Retention at the cis-Golgi and delayed degradation of tissue-non-specific alkaline phosphatase with an Asn153→Asp substitution, a cause of perinatal hypophosphatasia

Abstract

Tissue-non-specific alkaline phosphatase (TNSALP) is an ectoenzyme anchored to the plasma membrane via glycosylphosphatidylinositol (GPI). A TNSALP mutant with an Asn153→Asp (N153D) substitution was reported in a foetus diagnosed with perinatal hypophosphatasia (Mornet, Taillandier, Peyramaure, Kaper, Muller, Brenner, Bussiere, Freisinger, Godard, Merrer et al. (1998) Eur. J. Hum. Genet. 6, 308–314). When expressed ectopically in COS-1 cells, the wild-type TNSALP formed active non-covalently associated dimers, whereas TNSALP (N153D) formed aberrant disulphide-bonded high-molecular-mass aggregates devoid of enzyme activity. Cell-surface biotinylation and digestion with phosphatidylinositol-specific phospholipase C showed that TNSALP (N153D) failed to reach the cell surface. Instead, double immunofluorescence demonstrated that TNSALP (N153D) partially co-localized with a cis-Golgi marker (GM-130) at the steady-state. Upon treatment with brefeldin A, TNSALP (N153D) was still co-localized with GM-130, further supporting the finding that this mutant is localized in the cis-Golgi. Consistent with morphological results, pulse—chase experiments showed that newly synthesized TNSALP (N153D) remained endo-β-N-acetylglucosaminidase H-sensitive throughout the chase. Eventually, after a prolonged chase time, the mutant was found to be partly degraded in a proteasome-dependent manner. Since the mutant TNSALP was significantly labelled with [3H]ethanolamine, a component of GPI, comparable with the wild-type enzyme, it is unlikely that the abortive synthesis of the mutant is due to a defect in GPI-attachment. Interestingly, when asparagine was replaced by glutamine at position 153 (N153D), TNSALP (N153Q) was indistinguishable from the wild-type enzyme in terms of its molecular properties, suggesting the possible importance of amino acids with a polar amide group at position 153. Taken together, these findings indicate that replacing asparagine with aspartic acid at position 153 causes misfolding and incorrect assembly of TNSALP, which results in its retention at the cis-Golgi en route to the cell surface, followed by a delayed degradation, presumably as part of a quality-control process. We postulate that the molecular basis of the perinatal hypophosphatasia associated with TNSALP (N153D) is due to the absence of mature TNSALP at the cell surface.