ORIGINAL ARTICLE Tissue Distribution and Plasma Concentration of HumanGuanylin Yukari Date, Masamitsu Nakazato, Hideki Yamaguchi, Mikiya Miyazato and Shigeru Matsukura Guanylin, a peptide homologue of the bacterial heat-stable enterotoxins, is an endogenous activator ofguanylate cyclase C (GC-C). Wedetermined the tissue content and plasma concentra- tion of human guanylin, and its cellular source in the intestine. Humanguanylin is distributed widely from the duodenum to the rectum, the highest content being in the ileum and proximal colon. The plasma concentration of immunoreactive guanylin in the normal individuals tested was 30.3±3.7 fmol/ml (mean±SE) and that in patients with chronic renal failure was elevated with increasing serum creatinine concentration. Guanylin immunoreactivity was detected in the villus epithelial cells in the small intestine and these guanylin-containing cells were increased in number along the cephalocaudal axis of the gut. Guanylin was also present in Paneth cells in the small intestine and superficial epithelial cells in the large intestine. Guanylin mRNAwas detected in the intestine by the reverse transcription-polymerase chain reaction. Guanylin may have paracrine action on neighboring enterocytes, activating intestinal guanylate cyclase and thereby regulating intestinal fluid as well as electrolyte transport through the second messenger, cyclic GMP. (Internal Medicine 35: 171-175, 1996) Key words: guanylate cyclase C, enterotoxin, natriuretic polypeptide, electrolyte and water bal- ance, surface epithelial cell, Paneth cell Introduction coli,The heat-stableYercinia enterocoliticaenterotoxins and(STs)VibrioproducedcholeraebyareEscherichiaall 15-30 amino acid peptides. These enterotoxins cause acute diarrhea, whichdevelopingreportedlycountriesaccounts(1). STsforbindup toto intestine-specific50%of infant deathsrecep-in tor-guanylateand mediate the cyclasestimulation(GC-C)of Cl~in secretionthe brush andborderthe inhibitionmembrane (2).of Na+Guanylinand H2Oisabsorption,a 15-amino acidtherebypeptidecausingthat secretoryhas recentlydiarrheabeen colonpurifiedcarcinomafrom ratcelljejunumline whoseusingGC-Ca bioassayresponds ofto aSTsT84withhumanan aminoincreaseacidin intracellularsequence homologycyclic withGMP(CGMP)Escherichia(3). It colihas a ST.47% clase,Guanylinstimulatehas beenCl"shownsecretion,to activateand displaceintestinalST bindingguanylateto cy-T84 andcells electrolyte(4). It may transportbe a physiologicaland mayregulatorbe involvedof intestinalin the patho-fluid ingphysiologyof the physiologicalof diarrheal diseases.functions Aandmorepathophysiologicalcomplete understand-sig- requiresnificancedetailedof guanylinanalysisinofwaterits celland andelectrolytetissue distributionhomeostasisin Weprepared humans.an antiserum raised against synthetic rat guanylin bothand establishedhuman and rata radioimmunoassayguanylin, then identified(RIA) thathumancan quantifyguanylin andresultsits precursorsof the regionalin thedistributionintestine andplasmaand plasma(5).concentrationHere we reportof humanreverseguanylin.transcription-polymeraseWeinvestigated its genechainexpressionreaction (RT-PCR).using the Weofalsoguanylindeterminedin theimmunohistochemicallyintestine. the cellular source Materials and Methods TheQuantificationgastrointestinalof thetractguanyiinfrom thecontentstomachin totissuethe colon and kidneydiedwereof resectedischemic atheartautopsydisease.fromTissuesthree patientswere excisedwho andhad volumeimmediatelyof water heatedto inactivatefor 10 intrinsicminutes atproteinases.95-100°C inAftera 10-foldcool- From the Third Department of Internal Medicine, Miyazaki Medical College, Miyazaki Received for publication June 29, 1995; Accepted for publication December 5, 1995 Reprint requests should be addressed to Dr. Masamitsu Nakazato, the Third Department of Internal Medicine, Miyazaki Medical College, 5200 Kihara, Kiyotake, Miyazaki 889-16 Internal Medicine Vol. 35, No. 3 (March 1996) 171 Date et al ing the tissue samples to 4°C, CH3COOHand HC1were added MD).Reverse transcription was done for 30 minutes at 42°C to the respective final concentrations of 1 Mand 20 mM.The then incubation for 3 minutes at 94°C to inactivate the reverse tissues then were homogenized in a polytron for 10 minutes, transcriptase. The resulting CDNAwas subjected to PCRampli- after which the homogenate was centrifuged at 20,000 x g for fication with the 2 jiM sense and antisense primers for guanylin 30 minutes at 4°C. The supernatants were applied to Sep-Pak C- and (3-actin, and the 2.5 U Taq DNApolymerase (Perkin-Elmer 1 8 cartridges (Waters, Milford, CA), which were washed with Cetus, Foster City, CA). The sequences of the oligonucleotides 0.5 MCH3COOHand 10% acetonitrile (CH3CN) solution (first sense then antisense) for the guanylin and (3-actin PCRs containing 0. 