Molecular Vision 2021; 27:679-690
<http://www.molvis.org/molvis/v27/679>
Received 17 June 2021 | Accepted 05 December 2021 | Published 07 December 2021
Molecular Vision 2021; 27:679-690
<http://www.molvis.org/molvis/v27/679>
Received 17 June 2021 |
Accepted 05 December 2021 |
Published 07 December 2021
Shaohua Hu,1 Guohu Di,1,3 Xin Cao,1 Yaning Liu,1 Yihui Wang,1 Hui Zhao,4 Dianqiang Wang,2 Peng Chen1,3
The first two authors contributed equally to this article.
1Department of Human Anatomy, Histology and Embryology, School of Basic Medicine, Qingdao University, Qingdao, Shandong Province, China; 2Qingdao Aier Eye Hospital, Qingdao, China; 3Institute of Stem Cell Regeneration Medicine, School of Basic Medicine, Qingdao University, Qingdao, China; 4The 971 Hospital of the Chinese People’s Liberation Army Navy, Qingdao, Shandong Province, China
Correspondence to: Peng Chen, Department of Human Anatomy, Histology and Embryology, School of Basic Medicine, Qingdao University, 308 Ningxia Road, Qingdao 266071, China; Phone: (532) 8378-0061; FAX: (532) 83780081; email: chenpeng599205@126.com.
Purpose: AQP5−/− mice spontaneously exhibit dry eye symptoms. The purpose of this study was to assess the endoplasmic reticulum (ER) stress-mediated inflammation generated by a deficiency of aquaporin 5 (AQP5) in the lacrimal gland.
Methods: Hematoxylin and eosin (H&E) staining, Oil Red O staining, and transmission electron microscopy (TEM) analysis were performed to identify structural changes in lacrimal gland epithelial cells because of AQP5 deficiency. Corneal epithelial defects were assessed with sodium fluorescein staining. The expression profiles of mRNA and proteins were determined by quantitative real-time reverse transcription PCR (qRT-PCR) and western blot. Mice in the quercetin group were injected intraperitoneally with 40 mg/kg of quercetin, and the control group was injected with an equal volume of dimethyl sulfoxide (DMSO) for 4 weeks.
Results: Aqueous tear secretion fell at about 50% in 1- and 6-month-old AQP5−/− mice compared with that of AQP5+/+ mice. TEM showed that the ER structure was damaged. ER stress was significantly increased in the lacrimal gland of AQP5−/− mice. Lipid droplets accumulated in the matrix and acinar cells, and changes occurred in the lipid metabolism and gene expression levels for PPARα, CPT1α, and CPT2 in the AQP5−/− mice. Immune cell infiltration and increases in the gene expression levels of the chemokines CXCL1, CXCL2, and CCL5 were found in the lacrimal gland of AQP5−/− mice. Quercetin partially reversed ER stress levels, inflammation, and lipid accumulation, and it inhibited tear secretion.
Conclusions: The study data indicated that a deficiency of AQP5 induced pathophysiological changes and functional decompensation of the lacrimal gland. Quercetin may improve the inflammation in the lacrimal glands of AQP5−/− mice by regulating the ER stress levels.
The lacrimal gland is the organ that secretes tears, which are essential for the lubrication of the eye. Disruption of lacrimal fluid composition, production, or release results in dry eye, causing discomfort and damage to the ocular surface [1]. The lacrimal gland consists of the acini, ducts, nerves, myoepithelial cells (MECs), and plasma cells [2,3]. Acinar cells compose about 80% of the glands and secrete electrolytes, water, and proteins to form primary fluid. As the primary fluid moves along the duct system, the duct cells modify the primary fluid by secreting or absorbing electrolytes. The final lacrimal gland fluid is then secreted onto the surface of the eye. Lacrimal gland secretion is primarily under neural control, which is achieved through a neural reflex arc [4]. A decrease or lack of lacrimal gland secretion is the leading cause of aqueous tear–deficient dry eye syndrome [5]. Many causes, including autoimmune diseases (e.g., Sjögren syndrome [SS]) [6], allogeneic hematopoietic cell transplantation [7], ligation of the lacrimal gland duct [8], and aging [9], can induce lacrimal gland damage. These injuries can result in sufficient lacrimal gland malfunction to cause dry eye disease.
