TICAM-1/TRIF associates with Act1 and suppresses IL-17 receptor–mediated inflammatory responses

TICAM-1/TRIF, a TLR3 adaptor molecule, associates with Act1 to inhibit the interaction between IL-17RA and Act1, resulting in attenuated IL-17-mediated inflammatory responses.


Introduction
Pattern recognition receptors (PRRs) are essential for sensing pathogens and initiating the innate immune responses. Toll-like receptors, RIG-I-like receptors, and Nod-like receptors as well as cGAS and C-type lectins play crucial roles in detecting infection of microorganisms. TLRs require several adaptor molecules to induce the innate immune response. TICAM-1 (also called TRIF), is the sole adaptor of TLR3 and triggers the signal to induce the expression of type I IFN and pro-inflammatory cytokines in response to viral dsRNA (Yamamoto et al, 2002;Oshiumi et al, 2003a). MyD88, another adaptor molecule of TLRs, is involved in the signaling of all TLRs except TLR3 (Akira & Takeda, 2004;Kawai & Akira, 2011).
Both TICAM-1 and MyD88 have a toll/interleukin-1 receptor (TIR) domain essential for TLR binding. MyD88 binds to TLRs as well as IL-1R and IL-18R via its TIR domain; thus, MyD88 is involved not only in TLR signaling but also in IL-1 and IL-18 signaling (Adachi et al, 1998;Burns et al, 1998). TICAM-1 also uses TIR domain to binds TLR3 and to induce type I IFN expression in response to TLR3 ligands (Funami et al, 2008). Previous studies have shown that TLR3 and TICAM-1 play crucial roles in antiviral innate immune responses (Wang et al, 2004;Zhang et al, 2020). Because the TIR of TICAM-1 does not bind to those of IL-1 and IL-18, TICAM-1 is believed to not be involved in cytokine receptor-mediated signaling.
In addition to TLR3, TICAM-1 is required for other PRR-mediated signaling. TICAM-1 associates with TICAM-2 (also called TRAM) and functions as an adaptor of TLR4 (Fitzgerald et al, 2003;Oshiumi et al, 2003b). cGAS is a cytoplasmic viral dsDNA sensor and requires STING to induce type I IFN expression (Chan & Gack, 2016;Chen et al, 2016;Takashima et al, 2018). STING is reported to require TICAM-1 for triggering the innate immune responses (Wang et al, 2016). In addition, TICAM-1 is involved in the DDX1, DDX21, and DHX36 protein complex to sense cytoplasmic dsRNA and induces antiviral innate immune responses (Ruan et al, 2019). TICAM-1 functions as an adaptor for the PRRs and is crucial for antiviral innate immune responses, whereas it is also reported that Ticam-1 KO mice develop severe symptoms of experimental autoimmune encephalomyelitis (EAE), a mouse model of multiple sclerosis (Guo et al, 2008). But, it remains unclear whether TICAM-1 is directly involved in IL-17 signaling.

