Nucleoside analogue reverse transcriptase inhibitors (NRTIs) were the first drugs used in therapy for HIV infection. The development of new therapeutic compounds marked the beginning of the highly active antiretroviral therapy era in the management of HIV infection. Therapy combines typically NRTIs with either HIV protease inhibitors (PIs) or nonnucleoside reverse transcriptase inhibitors (NNRTIs). The benefits of the NRTI combination therapies in morbidity and mortality of HIV-infected patients are clear; however, adverse effects associated with the therapy have impaired the clinical management of the disease. Inhibition of DNA polymerase γ by NRTIs can cause mitochondrial dysfunction and cellular toxicity, and it seems to be the common pathway underlying the adverse effects of NRTIs on tissues. 1,2 Mitochondria are the main source of ATP by oxidative phosphorylation; therefore mitochondrial dysfunction leads to increased dependence on cytosolic glycolysis to obtain energy. This oxidative pathway results in an increased production and accumulation of lactate, which indicates mitochondrial dysfunction. NRTI-associated hyperlactatemia has been detected in HIV-infected patients. 3–5 In general, this finding represents a mild, asymptomatic, and nonprogressive hyperlactatemia. An approach for directly studying mitochondrial dysfunction is the measurement of mitochondrial membrane potential (Δψ) loss at cellular level. Depolarization of mitochondria is detected by using cationic lipophilic fluorochromes that enter in the mitochondria and are retained by the Δψ. Therefore, diminished fluorescence indicates a decreased mitochondrial potential and mitochondrial dysfunction. 6 Significant Δψ loss has been observed in peripheral blood lymphocytes (PBLs) during acute HIV syndrome 7 and chronic HIV-infected patients without antiretroviral treatment or taking zidovudine. 8,9
To our knowledge, no studies have been published on Δψ changes associated with NRTI combination therapy in peripheral lymphocytes in chronic HIV infection. The objective of this study was the detection of Δψ decreases in freshly collected peripheral blood lymphocytes from HIV-infected patients and to determine their association with blood lactate levels, clinical and virologic status, and antiretroviral therapy.
PATIENTS AND METHODS
This study was performed with the informed consent of 71 HIV-infected patients and 20 age-matched healthy controls. In the HIV-infected group, 20 patients were treatment naive and 51 were receiving antiretroviral therapy. Frequencies of zidovudine (AZT), stavudine (d4T), lamivudine (3TC), didanosine (ddI), and abacavir (ABC) combination therapies were: AZT, 3TC, PI/NNRTI (n = 10); AZT, ddI, PI (n = 2); d4T, 3TC, PI/NNRTI (n = 24); d4T, ddI, PI/NNRTI (n = 11); d4T, ddI, hydroxyurea (n = 1); d4T, ABC, PI (n = 2); and ddI, hydroxyurea (n = 1). Thirty-nine patients met clinical or laboratory parameters of AIDS: 2 of them showed CD4 T-cell count <199/μL and the remaining were included in clinical category C. Fourteen patients showed active opportunistic infection at the time of blood sampling. Patients showed HCO ≥ 20 mM and no symptoms associated with hyperlactatemia. Table 1 shows demographic, clinical, and laboratory characteristics of patients.
Mitochondrial function was studied by using the lipophilic cation 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolcarbocyanine iodide (JC-1; Molecular Probes, Eugene, OR). Samples were processed within 2 hours after venous blood collection as previously described. 7 Peripheral blood mononuclear cells (PBMCs) were obtained according to a standard method using gradient centrifugation on Ficoll (Sigma Chemical Co., St. Louis, MO). After washing twice with phosphate-buffered saline (PBS), cells were resuspended and adjusted at 0.5 × 106 cells/mL in complete medium: RPMI 1640 (Biowhittaker, Inc., Walkersville, MD) medium supplemented with 10% heat-inactivated fetal calf serum (Sigma), and 2 mM l-glutamine (Biowhittaker). PBMCs were incubated with 10 μM JC-1 in complete medium for 10 minutes at room temperature in the dark, washed twice with PBS, and resuspended in a total volume of 300 μL of PBS for acquisition with a FACScan flow cytometer and analyzed with CellQuest software (Becton Dickinson, San Jose, CA). Lymphocytes were gated on FSC/SSC plot, and at least 15,000 cells were analyzed. To set flow cytometer parameters and peripheral blood lymphocytes with mitochondrial dysfunction (PBLmd) analysis region, PBMCs were incubated with 5 μM valinomycin (Sigma), a K+ ionophore that collapsed mitochondra potential, washed with PBS, and stained with JC-1.
