Osa-Containing Brahma Complex Regulates Innate Immunity and the Expression of Metabolic in Drosophila

This information is current as Susanna Valanne, Mirva Järvelä-Stölting, Sanna-Kaisa E. of October 1, 2021. Harjula, Henna Myllymäki, Tiina S. Salminen and Mika Rämet J Immunol published online 20 March 2020 http://www.jimmunol.org/content/early/2020/03/19/jimmun ol.1900571 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2020/03/19/jimmunol.190057 Material 1.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 October 1, 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 Errata An erratum has been published regarding this article. Please see next page or: /content/206/4/917.full.pdf

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 © 2020 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published March 20, 2020, doi:10.4049/jimmunol.1900571 The Journal of Immunology

Osa-Containing Brahma Complex Regulates Innate Immunity and the Expression of Metabolic Genes in Drosophila

Susanna Valanne,*,1 Mirva Ja¨rvela¨-Sto¨lting,*,1 Sanna-Kaisa E. Harjula,* Henna Myllyma¨ki,*,2 Tiina S. Salminen,*,† and Mika Ra¨met*,‡,x,{

Negative regulation of innate immunity is essential to avoid autoinflammation. In Drosophila melanogaster, NF-kB signaling– mediated immune responses are negatively regulated at multiple levels. Using a Drosophila RNA interference in vitro screen, we identified a set of genes inhibiting immune activation. Four of these genes encode members of the chromatin remodeling Osa- containing Brahma (BAP) complex. Silencing additional two genes of the BAP complex was shown to have the same phenotype, confirming its role in immune regulation in vitro. In vivo, the knockdown of osa and brahma was shown to enhance the expression of the Toll pathway–mediated antimicrobial peptides when the flies were challenged with Gram-positive bacteria Micrococcus Downloaded from luteus. In this setting, osa knockdown had a particularly strong effect on immune effectors that are predominantly activated by the Imd pathway. Accordingly, Drosophila NF-kB Relish expression was increased by osa silencing. These transcriptional changes were associated with enhanced survival from M. luteus + E. faecalis infection. Besides regulating the expression of immune effector genes, osa RNA interference decreased the expression of a large group of genes involved in metabolism, particularly proteolysis. Of note, the expression of the recently characterized, immune-inducible Induced by Infection (IBIN) was diminished in osa knockdown flies. Although IBIN has been shown to modulate metabolism upon infection, the expression of selected Osa-regulated http://www.jimmunol.org/ metabolism genes was not rescued by overexpressing IBIN. We conclude that the BAP complex regulates expression of genes involved in metabolism at least partially independent or downstream of IBIN. Moreover, Osa affects the NF-kB–mediated immune response by regulating Drosophila NF-kB factor Relish expression. The Journal of Immunology, 2020, 204: 000–000.

he fruit fly Drosophila melanogaster is an established pathways, the Toll and the Immune deficiency (Imd) pathway (6–8). model to study mechanisms of innate immunity (1), in- The Imd pathway is activated when bacteria containing a DAP-type cluding both cellular and humoral responses (2, 3). Mo- peptidoglycan in their cell wall (9, 10) are recognized by the T by guest on October 1, 2021 lecular mechanisms of infection have been extensively studied by pathway receptor PGRP-LC (11–13). The activated signal is using the Drosophila model, and the evolutionary conservation of transduced forward to activate a dimer of NF-kB transcription immune signaling from fruit fly to human has been factors. The Toll pathway is activated by bacteria with a Lys-type demonstrated. Many sophisticated mechanisms are involved in the peptidoglycan in their cell wall (many Gram-positive bacteria) or activation of the cellular immune response when non-self material the fungal b-1,3-glucan (14–16). Activation elicits proteolytic are present, including recognition, phagocytosis, and encapsulation cascades of activator proteins and ultimately the cleavage of and the killing of parasites (4, 5). The humoral immune response Spa¨tzle (Spz) to produce an activated ligand of the pathway (17, leads to the production of antimicrobial peptides and is mainly 18). Activation of the Toll pathway also leads to the activation mediated by two evolutionarily conserved NF-kB signaling and dimerization of NF-kB transcription factors (reviewed in

*Laboratory of Experimental Immunology, Faculty of Medicine and Health Technol- performed the in vivo experiments; all authors analyzed their own data, S.V., M.J.-S., ogy, 33014 Tampere University, Tampere, Finland; †Laboratory of Mito-Immuno- T.S.S., and M.R. wrote the paper; and all authors read and approved the manuscript. Metabolism, Faculty of Medicine and Health Technology, 33014 Tampere University, The dataset presented in this article has been submitted to the Gene Expression Tampere, Finland; ‡PEDEGO Research Unit, Faculty of Medicine, 90014 University x omnibus under accession number GSE120443 (https://www.ncbi.nlm.nih.gov/geo/ of Oulu, Oulu, Finland; Medical Research Center Oulu, 90014 University of Oulu, { query/acc.cgi?acc=GSE120443). Oulu, Finland; and Department of Children and Adolescents, Oulu University Hos- pital, 90014 University of Oulu, Oulu, Finland Address correspondence and reprint requests to Prof. Mika Ra¨met, Laboratory of Experimental Immunology, Faculty of Medicine and Health Technology, Arvo 1S.V. and M.J.-S. contributed equally. Ylpo¨nKatu34,33014TampereUniversity,Tampere, Finland. E-mail address: 2Current address: University of Edinburgh Centre for Inflammation Research, mika.ramet@tuni.fi Queen’s Medical Research Institute, Edinburgh BioQuarter, Edinburgh, U.K. The online version of this article contains supplemental material. ORCIDs: 0000-0003-0898-8423 (S.V.); 0000-0002-5380-6628 (M.J.-S.); 0000-0003- Abbreviations used in this article: Bap60, Brahma associated 60 kD; 3750-2779 (S.-K.E.H.); 0000-0002-7232-0754 (T.S.S.). Bap111, Brahma associated protein 111 kD; BomS1, Bomanin Short 1; BomS2, Received for publication May 21, 2019. Accepted for publication February 12, 2020. Bomanin Short 2; brm, brahma; CecA1, Cecropin A1; dom, domino; DptB, Dipter- icin B; Drs-luc, Drosomycin-luciferase; FC, fold change; FDR, false discovery This work was supported by the Sigrid Juselius Foundation, the Academy of Finland rate; GO, ; IkB, inhibitor of kB; IBIN, Induced by Infection;Imd, (Grant 277495), the Competitive State Research Financing of the Expert Responsi- Immune deficiency; IntS5, Integrator 5; IntS6, Integrator 6; IntS8, Integrator 8; bility Area of Oulu University Hospital and the Tampere Tuberculosis Foundation (to mor, moira; p.i., postinfection; PB, polybromo; PCA, principal component analy- M.R.), the Academy of Finland (Grants 322732 and 328979) (to T.S.S.), and the sis; qRT-PCR, quantitative real-time PCR; RNAi, RNA interference; SNF, sucrose Maud Kuistila Memorial Foundation (to H.M.). The Tampere Drosophila Core Fa- nonfermenting; Snr1, Snf5-related 1; SP, serine protease; Spz, Spa¨tzle;SWI, cility, providing resources for fly work, is partially funded by Biocenter Finland. switching defective; VDRC, Vienna Drosophila Resource Center. S.V., M.J.-S., T.S.S., and M.R. designed the experiments; S.-K.E.H. and H.M. per- formed the in vitro RNA interference screen experiments; S.V., M.J.-S., and T.S.S. Copyright Ó 2020 by The American Association of Immunologists, Inc. 0022-1767/20/$37.50

