Vaccines have substantially reduced the burden of infectious diseases. In particular, the eradication of smallpox in 1980 through vaccination is one of the most significant medical achievements to date. However, many infectious diseases with worldwide significance are not currently preventable through vaccination. Traditional live-attenuated/inactivated whole-pathogen vaccines alone are sufficient to induce robust long-lasting immunity in mammals. However, these vaccines are unsuitable when natural infection does not confer long-term protection or when the pathogen cannot be grown in culture. Recombinant protein vaccines (composed of subunits or purified antigens) in isolation elicit only weak and short-lived immune responses [1]. Therefore, these vaccines must be delivered with an adjuvant to enhance and target the adaptive immune response to the vaccine antigens.

A vaccine-adjuvant is a substance or combination of substances added to a vaccine that enhance immunogenicity and contribute to an initial innate immune response by inducing an inflammatory reaction at the injection site. Thus, an adjuvant enhances the magnitude and durability of the vaccine and alters the effect of specific adaptive downstream immune responses to vaccine antigens without inducing a specific antigenic effect against itself ^{[2][3][4][5][6]}[2,3,4,5,6]. In addition, the use of adjuvants benefits vaccine product development in several ways, including reduced antigen dosing and fewer necessary immunizations, which can exert a potential effect on the global vaccine supply. Adjuvants can be classified into different category types, and herein, adjuvants arwe classified adjuvants on their the basis of their physicochemical properties into five families: chemicals (e.g., aluminum salts) [7], lipids (e.g., monophosphoryl lipid A [MPLA]) [8], polysaccharides (e.g., lipopolysaccharide [LPS]) [9], oligonucleotides (e.g., the CpG oligonucleotide) [10], and proteins (e.g., flagellin) [11]. In This entrythis review, we focuses on protein-based adjuvants (PBAs) isolated from different organisms that show significant immunostimulatory properties.

The development of new vaccine adjuvants has been considered one of the slowest processes in the history of medicine. Since the initial licensure of aluminum salts in the 1920s, it has remained for more than 70 years as the only adjuvant included in licensed vaccines against hepatitis B, diphtheria, tetanus, and pertussis, among others ^[6][7][12][6,7,12]. The main difficulties encountered in developing new adjuvants include limited understanding of their molecular complexity, including the molecular mechanisms of PBA action that induce an immune response. The PBAs can exhibit one or more of the following properties: (i) binding and activation of innate immune receptors, known as pattern recognition receptors (PRR), such as Toll like receptors (TLRs), C-type lectin receptors (CLRs), NOD-like receptors, and RIG-I-like receptors (retinoic acid-inducible gene-I-like receptors, RLRs) which recognized pathogen-associated molecular patterns (PAMPs) ^[1][6][11][1,6,11]; (ii) induction of proinflammatory cytokines such as IL-1β, IL-6, TNF-α, and IL-12 and chemokines such as IL-8, which recruit innate immune cells [13]; (iii) regulated expression of co-signaling molecules required for T-cell activation, such as the B7 family members, on professional antigen-presenting cells (APCs) [14]; (iv) induction of specific humoral-mediated (B cell) or cell-mediated (T-cell) immune responses [15]; and (v) induction of trained innate immunity, harnessing the activation state of APCs to enhance adaptive T-cell responses to both the specific antigen and PBA, generating a beneficial bystander effect [16]. A better understanding of these mechanisms will pave the way for the development of next-generation PBA that can stimulate and increase the magnitude and durability of the adaptive immune response to vaccine antigens [6].

