FTRCA1, a Species-Specific Member of finTRIM Family, Negatively Regulates Fish IFN Response through Autophage-Lysosomal Degradation of TBK1 This information is current as of September 29, 2021. Min Wu, Xiang Zhao, Xiu-Ying Gong, Yang Wang, Jian-Fang Gui and Yi-Bing Zhang J Immunol published online 8 March 2019 http://www.jimmunol.org/content/early/2019/03/05/jimmun ol.1801645 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2019/03/05/jimmunol.180164

Material 5.DCSupplemental http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 29, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2019 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published March 8, 2019, doi:10.4049/jimmunol.1801645 The Journal of Immunology

FTRCA1, a Species-Specific Member of finTRIM Family, Negatively Regulates Fish IFN Response through Autophage-Lysosomal Degradation of TBK1

Min Wu,*,† Xiang Zhao,*,† Xiu-Ying Gong,*,† Yang Wang,*,† Jian-Fang Gui,*,†,‡ and Yi-Bing Zhang*,†,‡,x

In mammals, tripartite motif (TRIM) proteins have emerged as pivotal players endowed with, directly, antiviral effects and, in- directly, modulatory capacity of the innate immune response. An unprecedented expansion of TRIM family has occurred in fish; however, the functional role of fish TRIM family members remains largely unknown. In this study, we identify a species-specific TRIM gene from crucian carp Carassius auratus, named FTRCA1, phylogenetically similar to the members of finTRIM,

a subfamily of TRIM exclusively in teleost fish. FTRCA1 is induced by IFN and IFN stimuli as a typical IFN-stimulated gene. Downloaded from Overexpression of FTRCA1 negatively regulates IFN antiviral response by inhibition of IRF3 phosphorylation; consistently, knockdown of FTRCA1 results in enhanced levels of IRF3 phosphorylation and also IFN expression following poly(I:C) trans- fection. Whereas FTRCA1 is associated with several pivotal signaling molecules of RIG-I–like receptor pathway, its association with TBK1 results in autophage-lysosomal degradation of TBK1, thus abrogating the downstream IFN induction. Interestingly, FTRCA1 is phosphorylated by TBK1, but this phosphorylation is not required for downregulation of TBK1 protein. Transfection

assays indicate that FTRCA1 is likely an E3 ligase with the requirement of RING finger domain, and deletion of N-terminal RING http://www.jimmunol.org/ domain or mutation of seven conservative sites abolishes the negative regulatory function of FTRCA1. Collectively, these results illuminate a novel finTRIM-mediated innate immune modulatory pathway, thus providing insights into species-specific regulation of fish IFN response. The Journal of Immunology, 2019, 202: 000–000.

n mammals, virus infection rapidly induces production of pattern recognition receptors. Retinoic acid–inducible gene type I IFNs, such as IFN-a/b, upon the recognition of virus- I (RIG-I)–like receptors (RLRs) and several TLR have been well specific pathogen-associated molecular patterns by host characterized as sensors of viral-derived nucleic acids (1). Gen-

I by guest on September 29, 2021 erally, they recognize intruding viral components, present at dif- ferent cellular compartments in immune cell lineages and other *State Key Laboratory of Freshwater Ecology and Biotechnology, Institute of cell types, to trigger distinct signaling pathways, but finally con- Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, China; †Department of verge on the activation of cytosolic protein kinases, TANK-binding Aquaculture, University of Chinese Academy of Sciences, Wuhan 430072, China; ‡The Innovation Academy of Seed Design, Chinese Academy of Sciences, Wuhan kinase 1 (TBK1), which enables IFN regulatory factor (IRF) 3/7 to 430072, China; and xKey Laboratory of Aquaculture Disease Control of Ministry of translocate into the nucleus, turning on the transcription of IFN-a/b. Agriculture, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, The secreted IFNs in turn induce the expression of IFN-stimulated China genes (ISGs) through JAK-STAT pathway (1). Whilst IFNs are ORCID: 0000-0001-7647-9335 (Y.-B.Z.). curial for the establishment of host antiviral state, overproduction of Received for publication December 19, 2018. Accepted for publication February 14, IFNs leads to the development of immunopathological conditions; 2019. therefore, host cells develop multiple mechanisms to precisely This work was supported by a grant from the National Key R&D Program of China (2018YFD0900302), grants from the National Natural Science Foundation modulate the IFN signaling for appropriate production of IFNs (2). (31572646 and 31772875), and a grant from the Freshwater Ecology and Biotech- Emerging evidence highlights the pivotal roles of tripartite motif nology Laboratory (2016FBZ01). (TRIM) proteins in the innate antiviral response (2–4). TRIM The sequence presented in this article has been submitted to GenBank under acces- proteins are defined by a consecutive TRIM from the N to C sion number MK239647. terminus, a really interesting new gene (RING) zinc finger do- Address correspondence and reprint requests to Dr. Yi-Bing Zhang, State Key Lab- oratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology, main, one or two B-boxes, and a coiled-coil domain, therefore also Chinese Academy of Sciences, Wuhan 430072, China. E-mail address: ybzhang@ihb. known as RBCC proteins (5). The RING domain has E3 ligase ac.cn activity responsible for the interaction with the ubiquitin- The online version of this article contains supplemental material. conjugating enzymes (E2) in the ubiquitination cascade process, Abbreviations used in this article: CAB, Crucian carp (C. auratus L.) blastula em- and the coiled-coil domain contributes to homo-interaction or bryonic; Co-IP, coimmunoprecipitation; EPC, epithelioma papulosum cyprini; oligomerization into high-order complexes (6). Mechanistically, a FTR, fish novel TRIM; GCRV, grass carp reovirus; IRF, IFN regulatory factor; ISG, IFN-stimulated gene; MAVS, mitochondrial antiviral signaling protein; substantial number of TRIM proteins limit viral replication by MDA5, melanoma-differentiation–associated gene 5; MITA, mediator of IRF3 directly targeting viral proteins. For example, TRIM22 disrupts activation; NC, nontargeting control siRNA; ORF, open reading frame; poly(I:C), poly- inosinic:polycytidylic acid; RIG-I, retinoic acid–inducible gene I; RING, really the proper trafficking of HIV protein Gag (7); TRIM69 mediates interesting new gene; RLR, RIG-I–like receptor; RT-qPCR, quantitative real-time DENV nonstructural protein 3 polyubiquitination for proteasomal PCR; siRNA, small interfering RNA; TBK1, TANK-binding kinase 1; TRIM, degradation (8). A subset of TRIMs, including TRIM25 (9) tripartite motif. and TRIM26 (10, 11), modulate the innate antiviral signaling Copyright Ó 2019 by The American Association of Immunologists, Inc. 0022-1767/19/$37.50 to promote or abrogate the production of IFN, respectively.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1801645 2 SPECIES-SPECIFIC REGULATION OF FISH IFN IMMUNE RESPONSE

Interestingly, TRIM5a serves as not only a potent retroviral re- UV-inactivated GCRV-infected CAB cells (32), a pair of primers was striction factor against HIV but also a potential modulator of designed for RACE PCR to clone the full-length cDNA of crucian carp antiviral innate immunity (12), indicating the function complexity FTRCA1 (GenBank accession no. MK239647, https://www.ncbi.nlm. nih.gov/nuccore/MK239647). Primers used for RACE PCR are listed of TRIM proteins. in Supplemental Table I. Using the FTRCA1 sequence as baits, A genome-wide survey reveals a rapid expansion of TRIM protein BLAST search was performed against the genome database family in vertebrates (13–16), which seems linked to the in- from National Center for Biotechnology Information, including human, creasingly complex immune system (17). In line with the notion, Homo sapien;zebrafish,Danio rerio;fugu,Takifugu rubripes; puffer fish, Tetraodon nigroviridis;medaka,Oryzias latipes;grasscarp, 75 TRIM family members have been identified in human, whereas Ctenopharyngodon idella; common carp, Cyprinus carpio; and barbel, the fruitfly has only 7 and the worm 18 (13, 15). Functionally, Sinocyclocheilus grahami. Because there is relatively perfect integrity of more than one third of human TRIM genes are transcriptionally genome databases and comprehensive information of TRIM family in upregulated by IFNs (18), and roughly half contribute to regu- human and zebrafish (15, 16), the searched human and zebrafish ho- lating innate immune response (4). Similar expansion happens to mologs most homologous to FTRCA1 were further used as baits for protein BLAST against other genome data to obtain many more se- teleost fish, with over 100–120 members identified in most species quences for phylogenetic analysis. The top hits from these queries were (13–15). The unprecedented and peculiar expansion of TRIM also used to reciprocally query an unpublished draft genome of gibel family in fish is highlighted by generation of a novel fish-specific carp (C. auratus gibello), a species belonging to a same genus Carassius TRIM subfamily, called finTRIM (fish novel TRIM [FTR]) (14). as crucian carp (C. auratus). Multiple alignments were performed with ClustalW2 and used to derive a phylogenetic tree by neighbor-joining FinTRIM family members are formed of N-terminal RBCC and methods in Geneious. length-variable C termini, including B30.2 domain (also known as

“PRY-SPRY” domain), which is found in many TRIMs with an- Plasmids Downloaded from tiviral function (7, 9, 19, 20). It is believed that selective pressures Expression plasmids Flag-FTRCA1 and Flag-FTRCA1-DRING were against virus infection make finTRIM family evolve fast, resulting generated by cloning the cDNA sequences (corresponding to aa 1–542 and in greatly varied gene number among fish species (14, 15, 21). For 60–542, respectively) with a stop codon into EcoRI and KpnI sites of 2 example, of the 84 FTR genes in zebrafish, most have no “one-to- pCMV-Flag, which is made by modifying pcDNA3.1/myc-His( ) with insertion of a Flag tag–containing sequence 59-GCTAGCCACCATG- one” orthologs in other fish species apart from zebrafish FTR82, GACTATAAGGACGATGATGACAAGAATTC-39 into NheI and EcoRI

