HIV-1 p17 matrix protein enhances type I interferon responses through the p17–OLA1–STING axis

Innate immunity plays an important role in the immune response and virus control (Zhang et al., 2020b). During HIV infection, the antiviral innate immune response is activated and triggers the type I interferon (IFN) response (Rasaiyaah et al., 2013). HIV infection is sensed in infected cells by cyclic GMP-AMP synthase (cGAS), which detects cytosolic DNA and initiates the catalytic synthesis of cyclic GMP-AMP (cGAMP) (Lahaye et al., 2013; Gao et al., 2013). As a second messenger, cGAMP binds and activates STING (also known as STING1), which recruits and activates TANK-binding kinase 1 (TBK1) and transcription factor IFN regulatory factor 3 (IRF3). The activated IRF3 translocates to nucleus and leads to the production of IFNs and other cytokines (Ishikawa et al., 2009).

Both HIV-1 and HIV-2 originated from simian immunodeficiency viruses (SIVs), and can cause acquired immunodeficiency syndrome (AIDS). HIV-1 has infected millions of people worldwide and shows a strong pathogenicity. In contrast, HIV-2 is generally considered a naturally attenuated form of HIV, which has been shown to induce higher neutralizing antibodies and to be better controlled in infected individuals (Duvall et al., 2008; Kong et al., 2012; Makvandi-Nejad and Rowland-Jones, 2015). In mature virions, HIV matrix protein (p17), which is myristoylated at the N-terminus, is a gag gene-encoded structural protein that locates at the inner leaflet of the viral membrane (Gottlinger et al., 1989). p17 participates in the viral replication and particle assembly, playing a key role throughout the virus life cycle (Facke et al., 1993).

Upon viral infection, p17 can promote T cell proliferation and increase release of proinflammatory cytokines. It is reported that HIV-1 p17 has a different effect from that of SIV p17 regarding signaling in B cells. HIV-1 p17 downregulates the phosphoinositide 3-kinase (PI3K)/Akt signaling pathway, whereas SIV p17 promotes the pathway in B cells (Caccuri et al., 2014). In addition, HIV-1 p17 binding to cysteine-X amino acid-cysteine receptor 1 (CXCR1) promotes human monocytes, sustaining inflammatory processes (Giagulli et al., 2012). HIV-1 p17 can increase proliferation of interleukin-2 (IL-2)-stimulated peripheral blood mononuclear cells (PBMCs) (De Francesco et al., 1998). Furthermore, p17 promotes various cells to release pro-inflammatory cytokines release and stimulates their proliferation to benefit HIV-1 replication (De Francesco et al., 1998; De Francesco et al., 2002).

Extracellular p17 has different biological functions that are directly or indirectly implicated in various diseases. p17 can cross the blood–brain barrier (BBB) and be detected in the central nervous system, leading to HIV-1-associated neurocognitive disorder development (Caccuri et al., 2022). Previous studies have shown that virion-free p17 can persist in the host for long times and is easily detected in different organs and tissues, even after successful antiretroviral therapy (Popovic et al., 2005; Mazzuca et al., 2017). Besides this, an interaction between HIV-1 p17 and OLA1 had already been discovered by 2012 (Jager et al., 2012), and HIV-1 p17 is implicated in T cell autophagy and proliferation by interacting with OLA1 under glucose starvation conditions (Lu et al., 2021).

OLA1 belongs to the Obg-like protein family and can hydrolyze both ATP and GTP (Koller-Eichhorn et al., 2007). OLA1 is composed of three domains, including the G domain, coiled-coil domain and ThrRS-GTPase-SpoT (TGS) domain (Koller-Eichhorn et al., 2007). The G domain bridges the coiled-coil domain and TGS domain, forming a positive charged cleft proposed to bind nucleic acids (Koller-Eichhorn et al., 2007). Furthermore, OLA1 can bind to heat shock 70 kDa protein 1A (HS71A, also known as HSPA1A) and regulate its stability, thus protecting cells from death (Mao et al., 2013). By promoting the generation of reactive oxygen species (ROS), OLA1 could negatively regulate the antioxidant response (Zhang et al., 2009). OLA1 is also closely associated with tumorigenesis. For instance, by interacting with breast cancer gene 1 (BRCA1), OLA1 is involved in centrosome regulation and tumorigenesis (Matsuzawa et al., 2014). In colon cancer cells, OLA1 knockdown inhibits cell proliferation and enhances the sensitivity of cell to doxorubicin-induced cell death (Sun et al., 2010). However, the role of OLA1 in pathogenic infection and the innate immune response is unclear.

