Expression And Significance Of Aquaporin-3 In Normal And Neoplastic Tissues Of Multiple Human Organs

The research on Aquaporins has been focused since Agre and MacKinnon wereawarded the Novel Prize in 2003. Aquaporins (AQPs), the water channel proteins,mediate the efficient movement of water across the membrane. To date, thirteenisoforms of aquaporins, AQP0-AQP12 have been characterized in mammals and fivein lung tissue (AQP1, 3, 4, 5 and 8). As to the distribution of AQPs, there is specialtybetween different kind of tissues, cells and sub-cells. AQPs not only plays an importantrole for the maintenance of water and material homeostasis in normal physiologic stage,but also facilitate trans-epithelial fluid transport (including water, glycerol, urea, etc.),involved in swelling of tissues under stress, facilitate cell migration and also beinvolved in neural signal transduction.The research about Aquaporins and tumor has also been focused more and more.AQP 1 has been found associated with nephradenoma, breast cancer, cholangiocellularcarcinoma and hemangioblastoma of CNS. AQP2 expressed in renal cell carcinoma.AQP3 can help for diagnosis of collecting duct carcinoma. AQP5 may be concernedwith the tumor genesis of ovary.Lung cancer is one of the most malignant tumors for human's life, whosemechanism of water and substance metabolism of the lung is still unknown. Toinvestigate the expression of AQPs in normal and neoplastic lung tissues can conduceto know about water and substance metabolism of the lung, and furthermore do good toknow about the tumorgenesis, development, diagnosis and prognosis of lung cancer.Differential diagnosis between tumors is very important for clinical pathology,especially for hunting the primary location when a metastatic carcinoma is met. As for the specialty of AQPs' distribution between different kind of tissues, cells and sub-cells,to investigate and compare the expression and distribution of AQPs will make sense fordiagnosis and differential diagnosis between different adenorcarcinomas.To evaluate the presence and expression pattern of AQP3 in human's normal andneoplastic tissues, we studied 159 pulmonary carcinoma tissues, 86 adenocarcinomatissues other than lung, and 2 cell lines (A549 and Sq-1) by immunohistochemistry,Western blot and RT-PCR, and finally discussed its clinicopathological significance.Materials and Methods1. Tissue Preparation: 159 cases of paraffin sections of lung carcinomascomprised of adenocarcinoma (84 cases), squamous cell carcinoma (42 cases), smallcell lung carcinoma (10 cases), large cell carcinoma (15 cases), pleomorphic carcinoma(4 cases), and metastatic colon carcinoma (4 cases), and 86 adenocarcinomas fromsome organs other than lung, including colon (15 specimens), stomach (5 specimens),prostate (5 specimens), breast (35 specimens), uterus body (11 specimens), ovary (10specimens) and thyroid (5 specimens) were chosen from the surgical files ofYamanashi University Hospital and Kofu City Hospital. All specimens routinelyprocessed after surgery and embedded in paraffin were available.2. Immunohistochemistry and immunofluorescence: EnVision Two-Stepmethod was used for immunohistochemistry staining. After Heat-induced epitoperetrieval carded out for 10 minutes at 121℃, samples were incubated with thepolyclonal, affinity-purified rabbit polyclonal antibodies to rat/human AQP3 at 1:500dilution for overnight at 4℃. Next steps were done as usual. Indirectimmunofluorescence staining was carried out basically as described previously, with aFITC--conjugated swine anti-rabbit IgG as the second antibody. For controls, PBS wasused instead of the primary antibody as a negative control. The tissue retrieved fromkidney was used as a positive control. The immunoreactivity was evaluated using ascale from-to 4+ (-, negative; 1+, positive of under 25% of cancer cells; 2+, positiveof 26-50% of cancer cells; 3+, positive of 51-75% of cancer cells; 4+, positive of more than 75% of cancer cells).3. Western blot analysis: Homogenized tissues in RIPA lysis buffer weresolubilized in sample buffer and heated to 70°for 10 min. These samples were loadedon a 13% polyacrylamide gel and transferred to an PVDF membrane. Unstainedmembranes were blocked in blotting buffer containing 5% nonfat dry milk in PBS0.05% Tween 20, followed by incubation with anti-AQP3 antibody diluted 1:2000 inblotting buffer at 4°overnight. The