In 1994 Badolato et al. were the first to describe a dose-dependent chemotactic effect of rhSAA on human polymorphonuclear leukocytes (PMNs) and peripheral blood mononuclear cells (PBMCs) both in vitro and in vivo. Moreover, rhSAA upregulated the expression of cell adhesion molecules including CD18/CD11b and CD18/CD11c. This finding was further complemented by adhesion assays that demonstrated increased adhesion of rhSAA-treated PMNs and PBMCs to endothelial cells (19). A year later, the same group demonstrated the inhibition of rhSAA-mediated human monocyte chemotaxis following pertussis toxin (PTx) treatment (20). While this narrowed down the receptor utilized by rhSAA during chemotaxis to a GPCR, it was only in 1999 that rhSAA was demonstrated to utilize FPR2 to chemoattract leukocytes (21). The in vitro recruitment of human neutrophils by rhSAA via FPR2 activation was later corroborated by Lee et al. (22). In addition, the chemoattractant effect of rhSAA on murine neutrophils has been demonstrated to occur via the murine counterpart of human FPR2 (23). In line with this, FRP2^-/^- mice show reduced overall cellular trafficking in an interleukin (IL)-1β-induced air pouch model following intravenous administration of rhSAA (24). Moreover, Chen et al. demonstrated the role of adipocyte-derived murine SAA3 in the recruitment of bone marrow-derived macrophages via FPR2 within the context of obesity (25). Over the years, the chemotactic effect of A-SAA has expanded to include more cell types such as mast cells, CD14⁺ monocytes, immature dendritic cells, macrophages, T-lymphocytes, smooth muscle cells, an epithelial ovarian cancer cell line (SKOV3) and endothelial cells (24, 26–31).

We recently provided additional evidence on the utilization of FPR2 by rhSAA1.1 during in vitro neutrophil chemotaxis using the selective FPR2 antagonist WRW4 (32). However, as previously discussed, a number of published data investigating the biological activity of recombinantly expressed SAA might in fact be mediated by contaminating bacterial antigens rather than being inherent to the APP itself (vide supra). This led us to question whether the FPR2-dependent chemotactic function of A-SAA might be mediated via contaminating formylated bacterial peptides. To this end, we investigated the in vivo chemotactic potential of rhSAA1.1 purified to homogeneity, via reversed-phase high-performance liquid chromatography, in C57BL/6 mice. We observed significant recruitment of neutrophils and mononuclear cells into the joint upon intra-articular injection of the purified rhSAA1.1, thereby solidifying the FPR2-mediated chemotactic function of A-SAA (7).

Synergy between GPCR ligands has been described within the framework of leukocyte chemotaxis. Indeed, chemokines have been shown to synergize not only with each other but also with other non-chemokine GPCR inflammatory ligands (33). The latter phenomenon is of interest in relation to A-SAA. Much to our surprise, we observed a synergistic effect between rhSAA1.1 and CXCL8 in the activation and recruitment of human neutrophils evidenced through neutrophil shape change and cell migration in the Boyden chamber chemotaxis assay, respectively (32). De Buck et al. recently purified a supposed monocyte chemotactic factor from bovine serum, which was later identified as a C-terminal fragment of SAA1 corresponding to the amino acids 46-122 of the mature protein. Interestingly, while this C-terminal fragment of bovine SAA1 lacked the capacity to induce human leukocyte recruitment as a single inducer, it synergized with CXCL8 and CCL3 in the recruitment of neutrophils and CD14⁺ monocytes, respectively. This synergistic effect was dependent upon FPR2 activation. Moreover, this bovine fragment could also synergize with murine CXCL6 in vivo to enhance the intraperitoneal recruitment of neutrophils in NMRI mice (34). In a similar manner, rhSAA1.1 peptides generated by matrix metalloproteinase (MMP)-9 corresponding to the amino acids 52-104 and 58-104 of SAA1.1 possess the capacity to synergize with CXCL8 in the activation and recruitment of human neutrophils, a finding that was not shared by their N-terminal counterpart, SAA1.1(1–51) (35).