2. PBAs as Agonists of Innate Immune Receptors

Since 1989, when Charles Janeway proposed the pivotal theory explaining the mechanism by which the innate immune system discriminates between autoantigens and pathogens, pattern recognition receptors (PRRs) have received a significant attention. Different PRRs families expressed on APCs recognize pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs), leading to their activation and subsequent influencing the type of adaptive immune response ^[17][18][22,23]. The TLR family has received particular attention. Notably, TLR1, TLR2, TLR4, TLR5, and TLR6 are expressed on the cell surface of APCs, whereas TLR3, TLR7, TLR8, and TLR9 are expressed in endosomes ^[19][24]. Through their ectodomains, TLRs bind a wide variety of pathogenic substances ^[20][25]. The TLR ligand induces the dimerization of TLRs and recruitment of adaptor proteins to intracellular Toll/interleukin-1 receptor (TIR) domains to initiate signaling ^[21][22][17,26]. The signaling cascades triggered via these TIR domains are mediated by specific adaptor molecules, including myeloid differentiation primary response 88 (MyD88); MyD88-adaptor-like (MAL), also known as TIR domain-containing adaptor protein (TIRAP); TIR domain-containing adapter-inducing IFNβ (TRIF); and TRIF-related adaptor molecule (TRAM) ^[22][26]. These adaptor proteins have TIR domains and establish TIR–TIR interactions that can be categorized as receptor–receptor, receptor–adaptor, and adaptor–adaptor interactions. Combinatorial recruitment of these adaptors via TIR–TIR interactions orchestrates downstream signaling, leading to the induction of proinflammatory gene expression ^[23][24][27,28]. The Tollike receptors TLR1 and TLR6 heterodimerize with TLR2 and engage in signaling through the MyD88 pathway to activate NF-κB and MAP kinases, which leads to cytokine secretion ^[6][21][25][6,17,29]. The receptors TLR4 and TLR5 are homodimers in the MyD88 signaling pathway. In addition, TLR4 engages with TRIF to induce type I interferon expression mediated through IRF ^[6][21][25][6,17,29]. Because of the roles they play in controlling APC-immunomodulatory functions, TLR agonists are considered promising vaccine adjuvants candidates. Most TLR-binding PBAs activate TLR2, TLR4, and TLR5 to mediate the innate immune response. In addition, other PBAs bind to C-type lectin receptors and the ganglioside GM1/GD1a to trigger immune effects, which are described in the following subsections. The PBAs can interact at with TLR2 (e.g., porins FomA from Fusobacterium nucleatum and major outer membrane protein [MOMP] from Shigella dysenteriae, AB-type toxins from Vibrio cholerae, lipoprotein OMP 16 from Bordetella pertussis); with TLR4 (e.g., high-mobility group box 1 [HMGB1], heat-shock-70-like protein 1 from humans, surface immunogenic protein [SIP] from Group B Streptococcus (GBS), pneumolysin from Streptococcus pneumoniae, hemocyanin from Concholepas concholepas and mistletoe lectin-I [ML-I] from Viscum album); with TLR5 (e.g., flagellin); and heterodimers TLR2–TLR1 (e.g., porin OmpU from V. cholerae) or TLR2–TLR6 (e.g., porin from S. dysenteriae.

