SERPINB10 contributes to asthma by inhibiting the apoptosis of allergenic Th2 cells

Female C57BL/6 J mice (6–8 weeks) were purchased from JSJ Laboratory (Shanghai, China) and bred under specific pathogen-free conditions at the animal center of Zhongshan Hospital, Fudan University (Shanghai, China). All experimental protocols were approved by the Animal Care and Use Committee of Zhongshan Hospital. Mice were anesthetized with isoflurane and administered, via the intratracheal route, adeno-associated virus (AAV) (30 µL; 6.32 × 10¹² viral particles/mL; Vigene Biosciences, Shandong, China) containing SERPINB10 short hairpin (sh)RNA or scrambled shRNA. The sequence of SERPINB10 shRNA was GCAGAACCACAATCTGTTAACTTCAAGAGAGTTAACAGATTGTGGTTCTGCTTTTTT. After 2 weeks, the mice were sacrificed to evaluate the knockdown efficiency of Serpinb10 AAV. Another batch of mice was used to establish an HDM asthma model. Mice were randomly divided into three groups: control group, asthma group and knockdown asthma group. Mice were sensitized by intranasal instillation of HDM extract (10 μg; Greer Laboratories, Lenoir, NC, USA) in 40 μL of phosphate-buffered saline (PBS) on days 0, 1, and 2. From day-8 to day-12, mice were challenged daily by intranasal administration of HDM (10 μg in 40 μL of PBS). Control mice were given, via the intranasal route, 40 μL of PBS during sensitization and challenge phases. Two weeks before the first sensitization, mice were administered 30 µL AAV containing SERPINB10 shRNA or scrambled shRNA. Mice were sacrificed for evaluation on day-14. There were five to six mice per group for each independent experiment.

RT-qPCR and Western blotting were performed as our previous study [12] and the protocol is given in the Additional file 2: Materials and Methods. The primer sequences of all genes for PCR are listed in Additional file 1: Table S1.

BALF was collected according to a method described previously [14]. Briefly, the trachea was cannulated through a 22-inch intravenous catheter and the lungs were lavaged with a total volume of 1 mL PBS for three successive aspirations to obtain BALF. The BALF was centrifuged at 500×g for 8 min at 4 ℃. The cell-free supernatant was collected for cytokine analyses using ELISA kits. Cell pellets were resuspended in PBS and the total cell number was counted using CellDrop® (DeNovix, Wilmington, DE, USA). Cells were analyzed by flow cytometry using PE-conjugated anti-SiglecF (eBioscience, San Diego, CA, USA), FITC-conjugated anti-CD3 (BD Biosciences, San Jose, CA, USA), APC-conjugated anti-CD11c (Multiscience, Zhejiang, China), FITC-conjugated anti-CD19 (BioLegend, San Diego, CA, USA), Percp-cy5.5-conjugated anti-Ly6G (BioLegend) and PE-cy7-conjugated anti-MHC II (BioLegend). The flow cytometry gating strategy was according to a method described previously [15].

We wished to calculate the number of Th cells in lungs. Lung tissues were immersed in Hank’s medium containing collagenase IV (1 mg/mL) and DNase I (20 μg/mL). Lung tissues were ground using a gentleMACS® Dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany) and then incubated with shaking at 100 rpm for 30 min at 37 ℃. Digested tissues were filtered through 70-μm nylon mesh, treated with red blood cell lysis buffer, and washed with staining buffer (PBS containing 2% fetal bovine serum). Cells were first incubated with purified anti-mouse CD16/32 (eBioscience) for 10 min (to block Fc receptors) and then stained with a mixture of Percp-cy5.5-conjugated anti-CD3e (BioLegend), BV510-conjugated anti-CD4 (BD Bioscience) and APC-cy7-conjugated anti-CD45 (BioLegend) for 30 min on ice. For intracellular staining, cells were fixed and permeabilized with Transcription Factor Buffer Set (BD Pharmingen, Franklin Lakes, NJ, USA) before addition of AF647-conjugated anti-GATA3 (BioLegend), BV421-conjugated anti- T-bet (BD Pharmingen) and PE-conjugated anti-active caspase-3 (BD Pharmingen). After washing, samples were analyzed by an Arial III flow cytometer (BD Biosciences) and data were analyzed using FlowJo® (Tree Star, Ashland, OR, USA).

The left lobes of mouse lungs were perfused with 0.3 ml 4% paraformaldehyde and then the left lobes of mouse lungs were isolated and fixed in 4% paraformaldehyde. Paraffin-embedded 5-μm lung sections were stained with hematoxylin and eosin (H&E) and periodic acid Schiff to assess infiltration of inflammatory cells, goblet-cell metaplasia, and mucus production.

Mouse ELISA kits for IL-4, IL-5 and IL-13 in BALF (R&D Systems, Minneapolis, MN, USA) and HDM-specific IgE in serum (JingKang Biotech, Shanghai, China) were used to measure protein expression according to manufacturer instructions. The cytokines in supernatants produced by polarized T cells were stained using the LEGENDplex® panel for human Th cytokines (BioLegend) and measured by flow cytometry.

