Expression and immunolocalization of aquaporin-7 in rat gastrointestinal tract

Biol. Cell (2005) 97, 605–613 (Printed in Great Britain) Research article

Expression and immunolocalizationof aquaporin-7 in rat gastrointestinaltractUmberto Laforenza*1, Giulia Gastaldi*, Monica Grazioli*, Emanuela Cova*, Simona Tritto*, Alide Faelli†,Giuseppe Calamita‡ and Ulderico Ventura**Department of Experimental Medicine, Section of Human Physiology, University of Pavia, Via Forlanini, 6, I-27100, Pavia, Italy, †Department

of Biomolecular Sciences and Biotechnologies, University of Milan, Milan, Italy, and ‡Department of General and Environmental Physiology,

University of Bari, Bari, Italy

Background information. In the gastrointestinal tract of mammals, water can either be secreted with digestive juicesor absorbed by the small and large intestine. Transcellular water movement can be mediated by the transmembraneprotein family of AQPs (aquaporins), as has also been recently identified in the gastrointestinal tract. However, thelocalization, expression and functioning of AQPs in the gastrointestinal tract have not been completely character-ized. For the present study, we investigated: (1) the expression of AQP7 in some portions of rat gastrointestinaltract by semiquantitative reverse transcriptase–PCR and by immunoblotting and (2) the cellular and subcellularlocalization of AQP7 by immunohistochemistry.

Results. AQP7 mRNA and proteins were highly expressed in the small intestine, weakly in the caecum, colonand rectum and were absent in the stomach. Immunoblotting analysis using rat gastrointestinal tract membranefractions showed two major bands corresponding to a molecular mass of approx. 34 and 40 kDa for the AQP7protein. No bands were observed when the anti-AQP7 antibody was preadsorbed with the immunizing peptide.Immunohistochemistry revealed strong AQP7 labelling in the surface epithelial cells of duodenum, jejunum, ileum,caecum, colon and rectum, whereas weak or no labelling was observed in the crypt cells. The labelling was manifestparticularly in the apical membrane but intracellular staining was also observed.

Conclusions. The results indicate that AQP7 is present in the small and large intestine. The higher expression ofAQP7 protein at the apical pole of the superficial epithelial cells suggests its involvement in rapid fluid movementthrough the villus epithelium.

IntroductionThe gastrointestinal tract is a well-known site forlarge fluid movement and its magnitude is secondonly to that of the kidney. Water transport is bi-directional: in humans approx. 9 l/day of water de-rived from diet and digestive juices are absorbed andapprox. 1–2 l/day are secreted with enteric juice(Powell, 1987; Ma and Verkman, 1999; Masyuket al., 2002).

1To whom correspondence should be addressed (email [email protected]).Key words: apical membrane, AQP7 protein, immunohistochemistry,RT–PCR, water channel.Abbreviations used: AQP, aquaporin; RT, reverse transcriptase.

Secretory and absorptive functions of the gas-trointestinal tract are vitally important for food di-gestion and to maintain the water and electrolytebalance respectively. In particular, osmoregulation isachieved by intestinal absorption; most of the water(∼84%) present in the entire intestine is absorbed inan iso-osmotic fashion by the small intestine and theremainder (∼16%) by the large intestine (Powell,1987). Nevertheless, the distal parts of intestine aremore efficient at absorbing water (and electrolytes)even against the osmotic gradient. Fluid recircula-tion takes place along the intestine, especially after ameal, because water is secreted in the upper part ofthe gastrointestinal tract allowing the rapid osmotic

www.biolcell.org | Volume 97 (8) | Pages 605–613 605

U. Laforenza and others

equilibration of intestinal contents and is continu-ously absorbed with hydrolysed nutrients (Powell,1987; Chang and Rao, 1994). According to theirtransepithelial electrical resistance characteristics, theepithelia from different portions of the gastrointes-tinal tract may be classified into three categories: leaky(small intestine), moderately tight (colon and gastricantrum) and tight (gastric fundus) (Powell, 1981).Water movement may follow different routes as vari-ous epithelia along the gastrointestinal tract have adifferent electrical resistance (Powell, 1981), waterpermeability of their apical and basolateral cell mem-branes (Powell, 1981; van Heeswijk and van Os,1986; Priver et al., 1993; Zeuthen, 1995, 1996;Carmosino et al., 2001) and osmotic gradients intheir lumen.

