091420.715 091420.660 032519.438 032519.469 132103.830 190503.200 160907*500 131521*030

Cellular localization of cystic fibrosis transmembrane regulator proteinin piglet and mouse intestineUrsula L. Hayden, Hannah V. Carey

Department of Comparative Biosciences, University of Wisconsin, School of Veterinary Medicine,2015 Linden Drive West, Madison, WI 53706, USA

&misc:Received: 4 May 1995 / Accepted: 6 September 1995

&p.1:Abstract. Antibodies raised against the cystic fibrosistransmembrane regulator protein (CFTR) were used tolocalize CFTR in intestinal tissues of piglets and mice.Positive staining for CFTR was detected in goblet cells ofboth species. A second population of epithelial cells ofunknown phenotype was also labeled by anti-CFTR anti-bodies. The labeling pattern was abolished by preincuba-tion of anti-CFTR antibodies with the immunogen orwhen non-immune IgG was used in place of anti-CFTRantibodies. These results support other studies that sug-gest that alterations in goblet cell function may be in-volved in the intestinal abnormalities associated with cys-tic fibrosis.

&kwd:Key words: Cystic fibrosis transmembrane regulator –Cystic fibrosis – Intestine – Mucus secretion – Pig–Mouse

Introduction

Cystic fibrosis (CF) is characterized by abnormalities ofthe respiratory, gastrointestinal and other organ systemsthat are lined with epithelial tissues. Manifestations ofthe disease include alterations in fluid and electrolytetransport and excessive mucus accumulation in the air-ways and intestinal lumen (Berschneider et al. 1988;O’Loughlin et al. 1991; McCray et al. 1992; Fuller andBenos 1992). Defective functioning of the ~170 kDaprotein known as the cystic fibrosis transmembrane reg-ulator (CFTR) has been implicated in the pathogenesisof CF (Fuller and Benos 1992; Clark et al. 1992). CFTRfunctions as an adenosine 3',5'-cyclic monophosphate(cAMP)-regulated Cl− channel that is responsible forvectorial secretion of salt and water (Anderson et al.1992). Transgenic mice that are deficient in the CFTRgene show clinical symptoms that mimic several aspectsof CF pathology, including abnormalities in intestinalstructure and function (Snouwaert et al. 1992). Yet, de-

spite intensive study, uncertainty still surrounds the tis-sue localization of CFTR in the gut and the specific celltype(s) with which it is associated. CFTR protein hasbeen localized on apical membranes of epithelial cells inthe small and large intestines, particularly in the cryptregion (Gaillard et al. 1994; Crawford et al. 1991). Insitu hybridization studies support the expression ofCFTR mainly in the intestinal cell population responsi-ble for Cl− secretion (i.e., crypt cells); however, CFTRmRNA is present in cells of unknown phenotype scat-tered throughout the villus epithelium of the human duo-denum and jejunum (Strong et al. 1994). Furthermore,CFTR protein has been detected in goblet cells of adultand fetal human intestine (Hoogeveen et al. 1991; Gail-lard et al. 1994). Knowledge of the distribution of CFTRin organs affected by CF is of obvious importance in un-derstanding the pathophysiology of the disease. Thus,the aim of the present study was to determine the local-ization of CFTR in the intestinal epithelium of mice andpiglets using anti-CFTR antibodies.

Materials and methods

Antibodies

The anti-CFTR antibody used is a rabbit polyclonal IgG, purifiedfrom sera by peptide affinity chromatography specific for peptide785 located in the R domain of the CFTR protein (Fuller et al.1992). The antibody recognizes a 170–180 kDa band in T84 cellsby immunoprecipitation. Specificity of the immunostaining wasconfirmed by prior incubation of antisera for 3 h at 37° C withpeptide 785 (100µg/ml), or by substituting the antisera with non-immune rabbit IgG. The monoclonal antibody against chromagra-nin A, a protein characteristic of enteroendocrine cells, was pur-chased from Boehringer Mannheim.

Tissue preparation and immunocytochemistry

Piglets were fed a milk replacer diet from days 10 to 21 and werethen euthanized by electrical shock. Mice were given food and wa-Correspondence to:H.V. Carey&/fn-block:

Cell Tissue Res (1996) 283:209–213

© Springer-Verlag 1996

210

ter ad libitum and were sacrificed by cervical dislocation. Seg-ments of mouse and piglet jejunum and distal colon obtained aftereuthanasia were fixed for 3–5 h at room temperature in 10% for-malin or Zamboni’s fixative (similar results were obtained usingboth fixatives). Paraffin-embedded tissues were sectioned (6µm)and mounted on poly-L-lysine coated slides, deparaffinized andthen rehydrated. Endogenous peroxidase activity was eliminatedby incubation in 0.03% hydrogen peroxide for 10 min. Sections

were then rinsed in phosphate-buffered saline (PBS, 0.01 M) andincubated in 2% goat serum for 30 min to block non-specific bind-ing sites. Sections were incubated overnight with anti-CFTR anti-bodies (10µg/ml) or non-immune rabbit IgG (10µg/ml) at 4° C.Tissues were then rinsed with PBS and incubated with biotinylatedgoat anti-rabbit IgG for 30 min at room temperature, followed by a30 min incubation with avidin-biotin complex (ABC Kit, VectorLabs). Immunoreactive sites were visualized using diaminobenzi-

