Abstract
Stimulator of interferon genes (STING) was initially described as a sensor of intracellular bacterial and viral DNA and a promising adjuvant target in innate immune cells; more recently STING has also been shown to detect endogenous DNA and play a role in tumor immunity and autoimmune disease development. Thus far STING has been studied in macrophages and dendritic cells. In this study, to our knowledge we provide the first evidence of STING activation in T cells, in which STING agonists not only provoke type I IFN production and IFN-stimulated gene expression, mirroring the response of innate cells, but are also capable of activating cell stress and death pathways. Our results suggest a re-evaluation of STING agonist–based therapies may be necessary to identify the possible effects on the T cell compartment. Conversely, the effects of STING on T cells could potentially be harnessed for therapeutic applications.
Introduction
Recognition of cytoplasmic nucleic acids by pattern recognition receptors (PRR) is crucial for cell defense and multiple pathways exist for this purpose, including the endoplasmic reticulum (ER) resident stimulator of interferon genes (STING). Unlike other nucleic acid sensors, STING does not directly bind DNA and instead recognizes cyclic dinucleotides (CDN) of either exogenous (e.g., bacterial) or endogenous origin (1). The latter, 2′3′cGAMP, is synthesized by cGAMP synthase upon binding cytoplasmic DNA resulting from cell damage, viral infection, or endogenous retroviruses (2). Murine, but not human, STING can also be activated by the synthetic small molecule DMXAA (3). Regardless of ligand, STING activation leads to strong type I IFN (IFN-I) production and increased expression of IFN-stimulated genes (ISG) (4). The two major IFN-I, IFN-α and β, alert other innate immune cells to detected threats and act in an autocrine manner to amplify the infected cell’s response. IFN responses provide crucial protection from many viral (and some bacterial) infections and STING agonists have been used as potent adjuvants to induce responses against model Ags and tumors (5).
Studies of STING primarily focus on its role in inducing macrophage and dendritic cell IFN-I responses to activate immediate innate defenses and direct subsequent effector T cell responses; thus, adaptive immune response differences in vivo following STING activation or deletion have been interpreted as the outcome of STING-dependent responses in innate cells influencing their activation of T cells. Whether STING could play a direct role in T cells has received little attention. Our interest was piqued by reported STING expression in the thymus and spleen when STING was first described (4) and STING mRNA expression in T cells (http://biogps.org). We asked what STING’s function might be in cells activated by TCR recognition of specific MHC peptide rather than PRR recognition of broad classes of pathogens in innate cells. One possibility was that STING does not directly activate T cells but does influence their behavior: several TLR, another class of PRR, are expressed by T cells and their stimulation in activated or memory CD4+ and CD8+ cells enhances proliferation and cytokine production (6), although TLR activation can also abrogate regulatory T cell suppressor function (7). We initially hypothesized that STING could have a similar modifying effect on T cell activation.
In this study, we show functional STING expression by T cells capable of initiating canonical IFN-I responses while also triggering T cell–specific responses that include increased expression of ER stress and cell death pathways in vitro. Many of these were augmented by concurrent TCR stimulation but STING activation alone induced large amounts of T cell death, a novel finding with implications for the development of therapies targeting STING.
Materials and Methods
Mice
B6 mice were from the Jackson Laboratory (Bar Harbor, ME); STING−/− mice were from Glen Barber and bred in house. For in vivo experiments, mice received 100 μg DMXAA i.v. in three doses over 2 d.
T cell purification and expansion
Total CD3+, CD4+, and CD8+ T cells were isolated from spleen and peripheral lymph nodes (pLN) using STEMCELL Technologies’ EasySep kits according to the manufacturer’s instructions. Typical purity was >97%. Expanded T cells were prepared from pLN cells using Mouse T activator CD3/CD28 Dynabeads (Thermo Fisher Scientific) with 50 U/ml recombinant IL-2.
T cell transfer experiment
CD3+ T cells were isolated from B6 mice expressing CD45.1 and 8 × 106 cells were adoptively transferred to CD45.2-expressing STING−/− mice. Following DMXAA treatment, CD3+CD45.1+ and CD3+CD45.2+ were separated by FACS for mRNA isolation.
