GPR35 (Ki = 12.8 nM)

The CID 2745687 (CID2745687) Ki for ERK1/2 phosphorylation using 1 μM Pamoic acid as the agonist is 18 nM[1]. In β-arrestin-2 interaction experiments, CID 2745687 (CID-2745687) is a strong antagonist limited to human GPR35 [2]. With an agonist concentration of 20 μM Zaprinast and the BRET-based GPR35-β-arrestin-2 interaction test, CID 2745687 exhibited moderate potency and concentration dependence at human GPR35, with a pIC50 of 6.70±0.09[2]. CID 2745687 (pIC50=6.27±0.08) completely counteracts Cromolyn disodium's agonistic effects [2].

---

Identification of CID2745687 as a GPR35 Antagonist with the Use of High Content Screening. [1]
There are no recognized GPR35 antagonists, but if such compounds were available, they would probably be useful as antitumor agents or as tool compounds in delineating the physiological role of GPR35 in animal models. To identify GPR35 antagonists, we assessed compounds for their ability to block 10 μM zaprinast-induced βarr2-GFP translocation. In a screen performed at the Sanford-Burnham Institute Molecular Libraries Probe Production Center, we assessed 293,187 compounds in an image-based high-content primary screen (PubChem AID 2508), and approximately 150 demonstrated low micromolar or better activities. Two structurally related hits, CID2745684 and CID2745687, are shown in Fig. 6A with the results of a confirmatory dose response performed down to 0.5 μM. The compounds' chemical identities and purities were confirmed using high-pressure liquid chromatography and mass spectrometry (the spectra are shown in Supplemental Fig. S3). Representative images from the high-throughput screen are shown in Supplemental Fig. S4.
The more potent of the two, CID2745687, was obtained as a dry powder, and its ability to antagonize GPR35 response was further evaluated against 1 μM pamoic acid. In Fig. 6B are representative images of βarr2-GFP recruitment in the presence of antagonist alone, pamoic acid alone, and pamoic acid in the presence of the antagonist where the inhibition of recruitment by CID2745687 is evident. In the βarr2-GFP trafficking assay (Fig. 6C, left), CID2745687 demonstrates a Ki of 12.8 nM (7.5–21.8) from three independent experiments. For ERK1/2 phosphorylation with 1 μM pamoic acid as the agonist (Fig. 6D, left), the CID2745687 Ki is 18 nM (9.1–35.7; n = 3). The antagonism of CID2745687 is also reversible (Fig. 6C, right) and competitive (Fig. 6D, right). To confirm that pamoic acid also activates GPR35 from other species, HEK293 cells were transiently cotransfected with plasmids for untagged mouse GPR35 and βarr2-GFP. Application of either pamoic acid (1 μM) or zaprinast (5 μM) induced trafficking of βarr2-GFP (Supplemental Fig. S5) was prevented by coincubation with CID2745687.

---

Antagonists of GPR35 Also Display Marked Species Ortholog Selectivity. [2]
Based on the marked species selectivity of certain agonist ligands, we next considered whether species selectivity might also be true of reported antagonists. CID2745687 (Fig. 1) is the only GPR35 antagonist detailed so far in the primary scientific literature (Zhao et al., 2010). Using the BRET-based GPR35-β-arrestin-2 interaction assay and an EC80 concentration of zaprinast (2 × 10−5 M) as agonist, CID-2745687 behaved as a moderately potent, concentration-dependent antagonist at human GPR35 with pIC50 = 6.70 ± 0.09 (mean ± S.E.M.; n = 9) (Fig. 5A). By contrast, even at 1 × 10−4 M CID-2745687 did not substantially block the agonist action of an EC80 concentration of zaprinast (4 × 10−6 M) at mouse GPR35 (Fig. 5A), and a similar inability to antagonize the effect of an EC80 concentration of zaprinast (4 × 10−7 M) was observed with rat GPR35 (Fig. 5A). This again was unexpected because Zhao et al. (2010) have reported CID-2745687 blocks the effects of zaprinast in a mouse pain model. Equivalent results were obtained with cromolyn disodium as agonist. Once again, a concentration-dependent inhibitory activity of CID-2745687 (pIC50 = 6.27 ± 0.08, mean ± S.E.M.; n = 9) to fully reverse the agonist action of an EC80 concentration of cromolyn disodium was observed only at the human ortholog (Fig. 5B). Although we were only able to explore the ability to inhibit the agonist action of pamoate at human GPR35, CID2745687 did so fully and in a concentration-dependent manner with pIC50 = 7.16 ± 0.12 (mean ± S.E.M.; n = 9) (Fig. 5C).

