Dietary targeting of TRPM8 rewires macrophage immunometabolism reducing colitis severity

Abstract

The interplay between diet, host genetics, microbiota, and immune system has a key role in the pathogenesis of inflammatory bowel disease (IBD). Although the causal pathophysiological mechanisms remain unknown, numerous dietary nutrients have been shown to regulate gut mucosal immune function, being effective in influencing innate or adaptive immunity. Here, we proved that transient receptor potential melastatin 8 (TRPM8), a non-selective cation channel, mediates LPS- evoked Ca²⁺ influx in macrophages leading to their activation. Additionally, we showed that TRPM8 is selectively blocked by the dietary flavonoid luteolin, which induced a pro-tolerogenic phenotype in pro-inflammatory macrophages. Accordingly, genetic deletion of Trpm8 in macrophages caused a deficit in the activation of pro-inflammatory metabolic and transcriptional reprogramming, leading to reduced production of key pro-inflammatory cytokines such as interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α. The TRPM8 anti-inflammatory effect was found to be dependent on lactate which in turn induces IL-10 gene expression. Adoptive transfer of TRPM8-deficient bone marrow in wild-type mice improved intestinal inflammation in a model of colitis. Accordingly, oral administration of luteolin protected mice against colitis through an impairment in the innate immune response. Our study reveals the potential of targeting TRPM8 through specific nutrient interventions to regulate immune function in sub-clinical scenarios or to treat inflammatory diseases, primarily driven by chronic immune responses, such as IBD.

Fig. 1. Luteolin is a new selective TRPM8 blocker.

From an evaluation of various diet-derived compounds affinity for TRPM8, luteolin emerged as a new selective TRPM8 blocker. A Scatter diagram of the theoretically estimated affinity of TRPM8 for 17 diet-derived compounds (Ki). Each point represents a natural compound. In red are reported rutin and luteolin. B Rutin structure and binding site on TRPM8. C Luteolin structure and binding site on TRPM8. D Concentration–response curve of luteolin (0.1–100 μM) on intracellular Ca²⁺ levels in TRPM8- HEK293, measured in the presence of icilin 0.25 μM, the TRPM8 reference agonist. Data represent the mean ± SEM of ≥5 determinations. E Effect of luteolin (0.1–30 μM) on intracellular Ca²⁺ levels on HEK293 cells over-expressing TRPM8, TRPV1, and TRPA1 in the presence or not of their respective agonists.

Fig. 2. TRPM8 channel activity controls macrophage response to LPS.

Over-expression of TRPM8 in pro-inflammatory macrophages is essential for LPS-induced intracellular Ca²⁺ increase. A Representative image of TRPM8 (orange), and 4′,6-diamidino-2-phenylindole (DAPI) (blue) immunofluorescence staining in naïve and pro-inflammatory WT BMDMs treated or not with luteolin (10 μM). Analysis was performed 18 h after polarization with LPS + IFN-γ. B Relative changes in the fluorescence of Fluo-4AM-loaded naïve BMDMs, reflecting changes in intracellular Ca²⁺ concentration [Ca^2+]i after icilin (40 μM) perfusion for 1 min. Data are presented as the average of n = 3 independent experiments. Error bars represent ±SEM (n = 58 and n = 60 for Vehicle and Luteolin 10 μM, respectively). C Representative fluorescence images from indicated time points in the experiments shown in (B). D Statistical representation of total peak area of Fluo-4 fluorescence after icilin perfusion from (B). Error bars represent ±SEM. P value was determined using unpaired Student’s ttest. *p < 0.05. E Relative changes in the fluorescence of Fluo-4AM-loaded naïve WT and Trpm8^−/− BMDMs, reflecting changes in intracellular Ca²⁺ concentration [Ca^2+]i after LPS (100 ng/mL) perfusion for 30 s. Data are presented as the average of n = 3 independent experiments. Error bars represent ±SEM (n = 86 and n = 112 for WT and Trpm8^−/−, respectively). F Quantification of basal Ca2+ levels in Fluo-4AM-loaded naïve WT and Trpm8^−/− BMDMs. Error bars represent ±SEM. P value was determined using unpaired Student’s ttest. ****p < 0.0001. G Representative fluorescence images from indicated time points in the experiments shown in (E). H Statistical representation of total peak area of Fluo-4 fluorescence after LPS perfusion from panel E. Error bars represent ±SEM. P value was determined using unpaired Student’s t-test. ****p < 0.0001. I Relative changes in the fluorescence of Fluo-4AM-loaded naïve BMDMs, reflecting changes in intracellular Ca²⁺ concentration [Ca²⁺]i BMDMs after LPS (100 ng/mL) perfusion for 30 s. Data are presented as the average of n = 3 independent experiments. Error bars represent ± SEM (n = 108 and n = 113 for Vehicle and Luteolin 10 μM, respectively). J Representative fluorescence images from indicated time points in the experiments shown in (I). K Statistical representation of total peak area of Fluo-4 fluorescence after LPS perfusion from panel I. Error bars represent ±SEM. P value was determined using unpaired Student’s t test. ****p < 0.0001.

