sign in sign up
<<
Home , Interleukin 3, TRAF5, TRAF6, YWHAZ

14-3-3ζ–TRAF5 axis governs -17A signaling

Jenna McGowana,1, Cara Petera,1, Joshua Kima, Sonam Poplib, Brent Veermana, Jessica Saul-McBethc, Heather Contic, Shondra M. Pruett-Millerd, Saurabh Chattopadhyayb, and Ritu Chakravartia,2

aDepartment of & Pharmacology, College of Medicine & Life Sciences, University of Toledo, Toledo, OH 43614; bDepartment of Medical Microbiology & Immunology, College of Medicine & Life Sciences, University of Toledo, Toledo, OH 43614; cDepartment of Biological Sciences, College of Natural Sciences & Mathematics, University of Toledo, Toledo, OH 43614; and dDepartment of Cell and Molecular Biology, St. Jude Children’s Research Hospital, Memphis, TN 38105

Edited by Vishva M. Dixit, Genentech, San Francisco, CA, and approved August 31, 2020 (received for review April 27, 2020) IL-17A is a therapeutic target in many autoimmune . Most of target , such as IL6 and IL8 (18). On the other hand, the nonhematopoietic cells express IL-17A receptors and respond to recruitment of TRAF2 and TRAF5 participates in the stabiliza- extracellular IL-17A by inducing proinflammatory . The IL- tion of mRNA transcripts of CXCL1 and CXCL5 (19). The 17A triggers two broad, TRAF6- and TRAF5- TRAF5/TRAF2-dependent mRNA stabilization pathway is inde- dependent, intracellular signaling pathways to produce representative pendent of the presence of TRAF6 (19); however, the role of cytokines (IL-6) and (CXCL-1), respectively. Our limited TRAF5/TRAF2 in the transcriptional pathway remains unknown. understanding of the cross-talk between these two branches has In addition, IL-17A activates MAP kinases, including ERK, p38, generated a crucial gap of knowledge, leading to therapeutics and JNK, that play essential roles in autoimmune diseases (20, indiscriminately blocking IL-17A and global inhibition of its target 21). Although several regulators of each transcriptional and genes. In previous work, we discovered an elevated expression of nontranscriptional branch are known (17), there is a knowledge 14-3-3 in inflammatory aortic , a rare au- gap in understanding of cross-talk between the two branches. toimmune disorder with increased levels of IL-17A. Here we report Identifying regulators that promote one branch over the other will that 14-3-3ζ is essential for IL-17 signaling by differentially regu- provide critical information to design targeted therapeutics. lating the signal-induced IL-6 and CXCL-1. Using genetically manip- To address this, we examined the role of 14-3-3ζ in the IL-17A ulated human and mouse cells, and ex vivo and in vivo rat models, signaling. 14-3-3ζ is one of the seven members of 14-3-3 family we uncovered a function of 14-3-3ζ. As a part of the molecular that interacts with 0.6% of the cellular proteome and performs a mechanism, we show that 14-3-3ζ interacts with several TRAF pro- range of context-dependent functions, including signaling IMMUNOLOGY AND teins; in particular, its interaction with TRAF5 and TRAF6 is increased (22–24). The 14-3-3ζ regulates the signal transduction of GM- in the presence of IL-17A. In contrast to TRAF6, we found TRAF5 to CSF, IL-3, IL-5, STAT3, Toll-like (TLR)3, and AMPK, be an endogenous suppressor of IL-17A–induced IL-6 production, an among others (25–29). Additionally, it supports the production effect countered by 14-3-3ζ. Furthermore, we observed that 14-3-3ζ of several cytokines, such as IL-13, IFN-γ, and IL-17A (30, 31). – interaction with TRAF proteins is required for the IL-17A induced IL- We previously identified increased 14-3-3 levels in the ζ 6 levels. Together, our results show that 14-3-3 is an essential com- aortic tissues of patients with large-vessel (32). Large- ponent of IL-17A signaling and IL-6 production, an effect that is vessel vasculitis is a presumed autoimmune disease with in- suppressed by TRAF5. To the best of our knowledge, this report creased infiltration of Th1 and Th17 cells in the affected artery, ζ – of the 14-3-3 -TRAF5 axis, which differentially regulates IL-17A induced resulting in increased levels of IL-17A and IL-6 (33–35). In the IL-6 and CXCL-1 production, is unique. Significance 14-3-3ζ | IL-17A | TRAF

Blocking IL-17A signaling is an attractive therapy in treating nterleukin-17A (IL-17A) is an inflammatory that is autoimmune diseases. Current IL-17A blockers target both Iassociated with autoimmune diseases and host-defense (1, 2). signaling branches, leading to inhibition of TRAF6-based IL-6 Increased IL-17A level in the sera of patients with rheumatoid – production, and TRAF5-based CXCL-1 production. Global sup- arthritis, psoriasis, and lupus, is well documented (3 5). The IL- pression of both branches is of concern due to the critical role 17A blocking therapy has been clinically successful to manage – IL-17A plays in the host-defense mechanisms. Our study shows psoriasis, but not other diseases (6 8). Current IL-17A blockers that 14-3-3ζ is necessary for IL-17A–stimulated IL-6 production. lack specificity and compromise the host-defense mechanisms (9, We also found that 14-3-3ζ represses TRAF5-mediated sup- 10). There is a need to design inhibitors that can target specific pression of IL-17A–stimulated IL-6 induction. Taken together, outcomes of IL-17A signaling. To achieve such a goal, we need our results show a unique cross-talk between two branches via to improve the understanding of IL-17A signal transduction and the 14-3-3ζ–TRAF5 axis that regulates the IL-17A signaling identify the regulatory mechanisms for each set of genes. output. These foundational results can lead to the design of a The most common targets of IL-17A signaling include NF- κ – unique inhibitor targeting a specific branch of IL-17A signaling B dependent cytokines, chemokines, antimicrobial peptides and obtaining the desired output. (AMP), and matrix metalloproteases (11, 12). IL-17A plays a significant role in up-regulating the production of inflammatory Author contributions: S.P., S.C., and R.C. designed research; J.M., C.P., J.K., S.P., B.V., cytokines via either transcriptional ( expression) or post- J.S.-M., H.C., and R.C. performed research; B.V., S.M.P.-M., and R.C. contributed new transcriptional (mRNA stabilization) branches (5, 13). IL-17A reagents/analytic tools; J.M., C.P., J.K., S.P., J.S.-M., H.C., S.C., and R.C. analyzed data; signal transduction occurs via IL-17 receptors consisting of the IL- and J.M., S.C., and R.C. wrote the paper. 17RA and IL-17RC subunits (14). While IL-17RA is ubiquitously The authors declare no competing interest. expressed, IL-17RC has limited expression to nonhematopoietic This article is a PNAS Direct Submission. epithelial and mesenchymal cells (15, 16). IL-17A binding to its Published under the PNAS license. receptor results in the recruitment of the cytosolic signaling 1J.M. and C.P. contributed equally to this work. complex consisting of Act1 and several TRAF proteins [see review 2To whom correspondence may be addressed. Email: [email protected]. for further details (17)]. Recruitment of TRAF6 results in the This article contains supporting information online at https://www.pnas.org/lookup/suppl/ activation of downstream molecules, including TAK1, eventually doi:10.1073/pnas.2008214117/-/DCSupplemental. leading to translocation of NF-κB to the nucleus and