---

Both CID2745687 and ML-145 Can Also Prevent Agonist-Induced Internalization of Human GPR35. [2]
It is noteworthy that both CID-2745687 and ML-145 (1 × 10−5 M) also fully blocked internalization of human FLAG-GPR35-eYFP in response to varying concentrations of zaprinast, cromolyn disodium, and pamoate (Fig. 7, A–C). This was not the case for either CID2745687 or ML-145 when tested against zaprinast at rat FLAG-GPR35-eYFP (Fig. 7D). At mouse FLAG-GPR35-eYFP, although ML-145 was without effect on the potency or effect of zaprinast (Fig. 7E), at 1 × 10−5 M CID-2745687 consistently produced a modest, but not statistically significant, decrease in potency but not a maximal effect of zaprinast (Fig. 7E). To explore this more fully, the ability of varying concentrations of ML-145 or CID-2745687 to prevent internalization of the human or mouse orthologs of GPR35 in response to an EC80 concentration of zaprinast was assessed. Here, both ML-145 (Fig. 8A) and CID-2745687 (Fig. 8B) were effective inhibitors of zaprinast at the human ortholog but neither produced a substantial effect at the mouse ortholog (Fig. 8, C and D). Equivalent results for both ML-145 (Fig. 8, A, and C) and CID-2745687 (Fig. 8, B and D) were produced when EC80 concentrations of cromolyn disodium (at both human and mouse GPR35) or pamoate (at the human ortholog) were used.

---

CID2745687 Inhibits the Action of Agonists at Human GPR35 by Distinct Mechanisms. [2]
To assess whether CID-2745687 also acted as a competitive antagonist, the ability of a range of concentrations of this ligand to alter the concentration-response curves of zaprinast, cromolyn disodium, and pamoate was also examined at human GPR35. For both zaprinast (Fig. 11D) and cromolyn disodium (Fig. 11E) such studies produced surmountable shifts in the potency of the agonist to higher concentrations. Global-fit analyses of such curves were also consistent with a competitive mode of antagonism with pA2 affinity values for CID-2745687 of 7.7 to 7.8 (1.6–1.9 × 10−8 M) and slope factor close to 1.0. By contrast, equivalent studies with pamoate produced very different results (Fig. 11F). Increasing concentrations of CID-2745687 had little effect on the EC50 of pamoate but rather decreased the maximal effect of the agonist, which is indicative of a noncompetitive mode of action.
Consistent with the observations that both zaprinast and cromolyn disodium were antagonized competitively by CID2745687 and ML-145 and are full agonists in the β-arrestin-2 interaction assay, the addition of a series of submaximally effective concentrations of cromolyn disodium did not alter the observed EC50 of zaprinast at human GPR35 (Fig. 12), suggesting the two agonists share a common or overlapping binding site (Jenkins et al., 2011).

A particular GPR35 antagonist, CID-2745687 (CID2745687; 1 mg/kg; taken orally daily for the last 4 weeks), counteracts the anti-fibrotic effects of lodoxamide[3].

---

Protective effect of lodoxamide on CCl4-induced liver fibrosis [3]
Intraperitoneal injection of CCl4 twice a week for eight weeks was performed to induce liver fibrosis. Lodoxamide and/or CID2745687 were administrated orally every day for the last 4 weeks. Body and liver weights were unchanged after administration of CCl4, lodoxamide, and CID2745687 (Supplementary Fig. 1A, 1B). The ratio of liver to body weight was also unchanged (Supplementary Fig. 1C). Serum AST and ALT levels were elevated by CCl4 treatment, implying liver injury, although these levels were unchanged after lodoxamide and CID2745687 treatment (Fig. 1).

