Steroid-Insensitive Gene Expression of Extracellular Matrix Components and Pro-fibrotic Factors in the Lung Associated with Airway Hyperresponsiveness in Murine Asthma

Abstract

Between 5 and 10% of asthma patients do not respond to glucocorticoid therapy. Experimental animal models are indispensable for investigating the pathogenesis of steroid-resistant asthma; however, the majority of murine asthma models respond well to glucocorticoids. We previously reported that multiple intratracheal administration of ovalbumin (OVA) at a high dose (500 µg/animal) induced steroid-insensitive airway eosinophilia and remodeling with lung fibrosis, whereas a low dose (5 µg/animal) caused steroid-sensitive responses. The aims of the present study were as follows: 1) to clarify whether airway hyperresponsiveness (AHR) in the two models is also insensitive and sensitive to a glucocorticoid, respectively, and 2) to identify steroid-insensitive genes encoding extracellular matrix (ECM) components and pro-fibrotic factors in the lung. In comparisons with non-challenged group, the 5- and 500-µg OVA groups both exhibited AHR to methacholine. Daily intraperitoneal treatment with dexamethasone (1 mg/kg) significantly suppressed the development of AHR in the 5-µg OVA group, but not in the 500-µg OVA group. Among genes encoding ECM components and pro-fibrotic factors, increased gene expressions of fibronectin and collagen types I, III, and IV as ECM components as well as 7 matrix metalloproteinases, tissue inhibitor of metalloproteinase-1, transforming growth factor-β1, and activin A/B as pro-fibrotic factors were insensitive to dexamethasone in the 500-µg OVA group, but were sensitive in the 5-µg OVA group. In conclusion, steroid-insensitive AHR developed in the 500-µg OVA group and steroid-insensitive genes encoding ECM components and pro-fibrotic factors were identified. Drugs targeting these molecules have potential in the treatment of steroid-resistant asthma.

INTRODUCTION

Glucocorticoids administered by inhalation or an oral route are the effective treatment for asthma. However, 5–10% patients of asthma do not respond to glucocorticoid therapy.^1,2) Animal models are indispensable for investigating the mechanisms underlying the pathogenesis of steroid-resistant asthma and developing new anti-asthma drugs; however, the majority of murine models respond well to glucocorticoids.³⁾ Therefore, we established a murine model of steroid-insensitive asthma by intratracheal administration of a high dose (500 µg/animal) of ovalbumin (OVA) in sensitized mice, and compared the pathogenesis with a steroid-sensitive model, in which a low dose (5 µg/animal) of OVA was administered.⁴⁾ In the steroid-insensitive model, development of airway eosinophilia and airway remodeling was not suppressed by a daily treatment with dexamethasone.⁴⁾ On the other hand, the airway hyperresponsiveness (AHR) test is important for diagnosing asthma. AHR is generally assessed by exposing patients to increasing doses of a bronchoconstrictive agent, and measuring airway parameters.⁵⁾

Lung fibrosis has been suggested as one of the causes of AHR⁶⁾ because fibrotic changes in lung tissue induce a loss of elasticity. We previously reported the induction of lung fibrosis in the 500-µg OVA model, but not in the 5-µg OVA model, and showed that fibrosis was insensitive to dexamethasone.⁴⁾ Components of extracellular matrix (ECM), including collagen, fibronectin, elastin, and laminin, are deposited in fibrotic lung tissue.⁷⁾ The degradation and deposition of ECM are induced by matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs), respectively.⁸⁾ Additionally, pro-fibrotic factors of the transforming growth factor (TGF) β superfamily have been shown to promote the production of collagens through increases in myofibroblasts.⁹⁾ Therefore, the present study also examined steroid-insensitive genes encoding ECM components and pro-fibrotic factors in the lung.

The aims of the present study were as follows: 1) to clarify whether AHR in the 5- and 500-µg OVA models is also insensitive and sensitive to a steroid, respectively, and 2) to identify the steroid-insensitive genes encoding ECM components and pro-fibrotic factors in the lung. The present results will contribute to the identification of targeting molecules for the development of new pharmacotherapies for steroid-resistant asthma associated with lung fibrosis and AHR.