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1900571 2 THE BAP COMPLEX IN DROSOPHILA IMMUNITY AND METABOLISM

Ref. 6). Relish is the NF-kB factor for the Imd pathway (19), negative control, GFP was carried out using the pMT/BiP/V5-His/GFP whereas Dorsal and Dif are NF-kB factors activated via the Toll plasmid (Invitrogen) as template. dsRNAs were produced from the nes- pathway (20, 21). Although NF-kB factors are normally thought ted PCR templates by in vitro transcription using the T7 MegaScript RNA polymerase (Ambion) according to the manufacturer’s instructions. to function as homodimers, it has been shown that heterodimers between different NF-kB factors do form (22, 23). This changes Luciferase reporter assay and dsRNA treatments the pattern of antimicrobial peptide expression regulated by Luciferase reporter assays were performed essentially as described earlier the NF-kB factors and produces a broader spectrum of host (37). Briefly, for activating the Toll pathway by transfection, S2 cells responses. were transfected either with 0.2 mgoftheToll10b plasmid, encoding a m Activation of immune responses requires strict control to avoid constitutively active form of the Toll receptor (39) or with 0.2 gthe SpzC106 construct pJM856 (kind gifts from J.-L. Imler), both under the costly autoinflammation and tissue damage in the host. For ex- control of a constitutively active Actin promoter. A total of 0.2 mgof ample in D. melanogaster, overactive immune response both the Drosomycin-luciferase (Drs-luc)plasmidand0.2mgoftheActin systemically and in the gut reduces life span (24, 25). The activity 5C-b-galactosidase reporter plasmid were transfected with the pathway- of NF-kBs is typically inhibited by the inhibitor of kB(IkB) inducing plasmids using the Fugene transfection reagent (Roche). The Toll pathway was also activated by the RNAi-mediated silencing of Cact, proteins containing multiple ankyrin repeats, which bind to the k m k the Drosophila I B homolog, by a cact dsRNA treatment (1 g/well). Rel-homology domains of their respective NF- B proteins. In the For measuring the activity of the Imd pathway, cells were transfected Drosophila Toll pathway, IkB Cactus (Cact) is the key inhibitor of with 0.1 mgofaAttacin A-luciferase plasmid and 0.1 mgofaActin NF-kB (26). The negative regulation of the Drosophila NF-kB 5C-b-galactosidase reporter plasmid, and the pathway was induced with activity takes place at multiple levels, and several negative a treatment with heat-killed Escherichia coli for24hpriortomeasuring reporter activities in cells. The dsRNAs directed against the studied regulators have been identified for the Imd pathway, including m genes and controls were added to the cells (1 g/well) together with the Downloaded from PGRP-LF, Pirk, Pickle, Zfh1, and the peptidoglycan degrading plasmid transfections. Primers used for dsRNA production are listed in PGRPs (27–32), summarized in Salminen and Ra¨met (33). Many Supplemental Table I. of these have been characterized at the molecular level, but the Seventy-two to ninety-six hours posttransfection, S2 cells were har- negative regulation of the Toll pathway is less well understood, vested by centrifugation and lysed in Passive Lysis Buffer (Promega). Luciferase activities from cleared lysates were measured with a Lumi- as Cact and Pellino (34) are the only well-characterized negative noskan luminometer (Thermo Fisher Scientific) using a luciferase assay regulators of the Toll pathway. system reagent (no. E1501; Promega) as a substrate. b-Galactosidase ac- In this study, based on a large-scale RNA interference (RNAi) tivities were measured using o-nitrophenyl-b-D-galactopyranoside http://www.jimmunol.org/ screen, we report nine genes that negatively regulate the Toll (ONPG; 1 mM final concentration) as a substrate, by carrying out a colorimetric measurement with a Multiskan spectrophotometer (Thermo pathway–mediated reporter activity in vitro. Four of these genes Fisher Scientific) at 420 nm. are members of the chromatin modifying Osa-containing Brahma (BAP) complex. Furthermore, we show that knocking down Expression of endogenous Drs in S2 cells upon the activation members of the BAP complex enhances the Toll pathway–mediated of the Toll pathway antimicrobial response both in vitro and in vivo. This is in line with For studying the endogenous expression of Drs in Drosophila S2 cells, cells earlier reports that have demonstrated the role of the Brahma were grown on six-well plates. Osa-containing brahma complex genes complex in regulating the activity of the Toll pathway in Drosophila (brahma [brm], osa, Brahma associated protein 111 kD [Bap111], and m by guest on October 1, 2021 (35, 36). When the transcriptome was analyzed from osa RNAi flies moira [mor]) were silenced by transfecting cells with dsRNAs (2 g/well) with the Fugene transfection reagent (Roche). The Toll pathway was ac- and controls with and without infection, we find that osa RNAi has tivated with cact dsRNA (2 mg/well). Cells were collected 48 h later by strong effects on metabolism, predominantly on protein catabolism. centrifugation at 5000 3 g for 3 min. Total RNA was extracted from the In our earlier report, we have shown that the expression of an cells with the TRI reagent as described below. The relative expression of immune-responsive gene Induced by Infection (IBIN)wasdepen- Drs (normalized to the expression of RpL32) was analyzed by quantitative real-time PCR (qRT-PCR) as described below. dent on osa expression and had effects on sugar metabolism. However, the expression of selected Osa-regulated genes involved Drosophila lines in metabolism was not rescued by overexpressing IBIN, indicating The UAS-GAL4–based system in Drosophila to generate silencing or two at least partially separate mechanisms. In addition, we show overexpression of genes in the F1 progeny (40) was used in most exper- that osa knockdown upregulates the Imd pathway target anti- iments. The brm (Vienna Drosophila Resource Center [VDRC] identifier microbial peptide genes in vivo, both upon Micrococcus luteus no. 37720), osa (no. 7810), osa (no. 330350), and MyD88 (no. 25399) cact (Gram-positive) and Enterobacter cloacae (Gram-negative) bacte- UAS-RNAi lines were obtained from the VDRC (www.vdrc.at). UAS- RNAi flies (5848R-3) were obtained from the National Institute of Ge- rial infection. Accordingly, Relish expression is enhanced upon osa netics Fly Stock Center in Japan. RelishE20 mutant flies were a kind gift knockdown both in M. luteus–infected and in E. cloacae–infected from Prof. D. Hultmark. Spz overexpressing flies [yw; P(w+-935; uas- flies, indicating that Osa regulates the expression of Relish. spz*) II (41)] and C564-GAL4 flies, driving the expression of the RNAi construct in the fat body (42) and some other tissues, were a kind gift from Prof. B. Lemaitre (Global Health Institute, Swiss Federal Institute of Materials and Methods Technology Lausanne, Switzerland). The w1118 line was used as a control, Cell culture and the C564-GAL4 driver was backcrossed to the w1118 background. Generating the IBIN OE constructs is described in Valanne et al. (43). To S2 cells were cultured in Schneider’s medium (Sigma-Aldrich) with 10% generate combination lines with the osaIR construct together with IBIN FBS (Sigma-Aldrich), 100 U/ml penicillin, and 100 mg/ml streptomycin overexpressing construct, two IBIN OE lines were used: IBIN OE#2, where (Biochrom, Berlin, Germany) at 25˚C, as previously described (37). the construct is in the second , and IBIN OE#7, where the construct is in the third chromosome. In the VDRC no. 7810 osaIR line, the dsRNA production construct is in the second chromosome. Therefore, IBIN OE#2; osaIR line #7 IR The templates for dsRNA synthesis were generated by a two-step PCR, was generated by combination crossing, whereas the line IBIN OE , osa was produced by recombination. essentially as described in Kleino et al. and Ra¨met et al. (37, 38), using primers for dsRNA production listed in Supplemental Table I. cDNA from Culturing and processing bacteria for infections S2 cells was used as a template for the first PCR with outer primers (I; Supplemental Table I). The nested PCR reactions were carried out using M. luteus and E. cloacae were cultured on Luria-Bertani agar plates under the first PCR products as templates. The following binding site for the T7 antibiotic selection, and Enterococcus faecalis was cultured in brain heart RNA polymerase was included in the T7-PCR primers (II; Supplemental infusion medium as described previously (43). Bacteria were collected Table I): 59-GAATTAATACGACTCACTATAGGGAGA-39. PCR for the from the plate or pelleted and mixed with glycerol to create thick paste. The Journal of Immunology 3

Fly infections analysis, the FC values were generated using a linear modeling process performed with Limma. The values were calculated in relation to a refer- Fly infections were carried out as described previously (43). Briefly, 0- to ence level group (either w, osaIR flies or w, osaIR infected flies in these 2-d-old male flies were collected and placed at +29˚C for 48 h, to maxi- analyses). The p values used are so called FDR p values, which are also mize the UAS-GAL4–mediated gene knockdown or overexpression. To produced by Limma. FDR p values were used to control the rate of false cause septic injury, flies were pricked into the thorax area with a sharpened positive findings in the result list and have been generally found more tungsten wire dipped in the thick bacterial paste. To activate the Toll robust compared with traditional p values. pathway, flies were infected with M. luteus and incubated at 25˚C. For Imd FlyMine (version 46.1, 2018, http://www.flymine.org) is an integrated pathway activation, the flies were infected with E. cloacae. To measure database for Drosophila genomics combining data from more than 30 data gene expression levels, infected flies and noninfected controls were incu- sources. In this study, FlyMine was used to extract gene expression pat- bated at 25˚C for the time required, after which they were collected by terns and functional annotation clusters from the RNA sequencing data. snap freezing them on dry ice or in liquid nitrogen and stored at 280˚C Gene lists that were created based on FDR p values (p , 0.001) between until RNA extraction. For the Toll pathway–mediated survival experi- chosen treatment comparisons were entered to FlyMine. Gene Ontology ments, M. luteus–infected flies (24 h, 25˚C) were subsequently infected (GO) lists were obtained based on similar gene expression patterns and with E. faecalis and incubated at room temperature. For the Imd pathway– functional characteristics such as molecular function, biological process, mediated survival experiments, flies were pricked with E. cloacae the same and cellular component. way as for the gene expression measurements. Statistics RNA extraction and qRT-PCR Statistical analyses of luciferase assay measurements and gene expression For RNA extraction from S2 cells, cells were grown on six-well plates, by qRT-PCR results were carried out using Student t test for two samples harvested by centrifugation as described above, and stored at 280˚C until assuming equal variances. Statistical analyses of survival experiments after RNA extraction. For extracting RNA from flies, 3 3 5 males per genotype pathogen exposure were carried out using the log-rank (Mantel–Cox) test