While A-SAA is largely regarded as an atherosclerosis disease biomarker (49), growing evidence suggests that SAA is not merely a passive participant but rather plays an active role in the formation of atherosclerotic plaques through the modulation of several disease-related pathways. The contribution of A-SAA to atherosclerotic plaque formation via FPR2 activation is discussed below.

Through A-SAA (SAA1.1 and SAA2.1) knockout (KO) in C57BL/6 mice, Smole et al. demonstrated a reduced type 2 immune response following house dust mite (HDM) exposure. Indeed, A-SAA knock-out (KO) mice showed reduced serum immunoglobulin E concentrations, bronchoalveolar lavage (BAL) eosinophilia, and goblet cell metaplasia of the airway lumen in comparison to wild-type (WT) animals. This occurred in line with notably lower numbers of dendritic cells and type 2 innate lymphoid cells and reduced Th2-cytokine production. Epithelial cell-derived IL-33 is important for the expansion of type 2 innate lymphoid cells that play a role in maintaining the local allergic response. Interestingly, the KO of A-SAA in C57BL/6 mice challenged with HDM brought about no change in baseline IL-33 levels in BAL fluid. On the contrary, WT mice showed a significant increase in IL-33 expression in BAL fluid upon exposure to HDM. Further in vitro experiments provided evidence of the mechanism by which A-SAA regulates the local allergic response. Indeed, HDM treatment of an airway epithelial cell line, BEAS-2B, concurrent with siRNA inhibiting SAA1, reduced expression of IL-33. Moreover, SAA1 was shown to bind to mite allergens Der p 13 and Blo t 13 upon which the formed complex activated FPR2 leading to the expression of IL-33 in the BEAS-2B cell line (59).

The FPR2-mediated anti-apoptotic function of rhSAA was initially demonstrated within the context of arthritic diseases, where RA-derived and OA-derived fibroblast-like synoviocytes (FLS) showed increased proliferation, following treatment with rhSAA. This proliferative effect was reduced upon the introduction of FPR2 siRNA. This was further complemented by MTT and DNA fragmentation assays, which illustrated the anti-apoptotic effect of rhSAA in response to serum-free starvation. Moreover, rhSAA protected against sodium nitroprusside-, nitric oxide- or anti-Fas IgM plus cycloheximide-induced apoptosis (60). In line with the aforementioned findings, García et al. demonstrated an anti-apoptotic effect of rSAA (unknown origin) on neutrophils. Indeed, the treatment of neutrophils with rSAA brought about an increase in the percentage of viable neutrophils relative to the DMSO control while decreasing the percentage of apoptotic and dead neutrophils. The anti-apoptotic effect of rhSAA1.1 extends beyond neutrophils to include monocytes. We have observed enhanced survival of monocytes following treatment with homogenously purified rhSAA1.1 (7). By delaying the apoptosis of neutrophils and monocytes, A-SAA prolongs their lifespan and contributes to the inflammatory response. In contrast, the pre-treatment of neutrophils with the FPR2 agonist BMS-986235 reduced the percentage of viable neutrophils and increased the percentage of apoptotic cells in response to rhSAA1.1, thereby implicating FPR2 in the rSAA-mediated anti-apoptotic effect (61). Thus, activation of FPR2 with pro-resolution agonists such as BMS-986235 or endogenous molecules such as 15-epi-LXA4 can override the anti-apoptotic signal from A-SAA and drive resolution (62).

