The NF-κB pathway plays a vital role in rat salivary gland atrophy model

Objective The aim of this study was to explore the histopathological and genetic changes in the submandibular glands after duct ligation and provide important clues to functional regeneration. Design We established a rat salivary gland duct ligation model and observed pathological changes in the rat submandibular gland on day 1 and weeks 1, 2, 3, and 4 using hematoxylin and eosin staining, Alcian blue–periodic acid Schiff staining, Masson staining, terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL), and immunohistochemical staining. RNA sequencing was performed on normal salivary glands and those from the ligation model after 1 week. Significantly differentially expressed genes were selected, and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed. Results Apoptosis levels and histological and functional KEGG pathway analyses showed that injury to the salivary gland after ligation gradually increased. The TGF-β pathway was activated and promoted fibrosis. RNA sequencing results and further verification of samples at week 1 showed that the NF-κB pathway plays a vital role in salivary gland atrophy. Conclusions Our results detailed the pathological changes in the submandibular gland after ligation and the important functions of the NF-κB pathway.


Introduction
Both the major and minor salivary glands produce and secrete digestive and protein-rich fluids in the oral cavity. However, three pairs of major salivary glands (parotid, submandibular, and sublingual) produce over 90% of the total saliva. Saliva assists in chewing and swallowing food, and digestive enzymes initiate digestion. Saliva is also important for speech and taste because it moisturizes the tongue and oral mucosa [1,2]. Healthy salivary glands are essential for maintaining oral and overall health. Salivary gland hypofunction is mainly manifested by a reduced salivary flow rate and reduced salivary secretion, which leads to oral mucosal dryness and discomfort, oral burning, and thirst. Long-term salivary gland hypofunction can alter the oral microenvironment, leading to severe dental erosion, mucosal lesions, or fungal infections [3]. Impairment of oral function may change a patient's eating preference, eventually leading to decreased masticatory muscle function, malabsorption, and nutritional deficiency [4].
Pathological conditions, such as obstructive factors, chronic heart failure, dementia, chronic kidney disease, psoriasis, and autoimmune diseases, such as Sjögren's syndrome and Hashimoto's disease, alter the function of salivary glands, which affects quality of life [5][6][7][8]. Obstructive sialadenitis and sialolithiasis induce chronic or acute inflammation and swelling of the salivary gland due to obstruction of the salivary duct [9]. Radiation therapy is commonly used to treat head and neck cancer and can impair the salivary glands in the radiation therapy region, eventually leading to salivary gland dysfunction and atrophy [10,11]. Autoimmune diseases, such as Sjögren's syndrome, induce lymphocytic infiltration and fibrosis, leading to salivary gland hypofunction [12].
Current clinical treatments for salivary gland injury temporarily alleviate but do not restore the secretory function of the salivary gland [13]. Thus, experimental animal salivary gland injury models are required to develop new therapeutic strategies. After ligation of the main excretory duct, salivary glands in animal models appear to have symptoms similar to those of clinical patients. A gradual increase in intraductal pressure is the main cause of salivary gland atrophy with duct ligation, affecting the acinar and ductal epithelial cells [14]. Cell proliferation markers are upregulated when mesenchymal stem cells are used to treat the necrotic tissue of salivary glands in a rat submandibular gland duct ligation model [15]. The regeneration process is initiated by duct ligation as part of self-repair, thus activating Wnt/β-catenin signaling, which then promotes regeneration of the salivary gland [16].
Ligation of the salivary gland causes progressive atrophy of acinar cells. However, previous studies are mostly descriptive, and the pathological changes and mechanisms of the salivary glands have not been extensively reported. Damage to the parotid gland of a rabbit after ligation is divided into four stages with overlapping features [17]. In a mouse submandibular gland ligation model, fibrosis-induced TGF-β signaling was observed [18]. The level of transcription factor p63, which promotes regeneration, is upregulated after duct ligation and irradiation [19].
Above all, it is difficult to imagine any therapeutic approach for the treatment of the pathology presented; however, knowing the mechanism is the first step. In the present study, we aimed to investigate the histopathological and genetic changes caused by salivary gland injury in a rat submandibular gland ligation model to generate data to help develop better therapies for salivary gland injuries.

