This is a self-archived – parallel published version of an original article. This version may differ from the original in pagination and typographic details. When using please cite the original. AUTHOR Syed Bilal Ahmad Andrabia, Kedar Batkulwara, Santosh D. Bhosalea, Robert Mouldera, Meraj Hasan Khana, Tanja Buchachera, Mohd Moin Khana, Ilona Arnkila, Omid Rasoola, Alexander Marsone, Ubaid Ullah Kalima, Riitta Lahesmaa TITLE HIC1 interacts with FOXP3 multi protein complex: Novel pleiotropic mechanisms to regulate human regulatory T cell differentiation and function YEAR 2023 DOI https://doi.org/10.1016/j.imlet.2023.09.001 VERSION Author’s accepted manuscript COPYRIGHT License: CC BY NC ND CITATION Andrabi, S.B.A., Batkulwar, K., Bhosale, S.D., Moulder, R., Khan, M.H., Buchacher, T., Khan, M.M., Arnkil, I., Rasool, O., Marson, A., Kalim, U.U., Lahesmaa, R., 2023. HIC1 interacts with FOXP3 multi protein complex: Novel pleiotropic mechanisms to regulate human regulatory T cell differentiation and function. Immunology Letters 263, 123–132. 1 HIC1 interacts with FOXP3 multi protein complex: novel pleiotropic mechanisms to 1 regulate human regulatory T cell differentiation and function 2 3 Syed Bilal Ahmad Andrabia,b*, Kedar Batkulwara,b*, Santosh D. Bhosalea,c, Robert Mouldera,b, 4 Meraj Hasan Khana,b, Tanja Buchachera,b, Mohd Moin Khana,b, Ilona Arnkila,b, Omid Rasoola,b, 5 Alexander Marsone,f, Ubaid Ullah Kalima,b, Riitta Lahesmaaa,b,d# 6 7 aTurku Bioscience Centre, University of Turku and Åbo Akademi University; Turku 20520, 8 Finland 9 bInFLAMES Research Flagship Center, University of Turku 10 cPrecision Biomarker Laboratories, Cedars-Sinai Medical Center, Los Angeles, CA 90048, 11 USA 12 dInstitute of Biomedicine, University of Turku. 13 eGladstone-UCSF Institute of Genomic Immunology, San Francisco, CA 94158, USA. 14 fDepartment of Medicine, University of California San Francisco, San Francisco, CA 94143, 15 USA. 16 17 *These authors contributed equally 18 19 #Correspondence 20 Professor Riitta Lahesmaa 21 Turku Bioscience Centre, 22 Tykistökatu 6A, Turku 20520, FINLAND 23 Email: rilahes@utu.fi 24 Corresponding author: rilahes@utu.fi 25 26 27 28 29 30 31 32 33 2 Highlights 34 • Systematic characterization of HIC1 interactome in regulatory T cells by Affinity 35 Purification-Mass Spectrometry 36 • HIC1 binds to the RUNX1 promoter and regulates its expression 37 • HIC1- a part of FOXP3-RUNX1-CBFB transcriptional complex 38 Abstract 39 Transcriptional repressor, hypermethylated in cancer 1 (HIC1) participates in a range of 40 important biological processes, such as tumor repression, immune suppression, embryonic 41 development and epigenetic gene regulation. Further to these, we previously demonstrated that 42 HIC1 provides a significant contribution to the function and development of regulatory T 43 (Treg) cells. However, the mechanism by which it regulates these processes was not apparent. 44 To address this question, we used affinity-purification mass spectrometry to characterize the 45 HIC1 interactome in human Treg cells. Altogether 61 high-confidence interactors were 46 identified, including IKZF3, which is a key transcription factor in the development of Treg 47 cells. The biological processes associated with these interacting proteins include protein 48 transport, mRNA processing, non-coding (ncRNA) transcription and RNA metabolism. The 49 results revealed that HIC1 is part of a FOXP3-RUNX1-CBFB protein complex that regulates 50 Treg signature genes thus improving our understanding of HIC1 function during early Treg 51 cell differentiation. 52 Graphical abstract 53 54 55 56 Keywords 57 FOXP3; HIC1; Immunoprecipitation; Interactome; iTreg; RUNX1. 58 59 60 61 62 3 Abbreviations 63 BCOR- BCL6 corepressor 64 CBFB- Core-binding factor subunit beta 65 ChIPseq- Chromatin immunoprecipitation sequencing 66 CTD- Carboxy-terminal domain 67 FOXP3- Forkhead box P3 68 GATA3- GATA Binding Protein 3 69 GO- Gene ontology 70 HCD- Higher energy collisional dissociation 71 HIC1- Hypermethylated in cancer 1 72 IKZF3- IKAROS Family Zinc Finger 3 73 IP- Immunoprecipitation 74 iTreg- In-vitro induced Treg cells 75 LFQ- Label-free quantification 76 MS- Mass spectrometry 77 m/z- Mass-to-charge ratio 78 PCR- Polymerase chain reaction 79 PLA- Proximity ligation assay 80 RUNX1- Runt-related transcription factor 1 81 SRM- Selected reaction monitoring 82 TBX21- T-Box transcription factor 21 83 TF- Transcription factor 84 Treg- Regulatory T cells 85 4 1. Introduction 86 Regulation of the immune system is critical for controlling autoimmunity and fighting cancer 87 [1]. Central to these processes, regulatory T (Treg) cells maintain immune tolerance and the 88 balance between pro- and anti-inflammatory responses [2]. The majority of Treg cells are 89 produced in the thymus (tTreg), while some acquire regulatory phenotype in the periphery 90 (pTreg) [3]. Treg cells can also be induced in vitro (iTreg) from naïve CD4+ T cells through 91 their activation in the presence of Treg cell polarizing cytokines [4]. The lineage-specific 92 transcription factor (TF) FOXP3 is essential for Treg cell differentiation, stability, and function 93 [5]. Besides FOXP3, several other TFs, e.g., IKZFs, NR4As, C-REL, NFAT, SMAD factors, 94 STAT5, and RUNX1 act in concert and play an important role in Treg cell differentiation [6–95 12]. 96 Recently, we characterized the role of a novel TF hypermethylated in cancer 1 (HIC1), in 97 human iTreg cell differentiation and showed that it contributes to suppressive function and 98 lineage specification of Treg cells without influencing expression of FOXP3 [13]. HIC1 is a 99 member of the Kruppel/Zinc Finger and BTB (POK/ZBTB) protein family and its expression 100 was reduced in various types of cancer [14]. In iTreg cells, HIC1 deficiency led to a 101 considerable loss of suppressive capability with a concomitant increase in the expression of 102 effector T cell associated genes including GATA3, TBX21 and IFNG [13]. However, the 103 molecular mechanisms by which HIC1 regulates the suppressive capacity of iTreg cells, were 104 lacking. As a follow up, to address this deficit, we comprehensively studied the protein 105 interaction network of HIC1 and validated the key interactors. The results indicate that HIC1 106 is a vital part of a protein complex that regulates Treg signature genes. This study sheds light 107 on the mechanism of HIC1 action during human Treg cell differentiation. 108 109 2. Materials and methods 110 2.1 CD4+ cell isolation and differentiation to iTreg Cells. 111 CD4+ T cells were isolated from human umbilical cord blood as described previously [13]. 112 Briefly, umbilical cord blood was layered on Ficoll (GE Healthcare, Cat# 17-1440-03) to 113 isolate white blood cells and the Dynal bead CD4+ isolation kit (Invitrogen, Cat# 11331D) was 114 used to isolate CD4+ T cells. CD25 MicroBeads II and LD columns (Miltenyi Biotec, Cat# 115 130-092-985 and Cat# 130042901, respectively) were used for selection of CD25– T-cells. 116 CD4+ CD25– cells from single donor or pool of multiple donors were used. For iTreg cultures, 117 cells were activated with plate-bound anti-CD3 (500 ng/well of 24-well culture plate; Beckman 118 5 Coulter, REF# IM-1304) and soluble anti-CD28 (500 ng/ml; Beckman Coulter, REF# IM1376) 119 antibodies in presence of cytokines i.e., IL-2 (12 ng/ml; R&D Systems); all-trans retinoic acid 120 (ATRA) (10 nM; Sigma-Aldrich); TGF-β1 (10 ng/ml; R&D Systems) and human serum (10%) 121 and cultured for 72 h at 37 °C in 5% CO2. 