crossmark
Analysis of the Endogenous Deoxynucleoside Triphosphate Pool in HIV-Positive and -Negative Individuals Receiving TenofovirEmtricitabine Xinhui Chen,a Jose R. Castillo-Mancilla,b Sharon M. Seifert,a Kevin B. McAllister,c Jia-Hua Zheng,a Lane R. Bushman,a Samantha MaWhinney,d Peter L. Andersona University of Colorado, Skaggs School of Pharmacy and Pharmaceutical Sciences, Department of Pharmaceutical Sciences, Aurora, Colorado, USAa; University of Colorado, School of Medicine, Division of Infectious Diseases, Aurora, Colorado, USAb; University of Colorado, School of Medicine, Aurora, Colorado, USAc; University of Colorado, Colorado School of Public Health, Department of Biostatistics and Informatics, Aurora, Colorado, USAd
Tenofovir (TFV) disoproxil fumarate (TDF) and emtricitabine (FTC), two nucleos(t)ide analogs (NA), are coformulated as an anti-HIV combination tablet for treatment and preexposure prophylaxis (PrEP). TDF/FTC may have effects on the deoxynucleoside triphosphate (dNTP) pool due to their similar structures and similar metabolic pathways. We carried out a comprehensive clinical study to characterize the effects of TDF/FTC on the endogenous dNTP pool, from baseline to 30 days of TDF/FTC therapy, in both treatment-naive HIV-positive and HIV-negative individuals. dATP, dCTP, dGTP, and TTP were quantified in peripheral blood mononuclear cells (PBMC) with a validated liquid chromatography-tandem mass spectrometry (LC-MS/MS) methodology. Forty individuals (19 HIV-positive) were enrolled and underwent a baseline visit and then received TDF/FTC for at least 30 days. Longitudinal measurements were analyzed using mixed-model segmented linear regression analysis. The dNTPs were reduced by 14% to 37% relative to the baseline level within 3 days in both HIV-negative and HIV-positive individuals (P < 0.003). These reductions persisted to various degrees at day 30. These findings indicate that dNTP pools are influenced by TDF/ FTC therapy. This may alter cellular homeostasis and could increase the antiviral effect through a more favorable analog/dNTP ratio. Further work is needed to elucidate mechanisms, to evaluate the clinical significance of these findings, and to further probe differences between HIV-negative and HIV-positive individuals. (This study has been registered at ClinicalTrials.gov under identifier NCT01040091.)
T
he coformulated medication consisting of 300 mg of tenofovir (TFV) disoproxil fumarate (TDF) and 200 mg of emtricitabine (FTC) is marketed as an antiviral combination tablet for treatment and preexposure prophylaxis (PrEP) of HIV infection (1). TFV is a nucleotide analog, and its diphosphate anabolite (TFVDP) has a structure similar to that of dATP; FTC is a nucleoside analog, and its trisphosphate (FTC-TP) has a structure similar to that of dCTP. TFV-DP and FTC-TP compete with dATP and dCTP (natural substrates) at the active site of HIV reverse transcriptase (RT), effectively inhibiting the biosynthesis of HIV genetic material. Once incorporated, they terminate the elongation of the HIV DNA chain due to the lack of a 3= hydroxyl group to incorporate the next component (2). The ratio between drug concentration and the corresponding deoxynucleoside triphosphate (dNTP) affects the pharmacologic efficacy of TDF/FTC as a high ratio has been associated with greater antiviral activity (3, 4). It is well known that nucleos(t)ide analogs (NAs) can affect the endogenous dNTP pool, including dATP and dCTP, as well as dGTP, and TTP. NAs may compete with the host enzyme system for phosphorylation, as well as influence the complex dNTP pool metabolism pathways. For example, in vitro, TFV or its anabolites competitively inhibit creatine kinase (CK), phosphoglycerate kinase (PGK), and purine nucleoside phosphorylase (PNP) (5, 6) and upregulate 5= ecto-nucleotidases and 5= nucleotidase (NT) (7). FTC and a similar deoxy-pyrimidine nucleoside analog, lamivudine (3TC), inhibit deoxycytidine kinase (dCK) (8, 9). In addition, zidovudine (AZT), a thymidine analog, has been shown to inhibit thymidylate kinase 2 (TK2) (10, 11). In HIV-infected patients, dATP increased during TDF/abacavir (ABC) dual therapy
September 2016 Volume 60 Number 9
compared to the level with TDF monotherapy (12) and decreased during didanosine (ddI)-containing therapy relative to the level with TDF-containing therapy (13). Taking these observations together, it is important to investigate the effects of TDF/FTC therapy on the endogenous dNTP pool in clinical settings. The use of TDF/FTC for PrEP, as well as long-term treatment, requires a low toxicity profile for an acceptable risk-benefit balance (14). TDF/FTC are generally safe and well tolerated (15); however, uncommon side effects have been reported, particularly in HIV-infected patients. For example, TDF is rarely associated with mitochondrial dysfunction in renal proximal tubule cells (16, 17). Mitochondrial dysfunction can arise from an imbalance among purine and pyrimidine nucleotides (11, 18, 19). Additionally, the TDF/ddI combination is correlated with a paradoxical CD4 T cell decline (20) that may be explained by PNP inhibition and subsequent apoptosis, which is associated with excessive in-
Received 11 May 2016 Accepted 23 June 2016 Accepted manuscript posted online 27 June 2016 Citation Chen X, Castillo-Mancilla JR, Seifert SM, Mcallister KB, Zheng J-H, Bushman LR, MaWhinney S, Anderson PL. 2016. Analysis of the endogenous deoxynucleoside triphosphate pool in HIV-positive and -negative individuals receiving tenofovir-emtricitabine. Antimicrob Agents Chemother 60:5387–5392. doi:10.1128/AAC.01019-16. Address correspondence to Peter L. Anderson, [email protected]. Supplemental material for this article may be found at http://dx.doi.org/10.1128 /AAC.01019-16. Copyright © 2016, American Society for Microbiology. All Rights Reserved.
Antimicrobial Agents and Chemotherapy
aac.asm.org
5387
Chen et al.
FIG 1 Clinical study design. TDF, tenofovir disoproxil fumarate; FTC, emtricitabine; EFV, efavirenz; 1 to 60, study visits in days.
tracellular dATP and dGTP (19, 21). FTC moderately reduces cell proliferation in vitro (22), which may be associated with imbalanced dNTP pools. The characterization of the dNTP pool changes in patients receiving TDF/FTC enables the quantitation of the analog/dNTP ratio for pharmacologic efficacy and provides possible mechanisms of adverse effects. The goal of this pharmacodynamic study was to investigate the time profile of the dNTP pool, from baseline to TDF/FTC pharmacological intracellular steady state, in both HIV-positive and HIV-negative individuals. MATERIALS AND METHODS Participants and study design. The clinical protocol was approved by the institutional review broad (IRB) of the University of Colorado, and participants provided informed consent (Cell-PrEP trial; ClinicalTrials.gov identifier NCT01040091). HIV-negative adults were enrolled in an intensive clinical pharmacology study of daily coformulated TDF/FTC treatment for 30 days, followed by 30 days of washout. HIV-positive adults initiated TDF/FTC/efavirenz (EFV) treatment for 60 days (and beyond). All participants were tested for hepatitis B virus and were excluded if they were positive. Individuals were excluded if they were pregnant (or planning pregnancy), breastfeeding, had a body weight of less than 110 pounds, a modification of the diet in renal disease (MDRD) estimated glomerular filtration rate (eGFR) of less than 60 ml/min/1.73 m2, a history of pathological bone fractures, an albuminuria creatinine ratio of more than 30, or a history of kidney disease. Participants’ age, weight, sex, race, and body mass index (BMI) were recorded upon enrollment in the study. Peripheral blood mononuclear cell (PBMC) samples were taken at baseline and at 8 h postdose on days 1, 3, 7, 20, 30, and 60 in all participants. The HIV-negative group had two additional visits on days 35 and 45 during the washout period. The study design is illustrated in Fig. 1. PBMC processing. A previously described method was used for PBMC processing (23). Blood was drawn into a cell preparation tube (CPT). After the sample was mixed, the tube was spun at 1,800 ⫻ g for 30 min at room temperature to separate plasma, PBMC, and red blood cells (RBC). The buffy layer (PBMCs) between the plasma and separation medium was collected into a 15-ml centrifuge tube. After RBC lysis to eliminate potential RBC contamination, the sample was washed with an equal volume of phosphate-buffered saline (PBS). The cell sample was then spun, and the cell pellet was resuspended in 5 ml of PBS for automated cell counting (Countess cell counter; Invitrogen/Thermo Fisher Scientific Corporation, Carlsbad, CA). Finally, the cells were spun again to pellet and lysed in 500 l of 70:30 methanol-water. The lysate was stored at ⫺80°C until analysis. The dNTP pool quantitation. The analytical method utilized a strong anion exchange isolation of mono-, di-, and triphosphates (MP, DP, and TP, respectively) from the intracellular matrix (23). The TP fraction was then dephosphorylated to the parent moiety (deoxynucleoside [dN]) yielding a molar equivalent to the original deoxynucleotide intracellular concentration. A reverse phase and weak cation exchange mixed-mode purification was then used for desaltation and concentration. Endogenous dNTP levels were then quantified using a validated liquid chromatography-tandem mass spectrometry (LC-MS/MS) method with an analytical range of 50 to 2,500 fmol/sample (24).