Animal model
Male Wistar rats aged 10 weeks (200-250 g) were purchased from SPF Biotechnology Co. (Beijing, China) and kept in the animal house of Capital Medical University under standard conditions. The rats had free access to food and water, and ligation was performed after 1 week of habituation. Animal experiments were performed with the approval of the Animal Welfare Management Association (approval number: AEEI-2020-124).
The number of animals used in this project was the number necessary to obtain valid and meaningful results by referring to a relevant study [20]. Each group required at least six rats. Animals were randomly separated into six groups: the NT group (without animal treatment) and five experimental groups (1 day, 1 week, 2 week, 3 week, and 4 week, with ligation of the left duct of the submandibular gland). Ligation of the submandibular gland duct was performed by a skilled oral clinician. Animals were anesthetized with ketamine and the neck region was disinfected. The submandibular gland, duct, peripheral vasculature, and nerves were exposed. We then used a non-absorbable suture to ligate the main duct. Animals were observed and sacrificed at different timepoints. The salivary flow rate was tested by injecting pilocarpine intraperitoneally (5 mg/kg), and saliva was collected into sterile 1.5-mL tubes for 5 min [21]. The submandibular glands on both sides were collected and weighted, and then divided into several sections for further experiments.

Hematoxylin and eosin (HE) staining and alcian blue-periodic acid schiff staining (AB-PAS)
Fresh samples were collected and fixed in 4% formalin for 48 h. Then, after dehydration and wax embedding, samples were sliced into 5-μm sections. Sections were dehydrated using graded ethanol and vitrified with dimethylbenzene. Hematoxylin and eosin staining was performed to observe histological changes in the submandibular gland after ligation. The AB-PAS was performed to detect salivary gland function following the standard procedure (G1285, Solarbio, Beijing, China). We captured five different views in each section from different animals. The acinar cells and AB-PAS-positive areas were calculated using Image J Pro Plus (National Institutes of Health, Bethesda, USA) based on morphology.
(caption on next page) Y. Yang et al.

Masson staining
The sections were dehydrated with gradient alcohol, as mentioned earlier. To check the degree of fibrosis, we used a Masson's trichrome staining kit (G1340, Solarbio). The nucleus was stained with hematoxylin, and muscle fiber was dyed with Masson red acid remix. Then, 2% glacial acetic acid, 1% phosphomolybdic acid aqueous solution, and aniline blue were used to stain collagenous fibers.

Terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL)
Paraffinized samples were sliced into 5-μm sections. TUNEL was performed using a One-Step TUNEL Apoptosis Assay Kit (C1090, Beyotime, Shanghai, China). Following the TUNEL reaction, the slides were treated with DAPI (F6057, Sigma, St. Loius, MO, USA) according to the manufacturer's instructions. Images were captured using a confocal microscope. The number of TUNEL-positive cells was calculated in six random fields per slide using ImageJ software.

RNA sequencing
We randomly selected transcriptomes from normal submandibular gland samples (NT group, n = 3) and 1-week ligated samples (1w group, n = 3) for RNA sequencing. A comparison was performed to identify genes that were differentially regulated in the NT group and 1w group. Preparation of transcriptome libraries and sequencing were performed by OE Biotech Co., Ltd. (Shanghai, China). A P-value <0.05 and fold change >1.5 or <0.75 was set as the threshold for significantly differential expression. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed respectively using R based on hypergeometric distribution.

Statistical analysis
Results are presented as mean ± standard error of the mean (SEM). All graphs were generated using GraphPad Prism 9 software (GraphPad Software). One-way analysis of variance (ANOVA) was used to compare the differences among multiple groups, and the unpaired Student's t-test was used to compare two groups. Bonferroni's multiple comparisons test was used for post-hoc analysis of the ANOVA results. P-values <0.05 were considered statistically significant.

Salivary gland function could partially recover at 1-week post-ligation
After duct ligation, the submandibular gland presented with characteristic injury. Because of the unilateral submandibular gland ligation, the weight of the ligated rats gradually increased. The salivary flow rate decreased significantly and did not recover. The mass of the submandibular gland increased significantly on day 1 and atrophied significantly on weeks 1, 2, 3, and 4 ( Supplementary Fig. 1).  The salivary gland acinar structures were obviously damaged at day 1, partially improved at week 1, and gradually worsened at weeks 2, 3, and 4. Interestingly, on day 1, the percentage of acinar cells was significantly reduced but partially recovered at week 1 (Fig. 1A,  E). In addition, micrographs in a lower magnification were supplied to show the microscopic anatomy of submandibular gland ( Supplementary Fig. 2). AB-PAS staining verified this phenomenon; samples had the smallest areas of blue staining on day 1 (Fig. 1B,  F). IF for AQP5 was significantly lower on day 1 after ligation compared to the normal group. Compared with day 1, AQP5 expression increased at week 1 and then gradually decreased over the following weeks (Fig. 1C, G). Apoptotic levels were determined using TUNEL. The number of apoptotic cells in the submandibular gland at week 4 after ligation significantly increased (Fig. 1D, H).

