Fig. 5. Mechanism of Lipid A. When TLR4 is activated by lipid A on the liposome vaccine,and binds to myeloid differentiation factor 2 (MD2) to form A complex, the signal is transduced to the Toll/IL-1R1 homologous region (TIR) domain, followed by two My88: myeloid differentiation factor 88 (My 88)-dependent and My88-independent or TIR-domain-containing adaptor (TRIF)-dependent pathways. My 88-dependent pathway: TLR binds to the TIR domain containing adaptor protein (TIRAP) of MyD88 and activates MyD88, forming an active TLR4/MyD88 complex that activates IL-1R-associated kinase (IRAK), which eventually causes nuclear factor-κB (NF-κB) into the nucleus and triggers subsequent reactions. My88-independent pathway or TRIF-dependent pathway: complex’s signal is transmitted to TIRF, the interferon transcription is regulated by activating TRIF to further activate the NF-κB. Impelling the body to produce inflammation-related factors.
Optimization of lipid A. It is difficult to obtain structurally homogeneous lipid A in varied forms from biological sources. Moreover, lipid A from natural sources may be contaminated by other components of the bacterial cell wall. Chemical synthesis as an effective tool can perfectly solve this problem. Previous review, from the perspective of chemical synthesis of each part in detail elaborated the lipid A chemical synthesis process and related derivatives, such as containing saturated phospholipids and double phosphate group of natural lipids, containing unsaturated double phosphate groups of lipid A, containing carboxylic acid derivatives, single phosphorus acyl lipid A and the phosphorylation of lipid A [63].
As the premise, various positions in the disaccharide backbone of lipid A with the same or different acyl lipid chain, thus lipid A and its analogues synthesis can be divided into two basic strategies. One is to build an asymmetric distribution of lipid chains, that is to construct general disaccharides skeleton and install the same length or different length of lipid acyl chain in a particular location. Another strategy is to synthesize a monosaccharide with acyl lipid chain and then stitch them into a disaccharide structure through glycosylation reaction. However, the latter is often hard to implement due to the steric hindrance. Shimoyama et al. synthesized Alcaligenes lipid A (3+3) (Fig. 4B) by the introduction of benzyl-protected β-hydroxy fatty acid into the disaccharide with an appropriate acylation reagent and repeated similar operations [58].
Jiang et al. provided a synthesis method of pentaerythritol lipid A (PET-LA) (Fig. 4B) which contains an amine-substituted pentaerythritol instead of a glucosamine unit found in natural lipid A [59].
After a period of research, Kawther et al. proved that PET-LA had powerful adjuvant properties in immunogenic mouse tumor models by loading CD40 and CD86 into polylactic acid-glycolic acid (PLGA) particles to stimulates the upregulation of CD40 and CD86 [66]. Compared to the glucosamine structure, PET can keep the anomeric phosphate group well, and because of its simple structure, the synthesis process is simplified and the molecular stability is also improved. Niels et al. studied the immune activity of lipid A analogue CRX-527 conjugated with long peptides in different ways to determine whether it had the potential to be vaccinated. CRX-527 (Fig. 4B), one of aminoalkyl glucosamine 4-phosphates (AGPs), which contains monosaccharide with a long-chain acyl group [60]. In vitro , ester-linked compounds performed better in inducing DC maturation, while ammonia-linked compounds showed greater activity in presentation. However, according to the data studied in vivo , it seemed that a conjugate of mounting the free carboxyl ligand at the N-terminus of the model peptide antigen can recruit enough protective immune response to effectively kill cancer cells and exert a powerful immune function for the development of anti-cancer.
Glucopyranosyl lipid adjuvant (GLA) is a synthetic non-toxic derivative of lipid A (Fig. 4B). When it is used as adjuvant and co-delivery with tumor antigen, it plays the role of TLR4 agonist. GLA targeting human DC and peripheral blood mononuclear cells (PBMC) has 10-100-fold more activity and safety than semi-synthetic MPLA. The intensity of immune-induced response is also different due to the change of the type of preparation. GLA prepared as stable oil-in-water emulsion (GLA-SE) formulations can induce stronger Th1 reaction than that prepare as an aqueous(GLA-AF) [61]. In a trial of an advanced solid tumor vaccine expressing NY-ESO-1, after stimulating the systemic innate immune response, locally intra-tumoral injection of GLA-SE can recruit NY-ESO-1-specific CD8+ T cells into the tumor [67]. In previous dose-increasing trials, GLA-SE at doses of 2-10 mg showed potential clinical benefits and reliable safety in the experimental patient population. Six of the eleven patients showed increased expression of CD4+ T cells after vaccination and four patients showed increased expression of CD8+T cells, but none of them showed a correlation between GLA-SE dose and induction of immune [68]. And when the vaccine is injected away from the tumor, Deepak et al. proved that the addition of GLA led to alters the kinetics of changes in T cell trafficking, and it can induce a massive proliferation of vaccine primed antigen-specific T cells in the spleen [8]. The most fundamental reason is that GLA precociously matures APC at the injection site, reduces the phagocytic function of APC and reduces the delivery of antigens to draining lymph nodes. But the use of GLA adjuvant can lead to a great number of antigen-specific T cells proliferation induced by vaccine in immune organs, driving more T cells into tumor tissue and infiltrating tumor T cells, resulting in stronger cytotoxicity.
