Differences in the intracellular accumulation of HIV protease inhibitors in vitro and the effect of active transport.

OBJECTIVES: To investigate the intracellular accumulation of HIV protease inhibitors (PI) and to assess the effect of active transport on this accumulation. METHODS: CEM cells were incubated with a PI for 18 h and the intracellular concentration determined using cell number and radioactivity. The effect of active transport was investigated using cells expressing P-glycoprotein (CEM(VBL)) and cells expressing multidrug resistance-associated protein 1 (MRP1; CEM(E1000)). Incubations were also carried out at 4 degrees C and in the presence of 2-deoxyglucose plus rotenone to examine the effect of inhibiting active transport. RESULTS: Nelfinavir (NFV) accumulated to the greatest extent (> 80-fold) followed by saquinavir (SQV; approximately 30-fold), ritonavir (RTV; 3-7-fold) and finally indinavir (IDV; extracellular equivalent to intracellular). In CEM(VBL) cells there was a significant reduction in the intracellular accumulation of NFV, SQV and RTV and in CEM(E1000) cells there was reduced accumulation of SQV and RTV. Inhibition of active transport processes caused a reduction in SQV and RTV accumulation but had no effect on IDV accumulation in all cell types. NFV accumulation was increased in CEM(VBL) cells as a result of inhibition of active transport. CONCLUSIONS: Marked differences can be detected in the intracellular accumulation of HIV PI drugs in vitro. Both P-glycoprotein and MRP1 may play a role in limiting the intracellular concentration of the PI and active influx mechanisms may contribute to drug accumulation.

The effect of zidovudine dose on the formation of intracellular phosphorylated metabolites.

OBJECTIVES: Zidovudine (ZDV) requires intracellular phosphorylation to ZDV triphosphate (ZDV-TP) prior to the inhibition of HIV replication. The effect of ZDV dose on the formation of intracellular phosphorylated metabolites may help define the optimum daily dose of ZDV, which is still unknown. DESIGN AND METHODS: The plasma and intracellular phosphorylated metabolite concentrations of ZDV were determined over a 12 h period following oral administration of 100 and 300 mg ZDV to 10 HIV-seropositive patients at steady state during two dosing regimens (i.e., 100 mg three times daily and 300 mg twice daily). The intracellular ZDV phosphates, ZDV monophosphate (ZDV-MP), ZDV diphosphate (ZDV-DP) and ZDV-TP were measured in peripheral blood mononuclear cells using a combination of high-performance liquid chromatography and radioimmunoassay. RESULTS: There was a greater than threefold increase in maximum plasma concentration (Cmax) following 300 mg ZDV when compared with 100 mg ZDV (mean +/- SD, 2.59 +/- 0.52 versus 0.70 +/- 0.14 mumol/l). The area under the concentration time curve (AUC0-12 h) was also significantly increased (4.59 +/- 0.79 versus 1.42 +/- 0.51 mumol/l x h) following 300 mg ZDV dose. For total intracellular ZDV phosphate metabolites the AUC0-12 h was doubled (7.64 +/- 3.67 versus 3.71 +/- 1.83 pmol/10(6) cells x h) in patients taking 300 mg ZDV compared with 100 mg. The AUC0-12 h for ZDV-MP was significantly increased at the higher dose (6.47 +/- 3.14 versus 2.77 +/- 1.70 pmol/10(6) cells x h), whereas the active moiety ZDV-TP was variable and not significantly different (0.42 +/- 0.42 versus 0.61 +/- 0.81 pmol/10(6) cells x h) following 100 and 300 mg ZDV. CONCLUSIONS: Administration of 100 mg ZDV orally produces significantly less of the potentially toxic metabolite, ZDV-MP, and comparative, although variable, concentrations of the active metabolite ZDV-TP when compared with 300 mg ZDV orally. This finding supports clinical data indicating the efficacy of low-dose (300 mg daily) ZDV. The measurement of intracellular phosphorylated metabolites advances our understanding of the clinical pharmacology of ZDV.

Determination of P-gp and MRP1 expression and function in peripheral blood mononuclear cells in vivo.