Results
Act1 physically interacts with the TICAM-1 adaptor TICAM-1, IL-17RA, and Act1 contain a TIR or SEFIR domain ( Fig S1A). We compared amino acid sequences of SEFIR and TIR domains of IL-17Rs and TLRs. A phylogenetic tree implied that the node of Act1 was relatively close to that of TICAM-1, TICAM-2, and TIRAP (Fig 1A), and alignment of SEFIR and TIR domains indicated similarities among the TIR and SEFIR domains (Fig 1B), as previously reported (Novatchkova et al, 2003;Mellett et al, 2015). We compared the expression profile of TICAM-1 to those of TLRs and IL-17RA. TLR4 and IL-17RA were expressed in various cell types, whereas TLR3 was specifically expressed in CD8 + DCs ( Fig 1C). This is consistent with previous observation that TLR3 is highly expressed in CD8 + DCs (Schulz et al, 2005). Although TICAM-1 expression level was high in CD8 + DCs, it was also expressed in other cell types (Fig 1C), implying the involvement of TICAM-1 in TLR3-independent pathways. Microarray data in Reference Database of Immune cells also show that the expression profiles of TLR3 and TLR4 are different from that of TICAM-1 ( Fig S1B).
Because TIR and SEFIR domains are protein-protein interaction domains, we investigated whether TICAM-1 binds to Act1. As expected, HA-tagged TICAM-1 was co-immunoprecipitated with FLAG-tagged Act1 ( Fig 1D). HA-tagged TIR domain of TICAM-1 (TICAM-1-TIR-HA), but not HA-tagged TICAM-1 without TIR domain (TICAM-1-ΔTIR-HA), was co-immunoprecipitated with FLAGtagged Act1 (Fig 1E). In addition, deletion of SEFIR domain of Act1 abrogated the interaction between TICAM-1 and Act1 ( Fig 1F). These data suggest that the TIR and SEFIR domains are important for the physical interaction. Next, we investigated the physical interaction of endogenous proteins in cells stimulated with IL-17A. Endogenous Act1 was co-immunoprecipitated with endogenous TICAM-1 in the presence and absence of IL-17A stimulation (Figs 1G and S1C). Endogenous TICAM-1 was also co-immunoprecipitated with endogenous Act1 (Fig 1H and I). These physical interactions between TICAM-1 and Act1 were a little reduced at later time points after IL-17A stimulation ( Fig 1G and H) (see the Discussion section). To further confirm the physical interaction between TICAM-1 and Act1, we performed a proximity ligation assay (PLA), which detects the colocalization of two proteins. As expected, the PLA signals showing the colocalization of FLAG-tagged Act1 with HA-tagged TICAM-1 were detected in the cytoplasm ( Fig S1D). Moreover, the PLA signals of colocalization between endogenous TICAM-1 and Act1 were also detected in the cytoplasm, and the number of signals was reduced after IL-17A stimulation ( Fig 1J). These data indicate that TICAM-1 binds to Act1 in resting and IL-17Astimulated cells.
Because previous studies have reported that Act1 directly stabilizes CXCL1 mRNA, thereby increasing the mRNA level (Herjan et al, 2018), we investigated whether TICAM-1 is involved in Act1-mediated stabilization of the mRNA. Transcriptions were inhibited by actinomycin D treatment, and the degradation of RNAs was determined by RT-qPCR. However, there were no significant differences in the degradation of CXCL1 mRNA in IL-17A-treated and untreated cells ( Fig S2B).