1% trifluoroacetic acid (TFA), and the peptides were guanylin, 5'-ATGAATGCCTTCCTGCTCTTCGCA-3' were eluted with 60% CH3CNsolution containing 0. 1% TFA. and 5'-TACAGGCAGCGTAGGCACAGATTT-3 ; p-actin, 5'- Cartridge eluates corresponding to 100 mgwet weight were ATCCAGGCTGTGCTATCCCTGTAC-3' and 5'-TCATTG- lyophilized and subjected to the RIA for guanylin. CCAATGGTGATGACCTGG-3'.The PCR conditions were 34 cycles of denaturation at 94°C for 1 minute, annealing at Determination of plasma guanylin concentration 58°C for 2 minutes, and extension at 72°C for 2 minutes. PCR Blood was obtained from 30 normal individuals ( 15 men and products were electrophoresed on a 2%agarose gel (FMC 15 women, aged 18^0 years) and 23 patients (ll men and 12 BioProducts, Rockland, ME)then made visible with ethidium women, aged 37-58 years) with chronic renal failure (CRF) bromideunderultraviolet. whose serum creatinine levels were above 1.5 mg/dl. These patients did not undergo hemodialysis. After the plasma was Immunohistochemistry diluted to one-half with 0.9% saline, it was applied to a Sep-Pak Humangastrointestinal tracts obtained from the same 3 C-18 cartridge which had been pre-equilibrated with 0.9% patients described above were fixed with 3.7% formaldehyde in saline. The cartridge was washedwith saline and 10%CH3CN 10 mMphosphate-buffered saline (PBS, pH 7.2) and, after solution containing 0.1% TFA. The adsorbed peptides were dehydration in a graded ethanol series, they were embedded in eluted with 60% CH3CNsolution containing 0. 1% TFA, after paraffin. Sections cut 3 jam thick were deparaffinized in xylene which the eluate was evaporated and used for the RIA. then rehydrated in a graded ethanol series, after which they were treated with 0.3% hydrogen peroxide for 30 minutes to inacti- RIA for guanylin vate endogenous peroxidases. Nonspecific binding was blocked The antiserum for rat guanylin was raised in NewZealand with normal goat serum. Anti-guanylin antiserum was used at white rabbits by the carbodiimide methodagainst synthetic rat the final dilution of 1/500 in 10 mMPBS containing 1% bovine guanylin conjugated with bovine thyroglobulin (5). It was used serum albumin (BSA) and 5% normal goat serum. The anti- at a final dilution of 1/1 ,000. Tyr°-rat guanylin was radioiodinated serum was allowed to react with the preparations overnight at by the lactoperoxidase method, after which the labeled peptide 4°C in a moist chamber. Sections first were incubated for 90 was purified by reverse-phase high performance liquid chroma- minutes with goat-biotinylated anti-rabbit IgG (Vectastain, tography (RP-HPLC). The half-maximum inhibition by human Vector Lab., Burlingame, CA) then for 45 minutes with peroxi- guanylin on the standard RIAcurve was 1 30 fmol/tube, and the dase-conjugated streptavidin (GIBCO BRL) diluted 1/200 in peptide was detectable at the low level of 1 1 fmol/tube (10% 10 mMPBS containing 1% BSA. They then were stained for 3 replacement). The antiserum did not crossreact with human minutes at room temperature with 20 mgof 3,3'-diamino- uroguanylin, ANP, BNP or CNP. The specificity of the antise- benzidine tetrahydrochloride and 0.006% hydrogen peroxide in rumfor guanylin molecules was confirmed by the analysis of 50 mMTris buffer solution (pH 7.2), after which they were immunoreactive substances in humanplasma and intestinal counterstained with methyl green. In the sequential double tissue extract by the use ofRP-HPLCcoupled with the RIA (5). staining for guanylin vs lysozyme, chromogranin A, serotonin, Therespective intra- and interassay coefficients of variation glucagon, pancreatic polypeptide, vasoactive intestinal polypep- were 3.2%and 4.2%.tide and somatostatin, guanylin first was stained using the streptavidin-peroxidase method, then the specimens were RT-PCRamplification of the guanylin transcript washed with 0.1 Mglycine-HCl buffer (pH 2.2) and stained The total RNAsof the humancolon, gastric antrum and with these proteins and peptides by the streptavidin-alkaline kidney were extracted by the acid-guanidium thiocyanate- phosphatase method using a Labeled Streptavidin Biotin Kit phenol-chloroform method (6). Three units of RNase-free and an AP Substrate Kit III (Vector Lab.). Control studies were DNase I (Pharmacia, Piscataway, NJ), 1 10 U RNase inhibitor, done with normal rabbit serum or anti-guanylin antiserum that 40 mMTris HC1 (pH 7.6) and 6 mMMgCl2 were added to the had been preabsorbed with 1 jig of human guanylin. 2.5 jig RNAsamples to digest the genomic DNAintroduced into the RNAfraction. The first strand CDNAwas synthesized Statistical analysis with 0.4 jig of the RNAsamples that had been treated with Comparison between groups of patients was done