Discerning the promising binding sites of S100/calgranulins and their therapeutic potential in atherosclerosis

Abstract

Introduction:

Atherosclerosis is a chronic inflammatory disease in which the members of S100 family proteins (calgranulins) bind with their receptors, particularly receptor for advanced glycation end products (RAGE) and toll-like receptor-4 (TLR-4) and play a key role in the pathogenesis and progression of disease. Thus, these proteins could be considered as potential biomarkers and therapeutic targets in the treatment of atherosclerotic inflammation.

Areas covered:

This review summarizes the pathology of S100A8, S100A9, and S100A12 in the development of atherosclerosis and reveals key structural features of these proteins which are potentially critical in their pathological effects. The article focuses on the translational significance of antagonizing these proteins by using small molecules in patent literature, clinical and preclinical studies and also discusses future approaches that could be employed to block these proteins in the treatment of atherosclerosis.

Expert Opinion:

Based on the critical role of S100/calgranulins in the regulation of atherosclerosis, these proteins are potential targets to develop better therapeutic options in the treatment of inflammatory diseases. However, further research is still needed to clarify their exact molecular mechanism by analyzing their detailed structural features that can expedite future research to develop novel therapeutics against these proteins to treat atherosclerotic inflammation.

1. INTRODUCTION

Cardiovascular disease (CVD) is the leading cause of death and is currently a major problem worldwide. The main pathological feature responsible for the pathogenesis of CVD is the narrowing of the arterial lumen restricting blood flow due to buildup of fats, cholesterol and and fibrous material in the innermost luminal wall of the arteries leading to atherosclerosis [1]. Atherosclerosis may affect any artery in the body, including arteries in the heart, arms, brain, pelvis, legs, and kidneys. The accumulation of fibrous material leads to the formation of occlusive atherosclerotic plaque following consequent inflammatory signals. The atherosclerotic plaques with time can become more fibrous and can accumulate more calcium mineral deposits in the arterial lumen, causes its narrowing to impede blood flow leading to tissue ischemia. Activated immune cells and activated smooth muscle cells (SMCs) are majorly associated with this disease condition [2]. The development of atherosclerotic plaque is a complex process that results from endothelial cell dysfunction, inflammatory cell recruitment, lipid accumulation, and connective tissue matrix deposition [3]. More specifically, the mechanism of atherosclerosis is a three-step process, i.e., initiation, progression, and rupture. The normal human artery is a trilaminar structural unit in which the outermost layer (adventitia) contains mast cells, nerve endings, and vasa vasorum that covers tunica media. Tunica media is the second inner layer and contains SMCs and extracellular matrix such as collagen, elastin, and various macromolecules. The innermost layer is called intima, where the development of atherosclerotic plaque occurs along with the medial layer.

Initially, low density lipoprotein (LDL) starts to accumulate in the innermost layer, which get oxidized by metal ion catalysis (Fenton reaction) and is modified to induce pro-inflammatory signals [4,5]. The integrins on classic monocytes in the blood vessels become activated and bind to adhesion molecules which are mainly expressed on the activated endothelial cells. This process results in the transmigration of circulating monocytes into the artery wall. These monocytes enter the intima and differentiate into macrophages that are thought to become foam cells by binding to oxidized LDL (oxLDL). Other cells, including T-lymphocytes which are lesser in number than monocytes in the bloodstream, B-lymphocytes and natural killer (NK) cells also enter the intima in response to the inflammation. This stimulates the migration of SMCs from media to intima. The SMCs and macrophages become engorged with lipid and contribute to lesion progression [6] (Figure 1). Most researchers considered that macrophages derived from blood monocytes are the only precursors of lipid foam cells, but in response to the inflammation metaplasia of SMCs may also give rise to foam cells [7].

Figure 1:

Role of S100/calgranulins in atherosclerosis

During the progression of atherosclerotic plaque, recruited SMCs produce extracellular matrix (ECM) molecules that contribute to make a thick intimal layer. T-cells mediators (including IFNℽ) impair the ability of SMCs to make interstitial collagen and to repair and maintain the fibrous cap. In addition, activated macrophages enhance the production of matrix metalloproteinases (MMPs) that also degrade the interstitial collagen and weakens the fibrous cap and thus plaque becomes susceptible to rupture, also known as unstable plaques [8-12]. These unstable plaques contain few SMCs with thin fibrous cap, little amount of collagen with more macrophages, while stable plaques are rich in SMCs with thick fibrous cap and collagen having few macrophages, thus less susceptible to rupture. The stability of plaques generally depends on matrix formation and ECM degradation [13].

In recent years, a variety of serum markers have been used to examine their association with atherosclerosis [14-18]. It was found that receptor for advanced glycation end products (RAGE) and toll-like receptor 4 (TLR-4) play an important role in several inflammatory conditions [19-21]. RAGE has been found in two forms: membrane bound (mRAGE) and soluble or secreted form (sRAGE), amongst which major form is the full-length protein mRAGE. Its soluble form functionally acts as a decoy for ligands [22]. Similarly, toll-like receptors also have various isoforms TLR-1 to TLR-13, amongst which TLR-4 located on the cell membrane has been found to regulate the inflammatory pathways. The increased protein levels of sRAGE and TLR-4 have been found in the serum of cardiovascular patients. Recent findings have clearly established a link between these two receptor proteins and atherosclerosis. RAGE expression is upregulated in the epithelial cells during inflammatory conditions and its soluble form (sRAGE) is secreted into the serum which is serving as a biomarker for the inflammatory conditions [22, 23].

