help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2003-0543
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Authier, F.
Right arrow Articles by Briand, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Authier, F.
Right arrow Articles by Briand, G.
Endocrinology Vol. 144, No. 12 5353-5364
Copyright © 2003 by The Endocrine Society

Endosomal Proteolysis of Glucagon at Neutral pH Generates the Bioactive Degradation Product Miniglucagon-(19–29)

François Authier, Pamela H. Cameron, Clémence Merlen, Mostafa Kouach and Gilbert Briand

Institut National de la Santé et de la Recherche Médicale Unité 510 (F.A., C.M.), Faculté de Pharmacie Paris XI, 92296 Châtenay-Malabry, France; Department of Anatomy and Cell Biology (P.H.C.), McGill University, Montréal, Québec, Canada H3A 2B2; and Laboratoire de Spectrométrie de Masse (M.K., G.B.), Faculté de Médecine, 59000 Lille, France

Address all correspondence and requests for reprints to: François Authier, Institut National de la Santé et de la Recherche Médicale Unité 510, Faculté de Pharmacie Paris XI, 5 rue Jean-Baptiste Clément, 92296 Châtenay-Malabry, France. E-mail: francois.authier{at}cep.u-psud.fr.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
We have investigated the proteolytic mechanisms of glucagon degradation within hepatic endosomes at neutral pH before lumen acidification. Hepatic endosomes incubated at neutral pH rapidly degraded native glucagon into 13 intermediate products, one of which corresponded to the bioactive fragment glucagon-(19–29) (miniglucagon). The serine protease inhibitor phenylmethylsulfonyl fluoride as well as the nonspecific protease inhibitor bacitracin inhibited the endosomal degradation of glucagon at pH 7. In purified endosomal fractions, miniglucagon endopeptidase was undetectable as evaluated by immunoblotting, and immunoprecipitation with antibodies to insulin-degrading enzyme, cathepsins B and D, or furin failed to remove the endosomal neutral glucagonase activity. Incubation of endosomal fractions and [125I]iodoglucagon with the zero-length bifunctional cross-linker 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide resulted in specific labeling of a 170-kDa polypeptide. The labeling was completely inhibited by unlabeled glucagon (IC50 value, 5 x 10-7 M) and bacitracin (IC50 value, 1 µg/ml), suggesting that it may correspond to a bacitracin-sensitive glucagon-degrading enzyme. Treatment of the 125I-labeled 170-kDa cross-linked polypeptide with N-glycanase demonstrated that the cross-linked complex contained approximately 30 kDa of N-linked oligosaccharides. Specific cross-linking of the 170-kDa polypeptide was also observed using [125I]Tyr12-miniglucagon as the radioligand. Together, these data suggest that the 170-kDa glycoprotein represents a novel glucagon-degrading activity that could mediate glucagon proteolysis within endosomes before the acidification step and generate the bioactive (19–29) miniglucagon peptide.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
GLUCAGON IS A 29-amino acid peptide hormone secreted by the {alpha}-cells of the pancreatic islets. Initially, the binding of glucagon to its cell surface receptor (1) triggers the G protein-mediated activation of adenylyl cyclase. This leads to an increase in intracellular cAMP followed by the activation of protein kinase A and, ultimately, an increase in glycogenolysis and a reduction in glycogen synthase activity in the liver (2). Over a period of 20–30 min, the agonist-receptor complex is internalized to a low-density endocytic compartment, physically separable from lysosomes and Golgi elements (3, 4, 5, 6). Finally, hepatic subcellular fractionation studies identified the endosomal compartment as a major site of degradation of internalized glucagon (5). Glucagon degradation within hepatic endosomes was optimal at acidic pH and functionally linked to ATP-dependent endosomal acidification (7). Recently, glucagon degradation has been attributed to membrane-bound forms of cathepsins B and D (8). In addition, another hepatic endosomal protease, insulin-degrading enzyme (IDE), has also been proposed to participate in endosomal clearance of glucagon at neutral pH (9, 10).

A large number of studies have proposed the plasma membrane as another physiological locus for glucagon degradation (reviewed in Ref. 11). At least four neutral proteolytic activities have been proposed to operate at the hepatic cell surface: 1) a tripeptidyl aminopeptidase that sequentially deletes three amino acids from the N terminus of glucagon (12); 2) a glucagon receptor-linked serine protease that cleaves at the internal peptide bond Tyr13-Leu14 (13); 3) dipeptidyl peptidase IV, a serine protease that releases dipeptides from the N terminus of glucagon and produces glucagon-(3–29) and glucagon-(5–29) (14); and 4) a bacitracin-sensitive metalloprotease, miniglucagon endopeptidase (MGE), which processes glucagon to its bioactive fragment glucagon-(19–29) (miniglucagon) (15, 16).

The miniglucagon fragment has its own biological activity and is a modulator of the action of glucagon, the mother molecule. At picomolar concentrations, it inhibits the hepatic plasma membrane Ca2+ pump without interfering with the adenylyl cyclase activity (17). At the same concentration, glucagon-(19–29) displays a negative inotropic effect on myocyte contraction (18) and the ability to inhibit insulin release through a Ca2+ pathway (19). The fragment appears to be produced either from the circulating glucagon at the surface of the target cells or inside the islets of Langerhans (15, 20).

It has been suggested that miniglucagon might be produced from circulating glucagon after interaction with glucagon target tissues, such as the liver, via the cell surface MGE (15). Thus, miniglucagon-like immunoreactivity appeared in the cell supernatant upon incubation of glucagon with HepG2 hepatoma cells, which do not contain high-affinity glucagon receptors (15). Purified from rat liver plasma membrane, the MGE activity, which was shown to be distinct from IDE, displayed a neutral pH optimum, a molecular mass of 100 kDa on SDS-PAGE, and a sensitivity to sulfhydryl-modifying reagents and chelating agents (16). However, the role of the plasma membrane as the physiological locus of miniglucagon production remains controversial due to inconsistencies in the data obtained with the in vivo and in vitro experiments on miniglucagon production. Thus, the absence of detectable amounts of native glucagon-(19–29) in rat plasma (15), and the very low peptide level generated upon incubation of glucagon with plasma membrane [1% of glucagon being converted into glucagon-(19–29) even in the presence of protease inhibitors (15)] make the involvement of the cell surface membrane in the generation of miniglucagon questionable. Moreover, the MGE enzyme has been shown to be competitively inhibited by insulin (which does not contain any dibasic sites) with an affinity higher than that for glucagon (16), indicating that it is not specific for the glucagon pathway. Pancreatic production of miniglucagon has been observed, in which the bioactive peptide is present in a stored form at molar concentrations in the range of 3–4% that of glucagon, and then released concomitant to glucagon secretion (20).

Consequently, the present study has attempted to evaluate the physiological relevance of the endosomal compartment to the generation of miniglucagon from internalized glucagon. We report here that hepatic endosomes contain a membrane serine endopeptidase activity that yields at neutral pH the glucagon-(19–29) peptide. Using immunological criteria, the data presented here demonstrate that the endosomal glucagon-(19–29)-generating enzyme is not related to the MGE activity previously suggested to be at the cell surface (15, 16), to IDE previously reported to be a candidate endosomal glucagon-degrading enzyme (10), nor to endosomal cathepsins B and D (8). Finally, affinity-labeling experiments using the zero-length bifunctional cross-linker 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) demonstrate the specific binding of [125I]iodoglucagon and [125I]Tyr12-miniglucagon to a 170-kDa endosomal membrane glycoprotein that fulfills the criteria expected for an endosomal-neutral glucagonase in terms of affinity and specificity toward the glucagon molecule, and sensitivity to the protease inhibitor bacitracin.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Peptides, antibodies, protein determination, peptide synthesis, antibody preparation, and materials
Porcine glucagon and recombinant human furin extracted from Sf-9 cells (2000 U/ml) were purchased from Sigma-Aldrich (St. Louis, MO). Rabbit antimouse cathepsin D R291 (8, 21) and antirat cathepsin B 7183 (8, 22) were obtained from Dr. J. S. Mort (Shriners Hospital for Crippled Children, Montréal, Québec, Canada) and used to immune deplete samples of native mature enzymes as described previously (8, 21, 22). Mouse monoclonal antibody 9B12 directed against the human IDE (23) was a gift from Dr. R. A. Roth (Stanford University, Stanford, CA). Rabbit polyclonal antiserum against rat liver 3-ketoacyl-coenzyme A (CoA) thiolase (thiolase) was a gift from Dr. R. A. Rachubinski (University of Alberta, Edmonton, Alberta, Canada). Rabbit polyclonal IgG H-220 raised against the C-terminal region of recombinant human furin (amino acids 575–794) and affinity-purified goat polyclonal IgG raised against human Hsp-60 were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Horseradish peroxidase-conjugated goat antirabbit IgG was from Bio-Rad (Hercules, CA). The protein content of isolated fractions was determined by the method of Lowry et al. (24). A peptide of 12 amino acids (AQDFVQWLMNTY) corresponding to the C-terminal region of human glucagon was synthesized with a tyrosine at the C terminus using a solid-phase procedure. The C-terminal tyrosine permitted radioiodination of the synthetic peptide. A peptide of 8 amino acids corresponding to the N-terminal residues (CREARPTTE) of rat MGE (16) was synthesized with a cysteine at the N terminus using a solid-phase procedure. The identity and purity of the synthetic peptides were confirmed by HPLC, amino acid, and mass spectrometry analyses. For antibody production, the CREARPTTE peptide was conjugated to BSA and injected into female New Zealand white rabbits. Nitrocellulose membranes and enhanced chemiluminescence detection kit were from Amersham (Arlington Heights, IL). Protein G-Sepharose was from Pharmacia (Peapack, NJ). Bacitracin, pepstatin-A, 1,10-phenanthroline, N-ethylmaleimide, p-hydroxymercuribenzoate, EDTA, EGTA, L-3-carboxy-2,3-trans-epoxypropionyl-leucylamido(4-guanidino)butane (E64), phenylmethylsulfonyl fluoride (PMSF), and benzamidine were from Sigma-Aldrich. N-Glycanase and GTP were from Roche Molecular Biochemicals (Basel, Switzerland). HPLC-grade acetonitrile and trifluoroacetic acid (TFA) were obtained from Baker Chemical Co. (Phillipsburg, NJ). All other chemicals were obtained from commercial sources and were of reagent grade.