P-glycoprotein (P-gp) and multidrug resistance-associated protein (MRP) mediate the efflux of many therapeutic agents and have been implicated in the treatment failure of many infectious diseases and cancers. The ability to characterise the expression and function of these transporters in vivo is important when assessing the pharmacological activity of drugs. We investigated some of the problems involved in screening the multidrug resistance status of individuals using flow cytometry. Expression of P-gp and MRP1 on the surface of lymphocytes isolated from blood samples (30 ml) was determined by indirect immunofluorescence. Functional ability was assessed by measuring the efflux of specific fluorescent dyes. Results were expressed as a mean fold increase in fluorescence from the isotype control (expression) and a change in fluorescence compared to the load (function). Using these assays, we determined the expression of P-gp to be 2.01+/-0.40, n=30 and MRP1 to be 1.46+/-0.23, n=25. Functional ability was 6.98+/-4.97, n=25 for P-gp and 1.55+/-0.25, n=25 for MRP1. The dye efflux studies were associated with a lack of specificity and a number of methodological difficulties. There was no correlation between the expression and function of P-gp (r=0.338; p=0.10) or MRP1 (r=0.283; p=0.17). Therefore, we considered determination of P-gp and MRP1 expression to be a more reproducible and accurate approach to clinical investigation into the role of multidrug resistance.

A simplified approach to determining P-glycoprotein expression in peripheral blood mononuclear cell subsets.

P-glycoprotein (P-gp), encoded by the MDR-1 (multidrug resistance) gene mediates the cellular efflux of several therapeutic agents with the potential of treatment failure. The differential expression of P-gp in many localised tissues and cells of the hematopoietic system implies diverse physiological and pharmacological roles. The exact function of P-gp involved in multidrug resistance remains unclear owing to the numerous discrepancies between different laboratories. The ability to characterise accurately P-gp expression has important clinical implications. However, a complete consensus recommendation regarding methods of P-gp detection has been difficult to reach. With the advancement in immune technology and new commercially available antibodies, we describe a simplified direct immunofluorescent assay capable of detecting surface P-gp expression in peripheral blood mononuclear cells (PBMCs) and subpopulations of lymphocytes in vivo by dual colour flow cytometry. Results were expressed as mean increase in fluorescence (MI) compared to isotypically matched controls. Using this assay, differential basal P-gp expression was found to exist in the following significant hierarchy CD56+ (MI=0.684+/-0.273; n=15)>CD8+ (MI=0.312+/-0.117; n=15)>CD4+ (MI=0.194+/-0.086; n=15). This method is rapid and reproducible and has potential use for in vitro and in vivo application.

Expression of P-glycoprotein and multidrug resistance-associated protein in healthy volunteers and HIV-infected patients.

Increased expression of the multidrug efflux transporters P-glycoprotein (P-gp) and multidrug resistance-associated protein (MRP) has been suggested as a potential mechanism for decreased drug availability at certain intracellular sites that provide sanctuary for HIV. Here we investigate the expression of these transporters in peripheral blood mononuclear cells (PBMCs) of HIV-infected patients and healthy volunteers. Venous blood (30 ml) was taken from healthy volunteers (n = 21) and HIV-infected patients (n = 21; 4 antiretroviral drug naive, 17 antiretroviral drug experienced). PBMCs were isolated and fixed. To assess P-gp expression, PBMCs were incubated with an isotype control antibody or an antibody directed to an external epitope of P-gp (UIC2). To assess MRP expression, cells were permeabilized before incubation with either a control antibody or an antibody directed to an internal epitope of MRP (MRPm5). After washing, a secondary phycoerythrin-bound antibody was incubated. After additional wash steps, samples were fixed and analyzed by flow cytometry. The median fluorescence intensity of 5000 events was recorded. Results are expressed as fold increase between isotype control and UIC2/MRPm5 samples. Expression of P-gp in HIV-infected patients (1.42 +/- 0.36) was significantly lower (p = 0.0021; 95% CI, -0.633 to -0.164) than in healthy volunteers (1.82 +/- 0.55). However, MRP expression was similar in HIV-infected patients (1.37 +/- 0.34) and healthy volunteers (1.37 +/- 0.21; p = 0.91; 95% CI, -0.148941 to 0.165191). We conclude that in HIV infection, P-gp expression in total PBMCs is reduced whereas MRP expression appears to be unaltered.

