Extracellular CIRP as an endogenous TREM-1 ligand to fuel inflammation in sepsis

Extracellular cold-inducible RNA-binding protein (eCIRP) is a recently-discovered damage-associated molecular pattern. Understanding the precise mechanism by which it exacerbates inflammation is essential. Here we identified that eCIRP is a new biologically active endogenous ligand of triggering receptor expressed on myeloid cells-1 (TREM-1), fueling inflammation in sepsis. Surface plasmon resonance revealed a strong binding affinity between eCIRP and TREM-1, and FRET assay confirmed eCIRP's interaction with TREM-1 in macrophages. Targeting TREM-1 by its siRNA or a decoy peptide LP17 or by using TREM-1-/- mice dramatically reduced eCIRP-induced inflammation. We developed a novel 7-aa peptide derived from human eCIRP, M3, which blocked the interaction of TREM-1 and eCIRP. M3 suppressed inflammation induced by eCIRP or agonist TREM-1 Ab crosslinking in murine macrophages or human peripheral blood monocytes. M3 also inhibited eCIRP-induced systemic inflammation and tissue injury. Treatment with M3 further protected mice from sepsis, improved acute lung injury, and increased survival. Thus, we have discovered a novel TREM-1 ligand and developed a new peptide M3 to block the eCIRP-TREM-1 interaction and improve the outcomes in sepsis.


Introduction
Pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) play leading roles in fueling inflammation in sepsis (1)(2)(3). Cold-inducible RNAbinding protein (CIRP) is a 18-kDa RNA chaperone protein (4). In addition to the passive release due to necrotic cell death, CIRP can be released extracellularly during sepsis, hemorrhage or ischemia-reperfusion injury, translocating from the nucleus to cytoplasmic stress granules before being released to the extracellular space (5). Extracellular CIRP (eCIRP) acts as a DAMP, causing inflammation and organ injury (5,6). Elevated plasma levels of eCIRP have been correlated with a poor prognosis in patients with sepsis (5,7). As a relatively new DAMP, a more thorough understating of eCIRP's pathobiology in inflammatory diseases is required to develop novel therapeutics.
Conversely, CIRP -/mice were protected from sepsis and ALI (5,6). Consistent with these findings, treatment of septic animals with a neutralizing antibody against eCIRP attenuated organ injury and prolonged survival (5,6). Collectively these findings indicate that eCIRP is a major contributing factor in the pathogenesis of sepsis; targeting eCIRP is a valid strategy to mitigate sepsis severity.
Triggering receptor expressed on myeloid cells-1 (TREM-1) is an innate immune receptor expressed primarily on neutrophils and macrophages (13). TREM-1 activation triggers inflammation independently (14), as well as by synergizing with the TLR4 pathways (14,15). TREM-1 activation leads to DNAX-activating protein of 12 kDa (DAP12) phosphorylation (13,15) which promotes activation of the tyrosine kinase Syk (16), resulting in the production of cytokines and chemokines (15,16). The ligands for TREM-1 remain elusive, with only high mobility group box 1 (HMGB1) (17), extracellular actin (18), and peptidoglycan recognition protein 1 (PGLYRP1) (19) identified thus far. The nature of the TREM-1 ligand(s) and mechanisms of TREM-1 signaling are not yet thoroughly explored. Identification of a new natural endogenous TREM-1 ligand will not only help improve our understanding of the pathophysiology of inflammatory diseases, but also discover new therapeutic avenues against those diseases.
Both eCIRP and TREM-1 are upregulated in sepsis to serve as mediators of inflammation (5,20), but their interaction has not been studied. In sepsis, both eCIRP and TREM-1 reside extracellularly (5,20), giving rise to the hypothesis that eCIRP could be a novel endogenous ligand of TREM-1. In this study, we have discovered that eCIRP is a new biologically active endogenous TREM-1 ligand and their interaction fuels inflammation. We have also developed a unique human eCIRP-derived ligand-dependent 7 amino acid peptide (RGFFRGG) to serve as an antagonist of TREM-1, named M3, and implemented M3 as a therapeutic in pre-clinical models of sepsis.

