Identification of multiple claudins in the rat epididymis

MOLECULAR REPRODUCTION AND DEVELOPMENT 73:580–588 (2006)

Identification of Multiple Claudinsin the Rat EpididymisMARY GREGORY AND DANIEL G. CYR*

INRS-Institut Armand-Frappier, Universite du Quebec, Pointe Claire, Quebec, Canada

ABSTRACT The luminal environment of theepididymis is highly specialized with specific proteins,ions, pH, etc. required for sperm maturation. Tightjunctions between epididymal principal cells are res-ponsible for the formation of the blood–epididymalbarrier, which regulates this luminal environment.Claudins (Cldns) are a recently discovered family oftransmembrane proteins and are essential componentsof tight junctions. Previous work from our laboratoryhas demonstrated the presence and localization ofCldn-1 in all regions of the rat epididymis. Theobjective of this study was to determine the presenceand localization of other Cldns in the epididymis. UsingRT-PCR we have identified mRNA transcripts for Cldn-3 through -9 in each region of the adult rat epididymis.Immunolocalization of Cldn-3, Cldn-4, and Cldn-5were done in adult as well as in 42- and 14-day-oldrats. Cldn-5 in adult rats was localized exclusively inblood vessels of the interstitium. Cldn-3 was localizedapically in the epididymal epithelium between adja-cent principal cells throughout the epididymis, wheretight junctions have been reported histologically. Therewere no differences in the localization of Cldn-3 inepididymides of rats at the different ages. In 14-day-old rats, Cldn-4 was localized all along the lateralplasma membrane between adjacent principal cells.The immunostaining was more pronounced in theproximal regions of the epididymis. In both 42-day-oldrats and adults, Cldn-4 was localized primarily toapical tight junctions between principal cells andstaining was more pronounced in the proximal regionof the epididymis. Cldn-16 transcripts were also identi-fied by RT-PCR. These transcripts were present in bothproximal and distal regions of the epididymis of young(Day 14 and 21) animals, but only in the proximal(initial segment) region of the adult epididymis. Thesedata indicate that epididymal tight junctions are com-posed of several Cldns, suggestive of a complex regu-lation of the blood–epididymal barrier. Mol. Reprod.Dev. 73: 580–588, 2006. � 2006 Wiley-Liss, Inc.

Key Words: tight junctions; development; principalcells; blood–epididymal barrier

INTRODUCTION

The production of mature sperm involves the differ-entiation of spermatogonia into spermatids (spermato-

genesis) and their maturation into spermatozoa in theseminiferous tubules of the testis (Clermont, 1972). Thefinal events of spermmaturation occur in the epididymiswhere spermatozoa acquire progressive motility andfertilizing ability (Robaire and Hermo, 1988). Sperma-tozoal maturation depends on several processes, includ-ing: interactions between spermatozoa and proteinssecreted by the epithelial principal cells of the epididy-mis; interactions between Golgi saccular elements ofthe sperm cytoplasmic droplet and the sperm plasmamembrane; and the changing environment in the lumenof the epididymis created by the blood–epididymalbarrier (Hoffer et al., 1973; Hamilton, 1975; Orgebin-Crist et al., 1975;Hinton, 1983, 1985;Hoffer andHinton,1984; Robaire and Hermo, 1988; Oko et al., 1993; Cyret al., 2001b).

Tight junctions between principal cells are responsi-ble for the formation of the blood–epididymal barrier(Cyr et al., 2001). The ultrastructural anatomy of theblood–epididymal barrier was first described by Friendand Gilula (1972) who reported the presence of a seriesof tight junctions between adjacent principal cells ofthe epididymis. Freeze fracture studies by Suzuki andNagano (1978) provided further evidence of a blood–epididymal barrier by showing the presence of tightjunctionmeshnetworks betweenprincipal cells. Indeed,Friend andGilula reported that tight junctions varied innumber of strands and complexity along the epididymis,being more extensive in the caput.

The epididymis is subdivided into four distinctregions: the initial segment, the caput, the corpus, andthe cauda epididymidis, and contains regional differ-ences in the ultrastructure of its tight junctions. Tightjunctions are composed of several proteins, includingoccludin, MAGUK proteins (notably, the family of ZO-1,-2, and -3), and JAM-associated proteins. Claudins(Cldns) are transmembrane proteins that have beenshown to be essential for tight junctions (Furuse et al.,

� 2006 WILEY-LISS, INC.

Grant sponsor: Toxic Substances Research Initiative (Health Canada);Grant sponsor: NSERC-CIHR Collaborative Research Grant.

