Acidied Pepsin Promotes Laryngeal Precancerosis by Upregulating H+/K+-ATPase and Activating Mitophagy in Laryngeal Epithelial Cells

Background: Although laryngopharyngeal reux (LPR) has been implicated in various upper aerodigestive tract and laryngeal diseases, the underlying mechanisms remain elusive. In this study, we investigated the role of gastric acidied pepsin in laryngeal precancerosis. Results: Acidied pepsin (pH=3) enhanced the growth and survival of mouse laryngeal epithelial cells in vitro and promoted laryngeal mucosal thickening and laryngeal epithelial cell growth in vivo. Furthermore, acidied pepsin promoted autophagy/mitophagy induction, accompanied by a signicant decrease in mitochondrial membrane potential (MMP). Inhibition of autophagy by chloroquine abolished the ability of acidied pepsin to promote mitophagy and cell growth in laryngeal epithelial cells. Additionally, chloroquine promoted cell apoptosis and further reduced MMP in laryngeal epithelial cells treated with acidied pepsin. The expression levels of pepsin and H + /K + -ATPase α and β subunits in 31 human laryngeal mucosa specimens were 51.6%, 48.4%, and 48.4%, respectively. Importantly, the pepsin level was correlated with the H + /K + -ATPase β subunit level. H + /K + -ATPase upregulation in laryngeal epithelial cells in response to acidied pepsin was essential for the mitophagy-promoting effect of acidied pepsin. H + /K + -ATPase knockout or inhibition further reduced MMP in the presence of acidied pepsin. Conclusions: Our ndings suggest that in an acidic environment, pepsin promotes laryngeal epithelial cell growth and survival by upregulating H+/K+-ATPase and activating mitophagy, potentially leading to laryngeal precancerosis.

and pepsin has been shown to promote cancer development and progression in rat models [15], suggesting that acidi ed pepsin may play a role in brosis and carcinogenesis. Additionally, salivary pepsin has emerged as a novel biomarker of LPR [16], impairing the function of the upper respiratory tract and initiating in ammatory changes within the larynx [17]. A recent study con rmed that LPR is common among patients with neoplastic lesions and is a signi cant risk factor for squamous cell carcinoma of the pharynx or larynx [18]. Given the pro-in ammatory and pro-tumorigenic effects of acidi ed pepsin, we hypothesized that pepsin is a key determinant of the LPR-mediated alternations in the laryngeal mucosa.
In the present study, we assessed the expression pattern and role of acidi ed pepsin in laryngeal epithelial cell apoptosis and proliferation. We also investigated the effect of acidi ed pepsin on mitophagy in laryngeal epithelial cells. Furthermore, we explored the relevance of H + /K + -ATPase in pepsin-mediated alterations in the proliferation and mitophagy of laryngeal epithelial cells. The ndings of this study provide further insight into the roles of acidi ed pepsin and H + /K + -ATPase in laryngeal cancer.

Results
Acidi ed pepsin enhances laryngeal epithelial cell growthin vitroandin vivo Non-acidic pepsin (pH = 7) has been shown to increase the proliferation of laryngeal carcinoma cells [21].
We previously showed that the pepsin level was elevated in vocal cord polyps and leukoplakia and was correlated with the grade of dysplasia [22]. Here, we assessed the effects of acidic pepsin on the growth and viability of laryngeal epithelial cells and found that at low (pH = 3) and moderate (pH = 5) pH, increasing pepsin levels signi cantly increased cell viability (Fig. 1A). However, neutral pepsin (pH = 7) and inactivated pepsin (pH = 7) did not affect laryngeal epithelial cell proliferation or viability. These ndings suggest that only activated pepsin can promote laryngeal epithelial cell proliferation. Thus, in subsequent experiments, we used acidi ed pepsin (pH = 3, 0.05-0.5 mg/mL).
