Ecrg4 peptide is the ligand of multiple scavenger receptors

Esophageal cancer-related gene 4 (Ecrg4) encodes a hormone-like peptide that is believed to be involved in a variety of physiological phenomena, including tumour suppression. Recent progress in the study of Ecrg4 has shown that Ecrg4 is a proinflammatory factor and induces the expression of several cytokines and chemokines in macrophages/microglia. However, the detailed molecular mechanisms of Ecrg4 signalling, especially the Ecrg4 receptors, remain poorly understood. Here, using retrovirus-mediated expression cloning, we identified lectin-like oxidised low-density lipoprotein receptor-1 (LOX-1) as a membrane protein that binds amino acid residues 71–132 of Ecrg4 (Ecrg4(71–132)). Moreover, in addition to LOX-1, several scavenger receptors, such as Scarf1, Cd36 and Stabilin-1, facilitated the efficient internalisation of Ecrg4(71–132) into cells. A broad competitive inhibitor of scavenger receptors, polyinosinic acid, reduced both the binding of Ecrg4(71–132) and the activation of NF-κB in microglia. This activation was dependent on MyD88, an adaptor protein that recruits signalling proteins to Toll-like receptors (TLRs), with the consequent induction of various immune responses. These data suggest that multiple scavenger receptors recognise Ecrg4(71–132) and transduce its signals, together with TLRs, in microglia.

ScienTific RepoRts | (2018) 8:4048 | DOI: 10.1038/s41598-018-22440-4 medium contained several bands that reacted with anti-Ecrg4 and anti-human Fc antibodies ( Supplementary  Fig. S1), although Ecrg4 was shown to be translocated to the cell surface and not released into the culture medium in HEK293 cells 14 . These mixed Ecrg4-Fc proteins were used for subsequent receptor screening.
We constructed a retrovirus-based expression library 18 prepared from the cDNA of CG4 cells and infected Ba/F3 cells with the virus. Using the Ecrg4-Fc protein, we sorted Ecrg4-Fc-binding cells by flow cytometry and expanded them. After 5 rounds of this sorting and expansion cycle, we obtained a single clone (designated BaF/3 (#BEB1) thereafter), which acquires a high binding capacity for Ecrg4 ( Supplementary Fig. S2). We further prepared an expression library from BaF/3 (#BEB1) mRNA and introduced the library into the parental Ba/F3 cells. After three cycles of sorting and expansion, we observed that over 40% of the cells were strongly bound to Ecrg4 (Fig. 1A). After we obtained Ecrg4-Fc-binding single-cell clones, we identified LOX-1, which was commonly expressed in most of the clones, using retrovirus-specific primers (Fig. 1B). LOX-1 is a type II transmembrane protein with a C-type lectin-like domain (CTLD) in the extracellular region, which is required for the recognition of its ligands, such as ox-LDL 19,20 and see (Fig. 1C). The results of a microarray indicated that the LOX-1 mRNA levels were much higher in the BaF/3 (#BEB1) cells than the parental Ba/F3 cells ( Supplementary Fig. S2).
To confirm that Ecrg4 binds LOX-1, we incubated LOX-1-overexpressing HEK293 T cells with Ecrg4-Fc to demonstrate the interaction between LOX-1 and ox-LDL 21 . We verified that Ecrg4-Fc clearly bound to the LOX-1-expressing cells, whereas it did not bind to the cells expressing CTLD-deleted LOX-1, revealing that CTLD is the Ecrg4-binding domain.

Activation of NF-κB by Ecrg4(71-132) is dependent on the MyD88 adaptor protein.
Previous studies have reported that several scavenger receptors recognise danger-associated molecular patterns (DAMPs) and pathogens 23,24 and trigger a proinflammatory response in combination with members of the TLR family. Since Ecrg4(133-148) has been shown to interact with TLR4, a member of the TLR family 17 , we investigated the ChemiDoc ™ MP Imaging system (Bio-Rad), and the relative band intensity ratio NF-κB p65(pSer536)/NF-κB p65 is shown (pSer536/cont). Representative data from three independent experiments are indicated. involvement of TLRs in the function of Ecrg4(71-132). We overexpressed either TLR4 or TLR2 in HEK293 T cells, incubated with Ecrg4(71-132), and tested its internalisation into the receptor-expressing cells. As shown in Fig. 4A, we found that neither TLR4 nor TLR2 internalised Ecrg4(71-132) as effectively as LOX-1 did. To clarify the role of TLRs in Ecrg4 signalling, we used T6167923, a chemical inhibitor for MyD88 that is essential for the TLR signalling pathway 30 . Microglia were preincubated with T6167923, stimulated with Ecrg4(71-132), and examined for p65 phosphorylation and Il6 expression. The blockade of MyD88 abolished both the phosphorylation of p65 and the increased expression of Il6 ( Fig. 4B and C). These results suggested that TLRs have important roles in Ecrg4(71-132) signalling, although they did not associate with Ecrg4(71-132).