MATERIALS AND METHODS

Sensitization and Challenge

As previously reported,⁴⁾ 5-week-old BALB/c mice (Japan SLC, Hamamatsu, Japan) were sensitized by intraperitoneal injections of OVA (Grade V; Sigma-Aldrich, St. Louis, MO, U.S.A.) +Al(OH)₃ on days 0, 14, and 28. Sensitized mice were then intratracheally challenged with OVA at 5 or 500 µg/animal on days 35, 36, 37, and 40 under inhalation anesthesia with isoflurane. Dexamethasone 21-phosphate (Sigma-Aldrich) was intraperitoneally administered once a day during the OVA challenges at 1 mg/kg. This animal study was approved by the Experimental Animal Research Committee of Setsunan University, and performed in accordance with the experimental animal guidelines of Setsunan University and the Japanese Ministry of Education, Culture, Sports, Science and Technology.

Measurement of AHR

AHR to methacholine was measured by forced oscillation technique using FlexiVent (Scireq, Montreal, QC, Canada) according to our previous study.¹⁰⁾ Twenty-four hours after the 4th challenge, mice were intraperitoneally anesthetized by pentobarbital (70 mg/kg) and xylazine (12 mg/kg). Under artificial ventilation, mice were challenged with methacholine mist. After each methacholine challenge, resistance (Rrs), elasticity (Ers) and compliance (Crs) of the respiratory system were assessed.

Total RNA Sequencing and Bioinformatics

Twenty-four hours after the 4th challenge, lung tissue was isolated under the anesthesia. Extraction and sequencing of total RNA, and bioinformatics were conducted as previously reported.⁷⁾ Differences were considered to be significant when the q-value <0.05.

Statistical Analysis

A one-way ANOVA was performed. If significant differences were detected, individual group differences were examined by Dunnett’s test or the Tukey–Kramer test. The unpaired t-test was used for comparisons between two groups. A probability value (p) <0.05 was considered to be significant. Statistical analyses were conducted using JMP Pro 15.1.0 (SAS Institute Japan, Tokyo, Japan).

RESULTS

AHR clearly developed in both models (Fig. 1): In comparisons with the non-challenged group, 5- and 500-µg OVA model mice both showed airway responses to methacholine in all parameters from lower concentrations. The degree of AHR in the 2 models was similar. The treatment with dexamethasone significantly suppressed AHR for all parameters in the 5-µg OVA model, but not in the 500-µg OVA model.

Fig. 1. Airway Hyperresponsiveness (AHR) to Methacholine in 5- and 500-µg OVA-Induced Asthma Models in Sensitized Mice

Sensitized mice were intratracheally challenged with OVA at 5 or 500 µg/animal 4 times. Twenty-four hours after the 4th challenge, mice were anesthetized with pentobarbital/xylazine and the exposed trachea was connected to a ventilator. Increasing doses of methacholine mist were administered at concentrations ranging between 0 and 50 mg/mL. After each inhalation of methacholine, changes in respiratory functions, namely, the resistance (Rrs, A), elasticity (Ers, B) and compliance (Crs, C) of the respiratory system, were measured. Each thick bar represents the mean of 9–13 animals.

RNA-sequencing analyses (Table 1) showed that levels of Col1a2 and Col3a1, genes encoding collagen types I and III were higher in the 5- and 500-µg OVA groups than in the non-challenged. The up-regulations were significantly suppressed by dexamethasone in the 5-µg OVA group, but not in the 500-µg OVA group. Col4a1 expression was not increased in the 5-µg OVA group, but decreased by dexamethasone to a lower level than the non-challenged. However, Col4a1 expression was up-regulated in the 500-µg OVA group, and the up-regulation was not significantly affected by dexamethasone. In contrast, levels of Col4a3, Col4a4, Col4a5, and Col4a6 were decreased by the OVA challenges. As another component of ECM, the expression of the fibronectin 1 gene, Fn1 was also sensitive to dexamethasone in the 5-µg OVA group, whereas it was not sensitive or was up-regulated by dexamethasone in the 500-µg OVA group. Regarding the elastin gene, the up-regulated expression of Eln in the 500-µg OVA group was suppressed by dexamethasone. The gene expression patterns of laminin in the 5 groups were similar to those of Col4a3, Col4a4, Col4a5 and Col4a6.