Water movement across the epithelium of the gas-trointestinal tract can hypothetically occur throughboth the paracellular and transcellular route. The lat-ter involves crossing the apical and basolateral mem-branes of the epithelial cells (Chang and Rao, 1994).Transcellular water transport may occur by three dif-ferent mechanisms: (1) passive diffusion across thelipid bilayer; (2) co-transport with ions and nutri-ents (Loo et al., 1996; Meinild et al., 1998; Wrightand Loo, 2000); and (3) diffusion across the waterchannels, AQPs (aquaporins) (Ma and Verkman,1999; Masyuk et al., 2002; Nielsen et al., 2002). Therelative contribution of these three mechanisms totransepithelial water movement along the gastro-intestinal tract is far from being elucidated. Specialattention should be paid to the role of the AQPs,involved, as in other organs, in the rapid and sub-stantial bidirectional movement of fluid given thateight different AQPs have recently been identified inthe gastrointestinal tract of mammals (Frigeri et al.,1995; Koyama et al., 1999; Ramirez-Lorcaet al., 1999; Kierbel et al., 2000; Calamita et al.,2001a; Elkjaer et al., 2001; Hatakeyama et al., 2001;Tani et al., 2001; Parvin et al., 2002; Okada et al.,2003; Mobasheri et al., 2004). However, their topo-graphical distribution, cellular localization, expres-sion and functioning in the gastrointestinal tract havenot yet been completely characterized.

AQP7, first cloned in the rat testis (Ishibashi et al.,1997), is a 269 amino acid protein belonging to theAQP family and to the aquaglyceroporin subgroup.It is permeable not only to water but also to smallmolecules such as glycerol, urea, arsenic compounds

and is insensitive to mercury chloride (Ishibashiet al., 1997; Borgnia et al., 1999; Liu et al., 2002).

Northern-blot analysis revealed that AQP7mRNA is also expressed in the rat kidney, heart,skeletal muscle, adipose tissues and small intestine(Ishibashi et al., 1997, 1998; Nejsum et al., 2000;Tsujikawa et al., 2003; Wakayama et al., 2004).Although the expression and localization of AQP7protein have been studied in the testis, kidney andskeletal muscle, no information is available regardingAQP7 at a protein level in the small intestine.

The purpose of the present research was to studythe expression and localization of AQP7 in the ratgastrointestinal tract, from stomach to rectum, underthe hypothesis that it could be involved in physiolo-gical mechanisms of fluid absorption and secretion.

In the present study, we examined the expres-sion of AQP7 mRNA and protein in different por-tions of the rat gastrointestinal tract by RT (reversetranscriptase)–PCR and immunoblotting, and thecellular and subcellular localization of AQP7 proteinby immunohistochemistry. The results reported hereprovide evidence that an apical AQP7 is expressed inthe superficial cells of both the small and large intes-tine and suggest a role for this AQP in the movementof water and small molecules through the intestinalepithelium.

A preliminary partial account of these resultswas presented at the 54th meeting of the ItalianPhysiological Society (Laforenza, U., Gastaldi, G.,Cova, E., Tritto, S., Grazioli, M., Ventura, U., 2003.Transepithelial water transport in the gastrointes-tinal system: role of aquaporins. Abstract presented atthe 54th Meeting of the Italian Physiological Society,pp. 83, no. 072).

ResultsRT–PCR analysis of AQP7 mRNA expressionin rat gastrointestinal tractSemiquantitative RT–PCR was performed to evalu-ate the expression and distribution of AQP7 mRNAin rat gastrointestinal tract. The expression of AQP7was normalized using β-actin as an internal stand-ard. The results of agarose gel electrophoresis ofrepresentative PCR reaction products are shownin Figure 1. Single bands of the expected size ofcDNA fragments were amplified (323 and 509 bp forAQP7 and β-actin respectively). Negative controls of

606 C© Portland Press 2005 | www.biolcell.org

Aquaporin-7 in the rat gastrointestinal tract Research article

Figure 1 Distribution of AQP7 mRNA expression indifferent portions of the rat gastrointestinal tractUpper panel: semiquantitative RT–PCR of total RNA (1 µg)

from indicated rat samples was performed by using specific

primers for AQP7 (see the Materials and methods section).

Lower panel: RT–PCR of β-actin used to normalize the ex-

pression of AQP7. The 323 and 509 bp bands correspond

to the AQP7 and β-actin-specific PCR products respectively.

Similar results were obtained for at least four different rat RNA

extracts.

RT–PCR experiments were always performed byomitting the RT (results not shown). The sequenceof the PCR products was completely identical withthe published sequence for AQP7.

The AQP7 mRNA was expressed in all the partsof the rat gastrointestinal tract except in the stomach(Figure 1).

Densitometric analysis of the bands showed thatthe AQP7 transcript was significantly higher in thesmall intestine (duodenum, jejunum and ileum)when compared with the large intestine (caecum, pro-ximal and distal colon and rectum). The values(means +− S.E.M.), expressed as the percentage of theAQP/β-actin densitometric ratio for four differentexperiments, were: 133∗ +− 29, duodenum; 161∗ +−39, jejunum; 130∗ +− 35, ileum; 57 +− 22, caecum;67 +− 21, proximal colon; 71 +− 33, distal colon; and71 +− 15, rectum. ∗P < 0.05 versus caecum, proximalcolon, distal colon and rectum (repeated measuringby ANOVA followed by Newman–Keuls’s Q test).

