Mitochondrial alterations have been widely associated with HIV infection1 and with antiretroviral (ARV) drugs2 in adulthood, even in asymptomatic patients.3 Although nucleoside reverse transcription inhibitors (NRTI) inhibit DNA polymerase gamma leading to mitochondrial dysfunction,4 non-nucleoside reverse transcription inhibitors and protease inhibitors (PIs) have also been described to cause mitochondrial apoptosis,2 although to a lesser extent. The combination of ARV drugs, known as highly active antiretroviral therapy (HAART), leads to a variety of important adverse secondary effects, including body fat abnormalities (BFA). Changes in body fat content and distribution prevalence rates in HIV-infected children range from 18% to 33%.5 BFA etiology and pathophysiology have not been completely elucidated. Both HIV and host-dependent factors are thought to be involved in this through premature aging, immune activation, and adipose tissue inflammation.6 Although some studies have not demonstrated the evidence of ARV class-specific effects,7 other studies support that BFA is an adverse event linked to ARV, specially NRTI and PI, which have been related to lipoatrophy and lipohypertrophy, respectively. All drug classes would potentially alter adipogenesis and adipocyte differentiation, but the main mechanism proposed to contribute to BFA development is NRTI capacity to inhibit DNA polymerase γ, first suggested by Brinkman in 1999.8 PIs can alter lipid and carbohydrate metabolism and have also been related to lipoatrophy through their proapoptotic and lipolytic effects in adipocytes.9
To date, very few studies have focused on mitochondrial function in HIV-infected children. Saitoh et al10 observed a mitochondrial DNA (mtDNA) depletion in the peripheral blood mononuclear cells (PBMC) of children receiving didanosine. Rosso et al11 did not identify mitochondrial benefits after switching from stavudine to tenofovir, a less toxic drug. Only one study has assessed possible associations between mitochondrial toxicity and BFA in children.12 Mitochondrial function and apoptosis in PBMC were studied in a group of HIV-infected children on HAART with and without BFA, as compared with a control group of uninfected children, and no differences were observed. Because a group of untreated HIV-infected children was lacking, an independent effect of HIV on mitochondria could not be assessed.
We hypothesized that HIV-infected children may show an HIV- and HAART-related mitochondrial dysfunction that may be more important in patients developing BFA. This impairment could be present at different molecular levels (genetic, translational, biochemical, or functional) and, therefore, the character of the lesion may be diffuse rather than localized.
MATERIALS AND METHODS
We designed a cross-sectional study in 69 HIV-infected children followed up in a tertiary care pediatric center in Barcelona, Spain and 24 uninfected healthy controls. HIV-infected patients were classified into 4 groups: 16 without BFA and off HAART (BFA–HAART–), 26 without BFA and on HAART (BFA–HAART+), 6 untreated patients presenting BFA (BFA+HAART–), and 21 patients with BFA on HAART (BFA+HAART+). Simultaneously, we also classified HIV-infected patients according to HIV viral load (CA HIV Monitor; Roche, Basel, Switzerland; limit <50 copies/mL), either undetectable (n = 39) or not (n = 30), to assess the independent effect of HIV on mitochondrial function. BFA was assessed by physical examination at the time of assessment and was defined as the presence or not of lipoatrophy (peripheral fat wasting in limbs, face, or buttocks), lipohypertrophy (central fat accumulation on neck, breasts, or abdomen), or a mixed pattern. Informed consent and local ethics committee approval were obtained.
Collection of Samples.
We collected 5 to 10 mL of venous peripheral blood and isolated their PBMC by a Ficoll density gradient centrifugation (Histopaque-1077; Sigma Diagnostics, St Louis, MO).
We determined the total protein content of PBMC by Bradford reagent method, and all PBMC values were normalized to protein amount.13
To characterize the presence of mitochondrial lesion at any level, we measured 9 mitochondrial features related to different aspects of mitochondrial status.
We estimated mitochondrial mass (MM) through the quantification of the porin content, also known as voltage-dependent anion channel protein (Calbiochem Anti-Porin 31HL; Darmstadt, Germany) by Western blot14 and through the estimation of citrate synthase (CS) activity (enzyme commission (EC) number 184.108.40.206) by spectrophotometry at 412 nm.14
Mitochondrial DNA Quantification.
