IMMUNOLOGY

Short Communication: Oxidative Stressin HIV-Infected Individuals: A Cross-Sectional Study

Ajay Wanchu, S.V. Rana, Suresh Pallikkuth, and Ravinder Kaur Sachdeva

Abstract

HIV infection increases the oxidative stress process, and antiretroviral combination therapy increases proteinoxidation and preexistent oxidative stress. The latter induces production of reactive oxygen species. Lipidperoxidation (LPO) is a means of determining oxidative stress. There is also a deficiency of glutathione in HIVinfection. Persistent oxidative load leads to an accelerated rate of consumption of glutathione (GSH). This studymeasured LPO and GSH levels in plasma of HIV-infected individuals with or without therapy and comparedthese with healthy controls. One hundred HIV-infected individuals and 30 healthy controls were included in thestudy. LPO and GSH levels were measured in plasma according to previously described methods. The meanlevel of LPO in HIV-infected individuals was 0.7� 0.1 mmol=ml (range, 0.5–0.9 mmol=ml), whereas the mean LPOlevel in controls was 0.3� 0.1 mmol=ml (range, 0.2–0.4 mmol=ml). The mean LPO levels were significantly higherin HIV-infected individuals as compared to healthy controls ( p value<0.0001). The mean GSH level in HIV-infected individuals was 0.06� 0.01 mmol=ml (range, 0.03–0.08). The mean GSH level in healthy controls was0.09� 0.01 mmol=ml (range, 0.05–0.1). The mean glutathione level in HIV-infected individuals was significantlylower in compared to healthy controls ( p value< 0.0001). There was a significant positive correlation betweenabsolute CD4 cells and GSH levels (r¼ 0.182, p¼ 0.045). There is increased oxidative stress in HIV-infectedpatients. Whether supplementation with antioxidants will reduce this oxidative stress is still unknown.

Introduction

HIV infection induces a wide array of immunolog-ical alterations resulting in the progressive develop-

ment of opportunistic infections and=or malignancies, whichresults in AIDS. Of the mechanisms contributing to this pro-gression, oxidative stress induced by the production of re-active oxygen species (ROS) may play a critical role in thestimulation of HIV replication and the development of im-munodeficiency.1,2

The infection is characterized by functional and structuralchanges related to the immunological system. Increasedproduction of ROS such as superoxide anion, hydroxyl radi-cal, and hydrogen peroxide may be related to an increasedactivation of polymorphonuclear leukocytes during infectionsor influenced by the prooxidant effect of proinflammatorycytokines produced by activated macrophages during thecourse of HIV infection.3

ROS can attack double bonds in polyunsaturated fatty ac-ids, inducing lipid peroxidation (LPO),4 which may result inmore oxidative cellular damage.5,6 Thus, measurement ofLPO is a means of determining oxidative stress. Such damage

may be prevented or moderated by a normal antioxidantdefense system that scavenges the ROS. Glutathione (g-glutamylcysteinylglycine, GSH), a tripeptide present in highconcentrations in all mammalian cells, has many criticalprotective and metabolic functions. It detoxifies electrophilicmetabolites of xenobiotics and protects cells from the toxiceffects of free radicals and reactive oxygen compounds.7 It isalso important in the immune response to infections and playsan important role in lymphocyte proliferation, antibody-dependent and cell-mediated cytotoxicity, and protectionof lymphocytes against superoxides that are produced todestroy invading pathogens.8,9

There have been numerous reports of GSH deficiency inHIV infection. The concentration of GSH is lower in plasma,lung epithelial lining fluid, and peripheral blood mononu-clear cells (PBMCs) of HIV-infected individuals.10–12 More-over, in vitro studies have shown that low GSH levels impair Tcell function12 and also promote HIV expression,13 suggestinga link between GSH deficiency and progression of HIV dis-ease. This was confirmed by a recent report of poor sur-vival rates of HIV-infected individuals with lower GSHlevels and improved survival when GSH was replenished.14

Department of Internal Medicine and Gastroenterology, PGIMER, Chandigarh, India.

