Modulating the RAGE‐Induced Inflammatory Response: Peptoids as RAGE Antagonists

While the primary pathology of Alzheimer's disease (AD) is defined by brain deposition of amyloid‐β (Aβ) plaques and tau neurofibrillary tangles, chronic inflammation has emerged as an important factor in AD etiology. Upregulated cell surface expression of the receptor for advanced glycation end‐products (RAGE), a key receptor of innate immune response, is reported in AD. In parallel, RAGE ligands, including Aβ aggregates, HMGB1, and S100B, are elevated in AD brain. Activation of RAGE by these ligands triggers release of inflammatory cytokines and upregulates cell surface RAGE. Despite such observation, there are currently no therapeutics that target RAGE for treatment of AD‐associated neuroinflammation. Peptoids, a novel class of potential AD therapeutics, display low toxicity, facile blood‐brain barrier permeability, and resistance to proteolytic degradation. In the current study, peptoids were designed to mimic Aβ, a ligand that binds the V‐domain of RAGE, and curtail RAGE inflammatory activation. We reveal the nanomolar binding capability of peptoids JPT1 and JPT1a to RAGE and demonstrate their ability to attenuate lipopolysaccharide‐induced pro‐inflammatory cytokine production as well as upregulation of RAGE cell surface expression. These results support RAGE antagonist peptoid‐based mimics as a prospective therapeutic strategy to counter neuroinflammation in AD and other neurodegenerative diseases.


Introduction
Alzheimer's disease (AD) is characterized by accumulation of amyloid-β protein (Aβ) plaques in frontal cortex and hippocampus. [1]While harmless in monomeric form, these proteins typically gain toxicity as aggregate species to promote cellular death in the brain.Moreover, recent studies suggest that neuroinflammation is strongly indicated in AD pathology.This inflammation is induced, in part, by upregulated cell surface expression and activation of the receptor for advanced glycation end-products (RAGE). [2]RAGE is a multi-ligand receptor that functions in the management of oxidative stress and chronic inflammation.Studies report an increased presence of RAGE ligands in AD brain, including Aβ oligomers and fibrils, S100 calcium-binding protein B (S100B), and high mobility group box 1 (HMGB1). [3]2b,5] Rise in the evidence of RAGE-associated neuroinflammation in AD has stimulated numerous studies aimed at identifying small molecule RAGE antagonists that are non-toxic and capable of crossing the blood-brain barrier (BBB).Small molecule FPS-ZM1 was selected via screening from a library of tertiary amines for its ability to bind RAGE-transfected cells and block RAGEÀ Aβ interaction. [6]This compound suppressed the level of neuroinflammation in cell culture and mouse models [3b,6-7] but did not progress to clinical trials.TTP488, also known as azeliragon, is another small molecule RAGE antagonist that was examined in human clinical trials as an AD drug.Despite its success in animal model studies as well as phase I and II clinical trials, development of azeliragon ended abruptly in phase III due to lack of efficacy in mild AD patients compared to the placebo group [8] coupled with the presentation of toxicity in patients when azeliragon was given at higher dosage. [9]eptoids, a type of peptidomimetic, present a novel class of therapeutics that can be designed as antagonists to RAGE.Peptoids are small chains with protein-like side chains to facilitate molecular recognition similar to peptides.However, while most peptides are vulnerable to degradation by proteases, peptoids can evade such degradation by repositioning the side-chains from the α-carbon, such as on peptides, to the amide nitrogen.Peptoids can influence signaling pathways and block programmed cellular death (apoptosis) as well as behave as anti-inflammatory agents.In addition, peptoids generally have low toxicity and demonstrate low immunogenicity. [10]Of importance to prospective AD therapies, peptoids have been observed to cross the BBB.For example, peptoid AIP1 demonstrated BBB permeability in a murine brain microvascular endothelial cell culture, [11] while peptoids SLKP and RA-1 demonstrated BBB permeability in vivo following intraperitoneal injection into a mouse model [12] and a Wistar rat model, [13] respectively.Peptoids' versatility and potential for intranasal administration make them excellent therapeutic candidates for AD as well as other inflammatory and neuronal diseases.The current study investigated two peptoids, JPT1 and JPT1a (Figure 1), designed as potential RAGE antagonists via mimic of Aβ, a ligand of the RAGE V-domain.Specifically, these peptoids mimic the Aβ aromatic, hydrophobic core, KLVFF, since a common feature of RAGE antagonists is an aromatic, hydrophobic region. [14]The JPT1 sequence in terms of corresponding amino acids, KIIFFIFF, includes four aromatic side chains.This peptoid utilizes chiral monomers to induce peptoid helicity in a manner that places two aromatic groups on each of two sides of the helix with spacing that mimics the position of aromatics within the β-sheet structure of Aβ aggregates.To probe the utility of this design, peptoid JPT1a comprises the same sequence as JPT1 but utilizes achiral groups to reduce helicity and thus disrupt the positioning of the aromatic residues while simultaneously imparting flexibility.The current study investigated the potential of these peptoids as non-toxic, RAGEtargeted therapeutics.Results demonstrate that these peptoids bind RAGE with nanomolar affinity and are non-toxic in macrophage culture.Moreover, these peptoids attenuate lipopolysaccharide (LPS)-induced pro-inflammatory cytokine release as well as reduce LPS-upregulated RAGE cell surface expression.Together, these findings identify RAGE-binding peptoid mimics as a promising, novel therapeutic approach to combat AD-associated neuroinflammation.

