Interaction of Calf Thymus DNAwith the Antiviral Drug Lamivudine

Nahid Shahabadi, Maryam Maghsudi, Maryam Mahdavi, and Mehdi Pourfoulad

One approach to accelerate the availability of new cancer drugs is to test drugs approved for other conditions asanticancer agents. In recent years, some researchers have shown that antiviral drugs, such as ritonavir, saqui-navir, and nelfinavir, inhibit the growth of over 60 cancer cell lines derived from nine different tumor types. Thisarticle studied the anticancer potential of an antiviral drug, lamivudine (LA). The interaction of LA and calfthymus DNA (CT-DNA) was studied using emission, absorption, circular dichroism (CD), and viscosity tech-niques. The binding constants evaluated from fluorescence data at different temperatures revealed that fluo-rescence enhancement is a static process that involves complex–DNA formation in the ground state. Further, theenthalpy and entropy of the reaction between the drug and CT-DNA showed DH < 0 ( - 126.38 – 0.61 kJ mol - 1)and DS < 0 ( - 352.17 – 2.1 J mol - 1 K - 1); therefore, van der Waals interactions or hydrogen bonds are the mainforces in the binding of LA to CT-DNA. The values of Kf clearly underscore the high affinity of LA to DNA. Inaddition, detectable changes in the CD spectrum of CT-DNA in the presence of LA indicated conformationalchanges. All these results showed that groove binding is the binding mode of this drug and CT-DNA.

Introduction

The development of anticancer drugs is slow andcostly. One approach to accelerate the availability of

new drugs is to reclassify drugs approved for other condi-tions as anticancer agents. In recent years, some researchershave shown that antiviral drugs, such as ritonavir, saqui-navir, and nelfinavir, inhibit the growth of over 60 cancer celllines derived from nine different tumor types (Gills et al.,2008).

Several nucleoside analogs, including lamivudine (LA)(Fig. 1), are used as antiviral drugs. The viral polymeraseincorporates these compounds with noncanonical bases.These compounds are activated in the cells by being con-verted into nucleotides; they are administered as nucleo-sides, because charged nucleotides cannot easily cross cellmembranes.

LA has become a main therapeutic option for treatinghepatitis B virus (HBV) infection ( Jonas et al., 2002). Its an-tiviral effects against HBV have been established both in vitroand in vivo (Mckenzie et al., 1995). Clinical trials revealed thatLA is effective in reducing HBV replication and in prevent-ing the progression of chronic liver disease (Liaw et al., 2004;Asselah et al., 2005).

Many antitumor and antiviral drugs and many carcino-gens act by binding within the minor groove of double-helical DNA, interfering with both replication and transcription.

Several of these are quite base specific, recognizing andbinding only to certain base sequences.

In this study, the interaction of LA with calf thymus DNA(CT-DNA) was investigated by (i) molecular spectroscopymethods, including ultraviolet (UV) spectrophotometry (Bi-ver et al., 2004), fluorescence (Liu et al., 1998), and circulardichroism (CD) spectropolarimetry (Mahadevan and Pala-niandavar, 1997, 1998), and (ii) dynamic viscosity measure-ments (Mitsopoulou et al., 2008).

Materials and Methods

Tris–HCl was purchased from Merck. Doubly distilleddeionized water was used throughout. Highly polymerizedCT-DNA was purchased from Sigma. LA was purchasedfrom NC IndTech Pvt. Ltd (India).

Absorbance spectra were recorded using an HP spec-trophotometer (Agilent 8453) equipped with a thermo-stated bath (Huber polysat cc1). Absorption titrationexperiments were conducted by keeping the concentrationof LA constant (5 · 10 - 5 M) while varying the DNA con-centration from 0 to 3 · 10 - 4 M (ri = [DNA]/[LA] = 0.0, 0.2,0.5, 0.75, 1, 1.5, 2, 3, 3.5, 4, 5, and 6). Absorbance valueswere recorded after each successive addition of DNAsolution, followed by an equilibration period. CD mea-surements were recorded on a JASCO ( J-810) spectro-polarimeter, keeping the concentration of DNA constant

Department of Inorganic Chemistry, Faculty of Chemistry, Razi University, Kermanshah, Iran.