T cell stimulation and proliferation
Purified or expanded T cells were activated with 10 μg/ml DMXAA unless otherwise indicated. For TCR stimulation, cells were added to plates coated overnight with 3 μg/ml anti-CD3 and anti-CD28 Abs; DMXAA and/or inhibitors were added with cells unless otherwise specified. Proliferation was determined by CFSE dilution in isolated CD3+ T cells after 3 d.
Immunoblots
Cell lysates were run on gradient gels, transferred to nitrocellulose membrane, and probed with primary Ab then fluorophore-conjugated secondary Ab. Fluorescence was read on a LI-COR Odyssey CLx at 700 and 800 nm.
Cytokine analysis
Supernatant cytokine concentration after 24 h was determined by sandwich ELISA (IFN-β, Santa Cruz and R&D Systems; IFN-γ, R&D Systems).
RT-PCR
cDNA was synthesized from Trizol-isolated RNA and SYBER green master mix (Fisher) was used to determine expression.
RNA sequencing
Trizol-isolated total RNA was used to construct a directional cDNA library (TrueSeq). Then 75 bp end-reads from cDNA libraries generated on MiSeq (Illumina) were aligned using TopHat2 and Cufflinks. The data are available at National Center for Biotechnology Information Gene Expression Omnibus under accession number GSE89361 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE89361).
Results and Discussion
STING expression and IFN production
We confirmed that murine T cells robustly expressed STING protein at levels comparable to, or higher than, macrophages (Fig. 1A) before testing their response to the STING-specific agonist DMXAA, which readily diffuses across the cell membrane and is a useful tool for experiments with primary T cells. Initially identified as an anti-vascular, pro-IFN cancer therapeutic (8), it was later shown to bind murine but not human (3) STING, activating the tank-binding protein 1 (TBK1)–IFN regulatory factor 3 (IRF3) axis and inducing IFN-I responses. Treatment with DMXAA induced STING-dependent expression of representative ISGs in B6 T cells (Fig. 1B). Because DMXAA is a small molecule and not a CDN, we also treated cells with thiol-modified CDN bisphosphorothioate analogue of cGAMP (R’S’cGAMP), which is known to activate macrophages without lipofection. Electroporation (Fig. 1D) with R’S’cGAMP robustly increased B6 T cell IFIT2 expression versus electroporation alone; lipofection had no effect (data not shown). Electroporation alone resulted in large amounts of cell death, so DMXAA was used in all subsequent experiments.
T cells exhibit a functional IFN-I response to STING agonists: (A) STING protein levels in unstimulated T cells and macrophages; ISG mRNA expression in T cells (B) after 4 h treatment with DMXAA, or (D) 5 h after electroporation with R’S’cGAMP. (C, E, and F) DMXAA-induced IFN-β and IFN-γ production, with and without anti-CD3 and anti-CD28, in total CD3+ or isolated CD4+ and CD8+ T cells. (G) TBK1-IRF3, NFKB, and MAPK phosphorylation in DMXAA-activated T cells. Results are representative of three to four independent experiments.
T cells exhibit a functional IFN-I response to STING agonists: (A) STING protein levels in unstimulated T cells and macrophages; ISG mRNA expression in T cells (B) after 4 h treatment with DMXAA, or (D) 5 h after electroporation with R’S’cGAMP. (C, E, and F) DMXAA-induced IFN-β and IFN-γ production, with and without anti-CD3 and anti-CD28, in total CD3+ or isolated CD4+ and CD8+ T cells. (G) TBK1-IRF3, NFKB, and MAPK phosphorylation in DMXAA-activated T cells. Results are representative of three to four independent experiments.
In addition to increasing ISG expression, DMXAA induced STING-dependent IFN-β and IFN-γ production (Fig. 1C), a striking result because IFN-I production is not typically associated with T cells, whereas IFN-γ is a major T cell cytokine. Moreover, anti-CD3/CD28 activation substantially increased IFN-β versus DMXAA alone (Fig. 1E). Although both CD4+ and CD8+ T cells produced STING-dependent IFN-β, only the former secreted IFN-γ (Fig. 1F).
To our knowledge, together these data provide evidence that activation of STING in T cells elicits an IFN-I response.
