NSC697923

Cross-talk between SUMOylation and ISGylation in response to interferon

Faten El-Asmia, Francis P. McManusb, Carlos Eduardo Brantis-de-Carvalhoa, Jose Carlos Valle-Casusod, Pierre Thibaultb,c,⁎, Mounira K. Chelbi-Alixa,⁎

Abstract

Interferon (IFN) plays a central role in regulating host immune response to viral pathogens through the induction of IFN-Stimulated Genes (ISGs). IFN also enhances cellular SUMOylation and ISGylation, though the functional interplay between these modifications remains unclear. Here, we used a system-level approach to profile global changes in protein abundance in SUMO3-expressing cells stimulated by IFNα. These analyses revealed the stabilization of several ISG factors including SAMHD1, MxB, GBP1, GBP5, Tetherin/BST2 and members of IFITM, IFIT and IFI families. This process was correlated with enhanced IFNα-induced anti-HIV-1 and HSV-1 activities. Also IFNα upregulated protein ISGylation through increased abundance of E2 conjugating enzyme UBE2L6, and E3 ISG15 ligases TRIM25 and HERC5. Remarkably, TRIM25 depletion blocked SUMO3-dependent protein stabilization in response to IFNα. Our data identify a new mechanism by which SUMO3 regulates ISG product stability and reinforces the relevance of the SUMO pathway in controlling both the expression and functions of the restriction factors and IFN antiviral response.

Keywords:
SUMO
IFN
ISG15
Ubiquitin
Restriction factors
HIV-1
HSV-1

1. Introduction

The innate immune response to viral infection relies on interferon (IFN) synthesis. IFNs act through autocrine and paracrine routes by binding to their receptors and activating the Jak/STAT pathway and thus inducing more than hundred IFN-Stimulated Genes (ISGs) [1–3]. Some of the ISG products mediate the IFN-induced antiviral state and are termed restriction factors [2,4]. In addition to their implication in the innate immune response, some restriction factors are constitutively expressed, and are also implicated in intrinsic antiviral activity. There is a growing body of evidence suggesting that SUMO (Small Ubiquitinrelated MOdifier) can modulate innate and intrinsic immunity by altering IFN synthesis, Jak/STAT signaling as well as the expression and function of some ISG restriction factors [5–10].
Many restriction factors, such as ProMyelocytic Leukemia (PML), TRIpartite Motif (TRIM) 5α, double-stranded RNA-dependent protein kinase (PKR), p53, Death domain associated protein (Daxx), SAM domain and HD domain-containing (SAMHD1), are known to be SUMOylated [6,11,12]. More recently, we reported that the myxovirus resistance protein A (MxA) is SUMOylated at lysine 48 [11] and is highly stabilized through its oligomerization in SUMO-expressing cells [8].
SUMO is a member of the ubiquitin-like family that plays an important role in a wide variety of biological processes, including protein subcellular localization, protein degradation, innate immunity and antiviral defense [6,13]. Several enzymes are involved in this process including SUMO proteases, an E1-activating enzyme (SAE1/SAE2), an E2-conjugating enzyme (UBC9) and several E3 ligases [14]. In humans, three SUMO paralogs (SUMO1, SUMO2 and SUMO3) are ubiquitously expressed and act as protein modifiers. The two highly homologous proteins SUMO2 and SUMO3, share 97% sequence identity, are collectively referred to as SUMO2/3 and only share 50% amino acid identity with SUMO1 [15]. SUMO is covalently conjugated to its substrates via an acceptor lysine (K) that is typically found within a consensus motif (ψKxE), where (ψ) is a hydrophobic amino acid and (x) any amino acid. SUMO1 and SUMO2/3 modify both common and different substrates and growing evidences show that they may have distinct functions [7,9,16–18]. A key difference amongst paralogs is the capability of SUMO2/3 to efficiently form highly branched poly SUMO chains that have the ability to recruit SUMO ubiquitin ligases such RING finger protein 4 (RNF4), thus forming SUMO-ubiquitin mixed chains and resulting in the proteasomal degradation of SUMO2/3 conjugated substrates [19]. The best studied case being the SUMO dependant degradation of PML in cells treated with arsenic trioxide [17,18] or with IFN [9].
In view of the importance of protein SUMOylation in innate and intrinsic immunity, the present work examines the repertoire of SUMO substrates modulated during the IFNα response. This was partly achieved through a targeted approach by specifically looking at the stability of restriction factors implicated in antiviral defense such as MxA, MxB, Guanylate Binding Protein 5 (GBP5), TRIM5α, Daxx and SAMHD1. We expanded the scope of the analysis by performing largescale proteomic analyses, which revealed the positive regulation of several other ISG products by SUMO3, including several proteins involved in ISG15 conjugation. Furthermore, we demonstrated the ability of SUMO3 to increase the antiviral activity of IFNα against Human immunodeficiency virus 1 (HIV-1) and Herpes Simplex Virus 1 (HSV-1), and uncovered an important cross-talk taking place between protein SUMOylation and ISGylation.

2. Materials and methods

2.1. Materials

Recombinant human IFNα2 was from Schering (USA) and human IFNγ from Roussel Uclaf (Romainville, France). Rabbit polyclonal antibodies raised against PML (Sc-5621), STAT1 (sc-345), SUMO1 (sc9060), Daxx (sc-7152), goat polyclonal anti-MxB (sc-47197), mouse monoclonal anti-IFI16 (1G7, sc-8023), mouse monoclonal anti-ubiquitin (P4D1, sc-8017) and rat monoclonal anti-GBP1 (1B1, sc-53857) antibodies were from Santa-Cruz Biotechnology (USA). Mouse anti6xHis antibody was from Clontech (USA), rabbit anti-MxA and antiGBP5 antibodies were from Proteintech (USA), mouse polyclonal antiSAMHD1 (Ab-67820) and mouse monoclonal anti-HSV-1 ICP0 (5H7) were from abcam (UK). Rabbit anti-phospho SAMHD1 (Thr592), rabbit anti-histone H3, monoclonal anti-UBC9 (D26F2) and monoclonal anti SUMO2/3 (18H8) antibodies were from Cell Signaling Technologies (USA). Rabbit anti-SAMD9 and monoclonal anti-αTubulin (DM1A) antibodies were from Sigma (USA), rabbit anti-TRIM25 antibody was from GeneTex (USA) and the mouse anti-HERC5 antibody from Abnova (Taiwan). Rabbit anti-TRIM5α and rabbit anti-TRIM22 antibodies were a gift from Paul Bieniasz (Aaron Diamond AIDS Research Center, New York, USA) and Nadir Mechti (Montpellier University, France), respectively. Mouse anti-ISG15 antibody was a gift from Ernest Borden (Center for Hematology and Oncology Molecular Therapeutics, Cleveland, USA) [20] and rabbit anti-RNF4 antibody was a gift from Jorma Palvimo (University of Eastern Finland). Recombinant SENP1, SENP2 and USP18 were from Boston biochem (USA). Peroxidase-coupled secondary antibodies were purchased from Santa Cruz Biotechnology. Plasmid transfections were performed using Fugene 6 from Promega (France). siRNA scramble and targeting HERC5, TRIM25, UBC9 and RNF4 were purchased from Dharmacon (USA) and transfected into cells using HiperFect transfection reagent (Qiagen, France).

2.2. Cells and viruses

Cervical cancer HeLa cells, Human glioblastoma astrocytoma U373MG and Human embryonic kidney (HEK) 293 cells were grown at 37 °C in DMEM supplemented with 10% foetal calf serum. HEK293 cells stably expressing 6xHisSUMO3-Q87R-Q88N (SUMO3m) were obtained as previously described [21]. Acute monocytic leukemia THP1 cells were purchased from ATCC and were grown in RPMI supplemented with 10% foetal calf serum. HeLa cells stably expressing each HisSUMO1 or His-SUMO3 were used as previously described [9]. The KOS strain of HSV-1 (1.7 108 pfu/ml) was used in this study. The viral titers were determined on U373MG cells by measuring the 50% tissue culture infective dose (TCID50). Vesicular Stomatitis Virus glycoprotein (VSVG)-pseudotyped retroviral vectors were generated by co-transfecting HEK293T cells with pVSV-G, a Gag-Pol expression plasmid and green fluorescent protein (GFP)-expressing retroviral vector, using the ProFection calcium phosphate kit (Promega).

