Protection from Pancreatitis by the Zymogen Granule Membrane Protein Integral Membrane-associated Protein-1
2002; Elsevier BV; Volume: 277; Issue: 52 Linguagem: Inglês
10.1074/jbc.m204159200
ISSN1083-351X
AutoresTakuji Imamura, Minoru Asada, Sherri K. Vogt, David A. Rudnick, Mark E. Lowe, Louis J. Muglia,
Tópico(s)Pancreatic function and diabetes
ResumoPancreatitis is a common disease with substantial morbidity and mortality. To better understand the mechanisms conferring sensitivity or resistance to pancreatitis, we have initiated the analysis of novel acinar cell proteins. Integral membrane-associated protein-1 (Itmap1) is a CUB (complement subcomponents C1r/C1s, sea urchin Uegf protein, bone morphogenetic protein-1) and zona pellucida (ZP) domain-containing protein we find prominently expressed in pancreatic acinar cells. Within the acinar cell, Itmap1 localizes to zymogen granule membranes. Although roles in epithelial polarity, granule assembly, and mucosal protection have been postulated for CUB/ZP proteins, in vivo functions for these molecules have not been proven. To determine the function of Itmap1, we generated Itmap1-deficient mice. Itmap1 −/− mice demonstrate increased severity of secretagogue- and diet-induced pancreatitis in comparison to Itmap1 +/+ mice. In contrast to previous animal models exhibiting altered severity of pancreatitis, Itmap1 deficiency results in impaired activation of trypsin, an enzyme believed critical for initiating a cascade of digestive zymogen activation during pancreatitis. Itmap1 deficiency does not alter zymogen granule size, appearance, or the composition of zymogen granule contents. Our results demonstrate that Itmap1 plays an essential role in trypsinogen activation and that both impaired and augmented trypsinogen activation can be associated with increased severity of pancreatitis. Pancreatitis is a common disease with substantial morbidity and mortality. To better understand the mechanisms conferring sensitivity or resistance to pancreatitis, we have initiated the analysis of novel acinar cell proteins. Integral membrane-associated protein-1 (Itmap1) is a CUB (complement subcomponents C1r/C1s, sea urchin Uegf protein, bone morphogenetic protein-1) and zona pellucida (ZP) domain-containing protein we find prominently expressed in pancreatic acinar cells. Within the acinar cell, Itmap1 localizes to zymogen granule membranes. Although roles in epithelial polarity, granule assembly, and mucosal protection have been postulated for CUB/ZP proteins, in vivo functions for these molecules have not been proven. To determine the function of Itmap1, we generated Itmap1-deficient mice. Itmap1 −/− mice demonstrate increased severity of secretagogue- and diet-induced pancreatitis in comparison to Itmap1 +/+ mice. In contrast to previous animal models exhibiting altered severity of pancreatitis, Itmap1 deficiency results in impaired activation of trypsin, an enzyme believed critical for initiating a cascade of digestive zymogen activation during pancreatitis. Itmap1 deficiency does not alter zymogen granule size, appearance, or the composition of zymogen granule contents. Our results demonstrate that Itmap1 plays an essential role in trypsinogen activation and that both impaired and augmented trypsinogen activation can be associated with increased severity of pancreatitis. zona pellucida choline-deficient, ethionine-supplemented complement subcomponents C1r/C1s, sea urchin Uegf protein, bone morphogenetic protein-1 embryonic stem integral membrane-associated protein-1 isoelectric focusing myeloperoxidase trypsinogen activation peptide phosphate-buffered saline 4-morpholinepropanesulfonic acid 4-morpholineethanesulfonic acid 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling Acute pancreatitis remains a significant cause of morbidity and mortality in the United States (1Steinberg W. Tenner S. N. Engl. J. Med. 1994; 330: 1198-1210Crossref PubMed Scopus (868) Google Scholar, 2Bettinger J.R. Grendell J.H. Pancreas. 