FTR83, and FTR84, which seem to be the most ancient ones and sites. Expression plasmids, RIG-I–myc and melanoma-differentiation– http://www.jimmunol.org/ have real orthologs in the main fish branches (14). associated gene 5 (MDA5)–myc, were made by insertion of the open In comparison with the wealth of information about TRIM- reading frames (ORF) of crucian carp RIG-I (GenBank accession no. JF970225, https://www.ncbi.nlm.nih.gov/nuccore/JF970225) and MDA5 mediated regulation of antiviral responses (2), the knowledge (GenBank accession no. JF970226, https://www.ncbi.nlm.nih.gov/nuccore/ about the functional role of fish TRIM family members is still JF970226) into BamHI and KpnI sites of pcDNA3.1/myc-His(2). EGFP- rudimentary. Fish possess conserved IFN antiviral response trig- myc were generated by using BamHI and HindIII sites of pcDNA3.1/ gered by TLR pathway and RLR pathway (22, 23). Although myc-His(2). HA-TBK1 was made by cloning of the ORF of crucian carp some fish TRIM genes appear to participate in fish antiviral re- TBK1 (GenBank accession no. JF970228, https://www.ncbi.nlm.nih.gov/ nuccore/JF970228) into XbaI and BamHI of pGN-HAM. For subcellular sponse (24–28), the molecular mechanisms involved remain un- localization assays, the ORF of TBK1 was subcloned into pDsRED1-C1 explored. A recent study shows that zebrafish FTR83, a relatively at XhoI and KpnI sites (pDsRED2-TBK1, encoding a fusion protein by guest on September 29, 2021 conserved member of finTRIM family, restricts viral infection DsRED2-TBK1) and the ORF of FTRCA1 into HindIII and KpnI sites of by triggering IFN response (21). In the current study, we identified pEGFP-N3 (pEGFP-FTRCA1, encoding a fusion protein FTRCA1- EGFP). Crucian carp IFN reporter plasmid IFNpro-luc was character- a species-specific finTRIM gene from crucian carp Carassius ized previously (29). Other plasmids, including crucian carp IFN, mediator auratus (according to the Chinese nomenclature) named FTRCA1 of IRF3 activation (MITA), TBK1-myc, IRF3-myc, IRF7-myc, zebrafish and determined its functional role in fish IFN response. Unlike mitochondrial antiviral signaling protein (MAVS), zebrafish IFN-w1pro-luc, FTR83, which exists in all fish species and is not induced by IFN zebrafish IFN-w3pro-luc, and some dominant negative mutant plasmids in- (14, 21), FTRCA1 is found in crucian carp but not in other fish cluding crucian carp TBK1-K38M, IRF3-DN, IRF7-DN, and STAT1-DC were described previously (29, 31, 33). species, even the ones belonging to the same genus Carassius with crucian carp, and meanwhile, it is induced by virus infection as a RNA extraction, cDNA synthesis, and quantitative typical ISG. Functionally, FTRCA1 negatively regulates fish IFN real-time PCR antiviral response by degradation of TBK1 through lysosomal- Total RNA extraction, cDNA sysnthesis, and quantitative real-time PCR dependent pathway. Our results reveal a fish species-specific (RT-qPCR) were performed as described previously (34). For RT-qPCR, all TRIM gene-mediated innate immune modulatory pathway, thus samples were analyzed in triplicate and the expression values, unless in- providing insights into species-specific regulation of fish IFN dicated, were normalized to b-actin. Primers used for RT-PCR analysis are listed in Supplemental Table I. response. Transfection and luciferase activity assays Materials and Methods Cells and virus Transfection was performed by FuGENE HD Transfection Reagent (Promega) according to the manufacturer’s protocol or our previous reports Crucian carp (C. auratus L.) blastula embryonic (CAB) cells and epithe- (29, 30, 34). Luciferase activity assays were performed by a Junior LB lioma papulosum cyprini (EPC) cells were cultured as described previ- 9509 luminometer (Berthold, Pforzheim, Germany) and normalized to ously (29, 30). HEK293T cells were originally obtained from American the amounts of Renilla luciferase activities as previously described Type Culture Collection and were cultured in DMEM (HyClone Labora- (29, 30, 33, 34). tories) with 10% FBS and grown at 37˚C in a humidified incubator containing 5% CO2. Grass carp reovirus (GCRV) was propagated in Small interfering RNA–mediated knockdown CAB cells and titered according to the method of Reed and Muench, by a Two FTRCA1-targeted small interfering RNAs (siRNAs; siRNA1 and tissue culture ID50 assay. Recombinant crucian carp IFN (GenBank acces- siRNA2) and a nontargeting control siRNA (NC) were designed by and sion no. AY452069, https://www.ncbi.nlm.nih.gov/nuccore/38455769) pro- purchased from RiboBio (China). The targeted sequences are 59-GGA- tein (rIFN) was generated by a prokaryotic expression system (31). CAGAGAAACAGAAACA-39 (siRNA 1) and 59-ACTCATTCTGGCA- Gene cloning and sequence analysis CAATAA-39 (siRNA 2), respectively. Because FTRCA1 is specific to crucian carp, knockdown of FTRCA1 was performed in CAB cells. Briefly, Based on an expression sequence tag retrieved from a suppressed sub- CAB cells seeded in 3.5-cm2 dishes overnight were transfected with 100 nM tractive cDNA library that was made with mRNAs derived from FTRCA1-specific siRNAs or NCs using Lipofectamine RNAiMAX (Life The Journal of Immunology 3

Technologies) according to the manufacturer’s instructions. Twenty-four investigated and have a regular nomenclature (15, 35), most of the hours later, the transfected cells were transfected again with 1 mg/ml searched homologs of FTRCA1 from other fish species are labeled polyinosinic:polycytidylic acid [poly(I:C); Sigma-Aldrich] for an addi- “TRIM16-like” (Supplemental Fig. 2), indicating that it is hard to tional 24 h by FuGENE HD Transfection Reagent (Promega). Knockdown efficiency of endogenous FTRCA1 was determined by RT-qPCR. define a given gene of finTRIM, likely due to the occurrence of genus- or species-specific expansion of finTRIM family after ra- Coimmunoprecipitation, Western blotting, and diation of teleosts (14, 15). Considering a relatively perfect in- ubiquitination assays tegrity of zebrafish genome, a simplified phylogenetic tree was Coimmunoprecipitation (Co-IP) assays and Western blotting were made to further determine whether there is a one-to-one ortholog performed as previously described (34). For ubiquitination assays, of FTRCA1 in zebrafish, by zebrafish sequences that are most 2 293T cells cultured in 10-cm dishes were transfected with the indicated homologous to FTRCA1 and several homologs searched from plasmids. Thirty hours later, the transfected cells were collected and lysed, and the cell supernatant was incubated with Ni2+-NTA His.Bind Resin grass carp, medaka, and fugu. As shown in Fig. 1B, the (Novagen) at 4˚C overnight. The samples were boiled together with searched finTRIM sequences are classified into two groups. One loading buffer (23; Beyotime Biotechnology) for 10 min, and then the group consists of FTR82, FTR83, and FTR84, with a definite ubiquitination was determined by Western blotting using the indicated orthologous relationship among four fish species. The other tag-specific Abs. The tag-specific Abs, including anti-myc, anti-His, and group is composed of FTRCA1 and sequences from zebrafish anti-Flag, are purchased from Cell Signaling Technology. Anti–crucian carpIRF3Abwasraisedbyimmunization of rabbit with prokaryotically and grass carp; obviously, they are not one-to-one orthologs to expressed crucian carp IRF3-DBD as described previously (29). Chloroquine each other, although some grass carp sequences are annotated was purchased from Cell Signaling Technology and MG132 from Calbiochem with similar names to zebrafish sequences. FTRCA1 is grouped

