HTLV-1 Tax impairs K63-linked ubiquitination of STING to evade host innate immunity

HTLV-1 Tax impairs K63-linked ubiquitination of STING to evade host innate immunity

Accepted Manuscript Title: HTLV-1 Tax impairs K63-linked ubiquitination of STING to evade host innate immunity Authors: Jie Wang, Shuai Yang, Lu Liu, ...

1MB Sizes 0 Downloads 59 Views

Accepted Manuscript Title: HTLV-1 Tax impairs K63-linked ubiquitination of STING to evade host innate immunity Authors: Jie Wang, Shuai Yang, Lu Liu, Hui Wang, Bo Yang PII: DOI: Reference:

S0168-1702(16)30512-3 http://dx.doi.org/doi:10.1016/j.virusres.2017.01.016 VIRUS 97059

To appear in:

Virus Research

Received date: Revised date: Accepted date:

18-8-2016 14-1-2017 15-1-2017

Please cite this article as: Wang, Jie, Yang, Shuai, Liu, Lu, Wang, Hui, Yang, Bo, HTLV1 Tax impairs K63-linked ubiquitination of STING to evade host innate immunity.Virus Research http://dx.doi.org/10.1016/j.virusres.2017.01.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

HTLV-1 Tax impairs K63-linked ubiquitination of STING to evade host innate immunity

Jie Wang*1, Shuai Yang*, Lu Liu*, Hui Wang*# and Bo Yang*#1

*Henan Collaborative Innovation Center of Molecular Diagnosis and Laboratory Medicine, School of Laboratory Medicine, Xinxiang Medical University, Xinxiang 453003, Henan Province, China

Address correspondence to: Dr Bo Yang, School of Laboratory Medicine, Xinxiang Medical University, Xinxiang, Henan Province, China. Email: [email protected] Or to Dr. Hui Wang, School of Laboratory Medicine, Xinxiang Medical University, Xinxiang, Henan Province, China. Email:[email protected].

1

These authors contributed equally to the work.

Highlights 

HTLV-1 protein expression was elevated in STING-knockdown macrophages.



Tax inhibited STING mediated production of type I IFN.



Tax interacted with STING and reduced the K63-linked ubiquitination of STING.

1

Abstract: The cellular antiviral innate immune system is essential for host defense and viruses have evolved a variety of strategies to evade the innate immunity. Human T lymphotropic virus type 1 (HTLV-1) belongs to the deltaretrovirus family and it can establish persistent infection in human beings for many years. However, how this virus evades the host innate immune responses remains unclear. Here we report a new strategy used by HTLV-1 to block innate immune responses. We observed that stimulator of interferon genes (STING) limited HTLV-1 protein expression and was critical to HTLV-1 reverse transcription intermediate (RTI) ssDNA90 triggered interferon (IFN)- production in phorbol12-myristate13-acetate (PMA)-differentiated THP1 (PMA-THP1) cells. The HTLV-1 protein Tax inhibited STING overexpression induced transcriptional activation of IFN-Tax also impaired poly(dA:dT), interferon stimulatory DNA (ISD) or cyclic GMP-AMP (cGAMP) -stimulated IFN-production, which was dependent on STING activation. Coimmunoprecipitation assays and confocal microscopy indicated that Tax was associated with STING in the same complex. Mechanistic studies suggested that Tax decreased the K63-linked ubiquitination of STING and disrupted the interactions between STING and TANK-binding kinase 1 (TBK1). These findings may shed more light on the molecular mechanisms underlying HTLV-1 infection.

Abbreviations

2

PAMPs, pathogen-associated molecular patterns; PRRs, pattern-recognition receptors; IFN-I, type I interferon; STING, stimulator of IFN genes; DAI, DNA-dependent activator of IFN regulatory factors; RNA-Pol III, RNA polymerase III; IFI16, IFN-gamma inducible factor 16; DDX41, DExD/ H-box helicase 41; cGAS, cyclic GMP-AMP synthase; cGAMP, cyclic GMP-AMP; ER, endoplasmic reticulum; TBK1, TANK-binding kinase 1; IKKi, inducible IkB kinase; IRF3/7, interferon regulatory factor 3/7; TRIM32/56, tripartite motif protein32/56; CREB, cyclic AMP responsive binding protein; NF-κB, nuclear factor kappa-B; TRIF, TIR domain-containing adaptor-inducing IFN-; RIP1, receptor-interacting protein kinase 1; RTI, reverse transcription intermediate; HTLV-1, Human T lymphotropic virus type 1; ATL, adult T-cell leukemia/lymphoma; HAM/TSP, HTLV-1-associated myelopathy/tropical spastic paraparesis; PMA-THP1, phorbol12-myristate13-acetate (PMA)-differentiated THP1 cells; AZT, azidothymidine; CHX, cycloheximide; ISD, interferon stimulatory DNA.

Key words: HTLV-1, Tax, STING, ubiquitination, innate immune responses, signal transduction

3

1. Introduction: The innate immune system functions as the first line of host defense against viral invasion. The initiation of innate immune responses relies on the recognition of pathogen-associated molecular patterns (PAMPs) by an array of pattern-recognition receptors (PRRs)(Broz and Monack, 2013). Different forms of DNA derived from viruses can act as PAMPs to induce activation of signaling pathways, leading to the production of type I interferon (IFN-I) and other antiviral innate immune responses(Keating et al., 2011). So far, several molecules have been identified as sensors for viral DNA, including DNA-dependent activator of IFN regulatory factors (DAI), RNA polymerase III (RNA-Pol III), IFN-gamma inducible factor 16 (IFI16), DExD/ H-box helicase 41 (DDX41), Ku70, stimulator of IFN genes (STING) and cyclic GMP-AMP synthase (cGAS)(Holm et al., 2013; Orzalli and Knipe, 2014; Paludan and Bowie, 2013). Among these DNA sensors, STING, also known as MITA/MPYS/ERIS, has emerged as central mediator in the cytosolic DNA-induced signaling pathways, either being an adaptor or directly sensing cytosolic dinucleotides(Abe et al., 2013; Ouyang et al., 2012; Zhang et al., 2013). Other DNA sensors, such as IFI16 and DDX41, interact with STING and trigger the transcription of IFN-I(Burdette and Vance, 2013). Another DNA sensor, cGAS, produces the second messenger cyclic GMP-AMP (cGAMP) after binding to cytosolic dsDNA, which is subsequently recognized by STING via direct binding, leading to IFN-I induction(Cai et al., 2014). Following

