Diabetologia (1988) 31:563-566 Diabetologia 9 Springer-Verlag 1988 Reviews Proglucagon-derived peptides: nomenclature, biosynthetic relationships and physiological roles J. M. Conlon Clinical Research Group for Gastrointestinal Endocrinology of the Max-Planck-Gesellschaft, University of Grttingen, FRG The decade of the 1980s has seen enormous advances becomes possible, therefore, to use an unambiguous in the field of molecular endocrinology brought about nomenclature that is based either on the amino acid in large measure by application of the methods of re- sequence of the prohormone or, when the site of cleav- combinant DNA technology. The increase in our age of the signal peptide is not known, on the se- knowledge and understanding of the glucagon-related quence of the pre-prohormone. Even when experimen- peptides provides an excellent illustration of this point. tal evidence is lacking, the site of signal peptide cleav- Cloning and sequence analysis of cDNAs and DNA age may be predicted with a high degree of probability fragments from genomic libraries has led to the eluci- [8I so that a nomenclature based upon the prohormone dation of the primary structure of the biosynthetic pre- structure is generally possible and preferred. cursor of glucagon (pre-proglucagon) from a variety of Human pre-proglucagon is a polypeptide of species including man [1-6]. With this information, it 179 amino acid residues that is proteolytically cleaved has been possible to interpret the complex structural by a signal peptidase at the site of the Gln2~ 21 relationships between the multiple forms of the gluca- bond [1]. Amino acid residues in the signal peptide re- gon-like peptides in tissues and in the circulation. gion are assigned negative numbers and the first (N- Probably the most exciting development in the field is terminal) residue in the proglucagon sequence is desig- the realisation that the pre-proglucagon gene in mam- nated + 1. Consequently, the amino acids in the region mals encodes, in addition to glucagon, two additional of the site of cleavage of the signal peptide are desig- peptides with structural similarity to glucagon, termed nated: Ser-3-Trp-Z-Gln-l-Argt-SerLLeu3. All proglu- glucagon-like peptide-I (GLP-1) and glucagon-like cagon-derived fragments are described by the residue peptide-2 (GLP-2). This review attempts to summarise number in proglucagon of their N- and C-terminal recent advances with a view to clarifying previous am- amino acids. The systematic and trivial names of the biguities regarding the nomenclature and biosynthetic major proglucagon-derived peptides present in tissues relationships between the peptides and speculates as to are compared in Figure 1 and their locations within the physiological role of the peptides. the proglucagon molecule are shown schematically. Although the systematic nomenclature is precise, it is somewhat unwieldy for repetitive use so that a strong Nomenclature of proglucagon-derived peptides case may be made for the use of a trivial name, provid- ing that the systematic name is referred to first. Thus, A previous attempt [7] to develop a workable nomen- proglucagon (1-69) becomes glicentin or enterogluca- clature for the classification of glucagon-related pep- gon and the name insulinotropin has been proposed tides was based upon a system that characterised a for proglucagon (78-108) [9]. peptide by its approximate molecular weight, generally determined by gel permeation chromatography, and by its reactivity towards antisera to glucagon of Biosynthetic relationships between proglucagon-derived defined regional specificity. Glucagon, for example, peptides was described as IRG 35~176This nomenclature is subject to severe limitations as the specificities of available The human genome probably contains only a single antisera are seldom totally clear and apparent molecu- copy of the pre-proglucagon gene that comprises five lar weights may be wildly inaccurate. The primary introns and six exons [1, 10]. However, discrepancies structures of the biosynthetic precursors of most of between the nucleotide sequences of the two reported neurohormonal peptides are now known from the nu- structures of the cloned pre-proglucagon gene suggest cleotide sequences of cDNAs or gene fragments. It the possibility of different alleles. Structural analysis 564 J.M.Conlon: Proglucagon-derived peptides Proglucagon Trivial Structural Representation Residues Name KR KRKR R RR (1 - 159 } Proglucagon H2N E II II II I If ~ ~COOH Glicentin 11-69) (enteroglucagon} r ii ii i {1-30) GRPP Fig. t. A schematic representation of (33-69) 0xyntomodulin III human proglucagon and the principal products of post-translational proces- {33-61) Glucagon I I sing of proglucagon. GRPP: glicentin- related pancreatic peptide; GLP: gluca- (72-I08) GLP-1 gon-like peptide, K: lysine, R: arginine. II J The structure of proglucagon shown is GLP- 1( 7-37 ) predicted from the nucleotide sequence ( 78-108) (insulinotropin) I I of the gene according to [1]. The struc- ture predicted