The endoplasmic reticulum (ER) is a dynamic cell lumen. Its main functions are secretion, synthesis, and folding of transmembrane proteins and regulation of calcium balance [10]. ER stress has been described in many diseases, including rheumatoid arthritis, neurodegenerative diseases, cardiovascular disorders, and inflammatory bowel disease [11,12]. Recently, it has been found that ER stress may be a pathophysiological change in the conjunctival pathogenesis of dry eye syndrome [13].
At the same time, inflammatory infiltration and fibrosis of lacrimal gland tissue could decrease tear secretion and eventually lead to dry eye [14]. Lipid metabolism involves the synthesis and degradation of lipids in cells [15,16]. The ER is the major organelle involved in lipid metabolism because it contains many relevant enzymes for this process. ER stress is a potential mechanistic link between excess nutrients and lipid accumulation, which is a crucial event in the development of minor labial salivary gland tissue [17,18]. It has also been shown that lymphocytic infiltration destruction of the acini increases with age, and abnormal lipid metabolism has been found in the lacrimal gland of a sleep deprivation–induced mouse dry eye model [19].
Thirteen known aquaporins (AQPs) are widely distributed in various organs and play different roles in the body. AQP5 has been detected in the salivary glands, lacrimal glands, cornea, and lungs [20]. In the salivary glands, AQP5 is localized at the apical membrane of serous-type acinar cells [21]. AQP5 knockout studies in mice confirm the critical role of AQP5 in water secretion in salivary gland acinar cells [22,23]. It has been proposed that AQP5 abnormalities may occur in SS and that aquaporin gene delivery to the lacrimal and salivary glands could increase fluid secretion.
Quercetin is a known phytochemical with antioxidant and anti-inflammatory properties [24,25]. It has been reported to directly inhibit the reactive oxygen species (ROS)-promoted activation of nuclear factor-κB (NF-κB) signaling [26,27]. Moreover, topical application of quercetin seems to suppress inflammation of the ocular surface and lacrimal gland [28]. The anti-inflammatory effect of quercetin has been reported in several studies, and dry eye can be effectively controlled in mice after alleviating lacrimal gland inflammation [29]. Found in natural botanical plants, quercetin is under basic and early stage clinical research for a variety of disease conditions [30]. It can protect RAW264.7 macrophages from glucosamine-induced lipid accumulation through the suppression of ER stress, as evidenced by reduced C/EBP homologous protein (CHOP) and activating transcription factor-6 (ATF6) levels [31]. We analyzed the anti-inflammatory effects and mechanisms of quercetin on ER stress-mediated inflammation raised by a deficiency of AQP5 in the lacrimal gland. Our results showed that in the lacrimal glands of AQP5−/− mice, ER stress and its induced inflammation were present in lacrimal epithelial cells, as was abnormal lipid metabolism. Quercetin partially reversed the inhibited ER stress levels, lipid accumulation, inflammation, and tear secretion.
Using clustered regularly interspaced short palindromic repeat - associated Cas9 (CRISPR/Cas9) technology, AQP5−/− mice were produced by the high-flux electric transfer of fertilized eggs from Cyagen Biosciences Inc. (Guangzhou, China) [32]. Age-matched AQP5+/+ and AQP5−/− mice were used for our study. Six-month-old AQP5−/− mice were divided into quercetin and control groups (n = 7 per group). Quercetin (purity ≥98%, Sigma, St. Louis, MO) was dissolved in dimethyl sulfoxide (DMSO; 100%; Solarbio, Beijing, China) at 100 mg/ml. Mice were injected intraperitoneally with quercetin in DMSO in the quercetin group (40 mg/kg, 7 μl per mouse) or DMSO alone in the control group (100% DMSO, 7 μl per mouse) once per day for 4 weeks.
Transmission electron microscopy (TEM) was conducted as described previously [33]. In brief, the lacrimal glands were cut into 1 mm3 blocks and fixed in prechilled 2.5% glutaraldehyde. After fixing and dehydration, the sections were double-stained with uranium and lead and observed under a transmission electron microscope (HITACHI, Tokyo, Japan).