T cell signaling in response to DMXAA
As reported in other cell types (9), DMXAA triggered TBK1 and IRF3 phosphorylation in B6 but not STING−/− T cells; it also increased p-p38 and p-p65 (Fig. 1G). Changes in phosphorylation of the NFKB inhibitory protein p105 have not been reported following STING activation, but because TCR activation triggers p105 phosphorylation and degradation to the active transcription factor p50 (10), we were curious whether DMXAA affected this pathway too. Although we detected an increase in p-p105 (Fig. 1G), there was no change in total p105 or p50 (Supplemental Fig. 1A). However, constitutive processing of p105 can obscure p50 detection even when it is transcriptionally active.
In contrast to macrophages, autocrine IFN appeared to have no effect on T cell STING signaling: T cells from mice lacking the IFN-I receptor (IFNAR−/−) phenocopied B6 T cells (Supplemental Fig. 1B).
Thus, although STING activation in T cells triggers many of the same pathways described in innate cells, it also appears to have unique effects that might result in T cell–specific outcomes.
STING activation is independent of, but augmented by, simultaneous TCR stimulation
The effect of TLR ligands on T cells requires prior or simultaneous TCR stimulation (11, 12), so we investigated whether concurrent STING and TCR activation had synergistic effects on signaling. When we activated T cells with plate-bound anti-CD3 and anti-CD28 and added DMXAA, we found only DMXAA increased STING-dependent p-TBK1 or p-IRF3 (Supplemental Fig. 1C). Other NFKB and MAPK responses overlapped, with apparent additive and possibly synergistic effects: p105 phosphorylation was weakly induced by TCR stimulation at 60 min in both B6 and STING−/− cells but was stronger and more rapid in B6 cells when DMXAA was added. p-p38 was similarly enhanced by DMXAA beyond TCR alone only in B6 T cells. But although both B6 and STING−/− T cells robustly increased p-ERK in response to anti-CD3 and -CD28, which in addition to the p-p65 findings indicated a functional TCR in STING−/− T cells, no additive effect was detected with DMXAA and TCR activation (Supplemental Fig. 1C).
Characterization of STING−/− mouse T cell compartment
STING−/− and B6 T cells exhibited similar levels of p-ERK and p-p65 following TCR stimulation, suggesting no activation defects occur with STING deletion. However, in vivo infection and immunization studies have described altered T cell responses in STING−/− mice, although this has been interpreted as an effect on APC that alters their interactions with T cells (13).
To rule out major differences in T cell development, we compared thymic and pLN CD4−CD8−, CD4+CD8+, and single positive CD4 and CD8 T cell populations, as well as levels of CD4 and CD8 TCR expression, and found no significant differences between B6 and STING−/− mice (Supplemental Fig. 2A, 2C). We likewise found no differences in naive, memory, or regulatory T cell populations (Supplemental Fig. 2B, 2E), and B6 and STING−/− T cells expressed comparable levels of CD69 and CD25 following TCR activation (Supplemental Fig. 2D). Coupled with the absence of TCR signaling differences, these data imply B6 and STING−/− T cells are functionally equivalent.
Effect of DMXAA on T cell proliferation
Our signaling data suggested possible synergy between TCR and STING-mediated signaling in B6 T cells, so we next tested whether adding DMXAA to TCR stimulation altered T cell activation or proliferation. DMXAA had no effect on CD69 expression after 16 h but after 40 h we noted reduced B6 CD25 expression (Supplemental Fig. 2D); we suspect this is due to DMXAA-induced cell death (see below). Although pLN T cells from both strains proliferated equally in response to TCR stimulation, DMXAA blocked this expansion in B6 but not STING−/− T cells (Fig. 2A). IFNAR−/− T cells also failed to expand in the presence of DMXAA, dismissing any contribution of autocrine IFN-I (14).
DMXAA inhibits TCR-induced proliferation independent of IFN-I. (A) Anti-CD3 and anti-CD28 induced proliferation of B6, KO, and IFNAR−/− CD3+ T cells in the presence of 10 μg/ml DMXAA. (B) Effect of adding DMXAA concurrent with TCR stimulation (day 0), or after 24 (day 1) or 48 (day 2) h. Results shown are representative of three independent experiments.
DMXAA inhibits TCR-induced proliferation independent of IFN-I. (A) Anti-CD3 and anti-CD28 induced proliferation of B6, KO, and IFNAR−/− CD3+ T cells in the presence of 10 μg/ml DMXAA. (B) Effect of adding DMXAA concurrent with TCR stimulation (day 0), or after 24 (day 1) or 48 (day 2) h. Results shown are representative of three independent experiments.