from the nucleotide se- [ 126-158) GLP-2 I I quence in [10] contains an additional C-terminal lysine residue has shown that pre-proglucagon mRNAs in the pan- enzyme, e.g. carboxypeptidase H. Chromatographic creas, ileum and colon are identical [11, 12] and an evidence has been provided to show that the glucagon- identical transcriptional start site is used in the three like peptide formerly described as IRG9~176176 probably tissues [12]. It has been concluded, therefore, that the represents proglucagon (1-61) [24]. This component is diversity of the molecular forms of the proglucagon- found in low concentration in the pancreas, gut and derived peptides in the pancreas and gut arise from plasma of healthy subjects but is elevated in the plas- different pathways of post-translational processing of ma of uraemic patients. the primary transcript [11]. As shown in Figure 1, The factors regulating the tissue-specific post-trans- human proglucagon contains several pairs of dibasic lational processing of proglucagon are not understood. amino acid residues (Lys-Arg and Arg-Arg) which can In rat islets, an unusual 0-glycosidic glycosylation of serve as potential recognition sites for a processing the prohormone takes place early in the biosynthetic enzyme [13]. In the A-cell of the pancreas, the prohor- process [25]. It has been shown that purified porcine mone is processed to proglucagon (1-30), which is also glicentin is a substrate for cyclic-AMP-dependent pro- referred to as glicentin-related pancreatic peptide tein kinase [26]. Thus, it is tempting to speculate that (GRPP) [14], glucagon, the hexapeptide proglucagon such post-translational modifications to individual (64-69) [15] but the remaining C-terminal fragment amino acids may be tissue-specific and alter the sus- proglucagon (72-158) is not processed further [16]. In ceptibility of processing sites towards the cleavage the L-cell of the gut, different dibasic residue cleavage enzyme(s). sites are used so that proglucagon is processed to pro- glucagon (1-69) (glicentin) [17], proglucagon (33-69), which has been termed oxyntomodulin [181, GLP-1 Physiological roles of the proglucagon-derived peptides (proglucagon (72-108)), an intervening peptide se- quence (proglucagon (111-123)) and GLP-2 (progluca- The biological significance of peptides derived from gon (126-158) [11, 19]. GLP-1 is a substrate for a sec- the N-terminal region of proglucagon remains unclear. ond proteolytic enzyme in the gut whose specificity in- Proglucagon (1-69) (glicentin) [27], glicentin-related volves cleavage at the site of single arginine residues pancreatic polypeptide [28], oxyntomodulin [29] and [20]. Consequently, the predominant molecular form the C-terminal octapeptide of oxyntomodulin [29] will of GLP in the intestine of the pig [21], rat [11] and hu- inhibit, with varying degrees of potency and effective- mans [22] is the truncated peptide, GLP-1 (7-37). The ness, pentagastrin-stimulated gastric acid secretion in predicted amino acid sequences of GLP-1 (7-37) and the rat. Oxyntomodulin (proglucagon (33-69)) is only proglucagon (111-123) terminate in a glycine residue, one-tenth as potent as glucagon in stimulating adeny- so that there is a strong probability that the peptides late cyclase in rat liver plasma membranes and one- contain an t~-amidated C-terminal residue. Ex- fifth as potent in potentiating insulin-release from the perimental confirmation of this hypothesis is required. perfused rat pancreas; but the peptide is 20-fold more Similarly, the predicted GLP-2 sequence terminates in potent than glucagon in inhibiting gastric acid secre- an arginine residue, so that, by analogy with other sys- tion [30]. Nevertheless, a clear demonstration that the tems [23], it is probable that this residue is removed by enterogastrone effect of proglucagon-derived peptides the action of an endogenous carboxypeptidase B-like is physiologically relevant in man is needed. The asser- J.M.Conlon: Proglucagon-derived peptides 565 tion that proglucagon (1-69) (enteroglucagon) is a tro- sites for GLP-1 (7-36) amide [39] and chemical cross- phic factor for the intestinal mucosa is based upon in- linking experiments have identified a binding protein direct evidence, such as the demonstration that plasma for the peptide in the plasma membrane of enteroglucagon levels correlate well with crypt cell mol.wt.63,000 (R.G6ke and J.M.Conlon, unpub- production rate in several animal models of mucosal lished data). Binding of GLP-1 (7-36) amide to the hypertrophy [31]. It has been shown, however, that cells resulted in an increase in the concentration of cy- continuous in vivo immunoneutralisation of enteroglu- clic AMP [39, 40] and stimulated insulin gene expres- cagon using a monoclonal antibody directed against sion [40]. GLP-1 (1-36) amide bound to the cells with the N-terminal to the central region of glucagon did an affinity that was approximately 200-fold less than not modify the adaptive response of the ileal remnant that of the truncated peptide.