4

activation, STING dimerizes and translocates from endoplasmic reticulum (ER), through the Golgi apparatus, and to the perinuclear microsome compartment where it engages in TANK-binding kinase 1 (TBK1) binding. The STING-TBK1 complex is required for TBK1 activation, which subsequently recruits and activates the transcriptional factor interferon regulatory factor (IRF) 3 (Barber, 2011; Tanaka and Chen, 2012). During the process, tripartite motif protein (TRIM) 32 and TRIM56-mediated K63-linked polyubiquitination of STING plays a critical role, which is a prerequisite for TBK1 recruitment and STING-triggered IFN-I induction (Tsuchida et al., 2010; Zhang et al., 2012). Retroviruses can trigger innate immune responses through DNA sensors (van Montfoort et al., 2014). There is an essential reverse transcription (RT) step in the life cycle of HIV-1 and other lentiviruses, leading to the production of a cDNA strand. The cDNA strand is used as a template to generate a proviral genome, which is then integrated into the genome of host cells. During this process, viral-derived DNA fragments accumulate in the cytosol, which can be recognized by the DNA sensors IFI16 and cGAS. IFI16 and cGAS interact with reverse transcription intermediates (RTIs) and recruit STING for downstream signal transduction (Gao et al., 2013; Jakobsen et al., 2013). Human T lymphotropic virus type 1 (HTLV-1) belongs to the deltaretrovirus family, which has been linked to multiple diseases, including the aggressive blood cancer, adult T-cell leukemia/lymphoma (ATL)(Ishitsuka and Tamura, 2014), and the chronic,

progressive

neurological

and

5

inflammatory

disease

termed

HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP)(Fuzii et al., 2014; Ishitsuka and Tamura, 2014; Yamano and Sato, 2012). The total number of global HTLV-1 infected people is very uncertain, because the widely used number of 10 to 20 million infected people is estimated from studies about 20 years before, with incomplete epidemiological studies in many endemic areas(Gessain and Cassar, 2012). It is not well understood how HTLV-1 causes diverse clinical diseases and why these diseases are caused many years after initial infection. One explanation is that the virus has evolved effective mechanisms to evade host antiviral immune responses. So, it is quite meaningful to explore the exact mechanisms and this may provide clues for treatment of HTLV-1 associated diseases. Although the exact cytosolic sensors for HTLV-1 remain unknown, it has been reported that RTI plays a role in triggering antiviral responses in HTLV-1 infected monocytes. A 90-nucleotide RTI from the U5 region of HTLV-1(ssDNA90), which was

transfected

into

monocytes,

stimulated

antiviral

responses

in

a

STING-IRF3-dependent manner, suggesting the important role of STING in HTLV-1 induced antiviral innate immunity(Sze et al., 2013). HTLV-1 encodes a critical transactivator, Tax, which plays a key role in promoting viral spread and inducing T cell transformation through multiple mechanisms, including activating some effectors, such as the cyclic AMP responsive binding protein (CREB) and CBP/p300, nuclear factor kappa-B (NF-κB) and so on(Currer et al., 2012; Matsuoka and Jeang, 2007). Given the fact that IFN-I restricts HTLV-1 replication(Cachat et al., 2013; Kinpara et al., 2009) and HTLV-1 has

6

evolved some methods to evade this restriction(Colisson et al., 2010; Feng and Ratner, 2008; Journo and Mahieux, 2011), it is reasonable to explore the role of Tax in IFN-I production and innate immune responses. However, the role of Tax on IFN-I production is controversial now. One group reported that Tax suppressed viral RNA triggered

innate

immune

signaling

pathways

by

interacting

with

TIR

domain-containing adaptor-inducing IFN- (TRIF) and receptor-interacting protein kinase (RIP) 1 to disrupt IRF7 activation(Hyun et al., 2015), whereas another article suggested that Tax could be recruited into the TBK1/IKKi complex as a scaffolding-adaptor protein to enhance IFN-I gene expression(Diani et al., 2015). In this study, we demonstrated that STING inhibited HTLV-1 protein expression in HTLV-1 infected PMA-THP1 cells and HTLV-1 protein Tax inhibited viral DNA triggered IFN-I production targeting STING. Tax interacted with STING and decreased its K63-linked ubiquitination, leading to reduced STING-TBK1 association and decreased IFN-I production. This work may uncover one of the mechanisms utilized by HTLV-1 to escape from innate immunity.

7

2. Materials and Methods 2.1 cDNA constructs and reagents Human cGAS and TBK1 were amplified by PCR using cDNA from HSV-1 infected HEK293 cells and were subsequently cloned into a pcDNA3.1-Flag/HA vector (Invitrogen). HTLV-1 protein Tax was amplified by PCR using cDNA from MT2 cells (HTLV-1 transformed T cell line). HA-Ubi, pNF-B-Luc, pISRE-Luc, pIFN--Luc, Flag-STING and its deletion mutants were obtained as described previously (Wang et al., 2013; Wang et al., 2015). The anti-HA antibody was obtained from Covance (HA.11; 16B2; CO-MMS-101R), and the anti-Flag (M2; F3165) antibody was purchased from Sigma-Aldrich. The anti-STING antibody (19851-1-AP) and -actin (60008-1-Ig) antibody were purchased from Proteintech. The anti-IRF3 antibody (sc-9082), anti-Tax antibody (sc-57872) and anti-Ubi (sc-8017) were obtained from Santa Cruz Biotechnology. The antibody specific for IRF3 phosphorylation at residue Ser396 (4947), the anti-p65 antibody (4764), and the phospho-p65 (Ser536) antibody (3033) were purchased from Cell Signaling Technology. The anti-HTLV-1-p19 (ab9080) antibody was obtained from Abcam. Poly(dA:dT) (tlrl-patn) was obtained from InvivoGen. PMA (S1819) was obtained from Beyotime. The 90-base-long HTLV-1 ssDNA90 is the reverse complement of the 5’UTR region (315–404) of complete HTLV-1 genome (NCBI) and was synthesized

from

the

Sangon

Biotech.

The

sequence

was

as

follows:

5'-CTGTGTACTAAATTTCTCTCCTGGAGAGTGCTATAGAATGGGCTGTCGC

8

TGGCTCCGAGCCAGCAGAGTTGCCGGTACTTGGCCGTGGGC-3'.

The

scrambled ssDNA90 as a control was also synthesized from the Sangon Biotech. The sequence

was

as

follows:

5'-ATTCAGCTCACGGCGTCGAGTGCTGCTCGATGGCTCCTTAGTCCTGCTA AGTCGAGGTGGCTAATCCGGTAGTCGGTCGGATGGAATTCG-3'.

2.2 Cell culture, transfection and stimulation HEK293T and HEK293 cells were cultured in DMEM. MT2, MT4 and THP1 cells were grown in RPMI 1640. All cells were supplemented with 10% FBS (Invitrogen), 4mM/L-glutamine, 100U/ml penicillin, and 100U/ml streptomycin under humidified conditions with 5% CO2 at 37°C. Transfection of HEK293T cells, HEK293 cells and THP1 cells was performed with Lipofectamine 2000 (Invitrogen). For stimulation, poly(dA:dT) (1g/ml) and ssDNA90 (0.5g/ml) were delivered into cells using Lipofectamine 2000.

