Bleomycin has antiviral properties against drug-resistant HIV strains and sensitises virus to currently used antiviral agents

International Journal of Antimicrobial Agents 27 (2006) 63–68

Bleomycin has antiviral properties against drug-resistant HIV strains and sensitises virus to currently used antiviral agents Niki A. Georgiou a , Tjomme van der Bruggen a , Derek M.C. Healy a,1 , Carla van Tienen a , Prim de Bie a,2 , Maroeska Oudshoorn a,3 , Joannes J.M. Marx a , B. Sweder van Asbeck b,∗ a

Eijkman-Winkler Center for Microbiology, Infectious Diseases and Inflammation, University Medical Center Utrecht, Utrecht, P.O. Box 85500, 3508 GA Utrecht, The Netherlands b Department of Internal Medicine, Room F02.126, University Medical Center Utrecht, Utrecht, P.O. Box 85500, 3508 GA Utrecht, The Netherlands Received 27 July 2005; accepted 3 September 2005

Abstract In this study we performed phenotypic assays to assess involvement of the cancer chemotherapeutic agent bleomycin (BLM) in replication inhibition of mutant HIV-1 viral strains. Three clinically relevant mutant HIV variants, including one containing the Q151M mutation conferring multinucleoside resistance, were equally as sensitive to BLM as the wild-type HXB2 strain. Long-term incubation of BLM with a wild-type HIVBa-L strain did not alter the sensitivity of the strain to BLM (IC50 of BLM 0.64 ␮M at the beginning of incubation to 0.58 ␮M). At the same point in time, resistance to lamivudine (3TC) and zidovudine (AZT) was noted. Interestingly, the BLM-treated virus showed hypersensitivity to both AZT and 3TC. Our results suggest a contribution of BLM in viral load reduction in patients receiving both anticancer and antiviral agents and harbouring both wild-type and resistant HIV strains. © 2005 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. Keywords: Bleomycin; HIV-1 replication inhibition; Mutant strains

1. Introduction The chemotherapeutic agent bleomycin (BLM) is used to treat various cancers, including Kaposi’s sarcoma and HIV-related Hodgkin’s lymphoma. Combination cancer chemotherapy and antiviral therapy in patients with HIVrelated lymphoma has been shown to be safe [1–3].

∗ Corresponding author. Tel.: +31 30 250 9111/7381; fax: +31 30 252 3741. E-mail address: [email protected] (B.S. van Asbeck). 1 Present address: Rabies & Wildlife Zoonoses Group, Virology Department, Veterinary Laboratories Agency – Weybridge, Woodham Lane, New Haw Addlestone, Surrey KT15 3NB, UK. 2 Present address: Department of Biomedical Genetics, University Medical Center Utrecht, Room KC.02.069.1, Lundlaan 6, 3584 EA Utrecht, The Netherlands. 3 Present address: Genmab, P.O. Box 85199, 3508 AD Utrecht, The Netherlands.

Pre-clinical development of novel drugs or study of the potential use of existing drugs with antiviral properties include an assessment of the in vitro drug susceptibility of highly drug-resistant HIV-1 isolates. In a recent study, some clinically relevant drug-resistant viruses, including one containing the Q151M mutation, were found to be resistant to a group of new experimental compounds in clinical development (seven reverse transcriptase and seven protease inhibitors) [4]. With such information at hand, the suitability of newly developed drugs can be predicted before they reach the patient. Previously we have reported the antiviral properties of BLM in HIV-1-infected human peripheral blood lymphocytes (PBLs) and macrophages [5]. BLM was also found to inhibit synergistically in vitro HIV-1 replication in a two-drug combination with either ritonavir (RTV), indinavir or zidovudine (AZT) [6]. In the current study, we applied current phenotypic assays used for the assessment of potential antiviral agents to assess the behaviour of BLM, an established anticancer agent with antiviral properties. Given

0924-8579/$ – see front matter © 2005 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2005.09.008