5388
aac.asm.org
Statistical analysis. Data were natural log (ln) transformed for statistical analysis. The intracellular levels of the dNTPs (dATP, dCTP, dGTP, and TTP) were reported as geometric means and 95% confidence intervals (CIs). All statistical analyses were performed using SAS (version 9.4; SAS Institute, Inc., Cary, NC, USA). To investigate the effects of HIV infection on the dNTP pool, the baseline data were analyzed. A Welch’s t test was used to compare the dNTPs between HIV-negative and HIV-positive groups, and an F test was used to compare the difference of variances. Linear regression was performed to probe for the effects of subject variables at baseline using continuous variables such as age, weight, and BMI, while a Welch’s t test and analysis of variance (ANOVA) were performed on categorical variables such as sex and race. To evaluate the effects of TDF/FTC on the dNTP pool, longitudinal measurements from baseline to TDF/FTC pharmacological steady state in PBMCs (day 0 to 30) were used in both HIV-positive and HIV-negative groups. Time profiles of the dNTP pool were described and analyzed using mixed-model segmented linear regression (SAS Mixed procedure) (model equations are given in the supplemental material) (25). The model estimated baseline, first slope, the change point (of two slopes), and the second slope. The change point was estimated by iteration (0 to 30 days, graduated by 0.05 in days), based on model goodness of fit (Akaike information criterion [AIC]) (26). The variance structures (e.g., compound symmetry, heterogeneous compound symmetry, autoregressive model, and unknown structure) were assessed and selected also by the model goodness of fit. The differences between the change point versus baseline, day 30 versus baseline, and the change point versus day 30 were tested using SAS Estimate statements, and results were summarized as percentage change. Subject covariates were tested using a likelihood ratio test (LRT). To evaluate the change of the dNTP pool during the washout period in the HIV-negative group, day 35, 45, and 60 data were compared to day 30 data, and day 60 data were also compared to the baseline level. A paired t test was used to compare the differences at two selected time points. To investigate the long-term treatment effects of TDF/FTC in the HIV-positive group, baseline and day 60 data were compared.
RESULTS
Study population. A total of 40 participants were enrolled in the study. Among them, 21 were HIV-negative and 19 were HIVpositive patients. Thirty-four subjects completed all study visits, but six participants, two HIV-negative and four HIV-positive subjects (one female and five males), stopped early. One participant (HIV-positive) had a baseline sample but withdrew prior to any dosing visits, and the other five withdrew prior to the day 30 visit. One HIV-positive participant did not have a baseline sample but completed all visits. The study drug was generally well tolerated. The demographic characteristics of the study population are summarized in Table 1. Baseline characteristics of the dNTP pool. Data are presented in geometric means and 95% CIs in Table 2. At baseline in PBMCs (n ⫽ 39), it was observed that the HIV-negative group generally had higher dNTP pool levels than the HIV-positive group, with the exception of dATP levels. However, these differences were not statistically significant (P ⱖ 0.21). Statistically significant differences in variances were observed in all four dNTPs (P ⱕ 0.04), with higher variability in the HIV-negative group. No subject covariates (including age, weight, BMI, gender, and race) had a statistically significant effect associated with baseline dNTP levels (P ⱖ 0.17). Changes in the dNTP pool on TDF/FTC therapy. Figure 2 shows data from each individual, the model predicted population means, and the 95% confidence intervals. Model parameter statis-
Antimicrobial Agents and Chemotherapy
September 2016 Volume 60 Number 9
Effects of TDF/FTC on dNTPs
TABLE 1 Demographic characteristics of the study population Parametera
Value for the parameter
HIV status (no. of patients [%]) Negative Positive
21 (53) 19 (47)
Gender (no. of patients [%]) Female Male
13 (33) 27 (67)
Race (no. of patients [%]) White Black Hispanic
19 (47) 16 (40) 5 (13)
Median eGFR (ml/min/1.73 m2 [range]) Median age (yr [range]) Median wt (kg [range]) Median BMI (kg/m2 [range]) No. of participants (no. of withdrawals)
93.3 (66.0–130.8) 31 (20–52) 81.1 (56.5–127) 26.6 (19.9–37.7) 40 (6)
a
eGFR, estimated glomerular filtration rate; BMI, body mass index.
tics are summarized in Table 3. For all four models, a heterogeneous, compound symmetry (CSH) covariate structure best described the data. Throughout the study, no effects of subject covariates on intercepts or slopes (as interaction terms) were observed (P ⱖ 0.24). Change in dCTP level. The change point of slopes for dCTP was day 1.55 with a decrease of 20% (95% CI, ⫺28% to ⫺11%; P ⬍ 0.0001) from baseline. Compared with the baseline, dCTP was decreased by 15% (95% CI, ⫺23% to ⫺6.4%; P ⫽ 0.001) at day 30. After day 1.55, concentrations increased slightly between the change point and day 30 but was not statistically significant (5.8%; 95% CI, ⫺2.4% to 15%; P ⫽ 0.17). Change in TTP level. The change point of slopes for TTP was reached at day 1.4 with a decrease of 37% (95% CI, ⫺46% to ⫺26%; P ⬍ 0.0001) from baseline. This 37% decrease remained at day 30 (95% CI, ⫺47% to ⫺27%; P ⫽ 0.001). The change between day 1.4 and day 30 was not statistically significant (⫺1.2%; 95% CI, ⫺15% to 14%; P ⫽ 0.86). Change in dATP level. The change point of slopes for dATP was reached at 2.9 days with a decrease of 14% (95% CI, ⫺21% to ⫺4.8%; P ⫽ 0.003) from baseline. After that, although concentrations increased slightly, no statistically significant change between day 2.9 and day 30 was detected (6.5%; 95% CI, ⫺2.3% to 16%; P ⫽ 0.15). A decreasing trend was observed when day 30 was compared to the baseline value, with a reduction of 8.3% (95% CI, ⫺17% to 0.80%; P ⫽ 0.07). Change in dGTP level. The change point of slopes for dGTP
was reached at day 1.35 of therapy with a decrease of 19% (95% CI, ⫺27% to ⫺8.9%; P ⫽ 0.0004) from baseline. Compared with the baseline, there was a decrease of 11% (95% CI, ⫺20% to ⫺0.67%; P ⫽ 0.03) at day 30. A statistically significant increase was detected from the change point (day 1.35) to day 30, amounting to 9.6% (95% CI, 0.61% to 19%; P ⫽ 0.04). Change in the dNTP pool in the washout period. The HIVnegative group was assigned to a 30-day washout period after 30 days of treatment. Compared to day 30 data, which represented the pharmacological steady state of TDF/FTC in PBMCs, deoxypyrimidine triphosphates increased throughout the washout period. A statistically significant increase of 20% (95% CI, 3.8% to 40%; P ⫽ 0.02) in dCTP was observed 15 days after the initiation of the washout period. TTP increased by 83% (95% CI, 42% to 136%; P ⫽ 0.0001) at day 5, 72% (95% CI, 34% to 121%; P ⫽ 0.0004) at day 15, and 82% (95% CI, 33% to 149%; P ⫽ 0.002) at day 30 of the washout period. However, for deoxy-purine triphosphates, statistically significant changes in dATP and dGTP were not observed (P ⱖ 0.38). Day 60 data compared to the baseline level was not statistically significant in dCTP (geometric means, 934 versus 1,038 fmol/106 cells; n ⫽ 12; P ⫽ 0.48) and TTP (geometric means, 496 versus 641 fmol/106 cells; n ⫽ 11; P ⫽ 0.14). However, statistically significant differences were observed for dATP and dGTP, with geometric means of 130 versus 181 fmol/ 106 cells (n ⫽ 10; P ⫽ 0.04) and 243 versus 343 fmol/106 cells (n ⫽ 10; P ⫽ 0.04). These results suggest that dCTP and TTP returned to baseline levels around 5 to 15 days after treatment discontinuation, while dATP and dGTP remained below the baseline level until at least day 30 of the washout period. Change in the dNTP pool after 2 months of treatment. In the HIV-positive group, the day 60 results were compared to those at baseline. No statistically significant difference was observed in all four components (n ⫽ 15; P ⱖ 0.28). The geometric means (95% CI) in femtomoles per million cells were 747 (631, 886) for dCTP, 354 (266, 470) for TTP, 159 (131, 193) for dATP, and 270 (226, 323) for dGTP, levels which were comparable to baseline levels (Table 1). This suggested that the dNTP pool recovered after treatment in HIV-infected patients. DISCUSSION
In this study, we used PBMCs to investigate the effects of TDF/ FTC on the dNTP pool of HIV-positive and -negative adults. PBMCs include cells that are the major target of TDF/FTC (e.g., CD4 T cells). Also, these nucleated cells represent dNTP pool kinetics driven by enzymes such as PNP, which have high activity in PBMCs (6). At baseline, we observed generally lower dNTPs (except for dATP), but not a statistically significantly difference, in treat-
TABLE 2 Baseline characteristics of the dNTP pool Geometric mean amt (fmol/106 cells [95% CI])a
Variance of log-transformed data
dNTP
HIV-negative group
HIV-positive group
P
HIV-negative group
HIV-positive group
Pb
dCTP TTP dATP dGTP
858 (710, 1037) 436 (322, 589) 150 (120, 187) 287 (234, 352)
804 (724, 893) 380 (312, 462) 154 (141, 167) 251 (230, 274)
0.53 0.43 0.83 0.21
0.17 0.44 0.24 0.20
0.045 0.16 0.029 0.031
0.007 0.04 0.0001 0.003
a b
CI, confidence interval. An F test was used to compare variances.
September 2016 Volume 60 Number 9
Antimicrobial Agents and Chemotherapy
aac.asm.org
5389
Chen et al.
FIG 2 The dNTP changes during TDF/FTC therapy. 95% CI, model 95% confidence interval.
ment-naive HIV-infected participants than in HIV-negative participants. However, the variance of the dNTP pool was higher in the HIV-negative group than in the HIV-positive group. Although a higher T-cell activation and cell proliferation might initially lead to dNTP pool upregulation (27, 28), the pathological chronic cell activation during HIV infection (29) may induce anergy in T cells (30). It is also reported that compensatory dNTP-consuming mechanisms associated with NTPDase-1 and SAMHD-1 are upregulated, especially in uncontrolled HIV-1 infection (31, 32). These biological effects may have reduced the variability in dNTPs in HIV-infected participants. However, the small sample size (n ⫽ 39) of our study limited the ability to probe possible effects of HIV infection and subject covariates on baseline dNTP pools. Future studies require a larger sample size to further address these questions. In HIV-infected patients, Hawkins et al. reported a lower
dATP level with didanosine (ddI) than with other treatments, including TDF-including treatment (medians, 67.2 versus 186 fmol/ 106 cells; P ⫽ 0.001), but could not exclude the possibility that TDF is also related to dATP reduction (13). Goicoechea et al. reported an increasing trend in dATP and dGTP with ABC/TDF dual therapy; however, in their 7-day monotherapy period, they observed a decreasing trend in dGTP (medians, 4,026 versus 2,464 fmol/106 cells; P ⫽ 0.68) (12). The study of Goicoechea et al. reported higher dATP and dGTP levels than the study by Hawkins et al. and the present study. In a macaque model, Durand-Gasselin et al. reported no change in dATP and dGTP levels from 2 to 24 h post-oral TDF and subcutaneous TFV treatment, but this study was limited by single-dose administration, a shorter observation period (ⱕ24 h), and the lack of a baseline observation (33). Taken together, the present study represents a novel assessment of dNTP changes in PBMCs during TDF/FTC therapy.