were collected and snap-frozen on dry ice or in liquid nitrogen. To start the with the GraphPad Prism software (version 5.02). The level of statistical Downloaded from RNA extraction, a sufficient amount of the TRI reagent (MRC; Thermo significance was established as p , 0.05. Fisher Scientific) was added to the frozen S2 cell pellet or to whole flies. S2 cells were homogenized in the TRI reagent by pipetting up and down Data availability for a minimum of 10 times. Whole flies were quickly defrosted and ho- Transcriptome analysis datasets (2) are available via the Gene Expression mogenized in the TRI reagent using a micropestle (Thermo Fisher Scien- Omnibus database. The first dataset entitled “Expression of immune-inducible tific). Subsequently, total were extracted according to manufacturer’s genes in Drosophila melanogaster whole fly samples upon M. luteus infection instructions. RNA pellets were dissolved in nuclease-free water, and the for 24 h” is publicly available via the following link: https://www.ncbi.nlm.nih. RNA concentrations and purity were measured using the Nano-Drop 2000 http://www.jimmunol.org/ gov/geo/query/acc.cgi?acc=GSE120387. equipment (Thermo Scientific). qRT-PCR was carried out from the total The second dataset with the title “The effect of osa RNAi with and RNA samples (∼40 ng/sample) using the iTaq Universal SYBR Green One- without infection” is available through the following link: https://www. step kit (Bio-Rad, Hercules, CA). Either a gene encoding for ribosomal ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE120443. protein L32 (RpL32) or a gene encoding a subunit of mitochondrial re- spiratory complex I, NADH dehydrogenase (ubiquinone) 39 kDa subunit (ND-39), (43) was used as a steadily expressing control gene to nor- Results malize differences in RNA amounts between samples. The primers used Members of the Osa-containing brahma complex regulate are listed in Supplemental Table I. NF-kB–mediated immunity in Drosophila in vitro Transcriptome analysis from total RNAs Large-scale RNAi screens in Drosophila S2 cells have been a by guest on October 1, 2021 Total RNAs from noninfected or M. luteus–infected (24 h) osa knockdown powerful tool for identifying genes involved in immune responses flies (C564 . osaIR) and controls (w, osaIR) were extracted with the TRI and other processes (13). We used this approach to identify regu- reagent. The resulting RNA samples were DNAse treated with the Rap- lators of the Drosophila NF-kB signaling pathway (37, 45). These idOut DNA removal kit (Thermo Scientific). RNA quality was assessed screens identified nine genes: osa, brm, Bap111, mor, domino with the Advanced Analytical Fragment Analyzer and found to be good. dom Integrator 5 IntS5 Integrator 6 IntS6 Integrator 8 Total RNA samples were pure and intact, and all samples were of similar ( ), ( ), ( ), quality. (IntS8), and PDGF- and VEGF-receptor related (Pvr)that The preparation of the RNA library and Illumina HiSeq 2500 sequencing negatively regulate the activity of the Toll pathway. When the were carried out in the Finnish Microarray and Sequencing Centre (Turku, Toll pathway–related immune response was activated in S2 Finland) as described previously (43). Briefly, RNA library was prepared cells either with Toll10b, a constitutively active form of the Toll according to Illumina TruSeq Stranded mRNA Sample Preparation Guide, C106 the mRNA were fragmented into small pieces, and the cleaved RNA receptor, or with an active form of the Spz ligand Spz , fragments were copied into first-strand cDNA using reverse transcriptase RNAi-mediated knockdown of these nine genes caused hyper- and random primers. The Unique Illumina TruSeq indexing adapters were activation of the Drs-luc reporter gene (Fig. 1A, 1B). dom is a ligated to samples for later sample pooling in one flow cell lane. The gene originally characterized to be required for cell proliferation, products were purified and enriched with PCR to create the final cDNA viability, hematopoiesis, and homeotic gene silencing (46, 47). library. The samples were normalized, pooled for clustering, and se- quenced with an Illumina HiSeq 2500 instrument using TruSeq v3 se- IntS5, IntS6, and IntS8 are members of the Integrator complex quencing chemistry. Paired-end sequencing with a 1 3 50 bp read length shown to be responsible for small nuclear RNA 39 end formation was used, followed by a 6-bp index run. The technical quality of the HiSeq (48). Pvr is a receptor protein–tyrosine kinase, the activation of 2500 run was good and the cluster amount was as expected. which has been shown to inhibit humoral responses while stimu- The principal component analysis (PCA) is an ordination technique complementary to clustering and it is used for reducing the dimensionality lating cellular responses (49). Accordingly, it has been shown earlier of large data sets. The PCA was carried out by transforming a large set of that Pvr knockdown increases the activity of the Imd pathway– variables to make a smaller set that still contains most of the information. In mediated Attacin-luciferase reporter response in Drosophila S2 cells this ordination technique, similar objects are placed near each other and (37). The mechanisms how these genes function as negative regu- dissimilar objects are placed further from each other. PCA was carried out lators of the Toll signaling remain to be investigated. and the PCA plot was drawn using normalized read values from the transcriptome data. Importantly, four of the nine genes identified as negatively To generate lists of up- and downregulated genes from the transcriptome regulating the Toll pathway were members of the Osa-containing data (Supplemental Table II), the following filtering criteria were used: fold Brahma (BAP) complex, namely osa, brm, Bap111, and mor, $ change (FC) 3.0 and false discovery rate (FDR)–corrected p value of marked with gray bars in Fig. 1 (50). This finding prompted us to ,0.001, which were calculated using the Limma package (Bioconductor) (44). In addition, genes whose transcript abundances were below 0.025 look at the role of the BAP complex in Drosophila immunity in reads across the treatments were excluded. FC describes the size of the more detail. To this end, we carried out targeted silencing of se- difference in gene expression between the compared groups. In our lected components of the two Drosophila Brm complexes (Fig. 1C), 4 THE BAP COMPLEX IN DROSOPHILA IMMUNITY AND METABOLISM Downloaded from http://www.jimmunol.org/

FIGURE 1. RNAi silencing of members of the BAP complex in Drosophila S2 cells hyperactivates the Toll pathway and reduces the activity of the Imd pathway by guest on October 1, 2021 in vitro. The Toll pathway was activated by (A and D) overexpression of Toll10b and (B and E) overexpression of SpzC106,afterwhichtheDrs-luc values were measured, normalized to Act-b-gal values, and plotted. (C) A schematic overview of the Drosophila Brahma complexes (51), with common components of the complexes marked in gray. Osa (striped background) is unique to the BAP complex, whereas PB and BAP170 (white background) define the PBAP complex. (F and G)Acact dsRNA treatment activates the expression of the (F) Drs-luc reporter as well as the (G) endogenous Drosomycin (Drs), and the signal is further enhanced by RNAi-mediated silencing of the BAP complex genes. (H) Knockdown of members of the BAP complex reduces the Imd pathway–mediated AttA-luc activity. BAP complex members are marked with gray bars and results are shown as an average 6 SD. Statistical analyses were carried out using Student t test for two samplesassumingequalvariances.Ineachgraph, the relative reporter activity value (A, B, D–F,andH)orexpressionvalue(G) of cells with an activated pathway treated with the negative dsRNA control (GFP) is set to 1. Asterisks indicate statistical significance: *p , 0.05, **p , 0.01, ***p , 0.001. n.s., not significant. as well as an additional dsRNA targeting osa (osaB). As shown in one pathway, we performed a double knockdown of either GFP or Fig. 1D and 1E, silencing osa expression using osaB resulted in osa together with other BAP complex members studied, namely elevation of both Toll10b and SpzC106-mediated Toll pathway ac- brm, mor, Bap111, BAP60,andSnr1. Simultaneous knockdown of tivity. In addition, silencing of Brahma associated protein 60 kD osa and other BAP complex members did not further increase the (Bap60)andSnf5-related 1 (Snr1) had a similar effect to the Drs-luc SpzC106-induced Drs-luc signal, indicating that the factors are activity, whereas silencing of polybromo (PB), a component of the functioning within the same pathway (Supplemental Fig. 1A). Our other Brm complex (PBAP; Fig. 1C), had no such effect (Fig. 1D, results demonstrate that also in the context of the Toll pathway, the 1E). To further study the role of the BAP complex members in the BAP complex appears to function as a chromatin modifier. Toll pathway, we activated the Toll pathway by RNAi-mediated When the Imd pathway was activated in S2 cells with heat-killed silencing of the Drosophila NF-kB inhibitor cact.Alsointhis E. coli for 24 h, silencing of members of the BAP complex caused case, the activity of Drs-luc was significantly higher when com- a reduction in the relative Attacin-luciferase (AttA-luc) values ponents of the BAP complex (osa, brm, Bap111, mor, Bap60,and (Fig. 1H). This is in line with results obtained by Bonnay and Snr1) were knocked down, whereas knocking down PB had no such coworkers (35), who showed that following an E. coli immune effect (Fig. 1F). In addition, the cact RNAi–induced endogenous challenge, RNAi silencing of brm, osa,andmor in S2 cells sig- gene expression of Drs in S2 cells was also monitored with qRT- nificantly reduced the induction of Attacin-A-luciferase compared PCR, showing that Drs levels were elevated upon RNAi silencing of with the negative control GFP. In addition, expression of the E. coli members of the BAP complex (Fig. 1G), indicating that the role of induced endogenous Diptericin B (DptB) was reduced upon the BAP complex is downstream of Cact. This is in line with pre- silencing of the BAP complex members, whereas Cecropin vious findings, showing that the BAP complex controls gene ex- A1 (CecA1) expression was elevated (Supplemental Fig. 1B, 1C). pression by modifying chromatin structure (50–52). Furthermore, to Bonnay and coworkers (35) have also previously reported that the determine if the identified factors function independently or within effect of silencing of the BAP complex to the Imd pathway–mediated The Journal of Immunology 5 antimicrobial peptides is very strongly context dependent. In con- per group were analyzed. A PCA (Fig. 3A) indicated that based on clusion, the BAP complex appears to regulate the Drosophila NF-kB the general expression patterns, the three biological replicates pathways Toll and Imd in S2 cells in vitro. within each treatment group fall close to each other. Silencing osa with the C564-GAL4 driver, which is strongly expressed in the fat Silencing of osa and brm enhances the Toll pathway–mediated body, had a modest but significant effect on osa expression at the immune response in vivo in Drosophila whole fly level (Fig. 3B). However, both the infection and the We investigated the in vivo role of the BAP complex in Drosophila C564-driven osa RNAi have strong effects on the general gene immunity, by following the effect of knocking down osa (osaIR) expression patterns compared with the control (w, osaIR) flies, and brm (brmIR). osaIR (no. 7810; GD-collection), osaIR (no. 330350; which is shown in the clustering of the samples in the PCA plot short hairpin RNA collection), brmIR (no. 37720; GD-collection), (Fig. 3A). Of note, silencing osa using the C564-GAL4 driver had MyD88IR (no. 25399; GD-collection), and cactIR (5848R-3; Kyoto no clear effect on the expression of other components of the BAP collection) flies were crossed with the C564-GAL4 driver, which is complex at the whole fly level. strongly expressed in the fat body (31, 42, 53). A background To analyze the differentially expressed transcripts between the control (VDRC identifier no. 60200) was included in the experi- treatments, the data were filtered to generate lists of up- and ments with the osaIR line from the VDRC short hairpin RNA col- downregulated genes (Supplemental Table II). Cluster analyses of lection (no. 330350). Three- to five-day-old males were infected the up- and downregulated genes in selected comparison groups with M. luteus, and RNA samples were collected 12 and 24 h were carried out using the FlyMine analysis tool. osa knockdown postinfection (p.i.). Silencing osa using the C564-GAL4 driver effect on downregulated genes without and with M. luteus 24 h p.i. (C564 . osaIR [no. 7810] and C564 . osaIR [no. 330350]) resulted (C564.osaIR versus w,osaIR) showed that the GO enrichment– in elevation of the Toll pathway target genes Bomanin Short 1 based biological process gave just one GO term after the Holm– Downloaded from (BomS1), Bomanin Short 2 (BomS2), and Drs, at both 12 h p.i. and Bonferroni test correction (p value , 0.05), namely Proteolysis. 24 h p.i. (Fig. 2A–F). Silencing of brm did not show a statistically Thirty (27%) genes from altogether 115 downregulated genes significant elevation of the Toll pathway target genes, although the from the C564.osaIR versus w,osaIR comparison had a function in trend was similar to that of osa silencing. As expected, silencing of proteolysis, and the corresponding comparison with infected flies MyD88 strongly inhibited the expression of BomS1, BomS2 and showed a similar pattern, as 39 (23%) of 166 downregulated genes Drs, whereas knocking down the negative regulator cact (C564 . had a function in proteolysis. From these two comparisons, four http://www.jimmunol.org/ cactIR) activated the pathway and high levels of BomS1, BomS2 and proteolysis related genes were shown to be specific to uninfected Drs expression was detected also in the uninfected flies (Fig. 2A–F). osaIR flies, 26 genes were common in both uninfected and To confirm the role of Osa in the Toll pathway, we generated flies M. luteus–infected osaIR flies, and 13 were specific to M. luteus– that overexpress a bona fide Toll agonist, Spz, together with the osa infected osaIR flies (Fig. 3C). b-Site APP-cleaving enzyme (Bace) RNAi construct (no. 7810). As expected, overexpression of Spz encodes an aspartic protease that cleaves amyloid precursor pro- caused induction of the Toll pathway targets BomS1, BomS2, and teins, and it had the strongest downregulation among the Prote- Drs (Fig. 2G–I). Expression of these genes was further en- olysis cluster in both uninfected and infected osaIR flies. hanced in Spz overexpressing flies with osa RNAi, showing that A majority, 63% (27/43) of the genes in the Proteolysis cluster by guest on October 1, 2021 silencing osa enhances Toll pathway target genes also in this was found to contain serine-type endopeptidase (i.e., serine pro- setting (Fig. 2G–I). tease [SP] activity) (56, 57), including 11 genes from the Jonah Next, we examined the effect of osa knockdown on the survival (Jon) gene family and five trypsin genes (epsilonTry, betaTry, of flies when exposed to pathogenic bacteria. Silencing osa using alphaTry, thetaTry, Try29F), which are putative digestive enzymes. the C564 driver causes problems in the fly development: a pro- Jonah and Trypsin genes, as well as the majority of the other SP portion of C564.osaIR (no. 7810) progeny flies show incomplete genes, are expressed and enriched in the midgut (Fig. 3C). Also a fusion of the abdominal epidermis (54, 55) (Supplemental Fig. 2A, cluster of six genes (Mal-A7, Amy-d, Amy-p, Mal-A6, Mal-A8,and see also Supplemental Fig. 2C and below), whereas a large pro- LManIII) that function in carbohydrate metabolic process was portion of C564.osaIR (no. 330350) flies have a deformed wing downregulated in uninfected osaIR flies and seven genes (Mal-A7, phenotype, edema, and markedly reduced viability (Supplemental Mal-A6, CG32444, UGP, Mal-A8, LManIII,andLManIV)inosaIR Fig. 2B). Therefore, the survival experiments were carried out only flies upon infection (C564.osaIR M. luteus 24 h versus w,osaIR with C564.osaIR (no. 7810) progeny flies. To monitor the Toll M. luteus 24 h). pathway–mediated survival, flies were first infected with M. luteus Based on the FlyMine analysis, osaIR caused upregulation of to activate the Toll pathway, and thereafter infected with E. faecalis, two gene clusters. The first of these gene clusters contain genes which causes a pathogenic infection by septic injury in Drosophila involved in chitin metabolism and cuticle development, upregu- (45). As is shown in Fig. 2J, C564-GAL4–driven osa RNAi im- lated in uninfected and infected osaIR flies (Supplemental Fig. proves the survival of flies from a M. luteus + E. faecalis infection 2C). This might be explained by a previously published finding compared with controls. No effect was observed in the Imd that osa RNAi and dalao/Bap111 RNAi, both members of the pathway–mediated survival with E. cloacae infection (Fig. 2K). BAP complex, cause a defect in the tergite pigmentation in the In conclusion, the BAP complex members Osa and Brm neg- abdomen of Drosophila adults (54). As mentioned above, we atively regulate the Toll pathway–mediated immunity in Dro- observed a similar phenotype in some C564.osaIR (no. 7810) sophila in vivo. flies (Supplemental Fig. 2A). However, this defect is more severe than a mere pigmentation defect in some of the flies, because Transcriptome profiling of osa knockdown flies confirms the visually, osa silencing causes incomplete fusion of the abdominal k function of Osa in the regulation of NF- B signaling and epidermis [the phenotype of which is demonstrated in e.g., (55)]. indicates a role in metabolism Thus, enhanced chitin metabolism is likely a mechanism to com- To understand the role of the BAP complex member osa in more pensate for problems in the cuticle formation. detail in Drosophila immunity, we carried out transcriptome Expectedly, a major cluster of upregulated genes upon in- analysis of M. luteus–infected and uninfected osa knockdown flies fection was predominantly involved with immune response, (C564.osaIR [no. 7810]) and controls. Three biological replicates with the following GO terms: response to other organism 6 THE BAP COMPLEX IN DROSOPHILA IMMUNITY AND METABOLISM