2.3 Immunoprecipitation and immunoblot analysis Immunoprecipitation and immunoblot analysis were performed as described previously (Wang et al., 2013). In short, either HEK293T or HEK293 cells were transfected with various combinations of plasmids. At 24h after the transfection, the cell lysates were prepared in lysis buffer containing 1.0% (vol/vol) Nonidet P40, 20 mM Tris-HCl, pH 8.0, 10%(vol/vol) glycerol, 150 mM NaCl, 0.2 mM Na3VO4, 1mM NaF, 0.1 mM sodium pyrophosphate and a protease inhibitor ‘cocktail’ (Roche). After

9

centrifugation for 20 min at 14,000g, supernatants were collected and incubated with the indicated antibody together with protein A/G Plus-agarose immunoprecipitation reagent (Santa Cruz Biotechnology) at 4°C for 3 h or overnight. After three washes, the immunoprecipitates were boiled in SDS sample buffer for 10 min and analyzed by immunoblot. For endogenous coimmunoprecipitation experiments, MT2 cells were either co-cultured with PMA-THP1 cells for 24h or left untreated. Then the cells were lysed and incubated with 0.5 l anti-STING antibody. The subsequent procedures were performed as described above.

2.4 Ex vivo infection of cells with HTLV-1 THP1 cells were pretreated with 100ng/ml PMA (Beyotime) for 24h, and then co-cultured with MT2, MT4 or control Jurkat cells for 24h at a ratio of 1:1 in RPMI 1640 with 10% FBS, 4 mM/L glutamine and antibiotics. The cells were washed with 1 mL PBS three times to remove the MT2, MT4 or Jurkat cells. Shake the cells gently when adding the PBS buffer. The adherent PMA-THP1 cells after three washes were then lysed for Western blot or Real-time PCR analysis.

2.5 Luciferase assays HEK293T or HEK293 cells were transfected with 50ng IFN-, ISRE or NF-B reporter plasmids and the indicated expression plasmids, together with 50ng Renilla luciferase plasmids as an internal control. The total DNA concentration was kept constant by supplementing with the empty vector pcDNA3.0. At 24h after the

10

transfection, the cells were either lysed in passive lysis buffer or transfected with poly(dA:dT) for an additional 24h and then lysed. The luciferase activity of the lysates was analyzed using the Dual-Luciferase Reporter Assay system (Promega).

2.6 Real-time PCR Total RNA was extracted from the cultured cells with TRIzol reagent as described by the manufacturer (Invitrogen). All gene transcripts were quantified by Real-time PCR with SYBR Green qPCR Master Mix using a 7500 Fast Real-time PCR system (Applied Biosystems). The relative fold induction was calculated using the 2-△△Ct method. The primers used for Real-time PCR were listed as follows: IFN-, forward,

5’-CAGCAATTTTCAGTGTCAGAAGCT-3’,

5’-TCATCCTGTCCTTGAGGCAGTAT-3’;

Tax,

reverse, forward, reverse,

5’-ATACCCAGTCTACGTGTTTGGAG-3’, 5’-CCGATAACGCGTCCATCGATG-3’;

ISG56,

forward, reverse,

5’-GCCATTTTCTTTGCTTCCCCTA-3’, 5’-TGCCCTTTTGTAGCCTCCTTG-3’;

and

5’-TCAACGACCACTTTGTCAAGCTCA-3’,

GAPDH,

forward, reverse,

5’-GCTGGTGGTCCAGGTCTTACT-3’.

2.7 RNA interference STING-Stealth-RNA interference was designed by the Invitrogen BLOCKiT RNAi Designer. The small interfering RNA (siRNA) sequences used were listed as

11

follows: ST1, forward, 5’-GAACCUGCAGAUGACAGCAGCUUCU-3’, reverse, 5’-AGAAGCUGCUGUCAUCUGCAGGUUC-3’;

ST2,

5’-GGCCCGGAUUCGAACUUACAAUCAG-3’,

forward, reverse,

5’-CUGAUUGUAAGUUCGAAUCCGGGCC-3’. The negative control siRNA was purchased from Invitrogen (catalog no.12935300). PMA-THP1 cells were transfected with siRNA using Lipofectamine 2000. At 24h after transfection, the cells were used for further experiments.

2.8 Confocal microscopy HEK293 cells were transfected with expressing plasmids for Cherry-Tax (Red) and YFP-STING (Green). At 24h after transfection, cells were fixed with 4% PFA in PBS and permeabilized with Triton X-100 and then blocked with 1% BSA in PBS. Nuclei were stained with 4, 6-diamidino-2-phenylindole (DAPI).

2.9 Statistics The data were presented as the means±SD from at least three independent experiments. The statistical comparisons between the different treatments were performed using the unpaired Student t test, and p<0.05 was considered statistically significant.

12

3. Results:

3.1 HTLV-1 protein expression is elevated in STING-knockdown macrophages

It has been clarified that HTLV-1 RTIs interact with STING to trigger formation of an IRF3-Bax complex, leading to apoptosis in monocytes during abortive HTLV-1 infection (Sze et al., 2013). However, in macrophages and monocyte-derived DC, HTLV-1 infection is not blocked by SAMHD1 and induces Tax production (Abe et al., 2013; Jones et al., 2008; Laguette et al., 2011). Given that STING is critical in the cytosolic DNA-induced signaling pathways, we wondered if STING affected HTLV-1 replication

in

macrophages.

We

used

phorbol12-myristate13-acetate

(PMA)-differentiated THP1 cells (a human macrophage-like cell line, PMA-THP1) to test this idea. We purchased two siRNA constructs and determined their effects on knockdown of STING. As shown in Fig. 1A and Fig. S1, both ST1 and ST2 could suppress the endogenous STING expression in PMA-THP1 cells. Then We co-cultured control and STING-knockdown PMA-THP1 cells with HTLV-1 transformed MT2 cells. Firstly, we examined the expressions of HTLV-1 protein Tax and p19 in PMA-THP1 cells by western blot. Compared to cells transfected with control siRNA, STING-silenced PMA-THP1 cells showed increased Tax and p19 expressions after co-culture with MT2 cells (Fig. 1A and Fig. S1). AZT, a nucleoside analog reverse transcriptase inhibitor, was used to suggest this co-culture with MT2

13

cells was a real infection and not passive transfer of viral proteins (Fig. S2). Similar results were observed in STING-silenced PMA-THP1 cells co-cultured with MT4 cells (another kind of HTLV-1 transformed cells) (Fig. 1B). Then we examined the effects of endogenous STING on the mRNA expression levels of Tax in PMA-THP1 cells after co-culture with MT2 cells. We observed that the Tax expression in mRNA levels was higher in STING-knockdown cells than in cells transfected with control siRNA (Fig. 1C). During viral infection, type I IFNs are important to protect cells from viral invasion by interfering with the replication of viruses. So we investigated the role of STING in HTLV-1 RTI stimulated IFN-I production and found that STING-knockdown impaired HTLV-1 RTI ssDNA90 induced IFN-production in mRNA levels (Fig. 1D). Scrambled ssDNA90 (whose sequence was generated by randomizing the U5 sequence) was used to confirm the specificity of HTLV-1 RTI to induce IFN- production. Collectively, these data suggest that STING can limit HTLV-1 protein expression and is critical to HTLV-1 RTI triggered IFN- production in macrophages.