Table 1. Gene Expression of Extracellular Matrix (ECM) Components

Gene symbol	Gene expression (FPKM)
	NC	5 µg OVA		5 µg OVA + DEX		500 µg OVA			500 µg OVA + DEX
			q-Value vs. NC		q-Value vs. 5 µg OVA		q-Value vs. NC	q-Value vs. 5 µg OVA		q-Value vs. 500 µg OVA
Collagen-related genes
Col1a1	7.75	19.6	0.0003	15.1	0.0025	37.6	0.0003	0.0003	31.6	0.0478
Col1a2	38.5	63.6	0.0003	50.8	0.0100	95.1	0.0003	0.0003	81.0	0.1041
Col3a1	73.0	254	0.0003	165	0.0003	317	0.0003	0.0925	291	0.6159
Col4a1	91.6	94.4	0.8532	63.5	0.0003	172	0.0003	0.0003	139	0.0850
Col4a2	61.7	53.0	0.1123	41.4	0.0034	101.2	0.0003	0.0003	81.4	0.0300
Col4a3	66.7	34.9	0.0003	33.0	0.6440	32.1	0.0003	0.4348	25.0	0.0034
Col4a4	20.2	10.4	0.0003	10.5	0.972	12.4	0.0003	0.0466	8.73	0.0003
Col4a5	34.9	18.8	0.0003	20.0	0.580	16.7	0.0003	0.2272	11.6	0.0003
Col4a6	0.636	0.304	0.0003	0.410	1.000	0.318	0.0003	1.0000	0.224	1.0000
Elastin-related gene
Eln	19.8	15.7	0.0067	12.5	0.0098	98.3	0.0003	0.0003	26.5	0.0003
Fibronectin-related gene
Fn1	32.8	155	0.0003	108	0.0010	202	0.0003	0.0477	279	0.0207
Laminin-related genes
Lama1	0.0377	0.0768	1.0000	0.0706	1.0000	0.149	1.0000	1.0000	0.0782	1.0000
Lama2	9.43	6.43	0.0003	7.27	0.1982	6.93	0.0003	0.4907	4.64	0.0003
Lama3	32.3	16.9	0.0003	18.5	0.3916	12.8	0.0003	0.0005	9.84	0.0016
Lama4	18.6	17.3	0.4830	18.6	0.4550	18.9	0.9073	0.3663	17.7	0.5632
Lama5	7.92	6.09	0.0012	7.04	0.1208	7.94	0.9863	0.0012	6.97	0.1581
Lamb1	19.0	14.5	0.0003	13.9	0.7305	19.7	0.7868	0.0003	15.8	0.0062
Lamb2	33.1	21.5	0.0003	24.7	0.1269	19.9	0.0003	0.4507	15.5	0.0031
Lamb3	22.5	15.2	0.0003	14.3	0.6031	15.6	0.0003	0.8714	12.7	0.0198
Lamc1	21.6	16.9	0.0027	17.1	0.9574	22.8	0.6354	0.0003	19.3	0.0516
Lamc2	37.6	30.3	0.0093	22.8	0.0003	39.9	0.5906	0.0010	36.3	0.3592
Lamc3	1.53	0.90	0.0003	1.03	0.3933	1.49	0.8883	0.0003	0.73	0.0003

The expressions of genes shown in bold font were suppressed by dexamethasone in the 5-µg OVA group, but not in the 500-µg OVA group. Genes detected at low levels of FPKM (less than 0.01, even in the 5-µg OVA group) are not shown. Each value represents the mean of 5 animals. NC: non-challenged.