(Germany). with a subset of finTRIMs including zebrafish FTR58, FTR57, Downloaded from Fluorescence microscopy FTR41, FTR42, and FTR94, indicating that FTRCA1 is likely a species-specific finTRIM. HEK293T cells were plated overnight on microscopic cover glasses in six-well plates and transfected with Red-TBK1 and FTRCA1-EGFP in- FTRCA1 is an ISG dividually or collectively by FuGENE HD Transfection Reagent (Promega). After 24 h posttransfection, the transfected cells were rinsed twice with PBS Considering that FTRCA1 was cloned from virally infected followed by fixation with 4% (v/v) paraformaldehyde for 15 min at room CAB cells, the transcript of FTRCA1 was investigated in CAB http://www.jimmunol.org/ temperature, washed with PBS, incubated with 0.2% Triton X-100 for cells following GCRV infection, rIFN treatment, and transfection 30 min, and finally stained with DAPI (100 ng/ml) for 10 min. Then, cells of poly(I:C). Compared with a basal level of FTRCA1 mRNA in were examined under a confocal microscope (Leica LCS SP8 confocal system. Objectives: 103/0.30 and 633/1.4 oil; analysis software: Leica resting cells, GCRV infection, rIFN treatment, or poly(I:C) trans- application suite X). fection resulted in an increase in FTRCA1 mRNA along with in- ducing time, showing a similar expression pattern to that of Mx1 Results (GenBank Accession no. AY303813, https://www.ncbi.nlm.nih.gov/ FTRCA1 is a species-specific TRIM belonging to FTR family nuccore/AY303813), a typical ISG (Fig. 1C). To further character- Based on an expressed sequence tag retrieved from a suppression ize the expression of FTRCA1, a 349-bp 59-flanking sequence was subtractive hybridization cDNA library made with mRNAs de- cloned by genome walking and used to make a promoter-driven by guest on September 29, 2021 rived from virally infected CAB cells (32), we further cloned a luciferase plasmid, FTRCA1pro-luc (Supplemental Fig. 3). As full-length cDNA (Supplemental Fig. 1A), which encodes a 542-aa shown in Fig. 1D, compared with the basal activity of pGL3- TRIM protein composed of an N-terminal RING finger domain, basic, FTRCA1 promoter-driven luciferase activity was signifi- two B-boxes, and a coiled-coil domain immediately preceded by cantly induced in CAB cells when treated with rIFN or poly(I:C) a C-terminal PRY-SPRY domain (Supplemental Fig. 1B). This (analogous to extracellular dsRNA), or transfected with poly(I:C) protein is most homologous to human TRIM16; however, human (analogous to intracellular dsRNA), indicating that FTRCA1 and zebrafish TRIM16s harbor a variant N terminus without promoter is activated by TLR- or RLR-triggered signaling. Con- the conserved RING domain (Fig. 1A). Genome-wide search sistently, transfection of CAB cells with pivotal molecules in- of homologs in zebrafish, D. rerio; fugu, T. rubripes; medaka, volved in RLR signaling, including RIG-I, MDA5, MITA, TBK1, O. latipes; grass carp C. Idella; common carp, C. carpio;and IRF3, and IRF7 (33, 36), effectively stimulated the activity of barbel, S. grahami and subsequent phylogenetic analysis FTRCA1pro-luc (Fig. 1E). Additionally, TBK1-directed activation showed that this protein is most homologous to finTRIM (FTR) of FTRCA1pro-luc was severely diminished by cotransfection of family members, and all FTR proteins are evolutionarily closest STAT1-DC (Fig. 1F), a dominant negative mutant of STAT1 that is to mammalian TRIM16, then to TRIM39, TRIM47, and TRIM25 involved in Jak-Stat signaling pathway (31). These results indicate (Supplemental Fig. 2). TRIM16s and finTRIMs cluster in a that FTRCA1 is an ISG. well-supported group, indicating that this sequence is not the fish ortholog of TRIM16. Notably, fish TRIM16s rather than finTRIMs FTRCA1 negatively regulates intracellular poly(I:C)-triggered have true orthologs in nonfish vertebrates (Supplemental Fig. 2), IFN response verifying a previous finding that FTR family is specific to fish We hypothesized that FTRCA1 plays a role in IFN response. To this genomes, and the family gene number varies greatly among fish end, we assessed its potential to affect crucian carp IFN promoter species (14). Therefore, we named this fish sequence FTRCA1 activation by luciferase assays. As shown in Fig. 2A, transfection (finTRIM C. auratus 1). of FTRCA1 alone had no effect on the activation of crucian carp We also failed to find a one-to-one ortholog of FTRCA1 by C. auratus IFN promoter-driven luciferase construct (CaIFNpro- searching a draft genome of gibel carp (C. auratus gibello), a luc) relative to transfection of empty vector pcDNA3.1; however, species belonging to a same genus, Carassius, as crucian carp this transfection significantly inhibited both extracellular and in- (C. auratus), because the five gibel carp sequences, which are tracellular poly(I:C)-driven crucian carp IFN promoter activity. most homologous to FTRCA1 through BLAST, did not form a Consistently, transfection of FTRCA1 in CAB cells resulted in an single clade with FTRCA1 (Supplemental Fig. 2). Fish TRIM attenuated induction of crucian carp IFN transcripts by extracel- genes are generally named after the human orthologs (14, 15). lular poly(I:C) or intracellular poly(I:C) (Fig. 2B). These results Whereas human and zebrafish TRIM proteins are comprehensively indicate a negative role of FTRCA1 in TLR- or RLR-triggered 4 SPECIES-SPECIFIC REGULATION OF FISH IFN IMMUNE RESPONSE Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 1. FTRCA1 is characterized as a crucian carp–specific ISG. (A) Schematic representation of crucian carp FTRCA1, FTRCA1-DRING, human TRIM16, and mouse TRIM16 proteins. (B) Phylogenetic tree analysis indicated that FTRCA1 is a crucian carp–specific gene of finTRIM subfamily. (C) FTRCA1 was transcriptionally induced by IFN and IFN stimuli. CAB cells seeded in six-well plates were transfected with poly(I:C) (1 mg/ml) or 3 treated with rIFN (5 ng/ml) or GCRV (10 tissue culture ID50 per milliliter) for the indicated times, followed by RT-qPCR analysis of FTRCA1 and Mx1 transcripts. Error bars represent SD obtained by measuring each sample in triplicate. (D) FTRCA1 promoter was activated by intracellular poly(I:C) or extracellular poly(I:C) or rIFN. CAB cells seeded in 24-well plates were cotransfected with FTRCA1 promoter-driven luciferase plasmid (FTRCA1pro- Luc, 250 ng) or control vector pGL3-Basic (250 ng). Renilla luciferase plasmid (pRL-TK, 25 ng) was included to normalize the expression level. Twenty- four hours later, the transfected cells were treated with 10 ng/ml rIFN, or 50 mg/ml poly(I:C) as extracellular dsRNA, or transfected again with 1 mg/ml poly(I:C) as intracellular dsRNA, or left untreated as a control (null) for additional 24 h, followed by detection of luciferase activity. (E) FTRCA1 promoter was activated by pivotal molecules of RLR signaling. CAB cells were transfected as in (D) with the indicated constructs, followed by luciferase activity at 48 h posttransfection. (F) TBK1-triggered activation of FTRCA1 promoter was mediated by Stat1 pathway. CAB cells were cotransfected with FTRCA1pro-Luc, together with TBK1 or pcDNA3.1 as negative control, also with each of the dominant negative plasmids, including IRF3-DN, IRF7-DN, and STAT1-DC (each 200 ng). Forty-eight hours later, the cells were collected for luciferase assays.

IFN response. Because TBK1 is a crucial kinase responsible for transcripts by intracellular poly(I:C). Expectedly, siRNA- IFN expression (33), we also detected the transcription of crucian mediated knockdown of FTRCA1 resulted in an enhanced tran- carp TBK1 gene. However, transfection of FTRCA1 did not alter scription of crucian carp IFN gene, but did not influence the the transcription levels of cellular TBK1 (Fig. 2B), which was transcription of crucian carp TBK1 (Fig. 2C). constitutively expressed and not induced by poly(I:C), similar to To determine the mechanism underlying FTRCA1-mediated our previous results (33). downregulation of IFN transcription, the endogenous IRF3 To further determine the function of FTRCA1 on the tran- protein was investigated in CAB cells under both conditions. scriptional change of endogenous IFN and TBK1, we used siRNA Crucian carp IRF3 protein was basally expressed in resting technology to knock down FTRCA1 expression in CAB cells CAB cells; transfection of poly(I:C) alone induced an increase followed by transfection of poly(I:C) (Fig. 2C). Compared with in the levels of a basally expressed form of IRF3 protein and also transfection of a NC, transfection of either of two FTRCA1-specific a bigger one (Fig. 2D), the latter of which has been confirmed siRNA sequences markedly decreased the levels of FTRCA1 as a phosphorylated form of crucian carp IRF3 protein (29, 37). The Journal of Immunology 5 Downloaded from http://www.jimmunol.org/

FIGURE 2. FTRCA1 negatively regulates intracellular poly(I:C)-triggered IFN response (A). Intracellular or extracellular poly(I:C)-triggered activation of crucian carp IFN promoter was inhibited by overexpression of FTRCA1. EPC cells seeded in 24-well plates overnight were transfected with CaIFNpro-Luc together with FTRCA1 or pcDNA3.1 (250 ng each) as negative control. Twenty-four hours later, the cells were transfected again with 1 mg/ml poly(I:C) as intracellular dsRNA or treated with 50 mg/ml poly(I:C) as extracelluar dsRNA for additional 24 h, followed by detection of luciferase activity. (B) Intracellular or extracellular poly(I:C)-induced IFN expression was inhibited by overexpression of FTRCA1. CAB cells seeded in 3.5-cm2 dishes overnight were transfected with FTRCA1 or pcDNA3.12 as control. Twenty-four hours later, the transfected cells were transfected by guest on September 29, 2021 again with 1 mg/ml poly(I:C) or treated with 50 mg/ml poly(I:C) for additional 24 h, followed by RT-qPCR analysis of the transcription of cellular TBK1 and IFN, and of the transfected FTRCA1 plasmid. (C) Intracellular poly(I:C)-induced IFN expression was enhanced by knockdown of FTRCA1. CAB cells were transfected with NC or FTRCA1-specific siRNA (siRNA1, siRNA2) and 24 h later, transfected again with 1 mg/ml poly(I:C) for additional 24 h. The endogenous FTRCA1, TBK1, and IFN mRNAs were determined by RT-qPCR. The relative expression was normalized to the expression of b-actin and represented as fold induction relative to the expression level in control cells, which was set to 1. Error bars represent SD obtained by measuring each sample in triplicate. (D) Intracellular poly(I:C)-induced IRF3 phosphorylation was blocked by overxpression of FTRCA1. CAB cells seeded in 3.5-cm2 dishes overnight were transfected as in Fig. 2B. (E) Intracellular poly(I:C)-induced IRF3 phosphorylation was enhanced by knockdown of FTRCA1. CAB cells were transfected as in Fig. 2C. Western blotting was used to detect the expression of endogenous IRF3 protein and the phosphorylated IRF3 protein by crucian carp IRF3 Ab (D and E).