Fig. 3. TRPM8 is required for pro-inflammatory polarization of macrophages.

TRPM8 blockade with luteolin, or its genetic ablation, blunt macrophage pro-inflammatory capacity. A Nitrite levels and TNF-α, IL-6, and IL-1β levels were measured in the supernatant of naïve and pro-inflammatory WT or Trpm8^−/− BMDMs. Analysis was performed 18 h after polarization with LPS + IFN-γ. Data are presented as the average of n ≥ 3 independent experiments, with each dot representing an individual mouse. Error bars represent mean ± SEM. P value was determined using Two-way ANOVA followed by Šídák’s multiple comparisons test. ***p < 0.001; ****p < 0.0001. B Nitrite levels and TNF-α, IL-6, and IL-1β levels were measured in the supernatant of pro-inflammatory WT or Trpm8^−/− BMDMs treated or not with luteolin (10 μM). Analysis was performed 18 h after polarization with LPS + IFN-γ. Data are presented as the average of n ≥ 3 independent experiments, with each dot representing an individual mouse. Error bars represent ±SEM. P value was determined using two-way ANOVA followed by Šídák’s multiple comparisons test. ns non significant,*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. C IL-6, IL-1β, and TNF-α mRNA relative expression in pro-inflammatory WT BMDMs treated or not with luteolin (10 μM). The analysis was performed after 0, 1, 3, 6, or 12 h of polarization with LPS + IFN-γ. Data are expressed as relative to basal levels (0 h timepoint) and are presented as the average of n = 5 independent experiments. Error bars represent ±SEM. P value was determined using multiple paired Student’s ttest. *p < 0.05; **p < 0.01. D IL-10 relative mRNA levels (left) and protein levels (right) measured in pro-inflammatory WT BMDMs treated or not with luteolin (10 μM). The analysis was performed after 0, 1, 3, 6, or 12 h of polarization with LPS + IFN-γ. mRNA data are expressed as relative to basal levels (0 h timepoint). All the data are presented as the average of n ≥ 5 independent experiments. Error bars represent ±SEM. P value was determined using paired multiple Student’s t test. *p < 0.05; **p < 0.01. E IL-10 relative mRNA levels measured in pro-inflammatory WT or Trpm8^−/− BMDMs treated or not with luteolin (10 μM). The analysis was performed after 0, 1, 3, 6, or 12 h of polarization with LPS + IFN-γ. Data are expressed as relative to basal levels (0 h timepoint) and are presented as the average of n = 5 independent experiments. Error bars represent ±SEM. P value was determined using multiple paired Student’s t test. *p < 0.05.

Fig. 4. TRPM8 regulates immunometabolic pathways in macrophages.

TRPM8 modulation with luteolin reduces the metabolic rewiring of macrophages, affecting the levels of the main immunoregulatory metabolites. A ECAR kinetics (left) and glycolytic parameters (right) of naïve and pro-inflammatory WT and Trpm8^−/− BMDMs treated or not with Luteolin (10 μM). Analysis was performed 18 h after polarization with LPS + IFN-γ. Data are presented as the average of n = 3 independent experiments. Error bars represent ±SEM. P value was determined using one-way ANOVA followed by Tukey’s multiple comparisons test. **p < 0.01; ****p < 0.0001. Score and loading plots derived from the PCA model (B) and succinate, lactate, glycine, glutamine, and glutamate levels (C) measured by NMR metabolomic analysis of pro-inflammatory RAW 264.7 treated or not with Luteolin (10 μM). Analysis was performed 18 h after polarization with LPS. C Data are presented as peak area normalized on total area of n = 3 independent experiments. Error bars represent ±SEM. P value was determined using unpaired Student’s t test. *p < 0.05; **p < 0.01.

Fig. 5. Blocking of TRPM8 induces lactate-mediated activation of IL-10 pathway.