Immunoblotting analysis of the AQP7 proteinin rat gastrointestinal tractTo determine the expression and distribution of theAQP7 protein along the rat gastrointestinal tract,total membrane fractions were analysed by immuno-blotting with affinity-purified antibodies againstrat AQP7. As observed for the RNA transcript, theAQP7 protein was detected in all the gastrointestinaltract zones except for the stomach (Figure 2A). The

Figure 2 Distribution of AQP7 protein expression in themembrane fraction of different portions of the ratgastrointestinal tractLanes were loaded with 70 µg of proteins. The immunoblots

were incubated with affinity-purified antibodies diluted 1:2000

and processed as described in the Materials and methods

section. (A) Anti-AQP7 immunoblot of membranes from dif-

ferent portions of the rat gastrointestinal tract. A major band

of 34 kDa and a minor band of 40 kDa were observed in

all gastrointestinal portions investigated, except in the stom-

ach. (B) Immunoblot of membrane fractions from rat jejunum

probed with anti-AQP7 (right) or with anti-AQP7 preadsorbed

with the immunizing peptide (left). No band was observed

when a preadsorbed anti-AQP7 antibody was used.

AQP7 protein distribution along the gastrointestinaltract showed higher expression in the small intestineportions followed by the colon, rectum and caecum.

Immunoblots of AQP7 protein showed a majorband of approx. 34 kDa and a faint band of approx.40 kDa (Figure 2A). The molecular masses were lar-ger than those expected by calculation of the molecu-lar mass (29 kDa) and probably represent productsof post-translational processing of a different type ofglycosylation to N-glycosylation, consistent with theabsence of an N-linked glycosylation consensus site(Ishibashi et al., 1997, 2000). Occasionally, a bandof lower molecular mass (∼29 kDa) was observed re-presenting the AQP7 non-glycosylated form (resultsnot shown).

When the antibody was preadsorbed with largeamounts of the immunizing peptide, the proteinbands completely disappeared, indicating the spe-cificity of the reaction (Figure 2B).

Immunohistochemical localization of the AQP7protein in rat gastrointestinal tractThe distribution and cellular and subcellular local-ization of the AQP7 protein in the gastrointestinal

www.biolcell.org | Volume 97 (8) | Pages 605–613 607

U. Laforenza and others

Figure 3 Immunohistochemical localization of the AQP7 protein in the stomach and duodenum of ratNo AQP7 staining was observed in the stomach (a). In the duodenum (c, e) the labelling was present in the upper part of the

villus (c) and decreased or absent in the crypt (e). Intense AQP7 immunoreactivity was observed in the apical part of the epithelial

cells (c; inset, arrows), but intracellular staining was also present. Goblet cells were not stained. Controls in which the primary

antibody was omitted show an absence of labelling (b, d and f). S, serosa; L, lumen. Magnification: ×200 (a, b), ×400 (c, d),

×300 (e, f), ×1000 (inset).

tract was investigated by immunohistochemistry, us-ing affinity-purified anti-AQP7 antibodies. In thestomach of a rat, no labelling was observed eitheron the surface epithelium or in the principal andoxyntic cells of the gastric glands (Figure 3a). Theseresults confirmed those obtained by RT–PCR and im-munoblotting. Immunolabelled controls were nega-tive (Figure 3b).

The AQP7 protein was detected in the mucosa ofboth the small and large intestine (Figures 3c and 4c).In the small intestine, AQP7 immunolabelling wasobserved on the superficial epithelial cells of the duo-denum (Figure 3c, inset), jejunum and ileum (resultsnot shown). The labelling showed a strong positivereaction in the cells of the upper part of the villus(Figure 3c, inset) but it was decreased or absent in

the crypt (Figure 3e). At high magnification, AQP7expression was more intense at the brush border mem-branes of the epithelial cells, even if an intracellularstaining was also present (Figure 3; inset). No stain-ing was observed at the basolateral membrane of theepithelial cells. Goblet cells did not express AQP7as expected from the constant interruption of theimmunostaining. Anti-AQP7 immunostaining wasabsent from controls where the primary antibodywas omitted (Figures 3b, 3d and 3f). Similar re-sults were also obtained when a preadsorbed affinity-purified anti-AQP7 antibody was used (results notshown).

In the large intestine, AQP7 immunostaining al-most exclusively labelled the monolayer of epithelialcells lining the tip of the crypt (Figures 4a, 4c, 4e and

608 C© Portland Press 2005 | www.biolcell.org

Aquaporin-7 in the rat gastrointestinal tract Research article

Figure 4 Immunohistochemical localization of the AQP7 protein in the rat caecum, proximal and distal colon andrectumFor the caecum (a), the proximal (c) and the distal (e) part of the colon, and the rectum (g), the labelling was present in the

surface epithelial cells of the crypt and almost absent in the epithelium at the base of the crypt. Intense AQP7 immunoreactivity

was observed in the apical part of the epithelial cells, but intracellular staining was also present. Arrows in the inset show

the staining of apical membranes in the rat distal colon at high magnification. Goblet cells of caecum, colon and rectum were

not stained. Controls in which the primary antibody was omitted show an absence of labelling (b, d, f and h). Magnifications:

×400 (a–d, f, h), ×300 (e, g), ×1000 (inset).