We then tested whether DMXAA added after initial TCR stimulation continued to block proliferation. DMXAA added on day 0 or day 1 completely inhibited T cell expansion, but if added on day 2 some early proliferation occurred (Fig. 2B). These findings suggested DMXAA might not alter early T cell activation but might initiate a STING-dependent anti-proliferative pathway.
STING-dependent activation of IFN, cell death, and ER stress pathways
To take an unbiased approach to STING activation in T cells, we performed RNA-sequencing on naive B6 and STING−/− CD3+ T cells stimulated with anti-CD3 and anti-CD28, DMXAA alone or both. In agreement with our RT-PCR results, DMXAA alone increased expression of a suite of ISG by B6 but not STING−/− T cells (Fig. 3C), with a particularly striking effect on IFN-α2, 4, and 5; transcripts that are not highly upregulated in myeloid cells by DMXAA. Adding TCR stimulation further increased expression of most ISGs, reinforcing the possibility of synergy between these pathways.
STING activation increases B6 T cell expression of apoptosis, cell stress-related genes in addition to dramatic increases in ISG expression. (A) Gene pathways activated in response to TCR or TCR+DMXAA stimulation as determined using RNA sequencing data clustered based on similarity of samples using GenomePattern software (Broad Institute). (B) Changes in T cell expression of anti- and proapoptotic and ER stress-related genes in T cells in response to TCR or TCR+DMXAA activation. (C) STING-dependent upregulation of IFNs and selected ISGs in T cells. (D) STING-dependent activation of proapoptotic pathways in T cells versus macrophages. TCR: anti-CD3 and anti-CD28. The RNA-sequencing experiment was done once.
STING activation increases B6 T cell expression of apoptosis, cell stress-related genes in addition to dramatic increases in ISG expression. (A) Gene pathways activated in response to TCR or TCR+DMXAA stimulation as determined using RNA sequencing data clustered based on similarity of samples using GenomePattern software (Broad Institute). (B) Changes in T cell expression of anti- and proapoptotic and ER stress-related genes in T cells in response to TCR or TCR+DMXAA activation. (C) STING-dependent upregulation of IFNs and selected ISGs in T cells. (D) STING-dependent activation of proapoptotic pathways in T cells versus macrophages. TCR: anti-CD3 and anti-CD28. The RNA-sequencing experiment was done once.
Pathway analysis of these data unexpectedly highlighted STING-dependent increases in apoptotic and caspase cascade pathways and decreases in IL-2 and cell cycle pathways with DMXAA (Fig. 3A). Although TCR activation alone in B6 and STING−/− T cells elicited comparable increases in anti-apoptotic (e.g., BCL2) and decreases in proapoptotic (e.g., BAX) gene expression, adding DMXAA reversed this trend in a STING-dependent manner (Fig 3B), substantially downregulating BCL2 and upregulating BAX only in B6 T cells. Strikingly, none of these prodeath pathways were activated by DMXAA in macrophages: comparison of pathway activation between macrophages and T cells showed greater similarities between STING−/− T cells and B6 macrophages than STING−/− and B6 T cells (Fig 3D). CDN-induced lethality has not been noted in innate immune cells and we accordingly saw no increase in proapoptotic gene expression by macrophages. We reasoned this might be because innate immune cells must survive STING-triggering threats to initiate further immune responses, but if STING is primarily detecting viral infection or internal DNA damage in T cells, the host might be better served by cell death.
Although STING-induced death is one explanation for the block we observed in B6 T cell proliferation, it might also be related to reduced expression of the cell cycle and IL2 pathways in response to DMXAA (Fig. 3A), with IL2 in particular having a critical role in T cell expansion. Both B6 and STING−/− cells sharply increased IL2 expression with TCR activation, but DMXAA blocked this only in B6 cells (Fig. 3A).
Our pathway analysis unexpectedly revealed strong upregulation of genes involved in the unfolded protein (UPR) response, notably Bip/HSPA5 and GADD34, in parallel with ISG upregulation (Fig. 3B). The build-up of unfolded or misfolded proteins within the ER leads to ER stress and blocked protein translation that can kill the cell. UPR activation resolves this stress and prevents death by increasing the folding capacity of the ER (15, 16). TCR stimulation by itself can induce the UPR as activated cells ramp up protein production for proliferation; at the same time, TCR signaling induces Ca2+ flux and perturbation of ER Ca2+ concentration is known to chemically interfere with protein folding and trigger the UPR. Activation-induced translocation of STING from the ER membrane could further perturb the Ca2+ balance and overwhelm the capacity of the UPR to resolve stress, resulting in cell death.