Peptoids JPT1 and JPT1a bind sRAGE with nanomolar affinity
Binding between peptoids and RAGE was evaluated using sRAGE, the soluble, extracellular region of RAGE that includes the V-domain capable of binding a multitude of RAGE ligands.Binding affinity was ascertained using an assay design established by Friguet et al. [15] and subsequently used by others. [16]ere, binding equilibrium was achieved in solution equilibrium reactions containing sRAGE and ligand, peptoid or azeliragon.Subsequently, unbound sRAGE was detected using ELISA such that bound sRAGE within solution equilibrium reactions could be calculated.Peptoids JPT1 and JPT1a exhibited K d values of 78 � 17 nM (Figure 2A) and 58 � 12 nM (Figure 2B), respectively.When evaluated within the same assay format, azeliragon exhibited a significantly higher (p < 0.0002) K d value of 239 � 34 nM (Figure 2C).These results demonstrate a high, nanomolar binding affinity between sRAGE and the designed peptoids and further suggest that the chiral structure of JPT1 is not necessary for a strong binding interaction.

Peptoids JPT1 and JPT1a do not exhibit toxicity
Tohoku Hospital Pediatrics-1 (THP-1) monocytes differentiated into macrophages were used to assess the effects of peptoids JPT1 and JPT1a on toxicity and RAGE function.THP-1 monocytes are derived from human acute monocytic leukemia and can differentiate into macrophages when incubated in the presence of phorbol 12-myristate 13-acetate (PMA).Such differentiated THP-1 macrophages are often used as a model for human microglia to study neurotoxic inflammatory responses. [17]As an active immune defense of the central nervous system, microglia function as macrophages in the brain and release cytokines in response to plaques, injuries, or infections.This makes THP-1 macrophages an ideal, affordable selection to study peptoid modulation of inflammatory stimuli such as LPS.
To establish that peptoids are non-toxic in the THP-1 macrophage cell culture system, cell viability was examined via measurement of metabolic activity using a 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT) reduction assay.Figure 3 demonstrates that cell viability was unaltered in the presence of up to 50 μM peptoid following 72 h treatment but was significantly reduced in the presence of 5-50 μM azeliragon.These results ensure that peptoid-induced toxicity will not occur within the duration of experiments that probe peptoid antagonistic capability.