www.pnas.org/cgi/doi/10.1073/pnas.2008214117 PNAS Latest Articles | 1of10 Downloaded by guest on September 29, 2021 present study, we identified a requirement of intracellular 14-3- drives Renilla luciferase by a NF-κB promoter, and a firefly lu- 3ζ in IL-17A signal transduction. We used 14-3-3ζ knockout ciferase driven by thymidine kinase promoter in the 14-3-3ζ (KO) murine fibroblasts and human epithelial cells to examine knockdown (KD) HeLa cells, and measured luciferase activity the role of 14-3-3ζ in IL-17A signaling. Our results show that 14- after the stimulation with IL-17A (39). The ablation of 14-3-3ζ 3-3ζ is essential for IL6, IL8, and Defb3 production in IL-17A resulted in significant suppression of the IL-17A–induced NF-κB signaling; however, it has an opposite effect on CXCL1 pro- activity in a luciferase assay (lanes 2 and 4, Fig. 2A). The com- duction. To validate our finding, we engineered 14-3-3ζ global KO plementation with HA-tagged 14-3-3ζ resulted in a substantial Lewis rats using CRISPR/Cas9 and subjected them to IL-17A rescue of NF-κB activity in 14-3-3ζKD cells (lanes 4 and 6, treatment, a technique successfully used to determine the Act1 Fig. 2A). Encouraged by these results, we focused on a specific role in IL-17A signal transduction (36). Similar to cells, the 14-3- stage of NF-κB activation, the nuclear translocation of the p65 3ζKO rats show a decrease in the IL-17A–stimulated IL-6 pro- subunit of NF-κB, in IL-17A–stimulated cells. IL-17A treatment of duction but an increase in CXCL-1 level. In the absence of 14-3- ARPE-19 cells resulted in the nuclear translocation of p65, which 3ζ, IL-17A–stimulated ERK phosphorylation (pERK) and p65 was inhibited in the 14-3-3ζKO ARPE-19 cells, analyzed by confocal translocation are decreased. The 14-3-3ζ interacts with TRAFs via microscopy (Fig. 2B). Similarly, we observed IL-17A–stimulated two of its TRAF-binding motifs (TBMs), particularly TRAF5, increase in nuclear p65+ cells in WT MEFs but not in 14-3-3ζKO which bound more strongly in the presence of IL-17A. The TRAF cells (Fig. 2C). To validate the microscopic results, we performed interaction with 14-3-3ζ plays a vital role in determining the effects subcellular fractionation and evaluated the nuclear translocation of on IL-6 production. We also uncovered a suppressive role of p65. Similar results were obtained using subcellular fractionation of TRAF5 on the IL-17A–induced IL-6 levels. To the best of our ARPE-19 cells and MEFs, confirming that 14-3-3ζ was required for knowledge, this report showing a role of 14-3-3ζ, and its activity IL-17A–induced p65 nuclear translocation (Fig. 2 D and E and SI with TRAF5 in regulating IL-17A signal transduction, is unique. Appendix,Fig.S3). Taken together, our results suggest that the role of 14-3-3ζ in IL-17A signaling is upstream of the p65 nuclear Results translocation event. 14-3-3ζ Is Required for IL-17A–Mediated Gene Induction. To inves- tigate the role of 14-3-3ζ in IL-17A signaling, we took advantage 14-3-3ζ Is Required for IL-17A–Induced pERK. IL-17A stimulation of the human retinal epithelial cells (ARPE-19) and mouse results in the activation of MAP kinases, which also activate NF- embryonic fibroblasts (MEFs) as in vitro cell models, which re- κB (1, 37). To evaluate whether 14-3-3ζ is required for IL- spond to exogenous IL-17A in the absence of prestimulation by 17A–induced pERK, we used WT and 14-3-3ζKO ARPE-19 cells. TNF-α (37). IL-17A treatment resulted in robust induction of IL6, IL-17A treatment, analyzed at two different time points, caused a which was suppressed by BV02, a pan 14-3-3 inhibitor (Fig. 1A). robust increase in pERK in the WT, which was significantly sup- This suggested that 14-3-3 proteins play a role in IL-17A signal pressedin14-3-3ζKO cells (Fig. 3A). We further examined this using transduction. To determine the isoform-specific role of the 14-3-3 MEFs, which also showed increased IL-17A–induced pERK in family of proteins, we generated 14-3-3ζKO ARPE-19 cells using WT cells, but not in the absence of 14-3-3ζ (Fig. 3B). To strengthen the CRISPR/Cas9 approach (SI Appendix,Fig.S1A). The loss of our observation, we tested additional doses of IL-17A, which still 14-3-3ζ in ARPE-19 did not affect the cell viability (SI Appendix, did not trigger pERK in 14-3-3ζKO ARPE-19 cells (Fig. 3C). To Fig. S1B) or the expression of two unrelated 14-3-3 isoforms, e and validate the 14-3-3ζ requirement for pERK, we rescued its ex- γ, measured at RNA and protein levels (Fig. 1B). Importantly, the pression in 14-3-3ζKD HeLa cells by lentiviral transduction. As 14-3-3ζKO cells displayed significant inhibition of IL-17A–induced expected, complementation of 14-3-3ζKD cells with Flag-tagged 14- IL-6 at both mRNA and protein levels (Fig. 1 C and E and SI 3-3ζ rescued the IL-17A–inducedpERKascomparedtoempty Appendix,Fig.S1C), suggesting a specific role of the 14-3-3ζ iso- vector control (Fig. 3D and SI Appendix,Fig.S4). Together, our form in IL-17A–induced IL-6 expression. We tested additional IL- results indicate that 14-3-3ζ is required for IL-17A–induced pERK 17A–induced genes (e.g., IL8), which were strongly inhibited in in human and mouse cells. 14-3-3ζKO cells (Fig. 1D). To validate these results in mouse cells, we examined WT or 14-3-3ζ Forms Complex with the TRAF Proteins. Our results showed 14-3-3ζKO MEFs (38). Similar to the human cells, 14-3-3ζKO that 14-3-3ζ participates in IL-17A signaling upstream of the MEFs exhibited a substantial reduction in IL-17A–induced Il6 activation of NF-κB and ERK, which led us to focus on the and Il8 levels (Fig. 1 F and G). To further solidify the uniquely TRAF proteins, which are recruited to IL-17RA and are es- identified role of 14-3-3ζ in IL-17A–mediated gene induction, we sential for signal transduction (19, 40, 41). We examined whether rescued 14-3-3ζ expression in 14-3-3ζKO ARPE-19 cells using len- the TRAF proteins—namely TRAF2, TRAF5, and TRAF6— tiviral transductions and tested the effect on the IL-17A–stimulated can interact with 14-3-3ζ. In HEK293T cells, the ectopically IL-6 or IL-8 production. As expected, the rescue of 14-3-3ζ ex- expressed Flag-tagged TRAF proteins were successfully pulled pression resulted in increased IL-17A–induced IL6 and IL8 by HA-tagged 14-3-3ζ (Fig. 4A), suggested that 14-3-3ζ might (Fig. 1 H and I). The loss of Il6 or Il8 induction was not due to any function at the level of IL-17RA–TRAF complex formation. change in IL-17 receptor expression on the 14-3-3ζKO cells (SI Next, we determined the interaction at endogenous levels using Appendix,Fig.S1D). To evaluate the specificity of 14-3-3ζ in IL-17A rat aortic lysates and coimmunoprecipitate (co-IP) 14-3-3ζ– signaling, we studied its role in IFN-γ signaling. The IFN-γ treat- bound proteins. Both TRAF5 and TRAF6 coimmunoprecipi- ment caused increased induction of IL6 in ARPE-19 cells, which tated with 14-3-3ζ (Fig. 4B). To evaluate whether 14-3-3ζ in- was unaffected by the presence of BV02 (SI Appendix,Fig.S2A). teracts with the TRAF proteins in a signal-dependent manner, Similarly, IFN-γ–induced Il6 was unchanged in 14- we used co-IP of endogenous 14-3-3ζ in unstimulated or IL- 3-3ζKO MEFs (SI Appendix,Fig.S2B). Treatment with IL-17A or 17A–stimulated ARPE-19 cells. We observed a robust IL- BV02 did not affect the 14-3-3ζ levels (SI Appendix,Fig.S2C). 17A–dependent increase in the interaction of TRAF5, but not Collectively, our results established a role of 14-3-3ζ in IL17 TRAF2, with 14-3-3ζ (Fig. 4C). Similarly, we observed increased A–mediated gene induction. interaction of TRAF6 with 14-3-3ζ in IL-17A–treated cells (Fig. 4D). We validated these results using confocal microscopy 14-3-3ζ Is Required for IL-17A–Induced NF-κB Activation. To inves- to evaluate the interaction between the endogenous proteins. tigate how 14-3-3ζ participates in IL-6 production, we examined Using IL-17A–stimulated ARPE-19 cells, we confirmed coloc- IL-17A–stimulated NF-κB activation, a step that is critical for alization of endogenous 14-3-3ζ with TRAF5, and 14-3-3ζ with gene induction (1, 18). We coexpressed a reporter plasmid that TRAF6 (Fig. 4E).

2of10 | www.pnas.org/cgi/doi/10.1073/pnas.2008214117 McGowan et al. Downloaded by guest on September 29, 2021 Fig. 1. 14-3-3ζ is required for IL-17A–mediated gene induction. (A) ARPE-19 cells, treated with hIL-17A in the presence of 14-3-3 pan inhibitor (BV02), were IMMUNOLOGY AND INFLAMMATION analyzed for IL6 mRNA levels by qRT-PCR. (B) CRISPR/Cas9-mediated 14-3-3ζKO ARPE-19 cells were analyzed for the endogenous levels of additional 14-3-3 isoforms, as indicated, using qRT-PCR (Upper) or immunoblot (Lower). (C and D) The WT and 14-3-3ζKO ARPE-19 cells, treated with hIL-17A, were analyzed for IL6 (C)orIL8 (D) mRNA levels using qRT-PCR. (E) The WT and 14-3-3ζKO ARPE-19 cells, treated with hIL-17A, were analyzed for IL-6 protein levels using ELISA. (F and G) The WT and 14-3-3ζKO MEFs, treated with mIL-17A, were analyzed for Il6 (F)orIl8 (G) mRNA levels using qRT-PCR. The loss of 14-3-3ζ protein in the KO MEFs is also shown as an immunoblot. (H and I) Rescuing Flag-tagged 14-3-3ζ (shown by immunoblot) by lentiviral transduction of 14-3-3ζKO ARPE-19 cells support IL-17A–induced IL6 (H) and IL8 (I) mRNA levels, analyzed by qRT-PCR. All results are representative of at least three experiments, *P < 0.05, ***P < 0.0005, and ****P < 0.0001; ns, not significant.