Animals and injections
In vivo procedures were approved by the institutional committee for use and care of experimental animals. Male Sprague Dawley rats (body weight, 180–200 g) were obtained from Charles River (St. Aubin Les Elbeufs, France) and were fasted for 18 h before being killed. Native glucagon (15 µg/100 g body weight) in 0.4 ml of 0.15 M NaCl was injected within 5 sec into the penile vein under light anesthesia with ether.

Isolation of subcellular fractions from rat liver
Subcellular fractionation was performed using established procedures (5, 7, 8, 21, 22). Following injection of human glucagon, animals were killed, and livers were rapidly removed and minced in isotonic ice-cold homogenization buffer as previously described (5, 7, 8, 21, 22).

Rat liver nuclear (N), large-granule (ML), microsomal (P), and cytosolic (S) fractions were isolated by differential centrifugation as previously described (5, 7, 25). The endosomal fraction (EN) was isolated by discontinuous sucrose gradient centrifugation and collected at the 0.25–1.0 M sucrose interface (5, 7, 8, 21, 22, 25). Plasma membrane was prepared according to the method of Neville (26) as described by Authier et al. (5, 27). Rough endoplasmic reticulum (ER) was prepared by the protocol of Walter and Blobel (28) as described previously (25) and used as described in Authier et al. (25). Purified peroxisomes (Per) were obtained as described by Bodnar and Rachubinski (29) with an 18.6 ± 0.5-fold cytochrome c oxidase enrichment and used as described in Authier et al. (25). Lysosomes (L2 fraction) were prepared by isopycnic centrifugation of the light mitochondrial fraction in a discontinuous metrizamide gradient according to the method of Wattiaux et al. (30) with a 53.2 ± 4.1-fold N-acetyl-ß-D-glucosaminidase enrichment (21).

Biochemical characterization
Triton X-114 (Sigma-Aldrich) extraction was carried out by the method of Bordier (31). N-Glycanase treatment of the [125I]glucagon/gp-170 cross-linked complex (see Cross-linking studies) was carried out essentially as previously described for endoglycosidase digestion of [125I]glucagon/glucagon receptor cross-linked complex (27). Cross-linked proteins (300 µg protein) were incubated at 37 C under constant shaking with 1 U N-glycanase for various times in 30 mM sodium phosphate buffer (pH 7.6) containing 20 mM EDTA, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate, and 0.5% 2-mercaptoethanol. The digestion was stopped with Laemmli sample buffer (32), and the samples were subjected to electrophoresis in an 8% acrylamide resolving gel.

Immunoblot analysis
Electrophoresed samples were transferred to nitrocellulose membranes (0.45 x 10-6 m) for 60 min at 380 mA in transfer buffer containing 25 mM Tris base and 192 mM glycine. The membranes were blocked by a 3-h incubation with 5% skim milk in 10 mM Tris-HCl (pH 7.5), 300 mM NaCl, and 0.05% Tween 20. The membranes were then incubated with primary antibody [rabbit polyclonal antisera against either mouse cathepsin D R291 (diluted 1:1500), rat MGE (diluted 1:200), rat liver 3-ketoacyl-CoA thiolase (diluted 1:300), affinity-purified goat polyclonal IgG against human Hsp-60 (diluted 1:500), or affinity-purified rabbit polyclonal IgG against human furin (diluted 1:500)] in the above buffer for 16 h at 4 C. The blots were then washed three times with 0.5% skim milk in 10 mM Tris-HCl (pH 7.5), 300 mM NaCl, and 0.05% Tween 20 over a period of 1 h at room temperature. The bound Ig was detected using horseradish peroxidase-conjugated goat antirabbit IgG or rabbit anti-goat IgG.

In vitro proteolysis of native glucagon by hepatic endosomes and furin
Hepatic endosomes (EN) ({approx}1 µg) were incubated for varying lengths of time at 37 C with 1 µM porcine glucagon in 200 µl of 50 mM citrate-phosphate buffer (pH 7–4), in the presence or absence of protease inhibitors. The samples were then acidified with acetic acid (15%) and immediately loaded onto a reversed phase-HPLC (RP-HPLC) column.

In some experiments, native glucagon was also digested in vitro with recombinant human furin. Glucagon (10-6 M) was incubated with 28 U/ml furin in 90 µl of 0.1 M HEPES buffer (pH 7.4) containing 5 mM CaCl2. After 15 min to 5 h of incubation at 37 C, the proteolytic reaction was stopped by adding acetic acid (15%), and the samples were analyzed by RP-HPLC.

Immunodepletion studies
In some experiments, EN was immunodepleted of active IDE, cathepsin B, cathepsin D, or furin before the digestion step by incubating EN (0.30 mg/ml) with antibodies coated to protein G-Sepharose beads for 16 h at 4 C in 800 µl of 20 mM sodium phosphate buffer (pH 7), containing 0.1% Sarkosyl-30. The fractions were then centrifuged for 5 min at 10,000 x gav, and the resultant immunodepleted supernatants were used in the glucagon degradation assay. The reaction was terminated by the addition of 15% acetic acid and immediately assayed by RP-HPLC.

In other experiments, the [125I]iodoglucagon/170-kDa cross-linked complex was incubated with anti-IDE or anti-furin antibodies coated to protein G-Sepharose for 15 h at 4 C in 600 µl of 30 mM sodium phosphate buffer (pH 7), containing 0.1% Triton X-100 (Sigma-Aldrich). The fractions were then centrifuged for 5 min at 10,000 x gav, and the resultant supernatants were subjected to SDS-PAGE as described above.

HPLC separation of unlabeled glucagon peptides
RP-HPLC was performed on a Beckman Coulter (Fullerton, CA) System Gold model 127 liquid chromatograph equipped with a Rheodyne sample injector fitted with a 0.5-ml loop and a µBondapak C18 column (Waters, Milford, MA; 0.39 x 30 cm; 10-µm particle size). Samples were chromatographed using a mixture of 0.1% TFA in water (solvent A) and 0.1% TFA in acetonitrile (solvent B) with a flow rate of 1 ml/min. Elution was carried out using two sequential linear gradients of 0–25% solvent B (10 min) and 25–37% solvent B (40 min), followed by an isocratic elution of 37% solvent B (10 min). Eluates were monitored on-line for absorbance at 214 nm with a LC-166 spectrophotometer (Beckman Coulter). The major components in the eluates were collected and submitted to mass spectrometry analysis.

Mass spectrometry
Samples were prepared and analyzed using ion spray mass spectrometry and HPLC-electron spray ionization mass spectrometry coupling as previously described (21, 33).