In this study, we found that HIV-1 p17, but not HIV-2 p17 or SIV p17, promotes innate immune responses to cytosolic DNA, and OLA1 is essential for the function of HIV-1 p17 in this process. Upon cGAMP treatment, knockout of OLA1 promotes the translocation of STING, which enhances the production of IFNs. By interacting with the STING C-terminal tail (CTT), OLA1 inhibits the interaction between STING and TBK1, which impairs phosphorylation of STING and downstream signaling. Moreover, the ATP and GTP hydrolytic activities of OLA1 are promoted by HIV-1 p17, but not HIV-2 p17 or SIV p17. Our results show that HIV-1 p17 promotes cGAMP-induced STING signaling by inhibiting the interaction between OLA1 and STING, whereas HIV-2 p17 or SIV p17 does not have any effect on STING-mediated innate immunity. This finding has important implications for our understanding of the differences among these lentiviruses.

HIV infection can induce an innate immune response via cGAS/STING pathway, and several proteins of HIV, such as p6, Vpx and Vif, are known to regulate the cGAS/STING signaling pathway, but the function of p17 remains enigmatic (Qian et al., 2023; Su et al., 2019; Wang et al., 2022). At the same time, it has been reported that different p17 might be employed with distinct functions (Caccuri et al., 2014). To explore the role of p17 in innate immunity, cell lines stably expressing p17 were treated with cGAMP and total RNA was extracted to perform quantitative reverse transcription PCR (qRT-PCR) analysis. Compared with HIV-2 p17 and SIV p17, HIV-1 p17 significantly enhanced the RNA levels of IFNB1(hereafter denoted IFN-β), ISG56 (also known as IFIT1), IP10 (CXCL10), ISG54 (IFIT2) and ISG15 after treatment with cGAMP (Fig. 1A,C,E; Fig. S1A). The protein expression levels of cell lines stably expressing HIV-1 p17, HIV-2 p17 or SIV p17 THP-1 were confirmed by western blotting (Fig. 1B,D,F; Fig. S1B). Furthermore, there was no significant difference between control group and the p17s in the mRNA levels of IFN-β or IP10 upon treatment with poly(I:C), which could activate the TBK1–IRF3 pathway independently of STING (Fig. S1C). These results demonstrate that HIV-1 p17, but not HIV-2 p17 or SIV p17, upregulates innate immune responses to cytosolic DNA.

OLA1, also known as DOC45 or GTP-binding protein 9, is a member of Obg GTPase family and has both ATPase and GTPase activities (Koller-Eichhorn et al., 2007). It has been reported that HIV-1 p17 can interact with OLA1 and inhibit T cell autophagic processes (Jager et al., 2012; Lu et al., 2021). GFP-Trap co-immunoprecipitation experiments in HEK293T cells showed that all three p17s interact with OLA1 (Fig. 2A) and the same results were obtained by using microscale thermophoresis (MST) analysis (Fig. 2B). Because OLA1 can hydrolyze both ATP and GTP, ATPase and GTPase activity assays were performed to investigate the influence of different p17 on enzymatic activity. Compared with HIV-2 p17 and SIV p17, using either ATP or GTP as a substrate, only HIV-1 p17 could promote OLA1 hydrolysis activity (Fig. 2C). Despite our best efforts, we failed to obtain the crystal structure of the p17s–OLA1 complex. Thus, we used AlphaFold to predict the influence of p17 binding on the structure of OLA1 (Jumper et al., 2021). The results showed that in the predicted p17–OLA1 complex, the OLA1 G2 domain has close contact with HIV-1 p17, but not with HIV-2 p17 or SIV p17 (Fig. 2D). Despite HIV-1 p17 having no direct contact with ATP/GTP-binding pocket, the change on G2 domain of OLA1 might influence OLA1 enzymatic activity distantly (Fig. 2D). These data suggest that although three p17s all interact with OLA1, only HIV-1 p17 can facilitate OLA1 hydrolytic activities on ATP and GTP.