To investigate if 14-3-3ζ directly interacts with TRAF5 or IL-17A–induced IL6 levels (SI Appendix, Fig. S6). As TRAF5 is TRAF6, we developed a cell-free in vitro interaction assay using known to promote IL-17A–stimulated CXCL1 induction (19), we the purified proteins from the transiently transfected HEK-293T measured the CXCL1 level in the TRAF5KD cells. As expected, cells (SI Appendix, Fig. S5). The TRAF5-bound FLAG-beads CXCL1 mRNA in TRAF5KD cells was significantly reduced in and 14-3-3ζ–bound HA-beads were washed with 300 mM salt comparison to NSC cells (Fig. 5C). The opposite effect of TRAF5 to remove any associated proteins to isolate TRAF5 and 14-3-3ζ on IL6 and CXCL1 gene induction in IL-17A signaling suggested proteins in near purity. The washed 14-3-3ζ tagged HA-beads were that TRAF5 may act as a suppressor of IL-6 production, but not used to affinity pull-down TRAF5 from the peptide elute of FLAG for CXCL-1. This intrigued us to evaluate the role of 14-3-3ζ in beads. The HA–14-3-3ζ beads were able to pull-down nearly puri- IL-17A–induced CXCL1 level. In contrast to TRAF5, 14-3-3ζ had fied TRAF5, suggesting that the two proteins can directly interact a suppressive effect on Cxcl1 induction (Fig. 5D). The opposite with each other in solution (Fig. 4F). Similarly, TRAF6 and 14-3-3ζ effects of 14-3-3ζ and TRAF5 on IL-17A–stimulated IL6,and showed the ability to interact directly with each other in the cell-free CXCL1 gene induction suggested the presence of a novel regu- in vitro interaction system (Fig. 4G). Since 14-3-3ζ is required in IL- latory axis that controls the signaling output. 17A signal transduction, we next addressed the functional role of To gain mechanistic insight into the TRAF5 role, we mea- 14-3-3ζ–TRAF complex in IL6 induction. sured the pERK levels, which were found to be increased in the TRAF5KD ARPE-19 cells (Fig. 5E). To examine if the increased 14-3-3ζ Suppresses the Role of TRAF5 in IL-17A Signaling. The in- pERK is responsible for the elevated IL6 induction, we used a well- crease in the 14-3-3ζ interaction with TRAF5 and TRAF6 upon known MEK inhibitor, U0126, which completely abolished the IL6 IL-17A stimulation led us to question its role in the downstream induction in TRAF5KD cells. This suggested that pERK is required signaling. Requirement of TRAF6 in IL-17A–stimulated gene for increased IL6 levels in IL-17A–stimulated TRAF5KD cells induction is well-established, but the role of TRAF5 remains (Fig. 5F). To test if the elevated IL6 in TRAF5KD is dependent on understudied (42). As TRAF6 and 14-3-3ζ have similar effects 14-3-3ζ, we generated 14-3-3ζKOTRAF5KD ARPE-19 cells that were on IL6 gene induction, we focused our attention to study the 14- deficient in both TRAF5 and 14-3-3ζ. The increase in IL- 3-3ζ-TRAF5 axis in the IL-17A–induced gene induction, and 17A–stimulated IL6 was completely absent in 14-3-3ζKOTRAF5KD used both IL6 and CXCL1 as readouts of IL-17A signaling. We cells (Fig. 5G). Collectively, we conclude that TRAF5 suppresses IL- used TRAF5KD ARPE-19 cells to examine the IL-6 production 17A–induced IL6 production by suppressing pERK, an effect which upon IL-17A treatment. Surprisingly, the IL-17A–induced IL6 is countered by 14-3-3ζ. level was significantly increased in the TRAF5KD ARPE-19 cells, compared to nonspecific target shRNA control (NSC) cells 14-3-3ζ Interaction with TRAF Is Required for IL-17–Induced IL-6 (Fig. 5 A and B). To rule out the cell-specific effect, we gener- Production. To understand role of the 14-3-3ζ–TRAF interac- ated TRAF5KD HeLa cells and confirmed TRAF5 KD promotes tion in IL-17A signaling, we questioned the basis of interaction

McGowan et al. PNAS Latest Articles | 3of10 Downloaded by guest on September 29, 2021 Fig. 2. 14-3-3ζ is required for IL-17A–mediated NF-κB activation. (A) Dual-luciferase activity was measured in NSC or 14-3-3ζKD HeLa cells, transiently coexpressing NF-κB, thymidine kinase (TK), and HA-tagged 14-3-3ζ, were treated with hIL-17A for 20 h. (B) WT or 14-3-3ζKO ARPE-19 cells, treated with hIL- 17A for 1 h, were immunostained for endogenous p65 and analyzed by confocal microscopy. (Scale bar: 10 μm.) (C) WT or 14-3-3ζKO MEFs, treated with mIL- 17A for 1 h, were immunostained for endogenous p65 and analyzed by confocal microscopy and the nuclear p65-expressing cells were quantified from at least 100 cells. (D and E) Nuclear fractions from WT and 14-3-3ζKO MEFs (D) or ARPE-19 cells (E), treated with IL-17A for 1 h, were analyzed for p65 and HDAC1 by immunoblots (shown in SI Appendix, Fig. S3), which were quantified using imageJ software. The results are representative of at least three experiments, *P < 0.05, **P < 0.005, and ***P < 0.0005.

between 14-3-3ζ and TRAF. TRAF proteins are known to rec- energy and kinetic binding constants (AYQE: ΔG −17.0 kcal/ ognize specific motifs, known as TBMs in their partners (43). We mol, Kd 3.5E-13 M, and SNEE: ΔG -15.6 kcal/mol, Kd 3.3E-12 found that 14-3-3ζ protein has two TBMs, 37SNEE and M) (SI Appendix, Table S1). 148AYQE (Fig. 6A). To assess whether the TBMs are required To examine whether these two TBMs are responsible for 14-3- for 14-3-3ζ–TRAF5 interaction, we performed in silico modeling 3ζ–TRAF5 binding, we generated site-directed mutants for both by docking the TRAF domain of TRAF5 (4GJH.pdb) on to 14-3- putative motifs (SNEE → SNAA, AYQE → AYAA) and tested 3ζ dimer (5D2D.pdb) using the ZDOCK server. We restricted these mutants for their ability to interact with TRAF5. Impor- the site of interaction to specifically these two TBMs on the 14-3- tantly, mutation of either of the two motifs (M1 and M2 mutants) 3ζ. The top 10 structures generated were screened using resulted in significant loss of interaction with TRAF5 (Fig. 6C). PRODIGY, and the structure (complex #5) with minimum in- To further confirm the loss of interaction, we used a proximity termolecular energy in the case of each motif was visualized ligation assay (PLA), which detects closely located proteins, and using Visual Molecular Dynamics (VMD) (Fig. 6B). For both established the binding of TRAF5 with WT but not with mutant motifs, the 14-3-3ζ–TRAF5 complex predicted comparable free 14-3-3ζ (Fig. 6D). To evaluate the functional consequence of this

Fig. 3. 14-3-3ζ is required for IL-17A–mediated pERK. (A–C) The WT and 14-3-3ζKO ARPE-19 (A and C) treated with hIL-17A, or MEFs treated with mIL-17A (B), for the indicated times and doses, were analyzed for pERK by immunoblot. (D) The 14-3-3ζ (Flag-tagged) restored 14-3-3ζKD HeLa cells, treated with IL-17A, were analyzed for pERK by immunoblot. EV, empty vector. All results are representative of at least three experiments, *P < 0.05, **P < 0.005, ***P < 0.0005, and ****P < 0.0001.