The S100 proteins were first reported in 1965 with the characteristic feature of their solubility in 100% saturated ammonium sulfate solution at neutral pH, thus named as “S100 protein” [24]. These proteins consist of 21 members which are mainly expressed in vertebrates and perform various intracellular and extracellular functions [25]. Among all the S100 proteins, S100A8, S100A9, and S100A12 are highly associated with cardiovascular disorders [18, 26-28]. These three members of the S100 protein family are commonly referred to as S100/calgranulins. S100A8 is also known as calgranulin A and myeloid-related protein 8 (MRP-8). Similarly, S100A9 is also called calgranulin B and MRP-14. Their non-covalently combined form is also known as calprotectin (MRP8/14) which is a 24 kDa heterodimer of S100A8 and S100A9 (S100A8/A9) [29, 30]. Likewise, calgranulin C or EN-RAGE are the other names of S100A12.

2. Expression and distribution of S100/calgranulins

S100A8 and S100A9 are derived from neutrophils and macrophages that are the major source of S100A8/A9 and participate in inflammatory processes. Neutrophils contain approximately 45% of S100A8 and S100A9 out of the total cytoplasmic proteins [31]. Neutrophils and myeloid-derived dendritic cells (DCs) have sufficient storage of S100A8/A9, while monocytes have low levels of these proteins. S100A8/A9 is significantly upregulated during trauma, heat, infection, stress, and various inflammatory processes [32].

S100A12 is mainly expressed in the cytoplasm of myeloid cells and acts as an inflammatory alarmin upon secretion. Surprisingly, its gene expression is greater in the classical monocytes than in non-classical monocytes [33, 34]. Its gene expression decreases during the maturation of monocytes to macrophages [35]. The S100A12 protein expression is markedly increased at the inflammatory sites and its concentration in serum might be useful to predict the inflammatory disease condition [36-40].

3. Biological functions of S100/calgranulins

Functionally, S100A8 can bind with TLR-4 that can crosstalk in the transmembrane signaling pathways with other receptors including opioid receptor, a G-protein coupled receptor [41], while only S100A9 has the ability to bind TLR-4. Their heterodimeric complex S100A8/A9 can interact with RAGE and scavenger receptors in addition to TLR-4 [42]. The binding of the ligands like S100A8/A9 and S100A12 with RAGE initiates endothelial cells, SMCs, and inflammatory pathways to secrete pro-inflammatory cytokines, which lead to leukocyte infiltration, oxidative stress, and vascular inflammatory response contributing to fibrous cap formation (Figure 1). Additionally, serum levels of S100A8/A9 and S100A12 have been found to be significantly higher in patients with coronary artery disease (CAD) compared to healthy subjects. Also, S100A8/A9 complex appears to be associated with elevated C-reactive protein suggesting that these proteins are novel biomarkers for atherosclerosis and a predictor of atherosclerotic plaque instability [43-45]. There are few review articles on S100 family proteins by other research investigators [45, 46]. However, this article specifically addresses the role of various structural features associated with direct binding to RAGE, and critically discussed the potential significance of antagonizing these binding interactions to treat atherosclerosis. To our knowledge, this is the first report that reveals together with the detailed available information on the crystal structure of S100/calgranulins that has great potential in research advancements to inhibit their functional complex formation or to antagonize their interactions with other proteins.

This compilation thus stands at the interface of biology and structural biochemistry. For this article, relevant literature published between the years 2015-2021 was considered and downloaded from online databases, including PubMed and Science direct.com. The patent literature of 2009 onwards was collected from ‘worldwide.espacenet.com’ and ‘google patents’. To search the articles and patents with English language, various combinations of keywords were used including: ‘S100A8/A9 + atherosclerosis’, ‘Calgranulins + atherosclerosis’, ‘S100A12 + atherosclerosis’, ‘S100A8 inhibitor’, ‘S100A9 inhibitor’, ‘S100A9 inhibitor’, and ‘Calgranulin inhibitor. After significant data collection, literature was critically reviewed and summarized in a systematic manner. For the proteins structural analysis, PDB codes for crystallized structures of S100 proteins were obtained from protein data bank (rcsb.org). Structural and binding sites analysis on S100 proteins was done by using SiteMap tool using Schrodinger 2020-4.

4. Structural features of S100/calgranulins

Crystal structures of these proteins have been reported in the recent past which has accelerated the research worldwide to understand their functions at the molecular level in relation to various disease conditions. It has been found that S100A8 and S100A9 are calcium ion (Ca²⁺) binding proteins which mainly exist in the form of heterodimer. The conformation and stability of S100A8 and S100A9 drastically depend upon the metal ion binding. The binding of Ca²⁺ triggers conformational changes that allow the interactions with other proteins. Binding of Zn²⁺ additionally modulates their conformational properties. By binding to these ions, heterodimers of these two proteins assemble into highly stable heterotetrameric complex (S100A8/A9) in a Ca²⁺ and Zn²⁺ dependent manner, and then fulfill other functional properties which are absent in monomeric and dimeric form [47, 48]. More specifically, at the site of inflammation S100A8 and S100A9 are present as heterodimer at low concentration of intracellular Ca²⁺ and these heterodimers at high Ca²⁺ concentration subsequently associate to S100A8/A9 tetramers. S100A8/A9 heterodimers have high binding affinity towards RAGE and TLR-4, thus promoting inflammatory processes. The binding sites within the S100A8/A9 tetramers interphase are hidden, thereby unable to bind to RAGE and TLR-4 and restricting the biological activity of inflammatory processes [49]. The structural modeling also revealed that S100A8/A9 heterodimer interaction with their receptor proteins RAGE and TLR-4 confirms identical binding regions for the homodimer of S100A9 as well as the heterodimer of S100A8/S100A9 [49]. The mass spectrometric analysis and density gradient centrifugation data suggested that Ca²⁺ tetramer formation is highly associated with the C-terminal calcium-binding site in S100A9 [50]. So, it is critical to closely analyze the structure of the S100A9 protein.