To investigate the function of OLA1 on innate immune responses to cytosolic DNA, we generated OLA1-knockout THP-1 cell lines, and control cell lines were transduced with scramble non-sense sgRNA (WT). After treatment with cGAMP, the total RNAs of OLA1-knockout cells and WT cells were extracted and detected by qRT-PCR. Compared with WT cells, OLA1-knockout cells showed a significant increase in the levels of IFN-β, ISG56, IP10, ISG54 and ISG15 (Fig. 3A; Fig. S2A). Western blotting showed that endogenous OLA1 was depleted efficiently (Fig. 3B), and the depletion did not affect cell proliferation (Fig. S2B). In the same way, after cGAMP treatment, in the THP-1 cell line stably overexpressing OLA1, the levels of IFN-β, ISG56, IP10, ISG54 and ISG15 were inhibited compared with the control group (Fig. 3C,D; Fig. S2C). In addition, OLA1 has no significant effect on the RNA levels of IFN-β or IP10 upon treatment with poly(I:C) (Fig. S2D,E). To eliminate the possibility of sgRNA mistargeting, OLA1 was reconstituted in the knockout THP-1 cell lines. After treatment with cGAMP, reconstituted OLA1 could rescue the OLA1-knockout phenotype. At the same time, we tested the effect of the ATP-binding ability on the function of OLA1, and the results showed that mutants with Phe¹²⁷ to alanine (F127A) or Asn²³⁰ to alanine (N230A) changes, both losing ATP-binding ability, could not rescue the OLA1-knockout phenotype (Fig. 3E,F; Fig. S2F). Next, we detected the influence of OLA1 on HIV-1 production and infection by inducing production of HIV-1 VSV-G- or Env-pseudovirus in OLA1-knockout HEK293T cells. The results showed that OLA1 knockout slightly decreased viral production (Fig. S3A–D). After infected with VSV-G-pseudovirus, knockout of OLA1 also mildly inhibited viral infection both in HEK293T cells and THP-1 cells (Fig. S3E,F). Collectively, these results indicate that OLA1 negatively regulates innate immune responses to cytosolic DNA.

After binding to cytosolic DNA, cGAS catalyzes the synthesis of cGAMP. When activated by cGAMP, the STING–TBK1–IRF3 signaling pathway is initiated and triggers the production of type I IFNs and many other cytokines. Thus, we suspect that OLA1 probably interacts with a member of this signaling complex. Indeed, the co-immunoprecipitation experiments showed that OLA1 interacts with STING, but not TBK1 or IRF3 (Fig. 4A,C,D). STING, located in endoplasmic reticulum (ER), consists of a transmembrane (TM) region (residues 1–134), a cyclic dinucleotide-binding domain (CBD, residues 135–342) and a CTT (residues 343–379). The CTT of STING is important for the downstream signaling pathway (Srikanth et al., 2019). GFP-Trap co-immunoprecipitation experiments were performed to investigate which domain mediates the interaction of STING with OLA1. Because STING resides on ER in resting cells and OLA1 is predominantly in the cytoplasm, we speculated that the interaction of the two proteins would not rely on the TM region of STING. GFP-Trap co-immunoprecipitation experiments confirmed that the CTT of STING was essential for OLA1 interaction (Fig. 4B). Moreover, to observe whether OLA1 and STING colocalize within cells, we established a HeLa and THP-1 cell line stably expressing STING–GFP. Using anti-OLA1 primary antibody and TRITC-conjugated secondary antibody to immunostain the endogenous OLA1 and then using confocal microscopy to observe, we showed that endogenous OLA1 colocalized with STING both in STING–GFP HeLa cells and THP-1 cells (Fig. 4E,F). Taken together, these data suggest that OLA1 interacts and colocalizes with STING.

