Share this article on:

Concentrations of zidovudine- and lamivudine-triphosphate according to cell type in HIV-seronegative adults

Anderson, Peter L; Zheng, Jia-Hua; King, Tracy; Bushman, Lane R; Predhomme, Julie; Meditz, Amie; Gerber, John; Fletcher, Courtney V

doi: 10.1097/QAD.0b013e3282741feb
Basic Science

Introduction: Concentrations of zidovudine (ZDV)- and lamivudine (3TC)-triphosphates (TP) have been quantified in unfractionated peripheral blood mononuclear cells (PBMC) from HIV+ patients. The objective of this study was to determine whether concentrations of ZDV-TP and 3TC-TP in PBMC reflect the concentrations within CD4 T cells in HIV-seronegative adults.

Methods: Volunteers had taken 300 mg of ZDV plus 150 mg of 3TC twice daily for ≥ 7 days. Blood (60 mL) was collected 2 or 5 h post observed dose. PBMC were processed into three cell fractions using CD4 magnetic immunobeads: CD4-purified cells; unfractionated PBMC; and CD4-depleted PBMC. TP were determined in each cell fraction with liquid chromatography–mass spectrometry and compared across cell types by non-parametric analyses.

Results: Six males and two females participated. The median (range) percentage of CD4 T cells (CD4%) in each fraction were: CD4-purified, 99%; unfractionated, 63% (range, 53–70); and CD4-depleted, 14% (range, 4–29). Corresponding median (range) ZDV-TP concentrations were 8.0 (5.3–10.3), 26.5 (12.9–42.2), and 34.2 (16.4–52.2) fmol/1 × 106 cells (Friedman P = 0.0008). The 3TC-TP values were 4.6 (2.3–6.7), 4.8 (3.5–8.8), and 6.8 (4.0–13.1) pmol//1 × 106 cells (Friedman P = 0.01). In mixed model analyses: ZDV-TP (fmol/1 × 106 cells) = 42–0.32 (CD4%); P < 0.001 and 3TC-TP (pmol/1 × 106 cells) = 7.3–0.03(CD4%); P = 0.003.

Conclusions: In HIV-seronegative volunteers, 3TC-TP concentrations in PBMC reflected the concentrations within CD4 T cells, but ZDV-TP concentrations were more than 70% lower in CD4 T cells than in PBMC. Thus, TP concentrations differ according to cell type in vivo with corresponding efficacy and toxicity implications for cells with low or high triphosphates.

From the University of Colorado Health Sciences Center, Denver, Colorado, USA.

Received 14 December, 2006

Revised 11 March, 2007

Accepted 22 May, 2007

Correspondence to P.L. Anderson, University of Colorado Health Sciences Center, Box C238, 4200 E. 9th Ave., Denver, CO 80262, USA. Tel: +1 303 315 1720; e-mail:

Back to Top | Article Outline


Nucleoside analog reverse transcriptase inhibitors (NRTI) require stepwise phosphorylation to the intracellular 5′'-triphosphate (TP), which is the pharmacologically active moiety [1]. NRTI-phosphates are ion-trapped in cells and the overall intracellular accumulation of the NRTI-TP depends on multiple and complex cellular enzyme systems [1–6]. The eventual active TP then blocks viral replication through inhibition of the HIV reverse transcriptase enzyme [7]. The active TP is also a potential source of toxicity through inhibition of mitochondrial DNA polymerase γ [8].

NRTI responses are therefore dependent upon a distinctive therapeutic index where the capacity of individual cells to accumulate the NRTI-TP can be an important factor that influences antiretroviral efficacy and toxicity at the cell or tissue level [3]. Unfortunately, studies in patients to address this fundamental pharmacologic concept have yet to be undertaken [2].

