Human Pancreatitis-associated Protein

Human Pancreatitis-associated ProteinMessenger RNA Cloning and Expression in Pancreatic Diseases

Beatrice Orelle, * Volker Keim, * * Luis Masciotra, Jean-Charles Dagorn, * and Juan-Lucio lovanna *

*U.315 Institut National de la Sante et de la Recherche Medicale, F-13009 Marseille, France; tUniversittt Heidelberg, II MedizinischeKlinik, Mannheim, Germany; and §Department ofPhysiology, Faculty ofMedicine, University ofBuenos Aires, Buenos Aires, Argentina

Abstract

A human pancreatic cDNA library was screened with thecDNA encoding rat "pancreatitis-associated protein" (PAP).The selected clone encoded a secretory protein structurally re-

lated to rat PAP. The protein had the same size as rat PAP andshowed 71% amino acid identity, the six half-cystines being inidentical positions. Domains of the proteins showing homolo-gies with calcium-dependent lectins were also conserved. In ad-dition, expression in pancreas of the genes encoding the humanprotein and rat PAP showed similar characteristics: both were

expressed at very low levels in control tissue and overexpressedduring the acute phase of pancreatitis, contrary to most secre-

tory products. The human protein was therefore named humanpancreatitis-associated protein (PAP-H). Antibodies raised toa synthetic peptide ofPAP-H detected a single band with an M,compatible with PAP-H in Western blot analysis of proteinsextracted from a pancreas presenting with acute pancreatitis.In that tissue, the protein could be immunolocalized to the api-cal regions of acinar cells. An immunoassay was also con-

structed to quantify the protein in serum. Elevated PAP-H lev-els were observed in patients with acute pancreatitis and insome patients with chronic pancreatitis. Values were close tobackground in healthy subjects and in patients with other ab-dominal diseases. These results confirm that PAP-H synthesisincreases during inflammation and suggest a possible use of theprotein as biological marker of acute pancreatitis. (J. Clin. In-vest. 1992. 90:2284-2291.) Key words: lectin * pancreatic in-flammation * acute pancreatitis * acute phase * biologicalmarker

Introduction

Acute pancreatitis is characterized by edema, leukocyte infil-tration, hemorrhage, and cellular necrosis ( ). When that epi-sode is not fatal, it is followed by a total recovery of the initialmorphology and functional capacity of the gland (1). Duringthe acute phase, damaged acinar cells show many ultrastruc-tural modifications such as fragmentation and dilatation oftherough endoplasmic reticulum, reduction in the number of zy-mogen granules, and formation of cytoplasmic vacuoles. Mor-

This work was presented in part at the Annual Meeting of the Ameri-can Gastroenterology Association, 19-22 May 1991.

Address reprint requests to J.-L. lovanna, U.3 15 INSERM, 46 Bou-levard de la Gaye, F- 13009 Marseille, France.

Receivedfor publication 30 December 1991 and in revisedform 17June 1992.

phological alterations are accompanied by important changesin secretory parameters. Studies in rats showed that contentand secretion of pancreatic secretory proteins were generallyreduced during the acute phase of pancreatitis (2), whereas alimited number of proteins were secreted in higher amounts(2-4). These modifications result from a change in the patternof gene expression programmed to help the tissue survive theacute phase of inflammation (2). The most dramatic increasewas observed for the "pancreatitis-associated protein" (PAP), 'which was merely detectable in control pancreas and repre-sented up to 5% of secreted protein at the climax of the acutephase. Return to control levels paralleled regression of tissueinflammation (5). The rapid and strong induction of the PAPgene is reminiscent of the response to stress of acute phaseproteins (3). Yet, contrary to such proteins, PAP is an exocrineprotein synthesized on the rough endoplasmic reticulum ofacinar cells and stored in zymogen granules before being se-creted, as demonstrated by subcellular fractionation and im-munoelectron microscopy (5). Several domains of homologywith calcium-dependent animal lectins were found in the pri-mary structure of rat PAP and the protein was shown to induceextensive bacterial aggregation in vitro (3), suggesting that it isinvolved in the control of bacterial proliferation. PAP appearstherefore as an important component of the mechanism ofdefense against pancreatic aggression which deserves beingcharacterized in human pancreas. It might also prove a veryuseful marker of pancreatic inflammation. Purification ofsucha protein by classical procedures is hindered by the hazardsassociated with pancreatic juice collection from patients pre-senting with acute pancreatitis. We developed an alternativestrategy based on the assumption that mRNAs encoding the ratand human PAPs were similar enough to allow screening ahuman cDNA library with the rat probe.

We report here the cloning and sequencing of human PAP(PAP-H) cDNA, from which was deduced the primary struc-ture ofthe protein. The human clone was used to monitor PAPgene expression in diseased pancreas. In addition, antibodies tothe synthetic amino-terminal peptide of PAP-H were used todetermine PAP-H levels during acute pancreatitis.

Methods

Library screening. A human pancreatic cDNA library in the expressionvector Xgt- 11 containing 1.4 x 106 different recombinant clones wasobtained from Clontech Laboratories, Inc. (Palo Alto, CA). That li-brary was screened by the plaque screening procedure (6), in condi-tions adapted to heterologous hybridization. The R4 insert (3), 32P-la-beled by random priming (7) to a sp act of 109 cpm/Ag, was used as

probe to screen 3 x 105 recombinant clones. That fragment corre-sponds to nucleotides 1-765 ofthe rat PAPcDNA (3). Duplicate filters

1. Abbreviations used in this paper: PAP, pancreatitis-associated pro-tein; PAP-H, human PAP; PCR, polymerase chain reaction.

2284 B. Orelle, V. Keim, L. Masciotra, J. C. Dagorn, and J. L. Iovanna

J. Clin. Invest.© The American Society for Clinical Investigation, Inc.0021-9738/92/12/2284/08 $2.00Volume 90, December 1992, 2284-2291

were prehybridized 4 h at 650C in a solution containing 6x standardsaline citrate (SSC) (SSC is 150 mM NaCI, 15 mM sodium citrate),0.1% BSA, 0.1% Ficoll, 0.1% polyvinylpyrrolidone, 0.5% SDS, and 100,ug/ ml of denatured herring sperm DNA. Overnight hybridization wasconducted in the same buffer in the presence of 5 x 106 cpm/ml ofcDNA probe, at 650C. Filters were washed 2 x 15 min with 6x SSC,0.1% SDS at 650C. A single clone (QH 1 ) was selected through threerounds of screening.

