Bcl2-like Protein 12 Plays a Critical Role in Development of Airway Allergy Through Inducing Aberrant TH2 Polarization.
Zhi-Qiang Liu,Ying Feng,Li-Hua Mo,Xian-Hai Zeng,Jiang-Qi Liu,Rui-Di Xie,Zhi-Gang Liu,Ping-Chang Yang,Guang-Ji Zhang,Shan-Dong Wu
DOI: https://doi.org/10.1016/j.jaci.2018.07.045
IF: 14.29
2019-01-01
Journal of Allergy and Clinical Immunology
Abstract:Although it is well understood that skewed TH2 polarization plays a critical role in the pathogenesis of allergic asthma,1Mazzarella G. Bianco A. Catena E. De Palma R. Abbate G.F. Th1/Th2 lymphocyte polarization in asthma.Allergy. 2000; 55: 6-9Crossref PubMed Scopus (132) Google Scholar the mechanism of TH2 polarization remains elusive.2Kidd P. Th1/Th2 balance: the hypothesis, its limitations, and implications for health and disease.Altern Med Rev. 2003; 8: 223-246PubMed Google Scholar Our recent studies showed that Bcl2-like protein 12 (Bcl2L12) was associated with pathogenesis of TH2-biased inflammation in the intestine by facilitating TH2 polarization.3Li M.G. Liu X.Y. Liu Z.Q. Hong J.Y. Liu J.Q. Zhou C.J. et al.Bcl2L12 contributes to Th2-biased inflammation in the intestinal mucosa by regulating CD4(+) T cell activities.J Immunol. 2018; 201: 725-733Crossref PubMed Scopus (15) Google Scholar Thus we hypothesize that Bcl2L12 might be also associated with the pathogenesis of allergic asthma. In this study we developed a mouse strain with Bcl2L12 knockout (KO) CD4+ T cells (ie, KO mice; see Fig E1 in this article's Online Repository at www.jacionline.org)3Li M.G. Liu X.Y. Liu Z.Q. Hong J.Y. Liu J.Q. Zhou C.J. et al.Bcl2L12 contributes to Th2-biased inflammation in the intestinal mucosa by regulating CD4(+) T cell activities.J Immunol. 2018; 201: 725-733Crossref PubMed Scopus (15) Google Scholar; KO mice and wild-type (WT) mice were sensitized to ovalbumin (OVA) to develop airway allergy according to published procedures.4Hong J.Y. Chung Y. Steenrod J. Chen Q. Lei J. Comstock A.T. et al.Macrophage activation state determines the response to rhinovirus infection in a mouse model of allergic asthma.Respir Res. 2014; 15: 63Crossref PubMed Scopus (35) Google Scholar After sensitization, WT mice showed asthma-like inflammation in the airway, including profound infiltration of mononuclear cells in lung tissue, increase in thickness of the basal membrane of bronchial walls (Fig 1, A), antigen-specific IgE in sera, and high levels of TH2 cytokines and mouse mast cell protease 1 in both sera and bronchoalveolar lavage (BAL) fluid, which were much lower in KO mice. The signature TH1 cytokines IFN-γ, IL-17, and TNF were also detected in sera and BAL fluid of both WT and KO mice, although levels were lower in sensitized WT mice compared with those in control mice (Fig 1, B-E). In BAL fluid total cell, eosinophil, lymphocyte, neutrophil, and macrophage counts were significantly greater in sensitized WT mice than those in sensitized KO mice (Fig 1, F-J). Lung resistance was much greater in sensitized WT mice than in sensitized KO mice (Fig 1, K). The data suggest that Bcl2L12 plays a critical role in development of asthma-like inflammation in the airway. BAL fluid was obtained from asthmatic patients and healthy subjects (see Table E1 for patient demographic data in this article’s Online Repository at www.jacionline.org). Cellular elements were prepared from BAL fluid samples. Compared with samples from healthy subjects, BAL fluid samples from asthmatic patients showed a TH2-dominant profile, as indicated by a higher frequency of TH2 cells and high levels of TH2 cytokines (Fig 2, A-C, and see Fig E2 in this article's Online Repository at www.jacionline.org). Expression of Bcl2L12 in CD4+ T cells isolated from BAL fluid samples was greater in the asthma group than that in the healthy group (Fig 2, D and E; see Table E2 for primers used in RT-qPCR in this article's Online Repository at www.jacionline.org). A positive correlation was identified between Bcl2L12 and IL-4 (r = 0.7172, P = .0196), IL-5 (r = 0.7545, P = .0117), and IL-13 (r = 0.8076, P = .0047) in BAL fluid CD4+ T cells collected from the asthma group (Fig 2, F-H). A negative correlation (r = 0.8953, P = .0005) was identified between Bcl2L12 expression in BAL fluid CD4+ T cells and FEV1 of asthmatic patients (Fig 2, I). Similar phenomena were