4of10 | www.pnas.org/cgi/doi/10.1073/pnas.2008214117 McGowan et al. Downloaded by guest on September 29, 2021 IMMUNOLOGY AND INFLAMMATION Fig. 4. 14-3-3ζ forms a complex with TRAF proteins in cells and tissues. (A) HEK293T cells were cotransfected with HA–14-3-3ζ and Flag-TRAFs, as indicated. The cell lysates were immunoprecipitated with anti-HA, and the IPs were analyzed by immunoblot. The expression of the proteins is shown in the input. (B)Rat aorticlysateswereimmunoprecipitatedwithanti–14-3-3ζ or a control IgG, and the IPs were analyzed by immunoblot, as indicated. (C and D)ARPE-19cells, treated with IL-17A, were immunoprecipitated with anti–14-3-3ζ or TRAF6, and the IPs were analyzed by immunoblot, as indicated. (E) Confocal analyses of TRAF6 and TRAF5 colocalization with 14-3-3ζ in hIL-17–treated ARPE-19 cells. (Scale bar, 10 μm.) (F and G)HA–14-3-3ζ or Flag-TRAF proteins were ectopically expressed in HEK293T cells, and isolated to near purity, and were subjected to cell-free interaction assay (SI Appendix,Fig.S5), and the complex was analyzed by co-IP.

interaction, we ectopically expressed the WT or mutant 14-3-3ζ in the in vitro studies, we measured IL-17A–mediated gene in- the 14-3-3ζKO ARPE-19 cells to measure the effect on IL- duction in the absence of any additional cytokines, such as TNF- 17A–induced IL6. The ectopic expression of WT, but not mu- α. First, we studied duodenum tissue from the heterozygous tant 14-3-3ζ, triggered strong IL6 mRNA induction (Fig. 6E), animals, which showed a strong reduction of 14-3-3ζ protein suggesting that 14-3-3ζ–TRAF interaction supports IL-17A–induced levels (Fig. 7C) and mimic our in vitro KD cell models. This IL6 levels. This decrease in the IL6 induction can be inter- allowed us to measure the effect of reduced 14-3-3ζ levels on the preted by the loss of 14-3-3ζ’s ability to either promote TRAF6- IL-17A–stimulated gene induction ex vivo. Similar to the cellular IL6 IL6 branch, or sequester TRAF5 to suppress its inhibition of studies, rat IL-17A–stimulated significant increase in Il6 in tis- induction. sues obtained from WT, but not from heterozygous animals D ζ ζ (Fig. 7 ). In contrast, tissues from heterozygous 14-3-3 animals Physiological Role of 14-3-3 in IL-17A Signaling. To evaluate the showed a significant increase in the Cxcl1 levels (Fig. 7E). Sec- physiological relevance of 14-3-3ζ in IL-17A signaling, we uti- KO ond, we performed a time course in 14-3-3ζ animals to mea- lized two approaches: Measuring the antifungal activity in a sure sera IL-6 level after the intraperitoneal injection of rat IL- newly designed bioassay, and the gene induction analyses using 17A, a strategy previously used to determine the critical com- ex vivo and in vivo animal models. To test the antifungal activity, we used the conditioned media from IL-17A–treated WT or 14- ponents of IL-17 signal transduction (36). At the basal level, both 3-3ζKO MEFs, and measured the Candida albicans killing. The WT and KO animals had similar levels of IL-6. Upon IL-17A conditioned media from IL-17A–treated WT, but not 14-3-3ζKO stimulation, at 12 h WT rats showed significantly higher IL-6 ζKO F MEFs, showed significant C. albicans killing in an in vitro bio- production compared to 14-3-3 rats (Fig. 7 ). By 72 h, a assay for antifungal activity (Fig. 7A) (44). Since the antifungal smaller yet significant difference between the WT and KO ani- ζKO role of IL-17A depends on the production of AMPs, we mea- mals was still present. In contrast to IL-6, 14-3-3 animals had sured the effect of 14-3-3ζ on the IL-17A–induced AMP, slightly higher (but not significant) CXCL-1 at 12 h, which was β-defensin 3 (BD3, encoded by Defb3). In contrast to the WT significantly increased by 72 h (Fig. 7G). Overall, results from MEFs, 14-3-3ζKO cells failed to induce Defb3 upon IL-17A in vivo studies validated our in vitro findings. Together, these stimulation (Fig. 7B). These results demonstrate a role of 14-3-3ζ results illuminate the critical role of 14-3-3ζ in IL-17A signal in IL-17A signaling and the resultant antifungal activity. transduction, and a point of cross-talk between the TRAF6- To study in vivo role of 14-3-3ζ in IL-17A signal transduction, dependent gene expression branch and the TRAF2/TRAF5-de- we engineered CRISPR/Cas9-based 14-3-3ζ global KO Lewis pendent mRNA stabilization branch of the pathway. rats (Fig. 7C). The 14-3-3ζKO rats show complete loss of the 14- 3-3ζ gene and protein in tissues (Fig. 7C). The fifth-generation Discussion 8-wk-old animals were used to assess the ex vivo and in vivo role In the present study, we identified 14-3-3ζ as a component of IL- of 14-3-3ζ in the IL-17A–stimulated gene induction. Similar to 17A signal transduction. Using in vitro, ex vivo, and in vivo

McGowan et al. PNAS Latest Articles | 5of10 Downloaded by guest on September 29, 2021 Fig. 5. TRAF5 suppresses IL-17A–stimulated 14-3-3ζ–dependent IL6 induction. (A and B) ARPE-19 cells, expressing shRNA against TRAF5 or a nonsilencing control (NSC) sequence, were analyzed for hIL-17A–induced IL6 by qRT-PCR (A) or ELISA (B). The KD of TRAF5 is shown by immunoblot (A). (C) The qRT-PCR analyses of CXCL1 mRNA in TRAF5KD ARPE-19 cells when compared to control cells, upon IL-17A treatment. (D) The qRT-PCR analyses of Cxcl1 mRNA induction in WT and 14-3-3ζKO MEFs, stimulated with mIL-17A. (E) IL-17A–stimulated pERK was analyzed in control (NSC) or TRAF5KD ARPE-19 cells by immunoblot. (F) The qRT-PCR analyses of IL6 induction in control (NSC) or TRAF5KD ARPE-19 cells in the presence or the absence of 30 μM of U0126 (MEK inhibitor), upon IL- 17A. (G) The ARPE-19–derived cells, as indicated, were analyzed for IL6 by qRT-PCR and the protein expression was validated in the immunoblot. All results are representative of at least three experiments, *P < 0.05, **P < 0.005, ***P < 0.0005, and ****P < 0.0001.

models, we confirmed the 14-3-3ζ requirement for IL-17A sig- rule out the contribution of costimuli, e.g., TNF-α, produced by naling to support IL-6 and IL-8 production, but not CXCL-1. myeloid cells, in these physiological models. The underlying mechanism of 14-3-3ζ’s role in the IL-17A sig- Like other cytokines signaling, IL-17A binding to the receptor naling depends, at least partly, upon its interaction with TRAFs, triggers a series of phosphorylation (Act1, MAPK, TPL2 kinase, which occurs upstream of pERK and p65 nuclear translocation IKK, Syk, TBK) and ubiquitination (TRAF6, HuR, USP25, events. We identified two putative TRAF-binding motifs in 14-3- A20)-based molecular events that play essential roles in the 3ζ, mutation of which disrupted the binding with TRAF5 and regulation of both de novo gene transcription and mRNA – reversed its promotional effect on IL-17A–stimulated IL-6 pro- stabilization-related outcomes (18, 47 49). Several regulators of IL6 ζ duction. Importantly, we also identified a suppressive impact of the TRAF6- branch include TRAF3, TRAF4, and IKB κ ζ TRAF5 on the IL-17A–induced IL-6 production, which was (inhibitor of NF- B- ) that affect the transcription of IL-17A dependent on the phospho-ERK and 14-3-3ζ. Our results sug- target genes (17). Similarly, specific regulators of the TRAF5- gest that increased TRAF5 interaction upon IL-17A stimulation CXCL1 branch include SF2, endoribonucleases MCPIP1 (Reg- is a possible mechanism by which 14-3-3ζ represses the TRAF5- nase-1), and Arid5a, which are responsible for regulating the mRNA stability of IL-17A target genes (19, 50, 51). It is known mediated suppression of IL-6 production. Overall, our results that TRAF6 plays no role in the TRAF5-CXCL1 branch; how- show a functional role of 14-3-3ζ in IL-17A signaling and a ever, the role of TRAF5 on the TRAF6-dependent IL6 induc- regulatory mechanism consisting of a TRAF5–14-3-3ζ axis that tion remains unknown (52). Because several growth factors can controls IL-17A signaling outcomes (Fig. 8). amplify the IL-17A signaling, it is reasonable to assume the Because IL-17A is a weak stimulator of NF-κB activity, cos- α presence of strict control by which both branches cross-talk and timulation with TNF- is often used as a strategy to activate the influence each other to regulate specific output. Our exciting IL-17A–mediated proinflammatory response in several in vitro IL6 α results show that loss of TRAF5 promotes the TRAF6- cellular models (40). We avoided the use of TNF- for our branch, suggesting that TRAF5 is an endogenous suppressor of studies because 14-3-3 proteins are known to participate in TNF- IL-17A–induced IL6 induction, which requires 14-3-3ζ and MEK α– κ mediated NF- B activity and mRNA stabilization (28, 29, 45, activity. It is important to note that TRAF5 as a negative regu- 46). Furthermore, TRAF proteins, including TRAF5, are critical lator of TLR signaling and suppressor of IL-6 production in B components of TNF signaling and the precise role of TRAF5 in lymphocytes has been previously reported (53). Similar to our IL-17A signaling must be evaluated in the absence of TNF-α. observation in TRAF5KD cells, TRAF5 KO animals also show Therefore, we preferred cells (ARPE-19 and MEFs) that can increased NF-κB activation and pERK (53, 54). Additional work produce small yet measurable outcomes of IL-17A stimuli in the is needed to examine if TRAF5 is a regulator that diverts the IL- absence of a costimulus. Similarly, we utilized IL-17A alone for 17A–ACT1 signal in favor of the mRNA stabilization branch the ex vivo and the in vivo experiments, which resulted in a (55). In contrast to TRAF5, the presence of 14-3-3ζ, is required maximum of fourfold cytokine induction. However, we cannot for de novo transcription of IL6 but not for CXCL1. The

