HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 11, 9437-9446, March 15, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/11/9437    most recent M110188200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Authier, F. Articles by Briand, G. Search for Related Content PubMed PubMed Citation Articles by Authier, F. Articles by Briand, G. Social Bookmarking                      What's this?

Endosomal Proteolysis of Internalized Insulin at the C-terminal Region of the B Chain by Cathepsin D^*

François Authier^§, Mourad Métioui, Sylvie Fabrega, Mostafa Kouach^¶, and Gilbert Briand^¶

From the  INSERM U510, Faculté de Pharmacie Paris XI, 92296 Châtenay-Malabry, France and ^¶ Laboratoire de Spectrométrie de Masse, Faculté de Médecine, 59000 Lille, France

Received for publication, October 23, 2001, and in revised form, December 20, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The endosomal compartment of hepatic parenchymal cells contains an acidic endopeptidase, endosomal acidic insulinase, which hydrolyzes internalized insulin and generates the major primary end product A^1-21-B^1-24 insulin resulting from a major cleavage at residues Phe^B24-Phe^B25. This study addresses the nature of the relevant endopeptidase activity in rat liver that is responsible for most receptor-mediated insulin degradation in vivo. The endosomal activity was shown to be aspartic acid protease cathepsin D (CD), based on biochemical similarities to purified CD in 1) the rate and site of substrate cleavage, 2) pH optimum, 3) sensitivity to pepstatin A, and 4) binding to pepstatin A-agarose. The identity of the protease was immunologically confirmed by removal of greater than 90% of the insulin-degrading activity associated with an endosomal lysate using polyclonal antibodies to CD. Moreover, the elution profile of the endosomal acidic insulinase activity on a gel-filtration TSK-GEL G3000 SW_XL high performance liquid chromatography column corresponded exactly with the elution profile of the immunoreactive 45-kDa mature form of endosomal CD. Using nondenaturating immunoprecipitation and immunoblotting procedures, other endosomal aspartic acid proteases such as cathepsin E and -site amyloid precursor protein-cleaving enzyme (BACE) were ruled out as candidate enzymes for the endosomal degradation of internalized insulin. Immunofluorescence studies showed a largely vesicular staining pattern for internalized insulin in rat hepatocytes that colocalized partially with CD. In vivo pepstatin A treatment was without any observable effect on the insulin receptor content of endosomes but augmented the phosphotyrosine content of the endosomal insulin receptor after insulin injection. These results suggest that CD is the endosomal acidic insulinase activity which catalyzes the rate-limiting step of the in vivo cleavage at the Phe^B24-Phe^B25 bond, generating the inactive A^1-21-B^1-24 insulin intermediate.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Proteins entering the endocytic pathway encounter an increasingly hydrolytic environment imposed by a progressive decrease in pH and an increase in protease concentrations (reviewed in Ref. 1). Ultimately, most are degraded in lysosomes to small peptides and free amino acids. For some, degradation takes place early in the endocytic pathway. This is the case for polypeptide hormones such as insulin (2-4), glucagon (5), and parathyroid hormone (6) and growth factors such as the epidermal growth factor (7) and insulin-like growth factor-I (IGF-I)¹ (8) as well as endocytosed protein antigens for major histocompatibility class II presentation (9) and plant toxins (10).

In liver parenchyma the endosomal degradation of internalized insulin is thought to occur after acidification of the endosomal lumen by a soluble enzyme termed endosomal acidic insulinase (EAI) (2, 3, 11), although this protease has yet to be identified. EAI, which was easily extracted by hypotonic shock from hepatic endosomes, displayed an acidic pH optimum for insulin binding and degradation (2, 3, 11). The events leading to the endosomal proteolysis of internalized insulin by EAI involve initial binding of the C-terminal region of the B chain of insulin to EAI followed by two initial proteolytic cleavages at Phe^B24-Phe^B25 and Phe^B25-Tyr^B26 peptide bonds (11). This is followed by the hydrolysis of seven peptide bonds in the C terminus of the insulin B chain and, in the central region of the insulin A and B chains, via an ordered sequential pathway. The high IC₅₀ value of insulin for EAI (2-4 × 10⁵ M) and the high proteolytic activity and affinity of EAI for insulin B chain and a B^22-30 insulin fragment suggest that EAI is a nonspecific peptidase whose physiological substrates other than insulin within the endosomal apparatus have yet to be defined (11). In support of this, we have recently demonstrated that cathepsin B, a general cysteine protease, down-regulates internalized epidermal growth factor and glucagon within rat liver endosomes (5, 7) and internalized IGF-I in MCF-7 and H-59 tumor cells (8) by inducing proteolytic cleavages of these ligands.

Previous attempts to purify and identify EAI have been made using the radiolabeled peptide [¹²⁵I-Tyr^A14]insulin to measure endosomal hydrolysis of insulin by monitoring the increase in soluble counts after precipitation of radioactive insulin with cold trichloroacetic acid (2, 3). Based on the trichloroacetic acid precipitation assay, endosomal proteolysis of [¹²⁵I-Tyr^A14]insulin was partially inhibited by chelating agents, thiol reagents, pepstatin A, leupeptin, and bacitracin (2). However, none of the protease inhibitors was individually capable of completely inhibiting [¹²⁵I-Tyr^A14]insulin proteolysis, suggesting the contribution of more than one protease to the processing of radioactive insulin. Later, using an alternate approach based on reverse-phase (RP) HPLC and mass spectrometry analysis in which hydrolysis was measured by the generation of degradation products from unlabeled HI, we elucidated the entire structure of nine early insulin intermediates generated within hepatic endosomes, all of which contained the A¹⁴ tyrosine residue in their sequence and displayed a molecular mass higher than 3125 Da (11). Indeed, all of the early insulin intermediates identified in this study were not trichloroacetic acid-soluble (11). Consequently, the trichloroacetic acid solubility of radioactive products, which greatly underestimates actual degradation, reflects additional and/or further processing of transient insulin intermediates and their late transformation into short end products and/or free amino acids by endosomal proteases that are possibly distinct from EAI.

