Proinsulin Targeting to the Regulated Pathway Is Not Impaired in Carboxypeptidase E-deficientCpe /Cpe Mice

Abstract

Sorting of proinsulin from the trans-Golgi network to secretory granules is critical for its conversion to insulin as well as for regulated insulin secretion. The proinsulin sorting mechanism is unknown. Recently, carboxypeptidase E (CPE) was proposed as a sorting receptor for prohormones. To know whether CPE is implicated in proinsulin sorting, pancreatic islets were isolated from CPE-deficient Cpe fat /Cpe fat mice and Cpe fat /+ controls, pulse-labeled ([3H]leucine), and then chased in basal medium (90 min) to examine constitutive secretion followed by medium with secretagogues (60 min) to stimulate regulated secretion. Secretion of labeled proinsulin via the constitutive pathway was <2% even inCpe fat /Cpe fat islets. After a 150-min chase, only 13% of radioactivity remained as proinsulin inCpe fat /+ islets compared with 46% inCpe fat /Cpe fat islets, reflecting slower conversion. Regulated secretion was stimulated to an equal extent from Cpe fat /+ andCpe fat /Cpe fat mice with 20% of the total content of labeled (pro)insulin released during the 60-min stimulatory period. It is concluded that in CPE-deficientCpe fat /Cpe fat mice, proinsulin is efficiently routed to the regulated pathway and its release can be effectively stimulated by secretagogues. CPE is thus not essential for sorting proinsulin to granules. Sorting of proinsulin from the trans-Golgi network to secretory granules is critical for its conversion to insulin as well as for regulated insulin secretion. The proinsulin sorting mechanism is unknown. Recently, carboxypeptidase E (CPE) was proposed as a sorting receptor for prohormones. To know whether CPE is implicated in proinsulin sorting, pancreatic islets were isolated from CPE-deficient Cpe fat /Cpe fat mice and Cpe fat /+ controls, pulse-labeled ([3H]leucine), and then chased in basal medium (90 min) to examine constitutive secretion followed by medium with secretagogues (60 min) to stimulate regulated secretion. Secretion of labeled proinsulin via the constitutive pathway was <2% even inCpe fat /Cpe fat islets. After a 150-min chase, only 13% of radioactivity remained as proinsulin inCpe fat /+ islets compared with 46% inCpe fat /Cpe fat islets, reflecting slower conversion. Regulated secretion was stimulated to an equal extent from Cpe fat /+ andCpe fat /Cpe fat mice with 20% of the total content of labeled (pro)insulin released during the 60-min stimulatory period. It is concluded that in CPE-deficientCpe fat /Cpe fat mice, proinsulin is efficiently routed to the regulated pathway and its release can be effectively stimulated by secretagogues. CPE is thus not essential for sorting proinsulin to granules. Proinsulin is directed to the regulated secretory pathway of the pancreatic β-cell and is converted to bioactive insulin in immature secretory granules (1Orci L. Ravazzola M. Amherdt M. Madsen O. Vassalli J.D. Perrelet A. Cell. 1985; 42: 671-681Abstract Full Text PDF PubMed Scopus (193) Google Scholar). At least three enzymes are needed for generating native insulin, namely the conversion endoproteases PC1 (also known as PC3) and PC2 (2Davidson H.W. Rhodes C.J. Hutton J.C. Nature. 1988; 333: 93-96Crossref PubMed Scopus (321) Google Scholar, 3Halban P.A. Irminger J.C. Biochem. J. 1994; 299: 1-18Crossref PubMed Scopus (288) Google Scholar) and carboxypeptidase E (also called carboxypeptidase H or CPE 1The abbreviations used are: CPE, carboxypeptidase E; TGN, trans-Golgi network; POMC, proopiomelanocortin; HPLC, high performance liquid chromatography; KRB, Krebs-Ringer bicarbonate. ) for trimming residual basic C-terminal residues (4Davidson H.W. Hutton J.C. Biochem. J. 1987; 245: 575-582Crossref PubMed Scopus (121) Google Scholar, 5Fricker L.D. Annu. Rev. Physiol. 