TICAM-1 suppresses IL-17RA-mediated signaling
Because the NF-κB pathway is activated through IL-17A stimulation (Wu et al, 2015), we assessed the role of TICAM-1 in IL-17A-mediated NF-κB activation. Phosphorylation of the IκB and NF-κB subunit p65 anti-TICAM-1 antibody. The proteins were subjected to SDS-PAGE and detected by Western blotting with indicated antibodies. The data were representative of three independent experiments. (H, I) HeLa cells were either left untreated or treated with 100 ng/ml IL-17A. (H) WCEs were immunoprecipitated with control IgG and anti-Act1 antibody (H). The data were representative of three independent experiments. The band intensities of immunoprecipitated TICAM-1 from unstimulated cells were measured and normalized to the average band intensity of immunoprecipitated TICAM-1 of samples immunoprecipitated with anti-Act1 antibody. (I) The results show ±SD means (n = 3 *P < 0.05; t test) (I). (J) HeLa cells were stimulated with 0 or 100 ng/ml of IL-17A for 15 min, and then proximity ligation assay was performed with anti-Act1 and anti-TICAM-1 antibodies (left panel). The proximity ligation assay signal intensities were normalized to those of DAPI (right panel). The results show means ± SD (*P < 0.05; t test). The white bar represents 10 μm. The results show ± SD means (n = 3 *P < 0.05; t test). (C, D) WT and TICAM-1 KO HeLa cells were stimulated with 50 ng/ml IL-17A, and the expression of CXCL1 and CXCL2 mRNA was determined by RT-qPCR. The results show ± SD means (n = 3 *P < 0.05; t test). (D) Another clone (KO2) was used in panel (D). The results show ± SD means (n = 3-4 *P < 0.05; t test). (E) Ticam-1 or negative control siRNA were transfected into L929 cells, which were then stimulated with IL-17A. The expression of Cxcl1 mRNA was determined by RT-qPCR. The results show ± SD means (n = 4 *P < 0.05; two-way ANOVA). (F) Ticam-1 or negative control siRNA were transfected into RAW264.7 cells, which were then stimulated with IL-17A. The expression of IL-6 mRNA was determined by RT-qPCR. The results show ± SD means (n = 3 *P < 0.05; t test). (G) WT or Ticam-1 KO MEFs were stimulated with IL-17A, and the expression of CXCL1 and CXCL2 mRNA was determined by RT-qPCR. (n = 4 *P < 0.05; two-way ANOVA). WT or TICAM-1 KO HeLa cells were stimulated with IL-17A (50 ng/ml) for the indicated time, and whole cell extracts (WCEs) were blotted with anti-phospho-IκB (p-IκB), IκB, TICAM-1, and β-actin (A) and anti-phospho-p65 (p-p65), p65, TICAM-1. and β-actin (B) antibodies as indicated. Each band intensity of p-IκB and p-p65 was measured and normalized to that of total IκB and p65, respectively. Fold increases were calculated by dividing the normalized intensity at each time point with that at 0 min in WT HeLa cells. The data were representative of two independent experiments. (C) WT and TICAM-1 KO HeLa cells were transfected with a reporter plasmid containing the NF-κB promoter and Renilla luciferase vector (internal control). After 18 h, cells were stimulated with 50 ng/ml IL-17A. NF-κB reporter activities were is a hallmark of NF-κB pathway activation (Karin & Ben-Neriah, 2000). IL-17A stimulation induced IκB and p65 phosphorylation (p-IκB and p-p65), which was augmented by TICAM-1 KO in HeLa cells (Fig 3A and B). Moreover, reporter gene analysis revealed that IL-17A-induced NF-κB reporter activity was significantly augmented by TICAM-1 KO (Fig 3C), suggesting that TICAM-1 suppresses the activation of NF-κB in response to IL-17A. MAP kinases, such as p38, are activated by IL-17A stimulation (Hata et al, 2002), and we found that p38 phosphorylation (p-p38) markedly increased by TICAM-1 KO in HeLa cells ( Fig 3D).
Next, we compared NF-κB activation in primary cells. Ticam-1 KO increased IL-17A-mediated p-p65 levels in MEFs ( Fig 3E). Collectively, these data indicate that TICAM-1 suppresses the activation of transcription factors downstream of IL-17RA, which is consistent with the notion that TICAM-1 suppresses IL-17RA-mediated cytokine expression.
Next, we investigated the post-translational modifications of TICAM-1 after IL-17A stimulation. First, we tried to detect polyubiquitin chains, but we could not detect polyubiquitination of TICAM-1 after IL-17A stimulation at least in our experimental condition ( Fig S3A). Second, we investigated the phosphorylation of TICAM-1 after IL-17A stimulation. To detect the phosphorylation, we used Phos-tag gel, in which phosphorylated proteins migrate more slowly than unphosphorylated proteins. However, we could not detect any delayed migration of TICAM-1 bands after IL-17A stimulation ( Fig S3B).

TICAM-1 disrupts the interaction between IL-17RA and Act1
Next, we investigated the underlying mechanism of how TICAM-1 inhibited IL-17RA signaling. We hypothesized that TICAM-1 would disrupt the IL-17RA-Act1 interaction. To test this hypothesis, we performed immunoprecipitation assays. Myc-tagged IL-17RA was co-immunoprecipitated with FLAG-tagged Act1, and TICAM-1 overexpression reduced the interaction between Act1 and IL-17RA in HEK293FT cells (Fig 4A). We repeated the experiments and confirmed that the reduction of the interaction was statistically significant ( Fig  4B). Next, endogenous IL-17RA was co-immunoprecipitated with endogenous Act1 in the presence and absence of IL-17A, and TICAM-1 overexpression inhibited the physical interaction between IL-17RA and Act1 in HeLa cells ( Fig 4C). Next, we investigated whether the TIR domain of TICAM-1 was responsible for the inhibition. Expression of the TICAM-1 TIR fragment inhibited the interaction between ectopically expressed Act1 and IL-17RA ( Fig S4). Moreover, the expression of the TIR domain inhibited the interaction between endogenous IL-17RA and Act1 upon IL-17A stimulation (Fig 4D), suggesting that the TIR domain is responsible for the inhibition. Conversely, TICAM-1 KO increased the interaction between endogenous Act1 and IL-17RA in the presence of IL-17A (Fig 4E).
We conducted a PLA to investigate whether TICAM-1 influence the colocalization of IL-17RA with Act1 upon IL-17A stimulation. The number of the PLA signals of IL-17RA and Act1 increased after IL-17A stimulation in WT HeLa cells (Fig 4F), and TICAM-1 KO further increased the number of the PLA signals upon IL-17A stimulation in HeLa cells (Fig 4G and H). These data indicate that TICAM-1 interferes with the physical interaction between Act1 and IL-17RA. This model is consistent with our observation that TICAM-1 KO enhanced IL-17RA-mediated cytokine expression.