122 123 2.2 Co-immunoprecipitation (Co-IP) 124 HIC1 immunoprecipitation (IP) was performed using anti-HIC1 antibody (Santa Cruz 125 Biotechnology, Cat# sc-271499) that recognizes 611-733 C-terminus region of human HIC1, 126 together with the respective control IgG antibody, in three biological replicates. Similarly, 127 reverse IPs were performed with anti-RUNX1 antibody (Santa Cruz Biotechnology, Cat# sc-128 365644), anti-FOXP3 antibody (eBioscience, Cat# 14-4776-82), anti-IKZF3 antibody (Abcam, 129 Cat# ab139408) and anti-CBFB antibody (Abcam, Cat # 62184). The Pierce™ MS-Compatible 130 Magnetic IP Kit (ThermoFisher Scientific, Cat# 90409) was used. The respective 131 Immunoglobulin G (IgGs) were used in control IPs. Briefly, differentiated iTreg cells were 132 harvested on ice and washed with PBS (2X). Cells were then re-suspended in ice-cold IP-MS 133 cell lysis buffer, followed by incubation on ice for 20 min with periodic mixing as 134 recommended by the manufacturer. The protein lysate was centrifuged at ~13,000 g for 10 min 135 to pellet down the cell debris. The supernatant was transferred to a new microcentrifuge tube 136 and protein concentration was determined by using a DC protein assay kit (BioRad, Cat# 500-137 0116). Cell lysates (500-1000 μg) were mixed with antibodies (1:50 dilution) and the 138 antibody/lysate solution was diluted with the IP-MS cell lysis buffer and incubated overnight 139 at +4 °C to form the immune complex, according to the manufacturers protocol. On the next 140 day, the immune complex was incubated with Protein A/G magnetic beads for 1 h at room 141 temperature (RT) followed by washing the beads and elution. The eluate containing the target 142 antigen was then transferred to a new low protein-binding microcentrifuge tube and dried by 143 vacuum concentrator prior to MS sample preparation and western blot. 144 145 2.3 Western blotting 146 Cells were resuspended in RIPA buffer (ThermoFisher, Cat# 89901) and sonicated on ice for 147 5-10 min using a Bioruptor® sonicator (Diagenode). The cell lysate was centrifuged at high 148 speed (16,000 g) and protein containing supernatant was transferred to a new tube. Protein 149 concentration was determined using a DC Protein assay (Bio-Rad, Cat# 500-0116). Equal 150 amounts of protein were loaded onto acrylamide gel (Bio-Rad Mini PROTEAN® TGX precast 151 gels). For the transfer of proteins to the PVDF membrane, mini transfer packs from Bio-Rad 152 6 were used. Primary/secondary antibody incubations were performed in 5% Bovine serum 153 albumin (BSA) in TBST buffer (0.1% Tween 20). The following antibodies were used, anti-154 RUNX1 antibody (Santa Cruz Biotechnology, Cat# sc-365644); anti-CBFB antibody (Abcam, 155 Cat # 62184); anti-HIC1 antibody (Santa Cruz Biotechnology, Cat# sc-271499); anti-IKZF3 156 antibody (Abcam, Cat# ab139408); anti-FOXP3 antibody (eBioscience, Cat# 14-4776-82) and 157 anti-Beta-actin antibody (Sigma, cat # A5441). 158 159 2.4 Sample preparation for mass-spectrometry analysis 160 The immunoprecipitated proteins (from both the HIC1 and IgG control bait) were digested with 161 trypsin. Briefly, urea buffer (8 M urea, 50mM Tris-HCl pH 8.0) was added to denature the 162 proteins, followed by addition of dithiothreitol to a final concentration of 10 mM and 163 incubation at 37 °C for 1 h for reduction. The reduced disulphide bridges were subsequently 164 alkylated using iodoacetamide (~14 mM) at RT for 30 min in dark condition. The samples were 165 diluted to a urea concentration of less than 1 M and digested with sequencing grade modified 166 trypsin (0.29 µg per sample) at 37 °C overnight (16 h). The tryptic digests were acidified and 167 desalted using in house made C18 Stage Tips (3M, Cat No 2215). The eluates from the Stage 168 Tips were dried in a vacuum centrifuge (Thermo Fisher Scientific) and stored at -80 °C until 169 further analysis. 170 171 2.5 LC-MS/MS analysis 172 The desalted tryptic peptides were reconstituted in formic acid/acetonitrile mixture and the 173 peptide amounts were estimated using a NanoDrop-1000 UV spectrophotometer (Thermo 174 Fisher Scientific). The samples were analysed by LC-MS/MS using an Easy-nLC 1200 coupled 175 to Q Exactive HF mass spectrometer (Thermo Fisher Scientific). A 20 x 0.1 mm i.d. Pre-176 column coupled with a 75 µm x 150 mm analytical column (both in-house packed with 5 µm 177 Reprosil C18; Dr Maisch GmbH) were used for sample loading and separation, respectively. 178 Peptides were eluted using a gradient from 5 to 36% B in 50 min at a flow rate of 300 nl/min. 179 Tandem mass spectra were acquired in positive ion mode using HCD fragmentation of the 15 180 most intense ions from each precursor scan. The Orbitrap resolution was set to 120,000 at m/z 181 200 for the full scan MS spectra, with a maximum injection time of 100 ms and a target value 182 of 3 x 106 in the 300-1650 m/z range. The tandem mass spectra were acquired at a resolution 183 of 15,000 at 200 m/z with a target value of 5 x 104 and a maximum injection time of 150 ms. 184 The lowest fixed first mass of 120 m/z was used and to the repeatedly identified peptides were 185 excluded for 20 s. The samples were analyzed in triplicate in randomized batch. 186 7 187 2.6 Data analysis 188 The mass spectrometry (MS) raw files were analyzed using MaxQuant software version 189 1.6.0.16 [15] and searched against a human UniProt FASTA sequence database (downloaded, 190 May 2019 and containing 20415 entries) with common contaminants using the Andromeda 191 search engine [16]. The specified search criteria were for trypsin digestion with up to two 192 missed cleavages, a fixed carbamidomethyl modification of cysteine residues and variable 193 modifications of methionine oxidation and N-terminal acetylation. The false discovery rate 194 (FDR) was set to 1% at the peptide and protein level. 195 The ‘proteinGroup.txt’ table generated from the search was filtered to remove contaminants, 196 proteins only identified by site and reverse hits using Perseus 1.6.2.3 [17]. Proteins identified 197 with two or more unique peptides were retained. The protein LFQ values were transformed to 198 log2 and the median values were calculated for the technical replicates. The data was further 199 filtered to remove the influence of inconsistently detected features, based on the inclusion 200 criteria of three valid values in at least one group, i.e. IgG or HIC1. For statistical analysis of 201 the data, the mass spectrometry interaction statistics (MiST) algorithm [18] was used. The 202 algorithm compares the results from IgG control to pulldowns bait and then computes a score 203 for each protein. The MiST scoring algorithm uses the protein signal intensity, reproducibility 204 and its specificity to the bait, to calculate a score ranging from 0 to 1. The criteria of a MiST 205 score, ≥ 0.75 with the HIC1 bait and ≤ 0.75 with IgG control bait was applied to define the 206 putative protein interactors [19]. Further filtering was made relative to an in-house database of 207 proteins frequently detected with IgG baits in similar experiments (i.e. human Th cells with 208 Dynabeads). Proteins detected in more than 70% of the previous measurements were excluded 209 from the analysis. Cytoscape was used to build the protein-protein interaction (PPI) network of 210 the identified interactors, combining existing PPI data from the STRING database [20]. The 211 mass spectrometry proteomics raw data have been deposited to the ProteomeXchange 212 Consortium via the PRIDE [21] partner repository with the dataset identifier PXD039337. 213 214 2.7 Selected reaction monitoring mass spectrometry (SRM-MS) 215 Selected reaction monitoring mass spectrometry was used to validate the relative abundance of 216 BCOR, FOXP3, RUNX1, CBFB, IKZF3 and HIC1 in immuno-precipitates from iTreg cells. 217 ACTB and GAPDH were also measured as a reference protein. Heavy-labeled synthetic 218 peptides (lysine 13C6 15N2 and arginine 13C6 15N4) were obtained for the targets of interest 219 (PEPotec, Grade 2, Thermo Fischer Scientific). For these validations, CD4+ CD25– cells 220 8 isolated from 3-5 donors were pooled and differentiated into iTreg cells followed by 221 Immunoprecipitation. Skyline software [22] was used to develop the SRM method, check the 222 peak integration, normalize the data and for statistical analysis. 