Mitoquinone mesylate attenuates brain inflammation in humanized mouse model of chronic HIV infection

Despite effective antiretroviral treatment (ART), people with HIV (PWH) are predisposed to age-related conditions, such as HIV-associated neurocognitive disorders (HAND) that may impact more than 25% of PWH [1]. Thus, ART alone is inadequate treatment for HAND and there is an unmet need for novel therapies for HAND. Persistent viral reservoirs, despite ART, impact innate immune cells and microglia leading to oxidative stress (OS), inflammation, immune activation and ultimately neurotoxicity [1] that contribute to pathogenesis of HAND [2]. The brain is enriched in mitochondria that are responsible for most of the cellular reactive oxygen species (mito-ROS) [2]. HIV-1 promotes aberrant production of mito-ROS that induce mitochondrial and cellular damage and inflammatory signaling pathways [3,4] that drive pathology of neurodegenerative disorders [2,5]. The mitochondrial antioxidant MitoQ (MitoQ Ltd, Auckland, New Zealand) can target the harmful production of mito-ROS and may also attenuate secretion of IL-1β and IL-6 that contribute to neuroinflammation in HAND [3,4]. Thus, we hypothesized that MitoQ may be effective in reducing neuroinflammation that leads to HAND. Herein, using a preclinical mouse model of chronic HIV, we demonstrate that MitoQ can attenuate proinflammatory cytokines and chemokines in the brain of HIV-1-infected humanized mice on ART.

NOD scid gamma (NSG) Bone Marrow and Liver Thymic (BLT) mice (equal number of males and females; n = 27) were generated, as described [6] in accordance with federal, state, and local-approved guidelines. NSG BLT mice were infected with dual-tropic HIV-1 89.6 and treated with daily ART [Tenofovir Disoproxil Fumarate (8.75 mg/kg)/Emtricitabine (13 mg/kg)/Raltegravir (17.5 mg/kg)] as previously described [6]. Mice were administered for 90 days 500 mmol/l MitoQ via drinking flavored water as previously [7]. Control mice did not receive MitoQ. Three groups were used: uninfected (group A: HIV⁻, n = 5), infected and on ART (group B: HIV⁺ART⁺, n = 12), and infected on ART and MitoQ (group C: HIV⁺ART⁺MitoQ⁺, n = 10). Brain human and murine cytokines and chemokines were determined using Luminex immunoassays according to the manufacturer’s instructions (R&D Systems, Minneapolis, MN, USA) and as described previously [6]. The Kruskal–Wallis test or the Mann–Whitney compared different groups. P values less than 0.05 by either test were deemed significant. All analyses were performed with Graphpad, v8.0 (GraphPad Software Inc., San Diego, CA, USA).

NSG BLT mice are a valuable model for HIV immunopathogenesis. Using sensitive Luminex immunoassays, we first assessed the direct impact of HIV and ART on protein levels of human cytokines and chemokines in NSG BLT mouse brain as a measure of neuroinflammation in HIV (Fig. 1a). Similar to our prior data that potent ART effectively suppressed HIV-1 replication in NSG BLT mice [6], all HIV⁺ART⁺ NSG BLT mice had suppressed plasma viremia after 4 weeks of potent ART (data not shown). NSG BLT HIV⁺ART⁺ mice had increased levels of the cytokines human (h) IL-1β (Fig. 1b), h-IL-8 (Fig. 1c) h-TNFα (Fig. 1d) (P < 0.01 for all comparisons) and a trend (P = 0.052) for increased levels of the chemokine h-CCL2 (Fig. 1e) in their brain compared with the uninfected mice after 90 days of ART. Compared with HIV-infected mice given ART, the addition of MitoQ 500 μmol/l in water for 60 days in HIV⁺ART⁺ mice (HIV⁺ART⁺ MitoQ⁺ mice) led to decreased levels of h-IL-1β, h-IL-6, h-TNFα, and h-CCL2 and murine (m-) m-IL-1β, m-IL-6 and m-TNFα in the brain (Fig. 1b-i) (P < 0.01 for all comparisons). No differences were observed in levels of human and murine IL-10, CCL3, CCL5, and CXCL10 among all groups (P > 0.05 for all comparisons) (Fig. 1c and j).

Fig. 1. MitoQ reduced proinflammatory cytokines and chemokines in the brain in a mouse model of chronic treated HIV-1 infection.

(a) Study design. NSG humanized mice were infected with HIV-1 for 90 days and treated with antiretroviral therapy (ART) alone or in combination with oral MitoQ given in water (500 μmol/l) for 60 days. Brain tissue from each mouse was collected and cytokines and chemokines that predict morbidity in chronic treated HIV-1 infection were determined in uninfected, HIV⁺ART⁺ and HIV⁺ART⁺MitoQ⁺ mice, by Luminex immunoassays. The first group was uninfected, white boxes and datapoints (n = 5). The second group included HIV-infected mice on potent ART with suppressed plasma viremia, light grey boxes and datapoints, HIV⁺ART⁺ (n = 12). The third group was infected, and received both ART and MitoQ, blue boxes and datapoints, HIV⁺ART⁺MitoQ⁺ (n = 10). Summary data of human (h-) IL-1β (b), h-IL-8, h-IL-10, h-CCL2, h-CCL3, h-CCL5, h-CXCL10 (c), h-TNFα (d), h-CCL2 (e), h-IL-6 (f), murine (m-) IL-1β (g), m-IL-6 (h), m-TNFα (i) and m-IL-10, m-CCL2, m-CCL3, m-CCL5, m-CXCL10 (j) in the brain of humanized mice. Data are presented as protein levels of each cytokine or chemokine (pg) normalized by the total protein amount in each tissue sample (per 10 μg) as determined by bicinchoninic acid (BCA) protein assay. Data represent box and whiskers with minimum, median and maximum values and each datapoint represents a mouse and the average of at least two replicates. Protein levels in HIV⁺ART⁺ and HIV⁺ART⁺MitoQ⁺ mice in (c) and (j) are presented as normalized data fold to the mean of the uninfected group. The Kruskal–Wallis test was used to compare three groups and the Mann–Whitney test was used to compare two groups (*P < 0.05, **P < 0.01, ***P < 0.001).

To our knowledge, this is the first preclinical demonstration of the potential therapeutic role of MitoQ for treatment of neuroinflammation, an established instigator of HAND, in chronic treated HIV. Herein, we showed that MitoQ effectively attenuated proinflammatory cytokines and chemokines (IL-1β, IL-6, TNFα, and CCL2) in the brain of HIV-1-infected BLT mice on potent ART. Notably, in addition to its antioxidant effect, MitoQ has anti-inflammatory effects, given that mito-ROS contribute to activation of the inflammasome and IL-1β [7], which collectively contribute to neurological impairment [8]. Consistent with our findings, prior in-vivo studies, demonstrated that MitoQ inhibited neuroinflammation in preclinical models of neurodegeneration [4,9].

Although HIV-1 does not infect murine microglia cells in BLT mice, the NSG BLT model can support HIV infection and can be used to monitor human immune responses to HIV-1 and ART. Further work is necessary to assess the impact of MitoQ in all known mediators of neuroinflammation. MitoQ is well tolerated and has been used as an oral diet supplement in humans for decades. It has also been used safely in clinical trials with no toxic effects [3,10]. Herein, we provide robust preclinical evidence that supports clinical trials to test whether MitoQ can be a useful treatment for neuroinflammation in HAND.

Data availability:

all data needed to understand and assess the conclusions of this research are available in the main text and supplementary materials. Raw datasets supporting the findings of this study are available from the corresponding author on reasonable request.