1988; 50: 309-321Crossref PubMed Scopus (322) Google Scholar, 6Guest P.C. Arden S.D. Rutherford N.G. Hutton J.C. Mol. Cell. Endocrinol. 1995; 113: 99-108Crossref PubMed Scopus (15) Google Scholar). Correct targeting to secretory granules is essential for the cell to produce and store bioactive insulin, which can then be released upon stimulation of the β-cell by glucose or other secretagogues. One of the crucial steps in targeting takes place in the TGN (trans-Golgi network) (7Orci L. Ravazzola M. Amherdt M. Perrelet A. Powell S.K. Quinn D.L. Moore H.P.H. Cell. 1987; 51: 1039-1051Abstract Full Text PDF PubMed Scopus (206) Google Scholar). The proprotein must somehow be recognized selectively in the TGN and then be transferred to the immature secretory granules, instead of being released through the constitutive or default pathway (3Halban P.A. Irminger J.C. Biochem. J. 1994; 299: 1-18Crossref PubMed Scopus (288) Google Scholar). One hypothesis (reviewed in Ref. 3Halban P.A. Irminger J.C. Biochem. J. 1994; 299: 1-18Crossref PubMed Scopus (288) Google Scholar) accounting for targeting within the TGN implicates a sorting receptor with a broad specificity in the TGN membrane, which recognizes a sorting domain and thereby binds proteins to be routed to the regulated pathway. Sorting domains have been reported for several prohormones including prosomatostatin (8Sevarino K.A. Stork P. Ventimiglia R. Mandel G. Goodman R.H. Cell. 1989; 57: 11-19Abstract Full Text PDF PubMed Scopus (82) Google Scholar, 9Stoller T.J. Shields D. J. Cell Biol. 1989; 108: 1647-1655Crossref PubMed Scopus (110) Google Scholar), chromogranin (10Parmer R.J. Xi X.P. Wu H.J. Helman L.J. Petz L.N. J. Clin. Invest. 1993; 92: 1042-1054Crossref PubMed Scopus (61) Google Scholar), and the basic proline-rich protein (11Castle A.M. Stahl L.E. Castle J.D. J. Biol. Chem. 1992; 267: 13093-13100Abstract Full Text PDF PubMed Google Scholar). An N-terminal sorting domain has been reported for proopiomelanocortin (POMC) (12Tam W.W. Andreasson K.I. Loh Y.P. Eur. J. Cell Biol. 1993; 62: 294-306PubMed Google Scholar, 13Cool D.R. Fenger M. Snell C.R. Loh Y.P. J. Biol. Chem. 1995; 270: 8723-8729Abstract Full Text Full Text PDF PubMed Scopus (139) Google Scholar), although this is controversial (14Roy P. Chevrier D. Fournier H. Chantal R. Zollinger M. Crine P. Boileau G. Mol. Cell Endocrinol. 1991; 82: 237-250Crossref PubMed Scopus (20) Google Scholar). A putative C-terminal sorting domain has also recently been proposed for the prohormone convertase PC2 (15Creemers J.W. Usac E.F. Bright N.A. Van de Loo J.W. Jansen E. Van de Ven W.J.M. Hutton J.C. J. Biol. Chem. 1996; 271: 25284-25291Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar). There is not as yet any direct experimental evidence for a sorting domain for proinsulin, although the comparison of structural features of many prohormones suggests that a region within the insulin B-chain may play such a role (16Kizer J.S. Tropsha L.A. Biochem. Biophys. Res. Commun. 1991; 174: 586-592Crossref PubMed Scopus (39) Google Scholar, 17Gorr S.U. Darling D.S. FEBS Lett. 1995; 361: 8-12Crossref PubMed Scopus (36) Google Scholar). Very recently, CPE has been proposed as the targeting receptor for POMC (18Cool D.R. Normant E. Shen F. Chen H.C. Pannell L. Zhang Y. Loh Y.P. Cell. 1997; 88: 73-83Abstract Full Text Full Text PDF PubMed Scopus (388) Google Scholar) with evidence that it interacts with the N-terminal sorting domain on this prohormone. Further, proinsulin and other proproteins, which are normally secreted through the regulated pathway but not proteins which are released through the constitutive pathway, were shown to displace POMC (or a peptide encompassing its sorting domain) from this putative sorting receptor (18Cool D.R. Normant E. Shen F. Chen H.C. Pannell L. Zhang Y. Loh Y.P. Cell. 1997; 88: 73-83Abstract Full Text Full Text PDF PubMed Scopus (388) Google Scholar). It was subsequently shown thatCpe fat /Cpe fat mice appear to misroute POMC, resulting in its abnormal conversion and secretion (19Shen F.S. Loh Y.P. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 5314-5319Crossref PubMed Scopus (69) Google Scholar). Obese Cpe fat /Cpe fat mice suffer from hyperproinsulinemia associated with expression of a mutated CPE, which results in destruction of the CPE protein in the rough endoplasmic reticulum (20Naggert J.K. Fricker L.D. Varlamov O. Nishina P.M. Rouillé Y. Steiner D.F. Carroll R.J. Paigen B.J. Leiter E.H. Nat. Genet. 1995; 10: 135-142Crossref PubMed Scopus (615) Google Scholar, 21Fricker L.D. Berman Y.L. Leiter E.H. Devi L.A. J. Biol. Chem. 1996; 271: 30619-30624Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar). The pancreatic β-cells of these mice are clearly granulated but the content of the granules appears to be less dense (20Naggert J.K. Fricker L.D. Varlamov O. Nishina P.M. Rouillé Y. Steiner D.F. Carroll R.J. Paigen B.J. Leiter E.H. Nat. Genet. 1995; 10: 135-142Crossref PubMed Scopus (615) Google Scholar), suggesting storage of proinsulin instead of mature insulin. Despite this observation, it has been suggested that proinsulin fails to be targeted to the regulated pathway in β-cells of these mice and that CPE is the sorting receptor not only for POMC but for other prohormones including proinsulin itself (18Cool D.R. Normant E. Shen F. Chen H.C. Pannell L. Zhang Y. Loh Y.P. Cell. 1997; 88: 73-83Abstract Full Text Full Text PDF PubMed Scopus (388) Google Scholar, 19Shen F.S. Loh Y.P. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 5314-5319Crossref PubMed Scopus (69) Google Scholar). Confronted by this discrepancy, we wanted to know whether CPE is indeed implicated in proinsulin sorting. To this end, we have examined whether proinsulin is efficiently targeted to the regulated secretory pathway or released in major part through the constitutive pathway in primary β-cells in islets isolated from CPE-deficientCpe fat mice. C57BLKS-Cpe fat /J male mice were purchased from The Jackson Laboratories, Bar Harbor, ME. Chemicals were from Fluka (Buchs, Switzerland). [3H]Leucine (120 Ci/mmol) was from ARC (St. Louis, MO). Collagenase A (catalog no. 17449) was purchased from Serva (Buchs, Switzerland). The pancreas was removed from mice at 8 to 11 weeks of age and digested in collagenase (1 mg/ml) for 18–20 min, and the islets were hand-picked under a dissecting microscope after coloration by dithizone. For each experiment, islets from 6 mice were pooled and cultured for 24 h in Dulbecco's modified Eagle's medium (8.3 mm glucose), 10% fetal calf serum before use to recover from the isolation procedure. The islets were incubated for 15 min in KRB (16.7 mm glucose) then labeled with 2 mCi/ml [3H]leucine for 20 min in KRB (16.7 mmglucose), washed three times, and chased in KRB (+ 1 mmunlabeled leucine) containing 1.7 mm glucose (“basal”) for 30-min intervals up to 90 min. After each interval, the medium was collected, and fresh medium was added. From 90 to 150 min of chase, the islets were incubated in KRB containing a secretagogue mixture consisting of 16.7 mm glucose, 10 mm Leu and Arg, 1 mm isobutylmethylxanthine, 10 μmforskolin, and 0.1 μm phorbol 12-myristate 13-acetate to stimulate release of granules from the regulated pathway. At 150 min, the medium was collected, and the cells were extracted in 1m acetic acid, 0.1% bovine serum albumin. The medium and the cell extracts were analyzed by reversed phase HPLC to separate and quantify radiolabeled insulin-related peptides as described in detail previously (22Sizonenko S.V. Halban P.A. Biochem. J. 1991; 278: 621-625Crossref PubMed Scopus (28) Google Scholar, 23Irminger J.C. Vollenweider F.M. Neerman-Arbez M. Halban P.A. J. Biol. Chem. 1994; 269: 1756-1762Abstract Full Text PDF PubMed Google Scholar). Intact mouse islets were isolated fromCpe fat /Cpe fat orCpe fat /+ control mice and