FIGURE 2. Knockdown of osa in- creases the Toll pathway–mediated antimicrobial peptide expression and survival in Drosophila in vivo. (A–F) Expression of the selected Toll path- way target genes was monitored in osa and brm knockdown flies and controls with and without M. luteus infection for 12 or 24 h. In each graph, the ex- pression value of w, osaIR flies in- fected for 12 h is set to 1. MyD88 Downloaded from RNAi was used as a control for Toll pathway function, and it prevents Toll pathway induction. Cact RNAi causes induction of the Toll pathway also without infection. (A and B) BomS1, (C and D) BomS2, and (E and F) Drs http://www.jimmunol.org/ expression. (G–I) Spz overexpression (C564.SpzOE) causes induction of Toll pathway targets; MyD88 RNAi pre- vents this induction. In each graph, the expression value of C564.SpzOE flies is set to 1. The expression of the Toll pathway target genes is further enhanced upon silencing of osa (no. 7810) also in this setting. (G) BomS1,

(H) BomS2,(I) Drs. Statistical anal- by guest on October 1, 2021 yses of differences in gene expression in the qRT-PCR measurements were carried out using Student t test for two samples assuming equal vari- ances. (J) Knockdown of osa using the no. 7810 RNAi line crossed over the C564-GAL4 driver increases survival from M. luteus + E. faecalis infection. (K) osa knockdown (C564.osaIR [no. 7810]) has no significant effect on survival upon E. cloacae infection. Statistical analyses of survival ex- periments were carried out using the log-rank (Mantel–cox) test.

(33 genes), defense response (29 genes), response to bacterium 24hversusw,osaIR M. luteus 24 h), the following cluster of (24 genes), defense response to Gram-positive bacterium (12 genes), genes was upregulated: defense response to other organism response to fungus (10 genes), and proteolysis (25 genes, in- (nine genes: AttA, DptB, DptA, CecA2, pirk, NimB3, AttC, cluding e.g., SPE). Expression of many of these was further Spn28Dc, PGRP-SB1). Selected immunity-related genes are enhanced upon osa silencing. In the comparison between in- explored in more detail below. These results confirm the earlier fected osaIR flies and infected controls (C564.osaIR M. luteus results obtained by qRT-PCR and demonstrate that Osa has a The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 3. Transcriptome analysis of infected and uninfected osa knockdown flies and controls. (A) PCA showing the positioning of samples based on the general gene expression patterns between treatments and biological replicates. (B) osa expression in different treatments showing the knockdown efficiency at the whole-fly level upon C564-GAL4-driven osa knockdown (C564.osaIR [no. 7810]). Statistical analysis was carried out using Student t test for two samples assuming equal variances. (C) Genes involved in metabolism, especially proteolysis, were strongly downregulated by osa knockdown in both uninfected and M. luteus–infected flies. The list of genes downregulated by osaIR was generated with the Limma Bioconductor package using the following filtering criteria: FC . 3.000, FDR-corrected p value ,0.001; excluding genes whose transcript abundances were below 0.025 reads across the treatments. general role in regulating the magnitude of the immune re- encoding gene CG44404 (IBIN)(43)wasidentified.Ofnote, sponse upon challenge with Gram-positive bacteria M. luteus. upon C564-GAL4–mediated silencing of osa, the expression of IBIN was diminished to ,1% of the induced expression Osa regulates the expression of the infection-inducible in the control flies (Fig. 4A). It was recently published that molecule IBIN IBIN is induced by both Gram-negative and Gram-positive When analyzing the genes that were most strongly induced by bacterial infection as well as by the parasitoid wasp Lep- M. luteus infection in the transcriptome data, a short peptide topilina boulardi and that its induction is dependent on osa 8 THE BAP COMPLEX IN DROSOPHILA IMMUNITY AND METABOLISM expression (43). We next investigated in more detail at which dependent on osa expression with both Gram-positive and Gram- time points Osa is needed for the expression of IBIN during the negative bacteria, at all time points tested. course of the infection. As shown in Fig. 4B and 4C, expression of osa is required for IBIN expression from the start of the infection Osa-mediated downregulation of sugar metabolism genes is in both M. luteus and E. cloacae infection models. A small amount not rescued by overexpression of IBIN of IBIN isexpressedbetween2and8htimepoints,butthisex- In addition to genes involved in proteolysis, osa RNAi also down- pression is completely abolished by 24 h p.i., when osa is silenced regulates sugar metabolism genes, such as Maltase-A6, -A7,and-A8, using the C564-GAL4 driver. We conclude that IBIN expression is as shown in the transcriptome data with the osaIR (no. 7810) flies Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 4. Expression of IBIN is dependent on osa expression and overexpression of IBIN does not rescue the osaIR-mediated downregulation of sugar metabolism genes. (A) IBIN is strongly induced by M. luteus infection and fully dependent on osa expression. (B and C) Expression of osa is required for IBIN expression starting from the onset of the infection in both M. luteus and E. cloacae infection models. (D) osa RNAi (C564.osaIR [no. 7810]) causes downregulation of genes involved in metabolism, including sugar metabolism genes Maltase-A6, -A7, and -A8, as assessed in the transcriptome analysis as normalized number of reads. Statistical analyses were carried out with the Limma Bioconductor package, the FDR-corrected p values are marked. (E) Also in C564.osaIR (no. 330350) flies, the expression of Maltase-A6, -A7, and -A8 genes is downregulated when measured by qRT-PCR. (F) The E. cloacae– induced endogenous expression of IBIN is abolished in the lines containing C564.osaIR (no. 7810). C564-GAL4–driven overexpression of IBIN is demonstrated from both IBIN OE#2 and IBIN OE#7 lines. (G) Maltase-A8 expression levels are strongly downregulated by osa RNAi (C564.osaIR [no. 7810]) and upregulated by IBIN OE#2 or IBIN OE#7 overexpression. However, overexpression of IBIN in the combination lines IBIN OE#2;osaIR (no. 7810) and IBIN OE#7,osaIR (no. 7810) does not rescue the osaIR-mediated downregulation of Maltase-A8. Statistical analyses of qRT-PCR results (E–G) were carried out using Student t test for two samples assuming equal variances. The Journal of Immunology 9