Biohazards associated with the experiments described in this publi-cation have been previously examined by the French National ControlCommittee.

Enzymatic amplificationfor sequencing QH1 DNA. About 0.5 jig ofQH 1 DNA was incubated in 50 ,1 of 50mM KCl, 10mM Tris HCl pH8.3, 1.25 mM MgCl2, 200 jM of each deoxynucleotide triphosphate,0.1% (wt/vol) gelatin and 2.5 U of Taq DNA Polymerase (Appligene,Illkirch, France) containing 20 pM ofeach ofthe synthetic oligonucleo-tides (A) 5'GGTGGCGACGACTCCTGGAGCCCG 3' and (B)5'TTGACACCAGACCAACTGGTAATG 3', corresponding to re-gions of the Xgt 11 DNA flanking the EcoRI cloning site. Polymerasechain reaction (PCR) was performed as follows for 25-30 cycles in athermocycler (Crocodile; Appligene): denaturation at 940C for 10 s,annealing at 55°C for 2 min, and DNA synthesis at 74°C for 3 min. Inthe last cycle DNA synthesis lasted 10 min. Amplified DNA was puri-fied by electroelution after separation on polyacrylamide gel followedby phenol/chloroform/isoamyl alcohol (50:48:2) extraction and eth-anol precipitation.

Generation ofsingle-strandedDNA byPCR. Asymmetric DNA am-plification was performed as described by Gyllenstein and Erlich (8).About 10 ng of amplified DNA was incubated in the PCR reactionmixture described above, with the exception that the concentrations ofsynthetic oligonucleotides were 50 pmol for A and 0.5 pmol for B. Asingle strand in the opposite direction was obtained with the reverseratio of oligonucleotide concentrations. The reaction mixture was sub-jected to 20 cycles of 10 s at 94°C, 2 min at 55°C, and 3 min at 74°C.

Sequencing single-stranded DNA. The amplification mixture wasadjusted to 300 ,il with H20, applied onto a 5-ml column (Bio-GelP-30; Bio-Rad Laboratories, Richmond, CA) and spun at 2,000 rpm ina rotor (HB-4; Sorvall Instruments Div., Du Pont Co., Newton, CT) toremove excess dNTP and buffer components and the DNA ethanolprecipitated. After resuspension in H20, the ssDNA was sequenced inboth directions with a DNA sequencing kit (Sequenase version 2.0;U.S. Biochemical Corp., Cleveland, OH) following recommendationsof the manufacturer and using appropriate synthetic oligonucleotidesas primers (see Results).

Sequence comparisons. The complete sequence of human PAPmRNA was compared with the sequences listed in Genbank (9). Thesearch was conducted with the BISANCE system (CITI2; Centre Inter-universitaire d'informatique, Paris, France) using the program ofGoadand Kanehisa ( 10). The complete amino acid sequence was comparedwith the sequences listed in the National Biochemical Research Foun-dation data bank, using the FASTP program (CITI2).

Genomic analysis. Human genomic DNA was purified from lym-phocytes ( 11 ) and digested in separate reactions with the restrictionendonucleases BamHI, EcoRI, PstI, HindIII, and Bgl II. The sampleswere fractionated by electrophoresis through a 1% agarose gel, and theDNA was transferred to a membrane (Hybond N; Amersham Int'l.,Amersham, UK). The membrane was baked in a vacuum oven at80°C for 2 h and prehybridized for 4 h at 65°C in 6x SSC, 0.1% BSA,0.1% Ficoll, 0.1% polyvinylpirrolidone, 0.5% SDS, and 100 ,g/ml de-natured herring sperm DNA. The membrane was hybridized in thesame solution containing as probe the QH 1 insert amplified by PCR( 12) and 32P-labeled (7) to a sp act of 1 X 109 cpm/,ug. The blot waswashed twice for 15 min at room temperature in 2X SSC, 0.1% SDS,and then twice for 30 min at 65°C in 0.1X SSC, 0.1% SDS beforeautoradiography.

RNA analysis. RNA was extracted from tissue fragments immedi-ately frozen in liquid nitrogen after surgery and stored at -80'C ( 13).Fragments of pancreas, liver, spleen, and kidney were obtained fromcadaver kidney transplant donors. Other fragments of pancreas were

obtained from the Service de Gastroenterologie, H6pital Sainte Mar-guerite, Marseille, France (Dr. J. Sahel). One of them was obtainedfrom a patient with a severe necrohemorrhagic pancreatitis who died 5d after the onset ofabdominal pain. Two samples were taken at surgeryfrom patients presenting with obstructive pancreatitis due to an adeno-carcinoma of the biliary tract and an adenocarcinoma of the head ofthe pancreas, respectively. Two samples were obtained from patientspresenting with chronic pancreatitis and one from a patient presentingwith a pancreatic endocrine tumor. Samples of pancreatic adenocarci-nomas were obtained from two patients. Samples were submitted toroutine histological examination. Frozen pellets of Capan I and Mia-Paca cells were a generous gift of Dr. A. Estival (U. 151 INSERM,Toulouse, France).

RNA samples were size fractionated on a formaldehyde agarose gelas described ( 11). Northern blot analysis was performed by blottingRNAs onto nylon membranes (Biodyne; Pall BioSupport Corp., GlenCove, NY). The filters were prehybridized then hybridized in 50%formamide, 5x sodium chloride sodium phosphate EDTA buffer ( 180mM NaCl, 1 mM EDTA, 10mM NaH2PO4, pH 7.5), 0.1% BSA, 0.1%Ficoll, 0.1% polyvinylpyrrolidone, 0.5% SDS, and 200,ug/ml of dena-tured herring sperm DNA at 420C in the presence of the 32P-labeledQH I insert. Then the filters were washed four times for 5 min at roomtemperature in 2X SSC, 0.1% SDS, twice for 15 min at 50'C in 0.1XSSC, 0.1% SDS, and for 30 min in 0.1 X SSC.

Production of antibodies to QHJ. The predicted amino-terminalpeptide (EEPQAY) of the protein was chemically synthesized (Neo-system, Strasbourg, France), coupled to BSA, hemoglobin, or lacto-ferrin as carriers using bis-diazotized benzidine ( 14), and used as im-munogen. Rabbits were injected subcutaneously with a suspension of100 ,g of the immunogen coupled to albumin (100 Mg in 1 ml saline)mixed with I ml CFA. The same injection, except that incompleteadjuvant was used, was repeated three times at 2-wk intervals with theantigen coupled with hemoglobin, lactoferrin, and albumin again. 10 dafter the last injection, antisera were collected by puncture of the earvein. Immunoglobulins were purified from the sera as already de-scribed (5).

