A Mitochondrial Micropeptide Is Required for Activation of the Nlrp3

A Mitochondrial Micropeptide Is Required for Activation of the Nlrp3 Inflammasome Ankit Bhatta, Maninjay Atianand, Zhaozhao Jiang, Juliet Crabtree, Juliana Blin and Katherine A. Fitzgerald This information is current as of September 27, 2021. J Immunol published online 13 December 2019 http://www.jimmunol.org/content/early/2019/12/12/jimmun ol.1900791 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2019/12/12/jimmunol.190079 Material 1.DCSupplemental

Why The JI? Submit online. http://www.jimmunol.org/

• 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

*average by guest on September 27, 2021

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 December 13, 2019, doi:10.4049/jimmunol.1900791 The Journal of Immunology

A Mitochondrial Micropeptide Is Required for Activation of the Nlrp3 Inﬂammasome

Ankit Bhatta,* Maninjay Atianand,† Zhaozhao Jiang,* Juliet Crabtree,* Juliana Blin,* and Katherine A. Fitzgerald*

Functional peptides encoded by short open reading frames are emerging as important mediators of fundamental biological processes. In this study, we identified a micropeptide produced from a putative long noncoding RNA (lncRNAs) that is important in controlling innate immunity. By studying lncRNAs in mice macrophages, we identified lncRNA 1810058I24Rik, which was downregulated in both human and murine myeloid cells exposed to LPS as well as other TLR ligands and inflammatory cytokines. Analysis of lncRNA 1810058I24Rik subcellular localization revealed that this transcript was localized in the cytosol, prompting us to evaluate its coding potential. In vitro translation with 35S-labeled methionine resulted in translation of a 47 aa micropeptide.

Microscopy and subcellular fractionation studies in macrophages demonstrated endogenous expression of this peptide on the Downloaded from mitochondrion. We thus named this gene mitochondrial micropeptide-47 (Mm47). Crispr–Cas9–mediated deletion of Mm47,as well as small interfering RNA studies in mice primary macrophages, showed that the transcriptional response downstream of TLR4 was intact in cells lacking Mm47. In contrast, Mm47-deficient or knockdown cells were compromised for Nlrp3 inflammasome responses. Activation of Nlrc4 or Aim2 inflammasomes were intact in cells lacking Mm47. This study therefore identifies, to our knowledge, a novel mitochondrial micropeptide Mm47 that is required for the activation of the Nlrp3 inflammasome. This

work further highlights the functional activity of short open reading frame–encoded peptides and underscores their importance in http://www.jimmunol.org/ innate immunity. The Journal of Immunology, 2020, 204: 000–000.

ibonucleic acid molecules can be divided into two cate- These sORF-encoded peptides (SEPs) or micropeptides are now gories: mRNAs and noncoding RNAs. A large proportion being recognized for their functions in development, differen- R of the mammalian genome is transcribed as long non- tiation, transport, and other fundamental biological processes coding RNAs (lncRNAs) (1). Recent studies have demonstrated (7, 9). Indeed, a recent study has revealed that a large number of diverse physiological functions for lncRNAs, including a growing transcripts currently annotated as lncRNAs are associated with appreciation for the role of these molecules in the immune system, ribosomes and, in some cases, are translated (8). Among these, in which they control the differentiation and/or activation of im- one such transcript, Aw112010, was found to be encoded from a by guest on September 27, 2021 mune cells (2–5). In the innate immune system, our group and noncanonical open reading frame (ORF) and found to play a others have found that lncRNAs act as positive or negative role in Salmonella typhimurium infection and intestinal in- regulators of inflammatory gene expression in a variety of im- flammation (8). Similarly, humanin, an antiapoptotic 24 aa SEP mune cells, including macrophages (4, 5). However, many an- localized on the mitochondria, prevents Bax translocation from notated lncRNAs have short open reading frames (sORFs) that the cytosol to the mitochondria (10). Other SEPs, such as Dworf may encode for functional proteins or small peptides (6–8). andMyoregulin,havealsobeenshowntoregulatecalcium transport in the endoplasmic reticulum (ER) (11). Collectively, these studies highlight important roles for SEPs in fundamental *Program in Innate Immunity, Department of Medicine, University of Massachusetts biological processes, including cell death and innate immunity. † Medical School, Worcester, MA 01605; and Department of Immunology, University Macrophages represent the first line of defense against infection of Pittsburgh, Pittsburgh, PA 15261 by detecting pathogens through germline-encoded pattern recog- ORCIDs: 0000-0003-4921-0087 (J.C.); 0000-0002-9628-8920 (J.B.). nition receptors. These pattern recognition receptors recognize Received for publication July 11, 2019. Accepted for publication November 10, specific pathogen-associated molecular patterns (PAMPs) such as 2019. LPS and trigger the expression of proinflammatory cytokines and This work was supported by grants from the National Institutes of Health (AI067497 to K.A.F.) and a T32 Training Grant (T32 AI095213 to A.B.). type I IFNs through the activation of the transcription factors k Address correspondence and reprint requests to Dr. Katherine A. Fitzgerald, University of NF- B and IRF3 (12, 13). Macrophages also upregulate the Massachusetts Medical School, Aaron Lazare Medical Research Building, Room 208, 364, expression of Nlrp3 and IL-1b, both of which are critical for Plantation Street, Worcester, MA 01605. E-mail address: kate.fi[email protected] mounting rapid immune responses during infection (14). Nlrp3 The online version of this article contains supplemental material. forms a multiprotein complex known as the inflammasome. Abbreviations used in this article: BMDM, bone marrow–derived macrophage; ER, Inflammasomes are cytoplasmic supramolecular complexes that endoplasmic reticulum; ETC, electron transport chain; Fwd, forward; Gsdmd, form in response to microbial as well as endogenous damage or gasdermin D; KO, knockout; lincRNA, intergenic lncRNA; lncRNA, long noncoding RNA; MEF, mouse embryonic fibroblast; Mm47, mitochondrial micropep- danger signals (13, 15–17). Inflammasomes activate inflamma- tide-47; mtROS, mitochondrial reactive oxygen species; NTC, nontargeting tory caspases such as caspase-1, which in turn controls the control; OCR, oxygen consumption rate; ORF, open reading frame; PAMP, b pathogen-associated molecular pattern; Poly(dA:dT), poly(deoxyadenylic-deoxy- protolytic maturation of the proinflammatory cytokines IL-1 thymidylic); Poly(I:C), polyinosinic-polycytidilic acid; qRT-PCR, quantitative RT- and IL-18. Caspase-1 also cleaves gasdermin D (Gsdmd) to PCR; Rev, reverse; SEP, sORF-encoded peptide; sgRNA, single-guide RNA; generate an N-terminal pore-forming fragment that facilitates siRNA, small interfering RNA; sORF, short open reading frame. cytokine release and pyroptotic cell death (18–20). The Nlrp3 Copyright Ó 2019 by The American Association of Immunologists, Inc. 0022-1767/19/$37.50 inflammasome is activated in response to a wide variety of

