ORIGINAL PAPER

Joo-Young Kim Æ Hye-Jin Shin Æ Tae-Hyun Kim

Kwan-Ho Cho Æ Kyung-Hwan Shin Æ Bu-Kyoung Kim

Ju-Won Roh Æ Sun Lee Æ Sang-Yoon Park

You-Jin Hwang Æ Inn-Oc Han

Tumor-associated carbonic anhydrases are linked to metastasesin primary cervical cancer

Received: 12 October 2005 / Accepted: 30 November 2005 / Published online: 14 January 2006� Springer-Verlag 2006

Abstract Purpose: Carbonic anhydrase IX (CA9) hasemerged as an important surrogate marker for hypoxiain solid tumors. CA12 shares a role with CA9 in acidi-fication of micromilieu but it is less strictly regulated byhypoxia than CA9. In this study, we investigatedexpression of CA9 and CA12 mRNA in primary cervicalcancer. We also examined whether CA9 expression canbe an indicator of reoxygenation of tumor by measuringits mRNA expression during fractionated radiotherapy.Methods: Tumor tissues were obtained from 59 patientswith uterine cervical cancer who underwent radiother-apy, and a second biopsy was taken after patients hadreceived either 10 or 20 Gy of radiation. The follow-upperiod ranged from 2.4 to 75 months (med-ian=23 months). The ratio of CA9 and b-actin mRNAexpression was determined both pre- and during radia-tion treatment by RT-PCR. Results: CA9 and CA12mRNA expression was detected in 62.7 and 88.1% oftumors (i.e. patients), respectively, and co-expressionwas observed in 61% of patients. Multivariate analysisrevealed that CA9 expression was the most significantfactor associated with metastasis-free survival(P=0.008, hazard ratio 34.8), whereas CA12 mRNAexpression was linked to a lower risk of metastasis

(P=0.007, hazard ratio of 0.07). Tumor CA9 expressionwas not altered following either 10 or 20 Gy of radio-therapy. Conclusion: The strong correlation betweenCA9 expression and metastasis suggests that CA9expression might be an important indicator for identi-fying patients who require more aggressive systemictherapy.

Keywords Hypoxia Æ Carbonic anhydrase IX (CA9) ÆXII (CA12) Æ Metastasis Æ Radiotherapy

Introduction

The presence of hypoxic conditions within solid tumorsis one of the most important factors affecting treatmentoutcome. As hypoxia-regulated molecular pathwayshave been unveiled in recent years, targeting hypoxia isincreasingly considered a sound and important strategyin treating cancer. Various experimental and clinicalstudies have shown that tumor cells under hypoxicconditions are resistant to radiation-induced cell killingcompared to cells in well-oxygenated conditions. Earlyanimal experiments showed that as the reoxygenationprocess occurs, the hypoxic proportion of a tumor canremain constant or even fall after radiation (Hall 2000;Van Putten and Kallman 1968). In clinical radiotherapypractice, it has been postulated that many types of tu-mors can be eradicated by step-wise killing of thereoxygenated proportion of a tumor using a fractionatedradiotherapy regimen over 6 or more weeks. However,reoxygenation has not been verified to occur in clinicaltumor samples.

Following sporadic efforts to find useful hypoxiamarkers, a recent study has identified 12 hypoxia-over-expressed genes (HOG) using serial analysis of geneexpression in response to hypoxic stimuli (Lal et al.2001). Among these genes, the isoenzymes of carbonicanhydrase (CA) families, CA9 and CA12, are shown asbeing induced under hypoxic conditions in many cancer

J.-Y. Kim and H.-J. Shin contributed equally to this work.

