Traditionally produced, natural archaeosomes are mixtures of glycolipids for which lot consistency and reproducibility of the formulations are difficult to control (16). To improve consistency of archaeosomes, simplified, semi-synthetic formulations have been developed, which consist of a single lipid called sulfated S-lactosylarchaeol (SLA), in which the sulfated saccharide group is covalently linked to the free sn-1 hydroxyl backbone of the archaeol core (14–16, 25). The chemical formula of SLA is 6’-sulfate-β-D-Galp-(1,4)-β-D-Glcp-(1,1)-archaeol (14, 15). SLA ensures consistency combined with an easy synthesis method and therefore reduced production cost (14).

Similar to traditional archaeosomes, SLA archaeosomes are capable of eliciting long-term cellular and humoral responses against various antigens and maintain a good safety profile (14–16, 26). They have been shown to be well tolerated without inducing adverse reactogenicity in mice at doses up to 10 times higher than those typically used in vaccines (14). Their adjuvant effect is comparable to or greater than that of widely used adjuvants, such as aluminum hydroxide, TLR3, TLR4 and TLR9 agonists, oil-in-water or water-in-oil formulations (11). SLA preparations cause intense local secretion of cytokines, retention of the antigen and immune cells at the vaccination site, and strong antigen uptake by APCs and other cells of the immune system, such as neutrophils and monocytes (14, 16). They are preferentially taken up by macrophages (16).

In addition to SLA archaeosomes, other semi-synthetic archaeosomes have also been tested in earlier studies as carriers and adjuvants for entrapped antigens (27). Semi-synthetic archaeosomes based on archaeol from Halobacterium salinarum to which various sugar moieties were grafted showed various types of immune responses to entrapped antigens. Immunization of mice with OVA entrapped into an archaetidylglycerophosphate-O-methyl archaeosome purified from Haloferax volcanii to which 45% gentiotriosylarchaeol, mannotriosylarchaeol or maltotriosylarchaeol were added resulted in enhanced CD8⁺ T cell and decreased antibody responses, compared to the same archaeosome formulation without the triglycosylarchaeols. In contrast, when all three triglycosylarchaeols were added to the archaeosome, the antibody response was less affected, while the enhanced CD8⁺ T cell response was still observed (28).

Among the archaeosome formulations there are traditional encapsulated archaeosomes, in which the antigen is released from the interior of the carrier, and admixed archaeosomes, in which the antigens are not entrapped but mixed with empty archaeosomes (18). Due to the differences in antigen localization on admixed archaeosomes compared to encapsulated archaeosomes, it is hypothesized that admixed archaeosomes may induce alternative antigen presentation pathways to those induced by encapsulated archaeosomes (16). Despite these differences, both formulations have repeatedly been shown to elicit comparable humoral and cellular responses and to induce strong antigen and immune cell retention at the vaccination site, with strong local cytokine secretion (15, 16, 26, 29). Similarly, by comparing three different formulations, Jia et al. also found that entrapped and admixed archaeosomes induced similar antibody and T cell responses to the test antigens OVA and hepatitis B surface antigen (18). However, in a B16- OVA melanoma model, the encapsulated archaeosomes appeared to have a greater ability to activate CD8⁺ T cells by dendritic cells in vitro than admixed formulations, yet in an in vivo murine tumor model the activity of both formulations was comparable with respect to survival (16). On the other hand, in a Human papilloma virus-16 (HPV-16) model, admixed preparations containing plasmid DNA encoding L1, E6 and E7 of HPV-16 induced higher IFN-γ response than encapsulated archaeosomes, but showed comparable effects on tumor size and survival of the mice (13).

Recent studies have shown that when OVA-containing SLA archaeosomes were co-administered with the TLR3 agonist poly(I:C), antibody responses to OVA were significantly enhanced after intramuscular administration over those obtained when OVA was formulated with SLA archaeosome or poly(I:C) alone (30). Antigen-specific cytotoxic T cell activity and IFN-γ production were also synergistically increased by the SLA archaeosome – poly(I:C) combination. A similar synergistic effect on antigen-specific cytotoxicity and IFN-γ production was also seen when the TLR9 agonist CpG instead of poly(I:C) was combined with the SLA archaeosomes. A synergistic effect of CpG or poly(I:C) with SLA archaeosomes could also be observed in a murine B16-OVA tumor model, in which the combination of SLA archaeosomes with poly(I:C) significantly delayed tumor growth and prolonged the mouse survival time compared to OVA formulations with SLA archaeosomes or poly(I:C) alone (31).