Changes in the gene expression of pro-fibrotic factors were analyzed (Table 2). Among MMP genes, the expressions of Mmp3, Mmp12, Mmp13, Mmp14, Mmp17, Mmp19, and Mmp25 were up-regulated in both the 5- and 500-µg OVA groups. These increases in the 5-µg OVA group were significantly suppressed by dexamethasone, whereas those in 500-µg OVA group were not suppressed or were further enhanced by dexamethasone. The expression profile of the TIMP-1 gene, Timp1 was similar to those of the 7 MMP gene. Regarding the TGF-β superfamily, the expressions of Tgfb1 and Inhba, encoding the respective TGF-β1 and an activin subunit, were suppressed in the 5-µg OVA group and were not suppressed or were up-regulated in the 500-µg OVA group by dexamethasone. On the other hand, the expression of Elane, a neutrophil-derived elastase gene, did not significantly differ among the 5 groups.

Table 2. Gene Expression of Fibrosis-Related Molecules

Gene symbol	Gene expression (FPKM)
	NC	5 µg OVA		5 µg OVA + DEX		500 µg OVA			500 µg OVA + DEX
			q-Value vs. NC		q-Value vs. 5 µg OVA		q-Value vs. NC	q-Value vs. 5 µg OVA		q-Value vs. 500 µg OVA
Matrix metalloproteinase (MMP) genes
Mmp2	22.8	30.4	0.0003	26.8	0.1648	37.9	0.0003	0.0085	23.6	0.0003
Mmp3	3.76	6.31	0.0003	3.61	0.0003	13.4	0.0003	0.0003	13.0	0.8627
Mmp8	3.90	3.29	0.2043	2.77	0.2297	11.6	0.0003	0.0003	12.6	0.5523
Mmp9	2.67	3.39	0.0465	3.22	0.7448	3.79	0.0012	0.4221	4.31	0.3378
Mmp11	12.9	8.42	0.0003	9.58	0.2724	7.99	0.0003	0.7199	6.12	0.0138
Mmp12	2.59	320	0.0003	88.2	0.0003	369	0.0003	0.3072	325	0.3890
Mmp13	0.240	2.85	0.0003	0.581	0.0003	8.34	0.0003	0.0003	12.2	0.0003
Mmp14	6.92	15.3	0.0003	10.1	0.0003	29.7	0.0003	0.0003	29.0	0.8451
Mmp15	14.7	9.14	0.0003	10.9	0.0528	8.23	0.0003	0.3202	6.42	0.0069
Mmp16	0.289	0.291	1.0000	0.238	1.0000	0.300	1.0000	1.0000	0.265	1.0000
Mmp17	0.372	0.528	0.0881	0.341	0.0282	0.542	0.0609	0.9387	0.544	0.9920
Mmp19	9.26	59.6	0.0003	35.5	0.0003	69.4	0.0003	0.1175	76.8	0.3519
Mmp21	0.103	0.214	1.0000	0.213	1.0000	0.197	1.0000	1.0000	0.230	1.0000
Mmp23	8.04	7.88	0.9170	8.30	0.7545	8.31	0.8613	0.7526	6.91	0.1339
Mmp24	0.127	0.078	1.0000	0.091	1.0000	0.113	1.0000	1.0000	0.093	1.0000
Mmp25	0.368	6.06	0.0003	3.83	0.0003	5.98	0.0003	0.9394	7.45	0.0246
Mmp27	0.268	0.380	1.0000	0.306	1.0000	1.03	0.0003	0.0003	0.577	0.0081
Mmp28	4.04	2.11	0.0003	1.73	0.2105	1.63	0.0003	0.0959	1.01	0.0034
Tissue inhibitors of metalloproteinase (TIMP) genes
Timp1	5.76	191	0.0003	86.0	0.0003	328	0.0003	0.0309	660	0.0044
Timp2	83.9	71.6	0.0859	72.7	0.9240	82.3	0.8967	0.1441	65.8	0.0119
Timp3	289	160	0.0003	184	0.2387	173	0.0003	0.5763	160	0.5398
Timp4	1.60	2.89	0.0086	2.45	0.5873	2.16	0.3298	0.2983	2.29	0.8748
Transforming growth factor (TGF)-β superfamily genes
Tgfb1	4.66	9.92	0.0003	7.59	0.0090	12.6	0.0003	0.0152	15.6	0.0272
Tgfb2	5.58	2.67	0.0003	3.79	0.0003	2.81	0.0003	0.7282	3.06	0.5266
Tgfb3	9.38	6.53	0.0003	6.82	0.7673	8.31	0.2620	0.0141	6.11	0.0016
Inhba	0.337	4.30	0.0003	1.64	0.0003	8.41	0.0003	0.0003	9.72	0.2598
Inhbb	2.71	2.64	0.8920	2.73	0.8417	4.86	0.0003	0.0003	4.76	0.9110
Elastase (neutrophil-derived) gene
Elane	0.027	0.142	1.0000	0.015	1.0000	0.151	1.0000	1.0000	0.044	1.0000