Li et al. demonstrated the capacity of A-SAA to regulate macrophage phenotype within the context of hepatocellular carcinoma (HCC) in a pleiotropic manner. The treatment of U937 macrophages with rSAA (unknown origin) skewed the phenotype towards an M2-like subset characterized by IL-10^high, IL-12p35^low and IL-23p19^high an effect that was mitigated by the FPR2 antagonist WRW4. Moreover, rSAA-treated macrophages showed an upregulation of IL-1β, IL-6, TNF-α and CXCL8. In vivo, LL-37, an FPR2 ligand, provoked a macrophage phenotype similar to that induced by rSAA in vitro. Indeed, analysis of tumor infiltrating macrophages from in situ tumor tissue following LL-37 treatment showed macrophages characterized by high IL-10 and low IL-12p35 expression levels, thereby implicating FPR2 signaling in regulating the phenotype of tumor-associated macrophages. In a paracrine manner, through its action on HCC cell lines, namely murine H22 and human HepG2, rSAA upregulated the expression of macrophage-colony stimulating factor (M-CSF) and CCL2, which in turn regulate macrophage phenotype, contributing to the tumor-associated macrophage population. Indeed, the treatment of U937 macrophages in conjugation with M-CSF and CCL2 skewed the population towards an M2-like phenotype characterized by high IL-10 and low IL-12p35 expression. Simultaneous treatment with M-CSF and CCL2 also provoked the expression of IL-1β, IL-6, TNF-α, CCL17, CXCL8 and CXCL13. The cascade of reactive oxygen species (ROS)-MAPK (ERK and p38)-NF-kB-mediated signaling was shown to be involved in the upregulation of M-CSF and CCL2 by rSAA in HCC cell lines. Furthermore, through FPR2 knockdown, using short-hairpin RNA, the authors demonstrated the involvement of FPR2 in rSAA-induced M-CSF and CCL2 expression in HCC cell lines (63). On the contrary, we recently demonstrated that pure rhSAA1.1 as a single stimulus does not modulate the macrophage phenotype in vitro (7). Nonetheless, Wang et al. demonstrated, through in vivo application of a specific anti-SAA antibody, modulation of macrophage polarization by A-SAA during hepatic inflammation in mice (64). The data of Wang et al. and Li et al. suggest that SAA might indirectly regulate macrophage polarization either through induction of immune modulators or through its cooperation with other immune modulators, as has been reported for interferon (IFN)-γ and LPS in the induction of M1 macrophages. Alternatively, SAA may exert a priming effect thereby augmenting the action of immune modulators. This has been described for granulocyte-macrophage colony-stimulating factor (GM-CSF) and M-CSF which prime macrophages towards an M1 or M2 phenotype, respectively (65).

Parathyroid hormone (PTH) regulates bone remodeling via its GPCR, parathyroid hormone 1 receptor (PTH1R), expressed on the surface of osteoblasts and osteoclasts, thereby playing a regulatory role in bone formation and resorption, respectively. Through its activation of PTH1R, PTH stimulates bone anabolism via cyclic adenosine monophosphate (cAMP) signaling. Conversely, to promote bone catabolism, PTH acts on osteoblastic lineage cells upregulating the expression of receptor activator of nuclear factor kappa-B ligand (RANKL) which promotes osteoclast differentiation from hematopoietic precursor cells (66). In the presence of Cox2, RANKL-treated bone marrow macrophages (BMMs) secrete a factor that impedes PTH-induced cAMP-mediated differentiation of primary osteoblasts (67). To further elaborate on the mechanism of RANKL-mediated bone resorption, Choudhary et al. revealed murine SAA3 as the PTH inhibitory factor. Indeed, WT RANKL-treated BMMs displayed a marked upregulation of SAA3 expression in a prostaglandin E₂ (PGE₂)-dependent manner, an effect that was not observed by Cox2 KO BMMs. To corroborate the role of SAA3 in impeding PTH-induced bone anabolism, SAA3 expression in BMMs was knocked down using short hairpin RNA, upon which the stimulatory effect of PTH on cAMP in primary osteoblasts was restored. Furthermore, to elucidate the receptor utilized by SAA during this process, the authors could reverse the inhibitory effect of rhSAA on PTH in primary osteoblasts using the selective FPR2 antagonist WRW4 (68).