Identification of eCIRP as a new TREM-1 ligand to promote inflammation
To study the direct interaction between eCIRP and TREM-1, we performed a surface plasmon resonance (SPR) assay, which demonstrated a strong binding between rmCIRP and rmTREM-1 with a KD of 11.7 × 10 -8 M (Fig 1A). An immunofluorescence study was performed to confirm the co-localization of eCIRP and TREM-1 in macrophages after rmCIRP stimulation. It clearly demonstrated the co-localization of rmCIRP and TREM1, as indicated by the yellow color in the merged image (Fig 1B). Conversely, rmCIRP did not co-localize with a negative control, the macrophage pan marker CD11b (Fig 1B). We next performed FRET analysis to quantitatively determine rmCIRP's association with TREM-1. FRET analysis revealed a clear association between rmCIRP and TREM-1 with an increase in FRET units of nearly 7-fold compared to rmCIRP's interaction with negative control CD11b in RAW 264.7 macrophages (Fig 1C), and nearly 10-fold in WT peritoneal macrophages (Fig 1C). These findings implicate that eCIRP is a novel TREM-1 ligand. We then studied the activation of downstream molecules DAP12 and Syk in macrophages treated with rmCIRP and found significant increase in the phosphorylation of DAP12 and Syk at 10 min after rmCIRP stimulation (Fig 1D, Supplemental Figure 1). We next confirmed the functional role of TREM-1 in eCIRP-mediated inflammation. We found that the siRNA-treated macrophages showed significant inhibition of TNF-α production following rmCIRP stimulation (Fig 1E). Similarly, the treatment of macrophages with LP17, an inhibitor of TREM-1 (21), dose-dependently inhibited rmCIRP-induced TNF-α production in RAW264.7 cells (Fig 1F). Conversely, the scramble peptide for LP17 did not show any inhibition of TNF-α production ( Fig 1F). Collectively, these data show that eCIRP specifically binds to TREM-1 in macrophages and induces TNF-α production. TREM-1 expression in macrophages is increased in sepsis (15). To explore the role of eCIRP on this increase, RAW264.7 cells and murine primary peritoneal macrophages were stimulated with rmCIRP. TREM-1 mRNA levels were increased 2.5-fold in rmCIRP-treated RAW264.7 cells as compared to PBS control (Supplemental Fig 2A). The protein levels of TREM-1 expression on the cell surface of both RAW264.7 cells and primary murine peritoneal macrophages treated with rmCIRP were significantly increased by 4.3 and 1.6-fold, respectively, compared to PBS control (Supplemental Fig 2B, C).

Pharmacologic inhibition of TREM-1 attenuates eCIRP-induced systemic inflammation and lung injury in mice
To further verify the interaction between TREM-1 and eCIRP in vivo, we exposed mice to rmCIRP and LP17, a TREM-1 decoy receptor which functions to inhibit TREM-1 mediated signaling (21). We proved that LP17 attenuated rmCIRP-induced systemic inflammation and lung injury in mice (Fig 2). Treatment of mice with LP17 dramatically reduced the levels of AST, ALT, and LDH in the serum of rmCIRP-injected mice compared to vehicle-treated rmCIRP-injected mice (Fig 2A-C). The levels of IL-6 and IL-1b in the serum were significantly decreased in LP17-treated mice (Fig 2D-E). LP17 treatment significantly attenuated TNF-a, IL-1b, and IL-6 mRNA (Fig 2F-H) and proteins in the lungs (Fig 2I-K). Similarly, treatment with LP17 significantly reduced the expression of chemokines MIP-2 and KC and adhesion molecules ICAM-1 and VCAM-1 mRNA in the lungs (Supplemental Fig 3). Histological images of lung tissue showed increased levels of alveolar congestion, exudate, interstitial and alveolar cellular infiltrates, intra-alveolar capillary hemorrhages, and damage of epithelial architecture, in rmCIRP-injected mice compared to sham mice (Fig 2L). LP17 treatment improved these histological injury parameters in rmCIRP-injected mice (Fig 2L). These histological changes were reflected in a significant decrease in lung tissue injury score in LP17-treated mice ( Fig   2M).

