| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Medical Physiology (R.U., C.F.D., J.J.H.) and Medical Anatomy (C.Ø.), the Panum Institute, University of Copenhagen, DK-2200 Copenhagen N., Denmark; and Department of Biochemistry and Molecular Biology (X.Z., D.F.S.), the Howard Hughes Medical Institute, University of Chicago, Chicago, Illinois 60637
Address all correspondence and requests for reprints to: Prof. Jens Juul Holst, Department of Medical Physiology, University of Copenhagen, the Panum Institute, Blegdamsvej 3, DK-2200 Copenhagen N., Denmark. E-mail: holst{at}mfi.ku.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) (9, 10, 11). Indeed, several experiments in cell lines expressing either PC1 or PC2 and transfected with proglucagon have shown an intestinal processing profile of proglucagon with expression of PC1 (12, 13, 14), and a pancreatic processing profile with expression of PC2 (15, 16). Furthermore, it is possible to produce a shift from an intestinal processing profile to a pancreatic processing profile when a cell line already expressing PC1 and proglucagon is cotransfected with PC2 and vice versa (12, 13). Because in vivo proglucagon has been shown to colocalize to a high degree with PC1 in the intestine (17, 18) and PC2 in the islet A cell (15, 19, 20), it seems possible that variations in expression levels of these two prohormone convertases with separate substrate specificities play an important role for the differential processing of proglucagon. It can, however, be argued that other unknown factors may also contribute to the intestinal processing profile because PC2 has been shown to colocalize with proglucagon in canine small intestine and in canine L cell cultures without concurrent production of significant amounts of glucagon (18). Furthermore, the more recently identified PC6A has been shown to be widely expressed throughout the intestine (21) and also to be localized to the regulated secretory pathway (22). PC2-deficient mice were recently demonstrated to exhibit a complex phenotype, in agreement with the suggested neural and endocrine biological functions of PC2. No mature glucagon was measurable in pancreatic extracts and plasma and, as a result, an accumulation of proglucagon occurred (8, 23). Hence, the mice have chronic fasting hypoglycemia and a reduced rise in blood glucose levels after an ip glucose tolerance test, despite a reduction in insulin levels. In addition, the pancreatic A cells showed hypertrophy and hyperplasia consistent with lack of inhibitory glucose and insulin feedback (23) and/or lack of negative feedback by glucagon (24). A PC1-deficient mouse has also recently been described (7). These animals exhibited multiple neural and endocrine prohormone processing defects, involving proinsulin, proopiomelanocortin, and pro-GHRH, as well as very low levels of intestinal GLP-1 and GLP-2 immunoreactivity. The present study provides a detailed profiling of proglucagon processing, revealing severe cleavage defects in the intestine but normal levels of fully processed glucagon in the pancreas of this PC1-deficient strain.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunohistochemistry
Specimens of upper and lower jejunum and ileum from three wild-type mice and four PC1 null mice were fixed in 4% buffered paraformaldehyde and embedded in paraffin. The tissue sections were dewaxed and subjected to antigen-retrieval by microwave irradiation in citrate buffer (pH 6.0) before the immunohistochemical stainings. For immunohistochemistry, we used a PC1 antiserum, RS-18B7, produced as described in the work of Scopsi et al. (17), but found to be more specific than the previously reported RS-20 antiserum (Zang, C., and D. F. Steiner, personal communication). For PC2 immunohistochemical stainings, we used the rabbit PC2Pep4 antiserum (17), and for proglucagon we used the side-viewing monoclonal mouse antibody GLPb-17F1A37B22C4 (F1). This antibody was a generous gift from Novo Nordisk (Bagsvaerd, Denmark) (25). In double-staining experiments, the PC1 antiserum was diluted 1:1000 and visualized using biotin antirabbit antibody (diluted 1:200; Dako, Glostrup, Denmark), and Texas red-labeled streptavidine (red fluorescence, diluted 1:200; Amersham Pharmacia Biotech, Allerod, Denmark). Simultaneously, the GLP-1 antibody F1 was diluted 1:1000 and visualized using digoxigenin-labeled antimouse antibody (diluted 1:25; Boehringer, Marburg, Germany) and fluorescein-labeled antidigoxigenin (diluted 1:25; green fluorescence, Boehringer). For immunohistochemistry, the PC2 was diluted 1:10,000, and visualized with biotin antirabbit and Texas red-labeled streptavidine as described above. For the double-stainings, we used the Mouse on Mouse (MOM) Kit (Vector Laboratories, Burlingame, CA) for dilution of antibodies and reagents, to allow for using a mouse primary antibody in mouse tissue.