PBLmds were considered to be those cells included in the region previously set with valinomycin that mantained normal light-scattering characteristics of lymphocytes.
Samples for lactate were collected and processed on the same day. Serum lactate levels were assessed by an enzymatic assay method and the laboratory range is 1.2–2.7 mM.
Statistical analysis was performed with SPSS 10.0 software (Chicago, IL). Median and interquartile ranges (IQRs) described continuous variables. Qualitative variables were compared with χ2 test. Intergroup comparisons of continuous variables were performed with the Mann-Whitney U test and bivariate correlation was analyzed with the Pearson correlation coefficient. Univariate logistic regression was performed to evaluate risk factors associated with high PBLmd or high lactate levels in HIV-infected patients. Variables tested were AIDS, on antiretroviral therapy, use of AZT, d4T, 3TC, ddI, or PIs, and CD4 T-cell count <200/μL.
A stepwise multivariate logistic analysis (forward: Wald) was performed to evaluate independent predictors for high PBLmd or high lactate levels in HIV-infected patients. Explanatory variables were: on antiretroviral therapy, use of AZT, d4T, 3TC, ddI, PIs, and CD4 T-cell count <200/μL. Variables with no significance at the 0.1 level were removed from the final model.
PBLmd and blood lactate values in healthy controls, untreated and treated HIV-infected patients, use of different NRTIs, and use of PIs are shown in Table 2. Use of PIs was considered in the analyses because PIs may inhibit Δψ decrease. 10 The following comparisons in PBLmd and lactate values were performed: healthy controls versus HIV-infected patients, untreated group versus use of AZT, d4T, 3TC, ddI, or PIs, use of AZT versus use of d4T, and use of 3TC versus use of ddI.
Significantly higher PBLmd percentages were found in patients than in healthy controls. Among NRTIs used in our study population, d4T-based therapy showed higher PBLmd percentages than the untreated group and the AZT-based therapy. Significantly lower blood lactate levels were found in untreated patients than in patients treated with antiretroviral therapy or d4T-based therapy.
Correlation analysis showed significant inverse correlations between PBLmd and CD4 T-cell percentage (r = −0.28, P = 0.02) and absolute count (r = −0.30, P = 0.01), but PBLmd did not correlate with age, duration of HIV infection, duration of current antiretroviral therapy, duration of d4T therapy, CD8 T-cell number, lactate levels, or plasma HIV RNA. No difference was found in PBLmd percentage between patients with undetectable (<1.7 log10 copies/mL; n = 30) and detectable (n = 41) viral load (1.3 [0.8–3.7] and 2.7 [1.0–3.7], respectively; P = 0.10).
To study whether advanced disease was associated with PBLmd, the PBLmds were compared in HIV-positive patients with (n = 39) and without (n = 32) AIDS. Patients with AIDS showed a higher percentage (2.7, IQR = 1.1–4.4) than HIV-positive patients without AIDS (1.4, IQR = 0.6–3.0; P = 0.012). In addition, patients with AIDS showed lower CD4 T-cell counts (157, IQR = 69–384) than those without (480, IQR = 275–745; P < 0.001). No differences in PBLmd between AIDS patients with and without active opportunistic diseases were found (P = 0.34).
High PBLmd percentages were defined as those above the healthy control mean +2 SD (2.8%). Demographic and laboratory data in patients with high and low PBLmds are shown in Table 3. HIV-infected patients with high PBLmds had significantly lower CD4 T-cell counts. Comparison of duration of d4T therapy between patients with high and low PBLmds showed no significant difference. Risk factors associated with high PBLmd were evaluated through univariate logistic regression (Table 4). Variables tested were AIDS, on antiretroviral therapy, use of AZT, d4T, 3TC, ddI, or PI, and CD4 T-cell count < 200/μL. Analysis showed that use of d4T (OR = 4.41, P = 0.006) and CD4 T-cell count < 200/μL (OR = 3.20, P = 0.026) were associated with high PBLmd percentage. In multivariate analysis, use of d4T (OR = 5.86, P = 0.003) and CD4 T-cell count < 200/μL (OR = 4.51, P = 0.012) were independent predictors of high PBLmd percentage (Table 5). No independent effect was observed in the use of ddI or 3TC after adjusting for the use of d4T.