Ligand radioiodination and preparation of purified [125I]Tyr-labeled isomers
Native glucagon and [Tyr12]-miniglucagon were radioiodinated by the lactoperoxidase method to specific activities of 60–80 µCi/µg as described previously (5, 7, 8, 27). Equimolar mixtures of [125I]Tyr10- and [125I]Tyr13-glucagon were purified by RP-HPLC as described previously (5). For [Tyr12]-miniglucagon, the iodination mixture was immediately loaded onto a RP-HPLC column (Waters µBondapak C18; 0.39 x 30 cm; 10-µm particle size) and chromatographed with a mixture of 0.1% TFA in water (solvent A) and 0.1% TFA in acetonitrile (solvent B) and a flow rate of 1.2 ml/min. Elution was carried out using a linear gradient of 0–50% over 60 min. Eluates were monitored on-line for radioactivity with a Berthold (Nashua, NH) LB 504 {gamma}-detector connected to an Apple IIc computer (Apple Computer, Cupertino, CA). The major component in the eluates (retention time, 52 min; see Fig. 8Go) was collected and freeze-dried.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 8. Covalent binding of [125I]Tyr12-miniglucagon to the 170-kDa endosomal protein. A, Tyr12-miniglucagon was iodinated using the lactoperoxidase method. The iodinated peptide was purified by RP-HPLC using a linear gradient of 0–50% acetonitrile. Under this condition, a major [125I]Tyr12-miniglucagon peptide was eluted with a retention time of 52 min. The dashed line indicates the position of native Tyr12-miniglucagon ({approx}42 min). B, [125I]Tyr12-miniglucagon (106 cpm) was cross-linked to proteins from the EN (0.4 mg/ml) using 50 mM EDC at pH 7 in the absence or presence of 10-5 M Tyr12-miniglucagon. The fractions were then subjected to SDS-PAGE in an 8% acrylamide resolving gel. Molecular mass markers are indicated to the left of the panel. The arrow indicates the mobility of the radioactive cross-linked complex ({approx}170 kDa).

 
Cross-linking studies
Affinity labeling of glucagon receptor by radiolabeled glucagon was performed as described by Authier et al. (27) with 1,4-difluoro-2,5-dinitrobenzene (DFDNB) (final concentration, 2 mM) as cross-linker. Affinity labeling of endosomal neutral glucagonase activity was performed using EDC as cross-linker. The EN (80 µg protein) was incubated in a final volume of 200 µl of 30 mM sodium phosphate buffer (pH 7), with 105–106 cpm of [125I]iodoglucagon or [125I]Tyr12-miniglucagon in the presence or absence of hormones, nucleotides, and protease inhibitors. After 30 min at 21 C, cross-linking was initiated by the addition of EDC to a final concentration of 10–50 mM and allowed to proceed for 30 min at 21 C. The cross-linking reaction was terminated by adding 2 µl of 1 M Tris-HCl (pH 7.4). The samples were heated for 1 min at 100 C in 1% sodium dodecyl sulfate and subjected to SDS-PAGE using an 8% acrylamide resolving gel according to the method of Laemmli (32).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Catalytic properties of endosomal neutral glucagonase activity
Figure 1Go shows the RP-HPLC elution profiles, measured by absorbance at 214 nm, of native glucagon peptide (A, HPLC profile A), and a mixture of glucagon and EN incubated for 60 min at pH 7 (A, HPLC profile B). Major intermediate peptide peaks were observed in addition to the undegraded glucagon peptide (peak 10), which had decreased in peak height (Fig. 1AGo, HPLC profile B).



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 1. Processing of glucagon at neutral pH by hepatic endosomes. A, Shown are representative RP-HPLC profiles resulting from the incubation of native glucagon (1 µM) with EN ({approx}1 µg) at 37 C for 60 min in 50 mM citrate-phosphate buffer (pH 7), in the absence (HPLC profile B) or presence of 1 mM PMSF (HPLC profile C) or 50 µg/ml bacitracin (HPLC profile D). All panels show absorbance profiles at 214 nm. Intact glucagon (G) had an elution time of 51 min (HPLC profile A). The endosomal proteins alone did not give any detectable peak (results not shown). Asterisks in HPLC profile C indicate absorbance peaks resulting from the presence of protease inhibitor in the incubation medium. B, EN ({approx}1 µg) was incubated with 1 µM native glucagon at 37 C for 60 min in 50 mM citrate-phosphate buffer (pH 7), in the absence or presence of 1 mM PMSF, 1 mM EDTA, 1 mM 1,10-phenanthroline, 5 mM EGTA, 0.1 µM E64, 1% Me2SO (DMSO), 5 µg/ml pepstatin-A, or 50 µg/ml bacitracin. At the end of the incubation, the proteolytic reaction was stopped with acetic acid (15%), and the incubation mixtures were analyzed by RP-HPLC as described in Materials and Methods. The rate of glucagon proteolysis was determined by following the disappearance of the peak area corresponding to the parent peptide. The results are expressed as glucagon degraded (percentage of control) and normalized to that seen in the absence of added compound. The results are the mean ± SD of three to five different experiments performed on ENs prepared from separate liver fractionations.

 
We evaluated the effect of various protease inhibitors on the neutral glucagon-degrading activity contained in hepatic endosomes (Fig. 1Go, A, HPLC profiles C and D, and B). The glucagonase activity was strongly inhibited (>88%) by the serine protease inhibitor PMSF (HPLC profile C) and the peptide antibiotic bacitracin (HPLC profile D). Metal-chelating agents such as EDTA, 1,10-phenanthroline, and EGTA had no significant effect.

Hepatic endosomes were next assessed for their ability to degrade native glucagon peptide in vitro at acidic (pH 4 and 5) and neutral pH (pH 7) (Fig. 2AGo). Degradation of glucagon was pH dependent with maximal degradation obtained at pH 4. However, a significant rate of hydrolysis of glucagon at pH 7 was observed and suggested the involvement of a specific endosomal neutral glucagonase activity (Fig. 2AGo).



View larger version (39K):
[in this window]
[in a new window]
  		FIG. 2. Assessment of endosomal neutral glucagonase activity. A, EN was incubated with 10-6 M glucagon for the indicated times at 37 C and pH 7, 5, or 4. The proteolytic reaction was stopped with acetic acid (15%), and the incubation mixtures were analyzed by RP-HPLC. Results are expressed as the amount of peptide degraded (percentage of control) at the end of incubation. Results are the mean of two (pH 4 and 5) and seven (pH 7) different experiments performed on ENs prepared from separate liver fractionations. B, EN was immunodepleted of active IDE (anti-IDE), cathepsin B (anti-CB), cathepsin D (anti-CD), or furin (anti-Furin) using monoclonal (IDE) or polyclonal antibodies (CB, CD, and furin) coated to protein G-Sepharose beads. After centrifugation, the resultant immunodepleted supernatants were tested for their ability to degrade native glucagon at pH 7 and 4 using RP-HPLC. The results were normalized (100%) to that seen in the absence of antibody. The results are the mean ± SD of three different experiments performed on ENs prepared from separate liver fractionations, except for the immunodepletion experiments done at pH 4 with anti-IDE antibody and at pH 7 with anti-CB and -CD antibodies, which are mean ± half-variation (n = 2).

 
Hepatic ENs are known to contain neutral and acidic peptidases such as IDE (10), cathepsin D (8, 21), cathepsin B (8, 22), ß-site APP cleaving enzyme (21), and furin (34, 35) as well as undefined tripeptidylaminopeptidase activity (8). We therefore used well-characterized antibodies to assess the neutral and acidic endosomal glucagonase activities (Fig. 2BGo). As expected, more than 50% of endosomal acidic glucagonase activity was depleted by anti-cathepsin B or anti-cathepsin D antibodies (8). However, immunoprecipitation with antibodies to cathepsin B, cathepsin D, IDE, or furin failed to remove the glucagonase activity observed at pH 7 (Fig. 2BGo).

Sites of cleavage of endosomal glucagon intermediates generated at neutral pH
Each of the major RP-HPLC peaks (see Fig. 1AGo, HPLC profile B) was analyzed using mass spectrometry to determine the molecular mass of peptide products (Table 1Go). Glucagon was cleaved at its N terminus at Ser2-Gln3, Gln3-Gly4, and Gly4-Thr5 bonds, and C terminus at Trp25-Leu26 and Leu26-Met27 bonds, thereby releasing the C-terminal peptides Gln3-Thr29 (peak 10), Gly4-Thr29 (product 9b), and Thr5-Thr29 (product 9c), and the N-terminal peptide His1-Trp25 (product 4a). Products 4b (Gln3-Trp25) and 6 (Gln3-Leu26) correspond to the above peptide products cleaved at both the N- and C-terminal regions. Other products revealed cleavage sites in the central region of glucagon between residues Lys12 and Tyr13, Asp15 and Ser16, Arg17 and Arg18, and Arg18 and Ala19, as evidenced by fragments Tyr13-Arg17 (peak 1), His1-Lys12 (peak 2), His1-Arg18 (product 3a), His1-Arg17 (product 3b), Arg18-Thr29 (peak 7), and Ala19-Thr29 (peak 8) (Table 1Go).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Masses and assigned structures of the cleavage products generated from native glucagon by endosomal neutral glucagonase activity

 
Identification of MGE as a mitochondrial protein
MGE, a neutral thiol-metalloendopeptidase of 100 kDa, has been proposed to process glucagon to its bioactive fragment glucagon-(19–29) (15, 16). Consequently, we attempted to evaluate the contribution of MGE to the endosomal degradation of glucagon at neutral pH by assessment of the content of MGE in hepatic endosomes using immunoblot analysis (Fig. 3Go). Established protocols were followed to isolate from rat liver homogenate (Hm) nuclear (N), large-granule (ML), microsomal (P), and cytosolic (S) fractions (Fig. 3AGo). Equal amounts of protein from these fractions were evaluated for their MGE content using a polyclonal antiserum produced against the N-terminal sequence REARPTTE of rat MGE (16) as described in Materials and Methods. The large-granule (ML) fraction contained the highest concentration of antigen (100 kDa), although immunoreactivity was also found in the low-speed nuclear (N) fraction (Fig. 3AGo). As shown in Fig. 3BGo, the 100-kDa polypeptide was not observed when the antiserum was preincubated with the antigenic peptide. Hence, the 100-kDa polypeptide detected in the ML fraction was not due to nonspecific binding.