223 The samples were prepared using the same digestion and desalting protocols used for 224 discovery. These were then spiked with synthetic heavy labelled analogues of the peptide 225 targets and a retention time standard (MSRT1, Sigma) for scheduled SRM. The LC-MS/MS 226 analyses were conducted using Easy-nLC 1000 liquid chromatography (Thermo Scientific) 227 coupled to a TSQ Vantage Triple Quadrupole mass spectrometer (Thermo Scientific). The 228 column configuration included a 20 x 0.1 mm i.d. pre-column in conjunction with a 150 mm x 229 75 µm i.d. analytical column, both packed with 5 µm Reprosil C18-bonded silica (Dr Maisch 230 GmbH). A separation gradient was used from 5% to 21% B in 11 min, then to 36% B in 9 min, 231 to 100% in 2 min, then ending with an 8 min isocratic period. A flow rate of 300 nl/min was 232 used, with the mobile phase compositions as indicate above. The raw SRM data are available 233 through Panorama (https://doi.org/10.1074/mcp.RA117.000543) with the dataset identifier 234 PXD038532. The estimated injected amount was 250 ng of endogenous sample, spiked with 235 50 fmol of heavy labelled peptides. Measured peptides used are listed in Table S1. 236 237 2.8 Proximity ligation assay (PLA) 238 PLA assay was performed following the manufacturer's protocol (Duolink®PLA, Sigma). The 239 Treg cells were fixed with 4% paraformaldehyde for 15 min at RT and then plated on Poly-L-240 lysine coated (10 μg/ml) coverslips using a cytospin for 5 min at 800 RPM. The cells were 241 permeabilized for 10 min with PBS containing 0.5% Triton X-100 at RT. A blocking solution 242 was added to the cells for 30 min at 37 °C, followed by incubation with primary antibodies (in 243 blocking solution) anti-HIC1 (Santa Cruz Biotechnology, Cat# sc-271499), anti-IKZF3 244 (Abcam, Cat# ab139408), anti- CBFB (Cell Signaling Technology, Cat# 62184), anti-FOXP3 245 (Invitrogen, Cat# PAI-806), anti-RUNX1 (Santa Cruz Biotechnology, Cat# sc-365644), anti-246 GFP (mouse) (Abcam, Cat# ab1218) and anti-GFP (rabbit) (Invitrogen Cat# A11122). After 247 washing with buffer A, PLA probes were incubated for 1 h at 37 °C, followed by a ligase 248 reaction step performed for 30 min at 37 °C. In the final step, polymerase solution was added 249 and the cells were incubated for 100 min at 37 °C for the amplification. All the incubations 250 were performed in a preheated humidity chamber. Post-amplification, the coverslips were 251 washed with buffer B from the PLA kit and mounted with Vectashield having DAPI (Vector 252 Laboratories). The PLA signal was detected by using a confocal microscope 3iCSU-W1 253 spinning disc microscope equipped with a 100x (NA 1.4 oil, Plan-Apochromat, M27) objective 254 9 and Evolve 512 EMCCD camera (Photometrics). PLA signals per cell were calculated by 255 dividing the amount of PLA signal dots in one field of view, determined using Cell Profiler 256 software [23]. 257 258 2.9 CRISPR-Cas9 mediated HIC1 ablation 259 The in vitro assembly of guide RNA (gRNA) with the Cas9 protein was carried out as described 260 previously [24]. Briefly, 80 μM of gRNA reagents were prepared by combining equimolar 261 amounts (1:1) of HIC1 targeting (5´-GCATGACAACCTGCTCAACC-3´) or non-targeting 262 control (NT) (5´-CGTTAATCGCGTATAATACG-3´) crisprRNA (crRNA) with trans-263 activating crispr RNA (tracrRNA) scaffold (both synthesized by IDT) followed by incubation 264 at 37 oC for 30 min. Assembled 80 μM gRNA was then mixed with equal volume of 40 μM 265 recombinant S. pyogenes Cas9-nuclear localization sequence (NLS) purified protein (QB3 266 Macro Lab, University of California, Berkeley) (2:1 gRNA to Cas9 molar ratio) and 1μl of 100 267 μM Ultramer DNA oligonucleotide enhancer (IDT), followed by incubation for 10 min at 37 268 °C for a final concentration of 20 μM CRISPR-Cas9 ribonucleoprotein (RNP). Freshly purified 269 CD4+ CD25– cells from 3 donors were resuspended in Opti-MEM™ (Gibco by Life 270 Technologies, Cat# 31985-047) and transfected with RNP complexes using Nucleofector 2C 271 system (Lonza) (4 × 106 cells per cuvette in 100 μL of Opti-MEM using U-014 nucleofection 272 program). One milliliter of pre-warmed culture media was added immediately after 273 nucleofection, and cells were then transferred into a 6-well plate, and additional culture media 274 was added to a final volume of 3 ml. After nucleofection, cells were rested for 24 h in RPMI 275 supplemented with 10% serum, followed by cell culturing under iTreg condition for 72 h, as 276 described above. 277 278 2.10 ChIPmentation-qPCR analysis 279 Tagmentation-based chromatin immunoprecipitation (ChIPmentation) was conducted as 280 described previously [25]. Briefly, fixed chromatin from differentiated Treg cells (three million 281 cells) was sonicated and immunoprecipitated with anti-HIC1 or control IgG antibody. Genomic 282 DNA was probed by qPCR for RUNX1 promoter region and negative control region, where 283 HIC1 does not bind. PCR primers for ChIPmentation are listed in Table S2. PCR primers were 284 designed by LightCycler® Probe Design Software from Roche. Real-time PCR was performed 285 using custom TaqMan Gene Expression Assay reagent on QuantStudio 12K Flex Real-Time 286 PCR System (Thermo Scientific). 287 288 10 2.11 Luciferase Assay 289 HIC1 binding on RUNX1 promoter was assessed using Dual-Luciferase® Reporter Assay by 290 cloning the ChIP-defined genomic region upstream of a minimal promoter driving a luciferase 291 gene (pGL4.10 [luc2/minP]; Promega). Wild type HIC1 binding motif of the RUNX1 proximal 292 promoter (5′-TGCCCTGG-3′) or the corresponding mutant target sequence (ΔRunx1: 5′-293 AAAAATGG-3) was cloned upstream of a minimal promoter in a luciferase reporter plasmid 294 pGL4.10. Näive CD4+ T cells were nucleofected with Runx1-pGL4minP or ΔRunx1-295 pGL4minP or empty pGL4minP construct and renilla luciferase plasmid [26]. Post 296 nucleofection, cells were rested and cultured under Treg polarizing condition for 72 h, after 297 which cells were collected and luciferase assay was performed using the Dual Luciferase 298 Reagents (Promega). Firefly luciferase activity was normalized to renilla luciferase activity for 299 each sample and expressed as fold change over empty pGL4-minP. ΔRUNX1-pGL4minP with 300 mutated promoter region served as a negative control. 301 302 2.12 Suppression assay 303 To evaluate the ability of iTreg cells to suppress the proliferation of effector cells (responder 304 cells), we used a mixed lymphocyte reaction. Responder cells (Tres) were CD4+CD25- cells 305 isolated from cord blood from three different donors with the Dynal CD4 Positive Isolation Kit 306 (Invitrogen) and CD25 depletion kit (Miltenyi Biotec) [13]. To reduce the variability among 307 different responder cell populations, a set of Tres were isolated and stored at -80°C in freezing 308 medium (90% FCS and 10% DMSO). On day 1 of the assay, Tres were labelled with cell trace 309 violet (EF670/CTV) (Thermo Fisher Scientific). Fifty thousand Tres were cultured in different 310 wells with Th0 or iTregs in ratios of 1:0.125, 1:0.25 or 1:0.5. The division of responder cells 311 was analyzed by dye dilution on day 4. The level of suppression was calculated using the 312 following formula: % Suppression = [Percent of dividing cells (Tres-iTregs)/percent of 313 dividing cells in Tres]*100. 314 315 2.13 Ethical approval 316 Usage of the blood of unknown donors is approved by the Ethics Committee, Hospital District 317 of Southwest Finland. 