Vanderbilt Institute of Chemical Biology

 

 

Discovery at the VICB

 

 

 

 

 

 

Targeting cGAS in Autoimmunity


By: Carol A. Rouzer, VICB Communications
Published:  October 9, 2017

 

 

New small molecule inhibitors of cyclic GMP-AMP synthase will serve as probes of the enzyme's role in autoimmunity and inflammation.  

 

Cells of the innate immune system express multiple receptors that recognize molecules associated with pathogens or damaged host tissue. One such molecule is cytosolic double-stranded DNA (dsDNA), an indicator of invasion by an intracellular pathogen (virus or bacteria) or of damage to the cell leading to displacement of dsDNA from the nucleus or mitochondria. An important receptor for cytosolic dsDNA is cyclic GMP-AMP synthase (cGAS). Binding of dsDNA to cGAS activates it to catalyze the production of 2´,3´-cyclic GAMP (cGAMP, c[G(2´,5´)pA(3´,5´)p] from GTP and ATP (Figure 1). The only known cyclic dinucleotide in metazoan organisms, cGAMP serves as a second messenger by binding to STimulator of INterferon Genes (STING), an endoplasmic reticulum-associated receptor protein. STING then activates Interferon Regulatory Factor 3 (IRF3), a transcription factor that promotes the biosynthesis of Interferon-β1 (IFNB1) and other cytokines.

 

The cGAS pathway plays an important role in the immune response to intracellular pathogens, and a cGAMP analog (ADU-S100) is currently being tested in clinical trials for anti-tumor activity. However, this pathway can also contribute to the pathogenesis of autoimmunity. A clear example is found in autoimmune disorders associated with an inactivating mutation in the gene encoding TREX1, the enzyme primarily responsible for degrading cytosolic dsDNA. Such mutations are found in patients suffering from Aicardi-Goutières syndrome and Chilblain lupus, diseases that bear some similarities to the more common systemic lupus erythematosis. Genetic knockout of cGAS, STING, or IRF3 production completely alleviates the symptoms of these diseases in mouse models of Trex1 deficiency. These findings indicate that cGAS is a key player in both positive and negative aspects of innate immunity, leading Vanderbilt Institute of Chemical Biology member Manuel Ascano, his collaborators Dinshaw Patel (Memorial Sloan-Kettering Cancer Center) and Fraser Glickman (The Rockefeller University), and their laboratories to develop small molecule cGAS inhibitors to probe the enzyme's function (J. Vincent et al. Nat. Commun., 2013, 8:750 DOI: 10.1038/s41467-017-00833-9).

 

 

 

FIGURE 1. Diagrammatic representation of the cGAS signaling pathway. DNA in the cytosol binds to cGAS, activating it to produce cGAMP from ATP and GTP. cGAMP binds to STING, leading to activation of IRB3 which triggers IFNB1 synthesis and release. Figure reproduced by permission from Macmillan, Ltd. from C. Cain. SciBX., 6(40), DOI 10.1038/scibx.2013.1117. Copyright 2013.

 

 

 

The investigators began their efforts by developing a RapidFire mass spectrometry-based high-throughput screen for cGAMP activity. They used the screen to evaluate a 123,306-member compound library, identifying 49 molecules of interest. Further screening led to the selection of 4 compounds that exhibited acceptable stability and structural diversity and inhibited cGAS with IC50 (the concentration that causes 50% inhibition) values of 0.03 to 1.9 μM.

 

Crystal structure data for one of the compounds, RU.365 (Figure 2), in complex with cGAS and dsDNA (Figure 3a) enabled the researchers to investigate its mode of interaction with the protein. They found that RU.365 bound in the active site pocket of the enzyme but filled only a portion of the pocket (Figure 3b) similar to cGAMP. Surrounded primarily by hydrophobic and aromatic amino acids, RU.365 formed key stacking interactions that placed its benzimidazole and part of its pyrazole ring between the guanidinium group of Arg-364 and the phenol ring of Tyr-421 (Figure 3c). Similar stacking occurs with the substrate ATP, 5´pppGpg (an analog of the reaction intermediate), and the product cGAMP. These interactions are likely of functional significance, as a double mutation of Arg-364 and Tyr-421 renders cGAS unable to respond to dsDNA with cGAMP synthesis. The investigators noted a surprising absence of hydrogen bond formation between RU.365 and cGAS despite the presence of multiple potential donors and acceptors in the inhibitor molecule. The crystal structure of a second inhibitor, RU.332 (Figure 2) in complex with cGAS and dsDNA revealed an almost identical binding interaction as that for RU.365. This was not surprising, as the only difference between the two molecules was the substitution of a benzothiazole ring in RU.332 for the benzimidazole in RU.365.


 

 

 

FIGURE 2. Structures of RU.365, RU.332, and RU.521 as labeled.

