LRRC8A:C/E Heteromeric Channels are Ubiquitous Transporters of cGAMP

SUMMARY

Extracellular 2’3’-cyclic-GMP-AMP (cGAMP) is an immunotransmitter exported by diseased cells and imported into host cells to activate the innate immune STING pathway. We previously identified SLC19A1 as a cGAMP importer, but its use across human cell lines is limited. Here, we identify LRRC8A heteromeric channels, better known as volume-regulated anion channels (VRAC), as widely-expressed cGAMP transporters. LRRC8A forms complexes with LRRC8C and/or LRRC8E, depending on their expression levels, to transport cGAMP and other 2’3’-cyclic dinucleotides. In contrast, LRRC8D inhibits cGAMP transport. We demonstrate that cGAMP is effluxed or influxed via LRRC8 channels, as dictated by the cGAMP electrochemical gradient. Activation of LRRC8A channels, which can occur under diverse stresses, strongly potentiates cGAMP transport. We identify activator sphingosine 1-phosphate and inhibitor DCPIB as chemical tools to manipulate channel-mediated cGAMP transport. Finally, LRRC8A channels are key cGAMP transporters in resting primary human vasculature cells and universal human cGAMP transporters when activated.

eTOC Blurb:

2’3’-cyclic-GMP-AMP (cGAMP) is a paracrine innate immune messenger. It is produced and exported by cells upon detection of cytosolic dsDNA and imported into neighboring cells to locally activate Stimulator of Interferon Genes (STING). Using a CRISPR screen, Lahey et al. identify LRRC8A channels as widely-expressed bi-directional transporters of cGAMP.

INTRODUCTION

A potent and versatile immune signaling molecule, 2’3’-cyclic-GMP-AMP (cGAMP) is a second messenger that couples detection of pathogen- or damage-associated threats to activation of innate immunity. cGAMP is produced when cyclic-AMP-GMP synthase (cGAS) detects double stranded DNA (dsDNA) in the cytosol of metazoan cells (Ablasser et al., 2013a; Diner et al., 2013; Gao et al., 2013; Wu et al., 2013; Zhang et al., 2013). Cytosolic dsDNA represents a conserved danger signal that can arise from diverse threats to a cell: dsDNA viruses (Li et al., 2013), retroviruses (Gao et al., 2015), bacteria (Manzanillo et al., 2012; Nandakumar et al., 2019), cancerous states (Bakhoum et al., 2018; Mackenzie et al., 2017), damage (Harding et al., 2017; Hatch et al., 2013; McArthur et al., 2018), and senescence (Dou et al., 2017; Glück et al., 2017; Yang et al., 2017). Once synthesized, cGAMP activates its receptor, Stimulator of Interferon Genes (STING) (Ablasser et al., 2013a; Diner et al., 2013; Zhang et al., 2013), to trigger activation of TBK1 and IRF3 signaling (Liu et al., 2015; Tanaka and Chen, 2012; Zhong et al., 2008). The resulting production of type I interferons and other cytokines then induces powerful inflammatory and defense responses (Sun et al., 2013). However, aberrant activation of this pathway, such as from DNAse deficiencies (Gao et al., 2015; Stetson et al., 2008) or mitochondrial damage (McArthur et al., 2018), can lead to deleterious effects such as autoimmune diseases (Crowl et al., 2017). In order to understand the crucial roles of the cGAS-STING axis in homeostasis and disease, it is imperative to determine how nature regulates signaling by the potent central messenger of the system, cGAMP.

cGAMP is not only a cell-intrinsic activator of STING, but also a paracrine signal that orchestrates larger-scale biological responses. During viral infection, cGAMP produced in infected cells passes through connexin gap junctions into immediately adjacent cells to activate local anti-viral innate immunity (Ablasser et al., 2013b). We recently reported another distinct cGAMP paracrine signaling mechanism: cGAMP is exported directly to the extracellular space in a soluble, non-membrane bound form (Carozza et al., 2020a). Moreover, irradiated cancer cells produce and export cGAMP, leading to an increase in innate immune cells in the tumor microenvironment and contributing to the curative effect of ionizing radiation in a STING-dependent manner. Extracellular cGAMP signaling has now been reported by others in the context of cancer, UV damage, and viral infections (Cordova et al., 2020; Zhou et al., 2020a, 2020b), and could, conceivably, operate in other physiological settings.

cGAMP is negatively charged, and as an extracellular messenger, must rely on facilitated mechanisms to cross the plasma membrane. We and others identified the reduced folate carrier SLC19A1 as the first importer of extracellular cGAMP (Luteijn et al., 2019; Ritchie et al., 2019a), however our studies indicated the presence of additional cGAMP import mechanisms. We were unable to identify a primary human cell type that uses SLC19A1 as its dominant cGAMP importer (Ritchie et al., 2019a). Our previous work also revealed that different cell types have varying degrees of sensitivity to extracellular cGAMP and that part of this difference was controlled at the level of import machinery. Furthermore, the identity of the cGAMP exporter(s) in cancer cells is currently unknown. Discovery of additional cGAMP transport mechanisms is imperative to understand the extent, selectivity, and regulation of extracellular cGAMP signaling in homeostasis and disease.

Here, through genetic, pharmacological, and electrophysiological approaches, we identify LRRC8A:C/E heteromeric channels as direct, bi-directional transporters of cGAMP. Interestingly, the transmembrane pore of LRRC8A channels shares rough structural similarities with that of connexins—gap junctions that are known to transport cGAMP(Ablasser et al., 2013b; Deneka et al., 2018; Kasuya et al., 2018; Kefauver et al., 2018). However, LRRC8A channels open to the extracellular space instead of linking together the cytosol between two cells. Since LRRC8A was identified as an essential component of the ubiquitous volume-regulated anion channel (VRAC) (Qiu et al., 2014; Voss et al., 2014), our understanding of its roles in biology are rapidly expanding beyond regulation of cell-intrinsic physiology; LRRC8A-containing channels are increasingly recognized to mediate cell-cell communication via extracellular transport of various signaling molecules (Chen et al., 2019; Osei-Owusu et al., 2018; Yang et al., 2019). We now report 2’3’-cyclic dinucleotides as an immunostimulatory substrate class for LRRC8A channels and characterize the regulation of this transport mechanism.

RESULTS

A Genome-Wide CRISPR Screen Identifies LRRC8A as a Positive Regulator of Extracellular cGAMP-Mediated STING Pathway Activation

We previously identified SLC19A1 as the first cGAMP importer by performing a whole-genome CRISPR knockout screen in the U937 monocyte-derived cell line (Ritchie et al., 2019a). While SLC19A1 was the dominant importer in U937 cells, the existence of additional cGAMP import pathways was evident from residual signal observed in SLC19A1^−/− cells: extracellular cGAMP treatment still led to activation of STING, phosphorylation of STING, the kinase TBK1, and the transcription factor IRF3, and ultimately cytokine production and cell death (Figure 1A) (Ritchie et al., 2019a). To identify SLC19A1-independent import mechanisms, we performed a whole-genome CRISPR knockout screen (Morgens et al., 2017) in U937 Cas9-SLC19A1^−/− cells (Figure S1A, S1B). We determined that ~30 μM extracellular cGAMP was a 50% lethal dose (LD₅₀) in U937 Cas9-SLC19A1^−/− cells at 48 hours (Figure S1C). We treated the CRISPR library daily for 12 days with cGAMP and passaged untreated cells in parallel as controls (Figure 1B). At the end of the selection, we isolated genomic DNA and sequenced the encoded sgRNAs. Fold changes of sgRNA sequences in cGAMP-treated cells versus controls were analyzed using the Cas9 high-Throughput maximum Likelihood Estimator (casTLE) statistical framework to calculate a confidence score and effect for each gene (Morgens et al., 2016). Key STING pathway components, TMEM173 (STING), TBK1, and IRF3, were reproducibly identified with high confidence scores (Figure 1C) and large positive effect sizes (Figure 1D), validating this screening approach. Interestingly, LRRC8A clustered near known STING pathway members (Figure 1C – 1D).

Figure 1. A Genome-Wide CRISPR Screen Identifies LRRC8A as a Positive Regulator of Extracellular cGAMP-Mediated STING Pathway Activation.

(A) Scheme of extracellular cGAMP-STING signaling in U937 SLC19A1^−/− cells. Imported cGAMP binds to and activates STING, which results in phosphorylation of STING, TBK1, and the transcription factor IRF3, and ultimately leads to interferon induction and cell death.

