A human infertility-associated KASH5 variant promotes mitochondrial localization

KASH5 is the most recently identified member of the KASH domain family of tail anchored, outer nuclear membrane (ONM) and endoplasmic reticulum (ER) proteins. During meiosis prophase I, KASH5 and SUN1 form a complex that spans the nuclear envelope and which links the telomeres of meiotic chromosomes to cytoplasmic dynein. This connection is essential for homologous chromosome dynamics and pairing. A recent study identified a variant in human KASH5 (L535Q) that correlated with male infertility associated with azoospermia. However, no molecular mechanism was described. Here, we report that this amino acid substitution, within the KASH5 transmembrane domain (TMD) has no predicted effects on secondary structure. However, the overall hydrophobicity of the L535Q TMD, is calculated to be lower than the wild-type KASH5, based on the GES (Goldman–Engelman–Steitz) amino acid hydrophobicity scale. This change in hydrophobicity profoundly affects the subcellular localization of KASH5. Through a series of amino acid substitution studies, we show that the L535Q substitution perturbs KASH5 localization to the ER and ONM and instead results in mistargeting to the mitochondria membrane. We suggest that this mislocalization accounts for the infertility and azoospermia phenotype in patients.

F1000Research TMATCH: A New Algorithm for Protein Alignments using amino-acid hydrophobicities

The identification of proteins of similar structure using sequence alignment is an important problem in bioinformatics. We decribe TMATCH, a basic dynamic programming alignment algorithm which can rapidly identify proteins of similar structure from a database. TMATCH was developed to utilize an optimal hydrophobicity metric for alignments traceable to fundamental properties of amino-acids. Standard alignment algorithms use affine gap penalties as contrasted with the TMATCH algorithm adaptation of local alignment score reinforcement of favorable diagonal paths (transitions) and punishment of unfavorable transitions paired with fixed gap opening penalties. The TMATCH algorithm is especially designed to take advantage of the extra information available within the hydrophobicity scale to detect homologies, as opposed to the probabilities derived from raw percent identities.