As anticipated, overexpression of FTRCA1 obviously decreased expression of FTRCA1 (Fig. 3A), in a dose-dependent manner the amount of crucian carp IRF3 protein that was induced by (Fig. 3B). Interestingly, FTRCA1 did not show any inhibitory intracellular poly(I:C), concomitantly with reduced levels of effect on IRF3-mediated promoter activation but did on IRF7- the phosphorylated IRF3 (Fig. 2D). Conversely, knockdown mediated IFN promoter activation (Fig. 3A, 3B). Zebrafish has of FTRCA1 promoted intracellular poly(I:C)-triggered IRF3 four type I IFN genes, IFN-w1–4 (38). Because zebrafish IFN-w1 protein induction and also the levels of IRF3 phosphorylation and IFN-w3 are differentially activated by fish IRF3 and IRF7, relative to transfection of NC (Fig. 2E). Together, these re- respectively (33), we examined whether FTRCA1 had differen- sults indicate that FTRCA1 negatively regulates intracellular tially regulatory roles in distinct fish IFN induction. To this end, poly(I:C)-triggered IFN response through impairing IRF3 luciferase assays were repeated with zebrafish IFN-w1 and IFN-w3 activation. promoter-driven luciferase constructs (DrIFN-w1pro-luc and DrIFN-w3pro-luc) instead of CaIFNpro-luc. Similar to our previ- FTRCA1 interacts with key components of RLR pathway to ous results (33, 37), fish IRF3 effectively activated DrIFN-w1 impair IFN response promoter but not DrIFN-w3 promoter, and the other fish RLR Because RLR signaling is involved in intracellular poly(I:C)- signaling molecules, including IRF7, gave a strong activation of triggered IFN response (22, 33), luciferase assays were further both zebrafish IFN promoters (Fig. 3C). Unexpectedly, transfec- used to investigate which step of RLR signaling was targeted by tion of FTRCA1 still exhibited the inhibitory effect on IRF7- FTRCA1. Crucian carp IFN promoter was effectively activated in mediated activation of both zebrafish IFN-w1andIFN-w3 EPC cells by overexpression of each pivotal RLR signaling mol- promoters but not on IRF3-mediated ones (Fig. 3C), indicating ecule, including RIG-I, MDA5, MAVS, MITA, TBK1, and IRF7; no regulatory selectivity of FTRCA1 on distinct fish IFN gene however, this activation was significantly inhibited by ectopic expression. 6 SPECIES-SPECIFIC REGULATION OF FISH IFN IMMUNE RESPONSE Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 3. FTRCA1 interacts with key components of RLR pathway to impair IFN response (A–C). Overexpression of FTRCA1 blocked the activation of crucian carp IFN promoter (A and B) or two zebrafish IFN promoters (C) by pivotal molecules of RLR signaling except for IRF3. EPC cells seeded in 24-well plates overnight were transfected with crucian carp IFN promoter plasmid (IFNpro-Luc; A and B) or two zebrafish IFN promoter plasmids (DrIFNw1pro-Luc and DrIFNw3pro-Luc; C), together with FTRCA1 and each of pivotal molecules of RLR signaling at a 200 ng (A) or increasing amounts (0, 100, 200, 300 ng; B and C). At 48 h posttransfection, the cells were collected for luciferase assays. (D and E) FTRCA1 interacts with key components of RLR pathway except IRF7. HEK293T cells seeded in 10-cm2 dishes overnight were transfected with 4 mg Flag-FTRCA1 and 4 mg pivotal molecules of RLR signaling, including RIG-I, MDA5, MITA, TBK1, IRF3, and IRF7 (D), or with 2 mg Flag-FTRCA1 together with 1 mg MITA-myc or 3 mg IRF7-myc (E). Two hundred nanograms EGFP-myc was transfected in parallel as control. Forty-eight hours later, Co-IP was performed with anti-Flag Ab and the immunoprecipitates were analyzed by Western blotting with anti-myc Ab. Arrow in (D) indicates a nonspecific band existed in all lanes.

Co-IP assays showed that FTRCA1 was associated with RIG-I, TBK1 by transfection of 293T cells with two constructs, MDA5,MITA,TBK1,andIRF3,butnotwithIRF7,in pEGFP-FTRCA1 and pDsRed2-TBK1, which encode fusion HEK293T cells when FTRCA1 was transfected with each plasmid proteins FTRCA1-EGFP (for FTRCA1) and DsRed2-TBK1 expressing RLR signaling components at a dose ratio of 1:1 (for TBK1), respectively. Confocal microscopy examination (Fig. 3D). To rule out a possibility that the failed detection of revealed that both FTRCA1 and TBK1 were dispersed through FTRCA1 to interact with fish IRF7 was due to a relatively weak the cytoplasm, showing a similar pattern with discrete cyto- expression of IRF7, we increased the dose ratio of IRF7 and plasmic speckles (Fig. 4A). Cotransfection of pEGFP-FTRCA1 FTRCA1 from 1:1 to 1.5:1 in next Co-IP assays; however, we still and pDsRed2-TBK1 showed colocalization of FTRCA1 and did not detect their interaction (Fig. 3E). In addition, we consis- TBK1 (Fig. 4B). tently observed a bigger FTRCA1 protein, which appeared in cells Subsequently, RT-qPCR was used to evaluate the effect of transfected with FTRCA1 and TBK1 together (Fig. 3D). These data FTRCA1onthetranscriptionofTBK1andIFNgeneinEPC promoted us to focus on TBK1 as a candidate target of FTRCA1. cells transfected with TBK1 alone or TBK1 and FTRCA1 to- gether. We designed primers to amplify the mRNAs derived FTRCA1 attenuates IFN response by promoting only from the transfected constructs expressing FTRCA1 or TBK1 degradation TBK1. A similar expression pattern was observed for the To determine the physiological function of FTRCA1 on TBK1, we transfected FTRCA1 and TBK1, which were increased ini- initially investigated the subcellular localization of FTRCA1 and tially, peaked at 12 h posttransfection, and decreased thereafter The Journal of Immunology 7

FIGURE 4. FTRCA1 and TBK1 are co- localized in cytoplasm. HEK293T cells plated onto glass coverslips in six-well plates overnight were transfected with 1 mg pEGFP-FTRCA1 and 1 mg pdsRED-TBK1 individually (A) or collectively (B). Empty

construct pdsRED-C1 and pEGFP were Downloaded from transfected in parallel as control. At 24 h posttransfection, the cells were fixed and examined using a confocal microscopy (original magnification 3400). DAPI staining shows the nuclei of cells. The last column shows a higher-power view of the area in the box. http://www.jimmunol.org/ by guest on September 29, 2021

(Fig. 5A, 5B). As expected, the endogenous EPC IFN chloroquine (inhibitors of autophage-lysosome–dependent was transcriptionally induced by overexpression of TBK1; degradation pathway) on FTRCA1-mediated downregulation of however, this induction was significantly impaired by coex- TBK1 protein in EPC cells. Whereas proteasomal inhibition pression of FTRCA1 (Fig. 5C). In contrast, overexpression MG132 had no effect on TBK1 degradation (Fig. 6A), inhibition of FTRCA1 did not alter the transfected TBK1 transcription of phagosome-lysosome fusion by either NH4Cl or chloroquine (Fig. 5B). resulted in a significant accumulation of TBK1 in the pres- We asked whether FTRCA1 might destabilize TBK1 protein. ence of FTRCA1 (Fig. 6B). Similar results were seen in Western blot analysis with tag-specific Abs showed that, HEK293T cells, showing that overexpression of FTRCA1 compared with the abundance of TBK1 protein in control EPC robustly targeted TBK1 for degradation in a dose-dependent cells transfected with empty vector pcDNA3.1, transfection of manner, which was profoundly blocked by addition of NH4Cl FTRCA1 resulted in consistently attenuated levels of TBK1 (Fig. 6C). These results suggest that FTRCA1 promotes protein, in a time-dependent fashion (Fig. 5D) and also in a dose- TBK1 degradation through autophage-lysosomal degradation dependent fashion (Fig. 5E). Similar results were obtained in pathway. HEK293T cells instead of EPC cells (Fig. 5F, 5G). As in FTRCA1 mediates TBK1 degradation independently of Fig. 3D, a bigger FTRCA1 protein was observed besides the its phosphorylation normal protein (Fig. 5D–G). These results indicate that FTRCA1 negatively regulates IFN response by targeting TBK1 for In the aforementioned Western blot experiments (Figs. 3D, degradation. 5D–G, 6A–C), two protein bands of FTRCA1 always appeared when FTRCA1 and TBK1 were overexpressed together, sug- FTRCA1 mediates TBK1 degradation through autophage- gesting that TBK1 may phosphorylate FTRCA1. In keeping lysosomal degradation pathway with this, EPC cells were transfected with FTRCA1 alone or To delineate the mechanisms involvedinFTRCA1interfering together with TBK1, and the whole-cell extracts were incu- with TBK1 expression during the posttranscriptional process, bated with or without l protein phosphatase. In the absence of we investigated the effect of MG132 (inhibitor of ubiquitin- TBK1, there appeared only a single protein corresponding to the proteasome-dependent degradation pathway) and NH4Cl, faster-migrating band; in the presence of TBK1, a slower-migrating 8 SPECIES-SPECIFIC REGULATION OF FISH IFN IMMUNE RESPONSE Downloaded from http://www.jimmunol.org/

FIGURE 5. FTRCA1 facilitates TBK1 degradation to downregulate TBK1-triggered IFN response (A–C). FTRCA1 downregulated TBK1- directed IFN gene transcription but did not influence TBK1 transcription. EPC cells seeded in 3.5-cm2 dishes were cotransfected with TBK1- myc and Flag-FTRCA1 (each 1 mg) for the indicated times, followed by RT-qPCR analysis of the mRNAs derived from the transfected plasmid FTRCA1 (A), the transfected plasmid TBK1 (B), and the cellular IFN (C). The relative expression was normalized to the expression of b-actin and represented as fold induction relative to the expression level in control cells, which was set to 1. Error bars represent SD obtained by by guest on September 29, 2021 measuring each sample in triplicate. (D and E) FTRCA1 targeted TBK1 for degradation in EPC cells. EPC cells seeded in 6-cm2 dishes were cotransfected with TBK1-myc and Flag-FTRCA1 (2 mg each) for different time points (D)orwithTBK1-myc(3mg) and increasing amounts of Flag-FTRCA1 (0, 0.5, 1, 2, 3 mg)for48h(E). Western blotting was used to analyze the expression of TBK1 and FTRCA1 proteins using corresponding Abs. (F and G) FTRCA1 targeted TBK1 for degradation in HEK293T cells. HEK293T cells seeded in 6-cm2 dishes were treated as in (D)and(E).