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1900791 2 INFLAMMASOME REGULATION BY sORF-ENCODED PEPTIDES microbial and endogenous danger signals, including nigericin, Reverse Transcription Supermix (1708840; Bio-Rad Laboratories). uric acid crystals, amyloid-b fibrils, and extracellular ATP. The Quantitative PCR on the cDNA was performed in Bio-Rad CFX96 Touch precise mechanisms leading to the formation of the Nlrp3 Real-time PCR using gene-specific primers. The genes were normalized to the housekeeping gene Gapdh. Mouse primers used are GAPDH Fwd 59- inflammasome in response to these diverse ligands are still TGGCAAAGTGGAGATTGTTGC-39, GAPDH Rev 59-AAGATGGTGA- unclear, although K+ efflux and, in some cases, mitochondria are TGGGCTTCCCG-39,MALAT1Fwd59-TTGGGACAGTGGACGTGTGG-39, important (15, 21, 22). MALAT1 Rev 59-TCAAGTGCCAGCAGACAGCA-39,Mm47Fwd59-GCT- 9 9 In this study, we have identified a SEP encoded from an an- CAGAACTATGAAATGCCAAAC-3 ,Mm47Rev5-GGTCTCAGAAGCA- GGTGGAC-39,IL-1b Fwd 59-GCCACCTTTTGACAGTGATGAG-39,IL-1b notated intergenic lncRNA (lincRNA) transcript that is regulated Rev 59-GTTTGGAAGCAGCCCTTCATC-39,Nlrp3Fwd59-CATGTTGCCT- in macrophages exposed to LPS. Loss of function studies identify GTTCTTCCAGAC-39,andNlrp3Rev59-CGGTTGGTGCTTAGACTTGAGA- a key role for this mitochondrial micropeptide-47 (Mm47) in 39.HumanprimersusedwereGAPDHFwd59-TGCAACAACCAACTGCTTA- controlling activation of the Nlrp3 inflammasome. CRISPR– 39, GAPDH Rev 59-AGAGGCAGGGATGATGTTC-39,Mm47Fwd59-CA- 9 9 Cas9 knockout (KO) of Mm47 or small interfering RNA (siRNA)– CCGACATCATGCTCGAGT-3 , and Mm47 Rev 5 -GCCAGATACATTCCA- ACCACGT-39. mediated suppression of Mm47 expression in primary cells resulted in impaired Nlrp3 inflammasome activation, leading Western blots b to reduced activation of caspase-1 and compromised IL-1 BMDMs stimulated as per the experiment were lysed in buffer containing secretion. The levels of Mm47 decrease following LPS stim- 20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40, and 5 mM EDTAwith ulation, suggesting that inflammasomes could be turned off fresh 13 Halt Protease Inhibitor mixture (no. 1861279; Promega). Ho- by eliminating this peptide. Given its mitochondrial locali- mogenized lysates were resolved on 14% SDS-PAGE and transferred to 0.2 mM PVDF membrane. Membranes were blocked with 5% nonfat dry zation and the functional link totheNlrp3inflammasome,we milk (w/v) and probed with Abs diluted in 13 PBS and 0.05% Tween-20 Downloaded from posit that Mm47 represents a novel signaling node in the (v/v). The Abs used were pro–IL-1b (AF-401-NA; R&D Systems), emerging cross-talk between mitochondria and the Nlrp3 caspase-1 (sc-514; Santa Cruz Biotechnology), Gsdmd (AB209845; inflammasome. Abcam), b-actin (Sigma), Nlrp3 (clone cryo-2; Enzo Life Sciences), Gapdh (G9295; Sigma), Flag (A8592; Sigma), KDEL (10C3; Enzo Life Sciences), Tom20 (AB186734; Abcam), VDAC (4866S; Cell Signaling Materials and Methods Technology), and HSP60 (13115; Santa Cruz Biotechnology). The Mm47 Ab was custom made by Thermo Fisher Scientific against the immuno- http://www.jimmunol.org/ Cell lines genic residue 22–47 of Mm47. Membranes were probed with HRP- Primary bone marrow–derived macrophages (BMDM) were generated conjugated anti-mouse (172-1011; Bio-Rad Laboratories) and anti-rabbit from C57/Bl6N mice or B6 mice expressing Cas9–GFP (no. 024858; The (170-6515; Bio-Rad Laboratories) or anti-goat (172-1034; Bio-Rad Lab- Jackson Laboratory) and used or immortalized for later. Briefly, bone oratories) and developed using ECL Chemiluminescent Substrate (Pierce marrow cells were cultured in DMEM supplemented with 10% FCS, 1% Biotechnology). penicillin/streptomycin, and 20% L929 conditioned media for 7 d to dif- Immunofluorescence and confocal microscopy ferentiate into BMDM cells. Immortalized BMDM were generated using J2 transforming retroviruses expressing Raf and Myc as described (23). MitoTracker Deep Red (M22426; Thermo Fisher Scientific) dye for mitochondrial staining was added to live cells as per the manual. Cells Knock down and KO

were fixed on eight-well chambered slides (155411; Lab-Tek) using by guest on September 27, 2021 3 Knock down of Mm47 was performed using Silencer Select small inhibitory ice-cold 4% paraformaldehyde for 15 min and washed with 1 PBS. 3 RNAs purchased from Thermo Fisher Scientific. Two nontargeting control The cells were permeabilized using 0.2% Triton X-100 in 1 PBS for (NTC) siRNAs (catalog no. 4390843, catalog no. 4390846; Thermo Fisher 15 min and blocked using 5% Normal Goat Serum (005-000-121; 3 Scientific) and three siRNAs targeting Mm47 (catalog no. 4390771; Thermo Jackson ImmunoResearch Laboratories) in 1 PBS and 0.2% Triton Fisher Scientific) were used. Lipofectamine and RNAiMAX (13778030) were X-100. Cells were incubated with Abs against Flag–Alexa Fluor 488 used to transfect 30 pmol siRNA in 1 3 106 BMDM cells. The cells were used (5407S; Cell Signaling Technology) or Mm47 (custom made; Thermo for experiments at 72 h posttransfection. Fisher Scientific) and anti-rabbit Alexa Fluor 488 (A11008; Thermo CRISPR–cas9–mediated KO of Mm47 was performed using single- Fisher Scientific) was used as a secondary Ab. The cells were washed guide RNAs (sgRNAs). Nontargeting sgRNAs used were NTC forward with PBS and incubated with DAPI at room temperature for 10 min (Fwd) 59-GGCGAGGTATTCGGCTCCGC-39 and NTC reverse (Rev) and washed again, followed by imaging using Leica SP8 Lightning 59-CGCGGAGCCGAATACCTCGC-39. The sgRNAs targeting Mm47 Confocal Microscope. 9 genomic region were sgRNA1 Fwd 5 -CAACGTGGTTGGAATGTATC- RACE 39, sgRNA1 Rev 59-GATACATTCCAACCACGTTG-39, sgRNA2 Fwd 59-TGAAGAGATTAAGAAGGACC-39, and sgRNA2 Rev 59-GGTCCTT- The 59- and 39-ends of Mm47 transcript were determined using FirstChoice CTTAATCTCTTCA-39. Lentivirus packaging the sgRNAs was used to RLM-RACE Kit (AM1700; Ambion), according to the manufacturer’s transduce immortalized BMDM expressing cas9–GFP. The cells were se- instructions. The 1810058I24Rik gene-specific primers used were Fwd 59- lected using 5 mg/ml puromycin and used for their respective experiments. GATAGCTGCTGGGGACTCAC-39 and Rev 59-CGTGGTTGGAATG- To rescue Mm47 expression, we created a new construct in which the TATCTGGCT-39. Pam sites were altered on the Mm47 DNA sequence to be resistant to cas9-mediated cleavage. Retrovirus was created using the empty Cell fractionation vector or CRISPR–cas9–resistant Mm47 and used to transduce KO cells, Primary BMDM cells were suspended in resuspension buffer (50 mM which were selected for stable lines using 2 mg/ml blasticidin. Tris-HCl [pH 7.4], 150 mM NaCl, and 10 mM MgCl2). To the cell Macrophage stimulation suspension, 10% Nonidet P-40 (v/v) was added, mixed, incubated on ice for 10 min, and centrifuged at 13,000 rpm 3 g for 30 s. The su- The cells were treated with LPS (100 ng/ml), polyinosinic-polycytidilic acid pernatant was used for cytosolic RNA, and the pellets were washed a [Poly(I:C)] (25 mg/ml), Pam3Csk4 (100 nM), CLO97 (200 ng/ml), TNF-a few times and used to extract nuclear RNA using QIAGEN RNAeasy (10 ng/ml), and IFN-a (500 U/ml) purchased from Sigma-Aldrich. Cell Kit. GAPDH and MALAT1 were used as cytoplasmic and nuclear lysate and supernatant were collected for RNA analysis and ELISA, re- controls, respectively. spectively. ELISA was performed for IL-1b, TNF-a, and IL-6 using kits available from R&D Systems. RNA was extracted from cells using the Custom RT2 profiler PCR array Aurum Kit (7326820; Bio-Rad Laboratories). Optimized real-time PCR primers were bound to 96-well plate by QIAGEN RNA and quantitative real-time PCR (330171) specific for 72 annotated lincRNA genes, eight housekeeping genes, six protein-coding inflammatory genes, and quantitative RT-PCR Total RNA was extracted from cells using Aurum Total RNA Mini Kit (qRT-PCR) positive and negative controls. The 2–DCT value for each gene (7326820; Bio-Rad Laboratories) or TRIzol (15596026; Invitrogen), was calculated against the average of b-actin and B2M for each treatment according to the product manual. The cDNA was synthesized using iScript condition. The Journal of Immunology 3