TICAM-1 deficiency promotes IL-17-induced inflammatory response in vivo
Our in vitro results indicate that TICAM-1 suppresses IL-17Amediated expression of cytokines and chemokines, such as CXCL1 and CXCL2, which play crucial roles in recruiting myeloid cells, such as CD11b + Gr1 + cells (Acharyya et al, 2012;Girolomoni et al, 2012). Next, we assessed the physiological significance of TICAM-1mediated inhibition of IL-17A signaling in vivo. We injected recombinant IL-17A intraperitoneally into WT and Ticam-1 KO mice and evaluated the accumulation of inflammatory cells in the peritoneal cavity. Absolute number of total cells increased in the peritoneal cavity upon IL-17A stimulation, and the accumulation was significantly augmented by TICAM-1 KO ( Fig 5A). Moreover, Ticam-1 KO mice displayed excessive accumulation of Gr-1 + CD11b + neutrophils in the peritoneal cavity (Fig 5B-D) as well as Gr-1 − CD11b middle CD11c − F4/80 − monocytes (Fig 5B and E) after peritoneal injection of recombinant IL-17A compared to WT. Furthermore, Ticam-1 KO mice intranasally injected with IL-17A also displayed excessive accumulation of Gr-1 + CD11b + neutrophils in the lung ( Fig  5F and G). We confirmed that Ticam-1 KO increased mRNA expression of CXCL1, CXCL2, and IL-6 in the lung in response to IL-17A administration (Fig 5H and I). Considering that TICAM-1 augmented the inflammatory response to IL-17A without TLR ligands, our data indicate a crucial role of TICAM-1 in regulating IL-17A-mediated immune response in vivo.
determined 6 h after IL-17A stimulation. The results show ± SD means (n = 3 *P < 0.05; t test). (D) WT or TICAM-1 KO HeLa cells were stimulated with IL-17A (50 ng/ml) for the indicated time, and WCEs were blotted with anti-phospho-p38 (p-p38), p38, TICAM-1, and β-actin antibodies as indicated. Each band intensity of phosphor-p38 (p-p38) was measured and normalized to that of total p38. Fold increases were calculated by dividing the normalized intensity at each time point with that at 0 min in WT HeLa cells. The data were representative of two independent experiments. (E) WT and Ticam-1 KO MEFs (E14) were stimulated with 50 ng/ml IL-17A for the indicated time. WCEs were blotted with anti-p-p65, p65, and β-actin antibodies. Each band intensity of p-p65 was measured and normalized to that of total p65. Fold increases were calculated by dividing the normalized intensity at each time point with that at 0 min in WT MEFs. The data were representative of two independent experiments. Next, we conducted mouse EAE assay. Clinical symptoms of EAE were exacerbated by Ticam-1 KO (Fig 6B). Absolute number of total cells in draining lymph nodes increased by Ticam-1 KO (Fig 6C), and spleen size was enlarged (Fig 6D) in EAE-induced Ticam-1 KO mice, suggesting that the inflammatory response is augmented in Ticam-1 KO mice during EAE pathogenesis.
Considering that TICAM-1 suppressed IL-17A-mediated chemokine expression, we expected that Ticam-1 KO would increase the accumulation of myeloid and lymphoid cells in EAE-induced mice. Flow cytometry analysis of the spinal cord demonstrated that the percentage of central nervous system-infiltrating CD4 + and Gr-1 − F4/80 + cells significantly increased in Ticam-1 KO mice compared with WT mice at 19 d post EAE induction (Fig 6E and F). Next, we investigated whether the severity of EAE in Ticam-1 KO mouse is dependent on IL-17RA-mediated signaling. We intraperitoneally injected anti-IL-17A blocking antibody during the induction of EAE and found that the blocking of IL-17A-mediated signaling inhibited EAE symptoms in both of WT and Ticam-1 KO mice (Fig 6G), suggesting that EAE was induced in an IL-17A-dependent manner both in WT and Ticam-1 KO mice in the experimental condition. Collectively, our data indicate that TICAM-1 is involved in IL-17Amediated immune responses in vivo.