Effect of ribavirin on zidovudine efficacy and toxicity in vitro: a concentration-dependent interaction.

Zidovudine (ZDV) is converted to its active triphosphate (ZDVTP) by intracellular kinases. The intermediate ZDV monophosphate (ZDVMP) is believed to play a major role in ZDV toxicity. Manipulation of ZDV phosphorylation is a possible therapeutic strategy for altering the risk-benefit ratio. Here we investigate whether combining RBV with ZDV is able to modulate efficacy and toxicity of ZDV. We have measured the intracellular activation of ZDV (0.3 microM) in the absence and presence of ribavirin (RBV; 2 and 20 microM) in Molt 4 and U937 cells. MTT cytotoxicity of ZDV (10-1000 microM) was also measured with and without RBV (2 microM) in Molt 4 and U937 cells. Measurement of endogenous deoxythymidine triphosphate (dTTP) allowed investigation of the dTTP/ZDVTP ratio. The antiviral efficacy of ZDV in combination with RBV (2 microM) was assessed by HIV p24 antigen measurements. In the presence of RBV (2 and 20 microM) a decrease in total ZDV phosphates was observed, owing mainly to an effect primarily on ZDVMP rather than the active ZDVTP. RBV also increased endogenous dTTP pools in both cell types, resulting in an increase in the dTTP/ZDVTP ratio. ZDV alone significantly reduced p24 antigen production, with an IC50 of 0.34 microM. Addition of RBV increased the IC50 approximately fivefold (1.52 microM). However, at higher concentrations of ZDV (10 and 100 microM) the antagonistic effect of RBV (2 microM) on ZDV was lost. The RBV-mediated decrease in ZDVMP may explain the reduction in ZDV toxicity when combined with RBV (2 microM). Cytotoxicity of ZDV was reduced in the presence of RBV (2 microM) at all concentrations in both cell lines, probably owing to saturation of ZDVTP formation. The interaction of ZDV and RBV is concentration dependent.

Intracellular activation of 2',3'-dideoxyinosine and drug interactions in vitro.

Didanosine (2',3'-dideoxyinosine; ddI) requires intracellular metabolism to its active triphosphate, 2',3'-dideoxyadenosine 5'-triphosphate (ddATP), to inhibit the replication of human immunodeficiency virus (HIV). We have investigated the metabolism of ddI to ddATP in the presence and absence of a range of compounds. In addition, we determined the levels of the endogenous competitor of ddATP, 2'-deoxyadenosine 5'-triphosphate (dATP), and calculated ddATP/dATP ratios. None of the nucleoside analogs studied had any effect on ddI phosphorylation at 1 and 10 microM concentrations. At 100 microM concentrations, ddC reduced total ddA phosphates (82% of control total ddA phosphates; p < 0.001). ZDV significantly decreased the levels of dATP, whereas ddC significantly increased dATP pools (e.g., at 100 microM ZDV, 82% of control dATP levels; p < 0.001). Hence, the ddATP/dATP ratio was increased in the presence of ZDV, but was decreased in the presence of ddC. Neither d4T nor 3TC affected the ddATP/dATP ratio. Deoxyinosine (dI) significantly reduced ddA phosphate production at 100 microM concentrations, with ddATP reduced to undetectable levels (p < 0.001). Hydroxyurea (HU) did not affect the activation of ddI, but significantly reduced dATP pools at 100 microM concentrations (67% of control dATP levels; p < 0.001), enhancing the ddATP/dATP ratio. ddA phosphate production was significantly reduced by pentoxyfylline (PXF) at 10 and 100 microM concentrations. dATP levels were unaffected, but the ddATP/dATP ratio was reduced. Finally, 8-aminoguanosine (8-AMG) had no effect on either ddI activation or dATP pools. These studies demonstrate the importance of determining both the active TP and the competing endogenous TP, as changes to the resulting ratio could alter the efficacy of the nucleoside analog in question.

Lamivudine (3TC) phosphorylation and drug interactions in vitro.