In order to measure histological changes induced by chronic liver fibrosis, the liver tissues were stained with H&E (Fig. 2). Compared to normal tissue (control), CCl4 induced severe injury, which appears as injured areas in the slide. Lodoxamide protected from the CCl4-induced injury and co-administration of CID2745687 inhibited lodoxamide-mediated protective effects (Fig. 2). Liver injury was semi-quantitatively evaluated using a subjective scale of 0–5, as previously described (Lim et al., 2016). Quantitative evaluation of liver injury confirmed the protective effect of lodoxamide and inhibition by CID2745687 (Fig. 2E).

In order to confirm liver fibrosis, Masson’s trichrome staining was also performed. As shown in Fig. 3, CCl4 induced fibrosis, as shown by blue-stained areas, while liver tissues from lodoxamide-treated mice exhibited less fibrosis. CID2745687 inhibited lodoxamide’s protective effects (Fig. 3). Degree of liver fibrosis was semi-quantitatively evaluated using a subjective scale of 0–5 in Ishak stage (Lim et al., 2016; Nallagangula et al., 2017). The result clearly showed the protective effect of lodoxamide and inhibition by CID2745687 (Fig. 3).

Next, changes in the mRNA levels of pro-fibrotic markers were measured in liver tissues. mRNA expression of the profibrotic markers, collagen Iα1, collagen Iα2, collagen IIIα1, αSMA, TIMP1, TGF-β1, and fibronectin in CCl4-treated liver tissues was measured by RT-PCR (Fig. 4A). mRNA levels of collagen Iα1, collagen Iα2, TIMP1, TGF-β1, and fibronectin were significantly elevated in livers of CCl4-treated mice, and CCl4-induced increase of collagen Iα1, collagen Iα2, and TGF-β1 was suppressed in lodoxamide-treated mice (Fig. 4B). CID2745687 co-treatment reversed the effects of lodoxamide treatment on TGF-β1 significantly, but not on others. Collagen IIIα1 and αSMA were unchanged by CCl4 treatment (Fig. 4A).

β-Arrestin Assay for Determining Receptor Responsiveness. [1]
U2OS cells transiently expressing human GPR35b and βarr2-GFP or HEK 293 cells transiently expressing mouse GPR35 and βarr2-GFP were used 48 h after transfection. U2OS cells permanently expressing HA-GPR35a and βarr2-GFP (UGPR35β) were used for most experiments. Cells were plated onto coverslips, placed in 24-well plates, and pretreated for 1 h with 0.02 mg/ml poly-d-lysine. Cells were maintained at 37°C in 5% CO2 until ready for experiments (80–85% confluent) and washed once with HBSS before drug application and experiments were performed in HBSS. Agonist-stimulated redistribution of βarr2-GFP was assessed after drug treatment for 40 min. Experiments involving antagonist were done with 15 min preincubation of antagonist for both the stable UGPR35β cells and the transiently transfected mouse GPR35 HEK293 cells. To examine reversibility of the antagonist, cells were preincubated with 100 nM CID2745687 for 10 min, then washed with HBSS five times for 5 min each before adding 1 μM pamoic acid. Cells were then fixed with 4% paraformaldehyde for 20 min at room temperature followed by three washes with HBSS. Glass coverslips were mounted on slides and were imaged on a fluorescence microscope using a 40× oil objective and 488 nm excitation for GFP.