Discussion
In this study, we demonstrated for the first time that Ecrg4(71-132) binds to multiple scavenger receptors that activate the innate immune responses 24 and maintain homeostasis by clearing modified lipoproteins, pathogens and apoptotic cells 23 . These results are consistent with our previous finding that Ecrg4 peptides induce the expression of proinflammatory factors and contribute to anti-glioma immunosurveillance 11 . The diverse combination of scavenger receptors, their ligands and co-factors generates the functional versatility via a variety of intracellular signalling pathways. For example, LOX-1, which binds with ox-LDL, rapidly increased reactive oxygen species levels and activated the NF-κB signalling pathway 31,32 , while with the outer membrane protein A (OmpA), LOX-1 triggered the innate immune response via a TLR2-dependent pathway 33 . Our findings suggest that scavenger receptors, in cooperation with MyD88 dependent molecules, such as TLR family receptors, activate the Ecrg4(71-132) signal pathway, similar to the activation of pathogen-triggered scavenger receptors [33][34][35][36][37] . We observed that Ecrg4-Fc did not activate NF-κB signaling in BaF/3 #BEB1 and 293 T cells that overexpressed LOX-1 alone, LOX1 and Tlr2, or LOX-1 and Tlr4, although it was internalised into the cells (data not shown). These results indicated that the scavenger receptors are essential for internalising the peptide but not for initiating signal transduction, and that LOX-1 and unknown factor are needed for Ecrg4-dependent activation of NF-κB signaling. Interestingly, we found that Ecrg4(133-148), which differs from Ecrg4(71-132) in its recognition by scavenger receptors examined in this manuscript, also induced the expression of cytokines via the MyD88-NF-κB signalling axis (Fig. 4B,C), suggesting that it transmits inflammatory signals through an unidentified component/receptor, although Ecrg4(133-148) was shown to interact with the TLR4/CD14/MD2 complex 17 . Thus, the detailed molecular mechanisms of Ecrg4 peptides should be determined in future studies.
The scavenger receptors recognise a wide range of ligands. For example, modified lipoproteins, polyanions (polyI and others), C-reactive protein (CRP), heat shock protein, apoptotic cells and bacteria bind to LOX-1 through the CTLD 20 . In this study, we showed that Ecrg4(71-132) bound to the CTLD of LOX-1. The calculated Kd value for the binding of ECRG4 and LOX-1 is similar to those for ox-LDL and CRP. The Kd for ox-LDL and LOX-1 was 1.7 × 10 −8 M when measured in living cells 38 and 1.8 × 10 −10 M when measured from the clustered CTLDs on a sensor chip 39 . The Kd for CRP and LOX-1 is reportedly 1.6 × 10 −7 M 40 . The crystal structure of the CTLD of LOX-1 has already been investigated 41 , and thus, the in silico characterisation of the binding mode or a structural analysis of the CTLD when complexed with short peptides may clarify the features of this ligand binding reaction. Scavenger receptors are also considered a subclass of the pattern recognition receptors because they recognise not only pathogen-associated molecular patterns (PAMPs) but also modified lipids and proteins that are considered DAMPs 23,24 . Ecrg4 peptides are endogenous proteins, and therefore, they should be classified as DAMPs. Investigation of the regulatory mechanisms of Ecrg4 gene expression may provide insight into Ecrg4 functions and cellular danger signals.
In conclusion, we demonstrated that Ecrg4 interacts directly with LOX-1 and is internalised via several scavenger receptors. Since combination and mechanisms of Ecrg4/receptors may be different in each physiological phenomena and disease, therefore, the detailed mechanisms should be solved in future studies. Our findings not only support the roles of Ecrg4 in tissue homeostasis and the inflammatory response but should promote further analyses of the physiological functions of Ecrg4 and the pathogenesis of multiple diseases.

