3.1. Vaccines Based on the Induction of Neutralizing Antibodies 1986–2003
Research into developing a vaccine against HIV began in 1986. The first trials were based on the assumption that the activation of neutralizing antibodies would be sufficient to protect against HIV infection. The first HIV vaccines were developed on the basis of recombinant DNA gene modification technology, similar to the hepatitis B vaccine licensed in 1986 [
24,
30,
31]. Between 1988 and 2008, various HIV vaccine designs were developed based on the viral envelope glycoproteins (mainly gp120 and gp160), which are responsible for binding the virus to target cells and serve as the primary targets for neutralizing antibodies. Among the first to begin testing VaxSyn (1988-2003) is the recombinant enveloped glycoprotein (rgpl60) created in the baculovirus-insect cell system. Subsequently, numerous envelope proteins were assessed in an additional 35 studies.
Collectively, these vaccine constructs induced binding and neutralizing antibodies, which were long-lasting, and also stimulated CD4+ T cell responses, but did not produce overt cytotoxic lymphocytes (CD 8+ CTL) responses [
32]. The induction of CTL responses, which was already perceived as a desirable characteristic of an HIV vaccine, could be achieved through the use of live vectors, especially poxvirus vectors [
33]. Live vectors were very appealing because they could present antigens to the immune system simulating natural infection. In 1986, two different groups reported the expression of the HIV envelope gene in vaccinia vectors [
34,
35].
One of them was HIVAC-1e (1988), a recombinant vaccinia virus designed to express HIV gp160. In the majority of subjects, T-cell responses were transient and no antibodies to HIV were detected. However, the appearance of a sustained T-cell response in a small proportion of subjects led researchers to the idea that antibody responses would be enhanced by priming with a recombinant vaccinia vector expressing the HIV-1 envelope and then with an envelope booster protein [
36]. VaxSyn + HIVAC-1e (1991) – combined approach. This trial was started with HIVAC-1e and then boosted with VaxSyn. The prime-boosting approach enhanced both the humoral and cellular immune response and induced neutralizing antibodies [
37,
38]. At this stage, one of the features of the vector used became apparent: there was a marked decrease in the immunogenicity of the vector in individuals previously vaccinated against smallpox [
31,
36].
Of course, the further refusal to use the cowpox vector was largely influenced by the fact that the introduction of such a vaccine is contraindicated for people who already have immunosuppression, which means that its use is unsafe in groups at high risk of HIV infection [
39,
40]. It was decided to use non-replicating poxvirus vectors in the early 1990s, based on two models, namely, a highly attenuated strain of vaccinia virus (NYVAC) or an avian poxvirus, canarypox (ALVAC), that is not able to replicate in mammalian cells [
33,
41].ALVAC-HIV Vector Vaccine, an HIV vaccine with the ALVAC vector, vCP125, expressing gp160 has been tested alone or as a combination primary booster with the adjuvant gp160 subunit. The vaccine significantly stimulated the response of neutralizing antibodies to protein administration and induced CTL activity [
42]. Other ALVAC vectors (vCP205, vCP300, vCP1433, vCP1452, and vCP1521) have been designed not only to express the HIV envelope, but also to express gag and other HIV genes to induce a broader cell-mediated immune response. It was vCP1521 that was then used in the most successful trial of RV144. [
43,
44]. Vax003-Vax004 ‒ (1994-) vaccines based on bivalent recombinant gp120 and alum provided protection to chimpanzees after HIV infection and were safe and immunogenic in phase 1/2 of human clinical trials [
45,
46]. As a result, high levels of both neutralizing antibodies (nAbs) binding (in VAX003 and VAX004) and antibody-dependent cellular inhibition of the virus (in VAX004) were observed, but both candidates were not effective in preventing HIV infection due to their narrow specificity [
45,
46].
3.1.1. Direct Cytotoxicity
Vaccines aimed at stimulating CTLs will cause pre-sensitized T-killers to immediately recognize HIV-infected cells and destroy them by the mechanism of direct cytotoxicity, preventing further replication of the virus. The mechanism of the cytotoxic effect is described below and in
Figure 2.