Next, we performed ow cytometry to investigate the effect of acidi ed pepsin on cell cycle in laryngeal epithelial cells. Although the population of cells in S phase was decreased under acidic conditions (pH = 3), treatment with high-dose acidi ed pepsin (pH = 3; 0.1-0.5 mg/mL) signi cantly increased the proportion of cells in S phase (P < 0.01; Fig. 1B). Cell proliferation was also analyzed by Ki67 staining. In line with the ow cytometry ndings, the number of Ki67-positive cells was lower under acidic than neutral conditions; medium and high doses of pepsin signi cantly increased the number of Ki67-positive cells (Fig. 1C). Low (0.05 mg/mL) and medium (0.1 mg/mL) concentrations of acidi ed pepsin had no effect on laryngeal epithelial cell apoptosis (Fig. 1D, E). However, a high dose (0.5 mg/mL) of acidi ed pepsin signi cantly increased the active caspase-3 level and promoted apoptosis in laryngeal epithelial cells (Fig. 1D, E). Given that the medium concentration (0.1 mg/mL) of pepsin promoted cell proliferation without signi cantly affecting apoptosis, we selected this dose to further explore the effect of pepsin on the growth of laryngeal epithelial cells.
To con rm the effects of pepsin on the laryngeal mucosa in vivo, we established an acid perfusion model using acidi ed pepsin and HCl (pH = 3), given that acidic medium (pH = 3) did not affect cell viability in laryngeal epithelial cells in vitro. Acidi ed pepsin signi cantly increased the thickness of the laryngeal mucosa and the number of Ki67-positive cells in the larynx ( Fig. 2A, B), in contrast to perfusion with HCl alone ( Fig. 2A, B). In accordance with our in vitro ndings, HCl and acidi ed pepsin did not affect the proportion of apoptotic cells in the laryngeal mucosa (Fig. 2C). These ndings suggest that acidi ed pepsin induces laryngeal epithelial cell proliferation without affecting cell survival.
2Acidi ed pepsin induces mitophagy and mitochondrial injury in laryngeal epithelial cells It has become increasingly evident that autophagy has a dual role in cell growth [23]. Mitophagy is a form of selective autophagy responsible for removing dysfunctional mitochondria via the autophagosomelysosome system [24]. To assess the effect of acidi ed pepsin on autophagy in laryngeal epithelial cells, we evaluated autophagic ux by measuring microtubule-associated protein light chain 3 (LC3) levels. Stronger LC3 signals were observed in acidi ed pepsin-treated cells than in those treated with acidic medium alone (Fig. 3A), suggesting that acidi ed pepsin promotes the formation of autolysosomes.
Consistently, Western blotting revealed that treatment with acidi ed pepsin increased the LC3II/LC3I ratio and decreased p62 levels in laryngeal epithelial cells (Fig. 3B). Additionally, acidi ed pepsin signi cantly increased the uorescence intensity of a mitophagy dye probe and the number of autophagosomes in mitochondria (Fig. 3C, D). These ndings indicate that acidi ed pepsin induces autophagy and mitophagy in laryngeal epithelial cells. Furthermore, we found that treatment with acidi ed pepsin signi cantly decreased MMP (Fig. 3E), which is essential for energy storage during oxidative phosphorylation and for the elimination of dysfunctional mitochondria [25]. Overall, these results suggest that acidi ed pepsin induces mitophagy and the accumulation of dysfunctional mitochondria accompanied by the reduction in MMP in laryngeal epithelial cells.
Autophagy inhibition abolishes acidi ed pepsin-induced cell growth in laryngeal epithelial cells Considering the role of mitophagy/autophagy in cell growth and differentiation [26], we investigated the relevance of autophagy in the growth-promoting effects of acidi ed pepsin in laryngeal epithelial cells. The autophagy inhibitor CQ increases the pH of lysosomes and prevents the fusion of lysosomes and autophagosomes. CQ treatment (at pH = 3) of laryngeal epithelial cells signi cantly decreased autophagic ux, as shown by the increased LC3II and p62 levels (Fig. 4A). Importantly, autophagy inhibition by CQ signi cantly abolished the effects of acidi ed pepsin on the LC3II/LC3I ratio and p62 levels (Fig. 4A). Consistently, CQ treatment signi cantly suppressed the increase in mitophagy in response to acidi ed pepsin (Fig. 4B). Although CQ treatment alone had little effect on the proportion of J-monomer-positive cells, cells treated with CQ combined with pepsin had a signi cantly higher number of J-monomer-positive cells than that of cells treated with acidi ed pepsin alone (Fig. 4C), suggesting that autophagy inhibition by CQ augments the effect of pepsin on MMP. Interestingly, CQ treatment signi cantly abolished the ability of acidi ed pepsin to promote the growth of laryngeal epithelial cells (Fig. 4D). Although acidi ed pepsin alone had no effect on laryngeal cell apoptosis, the combination of CQ with acidi ed pepsin profoundly enhanced cell apoptosis, as re ected by the increase in the cleaved caspase-3 level (Fig. 4E, F). These ndings suggest that mitophagy plays a critical role in acidi ed pepsin-induced laryngeal cell growth.