Subcellular fractionation 1810058I24Rik is transcribed from mouse chromosome 6. Mouse embryonic fibroblasts (MEFs) were used to perform sucrose dis- The 1810058I24Rik transcript is spliced and has three exons continuous gradient centrifugation to separate cytosolic, ER, and mito- (Supplemental Fig. 1A). Using 59- and 39-RACE, we determined chondrial fractions as described before (24). Briefly, MEFs were lysed in the full-length sequence (580 nt) of the spliced 1810058I24Rik MTE buffer (270 mM Mannitol plus 10 mM Tris plus 0.1 mM EDTA) and (Supplemental Fig. 1A, 1B). qRT-PCR analysis of RNA extracted 3 spun at 1400 rpm g for 10 min. The supernatant was spun at 15,000 rpm from mouse organs showed a broad expression profile of this 3 g for 10 min to separate crude ER fraction in the supernatant and mitochondrial fraction in the pellet. The crude ER fraction was further spun transcript in muscle, lung, spleen, liver, and kidney as well as the on a sucrose gradient of 1.3, 1.5, and 2 M at 152,000 rpm 3 g for 74 min small and large intestine (Fig. 1D). There was limited expression followed by collection, washing, and drying of the pure ER band. Simi- in brain and heart. larly, the crude mitochondrial fraction in the pellet was loaded to a sucrose gradient of 1 and 1.7 M and spun at 40,000 rpm 3 g for 30 min followed lncRNA 1810058I24Rik is downregulated by LPS in a by collection, washing, and drying of the pure mitochondrial band. All the TRIF-dependent manner high-speed centrifugation was performed in Beckman Coulter SW40-Ti. We next performed copy number analysis to define the abundance In vitro translation assay of this lncRNA. lncRNA 1810058I24Rik was expressed at ∼1250 Plasmids with full wild type sequence and frame-shifted mutant sequence of copies per cell in mouse macrophages, and these levels were re- Mm47 with T7 promoter tag were prepared. The PCR primers used for duced to ∼400 in response to LPS (Fig. 2A, 2B, Supplemental cloning are Mm47 XhoI Flag Fwd 59-ATCCTCGAG ATGGACTACAA- Fig. 1C). 1810058I24Rik is also conserved in humans, where we GGACGACGATGACAAGCTCCAGTTCCTGCTTGGATTTACTT-39,Mm47 XhoI Fwd 59-ATCCTCGAGATGCTCCAGTTCCTGCTTGGATTTACTT-39, found that it was also expressed at high levels in human mono- Mm47 BglII Rev 59-CATAGATCTTCAGGAACTAGGGGGCTTCTT- cytes (∼900 copies per cell) and dendritic cells, in which it also Downloaded from 39,Mm47BglIIFlagRev59-CATAGTCTTCACTTGTCATCGTCGT- exhibits a pattern of downregulation in response to LPS in both cell CCTTGTAGTCGGAACTAGGGGGCTTCTT-39, Mm47 XhoI T7 FS types (Supplemental Fig. 1D [monocytes] and Supplemental Fig. 9 Fwd 5 -ATCCTCGAGTAATACGACTCACTATAGATGACTCCAGTT- 1E [monocyte-derived dendritic cells]). We next sought to address CCTGCTTGGATTTACTT-39, and Mm47 XhoI T7 promoter Fwd 59- CTCGAGTAATACGACTCACTATAGGGACCTCTCACACCCTCCTCG- the mechanism involved in the downregulation of 1810058I24Rik. 39. MEGAscript Transcription kit (AM1334) was used to produce RNA. In wild type BMDM, the expression of 1810058I24Rik was reduced 35

Retic Lysate Kit (AM1200) and EasyTag Express S Protein Labeling in response to LPS stimulation. However, this downregulation was http://www.jimmunol.org/ 35 Mix, [ S] (catalog no. NEG772002MC) were used to generated radiola- lost in cells lacking TLR4 or TRIF but was still observed in cells beled Mm47 peptides. lacking MyD88, suggesting that the TLR4–TRIF pathway shuts Statistical analysis down expression of this transcript (Fig. 2C). Statistical analysis was performed with an unpaired, two-tailed Student t 1810058I24Rik encodes a highly conserved test in GraphPad Prism software, version 7.0. mitochondrial micropeptide Understanding the subcellular localization of a lncRNA can be Results informative. Frequently, these RNAs are localized in the nucleus, in

TLR4 signaling regulates expression of lncRNAs which they alter the expression of protein-coding genes by by guest on September 27, 2021 The macrophage transcriptome is dynamically regulated in re- controlling gene transcription. There is also evidence that sponse to microbial and endogenous stimuli (12, 25). LPS, which is lncRNAs can be localized and function in the cytosol and act as recognized by TLR4/MD2, leads to the induction of a transcrip- sponges for microRNAs (4, 5). Thus, we tested the subcellular tional program through NF-kB, IRFs, and STAT proteins. Previous location of the 1810058I24Rik transcript by generating cyto- work from our laboratory and others has demonstrated that TLRs solic and nuclear extracts, purifying RNA and performing also regulate the expression of lncRNAs, including both anti- qRT-PCR. Malat1, a well-characterized lncRNA that functions sense lncRNAs and lincRNAs. For example, lincRNAs such as in the nucleus, was detected in the nuclear fraction (28). In lincRNA-Cox2 (Ptgs2os2) and lincRNA-EPS are regulated in re- contrast, the 1810058I24Rik transcript was predominantly sponse to LPS and exhibit broad regulatory effects by controlling cytosolic (Fig. 3A). This was similar to the localization of the inflammatory gene expression (26, 27). These lncRNAs were mature GAPDH mRNA, a protein-coding gene that would be identified following transcriptional analysis of mouse BMDM translated in the cytosol. Given the cytosolic location of the stimulated with LPS using RNA sequencing (Fig. 1A). To further 1810058I24Rik transcript, we next evaluated the potential examine the role of lncRNAs in macrophages, in this study, we protein-coding capacity of this putative lncRNA because many generated a custom PCR array for an in-depth investigation of the lncRNAs are misannotated and may encode functional proteins or expression of 72 lncRNAs, including genes that were TLR4 reg- sORFs. The National Center for Biotechnology Information ORF- ulated (log2 fold change .2or,22) or abundantly expressed in finder program revealed three putative ORFs in 1810058I24Rik, BMDMs (26, 27). A heatmap showing the expression of this set of which included a 47-aa ORF that shows high conservation across 72 lncRNAs in mouse macrophages treated with LPS (TLR4), diverse species from Caenorhabditis elegans to humans (Fig. 3B). Pam3CSK4 (TLR2), Poly I:C (TLR3), CL097 (TLR7), or the In light of this observation, we wanted to address the possibility that cytokines TNF-a and type I IFN is shown (Fig. 1B). Consistent 1810058I24Rik may be a protein-coding gene. To test this, we with our previous studies, the expression of lincRNA-Cox2 was performed in vitro translation assays using 35S-labeled methi- induced by all of these ligands, whereas lincRNA-EPS was onine, which confirmed that the 1810058I24Rik RNA was in- downregulated in cells exposed to all of these stimuli. In ad- deed translated into ∼5.1 kDa peptide (Fig. 3C, Supplemental dition, among the downregulated lncRNAs was a gene that was Fig. 2A). We also generated a mutant with a frameshift mutation previously uncharacterized, called 1810058I24Rik. This tran- inserted immediately after the start site and found that this script was downregulated in response to multiple ligands mutation ablated translation. Next, we cloned the putative ORF (Fig. 1B). We confirmed the downregulation of this lncRNA in into a cDNA expression vector with a Flag tag at either the N response to LPS as well as other TLR ligands, as shown by terminus, C terminus, or both. We could detect Flag-tagged qRT-PCR (Fig. 1C). The RNA virus Sendai virus did not lead translated 1810058I24Rik only with C-terminal tags in MEFs to a downregulation of this RNA. and 293T human cells (Fig. 3D, Supplemental Fig. 2B, 2C). 4 INFLAMMASOME REGULATION BY sORF-ENCODED PEPTIDES