Immunodetection ofQHJ in pancreatic homogenates. Fragments ofpancreas ( 300 mg) were homogenized in 1 ml of 50mM Tris buffer,pH 8.0. The homogenate was centrifuged at 100,000 g for 2 h. 25 zd ofsupernatant was loaded onto a 15% polyacrylamide-SDS gel and pro-cessed for Western blotting with the antibodies to the NH2-terminalend of QH 1 as already described (5). Immunolocalization was per-formed in thin sections of pancreas using the peroxidase-antiperoxi-dase method of Sternberger et al. ( 15). The antiserum to the NH2-ter-minal end ofQH l was used at a 1:100 dilution.

Quantification ofPAP-H by competitive ELISA. The solid phasewas prepared by adding 100 Al of the peptide coupled to albumin (10Ag/ml) (14) in solution A (NaHCO3 100 mM, pH 8.5) into each wellof ELISA microplates (Nunc, Roskilde, Denmark). The plates wereincubated 2 h at 25°C in a humid atmosphere. The wells were thenwashed with buffer B (0.5% Tween 20 (vol/vol) in PBS). The antibodycapture assays were performed by incubating serum samples (50 yd, 25,l, 12.5 Al, and 6.25 tl) with purified IgG in 100 ,l of solution C (Tris100 mM, pH 7.4, 1% Tween 20, and 1.5% BSA), 2 h at 25°C in ahumid atmosphere. Incubation continued for 2 h in the antigen-coatedplates. The wells were washed with buffer B and fixed antibodies weredetected by goat anti-rabbit peroxidase-labeled IgG. Finally, the plateswere washed and the peroxidase reaction conducted as already de-scribed (5). Quantification was made by comparison with a calibrationcurve obtained with serial dilutions ofthe peptide. That ELISA allowedthe detection of 12.5 fmol ofQH 1 equivalent. Specificity of the QH 1assay in serum was controlled by fractionating serum from healthysubjects and patients with acute pancreatitis on an HPLC column(TSK 3000, System Gold apparatus; Beckman Instruments, Inc., Ful-lerton, CA). No immunoreactivity was observed in fractions from con-trol serum. In patients with acute pancreatitis, immunoreactivity ap-peared under a single peak and was completely inhibited by competi-tion with the synthetic QH 1 NH2-terminal peptide (not shown).

PAP-H was assayed in the serum ofpatients admitted to the Service

Human Pancreatitis-associated Protein 2285

cgggagagtgactcctgattgcctcctcaagtcgcagacact ATG CTGMet Leu

CCT CCC ATG GCC CTG CCC AGT GTA TCT TGG ATG CTG CTTPro Pro Met Ala Leu Pro Ser Val Ser Trp Met Leu Leu

TCC TGC CTC ATG CTG CTG TCT CAG GTT CAA GGT GAA GAA 126Ser Cys Leu Met Leu Leu Ser Gln Val Gln Gly Glu Glu 28

CCC CAG AGG GAA CTG CCC TCT GCA CGG ATC CGC TGT CCC 165Pro Gln Arg Glu Leu Pro Ser Ala Arg Ile Arg Cys Pro 41

AAA GGC TCC AAG GCC TAT GGC TCC CAC TGC TAT GCC TTG 204Lys Gly Ser Lys Ala Tyr Gly Ser His Cys Tyr Ala Leu 54

TTT TTG TCA CCA AAA TCC TGG ACA GAT GCA GAT CTG GCC 243Phe Leu Ser Pro Lys Ser Trp Thr Asp Ala Asp Leu Ala 67

TGC CAG AAG CGG CCC TCT GGA AAC CTG GTG TCT GTG CTC 282Cys Gln Lys Arg Pro Ser Gly Asn Leu Val Ser Val Leu 80

AGT GGG GCT GAG GGA TCC TTC GTG TCC TCC CTG GTG AAG 321Ser Gly Ala Glu Gly Ser Phe Val Ser Ser Leu Val Lys 93

AGC ATT GGT AAC AGC TAC TCA TAC GTC TGG ATT GGG CTC 360Ser Ile Gly Asn Ser Tyr Ser Tyr Val Trp Ile Gly Leu 106

CAT GAC CCC ACA CAG GGC ACC GAG CCC AAT GGA GAA GGT 399His Asp Pro Thr Gln Gly Thr Glu Pro Asn Gly Glu Gly 119

TGG GAG TGG AGT AGC AGT GAT GTG ATG AAT TAC TTT GCA 438Trp Glu Trp Ser Ser Ser Asp Val Met Asn Tyr Phe Ala 132

TGG GAG AGA AAT CCC TCC ACC ATC TCA AGC CCC GGC CAC 477Trp Glu Arg Asn Pro Ser Thr Ile Ser Ser Pro Gly His 145

TGT GCG AGC CTG TCG AGA AGC ACA GCA TTT CTG AGG TGG 516Cys Ala Ser Leu Ser Arg Ser Thr Ala Phe Leu Arg Trp 158

AAA GAT TAT AAC TGT AAT GTG AGG TTA CCC TAT GTC TGC 555Lys Asp Tyr Asn Cys Asn Val Arg Leu Pro Tyr Val Cys 171

AAG TTC ACT GAC tagtgcaggagggaagtcagcagcctgtgtttggt 602Lys Phe Thr Asp 175

gtgcaactcatcatgggcatgagaccagtgtgaggactcaccctggaagaga 654

atattcgcttaattcccccaacctgaccacctcattcttatctttcttctgt 706

ttcttcctccccgctagtcatttcagtctcttcattttgtcatacggcctaa 758

ggctttaaagagcaat.aaatttttagtctgcaaaaaaa 797

Figure 1. Nucleotide sequence ofQH I (PAP-H) mRNA and deducedamino acid sequence ofthe encoded preprotein. The initiation codon(nucleotides 43-45) was selected by analogy with rat PAP mRNAsequence (3). Noncoding sequences are in lowercase letters and thepolyadenylation site AATAAA is underlined. These sequence dataare available from EMBL/GenBank( under accession numberM8433.

de Gastroenterologie, H6pital Sainte Marguerite (Dr. J. Sahel) or theDepartement de Chirurgie, H6pital d'Adultes d'Avignon, France (Dr.G. Angelvin). Diagnosis ofacute pancreatitis was established by typicalabdominal pain, accompanied by increased serum amylase and lipaselevels and ultrasonography or computed tomography. Patients classi-fied in Fig. 10 in the group ofabdominal diseases other than pancreati-tis included seven main bile duct lithiasis, two cancers of the main bile

duct, four pancreatic cancers, two pancreatic cysts, three biliary steno-sis, five ulcerative colitis, one duodenal tumor, one liver cirrhosis, onehyperamylasemia of unknown origin, one renal insufficiency with in-creased amylasemia, and three duodenitis.

