2.1 Human immunodeficiency virus

HIV belongs to the genus Lentivirus in the family Retroviridae.15 Two types of genetically and antigenically different viruses are known as HIV-1 and HIV-2. The vast majority of HIV infections in the global pandemic are caused by HIV-1. Most HIV-2 cases are confined to some West African countries with their epicentre in Guinea-Bissau.16

HIV is present in body fluids as free virus particles and within infected immune cells and causes acquired immunodeficiency syndrome (AIDS). It primarily infects CD4+ T cells, macrophages and dendritic cells, in order to carry out its replication cycle. HIV infection is associated with a progressive decrease in CD4+ T-cell count and an increase in viral load. In the haematopoietic system, CD4+ T lymphocytes are the most visibly infected cell type since they express the CD4 molecule used by HIV as a receptor and can efficiently replicate the virus. Macrophages are also frequently found to be infected with HIV, but this infection may go unnoticed due to low viral production.17

HIV kills CD4+ T cells by three mechanisms: (a) by direct viral destruction of infected cells, (b) by increasing apoptosis rates in infected cells, and (c) by CD8 cytotoxic cell-mediated killing of infected CD4+ T cells. When CD4+ T-cell numbers drop below a critical level, cellular immunity is lost and the body becomes progressively more susceptible to opportunistic infections and neoplasms. The stage of infection, which presents different phases, can be determined by measuring the CD4+ T-cell counts and the viral load of the patients (Figure 2).

Different stages of HIV infection over time. The stages are (a) acute infection (also known as primary infection), which lasts for several weeks and it can include symptoms like fever, lymphadenopathy, pharyngitis, myalgia, or mouth and esophageal sores. (b) The latency stage involves few or no symptoms and can last from 2 weeks to 20 years or more. (c) AIDS defined by low CD4+ T cell counts <200/μl, increased viral loads, various infections opportunists and cancers1, 2

The transmission of HIV is greatly influenced by the amount of infectious virus particles in a body fluid and the extent of contact with that body fluid. Epidemiological studies during 1981 and 1982 indicated that the main routes of transmission of HIV were intimate sexual contact and contaminated blood. AIDS was initially described in homosexual and bisexual men and intravenous drug users, but its transmission as a result of heterosexual activity was also soon recognized. Furthermore, it became apparent that transfusion recipients and haemophiliacs could contract the disease by transfusion of blood or blood products and that mothers could transfer the causative agent to newborns as well. These three main means of transmission: parenteral, sexual and vertical (including during pregnancy at delivery and through breast milk) can be largely explained by the high concentrations of HIV in various body fluids. It is worth mentioning that the “optimal prevention scenario” would be the pregnant woman adheres to treatment, under regular care and with a suppressed HIV viral load of <50 copies of RNA/ml throughout pregnancy and lactation. When these criteria are met, the theoretical risk of mother-to-child transmission is practically zero.18, 19 Vertical transmission of HIV occurs in 11%–60% of children born to HIV-positive mothers.20 The latest findings suggest that the level of free infectious virus in maternal blood could predict the infection outcome of the newborn.2, 17 Vertical HIV transmission may decrease from 25% to 2% with the use of ARVs that include nucleoside analogues, during pregnancy. However, there is some evidence that exposure in the womb to nucleoside analogues can cause mitochondrial dysfunction symptoms in a small number of HIV-uninfected children.21 More studies are needed to elucidate the mechanisms of the mitochondrial dysfunction, focusing on exposure in utero, and to identify the importance of mitochondrial variations in children without clinical signs of mitochondrial dysfunction.21

2.1.1 HIV structure

HIV is a spherical particle 120 nm in diameter made up of three layers, including (i) a lipid envelope which is an external bilayer of phospholipids, coming from the infected host cell, containing class I and II histocompatibility antigens and some adhesion molecules that facilitate contact with target cells. It also contains 72 copies of a viral glycoprotein complex, called Env that protrudes through the surface to the outer environment. These glycoprotein complexes consist of a head of three gp120 glycoproteins and a body of three gp41 molecules, anchored to the molecules of the viral envelope. They allow the virus to bind and fuse cells to start the infectious cycle. Both surface proteins have been considered as possible targets for future development of treatments or vaccines.22 (ii) A capsid or matrix which is a spherical intermediate structure containing p17 protein and (iii) a nucleocapsid or nucleus which is an icosahedral internal structure, consisting of p24 protein. It contains the viral genome (two identical, single-stranded RNA molecules), the p9 and p7 nucleoproteins, and the machinery required for viral replication (reverse transcriptase, integrase and protease). As a retrovirus, its enzyme, reverse transcriptase, converts viral RNA into proviral DNA. The HIV genome is characterized by a high mutation rate, due to the errors of the reverse transcriptase enzyme during the back transcription of RNA into DNA, and the recombination capacity of the different viruses that can coexist within a cell. Consequently, HIV has high genetic variability, which hinders both the defence response of the immune system and the creation of an effective vaccine system against the virus.2