Lamivudine (2'-deoxy-3'-thiacytidine; 3TC) is a dideoxynucleoside analogue that inhibits the replication of human immunodeficiency virus (HIV). We are currently investigating the intracellular metabolism of 3TC to its active triphosphate (3TCTP) in peripheral blood mononuclear cells (PBMC) and a monocytic cell line (U937). Optimal phosphorylation of 3TC was achieved after incubation for 24 hr, with 3TC diphosphate (3TCDP) the predominant metabolite formed, in both cell types investigated. Further studies in PBMCs followed preincubation with the mitogen phytohaemagglutinin (PHA) for 72 hr. This enabled greater detection of phosphates, compared to resting cells. A 3TC concentration of 1 microM was chosen for future interaction studies, allowing good detection of 3TC and phosphates on radiochromatograms whilst being similar to the plasma level found in clinical studies (i.e. 3 microM). With a shift in treatment to combination therapy, it is essential that potential interactions between nucleoside analogues are investigated at the phosphorylation level, as this could affect antiviral activity. Both deoxycytidine (dC) and 2',3'-dideoxycytidine (ddC) significantly inhibited 3TC phosphorylation (e.g. at dC 100 microM, no 3TCTP was detected in PBMCs; P < 0.001, whereas 66% of control 3TCTP production was observed in U937 cells; P < 0.01). Zidovudine (ZDV) caused a small but significant reduction of 3TC phosphate production in both PBMCs and U937 cells. However, this may be due to toxicity or an effect on endogenous dCTP pools. Neither 2',3'-dideoxyinosine (ddI) or 2',3'-didehydro-2',3'-dideoxythymidine (d4T) significantly inhibited 3TC phosphorylation. These results suggest it would be better to coadminister two nucleoside analogues with different activation pathways.

Correlation between intracellular pharmacological activation of nucleoside analogues and HIV suppression in vitro.

Following intracellular activation of HIV nucleoside analogue reverse transcriptase inhibitors, their triphosphates (ddNTPs) compete with endogenous nucleoside triphosphates (dNTPs) for incorporation into proviral DNA. In this study we have examined the effect of combinations of two thymidine analogues, stavudine (d4T) and zidovudine (ZDV), and two cytidine analogues, lamivudine (3TC) and zalcitabine (ddC) on intracellular drug activation and on the relevant competing dNTP in uninfected and persistently HIV-infected cells. Endogenous triphosphates of deoxycytidine (dCTP) and deoxythymidine (dTTP) were measured using a template primer assay and the ratio of ddNTP:dNTP was calculated. Antiviral activity of two-drug combinations was also assayed by p24 ELISA. A significant reduction in d4T triphosphate (d4TTP) [0.11+/-0.09 pmol/10(6) cells to undetectable (<0.01); P=0.039] in the presence of equimolar concentrations of ZDV and d4T, resulted in a decrease in the d4TTP/dTTP ratio of 90%. ZDVTP/dTTP was not significantly altered in the presence of d4T. 3TC (10 microM) reduced total ddC phosphates by 57% and ddCTP/dCTP by 27%. 3TC phosphorylation was comparatively unaffected by ddC, up to a concentration of 10 microM ddC (>100 times the plasma concentration achieved following standard dosing). 3TC plus ddC resulted in greater p24 inhibition than 3TC or ddC alone (P<0.001). Combining one thymidine analogue (ZDV or d4T) with one cytidine analogue (3TC or ddC) resulted in greater inhibition of p24 inhibition than with any single agent. From a pharmacological viewpoint, the combination of ZDV plus d4T should be avoided, but in vitro the combination of 3TC plus ddC confers modest benefit over either drug alone. This in vitro study illustrates that decreases in ddNTP/dNTP are consistent with a reduction in antiviral effect.

Effect of ribavirin on intracellular and plasma pharmacokinetics of nucleoside reverse transcriptase inhibitors in patients with human immunodeficiency virus-hepatitis C virus coinfection: results of a randomized clinical study.