318 319 3. Results 320 11 3.1 Identification of HIC1 protein interaction network in iTreg cells using affinity 321 purification–mass spectrometry 322 To better understand how HIC1 regulates the development and function of human iTreg cells, 323 we performed HIC1 immunoprecipitation (IP) (Figure S1) followed by mass spectrometry 324 (MS) to identify the interacting partners of HIC1. We identified 61 proteins as high confidence 325 HIC1 protein interactors (Figure 1A; Table S3). Notably, RUNX1, CBFB, and IKZF3, which 326 have important roles in Treg cell development and function, were detected as components of 327 this interactome [27–30]. Additionally, many tRNA synthetases (EPRS, DARS, IARS, QARS, 328 RARS, MARS, KARS and LARS) and tRNA synthetase complex interacting multifunctional 329 proteins (AMP1 and AMP2) were identified as HIC1 interactors (Figure 1A; Table S3). 330 Interestingly, many of the interacting proteins are associated with RNA regulatory processes. 331 For instance RPRD1B and RPRD2 are known to preferentially bind to the phosphorylated CTD 332 of RNAP II, leading to decreased Ser-5 and Ser-7 phosphorylation of RNAP II at target gene 333 promoters and regulate gene transcription [31]. Likewise, PARP13 and splicing factor SRSF1 334 were also identified in the HIC1 interactome and are reported to function in the regulation of 335 RNA stability and splicing [32,33]. Similarly, we identified association of HIC1 with E3 336 Ubiquitin-Protein Ligase HERC2, a member of the HECT family of E3 ubiquitin-protein 337 ligases implicated in DNA damage repair responses [34]. Furthermore, BCOR (BCL-6 338 interacting corepressor) was among the detected HIC1 interacting proteins (Figure 1A; Table 339 S3). BCOR is known to interact with the BTB/POZ domain that is a specific feature of a small 340 subset of zinc finger proteins including BCL-6 and HIC1 [35]. 341 Ingenuity Pathways Analysis (IPA; Qiagen) was used to summarize the cellular locations of 342 the HIC1 interactors. The analysis revealed that the majority of the HIC1 interactors were 343 associated with nucleus, followed by the cytoplasm, whilst small fraction of interactors were 344 distributed in plasma membrane and extracellular space (Figure 1B). Although HIC1 is 345 primarily localized in nucleus [13], it might interact with cytoplasmic proteins after they are 346 translocated to nucleus or when HIC1 is in the cytoplasm at different stages of cell activation 347 and differentiation. Enzymes was the major functional class of the HIC1 interacting proteins 348 followed by transcriptional and translational regulators (Figure 1C). Additionally, the other 349 functions associated with HIC1 interacting proteins were related to cytokines, ion channels and 350 transmembrane receptor proteins (Figure 1C). Gene ontology (GO) enrichment analysis on 351 HIC1 interacting proteins revealed that most enriched biological processes were associated 352 with regulation of gene expression, tRNA amino-acetylation for protein translation, RNA 353 12 splicing, ncRNA transcription, RNA metabolism, cellular and metabolic process and protein 354 transport (Figure 1D). 355 356 3.2 Validation of the HIC1 protein interactors in iTreg cells 357 For validation of the HIC1 interactome results, we used three different independent approaches: 358 selected reaction monitoring-mass spectrometry (SRM-MS), proximity ligation assay (PLA) 359 combined with confocal microscopy, and co-immunoprecipitation (Co-IP)/reverse Co-IP 360 followed by western blotting. We selected CBFB, RUNX1, and IKZF3 since they are known 361 to play important role in Treg cell function and homeostasis [27,28,30,36,37]. Further, we also 362 included BCOR for SRM validation as it was among the top HIC1 interactor and is known to 363 play role in maintaining the lineage stability and suppressive function of Treg cells [38]. 364 Additionally, although FOXP3 was not among the identified HIC1 interacting proteins, we 365 included it in the validation analysis, as it is a well-known interactor of both RUNX1-CBFB 366 protein complex and IKZF3 [27,28,36]. As illustrated in Figure 2A, the SRM-MS analysis 367 confirmed HIC1 interaction with CBFB, RUNX1, IKZF3, BCOR and FOXP3. The interaction 368 of HIC1 with RUNX1, IKZF3, CBFB and FOXP3 was further confirmed using Co-IP and 369 PLA-confocal microscopy analysis (Figure 2 B-D). The quantification of PLA images was 370 performed and shown in Figure S2. 371 372 3.3 HIC1 binds to RUNX1 promoter and modulates its expression 373 In our earlier HIC1 Chromatin Immunoprecipitation-sequencing (ChIP-seq) analysis [13], 374 HIC1 binding was observed at the RUNX1 promoter region in iTreg cells (Figure 3A). To 375 verify the binding of HIC1 to RUNX1 promoter region, we performed ChIPmentation assay 376 followed by PCR. Higher fold enrichment of HIC1 at the RUNX1 promoter compared to 377 negative control region was observed (Figure 3B), thus confirming the binding of HIC1 to the 378 RUNX1 promoter. 379 The observed binding of HIC1 to the RUNX1 promoter suggested that it might be directly 380 involved in the control of RUNX1 transcription in iTreg cells. To validate this hypothesis, a 381 luciferase reporter assay was used. For these analyses, the wild-type (WT) or mutated binding 382 region of HIC1 at the RUNX1 promoter was cloned upstream of firefly luciferase. 383 As shown in Figure 3C, the luciferase activity was significantly higher with WT RUNX1 384 compared to mutated (ΔRUNX1) promoter construct. These findings suggest that HIC1 385 regulates transcriptional activation of RUNX1 by binding to its promoter region. Further, 386 CRISPR-Cas9-mediated HIC1 ablation in iTreg cells led to reduced RUNX1 expression in 387 13 three donors (Figure 4A-B). In addition, loss of Treg cell mediated suppression was verified in 388 the HIC1 deficient cells (Figure 4C), which is in line with our earlier study [13] . 389 Collectively, these data suggest that HIC1 regulates RUNX1 expression in iTreg cells. 390 RUNX1 is part of FOXP3 transcriptional complex and indispensable for Treg cell function 391 [27,28,30,37]. Based on these findings, it is tempting to speculate that the reduction of RUNX1 392 expression in HIC1-deficient iTreg cells will lead to less RUNX1 availability, which may result 393 in destabilization of the FOXP3-RUNX1-CBFB transcriptional complex and modulation of the 394 FOXP3-RUNX1-CBFB complex dependent transcription program with concomitant loss of 395 suppression ability (Figure 5A, B). 396 397 4. Discussion 398 Lineage specification factors play an essential role in Treg cell differentiation by regulating the 399 expression of a set of genes that define the functional and phenotypic properties of Treg cells. 400 Although FOXP3 is an important regulator of Treg cells, the differentiated status of Treg cells 401 is not determined solely by FOXP3 expression and other factors play important role in this 402 process [36,39]. Recently, we established the importance of HIC1 in the development and 403 suppressive function of iTreg cells [13]. The aim of this follow-up study was to elucidate the 404 interactome of HIC1 in iTreg cells, and through this determined protein network gain 405 mechanistic insights into how HIC1 regulates the development and suppressive capability of 406 iTreg cells. 407 FOXP3 mediates the expression of its target genes through its association with a diverse set of 408 binding partners [36,39,40]. This large complex includes transcriptional regulators and many 409 sequence-specific TFs, e.g., RUNX1, NFAT, EOS, pSTAT3, IRF4, T-bet, GATA-3, RORγt, 410 RORα, FOXO1 and FOXO3, SATB1 and HIF-1α [36,37,41–43]. 411 RUNX1-CBFB heterodimer protein complex is a component of the FOXP3 interacting protein 412 network and plays a pivotal role in the development and function of Treg cells [28]. Further, 413 IKZF3 plays important role in epigenetic regulation by recruitment of chromatin modifiers, 414 such as the nucleosome remodeling and deacetylase (NuRD) complex, to the locus of target 415 genes to alter their chromatin organisation and hence their gene expression. IKZF3 interacts 416 with FOXP3 to silence the transcriptional program of effector T cells and promotes the 417 differentiation of functional FOXP3+ Treg cells [27,29,36,39]. 