These concerns were addressed in this study by using native HI as the substrate and RP HPLC to improve the specificity of the degradation assay and identify the nature of EAI. An early time point of HI digestion was always chosen for RP HPLC analysis, when the two primary products A^1-21-B^1-24 and A^1-21-B^1-25 insulin were the most abundant. These improvements in the methodology used to evaluate the proteolytic activity have allowed us to identify the role of the aspartic acid protease cathepsin D in the endosomal processing of insulin.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Peptides, Antibodies, Protein Determination, Enzyme Assays, and Materials-- A chain and B chain from bovine insulin, human insulin, and bovine cathepsin D (EC 3.4.23.5), 15 units/mg, were from Sigma. Rabbit anti-mouse cathepsin D R291 (5), sheep anti-human cathepsin D M8147 (12, 13), rabbit anti-rat cathepsin B 7183 (5, 7, 14), and rabbit anti-rat cathepsin L R958 (5, 15) were obtained from Dr. J. S. Mort (Shriners Hospital for Crippled Children, Montreal, Quebec) and used to immune-deplete samples of native mature enzymes as described previously (5, 7). Rabbit polyclonal antisera directed against N- and C-terminal -site amyloid precursor protein-cleaving enzyme (BACE) peptides were obtained from Dr. R. W. Doms (University of Pennsylvania School of Medicine, Philadelphia, PA) and have been described previously (16). Affinity-purified goat antibody raised against human cathepsin E (sc-6508), which cross-reacts with cathepsin E of rat origin, was purchased from Santa Cruz Biotechnology Inc. Mouse monoclonal antibody 9B12 directed against the human insulin-degrading enzyme (17) was a kind gift from Dr. R. A. Roth (Stanford University, Stanford, CA). Polyclonal IgG against the human insulin receptor -subunit and mouse monoclonal anti-phosphotyrosine (clone 4G10) were purchased from Upstate Biotechnology Inc. Mouse monoclonal antibody directed against the rat early endosome antigen 1 (EEA1) was purchased from Transduction Laboratories. Horseradish peroxidase-conjugated goat anti-rabbit IgG and goat anti-mouse IgG were from Bio-Rad. The protein content of isolated fractions was determined by the method of Lowry et al. (18). N-Acetyl--D-glucosaminidase was assayed with p-nitrophenyl N-acetyl--D-glucosaminide as substrate according to Touster et al. (19). Acid phosphatase was assayed as described by Trouet (20). Nitrocellulose membranes and the Enhanced Chemiluminescence (ECL) detection kit were from Amersham Biosciences, Inc. Protein G-Sepharose was from Amersham Biosciences, Inc. CA074, which is a highly selective irreversible cathepsin B inhibitor, was purchased from Peptides International. Pepstatin A, pepstatin A-agarose, EDTA, 1,10-phenanthroline, E64, N-ethylmaleimide, p-hydroxymercuri phenylsulfonic acid, iodoacetamide, phenylmethylsulfonyl fluoride, and benzamidine were from Sigma. HPLC grade acetonitrile and trifluoroacetic acid were obtained from J. T. Baker Inc. All other chemicals were obtained from commercial sources and were of reagent grade.

Animals and Injections-- Male Sprague-Dawley rats, body weight 180-200 g, were obtained from Charles River France (St. Aubin Les Elbeufs, France) and were fasted for 18 h before sacrifice. Native insulin (15 µg/100 g of body weight) in 0.4 ml of 0.15 M NaCl was injected within 5 s into the penis vein under light anesthesia with ether. In some experiments, rats received an intraperitoneal injection of 2 mg of pepstatin A in 0.4 ml of 12.5% Me₂SO, 0.15 M NaCl 1 h before HI injection.

Isolation of Subcellular Fractions from Rat Liver-- Subcellular fractionation was performed using established procedures (2, 3, 5, 7, 11, 21-23). After injection of human insulin, animals were sacrificed, and livers were rapidly removed and minced in isotonic ice-cold homogenization buffer as previously described (2, 3, 5, 7, 11, 21-23).

Lysosomes were prepared by isopycnic centrifugation of the light mitochondrial fraction in a discontinuous metrizamide gradient according to the method of Wattiaux et al. (24). Lysosomes (L2 fraction) were collected at the 1.109-1.135 g/ml metrizamide interface (24). The endosomal fraction was isolated by discontinuous sucrose gradient centrifugation and collected at the 0.25-1.0 M sucrose interface (2, 3, 5, 7, 11, 21-23). The soluble extract from the endosomal fractions was isolated by freeze/thawing in 5 mM sodium phosphate, pH 7.4, and disrupted in the same hypotonic medium using a small Dounce homogenizer (15 strokes with Type A pestle) followed by centrifugation at 300,000 × g_av for 30 min as described previously (2, 3, 5, 7, 11, 25)_.

Endosomal fractions isolated after the injection of HI were suspended at 1.5 mg/ml in 62.5 mM Tris/HCl, pH 6.8, 2% SDS, 10% glycerol, 2% -mercaptoethanol and heated at 100 °C for 2 min. Samples were then subjected to SDS-PAGE, followed by Western blotting to determine the integrity and phosphotyrosine content of the internalized insulin receptor.

Immunoblot Analysis-- Electrophoresed samples were transferred to nitrocellulose membranes (0.45 × 10⁶ 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 or 2% bovine serum albumin (for phosphotyrosine immunoblots) 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, human BACE diluted 1:1000 or rat cathepsin L R958 diluted 1:200, affinity-purified polyclonal antibodies to insulin receptor diluted 1:1000, monoclonal anti-phosphotyrosine diluted 1:2500) in the above buffer for 16 h at 4 °C. After incubation, the blots were washed 3 times with 0.5% skim milk or 0.2% bovine serum albumin (for phosphotyrosine immunoblots) 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 immunoglobulin was detected using horseradish peroxidase-conjugated goat anti-rabbit IgG for the polyclonal antibody or horseradish peroxidase-conjugated goat anti-mouse IgG for the monoclonal anti-phosphotyrosine antibody.

In Vitro Proteolysis of Native Peptides-- Soluble endosomal extract (ENs) (1 ng) was incubated for varying lengths of time at 37 °C with 5 × 10⁵ M human insulin, insulin A chain, or insulin B chain in 200 × 10⁶ liters of 50 mM citrate-phosphate, pH 4, in the presence or absence of protease inhibitors. The samples were then acidified with acetic acid (15%) and immediately loaded onto a RP HPLC column. In some experiments, ENs was immunodepleted of active insulin-degrading enzyme, cathepsin L, cathepsin B, or cathepsin D before the digestion step by incubating ENs (0.15 mg/ml) with antibodies coated to protein G-Sepharose for 16 h at 4 °C in 800 × 10⁶ liters of 20 mM sodium phosphate buffer, pH 7. The fractions were then centrifuged for 5 min at 10,000 × g_av, and the resultant supernatants were adjusted to pH 4 with citrate-phosphate buffer and used in the HI degradation assay. The reaction was terminated by the addition of 15% acetic acid and immediately assayed by RP HPLC_.

Native HI, insulin A chain, and insulin B chain were also digested in vitro with bovine cathepsin D. Insulin peptides (5 × 10⁵ M) were incubated with 0.025 units/ml cathepsin D in 0.1 M citrate-phosphate, pH 4. After 5-60 min of incubation at 37 °C, the proteolytic reaction was stopped by adding acetic acid (15%), and the samples were immediately analyzed by RP HPLC.

HPLC Separation of Human Insulin, Insulin A Chain, and Insulin B Chain Peptides-- RP HPLC was performed on a Beckman Coulter System Gold model 127 liquid chromatograph equipped with a Rheodyne sample injector fitted with a 500-µl loop and a microBondapak C18 column (Waters, 0.39 × 30 cm; 10⁵ m particle size). Samples were chromatographed using a mixture of 0.1% trifluoroacetic acid in water (solvent A) and 0.1% trifluoroacetic acid in acetonitrile (solvent B) with a flow rate of 1 ml/min. Elution was carried out using two sequential linear gradients of 0-15% solvent B (5 min) and 15-39% solvent B (32 min) followed by an isocratic elution of 39% solvent B (13 min). Eluates were monitored on-line for absorbance at 214 nm with a LC-166 spectrophotometer (Beckman Coulter).