View larger version (9K):
[in this window]
[in a new window]
  		FIG. 3. Assessment of MGE by immunoblot analysis of subcellular fractions isolated from rat liver. A, Nuclear (N), mitochondrial-lysosomal (ML), microsomal (P), and cytosolic (S) fractions, as well as the isotonic homogenate (Hm) were evaluated by immunoblotting for their content of MGE using polyclonal antiserum made to a synthetic peptide of MGE (CREARPTTE). Thirty micrograms of protein were loaded onto each lane. B, The ML fraction was evaluated by immunoblotting for its content of MGE using the polyclonal anti-MGE peptide antiserum, which had been preincubated (+) or not (-) with the antigenic synthetic peptide CREARPTTE. Thirty micrograms of protein were loaded onto each lane. C, The ML fraction (Total) was partitioned using Triton X-114 into the aqueous (Aq) and detergent (Det) phases according to the procedure of Bordier (31 ). Samples were then processed for SDS-PAGE with 150 µg of protein loaded in lane Total. Gels were subjected to SDS-PAGE in an 8% acrylamide resolving gel, transferred to nitrocellulose, and immunoblotted with rabbit antiserum as described in Materials and Methods. Molecular mass markers are indicated to the left of each panel. The arrows indicate the mobility of immunoreactive MGE ({approx}100 kDa).

 
Phase partitioning with the detergent Triton X-114 was carried out on the ML fraction (Fig. 3CGo). Immunoblot analysis of the Triton X-114-extracted ML fraction using polyclonal anti-MGE revealed that the 100-kDa polypeptide preferentially partitioned into the aqueous phase.

Consequently, organelles expected to sediment in the ML fraction, i.e. plasma membrane (PM), endosomes (EN), endoplasmic reticulum (ER), mitochondria (Mit), peroxisomes (Per), and lysosomes (Lys) were partially purified using established procedures and evaluated for their content of MGE as assessed by immunoblotting with the polyclonal antiserum anti-MGE (Fig. 4Go). Immunoreactivity was restricted to the mitochondrial (Mit) fraction with none detectable in the plasma membrane (PM) fraction. Control immunoblots for the marker enzymes Hsp-60, thiolase (Th), and cathepsin D (CD) revealed the expected immunoreactivity, respectively, in mitochondria (Mit), peroxisomes (Per), and the endolysosomal compartment (Fig. 4Go).



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 4. Association of MGE with the mitochondrial fraction of rat liver. Plasma membrane (PM), endosomal (EN), rough endoplasmic reticulum (ER), mitochondrial (Mit), peroxisomal (Per), and lysosomal (Lys) fractions were evaluated by immunoblotting for their content of MGE, mitochondrial heat-shock protein 60 (Hsp 60), peroxisomal 3-ketoacyl-CoA thiolase (Th), and endolysosomal cathepsin D (CD). Sixty micrograms of protein were loaded onto each lane. Gels were subjected to SDS-PAGE in an 8% ({alpha}-MGE) or 12% ({alpha}-Hsp 60, {alpha}-Th, and {alpha}-CD) acrylamide resolving gel, transferred to nitrocellulose, and immunoblotted with their respective polyclonal antibodies as described in Materials and Methods. Molecular mass markers are indicated to the left of the upper blot. The arrows indicate the mobility of immunoreactive MGE ({approx}100 kDa), Hsp 60 ({approx}60 kDa), Th ({approx}41 kDa), and CD (64 kDa for the proenzyme and 45 kDa for the mature enzyme).

 
Using a variety of immunofluorescence protocols, the polyclonal anti-MGE antibody did not give any immunolocalization above that of nonimmune serum (results not shown). Hence, we were unable to evaluate morphologically the intracellular localization of MGE.

Identification of an endosomal glucagon-binding protein by affinity labeling
Cross-linking protocols have been successfully used to identify and characterize proteolytic activities degrading polypeptides such as insulin (25), glucagon (8), and epidermal growth factor (EGF) (22). Consequently, a cross-linking approach was attempted to identify the relevant protease(s) responsible for the endosomal proteolysis of glucagon at neutral pH. Chemical-affinity labeling was performed using the zero-length water-soluble cross-linker EDC after incubation of the EN with [125I]iodoglucagon for 30 min at 21 C and pH 7 (Fig. 5Go). Autoradiographic analysis of the cross-linked proteins following SDS-PAGE under nonreducing conditions revealed the presence of a 170-kDa protein (Fig. 5AGo, lane -). Under reducing conditions, the relative mobility of the 170-kDa protein was not altered (Fig. 5AGo, lane 2-mercaptoethanol). The binding of [125I]iodoglucagon to the 170-kDa protein was completely inhibited with 10-5 M unlabeled glucagon (Fig. 5AGo, lane glucagon), indicating specific binding of the glucagon peptide to the 170-kDa protein. The ability of unlabeled glucagon to compete for [125I]iodoglucagon cross-linking to the 170-kDa protein was dose dependent with labeling reduced by 22% at 10-7 M and 85% at 10-6 M (Fig. 5BGo). In contrast, unlabeled insulin (Fig. 5Go, A, lane insulin, and B) and GTP (A, lane GTP) had no discernible effect on the efficiency of cross-linking.



View larger version (38K):
[in this window]
[in a new window]
  		FIG. 5. Detection of an endosomal glucagon-binding protein by affinity labeling. A, [125I]Iodoglucagon (105–106 cpm) was cross-linked to proteins from the EN (0.4 mg/ml) with 20 mM EDC at pH 7 in the absence (-) or presence of 10-5 M glucagon, 1 mM GTP, or 10-5 M insulin. The fractions were then either reduced in 2% ß-mercaptoethanol (lane 2-mercaptoethanol) and/or subjected to SDS-PAGE in an 8% acrylamide resolving gel. B, Cross-linking was performed as described above in the absence (lanes -) or presence of 10-7–10-5 M glucagon and 10-6–10-5 M insulin. C, Cross-linked proteins (300 µg protein) were incubated with N-glycanase, and the samples were subjected to electrophoresis in an 8% acrylamide resolving gel. Molecular mass markers are indicated to the left of each panel. The arrows indicate the mobility of the intact ({approx}170 kDa) and deglycosylated ({approx}140 kDa) radioactive cross-linked complex.

 
Digestion of the radiolabeled cross-linked complex with N-glycanase reduced the mobility of the iodinated complex to an apparent molecular mass of 140 kDa (Fig. 5CGo). Hence, the 170-kDa glucagon-binding protein was N-glycosylated.

The 170-kDa endosomal protein displays affinity for bacitracin and is unrelated to glucagon receptor, IDE, and furin
Binding of [125I]iodoglucagon to the 170-kDa endosomal protein was competitively inhibited by the nonspecific protease inhibitor bacitracin (Fig. 6Go). Bacitracin reduced cross-linking by 92% at 5 µg/ml (Fig. 6BGo), whereas other protease inhibitors were not as effective at reducing the labeling (A). Hence, the 170-kDa protein likely corresponds to a bacitracin-sensitive endosomal glucagonase.



View larger version (49K):
[in this window]
[in a new window]
  		FIG. 6. Effect of protease inhibitors on the binding of [125I]iodoglucagon to the 170-kDa endosomal protein. [125I]Iodoglucagon (105–106 cpm) was cross-linked to proteins from the EN (0.4 mg/ml) with 20 mM EDC at pH 7 in the absence (lanes -) or presence of 0.1 mM N-ethylmaleimide (NEM), 0.1 mM 1,10-phenanthroline, 5 µg/ml pepstatin-A, 0.1 mM p-hydroxymercuribenzoate (pHMB), 50 µg/ml bacitracin, or 5 mM EGTA (A), and in the absence (lane -) or presence of bacitracin at the indicated concentrations (B). The fractions were then subjected to SDS-PAGE in an 8% acrylamide resolving gel. Molecular mass markers are indicated to the left of each panel. The arrows indicate the mobility of the radioactive cross-linked complex ({approx}170 kDa).

