Calcium-activated chloride channels (CLCA) are a recent addition to the family of chloride conductance proteins and participate in various physiological processes. Previously, CLCA isoforms have been identified with transepithelial transport as well as cell adhesion. The human isoform, human calcium-activated chloride channel 2 (hCLCA2), is believed to limit the effect of cystic fibrosis when expressed in the lung. Furthermore, hCLCA2 plays a key role in the adhesion of cancer cells to lung endothelia, and in the tumorigenicity of human breast cancer by virtue of its affinity for integrin ?4.

Members of the CLCA transport family have so far been located in human, murine and bovine epithelia and more recently hCLCA2 has been localised at the cytosolic aspect of basal epithelial cells. However, the functions of CLCA proteins and their association with integrin ?4 in normal tissue have yet to be fully identified. To understand better the function of hCLCA2 in human epithelia, quantitative measurements of hCLCA2 mRNA expression were performed in 15 different human epithelial tissues. Furthermore, double immunolabelling using an anti-hCLCA2 affinity-purified polyclonal antibody and monoclonal antibodies against integrin ?4 and collagen VII, both involved in cell-stroma adhesion, was carried out on several stratified epithelial tissues.

Purified mRNA from 15 human tissues, including uterus, trachea (BD Biosciences Clontech, San Jose, CA, USA), colon, brain (Ambion, Austin, TX, USA), vagina, oesophagus (Cosmo Bio, Tokyo, Japan), heart, skin, liver, spleen, stomach, lung, kidney (Stratagene, La Jolla, CA, USA) and cornea (mRNA extracted using the TRIzol method from excised corneal tissue, see below) (2 ?g of each) and 10 pmol oligo dT primer were heat-denatured in 5 ?l (total volume) at 70 °C for 2 min. After heat denaturation, reverse transcription (RT) reaction mixtures were assembled in 11 ?l (total volume), which included an additional 0.2 ?l of ribonuclease inhibitor (40 U/?l), all as specified by the manufacturer of the RT kit (Invitrogen, Carlsbad, CA, USA). The RT reaction mixtures were preheated to 40 °C for 2 min, after which 1 ?l of SuperScriptII RNase H^? (200 U/?l) was added. After a 60-min incubation at 40 °C, the reaction mixtures were cooled to 10 °C and 0.5 ?l of each RT reaction mixture was then used as the template for PCR as described below. For control RT reaction mixtures, SuperScriptII Rnase H^? was replaced by an equal volume of deionised water.

Specific primers for the hCLCA2 sequence were constructed with the sequence 5? GATGGGAGTACAGCTTCAAGA 3? (left) and 5? TTTTCCCTCTTTTCCACACC 3? (right). The final reaction conditions consisted of 0.5 ?l of RT template, left and right primers (10 pmol/?l), MgCl₂ (2.5 mM/?l), PCR buffer, deoxynucleoside triphosphate (2 mM/?l), Platinum Taq polymerase (5 U/?l) (Invitrogen) in 10 ?l of deionised water (total volume). RT-PCR amplification was facilitated by a touch-down method consisting of 3 cycles from 94 °C for 30 s, to 70 °C for 30 s and then repeated with 2 °C intervals until 62 °C followed by 2 sets of 3 cycles at 94 °C for 30 s, 60 °C (and then 58 °C) for 30 s and 72 °C for 30 s. Then, 25 cycles were performed at 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s, finishing with 6 min at 72 °C in 0.2-ml thin-walled tubes using a Biometra T-gradient Thermoblock. Control PCRs were the same as those described above. After RT-PCR amplification, 2 ?l of each reaction mixture was resolved on 1% agarose gel containing SYBR green (Molecular Probes, Eugene, OR, USA) (1:1000 dilution) in 1×Tris-acetate-EDTA running buffer. Images of each gel were captured by Polaroid film and digitised by a commercial flatbed scanner at 600 dpi. The quality of the cDNA products were assessed by measuring the presence of an ubiquitous housekeeping gene, ?-actin, amplified under the same conditions as above using the following primers 5? GGACTTCGAGCAAGAGATGG 3? (left) and 5? ATCTGCTGGAAGGTGGACAG 3? (right). Real-time PCR was performed using the cDNA products described above with probes against hCLCA2 and ?-actin (Applied Biosystems, Foster City, CA, USA), using a thermal profile of 50 °C for 2 min, 95 °C for 10 min followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. Real-time PCR measurements were detected by an ABI Prism 7000 (Applied Biosystems). For each of the 15 tissues, real-time PCR values against hCLCA2 were normalised by dividing by the corresponding values against ?-actin. Real-time PCR was repeated four times and the results averaged.