1991; 6: S2-S6Crossref PubMed Scopus (39) Google Scholar). These attacks often require hospitalization, and ∼10% of patients die. The treatment of pancreatitis centers on supportive care coupled with close observation for complications. No therapy alters the course of acute pancreatitis. The lack of therapy stems, at least in part, from a paucity of information about the events that led to pancreatic inflammation. Virtually all theories about the pathogenesis of acute pancreatitis have included autodigestion of the pancreas by the digestive enzymes normally synthesized in the pancreatic acinar cells (1Steinberg W. Tenner S. N. Engl. J. Med. 1994; 330: 1198-1210Crossref PubMed Scopus (868) Google Scholar, 2Bettinger J.R. Grendell J.H. Pancreas. 1991; 6: S2-S6Crossref PubMed Scopus (39) Google Scholar). In these models, some event triggers the inappropriate release of digestive enzymes into the parenchyma of the pancreas. Whatever the trigger, most authors believe that the insult causes the premature activation of trypsinogen to trypsin within the acinar cell. As trypsin accumulates, it activates other proenzymes (zymogens), and the combined action of these proteases and lipases further damages the pancreas. Several recent studies support the central role of trypsinogen activation in the pathophysiology of pancreatitis. Most patients with hereditary pancreatitis have a mutation in the gene encoding cationic trypsinogen (3Whitcomb D.C. Gorry M.C. Preston R.A. Furey W. Sossenheimer M.J. Ulrich C.D. Martin S.P. Gates L.K. Amann S.T. Toskes P.P. Liddle R. McGrath K. Uomo G. Post J.C. Ehrlich G.D. Nat. Genet. 1996; 14: 141-145Crossref PubMed Scopus (1342) Google Scholar). The mutation may allow trypsin to accumulate in the acinar cell and activate other digestive enzymes. Studies with mice deficient in cathepsin B, a lysosomal hydrolase that can convert trypsinogen to trypsin, support the importance of trypsinogen activation and the fusion of lysosomes and zymogen granules in the pathophysiology of pancreatitis (4Halangk W. Lerch M.M. Brandt-Nedelev B. Roth W. Ruthenbuerger M. Reinheckel T. Domschke W. Lippert H. Peters C. Deussing J. J. Clin. Invest. 2000; 106: 773-781Crossref PubMed Scopus (469) Google Scholar). The cathepsin B-deficient mice have decreased edema and cellular necrosis associated with decreased trypsinogen activation in an experimental model of pancreatitis induced by hyperstimulation with cerulein. To protect the pancreas from the inappropriate activation of trypsinogen and other zymogens, mechanisms have evolved that maintain the integrity of the pancreas (5Steer M. Go V. Dimagno E. Gardner J. Lebenthal E. Reber H. Scheele G. The Pancreas: Biology, Pathobiology, and Disease. Second Ed. Raven Press, New York, NY1993: 581-591Google Scholar). These mechanisms include the synthesis and secretion of digestive enzymes as inactive zymogens, the synthesis of protease inhibitors, the compartmentalization of proenzymes in specialized zymogen granules destined for trafficking to the apical plasma membrane, and the activation of proenzymes in the duodenum. Together, these protective mechanisms prevent or limit the cascade of proenzyme activation that results in tissue damage. For acute pancreatitis to develop, these protective mechanisms must fail and permit the untimely activation of digestive enzymes by trypsin. Only in the rare patient with hereditary pancreatitis is there a clue as to how the protective mechanisms might fail in acute pancreatitis (3Whitcomb D.C. Gorry M.C. Preston R.A. Furey W. Sossenheimer M.J. Ulrich C.D. Martin S.P. Gates L.K. Amann S.T. Toskes P.P. Liddle R. McGrath K. Uomo G. Post J.C. Ehrlich G.D. Nat. Genet. 