DNA AND CELL BIOLOGYVolume 31, Number 1, 2012ª Mary Ann Liebert, Inc.Pp. 122–127DOI: 10.1089/dna.2011.1228

122

(8 · 10 - 5 M) while varying the drug concentration (ri =[LA]/[DNA] = 0, 0.05, 0.1, 0.15, and 0.2).

Viscosity measurements were made using a viscosimeter(SCHOT AVS 450) maintained at 25.0�C – 0.5�C using aconstant temperature bath. The DNA concentration wasfixed at 5 · 10 - 5 M while varying the LA concentration (ri =[DNA]/[LA] = 0.0, 0.1, 0.3, 0.6, and 0.9), and the flow timewas measured with a digital stopwatch. The mean valuesof three measurements were used to evaluate the viscosity(Z) of the samples. The values for relative specific viscosity(Z/Z0)1/3, where Z0 and Z are the specific viscosity contri-butions of DNA in the absence (Z0) and presence of the drug(Z), respectively, were plotted against ri.

All fluorescence measurements were carried out with aJASCO spectrofluorometer (FP6200) by keeping the concen-tration of LA constant while varying the DNA concentrationfrom 0 to 30 · 10 - 5 M (ri = [DNA]/[LA] = 0.0, 0.5, 1.0, 1.5, 2, 3,4, 5, and 6) at four different temperatures (279, 293, 310, and318 K).

Standard deviations of measurements were calculatedfrom linear regression analyses as indicated by Diem andLentner (1970).

Results and Discussion

UV–visible spectroscopy

The application of electronic absorption spectroscopy inDNA binding studies is a useful technique. Figure 2 showsthe absorption spectra of LA in the absence and presence ofCT-DNA. Upon the addition of CT-DNA, the absorption

band of LA at about 272 nm shows hyperchromism and isaccompanied by a shift of 11 nm in lmax, from 272 to 261 nm,consistent with groove binding and leading to a small per-turbation. This hyperchromism can be attributed to externalcontact (surface binding) with the duplex. Other studies haveobserved similar hyperchromism (Kashanian et al., 2007; Xuet al., 2008).

Groove-binding molecules typically have unfused aro-matic ring systems linked by bonds with torsional freedomfor the molecules to adopt an appropriate conformation thatclosely matches the helical turn of DNA grooves (Strekowskiand Wilson, 2007).To illustrate this further, the intrinsicbinding constant, Kb, which indicates the binding strength ofthe LA with CT-DNA was determined from the spectral ti-tration data using the following equation (Pyle et al., 1989):

[DNA]=(ea� ef)¼ [DNA]=(eb� ef)þ 1=Kb(eb� ef) ð1Þ

where [DNA] is the concentration of DNA, and ef, ea, and eb

correspond to the extinction coefficient for the free LA, foreach addition of DNA to the drug, and for the drug in thefully bound form, respectively. A plot of [DNA]/(ea – ef)versus [DNA] gives Kb as the ratio of slope to the intercept.From the [DNA]/(ea–ef) versus [DNA] plot (Fig. 3), thebinding constant Kb for the drug was estimated to be(5 – 0.3) · 104 M - 1. Kb is lower than that reported for classicalintercalators (for ethidium bromide, Kb& 107 M - 1) (Coryet al., 1985). The observed binding constant is more inkeeping with groove binding, as observed in the literature(Erikson et al., 1992; Rajamanickam et al., 2000; Vaidyanathanand Nair, 2003; Lu et al., 2007).

Viscometric studies

To clarify further the interaction between the drug andDNA, viscosity measurements were carried out. Opticalphotophysical probes provide necessary, but not sufficient,clues to support a binding model. Hydrodynamic measure-

FIG. 2. Electronic absorption spectra for the titration of5.0 · 10 - 5 M LA with DNA (ri = 0.00, 0.5, 0.75, 1, 1.5, 2, 2.5, 3,3.5, 4, 5, and 6).

FIG. 1. The structure of LA. LA, lamivudine.

0

0.5

1

1.5

2

2.5

0 2 4 6 8

[DNA]¥108 M

[D

NA

] / (

e a-

e f) ×

105

FIG. 3. Plot of [DNA]/(ea – ef) versus [DNA] for the ab-sorption titration of CT-DNA with LA in Tris–HCl buffer.CT-DNA, calf thymus DNA.