To date, NRTI-TP in patients have only been quantified in peripheral blood mononuclear cells (PBMC), which contain monocytes, B-lymphocytes, and CD4/CD8 T-lymphocytes. Data in unfractionated PBMC from HIV-1-infected patients show elevated zidovudine (ZDV)- and lamivudine (3TC)-TP concentrations in severely CD4-depleted PBMC (clinically advanced patients) compared with PBMC with a higher CD4 cell percentage (clinically less advanced patients) [9,10]. The NRTI-TP concentration within CD4 cells is of specific interest as this is the cell type targeted by HIV. The goal of the present study was to determine whether concentrations of ZDV- and 3TC-TP in PBMC reflect the concentrations within CD4 cells from HIV-seronegative adults receiving ZDV and 3TC.

Back to Top | Article Outline


All subjects were participants in an ongoing study of ZDV- and 3TC-TP pharmacokinetics in PBMC. Subjects were eligible if they met the following criteria: 18–55 years of age; documented negative pregnancy, HIV ELISA, and Hepatitis B surface antigen tests; not receiving concomitant medications known to interact with ZDV and 3TC; and no preexisting laboratory abnormality outside Grade I according to specific guidelines [11]. The local Institutional Review Board (IRB) approved the study and written informed consent was obtained from all subjects.

Subjects received standard doses of ZDV (300 mg) and 3TC (150 mg), given as Combivir, twice daily for 12 days. Pharmacokinetic studies were performed during the first dose, and at day 3, day 7, and day 12 of therapy. Subjects arrived fasting to the Clinical Research Center on the morning before their prescribed morning dose, and a standardized breakfast was given (650 kcal; 15% protein, 45% fat, 40% carbohydrate). Upon completion of the meal, subjects were observed to take the ZDV/3TC dose and 19 mL blood were collected at 2, 5, and 8 h postdose.

For the present study, subjects provided an additional one-time blood draw of 60 mL under a protocol for PBMC donation to support NRTI-TP studies. This protocol was approved by the IRB for use in the parent study described above, and additional written informed consent was obtained from each participant. The 60 mL aliquot of blood was added to the 2- or 5-h postdose time point in the intensive pharmacokinetic visits on day 7 or day 12, while the subject was at pharmacokinetic steady state.

The blood was drawn into CPT tubes with heparin anticoagulant (Becton Dickinson, Franklin Lakes, New Jersey, USA) and PBMC were isolated according to the manufacturer's instructions. After washing and counting, a portion of PBMC was saved as unfractionated PBMC. The remaining PBMC were subjected to CD4 purification using microbeads with affinity to CD4 and a magnetic field (MACS, Miltenyi, Sunnyvale, California, USA) following the manufacturer's instructions. Briefly, PBMC were labeled with CD4 microbeads by incubation at 4°C for 15 min, washed, and loaded onto a magnetized separator column. CD4 cells were retained on the column and un-retained cells were collected as the CD4-depleted fraction. After removal of the column from the magnetic field, the CD4 cells were eluted and saved as the CD4-purified fraction. Each fraction was washed and an aliquot was set aside for flow cytometry (described below). All fractions were counted by the same laboratory scientist with a hemacytometer using Trypan Blue exclusion to assess cell viability. Fractions were lysed with 500 μl of 70: 30 methanol: water for storage at −80°C until analysis. Cells were kept on ice at every opportunity during the procedure.

Approximately 5 × 105 CD4-purified and CD4-depleted cells were labeled with directly conjugated antibodies: anti-CD4-fluroesceine isothiocyanate (FITC); anti-CD38-phycoerythrin (PE); anti-CD3-peridinin chlorophyll-a protein (perCP); and anti-HLA-DR-allophycocyanin (APC) antibodies (Becton Dickinson). The same procedure was carried out with 50 μl whole blood from the day 1 visit for the unfractionated PBMC fraction. Cells were examined on a FACSCalibur flow cytometer (Becton Dickinson) using a standard protocol and data were analyzed using CellQuest Pro software. The percentage of CD4 T cells was calculated by gating on the lymphocyte population and determining the fraction of CD3 cells that were also CD4. The percentage of activated CD4 T cells was also determined by examining the surface expression of HLA-DR/CD38 on the CD4 T cells.