1996; 14: 141-145Crossref PubMed Scopus (1342) Google Scholar). For the majority of patients with acute pancreatitis no mechanism is known. Other factors must predispose to premature trypsinogen activation or, alternatively, entirely distinct mechanisms may lead to pancreatitis in these patients. Further insight into these mechanisms may lead to novel strategies for identifying individuals at risk for pancreatitis and to more effective therapies for this potentially life-threatening disorder. To further understand the mechanisms involved in protecting the acinar cell from digestive damage, we have evaluated the function of a novel acinar cell zona pellucida (ZP)1 domain-containing protein Itmap1 (6Kasik J.W. Biochem. J. 1998; 330: 947-950Crossref PubMed Scopus (17) Google Scholar). ZP domain proteins such as uromodullin, the pancreatic ductal protein muclin, the zymogen granule protein GP2, and the inner ear protein β-tectorin are often found in fibrillar or gelatinous compartments of the extracellular matrix, with the ZP domain serving as a filament-organizing motif (7Killick R. Legan P. Malenczak C. Richardson G. J. Cell Biol. 1995; 129: 535-547Crossref PubMed Scopus (73) Google Scholar). This filamentous structure may contribute to mucosal protection by providing barrier function. Alternatively, ZP domain proteins have also been implicated in trafficking of secretory granules in the pancreas (8De Lisle R.C. Ziemer D. Eur. J. Cell Biol. 2000; 79: 892-904Crossref PubMed Scopus (20) Google Scholar). In this study, we demonstrate that Itmap1 is a protein tightly associated with zymogen granule membranes and that Itmap1 attenuates the severity of pancreatitis. We isolated an 861-bp cDNA encoding Itmap1 using gestation day 16 uterus poly(A)+ RNA as "tester" and gestation day 19 poly(A)+ RNA as "driver" in a PCR-Select cDNA subtraction kit (Clontech, Palo Alto, CA). Random-primer-labeled fragments of the Itmap1 cDNA were used to screen a murine 129Sv genomic DNA library (Stratagene, La Jolla, CA). Two genomic clones were isolated and used for sequence analysis and identification of intron-exon boundaries. To generate the vector for homologous recombination, a 3.3-kbXbaI-EcoRI fragment beginning in intron 1 was first cloned into XbaI- and EcoRI-digested pPNT (9Tybulewicz V. Crawford C. Jackson P. Bronson R. Mulligan R. Cell. 1991; 65: 1153-1163Abstract Full Text PDF PubMed Scopus (1160) Google Scholar) to generate pPNT.Itmap1–3′. Next, the 3.8-kbNotI-BamHI fragment ending 1.6-kb 5′ to the transcription initiation site was excised from pBluescript SK II+ as aNotI-XhoI fragment and cloned into pPNT.Itmap1-3′ to generate the final targeting vector pPNTΔItmap1. The TC1 line of embryonic stem (ES) cells (10Deng C. Wynshaw-Boris A. Shen M.M. Daugherty C. Ornitz D.M. Leder P. Genes Dev. 1994; 8: 3045-3057Crossref PubMed Scopus (635) Google Scholar) underwent electroporation with linearized pPNTΔItmap1 and selection as previously described (11Muglia L.J. Jenkins N.A. Gilbert D.J. Copeland N.G. Majzoub J.A. J. Clin. Invest. 