Anthony et al. demonstrated a correlation between the number of infiltrating neutrophils and the degree of SAA staining in lung tissue derived from COPD patients undergoing treatment for solitary peripheral carcinoma. Using an animal model where Balb/c mice were subjected to chronic intranasal administration of rhSAA, a direct relationship between rhSAA levels and pulmonary neutrophil infiltration was demonstrated. Furthermore, 6 hours following rhSAA administration, the levels of T-helper (Th) 17-polarizing cytokines namely IL-6, IL-1β and IL-23 peaked which was synonymous with an upregulated expression of IL-17A. The main cellular source of rhSAA-induced IL-17A were CD4⁺ T-cells, γδ T-cells, and Epcam⁺ CD45⁻ epithelial cells. To elaborate on the pathway leading to neutrophil recruitment, mice were treated with a neutralizing antibody against IL-17A, whereupon a downregulation of the murine neutrophil chemoattractant CXCL1 was observed. This occurred in congruence with reduced pulmonary neutrophil infiltration. Furthermore, the administration of 15-epi-LXA₄ could avert the inflammatory response mediated by rhSAA, thereby suggesting the involvement of FPR2 in the A-SAA-IL-17A axis (69). Following up on these findings Bozinovski et al. proposed that FPR2 could be a potential therapeutic target in the treatment of COPD (70).

Upregulated expression of A-SAA has been documented in tumors of varying origins including uterine carcinoma, lung cancer and nasopharyngeal carcinoma (71–73). In fact, tumor cells serve as a source of A-SAA during early carcinogenesis when the APR has not yet been activated (74). Moreover, a meta-analysis demonstrated a link between elevated A-SAA levels and diminished disease-free survival, progression-free survival and overall survival in patients suffering from distinct solid tumors (75). Multiple studies have pointed towards the role of A-SAA in various cancer-promoting functions. A-SAA was shown to contribute towards colitis-associated cancer in mice through enhanced inflammatory cytokine expression and macrophage infiltration (76). In addition, the role of SAA1 in pancreatic cancer was demonstrated wherein the knockdown of SAA1 in a human pancreatic cell line (PANC-1) improved response to chemotherapy, reduced epithelial-to-mesenchymal transition and reduced invasive capacity (77). Moreover, in a recent review, Fourie et al. discussed the contribution of SAA to breast cancer pathology via its role in inflammasome activation (78). However, it is currently unclear which receptor is utilized by A-SAA during some of these cancer-promoting functions. Nonetheless, FPR2 is a potential suspect. Indeed, FPR2 has been linked to multiple cancers (79–82) and has even been discussed as a potential therapeutic target (83).

PTM of inflammatory proteins serves as a way to modulate the instigated response. In particular, proteolytic processing is of relevance within an inflammatory setting. One such example is the activation of IL-1. Indeed, IL-1 is expressed as a precursor protein, which needs to undergo proteolytic processing to generate the active endogenous pyrogen (5, 84). On the other hand, the proteolytic processing of chemokines, which are not expressed as precursor proteins, serves to modulate their biological activity leading to divergent outcomes on their function (85). In comparison to chemokines, the study of PTM of A-SAA is still in its infancy with only a few reports on its processing. Stix et al. were the first to reveal the susceptibility of plasma-derived A-SAA to undergo proteolytic processing by MMPs, namely MMP-1, -2 and -3 which cleave intact A-SAA at multiple positions (86). Further studies by van der Hilst et al. demonstrated the differential capacity of MMP-1 to cleave different A-SAA isoforms, where rhSAA1.5 was observed to be more resistant to proteolytic processing by MMP-1 in comparison to rhSAA1.1. The cleavage sites reported by van der Hilst et al. were similar to those described by Stix et al. (87). We recently demonstrated the susceptibility of rhSAA1.1 to undergo proteolytic processing by MMP-9, which shared cleavage sites with MMP-1, -2 and -3 in this APP (35). Moreover, the processing of rhSAA by cathepsins B and L has been evidenced by Röken et al. Interestingly, those proteases produced highly distinct rhSAA fragments. In fact, cathepsin B was found to possess endoproteolytic, carboxypeptidase and minor aminopeptidase activity; whereas, cathepsin L was shown to possess only endoprotease activity (88). Besides cathepsins B and L, cathepsin D also cleaved rhSAA1.1. Furthermore, the incubation of rhSAA1.1 with a splenic extract derived from RA patients induced N-terminal processing which was inhibited following the addition of the aspartate protease inhibitor pepstatin. As such, an aspartate protease, presumed to be cathepsin D, was implicated in this amino-terminus processing of A-SAA (89). Moreover, the capacity of elastase to clip rhSAA1.1 was also demonstrated. Elastase processed rhSAA1.1 at its N- and C-terminus leading to the generation of a range of different peptides (90). Table 2 gives an overview of the A-SAA-derived cleavage products generated following incubation with proteases.