*Correspondence to: Dr. Daniel G. Cyr, INRS-Insitut ArmandFrappier, 245 Hymus boul., Pointe Claire, Quebec, H9R 1G6 Canada.E-mail: [email protected]

Received 17 October 2005; Accepted 26 December 2005Published online 17 February 2006 in Wiley InterScience(www.interscience.wiley.com).DOI 10.1002/mrd.20467

1998; Morita et al., 1999a; Sonoda et al., 1999). Inaddition to their contribution to the barrier function oftight junctions, Cldns have also been associated withspecific ion transport. (e.g., Cldn-4; Lapierre, 2000;Colegio et al., 2003; Mitic et al., 2003; Van Itallie andAnderson, 2004). In the epididymis, occludin andCldn-1are present in tight junctions between adjacent princi-pal cells. Furthermore, our studies indicate that Cldn-1is regulated by both androgens and testicular factors(Cyr et al., 2001b).Several studies have reported that tight junctions are

composedof variousCldnsandoccludin, and that eachofthe transmembrane proteinsmayplay a different role inthe physiology of the tight junctions. Kamimura et al.(2002) reported the presence of multiple Cldns in thetestis. These authors also suggested that a combinationof various Cldns (e.g., Cldn-1, -3, -7, -8) may have beenresponsible for observed differences in permeabilitybetween adjacent Sertoli cells versus tight junctions ofmicrovessels in the testis. Absence of or changes inexpression levels of several Cldns have also beenassociated with alterations in permeability. Studies byAmashehetal. (2002)have reported that overexpressionof Cldn-2 in MDCK cells resulted in decreased inter-cellular resistance across a cultured epithelium. Cldn-4has been shown to influence paracellular ion selectivityin MDCK cells, and to decrease paracellular Naþ

permeability (Van Itallie et al., 2003). Likewise, theoverexpression of Cldn-4 decreased paracellular electricconductance and increased intercellular resistance,suggesting a more impermeable tight junction (VanItallie et al., 2001). Finally, it has been reported thatcertain Cldns, such as Cldn-16, are involved in para-cellular transport and may contribute to the formationof pore-like structures within the junction to allow themovement of certain solutes between cells (Blanchardet al., 2001; Hirano et al., 2002; Kamimura et al., 2002).This is particularly important in epithelia, where cellu-lar barriers are responsible for creating specific micro-environments, such as the epididymal lumen. SinceCldns are transmembrane proteins necessary for cellu-lar recognition and the structural integrity of tight jun-ctions, and because they polymerize with one another toform the tight-junctional strands, the specific character-istics of tight junctions are therefore related to whichproteins comprise the tight junction. The composition ofthese junctions, or changes in their composition withrespect to development or physiological status, is there-fore likely to define the regulation of tight junctions.The objectives of this study were to establish which

Cldns other thanCldn-1 and -2were expressed in the ratepididymis, to determine their localization to the area ofthe blood–epididymal barrier, and to assess if theirexpression is altered during the formation of the blood–epididymal barrier.

MATERIALS AND METHODS

Animals and Experimental Protocol

Adult rats. Adult male Sprague–Dawley rats (350–400 g) were purchased fromCharles River Canada, Inc.,

(St. Constant, Que.). Rats were maintained under aconstant photoperiod of 12-hr light:12-hr dark andreceived food and water ad libitum. All animal protocolsused in this study were approved by the UniversityAnimal Care Committee.

At the time of sampling, rats were euthanized withCO2. Epididymides were dissected, weighed, and sub-divided into either two (proximal (initial segment,caput, and corpus epididymidis) or distal (cauda epidi-dymidis, epididymis) or four separate regions (initialsegment, caput, corpus, and cauda epididymidis), de-pending on the age of the rats at the time of sampling.Tissues were either prepared for cryosections asdescribed below or frozen in liquid nitrogen and storedat �808C for RNA analysis.

Postnatal development. To determine the postna-tal developmental expression ofCldns in the epididymis,timed-gestation female rats were purchased fromCharles River Canada, Inc. At the time of birth, pupswere sexed and random litters of 10 male pups wereplaced with each lactatingmother. Rats were weaned atDay24of age.ForRT-PCR, epididymideswere sectionedinto caput-corpus and cauda epididymidis as previouslydescribed (Cyr et al., 2001a). Tissues were frozen inliquid nitrogen and subsequently stored at �808C. Forimmunocytochemistry, rats were killed on Days 14, 21,42, and 91. Tissues were quickly excised, frozen in OCTcompound (Fisher Scientific,Ottawa,ON) on dry ice andstored at �808C until sectioning.

Amplification of MultipleClaudins in the Epididymis

Oligonucleotide primers for Cldns 3–9, 11, and 16were synthesized based on published sequences presentin GenBank. Primer sequences were designed usingOligo software and synthesized commercially (SheldonLaboratories, McGill University, Montreal, Que.). Totalcellular RNA was extracted from either the proximal ordistal epididymis according to the method of Chomc-zynski and Sacchi (1987). Total RNA (500 ng) wasreverse transcribed using an oligo d(T)16 primer. EachCldn was amplified using the specific primers describedin Table 1. PCR amplificationwas done using the follow-ing program: 728C, 5 min; 948C, 5 min; 30–35 cycles of958C, 30 sec; 568C, 30 sec; 948C, 30 sec; 728C, 5 min foraddition of poly-A overhangs, and 48C, 5 min. PCRproducts were then separated on a 1.5% agarose gel andvisualized with ethidium bromide on a Fluor-S Multi-imager (BioRad Laboratories, Mississauga, ON). Bandsof the predicted size were excised, the amplicons wereextracted, and cloned into plasmid. The modifiedplasmids were then amplified, purified, and sequenced.The identity of the amplified DNA was determinedbased on sequence homology with BLAST search(GenBank, NIH).

Immunohistochemistry

Tissue sections were removed from the�808C freezerand fixed in methanol or acetone at �208C for 10 min.