H + /K + -ATPase is essential for the acidi ed pepsin-mediated induction of autophagy/mitophagy The H + /K + -ATPase proton pump expressed in laryngeal cells has been implicated in the abnormal mucus secretion frequently seen in patients with chronic laryngitis, since laryngeal epithelial cells are more sensitive to alterations in pH than are esophageal epithelial cells [27,28]. To assess the relationship between H + /K + -ATPase and pepsin in normal laryngeal mucosa, we analyzed normal laryngeal mucosal specimens obtained from 31 patients who underwent 24 h combined multichannel intraluminal impedance and pH monitoring preoperatively. pH monitoring indicated GERD in seven patients. Six patients exhibited an RSI > 13, and two patients an RSF > 7. However, there were no signi cant correlations among pH monitoring ndings, RSI, and RSF (P > 0.05; Table 1). IHC revealed that pepsin and H + /K + -ATPase α and β subunits were expressed in 51.6%, 41.9%, and 48.4% of laryngeal mucosal cells, respectively (Fig. 5). There was a signi cant correlation between the levels of the H + /K + -ATPase β and α subunits, as well as between the levels Table 1 The correlations among pH monitoring ndings, RSI, and RSF of H + /K + -ATPase β and pepsin (P < 0.01, Table 2). IHC in tissues from an LPR mouse model con rmed the elevated expression of H + /K + -ATPase α and β subunits in the laryngeal mucosa in response to acidi ed pepsin (Fig. S1A, B). Consistently, mouse laryngeal epithelial cells treated with acidi ed pepsin exhibited a higher H + /K + -ATPase uorescence intensity (for both α and β subunits) than that in cells treated with acidic medium (pH = 3) alone (Fig. S1C, D). qRT-PCR and Western blotting con rmed the acidi ed pepsinmediated upregulation of H + /K + -ATPase α and β subunits at the mRNA and protein levels, respectively (Fig. S1E, F). Table 2 The correlations among pepsin, H+/K+-ATPase α and β subunits The β subunit of H + /K + -ATPase has a regulatory role, whereas the α subunit is the catalytic core of the pump. To investigate the relevance of H + /K + -ATPase in acidi ed pepsin-induced laryngeal epithelial cell growth, we established an H + /K + -ATPase-α-subunit-KO laryngeal cell line using CRISPR/Cas9 (Fig. S2A).
Expectedly, acidi ed pepsin did not upregulate the H + /K + -ATPase α subunit in the H + /K + -ATPase-KO laryngeal epithelial cells (Fig. S2A). In contrast to wild-type laryngeal epithelial cells, acidi ed pepsin exposure in H + /K + -ATPase-α-subunit-KO laryngeal epithelial cells did not promote cell growth but signi cantly enhanced cell apoptosis, as indicated by the increased level of cleaved caspase-3 ( Fig. S2B-D).
Given the similar effects of H + /K + -ATPase deletion and autophagy inhibition on pepsin-mediated cell growth, we investigated the effect of H + /K + -ATPase α subunit depletion on acidi ed pepsin-induced autophagy. Acidi ed pepsin treatment in wild-type laryngeal cells signi cantly increased H + /K + -ATPase expression and LC3 accumulation; in these cells, H + /K + -ATPase largely co-localized with LC3 ( Fig. 6A).
However, H + /K + -ATPase depletion abrogated the acidi ed pepsin-induced formation of autophagosomes and accumulation of LC3 (Fig. 6B). Additionally, acidi ed pepsin treatment in H + /K + -ATPase-KO laryngeal epithelial cells failed to induce mitophagy and reduce the p62 level (Fig. 6C, D). Consistent with the effects of CQ treatment, H + /K + -ATPase depletion reduced MMP in laryngeal epithelial cells, possibly due to the decrease in mitophagic activity (Fig. 6E). These ndings suggest that mitophagy induction in laryngeal epithelial cells in response to acidi ed pepsin is dependent on H + /K + -ATPase.
