ARTICLE

Ectonucleotidase expression profile and activity in humancervical cancer cell linesAline Beckenkamp, Danielle Bertodo Santana, Alessandra Nejar Bruno, Luciane Noal Calil,Emerson André Casali, Juliano Domiraci Paccez, Luiz F. Zerbini, Guido Lenz, Márcia R. Wink,and Andréia Buffon

Abstract: Cervical cancer is the third most frequent cancer in women worldwide. Adenine nucleotide signaling is modulated bythe ectonucleotidases that act in sequence, forming an enzymatic cascade. Considering the relationship between the purinergicsignaling and cancer, we studied the E-NTPDases, ecto-5=-nucleotidase, and E-NPPs in human cervical cancer cell lines andkeratinocytes. We evaluated the expression profiles of these enzymes using RT-PCR and quantitative real-time PCR analysis. Theactivities of these enzymes were examined using ATP, ADP, AMP, and p-nitrophenyl-5=-thymidine monophosphate (p-Nph-5=-TMP) as substrate, in a colorimetric assay. The extracellular adenine nucleotide hydrolysis was estimated by HPLC analysis. Thehydrolysis of all substrates exhibited a linear pattern and these activities were cation-dependent. An interesting difference in thedegradation rate was observed between cervical cancer cell lines SiHa, HeLa, and C33A and normal imortalized keratinocytes,HaCaT cells. The mRNA of ecto-5=-nucleotidase, E-NTPDases 5 and 6 were detectable in all cell lines, and the dominant geneexpressed was the Entpd 5 enzyme, in SiHa cell line (HPV16 positive). In accordance with this result, a higher hydrolysis activityfor UDP and GDP nucleotides was observed in the supernatant of the SiHa cells. Both normal and cancer cells presented activityand mRNAs of members of the NPP family. Considering that these enzymes exert an important catalytic activity, controllingpurinergic nucleotide concentrations in tumors, the presence of ectonucleotidases in cervical cancer cells can be important toregulate the levels of extracellular adenine nucleotides, limiting their effects.

Key words: cervical cancer, ectonucleotidases, purinergic signaling, biomarkers, NTPDase5.

Résumé : Le cancer du col utérin arrive au troisième rang sur le plan de la fréquence chez la femme, a travers le monde. Lasignalisation par les nucléotides d’adénine est modulée par les ectonucléotidases qui agissent en séquence pour former une cascadeenzymatique. Compte tenu de la relation qui existe entre la signalisation purinergique et le cancer, nous avons étudié les E-NTPDases,l’ecto-5=-nucléotidase et les E-NPP dans des lignées cellulaires de cancer du col utérin et des kératinocytes. Nous avons évalué les profilsd’expression de ces enzymes par RT-PCR et PCR quantitative en temps réel. L’activité de ces enzymes a été examinée par des essaiscolorimétriques en utilisant comme substrats l’ATP, l’ADP, l’AMP et le p-nitrophényl 5’-thymidine monophosphate (p-Nph-5=-TMP).L’hydrolyse des nucléotides d’adénine extracellulaires a été estimée par HPLC. L’hydrolyse de tous les substrats suivait un patronlinéaire et ces activités étaient dépendantes des cations. Une différence intéressante dans le taux de dégradation a été observée entreles lignées de cancer du col utérin SiHa, HeLa et C33A, et des kératinocytes normaux immortalisés HaCaT. L’ARNm de l’ecto-5=-nucléotidase et des E-NTPDases 5 et 6 était détectable dans toutes les lignées cellulaires, et Entpd 5 était le gène exprimé de façondominante dans la lignée SiHa (positive a VPH16). Conformément a ce résultat, une activité hydrolytique plus élevée pour l’UDP et leGDP était observée dans le surnageant des cellules SiHa. Tant les cellules normales que cancéreuses exprimaient les ARNm desmembres de la famille des NPP et présentaient l’activité enzymatique correspondante. Compte tenu du fait que ces enzymes exercentune activité catalytique importante, contrôlant les concentrations de nucléotides purinergiques dans les tumeurs, la présence desectonucléotidases dans les cellules de cancer du col utérin peut être importante afin de réguler les niveaux de nucléotides d’adénineextracellulaires, limitant leurs effets. [Traduit par la Rédaction]

Mots-clés : cancer du col utérin, ectonucléotidases, signalisation purinergique, biomarqueurs, NTPDase5.

Introduction

Cervical cancer is one of the most prevalent gynecological ma-lignancies in the world. The majority of cases of invasive cervixcancers occur by a slow evolution of the squamous intraepithelial

lesion. Cervical cancer remains a disease of high prevalence, inci-dence, and mortality (Saraiya et al. 2002). With respect to histo-logical classification, the most prevalent are the squamous cellcancers, and only a small portion is of glandular origin. The mainrisk factor for cervical cancer development is the human papil-

Received 20 May 2013. Revision received 20 November 2013. Accepted 13 December 2013.

A. Beckenkamp,* D.B. Santana,* L.N. Calil, and A. Buffon. LABC – Laboratory of Biochemical and Cytological Analysis, Analysis Department, Facultyof Pharmacy, Federal University of Rio Grande do Sul, Av. Ipiranga 2752, bairro Santana, CEP 90610-000, Porto Alegre, RS, Brazil.A.N. Bruno. Federal Institute of Education, Science and Technology of Rio Grande do Sul, Porto Alegre, RS, Brazil.E.A. Casali. Department of Biochemistry, Institute of Basic Health Sciences, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.J.D. Paccez and L.F. Zerbini. International Centre for Genetic Engineering and Biotechnology (ICGEB), Cancer Genomics Group, Cape Town, South Africa.G. Lenz. Laboratório de Sinalização Celular, Departamento de Biofísica, Federal University of Rio Grande do Sul, Porto Alegre, RS, Brazil.M.R. Wink. Laboratório de Biologia Celular, Department of Basic Health Sciences, Federal University of Health Sciences of Porto Alegre, RS, Brazil.Corresponding author: Andréia Buffon (e-mail: andreia.buffon@ufrgs.br).*These authors contributed equally to this work.

95

Biochem. Cell Biol. 92: 95–104 (2014) dx.doi.org/10.1139/bcb-2013-0051 Published at www.nrcresearchpress.com/bcb on 8 January 2014.

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

loma virus (HPV) infection, and many studies have indicated thatthe most frequent HPV types in cervical cancer cases are 16 and 18(Anttila et al. 2009; de Cremoux et al. 2009; Li et al. 2011). Primaryradiotherapy and concurrent platinum-based chemotherapy havebeen used as standard treatments for cervical cancer (Zagouriet al. 2012). However, many patients present recurrence, andmean survival in advanced cases is very short (Long et al. 2005).

Purinergic signaling can play a role in cell growth and death inthe human epidermis (Burnstock et al. 2012) and is involved inpathological processes, including cancer. The levels of extracellu-lar nucleotides are finely regulated through the hydrolysis per-formed by ecto-nucleotidases. Extracellular nucleotides act throughpurinergic receptors (P2-purinoceptors), subtypes P2X and P2Y.Ectonucleotidases and receptors were described to be altered indifferent tumor cells (Burnstock 2002; Hopfner et al. 2001;Tamajusuku et al. 2010; Wang et al. 2008; Yegutkin et al. 2011).