64

N.A. Georgiou et al. / International Journal of Antimicrobial Agents 27 (2006) 63–68

the fact that BLM is often included in the therapeutic regimen against HIV-related lymphomas, we first assessed the sensitivity of mutant HIV strains to this drug. The drug susceptibility of BLM was assessed in a range of clinically relevant drug-resistant viruses by infecting PBLs with either an isolate containing the Q151M mutation [7,8], a multiple resistant protease strain [9,10], or a prevalent AZT/lamivudine (3TC)-resistant isolate. All of the recombinant viral strains were derived from the HXB2-D wild-type and were sensitive to BLM. In vitro selection of mutations and characterisation of drug-resistant viral variants is a useful means of identifying the genetic and biochemical mechanisms of resistance to specific compounds as well as determining the time course for the development of resistance in the presence of an antiviral agent. We performed a conventional in vitro selection model to test a BLM-treated wild-type strain for its susceptibility to AZT, 3TC and RTV and the susceptibility of the AZT- and 3TC-treated strains to BLM. This model used peripheral blood mononuclear cells (PBMCs) and was run for 18 weeks; in parallel, long-term incubations with either AZT or 3TC were also performed as controls. At the time of onset of resistance to AZT and 3TC, the wild-type HIV strain remained sensitive to BLM. Interestingly, the BLMtreated virus became hypersensitive to AZT and 3TC and, although not significant, the corresponding 3TC- and AZTtreated strains were more susceptible to BLM than the wildtype.

2. Materials and methods 2.1. Compounds BLM was obtained from Asta Medica (Brussels, Belgium), RTV from Abbott Laboratories (Baar, Switzerland), AZT from Sigma Chemical Co. (St Louis, MO) and 3TC from GlaxoSmithKline (Hertfordshire, UK). BLM was dissolved in aliquots in phosphate-buffered saline and stored at −20 ◦ C until further use. AZT, 3TC and RTV were dissolved in dimethyl sulphoxide (DMSO) at a concentration of 20 mM or higher. Working concentrations of DMSO were <0.001%. 2.2. Cell isolations PBMC fractions were isolated from heparinised blood from HIV-1, HIV-2 and hepatitis B negative donors (Blood Bank, Utrecht, The Netherlands) by Ficoll-Isopaque gradient separation. Five to six donor PBMC fractions were pooled and the PBMC batch was then stored in aliquots in liquid nitrogen or in a −130 ◦ C freezer. When cells were needed, aliquots were thawed and cultured in RPMI1640 medium (Life Technologies Ltd., Paisley, UK) supplemented with 10% heat-inactivated foetal bovine serum (FBS) (Life Technologies Ltd.) and 10 ␮g/mL gentamicin (Life Technologies Ltd.). After 2 ␮g/mL phytohaemagglu-

tinin (PHA) stimulation for 4 days, the cells were washed twice to remove PHA and were further cultured in medium containing 10 U/mL human recombinant interleukin-2 (IL-2) (Boehringer, Mannheim, Germany). At the beginning of the experiment, viability of the cells was >95% as determined by trypan blue exclusion. When needed, PBLs were isolated freshly from buffy coats from HIV-1 seronegative donors by allowing the monocyte fraction of the PBMCs to adhere on fibronectin-coated flasks before the PBL fraction was harvested. PBLs were stimulated to proliferate for 3 days with 4 ␮g/mL PHA (Sigma Chemical Co.) in RPMI-1640 (Life Technologies Ltd.) medium supplemented with 10% heat inactivated FBS (Life Technologies Ltd.) and 10 ␮g/mL gentamicin. After PHA stimulation, PBLs were cultured in medium containing 10 U/mL human recombinant IL-2 (Boehringer). All incubations were carried out in flat-bottomed 96-well plates at 37 ◦ C, 5% CO2 and 95% air. 2.3. p24 measurements Virus in culture supernatant was inactivated in a final concentration of 0.05% empigen (Calbiochem-Novabiochem Co., La Jolla, CA) and by heating at 56 ◦ C for 30 min. The presence of HIV-1 in the inactivated supernatant was monitored by measuring the p24 core antigen levels by enzyme-linked immunosorbent assay (ELISA) as previously described [11]. 2.4. Viral stock production and 50% tissue culture infectious dose (TCID50 ) measurements For the drug susceptibility assays, the full-length molecular clone pHXB2 was used as the wild-type background clone for introduction of the different mutations. Virus stocks of the HXB2 virus and HXB2-derived mutant strains were prepared by infecting SupT1 cells. After observation of abundant syncytia, the supernatants were harvested and stored in aliquots at −80 ◦ C. HIV-1Ba-L stocks were prepared by infecting PBLs, harvesting the supernatant after 10–14 days and storing at −80 ◦ C until use. Virus titres (TCID50 /mL) were calculated in stimulated PBMCs infected for 7 days as previously described [12]. In short, the virus stock was titrated by use of a serial four-fold dilution in stimulated PBMCs. The cells and virus were cultured for 7 days at 37 ◦ C in a humidified atmosphere in 5% CO2. Virus production was then measured with a p24 ELISA as previously described [12]. Finally, TCID50 was calculated according to Reed and Muench [13]. Mutations in all three strains were confirmed by sequence analysis. 2.5. Mutant strains used for drug susceptibility assay 2.5.1. 41L + 184V + 215Y The mutations were generated by site-directed mutagenesis using the Muta-Gene M13 in vitro mutagenesis kit