STING activation triggers T cell death
Based on our RNA-sequencing data, we hypothesized that DMXAA might activate proapoptotic pathways in a STING-dependent manner, possibly by inducing a doomed UPR. To test this, TCR-activated T cells treated with DMXAA were stained with annexin V and propidium iodide (PI) to track cell death. The addition of DMXAA to anti-CD3 and anti-CD28 activated B6 T cells resulted in large numbers of annexin V+PI+ cells compared with their nonDMXAA counterparts but STING−/− T cells were largely annexin V−PI− regardless of DMXAA. We then treated cells with the pan-caspase inhibitor zVAD-fmk and/or the necroptosis inhibitor Nec1 to characterize the type of death (Fig. 4A). By itself neither was able to rescue DMXAA-treated B6 T cells but together they partially rescued STING-dependent T cell death, suggesting that denying DMXAA-treated T cells the apoptotic pathway may instead force death by necroptosis. This rescue appeared to be limited by the half-life of the inhibitors: neither zVAD, Nec1, nor a combination of the two rescued B6 cell proliferation when DMXAA was present (Fig. 4B).
DMXAA induces a STING-dependent ER stress response and cell death. zVAD and Nec1 together rescue DMXAA (10 μg/ml)-induced cell death after 24 h (A), but not proliferation at 72 h (B), in anti-CD3 and anti-CD28–activated T cells. (C) Cytotoxicity in CD3+ T cells of *DMXAA alone after 12 and 24 h is dose dependent. (D) The DMXAA-induced, STING-dependent increase in IFIT2 mirrors (E) the increased ratio of XBP-1s:XBP-1u and (F) decreased expression of anti-apoptotic BCL2 following DMXAA treatment. Results are representative of three independent experiments. (G) In vivo DMXAA treatment has no effect on pLN CD4+ and CD8+ T cell populations but (H) increases CD3+ T cell ISG expression independent of non-T cell IFN sources. Symbols represent individual mice.
DMXAA induces a STING-dependent ER stress response and cell death. zVAD and Nec1 together rescue DMXAA (10 μg/ml)-induced cell death after 24 h (A), but not proliferation at 72 h (B), in anti-CD3 and anti-CD28–activated T cells. (C) Cytotoxicity in CD3+ T cells of *DMXAA alone after 12 and 24 h is dose dependent. (D) The DMXAA-induced, STING-dependent increase in IFIT2 mirrors (E) the increased ratio of XBP-1s:XBP-1u and (F) decreased expression of anti-apoptotic BCL2 following DMXAA treatment. Results are representative of three independent experiments. (G) In vivo DMXAA treatment has no effect on pLN CD4+ and CD8+ T cell populations but (H) increases CD3+ T cell ISG expression independent of non-T cell IFN sources. Symbols represent individual mice.
To clarify whether DMXAA-induced cell death requires simultaneous TCR activation (17) we tested the effect of DMXAA alone at varying doses. Naive B6 T cells were largely dead after 12 h treatment with 10 μg/ml DMXAA, whereas 5 μg/ml DMXAA induced an equal degree of cell death after 24 h. DMXAA concentrations below 5 μg/ml had no effect on viability even after 24 h, whereas STING−/− T cells showed no increase in cell death at any dose or either time point (Fig. 4C).
We next investigated whether there is a link between STING, the UPR, and T cell death. In macrophages IFN production following PRR activation of the UPR requires activation of the transcription factor XBP-1 by unconventional splicing (18), so we examined whether STING activation by DMXAA induced XBP-1 splicing as part of the UPR in T cells. CD3+ T cells from B6 and STING−/− mice expressed comparable levels of unspliced XBP-1 (XBP-1u) that were unchanged by TCR and DMXAA activation, but DMXAA alone increased expression of spliced XBP-1 (XBP-1s) only in B6 T cells, resulting in a higher ratio of XBP-1s:XBP-1u that grew with the addition of anti-CD3 and anti-CD28 activation (Fig. 4E). Concurrent with this STING-dependent increase in the ratio of XBP-1s:XBP-1u was increased expression of IFIT2 and decreased expression of BCL-2 in B6 cells (Fig. 4D, 4F), confirming our RNA-sequencing data showing a correlation between increased ISG and UPR and decreased anti-apoptotic gene expression. Together these data support a link between STING activation in T cells, UPR induction, and subsequent cell death, possibly due to failed ER stress resolution.