The physiological action induced by purinergic signaling is reg-ulated by an enzymatic cascade, which includes the members ofthe ectonucleoside triphosphate diphosphohydrolases (E-NTPDase)family, ectonucleotide pyrophosphatase/phosphodiesterases (E-NPP) family, and ecto-5=-nucleotidase/CD73 (Zimmermann 1992;Zimmermann et al. 2001). The E-NTPDase family consists of 8 re-lated enzymes (E-NTPDase 1–8). E-NTPDase 1, 3, and 8 are con-nected to the extracellular side of the plasma membrane and arealso able to hydrolyze nucleoside triphosphates as efficiently asthey hydrolyze diphosphates. In turn, E-NTPDase 2 has muchgreater affinity for nucleoside triphosphates than for diphos-phates (Zimmermann et al. 2001). In this way, E-NTPDase enzymesconvert extracellular purine nucleotides, like adenosine triphos-phate (ATP) and adenosine diphosphate (ADP), to adenosine mono-phosphate (AMP), which is hydrolyzed by ecto-5=-nucleotidase toadenosine (Buffon et al. 2007a; Zimmermann 1992).

The ectonucleotide pyrophosphatase/phosphodiesterases (E-NPPs) constitute a group of 7 different enzymes (E-NPP 1–7) thatare located on the surface of the cell membrane. These proteinscarry out the hydrolysis of phosphodiester and pyrophosphatesbonds, forming monophosphated products. Among the possiblesubstrates are the adenine nucleotides, but only E-NPPs 1, 2, and 3can perform the hydrolysis of diadenosine polyphoshpates andp-Nph-5=-TMP (Buffon et al. 2010; Stefan et al. 2006).

The ectonucleotidases have been described as important notonly in regulating physiological processes but also in various pa-thologies, such as cancer. For instance, differential expression ofE-NTPDases and ecto-5=-nucleotidase/CD73 have been described inbladder cancer cells (Stella et al. 2010). In addition, recent studieshave addressed the involvement of purinergic system in plateletsof patients with cervical intraepithelial neoplasia and uterine cancer(Maldonado et al. 2010, 2012). However, these studies showed nosignificant changes in the ectonucleotidase activities with respectto the grade of neoplasia development. Although this enzymaticcascade has been evaluated in the circulation of patients withcervical intraepithelial neoplasia and uterine cancer (Maldonadoet al. 2012), at present no studies have reported the major role andexpression of these enzymes, which hydrolyze extracellular tri-and di-phosphonucleosides in cervical cancer cells.

In view of the involvement of the purinergic system in severalmalignant processes and of the need for more specific cervicalcancer markers, the aim of this research was to analyze the ade-nine nucleotide metabolism and expression of the ectonucleoti-dase family members in human cervical cancer cells. Therefore,the present study aims to understand the mechanisms responsi-ble for the development of the cervical carcinoma and to discovernovel molecular targets for prognostic evaluation of this malig-nancy.

Materials and methods

Cell linesThe HaCaT cells (non-tumoral control) and cervical human can-

cer cells SiHa (HPV 16-positive), HeLa (HPV 18-positive), and C33A(HPV-negative) were obtained from American Type Culture Collec-tion (Rockville, Md., USA). The cells were maintained in cultureflasks in Dulbecco’s Modified Eagle’s (DMEM) medium, supple-mented with 10% fetal bovine serum (FBS), in an atmosphere of 5%CO2/95% air at 37 °C. Cells were seeded in 24 multiwell plates (2 ×104 cells/well) and allowed to grow to confluence for the E-NTPDaseassays. For E-NPP assays, 5 × 103 cells/well were seeded in 96 mul-tiwell plates.

E-NTPDase activity assayThe ectonucleotidase activity was assessed using nucleotides

ATP, ADP, and AMP as substrate. A reaction medium containing20 mmol/L HEPES, 5 mmol/L KCl, 120 mmol/L NaCl, 10 mmol/Lglucose, and pH 7.4 was used to wash the cells 3 times and removethe culture medium. The enzymatic reaction was started by add-ing the substrate to a final concentration of 2 mmol/L, to a finalvolume of 200 �L for 15 min of reaction at 37 °C. The reaction wasinterrupted adding 150 �L of trichloracetic acid (TCA) 10% (v/v). Theamount of inorganic phosphate (Pi) released was measured at630 nm, according to the malachite green method, using a stan-dard curve as reference with KH2PO4 as a Pi standard. Specificactivities were expressed as nmol Pi released/min/mg of protein.

Ecto-nucleotide pyrophosphatase/phosphodiesterase (E-NPP)activity assay

E-NPP activity was assessed using p-nitrophenyl-5=-thymidinemonophosphate (p-Nph-5=-TMP)assubstrate.Thereactionmediumcon-taining 50 mmol/L Tris-HCl buffer, 5 mmol/L KCl, 135 mmol/LNaCl, 10 mmol/L glucose, pH 8.9 was used to wash the cells andremove the culture medium. The enzymatic reaction was startedby the addition of the p-Nph-5=-TMP to a final concentration of1.0 mmol/L, at a final volume of 40 �L. After 40 min of incubation,the addition of 30 �L of NaOH 0.2 mol/L stopped the reaction. Theamount of p-nitrophenol released was measured at 400 nm, usinga molar extinction coefficient of 18.8 × 10−3 (mol/L)–1·cm–1. Enzymeactivities were expressed as nmol of p-nitrophenol released·min–1·mg–1

of protein.

Divalent cation dependence and determination of kineticparameters

To investigate the cation dependence on E-NTPDase and E-NPPactivity, we tested the hydrolysis rate in the presence or absenceof divalent cations (Ca2+ and Mg2+) or EDTA (cation chelator). Inthe same way, the kinetic parameters, apparent Michaelis−Menten con-stants (Km, app), and the calculated maximum velocities (Vmax,app), were obtained using the Eadiee−Hofstee plot. These param-eters for the E-NTPDases were obtained by incubating the cellswith concentrations of nucleotides ranging from 0.05 to 2.0 mmol/L for15 min. For the E-NPP assay the concentration range was 0.025–2.5 mmol/L for SiHa, C33A, and HaCaT cell lines, and for the HeLathe range was of 0.010–1.5 mmol/L for 40 min. Controls to correctnon-enzymatic hydrolysis of substrate were prepared by the addi-tion of buffer only containing substrate, followed by the reactiondescribed above. All samples were prepared in triplicate.

UDP and GDP hydrolysis assay on adherent cells and insupernatant

Adherent cells at confluence in flasks of 25 cm2 were washedwith reaction medium to remove the culture medium with serum,and then 2 mL of reaction medium were added. After incubationin 5% CO2/95% air at 37 °C for 1 h, the supernatant was collectedand centrifuged for 5 min at 10 000g. An aliquot of supernatantwas incubated in the presence of substrates UDP and GDP2.0 mmol/L for 15 min. The enzyme activity was assayed, and the

96 Biochem. Cell Biol. Vol. 92, 2014

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

reaction rate determined as described above. The UDP and GDPhydrolysis activity on cells surface was performed as already de-scribed above.