ZDV- and 3TC-TP concentrations were quantified in all cell fractions with separate validated liquid chromatography-tandem mass spectroscopy (LC–MS–MS) methods. The complete method for ZDV-TP has been published [12]. The linear range of assay was 50–6400 fmol ZDV with a minimum quantifiable limit of 50 fmol (5 fmol/1 × 106 cells when 10 × 106 cells were analyzed). Both the between-day and within-day coefficients of variation (CV) for the quality controls during validation were < 20% at all concentrations. A direct LC–MS–MS method was developed and validated for 3TC-TP quantification. An internal standard, dideoxycytidine-TP, was added to cell lysates and samples were dried under nitrogen then reconstituted in HPLC-grade water. Chromatographic separation was performed on a Phenomenex Luna Phenyl-Hexyl, 2.0 × 150 mm column. The mobile phase consisted of 90% mobile phase A [20 mM triethylamine (TEA), 0.1% formic acid pH 9.0] and 10% mobile phase B (50: 50 acetonitrile: 40mM TEA, 0.2% formic acid pH 9.0). Detection and quantification of 3TC-TP and the internal standard were achieved by MS–MS detection in negative ion mode with electro-spray ionization. The assay was linear in the range of 5–500 pmol with a minimum quantifiable limit of 5 pmol (1 pmol/1 × 106 cells with 5 × 106 cells analyzed). Both the between-day and within-day CV and percentage deviation for the quality controls were < 15%.

A Friedman rank sum test was used initially to test if TP in any one of the cell fractions were different than in at least one of the other fractions. If the two-sided P value was < 0.05, Wilcoxon signed rank tests were used to make pairwise comparisons. Data are reported as median (range).

Back to Top | Article Outline


Six males and two females participated. Two volunteers were African–American (one female), four were Caucasian (one female), and one each was Hispanic and Asian. Their median age and weight were 27 (23–54) years and 80 (62–113) kg.

The range of time to completely process all blood samples was 3.1–3.5 h. All cell fractions were > 95% viable just before lysing and the three fractions were lysed at the same time. Cell counting was tested for precision and found to be reproducible with an average CV of 7%. The final cell counts (× 106) were 20 (13–37) for CD4-purified cells, 25 (13–37) for unfractionated PBMC, and 60 (47–136) for CD4-depleted PBMC. The percentage of CD4 T cells in each fraction was 99% (98–100) for CD4-purified cells, 63% (53–70) for unfractionated PBMC, and 14% (4–29) for CD4-depleted PBMC. Median HLA-DR and CD38 expression was 1–2% across all fractions with a range of 0–8%.

TP concentrations were quantified in all samples. ZDV-TP concentrations in the CD4-purified cells, unfractionated PBMC, and CD4-depleted PBMC were 8.0 (5.3–10.3), 26.5 (12.9–42.2) and 34.2 (16.4–52.2) fmol/1 × 106 cells, respectively (Freidman P = 0.0008). In the pairwise comparisons, statistically significant differences were observed for CD4-purified cells versus unfractionated PBMC and CD4-purified cells versus CD4-depleted PBMC (both P = 0.008), whereas that for unfractionated PBMC versus CD4-depleted PBMC was P = 0.05. The inter-subject variability in ZDV-TP was lower in the CD4-purified cells (CV = 22%) compared with the unfractionated PBMC (CV = 42%) and CD4-depleted PBMC (CV = 37%). In a mixed model to account for repeated measures across subjects, ZDV-TP (fmol/106 cells) = 42–0.32(CD4%); P < 0.001. Fig. 1 depicts the ZDV-TP concentrations according to cell type and % CD4 cells.

The 3TC-TP concentrations in the CD4-purified cells, unfractionated PBMC, and CD4-depleted PBMC were 4.6 (2.3–6.7), 4.8 (3.5–8.8), and 6.8 (4.0–13.1) pmol/1 × 106 cells, respectively (Friedman P = 0.01). In the pairwise comparisons, the significance levels were: CD4-purified cells versus unfractionated PBMC, P = 0.15; CD4-purified cells versus CD4-depleted PBMC, P = 0.008; and unfractionated PBMC versus CD4-depleted PBMC, P = 0.04. The inter-subject variability in 3TC-TP was highest in CD4-depleted PBMC (CV = 41%) and the same in CD4-purified cells and unfractionated PBMC (CV = 33%). In a similar mixed model, 3TC-TP (pmol/1 × 106 cells) = 7.3–0.03 (CD4%); P = 0.003. Fig. 2 shows the 3TC-TP concentrations according to cell type and % CD4 cells.