IDENTIFICATION OF MULTIPLE CLAUDINS 581

Sectionswere circled with a PAP pen (DAKO,Carpentia,CA) and washed in 1� PBS at 48C for 20 min. Slideswere blockedwithBSAplus 5%goat or donkey serum for15 min at 48C, followed by three 5-min washes in 1�PBS. Tissues were incubated with primary antibodies(Cldn-3 and Cldn-4, gifts of Dr. J. Anderson and Dr. C.Van Itallie, University of North Carolina; Cldn-6, gift ofDr. K. Turksen, University of Ottawa, Canada; andCldn-11, gift of Dr.M. Furuse, KyotoUniversity, Japan)for 90minat room temperature, andwashed three timesfor 5 min with 1� PBS. Tissues were then incubatedwith secondary antibodies (FITC-conjugated anti-goator anti-rabbit, and Texas Red-conjugated anti-goat oranti-rabbit IgG, Jackson Immunoresearch,West Grove,PA). For co-localization studies, primary and secondaryantibodies were applied sequentially, with 1� PBSwashes in between each antibody application, as des-cribed above. Following the final secondary antibodyapplication, slides were again washed three times in 1�PBS, and mounted with Vectastain-mounting mediumcontaining either DAPI or propidium iodide (VectorLaboratories, Burlington, ON).

RESULTS

Cldn Transcripts

To determine which Cldns were present in the ratepididymis we designed specific PCR primers for Cldn-3through 9 and Cldn-11 and -16 (Table 1). We initiallyselected Cldn-4 through 9 as these were also present inthe kidney, and Cldn-11 since its expression had beenreported in the testis (Gowet al., 1999;Kiuchi-Saishin etal., 2002; Reyes et al., 2002). Since the epididymis andkidney are both derived from the mesonephros, thisincreased the likelihood that these Cldns were alsopresent in the epididymis. Our results indicate that the

rat epididymis has mRNA transcripts for Cldn-3, Cldn-4, Cldn-5, Cldn-6, Cldn-7, Cldn-8, Cldn-9, Cldn-11(Fig. 1). Positive control tissues were: adult rat lung orkidney mRNA for Cldn-3 through 9, and Cldn-16; andrat testis for Cldn-11. In all cases, a transcript ofidentical size to the positive control was amplified ineach of the four epididymal segments. ThePCRproductswere sequenced and their identities confirmed by Blastsearch comparison in GenBank.

Immunolocalization of Cldns

Immunocytochemistry was done to localize Cldn-3through Cldn-6, and Cldn-11 in the rat epididymis.Results indicated that Cldn-3, Cldn-4, and Cldn-6 werepresent in the epithelium of the epididymis and werelocalized to the area of the tight junctions between

TABLE 1. Sequence of Specific Primers Used toAmplify Different Claudins (Cldns) in the Adult

Rat Epididymis

Claudin Primer sequence

CLDN-3-F 50-GCGCGTTTCGGCATTCATCCLDN-3-R 50-GGATCTTGGTGGGTGCGTACTTCTCLDN-4-F 50-CCTGGGAATCTCCTTGGCAGTCCLDN-4-R 50-GTTGCTGGCGGGGACAGAGCLDN-5-F 50-CCTGGGAATCTCCTTGGCAGTCCLDN-5-F 50-GTTGCTGGCGGGGACAGAGCLDN-6-F 50-GCAAATCTTGGGGATCGTCCTGCLDN-6-R 50-TCTTGGTGGGATATTCGGAGGGTCCLDN-7-F 50-CGAGCACTGCCATCCCTCAGCLDN-7-R 50-AGGAATTGGACTTGGGGTAAGAGCCLDN-8-F 50-ACCTACGCTCTTCAAATGGGCCLDN-8-R 50-GAGATCTCTTTTCGGCGTCGAACLDN-9-F 50-AGGCTGGCTAGGAACTTTGGTCTCCLDN-9-R 50-CCAGTCCCGAAGCACCTGAACCLDN-11-F 50-GTCGCAGCAGTGCTCGCAGCCGCTCCLDN-11-R 50-GTCCTTACCTGGAAGGATGAGGATGCLDN-16-F 50-TTCAGTACGCTGCCTGCTTCLDN-16-R 50-ACATTTTGGCTGTCTCTGTCCTA

Sequences were obtained based on known sequences forhuman andmice Cldns present in GenBank and derived usingthe Oligo software (Molecular Biology Insights, version 6.0).

Fig. 1. Agarose gel electrophoresis of Cldn-3, Cldn-4, Cldn-5, Cldn-6,Cldn-7, Cldn-8, Cldn-9, and Cldn-11. RT-PCR products from adult ratepididymis. Primer and PCR conditions are indicated in Table 1.Results indicate that all eight Cldns are present in the epididymis. Thedistribution of these along the epididymis varies according to eachCldn(IS, initial segment; CT, caput epididymidis; CS, corpus epididymidis;CA, cauda epididymidis). Positive control RT-PCR reactions areindicated in the Materials and Methods.