Karimi et al. (13) developed 2 vaccine formulations against HPV-16, which accounts for approximately 55-60% of all cervical cancer cases. The encapsulated and admixed formulations of the natural archaeosomes contained plasmid DNA encoding the capsid-building protein L1, as well as the E6 and E7 oncoproteins. Immunization of mice with either archaeosome-L1/E6/E7 induced humoral and cellular CD4⁺ and CD8⁺ responses, with high levels of IFN-γ, IL-2, IL-4 and IL-10. However, the admixed archaeosomes stimulated higher concentrations of these cytokines compared to the encapsulated archaeosomes. In addition, administration of archaeosomes to mice was associated with increased apoptosis of tumor cells, which reached 40 to 60% of the PBS controls, while the plasmid DNA alone led to 15 to 20% apoptosis (13). Empty archaeosomes were not tested for this readout. Furthermore, archaeosomes reduced and delayed the growth of tumors, as by day 28, the average tumor volume in mice immunized with archaeosomes-L1/E6/E7 was only 15% compared to that of the control mice. Recombinant plasmid DNA alone and empty archaeosomes led to a tumor size reduction of approximately 65% and 30%, respectively. Finally, while all mice in the control group had died by day 42, all mice immunized with archaeosomes were still alive at that time point. The recombinant plasmid alone and the empty archaeosome groups showed a slight delay in mortality compared to the PBS control (13).

When admixed and encapsulated SLA archaeosomes containing the HCV H77 E1/E2 heterodimer were compared to other adjuvant formulations, the admixed vaccine provided the strongest humoral response (26). Moreover, the SLA-admixed formulation induced antibodies that strongly neutralized HCV pseudoparticles and prevented infection of Huh7.5 liver cells, whereas the encapsulated formulation did not. However, the encapsulated formulation provided the strongest CD4⁺ cellular response. Neither preparation induced CD8⁺ cell responses in this model (26).

When encapsulated and admixed SLA preparations, natural encapsulated archaeosomes, and the squalene-based adjuvant AddaVax, carrying the influenza virus haemagglutinin were compared to each other, all preparations were found to elicit humoral responses against the antigen (15). However, the encapsulated SLA archaeosomes and AddaVax-formulated haemagglutinin provided stronger protection against influenza virus than M. smithii archaeosomes and prevented weight loss in influenza virus-infected mice, which averaged 10% in the natural archaeosome vaccinated group. Maternal vaccination to protect pups was also assessed using SLA and AddaVax formulations. Six-week old female mice were vaccinated with either encapsulated or admixed archaeosomes once before pregnancy or once before and once during pregnancy. The anti-HA antibody titers in the blood of the pups were higher when the mothers had received two doses and were about 10 times lower than in the blood of the mothers. Pups born to mothers vaccinated with encapsulated SLA archaeosomes had lower titers than those that were born to mothers vaccinated with admixed SLA archaeosomes. After one week, the pups and mothers were infected with the influenza virus H1N1 strain PR8 and protection against the disease in pups and mothers was evaluated. Both immunization protocols with the admixed archaeosomes induced strong protection against virus-induced mortality in both mothers and pups, while a single maternal immunization with the encapsulated archaeosomes did not protect the pups (15). There appeared thus to be a correlation between protection against weight loss/death and antibody titers in the pups. These titers did not reach the threshold levels in the pups born to mothers vaccinated once with the encapsulated archaeosomes, while the mothers had antibody levels high enough to be protected.

In order to evaluate the immunomodulatory properties of SLA archaeosomes and to compare them to other adjuvants, such as TLR3, TLR4 and TLR9 agonists, water-in-oil and oil-in-water emulsions, and aluminum hydroxide, a preparation containing encapsulated recombinant HBV surface antigen was produced. The formulation was found to induce strong humoral and CD8⁺ T cell responses and to stimulate the expression of IL-6, G-CSF, GM-CSF, KC, MCP-1 and MIP-2. The SLA archaeosomes showed strong adjuvant properties, similar to or stronger than other adjuvants and induced the strongest cytotoxic response among the tested formulations (11).

Four encapsulated natural archaeosomes containing separately Gp100 and TRP-2, a mixture of coentrapped melanoma antigens and a mixture of archaeosomal carriers of both proteins were reported to stimulate CD8⁺ T cell responses and to provide protection against cancer. Dipeptide vaccines were found to induce better and longer protection than monopeptide vaccines (37). Subsequently, it was shown that the TRP-2 protein encapsulated in the SLA archaeosome induced a strong CD8⁺ T cell response in mice and strong protection against B16 melanoma, comparable to that induced by the natural archaeosome formulation (25).

In the melanoma B16-OVA model, archaeosome vaccines against OVA and checkpoint inhibitor therapy consisting of anti-PD-1 and anti-CTLA-4 antibodies were combined. All formulations tested, i.e. admixed and entrapped SLA archaeosomes and encapsulated natural archaeosomes presenting OVA, enhanced the effects of checkpoint inhibition therapy by increasing the survival rate of mice and OVA-CD8⁺ T cell production and by reducing tumor growth in comparison to checkpoint inhibition therapy alone (29).

A vaccine formulation against T. cruzi, the parasite that causes Chagas disease, based on archaeosomes composed of Halorubrum tebenquichense lipids in which dissolved parasite homogenates had been encapsulated was found to be effective in stimulating specific humoral responses. In addition, immunized mice infected with T. cruzi trypomastigotes showed low blood parasite titers and high viability compared to the groups of unvaccinated mice (32).