The initial discovery of SAA was in fact evoked by the identification of amyloid A protein, an N-terminal A-SAA-derived fragment involved in the formation of amyloid A plaques, characteristic of amyloid A amyloidosis, providing early evidence that A-SAA undergoes PTM (91). In line with this, modified A-SAA variants have been detected in patients suffering from diseases with variable underlying inflammatory etiologies. Early studies by Migita et al. utilized an immunoblot analysis approach to discern modified A-SAA in sera derived from RA patients. In addition to the full-length protein, two distinct A-SAA peptides, a 6.0-kDa and a 4.5-kDa peptide were detected. Interestingly, the ratio of abundancy of SAA-derived fragments to total SAA was higher in patients suffering from amyloidosis (92). Yassine et al. characterized the PTM of A-SAA polymorphic variants in the plasma of type 2 diabetes mellitus patients. Removal of the N-terminal arginine or the arginine-serine dipeptide resulting in SAA1.1(2-104 and 3-104), SAA1.3(2-104 and 3-104) and SAA2.1(2-104 and 3-104) was detected. The ratio of abundancy to the intact variant was lower in diabetics in comparison to non-diabetics ( Table 3 ) (93). Furthermore, Tolson et al. identified three N-terminally processed variants of SAA1 including SAA1.1(2–104), SAA1.1(3-104) and SAA1.1(5-104) in the sera of renal carcinoma patients. Contrary to the observations made by Yassine et al., Tolson et al. did not detect N-terminally truncated A-SAA variants in healthy controls (94). Furthermore, Baba et al. described N-terminal truncation of A-SAA in non-amyloidogenic patients with an activated APR suggesting that this PTM is not unique to a particular disease aetiology (102). Also Trechevska and colleagues detected, N-terminal truncation of A-SAA in healthy populations, wherein they reported the truncation of up to five N-terminal amino acids of SAA1.1, SAA1.3, SAA2.1 and SAA2.2, thereby indicating that N-terminal processing of A-SAA occurs outside the context of disease (95). Besides N-terminal truncation, there is growing evidence indicating the occurrence of A-SAA-derived C-terminal peptides under inflammatory conditions. SAA1(92–104) was identified as a prognostic marker in the sera of renal carcinoma patients by Wood et al. (97). Furthermore, SAA1(98–104) has been detected in the synovial fluid of RA patients (98). SAA1(34-104) has been detected in the plasma of patients suffering from Kawasaki disease. Although detected in patients with the subacute disease, plasma levels of SAA1(34-104) were higher in patients at the acute stage of Kawasaki disease and the peptide was no longer detectable upon resolution of the disease (96). Evidence supporting the in vivo occurrence of A-SAA C-terminal peptides has also been reported in cows. De Buck et al. recently isolated bovine SAA1(46-112) from fetal calf serum, based upon biological activity, and subsequently characterized the bovine fragment (34).

