GPR35 agonists inhibit TRPA1-mediated colonic nociception through suppression of substance P release

Abstract

The development of nonopioid analgesics for the treatment of abdominal pain is a pressing clinical problem. To address this, we examined the expression of G i/o -coupled receptors, which typically inhibit nociceptor activation, in colonic sensory neurons. This led to the identification of the orphan receptor GPR35 as a visceral analgesic drug target because of its marked coexpression with transient receptor potential ankyrin 1 (TRPA1), a mediator of noxious mechanotransduction in the bowel. Building on in silico docking simulations, we confirmed that the mast cell stabiliser, cromolyn (CS), and phosphodiesterase inhibitor, zaprinast, are agonists at mouse GPR35, promoting the activation of different G i/o subunits. Pretreatment with either CS or zaprinast significantly attenuated TRPA1-mediated colonic nociceptor activation and prevented TRPA1-mediated mechanosensitisation. These effects were lost in tissue from GPR35 -/- mice and were shown to be mediated by inhibition of TRPA1-evoked substance P (SP) release. This observation highlights the pronociceptive effect of SP and its contribution to TRPA1-mediated colonic nociceptor activation and sensitisation. Consistent with this mechanism of action, we confirmed that TRPA1-mediated colonic contractions evoked by SP release were abolished by CS pretreatment in a GPR35-dependent manner. Our data demonstrate that GPR35 agonists prevent the activation and sensitisation of colonic nociceptors through the inhibition of TRPA1-mediated SP release. These findings highlight the potential of GPR35 agonists to deliver nonopioid analgesia for the treatment of abdominal pain.

Figure 1.

Expression of GPR35 in colonic sensory neurons. (A) Coexpression of transcripts encoding selected G_i/o-coupled receptors with Trpa1 in colonic sensory neurons. Inset: enlargement of top-right cluster of data points. Data in (A-D) redrawn from Hockley et al., 2019. (B) Expression (log [transcripts per million]) of transcripts encoding Trpa1 and Gpr35 in colonic sensory neurons. (C) Expression of transcripts encoding Trpa1 and Gpr35 in each subpopulation of colonic sensory neuron. (D) Coexpression of transcripts encoding Trpa1 and Gpr35 in colonic sensory neurons. m, mixed lumbar splanchnic and pelvic afferents; NF, neurofilament-expressing; NP, nonpeptidergic; p, pelvic afferents; PEP, peptidergic.

Figure 2.

Cromolyn and zaprinast are agonists of GPR35. (A) (i) Predicted binding mode of cromolyn at mouse GPR35 based on the 10 best ranked binding poses obtained through independent docking runs in GOLD v.5.3. (ii) Ligand interaction diagram generated using PoseView. The most representative pose for cromolyn is shown. Dotted black lines indicate hydrogen bonding, solid green lines represent hydrophobic interactions, and dotted green lines represent π-π stacking. (B) (i) Predicted binding mode of zaprinast at mouse GPR35 based on the 10 best ranked binding poses obtained through independent docking runs in GOLD v.5.3. (ii) Ligand interaction diagram generated using PoseView. The most representative pose for zaprinast is shown. Dotted black lines indicate hydrogen bonding, solid green lines represent hydrophobic interactions, and dotted green lines represent π-π stacking. (C-H) Dissociation of given G_i/o subunits in response to stimulation of human GPR35 with CS (red) or Zap (blue). Filled symbols show results from cells transfected with GPR35; open symbols show mock-transfected cells. Points are plotted as mean ± SEM and fit with a three-parameter log [agonist]-response Boltzmann curve. CS, cromolyn.

Figure 3.

TRPA1-mediated colonic afferent firing is attenuated by activation of GPR35. (A) Example rate histogram showing the change in afferent firing evoked by the application of ASP7663 (grey shaded region). (B) Example rate histogram showing the change in afferent firing evoked by the application of ASP7663 (grey shaded region) in the presence of cromolyn (CS, 100 µM, red bar). (C) Example rate histogram showing the change in afferent firing evoked by the application of ASP7663 (grey shaded region) in the presence of zaprinast (Zap, 100 µM, blue bar). (D) Grouped data showing the mean change in afferent firing evoked by ASP7336 (points indicate the mean; shaded regions show the standard error) in the presence of either CS (100 µM, red) or Zap (100 µM, blue). (E) Grouped data showing the peak change in afferent firing evoked by ASP7336 in the presence of either CS (100 µM, red) or Zap (100 µM, blue) in both wildtype (WT) and GPR35^−/− animals. Data for each genotype analysed using a one-way ANOVA with Dunnett post-hoc tests. TRPA1, transient receptor potential ankyrin 1.

Figure 4.