We quantified the amount of mtDNA, extracted by standard phenol–chloroform procedure, through quantitative real-time polymerase chain reaction (PCR; LightCycler FastStart DNA Master SYBR Green I; Roche Molecular Biochemicals, Mannheim, Germany) by 2 separate amplifications of the mitochondrial gene ND2 and the housekeeping nuclear gene 18SrRNA and the results were expressed as the ratio ND2/18SrRNA.14
Cytochrome c Oxidase Subunits II and IV of Complex IV.
We measured protein subunits content of complex IV (CIV) of mitochondrial respiratory chain (MRC); COXII, which is entirely encoded, transcribed, and translated in mitochondria and COXIV, which is nuclear encoded, and transcribed and translated in cytoplasm by Western blot analysis. Details of the experiment are extensively reported elsewhere.14
Enzymatic Activities Measurement.
We measured mitochondrial enzymatic activities glycerol-3-phosphate-dehydrogenase (G3PDH) (EC 220.127.116.11), succinate-ubiquinone reductase or complex II (CII) (EC 18.104.22.168), CII-ubiquinol-cytochrome c reductase or complex III (CIII) (EC 22.214.171.124), cytochrome c oxidase or CIV (EC 126.96.36.199), and the overall CI-III-IV enzymatic activity of the MRC.14 Enzymatic activities were expressed as nanomoles of oxidized substrate per minute and per milligram of protein. The enzymatic activities were normalized per mitochondrion by dividing absolute enzymatic activities per CS activity.
Statistical analyses were performed by Statistical Package for the Social Sciences version 15.0 for windows (SPSS, Inc., Chicago, IL). Results were expressed by means and standard error means. Normality of data was ascertained by Kolmogorov-Smirnov test. Comparisons between groups were carried out by using k-unrelated samples Kruskal-Wallis and 2-unrelated samples Mann-Whitney tests in case of nonparametric distribution. In case of normal data, parametric unrelated samples Student t test was used. P < 0.05 was considered statistically significant.
Characteristics of the patients included in the study were summarized in Table, Supplemental Digital Content 1, http://links.lww.com/INF/A883, which includes clinical and epidemiological data of the patients.
As expected, untreated patients showed higher viral loads than patients on HAART (4.18 ± 0.26 RNA HIV log vs. 0.68 ± 0.19; P < 0.001), but no differences in CD4+ T-cell counts were observed.
MM was maintained in all groups, measured either by CS or voltage-dependent anion channel (Fig., Supplemental Digital Content 2, http://links.lww.com/INF/A884, illustrates mitochondrial content estimated by both the methods).
No significant correlation between mtDNA and lactate levels, CD4+ T-cell counts, and the type of ARV therapy was found (data not shown).
A depletion in mtDNA was found in symptomatic BFA+ groups, either untreated or treated, with respect to controls (2.91 ± 0.43, 4.10 ± 0.41 vs. 5.62 ± 0.50; P = 0.014 and P = 0.024, respectively) (Fig., Supplemental Digital Content 3, http://links.lww.com/INF/A885, represents mtDNA amount in all the groups). Both untreated and treated HIV-infected children without BFA showed lower levels of mtDNA, but significant differences with respect to healthy controls were not attained (4.11 ± 0.36, 4.78 ± 0.44 vs. 5.62 ± 0.50; P = NS). When patients affected with BFA were classified according to viral load (detectable or not), the depletion in mtDNA remained significant in both the groups with respect to controls (3.81 ± 0.42, 3.89 ± 0.49 vs. 5.77 ± 0.49; P = 0.010 and P = 0.012, respectively; data not shown).
Expression of COXII and COXIV subunits of CIV (ratio COXII/COXIV) showed similar results between all groups (Fig., Supplemental Digital Content 4, http://links.lww.com/INF/A886, illustrates subunits COXII and COXIV content).
The measurement of isolated mitochondrial complexes G3PDH-CIII, CII, CII-III, and CIV of MRC (normalized per MM by dividing absolute values per CS activity) remained similar in all groups (Fig. 1A).