4g). No apparent differences were observed amongthe parts of the large intestine (caecum, proximalcolon, distal colon and rectum). The intensity ofAQP7 immunostaining decreased from the tip to-wards the base of the crypt. No labelling was de-tectable in the intercalated Goblet cells (Figures 4a,4c, 4e and 4g). At high magnification, the epithelialcells of the distal colon showed a strong AQP7 im-munostaining of the apical membranes (Figure 4, in-

set; arrows). Intracellular staining was also observed,whereas basolateral membranes were negative (Fig-ure 4, inset). Immunolabelled controls were nega-tive (Figures 4b, 4d, 4f and 4h).

DiscussionThe aim of the present study was to establish theexpression, topographical distribution, as well as

www.biolcell.org | Volume 97 (8) | Pages 605–613 609

U. Laforenza and others

the cellular and subcellular localization of the AQP7in the rat gastrointestinal tract. Until now, the pres-ence of AQP7 has been found in the small intestineonly at mRNA level (Ishibashi et al., 1997, 1998;Tsujikawa et al., 2003). The major findings of thepresent paper are that AQP7 is expressed particu-larly in the small intestine and in the large intes-tine of rats (Figures 1 and 2), and it is localizedat the apical membrane domain of the superficialcells of the gastrointestinal epithelium (Figures 3and 4).

The gastrointestinal epithelium has important bar-rier, absorptive and secretory functions for water andsolute transport. The paracellular and transcellularpathways of fluid transport have been extensivelystudied but their relative contributions have not beenfully clarified.

In the small intestinal epithelium, the paracellularroute was considered as the main route for water trans-port due to its low electrical resistance and high waterpermeability allowing rapid osmotic equilibration ofintestinal contents, and due to the low osmotic per-meability of brush border and basolateral membranes(Powell, 1981, 1987; Worman and Field, 1985; vanHeeswijk and van Os, 1986; Ma and Verkman, 1999;Masyuk et al., 2002).

However, the involvement of a transcellular path-way for water movement across the small intestinalepithelium cannot be ruled out since at least eightAQPs have been identified in the digestive system ofmammals, i.e. AQP1, AQP3, AQP4, AQP5, AQP7,AQP8, AQP9 and AQP10 (Frigeri et al., 1995;Koyama et al., 1999; Ramirez-Lorca et al., 1999;Kierbel et al., 2000; Calamita et al., 2001a;Elkjaer et al., 2001; Hatakeyama et al., 2001; Taniet al., 2001; Parvin et al., 2002; Okada et al., 2003;Mobasheri et al., 2004). Moreover, Zeuthen’s andWright’s groups (Loo et al., 1996; Zeuthen et al.,1997; Meinild et al., 1998; Wright and Loo, 2000)recently showed a sodium/glucose/water co-transportwith a stoichiometry ratio of 2:1:210 by using oocytesexpressing the human Na–glucose co-transporter,hSGLT1. This mechanism also occurs in the pres-ence of an adverse osmotic gradient and could beresponsible for the absorption of approx. 4 l/day ofwater in the small intestine.

The involvement of AQP proteins in the water fluxthrough the gastric and colonic epithelia cannot beexcluded since they are tight epithelia characterized

by high electrical resistance and low water permeab-ility (Powell, 1981).

Our results obtained from RT–PCR experimentsconfirmed the presence of AQP7 transcript in thesmall intestine of rats, as previously demonstratedby Northern-blot analysis (Ishibashi et al., 1998;Tsujikawa et al., 2003) (Figure 1). Densitometricanalysis of the bands showed no differences betweenthe duodenum, jejunum and ileum (see the Resultssection). AQP7 mRNA is also expressed in the cae-cum, proximal and distal colon and rectum, the in-tensities of the bands here being significantly lowerthan those of the small intestine. This result contra-dicts those presented by Ishibashi et al. (1998) whodid not find AQP7 mRNA in the rat colon. Never-theless, these differences could also be attributed tothe different techniques employed.

No signal was detected in the rat stomach. Hence,based on available results, we conclude that only twoAQPs (AQP3 and AQP4) are present in the stom-ach, both at the basolateral side of the cells (Ma andVerkman, 1999; Masyuk et al., 2002). Nevertheless,the existence of other AQPs on the apical membranesof gastric cells involved in gastric juice secretion can-not be excluded.

Similar results regarding the expression and distri-bution of AQP7 along the rat gastrointestinal tractwere obtained by immunoblotting of total membranefractions (Figure 2A).

As observed for the RNA transcript, the AQP7 pro-tein shows a higher expression in the small intestineportions followed by the colon, rectum and caecum.The pattern of a typical immunoblot exhibits a ma-jor band of an apparent molecular mass of 34 kDa(Figure 2A). The size of the band is larger than thatexpected (29 kDa), as already observed in rat kidneymembranes (Ishibashi et al., 2000). It most probablyrepresents a product of post-translational processingof a different type of glycosylation to N-glycosylation,which is consistent with the absence of an N-linkedglycosylation consensus site (Ishibashi et al., 1997).Frequently, a faint band of approx. 40 kDa is also ob-served but it completely disappears when the anti-body is preadsorbed with the immunizing peptide(Figures 2A and 2B). Further experiments need to beperformed to understand the role of this isoform.