 
Previous studies have described chemical cross-linking experiments using DFDNB as cross-linker to study the hepatic glucagon receptor (27). Accordingly, an iodinated complex of approximately 57 kDa was identified when cross-linking was done using an EN prepared from glucagon-injected rats (Fig. 7AGo, endosomes, lane -) or a PM fraction prepared from control rats (A, plasma membrane, lane -). In contrast to the 170-kDa endosomal glucagon-binding protein, cross-linking of [125I]iodoglucagon to the 57-kDa glucagon receptor was partially (Fig. 7AGo, endosomes) or totally inhibited (A, plasma membrane) by GTP. Moreover, bacitracin did not inhibit cross-linking of [125I]iodoglucagon to the glucagon receptor in the plasma membrane fraction (Fig. 7AGo). Hence, the 170-kDa endosomal glucagon-binding protein is structurally and functionally distinct from the glucagon receptor.



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 7. The 170-kDa endosomal protein is unrelated to glucagon receptor, IDE, and furin. A, Cross-linking of [125I]iodoglucagon to hepatic glucagon receptor. [125I]Iodoglucagon (106 cpm) was cross-linked to proteins from the EN (1 mg/ml), isolated from rats injected with 15 µg glucagon per 100 g body weight, using 2 mM DFDNB as the cross-linker in the absence (Endosomes, lane -) or presence of 0.1 mM GTP (Endosomes, lane GTP). Alternatively, [125I]iodoglucagon (106 cpm) was cross-linked to proteins from the plasma membrane fraction (0.1 mg/ml), prepared from noninjected rats, using 2 mM DFDNB in the absence (lane -) or presence of 0.1 mM GTP or 50 µg/ml bacitracin. The fractions were then subjected to SDS-PAGE in a 12% acrylamide resolving gel. Molecular mass markers are indicated to the left of the panel. The arrow indicates the mobility of the radioactive cross-linked complex ({approx}57 kDa). B, Effect of immunodepletion of IDE and furin on the endosomal 170-kDa polypeptide. [125I]Iodoglucagon (106 cpm) was cross-linked to proteins from the EN (0.5 mg/ml) with 20 mM EDC at pH 7. The fractions were then immunodepleted of IDE ({alpha}-IDE) or furin ({alpha}-Furin) in the presence of 0.1% Triton X-100. After immunoprecipitation and centrifugation, the resultant supernatants were subjected to SDS-PAGE in an 8% acrylamide resolving gel. Molecular mass markers are indicated to the left of the panel. The arrow indicates the mobility of the radioactive cross-linked complex ({approx}170 kDa). C, Assessment of endosomal furin by immunoblot analysis. The EN (50 µg) was evaluated by immunoblotting for its content of furin with polyclonal antibody (lane 1). Alternatively, native glucagon (30 µg) was incubated with the EN (100 µg of protein) in sodium phosphate buffer (pH 7). After 30 min at 21 C, cross-linking was performed using 20 mM EDC as described in Materials and Methods. The fraction was then subjected to SDS-PAGE in an 8% acrylamide resolving gel, transferred to nitrocellulose, and immunoblotted with anti-furin antibody (lane 2). Molecular mass markers are indicated to the left of the panel. The arrow indicates the mobility of the radioactive cross-linked complex ({approx}90 kDa).

 
Neutral proteases displaying a lower molecular mass than the 170-kDa cross-linked complex have been already identified in hepatic endosomes. Thus, we investigated whether EDC cross-linked glucagon to a multiprotein complex harboring polypeptides of a molecular mass lower than 170 kDa, such as IDE (110 kDa) and furin (85 kDa), using immunodepletion protocols (Fig. 7BGo). No detectable change in the 170-kDa cross-linked complex was observed after immunodepletion with antibodies to either furin or IDE (Fig. 7BGo, cf. lanes 2 and 3 with lane 1), ruling out IDE and furin as the endosomal glucagon-binding protein. Finally, to test whether furin was present in the EDC cross-linked complex, proteins from noniodinated glucagon/170-kDa polypeptide cross-linked complex were analyzed on Western blots using anti-furin antibody (Fig. 7CGo). Endosomal proteins alone (lane 1) or treated with glucagon and EDC (lane 2) exhibited an immunoreactive band at approximately 90 kDa with no immunoreactivity at 170 kDa. In addition, in vitro incubation of glucagon at 37 C and pH 7 with furin (28 U/ml) in the presence of 5 mM CaCl2 followed by RP-HPLC analysis did not reveal any detectable degradation of glucagon even up to 5 h of incubation (results not shown), suggesting that the endosomal endopeptidase is distinct from furin.

To investigate whether miniglucagon was a potential ligand for the 170-kDa endosomal glucagon-binding protein, we used a synthetic peptide corresponding to the 11 amino acids of the (19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29) miniglucagon peptide with an additional C-terminal tyrosine residue (designated Tyr12-miniglucagon) (Fig. 8Go). Iodination of Tyr12-miniglucagon was conducted using the lactoperoxidase method and the iodinated peptide was purified using the RP-HPLC method (Fig. 8AGo). [125I]Tyr12-miniglucagon (elution time, 52 min) was well resolved from native Tyr12-miniglucagon (elution time, 42 min) and free iodine (elution time, 4 min). The radiolabeled Tyr12-miniglucagon was then tested as a potential ligand for affinity labeling of the 170-kDa endosomal glucagon-binding protein using the cross-linker EDC (Fig. 8BGo). Cross-linking of the 170-kDa endosomal protein was observed (Fig. 8BGo, lane 1) and the cross-linking could be blocked with an excess of unlabeled Tyr12-miniglucagon (10-5 M) (B, lane 2). A diffuse radiolabeled band of 100–105 kDa was also observed and was not reduced by addition of unlabeled Tyr12-miniglucagon, suggesting that it may represent a nonspecifically bound protein(s).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Experiments using semipurified endosomal vesicles containing various internalized ligands have shown that proteolysis can be observed at acidic as well as neutral pH (reviewed in Ref. 36). These data demonstrate that proteases with a neutral pH optimum are present in endosomes. Thus, it has been proposed that ATP-dependent acidification of hepatic endosomes was not required for the initial step of insulin degradation, and that IDE, a neutral metalloendopeptidase, may operate in the endosomal lumen before the progressive decrease in endosomal pH (10). Using hepatic endosomes, Renfrew and Hubbard (37) have identified truncated forms of EGF that were generated by a carboxypeptidase B-like and a trypsin-like endosomal protease at neutral pH. Finally, using the plant toxin ricin A chain, Blum et al. (38) have identified an endosomal cathepsin that catalyzed toxin proteolysis at pH 4.5 and 7, and was primarily responsible for toxin cytotoxicity. Our results extend these observations to glucagon and suggest that endosomal processing of internalized ligands mediated by endosomal proteases with different pH optima may be regulated by changes in endosomal pH.

According to the sequences of peptide products, the main proteolytic activity participating in endosomal glucagon degradation at neutral pH can be attributed to an endopeptidase activity that generates eight major glucagon metabolites. Thus, endosomal proteolysis of glucagon at pH 7 resulted in three major cleavages in the N-terminal region (i.e. Ser2-Gln3, Gln3-Gly4, and Gly4-Thr5) as evidenced by the glucagon fragments (3–29), (4–29), and (5–29), and in the central region (i.e. Lys12-Tyr13, Arg17-Arg18, and Arg18-Ala19) as evidenced by the glucagon fragments (1–12), (1–17), (18–29), (1–18), and (19–29). Based on the location of the major cleavage sites, our data suggest that Ser2-Thr5 and Ser16-Ala19 may contain at least a portion of the residues important for the binding of glucagon to the neutral endosomal glucagonase. Other minor sites of cleavage were also identified at the C-terminal region (Trp25-Leu26 and Leu26-Met27 peptide bonds) and have been previously identified within hepatic endosomes at acidic pH (pH 4) (8). Consequently, we cannot be sure whether these two minor cleavages arise from the endosomal neutral glucagonase activity or from endosomal acidic cathepsins B and D, which may still remain active at neutral pH (22). In either case, it is clear that the endosomal neutral glucagonase activity mainly favors cleavage of glucagon at polar amino acid residues (N-terminal region) and basic amino acid residues (central region), and that it exhibits additional specificity that limits further proteolysis of the C-terminal (19–29) bioactive degradation product.