2.1.2 Replication cycle of HIV

Like all obligate intracellular pathogens, HIV must take advantage of the multiple functions of the host cell to replicate successfully.23 When HIV enters the target cell, the viral RNA genome becomes double-stranded proviral DNA and is imported into the cell nucleus and integrated into the cellular DNA by a virally encoded integrase and host cofactors.24 Once integrated, the virus can become dormant, allowing both the virus and its host cell to avoid detection by the immune system. Alternatively, the virus can be transcribed, producing new viral RNA and protein genomes that are packaged and released from the cell as new viral particles that begin the replication cycle. Importantly, most of the HIV proteins exert mitochondrial interactions, as depicted in Table 1. Mitochondrial changes derived from the viral interactions occur either in the host cells (mainly CD4+ T-cell lymphocytes and macrophages, but also other lymphocytes, neuronal and glial cells from the central nervous system (CNS), enterochromaffin cells from the gut and dendritic cells, including Langerhans cells,33 or bystander cells. Apoptosis of uninfected bystander cells is a key element of HIV pathogenesis and represents a driving force to the important CD4+ loss which cannot be explained only by the direct infection.34 While several viral proteins have been implicated in this process, the complex interaction between Env glycoprotein expressed on the surface of infected cells and the receptor and co-receptor expressing bystander cells has been proposed as a major mechanism. Laurent-Crawford et al.35 were the first to demonstrate that the HIV Env glycoprotein alone expressed on the surface of cells is capable of inducing cell death in neighbouring T cells. Importantly, the effects of HIV proteins and/or ARV on mitochondria may differ depending not only on whether the target is a host or a bystander cell, but also on the cell type. As an example of the latter, HIV gp120 and Tat have been shown to alter autophagy and mitophagy in neurons and Tat also alters mitophagy in microglial cells.36 Although this could certainly affect children, no data have been reported so far in paediatric population.

TABLE 1. Viral proteins of HIV and mitochondrial interactions in the host cells Type Protein Mechanism of action and mitochondrial interactions Structural Env17

Allows the virus to target and bind to specific cell types and infiltrate the cell membrane

Increases Bax (pro-apoptotic)

Decreases Bcl-2 (anti-apoptotic)

Activates mitochondrial apoptosis

Regulatory Tat17, 25-28

Reduces the expression of the mitochondrial superoxide dismutase 2 isoenzyme, (endogenous inhibitor of the permeability of the mitochondrial membrane) and triggers the loss of mitochondrial membrane potential

Increases Fas ligand expression in T cells, inducing apoptosis

Promotes Tat secretion by infected cells, promoting mitochondrial apoptosis in uninfected T cells

Induces apoptosis by a mechanism involving disruption of calcium homeostasis

Rev17

Ensures the replication of HIV in the infected cell

Targets the permeability transition pore, allowing the permeabilization of the mitochondrial membranes

Complementary Nef17

Regulates CD4+ expression on the cell surface

Disrupts T cell activation

Stimulates HIV infectivity

Vpr29-32

Blocks the cell cycle in G2

Blocks cell division

Prevents the activation of the complex p34cdc2/cyclin B, a known cell cycle regulator, required for entering into mitosis

Regulates apoptosis and transcriptional modulation of immune function

Vpu29

Promotes CD4+ modulation

Increases the release of virions

Is responsible for releasing the viral envelope, triggering the degradation of CD4+ molecules bound with Env