Cells and animals. HEK293 T cells and Plat-E cells were cultured in Dulbecco's modified Eagle's medium
(DMEM) containing 10% foetal calf serum (FCS). The rat oligodendrocyte precursor cell line CG4 was grown as described previously 5 . The Ba/F3 cells and a variant strain of BaF/3, BaF/3 (#BEB1) cells, which had an increased capacity to bind Ecrg4, were maintained in RPMI medium containing 10% FCS and 10 ng/mL recombinant mouse IL-3 (PeproTech). C57BL/6 mice were purchased from CLEA Japan. All mouse experiments were performed following the protocols approved by the Animal Care and Use Committee of Ehime University and Hokkaido University. Primary microglial cells were isolated from mouse pups as previously described 42 . Plasmids and other reagents. The DNA fragment encoding full-length mouse Ecrg4 was introduced into the pEF-Fc plasmid kindly provided by Dr. Nagata and used to obtain a mixture of the processed forms of the Ecrg4-Fc proteins. The expression vectors for Fc-fused Ecrg4(71-132) and Ecrg4(133-148) have been described previously 11 . For production of the C-terminal 6 × His-tagged proteins for SPR analysis, fragments of human with standard methods. The membranes were probed with antibodies against phospho-p65 (pSer536) and p65 (Cell Signaling Technology) overnight at 4 °C and then with horseradish-peroxidase-conjugated anti-rabbit IgG (GE Healthcare). Clarity Western ECL Substrate (Bio-Rad) was used for detection. In some case, the membranes were treated with stripping buffer (2% SDS, 100 mM 2-mercaptoethanol, 62.5 mM Tris-HCl, (pH 6.8)) to investigate phospho-p65 (pSer536) and p65 on the same blot.
Immunofluorescence staining. The binding and uptake of Fc-fused Ecrg4 into HEK293 T cells were determined with fluorescence microscopy. The cells were cultured on poly-l-lysine-coated glass coverslips and transiently transfected with plasmids expressing the indicated membrane proteins. Two days after transfection, the cells were incubated with several Ecrg4-Fc proteins at 37 °C for 2-3 h and then washed three times with PBS. The cells were then fixed with 4% paraformaldehyde in PBS for 10 min and treated with 0.5% Triton X-100 in PBS for 10 min. After blocking with 10% FCS, the coverslips were incubated with an anti-Flag antibody (M2, Sigma) and then with a FITC-conjugated F(ab')2 donkey anti-human IgG antibody (Jackson ImmunoResearch) and Alexa-594-conjugated goat anti-mouse IgG antibody (Thermo Fisher Scientific). The nuclei were counterstained with Hoechst. Fluorescent images were obtained with an Axio Imager A1 microscope (Carl Zeiss).  RNA preparation and quantitative real-time PCR. Total RNA was extracted with the RNeasy Mini Kit (Qiagen). The first-strand cDNA was synthesised with Transcriptor Reverse Transcriptase (Roche). Real-time PCR was performed with the StepOnePlus Real-Time PCR System (Thermo Fisher Scientific), according to the manufacturer's protocol. The sample values were normalised to those of the housekeeping gene β-actin with the ΔCt method. The primers used have been described previously 11 . Statistical analyses were carried out with Microsoft Excel software. Data are presented as the mean ± SD. Significant differences were determined using a two-sided Student's t test. A value of P < 0.05 was considered significant. DNA microarray analysis. Total RNA was prepared as described above, amplified and labelled with cyanine 3 using the one-colour Low Input Quick Amp Labeling Kit (Agilent Technologies) according to the manufacturer's instructions. The labelled cRNA was fragmented and hybridised to the Agilent Whole Mouse Genome Microarray 4 × 44 K ver. 2.0. After the microarrays were washed, they were scanned with an Agilent DNA Microarray Scanner. The intensity value for each scanned feature was quantified with the Agilent Feature Extraction software, which subtracted the background. Agilent GeneSpring GX version 12.1 was used for normalisation, as follows. First, signal intensities <1.0 were set to 1.0. Each chip was then normalised to the 75 th percentile of the measurements taken from that chip. Finally, the normalised intensities were log2-transformed. The microarray data have been submitted to the National Center for Biotechnology Information Gene Expression Omnibus (GEO) and are available under accession number GSE106400S.