ANTIVIRAL DRUGS AS ANTICANCER AGENTS 123

ments that are sensitive to the length change (i.e., viscosityand sedimentation) are regarded as the least ambiguous testsof a binding model in the absence of crystallographic struc-tural data. Intercalating agents are expected to elongate thedouble helix to accommodate the ligands in between thebases, leading to an increase in the viscosity of DNA. Incontrast, complexes that exclusively bind in the DNAgrooves by partial and/or nonclassical intercalation underthe same conditions typically cause less pronounced or nochange in DNA solution viscosity (Kelly et al., 1985). Thevalues of (Z/Z0)1/3 were plotted against [LA]/[DNA] (Fig.4). The results revealed that the drug effect causes an in-crease in DNA viscosity, which is consistent with the DNAgroove binding suggested above (Selvi and Palaniandavar,2002; Xi et al., 2009).

Fluorescence studies

As LA is luminescent in the absence of DNA, it shows anappreciable increase in emission upon the addition of CT-DNA (Fig. 5). Figure 5 shows that a regular increase in thefluorescence intensity of LA with a shift in fluorescenceemission maximum (612–613 nm) took place upon increasingthe concentration of DNA at 25.0�C at a pH of 7.2. Thesefluorescence enhancements indicate that the LA interactedwith DNA and that the quantum efficiency of LA was in-creased. Similar to a quenching process, the enhancementconstant can be determined using Eq. 2 (Shahabadi et al.,2009):

F0

F¼ 1�KE[E] (2)

If a dynamic process is part of the enhancing mechanism,Eq. 2 can be written as follows (Shahabadi et al., 2009):

F0

F¼ 1�KD[E]¼ 1�KBs0[E] (3)

where KD is the dynamic enhancement constant (similar to adynamic quenching constant), KB is the bimolecular en-hancement constant (similar to a bimolecular quenchingconstant), and t0 is the lifetime of the fluorophore in theabsence of the enhancer. The dynamic enhancement con-stants of LA at different temperatures were calculated usingEq. 3 (Table 1). As fluorescence lifetimes are typically near10 - 8 s, the bimolecular enhancement constant (KB) was cal-culated from KD = KBt0 (Table 1). By considering the equiv-alency of the bimolecular quenching and enhancementconstants, it can be seen that the latter is greater than thelargest possible value (1.0 · 1010 M - 1 s - 1) in an aqueousmedium. Thus, as a dynamic process does not initiate thefluorescence enhancement, it suggests that a static processinvolves complex formation in the ground state (Shahabadiet al., 2009).

Equilibrium binding titration

The binding constant (Kf) and the binding stoichiometry(n) for the complex formation between LA and DNA weremeasured using Eq. 4 (Eichhorn and Shin, 1968; Kennedyand Bryant, 1986):

Log (F0� F)=F¼ log Kf þ n log [DNA] (4)

FIG. 4. Effect of increasing amounts of LA on the viscosityof CT-DNA (5 · 10 - 5 M) in 10 mM Tris–HCl buffer (ri = 0.1,0.3, 0.6, and 0.9).

FIG. 5. Emission spectra of LA in Tris–HCl buffer in theabsence and presence of CT-DNA. ri = [DNA]/[drug] = 0.0,0.5, 1, 1.5, 2, 3, 4, 5, and 6.

Table 1. Dynamic Enhancement and Bimolecular

Enhancement Constants for the Interactions

Between Lamivudine and Calf Thymus DNAat Different Temperatures

KB KD Linear equation R2Temperature

(K)

13.24 · 1012 1.32 · 104 Y = 0.973–13,240X 0.92 27912.40 · 1012 1.24 · 104 Y = 0.963–12,405X 0.96 2939.98 · 1011 0.99 · 104 Y = 0.776–99,786X 0.98 3108.54 · 1011 0.85 · 104 Y = 0.631–8540X 0.95 318

124 SHAHABADI ET AL.

Here, F0 and F are the fluorescence intensities of thefluorophore in the absence and presence of different con-centrations of DNA, respectively.

Table 2 shows the linear equations of log (F–F0)/F versuslog [DNA] at different temperatures. The values of Kf un-derscore the remarkably high affinity of LA to DNA.