Results

Cloning human PAP messenger RNA. A radiolabeled rat PAPcDNA (R4) (3) was used to screen a human pancreatic cDNAlibrary in Xgt- 11 at relatively low stringency as requested by theheterologous nature of the probe. The only positive clone afterthree successive screenings (QH1) was selected and directlysequenced after single-strand generation by asymmetric PCRamplification (Fig. 1). Sequence was completed with the helpof five synthetic oligonucleotides, homologous, respectively, toregions 12 nucleotides upstream of the cloning site of the vec-tor and positions 168-186, 327-345, and 528-546 of QH1and, in the opposite direction, positions 798-765, 546-528,and 155-138 ofQH 1. The complete sequence comprised 790nucleotides, exclusive ofthe poly-A tail. A putative polyadenyl-ation signal (AATAAA) was present 13 nucleotides upstreamfrom the poly-A extension. The approximate length ofthe ma-ture transcript was estimated to 950 nucleotides by Northernblot analysis. The sequence reported here extends to 42 nucleo-tides in the nontranslated 5' end of the transcript.

Sequence comparisons ofQHI with PAP mRNA and oftheencoded protein with rat PAP and other lectins. Nucleotide se-quence comparison ofQH 1 and the coding region of rat PAPmRNA revealed 74% identity. Regions outside the coding re-gion only showed 55% identity (not shown).

A single open reading frame was found in QH 1 mRNA,encoding a polypeptide of 175 amino acids which showed anoverall similarity of 71% with rat PAP (Fig. 2). Positions ofthesix cysteine residues were conserved. In addition, the homologywith calcium-dependent animal lectins (C-type lectins) ob-served in several regions of the rat PAP sequence was alsofound in the QH 1 sequence. Fig. 3 shows the comparison ofthecarboxy-terminal portion ofQH 1 (residues 10-148) with sev-eral lectins including human lithostathine (16), human throm-bomodulin (17), chicken hepatic lectin (18), rat hepatic lectin(19), rat cartilage proteoglycan (20), human tetranectin (21),human IgE receptor (22), human chondroitin sulphate coreprotein (23), rat Kupffer cell receptor (24), Sarcophaga pere-grina lectin (25), dog pulmonary surfactant apoprotein (26),acorn barnacle lectin (27), and the A chain of the coagulationFactor IX/Factor X binding protein (28). Regions showinghomology corresponded to domains potentially involved inlectin activity (29). Comparison with human pancreatic lith-ostatine (16) (Fig. 3) revealed a high degree of similarity, asalready reported for the rat proteins (3). Again, similarity did

P ?OPIPHPr .PIP R . HRLF.PM.......

PrePAPR . HR L .F . V M S ~l S 0LtlS OVEP......P KR.E L. ...S.

SPKK I. ....

P.-R.C.. G.... .......S.....K..

..P. .. .. ..$ ... .. ..G...POA :-A:::::::::: :R :::: ::::::: ::: ::::::::::::K:- :: :.:-: '.W.:::::: :: -T::::::A: :L::::: ::::

MN~.........................................

Pr PAPS S S P ~~~..PrPAPR

LDGFCG$~~SRSSGFLR~RDTTCEVKLY::.::Y

1 - 351 - 35

36 - 70 Figure 2. Sequence compari-36 - 70 son of human (H) and rat

76 - 105 (R) prePAPs. Sequence76 - 105 alignment was obtained

without introducing dele-106 - 140 tions. Mature PAP-R starts106 - 140 with a Gln in position 27.141 - 175 Shaded areas correspond to141 - 175 amino acid identities.

2286 B. Orelle, V. Keim, L. Masciotra, J. C. Dagorn, and J. L. Iovanna

P,PHCRRII

HIP

Rs

)APH A ~R :::. :: K SK:-:P:~

[TM PA PQPGGR::RSQC VEH D C :A:LV:GY A'FLN::SQTCZR-.HL PCGAQ SRQWEY FE~r± C!:VV: FS.± MS FKKA:E Mp~-HL ~ NW E E S FAVt;(':PVKC::O2LENA -

kCAR DQ TC-: E:C! TK Q, H:E R E QOITN LQ V L G K H K L :-: T T~:T H :S D S:EXGER VCNT CP::E KWI NFQRK(CYVYFGKGT K:Q:W.:VHA:RYAkC:DDME -~ICSCP DTET:C-::DY .K-:W.HKFQGQ:C-:Y:KYFAHRRT1 DAAW:ERE C:RLQGA-LKCR VL Q i M Q DW ]K-: Y FN GKF :Y Y F SR D K WHE W:A E NF :C- V SQ0 GA--PL VP QLQKALD !~:REYLIETELKYN:i!HQAi:WHFEC~ARHCQ-

)P5A V L S Li E S ILV V CR KV FSS N A INFND IQE -CAGCAGC -kch D :::: LS S S KCt:c:E K A::E K Y4C:T:W..::E:V :E V T QAK* :O**ALBL T CP C N L LWQE£ GY,.W S....VTVH

....H....................... S: ........ Y S Y V. C . D ?A'SPH~ ~ ~>rn HLMTVR

. SVAADV.....................FP~ cAHL HI.. L.I.....Y.VO P M F IRN E - FI:D.E.N......~~HLHLVVVTSW K ~~~~~~~.jQ R~~~V 0 0 H MC P L N- ~ N..-.....

...H .SI. TPE...V K A0 D- Q R --.T.............LR.VGEA..I W ~ N M A -{XG~~R00VSIH S P K $~~~~ 0 C ~~ L T K HAS H TO - SW ~~~ 0 RN L C - --..

4CSCP ................-WI~ NDK --~~XCRHLASVTSQEF~~~~~~~QAFLV0ITNAVD - HWI~.............. -~~PLQLVTI~~~~~riADKNNAIIDI~~~~~~VKRVVCKSHN-LWL~~~~~~~. CNDEYSSS...)PAIAAMPKNAASVKQY--ALLEP~chHLSIESSEA~rV 0LVT0 MKRLD YIWCL VQC--

RHL -CHKVCTYT--FNRCPDYH

- G -P CT:GV-G:PF:............ NWR.CHT3PL DAL----DYYLRQSPTQA GN-S AE-IWSENNPD-::NYK-DPA - SC FQ MD AP -- -N TN YP EP CR --Lch----KVKQCNSF.WSDCSSV----SYENWIEAE SKT-~~~~~~~~~~~...........