1994; 93: 2066-2072Crossref PubMed Scopus (116) Google Scholar). ES clones having undergone appropriate homologous recombination were identified by Southern blot analysis employing a 240-bpEcoRI-BamHI probe external to theItmap1 genomic sequences in the targeting vector. Two independent clones were selected for blastocyst injection, each of which proved capable of germ line transmission. Mice used for these experiments were of a mixed 129 × Black Swiss background; similar results have been obtained on mice of mixed 129 × C57BL/6 or inbred 129SvJ backgrounds. A 488-bp cDNA fragment extending from nucleotide 1048 to 1535 of the Itmap1 cDNA (GenBankTM accession number U69699) was cloned into theXhoI and KpnI sites of the bacterial expression vector pBAD/HisC (Invitrogen, Carlsbad, CA) for generation of a recombinant protein fragment. After purification from arabinose-induced bacteria, the protein fragment was injected into two rabbits for generation of anti-Itmap1 polyclonal antisera. Serum from rabbit 67418 was utilized for the current studies at a dilution of 1:5000 on paraformaldehyde-fixed, paraffin-embedded tissues cut at 5 μm. Antibody binding was visualized with a goat anti-rabbit Cy3 secondary antibody. Sections were counterstained with 4′,6-diamidino-2-phenylindole for localization of cell nuclei. Tissues were fixed by immersion in diethylpyrocarbonate-treated 4% paraformaldehyde in PBS for 24 h at 4 °C. Samples were then cryopreserved in 10% sucrose in PBS, and embedded in OCT compound (Miles, Elkhart, IN) for sectioning on a cryostat. 14-μm sections were thaw-mounted onto Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA) and hybridized to an [α-33P]UTP-labeled 861-base Itmap1 riboprobe as previously described (12Muglia L.M. Schaefer M.L. Vogt S.K. Gurtner G. Imamura A. Muglia L.J. J. Neurosci. 1999; 19: 2051-2058Crossref PubMed Google Scholar). After washing, slides were exposed to Kodak NTB-2 emulsion (Eastman Kodak Co., Rochester, NY) for 3–10 days and developed. Immediately adjacent slides were counterstained with hematoxylin and eosin. For ultrastructural analysis, pancreata were minced into 1-mm3 pieces, immersion-fixed in 2.5% glutaraldehyde in 0.1 m sodium cacodylate buffer at 4 °C overnight, and postfixed in 1.25% osmium tetroxide. Samples were thin-sectioned in Polybed 812 (Polysciences, Warrington, PA), poststained with uranyl acetate and lead citrate, and visualized on a Zeiss 902 microscope (Zeiss, Thornwood, NY). For immunoelectron microscopy, pancreata were fixed in 8% paraformaldehyde-PBS overnight at 4 °C, rinsed three times in PBS, infused in 2 msucrose-polyvinylpyrrolidone, and processed for ultracryotomy. Ultrathin sections were prepared and incubated with blocking buffer containing 10% goat serum. Immunolabeling was done by incubating with the anti-Itmap1 antibody for 2 h, followed by secondary antibody (12-nm-gold-labeled goat anti-rabbit) for 1 h. After washing, sections were stained with uranyl acetate and embedded in methyl cellulose. Specimens were visualized with a Zeiss 902 microscope, and photographs were recorded with Kodak EM film. Pancreatic acinar cell zymogen granules were prepared from adult mice as previously described (13De Lisle R.C. Schulz I. Tyrakowski T. Haase W. Hopfer U. Am. J. Physiol. 1984; 246: G411-G418PubMed Google Scholar,14Wishart M.J. Andrews P.C. Nichols R. Blevins G.T. Logsdon C.D. Williams J.A. J. Biol. Chem. 1993; 268: 10303-10311Abstract Full Text PDF PubMed Google Scholar). Two to three pancreata were placed in ice-cold 250 mmsucrose, 5 mm MOPS (pH 7.0), 0.1 mmMgSO4 containing one MiniComplete tablet (Roche Molecular Biochemicals, Indianapolis, IN) per 10 ml of buffer. The pancreata were homogenized for 30 s with a Polytron on setting 8000 followed by four strokes with a loose-fitting glass Dounce homogenizer. The mixture was centrifuged at 150 × g for 15 min, and the supernatant was re-centrifuged at 1300 × g for 15 min. The pellet was mixed with 40% Percoll, 250 mm sucrose, 50 mm MES (pH 5.5), 0.1 mm MgSO4, 2 mm EGTA, and one MiniComplete tablet per 10 ml. The zymogen granules were banded by centrifugation at 100,000 × gfor 20 min. The zymogen granules were washed by dilution with homogenization buffer and centrifugation at 1300 × gfor 10 min. The granules were