2.1. TLR-2-Dependent Activation by PBAs

Because of its ability to recognize ligands as a heterodimer associated with TLR1 or TLR6, TLR2 detects many PAMPs from a wide variety of pathogens ^[26][30]. In addition, many reports have indicated that, after heterodimerization, TLR2 signaling triggers a pro- or anti-inflammatory response. The TLR2 interactions with PAMPs may lead to CD4+ T-helper lymphocyte (Th1 and Th2 cell) or regulatory T (Treg) cell differentiation [6]. Among PBA agonists implicated in the activation of TLR2 are bacterial pore-forming proteins known as porins [11]. Porins are outer-membrane proteins (OMPs) that form channels in Gram-negative bacteria, which regulates the entry of ions, such as K+ and Cl-, and small substrates ^[27][31]. The monomeric porin structure is approximately 48 kDa ^[28][32]. Porins are highly conserved among bacteria. All porins form homotrimers, although some dimeric or octameric porins have been described; these structural variations correlate with differential permeabilities of the OMP-formed pores ^[29][30][31][33,34,35]. Porins from Fusobacterium nucleatum (FomA), a human oral pathogen ^[32][36], induces IL-6 secretion and cell surface upregulation of CD86 and major histocompatibility complex (MHC) type II in splenic B cells. A recombinant form of FomA has been obtained, and its immunostimulatory properties, which are mediated through TLR2 signaling in vitro and in vivo, have been evaluated. Recombinant FomA induces a Th2-cell-type adjuvant effect characterized by enhanced production of OVA-specific IgG1 and IgG2b antibodies in C57BL/6 mice and enhanced secretion of IL-10 and IL-6 by splenic B cells ^[33][37]. The major OMP (MOMP) from Shigella flexneri activates TLR2, enhancing NF-kB and p38 MAP kinase activation ^[34][38]. In addition, the MOMP from Chlamydia trachomatis has been shown to induce IL-8 and IL-6 production in a TLR2/TLR1-dependent manner ^[35][39]. Porin from Shigella dysenteriae stimulated TLR2/TLR6 naïve CD4+ T-cells, B-cells, and macrophages ^[36][37][38][41,42,43], contributing to a Th1-cell immune response. In addition, the Vibrio cholerae porin OmpU mediates M1-polarization of macrophages/monocytes via TLR1/TLR2 activation ^[39][44]. The PorB porin from Neisseria meningitidis preferentially binds to the TLR2/TLR1 heterodimer compared to the TLR2/TLR6 heterodimer, upregulating CD86 expression in splenic B cells and NF-κB nuclear translocation in a HEK reporter cell line ^[40][45]. The PorB porin also enhances APC trafficking and cross-presentation and increases antigen deposition on germinal center follicular dendritic cells (DCs) ^[41][46]. Immunization against an inactivated H1N1 2009 pandemic influenza virus combined with Salmonella enterica serovar Typhi porins OmpC and OmpF as adjuvants, elicits a humoral response characterized by higher hemagglutinating anti-influenza IgG titers, antibody class switching rates, and affinity maturation. In addition, coadministration of OmpC and OmpF with unconjugated Vi capsular polysaccharide (a T-cell-independent antigen) induces higher IgG antibody titers and class switching rate in a murine model. The mechanism mediating these adjuvant effects might be related to the agonistic effect of S. typhi porins on TLR2 and TLR4 activity ^[42][47]. Tuberculosis (TB), caused by Mycobacterium tuberculosis, remains a major infectious disease worldwide. Early secreted antigenic target protein 6 (ESAT-6) is one of the most prominent antigens expressed by M. tuberculosis strain H37Rv. This PBA promotes a lung Th17 immune cell response in a TLR2-dependent manner. In addition, ESAT-6 induces IL-6 and TGF-β production by DCs ^[43][48]. Moreover, recombinant Brucella cell-surface protein 31 (rBCSP31) from Brucella abortus is a TLR2 and TLR4 protein agonist that induces TNF-α, IL-6, and IL-12p40 cytokine production by macrophages and a Th1-cell immune response ^[44][49]. Streptococcus pneumoniae, one of the leading causes of invasive bacterial disease worldwide, expresses two heat shock proteins that are important conserved virulence factors: DnaJ and pneumolysin (Ply), which are TLR4 and TLR2 ligands. The protein DnaJ is a member of the Hsp40 family and functions mainly as a molecular Hsp70 chaperone and thus participates in protein folding and assembly. Notably, Ply is a toxin that can be used as a carrier protein with future pneumococcal conjugate vaccines because of its immunogenic activity; however, it is highly toxic ^[45][52], which makes incorporating Ply into new vaccines a challenge. A fusion protein comprising two virulence factors of Streptococcus pneumoniae, DnaJ and a less-toxic Ply mutant (DnaJ-ΔA146Ply), induces the production of IL-12 and Th1 cell proliferation mediated via TLR2 in bone marrow-derived dendritic cells (BMDCs). In addition, in a preclinical model, DnaJ-ΔA146Ply confers protection against S. pneumoniae in a TLR2-dependent manner ^[46][50]. Another protein agonist of TLR2 in S. pneumoniae is recombinant endopeptidase O (rPepO), a pneumococcal virulence protein. The intratracheal instillation of rPepO protein results in a significant increase in IL-6, TNF-α, CXCL1, and CXCL10 production and neutrophil infiltration in mouse lungs. Interestingly, compared with wild-type mice, TLR2- or TLR4-deficient mice subjected to rPepO treatment show decreased cytokine production, reduced neutrophil infiltration, and intensified tissue injury. In addition, upon stimulation of peritoneal exudate macrophages (PEMs), rPepO induces IL-6, TNF-α, CXCL1, and CXCL10 production, which relies on the rapid phosphorylation of p38, protein kinase B (PKB, also known as Akt), and p65 in a TLR2-/TLR4-dependent manner ^[47][51]. Lipoproteins, such as BP1569 from B. pertussis, have emerged as novel TLR2 agonists. BP1569 has a molecular mass of 40 kDa and shares a sequence with lipoproteins from N. meningitidis, Burkholderia pseudomallei, and Haemophilis influenzae ^[48][49][53,54]. The three-dimensional structure of BP1569 and that of other lipoproteins of interest have not been characterized; however, comparative analyses of the amino acid sequences indicates that these proteins contain a positively charged N-terminal signal sequence, followed by a hydrophobic region and a lipobox sequence, which is acylated. Notably, the lipobox acyl group is essential for the immunostimulant effects of these lipoproteins since it directly interacts with TLR2 ^[50][55].

