Sage Journals: Discover world-class research

Abstract

Background

Allergic rhinitis (AR) is a common immunoglobulin E-mediated immune response involved various cell types, while the role of nasal fibroblasts (NFs) in the pathogenesis of AR is less understood.

Purpose

The study aimed to uncover the gene expression profile of AR-derived NFs and the potential mechanism for the changed phenotype of AR-NFs.

Research Design

The primary NFs were isolated from 3 AR patients (AR-NFs) and 3 controls (Ctrl-NFs), and the proliferation, migration and interleukins production abilities of NFs were detected respectively. RNA-sequence was used to identify differentially expressed genes (DEGs) in AR-NFs. Transcription factor (TF) regulatory network and bioinformatic analyses were both conducted to clarify the biological roles of DEGs including the TFs. The DEG with the highest validated |fold change (FC)| value, detected by qPCR, was selected for further confirmation.

Results

AR-NFs showed a higher proliferation and migration abilities as well as released higher levels of IL-33 and IL-6, compared to Ctrl-NFs. A total of 729 DEGs were screened out in AR-NFs. TF regulatory network indicated that BARX homeobox 1 (BARX1) and forkhead box L1 were the major node TFs. Bioinformatic analyses showed that a large number of DEGs including several target genes of BARX1 were both enriched cytokine-related GO terms, and immune- or inflammation-related pathways. BARX1 had the highest |FC| value, and silencing BARX1 in AR-NFs resulted in the significant downregulation of proliferation and migration abilities, and the production of interleukins.

Conclusions

Our study for the first time provided the gene expression profile of AR-derived NFs, and BARX1 could be developed as a potent target to alleviate the pathogenesis of AR.

Keywords

RNA-seq nasal fibroblasts allergic rhinitis BARX1 IL-33

Introduction

Allergic rhinitis (AR) is a common non-infectious inflammatory disease affecting 10–40% population worldwide.¹ AR is a specific immunoglobulin E (IgE)-mediated immune response against inhaled allergens, accompanied with classical symptoms including rhinorrhea, sneezing and nasal congestion, thus significantly affects quality of life and productivity at work or school.² Drug therapies such as anti-histamines, leukotriene receptor antagonists and nasal sprays, aiming to target mediators, cytokines, or nonspecific inflammation to attenuate allergic symptoms, are still current major treatment for AR patients.^3,4 Recent studies indicate that other non-immune cells, such as epithelial cells and fibroblasts, also play important roles in the allergic response. It is reported that, epithelial cell-derived cytokines, including thymic stromal lympho-poietin (TSLP), interleukin (IL)-25 and -33, are able to regulate the function of dendritic cells (DCs), T cells, and mast cells, resulting in the production of proinflammatory cytokines and even the initiation of type-2 immune responses.⁵ In addition, impairment of tight junctions molecules also plays causative roles in the pathogenesis of AR.⁶

Nasal fibroblasts (NFs) are an important cell component of nasal mucosa, and also are another notable population of non-immune cells in the pathogenesis of AR. Several studies reveal that NFs are not only the effector cells against allergens but also the mediators response to various inflammatory cytokines. For instance, exposure of NFs to urban particulate matter and diesel exhaust particles causes the upregulation of IL-6 and IL-8,^7,8 two important modulators of local inflammation.⁹ IL-6 is also thought to amplify the allergic inflammatory response.¹⁰ What is more, treatment of histamine, predominantly released from mast cells, also induces the release of IL-6 in NFs.¹¹ IL-17A and IL-21 are both involved in the pathogenesis of AR.^12,13 It is reported that, stimulation of NFs from AR patients with IL-17A significantly up-regulates the production of TSLP.¹⁴ Administration of IL-21 inhibits eotaxin release from NFs in murine AR models.¹⁵ The important role of NFs in the development of AR indicates that, target regulating the function of NFs is a necessary and vital research direction in the multi-targets treatment of AR. The current studies have uncovered relative pathways activated in the effector NFs,^8,16 but the deeper mechanism was not well understood.

In the present study, we aimed to investigate the phenotype and gene expression profile alterations of NFs isolated from AR patients (AR-NFs) compared to the control NFs (Ctrl-NFs). The biological roles of the differentially expressed genes (DEGs) in AR-NFs were then annotated by bioinformatic analyses, and transcription factors (TFs) regulatory network was also constructed. Finally, the role of interested DEG, BARX1 (BARX homeobox 1), playing important roles in tissue development and carcinogenesis,^17,18 on phenotype of NFs was selected for further verification.

Materials and methods

Patients and specimens

Inferior turbinate mucosa tissue from six donors was collected for the isolation of NFs. Three donors (2 males, 1 female; mean age 32.67 ± 6.03) diagnosed with persistent AR complicated with turbinate hypertrophy were sorted as the AR group, while another three donors (2 females, 1 male; 28.67 ± 4.51) without AR and required a corrective septorhinoplasty was recruited as the Ctrl group. None of the patients had a history of asthma, or infections of the respiratory tract. For AR patients, only those without the use of corticosteroids, antibiotics, leukotriene modifiers and anti-histamines within a month before surgery were included. The fresh specimen was reserved in sterile phate-buffered saline (PBS) and transported to the laboratory for cell isolation within 3 h. This study was approved by the Ethics Committee of Shanghai Tenth Peoples’ Hospital, Tongji University, and written informed consent was obtained from all patients. All procedures conformed to the principles of the Declaration of Helsinki.

Nasal fibroblasts isolation and culture

The isolation of NFs was carried out following previous studies^8,19 with some modifications. Fresh specimens were first washed with PBS supplemented with antibiotic-antimycotic solution (Gibco, Gaithersburg, MD) for three times to avoid contamination. Subsequently, the specimens were cut into small pieces (1–3 mm²) and digested in collagenase solution (500 U/mL, Sigma) at 37°C for 60 min. Next, the cells were collected by centrifugation at 1000 g for 5 min after the dissociated specimens were filtrated. The resulted NFs were cultured in DMEM medium containing 10% of fetal bovine serum (FBS, Invitrogen, Carlsbad, CA, USA), and 1% of penicillin/streptomycin (Invitrogen). The medium was changed every 3 days. All the cells were incubated at 37°C with 5% CO₂.