418 The results of this combination of bioanalytical methods, demonstrate that HIC1 is a part of 419 FOXP3-RUNX1-CBFB transcriptional complex. This FOXP3 protein complex is 420 14 indispensable for the expression of Treg signature genes and repressing the genes associated 421 with effector functions, thus maintaining suppressive ability of iTreg cells [27,36,39]. 422 Additionally, IKZF3 which is known to interact with FOXP3 and RUNX1 [36,44] was among 423 the top proteins identified in the HIC1 interactome. 424 Interestingly, FOXP3 was not detected as high confident interactor in discovery MS data but 425 was found to interact with HIC1 using SRM-based targeted MS data, western blot and PLA 426 assay. In fact, it was also detected in one of the MS replicates though not in the remaining two. 427 Discovery MS data was acquired in data-dependent acquisition mode, a semi-stochastic 428 method, wherein not all the proteins are detected all the time. This variability might explain 429 the absence of FOXP3 detection in the other two replicates. The confident interactors that were 430 identified in at least two out of three replicates were selected for further validations. Notably, 431 among the 61 high confidence HIC1 protein interactions identified, there were nine that are 432 also known FOXP3 partners (i.e. BCLAF1, CBFB, DARS, DHX30, IKZF3, POLR2A, 433 RBM8A and RUNX1) [36], which further support the idea that HIC1 is a part of FOXP3-434 RUNX1 complex. 435 Our previous results indicated that upon HIC1 silencing, FOXP3 expression remained 436 unchanged in iTreg cells [13]. Here, we demonstrate that HIC1 regulates the RUNX1 437 expression by binding to its promoter region. In the conditions of HIC1 deficiency, RUNX1 438 expression is reduced in iTreg cells. A lack of HIC1 and RUNX1 in HIC1 deficient cells may 439 result in the destabilization of the FOXP3 transcriptional complex thus altering the FOXP3-440 RUNX1-CBFB dependent transcriptional program, which may in turn diminish Treg-specific 441 gene expression and activate effector genes such as GATA3, TBX21 and IFNG with 442 concomitant loss of suppression ability as observed in our earlier study [13]. Thus, it seems 443 that HIC1 not only exerts control through direct binding to the RUNX1 promoter but also 444 engages in interactions with the RUNX1 protein within FOXP3-associated complexes. As a 445 consequence, HIC1 potentially governs the differentiation of Treg cells through a combination 446 of RUNX1-mediated direct and indirect mechanisms. 447 Interestingly, we found BCOR as one of the top HIC1 interactors. BCOR is a transcriptional 448 corepressor known to interact selectively with the BTB/POZ domain of the BCL6 449 transcriptional repressor which is involved in maintaining the lineage stability and regulating 450 suppressive function of Treg cells [38,45]. The interaction between BCOR and BCL6 results 451 in an increased transcriptional repression capacity of BCL6 [35]. Also, BCOR-mediated 452 repression is important for Th17 cell differentiation [46]. Additional studies are needed to 453 15 elucidate the role of BCOR-HIC1 interaction in context of iTreg cell differentiation and 454 function. 455 ARS-interacting multi-functional proteins (AIMP1, AIMP2) were also detected as interactors 456 of HIC1. Interestingly, it has been reported that AIMP1 enhances the differentiation of Treg 457 cells, while it has no effect on Th1, Th2, and Th17 cell differentiation [47]. 458 Further, among the identified HIC1 interactors were RPRD1B (CREPT), SRSF1, and HERC2 459 and PARP-1 (Figure 1A; Table S3). CREPT acts as an activator to promote transcriptional 460 activity of the β-catenin-TCF4 complex in response to Wnt signaling in tumors [48]. Notably, 461 it was reported that HIC1 also associates with β-catenin-TCF4 complex, and sequesters them 462 to HIC1 bodies and modulates the Wnt signalling [49]. Wnt–β-catenin signaling is known to 463 modulate function of human and murine Treg cells [50,51]. Based on these observations, HIC1 464 might contribute to Treg cell function by modulating Wnt–β-catenin signalling. However, 465 further studies are needed to understand the role of association of HIC1 and CREPT in the 466 context of Wnt–β-catenin signalling and iTreg cell differentiation. 467 SRSF1 is a member of the highly conserved serine/arginine (SR) family of RNA-binding 468 proteins [52]. SRSF1 controls post-transcriptional gene expression via pre-mRNA alternative 469 splicing, mRNA stability, and translation [33]. Recent studies show SRSF1 to be critical for 470 Treg cell function in mouse [53,54]. 471 HERC2 is a member of the HECT family of E3 ubiquitin-protein ligases and is implicated in 472 DNA damage repair responses by ubiquitinating processes [34]. Ubiquitin-mediated processes 473 influence the biology of Treg cells by modulating the signalling pathways critical for Foxp3 474 induction (e.g. TGFβ and NFκB signaling) or direct ubiquitination of the FOXP3 [55]. 475 Poly (ADP-ribose) polymerase (PARP)-1 mediates PolyADP-ribosylation which plays a key 476 role in the regulation of gene transcription in immune cells, including Treg cell differentiation 477 [56,57]. It would be of immense interest to establish the role of HIC1interaction with SRSF1, 478 HERC2 and PARP13 and how this association contributes to iTreg cell differentiation and 479 function. 480 481 5. Conclusions 482 We present the first characterisation of the HIC1 interactome, which includes several 483 noteworthy interactors. Gene ontology and pathway analysis of the associated proteins suggest 484 that HIC1 is involved in several functions that have not been previously implicated. Our results 485 indicate that HIC1 is a part of FOXP3 transcriptional complex comprised of RUNX1-CBFB 486 16 and IKZF3. We propose that the compromised Treg suppression ability observed in HIC1 487 deficient iTreg cells may be partly due to the perturbed activity of the FOXP3 transcriptional 488 complex. 489 490 Author contributions 491 The conception and design of the study, or acquisition of data, or analysis and interpretation of 492 data: SBAA, KB, OR, RM, SDB, IA, AM, TB, MHK, MMK, UUK, RL. 493 Drafting the article or revising it critically for important intellectual content: SBAA, KB, OR, 494 RM, UUK, RL. 495 Final approval of the version to be submitted: SBAA, KB, OR, RM, SDB, TB, MHK, AM, IA, 496 MMK, UUK, RL. 497 498 Acknowledgments 499 RL was supported by the Academy of Finland (AoF) Centre of Excellence in Molecular 500 Systems Immunology and Physiology Research (2012-2017) grant 250114; by the AoF grants 501 292335, 294337, 292482, 319280, 329277, 331793, 335435 and 31444; by grants from the 502 JDRF; the Novo Nordisk Foundation (grant NNF19OC0057218); the Sigrid Jusélius 503 Foundation; Jane and Aatos Erkko Foundation and the Finnish Cancer Foundation. Our 504 research is also supported by University of Turku, Åbo Akademi University, Turku Graduate 505 School, InFLAMES Flagship Programme of the Academy of Finland (decision number: 506 337530). SBAA, was supported by InFLAMES Postdoctoral fellowship Programme, Finnish 507 Cultural Foundation and Maud Kuistila Memorial Foundation. KB was supported by Orion 508 Research Foundation sr and Finnish Cultural Foundation. 509 510 Declaration of interest 511 Authors declare no competing interests. A.M. is a cofounder of Arsenal Biosciences, Spotlight 512 Therapeutics, and Survey Genomics, serves on the boards of directors at Spotlight Therapeutics 513 and Survey Genomics, is a board observer (and former member of the board of directors) at 514 Arsenal Biosciences, is a member of the scientific advisory boards of Arsenal Biosciences, 515 Spotlight Therapeutics, Survey Genomics, NewLimit, Amgen and Tenaya, owns stock in 516 Arsenal Biosciences, Spotlight Therapeutics, NewLimit, Survey Genomics, PACT Pharma, 517 and Tenaya and has received fees from Arsenal Biosciences, Spotlight Therapeutics, 518 NewLimit, 23andMe, PACT Pharma, Juno Therapeutics, Trizell, Vertex, Merck, Amgen, 519 Genentech, AlphaSights, Rupert Case Management, Bernstein and ALDA. A.M. is an investor 520 17 in and informal advisor to Offline Ventures and a client of EPIQ. The Marson laboratory has 521 received research support from Juno Therapeutics, Epinomics, Sanofi, GlaxoSmithKline, 522 Gilead and Anthem. 