The intracellular triphosphorylation and plasma pharmacokinetics of lamivudine (3TC), stavudine (d4T), and zidovudine (ZDV) were assessed in a pharmacokinetic substudy, in 56 human immunodeficiency virus-hepatitis C virus (HIV-HCV) coinfected patients receiving peginterferon alfa-2a (40KD) 180 microg/week plus either placebo or ribavirin (RBV) 800 mg/day in the AIDS PEGASYS Ribavirin International Coinfection Trial. There were no significant differences between patients treated with RBV and placebo in plasma pharmacokinetics parameters for the nucleoside reverse transcriptase inhibitors (NRTIs) at steady state (weeks 8 to 12): ratios of least squares mean of area under the plasma concentration-time curve (AUC(0-12 h)) were 1.17 (95% confidence interval, 0.91 to 1.51) for 3TC, 1.44 (95% confidence interval, 0.58 to 3.60) for d4T and 0.85 (95% confidence interval, 0.50 to 1.45) for ZDV, and ratios of least squares mean plasma C(max) were 1.33 (95% confidence interval, 0.99 to 1.78), 1.06 (95% confidence interval, 0.68 to 1.65), and 0.84 (95% confidence interval, 0.46 to 1.53), respectively. Concentrations of NRTI triphosphate (TP) metabolites in relation to those of the triphosphates of endogenous deoxythymidine-triphosphate (dTTP) and deoxcytidine-triphosphate (dCTP) were similar in the RBV and placebo groups. Differences (RBV to placebo) in least squares mean ratios of AUC(0-12 h) at steady state were 0.274 (95% confidence interval, -0.37 to 0.91) for 3TC-TP:dCTP, 0.009 (95% confidence interval, -0.06 to 0.08) for d4T-TP:dTTP, and -0.081 (95% confidence interval, -0.40 to 0.24) for ZDV-TP:dTTP. RBV did not adversely affect HIV-1 replication. In summary, RBV 800 mg/day administered in combination with peginterferon alfa-2a (40KD) does not significantly affect the intracellular phosphorylation or plasma pharmacokinetics of 3TC, d4T, and ZDV in HIV-HCV-coinfected patients.

Effects of drugs on 2',3'-dideoxy-2',3'-didehydrothymidine phosphorylation in vitro.

Drugs commonly administered to patients infected with the human immunodeficiency virus (HIV) have been studied for their propensity to alter the intracellular phosphorylation of the anti-HIV nucleoside analog stavudine (2',3'-dideoxy-2',3'-didehydrothymidine; d4T) in peripheral blood mononuclear cells (PBMCs) and U937 cells in vitro. PBMCs isolated from the blood of healthy volunteers were stimulated by the mitogen phytohemagglutinin (10 microg/ml) for 72 h. Stimulated PBMCs (3 x 10(6) cells/plate) were then incubated with [3H]d4T (0.65 microCi; 3 microM) and either acyclovir, dapsone, ddC, ddI, fluconazole, foscarnet, ganciclovir, itraconazole, lobucavir, ranitidine, ribavirin, rifampin, sorivudine, sulfamethoxazole, trimethoprim, lamivudine (3TC), zidovudine, or thymidine (30 and 300 microM) for 24 h. Doxorubicin and drugs showing some evidence of inhibition were also studied at 0.3 and 3 microM. Cells were extracted overnight with 60% methanol prior to analysis by radiometric high-performance liquid chromatography. Additional data for nine of the drugs were obtained by incubation with [3H]d4T in U937 cells for 24 h. The effect of d4T (0.2 to 20 microM) on zidovudine (0.65 microCi; 0.018 microCi) phosphorylation was also studied. Zidovudine significantly reduced d4T total phosphates in PBMCs and U937 cells (in PBMCs to 33% [P < 0.001] and 17% [P < 0.001] of that in control cells at 3 and 30 microM, respectively). A small reduction in zidovudine phosphorylation was seen with d4T but only at d4T:zidovudine ratios of 100 and 1,000. Of the other compounds screened, only thymidine, ribavirin, and doxorubicin produced inhibition of d4T phosphorylation in both PBMCs and U937 cells. However, doxorubicin was cytotoxic at 3 microM. The decrease in d4T phosphorylation in the presence of ribavirin is consistent with previous findings with zidovudine. Although ddC significantly inhibited the phosphorylation of d4T in PBMCs, this was not seen in U937 cells, and it is probable that the findings in PBMCs are related to mitochondrial toxicity [based on 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide cytotoxicity assay]. The only drugs screened which may interfere with d4T phosphorylation at clinically relevant concentrations were zidovudine, ribavirin, and doxorubicin.