STING, after activated by cGAMP, translocates from the ER to the Golgi and then relocates to perinuclear region of the cells (Ishikawa et al., 2009). To observe OLA1 localization in the process of STING translocation, anti-OLA1 primary antibody and TRITC-conjugated secondary antibody were used to immunostain endogenous OLA1 both in the STING-GFP HeLa and THP-1 cell line after treatment with cGAMP. The results showed that OLA1 colocalized with STING in small vesicles, but could not translocate with STING to perinuclear region, where large aggregates containing STING were accumulated (Fig. 5A,B). To gain insights into the influence of OLA1 on STING translocation, anti-STING primary antibody and TRITC-conjugated secondary antibody were used to immunostain endogenous STING both in WT and OLA1-knockout THP-1 cell lines after cGAMP treatment. Confocal microscopy results showed that the translocation of STING to the perinuclear region was increased in OLA1-knockout cells compared with WT cells (Fig. 5C,D). GFP-Trap co-immunoprecipitation experiments showed that cGAMP treatment did not influence the interaction between OLA1 and STING (Fig. 5E). Although the exact mechanism remains unclear, the phosphorylation of STING by TBK1 occurs during STING translocating to perinuclear region, which is essential for the downstream signaling activation (Tanaka and Chen, 2012). This prompted us to interrogate whether the interaction between STING and TBK1 was influenced by OLA1. To address this speculation, HeLa cells stably expressing STING–GFP were treated with cGAMP and GFP-Trap co-immunoprecipitation experiments were performed. As shown in Fig. 5F, the interaction between STING and TBK1 is inhibited by OLA1 upon cGAMP treatment. Consistent with this, phosphorylation of STING was promoted in OLA1-knockout THP-1 cells after cGAMP treatment (Fig. S4A,B). It has been demonstrated that STING is degraded after its activation, which is considered to be a feedback mechanism for termination of STING signaling. As the results shown in Fig. S4A,B, the degradation of STING was not affected by OLA1. Collectively, these data confirm that OLA1 translocates along with STING in small vesicles, which prevents the interaction between STING and TBK1, and inhibits STING phosphorylation.

As mentioned above, HIV-1 p17 promotes innate immune responses to cytosolic DNA and facilitates the hydrolytic activity of OLA1 (Figs 1 and 2). Hence, we wondered whether HIV-1 p17 exerts its immune-regulation function through OLA1. To verify this, WT and OLA1-knockout THP-1 cell lines that transiently expressed HIV-1 p17 were treated with cGAMP, then the total RNA was extracted and subjected to qRT-PCR analysis. WT cells expressing HIV-1 p17 had significantly increased mRNA levels of IFN-β, ISG56 and IP10 compared with levels in cells do not express HIV-1 p17. However, in OLA1-knockout cells, expressing HIV-1 p17 did not affect mRNA levels of IFN-β, ISG56 or IP10 (Fig. 6A,B), and the effect of overexpression of HIV-1 p17 was comparable to OLA1 knockout, as shown in Fig. 6A,B. Thus, HIV-1 p17 mainly promotes innate immune responses to cytosolic DNA by inhibiting OLA1 function. We have indicated that OLA1-knockout promotes STING translocation to perinuclear region. To gain insights into the influence of HIV-1 p17 on STING translocation, anti-STING antibody was used to immunostain endogenous STING both in control cells and THP-1 cells stably expressing HIV-1 p17 after cGAMP treatment. Confocal microscopy results showed that translocation of STING to the perinuclear region was increased in expressing HIV-1 p17 cells (Fig. 6C,D). Consistent with this, upon cGAMP treatment, the phosphorylation of STING was promoted by HIV-1 p17, but the degradation of STING was not affected (Fig. S4C,D). However, expressing HIV-1 p17 did not further affect the translocation of STING to the perinuclear region in OLA1-knockout cells (Fig. 6E,F). At the same time, GFP-Trap co-immunoprecipitation experiments suggest that HIV-1 p17 inhibits the interaction between OLA1 and STING (Fig. 6G). These results collectively indicate that HIV-1 p17 promotes cytosolic DNA-induced innate immune responses through inhibiting the interaction between OLA1 and STING.