Peptoids JPT1 and JPT1a attenuate LPS-induced cytokine release by THP-1 macrophages
The immunomodulatory capability of peptoids was examined by measuring the presence of inflammatory cytokines IL-1β, IL-6, and IL-8 in the supernatant harvested from THP-1 macrophages.When macrophages were exposed to a chronic, lowdose pro-inflammatory stimulus, 2 ng/mL LPS (positive control), significant levels of inflammatory cytokines were observed in the supernatant compared to cells exposed to the vehicle alone (negative control) (Figure 4A).In contrast, a dose-dependent attenuation for all three of these cytokines was observed when the macrophages exposed to LPS were simultaneously treated with peptoids (Figure 4B, C).Both JPT1 and JPT1a at 50 μM significantly reduced all three LPS-induced cytokines relative to the positive control.Additionally, the presence of 2 μM and 10 μM JPT1a as well as 10 μM JPT1 significantly diminished IL-6 production.THP-1 macrophages treated with peptoid alone displayed pro-inflammatory cytokine expression similar to the vehicle (Supplemental Figure S1), indicating that the peptoids alone do not elicit inflammatory response.These results demonstrate the ability of peptoids JPT1 and JPT1a to attenuate LPS-induced release of inflammatory cytokines and further suggest that the chiral structure of JPT1 is not necessary for antagonism of LPS-induced cytokine release.

Peptoids JPT1 and JPT1a attenuate LPS-induced selfupregulation of cell surface RAGE
Because activation of RAGE is known to upregulate RAGE expression itself, [18] the ability of peptoids to modulate cell surface RAGE expression was examined via immunocytochemistry.When THP-1 macrophages were treated with a chronic, low-dose proinflammatory stimulus, 2 ng/mL LPS (positive control), a significant upregulation of cell surface RAGE was observed (Figure 5, Figure 6a).Co-incubation simultaneously with 2 ng/mL LPS and   6b) and yielded an observable but not significant reduction at 10 μM and 2 μM peptoid (Figure 6b).Peptoids alone do not modulate cell surface RAGE expression (Supplemental Figure S2).These results demonstrate the ability of peptoids JPT1 and JPT1a to attenuate LPS-induced self-upregulation of cell surface RAGE and further suggest that the chiral structure of JPT1 is not necessary for antagonism of LPS-induced RAGE self-upregulation.