The crystal structure of S100A9 at 2.1Å resolution was first developed analyzed in 2002 by Itou et al. [51]. Structurally, it has a molecular mass of 13kDa and consists of 114 amino acid residues. Its extended C-terminus, considerably larger than other S100 family proteins make it distinguishable. S100A9 monomer consists of four α-helices i.e., H1 to H4, which form a unit of Helix-loop-Helix motifs, also known as EF-hand motifs (EF-1 and EF-2) (Figure 2A). At the N-terminus, EF-1 hand motif consists of three parts viz. helix H1 made up of Gln7 to Ser23 amino acid residues, a Ca²⁺ binding loop-1 having Val24 to Asn33 amino acid residues and helix H2 made up of Gln34 to Asp44 amino acids (Figure 2A). Similarly, EF-2 (at C-terminal) starts with helix H3 that contains amino acid residues Glu56 to Leu66, Ca²⁺ binding loop-2 having Asp67-Arg85 amino acids. These EF-1 and EF-2 hand motifs join together with the help of a hinge region which consists of 11 amino acid residues (Leu45-Asn55). In the first crystal structure, it could not be possible to analyze the longest C-terminal region of 28 residues after Thr87 due to poor electron density in this region. The presence of the C-terminal region on S100A9 was only confirmed through mass spectrometric analysis. Calcium ion mainly binds to EF-hand motifs. In the seven oxygen atoms of EF-1, four from carboxyl groups (−COOH) of Ser23, Leu26, His28, and Thr31 amino acids, and one from water molecule bind to Ca²⁺ ions. Similarly, in EF-2, oxygen atom present in amino acids Asp67, Asn69, Asp71, and Glu78 and a water molecule bind to Ca²⁺ [51]. S100A9 also present as a dimer and its dimerization is mainly achieved by anti-parallel hydrophobic interactions between α-helix H1 and H4 of each monomer. Interestingly, various hydrophobic amino acid residues (Phe19, Tyr22, Leu32, Phe37, Leu40, Val41, Phe48, Leu49, Val58, Ile59, Ile62, Met63, Leu66, Leu74, Phe79, and Leu82) within human S100A9 are highly important for its dimeric interactions [51].

Figure 2:

(A) Crystal structure of S100A9 protein (PDB: 1IRJ) and its various domains; (B) 2D and 3D binding modes of CHAPS within the RAGE V binding domain of S100A9 protein

A zwitterionic molecule, 3-[(3-cholamidopropyl) dimethylammonium]-1-propanesulfonate (CHAPS), induces the solubilization of membranous proteins and binds with hydrophobic cluster between two monomeric units of S100A9. In 2016, Chang et al. [52] revealed the molecular interactions between S100A9-RAGE V domain and S100A9-CHAPS through Heteronuclear Single Quantum Coherence Spectroscopy (HSQC) NMR experiments. They reported a putative model of Ca²⁺ bound mutant forming a complex with CHAPS and RAGE V domain. It was clear that most of the amino acid residues of S100A9 that can interact with the RAGE V domain are present in the hinge region (Phe48, Lys50, and Glu52) and the helix H4 (Met81, Leu82, Leu86, Thr87, Trp88, and Ala89) (Figure 2A). The dissociation constant (K_D) between the S100A9 and RAGE V domain was also calculated with the help of the HSQC titration spectrum which was found to be in the micromolar range (5.7 μM).

The hydrophobic region, as mentioned above in S100A9, consisting of helices H3 and H4 regions, provides a favorable binding site for CHAPS (Figure 2). The main amino acid residues Glu52, Ile62, Met81, Leu82, Thr87, and Trp88 were found responsible for these interactions (Figure 2B). CHAPS blocks the interactions between S100A9 and RAGE V domain which could be a novel route for drug development [52].

S100A12 is also a calcium-binding protein that expresses its bioactivity after Ca²⁺ ions bind to its EF-hand domains [53-55]. It has different structural states when bound to different metal ions and thus can express different biological functions [56, 57]. The X-ray crystal structure of S100A12 has been found in two crystal forms i.e., R3 and P2₁.R3 crystal structure is a dimer of S100A12 while P2₁ is a hexamer in which three dimers are arranged to form a complex [56] (Figure 3A). Calcium-binding with S100A12 explores its specific binding sites to activate various signaling pathways and is recognized as an important biomarker for various inflammatory diseases [58-60].

Figure 3:

(A) R3 and P2₁ crystal forms of S100A12, Monomeric unit of S100A12 and its RAGE V binding domain (PDB: 2M9G); (B) 2D (above) and 3D (below) interactions of Tranilast with S100A12.

