The experimental protocols were conducted in accordance with the National Institutes of Health guidelines. This study was reviewed and approved by the Institutional Animal Care and Use Committee of Beijing University of Chinese Medicine. Male guinea pigs (200–250 g) were provided by Cyagen Biosciences, Suzhou, China. They were randomly divided into the following groups: control, capsaicin (Solarbio, Beijing), capsaicin + sinomenine, capsaicin + si-SOX5, and capsaicin + sinomenine + SOX5. All guinea pigs were kept in IVC rearing cage in SPF animal rooms with 12 h light/12 h dark cycle at temperature of 20–26°C. There were two guinea pigs per cage, and each cage had autoclaved bedding materials (Cyagen Biosciences, Suzhou, China).

The guinea pig with enhanced cough sensitivity model was established as described earlier (Ji et al., 2017). In order to sensitize guinea pigs, cyclophosphamide (Solarbio, Beijing) was injected into the abdominal cavity at a dose of 30 mg/kg. Two days later, 2 mg ovalbumin and 100 mg aluminum hydroxide (Solarbio, Beijing) were injected into the abdominal cavity. After 3 weeks, 0.01 mg ovalbumin (Solarbio, Beijing) and 100 mg aluminum hydroxide were injected into the abdominal cavity. Three weeks after the enhancement of immunization, the guinea pigs were placed in self-made aerosol box, and stimulated with 1% ovalbumin solution atomized with Devilliis 646 atomizer for 90 s at an outflow rate of 0.037 ml/min). The number of coughs within 3 min was counted. Guinea pigs with cough reflex due to stimulation were considered as successful model of increased cough sensitivity (Figure 1H).

In all experiments, capsaicin was used to induced cough in guinea pigs. The guinea pigs were placed in transparent plastic bottles and each animal was exposed to capsaicin (50 μmol/L, 1 mL) vapor via an ultrasonic atomizer for 10 min. In the sinomenine-treated group, the guinea pigs were pretreated with sinomenine at a dose of 0.5/100 g body weight/day through intragastric administration for 7 days. The crude drug dose of sinomenine was 0.5 g sinomenine herbal powder/100 g body weight of guinea pig/day. Recombinant human SOX5 (2 mg/kg, 0.5 mg/ml in PBS) was purchased from R&D, and was administered via atomization inhalation with capsaicin (Figure 1H). Adenovirus overexpressing TRPV1 was administered via atomization inhalation 2 days before capsaicin treatment.

All animals were sacrificed under intraperitoneal anesthesia with 3% pentobarbital sodium at a dose of 90 mL/kg bwt, and the trachea and lung tissues were excised for use in follow-up studies. Portions of the lung tissues were fixed in 4% paraformaldehyde (Solarbio, Beijing) for 3 days, and embedded in paraffin for H&E and IHC staining. The other portions of lung tissues were used for the extractions of total RNA and protein. The sample size (n) for each experimental group/condition was 7 (n = 7).

Single chamber unrestricted body scanner (Buxco, United States) was used to determine cough reflex sensitivity (CRS) level of guinea pigs. The total number of cough was counted in 10-min duration, and it was used to reflect CRS level. Changes in air flow signal were measured using Buxco system. The area under the curve (V2) and the half peak conversion time of the air flow were calculated to judge whether it was cough, sneeze, or other irregular activities. Atomized capsaicin (50 μmol/L, 1 mL) was used to induce cough for 10 min, and cough count was recorded.

After determination of CRS (12 h), non-invasive animal lung function detector FinePointe^TM NAM (Buxco, United States) was used to determine enhanced pause (Penh) level. Aerosolized methacholine (200 mg/L, 100 μL) was used as a stimulant to induce airway response. Average value of Penh was recorded and converted to the percentage Penh value for saline group, expressed as Penh %, as an evaluation index of airway responsiveness.

Primary mouse hippocampal neuron cells (HNC) were isolated as described below: Neonatal guinea pigs were anesthetized with isoflurane and sterilized using 75% ethanol. Brain tissue was cut into small sections (1mm3) and digested with pancreatin (Beyotime Biotechnology, Beijing). The digested cell suspensions were centrifuged at 1000 rpm, and the cell pellet was resuspended in DMEM containing 10% FBS. The neuronal cells were cultured in cell incubator under 5% CO₂ for 48 h.

Next, Si-SOX5 and TRPV1 overexpression plasmids were produced by RiboBio (Guangzhou, China). Neuronal cells were transfected according to the manufacturer’s protocol using X-treme GENE siRNA transfection reagent (Roche, Germany). Six hours later, the transfection medium was changed to culture medium containing 10% FBS, and the cells were cultured for 48 h. In the capsaicin-treated groups, 50 μM capsaicin was administered for 3 h. For SOX5-related groups, cells were treated with SOX5 (2 ng/ml) for 24 h before capsaicin treatment.