582 M. GREGORY AND D.G. CYR

adjacent principal cells in the adult (Figs. 2–4). WhileCldn-3 and Cldn-4 were present between principal cellsin all regions of the epididymis, Cldn-6 was present onlyin the initial segment. Cldn-5 was not present in theepithelium of adult epididymis but localized to endothe-lial cells of blood vessels in the interstitial space(Fig. 4A). The expression of Cldn-3 and Cldn-4 alongthe epididymis appeared to be similar in the differentsegments of the adult epididymis. Cldn-3 and Cldn-4expression in the epididymis was also determined forrats of Days 14 (prior to barrier formation), 21 (at thecompletion of barrier formation), and 42 (entry ofspermatozoa in the epididymis) of age. Interestingly,there was a strong cytoplasmic reaction for Cldn-3 andCldn-4 throughout the epididymis of 14-day-old rats(Fig. 2A,B). By Day 21, both Cldns were localized to thearea of the tight junction (Fig. 2C,D). In 42-day-old rats,Cldn-3 expression was similar throughout the epididy-mis in the area of the tight junction, but Cldn-4immunoreaction was more intense in the proximalsegment of the epididymis, where spermatozoa werepresent (Fig. 3). Staining in adults (not shown) wascomparable to that of 42-day-old rats.

The expression of Cldn-6 only in the cauda epididy-midis of adult rat is particularly interesting (Fig. 4B).Previous studies had indicated that Cldn-6 was notpresent in tissues of adultmice,with the exception of thekidney, where low levels were amplified by RT-PCR(Turksen and Troy, 2001). Cldn-11 protein, present inthe testis (Fig. 4D), was not detectable in the epididymis(Fig. 4C), despite the presence of Cldn-11 mRNAtranscript (Fig. 1).

RT-PCR Development

Cldns-3, -4, -and -16 were studied as a function ofdevelopmental age. Results of RT-PCR of Cldn-3 andCldn-4 in young versus adult rats showed comparablemRNA levels for these Cldns throughout development,suggesting that mRNA levels for these Cldns are notregulated as a function of development. Cldn-16 mRNAlevels were distributed in both the proximal and distalregions of the epididymis in 14- and 21-day-old rats. Inadults, however, mRNA levels appeared to be very lowand restricted to the proximal region of the epididymis.Further analyses of the four different regions of theepididymis indicated that Cldn-16 transcripts were

Fig. 2. Immunocytochemical localization of Cldn-3 (A) and Cldn-4(B) in the epididymis of 14-day-old rat. At Day 14, the blood–epididymal barrier is not complete. Caput (A, B) epididymidis areshownas representative of immunostaining throughout the epididymisat this age. While both Cldn-3 (A) and Cldn-4 (B) are present in theepididymis, there is an important cytoplasmic reaction (arrows) at thisage, suggesting that Cldns have not yet completely localized to the areaof the tight junction. Magnification 640�. E, epithelial cells; IT,

intertubular space; L, lumen. Immunolocalization of Cldn-3 (C) andCldn-4 (D) in the epididymis of 21-day-old rat at the completion of theformation of the blood–epididymal barrier. Results indicate that by21 days of age, both Cldn-3 (C) and Cldn-4 (D) are localized (arrow) toarea of tight junction. Sections shown here are of the initial segment,but are representative of immunostaining observed in all other regionsof the rat epididymis at this age.Magnification 640�. P, principal cells;L, lumen; B, basal cells; IT, intertubular space.


expressed exclusively in the initial segment of theepididymis andwere not detectable in the caput, corpus,and cauda epididymidis.

DISCUSSION

There has been limited information on the proteinsthat are responsible for the formation of epididymaltight junctions and the blood–epididymal barrier. Ourprevious studies have shown that the transmembraneproteins occludin and Cldn-1 are both present inepididymal tight junctions. The present study indicatesthat there are transcripts for at least seven differentCldns in the adult rat epididymis. This suggests that the

composition of epididymal tight junctions is complexandthat epididymal tight junctions are likely to havedifferent roles with respect to the creation and main-tenance of the microenvironment of the lumen of theepididymis, which is necessary for sperm maturation.

It has recently been reportedbyGuanetal. (2005) thatthe epididymis ofWistar rats expressmultiple Cldns. Asin the present study, they observed the presence oftranscripts for Cldns-3, -4, -7, -8, and -11 (Guan et al.,2005). Interestingly they observed Cldn-6 expressiononly in 7-day-old rats and did not observe any Cldn-5,which we have shown in this study to be expressed inepididymal endothelial cells. They also indicate thatCldn-10 is expressed in the epididymis in both youngand adult rats. Likewise they report that Cldn-2 wasexpressed in the proximal initial segment of theepididymis. In a previous study we were unable toamplify Cldn-2 in adult rat epididymis (Gregory et al.,2001). Differences in the expression of these Cldns mayreflect differences between strains of rats or in assaysensitivity.

While RT-PCR indicates that there are at least sevendifferent Cldns in the epididymis of Sprague–Dawleyrats, not all Cldns are present in the epithelium of theepididymis. Cldn-5was localized exclusively to the bloodvessels present in the interstitial region of the epididy-mis (Fig. 5). Previous studies (Morita et al., 1999b) alsoreported that Cldn-5 is localized almost exclusively inendothelial cells that form blood vessels. The compart-mentalization of Cldn-5 in the circulatory system isinteresting and may be related to or responsible for thearchitecture of tight junctions in blood vessels. Tightjunctions of blood vessels are of the fascia occludenstype, so that these junctions do not occupy the entiremargin of the contiguous cell. Between these junctionsthere are slit-like clefts that allow the passage of fluidout of capillaries, for example. Therefore, the fact thatCldn-5 is localized to blood vessels suggests that it isnecessary for forming this type of architecture; hence, itis not surprising that this protein is present exclusivelyto blood vessels in the epididymis.