The dimer (R3) form of S100A12 is very similar to the structure of S100A9 dimer, the monomeric unit of which consists of two EF-hand motifs, four α-helices (H1-H4), two binding loops, and one hinge region (Figure 3A). EF-hand motifs also contain ion binding sites similar to that of S100A9 and having a central hydrophobic surface between both monomeric units of the dimer.

Similarly, in the P2₁ crystal form, three dimers form a hexameric complex with an external diameter of 60Å and contain dimers having a similar structure to that of its R3 crystal form. A single hexameric unit contains six Ca²⁺ ions that cause the three dimers to form a hexameric assembly [61]. The oligomeric state of S100A12 requires the presence of zinc ions [62], which means that this protein exists in a dimer form at a certain concentration of Ca²⁺ ions (1-10 mM). The putative binding sites lie on the outermost surface of this hexameric complex at which various ligands can bind to it with multimeric interactions. RAGE can bind to this surface which was confirmed in 2016 by Chiou et al. [63]. Prior to these findings Leclerc et al. had described the binding of the RAGE V domain to the S100A12 by using surface plasmon resonance (SPR) but these studies could not clear the molecular mechanism [64]. To understand the molecular mechanism of S100A12-RAGE bindings, Chiou et al. carried out 2D NMR experiments (HSQC) [63]. They also revealed that Tranilast (an anti-allergic drug; Figure 3) inhibited these interactions between S100A12 and RAGE V binding domain. From the NMR experiments, they revealed that the hydrophobic residues Lys46, Asn47, Ile48, Ile57, Gln59, Leu77, Ala79, Ile80, and Ala81 formed a hydrophobic region located at the hinge, H3 and H4 regions of S100A12 which is exposed to various interactions with RAGE V binding domain. Tranilast was also found to bind with the same hydrophobic interface between the S100A12 and RAGE V binding domain and inhibit their interactions. This drug molecule can bind to Glu9, Thr44, Ile48, Lys49, Leu77, Ala79, Ala81, Lys84 and Lys91 amino acid residues amongst which four (Ile48, Leu77, Ala79, Ala81) are also present at the interface of S100A12-RAGE V (Figure 3B). Tranilast is a hydrophobic molecule containing benzyl group. This might be the reason for its binding particularly to the hydrophobic surface of S100A12, which was corroborated through its hydrophobic interactions with Leu77 and Ala79 amino acid residues of S100A12.

The dissociation constant of 6.1 ± 1.4 μM was found for S100A12 binding with Tranilast and 3.1 ± 1.4 μM for S100A12 binding with RAGE V domain which means that Tranilast may be a prominent therapeutic molecule that is stable and can significantly inhibit these protein interactions [63].

5. S100/Calgranulin as potential targets in the treatment of various inflammatory conditions including atherosclerosis

S100A8/A9 and S100A12 have been found to be crucial proteins during the inflammatory processes, especially in atherosclerosis which suggests that therapeutic strategies targeting these molecular proteins could be beneficial in the treatment of inflammatory diseases over the traditional therapies.

In the inflammatory processes, Quinoline-3-carboxamide derivatives have been used for the last 30 years with encouraging outcomes in various inflammatory conditions like multiple sclerosis, systemic lupus erythematosus (SLE), and type 1 diabetes [65-67]. Quinoline compounds with completed clinical trials include Paquinimod, Laquinimod, and Tasquinimod (Figure 4). Interestingly, the previous compound used to treat multiple sclerosis was Linomide (Figure 4), for which phase III clinical trials were terminated due to its unwanted cardiovascular events (myocardial infarction; MI) [68]. The reason behind this adverse effect is still a quest. The exact mechanism of action for its beneficial effects is also still elusive but some experimental data showed that Linomide inhibits the endothelial cell proliferation and reduces the secretion of angiogenic factor TNF-α by tumor associated macrophages (TAMs) [69]. The newer compounds (Paquinimod, Laquinimod, and Tasquinimod) are the derivatives of Linomide in which methyl group on N-carboxamide is replaced with ethyl group. The in vivo data with these compounds showed that they did not exhibit a pro-inflammatory response of MI [70, 71]. Currently, Paquinimod (a potent S100A9 inhibitor), also known as ABR-215757, has been approved in the United States as an orphan drug for systemic sclerosis which has also been found to be safe for patients with lupus. In diabetes-related CVDs, Paquinimod showed significant vascular protection by targeting the S100A8/A9 complex [72].

Figure 4:

Various compounds that bind to S100/Calgranulins

Among these quinolone derivatives, Tasquinimod (Figure 4), an oral drug molecule, binds to S100A9 and S100A8/A9 complex in the presence of Cu²⁺ and Zn²⁺ ions and blocks the binding interactions between S100A9 and RAGE. The in vivo studies with Tasquinimod revealed that it blocks the further release of TNF-α [73]. During the development of atherosclerosis TNF-α probably initiates the inflammatory cascade inside the arterial wall and promotes the endothelial cell injury through recruitment of neutrophils, ultimately mediating tissue destruction. Therefore, Tasquinimod could also help to reduce the atherosclerotic inflammation [74].

Besides these, various anti-allergic drugs such as Tranilast, Amlexanox, and Cromolyn (Figure 4) can also bind to S100A12 but have not been tested so far in atherosclerosis either in experimental animal model or clinical studies in humans [75]. These small molecules could be of great potential to treat atherosclerosis as supported by their mechanism of action to bind with S100 proteins.