Total RNA was extracted from cells and tissue samples with TRIzol reagent (Invitrogen) according to the manufacturer’s protocol. The samples were treated with DNase I before reverse transcription into cDNA in line with the manufacturer’s protocol (iScript, Bio-Rad). Quantitative Real-Time PCR was performed using LightCycler 480 (Roche) and SYBR Green Master Mix (Roche) with corresponding primers in line with standard protocol. All assays were done at least in duplicate, and in a total of at least two independent assays. The relative expression levels were calculated using the 2^–ΔΔCt method. Data were analyzed by GraphPad 7. The sequences of primers used were: SOX5: Forward: CTGCCGCCATTGATGATTCC; Reverse: CCAGCCTTGTAGCTGAAACCA; TRPV1: Forward: CCGGCTTTTTGGGAAGGGT; Reverse: GAGACAGGTAG GTCCATCCAC.

The in vivo protein expression levels of SP and NKA in trachea and lung tissues were determined using standard immunoblotting protocol¹. The membranes were scanned with Odyssey Infrared Scanning System (Gene Co., Ltd., Hong Kong, China), and the blot results were analyzed using ImageJ software. The primary antibodies used were CD68 and GAPDH antibodies (Proteintech Group, Wuhan, China), as well as SOX5 and TRPV1, SP and NKA primary antibodies (Cell Signaling Technology, Danvers, MA, United States). The secondary antibodies (IRDye700/800 mouse and rabbit) were products of LICOR (Lincoln, NE, United States).

The psiCHECK-2 luciferase reporter plasmid (Transgen, Beijing) was inserted into the wild-type TRPV1 promotor or mutant TRPV1 promotor sequences that contained the putative binding sites of SOX5 to construct SOX5-WT or SOX5-Mut expression plasmid which was transfected with reporter vectors into neuronal cells. The cells were collected 48 h post-transfection and lysed, and luciferase activity (Promega) was assayed in the lysates.

Neuronal cells were plated in a 24-well cell culture plate. After capsaicin treatment, the cells were washed with PBS and fixed with 4% paraformaldehyde, followed by permeabilization with 0.2% Triton-X-100 solution in PBS. Next, the cells were blocked using goat serum, and incubated with SOX5 antibody at 4°C overnight, followed by incubation with FITC-conjugated goat anti-mouse antibodies for 1 h. After three washes with PBS, the cells were incubated with DAPI. Frozen sections of guinea pig were fixed in 4% paraformaldehyde, washed using PBS, and permeabilized with 0.5% Triton X-100. After washing three times, the blocked sections were treated with 50% goat serum. Then, the sections were incubated with SOX5 antibody overnight, incubated with secondary antibody, and stained with DAPI. Immunofluorescence was measured using an IX73 fluorescence microscope (Olympus, Valley, PA, United States).

Paraffin sections of lung tissue were dewaxed in xylene and descending series of ethanol. The sections were permeabilized using 0.5% Triton X-100. After washing three times, they were blocked with 50% goat serum. Then, the sections were incubated with SOX5 primary antibody overnight, followed by incubation with secondary antibody for 1 h at room temperature. Thereafter, the sections were DAPI-stained for 5 min at room temperature. The sections were photographed with light scope under an IX73 fluorescence microscope (Olympus, Valley, PA, United States), and analyzed using ImageJ software.

Hanks Balanced Salt Solution (HBSS) was used to dilute BALB cellProbe F3 solution 2000 times. This was used as working solution. The cells were added to 1 mL of working solution instead of culture media, and cultured at 37°C for 20 min. Then, the same volume of HBSS media containing 1% FBS was added to cells and incubated at 37°C for 40 min. Thereafter, the cells were digested, centrifugated and resuspended in 1 mL of HBSS, followed by measurement of fluorescence intensity (Promega).

All data are presented as a mean ± SEM. Statistical analyses were performed using the GraphPad Prism 8 software. Two-group comparison was done with two-tailed Student’s t-test, while multi-group comparison was done with one-way ANOVA. Values of p < 0.05 were considered as indicative of statistical significance.

To determine the effect of sinomenine on increased cough sensitivity induced by neurogenic inflammation, a guinea pig model of increased cough sensitivity was used. As shown in Figure 1A, capsaicin treatment increased the CRS of guinea pigs. In contrast, pretreatment with sinomenine suppressed CRS. Furthermore, compared to saline group, capsaicin treatment increased the Penh level (Figure 1B). However, sinomenine significantly reduced the level of Penh (Figure 1B). These data suggest that sinomenine can significantly attenuate capsaicin-induced high sensitivity of cough in guinea pigs. Results from H&E staining showed that sinomenine pre-administration reversed capsaicin-induced increase in inflammatory cell infiltration (Figure 1C). Immunohistochemical staining showed that atomized capsaicin treatment led to increased infiltration by monocytes/macrophages. However, sinomenine significantly reduced the level of CD68-positive monocytes/macrophages (Figures 1D,F). SP and NKA are two kinds of sensory neuropeptides which act as important inflammatory mediators that stimulate mast cells, affect the production of inflammatory factors by T-lymphocytes and B-lymphocytes, induce neurogenic inflammation, increase airway sensitivity, and cause persistent chronic cough (Wiesenfeld-Hallin and Xu, 1993; Bhatia, 2015; Malhotra, 2016; Mashaghi et al., 2016). Western blot analysis was performed to measure the expression levels of SP and NKA. As shown in Figures 1E,G, capsaicin upregulated the expressions of SP and NKA, while sinomenine pretreatment suppressed the expressions of SP and NKA. These data indicate that sinomenine inhibited capsaicin-induced increase in cough sensitivity in guinea pigs through mitigation of neurogenic inflammation.