The results of immunohistochemical studies clearlydemonstrate where the AQP7 protein is localized inthe rat gastrointestinal tract. AQP7 is found to be

610 C© Portland Press 2005 | www.biolcell.org

Aquaporin-7 in the rat gastrointestinal tract Research article

highly expressed on the superficial epithelial cellsof the small intestinal villus tip, and on the mono-layer of epithelial cells lining the tip of the cryptof the large intestine (Figures 3c and 4c). Moreover,AQP7 expression appears more intense at the brushborder membranes, even if intracellular staining isalso present (Figures 3 and 4, insets). This agreeswith the predominant but not exclusive AQP7 local-ization in the plasma membranes of spermatids andspermatozoa (Suzuki-Toyota et al., 1999; Calamitaet al., 2001b). AQP7 cellular localization at the brushborder membranes was also observed in the proximalstraight tubules (S3 segment) of the rat kidney byIshibashi et al. (2000). Hence, AQP7 may repres-ent the route for water transport through the apicalmembrane domain of rat intestinal epithelial cells.

Recently, AQP10, a new member of the aquagly-ceroporin family, was identified in the humansmall and large intestine (Hatakeyama et al., 2001;Mobasheri et al., 2004). Moreover, the localization onthe apical membranes of the absorptive cells suggeststhat AQP10 has an important role for water entryinto the small intestinal epithelial cells (Mobasheriet al., 2004).

From the results reported in the literature and thosepresented here, a model can be proposed for transcel-lular AQP-mediated water movement in the absorp-tive intestinal cells. Entry at the luminal (apical) sidetakes place through AQP7 in the rat and AQP10 inhumans, whereas the exiting at the basolateral sideis mediated in the small intestine by AQP3 in thevillus and AQP4 in the crypt cells, and in the largeintestine, by both AQP3 and AQP4 in the surfaceepithelial cells of the crypts (Ma and Verkman, 1999;Masyuk et al., 2002). In the rat rectum, immunohis-tochemical studies showed that AQP3 but not AQP4is expressed in the basolateral membrane of surfaceepithelial cells (Kierbel et al., 2000).

At present, the role of AQP8 in the gastrointes-tinal epithelium remains obscure due to its predom-inant localization in the intracellular compartments(Calamita et al., 2001a; Elkjaer et al., 2001). How-ever, AQP8 could be hypothetically redistributedto the apical membrane of the epithelial cells un-der certain stimuli (gastrointestinal hormones), asalready observed for rat hepatocytes (Garcia et al.,2001; Gradilone et al., 2003), and may play a rolein modulating water transport in the small and largeintestine.

It is intriguing that, except for AQP4, both AQPsfound in the apical membranes (AQP7 and AQP10)and the AQP3 found in the basolateral membranesof the gastrointestinal epithelial cells are multifunc-tional AQPs. Their low selectivity transport prop-erty may indicate an important role for these AQPs,not only in water movement but also in the absorp-tion and/or secretion of small molecules through thegastrointestinal epithelium.

To conclude, our results appear to support a rolefor AQPs in transcellular water movement in thegastrointestinal tract epithelium. The subcellular loc-alization of AQP7 suggests a pivotal role for this AQPin water transport at the luminal side of the rat in-testinal epithelium.

Materials and methodsAnimalsAdult Wistar albino rats (350–400 g of body weight) werehoused at the animal facility of the Department of ExperimentalMedicine, University of Pavia, cared for and killed according tothe current European legal Animal Practice requirements.

RNA isolation and RT–PCRTotal RNA was extracted from the gastrointestinal tissues us-ing RNA-BeeTM Isolation solvent (Tel-Test ‘B’, Friendwood,TX, U.S.A.). Single cDNA was synthesized from 1 µg of RNAusing random hexamers and M-MLV Reverse Transcriptase (In-vitrogen, Carlsbad, CA, U.S.A.). PCR (30 s at 96◦C, 30 s at60◦C and 30 s at 72◦C for 35 cycles) was performed as pre-viously described by Bione et al. (1996) on 2.5 µl of cDNAusing specific primers for rat AQP7 (Capurro et al., 2001). Re-verse transcription was always performed in the presence or ab-sence of an RT enzyme. The RT–PCR for AQP7 was normalizedusing β-actin as an internal standard (Calamita et al., 2001a).First, the sequences of the AQP bands were checked by usingthe Big dye terminator cycle sequencing kit (Applied Biosystem,Milan, Italy). PCR products were separated with agarose gel elec-trophoresis, stained with ethidium bromide and acquired withthe Image Master VDS (Amersham Biosciences Europe GMBH –Filiale Italiana, Milan, Italy). Densitometric analysis of the bandswas performed by the Total Lab V 1.11 computer program(Amersham Biosciences) and the results were expressed as apercentage of the AQP/β-actin densitometric ratio. The mol-ecular mass of the PCR products was compared with the DNAmolecular weight marker VIII (Roche Molecular Biochemicals,Monza, Italy).