FIGURE 1. Identification of Mm47 as a putative lincRNA regulated in Downloaded from the immune response. (A) Schematic detailing pipeline for identification of immune-regulated lincRNAs. (B) Quantitative profiling of candidate lincRNAs in response to various PAMPs and cytokines. Cells were

treated with LPS (100 ng/ml), Poly(I:C) http://www.jimmunol.org/ (25 mg/ml), Pam3Csk4 (100 nM), CLO97 (200 ng/ml), TNF-a (10 ng/ml), and IFN-a/b (500 U/ml). Upregulated genes are in red, and downregulated genes are in blue. Mm47 is highlighted in green. (C) qRT-PCR of Mm47 relative to Gapdh in response to bacterial and viral PAMPs. Cells were treated with Pam3Csk4 (100 nM), Poly(I:C) (25 mg/ml), LPS (100 ng/ml), CLO97 by guest on September 27, 2021 (200 ng/ml), CpGB (2 mg), and Sendai virus (400 hemagglutination unit). Data represent mean 6 SEM from three experiments. *p , 0.03, **p , 0.002, ***p , 0.001; ns, not signiﬁcant. (D) qRT-PCR showing expression level of Mm47 in wild type B6 mice tissues compared with housekeeping gene Gapdh. Data represent mean 6 SEM from three mice.

SignalP protein domain prediction suggested that the N terminus of 1810058I24Rik SEP. Finally, to formally deﬁne the presence 1810058I24Rik contained a potential signal sequence. The addition of this peptide in cells, we generated an Ab to detect the en- of the N-terminal tag may not be tolerable, causing degradation of dogenous 1810058I24Rik SEP. Consistent with our expression The Journal of Immunology 5

FIGURE 2. Mm47 transcript is downregulated in activated macrophage in Tlr4/MyD88-dependent manner. (A) qRT-PCR showing the copy number of mature transcripts of Mm47 in BMDM cells with and without LPS treatment. (B) qRT-PCR of Mm47 transcript with LPS treatment compared with Gapdh. (C) qRT-PCR showing Mm47 transcript downregulation in BMDM cells in response to 100 mg/ml LPS. (A–C) Data represent mean 6 SEM of three experiments. *p , 0.03, **p , 0.002; ns, not signiﬁcant.

data, endogenous 1810058I24Rik SEP was expressed in BMDM. separate pure cytosolic, ER, and mitochondrial fractions (33). Using The levels of this peptide decreased in response to LPS. the mitochondrial membrane protein Tom20 as a positive control, 1810058I24Rik SEP was abundant in macrophages and was de- Western blot analysis confirmed that the Mm47 peptide is pre- creased after 24 h post-LPS treatment (Fig. 3E). dominantly localized on the mitochondria (Fig. 4C). Further sub- We next aimed to further understand where the 1810058I24Rik fractionation of the mitochondrial fraction revealed that Mm47 is SEP was localized in cells. The SignalP program suggested that highly enriched on the outer membrane and intermembrane fraction Downloaded from 1810058I24Rik may be localized to the mitochondria, despite the of the mitochondrion, as indicated by the corresponding localization absence of a canonical mitochondrial targeting signal (29). A of VDAC, an outer membrane protein (Fig. 4D). Overall, these partial protein sequence of 1810058I24Rik had previously been studies indicate that the 1810058I24Rik transcript encodes a peptide reported in bovine heart and zebrafish mitochondria (30, 31). that is localized to the mitochondrion. To determine if the 1810058I24Rik peptide localized at the b mitochondria, we performed immunostaining on FLAG-tagged Mm47 is required for Nlrp3-dependent IL-1 maturation 1810058I24Rik SEP ectopically expressed in MEFs. This analy- We next examined the possibility that Mm47 played a role in http://www.jimmunol.org/ sis revealed that the 1810058I24Rik peptide colocalized with the regulating the magnitude or duration of the LPS response in mitochondrial marker MitoTracker Deep Red (Fig. 4A). Given macrophages. Using CRISPR–Cas9, we created Mm47 KO im- this localization, we propose to rename 1810058I24Rik SEP as mortalized mouse macrophages. Western blot analysis of BMDMs mitochondrial Mm47. We next confirmed the localization of en- expressing an NTC guide RNA and three guide RNAs targeting dogenous Mm47 in macrophages. Abs against the endogenous Mm47 confirmed KO of Mm47 in the three cell lines examined Mm47 again revealed mitochondrial localization in mouse BMDMs (Fig. 5A). We then monitored LPS-dependent transcriptional re- (Fig. 4B). The ER and mitochondria are known to intimately as- sponses in these cells using NanoString technology to simulta-

sociate, which may affect the immunoﬂuorescence imaging (32). neously detect the expression of 100 immune system–related by guest on September 27, 2021 Therefore, we employed sucrose gradient ultracentrifugation to genes. As shown in Fig. 5B, LPS treatment led to an increase in

FIGURE 3. Mm47 encodes for a stable micropeptide. (A) Subcellular fraction and qRT-PCR of Malat1, Gapdh, and Mm47 to test the localization of Mm47 transcript. Data represent mean 6 SEM of three independent experiments. (B) Mm47 micropeptide sequence alignment across species of different phyla. “*”, fully conserved residue; “:”, highly conserved residue and .0.5 Gonnet Pam 250 matrix; “.”, weakly conserved residue and ,0.5 Gonnet Pam 250 matrix. (C) In vitro translation assay using 35S-labeled methionine and SDS-PAGE electrophoresis showing 5.1-kDa band of Mm47 micropeptide in wild type (WT) sequence. Representative immunoblot of three independent experiments is shown. (D) Expression of Flag-tagged Mm47 micropeptide in MEFs and immunoblotting for Flag tag. Representative immunoblot of three independent experiments is shown. (E) Western blot of endogenous Mm47 micropeptide with LPS treatment. Representative immunoblot of three independent experiments is shown. 6 INFLAMMASOME REGULATION BY sORF-ENCODED PEPTIDES

a role in controlling the activation of the Nlrp3 inflammasome. Immortalized BMDM were primed with LPS for 3 h to upregulate IL-1b and Nlrp3 itself, followed by stimulation with nigericin, a potassium ionophore and bacterial pore-forming toxin that leads to the formation of the Nlrp3 inflammasome complex. LPS priming and nigericin stimulation led to robust secretion of IL-1b as measured by ELISA. This response was significantly abrogatedincellslackingMm47(Fig.5C).ThisMm47- dependent release of IL-1b was also abrogated in cells stimulated with another Nlrp3 activator, ATP (Fig. 5D). Indeed, this effect of Mm47 was specific for the Nlrp3 inflammasome as Mm47-deficient cells had comparable IL-1b release when cells were stimulated with poly(deoxyadenylic-deoxythymidylic) [Poly(dA:dT)], a dsDNA mimetic that activates the Aim2 inflammasome (Fig. 5E) or when cells were infected with S. typhimurium, a bacterial pathogen that activates the Nlrc4 inflammasome (Fig. 5F). We also performed rescue experiments by restoring Mm47 levels in the KO cells. This was achieved by

