Cleavage of proglucagon by the prohormone converting enzymes PCSK1 and PCSK2 is not dictated by the amino acid sequence immediately adjacent to the cleavage site

Abstract

The peptide hormones glucagon, glucagon‐like peptide‐1, and glucagon‐like peptide‐2 are all derived from proglucagon (PG), a 160 amino acid polypeptide containing multiple potential cleavage sites designated by either single or paired basic amino acids. Initial cleavage of PG takes place on the C‐terminal side of the K70‐R71 dibasic site, which liberates the intermediates glicentin and major proglucagon fragment (MPGF). My laboratory is interested in examining the specificity of the prohormone cleaving enzymes PCSK1 and PCSK2 using both PG and proinsulin (PI) as model substrates. PG is a useful model because it has one cleavage site (the glicentin/MPGF junction) recognized almost equally well by both enzymes and multiple potential cleavage sites that are not recognized by either enzyme. PI is a good model because it has one cleavage site (the B chain/C peptide junction) preferentially recognized by PCSK1 and a second cleavage site (the C peptide/A chain junction) preferentially recognized by PCSK2. My lab has previously described the expression and purification of these two model substrates. However, the purification of PG has presented a number of problems because of the presence of contaminants that obscure detection and participate in co‐precipitation during early stages of purification. I have now devised an improved, two‐step scheme using step elution from SP Sepharose as a capture step, followed by gradient elution from Q Sepharose as the major purification step. This method resolves the problems posed by the contaminants and generates PG samples suitable for analysis of the PCSK‐mediated cleavage reaction. An in vitro cleavage assay was then set up so that the PG substrate (~3 μM) incubated with either PCSK1 or PCSK2 (~100 nM at pH 6.0 or 5.4, respectively) results in ~30% cleavage at the glicentin/MPGF junction using either enzyme. As noted above, in contrast to the situation with PG, the normal maturation pathway for PI exhibits a marked preference for PCSK1‐mediated cleavage at the B chain/C peptide junction. To examine whether this cleavage selectivity is due to amino acids in the immediate vicinity of the cleavage site, the sequence EAED from the C‐terminal side of the B chain/C peptide cleavage site in PI was cloned into the C‐terminal side of the K70‐R71 cleavage site in PG. In contrast to expectations, this substitution resulted in a nearly 50% decrease in PG cleavage by PCSK1. Equally surprising was that this substitution had almost no effect on the efficiency of PCSK2‐mediated cleavage of the altered site, despite the fact that in the original context of the PI substrate, this sequence is recognized much less efficiently by PCSK2 than by PCSK1. Taken together, these results suggest that three‐dimensional features (secondary or tertiary structure) play a significant role in the cleavage site specificity exhibited by both PCSK1 and PCSK2.