AIDS RESEARCH AND HUMAN RETROVIRUSESVolume 25, Number 12, 2009ª Mary Ann Liebert, Inc.DOI: 10.1089=aid.2009.0062

1307

Alternatively, because GSH levels fall under conditions ofincreased oxidative stress, such as HIV infection, it can beproposed that a persistent oxidative load leads to an acceler-ated rate of consumption of GSH that is not matched by anequal increase in the rate of synthesis of the tripeptide.

The purpose of the present study was to measure theplasma levels of LPO and GSH in an HIV-positive population,either asymptomatic or with AIDS, and to compare these re-sults with those in age-matched apparently healthy controlsubjects.

Materials and Methods

This was a cross-sectional study. One hundred HIV-seropositive individuals were selected from the Immuno-deficiency Clinic of the Internal Medicine Department of thePost Graduate Institute of Medical Education & Research(PGIMER), Chandigarh, from January 2007 to June 2007.Thirty aged- and sex-matched apparently healthy seronega-tive individuals were included as healthy controls. Informedconsent was obtained from all individuals.

A complete medical history was taken and a physical ex-amination was carried out to screen patients for opportunisticinfections and coinfections. HIV-seropositive individualswere classified as per WHO staging. Two nucleoside reversetransciptase inhibitors and one nonnucleoside reverse tran-scriptase inhibitor were started as per WHO guidelines.15

Assay procedure

LPO and GSH analyses were performed as per the methodsdescribed by Ohkawa et al. and Beutler et al., respectively.16,17

Five milliliters of blood was drawn into pyrogen-free bloodcollection EDTA tubes. The sample was centrifuged andplasma was stored in 1-ml aliquots at �208C.

Measurement of LPO

For LPO, 100ml of plasma was incubated with 100ml ofdistilled water and 50ml sodium dodecyl sulfate (SDS) atroom temperature for 10 min. Following incubation, 375 ml of20% acetic acid and 375 ml of thiobarbituric acid (TBA) wereadded and the reaction mixture was boiled for 1 h, cooled, andthen extracted with 1 ml of an n-butanol-pyridine mixture toavoid turbidity. The upper layer of each sample was takenand absorbance at 532 nm was measured.16

Measurement of GSH

In brief, 0.5 ml of plasma was mixed with 1.5 ml of distilledwater and 3 ml of precipitating reagent in a test tube. Themixture was allowed to stand for 5 min and filtered. Onemilliliter of filtrate was taken in another test tube and 4 ml ofphosphate solution and 0.5 ml of 5,50-dithiobis-2-nitrobenzoicacid (DTNB) reagent were added. Optical density was mea-sured within 5 min at 412 nm against a test blank (containingdistilled water). GSH in a concentration of 20–140 mg was usedas a standard and the levels of GSH were expressed as mmol=gHb.17

Statistical analysis

LPO and GSH levels in HIV-infected individuals andhealthy controls were compared by unpaired t-test using

Graph pad prism software. Descriptive data were expressedas mean� SD for nonparametric distribution. Spearman rankcorrelation was used to correlate CD4 cell count with LPO andGSH levels. A p-value of less than 0.05 was considered sta-tistically significant.

Results

Sixty-nine HIV-seropositive males and 31 HIV-seropositivefemales were selected from the Immunodeficiency Clinic ofInternal Medicine department at PGIMER. The mean age ofthe HIV-seropositive individuals was 34.1� 7.3 years (range,22–53; median, 32 years). The mean hemoglobin of HIV-seropositive individuals was 11.8� 1.8 g=dl (range, 6.3–15.8 g=dl). The mean absolute CD4 was 207.1� 151.6 cells=ml(range, 16–827 cells=ml).

Opportunistic infections in the form of pulmonary tuber-culosis were seen in 12 patients: one had lymph nodetuberculosis and one had pulmonary tuberculosis with toxo-plasmosis, six patients presented with pneumocystis pneu-monia, five had oral candidiasis, three patients had herpeszoster, one had cryptosporidiosis, one presented with cryp-tococcosis, one had chronic diarrhea, and one patient pre-sented with progressive multifocal leukoencephalopathy.Thirty-four patients did not receive any medication. Fiftypatients received a nevirapine-based combination regimen[stavudine (d4T)=zidovudine (AZT), lamivudine (3TC), ne-virapine (NVP)], 15 patients received an efavirenz-basedcombination regimen (AZT, 3TC, EFV), and 1 received anelfinavir-based combination therapy (AZT, 3TC, NFV). Themean duration of treatment was 9.2� 7.0 months.