RIE 1:S.PLDT-: TH -YHT-N-E-T--- -TY SRNNNNWLDPAPH-$ S A R ~l'-..L......PyV......S....HTM-LCCPLCVAV~~~~~~~~~AAEATVPSEPIWFEQQCEVKADCFLCKFH~............

RCCHL IS T:CF D AH VW:---:CQQNV HV-Y--T--F--YYDVKMRUT. ..CC H D... ...V..........RCAR-ATCEDCVVMIWHER - CKWNDVPCNYQ~~~~~~~~~~~~~...... I'F:CKK...HTN-GK:ENCAV~~~~~~~~~SCAAN - CK~~~~~~POKRCRDK1~~~~~~-............HIGER .....V...S -....D K....C RP..KCR-CEREDCVH~~~~~Q-RMWN0NACCTAY-N..........SPL--HQE3{CVHIWDTKP - ----IQWNSNDCNVKM-CYICKPN...

SPLA Q-- KEQCVTES MYTDKNA -D CQWNN:iVV KHN-L :W::L G: GNICKFF,ASDRA RIAA M:S: PLCEK:R:EA YR.--::VN::C

........

AEL H IEQDC VVN D::--:QIFVT---TQNMKRD FYI RNKQNL -K

notextendthrough th NH.er.alpptd.o itotahneFg.5.hwwhich be-:Ars Qthe:inhiitorNActivityof4:Eth: poei owr casesN of chronc

CaCO..c.ystal.growth. .creatitis.but.n.Comparisonof.the.amio-termina end.ofrat.PA with..the .nomas,.or.panQH1seqence (Fi. 2) sugestedtht.the.tw proein were. levels... weecrprovided with: prppie fth aesz 26aioais, rncitcuPsinc thKetiNin Rnpoiin1RfQwasHtheV onlY: one.: toG human.: liVer Icofehestu Turl .eqiemnt o. reepidsi..ea m N frmt

chargedKain-trmnl, end andaVhGhyydoobcor CanI(Fg6(30adine mioci idnit.xtne.trug.h pre- Souher.bpeptdeceavae ste.o.preAP-.Cmue nlsso th .Suten.lo.

sequence-iniae that::maueQQ rtihad::VaGmolecularW: QH71R aftR retweihtof1656 ad aioeecri. pin of..9.. pepredasdeExpression- of thQH-G':.:KMRNAinDpanceasGandothe tis- Jtion ofYhight

sues. Pancreatic::RNDsfromuniesdtsu and acueKVnec- Q ments revealedrhemorrhagicpanceattisTCAweepoe inNorther blot:anal-t: T entinlowcopsi withR QH 1 antrp- oe-L E IcDNAGsHcotrlN(ig4.AQH1C LcaizaomRNwsotetctbl incotrl ancea, heeastrp- ie rase t t

Csinge I mRNA- wasabundant.: Conversly trypsnoge RPN used for detect

mRNA expression was low in the diseased pancreas, whereas blot analysis ofQH1 transcripts were very abundant. revealed a sing

10 - 4810 - 482 - 39

77 - 114148 - 18555 - 9245 - 82

159 - 196454 - 491409 - 446

1 - 3597 - 1331 - 371 - 39

49 - 8749 - 8440 - 79

115 - 145186 - 21693 - 12283 - 118

197 - 227492 - 521447 - 47836 - 74

134 - 16438 - 7440 - 73

88 - 11485 - 11080 - 113146 - 173217 - 246123 - 149119 - 146228 - 253522 - 548479 - 51175 - 104

165 - 19075 - 10174 - 98

115 - 148111 - 143114 - 152174 - 204247 - 278150 - 184147 - 179254 - 285549 - 581512 - 540105 - 137191 - 220101 - 12999 - 129

Figure 3. Sequence compari-son of PAP-H and severallectins. The PAP-H sequencewas aligned with sequencesof human pancreatic lithosta-thine (PSPH) ( 16) and thefollowing lectins: HTM, hu-man thrombomodulin (17);CHL, chicken hepatic lectin( 18); RHL, rat hepatic lectin(19); RCAR, rat cartilageproteoglycan (20); HTN,human tetranectin (21 );HIGER, human IgE receptor(22); HCSCP, human chon-droitin sulphate core protein(23); RKCR, rat Kupffer cellreceptor (24); SPL, Sarco-phaga peregrina lectin (25);DPSA, dog pulmonary sur-factant apoprotein (26); Ach,A chain of coagulation FactorIX/Factor X binding protein(28); ABL, acorn barnaclelectin (27). Shaded areascorrespond to amino acididentities. Amino acid num-bering of PAP-H is based onthe numbering of PAP-R(Fig. 2).

s that QH 1 transcripts could be detected in twopancreatitis and two cases of obstructive pan-

t in control pancreas, pancreatic adenocarci--reatic endocrine tumor. Trypsinogen I mRNAparable in all RNA samples studied. The QH 1d not be evidenced in mRNA extracted fromkidney, spleen, or mammary gland, nor inhe pancreatic cancer cell lines Mia-Paca and

lot analysis of the QHJ gene. Fig. 7 shows aanalysis of human genomic DNA probed withriction with five endonucleases. The blot wasscribed in Methods and washed under condi-ingency. The limited number ofrestriction frag-by the probe suggest that the QH 1 gene is pres-( numbers, perhaps as a single copy.InofQHI to the human pancreas. The antibod-synthetic NH2-terminal peptide ofQH 1 were

ng the protein in the diseased pancreas. Westernf pancreatic homogenate supernatant (Fig. 8)Ole polypeptide, with an apparent mol wt of

Human Pancreatitis-associated Protein 2287

Tg I1 2 3 4

1 2 34 5 6 7 Figure 6. Detection ofthe PAP-H transcriptin RNA from varioushuman tissues and celllines, by Northern blotanalysis. 10 ,g of RNAextracted from pancreaswith acute pancreatitis(lane 1), kidney (lane2), spleen (lane 3), liver(lane 4), mammarygland (lane 5), and thecell lines Mia-Paca (lane6) and Capan I (lane7) were processed as de-scribed in Methods andprobed with QH 1(PAP-H) cDNA.