Binding of Endosomal Acidic Insulinase to Pepstatin A-agarose-- ENs (0.15 mg/ml) was incubated for 15 h at 4 °C with pepstatin A-agarose in 20 mM citrate-phosphate buffer, pH 7. After centrifugation at 20,000 × g_av for 5 min, the resultant supernatant was tested for its ability to degrade 10⁶ M HI in 300 mM citrate-phosphate buffer, pH 4, at 37 °C for varying amounts of time, and samples were immediately analyzed by RP HPLC.

Characterization of Endosomal Acidic Insulinase Using Gel-filtration HPLC-- ENs was loaded onto a TSK-GEL G3000 SW_XL HPLC column (Tosoh Corp., 0.78 × 30 cm) equilibrated at 4 °C with 50 mM sodium phosphate buffer, pH 6. The column was washed with 30 ml of sodium phosphate buffer, pH 6, using a flow rate of 0.5 ml/min. Eluates were monitored on-line for absorbance at 214 nm with an LC-166 spectrophotometer (Beckman Coulter). Each fraction (0.5 ml) was immediately adjusted to pH 4 with 0.5 M citrate-phosphate buffer and evaluated for insulin-degrading activity by incubating with 5 × 10⁵ M HI at 37 °C for 10 min. The reaction was stopped by the addition of 15% acetic acid, and samples were analyzed by RP HPLC or subjected to SDS-PAGE followed by Western blotting to determine their cathepsin D content. In some experiments, fractions eluted from the gel-filtration HPLC column that contained insulin-degrading activity with the highest specific activity (fraction 20; see Fig. 4A) were incubated for 15 h at 4 °C with pepstatin A-agarose in 75 mM citrate-phosphate buffer, pH 7. After centrifugation at 20,000 × g_av for 5 min, the bound proteins were subjected to SDS-PAGE and Western blot analysis.

Mass Spectrometry-- Samples were prepared and analyzed using ion spray mass spectrometry and HPLC electron spray ionization mass spectrometry coupling as previously described (11).

Hepatocyte Culture-- Primary hepatocytes were isolated from male Sprague-Dawley rats weighing 180-200 g. Hepatocytes were prepared by collagenase perfusion and purified by Percoll gradient centrifugation (26) as modified by Balavoine et al. (27). Viability was >85% by trypan blue exclusion. Hepatocytes were suspended in William's E medium (Invitrogen) supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, and 1% glutamine and plated on type-I collagen-coated glass coverslips.

Immunofluorescence-- Hepatocytes grown on glass coverslips were washed once with phosphate-buffered saline (PBS) before incubation with serum-free William's E medium containing 2 µM human insulin. After 20 min at 37 °C, cells were washed 3 times with PBS and fixed with 3% paraformaldehyde in PBS for 20 min. Fixed cells were treated with 100 mM NH₄Cl for 15 min, washed with PBS, permeabilized with 0.1% Triton X-100 in PBS for 4 min, and blocked by a 20-min incubation with 10% horse serum in PBS. Staining consisted of mouse monoclonal antibody to human insulin (diluted 1:100) (Sigma), mouse monoclonal antibody to rat EEA1 (diluted 1:200) (Transduction Laboratories), and rabbit polyclonal antibody to mouse cathepsin D R291 (diluted 1:100) as the primary reagents, followed by Alexa Fluor 488-conjugated goat anti-mouse (diluted 1:100) (Molecular Probes) and Texas Red-conjugated goat anti-rabbit (diluted 1:400) (Jackson Immunoresearch). Laser-scanning confocal microscopy was done with a Zeiss LSM 510 confocal (Axiovert 100 M) inverted microscope equipped with a Zeiss X63/1.4 NA oil immersion objective lens (plan-Apochromat). Fluorescent images were acquired with argon (wavelength 488 nm) and helium neon (wavelength 543 nm) lasers. Simultaneous images corresponding to Alexa Fluor 488 and Texas Red fluorescence were obtained using the multi-tracking function of the microscope.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Catalytic Properties of Endosomal Acidic Insulinase-- As a substrate, radiolabeled HI may lack many structural features of the natural EAI substrate that might be important in directing protease interaction and cleavage. Moreover, the trichloroacetic acid precipitation assay gives only limited information about the enzymatic mechanism of insulin degradation where conditions are far from the initial reaction rate. The most appropriate substrate for assaying EAI activity is native HI, which contains the physiological prerequisites for protease recognition. We therefore decided to reevaluate the effect of various protease inhibitors on the insulin-degrading activity contained in endosomes using a highly specific RP HPLC assay (Fig. 1). A short time of digestion was always chosen in order to follow the major primary cleavage of HI at residues Phe^B24-Phe^B25, which generates the hexapeptide B^25-30 (Fig. 1B, middle panel, peak 2) and the remaining N-terminal residues B^1-24 connected to the intact A chain A^1-21 (A^1-21-B^1-24; Fig. 1B, middle panel, peak 3) (11). Correct endoproteolytic cleavage of HI at the Phe^B24-Phe^B25 peptide bond was confirmed using electron spray ionization-mass spectrometry analysis (results not shown) (11). Native HI was then digested with ENs at pH 4 and 37 °C for 15 min in the presence of protease inhibitors (Fig. 1A). The protease activity was strongly inhibited (>95%) by pepstatin A, an inhibitor of aspartic acid proteases (Fig. 1, A and B, lower panel). No effect was observed with metallo-, cysteine-, and serine-protease inhibitors (Fig. 1A). From these results it was concluded that the primary cleavages of native HI were due to aspartic acid protease(s).

View larger version (44K): [in this window] [in a new window]   Fig. 1.   Effect of protease inhibitors on endosomal insulin-degrading activity. ENs (~1 ng) was incubated with 106 M native HI at 37 °C for 15 min in 50 mM citrate-phosphate buffer, pH 4, in the absence or presence of 0.1-5 µg/ml pepstatin A (PA), 1% Me2SO (DMSO), 1 mM EDTA, 1 mM 1,10-phenanthroline, 107 M E64, 107 M CA074, 104 M N-ethylmaleimide, 104 M p-hydroxymercuriphenylsulfonic acid, 104 M iodoacetamide, 1 mM phenylmethylsulfonyl fluoride (PMSF), or 1 mM benzamidine (panel A). 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 under "Experimental Procedures." The rate of HI proteolysis was determined by following the disappearance of the peak area corresponding to the parent peptide. The results are expressed as HI degraded (% of control) and normalized to that seen in the absence of added compound. The results are the mean ± S.D. of three to five different experiments performed on endosomal fractions prepared from separate liver fractionations. Panel B shows representative RP HPLC profiles resulting from the incubation of HI with ENs in the presence or absence of 0.1 µg/ml PA. All panels show absorbance profiles at 214 nm. Intact HI had an elution time of 48 min. The endosomal proteins alone did not give any detectable peak (results not shown).