band was simultaneously generated and consistently enhanced FTRCA1 exerts its function dependent on the N-terminal along with transfection time (Fig. 7A). L protein phosphatase RING domain treatment did not affect the mobility of the faster-migrating pro- FTRCA1 belongs to the TRIM family, featuring a typical RING tein, but resulted in the disappearance of the slower-migrating domain in the N terminus (Fig. 1A). To determine the requirement band (Fig. 7A). Moreover, the slower-migrating protein was un- of FTRCA1 RING domain on TBK1 downregulation, the potential detectable when TBK1-K38M, a kinase inactivation construct of of FTRCA1-DRING to downregulate TBK1 was tested. Western TBK1 (33), was transfected instead of TBK1 (Fig. 7B). Notably, a blotting showed that FTRCA1-DRING lost the ability to mediate slower-migrating protein was also detectable in cells transfected TBK1 protein degradation (Fig. 8A) and also TBK1-K38M deg- with TBK1 and FTRCA1-DRING, an FTRCA1 mutant devoid of radation (Fig. 8B), although like wild-type FTRCA1, FTRCA1- N-terminal RING domain (Fig. 1A) but was undetectable in cells DRING was still phosphorylated by TBK1 (Fig. 8A) but not by transfected with TBK1-K38M instead of TBK1 (Fig. 7C). These TBK1-K38M (Fig. 8B). Consistently, FTRCA1-DRING failed to results suggest that FTRCA1 is phosphorylated by TBK1, and the downregulate TBK1- or IRF7-mediated IFN promoter activation phosphorylation occurs in the central or C-terminal domains of (Fig. 8C). These results suggest that RING domain is essential FTRCA1. for FTRCA1-targeted TBK1 degradation and also for FTRCA1- To determine whether such phosphorylation was essential for downregulated IFN response. FTRCA1 on TBK1 downregulation, FTRCA1 and TBK1-K38M were overexpressed together followed by detecting the sta- FTRCA1 does not polyubiquitinate TBK1 bility of TBK1-K38M. Surprisingly, whereas FTRCA1 could Cotransfection of HEK293T cells with His-ubiquitin and Flag- not be phosphorylated, it still exhibited the potential to facil- FTRCA1 followed by affinity purification of His-ubiquitin using itate TBK1-K38M degradation in a time-dependent fashion Ni2+-NTA resin showed that the wild-type FTRCA1 was ubiq- (Fig.7D)andalsoinadose-dependentfashion(Fig.7E), uitinated, but FTRCA1-DRING was not (Fig. 9A). Interestingly, indicating that FTRCA1 phosphorylation is not required for the nonubiquitinated FTRCA1 and FTRCA1-DRING were TBK1 degradation. also accumulated by Ni2+-NTA resin (upper panel in Fig. 9A), The Journal of Immunology 9 Downloaded from

FIGURE 6. FTRCA1 mediates TBK1 degradation through lysosomal degradation pathway (A and B). FTRCA1-targeted TBK1 degradation was blocked 2 2 in EPC cells by chloroquine and NH4Cl but not by MG132. EPC cells seeded in 6-cm dishes (A and right panel in B) or 3.5-cm dishes (left panel in B) were cotransfected with TBK1-myc and Flag-FTRCA1 at the indicated amounts. After 24 h, the transfected cells were treated with or without 20 mM

MG132 (A), 50 mM chloroquine (left panel in B), or 50 mM NH4Cl (right panel in B) for additional 9 h. Western blotting was used to analyze the expression http://www.jimmunol.org/ of TBK1 and FTRCA1 proteins using corresponding Abs (C). FTRCA1-targeted TBK1 degradation was blocked in HEK293T cells by NH4Cl. HEK293T cells seeded in 6-cm2 dishes were cotransfected with TBK1-myc (3 mg) and Flag-FTRCA1 at increasing amounts (0, 0.5, 1, 1.5, 2, 3 mg). After

24 h, the transfected cells were treated with or without 50 mM NH4Cl for additional 9 h. likely due to the existence of tandem histidines within FTRCA1 selected three mutants, H30A, C33A, and C36A, to test their ef- protein (Supplemental Fig. 1A), because His-specific Ab sometimes fects on TBK1 protein degradation and found that, as expected, could detect weak bands in Ni2+-NTA affinity-purified samples, they lost the ability to facilitate the degradation of TBK1 protein where FTRCA1 were largely concentrated, but not in input samples (Fig. 10C). (data not shown). Subsequent Co-IP assays showed that the Flag- by guest on September 29, 2021 specific Ab-precipitated FTRCA1 was also ubiquitinated (Fig. Discussion 9B). Considering that a key feature of TRIM proteins is their In mammals, many TRIM family members have been identified as N-terminal RING-finger E3 ligase activity (6, 39), these results modulators to regulate IFN antiviral signaling (2). TBK1 is an suggest that FTRCA1 is an E3 ligase, likely dependent on its essential protein kinase for phosphorylating IRF3 in IFN anti- RING domain. viral response; therefore, it is certainly an ideal target to be When His-ubiquitin and HA-TBK1 were transfected into modulated by TRIM proteins. Consistently, TRIM9s (45) and 293T cells together with or without Flag-FTRCA1, polyubiquitinated TRIM26 (11) facilitate RLR-triggered IFN antiviral response by TBK1 was detected (Fig. 9C). There was no difference in the targeting TBK1; instead, TRIM27 induces TBK1 degradation via polyubiquitination signalingofTBK1inthepresenceorabsenceof ubiquitin-proteasome pathway (46, 47). TRIM11 is believed to FTRCA1 and, similarly, polyubiquitinated TBK1-K38M was also interact with TBK1 and negatively regulates IFN-b production, detected but was not influenced by transfection of FTRCA1 although the detailed mechanism remains uncharacterized (48). (Fig. 9C), suggesting that FTRCA1 does not polyubiquitinate TBK1. In the current study, we identify a species-specific fish TRIM gene, FTRCA1, from crucian carp and establish that FTRCA1 The RING domain mutants of FTRCA1 lose the ability to negatively regulates IFN response by targeting TBK1 for negatively modulate TBK1-mediated IFN response lyososomal-dependent degradation. First, FTRCA1 is associated We hypothesized that the putative E3 ligase activity is required for with TBK1 by Co-IP assays (Fig. 3D) and subcellular colocali- FTRCA1 downregulating TBK1-mediated IFN response. To this zation analysis (Fig. 4). Secondly, TBK1-mediated IFN promoter end, we made seven single-site mutants of FTRCA1, including activation and IFN gene transcription are markedly abrogated by C13A, C16A, C28A, H30A, C33A, C36A, C52A, and one overexpression of FTRCA1 (Figs. 3A, 5C). Thirdly, although multiple-site mutant mut7 (Fig. 9D). These 7 aa are conserved in FTRCA1 does not alter TBK1 gene transcription, it targets the RING domains of FTRCA1 and the other TRIM proteins TBK1 protein for lyososomal-dependent degradation (Figs. 5, 6). (Fig. 9D) and are essential for E3 ligase activity in human Fourthly, FTRCA1 does not inhibit fish IRF3-mediated IFN TRIM23 (40, 41), human TRIM26 (11), human TRIM31 (42, 43), promoter activation (Fig. 3), but it exhibits a potential to block and human TRIM40 (44), respectively. Unlike wild-type IRF3 phosphorylation in intracellular poly(I:C)-triggered IFN FTRCA1, ubiquitination assays showed that these mutants all response, which is further corroborated by siRNA-mediated could not be ubiquitinated (Fig. 9E), indicating that these mutants knockdown analysis of FTRCA1 (Fig. 2). Considering that cannot have E3 ligase activity. Unlike wild-type FTRCA1, which FTRCA1 is significantly induced in response to virus infection or has the potential to attenuate TBK1-mediated IFN response rIFN (Fig. 1C), these results together suggest that FTRCA1 (Fig. 10A), all mutants lost the ability to inhibit the activation functions as a negative regulator to fine-tune fish IFN response in of fish IFN promoter by TBK1 (Fig. 10B). Finally, we randomly virally infected cells. 10 SPECIES-SPECIFIC REGULATION OF FISH IFN IMMUNE RESPONSE Downloaded from http://www.jimmunol.org/

FIGURE 7. FTRCA1 mediates TBK1 degradation independently of its phosphorylation (A and B). FTRCA1 was phosphorylated by TBK1 but not by TBK1-K38M. HEK293T cells seeded in 3.5-cm2 dishes were cotransfected with Flag-FTRCA1 and TBK1-myc (1 mg each) for the indicated times A (12, 24, 36 h) ( ) or with Flag-FTRCA1 and TBK1-myc (1 mg each), Flag-FTRCA1, and TBK1-K38M-myc (1 mg each) for the indicated times (0, 12, 24, by guest on September 29, 2021 48 h) (B). The cell lysates were incubated for 30 min with or without l phosphatase (L PP) at 30˚C before SDS-PAGE (A) or directly submitted to SDS- PAGE (B). (C) FTRCA1-DRING was phosphorylated by TBK1 but not by TBK1-K38M. EPC cells seeded in 6-cm2 dishes were transfected with Flag- FTRCA1-DRING (2 mg) together with TBK1-myc or TBK1-K38M-myc or pcDNA3.1 (each 2 mg) for the indicated times. (D and E) FTRCA1 mediated TBK1 degradation in a time-dependent fashion (D) and in a dose-dependent fashion (E). EPC cells seeded in 6-cm2 dishes were cotransfected with TBK1-K38M-myc and Flag-FTRCA1 or pcDNA3.1 (3 mg each) as control for the indicated times (0, 12, 24, 48 h) (D) or with TBK1-K38M-myc (3 mg) and Flag-FTRCA1 at increasing doses (from 0 to 3 mg) for 24 h (E). Western blotting was used to analyze the expression of TBK1, TBK1-K38M, FTRCA1, and FTRCA1-DRING proteins using tag Abs.