The viral cycle is divided into four stages: (i) fusion and entry of HIV: The gp120 glycoprotein binds CD4, undergoes a conformational change, and interacts with the cell co-receptor (CCR5 or CXCR4), prompting conformational changes in the viral gp41 glycoprotein.37 Subsequently, once fused with the cell membrane, HIV releases its genetic material (viral RNA) into the cytoplasm of the cell, along with viral proteins.2 (ii) Reverse transcription and integration of proviral DNA: single-stranded RNA is converted to double-stranded DNA through the activity of the viral reverse transcriptase enzyme. Viral proteins help double-stranded proviral DNA reach the nucleus and integrate into the cell genome, through the virus integrase enzyme. In the event of HIV entering a quiescent cell, the proviral DNA will accumulate in the cytoplasm without any integration, leading to latency. The latent provirus that exists as a reservoir within quiescent cells greatly hampers both the effective endogenous system and HIV treatment, as it avoids immune and exogenous control.2 (iii) Expression of the viral genome: once the proviral DNA has been integrated into the target cell nuclear genome, some viral proteins, along with cellular transcription factors, such as nuclear factor kappa, enhancer of the B-cell light chain activated κB (NF-κB), induce replication and transcription of the viral genome. Initially, transcription leads to the synthesis of HIV regulatory proteins (Tat, Rev, Vpr, Vpu and Nef).17, 25-27, 29-32, 38 Messenger RNA is produced as a single transcript that is transported to the cytoplasm and is processed into many different RNAs of different sizes. HIV protease is in charge of the conversion of a large protein precursor molecule to small active and functional molecules.2, 24 (iv) Assembly of new viral particles: all functional viral compounds are assembled giving rise to new viral particles that are released into the bloodstream, to infect other cells. The lifespan of HIV in plasma is 6 h. To maintain a constant viral concentration in the body,39, 40 new viral particles are produced daily. This fact makes it difficult to find an effective treatment against the virus.2

2.1.3 HIV in the paediatric population

The development of effective therapy for HIV infection has substantially reduced HIV-related morbidity and mortality, making HIV infection a chronic disease.41 The life expectancy of people with HIV has increased in countries where ARVs are widely used, although the continued spread of the pandemic has increased the number of people living with HIV. In 2018, around 1.7 million people contracted HIV globally, a 16% drop from 2010 that is driven, mostly, by steady progress in most of Eastern and Southern Africa. For example, South Africa has come a long way as it has significantly reduced new HIV infections (by more than 40%) and AIDS-related deaths (by approximately 40%) since 2010.42 In 2018, an estimated 37.9 million (32.7 million–44.0 million) people were living with HIV: 36.2 million (31.3 million–42.0 million) adults and 1.7 million (1.3 million–2.2 million) children (under the age of 15). Sixty-two percent (47%–75%) of adults over the age of 15 living with HIV had access to treatment, as did 54% (37%–73%) of children up to 14 years old. Importantly, since 2010, new HIV infections in children have decreased by 41%, from 280,000 (190,000–430,000) in 2010 to 160,000 (110,000–260,000) in 2018.42

In most cases, the diagnosis of vertical transmission of HIV is made in the first weeks of life: the viral genome is detected by polymerase chain reaction (PCR) in 93% of infected newborns at 15 days of life. The sensitivity and specificity of these tests increase to 96%–99% at the age of 1 month.43 Earlier diagnosis allows rapid implementation of ARV treatment in the acute stage of infection. In patients who have acquired HIV infection by vertical transmission, acute infection is not associated with the acute retroviral syndrome that occurs in 60% of newly infected adults.

As mentioned previously, HIV replicates in CD4+ T cells and progressively destroys the immune system. In children, since the immune system is not fully developed, immune suppression as well as AIDS develops faster than in adults. Consequently, in the first years of life, viral loads remain very high in plasma in the absence of ARV. The first symptoms of vertical HIV infection are usually nonspecific and develop during the first year of life. Opportunistic infections present in patients with severe immune suppression and, in most cases, have a worse evolution than in adults (such as pneumonia caused by Pneumocystis jirovecii).2

After vertical transmission, there are mainly two evolutionary patterns of progression of HIV infection: fast progressors (30%) and slow progressors (65%). Clinical manifestations during the first months of life will determine the prognosis. For example, HIV-associated encephalopathy and pneumonia caused by P. jirovecii are predictors of rapid progression, while chronic parotitis or lymphoid interstitial pneumonia is associated with slow progression. A third group of children (<5%) has also been described: very slow progressors, who remain with normal CD4+ T-cell counts and low viral loads for years, without any treatment.

2.1.4 Mitochondrial changes in HIV infection

Mitochondrial impairment was first associated with HIV in the 1990s,44 and in 2002, mtDNA depletion (a decrease in mtDNA copies) was described in mononuclear cells in the peripheral blood of HIV-infected patients who had never received ARV.45

HIV causes mitochondrial impairment by triggering apoptosis; many viral proteins are known to have the ability to induce apoptosis, as already mentioned above.46 HIV infection produces an increase in the levels of tumour necrosis factor α (TNFα), a cytokine produced in most inflammatory and immunological reactions, which is an apoptotic inducer. It occurs in lymphocytes as an anti-HIV response, and it also promotes HIV replication in T cells through activation of NF-κB transcription.47 In general, HIV-derived apoptosis affects infected and uninfected CD4+ T cells, contributing to leukopenia, typical of infected patients.48

Furthermore, and partly as a result of increased apoptosis, HIV-infected cells show an imbalance between oxidants and antioxidants.