Proton pump inhibition by pantoprazole suppresses mitophagy induction and laryngeal epithelial cell growth in response to acidi ed pepsin PPIs targeting H + /K + -ATPases are widely used to treat acid-related diseases, including LPR. To assess the effect of proton pump inhibition on acidi ed pepsin-mediated laryngeal epithelial cell growth, we treated laryngeal epithelial cells with different PPIs. Treatment with omeprazole (10 and 50 µg/mL), lansoprazole, rabeprazole, esomeprazole, and SCH-28080, but not pantoprazole, signi cantly reduced laryngeal epithelial cell growth (P < 0.05; Fig. S3). The combination of acidi ed pepsin with each of the PPIs signi cantly suppressed the pepsin-mediated increase in laryngeal epithelial cell growth (P < 0.01), with pantoprazole (50 µg/mL), rabeprazole, and esomeprazole providing the most potent growth inhibitory effects (Fig. S3). These results indicate that pantoprazole abrogates acidi ed pepsin-induced cell growth in laryngeal epithelial cells without signi cantly affecting the growth of normal laryngeal epithelial cells. Consistent with our ndings in H + /K + -ATPase-KO cells, pantoprazole treatment suppressed autophagy/mitophagy induction in response to acidi ed pepsin, re ected by the increased LC3II/LC3I ratio and p62 level (Fig. S4A, B). However, no changes in these indicators were observed in cells treated with pantoprazole alone.
Similarly, although treatment with pantoprazole alone did not affect MMP in laryngeal epithelial cells (at pH = 3), pantoprazole hindered the ability of acidi ed pepsin to decrease MMP in laryngeal epithelial cells (Fig. 7A). In addition, pantoprazole signi cantly abrogated the effect of acidi ed pepsin on cell viability in laryngeal epithelial cells (Fig. 7B). Consistently, cells treated with pantoprazole combined with acidi ed pepsin exhibited enhanced apoptosis and a higher level of cleaved caspase-3 compared with cells treated with pepsin alone (Fig. 7C, D). These ndings suggest that proton pump inhibition by pantoprazole suppresses mitophagy induction and laryngeal epithelial cell growth in response to acidi ed pepsin.
Pantoprazole suppresses acidi ed pepsin-induced cell proliferation in the laryngeal mucosain vivo Next, we assessed the effects of pantoprazole combined with acidi ed pepsin on laryngeal mucosal cell proliferation in vivo. Acidi ed pepsin alone resulted in profound laryngeal mucosa thickening and increased Ki67 expression in the laryngeal mucosa (Fig. 8A, B) but did not induce cell apoptosis in the laryngeal mucosa. The combination of pantoprazole with acidi ed pepsin signi cantly suppressed the thickening of the laryngeal mucosa and increased Ki67 expression (Fig. 8A-C). Additionally, pantoprazole markedly enhanced cell apoptosis in the laryngeal mucosa when used in combination with acidi ed pepsin (Fig. 8C). These results indicate that proton pump inhibition by pantoprazole suppresses acidi ed pepsin-induced laryngeal mucosa alterations by inducing apoptosis and inhibiting proliferation of laryngeal epithelial cells.

Discussion
In this study, we found that pepsin levels were elevated in normal laryngeal mucosal specimens obtained from laryngeal carcinoma patients and that acidi ed pepsin promoted laryngeal cell survival and proliferation via mitophagy induction. We also identi ed H + /K + -ATPase as a link between pepsin and mitophagy induction and showed that proton pump inhibition impaired laryngeal epithelial cell growth in the presence of acidi ed pepsin. PPIs are commonly used to treat acid-related diseases, including GERD and LPR [29]. However, the e cacy of PPIs in LPR remains poorly demonstrated due to the in uence of non-acid or mixed re ux, such as pepsin or bile salt re ux. The ndings presented here provide strong evidence supporting the ability of acidi ed pepsin to promote cell proliferation, which possibly hinders the therapeutic effect of PPIs on LPR. Our ndings also suggest that the combination of PPIs with pepsin-targeting agents may improve the treatment of LPR and prevent acidi ed pepsin-mediated in ammation and damage in laryngeal epithelial cells.