Figure 1a shows the scatter plot of LPO levels measured inplasma of 100 HIV-infected patients and 30 healthy controls.The mean level of LPO in HIV-infected individuals was0.7� 0.1 mmol=ml (range, 0.5–0.9 mmol=ml), whereas themean LPO level in controls was 0.3� 0.1 mmol=ml (range, 0.2–0.4 mmol=ml). The mean LPO levels were significantly higherin HIV-infected individuals as compared to healthy controls( p value< 0.0001) (Fig. 1a). Figure 1b shows the scatter plot ofGSH levels measured in plasma of HIV-infected patients andhealthy controls. The mean GSH level in HIV-infected indi-viduals was 0.06� 0.01 mmol=ml (range, 0.03–0.08). The meanGSH level in healthy controls was 0.09� 0.01 mmol=ml (range,0.05–0.1). The mean glutathione level was significantly lowerin HIV-infected individuals as compared to healthy controls( p value< 0.0001).

Figure 2 shows scatter plots depicting LPO and GSH levels,respectively, in cohorts that were not on antiretroviral therapy(ART) compared with those on various ART regimens. Therewas no statistically significant difference in the mean levels ofLPO ( p value 0.86) and mean levels of GSH ( p value 0.51) inthe HIV-infected group not on therapy versus the individualson various ART regimens taken together as one cohort. Also,no statistical significant difference was noted between meanLPO levels ( p value 0.68) and mean GSH levels ( p value 0.53)in HIV-infected individuals with and without tuberculosis.There was significant positive correlation between absoluteCD4 cells and GSH level (r¼ 0.182, p¼ 0.045); however, nosuch correlation was evident between CD4 cells and LPOlevels (r¼ 0.011, p¼ 0.912). The duration of therapy did notshow any correlation either with GSH levels (r¼ 0.01, p¼ 0.9)or with LPO levels (r¼ 0.08, p¼ 0.5). There was, however, an

1308 WANCHU ET AL.

inverse correlation noted between LPO levels and GSH levels(r¼ –0.5, p< 0.01).

Discussion

Oxidative stress is the common mediator of apoptotic(programmed) cell death in HIV=AIDS. Antiretroviral com-bination therapy increases protein oxidation as well as thelevel of oxidative stress already present in HIV infection.18

The use of highly active antiretroviral therapy (HAART) hasimproved the clinical evolution of these patients. However,some patients still show higher oxidative stress and other ef-fects of HAART, such as change in lipid and muscle metab-

olism.19 HIV-infected people tend to have subnormal GSHlevels in plasma, lung epithelial lining fluid, and PBMCs.10–12

Multiple mechanisms may contribute to systemic GSH defi-ciency in HIV disease, including excessive production of in-flammatory cytokines and use of GSH-depleting drugs suchas acetaminophen.20 Release of HIV-TAT (trans-acting tran-scriptional activator) blocks transcription of manganesesuperoxide dismutase, an enzyme that helps to prevent oxi-dative stress, and markedly decreases the activity of glucose-6-phosphate dehydrogenase, a key enzyme in pathways thatmaintain GSH in its reduced state.21

In this study, standardized methods by Ohkawa et al. andBeutler et al. were used to study LPO and GSH levels.16,17 The

FIG. 1. (a) Scatter plot showing LPO levels in plasma of 100 HIV-infected individuals and 30 healthy controls. Mean LPOlevels in plasma was significantly higher in HIV-infected individuals as compared to healthy controls ( p value< 0.0001).(b) Scatter plot showing GSH levels in plasma of 100 HIV-infected individuals and 30 healthy controls. Mean glutathionelevels in HIV-infected individuals was significantly lower in HIV-infected individuals as compared to healthy controls( p value< 0.0001)

FIG. 2. (a) Scatter plot showing LPO levels in plasma of HIV-infected individuals without ART, receivingAZT=d4T=3TC=NVP, receiving d4T=3TC=EFV, and receiving AZT=3TC=NFV. No statistically significant difference wasfound in the mean levels of LPO in HIV-infected individuals with or without therapy ( p value 0.86). (b) Scatter plot showingGSH levels in plasma of HIV-infected individuals without ART, receiving AZT=d4T=3TC=NVP, receiving d4T=3TC=EFV,and receiving AZT=3TC=NFV. There was no statistically significant difference in the mean levels of GSH in HIV-infectedindividuals with or without therapy ( p value 0.51).