523 524 References 525 [1] H. Nishikawa, S. Sakaguchi, Regulatory T cells in cancer immunotherapy, Curr. Opin. 526 Immunol. 27 (2014) 1–7. https://doi.org/10.1016/J.COI.2013.12.005. 527 [2] S. Sakaguchi, T. Yamaguchi, T. Nomura, M. Ono, Regulatory T Cells and Immune 528 Tolerance, Cell. 133 (2008) 775–787. https://doi.org/10.1016/j.cell.2008.05.009. 529 [3] D. Haribhai, J.B. Williams, S. Jia, D. Nickerson, E.G. Schmitt, B. Edwards, J. 530 Ziegelbauer, M. Yassai, S.H. Li, L.M. Relland, P.M. Wise, A. Chen, Y.Q. Zheng, P.M. 531 Simpson, J. Gorski, N.H. Salzman, M.J. Hessner, T.A. Chatila, C.B. Williams, A 532 Requisite Role for Induced Regulatory T Cells in Tolerance Based on Expanding 533 Antigen Receptor Diversity, Immunity. 35 (2011) 109–122. 534 https://doi.org/10.1016/J.IMMUNI.2011.03.029. 535 [4] A.K. Abbas, C. Benoist, J.A. Bluestone, D.J. Campbell, S. Ghosh, S. Hori, S. Jiang, 536 V.K. Kuchroo, D. Mathis, M.G. Roncarolo, A. Rudensky, S. Sakaguchi, E.M. 537 Shevach, D.A.A. Vignali, S.F. Ziegler, Regulatory T cells: Recommendations to 538 simplify the nomenclature, Nat. Immunol. 14 (2013) 307–308. 539 https://doi.org/10.1038/ni.2554. 540 [5] A.Y. Rudensky, Regulatory T cells and Foxp3, Immunol. Rev. 241 (2011) 260–268. 541 https://doi.org/10.1111/j.1600-065X.2011.01018.x. 542 [6] Y. Wu, M. Borde, V. Heissmeyer, M. Feuerer, A.D. Lapan, J.C. Stroud, D.L. Bates, L. 543 Guo, A. Han, S.F. Ziegler, D. Mathis, C. Benoist, L. Chen, A. Rao, FOXP3 Controls 544 Regulatory T Cell Function through Cooperation with NFAT, Cell. 126 (2006) 375–545 387. https://doi.org/10.1016/j.cell.2006.05.042. 546 [7] J.A. Hill, M. Feuerer, K. Tash, S. Haxhinasto, J. Perez, R. Melamed, D. Mathis, C. 547 Benoist, Foxp3 Transcription-Factor-Dependent and -Independent Regulation of the 548 Regulatory T Cell Transcriptional Signature, Immunity. 27 (2007) 786–800. 549 https://doi.org/10.1016/j.immuni.2007.09.010. 550 18 [8] T. Takimoto, Y. Wakabayashi, T. Sekiya, N. Inoue, R. Morita, K. Ichiyama, R. 551 Takahashi, M. Asakawa, G. Muto, T. Mori, E. Hasegawa, S. Shizuya, T. Hara, M. 552 Nomura, A. Yoshimura, Correction: Smad2 and Smad3 Are Redundantly Essential for 553 the TGF-β–Mediated Regulation of Regulatory T Plasticity and Th1 Development, J. 554 Immunol. 186 (2011) 632–632. https://doi.org/10.4049/jimmunol.1090121. 555 [9] T. Sekiya, H. Nakatsukasa, Q. Lu, A. Yoshimura, Roles of transcription factors and 556 epigenetic modifications in differentiation and maintenance of regulatory T cells, 557 Microbes Infect. 18 (2016) 378–386. https://doi.org/10.1016/J.MICINF.2016.02.004. 558 [10] T. Sekiya, I. Kashiwagi, N. Inoue, R. Morita, S. Hori, H. Waldmann, A.Y. Rudensky, 559 H. Ichinose, D. Metzger, P. Chambon, A. Yoshimura, The nuclear orphan receptor 560 Nr4a2 induces Foxp3 and regulates differentiation of CD4+ T cells, Nat. Commun. 2 561 (2011) 269. https://doi.org/10.1038/ncomms1272. 562 [11] T. Sekiya, I. Kashiwagi, R. Yoshida, T. Fukaya, R. Morita, A. Kimura, H. Ichinose, D. 563 Metzger, P. Chambon, A. Yoshimura, Nr4a receptors are essential for thymic 564 regulatory T cell development and immune homeostasis, Nat. Immunol. 14 (2013) 565 230–237. https://doi.org/10.1038/ni.2520. 566 [12] H. Oh, Y. Grinberg-Bleyer, W. Liao, D. Maloney, P. Wang, Z. Wu, J. Wang, D.M. 567 Bhatt, N. Heise, R.M. Schmid, M.S. Hayden, U. Klein, R. Rabadan, S. Ghosh, An NF-568 κB Transcription-Factor-Dependent Lineage-Specific Transcriptional Program 569 Promotes Regulatory T Cell Identity and Function, Immunity. 47 (2017) 450-465.e5. 570 https://doi.org/10.1016/j.immuni.2017.08.010. 571 [13] Ubaid Ullah, S.B.A. Andrabi, S.K. Tripathi, O. Dirasantha, K. Kanduri, S. Rautio, 572 C.C. Gross, S. Lehtimäki, K. Bala, J. Tuomisto, U. Bhatia, D. Chakroborty, L.L. Elo, 573 H. Lähdesmäki, H. Wiendl, O. Rasool, R. Lahesmaa, Transcriptional Repressor HIC1 574 Contributes to Suppressive Function of Human Induced Regulatory T Cells, Cell Rep. 575 22 (2018) 2094–2106. https://doi.org/10.1016/j.celrep.2018.01.070. 576 [14] A. Szczepny, K. Carey, L. McKenzie, W.S.N. Jayasekara, F. Rossello, A. Gonzalez-577 Rajal, A.S. McCaw, D. Popovski, D. Wang, A.J. Sadler, A. Mahar, P.A. Russell, G. 578 Wright, R.A. McCloy, D.J. Garama, D.J. Gough, S.B. Baylin, A. Burgess, J.E. Cain, 579 D.N. Watkins, The tumor suppressor Hic1 maintains chromosomal stability 580 independent of Tp53, Oncogene. 37 (2018) 1939–1948. 581 19 https://doi.org/10.1038/s41388-017-0022-1. 582 [15] J. Cox, M. Mann, MaxQuant enables high peptide identification rates, individualized 583 p.p.b.-range mass accuracies and proteome-wide protein quantification, Nat. 584 Biotechnol. 26 (2008) 1367–1372. https://doi.org/10.1038/nbt.1511. 585 [16] J. Cox, N. Neuhauser, A. Michalski, R.A. Scheltema, J. V. Olsen, M. Mann, 586 Andromeda: A peptide search engine integrated into the MaxQuant environment, J. 587 Proteome Res. 10 (2011) 1794–1805. https://doi.org/10.1021/pr101065j. 588 [17] S. Tyanova, T. Temu, P. Sinitcyn, A. Carlson, M.Y. Hein, T. Geiger, M. Mann, J. Cox, 589 The Perseus computational platform for comprehensive analysis of (prote)omics data, 590 Nat. Methods. 13 (2016) 731–740. https://doi.org/10.1038/nmeth.3901. 591 [18] S. Jäger, P. Cimermancic, N. Gulbahce, J.R. Johnson, K.E. McGovern, S.C. Clarke, M. 592 Shales, G. Mercenne, L. Pache, K. Li, H. Hernandez, G.M. Jang, S.L. Roth, E. Akiva, 593 J. Marlett, M. Stephens, I. D’Orso, J. Fernandes, M. Fahey, C. Mahon, A.J. 594 Oĝdonoghue, A. Todorovic, J.H. Morris, D.A. Maltby, T. Alber, G. Cagney, F.D. 595 Bushman, J.A. Young, S.K. Chanda, W.I. Sundquist, T. Kortemme, R.D. Hernandez, 596 C.S. Craik, A. Burlingame, A. Sali, A.D. Frankel, N.J. Krogan, Global landscape of 597 HIV-human protein complexes, Nature. 481 (2012) 365–370. 598 https://doi.org/10.1038/nature10719. 599 [19] D. Mellacheruvu, Z. Wright, A.L. Couzens, J.P. Lambert, N.A. St-Denis, T. Li, Y. V. 600 Miteva, S. Hauri, M.E. Sardiu, T.Y. Low, V.A. Halim, R.D. Bagshaw, N.C. Hubner, 601 A. Al-Hakim, A. Bouchard, D. Faubert, D. Fermin, W.H. Dunham, M. Goudreault, 602 Z.Y. Lin, B.G. Badillo, T. Pawson, D. Durocher, B. Coulombe, R. Aebersold, G. 603 Superti-Furga, J. Colinge, A.J.R. Heck, H. Choi, M. Gstaiger, S. Mohammed, I.M. 604 Cristea, K.L. Bennett, M.P. Washburn, B. Raught, R.M. Ewing, A.C. Gingras, A.I. 605 Nesvizhskii, The CRAPome: A contaminant repository for affinity purification-mass 606 spectrometry data, Nat. Methods. 10 (2013) 730–736. 607 https://doi.org/10.1038/nmeth.2557. 608 [20] M. Hill, D. Sais, T. Monteiro Marques, M. Gama Carvalho, N. Tran, Developing a 609 virus-microRNA interactome using cytoscape, MethodsX. 7 (2020) 100700. 610 https://doi.org/10.1016/j.mex.2019.10.011. 611 20 [21] Y. Perez-Riverol, J. Bai, C. Bandla, D. García-Seisdedos, S. Hewapathirana, S. 612 Kamatchinathan, D.J. Kundu, A. Prakash, A. Frericks-Zipper, M. Eisenacher, M. 613 Walzer, S. Wang, A. Brazma, J.A. Vizcaíno, The PRIDE database resources in 2022: a 614 hub for mass spectrometry-based proteomics evidences, Nucleic Acids Res. 50 (2022) 615 D543–D552. https://doi.org/10.1093/nar/gkab1038. 616 [22] B. MacLean, D.M. Tomazela, N. Shulman, M. Chambers, G.L. Finney, B. Frewen, R. 617 Kern, D.L. Tabb, D.C. Liebler, M.J. MacCoss, Skyline: An open source document 618 editor for creating and analyzing targeted proteomics experiments, Bioinformatics. 26 619 (2010) 966–968. https://doi.org/10.1093/bioinformatics/btq054. 