TREM-1 -/macrophages and mice produce decreased levels of pro-inflammatory cytokines upon eCIRP exposure
To definitively confirm the functional interaction between eCIRP and TREM-1, we performed both in vitro and in vivo studies using TREM-1 -/macrophages and mice. Primary peritoneal macrophages were isolated from both WT and TREM-1 -/mice and exposed to rmCIRP. IL-6 and TNF-a production after rmCIRP stimulation were significantly decreased in supernatant of TREM-1 -/macrophages as compared to WT macrophages (Fig 3A-B). Similarly, WT mice exposed to rmCIRP increased the serum levels of IL-6 and IL-1b by 2.3-and 9.1-fold as compared to their TREM-1 -/counterparts, respectively (Fig 3C-D). These data suggest that genetic ablation of TREM-1 controls eCIRP-induced inflammation.

An eCIRP-derived antagonist M3 inhibits the eCIRP-TREM-1 interaction
After establishing eCIRP as a novel TREM-1 ligand, we sought to identify a small molecule that inhibits the eCIRP-TREM-1 interaction. Employing the Protein Model Portal, part of the protein structure initiative knowledgebase, and Pep-Fold-3 (22), we compared structural images of a known TREM-1 ligand, murine PGLYRP1, and murine CIRP. We identified a section of CIRP with a similar structural form and associated similar amino acid sequences (Supplemental Fig   4A-B). We then synthesized a series of small peptides (M1, M2, M3) from this region of CIRP and tested their ability to inhibit TNF-a secretion in RAW264.7 cells stimulated with rmCIRP ( Fig 4A). We found that M3, a 7-aa peptide, consisting of the amino acid sequence of murine CIRP from 101-107 (RGFFRGG), which has 100% homology with human CIRP, demonstrated the greatest inhibitory effect on TNF-α production by the macrophages following rmCIRP treatment (Fig 4A). Using SPR, we identified considerable binding affinity between M3 and rmTREM-1 with a KD of 35.2 × 10 -6 M (Fig 4B). FRET assay was performed between rmCIRP and TREM-1 in both RAW264.7 cells and murine primary peritoneal macrophages in presence or absence of M3. Interestingly, we found that M3 was able to dramatically abrogate rmCIRP's binding to TREM-1 (Fig 4C-D). The agonist anti-TREM-1 Ab has been shown to induce TNF-α production through TREM-1 crosslinking (13). By inhibiting TREM-1-mediated inflammation caused solely by antibody induced receptor activation, in the absence of any other stimuli, we have demonstrated the independent role of M3 on TREM-1-mediated inflammation. We found that M3 demonstrated significant inhibition of TNF-α production by the macrophages treated with an agonist anti-TREM-1 Ab, indicating M3 specifically worked on TREM-1 (Fig 4E). On the other hand, the scramble peptides, M3-Sc1 and M3-Sc2 did not demonstrate any inhibition ( Fig 4E ). We next validated M3's inhibitory effect on controlling rmCIRP-treated inflammation in macrophages. We found that M3 inhibited rmCIRP-mediated TNF-a and IL-6 production by the RAW264.7 cells in a dose-dependent manner while the scramble peptides M3-Sc1 and M3-Sc2 were not able to inhibit rmCIRP-induced inflammation (Fig 4F-G). Similarly, M3 treatment was also able to significantly inhibit the production TNF-α in rmCIRP-stimulated primary human peripheral blood mononuclear cells (PBMC) obtained from healthy volunteers where the highest inhibition of 76% in TNF-α production occurred at 10 µg/ml of M3 treatment (Fig 4H).
Collectively, we have discovered a novel small peptide M3 which blocks the eCIRP-TREM-1 interaction and inhibits eCIRP-mediated inflammation.