2.2. TLR2-and Ganglioside-Dependent Activation by PBAs

AB-type toxins, such as cholera toxin (CT) in V. cholerae and heat-labile enterotoxin (LT) in Escherichia coli, have been extensively studied as mucosal adjuvants. Table 2 resumes PBAs of bacterial origin with TLR2 and GM1ganglioside adjuvant potential. The CT and LT show high amino acid sequence identity, and their three-dimensional structures are similar ^[51][52][53][56,57,58]. Cholera toxin is a hexamer formed by a single A subunit (28 kDa) and five B subunits (11 kDa each). The A subunit comprises A1 and A2 domains: A1 is a globular ADP-ribosylase whereas A2 is an extended alpha-helix. The A2 domain tethers the A subunit to the pentameric ring formed by the B subunits ^[53][58]. The B5 pentameric ring is essential to pathogenesis because it binds to glycosphingolipids on target cells, such as the ganglioside GM1, allowing CT endocytosis. This process promotes the release of A1 after disulfide bridges reduction and the subsequent recognition, processing, refolding, and activation of A1 enzymatic activity ^[54][55][59,60]. Similarly, the LT structure is an AB5 hexamer: a single A subunit with catalytic activity, and five B subunits form a ring with membrane-binding functions ^[52][57]. However, CT and LT present biophysical differences in terms of solvent-accessible contact area: CT has a higher contact area than LT, which limits the diffusion of water through CT-formed pores. These differences correlate with the different pathogenic effects of CT and LT ^[56][61]. Full-length CT and LT are pathogenic, and therefore, mucosal adjuvants are composed of less-toxic and less-allergenic derivatives of these proteins ^[57][62]. In one strategy, the catalytic activity of CT and LT is prevented by administering only the pentameric ring, which has immunomodulatory properties itself ^[58][63]. Another strategy involves site-directed mutagenesis to impair the catalytic activity of the A subunit. These mutants are powerful immunostimulants, but they might show residual activity depending on the amino acid residue substitution ^[59][64]. Cholera toxin enters the endoplasmic reticulum of immune cells through endosomes following binding to the ganglioside GM1 in mucosal membranes ^[60][65]. Then, the A1 subunit is released via disulfide bond reduction, and CTA1 is retroactively translocated to the cytosol ^[61][66]. Cytosolic CTA1 can bind to Gsα, catalyzing its ADP-ribosylation, and subsequently can elevate the 3′,5′-cyclic AMP (cAMP) concentration in a host cell ^[62][67]. Enhanced cAMP concentration mediated through cytosolic CTA1 induces the production of proinflammatory cytokines, including IL-1β, TNF-α, and IL-6, in DCs and CD4+ T-cells ^[63][68]. In a model of intranasal anthrax infection, CT has been shown to mediate the induction of IL-17-producing CD4+ Th17 cells ^[63][68]. Escherichia coli type I and type II LTs share many physiological and structural features. However, recent studies have shown that each toxin triggers unique signaling cascades, leading to different cellular responses ^[64][69]. The LTs interact with gangliosides, mediating the signaling between immune-competent cells, in which the composition of ganglioside species varies. However, LT B-pentamers typically interact with either gangliosides and/or TLRs. Studies have established that the B-pentamers in type II LTs (LT-IIa and LT-IIb) interact with TLR2, leading to the induction of IL-1β, IL-6, IL-8, and TNF-α expression in human THP-1 cells ^[64][69]. Furthermore, stimulation of human embryonic kidney (HEK)-293 cells that transiently express TLR1 and TLR2 has been shown to activate NF-κB-dependent luciferase gene expression ^[64][65][69,70]. In addition, studies have demonstrated the importance of the interaction between TLR2 and GD1a and a subunit of type IIb E. coli enterotoxin (LT-IIb-B5) ^[66][71]. Both LT-IIb-B5 and a defective GD1a-binding mutant (LT-IIb-B5[T13I]) binds TLR2 with moderate affinity. However, only the wild-type molecule demonstrates a significant increase in TLR2-binding activity in the presence of GD1a. Furthermore, fluorescence resonance energy transfer experiments have indicated that LT-IIb-B5 induces the recruitment of TLR2 and TLR1 to lipid rafts and clustering with GD1a, in contrast to the defective GD1a-binding mutant, which does not activate TLR2 signaling ^[67][72].