osmotically lysed and the membranes isolated and washed as described (14Wishart M.J. Andrews P.C. Nichols R. Blevins G.T. Logsdon C.D. Williams J.A. J. Biol. Chem. 1993; 268: 10303-10311Abstract Full Text PDF PubMed Google Scholar). The first post-lysis supernatant containing the granule contents was also saved for analysis. Enrichment of zymogen granule contents and membranes was confirmed determining the relative abundance of amylase in each preparation by Western blot analysis with an anti-amylase primary antibody (data not shown). Ten micrograms of total RNA was subjected to electrophoresis through 1.2% agarose-formaldehyde gels and transferred to nitrocellulose membranes. [α-32P]UTP-labeled RNA probes specific for mouse Itmap1 mRNA, cyclophilin mRNA, or 18 S ribosomal RNA were hybridized at 60 °C in 50% formamide-containing buffer as previously described (15Gross G. Imamura T. Vogt S.K. Wozniak D.F. Nelson D.M. Sadovsky Y. Muglia L.J. Am. J. Physiol. 2000; 278: R1415-R1423PubMed Google Scholar). Total cellular membrane proteins and zymogen granule membrane proteins and contents were subjected to 8% SDS-PAGE electrophoresis and then transferred to nitrocellulose membranes. Itmap1 was detected with our rabbit anti-Itmap1 primary antibody used at a 1:2000 dilution (total membrane proteins) or 1:5000 dilution (zymogen granule-enriched membranes and contents) and visualized using an enhanced chemiluminescent detection kit (Amersham Biosciences, Arlington Heights, IL). Zymogen granule glycoproteins were detected with peroxidase-conjugated concanavalin A and wheat-germ agglutinin after electrophoretic separation and transfer to nitrocellulose membranes (16Christie D.L. Palmer D.J. Biochem. J. 1990; 270: 57-61Crossref PubMed Scopus (9) Google Scholar). Ponceau S staining of membranes confirmed equivalent protein loading and transfer. Carbohydrate residues were removed from zymogen granule membrane glycoproteins by digestion with peptide:N-glycosidase F according to the manufacturer's recommended protocol (New England BioLabs, Inc., Beverly, MA). For Pronase digestion the zymogen granules were isolated and washed as above except that the last wash did not contain protease inhibitors. The granules were suspended in 1.0 ml of wash buffer divided into four 300-μl aliquots. The granules were incubated on ice for 10 min in buffer alone, buffer with 7% Nonidet P (NP)-40, buffer with 0.33 unit of Pronase, and buffer with 7% Nonidet P-40 and 0.33 unit of Pronase. The reaction was stopped by adding 700 μl of SDS-sample buffer containing one MiniComplete tablet (Roche Molecular Biochemicals, Indianapolis, IN) per 10 ml of buffer followed by boiling for 5 min. Forty micrograms of protein from zymogen granule contents were diluted in rehydration solution (Amersham Biosciences, Piscataway, NJ) containing 8m urea and 2% CHAPS and loaded onto pH 3–10 gradient isoelectric focusing (IEF) strips for first-dimension separation on the basis of IEF point. Samples were focused for ∼12,000 V-h to equilibrium. The focused samples were then loaded onto 4–12% gradient acrylamide gels for second dimension separation by SDS-PAGE on the basis of size. Gels were stained with Coomassie Brilliant Blue G-Colloidal (Sigma, St. Louis, MO). Individual protein spots on scanned images were quantitated densitometrically using Scion Image (Scion Corp., Frederick, MD) software from replicate gels. For cerulein-induced pancreatitis (17Lampel M. Kern H.F. Virch. Arch. A Pathol. Anat. Histol. 1977; 373: 97-117Crossref PubMed Scopus (587) Google Scholar, 18Adler G. Hupp T. Kern H.F. Virch. Arch. A Pathol. Anat. Histol. 1979; 382: 31-47Crossref PubMed Scopus (188) Google Scholar, 19Niederau C. Niederau M. Luthen R. Strohmeyer G. Ferrell L.D. Grendell J.H. Gastroenterology. 