Both HIV-1 and HIV-2 result in AIDS, but HIV-1 has a higher transmissibility and a stronger pathogenicity, which leads to millions of deaths worldwide (Soares et al., 2011; Drylewicz et al., 2008). The underlying mechanistic differences between the two lentiviruses remain incompletely understood. In this study, we compared p17 of HIV-1, HIV-2 and SIV, and found that only HIV-1 p17 upregulates innate immune responses to cytosolic DNA. HIV-1 p17 has been reported to enhance HIV replication (De Francesco et al., 1998), and our study shows that HIV-1 p17 promotes cellular innate immune responses. De Francesco et al. added recombinant p17 to cell medium to investigate the influence of p17 on cell proliferation and HIV replication. In this study, we generated cells line stably expressing p17 and tested the effect of p17 on the innate immune response upon cGAMP treatment. It is speculated that the different manipulation of p17, for example, extracellular supplementation versus intracellular expression, might lead to divergent influences on cellular activity and viral replication. At the same time, the hydrolysis activities of OLA1 on ATP and GTP are only promoted by HIV-1 p17, but not HIV-2 p17 or SIV p17. We also found that the ATP-binding ability of OLA1 is indispensable for its function in regulation of STING-mediated innate immune responses. AlphaFold analysis suggested that, although all three p17s interact with OLA1, the conformation of OLA1 at the G2 domain is only altered by HIV-1 p17. This might be the reason for the specific influence on OLA1 enzymatic activity by HIV-1 p17. OLA1 has functions in a variety of physiological processes (Sun et al., 2010; Mao et al., 2013; Zhang et al., 2009; Matsuzawa et al., 2014), but the specific role of OLA1 in innate immunity and pathogenic infection remains to be elucidated. In the current study, HIV-1 p17 promotes innate immune responses to cytosolic DNA via interacting with OLA1, and this interaction inhibits the ability of OLA1 to bind to STING. OLA1 inhibits STING trafficking by interacting with STING, but HIV-1 p17 could weaken the interaction between OLA1 and STING, thereby being involved in the regulation of innate immunity. We speculate that HIV-1 p17 binds to OLA1 and alters its conformation, resulting in the alteration of OLA1 function in innate immune responses to viral infection. Nevertheless, the precise role of the HIV-1 p17-OLA1 interaction needs to be further explored. Furthermore, whether the differences between the p17 interactions with OLA1 could contribute to distinct pathogenicity of the three lentiviruses also needs more investigation.

OLA1 has been reported to have various biological functions, including cell proliferation, intracellular homeostasis and autophagy (Ding et al., 2016; Mao et al., 2013; Lu et al., 2021). In the current study, OLA1 affects innate immune responses to cytosolic DNA through an OLA1–STING–TBK1 axis. OLA1 interacts with the CTT of STING, which is essential for TBK1 and IRF3 activation. Phosphorylation of STING is inhibited by OLA1 as a way of preventing the interaction between STING and TBK1, which causes the blockage of downstream signaling. In the immunofluorescence experiments, OLA1 was found to translocate with STING in early-stage small vesicles, but not later-stage perinuclear large aggregates. The specific mechanism that why OLA1 colocalizes with STING in small vesicles but is excluded from perinuclear large puncta needs further investigation.

The STING signaling pathway, which is implicated in the pathogenesis of colon cancer, cardiac hypertrophy, neuron degeneration and autoimmune diseases, has received a lot of attention in recent years (Lee et al., 2021; Oduro et al., 2022; Hinkle et al., 2022; Gao et al., 2015). After being activated, STING traffics from the ER to the perinuclear regions for recruiting and activating TBK1 and IRF3. Aberrant vesicular trafficking of STING leads to various diseases, and STING trafficking has become a new therapeutic target. It has been reported that there are many proteins involved in the regulation of STING translocation process. For example, ALG2 is involved in innate immune responses to cytosolic DNA by inhibiting STING trafficking (Ji et al., 2021). EGFR, STEEP and ACBD3 also function as regulators in the STING trafficking process (Wang et al., 2020; Zhang et al., 2020a; Motani et al., 2022). Furthermore, we indicate that OLA1 regulates STING-mediated innate immune responses independently of ACBD3 (Fig. S5). The specific mechanism of STING trafficking still needs to be elucidated. Here, HIV-1 p17 and OLA1 are identified as the proteins that influence STING translocation.

In conclusion, this study confirmed that HIV-1 p17, but not HIV-2 p17 or SIV p17, facilitates innate immune responses to cytosolic DNA. HIV-1 p17 interacts with OLA1 and inhibits the regulation of OLA1 on STING-mediated innate immunity. By interacting with STING, OLA1 suppresses translocation and phosphorylation of STING upon cGAMP treatment, resulting in suppression of the production of IFNs (Fig. 7). Furthermore, the ATP and GTP hydrolytic activities of OLA1 are enhanced by HIV-1 p17, but not HIV-2 p17 or SIV p17. This finding might provide a new therapeutic target for AIDS and STING-related immune diseases.