3.2 Tax suppresses STING-dependent transcriptional activation of IFN-

Next, we wondered if Tax affected STING signaling pathways. Using a luciferase assay, we found that coexpression of Tax and STING in HEK293T cells inhibited STING induced IFN- reporter activity in a dose-dependent manner (Fig. 2A). Previous research has demonstrated that induction of IFN- requires cooperation

14

of two transcription factors, IRF3 and NF-B (Seth et al., 2006). Consistently, overexpression of Tax inhibited STING induced ISRE (an enhancer motif for IRF3) activation in a dose-dependent manner (Fig. 2B). In reporter assays, transfection of STING alone only minimally activated the IFN- promoter. Co-expression of STING and its upstream signal, cGAS, activated the IFN- promoter to a higher degree. So we further confirmed our data by co-transfection of Tax with cGAS and STING. Consistently, Tax transfection inhibited IFN- reporter activation induced by co-transfection of cGAS and STING (Fig. 2C). In addition, Real-time PCR assays suggested that Tax transfection impaired STING induced IFN- and ISG56 production in mRNA levels (Fig. 2D, E). Furthermore, western blot assays showed that the phosphorylation of IRF3 triggered by STING overexpression or co-transfection of cGAS and STING was decreased in the presence of Tax (Fig. 2F and Fig. S3). Taken together, Tax appears to negatively regulate STING mediated transcriptional activation of IFN-.

3.3 Tax suppresses poly(dA:dT) induced IFN-I production

STING is reported to be essential in IFN-I production in response to transfected poly(dA:dT). So we determined whether Tax affected poly(dA:dT)-stimulated IFN-I production. Reporter assays indicated that Tax transfection dramatically reduced poly(dA:dT)-stimulated activation of the ISRE and IFN- reporter in HEK293 cells (Fig. 3A, B). However, Tax had no significant effects on the activation of the NF-B

15

reporter induced by cytoplasmic poly(dA:dT) (Fig. 3C). In addition, Real-time PCR results indicated that poly(dA:dT)-stimulated IFN- levels were significantly impaired at the mRNA levels in the presence of Tax (Fig. 3D). Similar results were obtained using interferon stimulatory DNA (ISD) or cGAMP stimulation, suggesting that Tax affected STING-dependent signaling pathways (Fig. S4). Moreover, the phosphorylation of IRF3 triggered by the transfection of poly(dA:dT) was inhibited by Tax expression, whereas the phosphorylation of p65 showed no noticeable difference with or without Tax expression (Fig. 3E and Fig. S5). These results suggest that Tax suppresses poly(dA:dT) induced IFN-I production.

3.4 Tax interacts with STING

Since our results suggested that Tax impaired STING-mediated IFN-I production, we examined whether Tax interacted with STING. pCMV-Tax and Flag-STING were co-transfected in HEK293T cells, and coimmunoprecipitation assays revealed that Tax interacted with STING (Fig. 4A, B). Next, we addressed whether this interaction occurred

under

physiological

conditions

and

performed

endogenous

coimmunoprecipitation experiments. As shown in Fig. 4C, endogenous Tax interacted with STING in MT2 cells. Then we examined the situation in HTLV-1 infected cells and the results indicated that Tax was associated with endogenous STING in PMA-THP1 cells after co-culture with MT2 cells (Fig. 4D). Confocal microscopy indicated that transfected Tax colocalized with STING in HEK293 cells and further

16

confirmed the interaction between Tax and STING (Fig. 4E). We then determined which regions of STING were required for association with Tax. Various Flag-tagged STING mutants (Fig. 4F) were expressed in HEK293T cells with pCMV-Tax, and the coimmunoprecipitation experiments were performed. As shown in Fig. 4G, the C-terminus of STING (residues 136-338) was sufficient to interact with Tax. Taken together, these data suggest that Tax interacts with STING and that the C-terminus of STING is required for this interaction.

3.5 Tax reduces the K63-linked ubiquitination of STING

Since we found that Tax interacted with STING, we next sought to determine the mechanism by which Tax impaired the STING-mediated signaling pathways. Firstly, we checked the stability of STING in the presence of Tax. When the plasmid expressing STING was co-transfected with the empty vector or Tax, Western blot analysis indicated that Tax had no effects on STING stability (Fig. 5A). We confirmed this result using the cycloheximide (CHX) treatment (Fig. S6). It has been reported that K63-linked ubiquitination of STING is an important modification during STING activation, so we detected whether the inhibition of STING signaling pathways was due to the change of ubiquitination. Tax was co-transfected with STING and HA-tagged ubiquitin. The ubiquitination of STING in the cells was examined by Western blot. As expected, Tax expression decreased the polyubiquitination of STING (Fig. 5B). Then we explored the type of ubquitination. We used expression plasmids

17

for ubiquitin mutants retaining only a single lysine residue, K48 (ubiquitin-K48) or K63 (ubiquitin-K63). As shown in Fig. 5B, immunoprecipitation and immunoblot analysis indicated that Tax inhibited STING ubiquitination linked by K63 but not K48. Then we examined the effects of Tax on endogenous STING ubiquitination and found that poly(dA:dT) stimulated STING ubiquitination was decreased in the presence of Tax (Fig. 5C). Additionally, similar effects were observed for endogenous STING polyubiquitination induced by ISD or cGAMP transfection (Fig. S7). It has been reported that activated STING translocates from ER, through Golgi apparatus, and to the perinuclear microsomes (Paludan and Bowie, 2013). So we investigated the effects of Tax on STING translocation. However, Tax affected the location of STING in Golgi very slightly (Fig. S8). Then, we examined whether Tax impaired the interaction between STING and TBK1, which was a critical step for signal transduction. The plasmids expressing the Flag-tagged STING and HA-tagged TBK1 were co-transfected together with or without pCMV-Tax into HEK293T cells, and coimmunoprecipitation assays revealed that Tax impaired the STING-TBK1 interaction (Fig. 5D). Then we examined the effects of Tax on endogenous interaction between STING and TBK1. As shown in Fig 5E, HEK293 cells transfected with pCMV-Tax exhibited decreased interaction between endogenous STING and TBK1 after poly(dA:dT) stimulation. Taken together, these data suggest that Tax inhibits the K63-linked ubiquitination of STING, which is critical to its association with TBK1 and the subsequent activation.