2.3. Purification of His6-tagged SUMO conjugates

HeLa-His-SUMO3 cells (107), untreated or treated with 1000 units/ mL of IFN for 20 h, were lysed in denaturating buffer A (6M guanidinium-HCl, 0.1 M Na2HPO4/NaH2PO4, 0.01 M Tris-HCl pH 8.0, 5 mM imidazole and 10 mM β-mercaptoethanol). After sonication, the lysates were mixed with 50 μL of Ni-NTA-agarose beads (Qiagen) for 3 h at room temperature. The beads were successively washed with buffer B (0.1% triton ×100; 8 M urea, 0.1 M Na2HPO4/NaH2PO4, 0.01 M TrisHCl pH 6.3, 10 mM β-mercaptoethanol), and subsequently eluted with 200 mM imidazole in 0.15 M Tris-HCl pH 6.7, 30% glycerol and 0.72 M β-mercaptoethanol.

2.4. SUMO and ISG15 deconjugation assays

2107 HeLa-SUMO3 cells were untreated or treated with IFNα for 20 h at 1000 units/mL. The Ni-NTA bound materials were eluted in elution buffer (6 M guanidine, 10 mM Tris-HCl, 100 mM NaH2PO4, 500 mM imidazole, pH 8.0), clarified through 0.45 μm nitrocellulose membranes and buffer exchanged into SENP deconjugation buffer (150 mM NaCl, 1% triton (v/v), 25 mM Tris-HCl, pH 8.0). Control, SENP1/2 and USP18 reaction samples were prepared by using from 1 mg of total cell extract (~25 μg of purified material). For the SENP1/2 treated samples, 0.3 μg of SENP1 and 0.3 μg of SENP2 were added and the total volume completed to 40 μL with SENP deconjugation buffer. For the USP18-treated samples, 0.1 μg of USP18 was added and the total volume completed to 40 μL with SENP deconjugation buffer. All samples were incubated at 37 °C for 1 h.

2.5. Real-time qPCR

HeLa-wt and HeLa-SUMO3 cells were untreated or treated with 1000 units/mL of IFN for 8 h. Total RNAs were extracted using the RNeasy Micro Kit (Qiagen GmbH, Hilden, Germany). RNA quantity and quality was measured using the Nanodrop 2000. Fluidigm gene expression technology was used to study the expression of the genes of interest (Fluidigm Corporation, USA). cDNA pre-amplification was performed with the PreAmp Master Mix according to the manufacturer’s directions, and were cycled using the recommended program for 18 cycles. Later, the pre-amplified sample was diluted 1:10 and the primers were mixed according to the manufactureŕs directions with the reagents from the Flex six Delta Gene Master kit and loaded in their chip inlets. The cycling program used consisted of 10 min at 95 °C followed by 40 cycles of 95 °C for 15 sec and 1 min at 60 °C. Data were analyzed using the BioMark Gene Expression Data Analysis software to obtain Ct values. Then the data was exported to excel to calculate ΔΔCt values using control sample and RPL13A as a reference gene.

2.6. Cell fractionation

For total cell extracts, cells were washed in PBS, scraped into Laemmli buffer. For cell fractionation, cells were lysed in hypotonic buffer (10 mM Tris-HCl, pH 7.65, 1.5 mM MgCl2, 1 mM DTT, 20 mM Nethylmaleimide, protease inhibitors) and centrifuged at 500g for 15 min. The supernatant constituted the cytoplasmic fraction and the pellet resuspended in Laemmli buffer constituted the nuclear fraction. Proteins of the different extracts were analyzed by western blot. For the analysis of the RIPA soluble and insoluble fractions, the cytoplasmic and nucleoplasm fractions were lysed in RIPA buffer for 20 min on ice followed by centrifugation at 12 000g for 15 min to separate the RIPA soluble fraction (R) from the pellet (P) fractions. The RIPA soluble and insoluble fractions were then supplemented with Laemmli buffer.

2.7. Immunoprecipitation assays

Cells (107) were incubated for 30 min at 4 °C in 0.5 mL of buffer containing 20 mM Tris-HCl (pH 7.4), 1 M NaCl, 5 mM MgCl2, 1% Triton X-100 and 1 mM phenylmethylsulfonyl fluoride (PMSF). After cell lysis, 50 μL was saved for the input and 2.4 mL of immunoprecipitation buffer (IB) (20 mM Tris-HCl [pH 7.4], 150 mM NaCl, 0.5% deoxycholate [DOC], 1% Triton X-100, 0.1% SDS, and 1 mM EDTA) was added to the remaining 450 μL. The anti-MxB or SAMHD1 antibodies were added, and the samples were incubated overnight at 4 °C. Protein G beads (Sigma) in IB were then added, and the samples were mixed for 2 h at room temperature. The beads were collected and washed four times with modified IB (IB plus 150 mM NaCl), and the bound proteins were subjected to Western blotting.

2.8. Proteomic workflow for SUMO3 stabilized substrates in the presence and absence of IFNα

For the SUMO3 stabilized and destabilized proteome; 106 HEK293 or HEK293-SUMO3m cells were grown for each biological replicate in either light (0Lys, 0Arg) or medium (4Lys, 6Arg) SILAC media according to table “Experiment 1” in Fig. 3A. For the SUMO3m + IFNα stabilized and destabilized proteome; 106 HEK293 or HEK293-SUMO3m cells were grown in either light or medium SILAC media according to table “Experiment 2” in Fig. 3A and treated with IFNα for 20 h at 1000 units/ mL prior to cell collection. Cell pellets were resuspended in 100 μL of proteome lysis buffer (8 M urea, 50 mM Tris-HCl, pH 7.6), sonicated for 3 sec at medium amplitude on ice before centrifugation at 12 000g for 10 min. The protein content of each sample was measured using Bradford’s assay. 30 μg of each channel were pooled according to the tables in Fig. 3A and resolved on a 4–12% SDS-PAGE. Each lane was cut into 8 fractions, the gels dehydrated in acetonitrile (ACN), the proteins reduced and alkylated using 5 mM TCEP and 20 mM 2-chloroacetamide for 30 min at RT. The gels were dehydrated using ACN and the proteins digested with 300 ng of trypsin per fraction for 16 h at 37 °C. The supernatants were transferred to new tubes and the peptides extracted two times with 90% ACN in water. The extracted peptides were pooled with the supernatant and dried in a speed vacuum.
The peptides were desalted on C18 STAGE-tips as described in the literature [22] using two layers of C18 extraction disks per tip. The tips were conditioned with 50 μL of 0.1% TFA, 50% acetonitrile followed by two washes with 50 μL of 0.1% TFA. The peptides were reconstituted in 50 μL of 0.1% TFA and loaded on the equilibrated tips. The tips were washed twice with 55 μL of 0.1% TFA and the peptides eluted from the tips with 50 μL of 0.1% TFA, 40% ACN. The peptides were dried down in a speed vacuum prior to their analysis by LC-MS/MS.