The influence of naturally occurring N-terminal truncation on the biological activity of intact A-SAA has not yet been investigated; however, other A-SAA-derived peptides have been studied. In 1998, intact rhSAA and multiple synthetic SAA1-derived peptides were investigated for their modulatory effect on N-formylmethionyl-leucyl-phenylalanine (fMLP)-induced neutrophil functions. Both intact rhSAA and A-SAA(1-14) were found to inhibit fMLP-induced myeloperoxidase release and neutrophil-directed migration suggesting an anti-inflammatory effect relayed by SAA and its derived N-terminal peptide (100). Nonetheless, a year later, SAA was identified as an FPR2 ligand (21). As such, the inhibitory effect brought about by rhSAA and A-SAA1(1-14) on the FPR2-mediated fMLP functions is likely due to receptor desensitization, thereby suggesting that A-SAA1(1-14) might be a potential FPR2 ligand. More recently, the inhibitory effect of synthetic SAA1 fragments (29-104, 27-90, 27-72, 11-72, 11-68 and 11-58) on the LPS-mediated inflammatory response in macrophages was demonstrated (99). Furthermore, Yavin et al. demonstrated upregulated expression of IFN-γ following the stimulation of human T-lymphocytes with SAA1(98-104). Moreover, SAA1(98-104) bound to T-lymphocytes in a specific and saturable manner (98) ( Table 3 ). Finally, A-SAA peptides play a role in chemotaxis by synergizing with chemokines to further enhance leukocyte recruitment to the site of inflammation (vide supra) (34, 35). Apart from the aforementioned studies, very little has been revealed regarding the biological activity conveyed by A-SAA proteolytic fragments.

Allosteric modulation of protein function was initially described in relation to enzymes and is now considered to play an integral role in the functionality of other proteins, among which are GPCRs (103). Allosteric ligands bind to active sites on receptors that are spatially and topographically distinct from those of the primary, orthosteric ligands. As such, the surface of GPCRs may reveal additional binding sites that have yet to be identified. Classically defined, an allosteric modulator is a ligand that does not play a role in receptor behavior in the absence of its orthosteric agonist, but rather serves to amplify the response generated by its orthosteric partner. On the contrary, an allosteric inhibitor locks the receptor in a conformation that inhibits the binding or the activity of its traditional orthosteric ligand (104). Nonetheless, allosteric agonists, molecules which can render a GPCR in its active state in the lack of the presence of an orthosteric ligand, have also been described (105). The nature of the interaction between A-SAA and FPR2 has been briefly investigated by Bena et al. (106). Using chimeric FPR2 clones, rhSAA was shown to engage extracellular loops 1 and 2 on FPR2 when inducing calcium fluxes. Extracellular loop 3 and the N-terminus of FPR2 were found to be dispensable for this SAA-mediated signaling (106). Recent cryogenic electron microscopy investigating the structure of the signaling complex of WKYMVm and FPR2 revealed the interaction of the peptide with extracellular loops 1 and 2 and the transmembrane domains 3-7 (107). Being a potent pan-agonist at FPR2, WKYMVm is regarded as an orthosteric FPR2 ligand (108–110). Although A-SAA and WKYMVm seem to share similar binding sites on FPR2, there is evidence to suggest that A-SAA does not bind the orthosteric binding site of FPR2. We and others have demonstrated the lack of competitive binding between A-SAA variants and WKYMVm (7, 45), as such, it is plausible that A-SAA functions as an allosteric FPR2 ligand.

In the late 1980s, circular-dichroism studies indicated that murine SAA2 and human SAA1 are helical in nature (111). In line with this, Lu and colleagues recently revealed the crystal structure of human SAA1.1, and showed that SAA1.1 is composed of a four-helix bundle fold stabilized by the C-terminal tail. In addition, the tendency of this A-SAA variant to form oligomers in solution, most notably a hexamer, was demonstrated; nonetheless, SAA1.1 was also found to exist as a monomer and dodecamer (112). Moreover, the capacity of murine SAA2.2 to exist in multiple oligomeric states, primarily as an octamer and a hexamer, has been demonstrated (113, 114). However, it is currently unknown whether A-SAA carries out its biological role as a monomer or rather through a higher oligomeric state. To render the matter more complex, human A-SAA and its murine counterpart have been described as intrinsically disordered proteins (IDP) (115, 116). Owing to their conformational plasticity, IDPs contain multiple motifs for interaction with their physiological partners, thereby allowing allosteric binding and regulation of distinct cellular signaling pathways (117). The structural instability of A-SAA may explain the capacity of this APP to instigate a wide array of biological functions. It is currently unknown whether the different oligomeric forms acquired by A-SAA play a role in the differential activation of FPR2. Moreover, since SAA is an IDP, the possibility that it may bind FPR2 in both its allosteric- and orthosteric-binding sites is not out of question. Indeed, the dual binding of FPR2 by A-SAA may explain the discrepancy observed in the literature regarding A-SAA-mediated functions. Indeed, Wold et al. described an alternative allosteric modulator-receptor complex instigating differential biology when compared to the native receptor (105). As such, further research is required to ascertain the structural folding of A-SAA when exerting its FPR2-mediated functions. Allosteric modulation of FPR by ligands other than A-SAA has been discussed in this review (vide infra).