The intracellular activation of lamivudine (3TC) and determination of 2'-deoxycytidine-5'-triphosphate (dCTP) pools in the presence and absence of various drugs in HepG2 cells.

AIMS: Lamivudine (3TC, 2'-deoxy-3'-thiacytidine) requires intracellular metabolism to its active 5'-triphosphate, 3TC-5'-triphosphate (3TCTP), to inhibit the replication of hepatitis B virus (HBV). We have investigated the activation of 3TC, in the presence and absence of a range of compounds, in HepG2 cells. The intracellular levels of the endogenous competitor of 3TCTP, 2'-deoxycytidine-5'-triphosphate (dCTP), were also determined and 3TCTP/dCTP ratios calculated. METHODS: The effects of a number of compounds on 3TC (3H; 1 microM) phosphorylation were investigated by radiometric h.p.l.c. dCTP levels were determined using a template primer extension assay. 3TCTP/dCTP ratios were calculated from these results. RESULTS: The phosphorylation of 3TC was significantly increased in the presence of either hydroxyurea (HU), methotrexate (MTX), or fludarabine (FLU). For example, at 100 microM HU, control 3TCTP levels were increased to 361% of control, whereas at 100 microM FLU, control 3TCTP levels were increased to 155%. dCTP pools were significantly reduced in the presence of HU and FLU, at 100 microM concentrations only. However, for all the above three compounds investigated, the ratio of 3TCTP/dCTP was favourably enhanced (e.g. at 1 microM MTX, 255% of control). Neither ganciclovir (GCV), lobucavir (LCV), penciclovir (PCV), adefovir dipivoxil (ADV), nor foscarnet (FOS) had any significant effects on 3TC phosphorylation or dCTP pools. CONCLUSIONS: These results suggest that the activity of 3TC may be potentiated when combined with one of the modulators studied. The lack of an interaction between 3TC and the other anti-HBV agents is reassuring. These in vitro studies can be used as an initial screen to examine potential interactions at the phosphorylation level.

Effect of alpha1-acid glycoprotein on the intracellular accumulation of the HIV protease inhibitors saquinavir, ritonavir and indinavir in vitro.

AIMS: Since alpha1-acid glycoprotein (AGP) levels may be raised during HIV infection, we have examined in vitro the effect of increasing the concentration of AGP on the intracellular accumulation of the HIV protease inhibitors saquinavir (SQV), ritonavir (RTV) and indinavir (IDV). METHODS: U937 cells (5 x 10(6) cells in 5 ml RPMI growth medium) were incubated at 37 degrees C for 18 h with [14C]-SQV (0.1 microCi), [3H]-RTV and [3H]-IDV (0.135 microCi) to a final concentration of 1 microM in the presence of 0, 0.5 and 2.0 mg x ml(-1) AGP. Following extraction in 60% methanol the intracellular drug concentration was determined by liquid scintillation counting. RESULTS: Binding to AGP (2.0 mg x ml(-1)) reduced the mean intracellular concentration of SQV from 31.5 microM to 7.4 microM (P < 0.0001; 95% CI 19.4-28.8). RTV concentration was also reduced (8.8 microM to 1.6 microM; P < 0.0001; 95% CI 5.4-9.0) as was the concentration of IDV (3.0 microM to 1.5 microM; P < 0.0001; 95% CI 1.1-1.9). CONCLUSIONS: Reduced intracellular protease inhibitor concentrations in the presence of increasing concentrations of AGP will certainly impact on the antiviral activity in vitro. However, since protease inhibitors are high clearance drugs, free drug concentration will likely remain unaffected in the presence of elevated AGP during chronic oral dosing although there will be an increase in total plasma drug concentration.

Effect of protease inhibitors on nucleoside analogue phosphorylation in vitro.

