Mutations in the novel mitochondrial protein REEP1 cause hereditary spastic paraplegia type 31

Abstract

Hereditary spastic paraplegia (HSP) comprises a group of clinically and genetically heterogeneous diseases that affect the upper motor neurons and their axonal projections. For the novel SPG31 locus on chromosome 2p12, we identified six different mutations in the receptor expression-enhancing protein 1 gene (REEP1). REEP1 mutations occurred in 6.5% of the patients with HSP in our sample, making it the third-most common HSP gene. We show that REEP1 is widely expressed and localizes to mitochondria, which underlines the importance of mitochondrial function in neurodegenerative disease.

Figure 1.

Overview of the SPG31 locus and identified genetic variation in REEP1. A, Within the minimal chromosomal SPG31 region, we screened several genes by direct sequencing analysis. B, REEP1 consists of seven exons (black boxes) and contains two predicted transmembrane domains (TM1 and TM2) and a “deleted in polyposis” domain (TB2/DP1/HVA22). The 3′ UTR comprises a conserved miRNA target site for miR-140 (MT). C, Sequencing traces of the identified unique mutations in six families with uncomplicated HSP. Sequencing analysis was performed using an ABI3730, following standard procedures. PCR primers are available from the authors on request. D, The miR-140 target site in the 3′ UTR is highly conserved. Two mutations within that site change nucleotides that provide G:U wobble base pairing: c.606+43G→U and c.606+50G→A (gray shading, blue letters, and arrows). E, The mature miR-140 gene is also highly conserved, including the two residues that would bind to the mutated nucleotides in the target site (gray shading and arrows). cen = centromere; tel = telomere.

Figure 2.

Different mutations identified in six unrelated HSP-affected families: DUK2369, DUK2354, DUK1959, DUK2036, DUK2189, and DUK2299. All families showed an uncomplicated HSP phenotype. The pedigrees present affected (blackened symbols) and unaffected (unblackened symbols) subjects. The clinical phenotypes were fully penetrant. All tested individuals are indicated (+/), as are mutations (/+) and wild-type genotypes (/−).

Figure 3.

REEP1 localized to mitochondria. A–F, Immunohistochemistry with two different REEP1 antibodies revealed colocalization with the marker Mitotracker Red. COS7 (kidney) and MN-1 (motor neuron) cells were maintained in Dulbecco’s modified Eagle medium with 10% fetal bovine serum and were incubated at 37°C supplemented with 5% CO₂. Cells were observed by confocal laser scan microscopy (Visitech model VT-Infinity). G and H, No colocalization was observed with microtubules and Golgi. I, COS7 cells were homogenized and mitochondria were separated from the cytosolic fraction by centrifugation. The REEP1 antibodies were probed against COS7 cell lysate on a western blot and showed a band of the expected size in the mitochondrial fraction. The blot was reprobed with porin, which constitutes a mitochondrial membrane protein, and β-actin. Note that the cytosolic fraction was highly enriched, as shown by the β-actin bands. J, RT-PCR on a series of tissues revealed ubiquitous expression. REEP1 was not expressed in peripheral blood and fibroblasts derived from a skin biopsy.

Figure 4.

Antibody design for REEP1. The proteins REEP1 through REEP4 are fairly conserved in human; thus, we chose two peptides in the less-conserved C terminal of REEP1 to be targeted by two polyclonal antibodies (frames). Two rabbits were injected with the synthetic peptides. Blood was collected after 50 d and was purified through keyhole-limpet-hemocyanin columns. All peptide synthesis work and antibody production was performed at Bethyl Laboratories.

Figure 5.

REEP1 is highly conserved throughout different species. Detected mutations in patients with HSP are highlighted in gray. Both transmembrane domains and the TB2/DP1/HVA22 domain are affected by two of the identified mutations. Stars highlight conserved residues.