TABLE 3 Summarization of model parameter statistics Parametera
dNTP
Estimate (95% CI)
P
Estimate (95% CI)
P
Estimate (95% CI)
P
Estimate (95% CI)
P
Change point (days)b
dCTP TTP dATP dGTP
6.73 (6.64, 6.83) 6.03 (5.86, 6.19) 5.08 (4.98, 5.17) 5.61 (5.50, 5.72)
⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001
⫺0.14 (⫺0.21, ⫺0.077) ⫺0.33 (⫺0.44, ⫺0.22) ⫺0.052 (⫺0.085, ⫺0.018) ⫺0.15 (⫺0.24, ⫺0.070)
⬍0.0001 ⬍0.0001 0.0026 0.0004
0.0020 (⫺0.00086, 0.0048) ⫺0.00044 (⫺0.0055, 0.0046) 0.0023 (⫺0.00085, 0.0055) 0.0031 (0.00018, 0.0061)
0.17 0.86 0.15 0.038
0.14 (0.077, 0.21) 0.33 (0.21, 0.44) 0.054 (0.019, 0.089) 0.16 (0.072, 0.24)
⬍0.0001 ⬍0.0001 0.0027 0.0003
1.55 1.4 2.9 1.35
Baseline
First slope
Second slope
Change in slope
a
Values are natural log transformed. 95% CI, 95% confidence interval of model estimates. Baseline values are given as ln(fmol/106 cells). Data for first slope, second slope, and change in slope are given as ln(fmol/106 cells)/day. b Estimated by iteration.
5390
aac.asm.org
Antimicrobial Agents and Chemotherapy
September 2016 Volume 60 Number 9
Effects of TDF/FTC on dNTPs
We observed that TDF/FTC therapy was associated with slight to moderate decreases in dNTP levels in both HIV-negative and HIV-positive individuals. A mixed-model segmented linear regression analysis was a useful approach to analyzing the dynamic changes in the dNTP pool during TDF/FTC treatment (34, 35). In comparison to nonlinear regression models, the statistical analysis of the second slope from the segmented linear regression model enabled us to investigate the possible change points and return of the dNTPs during therapy (a statistically significant return was detected in dGTP). This work indicates that the dNTPs decrease within a few days (⬍3 days) of TDF/FTC initiation. However, the change point was estimated based on the model goodness of fit, and these temporal approximations were limited by sampling times. The detected effects of TDF/FTC on the dNTP pool were (largest to smallest) TTP ⬎ dCTP ⬎ dGTP ⬎ dATP, indicating that TDF/FTC had larger effects on deoxy-pyrimidine triphosphate compounds (dCTP and TTP) than deoxy-purines (dGTP and dATP). If these were driven by FTC, it could due to higher intracellular concentrations of FTC triphosphate (FTC-TP) than of TFV diphosphate (TFV-DP) (36). We anticipate that lamivudine may induce similar effects, given the pharmacological similarity with emtricitabine (8, 9). Studies are needed to assess on the effects of other NA such as abacavir and zidovudine. In terms of potential mechanisms, the reduction of the dNTP pool could be explained by TFV-associated upregulation of the dephosphorylation enzymes 5=-ectonucleotidase and 5=-nucleotidase (7). TFV also suppresses the ATP-generating enzymes CK and PGK (5) and indirectly inhibits kinases that use ATP as the phosphate donor, such as nucleoside diphosphate kinase (NDPK) (37). Only in dGTP did we observe a statistically significant but minor return toward baseline values by day 30 after the change point was reached. Partial PNP inhibition by TFV phosphates might play a role in dGTP upregulation (38), perhaps counteracting some of the reduction effects. However, the decreased dATP and dGTP levels observed in this study do not support PNP inhibition as a dominant effect of TDF, which would theoretically induce an increase of dATP and dGTP. FTC effects may be partly explained by the inhibition of dCK, which had also been shown for another deoxycytidine analog, 3TC (8, 14). During the washout period in the HIV-negative group, dCTP and TTP returned to baseline levels, while dGTP and dATP stayed slightly below the baseline levels. TFV-DP exhibits a much longer intracellular half-life than FTC-TP, and previous research in this cohort observed quantifiable intracellular TFV-DP concentrations at day 30 of the washout phase, whereas FTC-TP was below the limit of quantitation (39). This is consistent with the suppression of dCTP and TTP being related to FTC-TP, while that of dATP and dGTP being related to TFV-DP. Future studies should investigate the individual effects of TDF and FTC on the dNTP pool in HIV-negative participants. For long-term treatment in the HIV-positive group, day 60 data showed less reduction in the dNTP pool level than the plateau or the change point level, which indicates that, in our participants, the endogenous dNTP pool slowly returned to baseline with longterm treatment. We still need more information to understand other factors that may affect the dNTP pool in PBMCs after longterm treatment in HIV-infected patients, such as disease progression reversal, immune system restoration, and immune cell activation changes (14, 40). In conclusion, during TDF/FTC therapy, the dNTP pool
September 2016 Volume 60 Number 9
slightly to moderately decreased within 3 days after the initiation of TDF/FTC therapy and generally remained below baseline at day 30. Potential clinical implications need to be evaluated in the future, including assessment of the analog/dNTP ratio over time, which influences the pharmacological efficacy of TDF/FTC (3, 4, 12). ACKNOWLEDGMENTS We thank Gilead Sciences, Inc., which provided the study drug. We also thank all of the volunteers who participated in the study, as well as the study personnel and Colorado Clinical Translational Research Center staff. This study was supported by NIH U01 AI84735, NIH U01 AI106499, T32 AI7447-23, and Colorado Clinical Translational Sciences Institute UL1 TR001082 and TL1 TR001081.
FUNDING INFORMATION This work, including the efforts of Xinhui Chen, José R. Castillo-Mancilla, Sharon M. Seifert, Kevin B. McAllister, Jia-Hua Zheng, Lane R. Bushman, Samantha MaWhinney, and Peter L. Anderson, was funded by HHS | NIH | NIH Office of the Director (OD) (U01 AI84735, U01 AI106499, T32 AI7447-23, UL1 TR001082, and TL1 TR001081).
REFERENCES 1. Grant RM, Lama JR, Anderson PL, McMahan V, Liu AY, Vargas L, Goicochea P, Casapia M, Guanira-Carranza JV, Ramirez-Cardich ME, Montoya-Herrera O, Fernandez T, Veloso VG, Buchbinder SP, Chariyalertsak S, Schechter M, Bekker LG, Mayer KH, Kallas EG, Amico KR, Mulligan K, Bushman LR, Hance RJ, Ganoza C, Defechereux P, Postle B, Wang F, McConnell JJ, Zheng JH, Lee J, Rooney JF, Jaffe HS, Martinez AI, Burns DN, Glidden DV, iPrEx Study Team. 2010. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med 363:2587–2599. http://dx.doi.org/10.1056/NEJMoa1011205. 2. Cihlar T, Ray AS. 2010. Nucleoside and nucleotide HIV reverse transcriptase inhibitors: 25 years after zidovudine. Antiviral Res 85:39 –58. http: //dx.doi.org/10.1016/j.antiviral.2009.09.014. 3. Margolis DM, Kewn S, Coull JJ, Ylisastigui L, Turner D, Wise H, Hossain MM, Lanier ER, Shaw LM, Back D. 2002. The addition of mycophenolate mofetil to antiretroviral therapy including abacavir is associated with depletion of intracellular deoxyguanosine triphosphate and a decrease in plasma HIV-1 RNA. J Acquir Immune Defic Syndr 31:45– 49. http://dx.doi.org/10.1097/00126334-200209010-00006. 4. Gao WY, Agbaria R, Driscoll JS, Mitsuya H. 1994. Divergent antihuman immunodeficiency virus activity and anabolic phosphorylation of 2=,3=-dideoxynucleoside analogs in resting and activated human cells. J Biol Chem 269:12633–12638. 5. Varga A, Graczer E, Chaloin L, Liliom K, Zavodszky P, Lionne C, Vas M. 2013. Selectivity of kinases on the activation of tenofovir, an anti-HIV agent. Eur J Pharm Sci 48:307–315. http://dx.doi.org/10.1016/j.ejps.2012 .11.007. 6. Ray AS, Olson L, Fridland A. 2004. Role of purine nucleoside phosphorylase in interactions between 2=,3=-dideoxyinosine and allopurinol, ganciclovir, or tenofovir. Antimicrob Agents Chemother 48:1089 –1095. http: //dx.doi.org/10.1128/AAC.48.4.1089-1095.2004. 7. Biswas N, Rodriguez-Garcia M, Crist SG, Shen Z, Bodwell JE, Fahey JV, Wira CR. 2013. Effect of tenofovir on nucleotidases and cytokines in HIV-1 target cells. PLoS One 8:e78814. http://dx.doi.org/10.1371/journal .pone.0078814. 8. Bethell R, De Muys J, Lippens J, Richard A, Hamelin B, Ren C, Collins P. 2007. In vitro interactions between apricitabine and other deoxycytidine analogues. Antimicrob Agents Chemother 51:2948 –2953. http://dx .doi.org/10.1128/AAC.01204-06. 9. Hazen R, Lanier ER. 2003. Relative anti-HIV-1 efficacy of lamivudine and emtricitabine in vitro is dependent on cell type. J Acquir Immune Defic Syndr 32:255–258. http://dx.doi.org/10.1097/00126334-200303010 -00003. 10. Morris GW, Iams TA, Slepchenko KG, McKee EE. 2009. Origin of pyrimidine deoxyribonucleotide pools in perfused rat heart: implications
Antimicrobial Agents and Chemotherapy
aac.asm.org
5391
Chen et al.