620 [23] A.E. Carpenter, T.R. Jones, M.R. Lamprecht, C. Clarke, I.H. Kang, O. Friman, D.A. 621 Guertin, J.H. Chang, R.A. Lindquist, J. Moffat, P. Golland, D.M. Sabatini, 622 CellProfiler: Image analysis software for identifying and quantifying cell phenotypes, 623 Genome Biol. 7 (2006) R100. https://doi.org/10.1186/gb-2006-7-10-r100. 624 [24] T.L. Roth, C. Puig-Saus, R. Yu, E. Shifrut, J. Carnevale, P.J. Li, J. Hiatt, J. Saco, P. 625 Krystofinski, H. Li, V. Tobin, D.N. Nguyen, M.R. Lee, A.L. Putnam, A.L. Ferris, J.W. 626 Chen, J.N. Schickel, L. Pellerin, D. Carmody, G. Alkorta-Aranburu, D. Del Gaudio, H. 627 Matsumoto, M. Morell, Y. Mao, M. Cho, R.M. Quadros, C.B. Gurumurthy, B. Smith, 628 M. Haugwitz, S.H. Hughes, J.S. Weissman, K. Schumann, J.H. Esensten, A.P. May, A. 629 Ashworth, G.M. Kupfer, S.A.W. Greeley, R. Bacchetta, E. Meffre, M.G. Roncarolo, 630 N. Romberg, K.C. Herold, A. Ribas, M.D. Leonetti, A. Marson, Reprogramming 631 human T cell function and specificity with non-viral genome targeting, Nature. 559 632 (2018) 405–409. https://doi.org/10.1038/s41586-018-0326-5. 633 [25] C. Schmidl, A.F. Rendeiro, N.C. Sheffield, C. Bock, ChIPmentation: Fast, robust, low-634 input ChIP-seq for histones and transcription factors, Nat. Methods. 12 (2015) 963–635 965. https://doi.org/10.1038/nmeth.3542. 636 [26] M.M. Khan, M.H. Khan, U.U. Kalim, S. Khan, S. Junttila, N. Paulin, L. Kong, O. 637 Rasool, L.L. Elo, R. Lahesmaa, Long Intergenic Noncoding RNA MIAT as a 638 Regulator of Human Th17 Cell Differentiation, Front. Immunol. 13 (2022) 856762. 639 https://doi.org/10.3389/fimmu.2022.856762. 640 [27] M. Ono, H. Yaguchi, N. Ohkura, I. Kitabayashi, Y. Nagamura, T. Nomura, Y. 641 Miyachi, T. Tsukada, S. Sakaguchi, Foxp3 controls regulatory T-cell function by 642 21 interacting with AML1/Runx1, Nature. 446 (2007) 685–689. 643 https://doi.org/10.1038/NATURE05673. 644 [28] A. Kitoh, M. Ono, Y. Naoe, N. Ohkura, T. Yamaguchi, H. Yaguchi, I. Kitabayashi, T. 645 Tsukada, T. Nomura, Y. Miyachi, I. Taniuchi, S. Sakaguchi, Indispensable Role of the 646 Runx1-Cbfβ Transcription Complex for In Vivo-Suppressive Function of FoxP3+ 647 Regulatory T Cells, Immunity. 31 (2009) 609–620. 648 https://doi.org/10.1016/J.IMMUNI.2009.09.003. 649 [29] H.K. Kwon, H.M. Chen, D. Mathis, C. Benoist, Different molecular complexes that 650 mediate transcriptional induction and repression by FoxP3, Nat. Immunol. 18 (2017) 651 1238–1248. https://doi.org/10.1038/ni.3835. 652 [30] F. Leng, W. Zhang, R.N. Ramirez, J. Leon, Y. Zhong, L. Hou, K. Yuki, J. van der 653 Veeken, A.Y. Rudensky, C. Benoist, S. Hur, The transcription factor FoxP3 can fold 654 into two dimerization states with divergent implications for regulatory T cell function 655 and immune homeostasis, Immunity. 55 (2022) 1354-1369.e8. 656 https://doi.org/10.1016/J.IMMUNI.2022.07.002. 657 [31] Z. Ni, J.B. Olsen, X. Guo, G. Zhong, E.D. Ruan, E. Marcon, P. Young, H. Guo, J. Li, 658 J. Moffat, A. Emili, J.F. Greenblatt, Control of the RNA polymerase II 659 phosphorylation state in promoter regions by CTD interaction domain-containing 660 proteins RPRD1A and RPRD1B, Transcription. 2 (2011) 237–242. 661 https://doi.org/10.4161/trns.2.5.17803. 662 [32] T. Todorova, F.J. Bock, P. Chang, PARP13 regulates cellular mRNA post-663 transcriptionally and functions as a pro-apoptotic factor by destabilizing TRAILR4 664 transcript, Nat. Commun. 5 (2014) 5362. https://doi.org/10.1038/ncomms6362. 665 [33] J.M. Howard, J.R. Sanford, The RNAissance family: SR proteins as multifaceted 666 regulators of gene expression, Wiley Interdiscip. Rev. RNA. 6 (2015) 93–110. 667 https://doi.org/10.1002/wrna.1260. 668 [34] S. Bekker-Jensen, J.R. Danielsen, K. Fugger, I. Gromova, A. Nerstedt, J. Bartek, J. 669 Lukas, N. Mailand, Erratum: HERC2 coordinates ubiquitin-dependent assembly of 670 DNA repair factors on damaged chromosomes (Nature Cell Biology (2010) 12 (80-671 81)), Nat. Cell Biol. 12 (2010). https://doi.org/10.1038/ncb0410-412. 672 22 [35] K.D. Huynh, W. Fischle, E. Verdin, V.J. Bardwell, BCoR, a novel corepressor 673 involved in BCL-6 repression, Genes Dev. 14 (2000) 1810–1823. 674 https://doi.org/10.1101/gad.14.14.1810. 675 [36] D. Rudra, P. Deroos, A. Chaudhry, R.E. Niec, A. Arvey, R.M. Samstein, C. Leslie, 676 S.A. Shaffer, D.R. Goodlett, A.Y. Rudensky, Transcription factor Foxp3 and its 677 protein partners form a complex regulatory network, Nat. Immunol. 13 (2012) 1010–678 1019. https://doi.org/10.1038/ni.2402. 679 [37] D. Rudra, T. Egawa, M.M.W. Chong, P. Treuting, D.R. Littman, A.Y. Rudensky, 680 Runx-CBFbeta complexes control expression of the transcription factor Foxp3 in 681 regulatory T cells., Nat. Immunol. 10 (2009). https://doi.org/10.1038/ni.1795. 682 [38] D. V. Sawant, H. Wu, W. Yao, S. Sehra, M.H. Kaplan, A.L. Dent, The transcriptional 683 repressor Bcl6 controls the stability of regulatory T cells by intrinsic and extrinsic 684 pathways, Immunology. 145 (2015) 11–23. https://doi.org/10.1111/imm.12393. 685 [39] M. Miyara, Y. Yoshioka, A. Kitoh, T. Shima, K. Wing, A. Niwa, C. Parizot, C. Taflin, 686 T. Heike, D. Valeyre, A. Mathian, T. Nakahata, T. Yamaguchi, T. Nomura, M. Ono, Z. 687 Amoura, G. Gorochov, S. Sakaguchi, Functional Delineation and Differentiation 688 Dynamics of Human CD4+ T Cells Expressing the FoxP3 Transcription Factor, 689 Immunity. 30 (2009) 899–911. https://doi.org/10.1016/j.immuni.2009.03.019. 690 [40] W. Fu, A. Ergun, T. Lu, J.A. Hill, S. Haxhinasto, M.S. Fassett, R. Gazit, S. Adoro, L. 691 Glimcher, S. Chan, P. Kastner, D. Rossi, J.J. Collins, D. Mathis, C. Benoist, A 692 multiply redundant genetic switch “locks in” the transcriptional signature of regulatory 693 T cells, Nat. Immunol. 13 (2012) 972–980. https://doi.org/10.1038/ni.2420. 694 [41] Y. Wu, M. Borde, V. Heissmeyer, M. Feuerer, A.D. Lapan, J.C. Stroud, D.L. Bates, L. 695 Guo, A. Han, S.F. Ziegler, D. Mathis, C. Benoist, L. Chen, A. Rao, FOXP3 Controls 696 Regulatory T Cell Function through Cooperation with NFAT, Cell. 126 (2006) 375–697 387. https://doi.org/10.1016/j.cell.2006.05.042. 698 [42] E.A. Wohlfert, J.R. Grainger, N. Bouladoux, J.E. Konkel, G. Oldenhove, C.H. Ribeiro, 699 J.A. Hall, R. Yagi, S. Naik, R. Bhairavabhotla, W.E. Paul, R. Bosselut, G. Wei, K. 700 Zhao, M. Oukka, J. Zhu, Y. Belkaid, GATA3 controls Foxp3+ regulatory T cell fate 701 during inflammation in mice, J. Clin. Invest. 121 (2011) 4503–4515. 702 23 https://doi.org/10.1172/JCI57456. 703 [43] E. Bettelli, M. Dastrange, M. Oukka, Foxp3 interacts with nuclear factor of activated T 704 cells and NF-κB to repress cytokine gene expression and effector functions of T helper 705 cells, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 5138–5143. 706 https://doi.org/10.1073/PNAS.0501675102. 707 [44] N. Zhou, A. Gutierrez-Uzquiza, X.Y. Zheng, R. Chang, D.T. Vogl, A.L. Garfall, L. 708 Bernabei, A. Saraf, L. Florens, M.P. Washburn, A. Illendula, J.H. Bushweller, L. 709 Busino, RUNX proteins desensitize multiple myeloma to lenalidomide via protecting 710 IKZFs from degradation, Leukemia. 33 (2019) 2006–2021. 711 https://doi.org/10.1038/S41375-019-0403-2. 712 [45] Y. Li, Z. Wang, H. Lin, L. Wang, X. Chen, Q. Liu, Q. Zuo, J. Hu, H. Wang, J. Guo, L. 713 Xie, J. Tang, Z. Li, L. Hu, L. Xu, X. Zhou, L. Ye, Q. Huang, L. Xu, Bcl6 Preserves the 714 Suppressive Function of Regulatory T Cells During Tumorigenesis, Front. Immunol. 715 11 (2020) 806. https://doi.org/10.3389/fimmu.2020.00806. 716 [46] J.A. Kotov, D.I. Kotov, J.L. Linehan, V.J. Bardwell, M.D. Gearhart, M.K. Jenkins, 717 BCL6 corepressor contributes to Th17 cell formation by inhibiting Th17 fate 718 suppressors, J. Exp. Med. 216 (2019) 1450–1464. 719 https://doi.org/10.1084/jem.20182376. 720 [47] M.S. Kim, A. Lee, D. Cho, T.S. Kim, AIMP1 regulates TCR signaling and induces 721 differentiation of regulatory T cells by interfering with lipid raft association, Biochem. 722 Biophys. Res. Commun. 