Different glycosylation requirements for the synthesis of enzymatically active angiotensin-converting enzyme in mammalian cells and yeast.

For facilitating crystallization and structural studies of the testicular isozyme of angiotensin-converting enzyme (ACE,), we attempted the production of enzymatically active ACET proteins which are unglycosylated or underglycosylated. Expression in Escherichia coli of the rabbit ACET cDNA resulted in the synthesis of an unglycosylated but inactive protein. Similarly, unglycosylated ACET synthesized in HeLa cells, by using a cDNA in which all five potential N-glycosylation sites had been mutated, was inactive and rapidly degraded. Several ACET variants carrying mutations in one or more of the potential N-glycosylation sites were used to examine the role of glycosylation at specific sites on ACET synthesis, transport to the cell surface, cleavage processing, and enzyme activity. These experiments demonstrated that allowing glycosylation only at the first or the second site, as counted from the NH2 terminus, was sufficient for normal synthesis and processing of active ACET. In contrast, ACETg3, which had only the third glycosylation site available, was unglycosylated, enzymatically inactive and rapidly degraded. N-Glycosylated ACET could also be produced in yeast. Surprisingly, the mutant ACETg3 was synthesized, N-glycosylated, and properly transported in yeast. Wild type and mutant ACE proteins were cleavage-secreted from yeast and enzymatically active.