(Fig. 4D). The effect of osa silencing to these genes was further To further investigate the role of Osa in the expression of an- confirmed by quantitative PCR using the osaIR (no. 330350) line timicrobial peptide genes, we infected osa RNAi flies and con- (Fig. 4E). Because IBIN enhances the expression of a group of trols either with Gram-negative (E. cloacae) or Gram-positive genes involved in sugar metabolism (43), we tested if over- (M. luteus) bacteria and used quantitative PCR to measure the expression of IBIN rescues the osa RNAi-mediated downregula- expression levels of the typical Imd pathway target antimicrobial tion of selected sugar metabolism genes. To this end, we generated peptides (AttC, DptB, and CecA1) as well as of the known Toll combination flies with the osaIR (no. 7810) construct and an IBIN pathway target gene BomS1 (Fig. 5D–G). Upon infection with overexpression construct, using two separate IBIN construct in- either E. cloacae or M. luteus, the C564 . osaIR flies showed an sertions, namely IBIN OE#2 and IBIN OE#7. C564-GAL4–driven upregulation of all of the genes studied, compared with controls. overexpression of IBIN is demonstrated from both lines, but the Of note, the inductions of AttC, DptB, and CecA1 were totally expression is stronger in the line IBIN OE#7 compared with the Relish dependent with both infection models (Fig. 5D–F). How- IBIN OE#2 line (Fig. 4F). ever, the induction of BomS1 with the Gram-positive bacteria As demonstrated in Fig. 4F, the E. cloacae–induced endogenous M. luteus was Relish independent, whereas with the Gram-negative expression of IBIN is abolished in the lines containing C564.osaIR. E. cloacae, the induction was Relish dependent (Fig. 5G). Accordingly, when overexpressed at the C564.osaIR background, Based on the RNA sequencing data, Relish expression is altered IBIN expression is at the same level in uninfected and infected upon both infection and osa RNAi. To this end, we investigated samples (Fig. 4F). In line with the transcriptome data, the Maltase-A8 the role of Relish in more detail in the osa knockdown flies and expression levels are strongly downregulated by osa RNAi also when controls. It has previously been shown that in addition to the well- measured with quantitative PCR (Fig. 4G). As shown previously documented posttranslational activation events of Relish (62), Downloaded from (43), Maltase-A8 expression levels are upregulated by IBIN over- elevation in the full-length Relish expression induces Diptericin expression, in both of the lines used (IBIN OE#2 or IBIN OE#7) expression (63) and is needed for Attacin A/B inducibility in (Fig. 4F). However, overexpression of IBIN, either from the line IBIN Relish mutants (64). Therefore, we measured the Relish expres- OE#2 or IBIN OE#7, does not rescue the downregulation of Maltase- sion levels in the E. cloacae (24 h)– or M. luteus (24 h)–infected A8 (Fig. 4G). We conclude that the mechanism of osaIR-mediated and uninfected flies with and without osa knockdown (Fig. 5H). In downregulation of genes involved in sugar metabolism differs from Fig. 5H, it is shown that Relish expression is enhanced upon osa the mechanism of IBIN action or is downstream of IBIN. knockdown both in M. luteus–infected and in E. cloacae–infected http://www.jimmunol.org/ flies as well as in the uninfected flies. This points to the conclusion Osa regulates the expression of the Drosophila NF-kB factors that Osa regulates the expression of Relish and thus affects the Relish and Dif balance of Drosophila NF-kB gene expression (22). It appears that Because of our data showing that osa RNAi causes elevation of upon osa RNAi in infected flies, the favored NF-kB factor is selected Toll pathway target genes, namely Drs, BomS1, and Relish, at least transcriptionally, which then leads to enhanced BomS2 (Fig. 2), we investigated the Toll pathway targets on a expression of the classical Imd pathway target genes including more global scale in the transcriptome analysis. As shown in AttC, DptB, and CecA1. Fig. 5A, the target genes of the Toll pathway were activated by In conclusion, we have shown that the chromatin remodeling by guest on October 1, 2021 M. luteus infection, and the expression of most of these genes Osa-containing Brahma complex regulates the expression of genes was further enhanced with osa RNAi. In addition to the clas- involved in metabolism at least partially independent or down- sical Toll pathway target genes Drs and mtk, the Toll pathway stream of IBIN. Furthermore, the induced expression of target target genes include a recently expanded family of Bomanin genes of the Toll pathway is enhanced by osa knockdown both (Bom) genes (S. A. Wasserman, personal communication to in vitro and in vivo. In addition, osa knockdown increases the Imd FlyBase, FBrf0243179), some of which were previously des- pathway target antimicrobial peptide gene expression in vivo, both ignated as immune induced genes (58). Among the Toll targets upon M. luteus and E. cloacae infection, and this is associated are also newly named Daisho genes Daisho1 (dso1,previously with enhanced expression of Relish. IM4)andDaisho2 (dso2, previously IM14) (59). Although most of the individual changes in the expression of Toll path- Discussion way target genes did not show statistically significant differ- In our present work, we have shown that the Osa-containing ence after FDR correction, these overall changes show general Brahma complex (BAP complex) negatively regulates the Toll enhancement of the Toll pathway activation by osa RNAi. pathway–mediated immunity in in vitro and in vivo Drosophila. Importantly, many of the target genes induced predominantly Moreover, in vivo, C564-GAL4–mediated osa knockdown had a via the Imd pathway were also induced by the M. luteus in- particularly strong effect on target genes that are mainly activated fection, and their activation was hyperactivated upon osa RNAi via the Imd pathway. Osa regulates the expression of Relish, (Fig. 5B). Interestingly, expression of PGRP-SA, which encodes which points to regulating the balance between the Drosophila for a peptidoglycan recognition protein in the Toll pathway (14), NF-kB transcription factors Relish and Dif. In addition to its ef- and PGRP-SB1, a gene induced via the Imd pathway, encoding a fects on immune genes, osa RNAi downregulated the expression catalytically active PGRP but with no confirmed function in im- of a cluster of genes shown to be involved in metabolism, in munity (60), are strongly affected by osa RNAi. The expression of particular proteolysis. PGRP-SD, a secreted pattern-recognition receptor upstream of The Drosophila Brahma protein is a member of the switching PGRP-LC that enhances the Imd pathway activation (61), is in- defective (SWI)/sucrose nonfermenting (SNF) complex, which is duced by M. luteus infection but downregulated by osa knock- an evolutionarily conserved chromatin remodeling complex orig- down. As shown in Fig. 5C, in the M. luteus–infected flies, osa inally identified in yeast that uses the energy from ATP hydrolysis RNAi has an opposite effect on the expression of Dif and Relish. to modulate chromatin structure. Brahma forms complexes with Upon osa knockdown the expression of Relish is enhanced, other complex subunits, including Osa and Polybromo, that as- whereas the expression of Dif is decreased. The expression of sociate with Brahma in a mutually exclusive manner. This gives dorsal, although slightly elevated by M. luteus infection, is not rise to two variants of the Brahma complex, BAP and PBAP affected by osa knockdown in the infected flies (Fig. 5C). (51, 52, 65). The BAP complex is defined by the presence of Osa 10 THE BAP COMPLEX IN DROSOPHILA IMMUNITY AND METABOLISM Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 5. osa silencing elevates the expression levels of both the Toll and the Imd pathway target antimicrobial peptides by enhancing the expression of Relish.(A and B) Expression of selected genes induced by infection in the transcriptome analysis. (Ai and Aii) Toll pathway target genes; (Bi, Bii,andBiii)Imd pathway-mediated antimicrobial peptides and other Imd pathway target genes. (C)ExpressionoftheDrosophila NF-kB factors in the transcriptome data. (A–C) Statistically significant differences in the transcriptome data were calculated with the Limma Bioconductor package, and FDR-corrected, statistically significant p values (p , 0.001) are marked. (D–G) Expression of the selected NF-kB target genes as measured by qRT-PCR from uninfected flies and flies with M. luteus infection (24 h; validation of the transcriptome analysis data) or E. cloacae infection (24 h). In (D)–(G), RelE20 flies succumbed to E. cloacae infection prior to the sampling time point (24 h, marked with no. ). (H)ExpressionofRelish is elevated in C564.osaIR (no. 7810) flies upon both M. luteus and E. cloacae infection as well as in uninfected samples. Statistical analyses of qRT-PCR results (D–H) were carried out using Student t test for two samples assuming equal variances. The Journal of Immunology 11

(50, 66), whereas Polybromo/BAP180 and BAP170 specify the for 48 h prior to infection experiments, and the flies were able to PBAP complex (51). In yeast, members of the SWI/SNF complex sustain this without problems. were originally identified as required for the expression of the The interplay between the BAP or PBAP complex and the Imd/ mating type gene HO or the sucrose fermentation gene SUC2. Relish pathway has also been investigated. Using a genome-wide Later, it became evident that the SWI/SNF complexes affect the approach, Bonnay and coworkers (35) showed that the conserved expression of numerous other genes, and they are needed for al- nuclear protein Akirin is a NF-kB cofactor required for the acti- tering the organized chromatin structure so that transcription factors vation of a subset of Relish-dependent genes. BAP60, a compo- can access their promoters (reviewed by Ref. 67). Whole genome nent of the BAP complex, binds to Akirin upon an immune expression profiling indicated that BAP and PBAP control different challenge. Akirin, together with DMAP1, has been shown to in- cellular processes; BAP activates the transcription of a key mitotic teract also with the Brahma-associated protein BAP55 (69), and regulator stg/cdc25, and the binding of BAP to the stg/cdc25 pro- Akirin has been shown to facilitate an interaction between Twist moter is critically dependent on Osa. Both complexes associate and the Brm-containing complexes to promote gene expression in mainly with open, hyperacetylated chromatin regions (52). development (70). In addition, when looking at protein–protein In this study we show that at the transcriptome level, the Osa/ interactions in the Imd pathway, members of the Brm complex BAP complex strongly regulates the expression of metabolic (BAP60, BAP55; Moira) were identified (71). Furthermore, He genes. There are no published studies about the role of Osa/BAP and coworkers (72) have recently shown that Polybromo/BAP180, complex in metabolism. In this study, we have demonstrated that the core member of the PBAP complex, is needed to maintain osa knockdown has marked effects on genes important for homeostasis in the Drosophila intestine upon immune activation metabolism, mainly proteolysis but also genes important for sugar —bap180 mutant flies have a hyperactivated immune response in catabolism. The effect of osa knockdown on the metabolic genes the midgut in the Erwinia carotovora carotovora 15 (Ecc15) in- Downloaded from is seen in both uninfected and infected flies. The majority of the fection model. In these conditions, Bap180 would act as a tran- proteolytic genes in the clusters identified were found or predicted scriptional repressor, so the hyperactivation in bap180 mutants is to have SP activity. SP and their homologs comprise a large family proposed to happen via increased eiger expression without the of genes in the Drosophila genome, with diverse activating repressor being present. Interestingly, Bap180 was shown to functions (56, 57). Functions of SPs range from digestion of di- physically interact with Relish upon Imd pathway activation (72). etary proteins to cascades generating, for example, active cyto- In this current study, we have identified a novel mechanism http://www.jimmunol.org/ kines. Many SPs and SP homologs contain a CLIP domain, which regarding how innate immune response is regulated in Drosophila. is thought to regulate the SP activity (56). In our transcriptome The close connection between immune reaction and metabolic analysis, the proteolysis cluster contained Trypsin genes, Jonah changes has become evident. More detailed understanding about genes and a more diverse group of genes with predicted SP ac- pathways, mechanisms, and molecules that regulate these re- tivity, most of which, according to FlyBase expression data, have sponses will help us to understand reasons for harmful metabolic strong expression in the midgut. Moreover, the Osa-regulated changes that are associated with chronic inflammation. genes involved in sugar catabolism are strongly expressed in the gut. This points to Osa’s regulatory role in the catabolism of di-