Fibrosis in the submandibular gland was associated with the TGF-β pathway
The submandibular glands showed varying degrees of fibrosis after duct ligation. Masson staining showed that the collagen fibers (blue area) gradually increased and reached nearly 30% of the total area at week 4 ( Fig. 2A and B). The microvasculature was checked using CD31 detection, which was present at a low level in the normal and ligation groups at day 1 and week 1. CD31 expression increased from week 2 after ligation and was maintained at a high level for the following weeks ( Fig. 2C and D). The results of TGF-β IHC were similar to those of CD31, which indicated that fibrosis started at least 1 week after ligation (Fig. 2E and F).

Oxidative metabolism was inhibited and inflammation was enhanced 1-week post-ligation
In the present study, we observed pathological changes in the rat submandibular gland at different timepoints after ligation. Based on our previous results, we found that there was some recovery from gland injury after 1 week. Detecting gene expression at this timepoint could help to capture the mechanism of slight recovery. Thus, to explore the possible mechanisms that lead to pathological changes in the submandibular gland after duct ligation, 1w and NT submandibular gland samples were selected for RNA sequencing and further verification. Samples from these two groups were separated using clustering analysis (Fig. 3A). The PC values are provided in Supplementary Table 1. There were 2827 upregulated DEGs and 2074 downregulated DEGs between the 1w and NT groups. The different genes were screened out with a P value < 0.05 ( Fig. 3B and C). There were 40 DEGs (including klk1c10, LOC100911689, LOC103690048, klk1c12, LOC108348116, and LOC108348007) that were only expressed in the NT group, and 84 DEGs (including Ms4a4c, A2m, Ndnf, Sirpb2l1, and Lilrc2) were only expressed in the 1w group. The top 30 upregulated and downregulated genes in the 1w group compared with the NT group are listed in Tables 1 and 2, respectively.
The downregulated genes included those encoding major salivary families of specific secretory proteins, like Spt1 and Csap1. The expression of tissue-specific genes, such as kallikrein-related genes (Klk1c2, Klk1c8, Klk1b3, Klk1c9, Klk1, Klk1c6, Klk15, LOC687880, Klk1c3) decreased in the 1w group (Table 1). These genes mainly participate in processes such as protein K69-linked methylation, acetyl-CoA biosynthesis, the proton-transporting ATP synthase complex, the mitochondrial isocitrate dehydrogenase complex (NAD+), 2,4-dienoyl-CoA reductase (NADPH) activity, NADH dehydrogenase activity, and NAD binding (Fig. 3D). The upregulated genes included Card14, Sfrp4, Lox, Col3a1, and Adam8 (Table 2). These genes participate in processes such as collagen receptor activity, the Fc-epsilon receptor I complex, and positive regulation of the Toll-like receptor 7 signaling pathway (Fig. 3E). KEGG analysis showed that salivary secretion function was downregulated compared with that in normal samples (Fig. 4A). Inflammation-associated indexes, such as leukocyte transendothelial migration, cytokine-cytokine receptor interaction pathway, and NF-κB, Th1, Th2, and Th17 cell differentiation, were upregulated (Fig. 4B). There were some overlapping genes in these pathways, such as Cxcl12, Nfkb1, and Vcam1 (See Supplementary Table 2). The NF-κB pathway and inflammation play a vital role in the pathological process of the ligated submandibular gland The inflammatory pathway was examined using samples from the normal group and 1w samples of the submandibular gland after ligation. The levels of IL-2, IL-6, TNF-α, and p-p65 were significantly increased in 1w samples compared with the normal gland using IHC, which confirmed our gene sequencing results (Fig. 5A-H). Members of the NF-κB pathway, such as p-IκKα, IκKα, p-IκBα, IκBα, p-p65, and p65, were significantly upregulated in 1w samples, indicating that inflammatory factors are important in the pathological process of the submandibular gland after duct ligation ( Fig. 5I and J).

