Blockade of b-catenin signaling attenuates toluene diisocyanate-induced experimental asthma
Abstract
Background: Aberrant activation of b-catenin signaling by both WNT-dependent and WNT-independent pathways has been demonstrated in asthmatic airways, which is thought to contribute critically in remodeling of the airways. Yet, the exact role of b-catenin in asthma is very poorly defined. As we have previously reported abnormal expression of b-catenin in a toluene diisocyanate (TDI)- induced asthma model, in this study, we evaluated the therapeutic efficacy of two small molecules XAV-939 and ICG-001 in TDI-asthmatic male BALB/c mice, which selectively block b-catenin-mediated transcription.
Methods: Male BALB/c mice were sensitized and challenged with TDI to generate a chemically induced asthma model. Inhibitors of b-catenin, XAV-939, and ICG- 001 were respectively given to the mice through intraperitoneally injection.
Results: TDI exposure led to a significantly increased activity of b-catenin, which was then confirmed by a luciferase assay in 16HBE transfected with the TOP- Flash reporter plasmid. Treatment with either XAV-939 or ICG-001 effectively inhibited activation of b-catenin and downregulated mRNA expression of b-cate- nin-targeted genes in TDI-asthmatic mice, paralleled by dramatically attenuated TDI-induced hyperresponsiveness and inflammation of the airway, alleviated air- way goblet cell metaplasia and collagen deposition, decreased Th2 inflammation, as well as lower levels of TGFb1, VEGF, HMGB1, and IL-1b.
Conclusion: The results showed that b-catenin is a principal mediator of TDI- induced asthma, proposing b-catenin as a promising therapeutic target in asthma.
b-catenin is an evolutionary conserved molecule that exerts dual roles in cellular signaling. As a membrane-bound pro- tein, b-catenin constitutes a key component of adherens junc- tions where it interacts with the cadherins to connect them to the cytoskeleton (1). Besides, cytosolic and nuclear b-catenin is the pivotal effector of the WNT signaling. The canonical WNT/b-catenin pathway initiates a signaling cascade funda- mental in regulating various biologic processes such as organ development, tissue homeostasis, and pathogenesis of human diseases (2). The hallmark of this pathway is that it activates the transcriptional role of b-catenin. In the absence of Wnt signals, a multiprotein destruction complex including casein kinase I, glycogen synthase kinase-3b (GSK3b), axin, and the adenomatous polyposis coli phosphorylates b-catenin at ser- ine residues in the N-terminus to promote its ubiquitination and proteasomal destruction. Activation of Wnt signaling inhibits this degradative process by binding to Frizzled recep- tors and signaling through the associated low-density lipoprotein-related proteins, LRP5/6, thereby allowing b-cate- nin to accumulate in the cytosol (active) and enter the nucleus, where it binds to members of the T-cell factor (TCF)/lymphoid enhancer-binding factor family of transcrip- tion factors (3, 4). Nuclear b-catenin/TCF then assembles a transcriptionally active complex by recruiting the transcrip- tional coactivators cAMP response element-binding protein (CBP) or its closely related protein p300, as well as other components of the basal transcription machinery, to stimu- late transcription of various target genes (5).
Activation of b-catenin signaling has been demonstrated in airway smooth muscle cells and airway epithelium in both in vivo and in vitro models of asthma (6–8). Recently, b-cate- nin has been proposed as a regulator and therapeutic target for asthmatic airway remodeling (9). Using small interfering RNA targeting b-catenin not only decreased smooth muscle hyperplasia, subepithelial fibrosis, and eosinophil numbers in bronchoalveolar lavage fluid (BALF) in an ovalbumin (OVA)-induced chronic asthma model (10), but also inhibited LPS/lipoteichoic acid-induced upregulation of IL-6, IL-8, and IL-1b in airway epithelial cell lines (11, 12), suggesting critical roles for b-catenin in asthma pathogenesis. In a previ- ous study, we have demonstrated aberrant activation of b- catenin signaling in a toluene diisocyanate (TDI)-induced asthma model (13). Yet, how b-catenin signaling contributes to the pathology of TDI asthma still remains unclear.