N.A. Georgiou et al. / International Journal of Antimicrobial Agents 27 (2006) 63–68

(Bio-Rad, Hercules, CA) (strain kindly supplied by Dr W. Keulen, Utrecht, The Netherlands). The phenotype of this strain was confirmed by drug susceptibility assays (data not shown). 2.5.2. R460.6 clone This multiple resistance protease clone was isolated from a patient treated with RTV [9]. Cloned-amplified RNA from patient serum was used to generate this recombinant protease-resistant virus by introducing viral protease sequences derived from serum into a protease-deleted HIV-1 clone (HXB2
pro) by homologous recombination [9,10]. 2.5.3. HIV-1GW980 This multiple resistant isolate possesses the complete Q151M multinucleoside resistance complex [7] and confers partial resistance to at least four different nucleoside analogues (AZT, ddI, ddC and d4T). This is the first mutation to develop in patient HIV-1 strains developing the other multidrug resistance mutations [8] (kindly supplied by Dr R.W. Shafer, Stanford, CA). 2.6. Drug susceptibility assay Stimulated PBMCs were infected for 2 h with either the wild-type HXB2 strain or the mutant strains 41L + 184V + 215Y, R460.6 or HIV-1GW980 at a multiplicity of infection (MOI) of 0.001 at 37 ◦ C. The cells were then washed twice to remove excess virus and subsequently incubated with various concentrations of either BLM, AZT or 3TC. After 7 days compound incubations, supernatant samples were collected for p24 ELISA measurements and the drug concentration achieving a p24 inhibition of 50% (IC50 ) for each drug was determined using the computer program CalcuSyn [14,15].

65

the drug did not allow concentrations higher than 1 ␮M. At every new passage the virus was also incubated with the drug concentration of the previous passage as a back-up. On average, cells and drugs were refreshed every 2 weeks. This in vitro model was run for 18 weeks; nine passages were performed and the final concentrations of AZT and 3TC reached were 160 nM and 960 nM, respectively. 2.8. Propagation of virus and determination of IC50 After 18 weeks of exposure to either BLM, AZT or 3TC, the treated and untreated (wild-type control) viruses were collected. The treated and untreated HIVBa-L viruses were propagated in PBLs, as described in Section 2.4. The TCID50 of the treated and untreated viruses were measured by titration in PBLs to calculate the infectivity of the different virus batches. Once the TCID50 was known for all treated and untreated strains, it was possible to standardise for the amount of infectious dose of each treated virus. Freshly isolated and stimulated PBLs were infected with each strain at a MOI of 0.005 for 2 h before excess virus was washed away. Subsequent drug incubations were performed with either BLM, 3TC or AZT in all different virus batches (treated and untreated) for 7 days before p24 samples were collected and the IC50 of each drug in each virus batch was measured using the computer program CalcuSyn [14,15]. 2.9. Cellular viability Simultaneously to the drug susceptibility assays, cellular viability was monitored after incubation with the various drugs by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (Sigma Chemical Co.) assay [17].