11.
12.
13.
14.
15.
16.
17. 18.
19. 20. 21.
22.
23.
24.
25.
for 3=-azido-3=-deoxythymidine-dependent cardiotoxicity. Biochem J 422:513–520. http://dx.doi.org/10.1042/BJ20082427. Rylova SN, Albertioni F, Flygh G, Eriksson S. 2005. Activity profiles of deoxynucleoside kinases and 5=-nucleotidases in cultured adipocytes and myoblastic cells: insights into mitochondrial toxicity of nucleoside analogs. Biochem Pharmacol 69:951–960. http://dx.doi.org/10.1016/j.bcp .2004.12.010. Goicoechea M, Jain S, Bi L, Kemper C, Daar ES, Diamond C, Ha B, Flaherty J, Sun S, Richman D, Louie S, Haubrich R, California Collaborative Treatment Group. 2010. Abacavir and tenofovir disoproxil fumarate co-administration results in a nonadditive antiviral effect in HIV-1infected patients. AIDS 24:707–716. http://dx.doi.org/10.1097/QAD .0b013e32833676eb. Hawkins T, Veikley W, Durand-Gasselin L, Babusis D, Reddy YS, Flaherty JF, Ray AS. 2011. Intracellular nucleotide levels during coadministration of tenofovir disoproxil fumarate and didanosine in HIV-1infected patients. Antimicrob Agents Chemother 55:1549 –1555. http://dx .doi.org/10.1128/AAC.00910-10. Anderson PL, Kiser JJ, Gardner EM, Rower JE, Meditz A, Grant RM. 2011. Pharmacological considerations for tenofovir and emtricitabine to prevent HIV infection. J Antimicrob Chemother 66:240 –250. http://dx .doi.org/10.1093/jac/dkq447. Mugwanya KK, Baeten JM. 2016. Safety of oral tenofovir disoproxil fumarate-based pre-exposure prophylaxis for HIV prevention. Expert Opin Drug Saf 15:265–273. http://dx.doi.org/10.1517/14740338.2016 .1128412. Woodward CL, Hall AM, Williams IG, Madge S, Copas A, Nair D, Edwards SG, Johnson MA, Connolly JO. 2009. Tenofovir-associated renal and bone toxicity. HIV Med 10:482– 487. http://dx.doi.org/10.1111 /j.1468-1293.2009.00716.x. Kakuda TN. 2000. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clin Ther 22:685– 708. http://dx.doi.org/10.1016/S0149-2918(00)90004-3. Desler C, Munch-Petersen B, Stevnsner T, Matsui S, Kulawiec M, Singh KK, Rasmussen LJ. 2007. Mitochondria as determinant of nucleotide pools and chromosomal stability. Mutat Res 625:112–124. http://dx.doi .org/10.1016/j.mrfmmm.2007.06.002. Oliver FJ, Collins MK, Lopez-Rivas A. 1996. dNTP pools imbalance as a signal to initiate apoptosis. Experientia 52:995–1000. http://dx.doi.org/10 .1007/BF01920108. Kakuda TN, Anderson PL, Becker SL. 2004. CD4 cell decline with didanosine and tenofovir and failure of triple nucleoside/nucleotide regimens may be related. AIDS 18:2442–2444. Bantia S, Ananth SL, Parker CD, Horn LL, Upshaw R. 2003. Mechanism of inhibition of T-acute lymphoblastic leukemia cells by PNP inhibitor— BCX-1777. Int Immunopharmacol 3:879 – 887. http://dx.doi.org/10.1016 /S1567-5769(03)00076-6. Venhoff N, Setzer B, Melkaoui K, Walker UA. 2007. Mitochondrial toxicity of tenofovir, emtricitabine and abacavir alone and in combination with additional nucleoside reverse transcriptase inhibitors. Antivir Ther 12:1075–1085. Bushman LR, Kiser JJ, Rower JE, Klein B, Zheng JH, Ray ML, Anderson PL. 2011. Determination of nucleoside analog mono-, di-, and triphosphates in cellular matrix by solid phase extraction and ultra-sensitive LC-MS/MS detection. J Pharm Biomed Anal 56:390 – 401. http://dx.doi .org/10.1016/j.jpba.2011.05.039. Chen X, McAllister KB, Zheng J-H, Castillo-Mancilla J, Klein B, Guida LA, Seifert SM, Bushman LR, Anderson PL. 2014. LC-MS/MS analysis of endogenous deoxy-nucleotide-triphosphate (dNTP) pools in HIVnegative versus HIV-positive adults, abstr PP-12. 15th Int Workshop Clin Pharmacol HIV Hepat Ther, Washington, DC, 19 to 21 May 2014. Torti C, Lapadula G, Barreiro P, Soriano V, Mandalia S, De Silvestri A, Suter F, Maggiolo F, Antinori A, Antonucci F, Maserati R, El Hamad I, Pierotti P, Sighinolfi L, Migliorino G, Ladisa N, Carosi G. 2007. CD4⫹
5392
aac.asm.org
26.
27. 28. 29. 30. 31.
32. 33.
34.
35. 36.
37.
38. 39.
40.
T cell evolution and predictors of its trend before and after tenofovir/ didanosine backbone in the presence of sustained undetectable HIV plasma viral load. J Antimicrob Chemother 59:1141–1147. http://dx.doi .org/10.1093/jac/dkm100. Greene ME, Rolfson O, Garellick G, Gordon M, Nemes S. 2015. Improved statistical analysis of pre- and post-treatment patient-reported outcome measures (PROMs): the applicability of piecewise linear regression splines. Qual Life Res 24:567–573. http://dx.doi.org/10.1007/s11136 -014-0808-3. Nicander B, Reichard P. 1985. Relations between synthesis of deoxyribonucleotides and DNA replication in 3T6 fibroblasts. J Biol Chem 260: 5376 –5381. Galmarini CM, Mackey JR, Dumontet C. 2001. Nucleoside analogues: mechanisms of drug resistance and reversal strategies. Leukemia 15:875– 890. http://dx.doi.org/10.1038/sj.leu.2402114. Silvestri G, Feinberg MB. 2003. Turnover of lymphocytes and conceptual paradigms in HIV infection. J Clin Invest 112:821– 824. http://dx.doi.org /10.1172/JCI19799. Gougeon ML. 1995. T cell apoptosis as a consequence of chronic activation of the immune system in HIV infection. Adv Exp Med Biol 374:121– 127. http://dx.doi.org/10.1007/978-1-4615-1995-9_11. Leal DB, Streher CA, Bertoncheli CDM, Carli LF, Leal CA, da Silva JE, Morsch VM, Schetinger MR. 2005. HIV infection is associated with increased NTPDase activity that correlates with CD39-positive lymphocytes. Biochim Biophys Acta 1746:129 –134. http://dx.doi.org/10.1016/j .bbamcr.2005.10.009. Ahn J. 2016. Functional organization of human SAMHD1 and mechanisms of HIV-1 restriction. Biol Chem 397:373–379. http://dx.doi.org/10 .1515/hsz-2015-0260. Durand-Gasselin L, Van Rompay KK, Vela JE, Henne IN, Lee WA, Rhodes GR, Ray AS. 2009. Nucleotide analogue prodrug tenofovir disoproxil enhances lymphoid cell loading following oral administration in monkeys. Mol Pharm 6:1145–1151. http://dx.doi.org/10.1021 /mp900036s. Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. 2002. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther 27:299 –309. http://dx.doi.org/10.1046/j.1365 -2710.2002.00430.x. Detry MA, Ma Y. 2016. Analyzing repeated measurements using mixed models. JAMA 315:407– 408. http://dx.doi.org/10.1001/jama.2015.19394. Anderson PL, Glidden DV, Liu A, Buchbinder S, Lama JR, Guanira JV, McMahan V, Bushman LR, Casapia M, Montoya-Herrera O, Veloso VG, Mayer KH, Chariyalertsak S, Schechter M, Bekker LG, Kallas EG, Grant RM, iPrEx Study Team. 2012. Emtricitabine-tenofovir concentrations and pre-exposure prophylaxis efficacy in men who have sex with men. Sci Transl Med 4:151ra125. Yegutkin GG. 2014. Enzymes involved in metabolism of extracellular nucleotides and nucleosides: functional implications and measurement of activities. Crit Rev Biochem Mol Biol 49:473– 497. http://dx.doi.org/10 .3109/10409238.2014.953627. Vela JE, Miller MD, Rhodes GR, Ray AS. 2008. Effect of nucleoside and nucleotide reverse transcriptase inhibitors of HIV on endogenous nucleotide pools. Antivir Ther 13:789 –797. Seifert SM, Glidden DV, Meditz AL, Castillo-Mancilla JR, Gardner EM, Predhomme JA, Rower C, Klein B, Kerr BJ, Guida LA, Zheng JH, Bushman LR, Anderson PL. 2015. Dose response for starting and stopping HIV preexposure prophylaxis for men who have sex with men. Clin Infect Dis 60:804 – 810. http://dx.doi.org/10.1093/cid/ciu916. Castillo-Mancilla JR, Meditz A, Wilson C, Zheng JH, Palmer BE, Lee EJ, Gardner EM, Seifert S, Kerr B, Bushman LR, MaWhinney S, Anderson PL. 2015. Reduced immune activation during tenofovir-emtricitabine therapy in HIV-negative individuals. J Acquir Immune Defic Syndr 68: 495–501. http://dx.doi.org/10.1097/QAI.0000000000000529.