Identification of Aspartic Acid Proteases Associated with Hepatic Endosomes-- Of the intracellular aspartic acid proteases characterized to date in mammalian cells, the most obvious candidates for our observed results are cathepsin D, cathepsin E, or BACE (28, 29). We were unable to detect cathepsin E in hepatic endosomes by Western blot analysis (results not shown), confirming the little or no detectable expression of the protease previously reported in rat liver parenchyma (29). ENs was then evaluated for its content of cathepsin D and BACE by immunoblotting with well characterized polyclonal antibodies (Fig. 2) (5, 16). The soluble hypotonic extract from endosomes (ENs) showed intense immunoreactivity for the 45-kDa mature cathepsin D enzyme and the 64-kDa cathepsin D proenzyme. By contrast, using anti-BACE antibodies directed against N- (-BACE N) and C-terminal (-BACE C) BACE peptides (16), BACE immunoreactivity (68-kDa) was restricted to endosomal membranes (ENm) with none detectable in the luminal content (ENs), confirming that BACE is an integral membrane protein (30).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Association of aspartyl proteases with endosomal fractions. Hepatic endosomes were disrupted by hypotonic shock. The lysate supernatants (ENs) from the 300,000 × gav × 60 min centrifugation and their respective membrane pellets (endosomal membranes (ENm)) were prepared, adjusted to the same volume, subjected to SDS-PAGE, transferred to nitrocellulose, and immunoblotted with rabbit polyclonal antisera against N-terminal (N) and C-terminal (C) human BACE peptides (16) or mouse liver CD (5). Molecular mass markers are indicated to the left of each panel.

The possibility of lysosome contamination of our endosomal fraction has been assessed in Fig. 3. We compared enzyme distribution in endosomal and lysosomal fractions of two lysosomal enzyme markers, acid phosphatase and N-acetyl--D-glucosaminidase (Fig. 3A). The endosomal fraction revealed no significant enrichment of N-acetyl--D-glucosaminidase (relative specific activity = 1.5) and acid phosphatase (relative specific activity = 2.2) with the yield of enzymes accounting for <0.2% that of homogenate. On the other hand, the L2 lysosomal fraction was more than 60 times enriched in lysosomal enzyme markers (Fig. 3A). Overall, the enrichment of radiolabeled [¹²⁵I]Tyr^A14 human insulin in endosomal fractions at the early time (4 min) of endocytosis (relative specific activity = 92.4 ± 6.38) was far greater than that of any of the enzymes assayed (results not shown; Ref. 3). As well, we used the well characterized anti-cathepsin antibodies to assess the distribution of precursor and mature forms of cathepsins in hepatic endosomes (EN) and lysosomes (L2) (Fig. 3B). The 37-kDa procathepsin L and 64-kDa procathepsin D proenzymes were exclusively identified in endosomes, whereas lysosomes showed strong immunoreactivity only for mature cathepsin L (29- and 22-kDa species) and cathepsin D (45- and 31-kDa species). The distinct distribution of procathepsin L to endosomes and mature 29-kDa single-chain and 22-kDa heavy-chain cathepsin L to lysosomes attests to the lack of contamination of our endosomal fraction by lysosomes (Fig. 3B). These data essentially agree with previously published studies (5, 21).

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Distribution of lysosomal enzymes in endosomal and lysosomal fractions from rat liver. Endosomal (EN) and lysosomal (L2) fractions were isolated from control rats. Panel A, acid phosphatase and N-acetyl--D-glucosaminidase activities were determined, and results are expressed as relative specific activity, i.e. specific activity found in the subcellular fraction/specific activity measured in the homogenate. Recoveries (% of the constituent present in the homogenate) of lysosomal marker enzymes in endosomal fraction were 0.17% ± 0.01 acid phosphatase and 0.113% ± 0.01 N-acetyl--D-glucosaminidase. The results are the mean ± S.D. of three to six separate experiments. Panel B, endosomal (EN) and lysosomal (L2) fractions were subjected to SDS-PAGE on a 12% acrylamide-resolving gel for the anti-CL immunoblot and 8% acrylamide-resolving gel for the anti-CD immunoblot, transferred to nitrocellulose, and immunoblotted with polyclonal antisera against mouse liver CD or rat cathepsin L (CL). Each lane contained 80 µg of endosomal protein or 5 µg of lysosomal protein. Molecular mass markers are indicated to the left of the panels. The arrows indicate the mobility of immunoreactive proforms and mature forms of cathepsins L and D.

Identification of Endosomal Acidic Insulinase as Cathepsin D-- The aspartic acid proteolytic activity associated with soluble endosomal proteins that generates the major primary cleavage Phe^B24-Phe^B25 after digestion of native HI was further purified on a TSK-GEL G3000 HPLC column (Fig. 4). Fractions eluted from the gel-filtration HPLC column were characterized for their insulinase activity at pH 4 using native HI followed by RP HPLC (Fig. 4A). In addition, fractions 13-25 were assayed for their content of cathepsin D by Western blotting using anti-cathepsin D antibody (Fig. 4B). The peak of acidic insulinase activity (fraction 20) coincided with elution of two cathepsin D immunoreactive polypeptides of 45 kDa (major peptide; fractions 19-21) and 31 kDa (minor peptide; fractions 20-21). The elution of the inactive 64-kDa cathepsin D proenzyme (fractions 16-18) was clearly separated from the mature active enzyme (Fig. 4B).

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Characterization of endosomal acidic insulinase activity by gel-filtration HPLC. ENs (230 µg) was applied to a TSK-GEL G3000 HPLC column. Panel A shows the absorbance profile at 214 nm. The eluted fractions were also tested for their ability to degrade 5 × 105 M HI at 37 °C and pH 4 and analyzed using RP HPLC. In panel B, fractions 13-25 were evaluated for their content of cathepsin D by immunoblotting with polyclonal anti-mouse cathepsin D antibody R291. Molecular mass markers are indicated to the left of panel B.

To determine whether other endosomal proteases in addition to aspartic acid proteases may participate in the primary cleavage of the insulin B chain at the Phe^B24-Phe^B25 peptide bond, ENs was incubated with pepstatin A-agarose at pH 7, and the absorbed proteins were precipitated by centrifugation (Fig. 5). Analysis of the bead supernatant showed that depletion of pepstatin A-binding proteins prevented proteolysis of native HI as measured at pH 4 and 15 or 30 min of incubation (Fig. 5A). To test whether pepstatin A-agarose bound nonspecifically to other endosomal proteins, we determined the protein concentration of the endosomal fractions before and after the pepstatin A-agarose precipitation step (Fig. 5B), and we could not detect any significant change in protein concentration. To gain more information regarding the specificity of the insulinase-pepstatin A interaction, the HPLC fraction from the TSK-GEL G3000 column with the highest acidic insulinase activity (fraction 20, see Fig. 4A) was incubated with pepstatin A-agarose beads (Fig. 5C). The bead precipitate was then subjected to SDS-PAGE and Western blotting using anti-mouse cathepsin D antibody (Fig. 5C, PA-ag pellet), revealing the presence of the two active forms of cathepsin D at 45 and 31 kDa. Hence, the EAI activity present in ENs likely corresponds to an active form of cathepsin D. 

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Affinity purification of a soluble endosomal pepstatin A-sensitive protease. Panel A, ENs was incubated for 15 h at 4 °C with pepstatin A-agarose in 20 mM citrate-phosphate buffer, pH 7. After centrifugation at 20,000 × gav for 5 min, the resultant supernatant was tested for its ability to degrade 106 M HI in 300 mM citrate-phosphate buffer, pH 4, for 15 or 30 min of incubation at 37 °C (PA-ag supernatant). The amount of protein recovered in the supernatant (PA-ag supernatant) as well as in the starting material (ENs) is presented in panel B. Panel C, the fraction eluted from the gel-filtration HPLC column that contained insulin-degrading activity with the highest specific activity (fraction 20; see Fig. 4A) was incubated with pepstatin A-agarose beads using the same procedure as described for panel A. Proteins associated with the pellet (lane PA-ag pellet) were eluted from the pepstatin A-agarose beads under denaturing conditions by suspending the beads in Laemmli buffer and heating at 100 °C for 5 min. The samples were then subjected to SDS-PAGE, transferred to nitrocellulose, and immunoblotted with anti-mouse cathepsin D antibody R291. The parent soluble endosomal fraction is also shown (lane ENs). Molecular mass markers are indicated to the left of panel C.