2.3. TLR4-Dependent Activation by PBAs

The receptor TLR4 is a promising target for immunomodulation, partially due to the success of the GlaxoSmithKline-produced MPLA adjuvant, the first TLR agonist approved by the Food and Drug Administration (FDA) for use in the development of new vaccines. Moreover, the adjuvant AS01, a mixture of MPL and QS21 (purified saponin from Quillaja saponaria), activates TLR4, stimulating a Th1-cell-type response, which can trigger the activation of CD8+ T-lymphocytes, showing potential applications to vaccines for malaria and herpes zoster ^[68][73]. There are several TLR4-dependent PBAs that exhibit great immunogenicity amongst them cell death derived adjuvants (HMGB1 and Hsp70L1), SIP from GBS, virulence factors DnaJ and Ply from S. pneumoniae, Omp16 from Brucella abortus, proteins from M. tuberculosis (RpfE, HBHA, Rv0652 and GrpE), hemocyanins and plant lectins.  It has been described that adjuvants such as Alum or MF59 cause local tissue damage and cell death creating a local pro-inflammatory milieu to recruit immune cells ^{[3][69][70][71]}[3,74,75,76]. In this context, the extracellular release of HMGB1 can activate DCs to stimulate adaptive immunity ^[72][77]. Another mammalian protein regulating cell death is Hsp70-like protein 1 (Hsp70L1), promotes the production of TNF-α, IL-1, and IL-12p70 and the expression of surface markers such as CD40, CD80, and CD86 in bone marrow-derived DCs ^[73][78]. On the other hand, in a murine model, Hsp70L1 generates a specific Th1-cell-triggered immune response against carcinoembryonic antigen (CEA) ^[73][78]. The Surface Immunogenic Protein (SIP) in in Group B Streptococcus (GBS) has a molecular mass of 53 kDa, and its amino acid sequence is highly conserved among different species. The main secondary SIP structure consists of β-sheets, but the three-dimensional structure has not been characterized ^[74][79]. Similarly, the physiological function of SIP remains unknown; however, this protein is exposed to GBS surface and can be secreted ^[75][76][80,81]. Furthermore, a recombinant surface immunogenic protein in GBS (rSIP), expressed by E. coli and Pichia pastoris has shown immunomodulatory properties as a TLR4 agonist protein adjuvant ^[74][77][79,82]. TheOur data show that rSIP stimulates innate immune cells as an adjuvant to induce Th1 adaptive immune responses and is an oral mucosal vaccine candidate against GBS ^{[77][78][79][80]}[82,83,84,85]. Virulence factors of S. pneumoniae, such as DnaJ ^[81][86] and Ply ^[47][82][51,87], have been shown to act as TLR4 ligands. In this context, pneumolysin stimulates TNF-α and IL-6 in wild-type macrophages but not in macrophages in which MyD88 is deleted. Moreover, macrophages that carry a spontaneous mutation in TLR4 (P712H) are hyporesponsive to Ply. Recombinant DnaJ induces BMDC activation and maturation mediated via TLR4 and activated MAP kinase, NF-κB, and PI3K-Akt pathways. In addition, rDnaJ-treated BMDCs effectively stimulated naïve CD4+ T-cells to secrete IFN-γ and IL-17A. Moreover, the fusion of DnaJ and a less toxic Ply mutant (1A146Ply-) DnaJ-1A146Ply induces TLR4-dependent Th1- and TH17-cell-like responses against S. pneumoniae ^[83][88]. The outer membrane protein (Omp)16 lipoprotein is another TLR4 ligand, from Brucella abortus, that stimulates DCs and macrophages in vitro and induces a protective TLR4-dependent Th1-cell immune response against B. abortus infection ^[84][89]. Wild-type macrophages and BMDCs pulsed with Omp16 showed a significant increase in IL-12 and TNF-α expression compared to cells from TLR4-deficient mice. In addition, immunization with Omp16 generates Th1-cellimmune responses characterized by the secretion of IFN-γ by murine splenocytes ^[84][89]. Mycobacterium tuberculosis harbors four TLR4-dependent protein ligands including RpfE, HBHA, Rv0652 and Grp. Resuscitation-promoting factor E (RpfE), a latency-associated member of the Rpf family, promotes naïve CD4+ T-cell differentiation toward Th1 and Th17 cells. The RpfE induces DC maturation by increasing the expression of surface molecules such as CD86 and CD80 and the production of IL-6, IL-1β, IL-23p19, IL-12p70, and TNF-α ^[85][91]. In this context, RpfE activates ERK, p38 MAPKs, and NF-κB signaling after TLR4 binding ^[85][91]. Heparin-binding hemagglutinin adhesin (HBHA) induces DC maturation in a TLR4-dependent manner that is characterized by the expression of CCR7, CD40, CD80, CD86, MHC class I and II, and the proinflammatory cytokines IL-6, IL-12, IL-1β, and TNF, leading to a Th1-clell immune response. In addition, mechanistic investigations have established that the MyD88 and TRIF signaling pathways downstream of TLR4 mediate the secretion of HBHA-induced proinflammatory cytokines ^[86][92]. The 50S ribosomal protein L7/L12 (RPLL) Rv0652 modulates DC maturation and proinflammatory cytokine production (TNF-α, IL-1β, and IL-6) partially mediated through the TLR4/MyD88 signaling pathway. Moreover, DCs pulsed with Rv0652 plus OVA exhibits an induced OVA-specific CD8+ T-cell response, slowed tumor growth, and prolonged long-term survival in an OVA-expressing E.G7 thymoma murine model ^[87][93]. In addition, GrpE, a cofactor of heat-shock protein 70 (HSP70), promotes Th1-cell-type immunity by interacting with TLR4 located on DCs. These effects of GrpE on DC activation are mediated by the downstream activation of the MyD88, TRIF, MAPK, and