also found in peripheral blood CD4+ T cells of the same group of asthmatic patients (see Fig E3 in this article's Online Repository at www.jacionline.org). No correlation was detected between Bcl2L12 and IFN-γ, IL-17, or TNF (data not shown). These data imply that Bcl2L12 might be involved in the development of the skewed TH2 response in asthmatic patients. CD4+ T cells were isolated from blood samples collected from asthmatic patients and healthy subjects to gain insight into the mechanism by which Bcl2L12 regulates TH2 response. After stimulation with activators in culture for 4 days, more IL-4+ T cells were induced in naive CD4+ T cells collected from asthmatic patients than those from healthy subjects, which was abolished by knocking down Bcl2L12 expression. Overexpression of Bcl2L12 induced naive CD4+ T cells to differentiate into TH2 cells (Fig 2, J, and see Fig E4 in this article's Online Repository at www.jacionline.org). Results of co-immunoprecipitation showed that Bcl2L12 formed a complex with GATA-3 in CD4+ T cells; the quantity of the complex was greater in samples collected from asthmatic patients than that from healthy subjects (Fig 2, K). Binding between Bcl2L12 and GATA-3 was verified in in vitro experiments. HEK293 cells were transfected with Bcl2L12-expressing and GATA-3–expressing plasmids. A complex of rBcl2L12 and rGATA-3 was detected in the HEK293 cells (Fig 2, L). By using the co-chromatin immunoprecipitation (co-ChIP) approach, the complex of Bcl2L12 and GATA-3 was detected in the Il4 promoter locus, which was much greater in CD4+ T cells collected from asthmatic patients than that from healthy subjects (Fig 2, M). To corroborate these data, CD4+ T cells were isolated from the spleens of WT and KO mice treated with airway sensitization procedures. The cells were analyzed by using co-ChIP. The results showed that the Bcl2L12 and GATA-3 complex was detected in the Il4 promoter locus of CD4+ T cells isolated from sensitized WT mice. Inhibition of Bcl2L12 suppressed levels of GATA-3 but did not affect its baseline in the Il4 promoter locus of CD4+ T cells (Fig 2, N). Data demonstrate that Bcl2L12 plays an important role in TH2 polarization in asthmatic patients by modulating the activities of GATA-3. The data indicate that Bcl2L12 might be a new member of the molecular family in regulation of GATA-3 activities in CD4+ T cells. Because of the proline-rich feature Bcl2L12 confers on binding ability,5Nedved M.L. Gottlieb P.A. Moe G.R. CD and DNA binding studies of a proline repeat-containing segment of the replication arrest protein Tus.Nucleic Acids Res. 1994; 22: 5024-5030Crossref PubMed Scopus (9) Google Scholar, 6Scorilas A. Kyriakopoulou L. Yousef G.M. Ashworth L.K. Kwamie A. Diamandis E.P. Molecular cloning, physical mapping, and expression analysis of a novel gene, BCL2L12, encoding a proline-rich protein with a highly conserved BH2 domain of the Bcl-2 family.Genomics. 2001; 72: 217-221Crossref PubMed Scopus (81) Google Scholar it can contribute to formation of the Bcl2L12 and GATA-3 complex. Because GATA-3 is the major transcription factor of IL-4, these results suggest that physical contact between Bcl2L12 and GATA-3 enhances expression of IL-4 in CD4+ T cells. Overproduction of IL-4 is one of the main features of TH2-biased inflammation, such as allergic asthma. GATA-3 is the essential transcription factor of IL-4. Thus it is conceivable that inhibition of GATA-3 might attenuate or inhibit the TH2-biased inflammation.7Krug N. Hohlfeld J.M. Kirsten A.M. Kornmann O. Beeh K.M. Kappeler D. et al.Allergen-induced asthmatic responses modified by a GATA3-specific DNAzyme.N Engl J Med. 2015; 372: 1987-1995Crossref PubMed Scopus (226) Google Scholar However, IL-4 is also an important cytokine in antibody production, which plays a crucial role in the defensive functions of the body. The present data indicate that inhibition of Bcl2L12 can inhibit binding between GATA-3 and the Il4 promoter to suppress IL-4 expression but does not affect its baseline expression. Data also suggest that inhibiting Bcl2L12 can inhibit the unnecessary amount of IL-4 and not affect its physiologic functions. Bcl2L12 antibody was purchased from Abcam (Cambridge, Mass). The reagent kit for Bcl2L12 RNA interference and GATA-3 antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, Calif). ELISA kits for specific IgE, mouse mast cell protease 1, IL-4, IL-5, IL-13, IFN-γ, IL-17, and TNF were purchased from Biomart (Beijing, China). Fluorescence-labeled antibodies (for flow cytometry) for CD3, CD4, IL-4, IL-5, IL-13, IFN-γ, IL-17, and TNF were purchased from BD Biosciences (Franklin Lakes, NJ). Immune cell isolation kits were purchased from Miltenyi Biotech (San Diego, Calif). Reagents for real-time quantitative RT-PCR, Western blotting, and gene transfection were purchased from Invitrogen (Carlsbad, Calif). Phorbol 12-myristate 13-acetate, ionomycin, reagents, and materials for immunoprecipitation and ChIP were purchased from Sigma-Aldrich (St Louis, Mo). Patients with perennial asthma were recruited to this study at our affiliated hospitals. Diagnosis and management of asthma were carried out by the physicians of our hospitals, according to routine procedures. Demographic data of the patients are presented in Table E1. Patients with any of the following conditions were excluded from this study: severe organ diseases, cancer, and autoimmune diseases, as well as those using immune suppressors. Healthy subjects were also recruited to be used as control subjects. Use of human tissues in the present study was approved by the Human Ethics Committee at Shenzhen University. Informed written consent was obtained from each human subject.Table E1Clinical characteristics of study groupsHealthy subjectsAsthmatic patientsNo. of subjects3030Age (y)35.6 ± 6.236.5 ± 7.3Sex (female/male)15/1515/15Body mass index (kg/m2)25.8 ± 4.626.2 ± 5.1FEV1 (% predicted)111 ± 12.383 ± 14.5∗P ≤ .05 compared with the healthy group (t test).FEV1/FVC ratio (%)86 ± 6.170 ± 8.3∗P ≤ .05 compared with the healthy group (t test).Total IgE (IU/mL)†Median (lower quartile–upper quartile).23 (5.5–41.5)168 (85.5–285.3)∗P ≤ .05 compared with the healthy group (t test).DME-specific IgE (IU/mL)UD68.5 (46.5-166.8)Eosinophil count (/μL)†Median (lower quartile–upper quartile).98 (71–125)226 (113–324)∗P ≤ .05 compared with the healthy group (t test).C-reactive protein (mg/L)†Median (lower quartile–upper quartile).0.91 (0.38–2.5)1.5 (0.7–4.8)∗P ≤ .05 compared with the healthy group (t test).DME-SPT (mm)UD8.8 ± 3.5With allergic rhinitis06 (20%)With allergic dermatitis04 (13%)Values shown are means ± SDs.DME, Dust mite extract; SPT, skin prick test; UD, undetectable.∗ P ≤ .05 compared with the healthy group (t test).† Median (lower quartile–upper quartile). Open table in a new tab Values shown are means ± SDs. DME, Dust mite extract; SPT, skin prick test; UD, undetectable. Among the human subjects, BAL was performed in 10 asthmatic patients and 10 healthy subjects by using fiberoptic bronchoscopy (Olympus B2-10; Olympus, Tokyo, Japan), according to our routine procedures, which were also reported by others.E1Deshane J.S. Redden D.T. Zeng M. Spell M.L. Zmijewski J.W. Anderson J.T. et al.Subsets of airway myeloid-derived regulatory cells distinguish mild asthma from chronic obstructive pulmonary disease.J Allergy Clin Immunol. 2015; 135: 413-424.e15Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar Briefly, by using four 50-mL aliquots of warm saline instilled at the right-middle lobe of the lung, about 100 mL of fluid was recovered from each subject. Samples were centrifuged at 400g for 5 minutes at 4°C. Supernatants were collected and stored at −80°C until use. Cell pellets were resuspended in RPMI 1640 for cytologic analysis. Lungs of each mouse were flushed with PBS (contEaining 0.5% FBS BSA in 1 mL of PBS). BAL fluid was obtained after lavage and centrifuged at 400g for 5 minutes at 4°C. Pellets were resuspended in 50 μL of PBS, and cytospin slides of BAL fluid cells were made with a cytospin device (Shandon Southern Instruments, Runcorn, United Kingdom). Slides were stained with May-Grunwald Giemsa. Cell counts were performed by using a light microscope. Absolute numbers and percentages of eosinophils, lymphocytes, neutrophils, and monocytes/macrophages were quantified. The supernatant was collected and stored at −80°C until use. All animal experiments were approved by the Shenzhen University Institutional Animal Care and Use Committee. Male BALB/c mice (WT mice; provided by the Guangdong Experimental Animal Center, Guangzhou, China) and mice with