18

4. Discussion: It has been controversial whether Tax impairs or enhances IFN-I production during HTLV-1 infection. A previous study by Jinhee Hyun et al. reported that Tax could potently inhibit innate immune signaling events triggered by dsRNA. Their research mainly focused on RIG-I and TRIF mediated IFN-I production, which demonstrated that Tax inhibited these pathways by binding to the RHIM domains of TRIF and RIP1 to disrupt IRF7 activity (Hyun et al., 2015). Recently, some studies indicated that as a kind of retrovirus, HTLV-1 might be recognized by DNA sensors, which might play an important role in HTLV-1 infection (Sze et al., 2013). In our study, we found that HTLV-1 could escape from the host innate immune response by blocking IFN- production mediated by STING, a critical molecule in the host cytosolic DNA-sensing pathways. Our results indicated that the HTLV-1 protein Tax interacted with STING and inhibited its K63-linked ubiquitination, which results in attenuation of STING function and inhibition of IRF3 activation, leading to decreased IFN- production and the associated antiviral response. Surprisingly, Erica Diani et al. proposed that Tax might be recruited into the TBK1/IKKi complexes as a scaffolding-adaptor protein to enhance IFN-I production (Diani et al., 2015). Erica Diani et al. used an in vitro reconstitution system, HEK293T cells expressing Tax and other plasmids, to show that Tax interacted with TBK1/IKKi and promoted TBK1/IKKi induced activation of IFN-In our system, we tried to explore the effects of Tax on endogenous STING-dependent signaling pathways induced by poly(dA:dT),

19

cGAMP or ISD stimulation. Furthermore, we examined the endogenous interaction of Tax and STING in MT2 cells and MT2-co-cultured PMA-THP1 cells. We also noted that Erica Diani et al. transfected only 2.5 to 20ng Tax plasmids into HEK293 cells and luciferase assays were performed. In our luciferase assay system, we did the dose-dependent experiments with 100 to 800ng Tax plasmids and we showed the increasing expression of Tax by western blot. A possible explanation for the conflicts is that the role of Tax in regulating host innate responses is very complicated, and needs further examination, especially in the in vivo environment during HTLV-1 infection. It is well accepted that STING is critically involved in IFN-I induction and host antiviral innate immune responses against multiple intracellular DNA, including the reverse transcription intermediates produced in the life cycle of HIV-1 and other lentiviruses. It has been reported that the HTLV-1 RTIs interact with STING to trigger formation of an IRF3-Bax complex, leading to apoptosis during abortive infection of monocytes (Sze et al., 2013). Our data indicated that in macrophages, STING also played an important role against HTLV-1 infection. STING-knockdown promoted the expression of HTLV-1 protein p19 and Tax. In addition, STING-knockdown inhibited HTLV-1 RTI ssDNA90 induced IFN-I production. So it was meaningful to explore whether HTLV-1 evaded innate immune responses targeting STING. Given the key role of STING in the host antiviral immune responses, many viruses have evolved the ability to target STING for inhibiting IFN-I production and subverting the host cellular defense system (Chan and Gack, 2016; Christensen and

20

Paludan, 2016). It has been reported that Dengue virus NS2B3 protease inhibits IFN- production by cleavage of STING (Aguirre et al., 2012). HBV, which is a DNA virus that causes chronic hepatitis, disrupts K63-linked ubiquitination of STING and inhibits STING-mediated signaling through the viral Pol protein(Liu et al., 2015). The HCV NS4B suppresses STING-mediated signaling by disrupting the interaction of STING and TBK1 (Ding et al., 2013). Our data showed that HTLV-1 Tax protein inhibited STING mediated ISRE and IFN- reporter activation, IFN- and ISG56 production, and IRF3 phosphorylation. Further more, Tax blocked poly(dA:dT) induced IFN-I production, which was well known to be a STING dependent signaling pathway. It has been reported that during STING activation, several cellular biological events, such as K63-linked ubiquitination and its recruitment of TBK1, are critical for STING-mediated signal transduction. Our data demonstrated that HTLV-1 protein Tax interacted with STING and markedly impaired its K63-linked ubiquitination. This interaction also attenuated the recruitment of TBK1 by STING, which was required for the downstream signal transduction of STING and was considered to be a process that was regulated by STING ubiquitination. However, Tax itself has not been reported as a deubiquitinase, then how could Tax impair the ubiquitination of STING? Our data only described that Tax was associated with STING in a complex and could not determine whether Tax interacted with STING directly or indirectly. Maybe other proteins were involved in the same complex. It has been reported that two ubiquitin ligases TRIM32 and TRIM56 mediate K63-linked ubiquitination of STING and potentiate IFN- induction

21

(Tsuchida et al., 2010; Zhang et al., 2012). Interestingly, domain mapping experiments indicated that the N-terminal transmembrane domain-containing fragment of STING was important for its interaction with TRIM32 while the C-terminal region of STING interacted with TRIM56. Our research demonstrated that the N-terminal transmembrane domains of STING were not necessary for the binding of Tax. So it is possible that Tax attenuates the K63-linked ubiquitination of STING by competing with TRIM56 for the C-terminal region of STING and further research may be needed to clarify this question. While we were finishing up the manuscript, Chun-Kit Yuen et al. reported that Tax suppressed IFN-I production through interaction with and inhibition of TBK1 kinase and suggested that Tax blocked IFN-I production at the step of IRF3 phosphorylation (Yuen et al., 2016). However, our findings suggest that Tax inhibits IFN-I production targeting STING. So, here we reported a different mechanism used by HTLV-1 to evade innate immune responses. Taken together, our results provide a new mechanism by which HTLV-1 escape from the host antiviral innate immune responses. Our data suggest that the HTLV-1 protein Tax inhibits STING mediated IFN- induction by disrupting its K63-linked ubiquitination. We believe these findings may expand our knowledge on the molecular mechanisms by which HTLV-1 counteracts the host antiviral innate immunity and may provide new target for designing novel therapeutic interventions to treat HTLV-1 associated diseases.

22

5. Conclusions: Expression of HTLV-1 protein was elevated in STING-knockdown macrophages, suggesting STING had an essential role in limiting HTLV-1 infection. It was reasonable that HTLV-1 used Tax to evade host defense. We demonstrated that Tax interacted with STING and inhibited STING mediated IFN-I production. Tax decreased the K63-linked ubiquitination of STING and disrupted the interactions between STING and TBK1. Our research may present a new mechanism by which HTLV-1 controls interferon responses.