As further evidence, we used well characterized polyclonal antibodies to mature cathepsin D and its proform (5) to deplete cathepsin D from ENs (Fig. 6). Quantitative immunoprecipitation of cathepsin D using antibodies directed against the mouse and human enzyme removed greater than 90% of endosomal HI-degrading activity from ENs based on the Phe^B24-Phe^B25 peptide bond cleavage assay at acidic pH (Fig. 6, A and B). Immunodepletion of ENs with antibodies to cathepsins L and B and insulin-degrading enzyme failed to remove the insulinase activity as measured at pH 4 (Fig. 6A).

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Effect of immunodepletion of endosomal proteases on endosomal insulin-degrading activity. Panel A, ENs were immunodepleted of active insulin-degrading enzyme (-IDE), cathepsin L (-CL), cathepsin B (-CB), or cathepsin D (-CD) using monoclonal (insulin-degrading enzyme) or polyclonal antibodies (cathepsin L, cathepsin B, and CD) coated to protein G-Sepharose. After centrifugation, the resultant supernatants were tested for their ability to degrade native HI at pH 4 and analyzed by RP HPLC. The results were normalized to that seen in the absence of antibody and are the mean ± S.D. of four experiments. Panel B, ENs were immunodepleted of active cathepsin D using polyclonal antibodies directed against mouse and human cathepsin D, and the resultant supernatants were tested for their ability to degrade 5 × 105 M HI in 50 mM citrate-phosphate buffer, pH 4, after 20 min incubation at 37 °C. The proteolytic reaction was stopped with acetic acid (15%), and the samples were analyzed by RP HPLC. All panels show absorbance profiles at 214 nm.

Digestion of Insulin Peptides with Cathepsin D-- To determine whether cathepsin D is capable of cleaving HI at the same primary sites as those observed with ENs (11), HI was subjected to in vitro digestion with bovine cathepsin D at pH 4. The digestion products were separated by RP HPLC, and the resulting chromatogram is shown in Fig. 7 (lower panel, HI + CD). RP HPLC analysis revealed four proteolytic products (peaks 1-4) with retention times identical to those seen with ENs (upper panel, HI + ENs) (11). Each of the major HPLC peaks for HI plus CD was analyzed using mass spectrometry to determine the molecular masses of the peptide products. Table I lists the peptide peaks (HPLC pools), their retention times, theoretical and experimental molecular masses, and structures. Peak 2 corresponded to a mixture of the hexapeptide B^25-30 (755.5 Da) and pentapeptide B^26-30 (608.2 Da). Peak 4 was a mixture of the remaining N-terminal residues B^1-24 and B^1-25 connected to the intact A-chain A^1-21 (A^1-21-B^1-24, 5087 Da, and A^1-21-B^1-25, 5216 Da). Peak 3 (5069 Da) was the same as the A^1-21-B^1-24 peptide with an internal peptide bond cleaved. The HPLC pool 1 contained the B^27-30 peptide (445.2 Da), thereby indicating the removal of the Tyr^B26 residue from peptide 2a. The full-length HI substrate (peptide 4c; 5806 Da) coeluted with products 4a and b with a retention time of 48 min. Thus, HI was similarly cleaved by cathepsin D (Table I) and ENs (11) at the two primary sites, Phe^B24-Phe^B25 and Phe^B25-Tyr^B26, both located in the C terminus of the B chain. Digestion with human cathepsin D (not shown) gave identical results to that of bovine cathepsin D, indicating that the cleavage specificity of this enzyme is conserved across species.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   HPLC profiles of native human insulin digested in vitro with a soluble endosomal extract or cathepsin D. HI (5 × 105 M) was treated in vitro with ENs or 0.025 units/ml cathepsin D in 0.1 M citrate-phosphate buffer, pH 4. After 15 min (HI + ENs) or 60 min (HI + CD) at 37 °C, the proteolytic reaction was stopped by adding acetic acid (15%), and the hydrolysis products were immediately analyzed by RP HPLC. The major degradation products, pools 1-4, were collected from the elution profile HI + CD as indicated. These pools were subjected to mass spectrometry analyses (see Table I).

We have previously shown that the insulin B chain represents a high affinity substrate for EAI (11). Using RP HPLC analysis of digestions performed under initial rate conditions, we have compared the rate of hydrolysis of intact HI and the individual insulin A and B chains by ENs or pure cathepsin D (Fig. 8). As previously reported (11), ENs degraded insulin B chain more efficiently than for intact HI, whereas the rate of hydrolysis of insulin A chain was lower than that of HI (Fig. 8A). Pure cathepsin D was found to degrade insulin peptides in a manner similar to ENs (i.e. insulin B chain > HI > insulin A-chain, Fig. 8B). These results support the hypothesis that EAI and endosomal cathepsin D are the same enzyme.

View larger version (31K): [in this window] [in a new window]   Fig. 8.   Proteolysis of native human insulin, insulin A-chain, and insulin B chain by ENs or cathepsin D. Native HI, insulin A chain and insulin B chain were incubated in vitro with ENs (panel A) or 0.025 units/ml cathepsin D (panel B) in 0.1 M citrate-phosphate buffer, pH 4. After 5-120 min at 37 °C, the proteolytic reaction was stopped by adding acetic acid (15%), and the incubation mixture was analyzed by RP HPLC. The rate of hydrolysis of each peptide was determined by following the disappearance of the peak area, corresponding to the parent peptide. The results are the mean ± S.D. of three different experiments performed on endosomal fractions prepared from separate liver fractionations.

Intracellular Colocalization of Human Insulin with Cathepsin D-- To strengthen the physiological relevance of our observations obtained with cell-free endosomes, we studied the cellular localization of internalized HI and cathepsin D in intact hepatocytes by confocal microscopy coupled to immunostaining (Fig. 9). Hepatocytes were incubated with 2 × 10⁶ M HI for 20 min, when most of the internalized ligand would be located in the endosomes (3), and the subcellular localization of both HI and cathepsin D were revealed by indirect immunofluorescence (Fig. 9A). Antibody to HI (in green) demonstrated a highly punctate staining pattern reminiscent of vesicular compartments. Costaining with antibody directed against mature and precursor cathepsin D enzyme (in red) revealed a partial colocalization (in yellow) with the intracellular protease. HI co-localization did not overlap completely with intracellular cathepsin D, although the extent of signal overlap was quite significant (Fig. 9A).