Extraction of intestinal proglucagon
The frozen tissue was extracted at neutral pH as previously described [extraction of small samples and acidic peptides, (26)]. In short, frozen tissue was submerged in boiling water (10 ml/g tissue), boiled for 8 min, homogenized with a Polytron PT 3000 homogenizer (Kinematica AG, Littau/Luzern, Switzerland), and centrifuged. The supernatant was retained for gel filtration.
Extraction of intestinal and pancreatic proglucagon-derived peptides
The tissue extraction method has been described previously [method for neutral or basic peptides (26)]. Briefly, frozen pieces of intestine were homogenized in four volumes of acid ethanol. The homogenate was centrifuged, and the supernatant evaporated to one third with N2, neutralized to pH 4 with NaOH, diluted seven times with 40 mmol/liter sodium phosphate buffer (pH 7.5) containing, in addition, 0.1% wt/vol human serum albumin, 0.1 mol/liter NaCl, 0.6 mmol/liter thimerosal, and 10 mmol/liter EDTA.
Gel filtration
The extracts were applied to a K16/100 Sephadex G50, fine-grade column (Amersham Pharmacia Biotech, Hoersholm, Denmark), equilibrated, and eluted with 40 mmol/liter sodium phosphate buffer (pH 7.5) containing, in addition, 0.1% wt/vol human serum albumin, 0.1 mol/liter NaCl, 0.6 mmol/liter thimerosal, and 10 mmol/liter EDTA with a flow rate of 0.4 ml/min at 4 C. The void volume (V0) and the available inner volume (Vt) of the gel bed were determined from elution positions of 125I-albumin and 22NaCl, both added in trace amounts to the extracts before gel filtration for internal calibration. The coefficient of distribution (Kd) was calculated using the formula Kd = (Ve - V0)/(Vt - V0) where Ve is the elution volume of the peak investigated. When analyzing for proglucagon, only 22NaCl was added, and the coefficient of distribution was calculated from this tracer and previous calibration. Synthetic glicentin, oxyntomodulin, glucagon, GLP-1(7-36NH2), and GLP-2 were subjected to gel filtration under similar conditions. Eluted fractions were assayed by RIA.
Reverse-phase HPLC
Aliquots of gel filtration fractions containing glucagon, GLP-1, or GLP-2 immunoreactivity were pooled, acidified with 10% trifluoroacetic acid (TFA; Rathburn, Walkerburn, UK) and applied to HPLC using a Vydac C18 column (Microlab Aarhus A/S, Hoejbjerg, Denmark). The column was eluted with a linear gradient of acetonitrile (AcN; Rathburn) in 0.1% TFA (0–75% AcN over 75 min followed by 75–74% AcN for 5 min; 1 ml/min). Eluted fractions were dried in a Heto vacuum centrifuge (Heto-Holten A/S, Alleroed, Denmark), reconstituted, and assayed by RIA.
RIA
Three different RIAs for GLP-1 immunoreactivity were used. Briefly, antiserum 2135 (27, 28) is side-viewing and measures all molecules that contain the midsequence of GLP-1 (Fig. 1
). It reacts equally with all N- and C-terminal truncated or extended forms of GLP-1 and can, therefore, be said to be largely processing-independent. Antiserum 93242 (29) is specific for the N terminus and reacts minimally with N terminally extended or truncated forms of GLP-1. Antiserum 89390 (30) has an absolute requirement for the intact amidated C terminus. GLP-1(7-36NH2) was used in all assays for standard and tracer preparation.