Luteolin, via TRPM8, increases lactate levels necessary for IL-10 pathway activation. A Relative lactate levels measured by GC-MS metabolomic analysis of pro-inflammatory BMDMs treated or not with Luteolin (10 μM). The analysis was performed after 3 or 6 h of polarization with LPS + IFN-γ. Data are presented as peak area normalized on μg of protein of n = 3 independent experiments, each performed in technical triplicate. Error bars represent ±SEM. P value was determined using unpaired Student’s t test. *p < 0.05. B Relative lactate levels measured by GC-MS metabolomic analysis of WT and Trpm8^−/− BMDMs. The analysis was performed in basal conditions (t0) and after 3 or 6 h of polarization with LPS + IFN-γ. Data are presented as peak area normalized on μg of protein of n = 3 independent experiments, each performed in technical triplicate. Error bars represent ± SEM. P value was determined using unpaired Student’s t test. *p < 0.05; **p < 0.01. IL-10 (C) and IL-6, TNF-α, and IL-1β (D) mRNA relative expression measured in pro-inflammatory WT BMDMs treated or not with luteolin (10 μM) and LDH-a inhibitor GSK2837808A (10 μM). The analysis was performed after 0, 1, 3, or 6 h of polarization with LPS + IFN-γ. Data are expressed as relative to basal levels (0 h timepoint) and are presented as the average of n = 5 independent experiments. Error bars represent ±SEM. P value was determined using multiple paired Student’s t test. *p < 0.05; **p < 0.01; ***p < 0.001 Pro-inflammatory+Lut10μM vs Pro-inflammatory; #p < 0.05 Pro-inflammatory+LDHAi+Lut10μM vs Pro-inflammatory+LDHAi. E IL-6, IL-1β, and TNF-α levels measured in the supernatant of IL-10^fl/fl WT and IL-10^fl/fl LysMCre BMDMs treated or not with luteolin (10 μM), after 3, 6, or 12 h of polarization with LPS + IFN-γ. Data are presented as the average of n = 5 independent experiments. Error bars represent ±SEM. P value was determined using multiple paired Student’s t test. *p < 0.05; ** p < 0.01 vs IL-10^fl/fl WT pro-inflammatory. F ECAR kinetics (left) and glycolytic parameters (right) of naïve and pro-inflammatory IL-10^fl/fl WT and IL-10^fl/fl LysMCre BMDMs treated or not with Luteolin (10 μM). Analysis was performed 18 h after polarization with LPS + IFN-γ. Data are presented as the average of n = 3 independent experiments. Error bars represent ±SEM. P value was determined using one-way ANOVA followed by Tukey’s multiple comparisons test. n.s. non-significant, *p < 0.05; **p < 0.01; ****p < 0.0001.

Fig. 6. Blocking TRPM8 in mice ameliorates DSS-induced colitis.

WT mice reconstituted with Trpm8^−/− BM cells or supplemented with oral luteolin are resistant to DSS-induced colitis by reduced recruitment of inflammatory myeloid cells. A Schematic representation of DSS protocol and timeline. B Disease activity index (DAI) of colitis severity. Data are presented as the average of n = 5 mice for each experimental group. Error bars represent ±SEM. P value was determined using Student’s t test. *p < 0.05. C Spleen weight (g) and weight/length ratio (mg/cm) of colon tissue in WT mice reconstituted or not with Trpm8^−/− BM. Error bars represent ±SEM. P value was determined using unpaired Student’s t-test. **p < 0.01. D Histological score (left) and representative H&E images (right) of colon sections of WT mice reconstituted or not with Trpm8^−/− BM. Error bars represent mean ± SEM. P value was determined using an unpaired Student’s t test. *p < 0.05. E tSNE + FlowSOM analysis performed on CD45⁺ lineage of lamina propria of colon samples from WT mice subjected to DSS administration (n = 3 mice). F Heatmap representing Trpm8 expression throughout the clusters of CD45⁺ lineage from lamina propria of colon samples from WT mice subjected to DSS administration (n = 3 mice). G Expression of CD45, Ly6C, MHCII and CD206 by cells isolated from colonic lamina propria of WT mice reconstituted or not with Trpm8^−/− BM. Data are presented as the average of n = 3 mice for each experimental group. Error bars represent ±SEM. P value was determined using unpaired Student’s t test. ns non-significant; *p < 0.05; **p < 0.01. H Schematic representation of DSS protocol and timeline. I Mice body weight gain/loss. Data are presented as the average of n = 10 mice for each experimental group. Error bars represent ±SEM. P value was determined using unpaired Student’s t test. **p < 0.01. J Disease activity index (DAI) of colitis severity. Data are presented as the average of n = 10 mice for each experimental group. Error bars represent ±SEM. P value was determined using Student’s t test. ****p < 0.0001. K Histological score (left) and representative H&E images (right) of mice colon sections. Error bars represent mean ± SEM. P value was determined using an unpaired Student’s t test. ***p < 0.001. L Representative confocal images of TRPM8 (magenta), IBA-1 (yellow), and DAPI (cyan) immunostaining in DSS+Vehicle or DSS+Luteolin treated mice, collected 15 dpt. Analyses were carried out on five mice for each experimental group. Representative flow plots and count of CD45 (M) and neutrophils (N) populations in lamina propria of mice colon. Data are presented as the average of n = 5 mice for each experimental group. Error bars represent ±SEM. P value was determined using unpaired Student’s t test. **p < 0.01. Representative flow plots and expression of Ly6C and MHCII (O) and CCR2 and CD206 (P) by monocytes isolated from lamina propria of mice colon. Data are presented as the average of n = 5 mice for each experimental group. Error bars represent ±SEM. P value was determined using unpaired Student’s t-test. *p < 0.05; **p < 0.01; ****p < 0.0001.