It remains unclear as to how the endosomal (19–29) fragment would come into contact with the plasma membrane Ca2+-Mg2+-adenosine triphosphatase. Changes in the biological activity of molecules following processing in endosomes have previously been observed for PTH-(1–84) (Ref. 39), as well as endocytosed protein antigens for major histocompatibility class II presentation (40), with bioactive fragments released from the cells without delivery to lysosomes. Interestingly, activation of PTH-(1–84) to its bioactive peptide PTH-(1–34), a potent inhibitor of the (Ca2+-Mg2+)-adenosine triphosphatase activity (as is miniglucagon) at a concentration of 100 nM (41), is mediated in endosomes by cathepsin D and released to the extracellular medium within 10–15 min (39). Our previous work on the glucagon receptor in rat liver showed that internalized endosomal glucagon receptors were recycled back to the cell surface with changes in glucagon-binding activity in plasma membrane completely reversed in less than 2 h (27). Thus, a retroendocytosis pathway, previously described for various ligands such as insulin (42), may also apply to the endosome-associated miniglucagon. In support of this hypothesis, biochemical assays have shown that plasma membranes were able to fuse with early endosomes containing previously internalized ligands (43). Interestingly, using circular dichroism and fluorescence spectroscopy, it has been shown that the Ca2+-binding capacity of glucagon is maintained in the (19–29) fragment, and that both peptides display some ability to penetrate the lipid bilayer in a hydrophobic environment with the bound Ca2+ ion (44). Thus, translocation of the (19–29) bioactive glucagon fragment across the endosomal membrane, comparable to that observed for bacterial and plant toxins (45), cannot be formally excluded. Additionally, the endocytic breakdown pathway may also represent an intermediate stage before endolysosomal transfer and subsequent lysosomal degradation of glucagon metabolites.

Cross-linking has been shown to be effective in the detection of non-covalently associated radiolabeled substrate-protease complexes (8, 25). Thus, following incubation of cytosol with [125I]iodoinsulin (25) or [125I]iodoglucagon (8) at neutral pH in the presence of disuccinimidyl suberate or bis(sulfosuccinimidyl) suberate (BS3), radiolabeled ligands were found in cross-linked IDE complexes of 110 kDa. Using a similar approach, we have previously reported the affinity labeling at acidic pH of [125I]iodo-EGF and [125I]iodoglucagon to endosomal cathepsin B using BS3 as the cross-linker (8, 22). We found that both disuccinimidyl suberate and BS3 were ineffective at cross-linking [125I]iodoglucagon to endosomal neutral glucagonase or to glucagon receptor at pH 7 (results not shown). However, we did detect a 170-kDa endosomal glucagon-binding protein using the zero-length water-soluble cross-linker EDC, which catalyzes the formation of peptide bonds between amino and carboxyl groups. This chemical cross-linker has been used extensively to identify polypeptides that are in close proximity such as skeletal essential light chain-1 and F-actin (46) or profilin and ß/{gamma}-actin (47). The 170-kDa affinity-labeled protein we have identified in this present study is likely to be the endosomal neutral glucagonase, because 1) [125I]iodoglucagon bound to the protein with a high affinity (IC50 {approx} 5 x 10-7 M) and specificity, and 2) the binding of [125I]iodoglucagon to the protein was completely abolished by a low concentration of bacitracin, a known inhibitor of endosomal neutral glucagonase. Other characteristics of the 170-kDa glucagon-binding protein include the following: 1) lack of sensitivity to GTP, 2) sensitivity to N-glycanase, 3) capacity to specifically bind the radiolabeled [Tyr12]-miniglucagon derivate, and 4) inability to be linked covalently to radiolabeled glucagon using DFDNB as the cross-linker. With the exception of N-glycosylation, all of these characteristics clearly distinguish the 170-kDa glucagon-binding protein from the glucagon receptor.

In agreement with our demonstration of the presence of a neutral basic processing activity within hepatic endosomes, the dibasic substrate human proinsulin and the monobasic substrate chicken proalbumin were shown to be processed, respectively, at the Arg-Arg and Arg-Phe-Ala-Arg sites in Triton X-100-solubilized liver endosomal extracts at pH 6 (34). In these studies, proinsulin and proalbumin processing was attributed to Ca2+-dependent furin, a subtilisin-like endoprotease that processes a number of proteins in liver with a preference for dibasic residues preceded by an additional basic residue at the -4 position (48). In intracellular vesicles, the same transmembrane serine protease was shown to cleave internalized Pseudomonas exotoxin A after Arg279, generating the active 37-kDa carboxyl-terminal fragment that translocates from endocytic vesicles to the cytosol (49). A role for furin in the intracellular processing of endocytosed Diphtheria toxin at Arg190-Val191-Arg192-Arg193 and anthrax toxin at Arg164-Lys165-Lys166-Arg167 has also been demonstrated (50). Finally, immunoblotting has detected furin in rat liver Golgi-ENs (this study; Ref. 35), and endogenous furin has been found within early endosomes of various cells (51). These data are consistent with the endocytosis of cell surface furin to early endosomes (52).

The endosomal glucagon-degrading enzyme described here differs from furin proteolytic activity in several respects: 1) the endosomal neutral glucagonase activity was not inhibited by metal-chelating reagents such as EDTA and EGTA, whereas Ca2+ was strictly required for furin activity; 2) quantitative immunodepletion of endosomal furin failed to reduce neutral glucagonase activity; 3) no degradation products were observed when pure active furin was incubated with glucagon under conditions for which cleavage of various furin substrates were reported (53); 4) furin recognition sequences (-Arg-X-Lys/Arg-Arg- or -Lys/Arg-X-X-X-Lys/Arg-Arg-) are not present in the glucagon molecule, especially a basic residue at the -4 or -6 positions (54); and 5) with the exception of the Arg18-Ala19 cleavage site, the other five major cleavages do not occur at a dibasic peptide bond. In addition, the possibility that the 170-kDa glycoprotein that was linked covalently to [125I]iodoglucagon represented furin was ruled out by the following: 1) its molecular mass (170 kDa) being distinct from that of furin, which is reported to be approximately 85–90 kDa by SDS-PAGE as shown by ourselves in the present study and others (53); 2) the absence of an effect with EGTA-treated endosomes on the 125I labeling of the 170-kDa complex when compared with experiments omitting this furin inhibitor; and 3) the failure to immunoprecipitate the 170-kDa glycoprotein using anti-furin antibody.

Using similar criteria, other glucagon-degrading enzymes previously proposed as being physiologically relevant to the mechanism of glucagon clearance from liver and plasma serum differ from the neutral glucagonase described here. Thus, the neutral glucagon-processing enzyme differs from the following: 1) IDE, a metalloendopeptidase of 110 kDa that generates glucagon degradation products that are structurally different from those described in the present study (9) and 2) the serine protease dipeptidyl peptidase IV, which is relatively insensitive to PMSF (55), displays a molecular mass of 105 kDa in rat hepatocytes (56), and removes the N-terminal dipeptides His1-Ser2 and Gln3-Gly4 from native glucagon (14). However, other members of the subtilisin-like proprotein convertase (PCs) (i.e. PC1, PC2, paired basic amino acid converting enzyme 4, PC4, PC5, and PC7) may be putative candidates, but their presence in the endosomal compartment of hepatic parenchyma and their catalytic activity toward the mature glucagon hormone have not been investigated (57, 58).

Cathepsin B (Enzyme Commission no. 3.4.22.1), a rat liver endosomal/lysosomal cysteine protease that processes at acidic pH internalized glucagon (8), EGF (22), and presumably IGF-I (59), is active in early endosomes between pH 4 and 7 (38). It has both endo- and exopeptidase activity, with the exopeptidase activity optimal at pH 5 or less, and the endopeptidase activity maximal at pH 7 (36). Although this protease was responsible for proteolytic cleavage of ricin A chain in macrophage endosomes at both neutral and acidic pH (38), our results indicate that its proteolytic activity toward the glucagon molecule was restricted to acidic pH as indicated by the following observations: 1) lack of inhibition of endosomal neutral glucagonase by the cysteine protease inhibitor E64, a characteristic inhibitor of cathepsin B (22); and 2) lack of cross-reactivity of endosomal neutral glucagonase toward a specific cathepsin B antibody (22).

A rat liver membrane protease that cleaves glucagon and produces glucagon-(19–29) has been previously isolated from plasma membrane fractions (16). The purified enzyme displayed a molecular mass of 100 kDa, was inhibited by both sulfhydryl-modifying reagents and chelating agents, and was present at highest levels in the liver (15, 16). The N-terminal sequence of the purified protein was shown to be distinct from that of IDE (16). Using a rabbit antiserum directed against the N-terminal sequence of rat liver MGE (see Materials and Methods and Ref. 16), we show in the present study that MGE is absent from hepatic plasma membranes using immunoblot analysis, but present in the low-speed particulate N and ML fractions and in highly purified mitochondria. Thus, in contrast to the report of Blache et al. (16), we were unable to confirm the presence of MGE in hepatic plasma membranes. In the study of Blache et al. (16), no comparison was made with MGE present in other liver subcellular fractions suggesting that the plasma membrane-MGE association may have derived from organelle contamination. Moreover, Blache et al. (16) did not demonstrate that the N-terminal sequence obtained from the purified protein and used to raise the rabbit antiserum did in fact come from the protease, because the antibody to this peptide could not precipitate the proteolytic activity. Studies investigating the fate of iodinated and colloidal gold-labeled glucagon in hepatocytes using quantitative electron microscope autoradiography (3, 4) have never reported endocytosed glucagon in mitochondria, whereas an endosomal location for internalized glucagon has been clearly established using both morphological (3, 4) and biochemical approaches (5). Whatever the nature of the 100-kDa protein recognized by both the antisera used in our study and that of Blache et al. (16), our demonstration of the mitochondrial location of the antigenic protein suggests a function related to mitochondrial physiology that is clearly distinct from that of degrading polypeptide hormones such as glucagon.