Intact GLP-2 was measured using antiserum 92160, which is N-terminally directed and cross-reacts minimally with N-terminally extended or truncated forms of GLP-2 (31). For standards, we used recombinant human GLP-2(1-33), and the tracer was bovine GLP-2 with Thr12
Tyr12 substitution, 125I-labeled using the standard stoichiometric chloramine T method, as described elsewhere (26). Total GLP-2 immunoreactivity was measured using the monoclonal mouse side-viewing antiserum HYB 312–01 (produced in collaboration with Statens Serum Institut, Copenhagen, Denmark), which detects all N- and C-terminally truncated or extended forms of GLP-2 (Fig. 1). As standards, recombinant human GLP-2(1-33) was used, and the tracer was rat GLP-2, with Asp33Tyr33 substitution for iodination. In agreement with its specificity for a midregion of GLP-2, the antiserum does not bind the Tyr12-GLP-2 tracer.
Glucagon immunoreactivity was measured using two antisera as previously described (32). Antiserum 4304 detects all peptides containing the 6-15 sequence of glucagon including glucagon itself, glicentin, and oxyntomodulin (Fig. 1). Antiserum 4305 is specific for the C terminus of glucagon, it only detects peptides with fully processed C terminus of glucagon, and it reacts with neither glicentin nor oxyntomodulin.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A–C), but not in the null mice, where PC1 was not present (Fig. 2, D–F).
|
, A–C). Applying all three assays to intestinal extracts, the null mice presented with a shift from mature, fully processed peptides to a peptide with early elution position at Kd 0.14 for all assays [2135: 0.14 ± 0.05; HYB 312-01: 0.13 ± 0.03; 4304: 0.14 ± 0.05 (mean ± SD), not significant]. The peak was identified as proglucagon because of its large molecular size and the presence of glucagon, GLP-1, and GLP-2 immunoreactivity. These results indicate that the PC1 -/- mice have a severe block in their proglucagon processing and suggest accumulation of proglucagon in secretory granules of the L cell.
|
, A–D) using the antisera 4304 for total glucagon immunoreactivity, 93242 and 89390 for intact and amidated forms of GLP-1, respectively, and 92160 directed against the N terminus of GLP-2. These extracts confirmed the presence of mature proglucagon-derived peptides in the wild-type mice and a processing defect in the PC1-deficient animals. In support of previous studies performed in cell lines suggesting that the K70-R71 site (Fig. 1) can be processed by more than one endoprotease (12, 13, 14, 33), our results indicate that partial processing at K70-R71 may occur, because the null mice appear to produce small amounts of glicentin as indicated by gel filtration (Fig. 4A) and by HPLC (data not shown). Processing leading to the formation of glicentin would be expected to result in the formation of similar amounts of major proglucagon fragment (MPGF). If present in the extracts, this component (MPGF) would probably elute close to the proglucagon peak and, therefore, escape detection. Having established a severe processing defect, we decided to examine whether the PC1-deficient mice were able to produce any mature intestinal peptides (GLP-1 and GLP-2) at all. Two PC1 null mice showed low but correctly positioned levels of GLP-1 and GLP-2 immunoreactivity, and one showed only low levels of GLP-2 immunoreactivity. Accordingly, gel filtration fractions containing GLP-1 or GLP-2 immunoreactivity collected at Kd values corresponding to wild-type immunoreactive peptides, were pooled and subjected to HPLC, and were subsequently analyzed for total GLP-1 (code 2135) and total GLP-2 (HYB 312-01) immunoreactivity (Fig. 5). GLP-1 immunoreactivity eluted in more than one position, and relevant fractions were subsequently analyzed for N-terminal immunoreactivity (code 93242) and with an antiserum specific for the amidated C terminus (code 89390). The peak discovered in the wild-type was found to consist solely of GLP-1(7-36NH2). However, N- and C-terminal GLP-1 immunoreactivity did not elute in the same fractions in the null mice. In contrast, low levels of GLP-2 from PC1 null mice were found to elute at the same elution position as in the control mice.
|
|
). Furthermore, the gel filtration profile of acidic extract from wild-type mice showed an early peak of C-terminal glucagon immunoreactivity (Kd, 0.39). This may correspond to the 9-kDa peptide (glicentin-related pancreatic polypeptide-glucagon) and was not found in the PC1 null mice. Aliquots of PC1 -/- gel filtration fractions containing C-terminal glucagon immunoreactivity at the glucagon standard elution position were pooled and subjected to HPLC. The peptide was found to elute at the same position as aliquots of PC1 +/+ pancreas gel filtration fractions containing total glucagon immunoreactivity (code 4304) and glucagon standard. Thus, we identified the elevated levels of C-terminal immunoreactive peptide in the intestinal, acidic extract from null mice as mainly mature glucagon.