Significant advances have been made in the generation of small-molecule inhibitors that target b-catenin directly or indirectly. Recent studies have identified ICG-001, a pep- tidomimetic small molecule that selectively inhibits b-catenin signaling in a CBP-dependent fashion. Mechanistically, ICG- 001 competes with b-catenin for binding to CBP to prevent b- catenin/CBP complex formation, thereby inhibiting b-catenin- mediated gene expression (14–17). Another example is XAV- 939. Pharmacologic inhibition of b-catenin signaling by XAV- 939 stabilizes the b-catenin destruction complex, decreased cytoplasmic retention and nuclear import of b-catenin, and efficiently inhibited b-catenin-targeted gene transcription (18). These two offer almost perfect approaches to selectively dis- rupt the signaling role of b-catenin without affecting its adhe- sive function. Both have proved great capacity to suppress organ fibrosis in animal models (19–22). Therefore, in this study, we intended to examine the therapeutic effects of ICG- 001 and XAV-939 in our asthma model to investigate deeper mechanisms for b-catenin signaling in TDI-induced asthma.
Methods
Reagents
Toluene diisocyanate (toluene-2,4-diisocyanate, ≥98.0%) was obtained from Sigma-Aldrich (Shanghai, China). The vehicle (AOO) used to dissolve TDI consists of a mixture of two vol- umes of acetone and three volumes of olive oil for sensitiza- tion, and one volume of acetone and four volumes of olive oil for the challenge. ELISA kit for total IgE, as well as multiplex immunoassay kit for detecting IFN-c, IL-4, IL-5, IL-13, VEGF, and IL-25, was from eBioscience (San Diego, CA, USA). Rabbit anti-b-catenin and anti-IL-1b antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-non-phospho-b-catenin (Ser33/37/Thr41) antibody was from Cell Signaling Technology. Anti-HMGB1, anti-VEGF, and anti-TGFb1 antibodies were bought from ABclonal Biotech (Beijing, China).
Animals and in vivo treatment with b-catenin signaling inhibitors
Male BALB/c mice (6–8 weeks) were purchased from South- ern Medical University. The mice were housed in a SPF facil- ity with 12-h dark/light cycles and fed with sterile water and irradiated food ad libitum. All studies were conducted in accordance with the committee of Southern Medical Univer- sity on the use and care of animals. The protocols were approved by the Animal Subjects Committee of Nanfang Hospital. In the first protocol, mice were randomized to the following groups: (i) AOO-sensitized, AOO-challenged, and DMSO-treated (AOO/AOO group); (ii) TDI-sensitized, TDI-challenged, and DMSO-treated (TDI/TDI group); (iii) TDI-sensitized, TDI-challenged, and XAV-939-treated (TDI/ TDI + XAV-939 group); (iv) TDI-sensitized, TDI-challenged, and ICG-001-treated (TDI/TDI + ICG-001 group). In the second protocol, to explore the role of b-catenin in the toxic response of TDI, mice were allocated to the paralleled four groups: (i) none-sensitized, AOO-challenged, and DMSO- treated (-/AOO group); (ii) none-sensitized, TDI-challenged, and DMSO-treated (-/TDI group); (iii) none-sensitized, TDI- challenged, and XAV-939-treated (-/TDI + XAV-939 group); (iv) none-sensitized, TDI-challenged, and ICG-001-treated (-/ TDI + ICG-001 group). Animal models were generated as described elsewhere (13). Briefly, on days 1 and 8, mice were dermally sensitized with 0.3% TDI on the dorsum of both ears (20 ll per ear). On days 15, 18, and 21, the mice were placed in a horizontal rectangle chamber and challenged for 3 h each time with 3% TDI through compressed air nebu- lization (NE-C28; Omron, Tokyo, Japan). As a control, mice were sensitized and challenged by the same procedures with the same amount of AOO. The none-sensitized mice only underwent airway challenge twice on days 15 and 18. XAV- 939 (2.5 mg/kg, i.p.; Selleck Chem, Shanghai, China) (22) and ICG-001 (5 mg/kg, i.p.; Selleck) (19) dissolved in DMSO and diluted with PBS (pH = 7.4) were, respectively, given to the mice once daily, beginning immediately after the first challenge to the last day of challenge. Sham mice received the same volume of vehicle by comparison.