Thermodynamic studies

According to the thermodynamic data, interpreted asfollows, the model of the interaction between a drug andbiomolecule may be (1) DH > 0 and DS > 0, hydrophobic for-ces; (2) DH < 0 and DS < 0, van der Waals interactions andhydrogen bonds; and (3) DH < 0 and DS > 0, electrostatic in-teractions (Shahabadi and Fatahi, 2010). To elucidate theinteraction of our complex with DNA, the thermodynamicparameters were calculated. The plot of ln Kf versus 1/T (Fig.6; Eq. 5) allows the determination of DH and DS. If thetemperature does not vary significantly, the enthalpy changecan be regarded as a constant. Based on the binding con-stants at different temperatures, Eqs. 5 and 6 can estimate thefree energy change:

LnK¼ � DH

RTþ DS

Rð5Þ

DG¼DH�TDS¼ �RTLnK ð6Þ

We found that DH < 0 and DS < 0; therefore, van der Waalsinteractions or hydrogen bonds are the main forces in thebinding of the LA to CT-DNA. (Table 3).

CD spectral studies

CD spectral techniques give us useful information on howthe binding of the metal complex to DNA influences theconformation of DNA. The observed CD spectrum of CT-DNA consists of a positive band at 275 nm due to basestacking and a negative band at 245 nm due to helicity,which is characteristic of DNA in the right-handed B form( Johnson, 1994). Although groove binding and electrostaticinteraction of small molecules with DNA show little or noperturbations on the base stacking and helicity bands, in-tercalation enhances the intensities of both the bands, stabi-lizing the right-handed B conformation of CT-DNA. Figure 7shows the CD spectra of DNA taken after incubation of thedrug with CT-DNA.

The intensities of both the negative and positive bandssignificantly decrease (shifting to zero levels). This suggeststhat the DNA binding of the drug induces conformationalchanges, including the conversion from a more B-like to amore C-like structure within the DNA molecule (Mahadevanand Palaniandavar, 1998). These changes are indicative of anonintercalative mode of binding of this drug and offersupport of its groove binding nature (Maheswari andPalaniandavar, 2004).

Conclusion

The antiviral drug LA exhibits high binding affinity forCT-DNA. Different instrumental methods were used to in-vestigate the interaction mechanism. The results support thenotion that the drug can bind to CT-DNA. The absorption

Table 2. Linear Equations of Log (F - F0)/F Versus

Log [DNA] and Kf of Lamivudine with DNAat Different Temperatures

Temperature (K) Linear equation R2 Kf Log Kf

279 Y = 1.21X + 5.12 0.942 1.32 · 105 5.12293 Y = 1.02X + 4.27 0.92 1.86 · 104 4.27310 Y = 0.78X + 3.26 0.975 1.82 · 103 3.26318 Y = 0.51X + 2.02 0.953 1.04 · 102 2.02

0

1

2

3

4

5

6

3.1 3.2 3.3 3.4 3.5 3.6 3.7

1000 1/T

Ln

K

FIG. 6. Van’t Hoff plot for the interaction of LA and CT-DNA at pH 7.2.

Table 3. Thermodynamic Parameters

and Binding Constants for the Binding

of Lamivudine to Calf Thymus DNA

T (K)Log

K(M - 1)DGo

(kJ mol - 1) DHo (kJ mol - 1)DSo

( J mol - 1 K - 1)

279 5.12 - 28.12 – 1.1 - 126.377 – 0.61 - 352.173 – 2.1293 4.27 - 23.19 – 0.5 - 126.377 – 0.61 - 352.173 – 2.1310 3.26 - 17.21 – 0.8 - 126.377 – 0.61 - 352.173 – 2.1318 2.02 - 14.38 – 1.2 - 126.377 – 0.61 - 352.173 – 2.1

FIG. 7. Circular dichroism spectra of DNA (8.0 · 10 - 5) in10 mM Tris–HCl buffer, in the presence of increasingamounts of LA (ri = [drug]/[DNA] = 0.0, 0.05, 0.1, 0.15,and 0.2).

ANTIVIRAL DRUGS AS ANTICANCER AGENTS 125

spectrum of the drug shows that as the concentration ofDNA increases, a large degree of hypochromism develops inthe spectrum. This hyperchromism can be attributed to ex-ternal contact (surface binding) with the duplex. The ob-served binding constant (Kb = 5 · 104 M - 1) is in keeping withgroove binding. The fluorescence studies showed an appre-ciable increase in the drug emission upon addition of DNA.The positive slope in the Van’t Hoff plot indicates thatthe reaction between LA and DNA was enthalpy favored( - 126.377 kJ mol - 1). CD results showed deep conforma-tional changes in the CT-DNA double helix upon bindingwith the drug. The results of viscosimetry revealed that thedrug effects show a relative increase in DNA viscosity,which is consistent with DNA groove binding. This studyis expected to provide greater insight into the use of an-tiviral drugs as anticancer drugs; further studies are inprogress.