Cldn-3 has been described as a constitutive tightjunctional transmembrane protein, as it always co-localized with occludin in tight junctions ofMDCK cells.However, the immunolocalization of Cldn-3 and Cldn-4in 14-day-old rats indicated that at this stage ofdevelopment, both Cldn-3 and -4 exhibit strong cyto-plasmic immunostaining.While both Cldn-3 andCldn-4are present along the lateral plasma membrane of theepithelial cells which line the epididymal lumen, thecytoplasmic localization of Cldn-3 and Cldn-4 indicatesthat the targeting of these two proteins towards theapical plasma membrane of the epithelial cells is notcomplete at this age. This is similar to the observationbyGuan et al. (2005) for Cldn-10, which reported cytoplas-mic immunostaining. Thus, it would appear that thisphenomenon is not specific to Cldns-3 and -4, but alsoapplies to other Cldn family members, including Cldn-10. Previous studies have suggested that ZO-1 isinvolved in the targeting of Cldns to the area of the

Fig. 3. Immunocytochemical localization of Cldn-3 (A–D) andCldn-4 (E–H) in the epididymis of 42-day-old rat. In all regions of theepididymis (initial segment (A,E), caput (B,F), corpus (C,G) and cauda(D, H) epididymis), Cldn-3 and Cldn-4 were localized to the apicalregions of the epithelial principal cells that line the lumen of theepididymis (arrows). Interestingly, Cldn-4 immunostaining was muchmore intense in the initial segment and proximal caput (E, F) of theepididymis where spermatozoa were present. Magnification 640�. P,principal cells; B, basal cells; IT, intertubular space; L, lumen; S,spermatozoa.


tight junction (Itoh et al., 1999). Furthermore, we haverecently shown that ZO-1 immunoprecipitated with b-catenin in the epididymis of young rats but that thisassociation decreased as a function of development. Inyoung rats ZO-1 was not localized exclusively toepididymal tight junctions but rather all along thelateral plasma membrane (DeBellefeuille et al., 2003).The fact that ZO-1 at this age has not exclusivelylocalized to the tight junction suggests that Cldn-3 andCldn-4 have also not yet been targeted to the tightjunction. This would imply that in addition to ZO-1,there are other cytoplasmic proteins that either need tobe activated (phosphorylated) or transported to the tightjunction in order for these Cldns to be targeted to thetight junction. This observation is intriguing andsuggests that one of the regulatory mechanisms for theformation of the blood–epididymal barrier is not onlythe expression of specific genes but also their targetingto the tight junction. This notion is supported by somestudies, which have reported direct interactionsbetween the carboxyl termini ofCldnsandZO-molecules(Itoh et al., 1999).

By Day 21, both Cldn-3 and Cldn-4 are present at theepididymal tight junction. This suggests that at the timewhen the blood–epididymal barrier is complete, theproteins that comprise the epididymal tight junctionshave been targeted to the tight junction. By Day 42 theimmunostaining of Cldn-3 and Cldn-4 resembles that ofadult rats, indicating that at this time the epididymaltight junctions are similar to those of adults. Ultra-structural studies on the rat epididymis have shownthat by Day 40 the epididymal tight junctions haveassumed an adult-like appearance and that the numberof tight junctional strands appears to have stabilized(DeBellefeuille et al., 2003).The more intense Cldn-4immunostaining in the proximal segment of the epidi-dymis in 42-day-old rats, where spermatozoa werepresent (Fig. 3E,F), suggests that factors associatedwith the entry of spermatozoa may influence Cldn-4expression levels. SinceCldn-4 has been associatedwithdecreased membrane permeability to sodium ions inMDCK cells, it is tempting to speculate that itsincreased expression is related to a decrease in cationpermeability across the blood–epididymal barrier,

Fig. 4. Immunolocalization of Cldn-5 (A) andCldn-6 (B) in the adultepididymis. Cldn-5 (A) was localized exclusively to endothelial cells ofbloodvessels andno immunostainingwas observed in the epithelium inall regions of the epididymis. A photomicrograph from the caputepididymidis is shown.Cldn-6was present in the area of tight junctions(arrows) but little reaction was observed in the initial segment andcaput regions (not shown). Amore intense reactionwas observed in thecorpus and cauda epididymidis, shownhere (B).Magnification 640�. S,

spermatozoa; P, principal cells; IT, intertubular space; BV, bloodvessel. Immunocytochemical localization of Cldn-11 in the epididymis(C) and testis (D) of an adult rat. Date indicate that in the testis, Cldn-11 is present in the seminiferous tubule (ST) of adult rats betweenadjacent Sertoli cells (arrow). There was no detectable Cldn-11immunostaining in any region of the epididymis as demonstrated bythe photomicrograph of the caput epididymidis. L, lumen; P, principalcells; b, basal cells; IT, interstitial space; Ly, Leydig cells; n, nucleus.


thereby contributing to the lumenal environmentnecessary for spermatozoal maturation (Van Itallieet al., 2003).