Membrane preparation and immunoblottingThe mucosal scraping from different parts of the gastrointestinaltract were homogenized with a Teflon glass Potter–Elvehjem-type homogenizer (Kontes, Vineland, NJ, U.S.A.) in a solutioncontaining 100 mM NaCl, 1 mM EDTA, 10 mM Tris/HCl(pH 6.8) and 0.1 mg/ml PMSF. After centrifugation at 100000 gfor 60 min at 4◦C, the pellets were suspended in the homo-genization buffer and treated as described previously (Laemmli

www.biolcell.org | Volume 97 (8) | Pages 605–613 611

U. Laforenza and others

et al., 1970). Solubilized proteins (70 µg) were subjectedto SDS/PAGE (12.5% polyacrylamide) and transferred on tothe Hybond ECL® nitrocellulose membrane (Amersham Bios-ciences) by electroelution. After 3 h blocking with Tris-bufferedsaline containing 5% (w/v) non-fat dried milk and 0.1% Tween20 (blocking solution), the membranes were incubated overnightwith affinity-purified antibody to AQP7, raised against the syn-thetic peptide (19 amino acids) of the N-terminal cytoplas-mic domain of rat AQP7 (Alpha Diagnostic International, SanAntonio, TX, U.S.A.), and then diluted 1:2000 in the block-ing solution. The membranes were washed and incubated for1 h with peroxidase-conjugated goat anti-rabbit IgG (1:3000in blocking solution) (Amersham Biosciences). The bandswere detected with ECL® Western-blotting detection system(Amersham Biosciences) and analysed as indicated above forPCR gels. The control experiment was performed using an anti-body preadsorbed with a 20-fold molar excess of the immuniz-ing peptide (Alpha Diagnostic International). The ChemiBlotTM

Molecular Weight Markers were used to accurately estimatethe molecular mass and were used as a positive control for theimmunoblot analysis (Chemicon International, Temecula, CA,U.S.A.).

ImmunohistochemistryThe rats were anesthetized with 2-bromo-2-chloro-1,1,1-tri-fluoroethane and intracardially perfused with acetate-buffered4% (w/v) formalin. Gastrointestinal tissues were removed, post-fixed for 1 h and processed into paraffin. Serial paraffin sections(5 µm) were brought to water and treated with 3% (w/v) H2O2for 10 min at room temperature (22◦C) to block the endogenousperoxidases. After washing for 5 min with PBS, sections wereblocked with 3% (w/v) BSA in PBS for 30 min at room tem-perature. Sections were incubated for 2 h at room temperaturewith affinity-purified anti-AQP7 antibodies diluted 1:2000 inPBS containing 1% BSA. After three 5 min washes with PBScontaining 1% BSA, the sections were first incubated for 15 minat room temperature with biotinylated anti-rabbit IgG and thenwashed three times with PBS containing 1% BSA for 15 min atroom temperature with HRP-conjugated streptavidin (Univer-sal DAKO LSAB® + kit, peroxidase, K0679; DakoCytomation,Milan, Italy). The reaction was visualized by incubation witha DakoCytomation 3,3′-diaminobenzidine chromogen solution.The sections were counterstained with haematoxylin and moun-ted on DPX (Merck Eurolab, Milan, Italy). Control experimentswere performed simultaneously using antibodies preadsorbedwith an immunizing peptide or omitting the primary antibody.

The immunostained slides were examined by light micro-scopy using an Olympus BX41 and the digital images capturedwith an Olympus Camedia C-5050 zoom digital camera (Olym-pus Italia, Milan, Italy).

Protein contentThe protein content was determined using a method describedby Lowry et al. (1951) using BSA as a standard.

StatisticsThe significance of the differences of the means for the RT–PCR densitometric analysis was evaluated by using the ANOVAmethod followed by Newman–Keuls’s Q test. The statistical

tests were performed with a computerized program (Glantz,1988).

AcknowledgementsThis work was supported by the University of PaviaFAR 2004.

ReferencesBione, S., D’Adamo, P., Maestrini, E., Gedeon, A.K., Bolhuis, P.A. and

Toniolo, D. (1996) A novel X-linked gene, G4.5. is responsible forBarth syndrome. Nat. Genet. 12, 385–389

Borgnia, M., Nielsen, S., Engel, A. and Agre, P. (1999) Cellularand molecular biology of the aquaporin water channels.Annu. Rev. Biochem. 68, 425–458

Calamita, G., Mazzone, A., Bizzoca, A., Cavalier, A., Cassano, G.,Thomas, D. and Svelto, M. (2001a) Expression andimmunolocalization of the aquaporin-8 water channel in ratgastrointestinal tract. Eur. J. Cell Biol. 80, 711–719

Calamita, G., Mazzone, A., Cho, Y.S., Valenti, G. and Svelto, M.(2001b) Expression and localization of the aquaporin-8 waterchannel in rat testis. Biol. Reprod. 64, 1660–1666

