research article prognostic value of combined aquaporin 3...

Hindawi Publishing CorporationBioMed Research InternationalVolume 2013 Article ID 206525 7 pageshttpdxdoiorg1011552013206525

Research ArticlePrognostic Value of Combined Aquaporin 3 and Aquaporin 5Overexpression in Hepatocellular Carcinoma

Xiaodong Guo1 Ting Sun2 Mei Yang1 Zhiyan Li1 Zhiwei Li1 and Yuejuan Gao1

1 302 Hospital of PLA Beijing 100039 China2Navy General Hospital Beijing 100048 China

Correspondence should be addressed to Zhiyan Li guoxd 1981163com

Received 11 May 2013 Revised 29 August 2013 Accepted 2 September 2013

Academic Editor Koichiro Wada

Copyright copy 2013 Xiaodong Guo et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Background Aquaporin (AQP) 3 and AQP 5 are involved in tumorigenesis and tumor progression of several tumor typesAim To investigate expression patterns and clinical significance of AQP3 and AQP5 in hepatocellular carcinoma (HCC)Methods Immunohistochemistry was performed to detect the expression of AQP3 and AQP5 in HCC tissues ResultsImmunohistochemistry analysis showed the increased expression of AQP3 and AQP5 protein levels in HCC tissues compared withtheir adjacent nonneoplastic tissues (both 119875 lt 0001) In addition combined AQP3 and AQP5 protein expression was significantlyassociated with serum AFP (119875 = 0008) tumor stage (119875 = 0006) and tumor grade (119875 = 0006) Moreover HCC patients highlyexpressing both AQP3 and AQP5 proteins had worse 5-year disease-free survival and 5-year overall survival (119875 = 0002 and 0005resp) Furthermore the Cox proportional hazards model showed that combined AQP3 and AQP5 protein expression was anindependent poor prognostic factor for both 5-year disease-free survival (119875 = 0009) and 5-year overall survival (119875 = 001) inHCC Conclusion Our data suggest for the first time that the aberrant expression of AQP3 and AQP5 proteins may be stronglyrelated to tumor progression and prognosis in patients with HCC The overexpression of AQP3 in combination with upregulationof AQP5 may be an unfavorable prognostic factor for HCC

1 Introduction

Hepatocellular carcinoma (HCC) represents one of the mostcommon tumors worldwide and ranks as the third causeof cancer-related death especially in east Asia and sub-Saharan Africa [1] In China HCC accounts for about 110000deaths annually and is the second leading cause of cancer-related death among men [2] In the USA and Europe theincidence of HCC has also been increasing in the recentyears As a highly aggressive solid tumor HCC is character-ized by fast infiltrating growth early metastasis high-grademalignancy and poor prognosis Despite improvements intreatment modalities during the past few decades the long-term survival remains unsatisfactory mainly because of ahigh incidence of postoperative metastasis and recurrenceand the high resistance of HCC to chemotherapy [3] Sincetumor progression of HCC is a complicated process that isassociated with cumulative genomic alterations the abnor-mal expression of oncogenes and tumor suppressors may

be responsible for the development of HCC Thus it isnecessary to perform further insight into the genes involvedin hepatocarcinogenesis and to identify novel markers forHCC diagnosis and prognosis

Aquaporins (AQPs) are a family of small (sim30 kDamonomer) hydrophobic channel-forming membrane pro-teins that are expressed widely in the animal and plant king-doms and they are involved in the transepithelial fluid trans-port occurring in the urinary concentrating mechanism andglandular fluid secretion [4] There are 13 members (AQP0simAQP12) having been identified so far in mammals Amongthem AQP0 AQP1 AQP2 AQP4 AQP5 AQP6 and AQP8are primarily water selective whereas AQP3 AQP7 AQP9AQP10 and AQP12 also transport glycerol and possibly othersmall solutes In addition to the classical function as osmot-ically driven transepithelial and transcellular water trans-porters AQPs are also involved in swelling of tissues understress as in the injured cornea and the brain in stroke tumorand infection [5 6] In particular accumulating evidence

2 BioMed Research International

suggests the diagnostic and prognostic value of the aberrantexpression of AQPs AQP expression in some reports is asso-ciated with tumor progression In liver tumors Mazal et al[7] demonstrated that AQP1 expression in HCC tissues waslower than that in cholangiocarcinoma suggesting that AQP1might be a highly selective marker for differentiated cholan-giocytes and can be very helpful in the differential diagnosisof liver tumors Padma et al [8] reported that human HCCmay be characterized by altered AQP9 expression and thatAQP9 localization in the nontumourigenic liver mass maybe dependent on underlying liver pathology Jablonski et al[9] found that AQP8 expression and AQP9 expression weresignificantly decreased in HCC versus normal liver Thesefindings implicated that the AQPs might play an importantrole in human HCC