Proliferation assay

Cells were collected and resuspended at a density of 3 × 10⁴ cells/ml, and 100 µl of cells was then seeded into each well of the 96-well plate. Cell viability was detected daily for 7 days by MTT assay. Briefly, 10 µl of 5 mg/mL MTT solution (Sigma-Aldrich; Merck KGaA, Darmstadt, German) was added to each well and the cells were incubated at 37°C for 3 h. The absorbance of each well was detected at 490 nm after the formazan was totally resolved in the DMSO solution. Proliferation rate was calculated using the following formula: proliferation rate = A(day n)/A(day 1), n = 1–7.

Transwell assay

Briefly, cells at a density of 4 × 10⁵ cells/mL was resuspended in the FBS-free medium. Then 100 µl of the cells were seeded into the upper chamber of the transwell (Costar, Cambridge, USA), and another 600 µL complete medium was added to the lower chamber. After 24 h, cells retained in the inner side of the membrane were removed, and those migrated through the membrane were fixed with methanol for 20 min and stained with crystal violet solution for another 20 min. Finally, the visual fields of each chamber were randomly selected and photographed by an inverted microscope.

Enzyme-linked immunosorbent assay

Cell supernatant was collected to detect the level of IL-33 and IL-6 using human IL-33, IL ELISA kits (ABclonal, Wuhan, China), following the manufacturer’s instructions. Optical density values were measured at 450 nm. The concentration of each cytokines was calculated using standard curve of corresponding recombinant protein.

qPCR

Total mRNA of cells was extracted using a Universal RNA Extraction Kit (TAKARA, Dalian, China) and cDNA was synthesized using a RT-PCR Quick Master Mix Kit (Toyobo, Osaka, Japan) following the separate protocols. SYBR Premix Ex Taq^™ II kit (TAKARA) was used to prepare the qPCR reaction system (20 µL), containing 1 µL of cDNA, 0.5 µL of each primer, 10 µL of SYBR Green, and 8 µL of deionized water. Thermal cycling was as follows: 94°C for 300 s, then 40 of cycles at 94°C for 5 s, and 60°C for 60 s. Data were analyzed using the 2^−ΔΔCt method and normalized to GAPDH levels. Sequences of gene primers were synthesized by Sangon Biotech Co., Ltd., (Shanghai, China) and presented in Table 1.

Table 1.

Primer sequence used for qPCR.

Genes	Primer sequences
IL-33	Forward	GTGACGGTGTTGATGGTAAGAT
IL-33	Reverse	AGCTCCACAGAGTGTTCCTTG
IL-6	Forward	CCTGAACCTTCCAAAGATGGC
IL-6	Reverse	TTCACCAGGCAAGTCTCCTCA
BARX1	Forward	TTCCACGCCGGACAGAATAGA
BARX1	Reverse	AGTAAGCTGCTCGCTCGTTG
TMEM176B	Forward	ATGACGCAAAACACGGTGATT
TMEM176B	Reverse	GCAGTTGTGTCAAAGCTGACT
GRIK5	Forward	GATCAACGGGATCATCGAGGT
GRIK5	Reverse	GTGTCCGTGGTCTCGTACTG
LAMP3	Forward	GCGTCCCTGGCCGTAATTT
LAMP3	Reverse	TGCTTGCTTAGCTGGTTGCT
AMTN	Forward	TGTCTTCTAGGATCAACTCGGT
AMTN	Reverse	TGGTTTGGTAGTGTTCCCTGA
GAPDH	Forward	GGAGCGAGATCCCTCCAAAAT
GAPDH	Reverse	GGCTGTTGTCATACTTCTCATGG

Ribonucleic Acid (RNA)-sequence

Total RNA of cells in AR (n = 3) and Ctrl (n = 3) groups was collected by Trizol reagent (Invitrogen, Carlsbad, CA, USA), and isolated using RNeasy mini kit (Qiagen, Germany). Complimentary DNA (cDNA) libraries were synthesized following the protocols of TruSeq^™ RNA Sample Preparation Kit (Illumina, USA), which included mRNA purification and fragmentation, cDNA synthesis and end-repair, as well as the PCR enrichment. Next, the libraries were quantified by Qubit^®2.0 Fluorometer (Life Technologies, USA) and validated by Agilent 2100 bioanalyzer (Agilent Technologies, USA) to confirm the insert size and calculate the mole concentration. Cluster was generated by cBot with the library diluted to 10 p.m. and then sequenced on the Illumina NovaSeq 6000 (Illumina). Gene abundance was expressed as fragments per kilobase of exon per million reads mapped (FPKM). Stringtie software was used to count the fragment within each gene, and TMM algorithm was used for normalization. Differentially expressed genes (DEGs) were screened under the thresholds of |fold change (FC)| value >1.5 and p value < 0.05.

Bioinformatic analyses

Gene Ontology (GO) analysis (https://david.ncifcrf.gov/home.jsp) for biological processes, cellular components and molecular function and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis (http://www.genome.ad.jp/kegg) were performed, respectively, to annotated the biological role of the DEGs. Enrichment analysis was further performed and visualized using R package.

Cell transfection

AR-NFs seeded in 6-well with a confluence of 80% was prepared, then small interference RNA targeting BARX1 (si-BARX1) was transfected into the cells using lipofectamine 2000 (Invitrogen) following the instructions. After 48 h, cells were collected for qPCR verification. si-BARX1 (5’-GCU GGA GAA ACG CUU CGA GAA-3’) and the negative control (si-NC, 5’-UUC UCC GAA CGU GUC ACG UTT-3’) were synthesized by Geneseed Biotech (Guangzhou, China).

Statistical analysis

Results were presented as mean ± standard deviation. The statistical significance of differences between two groups was assessed unpaired t-test or one-way analysis of variance (ANOVA). Triplicate wells were prepared for each condition in each experiment. p < 0.05 was considered statistically significant.

Results

Phenotype detection of allergic rhinitis-nasal fibroblasts in proliferation, migration abilities, and production of cytokines

As shown in Figure 1(a), NFs isolated from two groups were both shuttle-shaped, and AR-NFs were slender spindle in morphology. MTT results showed that the proliferation rate of AR-NFs was significantly enhanced compared to the controls (Figure 1(b)). Transwell assay indicated that AR group-derived NFs showed a higher migration ability than Ctrl-NFs (Figure 1(c)). In addition, the supernatant IL-33 and IL-6 levels of AR-NFs were both significantly higher than those of the controls according to the ELISA analysis result (Figure 1(d)). Further, qPCR detection indicated that the mRNA expression levels of the two interleukins were also significantly enhanced in AR-NFs (Figure 1(e)), consistent to the ELISA result.