Bcl2L12 KO CD4+ T cells (provided by the Institute of Bipplatrics at Chinese Academy of Agricultural Sciences, Beijing, China; Fig E1) were injected intraperitoneally on days 0, 7, and 14 with OVA (100 μg mixed in 0.2 mL of alum) or PBS (control). The mice were challenged intranasally with 50 μL of 100 μg of OVA or PBS daily from days 21 to 27. Mice were killed on day 28.Fig E2BAL fluid CD4+ T-cell phenotypes. BAL fluid samples were collected from asthmatic patients (n = 10) and healthy subjects (n = 10). BAL fluid cells were prepared and analyzed by using flow cytometry. A, CD3+CD4+ T cells were gated. B, Gated histograms indicate the frequency of CD4+ T-cell phenotypes in BAL fluid cells.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E3Correlation between Bcl2L12 and FEV1 and TH2 cytokines in peripheral blood CD4+ T cells of asthmatic patients. Blood samples were collected from healthy subjects (n = 30) and asthmatic patients (n = 30). PBMCs were prepared and analyzed by using flow cytometry. A-C, Gated dot plots show CD4+ T cells in PBMCs. Gated histograms show the frequency of CD4+ T-cell phenotypes. Bars show summarized frequency data of CD4+ T-cell phenotypes. D and E, Levels of mRNA (Fig E3, D) and protein (Fig E3, E) in peripheral CD4+ T cells. F, Integrated density of the immunoblots from Fig E3, E. G, Negative correlation between Bcl2L12 mRNA in peripheral CD4+ T cells and FEV1 of asthmatic patients. H, mRNA levels of cytokines in CD4+ T cells (isolated from PBMCs). I, Correlation between Bcl2L12 mRNA and mRNA of TH2 cytokines in peripheral CD4+ T cells. Data in bars are presented as means ± SDs. *P < .01 compared with the healthy group. Samples from individual subjects were analyzed separately.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E4Assessment of the role of Bcl212 in TH2 cell differentiation. Naive CD4+ T cells (CD3+CD4+CD25−CD62L+) were prepared from PBMCs isolated from healthy subjects (n = 30) and asthmatic patients (n = 30). Cells were treated with procedures as denoted above each subpanel and exposed to activators (phorbol 12-myristate 13-acetate, 50 ng/mL; ionomycin, 0.5 μg/mL) in culture for 4 days. A, Gated flow cytometric histograms show representative data of the frequency of TH2 cells generated from naive CD4+ T cells from asthmatic patients. B, Gated flow cytometric histograms show representative data of the generated TH2 cells. C, Immunoblots show the results of Bcl2L12 RNA interference. D, Gated flow cytometric histograms show representative data of the frequency of generated TH2 cells. E, Results of Bcl2L12 overexpression in CD4+ T cells by transfecting with Bcl2L12-expressing plasmids.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Cytokines in BAL fluid and serum were quantified by means of ELISA with relevant reagent kits, according to the manufacturer's instructions. Blood samples were collected from each human subject by means of ulnar vein puncture. PBMCs were isolated from blood samples by means of density gradient centrifugation. Respiratory mechanics were evaluated by using the flexiVent system, according to the published procedures,E2Robichaud A. Fereydoonzad L. Urovitch I.B. Brunet J.D. Comparative study of three flexiVent system configurations using mechanical test loads.Exp Lung Res. 2015; 41: 84-92Crossref PubMed Scopus (11) Google Scholar to evaluate lung function in the mice. Mice were anaesthetized by means of intraperitoneal injection with xylaxine and sodium pentobarbital. Mice were treated with increasing concentrations of aerosolized methacholine; measurements were performed with the flexiVent small-animal ventilator (SCIREQ, Montreal, Quebec, Canada). Immune cells were isolated from PBMCs by means of magnetic cell sorting (MACS) with relevant reagent kits, according to the manufacturer's instructions. Purity of the isolated cells was checked by using flow cytometry. If the purity was less than 95%, MACS was performed again with the cells. On surface staining, cells were stained with antibodies (labeled with fluorochromes) of interest or isotype IgG for 30 minutes at 4°C. If intracellular staining was required, cells were treated with fixative/permeable reagents and stained with antibodies (labeled with fluorochromes) of interest or isotype IgG for 30 minutes at 4°C. Cells