6of10 | www.pnas.org/cgi/doi/10.1073/pnas.2008214117 McGowan et al. Downloaded by guest on September 29, 2021 Fig. 6. The interaction with TRAF is required for 14-3-3ζ–mediated IL6 induction by IL-17A. (A) Two putative TRAF5 binding motifs (in red) of 14-3-3ζ and the mutants (M1 and M2) are shown. (B) The modeled structures of the TRAF-domain of TRAF5 (4GJH.pdb, pink color) complex with the dimeric 14-3-3ζ IMMUNOLOGY AND INFLAMMATION (5D2D.pdb, gray color) with restricted interaction to either of the two TBMs, SNEE or AYQE, are shown. The TBMs of 14-3-3ζ are shown in yellow color. (C) HEK293T cells, coexpressing V5-TRAF5 or Flag–14-3-3ζ, as indicated, were immunorecipitated with anti-Flag and elutes were analyzed by immunoblot. (D) HEK293T cells, coexpressing Flag-TRAF5 and HA–14-3-3ζ (WT or M2), were analyzed by PLA followed by confocal microscopy. (Scale bar, 5 μm.) (E) ARPE-19 cells, ectopically expressing 14-3-3ζ (WT, M1 or M2), were analyzed for IL-17A–induced IL6 by qRT-PCR. The expression of Flag-tagged 14-3-3ζ is shown by immunoblot. All results are representative of at least three experiments, ****P < 0.005.

opposite effects by 14-3-3ζ and TRAF5 on IL-17A–stimulated two critical regulators of IL-17A signaling that can alter IL-6 and IL-6 and CXCL-1 production suggest the presence of counter CXCL-1 levels. Future studies will be directed to investigate mechanisms to regulate the IL-17A–TRAF6 branch. These re- each of the possibilities and determine their contribution in sults show the presence of the 14-3-3ζ–TRAF5 axis as a regu- regulating IL-17A signaling. Whether a pathway-specific role of latory mechanism of IL-17A signaling. 14-3-3ζ exists to modulate the two TRAF branches is an in- The 14-3-3 proteins, particularly 14-3-3ζ, are known for their triguing point and will require 14-3-3ζ mutants that separate role in supporting cellular signaling and host defense (24, 27). these functions. Our structural modeling studies may help Our results show that 14-3-3ζ is important for IL-17A–stimulated identify such mutants in the future. In addition, several 14-3-3 Defb3, which is important for the antifungal activity of IL-17A. proteins are known to stabilize the phospho-proteins and are The in vivo antifungal effect of IL-17A depends on the pro- critical regulators of MAPK (62). Our result shows that 14-3-3ζ duction of a variety of AMP, neutrophil attracting chemokines, activity in IL-17A signal transduction is upstream of pERK; development of functional natural killer cells, and proin- therefore, it will be interesting to examine 14-3-3ζ’s role in the flammatory cytokines (56–58). Our straightforward in vitro assay regulating MAPK activity to promote IL-17A signaling. allowed us to evaluate the role of 14-3-3ζ in IL-17A–induced Overall, our results show 14-3-3ζ is an essential component of Defb3 production and fungicidal activity in the absence of neu- the IL-17A signaling pathway that promotes IL-6 production and trophils or other immune effector cells. antifungal activity via activating NF-κB. Importantly, we dis- Like IL-17A, IL-6 is also essential for autoimmune disease and covered, as a part of the molecular mechanism, 14-3-3ζ–TRAF5 host-defense (59–61). We show that interaction between the 14- is a regulatory axis in determining the functional outcome of the 3-3ζ and TRAF proteins is vital for IL-6 production. 14-3-3ζ can IL-17A–TRAF6 branch. Since this axis regulates IL-6 and interact with TRAF5, TRAF2, and TRAF6, and this interaction CXCL-1 production, its role in several immune diseases, in- plays a critical role in IL-17A–stimulated IL-6 production. Be- cluding rheumatoid arthritis, lupus, host-defense, and so forth, is cause mutation of the TRAF-binding motif can also disrupt 14-3- implied. Our goal is to achieve in-depth insight into IL-17A 3ζ binding to other TRAFs, an additional role of 14-3-3ζ in IL- signaling and uncover regulators that can help fine tune the 17A signaling cannot be ignored; specifically, we observed that outcomes of signal transduction. We expect that further insight ζ– TRAF6 interacts with 14-3-3ζ in tissue lysates and IL- of the 14-3-3 TRAF5 axis will make the basis of future thera- 17A–treated cells. Since TRAF5 seems to suppress IL-6 pro- peutics to promote desired gene outcomes of IL-17A signaling. duction that requires 14-3-3ζ, two possibilities may explain its Materials and Methods function: 1) IL-17A-induced increased interaction with TRAF5 is a mechanism by which 14-3-3ζ acts to sequester and inhibit Cell Culture and Reagents. The HEK-293T, HeLa, and MEFs were maintained in DMEM containing 10% FBS, penicillin, and streptomycin. The human ARPE- TRAF5 function, which results in the increased IL-6 production; ’ ζ 19 cells were maintained in DMEM-Ham s F-12 media containing 10% FBS, or 2) the 14-3-3 interaction with TRAF6 is required to promote penicillin, and streptomycin. All cell lines used in this study were maintained IL-6 production where TRAF5 sequesters 14-3-3ζ to act as a in the authors’ laboratory at the University of Toledo. Unless stated other- brake on IL-6 production. In either case, 14-3-3ζ and TRAF5 are wise, IL-17A stimulation in ARPE and MEFs was performed at 50 ng/mL of

McGowan et al. PNAS Latest Articles | 7of10 Downloaded by guest on September 29, 2021 Fig. 7. Functional significance of 14-3-3ζ participation in IL-17A signaling. (A) Culture supernatants from either untreated or IL-17A–treated WT or 14-3-3ζKO MEFs were analyzed for the killing of C. albicans.(B)WTor14-3-3ζKO MEFs, treated with mIL-17A, were analyzed for Defb3 by qRT-PCR. (C) The schematic design of 14-3-3ζ KO Lewis rat generation is shown; the exon 3 of rat 14-3-3ζ was targeted by the gRNAs, as indicated. A representative genotyping result is shown from − − − the tail DNA isolated from the indicated rat strains (+/+, +/ ,and / ). An immunoblot of 14-3-3ζ from the tail biopsy of the indicated strains is shown. (D and E)The duodenum of WT or heterozygous 14-3-3ζ Lewis rats were either untreated or treated with rat (r) IL-17A for 20 h, and the Il6 (D)andCxcl1 (E) mRNA levels were analyzed by qRT-PCR. (F and G) The WT and 14-3-3ζKO rats (n = 8) were injected with rIL-17A intraperitoneally for the indicated times, and cytokine (Il-6 and Cxcl-1) levels in the sera were analyzed by ELISA. The results are representative of at least three experiments, *P < 0.05, **P < 0.005, ***P < 0.0005, and ****P < 0.0001.