HPLC analysisThe cell monolayer was washed 3 times with the same incuba-

tion buffer used in the ectonucleotidase activity assay, with addi-tion of Mg+2 4 mmol/L. The addition of the nucleotide (ATP) to afinal concentration of 100 �mol/L and to a final volume of 200 �Lstarted the reaction. After assigned incubation times were elapsed(0, 5, 10, 30, 60, 90, 120, and 180 min), the reaction was stoppedon ice. The supernatant was centrifuged at 4 °C for 10 min at12 000g. For the analysis, aliquots of 40 �L were applied to areversed-phase HPLC system (Shimadzu, Japan), using a C18 Rest-eck column of 250 mm × 4.6 mm, poro size 5 �m (USA) at 260 nmwith a mobile phase containing 60 mmol/L KH2PO4, 5.0 mmol/Ltetrabutylammonium chloride, pH 6.0, in 30% methanol, as pre-viously described (Casali et al. 2003). The peaks were identified bytheir retention time and by comparison with standards. The re-sults were expressed as the amount of the different compounds ina given incubation time. All incubations were carried out in trip-licate, and the controls to correct non-enzymatic hydrolysis ofnucleotides were performed by measuring the peaks detected forthe same reaction medium, but incubated without cells. The con-trol for cellular purine secretion was performed by incubatingthe cells without substrate under the same conditions describedabove.

RT-PCR analysisTotal RNA from cervical carcinoma cell cultures were isolated

with Trizol LS reagent (Life Technologies, Carlsbad, Calif., USA), inaccordance with the manufacturer’s instructions. The cDNA spe-cies were synthesized with M-MLV reverse transcriptase (Promega,Madison, Wis, USA) from 5 �g of total RNA to a final volume of25 �L with a random hexamer primer in accordance with themanufacturer’s instructions, and as a control of this synthesis, aPCR with the primer of GAPDH was performed. One microliter ofthe RT reaction mix was used as a template for PCR in a totalvolume of 25 �L, containing a 1 �L of each primer (Table 1),50 �mol/L dNTPs, and 1.25 units of TaqDNA polymerase (LudwigBiotec, Porto Alegre, RS, BRA) in the supplied reaction buffer. ThePCR was run for 35 cycles, and the cycling conditions were asfollows: 1 min at 95 °C, 1 min at 94 °C, 1 min at annealing temper-ature, 1 min at 72 °C, and a final 10 min extension at 72 °C. Tenmicroliters of the PCR reaction was analyzed on a 1.3% agarose gelcontaining 5 �L of ethidium bromide and visualized under ultra-violet light. Negative controls were performed with distilled wateras a template, and positive controls were plasmids with cDNAsequences for all E-NTPDases, except for ecto–5=-nucleotidase(CD73), in which line U138MG was used as positive control.

Real-time PCR analysisTotal RNA and cDNA were generated as described in RT-PCR

analysis. SYBR Green I-based real-time PCR was carried out onLight cycler II (Roche), as previously described (Buffon et al. 2010).All PCR mixtures were performed using the Kapa SyBR FAST Uni-versal qPCR KIT (kappa biosystems) following the manufacturer’srecommendation, with 2 �L cDNA to a 20 �L final volume reactionmix. The samples were loaded into wells for Low Profile 96-wellmicroplates (Roche). After an initial denaturing step for 45 s at94 °C, conditions for cycling were 40 cycles of 30 s at 94 °C, 30 s at58 °C, 30 s at 72 °C. The fluorescence signal was measured rightafter incubation for 5 s at 72 °C following the extension step,which eliminates possible primer dimer detection. At the end ofthe PCR cycles, a melting curve was generated to identify specific-

ity of the PCR product. For each run, serial dilutions of humanGAPDH plasmids were used as standards for quantitative mea-surement of the amount of amplified DNA. In addition, to norm-alize each sample, GAPDH primers were used to measure theamount of GAPDH cDNA. All samples were run in triplicate, andthe data were presented as ratio of enzymes/GAPDH. The primersused for real-time PCR were the same as used in the RT-PCR anal-ysis (Table 1).

Cellular integrityTo evaluate cellular integrity during the experiments, we deter-

mined the activity of the cytosolic enzyme lactate dehydrogenase(LDH). The activity of LDH released into the supernatant was mea-sured using the test kit LDH Liquiform™ (Labtest Diagnostica, MG,Brazil) and compared with the activity of cells lysed with 1.8%Triton X-100 (Buffon et al. 2007a).

Protein determinationAfter the E-NTPDase and E-NPP activity assays, 200 �L of NaOH

was added to each well, and plates were kept refrigerated over-night. An aliquot of 50 �L was used for protein determination.Protein was measured according to the Bradford method, usingbovine serum albumin as standard (Bradford 1976). For experi-ments in supernatant of adherent cells, protein content was mea-sured using the Sensiprot® kit (Labtest Diagnostica, MG, Brazil).

Statistical analysisThe results are expressed as means ± SD. Data were analyzed by

one-way analysis of variance (ANOVA), followed by the Tukey’stest using the program Graph Pad Prism 5. Values of p < 0.05 wereconsidered significant.

Results

Cellular integrityLDH activity was used as a marker of cellular integrity. It was

measured in the cells after the incubation with adenine nucleo-tides and p-Nph-5=-TMP. These activities were compared with theactivity of cells lysed with 1.8% Triton X-100 (positive control). Theresults point to approximately 95% of the cells remaining intactafter the incubation (data not shown).

ATP, ADP, and AMP hydrolysisInitially we investigated the conditions of ATP, ADP, and AMP

hydrolysis activities as a function of time. The cells were incu-bated as described in materials and methods, with 2.0 mmol/L ofATP, ADP, or AMP and presented a linear profile for up to 40 minof incubation, for all cell lines studied (Supplementary data Fig. S1)1.While an increased specific activity was observed for ATP hy-drolysis by SiHa (HPV16 positive) cells, HaCaT keratinocytes pre-sented higher ADPase and AMPase activities when compared withcervical cancer cells (Table 2). Moreover, all cervical cancer celllines tested showed an interesting difference in the rate of degra-dation of extracellular nucleotides, without a specific pattern ofATP, ADP, and AMP hydrolysis. The cell line of immortalized ker-atinocytes, HaCaT, exhibited hydrolysis rate for extracellular nucleo-tides in the order AMP > ATP > ADP. The extracellular hydrolysisby SiHa and HeLa followed the order ATP > AMP > ADP, though forC33A there was no difference in the rate of hydrolysis of ATP andADP, whereas AMP was hydrolyzed less efficiently.

The hydrolysis of p-Nph-5=-TMPThe possible existence of members of the E-NPP family, which

participate in nucleotide hydrolyses, also was evaluated. The re-sults demonstrate that all cancer cell lines and keratinocytes arecapable of hydrolyzing the E-NPP artificial substrate p-Nph-5=-

1Supplementary data are available with the article through the journal Web site at http://nrcresearchpress.com/doi/suppl/10.1139/bcb-2013-0051.

Beckenkamp et al. 97

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

TMP, with a linear behavior of up to 60 min of incubation (Sup-plementary data Fig. S2)1. Moreover, tumoral cells showed lowerenzymatic activity when compared with HaCaT cell line (Table 2).