Derivatives of lipid A. All lipid A contain a highly similar core structure of β-1,6-linked disaccharide of D-glucosamine. Specially, subtle changes in chemical structure can result in dramatically different immune activities. An understanding of structure–activity relationships help to provide a better choice when developing a cancer vaccine. In previous studies, many factors such as the number and length of the acyl chain, the structure and substituent pattern of the disaccharide scaffold, the length and phosphorylation state of the acyl chain at positions 1 and 40 of the disaccharide backbone had a strong effect on the activation of the immune response [69].
Gram-negative bacteria have evolved the diversified of their lipid A structures. A typical lipid containing six acyl chains exhibits severe immunological and endotoxic activities, such as asymmetrically distributed (4+2) from E. coli[70], or symmetrically (3+3) from Neisseria meningitidis[71] or Alcaligenes [58, 72]. However, removal of the phosphate group at the O-1-position to generate monophosphoryl lipid A (MPLA) derived from Salmonella Minnesota (Fig. 4B), which greatly eliminates the endotoxin properties without affecting its immune-stimulating activity, has been approved by the FDA for using as an adjuvant vaccine [62]. Most studies have tried to prove that MPLA is a favorable candidate for the formulation of different cancer vaccines with better immune potency.
Combination strategy of lipid A. The construction of a multi-functional liposome vaccine delivery platform with a variety of immune stimulants is the focus of the current preclinical research, researchers are committed to adding vaccine adjuvants and antigen delivery to achieve preferable anti-tumor effects. In order to improve the immunogenicity of the vaccine and the survival rate of cancer patients, the combination strategy of lipid A is a good choice. The combined use of adjuvants achieved dual effects, and the combination of cancer drugs expanded the therapeutic effect of drugs.
Currently, mature adjuvant combination formulations have been tested in preclinical or clinical trials of cancer vaccines. For example, in AS01, MPLA is combined with QS-21, a triterpenoid glycoside saponin extracted from the bark of Quillaja Saponaria Molina, which is a powerful adjuvant [73]. AS02 precipitates MPLA and QS-21 in a stable oil-in-water emulsion [74]. The TLR9 agonist CpG ODN was added to the existing formula, and then the whole was wrapped in lipids to form an AS15 combination [75]. Based on these approved formulations, scientists have developed several new adjuvant coordination strategies. Zhu et al. synergistically activated the immune response with TLR7/8 agonist imiquimod and TLR4 agonist MPLA. From immune study in vivo , the addition of combined adjuvant effectively improved the proliferation and activation of immune cells, especially antigen-specific T lymphocytes. Surprisingly, the mice received the tumor challenge again after vaccination, of which 2/8 mice maintain a tumor-free state to 92 days [76]. The vaccination greatly prolonged the median survival time of the mice. Both complex vaccine adjuvants ((HA/MPLA/QS21), HMQ and (HA/MPLA/R837), HMR) were prepared by Shin, which exerted important immunostimulatory effects in vitro and essential antitumor therapy in vivo [77]. With the help of HA, HMQ or HMR were easily prepared into aqueous solution and freeze-dried in the form of powder, showing high stability, even when stored at room temperature. TLR4 agonist MPLA and 9 agonist oligonucleotide CpG combined within vitro stimulating DC secretion IL-12, TNF-α, and proliferation of the same foreign CD8+ T cells have unique intensive effects, producing adding adjuvant effect [78]. The combination of TLR agonists and non-TLR agonists also presented an effective power [62]. The strong anti-tumor cytotoxic response dependent on natural killer cells can be induced by α-galactose ceramide (α-GalCer) and when cooperated with MPLA and DNA antigens, multiple immune targets are activated at the same time, resulting in more lasting immunotherapy [79]. The cell killing activity of mice generated by the combined strategy is more than twice that of the single adjuvant. Not only that, the use of combined adjuvants can also reduce the amount of antigen in the vaccine to avoid possible problems and obstacles in the use of high doses of antigen.
The efficacy can also be amplified in combination with commonly used cancer drugs. Doxorubicin (DOX), as a broad-spectrum anti-tumor chemotherapy drug commonly used in clinical practice, promoted immunogenic cell death by increasing the expression of tumor antigen in vivo , thus triggering immune response. However, its anti-tumor effect is limited, and it is also necessary to maintain high doses of drug concentrations in the tumor site. This requires increased DOX administration, which inevitably causes drug toxicity and drug resistance problems. The addition of MPLA assists in the immune response triggered by DOX. The joint processing of MPLA-DOX significantly increases the number of invasive CD8+ or CD4+ T cells, which basically inhibits tumor growth [80]. Immune checkpoint inhibitors can inhibit the related receptors PD-L1 or PD-1, which are a successful treatment for patients with advanced melanoma. When combined with MPLA, MPLA-induced interferon γ (IFN-γ) production by T cells can further promote PD-L1 expression, promote the proliferation and CTL of the infiltrated regulatory T cell, thereby finally eradicating the primary tumor and inhibiting tumor transfer and recurrence [81, 82]. Therefore, the combined strategy can make the tumor treatment effects better, which provides a desired strategy for research work.