NF-κB signaling pathways ^[88][94]. Regarding hemocyanins, theour data show the involvement of TLR4 in a hemocyanin-mediated proinflammatory response in APCs ^[89][90][98,99]. Mollusk hemocyanins are large oligomeric glycoproteins widely used as adjuvants, peptide and hapten carriers, and nonspecific natural immunostimulants in certain tumor therapies ^{[91][92][93][94]}[100,112,113,114]. The biochemical and biophysical attributes of several hemocyanins with immunomodulatory properties purified from wild gastropods, including Keyhole limpet hemocyanin (KLH) from Megathura crenulata which has two immunization forms (high-molecular-weight and subunit clinical-grade formulations), CCH from C. concholepas, FLH from Fissurella latimarginata, RtH from Rapana thomasiana, HtH from Haliotis tuberculata, and HlH from Helix lucorum, have been characterized ^{[91][92][93][94]}[100,112,113,114]. The molecular mass of the hemocyanin oligomeric structure in these species is as high as 8 MDa, and each hemocyanin forms a cylinder of approximately 35 nm in diameter known as a didecamer. Each didecamer comprises 20 subunits (approximately 400 kDa each). However, fewer decamers or multimers have been identified. The hemocyanin subunits contain eight globular domains known as functional units (FUs), and each FU includes an active site in which oxygen reversibly binds ^{[95][96][97][98]}[115,116,117,118]. Interestingly, hemocyanins consist of one or two types of subunits; for example, FLH is composed of only one subunit, whereas KLH, CCH, RtH, HtH and HlH are each composed of two subunits. Furthermore, these proteins fold into a homodidecamer, such as KLH, or heterodidecamer, such as CCH ^{[99][100][101]}[101,102,103]. Another essential feature of hemocyanins is an abundant glycan content, which reaches 3–4% w/w. In the case of KLH, cryogenic electron microscopy analyses have shown that carbohydrates are localized on the rim of the cylinder wall and on the wall surface of the molecule ^[96][102][116,119]. The N- and O-linked carbohydrates are added glycosylation, and their abundance and heterogenicity vary among species, but are primarily mannose-rich N-glycans and N-mixed carbohydrates with fucose, galactose, GlcNAc, and glycosylation branches that are not found in mammals ^[89][98]. Glycosylation is essential for the stability of the hemocyanin oligomeric structure ^{[89][103][104][105]}[98,120,121,122]. Furthermore, glycosylation contributes to hemocyanin-induced immunostimulant properties, highlighting the relevance of posttranslational modifications in adjuvant proteins ^{[89][104][106][107]}[98,104,121,123]. Notably, rwesearchers confirmed glycan-dependent binding of hemocyanins to chimeric TLR4 in vitro and in vivo ^[89][90][98,99]. Indeed, DCs from mice with deficient MyD88 expression are partially activated by FLH, suggesting a role played by the TLR pathway in hemocyanin recognition leading to APC activation. Moreover, hemocyanin-induced proinflammatory cytokine secretion is impaired in several models of APCs lacking functional TLR4. Furthermore, rwesearchers have shown that KLH and FLH induced TLR4-dependent ERK1/2 phosphorylation, a key event in the TLR4 signaling pathway ^[90][99]. Plant lectins selectively bind to carbohydrate motifs, and they can be purified by affinity chromatography using different ligands ^[108][109][124,125]. Mistletoe lectin-I (ML-I) and jacalin have been studied as potential mucosal adjuvants because their derivatives bind to mono- or oligosaccharides on mucins or IgA molecules and target the mucosal epithelium ^[110][107]. Soybean agglutinin (SBA) is a nonfibrous carbohydrate-related protein with a molecular mass of 120 kDa and the main non-nutritional factor in soybean. Specifically, SBA is a tetrameric N-acetyl-D-galactosamine (GalNAc) and galactose-specific lectin that forms a unique cross-linked composite with a variety of naturally occurring and synthetic multi-antennary carbohydrates with terminal GalNAc or galactose residues ^[111][106]. Regarding their mechanism of action, interactions between glycosylated TLR receptors and certain lectin types on APCs has been identified ^[112][126]. For instance, SBA stimulates TLR4 in a reporter cell line. This effect was specific for TLR4, and no agonist effect was observed for TLR-2/6, -3, -5, -7, -8, and -9 ^[111][106]. A single intratumoral injection of recombinant Mistletoe lectin (Aviscumine) prolongs the median survival of glioma-bearing mice ^[113][108]. The Lavelle group showed, for the first time, that ML-1 exhibits high mucosal adjuvant activity when administered with herpes simplex virus glycoprotein D2, showing an enhanced type Th2-cell response to the bystander antigen ^[114][109]. This outcome suggests that ML-I provides a platform for the generation of effective mucosal adjuvants due to its ability to penetrate the gut, where it induces mucosal and systemic immune responses ^{[114][115][116]}[109,110,111]. In the case of Jacalin, early studies have shown its adjuvant effects on humoral immune responses in mice immunized with a lysate or viable epimastigote forms of Trypanosoma cruzi, which result in a marked increase in the levels of anti-T. cruzi antibodies ^[108][124]. However, another plant lectin, ArtinM, administered to mice immunized against neospora, a dog parasite, shows a more significant immunostimulatory and adjuvant effect than Jacalin ^[117][127].