Cation dependenceWe tested the hydrolysis rate of the substrates for E-NTPDases

(ATP, ADP, and AMP) and E-NPPs (p-Nph-5=-TMP) in the presence orabsence of divalent cations (Ca2+ and Mg2+) or EDTA (cation che-lator), to evaluate the possibility of cation dependence for theactivities of E-NTPDases, ecto-5=nucleotidase and E-NPPs in cervi-cal cancer cells. In the presence of the chelator, EDTA, the hy-drolysis of all substrates tested was lower when compared to thecontrol (without addition of divalent cations) for all cell lines(Supplementary data Figs. S3–S7)1. The hydrolysis of substrateswas increased in the presence of Ca+2 and Mg+2, when comparedto the control group, and the stimulation was most evident whenMg2+ was used for E-NTPDases and most evident when Ca+2 wasused for E-NPPs. Then, for the following experiments with E-NTPDaseactivities we used 4 mmol/L of Mg2+, and for E-NPPs activities weused 6 mmol/L of Ca+2.

Kinetic parametersThe Eadie–Hofstee plots showed that all enzymatic activities

increased with increasing nucleotide concentration until satura-tion (Supplementary data Figs. S8–S12)1. The Michaelis–Mentenconstants values (Km) calculated by linear regression and the max-imum velocities (Vmax) for the adenine nucleotides and p-Nph-5=-

TMP hydrolysis, by all cell lines, are shown in Table 3 and Table 4,respectively.

Ectonucleotidase mRNA expressionWe also investigated the major nucleotide degradation en-

zymes expressed by the immortalized keratinocytes and cervicalcancer cells by RT-PCR and quantitative real-time RT-PCR analysis(Fig. 1). mRNAs of the Entpd 5 and 6 and ecto-5=-nucleotidase/CD73were detected at different levels of expression in the cervical can-cer cells and HaCaT (Fig. 1) and Entpd 5 was highly expressed inSiHa cell line (HPV16 positive). The signal for E-NTPDase1, 2, 3, and8 was almost undetectable. Moreover, in accordance with the en-zymatic activity, SiHa and HeLa cancer cell lines exhibited similarlevels of ecto-5=-nucleotidase/CD73 expression, whereas in C33Acells the expression of this gene was almost undetectable. In ac-cordance, the AMP hydrolysis was also lower in the C33A cells.

The E-NPPs 1, 2, and 3 family members, also involved in thepurinergic signaling modulation, presented a mRNA expressionprofile that was different between cervical cancer cells and non-tumoral cell line HaCaT (Fig. 2), suggesting the presence of anextracelullar enzimatic complex that regulates nucleotide avail-ability in a different way in cancer cells. The presence of NPPs,besides their mRNA expression, could be detected by the enzy-matic activity measured using the specific artificial substratep-Nph-5=-TMP. According to kinetics results presented in theTable 2, the level of mRNA expression was matched by the enzy-matic activities measured on cells surface, where the enzymaticactivity of SiHa was higher than HeLa and C33A (SiHa > HeLa > C33A).

Membrane bound and soluble diphosphatase activityConsidering the preference of the NTPDase5, which degrades

for UDP and GDP, and the possibility that this enzyme could besecreted, we determined the rates of hydrolysis of these nucleo-tides on cells surface and in their supernatant (Table 5). Interest-ing all cells lead to high levels of hydrolysis activity for bothnucleotides (UDP > GDP) in their supernatant, suggesting thatNTPDase5 could be secreted in culture Among the cells, SiHashowed the highest degradation rates, in agreement with themRNA level quantified by real time PCR (Fig. 1b).

Table 1. Primer sequences for ectonucleotidases, annealing temperatures (Ta) and fragment sizes.

Gene GenBank number Primer sequence (5= to 3=) Ta (°C)Fragmentsize (bp)

NTPDase1 F NM_001776.5 CTACCCCTTTGACTTCCA 58 176NTPDase1 R CTCCCCCAAGGTCCAAAGCNTPDase2 F NM_001246.3 GGGTGCACGCATCCTCTCG 58 206NTPDase2 R CCTCGCTGGCTCTGTCCTCNTPDase3 F NM_001248.2 TACCGAACTCCAACCATCA 58 310NTPDase3 R CCTTGACTTTTTGCATACANTPDase5 F NM_001249.2 CAGGTCAGCTGCATGGCCACA 58 226NTPDase5 R TCCAGGGCTCCCAGGGTTGCNTPDase6 F NM_001114089.1 CAGCTGCAGACGGGCACGAG 58 154NTPDase6 R GCAGAAAGACCTGGCTTCACTGCTNTPDase8 F NM_001033113.1 GCGGACACAGAAGCGTCTAA 58 230NTPDase8 R ACTTGATGTCTGTGGGCAGGEcto-5=NT F NM_001204813.1 GCTGGCGCCTGGGAGCTTAC 58 176Ecto-5=NT R TCCAGCAGCAGCACGTTGGGNPP1 F NM_006208.2 GGGGAGGAGCCGCTGGAGAA 58 213NPP1 R ACAGGCAGCATCACAGCGACANPP2 F NM_001130863.2 GCTTGGCCAGGGGAGACTGC 58 158NPP2 R TGCACGGAAGCCATCCACGGNPP3 F NM_005021.3 AACTGCCCTGGGTGGCTGGA 58 134NPP3 R CGGACCCGGGCAATGTGAGCGAPDH F NM_001256799.1 CAAAGTTGTCATGGATGACC 58 195GAPDH R CCATGGAGAAGGCTGGGG

Note: F = Forward, R = Reverse.

Table 2. Adenine nucleotides and p-Nph-5=-TMP hydrolysis in keratin-ocytes and cervical cancer cell lines.

Cell line ATP ADP AMP p-Nph-5=TMP

HaCaT 12.45±0.86 11.11±0.30 15.59±0.39 2.35±0.04SiHa 16.69±0.34* 9.48±0.52 10.92±0.01* 1.36±0.02*HeLa 10.57±0.84 8.16±0.16* 9.22±0.39* 0.62±0.01*C33A 5.58±0.04* 5.56±0.66* 3.84±0.05* 0.30±0.01*

Note: Mean values ± SD is expressed in specific activity in nmol of Piliberated·min–1·mg–1 of protein and nmol of p-nitrophenol liberated·min–1·mg–1

of protein, for adenine nucleotides and p-Nph-5=-TMP hydrolysis, respectively.The values are representative of 3 different experiments. Data were comparedby ANOVA followed by Tukey's test. *p < 0.05, represents statistical significancewhen cancer cells (SiHa, HeLa, and C33 A) were compared to keratinocytes (HaCaT).

98 Biochem. Cell Biol. Vol. 92, 2014

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

Metabolism of extracellular ATPThe pattern for extracellular ATP metabolism measured in im-

mortalized keratinocytes and cervical cancer cells was analyzedby HPLC. These cells present all enzymes to the complete adeninecatabolism, as can be seen by the compounds formed from ATP(Tables S1 and S2 in Supplementary data)1. Although there weresome minor differences between the cells studied, their mainpattern of substrate hydrolysis remained similar (Fig. 3). After180 min, all cell lines hydrolyzed ATP to their metabolites, pro-ducing low amounts of AMP. In agreement with the low enzy-matic activities, the mRNA expression for E-NTPDase1, 2, 3, and 8,which are important enzymes responsible for ATP and ADP extra-cellular hydrolysis, was almost undetectable in cancer cells (Fig. 1).