Figure 1.

Phenotype detection of AR-NFs in proliferation, migration abilities and production of cytokines. (A) Cellular morphology of NFs; (B) MTT assay was used to detect the proliferation rate of NFs; (C) Transwell assay was used to evaluate the migration ability; (D) ELISA assay was performed to detect supernatant IL-33 and IL-6 of NFs; (E) qPCR was used to detect mRNA expression levels of IL-33 and IL-6 in NFs. AR: allergic rhinitis; Ctrl: control; NFs: nasal fibroblasts. *p < 0.05.

RNA-seq analysis of differentially expressed genes in allergic rhinitis-derived nasal fibroblasts

A total of 729 DEGs, including 341 upregulated (red dots) and 388 down-regulated genes (blue dots), were identified in AR-NFs compared to Ctrl-NFs by RNA-seq in the thresholds of |FC| value >1.5 and p value < 0.05 (Figure 2(a)). |FC| value of most DEGs ranged from 1.5–4. A small amount of the DEGs was over eight and the dots could be found in the outer vertical border of Figure 2(a). The heap-map of DEG presented the expression profile of each NF sample, and the overall gene expression trend of three samples in each group was basically consistent (Figure 2(b)). The FC value of the top 20 DEGs were also listed in Table 2. As we could see, 70% of the DEGs were significantly upregulated, and BARX1 had the highest FC value (1705.1) (Table 2). In addition, the FC value of IL-33 reached 20.305, consistent with the increased production of IL-33 in ELISA result.

Figure 2.

RNA-seq analysis of DEGs in AR-NFs. (A) RNA-seq was used to screen DEGs in AR-NFs compared to Ctrl-NFs under the thresholds of |FC| value >1.5 and p value < 0.05, and volcano plot presented all the upregulated (red dots) and down-regulated genes (blue dots); (B) The fold change of all DEGs in each sample were presented in the heat-map. AR: allergic rhinitis; Ctrl: control; NFs: nasal fibroblasts. DEGs: differentially expressed genes.

Table 2.

Top 20 differentially expressed genes in allergic rhinitis-nasal fibroblasts.

	Gene ID	Gene name	Locus	log2FC	log2FC abs	FC abs	p value	Up/down
1	ENSG00000131668	BARX1	9:93,951,627–93,955,355	10.736	10.736	1705.109	2.810E-59	Up
2	ENSG00000106565	TMEM176 B	7:1,50,791,285–150,801,360	6.377	6.377	83.088	2.062E-07	Up
3	ENSG00000105737	GRIK5	19:41,998,321–42,069,498	5.644	5.644	49.988	5.863E-07	Up
4	ENSG00000078081	LAMP3	3:1,83122215–183,163,839	−5.520	5.520	45.876	3.170E-08	Down
5	ENSG00000187689	AMTN	4:70,518,569–70,532,743	5.274	5.274	38.695	7.495E-04	Up
6	ENSG00000285130	AL358113.1	9:69,035,747–69,255,187	5.197	5.197	36.685	2.009E-05	Up
7	ENSG00000257390	AC023055.1	12:55,757,275–55,827,546	4.843	4.843	28.704	3.726E-04	Up
8	ENSG00000243137	PSG4	19:43,192,702–43,207,299	−4.763	4.763	27.144	9.986E-03	Down
9	ENSG00000102683	SGCG	13:23,180,979–23,325,162	−4.750	4.750	26.906	3.391E-03	Down
10	ENSG00000002745	WNT16	7:1,21,325,367–121,341,104	−4.646	4.646	25.041	1.319E-05	Down
11	ENSG00000288634	AC079741.2	7:1,12,450,487–112,450,645	−4.519	4.519	22.934	3.908E-03	Down
12	ENSG00000205403	CFI	4:1,09,740,694–109,802,179	4.405	4.405	21.181	4.322E-08	Up
13	ENSG00000267368	UPK3BL1	7:1,02,637,025–102,642,791	−4.401	4.401	21.134	2.795E-04	Down
14	ENSG00000137033	IL33	9:6,215,786–6,257,983	4.344	4.344	20.305	1.007E-06	Up
15	ENSG00000272398	CD24	6:1,06,969,831–106,975,627	4.178	4.178	18.097	1.183E-04	Up
16	ENSG00000179772	FOXS1	20:31,844,303–31,845,604	4.075	4.075	16.856	8.115E-03	Up
17	ENSG00000240764	PCDHGC5	5:1,41,489,081–141,512,975	3.971	3.971	15.685	2.908E-03	Up
18	ENSG00000156150	ALX3	1:1,10,059,870–110,070,672	3.916	3.916	15.092	8.462E-08	Up
19	ENSG00000160180	TFF3	21:42,311,667–42,315,651	3.894	3.894	14.866	1.608E-05	Up
20	ENSG00000104435	STMN2	8:79,611,117–79,666,158	3.700	3.700	12.994	3.275E-04	Up

Transcription factor regulatory network of differentially expressed genes

For a better understanding the connections between those DEGs, TF regulatory network was constructed. As we could see in Figure 3, the network covered seven of the TFs and 41 of the targeted DEGs. And one of the TFs, BARXs, had 13 of target genes, including, CDK11A, LEPR, CTH, GBP2, Vascular cell adhesion molecule 1 (VCAM1), PTGS2, and PPP1R12B, et al. Another TF forkhead box L1 (FOXL1), with the log2 FC = −1.89, shared eight target genes with BARX1, and also regulated the transcription of another 19 genes.

Figure 3.

TF regulatory network of DEGs. TF: transcription factors; DEGs: differentially expressed genes.