Fig. 1a–d.Jejunal tissues immunostained for CFTR. a Piglet; b mouse. Adjacent sections were incubated with non-immune rabbit IgG: cpiglet, d mouse. Scale bars:50 µm except for a, which is 20µm&/fig.c:

dine as chromogen. Some sections were incubated with both anti-CFTR and anti-chromogranin A antibodies. Chromogranin A wasvisualized using secondary antibodies conjugated to FITC.

Results

The majority of cells in piglet and mouse jejunum la-beled by anti-CFTR antibodies were characterized byprominent apically-located mucus granules, suggesting

that they were goblet cells (Figs. 1a, 1b). Other CFTR-positive cells that lacked a large mucus granule (e.g.,Fig. 1b) may represent the “uncharacterized” subset ofepithelial cells recently described by Ameen et al.(1995). Closer examination of CFTR-labeled goblet cellsin piglet jejunum revealed labeling that was restricted tothe base of the goblet cells, leaving the mucus granulesunstained (Fig. 2). No positive staining was detectedwhen non-immune rabbit IgG was used in place of anti-CFTR antibodies (Figs. 1c, 1d). Furthermore, no stain-

211

Fig. 2. Higher magnification of CFTR-labeled goblet cells in pig-let jejunum. Scale bar:10 µm&/fig.c:

Fig. 3. Preincubation of the anti-CFTR antibody with peptide 785resulted in negative immunostaining in piglet jejunum. Tissue sec-tion is from same animal as used in Fig. 1. Scale bar:50 µm&/fig.c:

Fig. 4 a, b. Adjacent sections of piglet distal colon immuno-stained for a CFTR and b non-immune rabbit IgG. Scale bar:20µm for a and 50µm for b

212

ing of piglet jejunum was observed when anti-CFTR an-tibodies were pre-incubated with peptide 785 (Fig. 3). Inpiglet distal colon CFTR was detected in the crypt re-gion in cells of goblet-like appearance and in cells of un-known phenotype (Fig. 4a). No staining was observedwhen colonic tissues were incubated with non-immunerabbit IgG (Fig. 4b). No co-localization of CFTR andchromagranin A was observed in either jejunum or colon(data not shown).

Discussion

Cystic fibrosis, the most common lethal genetic defect inthe Caucasian population, is characterized by chroniclung infections and epithelial cell dysfunction in severalorgan systems (Fuller and Benos 1992). Mutations in theCF gene product, CFTR, appear to be responsible formost if not all of the manifestations of the disease. Theaim of this study was to determine the cellular distribu-tion of CFTR in piglet and mouse intestinal epithelia.This work focused on the intestinal tract because intesti-nal obstruction due to mechanism ileus is an early com-plicaton of CF in newborns. Piglets were used becausethey represent an ideal model for gastrointestinal studiesin neonates (Moughan et al. 1992). Mice were utilizedbecause of the development of the CFTR (−/−) mousethat contains mutations in the CFTR gene, leading to re-duced expression of CFTR and clinical symptoms thatmimic several aspects of CF pathology (e.g., Snouwaertet al. 1992).

Immunochemical localization of CFTR was detectedin goblet cells of the intestinal epithelium in both speciesstudied. This finding is of interest because one hallmarkof cystic fibrosis is the excessive accumulation of mucuswithin the intestinal tract. Indeed, expression of humanCFTR in CF mutant mice partially corrected the intesti-nal goblet cell hyperplasia characteristic of that strain(Zhou et al. 1994). Recent evidence suggests that mucinsynthesis and secretion is regulated in part by CFTR(Kuver et al. 1994) and that CFTR may provide a func-tional link between Cl− and mucus secretion (Englehardtet al. 1994), although this may not be the case for all celltypes (Jarry et al. 1994). The finding that the human co-lonic cell line HT29-Cl.16E expresses mucin exocytosisand a cAMP-regulated Cl− conductance (probably medi-ated by CFTR) provides further support for the presenceof CFTR in goblet cells (Jarry et al. 1994). Thus, it islikely that defective CFTR function in intestinal gobletcells contributes to the pathogenesis of CF.

It is unclear why the anti-CFTR antibodies failed tolabel apical membranes of intestinal epithelial cells, giv-en that other studies have demonstrated an apical local-ization of CFTR (see Introduction) and the evidence thatCFTR regulates an apical Cl− conductance in epithelia(Anderson et al. 1992; Fuller and Benos 1992). Howev-er, our results are consistent with those of Ameen et al.(1995) who did not detect CFTR on apical membranesof rat and human enterocytes but did find CFTR in scat-tered cells of unknown phenotype. Our results also agree

with other reports that demonstrate CFTR protein in in-testinal goblet cells (Hoogeveen et al. 1991; Gaillard etal. 1994). The lack of co-localization between CFTR andchromogranin A suggests that these cells are not entero-endocrine cells. Further studies are needed to elucidatethe function of non-goblet epithelial cells labeled by an-ti-CFTR antibodies. However, the present results do sup-port the idea that intestinal goblet cell function may beregulated by CFTR in both normal and pathophysiologi-cal states.