HIV-1 p17 gene was obtained from an HIV packaging plasmid NLENY1-ES-IRES (a gift from Dr David Levy, Department of Medicine, University of Alabama at Birmingham, USA). HIV-2 p17 and SIV p17 genes were synthesized by Beijing Aoke Biotechnology Co., Ltd (Beijing, China). For co-immunoprecipitation, HIV-1 p17, HIV-2 p17 and SIV p17 were cloned into pEGFP-N1 vector (Clontech). HIV-1 p17 was fused with a GST tag at the N-terminus and cloned into pQCXIP2.0 (a gift from Dr Wentao Qiao, Nankai University, Tianjin, China). For stable expression in mammalian cells, HIV-1 p17, HIV-2 p17 and SIV p17 fused with a 3HA tag at N-terminus for detection by western blotting, were cloned into pQCXIP2.0. For prokaryotic expression, HIV-1 p17, HIV-2 p17 and SIV p17 were cloned into pET30 (Novagen). OLA1 cDNA was a gift from Dr Jiahuai Han (Xiamen University, Xiamen, China). For co-immunoprecipitation, OLA1 was cloned into pEGFP-N1 vector, or fused with a Flag tag at the N-terminus and cloned into pQCXIP2.0. For prokaryotic expression, OLA1 was cloned into pET30. STING cDNA was a gift from Dr Liangguo Xu (Jiangxi Normal University, Nanchang, China). For co-immunoprecipitation, STING was cloned into pCMV-3HA (a gift from Dr Wentao Qiao). STING(135-379), STING(135-342) and STING(343-379), fused to a GST tag at the N-terminus, were cloned into pQCXIP2.0. For stable expression in mammalian cells, STING was fused with a GFP tag at the C-terminus and cloned into pQCXIP2.0. IRF3 cDNA was a gift from Dr Yushan Zhu (Nankai University, Tianjin, China). For co-immunoprecipitation, IRF3 was cloned into pCMV-3HA. TBK1 cDNA was a gift from Dr Yushan Zhu (Nankai University, Tianjin, China). For co-immunoprecipitation, TBK1 was cloned into pCMV-3HA. ACBD3 cDNA was obtained from our THP-1 cDNA library. For eukaryotic expression, ACBD3 was cloned into pCMV-3HA.

Single-guide RNAs (sgRNAs) were designed to target the human OLA1. The related sgRNA sequences were cloned into a lentiCRISPR vector. The sgRNA-resistant OLA1 was generated by overlap PCR. The pVSV-G plasmid was a gift from Dr David Levy (University of Alabama at Birmingham, Birmingham, Alabama, USA). The pCMV-promoted Gag-pol expression plasmid for MLV retrovirus packaging was a gift from Dr Wentao Qiao (Nankai University, Tianjin, China). The lentiCRISPR and psPAX2 vectors were gifts from Dr Lin Liu (Nankai University, Tianjin, China). Sequences of sgRNA targeting OLA1 were as following: sgRNA-1: 5′-TCTCTTACCTTGCTGGTTTG-3′, sgRNA-2: 5′-TCTCATATTAGTGCCTGTGA-3′.

6×His-tagged HIV-1 p17, HIV-2 p17, SIV p17 or OLA1 was expressed in Escherichia coli BL21 (DE3) strain (Novagen) and cultured in Luria-Bertani (LB) medium (BD Biosciences), with addition of kanamycin (50 μg ml⁻¹, Sangon Biotech). Cells were grown at 37°C until the optical density of the culture reached 0.8 at a wavelength of 600 nm (OD₆₀₀). Protein expression was induced with 200 μM isopropyl β-D-thiogalactopyranoside (IPTG) at 16°C for 12 h. The cells were harvested and resuspended in binding buffer (50 mM Tris-HCl, 500 mM NaCl, pH 8.0), then lysed by sonication and clarified by centrifugation (18,000 g, 40 min, 4°C). 6×His-tagged proteins were purified using Ni-NTA affinity resin (GE Healthcare). For MST, 6×His-tagged OLA1 (104 nM) was labeled as the target protein and mixed with serially diluted non-labeled 6×His-tagged HIV-1 p17 (643 μM), HIV-2 p17 (321 μM) or SIV p17 (172 μM) in a 1:1 ratio. Experiments were performed in standard treated capillaries (NanoTemper Technologies) using a Monolith™ NT.115 instrument (NanoTemper Technologies). For ATPase and GTPase activity assays, the purified proteins were replaced with binding buffer to avoid interference from imidazole. The ATPase and GTPase activities were assayed with a Ca²⁺ Mg²⁺ ATPase kit (Solarbio), and the experiments were performed according to the manufacturer's instructions.