Bioinformatic analyses of differentially expressed genes

Gene ontology enrichment analysis showed that many DEGs were involved in protein synthesis, including protein folding in endoplasmic reticulum, ATF6-mediated unfolded protein response, and PERK-mediated unfolded protein response (Figure 4). Some DEGs were enriched in positive regulation of vascular associated smooth muscle cell migration, and transforming growth factor beta receptor binding (Figure 4). Since little relation of the top 30 GO terms could be directly connected with the production of the cytokines, the keywords, “cytokine,” “chemokine,” and “immune,” were used to screen relevant biological processes and DEGs. As a result, six GO terms were screened out under the threshold of p value < 0.05 (Table 3). DEGs enriched in cytokine-related processes (GO:0034097, GO:0071345) reached ∼90–100, including some chemokine/cytokine and the receptors, IL33, CXCL6, IL7R, and CCR10, as well as some targets of BARX1, such as VCAM1, GBP2, CTH, PTGS2, and LEPR (Table 3). Additionally, 64 DEGs were involved in immune system development, including BARX1, and IL7R, and VCAM1 (Table 3). GO classification and enrichment analyses of all DEGs were presented in Supplementary Table S1.

Figure 4.

Top 30 GO terms of DEGs enriched in. circle, triangle, and square, represents biological process, cellular component, and molecular function. The size of the icon indicates gene number enriched in the GO term. The color of each icon indicates the q value. DEGs: differentially expressed genes.

Table 3.

Differentially expressed genes enriched in relevant GO terms.

Key words	GO ID	GO term	Type	Up_ gene	Down_gene	gene_Up_list	gene_Down_list	rich_factor	p value
Cytokine	GO:0019955	Cytokine binding	Molecular function	10	5	TGFB3, PDPN, CCR10, KIT, A2M	GREM1, ITGA4, TGFBR3, TRIM16, LEPR	2.39	.0095
	GO:0034097	Response to cytokine	Biological process	63	35	TUBA1B, IL7R, BIRC5, HLA-C, KRT18, GBP4, SFRP1, VCAM1, CCR10, BST2, IFNGR2, MX2, MYBL2, MX1, TRIM62, RAB43, FOXC1, SSTR1, MMP3, IFI6, ACP5, ISG15, KIT, GBP2, CXCL6, COL3A1, IFITM1, TNFRSF6B, CARD16, IL33, SOCS2, OAS3, GGT5, LAMA5, CD24	CTH, GPER1, IFI30, TIMP3, EREG, LAMP3, FAS, EDA2R, ITGA4, HK2, SLIT3, PDIA3, NFAT5, LOX, LEPR, SERPINB2, YTHDF3, HSP90B1, EGR1, KLF4, UBE2V1, GSDME, PPARG, JUN, MME, XBP1, PTGS2, HSPA5	1.64	.0002
	GO:0071345	cellular response to cytokine stimulus	Biological process	57	32	IFNGR2, CCR10, BST2, MX2, MX1, MYBL2, IL7R, BIRC5, TUBA1B, VCAM1, KRT18, HLA-C, SFRP1, GBP4, SOCS2, IL33, OAS3, CARD16, TNFRSF6B, IFITM1, CD24, LAMA5, MMP3, SSTR1, RAB43, TRIM62, FOXC1, CXCL6, IFI6, GBP2, KIT, ISG15	ITGA4, EDA2R, FAS, PDIA3, HK2, SLIT3, IFI30, GPER1, CTH, EREG, MME, XBP1, PTGS2, HSPA5, YTHDF3, SERPINB2, EGR1, HSP90B1, LOX, NFAT5, LEPR, UBE2V1, PPARG, GSDME, KLF4	1.60	.0009
	GO:0019221	Cytokine-mediated signaling pathway	Biological process	39	26	FOXC1, TRIM62, BIRC5, MMP3, IL7R, ISG15, KIT, GBP2, HLA-C, IFI6, KRT18, VCAM1, CXCL6, IFITM1, TNFRSF6B, CARD16, BST2, CCR10, IFNGR2, OAS3, IL33, SOCS2, MX1, LAMA5, MX2, CD24	LEPR, HSP90B1, EGR1, SERPINB2, YTHDF3, IFI30, EREG, PPARG, UBE2V1, FAS, EDA2R, SLIT3, PTGS2	1.50	.0148
Chemokine	GO:0019956	Chemokine binding	Molecular function	4	3	A2M, PDPN, CCR10	ITGA4	4.34	.0069
Immune	GO:0002520	Immune system development	Biological process	45	19	BARX1, FSTL3, SOX4, TSPAN2, THRA, IL7R, VCAM1, TCTA, SIX1, SFRP1, TCF12, SLC40A1, PDPN, TMEM176 B, FLVCR1, FOXC1, HLX, ISG15, KIT	ITGA4, MKNK2, FLT1, H4C8, STK4, CD79 B, TGFBR3, RBP1, DNAJB9, EVI2B, CEBPG, JUN, XBP1, CMTM7, EXOSC6, EGR1, LOX, SYVN1, LEPR, MFAP5, FOXL1, RASSF2, PPARG, TOX, KLF4, DYRK3	1.48	.0108

Kyoto encyclopedia of genes and genomes classification indicated that many DEGs were involved in immune and endocrine systems, and some DEGs were involved in cancer-related disease in the class of human disease, which might be related to the enhanced proliferation and migration abilities of AR-NFs (Figure 5). Moreover, 80 DEGs, ∼11% of all DEGs, were involved in signal transduction (Figure 5), indicating a comprehensive dysregulation of pathways in AR-derived NFs, compared to the controls. DEGs enriched in 20 sub-pathways of immune system, 23 sub-pathways of signal transduction, and another four pathways we interested in were listed in Table 4. PTGS2 and VCAM1 were both involved in NF-kappa B (NF-κB) and TNF signaling pathways, and protein phosphatase 1 regulatory subunit 12B (PPP1R12B) was associated with regulation of actin cytoskeleton, which belongs to cell motility pathway (Table 4). Enrichment analysis showed that, many DEGs were enriched in protein processing in endoplasmic reticulum and ECM-receptor interaction, and some of them were involved in hippo signaling pathway-multiple species and p53 signaling pathway (Supplementary Figure S1). KEGG classification and enrichment analyses of all DEGs were provided in Supplementary Table S2.

Figure 5.

KEGG classification of all DEGs. All the KEGG pathways were sorted into six groups marked in different colors. The length of each bar indicates gene number enriched in the pathway. DEGs: differentially expressed genes.

Table 4.

Differentially expressed genes enriched in relevant KEGG pathways.