2.4. TLR5-Dependent Activation by PBAs

The receptor TLR5 responds to flagellin from β- and γ-proteobacteria ^[118][128]. The product encoded by the flagellin gene in Salmonella binds to TLR5, inducing MyD88-dependent signaling with subsequent activation of the NF-kB pathway in epithelial cells, monocytes, and DCs ^[119][129]. Flagellin is the structural component of the flagellum, whose primary function is bacterial motility ^[120][130]. The amino acid sequences of the N- and C-terminal regions are conserved among different bacteria. These regions are critical for flagellin oligomerization ^[121][122][131,132]. In contrast, the central domains are highly variable and might contribute to protein stability; the N- or C-terminus of full-length flagellin has been reported to be a safe and nontoxic adjuvant of AB-type toxins in protein form ^{[123][124][125]}[133,134,135]. However, high doses are related to systemic inflammation and liver injury. Therefore, flagellin has been produced as a recombinant protein and fused or co-expressed with different antigens to boost the immune response mediated by TLR5 ^[126][127][136,137]. The TLR5 ligand flagellin has been studied to determine its usefulness as an adjuvant. The first evidence of flagellin proinflammatory activity was observed with Salmonella flagellin, which was shown to be a potent inducer of cytokines at sub-nanomolar concentrations in a promonocytic cell line ^{[128][129][130]}[138,139,140]. Flagellin can profoundly activate migratory lung DC (migDC) subsets and upregulate CD40, CD80, CD86, and CCR7 in these DCs ^[131][132][141,142]. The adjuvant activity of flagellin has been shown in a mouse models of infectious diseases, leading to adequate protection against infection, indicating that flagellin can be used as a carrier for peptides derived from influenza virus ^[133][143] and in an experimental vaccine for Schistosoma mansoni ^[133][143]. Subsequently, many reports have described flagellin as a vaccine adjuvant and carrier in preclinical studies with several antigens of microbial origin, including vaccinia virus ^[134][144], the parasite Plasmodium falciparum ^[135][136][145,146] and the HIV gp40 protein ^[137][147]. Notably, flagellin has been co-administered with allergens to inhibit airway allergic disease in a murine model of allergic rhinitis ^[138][139][148,149]. Finally, flagellin has been used as an adjuvant in a recombinant hemagglutinin (HA) fusion vaccine (VAX125), inducing higher antibody titers in humans than the vaccine alone ^[140][141][150,151]. Mycoplasma hyopneumoniae is the etiological agent of porcine enzootic pneumonia. The P97 protein can mediate microbial adhesion to epithelial cells in the respiratory tract. Remarkably, recombinant expression of the P97 C-terminal domain triggers concentration-dependent TLR5 activation, similar to flagellin, and stimulates the production of IL-8 in HEK-Blue mTLR5 cells. Mice immunized with P97c fused to the ectodomain matrix 2 protein (M2e) of influenza A virus exhibit a high antibody titer against the M2e epitope, which is associated with a hybrid Th1-/Th2-cell immune response ^[142][152].

2.5. Interaction of PBAs with C-Type Lectin Receptors

C-type lectin receptors comprise several families of receptors, including collectins, selectins, endocytic receptors, and proteoglycans, whose interactions with their glycosylate ligands can be calcium-dependent or calcium-independent ^[143][156]. The CLRs possess one or more carbohydrate recognition domains (CRDs) or C-type lectin-like domains (CTLDs), which recognize other noncarbohydrate agonists ^[144][157]. Upon ligand recognition, antigen is internalized, processed, and presented, inducing intracellular signaling pathways that regulate cellular function ^[145][158]. However, few reports on the roles played by CLRs in PBA recognition and modulation of immune responses have been published. Moreover, some studies have suggested that CLRs are dispensable for some PBAs ^[146][147][159,160], whereas for hemocyanins are indispensable ^[145][158]. Due to the multivalent nature of their glycosylated residues, hemocyanins, in contrast to other PBAs, interact not only with TLR4 as previously described but also with several CLRs. Indeed, rwesearchers have previously shown that murine APCs internalize hemocyanins in a glycosylation-dependent manner through receptor-mediated endocytosis with proteins that contain a CTLD, such as the mannose receptor (MR) and macrophage galactose lectin (MGL) ^[89][90][98,99]. Similarly, reswearchers have observed that hemocyanins directly bind to CRDs in the MR and DC-SIGN (a DC-specific ICAM-3–grabbing nonintegrin) with high affinity constants, colocalizing with these receptors after being internalized into human DCs through clathrin-mediated endocytosis ^[106][104]. Notably, MR lacks a cytoplasmic domain and therefore cannot transduce external signals to intracellular pathways, requiring its cooperation with other innate immune receptors such as TLR4 ^[90][99]. Importantly, RtH presents significant adjuvant properties when administered to mice in conjunction with bacterial or viral proteins, a toxoid, and an influenza preparation ^[148][161]. In the same antigen preparations, HtH increases anti-toxoid IgG antibodies in the serum of mice to levels comparable to those produced by mice that receive toxoid Al(OH)₃. In addition, HtH induce a strong anti-influenza cytotoxic response ^[149][162]. The FLH shows better immunogenic capabilities than CCH and KLH, exhibiting significant antitumor activity in a B16F10 mouse melanoma model ^[99][101] and murine model of oral cancer ^[150][163]. The interaction of hemocyanins with various receptors endows them with advantages as adjuvants because they can activate different signal transduction pathways, leading to potent immunostimulatory effects.