Figure 4. Detection of the PAP-H transcript in pancreatic RNA fromcontrol tissue and acute pancreatitis by Northern blot analysis. 25 'gof pancreatic RNA from acute pancreatitis (lanes I and 3) or controltissue (lanes 2 and 4) were submitted to electrophoresis, transferredonto nylon membranes, and probed with QH 1 (PAP-H) cDNA (leftpanel) or human trypsynogen I cDNA (right panel). In each panel,lanes I and 2 corresponded to a 4-h exposure and lanes 3 and 4 to a3-d exposure.

- 16,000, compatible with the predicted size of mature QH 1.No signal was visible with normal pancreas. Immunolocaliza-tion on pancreatic sections from patients with acute pancreati-tis (Fig. 9) revealed that the apical regions of acinar cells werestrongly labeled whereas duct cells and fibroblasts were notstained. Again, no signal could be seen on sections of undis-eased pancreas.

Figure 5. Detection ofA the PAP-H transcript

in pancreatic RNA frompatients with variousdiseases by Northernblot analysis. 25 tgRNA were loaded ineach lane. Lanes I and2, control tissue; lanes3 and 4, pancreatic ade-nocarcinoma; lanes 5

B and 6, chronic pancre-atitis; lanes 7 and 9, ob-structive pancreatitis;lane 8, endocrine tu-mor. The filter wasprobed with QH 1(PAP-H) cDNA (A).The same filter was thendehybridized andprobed with humantrypsinogen I cDNA(B).

Serum concentration ofthe QH1 protein. Serum concentra-tion of QH 1 was measured in healthy subjects and patientspresenting with various pancreatic diseases (Fig. 10). QH1concentration was below 0.6 pmol/ml in all samples fromhealthy subjects and from 26 patients with other abdominaldiseases whose diagnoses are listed in Methods. In that group,two out of seven patients with stones in the common bile ductand two patients with cancer of the main bile duct showedmoderate increase. On the contrary, all samples from patientswith acute pancreatitis ( 19 cases) showed higher concentra-tions, values ranging from 15 to 82 pmol/ml. 7 of 13 patientswith chronic pancreatitis showed elevated values of serumQH 1 protein. In four of them hospitalized for recurrent pain,values were comparable to those ofpatients with acute pancre-atitis.

Discussion

Management ofacute pancreatitis has seen little improvementin the last decade. Future prospects seem better, however, be-cause ofthe wealth of information recently obtained in experi-mental models, concerning regulation of gene expression (2),modifications in hormonal response (31), and alterations inthe secretory pathway taking place during pancreatic inflamma-tion (32). A novel secretory protein (PAP), whose expressionincreases sharply during acute pancreatitis contrary to othersecretory products (5), was also described in the rat. Such aprotein might be ofclinical relevance as potential marker ofthedisease. Because many rat secretory proteins have their equiva-lent in human pancreas (33) we investigated whether a proteinsimilar to PAP was expressed in human tissue during pancre-atic inflammation. A classical strategy would have been to lookfor a PAP-H in the pancreatic juice of patients with acute pan-creatitis, but juice collection in those patients is impossible forethical reasons. We developed another strategy based on theassumption that if PAP-H does exist, human and rat PAPmRNA sequences should be similar enough to allow detectionby heterologous screening.

A single full-length cDNA (QH 1) was selected in the hu-man cDNA library upon screening with the rat PAP cDNA andsequencing (Fig. 1). Analysis of its nucleotide sequence re-vealed that the region homologous to rat PAP cDNA, which

2288 B. Orelle, V. Keim, L. Masciotra, J. C. Dagorn, and J. L. Iovanna

PAP:: :.:1. 2

.. .....

.;.. .... ....

.: ;. :. ::.. ..

..

:::

*: ::* : . :: ... ..

:... .. ::

::::::::

*: :: .. ::.:: ... . : : ::............ ... . .. .. ...*: .:: .:.. ..

........ .... .. .. ....

:.: ....

*::::::.

3 4

1 2 3 4 5 6 7 8 9. . . . . . . . ... ... ..... .......... ... .* . X* : ......... a,..j.,,,,,;....0E.E

T

wi

x~

CaCBQ C * S

XLLJX Z£

kb

400.

w9We-Sd,

4 I~~~~~~~~~~~~~~~~~~~db~~~~~~~~

bers, probably as a single copy, as shown for rat PAP (unpub-lished observations).

Whether the QH 1 protein shares with rat PAP the ability toinduce bacterial aggregation is impossible to verify at this time,because the purified protein is not available. It is noteworthy,however, that the QH 1 protein has retained the structural do-mains specific of calcium-dependent lectins (Fig. 3), alreadycharacterized in rat PAP, which suggests a similar function.

The most important characteristic of the rat PAP gene, be-cause of its potential implications, is a very low expression innormal pancreas and a dramatic increase during the acutephase of the disease. Presence of a single clone encoding QH 1in the cDNA library constructed with mRNA from controltissue was a first indication that the QH 1 transcript was alsoexpressed at a low level in the absence ofinflammation. In fact,that transcript was not detectable by Northern blot analysis incontrol RNA, in conditions where trypsinogen gave a strongsignal (Figs. 4 and 5). In a similar analysis of RNA extractedfrom a tissue fragment resected during necrohemorrhagic pan-creatitis, the opposite situation was observed with a faint signalfor trypsinogen and a strong hybridization with QH 1 (Fig. 4).Decreased expression of trypsinogen and other pancreatic en-zymes was reported in the rat during acute pancreatitis andconsidered as part of a defense mechanism including PAPoverexpression (2). A similar mechanism might thus exist inhumans, QH 1 overexpression during acute pancreatitis beingcomparable to that ofrat PAP during the acute phase ofexperi-mental pancreatitis. On the other hand, QH 1 expression wasnot detectable in human pancreatic adenocarcinoma or in apancreatic endocrine tumor, nor could it be detected in humancancer cell lines ofpancreatic origin (Fig. 6) but was induced atlow level in obstructive pancreatitis and some cases of chronicpancreatitis (Fig. 5). Such inductions are therefore probablycaused by the focal inflammatory lesions associated with bothdiseases rather that to the primary diseases themselves. No ex-

A BkD

Figure 7. Analysis of the PAP-H gene(s) by Southern blotting. Frag-ments of human genomic DNA generated by endonuclease restrictionwere probed with QH I (PAP-H) cDNA. Migration of size markersis indicated on the left.

allowed selective hybridization, extended from around nucleo-tide 45 to nucleotide 565, which corresponded in fact to thecoding region of the transcript. The protein encoded by QH 1

contained the same number ofamino acids as rat PAP (Fig. 2)and the six cysteine residues were located in identical positionssuggesting a similar organization ofthe disulphide bridges. Thetwo sequences had an overall amino acid identity of 71%,which is in the range of values obtained when comparing re-

lated pancreatic proteins in rats and in humans (34). Relativeconservation ofthe coding regions ofQH1 and rat PAP mRNAand lack of homology between their noncoding regions couldreflect differences in selective pressure. Genomic analysis (Fig.7) revealed that the QH 1 gene was present in low copy num-

21'-

Figure 8. Immunodetection ofPAP inpancreatic homogenate from acutepancreatitis (A) and control tissue(B). The antibody raised to theNH2-terminal peptide of PAP-H was

used to detect the protein in pancre-atic homogenate supernatant (West-ern blot analysis). Position of molec-ular weight markers is indicated tothe left.

