iconstar paper   HIV Articles<  
Back grey arrow rt.gif
Mitochondrial toxicity and caspase activation in HIV pregnant women
  Download the PDF here
30 August 2016
J. Cell. Mol. Med. Vol 21, No 1, 2017
Sandra Hernandez a, b, c, #, Constanza Moren b, c, #, Marc Catalan-Garcıa b, c, Marta Lopez a, c,
Mariona Guitart-Mampel b, c, *, Oriol Coll d, Laura Garcia a, c, Jose Milisenda b, c, Angela Justamante b, c, Josep Maria Gatell e, Francesc Cardellach b, c, Eduard Gratacos a, c,
Oscar Miro b, c, Gloria Garrabou b, c
a Maternal-Fetal Medicine Department, Clinical Institute of Gynecology, Obstetrics and Neonatology, Hospital Clinic of Barcelona,
Barcelona, Spain
b Muscle Research and Mitochondrial Function Laboratory, Cellex-IDIBAPS, Faculty of Medicine-University of Barcelona,
Hospital Clinic of Barcelona, Barcelona, Spain
c Centro de Investigacion Biomedica en Red de Enfermedades Raras (CIBERER), Madrid, Spain d Clinica Eugin, Barcelona, Spain
e Infectious Disease, Hospital Clinic of Barcelona, Barcelona, Spain
Mitochondrial disturbances in HIV

To assess the impact of HIV-infection and highly active anti-retroviral treatment in mitochondria and apoptotic activation of caspases during pregnancy and their association with adverse perinatal outcome. Changes of mitochondrial parameters and apoptotic caspase activation in maternal peripheral blood mononuclear cells were compared at first trimester of pregnancy and delivery in 27 HIV-infected and -treated pregnant women versus 24 uninfected pregnant controls. We correlated immunovirological, therapeutic and perinatal outcome with experimental findings: mitochondrial DNA (mtDNA) content, mitochondrial protein synthesis, mitochondrial function and apoptotic caspase activation. The HIV pregnancies showed increased adverse perinatal outcome (OR: 4.81 [1.14-20.16]; P < 0.05) and decreased mtDNA content (42.66 ± 5.94%, P < 0.01) compared to controls, even higher in naïve participants. This depletion caused a correlated decrease in mitochondrial protein synthesis (12.82 ± 5.73%, P < 0.01) and function (20.50 ± 10.14%, P < 0.001), not observed in controls. Along pregnancy, apoptotic caspase-3 activation increased 63.64 ± 45.45% in controls (P < 0.001) and 100.00 ± 47.37% in HIV-pregnancies (P < 0.001), in correlation with longer exposure to nucleoside analogues. HIV-infected women showed increased obstetric problems and declined genetic and functional mitochondrial parameters during pregnancy, especially those firstly exposed to anti-retrovirals. The apoptotic activation of caspases along pregnancy is emphasized in HIV pregnancies promoted by nucleoside analogues. However, we could not demonstrate direct mitochondrial or apoptotic implication in adverse obstetric outcome probably because of the reduced sample size.
The effectiveness of highly active antiretroviral treatment (HAART) regimens in reducing mother-to-child vertical transmission (MTCT) of HIV-infection and in delaying disease progression has been demonstrated and should therefore be offered to all pregnancies [1-3]. Anti-retroviral (ARV) treatment comprised of two nucleoside reverse transcriptase inhibitors (NRTIs) and non-nucleoside reverse transcriptase inhibitor (NNRTI) or protease inhibitor (PI) is recommended in pregnancy by the United States, World Health Organization and European guidelines [4, 5]. Additional measures including selective caesarean or avoidance of breastfeeding are strongly recommended. However, adverse pregnancy outcome have been increasingly reported by several observational studies in HIV-infected women exposed to HAART [6-11]. Anti-retrovirals and HIV-infection have been associated with pre-eclampsia, stillbirth, pre-term labour, low birth weight and intrauterine growth restriction (IUGR) [12-17].
One of the most serious complications associated with ARVs is mitochondrial toxicity. Mitochondrial-derived clinical effects of NRTI have been firmly established in HIV-infected non-pregnant adults [18-20]. These negative effects depend on the capacity of NRTIs to inhibit DNA polymerase gamma, the enzyme devoted to mitochondrial DNA (mtDNA) replication, leading to a decrease in mtDNA copy number and quality, which may, finally, cause mitochondrial dysfunction [18]. Depletion of mtDNA has been extensively described in different tissues of human and animal models (placenta, foetal cord blood, heart, adipose tissue, skeletal muscle, brain and kidney, among others) leading to mitochondrial morphologic, metabolic and energetic abnormalities [21-25]. Such mitochondrial disturbances are enhanced by HIV, which has been additionally blamed for triggering apoptosis [26]. Associated clinical syndromes have been expanded to include lipoatrophy, peripheral neuropathy, cardiomyopathy, lactic acidosis and hepatic steatosis [27, 28]. Moreover, accelerated 'mitochondrial aging' associated with ARV may contribute to cardiovascular disease, malignancies and frailty [29].