Acknowledgements This work was supported by grants from the National Natural Science Foundation of China (31400776, U1504811, 31600697), Key scientific research projects in Universities of Henan Province (15A310023, 16A31003), and Scientific Research of Xinxiang Medical University (2013QN116, 2014QN156). References: Abe, T., Harashima, A., Xia, T., Konno, H., Konno, K., Morales, A., Ahn, J., Gutman, D., Barber, G.N., 2013. STING recognition of cytoplasmic DNA instigates cellular defense. Molecular cell 50(1), 5-15. Aguirre, S., Maestre, A.M., Pagni, S., Patel, J.R., Savage, T., Gutman, D., Maringer, K., Bernal-Rubio, D., Shabman, R.S., Simon, V., Rodriguez-Madoz, J.R., Mulder, L.C., Barber, G.N., Fernandez-Sesma, A., 2012. DENV inhibits type I IFN production in infected cells by cleaving human STING. PLoS Pathog 8(10), e1002934. Barber, G.N., 2011. STING-dependent signaling. Nature immunology 12(10), 929-930. Broz, P., Monack, D.M., 2013. Newly described pattern recognition receptors team up against intracellular pathogens. Nature reviews 13(8), 551-565. Burdette, D.L., Vance, R.E., 2013. STING and the innate immune response to nucleic acids in the cytosol. Nature immunology 14(1), 19-26.

23

Cachat, A., Chevalier, S.A., Alais, S., Ko, N.L., Ratner, L., Journo, C., Dutartre, H., Mahieux, R., 2013. Alpha interferon restricts human T-lymphotropic virus type 1 and 2 de novo infection through PKR activation. J Virol 87(24), 13386-13396. Cai, X., Chiu, Y.H., Chen, Z.J., 2014. The cGAS-cGAMP-STING pathway of cytosolic DNA sensing and signaling. Molecular cell 54(2), 289-296. Chan, Y.K., Gack, M.U., 2016. Viral evasion of intracellular DNA and RNA sensing. Nature reviews 14(6), 360-373. Christensen, M.H., Paludan, S.R., 2016. Viral evasion of DNA-stimulated innate immune responses. Cell Mol Immunol. Colisson, R., Barblu, L., Gras, C., Raynaud, F., Hadj-Slimane, R., Pique, C., Hermine, O., Lepelletier, Y., Herbeuval, J.P., 2010. Free HTLV-1 induces TLR7-dependent innate immune response and TRAIL relocalization in killer plasmacytoid dendritic cells. Blood 115(11), 2177-2185. Currer, R., Van Duyne, R., Jaworski, E., Guendel, I., Sampey, G., Das, R., Narayanan, A., Kashanchi, F., 2012. HTLV tax: a fascinating multifunctional co-regulator of viral and cellular pathways. Front Microbiol 3, 406. Diani, E., Avesani, F., Bergamo, E., Cremonese, G., Bertazzoni, U., Romanelli, M.G., 2015. HTLV-1 Tax protein recruitment into IKKepsilon and TBK1 kinase complexes enhances IFN-I expression. Virology 476, 92-99. Ding, Q., Cao, X., Lu, J., Huang, B., Liu, Y.J., Kato, N., Shu, H.B., Zhong, J., 2013. Hepatitis C virus NS4B blocks the interaction of STING and TBK1 to evade host innate immunity. J Hepatol 59(1), 52-58. Feng, X., Ratner, L., 2008. Human T-cell leukemia virus type 1 blunts signaling by interferon alpha. Virology 374(1), 210-216. Fuzii, H.T., da Silva Dias, G.A., de Barros, R.J., Falcao, L.F., Quaresma, J.A., 2014. Immunopathogenesis

of

HTLV-1-assoaciated

myelopathy/tropical

spastic

paraparesis

(HAM/TSP). Life Sci 104(1-2), 9-14. Gao, D., Wu, J., Wu, Y.T., Du, F., Aroh, C., Yan, N., Sun, L., Chen, Z.J., 2013. Cyclic GMP-AMP synthase is an innate immune sensor of HIV and other retroviruses. Science 341(6148), 903-906. Gessain, A., Cassar, O., 2012. Epidemiological Aspects and World Distribution of HTLV-1 Infection. Front Microbiol 3, 388. Holm, C.K., Paludan, S.R., Fitzgerald, K.A., 2013. DNA recognition in immunity and disease. Current opinion in immunology 25(1), 13-18. Hyun, J., Ramos, J.C., Toomey, N., Balachandran, S., Lavorgna, A., Harhaj, E., Barber, G.N., 2015. Oncogenic human T-cell lymphotropic virus type 1 tax suppression of primary innate immune signaling pathways. J Virol 89(9), 4880-4893. Ishitsuka, K., Tamura, K., 2014. Human T-cell leukaemia virus type I and adult T-cell leukaemia-lymphoma. Lancet Oncol 15(11), e517-526. Jakobsen, M.R., Bak, R.O., Andersen, A., Berg, R.K., Jensen, S.B., Tengchuan, J., Laustsen, A., Hansen, K., Ostergaard, L., Fitzgerald, K.A., Xiao, T.S., Mikkelsen, J.G., Mogensen, T.H., Paludan, S.R., 2013. IFI16 senses DNA forms of the lentiviral replication cycle and controls HIV-1 replication. Proceedings of the National Academy of Sciences of the United States of America 110(48), E4571-4580. Jones, K.S., Petrow-Sadowski, C., Huang, Y.K., Bertolette, D.C., Ruscetti, F.W., 2008. Cell-free