CTLs secrete perforin proteins, which are incorporated into the cell membrane and oligomerize therein to form a pore. Next, CTLs secrete granzyme B, which enters the cell through the previously created pore. The cell cytosol contains procaspase 7, which consists of prodomain, large subunit and small subunit. Granzyme B proteolytically detaches the prodomain procaspase from the large subunit and also detaches the large subunit from the small subunit. Two executioner procaspase 7 molecules are required to form an "active" caspase. As a result, the two large and two small procaspase subunits join together to form a heterotetramer, which is active executioner caspase 7. Executioner caspase 7 starts proteolytically destroying the cell proteins leading to cell death.
It is important to note that proapoptotic proteins such as Bax and Bak are constantly present in the cell. Moreover, Bak is built into the outer mitochondrial membrane and Bax is freely present in the cell cytoplasm. The Bax molecule oligomerizes with Bak on the outer mitochondrial membrane forming a large pore in it through which cytochrome c escapes from the intermembrane space into the cell cytosol and triggers apoptosis. However, in the normal cell cytoplasm there are antiapoptotic proteins such as Bcl2 and BCl-XL, which bind to Bax and Bak preventing their oligomerization. Granzyme B proteolytically destroys Bcl2 and BCl-XL proteins and thus creates conditions for cytochrome c release into the cell cytosol [
47].
The cell cytosol contains Apoptotic protease activating factor 1 (APAF1) which has a caspase recruitment domain (CARD) that binds to the active center of APAF1 itself and blocks it. But cytochrome c, once in the cytosol of the cell, pushes CARD from the active center of APAF1 and binds itself to it instead of CARD. Further, about 7 APAF1 molecules with active centers bound by cytochrome c bind to each other using their own CARD domains, forming the apoptosome. Further, CARDs associated with prodomain of seven initiator procaspase 9 interact with the CARDs that are part of the apoptosome. As a result of this interaction, the small subunit detaches from the large subunit of procaspase 9, and the large subunit also detaches from the prodomain. From there, a heterotetramer consisting of two large and two small subunits is assembled with the formation of the "active" initiator caspase 9, which itself will start proteolytically activating Caspase 3 and Caspase 6, which are the executioner caspases. Caspase 3 and Caspase 6 produce proteolysis of the cell proteins, which leads to cell death.
Perforins form a pore in the plasma membrane. Through this pore Granzyme B enter the cell. Granzyme B activates procaspase 7. Caspase 7 starts proteolytically destroying the cell proteins leading to cell death. Antiapoptotic proteins Bcl2 and BCl-XL bind to Bax and Bak preventing their oligomerization. Granzyme B proteolytically destroys Bcl2 and BCl-XL proteins and thus creates conditions for cytochrome c release into the cell cytosol. Cytochrome c binds to the APAF1 molecule. Further, about 7 APAF1 molecules with active centers bound by cytochrome c bind to each other using their own CARD domains, forming the apoptosome. Apoptosome activates initiator caspase 9, which in turn activates executioner caspases 3 and 6. Caspase 3 and Caspase 6 produce proteolysis of the cell proteins, which leads to cell death [
48].
3.1.2. Chemokine-Mediated HIV Suppression
Chemokines based vaccines, especially CCL4 based vaccines will lead to CCR5 desensitization. This would result in HIV having no entry gate for T-helper cells to be infected. Although this type of vaccines will not protect somatic cells from infection, they will prevent the development of AIDS. The mechanism of CCR5 desensitization is described below and in
Figure 3.
Being a ligand for CCR5, CCL4 in high amount can overstimulate CCR5, that will lead to the phosphorylation CCR5 by CD28-induced G-protein-coupled receptor kinase 2 (GRK2). b-arrestin binds to the phosphorylated inorganic phosphate molecules. Then, b-arrestin changes its conformation and releases a site for binding to the Adaptor protein complex 2 (AP-2 complex). However, this is not enough for the AP-2 complex to join the b-arrestin. The AP-2 complex consists of four proteins called adaptins: two large adaptins (α and β), a medium adaptin (μ), and a small adaptin (σ). The α and β adaptins have so-called ’ears’ that form the appendage domain. The remaining part of α and β adaptins, as well as μ and σ adaptins together form the core domain. In general, the AP-2 complex has one core domain and two appendage domains. The core domain binds to synaptotagmin ½, which is embedded in the cell membrane. It is important to know that this interaction is stable only if PIP2 is present in the cell membrane with which the AP-2 complex also binds. After the previously mentioned interactions, the AP-2 complex changes its conformation and is activated, which leads to the interaction of μ-adaptin with b-arrestin.
