The TRPA1 Agonist Cinnamaldehyde Induces the Secretion of HCO3- by the Porcine Colon

Abstract

A therapeutic potential of the TRPA1 channel agonist cinnamaldehyde for use in inflammatory bowel disease is emerging, but the mechanisms are unclear. Semi-quantitative qPCR of various parts of the porcine gastrointestinal tract showed that mRNA for TRPA1 was highest in the colonic mucosa. In Ussing chambers, 1 mmol·L^-1 cinnamaldehyde induced increases in short circuit current (ΔI_sc) and conductance (ΔG_t) across the colon that were higher than those across the jejunum or after 1 mmol·L^-1 thymol. Lidocaine, amiloride or bumetanide did not change the response. The application of 1 mmol·L^-1 quinidine or the bilateral replacement of 120 Na⁺, 120 Cl^- or 25 HCO₃^- reduced ΔG_t, while the removal of Ca²⁺ enhanced ΔG_t with ΔI_sc numerically higher. ΔI_sc decreased after 0.5 NPPB, 0.01 indometacin and the bilateral replacement of 120 Na⁺ or 25 HCO₃^-. The removal of 120 Cl^- had no effect. Cinnamaldehyde also activates TRPV3, but comparative measurements involving patch clamp experiments on overexpressing cells demonstrated that much higher concentrations are required. We suggest that cinnamaldehyde stimulates the secretion of HCO₃^- via apical CFTR and basolateral Na⁺-HCO₃^- cotransport, preventing acidosis and damage to the epithelium and the colonic microbiome. Signaling may involve the opening of TRPA1, depolarization of the epithelium and a rise in PGE2 following a lower uptake of prostaglandins via OATP2A1.

Keywords: TRPA1; TRPV3; Ussing chamber; cinnamaldehyde; colon; colonic buffering; epithelial transport; essential oils; intestine; patch clamp; pig; prostaglandin.

Figure 1

Relative mRNA expression of TRPA1 in the mucosa of the stomach (fundus and cardia), duodenum, jejunum, ileum, caecum and colon, and associated muscle layers of four young pigs from a controlled in-house study. Normalization was performed to the reference genes ACTB, GAPDH and YWHAZ, with scaling to the mean values of all of the samples. The letters above the bars indicate statistically significant differences between those bars that do not share a letter (p < 0.05) (n.d. = not detected).

Figure 2

Effect of the TRPA1 agonist cinnamaldehyde at a concentration of 100 µmol·L⁻¹ (a,c) and 1 mmol·L⁻¹ (b,d) on I_sc and G_t in the Ussing chamber using the colonic tissue of young pigs (controlled in-house study). While the smaller concentration of 100 µmol·L⁻¹ was insufficient, after the addition of 1 mmol·L⁻¹, a significant increase of the I_sc and G_t could be observed. After washout, the values dropped again. N/n = the number of animals/number of tissues, which was identical for G_t and I_sc.

Figure 3

Effect of the TRPA1 agonist cinnamaldehyde (CIN) at a concentration of 1 mmol·L⁻¹ after the mucosal (a,d), serosal (b,e) or bilateral addition (c,f) to I_sc and G_t in the Ussing chamber using the colonic tissue of pigs. After mucosal and bilateral addition, a significant increase in I_sc and G_t was observed, although the effects were absent after the serosal addition (details see text). In this figure, some tissues were obtained from older, larger pigs from a commercial slaughterhouse (blue lines), which showed a strong, sustained increase, while the black lines reflect the frequently biphasic responses that were seen in younger and smaller pigs slaughtered within a controlled study, as in the rest of the manuscript.

Figure 4

Comparison of the effect of cinnamaldehyde (1 mmol·L⁻¹) on the short circuit current (ΔI_sc) and the conductance (ΔG_t) of the colonic mucosa from young pigs (controlled in-house study) in Ussing chambers after preincubation with different blockers, or after the ion replacement. In order to allow a comparison of the data from different sets of experiments, the differences (deltas) were calculated by subtracting the peak value of the cinnamaldehyde response in the 15 min period after addition of cinnamaldehyde from the baseline value before addition. The delta values of the control tissues of each animal were set to 100%, and the deltas of the treated tissues were calculated as a percentage of the control tissues of the animal in question. Significant differences versus the control group are marked as *, †, or ‡ (p < 0.05, p < 0.01 or p < 0.001). For the concentrations of the blockers and the composition of the solutions used, see the results and the Supplementary Materials. N/n = number of animals/number of tissues, which were identical for G_t and I_sc; muc = mucosal; bilat = bilateral.

Figure 5

Effect of a bilateral application of thymol on the I_sc and G_t of the colon of ten young pigs (controlled in-house study). The data are given as means ± SEM. In some of the tissues, an increase (“up“) in I_sc could be observed after the addition of thymol (a), whereas in other tissues a decrease or no effect (“down“) was observed (b). An increase in G_t was observed in all of the tissues after the addition of thymol, ruling out barrier effects for increases in I_sc (c,d). The significance bars within graphs compare values taken immediately prior to addition of the agonist and after an incubation of 10 min. The significance bars between graphs indicate that in the colon, there was no difference between the “up” and the “down” groups before thymol was added. Significant differences are marked as *, †, or ‡ (p < 0.05, p < 0.01 or p < 0.001).

