Diisocyanates influence models of atopic dermatitis through direct activation of TRPA1

Abstract

We recently used EPA databases to identify that isocyanates, most notably toluene diisocyanate (TDI), were the pollutant class with the strongest spatiotemporal and epidemiologic association with atopic dermatitis (AD). Our findings demonstrated that isocyanates like TDI disrupted lipid homeostasis and modeled benefit in commensal bacteria like Roseomonas mucosa through disrupting nitrogen fixation. However, TDI has also been established to activate transient receptor potential ankyrin 1 (TRPA1) in mice and thus could directly contribute to AD through induction of itch, rash, and psychological stress. Using cell culture and mouse models, we now demonstrate that TDI induced skin inflammation in mice as well as calcium influx in human neurons; each of these findings were dependent on TRPA1. Furthermore, TRPA1 blockade synergized with R. mucosa treatment in mice to improve TDI-independent models of AD. Finally, we show that the cellular effects of TRPA1 are related to shifting the balance of the tyrosine metabolites epinephrine and dopamine. This work provides added insight into the potential role, and therapeutic potential, or TRPA1 in the pathogenesis of AD.

Conflict of interest statement

The authors have declared that no competing interests exist.

Copyright: This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

Figures

Fig 1. Toluene diisocyanate (TDI) induced dermatitis in mice is dependent on TRPA1.

(A) Overview of model where TDI was applied to B6129PF2/J wild type (WT) or TRPA1 knockout (TRPA1-/-) mouse ears every other day for 3 weeks. (B-D) Resultant thickness for each mouse (B), representative photos (C) and H&E histology (D) are shown. Note that both the B6129PF2/J control and TRPA1-/- mice have natural variation in coloration, thus a representative image from various mice are shown to encompass the range in these strains. (E-F) Schwann cells stimulated with either TDI (at indicated final concentrations) or toluene (arrow indicates when TDI was added). In select conditions pre-incubation with the TRPA1 inhibitor HC030031 either 30 minutes prior (preHC), immediately prior (HC), or both (preHC-HC). Mean calcium flux fluorescent intensity (E) and representative image at peak brightness (F) are shown. Full data set and images can be found in S1 Fig and S1 Video. Data represent three independent experiments and are displayed as mean+SEM (B) or mean (E). *** indicated p

Fig 2. TRPA1 blockade improves non-TDI models of atopic dermatitis.

(A) Overview of MC903 model where topical application occurred daily for 8 days before topical Roseomonas mucosa, cardamonin (Card), or ground cardamom seeds (Spice) were applied for three days. Ears were assessed 3 days after treatment; resultant ear thickness (B) and representative histology (C) are shown (N = 8–19 per group with each dot representing one mouse). (D) R. mucosa was grown in broth with either diluent, cardamonin or ground cardamom seeds (N = 1 isolate in triplicate wells). Resultant impact on optical density is shown. Data represent a combination of (B) or representative of (C-D) two or more independent experiments and are displayed as mean+SEM. ns = not significant, * = p <0.05, ** = p <0.01, *** = p <0.001, **** = p<0.0001 as determined by ANOVA for all displayed items.

Fig 3. TRPA1-/- have differing metabolic response to TDI exposure.

Ears from mice treated with TDI as in Fig 1 were collected and examined by imagine mass spec. (A) Segmentation for representative ear, (B) NMDS similarly plot, and all of the metabolites that were identified by both m/z and collisional cross section (C) are shown. Data represent three independent experiments and calculated as NMDS (B) or ANOVA (C). WT = wild type mice.

Fig 4. TRPA1 blockade alters metabolic response to AD models in mice.

Ears from mice treated with TDI as in Fig 3 were collected and examined by imagine mass spec. (A) Segmentation for representative ear, (B) NMDS similarly plot, and all of the metabolites that were identified by both m/z and collisional cross section (C) are shown. Data represent two independent experiments and calculated using ANOVA. Dil = diluent, Card = cardamonin, Spice = ground cardamom seeds, Rm = Roseomonas mucosa, Rm+C = R. mucosa and Card, Rm + S = R. mucosa and Spice.

Fig 5. Metabolic influence of TRPA1 is concentrated in tyrosine metabolism.

(A-C) Individual pathways as indicated from Metaboanalyst taken from metabolites identified as different between wild type and TRPA1-/- mice in response to TDI (A), R. mucosa treated versus R. mucosa plus Cardamonin (RmC; B), and R. mucosa treated versus R. mucosa plus ground cardamom seeds (RmS; C). (D) Summarizing all pathways impacted by index of pathway significance (IPS) for each condition versus its diluent control. (E, F) Percent of starting wound closure over time for keratinocytes (E) and Schwann neuron cells (F) incubated with diluent, the TRAP1 agonist cinnamaldehyde (Cinn) or TRPA1 blocker HC030031 (N = 1 cells line each in triplicate wells). Significance determined by comparison of area under the curve with 95% confidence intervals by PRISM. (G) Schwann cells in scratch assay with Cinn with or without addition of anti-dopaminergic haloperidol (halo) or anti-adrenergic sotalol (N = 1 cells line each in triplicate wells). Data represent two or more independent experiments and are shown as mean ± SEM. * = p < 0.05 for comparison of area under the curve for wound closure as indicated.