Antimicrobial Agents and Chemotherapy
September 2016 Volume 60 Number 9
Analysis of the Endogenous Deoxynucleoside Triphosphate Pool in HIV-Positive and -Negative Individuals Receiving TenofovirEmtricitabine Xinhui Chen,a Jose R. Castillo-Mancilla,b Sharon M. Seifert,a Kevin B. McAllister,c Jia-Hua Zheng,a Lane R. Bushman,a Samantha MaWhinney,d Peter L. Andersona University of Colorado, Skaggs School of Pharmacy and Pharmaceutical Sciences, Department of Pharmaceutical Sciences, Aurora, Colorado, USAa; University of Colorado, School of Medicine, Division of Infectious Diseases, Aurora, Colorado, USAb; University of Colorado, School of Medicine, Aurora, Colorado, USAc; University of Colorado, Colorado School of Public Health, Department of Biostatistics and Informatics, Aurora, Colorado, USAd
Tenofovir (TFV) disoproxil fumarate (TDF) and emtricitabine (FTC), two nucleos(t)ide analogs (NA), are coformulated as an anti-HIV combination tablet for treatment and preexposure prophylaxis (PrEP). TDF/FTC may have effects on the deoxynucleoside triphosphate (dNTP) pool due to their similar structures and similar metabolic pathways. We carried out a comprehensive clinical study to characterize the effects of TDF/FTC on the endogenous dNTP pool, from baseline to 30 days of TDF/FTC therapy, in both treatment-naive HIV-positive and HIV-negative individuals. dATP, dCTP, dGTP, and TTP were quantified in peripheral blood mononuclear cells (PBMC) with a validated liquid chromatography-tandem mass spectrometry (LC-MS/MS) methodology. Forty individuals (19 HIV-positive) were enrolled and underwent a baseline visit and then received TDF/FTC for at least 30 days. Longitudinal measurements were analyzed using mixed-model segmented linear regression analysis. The dNTPs were reduced by 14% to 37% relative to the baseline level within 3 days in both HIV-negative and HIV-positive individuals (P < 0.003). These reductions persisted to various degrees at day 30. These findings indicate that dNTP pools are influenced by TDF/ FTC therapy. This may alter cellular homeostasis and could increase the antiviral effect through a more favorable analog/dNTP ratio. Further work is needed to elucidate mechanisms, to evaluate the clinical significance of these findings, and to further probe differences between HIV-negative and HIV-positive individuals. (This study has been registered at ClinicalTrials.gov under identifier NCT01040091.)
T
he coformulated medication consisting of 300 mg of tenofovir (TFV) disoproxil fumarate (TDF) and 200 mg of emtricitabine (FTC) is marketed as an antiviral combination tablet for treatment and preexposure prophylaxis (PrEP) of HIV infection (1). TFV is a nucleotide analog, and its diphosphate anabolite (TFVDP) has a structure similar to that of dATP; FTC is a nucleoside analog, and its trisphosphate (FTC-TP) has a structure similar to that of dCTP. TFV-DP and FTC-TP compete with dATP and dCTP (natural substrates) at the active site of HIV reverse transcriptase (RT), effectively inhibiting the biosynthesis of HIV genetic material. Once incorporated, they terminate the elongation of the HIV DNA chain due to the lack of a 3= hydroxyl group to incorporate the next component (2). The ratio between drug concentration and the corresponding deoxynucleoside triphosphate (dNTP) affects the pharmacologic efficacy of TDF/FTC as a high ratio has been associated with greater antiviral activity (3, 4). It is well known that nucleos(t)ide analogs (NAs) can affect the endogenous dNTP pool, including dATP and dCTP, as well as dGTP, and TTP. NAs may compete with the host enzyme system for phosphorylation, as well as influence the complex dNTP pool metabolism pathways. For example, in vitro, TFV or its anabolites competitively inhibit creatine kinase (CK), phosphoglycerate kinase (PGK), and purine nucleoside phosphorylase (PNP) (5, 6) and upregulate 5= ecto-nucleotidases and 5= nucleotidase (NT) (7). FTC and a similar deoxy-pyrimidine nucleoside analog, lamivudine (3TC), inhibit deoxycytidine kinase (dCK) (8, 9). In addition, zidovudine (AZT), a thymidine analog, has been shown to inhibit thymidylate kinase 2 (TK2) (10, 11). In HIV-infected patients, dATP increased during TDF/abacavir (ABC) dual therapy
September 2016 Volume 60 Number 9
compared to the level with TDF monotherapy (12) and decreased during didanosine (ddI)-containing therapy relative to the level with TDF-containing therapy (13). Taking these observations together, it is important to investigate the effects of TDF/FTC therapy on the endogenous dNTP pool in clinical settings. The use of TDF/FTC for PrEP, as well as long-term treatment, requires a low toxicity profile for an acceptable risk-benefit balance (14). TDF/FTC are generally safe and well tolerated (15); however, uncommon side effects have been reported, particularly in HIV-infected patients. For example, TDF is rarely associated with mitochondrial dysfunction in renal proximal tubule cells (16, 17). Mitochondrial dysfunction can arise from an imbalance among purine and pyrimidine nucleotides (11, 18, 19). Additionally, the TDF/ddI combination is correlated with a paradoxical CD4 T cell decline (20) that may be explained by PNP inhibition and subsequent apoptosis, which is associated with excessive in-
Received 11 May 2016 Accepted 23 June 2016 Accepted manuscript posted online 27 June 2016 Citation Chen X, Castillo-Mancilla JR, Seifert SM, Mcallister KB, Zheng J-H, Bushman LR, MaWhinney S, Anderson PL. 2016. Analysis of the endogenous deoxynucleoside triphosphate pool in HIV-positive and -negative individuals receiving tenofovir-emtricitabine. Antimicrob Agents Chemother 60:5387–5392. doi:10.1128/AAC.01019-16. Address correspondence to Peter L. Anderson, [email protected]. Supplemental material for this article may be found at http://dx.doi.org/10.1128 /AAC.01019-16. Copyright © 2016, American Society for Microbiology. All Rights Reserved.
Antimicrobial Agents and Chemotherapy
aac.asm.org
5387
Chen et al.
FIG 1 Clinical study design. TDF, tenofovir disoproxil fumarate; FTC, emtricitabine; EFV, efavirenz; 1 to 60, study visits in days.
tracellular dATP and dGTP (19, 21). FTC moderately reduces cell proliferation in vitro (22), which may be associated with imbalanced dNTP pools. The characterization of the dNTP pool changes in patients receiving TDF/FTC enables the quantitation of the analog/dNTP ratio for pharmacologic efficacy and provides possible mechanisms of adverse effects. The goal of this pharmacodynamic study was to investigate the time profile of the dNTP pool, from baseline to TDF/FTC pharmacological intracellular steady state, in both HIV-positive and HIV-negative individuals. MATERIALS AND METHODS Participants and study design. The clinical protocol was approved by the institutional review broad (IRB) of the University of Colorado, and participants provided informed consent (Cell-PrEP trial; ClinicalTrials.gov identifier NCT01040091). HIV-negative adults were enrolled in an intensive clinical pharmacology study of daily coformulated TDF/FTC treatment for 30 days, followed by 30 days of washout. HIV-positive adults initiated TDF/FTC/efavirenz (EFV) treatment for 60 days (and beyond). All participants were tested for hepatitis B virus and were excluded if they were positive. Individuals were excluded if they were pregnant (or planning pregnancy), breastfeeding, had a body weight of less than 110 pounds, a modification of the diet in renal disease (MDRD) estimated glomerular filtration rate (eGFR) of less than 60 ml/min/1.73 m2, a history of pathological bone fractures, an albuminuria creatinine ratio of more than 30, or a history of kidney disease. Participants’ age, weight, sex, race, and body mass index (BMI) were recorded upon enrollment in the study. Peripheral blood mononuclear cell (PBMC) samples were taken at baseline and at 8 h postdose on days 1, 3, 7, 20, 30, and 60 in all participants. The HIV-negative group had two additional visits on days 35 and 45 during the washout period. The study design is illustrated in Fig. 1. PBMC processing. A previously described method was used for PBMC processing (23). Blood was drawn into a cell preparation tube (CPT). After the sample was mixed, the tube was spun at 1,800 ⫻ g for 30 min at room temperature to separate plasma, PBMC, and red blood cells (RBC). The buffy layer (PBMCs) between the plasma and separation medium was collected into a 15-ml centrifuge tube. After RBC lysis to eliminate potential RBC contamination, the sample was washed with an equal volume of phosphate-buffered saline (PBS). The cell sample was then spun, and the cell pellet was resuspended in 5 ml of PBS for automated cell counting (Countess cell counter; Invitrogen/Thermo Fisher Scientific Corporation, Carlsbad, CA). Finally, the cells were spun again to pellet and lysed in 500 l of 70:30 methanol-water. The lysate was stored at ⫺80°C until analysis. The dNTP pool quantitation. The analytical method utilized a strong anion exchange isolation of mono-, di-, and triphosphates (MP, DP, and TP, respectively) from the intracellular matrix (23). The TP fraction was then dephosphorylated to the parent moiety (deoxynucleoside [dN]) yielding a molar equivalent to the original deoxynucleotide intracellular concentration. A reverse phase and weak cation exchange mixed-mode purification was then used for desaltation and concentration. Endogenous dNTP levels were then quantified using a validated liquid chromatography-tandem mass spectrometry (LC-MS/MS) method with an analytical range of 50 to 2,500 fmol/sample (24).