Despite the dramatic evidence of ER stress linked T cell death in vitro, when mice were injected i.v. with DMXAA on two successive days we saw no changes in pLN or spleen-naive and memory T cell populations indicative of cell death (Fig. 4G, data not shown). However, we did note dramatic increases in ISG expression in response to DMXAA. To ensure these changes were independent of IFN-I production from non-T cells, we transferred STING+/+ CD3+ T cells expressing CD45.1 into STING−/− recipients that express CD45.2. After DMXAA administration only CD3+CD45.1+ T cells exhibited increased ISG expression (Fig. 4H). The absence of T cell death in our in vivo model could be due to a dosage effect, with T cells being exposed to a lower local concentration of DMXAA than in our in vitro experiments. Alternatively, T cells may be receiving survival signals from other cell populations in vivo that are unavailable in vitro.
We describe in this study two novel findings: first, analogous to its effects in innate immune cells, the STING pathway induces a IFN-I response in T cells. This adds to a growing body of evidence that PRR are functional in T cells, though unlike other PRR our data uniquely shows a TCR activation–independent response to STING agonists. In addition, we report STING-dependent UPR induction and cell death in T cells that has not been described in myeloid cells.
The functionality and role of PRR in T cells has been a subject of study and some debate for many years. T cells are known to express various PRR that provide costimulatory signals (19, 20), but TCR-independent cytosolic DNA-sensing could be uniquely useful to allow surveillance of infection or DNA damage in T cells and enable self-destruction, preventing infected or damaged and potentially cancerous T cells from proliferating and causing disease. These findings complement a recent report by Tang et al. (21) demonstrating CDN-induced apoptosis in B cells and B cell lymphomas. But although they concluded this response was B cell–specific because no effect on T cells was observed with in vitro cGAMP treatment, our DMXAA and R’S’cGAMP experiments conclusively demonstrate T cells are responsive to STING activation.
Why the T cell response to STING activation in vitro results in cell death but in vivo leads to IFN-I responses requires further investigation.
These initial findings caution that careful study of STING agonist effects on T cells will be necessary whenever these agonists are tested for therapeutic effects. Boosting T cell IFN-I production with STING agonists, particularly when chimeric Ag receptor T cells are transferred to patients, could dramatically improve anti-tumor responses. Likewise, the inclusion of a DMXAA analog as a vaccine adjuvant may be particularly useful when the desired outcome is an IFN-dominated response. However, more long-term effects of STING activation on T cells remain to be studied. B cell lymphoma death in response to injected cGAMP was studied after weeks of multiple injections (21) and a similar timeline might result in T cell stress and death more similar to our in vitro experiments instead of the IFN-I response we found in our short-term in vivo experiment.
Overall, the work presented in this study strongly suggests that STING has its own unique functional outcomes in an adaptive immune cell type, and it may yet have tricks up its sleeve awaiting our discovery.
Footnotes
This work was supported by National Institutes of Health Grants AI-056234, AI-119833, and AI-126050 to A.P.; Russian Science Fund Project 15-15-00100 (RNA-sequencing) to A.P.; and by National Institutes of Health Grant T32-AI-007077 to the Immunology Graduate Program (Tufts University).
The sequences presented in this article have been submitted to the National Center for Biotechnology Information Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo) under accession number GSE89361.
The online version of this article contains supplemental material.
Abbreviations used in this article:
- CDN
cyclic dinucleotide
- ER
endoplasmic reticulum
- IFN-I
type I IFN
- IRF3
IFN regulatory factor 3
- ISG
IFN-stimulated gene
- pLN
peripheral lymph node
- PI
propidium iodide
- PRR
pattern recognition receptor
- R’S’cGAMP
bisphosphorothioate analogue of cGAMP
- STING
stimulator of interferon genes
- TBK1
tank-binding protein 1
- UPR
unfolded protein
- XBP-1s
spliced XBP-1
- XBP-1u
unspliced XBP-1.
References
Disclosures
The authors have no financial conflicts of interest.