Blockade of S100A8/A9 complex secretion during sepsis may be a prominent therapeutic approach, as it has been found that surviving patients were shown to have a decreased level of S100A8/A9 complex compared to that of non-survivors [76]. Similarly, targeting S100A9 could treat lung inflammation and its associated disease during infection with influenza A virus (IAV) [77].

Interestingly, in rheumatoid arthritis (RA) patients, S100 proteins can be targeted in therapeutic approaches. Treatment with anti-S100A9 antibody improves the condition of RA patients by 50% in terms of clinical scoring [78]. In murine models of RA, inhibition of the S100A8/A9 complex can ameliorate serious inflammation, therefore, S100 proteins could be potential targets for human RA patients [79]. The inhibition of S100A8 and S100A9 could also be good targets to treat obesity-induced chronic inflammation, by inhibiting the further release of TLR-4 and IL-1β [80-82]. There is a recent report that provides novel evidence for the primary role of neutrophil derived S100A8/A9 proteins confirming the nature of the ensuing inflammatory response after myocardial injury. The study highlights that therapeutic strategies aimed at disruption of S100A8/A9 signaling or their downstream mediators in neutrophils suppress granulopoiesis and thus can improve cardiac function in patients with atherosclerosis [83].

Furthermore, the potential involvement of S100/calgranulins in atherogenesis, ischemia-associated myocardial inflammation, plaque vulnerability, and heart failure suggests that these proteins might serve as therapeutic targets in CVD [84]. S100A8/A9 can be potential targets for ruptured intracranial aneurysms and acute coronary syndrome [85, 86]. Synthetic derivatives of quinolone-3-carboxamide have been found to reduce inflammation, atherosclerotic lesion, and vulnerability features in S100A12 transgenic hyperlipidemic ApoE^−/− mice [87]. Regulation of S100A8-SAA3-LOX-1 cascade may improve the stability of atherosclerosis in CVD and decrease the chances of further cardiovascular events [88, 89].

In Alzheimer’s disease (AD), upregulation of particularly S100A9 protein may increase the β-amyloid production and their deposition in the brain by stimulating neuro-inflammatory cascade. Therefore, blocking the S100A9 in the chronic inflammatory phase of AD could be a better treatment approach for AD [90-92].

Kawano et al. also designed an antiplatelet vaccine that can inhibit S100A9/CD36 signaling in the mice model [93, 94]. These investigators reported that the antithrombotic effect of the vaccine lasts for 84 days after the last day of vaccination which could support it as a potential treatment therapy for inflammatory diseases. Inhibition of the anti-thrombotic effect with an antiplatelet vaccine that inhibits S100A9/CD36 signaling is central to the pathophysiology of acute coronary syndrome.

In 2019, Duvetorp and colleagues [95] reported that narrowband ultraviolet B (UVB) treatment is also a better option to treat inflammatory disease conditions through targeting S100A8/A9. In their study, UVB treatment was given to the patients with chronic plaque type psoriasis, and S100A8, S100A9 gene, and S100A8/A9 protein expression was analyzed. UVB treatment significantly reduced gene as well as protein expression only in skin lesion but protein level in serum was found unchanged. This suggests that serum S100A8/A9 level does not associate with psoriasis skin keratinocytes [95].

Recently in 2020, Guo et al. [96] found that Paquinimod also could reduce the inflammatory response in SASR-CoV-2 infected animals. They reported 100% survival of lethal coronavirus infected mice. Number of neutrophils that trigger uncontrolled inflammation was dramatically upregulated in COVID-19 infection. Treatment with Paquinimod reduced neutrophilia and induced antiviral responses, suggesting the key roles of S100A8/A9 and noncanonical neutrophils in the pathogenesis of COVID-19, highlighting new opportunities for therapeutic intervention [96]. It was also found that in in vitro experiments, Paquinimod also binds to S100A12 with a different kinetic profile than S100A9 [87]. Intriguingly, when S100A12 transgenic ApoE^−/− mice having atherosclerotic disease were treated with Paquinimod, a 20% reduction in lesion size was found compared to placebo-treated mice. Additionally, expression of leukocyte production was reduced to 55-60% in Paquinimod-treated mice [97]. Interestingly, S100A12 also antagonize the interactions between S100A9 and RAGE V domain with micromolar range suggesting that S100A12 itself could also be used as a therapeutic protein against inflammatory diseases, but it was only confirmed so far in cancer treatment [98]. A similar approach can be used in preventing atherosclerosis as laquinimod has been found to reduce the expression of inflammatory cytokines and chemokines, such as IL-6, HMGB1 and MCP-1, on human aortic endothelial in a TNF-α-induced atherosclerotic microenvironment [99].

Similarly, one anti-inflammatory drug currently being tested in various clinical trials for atherosclerosis is Methotrexate (Figure 4). The anti-inflammatory effect of this drug is multifactorial and reduces the circulating level of S100A12 in patients with systemic arthritis [36].