The cell cytosol contains clathrin protein, which consists of a heavy chain comprising three domains: proximal, distal, and N-terminal, and a light chain bound to the proximal domain of the heavy chain. Clathrin is not found as a monomer in the cell cytosol. The three clathrin molecules bind to each other using the proximal heavy chain domain to form claritine triskelion, which is present in the cell cytosol. The claritine triskelion then binds to the appendage domains of the AP-2 complex, which will lead to polymerization of the triskelion, forming a closed three-dimensional network resembling a soccer ball. As a result of these processes, the plasma membrane is woven inward to form the Clatrin coated vesicle bud [
49,
50].
The cell requires Ca2+ for normal clathrin-dependent endocytosis. Ca2+ ions bind to calmodulin, forming the calmodulin/4Ca2+ complex. In turn, the calmodulin/4Ca2+ complex binds to calcineurin activating it. In the cytoplasm of the cell there is a protein dynamin 1 with which an inorganic phosphate molecule is bound and inhibiting it. Calcineurin dephosphorylates dynamin 1 and thereby activates it. Once activated, dynamin 1 binds to the "neck" of the Clatrin coated vesicle bud and oligomerizes, covering it completely. dynamin 1 has GTPase activity and having enveloped the "neck" of the clathrin coated vesicle bud it begins to hydrolyze GTP to GDP and inorganic phosphate, releasing energy which is used to enhance the torsion of oligomerized dynamin 1 around the "neck", narrowing and eventually cutting it, forming the clathrin coated endocytic vesicle. In the process of cutting the neck, endophilin also plays an important role, which is normally in the homodimerized state and also binds to the "neck" and then polymerizes around it in the form of a ring. This happens because endophilin has "+" charged amino acids, which are attracted to the "-" charged heads of the cell membrane phospholipids.
Phosphorylation of CCR5 by GRK2. B-arrestin binds with the inorganic phosphate molecules of CCR5. Then AP-2 complex binds with B-arrestin. To form a stable complex, AP-2 interacts with SYT and PIP2, embedded in the plasma membrane. Three clatrin molecules bind to each other and form clatrin triskelion. The clatrin triskelion then binds to the AP-2 complex. As a result of these processes, the plasma membrane is woven inward to form the Clatrin coated vesicle bud. Ca
2+ ions bind to calmodulin, forming the calmodulin/4Ca
2+ complex. The calmodulin/4Ca
2+ complex binds to calcineurin and activates it. Calcineurin dephosphorylates dynamin 1 and thereby activates it. Once activated, dynamin 1 binds to the "neck" of the Clatrin coated vesicle bud and oligomerizes, covering it completely. Having enveloped the "neck" of the clathrin coated vesicle bud dynamin 1 begins to hydrolyze GTP to GDP and inorganic phosphate, releasing energy which is used to enhance the torsion of oligomerized dynamin 1 around the "neck", narrowing and eventually cutting it, forming the clathrin coated endocytic vesicle. Endophilin also plays an important role in forming the clathrin coated endocytic vesicle [
51,
52].
3.2. Stimulation of T-Cell Immune (1995–2007)
Further study of the problem and analysis of the unsatisfactory results of previous years of research prompted scientists to use the T-cell immune response. In the corresponding period of time, the degree of knowledge of the virus also increased. The leading role of CD8+ T-cells in the containment of virus replication at one of the stages of its development in the body has been proven.
Subsequent studies have been conducted with increasing reliance specifically (and often, only) on the T-response [
53,
54]. Candidate model vaccine vectors used for T-cell vaccines were live recombinant viral vectors, mainly pox and adenovirus vectors (particularly replication-defective adenovirus 5 (Ad5)), as well as DNA vaccines [
55].
STEP trial - HVTN 502 – HVTN 503 (Phambili) - recombinant replication-defective adenovirus type 5 (Ad5) vector vaccines: MRKAd5 HIV-1 gag/pol/nef clade B vaccine. Both trials were primarily aimed at reducing post-infection viral load, but were interrupted early because those who were vaccinated had an increased risk of infection [
31,
56,
57].