(B) CRISPR screen strategy. U937 Cas9-SLC19A1^−/− cells were transduced with a whole-genome sgRNA lentiviral library. In two replicates, library cells were treated with cGAMP or untreated for 12 days. Genomic DNA was harvested, deep sequenced, and analyzed for sgRNA depletion or enrichment.

(C) Plot of casTLE score for each gene in replicate 1 versus replicate 2 with displayed correlation across all genes (R² = 0.86). Top scoring hits are annotated.

(D) Volcano plot of casTLE effect size versus false discovery rate (FDR) for all genes across replicates. Top hits and all LRRC8 genes are annotated. A 5% FDR significance threshold is indicated.

See also Figure S1.

LRRC8A and LRRC8 Paralogs Differentially Facilitate cGAMP Import

LRRC8A was recently identified as an essential component of the volume-regulated anion channel (VRAC) that resides on the plasma membrane (Qiu et al., 2014; Voss et al., 2014). Upon cell swelling or sensing of various stimuli, opening of VRAC leads to outward flow of chloride, organic osmolytes, and water to facilitate regulatory volume decrease (Chen et al., 2019; Osei-Owusu et al., 2018; Strange et al., 2019). In order to form channels with functional VRAC activity in cells, LRRC8A must associate with one or more paralogous proteins, LRRC8B–E (Voss et al., 2014). These LRRC8 complexes are reported to be heteromeric hexamers in which subunit stoichiometry may be variable (Gaitán-Peñas et al., 2016; Lutter et al., 2017). Depending on the paralog(s) in complex with LRRC8A, the resulting channel exhibits different properties (Syeda et al., 2016) and can transport different substrates. While complexes containing LRRC8B–E mediate inorganic anion flux, LRRC8A:C, LRRC8A:D, and LRRC8A:E complexes transport larger anionic substrates such as glutamate and aspartate (Gaitán-Peñas et al., 2016; Lutter et al., 2017; Planells-Cases et al., 2015; Schober et al., 2017). LRRC8A:C and LRRC8A:E channels have been reported to transport ATP (Gaitán-Peñas et al., 2016). Distinct characteristics of LRRC8D-containing channels include mediating efflux of large uncharged or positively charged cellular osmolytes (Lutter et al., 2017; Schober et al., 2017) and import of the drugs blasticidin, cisplatin, and carboplatin (Lee et al., 2014; Planells-Cases et al., 2015). It is therefore likely that LRRC8A forms heteromeric channels with other LRRC8 paralogs to regulate cGAMP transport.

As the obligatory component of the channel, LRRC8A exerted the strongest effect among the paralogs in the screen (Figure 1D). LRRC8C had a moderate positive effect while LRRC8D had a negative effect. LRRC8B and LRRC8E fell outside of the 5% false discovery rate (FDR) threshold and demonstrated no or minimal effect, respectively, which may result in part from low transcription of LRRC8B and LRRC8E in the U937 line (Morgens et al., 2017) (Figure S1D).

To validate the screen result and begin investigating the role of LRRC8 paralogs in extracellular cGAMP signaling, we individually knocked out each gene in U937 SLC19A1^−/− cells using multiple gene-specific sgRNAs. We also used a non-targeting scrambled sgRNA to serve as a negative control. We then isolated and validated single cell clones, treated them with extracellular cGAMP, and measured phosphorylation of IRF3 (p-IRF3) to assess STING pathway activation (Figure S2). LRRC8A knockout clones exhibited a ~40% reduction in p-IRF3 relative to total IRF3 as compared to scramble controls, while LRRC8C knockout clones exhibited a ~30% reduction (Figure 2A). In contrast, we observed a ~40% increase in p-IRF3 signaling in LRRC8D knockout clones. Negligible differences were observed in LRRC8B and LRRC8E knockouts, both of which are lowly expressed at the transcript level. Together, these data validate the effects observed in the whole-genome screen that LRRC8 paralogs play differential regulatory roles in extracellular cGAMP signaling.

Figure 2. LRRC8A and LRRC8 Paralogs Differentially Facilitate cGAMP Import.

(A) U937 SLC19A1^−/− subclones expressing scramble or LRRC8A–LRRC8E sgRNAs were treated with 100 μM cGAMP for 4 h and signaling was assessed by Western blot (see Figure S2 for display of all blots). Summarized results for unique subclones (confirmed heterozygous or homozygous knockouts) are plotted relative to the respective scramble control average (n = 4–9 subclones tested in 1 experimental replicate). Significance calculated by two-tailed t-test; *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

(B) U937 SLC19A1^−/−-scramble or -LRRC8A^−/− through -LRRC8E^−/− clones were electroporated with 100 nM cGAMP, cultured for 2 h, and signaling was assessed by Western blot (n = 2 biological replicates).

For (A)–(B), plotted range is mean ± SD.

See also Figure S2.

We then tested whether LRRC8 paralogs affect the response to exogenous cGAMP at the level of import or by altering downstream STING pathway signaling. When cGAMP was electroporated into the cells, bypassing the need for specific importers, LRRC8A^−/− through LRRC8E^−/− clones no longer exhibited differential p-IRF3 signaling compared to scramble controls (Figure 2B). Taken together, our data support a model in which LRRC8A, likely in complex with LRRC8C, facilitates the transport of extracellular cGAMP across the plasma membrane into U937 cells. Our results also demonstrate that LRRC8D inhibits extracellular cGAMP-mediated activation of STING in a plasma membrane transport-dependent manner. As to potential molecular mechanisms of the latter finding, we hypothesized that LRRC8D could negatively regulate cGAMP import by i) promoting cGAMP export to counteract import or ii) inhibiting both directions of cGAMP transport, two scenarios that we are able to distinguish between later.

LRRC8A-Containing VRAC Channels Directly Transport cGAMP

We next sought to determine whether cGAMP is directly transported by LRRC8A channels. Alternatively, VRAC activation could alter membrane potential to influence cGAMP transport through another route not revealed by our CRISPR screen. Much work dissecting VRAC function has been performed in HEK293 cell lineages (Osei-Owusu et al., 2018; Pedersen et al., 2015), which we determined respond to extracellular cGAMP in an SLC19A1-independent manner (Ritchie et al., 2019a). Following the generation and validation of LRRC8 paralog knockout pools (Figure S3A), we tested whether these channels facilitate cGAMP import in HEK293 cells. Using the most upstream reporter of cGAMP import, phosphorylation of STING (p-STING) following cGAMP binding, we observed that HEK293 LRRC8A knockout pools exhibited a ~20% decrease in cellular response to extracellular cGAMP compared to controls (Figure 3A). Conversely, LRRC8D knockout pools exhibited a ~30% increase in extracellular cGAMP response. These results indicate that LRRC8A-containing channels also facilitate import of cGAMP in HEK293 cells, although they do not account for the dominant mechanism. Consistent with our observations in U937 cells, LRRC8A and LRRC8D affect the cGAMP import process in opposing directions in HEK293 cells.

Figure 3. LRRC8A-Containing VRAC Channels Directly Transport cGAMP.

(A) HEK293 scramble or LRRC8A–LRRC8E knockout pools were treated with 100 μM cGAMP for 2 h and signaling was assessed by Western blot. Representative blots displayed with quantification (n = 4 biological replicates).

(B) HEK293 scramble or LRRC8A–LRRC8E knockout pools were treated with 20 μM cGAMP for 1 h in isotonic or hypotonic solution and signaling assessed by Western blot. Representative blots displayed with quantification (n = 3 biological replicates).

(C) Scramble or LRRC8A knockout cells were transfected for expression of mSTING. After 24 h, cells were treated with 250 μM DMXAA or 100 μM GAMP in isotonic or hypotonic buffer for 1 h. Activation of p-mSTING or p-hSTING signaling was assessed by Western blot. Representative blots displayed with quantification (n = 2 technical replicates). ND = not detected.

(D) Extracellular cGAMP concentrations from HEK293 scramble or LRRC8A–LRRC8E knockout pools were measured following cGAS plasmid transfection from media (24 h) or after hypotonic buffer stimulation (20 min) using STING-CAP and the cGAMP-Luc assay (n = 3 biological replicates).

(E) Representative whole-cell patch clamp current traces measured in HEK293 cells upon hypotonic stimulation in the absence or presence of cGAMP (100 μM). Overlays of the current-voltage relationship at the indicated points are also displayed. For additional traces, see Figure S3.

(F) Summary data for inhibition of VRAC current by 100 μM extracellular cGAMP at +150 mV (n = 8 measured in 5 cells).

For (A)–(D), and (F), plotted range is mean ± SD. Significance calculated using two-tailed t-test; *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, not significant (NS).

See also Figure S3.