2.9. SUMO proteomic workflow to quantify IFNα induced protein

SUMOylation events 108 cells were grown in either light (0Lys, 0Arg) or heavy (8Lys, 10Arg) SILAC media for each biological replicate and treated with IFNα for 20 h at 1000 units/mL or mock treated with PBS according to the table in Fig. S4A. Cell pellets were resuspended in 2 mL of buffer A (6 M guanidine, 10 mM Tris-HCl, 100 mM NaH2PO4, pH 8.0), sonicated for 10 sec at medium amplitude before centrifugation at 12 000g for 10 min. The protein content of each sample was measured using Bradford’s assay. 10 mg of each channel were pooled according to the table in Fig. S4A. The samples were supplemented with β-mercaptoethanol, 2-chloroacetamide and imidazole to 5 mM, 20 mM and 10 mM, respectively. Samples were allowed to react for 30 min at RT. 0.8 mL of Ni-NTA slurry was added to each sample. Proteins were allowed to bind to the resin overnight at 4 °C. The resin washed with 10 mL of buffer A, followed by 5 washed with 10 mL of buffer B (8 M urea, 100 mM NaH2PO4, 10 mM Tris-HCl, 5 mM 2-mercaptoethanol, 20 mM imidazole, pH 6.3). Beads were washed once with 5 mL of 50 mM ammonium bicarbonate and finally resuspended in 5 mL of 50 mM ammonium bicarbonate. 20 μg of trypsin was added to each sample and allowed to react overnight at 37 °C. Enriched peptides were acidified by adding TFA to a final concentration of 1%, desalted on 1 cc HLB cartridges as per the manufacturer’s instructions, eluted into microfuge tubes and lyophilized by speed vacuum.
For each sample, 180 μL of PureProteome protein A magnetic beads were incubated in PBS with 180 μg of anti-K(NQTGG) antibody for 1 h at 4 °C. The beads were washed three times with 200 mM triethanolamine, pH 8.3. 1.8 mL of 5 mM DMP in 200 mM triethanolamine, pH 8.3 was added for each sample and incubated 1 h at room temperature to covalently attach the antibody to the beads. After 1 h, 90 µL of 1 M Tris-HCl, pH 8 was added to each sample and the mixture incubated for 30 min at RT. The supernatant was removed and the antibody-linked beads were washed three times with ice cold PBS and once with PBS containing 50% glycerol.
The Ni-NTA enriched peptide mixtures were reconstituted in 500 µL of PBS containing 50% glycerol. The anti-K(NQTGG) bound beads were added to each sample and incubated for 1 h at 4 °C. The flow through was set aside and the beads washed two times with 1 mL of PBS followed by two washes with 1 mL of 0.1X PBS. The SUMO remnant modified peptides were eluted from the beads with two sequential elutions with 100 μL of 0.2% formic acid. The eluates were dried by speed vacuum.

2.10. Workflow for increased depth of SUMO proteome

1.5 108 cells were grown in DMEM and treated with IFNα for 20 h at 1000 units/mL and 16 h with 10 μM MG132. Cell pellets were resuspended in 8 mL of buffer A. 80 mg of protein extract were processed as indicated above with the appropriate scale up to all materials. NiNTA enrich proteins were digested with trypsin and fractionated into 5 fractions using high pH reverse phase fractionation on a C18 cartridge (WAT036820). Peptides were loaded on the cartridge in 0.2% formic acid and eluted in 10 mM NH4HCO2 pH 10 containing increasing concentrations of ACN. Peptides were eluted using 5% ACN steps from 5% to 80% ACN, providing 15 fractions. Every fifth fraction was pooled to prepare 5 concatenated fractions in total. Immunoprecipitations with 100 μg of K(NQTGG) antibody were performed on each fraction, as described above. The SUMO remnant modified peptides were eluted from the beads and dried by speed vacuum.