Of 13AQP varieties AQP1 -8 and -9were downregulatedin human HCC tissues [7ndash9] The functional role of otherAQPmembers in HCC has not been fully elucidated Amongthem AQP3 cloned in 1994 functions as a membranechannel of water and other small solutes such as glyceroland urea and plays a major role in fluid homeostasis [10]AQP5 cloned from the salivary gland is a 27-kDa proteinthat is known as an exocrinetype water channel with a uniquetissue expression [11] AQP3 that transports not only waterbut also glycerol and urea is known to be expressed in kidneyskin lung and gastrointestinal tracts [10] and AQP5 thattransports only water is known to be expressed in variousorgans such as lung and saliva gland [12] Recent studies haveshowed the altered expression of AQP3 and AQP5 in varioustumor types However little is known about expression andprecise role of the two proteins on HCC Therefore the aimof this study was to investigate the expression patterns andclinical significance of AQP3 and AQP5 in human HCC

2 Materials and Methods

21 Patients and Tissue Samples The study was approved bythe Research Ethics Committee of 302nd Hospital of PLABeijing China Informed consentwas obtained fromall of thepatients All specimens were handled and made anonymousaccording to the ethical and legal standards

A total of 130 patients with primaryHCCwho underwenta curative liver resection at the 302nd Hospital of PLABeijing China were included in this retrospective studyTissues used in the study were retrieved from the tissue bankof the Department of Pathology at the 302nd Hospital ofPLA These patients were diagnosed as HCC between 2001and 2006 None of the patients recruited in this study hadchemotherapy or radiotherapy before the surgery HCC diag-nosis was based on the World Health Organization (WHO)criteria Tumor differentiation was defined according to theEdmondson grading system Liver function was assessedusing the Child-Pugh scoring system Tumor staging wasdetermined according to the sixth edition of the tumor-node-metastasis (TNM) classification of the InternationalUnion against CancerThe clinicopathological features of 130patients are summarized in Table 1

The median follow-up period was 86 years Postoper-ative surveillance included routine clinical and laboratory

examinations every third month computed tomographyscans of the abdomen and radiographs of the chest everythird month After 5 years the examination interval wasextended to 12 months

22 Immunohistochemistry Analysis AQP3 expression andAQP5 expression were immunohistochemically evaluated inparaffin-embedded specimens of 130 patients with HCCSurgical specimens were fixed in 10 formalin embed-ded in paraffin and sectioned at a 4120583m thickness Forheat-induced epitope retrieval deparaffinized sections weresoaked in 10mM citrate buffer (pH 60) and treated at 95∘Cfor 30min using the microwave oven method Immuno-histochemical staining was performed using the avidin-biotin immunoperoxidase technique according to our pre-vious studies [13ndash15] The activity of endogenous peroxidasewas blocked by incubation with 03 H

2O2in methanol

for 15min and nonspecific immunoglobulin binding wasblocked by incubation with 10 normal goat serum for10min Sections were incubated at room temperature for 4 hwith anti-AQP3 rabbit polyclonal antibody (sc-20811 SantaCruz Biotechnology Inc USA) or with anti-AQP5 rabbitpolyclonal antibody (sc-28628 Santa Cruz BiotechnologyInc USA) at a 1 100 or 1 150 dilution and they werethen rinsed and incubated for 30min with a biotinylatedsecond antibody After washing the sections were incubatedfor 30min with horseradish peroxidase-conjugated strepta-vidin and were finally treated with 331015840-diaminobenzidinetetrahydrochloride in 001 H

2O2for 10min The slides

were counterstained with Meyerrsquos hematoxylin The negativecontrols were processed in a similarmanner with PBS insteadof primary antibody The positive AQP3 expression andAQP5 expression confirmed by western blotting were usedas positive controls for immunostaining

Following a hematoxylin counterstaining immunostain-ing was scored by two independent experienced pathologistswho were blinded to the clinicopathological parameters andclinical outcomes of the patients The scores of the twopathologists were compared and any discrepant scores weretrained through reexamining the staining by both patholo-gists to achieve a consensus score The number of positive-staining cells showing immunoreactivity in the cytoplasmand cell membrane for both AQP3 and AQP5 in ten repre-sentative microscopic fields was counted and the percentageof positive cells was calculated The percentage scoring ofimmunoreactive tumor cells was as follows 0 (0) 1 (1ndash10) 2 (11ndash50) and 3 (gt50) The staining intensity wasvisually scored and stratified as follows 0 (negative) 1 (weak)2 (moderate) and 3 (strong) A final score was obtained foreach case by multiplying the percentage and the intensityscore Therefore tumors with a multiplied score exceedingmedian of total scores for AQP3 or AQP5 were deemed tobe low expressions of AQP3 or AQP5 all other scores wereconsidered to be high expressions of AQP3 or AQP5

23 Statistical Analysis The software of SPSS version 130 forWindows (SPSS Inc IL USA) and SAS 91 (SAS InstituteCary NC) was used for statistical analysis The chi-squaredtest was used to show differences in categorical variables

BioMed Research International 3

Table 1 Association of AQP3 and AQP5 expression with clinicopathological features of 130 hepatocellular carcinoma patients