AIMS: Combination antiretroviral therapy for human immunodeficiency virus (HIV) infection now involves both nucleoside analogues and protease inhibitors. Since intracellular phosphorylation is essential for the activity of all the nucleoside analogues this study was designed to investigate interactions with protease inhibitors at the intracellular level which may alter antiviral efficacy. METHODS: PHA-stimulated PBMCs (3 x 10[6] cell/plate) and U937 cells (4 x 10[6] cells/plate) were incubated with either radiolabelled zidovudine (ZDV), stavudine (d4T), zalcitabine (ddC), lamivudine (3TC) or didanosine (ddI) in the presence and absence of the protease inhibitors, indinavir, ritonavir, and saquinavir (0.1-10 microM) for 24 h. Cells were extracted overnight prior to analysis by radiometric h.p.l.c. Intracellular phosphates were standardised to pmol per million cells. RESULTS: None of the three protease inhibitors tested had any significant effect on the intracellular phosphorylation of the five nucleoside analogues. It is particularly important to focus on the active triphosphate anabolites and data for control vs ritonavir (10 microM) incubations in U937 cells were as follows: ZDVTP, 0.19 +/- 0.02 vs 0.21 +/- 0.2 pmol/10(6) cells (mean +/- s.d.; n = 5); d4TTP, 0.30 +/- 0.13 vs 0.27 +/- 0.26; 3TCTP, 0.32 +/- 0.12 vs 0.26 +/- 0.19; ddCTP, 0.07 +/- 0.04 vs 0.06 +/- 0.02, ddATP, 0.014 +/- 0.003 vs 0.018 +/- 0.006 pmol/10(6) cells. CONCLUSIONS: The protease inhibitors, indinavir, ritonavir and saquinavir have no effect on the enzymes responsible for phosphorylation. Combining protease inhibitors and nucleoside analogues should not lead to any intracellular interactions in vivo.

Drug interactions with zidovudine phosphorylation in vitro.

We have investigated the effect of a range of drugs (some commonly coadministered with zidovudine [ZDV] to human immunodeficiency virus-positive patients) on intracellular phosphorylation of ZDV by stimulated peripheral blood mononuclear cells, Molt 4 cells, and U937 cells in vitro. Of the drugs tested (azoles, antiviral agents, antibiotics, and anticancer agents), only doxorubicin and ribavirin caused inhibition of anabolite formation as measured by high-performance liquid chromatography. This in vitro approach may provide important leads to potential interactions at the phosphorylation level in patients with human immunodeficiency virus disease. It is reassuring that so many commonly administered drugs do not alter ZDV phosphorylation.

Zidovudine phosphorylation and mitochondrial toxicity in vitro.

Zidovudine (ZDV) is a thymidine analogue activated to its triphosphate (ZDVTP) by the host's intracellular enzymes. The initial phosphorylation step is conversion to ZDV monophosphate (ZDVMP). The poor affinity of ZDVMP for thymidylate kinase results in intracellular accumulation of ZDVMP. Clinical use of ZDV is associated with cytotoxicity, thought to be mediated through mitochondrial damage. It has been suggested that ZDV cytotoxicity correlates with intracellular ZDVMP. Here we have further studied the role of ZDVMP in cytotoxicity and some of the mechanisms involved. Intracellular metabolism of ZDV in five lymphocyte/monocyte cell lines, U937, BSM, MOLT 4, JJAHN, and RAJI (4 x 10(6) cells), was investigated following 24 h incubation with [(3)H]ZDV (1.2 microCi; 0.1 microM) and cytotoxicity was determined by the MTT assay. Cytotoxicity was closely related to intracellular concentrations of the major metabolite (ZDVMP) but not with the active metabolite ZDVTP. ZDVMP was the only metabolite detected following incubation of viable mitochondria isolated from U937 cells with ZDV (1.2 microCi; 0.1 microM; 24 h) with mitochondrial levels of 0.27 +/- 0.11 pmol/microg protein (mean +/- SD; n = 3). No MTT toxicity was seen in isolated mitochondria. Following phytohemagglutinin (PHA) stimulation of peripheral blood mononuclear cells there was an increase in ZDV cytotoxicity compared to unstimulated cells. The results suggest that the mitochondrial isozyme of thymidine kinase (TK2) plays only a minor part in ZDVMP formation. Following PHA stimulation, activation of the cytosolic thymidine kinase isozyme (TK1) is associated with increased toxicity of ZDV. We conclude that ZDVMP responsible for mitochondrial toxicity is formed in the cytosol.