6. PATENT LITERATURE

6.1. S100/Calgranulins antagonists against atherosclerosis

Palefsky and colleagues in 2009 [100] reported compositions which can treat inflammatory disorders including atherosclerosis by using nucleic acid sequence encoding mutant human S100A8 and S100A9 proteins. This mutation was done in order to hamper the dimerization process of these proteins (responsible for inflammatory processes). Mutations were done by the replacement of Cysteine amino acid at position 42 with Alanine in human S100A8 and replacement of Methionine amino acid at position 61, 81 and 83 with Alanine in human S100A9. The administration of these mutated proteins significantly reduced LPS-induced inflammatory response when tested in a rat model of inflammation, suggesting that targeting calgranulin is important in the treatment of atherosclerosis. More specifically, sterile air was injected subcutaneously to the back of animals to form an air pouch and then animals were used for the assessment of leukocyte recruitment. LPS and mutated protein treatment was given and after three hours leukocytes were harvested from the pouch, processed and quantified by flow cytometry. It was found that leukocyte migration was significantly abrogated with the modified S100 proteins. Similarly in 2015, Roth and co-workers published a patent claiming antibodies that can inhibit the formation of complex between S100A8 and S100A9 [101]. These antibodies can bind to amino acid positions 55-71 of human S100A8 and similarly can bind with human S100A9 at the amino acid positions 63-79 or 73-85. These amino acid positions are mainly responsible for their dimerization process. Reported antibody consists of a peptide of 15 amino acid sequences i.e. KGLSFEEFIMLMARL. In vitro and in silico studies on this peptide revealed that it can be used as effective therapeutics in the treatment of inflammatory disorders associated with S100A8 and S100A9 proteins [101].

Inhibition of the S100A9 protein by small interfering RNA (siRNA) when transfected into human macrophages showed decrease in calcification in atherosclerotic mice. Aikawa and co-workers in 2015 reported a reduced incidence of calcification in siRNA-treated mice compared to WT mice [102]. Similarly, Foell and colleagues published a patent in 2016 claiming that a monoclonal antibody (mAb:11G7A1) that specifically binds to hexameric complex of S100A12 inhibits its interactions with TLR-4. It was found that 66 nM concentration of this antibody can effectively neutralizes TNF-α released by LPS-induced human monocytes. Accordingly, this antibody could have a potential to treat immune-inflammatory disorders including atherosclerosis [103].

In 2010, Mjalli and colleagues [104] published a patent claiming the modulatory effect of carboxamide derivatives in the interactions between S100/calgranulins (specifically EN-RAGE) for the treatment of inflammatory diseases. It was found that compound 1 (Figure 5) significantly inhibited the interactions between S100A12 and RAGE with the IC₅₀ value below 0.5 μM, suggesting that compound 1 could be a hit lead for further optimization in the development of therapeutics for inflammatory diseases.

Figure 5:

Synthetic compounds as antagonists of S100/calgranulins

Since S100A12 is not present in mice, transgenic mice have been developed expressing human S100A12 in SMCs. ApoE-deficient S100A12 transgenic mice showed 1.4-fold increase in atherosclerotic plaque, necrotic core, calcified plaque area and significant reduction in extracellular matrix compared to ApoE-deficient wild type mice [105]. The underlying process of promoting the vascular calcification could be mediated by binding of S100 proteins to a dominant group of proteins in the matrix vesicles [106, 107].

It has also been revealed that in streptozotocin-induced diabetic mice, islet cell destruction increased the number of circulating proinflammatory monocytes and neutrophils which was found due to the overproduction of granulocyte macrophage progenitors and common myeloid progenitors. This myelopoiesis in diabetic mice was found to be driven by S100 proteins and RAGE as confirmed by wild type and myd88 deficient bone marrow, while RAGE-deficient cell does not show excessive proliferation. It was found noteworthy that in LDLR−/− atherosclerotic mouse model, anti-diabetic therapy corrected the myelopoiesis, reduced circulating Chi cells and their entry into atherosclerotic lesions, ultimately reducing the atherosclerotic lesions [108]. These experimental models reveal that hyperglycemia promotes monocytes induction as a contributing factor in the pathogenesis of atherosclerosis and offers a rationale for the development of various strategies aimed at therapeutic targets linking hyperglycemia to S100 proteins and RAGE-mediated atherosclerosis. In postulation, Dutta et al. confirmed enhanced atherosclerosis after experimental myocardial infarction in ApoE null mice and suggested the pathway as a potential therapeutic target to attenuate the progression of atherosclerosis after myocardial infarction [109].

6.2. S100/Calgranulin antagonists against various inflammatory diseases other than atherosclerosis

Various compounds and compositions including antibodies targeting S100/calgranulins have been tested so far by various research groups in several inflammatory diseases other than atherosclerosis which indicate the imperative role of these proteins in a multitude of inflammatory comorbidities.

In 2013, Tessier and colleagues published a patent describing that inhibiting the S100A8, S100A9 and S100A12 by using antibodies could be an effective treatment for the inflammation in gouty arthritis (a condition of crystallization of monosodium urate: MSU in articulation) [110]. These investigators reported that the injection of MSU stimulates the release of calgranulins in the air pouches of mice which increases neutrophil accumulation at the site of inflammation. The treatment with polyclonal antibodies against these proteins significantly reduced the accumulation of neutrophils suggesting the beneficial role of inhibiting calgranulins in the inflammation related to arthritis [110]. Likewise, a research group from Laval University reported significant proliferation of bone marrow cells when incubated with S100A8 and S100A9. Proliferation was also found when bone marrow cells isolated from acute myeloid leukemia (AML) and chronic myeloid leukemia (CML) patients were incubated with these proteins. The antibodies against S100A8 and S100A9 inhibited the proliferation of leukemia cells [111]. These findings support the inhibitory effect of S100A8 and S100A9 antibodies in the growth of leukemia cancer cells.