HVTN 505 - DNA vaccine expressing clade B gag/pol/nef and env from branches A, B and C, which was used to prime rAd5 multiclade vaccines. The goal was to activate both arms of immunity. An increased risk of infection was not observed, but the vaccination schedule was not shown to be effective [
58,
59,
60].
RV-144 - the study design consisted of a recombinant canarypox vector vaccine, ALVAC-HIV (vCP1521), expressing Env (branch E), group specific antigen (Gag) (branch B) and protease (Pro) (branch B), and AIDSVAX B vaccines/E with an alum adjuvant and a bivalent subunit vaccine based on HIV glycoprotein 120 (gp120). The vaccine showed significant effectiveness in protecting against HIV infection - 31.2% in 42 months after vaccination. Vaccine-induced reactions included binding of IgG antibodies to HIV Env variable loops 1 and 2 (V1V2) and antibody-dependent cellular cytotoxicity (ADCC) in low-IgA vaccine recipients. The study showed that the size and multifunctionality of Env-specific CD4+ T cells play a role in reducing the risk of HIV infection [
61,
62]. RV305- RV306- Evaluation of the immune response generated in recent booster vaccine recipients compared to RV144 vaccine recipients. The assumption was that late boosters will elicit a sustained immune response [
62,
63].
HVTN 097 - the schedule included 2 doses of ALVAC-HIV (vCP1521) HIV vaccine followed by 2 booster doses of AIDSVAX B/E. Tetanus immunization was included in the schedule to avoid cross-reaction to the vaccine [
64]. The primary booster schedule induced mainly Env-specific CD4+ T cell responses at significantly higher levels compared to RV144 vaccine recipients (RV144 = 36.4%; HVTN 097 = 51.9%). IgG antibodies recognizing the V1V2 region and IgG3 binding antibody responses to gp120 and V1V2 antigens were also significantly higher among HVTN 097 vaccine recipients compared to RV144 recipients. ADCC antibody responses were also higher in HVTN 097 than in RV144 [
64].
HVTN 100 is an ALVAC-HIV vector (vCP2438) expressing HIV gp120 Env (C branch of ZM96), gp41 Env, Gag and Pro (all B branches), and the bivalent protein vaccine with gp120 subtype C and MF59 adjuvant. The vaccine induced a greater frequency of IgG3 responses to Env gp120, significantly higher CD4+ T cell responses, and gp120 binding antibody responses compared to the RV144 schedule [
65]. HVTN 100 exceeded levels that were modeled as necessary for protection.
HVTN 702 - evaluation of the efficacy, safety and tolerability of ALVAC-HIV (vCP2438) plus a bivalent gp120/MF59 subtype C vaccination regimen with primary booster. The vaccination schedule used in HVTN 702 has been modified to improve the efficiency and robustness of immune responses compared to RV144. The study was terminated due to the low efficacy of the vaccine [
65,
66].
The RV144 trial showed the best result ever for an HIV vaccine development approach, subsequent studies based on RV144 designed to enhance efficacy came to mixed conclusions. However, in all subsequent studies, special attention is paid to the endemicity of the HIV isolate (
Table 2) [
67].
3.3. Mosaic HIV Vaccines
Polyvalent HIV-1 Gag, Pol and Env mosaic immunogens expressed by replication-incompetent adenovirus serotype 26 (Ad26) vector and optimized to provide maximum coverage of potential T-cell epitopes. The mosaic HIV vaccine causes the formation of multifunctional antibodies - neutralizing, non-neutralizing and binding. This approach solves the problem of narrow specificity of the candidate vaccine and affects different circulating virus isolates [
68]. However, large clinical studies of mosaic HIV vaccines have not shown their effectiveness, although some of these studies are currently underway (
Table 3).
3.4. mRNA HIV Vaccine
mRNA platforms are designed to deliver a piece of genetic material that instructs the body to make a protein fragment of a target pathogen (such as HIV), which the immune system will hopefully recognize and mount a defense against [
69].
mRNA technology may be an important step forward to speed identification of the right target antigens for a protective response, but it alone does not address other challenges associated with HIV vaccine development, such as what antigen will be right. B-cells play an important role in the action of these vaccines, which implement their immune response through the BCR. The mechanism of mRNA HIV vaccine is discussed below and in
Figure 4.