Discussion
The prevalence of RAGE-associated neuroinflammation in AD, [19] the correlation of inflammation to AD disease severity, [19a,20] and the increase in the presence of inflammatory markers in AD brain [19b,21] all indicate that RAGE-associated inflammation plays a significant role in the progression of AD.Therefore, interrupting interaction between RAGE and ligands that are increased in AD brain, thus attenuating the consequent inflammatory cascade, is central to identifying new drug targets.Indeed, Aβ  binds RAGE within the V-domain, [22] and this binding activates NF-kB signaling that leads to release of proinflammatory cytokines in AD brain. [23]Relative to other small molecule RAGE antagonists, peptoids designed to mimic the aromatic, hydrophobic core of Aβ (KLVFF) present a novel therapeutic approach to target RAGE.This study provides evidence that peptoidbased mimics JPT1 and JPT1a both bind RAGE with nanomolar affinity to attenuate both RAGE-induced inflammatory response and RAGE upregulation.Moreover, the similar binding affinity and antagonistic activity indicate that the chiral structure of JPT1 is not critical, which could result from the flexibility of JPT1a enabling it to assume the conformation needed to bind and serve as an antagonist.
A high binding affinity is important for a drug candidate to effectively bind RAGE within the in vivo environment.Peptoids JPT1 and JPT1a both demonstrated nanomolar binding affinity to sRAGE at 78 � 17 nM and 58 � 12 nM, respectively, in an equilibrium binding assay format (Figure 2A, B).This affinity is significantly higher (p < 0.0002) than RAGE-targeted small molecule and AD clinical trial drug azeliragon, which exhibited a K d of 239 � 34 nM (Figure 2C), falling within the reported K d range for azeliragon of 12.7-500 nM. [24]The affinity of peptoids for sRAGE is slightly lower than that of RAGE-targeted small molecule FPS-ZM1, reported to bind RAGE with a K d of 15 nM. [25]owever, this compound may not have advanced to clinical trials as an AD therapeutic due to toxicity. [26]14a] The zero-to low-toxic nature of other peptoid drug candidates [27] suggests that these Aβ-mimic peptoids hold potential to present a lower toxicity drug candidate than prior RAGE antagonists.In fact, peptoids JPT1 and JPT1a exhibited no toxic effect on human THP-1 macrophages during 72 h treatment and at doses as high as 50 μM (Figure 3).Other small molecule RAGE antagonists designed to offset inflammatory responses were limited by their toxicity.For example, when viability was tested in human umbilical vein endothelial cells (HUVECs) treated with 0-20 nM FPS-ZM1, this RAGE antagonist showed no toxicity at concentrations � 1 nM, but a significant decrease in HUVEC viability was noted at concentrations above 1 nM, [26] potentially impairing the ability of this compound to advance to clinical trials.When THP-1 macrophages were treated with 1-50 μM azeliragon over 72 h, cell viability was significantly reduced at azeliragon concentrations as low as 5 μM and by up to 51 % at the highest dose tested, indicating that azeliragon induces cellular toxicity (Figure 3).Azeliragon abruptly stopped advancing at phase III of the clinical trial due to its toxic effect when increased to a higher dosage. [9,28]hen peptoids were examined for their ability to attenuate inflammatory response, JPT1 and JPT1a dose-dependently reduced proinflammatory cytokine excretion by LPS-treated THP-1 macrophages (Figure 4).RAGE antagonist FPS-ZM1 also suppressed LPS-induced inflammatory cytokine release in an acute lung injury (ALI) murine model [29] as well as in both BV-2 cells and primary microglial cells. [30]In particular, levels of IL-1β, IL-6, and TNF-α were reduced.Azeliragon also reduced serum cytokines in mice to improve xenograft rejection by T cells, with reduction in levels of IL-1β, IL-10, and IL-17A. [31]Such studies indicate that inhibiting interactions between RAGE and its ligands can affect the downward stream of proinflammatory cytokine production and thus attenuate negative pathological inflammatory pathways.The parallel reduction of cytokine production by peptoids in cell culture suggests that they possess potential to affect inflammation-associated disease states.
Peptoids JPT1 and JPT1a also exhibited the capability to attenuate LPS-induced self-upregulation of RAGE (Figure 5, Figure 6).5c,32] However, in a chronic inflammatory disease state, wherein RAGE ligands are upregulated, the constant interaction of RAGE and these ligands additionally upregulates RAGE expression, creating a positive feedback loop. [33]Subduing RAGE expression is not only critical in influencing the subsequent release of inflammatory markers but also in preventing an undesirable cycle of inflammatory responses. [34]Studies have demonstrated the importance of reducing RAGE self-upregulation following LPS-induced inflammation.For instance, when a mouse model was administered RAGE antagonist treatments of FPS-ZM1 or azeliragon following ALI induced via intratracheal instillation of LPS, both treatments attenuated LPS-induced RAGE expression. [29]Similarly, reduction of RAGE self-upregulation by peptoids may enhance its ability to attenuate negative, chronic inflammatory disease states.