Acknowledgments by guest on October 1, 2021 etary molecules. A role for the Osa-dependent, short-polypeptide We thank Tuula Myllyma¨ki (Tampere University) for technical assistance IBIN in enhancing the catabolism of starch was also suggested and all the members of the Experimental Immunology research group for recently, because IBIN overexpression had an enhancing effect on insightful scientific discussions. Tampere Drosophila Core Facility is ac- the expression of sugar metabolism genes, including several knowledged for providing resources for fly work. maltases (43). However, IBIN overexpression did not rescue the osa knockdown–mediated downregulation of maltases, indicating Disclosures that the pathways involved are at least partially independent. The authors have no financial conflicts of interest The BAP or PBAP complexes in Drosophila NF-kB–mediated immunity have been studied previously. The interplay between the References Toll pathway and the BAP complex in Drosophila has been in- 1. Ra¨met, M. 2012. The fruit fly Drosophila melanogaster unfolds the secrets of vestigated in two separate studies. In a study by Ku¨ttenkeuler and innate immunity. Acta Paediatr. 101: 900–905. coworkers (36), a subset of a genome-wide RNAi library was used 2. Hultmark, D. 2003. Drosophila immunity: paths and patterns. Curr. Opin. Immunol. 15: 12–19. to search for transcriptional modulators of the expression of Toll 3. Lemaitre, B., and J. Hoffmann. 2007. The host defense of Drosophila mela- target genes. In addition to their main finding identifying Deaf1 as nogaster. Annu. Rev. Immunol. 25: 697–743. an important transcription factor of the Toll pathway, members of 4. Ulvila, J., L. M. Vanha-Aho, and M. Ra¨met. 2011. Drosophila phagocytosis - still many unknowns under the surface. APMIS 119: 651–662. the BAP complex Brm and Dalao (Bap111) were identified as 5. Banerjee, U., J. R. Girard, L. M. Goins, and C. M. Spratford. 2019. Drosophila negative regulators of the Toll pathway (36). Also, Bonnay and as a genetic model for hematopoiesis. Genetics 211: 367–417. coworkers (35) showed that the expression of the Drs-luc reporter 6. Valanne, S., J. H. Wang, and M. Ra¨met. 2011. The Drosophila Toll signaling pathway. J. Immunol. 186: 649–656. was enhanced when the BAP/PBAP complex components bap55, 7. Myllyma¨ki, H., S. Valanne, and M. Ra¨met. 2014. The Drosophila imd signaling bap60, brahma, mor, osa, snr-1,orPB were knocked down by pathway. J. Immunol. 192: 3455–3462. 8. Lemaitre, B., E. Kromer-Metzger, L. Michaut, E. Nicolas, M. Meister, RNAi in S2 cells. However, in both studies, the role of the Toll P. Georgel, J. M. Reichhart, and J. A. Hoffmann. 1995. A recessive mutation, pathway has been studied only in vitro. Bonnay and coworkers immune deficiency (imd), defines two distinct control pathways in the Dro- (35) had difficulties obtaining progeny with the osa VDRC RNAi sophila host defense. Proc. Natl. Acad. Sci. USA 92: 9465–9469. 9. Kaneko, T., W. E. Goldman, P. Mellroth, H. Steiner, K. Fukase, S. Kusumoto, line that was used also in this study (no. 7810), using the C564- W. Harley, A. Fox, D. Golenbock, and N. Silverman. 2004. Monomeric and GAL4 driver. We obtained a fair amount of C564.osaIR progeny polymeric gram-negative peptidoglycan but not purified LPS stimulate the when flies were grown at +25˚C, but it was not possible to grow Drosophila IMD pathway. Immunity 20: 637–649. 10. Leulier, F., C. Parquet, S. Pili-Floury, J. H. Ryu, M. Caroff, W. J. Lee, D. Mengin- the flies at +29˚C through the development. Also, no progeny Lecreulx, and B. Lemaitre. 2003. The Drosophila immune system detects bacteria eclosed when using the fatbody-specific Fb-GAL4 driver (68) through specific peptidoglycan recognition. Nat. Immunol. 4: 478–484. 11. Choe, K. M., T. Werner, S. Sto¨ven, D. Hultmark, and K. V. Anderson. 2002. (http://flybase.org/reports/FBti0013267). To maximize the effect Requirement for a peptidoglycan recognition protein (PGRP) in Relish activation of the UAS-GAL4 based RNAi, adult flies were placed at +29˚C and antibacterial immune responses in Drosophila. Science 296: 359–362. 12 THE BAP COMPLEX IN DROSOPHILA IMMUNITY AND METABOLISM

12. Gottar, M., V. Gobert, T. Michel, M. Belvin, G. Duyk, J. A. Hoffmann, 39. Rosetto, M., Y. Engstro¨m, C. T. Baldari, J. L. Telford, and D. Hultmark. 1995. D. Ferrandon, and J. Royet. 2002. The Drosophila immune response against Signals from the IL-1 receptor homolog, Toll, can activate an immune re- Gram-negative bacteria is mediated by a peptidoglycan recognition protein. sponse in a Drosophila hemocyte cell line. Biochem. Biophys. Res. Commun. Nature 416: 640–644. 209: 111–116. 13. Ra¨met, M., P. Manfruelli, A. Pearson, B. Mathey-Prevot, and R. A. Ezekowitz. 40. Brand, A. H., and N. Perrimon. 1993. Targeted gene expression as a means of 2002. Functional genomic analysis of phagocytosis and identification of a altering cell fates and generating dominant phenotypes. Development 118: Drosophila receptor for E. coli. Nature 416: 644–648. 401–415. 14. Michel, T., J. M. Reichhart, J. A. Hoffmann, and J. Royet. 2001. Drosophila Toll 41. Kambris, Z., S. Brun, I. H. Jang, H. J. Nam, Y. Romeo, K. Takahashi, W. J. Lee, is activated by Gram-positive bacteria through a circulating peptidoglycan rec- R. Ueda, and B. Lemaitre. 2006. Drosophila immunity: a large-scale in vivo ognition protein. Nature 414: 756–759. RNAi screen identifies five serine proteases required for Toll activation. Curr. 15. Gobert, V., M. Gottar, A. A. Matskevich, S. Rutschmann, J. Royet, M. Belvin, Biol. 16: 808–813. J. A. Hoffmann, and D. Ferrandon. 2003. Dual activation of the Drosophila toll 42. Hrdlicka, L., M. Gibson, A. Kiger, C. Micchelli, M. Schober, F. Scho¨ck, and pathway by two pattern recognition receptors. Science 302: 2126–2130. N. Perrimon. 2002. Analysis of twenty-four Gal4 lines in Drosophila mela- 16. Gottar, M., V. Gobert, A. A. Matskevich, J. M. Reichhart, C. Wang, T. M. Butt, nogaster. Genesis 34: 51–57. M. Belvin, J. A. Hoffmann, and D. Ferrandon. 2006. Dual detection of fungal 43. Valanne, S., T. S. Salminen, M. Ja¨rvela¨-Sto¨lting, L. Vesala, and M. Ra¨met. 2019. infections in Drosophila via recognition of glucans and sensing of virulence Immune-inducible non-coding RNA molecule lincRNA-IBIN connects immu- factors. Cell 127: 1425–1437. nity and metabolism in Drosophila melanogaster. [Published erratum appears in 17. DeLotto, Y., and R. DeLotto. 1998. Proteolytic processing of the Drosophila 2019 PLoS Pathog. 15: e1008088.] PLoS Pathog. 15: e1007504. Spa¨tzle protein by easter generates a dimeric NGF-like molecule with ventral- 44. Smyth, G. K. 2005. Limma: linear models for microarray data. In Bio- ising activity. Mech. Dev. 72: 141–148. informatics and computational biology solutions using R and Bioconductor. 18. Jang, I. H., N. Chosa, S. H. Kim, H. J. Nam, B. Lemaitre, M. Ochiai,Z.Kambris, R. Gentleman, V. Carey, S. Dudoit, R. Irizarry, and W. Huber, eds. Springer, S. Brun, C. Hashimoto, M. Ashida, et al. 2006. A Spa¨tzle-processing enzyme required New York, p. 397–420. for toll signaling activation in Drosophila innate immunity. Dev. Cell 10: 45–55. 45.Valanne,S.,H.Myllyma¨ki, J. Kallio, M. R. Schmid, A. Kleino, A. Muruma¨gi, 19. Dushay, M. S., B. Asling, and D. Hultmark. 1996. Origins of immunity: relish, a L. Airaksinen, T. Kotipelto, M. Kaustio, J. Ulvila, et al. 2010. Genome-wide compound Rel-like gene in the antibacterial defense of Drosophila. Proc. Natl. RNA interference in Drosophila cells identifies G protein-coupled receptor Acad. Sci. USA 93: 10343–10347. kinase 2 as a conserved regulator of NF-kappaB signaling. J. Immunol. 184: 20. Ip, Y. T., M. Reach, Y. Engstro¨m, L. Kadalayil, H. Cai, S. Gonza´lez-Crespo, 6188–6198. Downloaded from K. Tatei, and M. Levine. 1993. Dif, a dorsal-related gene that mediates an im- 46. Braun, A., B. Lemaitre, R. Lanot, D. Zachary, and M. Meister. 1997. Drosophila mune response in Drosophila. Cell 75: 753–763. immunity: analysis of larval hemocytes by P-element-mediated enhancer trap. 21. Reichhart, J. M., P. Georgel, M. Meister, B. Lemaitre, C. Kappler, and Genetics 147: 623–634. J. A. Hoffmann. 1993. Expression and nuclear translocation of the rel/NF- 47. Ruhf, M. L., A. Braun, O. Papoulas, J. W. Tamkun, N. Randsholt, and kappa B-related morphogen dorsal during the immune response of Drosophila. M. Meister. 2001. The domino gene of Drosophila encodes novel members of the C. R. Acad. Sci. III 316: 1218–1224. SWI2/SNF2 family of DNA-dependent ATPases, which contribute to the si- 22. Tanji, T., E. Y. Yun, and Y. T. Ip. 2010. Heterodimers of NF-kappaB transcription lencing of homeotic genes. Development 128: 1429–1441.