	Pathway_3	Pathway_3_num	All_diff_gene_list	Pathway_2
1	B Cell receptor signaling pathway	3	JUN, IFITM1, CD79 B	Immune system
2	Complement and coagulation cascades	4	SERPINB2, CFI, F2RL2, A2M	Immune system
3	RIG-I-like receptor signaling pathway	2	ISG15, PIN1	Immune system
4	Leukocyte transendothelial migration	4	CLDN11, VCAM1, ITGA4, F11 R	Immune system
5	Intestinal immune network for IgA production	3	CCL28, ITGA4, CCR10	Immune system
6	Th17 cell differentiation	2	JUN, IFNGR2	Immune system
7	Platelet activation	3	VAMP8, COL3A1, GUCY1A2	Immune system
8	Chemokine signaling pathway	4	CXCL6, SHC2, CCL28, CCR10	Immune system
9	Th1 and Th2 cell differentiation	2	JUN, IFNGR2	Immune system
10	IL-17 signaling pathway	5	HSP90B1, CXCL6, JUN, MMP3, PTGS2	Immune system
11	Antigen processing and presentation	5	PDIA3, HLA-C, CALR, IFI30, HSPA5,	Immune system
12	NOD-like receptor signaling pathway	10	TRPV2, JUN, OAS3, CARD16, ANTXR2, GABARAPL1, TXNIP, GBP2, GBP4, TXN	Immune system
13	T Cell receptor signaling pathway	2	JUN, NCK2	Immune system
14	Cytosolic DNA-sensing pathway	1	IL33	Immune system
15	Toll-like receptor signaling pathway	1	JUN	Immune system
16	Natural killer cell mediated cytotoxicity	6	SHC2, HLA-C, TNFSF10, IFNGR2, FAS, TNFRSF10 B	Immune system
17	Hematopoietic cell lineage	5	IL7R, MME, KIT, ITGA4, CD24	Immune system
18	C-type lectin receptor signaling pathway	2	JUN, PTGS2	Immune system
19	Neutrophil extracellular trap formation	3	H4C8, CLCN4, H2AX	Immune system
20	Fc gamma R-mediated phagocytosis	1	MARCKSL1	Immune system
1	MAPK signaling pathway	14	FLT1, JUN, EREG, MKNK2, TGFB3, KIT, DDIT3, STK4, VEGFD, FAS, FGF7, MAX, NGFR, FGF5	Signal transduction
2	Phosphatidylinositol signaling system	1	ITPKA	Signal transduction
3	NF-kappa B signaling pathway	3	EDA2R, PTGS2, VCAM1	Signal transduction
4	Hippo signaling pathway	4	FRMD6, BIRC5, TGFB3, WNT16	Signal transduction
5	TNF signaling pathway	7	CXCL6, JUN, MMP3, PTGS2, VCAM1, VEGFD, FAS	Signal transduction
6	cAMP signaling pathway	7	HCAR1, JUN, ARAP3, ADRB2, SSTR1, ATP2B1, GABBR2	Signal transduction
7	ErbB signaling pathway	5	SHC2, JUN, BTC, EREG, NCK2	Signal transduction
8	Calcium signaling pathway	9	FLT1, HTR2B, ADRB2, ATP2B1, ITPKA, VEGFD, FGF7, AGTR1, FGF5	Signal transduction
9	JAK-STAT signaling pathway	5	IL7R, IFNGR2, LEPR, SOCS2, FHL1	Signal transduction
10	HIF-1 signaling pathway	5	FLT1, HK2, MKNK2, IFNGR2, SLC2A1	Signal transduction
11	Sphingolipid signaling pathway	1	NSMAF	Signal transduction
12	Phospholipase D signaling pathway	5	LPAR1, SHC2, AGT, KIT, AGTR1	Signal transduction
13	TGF-beta signaling pathway	3	TGFB3, SMAD6, GREM1	Signal transduction
14	mTOR signaling pathway	4	SLC7A5, SLC3A2, CAB39 L, WNT16	Signal transduction
15	Ras signaling pathway	13	FLT1, PLA1A, SHC2, RAB5C, RASA4, RASA4B, KIT, STK4, VEGFD, FGF7, NGFR, PLAAT3, FGF5	Signal transduction
16	cGMP-PKG signaling pathway	7	MEF2A, PDE5A, ADRB2, ATP2B1, GUCY1A2, KCNJ8, AGTR1	Signal transduction
17	PI3K-Akt signaling pathway	20	LPAR1, IL7R, LAMA5, FLT1, HSP90B1, LAMC2, LAMA2, EREG, ITGB4, THBS2, KIT, COL4A2, VEGFD, COL4A1, FGF7, ITGA4, NGFR, TNC, FGF5, ITGA11	Signal transduction
18	AMPK signaling pathway	3	PPARG, CAB39 L, LEPR	Signal transduction
19	Wnt signaling pathway	4	JUN, SFRP1, RSPO3, WNT16	Signal transduction
20	Rap1 signaling pathway	9	LPAR1, FLT1, ARAP3, KIT, VEGFD, FGF7, NGFR, FYB1, FGF5	Signal transduction
21	VEGF signaling pathway	2	SHC2, PTGS2	Signal transduction
22	FoxO signaling pathway	6	IL7R, FBXO32, TNFSF10, GABARAPL1, TGFB3, STK4	Signal transduction
23	Apelin signaling pathway	5	MEF2A, GABARAPL1, EGR1, APLN, AGTR1	Signal transduction
1	Cytokine-cytokine receptor interaction	14	IL7R, CXCL6, TNFRSF6B, EDA2R, TNFSF10, CCL28, TGFB3, IL33, IFNGR2, FAS, TNFRSF10 B, LEPR, NGFR, CCR10	Signaling molecules and interaction
2	p53 signaling pathway	6	PMAIP1, IGFBP3, SFN, STEAP3, FAS, TNFRSF10 B	Cell growth and death
3	Cell cycle	4	MCM5, SFN, TGFB3, ORC1	Cell growth and death
4	Regulation of actin cytoskeleton	8	LPAR1, PPP1R12 B, MYH10, ITGB4, FGF7, ITGA4, FGF5, ITGA11	Cell motility

Silence of BARX1 reversed the phenotype of allergic rhinitis-nasal fibroblasts

To investigate the potential mechanism for the phenotype alteration of AR-NFs, qPCR validation was used to detect the FC of the top 5 DEGs. As we could see in Figure 6(a), the log2FC value of the 5 DEG was basically consistent to the RNA-seq result, and BARX1 had the highest validated FC value. Therefore, BARX1 was selected for further function investigation. Next, si-BARX1 was transfected to the AR-NFs, and qPCR validation showed that the expression level of BARX1 was decreased to ∼25%, indicating an effective interference of BARX1 in AR-NFs (Figure 6(b)). Phenotype analyses indicated that silence of BARX1 significantly inhibited the proliferation and migration ability of AR-NFs (Figure 6(c)). What is more, the supernatant IL-33 and IL-6 as well as the mRNA levels of the two interleukins were both reduced after silencing BARX1 in AR-NFs (Figures 6(d) and (e)).