In summary, we have identified an endosomal membrane serine glucagonase that binds specifically to glucagon and bacitracin, and transforms glucagon to glucagon-(19–29) at neutral pH. In various studies, the purification and subsequent characterization of bacitracin-sensitive proteases have successfully been made using affinity chromatography of proteolytic activities on bacitracin-agarose (60). Remarkably, bacitracin-agarose has been extensively used for affinity chromatography of serine proteinases, such as homologs of eubacterial subtilisins from Halobacterium mediterranei (61). Development of such purification protocols should enable us to address the physiological significance of the bacitracin-sensitive endosomal serine glucagonase in the intracellular metabolism of glucagon. Moreover, the presence of glycosylated side chains predicts that immobilized lectins will also be a useful tool in the purification of the 170-kDa endosomal glucagon-binding protein.


   Footnotes
 
Abbreviations: BS3, Bis(sulfosuccinimidyl) suberate; CoA, coenzyme A; DFDNB, 1,4-difluoro-2,5-dinitrobenzene; E64, L-3-carboxy-2,3-trans-epoxypropionyl-leucylamido(4-guanidino)butane; EDC, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide; EGF, epidermal growth factor; EN, endosomal fraction; IDE, insulin-degrading enzyme; MGE, miniglucagon endopeptidase; PC, proprotein convertase; PMSF, phenylmethylsulfonyl fluoride; RP-HPLC, reversed phase-HPLC; TFA, trifluoroacetic acid.

Received April 30, 2003.