We hypothesized that if cGAMP is a substrate of LRRC8A, stimulating opening of the channel should increase influx of extracellular cGAMP and downstream STING signaling. VRAC activation and induction of a LRRC8A-dependent VRAC current is routinely achieved by application of hypotonic extracellular buffer (Qiu et al., 2014; Voss et al., 2014). Indeed, delivery of extracellular cGAMP in a hypotonic buffer drastically increased STING activation compared to delivery in isotonic buffer (Figure 3B). This potentiation was completely abolished in LRRC8A knockout cells and significantly reduced in LRRC8C and LRRC8E knockout cells relative to control cells (~40% and ~65% reductions, respectively). In contrast, a trend towards a small but reproducible increase in hypotonic response was observed in the LRRC8D knockout pool. These results demonstrate that activation of channels containing LRRC8A, likely in complex with LRRC8C and/or LRRC8E, potentiates cellular response to extracellular cGAMP. To test whether hypotonic buffer increases cGAMP signaling by promoting cellular uptake of the molecule, we utilized a membrane-permeable small molecule mouse STING agonist, DMXAA, as a negative control (Conlon et al., 2013; Kim et al., 2013). We over-expressed mouse STING (mSTING) in HEK293 cells, which express human STING (hSTING) naturally, and treated with DMXAA or cGAMP in isotonic or hypotonic solution. Hypotonic buffer strongly potentiated cGAMP-, but not DMXAA-induced, mSTING activation and this increase was mostly abolished in LRRC8A knockout pools (Figure 3C). While hSTING was not activated by DMXAA treatment, as expected, hypotonicity again potentiated cGAMP activation of hSTING in a LRRC8A-dependent manner. Collectively, these data demonstrate that LRRC8A channel activation boosts signaling for one STING agonist (cGAMP) but not another (DMXAA), and suggest that LRRC8A channels act by increasing cGAMP transport as opposed to altering STING signaling.

We then sought to directly measure LRRC8A-dependent cGAMP transport. While direct detection of cGAMP import into the cell cytosol is technically challenging (Ritchie et al., 2019b), we established methods to measure cGAMP export into cell culture media with nanomolar sensitivity (Carozza et al., 2020a; Mardjuki et al., 2020). When synthesized intracellularly, both the voltage and chemical gradients for negatively charged cGAMP favor exit from the cell. If cGAMP export occurs via a channel, the flux should be regulated by channel activation state. VRAC-mediated efflux of ATP has been demonstrated upon channel opening (Dunn et al., 2020; Gaitán-Peñas et al., 2016; Hisadome et al., 2002). To determine if LRRC8-containing VRAC channels also mediate cGAMP export, we stimulated intracellular cGAMP production in HEK293 scramble and LRRC8A–E knockout pools by transfecting cells with a cGAS expression plasmid in the presence of an inhibitor of extracellular cGAMP hydrolysis (Carozza et al., 2020a, 2020b). After 24 hours, we harvested the media from each condition, treated cells with a hypotonic buffer for 20 minutes to activate channel opening, and measured cGAMP concentrations in the extracellular solutions using a coupled enzyme assay, cGAMP-Luc (Mardjuki et al., 2020). cGAMP was secreted into the media at similar levels from control and LRRC8 knockout pools, suggesting that LRRC8 channels are not major cGAMP exporters in HEK293 cells under resting conditions (Figure 3D). In contrast, application of hypotonic buffer robustly stimulated LRRC8A- and LRRC8E-dependent cGAMP export (Figure 3D), strongly indicating that LRRC8A-containing channels, when opened, mediate cGAMP export. In contrast, hypotonicity-stimulated cGAMP export was increased in LRRC8D knockout cells to a small but statistically significant extent, signifying that LRRC8D inhibits cGAMP export. Similar to LRRC8D, LRRC8B knockout also increased hypotonicity-stimulated cGAMP export from HEK cells. While reported LRRC8B expression is low and variable across cell lines (Figure S4B), if present at a certain threshold, LRRC8B appears to act similarly to LRRC8D. We observe that LRRC8D acts to decrease both cGAMP import (Figure 2A, 3A) and export (Figure 3D), indicating that LRRC8D generally inhibits transport of the negatively charged molecule in a manner independent from direction of flux. Given that LRRC8D complexes with LRRC8A, we propose a model in which decreased transport results from the inhibitory paralog diluting or poisoning active cGAMP transporting LRRC8A:C/E complexes.

We next tested whether cGAMP directly interacts with LRRC8A channels. Previous electrophysiology studies demonstrated that extracellular ATP inhibits VRAC-mediated Cl⁻ influx via direct pore block (Jackson and Strange, 1995; Kefauver et al., 2018; Tsumura et al., 1996). We tested whether extracellular cGAMP also blocks Cl⁻ influx through VRAC. Whole-cell patch clamp analysis of HEK293 cells reveals that the characteristic, hypotonic solution-induced VRAC current measured at positive membrane potential is inhibited by introduction of cGAMP in the bath solution (Figure 3E, 3F, and Figure S3B). Similar to inhibition by ATP (Figure S3C), inhibition by cGAMP is rapid and reversible, providing strong support for the conclusion that cGAMP directly interacts with the LRRC8A-containing VRAC channel. The observed inhibition is also consistent with cGAMP being directly transported by the LRRC8A-containing channels: larger substrates that are transported slowly generally act as inhibitors (“permeating blockers”) of fast traveling, small substrates through channels (Woodhull, 1973) – ATP was shown to be such a permeating blocker of Cl⁻ influx through VRAC (Hisadome et al., 2002). Therefore, it is likely that cGAMP, like ATP, permeates through the LRRC8A channel.

The LRRC8A:C Channel Is a Major cGAMP Importer in Microvasculature Cells

We then sought to identify cell lineages in which LRRC8 channels play a dominant role in cGAMP transport. Based on our LRRC8 subunit analyses in U937 and HEK293 cells, cell types expressing high levels of LRRC8A (and either LRRC8C or LRRC8E), but low levels of LRRC8D, are predicted to import cGAMP effectively. Upon evaluation of RNA expression levels across cell types in The Human Protein Atlas (proteinatlas.org), we identified that several vasculature cell lineages fit such a profile (Figure S4A). TIME cells, an immortalized human microvascular endothelial cell (HMVEC) line, express LRRC8A transcripts at a 15.2-fold higher level than U937 cells while expressing LRRC8D transcripts at a 4.3-fold lower level (Figure S4B). Moreover, LRRC8C is expressed at a 1.6-fold higher level in TIME cells than U937 cells. Indeed, TIME cells responded to extracellular cGAMP with an EC₅₀ of ~24 μM (Figure S4C), which is more sensitive than most cell lines we have evaluated to date, including U937 cells with an EC₅₀ of 270 μM (Ritchie et al., 2019a). To determine if cGAMP import depends on LRRC8 complexes in these cells, we knocked out LRRC8A–E using CRISPR, as well as the first identified importer, SLC19A1 (Figure S4D). While no significant effect was observed upon SLC19A1 knockout, LRRC8A or LRRC8C knockout in TIME cells resulted in loss of the majority of response to extracellular cGAMP (Figure 4A), but not to transfected cGAMP (Figure S4E), supporting a role for LRRC8A and LRRC8C in cGAMP transport across the plasma membrane in this cell line. LRRC8D knockout cells exhibited an increased p-STING response, the amplitude of which is lower than in U937 cells, likely reflective of lower LRRC8D expression in TIME cells. Overall, reported mRNA levels appear to be a reasonable proxy for channel protein expression across the cell lines studied here, given that observed effect sizes upon knockout trend with transcript expression (Figure S4F). Further opening the channel by hypotonic treatment led to a ~40-fold increase in p-STING signaling in response to extracellular cGAMP treatment, which was significantly ablated in LRRC8A and LRRC8C knockout cells (Figure 4B). Additionally, knockout of both LRRC8A and LRRC8C did not have an additive effect compared to single knockouts (Figure 4C), consistent with LRRC8A and LRRC8C forming heteromeric complexes to import cGAMP in TIME cells.

Figure 4. The LRRC8A:C Channel Is a Major cGAMP Importer in Microvasculature Cells.

(A) TIME LRRC8A–E and SLC19A1 knockout pools were treated with 50 μM cGAMP for 2 h and signaling was assessed by Western blot (n = 3 biological replicates).

(B) TIME scramble or LRRC8A–LRRC8E knockout pools were treated with 10 μM cGAMP for 1 h in isotonic or hypotonic solution and signaling assessed by Western blot (n = 3 biological replicates).

(C) TIME LRRC8A, LRRC8C, and LRRC8A:C knockout pools were treated with 50 μM cGAMP for 2 h and signaling was assessed by Western blot (n = 3 biological replicates).