Figure 6.

Silence of BARX1 reversed the phenotype of AR-NFs. (A) qPCR detection was used to verify the fold change of top 5 DEGs in AR-NFs compared to Ctrl-NFs; (B) qPCR analysis of the mRNA level of BARX1 in AR-NFs after silence of BARX1; (C) MTT assay detected the effect of silencing BARX1 on the proliferation ability of AR-NFs; (D) Transwell assay evaluated the effect of silencing BARX1 on the migration ability of AR-NFs; (E) ELISA assay detected the effect of silencing BARX1 on the production of IL-33 and IL-6 from AR-NFs; (F) qPCR was used to detect mRNA expression levels of IL-33 and IL-6 after silencing BRAX1 in AR-NFs. AR: allergic rhinitis; Ctrl: control; NFs: nasal fibroblasts. *p < 0.05.

Discussion

Allergic rhinitis is a common chronic allergic respiratory disorder involved in various cells and mediators. The roles of the immune cells including DCs, T cells, and mast cells, have been well understood,² and the crosstalk between epithelial cells and immune cells is also well-clarified in the pathogenesis of AR.⁵ However, the role of NFs in AR is less investigated. The present study uncovered the gene expression profile of AR-derived NFs for the first time, and also provided a potential treatment target to reverse the special phenotype AR-NFs.

Nasal fibroblasts are not only the major support cells of nasal mucosa but also could release various cytokines and chemokines to involve in the immune reaction. Several studies have proved that NFs are the important effector cells against the allergens and cytokines through some treatment to normal or AR-NFs. Different from the previous studies, our study focused on both normal and AR-derived NFs, and the direct identification of DEGs between the two NFs without any treatment helps to uncover the actual gene alterations in AR-NFs. Phenotype detection results indicated that AR-NFs showed a higher proliferation and migration abilities, compared to the controls, which is similar to the result of a previous study.²⁰ IL-33 is a member of IL-1 family and also a ligand for the receptor ST2, widely expressed on Th2 cells, mast cells, eosinophils and basophils, and epithelial cells.²¹ Therefore, the increased release of IL-33 generally results in the production of proinflammation cytokines or the activation of the above cells.²² IL-6 is recognized as a marker of allergic asthma,²³ and is also the major regulator in the differentiation and function of Th17 cells,²⁴ which play an important role in the Th2-mediated allergic response through the production of IL-17.²⁵ The elevated levels of supernatant IL-33 and IL-6 from AR-NFs indicates an extensive immune-regulatory and also important proinflammatory roles of NFs in the development of AR. Inhibition the expression of these cytokines released from NFs is helpful to alleviate the progression of AR.

RNA-seq identified 729 DEGs in AR-NFs compared to the controls. To date, there is no any reports about the gene expression profile of AR-derived NFs, and only the dysregulated genes in nasal mucosa tissue or airway epithelium cells from AR patients are identified in some researches.^26,27 Under the thresholds of |FC| value >2 and p value < 0.05, a total of 127 mRNAs are screened out in three of AR nasal mucosa tissue compared to another three of healthy controls through microarray analysis.²⁶ Obviously, the DEGs identified in our RNA-seq result were far more than the 127 DEGs. Apart from the difference in sample sources (NFs vs nasal mucosa tissue), the major reasons were the differences in methods and thresholds. The original RNA-seq analysis result covers more than 40,000 genes, while only ∼26,000 probes are included in the mRNA microarray. Additionally, |FC| value used in our study is 1.5, lower than value of 2. The DEGs screened out in our NFs samples was also compared to the previous study. Interestingly, some genes such as IL33, ZFP90, MTHFD2, and BST2, were both dysregulated in AR nasal mucosa tissue²⁶ and our NFs samples, which indicate the coincidence of the two results to some extends.

The construction of TF regulatory network presented the regulatory relationships among the DEGs. One of the TF, BARX1 with highest FC value (1705.1) and various targets attracted our attention. Another TF, FOXL1, also play an important role in the complex transcription regulatory network, but its FC value was only 3.7, making it difficult to obtain a relative consistent validated result. Therefore, we focused on the top 5 DEGs, and qPCR validation showed that the resulted FC values were basically consistent the RNA-seq results. The selection of BARX1 for next confirmation is based on the considerations of both validated |FC| values and the direct or indirect biological roles. We checked the bioinformatic analyses of the top five candidates in the GO terms and KEGG pathways we interested in (Tables 3 and 4), and found few direct relationships of all the 5 DEGs. Different from other 4 DEGs without target genes, some target genes of BARX1 were enriched in cytokine-related biological processes and classical inflammatory pathways. Therefore, BARX1 with the highest validated FC value and the important biological roles were chosen out.

BARX1 is first studied in developmental biology and is proven to play vital roles in the development of molar teeth, stomach,²⁸ and muscle.¹⁷ The role of BARX1 in immune or inflammation reactions is still a blank. Recent studies showed that BARX1 could also participate in the progression of various cancers including clear cell renal cell carcinoma,¹⁸ endometrial carcinoma,²⁹ and gastrointestinal stromal tumor.³⁰ Overexpression or silence of BARX1 in cancer cells resulted in significant phenotype alterations in proliferation and migration abilities.^18,29,30 Similarly, in the present study, silence of BARX1 significantly inhibited the proliferation and migration abilities of AR-NFs. Additionally, supernatant IL-33 and IL-6 levels were also reduced after silencing BARX1. Since little clues could be found between BARX1 and related bioinformatic annotation, we focused on its target genes.