Back to Top | Article Outline


In HIV-seronegative adults receiving ZDV and 3TC, we found that 3TC-TP concentrations in PBMC reflected concentrations within CD4 cells, but ZDV-TP concentrations were ≥70% lower in CD4 cells versus the other cell fractions (P = 0.008). TP concentrations for both drugs generally decreased as the CD4% in the PBMC increased (Figs 1 and 2). The ZDV- and 3TC-TP concentrations in unfractionated PBMC in this study were comparable to those in other studies of healthy volunteers [9,13,14]. Studies are now needed to evaluate other NRTI-TP in CD4 cells and to investigate other cell types from PBMC or other tissues.

The impetus for this study arose from a previous observation that ZDV-TP concentrations in unfractionated PBMC were approximately threefold higher in patients with depleted CD4 cells (< 100 CD4/μl) versus those with relatively preserved CD4 cells (> 100 CD4/μl) [10]. These high TP concentrations subsequently declined by two-thirds in unfractionated PBMC during the course of antiretroviral therapy in association with therapeutic response and a rise in CD4 cells. The same relationship was evident for 3TC-TP, but to a lesser degree. Interestingly, these TP fluctuations were not observed in subjects who initiated therapy with > 100 CD4 cells at baseline. The present study was designed to determine whether ZDV- and 3TC-TP in PBMC reflect the concentrations in CD4 cells in vivo. Figs 1b and 2b can be used to inform about how ZDV- and 3TC-TP change as a function of the CD4 percentage in the PBMC sample. The patients in the previous study had approximately 25 CD4/μl (6% CD4) at baseline, which increased to about 194/μl (14%) at 1 year [10]. According to the equation shown in Fig. 1b, predicted ZDV-TP concentrations would change from approximately 40 to 38 fmol/1 × 106 cells in PBMC with 6% CD4 cells versus 14% CD4 cells, which is only a ∼5% difference. This alone does not adequately explain the threefold ZDV-TP differences observed in the patients from the previous study [10].

Another hypothesis to explain higher ZDV- and 3TC-TP in advanced HIV patients is elevated cellular activation associated advanced disease. Cell activation has been shown to increase ZDV and 3TC phosphorylation in vitro (nevertheless, the anti-HIV activity of 3TC-TP may be better in resting cells due to relatively low deoxycytidine-TP in these cells) [9,10,15,16]. While this concept of activation-dependent NRTI-TP concentrations is well accepted in vitro, it is not known how this translates to patients. In the present study, we investigated HIV-negative volunteers with low cellular activation (median of 1–2% HLA-DR and CD38 CD4 cells). Other studies have shown that ZDV-monophosphate, TP, and total phosphates are lower in PBMC from HIV-negative volunteers versus HIV-infected persons by 70–90%, which is consistent with HIV-associated cellular activation increasing ZDV phosphorylation in vivo [9,14]. These findings are also consistent with another study that measured lower thymidine kinase 1 activity (TK1) in unfractionated PBMC of HIV-negative volunteers versus HIV-infected patients [17]. TK1 catalyzes the conversion of ZDV to ZDV-monophosphate. The authors found no correlation between TK1 activity and CD4 cell count in this study. Taken together, cellular activation and increased TK1 activity (and/or other enzyme activity) associated with HIV infection provides a plausible explanation for the large elevations in ZDV-phosphates among HIV-infected patients compared with HIV-negative volunteers, and in patients with clinically-advanced versus clinically less advanced disease.