Tears were measured by inserting cotton phenol red threads (Zone-Quick, AYUMI Pharmaceutical, Tokyo, Japan) into the lateral canthus of the eye for 20 s in unanesthetized animals, as described previously [33]. A 1% fluorescein solution (Sigma) was applied to the cornea for 2 min. The eyes were rinsed with PBS and examined using a cobalt blue light from a slit-lamp scope (66 Vision Tech. Co., Ltd., Suzhou, China).
Mice were sacrificed by cervical dislocation, and the lacrimal glands were fixed in 10% buffered formalin and embedded in paraffin. Paraffin sections (4 μm) were stained with hematoxylin and eosin (H&E) and observed under a light microscope (CKX53, Olympus, Tokyo, Japan). Lacrimal gland cryosections (6 μm) were fixed in 4% paraformaldehyde and then immunostained with anti-AQP5 antibody (1:100, Abcam, Cambridge, UK) and anti-α-smooth muscle actin (anti-α-SMA; 1:100, Abcam). The samples were then incubated with secondary antibodies (1:200, Thermo Fisher Scientific, Rockford, IL) and 4’6-diamidino-2-phenylindole (DAPI; 1:500, Beyotime Biotechnology, Shanghai, China). Sections were sliced and immunostained with F4/80 (1:100, Affymetrix, Inc., San Diego, CA), Ly6g (1:100, Biolegend, CA), and CD4+ (1:100, Affymetrix, Inc.). A laser confocal microscope (Olympus) was used to view the images. The immunofluorescence results were analyzed using ImageJ (National Institutes of Health, Bethesda, MA) software.
All 6-mm-thick frozen sections were covered with 4% paraformaldehyde (Solarbio) for 25 min and rinsed twice with PBS for 5 min each time. The sections were stained with filterable Oil Red O (storage liquid:distilled water = 3:2, Cyagen Biosciences Inc.) for 30 min. They were then rinsed three times with PBS for 5 min each time. Following this, the sections were stained with Mayer’s hematoxylin (Sigma) for 1 min and washed with water for 3 min. Glycerol (Servicebio, Wuhan, China) was used for the gelatin sealing piece. Six different visual fields were selected from each group, and the positive density of the images was quantitatively analyzed using ImageJ.
The GeneJET RNA Purification Kit (Thermo Fisher Scientific, Ottawa, Canada) was used for the isolation of total RNA. The specific primers for detection of CXCL1, CXCL2, CCL5, CPT1α, CPT2, PPARα, MCP-1, IL-1β, IL-6, TNF-α, and GAPDH were designed and synthesized by Invitrogen (Shanghai, China). The cDNAs of each sample were synthesized using a PrimeScript RT reagent kit (TaKaRa, Tokyo, Japan). Quantitative PCR was conducted using SYBR Green (Vazyme, Nanjing, China) and a Bio-Rad CFX 96 Machine (Bio-Rad, Hercules, CA). The expression levels of each mRNA were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression levels. The relative expression level was calculated using the 2-ΔΔCT method. The primers are listed in Table 1.
Lacrimal gland tissue was extracted with Radioimmunoprecipitation Assay (RIPA) Lysis Buffer (Beyotime) mixed with phenylmethylsulfonyl fluoride (PMSF; Solarbio). Protein extracts were resolved via 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE; EPizyme, Shanghai, China) and then electrotransferred to a polyvinylidene difluoride membrane. After blocking, the membranes were probed with primary antibodies to AQP5 (1:2000, Abcam), α-SMA (1:2000, Abcam), GAPDH (1:3000, Kangchen, Shanghai, China), CHOP (1:500, Affinity Biosciences, Wuhan, China), Bax (1:1000, Affinity Biosciences), Bcl-2 (1:2000, Affinity Biosciences), Caspase12 (1:1000, Affinity Biosciences), and GRP78 (1:1000, Affinity Biosciences). Horseradish peroxidase–conjugated secondary antibodies were used, and the bands were developed using a Western Bright Electrochemiluminescence (ECL) substrate (Applygen, Beijing, China). GAPDH was used for normalization of expression [34].
All data are presented as the mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was applied to evaluate the significance between different groups, and p<0.05 was considered statistically significant. The statistical analysis was performed with Prism software version 8.0 (GraphPad Software, La Jolla, CA).