J.-Y. Kim Æ H.-J. Shin Æ T.-H. Kim Æ K.-H. Cho Æ K.-H. ShinB.-K. Kim Æ J.-W. Roh Æ S. Lee Æ S.-Y. Park Æ I.-O. Han (&)Research Institute, National Cancer Center,Gyeonggi, Goyang, KoreaE-mail: iohan@inha.ac.krTel.: +82-31-9201724Fax: +82-31-9200149

Y.-J. HwangDivision of Biological Science, Gachon Medical School,Inchon, Korea

I.-O. HanDepartment of Physiology and Biophysics, Inha University,College of Medicine, Incheon, Korea

J Cancer Res Clin Oncol (2006) 132: 302–308DOI 10.1007/s00432-005-0068-2

cell lines. CA9 protein is expressed in the perinecroticregion of solid tumors and co-localizes with pimoni-dazole staining, a well-known hypoxia indicator (Vor-dermark and Brown 2003). In addition, CA9 expressionis more specifically and rapidly regulated by oxygenconcentration compared to other HOG such as glucosetransporter-1 (GLUT-1) and -3 and vascular endothelialgrowth factor (VEGF), the expression of which isheavily regulated by a number of other signaling path-ways (Harris 2002; Lal et al. 2001; Semenza 2003). Al-though both CA9 and CA12 are downregulated byVHL, the transcription of which are considered to beunder control of HIF-1a, the expression and the tissuedistribution of CA12 gene are not associated with theexpression of CA9 (Ivanov et al. 1998). Unlike to CA9,which is usually found in tumor tissues, CA12 is dis-tributed in a variety of normal tissues and its expressionbecomes stronger in tumors originating from the sametissues (Ivanov et al. 2001).

Identifying CA9 expression might be particularlymeaningful in cervical cancer due to the possible linkbetween CA9 induction and human papilloma virus(HPV) infection. CA9 was cloned from HeLa cells in1994 and has been known to be the only tumor-associ-ated member of the CA family, which catalyzes thecarbon dioxide to carbonic acid conversion (Opavskyet al. 1996). CA9 is subsequently shown to be involved incell–cell and cell–matrix interactions that affect loss ofcontact inhibition and anchorage-independent cellgrowth in MN/CA9 gene-transfected C33A and SiHacervical cancer cell lines (Lieskovska et al. 1999). Inaddition, CA9 is considered as a biomarker of HPVinfection, along with P16 and cyclin E (Keating et al.2001; Resnick et al. 1996), and these molecules can beused as novel indicators for screening and diagnosis ofHPV infection.

In the present study, we examined whether expressionof CA9 and CA12 were prognostic indicators for pa-tients with uterine cervical cancer. As a potential mea-sure of reoxygenation, we determined the kinetics ofCA9 expression in individual tumors during fractionatedradiotherapy. The results in this paper demonstratedthat CA9 expression is linked to distant metastasis inpatients with uterine cervical cancer. The data furthersuggest that patients with high CA9 expression poten-tially have a considerable hypoxic tumor burden and areprone to fail at distant sites and hence should beadministered more aggressive systemic treatments.

Methods and materials

Patients and clinical specimen

Pre-treatment cervical cancer tissues were obtained fromthe 59 patients who underwent radiotherapy as theirmain treatment modality. Our study was approved bythe Institutional Review Board on the experimentalstudies and informed consent was obtained from each

patient or a responsible relative accordingly. All patientshad newly diagnosed, histologically proven squamouscell carcinoma or adenocarcinoma of the cervix. In thesepatients, 12 and 32 patients undertook an additionalbiopsy at a cumulative radiation dose of 10 and 20 Gy,respectively. We used 3–4 mm width tumor tissue andobserved the frozen cut of each tissue under microscopebefore RNA is extracted to minimize the possibility ofnormal tissue contamination. This second biopsy wasobtained to detect any change in hypoxia-related genesduring fractionated radiotherapy. Multiple biopsies weretaken avoiding the grossly necrotic portion and the tu-mor tissues thus obtained were fixed in the 4% neutralformalin and were snap-frozen in liquid nitrogen.Chemical coagulation was avoided as much as possiblefor later biopsy procedures. All tumor sections were H-Estained and were subsequently examined for the pres-ence of tumor cells by a pathologist.

RNA preparation

About 3–4 mm length of the tumor tissues were takenfrom the biopsy specimens and were kept at �80�C inTrizol 500 ll until it was further processed. For RNApreparation, samples were chopped and minced, fol-lowed by addition of 200 ll of chloroform. The sampleswere incubated in ice for 5 min and then centrifuged toseparate the aqueous phase. The upper phase was care-fully removed and RNA was precipitated by adding 1 mlof isopropanol followed by centrifugation. The RNApellet thus obtained was rinsed with 70% alcohol inDEPC water. Finally, the pellet was air-dried and wasthen solubilized with RNase-free water. RNA concen-tration was measured by spectrophotometry with opticaldensity read at 260 nm.