TRPA1 contributes to colonic afferent mechanosensitivity. (A) Example recording showing afferent firing during 5 successive ramp distensions with AM0902 (1 mM) applied before and during the third ramp distension. (B) Grouped data showing the percentage change in afferent firing during the third ramp distension in vehicle- (black), AM0902- (grey), CS- (red), and Zap- (blue) treated tissue. Two-way repeated measures ANOVA with Dunnett post-hoc tests. The percentage change in firing is relative to peak firing during the preceding ramp distension. (C) Grouped data showing the peak change in afferent firing during the third ramp distension in vehicle-, AM0902-, CS-, and Zap-treated tissue. One-way ANOVA with Dunnett post-hoc tests. (D) Example recording showing afferent firing during 3 consecutive ramp distensions with CS applied before and during the third ramp distension. (E) Example recording showing afferent firing during 3 consecutive ramp distensions with Zap applied before and during the third ramp distension. CS, cromolyn; TRPA1, transient receptor potential ankyrin 1.

Figure 5.

GPR35 activation inhibits TRPA1-induced mechanical hypersensitivity. (A) (i) Example recording showing the sensitisation of the afferent response to ramp distension of the colon by ASP7663 in tissue from a wildtype mouse. (ii) Grouped data showing the change in afferent firing during ramp distension of the colon before and after ASP7663 application. Two-way repeated measures ANOVA. (iii) Grouped data showing the peak change in afferent firing during ramp distension of the colon before and after ASP7663 application. Two-tailed ratio-paired t test. (B) (i) Example recording showing the sensitisation of the afferent response to ramp distension of the colon by ASP7663 in tissue from a GPR35^−/− mouse. (ii) Grouped data showing the change in afferent firing during ramp distension of the colon before and after ASP7663 application. Two-way repeated measures ANOVA. (iii) Grouped data showing the peak change in afferent firing during ramp distension of the colon before and after ASP7663 application. Two-tailed ratio-paired t test. (C) Example recording from tissue from a wildtype mouse showing the inhibition of ASP7663-induced afferent firing and sensitisation of the response to ramp distension by pretreatment with CS. (D) Example recording from tissue from a GPR35^−/− mouse showing the loss of the inhibitory effect of CS on ASP7663-induced afferent firing and sensitisation of the response to ramp distension. (E) Grouped data showing the change in afferent firing during the third (post-ASP7663) ramp distension in tissue from wildtype mice treated with ASP7663 with or without CS. Two-way repeated measures ANOVA with Holm–Sidak post-hoc tests. (F) Grouped data showing the change in afferent firing during the third (post-ASP7663) ramp distension in tissue from GPR35^−/− mice treated with ASP7663 with or without CS. Two-way repeated measures ANOVA with Holm–Sidak post-hoc tests. (G) Grouped data showing the change in afferent firing during the third (post-ASP7663) ramp distension in tissue from wildtype mice treated with ASP7663 with or without Zap. Two-way repeated measures ANOVA with Holm–Sidak post-hoc tests. (H) Grouped data showing the peak change in afferent firing during the third (postvehicle or -ASP7663) ramp distension. One-way ANOVA with Bonferroni post-hoc tests. CS, cromolyn; TRPA1, transient receptor potential ankyrin 1.

Figure 6.

SP stimulates colonic afferents and cultured sensory neurons. (A) Coexpression of Trpa1 and Tac1 transcripts in colonic sensory neurons. Data redrawn from Hockley et al., 2019. (B) Grouped data showing the change in afferent firing rate after the application of ASP7663 (30 µM) alone or after tissue pretreatment with aprepitant (10 µM). (C) Grouped data showing the peak change in afferent firing rate after the application of ASP7663 (30 µM) alone or after tissue pretreatment with aprepitant (10 µM). Two-tailed unpaired t test. (D) Grouped data showing the change in afferent firing rate after the application of 3, 9, or 15 µM of SP. (E) Grouped data showing the change in afferent firing rate after the application of SP with thiorphan (TH) and captopril (Capt) in the absence and presence of aprepitant. (F) Grouped data showing the peak change in afferent firing rate from the experiments shown in (D and E). One-way ANOVA with Bonferroni post-hoc tests. (G) Example Fluo-4 fluorescence traces showing 3 distinct response profiles after the application of SP and capsaicin. (H) (Left) Grouped data showing the proportion of neurons which responded to SP alone, capsaicin alone, or to both SP and capsaicin. (Right) Grouped data showing the proportion of SP-sensitive neurons which were cosensitive to capsaicin. (I) Histogram showing the distribution of sensory neuron soma size for SP-sensitive (purple) and SP-insensitive (grey) neurons. Median soma areas compared using a Mann–Whitney U test. SP, substance P.

Figure 7.