M3 inhibits eCIRP or LPS-induced inflammation in mice
Having established M3 as an antagonist of eCIRP and TREM-1 binding, we sought to evaluate its efficacy in reducing pulmoanary and systemic inflammation in vivo. Administration of rmCIRP in healthy mice dramatically increased lung mRNA and protein levels of TNF-a, IL-1b, and IL-6 while M3 treatment reduced their expression (Fig 5A-B, E-F, I-J). rmCIRP administration also increased serum levels of IL-6 and IL-1b, and M3 treatment decreased these levels (Fig 5C-D). LPS is known to stimulate eCIRP release by the macrophages (5). Using an endotoxemia model, we demonstrated that M3 treatment was able to reduce serum levels of IL-6 and TNF-a (Fig 5G-H). Administration of M3 during endotoxemia increased the 7-day survival from 70% to 100% (Fig 5K). Therefore, TREM-1 is involved in eCIRP-mediated inflammation and tissue injury, and blockade of the eCIRP-TREM-1 interaction attenuates inflammation.

M3 protects mice from polymicrobial sepsis
We next evaluated the efficacy of M3 in a more clinically relevant model of sepsis utilizing CLP in mice. A double puncture, high mortaility CLP resulted in increases in organ injury markers AST and LDH in the serum at 20 hours, while the mice treated with M3 simultaneously had a significant decrease in their levels (Fig 6A-B). Similarly, serum levels of IL-6 and TNF-a were elevated by CLP, however M3 treatment significantly reduced these levels (Fig 6C-D).
Expression of the pro-inflammatory cytokines IL-6, TNF-a, and chemokine KC mRNA in lung tissues were increased in CLP-induced sepsis, and were reduced with M3 treatment (Fig 6E-G).
Histological images of lung tissue in CLP mice displayed significant damage, with increased levels of alveolar congestion, proteinaceous debris, interstitial and alveolar neutrophil infiltration, intra-alveolar capillary hemorrhages, and damage of epithelial architecture (Fig 6I ).
M3 treatment improved these histological injury parameters in septic mice (Fig 6I). These histological changes were reflected in a significant decrease in lung tissue injury score in M3-treated mice compared to vehicle mice (Fig 6J). To determine if M3 was able to improve survival in sepsis, mice were subjected to reduced severity, single puncture model of CLP and randomized to simultaneous treatment with M3 or vehicle. We found that M3 treatment increased the survival rate from 45% to 80% at day 10 after CLP ( Fig 6H). In addition, in order to test the efficacy of M3 when given in a delayed manner after sepsis induction, mice were subjected to an increased severity, but still single puncture, CLP. M3 injection was given i.p. 90 minutes after CLP. M3 was able to improve 10-day survival from 10% to 25% (p = 0.02) (Supplemental Fig 5). In summary, the eCIRP released in sepsis binds to TREM-1 and potentiates proinflammatory signaling. Administration of M3, by inhibiting the eCIRP-TREM-1 interaction, exhibits excellent therapeutic potential against murine polymicrobial sepsis.