hIL-17 or 25 ng/mL of mIL-17, respectively, for 20 min for signaling, 60 min Cell Lysis and IP. Immunoblot analyses were performed using previously for p65 nuclear translocation, and 20 h for gene induction, unless specified described procedures (64, 65). Briefly, the cells were lysed in 50 mM Tris otherwise. The HeLa cells were pretreated with TNF-α for 30 min, followed buffer, pH 7.4 containing 150 mM of NaCl, 0.1% Triton X-100, 1 mM sodium by IL-17A for 20 min. The antibodies against the specific proteins were orthovanadate, 10 mM of sodium fluoride, 10 mM of β-glycerophosphate, obtained as indicated: Antiphospho-ERK ( #4370), anti-ERK 5 mM sodium pyrophosphate, protease and phosphatase inhibitors (Roche). μ μ (Cell Signaling #4695), anti-TRAF5 (Santa Cruz #74502, CST#41658), anti- For IP, either 2 g of antibody followed by Protein G Sepharose or 10 Lof tagged beads were added to the cell lysate and incubated overnight at 4 °C. TRAF6 (CST#8028), anti-TRAF2 (CST#4724), anti–14-3-3ζ (Santa Cruz #sc- Total protein extracts or pull-down beads were analyzed by SDS/PAGE, 1019, CST#7413), anti–β-tubulin (Abcam #ab15568), anti-GAPDH (Fitzgerald followed by immunoblot. 10R-G109A), anti-actin (Sigma-Aldrich #A5441), anti-HDAC (Santa Cruz #sc- 81598), anti-V5 (Thermo Fisher Scientific #R960-25). The IL-17A (human, rat, ELISA. ELISA kits for rat IL-6, rat CXCL-1, human IL-6 (R&D Systems) were used and murine) and TNF-α (human) were obtained from R&D Systems. All other as per the recommended protocols. Conditioned media from same number chemicals were purchased from Fisher Scientific. of cells or equal volume of animal sera was used for comparison between the WT and KO set of samples. Generation of CRISPR/Cas9-Mediated KO Cells. ARPE-19 cells were transfected ζ– with either control (sc-418922) or 14-3-3 specific (sc-400490) CRISPR/Cas9 In Vitro Protein–Protein Interaction Assay. HEK293T cells transiently express- plasmids (Santa Cruz) using Lipofectamine 2000 (Thermo Fisher Scientific). ing either HA-tagged 14-3-3ζ or Flag-tagged TRAF5 were lysed in EPPS buffer Transfected cells were sorted for high GFP-expressers using flow cytometry, containing protease inhibitors by five freeze-thaw cycles and lysates were and the GFP-expressing cells were expanded to isolate individual clones. immunoprecipitated separately with HA or Flag beads overnight at 4 °C. These clones were screened for 14-3-3ζ protein levels by immunoblot, and Beads were washed thrice with EPPS buffer containing 300 mM NaCl. The the clones with no 14-3-3ζ protein expression were expanded and further TRAF5 elute, from beads using Flag peptide, was incubated with washed HA validated to ensure the gene KO. beads bound to 14-3-3ζ or HA beads alone, for 2 h at room temperature. Beads were washed twice with EPPS buffer and once with RIPA buffer be- Lentiviral Transductions to Generate KD/Rescue Gene Expression. The lentivirus fore performing final elution for bound proteins by boiling in 2× SDS sample particles were generated using HEK-293T cells cotransfected with the plas- loading buffer. mid (TRAF5 or 14-3-3ζ shRNA from Millipore-Sigma, or lentiviral plasmid carrying 14-3-3ζ) and packaging plasmids, as approved by the University of RNA Isolation and qRT-PCR Analyses. Total RNA was isolated using TRIzol Toledo Institutional Review Board. The lentiviral particles with polybrene (Invitrogen), cDNA was prepared using ImProm-II Reverse Transcription Kit (Promega), and the cDNA was analyzed using Radiant SYBR Green PCR mix were used to transduce ARPE-19 or HeLa cells using a method as described (Alkali Scientific) in Roche LightCycler 96 instrument and analyzed with the previously (63). LightCycler 480 Software, v1.5. The expression levels of the mRNAs were normalized to 18S rRNA. For the qRT-PCR analyses of the respective genes, NF-κB Activity Measurement. The HeLa cells were transfected with 100 ng of the following primers were used: pNL3.2 NF-κB–RE, 10 ng of firefly luciferase tagged thymidine kinase, and different amounts of HA-tagged 14-3-3ζ, with Lipofectamine 2000. After IL6- GTAGCCGCCCCACACAGA/CATGTCTCCTTTCTCAGGGCTG α 24 h of transfection, cells were treated with 10 units of TNF- and 25 ng/mL CXCL1- AACCGAAGTCATAGCCACAC/GTTGGATTTGTCACTGTTCAGC, of hIL-17A and incubated for 20 h. Cells were lysed, and NF-κB activity was YWHAZ-ACCAGTATGTAGGCAGTTTTC/GGATCCATGGATAAAGAGCTGGTT, measured using the Nano-Glo Dual-Luciferase reporter system as per the manufacturer’s protocol (Promega). YWHAG- CCTGCATGAAGTCTGTAA/GAGCTACGGGCTACAACA

8of10 | www.pnas.org/cgi/doi/10.1073/pnas.2008214117 McGowan et al. Downloaded by guest on September 29, 2021 intermolecular energy constants of each simulated complex (67, 68). The simulated complexes with the least energy were visualized using VMD software. The TRAF-binding motifs are shown as a yellow highlight on the gray 14-3-3ζ dimer, while TRAF5 is shown in pink.

Antifungal Assay. C. albicans cultures were grown to an OD600 of ∼0.8 prior to adding conditioned media of MEFs or 14-3-3ζKO MEFs that were treated with IL-17A for 48 h. Aliquots of 45 μL of fungal cells were treated with 5-μL conditioned media for 1 h at 37 °C. After treatment, samples were serially diluted and plated for enumeration. The percent survival was determined by dividing the CFU of treated samples by those of untreated samples times 100. Assays were performed in triplicate and P values were calculated from all three experiments using ANOVA with Tukey’s post hoc test. MEF pellets were used for RNA analysis.

− − − − 14-3-3ζ / (Ywhaz / ) Lewis Rat Generation. KO rats were prepared as de- scribed previously (69). In brief, two single-guide RNAs (sgRNAs) were designed to have at least 3 bp of mismatch between the target site and any other site in the rat genome. Ribonucleoprotein (RNP) complexes were formed by incubating the 30 ng/mcL gRNA#1 (5′-CAGCAACCUCAGCCAAGU- AG-3′) or 30 ng/mcL gRNA#2 (5′-GCUUUCUGGCUGCGAAGCAU-3′) with 50 ng/mcL Cas9 protein (St. Jude Protein Production Core) at the room tem- perature for 10 min. The RNP were combined in a 1:1 ratio and placed on ice until rat zygotes microinjections at the Transgenic Animal Model Facility at the University of Michigan. Rat zygotes for microinjection were obtained by mating superovulated Lewis rat females with males of the same strain (LEW/ Fig. 8. A proposed model for the identified 14-3-3ζ–TRAF5 axis in regulat- Crl, strain code; Charles River Laboratory) and used for microinjection. as ing IL-17A signaling pathways. IL-17A binding triggers a receptor complex (IL- 17RA–ACT1–TRAF) activation that transduces the signal to generate transcrip- described previously (70). Animals were housed in American Association for tional output (IL-6/IL-8/CXCL-1). Our results show that 14-3-3ζ is a TRAF-binding Accreditation of Laboratory Animal Care-accredited animal care facility at IMMUNOLOGY AND INFLAMMATION component that is needed for IL-17A-stimulated pERK, NF-κB activation, and IL-6/ the University of Michigan or the University of Toledo under the Institu- IL-8 production, but not for CXCL-1. Importantly, we uncovered a role of TRAF5 tional Animal Care and Use Committee-approved protocols. One-hundred in suppressing the IL-17A–stimulated IL-6 production. The IL-17A stimulation sixty rat zygotes were microinjected; 349 survived injection and were increases the 14-3-3ζ–TRAF5 interaction that potentially inhibits the TRAF5 transferred to pseudopregnant Sprague-Dawley females [Crl:CD(SD), strain function and supports the IL-6 production. Overall, our model suggests that IL- code 001; Charles River Laboratory]. A total of 20 G0 rat pups were obtained, ζ 17A stimulation increases 14-3-3 interaction to sequester TRAF5, which results in of which 9 were found to be homozygous null, and 11 were heterozygous ’ increased IL-6 levels. These results have broad implications covering IL-17A srole for the Ywhaz gene. All heterozygous G0 pups were mated with WT Lewis in antifungal immunity to regulation of inflammation. rats (Charles River Laboratory) to obtain germ-line transmission of gene KO. Ear or tailpiece was used for DNA extraction and genotype of each gener- il6- ACACATGTTCTCTGGGAAATCGT/AAGTGCATCATCGTTGTTCATACA, ation of animals was determined by PCR. Three sets of YWHAZ primers (TCT- TTTCTTCTGAAGTCTCTTTTGG, ATATAAGCTCAACCAAGCAAAACC, CTACAC- Cxcl1- TGAGCTGCGCTGTCAGTGCCT/AGAAGCCAGCGTTCACCAGA. TTTTGAGACAGCATTGG, GATCACCATCTTGTGATTTCCTC, CTCTCCAATGCA- Defb3 was purchased from Qiagen (PPM25409B). GTTACATCATC, CACACTAGTCACTCGCTGTTGTC) were used to identify the 170-, 549-, and 1,026-bp product for the WT gene. The heterozygous animals Confocal Microscopy. ARPE-19 cells were grown on coverslips and treated were maintained for breeding and to obtain KO animals. The generation #5 with IL-17A for 1 h. The treated cells were fixed in 4% paraformaldehyde animals were utilized for this study. (Electron Microscopy Sciences #15710), permeabilized in 0.2% Triton X-100 All animals were housed in a standard animal care facility with controlled (Fisher Scientific #9002-93-1), and immunostained with primary antibody temperature and unlimited access to food and water. Age- and sex-matched followed by Alexa Fluor-conjugated secondary antibody (Invitrogen #A- rats were used in the study. For in vivo experiments, animals were injected 11004). The coverslips were mounted onto microscopy slides using Vecta- intraperitoneally with 500 ng of rat IL-17A, and blood was collected via Shield/DAPI (Vector Laboratories #H-1200) and analyzed using an Olympus saphenous vein at 0, 4, 12, and 72 h. For the ex vivo experiments, 12-wk-old confocal microscope and Olympus Fluoview FV1000 software. WT and heterozygous male animals were killed to collect duodenum under aseptic environment. Ingest from duodenum was removed and cleaned with Proximity Ligation Assay. For PLA, 293T cells were transfected with Flag- sterile PBS containing antibiotics (71). Tissue was cut into 5- to 8-mm size and TRAF5, HA–14-3-3ζ constructs, and after 24 h, posttransfection cells were fixed with 4% paraformaldehyde (Electron Microscopy Sciences #15710). incubated with 10% FBS and antibiotic-containing DMEM for few hours Afterward, cells were permeabilized with 0.2% Triton X-100 (Fisher Scientific before adding rat IL-17A at 50 ng/mL followed by incubation for 20 h. #9002-93-1) for 20 min and blocked with 2.5% BSA for 1 h. Immunostaining was done with anti-HA (CST#3724S, 1/250 dilution) and anti-Flag (Sigma# Statistical Analysis. All experiments were performed at least thrice unless F1804, 1/50 dilution) antibodies overnight at 4 °C and cells were proceeded for stated otherwise. Depending upon the number of sets for comparison, either incubation with PLA probes (Sigma#DUO92008-3, DUO92004, DUO92002) and unpaired Student’s t test or ordinary one-way ANOVA was used. subsequent steps as per the manufacturer’s instructions. The coverslips were mounted onto microscopy slides using VectaShield/DAPI (Vector Laboratories Data Availability. All study data are included in the article and SI Appendix. #H-1200) and analyzed using an Olympus confocal microscope and Olympus Fluoview FV1000 software. ACKNOWLEDGMENTS. We thank Dr. Lopez for sharing the 14-3-3ζKO mouse embryonic fibroblasts; Dr. Bina Joe, Dr. Cheng Xi, Dr. Thom Saunders, and Modeling of TRAF-14-3-3ζ Binding. A three-dimensional structural model of Ms. Blair Bell, who helped us in generating the animal model; and several pre- 14-3-3ζ-TRAF5 complex was generated using the following PDB structure sent and former laboratory members in shaping the project, and in particular files 5D2D and 4GJH for 14-3-3ζ and TRAF5, respectively. For docking, we Ms. Anna Glanz, who provided technical assistance. We acknowledge funding used the ZDOCK server (zdock.umassmed.edu/) (66). We restricted the TRAF5 from American Heart Association Grant 15SDG25008025 (to R.C.), NIH/NIDCR binding to the TRAF motifs, SNEE or AYQE, on the 14-3-3ζ for docking. The R15DE026898-01 (to H.C.), 1R01DE027343-01A1 (to H.C.), and the University of top 10 simulated complexes were processed with PRODIGY to determine the Toledo College of Medicine and Life Sciences startup funds (R.C.).