Most TRIM proteins have E3 liagase activity (6). FTRCA1 they cannot have E3 ligase activity. Functionally, human TRIM16 rather than FTRCA1-DRING can be ubiquitinated by transfection is shown to play roles in secretory autophagy and carcinogenesis assays (Fig. 9), suggesting that FTRCA1 is likely an E3 ligase (50), and no antiviral roles have been characterized so far. These with the requirement of RING finger domain. It is noted that this results indicate that despite high resemblance to TRIM16, notion still has to be further verified by a standard in vitro assay FTRCA1 has evolutionarily developed distinct structure and (41), by incubation of only FTRCA1 protein and appropriate function. E1 and E2, because transfection assays cannot exclude whether The ubiquitin-proteasome and autophagy-lysosomal pathways FTRCA1 has recruited another E3 ligase for the observed ubiq- are two major degradation systems responsible for cellular protein uitination. Apart from finTRIM family members, FTRCA1 is most degradation, in which ubiquitination is commonly used as a deg- similar to TRIM16 (Fig. 1B). Interestingly, human TRIM16 does radation signal (51). Given that FTRCA1 is an E3 ligase, it does not have the N-terminal RING domain like FTRCA1 (Fig. 1A), not follow a mechanism to downregulate TBK1 protein by the but it still is polyubiquitinated by transfection assays, indicating ubiquitin-proteasome system, in which the K48-linked polyubiquintin that human TRIM16 is likely an E3 ligase (39). Unlike FTRCA1, chain is catalyzed to the targeted substrate protein. This notion is the E3 ligase activity of human TRIM16 is ascribed to its B-box supported by the findings that FTACA1 cannot enhance TBK1 domains, because they adopt RING-like folds showing a similar polyubiquitination (Fig. 9C) and that MG132 does not block the tertiary structure as a result of being cysteine-rich (49). Obviously, effect of FTRCA1 on TBK1 protein degradation (Fig. 6A). Instead, the RING domain of FTRCA1 follows this rule, as evidenced by both inhibitors of autophagosome-lysosome fusion, NH4Cl and the findings that the FTRCA1 mutants with individual and com- chloroquine, can rescue TBK1 degradation by FTRCA1 (Fig. 6), bined mutation of six conservative cysteine sites and one histidine supporting the notion that FTRCA1 might act as a cargo receptor site within RING domain (Fig. 9E), and FTRCA1-DRING, a mu- that directly recognizes TBK1 and targets it for degradation by tant devoid of RING domain and retaining the intact B-box do- autophagy-lysosomal pathway. Whereas FTRCA1 does not partic- mains (Fig. 9A), cannot be ubiquintinated any more, indicating ipate in TBK1 polyubiquitination, it is likely that the E3 ligase The Journal of Immunology 11 Downloaded from

FIGURE 8. FTRCA1 exerts its function dependent on the N-terminal RING domain (A). TBK1 degradation was targeted by FTRCA1 but not by FTRCA1-DRING. HEK293T cells seeded in 6-cm2 dishes were cotransfected with TBK1-myc and Flag-FTRCA1 or Flag-FTRCA1-DRING or pcDNA3.1 (2 mg each) for the indicated times (0, 12, 24, 48 h). Western blotting was performed by tag Abs. (B) TBK1-K38M degradation was targeted by FTRCA1 but not by FTRCA1-DRING. EPC cells seeded in 6-cm2 dishes were cotransfected as in (A) with the indicated plasmids for the indicated times (0, 12, 24, http://www.jimmunol.org/ 48 h). Western blotting was performed by tag Abs. (C) FTRCA1-DRING lost the ability to attenuate the activation of crucican carp IFN promoter by TBK1 and IRF7. EPC cells seeded in 24-well plates overnight were cotransfected with CaIFNpro-Luc, TBK1, or IRF3 or IRF7 (200 ng each), together with FTRCA1 (upper panel) or FTRCA1-DRING (lower panel) at increasing amounts (0, 100, 400 ng). At 48 h posttransfection, the cells were collected for luciferase assays. activity of FTRCA1 is required for its facilitating TBK1 degra- also facilitating the K27-linked polyubiquitination chains of dation, because deletion of N-terminal RING domain or mutation MAVS during antiviral response (19, 20). Mouse TRIM38 neg- of seven conservative sites within this domain cannot retain the atively regulates TLR3/4-triggered signaling by degrading TRIF by guest on September 29, 2021 E3 ligase activity of FTRCA1 and abrogates the potential of through proteasome pathway, and in contrast, it also mediates FTRCA1 to degrade TBK1 and block IFN response (Figs. 8, 10). Tab2 degradation in a lysosomal-dependent process to down- These results highlight the importance of structural integration regulate TNF-a/IL-1b signaling (53). Moreover, TRIM proteins and E3 ligase activity of FTRCA1 for its negative regulatory have recently been shown to serve as autophagic receptors to function. directly bridge their degradative targets with the core autophagic Our results reveal the functional complexity of FTRCA1 in machinery in response to viral infection (40, 54–56). These re- regulation of fish IFN immune response. Either intracellular sults highlight that TRIM proteins act as a scaffold for the as- poly(I:C)– or extracellular poly(I:C)–triggered IFN response is sembly of larger multiprotein complexes, although the detailed negatively modulated by overexpression of FTRCA1 (Fig. 2A), mechanisms remain largely unexplored (17). indicating that FTRCA1 also plays a negatively regulatory role Whereas TRIM family is widely distributed in metazoans, it in TLR-mediated IFN signaling. In addition, further experiments has been greatly diversified in vertebrates, especially in fish (15). need to delineate the physiological relevance of TBK1-mediated Species-specific TRIM genes exist in mouse and human ge- FTRCA1 phosphorylation, which seems to be dispensable for nomes, although most of human TRIM genes have strict 1:1 TBK1 degradation in the current study (Fig. 7). Besides TBK1, mouse orthologs (13). Systematic analysis of all known 75 hu- FTRCA1 interacts with other RLR signaling molecules, including man TRIMs reveals that half of them function at multiple levels RIG-I, MDA5, and MITA (Fig. 3D), and actually, it is able to in innate signaling pathways and, actually, their transcription and inhibit IFN promoter activation triggered by each of them (Fig. 3), localization are influenced by virus infection, strengthening the indicating a possibility that FTRCA1 also targets these signaling notion that TRIM family is evolved as components of innate factors to negatively modulate IFN response. Surprisingly, antiviral immunity (4). Subsequent survey of the youngest hu- FTRCA1 is not associated with IRF7 but still negatively regulates man TRIM genes has identified 11 genes specific to humans and IRF7-mediated IFN signaling, which is quite opposite to IRF3 African apes and another seven specific to humans; interestingly, (Fig. 3C–E). These differences promoted us to focus on TBK1, these gene copies are variable even within the human population upstream of IRF3/7, as a candidate target of FTRCA1, thus re- (16), indicating a constant expansion of species-specific TRIM vealing the mechanism whereby FTRCA1 degrades TBK1 to genes in the recent evolutionary time, during the speciation of downregulate IFN signaling. Therefore, FTRCA1 appears to act as the great apes (6–7 million years ago) (57). These youngest an essential negative modulator to play broad and complex reg- genes, generating through segmental duplication, might encode ulatory roles in RLR- and TLR-triggered IFN response. Similar new antiviral effectors that are selected to combat new and evolving functional complexity is seen for mammalian TRIMs. Apart from pathogens (16). directing the virions to proteasome (52), human TRIM21 serves as Although the homologous genes of FTRCA1 still have to be a positive regulator for not only sustaining IRF3 activation but discovered in the future, the current data from available fish 12 SPECIES-SPECIFIC REGULATION OF FISH IFN IMMUNE RESPONSE Downloaded from http://www.jimmunol.org/