Interestingly, some slight differences in adenosine productionwere observed, mainly in HaCaT and SiHa cells, where a detect-able amount of AMP nucleoside was observed after 30 min ofincubation, compared with C33A and HeLa cells. This result is inagreement with the specific activities for AMP hydrolysis, andecto-5=-nucleotidase/CD73 mRNA expression, which are higher inHaCaT and SiHa cells (Table 2 and Fig. 1).

DiscussionSeveral studies have demonstrated the relationship between

ectonucleotidases and mechanisms of tumor progression (Braganholet al. 2009; Cappellari et al. 2012; Feng et al. 2011; Spychala et al.2004). However, few reports in the literature have addressed therelationship between these enzymes and cervical cancer. In the

present study, we performed an analysis of the enzymatic activityand the expression of the E-NTPDases, ecto-5=-nucleotidase, andE-NPPs in human cervical cancer cells (HeLa, SiHa, and C33A) aswell in immortalized keratinocytes (HaCaT). It is the first identifi-cation of ectonucleotidases in cervical cancer cells. NTPDases 4and 7 were not investigated in this study, as these proteins areexclusively intracellular and thus cannot be responsible for nucle-otide degradation at the surface of cervical cancer cells (Biederbicket al. 2000; Shi et al. 2001).

The enzymatic activity analysis revealed that all cell lines me-tabolize adenine nucleotides, according to different patterns ofextracellular hydrolysis, and presented low ATP and ADP hy-drolytic activity when compared to other tumoral cells (Braganholet al. 2009; Buffon et al. 2007a). The enzymatic activities for ade-nine nucleotides hydrolysis were shown to be cation-dependent,being activated by divalent cations, such as Ca2+ and Mg2+. Thisdependence was confirmed by the reduction in enzyme activitywhen a chelating agent, EDTA, was added to the system. The bestactivator for the nucleotide hydrolysis by E-NTPDase for all thestudied cells was Mg2+, which was used for the experiments thatfollowed (at the concentration of 4 mmol/L). In addition, the HPLCanalysis indicated that all cell lines slowly hydrolyze ATP, which isconsistent with the low mRNA expression of the key enzymes hy-drolyzing ATP, E-NTPDase1, 2, and 3, observed in this study.

Ectonucleotide pyrophosphatase/phosphodiesterase activitywas also detected, based on its ability to hydrolyze the substratep-Nph-5=-TMP. However, this activity is very low in the optimal pHfor these enzymes (pH 8.9), if compared with descriptions forother tumor cells (Buffon et al. 2010; Stella et al. 2010). The extra-cellular p-Nph-5=-TMP hydrolysis presented Ca2+ as the best cofac-tor. This finding is in agreement with previously described findings forthese enzymes in cancer cells (Buffon et al. 2007a, 2007b).

In this study, we reported for the first time in cervical cancercell lines, SiHa, HeLa, and C33A, the specific mRNA for ectonucle-otidase members. The NTPDases5 and 6, ecto-5=-nucleotidase/CD73, and also NPP1, 2, and 3 genes were successfully amplified, atdistinct levels. The highest expression was observed for the en-zymes NTPDase5 and NPP2 in the SiHa cells, which present theHPV16 copies. The higher expression of NTPDase5, which presentsa preference for nucleoside diphosphates, corroborates the activ-ity of UDP and GDP hydrolysis measured on cells surface. Veryinterestingly, this activity was detected in high levels on superna-tant, suggesting that the enzyme is possibly released in its solubleform from cells and may therefore affect the tumors’ microenvi-ronment.

In fact, an important finding of this study was the increasedexpression of NTPDase5 in SiHa cells but not in the immortalizedkeratinocytes HaCaT, a non-tumoral control. This soluble proteinhas been identified as PCPH, a known human proto-oncogene(Paez et al. 2001), and previously published data have reported themutated or deregulated expression of this enzyme in human can-cer progression (Blanquez et al. 2004; Read et al. 2009; Villar et al.2007). Recently, E-NTPDase5 was also described as an importantlink in the PI3K/PTEN pathway. Besides promoting cell growthand survival, E-NTPDase5 seems to mediate a compensatory in-crease in aerobic glycolysis, the Warburg effect, and has beenproposed as a potential target in anticancer treatment (Fang et al.2010). The presence of this proto-oncogene in SiHa cell line may bea characteristic of cervical cancer cells, possibly linked to the HPV16 activation pathways. In fact, the exposure of human keratino-cytes to HPV16 pseudovirions stimulates the Akt signaling path-way by activation of PI3K protein and the EGFR receptor (Surviladzeet al. 2013). Furthermore, differential regulation of Akt/PKB path-way members was also observed in transfected cells with E6 viralproteins, involved in human epithelial cell immortalization andtransformation (Contreras-Paredes et al. 2009). Therefore theNTPDase5 may be linked with the expression of oncogenic viralproteins (as high risk HPV16) in cervical cancer development,

Table 3. Kinetic parameters of nucleotide hydrolysis inkeratinocytes and cervical cancer cell lines.

Cell linenucleotide Km (�mol/L)

Vmax (nmol ofPi·min–1·mg–1

of protein)

HaCaT ATP 173.5±3.5 15.5±0.1ADP 282.0±36.7 11.4±0.4AMP 552.5±62.9 16.5±0.1

SiHa ATP 109.5±17.6 18.4±1.8ADP 180.0±43.8 10.0±0.6AMP 131.0±9.9 12.7±0.1

HeLa ATP 270.0±11.3 12.4±0.1ADP 113.0±4.2 8.7±0.2AMP 28.5±3.5 8.5±0.1

C33A ATP 292.0±36.7 6.1±0.5ADP 433.5±53.0 8.0±1.4AMP 89.5±36.0 4.5±0.2

Note: Km and Vmax values obtained from Eadie–Hofstee plotsfor extracellular hydrolysis of ATP, ADP, and AMP by immortal-ized keratinocytes and cervical cancer cells. Reaction rate wasmeasured by released Pi, as described in Material and methods.Results were obtained with a nucleotide concentration rangingfrom 0.05 to 2.0 mmol/L for ATP, ADP, and AMP, plus 4.0 mmol/L Mg2+.

Table 4. Kinetic parameters of p-Nph-5=-TMP hydrolysisin keratinocytes and cervical cancer cell lines.

Cell line Km (�mol/L)

Vmax (nmol ofp-nitrophenol·min–1·mg–1

of protein)

HaCaT 78.8±13.5 3.8±0.1SiHa 24.1±0.4 1.4±0.1HeLa 30.7±0.1 0.9±0.1C33A 37.8±0.4 0.6±0.1

Note: Km and Vmax values obtained from Eadie–Hofstee plotsfor extracellular hydrolysis of p-Nph-5=-TMP by immortalized ker-atinocytes and cervical cancer cells. Reaction rate was measuredby released p-nitrophenol, as described in Material and methods.Results were obtained with a substrate concentration rangingfrom 0.025 to 2.5 mmol/L for SiHa, C33A, and HaCaT, and for HeLathe range was 0.01 to 1.5 mmol/, plus 6.0 mmol/L Ca2+.