Assessment of airway hyperresponsiveness
Airway reactivity to methacholine was assessed on day 22. Measurements of lung resistance (RL) were performed on anaesthetized, intubated, and mechanically ventilated (Buxco Electronics, Troy, NY, USA) mice in response to increasing doses of aerosolized methacholine (Mch; 3.12, 6.25, 12.5 and 25 mg/ml). Measurements of RL were performed every 5 min following each nebulization step.
Quantification of serum IgE, preparation of lymphocyte supernatants, and analysis of bronchoalveolar lavage fluid
One day after measuring airway parameters, mice were killed with overdose of pentobarbital, then blood samples were immediately taken, rest at room temperature for 2 h, then centrifuged (3000 g, 20 min) and supernatants were harvested for detection of total IgE by ELISA according to the manu- facturers’ instructions.
As soon as blood was taken, cervical lymph nodes were dissected and processed, and supernatants of cultured lym- phocytes were obtained as previously described (23), as well as bronchoalveolar lavage fluid (BALF). Total cells in BALF were counted, and a cytospin sample was prepared and stained with hematoxylin and eosin (H&E) for blinded assessment of differential cell percentages in BALF. Then, the remaining fluids were centrifuged (1000 g, 10 min) and supernatants were stored at —80°C. VEGF, IL-25 in super- natants of BALF, and IFN-c, IL-4, IL-5, and IL-13 in super- natants of cultured lymphocytes were detected by a multiplex immunoassay kit according to procedures recommended by the manufacturer.
Histopathologic examination of the lung
Left lungs were isolated and fixed with 4% neutral formalin, paraffin-embedded, cut in 4-lm sections, and stained with H&E, periodic acid–Schiff (PAS), and Masson’s trichrome for blinded histopathologic assessment. Lung inflammation and airway goblet cell metaplasia were semiquantified as pre- viously described (23). The presence of collagen fibril deposi- tion with Masson’s staining was analyzed by IMAGE-PRO PLUS software (Media Cybernetics, Silver Spring, USA). Scoring was performed at a magnification of 2009 by examining atleast 40 image fields of 20 slices from eight mice per group. Sections were assigned a random code to blind the examiner to the identity of each specimen.
Western blot and immunohistochemistry
Pulmonary expression of non-phospho (active)-b-catenin (Ser33/37/Thr41), HMGB1, TGFb1, and IL-1b was detected by Western blot. Whole lung extracts mixed with 59 SDS loading buffer were separated by 10% SDS–polyacrylamide gel and transferred onto PVDF membranes. Membranes were then probed with anti-b-catenin, anti-non-phospho-b-catenin (Ser33/37/Thr41), anti-TGFb1, anti-HMGB1, and anti-IL-1b antibodies with indicated dilutions. After incubation with an IRDye® 680WC-conjugated secondary antibody (LI-COR Biosciences, Lincoln, NE), immunoreactive bands were exposed to Odyssey® CLx Imager (LI-COR Biosciences, Lin- coln, NE) for image capture. Data analysis was performed with ODYSSEY software (LI-COR Biosciences, Lincoln, NE).
For immunohistochemistry of b-catenin and VEGF, lung sections were deparaffinized, then subject to antigen retrieval. Samples were treated with H2O2 for 15 min to block endoge- nous peroxidase, and then incubated overnight at 4°C in rec- ommended dilutions of anti-b-catenin and anti-VEGF antibodies. After washing with PBS, slices were incubated with a secondary antibody for 20 min at room temperature. Signals were visualized with DAB.