2.11. Mass spectrometry analysis

For the proteome samples, the peptides were reconstituted in 50 µL of water containing 0.2% formic acid and 10 µL of this mixture was injected by LC-MS/MS using an Orbitrap Q-Exactive HF (Thermo Fisher Scientific) coupled to a Proxeon Easy-nLC 1000. Peptides were resolved on a 150 μm × 20 cm nano LC column (Jupiter C18, 3 μm, 300 A,
Phenomenex). The separation was performed on a linear gradient from 4% to 40% acetonitrile, 0.2% formic acid over 225 min at 600 nL/min. Full MS scans were performed on ions from m/z 350 to m/z 1300 at resolution 60 000 at m/z 200, with a target AGC of 5E6 and a maximum injection time of 100 ms. MS/MS scans were acquired in HCD mode with a normalized collision energy of 27 and resolution 15 000 using a Top 15 method, with a target AGC of 2E4 and a maximum injection time of 400 ms. The MS/MS triggering threshold was set at 5E4 and the dynamic exclusion of previously acquired precursor was enabled for 25 s within a mass range of ± 0.8 Da. Ions with charges < 2 were excluded from triggering MS2 events. For the SUMO proteome samples the peptides were reconstituted in 25 µL of water containing 4% formic acid and 10 µL of this mixture was analyzed by LC-MS/MS. Peptides were resolved on a 150 μm × 20 cm LC column (Jupiter C18, 3 μm, 300 A, Phenomenex). The separation was performed on a linear gradient from 5% to 30% ACN, 0.2% formic acid over 106 min at 600 nL/min. Full MS scans were performed on ion from m/z 350 to m/z 1200 at resolution 120 000 at m/z 200, with a target AGC of 5E6 and a maximum injection time of 50 ms. MS/MS scans were acquired in HCD mode with a normalized collision energy of 25 and resolution 30 000 using a Top 5 method, with a target AGC of 2E4 and a maximum injection time of 1 s. The MS/MS triggering threshold was set at 5E3 and the dynamic exclusion of previously acquired precursor was enabled for 15 s within a mass range of ± 0.8 Da. Ions with charges > 5 and < 2 were excluded from triggering MS2 events. 2.12. Data processing MS/MS spectra were searched against Uniprot/SwissProt database including Isoforms (released on June 08, 2018, 20 349 entries) using MaxQuant (version 1.6.2.1) [23]. The precursor ion tolerance was set to 20 and 4.5 ppm for the first and main searches, respectively. The MS/ MS spectra search were set to a mass tolerance of 20 ppm. For the proteome samples: The maximum number of missed cleavages was set to 2 using trypsin as protease. Arg6 and Lys4 were included in the search for the medium labeled SILAC samples. Carbamidomethylation (C) was set as fixed modification and acetylation (Protein N term) and oxidation (M) were set as variable modifications. For the SUMO proteome samples: The maximum number of missed cleavage was set to 3 using trypsin/P as protease. Arg10 and Lys8 were included in the search for the heavy labeled SILAC samples. Carbamidomethylation (C) was set as fixed modification and acetylation (Protein N term), phosphorylation (STY), oxidation (M), deamidation (NQ), and NQTGG (K) for SUMO3 remnant were set as variable modifications. The match between runs function was enabled with a 20 min alignment window and 0.7 min match time window. Protein and peptides were filtered with a 1% FDR using the reverse database as decoy. The MS/MS spectra for modified peptides with an Andromeda score below 40 (default values) were discarded from further analysis. 3. Results 3.1. Differential effects of SUMO paralogs on ISG restriction factors upon IFNα stimulation Recently, we reported that MxA protein expression is higher in SUMO1- and SUMO3-expressing HeLa cells than in wild type (wt) cells [8]. Analysis of extracts from HeLa cells treated for 20 h with 1000 units/mL of IFNα revealed that MxA protein was enhanced in SUMO1expressing cells, and that a much higher increase was observed in cells expressing SUMO3 compared to wt cells (Fig. 1A). In contrast, the levels of MxB protein and Guanylate Binding Protein (GBP5), other members of GTPase family, were not affected by SUMO1 or SUMO3 in the absence of IFNα (Fig. 1B). However, IFNα induced a significant increase of MxB and GBP5 in HeLa-SUMO3 cells whereas the increase of MxB and GBP5 proteins was similar in HeLa-wt and HeLa-SUMO1 cells (Fig. 1B). Upregulation of MxB in HeLa-SUMO3 cells was also observed upon treatment with lower doses of IFNα (Fig. S1A). Overall, these results revealed that SUMO3 and not SUMO1 drastically increased MxA, MxB and GBP5 protein expression in response to IFNα. We reported previously that SUMO3 expression in HeLa cells treated with IFNα abrogates the increase of PML and Sp100 proteins due to their RNF4-dependent proteasomal degradation [9]. To gain further insights into the effect of SUMO3 on ISG restriction factors that exhibit antiviral activity, we analyzed the expression of TRIM5α, PML, Daxx, TRIM22, STAT1 and SAMHD1 by western blot. As expected, their expressions were enhanced in wt cells treated with IFNα (Fig. 1C and 1D) [24–28]. In the absence of IFNα, SUMO3 increased the expression of TRIM5α (Fig. 1C) and SAMHD1 (Fig. 1d), but did not significantly alter the expression of PML, Daxx and TRIM22 (Fig. 1C and 1D). Remarkably, western blot analysis revealed that upon IFNα treatment, SUMO3 had three different effects on these restriction factors, indeed, SUMO3 blocked the increase of TRIM5α, Daxx and PML (Fig. 1C), did not alter the increase of TRIM22 (Fig. 1D), and significantly enhanced the abundance of SAMHD1 and to a lower extent that of STAT1 (Fig. 1D). Similar to MxB and GBP5 (Fig. 1B), SAMHD1 expression was upregulated in IFNα-treated HeLa cells expressing SUMO3 and not SUMO1 (Fig. 1E). MxB and SAMHD1 levels were also enhanced in response to IFNα in HEK293 cells expressing SUMO3 compared to HEK293 parental cell line (Fig. S1B). On a side note, we found that phosphorylation at Thr592 of SAMHD1 was increased in response to IFNα, and was further enhanced in HeLa-SUMO3 cells (Fig. 1D). The enhancement of SAMHD1 phosphorylation in HeLa and THP-1 cells treated with IFNα or IFNγ was correlated with its increased protein expression (Fig. S1C). This is consistent with a recent report showing a cell type-specific regulation of SAMHD1 phosphorylation in response to IFN [29]. To determine if SUMO3 dependent changes of restriction factor abundances were correlated with their transcript profiles, we used RTqPCR to quantify their mRNA levels in extracts from HeLa-wt and HeLaSUMO3 cells untreated or treated with IFNα for 8 h (Fig. S2). The mRNA expressions of MX1, MX2, SAMHD1, PML/TRIM19, TRIM5α, TRIM22, Daxx and STAT1 were induced by IFNα, but were refractory to SUMO3 expression (Fig. S2). These results are in agreement with previous findings showing that overexpression of SUMO does not affect transcriptional responses of IFNα [9], and suggest that changes in protein expression of ISG restriction factors are regulated by posttranscriptional events. 3.2. SUMO3 enhances in response to IFNα MxB and SAMHD1 interaction and increases anti-HIV-1 and anti-HSV-1 activities We showed in Fig. 1 that in response to IFNα, SUMO3 expression enhanced many restriction factors known to confer resistance to HIV-1 and/or HSV-1 such as MxB, SAMHD1 and GBP5 [30–35]. Also we found that modified forms of MxB and SAMHD1 were only expressed in the nucleus of IFNα-treated HeLa-SUMO3 cells (Fig. S1D). To corroborate these observations, we investigated by co-immunoprecipitation a possible interaction between MxB, SAMHD1 and SUMO3. Untreated and IFNα-treated HeLa-wt and HeLa-SUMO3 cells were lysed and separated into two fractions. One fraction was immunoprecipitated with anti-MxB antibodies and analyzed by western blot using anti-SAMHD1 or antiSUMO2/3 antibodies (Fig. 2A). The second fraction was immunoprecipitated with anti-SAMHD1 antibodies and analyzed by western blot, using anti-MxB or anti-SUMO2/3 antibodies (Fig. 2B). MxB antiserum precipitated SAMHD1 and, reciprocally, only from fractions derived from IFNα-treated SUMO3 cells. In addition, probing the blots with anti-SUMO2/3 antibodies revealed that in SUMO3 expressing cells, IFNα enhanced the interaction of either MxB (Fig. 2A) or SAMHD1 (Fig. 2B) with their SUMOylated partners. Taken together these results show that MxB and SAMHD1 interact together and with SUMO3 in response to IFNα, suggesting that under certain conditions MxB and SAMHD1 cooperate to confer viral resistance in IFNα-treated cells [36]. Next, we evaluated whether MxB and SAMHD1 were covalently conjugated to SUMO3 and whether IFNα treatment enhanced their modification. MxB and SAMHD1 immunoblots of Ni-NTA affinity-purified extracts from HeLa-SUMO3 cells showed that both proteins are modified by SUMO3 and that IFNα enhanced their SUMOylation (Fig. 2C). Taken together, these results show that IFNα enhanced the interaction between MxB and SAMHD1 and promoted their SUMOylation. To determine if there is a correlation between the increase in the expression of these proteins and viral inhibition, we analyzed the effect of IFNα in wt and SUMO3-expressing cells on HIV-1 and HSV-1 replication. The anti-HIV-1 effect of IFNα was higher in SUMO3-expressing cells compared to wt cells (Fig. 2D). Also, IFNα inhibited HSV-1 replication by 30-fold in wt cells and by more than 400-fold in SUMO3expressing cells (Fig. 2E). The inhibition of HSV-1 was correlated with a higher decrease in the expression of the viral protein ICP0 in IFNαtreated SUMO3 cells compared to IFNα-treated wt cells (Fig. 2F). Thus, SUMO3 and IFNα cooperate to enhance the expression of restriction factors that are essential to confer viral resistance. 