Capurro, C., Rivarola, V., Kierbel, A., Escoubet, B., Barman, N.,Blot-Chabaud, M. and Parisi, M. (2001) Vasopressin regulateswater flow in a rat cortical collecting duct cell line not containingknown aquaporins. J. Membr. Biol. 179, 63–70

Carmosino, M., Procino, G., Nicchia, G.P., Mannucci, R., Verbavatz,J.M., Gobin, R., Svelto, M. and Valenti, G. (2001) Histaminetreatment induces rearrangements of orthogonal arrays of particles(OAPs) in human AQP4-expressing gastric cells. J. Cell Biol. 154,1235–1243

Chang, E. and Rao, M.C. (1994) Intestinal water and electrolytetransport. Mechanisms of physiological and adaptive responses.In Physiology of the Gastrointestinal Tract (Johnson, L.R., ed.),pp. 2040–2043, Raven Press, New York

Elkjaer, M.L., Nejsum, L.N., Gresz, V., Kwon, T.H., Jensen, U.B.,Frokiaer, J. and Nielsen, S. (2001) Immunolocalization ofaquaporin-8 in rat kidney, gastrointestinal tract, testis, and airways.Am. J. Physiol. Renal Physiol. 281, F1047–F1057

Frigeri, A., Gropper, M.A., Brown, D. and Verkman, A.S. (1995)Localization of MIWC and GLIP water channel homologs inneuromuscular, epithelial and glandular tissues. J. Cell Sci. 108,2993–3002

Garcia, F., Kierbel, A., Larocca, M.C., Graditone, S.A., Sprinter, P.,LaRusso, N.F. and Marinelli, R.A. (2001) The water channelaquaporin-8 is mainly intracellular in rat hepatocytes, and itsplasma membrane insertion is stimulated by cyclic AMP.J. Biol. Chem. 276, 12147–12152

Glantz, S.A. (1988) Statistica per Discipline Biomediche: ProgrammaApplicativo, 2nd edn, McGraw-Hill Libri Italia, Milan

Gradilone, S.A., Garcia, F., Huebert, R.C., Tietz, P.S., Larocca, M.C.,Kierbel, A., Carreras, F.I., Larusso, N.F. and Marinelli, R.A. (2003)Glucagon induces the plasma membrane insertion of functionalaquaporin-8 water channels in isolated rat hepatocytes.Hepatology 37, 1435–1441

Hatakeyama, S., Yoshida, Y., Tani, T., Koyama, Y., Nihei, K.,Ohshiro, K., Kamiie, J.I., Yaoita, E., Suda, T., Hatakeyama, K. et al.(2001) Cloning of a new aquaporin (AQP10) abundantly expressedin duodenum and jejunum. Biochem. Biophys. Res. Commun. 287,814–819

Ishibashi, K., Kuwahara, M., Gu, Y., Kageyama, Y., Tohsaka, A.,Suzuki, F., Marmo, F. and Sasaki, S. (1997) Cloning and functionalexpression of a new water channel abundantly expressed in thetestis permeable to water, glycerol, and urea. J. Biol. Chem. 272,20782–20786

612 C© Portland Press 2005 | www.biolcell.org

Aquaporin-7 in the rat gastrointestinal tract Research article

Ishibashi, K., Yamauchi, K., Kageyama, Y., Saito-Ohara, F.,Ikeuchi, T., Marumo, F. and Sasaki, S. (1998) Molecularcharacterization of human aquaporin-7 gene and its chromosomalmapping. Biochim. Biophys. Acta 1399, 62–66

Ishibashi, K., Imai, M. and Sasaki, S. (2000) Cellular localization ofaquaporin 7 in the rat kidney. Exp. Nephrol. 8, 252–257

Kierbel, A., Capurro, C., Pisam, M., Gobin, R., Chrinstensen, B.M.,Nielsen, S. and Parisi, M. (2000) Effects of medium hypertonicityon water permeability in the mammalian rectum: ultrastructural andmolecular correlates. Pflugers Arch. 440, 609–618

Koyama, Y., Yamamoto, T., Tani, T., Nihei, K., Kondo, D.,Funaki, H., Yaoita, E., Kawasaki, K., Sato, N., Hatakeyama, K.et al. (1999) Expression and localization of aquaporins in ratgastrointestinal tract. Am. J. Physiol. Cell Physiol. 276,C621–C627

Laemmli, U.K. (1970) Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London) 227,680–685

Liu, Z., Shen, J., Carbrey, J.M., Mukhopadhyay, R., Agre, P.and Rosen, B.P. (2002) Arsenite transport by mammalianaquaglyceroporins AQP7 and AQP9. Proc. Natl.Acad. Sci. U.S.A. 99, 6053–6058

Loo, D.D.F., Zeuthen, T., Chandy, G. and Wright, E.M. (1996)Cotransport of water by the Na+/glucose cotransporter. Proc. Natl.Acad. Sci. U.S.A. 93, 13367–13370

Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall, R.J. (1951)Protein measurement with the Folin phenol reagent. J. Biol. Chem.193, 443–449

Ma, T. and Verkman, A.S. (1999) Aquaporin water channels ingastrointestinal physiology. J. Physiol. (Cambridge, U.K.) 517,317–326