Human Pancreatitis-associated Protein 2289

21.0

4.2.'3.5

2.0 a"

1.5 a"

94'-67'-

Figure 9. Immunolocalization of PAP-H in pancreatic tissue. The protein was immunodetected in thin sections of normal pancreas (A) andpancreas with acute pancreatitis (B), with the antibody to the NH2-terminal peptide of the protein.

pression was observed in liver, spleen, kidney, or mammarygland (Fig. 6).

Further characterization of the QH1 protein was under-taken with antibodies raised to a synthetic peptide in theamino-terminal region ofthe QH1 sequence. Western blot anal-ysis of proteins from pancreatic homogenates of patients with

20 0

0

0

0

10

0 ~(32) 6 - 2

A B C D

Figure 10. Quantifica-tion ofPAP-H in serum.PAP-H was assayed (A)in healthy individuals(control group, n = 34);(B) in patients present-ing with acute pancre-atitis (n = 19); (C) inpatients with chronicpancreatitis (n = 13);and (D) in patients withother abdominal dis-eases (n = 30, corre-sponding diagnoses arelisted in Methods).PAP-H concentrationsare given in picomolesper milliliter (1 pmol= 16.5 ng). Figures inbrackets correspond tothe number of cases ineach group with PAP-Hlevels < 0.6 pmol/ml.

acute pancreatitis revealed a single band, at a position compati-ble with the size of QH 1 ( 16 kD), whereas no signal could bedetected in homogenates ofnormal tissue (Fig. 8). The proteincould also be immunolocalized to the apical region of pancre-atic acinar cells, as expected for a secretory protein (Fig. 9).Altogether, these results indicate that QH 1 and PAP are verysimilar in their structure and in their pattern of expressionduring acute pancreatitis. QH 1 represents, therefore, an equiva-lent of rat PAP in human and will be henceforth namedPAP-H.

The limited number of tissue samples available for acutepancreatitis made it difficult to estimate the range of PAP-Hexpression during pancreatic inflammation, information ob-tained in the rat by measuring the rate ofPAP synthesis (3, 5).We used an indirect approach, based on the fact that all pancre-atic secretory proteins reach the bloodstream during the acutephase of pancreatitis. Measuring the increase in serum PAP-Hconcentration would provide therefore a conservative estimateof PAP-H overexpression in pancreas. An immunoassay wasconstructed using the antibody to the NH2-terminal end oftheprotein and serum PAP-H was quantified in healthy subjectsand a series of patients with pancreatitis and other abdominaldiseases (Fig. 10). PAP-H serum levels were at near back-ground values in healthy subjects and in patients with abdomi-nal diseases other than pancreatitis but increased dramaticallyin all cases ofacute pancreatitis and during acute exacerbationsofchronic pancreatitis. Increases were 25-140 times over back-ground, in good agreement with the 100-300 times increasesobserved in rat (3) and data on PAP-H gene expression (Figs. 4and 5). These results confirm the important overexpression of

2290 B. Orelle, V. Keim, L. Masciotra, J. C. Dagorn, and J. L. Iovanna

0

0

0

100

90

80-

70

60

E 500ECf 40-

30

00

0,

0,

OD

0

0

the protein during pancreatic inflammation. They also suggestthat PAP-H, which is not detectable in serum ofhealthy individ-uals and appears in large amounts in patients with pancreaticinflammation, might be a very useful marker of acute pancre-atitis.

Acknowledgments

Technical assistance of P. Garrido and A. Sansonetti is gratefully ac-knowledged.

J. L. lovanna was supported by a grant from the Fondation pour laRecherche Medicale (Paris) and V. Keim was supported by grant Ke347/3-1 from the Deutsche Forschungsgemeinschaft.

References

1. Gyr, K. E., M. Singer, and H. Sarles. 1984. Revised classification ofpancre-atitis. In Pancreatitis, Concepts and Classification. K. E. Gyr, M. Singer, and H.Sarles, editors. Excerpta Medica, Amsterdam. XXIII-XXV.

2. lovanna, J., V. Keim, R. Michel, and J. C. Dagorn. 1991. Pancreatic geneexpression is altered during acute experimental pancreatitis in the rat. Am. J.Physiol. 261:G485-G489.

3. lovanna, J., B. Orelle, V. Keim, and J. C. Dagorn. 1991. Messenger RNAsequence and expression of rat Pancreatitis-Associated Protein (PAP), a lectin-related protein overexpressed during acute experimental pancreatitis. J. Biol.Chem. 266:24664-24669.

4. Rouquier, S., J. M. Verdier, J. Iovanna, J. C. Dagorn, and D. Giorgi. 1991.Rat pancreatic stone protein messenger RNA. Abundant expression in matureexocrine cells, regulation by food content, and sequence identity with the endo-crine reg transcript. J. Biol. Chem. 266:786-791.

5. Keim, V., J. lovanna, G. Rohr, K. H. Usadel, and J. C. Dagorn. 1991.Characterization ofa rat pancreatic secretory protein associated with pancreatitis.Gastroenterology. 100:775-782.

6. Huynh, T. V., R. A. Young, and R. W. Davies. 1985. Constructing andscreening cDNA libraries in Xgt 10 and Xgt 11. In DNA Cloning: A PracticalApproach. D. M. Glover, editor. IRL Press Ltd., Oxford. I:49-78.

7. Feimberg, A. P., and B. Volgestein. 1983. A technique for radiolabellingDNA restriction endonuclease fragments to high specific activity. Anal. Biochem.132:6-13.

8. Gyllenstein, U. B., and H. A. Erlich. 1988. Generation of single-strandedDNA by the polymerase chain reaction and its application to direct sequencing ofthe HLA-dqa locus. Proc. Natl. Acad. Sci. USA. 85:7652-7656.