(D) Cryo-EM structure of human homo-hexameric LRRC8A (PDB: 5ZSU) (Kasuya et al., 2018). Surface display of the complex, with ribbon representation of two opposing LRRC8A subunits and zoom to pore gating residue R103.

(E) TIME knockout cells were induced to express Flag-tagged LRRC8A, LRRC8A-R103L, or LRRC8C, then treated with 50 μM cGAMP for 2 h and signaling assessed by Western blot (n = 3 biological replicates).

For (A)–(C) and (E), representative blots are shown with quantification of all experiments and range plotted as mean ± SD. Significance calculated using two-tailed t-test; *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, not significant (NS).

See also Figure S4.

We then probed how LRRC8A channels may interact with cGAMP. Recent cryo-electron microscopy structures of LRRC8A homomeric complexes (Deneka et al., 2018; Kasuya et al., 2018; Kefauver et al., 2018) revealed that the narrowest region of the LRRC8A channel is gated by an arginine residue (R103), creating a strong positive charge potential at the extracellular pore (Figure 4D). Both LRRC8C and LRRC8E are predicted to have a leucine residue at the corresponding position. The phosphodiester ring of cGAMP is double-negatively charged, therefore we predicted that R103 contributed by the LRRC8A subunit is crucial for cGAMP recognition. To test this hypothesis, we evaluated whether rescued expression of wild-type proteins and R103L substituted LRRC8A restored signaling in the corresponding knockout TIME cells. We generated stable cell lines in which LRRC8A-Flag and LRRC8C-Flag expression is driven from a tetracycline-inducible promoter. Unfortunately, we were not able to overexpress LRRC8D to probe its inhibitory role in these cells as its expression was consistently low or undetectable. We observed concentration-dependent increases in extracellular cGAMP response upon rescue of LRRC8A and LRRC8C protein expression (Figure 4E). In contrast, overexpression of LRRC8A-R103L in LRRC8A knockout cells failed to rescue response to extracellular cGAMP (Figure 4E). This result supports a role for the pore-gating residue R103 in cGAMP recognition in a manner that is consistent with charge-charge interactions regulating channel access.

LRRC8A Channels Transport Other 2’3’-Cyclic Dinucleotides

We next evaluated the selectivity of LRRC8A complexes against a broad panel of natural and synthetic cyclic dinucleotides (CDNs) (Figure 5A). Synthetic nonhydrolyzable phosphorothioate cGAMP analogs 2’3’-cG^SA^SMP (Li et al., 2014) and 2’3’-CDA^S (Corrales et al., 2015) have been designed as potent STING agonists for innate immunotherapy. 2’3’-CDA^S, also known as ADU-S100, is undergoing evaluation in clinical trials for treatment of metastatic cancers (ClinicalTrials.gov: NCT02675439, NCT03172936, and NCT03937141). Additionally, bacteria produce CDN second messengers with a different 3’3’-phosphodiester linkage between nucleotides. Bacterial CDNs include 3’3’-cyclic-GMP-AMP (3’3’-cGAMP), 3’3’-cyclic-di-GMP (3’3’-CDG), and 3’3’-cyclic-di-AMP (3’3’-CDA) (Burdette et al., 2011; Woodward et al., 2010; Zhang et al., 2013), which can trigger host STING signaling upon infection (Dey et al., 2015; Sauer et al., 2010; Woodward et al., 2010). We observed that LRRC8A channels account for the majority of cGAMP, 2’3’-cG^SA^SMP, and 2’3’-CDA^S import in TIME cells at the evaluated concentrations, while playing a minor role in 2’3’-CDA import (Figure 5B). No prominent LRRC8A-dependent difference in signaling was observed for the tested 3’3’-CDNs. While this could indicate that 3’3’-CDNs are not channel substrates, it could also result from their weaker agonism of STING by reducing the sensitivity of p-STING detection, or perhaps by limiting potential STING signaling-stimulated opening of LRRC8A channels. Direct detection of 3’3’-CDN import in future studies can discriminate between these possibilities.

Figure 5. LRRC8A Channels Transport Other 2’3’-Cyclic Dinucleotides.

(A) Chemical structures of 2’3’- and 3’3’-cyclic dinucleotide (CDN) STING agonists.

(B) TIME scramble and LRRC8A knockout pools were treated with 50 μM cGAMP, 25 μM 2’3’-cG^SA^SMP, 100 μM 2’3’-CDA, 25 μM 2’3’-CDA^S, 100 μM 3’3’-cGAMP, 200 μM 3’3’-CDA, or 200 μM 3’3’-CDG for 2 h and signaling assessed by Western blot (n = 2 biological replicates). Representative blots are shown with quantification of all experiments and range plotted as mean ± SD. ND = not detected.

(C) TIME scramble control or LRRC8A knockout pools were treated with cGAMP or 2’3’-cG^SA^SMP at the indicated concentrations for 3 h. Induction of IFNB1 relative to untreated cells was quantified by RT-qPCR (Figure S5) and fold change in IFNB1 induction between scramble versus LRRC8A-KO cells calculated. Plotted range is mean ± SD (n=2 biological replicates).

See also Figure S5.

Nevertheless, it appeared that one or more other import mechanisms exist for 2’3’- and 3’3’-CDNs in TIME cells. We therefore tested the concentration dependence of LRRC8 channel-mediated transport for cGAMP and degradation-resistant 2’3’-cG^SA^SMP. We treated cells with serial dilutions of both substrates down to ~1 μM, measured upregulation of interferon β-encoding RNA transcripts (IFNB1) by RT-qPCR (Figure S5), and evaluated the relative fold change in response between scramble control and LRRC8A knockout cells. As expected, higher concentrations of extracellular cGAMP and 2’3’-cG^SA^SMP stimulated robust IFNB1 induction in control cells but IFNB1 induction was decreased in cells lacking LRRC8A (Figure 5C). Interestingly, at low CDN concentrations the relative responses invert with LRRC8A knockout cells being more activated by both CDNs, represented by ratios less than one. These data imply that an unknown importer is dominant at low extracellular CDN concentrations and that, within this regime, LRRC8A channels contribute to cellular export of cGAMP. This behavior is consistent with the electrochemical forces that dictate direction of anion flow through a channel. The negative resting membrane voltage gradient of a cell favors cGAMP efflux owing to the molecule’s double negative charge. However, if cGAMP is sufficiently concentrated in the extracellular space, the chemical gradient then drives influx of cGAMP. In total, our results suggest that the LRRC8A complex is an importer of cGAMP at high extracellular concentrations, but acts as an exporter in cells already loaded with cGAMP.

cGAMP Import by the LRRC8A:C Channel Can Be Potentiated by Sphingosine-1-Phosphate and Inhibited by DCPIB

After establishing a role for LRRC8A:C-mediated cGAMP transport in TIME cells, we next sought to identify chemical tools to regulate cGAMP flux via this mechanism. VRAC is activated downstream of sphingosine-1-phosphate (S1P) signaling to one of its G protein-coupled receptors, S1PR1 (Burow et al., 2014). In addition, endothelial cells are exposed to S1P in the blood and lymph (at ~1 μM and ~100 nM, respectively), and also produce S1P themselves (Cartier and Hla, 2019). We hypothesized that increasing S1P levels within this physiologic range would further increase cGAMP uptake by TIME cells in vitro. Additionally, we sought to determine the effect of DCPIB (Figure 6A), a known small molecule VRAC inhibitor (Decher et al., 2001; Qiu et al., 2014), on the response to extracellular cGAMP. The recently solved structure of homomeric LRRC8A in complex with DCPIB reveals that the inhibitor binds within the narrow pore formed by R103, acting like a “cork in the bottle” (Kern et al., 2019). Since cGAMP uptake also requires R103, we anticipated that DCPIB should inhibit cGAMP import. We observed that S1P increased extracellular cGAMP signaling by 2- to 3-fold in a dose- and LRRC8A-dependent manner (Figure 6B), but did not lead to STING activation in the absence of cGAMP (Figure S6). Conversely, DCPIB significantly inhibited LRRC8A-dependent extracellular cGAMP signaling under basal and S1P-stimulated conditions. Our results demonstrate that LRRC8A:C channel uptake of extracellular cGAMP into TIME cells is potentiated by nanomolar concentrations of S1P. In addition, DCPIB can be used as a tool to block LRRC8A-dependent cGAMP import.

Figure 6. Pharmacologic and Genetic Perturbations Reveal LRRC8A-Containing Channels are Major cGAMP Transporters in Primary Human Endothelial Cells.

(A) Chemical structure of DCPIB.