Clinicopathological features Case AQP3-high (n ) P AQP5-high (n ) P AQP3-highAQP5-high (n ) PAge (years)le50 72 47 (6528) NS 40 (5556) NS 30 (4167) NSgt50 58 35 (6034) 33 (5690) 22 (3793)

GenderMale 96 62 (6458) NS 54 (5625) NS 40 (4167) NSFemale 34 20 (5882) 19 (5588) 12 (3529)

Serum AFPPositive 72 64 (8889) 0002 43 (5972) NS 45 (6250) 0008Negative 58 18 (3103) 30 (5172) 10 (1724)

Tumor stageT1 23 0 (0)

0005

0 (0)

0008

0 (0)

0006T2 40 21 (5250) 17 (4250) 10 (2500)T3 52 46 (8846) 40 (7692) 27 (5192)T4 15 15 (10000) 15 (10000) 15 (10000)

Tumor gradeG1 31 20 (6452)

NS13 (4194)

00093 (968)

0006G2 76 49 (6447) 40 (5263) 29 (3816)G3 23 13 (5652) 20 (8696) 20 (8696)

Growth patternTrabecular 101 63 (6238) NS 54 (5247) NS 40 (3960) NSNontrabecular 29 19 (6552) 19 (6552) 12 (4138)

CirrhosisYes 86 46 (5349) NS 47 (5465) NS 34 (3953) NSNo 44 26 (5909) 26 (5909) 18 (4091)

Underlying liver diseaseAlcoholic 25 15 (6000)

NS

15 (6000)

NS

7 (280)

NSHepatitis B 49 28 (5714) 28 (5714) 18 (3673)Hepatitis C 35 27 (7714) 18 (5143) 15 (4286)Unknown 21 12 (5714) 12 (5714) 12 (5714)

Note ldquoNSrdquo refers to that the differences among groups have no statistical significance

Correlations betweenAQP3 expression andAQP5 expressionwere calculated using Spearmanrsquos correlation Patient survivaland the differences in patient survival were determined bythe Kaplan-Meier method and the log-rank test respectivelyA Cox regression analysis (proportional hazard model) wasperformed for themultivariate analyses of prognostic factorsDifferences were considered statistically significant when 119875was less than 005

3 Results

31 Expression Patterns and Subcellular Localization of AQP3and AQP5 Proteins in HCC The subcellular localizationand the expression pattern of AQP3 and AQP5 proteinsin 130 self-pairs of HCC and adjacent nonneoplastic livertissueswere observed by the immunohistochemistry analysisAs shown in Figure 1 both AQP3 positive staining andAQP5 positive staining were localized in the cytoplasm andmembrane of tumor cells in HCC tissues Compared with theadjacent nonneoplastic tissues the immunohistochemistryscores of AQP3 (mean plusmn SD 561 plusmn 023 versus 218 plusmn 009

119875 lt 0001) and AQP5 (mean plusmn SD 628 plusmn 036 versus216 plusmn 005 119875 lt 0001) proteins were both significantlyincreased in HCC tissues In addition the expression levelsof AQP3 and AQP5 in 130 HCC cases were summarized inTable 2 Based on the scoring system used in the presentstudy 52 (4000) cases were both high expression of AQP3and AQP5 27 (2077) cases were both low expression ofAQP3 and AQP5 30 (2308) cases were AQP3 high andAQP5 low expression and 21 (1615) cases were AQP3 lowand AQP5 high expression As determined by Spearmanrsquoscorrelation the AQP3 expression was significantly associatedwith the AQP5 expression (119903 = 076 119875 = 001 Table 2)

32 Association of AQP3 and AQP5 Protein Expression withthe Clinicopathological Features of HCC To evaluate whetherAQP3 protein expression and AQP5 protein expression wereassociated with clinicopathological features of patients withHCC we correlated immunohistochemical AQP3 and AQP5staining results with tumor stage tumor grade serum AFPlevel presence of cirrhosis and underlying liver diseaseincluding alcohol abuse viral hepatitis B and C sex and


(b)(a)

Figure 1 Representative immunohistochemical images of AQP3 (a) expression and AQP5 (b) expression in HCC tissues (originalmagnification times400) AQP3 and AQP5 positive staining results were both indicated by numerous yellowish granules in the cytoplasm andmembrane of tumor cells in HCC tissues

Table 2 Expression of AQP3 and AQP5 proteins in 130 hepatocel-lular carcinoma patients

AQP5 expressionP

High (119899 = 82) Low (119899 = 48)

AQP3 expressionHigh (119899 = 73) 52 21 001Low (119899 = 57) 30 27

age (Table 1) According to the results we found that theexpression levels of AQP3 protein in HCC tissues with thehigher tumor stage (T3sim4) and the positive serum AFP levelwere significantly lower than those with the lower tumorstage (T1sim2 119875 = 0005 Table 1) and the negative serumAFP level (119875 = 0002 Table 1) respectively In addition thefrequencies of aberrant AQP5 expressionwere higher inHCCtissues with higher tumor stage (T3sim4) than those with lowertumor stage (119875 = 0008 Table 1) AQP5 overexpression wasalso observed more frequently in HCC tissues with hightumor grade than those with low grade (119875 = 0009 Table 1)Moreover combinedAQP3 andAQP5protein expressionwassignificantly associated with serum AFP (119875 = 0008 Table 1)tumor stage (119875 = 0006 Table 1) and tumor grade (119875 = 0006Table 1)