However, the assessment of overall functionality of MRC through the measurement of CI-III-IV/CS enzymatic activity showed at least a 50% reduction of the activity in 3 groups with respect to healthy patients (0.023 ± 0.004 in BFA-HAART+, 0.026 ± 0.004 in BFA+HAART-, and 0.014 ± 0.003 in BFA+HAART+ vs. 0.05 ± 0.01; P = 0.025, P = 0.041, and P = 0.004, respectively) (Fig. 1B). Global CI-III-IV/CS activity of BFA-affected patients was also decreased both in patients with detectable and undetectable viral loads with respect to controls (0.020 ± 0.004, 0.03 ± 0.012 vs. 0.05 ± 0.01; P = 0.036 and P = 0.012, respectively).
Among BFA+ patients, those affected with lipoatrophy presented lower mtDNA levels than lipohypertrophic ones (3.19 ± 0.43, 4.29 ± 0.47; P = 0.018), and both the groups presented a decrease in CI-III-IV/CS enzymatic activity that was 2-fold higher in children with lipohypertrophy (0.025 ± 0.004, 0.012 ± 0.002; P = 0.034) (Fig. 2).
Data associating BFA with mitochondrial toxicity in adults are controversial. In some studies, mtDNA depletion has been observed in fat and PBMC from adult patients with BFA.15 In other studies, fat metabolism and mitochondrial function were normal or even increased in this population.16 In the present study, HIV-infected children showed lower mtDNA levels and a reduction in mitochondrial functionality, as assessed per CI-III-IV/CS enzymatic activity, with respect to healthy controls. These findings were more pronounced in patients affected with BFA, regardless of the use of HAART. Even in the presence of mtDNA depletion, no further decrease of subunits COXII and COXIV from MRC CIV was detected. Thus, there was not a deficit in the translational pathway from mitochondrial genome to subunit II or from nuclear genome to subunit IV of MRC CIV. Therefore, the CI-III-IV dysfunction we describe could rather be due to an independent pathway different from mitochondrial genetics or due to a dysfunction of the MRC, not related to any alteration in CIV.
Of note, mitochondrial parameters were differently affected in children presenting with lipohypertrophy (higher CI-III-IV dysfunction) and lipoatrophy (higher mtDNA depletion), suggesting different pathogenic pathways in these different presentations of BFA. To our knowledge, this has not been previously reported, but the fact that most patients were receiving a PI-based regimen at the time of assessment suggests that these findings are more likely related to genetic variability rather than to class-specific toxicity. To date, only one study has evaluated the association of mitochondrial toxicity and BFA in children.12 Cossarizza et al did not find any differences in PBMC mtDNA copies, mitochondrial functionality, or in the trend to undergo apoptosis in HIV-infected children with or without BFA and also in uninfected controls. In our series, we included a group of untreated patients, either presenting BFA or not, and we observed that the presence of genetic and functional mitochondrial damage arises in symptomatic BFA+ patients regardless of the administration of therapy. Mitochondrial alterations were found in both BFA+HAART– and BFA+HAART+ groups of children, at both genetic and functional levels. In patients with long-term undetectable viral load after HAART implementation, the mitochondrial dysfunction should probably be ascribed to ARV-related toxicity, whereas in the patients who are off HAART, the HIV-related mitochondrial damage should also be taken into account.
Similar previous findings of mitochondrial genetic and functional impairment in adipose tissue from HIV-infected adults corroborate the potential mitochondrial basis for BFA development.17 Accordingly, Milazzo et al18 have recently observed a protective role of antioxidant supplementation on mitochondrial function in HIV-infected patients presenting lipoatrophy.
As opposite to Cossarizza et al,12 our findings suggest that the use of PBMC may be an adequate indicator for the assessment of the mitochondrial toxicity that remains at the etiopathogenic basis of BFA in HIV-infected children. Many other studies have previously validated the use of PBMC model, as a noninvasive approach, especially important in children, in the assessment of HIV- and HAART-mediated mitochondrial toxicity.1,10,14,19,20 Small numbers, the heterogeneity in patients characteristics, and the cross-sectional design are obvious methodologic limitations of our study; in fact, we were not able to identify any of the acquired or inherited factors that have been associated with the different patterns of BFA, such as the use of stavudine, AIDS, nadir CD4 counts, or IL-1beta (+3954C/T) polymorphism.21
According to our findings, HIV and HAART exert similar mitochondrial toxicity. It is important to keep attention at the secondary events resulting from ARV drugs. It is remarkable fact that mitochondrial toxicity derived from ARV drugs seems to be reversible, once the toxic agent is interrupted.19
Our results indicate that mitochondrial function is more severely affected in BFA+ patients than in HIV-treated patients without BFA, suggesting a mitochondrial etiology for BFA. Further studies of mitochondrial function and evolution on children over time, considering the different HAART regimens, should be addressed to ascertain more secure ARV regimens and reduce BFA development.