3. Results 2.7. In vitro selection model 3.1. Drug susceptibility to BLM of mutant HIV strains Long-term incubations of wild-type virus with either 3TC, AZT or BLM were run simultaneously for 18 weeks. Resistance to 3TC is known to arise within a few weeks in vitro [16]. Stimulated PBMCs were infected with the HIV-1Ba-L virus for 2 h at 37 ◦ C, 5% CO2 at a MOI of 0.001. After washing excess virus away, the cells were seeded in 24-well plates at a concentration of 0.5 × 106 /mL (1 mL per well) and BLM was added at the IC50 concentration of 0.5 ␮M. 3TC and AZT were added at 1× their IC50 concentrations, which corresponds to 20 nM for AZT and 30 nM for 3TC. No drugs were added to two wells as virus controls. p24 production was checked to determine whether passage of the virus was possible (threshold, 2000 pg p24/mL). With every passage, freshly stimulated PBMCs were infected with supernatant (containing the virus) from the previous passage, and medium and cells from the previous passage were stored at −80 ◦ C. During selection, the drug concentration of 3TC and AZT was increased with each new passage, but not BLM. Toxicity of

Recombinant HIV-1 variants carrying drug-resistance mutations were employed to test the susceptibility of these strains to BLM in PBMCs. All of the recombinant viral strains were derived from the wild-type strain HXB2-D. The sensitivities of each mutant strain to BLM in PBMCs are shown in Table 1. The mean IC50 values ± S.E.M. are shown and, as can be seen in Table 1, the IC50 values of BLM in the three resistant HIV-1 strains tested were not statistically different from the wild-type HXB2-D strain (IC50 = 0.49 ␮M), suggesting that all mutant strains investigated remained sensitive to the antiviral effects of BLM. 3.2. In vitro selection model When HIV-1Ba-L was treated with either 3TC or AZT for a period of 18 weeks, the virus became resistant to both drugs. In the same setting, virally infected cells were also incubated

N.A. Georgiou et al. / International Journal of Antimicrobial Agents 27 (2006) 63–68

66

Table 1 Effect of bleomycin (BLM) on recombinant drug-resistant strains of HXB2-D in peripheral blood mononuclear cells Recombinant variant

Resistance

BLM (␮M)a

Significance (P-value)b

HXB2-D R460.6 41L + 184V + 215Y HIV-1GW980

Multiple protease resistance 3TC/AZT Multinucleoside resistance

0.49 ± 0.39 0.77 ± 0.40 0.28 ± 0.16 0.28 ± 0.18

0.59* 0.73* 0.55#

Results are the average of five experiments performed in duplicate ± S.E.M. 3TC, lamivudine; AZT, zidovudine. a BLM concentrations represent the IC values (i.e. concentrations of BLM necessary to achieve 50% p24 inhibition after 7 days of incubation with peripheral 50 blood mononuclear cells). b Statistical analysis using * paired Student’s t-test or # Student’s t-test (N = 3).

in the presence of BLM for 18 weeks. In the wild-type control strain (which was incubated in the absence of any drugs throughout the in vitro model), the IC50 of BLM, AZT, 3TC and RTV was found to be 0.64, 0.05, 0.068 and 0.029 ␮M, respectively. In the BLM-treated strain, the IC50 of BLM was 0.58 ␮M, which was not significantly different to that found in the control strain (P = 0.92). The IC50 of AZT in the AZT-treated strain increased from 0.05 ␮M in the control strain to 0.69 ␮M (P = 0.37). In contrast to both BLM and AZT IC50 values in their respective same-drug-treated strains, the IC50 of 3TC increased 85-fold from 0.068 ␮M in the control strain to 5.72 ␮M in the 3TC-treated strain (P = 0.02), suggesting the onset of drug resistance. Interestingly, the virus that had been treated with BLM was more sensitive to the antiviral effects of both AZT and 3TC, with the IC50 of these compounds falling significantly from 0.05 to 0.01 ␮M (P = 0.002) for AZT and from 0.068 to 0.017 ␮M (P = 0.036) for 3TC in the BLM-treated strain (Table 2). The IC50 of the protease inhibitor RTV in the wild-type strain and the BLM-treated strain was 0.029 and 0.012 ␮M, respectively (P = 0.287). There seemed to be increased sensitivity of both AZT-treated (P = 0.1) and 3TC-treated (P = 0.07) virus to BLM; however, this was not significant.