Discussion
Extracellular CIRP has been recently identified as a DAMP that is released during sepsis, shock, and ischemia/reperfusion injury (5,6). However, its molecular mechanism to induce inflammation still remains enigmatic. Despite identification of the TREM-1 receptor nearly two decades ago (13), its ligand(s) are not well elucidated. Our study fills a significant previous knowledge gap by establishing a novel link between eCIRP and TREM-1 demonstrated by in vitro studies using murine and primary human macrophages and in several in vivo models. We summarized the overall findings in Fig 7, which demonstrates that eCIRP released during sepsis binds to TREM-1, serving as a novel biologically active endogenous TREM-1 ligand. This binding leads to the activation of intracellular DAP12 and Syk and increased production of inflammatory mediators to cause hyperinflammation and tissue injury. Targeting the interaction between eCIRP and TREM-1 by a small peptide M3 derived from human eCIRP is protective in sepsis.
We have, for the first time, identified eCIRP as an endogenous ligand for TREM-1. In other words, TREM-1 is a new receptor for eCIRP. Previously, TLR4 was the only identified receptor for eCIRP (5,6). We hypothesized that, like most DAMPs, eCIRP has several receptors for signal transduction. HMGB1, for example, is known to potentiate signals through TLR2, TLR4, receptor for advanced glycation end product (RAGE), TREM-1, and CD163 (23-25).
Using SPR we found a strong KD of 11.7 x 10 -8 M of binding between eCIRP and TREM-1, which is comparable to the dissociation constant of the previously identified receptor for eCIRP, TLR4 (5). Of the known ligands for TREM-1: PGLYRP1 (19), extracellular actin (18), and HMGB1 (17), SPR has been used between TREM-1 and PGLYRP1 and HMGB1. Surprisingly, the KD between eCIRP and TREM-1 (11.7 × 10 -8 M) was more than two folds lower than the KD between HMGB1 and TREM-1 (35.4 x 10 -6 M) (17), indicating higher binding affinity of eCIRP to TREM-1 than between HMGB1 and TREM-1. In addition to SPR, several additional approaches were taken to confirm the physical interaction of eCIRP and TREM-1, including confocal microscopy and FRET analysis.
After demonstrating eCIRP's binding to TREM-1, the focus was shifted to determining if eCIRP's binding to TREM-1 resulted in fueling inflammation and causing tissue injury. This was a crucial step, as we previously demonstrated that although SPR displays binding between eCIRP and TLR2 as well as RAGE, there are no functional outcomes of this association (5). Inhibition of TREM-1 by siRNA showed some, but not complete, reduction in the production of proinflammatory cytokine by the macrophages treated with eCIRP. This could be due to off target effects of TREM-1 siRNA on other TREMs, although this remains speculative. TREM family contains TREM-1, TREM-2, and, in mice, TREM-3 (26). Although TREM-1 contains an immunoreceptor tyrosine-based activation motif (ITAM), additional TREM family members contain an immunoreceptor tyrosine-based inhibitory motif (ITIM), and thus they could play anti-inflammatory role (27,28). In addition, macrophages deficient in TREM-2 are hyperresponsive to TLR ligands (29). We therefore used another approach to inhibit TREM-1 by LP17 which abrogated eCIRP-induced inflammation and ALI, confirming the notion that the eCIRP-TREM-1 interaction is functionally dynamic. LP17 is a 17 amino acid peptide whose sequence is taken from TREM-1; it is thought to function as a decoy receptor (21). We attempted to determine the interaction between LP17 and eCIRP using SPR and demonstrated a KD of 48 x 10 -6 M between these two molecules (data not shown). This could be artificially elevated due to conditions present in vivo that we cannot reproduce artificially, such as involvement of other biological co-factors, temperature, and molecular stability. The TREM-1 antagonists function primarily as decoy receptors and to prevent receptor dimerization (21,30). Additionally, Sigalov et al has developed a ligand-independent inhibitor of TREM-1 (31,32). Although use of these peptides in animal models of inflammatory diseases has resulted in decreased inflammation and can be absorbed directly into the systemic circulation via blood capillaries, whereas larger peptides are absorbed initially through the lymphatic system or in a combination of both blood and lymphatic absorption (33,34). Additionally, because patient compliance is increased with non-invasive administration, small peptides have an advantage over larger peptides as they can be administered through mucosal absorption. In the current study, we demonstrated M3 had an excellent inhibitory role in mitigating inflammation not only in macrophage cells but also in bacterial sepsis. Prior to usage of M3 in large animal models or human clinical trials, further studies on the stability, half-life, and safety would be required. However, at the doses used here, M3 did not demonstrate immunogenicity or tissue injury as demonstrated by cytokine levels, organ injury markers, or histologic analysis. In addition, cell proliferation assays in the presence of 100 µg/mL of M3 peptide -a dose 10x higher than doses used in vitro in this studydemonstrated no toxicity (Supplemental Fig 6).
In addition to eCIRP's interaction with TREM-1, we also found that the expression of TREM-1 in macrophages was increased following treatment with eCIRP. Several studies have established the connection between TLR4 and TREM-1 in immune cells (14,35,36). The cellular consequences of TREM-1 activation following treatment with anti-TREM-1 Ab were examined in terms of global gene expression and compared with cells receiving LPS and also cells receiving a combined treatment of anti-TREM-1 plus LPS (14). The results indicated that besides having the crosstalk with TLR4, TREM-1 signaling independently promoted the expression of some unique genes (14), implying the direct impact of TREM-1 pathway in inflammation. As previously discussed, like other DAMPs, eCIRP has been shown to recognize TLR4 (5). TLR4 activation leads to upregulation of TREM-1 expression which is MyD88dependent and involves transcription factors NF-κB, PU.1 and AP1 (37). Additionally, simultaneous activation of TREM-1 and TLR4 leads to synergistic production of proinflammatory mediators through common signaling pathway activation including PI3K, ERK1/2, IRAK1 and NF-κB activation (13)(14)(15)(16). Thus, the upregulation of TREM-1 expression in macrophages after treatment with eCIRP could be promoted through both TLR4 and TREM-1dependent pathways. The connection between TLR4 and TREM-1 has been shown in neutrophils: following LPS stimulation of neutrophils, TREM-1 was found to be co-localized with TLR4 (36). Genetic deletion of TREM-1 down-regulates the expression of several genes implicated in the TLR4 pathway including MyD88, CD14 and IκBα (35). Our data demonstrating the eCIRP-induced expression of pro-inflammatory cytokines by the macrophages also supports the synergistic effects of TREM-1 and TLR4. However, to demonstrate the independent pro-inflammatory role of the TREM-1 in eCIRP-mediated inflammation, we activated TREM-1 using agonist antibody crosslinking (13,14). TREM-1 activation resulted in increased levels of TNF-α; however, TNF-α production was diminished in cells treated with M3.
M3 interfered with the crosslinking effect demonstrating a TLR4 independent interaction between eCIRP and TREM-1.
In summary, we have discovered that eCIRP is a new ligand of TREM-1 in macrophages. Animals were randomly assigned to sham, vehicle, or treatment group. Every attempt was made to limit the number of animals utilized per experiment.
Only male mice were used in this study. Studies have found sex-specific differences in sepsis (38). It has been reported that male and female sex steroids exhibit diverse immunemodulating functions in both humoral and cell-mediated immune responses under normal conditions and various disease processes (38). Male sex hormones have been shown to suppress cell-mediated immune responses and cardiovascular function while female sex hormones have been shown to be protective. Survival rates after CLP have been shown to be greatly increased in female mice as compared to male mice (39). Given the impact of sex on sepsis pathogenesis, to generate reliable and consistent findings, only male mice were used.