Maternal death has been described as a result of lactic acidosis in women receiving long-term treatment with a combination of NRTI [30, 31], especially in the third trimester of pregnancy and in ARV schedules including two NRTIs. Fortunately, severe maternal mitochondrial toxicity associated with NRTI during pregnancy appears to be rare and is reversible on treatment discontinuation. However, milder forms of mitochondrial toxicity are commonly reported and may have future long-term effects.
Although ARV use during pregnancy is considered safe, data on ARV and pregnancy, especially in HAART, are insufficient, and safety and long-term health consequences are currently unknown. The few studies on long-term HAART exposure have focused on foetal and perinatal ARV effects [32-37], but rarely on maternal-related problems which may, in turn, affect children.
We recently described that HAART toxicity may cause subclinical mitochondrial damage in pregnant women and their newborn [38] by reducing mtDNA levels, mitochondrial protein synthesis and mitochondrial function. Additionally, increased apoptosis through caspase-3 activation was observed in HIV-pregnant women, but not in their children, cross-sectionally, at delivery.
The aim of this study was to investigate the impact of HIV-infection and HAART on mitochondria and apoptotic caspase activation during pregnancy to assess their implication in the increase of adverse perinatal outcome characteristic of HIV-pregnancies to establish potential prenatal prognosis markers.
Anti-retrovirals are indispensable in the treatment and prevention of HIV infection. Although their use during pregnancy is considered safe, there are still lingering concerns about long-term health consequences.
Several studies have demonstrated mitochondrial toxicity in animal models, HIV-infected infants and adults on NRTI therapy and newborns exposed in utero to ARVs [22, 24, 37, 38, 47], but it is currently unknown whether HIV-pregnancy may be an additional risk for the onset of mitochondrial toxicity.
Several physiopathological mechanisms have been proposed to explain the adverse clinical effects in HIV pregnancies. Mitochondrial bioenergetics conditions foetal growth and early postnatal adaptation. Given that mitochondria are exclusively inherited from the maternal ovum, early exposure of the ova and the mitochondria to ARV affects foetal development. We have demonstrated that oocytes from infertile HIV-infected HAART-treated women show a decreased mtDNA content that could explain their poor reproductive outcome [19]. Several mitochondrial alterations are associated with initiation of mitochondrial biogenesis and activation during early states of embryo development. The mtDNA content and expression levels of genes involved in the maintenance and regulation of biogenesis change during human foetal development. The foetus is exposed to the potentially stressful in uteroenvironment created by maternal ARV-associated metabolic toxicities and placental transferred ARVs at critical points in its development.
Colleoni et al. have recently described significantly decreased mtDNA in non-infected women carrying IUGR fetuses compared to control pregnancies. The authors concluded that in HIV-uninfected pregnancies mtDNA content and mitochondrial function may help recognizing adverse perinatal outcome [42]. It is unknown whether mitochondrial depletion is a predictor factor for poor pregnancy outcome in HIV patients. Indeed, only two studies have evaluated the association between mitochondrial toxicity and poor pregnancy outcome in HIV-infected pregnant women on HAART. Both analyse mitochondrial status in the third trimester of gestation. Nasi et al. described a decrease in mtDNA content in subcutaneous fat of HIV-pregnant women taking ARV compared to uninfected women without ARV [43]. The second report, published by our group, described that HAART toxicity may cause subclinical mitochondrial damage in blood of HIV-pregnant women and their newborn compared to uninfected pregnancies [27]. To our knowledge, the present work is the first longitudinal study investigating the evolution of mtDNA content, mitochondrial protein synthesis, mitochondrial function and apoptotic caspase activation in HIV-infected women during pregnancy or the potential use of these parameters as pronostics factors to predict adverse perinatal outcome in HIV pregnancies.
In this study we found a higher prevalence of adverse perinatal outcome in the HIV cohort compared with controls, thereby confirming the deleterious effect of both the virus and the treatment on foetal development as well as validating our sample.
During pregnancy, we observed a progressive decrease in mtDNA content which was significantly higher in HIV-infected pregnant women. Our results agree with two previous studies performed in pregnant women which evaluated blood mtDNA content along pregnancy [38-42]. Colleoni et al. reported a significant decrease in mtDNA content in the blood of uninfected pregnant women in the first, second and third trimesters compared to non-pregnant women [42]. This decrease was higher in HIV-pregnancies [38] and, according to the present findings, would be partially developed along pregnancy.
The mtDNA depletion observed in our HIV cohort was even greater in naïve pregnant women who started HAART during the second trimester of gestation compared with pregnancies under HAART prior to conception. This finding could be explained by the initiation of ARV, abruptly increasing mitochondrial toxicity and resulting in a dramatic decrease in mtDNA content which is maintained over time. This theory is in agreement with a previously published study by our group [44].
In our study we have demonstrated that a decrease in mtDNA leads to a significant reduction in downstream mitochondrial protein synthesis and mitochondrial function in pregnancies complicated with HIV infection and ARV. Correlations were found between genetic and functional mitochondrial parameters demonstrating (i) that proper mitochondrial functionalism relies on proper levels of mtDNA copies and (ii) the etiopathogenic cause of the dysfunction observed is because of NRTI toxicity and interference in mtDNA replication.