Low or high NRTI-TP levels in different cell types and/or according to cellular-activation in patients has several potentially important clinical implications. First, NRTI-TP concentrations within CD4 cells represent the actual site of action and should best correspond with efficacy. The present study showed that ZDV-TP concentrations in CD4-purified cells of HIV-negative volunteers were ≥70% lower than in unfractionated PBMC and CD4-depleted PBMC. Although ZDV has proven prophylactic efficacy in HIV negative persons, this finding suggests that NRTI-TP that concentrate better in CD4 cells in vivo, such as shown for 3TC-TP in this study, may substantially improve antiviral effect. Second, additional work is now needed to determine whether cellular activation in HIV-infected patients affects TP concentrations. The importance of this concept may be evident in the surprisingly poor efficacy of the triple-NRTI regimens that included tenofovir plus 3TC, and either didanosine or abacavir [18,19]. None of the TP of these drugs are preferentially accumulated in activated cells (relative to the endogenous counterparts), which allows for the possibility that low NRTI-TP in the activated cell compartment led to resistance and regimen failure [20,21]. In support of this, other studies have shown better virologic outcomes with triple-NRTI regimens including ZDV, which targets activated cells [22]. Another clinical implication of low TP in certain cellular compartments is persistent and ongoing rounds of HIV replication even in patients with < 50 copies/ml. Future research should investigate whether low NRTI-TP levels in certain cell types may provide an explanation for ongoing viremia. Lastly, NRTI-TP concentrations in certain tissues that significantly exceed concentrations in other tissues would provide a pharmacologic basis for tissue-dependent toxicities such as, lipoatrophy, peripheral neuropathy, and pancreatitis [3].

In summary, we found differences in ZDV- and 3TC-TP concentrations according to cell type in HIV-seronegative adults. Further studies are needed to establish the mechanism and clinical implications of these findings with the ultimate goal of utilizing NRTI in the most informed and rational manner possible.

Back to Top | Article Outline


Supported by Grants R01 AI64029 (PLA), AI33835 (CVF), P30-AI054907 (PLA and CVF), the General Clinical Research Center Grant RR000051, and the American Association of Colleges of Pharmacy (PLA). The authors thank M. Gerschenson, K. Lichtenstein, E. Connick, E. Gardner, T. Campbell, and S. MaWhinney for assistance in the design and implementation of the main study, B. Palmer for assistance with flow cytometry, A. Le for assistance with the MACS procedure, and the study subjects for volunteering to participate.