ectopically expressing Mm47 in a form that was resistant to Downloaded from Cas9-directed cleavage. This approach led to restoration of Mm47 at levels observed in wild type cells (Fig. 5G). These restored cells were fully competent for LPS/nigericin-induced secretion of IL-1b (Fig. 5H, Supplemental Fig. 3A). Together, these observations indicate that Mm47 controls the activation of

the Nlrp3 inflammasome in a highly specific manner. http://www.jimmunol.org/ Because mitochondrial reactive oxygen species (mtROS) contributes to the priming of macrophages via Hif1a-mediated transcription of pro–IL-1b as well as activation of the Nlrp3 inflammasome, we tested the levels of mtROS in cells (34–36). We did not observe a significant increase in mtROS in LPS-treated cells, and there was no difference between Mm47-deficient and wild type cells (Supplemental Fig. 3B). Similarly, alteration in mitochondrial electron transport chain (ETC) function affects ATP levels, which in turn alter Nlrp3 activation (37). Therefore, we by guest on September 27, 2021 tested the mitochondrial oxygen consumption rate (OCR) as a proxy for mitochondrial ATP. We notice that the Mm47-deficient cells had reduced basal OCR. However, after the addition of chemical inhibitors of ETC, we observed similar levels of OCR in both wild type and Mm47 KO cells with reduced maximal OCR in FIGURE 4. Mm47 micropeptide is localized on the mitochondria. (A) LPS-exposed cells (Supplemental Fig. 3C). We conclude from Ectopic expression of Flag-tagged Mm47 in MEF followed by immuno- these studies that Mm47 does not contribute to the generation of fluorescence imaging of Flag-Mm47 (green), MitoTracker Deep Red (red) ATP during mitochondrial ETC stress. for mitochondria, and DAPI (blue) for nucleus. Representative images To further corroborate the finding that Mm47 impacts Nlrp3 from three independent experiments are shown. (B) Immunofluorescence inflammasome-mediated IL-1b production, we employed an in- imaging of endogenous Mm47, MitoTracker Deep Red for mitochondria, dependent siRNA approach in primary BMDM. BMDM were and DAPI for nucleus. Representative images from three independent transfected with two nontargeting siRNAs or with three Mm47- C experiments are shown. ( ) Subcellular fraction of the cytosolic, ER, and targeting siRNAs. Analysis of LPS and nigericin responses in mitochondrial fraction followed by immunoblot for endogenous Mm47, these cells showed that siRNAs targeting Mm47 led to a signifi- mitochondrial outer membrane protein Tom20, ER membrane protein cant reduction in IL-1b in response to nigericin treatment KDEL, and cytosolic protein Gapdh. Representative immunoblot of three b independent experiments is shown. (D) Suborganelle fractionation of the (Fig. 6A). Meanwhile, release of IL-1 was comparable between outer and inner mitochondrial compartment followed by immunoblot for nontargeting siRNA and Mm47-targeting siRNA-transfected cells endogenous Mm47, outer mitochondrial membrane protein VDAC, and when cells were stimulated with dsDNA (poly (dA:dT)) or in- mitochondrial matrix protein HSp60. fected with S. typhimurium (Fig. 6B, 6C). Immunoblotting of cell lysates and supernatants revealed that Mm47-targeting siRNAs led to reduced processing of caspase-1 p20 and mature IL-1b the expression of a broad panel of immune genes in the control (p17) (Fig. 6D). We also measured mRNA levels of pro–IL-1b lines after 2 and 6 h of LPS. These responses were largely intact in or Nlrp3 itself in Mm47–siRNA–targeted cells, and in both macrophages lacking Mm47. These results suggest that Mm47 cases, the responses were intact (Fig. 6E, Supplemental Fig. does not impact the transcriptional response induced in macro- 4A). Furthermore, the Mm47–siRNA–targeted cells displayed phages following LPS stimulation. reduced levels of processed Gsdmd compared with controls. Given the localization of Mm47 on the mitochondria and the This effect on Gsdmd, however, did not have a significant literature that reveals how damaged or defective mitochondria impact on cell death (Supplemental Fig. 4B, 4C). Collectively, contribute to the activation of inflammation through the engage- these findings reveal that expression of Mm47 is essential to ment of Nlrp3, we next examined the possibility that Mm47 played facilitate the activation of the Nlrp3 inflammasome and the The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 5. Mm47-deficient immortalized BMDM have impaired Nlrp3 pathway. (A) Western blot for endogenous Mm47 after CRISPR–Cas9–mediated KO. (B) Heatmap of inflammatory genes tested using NanoString code set to compare the response of NTC cells and Mm47 KO cells to 100 ng/ml LPS treatment for 2 and 6 h. Transcription change shown as log2 fold change compared with NTC. (C) ELISA for secreted IL-1b in nontargeted and CRISPR– cas9–mediated KO of Mm47 in BMDM after priming with 100 ng/ml LPS and stimulation with 5 mM nigericin. (D) ELISA for secreted IL-1b in nontargeted and CRISPR–cas9–mediated KO of Mm47 in BMDM after priming with 100 ng/ml LPS and stimulation with 5 mMATP.(E) ELISA to test AIM2 receptor–mediated secretion of IL-1b in BMDM with or without Mm47 KO upon 100 ng/ml LPS priming and 2 mg Poly(dA:dT) transfection. (F) ELISA to test NLRC4 receptor–mediated IL-1b secretion in primary BMDM with or without Mm47 KO upon S. typhimurium infection for 6 h. (G) Western blot of Mm47 to test restoration of CRISPR–cas9–resistant Mm47 expression using lentiviral transduction. The control cells were transduced with empty vector (EV), and rescued cells (R) were transduced with CRISPR–cas9–resistant Mm47. (H) ELISA testing secreted IL-1b expression in control and CRISPR–cas9–resistant Mm47-reconstituted cells upon priming with 100 ng/ml LPS and stimulation with 5 mM nigericin. (A–F) Data represent mean 6 SEM of three experiments. **p , 0.002, ***p , 0.001; ns, not significant. 8 INFLAMMASOME REGULATION BY sORF-ENCODED PEPTIDES Downloaded from http://www.jimmunol.org/