Under pathological conditions, mitochondrion triggers apoptosis. Consequently, any mitochondrial disarrangement may have fatal cell consequences. An important pre-requisite for a successful pregnancy is that the maternal immune system does not reject the foetus and thus, cellular immune response could be essential. Apoptosis has also been shown to play an important role in promoting maternal immune tolerance to paternal antigens expressed by trophoblastic cells [45], which is a physiological process during pregnancy. While apoptosis is thought to be important as a normal physiological feature for foetal or placental development, enhanced levels may also be involved in the pathological conditions. Higher apoptosis levels may have implications in adverse perinatal outcome. A greater incidence of apoptosis has been observed in conditions such as pre-eclampsia and IUGR, suggesting that appropriate regulation of apoptosis is important for normal pregnancy [46].
Blood cells apoptotic activation of caspase-3 increased along pregnancy in both our cohort of HIV-patients under treatment and in controls. In uninfected pregnancies the apoptotic rate of caspase-3 activation probably increases as a physiological mechanism to delete newborn cells from the maternal blood [45]. However, this increase was significantly enhanced in HIV pregnancies in concordance with accumulated, previous NRTI exposure. Apoptosis of uninfected cells is a key element of HIV pathogenesis and is believed to be the driving force behind the selective depletion of CD4+ T cells leading to immunodeficiency. We found that HIV infection and ARV have a significant impact on apoptotic activation of PBMC's caspases. In the first trimester of gestation HIV-positive women showed higher levels of PBMC apoptotic caspase-3 activation specially in those with longer exposure to HIV or NRTI. Additionally, at delivery, HIV-infected women longer exposed to NRTI showed higher levels of PBMC caspase-3 activation. These results are in agreement with previously published work [38].
However, we did not find any association between mtDNA content, mitochondrial protein synthesis, mitochondrial function or apoptotic caspase activation and adverse perinatal outcome in HIV patients. We found an increased prevalence of adverse pregnancy events in HIV pregnancies and enhanced trends towards mitochondrial impairment and apoptotic caspase activation along pregnancy in accordance to HIV or NRTI exposure, but did not find a significant association between these molecular findings and poor obstetric outcome. We did not find the presence of different patterns of mitochondrial toxicity and apoptotic caspase activation in HIV pregnancies correlating with a distinct pattern of clinical expression, probably because of the small sample size, which made patient stratification and statistical findings difficult. Other constraints of this study may be the presence of different types of HAART and time-exposure to HIV or ARVs, characteristic of observational studies and personalized treatment interventions, which on the other hand may enclose findings to reality. The inclusion of a control group of non-pregnant HIV-infected and treated women may have been useful to assess longitudinal mitochondrial and apoptotic toxicity of HIV and HAART without potential gestation interference. However, such studies are extensively documented in the bibliography [47] and, additionally, the interest of the present work was focused on obstetric problems and, thus, in pregnant women. Additionally, to overcome methodological pitfalls and strength, the significance of reported findings, the deregulation of bioenergetics capacity or apoptotic status in studied patients was assessed, in parallel, by Western Blotting and functional measures, which rendered similar results. However, detailed mechanistic pathways underlying HIV- and HAART-associated toxicity in HIV pregnancies should be further elucidated.
In conclusion, although pathogenetically plausible with these findings, it is not yet possible to prove a cause-effect association of adverse perinatal outcome with mitochondrial and apoptotic toxicity of HIV and NRTIs exposure during pregnancy. As a consequence, we did not succeed in the secondary objective of this study; identify epidemiological risk factors and prognostic markers of mitochondrial toxicity or apoptotic caspase activation for potentially associated poor clinical results in HIV-infected mothers.
Several groups have proposed that monitoring possible markers of mitochondrial dysfunction in peripheral blood of pregnant women may be useful for detecting preclinical NRTI toxicity. Furthermore, prospective studies in HIV pregnancies under HAART are needed, to determine whether the incidence of mitochondrial disorders differs according to the regimen used and to develop predictive models to identify mothers-infants at highest risk. In an era of expanding treatment options, minimizing toxicities is possible and necessary. The role of mitochondria and apoptosis in physiological conditions must also be clarified to define the relevance of mitochondrial or apoptotic alterations in HIV pregnancies. As a proof-of-concept, the present study has been conducted in a single-site centre and in a limited population. The short- and long-term health consequences of mitochondrial toxicity and apoptosis in HIV pregnancies should be further investigated in larger cohorts. There is a crucial need to fully understand the scope and depth of this problem through continued basic and clinical research evaluating the effects of foetal and maternal ARV exposure to better understand the morbidity associated with mitochondrial toxicity or apoptosis in pregnant women exposed to HIV and HAART. The current challenge is to design new ARV schedules with reduced harmful mitochondrial and apoptotic effects.
  iconpaperstack View Older Articles<   Back to Top<   www.natap.org<