Conclusions
This study provides evidence that peptoid-based mimics of the aromatic, hydrophobic core of Aβ interact with RAGE to alter inflammatory response.The potential contribution of these peptoids to attenuate neuroinflammation in the progression of AD presents a novel class of therapeutics.Results from the current study demonstrate that peptoids JPT1 and JPT1a exhibit both nanomolar binding affinity to RAGE and the capability to significantly reduce LPS-induced cytokine release and RAGE self-upregulation at non-toxic concentrations.Similar binding affinity and antagonist activity by both the chiral and achiral design indicates that peptoid chirality is not essential and prompts future experiments that will further probe key structural elements.These findings extend previous studies reporting the importance of exploring RAGE antagonists to offset RAGE-associated inflammation in AD.Furthermore, a previous study evidenced the ability of these peptoids to reduce Aβ aggregation, [35] indicating that peptoids could serve as a dual-target drug.Future cell culture studies will further probe the involvement of NF-kB signaling in the attenuation of RAGE-induced responses, and future in vivo studies will translate these in vitro findings to an in vivo environment and chronic inflammatory disease state.These peptoids also have the potential to block other ligands of RAGE that bind within the V-domain, such as S100B and HMGB1.Such capabilities would extend the potential therapeutic benefits of RAGE antagonist peptoids to other inflammation associated diseases and conditions.

Peptoid synthesis, purification, and characterization
Peptoids JPT1 and JPT1a were synthesized via a two-step process as described previously. [36]Briefly, after rink amide resin was swelled with DMF, and the fluorenylmethoxycarbonyl protecting group was removed, secondary amine was acylated, and side chain amines were added sequentially.This process was repeated until the desired sequence was achieved.Peptoid was removed from the resin with a solution of TFA, triisopropylsilane, and water, diluted in acetonitrile-water, and purified via preparative high performance liquid chromatography (HPLC).Molecular weight (Figure 1) was confirmed using mass spectrometry, and purity (> 98 %) was confirmed via analytical HPLC (Waters 2795 Separations Module) equipped with a Duragel G C18 150 2.1 mm column (Peeke Scientific, Redwood City, CA) and using a linear gradient of 5 to 95 % solvent D in solvent C (solvent C: water, 0.1 % TFA; solvent D: acetonitrile, 0.1 % TFA) over 30 min.Purified peptoid solutions were lyophilized and stored at À 20 °C.At the time of experimentation, lyophilized peptoids were reconstituted in vehicle (PBS) at a concentration of 500 μM.

Solution equilibrium binding assay
sRAGE, the extracellular portion of RAGE that incorporates the Vbinding domain, was used to enable binding measurements in solution.Using an approach established by Friguet et al. [15] and subsequently utilized by others, [16] binding was allowed to reach equilibrium within solution binding reactions; subsequently, unbound sRAGE within these reactions was detected via ELISA wherein the same ligand represented in solution binding reactions, peptoid or azeliragon, was used as the capture agent.Solution binding reactions were prepared using kit-included dilution buffer to contain 20 nM sRAGE and 0-2000 nM ligand, peptoid or azeliragon.The reactions were vortexed gently and incubated overnight at 4 °C to reach binding equilibrium.Solution binding reactions were then applied in triplicate to clear flatbottomed MaxiSorp TM 96-well plates (VWR) coated (overnight, 4 °C) with 250 nM of the same ligand represented in solution binding reactions, peptoid or azeliragon, to allow capture of only unbound sRAGE.Reactions were removed following 5 min on an orbital shaker at 25 °C.This incubation time was experimentally determined to maintain binding equilibrium within the solution. [15]Plates were washed three times immediately after reaction removal and incubated with kit-included sRAGE detection antibody and antirabbit IgG-HRP conjugate in HRP diluent solution with intermediate wash steps (three times).Fresh 5,5'-tetramethylbenzidine (TMB) substrate was applied directly to wells prior to cessation with kitincluded stop solution.Absorbance was assessed at 450 nm using a SpectraMax microplate reader (Molecular Devices, San Jose, CA).
Absorbance data were used to calculate the concentration of sRAGE bound to ligand, peptoid or azeliragon, within the solution binding reaction (Equation 1), where A 0 is the absorbance generated from solution binding reactions containing 20 nM sRAGE alone, A i is the individual absorbance generated from solution binding reactions containing varying concentrations of ligand, peptoid or azeliragon, and A ∞ is the absorbance generated from solution binding reactions containing ligand, peptoid or azeliragon, in excess.[X] total represents the total molar concentration of sRAGE epitopes on peptoid or azeliragon, for which 1 : 1 binding was assumed.
Data were plotted as the concentration of sRAGE bound to ligand, peptoid or azeliragon, in the solution binding reactions (Equation 1) versus the total concentration of ligand, peptoid or azeliragon, present in the solution binding reaction.GraphPad Prism 9.4.0 was used to perform least squares regression for a one site, specific binding model to determine K d .