FIGURE 6. Mm47-deficient BMDM have impaired Nlrp3 pathway. (A) ELISA testing the secreted IL-1b levels upon priming with 100 ng/ml LPS and treatment with 5 mM nigericin in nontargeted or siRNA-mediated knock down of Mm47 in primary BMDM. Two nontargeting lines (NT1 and NT2) were used as a control, and three independent siRNAs targeting Mm47 were used. (B) ELISA testing AIM2 receptor–mediated secretion of IL-1b in primary BMDM with or without Mm47 knock down upon LPS priming followed by 2 mg Poly(dA:dT) transfection. (C) ELISA testing NLRC4 receptor–mediated by guest on September 27, 2021 IL-1b secretion in primary BMDM with or without Mm47 knock down upon S. typhimurium infection for 6 h. (D) Mm47 was knocked down using siRNA or nontargeting siRNA in primary BMDM cells. Cells were challenged as indicated, and Western bot was performed to procaspase-1, cleaved caspase-1, pro–IL-1b, and cleaved IL-1b. Representative images from three independent experiments are shown. (E) qRT-PCR showing mRNA levels of IL-1b and Nlrp3 after priming primary BMDM with 100 ng/ml LPS for 2 h with or without Mm47 knock down. (A–D) Data represent mean 6 SEM of three experiments. *p , 0.03, ***p , 0.001; ns, not significant. release of IL-1b and Gsdmd processing with a modest impact which were predicted to be translated from non-AUG start sites (8). on cell death. This study, among others, underscores the need for independently authenticating the coding potential of cytosolic lncRNAs and Discussion determining to what extent these RNAs produce peptides. An accumulating body of evidence indicates that a large number In a surprising twist on our initial focus on lncRNAs, in this of genes are misannotated as lncRNAs, many of which produce study, we have identified Mm47, a highly conserved 47 aa–long small proteins with important biological functions (38). Studies on mitochondrial micropeptide, which is produced from an annotated the functions of these SEPs are currently limited because in-depth lncRNA. Further, we provide clear genetic evidence demonstrat- experimental investigation of individual genes is required to un- ing that Mm47 regulates Nlrp3 inflammasome function in mac- cover their functions. Recent computational and genomic ad- rophages. Mitochondria provide a signaling hub to integrate vances have allowed the identification and mapping of SEPs (7). metabolic and inflammatory capability of immune cells. Such Several recent studies using ribosome profiling and mass spec- integration is provided through innate immune receptors, includ- trometry–based detection of peptides have revealed that many ing TLRs and the Nlrp3 inflammasome, which respond to altered genes currently annotated as lncRNAs produce small proteins mitochondrial function and/or mitochondrial-derived molecules (7, 38). These include ORFs that are translated from canonical released from damaged organelles. In response to stress, mito- AUG or noncanonical start sites (6, 7, 9). In macrophages, a novel chondria release molecules, such as cardiolipin and mitochondrial gene AW112010 was recently identified as a SEP translated from DNA, both of which are sensed by the Nlrp3 inflammasome as a noncanonical start site of an annotated lncRNA gene. AW112010 well as by other innate sensors such as cyclic GMP–AMP syn- was shown to be regulated in innate cells such as macrophages thase (22, 39, 40). Mm47-deficient cells exhibit impaired Nlrp3- following microbial infection, and genetic loss of function models mediated inflammasome responses. This defect was remarkably in mice showed this SEP was important in mucosal immunity by specific for Nlrp3, as neither Aim2 nor Nlrc4 inflammasomes were controlling S. typhimurium infection and altering the susceptibility impacted by the loss of Mm47. Importantly, genetic comple- to colitis in mice. The same study also identified three ORFs of mentation of the Mm47 coding DNA sequence in Mm47 KO cells the 1810058I24Rik gene in their ribosome profiling studies, two of restored Nlrp3-mediated inflammasome responses. This indicates The Journal of Immunology 9 that the loss of Nlrp3 activity in the KOs is truly due to Mm47 and 2. Wang, K. C., and H. Y. Chang. 2011. Molecular mechanisms of long noncoding RNAs. Mol. Cell 43: 904–914. not an additional impairment in these cells. Further, the ability of 3. Geisler, S., and J. Coller. 2013. RNA in unexpected places: long non-coding the micropeptide ORF to rescue inflammasome defects also sup- RNA functions in diverse cellular contexts. Nat. Rev. Mol. Cell Biol. 14: ports the importance of the peptide product in this biological 699–712. 4. Atianand, M. K., D. R. Caffrey, and K. A. Fitzgerald. 2017. Immunobiology of context. long noncoding RNAs. Annu. Rev. Immunol. 35: 177–198. Exactly how Mm47 on the mitochondrion impacts the activation 5. Agliano, F., V. A. Rathinam, A. E. Medvedev, S. K. Vanaja, and A. T. Vella. of the Nlrp3 inflammasome is still an open question. Our results 2019. Long noncoding RNAs in host-pathogen interactions. Trends Immunol. 40: 492–510. indicate that Mm47 does not act at the level of inflammasome 6. Ingolia, N. T., L. F. Lareau, and J. S. Weissman. 2011. Ribosome profiling of priming because LPS-driven expression of pro–IL-1b and of Nlrp3 mouse embryonic stem cells reveals the complexity and dynamics of mammalian were intact in Mm47-deficient cells as was the broader transcrip- proteomes. Cell 147: 789–802. 7. Makarewich, C. A., and E. N. Olson. 2017. Mining for micropeptides. Trends tional program activated downstream of TLR4. Mm47 appears to Cell Biol. 27: 685–696. act at the level of inflammasome licensing or activation, although 8. Jackson, R., L. Kroehling, A. Khitun, W. Bailis, A. Jarret, A. G. York, how this occurs is not clear from our work. Activation of O. M. Khan, J. R. Brewer, M. H. Skadow, C. Duizer, et al. 2018. The translation of non-canonical open reading frames controls mucosal immunity. Nature 564: caspase-1 as measured by monitoring the formation of the p20 434–438. fragment and generation of the IL-1b p17 was reduced in cell 9. Andrews, S. J., and J. A. Rothnagel. 2014. Emerging evidence for functional lacking Mm47. Caspase-1 cleaves pro–IL-1b and Gsdmd in peptides encoded by short open reading frames. [Published erratum appears in b 2014 Nat. Rev. Genet. 15: 286.] Nat. Rev. Genet. 15: 193–204. canonical inflammasome activation (18). The levels of IL-1 10. Guo, B., D. Zhai, E. Cabezas, K. Welsh, S. Nouraini, A. C. Satterthwait, and p17 and Gsdmd p30 were both reduced in Mm47-targeted cells; J. C. Reed. 2003. Humanin peptide suppresses apoptosis by interfering with Bax however, the kinetics of cell death was similar to nontargeted activation. Nature 423: 456–461. Downloaded from 11. Nelson, B. R., C. A. Makarewich, D. M. Anderson, B. R. Winders, cells. Despite the reduction in Gsdmd processing, Mm47- C. D. Troupes, F. Wu, A. L. Reese, J. R. McAnally, X. Chen, E. T. Kavalali, et al. targeted cells still underwent pyroptosis. Previous studies have 2016. A peptide encoded by a transcript annotated as long noncoding RNA shown that although Gsdmd is a major executioner of cell death enhances SERCA activity in muscle. Science 351: 271–275. 12. Satoh, T., and S. Akira. 2016. Toll-like receptor signaling and its inducible in the Nlrp3 pathway, Gsdmd-independent cell death pathways proteins. Microbiol. Spectr. 4: DOI: 10.1128/microbiolspec.MCHD-0040-2016. are also involved (18). It is possible in Mm47-targeted cells that 13. Takeda, K., and S. Akira. 2015. Toll-like receptors. Curr. Protoc. Immunol. 109: 14.12.1–14.12.10. caspase-8 activation leads to the cell death observed (41, 42). 14. Bauernfeind, F. G., G. Horvath, A. Stutz, E. S. Alnemri, K. MacDonald, http://www.jimmunol.org/ Although there has been considerable progress in understanding D. Speert, T. Fernandes-Alnemri, J. Wu, B. G. Monks, K. A. Fitzgerald, et al. inflammasome biology, particularly the importance of Nlrp3 in 2009. Cutting edge: NF-kappaB activating pattern recognition and cytokine receptors license NLRP3 inflammasome activation by regulating NLRP3 expres- a diverse range of cellular processes and diseases, the precise sion. J. Immunol. 183: 787–791. mechanisms that coordinate Nlrp3 inflammasome complex 15. Rathinam, V. A., and K. A. Fitzgerald. 2016. Inflammasome complexes: formation are very poorly understood. A unifying mechanism of emerging mechanisms and effector functions. Cell 165: 792–800. 16. Broz, P., and V. M. Dixit. 2016. Inflammasomes: mechanism of assembly, reg- how Nlrp3 is activated is not known, and therefore, it is difficult ulation and signalling. Nat. Rev. Immunol. 16: 407–420. to mechanistically define how Mm47 alters these processes. We 17. Lamkanfi, M., and V. M. Dixit. 2014. Mechanisms and functions of inflamma- also do not understand the impact of Mm47 deficiency on nor- somes. Cell 157: 1013–1022. 18. Kayagaki, N., I. B. Stowe, B. L. Lee, K. O’Rourke, K. Anderson, S. Warming, mal mitochondrial function. When we overexpress Mm47 in T. Cuellar, B. Haley, M. Roose-Girma, Q. T. Phung, et al. 2015. Caspase-11 cleaves by guest on September 27, 2021 murine embryonic fibroblasts, the morphology of the mito- gasdermin D for non-canonical inflammasome signalling. Nature 526: 666–671. 19. Shi, J., W. Gao, and F. Shao. 2017. Pyroptosis: gasdermin-mediated programmed chondria was altered. This could indicate that Mm47 has a role necrotic cell death. Trends Biochem. Sci. 42: 245–254. in mitochondrial fission during stress that may directly or in- 20. Shi, J., Y. Zhao, K. Wang, X. Shi, Y. Wang, H. Huang, Y. Zhuang, T. Cai, directly impact leakage of mitochondrial DNA or cardiolipins F. Wang, and F. Shao. 2015. Cleavage of GSDMD by inflammatory caspases determines pyroptotic cell death. Nature 526: 660–665. into the cytosol. 21. Latz, E., T. S. Xiao, and A. Stutz. 2013. Activation and regulation of the Although we do not understand how Mm47 interfaces with the inflammasomes. Nat. Rev. Immunol. 13: 397–411. inflammasome, we did find that after exposure to LPS, both the 22. Rongvaux, A. 2018. Innate immunity and tolerance toward mitochondria. Mi- tochondrion 41: 14–20. 1810058I24Rik gene and the Mm47 polypeptide are reduced. At 23. Roberson, S. M., and W. S. Walker. 1988. Immortalization of cloned mouse the protein level, the decrease in expression of Mm47 was observed splenic macrophages with a retrovirus containing the v-raf/mil and v-myc on- cogenes. Cell. Immunol. 116: 341–351. after prolonged exposure (24–48 h) of cells to LPS. It is possible 24. Bozidis, P., C. D. Williamson, and A. M. Colberg-Poley. 2007. Isolation of that this temporal downregulation of Mm47 protein removes this endoplasmic reticulum, mitochondria, and mitochondria-associated membrane essential factor on the mitochondrion and could represent a mech- fractions from transfected cells and from human cytomegalovirus-infected primary fibroblasts. Curr. Protoc. Cell Biol. Chapter 3: Unit 3.27. anism to curb inflammasome activation to resolve inflammation. 25. Medzhitov, R., and T. Horng. 2009. Transcriptional control of the inflammatory response. Nat. Rev. Immunol. 9: 692–703. 26. Carpenter, S., D. Aiello, M. K. Atianand, E. P. Ricci, P. Gandhi, L. L. Hall, Acknowledgments M. Byron, B. Monks, M. Henry-Bezy, J. B. Lawrence, et al. 2013. A long We thank the members of the Fitzgerald laboratory for suggestions and noncoding RNA mediates both activation and repression of immune response genes. Science 341: 789–792. comments, especially Fiachra Humphries and Shiuli Agarwal. We thank 27. Atianand, M. K., W. Hu, A. T. Satpathy, Y. Shen, E. P. Ricci, J. R. Alvarez- Cole Haynes (University of Massachusetts Medical School, Worcester, Dominguez, A. Bhatta, S. A. Schattgen, J. D. McGowan, J. Blin, et al. 2016. A MA) for guidance on mitochondrial biology. We also thank Richard long noncoding RNA lincRNA-EPS acts as a transcriptional brake to restrain Kandasamy (Norwegian University of Science and Technology, Trondheim, inflammation. Cell 165: 1672–1685. 28. Tripathi, V., J. D. Ellis, Z. Shen, D. Y. Song, Q. Pan, A. T. Watt, S. M. Freier, Norway) for helping with manuscript editing. C. F. Bennett, A. Sharma, P. A. Bubulya, et al. 2010. The nuclear-retained noncoding RNA MALAT1 regulates alternative splicing by modulating SR splicing factor phosphorylation. Mol. Cell 39: 925–938. Disclosures 29. Fukasawa, Y., J. Tsuji, S. C. Fu, K. Tomii, P. Horton, and K. Imai. 2015. The authors have no financial conflicts of interest. MitoFates: improved prediction of mitochondrial targeting sequences and their cleavage sites. Mol. Cell. Proteomics 14: 1113–1126. 30. Zhang, D., Y. Xi, M. L. Coccimiglio, J. A. Mennigen, M. G. Jonz, M. Ekker, and V. L. Trudeau. 2012. Functional prediction and physiological characterization of References a novel short trans-membrane protein 1 as a subunit of mitochondrial respiratory 1. Derrien, T., R. Johnson, G. Bussotti, A. Tanzer, S. Djebali, H. Tilgner, complexes. Physiol. Genomics 44: 1133–1140. G. Guernec, D. Martin, A. Merkel, D. G. Knowles, et al. 2012. The GENCODE 31. Carroll, J., M. C. Altman, I. M. Fearnley, and J. E. Walker. 2007. Identification of v7 catalog of human long noncoding RNAs: analysis of their gene structure, membrane proteins by tandem mass spectrometry of protein ions. Proc. Natl. evolution, and expression. Genome Res. 22: 1775–1789. Acad. Sci. USA 104: 14330–14335. 10 INFLAMMASOME REGULATION BY sORF-ENCODED PEPTIDES