Cldn-6 was detected by RT-PCR in all regions of theadult rat epididymis. However, while transcripts forCldn-6 were detected in all regions of the adultepididymis, Cldn-6 protein was expressed only in thedistal (cauda) epididymis, according to our immunos-taining studies (Fig. 4B). This corresponds in part toobservations by Turksen and Troy (2001), who did notdetect Cldn-6 in adult mice tissue. Little is knownregarding the role ofCldn-6 in tissues.TurksenandTroy(2001) demonstrated that Cldn-6 was developmentallyregulated in mouse embryonic epithelium. Their studyalso indicated that Cldn-6 was not present in tissues ofadult mice, with the exception of the kidney, where lowlevels were amplified by RT-PCR. It should be noted,however, that adult epididymis was not tested in thatstudy. The same group recently demonstrated thatoverexpression of Cldn-6 in mice leads to a defectiveepidermal permeability barrier and subsequent death48 hr after birth (Troy et al., 2005; Turksen and Troy,

2002). The expression of Cldn-6 in the cauda epididymi-dis of adult rat is particularly interesting, but itssignificance in this region of the adult rat remainsunclear.

Cldns-7, -8, and -9 were also detected by RT-PCR, tovarying degrees, in all segments of adult rat epididymis,with results comparable to those reported byGuan et al.(2005). These Cldns have been less extensively studied,and little is known regarding their specific role(s) intissues. It has been shown that suppression of occludinhas been associated with decreases in the expression ofCldns-1 and -7 (Yu et al., 2005). A recent study, in whichCldn-7 was overexpressed in LLC-PK1 cells, suggestedthat, as with certain other Cldn familymembers, Cldn-7may form a paracellular barrier to chloride ions, but aparacellular channel for sodium ions. Cldn-7 has beenreported recently to co-localize with ZO-1 in mosttight junctions of the mouse epididymal epithelium(Blackman et al, 2005). Cldn-8 has also been implicatedin the formation of a paracellular cation barrier,suggesting either a functional redundancy or a cell- ortissue-specificity of certain Cldns (Li et al., 2005).

Cldn-11 transcripts were detectable in the epididy-mis; however, there was no detectable Cldn-11 protein,despite its presence in the testis (Fig. 4). Cldn-11transcripts were also observed in several regions,namely, initial segment, caput, and corpus, of the ratepididymis, as reported by Guan et al. (2005), andexhibited a similar pattern to that observed in our study.Cldn-11 has been reported as being necessary for theformation of tight junctions in the testis (Gow et al.,1999). This is not the first time that differences in theexpression of junctional proteins between the testis andepididymis have been reported. Studies with the celladhesion protein, N-cadherin, indicated that it wasexpressed in the testis between developing germ cellsand Sertoli cells, but was not expressed in the epididy-mis (Cyr et al., 1992).

The expression ofCldn-16, alsoknownasparacellin-1,appears to vary considerably during development(Fig. 6). In young animals Cldn-16 is present in boththe proximal and distal regions of the epididymis.However, as the animals age, Cldn-16 appears to belocalized exclusively to the initial segment of theepididymis. Cldn-16 has been implicated in paracellulartransport, particularly that of magnesium and othercations (Van Itallie et al., 2001). Absence ormutations ofCldn-16 in bothhumanandbovine species have resultedin deficiencies or impaired magnesium and potassiumhandling in the kidney (Hirano et al., 2002;Weber et al.,2001). In the epididymis, the initial segment reabsorbsimportant quantities of water and ions from the fluidthat carries the sperm into the epididymis (Turner,2002). Potassium concentrations in epididymal fluid re-main fairly constant in the initial segment and increaseonly slightly in the distal epididymal regions (Turner,2002). It is therefore possible that one of the roles ofCldn-16 in the initial segment of the epididymis is toprevent the loss of potassium and magnesium into theepididymal fluid, particularly from the initial segment,

Fig. 6. Agarose gel electrophoresis of RT-PCR for Cldn-16 in theproximal (PE, initial segment caput corpus) and distal (DE, cauda)epididymal regions of 14 and 21 days of age as well as in adult. In adultrats RNAwas isolated from the initial segment (IS), caput (CT), corpus(CS), and cauda (CA) epididymidis. Cldn-16 was expressed throughoutthe epididymis in 14- and21-day-old rats but only in the initial segmentof adult rats. WC is a water negative control.

Fig. 5. Agarose gel electrophoresis of RT-PCR for Cldn-3, Cldn-4 inproximal (P, initial segment caput corpus) and distal (D, cauda)epididymal regions at different developmental ages. RT-PCR forGAPDH was performed to correct for loading differences. The dataindicate that mRNA levels for Cldn-3 and Cldn-4 are comparablethroughout postnatal development.


where there is considerable water and ion flux. Furtherstudies will be needed to test this hypothesis.In conclusion, these results indicate that epididymal

tight junctions are comprised of at least seven differentCldns. Furthermore, developmental studies indicatethat the targeting ofCldn-3andCldn-4maybe related tofactors, which regulate the formation of the blood–epididymal barrier. Finally, regional differences in theexpression of Cldn-16 from young animals as comparedto adults suggest that Cldn-16 may be important inmaintaining magnesium and/or potassium levels in theinitial segment of the epididymis following the forma-tion of the blood–epididymal barrier, thereby contribut-ing to the creation of specific ionic balances in thelumenal microenvironment.