Pharmacokinetics of zidovudine phosphorylation in human immunodeficiency virus-positive thai patients and healthy volunteers.

We assessed the pharmacokinetics of zidovudine (ZDV) in plasma and intracellular ZDV phosphate anabolites in peripheral blood mononuclear cells in Thai human immunodeficiency virus (HIV) type 1-infected patients and healthy volunteers. The plasma ZDV area under the concentration-time curve from 0 to 6 h (AUC(0-6)) was similar in patients and healthy volunteers (32.77 and 22.77 micromol/liter. h, respectively; confidence interval, -3.37 to 19. 92). Although the concentration of ZDV triphosphate (ZDVTP) was similar in the two groups, the ZDV monophosphate (ZDVMP) AUC(0-6) was significantly greater in HIV patients (1.12 pmol/10(6) cells) than in healthy volunteers (0.15 pmol/10(6) cells). In agreement with previously published data obtained with Caucasians, the significant difference in intracellular phosphorylation in Thai volunteers and HIV patients is primarily due to ZDVMP. Comparing the data from this study with the data obtained with Caucasians suggests no marked ethnic differences in ZDV phosphorylation.

Effect of protease inhibitor-containing regimens on lymphocyte multidrug resistance transporter expression.

BACKGROUND: Increased expression of multidrug resistance transporters, such as P-glycoprotein (P-gp), has been suggested as a potential mechanism for decreased protease inhibitor (PI) availability at certain intracellular sites and tissue compartments. OBJECTIVES: To investigate the effect of PIs on the surface lymphocyte expression of P-gp in vitro and in vivo. PATIENTS AND METHODS: Peripheral blood mononuclear cells (PBMCs) were isolated from healthy subjects (n = 15) and incubated (72 h) with 10 microM of each PI studied (saquinavir, ritonavir, nelfinavir, indinavir, amprenavir and lopinavir), or dimethyl sulphoxide (DMSO) control. PBMCs were also isolated from HIV-infected subjects (n = 50; viral load <50 copies/mL) on a PI- or a non-PI-containing combination antiretroviral regimen. P-gp expression was analysed by flow cytometry. RESULTS: No differences in surface P-gp expression on lymphocytes, CD4+ or CD8+ lymphocyte subsets were observed following incubation with 10 microM saquinavir, ritonavir, indinavir, amprenavir or lopinavir in vitro. Nelfinavir, however, increased P-gp expression. In vivo, no difference in P-gp expression on total lymphocytes was observed between patients receiving a PI-containing regimen [saquinavir n = 9, ritonavir n = 6, nelfinavir n = 7, indinavir n = 7 and lopinavir/ritonavir n = 13, and two nucleoside reverse transcriptase inhibitors (NRTIs)] and patients receiving a control regimen of three NRTIs alone (n = 8). CONCLUSION: This study suggests that, of the PIs, only nelfinavir increases P-gp expression in vitro, and in vivo the PI class of antiretrovirals do not increase P-gp expression on lymphocytes. It is clear that factors other than PI induction are important in the inter-individual variability in the lymphocyte expression of P-gp.

P-glycoprotein and MRP1 expression and reduced ritonavir and saquinavir accumulation in HIV-infected individuals.

OBJECTIVES: Efflux transporters may play a role in lowering intracellular drug concentrations. As the HIV protease inhibitors are substrates for the efflux transporters P-glycoprotein and MRP, we wished to investigate whether differences in expression of these transporters on human lymphocytes correlated with intracellular concentrations of ritonavir and saquinavir. PATIENTS AND METHODS: Drug efflux transporter expression (P-glycoprotein and MRP1) on peripheral blood mononuclear cells isolated from HIV-positive patients was investigated using flow cytometry. In addition, plasma and intracellular ritonavir and saquinavir concentrations were measured by HPLC/mass spectrometry. The ratio of intracellular:plasma drug concentration was used to quantify intracellular drug accumulation. RESULTS: Patients with lower MRP1 expression (