Besides antibodies, various research groups used synthetic compounds to inhibit the function of S100/calgranulins, and these compounds could act as hit lead molecules for further designing of novel and potent antagonists for S100/calgranulins. Wellmar and co-workers [112] published a patent on N-(2-oxo-3-(trifluoromethyl)-2,3-dihydro-1H-benzo[d]imidazo[1,2-a]imidazol-3-yl) amide derivatives as potent inhibitors of interactions between S100A9 and RAGE or TLR-4 in the treatment of cancer. Among all the derivatives compound 2 (ABR239071) (Figure 5) was found to be effective in terms of reducing the tumor size (50% reduction) when tested in a mice model of tumor when administered 30 mg/kg (daily oral dose for 16 days). It was also revealed that compound 2 competitively binds to S100A9 in the presence of RAGE and TLR-4 and it inhibits RAGE with an IC₅₀ value of 3.13 μM [112]. Same research group published more patent on imidazo-[2,1-B]-thiazole and 5,6-dihydroimidazo-[2,1-b]-thiazole derivatives and claimed to be potent series of inhibitors of S100A9 in the treatment of cancer. Binding of the compounds with S100A9 was examined by using surface plasmon resonance. Inhibition between S100A9 and RAGE was evaluated by using amplified luminescent proximity homogenous assay in which interaction of biotinylated hS100A9 to hRAGE-Fc generated high luminescent signal and inhibited the complex resulting in decreased signal. Almost all the compounds showed good-to-moderate inhibition amongst which compound 3 displayed prominent inhibition of interactions between S100A9-RAGE with an IC₅₀ value of 0.07 μM [113].

Fritzson and colleagues [114] also published a patent claiming N-(heteroaryl)-sulfonamide synthetic derivatives as potent inhibitors of interactions between S100A9 and RAGE or TLR-4 in the treatment of cancer. Majority of the compounds significantly inhibited the interactions between S100A9 and RAGE with the IC₅₀ values ranging from 0.12 to 4.6 μM. Among these compounds three most potent molecules (compounds 4-6) were administered to C57BL/6 mice to evaluate their anti-tumor effect against murine MC38 colon adenocarcinoma cells when given subcutaneously. These compounds significantly reduced the tumor size when administered 1 mg/kg i.v dose or 5 mg/kg of oral dose [114].

Recently in 2020, Korkaya and colleagues [115] reported some synthetic compounds that have potential to inhibit calprotectin and claimed to be used in the treatment of tumor metastasis. Compounds showed anticancer effect by inhibiting the induction of myeloid derived suppressor cells (MDSC). They first evaluated the designed library of compounds by taking a look into the crystal structure of S100A8/S100A9. Among the designed compounds forty compounds were identified with best binding in silico. In vitro assay was performed to evaluate the inhibitory effect of forty compounds on MDSC differentiation which revealed three hit lead compounds 7-9 that showed significant inhibition (Figure 5). Upon further examination the compounds showed anticancer potential by inhibiting the MDSC-related S100A8 and S100A9 genes. Among these three potent molecules, the compound 8 displayed reasonable suppression of tumor growth in a mouse model [115].

Overall, these studies suggest that targeting S100A8/A9 and S100A12 is a potential therapeutic strategy to attenuate chronic inflammation in various inflammatory diseases including atherosclerosis. Further, Quinoline-3-carboxamide derivatives effectively target S100A8/A9 and S100A12 and result in suppression of inflammation with beneficial clinical outcome in various inflammatory diseases. However, no effective and efficient derivative has been reported to target the chronic inflammation in atherosclerotic plaque which leads to destabilization of the plaque resulting into unstable plaque. Thus, there is a need to develop better derivatives or small molecules to target S100 proteins especially S100A8, S100A9, and S100A12 in atherosclerosis.

7. CONCLUSION

From a large amount of experimental data, it can be concluded that S100A8, S100A9, their complex, and S100A12 are not only inflammatory markers but also are the participants in the pathogenesis of atherosclerosis. This suggests their critical importance as a target in developing better therapeutic strategies in the treatment of atherosclerosis. A number of small molecule inhibitors are currently used in clinical and preclinical research that inhibit these proteins but there is still a need to look deeper into their molecular mechanism of action to further accelerate the research. A critical review of the patent literature also suggests that no effective synthetic derivative or chemical compositions has yet been reported that could be used in the treatment of chronic inflammation particularly in atherosclerosis. Thus, there is a pressing need for the development of therapeutic chemical compositions or small molecules that can target S100/calgranulins to treat atherosclerotic inflammation.

8. EXPERT OPINION

In almost every inflammatory condition, S100A8/A9 and S100A12 have been found to be significantly increased as alarmins of inflammation. Despite significant information showing the importance of these proteins in various biological functions related to atherosclerosis, the defense mechanism of S100/calgranulins is still not very clear. Although there are many studies focusing on S100A8 and S100A9, unfortunately there are limited reports on detailed functional studies on their complex formation and their biological functions (S100A8/A9). The biological functioning of a particular protein depends upon structural features, net charge, amino acid sequence, and polarity of its specific binding site or surface. Therefore, it is critical to analyze various structural features of proteins to block their interactions with their partner proteins to develop therapeutics in the treatment of diseases. Overall, we are still striving hard to unravel the precise functional mechanism for S100/calgranulins.