3. PBAs and the Adaptive Immune Response

Many adjuvants primarily target DCs to induce cellular activation, including antigen presentation. Pattern recognition receptors s enhance the expression of CCR7 on APCs, which promotes their migration to draining lymph nodes, at which time, protein antigens are processed and loaded onto the MHC, facilitating the signaling required to activate naïve antigen-specific Th cells ^[151][164]. Upon activation, DCs upregulate the expression of MHC-I/MHC-II and costimulatory molecules and release cytokines and chemokines that polarize T-cells toward acquisition of a Th1, Th2, or Th17 phenotype. Specifically, IL-12 promotes the acquisition of the Th1 phenotype, primarily contributing to cellular immunity ^[152][153][165,166]. In contrast, IL-4 and IL-10 in the absence of IL-12 promote acquisition of the Th2 phenotype, which stimulates humoral immunity ^[152][153][165,166]. The Th17 phenotype acquisition is promoted by TGF-β and IL-6, which are important to mucosal immunity and protect against bacterial and fungal infection ^[154][167]. A specific cytokine profile is required to overcome immune tolerance and is controlled by Tregs ^[155][168]. Activated Th-cells upregulate CXCR5 expression, which mediates Th-cell migration to the interface between the B-cell follicle and T-cell areas. Th-cells express IL-21 and CD40 L, which stimulate the clonal expansion of antigen-activated B cells ^[156][169]. Antigen-specific B-cells can migrate to the medullary cord and differentiate into short-lived plasma cells. In contrast, other activated B-cells migrate to B-cell follicles and form germ centers (GCs) ^[157][170]. In GCs, B-cells differentiate into recirculating memory B-cells or long-lived plasma cells (LLPCs) that migrate to the bone marrow ^[158][171]. This process results in affinity maturation during the antibody response, which generates sustained antibody responses [6]. The PBAs can regulate lymphocyte function and adaptive immune responses at different levels. The effect of a PBA on the adaptive immune response is directly related to its immunogenic potential, which is associated with the type of cognate PRR, type of cell, and several tissue-specific factors. For instance, the immunological potency of TLR/CLR agonists has been reported to vary depending on the cell type ^[68][73]; that is, certain MPLA formulations behave as full agonists or partial agonists depending on whether they target human or murine TLR4. In addition, most TLR/CLR agonists induce the expression and secretion of proinflammatory cytokines from Th1-/Th2-/Th17-cell types to different degrees and in different timeframes, suggesting differences in the signal transduction pathways that they trigger. Importantly, antibodies are the crucial effector molecules necessarily induced by vaccines because antibodies trigger immune responses (neutralization, opsonization, complement activation, and antibody-dependent cellular cytotoxicity). Therefore, as thymus-dependent antigens, PBAs, contribute directly or indirectly these immune responses. Furthermore, with the development of bioinformatics tools and information on the sequences of protein antigens, T-cell receptor (TCR) and B-cell receptor (BCR) epitopes can be designed for epitope-based vaccines ^[159][172]. This approach endows PBAs with a significant advantage over other types of adjuvants since many structures can be epitopes, and PBAs are the only adjuvants that can directly induce cellular and humoral immunity. Because only MHC-peptide complexes can bind TCRs and activate T-cells, T-cell epitopes in PBAs can be identified or designed to be specific to MHC molecules expressed by a cell. For example, T and B epitopes in flagellin from B. pseudomallei have been predicted, and the predicted peptides have been synthesized and characterized using bioinformatics tools. As a result, two of the produced peptides include dominant immunoreactive epitopes, which elicit cytokine production in human peripheral blood mononuclear cells ^[160][173]. Undoubtedly, the limitations associated with this type of approach need be considered to prevent antigen-associated complications that may result in adverse effects ^[161][174].