514 (2019) 875–880. 723 https://doi.org/10.1016/J.BBRC.2019.05.040. 724 [48] Y. Zhang, C. Liu, X. Duan, F. Ren, S. Li, Z. Jin, Y. Wang, Y. Feng, Z. Liu, Z. Chang, 725 CREPT/RPRD1B, a recently identified novel protein highly expressed in tumors, 726 enhances the β-catenin·TCF4 transcriptional activity in response to Wnt signaling, J. 727 Biol. Chem. 289 (2014) 22589–22599. https://doi.org/10.1074/jbc.M114.560979. 728 [49] T. Valenta, J. Lukas, L. Doubravska, B. Fafilek, V. Korinek, HIC1 attenuates Wnt 729 signaling by recruitment of TCF-4 and β-catenin to the nuclear bodies, EMBO J. 25 730 (2006) 2326–2337. https://doi.org/10.1038/sj.emboj.7601147. 731 [50] J. Quandt, S. Arnovitz, L. Haghi, J. Woehlk, A. Mohsin, M. Okoreeh, P.S. Mathur, 732 24 A.O. Emmanuel, A. Osman, M. Krishnan, S.B. Morin, A.T. Pearson, R.F. Sweis, J. 733 Pekow, C.R. Weber, K. Khazaie, F. Gounari, Wnt–β-catenin activation epigenetically 734 reprograms Treg cells in inflammatory bowel disease and dysplastic progression, Nat. 735 Immunol. 22 (2021) 471–484. https://doi.org/10.1038/s41590-021-00889-2. 736 [51] J. vanLoosdregt, V. Fleskens, M.M. Tiemessen, M. Mokry, R. vanBoxtel, J. Meerding, 737 C.E.G.M. Pals, D. Kurek, M.R.M. Baert, E.M. Delemarre, A. Gröne, M.J.A.G. 738 Koerkamp, A.J.A.M. Sijts, E.E.S. Nieuwenhuis, M.M. Maurice, J.H. vanEs, D. 739 tenBerge, F.C. Holstege, F.J.T. Staal, D.M.W. Zaiss, B.J. Prakken, P.J. Coffer, 740 Canonical Wnt Signaling Negatively Modulates Regulatory T Cell Function, 741 Immunity. 39 (2013) 298–310. https://doi.org/10.1016/J.IMMUNI.2013.07.019. 742 [52] S. Das, A.R. Krainer, Emerging functions of SRSF1, splicing factor and oncoprotein, 743 in RNA metabolism and cancer, Mol. Cancer Res. 12 (2014) 1195–1204. 744 https://doi.org/10.1158/1541-7786.MCR-14-0131. 745 [53] T. Katsuyama, V.R. Moulton, Splicing factor SRSF1 is indispensable for regulatory 746 T cell homeostasis and function, Cell Rep. 36 (2021) 109339. 747 https://doi.org/10.1016/J.CELREP.2021.109339. 748 [54] M.F. Cassidy, Z.T. Herbert, V.R. Moulton, Splicing factor SRSF1 controls 749 autoimmune-related molecular pathways in regulatory T cells distinct from FoxP3, 750 Mol. Immunol. 152 (2022) 140–152. https://doi.org/10.1016/j.molimm.2022.10.017. 751 [55] J. Barbi, D.M. Pardoll, F. Pan, Ubiquitin-dependent regulation of Foxp3 and Treg 752 function, Immunol. Rev. 266 (2015) 27–45. https://doi.org/10.1111/imr.12312. 753 [56] A.R. Fehr, S.A. Singh, C.M. Kerr, S. Mukai, H. Higashi, M. Aikawa, The impact of 754 PARPs and ADP-ribosylation on inflammation and host-pathogen interactions, Genes 755 Dev. 34 (2020) 341–359. https://doi.org/10.1101/gad.334425.119. 756 [57] X. Luo, J. Nie, S. Wang, Z. Chen, W. Chen, D. Li, H. Hu, B. Li, Poly(ADP-757 ribosyl)ation of FOXP3 Protein Mediated by PARP-1 Protein regulates the function of 758 regulatory t cells, J. Biol. Chem. 290 (2015) 28675–28682. 759 https://doi.org/10.1074/jbc.M115.661611. 760 761 25 Figures and figure legends 762 763 Figure 1: The HIC1 interactome. Following affinity purification and mass spectrometry (AP-764 MS), statistical analysis of interactions using the MIST algorithm and filtering for common 765 contaminants, a panel of interacting proteins was identified. The proteins are shown as a heat-766 map (A), clustered on the basis of their Z-score normalized LFQ intensities from the bait and 767 control measurements. Representations of their cellular localization (B) and functional classes 768 (C) are shown as pie charts. The interactors are further represented as an interaction network 769 (D), with the enriched biological processes indicated by the use of colored circles and strength 770 of the enrichment indicated by the size of the peripheral circles. Further representation of this 771 data is provided in Table S4. 772 773 A B C D 26 774 Figure 2: HIC1 interactome validation by SRM-MS, IP-WB, reverse IP-WB and PLA assay. 775 (A) SRM-MS validation of the candidate HIC1 interacting partners. HIC1 IP was performed 776 using HIC1 antibody in 72 h polarized iTreg cells and six proteins were validated by targeted 777 proteomics. Averaged results from three replicates are presented in the form of a box plot. 778 Statistical analysis was made by a two-tailed paired student's T-test, **p-value represents < 779 0.005; and ***p-value represents < 0.0005. (B) Validation of HIC1 protein interaction by IP-780 western blot. HIC1 IP followed by WB detection with RUNX1, IKZF3, CBFB and FOXP3 781 antibodies was performed to validate the HIC1 interaction with RUNX1, IKZF3, CBFB and 782 FOXP3. Total cell lysate (Input), control IP (IgG) and HIC1 IP are shown in the blots. A 783 conformational specific secondary antibody was used to probe proteins without interference 784 D 27 from the denatured IgG heavy (50 kDa) and light chains (25 kDa). A representative WB of 785 three biological replicates is shown. (C) RUNX1-IP, IKZF3-IP, CBFB-IP and FOXP3-IP were 786 performed using iTreg cell lysate followed by WB detection with HIC1 antibody to validate 787 the HIC1 interaction with RUNX1, IKZF3, CBFB and FOXP3. A representative western blot 788 of three biological replicates is shown. (D) Proximity ligation assay (PLA) with indicated 789 antibody pairs in 72 h polarized iTreg cells. The GFP antibody was used as a negative control. 790 DAPI was used to stain the nuclei. The scale bar is 7 μm. 791 792 28 793 794 795 796 797 798 Figure 3: ChIPseq, Luciferase and ChIP-qPCR assays demonstrate the recruitment of HIC1 to 799 the RUNX1 promoter in iTreg cells. (A) ChIP-Seq analysis identifies binding sites of HIC1 at 800 RUNX1 promoter region in iTreg cells. UCSC genome browser snap shots for the RUNX1 gene 801 that demonstrates the HIC1 binding peaks as indicated in enclosed box. (B) ChIP-qPCR assay 802 was performed in iTreg cells using anti-HIC1 IgG (HIC1). Fold enrichment of HIC1 binding 803 over a negative control (NC) region, where HIC1 does not bind, was measured. (C) Relative 804 luciferase activities were measured after co-transfection of RUNX1 promoter (Runx1) or 805 mutated Runx1 promoter (ΔRunx1) constructs along with the pRL-TK expression plasmid into 806 iTreg cells. Statistical significance calculated using Student's t-test (two-tailed paired) where 807 *denotes p value <0.05, ***denotes p value <0.0005. 808 809 A B C NC HIC1 0 50 100 150 Fo ld e nr ic hm en t ✱ 29 810 811 Figure 4: HIC1 deficiency leads to reduced RUNX1 expression. (A) HIC1 was targeted with 812 CRISPR-Cas9 RNPs in three individual donors, and HIC1, RUNX1 and Actin expression 813 levels were measured by WB. (B) The dot plot shows the quantification of the HIC1 and 814 RUNX1 levels in HIC1 depleted and Control (NT) iTreg cells. Quantification was performed 815 using Image Studio Lite software Ver 5.2. HIC1 and RUNX1 levels were normalised to Actin 816 levels. Statistical significance was calculated using Student's t-test (two-tailed paired) where 817 **denotes p value <0.005, ***denotes p value <0.0005. (C) The dot plot shows the 818 proliferation of responder cells at a responder/suppressor ratio of 1:1, 1:0.5 and 1:0.25 after 72 819 h of activation in presence of HIC1-sufficient (NT) or HIC1-deficient (HIC1-KO) iTregs. The 820 percentage of suppression was calculated using the following formula: % suppression = [% of 821 dividing cells (Tres-iTreg)/% of dividing cells in Tres] × 100. Statistical significance was 822 calculated using Student's t-test (two-tailed paired) where *: p<0.05. 823 824 30 825 826 Figure 5: HIC1 interacts with FOXP3-RUNX1-CBFB protein complex. (A) FOXP3-RUNX1- 827 CBFB-HIC1 protein complex regulates expression of downstream target genes and controls 828 Treg suppression. (B) HIC1 deficiency results in down-regulation of RUNX1, perhaps 829 destabilizes the FOXP3 protein complex, leading to dysregulation in the expression of target 830 genes, and ultimately causing a loss of Treg suppression. 831 832 A B