(B) TIME scramble and LRRC8A knockout pools were treated with cGAMP (50 μM), with or without S1P (10 or 100 nM) and DCPIB (20 μM), for 1 h and signaling assessed by Western blot (n = 3 biological replicates). Significance calculated using two-tailed t-test; *P ≤ 0.05, **P ≤ 0.01.

(C) Following siRNA knockdown of LRRC8A expression in HUVEC cells pooled from six donors, cells were treated with cGAMP (100 μM) for 2 h. Western blots shown with quantification (n = 1).

(D) HUVEC cells pooled from 6 donors were treated with cGAMP (50 μM), with or without S1P (10 or 100 nM) and DCPIB (20 μM) for 1 h and signaling assed by Western blot (n = 2 biological replicates).

For (B) and (D), representative blots are shown with quantification of all experiments and range plotted as mean ± SD.

See also Figure S6.

Pharmacologic and Genetic Perturbations Reveal LRRC8A-Containing Channels are Major cGAMP Transporters in Primary Human Endothelial Cells

Numerous studies have revealed that stromal cells are key responders to paracrine or pharmacologic STING activation. In the case of cancer, the stromal (non-immune) compartment helps contribute to generation of downstream anti-tumor immune signaling cascades. For example, during early engraftment in a B16.F10 melanoma model, spontaneous STING activation and interferon production are detected in endothelial cells prior to infiltration of dendritic cells and CD8⁺ T cells to the tumor microenvironment (Demaria et al., 2015). Additionally, induction of a tumor necrosis phenotype upon injection of 2’3’-CDA^S into B16.F10 tumors depends on STING signaling in the stromal compartment (Francica et al., 2018). Elucidating the uptake mechanisms of extracellular cGAMP and its analogs into stroma, such as vasculature and connective tissues, is therefore necessary to understand the selectivity and regulation of this biologically important STING signaling axis.

The key import role of LRRC8A:C observed in TIME cells, an immortalized microvasculature endothelial lineage, strongly suggested that cGAMP import in primary human endothelial cells could also occur through LRRC8A channels. siRNA knockdown in primary human umbilical vein endothelial cells (HUVEC) pooled from six donors decreased LRRC8A protein levels by ~70–90% (Figure 6C). As predicted, all siRNA treatments decreased extracellular cGAMP signaling, with oligo-9, -10, and -11 knocking down p-IRF3 signaling by more than 50%. These results demonstrate that the LRRC8A channel is not only a major cGAMP importer in immortalized human microvasculature cells, but also in primary HUVECs.

Given the modulation of LRRC8A-mediated cGAMP import that we achieved using S1P and DCPIB in the microvasculature line, we tested whether these same tools could effectively regulate cGAMP import in primary human endothelial cells. Indeed, the presence of physiological S1P concentrations increased extracellular cGAMP signaling in HUVEC cells by up to 40% (Figure 6D). DCPIB again inhibited both basal and S1P-potentiated cGAMP import. In total, these data establish pharmacologic modulation of the LRRC8A channel as a potential strategy to control cGAMP import into human vasculature.

DISCUSSION

Here, we report identification of LRRC8A:C/E heteromeric channels as the second known transport mechanism of extracellular cGAMP. The LRRC8 family was recently discovered to form the channels responsible for VRAC activity after three decades of active searching (Cahalan and Lewis, 1988; Grinstein et al., 1982; Qiu et al., 2014; Voss et al., 2014). Following the molecular identification of VRAC, detailed mechanistic studies became possible and are rapidly providing new insights. Beyond small anions, LRRC8A channels transport metabolites, neurotransmitters, and small molecule drugs (Lee et al., 2014; Lutter et al., 2017; Yang et al., 2019). It is becoming clear that VRAC’s namesake function in regulation of cell volume represents a subset of its physiological roles. LRRC8A has been linked to cellular development, migration, and apoptosis, as well as cell-to-cell communication (Chen et al., 2019; Osei-Owusu et al., 2018).

Structural studies recently revealed notable similarities and differences between two complexes that transport cGAMP: LRRC8A-containing channels and connexins. In the solved homomeric structures of LRRC8A and connexin Cx26, both form hexameric complexes with the four alpha helices of the respective transmembrane regions forming highly analogous secondary structures (Deneka et al., 2018; Maeda et al., 2009). This homology supports the idea that substrates of these complexes, such as cGAMP, could overlap. In the case of gap junctions, two connexin complexes on neighboring cells associate to form a channel between the two plasma membranes that allows direct cytosol-to-cytosol transport. In contrast, a single LRRC8A complex forms a channel that links the cytosol to the extracellular space and is gated by a narrow extracellular pore. While gap junction-mediated cGAMP transfer was known to enable rapid signal transduction upon viral infection in tightly packed cell matrices (Ablasser et al., 2013b), LRRC8A-mediated transport is suited to enable longer-range cGAMP communication through the extracellular milieu to non-adjacent or migrating cells. In this manner, these large channel complexes are poised to play complementary and non-overlapping roles in paracrine cGAMP signaling.

Once the LRRC8A channel is open, the direction of a substrate’s flux is dictated by the molecule’s electrochemical gradient across the plasma membrane. Indeed, other than serving as a cGAMP importer, we demonstrate that the LRRC8A channel, when activated, exports cGAMP out of HEK cells overexpressing cGAS. Therefore, LRRC8A is the first reported cGAMP exporter. While this manuscript was under review, a new report was published in which LRRC8A:E channels were identified as important for the elicitation of immunity against DNA viruses in murine models, specifically by mediating cGAMP transfer from infected to responding cells (Zhou et al., 2020a). In the setting of HSV-1 infection, Zhou et al. measured cGAMP export from murine embryonic fibroblasts in an LRRC8A-dependent manner, although ~2/3 of the total cGAMP release occurred by an unknown mechanism. Using only transfected DNA as a stimulus for cGAMP production, we do not observe a significant role for LRRC8A–E in basal cGAMP export from HEK293 cells. However, activation of VRAC with hypotonic solution leads to robust LRRC8A- and LRRC8E-dependent cGAMP efflux from the cells. The discrepancies observed by our two groups suggest that regulation of channel activation by cell-intrinsic mechanisms, or cell-extrinsic factors (e.g., viral elements), likely determines the dominance of LRRC8A-mediated cGAMP export. While the mechanism of LRRC8 channel opening remains unsolved to date, a wide range of stimuli have been reported to regulate channel activation, including but not limited to purinergic signaling, calcium signaling, apoptosis-inducing drugs, TNFα, PI3K-Akt, and oxidative stress (Chen et al., 2019). We speculate that such factors, several of which have been associated with STING signaling, may promote cGAMP release in physiological contexts.

We also provide insights into how the subunit composition of the heteromeric LRRC8 complex influences cGAMP transport and demonstrate that paralog use varies in a cell type-specific manner. LRRC8A-mediated VRAC function is reported to be ubiquitous across cell types, and we observe a contribution of LRRC8A to cGAMP signaling in all human cell lines evaluated. In contrast to the report by Zhou et al., we determine that both LRRC8C and LRRC8E, when complexed with the obligatory LRRC8A subunit, transport cGAMP. While LRRC8C and LRRC8E exhibit high homology (Abascal and Zardoya, 2012) and share common substrates such as aspartate and ATP (Gaitán-Peñas et al., 2016; Lutter et al., 2017; Schober et al., 2017), differences in their substrate transport dynamics have been established. At strongly depolarized membrane potentials, LRRC8E rapidly promotes channel inactivation while LRRC8C is associated with slowing inactivation (Voss et al., 2014). In addition, hypotonicity-induced ATP export occurs more readily through LRRC8A:E than LRRC8A:C channels when heterologously expressed in Xenopus oocytes (Gaitán-Peñas et al., 2016). Beyond channel properties, differences in tissue distribution and abundance have been reported, with LRRC8E expression more restricted than that of LRRC8C (Pervaiz et al., 2019), and this likely determines paralog use for cGAMP transport. In fact, differences in paralog expression between cell lines may help explain why extracellular cGAMP-dependent block of VRAC current was not observed in HeLa cells by Zhou et al., but was observed in our hands using HEK293 cells (which are reported to express higher levels of both LRRC8C and LRRC8E compared to HeLa cells, in addition to lower levels of LRRC8D) (Figure S4B). Finally, it is possible that certain signaling networks may activate channels containing LRRC8C and not LRRC8E (or vice versa), which could enable differential regulation of cGAMP transport. Despite the current discrepancies in reported paralog use (murine LRRC8A:E vs. human LRRC8A:C/E), the identification of a cGAMP transport mechanism that is conserved between mice and humans is noteworthy in light of the species-specific differences in STING signaling that have complicated research and therapeutic developments in the field. For example, DMXAA is a potent anti-cancer murine STING agonist, but it does not activate human STING (Conlon et al., 2013; Kim et al., 2013). Human SLC19A1 imports cGAMP, but mouse SLC19A1 does not (Luteijn et al., 2019; Ritchie et al., 2019a). The discovery of a cGAMP transporter that spans species will enable study of the diverse in vivo contexts in which this mechanism may regulate STING signaling and present opportunities to translate those discoveries into advancements for human therapeutics.