RT-PCR of the CA9, CA12, and VEGF genes

Total RNA (2 lg) was reverse transcribed for 1 h at37�C in a reaction mixture containing 5 U RNase(Amersham, Piscataway, NJ, USA), 0.5 mM dNTP(Boeringer Mannheim, Indianapolis, IN, USA), 2 lMrandom hexamer (Stratagene, La Jolla, CA, USA), RTbuffer, and 5 U reverse transcriptase (Quiagen). Thefirst-strand cDNA synthesis was performed with theoligo (dT) primed first-strand synthesis in a total volumeof 10 ll with a reverse transcriptase. The product of theRT-PCR 1 ll was mixed with 10· buffer (MgCl2), 3 llof 2.5 mM dNTP, 1 ll of 10 pmol forward and reverseprimer, 1 ll of Taq polymerase, and finally RNase andDNase-free water 20 ll was mixed to make a total vol-ume of 30 ll. The reaction mixture was subjected toPCR at 94�C 5 min, 94�C 30 s, 62�C 30 s (57�C for CA12 and 55�C for VEGF), 72�C 30 s for 30 cycles, and72�C 7 min. The primers were CA9, forward, 5¢-TA-AGCAGCTCCACACCCTCT-3¢, and reverse, 5¢-TCTCATCTGCACAAGGAACG-3¢; CA12, forward,

303

5¢-ATGGCAGGTTCAAGTTCCAC-3¢, and reverse, 5¢-TCGGAACTCATGTCTCCTCC-3¢; VEGF, forward,5¢-GTGGACATCTTCCAGGAGTA-3¢ and reverse, 5¢-ATCTGCAAGTACGTTCGTTT-3¢) and GAPDH,forward, 5¢-ATGTACGTAGCCATCCATCCAGGC-3¢, reverse, 5¢-AGGAAGGAAGGCTGGAAGAG-3¢.Analysis of the resulting PCR products on 1% agarosegels showed single-band amplification products withexpected sizes.

Real-time PCR of CA9 from the serially obtained tumortissues

CA9 DNA amplification was carried out using CA9cDNA template obtained from the mRNA producedfrom the RT-PCR procedure. Real-time PCR reactionswere set up in a reaction volume of 15 ll using theTaqMan Universal PCR Master Mix (PE Applied Bio-systems, CA, USA). A reporter dye FAM (6-carboxy-fluorescein) was covalently attached to the 5¢ end and aquencher dye TAMRA (6-carboxy-tetramethyl-rhoda-mine) was incorporated into the 3¢ end of TaqManprobe with sequence of TGAACTTCCGA GCGACG-CAGCC for CA9. PCR primers of CA9 was 5¢-AC-CTGGTGACTCTC GGCTACAG-3¢as a forwardprimer and 5¢-TTTGAATGGGCGAGTGATTG-3¢as areverse primer. As an endogenous control, real-timePCR analysis was performed on the b-actin gene in thesame cDNA samples for relative gene expression quan-tification. The b-actin gene primers and probe labeledwith VICTM dye and MGB/non-fluorescent quencherwere provided by Applied Biosystems. DNA amplifica-tion was done in a 96-well reaction plate format in anABI PRISM� 7900HT Sequence Detection system (PEApplied Biosystems). Both CA9 and b-actin PCR reac-tions were carried out in triplicate. Multiple negativewater blanks were included in every analysis. Relativestandard curves were established from the control DNAextracted from HeLa cells (1 lg cDNA/ll). The controlcDNA was provided as 1 lg/ul and was diluted from1,000 to 0.01 ng/ll. Two-hundred ng/ll of samplecDNA was loaded and the reactions were run on ABIPRISM� 7900HT Sequence Detection system pro-grammed to 5-min initial step at 95�C followed by 40cycles of 15 s at 95�C and 60 s at 60�C. The mean Ctvalues and quantities in log linear range were used forquantification of CA9 gene and control gene.