membranes were then washed in blotting buffer,incubated for 60 min with 1:2,000 dilution in blotting buffer of goat anti-rabbit IgGconjugated to horseradish peroxidase, washed in PBS-Tween, and detected using ECLtechnique. Stained bands were scanned and pixel intensity was quantified on NIHImage.4. RNA extraction and RT-PCR analysis: Total RNA was prepared from frozentissues, paraffin-embedded tissues and cell lines using the acidguanidine-phenol-chloroform RNA extraction system according to the manufacturer'sinstructions. The quantity of total RNA was first determined by OD260 measurement,and the quality of total RNA was estimated by 1.5% agarose gel electrophoresis.RT-PCR was performed according to the manufacturer's instructions. The PCR protocolwas 8 minutes at 94℃; 30 seconds at 94℃(denature), 1 minute at 57℃(annealing)and 30 seconds at 72°C (extension) for 35 cycles; and 10 minutes at 72°C. Theprimers' sequence is: AQP3 forward 5'- GAAGGTGAAGGTCGGAGTC- 3', reverse5'-GAAGATGGTGATGGGATTTC-3'(180bp), GAPDH forward 5'-ATCAACCTGGCCTTTGGCTT-3', reverse 5'-TATTCCAGCACCCAAGAAGGCTT-3'(226bp).5. Cell culture: Two cell lines consists of a lung adenocarcinoma cell line (A549)and a human lung squamous cell carcinoma cell line (Sq-1), which were provided fromCell Resource Center for Biomedical Research of Tohoku University. Both cells weremaintained in RMPI 1640 culture medium containing penicillin G (100 U/ml),streptomycin (100 mg/ml), and 10% FCS. Cells were routinely cultivated in a humidified atmosphere of 5% CO2 and 95% air at 37℃.6. Statistics analysis: x~2 test was employed to analyze the data by the SPSS 12.0software. P＜0.05 was considered as statistical significance.Results1. The expression of AQP3 in normal lung tissues: In the pseudostratified epitheliaof bronchi, AQP3 was highly expressed in the cytoplasmic membrane of basal cells,faint or absent in ciliated columnar cells or goblet cells, and partly seen in the mucinsecreting cells of some bronchial glands. In bronchioles, AQP3 immunopositivity wasobserved in all columnar cells facing the lumen as well as in basal cells. AQP3 wasevident in the mucin secreting cells of some bronchial glands (Figure 2C). In alveolartissue, AQP3 was specially expressed in typeⅡalveolar cells.2. AQP3 was observed in 47.8% (76/159) of the lung carcinoma tissues. Thehighest and strongest immunoreactivity of AQP3 was observed in adenocarcinomas(70.2%, 59/84), while lower and fainter immunoreactivity was observed in squamouscell carcinomas (35.8%) and large cell carcinoma (13.4%). No specificimmunoreactivity of AQP3 was observed in small cell carcinoma, pleomorphiccarcinoma and metastastic colon carcinoma.3. In the subtypes of lung adenocarcinoma, the frequency of AQP3immunoreactivitywas: 100% (12/12) in bronchioloalveolar subtype, 82.1% (32/39)in papillary subtype, 52.9% (9/17) in acinar subtype and 25.0% (4/16) solid with mucinsubtype, respectively.4. By Statistics analysis, AQP3 immunopositivity was positively correlated withclinical stage (x~2=37.86, P＜0.01) and negatively correlated with tumor differentiation(x~2=21.22, P＜0.01) of lung adenocarcinoma.5. Western blotting and RT-PCR analysis furtherly confirmed the protein andmRNA expression of AQP3 in cell line and tissue sample of lung carcinoma.6. The frequency of AQP3 immunoreactivity in normal and neoplastic tissues ofseven organs were : both of 100%(5/5) in prostate, 75.0% (6/8) and 14.3% (5/35) in breast, 60%(3/5) and 80.0%(4/5) in stomach, 14.3% (3/21) and 27.2%(3/11) inendometrium, and 37.5% (3/8) and 20.0% (3/15) in colon, respectively. No specificimmunoreactivity of AQP3 was observed in ovary and thyroid adenocarcinoma.Conclusions1. AQP3 distributes widely and can play an important role for the maintenance ofwater homeostasis in the respiratory tract.2. AQP3 is expressed in lung carcinoma in different extension, highly expressed inlung adenocarcinoma, especially in bronchi-alveolar subtype. The immunoreactivity ofAQP3 is positively correlated with clinical stage and negatively correlated with tumordifferentiation of lung adenocarcinoma by statistics analysis, suggesting that AQP3may be concerned with the development of lung adenocarcinoma.3. That no specific immunoreactivity of AQP3 was observed in ovary and thyroidadenocarcinoma makes AQP3 to be a candidate marker to exclude ovary and thyroid inidentifying the origin of a metastasitic adenocarcinoma.