Treatment of mouse lung for qPCR
Expression of MMP-2, MMP-9, fibronectin, TGFb1, and VEGF was detected by the SYBR green-based PCR tech- nique. Total lung RNA was isolated using an RNAiso Plus kit (Takara, Guangzhou, China). RNA samples were then reverse-transcribed into first-strand cDNA using the Prime- ScriptTM RT reagent kit from Takara. A model LightCycler 480 Fast Real-Time PCR System was used. Primers for amplifying GAPDH cDNA were 50-AAGAGGGATGCTGCCCTTAC-30 (forward) and 50-CCAATACGGCCAAATCCGTTC-30 (reverse); primers for MMP-2 were 50-TGCAGG AGACAAGTTCTGGAG-30 (forward) and 50-GTAGCTATGACCACCACCCTG-30 (reverse); primers for MMP-9 were 50-CGTGTCTGGAGATTCGACTTGA-30 (forward) and 50-TGGTTCACCTCATGGTCCAC-30 (reverse); primers for fibronectin were 50-GAAATCTGCAGCCTGGGTCT-30 (forward) and 50-ACACCCAGCTTGAAGCCAAT-30 (reverse); primers for TGFb1 were 50-ACGTCACTGGAGTTGTACG G-30 (forward) and 50-GGGGCTGATCCCGTTGATT-30 (reverse); primers for VEGF were 50-CACTGGACCC TGGCTTTACT-30 (forward) and 50-ACTTGATCACTTCATGGGACTTCT-30 (reverse). The amplification protocol was set as follows: 95°C denaturation for 30 s followed by 40 cycles of 15 s denaturation at 95°C, 1 min of annealing/ extension, and data collection at 60°C.
Cell culture and TOPFlash luciferase assay
This part of methods was shown in the Supporting information.
Statistics
Statistical analysis was performed using SPSS version 20.0 (SPSS Inc., Chicago, Illinois, US). Data were expressed as mean SE, and comparisons among groups were analyzed by one-way analysis of variance (ANOVA) accompanied by Bonferroni post hoc test (equal variances assumed) or Dun- nett’s T3 (equal variances not assumed) post hoc tests for multiple comparisons. P < 0.05 was considered statistically significant. Results XAV-939 and ICG-001 inhibited activation of b-catenin Consistent with our previous study (13), TDI exposure leads to aberrant distribution of b-catenin in the cytoplasm and nucleus, which is most prominent in bronchial epithelial cells (Fig. 1A), while b-catenin in the alveolar region showed no sig- nificant alterations (data not shown). In line with this is an increased level of non-phospho-b-catenin (Ser33/37/Thr41) in the asthmatic lung (Fig. 1B). Next, we performed transient transfection reporter assay with the well-documented TOP- Flash reporter in the bronchial epithelial cell line 16HBE. As can be seen in Fig. 1D, TDI–HSA induced a significant increased luciferase activity in 16HBE. Together, these in vivo and in vitro experiments supported an active b-catenin signal- ing in TDI asthma, which can be suppressed by treatment with either XAV-939 or ICG-001. XAV-939 and ICG-001 blocked b-catenin-mediated gene expression and attenuated airway remodeling Quantitative PCR (qPCR) demonstrated that TDI challenge following dermal sensitization significantly increased the pul- monary expression of a number of well-documented b-catenin target genes also associated with remodeling (MMP- 2, MMP-9, fibronectin, TGFb1, and VEGF) [http://web.stan ford.edu/group/nusselab/cgi-bin/wnt/target_genes]. The increased expression of these genes was almost eliminated in TDI- treated mice given XAV-939 or ICG-001 (Fig. 1C). Protein expression of TGFb1, a critical mediator involved in asthmatic airway remodeling (24), was also upregulated in the lungs of TDI-inhaled mice, but reduced by XAV-939 and ICG-001 (Fig. 2E). We next investigated whether treatment with XAV-939 or ICG-001 could prevent airway remodeling induced by TDI. As seen by histopathology and morphometry, inhalation of 3% TDI 3 h daily on days 15, 18 and 21 following sensitiza- tion induced extensive deposition of collagen and pronounced goblet cell metaplasia in the airway compared with vehicle control. Administration of XAV-939 (2.5 mg/kg/day) or ICG-001 (5 mg/kg/day), beginning the first