cultured for 24 h before use. The islets were pulse-labeled (20 min, [3H]Leu) and chased for 30-min intervals up to 90 min in basal medium containing 1.67 mm glucose. During these initial 90 min of chase, it was reasoned that under nonstimulatory conditions any newly synthesized (labeled) proinsulin in the medium would have been secreted predominantly by the constitutive pathway. At 90 min, medium containing the secretagogue mixture (see “Experimental Procedures”) was added, and the islets were incubated for an additional 60 min. Secretion of labeled proinsulin or insulin during this period reflects in major part that released via the regulated pathway. At the end of the total chase period of 150 min, the islets were extracted in acid. At this time, the majority of labeled proinsulin/insulin can be assumed to be stored in granules of the regulated secretory pathway; any proinsulin diverted to the constitutive pathway would thus have been secreted within this 150-min period. The major products observed in the islet cell extracts of control (Cpe fat /+) mice were insulin I and II. There was thus only a small amount (13% of total) of labeled proinsulin left in these cells after 150 min of chase (Fig.1 C). This is quite in keeping with proinsulin having been directed to secretory granules in which it had been converted to insulin and stored. InCpe fat /Cpe fat islets, by contrast, the major radioactive peak was proinsulin (Fig.2 C), accounting for 46.3 ± 2.4% of total proinsulin- and insulin-related radioactivity. The presence of such a secretory product in cells after 150 min is suggestive of its being stored in granules of the regulated pathway, given that no such storage compartment exists for the constitutive pathway but indicates unusually slow proinsulin conversion in keeping with published data (20Naggert J.K. Fricker L.D. Varlamov O. Nishina P.M. Rouillé Y. Steiner D.F. Carroll R.J. Paigen B.J. Leiter E.H. Nat. Genet. 1995; 10: 135-142Crossref PubMed Scopus (615) Google Scholar). The other minor radioactive peaks could correspond to arginine-extended forms of insulin or other proinsulin conversion intermediates. Given that these minor peaks each accounted for no more than 10% of the total labeled products and that the present study addressed proinsulin sorting rather than processing, they were not analyzed further. The reason for slow proinsulin conversion inCpe fat /Cpe fat mouse islets is unknown. Although missorting to the constitutive pathway could account for poor proinsulin conversion, our data do not support this model. It therefore appears as though lack of CPE activity in granules in some way inhibits activity of the conversion endoproteases per se. Most recently, a study in β-cell lines derived from theCpe fat /Cpe fat mice (NIT-2 and NIT-3 cells) confirmed that this CPE mutation leads to inhibition of proinsulin conversion while suggesting that at least some proinsulin was directed to granules. 2O. Varlamov, L. D. Fricker, H. Furukawa, D. F. Steiner, S. H. Langley, and E. H. Leiter, submitted for publication. Figure 2HPLC analysis of labeled (pro)insulin-like products from Cpe fat /Cpe fat islets. The islets were labeled and chased as in Fig. 1.A, combined basal medium from 30–90 min of chase;B, 90–150-min stimulated medium; C, cell extract. The elution times for authentic standards are as indicated by the arrows: 1, insulin I; 2, insulin II; 3, proinsulins I + II. Representative of three independent experiments.