Figure 6

(a) Immunohistochemical staining of HEK-293 cells transfected with a vector for the simultaneous overexpression of human TRPV3 and green fluorescent protein (GFP, green), stained with a specific antibody against TRPV3 (red). The cell nuclei were stained with DAPI (blue). (b) An original recording of a patch clamp measurement of an hTRPV3 HEK-293 cell at 23 °C. No visible response was seen after the addition of 5 mmol·L⁻¹ cinnamaldehyde (CIN), whereas 1 mmol·L⁻¹ thymol (THY) elicited a clear response. (c) A patch clamp measurement of an hTRPV3 HEK-293 cell at 37 °C. The concentration of cinnamaldehyde had to be elevated to 5 mmol·L⁻¹ before a small response could be observed. In contrast, at 1 mmol·L⁻¹, the effects of the thymol were very strong. (d) A boxplot of patch clamp data from hTRPV3 HEK-293 cells at 37 °C and 23 °C, and from control cells transfected with the empty vector (MT, 37 °C) at +100 mV and −120 mV. Within a group, significant differences after the addition of cinnamaldehyde or thymol and the subsequent washouts (NaCl(2) and NaCl(3)) are indicated via different letters above the bars. Comparisons between the groups are given in the main text (e) Data from Ussing chamber experiments from a subset of seven young pigs from Figure 5 (controlled in-house study), treated in parallel with either thymol (N/n = 7/13) or cinnamaldehyde (N/n = 7/14). In the native colonic tissues, 1 mmol·L⁻¹ of cinnamaldehyde was sufficient to induce a clear change in the short circuit current and the conductance, which rose by ΔI_sc and ΔG_t, respectively. For comparison, the data for thymol (1 mmol·L⁻¹) are also shown. Here, the I_sc responses were clearly smaller, but diverse, with some tissues showing an increase in I_sc (“up”, N/n = 4/7) and others a decrease (“down”, N/n = 4/6). Bars that do not share a letter are significantly different.

Figure 7

Model of the effects of cinnamaldehyde on the colon, based on the present study and the current literature. (a) Cinnamaldehyde opens apical, non-selective TRPA1 channels in the colonic mucosa near the lumen (). The selectivity filter of the channel allows both the influx of Na⁺ and Ca²⁺, and a smaller efflux of K⁺, so that the effects on I_sc are small. However, the cell is depolarised and a significant increase in conductance ΔG_t is observed, which is reduced by quinidine and enhanced by the removal of divalent cations. The effects of cinnamaldehyde on TRPV3 (➀), which favors efflux of K⁺ over influx of Na⁺ and Ca²⁺, are discrete. Thymol opens both channels. (b) Prostanoids such as PGE2 are anions that are synthesized from membrane phospholipids via cyclooxygenase-mediated pathways and secreted into the extracellular space via pathways that are being explored (➁). For prostanoid signalling to end, the anionic prostaglandin has to be taken up into the cytosol via an electrogenic anion exchanger, OATP2A1 (SLCOA1) (➂), after which the prostaglandin is degraded by cytosolic enzymes. Due to the electrogenic nature of the cotransporter, the depolarization of the cellular membrane, as occurs after the opening of TRPA1 channels via cinnamaldehyde (), decreases the uptake of prostaglandins and thus increases the extracellular prostaglandin concentration. (c) After the binding of the PGE2 to EP4 receptors (➃) expressed by the colonic mucosa, adenylyl cyclase is stimulated, resulting in rising levels of cAMP that open apical CFTR channels (➄). Other anion channels may contribute to the secretion of HCO₃⁻, which is driven by the uptake of Na⁺ via basolateral NBCn1 (Slc4a7), NBCe1 (SLC4A4), or NBCe2 (Slc4a5) at a ratio of 1, 2 or 3 HCO₃⁻ for each Na⁺ (➅). Most of the NPPB-sensitive rise in I_sc that is observed after the activation of TRPA1 via cinnamaldehyde can be explained by this mechanism. The secretion of HCO₃⁻ is important for the buffering of protons formed in the fermentational process (➆), and for the unfolding of mucines in the mucus layer, thus protecting the epithelium. Energy-rich short chain fatty acid anions (SCFA⁻) are absorbed via various transport proteins (➇) without challenging cytosolic pH homeostasis. In physiological concentrations, prostaglandins are also thought to have barrier-enhancing properties through interaction with tight junction proteins (➈). Possibly, the secretion of HCO₃⁻ is highest in cells near the surface, while in the crypts, the expression of NKCC1 (SLC12A2) (➉) predominates. The latter pathway leads to the secretion of Cl⁻ via CFTR, which can result in diarrhea when cAMP levels are pathologically high. Because the gradients favor a unilateral efflux of anions, the opening of CFTR will have higher effects on ΔI_sc than those after the opening of TRPA1.