3.3. SUMO3 expression stabilizes several ISG products Having uncovered by western blot analyses that SUMO3 expression led to changes in the abundance of several ISG products in HeLa cells, we sought to identify in a non-biased manner if other proteins are stabilized/destabilized by SUMO3 during IFNα treatment. Accordingly, we performed system-wide proteomic analyses in HEK293 cells stably expressing 6xHisSUMO-3-Q87R-Q88N cells [21]. We refer to this protein as SUMO3m to indicate the presence of the Nterminal His6 tag and the insertion of a strategically located tryptic cleavage site at the C terminus to facilitate the recovery and identification of SUMOylated peptides by affinity enrichment and mass spectrometry. Two separate experiments were developed to profile changes in protein abundance of HEK293-SUMO3m cells untreated or treated with IFNα (Fig. 3A). This workflow permits to distinguish which proteins are selectively stabilized/destabilized by SUMO3 in response to IFNα from those that are regulated by SUMO3 independently of the IFNα treatment. The same workflow was used for both proteomic experiments and was conducted in parallel on four biological replicates. HEK293-wt or HEK293-SUMO3m cells were grown in either light or medium media and combined in a 1:1 fashion based on protein content as indicated in Fig. 3A. The combined protein extracts were separated by SDS-PAGE and sliced into 8 separate bands. The proteins were digested in gel with trypsin, desalted on C18 STAGE-tips and analyzed by LC-MS/MS. Altogether we identified 6736 proteins, of which 5187 proteins were quantified across at least 3 out of 4 biological replicates. The overexpression of SUMO3m in HEK293 cells led to the regulation of 1131 proteins, of which 579 and 551 proteins were up- and downregulated respectively (Fig. 3B, and Table S1). SUMO3 expression enhanced the abundance of several proteins involved in intrinsic antiviral activity (Orange circles on Fig. 3B), including, IFIT1, IFIT5, SAMHD1, GBP1 and PKR (Table 1, Table S1, Table S2, Fig. 3B). A similar overall trend in the changes in protein levels was observed when HEK293-wt and HEK293-SUMO3m cells were treated with IFNα. The elevated levels of SUMO3 lead to changes in the abundance of 1170 proteins, 583 of which were upregulated and 586 were downregulated (Table 1, Table S1). Although the overall trend was similar in both sets of experiments the biological outcome was different. Indeed, SUMO3 enhanced the abundance of proteins involved in the type I IFN response as shown by the orange circles in Fig. 3C. Notably, we observed that SAMHD1 was stabilized by SUMO3 during IFNα treatment, in line with our previous results (Fig. 1). Importantly, in addition to MxA, MxB, GBP5 and SAMHD1 that were shown by western blot analysis to be upregulated in IFNα-treated SUMO3 cells (Fig. 1), LC-MS/MS analyses revealed that several restriction factors playing key roles in antiviral defense were also highly expressed, including PML, GBP1, TRIM21, SAMHD1, Tetherin/BST2, PKR, and members of IFITM, IFIT and IFI families (Table 1, Table S1, Fig. 3C). Note that in response to IFNα, PML was downregulated in HeLa-SUMO3 cells (Fig. 1C) and upregulated in HEK293-SUMO3m cells (Fig. 3C), suggesting a differential role for SUMOylation in the processing of PML nuclear bodies in a cell type dependent manner. Moreover, nearly all components of the ISG15 conjugation machinery (UBE2L6, TRIM25 and HERC5) were stabilized by SUMO3 in IFNα-treated HEK293 cells, except for ISG15 that was downregulated. This apparent decrease in ISG15 abundance was ultimately due to the enhanced ISGylation of protein substrates and not to a reduced ISG15 expression (Fig. 4A). Using a two-sample t-test to compare changes in protein abundance in experiment 2 to those of experiment 1, we identified 41 proteins that were upregulated in IFNα-treated HEK293-SUMO3m cells (Table S1, Fig. S3) including several restriction factors (Table 1). The two samples t-test allowed us to determine if a positive regulation occurred between SUMO3 expression and IFNα treatment, and if proteins that were stabilized by SUMO3 expression (experiment 1) were further stabilized by the addition of IFNα (experiment 2). We also analyzed proteins that were either stabilized or destabilized by SUMO3 in the absence or the presence of IFNα. This list of upregulated proteins includes ISG15 ligases (TRIM25 and HERC5) and proteins involved in IFN synthesis, IFN signaling and antiviral defense (Table 1, Fig. 3, Fig. S3, Tables S1 and S2). Having shown that several proteins involved in the IFN pathway were stabilized by SUMO3 expression in response to IFNα (Fig. 1), and that SUMO3 expression did not alter their mRNA level (8 h post-IFNα treatment) (Fig. S2), we set out to determine if proteins that are stabilized by SUMO3 are themselves directly modified by SUMO3. Although we previously studied the changes in the SUMO proteome in response to IFNα [37], these experiments were conducted in technical replicates only and at different time points (0.75 h and 16 h) than those used in the present study (20 h). Accordingly, we used a SUMO proteomic strategy [38,39] where biological replicates of HEK293SUMO3m cells were expanded in either light or heavy media and treated for 20 h with IFNα at 1000 units/mL (Fig. S4A). Heavy and light labeled cell lysates (10 mg each) were pooled and SUMOylated proteins were enriched on Ni-NTA beads, digested on beads with trypsin and the resulting peptides desalted by reverse phase chromatography. The SUMO modified peptides were affinity purified using an anti-K (NQTGG) antibody that recognized the SUMO remnant produced on the SUMO modified lysine residue upon tryptic digestion. The corresponding peptides were fractionated by strong cation exchange (SCX) chromatography and analyzed by LC-MS/MS. The results from the SUMO proteome analysis are depicted on the volcano plot in Fig. S4B. In total, 148 SUMOylation sites were regulated by IFNα (Table S3). The SUMOylation abundance decreased on 47 of these sites while 101 sites increased in SUMOylation. We noted a dramatic increase in PML SUMOylation at multiple sites and of STAT1 SUMOylation at Lys703, consistent with our previous report for cells treated for 16 h with IFNα [37]. The enhanced coverage of the present SUMO proteome study enabled the profiling of PML SUMOylation on 9 different lysine residues (Lys65, Lys380, Lys394, Lys400, Lys476, Lys478, Lys487, Lys490 and Lys497). In addition, we were able to quantify the increased SUMOylation of SAMHD1 at Lys622 that was undetected in our previous study [12]. Of note, the SUMOylation level at Lys469 of SAMHD1 was not regulated by the IFNα treatment. Lastly, we identified 7 SUMOylation sites on the RNA-specific adenosine deaminase ADAR1, where both Lys384 and Lys418 increased in SUMOylation in response to the IFNα treatment. The increase in SUMOylation on different proteins was low relative to the number of proteins that were stabilized by SUMO3 overexpression in the proteome studies (Fig. 3C). This low number of identified SUMOylation events on IFNα-upregulated proteins may not be a consequence of a reduced depth of the proteome coverage in the SUMO proteomic analysis since the SUMO proteome actually provided a modest increase in the depth of the proteome coverage as highlighted in the circled portion of Fig. S4C [40]. To address this concern, we also performed a deeper profiling of the SUMO proteome by using 8 times more materials and inhibiting the proteasome with MG132 during the IFN treatment to identify other IFN-regulated proteins that are SUMOylated at low levels (Table S4). Indeed, with this more sensitive approach we identified over 50 SUMOylation sites on 14 ISG products including the restriction factors PML, SAMHD1, PKR, ISG20, ADAR1 and IFIT1 (summarized in Table 1 and Table S4). Taken together, these results suggest that several proteins involved in the antiviral property of type I IFN are SUMOylated. SUMO3 could also be implicated in stabilizing some ISG products by a mechanism that is independent of their direct modification by SUMO3. It should be noted that key players of type I IFN signaling namely IRF9, STAT1 and STAT2 were highly stabilized after IFNα treatment, and could participate in the increase of ISG products in a SUMO-independent manner. However, we cannot exclude that the stability imparted by SUMO3 expression could be a consequence of a non-covalent interaction with SUMO3. 