33 Prognostic Values of AQP3 and AQP5 Protein Expressionin HCC Five-year disease-free survival was observed in 30(2308) patients whereas in 100 (7692) patients diseaserecurred and 88 (6769) even died during a 5-year follow-upperiod We observed a trend that 5-year disease-free survivalin the group with high AQP3 expression was significantlypoorer than that in the group with low AQP3 expression(119875 = 0005 log-rank test Figure 2(a)) Additionally theKaplan-Meier plot of 5-year overall survival curves stratifiedby AQP3 expression was shown in Figure 2(b) A significantrelationship was found between AQP3 expression and 5-year overall survival (119875 = 0008 log-rank test Figure 2(b))Similar withAQP3 the disease-free survival (Figure 2(c)119875 =0002) and overall survival (Figure 2(d) 119875 = 0006) of HCCpatients with high AQP5 expression were both significantly

shorter than those with low AQP5 expression Moreover theassociation between coexpression of AQP3AQP5 and thesurvival rates was tested by the method of Kaplan-Meier TheChi-square value by log-rank test (Mantel-Cox) indicated asignificant difference among different groups with regard tothe conjoined expression status of AQP3AQP5 (Figures 2(e)and 2(f)) The results by pairwise comparisons showed thatthe statistically significant difference of disease-free survivaland overall survival existed between AQP3-highAQP5-highpatients and any of other three groups (119875 = 0002 and 0005resp) In all four groups AQP3-highAQP5-high patientshad the poorest prognosis

Furthermore in a multivariate Cox model includingserum AFP tumor stage tumor grading presence of cir-rhosis gender age AQP3 expression AQP5 expression andcombined AQP3AQP5 expression we found that AQP3expression (both 119875 = 001 Table 3) AQP5 expression (119875 =0006 and 001 Table 3) and combined AQP3AQP5 expres-sion (119875 = 0009 and 001 Table 3) were independent poorprognostic factors for both 5-year disease-free survival and5-year overall survival in HCC

4 Discussion

In the current study we determined the expression patternsof AQP3 and AQP5 proteins in 130 HCC tissues and pairedadjacent nonneoplastic tissues using immunohistochemistryanalysis We confirmed that the overexpression of AQP3mainly occurred in the cytoplasm and cellmembrane inHCCtissues relative to adjacent nonneoplastic tissues and thatAQP5 expression was markedly upregulated in HCC tissuescompared with paired adjacent nonneoplastic tissues Theincreased expression of both AQP3 and AQP5 proteins wassignificantly associated with aggressive clinicopathologicalfeatures of HCCWe observed the coexpression of AQP3 andAQP5 to be associated with tumor stage tumor grade tumormetastasis and patient prognosis Taken together our resultssuggest for the first time that the coexpression of AQP3 andAQP5 proteins may be a useful diagnostic and prognosticmarker in HCC patients


10

08

04

06

02

00

Low AQP3 expression

P = 0005

High AQP3 expression

0 12 24 36 48 60

Months after surgery

Five

-yea

r dise

ase-

free s

urvi

val

(a)

Low AQP3 expression

P = 0008


10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r ove

rall

surv

ival

(b)

Low AQP5 expression


P = 0002

10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(c)

Low AQP5 expression


10

08

04

06

02

00

0 12 24 36 48 60

P = 0006


Five

-yea

r ove

rall

surv

ival

(d)

P = 0002

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-low

AQP3-lowAQP5-high

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r dise

ase-

free s

urvi

val

(e)

P = 0005

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-high

AQP3-lowAQP5-low

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r ove

rall

surv

ival

(f)

Figure 2 Disease-free survival and overall survival curves for two groups defined by low and high expression of AQP3 ((a) and (b)) or AQP5((c) and (d)) and for four groups defined by combined expression of AQP3 and AQP5 ((e) and (f)) in patients with HCC The patients withhigh AQP3 and AQP5 expression had a significantly shorter 5-year overall and disease-free survival rate than those with lowAQP3 and AQP5expression (119875 = 0006 and 119875 = 001 resp) In addition the results by pairwise comparisons showed that the statistically significant differenceof overall and disease-free survival existed between AQP3-highAQP5-high patients and any of other three groups (119875 = 0002 and 0005resp) In all four groups AQP3-highAQP5-high patients had the poorest prognosis


Table 3 Multivariate survival analysis of five-year overall and disease-free survival in 130 patients with hepatocellular carcinoma

Features Five-year overall survival Five-year disease-free survivalHR 95 CI P HR 95 CI P