1990; 99: 1120-1127Abstract Full Text PDF PubMed Scopus (188) Google Scholar), 2- to 3-month-old male Itmap1 +/+ andItmap1 −/− mice were injected with seven doses of either 50 μg/kg cerulein in normal saline or normal saline at hourly intervals following an overnight fast and ad libitumaccess to water. 8 h (vehicle and cerulein-injected mice) and 24 h (cerulein-injected mice) after the initial injection, blood was obtained by retro-orbital phlebotomy for measurement of amylase and lipase activity on a Vitros 250 analyzer (Ortho Clinical Diagnostics, Rochester, NY) using reagents supplied by the instrument manufacturer (n = 6–9 per group). 8 h after the initial injections, an additional n = 4–6 mice were euthanized, and their pancreata were isolated and weighed. Pancreata were then immersion-fixed in 4% paraformaldehyde, embedded in paraffin, cut into 5-μm sections, and stained with hematoxylin and eosin. Photographs of coded sections (4–6 fields per mouse) were examined by two independent observers and scored for necrosis using the following semi-quantitative scale: 0, no necrosis; 1, periductal necrosis ( 20%). Additional sections were evaluated for acinar cell apoptosis by TUNEL-staining employing an Apoptag peroxidase kit (Intergen Co., Purchase, NY). Sections were counterstained with methyl green after the peroxidase reaction. Quantitation of apoptotic cells was performed on n = 4 saline-treated mice andn = 4–5 cerulein-treated mice. The average number of apoptotic nuclei per ×400 field was calculated from three (saline) or eight (cerulein) random, independent fields per mouse. For the measurement of intra-pancreatic enzyme activities, the pancreas was removed at the indicated times after initiation of cerulein injections and immediately frozen in liquid nitrogen and stored at −80 °C. For the measurement of trypsin activity, the tissue was thawed and homogenized in ice-cold 5 mm MOPS (pH 6.5), 1 mm MgSO4, and 250 mm sucrose (4Halangk W. Lerch M.M. Brandt-Nedelev B. Roth W. Ruthenbuerger M. Reinheckel T. Domschke W. Lippert H. Peters C. Deussing J. J. Clin. Invest. 2000; 106: 773-781Crossref PubMed Scopus (469) Google Scholar). The sample was sonicated for 30 s and centrifuged for 5 min at 16,000 × g. The same buffer supplemented with 1 mm EDTA and 0.1% Triton X-100 was used to prepare the sample for assay of the trypsinogen activation peptide (TAP). After homogenization, the sample was boiled for 10 min, and a supernatant was prepared by centrifugation at 16,000 × g for 5 min. For myeloperoxidase (MPO) assays, pancreatic tissue was homogenized in 20 mm potassium phosphate (pH 7.0) and centrifuged for 10 min at 10,000 × g (4Halangk W. Lerch M.M. Brandt-Nedelev B. Roth W. Ruthenbuerger M. Reinheckel T. Domschke W. Lippert H. Peters C. Deussing J. J. Clin. Invest. 2000; 106: 773-781Crossref PubMed Scopus (469) Google Scholar). The pellet was resuspended in 50 mm potassium phosphate buffer (pH 6.0) containing 0.5% cetyltrimethylammonium bromide. Afterward the sample was frozen and thawed four times and sonicated for 10 s, and the supernatant was prepared by centrifugation at 10,000 ×g for 5 min. Protein concentrations in the samples were determined by the BCA method (Pierce, Rockford, IL). Trypsin, TAP, and MPO were measured by published methods (4Halangk W. Lerch M.M. Brandt-Nedelev B. Roth W. Ruthenbuerger M. Reinheckel T. Domschke W. Lippert H. Peters C. Deussing J. J. Clin. Invest. 