Very interesting is the number of emerging reports that have delineated the role of A-SAA in orchestrating disease pathology through the KO of A-SAA in animal models of disease (6). For instance, Sano and colleagues described the role of SAA1/2 in regulating the T-helper 17 (Th17) cell-mediated immune response following colonization of mice with segmented filamentous bacteria (SFB). SAA1/2 was shown to not only upregulate the expression of IL-17A and IL-17F by Th17 cells but also to enhance the proliferation of IL-17-expressing RORγt⁺ Th17 cells (118). This is in agreement with the findings of Anthony et al. who demonstrated the role of the A-SAA-IL-17A axis in COPD via FPR2 (vide supra) (69). Moreover, the in vivo chemoattractant role of A-SAA was demonstrated within the context of apical periodontitis by Hirai et al. Indeed, following pulpal infection, SAA1/2 KO mice displayed a reduction in myeloid cell infiltration when compared to wild-type mice (119).

Although A-SAA is predominantly discussed in the literature as a pro-inflammatory mediator, a growing body of research suggests that a rather dual function is relayed by A-SAA during inflammation. For instance, Murdoch et al. showed that despite being essential for the maturation and recruitment of neutrophils in zebrafish, SAA also serves to confine neutrophil-mediated inflammation by reducing neutrophil bactericidal activity and the expression of inflammatory markers (120). Nonetheless, for the majority of these A-SAA-mediated roles, no receptor interaction has been assigned. In addition, as the biological characterization of A-SAA has been hampered by low-quality commercial A-SAA preparations (vide supra), the biological function of A-SAA is yet to be accurately elucidated. As such, further exploration of the A-SAA-FPR2 interaction is likely to yield insight into the role of this interaction in disease pathology. Indeed, due to its promiscuous cellular expression profile, FPR2 is linked to various pathophysiological processes mainly through its role in cellular chemotaxis. Moreover, its capacity to recognize structurally diverse ligands has implicated FPR2 in the pathology of diseases displaying variable underlying etiologies (121–123). Apart from its role in disease pathology, FPR2 is also described to instigate the resolution of disease (124). Hence, FPR2 might serve as a potential drug target. GPCRs constitute the largest class of druggable proteins within the human genome and remain a promising target for future drug development (125). Indeed, pharmaceuticals targeting GPCRs account for approximately 35% of FDA-approved drugs (126). Targeting chemoattractant GPCRs has been successful in some cases. For instance, Maraviroc, an allosteric modulator of the chemokine receptor CCR5, is currently marketed as an anti-HIV agent (105). Taken together, this leads us to the question whether A-SAA-derived peptides may serve as potential FPR2-targted therapeutic agents. Peptide drugs offer certain advantages. They display higher specificity and efficacy and lower immunogenicity and are hence safer with a rather tolerable side effect profile. Moreover, peptide drugs are cheaper to produce than, e.g. monoclonal antibodies. One major drawback observed with peptide drugs is the lack of in vivo stability conferred by secondary and tertiary structures leading to poor bioavailability. Nonetheless, extensive research continues to be carried out to produce peptide-based drugs with optimal pharmacokinetics (127).