For instance, VCAM1, GBP2, CTH, PTGS2, and LEPR were enriched in cytokine-related biological processes. The dysregulation of these genes might play some roles in the development of AR. KEGG pathway analysis showed that VCAM1 and PTGS2 were enriched in NF-κB and TNF signaling pathways. NF-κB pathway is an important mediator in the activation of immune and inflammation responses,^31,32 and local NF-κB inhibition suppresses the allergic response in a mouse AR model along with reduced level of IL-6.³³ VCAM1 is a cell surface glycoprotein of the immunoglobulin gene superfamily, and is expressed in endothelial cells, epithelial cells, dendritic cells,³⁴ and fibroblasts.³⁵ The expression of VCAM1 could be induced by cytokine, and play an important role in recruiting immune cells to sites of active inflammation.^34,36 Notably, it is reported that the activity of VCAM1 promoter could also be regulated by NF-κB (p65) in endothelial cells from atherosclerotic plaques,³⁷ and the upregulated VCAM1 in our RNA-seq result might be attributed to the activation of NF-κB pathway. Therefore, the increased IL-6 level in AR-NFs might be attributed to the activation of NF-κB pathway, and the reduced IL-6 in BARX1 silenced AR-NFs might be related to the inhibition of the pathway.

Another target gene PPP1R12B was enriched in cell motility-related pathway (regulation of actin cytoskeleton). PPP1R12B, one of the regulatory subunits of protein phosphatase 1 (PP1), is predominantly expressed in cardiac/skeletal muscle and brain.³⁸ Recent studies indicate that, PPP1R12B is closely associated with the cell migration behavior. For instance, loss of PPP1R12B causes breast cancer cells to exert higher traction forces,³⁹ and PPP1R12B knockdown significantly abrogated pseudopodium enriched atypical kinase 1-mediated tumor suppressive effects, including proliferation and migration abilities.⁴⁰ These evidences showed that PPP1R12B plays important roles in modulating cellular behaviors, and the role of BARX1 on the proliferation and migration behaviors of AR-NFs might be attributed to the transcriptional regulation of PPP1R12B.

In conclusion, our study provided a gene expression profile of AR-derived NFs, and proved that upregulation of BARX1 was responsible for phenotype alteration of AR-NFs. The potential mechanism of BARX1 might be related to the regulation of NF-κB pathway or the transcriptional regulation of PPP1R12B. Additionally, the DEGs identified in AR-NFs also provided a large number of candidate genes to further investigate the role of NFs in the pathogenesis of AR.

Supplemental Material

sj-pdf-1-het-10.1177_09603271211069038 – Supplemental Material for A comprehensive gene expression profile of allergic rhinitis-derived nasal fibroblasts and the potential mechanism for its phenotype

Supplemental Material, sj-pdf-1-het-10.1177_09603271211069038 for A comprehensive gene expression profile of allergic rhinitis-derived nasal fibroblasts and the potential mechanism for its phenotype by Zhengwen Li, Wentao Zou, Jingwen Sun, Shuang Zhou, Yue Zhou, Xiaojing Cai and Jiaxiong Zhang in Human & Experimental Toxicology

Supplemental Material

sj-pdf-2-het-10.1177_09603271211069038 – Supplemental Material for A comprehensive gene expression profile of allergic rhinitis-derived nasal fibroblasts and the potential mechanism for its phenotype

Supplemental Material, sj-pdf-2-het-10.1177_09603271211069038 for A comprehensive gene expression profile of allergic rhinitis-derived nasal fibroblasts and the potential mechanism for its phenotype by Zhengwen Li, Wentao Zou, Jingwen Sun, Shuang Zhou, Yue Zhou, Xiaojing Cai and Jiaxiong Zhang in Human & Experimental Toxicology

Supplemental Material

sj-pdf-3-het-10.1177_09603271211069038 – Supplemental Material for A comprehensive gene expression profile of allergic rhinitis-derived nasal fibroblasts and the potential mechanism for its phenotype

Supplemental Material, sj-pdf-3-het-10.1177_09603271211069038 for A comprehensive gene expression profile of allergic rhinitis-derived nasal fibroblasts and the potential mechanism for its phenotype by Zhengwen Li, Wentao Zou, Jingwen Sun, Shuang Zhou, Yue Zhou, Xiaojing Cai and Jiaxiong Zhang in Human & Experimental Toxicology

Footnotes

Author Contributions

JZ and XC designed the study.

ZL and WZ performed the experiments.

JS and SZ collected the data.

ZL and YZ analyzed the data.

All authors contributed to preparation of the article and approved the final version.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Ethics approval

This study was approved by the Ethics Committee of Shanghai Tenth Peoples’ Hospital, Tongji University, and written informed consent was obtained from all patients. All procedures conformed to the principles of the Declaration of Helsinki.

Data availability statement

The data supporting the conclusions of the study are available from the corresponding author upon request.

ORCID iD

Jiaxiong Zhang

Supplemental material

Supplemental material for this article is available online.

References

Bernstein

Schwartz

Bernstein

. Allergic rhinitis. Immunol Allergy Clin North Am 2016; 36: 261–278.

Greiner

Hellings

Rotiroti

, et al. Allergic rhinitis. Lancet 2011; 378: 2112–2122.

Sur

Plesa

. Treatment of allergic rhinitis. Am Fam Physician 2015; 92: 985–992.

Tomazic

Lang-Loidolt

. Current and emerging pharmacotherapy for pediatric allergic rhinitis. Expert Opin Pharmacother 2021; 22: 849–855.

Kamekura

Yamashita

Jitsukawa

, et al. Role of crosstalk between epithelial and immune cells, the epimmunome, in allergic rhinitis pathogenesis. Adv Otorhinolaryngol 2016; 77: 75–82.

Nur Husna

Tan

Md Shukri

, et al. Nasal epithelial barrier integrity and tight junctions disruption in allergic rhinitis: overview and pathogenic insights. Front Immunol 2021; 12: 663626.

Kim

Cho

Park

, et al. Diesel exhaust particles upregulate interleukins IL-6 and IL-8 in nasal fibroblasts. PLoS One 2016; 11: e0157058.

Lee

Choi

, et al. The effect of urban particulate matter on cultured human nasal fibroblasts. Int Forum Allergy Rhinol 2018; 8: 993–1000.

Roy

Kumar

, et al. Acute phase proteins and their potential role as an indicator for fish health and in diagnosis of fish diseases. Protein Pept Lett 2017; 24: 78–89.

10.

Wong

, et al. Proinflammatory cytokines (IL-17, IL-6, IL-18 and IL-12) and Th cytokines (IFN-γ, IL-4, IL-10 and IL-13) in patients with allergic asthma. Clin Exp Immunol 2001; 125: 177–183.