ACKNOWLEDGMENTS

Dr. J. Anderson and Dr. C. Van Itallie (University ofNorth Carolina) are thanked for their generous gifts ofCldn-3 and Cldn-4 antisera. Dr. K. Turksen (Universityof Ottawa) is thanked for Cldn-6 antiserum, and Dr. M.Furuse (Kyoto University) is thanked for Cldn-11antiserum. This study was supported in part by theToxic Substances Research Initiative (Health Canada)and a NSERC-CIHR collaborative research grant.

REFERENCES

Amasheh S, Meiri N, Gitter AH, Schoneberg T, Mankertz J, SchulzkeJD, Fromm M. 2002. Claudin-2 expression induces cation-selectivechannels in tight junctions of epithelial cells. J Cell Sci 115:4969–4976.

Blackman B, Russell T, Nordeen SK, Medina D, Neville MC. 2005.Claudin 7 expression and localization in the normal murinemammary gland and murine mammary tumors. Breast Cancer Res7:R248–R255.

Blanchard A, Jeunemaitre X, Coudol P, Dechaux M, Froissart M, MayA,DemontisR, FournierA, PaillardM,Houillier P. 2001. Paracellin-1is critical for magnesium and calcium reabsorption in the humanthick ascending limb of Henle. Kidney Int 59:2206–2211.

ChomczynskiP, SacchiN. 1987. Single-stepmethod ofRNA isolationbyacid guanidinium thiocyanate-phenol-chloroform extraction. AnalBichem 162:156–159.

Clermont Y. 1972. Kinetics of spermatogenesis in mammals: Semi-niferous epithelium cycle and spermatogonial renewal. Physiol Rev52:198–236.

ColegioOR,Van ItallieCM,McCreaHJ,RahnerC,Anderson JM. 2003.Claudins create charge-selective channels in the paracellular path-way between epithelial cells. Am J Physiol Cell Physiol 283:C142–C147.

Cyr DG, Blaschuck OW, Robaire B. 1992. Identification and develop-mental regulation of cadherin messenger ribonucleic acids in the rattestis. Endocrinology 131:139–145.

Cyr DG, Dufresne J, Pillet S, Hermo L. 2001a. Immunolocalizationand regulation of metallothionein in the rat epididymis. J Androl 22:124–135.

Cyr DG, Finnson KW, Dufresne J, Gregory M. 2001b. Cellularinteractionsand theblood–epididymalbarrier. In:RobaireB,HintonBT, editors. The epididymis: Frommolecules to clinical practice.NewYork: Kluwer Academic/Plenum Press. pp 103–118.

DeBellefeuille S, Hermo L, Gregory M, Dufresne J, Cyr DG. 2003.Catenins in the rat epididymis: Their expression and regulation inadulthood and during postnatal development. Endocrinology 144:5040–5049.

Friend DS, Gilula NB. 1972. Variations in tight junctions and gapjunctions in mammalian tissues. J Cell Biol 53:758–776.

Furuse M, Fujita K, Hiragi T, Fujimoto K, Tsukita S. 1998. Claudin-1and -2: Novel integral membrane proteins localizing at tightjunctions with no sequence similarity to occludin. J Cell Biol 141:1539–1550.

Gow A, Southwood CM, Li JS, Pariali M, Riordan GP, Brodie SE,Danias J, Bronstein JM, Kachar B, Lazzarini RA. 1999. CNSmyelinand Sertoli cell tight junction strands are absent in Osp/claudin-11null mice. Cell 99:649–659.

Gregory M, Dufresne J, Hermo L, Cyr D. 2001. Claudin-1 is notrestricted to the tight junctions in the rat epididymis. Endocrinology142:854–863.

Guan X, Inai T, Shibata Y. 2005. Segment-specific expression of tightjunction proteins, claudin-2 and -10, in the rat epididymal epithe-lium. Arch Histol Cytol 68:213–225.

Hamilton DW. 1975. Structure and function of the epithelium liningthe ductuli efferentes, ductus epididymis and ductus deferens in therat. In: Greep RO, Astwood EB, editors. Handbook of physiology,Section 7, Volume 5. Washington, DC: American PhysiologicalSociety. pp 303–317.

Hinton BT. 1983. The epithelium of the caput epididymidis: Itspermeability and its role in the formation of a specialized micro-environment. Ann NY Acad Sci 383:462–463.

Hinton BT. 1985. The blood–epididymis barrier. In: Lobl TJ, HafezESE, editors. Male fertility and its regulation. Boston: MTP PressLtd. pp 371–382.

Hirano T, Hirotsune S, Sasaki S, Kikuchi T, Sugimoto Y. 2002. A newdeletion mutation in bovine Claudin-16 (CLDN-16) deficiency anddiagnosis. Anim Genet 33:118–122.

Hoffer AP, Hinton BT. 1984. Morphological evidence for a bloodepididymis barrier and the effects of gossypol on its integrity. BiolReprod 30:991–1004.

Hoffer AP, Hamilton DW, Fawcett DW. 1973. The ultrastructure of theprincipal cells and intraepithelial leucocytes in the initial segment ofthe rat epididymis. Anat Rec 175:169–201.

Itoh M, Furuse M, Morita K, Kubota K, Saitou M, Tsukita S. 1999.Direct binding of three tight junction-associated MAGUKs, ZO-1,ZO-2, and ZO-3, with the COOH termini of claudins. J Cell Biol147:1351–1363.