Moreover, this work reveals that LRRC8D plays an inhibitory role in cGAMP transport that counters the action of LRRC8C and LRRC8E. Mechanistically, such inhibition could arise because inclusion of LRRC8D into complexes with LRRC8A:C/E reduces or abolishes cGAMP transport efficiency, or alters channel activation. In addition, formation of separate LRRC8A:D complexes that poorly conduct cGAMP could sequester LRRC8A and decrease the concentration of productive LRRC8A:C/E units. LRRC8-Dcontaining complexes transport a diverse range of substrates, including molecules with net neutral, positive, or negative charge and of varying sizes (Gaitán-Peñas et al., 2016; Lee et al., 2014; Lutter et al., 2017; Planells-Cases et al., 2015; Schober et al., 2017). Others have demonstrated that increasing the anionic character of a substrate decreases the role that LRRC8D plays in its transport: when the zwitterion taurine is neutrally charged, its export is LRRC8A- and LRRC8D-dependent, but when modulated to a negatively charged molecule, transport becomes LRRC8D-independent but remains LRRC8A-dependent (Schober et al., 2017). This finding suggests that some negatively charged substrates are not transported via LRRC8A:D complexes, but rather through LRRC8A:C/E channels, which may incorporate LRRC8D to some degree. Our studies, which identify a molecular composition of LRRC8A:C/E for cGAMP transport and a competing role for LRRC8D, are largely consistent with the above model. While we did not observe a dominant regulatory role for LRRC8B here, we speculate it could play an inhibitory role similar to LRRC8D if its expression reduces formation of productive LRRC8A:C/E channels. Building from the requirement of a specific LRRC8A heteromeric complex and differential expression profiles of the paralogs across cell types, we noticed that human vasculature cells express high levels of LRRC8A and LRRC8C, but low levels of LRRC8D. Indeed, an immortalized microvasculature cell line and primary human endothelial cells use LRRC8A channels as a key cGAMP transport mechanism. We predict that other cell lineages that share expression profiles enriched for LRRC8A:C/E can also transport cGAMP via this route.

LRRC8A channels represent the first identification of a major cGAMP transport mechanism in primary human cells, since the previously identified cGAMP importer SLC19A1 plays limited roles in tested primary cells (Ritchie et al., 2019a). Our identification of LRRC8A channels as a key cGAMP transporter in vasculature cells also provides a molecular mechanism for sensing other 2’3’-CDNs, such as therapeutics. Tumor vascularization is required to meet high metabolic demand and enable tumor outgrowth, and therefore has been a target for anti-cancer drug development for decades. Tumor vasculature-disrupting agents have shown promising results in treatment of solid tumors by shutting down blood flow and causing massive tumor necrosis (Tozer et al., 2005). It was recently shown that the ability of the cGAMP analog 2’3’-CDA^S to induce tumor necrosis depends on activation of STING in stromal cells and, moreover, that signaling between the stromal and hematopoietic compartments is beneficial for generation of anti-cancer immunity (Francica et al., 2018). With 2’3’-CDA^S currently undergoing evaluation in clinical trials for treatment of metastatic cancers, establishing how the investigational new drug reaches different effector cells will aid understanding of its efficacy and potential off-target effects. Our results suggest that LRRC8A channels may represent a major uptake mechanism of 2’3’-CDA^S injected into tumor vessels or accessible to the microvasculature. Building upon the discoveries described here, modulation of LRRC8A:C/E channel activity represents a novel pharmacologic strategy to selectively increase or decrease extracellular CDN to STING signaling, especially in vasculature, and may hold the potential to boost elicitation of anti-tumor responses or dampen unwanted inflammatory responses in autoimmune diseases.

STAR METHODS

RESOURCE AVAILABILITY

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Dr. Lingyin Li (lingyinl@stanford.edu).

Materials Availability

All unique/stable reagents generated in this study are available from the Lead Contact without restriction.

Data and Code Availability

CRISPR screen results were analyzed using the Cas9 high-Throughput maximum Likelihood Estimator (casTLE) statistical framework (Morgens et al., 2016), available at https://bitbucket.org/dmorgens/castle. CRISPR screen sgRNA counts data are available in Table S3. Complete, uncropped versions of all western blots appearing in this paper are accessible from Mendeley Data at http://dx.doi.org/10.17632/g7rwfvgjbm.1.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Mammalian tissue culture

Unless otherwise noted, cell lines were obtained from ATCC. U937 cells were maintained in RPMI (Cellgro) supplemented with 10% heat-inactivated FBS (Atlanta Biologicals) and 1% penicillin-streptomycin (Gibco). HEK293 cells were maintained in DMEM (Cellgro) supplemented with 10% FBS and 1% penicillin-streptomycin. Telomerase-immortalized human microvascular endothelium (TIME) cells line were maintained in vascular cell basal media supplemented with microvasculature endothelial cell growth kit-VEGF (ATCC) and 0.1% penicillin-streptomycin. Human umbilical vein endothelial cells (HUVEC) pooled from six different donors were purchased from Lonza and maintained in EGM-2 supplemented growth media (Lonza). All cells were maintained in a 5% CO₂ incubator at 37 °C.

METHOD DETAILS

Reagents and antibodies

2’3’-cyclic-GMP-AMP (cGAMP) was synthesized and purified in-house as previously described (Ritchie et al., 2019a). DMXAA, 2’3’-bisphosphorothioate-cyclic-GMP-AMP (2’3’-cG^SA^SMP), 2’3’-cyclic-di-AMP (2’3’-CDA), 2’3’-bisphosphorothioate-cyclic-di-AMP (2’3’-CDA^S), 3’3’-cyclic-GMP-AMP (3’3’-cGAMP), 3’3’-cyclic-di-AMP (3’3’-CDA), and 3’3’-cyclic-di-GMP (3’3’-CDG) were purchased from Invivogen and reconstituted in endotoxin-free water. Sphingosine-1-phosphate (d18:1) was purchased from Cayman Chemical. 4-[(2-Butyl-6,7-dichloro-2-cyclopentyl-2,3-dihydro-1-oxo-1H-inden-5-yl)oxy] butanoic acid (DCPIB) was purchased from Tocris Bioscience and reconstituted to 50 mM in DMSO. Antibodies and dilutions used for Western blotting are listed in Table S1.

Generation of CRISPR edited cell lines

LentiCRISPR v2 (Addgene) was used as the 3^rd-generation lentiviral backbone for all knockout lines. Sequences for all sgRNAs used in this study are listed in Table S2. The guide sequences were cloned into the lentiviral backbone using the Lentiviral CRISPR Toolbox protocol from the Zhang Lab at MIT (Sanjana et al., 2014; Shalem et al., 2014). Lentiviral packaging plasmids (pHDM-G, pHDM-Hgmp2, pHDM-tat1b, and pRC/CMV-rev1b) were purchased from Harvard Medical School. 500 ng of the lentiviral backbone plasmid containing the guide sequence and 500 ng of each of the packaging plasmids were transfected into HEK 293T cells using FuGENE 6 transfection reagent (Promega). The viral media was exchanged after 24 hours, harvested after 48 hours and passed through a 0.45 μm filter. U937 cells were transduced by spin infection in which cells were suspended in viral media with 8 μg/mL polybrene (Sigma Aldrich), centrifuged at 1000 × g for 1 hour, then resuspended in fresh media. HEK293 and TIME cell lines were reverse transduced by trypsinizing adherent cells and adding cell suspensions to viral media supplemented with 8 μg/mL polybrene. All lines were put under relevant antibiotic selection beginning 72-hours post-transduction and lasting until control (untransduced) cells completely died.

Cell viability measurement

U937 cells were treated with the indicated concentrations of cGAMP for 24 or 48 hours in duplicate. Cell density and viability was determined by hemocytometer and trypan blue staining.