4. Discussion The involvement of cancer chemotherapeutic agents in viral load reduction in HIV-infected patients receiving both antiviral and anticancer agents has not been structurally

addressed in the clinic. It has previously been shown that BLM has antiviral properties and can synergise with known antiviral agents to inhibit HIV-1 replication [5,6]. In this study, resistant HIV-1 strains (including one containing the Q151M mutation) were equally sensitive to BLM in human PBMCs compared with the wild-type HXB2-D strain (Table 1). This suggests that in patients harbouring various mutations to known antiviral agents, BLM would still be effective. Based on the high turnover rate of the virus in HIV-positive individuals and the high mutation rate of the reverse transcriptase enzyme, variants resulting from single nucleotide substitutions can be produced on a daily basis [18]. Thus, drug-resistant variants may pre-exist in the virus population before the start of antiviral therapy. Indeed, mutations associated with reduced sensitivity to reverse transcriptase inhibitors and protease inhibitors have been detected in treatment-naive patients [19–21]. Concentrations of BLM higher than 1 ␮M were not used in these experiments. It is well known from oncology that BLM administration should not exceed a cumulative dose of 450 mg [22] owing to toxicity. A dose of 30 mg BLM per day for 4 days results in steady-state concentrations in plasma of between 0.1 and 0.3 ␮g/mL (0.07–0.21 ␮M) [23,24]. Effective antiviral concentrations of BLM achieving 50% reduction in viral p24 production (EC50 ) ranged from 0.49 to 0.56 ␮M in different strains. Concentrations resulting in 50% cellular viability (CC50 ) ranged from 3.08 to 7.49 ␮M and Selectivity Index (CC50 /EC50 ) ranged from 9 to 13 [6]. EC50 values of BLM in combination with either AZT, indinavir or RTV fall 16-, 4- and 4-fold, respectively, as BLM

Table 2 Drug susceptibility testing of HIV-1Ba-L virus treated for a period of 18 weeks with either BLM, AZT or 3TC Drug

HIV-1Ba-L long-term treatment with drugs Control

BLM-treated

AZT-treated

3TC-treated

BLM (␮M) AZT (␮M) 3TC (␮M) RTV (␮M)

0.64 ± 0.24 0.05 ± 0.007 0.068 ± 0.018 0.029 ± 0.014

0.58 ± 0.52 (P = 0.92) 0.01 ± 0.003 (P = 0.002) 0.017 ± 0.006 (P = 0.036) 0.012 ± 0.004 (P = 0.287)

0.154 ± 0.07 (P = 0.10) 0.69 ± 0.66 (P = 0.37) – –

0.112 ± 0.021 (P = 0.07) – 5.72 ± 1.84 (P = 0.02) –

BLM, bleomycin; AZT, zidovudine; 3TC, lamivudine; RTV, ritonavir; IC50 , drug concentration achieving a p24 inhibition of 50%. Concentrations represent IC50 values, and in parentheses are the P-values of each drug compared with its IC50 concentration in the control wild-type virus. Peripheral blood lymphocytes were infected with the corresponding viruses for 2 h at a multiplicity of infection (MOI) of 0.005 before excess virus was washed off and the various drug concentrations titrated. After 7 days of incubation, IC50 was calculated from p24 supernatant values. Results are the average of four different experiments and statistical analysis was performed by Student’s t-test.