---

In-Cell Western Assay for ERK activity. [1]
Cells were grown to confluence in 96-well plates and serum-starved overnight before assay. After drug treatment, the medium was removed, and 4% paraformaldehyde in PBS was added to fix cells for 20 min at room temperature. PTX, 200 ng/ml, was incubated with the cells for 3 h before drug treatment. The CID2745687 antagonist was coapplied with agonist. In the competitive antagonist assay, 300 nM CID2745687 was coapplied together with a serial dilution of pamoic acid. Cells were then permeabilized by 0.1% Triton X-100 in PBS for five washes, 5 min per wash. LI-COR blocking buffer was added, and samples were shaken on a rotator for 1 h. Primary antibodies against phospho-ERK1/2 (1:100) were applied overnight in a cold room, and then secondary antibodies goat anti-rabbit 800CW (1:800) were applied for 2 h at room temperature. Sapphire700 (1:1000; LI-COR) and DRAQ5 (1:2000) were added together with the secondary antibodies for normalization. The plate was dried and then scanned using a LI-COR Odyssey IR Imager set at 169 μM resolution, 3 focus offset, and 4.5 to 6 intensity.

Animal/Disease Models: Sixweeks old male C57BL/6 mice[3]
Doses: 1 mg/kg
Route of Administration: Oral administration, every day for 4 weeks
Experimental Results: Inhibited Lodoxamide-mediated protective effects.

---

Induction of hepatic fibrosis in C57BL/6 mice [3]
Six-week-old male C57BL/6 mice were housed under standard laboratory conditions (22°C ± 2°C, 12-h light/dark cycles) with free access to food and water in the laboratory animal facility at PNU. In this study, seven-week-old male C57BL/6 mice were randomly divided into 4 groups : control (n=5), in which mice were intraperitoneal (i.p.) injected a vehicle for 8 weeks; CCl4 (n=5), in which mice were i.p. injected a CCl4 (5 ml/kg, CCl4:con oil=2:8) twice a week for 8 weeks; CCl4 plus lodoxamide (n=5), in which mice were i.p. injected a CCl4 two times a week for 8 weeks and oral administration with lodoxamide (1 mg/kg) every day of the last 4 weeks; and CCl4 plus lodoxamide and CID2745687 (n=5), in which mice were i.p. injected a CCl4 two times a week for 8 weeks, and oral administration of lodoxamide (1 mg/kg) and CID2745687 (1 mg/kg) every day of the last 4 weeks.

[1]. Targeting of the orphan receptor GPR35 by pamoic acid: a potent activator of extracellular signal-regulated kinase and β-arrestin2 with antinociceptive activity. Mol Pharmacol. 2010 Oct;78(4):560-8.

[2]. Antagonists of GPR35 display high species ortholog selectivity and varying modes of action. J Pharmacol Exp Ther. 2012 Dec;343(3):683-95.

[3]. Lodoxamide Attenuates Hepatic Fibrosis in Mice: Involvement of GPR35. Biomol Ther (Seoul). 2019 Jun 13;28(1):92-97.

5-[[[(tert-butylamino)-sulfanylidenemethyl]hydrazinylidene]methyl]-1-(2,4-difluorophenyl)-4-pyrazolecarboxylic acid methyl ester is a ring assembly and a member of pyrazoles.

---

Known agonists of the orphan receptor GPR35 are kynurenic acid, zaprinast, 5-nitro-2-(3-phenylproplyamino) benzoic acid, and lysophosphatidic acids. Their relatively low affinities for GPR35 and prominent off-target effects at other pathways, however, diminish their utility for understanding GPR35 signaling and for identifying potential therapeutic uses of GPR35. In a screen of the Prestwick Library of drugs and drug-like compounds, we have found that pamoic acid is a potent GPR35 agonist. Pamoic acid is considered by the Food and Drug Administration as an inactive compound that enables long-acting formulations of numerous drugs, such as the antihelminthics oxantel pamoate and pyrantel pamoate; the psychoactive compounds hydroxyzine pamoate (Vistaril) and imipramine pamoate (Tofranil-PM); and the peptide hormones triptorelin pamoate (Trelstar) and octreotide pamoate (OncoLar). We have found that pamoic acid induces a G(i/o)-linked, GPR35-mediated increase in the phosphorylation of extracellular signal-regulated kinase 1/2, recruitment of β-arrestin2 to GPR35, and internalization of GPR35. In mice, it attenuates visceral pain perception, indicating an antinociceptive effect, possibly through GPR35 receptors. We have also identified in collaboration with the Sanford-Burnham Institute Molecular Libraries Probe Production Center new classes of GPR35 antagonist compounds, including the nanomolar potency antagonist methyl-5-[(tert-butylcarbamothioylhydrazinylidene)methyl]-1-(2,4-difluorophenyl)pyrazole-4-carboxylate (CID2745687). Pamoic acid and potent antagonists such as CID2745687 present novel opportunities for expanding the chemical space of GPR35, elucidating GPR35 pharmacology, and stimulating GPR35-associated drug development. Our results indicate that the unexpected biological functions of pamoic acid may yield potential new uses for a common drug constituent. [1]