factors DIF and Relish regulate antimicrobial peptide genes in Drosophila. Proc. 48. Chen, J., and E. J. Wagner. 2010. snRNA 39 end formation: the dawn of the http://www.jimmunol.org/ Natl. Acad. Sci. USA 107: 14715–14720. Integrator complex. Biochem. Soc. Trans. 38: 1082–1087. 23. Tanji, T., X. Hu, A. N. Weber, and Y. T. Ip. 2007. Toll and IMD pathways 49. Tsuzuki, S., H. Matsumoto, S. Furihata, M. Ryuda, H. Tanaka, E. J. Sung, synergistically activate an innate immune response in Drosophila melanogaster. G. S. Bird, Y. Zhou, S. B. Shears, and Y. Hayakawa. 2014. Switching between Mol. Cell. Biol. 27: 4578–4588. humoral and cellular immune responses in Drosophila is guided by the cytokine 24. Fernando, M. D., I. Kounatidis, and P. Ligoxygakis. 2014. Loss of Trabid, a new GBP. Nat. Commun. 5: 4628. negative regulator of the drosophila immune-deficiency pathway at the level of 50. Collins, R. T., T. Furukawa, N. Tanese, and J. E. Treisman. 1999. Osa associates TAK1, reduces life span. PLoS Genet. 10: e1004117. with the Brahma chromatin remodeling complex and promotes the activation of 25. Dantoft, W., D. Lundin, S. S. Esfahani, and Y. Engstro¨m. 2016. The POU/Oct some target genes. EMBO J. 18: 7029–7040. transcription factor Pdm1/nub is necessary for a beneficial gut microbiota and 51. Mohrmann, L., K. Langenberg, J. Krijgsveld, A. J. Kal, A. J. Heck, and normal lifespan of Drosophila. J. Innate Immun. 8: 412–426. C. P. Verrijzer. 2004. Differential targeting of two distinct SWI/SNF- 26. Geisler, R., A. Bergmann, Y. Hiromi, and C. Nu¨sslein-Volhard. 1992. cactus, a related Drosophila chromatin-remodeling complexes. Mol. Cell. Biol. 24:

gene involved in dorsoventral pattern formation of Drosophila, is related to the I 3077–3088. by guest on October 1, 2021 kappa B gene family of vertebrates. Cell 71: 613–621. 52. Moshkin, Y. M., L. Mohrmann, W. F. van Ijcken, and C. P. Verrijzer. 2007. 27. Kleino, A., H. Myllyma¨ki, J. Kallio, L. M. Vanha-aho, K. Oksanen, J. Ulvila, Functional differentiation of SWI/SNF remodelers in transcription and cell cycle D. Hultmark, S. Valanne, and M. Ra¨met. 2008. Pirk is a negative regulator of the control. Mol. Cell. Biol. 27: 651–661. Drosophila Imd pathway. J. Immunol. 180: 5413–5422. 53. Paredes, J. C., D. P. Welchman, M. Poidevin, and B. Lemaitre. 2011. Negative 28. Maillet, F., V. Bischoff, C. Vignal, J. Hoffmann, and J. Royet. 2008. The Dro- regulation by amidase PGRPs shapes the Drosophila antibacterial response and sophila peptidoglycan recognition protein PGRP-LF blocks PGRP-LC and IMD/ protects the fly from innocuous infection. Immunity 35: 770–779. JNK pathway activation. Cell Host Microbe 3: 293–303. 54. Rogers, W. A., S. Grover, S. J. Stringer, J. Parks, M. Rebeiz, and T. M. Williams. 29. Morris, O., X. Liu, C. Domingues, C. Runchel, A. Chai, S. Basith, T. Tenev, 2014. A survey of the trans-regulatory landscape for Drosophila melanogaster H. Chen, S. Choi, G. Pennetta, et al. 2016. Signal integration by the IkB protein abdominal pigmentation. Dev. Biol. 385: 417–432. pickle shapes Drosophila innate host defense. Cell Host Microbe 20: 283–295. 55. Sekyrova, P., D. Bohmann, M. Jindra, and M. Uhlirova. 2010. Interaction be- 30. Myllyma¨ki, H., and M. Ra¨met. 2013. Transcription factor zfh1 downregulates tween Drosophila bZIP proteins Atf3 and Jun prevents replacement of epithelial Drosophila Imd pathway. Dev. Comp. Immunol. 39: 188–197. cells during metamorphosis. Development 137: 141–150. 31. Zaidman-Re´my, A., M. Herve´, M. Poidevin, S. Pili-Floury, M. S. Kim, 56. Ross, J., H. Jiang, M. R. Kanost, and Y. Wang. 2003. Serine proteases and their D. Blanot, B. H. Oh, R. Ueda, D. Mengin-Lecreulx, and B. Lemaitre. 2006. The homologs in the Drosophila melanogaster genome: an initial analysis of se- Drosophila amidase PGRP-LB modulates the immune response to bacterial in- quence conservation and phylogenetic relationships. Gene 304: 117–131. fection. Immunity 24: 463–473. 57. Cao, X., and H. Jiang. 2018. Building a platform for predicting functions of 32. Kallio, J., A. Leinonen, J. Ulvila, S. Valanne, R. A. Ezekowitz, and M. Ra¨met. serine protease-related proteins in Drosophila melanogaster and other insects. 2005. Functional analysis of immune response genes in Drosophila identifies Insect Biochem. Mol. Biol. 103: 53–69. JNK pathway as a regulator of antimicrobial peptide gene expression in S2 cells. 58. Lindsay, S. A., S. J. H. Lin, and S. A. Wasserman. 2018. Short-form bomanins Microbes Infect. 7: 811–819. mediate humoral immunity in Drosophila. J. Innate Immun. 10: 306–314. 33. Salminen, T. S., and M. Ra¨met. 2016. Pickle flavors relish in Drosophila im- 59. Cohen, L. B., S. A. Lindsay, Y. Xu, S. J. H. Lin, and S. A. Wasserman. 2020. The munity. Cell Host Microbe 20: 273–274. Daisho peptides mediate Drosophila defense against a subset of filamentous 34. Ji, S., M. Sun, X. Zheng, L. Li, L. Sun, D. Chen, and Q. Sun. 2014. Cell-surface fungi. Front. Immunol. 11: 9. localization of Pellino antagonizes Toll-mediated innate immune signalling by 60. Zaidman-Re´my, A., M. Poidevin, M. Herve´, D. P. Welchman, J. C. Paredes, controlling MyD88 turnover in Drosophila. Nat. Commun. 5: 3458. C. Fahlander, H. Steiner, D. Mengin-Lecreulx, and B. Lemaitre. 2011. Dro- 35. Bonnay, F., X. H. Nguyen, E. Cohen-Berros, L. Troxler, E. Batsche, J. Camonis, sophila immunity: analysis of PGRP-SB1 expression, enzymatic activity and O. Takeuchi, J. M. Reichhart, and N. Matt. 2014. Akirin specifies NF-kB se- function. PLoS One 6: e17231. lectivity of Drosophila innate immune response via chromatin remodeling. 61. Iatsenko, I., S. Kondo, D. Mengin-Lecreulx, and B. Lemaitre. 2016. PGRP-SD, EMBO J. 33: 2349–2362. an extracellular pattern-recognition receptor, enhances peptidoglycan-mediated 36. Ku¨ttenkeuler, D., N. Pelte, A. Ragab, V. Gesellchen, L. Schneider, C. Blass, activation of the Drosophila imd pathway. Immunity 45: 1013–1023. E. Axelsson, W. Huber, and M. Boutros. 2010. A large-scale RNAi screen 62. Sto¨ven, S., I. Ando, L. Kadalayil, Y. Engstro¨m, and D. Hultmark. 2000. Acti- identifies Deaf1 as a regulator of innate immune responses in Drosophila. vation of the Drosophila NF-kappaB factor Relish by rapid endoproteolytic J. Innate Immun. 2: 181–194. cleavage. EMBO Rep. 1: 347–352. 37. Kleino, A., S. Valanne, J. Ulvila, J. Kallio, H. Myllyma¨ki, H. Enwald, S. Sto¨ven, 63. Khush, R. S., W. D. Cornwell, J. N. Uram, and B. Lemaitre. 2002. A ubiquitin- M. Poidevin, R. Ueda, D. Hultmark, et al. 2005. Inhibitor of apoptosis 2 and proteasome pathway represses the Drosophila immune deficiency signaling TAK1-binding protein are components of the Drosophila Imd pathway. EMBO J. cascade. Curr. Biol. 12: 1728–1737. 24: 3423–3434. 64. Wiklund, M. L., S. Steinert, A. Junell, D. Hultmark, and S. Sto¨ven. 2009. The 38. Ra¨met, M., A. Pearson, P. Manfruelli, X. Li, H. Koziel, V. Go¨bel, E. Chung, N-terminal half of the Drosophila Rel/NF-kappaB factor Relish, REL-68, M. Krieger, and R. A. Ezekowitz. 2001. Drosophila scavenger receptor CI is a constitutively activates transcription of specific Relish target genes. Dev. pattern recognition receptor for bacteria. Immunity 15: 1027–1038. Comp. Immunol. 33: 690–696. The Journal of Immunology 13