24

HTLV-1 infects dendritic cells leading to transmission and transformation of CD4(+) T cells. Nat Med 14(4), 429-436. Journo, C., Mahieux, R., 2011. HTLV-1 and innate immunity. Viruses 3(8), 1374-1394. Keating, S.E., Baran, M., Bowie, A.G., 2011. Cytosolic DNA sensors regulating type I interferon induction. Trends Immunol 32(12), 574-581. Kinpara, S., Hasegawa, A., Utsunomiya, A., Nishitsuji, H., Furukawa, H., Masuda, T., Kannagi, M., 2009. Stromal cell-mediated suppression of human T-cell leukemia virus type 1 expression in vitro and in vivo by type I interferon. J Virol 83(10), 5101-5108. Laguette, N., Sobhian, B., Casartelli, N., Ringeard, M., Chable-Bessia, C., Segeral, E., Yatim, A., Emiliani, S., Schwartz, O., Benkirane, M., 2011. SAMHD1 is the dendritic- and myeloid-cell-specific HIV-1 restriction factor counteracted by Vpx. Nature 474(7353), 654-657. Liu, Y., Li, J., Chen, J., Li, Y., Wang, W., Du, X., Song, W., Zhang, W., Lin, L., Yuan, Z., 2015. Hepatitis B virus polymerase disrupts K63-linked ubiquitination of STING to block innate cytosolic DNA-sensing pathways. J Virol 89(4), 2287-2300. Matsuoka, M., Jeang, K.T., 2007. Human T-cell leukaemia virus type 1 (HTLV-1) infectivity and cellular transformation. Nat Rev Cancer 7(4), 270-280. Orzalli, M.H., Knipe, D.M., 2014. Cellular sensing of viral DNA and viral evasion mechanisms. Annu Rev Microbiol 68, 477-492. Ouyang, S., Song, X., Wang, Y., Ru, H., Shaw, N., Jiang, Y., Niu, F., Zhu, Y., Qiu, W., Parvatiyar, K., Li, Y., Zhang, R., Cheng, G., Liu, Z.J., 2012. Structural analysis of the STING adaptor protein reveals a hydrophobic dimer interface and mode of cyclic di-GMP binding. Immunity 36(6), 1073-1086. Paludan, S.R., Bowie, A.G., 2013. Immune sensing of DNA. Immunity 38(5), 870-880. Seth, R.B., Sun, L., Chen, Z.J., 2006. Antiviral innate immunity pathways. Cell Res 16(2), 141-147. Sze, A., Belgnaoui, S.M., Olagnier, D., Lin, R., Hiscott, J., van Grevenynghe, J., 2013. Host restriction factor SAMHD1 limits human T cell leukemia virus type 1 infection of monocytes via STING-mediated apoptosis. Cell Host Microbe 14(4), 422-434. Tanaka, Y., Chen, Z.J., 2012. STING specifies IRF3 phosphorylation by TBK1 in the cytosolic DNA signaling pathway. Sci Signal 5(214), ra20. Tsuchida, T., Zou, J., Saitoh, T., Kumar, H., Abe, T., Matsuura, Y., Kawai, T., Akira, S., 2010. The ubiquitin ligase TRIM56 regulates innate immune responses to intracellular double-stranded DNA. Immunity 33(5), 765-776. van Montfoort, N., Olagnier, D., Hiscott, J., 2014. Unmasking immune sensing of retroviruses: interplay between innate sensors and host effectors. Cytokine Growth Factor Rev 25(6), 657-668. Wang, J., Yang, B., Hu, Y., Zheng, Y., Zhou, H., Wang, Y., Ma, Y., Mao, K., Yang, L., Lin, G., Ji, Y., Wu, X., Sun, B., 2013. Negative regulation of Nmi on virus-triggered type I IFN production by targeting IRF7. J Immunol 191(6), 3393-3399. Wang, Y., Lian, Q., Yang, B., Yan, S., Zhou, H., He, L., Lin, G., Lian, Z., Jiang, Z., Sun, B., 2015. TRIM30alpha Is a Negative-Feedback Regulator of the Intracellular DNA and DNA Virus-Triggered Response by Targeting STING. PLoS Pathog 11(6), e1005012. Yamano, Y., Sato, T., 2012. Clinical pathophysiology of human T-lymphotropic virus-type 1-associated myelopathy/tropical spastic paraparesis. Front Microbiol 3, 389.

25

Yuen, C.K., Chan, C.P., Fung, S.Y., Wang, P.H., Wong, W.M., Tang, H.M., Yuen, K.S., Jin, D.Y., Kok, K.H., 2016. Suppression of Type I Interferon Production by Human T-Cell Leukemia Virus Type 1 Oncoprotein Tax through Inhibition of IRF3 Phosphorylation. J Virol 90(8), 3902-3912. Zhang, J., Hu, M.M., Wang, Y.Y., Shu, H.B., 2012. TRIM32 protein modulates type I interferon induction and cellular antiviral response by targeting MITA/STING protein for K63-linked ubiquitination. The Journal of biological chemistry 287(34), 28646-28655. Zhang, X., Shi, H., Wu, J., Sun, L., Chen, C., Chen, Z.J., 2013. Cyclic GMP-AMP containing mixed phosphodiester linkages is an endogenous high-affinity ligand for STING. Molecular cell 51(2), 226-235.

26

Figures and figure legends Figure 1 STING limits HTLV-1 replication.

(A) THP1 cells were pretreated with 100ng/ml PMA for 24h (PMA-THP1), and then transfected with control siRNA (SC) or STING-specific siRNA (ST1, ST2). At 24h after transfection, PMA-THP1 cells were co-cultured with MT2 cells for another 24h. Then the cells were washed with PBS three times to remove MT2 cells and lysed for immunoblot assays. -actin was used as a loading control. (B) PMA-THP1 cells were transfected with SC or ST2. At 24h after transfection, PMA-THP1 cells were co-cultured with MT2 or MT4 cells for another 24h. Then the cells were washed with PBS three times to remove MT2 or MT4 cells and lysed for immunoblot assays. -actin was used as a loading control. (C) After transfected with SC or ST2 for 24h, PMA-THP1 cells were co-cultured with MT2 cells for 24h. Then the cells were washed with PBS three times to remove MT2 cells, and the Real-time PCR analysis of Tax mRNA was performed. (D) PMA-THP1 cells were transfected with SC or ST2, and then stimulated with ssDNA90 or scrambled ssDNA90 for 8h. The Real-time 27

PCR analysis of IFN- mRNA was performed. scram=scrambled ssDNA90. The data were representative of three independent experiments and were presented as means± SD (n = 3). *p<0.05, **p<0.01.

Figure 2 Tax inhibits STING-induced signaling pathways.

(A)HEK293T cells were transfected with empty vector (-) or 200ng Flag-STING plus IFN- promoter, together with empty vector (-) or increased amounts (100, 200, 400, 800ng) of pCMV-Tax plasmids. Luciferase assays were performed 24h after transfection. Results were presented as fold induction relative to cells with no STING transfection. (B)HEK293T cells were transfected with empty vector (-) or 200ng Flag-STING plus ISRE promoter, together with empty vector (-) or increased amounts (100, 200, 400, 800ng) of pCMV-Tax plasmids. Luciferase assays were performed 24h after transfection. Results were presented as fold induction relative to cells with no STING transfection. (C)HEK293T cells were transfected with indicated plasmids,

28

including

200ng

pCMV-Tax,

200ng

Flag-STING,

200ng

Flag-cGAS

and

IFN-promoter. Luciferase assays were performed 24h after transfection. Results were presented as fold induction relative to cells with empty vector transfection. (D and E) Real-time PCR analysis of IFN-D) and ISG56 (E) mRNA in HEK293T cells transfected with indicated plasmids. (F) HEK293T cells were transfected with pCMV-Tax, Flag-STING, Flag-cGAS or empty vector. At 24h after transfection, cell lysates were immunoblotted with anti-IRF3, anti-p-IRF3, anti-Flag, anti--actin or anti-Tax as indicated. The data were representative of three independent experiments and were presented as means±SD (n = 3). *p<0.05, **p<0.01.

29

Figure 3 Tax inhibits poly(dA:dT)-triggered signaling pathways.

(A, B and C) HEK293 cells were transfected with 200ng pCMV-Tax or empty vector, together with an ISRE (A), IFN- (B) or NF-B (C) luciferase reporter. At 24h after transfection, the cells were stimulated with 1g/ml poly(dA:dT) for another 24h. The cell lysates were subjected to luciferase assays. Results were presented as fold induction relative to untreated cells. (D) HEK293 cells were transfected with pCMV-Tax or empty vector. At 24h after transfection, the cells were stimulated with 1g/ml poly(dA:dT) for 8h. The cell lysates were subjected to Real-time PCR 30

analysis for IFN-mRNA. The mRNA results were relative to those of the untreated cells. (E) HEK293 cells were transfected with 500ng pCMV-Tax or empty vector. At 24h after transfection, the cells were stimulated with 1g/ml poly(dA:dT) for 8h. The total and phosphorylated IRF3 and p65 were analyzed by immunoblot analysis. The data were representative of three independent experiments and were presented as means±SD (n = 3). *p<0.05, **p<0.01.