9. Wilbur, W. J., and D. J. Lipman. 1983. Rapid similarity searches ofnucleicacids and protein data banks. Proc. Nall. Acad. Sci. USA. 80:726-730.

10. Goad, W. B., and M. I. Kanehisa. 1982. Pattern recognition in nucleicacid sequences. I. A general method for finding local homologies and symmetries.Nucleic Acids Res. 10:247-263.

1 1. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. In Molecular Cloning:A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.545 pp.

12. Dorfman, D. M., L. I. Zon, and S. H. Orkin. 1989. Rapid amplification ofXgt 11 bacteriophage library inserts from plaques using the polymerase chainreaction (PCR). Biotechniques. 7:568-570.

13. Chirgwin, J. M., A. Przybyla, R. J. MacDonald, and W. J. Rutter. 1979.Isolation of biologically active ribonucleic acid from sources enriched in ribonu-clease. Biochemistry. 24:5294-5299.

14. Harlow, E., and D. Lane. 1988. In Antibodies: A Laboratory Manual.Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 86-87.

15. Sternberger, L. A., P. H. Hardy, J. J. Cuculis, and H. G. Meyer. 1970. Theunlabelled antibody enzyme method ofimmunochemistry. Preparation and prop-erties ofsoluble antigen-antibody complex (horseradish peroxidase-antihorserad-

ish peroxidase) and its use in identification of spirochetes. J. Histochem. Cyto-chem. 18:315-333.

16. Giorgi, D., J. P. Bernard, S. Rouquier, J. lovanna, H. Sarles, and J. C.Dagorn. 1989. Secretory pancreatic stone protein messenger RNA. Nucleotidesequence and expression in chronic calcifying pancreatitis. J. Clin. Invest.84:100-106.

17. Wen, D., W. Dittman, Y. Ye, L. Deaven, P. Majerus, and E. Sadler. 1987.Human thrombomodulin: complete cDNA sequence and chromosome localiza-tion of the gene. Biochemistry. 26:4350-4357.

18. Drickamer, K. Complete amino acid sequence ofa membrane receptor forglycoproteins. Sequence or the chicken hepatic lectin. J. Biol. Chem. 256:5827-5839.

19. Drickamer, K., J. Mamon, G. Binn, and J. Leung. 1984. Primary structureof the rat liver asialoglycoprotein receptor. Structural evidence for multiple poli-peptide species. J. Biol. Chem. 259:770-778.

20. Doege, K., P. Fernandez, J. Hossell, M. Sasaki, and Y. Yamada. 1986.Partial cDNA sequence encoding a globular domain at the c-terminus of the ratcartilage proteoglycan. J. Biol. Chem. 261:8108-81 1 1.

21. Fuhlendorff, J., I. Clemmensen, and S. Magnusson. 1987. Primary struc-ture of tetranectin, a plasminogen kringle 4 binding protein: homology with asia-loglycoprotein receptors and cartilage proteoglycan core protein. Biochemistry.26:6757-6764.

22. Kikutani, H., S. Inui, R. Sato, E. Barsumian, H. Owaki, K. Yamasaki, T.Kaisho, N. Uchibayashi, R. Hardi, T. Hirano, et al. 1986. Molecular structure ofhuman lymphocyte receptor for immunoglobulin E. Cell. 47:657-665.

23. Krusius, T., K. Gehlsen, and E. Ruoslahti. 1987. A fibroblast chondroitinsulphate proteoglycan core protein contains lectin-like and growth factor-likesequences. J. Biol. Chem. 262:13120-13125.

24. Hoyle, G., and R. Hill. 1988. Molecular cloning and sequencing of acDNA for a carbohydrate binding receptor unique to rat Kupffer cell. J. Biol.Chem. 263:7487-7492.

25. Takahashi, H., H. Komano, N. Kawagushi, N. Kitamura, S. Nakanishi,and S. Natori. 1985. Cloning and sequencing of cDNA of sarcophaga peregrinahumoral lectin induced on injury of the body wall. J. Biol. Chem. 260:1228-1233.

26. Benson, B., S. Hawgood, J. Schilling, J. Clements, D. Damm, B. Cordel,and T. White. 1985. Structure ofcanine pulmonary apoprotein cDNA and com-plete amino acid sequence. Proc. Natl. Acad. Sci. USA. 82:6379-6387.

27. Muramoto, K., and H. Kamiya. 1986. The amino-acid sequence of alectin of acorn barnacle megabalanus rosa. Biochim. Biophys. Acta. 874:285-295.

28. Atoda, H., M. Hyuga, and T. Morita. 1991. The primary structure ofcoagulation factor IX/factor X-binding protein isolated from the venom of Tri-meresurusflavovirides. Homology with asialoglycoprotein receptors, proteogly-can core protein, tetranectin and lymphocyte Fc receptor for immunoglobulin E.J. Biol. Chem. 266:14903-14911.

29. Drickamer, K. 1988. Two distinct classes of carbohydrate-recognitiondomains in animal lectins. J. Biol. Chem. 263:9557-9560.

30. Von Heijne, G. 1984. Analysis of the distribution of charged residues inthe N-terminal region of signal sequences: implications for protein export inprokariotic and eukariotic cells. EMBO (Eur. Mol. Biol. Organ) J. 3:2315-2318.

31. Niederau, C., M. Niederau, R. Luthen, G. Strohmeyer, L. Ferrell, and J.Grendell. 1990. Pancreatic exocrine secretion in acute experimental pancreatitis.Gastroenterology. 99:1120-1127.

32. Steer, M., and J. Meldolesi. 1989. The cell biology ofacute pancreatitis. N.Engl. J. Med. 316:144-150.

33. Scheele, G. 1986. Two-dimensional electrophoresis in the analysis ofexo-crine pancreatic proteins. In The Exocrine Pancreas. Biology, Pathobiology andDiseases. V. L. W. Go, J. D. Gardner, F. P. Brooks, E. Liebenthal, E. P. DiMagno, and G. A. Scheele, editors. Raven Press Ltd., New York. 185-192.

34. Scheele, G. 1986. Regulation of gene expression in the exocrine pancreas.In The Exocrine Pancreas. Biology, Pathobiology and Diseases. V. L. W. Go,J. D. Gardner, F. P. Brooks, E. Liebenthal, E. P. Di Magno, and G. A. Scheele,editors. Raven Press Ltd., New York. 55-67.

Human Pancreatitis-associated Protein 2291