View larger version (49K): [in this window] [in a new window]   Fig. 9.   Partial colocalization of internalized insulin and cathepsin D in hepatocytes. Hepatocytes were treated with 2 × 106 M HI for 20 min, fixed with paraformaldehyde, and permeabilized with Triton X-100 before staining. Panel A, HI is shown in green, cathepsin D is shown in red, and merged images on the far right indicate the extent of colocalization (yellow). Anti-HI antibody gave a typical endosomal vesicular pattern that colocalized partially with intracellular cathepsin D. Scale bar, 7.6 µm. Two cells with a typical fluorescent staining pattern are presented. Panel B, cathepsin D is shown in red, EEA1 is shown in green, and the merged image on the far right indicates the extent of colocalization (yellow). Scale bar, 8.5 µm. Fluorescent images were captured at two emission wavelengths (488 and 543 nm).

The localization of cathepsin D in endocytic vesicles was next examined by double-staining immunofluorescence with the early endosome marker EEA1 (Fig. 9B). A partial immunoreactivity of cathepsin D was observed in the early endocytic vesicles identified by the presence of EEA1. These results confirm a minor but genuine endosomal location for the lysosomal aspartic acid protease. Moreover, the previous biochemical and morphological demonstrations that internalized insulin in hepatocytes accumulated into endocytic components clearly distinguishable from lysosomes (reviewed in Ref. 31) combined with the biochemical (this study; Refs. 5 and 32) and morphological (this study; Ref. 33) demonstrations of the presence of cathepsin D within hepatic endosomes suggest that the partial colocalization of internalized insulin and cathepsin D observed in Fig. 9A occurred most probably within endocytic vesicles.

Effect of Pepstatin A on the Fate of Insulin Receptor within Hepatic Endosomes in Vivo-- The level of insulin receptor internalization and the tyrosine phosphorylation state of the internalized insulin receptor could potentially be affected by pepstatin A, which affects endosomal proteolysis of HI. Therefore, the effect of pepstatin A on endosomal insulin receptor signal transduction was next examined (Fig. 10). Animals were administered an intraperitoneal injection of either 12.5% Me₂SO, 0.15 M NaCl or 2 mg of pepstatin A diluted in 12.5% Me₂SO, 0.15 M NaCl 1 h before native HI administration, and animals were sacrificed 2-60 min after HI administration. Hepatic endosomes were prepared, and the amount of internalized insulin receptor and its phosphotyrosine content was determined by Western blot analyses. A time-dependent increase in insulin receptor content was observed in endosomal fractions isolated 2-30 min after administration of a single receptor-saturating dose of HI (15 µg/100 g of body weight) as described previously (3) (Fig. 10, lower panel). No difference in the time nor extent of receptor internalization was observed after pepstatin A treatment. In contrast, a difference was observed for tyrosine phosphorylation of the insulin receptor -subunit. Insulin administration led to a brief burst (2 min) of tyrosine autophosphorylation of the insulin receptor, and pepstatin A treatment could extend this time to 5 min (Fig. 10, upper panel).

View larger version (65K): [in this window] [in a new window]   Fig. 10.   Effect of pepstatin A treatment on the internalization of tyrosine-phosphorylated insulin receptor after administration of human insulin in vivo. Rats were injected with native HI (15 µg/100 g of body weight) having first received an intraperitoneal injection of either 12.5% Me2SO, 0.15 M NaCl or 2 mg of pepstatin A in 12.5% Me2SO, 0.15 M NaCl 1 h before ligand administration. Endosomal fractions were isolated at the indicated times and evaluated by Western blotting for their content of insulin receptor using polyclonal antibody to the insulin receptor -subunit (Anti-IR) and for their immunoreactivity to monoclonal antibody against phosphotyrosine (Anti-PY). Each lane contained 60 µg of endosomal protein. The arrows indicate the mobility of the -subunit of the insulin receptor (94 kDa). Molecular mass markers are indicated to the left of the panels.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The soluble aspartic acid protease cathepsin D has been identified as the insulin-degrading activity responsible for the primary cleavages of internalized insulin within hepatic endosomes. In this study, we sought to characterize the proteolytic function of endosomal cathepsin D toward the insulin substrate through a series of biochemical and morphological investigations. Using similar approaches, we have previously reported that hepatic endosomes contain the cysteine protease cathepsin B, that processes internalized glucagon (5), epidermal growth factor (7), and IGF-I (8). Identically, we have recently shown that CA074-methyl ester, a cathepsin B inhibitor, prevented endosomal processing of internalized IGF-I in MCF-7 and H-59 tumor cells (8). Consistent with our previous studies, we concluded in this present work that the EAI activity was likely the aspartic acid protease cathepsin D, as indicated by the following observations. (a) The inhibitor profile and the pH optimum of the endosomal proteolytic activity were similar to those of cathepsin D; (b) immunodepletion of cathepsin D from ENs led to a loss of EAI activity; (c) the EAI activity was specifically absorbed to pepstatin A-agarose; (d) Western blot experiments demonstrated the presence of the 45-kDa active form of cathepsin D as well as its 64-kDa inactive precursor in the endosomal lumen; (e) using confocal immunofluorescent microscopy, internalization of HI into rat hepatocytes led to partial colocalization of internalized insulin with cathepsin D and cathepsin D with the early endocytic marker EEA1; and (f) pure cathepsin D produced a cleavage pattern for HI that was similar to that generated using the endosomal proteolytic activity.

That the presence of active cathepsin D within hepatic endosomes was due to lysosomal contamination of our endosomal fraction is highly unlikely. Thus, the modest content of N-acetyl--D-glucosaminidase and acid phosphatase in our endosomal fraction points to minimal contamination by lysosomes and confirms our earlier observations (21). Moreover, mature lysosomal cathepsin L was undetectable in our endosomal fraction although its inactive precursor was easily identified. Finally, immunofluorescence studies performed on rat hepatocytes revealed that the endosomal marker protein EEA1 was present in some cathepsin D-positive structures, suggesting a genuine endosomal localization for the cathepsin D enzyme. These data conform well to previous biochemical (5, 32) and morphological evidence (33) that active cathepsin D was selectively retained within hepatic endosomes.

Although the radiolabeled peptide [¹²⁵I-Tyr^A14]HI can be used to assay insulin degradation at the endosomal locus (2-4), caution should be used when interpreting this cleavage activity as the true EAI activity (11). First, radioactive HI might not display all the primary, secondary, and tertiary structural elements of the natural peptide substrate (34). Thus, the radioactive iodine on the molecule might mask the substrate recognition site and the sites of cleavage for the endosomal proteolytic system. Secondly, although a trichloroacetic acid precipitation assay has been developed for the degradation analysis of radioactive insulin (2-4, 11), the limitation of this type of assay is that it greatly underestimates actual degradation and measures the proteolytic activities under conditions far from the initial rate of the reaction. In a complex system where intermediate cleavage products can be further processed, it is difficult to obtain any meaningful information on the enzyme of interest. The trichloroacetic acid precipitation assay is, therefore, inherently unsuitable to screening for a specific protease in endosomal lysates, where many other proteases with a broad substrate specificity are likely to be present (5, 7). Thus, we have developed an RP HPLC assay based on measurement of the initial degradation step of native HI by endosomal lysates, with the subsequent identification of the early insulin intermediates. The improved specificity of this method allowed for the rigorous screening of EAI activity. However, we cannot exclude that very late cleavages of small insulin intermediates observed within hepatic endosomes (11) may be produced by other endosomal endo- and/or exopeptidases such as cathepsin B (5, 7), insulin-degrading enzyme (35), or other as yet unidentified endosomal proteolytic activities.

