Snowflake, albinism and conservation

Abstract

“the molecular explanation of his rare genetic condition remained elusive” The genetic basis for Snowflake's albinism intrigued many of us, but the molecular explanation of his rare genetic condition remained elusive for many years. At first, no nucleotide changes were found in the coding region of the tyrosinase gene (TYR), whose mutations are associated with the commonest type of albinism in humans, OCA1, even though the complete lack of tyrosinase activity was also reported (Martínez-Arias et al., 2000). In 2004, my laboratory at CNB-CSIC was granted access to the fixed eyes of Snowflake, in paraffin blocks, and the subsequent histological analysis suggested the presences of traces of pigmentation in the retina. At the time, we and others had described the presence of relevant regulatory elements within the 5′ region of the human TYR gene, homologous to those found in the mouse Tyr locus (Regales et al., 2003). Therefore, we hypothesized that Snowflake could be a OCA1B case, where some residual TYR activity is observed, and where mutations in non-coding regions could explain its albinism. However, despite identifying numerous DNA sequence heterogeneities and heterozygosities in the 5′ region of the gorilla TYR locus, compared with homologous sequences from other, pigmented gorillas, we did not find any obvious mutation that was likely to be responsible for the albino phenotype of Snowflake (Roy et al., 2004). Almost 10 yr later, a large collaborative team led by Tomas Marques-Bonet, at UPF in Barcelona, has eventually deciphered the genetic cause of the oculocutaneous albinism in Snowflake through an impressive massive DNA sequencing effort. They applied a whole genome sequencing approach and aligned Snowflake's genome with the human reference genome to describe more than 73 000 homozygous non-synonymous species differences. Of those, 20 appeared to be located within genes associated with OCA and one of them mapped within the last exon of the SLC45A2 locus, resulting in an amino acid exchange (p.G518R) in one of the twelve transmembrane regions of this protein, predicted to function as a solute carrier. Mutations in the SLC45A2 gene (previously known as MATP, for membrane-associated transport protein) are associated with oculocutaneous albinism type 4 (OCA4) (Newton et al., 2001), one of the most common types of albinism in Japan but normally rare elsewhere (Mártinez-García and Montoliu, 2013). The authors of this interesting and conclusive report on the albinism of Snowflake went on in their analyses of the SLC45A2 mutation. To date, the homologous mutation p.G518R has not been reported in humans, nor its association with OCA4 (The Human Gene Mutation Database, HGMD 2013.1). However, this Glycine 518 residue appears to be conserved through all available vertebrate genomes investigated, hence suggesting a relevant functional role at this position, plus it is true that other glycine-to-arginine amino acid changes in the SLC45A2 protein have been already associated with OCA4 (data from HGMD 2013.1: p.G44R; p.G89R; p.G110R; p.G349R; p.G370R; p.G404R). Some indirect evidence of an altered association of the mutant protein with membranes was also reported by the authors. However, it is probably expected, for a protein with 12 transmembrane domains, that it would remain associated with membranes, even in the presence of some disruptive mutations, although these minor topological alterations might actually result in a loss-of-function phenotype. The ultimate proof should be provided by engineering this mutation in an experimental animal model (i.e. mouse or zebrafish) and/or, simply, by reporting its presence in human subjects diagnosed as OCA4. Finally, the authors investigated the landscape where this SLC45A2 mutation was found in Snowflake's genome and showed that it is nicely located within a large run (40 Mbps) of homozygosity, orthologous to human chromosome 5 (where the human SLC45A2 gene is located), indicating this allele was inside a block identical by descent, characteristic for Mendelian recessive disorders. Furthermore, by investigating the patterns of heterozygosity in the Snowflake's genome, the authors inferred that it was likely a result of inbreeding. Through computational simulations they concluded that the most probable scenarios for Snowflake's parents were uncle/niece or aunt/nephew.