Beckenkamp et al. 99

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

Fig. 1. (a) RT-PCR analysis of NTPDases and CD73 expression by immortalized keratinocytes (HaCaT) and cervical cancer cell lines (SiHa, HeLa,and C33A). Total RNA was isolated from cell cultures, and the cDNA were analyzed by PCR with primers for the NTPDases and CD73, asdescribed in Materials and methods. Plasmids containing the sequences of all NTPDases were used as positive controls (PC), and U138MG wasused as PC for CD73. The length (bp) of the PCR products obtained with each pair of primers is given in each figure. (b) Quantitativeexpression of NTPDases and CD73 in immortalized keratinocytes (HaCaT) and cervical cancer cell lines (SiHa, HeLa, and C33A) was analyzed byreal-time PCR, as described in Materials and methods. Results are presented as the ratio of cDNA of enzymes/GAPDH. Bars representmean ± SD for 2 experiments.

100 Biochem. Cell Biol. Vol. 92, 2014

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

regulating cell transformation. However, the real role played bythis enzyme remains to be elucidated, and future studies will aimto address its significance in cervical cancer.

We cannot disregard the possibility that other NTPDases can besecreted from cervical cancer cells nor its contribution for extra-cellular hydrolysis of nucleoside diphosphates. In this sense, wealso detected the mRNA of NTPDase6 that can be secreted andhas a preference for nucleoside diphosphates (Braun et al. 2000;Hicks-Berger et al. 2000; Ivanenkov et al, 2003). However, thisenzyme has been described to show a preference for GDP and IDPin relation to UDP (Ivanenkov et al. 2003; Braun et al. 2000;Hicks-Berger et al. 2000), and we observed a higher UDP degrada-tion. mRNA of NTPDase6 was also less expressed when compared

Fig. 2. (a) RT-PCR analysis of NPPs expression by immortalized keratinocytes (HaCaT) and cervical cancer cell lines (SiHa, HeLa, and C33A).Total RNA was isolated from cell cultures, and the cDNA were analyzed by PCR with primers for the NPPs, as described in Materials andmethods. The length (bp) of the PCR products obtained with each pair of primers is given in each figure. (b) Quantitative expression of NPPs inimmortalized keratinocytes (HaCaT) and cervical cancer cell lines (SiHa, HeLa, and C33A) was analyzed by real-time PCR, as described inMaterials and methods. Results are presented as the ratio of cDNA of enzymes/GAPDH. Bars represent mean ± SD for 2 experiments.

Table 5. UDP and GDP hydrolysis activity on keratinocytes and cervi-cal cancer cell lines surface and supernatant.

UDP GDP

Cell line Cell surface Supernatant Cell surface Supernatant

HaCaT 8.21±0.21 39.14±0.33 6.73±0.23 34.77±0.04SiHa 9.21±0.29 50.75±1.63* 7.05±0.52 43.75±0.11*HeLa 6.64±0.21* 38.93±1.33 6.42±0.64 28.37±1.10*C33A 5.78±0.39* 34.03±0.81* 5.98±0.23 25.83±0.79*

Note: Mean values ± SD is expressed in specific activity (nmol of Piliberated·min–1·mg–1 of protein) for UDP and GDP hydrolysis. The values arerepresentative of 3 different experiments. Data were compared by ANOVA fol-lowed by Tukey's test. *p<0.05, represents statistical significance when cancercells (SiHa, HeLa, and C33 A) were compared to keratinocytes (HaCaT).

Beckenkamp et al. 101

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

to NTPDase5 in the majority of cells. It is possible, therefore, thatthe dinucleotide activity observed in the supernatant is fromNTPDase5 and 6, but the higher UDPase activity and higher ex-pression argues in favor of NTPDase5 as the major responsible forthis activity in SiHa cells.

The naturally immortalized human cell line, HaCaT, used as acontrol of non-tumor cell in this study, is known to have normalmorphogenesis, differentiation features, and is grown in tradi-tional culture media (Deyrieux and Wilson 2007). According toour results, a higher level of extracellular hydrolysis of AMP byHaCaT cells was observed in relation to the cervical cancer cells.The immortalized keratinocytes also metabolizes extracellularATP without the transitory accumulation of AMP and productiondetectable quantities of adenosine. This results correlates withthe high activity of hydrolysis of this nucleotide and with theimportant increased expression of CD73/ecto-5=-nucleotidase mRNA bynon-tumoral HaCaT cells in real time-PCR analysis. The involve-ment of this enzyme in tumor cells has been reported (Cappellariet al. 2012; Stagg et al. 2012; Wu et al. 2012), but its activity isvariable. In addition to the important role in the generation ofextracellular adenosine, this enzyme may also be involved in cell–cell and cell–matrix adhesion, mediating tumor invasiveness(Bavaresco et al. 2008; Zhang 2012). However, the low expression

and activity in cervical cancer cells, either limiting the effectsmediated by adenosine or interfering with the process of celladhesion, is unclear and should be further investigated.

Earlier studies have shown alterations in platelets E-NTPDaseand 5=-nucleotidase activities in patients previously treated foruterine cervical neoplasia with either conization or radiotherapy(RTX) but are not sensitive with respect to the grade of neoplasiadevelopment (Maldonado et al. 2012). More specifically, the pu-rinergic receptor expression also has been described in cervicalcancer tissues. P2X7 expression is lower in uterine epithelial can-cer tissues, compared with the corresponding normal tissues, sug-gesting that the mRNA and protein levels of P2X7 could be used asa biomarker to differentiate normal and cancer uterine epithelialtissues (Li et al. 2006). Therefore, considering the expression of thereceptors in cervical cancer cells, the importance of this enzy-matic cascade, which modulates the agonist levels, should be con-sidered.

In conclusion, our results demonstrate that cervical cancer cellsexpress ectonucleotidase members at different levels and alsoshow different patterns of hydrolysis of adenine nucleotides, reg-ulating levels of extracellular nucleotides. The enzyme with high-est expression on tumor cells analyzed specifically in the SiHa cellline, but not in the control, was a proto-oncogene, E-NTPDase5. To

Fig. 3. Hydrolysis of extracellular ATP and product formation by immortalized keratinocytes HaCaT (a), and of cervical cancer cells SiHa (b),HeLa (c), and C33A (d). Keratinocytes and cervical cancer cell lines were incubated with 100 �mol/L of ATP, as described in Materials andmethods. The amount of hypoxantine, xanthine, and uric acid was omitted from the graphics to facilitate visualization. The amount of allproducts formed from ATP degradation as well as the cells basal secretion can be seen in the Tables S1 and S2 in the Supplementary data.1

102 Biochem. Cell Biol. Vol. 92, 2014

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

our knowledge, this is the first report showing the extracellularadenine nucleotide metabolism and ectonucleotidases expressionin cervix tumor cells, and further studies are needed to elucidatethe role of this family of enzymes in cervical cancer.

AcknowledgementsThis work was supported by grants from Fundação de Amparo a

Pesquisa do Estado do Rio Grande do Sul (FAPERGS) and fromCNPq-Brasil.

ReferencesAnttila, A., von Karsa, L., Aasmaa, A., Fender, M., Patnick, J., Rebolj, M., et al.

2009. Cervical cancer screening policies and coverage in Europe. Eur. J. Can-cer, 45: 2649–2658. doi:10.1016/j.ejca.2009.07.020. PMID:19699081.

Bavaresco, L., Bernardi, A., Braganhol, E., Cappellari, A.R., Rockenbach, L.,Farias, P.F., et al. 2008. The role of ecto-5=-nucleotidase/CD73 in glioma cellline proliferation. Mol. Cell. Biochem. 319: 61–68. doi:10.1007/s11010-008-9877-3.PMID:18636315.

