Functional MAPT haplotypes: bridging the gap between genotype and neuropathology

Abstract

The microtubule-associated protein tau (MAPT) locus has long been associated with sporadic neurodegenerative disease, notably progressive supranuclear palsy and corticobasal degeneration, and more recently with Alzheimer's disease and Parkinson's disease. However, the functional biological mechanisms behind the genetic association have only now started to emerge. The genomic architecture in the region spanning MAPT is highly complex, and includes a approximately 1.8 Mb block of linkage disequilibrium (LD). The region is divided into two major haplotypes, H1 and H2, defined by numerous single nucleotide polymorphisms and a 900 kb inversion which suppresses recombination. Fine mapping of the MAPT region has identified sub-clades of the MAPT H1 haplotype which are specifically associated with neurodegenerative disease. Here we briefly review the role of MAPT in sporadic and familial neurodegenerative disease, and then discuss recent work which, for the first time, proposes functional mechanisms to link MAPT haplotypes with the neuropathology seen in patients.

Figure 1

MAPT locus consists of 16 exons of which exons 2, 3 and 10 are alternatively spliced in the adult CNS (grey or black stripes). Top: Exons 2 and 3 code for the N-terminal projection domain (black stripes). Exons 9 −12 imperfect repeats each coding for a microtubule-binding domain (grey). Exons 4A and 8 (white) are absent from the CNS, but exon 4A is expressed in the peripheral nervous system. Bottom: Six tau isoforms are expressed in the adult CNS. Alternative splicing of exons 2 and 3 results in proteins with 0, 1 or 2 N-terminal inserts (ON, 1N, 2N). Splicing of exon 10 generates proteins with either 3 or 4 microtubule-binding repeats (3R or 4R tau protein).

Figure 2

Cis-elements affecting splicing of MAPT exon 10. Silencer (red) and enhancer (green) sequences within and surrounding exon 10 are indicated. Exon 10 sequence is designated by capital letters and intronic sequence is indicated by small case letters. (A) Several cis-elements affect MAPT exon 10 splicing, listed 5′ to 3′: a SC35-like enhancer, a polypurine enhancer (PPE), an A/C-rich enhancer (ACE), an exonic splicing silencer (ESS), an exonic splicing enhancer (ESE), an intronic splicing silencer (ISS) and an intronic splicing modulator (ISM). Select FTDP-17 and PSP mutations are also shown. (B) Alternative splicing regulation at the tau exon 10 5′ splice site. Two of the variant stem-loop structures proposed to regulate splicing at the tau exon 5′ splice site (Hutton et al. 1998; Spillantini et al. 1998). (C) The linear model of tau exon 10 5′ splicing regulation postulates that binding of trans-acting factors to the cis-regulatory elements mediates exon 10 alternative splicing. Shown here is a trans-acting factor bound to the ISM. This ISM binding protein sterically hinders a trans-acting silencer (ISS binding protein) from binding the silencing cis-element, thereby allowing access of the U1 snRNP to the 5′ splice site (D’Souza and Schellenberg 2002; D’Souza and Schellenberg 2005).

Figure 3

MAPT Haplotypes. (A) The MAPT locus has been divided into two major haplotypes, H1 and H2. MAPT falls within the largest known block of LD in the human genome, spanning ~1.8 Mb. There is a 900 kb inversion of the H2 haplotype with respect to the H1 haplotype, covering a region containing MAPT, IMP5, CRHR1 and NSF. (B) Association of the dinucleotide marker A0 in MAPT intron 9 to PSP was the first genetic link between the disease and tau. This association was extended to cover the entire locus, using a 238 bp deletion (+/− 238 bp) and 8 SNPs (SNP 1 – SNP 13) to tag the haplotypes (Baker et al. 1999). Sub-haplotypes of H1 have since been identified using several haplotype tagging SNPs (rs1467967, rs24557, rs3875883, rs2471738, rs7521) (Pittman et al. 2004).