HEK293T, HeLa, TZM-bl and THP-1 cells were obtained from the American Type Culture Collection (ATCC) and all cells were grown at 37°C and 5% CO₂. HEK293T, HeLa and TZM-bl cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Corning) containing 10% fetal bovine serum (FBS; ExCell Bio). THP-1 cells were cultured in Roswell Park Memorial Institute (RPMI)-1640 medium (Corning) containing 10% FBS. Cells were seeded, cultured for 18–24 h and reached a confluence of 60% before transfection with plasmids using polyethylenimine (PEI). THP-1 cells were transfected with poly(I:C) (Invivogen, tlrl-picw) by Lipofectamine 2000 (Thermo Fisher Scientific, 11668019).

The experiments were performed as previously described (Ji et al., 2023a). Briefly, to produce lentiviruses, HEK293T cells were seeded on 10 cm plates for 24 h and reached a confluence of 70–80% before transfection. 6 µg retroviral vector, 6 µg pCMV-MLV-gag-pol or psPAX2 and 3 µg pCMV-VSV-G were transfected together using PEI. Lentiviruses were harvested by centrifugation for 15 min at 1600 g after transfection for 48 h, then filtered through 0.22 µm filters and concentrated by ViraTrap™ Lentivirus Concentration Reagent (Biomiga). THP-1 cells were infected with lentivirus particles for 6 h, then cultured with fresh medium. For generation of OLA1-knockout THP-1 cell lines, sgRNAs targeting OLA1 were inserted into lentiCRISPR vector. Lentiviruses were produced in HEK293T cells as described above. THP-1 cells were infected by lentivirus for 48 h. Then infected cells were selected in 96-well plates with 2 µg ml⁻¹ puromycin for 2 weeks. Single colonies were grown and analyzed by western blotting. To generate cell lines that stably expressed HIV-1 p17, HIV-2 p17, SIV p17, STING–GFP or overexpressed OLA1, HeLa or THP-1 cells were infected with corresponding lentivirus for 48 h and then selected and analyzed as described above.

For producing pseudotyped HIV-1, pSRHS-Env or VSV-G and NLENY1-ES-IRES were transfected into OLA1-knockout HEK293T cell lines using PEI. THP-1, TZM-bl or HEK293T cells were infected with pseudotyped HIV-1 virus for 6 h, then cultured with fresh medium for 48 h.

To analyze protein levels, cells were rinsed with PBS and then scraped into a RIPA buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) for 30 min at 4°C. Then the cells lysates were boiled at 100°C for 10 min with SDS loading buffer and analyzed by SDS-PAGE. The proteins were transferred to 0.45 μm PVDF membranes for western blotting analysis. Primary antibodies used were: anti-GAPDH (Santa Cruz Biotechnology, sc-32233, 1:1000), anti-HA (Sigma, H3663, 1:2000), anti-GFP (Santa Cruz Biotechnology, sc-9996, 1:1000), anti-GST (Utibody, UM3005, 1:2000), anti-Flag (Sigma, F1804, 1:2000), anti-OLA1 (Proteintech, 16371-1-AP, 1:1000), anti-STING (Proteintech, 19851-1-AP, 1:1000), anti-p-STING (Cell Signaling Technology, 50907, 1:1000) and anti-HIV-1 p24 (Millipore, Mab8790, 1:1000). Secondary antibodies used were: goat anti-mouse IgG-HRP (Sungene Biotech) and goat anti-rabbit IgG-HRP (Sungene Biotech).

HEK293T or HeLa cells were transfected with indicated plasmids for 48 h, then cells were harvested and lysed by sonication in RIPA buffer. The cell lysates were clarified by centrifugation (12,000 g, 30 min, 4°C), then the supernatant was incubated with GFP-Trap Sepharose resin (Chromotek) for 4 h at 4°C. Immunopellets were washed five times. The precipitated proteins were treated with the SDS loading buffer at 100°C for 10 min and detected by western blotting as described above.

TZM-bl cells were infected by VSV-G-pseudotyped HIV-1 virus or Env-pseudotyped HIV-1 virus for 48 h. Cells were rinsed with PBS and then lysis buffer (Promega) was added to lyse cells. The cell lysates were centrifuged (12,000 g, 30 min, 4°C) and incubated with luciferin reagent (Promega) and then luciferase signals were measured following the manufacturer's protocols.