Biederbick, A., Kosan, C., Kunz, J., and Elsasser, H. 2000. First apyrase splicevariants have different enzymatic properties. J. Biol. Chem. 275: 19018–19024.doi:10.1074/jbc.M001245200. PMID:10858452.

Blanquez, M.J., Arenas, M.I., Conde, I., Tirado, O.M., Paniagua, R., and Notario, V.2004. Deregulated expression of the PCPH proto-oncogene in human breastcancers. Int. J. Oncol. 25: 821–830. PMID:15375529.

Bradford, M.M. 1976. A rapid and sensitive method for the quantitation of mi-crogram quantities of protein utilizing the principle of protein-dye binding.Anal. Biochem. 72: 248–254. doi:10.1016/0003-2697(76)90527-3. PMID:942051.

Braganhol, E., Morrone, F.B., Bernardi, A., Huppes, D., Meurer, L.,Edelweiss, M.I., et al. 2009. Selective NTPDase2 expression modulates in vivorat glioma growth. Cancer Sci. 100: 1434–1442. doi:10.1111/j.1349-7006.2009.01219.x. PMID:19558578.

Braun, N., Fengler, S., Ebeling, C., Servos, J., and Zimmermann, H. 2000. Se-quencing, functional expression and characterization of rat NTPDase6, anucleoside diphosphatase and novel member of the ecto-nucleoside triphos-phate diphosphohydrolase family. Biochem. J. 3: 639–647. doi:10.1042/0264-6021:3510639. PMID:11042118.

Buffon, A., Ribeiro, V.B., Wink, M.R., Casali, E.A., and Sarkis, J.J. 2007a. Nucleo-tide metabolizing ecto-enzymes in Walker 256 tumor cells: molecular iden-tification, kinetic characterization and biochemical properties. Life Sci. 80:950–958. doi:10.1016/j.lfs.2006.11.024. PMID:17169379.

Buffon, A., Wink, M.R., Ribeiro, B.V., Casali, E.A., Libermann, T.A., Zerbini, L.F.,et al. 2007b. NTPDase and 5= ecto-nucleotidase expression profiles and thepattern of extracellular ATP metabolism in the Walker 256 tumor. Biochim.Biophys. Acta, 1770: 1259–1265. PMID:17574764.

Buffon, A., Casali, E.A., Cardoso, V.V., Zerbini, L.F., Robson, S.C., Sarkis, J.J., andWink,M.R.2010.Differentialexpressionofnucleotidepyrophosphatase/phos-phodiesterases by Walker 256 mammary cancer cells in solid tumors andmalignant ascites. Life Sci. 86: 435–440. doi:10.1016/j.lfs.2010.01.015. PMID:20117116.

Burnstock, G. 2002. Potential therapeutic targets in the rapidly expanding fieldof purinergic signalling. Clin. Med. 2: 45–53. doi:10.7861/clinmedicine.2-1-45.PMID:11871639.

Burnstock, G., Knight, G.E., and Greig, A.V. 2012. Purinergic signaling in healthyand diseased skin. J. Invest. Dermatol. 132: 526–546. doi:10.1038/jid.2011.344.PMID:22158558.

Cappellari, A.R., Vasques, G.J., Bavaresco, L., Braganhol, E., and Battastini, A.M.2012. Involvement of ecto-5=-nucleotidase/CD73 in U138MG glioma cell adhe-sion. Mol. Cell. Biochem. 359: 315–322. doi:10.1007/s11010-011-1025-9. PMID:21858682.

Casali, E.A., de Souza, L.F., Gelain, D.P., Kaiser, G.R., Battastini, A.M., andSarkis, J.J. 2003. Changes in ectonucleotidase activities in rat Sertoli cellsduring sexual maturation. Mol. Cell. Biochem. 247: 111–119. doi:10.1023/A:1024150619490. PMID:12841638.

Contreras-Paredes, A., De la Cruz-Hernández, E., Martínez-Ramírez, I.,Dueñas-González, A., and Lizano, M. 2009. E6 variants of human papilloma-virus 18 differentially modulate the protein kinase B/phosphatidylinositol3-kinase (akt/PI3K) signalingpathway.Virology,383: 78–85.doi:10.1016/j.virol.2008.09.040. PMID:19007961.

de Cremoux, P., de la Rochefordiere, A., Savignoni, A., Kirova, Y., Alran, S.,Fourchotte, V., et al. 2009. Different outcome of invasive cervical cancerassociated with high-risk versus intermediate-risk HPV genotype. In. J. Can-cer, 124: 778–782. doi:10.1002/ijc.24075. PMID:19035459.

Deyrieux, A.F., and Wilson, V.G. 2007. In vitro culture conditions to study kera-tinocyte differentiation using the HaCaT cell line. Cytotechnology, 54: 77–83.doi:10.1007/s10616-007-9076-1. PMID:19003021.

Fang, M., Shen, Z., Huang, S., Zhao, L., Chen, S., Mak, T.W., and Wang, X. 2010.The ER UDPase ENTPD5 promotes protein N-glycosylation, the Warburg ef-fect, and proliferation in the PTEN pathway. Cell, 143: 711–724. doi:10.1016/j.cell.2010.10.010. PMID:21074248.

Feng, L., Sun, X., Csizmadia, E., Han, L., Bian, S., Murakami, T., et al. 2011.Vascular CD39/ENTPD1 directly promotes tumor cell growth by scavenging

extracellular adenosine triphosphate. Neoplasia, 13: 206–216. PMID:21390184.

Hicks-Berger, C.A., Chadwick, B.P., Frischauf, A.M., and Kirley, T.L. 2000. Expres-sion and characterization of soluble and membrane-bound human nucleosidetriphosphate diphosphohydrolase 6 (CD39L2). J. Biol. Chem. 3: 34041–34045.doi:10.1074/jbc.M004723200. PMID:10948193.

Hopfner, M., Maaser, K., Barthel, B., von Lampe, B., Hanski, C., Riecken, E.O.,et al. 2001. Growth inhibition and apoptosis induced by P2Y2 receptors inhuman colorectal carcinoma cells: involvement of intracellular calcium andcyclic adenosine monophosphate. Int. J. Colorectal Dis. 16: 154–166. doi:10.1007/s003840100302. PMID:11459289.

Ivanenkov, V.V., Murphy-Piedmonte, D.M., and Kirley, T.L. 2003. Bacterial ex-pression, characterization, and disulfide bond determination of solublehuman NTPDase6 (CD39L2) nucleotidase: implications for structure and func-tion. Biochemistry, 14: 11726–11735. doi:10.1021/bi035137r. PMID:14529283.

Li, X., Zhou, L., Feng, Y.H., Abdul-Karim, F.W., and Gorodeski, G.I. 2006. The P2X7receptor: a novel biomarker of uterine epithelial cancers. Cancer Epidemiol.Biomarkers Prev. 15: 1906–1913. doi:10.1158/1055-9965.EPI-06-0407. PMID:17035398.

Li, N., Franceschi, S., Howell-Jones, R., Snijders, P.J., and Clifford, G.M. 2011.Human papillomavirus type distribution in 30,848 invasive cervical cancersworldwide: Variation by geographical region, histological type and year ofpublication. Int. J. Cancer, 128: 927–935. doi:10.1002/ijc.25396. PMID:20473886.