Acknowledgment

Financial support from the Razi University ResearchCenter is gratefully acknowledged.

Disclosure Statement

No competing financial interests exist.

References

Asselah, T., Ripault, M.P., and Castelnau, C. (2005). The currentstatus of antiviral therapy of chronic hepatitis B. J Clin Viral34, S115–S124.

Biver, T., Secco, F., Tine, M.R., and Venturini, M. (2004). Kineticsand equilibria for the formation of a new DNA metal-inter-calator: the cyclic polyamine Neotrien/copper (II) complex. JInorg Biochem 98, 33–40.

Cory, M., Mckee, D.D., Kagan, J., Henry, D.W., and Miller, J.A.(1985). Synthesis and DNA binding properties of bifunctionalintrecalators. Comparison of polymethylene and diphenylether chains connecting phenanthridine. J Am Chem Soc 107,

2528–2536.Diem, K., and Lentner, C. (1970). Scientific Tables, 7th ed. (Ciba-

Geigy, Basal, Switserland).Eichhorn, G.L., and Shin, Y.A. (1968). Interaction of metal ions

with polynucleotides and related compounds. XII. The relativeeffect of various metal ions on DNA helicity. J Am Chem Soc90, 7323–7328.

Erikson, M., Leijon, M., Hirot, C., Norden, B., and Graeslud, A.(1992). Minor groove binding of [Ru(phen)3]2 + to[d(CGCGATCGCG)]2 evidenced by two-dimensional NMR. JAm Chem Soc 114, 4933–4934.

Gills, J.J., Lopoiccolo, J., and Dennis, P.A. (2008). Nelfinavir anew anti-cancer drug with pleiotropic effects and many pathsto autophagy. Autophagy 4, 107–109.

Johnson, W.C. (1994). Circular Dichroism: Principles and Applica-tions. K. Nakanishi, N. Berova, and R.W. Woody, eds. (VCH,New York), pp. 523–540.

Jonas, M.M., Kelley, D.A., Mizerski, J., Badia, I.B., Areias, J.A.,Schwarz. K. B., Little, N.R., Greensmith, M.J., Gardner, S.D.,Bell, M.S., and Sokal, E.M. (2002). Clinical trial of lamivudine inchildren with chronic hepatitis B. N Engl J Med 346, 1706–1713.

Kashanian, S., Gholivand, M.B., Ahmadi, F., Taravati, A., andHossinzadeh Colagar, A. (2007). DNA interaction with Al-

N,N’-bis(salicylidene)2,2’-phenylendiamine complex. Spectro-chim Acta A 67, 472–478.

Kelly, T.M., Tossi, A.B., McConnell, D.J., and Strekas, T.C.(1985). A study of the interactions of some poly-pyridylruthenium(II) complexes with DNA using fluores-cence spectroscopy, topoisomerisation and thermaldenaturation. Nucleic Acids Res 13, 6017–6034.

Kennedy, S.D., and Bryant, R.G. (1986). Manganese-deoxyribonucleic acid binding modes. Nuclear magnetic re-laxation dispersion results. Biophys J 50, 669–676.

Liaw, Y.F., Sung, J.J.Y., and Chow, W.C., Farrell, G., Lee, C.-Z.,Yuen, H., Tanwandee, T., Tao, Q.-M., Shue, K., Keene, O.N.,Dixon, J.S., Gray, D.S., and Sabbat, J. (2004). Lamivudine forpatients with chronic hepatitis B and advanced liver disease.N Engl J Med 351, 1521–1531.

Liu, C., Zhou, J., and Xu, H. (1998). Interaction of the copper(II)macrocyclic complexes with DNA studied by fluorescencequenching of ethidium. J Inorg Biochem 71, 1–6.