&p.2:Acknowledgements.The authors thank Drs. Dale Benos and Cath-erine Fuller for the anti-CFTR antibodies, the 785 peptide andhelpful discussions.

References

Ameen NA, Ardito T, Kashgarian M, Marino CR (1995) A uniquesubset of rat and human intestinal villus cells express the cys-tic fibrosis transmembrane conductance regulator. Gastroen-terology 108:1016–1023

Anderson MP, Sheppard DN, Berger HA, Welsh MJ (1992) Chlo-ride channels in the apical membrane of normal and cystic fi-brosis airway and intestinal epithelia. Am J Physiol 263 (LungCell Mol Physiol 7):L1–L14

Berschneider HM, Knowles MR, Azizkhan RG, Boucher RC, To-bey NA, Orlando RC, Powell DW (1988) Altered intestinalchloride transport in cystic fibrosis. FASEB J 2:2625–2629

Clarke LL, Grubb BR, Gabriel SE, Smithies O, Koller BH, Bou-cher RC (1992) Defective epithelial chloride transport in agene-targeted mouse model of cystic fibrosis. Science 257:1125–1128

Crawford I, Maloney PC, Zeitlin PL, Gugging WB, Hyde SC,Turley H, Gatter KC, Harris A, Higgins CF (1991) Immuno-cytochemical localization of the cystic fibrosis gene productCFTR. Proc Natl Acad Sci USA 88:9262–9266

Englehardt JF, Smith SS, Allen E, Yankaskas JR, Dawson DC,Wilson JM (1994) Coupled secretion of chloride and mucus inskin of Xenopus laevis:possible role for CFTR. Am J Physiol267 (Cell Physiol 36):C491–C500

Fuller CM, Benos DJ (1992) CFTR! Am J Physiol 263 (CellPhysiol 32):C267–C286

Fuller CM, Howard MB, Bedwell DM, Fizzel RA, Benos DJ(1992) Antibodies against the cystic fibrosis transmembraneregulator. Am J Physiol 262 (Cell Physiol 31):C396–C404

Gaillard D, Ruocco S, Lallemand A, Dalemans W, Hinnrasky J,Puchelle E (1994) Immunohistochemical localization of cysticfibrosis transmembrane conductance regulator in human fetalairway and digestive mucosa. Ped Res 36:137–143

Hoogeveen AT, Keulemans J, Willwmsen R, Scholte BJ, Bijman J,Edixhoven MJ, DeJonge HR, Galjaard H (1991) Immunologi-cal localization of cystic fibrosis candidate gene products. ExpCell Res 193:435–437

Jarry A, Merlin D, Hopfer U, Laboisse CL (1994) Cyclic AMP-induced mucin exocytosis is independent of Cl− movement inhuman colonic epithelial cells (HT29-Cl.16E). Biochem J304:675–678

Kuver R, Ramesh N, Lau S, Savard C, Lee SP, Osborne WRA(1994) Constitutive mucin secretion linked to CFTR expres-sion. Biochem Biophys Res Comm 203:1457–1462

McCray PB, Reenstra WW, Louie E, Johnson J, Bettencourt JD,Bastacky J (1992) Expression of CFTR and presence ofcAMP-mediated fluid secretion in human fetal lung. Am JPhysiol (Lung Cell Mol Physiol 6) 262:L472–L481

Moughan PJ, Birtles MJ, Cranwell PD, Smith WC, Pedraza(1992) The piglet as a model animal for studying aspects of

digestion and absorption in milk-fed human infants. In: Si-mopoulos AP (ed) Nutritional triggers for health and in dis-ease. World Rev Nutr Diet, vol 67:40–113. Basel, Karger

O’Loughlin EV, Hunt DM, Gaskin KJ, Steil D, Bruzuszcak IM,Martin HCO, Bambach C, Smith R (1991) Abnormal epithe-lial transport in cystic fibrosis jejunum. Am J Physiol 260(Gastrointest Liver Physiol 23):G758–G763

Snouwaert JN, Brigman KK, Latour AM, Malouf NN, BoucherRC, Smithies O, Koller BH (1992) An animal model for cysticfibrosis made by gene targeting. Science 257:1083–1088

Strong TV, Boehm K, Collins FS (1994) Localization of cystic fi-brosis transmembrane conductance regulator mRNA in the hu-man gastrointestinal tract by in situ hybridization. J Clin In-vest 93:347–354

Zhou L, Dey CR, Wert SE, Du Vall MD, Frizzell RA, Whitsett JA(1994) Correction of lethal intestinal defect in a mouse modelof cystic fibrosis by human CFTR. Science 266:1705–1708

213