N.A. Georgiou et al. / International Journal of Antimicrobial Agents 27 (2006) 63–68

acts synergistically with each of these compounds to lower HIV-1 replication [6]. Of interest is the increased sensitivity of the BLM-treated virus to both AZT and 3TC. We saw a significant four- to fivefold reduction in the effective dose of either 3TC or AZT in the BLM-treated strain (Table 2). BLM is known to bind to and damage DNA and RNA [25,26]. In addition to the ironand oxidative-dependent viral DNA damaging properties of BLM, the drug was found to decrease virus infectivity by 38% [27]. One can speculate that damage to the virus or any weak disruption of viral proteins could result in this hypersensitivity to AZT and 3TC. In a recent study, the efficacy and safety of the chemotherapeutic regimen BLM, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine and prednisone (BEACOPP) was assessed in HIV-infected patients with Hodgkin’s lymphoma [28] with and without additional antiretroviral therapy. In this small study population of 12 patients, the courses of CD4+ counts of patients with and without antiretroviral therapy were very similar, and in all patients with detectable RNA load not receiving concomitant antiviral therapy the RNA load was decreased 6 months after the last cycle of chemotherapy. The reduction of RNA load in the patients receiving only cancer chemotherapy is in agreement with our in vitro findings. Recently, the potential use of combining the anticancer cytotoxic drug 6-thioguanine with combination antiviral chemotherapy in eradicating HIV1 was reported [29]. Given the cumulative toxicity of BLM, we do not suggest incorporation of this drug as a cocktail in antiviral therapy. However, we believe that interaction between HIV and cancer chemotherapeutic agents and the contribution of the latter group in viral load reduction in patients receiving the combination of these two groups of drugs merits further investigation. The results described in this study predict the involvement of cancer chemotherapeutic agents such as BLM in viral load reduction of both wild-type and drug-resistant strains. Acknowledgments This work was supported by EC grant BMH4-CT97-2149, BIOMED 2, awarded to Prof. J.J.M. Marx and Dr B.S. van Asbeck. The authors wish to thank Albert van Wijk and Loek de Graaf for technical assistance. References [1] Ratner L, Lee J, Tang S, et al. Chemotherapy for human immunodeficiency virus-associated non-Hodgkin’s lymphoma in combination with highly active antiretroviral therapy. J Clin Oncol 2001;19:2171–8. [2] Sparano JA, Wiernik PH, Hu X, et al. Pilot trial of infusional cyclophosphamide, doxorubicin, and etoposide plus didanosine and filgrastim in patients with human immunodeficiency virus-associated non-Hodgkin’s lymphoma. J Clin Oncol 1996;14:3026–35.