Accepted for publication August 22, 2003.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Jelinek LJ, Lok S, Rosenberg GB, Smith RA, Grant FJ, Biggs S, Bensch PA, Kuijper JL, Sheppard PO, Sprecher CA, O’Hara PJ, Foster D, Walker KM, Chen LHJ, McKernan PA, Kindsvogel W 1993 Expression cloning and signaling properties of the rat glucagon receptor. Science 259:1614–1616[Abstract/Free Full Text]
  2. Exton JH 1988 Mechanisms of action of glucagon. In: Cooke BA, King RJB, van Der Molen HJ, eds. Hormones and their actions. Part II. Berlin: Springer Verlag; 231–264
  3. Watanabe J, Kanai K, Kanamura S 1988 Glucagon receptors in endothelial and Kupffer cells of mouse liver. J Histochem Cytochem 36:1081–1089[Abstract]
  4. Barazzone P, Gorden P, Carpentier JL, Orci L, Freychet P, Canivet B 1980 Binding, internalization, and lysosomal association of 125I-glucagon in isolated rat hepatocytes. A quantitative electron microscope autoradiographic study. J Clin Invest 66:1081–1093
  5. Authier F, Janicot M, Lederer F, Desbuquois B 1990 Fate of injected glucagon taken up by rat liver in vivo. Biochem J 272:703–712[Medline]
  6. Buggy JJ, Heurich RO, MacDougall M, Kelley KA, Livingston JN, Yoo-Warren H, Rossomando AJ 1997 Role of the glucagon receptor COOH-terminal domain in glucagon-mediated signaling and receptor internalization. Diabetes 46:1400–1405[Abstract]
  7. Authier F, Desbuquois B 1991 Degradation of glucagon in isolated liver endosomes. Biochem J 280:211–218
  8. Authier F, Mort JS, Bell AW, Posner BI, Bergeron JJM 1995 Proteolysis of glucagon within hepatic endosomes by membrane-associated cathepsins B and D. J Biol Chem 270:15798–15807[Abstract/Free Full Text]
  9. Rose K, Savoy LA, Muir AV, Davies JG, Offord RE, Turcatti G 1988 Insulin proteinase liberates from glucagon a fragment known to have enhanced activity against Ca2+ + Mg2+-dependent ATPase. Biochem J 256:847–851[Medline]
  10. Hamel FG, Mahoney MJ, Duckworth WC 1991 Degradation of intraendosomal insulin by insulin-degrading enzyme without acidification. Diabetes 40:436–443[Abstract]
  11. Authier F, Posner BI, Bergeron JJM 1994 Hepatic endosomes are a major physiological locus of insulin and glucagon degradation in vivo. In: Ciechanover A, Schwartz AL, eds. Cellular proteolytic systems. Vol 15. New York: Wiley-Liss; 89–113
  12. Hagopian WA, Tager HS 1984 Receptor binding and cell-mediated metabolism of [125I]monoiodoglucagon by isolated canine hepatocytes. J Biol Chem 259:8986–8993[Abstract/Free Full Text]
  13. Sheetz MJ, Tager HS 1988 Characterization of a glucagon receptor-linked protease from canine hepatic plasma membranes. J Biol Chem 263:19210–19217[Abstract/Free Full Text]
  14. Hinke SA, Pospisilik JA, Demuth HU, Mannhart S, Kühn-Wache K, Hoffmann T, Nishimura E, Pederson RA, McIntosh CHS 2000 Dipeptidyl peptidase IV (DPIV/CD26) degradation of glucagon. Characterization of glucagon degradation products and DPIV-resistant analogs. J Biol Chem 275:3827–3834[Abstract/Free Full Text]
  15. Blache P, Kervran A, Dufour M, Martinez J, Le-Nguyen D, Lotersztajn S, Pavoine C, Pecker F, Bataille D 1990 Glucagon-(19–29), a Ca2+ pump inhibitory peptide, is processed from glucagon in the rat liver plasma membrane by a thiol endopeptidase. J Biol Chem 265:21514–21519[Abstract/Free Full Text]
  16. Blache P, Kervran A, Le-Nguyen D, Dufour M, Cohen-Solal A, Duckworth W, Bataille D 1993 Endopeptidase from rat liver membranes, which generates miniglucagon from glucagon. J Biol Chem 268:21748–21753[Abstract/Free Full Text]
  17. Mallat A, Pavoine C, Dufour M, Lotersztajn S, Bataille D, Pecker F 1987 A glucagon fragment is responsible for the inhibition of the liver Ca2+ pump by glucagon. Nature 325:620–622[CrossRef][Medline]
  18. Pavoine C, Brechler V, Kervran A, Blache P, Le-Nguyen D, Laurent S, Bataille D, Pecker F 1991 Miniglucagon [glucagon-(19–29)] is a component of the positive inotropic effect of glucagon. Am J Physiol 260:C993–C999
  19. Dalle S, Smith P, Blache P, Le-Nguyen D, Le Brigand L, Bergeron F, Ashcroft FM, Bataille D 1999 Miniglucagon (glucagon 19–29), a potent and efficient inhibitor of secretagogue-induced insulin release through a Ca2+ pathway. J Biol Chem 274:10869–10876[Abstract/Free Full Text]
  20. Dalle S, Fontés G, Lajoix AD, Le Brigand L, Gross R, Ribes G, Dufour M, Barry L, Le-Nguyen D, Bataille D 2002 Miniglucagon (glucagon 19–29). A novel regulator of the pancreatic islet physiology. Diabetes 51:406–412[Abstract/Free Full Text]
  21. Authier F, Métioui M, Fabrega S, Kouach M, Briand G 2002 Endosomal proteolysis of internalized insulin at the C-terminal region of the B chain by cathepsin D. J Biol Chem 277:9437–9446[Abstract/Free Full Text]
  22. Authier F, Métioui M, Bell AW, Mort JS 1999 Negative regulation of epidermal growth factor signaling by selective proteolytic mechanisms in the endosome mediated by cathepsin B. J Biol Chem 274:33723–33731[Abstract/Free Full Text]
  23. Shii K, Roth RA 1986 Inhibition of insulin degradation by hepatoma cells after microinjection of monoclonal antibodies to a specific cytosolic protease. Proc Natl Acad Sci USA 83:4147–4151[Abstract/Free Full Text]
  24. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ 1951 Protein measurement with the folin phenol reagent. J Biol Chem 193:265–275[Free Full Text]
  25. Authier F, Rachubinski RA, Posner BI, Bergeron JJM 1994 Endosomal proteolysis of insulin by an acidic thiol metalloprotease unrelated to insulin degrading enzyme. J Biol Chem 269:3010–3016[Abstract/Free Full Text]
  26. Neville DM 1968 Isolation of an organ specific protein antigen from cell surface membrane of rat liver. Biochim Biophys Acta 154:540–552[Medline]
  27. Authier F, Desbuquois B, de Galle B 1992 Ligand-mediated internalization of glucagon receptors in intact rat liver. Endocrinology 131:447–457[Abstract/Free Full Text]
  28. Walter P, Blobel G 1983 Preparation of microsomal membranes for cotranslational protein translocation. Methods Enzymol 96:84–93[Medline]
  29. Bodnar AG, Rachubinski RA 1991 Characterization of the integral membrane polypeptides of rat liver peroxisomes isolated from untreated and clofibrate-treated rats. Biochem Cell Biol 69:499–508[Medline]
  30. Wattiaux R, Wattiaux-De Coninck S, Ronveaux-Dupal MF, Dubois F 1978 Isolation of rat liver lysosomes by isopycnic centrifugation in a metrizamide gradient. J Cell Biol 78:349–368[Abstract/Free Full Text]
  31. Bordier C 1981 Phase separation of integral membrane proteins in Triton X-114 solution. J Biol Chem 256:1604–1607[Abstract/Free Full Text]
  32. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685[CrossRef][Medline]
  33. Authier F, Danielsen GM, Kouach M, Briand G, Chauvet G 2001 Identification of insulin domains important for binding to and degradation by endosomal acidic insulinase. Endocrinology 142:276–289[Abstract/Free Full Text]
  34. Brennan SO, Peach RJ 1991 The processing of human proinsulin and chicken proalbumin by rat hepatic vesicles suggests a convertase specific for X-Y-Arg-Arg or Arg-X-Y-Arg sequences. J Biol Chem 266:21504–21508[Abstract/Free Full Text]
  35. Alarcon C, Cheatham B, Lincoln B, Kahn CR, Siddle K, Rhodes CJ 1994 A Kex2-related endopeptidase activity present in rat liver specifically processes the insulin proreceptor. Biochem J 301:257–265
  36. Berg T, Gjøen T, Bakke O 1995 Physiological functions of endosomal proteolysis. Biochem J 307:313–326
  37. Renfrew CA, Hubbard AL 1991 Sequential processing of epidermal growth factor in early and late endosomes of rat liver. J Biol Chem 266:4348–4356[Abstract/Free Full Text]
  38. Blum JS, Fiani ML, Stahl PD 1991 Proteolytic cleavage of ricin A chain in endosomal vesicles. Evidence for the action of endosomal proteases at both neutral and acidic pH. J Biol Chem 266:22091–22095[Abstract/Free Full Text]
  39. Diment S, Martin KJ, Stahl PD 1989 Cleavage of parathyroid hormone in macrophage endosomes illustrates a novel pathway for intracellular processing of proteins. J Biol Chem 264:13403–13406[Abstract/Free Full Text]
  40. Chapman HA 1998 Endosomal proteolysis and MHC class II function. Curr Opin Immunol 10:93–102[CrossRef][Medline]
  41. McKenzie R, Lotersztajn S, Pavoine C, Pecker F, Epand RM, Orlowski RC 1990 Inhibition of the calcium pump by human parathyroid hormone-(1–34) and human calcitonin in liver plasma membranes. Biochem J 266:817–822[Medline]
  42. Duckworth WC, Bennett RG, Hamel FG 1998 Insulin degradation: progress and potential. Endocr Rev 19:608–624[Abstract/Free Full Text]
  43. Mayorga LS, Diaz R, Stahl PD 1988 Plasma membrane-derived vesicles containing receptor-ligand complexes are fusogenic with early endosomes in a cell-free system. J Biol Chem 263:17213–17216[Abstract/Free Full Text]
  44. Brimble KS, Ananthanarayanan VS 1993 Calcium binding and translocation properties of glucagon and its fragments. Biochemistry 32:1632–1640[CrossRef][Medline]
  45. Falnes PO, Sandvig K 2000 Penetration of protein toxins into cells. Curr Opin Cell Biol 12:407–413[CrossRef][Medline]
  46. Andreev OA, Saraswat LD, Lowey S, Slaughter C, Borejdo J 1999 Interaction of the N-terminus of chicken skeletal essential light chain 1 with F-actin. Biochemistry 38:2480–2485[CrossRef][Medline]
  47. Nyman T, Page R, Schutt CE, Karlsson R, Lindberg U 2002 A cross-linked profilin-actin heterodimer interferes with elongation at the fast-growing end of F-actin. J Biol Chem 277:15828–15833[Abstract/Free Full Text]
  48. Nakayama K 1997 Furin: a mammalian subtilisin/Kex2p-like endoprotease involved in processing of a wide variety of precursor proteins. Biochem J 327:625–635
  49. Ogata M, Fryling CM, Pastan I, FitzGerald DJ 1992 Cell-mediated cleavage of Pseudomonas exotoxin between Arg279 and Gly280 generates the enzymatically active fragment which translocates to the cytosol. J Biol Chem 267:25396–25401[Abstract/Free Full Text]
  50. Gordon VM, Klimpel KR, Arora N, Henderson MA, Leppla SH 1995 Proteolytic activation of bacterial toxins by eukaryotic cells is performed by furin and by additional cellular proteases. Infect Immun 63:82–87[Abstract]
  51. Molloy SS, Thomas L, Kamibayashi C, Mumby MC, Thomas G 1998 Regulation of endosome sorting by a specific PP2A isoform. J Cell Biol 142:1399–1411[Abstract/Free Full Text]
  52. Schafer W, Stroh A, Berghofer S, Seiler J, Vey M, Kruse ML, Kern HF, Klenk HD, Garten W 1995 Two independent targeting signals in the cytoplasmic domain determine trans-Golgi network localization and endosomal trafficking of the proprotein convertase furin. EMBO J 14:2424–2435[Medline]
  53. Chiron MF, Fryling CM, FitzGerald DJ 1994 Cleavage of pseudomonas exotoxin and diphtheria toxin by a furin-like enzyme prepared from beef liver. J Biol Chem 269:18167–18176[Abstract/Free Full Text]
  54. Nakayama K, Watanabe T, Nakagawa T, Kim WS, Nagahama M, Hosaka M, Hatsuzawa K, Kondoh-Hashiba K, Murakami K 1992 Consensus sequence for precursor processing at mono-arginyl sites. Evidence for the involvement of a Kex2-like endoprotease in precursor cleavages at both dibasic and mono-arginyl sites. J Biol Chem 267:16335–16340[Abstract/Free Full Text]
  55. Mentlein R 1999 Dipeptidyl-peptidase IV (CD26)-role in the inactivation of regulatory peptides. Regul Pept 85:9–24[CrossRef][Medline]
  56. Loch N, Tauber R, Becker A, Hartel-Schenk S, Reutter W 1992 Biosynthesis and metabolism of dipeptidylpeptidase IV in primary cultured rat hepatocytes and Morris hepatoma 7777 cells. Eur J Biochem 210:161–168[Medline]
  57. Steiner DF 1998 The proprotein convertases. Curr Opin Chem Biol 2:31–39[CrossRef][Medline]
  58. Seidah NG, Prat A 2002 Precursor convertases in the secretory pathway, cytosol and extracellular milieu. Essays Biochem 38:79–94[Medline]
  59. Navab R, Chevet E, Authier F, Di Guglielmo GM, Bergeron JJ, Brodt P 2001 Inhibition of endosomal insulin-like growth factor-I processing by cysteine proteinase inhibitors blocks receptor-mediated functions. J Biol Chem 276:13644–13649[Abstract/Free Full Text]
  60. Stepanov VM, Rudenskaya GN 1983 Proteinase affinity chromatography on bacitracin-Sepharose. J Appl Biochem 5:420–428[Medline]
  61. Stepanov VM, Rudenskaya GN, Revina LP, Gryaznova YB, Lysogorskaya EN, Filippova IY, Ivanova II 1992 A serine proteinase of an archaebacterium, Halobacterium mediterranei. A homologue of eubacterial subtilisins. Biochem J 285:281–286



This article has been cited by other articles:


Home page
 EndocrinologyHome page
G. Fontes, A.-D. Lajoix, F. Bergeron, S. Cadel, A. Prat, T. Foulon, R. Gross, S. Dalle, D. Le-Nguyen, F. Tribillac, et al.
Miniglucagon (MG)-Generating Endopeptidase, which Processes Glucagon into MG, Is Composed of N-Arginine Dibasic Convertase and Aminopeptidase B
Endocrinology, February 1, 2005; 146(2): 702 - 712.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Authier, F.
Right arrow Articles by Briand, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Authier, F.
Right arrow Articles by Briand, G.

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