day of TDI inhalation and continuing for a week, significantly reduced collagen deposition around the airways and epithelial goblet cell hyperplasia as assessed by Masson’s trichrome and PAS staining (Fig. 2A–D). Figure 2 XAV-939 and ICG-001 ameliorated toluene diisocyanate (TDI)-induced asthmatic airway remodeling. (A) Representative Masson’s trichrome-stained lung sections and subsequent semi- quantification (C) showed less collagen deposition around the air- way after mice were treated with XAV/939/ICG-001. Original magnification was 2009. n = 8–10. (B) Representative periodic acid–Schiff (PAS)-stained lung sections and subsequent semiquantification (D) showed the TDI-induced goblet cell meta- plasia was significantly reduced by XAV-939/ICG-001. The red arrows indicate PAS-stained positive epithelia. Original magnifica- tion was 2009. n = 8–10. (E) Level of TGFb1 in the whole lung detected by Western blot. n = 5–6. *P < 0.05 compared with AOO/AOO group; #P < 0.05 compared with TDI/TDI group. XAV-939 and ICG-001 decreased airway hyperresponsiveness and airway inflammation induced by TDI Mice underwent TDI sensitization and challenge displayed increased reactivity to methacholine when compared with those AOO-treated (Fig. 3F), together with a dense accu- mulation of inflammatory cells around the airway, and numerous neutrophils and eosinophils infiltrating into the airway as shown by histology and BALF analysis (Fig. 3A–E). Daily intraperitoneal injection with either XAV-939 or ICG-001 since the first day of challenge dramatically decreased TDI-induced airway hyperrespon- siveness (AHR) and airway inflammation. Higher total serum IgE, a marker of sensitization, was also detected in TDI-asthmatic mice, but was not affected by XAV-939 nor ICG-001 (Fig. 3G). XAV-939 and ICG-001 diminished Th1/Th2 responses Dysregulation of Th1 and Th2 responses was observed in TDI-sensitized and challenged mice, with raised levels of IFN-c (Th1-related) and IL-4, IL-5, IL-13 (Th2-related) in supernatants of cultured lymphocytes, which was greatly recovered by XAV-939 and ICG-001 (Fig. 3H). Yet, we failed to detect IL-25 in BALF, which is a typical epithelia- derived Th2 cytokine. XAV-939 and ICG-001 downregulated expression of VEGF and HMGB1, IL-1b in TDI asthma VEGF is a canonical b-catenin-targeted gene that partici- pates critically in TDI-induced asthma (25). Immunohisto- chemistry revealed relatively faint staining in AOO-exposed lung, but abundant expression after TDI treatment, with robust immunoreactivity in the airway epithelia, airway smooth muscles, and the infiltrating inflammatory cells (Fig. 4A). In line with this, there is a robust increase of VEGF in BALF after TDI inhalation (Fig. 4B). Blockade of b-catenin-targeted transcriptional signaling with XAV- 939 and ICG-001 effectively inhibited VEGF secretion into BALF as well as its expression in the lung. Similar results were obtained when pulmonary HMGB1 and IL-1b were detected by Western blot (Fig. 4C). XAV-939/ICG-001 ameliorated TDI-challenged airway inflammation without dermal sensitization To investigate the effect of signaling inhibition on the toxic response of TDI, another group of mice underwent TDI chal- lenge without prior dermal sensitization were treated with XAV-939 or ICG-001, respectively. As with previously reported (13), inhalation of 3% TDI 3 h daily on days 15 and 18 without dermal sensitization was not sufficient to generate asthma-like responses. Indeed, only slight inflammatory infil- tration around the airway accompanied by mild epithelial hyperplasia was seen in this group of mice, which was completely abolished by daily injection with XAV-939 (2.5 mg/kg) or ICG-001 (5 mg/kg) (Fig. 5A). In line with this, immunohistochemistry showed that treatment with XAV-939/ ICG-001 also ameliorated the diffused