Gastric pepsin internalized by airway epithelial cells promotes cell damage, oxidative stress, and in ammation, which could lead to cancer development [30,31]. Acid and pepsin have also been shown to induce gastric epithelial cell damage in rat models, and administration of epidermal growth factor or sucralfate prevented acid/pepsin-induced damage in rat gastric cells [32]. Furthermore, pepsin has been demonstrated to increase the risk of laryngeal carcinoma development by promoting IL-8-induced epithelial-mesenchymal transition [33]. Non-acidic pepsin (pH = 7) has also been shown to increase the proliferation of laryngeal epithelial cells and human hypopharyngeal squamous cell carcinoma cells [21]. In contrast to these reports, we found that non-acidic pepsin had no effect on the viability of laryngeal epithelial cells. On the other hand, we found that a high dose of acidic pepsin promoted laryngeal epithelial cell proliferation and survival. Although low doses of acidic pepsin enhanced the proliferation of laryngeal epithelial cells, they failed to promote cell survival in vitro or in vivo. These ndings suggest that the effects of acidi ed pepsin on laryngeal epithelial cell viability are concentration dependent.
Autophagy exerted cytoprotective effects in laryngeal squamous cell carcinoma cells treated with recombinant human arginase, which induced cytotoxicity [34]. Similarly, mitophagy induction after exposure to hydrogen peroxide has been demonstrated to drive cell survival in laryngeal cancer cells, promoting laryngeal cancer progression [35]. By activating autophagy and inhibiting the release of gliadin peptides, gastrointestinal and pancreatic enzymes alleviate celiac disease [36]. Here, we show that treatment with acidi ed pepsin promoted autophagy and mitophagy in laryngeal epithelial cells. Therefore, the ability of pepsin to promote laryngeal epithelial cell survival might be linked to the induction of mitophagy.
Alterations in MMP strongly indicate cell dysfunction [37]. Notably, MMP disruption has been shown to induce mitochondrial damage and mitophagy in human gastric epithelial cells [38]. In this study, we found that, in addition to inducing mitophagy, pepsin treatment signi cantly reduced MMP in laryngeal epithelial cells. Mitochondrial dysfunction due to alterations in MMP has been shown to promote mitochondrial biogenesis and induce mitophagy [39,40], further supporting that the reduction in MMP by acidi ed pepsin might be linked to mitophagy induction. Mitophagy is critical for removing dysfunctional mitochondria, promoting cell survival, and protecting cells from oxidative stress [41][42][43]. Typically, MMP alterations activate the mitochondria-mediated apoptosis pathway [44]. Additionally, mitophagy inhibition has been shown to induce apoptosis in epithelial cells [45]. In this study, we found that mitophagy inhibition using CQ promoted the accumulation of dysfunctional mitochondria, further reducing the MMP and ultimately causing apoptosis. Thus, pharmacological inhibition of autophagy may represent a promising strategy to treat laryngeal cancer [46]. Additionally, mitophagy inhibition may suppress acidi ed pepsin-induced cell proliferation, preventing the development of precancerous lesions in the larynx.
The H + /K + -ATPase proton pump comprises α and β subunits [47]. Gastric H + /K + -ATPase plays a key role in regulating gastric acid secretion [48]. H + /K + -ATPase is strongly expressed in the laryngeal mucosa [27,49] and is elevated in LPR and laryngeal cancer patients [28]. In this study, we con rmed high H + /K + -ATPase expression levels in the laryngeal mucosa. Interestingly, a diet high in phytate has been shown to reduce pepsin activity, thereby downregulating H + /K + -ATPase expression [50]. Consistent with these ndings, we found that pepsin treatment increased H + /K + -ATPase levels. Similar to H + /K + -ATPase, ATP13A2 is a lysosomal P-type ATPase. Notably, ATP13A2 de ciency has been shown to decrease autophagic ux accompanied by accumulation of mitochondrial mass, implying that proton pump inhibition induces mitochondrial injury [51]. Here, we show that H+/K+-ATPase depletion markedly reduced autolysosome formation and autophagic ux in pepsin-treated cells, as well as promoted mitochondrial damage and MMP disruption, which led to apoptosis. These ndings strongly suggest that acidi ed pepsin-induced mitophagy is dependent on H + /K + -ATPase.