Discussion
TICAM-1 has been reported to be the solo adaptor molecule of TLR3, and in this study, we elucidated another role of TICAM-1 in IL-17A-mediated signaling. Previously, Mellett et al have reported a cross talk between TLRs and IL-17Rs, in which the SEFIR domain of IL-17RD inhibits TLR signaling (Mellett et al, 2015). Our findings provide another mechanism underlying the cross talk between TLRs and IL-17Rs. It is expected that the difference in TICAM-1 expression levels would make a difference in the cytokine expression profile after stimulations with TLR3 and/or IL-17RA ligands.
TIR domain resembles the SEFIR domain in structure (Maitra et al, 2007). Since TIR and SEFIR domains are responsible for protein-protein interactions, we hypothesized that TICAM-1 would associate with SEFIR domain containing proteins and found the interaction between TICAM-1 and Act1. Although TICAM-1 promotes TLR3-mediated type I IFN expression, TICAM-1 inhibited the Act1-IL-17RA interaction, thereby attenuating IL-17A-mediated cytokine and chemokine expression. Thus, expression levels of TICAM-1 are expected to determine the strengths of TLR3 and IL-17RA signaling. In addition, we found that IL-17A stimulation reduced the interaction between TICAM-1 and Act1. This is correlated with previous model that IL-17A induces the interaction between IL-17RA and Act1 (Qian et al, 2007).
It is still unclear how TICAM-1 inhibits the interaction between IL-17RA and Act1. It is possible that TICAM-1 directly disrupt the physical interaction between IL-17RA and Act1. Another possibility is that TICAM-1 sequestrates Act1, thereby keeping Act1 away from IL-17RA. In our experimental conditions, Act1 bound to IL-17RA immediately after IL-17A stimulation, whereas TICAM-1 dissociated from Act1 at later time points. Thus, dissociation of TICAM-1 from Act1 would not be a prerequisite for the binding of Act1 to IL-17RA, but it might modulate the binding stability of Act1 to IL-17RA. Another possibility is that other factors, such as TICAM-2/TRAM, affected their dissociation and/or associations. It is also possible that post-translational modification affected the TICAM-1-mediated inhibition. Some experimental conditions, such as transfection of plasmids, might affect the sensitivity of cells to IL-17A. Further studies are required to fully reveal the underlying mechanism.
IL-17A triggers the expression of inflammatory responses, such as chemokine expression, leading to accumulation of myeloid and lymphoid cells. Because TICAM-1 attenuated the interaction between IL-17RA and Act1, it was expected that Ticam-1 KO would accelerate recruitment and accumulation of the immune cells. Indeed, Ticam-1 KO increased accumulation and/or recruitment of myeloid cells. These observations indicate an important in vivo role of TICAM-1-mediated inhibition of the interaction in ameliorating excessive IL-17A-mediated inflammatory responses.
IL-17A are important for hyper sensitivities and autoimmune diseases, and it has been reported that IL-17A promotes mouse EAE (Ogura et al, 2008). Our study showed that Ticam-1 KO exacerbated mouse EAE. It is still possible that Ticam-1 KO affected mouse EAE via TICAM-1-dependent TLR responses. For instance, a previous study has shown that TICAM-1-dependent type I IFN expression promotes the production of anti-inflammatory cytokine IL-27 that attenuates IL-17 production, thereby ameliorating mouse EAE (Guo et al, 2008). However, we prefer the interpretation that not only TICAM-1-dependent type I IFN production but also TICAM-1mediated inhibition of IL-17RA and Act1 interaction ameliorate mouse EAE. This is because our in vitro studies showed that TICAM-1 inhibited the interaction between Act1 and IL-17RA even in the absence of any TLR simulation. Moreover, IL-17A-induced myeloid cell accumulation in vivo was enhanced by Ticam-1 KO in the absence of TLR stimulation. Therefore, we expect that TICAM-1 would directly and indirectly inhibit IL-17RA signaling.
Several viral infections reportedly trigger the development of IL-17-mediated autoimmune disease (Getts et al, 2013). Because several viruses target TICAM-1 to infect host cells, our findings imply that viruses might affect not only TLR but also IL-17 signaling pathways. IL-17RA signaling is also involved in the pathogenesis of acute respiratory distress syndrome (ARDS) caused by viral infection (Crowe et al, 2009). Cytokine release syndrome in COVID-19 patients sometimes results in ARDS because of pulmonary edema  (F, G) WT and Ticam-1 KO female mice were intranasally injected with PBS or IL-17A (2 μg). 5 h after injection, the lung was excised, and cells were prepared. Intravascular CD45 staining was performed just before mice were euthanized. (F, G) Representative plots (F) and and lung failure. Several groups reported that peripheral Th17 cells and IL-17A in serum are higher in COVID-19 patients than healthy donors (Xu et al, 2020;Ghazavi et al, 2021). PRRs and these adaptor molecules are also involved in ARDS caused by viral infection, and it has been reported that TICAM-1 had a protective role in ARDS caused by SARS-CoV, showing increased mortality, weight loss, and reduced lung function in Ticam-1 KO mice (Totura et al, 2015). Further studies are required to assess physiological significance of TICAM-1-mediated regulation of IL-17A signaling in ARDS. Recently, IL-17-dependent immune responses have been intensively studied in viral infectious diseases (Parrot et al, 2020;Pierce et al, 2020). Thus, the cross talk between TLR3 and IL-17A signaling via TICAM-1 is expected to be important to understand the pathogenesis of those serious infectious diseases, including recent coronavirus disease 2019.