expression of b-catenin in the airway epithelium (Fig. 5B). Both inhibitors (with aforementioned doses daily for a week) had no effects on nor- mal BALB/c mice aged 6–8 weeks (data not shown). Discussion In this study, we showed that b-catenin signaling is activated in a murine model of TDI-induced asthma. Selective blockade of b-catenin-mediated transcriptional signaling with XAV-939 and ICG-001 significantly relieved TDI-induced airway remodeling, and for the first time, we demonstrated that b-catenin mediates AHR and inflammation, dysregulated Th1/Th2 responses, and upregulation of TGFb1, VEGF, HMGB1, and IL-1b in asthma. It is well established that aberrant Wnt/b-catenin signaling underlines a wide range of human pathologies, including can- cer, metabolic, inflammatory and fibrotic diseases (26). Recent studies have suggested b-catenin as a therapeutic tar- get in airway remodeling of asthma due to its preeminent roles in cell proliferation, survival, repair, and regeneration (9). Interrupting b-catenin by siRNA technique in an OVA- induced asthma model decreased smooth muscle hyperplasia and subepithelial fibrosis (10), which are key features of airway remodeling. As we have previously demonstrated abnormal expression of b-catenin in TDI-induced asthma model, we hypothesized that b-catenin signal might be important for TDI-induced inflammation and remodeling. In the present study, an increased level of non-phospho-(active) b-catenin (Ser33/37/Thr41) was found in lungs of TDI-treated mice, along with diffused expression in the cyto- plasm and nucleus mainly in the airway epithelia, implying activated b-catenin signaling in vivo. Subsequent luciferase assay in 16HBE fueled our hypothesis that TDI induces bona fide b-catenin signaling activation in airway epithelia. Animals treated with the tankyrase inhibitor XAV-939 had enhanced b-catenin phosphorylation at Ser33/37/Thr41, which triggers ubiquitination and degradation of the protein (3). Although the direct mechanistic function of ICG-001 is not to degrade cytoplasmic b-catenin to inhibit activation, different from that of XAV-939, ICG-001 still downregu- lated non-phospho-b-catenin and rescued the abnormal immunostaining of b-catenin in the airway. This could be explained by a positive feedback response induced by ICG- 001-targeted genes. For instance, TGFb1, whose mRNA and protein levels were both suppressed by ICG-001 in our asthma model, was known to activate b-catenin signaling in airway smooth muscles and bronchial epithelial cells (7, 27). As with the study of Kwak et al. (10), we found that block- ade of b-catenin-mediated transcriptional signaling with XAV-939 or ICG-001 significantly inhibited TDI-induced airway remodeling, manifested by downregulated gene expression associated with remodeling and fibrosis, decreased collagen deposition around the airway, and sligh- ter airway goblet cell hyperplasia. In addition to remodeling, another central feature of the airways in TDI-induced asthma is inflammation, which is dominated by a marked infiltration of neutrophils, a slight infiltration of eosinophils into the airways, and a mixed Th1/ Th2 response (28). Yet currently, there are limited evidence suggesting a direct link between b-catenin and airway inflam- mation. In 2014, researchers from Korea found for the first time that using siRNA downregulating b-catenin inhibited activation of NF-jB and expression of IL-6, IL-8, MCP-1, and TNF-a induced by house dust mite (HDM) in THP-1 (29). Subsequently, the same research team reported that b- catenin is responsible for LPS-induced NF-jB activation and inflammatory cytokine expression in bronchial epithelia (11). Their contributions shed light on a probably underestimate function of b-catenin in inflammation. In this study, both inhibitors displayed high capacity to blunt