Substance P signalling is required for ASP7663-induced mechanical hypersensitivity. (A) Example recording showing afferent firing during 3 consecutive ramp distensions of the colon, with SP, thiorphan, and captopril applied between the second and third ramps. (B) Grouped data showing the percentage change in afferent firing during the third ramp distension after pretreatment with either thiorphan and captopril alone or thiorphan and captopril with 15 µM of SP. Two-way repeated measures ANOVA with Holm–Sidak post-hoc tests. (C) Grouped data showing the peak percentage change in afferent firing from the experiments shown in (A and B). Two-tailed unpaired t test. (D) Example recording showing afferent firing during 3 consecutive ramp distensions of the colon with aprepitant and ASP7663 applied between the second and third ramps. (E) Grouped data showing the percentage change in afferent firing during the third ramp distension after pretreatment with ASP7663 with or without aprepitant. Two-way repeated measures ANOVA with Holm–Sidak post-hoc tests. (F) Grouped data showing the peak percentage change in afferent firing from the experiments shown in (D and E). Two-tailed unpaired t test. (G) Example recording showing afferent firing during 3 ramp distensions with aprepitant alone applied between the second and third ramps. (H) Grouped data showing the change in afferent firing rate during the second (preaprepitant) and third (postaprepitant) ramp distensions. Two-way repeated measures ANOVA with Holm–Sidak post-hoc tests. (I) Grouped data showing the peak change in afferent firing rate during the second (preaprepitant) and third (postaprepitant) ramp distensions. Two-tailed paired t test. SP, substance P.

Figure 8.

TRPA1 activation evokes SP release in the colon. (A) Schematic showing the experimental protocol for measuring SP release in the colon. (B) Grouped data showing the fold change in SP (relative to a tissue-matched unstimulated control experiment) after incubation of tissue with ASP7663 alone or in the presence of AM0902 or CS in wildtype and GPR35^−/− tissue. One-way ANOVA with Bonferroni post-hoc tests. (C) Example recordings showing the reduction in colon length by ASP7663 (shown as a reduction in voltage across a transducer attached to one end of the colon; voltages are shown relative to that evoked by 10 µM of ACh). The effect of ASP7663 was blocked by both AM0902 and aprepitant. (D) Example recordings showing the effect of CS pretreatment on ASP7663-induced colonic contraction in tissue from wildtype and GPR35^−/− animals. CS failed to inhibit ASP7663-induced contraction in tissue lacking GPR35. (E) Grouped data showing the peak percentage change in transducer voltage (relative to ACh) for the experiments shown in (C and D). One-way ANOVA with Bonferroni post-hoc tests. CS, cromolyn; SP, substance P; TRPA1, transient receptor potential ankyrin 1.

Figure 9.

CS inhibits ASP7663-induced afferent activity and mechanical hypersensitivity in tissue from female mice. (A) Grouped data showing the change in afferent firing after the application of ASP7663 (30 µM) to tissue preincubated with vehicle (black) or CS (100 µM, pink). Solid points show the mean change in afferent firing; shaded regions show SEM. (B) Grouped data showing the change in afferent firing evoked by ASP7663 in vehicle- and CS-treated tissue. Data analysed using a two-tailed Mann–Whitney U test. (C) Grouped data showing the peak change in afferent firing during the third (post-ASP7663) ramp distension in tissue treated with ASP7663 with or without CS. Two-way repeated measures ANOVA. (D) Grouped data showing the peak percentage change in afferent firing during the third (post-ASP7333) ramp distension in tissue treated with ASP7663 alone (green) or ASP7663 and CS (red). Two-tailed unpaired t test. CS, cromolyn.

Figure 10.

Prolonged colonic distension evoked SP-dependent colonic afferent activity. (A) Example rate histogram (top) and colonic luminal pressure trace (bottom) showing the afferent response to slow ramp distension to 40 mm Hg (Dist. 2) and 120 mm Hg (Dist. 3) in the presence of vehicle (1:1000 DMSO). (B) Example rate histogram (top) and colonic luminal pressure trace (bottom) showing the afferent response to slow ramp distension to 40 mm Hg (Dist. 2) and 120 mm Hg (Dist. 3) in the presence of aprepitant (10 µM). (C) Example rate histogram (top) and colonic luminal pressure trace (bottom) showing the afferent response to slow ramp distension to 40 mm Hg (Dist. 2) and 120 mm Hg (Dist. 3) in the presence of CS (100 µM). (D) Grouped data showing the percentage change in afferent activity during the third ramp distension (to 120 mm Hg) in the presence of vehicle (black), aprepitant (pink), and CS (red). Two-way repeated measures ANOVA with Holm–Sidak post-hoc tests. (E) Grouped data showing the area under the pressure-response curves (AUC) in (D). One-way ANOVA with Dunnett post-hoc tests. CS, cromolyn; SP, substance P.