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The mere presence of proinsulin in a putative storage compartment is not proof in itself of sorting to the regulated pathway. This can only be addressed by following the kinetics of secretion of newly synthesized (labeled) proinsulin/insulin. Very little proinsulin could be detected in the basal medium of either control (Cpe fat /+) orCpe fat /Cpe fat islets, indicating that constitutive release was minimal in both cases. Essentially no detectable proinsulin was released during the initial 30 min of chase. Thereafter, from 30–90 min of chase under basal conditions, there was 1% proinsulin (of total content) released from theCpe fat /+ islets and 1.5% from theCpe fat /Cpe fat islets (Figs.1 A and 2 A, TableI). A value of 1% is close to that reported by us previously from rat islets (24Rhodes C.J. Halban P.A. J. Cell Biol. 1987; 105: 145-153Crossref PubMed Scopus (158) Google Scholar) and reflects remarkably efficient proinsulin sorting, notably even inCpe fat /Cpe fat mice.Table IPercentage of total (pro)insulin-related productsChase medium 30–90 min basalChase medium 90–150-min stimulatedCell extract at 150 min%%%Cpe fat /+121.477.6Cpe fat /Cpe fat1.5 ± 0.420.4 ± 5.178.1 ± 5.5The medium and the cell extracts were analyzed on HPLC as described in Fig. 1. The radioactivity eluting from HPLC as proinsulin/insulin is expressed as a percentage of the total proinsulin/insulin radioactivity found in the chase media (basal plus stimulated) plus that found in the cell extracts. Data are mean ± S.E. for three independent experiments for Cpe fat /Cpe fat islets and one of two independent experiments for controls. Open table in a new tab The medium and the cell extracts were analyzed on HPLC as described in Fig. 1. The radioactivity eluting from HPLC as proinsulin/insulin is expressed as a percentage of the total proinsulin/insulin radioactivity found in the chase media (basal plus stimulated) plus that found in the cell extracts. Data are mean ± S.E. for three independent experiments for Cpe fat /Cpe fat islets and one of two independent experiments for controls. The hallmark of the regulated pathway is its sensitivity toward secretagogues. Release of proinsulin/insulin was stimulated to a comparable extent from control (Cpe fat /+) andCpe fat /Cpe fat islets. After 60 min of stimulation, 20% of total (cellular) (pro)insulin had been released from the islets of both groups (Table I). Furthermore, the HPLC profile for labeled products released following stimulation was essentially identical to that of stored cellular products (Fig. 1,B versus C, and Fig. 2, B versus C). Taken together with the values for constitutive release described above, these data indicate clearly that the overwhelming majority of proinsulin, even in Cpe fat /Cpe fat islets, had been correctly sorted to granules. We conclude that even if proinsulin is not processed rapidly or extensively in the CPE-deficientCpe fat /Cpe fat mice, it is efficiently targeted to the regulated pathway and stored in secretory granules. The lack of CPE does not have any apparent influence on proinsulin routing and therefore it is most unlikely that CPE is the regulated secretory pathway sorting receptor for proinsulin, at least in the Cpe fat /Cpe fat mouse model. In a very recent report, the putative role of CPE as a sorting receptor has been questioned on theoretical grounds (25Thiele C. Gerdes H.H. Huttner W.B. Curr. Biol. 1997; 7: R496-R500Abstract Full Text Full Text PDF PubMed Google Scholar). It cannot be excluded at this point that different classes of sorting receptors or even different sorting mechanisms can co-exist, responsible for targeting of different subsets of prohormones. Although it remains to be explained why proinsulin is so efficient in displacing POMC from CPE (18Cool D.R. Normant E. Shen F. Chen H.C. Pannell L. Zhang Y. Loh Y.P. Cell. 1997; 88: 73-83Abstract Full Text Full Text PDF PubMed Scopus (388) Google Scholar), the physiological relevance of this finding must be revisited in the light of the present study, which demonstrates unequivocally that CPE is not indispensable for proinsulin sorting to the regulated pathway. We thank Dr. Edward Leiter (The Jackson Laboratories, Bar Harbor, ME) for sending a copy of the study by Varlamov et al. 2 prior to publication.