Another HIV-associated toxic effect is Ca2+ overload and activation of nitric oxide synthase (NOS). This enzyme, which catalyses the formation of nitric oxide (NO) from L-arginine, can be expressed in neurons (nNOS or NOS-1), as well as by activated microglia (iNOS or NOS-2). Increases in NO can react with cellular superoxide to form peroxynitrite and promote various forms of neurodegenerative diseases.49 Tat affects both iNOS and nNOS, increasing Ca2+ by releasing intracellular deposits as well as through Ca2+ entry, induced by activation of N-methyl-D-aspartic acid (NMDA) receptors. The toxic effects of Ca2+-induced increases in Tat are mitigated by Ca2+ chelators, as well as inhibitors of Ca2+ absorption in mitochondria,28 supporting the role of Ca2+ dysregulation and Tat neurotoxicity. In addition to the Ca2+ channels of the plasma membrane, eukaryotic cells control Ca2+ homeostasis through Ca2+ channels, located in the ER, mitochondria and other organelles, through Ca2+ buffering proteins, and systems for extrusion and sequestration of Ca2+.50 Therefore, it is important to consider that Tat may also affect Ca2+ homeostasis in a manner independent of the NMDA receptor. In fact, Tat depletes both mitochondrial and ER Ca2+ by activating ryanodine receptors.51 Furthermore, Tat appears to increase Ca2+ by activating L-type channels.52 Thus, Tat appears to disrupt Ca2+ homeostasis by affecting both ER and other Ca2+-controlling organelles and Ca2+ regulatory systems located in the plasma membrane.53 Regardless of the mechanisms, all evidence points to altered Ca2+ homeostasis as one of the main mechanisms of Tat neurotoxicity. It should be mentioned that viral gp120 appears to modulate Ca2+ by a different mechanism. In fact, in contrast to Tat, gp120 increases Ca2+ mainly by mobilizing calcium deposits sensitive to inositol triphosphate (IP3).54 Because viral protein-induced mitochondrial toxicity has been repeatedly associated with disruption of Ca2+ homeostasis, it is not surprising that indirect ways to prevent Tat or gp120 toxicity include receptor-mediated blocking of Ca2+ entry. This includes the reduction of NMDA receptor activation by mild receptor antagonists, such as memantine, which protects neuronal function against gp120-mediated toxicity.55

NO has antiviral effects and increases within the cell in the presence of HIV, however, NO and ONOO− contribute to oxidative damage to cells and direct inhibition of mitochondrial respiration.56

Since changes in Ca2+ homeostasis, some of them above explained, have an influence in mitochondrial dynamics.57 Mitochondrial dynamics are also affected by HIV. Mitofusin 1 (Mfn1) and mitofusin 2 (Mfn2) are required to promote fusion of two neighbouring mitochondria. In contrast, mitochondrial fission is mediated by dynamin-related protein 1 (Drp1), which divides a mitochondrion into two. Fission helps by splitting healthy from defective mitochondria. Damaged mitochondria are then recycled or degraded through mitophagy; otherwise, apoptosis begins. In fact, both Tat and gp120 from HIV promote mitochondrial fragmentation (fission) and mitophagy alterations in human neurons.58

The mitochondrial dynamics of fusion and fission, estimated by Mfn2/β-actin and Drp1/β-actin contents, are decreased in the placenta of HIV-infected pregnant women, although there is a lack of information as to whether the newborn continues to present such alterations or not.59

Several HIV proteins activate key components of the transient permeability transition pore (PTP), leading to mitochondrial membrane depolarization. Furthermore, Tat can cause the translocation of Bim, a member of the pro-apoptotic family Bcl-2, from microtubules to mitochondria, where it induces PTP. Acute Ca2+ overload caused by Tat can also trigger the formation of PTP complexes. Furthermore, this protein can promote mitochondrially induced apoptosis.60

The use of ARVs minimizes HIV-related mitochondrial deterioration by decreasing viral load to undetectable levels. However, ARVs are also linked to side effects, as described in the following section. Therefore, mitochondrial toxicity is ultimately determined by both viral load and ARV exposure. In clinical practice, it is often difficult to differentiate whether mitochondrial abnormalities are related to HIV itself, or to ARVs.2

2.1.5 ARV treatment in the paediatric population and mitochondrial involvement

The different families of ARVs and the site of action are shown (Figure 3). Currently, ARV implementation has dramatically improved mortality and morbidity from HIV infection by decreasing viral load to undetectable levels and increasing CD4+ T-cell counts. In addition, simplification of therapeutic administration has led to better adherence to therapy. In developed countries, due to ARV administration, HIV infection is considered a chronic disease rather than a lethal infection.2 Importantly, several anti-HIV drugs may also lead to mitochondrial alterations at different levels, which has been summarized, including paediatric studies (Table 2).