Figure 2 .
Figure 2. Binding affinity of peptoids and azeliragon for sRAGE.A solution equilibrium binding assay was performed to determine the binding affinity between sRAGE and JPT1 (panel A), JPT1a (panel B), and azeliragon (panel C). 20 nM sRAGE and 0-2000 nM ligand, peptoid or azeliragon, were allowed to achieve binding equilibrium in solution.sRAGE bound to ligand, peptoid or azeliragon, was quantified and plotted versus the varying concentration of ligand in solution.A nonlinear fit (solid lines) was performed to calculate K d values of 78 � 17 nM, 58 � 12 nM, and 239 � 34 nM for JPT1, JPT1a, and azeliragon, respectively.95 % confidence intervals are shown as dashed lines.Individual symbols represent independent experiments.

Figure 4 .
Figure 4. Effect of peptoids on LPS-induced pro-inflammatory cytokine release by THP-1 macrophages.THP-1 macrophages were exposed to vehicle (panel A, negative control, light grey) or stimulated via 72 h incubation with 2 ng/mL LPS alone (panel A, positive control, dark grey) or stimulated simultaneously with 2 ng/mL LPS and 2 μM (solid), 10 μM (stripe), or 50 μM (pattern) JPT1 (panel B) or JPT1a (panel C).Supernatant media was collected, and levels of cytokines IL-1β, IL-6, and IL-8 were measured via ELISA.Results are reported as cytokine concentration (panel A) or as a fraction of the positive control, indicated by the dashed line at y = 1 (panels B, C).Error bars indicate SEM; n = 3-4 from 3-4 independent experiments.* p < 0.05, **** p < 0.0001 vs. vehicle (panel A) or vs. positive control (panels B, C).

Figure 5 .
Figure5.Effect of peptoids on LPS-induced upregulation of cell surface RAGE.THP-1 macrophages were exposed to vehicle (negative control, first row) or stimulated via 72 h incubation with 2 ng/mL LPS alone (positive control, second row), with 2 ng/mL LPS in the presence of JPT1 at 2 μM (third row) or 50 μM (fifth row), or with 2 ng/mL LPS in the presence of JPT1a at 2 μM (fourth row) or 50 μM (sixth row).Cells were stained with DAPI (blue, first column), phalloidin (green, second column), and anti-RAGE antibody (red, third column) to detect nucleus, cytoskeleton, and cell surface RAGE, respectively.Images were acquired via confocal microscopy using a plan-neofluar 40X/1.3 oil DIC immersion objective.Images are representative of 3 images acquired from duplicate treatments within each of 3 independent experiments.

Figure 6 .
Figure 6.Quantified effect of peptoids on LPS-induced upregulation of cell surface RAGE.THP-1 macrophages were treated (72 h), stained, and evaluated via immunocytochemistry as described in Figure 5. (A) THP-1 macrophages were exposed to vehicle (negative control, light grey) or stimulated with 2 ng/mL LPS alone (positive control, dark grey).RAGE expression as a percentage of cell area was calculated using a custom Matlab routine.(B) THP-1 macrophages were stimulated with 2 ng/mL LPS in the presence of 2 μM, 10 μM, or 50 μM JPT1 (black) or JPT1a (white).RAGE expression as a percentage of cell area is reported as the fraction of that observed in the positive control, indicated by the dashed line at y = 1.Error bars indicate SEM; n = 2 from 3 independent experiments.*** p < 0.001, **** p < 0.0001 vs. vehicle (panel A) or vs. positive control (panel B).