Phagophores play an essential role in autophagy [52]. After nucleation, the membrane of phagophores expands to generate autophagosomes, which mature into autolysosomes that function to degrade different cellular components [53]. During autophagy, cytosolic LC3I is conjugated to phosphatidylethanolamine to form LC3II, which is recruited to the autophagosomal membrane [54]. After fusing with lysosomes, LC3II in the autolysosomal lumen is degraded along with other intraautophagosomal components [55]. p62, commonly used as a marker of autophagy, is also degraded during autophagy [56]. In this study, we found that acidi ed pepsin increased LC3II and decreased p62 levels. Moreover, we found that LC3 co-localized with H + /K + -ATPase in cells undergoing autophagy, suggesting the involvement of H + /K + -ATPase in the process of autophagosome/autolysosome formation in cells treated with acidi ed pepsin. Inhibition of autolysosome formation by CQ results in the accumulation of p62 and LC3II [57]. Here, we show that H + /K + -ATPase depletion abrogated mitophagy induction induced by acidi ed pepsin, as indicated by the increased levels of LC3II and p62. It is possible that H + /K + -ATPase depletion inhibits mitophagy by impairing autolysosome formation. However, future studies are required to elucidate the role of H + /K + -ATPase in autolysosome formation.
Additionally, proton pump inhibition by pantoprazole inhibited mitophagy, enhanced mitochondrial damage, and promoted mitochondrial-mediated apoptosis in laryngeal epithelial cells treated with acidi ed pepsin. Consistent with these ndings, pantoprazole has been previously shown to augment the anti-tumor effect of docetaxel by inhibiting autophagy [58]. Hence, PPIs may prevent the development of LPR-associated laryngeal precancerous lesions by activating mitochondrial-mediated apoptosis. However, the ability of PPIs to prevent laryngeal cancer development and progression requires further investigation. Furthermore, given the importance of H + /K + -ATPase in gastric acid release, the role of H + /K + -ATPase in establishing an acidic environment in the larynx and the relevance of this effect in the growth-promoting effects of pepsin merit further investigation.

Conclusions
In this study, we found that pepsin levels were elevated in the laryngeal mucosa, promoting laryngeal epithelial cell growth and survival by activating autophagy and mitophagy under acidic conditions. We also found that acidi ed pepsin-mediated mitophagy ux was dependent on the proton pump H + /K + -ATPase. These ndings strongly support that inhibition of mitophagy and proton pumps may represent a promising approach to treat patients with LPR and prevent the development of laryngeal cancer.

Methods
This study was approved by the Institutional Review Board of The First A liated Hospital, College of Medicine, Zhejiang University, China. All authors had access to the study data and reviewed and approved the nal manuscript. The informed consent was obtained for experimentation with human subjects. The animals' care and use involving experiments were in accordance with institution guidelines for the care and use of laboratory animals.

Patients and patient samples
Specimens (n=31) were acquired from normal laryngeal mucosa far from the negative margin during open partial or total laryngectomy for tumor excision or microsurgery for benign laryngeal lesions. Operations were performed in the Department of Otolaryngology of The First A liated Hospital, College of Medicine, Zhejiang University between October 2018 and May 2019. Specimens were frozen immediately in liquid nitrogen following surgical removal and stored until further analysis. Among the 31 patients, 28 were males, and 3 were females, with a mean age of 58.1 (range, 32-81) years. Exclusion criteria were as follows: 1) preoperative radiotherapy, chemotherapy, or immunotherapy, 2) treatment with PPIs, and 3) underlying metabolic or autoimmune diseases.
All patients voluntarily received 24 h combined multichannel intraluminal impedance and pH monitoring (MMS Ohmega Ambulatory Impedance and pH Recorder, the Netherlands preoperatively (after 4 h of fasting) and completed the self-administered re ux symptom index (RSI; Chinese version, 2015), which contains nine items. The maximum RSI was 45, with an RSI >13 considered suggestive of LPR. All patients underwent laryngoscopic examination and evaluation for the re ux nding score (RFS). The RFS is based on eight components (maximum score, 26), and an RFS >7 was considered suggestive of LPR. Abnormal gastroesophageal re ux was de ned as a DeMeester score of ≥14.7 and syndrome association probability of ≥95% (gastroesophageal acid re ux) or syndrome association probability ≥95% (abnormal non-acid gastroesophageal re ux).