Contact hypersensitivity assay and in vivo IL-17 stimulation
Age-(8-12 wk) and sex-matched mice were sensitized by applying 50 μl of freshly prepared 0.5% of 1-fluoro-2,4-dinitrobenzene (DNFB) in acetone: olive oil (4:1) to the shaved abdomen. 4 d later the right ears were challenged with 10 μl of freshly prepared 1% of DNFB. Right and left ear thickness before and 24 h after the challenge was measured. The following formula was used for calculating ear swelling (%). % Ear swelling = (right ear thickness/left ear thickness) × 100. For in vivo IL-17 stimulation, mouse recombinant IL-17A (0.5 or 5 μg) or vehicle control (PBS) was intraperitoneally injected or intranasally inoculated into the mice, and 5 h later peritoneal exudate cells or lung tissues were harvested for further analysis, respectively.

EAE induction
EAE induction was previously described. Briefly, age-(8-12 wk) and sex-matched mice were immunized with 200 μg MOG 35-55 peptide (Scrum) and complete Freund's adjuvant containing 5 mg/ml (0.5 mg/mouse) heat-killed Mycobacterium tuberculosis (BD). Mice were injected with this emulsion at the tail base. Pertussis toxin (500 ng/mouse) was injected intraperitoneally on days 0 and 2. For IL-17A neutralization study, mice were treated intraperitoneally with 100 μg of anti-IL-17A (clone 17F3; BioXCell) or mouse IgG 1 isotype control antibody (clone MOPC-21; BioXCell) every 3 d, starting at day −1 of EAE immunization until day 14. Immunized mice were assessed everyday by clinical scores as follows: no clinical signs, 0; partially limp tail, 0.5; completely limp tail, 1; hind leg inhibition (hind legs fall when mice are dropped on a wire rack), 1.5; dragging of one hind leg, 2; dragging of both hind legs or paralysis of one hind leg, 2.5; paralysis of both hind legs, 3; paralysis of both hind legs and hind legs are together on one side of the body, 3.5; limp tail, complete hind legs, and mouse is minimally moving around the cage, 4; mouse is found dead due to paralysis, 5.