Age 1132 0316ndash3516 0192 1536 0322ndash3736 0125Gender 1191 0345ndash3857 0136 1559 0357ndash3831 0131Serum AFP 1931 0685ndash4056 0063 1953 0615ndash4273 0062Tumor stage 2879 1366ndash5196 0009 2686 1386ndash6009 001Tumor grade 1563 0609ndash4088 0081 1551 0607ndash4466 0086Presence of cirrhosis 1919 0738ndash4102 0063 1921 0793ndash4219 0062AQP3 expression 5398 1312ndash11338 001 5200 1343ndash11186 001AQP5 expression 8476 1993ndash17286 0006 5936 1312ndash12588 001AQP3AQP5 expression 6982 1601ndash15193 0009 5695 1381ndash11902 001

AQPs are water channel proteins that facilitate trans-cellular water movements [16] Accumulating evidence sug-gests that AQPs are involved in cell migration and prolifera-tion adding them to an expanding list of effectors in tumorpathologyThe aberrant expression of human AQPs has beenreported to be associated with various cancers In particularIshimoto et al [17] demonstrated that the overexpression ofboth AQP3 andAQP5was immunohistochemically observedon tumor cells in squamous cell carcinoma whereas ade-noid cystic carcinoma cells were faintly stained with thoseantibodies against AQPs Guo et al [13] indicated that theupregulation of AQP3 and AQP5 in lung cancer cells maybe mostly associated with cellular differentiation Based onRT-PCR analysis Guo et al [14] reported that AQP3 andAQP5 exhibited differential expression between human gas-tric carcinomas and corresponding normal tissues whichwasconfirmed byWestern blot analyses Guo et al [15] found thatthe expression of AQP5 was significantly decreased ovariancancer High expression level of AQP3 was also observedby Kusayama et al [18] in tumor areas of human primarysquamous cell carcinoma such as esophageal and lingualcancers and lymph node metastasis but it was not observedin normal areas In the present study our immunohistochem-istry analysis showed that AQP3 and AQP5 proteins wereboth upregulated in HCC tissues compared with the normalcontrolsThese findings revealed the altered expression AQPsin several types of tumors upon their specific expressionpatterns

In addition to the above expression studies there havebeen some studies on the role of AQPs in human carcino-genesis that have been alluded For example Sekine et al[19] found that the survival of biliary tract carcinoma patientswith high AQP5 expression was longer compared to thatof patients with low AQP5 expression Coxrsquos proportionalhazard model revealed that AQP5 expression was an inde-pendent prognostic factor and Chi-square analysis revealedthat high AQP5 expression correlated to small tumor size inbiliary tract carcinoma patients AQP5 expression in coloncancer cell lines and human colon cancer tissues may beassociated with cell proliferation and metastasis to liver [20]Zhang et al [21] showed that the high AQP5 protein expres-sion in intestinal type of adenocarcinoma was significantlyassociated with lymph node metastasis and lymphovascular

invasion in patients Watanabe et al [22] also found thatupregulation of AQP5 might be involved in differentiationof human gastric cancer cells Yang et al [23] also havepreviously found that AQP5 expression in ovarian malignantand borderline tumors was significantly higher than thatof benign ovarian tumors and normal ovarian tissue andthat the increased AQP5 protein level was associated withlymph node metastasis and ascites Li et al [24] reported thatAQP3 overexpression could facilitate colorectal carcinomacell migration and that AQP3 may be considered a potentialindicator and therapeutic target for colon tumor metastasisand prognosis Otto et al [25] indicated that loss of AQP3protein expression in pT1 bladder cancer may play a keyrole in disease progression and is associated with worseprogression-free survival In this study we also found thatthe overexpression of both AQP3 and AQP5 was associatedwith advanced tumor stage positive distant metastasis andunfavorable prognosis Notably patients with AQP3 overex-pression in combinationwithAQP5upregulation had aworseprognosis than all of the other patients From these resultswe suggest that AQP3 and AQP5 may serve as molecularprognostic markers for HCC and that AQP3 overexpressionin combination with AQP5 upregulation may be associatedwith even worse prognosis of HCC patients Overall theseresults indicated that AQP3 and AQP5 are involved in thedevelopment of several tumor types especially in HCCbut the two proteins function as tumor promoter or tumorsuppressor in different tumor types

In conclusion our data suggest for the first time thatthe aberrant expression of AQP3 and AQP5 proteins may bestrongly related to tumor progression and prognosis inpatients with HCC The overexpression of AQP3 in combi-nation with upregulation of AQP5 may be an unfavorableprognostic factor for HCC Although the role of AQP3 andAQP5 in human tumor pathology has been explored exten-sively their molecular mechanisms in different tumor typeshave not been fully elucidated Further studies are needed toinvestigate the precisemechanisms of AQP3 andAQP5 in theprogression of HCC

Conflict of Interests

The authors declare that they have no conflict of interests


Authorsrsquo Contribution

Xiaodong Guo Ting Sun and Mei Yang contributed equallyto this study

References

[1] M Dhir E R Lyden L M Smith and C Are ldquoComparisonof outcomes of transplantation and resection in patients withearly hepatocellular carcinoma a meta-analysisrdquo HPB vol 14pp 635ndash645 2012