Despite the limitations of the trichloroacetic acid precipitation assay, an accurate examination of the early proteolytic cleavages generated in vivo has been possible with the use of monoiodinated insulin isomers combined with RP HPLC and radiosequencing procedures (35-38). Using this approach, endosome-associated degradation products extracted at an early time of insulin endocytosis displayed intact A chain and cleavages in the B chain at B²⁴-B²⁵, B²³-B²⁴, B¹⁶-B¹⁷, and B¹⁴-B¹⁵ bonds (35-38). Structural and kinetic analyses of in vivo intermediates isolated from the intraendosomal compartment have allowed the determination of an ordered sequence involving an initial cleavage in the B chain at 24-25, which produces des(hexapeptide)insulin (38). This process has been supported by our in vitro study characterizing endosomal proteolysis of native HI using a soluble endosomal extract from rat liver parenchyma (11). This study showed initial B²⁴-B²⁵ cleavage with the parallel production of des(hexapeptide)-HI and the C-terminal B chain hexapeptide (B^25-30) (11). Therefore, we elected to pursue the in vitro characterization of EAI using the most sensitive assay for insulin degradation by following the primary in vivo cleavage between Phe^B24 and Phe^B25 residues.

The two primary end products of HI that are generated using cell-free endosomes (this study; Ref. 11) and pure cathepsin D (this study) result from proteolytic cleavages occurring at residues Phe^B24-Phe^B25 (major pathway) and Phe^B25-Tyr^B26 (minor pathway), two sites that fit the specificity pattern for sites cleaved by cathepsin D (39). Sequences readily cleaved by cathepsin D contain apolar or hydrophobic residues situated on both sides of the cleavage site, with the Phe-Phe bond strongly favored and, in most cases, a hydrophobic residue in position P₁ (39). Thus, the sites in HI that were cleaved to produce the two primary end products A^1-21-B^1-24 and A^1-21-B^1-25 satisfy all the requirements for a cathepsin D cleavage site. Accordingly, characterization of the residues in positions P₁ and P'₁, determined with synthetic substrates (39), explains the highly specific primary cleavages obtained using natural and physiological substrates by cathepsin D. Thus, cathepsin D appears to mediate the selective cleavage of parathyroid hormone at the Phe³⁴-Val³⁵ peptide bond (6), human invariant chain at the Leu¹⁷⁴-Phe¹⁷⁵ peptide bond (40), the A1 protein from myelin at the Phe⁴²-Phe⁴³ peptide bond (41), -1,4-N-acetylgalactosaminyltransferase at the Leu²³-Tyr²⁴ peptide bond (42), and -amyloid precursor protein at various hydrophobic peptide bonds such as between the Phe⁹³ and Phe⁹⁴ residues (43). Finally, kallistatin, a serpin that specifically inhibits human tissue kallikrein, was recently demonstrated to be cleaved at the Phe-Phe bond in its reactive site loop by cathepsin D (44).

Studies on the structural requirements for HI/EAI binding have also identified a highly specific molecular interaction at the C-terminal B²²-B³⁰ region of the insulin B chain, a region that contains the aromatic locus Phe^B24-Phe^B25-Tyr^B26 (11). Thus, the hydrophobic Phe^B24-Phe^B25-Tyr^B26 tripeptide might also dictate HI binding to cathepsin D. Unexpectedly, binding (11) and cleavage (8) of the insulin-related IGF-I polypeptide by EAI activity and aspartic acid proteases were not detected upon incubation of this growth factor with hepatic ENs and tumor cell lysates. One explanation might be the difference in primary sequence between positions B²⁴-B²⁶ (Phe-Phe-Tyr) of the insulin B chain and the corresponding positions 23-25 (Phe-Tyr-Phe) of IGF-I, which displays a simple inversion of the insulin Phe-Tyr sequence and lacks the unique Phe-Phe bond (45).

The present studies indicate that internalized insulin is a physiological substrate for the endosomal protease cathepsin D. Cathepsin D has been identified in endosomes of rabbit alveolar macrophages by biosynthetic studies and affinity chromatography purification (46), in hepatic endosomes by Western blotting (5, 32) and immunocytochemistry (33), in multivesicular endosomes of macrophages (47) and B-lymphoblastoid cells (48) by immunocytochemistry, and in the late endosomal Rab7-containing compartment by flow cytometric sorting (49). Consistent with previous studies on the subcellular distribution and molecular species of cathepsins in liver parenchyma (5, 32) and other studies on the biogenesis of cathepsin D (46, 49, 50), the cathepsin D species identified in endosomes consisted of both the inactive precursor and active mature forms. These data suggest that significant processing of lysosomal hydrolases may occur within endosomes (46). The molecular mechanisms responsible for targeting and retaining cathepsin D and its proform in endosomes are not clear. Morphological and biochemical analyses, which focused on the localization and targeting of the cation-independent mannose 6-phosphate receptor (CI-MPR) in stable BHK cell lines and one of its ligands, cathepsin D, have shown that a significant fraction of CI-MPR and cathepsin D traffic from the trans-Golgi network to late endosomes/lysosomes via early endosomes (51). Alternatively, mannose 6-phosphate receptor-independent membrane association of cathepsin D has been reported within acidic endosomes in macrophages (46), in HepG2 cells (50), in human B-lymphocytes (52), and in NIH 3T3 mouse fibroblasts (53).

In the present study, we provide also the first evidence for endosomal compartmentalization of the recently identified -secretase protein BACE in hepatocytes. By immunoblot analysis, we confirmed earlier biochemical and immunofluorescence studies that had localized BACE to endosomes, one of several intracellular sites where amyloid  is thought to be produced (28). The accumulation of BACE in the endosomal system appears to be governed largely by a cytoplasmic dileucine motif (16). Although the evidence is quite compelling that BACE is the -secretase in Alzheimer's disease (16, 28, 30), it has not been established whether other soluble or membrane-bound proteins, related or not to -amyloid precursor protein, are subject to endosomal cleavage by BACE. The previously reported transmembrane property of BACE (30), confirmed by us in this study, provides evidence against a role for BACE in the proteolysis of free insulin within the endosomal lumen. However, the type-I integral membrane protein BACE may be a good candidate protease for the endosomal degradation of membrane-associated peptides.

Our present work combined with our previous studies that provided the first evidence for endosomal compartmentalization of cathepsin B in hepatocytes (5, 7) is in total agreement with other studies suggesting that 20-40% of newly synthesized lysosomal hydrolases can be detected in early endosomes (54). Recently, cell fractionation studies performed on stable BHK cell lines reveal cathepsin D comigration with Rab5 and EEA1, established markers of early endosomes, as well as the sensitivity of cathepsin D to horseradish peroxidase/diaminobenzidine-mediated cross-linking of endosomal proteins (51). Our immunofluorescence studies confirm that a small amount of cathepsin D can be detected in early endocytic EEA1-positive vesicles. In the study of Press et al. (51), the potential significance of cathepsin D delivery to early endosomes is discussed in light of its involvement in cellular growth control and differentiation. Indeed, increased extracellular levels of various cathepsins, especially cathepsins D and B, have been correlated to metastasis and a number of human tumors (55). Thus, endosomal cathepsin D may modulate the amount of these molecules delivered to the cell surface versus the amount transported toward lysosomes (51). Even cell death was recently shown to be influenced by endosomal/lysosomal cathepsin D in a study showing its involvement in cell apoptosis induced by oxidative stress (56). Our studies provide important clues regarding another function of endosomal cathepsin D related to the proteolysis of internalized ligands, with the subsequent regulation of intracellular receptor signaling and trafficking at the endosomal locus (1, 3, 5).