Site of action of the different types of antiretroviral treatment within the host cell during HIV replication. Fusion and entrance inhibitors block the fusion and entrance of the virus in the host cell. Reverse transcriptase inhibitors block the retrotranscription from viral RNA to DNA. Integrase inhibitors inhibit the integration of proviral DNA into the cell nuclear genome. Protease inhibitors block the protease enzyme and therefore the assembly of the virions. Post-attachment inhibitors block the HIV from attaching the CCR5 and CXCR4 co-receptors of the host cell

TABLE 2. HIV antiretroviral agents and derived mitochondrial dysfunction including paediatric studies Antiretroviral family Characteristics Mechanism of action Mitochondrial dysfunction Clinical secondary effects Paediatric studies Nucleoside/nucleotide reverse transcriptase inhibitors (NRTI), e.g., ABC, FTC, 3TC, TDF and ZDV2, 10, 61

Antagonists of natural nucleosides: adenine, thymine, cytosine and guanine

Adequate resistance profile

Excellent tolerability

High bioavailability

Once daily treatment (except for ZDV)

Interfere with reverse transcriptase protein of HIV, which is necessary for viral replication

Inhibition of mtDNA polymerase gamma

mtDNA depletion

(a)

by means of a direct inhibition of DNA polymerase

(b)

By inducing errors during replication

(c)

By reducing exonuclease repair capacity

Decrease of mtDNA encoded proteins

General dysfunction of MRC

Direct inhibition of complexes of MRC (I–IV)

Decreased levels of ATP

ROS production

Decrease of mitochondrial membrane potential Δψm

Impairment of ADP/ATP translocase

Impairment of fatty acid oxidation62-65

NAD+/NADH impairment

Increased apoptosis

Overexpression of the Fas receptor66

Lactic acidosis, polyneuropathy, pancreatitis or lipodystrophy, among others

Complex and multifactorial mechanism: Genetic predisposition, dose and type of NRTI and duration of exposure

ZDV increases the risk of decreased blood mtDNA content which may be associated with altered mitochondrial fuel in infants67

Non-nucleoside reverse transcriptase inhibitors (NNRTI) e.g., EFV, ETR, NVP, RPV and DOR2, 10, 61

They do not need to compete with natural nucleosides

They are activated within the cell, directly interacting with viral reverse transcriptase and blocking its activity2, 68, 69

Stops HIV replication within cells by inhibiting the reverse transcriptase protein of HIV Mitochondrial dysfunction through bioenergetics stress (e.g., EFV has been associated to alterations in MRC in cultured glial cells and neurons70 NVP and EFV have been associated with hepatotoxicity In an urban area of Togo, the resistance of children with HIV type 1 treated with two NRTIs and one NNRTI showed mutations related to NNRTI class, with 100% mutations for EFV and NVP. The need to use PI is shown in most children treated with NNRTI71-73 Protease inhibitors (PI) e.g. LPV/rtv, ATV/rtv, DRV/rtv 2, 10, 61

Block maturation and activation of viral proteins (in an advanced stage of the viral cycle)

Metabolization by cytochrome P450, therefore, pharmacokinetic interaction with other drugs is common

Inhibit protease activity of HIV, a protein required for viral replication

Mitochondrial network fragmentation

Mitochondrial Ca2+ accumulation

Apoptosis

ROS production

Alterations of glucose and lipid metabolism2, 74

Peripheral neuropathy,75 lipodystrophy, metabolic syndrome, insulin resistance, diabetes, or cardiovascular risk Some studies report low tolerability, problems of adherence and development of resistance to treatment in children76 Integrase inhibitors (II) e.g., RAL, DTG and EVG2, 10, 61 Inhibit the integration of the viral genome into the nuclear genome of the cell Interfere with the viral enzyme integrase, which is needed to insert HIV genetic material into genetic material of human cells Expected cytotoxicity is low for most of them, as they suppress the viral cycle at very early stages Severe skin reactions, allergic reactions and liver disorders WHO recommends regimens based on DTG, once formulations suitable for children are widely implemented and available, as well as ongoing dosage and safety studies are completed; this will significantly ameliorate treatment outcomes76 Fusion inhibitors (FI) e.g., T-20 2, 10, 61

Block the fusion between HIV membrane and the target cell

Lim