Lu, H.L., Liang, J.J., Zeng, Z.Z., Xi, P.X., Liu, X.H., Chen, F.J.,and Xu, Z.H. (2007) Three salicylaldehyde derivative Schiffbase Zn(II) complexes: synthesis, DNA binding and hy-droxyl radical scavenging capacity Trans Met Chem 32,

564–569.Mahadevan, S., and Palaniandavar, M. (1997). Spectroscopic and

voltammetric studies of copper (II) complexes of bis(pyrid-2-yl)-di/trithia ligands bound to calf thymus DNA. Inorg ChimActa 254, 291–302.

Mahadevan, S., and Palaniandavar, M. (1998). Spectral andelectrochemical behavior of copper (ii) - phenanthrolinesbound to calf thymus DNA. [(5, 6-dimethyl-OP)2Cu]2 + (5,6-dimethyl-OP = 5,6-dimethyl-1,10-phenanthroline) induces aconformational transition from B to Z DNA. Inorg Chem 37,

693–700.Maheswari, P.U., and Palaniandavar, M. (2004). DNA binding

and cleavage properties of certain tetrammine ruthenium (II)complexes of modified 1,10-phenanthrolines–effect of hydro-gen-bonding on DNA-binding affinity. J Inorg Biochem 98,

219–230.McKenzie, M.D., Fried, M.W., Sallie, R., Conjeevaram, H., Di

Bisceglie, A.M., Park, Y., Savarese, B., Kleiner, D., Tsokos, M.,Luciano, C., Pruett, T., Stotka, J.L., Straus, S.E., and Hoofnagle,J.H. (1995). Hepatic failure and lactic acidosis due to fialur-idine (FIAU), an investigational nucleoside analogue forchronic hepatitis B. N Engl J Med 333, 1099–1105.

Mitsopoulou, C.A., Dagas, C.E., and Makedonas, C. (2008).Characterization and DNA interaction of the Pt(II)(pq)(bdt)complex: a theoretical and experimental research. Inorg ChimActa 361, 1973–1982.

Pyle, A.M., Rehmann, J.P., Meshoyrer, R., Kumar, C.V., Turro,N.J., and Barton, J.K. (1989). Mixed-ligand complexes of ru-thenium (ii): factors governing binding to DNA. J Am ChemSoc 111, 3051.

Selvi, P.T., and Palaniandavar, M. (2002). Spectral, viscometricand electrochemical studies on mixed ligand cobalt (III)complexes of certain diimine ligands bound to calf thymusDNA. Inorg Chim Acta 337, 420–428.

Shahabadi, N., and Fatahi, A. (2010). Multispectroscopic DNA-binding studies of a tris-chelate nickel (II) complex containing4, 7-diphenyl 1, 10-phenanthroline ligands. J Mol Struct 970,

90–95.Shahabadi, N., Kashanian, S., and Darabi, F. (2009). In vitro

study of DNA interaction with a water-soluble dinitrogenschiff base. DNA Cell Biol 28, 589–596.

126 SHAHABADI ET AL.

Strekowski, L., and Wilson, B. (2007). Noncovalent interactionswith DNA: an overview. Mutat Res 3, 623.

Vaidyanathan, V.G., and Nair, B.U. (2003). Photooxidation of DNAby cobalt (II) tridentates complex. J Inorg Biochem 94, 121–126.

Xi, P.X., Xu, Z., Liu, X.H., Chen, F.J., Zeng, Z.Z., Zhang, X.W.,and Liu. Y. (2009). Synthesis, characterization, antioxidantactivity and DNA-binding studies of three rare earth (iii)complexes with 1-(4-aminoantipyrine)-3-tosylurea ligand. JFluoresc 19, 63–72.

Xu, Z.H., Chen, F.J., Xi, P.X., Liu, X.H., and Zeng, Z.Z. (2008).Synthesis characterization and DNA-binding properties of thecobalt (II) and nickel (II) complexes with salicylaldehyde 2-phenylquinoline-4-carboylhydrazone. J Photochem PhotobiolA: Chem 196, 77–83.

Address correspondence to:Nahid Shahabadi, M.Sc., Ph.D.

Department of Inorganic ChemistryFaculty of Chemistry

Razi UniversityKermanshah 74155

Iran

E-mail: nahidshahabadi@yahoo.com

Received for publication January 18, 2011; received in re-vised form May 7, 2011; accepted May 10, 2011.

ANTIVIRAL DRUGS AS ANTICANCER AGENTS 127