The expressions of genes shown in bold font were suppressed by dexamethasone in the 5-µg OVA group, but not in the 500-µg OVA group. Genes detected at low levels of FPKM (less than 0.01, even in the 5-µg OVA group) are not shown. Each value represents the mean of 5 animals. NC: non-challenged.

DISCUSSION

The present results clearly showed AHR in both the 5- and 500-µg OVA groups, which was sensitive and insensitive to dexamethasone, respectively. Among the airway physiological parameters examined herein, Ers and Crs reflect the rigidity and elasticity and of the respiratory system, respectively. Since lung fibrosis induced in the 500-µg OVA model⁷⁾ is related to the rigidity and elasticity of the lung, the present study focused on the accumulation of ECM components and production of pro-fibrotic factors in the lung, and identified increases in the expression of steroid-insensitive genes, thereby providing insights into the candidate molecules responsible for the induction of steroid insensitivity. The results showed that the increased expressions of genes encoding collagens and fibronectin-1, and 7 MMPs, TIMP-1, TGF-β1, and activin A/B were insensitive to dexamethasone in the 500-µg OVA model, but sensitive in the 5-µg OVA group. Therefore, those ECM components and pro-fibrotic factors may be the candidate molecules.

Consistent with the present results, the expression of the ECM components, collagen types I, III, and IV, and fibronection-1 in the lung was up-regulated in asthma patients.⁷⁾ Furthermore, the mRNAs of collagen types I and III were up-regulated in the lung of a murine asthma model.¹¹⁾ The deposition of ECM components in the airway wall, such as the basement membrane and interstitial matrix, is regulated by various pro-fibrotic factors. Among them, the balance between MMP-9 and TIMP-1 is informative; when the expression of TIMP-1 increases over that of MMP-9, ECM deposition progresses, resulting in the development of fibrosis.⁸⁾ In the present study, the gene expression (FPKM) of Timp1 was 50- to 150-fold higher than that of Mmp9, indicating the predominant role of TIMP-1 in the accumulation of collagens and fibronectin-1.

Among MMP genes, the expression of Mmp12, a gene encoding macrophage elastase MMP-12,¹²⁾ was the highest. A literature reported that MMP-12 was produced from airway epithelial cells in a murine asthma model, and an MMP-12 inhibitor effectively suppressed the development of AHR and lung fibrosis.¹³⁾ The up-regulation of Mmp12 in our model was also insensitive to dexamethasone in the 500-µg OVA model. Regarding the substrate of elastases, the increase in the expression of Eln, an elastin gene, was significantly reduced by dexamethasone in the 500-µg OVA model. The degradation of elastin may be due to the up-regulated expression of MMP-12 in the model. In contrast, Elane, the gene encoding neutrophil-derived elastase, did not appear to be expressed in the models, implying the minor role of neutrophils in the development of lung fibrosis.

The following mechanisms have been proposed for the development of steroid insensitivity: 1) a decrease in the expression of glucocorticoid receptor (GR) α, 2) an impairment in GRα translocation into the nucleus, 3) the expression of GRβ, a dominant negative for GRα, and 4) decreases in histone deacetylase.¹⁴⁾ In our steroid-insensitive model, the gene encoding anti-apoptotic factor, Bcl-xL was up-regulated in the lung (data not shown). Therefore, glucocorticoid-mediated apoptosis in immune cells may have been blocked by Bcl-xL.