3.4. IFNα promotes protein ISGylation by TRIM25 in a SUMO3 dependent manner In accordance with the mass spectrometry results, our western blot analysis showed that p53 and the free ISG15 protein were downregulated in IFNα-treated HEK293-SUMO3m cells (Table S1 and Fig. 4A). Remarkably, the decrease of free ISG15 in these cells was accompanied by a drastic increase of global cellular ISGylation (Fig. 4A). In HeLa cells, SUMO3 expression enhanced both the expression of free ISG15 and IFNα-induced ISGylation (Fig. 4B). Importantly, the upregulation of ISGylation in response to IFNα was observed in SUMO3- but not in SUMO1-expressing HeLa cells (Fig. 4B). In addition, we validated that both GBP1 (Fig. 4C) and Sterile Alpha Motif containing domain 9 (SAMD9) (Fig. 4C) were more abundant in response to IFNα in cells expressing SUMO3 than in wt cells. We also validated the downregulation of IFI16 in IFNα-treated HeLa-SUMO3 compared to IFNα-treated HeLa-wt cells (Fig. 4C). Furthermore, western blot analysis of Ni-NTA enriched samples revealed that IFNα drastically enhanced global cellular ISGylation and ubiquitination in cells overexpressing SUMO3 (Fig. 4D), demonstrating that ISG15 may modify SUMO2/3 conjugated proteins as previously shown for ubiquitin [18]. To further evaluate the cross-talk between SUMOylation and ISGylation, we analyzed the effect of UBC9 depletion on ISGylation in IFNα-treated HeLa-SUMO3 cells (Fig. 4E). As expected, depletion of UBC9 decreased cellular SUMOylation and remarkably resulted in a decrease of global cellular ISGylation (Fig. 4E). Next, we performed a western blot to determine the levels of SUMOylated and ISGylated proteins in the inputs and the Ni-NTA flow through of IFNα-treated SUMO3 cells (Fig. 4F). Analysis of the Ni-NTA flow through samples revealed the absence of SUMOylated proteins and the presence of ISGylated proteins, demonstrating that ISG15 is conjugated to substrates independently to their SUMOylation. Further, we invoked the use of the SUMO and ISG15 deconjugation assays to ascertain whether ISG15 conjugation occurs on the SUMO3 moiety of the modified proteins. To this end, we removed SUMO by adding SENP1/2 and ISG15 by adding USP18 to the Ni-NTA enriched substrates and probed for SUMO2/3 and ISG15 conjugation levels (Fig. 4G). After the deconjugation assays, we observed a decrease in the conjugated ISG15 levels suggesting that a small portion of ISG15 form mixed SUMO chains while a larger portion of ISG15 occupy other sites on SUMOylated proteins. Taken together, these results indicate that the ISGylation occurs in SUMO-dependent and independent manners. Therefore, enhanced global ISGylation in IFNα-treated SUMO3 cells could also take place on non-SUMOylated substrates and could prevent their proteasomal degradation. Furthermore, the levels of the ISG15 E3 ligases (TRIM25 and HERC5) were increased upon IFNα treatment for both HEK293SUMO3m and HeLa-SUMO3 cells (Fig. 3C, Fig. 5A, Table S1). SiRNA mediated depletion of HERC5 or TRIM25 in HeLa-SUMO3 cells led to the expected decreased in global ISGylation, but did not affect global protein SUMOylation (Fig. 5B). Remarkably, TRIM25 knockdown reduced the upregulation of SAMHD1, GBP5 and MxB in IFNα-treated SUMO3 cells, whereas HERC5 depletion did not alter the expression of these restriction factors. TRIM25 depletion in IFNα-treated SUMO3 cells did not alter the downregulation of IFI16 (Fig. S5), whereas RNF4 depletion prevented IFI16 degradation, thus restoring an increase of IFI16 protein in response to IFNα (Fig. 5C). Taken together, these results show that TRIM25 and RNF4, respectively, drive the SUMO3-dependant stabilization and destabilisation of target proteins in response to IFNα. Even though we did not identified TRIM25 SUMOylation sites by proteomic analysis, TRIM25 immunoblot of Ni-NTA affinity-purified extracts from HeLa-SUMO3 cells showed that TRIM25 was modified by SUMO3 and that IFNα enhanced its SUMOylation (Fig. 5D). To determine whether the SUMOylation of TRIM25 might modify its cellular localization we analyzed by western blot the RIPA-soluble (the cytoplasm and most of the nucleoplasm) and insoluble (the nuclear matrix and some chromatin components) fractions from HeLa-wt and HeLaSUMO3 cells untreated or treated with IFNα (Fig. 5E). The unmodified form of TRIM25 was found in the RIPA soluble fractions of all samples. In contrast, the SUMO-modified TRIM25 was found only in the RIPAinsoluble fraction of IFNα-treated HeLa-SUMO3 cells (Fig. 5E). SUMO2/3 and ISG15 immunoblots revealed that in IFNα-treated HeLaSUMO3 cells (Fig. 5E) the RIPA-insoluble fraction contained, in addition to SUMOylated TRIM25, most of the SUMOylated proteins and a small portion of ISGylated proteins. These results revealed that IFNα enhanced TRIM25 SUMOylation and shifted SUMO-modified TRIM25 to the nuclear matrix where most of SUMOylated proteins and a small portion of ISGylated proteins were also found. 4. Discussion IFN plays important roles on innate and adaptive immune cells during viral infections by inducing a program of gene transcription that regulates key mediators of antiviral response. Increasing evidences also indicate that SUMOylation can play a role in the regulation of innate immunity and the synthesis of IFN upon viral infection [41–43], thus suggesting a possible interplay between IFN and Ubiquitin-like modifiers (ULMs). Indeed, previous reports indicated that IFN was shown to enhance global levels of cellular SUMOylation [37], ubiquitination [44] and ISGylation [45]. Here, we report that stable expression of SUMO3, but not SUMO1, upregulated IFNα-induced global cellular ubiquitination and ISGylation, altering expression levels of several ISG products. Our large-scale proteomic analyses revealed that in response to IFNα, SUMO3 overexpression led to the down-regulation of 586 proteins such as IFI16, and that IFI16 expression was restored upon RNF4 depletion as previously shown for PML and Sp100 [9]. Interestingly, SUMO3 overexpression positively regulated the expression of 583 proteins in response to IFNα, including several proteins playing key roles in IFN production, IFN signaling, antiviral defense and ISG15 pathway. Indeed, some of these upregulated proteins such as GBP1, GBP5, IFITM1, IFITM2, IFITM3, IFIT1, IFIT3, IFIT5, PKR, SAMHD1, Tetherin/BST2 and MxB act as restriction factors against HSV-1 and/or HIV-1 infections [30,32,34,35,46–51]. In addition, the increase in abundance of these restriction factors in IFNα-treated SUMO3 expressing cells was correlated with an enhancement of IFNα-induced anti-HIV-1 and antiHSV-1 activities. Also, we report that MxB interacted with SAMHD1 in IFNα-treated HeLa-SUMO3 cells and that IFNα enhanced their SUMOylation, suggesting that proteins stabilized by SUMO3 during the IFNα response may be part of several interacting complexes. Therefore, MxB in conjunction with SAMHD1 and SUMO3 could contribute to IFNα-induced antiviral defense. However, the extent of their respective contributions in response to IFNα is difficult to assess due to the growing number of the restriction factors and the evasion strategies elaborated by viruses[52–54]. SUMO3 overexpression did not alter the transcriptional response or increase ISG mRNAs 8 h post-IFNα treatment [9], but upregulated STAT1, STAT2 and IRF9 protein levels 20 h post-treatment. The upregulation of these key players involved in type I IFN signaling could enhance ISG products independently of their