OXIDATIVE STRESS IN HIV INFECTION 1309

increased LPO levels and reduction in GSH levels in HIV-infected individuals as compared to healthy controls areconsistent with those of many studies published previously.The results showed that in HIV-infected individuals oxidativestress was significantly higher than in noninfected individu-als. However, there was no significant difference found in thelevels of LPO and GSH between HIV-infected individualsreceiving therapy and those without therapy. This is consis-tent with the result of previous study from Southern India.21

A study by Aukrust et al. had also shown that duringHAART, increases in CD4 cell count were accompanied byimprovement in GSH-redox status and an increase in thesubnormal levels of antioxidant vitamins, but HAART did notinduce full normalization of these parameters.19 Jahoor et al.had shown that GSH deficiency of HIV-infected individuals isdue in part to a reduced synthesis rate secondary to a shortagein cysteine availability. There have been numerous reports ofincreased LPO and GSH deficiency in HIV infection.22 Allardet al. had also shown lower concentrations of antioxidantcompounds and higher contents of the products of oxidativereactions in HIV-infected individuals than in noninfectedindividuals.23 Also, in the present study, no statistically sig-nificant difference was noted between LPO levels in HIV-infected individuals with and without tuberculosis ( p value0.68). This might be because all individuals with tuberculosisreceived antitubercular therapy and none of the individualenrolled had active tuberculosis.

Oxidative stress may decline with treatment in individualswith dual infection. The mechanism underlying the increasedoxidative stress in the HIV population remains unclear.In addition to an excessive production of reactive oxygenspecies, which may be explained by polymorphonuclearleukocyte activation during infectious conditions or by aprooxidant effect of tumor necrosis factor-a produced by ac-tivated macrophages, a weakened antioxidant defense systemmay play a role.22 Because it is known that GSH peroxidaseplays a central role in the metabolism of reactive oxygenspecies, antioxidant supplements may have an effect on oxi-dative stress in HIV-infected individuals.24,25

It is possible that maintenance of antioxidant defenses willbe more important as survival is prolonged indefinitely, be-cause the accumulation of tissue damage due to oxidativestress may take a long time to manifest. It is also possible thatantioxidant therapy would be more beneficial in acute ratherthan chronic conditions because oxidative stress is muchgreater during acute processes, such as opportunistic infec-tions, than during periods of clinical stability.

Thus, the potential for therapeutic benefit from antioxidanttherapies remains questionable. The relationship betweenHIV viral content and oxidative stress as well as the effect ofcurrent antiretroviral therapies on oxidative stress need to bedetermined.

Disclosure Statement

No competing financial interests exist.

References

1. Wong GHW, McHugh T, Weber R, and Goeddel DV: Tumornecrosis factor alpha selectively sensitizes human immuno-deficiency virus infected cells to heat and radiation. ProcNatl Acad Sci USA 1991;88:4372–4376.

2. Schreck R, Rieber P, and Baeuerle PA: Reactive oxygen in-termediates as apparently widely used messengers in theactivation of the NF–kB transcription factor and HIV–1.EMBO J 1991;10:2247–2258.

3. Das UN, Podma M, Sogar PS, Ramesh G, and Koratkar R:Stimulation of free radical generation in human leucocytesby various agents including tumor necrosis factor is acalmodulin-dependent process. Biochem Biophys Res Com-mun 1990;167:1030–1036.

4. McCord JM: The superoxide free radical: Its biochemistryand pathophysiology. Surgery 1983;94:412–414.

5. Freeman BA and Crapo JD: Biology of disease. Free radicalsand tissue injury. Lab Invest 1983;47:412–426.

6. Halliwell B: Oxidants and human disease: Some new con-cepts. FASEB J 1987;1:358–364.

7. Beutler E: Nutritional and metabolic aspects of glutathione.Annu Rev Nutr 1989;9:287–302.

8. Droge W, Eck HP, Gmunder H, and Mihm S: Modulation oflymphocyte functions and immune responses by cysteineand cysteine derivatives. Am J Med Suppl 1991;3C:140S–144S.