[2] A X Zhu ldquoMolecularly targeted therapy for advanced hepato-cellular carcinoma in 2012 current status and future perspec-tivesrdquo Seminars in Oncology vol 39 pp 493ndash502 2012

[3] S Tanaka and S Arii ldquoMolecular targeted therapies in hepato-cellular carcinomardquo Seminars in Oncology vol 39 pp 486ndash4922012

[4] V J Huber M Tsujita and T Nakada ldquoAquaporins in drug dis-covery and pharmacotherapyrdquo Molecular Aspects of Medicinevol 33 pp 691ndash703 2012

[5] A S Verkman ldquoAquaporins in clinical medicinerdquo AnnualReview of Medicine vol 63 pp 303ndash316 2012

[6] A S Verkman ldquoAquaporins at a glancerdquo Journal of Cell Sciencevol 124 no 13 pp 2107ndash2112 2011

[7] P R Mazal M Susani F Wrba and A Haitel ldquoDiagnosticsignificance of aquaporin-1 in liver tumorsrdquo Human Pathologyvol 36 no 11 pp 1226ndash1231 2005

[8] S Padma A M Smeltz P M Banks D A Iannitti and I HMcKillop ldquoAltered aquaporin 9 expression and localization inhuman hepatocellular carcinomardquo HPB vol 11 no 1 pp 66ndash74 2009

[9] E M Jablonski M A Mattocks E Sokolov et al ldquoDecreasedaquaporin expression leads to increased resistance to apoptosisin hepatocellular carcinomardquo Cancer Letters vol 250 no 1 pp36ndash46 2007

[10] M Hara-Chikuma and A S Verkman ldquoAquaporin-3 functionsas a glycerol transporter inmammalian skinrdquoBiology of the Cellvol 97 no 7 pp 479ndash486 2005

[11] R Inoue E Sohara T Rai et al ldquoImmunolocalization andtranslocation of aquaporin-5 water channel in sweat glandsrdquoJournal of Dermatological Science vol 70 pp 26ndash33 2013

[12] T Matsuzaki T Susa K Shimizu et al ldquoFunction of themembrane water channel aquaporin-5 in the salivary glandrdquoActa Histochemica et Cytochemica vol 45 pp 251ndash259 2012

[13] X Guo L Xiong L Zou and J Zhao ldquoUpregulation of bonemorphogenetic protein 4 is associated with poor prognosis inpatients with hepatocellular carcinomardquo Pathology and Oncol-ogy Research vol 18 pp 635ndash640 2012

[14] X Guo L Xiong T Sun et al ldquoExpression features of SOX9associate with tumor progression and poor prognosis of hep-atocellular carcinomardquo Diagnostic Pathology vol 7 article 442012

[15] X Guo L Xiong L Zou et al ldquoL1 cell adhesion moleculeoverexpression in hepatocellular carcinoma associates withadvanced tumor progression and poor patient survivalrdquo Diag-nostic Pathology vol 7 article 96 2012

[16] C Hachez and F Chaumont ldquoAquaporins a family of highlyregulated multifunctional channelsrdquo Advances in ExperimentalMedicine and Biology vol 679 pp 1ndash17 2010

[17] S Ishimoto KWada Y Usami et al ldquoDifferential expression ofaquaporin 5 and aquaporin 3 in squamous cell carcinoma and

adenoid cystic carcinomardquo International Journal of Oncologyvol 41 pp 67ndash75 2012

[18] M Kusayama K Wada M Nagata et al ldquoCritical role ofaquaporin 3 on growth of human esophageal and oral squamouscell carcinomardquo Cancer Science vol 102 no 6 pp 1128ndash11362011

[19] S Sekine Y Shimada T Nagata et al ldquoPrognostic significanceof aquaporins in human biliary tract carcinomardquo OncologyReports vol 27 pp 1741ndash1747 2012

[20] K K Sung K C Young JWoo et al ldquoRole of human aquaporin5 in colorectal carcinogenesisrdquo American Journal of Pathologyvol 173 no 2 pp 518ndash525 2008

[21] Z-Q Zhang Z-X Zhu C-X Bai and Z-H Chen ldquoAquaporin5 expression increases mucin production in lung adenocarci-nomardquo Oncology Reports vol 25 no 6 pp 1645ndash1650 2011

[22] T Watanabe T Fujii T Oya et al ldquoInvolvement of aquaporin-5 in differentiation of human gastric cancer cellsrdquo Journal ofPhysiological Sciences vol 59 no 2 pp 113ndash122 2009

[23] J-H Yang Y-F Shi Q Cheng and L Deng ldquoExpression andlocalization of aquaporin-5 in the epithelial ovarian tumorsrdquoGynecologic Oncology vol 100 no 2 pp 294ndash299 2006

[24] A Li D Lu Y Zhang et al ldquoCritical role of aquaporin-3in epidermal growth factor-induced migration of colorectalcarcinoma cells and its clinical significancerdquo Oncology Reportsvol 29 pp 535ndash540 2013