67

[3] Levine AM, Li P, Cheung T, et al. Chemotherapy consisting of doxorubicin, bleomycin, vinblastine, and dacarbazine with granulocytecolony-stimulating factor in HIV-infected patients with newly diagnosed Hodgkin’s disease: a prospective, multi-institutional AIDS clinical trials group study (ACTG 149). J Acquir Immune Defic Syndr 2000;24:444–50. [4] Palmer S, Shafer RW, Merigan TC. Highly drug-resistant HIV-1 clinical isolates are cross-resistant to many antiretroviral compounds in current clinical development. AIDS 1999;13:661–7. [5] Georgiou NA, van der Bruggen T, Oudshoorn M, et al. Inhibition of human immunodeficiency virus type 1 replication in human mononuclear blood cells by the iron chelators deferoxamine, deferiprone, and bleomycin. J Infect Dis 2000;181:484–90. [6] Georgiou NA, van der Bruggen T, Jansen CA, et al. The chemotherapeutic agent bleomycin in a two-drug combination with zidovudine, ritonavir or indinavir synergistically inhibits HIV Type-1 replication in peripheral blood lymphocytes. Int J Antimicrob Agents 2001;18:513–8. [7] Coakley EP, Gillis JM, Hammer SM. Phenotypic and genotypic resistance patterns of HIV-1 isolates derived from individuals treated with didanosine and stavudine. AIDS 2000;14:F9–15. [8] Iversen AK, Shafer RW, Wehrly K, et al. Multidrug-resistant human immunodeficiency virus type 1 strains resulting from combination antiretroviral therapy. J Virol 1996;70:1086–90. [9] Nijhuis M, Schuurman R, de Jong D, et al. Increased fitness of drug resistant HIV-1 protease as a result of acquisition of compensatory mutations during suboptimal therapy. AIDS 1999;13:2349– 59. [10] Maschera B, Furfine E, Blair ED. Analysis of resistance to human immunodeficiency virus type 1 protease inhibitors by using matched bacterial expression and proviral infection vectors. J Virol 1995;69:5431–6. [11] Moore JP, McKeating JA, Weiss RA, Sattentau QJ. Dissociation of gp120 from HIV-1 virions induced by soluble CD4. Science 1990;250:1139–42. [12] Japour AJ, Mayers DL, Johnson VA, et al. Standardized peripheral blood mononuclear cell culture assay for determination of drug susceptibilities of clinical human immunodeficiency virus type 1 isolates. The RV-43 Study Group, the AIDS Clinical Trials Group Virology Committee Resistance Working Group. Antimicrob Agents Chemother 1993;37:1095–101. [13] Reed LJ, Muench H. A simple method of estimating fifty percent endpoints. Am J Hyg 1938;27:493–7. [14] Chou TC, Hayball MP. CalcuSyn: Windows software for dose effect analysis. Cambridge, UK: BIOSOFT; 1996. [15] Chou TC, Talalay P. Quantitative analysis of dose–effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984;22:27–55. [16] Keulen W, Back NK, van Wijk A, Boucher CA, Berkhout B. Initial appearance of the 184Ile variant in lamivudine-treated patients is caused by the mutational bias of human immunodeficiency virus type 1 reverse transcriptase. J Virol 1997;71:3346– 50. [17] Pauwels R, Balzarini J, Baba M, et al. Rapid and automated tetrazolium-based colorimetric assay for the detection of anti-HIV compounds. J Virol Methods 1988;20:309–21. [18] Coffin JM. HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 1995;267:483–9. [19] de Jong MD, Schuurman R, Lange JM, Boucher CA. Replication of a pre-existing resistant HIV-1 subpopulation in vivo after introduction of a strong selective drug pressure. Antivir Ther 1996;1:33– 41. [20] Lech WJ, Wang G, Yang YL, et al. In vivo sequence diversity of the protease of human immunodeficiency virus type 1: presence of protease inhibitor-resistant variants in untreated subjects. J Virol 1996;70:2038–43.

68

N.A. Georgiou et al. / International Journal of Antimicrobial Agents 27 (2006) 63–68

[21] Najera I, Holguin A, Quinones-Mateu ME, et al. Pol gene quasispecies of human immunodeficiency virus: mutations associated with drug resistance in virus from patients undergoing no drug therapy. J Virol 1995;69:23–31. [22] Blum RH, Carter SK, Agre K. A clinical review of bleomycin—a new antineoplastic agent. Cancer 1973;31:903–14. [23] Alberts DS, Chen HS, Liu R, et al. Bleomycin pharmacokinetics in man. I. Intravenous administration. Cancer Chemother Pharmacol 1978;1:177–81. [24] Alberts DS, Chen HS, Mayersohn M, Perrier D, Moon TE, Gross JF. Bleomycin pharmacokinetics in man. II. Intracavitary administration. Cancer Chemother Pharmacol 1979;2:127–32. [25] Carter BJ, de Vroom E, Long EC, van der Marel GA, van Boom JH, Hecht SM. Site-specific cleavage of RNA by Fe(II)·bleomycin. Proc Natl Acad Sci USA 1990;87:9373–7.

[26] Holmes CE, Carter BJ, Hecht SM. Characterization of iron(II)·bleomycin-mediated RNA strand scission. Biochemistry 1993;32:4293–307. [27] Georgiou NA, van der BT, Oudshoorn M, et al. Mechanism of inhibition of the human immunodeficiency virus type 1 by the oxygen radical generating agent bleomycin. Antiviral Res 2004;63:97– 106. [28] Hartmann P, Rehwald U, Salzberger B, et al. BEACOPP therapeutic regimen for patients with Hodgkin’s disease and HIV infection. Ann Oncol 2003;14:1562–9. [29] Yang QE, Li KG, Mikovits JA. Eradication of human immunodeficiency virus type 1-infected cells by a combination of antimetabolic cytotoxic chemotherapy and antiviral chemotherapy in vitro: a pilot study. J Infect Dis 2002;186:706–9.