2000; 106: 773-781Crossref PubMed Scopus (469) Google Scholar). Relative amounts of trypsinogen were determined by protein immunoblot of pancreatic extracts with a rabbit polyclonal antibody against trypsinogen (Abcam Ltd., Cambridge, UK). Relative abundance of trypsinogen on scanned immunoblots was determined by densitometric analysis with results expressed as normalized absorbance/mg of total protein. Cathepsin B activity was determined with α-N-benzyloxycarbonyl-Arg-Arg-β-naphthylamide as described previously (20McDonald M.K. Ellis S. Life Sci. 1975; 17: 1269-1276Crossref PubMed Scopus (159) Google Scholar). Cathepsin B activity is reported as relative units (change in fluorescence/min)/mg of total protein. Each assay was done in triplicate on samples isolated from three different mice of each genotype at each time point. To evaluate cerulein-induced signaling in pancreata from mice of each genotype, the cerulein concentration dependence of amylase secretion from pancreatic snips harvested 1–2 h prior to stimulation was measured in triplicate as previously described (21Jaffrey C. Eichenbaum D. Denham D.W. Norman J. Pancreas. 1999; 19: 377-381Crossref PubMed Scopus (7) Google Scholar). Diet-induced pancreatitis was precipitated by placing 2- to 3-month-oldItmap1 +/+ (n = 17) andItmap1 −/− (n = 18) female mice (body weight, 25 ± 0.5 g) on choline-deficient, 0.5% ethionine-supplemented (CDE) chow (Dyets, Inc., Bethlehem, PA) as previously described (22Lombardi B. Rao N.K. Am. J. Pathol. 1975; 81: 87-99PubMed Google Scholar, 23Lombardi B. Estes L.W. Longnecker D.S. Am. J. Pathol. 1975; 79: 465-480PubMed Google Scholar). We chose to test this age range of mice, because they are somewhat more resistant to morbidity in this model of pancreatitis. After an overnight fast, mice received the CDE chow for 48 h and then were returned to normal rodent chow. Samples for serum enzymes (n = 8 per genotype) and pancreatic histology (n = 3 per genotype) were obtained 72 h after the initiation of the CDE diet. All results are expressed as mean ± S.E. unless otherwise indicated. Statistical analysis was by analysis of variance, with p ≤ 0.05 considered significant. Differences in mortality were assessed for significance by Chi-square analysis. Statistical analysis of the semi-quantitative assessment of acinar cell necrosis was by Mann-Whitney rank sum. Itmap1 was first identified as a novel ZP protein induced specifically in late gestation mouse uterus (6Kasik J.W. Biochem. J. 1998; 330: 947-950Crossref PubMed Scopus (17) Google Scholar). In accord with this previous report, we found no expression of Itmap1 in non-gravid mouse uterus and high level expression in mid-to-late gestation uterus (Fig. 1 a). Within the uterus, Itmap1 expression localized exclusively to the epithelium (Fig. 1 b). A general tissue survey in non-gravid mice was notable for high levels of Itmap1 expression in the pancreas, a site not previously evaluated for Itmap1 expression. In situhybridization to histological sections of pancreas demonstrated Itmap1 mRNA expression in acinar cells but not ductal epithelium (Fig. 1 c). Other tissues, including spleen, stomach, kidney, thymus, ileum, and colon did not express Itmap1 mRNA (data not shown). To define the cellular distribution of Itmap1, we generated a rabbit polyclonal anti-serum against the junction of the CUB and ZP domains of Itmap1. By immunofluorescence, this antiserum detected Itmap1 immunoreactivity throughout the acinar cell population of the pancreas (Fig. 2 a). The staining co-localized with zymogen granules. We confirmed that Itmap1 resides in the zymogen granules by immunoelectron microscopy (Fig. 2 b). Grains were only associated with zymogen granules. In the uterus, the anti-Itmap1 antibody also stained granular structures in the apical region of the epithelium (data not shown). This pattern of expression was specific for Itmap1 and not other ZP domain proteins, becauseItmap1 −/− mice (described below) failed to demonstrate any immunoreactivity. The amino acid sequence of Itmap1 contains a