---

Variation in pharmacology and function of ligands at species orthologs can be a confounding feature in understanding the biology and role of poorly characterized receptors. Substantial selectivity in potency of a number of GPR35 agonists has previously been demonstrated between human and rat orthologs of this G protein-coupled receptor. Via a bioluminescence resonance energy transfer-based assay of induced interactions between GPR35 and β-arrestin-2, addition of the mouse ortholog to such studies indicated that, as for the rat ortholog, murine GPR35 displayed very low potency for pamoate, whereas potency for the reference GPR35 agonist zaprinast was intermediate between the rat and human orthologs. This pattern was replicated in receptor internalization and G protein activation assays. The effectiveness and mode of action of two recently reported GPR35 antagonists, methyl-5-[(tert-butylcarbamothioylhydrazinylidene)methyl]-1-(2,4-difluorophenyl)pyrazole-4-carboxylate (CID2745687) and 2-hydroxy-4-[4-(5Z)-5-[(E)-2-methyl-3-phenylprop-2-enylidene]-4-oxo-2-sulfanylidene-1,3-thiazolidin-3-yl]butanoylamino)benzoic acid (ML-145), were investigated. Both CID-2745687 and ML-145 competitively inhibited the effects at human GPR35 of cromolyn disodium and zaprinast, two agonists that share an overlapping binding site. By contrast, although ML-145 also competitively antagonized the effects of pamoate, CID-2745687 acted in a noncompetitive fashion. Neither ML-145 nor CID-2745687 was able to effectively antagonize the agonist effects of either zaprinast or cromolyn disodium at either rodent ortholog of GPR35. These studies demonstrate that marked species selectivity of ligands at GPR35 is not restricted to agonists and considerable care is required to select appropriate ligands to explore the function of GPR35 in nonhuman cells and tissues. [2]

---

Although it is also possible that nonreceptor accessory proteins might alter the pharmacology of GPR35 in a species-dependent manner, the current studies re-emphasize the need to perform standard but insightful pharmacological analyses to fully understand both the potential and the potential limitations of novel ligands identified via various screens and their detailed mode of action. The current studies define that, despite their high affinity at human GPR35, neither CID2745687 nor ML-145 are useful pharmacological antagonists for probing the functions of GPR35 in either mouse or rat models of physiology and disease, and pamoate is not a high-potency agonist at rodent orthologs of this receptor.[2]

---

A previous pharmacogenomic analysis identified cromolyn, an anti-allergic drug, as an effective anti-fibrotic agent that acts on hepatocytes and stellate cells. Furthermore, cromolyn was shown to be a G protein-coupled receptor 35 (GPR35) agonist. However, it has not been studied whether anti-fibrotic effects are mediated by GPR35. Therefore, in this study, the role of GPR35 in hepatic fibrosis was investigated through the use of lodoxamide, another anti-allergic drug and a potent GPR35 agonist. Longterm treatment with carbon tetrachloride induced hepatic fibrosis, which was inhibited by treatment with lodoxamide. Furthermore, CID2745687, a specific GPR35 antagonist, reversed lodoxamide-mediated anti-fibrotic effects. In addition, lodoxamide treatment showed significant effects on the mRNA expression of collagen Iα1, collagen Iα2, and TGF-β1 in the extracellular matrix. However, a transforming growth factor α (TGF-α) shedding assay revealed lodoxamide not to be a potent agonist of mouse GPR35 in vitro. Therefore, these results showed anti-fibrotic effects of lodoxamide in mice and raise concerns how lodoxamide protects against liver fibrosis in vivo and whether GPR35 is involved in the action.[3]