|
), and the findings were confirmed by HPLC (data not shown), whereas processing to both GLP-1 and GLP-2 was negligible in wild-type as well as PC1 null animals. This supports previous results indicating that PC2 is the essential convertase responsible for the formation of glucagon in the A cell by cleaving the dibasic sites K70-R71, K31-R32, and K62-R63 (8, 23) (Fig. 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Based on the previous evidence, it has been speculated that PC1 or PC2 is needed for the first cleavage at the intermediate site K70-R71. Thereafter, the L cell-specific processing relies on PC1; however, this has never been tested in vivo. The PC1 null mice studied here accumulate proglucagon in the intestine, and only produce a partial cleavage at the interdomain site K70-R71, resulting in low levels of glicentin. This reduction is consistent with previous cell line studies and in vivo results indicating that this site is the most readily accessible in the intact precursor and is rapidly cleaved by either PC1 or PC2 (8, 12, 14, 33). The observed partial cleavage of K70-R71 in the PC1 null mice is, however, not likely to be the result of low levels of PC2, because this was undetectable by immunohistochemistry, but could be mediated by other convertases like furin or PC6A.
In vivo, the C-terminal domain of intestinal proglucagon is processed to GLP-1(7-36NH2) and GLP-2 (9, 11). This processing profile is reproduced in the wild-type mice, whereas the PC1 null mice have a severe defect in cleavage of the relevant sites resulting in nearly no N- or C-terminal GLP-1 and low levels of N-terminal GLP-2 immunoreactivity upon gel filtration analysis. These results are in agreement with previous experiments performed in cell-free systems and in various cell lines used as a model for the intestinal L cell (12, 13, 14, 34). Despite this quantitative defect, we identified small amounts of GLP-1 and GLP-2 immunoreactivity by HPLC analysis. However, when analyzing the GLP-1 immunoreactivity in detail, we found a complete block in the production of bioactive GLP-1(7-36NH2) in the null mice. The total GLP-1 immunoreactivity, as shown in Fig. 5, may represent low levels of GLP-1(1-37), GLP-1(7-37), and GLP-1(1-36NH2) in agreement with previous studies (35), or degraded forms containing the GLP-1 sequence. Although a lack of PC1 message or protein by Northern and Western blotting in hypothalamus and whole pituitary, respectively, has previously been reported (7), the finding of low levels of GLP-2 could still be argued to result from residual low levels of functional intestinal PC1. To rule out this possibility, we performed immunohistochemistry and confirmed that PC1 is present in proglucagon-producing cells in the small intestine of control mice, yet absent in the small intestine of the null mice. Thus, the finding of low levels of GLP-2 immunoreactivity suggests that other convertases are able to process the relevant proglucagon cleavage site. This is, however, inefficient.
The question of whether PC1 in vivo is able to cleave at the monobasic site R77 has been enigmatic. In cell lines and in cell-free studies, PC1 has been shown to cleave a monobasic site in the processing of prodynorphin (36), which, like the monobasic proglucagon site R77, lacks both a P2 basic and a P4 arginine residue. All cell lines expressing PC1 and transfected with proglucagon, which have been examined, have been found to be able to process proglucagon to GLP-1(7-37/36NH2) (12). In addition, primary rat A cells in culture, which normally express MPGF and not GLP-1(7-36NH2), were shown to produce GLP-1(7-36NH2) after transfection with PC1 (13). These data imply that PC1 could be the convertase responsible for the in vivo cleavage of the proglucagon processing residues, including the monobasic site. Results of in vitro cell-free studies have, however, been conflicting. When PC1 was incubated with GLP-1(1-37), no conversion to the intact GLP-1(7-37) was initially found (12). Only when PC1 in large amounts was incubated for 16 h with GLP-1(1-37), was a partial conversion observed (13). Consequently, it has been proposed that GLP- 1(1-37) is a poor substrate for PC1 monobasic cleavage, in agreement with the findings in pulse-chase experiments indicating that MPGF may be the natural substrate in vivo (13). We found markedly decreased levels of N-terminal GLP-1 immunoreactivity in the PC1 null mice by gel filtration and HPLC, indicating that PC1 is necessary for the monobasic cleavage and, thereby, may play a role in the processing of all the cleavage sites to generate the normal intestinal proglucagon processing profile.