Discussion
Ligation of the submandibular gland in rats results in a complete acute injury with rapid pathological changes. Clinical patients mostly experience long-term chronic stimulation, eventually leading to gland atrophy and loss of function. Ligation of the submandibular gland cannot completely simulate clinical pathological changes. However, pathological changes in the submandibular gland after ligation are characteristic. Eventually, it leads to atrophy and hypofunction of the submandibular gland. The pathological changes have certain similarities. In the present study, we explored pathological changes in the rat submandibular gland after duct ligation on day 1 and weeks 1, 2, 3, and 4. In addition, the specific genes between the normal and 1-week samples were screened using RNA sequencing. Our results showed a trend of pathological changes and the vital role of the NF-κB pathway.
The number of acinar cells and secretory function of the submandibular gland detected by AQP5 significantly decreased on day 1 after duct ligation. Although the submandibular gland partially recovered from the injury at week 1, it gradually became damaged over time. These results show that the salivary glands were injured immediately and severely within a short time after ligation. However, based on self-repairing capacity, the injury was slightly alleviated after the acute phase [13]. During the chronic phase, the damage to the submandibular gland progressively worsened. These pathological changes indicate that acute injury to the submandibular gland is rapid and serious.
Fibrosis of the salivary gland after ligation is associated with the TGF-β pathway, which drives fibrosis in most forms of chronic kidney disease. Inhibition of the TGF-β pathway can alleviate renal fibrosis by limiting the proliferation of fibroblast-type cells [22]. In addition, the TGF-β/SMAD pathway plays a vital role in the pathogenesis of hepatic fibrosis, and targeting TGF-β/SMAD signaling is a potential treatment for hepatic fibrosis [23]. Fibrosis is a common physiological and pathological condition of the salivary gland that includes sialadenitis, radiation-induced salivary gland dysfunction, and Sjögren's syndrome [24][25][26]. Targeting TGF-β pathways for the treatment of salivary gland fibrosis has been studied; however, satisfactory clinical results have not yet been achieved. TGF-β and CD31 expression in the rat submandibular gland after duct ligation, which induce fibrosis, gradually increased. Inhibition of TGF-β signaling is difficult because of its pleiotropic expression and broad distribution in the salivary glands [27]. Thus, it is important to study the fibrotic mechanism of the salivary glands after duct ligation. Our results confirm the vital role of the TGF-β pathway in fibrotic changes in the rat submandibular gland.
RNA sequencing results showed that inflammatory factors were significantly upregulated in salivary glands 1 week after duct ligation. Inflammation is an important trigger for tissue damage and fibrosis. After salivary duct ligation, different cell types in the immune system are activated, which recruit multiple inflammatory factors [28]. Our sequencing results showed that Th1, Th2, and Th17 cell differentiation and the NF-κB pathway were upregulated in the week-1 samples after duct ligation. Based on our analysis, ligation may reduce oxidative metabolic capacity, activate the NF-κB pathway, and induce inflammation, leading to salivary gland injury. Significantly increased expression of the inflammatory signaling pathway was the major cause of hyposalivation. A previous study showed that although inflammatory reactions in the ligated gland are eliminated by dexamethasone, the injured salivary function did not improved [29]. This suggests that inflammation is not only the cause of injury but may also be a trigger point for repair. This can be further verified by single-cell sequencing or in large animals. Further submandibular gland ligation-deligation models are also required to analyze genetic changes. Changes in injury and regeneration were analyzed in combination. Thus, further studies are needed to elucidate the mechanism underlying the induction of inflammation in ligated salivary glands.

Conclusions
The clinical changes induced by salivary gland disease are different from those in the present experimental duct ligation model, to some extent. Our results detailed pathological changes in the submandibular gland after ligation. Fibrotic processes in the submandibular gland are associated with the TGF-β pathway. The NF-κB pathway and inflammation play vital roles in the pathological processes of the ligated submandibular gland. Moreover, these data could preliminarily provide a pathological basis that explains the changes in the rat submandibular gland after ligation.

Author contribution statement
Yang Yang: Conceived and designed the experiments, Performed the experiments, Analyzed and interpreted the data, Wrote the paper.Yang Zi: Performed the experiments, Analyzed and interpreted the data, Wrote the paper.Du Conglin: Performed the experiments, Analyzed and interpreted the data.Zhang Chunmei: Contributed reagents, materials, analysis tools or data, Wrote the paper. Hu Liang: Conceived and designed the experiments, Analyzed and interpreted the data, Wrote the paper.Wang Songlin: Conceived and designed the experiments, Contributed reagents, materials, analysis tools or data, Wrote the paper.