Animal models
Seven-week-old male C57BL/6 mice (20-25 g) were purchased from Hubei Provincial Center for Disease Control and Prevention. Mice were housed in a speci c-pathogen-free room with a controlled temperature (20 ± 2°C). The LPR animal model was established as described previously [19,20]. All mice were maintained under a 12 h light/dark cycle with free access to standard rodent chow and water. After 1 week of acclimation, the mice were randomly divided into three groups: control (treated with PBS), HCl (pH=3) treatment, and acidi ed pepsin (pH=3) treatment. HCl and acidi ed pepsin were administered by esophageal acid perfusion. Brie y, mice were anesthetized by intraperitoneal injection of pentobarbital sodium (Sigma, P3761). Subsequently, 0.15 mL HCl (1 M, pH=3) with or without 2.5 mg/g pepsin (SINOPHARM, 7647-01-0) was perfused (8 drops/min) into the mid-and lower esophagus using a feeding tube. Drug perfusion was performed twice daily for 45 consecutive days. Control mice were perfused with PBS instead of HCl. After treatment, throat specimens were obtained and washed with 0.9% saline solution. The specimens were stored at −80°C until further analysis.
For H&E staining, 5-μM-thick sections were depara nized, rehydrated, and stained with hematoxylin (Sigma, H9627) for 5 min. After plating in HCl alcohol for 10 s, the sections were stained with 0.5% eosin for 1 min, incubated with ammonium hydroxide for 10 s, washed under running water, and mounted in neutral balsam. Experienced pathologists evaluated the histology.
Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay Apoptosis in the throat tissues of mice was assessed using the In Situ Cell Death Detection Kit-POD (Roche, Shanghai, China). Brie y, 5-μM-thick sections were depara nized, rehydrated, and incubated with proteinase for 25 min. After a 20-min incubation in cell-penetrating solution, sections were incubated in TUNEL reaction mixture at 37°C for 2 h in the dark. Subsequently, the sections were stained with 4,'6diamidino-2-phenylindole (DAPI), mounted in mounting medium, and observed under a uorescence microscope.
Isolation oflaryngeal epithelial cells Primary laryngeal epithelial cells were isolated from C57BL/6 mice. Brie y, mice were anesthetized by intraperitoneal injection of pentobarbital sodium (Sigma, P3761). Laryngeal mucosal specimens were obtained under a dissecting microscope and washed with Hanks' solution on ice. After cutting into 1 mm 3 blocks, the specimens were treated with 10 mg/mL dispase II (Sigma-Aldrich, St. Louis, MO, USA) for 48 h at 4°C. The cell pellet was enzymatically dissociated in 0.05% Trypsin/EDTA at 37°C for 2-3 min and resuspended in 2 mL RPMI-1640 (SH30809.01; HyClone, USA) medium. The medium was replaced every 3 days, and the cells were expanded upon reaching 90% con uence.

Generation of H + /K + -ATPase-α-knockout (KO) cells and treatment
To establish an H + /K + -ATPase-α-subunit-KO cell line using CRISPR, we cloned a guide RNA (gRNA) targeting the H + /K + -ATPase α subunit (designed and synthesized by RIO Biotech) into the pUC57-T7-gRNA plasmid. The gRNA was ampli ed (forward, CACCGTATCAGACCAGCGCCACCA; reverse, AAACTGGTGGCGCTGGTCTGATAC), and the target and control gRNAs were cloned into the BsaI sites of the pUC57-T7-gRNA plasmid. Plasmids were transfected into laryngeal epithelial cells using Lipofectamine 2000 reagent (Invitrogen). After selecting for the transfected (puromycin-resistant) cells and con rming the knockout e ciency by western blotting, cells were maintained in RPMI-1640 supplemented with 10% fetal bovine serum (HyClone, Waltham, MA, USA) at 37°C in a 5% CO 2 humidi ed atmosphere. KO cells were treated with acidic medium (pH=3 or 5) or normal medium (pH=7) with or without pepsin for 1 h, followed by a 24 h incubation in fresh RPMI-1640. Pepsin was inactivated using 10 M NaOH (pH=8) at 37°C for 30 min, and the pH of the medium was adjusted to 7 using 1 M HCl. To induce autophagy and inhibit proton pumps, we also treated cells with 20 μM chloroquine (CQ) and 10 μg/mL pantoprazole, respectively.