Vaginal, skin, corneal, oesophageal and laryngeal samples were obtained for immunohistochemistry from healthy tissue adjacent to excised tumors during surgery on five individual patients at the Kyoto Prefectural University of Medicine, Japan. Prior informed consent was obtained from each patient after a detailed explanation of the procedures in accordance with the tenets of the Declaration of Helsinki. Immediately after excision, specimens were frozen in Tissue Tek (Sakura Fineteck, Torrance, CA, USA). Cryostat sections, 7-?m thick, were collected on silanised glass slides and fixed in Zamboni fixative at 4 °C for 10 min followed by multiple washes in 0.01 M PBS. Non-specific antibody binding sites were blocked by incubation for 30 min with 0.01 M PBS containing 1% BSA. Then, sections were coincubated with anti-hCLCA2 antibody (1:10,000 dilution) and either anti-integrin ?4 (1:500 dilution; Chemicon, Temecula, CA, USA) or anti-collagen VII (1:200 dilution; Chemicon) for 1 h at 25 °C followed by 3×5 min washes in 0.01 M PBS. Preparation and characterisation of the hCLCA2 polyclonal antibody has been described previously. Sections were then incubated at room temperature for 1 h with a mixture of equal amounts of Alexa Fluor 488-conjugated anti-rabbit IgG antibody and Alexa Fluor 594-conjugated anti-mouse IgG antibody (1:1000 dilution; Molecular Probes). After washing with PBS, sections were mounted in medium containing DAPI (Vector, Burlingame, CA, USA), and examined by fluorescence microscopy (AX70; Olympus, Tokyo, Japan). Negative controls were performed by replacing the primary antibodies with 0.01 M PBS.

hCLCA2 was detected by RT-PCR in each of the 15 human epithelial tissues tested except brain, heart, kidney, stomach, liver and spleen. Furthermore, a more quantitative assessment of hCLCA2 expression by real-time PCR subsequently found that of the tissues tested those that contained stratified epithelium (cornea, oesophagus, larynx, skin and vagina) had a significantly higher level of hCLCA2 gene expression than those tissues which contain only epithelial monolayers.

Stratified epithelial tissues, shown by real-time PCR to contain high amounts of hCLCA2, were subjected to a double immunofluorescent labelling study to examine the colocalisation of hCLCA2, integrin ?4 and collagen VII. The resulting expression patterns in corneal, skin, laryngeal, vaginal and oesophageal epithelia displayed a distinct colocalisation between basal cell expression of hCLCA2 and integrin ?4, whereas collagen VII expression was separately resolved and found to be directly below, parallel and adjacent to hCLCA2 expression.

Double immunofluorescence labelling of stratified epithelia from the human tissues: cornea, skin, larynx, vagina and oesophagus. Anti-hCLCA2 (green) and anti-integrin ?4 (red) (column 1) show a distinct colocalisation (yellow) along the basal side of the basal epithelial cells. Anti-hCLCA2 (green) and anti-collagen VII (red) are similarly distributed but can be separately resolved (column 2) suggesting hCLCA2 and integrin ?4 are spatially more closely related. Column 3 shows the results of control incubations. Scale bars, 50 ?m.