McGowan et al. PNAS Latest Articles | 9of10 Downloaded by guest on September 29, 2021 1. N. Amatya, A. V. Garg, S. L. Gaffen, IL-17 signaling: The yin and the yang. Trends 35. B. Benhuri, A. ELJack, B. Kahaleh, R. Chakravarti, Mechanism and biomarkers in Immunol. 38, 310–322 (2017). aortitis—A review. J. Mol. Med. (Berl.) 98,11–23 (2020). 2. H. Ishigame et al., Differential roles of interleukin-17A and -17F in host defense 36. Y. Qian et al., The adaptor Act1 is required for -dependent signaling as- against mucoepithelial bacterial and allergic responses. Immunity 30, sociated with autoimmune and inflammatory disease. Nat. Immunol. 8, 247 –256 (2007). 108–119 (2009). 37. Y. Chen, A. Kijlstra, Y. Chen, P. Yang, IL-17A stimulates the production of inflam- 3. C. Leonardi et al., Anti-interleukin-17 monoclonal antibody in chronic matory mediators via Erk1/2, p38 MAPK, PI3K/Akt, and NF-κB pathways in ARPE-19 plaque psoriasis. N. Engl. J. Med. 366, 1190–1199 (2012). cells. Mol. Vis. 17, 3072–3077 (2011). 4. L. G. van Baarsen et al., Heterogeneous expression pattern of interleukin 17A (IL-17A), 38. G. E. Lim et al., 14-3-3ζ coordinates adipogenesis of visceral fat. Nat. Commun. 6, 7671 (2015). IL-17F and their receptors in synovium of rheumatoid arthritis, psoriatic arthritis and 39. P. Rodrigues et al., NF-κB-Dependent lymphoid enhancer co-option promotes renal osteoarthritis: Possible explanation for nonresponse to anti-IL-17 therapy? Arthritis carcinoma metastasis. Discov. 8,850–865 (2018). Res. Ther. 16, 426 (2014). 40. S. Swaidani, C. Liu, J. Zhao, K. Bulek, X. Li, TRAF regulation of IL-17 cytokine signaling. 5. L. Pacheco-Lugo et al., Plasma cytokines as potential biomarkers of kidney damage in Front. Immunol. 10, 1293 (2019). patients with systemic lupus erythematosus. Lupus 28,34–43 (2019). 41. R. Schwandner, K. Yamaguchi, Z. Cao, Requirement of tumor factor receptor- 6. G. E. Fragoulis, S. Siebert, I. B. McInnes, Therapeutic targeting of IL-17 and IL-23 cy- associated factor (TRAF)6 in interleukin 17 signal transduction. J. Exp. Med. 191, tokines in immune-mediated diseases. Annu. Rev. Med. 67, 337–353 (2016). 1233–1240 (2000). 7. W. Hueber et al.;SecukinumabinCrohn’s Disease Study Group, , a human 42. F. Qu et al., TRAF6-dependent Act1 phosphorylation by the IκB kinase-related kinases anti-IL-17A monoclonal antibody, for moderate to severe crohn’s disease: Unexpected suppresses interleukin-17-induced NF-κB activation. Mol. Cell. Biol. 32,3925–3937 (2012). results of a randomised, double-blind placebo-controlled trial. Gut 61,1693–1700 (2012). 43. M. Schneider et al., The innate immune sensor NLRC3 attenuates Toll-like receptor 8. M. C. Genovese et al., Efficacy and safety of secukinumab in patients with rheumatoid signaling via modification of the signaling adaptor TRAF6 and NF- arthritis: A phase II, dose-finding, double-blind, randomised, placebo controlled κB. Nat. Immunol. 13, 823–831 (2012). study. Ann. Rheum. Dis. 72, 863–869 (2013). 44. H. R. Conti et al., IL-17 receptor signaling in oral epithelial cells is critical for protection 9. K. M. Robinson, M. L. Manni, P. S. Biswas, J. F. Alcorn, Clinical consequences of tar- against oropharyngeal candidiasis. Cell Host Microbe 20, 606–617 (2016). geting IL-17 and TH17 in autoimmune and allergic disorders. Curr. Allergy Asthma 45. C. Aguilera et al., Efficient nuclear export of p65-IkappaBalpha complexes requires Rep. 13, 587–595 (2013). 14-3-3 proteins. J. Cell Sci. 119, 3695–3704 (2006). 10. D. M. Saunte, U. Mrowietz, L. Puig, C. Zachariae, Candida in patients with 46. M. Lu et al., The attenuation of 14-3-3ζ is involved in the caffeic acid-blocked psoriasis and psoriatic arthritis treated with interleukin-17 inhibitors and their prac- lipopolysaccharide-stimulated inflammatory response in RAW264.7 . tical management. Br. J. Dermatol. 177,47–62 (2017). Inflammation 40, 1753–1760 (2017). 11. R. M. Onishi, S. L. Gaffen, Interleukin-17 and its target genes: Mechanisms of 47. S. D. Levin, IL-17 receptor signaling: Ubiquitin gets in on the act. Sci. Signal. 2, pe64 (2009). interleukin-17 function in disease. Immunology 129, 311–321 (2010). 48. J. A. Cruz et al., Interleukin-17 signaling triggers degradation of the constitutive NF- 12. H. R. Conti, S. L. Gaffen, IL-17-Mediated immunity to the opportunistic fungal path- κB inhibitor ABIN-1. Immunohorizons 1, 133–141 (2017). ogen Candida albicans. J. Immunol. 195, 780–788 (2015). 49. F. Huang et al., Requirement for both JAK-mediated PI3K signaling and ACT1/TRAF6/ 13. C. Gu, L. Wu, X. Li, IL-17 family: Cytokines, receptors and signaling. Cytokine 64, TAK1-dependent NF-kappaB activation by IL-17A in enhancing cytokine expression in 477–485 (2013). human airway epithelial cells. J. Immunol. 179, 6504–6513 (2007). 14. D. Toy et al., Cutting edge: Interleukin 17 signals through a heteromeric receptor 50. L. Monin et al., MCPIP1/Regnase-1 restricts IL-17A- and IL-17C-dependent skin in- complex. J. Immunol. 177,36–39 (2006). flammation. J. Immunol. 198, 767–775 (2017). 15. Z. Yao et al., Herpesvirus Saimiri encodes a new cytokine, IL-17, which binds to a novel 51. N. Amatya et al., IL-17 integrates multiple self-reinforcing, feed-forward mechanisms . Immunity 3, 811–821 (1995). through the RNA binding protein Arid5a. Sci. Signal. 11, eaat4617 (2018). 16. D. Ge, Z. You, Expression of interleukin-17RC protein in normal human tissues. Int. 52. J. Hartupee et al., IL-17 signaling for mRNA stabilization does not require TNF Arch. Med. 1, 19 (2008). receptor-associated factor 6. J. Immunol. 182, 1660–1666 (2009). 17. X. Li, R. Bechara, J. Zhao, M. J. McGeachy, S. L. Gaffen, IL-17 receptor-based signaling 53. C. M. Buchta, G. A. Bishop, TRAF5 negatively regulates TLR signaling in B lymphocytes. and implications for disease. Nat. Immunol. 20, 1594–1602 (2019). J. Immunol. 192, 145–150 (2014). 18. X. Song, Y. Qian, The activation and regulation of IL-17 receptor mediated signaling. 