B cell antigen receptor (BCR) is embedded in the membrane of B cells. The BCR is composed of membrane immunoglobulin (mIg) molecules and associated Igα/Igβ (CD79a/CD79b) heterodimers (α/β). The mIg subunits bind antigen, resulting in receptor aggregation, while the α/β subunits transduce signals to the cell interior. BCR aggregation rapidly activates the Src family of Lyn kinases. The cytoplasmic part of CD79a/b molecules contains immunoreceptor tyrosine-based activation motifs (ITAMs). The Lyn protein, which phosphorylates CD79a/b by the OH groups of the two tyrosine residues that make up the ITAMs, is embedded in the membrane of B cells. The Syk protein binds to the phosphorylated tyrosine residues using SRC homology 2 (SH2) domain, after which it phosphorylates itself by the OH groups of tyrosine (autophosphorylation), leading to its own activation. It is important to note that the Lyn protein can also activate Syk by phosphorylation [
70].
The adaptor protein BLNK (B-cell linker protein) binds to the phosphorylated tyrosine residues of CD79a/b using the SH2 domain. Next, Syk phosphorylates the BLNK molecule by the tyrosine OH-group, facilitating the binding of phospholipase Cγ2 (PLCγ2) to it. But this is not enough to activate PLCγ2. The fact is that the Bruton’s tyrosine kinase (BTK) molecule binds to the phosphorylated tyrosine residues of BLNK protein using its SH2 domain, after which BTK binds to phosphatidylinositol (3,4,5)-trisphosphate (PIP3), which is a phospholipid of the inner layer of the cell membrane using pleckstrin homology (PH) domain. The interaction of the PH domain with PIP3 leads to full activation of BTK protein. As a result, BTK phosphorylates and thereby activates PLCγ2. PLCγ2 begins to hydrolyze PIP2 (phosphotidylinositol diphosphate), which is a typical representative of cell membrane phospholipids, to inositol triphosphate (phospholipid head) and diacylglycerol (phospholipid tail). DAG remains in the cell membrane. IP3 binds to IP3 receptors located on the ER membrane and stimulates them. It is important to note that IP3 receptors are Ca
2+ channels and that after interaction with IP3 they open, promoting the release of large amounts of Ca
2+ ions from the ER into the cell cytosol along a concentration gradient. Ca
2+ ions and DAG together activate PKCβ (Proteinkinase Cβ). In turn, Ca
2+ binds to calmodulin, forming the calmodulin/4Ca
2+ complex. In the quiescent state, the cell cytoplasm contains the molecule Nuclear factor of activated T-cells (NFAT), which is in a phosphorylated state and is a transcription factor. As long as NFAT is in the phosphorylated state it cannot pass into the nucleoplasm. The calmodulin/4Ca
2+ complex binds to the protein calcineurine resulting in its activation. Calcineurine, a protein phosphatase, dephosphorylates NFAT and thereby changes its conformation, leading to the availability of nuclear localization signals (NLS) and transport of this transcription factor into the nucleoplasm [
71].
The cell cytosol contains a caspase recruitment domain family member 11 (CARD11 or CARMA1), which is in the inactive, "closed" state. PKCβ phosphorylates CARMA1 by the serine/threonine OH-group and thereby promotes a change in the conformation of this molecule and its transition from the "closed" to the "open" state. There is a BCL10-MALT1 heterodimer (B cell CLL/lymphoma 10-MALT1 paracaspase heterodimer) in the cytoplasm. After CARMA1 is in the "open" state, it is bound to BCL10 molecule through caspase recruitment domain (CARD)-CARD domain interactions, which is a part of BCL10-MALT1 heterodimer, forming CBM signalosome complex (The caspase recruitment domain family member 11 (CARD11 or CARMA1)-B cell CLL/lymphoma 10 (BCL10)-MALT1 paracaspase (MALT1)). Once this complex is formed, the MALT1 molecule begins to oligomerize, leading to the recruitment of tumor necrosis factor receptor-associated factor 6 (TRAF6) protein, which is an E3 ubiquitin ligase. The cell cytoplasm contains the molecule IKK (IκB kinase), which consists of three subunits: IKK-α, IKK-β, and IKK-γ (NEMO-NF-kappa-B essential modulator). TRAF6 modifies IKKγ with Lys63-polyUb chains, recruiting the kinase TAK1 through its ubiquitin-binding cofactors TAB2 and TAB3. TAK1 then phosphorylates IKK-β, leading to activation of the IKK molecule. It is important to mention that the B-cell cytosol contains the NF-κb molecule, which consists of three subunits: IkBα (Inhibitor of Nuclear Factor (NF)-Κb α isoform), p50, and RelA. In the resting state, IkBα masks the NLS (nuclear localization sequence) of p50 and RelA proteins, preventing transport of these subunits as a heterodimer from the cell cytosol into the nucleoplasm. However, upon activation, IKK starts phosphorylating the IkBα subunit by the OH-groups of the serine/threonine residues. Further, the enzyme E3 ubiquitin ligase binds to the phosphorylated serine/threonine residues, promoting ubiquitination and further proteosomal degradation of IkBα. As a result, NLS is unmasked and the p50/RelA heterodimer is transferred from the cell cytoplasm into the nucleoplasm [
72].