Cathepsin D induces the proteolytic cleavage of several other polypeptide hormones, proteins, and plant toxins within endocytic vesicles. Thus, we have previously reported that internalized glucagon is partially processed within hepatic endosomes by the endopeptidase activity of cathepsin D (5). Also, proteolysis of mannose-bovine serum albumin and parathyroid hormone-(1-84), with the subsequent release of the bioactive peptide parathyroid hormone-(1-34), was primarily mediated by cathepsin D within macrophage endosomes (6). The participation of cathepsin D in the endosomal processing, membrane translocation, and cytotoxicity of ricin A chain has also been demonstrated (10). Studies have also suggested the potential role of cathepsin D in conjunction with cathepsins S and L in degrading the invariant chain and endocytosed antigens within antigen-presenting cells (57). Finally, some properties associated with cathepsin D favor its putative role in the amyloidogenic processing of -amyloid precursor protein within the endo-lysosomal compartment (43, 58).

In this study, we have shown that in vivo administration of pepstatin A, a compound that inhibits endosomal degradation of insulin in vitro, was without apparent effect on the level of insulin receptors present in endosomes but induced a more prolonged tyrosine autophosphorylation of the insulin receptor -subunit. The concentration of pepstatin A required to demonstrate this same phenomenon in vitro was in the order of 1 µg/ml and induced a complete inhibitory effect. The modest effect of pepstatin A observed in vivo probably reflects its low degree of cell and organelle penetrance (59, 60) and also its inability to affect the dissociation of internalized insulin from its endosomal receptor, which attenuates any further ligand-driven receptor re-phosphorylation (3). Moreover, after intravenous administration in rats, pepstatin A is rapidly cleared from the blood by the kidneys (61), confirming that it requires very high concentrations to be effective in vivo (60). Consequently, because cells are rather impermeable to pepstatin A, it was necessary to treat rats with high concentrations (>1 mg) in the presence of Me₂SO to obtain effective intracellular concentrations. The development of specific and potent inhibitors of CD activity will greatly advance the elucidation of the function of the endosomal proteolysis of internalized insulin.

Some studies point out that the defects in the metabolism of intracellular insulin observed in non-insulin-dependent diabetes mellitus could be localized within the endosomal apparatus and are caused mainly by a defective acidification of its interior (62, 63). Studies are under way to determine whether the pathological decrease in endosomal proteolysis of insulin observed in non-insulin-dependent diabetes mellitus may also be caused by alterations in the endosomal targeting or activity of the pepstatin A-sensitive CD activity.

	ACKNOWLEDGEMENTS

We thank Pamela H. Cameron (McGill University, Montreal, Quebec, Canada) for reviewing the manuscript. We thank Dr. R. A. Roth (Stanford University, Stanford, CA), Dr. J. S. Mort (Shriners Hospital for Crippled Children, Montreal, Quebec, Canada), Dr. R. W. Doms (University of Pennsylvania, Philadelphia, PA) for kind gifts of anti-insulin-degrading enzyme, -cathepsins, and -BACE antibodies, respectively. We thank Dr. F. Daniel and P. Perron (U327 INSERM, Paris, France) for assistance in the preparation of primary hepatocytes. We thank Dr. V. Nicolas (IFR 75 INSERM, Faculté de Pharmacie, Châtenay-Malabry, France) for assistance in confocal microscopy.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: INSERM U510, Faculté de Pharmacie Paris XI, 5 rue Jean-Baptiste Clément, 92296 Châtenay-Malabry, France. Tel.: 33.1.46835843; Fax: 33.1.46835844; E-mail: francois.authier@cep.u-psud.fr.

Published, JBC Papers in Press, January 4, 2002, DOI 10.1074/jbc.M110188200

	ABBREVIATIONS

The abbreviations used are: IGF-I, insulin-like growth factor I; BACE, -site amyloid precursor protein-cleaving enzyme; EAI, endosomal acidic insulinase; EEA1, early endosome antigen 1; ENs, soluble endosomal extract; HI, human insulin; RP HPLC, reverse-phase high pressure liquid chromatography; CD, cathepsin D; PBS, phosphate-buffered saline.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				X.-L. Chen, B.-B. Xie, F. Bian, G.-Y. Zhao, H.-L. Zhao, H.-L. He, B.-C. Zhou, and Y.-Z. Zhang Ecological Function of Myroilysin, a Novel Bacterial M12 Metalloprotease with Elastinolytic Activity and a Synergistic Role in Collagen Hydrolysis, in Biodegradation of Deep-Sea High-Molecular-Weight Organic Nitrogen Appl. Envir. Microbiol., April 1, 2009; 75(7): 1838 - 1844. [Abstract] [Full Text] [PDF]

				M. Yang, J. Sun, T. Zhang, J. Liu, J. Zhang, M. A. Shi, F. Darakhshan, M. Guerre-Millo, K. Clement, B. D. Gelb, et al. Deficiency and Inhibition of Cathepsin K Reduce Body Weight Gain and Increase Glucose Metabolism in Mice Arterioscler. Thromb. Vasc. Biol., December 1, 2008; 28(12): 2202 - 2208. [Abstract] [Full Text] [PDF]

				T. Tamura, P. Thotakura, T. S. Tanaka, M. S. H. Ko, and K. Ozato Identification of target genes and a unique cis element regulated by IRF-8 in developing macrophages Blood, September 15, 2005; 106(6): 1938 - 1947. [Abstract] [Full Text] [PDF]

				S. Diederich, M. Moll, H.-D. Klenk, and A. Maisner The Nipah Virus Fusion Protein Is Cleaved within the Endosomal Compartment J. Biol. Chem., August 19, 2005; 280(33): 29899 - 29903. [Abstract] [Full Text] [PDF]

				A. Balbis, G. Baquiran, V. Dumas, and B. I. Posner Effect of Inhibiting Vacuolar Acidification on Insulin Signaling in Hepatocytes J. Biol. Chem., March 26, 2004; 279(13): 12777 - 12785. [Abstract] [Full Text] [PDF]

				F. Authier, P. H. Cameron, C. Merlen, M. Kouach, and G. Briand Endosomal Proteolysis of Glucagon at Neutral pH Generates the Bioactive Degradation Product Miniglucagon-(19-29) Endocrinology, December 1, 2003; 144(12): 5353 - 5364. [Abstract] [Full Text] [PDF]

				B. Desbuquois, G. Chauvet, M. Kouach, and F. Authier Cell Itinerary and Metabolic Fate of Proinsulin in Rat Liver: In Vivo and in Vitro Studies Endocrinology, December 1, 2003; 144(12): 5308 - 5321. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/11/9437    most recent M110188200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Authier, F. Articles by Briand, G. Search for Related Content PubMed PubMed Citation Articles by Authier, F. Articles by Briand, G. Social Bookmarking                      What's this?