Upon examining the processing of the intestinal proglucagon N-terminal domain, we found an elevated formation of glucagon in the PC1 nulls. The significance of this finding is, at present, unclear. PC2 has previously been reported to be present in the L cells of the dog (18). However, we were unable to find intestinal PC2 in either wild-type or null mice by immunohistochemistry (data not shown).
When measuring the pancreatic processing to glucagon, we found similar amounts in the pancreas of PC1 nulls and wild-type mice. It has been proposed that PC2 is insufficient for cleavage of glucagon from proglucagon in vivo, because the cleavage of the K31-R32 site, flanking the N-terminal of mature glucagon, by PC2 is inefficient (14, 37). It has, therefore, been suggested that other convertases, including PC1, may be involved in the processing of glucagon from proglucagon. This study shows that PC1 is not necessary for production of glucagon.
In conclusion, these results strongly support the role of PC1 as essential for cleavage of all the intestinal proglucagon processing sites in vivo and for the production of the bioactive hormones GLP-1 and GLP-2.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AcN, Acetonitrile; GLP, glucagon-like peptide; MPGF, major proglucagon fragment; PC, prohormone convertase; TFA, trifluoroacetic acid.
Received June 26, 2003.
Accepted for publication November 10, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TC1–6 cells. Proc Natl Acad Sci USA 91:3242–3246 cell hyperplasia in glucagon receptor knockout mice. Proc Natl Acad Sci USA 100:1438–1443This article has been cited by other articles:
 |
|
 |
|
  W. Kim and J. M. Egan The Role of Incretins in Glucose Homeostasis and Diabetes Treatment Pharmacol. Rev., December 1, 2008; 60(4): 470 - 512. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Holst The Physiology of Glucagon-like Peptide 1 Physiol Rev, October 1, 2007; 87(4): 1409 - 1439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. E. Pritchard and A. White Neuropeptide Processing and Its Impact on Melanocortin Pathways Endocrinology, September 1, 2007; 148(9): 4201 - 4207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Simonsen, S. Pilgaard, C. Orskov, M. M. Rosenkilde, B. Hartmann, J. J. Holst, and C. F. Deacon Exendin-4, but not dipeptidyl peptidase IV inhibition, increases small intestinal mass in GK rats Am J Physiol Gastrointest Liver Physiol, July 1, 2007; 293(1): G288 - G295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Gromada, I. Franklin, and C. B. Wollheim {alpha}-Cells of the Endocrine Pancreas: 35 Years of Research but the Enigma Remains Endocr. Rev., February 1, 2007; 28(1): 84 - 116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ugleholdt, M.-L. H. Poulsen, P. J. Holst, J.-C. Irminger, C. Orskov, J. Pedersen, M. M. Rosenkilde, X. Zhu, D. F. Steiner, and J. J. Holst Prohormone Convertase 1/3 Is Essential for Processing of the Glucose-dependent Insulinotropic Polypeptide Precursor J. Biol. Chem., April 21, 2006; 281(16): 11050 - 11057. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Dey, G. M. Lipkind, Y. Rouille, C. Norrbom, J. Stein, C. Zhang, R. Carroll, and D. F. Steiner Significance of Prohormone Convertase 2, PC2, Mediated Initial Cleavage at the Proglucagon Interdomain Site, Lys70-Arg71, to Generate Glucagon Endocrinology, February 1, 2005; 146(2): 713 - 727. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. C. Webb, A. Dey, J. Wang, J. Stein, M. Milewski, and D. F. Steiner Altered Proglucagon Processing in an {alpha}-Cell Line Derived from Prohormone Convertase 2 Null Mouse Islets J. Biol. Chem., July 23, 2004; 279(30): 31068 - 31075. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