were analyzed with a flow cytometer (FACSCanto II; BD Biosciences). Data were processed with the software package FlowJo (Treestar, Ashland, Ore). Isotype IgG staining data were used as gating references. Cells were collected from relevant experiments. Total RNA was extracted from cells with TRIzol reagents (Thermo Fisher, Waltham, Mass). cDNA was converted with RNA samples with a reverse transcription kit, according to the manufacturer's instructions. Samples were amplified in a quantitative PCR device (CFX96 Touch; Bio-Rad Laboratories, Hercules, Calif) with SYBR Green Master Mix in the presence of relevant primers (Table E2, see below). Results were normalized to fold change against the housekeeping gene β-actin.Table E2Primers used in real-time quantitative RT-PCRMoleculesForwardReverseBcl2L12cttctatgctttggtggccctacagaacagctccaccaggIL-4gcagttctacagccaccatgactctggttggcttccttcaIL-5gccatccccacagaaattccccccttgcacagtttgactcIL-13catccgctcctcaatcctctgatgctccataccatgctgcIFN-γgtcctccatgagctgatccagtttctcccaccctctcctcIL-17tgtgatctgggaggcaaagtcccacggacaccagtatcttTNFgtcaacctcctctctgccatccaaagtagacctgcccagap53tggccatctacaagcagtcaggtacagtcagagccaacctFasgtgctttgcttagggttcccaacttgcacttctggccatgβ-Actincatggaatcctgtggcatcccacacagagtacttgcgctc Open table in a new tab Cells were collected from relevant experiments and lysed with lysis buffer. Samples were centrifuged for 10 minutes at 13,000 rpm. Supernatants were collected and used as cytosolic extracts. Pellets were resuspended in nuclear lysing buffer for 30 minutes. Samples were centrifuged for 10 minutes at 13,000 rpm. Supernatants were collected and used as nuclear extracts. All procedures were performed at 4°C. Proteins were fractioned by using SDS-PAGE and transferred onto a polyvinylidene difluoride membrane. After blocking with 5% skim milk solution for 30 minutes, the membrane was incubated with the first antibodies of interest or isotype IgG overnight at 4°C, followed by incubating with the second antibodies (labeled with peroxidase) for 2 hours at room temperature. The membrane was washed with Tris-buffered saline–Tween 20 after each incubation. Immunoblots on the membrane were developed with enhanced chemiluminescence and photographed with an image device (UVI, Cambridge, United Kingdom). CD4+ T cells were isolated from PBMCs of asthmatic patients by using MACS. Cells were treated with a Bcl2L12 short hairpin RNA reagent kit (Santa Cruz Biotech, Santa Cruz, Calif), according to the manufacturer's instructions. The effects of RNA interference were assessed 48 hours later by using Western blotting. Naive CD4+CD25− T cells were isolated from PBMCs of healthy subjects by using MACS. Cells were transfected with the full-length Bcl2L12 gene-carrying plasmids or control plasmids (provided by the Sangon Biotech, Shanghai, China) by means of electroporation with a Bio-Rad Gene Pulse set at 950 μF and 280 V (Bio-Rad Laboratories). Bcl2L12 expression was assessed in cells 48 hours later by using Western blotting. Protein samples were precleared by means of incubation of protein G agarose beads for 2 hours. The supernatant was incubated with antibodies of interest or isotype IgG overnight. Agarose beads were collected by means of centrifugation. Protein complexes on the beads were eluted with an eluting buffer and analyzed by means of Western blotting. All procedures were performed at 4°C. Cells were collected from relevant experiments and fixed with 1% formalin for 15 minutes. Cells were lysed with a lysis buffer, followed by sonicating to shear DNA into small pieces. Samples were then processed with immunoprecipitation procedures. After eluting from the agarose beads, DNA was recovered from samples with a DNA recovering reagent kit (QIAGEN Sciences, Germantown, Md), according to the manufacturer's instructions. DNA was analyzed by using qPCR in the presence of primers of the Il4 promoter (ggcctctcccttctatgcaa and gattgtcagtcacttggggc). Results were normalized to fold change against the input. Data are presented as means ± SDs. The difference between 2 groups was determined by using the Student t test or ANOVA, followed by the Dunnett t test or SNK test for multiple comparisons. Correlation between 2 groups was tested with the Pearson correlation assay. A P value of less than .05 was set as a significance criterion.