9. Smyth MJ: Glutathione modulates activation-dependentproliferation of human peripheral blood lymphocyte popu-lations without regulating their activated function. J Im-munol 1991;146:1921–1927.

10. Buhl R, Holroyd KJ, Mastrangeli A, et al.: Systemic gluta-thione deficiency in symptom-free HIV-seropositive indi-viduals. Lancet 1989;2:1294–1298.

11. De Quay B, Malinverni R, and Lauterberg BH: Glutathionedepletion in HIV-infected patients: Role of cysteine defi-ciency and effect of oral N-acetylcysteine. AIDS 1992;6:815–819.

12. Roederer M F, Staal JT, Anderson M, et al.: Dysregulation ofleukocyte glutathione in AIDS. Ann NY Acad Sci 1993;677:113–125.

13. Kalebic T, Kinter A, Pol G, Anderson ME, Meister A, andFauci AS: Suppression of human immunodeficiency virusexpression in chronically infected monocytic cells by gluta-thione, glutathione ester, and N-acetylcysteine. Proc NatlAcad Sci USA 1991;88:986–990.

14. Herzenberg LA, De Rosa SC, Dubs JG, et al.: Glutathionedeficiency is associated with impaired survival in HIV dis-ease. Proc Natl Acad Sci USA 1997;94:1967–1972.

15. World Health Organization: Scaling up antiretroviral ther-apy in resource-limited settings. Available at www.who.int=docstore=hiv=scaling=summ.html (accessed on 5 March2009).

16. Ohkawa H, Oshishi N and Yagi H: Reaction of linolenic acidhydroperoxide with TBA: J Lipid Res 1978;19:1053–1057.

17. Beutler E, Ducon O, and Kelly BM: Improved method for thedetermination of blood glutathione. J Lab Clin Med 1963;61:882–890.

18. Ngondi JL, Oben J, Etame LH, Forkah DM, and Mbanya D:The effect of different combination therapies on oxidativestress markers in HIV infected patients in Cameroon. AIDSRes Ther 2006;3:19.

19. Aukrust P, Muller F, Svardal A, Ueland T, Berge RK, andFroland SS: Disturbed glutathione metabolism and de-creased antioxidants levels in human immunodeficiencyvirus-infected patients during highly active antiretroviraltherapy––Potential immunomodulatory effects of antioxi-dants. J Infect Dis 2003;188:232–238.

20. Staal FJT, Ela SW, Roederer R, Anderson MT, HerzenbergLA, and Herzenberg LA: Glutathione deficiency and human

1310 WANCHU ET AL.

immunodeficiency virus infection. Lancet 1992;339;909–912.

21. Sundaram M, Saghayam S, Priya B, et al.: Changes in anti-oxidant profile among HIV-infected individuals on generichighly active antiretroviral therapy in Southern India. Int JInfect Dis 2008;12(6):e61–e66. Epub July 14, 2008.

22. Jahoor F, Jackson A, Gazzard B, et al.: Erythrocyte glutathi-one deficiency in symptom-free HIV infection is associatedwith decreased synthesis rate. Am J Physiol EndocrinolMetab 1999;276:205–211.

23. Allard JP, Aghdassi E, Chau J, Salit I, and Walmsley S:Oxidative stress and plasma antioxidant micronutrients inhumans with HIV infection. Am J Clin Nutr 1998;67:143–147.

24. Buhl R, Jaffe A, Holyroyd KJ, et al.: Systemic glutathionedeficiency in symptom-free HIV sero-positive patients.Lancet 1989;2:1294–1296.

25. Bogden JD, Baker H, Frank O, et al.: Micronutrient status andhuman immunodeficiency virus (HIV) infection. Ann NYAcad Sci 1990;587:189–195.

Address correspondence to:Ajay Wanchu

Department of Internal MedicinePGIMER

Chandigarh 160012, India

E-mail: awanchu@yahoo.com

OXIDATIVE STRESS IN HIV INFECTION 1311