[25] W Otto P C Rubenwolf M Burger et al ldquoLoss of aquaporin 3protein expression constitutes an independent prognostic factorfor progression-free survival an immunohistochemical studyon stage pT1 urothelial bladder cancerrdquo BMC Cancer vol 12article 459 2012

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


suggests the diagnostic and prognostic value of the aberrantexpression of AQPs AQP expression in some reports is asso-ciated with tumor progression In liver tumors Mazal et al[7] demonstrated that AQP1 expression in HCC tissues waslower than that in cholangiocarcinoma suggesting that AQP1might be a highly selective marker for differentiated cholan-giocytes and can be very helpful in the differential diagnosisof liver tumors Padma et al [8] reported that human HCCmay be characterized by altered AQP9 expression and thatAQP9 localization in the nontumourigenic liver mass maybe dependent on underlying liver pathology Jablonski et al[9] found that AQP8 expression and AQP9 expression weresignificantly decreased in HCC versus normal liver Thesefindings implicated that the AQPs might play an importantrole in human HCC

Of 13AQP varieties AQP1 -8 and -9were downregulatedin human HCC tissues [7ndash9] The functional role of otherAQPmembers in HCC has not been fully elucidated Amongthem AQP3 cloned in 1994 functions as a membranechannel of water and other small solutes such as glyceroland urea and plays a major role in fluid homeostasis [10]AQP5 cloned from the salivary gland is a 27-kDa proteinthat is known as an exocrinetype water channel with a uniquetissue expression [11] AQP3 that transports not only waterbut also glycerol and urea is known to be expressed in kidneyskin lung and gastrointestinal tracts [10] and AQP5 thattransports only water is known to be expressed in variousorgans such as lung and saliva gland [12] Recent studies haveshowed the altered expression of AQP3 and AQP5 in varioustumor types However little is known about expression andprecise role of the two proteins on HCC Therefore the aimof this study was to investigate the expression patterns andclinical significance of AQP3 and AQP5 in human HCC

2 Materials and Methods

21 Patients and Tissue Samples The study was approved bythe Research Ethics Committee of 302nd Hospital of PLABeijing China Informed consentwas obtained fromall of thepatients All specimens were handled and made anonymousaccording to the ethical and legal standards

A total of 130 patients with primaryHCCwho underwenta curative liver resection at the 302nd Hospital of PLABeijing China were included in this retrospective studyTissues used in the study were retrieved from the tissue bankof the Department of Pathology at the 302nd Hospital ofPLA These patients were diagnosed as HCC between 2001and 2006 None of the patients recruited in this study hadchemotherapy or radiotherapy before the surgery HCC diag-nosis was based on the World Health Organization (WHO)criteria Tumor differentiation was defined according to theEdmondson grading system Liver function was assessedusing the Child-Pugh scoring system Tumor staging wasdetermined according to the sixth edition of the tumor-node-metastasis (TNM) classification of the InternationalUnion against CancerThe clinicopathological features of 130patients are summarized in Table 1

The median follow-up period was 86 years Postoper-ative surveillance included routine clinical and laboratory

examinations every third month computed tomographyscans of the abdomen and radiographs of the chest everythird month After 5 years the examination interval wasextended to 12 months

22 Immunohistochemistry Analysis AQP3 expression andAQP5 expression were immunohistochemically evaluated inparaffin-embedded specimens of 130 patients with HCCSurgical specimens were fixed in 10 formalin embed-ded in paraffin and sectioned at a 4120583m thickness Forheat-induced epitope retrieval deparaffinized sections weresoaked in 10mM citrate buffer (pH 60) and treated at 95∘Cfor 30min using the microwave oven method Immuno-histochemical staining was performed using the avidin-biotin immunoperoxidase technique according to our pre-vious studies [13ndash15] The activity of endogenous peroxidasewas blocked by incubation with 03 H

2O2in methanol

for 15min and nonspecific immunoglobulin binding wasblocked by incubation with 10 normal goat serum for10min Sections were incubated at room temperature for 4 hwith anti-AQP3 rabbit polyclonal antibody (sc-20811 SantaCruz Biotechnology Inc USA) or with anti-AQP5 rabbitpolyclonal antibody (sc-28628 Santa Cruz BiotechnologyInc USA) at a 1 100 or 1 150 dilution and they werethen rinsed and incubated for 30min with a biotinylatedsecond antibody After washing the sections were incubatedfor 30min with horseradish peroxidase-conjugated strepta-vidin and were finally treated with 331015840-diaminobenzidinetetrahydrochloride in 001 H

2O2for 10min The slides

were counterstained with Meyerrsquos hematoxylin The negativecontrols were processed in a similarmanner with PBS insteadof primary antibody The positive AQP3 expression andAQP5 expression confirmed by western blotting were usedas positive controls for immunostaining