32. Rowland, A. A., and G. K. Voeltz. 2012. Endoplasmic reticulum-mitochondria and D. Sancho. 2016. Mitochondrial respiratory-chain adaptations in mac- contacts: function of the junction. Nat. Rev. Mol. Cell Biol. 13: 607–625. rophages contribute to antibacterial host defense. Nat. Immunol. 17: 1037– 33. Williamson, C. D., D. S. Wong, P. Bozidis, A. Zhang, and A. M. Colberg-Poley. 1045. 2015. Isolation of endoplasmic reticulum, mitochondria, and mitochondria- 38. Pueyo, J. I., E. G. Magny, and J. P. Couso. 2016. New peptides under the s(ORF) associated membrane and detergent resistant membrane fractions from trans- ace of the genome. Trends Biochem. Sci. 41: 665–678. fected cells and from human cytomegalovirus-infected primary fibroblasts. Curr. 39. Zhong, Z., S. Liang, E. Sanchez-Lopez, F. He, S. Shalapour, X. J. Lin, J. Wong, Protoc. Cell Biol. 68: 3.27.1–3.27.33. S. Ding, E. Seki, B. Schnabl, et al. 2018. New mitochondrial DNA synthesis 34. Zhou, R., A. S. Yazdi, P. Menu, and J. Tschopp. 2011. A role for mitochondria in enables NLRP3 inflammasome activation. Nature 560: 198–203. NLRP3 inflammasome activation. [Published erratum appears in 2011 Nature 40. Iyer, S. S., Q. He, J. R. Janczy, E. I. Elliott, Z. Zhong, A. K. Olivier, 475: 122.] Nature 469: 221–225. J. J. Sadler, V. Knepper-Adrian, R. Han, L. Qiao, et al. 2013. Mitochondrial 35. Mills, E. L., B. Kelly, A. Logan, A. S. H. Costa, M. Varma, C. E. Bryant, cardiolipin is required for Nlrp3 inflammasome activation. Immunity 39: P. Tourlomousis, J. H. M. Da¨britz, E. Gottlieb, I. Latorre, et al. 2016. Succinate 311–323. dehydrogenase supports metabolic repurposing of mitochondria to drive in- 41. Lee, B. L., K. M. Mirrashidi, I. B. Stowe, S. K. Kummerfeld, C. Watanabe, flammatory macrophages. Cell 167: 457–470.e13. B. Haley, T. L. Cuellar, M. Reichelt, and N. Kayagaki. 2018. ASC- and caspase- 36. Herb, M., A. Gluschko, K. Wiegmann, A. Farid, A. Wolf, O. Utermo¨hlen, 8-dependent apoptotic pathway diverges from the NLRC4 inflammasome in O. Krut, M. Kro¨nke, and M. Schramm. 2019. Mitochondrial reactive oxygen macrophages. Sci. Rep. 8: 3788. species enable proinflammatory signaling through disulfide linkage of NEMO. 42. Schneider, K. S., C. J. Groß, R. F. Dreier, B. S. Saller, R. Mishra, O. Gorka, Sci. Signal. 12: eaar5926. R. Heilig, E. Meunier, M. S. Dick, T. Ćikovic´, et al. 2017. The inflammasome 37. Garaude, J., R. Acıń-Pe´rez, S. Martıńez-Cano, M. Enamorado, M. Ugolini, drives GSDMD-independent secondary pyroptosis and IL-1 release in the ab- E. Nistal-Villań, S. Herva´s-Stubbs, P. Pelegrıń, L. E. Sander, J. A. Enrı´quez, sence of caspase-1 protease activity. Cell Rep. 21: 3846–3859. Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021 Figure 1S:

A. 1810058I24Rik genomic locus E1 E2 E3 (8.9kb) on chromosome 6

1810058I24Rik RNA Transcript (580bp) siRNA1 siRNA2 siRNA3 (77-97bp) (133-153bp) (481-501bp)

B. ACCTCTCACACCCTCCTCGCCCTCCCCACCGACATCATGCTCCAGTTCCTGCTTGGATTTACTTTGGGCAACGTGGTTGGAATGTATCTGGCTCAGAACTATGAAATGCCAAAC ORF Exon 1 ORF Exon 2 Forward primer 1 siRNA1 target Reverse primer 1 sgRNA1 target Forward primer 2 sgRNA2 target

CTGGCTAAAAAACTTGAAGAGATTAAGAAGGACCTGGAAGCCAAGAAGAAGCCCCCTAGTTCCTGATGCCTGCCTGGCACTGCAATCTGGACCCACCAGTTCCATCCTGGGG ORF exon 3 siRNA2 target

GGCCTGTCTCCTTCACAGCCACGTCCAAAGCTGTGTTCCTCTTTGGTCTCAGACCCTGCGCGTTGCTGAGGCTCCAGCACGGGCCAGAAGTGGAGTTAGCTTTCCGTTTCCAG

CATCTCCCCGTTCCCGTATCTTCCGTCCACCTGCTTCTGAGACCTAAGCCTGGAACCATGGTGCTAGGTTAGCAGCCGCAACCATGACAGGACTCACTCGCCTGTGCTCTGGG Reverse primer 2

ATTCCTACCGCCTACTGTCAAGAAATGAATGGATTTGGATAGCTGCTGGGGACTCACTGTCTAGCAGTGATAACAGGAGCTTGCTAGACAAACTAAACTGTGTTAAAGTCATT siRNA3 target

AAAGTCAGTATCTTC

C. D. E. 1200 1.2

40 900 0.9 y = -1.035ln(x) + 35.347 Ct 30 R² = 0.91975 600 0.6

20 (LPS/NT )

10 300 0.3 1810058I24Rik copy number/ cel l

0 1810058I24Rik mRNA Fold change 0.0 0 2 109 4 109 6 109 LPS 0 100ng/mL NT 4hr NT 2hr 6hr 24hr RNA copy number

Figure 1S: a) Schematic diagram showing Mm47 genomic locus, and mature RNA transcript along with qPCR primers and siRNA target sites. The translating ORF is highlighted in red. b) Mature RNA transcript sequence of Mm47 showing ORF, siRNA target site, single-guide RNA target sites and qPCR primer sites. c) Standard curve for copy number analysis of Mm47. d) qRT-PCR in LPS treated human monocytes for copy number of Mm47 transcript. Data represents mean±SEM of two experiments. e) qRT-PCR in LPS treated human dendritic cells for fold change in Mm47 transcript level. Data represents mean ±SEM of two experiments. Figure 2S B. A. E1 E2 E3 1810058I24Rik T7 T7 Flag 1810058I24Rik ORF locus 5’ UTR 3’ UTR ATG

T7 T7 1810058I24Rik 5’ UTR 3’ UTR Flag 1810058I24Rik ORF Flag RNA 1810058I24Rik ATG ATG

1810058I24Rik T7 T7 mutant 5’ UTR 1810058I24Rik 3’ UTR 1810058I24Rik ORF Flag FS ATG

C. -Flag WT (neg) WT -Flag 1810058I24Rik Flag- 1810058I24Rik Flag- 1810058I24Rik 1810058I24Rik

~5-6 kDa β-actin

Figure 2S: a) Schematic for wildtype and frame-shift mutant of Mm47 transcript used for in vitro translation assay. One nucleotide base was inserted immediately after the start site to create a frameshift. b) Schematic for ectopically expressed Mm47 with C-terminal and Nterminal Flag-tag from an expression vector with a T7 promoter. c) Expression of Flag-tagged Mm47 in 293T cells followed by immunoblot for Flag-tag. Representative immunoblot of three independent experiment is shown. Figure 3S: A...... ACCGACATCATGCTCCAGTTCCTGCTTGGATTTACTTTGGGCAACGTGGTTGGAATGTATCTGG CTCAGAACTATGAAATGCCAAACCTGGCTAAAA ORF Exon 1 ORF Exon 2 sgRNA1 target PAM sgRNA2 target PAM

AACTTGAAGAGATTAAGAAGGACCTGGAAGCCAAGAAGAAGCCCCCTAGTTCCTGATGCCTGCCTGGCACTG...... ORF exon 3

B. C. NTC Mm47 KO NT NTC LPS Mm47 KO LPS NT 500 LPS 6h Oligo FCCP Ant/Rot 2.0 ns 400

1.5 300

1.0 200

mtROS fold change 0.5 OCR (pmol/min ) 100

0.0 0 NTC Mm47 KO 0 50 100 Time (minutes)

Figure 3S: a) Schematic diagram showing PAM (changed from TGG to TAG) in ectopically expressed Mm47 to rescue the expression in CRISPR-mediated knock out cells. b) Relative mtROS amounts in non- and LPS-primed wildtype and Mm47-KO BMDM cells. MitoSox reagent (ThermoFisher M36008) was added to cells at 5 uM for 30 minutes before FACs. Cells were gated for live cells for mean fluorescence intensity, and fold change was calculated against untreated control. Data are mean ±SEM of three independent experiments. c) Real-time Oxygen consumption rate of NTC and CRISPR-cas9-mediated knockout BMDM cells at basal condition and with LPS treatment for 3 hours after oligomycin, FCCP, and antimycin and rotenone (Ant/Rot). Seahorse XF96 analyser was used to measure the OCR of cells. Data are mean ±SEM of three independent experiments. Figure 4S: A. siNT1 siNT2 Mm47 siRNA1 Mm47 siRNA2 B. Mm47 KD siNT1 siNT2 siRNA1 siRNA2 8000 Media + + + + + + + + + + + + LPS - + + - + + - + + - + + Nigericin - - + - - + - - + - - + GSDMD FL 6000 GSDMD p30

Asc (/siNT1 ) 4000 Mm47

β-actin

IL1b mRN A 2000

C. siNT UT Mm47 siRNA UT 0 siNT LPS Mm47 siRNA LPS NT LPS LPS LPS LPS siNT LPS+nig Mm47 siRNA LPS+nig 0.1ng/mL 1ng/mL 10ng/mL 100ng/mL 80 ns ns 15 60

10 (/siNT1 ) 40

5 20 Sytox/Hoest Percentag e NLRP3 mRN A

0 0 NT LPS LPS LPS LPS 0 20 40 60 0.1ng/mL 1ng/mL 10ng/mL 100ng/mL Minutes

4000

3000 (/siNT1 ) 2000

IL-6 mRN A 1000

0 NT LPS LPS LPS LPS 0.1ng/mL 1ng/mL 10ng/mL 100ng/mL Figure 4S: a) Primary BMDM cells with Mm47 knock-down or non-targeting siRNA were primed with LPS and at 0.1n/mL, 1ng/mL. 10ng/mL and 100ng/mL followed by qRT-PCR for IL1 (upper), Nlrp3 (middle) and IL6 (lower) relative to control (siNT1) untreated cells. b) Mm47 was knocked down using siRNA or non-targeting siRNA in primary BMDM cells. Cells were challenged as indicated and Western blot was performed for full length and cleaved Gasdermin D, Asc, Mm47 and -actin. Representative images from two independent experiments is shown. c) Mm47 was knocked down using siRNA or non-targeting siRNA in primary BMDM cells. Cells were challenged as indicated. Sytox orange and Hoest dye was added to the cells and cell death was measured over time as a ratio of Sytox/Hoest fluorescence using Cytation 5 imager. Average values of the knock down and non-targeting cells are plotted in graph.