Back to Top | Article Outline


1. Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clin Ther 2000; 22:685–708.
2. Hoggard PG, Back DJ. Intracellular pharmacology of nucleoside analogues and protease inhibitors: role of transporter molecules. Curr Opin Infect Dis 2002; 15:3–8.
3. Anderson PL, Kakuda TN, Lichtenstein KA. The cellular pharmacology of nucleoside- and nucleotide-analogue reverse-transcriptase inhibitors and its relationship to clinical toxicities. Clin Infect Dis 2004; 38:743–753.
4. Yao SY, Ng AM, Sundaram M, Cass CE, Baldwin SA, Young JD. Transport of antiviral 3′-deoxy-nucleoside drugs by recombinant human and rat equilibrative, nitrobenzylthioinosine (NBMPR)-insensitive (ENT2) nucleoside transporter proteins produced in Xenopus oocytes. Mol Membr Biol 2001; 18:161–167.
5. Yao SY, Ng AM, Loewen SK, Cass CE, Baldwin SA, Young JD. An ancient prevertebrate Na+-nucleoside cotransporter (hfCNT) from the Pacific hagfish (Eptatretus stouti) 10.1152/ajpcell.00587.2001. Am J Physiol Cell Physiol 2002; 283:C155–C168.
6. Schuetz JD, Connelly MC, Sun D, Paibir SG, Flynn PM, Srinivas RV, et al. MRP4: A previously unidentified factor in resistance to nucleoside-based antiviral drugs. Nat Med 1999; 5:1048–1051.
7. Furman PA, Fyfe JA, St Clair MH, Weinhold K, Rideout JL, Freeman GA, et al. Phosphorylation of 3′-azido-3′-deoxythymidine and selective interaction of the 5′-triphosphate with human immunodeficiency virus reverse transcriptase. Proc Natl Acad Sci USA 1986; 83:8333–8337.
8. Johnson AA, Ray AS, Hanes J, Suo Z, Colacino JM, Anderson KS, et al. Toxicity of antiviral nucleoside analogs and the human mitochondrial DNA polymerase. J Biol Chem 2001; 276:40847–40857.
9. Barry M, Wild M, Veal G, Back D, Breckenridge A, Fox R, et al. Zidovudine phosphorylation in HIV-infected patients and seronegative volunteers. AIDS 1994; 8:F1–5.
10. Anderson PL, Kakuda TN, Kawle S, Fletcher CV. Antiviral dynamics and sex differences of zidovudine and lamivudine triphosphate concentrations in HIV-infected individuals. AIDS 2003; 17:2159–2168.
11. Division of AIDS table for grading severity of adult adverse experiences. National Institutes of Allergy and Infectious Disease. December 2004.
12. King T, Bushman L, Anderson PL, Delahunty T, Ray M, Fletcher CV. Quantitation of zidovudine triphosphate concentrations from human peripheral blood mononuclear cells by anion exchange solid phase extraction and liquid chromatography-tandem mass spectroscopy; an indirect quantitation methodology. J Chromatogr B Analyt Technol Biomed Life Sci 2006; 831(1–2):248–257.
13. Yuen GJ, Lou Y, Bumgarner NF, Bishop JP, Smith GA, Otto VR, et al. Equivalent steady-state pharmacokinetics of lamivudine in plasma and lamivudine triphosphate within cells following administration of lamivudine at 300 milligrams once daily and 150 milligrams twice daily. Antimicrob Agents Chemother 2004; 48:176–182.
14. Wattanagoon Y, Na Bangchang K, Hoggard PG, Khoo SH, Gibbons SE, Phiboonbhanakit D, et al. Pharmacokinetics of zidovudine phosphorylation in human immunodeficiency virus-positive Thai patients and healthy volunteers. Antimicrob Agents Chemother 2000; 44:1986–1989.
15. Gao WY, Shirasaka T, Johns DG, Broder S, Mitsuya H. Differential phosphorylation of azidothymidine, dideoxycytidine, and dideoxyinosine in resting and activated peripheral blood mononuclear cells. J Clin Invest 1993; 91:2326–2333.
16. Gao WY, Agbaria R, Driscoll JS, Mitsuya H. Divergent antihuman immunodeficiency virus activity and anabolic phosphorylation of 2′,3′-dideoxynucleoside analogs in resting and activated human cells. J Biol Chem 1994; 269:12633–12638.
17. Turriziani O, Butera O, Gianotti N, Parisi SG, Mazzi R, Girardi E, et al. Thymidine kinase and deoxycytidine kinase activity in mononuclear cells from antiretroviral-naive HIV-infected patients. AIDS 2005; 19:473–479.
18. Gallant JE, Rodriguez AE, Weinberg WG, Young B, Berger DS, Lim ML, et al. Early virologic nonresponse to tenofovir, abacavir, and lamivudine in HIV-infected antiretroviral-naive subjects. J Infect Dis 2005; 192:1921–1930.
19. Jemsek J, Hutcherson P, Harper E. Poor virologic responses and early emergence of resistance in treatment naive, HIV-infected patients receiving a once daily triple nucleoside regimen of didanosine, lamivudine and tenofovir DF. Eleventh Conference on Retroviruses and Opportunistic Infections. San Francisco, February 2004 [abstract 51].
20. Anderson PL, Kakuda TN. Comment on: suboptimal CD4 gains in HIV-infected patients receiving didanosine plus tenofovir. J Antimicrob Chemother 2006; 58:220–221.
21. Kuritzkes DR. Less than the sum of its parts: failure of a tenofovir-abacavir-lamivudine triple-nucleoside regimen. J Infect Dis 2005; 192:1867–1868.
22. Gulick RM, Ribaudo HJ, Shikuma CM, Lustgarten S, Squires KE, Meyer WA III, et al, and the AIDS Clinical Trials Group Study A5095 Team. Triple-nucleoside regimens versus efavirenz-containing regimens for the initial treatment of HIV-1 infection 10.1056/NEJMoa031772. N Engl J Med 2004; 350:1850–1861.
© 2007 Lippincott Williams & Wilkins, Inc.