Kamimura Y, Chiba H, Utsumi H, Gotoh T, Tobioka H, Sawada N.2002. Barrier function of microvessels and roles of glial cell line-derived neurotrophic factor in the rat testis. Med Electron Microsc35:139–145.

Kiuchi-Saishin Y, Gotoh S, Furuse M, Takasuga A, Tano Y, Tsukita S.2002. Differential expression patterns of claudins, tight junctionmembrane proteins, in mouse nephron segments. J Am Soc Nephrol13:875–886.

Lapierre LA. 2000. The molecular structure of the tight junction. AdvDrug Deliv Rev 41:255–264.

Li Y, Fanning AS, Anderson JM, Lavie A. 2005. Structure of theconserved cytoplasmic C-terminal domain of occludin: Identificationof the ZO-1 binding surface. J Mol Biol 352:151–164.

Mitic LL, Unger VM, Anderson JM. 2003. Expression, solubilization,and biochemical characterization of the tight junction transmem-brane protein claudin-4. Protein Sci 12:218–227.

Morita K, FuruseM, Fujimoto K, Tsukita S. 1999a. Claudin multigenefamily encoding four-transmembrane domain protein components oftight junction strands. Proc Natl Acad Sci USA 96:511–516.

Morita K, Sasaki H, Furuse M, Tsukita S. 1999b. Endothelial claudin:Claudin-5/TMVCF constitutes tight junction strands in endothelialcells. J Cell Biol 147:185–194.

Oko R, Hermo L, Chan PTK, Fazel A, Bergeron JJM. 1993. Thecytoplasmic droplet of rat epididymal spermatozoa containssaccular elements with Golgi characteristics. J Cell Biol 123:809–821.

Orgebin-CristM-C, Danzo BJ, Davies J. 1975. Endocrine control of thedevelopment and maintenance of sperm fertilizing ability in theepididymis. In: Greep RO, Astwood EB, editors. Handbook ofphysiology, section 7, Volume 5. Washington, DC: AmericanPhysiological Society. pp 319–338.

Reyes JL, LamasM,Martin D, del CarmenNamoradoM, Islas S, LunaJ, Tauc M, Gonzalez-Mariscal L. 2002. The renal segmentaldistribution of claudins changes with development. Kidney Int 62:476–487.


Robaire B, Hermo L. 1988. Efferent ducts, epididymis, and vasdeferens: Structure, functions, and their regulation. In: Knobil E,Neill J, editors. The physiology of reproduction. New York: RavenPress Ltd. pp 999–1080.

Sonoda N, Furuse M, Sasaki H, Yonemura S, Katahira J, Horiguchi Y,Tsukita S. 1999. Clostridium perfringens enterotoxin fragmentremoves specific claudins from tight junction strands: Evidence fordirect involvement of claudins in tight junction barrier. J Cell Biol147:195–204.

Suzuki F, Nagano T. 1978. Development of tight junctions in the caputepididymal epithelium of the mouse. Dev Biol 63:321–334.

Troy TC, Rahbar R, Arabzadeh A, Cheung RM, Turksen K. 2005.Delayed epidermal permeability barrier formation and hair follicleaberrations in Inv-Cldn-6 mice. Mech Dev 122:805–819.

TurksenK,TroyTC. 2001.Claudin-6: Anovel tight junctionmolecule isdevelopmentally regulated inmouse embryonic epithelium.DevDyn222:292–300.

Turksen K, Troy TC. 2002. Permeability barrier dysfunction in trans-genic mice overexpressing claudin-6. Development 129:1775–1784.

Turner TT. 2002. Necessities potions: Inorganic ions and small organicmolecules in the epididymal lumen. In: Robaire B, Hinton BT,

editors. The epididymis: From molecules to clinical practice. NewYork: Plenum Press. pp 131–150.

Van Itallie CM, Anderson JM. 2004. The role of claudins indetermining paracellular charge selectivity. Proc Am Thorac Soc 1:38–41.

Van Itallie C, Mitic LL, Anderson JM. 2001. Occludin and claudins intight junctions. In: CerejiedoM,Anderson J, editors. Tight junctions.2nd edn. Boca Raton, Florida: CRC Press. pp 213–230.

Van Itallie CM, Fanning AS, Anderson JM. 2003. Reversal of chargeselectivity in cation or anion-selective epithelial lines by expressionof different claudins. Am J Physiol Renal Physiol 285:F1078–F1084.

Weber S, Schneider L, Peters M, Misselwitz J, Ronnefarth G, BoswaldM, Bonzel KE, Seeman T, Sulakova T, Kuwertz-Broking E, GregoricA, Palcoux JB, Tasic V,Manz F, Scharer K, SeyberthHW,KonradM.2001. Novel paracellin-1 mutations in 25 families with familialhypomagnesemia with hypercalciuria and nephrocalcinosis. J AmSoc Nephrol 12:1872–1881.

Yu AS, McCarthy KM, Francis SA, McCormack JM, Lai J, Rogers RA,Lynch RD, Schneeberger EE. 2005. Knockdown of occludin expres-sion leads to diverse phenotypic alterations in epithelial cells. AmJ Physiol Cell Physiol 288:1231–1241.


Identification of multiple claudins in the rat epididymis

Documents

Transcript of Identification of multiple claudins in the rat epididymis