U937 SLC19A1^−/− CRISPR knockout library generation

Creation of the whole-genome CRISPR sgRNA library was previously described by Morgens et al. (2017). Briefly, a whole-genome library of exon-targeting sgRNAs were designed, with the goal of minimizing off-target effects and maximizing gene disruption. The top 10 sgRNA sequences for each gene were included in the library, along with thousands of safe-targeting and non-targeting negative controls. The library was cloned into a lentiviral vector, pMCB320, which also expresses mCherry and a puromycin resistance cassette. The U937 Cas9-SLC19A1^−/− line was generated by lentiviral transduction as described above. Knockout of SLC19A1 was confirmed by genomic DNA sequencing (Figure S1A) and Cas9 function was confirmed by the efficiency of knocking out GFP expression using a sgRNA sequence against GFP confirmed by using a sgRNA sequence against GFP (Figure S1B). U937 Cas9-SLC19A1^−/− cells were infected with the lentiviral CRISPR sgRNA library and selected with puromycin.

CRISPR screen

The U937 CRISPR knockout library line was grown in 4 spinner flasks (1 L), with 2 flasks serving as untreated controls and 2 flasks receiving cGAMP treatment. Throughout the screen all of the samples were split daily to keep the cell density at 250 million cells per 500 mL, which corresponded to 1,000 cells per guide in the untreated samples. The experimental samples were treated daily with enough cGAMP to reduce treated cell fold growth by 50% as compared to the control samples. Initial treatments began at 15 μM for two days and were then titrated to 30 μM. After a difference of ten population doublings was achieved between experimental and control samples (12 days), the genomic DNA was extracted using a Qiagen Blood Maxi Kit. The library was sequenced using a NextSeq 500/550 Mid Output v2 kit (Illumina).

Sequencing of gDNA to confirm CRISPR editing

gDNA was isolated from cell lines using the QIAamp DNA Mini Kit (Qiagen). Sequences of each gene loci flanking the sgRNA target sequence were PCR amplified. PCR products were purified and submitted for Sanger sequencing analysis. Control (unedited) and experimental (edited) sequence traces were analyzed using the Inference of CRISPR Edits (ICE) version2 software tool (Synthego). ICE knockout scores represent the proportion of sequences in which a coding frameshift or >21-bp insertion/deletion are detected.

Electroporation of STING agonists

U937 cells were resuspended in 100 μL electroporation solution (90 mM Na₂HPO₄, 90 mM NaH₂PO₄, 5 mM KCl, 10 mM MgCl₂, 10 mM sodium succinate) with the appropriate concentration of STING agonist. Cells were then transferred to a cuvette with a 0.2 cm electrode gap (Bio-Rad) and electroporated using program U-013 on a Nucleofector II device (Lonza). Following electroporation, cells were transferred to media and cultured as indicated. Cell lysates were then harvested for analysis.

Extracellular cGAMP treatments with manipulation of buffer osmolarity

HEK293 cell lines were seeded in 12-well plates at 300,000 total cells one day before treatment. Following removal of complete DMEM media, HEK293 cells were treated for 1 h with 20 μM cGAMP in either isotonic (150 mM NaCl, 6 mM KCl, 1 mM MgCl_2, 1.5 mM CaCl₂, 10 mM glucose, 10 mM HEPES, pH 7.4, ~320 mOsm) or hypotonic buffer (105 mM NaCl, 6 mM KCl, 1 mM MgCl_2, 1.5 mM CaCl₂, 10 mM glucose, 10 mM HEPES, pH 7.4, ~240 mOsm). TIME cell lines were seeded in 12-well plates at 100,000 total cells one day before treatment. Following removal of vascular cell complete media, TIME cells were treated for 1 h with 10 μM cGAMP in either isotonic (125 mM NaCl, 6 mM KCl, 1 mM MgCl_2, 1.5 mM CaCl₂, 10 mM glucose, 10 mM HEPES, pH 7.4, ~270 mOsm) or hypotonic buffer (60 mM NaCl, 6 mM KCl, 1 mM MgCl_2, 1.5 mM CaCl₂, 10 mM glucose, 10 mM HEPES, pH 7.4, ~155 mOsm).

Stimulation of cGAMP synthesis and measurement of export in HEK293 cells

HEK293 cell lines were seeded in PurCol-coated (Advanced BioMatrix) 6-well plates at 300,000 total cells in 2 mL media one day before transfection. At the start of the experiment, media was gently removed and replaced with complete DMEM supplemented with ENPP1 inhibitor STF-1084 for 50 μM final. Cells were then transfected with 1500 ng pcDNA-FLAG-HA-sscGAS plasmid complexed with FuGene 6 reagent (Promega) according to manufacturer’s instructions or treated with FuGene 6 alone as a negative control. After 24 hours of incubation, media from each condition was harvested and cells were immediately incubated for 20 minutes with hypotonic buffer (60 mM NaCl, 6 mM KCl, 1 mM MgCl_2, 1.5 mM CaCl₂, 10 mM glucose, 10 mM HEPES, pH 7.4, ~155 mOsm), supplemented with 2 μM STF-1084. Following collection, media and buffer samples were centrifuged at 1000 × g for 5 minutes and supernatants collected. Plated cells were lysed in 200 mM NaOH and total protein concentration determined by BCA assay (Thermo Fisher). Extracellular cGAMP was quantified using STING-CAP and cGAMP-Luc (Mardjuki et al., 2020).

Electrophysiology

Endogenous VRAC currents from HEK293 cells were measured by whole-cell patch-clamp. Cells were transferred from culture dishes to the electrophysiology recording chamber following treatment with 0.5 mg/mL trypsin (Sigma-Aldrich) for 2 minutes. External recording solution contained 88 mM NaCl, 10 mM HEPES with either 110 mM mannitol (isotonic, 300 mOsm/kg) or 30 mM mannitol (hypotonic, 230mOsm/kg), pH 7.4 by NaOH. The internal (pipette) solution contained 130 mM CsCl, 10 mM HEPES, 4 mM Mg-ATP, pH 7.3. Borosilicate glass micropipettes (Sutter Instruments, Novato, CA) were pulled and fire-polished to a tip diameter with a typical resistance of 1.5–3.0 MΩ. Data were acquired using an Axopatch-200B amplifier (Axon Instruments, Union City, CA) and InstruTECH ITC-16 interface (HEKA Instruments, Holliston, MA), with a sampling rate of 5 kHz and filtering at 1 kHz. Igor Pro (WaveMetrics, Portland, OR) software was used for stimulation and data collection. Currents were measured in response to voltage ramps from -150 to 150 mV over 1.0 s, with an inter-ramp interval of 10 s and a holding potential of 0 mV. All recordings were carried out at room temperature (20 – 22 °C).

Lipofection of cGAMP

Lipofectamine 3000 (Thermo Fisher Scientific) was diluted into Opti-MEM (3 μL into 50 μL), P3000 was diluted into Opti-MEM (2 μL into 50 uL) in the absence or presence of cGAMP (for 1 μM final in experiment), and then dilutions were combined and incubated for 15 minutes. Lipid complexes (100 μL) were added dropwise to plated cells in 900 uL media. In parallel, free cGAMP was added for 1 μM final in the absence of lipid reagent.

Inducible expression of LRRC8A and LRRC8C

Gene fragments encoding LRRC8A (Uniprot Q8IWT6) and LRRC8C (Uniprot Q8TDW0) were commercially synthesized and cloned into plasmids (Twist Bioscience). Synonymous DNA mutations were introduced by QuickChange to ablate downstream Cas9-sgRNA targeting at each site (LRRC8A: GGATCCTGAAGCCGTGGT to GCATTTTAAAACCATGGT, LRRC8C: GTTATGAGCGAGCCCTCCAC to GCTACGAACGCGCGTTACAT). Additionally, QuickChange was used to generate a sequence with a LRRC8A-R103L encoding substitution (CGG to CTG). A parent pLVX-TetOne plasmid (Takara Bio) was modified by 1) insertion of a GGSG-FLAG encoding sequence to flank the multiple cloning site and 2) insertion of SV40 promoter and hygromycin resistance factor encoding sequences were cloned in following the TetOn 3G element, yielding a pLVX-TetOne-FLAG-Hygro plasmid. LRRC8A, LRRC8A-R103L, and LRRC8C DNA fragments (all with ablated sgRNA sites) were then cloned into pLVX-TetOne-FLAG-Hygro backbone by Gibson assembly. Lentiviral packaging of constructs was performed as described above and reverse transduced into LRRC8A^−/− or LRRC8C^−/− TIME cells. Cells were selected with hygromycin for 2 weeks. Doxycycline was added to cultures for 36 hours at the indicated concentrations before each experiment to induce expression.