CCK-8 assay
Cells were seeded in 96-well plates (5×10 5 /well) and treated with acidic medium containing pepsin and CQ/pantoprazole for 24 h. Subsequently, 20 µL cell counting solution (Beyotime Biological Technology Co. Ltd, C0037) was added, and the cells were incubated in the dark for an additional 4 h. The optical absorbance at 450 nm was measured using the Spectra Plus microplate reader (Multiskan MK3, Thermo).

Flow cytometry
For cell cycle analysis, we xed cells in 700 µL ice-cold 80% ethanol at 4°C for at least 4 h. After incubating with 10 μL RNase (1 mg/mL) and 10 μL propidium iodide (APOAF, Sigma) in the dark, the cells were analyzed by ow cytometry. For apoptosis measurement, we incubated the cells in 500 µL Annexin V Binding Buffer. After adding 5 μL Annexin V-FITC and 5 μL propidium iodide, the cells were incubated for 10 min at room temperature in the dark. The percentage of apoptotic cells was determined by ow cytometry. To measure mitochondrial membrane potential (MMP), we incubated the cells in 1000 μL JC-1 solution (Beyotime Biological Technology) for 20 min at 37°C. After centrifugation, the cells were resuspended in 200 μL 10× incubation buffer and analyzed by ow cytometry. All ow cytometry data were analyzed using ModFit LT software (Becton Dickinson, Mountain View, CA, USA).
Glyceraldehyde 3-phosphate dehydrogenase (GAPDH; ab181602, Abcam) served as a loading control. After incubating with the appropriate secondary antibodies for 1 h, the signal was developed using an enhanced chemiluminescence assay kit (Beyotime Biological Technology) and visualized using the ChemiDoc XRS+ System (Bio-Rad Laboratories, Hercules, CA, USA).
Immuno uorescence (IF) assay LC3 and H + /K + -ATPase (α subunit) expression was assessed by IF. Brie y, cells were grown on glass slides for 24 h, xed in 3% paraformaldehyde for 30 min at 4°C, and neutralized with 50 mM NH 4 Cl. After permeabilizing with 0.1% Triton X-100 for 15 min and washing with PBS, cells were incubated with antibodies against H + /K + -ATPase (α subunit) and LC3 at room temperature for 1 h, followed by incubation with the appropriate secondary antibody for 1 h and then with DAPI for 5 min. Stained cells were visualized under the LSM5 EXCITER laser scanning confocal microscope (Carl Zeiss, Oberkochen, Germany).
GFP-RFP-LC3 expression was also assessed by IF. Cells were treated with pepsin, CQ, and pantoprazole, followed by a 24 h incubation with adenoviruses carrying GFP-RFP-LC3. After xing with 4% paraformaldehyde, LC3 expression was assessed by confocal laser scanning microscopy (LSM 800; Carl Zeiss). For analysis of mitophagy, cells were seeded on glass slides and, after the appropriate treatments, incubated with 100 nmol/L mitophagy dye for 30 min at 37°C. Fluorescence was measured by confocal laser scanning microscopy.

Transmission electron microscopy
Cells grown on glass slides were xed in 2.5% glutaraldehyde, post-xed in 1% osmium tetroxide, and gradually dehydrated in ethanol and acetone. Subsequently, the cells were embedded in epoxy resin and stained with uranyl acetate and lead citrate. Autophagy was assessed by transmission electron microscopy (HITACHI, HT7700-SS, Japan).

Statistical analysis
All experiments were independently performed at least three times. Data are expressed as means ± standard error. Statistical analyses were performed using SPSS 25.0 software (IBM Corp., Armonk, NY, USA) or GraphPad Prism 6.0 (GraphPad Software Inc., San Diego, CA, USA). IHC and IF images were analyzed using Image-Pro Plus 6.0 software. Statistical signi cance was determined using one-way analysis of variance (ANOVA) or Student's t-test. P-values <0.05 were considered statistically signi cant.