Statistical method

Overall survival, recurrence-free survival, and metasta-sis-free survival were analyzed by Kaplan–Meier meth-od. Univariate analysis was made on the relationship ofCA9, CA12, VEGF, and other individual variables withrespect to the above survival parameters using the log-rank test. Multivariate analysis for the same variableswas also undertaken with the Cox model using the

general strategy for model selection involving the hier-archic principle with the likelihood ratio chi-square test.The correlations of the expression of CA9, CA12, andVEGF were made by Spearmann rank correlation test.Pre-treatment and post-treatment CA9 expression wascompared by paired T-test. All statistical analyses wereperformed with SAS programs (version 8.01, SASInstitute Inc., Cary, NC, USA).

Results

Association between CA9, CA12, and VEGF geneexpression and clinical outcome

The 5-year overall survival, local recurrence-free sur-vival, and metastasis-free survival rates for all patientswere 73.4, 69, and 82%, respectively. Distant metastasisdeveloped in 14 of 59 patients (23.7%). Using RT-PCR,biopsy tissues from individual tumors were measured forCA9, CA12 and VEGF mRNA expression, and wereanalyzed in terms of clinical outcomes. CA9 and CA12mRNA was detected in 63% (37/59) and 88% (52/59) oftumors (i.e. patients), respectively, and VEGF expres-sion was detected in 58% (54/59) of tumors. CA9expression was closely correlated to that of CA12 andVEGF (r=0.75, P<0.001). Analysis of clinical out-comes showed that overall survival and local recurrencewere not linked to mRNA expression of any of the threegenes. Univariate analysis showed tumor size and clini-cal stage were significant factors related to overall sur-vival of patients (Table. 1, 2, 3), while multivariateanalysis showed tumor size was the only significantfactor influencing local control of the disease (P=0.008,hazard ratio 3). In addition, multivariate analysisshowed that CA9 mRNA expression was the single mostimportant factor associated with the occurrence of dis-tant metastasis (P=0.008, hazard ratio of 34.8; Fig. 1a).In remarkable contrast, CA12 mRNA expression wasassociated with less likelihood of metastasis (P=0.007,hazard ratio of 0.07; Figs. 1b). VEGF mRNA expres-sion was not related to any clinical end-point.

The effect of radiotherapy on CA9 gene expression

Using RT-PCR and real-time quantitative PCR, CA9mRNA expression in tumor samples was measuredquantitatively both prior to radiotherapy and afterreceiving either 10 or 20 Gy radiation (two separatepatient groups). We found that after receiving 10 Gytreatment there was no difference in CA9 mRNA levelsin biopsy samples compared to levels in tissue takenprior to treatment. Although there was considerabledecrease in tumor CA9 mRNA expression following20 Gy irradiation, this change was similar to that ob-served in expression of the comparison control gene b-actin. The mean ratios of CA9 to b-actin mRNAexpression were 1.18 at pre-treatment and 1.03 at the

304

completion of 20 Gy fractionated radiation therapy(P=0.22). The corresponding values for 10 Gy were 1.1and 1.0, respectively (Fig. 2).

Discussion

The aims of this study were to investigate whetherexpression of CA9 and/or CA12 in tumors was a prog-nostic indicator in primary cervical cancer, and whether

CA9 could be used as a molecular marker for tissuereoxygenation. The study undertook a retrospectiveanalysis of 59 patients with cervical carcinoma andanalyzed CA9, CA12, and VEGF mRNA expression byRT-PCR in both pre- and post-irradiated tumor biopsytissues. Expression of VEGF, a powerful mitogen forendothelial cell proliferation, is induced by hypoxia andwas used as a comparison. Among a variety of param-eters examined, we found that CA9 was the most sig-nificant factor predicting development of distantmetastases. Studies on the clinical implication of CA9expression in various tumor tissues show a variety ofrelationships with different clinical end-points (Vorder-mark and Brown 2003). A study similar to the presentone examined CA9 expression in uterine cervical cancerand reported results consistent with those reported herein that positive CA9 immunohistochemical staining isstrongly associated with disease-free and metastasis-freesurvival (Loncaster et al. 2001).