1.Côté H, Brumme ZL, Craib KJ, et al. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. N Engl J Med
2.Pilon AA, Lum JJ, Sanchez-Dardon J, et al. Induction of apoptosis by a nonnucleoside human immunodeficiency virus type 1 reverse transcriptase inhibitor. Antimicrob Agents Chemother
3.Casademont J, Barrientos A, Grau JM, et al. The effect of zidovudine on skeletal muscle mtDNA in HIV-1 infected patients with mild or no muscle dysfunction. Brain
4.Dalakas MC, Illa I, Pezeshkpour GH, et al. Mitochondrial myopathy caused by long-term zidovudine therapy. New Engl J Med
5.Viganò A, Giacomet V. Nucleoside analogues toxicities relate to mitochondrial dysfunction: focus on HIV-infected children. Antivir Ther
. 2005;10(suppl 2).
6.Caron-Debarle M, Lagathu C, Boccara C, et al. HIV-associated lipodystrophy: from fat injury to premature aging. Trends Mol Med
7.Lichtenstein KA, Delaney KM, Armon C, et al. Incidence of and risk factor for lipoatrophy (abnormal fat loss) in ambulatory HIV-1-infected patients. J Acquir Immune Defic Syndr
8.Brinkman K, Smeitink JA, Romijn JA, et al. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet
9.Kim R, Rustein RM. Impact on antiretroviral therapy on growth, body composition and metabolism in pediatric HIV patients. Paediatr Drugs
10.Saitoh A, Fenton T, Alvero C, et al. Impact of nucleoside reverse transcriptase inhibitors on mitochondrial in human immunodeficiency virus type 1-infected children receiving highly active antiretroviral therapy. Antimicrob Agents Chemother
11.Rosso R, Nasi M, Di Biagio A, et al. Effects of the change from stavudine to tenofovir in human immunodeficiency virus-infected children treated with highly active antiretroviral therapy: studies on mitochondrial toxicity and thymic function. Pediatr Infect Dis J
12.Cossarizza A, Pinti M, Moretti L, et al. Mitochondrial functionality and mitochondrial DNA content in lymphocytes of vertically infected human immunodeficiency virus-positive children with highly active antirretroviral therapy- related lipodystrophy. J Infect Dis
13.Bradford M. A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem
14.Miró Ò, López S, Rodríguez de la Concepción M, et al. Upregulatory mechanisms compensate for mitochondrial DNA depletion in asymptomatic individuals receiving stavudine plus didanosine. J Acquir Immune Defic Syndr.
15.Nolan D, Hammond E, Martin A, et al. NRTI therapy is associated with mitochondrial DNA depletion, increased mitochondrial biogenesis and subcutaneous fat wasting in HIV infected patients (abstract 522). In: Final Program and abstracts of the 1st IAS Conference on HIV Pathogenesis and Treatment; Buenos Aires, Argentina; 2001.
16.Sekhar RV, Jahoor F, Visnegarwala F, et al. Sysregulation of lipid turnover is a key defect in the HIV lipodystrophy syndrome. Antivir Ther
. 2000;5(suppl 5):38.
17.Garrabou G, López S, Morén C, et al. Mitochondrial damage in adipose tissue of untreated HIV-infected patients. AIDS
18.Milazzo L, Menzaghi B, Caramma I, et al. Effect of antioxidants on mitochondrial function in HIV-1-related lipoatrophy: a pilot study. AIDS Res Hum Retroviruses
19.Noguera A, Morén C, Rovira N, et al. Evolution of mitochondrial DNA content after planned treatment interruption of HIV-infected pediatric patients. AIDS Res Hum Retroviruses
20.Morén C, Noguera-Julian A, Rovira N, et al. Mitochondrial assessment in asymptomatic HIV-infected pediatric patients on Highly Active Antiretroviral Therapy. Antivir Ther
. In press.
21.Asensi V, Rego C, Montes AH, et al. IL-1beta (+3954C/T) polymorphism could protect human immunodeficiency virus (HIV)-infected patients on highly active antiretroviral treatment (HAART) against lipodystrophic syndrome. Genet Med