IFNB1 qPCR

TIME cells were incubated with the indicated concentration of cGAMP, 2’3’-cG^SA^SMP, or untreated for 3 h. Following treatment, cells were lysed with 0.4 mL TRIzol (Invitrogen) and total RNA from cells was extracted following manufacturer’s instructions. RNA was reverse transcribed in 20 μL reactions containing 500 ng total RNA, 100 pmol Random Hexamer Primers (Thermo Scientific), 0.5 mM dNTPs (NEB), 40 U RNaseOUT (Invitrogen), 1x Maxima RT Buffer (Thermo Scientific), and 200 U Maxima Reverse Transcriptase (Thermo Scientific). RT reactions were incubated for 10 min at 25°C, then for 30 min at 50°C, and terminated by incubating for 5 min at 85°C. To quantify transcript levels, 10 μL reactions were set up containing 1x GreenStar Master Mix (Bioneer), 10x ROX dye (Bioneer), 100 nM forward and reverse primers, and 0.7 μL of RT reactions. To determine Ct values, reactions were run on a ViiA 7 Real-Time PCR System (Applied Biosystems) using the following program: ramp to 50°C (1.6 °C/s), incubate for 2 min, ramp to 95°C (1.6°C/s), incubate for 10 min; then 40 cycles of ramp to 95 °C (1.6°C/s) and incubate for 15 sec, ramp down to 60 °C (1.6°C/s) and incubate for 1 min. The following primers sets were used: IFNB1 transcript 5’-AAACTCATGAGCAGTCTGCA-3’ (forward), 5’-AGGAGATCTTCAGTTTCGGAGG-3’ (reverse) (Richtsteiger et al., 2003); ACTB transcript 5’-GGCATCCTCACCCTGAAGTA-3’ (forward), 5’-AGAGGCGTACAGGGATAGCA-3’ (reverse).

S1P Reconstitution

Lipid was solubilized in warm methanol, aliquoted, then solvent evaporated under nitrogen before storage at −20*C. At the start of each experiment, S1P was freshly reconstituted to 100 μM with 4 mg/mL fatty acid-free human serum albumin carrier protein (Millipore Sigma) in PBS.

siRNA knockdown of LRRC8A in HUVEC

Four ON-TARGETplus siRNAs for knockdown of LRRC8A were purchased from Dharmacon, along with non-targeting control siRNA. HUVEC cells were seeded the night before transfection in 6-well plates at 2.5×10⁵ total cells in 2 mL media, with a change to fresh media on the day of knockdown. Following manufacturer’s instructions DharmaFECT4 (6 uL) was complexed with siRNA (10 uL of 5 uM stock) in Opti-MEM to yield 200 uL of complexes then used to transfect cells. Cells were split on day three post-transfection and tested for cGAMP response on day four.

QUANTIFICATION AND STATISTICAL ANALYSIS

For CRISPR screen sequencing data, experimental and control samples were analyzed using casTLE (Morgens et al., 2016), to determine confidence scores and effect sizes. P-values were adjusted for multiple test corrections using the Benjamini-Hochberg method to determine false discovery rates. All other statistical analyses were performed using GraphPad Prism 8.3.1. For experiments involving Western blots, densitometric measurements of protein bands were made using ImageJ 1.52a. Measurements of phosphorylated proteins (p-IRF3 and p-STING) were normalized to matched total protein (IRF3 or STING) or tubulin, as indicated. Phoshpo to total protein ratios were evaluated for significance using the indicated method and then plotted relative to the respective scramble control value.

Supplementary Material

2

Table S3. CRISPR screen sgRNA counts, Related to STAR Methods

Deep sequencing data were analyzed using casTLE (Morgens et al., 2016), which yielded raw count values listed for each indicated treatment condition.

KEY RESOURCES TABLE.

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Rabbit monoclonal anti-phospho-STING (Ser366) (clone D7C3S)	Cell Signaling Technology	Cat#19781; RRID: AB_2737062
Rabbit monoclonal anti-phospho-STING (Ser365) (clone D8F4W)	Cell Signaling Technology	Cat#72971; RRID: AB_2799831
Rabbit monoclonal anti-phospho-IRF3 (Ser396) (clone 4D4G)	Cell Signaling Technology	Cat#4947; RRID: AB_823547
Rabbit monoclonal anti-STING (clone D2P2F)	Cell Signaling Technology	Cat#13647; RRID: AB_2732796
Rabbit monoclonal anti-IRF3 (clone D83B9)	Cell Signaling Technology	Cat#4302; RRID: AB_1904036
Mouse monoclonal anti-alpha-Tubulin (clone DM1A)	Cell Signaling Technology	Cat#3873; RRID: AB_1904178
Rabbit polyclonal anti-DYKDDDDK Tag (“FLAG”)	Cell Signaling Technology	Cat#2368; RRID: AB_2217020
Mouse monoclonal anti-LRRC8A (clone 8H9)	Millipore Sigma	Cat#SAB1412855; RRID: AB_2864352
Rabbit polyclonal anti-LRRC8C	MyBioSource	Cat#MBS711316; RRID: AB_2864353
Goat anti-Mouse IgG (H+L) IRDye 680RD	LI-COR Biosciences	Cat#926-68070; RRID: AB_10956588
Goat anti-Rabbit IgG (H+L) IRDye 800CW	LI-COR Biosciences	Cat#925-32211; RRID: AB_2651127
Biological Samples
HUVEC, from pooled donors	Lonza	Cat#C2519A
Chemicals, Peptides, and Recombinant Proteins
2′3′-cGAMP (cGAMP)	This paper	N/A
2′3′-cGAM(PS)2 (Rp/Sp) (2′3′-cGSASMP)	Invivogen	Cat#tlrl-nacga2srs
2′3′-c-di-AMP (2′3′-CDA)	Invivogen	Cat#tlrl-nacda23
2′3′-c-di-AM(PS)2 (Rp,Rp) (2′3′-CDAS)	Invivogen	Cat#tlrl-nacda2r-01
3′3′-cGAMP	Invivogen	Cat#tlrl-nacga
c-di-AMP (3′3′-CDA)	Invivogen	Cat#tlrl-nacda
c-di-GMP (3′3′-CDG)	Invivogen	Cat#tlrl-nacdg
DMXAA	Invivogen	Cat#tlrl-dmx
Sphingosine-1-phosphate (d18:1)	Caymen Chemical	Cat#625701
DCPIB	Tocris	Cat#1540
Critical Commercial Assays
NextSeq 500/550 mid output v2 kit	Illumina	Cat#FC-404-2001
CellTiter-Glo One Solution Assay	Promega	Cat#G8461
Deposited Data
Uncropped Western blots	This paper; Mendeley data	http://dx.doi.org/10.17632/g7rwfvgjbm.1
Experimental Models: Cell Lines
U937	ATCC	Cat#CRL-1593.2; RRID: CVCL_0007
HEK-293	ATCC	Cat#CRL-1573; RRID: CVCL_0045
TIME	ATCC	Cat#CRL-4025; RRID: CVCL_0047
Oligonucleotides
See Table S1 for qPCR primers	Richtsteiger 2003	N/A
Also see Table S1 for all other oligos	This paper	N/A
Recombinant DNA
lentiCRISPR v2	Sanjana 2014	Addgene 52961; RRID: Addgene_52961
pHDM-G	Harvard PlasmID	EvNO00061606
pHDM-Hgpm2	Harvard PlasmID	EvNO00061607
pHDM-tat1b	Harvard PlasmID	EvNO00061608
pRC/CMV-rev1b	Harvard PlasmID	EvNO00061616
pDB-His-MBP-sscGAS	Ritchie 2019	N/A
pMCB320	Han 2017	RRID: Addgene_89359
pDB-His-MBP-sscGAS	Ritchie 2019	N/A
pLVX-TetOne-LRRC8A-FLAG-Hygro	This paper	N/A
pLVX-TetOne-LRRC8A-R103L-FLAG-Hygro	This paper	N/A
pLVX-TetOne-LRRC8C-FLAG-Hygro	This paper	N/A
Software and Algorithms
casTLE	Morgens et al 2016	https://bitbucket.org/dmorgens/castle
Inference of CRISPR Edits (ICE), v2	Synthego	https://ice.synthego.com/#/
ImageJ 1.51	Schneider 2012	https://imagej.nih.gov/ij/
Prism 8.4	Graph Pad	https://www.graphpad.com/scientific-software/prism/

Highlights:

• A CRISPR screen identifies human LRRC8A heteromeric channels as cGAMP transporters

• Expression of LRRC8A and LRRC8C/E promotes cGAMP transport while LRRC8D inhibits it

• Activating LRRC8A channels potentiates cGAMP transport

• Resting primary human vasculature cells use LRRC8A channels to import cGAMP