The reasons why CA9 expression affects uterine cer-vical cancermetastasis are currently unknown.Olive et al.(2001) showed a direct relationship between CA9expression and the hypoxic status of cells using in vitroand in vivo experimental system, and that cells with highCA9 expression are radioresistant. If the high CA9expression mediates the radioresistance of tumor cells bya certain mechanism, it might be expected to predict localrecurrence. However, our results suggest that CA9expression is strongly related to the enhanced metastaticpotential of hypoxic tumors and not to the local control ofthese tumors by radiotherapy. It is highly likely that CA9expression may have an important function in modulat-ing tumor invasion and metastasis. The role of CA9 inacidifying extracellular matrices can be considered as areason why CA9-expressing tumor cells are more meta-

Table 2 Univariate analysis ofclinical outcome with variables

OS overall survival, RFS recu-rrence-free survival, MFS met-astasis-free survival

Factors OS RFS MFS

5-year OS P value 5-year RFS P value 5-year MFS P value

Age£ 60 63 0.25 64 0.40 76 0.76>60 79 70 75Tumor size>4 cm 83 0.04 57 0.02 60 0.15£ 4 cm 62 80 86

StageI 100 0.02 100 0.003 100 0.03IIA 61 82 58IIB 84 60 88III 25 44 83IV 0 0 0CA IX� 74 0.83 66 0.98 90 0.07+ 72 70 65CA XII� 69 0.82 100 0.13 57 0.20+ 73 63 78VEGF� 74 0.99 66 0.89 82 0.40+ 70 71 67

Table 1 Patients characteristics (N=59)

Characteristics Number (%)

Age£ 60 23 (39)>60 36 (61)Tumor size

‡4 cm 30 (49)<4 cm 29 (51)StageI 6 (10)IIA 16 (27)IIB 25 (42)III 6 (10)IV 6 (10)CA IX� 22 (37)+ 37 (63)CAXII� 7 (12)+ 52 (88)VEGF� 5 (8.3)+ 54 (90)CA IX and VEGF++ 33 (56)Others 26 (34)

305

static; a low pHmicroenvironment assists in extracellularmatrix degradation, which increases the likelihood ofblood vessel invasion. Identification of the exact mecha-nisms underlying metastases associated with CA9-over-expressing cervical cancers awaits further studies.

Intriguingly, the present study showed that CA12 andCA9 expression had very different associations withmetastasis. It has been reported that CA12 expression islinked to more differentiated, less invasive status of thetumor cells of the human breast cancer tissues (Wykoffet al. 2001). Besides, CA12 expression became strongerwith increasing age of mouse embryos, indicatingdevelopmental regulation of this protein (Halmi et al.2004). Usually, CA12 is distributed in a variety of nor-mal tissues and its expression becomes stronger in tu-mors originating from those same tissues (Ivanov et al.2001). It is likely that CA12 has an unknown role intumor progression and differentiation in its own right,which is not related to hypoxia-associated pathways.Despite the above evidences, which support our findingson the inverse correlation of CA12 expression and tumormetastasis, our data need to be interpreted with cautionwhen considering the relatively small number of patientswith CA12 negative tumors in the present study.

Another important aim of this study was to define theCA9 mRNA levels within the tumor during radiother-apy. We have observed that CA9 and CA12 mRNA

began to be expressed once oxygen concentrations fellbelow 5% O2 in the cell culture system (data not shown).Five percent oxygen level is often found in normal tis-sues with low oxygen pressure as well as in intermedi-ately hypoxic tumor tissues. Hence, CA9 mRNAexpression can represent a broad range of hypoxia.Radiobiological modeling shows that the presence ofcells exposed to the intermediate hypoxia between 0.5and 20 mmHg O2 can be very important in low dose(1.8–2 Gy) fractionated radiotherapy as opposed to alarge single dose of radiation (Wouters and Brown1997). By using RT-PCR rather than immunohisto-chemistry for CA expression, we expected to measure itsoverall level, rather than the hypoxic subvolume withinthe tumors. It has been hypothesized that the hypoxicfraction of tumors remains constant, or even decreases,during fractionated radiotherapy, with each radiother-apy fraction killing only the well-oxygenized proportionof tumor cells (Hall 2000; Van Putten and Kallman1968). The present study shows that while CA9 expres-sion decreased significantly during radiotherapy, thisreduction was the same as that for the b-actin controlgene. The reduction in CA9 expression appeared to bemostly due to the reduced tumor cell burden followingradiation. Comparison between the expression levels inbiopsies taken at two different times is not likely torepresent the whole tumor due to inherent limitation