5388
aac.asm.org
Statistical analysis. Data were natural log (ln) transformed for statistical analysis. The intracellular levels of the dNTPs (dATP, dCTP, dGTP, and TTP) were reported as geometric means and 95% confidence intervals (CIs). All statistical analyses were performed using SAS (version 9.4; SAS Institute, Inc., Cary, NC, USA). To investigate the effects of HIV infection on the dNTP pool, the baseline data were analyzed. A Welch’s t test was used to compare the dNTPs between HIV-negative and HIV-positive groups, and an F test was used to compare the difference of variances. Linear regression was performed to probe for the effects of subject variables at baseline using continuous variables such as age, weight, and BMI, while a Welch’s t test and analysis of variance (ANOVA) were performed on categorical variables such as sex and race. To evaluate the effects of TDF/FTC on the dNTP pool, longitudinal measurements from baseline to TDF/FTC pharmacological steady state in PBMCs (day 0 to 30) were used in both HIV-positive and HIV-negative groups. Time profiles of the dNTP pool were described and analyzed using mixed-model segmented linear regression (SAS Mixed procedure) (model equations are given in the supplemental material) (25). The model estimated baseline, first slope, the change point (of two slopes), and the second slope. The change point was estimated by iteration (0 to 30 days, graduated by 0.05 in days), based on model goodness of fit (Akaike information criterion [AIC]) (26). The variance structures (e.g., compound symmetry, heterogeneous compound symmetry, autoregressive model, and unknown structure) were assessed and selected also by the model goodness of fit. The differences between the change point versus baseline, day 30 versus baseline, and the change point versus day 30 were tested using SAS Estimate statements, and results were summarized as percentage change. Subject covariates were tested using a likelihood ratio test (LRT). To evaluate the change of the dNTP pool during the washout period in the HIV-negative group, day 35, 45, and 60 data were compared to day 30 data, and day 60 data were also compared to the baseline level. A paired t test was used to compare the differences at two selected time points. To investigate the long-term treatment effects of TDF/FTC in the HIV-positive group, baseline and day 60 data were compared.
RESULTS
Study population. A total of 40 participants were enrolled in the study. Among them, 21 were HIV-negative and 19 were HIVpositive patients. Thirty-four subjects completed all study visits, but six participants, two HIV-negative and four HIV-positive subjects (one female and five males), stopped early. One participant (HIV-positive) had a baseline sample but withdrew prior to any dosing visits, and the other five withdrew prior to the day 30 visit. One HIV-positive participant did not have a baseline sample but completed all visits. The study drug was generally well tolerated. The demographic characteristics of the study population are summarized in Table 1. Baseline characteristics of the dNTP pool. Data are presented in geometric means and 95% CIs in Table 2. At baseline in PBMCs (n ⫽ 39), it was observed that the HIV-negative group generally had higher dNTP pool levels than the HIV-positive group, with the exception of dATP levels. However, these differences were not statistically significant (P ⱖ 0.21). Statistically significant differences in variances were observed in all four dNTPs (P ⱕ 0.04), with higher variability in the HIV-negative group. No subject covariates (including age, weight, BMI, gender, and race) had a statistically significant effect associated with baseline dNTP levels (P ⱖ 0.17). Changes in the dNTP pool on TDF/FTC therapy. Figure 2 shows data from each individual, the model predicted population means, and the 95% confidence intervals. Model parameter statis-
Antimicrobial Agents and Chemotherapy
September 2016 Volume 60 Number 9
Effects of TDF/FTC on dNTPs
TABLE 1 Demographic characteristics of the study population Parametera
Value for the parameter
HIV status (no. of patients [%]) Negative Positive
21 (53) 19 (47)
Gender (no. of patients [%]) Female Male
13 (33) 27 (67)
Race (no. of patients [%]) White Black Hispanic
19 (47) 16 (40) 5 (13)
Median eGFR (ml/min/1.73 m2 [range]) Median age (yr [range]) Median wt (kg [range]) Median BMI (kg/m2 [range]) No. of participants (no. of withdrawals)
93.3 (66.0–130.8) 31 (20–52) 81.1 (56.5–127) 26.6 (19.9–37.7) 40 (6)
a
eGFR, estimated glomerular filtration rate; BMI, body mass index.
tics are summarized in Table 3. For all four models, a heterogeneous, compound symmetry (CSH) covariate structure best described the data. Throughout the study, no effects of subject covariates on intercepts or slopes (as interaction terms) were observed (P ⱖ 0.24). Change in dCTP level. The change point of slopes for dCTP was day 1.55 with a decrease of 20% (95% CI, ⫺28% to ⫺11%; P ⬍ 0.0001) from baseline. Compared with the baseline, dCTP was decreased by 15% (95% CI, ⫺23% to ⫺6.4%; P ⫽ 0.001) at day 30. After day 1.55, concentrations increased slightly between the change point and day 30 but was not statistically significant (5.8%; 95% CI, ⫺2.4% to 15%; P ⫽ 0.17). Change in TTP level. The change point of slopes for TTP was reached at day 1.4 with a decrease of 37% (95% CI, ⫺46% to ⫺26%; P ⬍ 0.0001) from baseline. This 37% decrease remained at day 30 (95% CI, ⫺47% to ⫺27%; P ⫽ 0.001). The change between day 1.4 and day 30 was not statistically significant (⫺1.2%; 95% CI, ⫺15% to 14%; P ⫽ 0.86). Change in dATP level. The change point of slopes for dATP was reached at 2.9 days with a decrease of 14% (95% CI, ⫺21% to ⫺4.8%; P ⫽ 0.003) from baseline. After that, although concentrations increased slightly, no statistically significant change between day 2.9 and day 30 was detected (6.5%; 95% CI, ⫺2.3% to 16%; P ⫽ 0.15). A decreasing trend was observed when day 30 was compared to the baseline value, with a reduction of 8.3% (95% CI, ⫺17% to 0.80%; P ⫽ 0.07). Change in dGTP level. The change point of slopes for dGTP
was reached at day 1.35 of therapy with a decrease of 19% (95% CI, ⫺27% to ⫺8.9%; P ⫽ 0.0004) from baseline. Compared with the baseline, there was a decrease of 11% (95% CI, ⫺20% to ⫺0.67%; P ⫽ 0.03) at day 30. A statistically significant increase was detected from the change point (day 1.35) to day 30, amounting to 9.6% (95% CI, 0.61% to 19%; P ⫽ 0.04). Change in the dNTP pool in the washout period. The HIVnegative group was assigned to a 30-day washout period after 30 days of treatment. Compared to day 30 data, which represented the pharmacological steady state of TDF/FTC in PBMCs, deoxypyrimidine triphosphates increased throughout the washout period. A statistically significant increase of 20% (95% CI, 3.8% to 40%; P ⫽ 0.02) in dCTP was observed 15 days after the initiation of the washout period. TTP increased by 83% (95% CI, 42% to 136%; P ⫽ 0.0001) at day 5, 72% (95% CI, 34% to 121%; P ⫽ 0.0004) at day 15, and 82% (95% CI, 33% to 149%; P ⫽ 0.002) at day 30 of the washout period. However, for deoxy-purine triphosphates, statistically significant changes in dATP and dGTP were not observed (P ⱖ 0.38). Day 60 data compared to the baseline level was not statistically significant in dCTP (geometric means, 934 versus 1,038 fmol/106 cells; n ⫽ 12; P ⫽ 0.48) and TTP (geometric means, 496 versus 641 fmol/106 cells; n ⫽ 11; P ⫽ 0.14). However, statistically significant differences were observed for dATP and dGTP, with geometric means of 130 versus 181 fmol/ 106 cells (n ⫽ 10; P ⫽ 0.04) and 243 versus 343 fmol/106 cells (n ⫽ 10; P ⫽ 0.04). These results suggest that dCTP and TTP returned to baseline levels around 5 to 15 days after treatment discontinuation, while dATP and dGTP remained below the baseline level until at least day 30 of the washout period. Change in the dNTP pool after 2 months of treatment. In the HIV-positive group, the day 60 results were compared to those at baseline. No statistically significant difference was observed in all four components (n ⫽ 15; P ⱖ 0.28). The geometric means (95% CI) in femtomoles per million cells were 747 (631, 886) for dCTP, 354 (266, 470) for TTP, 159 (131, 193) for dATP, and 270 (226, 323) for dGTP, levels which were comparable to baseline levels (Table 1). This suggested that the dNTP pool recovered after treatment in HIV-infected patients. DISCUSSION
In this study, we used PBMCs to investigate the effects of TDF/ FTC on the dNTP pool of HIV-positive and -negative adults. PBMCs include cells that are the major target of TDF/FTC (e.g., CD4 T cells). Also, these nucleated cells represent dNTP pool kinetics driven by enzymes such as PNP, which have high activity in PBMCs (6). At baseline, we observed generally lower dNTPs (except for dATP), but not a statistically significantly difference, in treat-
TABLE 2 Baseline characteristics of the dNTP pool Geometric mean amt (fmol/106 cells [95% CI])a
Variance of log-transformed data
dNTP
HIV-negative group
HIV-positive group
P
HIV-negative group
HIV-positive group
Pb
dCTP TTP dATP dGTP
858 (710, 1037) 436 (322, 589) 150 (120, 187) 287 (234, 352)
804 (724, 893) 380 (312, 462) 154 (141, 167) 251 (230, 274)
0.53 0.43 0.83 0.21
0.17 0.44 0.24 0.20
0.045 0.16 0.029 0.031
0.007 0.04 0.0001 0.003
a b
CI, confidence interval. An F test was used to compare variances.
September 2016 Volume 60 Number 9
Antimicrobial Agents and Chemotherapy
aac.asm.org
5389
Chen et al.
FIG 2 The dNTP changes during TDF/FTC therapy. 95% CI, model 95% confidence interval.
ment-naive HIV-infected participants than in HIV-negative participants. However, the variance of the dNTP pool was higher in the HIV-negative group than in the HIV-positive group. Although a higher T-cell activation and cell proliferation might initially lead to dNTP pool upregulation (27, 28), the pathological chronic cell activation during HIV infection (29) may induce anergy in T cells (30). It is also reported that compensatory dNTP-consuming mechanisms associated with NTPDase-1 and SAMHD-1 are upregulated, especially in uncontrolled HIV-1 infection (31, 32). These biological effects may have reduced the variability in dNTPs in HIV-infected participants. However, the small sample size (n ⫽ 39) of our study limited the ability to probe possible effects of HIV infection and subject covariates on baseline dNTP pools. Future studies require a larger sample size to further address these questions. In HIV-infected patients, Hawkins et al. reported a lower
dATP level with didanosine (ddI) than with other treatments, including TDF-including treatment (medians, 67.2 versus 186 fmol/ 106 cells; P ⫽ 0.001), but could not exclude the possibility that TDF is also related to dATP reduction (13). Goicoechea et al. reported an increasing trend in dATP and dGTP with ABC/TDF dual therapy; however, in their 7-day monotherapy period, they observed a decreasing trend in dGTP (medians, 4,026 versus 2,464 fmol/106 cells; P ⫽ 0.68) (12). The study of Goicoechea et al. reported higher dATP and dGTP levels than the study by Hawkins et al. and the present study. In a macaque model, Durand-Gasselin et al. reported no change in dATP and dGTP levels from 2 to 24 h post-oral TDF and subcutaneous TFV treatment, but this study was limited by single-dose administration, a shorter observation period (ⱕ24 h), and the lack of a baseline observation (33). Taken together, the present study represents a novel assessment of dNTP changes in PBMCs during TDF/FTC therapy.