TDI-induced air- way inflammation, with smaller numbers of neutrophils and eosinophils recruited into the airway, decreased levels of inflammatory cytokines such as VEGF, HMGB1, and IL-1b, and a compromised Th1/Th2 response, indicating that activa- tion of b-catenin signaling is critically engaged in TDI- induced inflammation. In contrast, Trischler et al. (30) found that genetic loss of Wnt10b led to augmented inflammation in an HDM asthma model. Their finding seems opposite to ours given the central role of b-catenin in Wnt signaling. Yet, it should be noted that ligation of Wnt ligands to core- ceptors would result in not only the canonical Wnt/b-catenin signaling but also pathways independent of b-catenin (31). So, it is easy to understand that inhibition of one Wnt ligand may result in almost opposite outcomes from inhibition of Wnt/b-catenin signaling. Although b-catenin signaling is proposed to contribute to TDI-induced airway inflammation, the underlying mecha- nisms still remain obscure. Vascular endothelial growth fac- tor (VEGF) has long been identified as a prototypical target gene of b-catenin (32). It is one of the most potent proangio- genic cytokines and occupies a central role in the process of angiogenesis as well as tissue remodeling (33). Besides, stud- ies have demonstrated that VEGF also contributes critically to airway inflammation, airway remodeling, and Th2 inflam- mation in the lung (34, 35). Increased expression of VEGF was found in patients with TDI-induced asthma as well as animal models (25, 36). Blocking VEGF signaling with VEGFR inhibitors reduced AHR and inflammation induced by TDI in a murine model (25). And we found increased expression of VEGF induced by TDI–HSA conjugate in our previous study (37). These studies suggest that VEGF is one of the major determinants of TDI-induced asthma. In consis- tent with these findings, we detected higher gene and protein levels of VEGF in TDI-sensitized and challenged mice when compared with the AOO-treated. Of note, decreased bron- chial hyperresponsiveness, lung neutrophilic, eosinophilic, and Th2 inflammation were coupled with lower levels of VEGF when mice were treated with XAV939 or ICG-001, suggesting that b-catenin signaling might modulate TDI- induced airway inflammation through a VEGF-dependent mechanism, which needs to be further confirmed in our future studies. There are other important mediators underlying the etiol- ogy of TDI-induced asthma. Matrix metalloproteinases (MMPs) are proteolytic enzymes involved in extracellular matrix turnover that play an essential role in physiological tissue remodeling and repair (38). Members of this family include MMP-2 and MMP-9. Both are well-known down- stream target genes of b-catenin signaling (http://web.stan ford.edu/group/nusselab/cgi-bin/wnt/target_genes) and potent chemotactic factors that mediate transmigration of neu- trophils and eosinophils to the sites of inflammation (38, 39). In agreement with the results of Lee et al. (39), we detected fold increases in mRNA expression of MMP-2 and MMP-9 in TDI-asthmatic mice, which were extensively sup- pressed by XAV-939 and ICG-001, indicating a possible b- catenin/MMPs-mediated neutrophilic and eosinophilic air- way inflammation pathway initiated by TDI. Besides, we also found that XAV-939 or ICG-001 decreased protein expression of HMGB1 and IL-1b in lungs of mice subjected to TDI sensitization of challenge, both of which are critical regulators in TDI-induced asthmatic responses (23, 40). This suggests that b-catenin might mediate airway inflammation through modulation of HMGB1 and IL-1b.In summary, our studies demonstrated that blocking b-catenin signaling attenuates AHR, inflammation, and remodeling in a TDI-induced murine asthma model, provid- ing novel evidence for targeting b-catenin in asthma.