BA

Met

asta

sis-

Fre

e S

urvi

val

0.0

0.2

0.4

0.6

0.8

1.0

1 2 3 4 5 6 7Years

5 Yr 82% (n=37)CA9 (-)

0.0

0.2

0.4

0.6

0.8

1.0

1 2 3 4 5 6 7

5 Yr 83% (n=52)CA12 (+)

Years

Met

asta

sis-

Fre

e S

urvi

val

CA9 (+) 5 Yr 67% (n=21)

CA12 (-) 5 Yr 57% (n=7)

Fig. 1 CA9 and CA12 mRNA expression was examined by RT-PCR in pre-treatment tumor specimens of 59 patients and the resultwas analyzed with respect to the survival parameters. AMetastasis-free survival of the patients with positive CA9 expression (5-year

rate 67%, n=37) versus negative CA9 expression (5-year rate 82%,n=21). B Metastasis-free survival of the patients with positiveCA12 expression (5-year rate 83%, n=52) versus negative CA12expression (5-year rate 57%, n=7)

Table 3 Multivariate analysis of clinical outcome with various parameters

Factors Overall survival Recurrence-free survival Metastasis-free survival

P value Hazard ratio P value Hazard ratio P value Hazard ratio

Age ( £ 60 vs. >60) 0.54 0.70 0.10 0.10 0.85 0.89Tumor size ( £ 4 cm vs. >4 cm) 0.18 2.45 0.08 3.14 0.25 2.31Stage (I–IV) 0.29 1.40 0.11 1.64 0.46 1.3CA IX (� vs. +) 0.98 1.52 0.74 0.78 0.008 34.80CA XII (� vs. +) 0.47 2.84 0.99 1.00 0.007 0.07VEGF (� vs. +) 0.44 0.56 0.49 0.60 0.28 0.41

306

using human tissue samples. An examination of acomprehensive set of HOGs might be required to ex-plore the changes in hypoxia levels in human tumors.Furthermore, analysis of heavily irradiated human tu-mor specimens can be affected by the number of residualtumor cells as well as the changes in structure andintegrity of subcellular macromolecules after irradiation.In our experiments, many biopsies collected following20 Gy treatment had to be excluded from analysis asthey did not contain enough tumor cells or did not allowfor the extraction of enough RNA of good quality.

In conclusion, we examined the expression of CA9 andCA12 in human cervical tumors. Measuring CAs mRNAexpressions using RT-PCR was sensitive, reproducibleand simple, and yielded significant prognostic informa-tion. CA9 expression was strongly associated with thedevelopment of distant metastases, while CA12 expres-sion was associated with less such metastases. A largerstudy would probably find that CA9 expression is linkedto overall survival in uterine cervical cancer patients andmay also provide further information regarding theinteresting observation that CA12 expression appears tobe associated with less metastasis in these patients. Wepropose that CA9 expression be used to identify patientswho require more intensive systemic therapies due totheir increased likelihood of failure at distant sites.

Acknowledgements This work was supported by the NationalCancer Center Grant 0310150.

References

Hall EJ (2000) Oxygen effect and reoxygenation. In: Radiobiologyfor radiologists. Lippincott Williams and Wilkins, New York,pp 103–106

Halmi P, Lehtonen J, Waheed A, Sly WS, Parkkila S (2004)Expression of hypoxia-inducible, membrane-bound carbonic

anhydrase isozyme XII in mouse tissues. Anat Rec 277A:171–177

Harris AL (2002) Hypoxia—a key regulatory factor in tumourgrowth. Nat Rev Cancer 2:38–47

Ivanov SV, Kuzmin I, Wei MH, Pack S, Geil L, Johnson BE,Stanbridge EJ, Lerman MI (1998) Down-regulation of trans-membrane carbonic anhydrases in renal cell carcinoma cell linesby wild-type von Hippel-Lindau transgenes. Proc Natl Acad SciUSA 95:12596–12601