SUMOylation, and may explain why so few SUMOylated targets varied in response to IFNα while over 500 proteins changed in abundance during the same time period. We have shown previously that under proteasome inhibition SAMHD1 is SUMOylated at various sites (K66, K148, K446, K455, K467, K469, K484, K492, K494, K534, K544, K595, K622) [12]. Our SUMO proteome revealed that 101 SUMOylation sites were increased in response to IFNα, including SAMHD1 Lys622, STAT1 Lys703 and several sites on PML (Lys65, Lys160, Lys209, Lys226, Lys337, Lys380, Lys394, Lys400, Lys401, Lys476, Lys478, Lys487, Lys490, Lys497 and Lys616). In addition, we identified 7 SUMOylation sites on ADAR1, where both Lys384 and Lys418 SUMOylation increased in response to IFNα. Furthermore, 50 other SUMOylation sites were enhanced by IFNα in the presence of MG132 including PKR (Lys256), IFIT1 (Lys199, Lys336, Lys370, Lys407), IRF1 (Lys117) and ISG20 (Lys95). Further experiments are needed to determine whether the SUMOylation of the restriction factors such as SAMHD1, IFIT1, ADAR1 is required for their antiviral activities. In the absence of IFN, ectopic expression of SUMO3 stabilized many proteins including the restriction factors implicated in antiviral defense such as SAMHD1, TRIM5α, PKR, IFIT5. This finding is consistent with recent reports indicating that SUMOylated MxA [11] is stabilized in SUMO-expressing cells and mediates a SUMO-induced Vesicular Stomatitis Virus resistance, independently of IFN production [8]. Importantly, this study highlights an unsuspected interplay between SUMOylation and ISGylation, whereby SUMO3 overexpression led to a stabilization of the ISG15 E2 conjugating enzyme UBE2L6 and the E3 ISG15 ligases TRIM25 and HERC5. Interestingly, IFNα enhanced TRIM25 SUMOylation, localizing its SUMOylated form to the nuclear matrix, where most of the SUMOylated proteins and a small portion of the ISGylated proteins reside. Furthermore, TRIM25 depletion in IFNαtreated SUMO3 cells impaired protein ISGylation, and reduced the abundance of SAMHD1, GBP5 and MxB proteins, demonstrating that TRIM25 is a key player in the SUMO3 mediated protein stabilization. The functional significance of protein ISGylation is not fully understood. Unlike ubiquitination, ISGylation seems to counteract proteasome-dependent protein degradation via the conjugation of ISG15 to different E2 and E3 ubiquitin-conjugating enzymes [55] or via the formation of mixed ubiquitin-ISG15 chains [56], thus resulting in a decrease of polyubiquitinated protein levels and protein protection from proteasomal degradation. Many of the ISG products stabilized by SUMO3 have been previously shown to be ISGylated such as MxA, GBP1, SAMHD1, PKR, IFIT1, IFIT3, UBE2L6 and HERC5 [57,58]. Therefore enhanced IFNα-induced global ISGylation could also directly protect some proteins from proteasomal degradation. SUMO1 and SUMO2/3 modify both mutual and distinct substrates and several reports suggest that they may have individual functions [5,7,9,16,18,59,60]. Here we provide a novel differential effect of SUMO1 and SUMO3 on the positive regulation of ISG products and IFNα-induced ISGylation. In addition, enhanced IFNα-induced ISGylation in SUMO3-expressing cells targeted both SUMOylated and non SUMOylated proteins protecting them from degradation. While SUMO2/3 was previously reported to target proteins for RNF4-dependent proteasomal degradation [7,9,18], our results highlighted a novel mechanism by which SUMO3 can stabilize several restriction factors in a TRIM25-dependent fashion following cell stimulation by IFNα. Collectively, our findings led to the following proposal where (i) IFNα enhances cellular SUMOylation that triggers IFNα-induced ubiquitination and ISGylation of various proteins resulting in different outcomes; (ii) polySUMO chains can interact with RNF4 resulting in the ubiquitination and degradation of SUMOylated substrates; (iii) polySUMO chains can interact with TRIM25 resulting in the ISGylation and stabilization of SUMOylated proteins; (iv) ubiquitination and ISGylation can also occur on proteins independently of their SUMOylation. These novel findings reinforce the relevance of the SUMO pathway in controlling both the stability and possibly the functions of specific restriction factors to mediate the IFN response. References [1] L.B. Ivashkiv, L.T. Donlin, Regulation of type I interferon responses, Nat. Rev.Immunol. 14 (1) (2014) 36–49. [2] W.M. Schneider, M.D. Chevillotte, C.M. Rice, Interferon-stimulated genes: a complex web of host defenses, Annu. Rev. Immunol. 32 (2014) 513–545. [3] W. Wang, L. Xu, J. Su, M.P. Peppelenbosch, Q. Pan, Transcriptional regulation of antiviral interferon-stimulated genes, Trends Microbiol. 25 (7) (2017) 573–584. [4] M.K. Chelbi-Alix, J. Wietzerbin, Interferon, a growing cytokine family: 50 years of interferon research, Biochimie 89 (6–7) (2007) 713–718. [5] J.T. Crowl, D.B. Stetson, SUMO2 and SUMO3 redundantly prevent a noncanonical type I interferon response, Proc. Natl. Acad. Sci. USA 115 (26) (2018) 6798–6803. [6] Z. Hannoun, G. Maarifi, M.K. Chelbi-Alix, The implication of SUMO in intrinsic and innate immunity, Cytokine Growth Factor Rev. 29 (2016) 3–16. [7] G. Maarifi, F. El Asmi, M.A. Maroui, L. Dianoux, M.K. Chelbi-Alix, Differential effects of SUMO1 and SUMO3 on PKR activation and stability, Sci. Rep. 8 (1) (2018) 1277. [8] G. Maarifi, Z. Hannoun, M.C. Geoffroy, F. El Asmi, K. Zarrouk, S. Nisole, D. Blondel, M.K. Chelbi-Alix, MxA mediates SUMO-induced resistance to vesicular stomatitis virus, J. Virol. 90 (14) (2016) 6598–6610. [9] G. Maarifi, M.A. Maroui, J. Dutrieux, L. Dianoux, S. Nisole, M.K. Chelbi-Alix, Small ubiquitin-like modifier alters IFN response, J. Immunol. 195 (5) (2015) 2312–2324. [10] U. Sahin, O. Ferhi, X. Carnec, A. Zamborlini, L. Peres, F. Jollivet, A. VitalianoPrunier, H. de The, V. Lallemand-Breitenbach, Interferon controls SUMO NSC697923 availability via the Lin28 and let-7 axis to impede virus replication, Nat. Commun. 5 (2014) 4187.
[11] C.E. Brantis-de-Carvalho, G. Maarifi, P.E. Goncalves Boldrin, C.F. Zanelli, S. Nisole, M.K. Chelbi-Alix, S.R. Valentini, MxA interacts with and is modified by the SUMOylation machinery, Exp. Cell Res. 330 (1) (2015) 151–163.
[12] F. Lamoliatte, F.P. McManus, G. Maarifi, M.K. Chelbi-Alix, P. Thibault, Uncovering the SUMOylation and ubiquitylation crosstalk in human cells using sequential peptide immunopurification, Nat. Commun. 8 (2017) 14109.
[13] R.D. Everett, C. Boutell, B.G. Hale, Interplay between viruses and host sumoylation pathways, Nat. Rev. Microbiol. 11 (6) (2013) 400–411.
[14] E.S. Johnson, Protein modification by SUMO, Annu. Rev. Biochem. 73 (2004) 355–382.
[15] J.R. Gareau, C.D. Lima, The SUMO pathway: emerging mechanisms that shape specificity, conjugation and recognition, Nat. Rev. Mol. Cell Biol. 11 (12) (2010) 861–871.
[16] A. Flotho, F. Melchior, Sumoylation: a regulatory protein modification in health and disease, Annu. Rev. Biochem. 82 (2013) 357–385.
[17] V. Lallemand-Breitenbach, M. Jeanne, S. Benhenda, R. Nasr, M. Lei, L. Peres, J. Zhou, J. Zhu, B. Raught, H. de The, Arsenic degrades PML or PML-RARalpha through a SUMO-triggered RNF4/ubiquitin-mediated pathway, Nat. Cell Biol. 10 (5) (2008) 547–555.
[18] M.H. Tatham, M.C. Geoffroy, L. Shen, A. Plechanovova, N. Hattersley, E.G. Jaffray, J.J. Palvimo, R.T. Hay, RNF4 is a poly-SUMO-specific E3 ubiquitin ligase required for arsenic-induced PML degradation, Nat. Cell Biol. 10 (5) (2008) 538–546.
[19] M.C. Geoffroy, R.T. Hay, An additional role for SUMO in ubiquitin-mediated proteolysis, Nat. Rev. Mol. Cell Biol. 10 (8) (2009) 564–568.
[20] J. D’Cunha, S. Ramanujam, R.J. Wagner, P.L. Witt, E. Knight Jr., E.C. Borden, In vitro and in vivo secretion of human ISG15, an IFN-induced immunomodulatory cytokine, J. Immunol. 157 (9) (1996) 4100–4108.
[21] F. Galisson, L. Mahrouche, M. Courcelles, E. Bonneil, S. Meloche, M.K. Chelbi-Alix, P. Thibault, A novel proteomics approach to identify SUMOylated proteins and their modification sites in human cells, Mol. Cell. Proteomics 10 (2) (2011) M110–004796.
[22] J. Rappsilber, M. Mann, Y. Ishihama, Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using StageTips, Nat.Protoc. 2 (8) (2007) 1896–1906.
[23] J. Cox, M. Mann, MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification, Nat. Biotechnol. 26 (12) (2008) 1367–1372.
[24] K. Asaoka, K. Ikeda, T. Hishinuma, K. Horie-Inoue, S. Takeda, S. Inoue, A retrovirus restriction factor TRIM5alpha is transcriptionally regulated by interferons, Biochem. Biophys. Res. Commun. 338 (4) (2005) 1950–1956.