Masyuk, A.I., Marinelli, R.A. and LaRusso, N.F. (2002) Watertransport by epithelia of the digestive tract. Gastroenterology 122,545–562

Meinild, A.K., Klaerke, D.A., Loo, D.D.F., Wright, E.M. and Zeuthen, T.(1998) The human Na+-glucose cotransporter is a molecular waterpump. J. Physiol. (Cambridge, U.K.) 508, 15–21

Mobasheri, A., Shakibaei, M. and Marples, D. (2004)Immunohistochemical localization of aquaporin 10 in the apicalmembranes of the human ileum: a potential pathway for luminalwater and small solute absorption. Histochem. Cell Biol. 121,463–471

Nejsum, L.N., Elkjaer, M., Hager, H., Frokiaer, J., Kwon, T.H. andNielsen, S. (2000) Localization of aquaporin-7 in rat and mousekidney using RT–PCR, immunoblotting, and immunocytochemistry.Biochem. Biophys. Res. Commun. 277, 164–170

Nielsen, S., Frokiaer, J., Marples, D., Kwon, T.H., Agre, P. andKnepper, M.A. (2002) Aquaporins in the kidney: from molecules tomedicine. Physiol. Rev. 82, 205–244

Okada, S., Misaka, T., Matsumoto, I., Watanabe, H. and Abe, K.(2003) Aquaporin-9 is expressed in a mucus-secreting goblet cellsubset in the small intestine. FEBS Lett. 540, 157–162

Parvin, M.N., Tsumura, K., Akamatsu, T., Kanamori, N. and Hosoi, K.(2002) Expression and localization of AQP5 in the stomach andduodenum of the rat. Biochim. Biophys. Acta 1542, 116–124

Powell, D.W. (1981) Barrier function of epithelia. Am. J. Physiol.Gastrointest. Liver Physiol. 241, G275–G288

Powell, D.W. (1987) Intestinal water and electrolyte transport.In Physiology of the Gastrointestinal Tract (Johnson, L.R., ed.),pp. 1267–1305, Raven Press, New York

Priver, N.A., Rabon, E.C. and Zeidel, M.L. (1993) Apical membraneof the gastric parietal cell: water, proton, and nonelectrolytepermeabilities. Biochemistry 32, 2459–2468

Ramirez-Lorca, R., Vizuete, M.L., Venero, J.L., Revuelta, M., Cano, J.,Ilundain, A.A. and Echevarria, M. (1999) Localization ofaquaporin-3 mRNA and protein along the gastrointestinal tractof Wistar rats. Pflugers Arch. 438, 94–100

Suzuki-Toyota, F., Ishibashi, K. and Yuasa, S. (1999)Immunohistochemical localization of a water channel, aquaporin 7(AQP7), in the rat testis. Cell Tissue Res. 295, 279–285

Tani, T., Koyama, Y., Nihei, K., Hatakeyama, S., Ohshiro, K.,Yoshida, Y., Yaoita, E., Sakai, Y., Hatakeyama, K. and Yamamoto, T.(2001) Immunolocalization of aquaporin-8 in rat digestive organsand testis. Arch. Histol. Cytol. 64, 159–168

Tsujikawa, T., Itoh, A., Fukunaga, T., Satoh, J., Yasuoka, T. andFujiyama, Y. (2003) Alteration of aquaporin mRNA expression aftersmall bowel resection in the rat residual ileum and colon.J. Gastroenterol. Hepatol. 18, 803–808

van Heeswijk, M.P.E. and van Os, C.H. (1986) Osmotic waterpermeabilities of brush border and basolateral membrane vesiclesfrom rat renal cortex and small intestine. J. Membr. Biol. 92,183–193

Wakayama, Y., Inoue, M., Kojima, H., Jimi, T., Shibuya, S., Hara, H.and Oniki, H. (2004) Expression and localization of aquaporin 7 innormal skeletal myofiber. Cell Tissue Res. 316, 123–129

Worman, H.J. and Field, M. (1985) Osmotic water permeability ofsmall intestinal brush-border membranes. J. Membr. Biol. 87,233–239

Wright, E.M. and Loo, D.D.F. (2000) Coupling between Na+, sugarand water transport across the intestine. Ann. N. Y. Acad. Sci. 915,54–66

Zeuthen, T. (1995) Molecular mechanisms for passive and activetransport of water. Int. Rev. Cytol. 160, 99–161

Zeuthen, T. (1996) Molecular Mechanisms of Water Transport,Springer, New York, R.G. Landes Company, Texas, 170 pp

Zeuthen, T., Meinild, A.K., Klaerke, D., Wright, E.M., Belhage, B., Loo,D.D.F. and Litman, T. (1997) Water transport by the Na+/glucosecotransporter under isotonic conditions. Biol. Cell 89, 307–312

Received 2 August 2004/12 October 2004; accepted 12 October 2004

Published as Immediate Publication 9 June 2005, DOI 10.1042/BC20040090

www.biolcell.org | Volume 97 (8) | Pages 605–613 613