Similar analysis was performed for high blood lactate levels defined as those above the healthy control mean +2 SD, that is, 2.7 mM. All untreated patients showed lactate levels <2.7 mM. No significant risk factor was found by univariate analysis (Table 6), and no model was determined in the multivariate logistic regression analysis for high blood lactate levels, suggesting that high lactate is independent of the type of NRTI used in our study population.
Mitochondrial toxicity associated with NRTI use represents a growing important issue in the clinical management of HIV-infected patients in the recent years. The need to identify and monitor mitochondrial toxicity via laboratory markers has led to the development of assays for measurement of mitochondrial DNA depletion. 11–14 Another laboratory approach to identify NRTI-associated mitochondrial toxicity is based on the measurement of decreased Δψ as a marker of mitochondrial dysfuntion. Previous reports have reported significant Δψ decrease in lymphocytes from acutely HIV-infected patients 7 and from untreated and AZT-treated patients. 8 A recent report described an in vitro assay that predicts NRTI mitochondrial toxicity by detecting changes in Δψ using flow cytometry. 15 The authors showed Δψ changes using JC-1 and flow cytometry in pancreatic and hepatic human cells induced by hydroxyurea and ddI. Our data indicate that this method can be applied to the study of ex vivo mitochondrial dysfunction in HIV infection. We have observed that PBLmd percentage is higher in HIV-infected patients than in controls and is higher in AIDS than in non-AIDS patients and is inversely correlated with CD4 T-cell number. However, we have found considerably lower PBLmd percentages than those previously reported, 9 probably due to differences in sensitivity and specificity of the probes used for the detection of Δψ changes. 16,17
Taking into account that CD4 T-cell count <200/μL is an independent risk factor for mitochondrial dysfunction and that patients with AIDS showed significantly lower CD4 T-cell counts, this finding suggests that mitochondrial alterations may be involved in the immune suppression in advanced HIV infection. Although we did not find any association between viral load and PBLmd, HIV-1 viral load could play a role in the Δψ decrease via increased cytokine production or binding of viral proteins to their cellular receptors. 18–20
Patients receiving antiretroviral therapy showed significant higher PBLmd percentages than untreated patients. Among the antiretroviral drugs, the use of d4T is independently associated with high PBLmd percentages. Differences between NRTIs could be explained by their different potencies for inducing mitochondrial dysfunction 21,22 and their ability to alter mitochondrial components other than polymerase γ. 23,24
NRTI-induced mitochondrial toxicity is thought to cause the treatment-associated hyperlactatemia in HIV-infected patients. Within the spectrum of clinical manifestation appreciated in the NRTI-induced hyperlactatemia, 3 our treated patients fell into mild, asymptomatic hyperlactatemia range. The finding of higher blood lactate levels in the use of d4T compared with untreated patients is in agreement with previously published reports. 25,26 Despite the relationship between mitochondrial dysfunction and increased lactate production, we did not find any association between PBLmd percentages and blood lactate. This could be because blood lactate reflects the balance between the systemic production and clearance of lactate, while our data on mitochondrial dysfunction are peripheral lymphocyte specific. It is important to note that our patients showed compensated hyperlactatemia, and no data can be shown on PBLmd and lactic acidosis. On the other hand, it remains to be determined whether successive detection of high PBLmd percentage in longitudinal studies could be associated with a greater risk for symptomatic hyperlactatemia.
To sum up, the study of mitochondrial dysfunction by flow cytometry can be a useful method to detect NRTI-induced toxicity in HIV-infected patients, and other studies are required to assess its clinical relevance.
1. Brinkman K, Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clin Ther. 2000; 22:685–708.
2. White AJ. Mitochondrial toxicity and HIV therapy. Sex Transm Infect. 2001; 77:158–173.
3. John M, Mallal S. Hyperlactatemia syndromes in people with HIV infection
. Curr Opin Infect Dis. 2002; 15:23–29.
4. John M, Moore CB, James IR, et al. Chronic hyperlactatemia in HIV-infected patients taking antiretroviral therapy
. 2001; 15:717–723.
5. Lonergan JT, Behling C, Poza G, et al. Hyperlactatemia and hepatic abnormalities in 10 HIV-infected patients receiving nucleoside analogue combination regimens. Clin Infect Dis. 2000; 31:162–166.