Patients with asthma were diagnosed by a physician according to Global Initiative for Asthma guidelines. All asthma patients were clinically diagnosed with symptomatic asthma and demonstrated evidence of a hyperresponsive airway (provocative dose of methacholine causing a 20% drop in FEV1 < 2.5 mg) and/or bronchodilator responsiveness (> 12% improvement in FEV1% predicted following inhalation of 200 μg of salbutamol). None of the patients had ever received inhaled or oral corticosteroid or leukotriene antagonist. Demographic information (Table 1) and blood samples from 16 patients were collected for study. Written informed consent was obtained from all patients. This study was approved by the ethics committee of Zhongshan hospital, Fudan University.

Heparinized venous blood was collected from patients with asthma. Then, it was diluted (1:1) with PBS and layered on Lymphoprep® (StemCell Technologies, Vancouver, Canada) density-gradient medium and centrifuged for 20 min at 800×g at room temperature. The layer of peripheral-blood mononuclear cells was collected, washed and resuspended in RoboSep® Buffer (StemCell Technologies). Naïve CD4⁺CD45RA⁺CD45RO⁻ T cells were negatively selected and enriched using EasySep® Human Naïve CD4⁺ T Cell Isolation Kit II (StemCell Technologies) according to manufacturer instructions. The purity of the final isolated fraction (as determined by flow cytometry using FITC-conjugated anti-CD4, PE conjugated anti-CD45RA and APC-conjugated anti-CD45RO (BioLegend)) was 97%. Purified naïve CD4⁺ T cells were cultured in ImmunoCult®-XF Cell Expansion Medium (StemCell Technologies) and stimulated with plate-bound anti-CD3 (2 μg/mL) and anti-CD28 (4 μg/mL; Peprotech-BioGems, Westlake Village, CA, USA). ImmunoCult Human Th1 Differentiation Supplement (StemCell Technologies) and ImmunoCult Human Th2 Differentiation Supplement (StemCell Technologies) were added to direct the differentiation of Th1 and Th2 cells, respectively. After 3–4 days, cells were expanded under identical conditions in the absence of anti-CD3 and anti-CD28. Then, cells were re-stimulated every 7 days. If required, cells were activated with Leukocyte Activation Cocktail (BD Pharmingen) for 6 h.

Lentivirus containing SERPINB10 shRNA or scrambled shRNA were used to transduce CD4⁺ T cells. The sequence of SERPINB10 shRNA was GCCTGTTAACTTTGTGGAA. Naïve CD4⁺CD45RA⁺CD45RO⁻ T cells were stimulated with plate-bound anti-CD3 and anti-CD28. After 48 h, they were transduced with lentivirus (multiplicity of infection = 100) by centrifugation at 500×g for 90 min at room temperature in polybrene (6 μg/mL) and then cultured at 37℃ in a chamber containing 5% CO₂. After 3 days, cells were analyzed by flow cytometry for expression of CD4 and green fluorescent protein (GFP). If required, GFP⁺ cells were sorted by flow cytometry and cultured under polarization conditions.

Data are the mean ± SEM and were analyzed using Prism 8 (Graph Pad, San Diego, CA, USA). Differences were assessed using the unpaired Student’s t-test between two groups, and one-way analysis of variance with Tukey’s multiple comparison test among three groups. P < 0.05 was considered significant.

We wished to investigate the role of SERPINB10 in the Th2 response of allergic asthma. We used an HDM-induced asthma model and knocked down SERPINB10 expression by transfection with AAV against SERPINB10 (Fig. 1a). On day-0, some mice were euthanized and lung tissues were collected for analyses. Expression of protein and transcript levels of SERPINB10 were suppressed by 74.2% by SERPINB10 AAV. Negative control AAV had no effect on SERPINB10 expression (Fig. 1b, c). On day-14, expression of protein and transcript levels of SERPINB10 were increased significantly after HDM sensitization and challenge, and were suppressed after knockdown of SERPINB10 expression (Fig. 1d, e). Our results suggested that SERPINB10 AAV could inhibit baseline and HDM-induced SERPINB10 expression in mouse lungs effectively and continuously.

We used flow cytometry to analyze the inflammatory cells in BALF. The gating strategy is shown in Fig. 2a. The number of total cells, lymphocytes, eosinophils, and neutrophils was increased significantly after exposure to HDM, thereby suggesting that HDM induced allergic inflammation in mouse airways. Knockdown of SERPINB10 expression reduced the number of total cells, lymphocytes, eosinophils, and neutrophils in BALF (Fig. 2b). In support of these observations, histology of lung tissues showed that mice with knockdown of SERPINB10 expression had less peri-bronchial infiltration of inflammatory cells after HDM exposure (Fig. 2c). Goblet-cell hyperplasia and mucus secretion were increased after HDM challenge but were reduced significantly after knockdown of SERPINB10 expression (Fig. 2d). Expression of the Th2 cytokines IL-4, IL-5 and IL-13 in BALF was increased in mice challenged with HDM compared with those challenged with PBS. Knockdown of SERPINB10 expression diminished HDM-induced secretion of Th2 cytokines significantly (Fig. 2e). We also documented reduced expression of HDM-specific IgE in the sera of SERPINB10-knockdown mice (Fig. 2f). Our data suggested that SERPINB10 contributed to HDM-induced airway inflammation and the Th2 response.

