A biological function for glucagon-like peptide-2

The cloning of the glucagon gene from anglerfish, rodents and humans ((1) and references therein) in the early 1980s yielded the surprising finding that the glucagon gene not only encoded the known peptide hormone glucagon, but also contained sequences for two additional peptides, which, because of their homology to glucagon, were named glucagon-like peptide (GLP)-1 and -2 (Fig. 1). The protein product of the gene, preproglucagon, is processed in a tissuespecific manner to glucagon and the major proglucagon fragment in pancreatic endocrine a-cells or to glicentin, oxyntomodulin, GLP-1, intervening peptide 2 (IP-2), and GLP-2 in intestinal L-cells and certain neurons of the brain (Fig. 1). GLP-1 has a high sequence homology to the incretin hormone, gastric inhibitory polypeptide (GIP), and thus, at the time of its identification, was already hypothesized to represent a second incretin hormone (1). Subsequent studies did, indeed, reveal an incretin effect of GLP-1 (2). Furthermore, GLP-1 has been shown to be involved in satiety control and gastric emptying (3, 4). GLP-2 is co-secreted with GLP-1. Fasting levels in human plasma are 151 6 14 pmol/l and these increase to 225 6 15 pmol/l by 2 h after a mixed meal (5). In contrast, GLP-2, a 33-amino acid peptide, has only recently been ascribed a biological function, elucidated in a pioneering effort by Dr Drucker’s group in Toronto. The background of their studies were earlier observations that over-expression of the glucagon gene in rodents was associated with bowel growth and regeneration (6, 7). Furthermore, two reports of patients presenting with proglucagon-expressing tumors in the kidney, increased circulating concentrations of the proglucagon-derived peptides (PGDP), and small-bowel hypertrophy that regressed after tumor resection, suggested a link between peptide products derived from proglucagon and the control of small-bowel epithelium proliferation (8, 9). Small-bowel epithelial proliferation can also be reproduced in glucagon promoter‐SV40 T antigen transgenic mice, which develop proglucagon-producing tumors and increased plasma concentrations of PGDP (10‐12). When nude mice are injected subcutaneously with proglucagonproducing tumors, circulating PGDP are increased and the small bowel appears enlarged (13). To identify further which PGDP was leading to the intestinal enlargement, each PGDP was injected separately to normal mice. Whereas GLP-1 and IP2 had no effect, glicentin treatment led to a small increase in bowel weight. Remarkably, GLP-2 treatment resulted in a 50% increase in total small-bowel wet and dry weight. GLP-2 increased mucosal thickness, which was attributable to an increase in villus height; crypt depth and muscle thickness did not change. Markers for cell proliferation were increased in epithelial cells of the crypts and also extended to the lower portions of the villus surfaces. Remarkably, the effects of GLP-2 were confined to the crypt and villus epithelium, and no changes were detectable in other layers of the intestine or in other organs. Changes in intestinal epithelial proliferation markers were already evident after a single dose of GLP-2, and morphologic changes were detectable after a treatment period of as little as 4 days (13). Food intake and body weight were similar between control and GLP-2 treated animals, excluding the possibility that increased enteral nutrient ingestion attributable to GLP-2 may