FIGURE 9. FTRCA1 does not polyubiquitinate TBK1 (A and B). Polyubiquitination of FTRCA1. HEK293T cells seeded in 10-cm2 dishes were by guest on September 29, 2021 transfected with His-ubiquitin (5 mg) and Flag-FTRCA1 (4 mg) individually or collectively, His-ubiquitin (5 mg) and Flag-FTRCA1DR(4mg) (A), or with ubiquitin in the absence or presence of Flag-FTRCA1 (B). Thirty hours later, the cells were collected and the cell lysates were incubated with Ni2-NTA overnight (A) or were immunoprecipitaed by Ab specific to Flag tag (B). Western blotting was used to detect the polyubiquitination of FTRCA1 using His-specific Ab (A)orubiquitin-specificAb(B). (C) Western blot analysis of TBK1 polyubiquitination by FTRCA1. HEK293T cells seeded in 10-cm2 dishes were transfected with His-ubiquitin (5 mg), together with either or both of TBK1 and FTRCA1, or either or both of TBK1- K38M and FTRCA1 (3 mg each) for 30 h. The cell lysates were incubated with Ni2-NTA overnight. The ubiquitination of TBK1 was determined by Western blotting. (D) Multiple alignments of the RING finger domains from FTRCA1, several known TRIMs. The arrows indicate the seven conserved amino acids, which are selected for mutation to alanine individually or collectively. (E) Individual or combined mutation of seven conserved amino acids within the RING domain of FTRCA1 resulted in a diminished E3 ligase activity. HEK293T cells in 10-cm2 dishes were transfected as in (A) with the indicated plasmids for 30 h. genomes strongly indicate that FTRCA1 is a crucian carp–specific development of immunopathological conditions. Based on ap- TRIM gene, which belongs to finTRIM subfamily (Fig. 1B). parent phylogenetic age, only three zebrafish finTRIM genes, in- Similar to the human-specific TRIM genes (16), finTRIM cluding FTR82/83/84, have orthologs in other fish species, and genes exist exclusively in fish genomes, and most of them are most of zebrafish finTRIM genes arise by genus- or species-specific transcriptionally induced by virus infection (14). Compre- duplication genes, as evidenced by the fact that no clear con- hensive analyses of the available fish genome data reveal that served synteny is found between the loci harboring these fin- finTRIM family has undergone species-dependent expansion TRIMs in different fish species (14). Unlike FTRCA1, zebrafish by diverse mechanisms of gene spreading, resulting in variable FTR83 is highly expressed at a higher constitutive level and is gene numbers and species-specific TRIMs in different fish not induced by viral infection or IFN treatment, but it plays an species (,10 to more than 80) (14, 15). Because of high di- essential role in restricting viral infection by triggering IFN versification of selection pressures exerted by fish species– response (21). These results support the notion that fish-specific specific viruses that are always evolving (58), such ongoing finTRIM proteins participate in innate antiviral immunity, re- expansion might serve as a gene reservoir, allowing a given gardless of its expression feature. Actually, the IFN-inducible fish species to easily acquire new antiviral genes toward pre- feature is not a prerequisite for their antiviral role, such as sent and new viral infection, certainly including the kind of mammalian MAVS, TBK1, and IRF3, which are not induced by genes, such as FTRCA1, that are responsible for the balance of IFN but function as pivotal components of RLR-IFN signaling host innate immunity. (29). Because FTRCA1 is a TRIM specific to crucian carp and Accordingly, our results provide a good example of how an functions as a negative regulator of fish IFN signaling, the data FTR gene regulates fish IFN response toward viral infection, described in this article indicate a recent and ongoing expansion which benefits to avoid overproduction of IFNs, leading to the of finTRIM as part of the innate immune system, providing The Journal of Immunology 13 Downloaded from http://www.jimmunol.org/

FIGURE 10. The RING domain mutants of FTRCA1 lose the ability to negatively modulate TBK1-mediated IFN response. (A and B) Ligase-inactive mutants of FTRCA1 lost the ability to block IFN response. EPC cells seeded in 24-well plates overnight were cotransfected with CaIFNpro-Luc, TBK1, or pcDNA3.1 (200 ng each) together with each of ligase-inactive mutants of FTRCA1 at increasing amounts (0, 100, 200, 300 ng). Forty-eight hours later, the cells were collected for luciferase assays. (C) Ligase-inactive mutants of FTRCA1 lost the ability to degrade TBK1 protein. EPC cells seeded in 3.5-cm2

dishes were transfected with TBK1-myc (1 mg) together with each of ligase-inactive mutants of FTRCA1 at increasing doses (0, 0.5, 1 mg) for 24 h. by guest on September 29, 2021 Western blotting was performed by tag Abs. insights into species-specific evolving regulation of fish IFN an- 8. Wang, K., C. Zou, X. Wang, C. Huang, T. Feng, W. Pan, Q. Wu, P. Wang, and J. Dai. 2018. Interferon-stimulated TRIM69 interrupts dengue virus replication tiviral response. by ubiquitinating viral nonstructural protein 3. PLoS Pathog. 14: e1007287. 9. Zeng, W., L. Sun, X. Jiang, X. Chen, F. Hou, A. Adhikari, M. Xu, and Z. J. Chen. 2010. Reconstitution of the RIG-I pathway reveals a signaling role of unan- Acknowledgments chored polyubiquitin chains in innate immunity. Cell 141: 315–330. We thank Dr. and Prof. Xiaoqin Xia (Institute of Hydrobiology, Chinese 10. Wang, P., W. Zhao, K. Zhao, L. Zhang, and C. Gao. 2015. TRIM26 negatively Academy of Sciences) for searching homologous genes of FTRCA1 from regulates interferon-b production and antiviral response through polyubiquitination and degradation of nuclear IRF3. PLoS Pathog. 11: e1004726. grass carp genome. 11. Ran, Y., J. Zhang, L. L. Liu, Z. Y. Pan, Y. Nie, H. Y. Zhang, and Y. Y. Wang. 2016. Autoubiquitination of TRIM26 links TBK1 to NEMO in RLR-mediated innate antiviral immune response. J. Mol. Cell Biol. 8: 31–43. Disclosures 12. Pertel, T., S. Hausmann, D. Morger, S. Zu¨ger, J. Guerra, J. Lascano, C. Reinhard, The authors have no financial conflicts of interest. F. A. Santoni, P. D. Uchil, L. Chatel, et al. 2011. TRIM5 is an innate immune sensor for the retrovirus capsid lattice. Nature 472: 361–365. 13. Sardiello, M., S. Cairo, B. Fontanella, A. Ballabio, and G. Meroni. 2008. Ge- nomic analysis of the TRIM family reveals two groups of genes with distinct References evolutionary properties. BMC Evol. Biol. 8: 225. 1. Schlee, M., and G. Hartmann. 2016. Discriminating self from non-self in nucleic 14. van der Aa, L. M., J. P. Levraud, M. Yahmi, E. Lauret, V. Briolat, P. Herbomel, acid sensing. Nat. Rev. Immunol. 16: 566–580. A. Benmansour, and P. Boudinot. 2009. A large new subset of TRIM genes 2. van Gent, M., K. M. J. Sparrer, and M. U. Gack. 2018. TRIM proteins and their highly diversified by duplication and positive selection in teleost fish. BMC Biol. roles in antiviral host defenses. Annu. Rev. Virol. 5: 385–405. 7: 7. 3. Ozato, K., D. M. Shin, T. H. Chang, and H. C. Morse, III. 2008. TRIM 15. Boudinot, P., L. M. van der Aa, L. Jouneau, L. Du Pasquier, P. Pontarotti, family proteins and their emerging roles in innate immunity. Nat. Rev. Immunol. V. Briolat, A. Benmansour, and J. P. Levraud. 2011. Origin and evolution of 8: 849–860. TRIM proteins: new insights from the complete TRIM repertoire of zebrafish 4. Versteeg, G. A., R. Rajsbaum, M. T. Sa´nchez-Aparicio, A. M. Maestre, and pufferfish. PLoS One 6: e22022. J. Valdiviezo, M. Shi, K. S. Inn, A. Fernandez-Sesma, J. Jung, and A. Garcı´a- 16. Han, K., D. I. Lou, and S. L. Sawyer. 2011. Identification of a genomic reservoir Sastre. 2013. The E3-ligase TRIM family of proteins regulates signaling pathways for new TRIM genes in primate genomes. PLoS Genet. 7: e1002388. triggered by innate immune pattern-recognition receptors. Immunity 38: 384–398. 17. Rajsbaum, R., A. Garcı´a-Sastre, and G. A. Versteeg. 2014. TRIMmunity: the 5. Reymond, A., G. Meroni, A. Fantozzi, G. Merla, S. Cairo, L. Luzi, D. Riganelli, roles of the TRIM E3-ubiquitin ligase family in innate antiviral immunity. E. Zanaria, S. Messali, S. Cainarca, et al. 2001. The tripartite motif family J. Mol. Biol. 426: 1265–1284. identifies cell compartments. EMBO J. 20: 2140–2151. 18. Carthagena, L., A. Bergamaschi, J. M. Luna, A. David, P. D. Uchil, F. Margottin- 6. Deshaies, R. J., and C. A. Joazeiro. 2009. RING domain E3 ubiquitin ligases. Goguet, W. Mothes, U. Hazan, C. Transy, G. Pancino, and S. Nisole. 2009. Annu. Rev. Biochem. 78: 399–434. Human TRIM gene expression in response to interferons. PLoS One 4: e4894. 7. Barr, S. D., J. R. Smiley, and F. D. Bushman. 2008. The interferon response 19. Yang, K., H. X. Shi, X. Y. Liu, Y. F. Shan, B. Wei, S. Chen, and C. Wang. 2009. inhibits HIV particle production by induction of TRIM22. PLoS Pathog. 4: TRIM21 is essential to sustain IFN regulatory factor 3 activation during antiviral e1000007. response. J. Immunol. 182: 3782–3792. 14 SPECIES-SPECIFIC REGULATION OF FISH IFN IMMUNE RESPONSE