For S100A8/A9 protein complex, the binding of RAGE can be blocked by using CHAPS (Figure 4) and quinoline-3-carboxamide derivatives [52, 73]. However, the particular site(s) of S100A8/A9 to which these derivatives can bind remain unclear. Adequate studies on CHAPS have been published that defines its binding mechanism and various binding interactions on the hydrophobic surface of S100A9 [52]. Similarly, for S100A12, SPR studies revealed the binding of Paquinimod to this protein [45], but it is still unclear that which amino acid residues of S100A12 are involved in binding with Paquinimod to block the interactions between S100A12 and RAGE V domain. Tranilast can block these interactions by binding to the same hydrophobic surface of S100A12 which is exposed for RAGE bindings [63]. Based on the available information on S100/calgranulins, these proteins have more than single binding sites where a molecule can fit, which can change their functional conformation. Several binding sites on S100A9 dimeric complex and S100A12 proteins were found by performing SiteMap using Schrodinger 2020-4 tools [116-118]. S100A9 may have four binding sites other than CHAPS binding surface. Three binding sites have good draggability score (D. Score) and site score (S. score) which are depicted in Table 1, amongst which CHAPS can bind to the amino acid residues of Site 3. It is noteworthy that Site 1 has a better D and S score than Site 3 (CHAPS binding surface) which suggests that there may be a better binding site to target on S100A9 protein to further develop novel inhibitory molecules that can bind to S100A9 and block the formation of S100A8/A9 heterodimeric complex or can block the RAGE and TLR-4 interactions. Various structural features and functions of these binding sites are summarized in Table 1.

Table 1:

Various binding sites available on S100A9 and S100A12 proteins and their structural and functional information

Possible shallow binding sites available on S100A9
Site	Site Score	D. Score	Volume	Active/ND	Function	Amino acids
Site_1	0.945	1.086	147.49	ND	Unknown (by targeting this shallow site may change the conformation of other potential sites that can block protein-protein interactions)	Cys3, Lys4-Arg8, Arg10-Thr12, Thr14, Ile15, Arg40, Lys41, Asp42, Leu43, Gln44, Gln71, Phe74, Ser75, Phe77, Glu78, Phe79, Met81, Leu82, Arg85
Site_3	0.851	0.955	81.977	Active	RAGE or CHAPS binding site (responsible for triggering inflammatory response) [52]	Asp42, Leu43, Asp44, Leu45, Gln46, Asn47, Phe48, Leu49, Glu52, Asn53, Glu56, Lys57, Glu60, Ile62, Met63, Ile80, Met81, Leu82, Leu86, Thr87, Trp88, Ala89
Site_2	0.844	0.960	87.808	Active	Consisting of Ca2+ ions binding loops. Site is responsible for the formation of heterodimer and heterotetramer with S100A8 (depends upon Ca2+ ion concentration) [49, 51]	Glu9, Ile12, Glu13, Thr14, Ile16, Asn17, Thr18, His20, Gln21, Tyr22, Ser23, Val24-Pro29, Thr31, Gln32, Phe35, Lys36, Val39, Arg40, Leu66, Asp67, Thr68, Asp69, Asp71, Glu78
Site_4	0.625	0.684	31.899	ND	Unknown	His26, Pro27, His61, Ile62, Met63, Glu64, Leu66, Thr68, Ala70, Lys72, Gln73, Phe76
Site_5	0.541	0.577	17.836	ND	Unknown	Pro29, Asp30, Thr31, Gln34, Lys51, Val58, Ile59, Ile62, Asp65, Asp67
Possible binding sites available on S100A12
Site	Site Score	D. Score	Volume	Active/ND	Function	Amino acids
Site_1	0.881	0.947	124.509	Active	Bind to RAGE and triggers inflammatory response [63]	His7, Leu8, Ile11, Val12, Ile14, Phe15, Leu41, Thr44, Phe71, Phe74, Ile75, Leu77, Val78, Ala81, Leu82
Site_2	0.611	0.551	53.165	ND	Targetting this site may change the conformation of protein and thus can block the various protein-protein interactions.	Ile45, Lys46, Val53, Glu56, Ile57, Ile80, His90, Lys91

CHAPS, 3-[(3-cholamidopropyl) dimethylammonium]-1-propanesulfonate; ND, Not Determined (Not reported); RAGE, receptor for advanced glycation end products

Similarly, except the binding sites of Tranilast on S100A12, SiteMap analysis revealed other binding sites that can be targeted to develop therapeutic compounds to block the interactions of S100A12 with its partner proteins (Table 1).

Thus, based on the findings of SiteMap analysis, further investigations with carefully designed experiments are warranted to identify novel binding sites and cellular response to better understand the cellular and molecular mechanisms on the role of S100/calgranulins in chronic inflammatory conditions and develop novel therapeutics in the treatment of atherosclerosis.

ARTICLE HIGHLIGHTS.

• The pathological role of S100/calgranulins in atherosclerosis is critically analyzed.

• Expression, distribution, and structural information of S100A8/A9 and S100A12 proteins are reviewed.

• Various small molecules and their effects in clinical and preclinical studies in antagonizing the S100/calgranulins have been disclosed.

• Patent literature is provided with description on the importance of antagonizing S100/calgranulins by using reported synthetic molecules and antibody compositions.

• Gap in our knowledge and future directions for further research are highlighted to define the role and significance of antagonizing S100/calgranulins in various inflammatory diseases.