54. J. Shang et al., Disruption of receptor-associated factor 5 ex- Cytokine 62, 175–182 (2013). acerbates murine experimental colitis via regulating T helper cell-mediated inflam- 19. D. Sun et al., Treatment with IL-17 prolongs the half-life of CXCL1 mRNA mation. Mediators Inflamm. 2016, 9453745 (2016). via the adaptor TRAF5 and the splicing-regulatory factor SF2 (ASF). Nat. Immunol. 12, 55. T. Herjan et al., IL-17-receptor-associated adaptor Act1 directly stabilizes mRNAs to 853–860 (2011). mediate IL-17 inflammatory signaling. Nat. Immunol. 19, 354–365 (2018). 20. G. Huang, Y. Wang, P. Vogel, H. Chi, Control of IL-17 receptor signaling and tissue 56. B. G. Mengesha, H. R. Conti, The role of IL-17 in protection against mucosal Candida inflammation by the p38α-MKP-1 signaling axis in a mouse model of multiple scle- infections. J. Fungi (Basel) 3, 52 (2017). rosis. Sci. Signal. 8, ra24 (2015). 57. K. Trautwein-Weidner, A. Gladiator, S. Nur, P. Diethelm, S. LeibundGut-Landmann, IL- 21. Y. Xiao et al., TPL2 mediates autoimmune inflammation through activation of the 17-mediated antifungal defense in the oral mucosa is independent of neutrophils. TAK1 axis of IL-17 signaling. J. Exp. Med. 211, 1689–1702 (2014). Mucosal Immunol. 8, 221–231 (2015). 22. T. Öhman et al., Phosphoproteomics combined with quantitative 14-3-3-affinity 58. E. Bär, P. G. Whitney, K. Moor, C. Reis e Sousa, S. LeibundGut-Landmann, IL-17 reg- capture identifies SIRT1 and RAI as novel regulators of cytosolic double-stranded RNA ulates systemic fungal immunity by controlling the functional competence of NK cells. recognition pathway. Mol. Cell. Proteomics 13, 2604–2617 (2014). Immunity 40, 117–127 (2014). 23. K. L. Pennington, T. Y. Chan, M. P. Torres, J. L. Andersen, The dynamic and stress- 59. J. Hoge et al., IL-6 controls the innate immune response against Listeria mono- adaptive signaling hub of 14-3-3: Emerging mechanisms of regulation and context- cytogenes via classical IL-6 signaling. J. Immunol. 190, 703–711 (2013). dependent protein-protein interactions. Oncogene 37, 5587–5604 (2018). 60. K. Van Der Geest et al., Increased IL-6 production by effector B cells in giant cell ar- 24. D. Thomas, M. Guthridge, J. Woodcock, A. Lopez, 14-3-3 protein signaling in devel- teritis and polymyalgia rheumatica. Ann. Rheum. Dis. 66, 73 (2014). opment and responses. Curr. Top. Dev. Biol. 67, 285–303 (2005). 61. S. A. Jones, B. J. Jenkins, Recent insights into targeting the IL-6 cytokine family in 25. X. Han, Y. Han, H. Jiao, Y. Jie, 14-3-3ζ regulates immune response through Stat3 inflammatory diseases and cancer. Nat. Rev. Immunol. 18, 773–789 (2018). signaling in oral squamous cell carcinoma. Mol. Cells 38, 112–121 (2015). 62. M. Gaestel, MAPKAP kinases—MKs—Two’s company, three’s a crowd. Nat. Rev. Mol. 26. K. Funami, M. Matsumoto, C. Obuse, T. Seya, 14-3-3-zeta participates in TLR3- Cell Biol. 7, 120–130 (2006). mediated TICAM-1 signal-platform formation. Mol. Immunol. 73,60–68 (2016). 63. G. Subramanian et al., A new mechanism of ’s antiviral action: Induction of 27. F. C. Stomski et al., Identification of a 14-3-3 binding sequence in the common beta chain of , essential for paramyxovirus replication, is inhibited by the interferon the - colony-stimulating factor (GM-CSF), interleukin-3 (IL-3), and stimulated gene, TDRD7. PLoS Pathog. 14, e1006877 (2018). IL-5 receptors that is serine-phosphorylated by GM-CSF. Blood 94, 1933–1942 (1999). 64. S. Chattopadhyay et al., Viral is induced by IRF-3-mediated activation of 28. S. Zuo et al., 14-3-3 epsilon dynamically interacts with key components of mitogen- Bax. EMBO J. 29, 1762–1773 (2010). activated protein kinase signal module for selective modulation of the 65. S. Chattopadhyay, T. Kuzmanovic, Y. Zhang, J. L. Wetzel, G. C. Sen, Ubiquitination of TNF-alpha-induced time course-dependent NF-kappaB activity. J. Proteome Res. 9, the transcription factor IRF-3 activates RIPA, the apoptotic pathway that protects 3465–3478 (2010). mice from viral pathogenesis. Immunity 44, 1151–1161 (2016). 29. L. Sun et al., Tristetraprolin (TTP)-14-3-3 complex formation protects TTP from de- 66. B. G. Pierce et al., ZDOCK server: Interactive docking prediction of protein-protein phosphorylation by protein phosphatase 2a and stabilizes tumor necrosis factor-alpha complexes and symmetric multimers. Bioinformatics 30, 1771–1773 (2014). mRNA. J. Biol. Chem. 282, 3766–3777 (2007). 67. A. Vangone, A. M. Bonvin, Contacts-based prediction of binding affinity in protein- 30. S. Cascio et al., 14-3-3z sequesters cytosolic T-bet, upregulating IL-13 levels in TC2 and protein complexes. eLife 4, e07454 (2015). CD8(+) lymphocytes from patients with scleroderma. J. Allergy Clin. Immunol. 142, 68. L. C. Xue, J. P. Rodrigues, P. L. Kastritis, A. M. Bonvin, A. Vangone, PRODIGY: A web 109–19.e6 (2018). server for predicting the binding affinity of protein-protein complexes. Bio- 31. J. McGowan, C. Peter, S. Chattopadhyay, R. Chakravarti, 14-3-3zeta: A novel immunogen informatics 32, 3676–3678 (2016). promotes production. Frontiers Immunol. 10, 1553 (2019). 69. S. W. Watts et al., The chemerin knockout rat reveals chemerin dependence in fe- 32. R. Chakravarti et al., 14-3-3 in thoracic aortic aneurysms: Identification of a novel male, but not male, experimental hypertension. FASEB J. 32, fj201800479 (2018). autoantigen in large vessel vasculitis. Arthritis Rheumatol. 67, 1913–1921 (2015). 70. W. E. Filipiak, T. L. Saunders, Advances in transgenic rat production. Transgenic Res. 33. J. Deng, B. R. Younge, R. A. Olshen, J. J. Goronzy, C. M. Weyand, Th17 and Th1 T-cell 15, 673–686 (2006). responses in giant cell arteritis. Circulation 121, 906–915 (2010). 71. B. A Rasmussen et al., Duodenal activation of cAMP-dependent protein kinase in- 34. D. Saadoun et al., Th1 and Th17 cytokines drive inflammation in Takayasu arteritis. duces vagal afferent firing and lowers production in rats. Gastroenterology Arthritis Rheumatol. 67, 1353–1360 (2015). 142, 834–43.e3 (2012).

10 of 10 | www.pnas.org/cgi/doi/10.1073/pnas.2008214117 McGowan et al. Downloaded by guest on September 29, 2021