Long, H.J., III, Bundy, B.N., Grendys, E.C., Jr., Benda, J.A., McMeekin, D.S.,Sorosky, J., et al.; Gynecologic Oncology Group Study. 2005. Randomizedphase III trial of cisplatin with or without topotecan in carcinoma of theuterine cervix: a Gynecologic Oncology Group Study. J. Clin. Oncol. 23: 4626–4633. doi:10.1200/JCO.2005.10.021. PMID:15911865.

Maldonado, P.A., Negrini, L.A., Ethur Jda, S., Oliveira, L., Correa Mde, C.,Becker, L.V., et al. 2010. Nucleotide degrading enzymes in platelets fromuterine cervical neoplasia patients treated with conization or radiotherapy.Biomed. Pharmacother. 64: 499–504. doi:10.1016/j.biopha.2010.02.007. PMID:20347573.

Maldonado, P.A., Pimentel, V.C., Negrini, L.A., Morsch, V.M., and Schetinger, M.R. 2012.tk;1Role of the purinergic system in patients with cervical intraepithelialneoplasia and uterine cancer. Biomed. Pharmacother. 66: 6–11. doi:10.1016/j.biopha.2011.09.007. PMID:22257696.

Paez, J.G., Recio, J.A., Rouzaut, A., and Notario, V. 2001. Identity between thePCPH proto-oncogene and the CD39L4 (ENTPD5) ectonucleoside triphos-phate diphosphohydrolase gene. Int. J. Oncol. 19: 1249–1254. PMID:11713596.

Read, R., Hansen, G., Kramer, J., Finch, R., Li, L., and Vogel, P. 2009. Ectonucleo-side triphosphate diphosphohydrolase type 5 (Entpd5)-deficient mice de-velop progressive hepatopathy, hepatocellular tumors, and spermatogenicarrest. Vet. Pathol. 46: 491–504. doi:10.1354/vp.08-VP-0201-R-AM. PMID:19176496.

Saraiya, M., Lee, N.C., Blackman, D., Smith, M.J., Morrow, B., and McKenna, M.T.2002. Observations from the CDC. An assessment of Pap smears and hyster-ectomies among women in the United States. J. Womens Health Gend. BasedMed. 11: 103–109. doi:10.1089/152460902753645245. PMID:11975858.

Shi, J., Kukar, T., Wang, C., Li, Q., Cruz, P.E., Davoodi-Semiromi, A., et al. 2001.Molecular cloning and characterization of a novel mammalian endo-apyrase(LALP1). J. Biol. Chem. 276: 17474–17478. doi:10.1074/jbc.M011569200. PMID:11278936.

Spychala, J., Lazarowski, E., Ostapkowicz, A., Ayscue, L.H., Jin, A., andMitchell, B.S. 2004. Role of estrogen receptor in the regulation of ecto-5=-nucleotidase and adenosine in breast cancer. Clin. Cancer Res. 10: 708–717.doi:10.1158/1078-0432.CCR-0811-03. PMID:14760094.

Stagg, J., Beavis, P.A., Divisekera, U., Liu, M.C., Moller, A., Darcy, P.K., andSmyth, M.J. 2012. CD73-deficient mice are resistant to carcinogenesis. CancerRes. 72: 2190–2196. doi:10.1158/0008-5472.CAN-12-0420. PMID:22396496.

Stefan, C., Jansen, S., and Bollen, M. 2006. Modulation of purinergic signaling byNPP-type ectophosphodiesterases. Purinergic Signal. 2: 361–370. doi:10.1007/s11302-005-5303-4. PMID:18404476.

Stella, J., Bavaresco, L., Braganhol, E., Rockenbach, L., Farias, P.F., Wink, M.R.,et al. 2010. Differential ectonucleotidase expression in human bladder cancercell lines. Urol. Oncol. 28: 260–267. doi:10.1016/j.urolonc.2009.01.035. PMID:19372055.

Surviladze, Z., Sterk, R.T., DeHaro, S.A., and Ozbun, M.A. 2013. Cellular entry ofhuman papillomavirus type 16 involves activation of the phosphatidylinosi-tol 3-kinase/Akt/mTOR pathway and inhibition of autophagy. J. Virol. 87:2508–2517. doi:10.1128/JVI.02319-12. PMID:23255786.

Tamajusuku, A.S., Villodre, E.S., Paulus, R., Coutinho-Silva, R., Battasstini, A.M.,Wink, M.R., and Lenz, G. 2010. Characterization of ATP-induced cell death inthe GL261 mouse glioma. J. Cell. Biochem. 109: 983–991. doi:10.1002/jcb.22478. PMID:20069573.

Villar, J., Arenas, M.I., MacCarthy, C.M., Blanquez, M.J., Tirado, O.M., andNotario, V. 2007. PCPH/ENTPD5 expression enhances the invasiveness of hu-man prostate cancer cells by a protein kinase C delta-dependent mechanism.Cancer Res. 67: 10859–10868. doi:10.1158/0008-5472.CAN-07-2041. PMID:18006831.

Wang, L., Zhou, X., Zhou, T., Ma, D., Chen, S., Zhi, X., et al. 2008. Ecto-5=-nucleotidase promotes invasion, migration and adhesion of human breastcancer cells. J. Cancer Res. Clin. Oncol. 134: 365–372. doi:10.1007/s00432-007-0292-z. PMID:17671792.

Wu, X.R., He, X.S., Chen, Y.F., Yuan, R.X., Zeng, Y., Lian, L., et al. 2012. High

Beckenkamp et al. 103

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.

expression of CD73 as a poor prognostic biomarker in human colorectalcancer. J. Surg. Oncol. 106: 130–137. doi:10.1002/jso.23056. PMID:22287455.

Yegutkin, G.G., Marttila-Ichihara, F., Karikoski, M., Niemela, J., Laurila, J.P.,Elima, K., et al. 2011. Altered purinergic signaling in CD73-deficient miceinhibits tumor progression. Eur. J. Immunol. 41: 1231–1241. doi:10.1002/eji.201041292. PMID:21469131.

Zagouri, F., Sergentanis, T.N., Chrysikos, D., Filipits, M., and Bartsch, R. 2012.Molecularly targeted therapies in cervical cancer. A systematic review. Gyne-col. Oncol. 126: 291–303. doi:10.1016/j.ygyno.2012.04.007. PMID:22504292.

Zhang, B. 2012. CD73 promotes tumor growth and metastasis. Oncoimmunol-ogy, 1: 67–70. doi:10.4161/onci.1.1.18068. PMID:22720214.

Zimmermann, H. 1992. 5=-Nucleotidase: molecular structure and functional as-pects. Biochem. J. 285: 345–365. PMID:1637327.

Zimmermann, H.G., Neuneier, R., and Grothmann, R. 2001. Multi-agent model-ing of multiple FX-markets by neural networks. IEEE Trans. Neural Netw. 12:735–743. doi:10.1109/72.935087. PMID:18249909.

104 Biochem. Cell Biol. Vol. 92, 2014

Published by NRC Research Press

Bio

chem

. Cel

l Bio

l. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y Y

OR

K U

NIV

on

06/0

4/14

For

pers

onal

use

onl

y.