20. Xue, B., H. Li, M. Guo, J. Wang, Y. Xu, X. Zou, R. Deng, G. Li, and H. Zhu. 40. Sparrer, K. M. J., S. Gableske, M. A. Zurenski, Z. M. Parker, F. Full, 2018. TRIM21 promotes innate immune response to RNA viral infection through G. J. Baumgart, J. Kato, G. Pacheco-Rodriguez, C. Liang, O. Pornillos, et al. Lys27-linked polyubiquitination of MAVS. J. Virol. 92: e00321-18. 2017. TRIM23 mediates virus-induced autophagy via activation of TBK1. Nat. 21. Langevin, C., E. Aleksejeva, A. Houel, V. Briolat, C. Torhy, A. Lunazzi, Microbiol. 2: 1543–1557. J. P. Levraud, and P. Boudinot. 2017. FTR83, a member of the large fish-specific 41. Vichi, A., D. M. Payne, G. Pacheco-Rodriguez, J. Moss, and M. Vaughan. 2005. finTRIM family, triggers IFN pathway and counters viral infection. Front. E3 ubiquitin ligase activity of the trifunctional ARD1 (ADP-ribosylation factor Immunol. 8: 617. domain protein 1). Proc. Natl. Acad. Sci. USA 102: 1945–1950. 22. Zhang, Y. B., and J. F. Gui. 2012. Molecular regulation of interferon antiviral 42. Song, H., B. Liu, W. Huai, Z. Yu, W. Wang, J. Zhao, L. Han, G. Jiang, L. Zhang, response in fish. Dev. Comp. Immunol. 38: 193–202. C. Gao, and W. Zhao. 2016. The E3 ubiquitin ligase TRIM31 attenuates NLRP3 23. Boudinot, P., C. Langevin, C. J. Secombes, and J. P. Levraud. 2016. The peculiar inflammasome activation by promoting proteasomal degradation of NLRP3. Nat. characteristics of fish type I interferons. Viruses 8: E298. Commun. 7: 13727. 24. Huang, Y., M. Yang, Y. Yu, Y. Yang, L. Zhou, X. Huang, and Q. Qin. 2016. 43. Liu, B., M. Zhang, H. Chu, H. Zhang, H. Wu, G. Song, P. Wang, K. Zhao, J. Hou, Grouper TRIM13 exerts negative regulation of antiviral immune response X. Wang, et al. 2017. The ubiquitin E3 ligase TRIM31 promotes aggregation and against nodavirus. Fish Shellfish Immunol. 55: 106–115. activation of the signaling adaptor MAVS through Lys63-linked polyubiquitination. 25. Huang, Y., Y. Yu, Y. Yang, M. Yang, L. Zhou, X. Huang, and Q. Qin. 2016. Fish Nat. Immunol. 18: 214–224. TRIM8 exerts antiviral roles through regulation of the proinflammatory factors 44. Zhang, Q., F. Meng, S. Chen, S. W. Plouffe, S. Wu, S. Liu, X. Li, R. Zhou, and interferon signaling. Fish Shellfish Immunol. 54: 435–444. J. Wang, B. Zhao, et al. 2017. Hippo signalling governs cytosolic nucleic 26. Wang, W., Y. Huang, Y. Yu, Y. Yang, M. Xu, X. Chen, S. Ni, Q. Qin, and X. Huang. acid sensing through YAP/TAZ-mediated TBK1 blockade. Nat. Cell Biol. 19: 2016. Fish TRIM39 regulates cell cycle progression and exerts its antiviral function 362–374. against iridovirus and nodavirus. Fish Shellfish Immunol. 50: 1–10. 45. Qin, Y., Q. Liu, S. Tian, W. Xie, J. Cui, and R. F. Wang. 2016. TRIM9 short 27. Yu, Y., X. Huang, J. Zhang, J. Liu, Y. Hu, Y. Yang, J. Cai, Y. Huang, and Q. Qin. isoform preferentially promotes DNA and RNA virus-induced production of type 2016. Fish TRIM16L exerts negative regulation on antiviral immune response I interferon by recruiting GSK3b to TBK1. Cell Res. 26: 613–628. against grouper iridoviruses. Fish Shellfish Immunol. 59: 256–267. 46. Zheng, Q., J. Hou, Y. Zhou, Y. Yang, B. Xie, and X. Cao. 2015. Siglec1 sup- 28. Kim, J., J. W. Kim, D. G. Kim, B. H. Nam, Y. O. Kim, J. Y. Park, and H. J. Kong. presses antiviral innate immune response by inducing TBK1 degradation via the 2018. Molecular characterization of Rhodeus uyekii tripartite motif protein 1 ubiquitin ligase TRIM27. Cell Res. 25: 1121–1136. (TRIM1) involved in IFN-g/LPS-induced NF-kB signaling. Fish Shellfish 47. Cai, J., H. Y. Chen, S. J. Peng, J. L. Meng, Y. Wang, Y. Zhou, X. P. Qian, Downloaded from Immunol. 79: 42–51. X. Y. Sun, X. W. Pang, Y. Zhang, and J. Zhang. 2018. USP7-TRIM27 axis 29. Sun, F., Y. B. Zhang, T. K. Liu, L. Gan, F. F. Yu, Y. Liu, and J. F. Gui. 2010. negatively modulates antiviral type I IFN signaling. FASEB J. 32: 5238–5249. Characterization of fish IRF3 as an IFN-inducible protein reveals evolving 48. Lee, Y., B. Song, C. Park, and K. S. Kwon. 2013. TRIM11 negatively regulates regulation of IFN response in vertebrates. J. Immunol. 185: 7573–7582. IFNb production and antiviral activity by targeting TBK1. PLoS One 8: e63255. 30. Liu, T. K., Y. B. Zhang, Y. Liu, F. Sun, and J. F. Gui. 2011. Cooperative roles of 49. Massiah, M. A., J. A. Matts, K. M. Short, B. N. Simmons, S. Singireddy, Z. Yi, fish protein kinase containing Z-DNA binding domains and double-stranded and T. C. Cox. 2007. Solution structure of the MID1 B-box2 CHC(D/C)C(2)H(2) RNA-dependent protein kinase in interferon-mediated antiviral response. zinc-binding domain: insights into an evolutionarily conserved RING fold.

J. Virol. 85: 12769–12780. J. Mol. Biol. 369: 1–10. http://www.jimmunol.org/ 31. Yu, F.-F., Y.-B. Zhang, T.-K. Liu, Y. Liu, F. Sun, J. Jiang, and J.-F. Gui. 2010. 50. Jena, K. K., S. P. Kolapalli, S. Mehto, P. Nath, B. Das, P. K. Sahoo, A. Ahad, Fish virus-induced interferon exerts antiviral function through Stat1 pathway. G. H. Syed, S. K. Raghav, S. Senapati, et al. 2018. TRIM16 controls assembly Mol. Immunol. 47: 2330–2341. and degradation of protein aggregates by modulating the p62-NRF2 axis and 32. Zhang, Y. B., J. Jiang, Y. D. Chen, R. Zhu, Y. Shi, Q. Y. Zhang, and J. F. Gui. 2007. autophagy. EMBO J. 37: e98358. The innate immune response to grass carp hemorrhagic virus (GCHV) in cultured 51. Dikic, I. 2017. Proteasomal and autophagic degradation systems. Annu. Rev. Carassius auratus blastulae (CAB) cells. Dev. Comp. Immunol. 31: 232–243. Biochem. 86: 193–224. 33. Sun, F., Y. B. Zhang, T. K. Liu, J. Shi, B. Wang, and J. F. Gui. 2011. Fish MITA 52. Mallery, D. L., W. A. McEwan, S. R. Bidgood, G. J. Towers, C. M. Johnson, serves as a mediator for distinct fish IFN gene activation dependent on IRF3 or and L. C. James. 2010. Antibodies mediate intracellular immunity through IRF7. J. Immunol. 187: 2531–2539. tripartite motif-containing 21 (TRIM21). Proc.Natl.Acad.Sci.USA107: 34. Zhang, Q. M., X. Zhao, Z. Li, M. Wu, J. F. Gui, and Y. B. Zhang. 2018. Al- 19985–19990. ternative splicing transcripts of zebrafish LGP2 gene differentially contribute to 53. Hu, M. M., X. Q. Xie, Q. Yang, C. Y. Liao, W. Ye, H. Lin, and H. B. Shu. 2015.

IFN antiviral response. J. Immunol. 200: 688–703. TRIM38 negatively regulates TLR3/4-mediated innate immune and inflamma- by guest on September 29, 2021 35. Short, K. M., and T. C. Cox. 2006. Subclassification of the RBCC/TRIM su- tory responses by two sequential and distinct mechanisms. J. Immunol. 195: perfamily reveals a novel motif necessary for microtubule binding. J. Biol. 4415–4425. Chem. 281: 8970–8980. 54. Mandell, M. A., A. Jain, J. Arko-Mensah, S. Chauhan, T. Kimura, C. Dinkins, 36. Feng, H., Y. B. Zhang, Q. M. Zhang, Z. Li, Q. Y. Zhang, and J. F. Gui. 2015. Zebrafish G. Silvestri, J. Mu¨nch, F. Kirchhoff, A. Simonsen, et al. 2014. TRIM proteins IRF1 regulates IFN antiviral response through binding to IFNf1andIFNf3pro- regulate autophagy and can target autophagic substrates by direct recognition. moters downstream of MyD88 signaling. J. Immunol. 194: 1225–1238. Dev. Cell 30: 394–409. 37. Feng, H., Q. M. Zhang, Y. B. Zhang, Z. Li, J. Zhang, Y. W. Xiong, M. Wu, and 55. Kimura, T., A. Jain, S. W. Choi, M. A. Mandell, K. Schroder, T. Johansen, and J. F. Gui. 2016. Zebrafish IRF1, IRF3, and IRF7 differentially regulate IFNF1 V. Deretic. 2015. TRIM-mediated precision autophagy targets cytoplasmic and IFNF3 expression through assembly of homo- or heteroprotein complexes. regulators of innate immunity. J. Cell Biol. 210: 973–989. J. Immunol. 197: 1893–1904. 56. Kimura, T., A. Jain, S. W. Choi, M. A. Mandell, T. Johansen, and V. Deretic. 38. Aggad, D., M. Mazel, P. Boudinot, K. E. Mogensen, O. J. Hamming, 2017. TRIM-directed selective autophagy regulates immune activation. Autophagy R. Hartmann, S. Kotenko, P. Herbomel, G. Lutfalla, and J. P. Levraud. 2009. The 13: 989–990. two groups of zebrafish virus-induced interferons signal via distinct receptors 57. Perelman, P., W. E. Johnson, C. Roos, H. N. Seua´nez, J. E. Horvath, with specific and shared chains. J. Immunol. 183: 3924–3931. M. A. Moreira, B. Kessing, J. Pontius, M. Roelke, Y. Rumpler, et al. 2011. A 39. Bell, J. L., A. Malyukova, J. K. Holien, J. Koach, M. W. Parker, M. Kavallaris, molecular phylogeny of living primates. PLoS Genet. 7: e1001342. G. M. Marshall, and B. B. Cheung. 2012. TRIM16 acts as an E3 ubiquitin ligase 58. Zhang, Q. Y., and J. F. Gui. 2018. Diversity, evolutionary contribution and and can heterodimerize with other TRIM family members. PLoS One 7: e37470. ecological roles of aquatic viruses. Sci. China Life Sci. 61: 1486–1502.