Western blot analysis and immunoprecipitation
Cells were lysed with lysis buffer (10 mM Hepes, 50 mM NaCl, 1 mM EDTA, 10% glycerol, 1% Triton, 30 mM NaF, and 1 mM Na 3 VO 4 ) in the presence of protease inhibitor cocktail tablet (Sigma-Aldrich). For immunoprecipitation, lysates were incubated with Protein G Sepharose beads and the antibody overnight. Samples were boiled and separated to Tris-glycine SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were incubated with the primary antibody and HRP-conjugated secondary antibodies (GE Healthcare). Immunoblots were visualized using the Amersham ECL Prime Western blotting detection reagent (GE Healthcare) and Bio-Rad ChemiDoc touch. The band intensity was measured by Image Lab software.

Reporter gene assay
HeLa cells were co-transfected with NF-κB reporter (100 ng/ml) and phRL-TK (10 ng/well Promega) plasmids. The phRL-TK vector encodes Renilla luciferase. After 18 h, cells were stimulated with 50 ng/ml IL-17A for 6 h and lysed. Luciferase and Renilla luciferase activities were determined using the Dual-Luciferase Reporter Assay system (Promega). Luminescence was detected by using a luminometer (Cat. no. AB-2270; ATTO). Luciferase activity was normalized to Renilla luciferase activity.

PLA
HeLa cells cultured on glass bottom plate were fixed with 4% PFA and were permeabilized with 0.3% Triton X-100 in PBS. PLA signals were detected by Duolink in situ PLA kit, according to the manufacturer's instruction (Sigma-Aldrich). In brief, HeLa cells were incubated with first antibody, and then secondary antibodies conjugated with the PLA probes were added. If the PLA probes are in proximity, the DNA synthesis is initiated. Finally, fluorescent probes that bind to the amplified DNA were added. We detected PLA signal intensity using microscopy software (BZ-X Analyzer) and normalized DAPI numbers.

Quantitative real-time PCR
Total RNA was isolated using the TRI reagent (MOR), according to the manufacturer's instruction, and incubated for 5 min at 65°C. cDNA was obtained using the ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO). qPCR was performed using the Power SYBR Green PCR Master Mix (Life Technologies) and Step-one Realtime PCR system (Life Technologies). Target mRNA expressions was determined by using the following primers: human ( For measurement of Cxcl1 RNA stability, cells were pretreated for 1 h with TNF-α (10 ng/ml), followed by the treatment with actinomycin D (5 μg/ml) alone or in combination with IL-17 stimulation (50 ng/ml) for various time. The decay of Cxcl1 mRNA levels was calculated using the value of relative expression of Cxcl1 mRNA at 0 min (starting time point) of each stimulatory condition as 100%.

Statistical analysis
Data were analyzed using unpaired t test when comparing two experimental groups. The Mann-Whitney test was used as a nonparametric alternative to t test. Multiple comparisons were performed using one-way ANOVA followed by the Tukey-Kramer monocytes (WT n = 8, KO n = 5), and Gr-1 + F4/80 − neutrophils (WT n = 10, KO n = 7) in the spinal cord on day 19 after immunization. Cells were collected and then stained with indicated antibodies, and were subjected to flow cytometry analyses. Samples were extracted from WT and Ticam-1 KO mice with EAE clinical score S 1. (G) EAE scores of WT and Ticam-1 KO male mice injected with 100 μg of anti-IL-17A (WT n = 18, KO n = 18) or isotype control antibody (WT n = 18, KO n = 19) on days −1, 2, 5, 8, 11, and 14 of EAE. EAE scores of WT and Ticam-1 KO mice (isotype control) at each time point were compared for statistical analyses (*P < 0.05, two-way ANOVA). Data in (B) show mean ± SD pooled from two experiments. Data in (F) show mean ± SD pooled from three experiments. Data in (G) show mean ± SD pooled from four experiments. P-values calculated using t test except for (A, B, and F). post hoc test. These analyses were performed using Prism ver.4.0. Error bars represent SD. P-values less than 0.05 were considered statistically significant.

Data Availability
Data and details of this study are available from the corresponding authors upon request.