Animal model of polymicrobial sepsis
Mice were anesthetized with inhaled isoflurane and placed in supine position. Cecal ligation and puncture (CLP) was preformed through a midline laparotomy (40,41). Briefly, the abdomen was shaved and disinfected. A 2-cm incision was created, the cecum was exposed, and ligated with a Analgesics and sedatives to mitigate pain and discomfort in septic mice can modulate immune responses in sepsis. Studies have shown that the levels of proinflammatory cytokines were reduced by fentanyl and/or midazolam, which may contribute to the beneficial effects of these medications in septic mice. Additionally, analgesia can reduce inflammatory responses in septic mice, resulting in immunosedation and contributing to improved outcomes and mortality rates in murine sepsis (42). While we did use antibiotics for the survival studies, we did not use antibiotics for the short-term, 20 hour, experiments. The inflammatory response in sepsis depends, at least in part, on bacterial load. Since the use of antibiotics reduce bacterial contents ,and therefore inflammation, antibiotics were not used in the short-term experiments in order to allow for a rapid, robust, inflammatory response. All experiments were in compliance with IACUC guidelines. Additional environmental enrichment and comfort was provided to all mice in the form of pressed cotton squares (43) and Shepard Shacks (44).

In vivo administration of rmCIRP, LP17 and M3
rmCIRP was produced in our lab as previously described (5). rmCIRP at a dose of 5 mg/kg BW or normal saline was administered intravenously (i.v.) via retro-orbital injection using an h nonadherent cells were removed and adherent cells, primarily macrophages, were cultured overnight prior to use. All cultured media was supplemented with 10% heat-inactivated fetal bovine serum (FBS, MP Biomedicals, Solon, OH), 1% penicillin-streptomycin and 2 mM glutamine. Cells were maintained in a humidified incubator with 5% CO 2 at 37 °C.

Determination of organ injury markers
Serum levels of lactate dehydrogenase (LDH), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and lactate were determined using specific colorimetric enzymatic assays (Pointe Scientific, Canton, MI) according to the manufacturer's instructions.