The identification and characterization of A-SAA-derived peptides is certainly not yet fully accomplished. We hypothesize that under inflammatory conditions, upregulated proteases process A-SAA, thereby generating a range of peptides, which display variable functions to modify the inflammatory reaction ( Figure 1 ). As mentioned, SAA1.1-derived C-terminal peptides have been shown to magnify chemokine-induced chemotactic responses via their binding to FPR2 (vide supra). As such, further research is warranted to identify novel A-SAA-derived peptides, which could be further developed as drugs following their biological characterization. For instance, Jana et al. demonstrated the capacity of SAA1(1–5) to destabilize SAA fibrils, thereby placing the peptide as a promising lead drug for the treatment of amyloid A amyloidosis, a common secondary disease occurring within the context of uncontrolled inflammation (101).

Concerning FPR2, a growing body of reports have discussed its role as a potential therapeutic target using peptide-based drugs. The FPR2 ligand CGEN-855A, a 21 amino acid peptide, was demonstrated to provide an anti-inflammatory effect in a zymosan-induced air pouch model of inflammation and provided cardioprotection against ischemia reperfusion-induced injury via reduced PMN infiltration to the injured organ (128). CGEN-855A is reported to have reached preclinical development (124). Moreover, in a recent review by Ma et al., the orthosteric FPR2 agonistic hexapeptide, WKYMVm, has been discussed as a potential therapeutic to treat inflammation, cancer, angiogenesis, tissue repair, neurodegenerative diseases, vaccine development, insulin resistance and osteolytic diseases (83).

Despite the clinical success of orthosteric GPCR ligands, one major issue observed with such therapeutics remains to be the high degree of homology in the orthosteric binding sites of closely associated GPCRs thereby leading to off-target side effects. On this account, current research is directed towards the design and development of allosteric modulators of GPCRs (129). One assumed outcome of investigating A-SAA-derived peptides and their link to FPR2 activity is the identification of novel allosteric modulators of FPR2. The notion of allosteric modulation of FPR2 by peptides has been previously discussed in the literature. Zhang et al. demonstrated biased allosteric modulation of FPR2 by Ac_2-26, an N-terminal peptide of AnxAI, wherein the AnxAI-derived peptide-induced FPR2 conformational changes that are thought to lower the binding and to inhibit the pro-inflammatory effect induced by WKYMVm. In sharp contrast, Aβ₄₂, a peptide detected in the neuritic plaques in Alzheimer’s disease, brought about conformational changes that improved the binding affinity of WKYMVm hence exacerbating its pro-inflammatory function (130).

Pepducins represent a promising class of molecules that enable the study of GPCRs and have therapeutic potential. Pepducins are lipopeptides with an N-terminal lipid, typically, palmitate, linked to a short peptide wherein the peptide sequence is derived from the intracellular loop of the GPCR that it is presumably thought to target. The lipid moiety instigates the translocation of the pepducin across the plasma membrane where the pepducin functions to modulate receptor signaling (131). F2Pal₁₆, derived from the third intracellular loop of FPR2, was demonstrated to activate neutrophils leading to enhanced inflammatory function (132). Nonetheless, the notion that pepducins will selectively target GPCRs from which they are derived does not always apply. In a report by Winther et al., an FPR1-derived pepducin, F1Pal₁₆ demonstrated an inhibitory effect on FPR2-mediated neutrophil inflammation in response to a diverse set of ligands (MMK-1, WKYMVm, F2Pal₁₀ and PSMα2) specific to the receptor suggesting that the pepducin displays a general inhibitory effect on FPR2. Whereas F1Pal₁₆ was shown to compete with conventional FPR2 agonists, namely WKYMVm, for binding, it is currently unclear whether the inhibitory effect is mediated via competitive binding or rather through allosteric modulation inducing conformational changes that reduce the binding of the conventional agonist (133). Moreover, a protease-activated receptor (PAR) 4-derived pepducin, P4Pal₁₀, was found to inhibit FPR2-mediated Gαi signaling in neutrophils following stimulation with WKYMVm. In addition, the pepducin did not compete with WKYMVm suggesting that its inhibitory effect is not relayed via direct interaction with the conventional agonist binding site (134). Finally, allosteric modulation of FPR2 on neutrophils using pepducins has also been discussed as a potential antibacterial strategy to solve the ongoing issue of antibiotic resistance (135).