The experiments were performed as previously described (Ji et al., 2023b). Briefly, the cGAMP (InvivoGen) was delivered into HEK293T or HeLa cells by permeabilization with digitonin (10 μg ml⁻¹) for 30 min at 37°C in buffer A (50 mM Hepes, pH 7.2, 100 mM KCl, 3 mM MgCl₂, 0.1 mM DTT, 85 mM sucrose, 0.2% BSA, 1 mM ATP). cGAMP was delivered into THP-1 cells by permeabilization with perfringolysin O (PFO, 0.01 μg ml⁻¹) for 30 min at 37°C in buffer A.

HeLa STING–GFP cells (4×10⁵/well) were grown on glass coverslips in 12-well plates for 24 h before cGAMP treatment. Cells were fixed with 4% paraformaldehyde for 20 min followed by permeabilization with 0.2% Triton X-100 for 10 min. After washing with PBS, cells were blocked in 3% bovine serum albumin (BSA) for 2 h. Subsequently, cells were incubated with primary antibody (as described above) for 2 h followed by incubation with TRITC-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch Laboratories, 1:100) for 2 h. THP-1 cells were processed as previously described (Ji et al., 2023a). Briefly, THP-1 STING-GFP cells (2×10⁶/well) were added on glass coverslips after cGAMP treatment. Cells were fixed at 120 g for 20 min with 4% paraformaldehyde, and then permeabilized with 0.2% Triton X-100 at 120 g for 10 min. After washing with PBS, cells were blocked in 3% BSA for 2 h. Subsequently, cells were incubated with primary antibody for 2 h followed by incubation with TRITC-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch Laboratories, 1:100) for 2 h. Images were captured with Leica TCS SP5 confocal laser scanning microscope. ImageJ was used to quantify fluorescence intensity.

Cells were washed with PBS and lysed in TRIzol reagent (Invitrogen) for 15 min, then mixed with chloroform and shaken for 15 s. Subsequently, the samples were centrifuged (12,000 g, 15 min, 4°C), and the aqueous upper phase was transferred into a new tube, then isopropyl alcohol was added for total RNA precipitation. The RNA precipitate was dissolved in RNase-free ddH₂O. Reverse transcription was performed according to the manufacturer's instructions of the RT-PCR kit (Abm, Canada). Quantitative RT-PCR was completed using a two-step real-time PCR kit (SYBR Green, GenStar). The primers used (5′-3′) were: ACTIN (forward primer: ACGTTGCTATCCAGGCTGTG, reverse primer: GAGGGCATACCCCTCGTAGA), IFN-β (forward primer: ACGCCGCATTGACCATCTAT, reverse primer: GTCTCATTCCAGCCAGTGCT), ISG56 (forward primer: CCTCCTTGGGTTCGTCTACA, reverse primer: AGTGGCTGATATCTGGGTGC), IP10 (forward primer: GTGGCATTCAAGGAGTACCTC, reverse primer: TGATGGCCTTCGATTCTGGAT), ISG54 (forward primer: AAGAGTGCAGCTGCCTGAAC, reverse primer: CCTCCATCAAGTTCCAGGTG), ISG15 (forward primer: GGTGGACAAATGCGACGAA, reverse primer: TGCTGCGGCCCTTGTTAT).

All data were obtained from at least three biologically independent replicate experiments. Data are shown as mean±s.d. GraphPad Prism 8 software (GraphPad) was used for statistical analysis. A Student's t-test (unpaired and two-tailed) was used to analyze single comparisons between two independent groups. For multiple comparisons, one-way ANOVA with Tukey's multiple comparisons test was used for statistical analysis.

We thank Dr David Levy (University of Alabama at Birmingham, Birmingham, USA), Dr Jiahuai Han (Xiamen University, Xiamen, China), Dr Wentao Qiao (Nankai University, Tianjin, China), Dr Liangguo Xu (Jiangxi Normal University, Nanchang, China), Dr Yushan Zhu (Nankai University, Tianjin, China) and Dr Lin Liu (Nankai University, Tianjin, China) for providing plasmids. We thank the Sharing of Apparatus Management Platform (Nankai University, Tianjin, China) for providing the laser scanning confocal microscopy Monolith™ NT.115 instrument for MST experiments and the Tannon-5500 gel imager.