Growing evidence indicate that TRPV1 plays important role in cough and other airway diseases (Lee et al., 2011; Millqvist, 2016). To determine if TRPV1 was involved in the regulation of cough by sinomenine, western blot was performed to determine the expression levels of TRPV1 in trachea and lung tissues. Compared with saline group, atomized capsaicin treatment significantly elevated the protein expression of TRPV1. In contrast, sinomenine pretreatment reduced the protein level of TRPV1 (Figure 2A). In addition, mRNA expression level of TRPV1 is shown in Figure 2B. Previous studies indicated that TRPV1 regulated intracellular calcium levels. As shown in Figure 2C, a specific calcium probe was used to determine changes in intracellular calcium levels after atomized capsaicin treatment as described earlier (Barreto-Chang and Dolmetsch, 2009). Results showed that sinomenine pretreatment significantly reduced the capsaicin-induced upregulation of calcium level. Furthermore, western blot analysis showed that overexpression of TRPV1 abolished the inhibitory effect of sinomenine on expressions of SP and NKA (Figure 2D). Moreover, overexpression of TRPV1 reversed the protective effect of sinomenine pretreatment and increased the levels of Penh while reducing CRS level in the guinea pigs (Figures 2E,F).

It is known that SOX5, a transcription factor expressed widely during development in several tissues, is a member of Sry-related HMG-box family (Zhang and Liu, 2017; Hu et al., 2018). Prediction data from Genecards² showed a binding site of SOX5 on the sequence of TRPV1. Firstly, the expression levels of SOX5 in neuronal cells and trachea and lung tissues were determined. Western blot analysis showed that the protein level of SOX5 was increased in trachea and lung tissues, and in neuronal cells after capsaicin treatment (Figures 3A,C). In addition, capsaicin administration upregulated the transcription levels of SOX5 gene in trachea and lung tissues, and in neuronal cells (Figures 3B,D). Furthermore, immunological staining showed that capsaicin treatment significantly elevated the expression level of SOX5 (Figures 3E,F). Then, the efficiency of SOX5 knockdown was determined. The si-RNA of SOX5 significantly decreased the mRNA level of SOX5 (Figure 4A). Capsaicin was added to neuronal cells which were transfected with si-SOX5 to block the expression of SOX5. Western blot analysis showed that capsaicin increased the protein level of TRPV1 (Figures 4B,C). However, siRNA interference of SOX5 significantly reduced the expression of TRPV1, suggesting that SOX5 is an important regulator of TRPV1. Moreover, luciferase assay indicated that SOX5 had no effect on the vector carrying mutant binding site of TRPV1 (TRPV1-Mut) which inhibited the activity of wild-type TRPV1 luciferase vector (TRPV1-WT) (Figure 4D). These results indicate that SOX5 transcription regulated TRPV1 expression.

Western blot analysis showed that sinomenine suppressed the level of SOX5, but SOX5 transfection significantly reversed the inhibitory effect of sinomenine on TRPV1 expression in neuronal cells (Figures 5A,B). Moreover, western blot analysis showed that overexpression of SOX5 abolished the effect of sinomenine and enhanced expressions of SP and NKA (Figures 5C,D). Intracellular calcium, which was reduced by sinomenine administration, was also upregulated after the forced expression of SOX5 (Figure 5E). These data indicate that sinomenine inhibited the expression of TRPV1 via downregulation of the protein level of SOX5 which acted as a transcription factor of TRPV1. Then, it was determined whether sinomenine regulated increased cough sensitivity in guinea pigs through SOX5/TRPV1. Western blot analysis showed that capsaicin-mediated upregulations of SOX5 and TRPV1 were attenuated after treatment of sinomenine, but these were reversed by SOX5 overexpression in lung tissues (Figures 6A,B). As shown in Figure 6C, CRS level was upregulated by capsaicin, but it was inhibited in sinomenine pretreatment group (Figure 6C). However, SOX5 attenuated the effect of sinomenine and increased CRS level (Figure 6C). Penh level was also measured. It was revealed that SOX5 significantly increased the level of Penh, but this effect was decreased by sinomenine pretreatment (Figure 6D). Besides, results from H&E staining showed that SOX5 abolished the sinomenine-induced inhibition of inflammatory cell infiltration (Figure 6E). Immunohistochemical staining analysis indicated that the number of CD68-positive monocytes/macrophages was also increased by SOX5 which reversed the inhibitory effect of sinomenine (Figures 6F,I). As shown in Figures 6G,H, the expressions of SP and NKA were also restored through SOX5 over-expression.