Figure 4 Tax interacts with STING

(A) HEK293T cells were transfected with Flag-tagged STING and pCMV-Tax or empty vector plasmids. The cell lysates were immunoprecititated (IP) with anti-Tax and immunoblotted (IB) with anti-Tax or anti-Flag as indicated. (B) HEK293T cells were transfected with pCMV-Tax and Flag-tagged STING or empty vector plasmids. 31

The cell lysates were immunoprecititated with anti-Flag and immunoblotted with anti-Tax or anti-Flag as indicated. (C) The MT2 cell lysates were immunoprecipitated with anti-Tax or control IgG, and immunoblotted with anti-Tax or anti-STING. (D) PMA-THP1 cells were co-cultured with MT2 cells for 24h. The cell lysates were then immunoprecipitated with anti-STING Ab or control IgG, and immunoblotted with anti-Tax or anti-STING as indicated. (E) HEK293 cells were transfected for 24h with Cherry-Tax (Red) and YFP-STING (Green). Then the cells were prepared for confocal microscopy. Nuclei were stained with DAPI. (F) A schematic representation of the STING and its mutants. (G) HEK293T cells were transfected with the indicated plasmids. 24h after transfection, the cells were subjected to immunoprecipitation and immunoblot analysis. The data were representative of three independent experiments.

Figure 5 Tax reduces the K63-linked ubiquitination of STING

(A)HEK293T cells were transfected with Flag-STING and increased amounts of 32

pCMV-Tax plasmids (0, 0.2, 0.4 and 0.6g). At 24h after transfection, cell lysates were immunoblotted with anti-Flag, anti-Tax or anti--action as indicated. (B) HEK293T cells were transfected with various combinations of plasmids, including Flag-STING, pCMV-Tax, HA-Ubi, HA-K48-Ubi, and HA-K63-Ubi. At 24h after transfection, cell lysates were immunoprecipitated with anti-Flag and immunoblotted with anti-HA, anti-Flag or anti-Tax as indicated. (C) HEK293 cells transfected with the indicated plasmids were stimulated with poly(dA:dT) (1g/ml) for 8h. Cell lysates were immunoprecipitated with anti-STING, and immunoblotted with anti-Ubi, anti-K63 or anti-K48 as indicated. (D) HEK293T cells were transfected with the indicated plasmids, including Flag-STING, pCMV-Tax and HA-TBK1. At 24h after transfection, cell lysates were immunoprecipitated with anti-Flag and immunoblotted with anti-HA, anti-Flag or anti-Tax as indicated. (E) HEK293 cells were transfected with pCMV-Tax or empty vector as indicated. At 24h after transfection, the cells were stimulated with 1g/ml poly(dA:dT) for 8h. Cell lysates were immunoprecipitated with anti-STING or control IgG and immunoblotted with anti-TBK1, anti-STING or anti-Tax as indicated. The data were representative of three independent experiments.

33

Supplemental figures and figure legends Fig. S1

Figure S1 Densitometric analysis to quantify ratio of Tax/p19/STING to-actin in Fig 1A. The data were representative of three independent experiments and were presented as means±SD (n = 3). *p<0.05.

Fig. S2

Figure S2 Effects of STING-siRNA on p19 expression in the presence of AZT. PMA-THP1 cells were transfected with control siRNA (SC) or STING-specific siRNA (ST2). 24h later, the cells were pretreated with or without 50M azidothymidine (AZT) for 30 min and then co-cultured with MT2 cells (with or

34

without 50M AZT as indicated). 24h later, the cells were washed with PBS three times to remove MT2 cells and lysed for immunoblot assays. -actin was used as a loading control. The data were representative of three independent experiments.

Fig. S3

Figure S3 Densitometric analysis to quantify ratio of phosphorylated IRF3 to-actin in Fig 2F. The data were representative of three independent experiments and were presented as means±SD (n = 3). *p<0.05.

Fig. S4

35

Figure S4 Tax inhibits ISD or cGAMP induced IFN- production. HEK293 cells were transfected with pCMV-Tax or empty vector. At 24h after transfection, the cells were stimulated with 1g/ml ISD or cGAMP for 8h. The cell lysates were subjected to Real-time PCR analysis for IFN-mRNA. The mRNA results were relative to those of the untreated cells. The data were representative of three independent experiments and were presented as means±SD (n = 3). *p<0.05, ***p<0.001.

Fig. S5

Figure S5 Densitometric analysis to quantify ratio of phosphorylated IRF3 to-actin in Fig 3E. The data were representative of three independent experiments and were presented as means±SD (n = 3). *p<0.05.

Fig. S6

36

Figure S6 Effects of Tax on STING stability in the presence of CHX. HEK293T cells were transfected with 0.4g Flag-STING and 0.4g pCMV-Tax (+) or empty vector (-). At 24h after transfection, the cells were treated with 50g/ml CHX for indicated times. The cell lysates were immunoblotted with anti-Flag, anti-Tax or anti--actin as indicated. Actin was used as a loading control. The data were representative of three independent experiments.

Fig. S7

37

Figure S7 Tax inhibits ISD or cGAMP induced STING polyubiquitination. HEK293 cells transfected with the indicated plasmids were stimulated with 1g/ml ISD or cGAMP for 8h. Cell lysates were immunoprecipitated with anti-STING, and immunoblotted with anti-Ubi, or anti-K63 as indicated. The data were representative of three independent experiments. Fig. S8

38

Figure S8 The effect of Tax on STING location. Confocal microscopy of HEK293 cells transfected for 24h with YFP-STING and Tax or Vector (Vec), and then mock treated or stimulated for 4h with 1 μg/ml poly(dA:dT) (indicated as dA:dT in the figure). Nuclei were stained with the DNA intercalating dye DAPI. Staining of calnexin or gm130 served as a marker of the ER or Golgi body, respectively. The data were representative of three independent experiments.

Supplementary materials and methods Reagents AZT (A2169) and CHX (R750107) were obtained from Sigma-Aldrich. ISD (tlrl-isdn) and 2`3`-cGAMP (tlrl-nacga23) were obtained from InvivoGen.

Confocal microscopy

39

HEK293 cells were transfected with expressing plasmids for Tax and YFP-STING. At 24h after transfection, cells were fixed with 4% PFA in PBS and permeabilized with Triton X-100 and then blocked with 1% BSA in PBS. Staining of calnexin or gm130 served as a marker of the ER or Golgi body, respectively. Nuclei were stained with 4, 6-diamidino-2-phenylindole (DAPI).

40