 

 

 

FIGURE 3. (A). Representation of the crystal structure of the ternary complex of cGAS (violet), dsDNA (light orange), and RU.365 (yellow). (B) Space-filling close-up representation of RU.365 (yellow) in the cGAS (violet) pocket. The full extent of the pocket is outlined in black dashed lines. (c) Close-up representation of the amino acids (violet) surrounding RU.365 (yellow) in the cGAS pocket. Figure reproduced under the Creative Commons Attribution 4.0 International License from J. Vincent et al. Nat. Commun., 2013, 8:750 DOI: 10.1038/s41467-017-00833-9).

 

 

To obtain inhibitors of higher potency, the researchers embarked on a structure-guided medicinal chemistry effort. The result was the discovery of RU.521 (Figure 2), an analog of RU.365 bearing dichloro substitution on the benzimidazole ring. RU.521 (IC50 = 0.11 μM) was substantially more potent than either RU.365 (IC50 = 1.9 μM) or RU.322 (IC50 = 1.8 μM). The crystal structure of this inhibitor in complex with cGAS and dsDNA revealed that the chlorine atoms burrowed more deeply into the active site pocket of the enzyme, displacing the benzimidazole ring outward and increasing its stacking interaction with Arg-364 and Tyr-421. In addition, RU.521 established a water-mediated hydrogen bonding interaction between the carbonyl oxygen of its phthalide ring and the backbone of Gly-293 and side chain of Lys-360 in the enzyme's active site pocket. Thus, the structural data could explain the increased potency of RU.521 as compared to RU.365.

 

Superposition of the structures of the inhibitors in complex with the enzyme over the structure of inactive cGAS (lacking bound dsDNA) revealed that all three should be able to bind well in the much smaller pocket of the inactive enzyme. The researchers confirmed this observation in the case of RU.365 by performing isothermal calorimetry to reveal similar binding affinities to cGAS in the presence or absence of dsDNA (Kd = 64 nM and 98 nM, respectively). Isothermal calorimetry also confirmed a higher binding affinity of RU.521 (Kd = 35 nM) as compared to that of RU.365 with dsDNA-bound cGAS.

 

To better understand their mechanism of inhibition, the investigators superimposed the crystal structures of RU.365 or RU.521 complexed with cGAS and dsDNA over those of the enzyme and dsDNA complexed to GTP, 5´pppGpG, or cGAMP. In every case, the inhibitor clashed severely with the alternative ligand in the enzyme's active site pocket (Figure 4). These results indicated that the inhibitors could conflict with enzyme activity at any point along the reaction pathway. Isothermal calorimetry confirmed that the presence of inhibitor decreases the enzyme's affinity for GTP or ATP, suggesting the possibility of competitive inhibition. However, kinetic studies revealed a decrease in kcat but no change in KM for the reaction in the presence of inhibitor, a finding consistent with noncompetitive rather than competitive inhibition.

 

 

Figure 4. Superposition of the structure of RU.521 with that of ATP (left), 5´pppGpG (a transition state analog, center) and cGMP (right) in complex with cGAS and dsDNA. Figure reproduced under the Creative Commons Attribution 4.0 International License from J. Vincent et al. Nat. Commun., 2013, 8:750 DOI: 10.1038/s41467-017-00833-9).

 

 

The RAW264.7 (RAW) murine macrophage-like cell line responds to the introduction of dsDNA with a strong upregulation in IFNB1 production, suggesting the presence of a robust cGAS-dependent signaling pathway. The researchers used RAW cells expressing an interferon-responsive element coupled to a luciferase gene to study the effects of the inhibitors on the cells' response to dsDNA. The compounds tested were active, exhibiting IC50 values in the range of 700 nM (RU.521) to 13.6 μM (RU.365). The investigators also explored the ability of several of the inhibitors to directly or indirectly affect other inflammatory signaling pathways. The results showed that RU.521 was the most specific of all compounds tested, having no effects outside of cGAS-dependent signaling suppression.

 

Mice genetically lacking the gene that encodes Trex1 serve as a model for Aicardi-Goutières syndrome and Chilblain lupus. The investigators showed that RU.521 reduces the production of IFNB1 in bone marrow-derived macrophages from these mice. These findings demonstrate the ability of the inhibitor to block cGAS signaling that results from naturally acquired aberrantly localized dsDNA.

 

The results confirm RU.521 as a potent and selective inhibitor of cGAS that is active both in vitro and in cells. Although it is possible that the compound exerts off-target effects that are currently unknown, the data presented here indicate that it should be a valuable probe with which to explore the role of cGAS in the response of the innate immune system in a wide variety of circumstances including invasion by an intracellular pathogen, development of some types of cancer, and a range of autoimmune disorders.

 

 

ViewNature Communications article: Small molecule inhibition of cGAS reduces interferon expression in primary macrophages from autoimmune mice

 

TwitterYouTube

 

 

The Vanderbilt Institute of Chemical Biology, 896 Preston Building, Nashville, TN 37232-6304, phone 866.303 VICB (8422), fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2014 by Vanderbilt University Medical Center

  • Areas of Interest