11.

Park

Cho

, et al. Histamine promotes the release of interleukin-6 via the H1R/p38 and NF-κB pathways in nasal fibroblasts. Allergy Asthma Immunol Res 2014; 6: 567–572.

12.

Gong

Pan

, et al. The emerging role of interleukin-21 in allergic diseases (review). Biomed Rep 2013; 1: 837–839.

13.

Jung

Shim

, et al. Diesel exhaust particles increase nasal symptoms and IL-17A in house dust mite-induced allergic mice. Sci Rep 2021; 11: 16300.

14.

Wang

Zhang

, et al. Interleukin-17A up-regulates thymic stromal lymphopoietin production by nasal fibroblasts from patients with allergic rhinitis. Eur Arch Otorhinolaryngol 2021; 278: 127–133.

15.

Hiromura

Kishida

Nakano

, et al. IL-21 administration into the nostril alleviates murine allergic rhinitis. J Immunol 2007; 179: 7157–7165.

16.

Zhuang

Qiu

, et al. Effect of xingbi gel nasal drops on fyn-STAT5 pathway in nasal mucosa fibroblasts of guinea pigs with allergic rhinitis. Evid Based Complem Alternat Med 2021; 2021: 6686815.

17.

Makarenkova

Meech

. Barx homeobox family in muscle development and regeneration. Int Rev Cell Mol Biol 2012; 297: 117–173.

18.

Sun

Zhang

, et al. Transcription factors BARX1 and DLX4 contribute to progression of clear cell renal cell carcinoma via promoting proliferation and epithelial-mesenchymal transition. Front Mol Biosci 2021; 8: 626328.

19.

Terada

Hamano

Nomura

, et al. Interleukin-13 and tumour necrosis factor-α synergistically induce eotaxin production in human nasal fibroblasts. Clin Exp Allergy 2000; 30: 348–355.

20.

Park

Jun

Kim

, et al. Increased expression of YKL-40 in mild and moderate/severe persistent allergic rhinitis and its possible contribution to remodeling of nasal mucosa. Am J Rhinol Allergy 2013; 27: 372–380.

21.

Hong

Kim

Lin

. Interleukin-33 and ST2 signaling in tumor microenvironment. J Interferon Cytokine Res 2019; 39: 61–71.

22.

Gupta

Dwivedi

. Pathophysiology of IL-33 and IL-17 in allergic disorders. Cytokine Growth Factor Rev 2017; 38: 22–36.

23.

Zagorska

Grzela

Kulus

, et al. Nitric oxide, IL-6 and IL-13 are increased in the exhaled breath condensates of children with allergic rhinitis. Acta Paediatr 2014; 103: e148–e153.

24.

Piao

Song

Lee

, et al. Saikosaponin a ameliorates nasal inflammation by suppressing IL-6/ROR-γt/STAT3/IL-17/NF-κB pathway in OVA-induced allergic rhinitis. Chem Biol Interact 2020; 315: 108874.

25.

Meng

Shi

Zhang

, et al. The role of Th17 cells and IL-17 in Th2 immune responses of allergic conjunctivitis. J Ophthalmol 2020; 2020: 6917185.

26.

Wei

Wang

, et al. Genome-wide analysis of long noncoding RNA expression profile in nasal mucosa with allergic rhinitis. BMC Med Genomics 2021; 14: 100.

27.

Zhang

Huang

Chen

, et al. Identification of key genes in allergic rhinitis by bioinformatics analysis. J Int Med Res 2021; 49: 3000605211029521.

28.

Miletich

Buchner

Sharpe

. Barx1 and evolutionary changes in feeding. J Anat 2005; 207: 619–622.

29.

Yang

, et al. BarH-like homeobox 1 induces the progression of cell malignant phenotype in endometrial carcinoma through the regulation of ERK/MEK signaling pathway. Reprod Biol 2021; 21: 100502.

30.

Hemming

Coy

Lin

, et al. HAND1 and BARX1 act as transcriptional and anatomic determinants of malignancy in gastrointestinal stromal tumor. Clin Cancer Res 2021; 27: 1706–1719.

31.

Sun

. The non-canonical NF-κB pathway in immunity and inflammation. Nat Rev Immunol 2017; 17: 545–558.

32.

Pflug

Sitcheran

. Targeting NF-κB-inducing kinase (NIK) in immunity, inflammation, and cancer. Int J Mol Sci 2020; 21: 8470.

33.

Wee

Zhang

Rhee

, et al. Inhibition of allergic response by intranasal selective NF-κB decoy oligodeoxynucleotides in a murine model of allergic rhinitis. Allergy Asthma Immunol Res 2017; 9: 61–69.

34.

Kelley

Collins

, et al. IκB kinase is critical for TNF-α-induced VCAM1 gene expression in renal tubular epithelial cells. J Immunol 2001; 166: 6839–6846.

35.

Iwamiya

Segard

Matsuoka

, et al. Human cardiac fibroblasts expressing VCAM1 improve heart function in postinfarct heart failure rat models by stimulating lymphangiogenesis. PLoS One 2020; 15: e0237810.

36.

Agassandian

Tedrow

Sembrat

, et al. VCAM-1 is a TGF-β1 inducible gene upregulated in idiopathic pulmonary fibrosis. Cell Signal 2015; 27: 2467–2473.

37.

Yang

, et al. MiR-520b inhibits endothelial activation by targeting NF-κB p65-VCAM1 axis. Biochem Pharmacol 2021; 188: 114540.

38.

Grassie

Moffat

Walsh

, et al. The myosin phosphatase targeting protein (MYPT) family: a regulated mechanism for achieving substrate specificity of the catalytic subunit of protein phosphatase type 1δ. Arch Biochem Biophys 2011; 510: 147–159.

39.

Fokkelman

Balcioglu

Klip

, et al. Cellular adhesome screen identifies critical modulators of focal adhesion dynamics, cellular traction forces and cell migration behaviour. Sci Rep 2016; 6: 31707.

40.

Ding

Tang

, et al. The PEAK1-PPP1R12B axis inhibits tumor growth and metastasis by regulating Grb2/PI3K/Akt signalling in colorectal cancer. Cancer Lett 2019; 442: 383–395.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.41 MB

1.74 MB

0.36 MB