TABLE 3 Summarization of model parameter statistics Parametera
dNTP
Estimate (95% CI)
P
Estimate (95% CI)
P
Estimate (95% CI)
P
Estimate (95% CI)
P
Change point (days)b
dCTP TTP dATP dGTP
6.73 (6.64, 6.83) 6.03 (5.86, 6.19) 5.08 (4.98, 5.17) 5.61 (5.50, 5.72)
⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001
⫺0.14 (⫺0.21, ⫺0.077) ⫺0.33 (⫺0.44, ⫺0.22) ⫺0.052 (⫺0.085, ⫺0.018) ⫺0.15 (⫺0.24, ⫺0.070)
⬍0.0001 ⬍0.0001 0.0026 0.0004
0.0020 (⫺0.00086, 0.0048) ⫺0.00044 (⫺0.0055, 0.0046) 0.0023 (⫺0.00085, 0.0055) 0.0031 (0.00018, 0.0061)
0.17 0.86 0.15 0.038
0.14 (0.077, 0.21) 0.33 (0.21, 0.44) 0.054 (0.019, 0.089) 0.16 (0.072, 0.24)
⬍0.0001 ⬍0.0001 0.0027 0.0003
1.55 1.4 2.9 1.35
Baseline
First slope
Second slope
Change in slope
a
Values are natural log transformed. 95% CI, 95% confidence interval of model estimates. Baseline values are given as ln(fmol/106 cells). Data for first slope, second slope, and change in slope are given as ln(fmol/106 cells)/day. b Estimated by iteration.
5390
aac.asm.org
Antimicrobial Agents and Chemotherapy
September 2016 Volume 60 Number 9
Effects of TDF/FTC on dNTPs
We observed that TDF/FTC therapy was associated with slight to moderate decreases in dNTP levels in both HIV-negative and HIV-positive individuals. A mixed-model segmented linear regression analysis was a useful approach to analyzing the dynamic changes in the dNTP pool during TDF/FTC treatment (34, 35). In comparison to nonlinear regression models, the statistical analysis of the second slope from the segmented linear regression model enabled us to investigate the possible change points and return of the dNTPs during therapy (a statistically significant return was detected in dGTP). This work indicates that the dNTPs decrease within a few days (⬍3 days) of TDF/FTC initiation. However, the change point was estimated based on the model goodness of fit, and these temporal approximations were limited by sampling times. The detected effects of TDF/FTC on the dNTP pool were (largest to smallest) TTP ⬎ dCTP ⬎ dGTP ⬎ dATP, indicating that TDF/FTC had larger effects on deoxy-pyrimidine triphosphate compounds (dCTP and TTP) than deoxy-purines (dGTP and dATP). If these were driven by FTC, it could due to higher intracellular concentrations of FTC triphosphate (FTC-TP) than of TFV diphosphate (TFV-DP) (36). We anticipate that lamivudine may induce similar effects, given the pharmacological similarity with emtricitabine (8, 9). Studies are needed to assess on the effects of other NA such as abacavir and zidovudine. In terms of potential mechanisms, the reduction of the dNTP pool could be explained by TFV-associated upregulation of the dephosphorylation enzymes 5=-ectonucleotidase and 5=-nucleotidase (7). TFV also suppresses the ATP-generating enzymes CK and PGK (5) and indirectly inhibits kinases that use ATP as the phosphate donor, such as nucleoside diphosphate kinase (NDPK) (37). Only in dGTP did we observe a statistically significant but minor return toward baseline values by day 30 after the change point was reached. Partial PNP inhibition by TFV phosphates might play a role in dGTP upregulation (38), perhaps counteracting some of the reduction effects. However, the decreased dATP and dGTP levels observed in this study do not support PNP inhibition as a dominant effect of TDF, which would theoretically induce an increase of dATP and dGTP. FTC effects may be partly explained by the inhibition of dCK, which had also been shown for another deoxycytidine analog, 3TC (8, 14). During the washout period in the HIV-negative group, dCTP and TTP returned to baseline levels, while dGTP and dATP stayed slightly below the baseline levels. TFV-DP exhibits a much longer intracellular half-life than FTC-TP, and previous research in this cohort observed quantifiable intracellular TFV-DP concentrations at day 30 of the washout phase, whereas FTC-TP was below the limit of quantitation (39). This is consistent with the suppression of dCTP and TTP being related to FTC-TP, while that of dATP and dGTP being related to TFV-DP. Future studies should investigate the individual effects of TDF and FTC on the dNTP pool in HIV-negative participants. For long-term treatment in the HIV-positive group, day 60 data showed less reduction in the dNTP pool level than the plateau or the change point level, which indicates that, in our participants, the endogenous dNTP pool slowly returned to baseline with longterm treatment. We still need more information to understand other factors that may affect the dNTP pool in PBMCs after longterm treatment in HIV-infected patients, such as disease progression reversal, immune system restoration, and immune cell activation changes (14, 40). In conclusion, during TDF/FTC therapy, the dNTP pool
September 2016 Volume 60 Number 9
slightly to moderately decreased within 3 days after the initiation of TDF/FTC therapy and generally remained below baseline at day 30. Potential clinical implications need to be evaluated in the future, including assessment of the analog/dNTP ratio over time, which influences the pharmacological efficacy of TDF/FTC (3, 4, 12). ACKNOWLEDGMENTS We thank Gilead Sciences, Inc., which provided the study drug. We also thank all of the volunteers who participated in the study, as well as the study personnel and Colorado Clinical Translational Research Center staff. This study was supported by NIH U01 AI84735, NIH U01 AI106499, T32 AI7447-23, and Colorado Clinical Translational Sciences Institute UL1 TR001082 and TL1 TR001081.
FUNDING INFORMATION This work, including the efforts of Xinhui Chen, José R. Castillo-Mancilla, Sharon M. Seifert, Kevin B. McAllister, Jia-Hua Zheng, Lane R. Bushman, Samantha MaWhinney, and Peter L. Anderson, was funded by HHS | NIH | NIH Office of the Director (OD) (U01 AI84735, U01 AI106499, T32 AI7447-23, UL1 TR001082, and TL1 TR001081).
REFERENCES 1. Grant RM, Lama JR, Anderson PL, McMahan V, Liu AY, Vargas L, Goicochea P, Casapia M, Guanira-Carranza JV, Ramirez-Cardich ME, Montoya-Herrera O, Fernandez T, Veloso VG, Buchbinder SP, Chariyalertsak S, Schechter M, Bekker LG, Mayer KH, Kallas EG, Amico KR, Mulligan K, Bushman LR, Hance RJ, Ganoza C, Defechereux P, Postle B, Wang F, McConnell JJ, Zheng JH, Lee J, Rooney JF, Jaffe HS, Martinez AI, Burns DN, Glidden DV, iPrEx Study Team. 2010. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med 363:2587–2599. http://dx.doi.org/10.1056/NEJMoa1011205. 2. Cihlar T, Ray AS. 2010. Nucleoside and nucleotide HIV reverse transcriptase inhibitors: 25 years after zidovudine. Antiviral Res 85:39 –58. http: //dx.doi.org/10.1016/j.antiviral.2009.09.014. 3. Margolis DM, Kewn S, Coull JJ, Ylisastigui L, Turner D, Wise H, Hossain MM, Lanier ER, Shaw LM, Back D. 2002. The addition of mycophenolate mofetil to antiretroviral therapy including abacavir is associated with depletion of intracellular deoxyguanosine triphosphate and a decrease in plasma HIV-1 RNA. J Acquir Immune Defic Syndr 31:45– 49. http://dx.doi.org/10.1097/00126334-200209010-00006. 4. Gao WY, Agbaria R, Driscoll JS, Mitsuya H. 1994. Divergent antihuman immunodeficiency virus activity and anabolic phosphorylation of 2=,3=-dideoxynucleoside analogs in resting and activated human cells. J Biol Chem 269:12633–12638. 5. Varga A, Graczer E, Chaloin L, Liliom K, Zavodszky P, Lionne C, Vas M. 2013. Selectivity of kinases on the activation of tenofovir, an anti-HIV agent. Eur J Pharm Sci 48:307–315. http://dx.doi.org/10.1016/j.ejps.2012 .11.007. 6. Ray AS, Olson L, Fridland A. 2004. Role of purine nucleoside phosphorylase in interactions between 2=,3=-dideoxyinosine and allopurinol, ganciclovir, or tenofovir. Antimicrob Agents Chemother 48:1089 –1095. http: //dx.doi.org/10.1128/AAC.48.4.1089-1095.2004. 7. Biswas N, Rodriguez-Garcia M, Crist SG, Shen Z, Bodwell JE, Fahey JV, Wira CR. 2013. Effect of tenofovir on nucleotidases and cytokines in HIV-1 target cells. PLoS One 8:e78814. http://dx.doi.org/10.1371/journal .pone.0078814. 8. Bethell R, De Muys J, Lippens J, Richard A, Hamelin B, Ren C, Collins P. 2007. In vitro interactions between apricitabine and other deoxycytidine analogues. Antimicrob Agents Chemother 51:2948 –2953. http://dx .doi.org/10.1128/AAC.01204-06. 9. Hazen R, Lanier ER. 2003. Relative anti-HIV-1 efficacy of lamivudine and emtricitabine in vitro is dependent on cell type. J Acquir Immune Defic Syndr 32:255–258. http://dx.doi.org/10.1097/00126334-200303010 -00003. 10. Morris GW, Iams TA, Slepchenko KG, McKee EE. 2009. Origin of pyrimidine deoxyribonucleotide pools in perfused rat heart: implications
Antimicrobial Agents and Chemotherapy
aac.asm.org
5391
Chen et al.
