = noradrenaline, A3 = dopamine, A4 = serotonine, A5 = glycine2, A6 = taurine, A7 = proline, A9 = glycine 1, A11 = GABA3, A12 = betaine/GABA, A13 = GABA2, and A14 = ATB⁰⁺). In the protein sequence of A5, certain amino acids (20–46, 51–93, 116–127, 143–152, and 168–184) were cut out, because these stretches occur only in SLC6A5 and would make the figure disorganized. TMI–TMXII = putative transmembrane domain, predicted by ExPASy (Swiss-Prot S6A8_Human P48029); ♦ = putative cAMP-PK phosphorylation site (2×); ▴ = putative N-glycosylation sites (3×); ▪ = putative Leu zipper motif (4×).

Figure 3

Pedigree charts of families in which segregation could be studied. (For designation of the variants, see table 1.)