Lung pathohistology
Lung tissues were fixed in 10% formalin prior to being embedded in paraffin. Tissues were cut into 5 μm sections and stained with hematoxylin and eosin (H&E). Slides were evaluated under light microscopy to evaluate the degree of lung injury. Scoring was done using a system created by the American Thoracic Society (46). Scores ranged from zero to one and were based on the presence of proteinaceous debris in the airspaces, the degree of septal thickening, and neutrophil infiltration in the alveolar and interstitial spaces. The average score per field was calculated at 400× magnification.

Assessment of TREM-1 expression in macrophages by flow cytometry
To detect TREM-1 expression on the surface of macrophages, a total of 1 × 10 6 RAW264.7 or primary peritoneal macrophages were washed with FACS buffer containing PBS with 2% FBS and stained with APC anti-mouse TREM-1 Ab (clone: 174021, R&D systems). Unstained cells were used as a negative control to establish the flow cytometer voltage setting. Acquisition was performed on 10,000 events using a BD LSR Fortessa flow cytometer (BD Biosciences, San Jose, CA) and data were analyzed with FlowJo software (Tree Star, Ashland, OR).

Real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR)
Total RNA was extracted from tissue using Trizol reagent (Invitrogen, Carlsbad, CA). cDNA was synthesized using MLV reverse transcriptase (Applied Biosystems, Foster City, CA). PCR reactions were carried out in 20 μl of a final volume of 0.08 μM of each forward and reverse primer (Supplemental Table 1), cDNA, water, and SYBR Green PCR master mix (Applied Biosystems). Amplification and analysis was conducted in a Step One Plus real-time PCR machine (Applied Biosystems). Mouse β-actin mRNA was used as an internal control for amplification and relative gene expression levels were calculated using the DDCT method.
Relative expression of mRNA was expressed as fold change in comparison with sham tissues.

Stimulation of cytokine production by TREM-1 engagement
To activate RAW264.7 cells through TREM-1, 96-well flat bottom plates were pre-coated with 20 μg/ml of an agonist anti-TREM-1 mAb (clone 174031; R&D Systems) overnight at 37°C, similar to previously described protocols (13,15). The wells were washed with sterile PBS and 5 × 10 4 cells/well were plated. Cells treated with M3 or scramble M3-Sc1 were premixed with the peptide for 30 min before adding to the well. TNF-a production was measured in the culture supernatants after an additional 24 h of incubation.

Surface plasmon resonance
To examine the direct interaction between eCIRP and TREM-1, surface plasmon resonance objective (Zeiss, Oberkochen, Germ). Images were analyzed and quantified by using the ZenBlue software (Zeiss, Oberkochen, Germany).

FRET analyses
FRET analysis was performed as described previously (47

Cell proliferation assay
To confirm that there were no cytotoxic effects from LP17 or M3, a cell proliferation assay (CellTiter 96 AQueous Non-Radioactive Cell Proliferation Assay, Promega, Madison WI) was performed according to the manufactor instructions. Briefly, RAW264.7 cells were plated in 96well plates and treated with PBS, 1 ug/mL CIRP, or 1 ug/mL CIRP in addition to various doses of the inhibitory peptide. Cell viability was determined by colometric analysis at 490 nm.

Statistical analysis
Data represented in the figures are expressed as mean ± SE. All data has been tested for normality using the Kolmogorov-Smirnov Test of Normality. Normally distrubted data was analyzed using One-way ANOVA for comparison among multiple groups and the significance between individual groups was determined by the Tukey method. The two-tailed Student's t test was applied for two-group comparisons. Nonparametic one-way comparison among multiple groups was performed using the Kruskal-Wallis test with a Dunn's multiple comparison test. The specific tests used for each graph are identified in the figure ledgends. Significance was considered for p ≤ 0.05 between study groups. Data analyses were carried out using GraphPad Prism graphing and statistical software (GraphPad Software, San Diego, CA).

Study approval
All experiments were performed in accordance with the guidelines for the use of experimental animals by the NIH and were approved (protocol number: 2018-026) by the Institutional Animal Care and Use Committee.

Grants
This study was supported by the National Institutes of Health (NIH) grants R35GM118337 (P.W.) and R01GM129633 (M.A.).