11.
12.
13.
14.
15.
16.
17. 18.
19. 20. 21.
22.
23.
24.
25.
for 3=-azido-3=-deoxythymidine-dependent cardiotoxicity. Biochem J 422:513–520. http://dx.doi.org/10.1042/BJ20082427. Rylova SN, Albertioni F, Flygh G, Eriksson S. 2005. Activity profiles of deoxynucleoside kinases and 5=-nucleotidases in cultured adipocytes and myoblastic cells: insights into mitochondrial toxicity of nucleoside analogs. Biochem Pharmacol 69:951–960. http://dx.doi.org/10.1016/j.bcp .2004.12.010. Goicoechea M, Jain S, Bi L, Kemper C, Daar ES, Diamond C, Ha B, Flaherty J, Sun S, Richman D, Louie S, Haubrich R, California Collaborative Treatment Group. 2010. Abacavir and tenofovir disoproxil fumarate co-administration results in a nonadditive antiviral effect in HIV-1infected patients. AIDS 24:707–716. http://dx.doi.org/10.1097/QAD .0b013e32833676eb. Hawkins T, Veikley W, Durand-Gasselin L, Babusis D, Reddy YS, Flaherty JF, Ray AS. 2011. Intracellular nucleotide levels during coadministration of tenofovir disoproxil fumarate and didanosine in HIV-1infected patients. Antimicrob Agents Chemother 55:1549 –1555. http://dx .doi.org/10.1128/AAC.00910-10. Anderson PL, Kiser JJ, Gardner EM, Rower JE, Meditz A, Grant RM. 2011. Pharmacological considerations for tenofovir and emtricitabine to prevent HIV infection. J Antimicrob Chemother 66:240 –250. http://dx .doi.org/10.1093/jac/dkq447. Mugwanya KK, Baeten JM. 2016. Safety of oral tenofovir disoproxil fumarate-based pre-exposure prophylaxis for HIV prevention. Expert Opin Drug Saf 15:265–273. http://dx.doi.org/10.1517/14740338.2016 .1128412. Woodward CL, Hall AM, Williams IG, Madge S, Copas A, Nair D, Edwards SG, Johnson MA, Connolly JO. 2009. Tenofovir-associated renal and bone toxicity. HIV Med 10:482– 487. http://dx.doi.org/10.1111 /j.1468-1293.2009.00716.x. Kakuda TN. 2000. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clin Ther 22:685– 708. http://dx.doi.org/10.1016/S0149-2918(00)90004-3. Desler C, Munch-Petersen B, Stevnsner T, Matsui S, Kulawiec M, Singh KK, Rasmussen LJ. 2007. Mitochondria as determinant of nucleotide pools and chromosomal stability. Mutat Res 625:112–124. http://dx.doi .org/10.1016/j.mrfmmm.2007.06.002. Oliver FJ, Collins MK, Lopez-Rivas A. 1996. dNTP pools imbalance as a signal to initiate apoptosis. Experientia 52:995–1000. http://dx.doi.org/10 .1007/BF01920108. Kakuda TN, Anderson PL, Becker SL. 2004. CD4 cell decline with didanosine and tenofovir and failure of triple nucleoside/nucleotide regimens may be related. AIDS 18:2442–2444. Bantia S, Ananth SL, Parker CD, Horn LL, Upshaw R. 2003. Mechanism of inhibition of T-acute lymphoblastic leukemia cells by PNP inhibitor— BCX-1777. Int Immunopharmacol 3:879 – 887. http://dx.doi.org/10.1016 /S1567-5769(03)00076-6. Venhoff N, Setzer B, Melkaoui K, Walker UA. 2007. Mitochondrial toxicity of tenofovir, emtricitabine and abacavir alone and in combination with additional nucleoside reverse transcriptase inhibitors. Antivir Ther 12:1075–1085. Bushman LR, Kiser JJ, Rower JE, Klein B, Zheng JH, Ray ML, Anderson PL. 2011. Determination of nucleoside analog mono-, di-, and triphosphates in cellular matrix by solid phase extraction and ultra-sensitive LC-MS/MS detection. J Pharm Biomed Anal 56:390 – 401. http://dx.doi .org/10.1016/j.jpba.2011.05.039. Chen X, McAllister KB, Zheng J-H, Castillo-Mancilla J, Klein B, Guida LA, Seifert SM, Bushman LR, Anderson PL. 2014. LC-MS/MS analysis of endogenous deoxy-nucleotide-triphosphate (dNTP) pools in HIVnegative versus HIV-positive adults, abstr PP-12. 15th Int Workshop Clin Pharmacol HIV Hepat Ther, Washington, DC, 19 to 21 May 2014. Torti C, Lapadula G, Barreiro P, Soriano V, Mandalia S, De Silvestri A, Suter F, Maggiolo F, Antinori A, Antonucci F, Maserati R, El Hamad I, Pierotti P, Sighinolfi L, Migliorino G, Ladisa N, Carosi G. 2007. CD4⫹
5392
aac.asm.org
26.
27. 28. 29. 30. 31.
32. 33.
34.
35. 36.
37.
38. 39.
40.
T cell evolution and predictors of its trend before and after tenofovir/ didanosine backbone in the presence of sustained undetectable HIV plasma viral load. J Antimicrob Chemother 59:1141–1147. http://dx.doi .org/10.1093/jac/dkm100. Greene ME, Rolfson O, Garellick G, Gordon M, Nemes S. 2015. Improved statistical analysis of pre- and post-treatment patient-reported outcome measures (PROMs): the applicability of piecewise linear regression splines. Qual Life Res 24:567–573. http://dx.doi.org/10.1007/s11136 -014-0808-3. Nicander B, Reichard P. 1985. Relations between synthesis of deoxyribonucleotides and DNA replication in 3T6 fibroblasts. J Biol Chem 260: 5376 –5381. Galmarini CM, Mackey JR, Dumontet C. 2001. Nucleoside analogues: mechanisms of drug resistance and reversal strategies. Leukemia 15:875– 890. http://dx.doi.org/10.1038/sj.leu.2402114. Silvestri G, Feinberg MB. 2003. Turnover of lymphocytes and conceptual paradigms in HIV infection. J Clin Invest 112:821– 824. http://dx.doi.org /10.1172/JCI19799. Gougeon ML. 1995. T cell apoptosis as a consequence of chronic activation of the immune system in HIV infection. Adv Exp Med Biol 374:121– 127. http://dx.doi.org/10.1007/978-1-4615-1995-9_11. Leal DB, Streher CA, Bertoncheli CDM, Carli LF, Leal CA, da Silva JE, Morsch VM, Schetinger MR. 2005. HIV infection is associated with increased NTPDase activity that correlates with CD39-positive lymphocytes. Biochim Biophys Acta 1746:129 –134. http://dx.doi.org/10.1016/j .bbamcr.2005.10.009. Ahn J. 2016. Functional organization of human SAMHD1 and mechanisms of HIV-1 restriction. Biol Chem 397:373–379. http://dx.doi.org/10 .1515/hsz-2015-0260. Durand-Gasselin L, Van Rompay KK, Vela JE, Henne IN, Lee WA, Rhodes GR, Ray AS. 2009. Nucleotide analogue prodrug tenofovir disoproxil enhances lymphoid cell loading following oral administration in monkeys. Mol Pharm 6:1145–1151. http://dx.doi.org/10.1021 /mp900036s. Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. 2002. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther 27:299 –309. http://dx.doi.org/10.1046/j.1365 -2710.2002.00430.x. Detry MA, Ma Y. 2016. Analyzing repeated measurements using mixed models. JAMA 315:407– 408. http://dx.doi.org/10.1001/jama.2015.19394. Anderson PL, Glidden DV, Liu A, Buchbinder S, Lama JR, Guanira JV, McMahan V, Bushman LR, Casapia M, Montoya-Herrera O, Veloso VG, Mayer KH, Chariyalertsak S, Schechter M, Bekker LG, Kallas EG, Grant RM, iPrEx Study Team. 2012. Emtricitabine-tenofovir concentrations and pre-exposure prophylaxis efficacy in men who have sex with men. Sci Transl Med 4:151ra125. Yegutkin GG. 2014. Enzymes involved in metabolism of extracellular nucleotides and nucleosides: functional implications and measurement of activities. Crit Rev Biochem Mol Biol 49:473– 497. http://dx.doi.org/10 .3109/10409238.2014.953627. Vela JE, Miller MD, Rhodes GR, Ray AS. 2008. Effect of nucleoside and nucleotide reverse transcriptase inhibitors of HIV on endogenous nucleotide pools. Antivir Ther 13:789 –797. Seifert SM, Glidden DV, Meditz AL, Castillo-Mancilla JR, Gardner EM, Predhomme JA, Rower C, Klein B, Kerr BJ, Guida LA, Zheng JH, Bushman LR, Anderson PL. 2015. Dose response for starting and stopping HIV preexposure prophylaxis for men who have sex with men. Clin Infect Dis 60:804 – 810. http://dx.doi.org/10.1093/cid/ciu916. Castillo-Mancilla JR, Meditz A, Wilson C, Zheng JH, Palmer BE, Lee EJ, Gardner EM, Seifert S, Kerr B, Bushman LR, MaWhinney S, Anderson PL. 2015. Reduced immune activation during tenofovir-emtricitabine therapy in HIV-negative individuals. J Acquir Immune Defic Syndr 68: 495–501. http://dx.doi.org/10.1097/QAI.0000000000000529.
Antimicrobial Agents and Chemotherapy
September 2016 Volume 60 Number 9