Ivanov S, Liao SY, Ivanova A, Danilkovitch-Miagkova A, Tar-asova N, Weirich G, Merrill MJ, Proescholdt MA, Oldfield EH,Lee J, Zavada J, Waheed A, Sly W, Lerman MI, Stanbridge EJ(2001) Expression of hypoxia-inducible cell-surface transmem-brane carbonic anhydrases in human cancer. Am J Pathol158:905–919

Keating JT, Ince T, Crum CP (2001) Surrogate biomarkers of HPVinfection in cervical neoplasia screening and diagnosis. AdvAnat Pathol 8:83–92

Lal A, Peters H, St Croix B, Haroon ZA, Dewhirst MW, Straus-berg RL, Kaanders JH, van der Kogel AJ, Riggins GJ (2001)Transcriptional response to hypoxia in human tumors. J NatlCancer Inst 93:1337–1343

Lieskovska J, Opavsky R, Zacikova L, Glasova M, Pastorek J,Pastorekova S (1999) Study of in vitro conditions modulatingexpression of MN/CA IX protein in human cell lines derivedfrom cervical carcinoma. Neoplasma 46:17–24

Loncaster JA, Harris AL, Davidson SE, Logue JP, Hunter RD,Wycoff CC, Pastorek J, Ratcliffe PJ, Stratford IJ, West CM(2001) Carbonic anhydrase (CA IX) expression, a potential newintrinsic marker of hypoxia: correlations with tumor oxygenmeasurements and prognosis in locally advanced carcinoma ofthe cervix. Cancer Res 61:6394–6399

Olive PL, Aquino-Parsons C, MacPhail SH, Liao SY, Raleigh JA,Lerman MI, Stanbridge EJ (2001) Carbonic anhydrase 9 as anendogenous marker for hypoxic cells in cervical cancer. CancerRes 61:8924–8929

Opavsky R, Pastorekova S, Zelnik V, Gibadulinova A, StanbridgeEJ, Zavada J, Kettmann R, Pastorek J (1996) Human MN/CA9gene, a novel member of the carbonic anhydrase family:structure and exon to protein domain relationships. Genomics33:480–487

Resnick M, Lester S, Tate JE, Sheets EE, Sparks C, Crum CP(1996) Viral and histopathologic correlates of MN and MIB-1expression in cervical intraepithelial neoplasia. Hum Pathol27:234–239

10Gy 20Gy0.90

0.95

1.00

1.05

1.10

1.15

1.20

pre-treatment during-treatment

(Radiation Dose)

1.00 1.01

n=12(p=0.79)

n=32(p=0.38)

1.181.03

CA

9/β

-act

in m

RN

A

Fig. 2 Change in the expressionof CA9 mRNA. A secondbiopsy was obtained at 10 or20 Gy of cumulative radiationdose in 12 and 32 patients,respectively, and their CA9mRNA expression wasexamined by RT-PCR. Openbox pre-treatment, shaded boxCA9 mRNA at 10 Gy (Lt) and20 Gy (Rt), respectively. Thebottom and the top of each boxrepresent 25 percentile and 75percentile of the observed CA9/b-actin mRNA ratio. The figureand the marks in each boxrepresent the median value ofeach group

307

Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat RevCancer 3:721–732

Van Putten LM, Kallman RF (1968) Oxygenation status of atransplantable tumor during fractionated radiation therapy.J Natl Cancer Inst 40:441–451

Vordermark D, Brown JM (2003) Endogenous markers of tumorhypoxia predictors of clinical radiation resistance? StrahlentherOnkol 179:801–811

Wouters BG, Brown JM (1997) Cells at intermediate oxygen levelscan be more important than the ‘‘hypoxic fraction’’ in deter-mining tumor response to fractionated radiotherapy. RadiatRes 147:541–550

Wykoff CC, Beasley N, Watson PH, Campo L, Chia SK, EnglishR, Pastorek J, Sly WS, Ratcliffe P, Harris AL (2001) Expressionof the hypoxia-inducible and tumor-associated carbonic an-hydrases in ductal carcinoma in situ of the breast. Am J Pathol158:1011–1019

308