Perera et al. (35) prepared three vaccine formulations targeting Cathepsin B (SmCB), the most abundant cysteine protease in schistosomula and adult S. mansoni by combining the antigen with SLA archaeosomes or AddaWax™. Both vaccines provided 40% protection and reduced adult worm burden, liver and intestinal egg counts, with the SLA archaeosome formulation reaching 60.5% reduction and AddaWax 86.8%. Both formulations also induced humoral and cellular immune responses. The SLA archaeosome formation increased IL-3 and Th1 cytokine levels, while AddaWax stimulated Th17 and Th2 cytokine production.

In a vaccine formulation against L. monocytogenes, secretory proteins obtained from the culture supernatant of this pathogen were encapsulated in archaeosomes composed of H. salinarum lipids. Vaccination with this formulation resulted in increased production of the Th1 cytokines IFN-γ and IL-12, induction of humoral and cytotoxic T cell responses with the generation of T cell memory, without inducing IL-4. One week after infection with L. monocytogenes, a significant reduction in bacterial burden in the liver and spleen was noticed in the vaccinated animals compared to the controls, and after further 4 weeks, all vaccinated mice, unlike the control group, were protected against death (33).

A vaccine formulation against M. tuberculosis based on natural archaeosomes encapsulated with the Rv3619c antigen was compared to the most commonly used vaccine, the Bacille Calmette-Guérin, and shown to provide stronger IFN-γ and IL-12 responses, as well as proliferation of Rv3619-specific T cells, higher concentrations of co-stimulatory markers, stronger effector memory T cell responses and significantly greater reduction in bacterial load in the lungs and spleen after M. tuberculosis challenge (34).

The immune responses against the cholera toxin subunit B were also compared between preparations using liposomal and archaeosomal carriers. Archaeosomal vaccines prepared from M. smithii lipids appeared to induce significantly higher humoral responses compared to liposomes and similar responses compared to Freund’s complete adjuvant. Moreover, archaeosomal vaccines only required two administrations to achieve maximum antibody titers in mice (38).

In the case of SIV, fragments of Gag, Tat, Rev and Nef1 were expressed on the surface of GVNPs. Recombinant GVNPs presenting fragments of Gag (17, 168 or 235 amino acids-long peptides) induced IgG production and long-term immune memory. Anti-Gag antibodies were detected 120 days after a booster injection (36, 39). The fragments of Tat (50 amino acids), Rev (81 amino acids) and Nef1 (214 amino acids) were also expressed on GVNP. The recombinant GVNPs induced immune responses at levels similar to those obtained by Gag-GVNPs (36). During immunization with recombinant Tat-GVNP, Rev-GVNP and Nef1-GVNP vaccines, IL-10, IL-12 and IL-18 were produced (36, 39). The strongest immune response was shown after immunization with the recombinant Tat-GVNP vaccine (39).

To stimulate immunity against S. enterica serovar Typhimurium, GVNPs containing 2 fragments of SopB (100 amino acid-long SopB4 and 165 amino acid-long SopB5) have been prepared. Mice primed with an attenuated vaccine against S. typhimurium, followed by boosting with 100 µg of SopB4-GVNP or SopB5-GVNP 7 and 14 days later, developed a strong and long-lasting immune response against SopB and increased levels of the pro-inflammatory cytokines IFN-γ, IL-2 and IL-9, as well as GM-CSF in the serum. One week after oral challenge with 10⁷ colony-forming units of virulent Salmonella the bacterial loads in mesenteric lymph nodes, liver and spleen were at least 2 orders of magnitude lower than in mice boosted with non-recombinant GVNP. CD4⁺ T cell levels in the spleen were increased 2- to 4-fold compared to non-recombinant GVNP-boosted controls. Moreover, mice boosted with SopB5-GVNP survived 3-5 days longer than those boosted with non-recombinant GVNP (36, 42).

For the expression of C. trachomatis antigens on the GVNP surface, protein fragments of the major outer membrane protein MOMP (48 and 69 amino acids), the outer membrane B complex OcmB (162 and 144 amino acids) and the outer membrane polymorphic protein PompD (173 and 222 amino acids) have been used (36, 39). Immunostaining confirmed that GVNPs presenting the C. trachomatis antigens were absorbed by human foreskin fibroblast cells, in which they were gradually disintegrated and finally exposed on the surface of the fibroblasts (36, 39). Recombinant GVNPs were shown to engage TLR-4 and TLR-5 and to stimulate the production of TNF-α, IL-1β, IL-6 and IL-12 (36, 42).

The highly conserved 15 amino acid-long fragment of the P. falciparum enolase (14 out of the 15 residues are identical between the P. falciparum and the Plasmodium yoelii peptide) was also expressed on GVNPs. Mice immunized with the recombinant GNVPs and then challenged with the murine Plasmodium yoelii parasite showed a lower degree of parasitemia and a longer survival rate compared to mice immunized with non-recombinant GVNPs (39). As P. falciparum does not infect mice, P. yoelli is routinely used for studies in mouse models (43). Likewise, the P. falciparum circumsporozoite protein was successfully expressed on the GVNP surface. However, the induced immune responses to this antigen have not yet been assessed (36, 44).