The B-cell membrane contains a CD19 protein, the intracellular part of which is phosphorylated by Lyn protein by the OH-group of tyrosine. The phosphorylated tyrosine residues of CD19 are bound to the BCAP (B cell adaptor molecule for PI3K) molecule, which recruits the p85 domain of PI3K protein to itself. This interaction leads to the activation of the p110 domain of PI3K, which with its catalytic activity phosphorylates phosphatidylinositol-4,5-bisphosphate (PIP2) to produce and accumulate phosphatidylinositol-3,4,5-trisphosphate (PIP3). On the one hand, PIP3 is important for binding the BTK molecule to the inner surface of the cell membrane, and on the other hand, it is important for the activation of the Akt/mTOR pathway.
BCR aggregation rapidly activates the Src family of Lyn kinases. The Lyn protein phosphorylates CD79a/b by the OH groups of the two tyrosine residues. The Syk protein binds to the phosphorylated tyrosine residues using SH2 domain, after which it phosphorylates itself by the OH groups of tyrosine (autophosphorylation), leading to its own activation. The adaptor protein BLNK binds to the phosphorylated tyrosine residues of CD79a/b using the SH2 domain. Syk phosphorylates the BLNK molecule, facilitating the binding of PLCγ2 to it. BTK molecule binds to the phosphorylated tyrosine residues of BLNK protein using its SH2 domain, after which BTK binds to PIP3 using PH domain. As a result, BTK phosphorylates and thereby activates PLCγ2. PLCγ2 begins to hydrolyze PIP2 to IP3 and DAG. IP3 contributes to the output of Ca
2+ form the ER. Ca
2+ ions and DAG together activate PKCβ. Ca
2+ binds to calmodulin, forming the calmodulin/4Ca
2+ complex. The calmodulin/4Ca
2+ complex binds to the protein calcineurine resulting in its activation. Calcineurine, a protein phosphatase, dephosphorylates NFAT and thereby leads to the transport of this transcription factor into the nucleoplasm. PKCβ phosphorylates CARMA1. CARMA1 binds to BCL-10 molecule, which is a part of BCL-10-MALT1 heterodimer, forming CBM signalosome complex. Once this complex is formed, the MALT1 molecule begins to oligomerize, leading to the recruitment of TRAF6 protein. TRAF6 modifies IKKγ with Lys63-polyUb chains, recruiting the kinase TAK1 through its ubiquitin-binding cofactors TAB2 and TAB3. TAK1 then phosphorylates IKK-β, leading to activation of the IKK molecule. IKK starts phosphorylating the IkBα subunit of NF-κb. Further, the enzyme E3 ubiquitin ligase binds to the phosphate groups of IkBα, promoting ubiquitination and further proteosomal degradation of IkBα. As a result, NLS is unmasked and the p50/RelA heterodimer is transferred from the cell cytoplasm into the nucleoplasm. Intracellular part of CD19 protein is phosphorylated by Lyn. With that phosphate groups BCAP protein binds. BCAP recruits PI3K, which phosphorylates PIP2 to PIP3. On the one hand, PIP3 is important for binding the BTK molecule to the inner surface of the cell membrane, and on the other hand, it is important for the activation of the Akt/mTOR pathway [
70,
73].
The
Table 4 below summarizes ongoing vaccine trials on the mRNA platform. The difference is in the antigens, the viral proteins that the immune system targets.