[25] M.K. Chelbi-Alix, L. Pelicano, F. Quignon, M.H. Koken, L. Venturini, M. Stadler, J. Pavlovic, L. Degos, H. de The, Induction of the PML protein by interferons in normal and APL cells, Leukemia 9 (12) (1995) 2027–2033.
[26] K. Ota, T. Matsumiya, H. Sakuraba, T. Imaizumi, H. Yoshida, S. Kawaguchi, S. Fukuda, K. Satoh, Interferon-alpha2b induces p21cip1/waf1 degradation and cell proliferation in HeLa cells, Cell Cycle 9 (1) (2010) 131–139.
[27] K. Shimoda, K. Kamesaki, A. Numata, K. Aoki, T. Matsuda, K. Oritani, S. Tamiya,K. Kato, K. Takase, R. Imamura, T. Yamamoto, T. Miyamoto, K. Nagafuji, H. Gondo, S. Nagafuchi, K. Nakayama, M. Harada, Cutting edge: tyk2 is required for the induction and nuclear translocation of Daxx which regulates IFN-alpha-induced suppression of B lymphocyte formation, J. Immunol. 169 (9) (2002) 4707–4711.
[28] C. Tissot, N. Mechti, Molecular cloning of a new interferon-induced factor that represses human immunodeficiency virus type 1 long terminal repeat expression, J.Biol. Chem. 270 (25) (1995) 14891–14898.
[29] K. Schott, N.V. Fuchs, R. Derua, B. Mahboubi, E. Schnellbacher, J. Seifried, C. Tondera, H. Schmitz, C. Shepard, A. Brandariz-Nunez, F. Diaz-Griffero, A. Reuter, B. Kim, V. Janssens, R. Konig, Dephosphorylation of the HIV-1 restriction factor SAMHD1 is mediated by PP2A-B55alpha holoenzymes during mitotic exit, Nat. Commun. 9 (1) (2018) 2227.
[30] M. Crameri, M. Bauer, N. Caduff, R. Walker, F. Steiner, F.D. Franzoso, C. Gujer, K. Boucke, T. Kucera, A. Zbinden, C. Munz, C. Fraefel, U.F. Greber, J. Pavlovic, MxB is an interferon-induced restriction factor of human herpesviruses, Nat. Commun. 9 (1) (2018) 1980.
[31] D.C. Goldstone, V. Ennis-Adeniran, J.J. Hedden, H.C. Groom, G.I. Rice, E. Christodoulou, P.A. Walker, G. Kelly, L.F. Haire, M.W. Yap, L.P. de Carvalho, J.P. Stoye, Y.J. Crow, I.A. Taylor, M. Webb, HIV-1 restriction factor SAMHD1 is a deoxynucleoside triphosphate triphosphohydrolase, Nature 480 (7377) (2011) 379–382.
[32] C. Goujon, O. Moncorge, H. Bauby, T. Doyle, C.C. Ward, T. Schaller, S. Hue, W.S. Barclay, R. Schulz, M.H. Malim, Human MX2 is an interferon-induced postentry inhibitor of HIV-1 infection, Nature 502 (7472) (2013) 559–562.
[33] J.A. Hollenbaugh, P. Gee, J. Baker, M.B. Daly, S.M. Amie, J. Tate, N. Kasai, Y. Kanemura, D.H. Kim, B.M. Ward, Y. Koyanagi, B. Kim, Host factor SAMHD1 restricts DNA viruses in non-dividing myeloid cells, PLoS Pathog. 9 (6) (2013) e1003481.
[34] M. Kane, S.S. Yadav, J. Bitzegeio, S.B. Kutluay, T. Zang, S.J. Wilson, J.W. Schoggins, C.M. Rice, M. Yamashita, T. Hatziioannou, P.D. Bieniasz, MX2 is an interferoninduced inhibitor of HIV-1 infection, Nature 502 (7472) (2013) 563–566.
[35] C. Krapp, D. Hotter, A. Gawanbacht, P.J. McLaren, S.F. Kluge, C.M. Sturzel, K.Mack, E. Reith, S. Engelhart, A. Ciuffi, V. Hornung, D. Sauter, A. Telenti, F. Kirchhoff, Guanylate Binding Protein (GBP) 5 Is an Interferon-Inducible Inhibitor of HIV-1 Infectivity, Cell Host Microbe 19 (4) (2016) 504–514.
[36] C. Buffone, J. Kutzner, S. Opp, A. Martinez-Lopez, A. Selyutina, S.A. Coggings, L.R. Studdard, L. Ding, B. Kim, P. Spearman, T. Schaller, F. Diaz-Griffero, The ability of SAMHD1 to block HIV-1 but not SIV requires expression of MxB, Virology 531 (2019) 260–268.
[37] M.A. Maroui, G. Maarifi, F.P. McManus, F. Lamoliatte, P. Thibault, M.K. ChelbiAlix, Promyelocytic leukemia protein (PML) requirement for interferon-induced global cellular SUMOylation, Mol. Cell. Proteom. 17 (6) (2018) 1196–1208.
[38] F. Lamoliatte, D. Caron, C. Durette, L. Mahrouche, M.A. Maroui, O. Caron-Lizotte, E. Bonneil, M.K. Chelbi-Alix, P. Thibault, Large-scale analysis of lysine SUMOylation by SUMO remnant immunoaffinity profiling, Nat. Commun. 5 (2014) 5409.
[39] F.P. McManus, F. Lamoliatte, P. Thibault, Identification of cross talk between SUMOylation and ubiquitylation using a sequential peptide immunopurification approach, Nat. Protoc. 12 (11) (2017) 2342–2358.
[40] J.R. Wisniewski, M.Y. Hein, J. Cox, M. Mann, A “proteomic ruler” for protein copy number and concentration estimation without spike-in standards, Mol. Cell.Proteom. 13 (12) (2014) 3497–3506.
[41] T. Kubota, M. Matsuoka, T.H. Chang, P. Tailor, T. Sasaki, M. Tashiro, A. Kato, K. Ozato, Virus infection triggers SUMOylation of IRF3 and IRF7, leading to the negative regulation of type I interferon gene expression, J. Biol. Chem. 283 (37) (2008) 25660–25670.
[42] K. Nakagawa, H. Yokosawa, PIAS3 induces SUMO-1 modification and transcriptional repression of IRF-1, FEBS Lett. 530 (1–3) (2002) 204–208.
[43] Y. Ran, T.T. Liu, Q. Zhou, S. Li, A.P. Mao, Y. Li, L.J. Liu, J.K. Cheng, H.B. Shu, SENP2 negatively regulates cellular antiviral response by deSUMOylating IRF3 and conditioning it for ubiquitination and degradation, J. Mol. Cell. Biol. 3 (5) (2011) 283–292.
[44] U. Seifert, L.P. Bialy, F. Ebstein, D. Bech-Otschir, A. Voigt, F. Schroter, T. Prozorovski, N. Lange, J. Steffen, M. Rieger, U. Kuckelkorn, O. Aktas, P.M. Kloetzel, E. Kruger, Immunoproteasomes preserve protein homeostasis upon interferon-induced oxidative stress, Cell 142 (4) (2010) 613–624.
[45] R.N. Harty, P.M. Pitha, A. Okumura, Antiviral activity of innate immune protein ISG15, J. Innate Immun. 1 (5) (2009) 397–404.
[46] K. Al-khatib, B.R. Williams, R.H. Silverman, W. Halford, D.J. Carr, The murine double-stranded RNA-dependent protein kinase PKR and the murine 2′,5′-oligoadenylate synthetase-dependent RNase L are required for IFN-beta-mediated resistance against herpes simplex virus type 1 in primary trigeminal ganglion culture, Virology 313 (1) (2003) 126–135.
[47] J. Dutrieux, G. Maarifi, D.M. Portilho, N.J. Arhel, M.K. Chelbi-Alix, S. Nisole, PML/ TRIM19-dependent inhibition of retroviral reverse-transcription by Daxx, PLoS Pathog. 11 (11) (2015) e1005280.
[48] M. Glass, R.D. Everett, Components of promyelocytic leukemia nuclear bodies (ND10) act cooperatively to repress herpesvirus infection, J. Virol. 87 (4) (2013) 2174–2185.
[49] K.E. Johnson, V. Bottero, S. Flaherty, S. Dutta, V.V. Singh, B. Chandran, IFI16 restricts HSV-1 replication by accumulating on the hsv-1 genome, repressing HSV-1 gene expression, and directly or indirectly modulating histone modifications, PLoS Pathog. 10 (11) (2014) e1004503.
[50] E.T. Kim, T.E. White, A. Brandariz-Nunez, F. Diaz-Griffero, M.D. Weitzman, SAMHD1 restricts herpes simplex virus 1 in macrophages by limiting DNA replication, J. Virol. 87 (23) (2013) 12949–12956.
[51] M.H. Orzalli, S.E. Conwell, C. Berrios, J.A. DeCaprio, D.M. Knipe, Nuclear interferon-inducible protein 16 promotes silencing of herpesviral and transfected DNA, Proc. Natl. Acad. Sci. USA 110 (47) (2013) E4492–E4501.
[52] M.K. Chelbi-Alix, H. de The, Herpes virus induced proteasome-dependent degradation of the nuclear bodies-associated PML and Sp100 proteins, Oncogene 18 (4) (1999) 935–941.
[53] S.R. Paludan, A.G. Bowie, Immune sensing of DNA, Immunity 38 (5) (2013) 870–880.
[54] P. Vandevenne, C. Sadzot-Delvaux, J. Piette, Innate immune response and viral interference strategies developed by human herpesviruses, Biochem. Pharmacol. 80 (12) (2010) 1955–1972.
[55] A. Okumura, P.M. Pitha, R.N. Harty, ISG15 inhibits Ebola VP40 VLP budding in an L-domain-dependent manner by blocking Nedd4 ligase activity, Proc. Natl. Acad.Sci. USA 105 (10) (2008) 3974–3979.
[56] J.B. Fan, K. Arimoto, K. Motamedchaboki, M. Yan, D.A. Wolf, D.E. Zhang, Identification and characterization of a novel ISG15-ubiquitin mixed chain and its role in regulating protein homeostasis, Sci. Rep. 5 (2015) 12704.
[57] N.V. Giannakopoulos, J.K. Luo, V. Papov, W. Zou, D.J. Lenschow, B.S. Jacobs, E.C. Borden, J. Li, H.W. Virgin, D.E. Zhang, Proteomic identification of proteins conjugated to ISG15 in mouse and human cells, Biochem. Biophys. Res. Commun. 336 (2) (2005) 496–506.
[58] C. Zhao, C. Denison, J.M. Huibregtse, S. Gygi, R.M. Krug, Human ISG15 conjugation targets both IFN-induced and constitutively expressed proteins functioning in diverse cellular pathways, Proc. Natl. Acad. Sci. USA 102 (29) (2005) 10200–10205. [59] N. Kolli, J. Mikolajczyk, M. Drag, D. Mukhopadhyay, N. Moffatt, M. Dasso, G. Salvesen, K.D. Wilkinson, Distribution and paralogue specificity of mammalian deSUMOylating enzymes, Biochem. J. 430 (2) (2010) 335–344.
[60] H. Saitoh, J. Hinchey, Functional heterogeneity of small ubiquitin-related protein modifiers SUMO-1 versus SUMO-2/3, J. Biol. Chem. 275 (9) (2000) 6252–6258.