We further explored the reasons for the defective Th2 response in SERPINB10-knockdown mice. Lung tissues were prepared into a single-cell suspension and the number of Th cells in lung tissue measured by flow cytometry (Fig. 3a). The number of Th1 and Th2 cells in lung tissues was increased in mice challenged with HDM compared with those challenged with PBS. However, the increase in the number of Th2 cells was much higher than that of Th1 cells. After HDM exposure, SERPINB10-knockdown mice had diminished numbers of Th2 cells, but similar numbers of Th1 cells, compared with those of negative-control mice (Fig. 3b). Caspase-3 is known as an executioner caspase in apoptosis because of its role in coordinating the destruction of cellular structures such as DNA fragmentation or degradation of cytoskeletal proteins.Active caspase-3 is a marker for cells undergoing apoptosis [16]. Intracellular staining for activated caspase-3 revealed that more Th1 cells underwent apoptosis compared with Th2 cells. However, a significant difference in the percentage of activated caspase-3-positive Th1 cells was not observed between SERPINB10-knockdown mice and control mice after HDM exposure. In contrast, Th2 cells in SERPINB10-knockdown mice were more susceptible to apoptosis than those in control mice (Fig. 3c). Our data suggested that SERPINB10 may contribute to airway inflammation and the Th2 response of asthma by regulating apoptosis of Th2 cells.

We further investigated the role of SERPINB10 in polarized Th2 cells in vitro. Naïve CD4 T cells (CD4⁺CD45RA⁺CD45RO⁻) were magnetic beads sorted from the peripheral blood of asthma patients and then cultured under appropriate polarizing conditions to generate Th1 and Th2 cells. Upon completion of polarization, cells were activated with Leukocyte Activation Cocktail and verified using intracellular staining and flow cytometry to detect signature cytokines (Fig. 4a).

Approximately 60% of CD4 T cells could be polarized into Th1 or Th2 cells under appropriate polarization conditions (Fig. 4a). Polarized Th1 cells were confirmed by upregulation of transcript levels of interferon (IFN)-γ and T-bet compared with that in polarized Th2 cells. In contrast, polarized Th2 cells showed upregulated transcript levels of IL-4 and GATA-3 compared with those in polarized Th1 cells (Fig. 4b). Cytokine levels in cell-culture supernatants were measured by LEGENDplex. Polarized Th1 cells secreted more IL-6 and TNF-α than polarized Th2 cells. IL-2 and IL-10 had similar expression trends but were not significant. Polarized Th2 cells produced more IL-4, IL-5 and IL-13 than polarized Th1 cells (Fig. 4c).

Next, we determined whether the SERPINB10 transcript level was increased in Th1 or Th2 cells. Stimulation of T-cell receptors (TCRs) with anti-CD3 antibody caused upregulation of SERPINB10 expression in polarized Th2 cells but not in polarized Th1 cells. Moreover, stimulation through the IFN-γ receptor did not induce an increase in SERPINB10 transcript levels in polarized Th1 cells, whereas stimulation of IL-4 receptors resulted in upregulation of SERPINB10 expression in polarized Th2 cells, though not significantly (Fig. 4d). Our data demonstrated that stimulation through TCRs resulted in upregulation of SERPINB10 expression in polarized Th2 cells, but not in Th1 cells.

To clarify further the functional importance of upregulation of SERPINB10 expression in Th2 cells, we sorted naïve CD4 T cells from the peripheral blood of asthma patients and transduced them with lentivirus encoding control shRNA or shRNA specific for SERPINB10. Transduced cells were identified based on GFP expression by microscope observation and flow cytometry (Fig. 5a, b). Approximately 60% of total cells were transduced with lentivirus (Fig. 5c). Levels of SERPINB10 transcripts decreased by 60% after transduction of lentivirus (Fig. 5d). GFP⁺ cells were sorted by flow cytometry and then cultured under polarization conditions. Upon completion of polarization, we did not observe a significant difference in the percentage of Th1 cells between the SERPINB10-knockdown group and control group. Similarly, there was no significant difference in the percentage of Th2 cells between these two groups (Fig. 5e). However, when cells were expanded under an identical culture condition and rested until day-8, the percentage of Th2 cells in the SERPINB10-knockdown group was decreased significantly compared with that in the control group (Fig. 5f). Phosphatidylserine expression on plasma membranes was measured by Annexin-V staining and revealed that Th2 cells with knockdown of SERPINB10 expression were more susceptible to apoptosis than control Th2 cells (Fig. 5g). However, a significant difference in the percentage of Th1 cells (Fig. 5f) and percentage of Annexin V-positive Th1 cells (Fig. 5g) between two groups was not observed. Our data demonstrated that SERPINB10 could protect Th2 cells from apoptosis.