In conclusion, steroid-insensitive AHR developed in the 500-µg OVA group. The expressions of genes encoding collagen types I, III, and IV, fibronection-1, 7 MMPs, TIMP-1, TGF-β1, and activins were insensitive to the steroid treatment in the 500-µg OVA group. The pro-fibrotic molecules may be responsible for the development of steroid insensitivity. The targeting of these molecules may be useful in the development of new pharmacotherapies for steroid-resistant asthma.

Acknowledgments

This study was partly supported by JSPS KAKENHI Grant Number: 20K07301 (to T.N.), and Platform Project for Supporting in Drug Discovery and Life Science Research (Basis for Supporting Innovative Drug Discovery and Life Science Research, BINDS) from Japan Agency for Medical Research and Development (AMED).

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Holgate ST, Polosa R. The mechanisms, diagnosis, and management of severe asthma in adults. Lancet, 368, 780–793 (2006).
• 2) Nabe T. Steroid-resistant asthma and neutrophils. Biol. Pharm. Bull., 43, 31–35 (2020).
• 3) De Bie JJ, Hessel EM, Van Ark I, Van Esch B, Hofman G, Nijkamp FP, Van Oosterhout AJ. Effect of dexamethasone and endogenous corticosterone on airway hyperresponsiveness and eosinophilia in the mouse. Br. J. Pharmacol., 119, 1484–1490 (1996).
• 4) Matsuda M, Tanaka Y, Shimora H, Takemoto N, Nomura M, Terakawa R, Hashimoto K, Sakae H, Kanda A, Iwai H, Kitatani K, Nabe T. Pathogenic changes in group 2 innate lymphoid cells (ILC2s) in a steroid-insensitive asthma model of mice. Eur. J. Pharmacol., 916, 174732 (2022).
• 5) Kraemer R, Smith H-J, Sigrist T, Giger G, Keller R, Frey M. Diagnostic accuracy of methacholine challenge tests assessing airway hyperreactivity in asthmatic patients—a multifunctional approach. Respir. Res., 17, 154 (2016).
• 6) Roche WR, Beasley R, Williams JH, Holgate ST. Subepithelial fibrosis in the bronchi of asthmatics. Lancet, 333, 520–524 (1989).
• 7) Gu B-H, Madison MC, Corry D, Kheradmand F. Matrix remodeling in chronic lung diseases. Matrix Biol., 73, 52–63 (2018).
• 8) Vignola AM, Riccobono L, Mirabella A, Profita M, Chanez P, Bellia V, Mautino G, D’accardi P, Bousquet J, Bonsignore G. Sputum metalloproteinase-9/tissue inhibitor of metalloproteinase-1 ratio correlates with airflow obstruction in asthma and chronic bronchitis. Am. J. Respir. Crit. Care Med., 158, 1945–1950 (1998).
• 9) Hardy CL, Rolland JM, O’Hehir RE. The immunoregulatory and fibrotic roles of activin A in allergic asthma. Clin. Exp. Allergy, 45, 1510–1522 (2015).
• 10) Matsuda M, Morie Y, Oze H, Doi K, Tsutsumi T, Hamaguchi J, Inaba M, Nabe T. Phenotype analyses of IL-10-producing Foxp3⁻ CD4⁺ T cells increased by subcutaneous immunotherapy in allergic airway inflammation. Int. Immunopharmacol., 61, 297–305 (2018).
• 11) Kenyon NJ, Ward RW, Last JA. Airway fibrosis in a mouse model of airway inflammation. Toxicol. Appl. Pharmacol., 186, 90–100 (2003).
• 12) Lagente V, Le Quement C, Boichot E. Macrophage metalloelastase (MMP-12) as a target for inflammatory respiratory diseases. Expert Opin. Ther. Targets, 13, 287–295 (2009).
• 13) Ravanetti L, Dekker T, Guo L, Dijkhuis A, Dierdorp BS, Diamant Z, Florquin S, Lutter R. Efficacy of FP-025: a novel matrix metalloproteinase-12 (MMP-12) inhibitor in murine allergic asthma. Allergy, 78, 559–562 (2023).
• 14) Barnes PJ. Corticosteroid resistance in patients with asthma and chronic obstructive pulmonary disease. J. Allergy Clin. Immunol., 131, 636–645 (2013).