65. Meier, K., and A. Brehm. 2014. Chromatin regulation: how complex does it get? factor Relish in the Drosophila innate immune response. J. Biol. Chem. 289: Epigenetics 9: 1485–1495. 20470–20476. 66. Kal, A. J., T. Mahmoudi, N. B. Zak, and C. P. Verrijzer. 2000. The Drosophila 70. Nowak, S. J., H. Aihara, K. Gonzalez, Y. Nibu, and M. K. Baylies. 2012. Akirin brahma complex is an essential coactivator for the trithorax group protein zeste. links twist-regulated transcription with the Brahma chromatin remodeling Genes Dev. 14: 1058–1071. complex during embryogenesis. PLoS Genet. 8: e1002547. 67. Kingston, R. E., and G. J. Narlikar. 1999. ATP-dependent remodeling and 71. Fukuyama, H., Y. Verdier, Y. Guan, C. Makino-Okamura, V. Shilova, X. Liu, acetylation as regulators of chromatin fluidity. Genes Dev. 13: 2339–2352. E. Maksoud, J. Matsubayashi, I. Haddad, K. Spirohn, et al. 2013. Landscape of 68. Gro¨nke, S., M. Beller, S. Fellert, H. Ramakrishnan, H. Ja¨ckle, and protein-protein interactions in Drosophila immune deficiency signaling during R. P. Ku¨hnlein. 2003. Control of fat storage by a Drosophila PAT domain protein. bacterial challenge. Proc. Natl. Acad. Sci. USA 110: 10717–10722. Curr. Biol. 13: 603–606. 72. He, X., J. Yu, M. Wang, Y. Cheng, Y. Han, S. Yang, G. Shi, L. Sun, Y. Fang, 69. Goto, A., H. Fukuyama, J. L. Imler, and J. A. Hoffmann. 2014. The chromatin S. T. Gong, et al. 2017. Bap180/Baf180 is required to maintain homeostasis of in- regulator DMAP1 modulates activity of the nuclear factor B (NF-B) transcription testinal innate immune response in Drosophila and mice. Nat. Microbiol. 2: 17056. Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021 The Journal of Immunology 917

Corrections

Valanne, S., M. Ja¨rvela¨-Sto¨lting, S.-K. E. Harjula, H. Myllyma¨ki, T. S. Salminen, and M. Ra¨met. 2020. Osa-containing Brahma complex regulates innate immunity and the expression of metabolic genes in Drosophila. J. Immunol. 204: 2143–2155.

The keys for the graph in Fig. 2J and 2K were incorrect as originally published. The key for Fig. 2J should have indicated a black solid line for C564.osaIR (#7810), a black dotted line for C564, w1118, and a gray line for w1118, osaIR (#7810). Similarly, the key for Fig. 2K should have indicated a black solid line for C564.osaIR (#7810), a black dotted line for C564, w1118, a gray line for w1118, osaIR (#7810), and a black dashed line for RelishE20.

The corrected version of Fig. 2 is shown below. The figure legend is correct as published and is shown below for reference. The figure has been corrected in the online version of the article, which now differs from the print version as originally published.

Copyright Ó 2021 by The American Association of Immunologists, Inc. 0022-1767/21/$37.50 918 CORRECTIONS

FIGURE 2. Knockdown of osa increases the Toll pathway–mediated antimicrobial peptide expression and survival in Drosophila in vivo. (A–F) Expression of the selected Toll pathway target genes was monitored in osa and brm knockdown flies and controls with and without M. luteus infection for 12 or 24 h. In each graph, the expression value of w, osaIR flies infected for 12 h is set to 1. MyD88 RNAi was used as a control for Toll pathway function, and it prevents Toll pathway induction. Cact RNAi causes induction of the Toll pathway also without infection. (A and B) BomS1,(C and D) BomS2, and (E and F) Drs expression. (G–I) Spz overexpression (C564.SpzOE) cau- ses induction of Toll pathway targets; MyD88 RNAi prevents this induction. In each graph, the expression value of C564.SpzOE flies is set to 1. The expression of the Toll pathway target genes is further enhanced upon silencing of osa (no. 7810) also in this setting. (G) BomS1,(H) BomS2,(I) Drs. Statistical analy- ses of differences in gene expression in the qRT-PCR measurements were carried out us- ing Student t test for two samples assuming equal variances. (J) Knockdown of osa us- ing the no. 7810 RNAi line crossed over the C564-GAL4 driver increases survival from M. luteus 1 E. faecalis infection. (K) osa knockdown (C564.osaIR [no. 7810]) has no significant effect on survival upon E. cloacae infection. Statistical analyses of survival ex- periments were carried out using the log-rank (Mantel–Cox) test.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.2001320 Table S1. Primers for dsRNA production and qRT-PCR. Primers for dsRNA design and synthesis are indicated as I = outer primers, II = nested primers with the T7 binding site.

Primer/ prod. Gene Forward, 5’3’ Reverse, 5’3’ (bp) Target dsRNA primers GFP T7 + GCTCGGGAGATCTCC T7 + CTAGACTCGAGCGGC 777 dsRNA synthesis cactus (I) GTTGATAGAACTGCTCCC AGCAGCGGAGGCAGCAAC 659 nested PCR cactus (II) T7 + AGCGGATGATGTTGC T7 + GACTGCAGCTGCAGC 573 dsRNA synthesis brm (I) GGACATGTCGCACTTGATG GATTGGATCGACGACGAG 1002 nested PCR brm (II) T7 + CTTGTGCAGACGACG T7 + TGTACCACTCTCCAC 484 dsRNA synthesis Bap111 (I) ACCAAGTCGTCGAGCAGC CGCATTCGCATTCGGTTC 795 nested PCR Bap111 (II) T7 + AAGGCTACCCAAGCC T7 + CCAAGTGCTTGGTCG 410 dsRNA synthesis mor (I) CAACAGCAGAGTCTCCTG GCGACAACGACATGATGC 848 nested PCR mor (II) T7 + GGTCCATTCACGAGC T7 + GTGATCGAGAAACGGG 573 dsRNA synthesis osa A (I) CAGGCGCAGACAGACAAGAC GGATGCTGCTGTTGCTGCTG 646 nested PCR osa A (II) T7 + GCGCAGACAGACAAG T7 + GTTGTTGGGTGTGGG 473 dsRNA synthesis osa B (I) CATCCGGCGAAGATCCACA TGAGTTCGTTGTTCCAGCATCC 683 nested PCR osa B (II) T7 + CAGTTGTGATGCATCC T7 + CCAAATGTCCACCAC 567 dsRNA synthesis BAP60 (I) GCAACGCTTTGCACCTGG CCAGCTCGATAACCAGCGAC 731 nested PCR BAP60 (II) T7 + AAGCGCCGGTGCAGT T7 + ACGACGAAAACTTGCG 684 dsRNA synthesis Snr1 (I) TACGTGGGCTCGGAAGTGG GTTCAGCTTAACAATGACCCGC 690 nested PCR Snr1 (II) T7 + ACATGCGCCACTTCC T7 + CAGGTCTCCTCGAGGA 637 dsRNA synthesis Rel (I) CCAGCACCAGTGGCTATAGC GCTCATCGTTGCCCATCACC 916 nested PCR Rel (II) T7 + GCAAACGGACTTCGC T7 + CTCACGCTCTGTCTC 635 dsRNA synthesis PB (I) CCTCGTACTATGACGTGGTGGT CCGAGCTGTCAACCTCCTC 729 nested PCR PB (II) T7 + CTGATGGCCGACCTG T7 + CTCCAGCTCAATACGC 555 dsRNA synthesis dom (I) TACCACCTGGCTATCTGG GAGCATCACAGACGAAGG 815 nested PCR dom (II) T7 + CTGATGACAGCGATG T7 + GTGCTGGTACTCACG 569 dsRNA synthesis Pvr (I) GGATTGGGTCGGCAGTAG CACCACGCTTAAGTGCG 679 nested PCR Pvr (II) T7 + TCCACGATGCTTCGG T7 + GATGAGCTGGTGGAC 619 dsRNA synthesis qRT-PCR primers Drs ATGATGCAGATCAAGTACTTG GCATCCTTCGCACCAGC 210 qRT-PCR IM1 CTCGGTCTGCTGGCTGTGGC CCGTGGACATTGCACACCC 95 qRT-PCR IM2 CTTCTCAGTCGTCACCGTCC TCCACCGTGCACATTGCAG 124 qRT-PCR

RpL32 GGTTACGGATCGAACAAGCG TTCTGCATGAGCAGGACCTC 101 qRT-PCR, control gene ND-39 ACCGACAAGGTTCTGACTGG CTCCGCTTAGGCAAACAGAC 201 qRT-PCR, control gene IBIN CAACTGCTGCCAATCCTCG GCCTGGGATCGTAGTCACTT 103 qRT-PCR Mal-A6 AACACTCAGCTCAATGCGACG CATCTGCAGCATATTCATCAAGTCG 165 qRT-PCR Mal-A7 GAAAATCTATAAGGCGCTGGTGG GGATGCAGAGCCGGAGATG 127 qRT-PCR Mal-A8 CACTGCCTCCGCTTTTTGAG CGTGGTGGTCAGATAGTCGC 110 qRT-PCR AttC CATCGTTGGCGTACTTGGC TTGCTGGAAGCTATCCCGC 353 qRT-PCR DptB GACTGGCTTGTGCCTTC CCTGAAGGTATACACTCC 326 qRT-PCR CecA1 CGTCGCTCTCATTCTGGC GTTGCGGCGACATTGGC 153 qRT-PCR Rel CCAATATGCCATTGTGTGC GTCGGCGTTTCCTCGAC 167 qRT-PCR

Fig. S1: Members of the BAP complex regulate NF-κB-mediated immunity in Drosophila S2 cells in vitro. A) Silencing of the BAP complex members elevates the SpzC106-induced Drs-luc activity (gray bars). Double knockdown of osa and other BAP complex members does not further increase the Drs-luc signal (black bars). B) Expression of E. coli induced endogenous DptB is reduced upon silencing of the BAP complex members. C) Expression of E. coli induced endogenous CecA1 is elevated upon silencing of the BAP complex members.

Figure S2: Silencing osa in Drosophila larvae and adults using the C564-GAL4 driver causes some problems in the fly development. A) osa silencing using the osaIR (#7810) line causes incomplete fusion of the abdominal epidermis in some C564> osaIR (#7810) flies (red arrow). B) Many C564>osaIR (#330350) flies have a deformed wing phenotype (upper red arrow), edema and deformations in the striping pattern of the abdomen (lower red arrow). C) Expression of chitin metabolism and cuticle development related genes is enhanced upon osa knockdown in C564> osaIR (#7810) flies.