6. Poot M, Pierce RH. Analysis of mitochondria by flow cytometry. Methods Cell Biol. 2001; 64:117–128.
7. Cossarizza A, Mussini C, Mongiardo N, et al. Mitochondria alterations and dramatic tendency to undergo apoptosis in peripheral blood lymphocytes
during acute HIV syndrome. AIDS
. 1997; 11:19–26.
8. Macho A, Castedo M, Marchetti P, et al. Mitochondrial dysfunctions in circulating T lymphocytes from human immunodeficiency virus-1 carriers. Blood. 1995; 86:2481–2487.
9. Moretti S. Marcellini S, Boschini A, et al. Apoptosis and apoptosis-associated perturbations of peripheral blood lymphocytes
during HIV infection
: comparison between AIDS
patients and asymptomatic long-term non-progressors. Clin Exp Immunol. 2000; 122:364–373.
10. Phenix BN, Lum JJ, Nie Z, et al. Antiapoptotic mechanism of HIV protease inhibitors: preventing mitochondrial transmembrane potential loss. Blood. 2001; 98:1078–1085.
11. Walker UA, Bickel M, Lutke Volksbeck SI, et al. Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. J Acquir Immune Defic Syndr. 2002; 29:117–121.
12. Shikuma CM, Hu N, Milne C, et al. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS
. 2001; 15:1801–1809.
13. Dalakas MC, Semino-Mora C, Leon-Monzon M. Mitochondrial alterations with mitochondrial DNA depletion in the nerves of AIDS
patients with peripheral neuropathy induced by 2′3′-dideoxycytidine (ddC). Lab Invest. 2001; 81:1537–1544.
14. Cote HC, Brumme ZL, Craib KJ, et al. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. N Engl J Med. 2002; 346:811–820.
15. Foli A, Benvenuto F, Piccinini G, et al. Direct analysis of mitochondrial toxicity of antiretroviral drugs. AIDS
. 2001; 15:1687–1694.
16. Salvioli S, Ardizzoni A, Franceschi C, et al. JC-1, but not DiOC6
(3) or rhodamine 123, is a reliable fluorescent probe to assess Δψ in intact cells: implications for studies on mitochondrial functionality during apoptosis. FEBS Lett. 1997; 411:77–82.
17. Sabnis RW, Deligeorgiev TG, Jachak MN, et al. DiOC6(3): a useful dye for staining the endoplasmic reticulum. Biotech Histochem. 1997; 72:253–258.
18. Petit F, Arnoult D, Lelievre JD, et al. Productive HIV-1 infection of primary CD4+ T cells induces mitochondrial membrane permeabilization leading to a caspase-independent cell death. J Biol Chem. 2002; 277:1477–1487.
19. Roggero R, Robert-Hebmann V, Harrington S, et al. Binding of human immunodeficiency virus type 1 gp120 to CXCR4 induces mitochondrial transmembrane depolarization and cytochrome c-mediated apoptosis independently of Fas signaling. J Virol. 2001; 75:7637–7650.
20. Roumier T, Vieira HL, Castedo M, et al. The C-terminal moiety of HIV-1 Vpr induces cell death via a caspase-independent mitochondrial pathway. Cell Death Differ. 2002; 9:1212–1219.
21. Brinkman K, ter Hofstede HJM, Burguer DM, et al. Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity as common pathway. AIDS
. 1998; 12:1735–1744.
22. Johnson AA, Ray AS, Hanes J, et al. Toxicity of antiviral nucleoside analogs and the human mitochondrial DNA polymerase. J Biol Chem. 2001; 276:40847–40857.
23. Barile M, Valenti D, Quagliarello E, et al. Mitochondria as cell targets of AZT (zidovudine). Gen Pharmacol. 1998; 31:531–538.
24. Masini A, Scotti C, Caligaro A, et al. Zidovudine-induced experimental myopathy: dual mechanism of mitochondrial damage. J Neurol Sci. 1999; 166:131–140.
25. Boubaker K, Flepp M, Sudre P, et al. Hyperlactatemia and antiretroviral therapy
: the Swiss HIV Cohort Study. Clin Infect Dis. 2001; 33:1931–1937.
26. John M, Moore CB, James IR, et al. Chronic hyperlactatemia in HIV-infected patients taking antiretroviral therapy
. 2001; 15:717–723.