---

CID2745687 was reported as a potent and effective antagonist on human GPR35 but not of rodent GPR35 orthologs (Jenkins et al., 2012). However, in other reports CID274568 was reported to inhibit agonist-induced activation of mouse GPR35 in HEK293 cells, mouse astrocytes, and mouse colon epithelial cells (Zhao et al., 2010; Berlinguer-Palmini et al., 2013; Tsukahara et al., 2017). Although inhibitory effect of CID2745687 on lodoxamide action was observed in mice. we cannot exclude the possibility of off-target effects of CID2745687. This may suggest that lodoxamide and CID2745687 act as an agonist and an antagonist, respectively, of the same unknown target in mice. In our previous study on hepatocytes, CID2745687 reversed the effects of lodoxamide when administrated at a dose of 10 times greater (10 mg/kg) than that of lodoxamide (1 mg/kg) in vivo (Nam et al., 2019). However, in the present study, the same dose of CID2745687 effectively reversed the effects of lodoxamide, suggesting different affinities between the two experimental models of hepatic steatosis and hepatic fibrosis. Further investigation is necessary to elucidate the different efficacies of CID2745687. In summary, lodoxamide attenuates CCl4-induced liver fibrosis in mice and CID2745687 reverses the lodoxamide’s effect. However, involvement of mouse GPR35 in the effects is questioned in AP-TGF-α shedding assay. Therefore, we reported the results to accelerate drug development for liver fibrosis and raise concerns how lodoxamide protects against liver fibrosis in vivo and whether GPR35 is involved in the action. [3]

C17H19F2N5O2S

395.43

395.122

C, 51.64; H, 4.84; F, 9.61; N, 17.71; O, 8.09; S, 8.11

264233-05-8

9581011

Off-white to light yellow solid powder

1.3±0.1 g/cm3

493.6±55.0 °C at 760 mmHg

252.3±31.5 °C

0.0±1.3 mmHg at 25°C

1.591

3.05

2

7

6

27

572

0

CC(C)(C)NC(=S)N/N=C/C1=C(C=NN1C2=C(C=C(C=C2)F)F)C(=O)OC

CYNLZIBKERMMOA-AWQFTUOYSA-N

InChI=1S/C17H19F2N5O2S/c1-17(2,3)22-16(27)23-20-9-14-11(15(25)26-4)8-21-24(14)13-6-5-10(18)7-12(13)19/h5-9H,1-4H3,(H2,22,23,27)/b20-9+

methyl 5-[(E)-(tert-butylcarbamothioylhydrazinylidene)methyl]-1-(2,4-difluorophenyl)pyrazole-4-carboxylate

264233-05-8; CID 2745687; ML194; MLS000834953; methyl 5-[(E)-(tert-butylcarbamothioylhydrazinylidene)methyl]-1-(2,4-difluorophenyl)pyrazole-4-carboxylate; CHEMBL1708510; CID 27456; SMR000461569;

2934.99.9001

Powder      -20°C    3 years

                     4°C     2 years

In solvent   -80°C    6 months

                  -20°C    1 month

Room temperature (This product is stable at ambient temperature for a few days during ordinary shipping and time spent in Customs)

DMSO: 125 mg/mL (316.11 mM)

Solubility in Formulation 1: ≥ 2.08 mg/mL (5.26 mM) (saturation unknown) in 10% DMSO + 90% Corn Oil (add these co-solvents sequentially from left to right, and one by one), clear solution.
For example, if 1 mL of working solution is to be prepared, you can add 100 μL of 20.8 mg/mL clear DMSO stock solution to 900 μL of corn oil and mix evenly.

---

 (Please use freshly prepared in vivo formulations for optimal results.)