predicted signal peptide and transmembrane domain indicating that Itmap1 should localize to a cellular membrane. To determine if Itmap1 associates with membranes, we performed Western blots on preparations of total membranes and of zymogen granule-enriched membranes from both wild type andItmap1 −/− mice. Our antibody detected a strongly positive 110-kDa protein in total cellular membrane preparations from wild type pancreas and uterus (Fig. 3 a). In contrast, only weak, nonspecific bands were detected in membranes prepared fromItmap1 −/− mice (Fig. 3 a) and when pre-immune serum was substituted for the anti-Itmap1 antibody (data not shown). A protein of identical size was also present in alkali-extracted zymogen granule-enriched membranes from wild type but not Itmap1-deficient mice (Fig. 3 b). Itmap1 reactivity was not present in zymogen granule contents. Thus, Itmap1 is tightly associated with zymogen granule membranes, and, by virtue of its predicted transmembrane domain, likely to be an integral membrane protein. To determine whether the larger amino-terminal CUB/ZP portion of Itmap1 resides within or outside the zymogen granule, we subjected isolated zymogen granules to protease digestion with or without permeabilization with the detergent Nonidet P-40. Itmap1 immunoreactivity was abolished with Pronase treatment only after permeabilization of the zymogen granules with detergent, demonstrating that the CUB and ZP domains reside within the zymogen granule (Fig. 3 c). Because the Itmap1 cDNA sequence predicts a product of 69 kDa and our antiserum recognizes a product of ∼110 kDa, we investigated the possibility that post-translational glycosylation of Itmap1 explained the size difference. Western blot analysis of zymogen granule membrane proteins with the anti-Itmap1 antibody after digestion with peptide:N-glycosidase F resulted in disappearance of the 110-kDa protein band, and appearance of a new 69-kDa immunoreactive band, confirming that the observed size of mature Itmap1 results from glycosylation (Fig. 3 d). To determine the role of Itmap1 in zymogen granule functionin vivo, we generated Itmap1 −/−mice by homologous recombination in embryonic stem cells. TheItmap1 gene consists of eight exons encompassing ∼13 kb of mouse genomic DNA (Fig. 4 a). In our targeting vector, we replaced exon 1, encoding the translation start site and first 27 amino acids, and 1.6 kb 5′ to the transcription initiation region, with a phosphoglycerate kinase-neomycin resistance cassette (Fig. 4 b). Two independently targeted clones were selected for microinjection, and both proved capable of germ line transmission. Similar results were obtained withItmap1 −/− mice arising from each clone. To ensure that deletion of the Itmap1 transcription promoter region and exon 1 resulted in a null allele, we performed RNA blot analysis of uterus RNA from gravid mice at 18.5 days of gestation utilizing a hybridization probe detecting mRNA sequences 3′ to the deleted region. Robust Itmap1 mRNA expression was found inItmap1 +/+ mice, and Itmap1 mRNA was absent in Itmap1 −/− mice (Fig. 4 c).Itmap1 −/− mice exhibit normal growth, longevity, and fertility when compared withItmap1 +/+ littermates.Itmap1 −/− females have normal timing for parturition, reproducibly delivering viable litters at 19.5 days of gestation. Zymogen granule size (Itmap1 +/+0.81 ± 0.02 μm versus Itmap1 −/− 0.78 ± 0.2 μm;n = 3 mice per group, 40 granules measured per animal) and appearance by electron microscopy did not differ betweenItmap1 +/+ and Itmap1 −/−mice (Fig. 4 d). Recent studies evaluating other pancreatic ZP proteins suggest that they may be involved in the intracellular trafficking of secretory granules and in the regulated exocrine transport of dig
Referência(s)