Following a hematoxylin counterstaining immunostain-ing was scored by two independent experienced pathologistswho were blinded to the clinicopathological parameters andclinical outcomes of the patients The scores of the twopathologists were compared and any discrepant scores weretrained through reexamining the staining by both patholo-gists to achieve a consensus score The number of positive-staining cells showing immunoreactivity in the cytoplasmand cell membrane for both AQP3 and AQP5 in ten repre-sentative microscopic fields was counted and the percentageof positive cells was calculated The percentage scoring ofimmunoreactive tumor cells was as follows 0 (0) 1 (1ndash10) 2 (11ndash50) and 3 (gt50) The staining intensity wasvisually scored and stratified as follows 0 (negative) 1 (weak)2 (moderate) and 3 (strong) A final score was obtained foreach case by multiplying the percentage and the intensityscore Therefore tumors with a multiplied score exceedingmedian of total scores for AQP3 or AQP5 were deemed tobe low expressions of AQP3 or AQP5 all other scores wereconsidered to be high expressions of AQP3 or AQP5

23 Statistical Analysis The software of SPSS version 130 forWindows (SPSS Inc IL USA) and SAS 91 (SAS InstituteCary NC) was used for statistical analysis The chi-squaredtest was used to show differences in categorical variables







0005

0 (0)

0008

0 (0)

0006T2 40 21 (5250) 17 (4250) 10 (2500)T3 52 46 (8846) 40 (7692) 27 (5192)T4 15 15 (10000) 15 (10000) 15 (10000)


NS13 (4194)

00093 (968)

0006G2 76 49 (6447) 40 (5263) 29 (3816)G3 23 13 (5652) 20 (8696) 20 (8696)




NS

15 (6000)

NS

7 (280)




3 Results





(b)(a)



AQP5 expressionP

High (119899 = 82) Low (119899 = 48)






4 Discussion



10

08

04

06

02

00

Low AQP3 expression

P = 0005


0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(a)

Low AQP3 expression

P = 0008


10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r ove

rall

surv

ival

(b)

Low AQP5 expression


P = 0002

10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(c)

Low AQP5 expression


10

08

04

06

02

00

0 12 24 36 48 60

P = 0006


Five

-yea

r ove

rall

surv

ival

(d)

P = 0002

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-low

AQP3-lowAQP5-high

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r dise

ase-

free s

urvi

val

(e)

P = 0005

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-high

AQP3-lowAQP5-low

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r ove

rall

surv

ival

(f)















References
































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of






















0005

0 (0)

0008

0 (0)

0006T2 40 21 (5250) 17 (4250) 10 (2500)T3 52 46 (8846) 40 (7692) 27 (5192)T4 15 15 (10000) 15 (10000) 15 (10000)


NS13 (4194)

00093 (968)

0006G2 76 49 (6447) 40 (5263) 29 (3816)G3 23 13 (5652) 20 (8696) 20 (8696)




NS

15 (6000)

NS

7 (280)




3 Results





(b)(a)



AQP5 expressionP

High (119899 = 82) Low (119899 = 48)






4 Discussion



10

08

04

06

02

00

Low AQP3 expression

P = 0005


0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(a)

Low AQP3 expression

P = 0008


10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r ove

rall

surv

ival

(b)

Low AQP5 expression


P = 0002

10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(c)

Low AQP5 expression


10

08

04

06

02

00

0 12 24 36 48 60

P = 0006


Five

-yea

r ove

rall

surv

ival

(d)

P = 0002

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-low

AQP3-lowAQP5-high

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r dise

ase-

free s

urvi

val

(e)

P = 0005

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-high

AQP3-lowAQP5-low

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r ove

rall

surv

ival

(f)















References
































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















(b)(a)



AQP5 expressionP

High (119899 = 82) Low (119899 = 48)






4 Discussion



10

08

04

06

02

00

Low AQP3 expression

P = 0005


0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(a)

Low AQP3 expression

P = 0008


10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r ove

rall

surv

ival

(b)

Low AQP5 expression


P = 0002

10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(c)

Low AQP5 expression


10

08

04

06

02

00

0 12 24 36 48 60

P = 0006


Five

-yea

r ove

rall

surv

ival

(d)

P = 0002

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-low

AQP3-lowAQP5-high

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r dise

ase-

free s

urvi

val

(e)

P = 0005

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-high

AQP3-lowAQP5-low

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r ove

rall

surv

ival

(f)















References
































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















10

08

04

06

02

00

Low AQP3 expression

P = 0005


0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(a)

Low AQP3 expression

P = 0008


10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r ove

rall

surv

ival

(b)

Low AQP5 expression


P = 0002

10

08

04

06

02

00

0 12 24 36 48 60


Five

-yea

r dise

ase-

free s

urvi

val

(c)

Low AQP5 expression


10

08

04

06

02

00

0 12 24 36 48 60

P = 0006


Five

-yea

r ove

rall

surv

ival

(d)

P = 0002

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-low

AQP3-lowAQP5-high

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r dise

ase-

free s

urvi

val

(e)

P = 0005

10

08

04

06

02

00

0 12 24 36 48 60

AQP3-lowAQP5-high

AQP3-lowAQP5-low

AQP3-highAQP5-high

AQP3-highAQP5-low


Five

-yea

r ove

rall

surv

ival

(f)















References
































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





























References
































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





















of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















research article prognostic value of combined aquaporin 3...

Documents