Immunofluorescence staining showed that AQP5 was normally expressed in the apical membrane of acinar cells in the lacrimal glands of AQP5+/+ mice (Figure 1A). Immunofluorescence staining and western blot demonstrated that no expression of AQP5 was detected in the lacrimal glands of AQP5−/− mice (Figure 1A,B). We found that there was no significant difference in body weight between the two groups (1 month old, p = 0.5806; 6 months old, p = 0.2528; n = 10 per group; Figure 1C). The weight of the lacrimal gland was greater in the 1- and 6-month-old AQP5−/− mice than it was in the age-matched AQP5+/+ mice (Figure 1D,E). The tear volume of AQP5−/− mice was significantly lower than that of AQP5+/+ mice (Figure 1F). Spontaneous punctate corneal epithelial defects in AQP5−/− mice were observed under a slit lamp (Figure 1G).
TEM images confirmed that the number and structure of organelles in the lacrimal glands of AQP5+/+ mice were normal. Rough ER was dilated and mostly short, and the amount of rough ER was lower in the lacrimal glands of AQP5−/− mice (Figure 2A). To explore the underlying mechanisms of AQP5 deficiency affecting structural ER changes, we investigated whether AQP5 regulated ER stress in the lacrimal gland. The ER stress markers GRP78 and transcription factor CHOP were elevated in the lacrimal glands, suggesting that excessive ER stress was induced by AQP5 deficiency. We also detected ER stress-mediated apoptosis in AQP5−/− mice. The results showed that cleaved Caspase12, a caspase that activates ER stress-mediated apoptosis, was heightened in AQP5−/− mice (Figure 2B,C). We also found that the expression of the apoptosis regulator Bax was increased in AQP5−/− mice (Figure 2B,C). In contrast, the expression of Bcl-2, a cellular protein that inhibits apoptosis, decreased significantly in AQP5−/− mice (Figure 2B,C).
Using Oil Red O staining, an accumulation of neutral triglycerides in the acinar of AQP5−/− mice was observed (Figure 3A). The intensity increased from 1,545.55 ± 523.97 (1 month) and 5,769.42 ± 1,140.71 (6 months) in AQP5+/+ mice to 4,365.87 ± 1,383.92 (1 month) and 8,532.98 ± 1,590.97 (6 months) in AQP5−/− mice (n = 6 per group; Figure 3A,B). Real-time PCR was performed with lipid metabolism genes PPAR, CPT1α, and CPT2. It was found that PPARα and CPT1α were downregulated in the lacrimal glands of 1- and 6-month-old AQP5−/− mice compared with age-matched AQP5+/+ mice (n = 3 per group; Figure 3C,D). CPT2 was downregulated in the lacrimal glands of 1-month-old AQP5−/− old mice (n = 3 per group; Figure 3E).
Inflammatory cells infiltrated around the blood vessels in the lacrimal gland of 6-month-old AQP5−/− mice (Figure 4Ai). Immunofluorescence staining identified time-dependent increases in CD4+ T lymphocyte density around the acini in lacrimal glands of 1- and 6-month-old AQP5−/− mice compared with the age-matched AQP5+/+ mice (Figure 4Aii,Bi). Neutrophil infiltration identified by Ly6g staining increased in the lacrimal gland of 6-month-old AQP5−/− mice (Figure 4Aiii,Bii). F4/80 staining increased the mononuclear macrophage density in the lacrimal gland in 1- and 6-month-old AQP5−/− mice (Figure 4Aiv,Biii). Expression of proinflammatory factors, such as IL-6, IL-1β, MCP-1, and TNF-α, significantly increased in the lacrimal glands of 1- and 6-month-old AQP5−/− mice compared with the age-matched AQP5+/+ mice (Figure 4C). In parallel, chemokine CXCL1, CXCL2, and CCL5 expression levels significantly increased in 1- and 6-month-old AQP5−/− mice (Figure 4C).
α-SMA is a MEC marker [35]. In this study, α-SMA expression showed an obvious decline in the lacrimal glands of AQP5−/− mice (Figure 5A). According to the analysis of α-SMA staining, the intensity decreased from 10.93 ± 0.76 (1 month) and 10.99 ± 2.52 (6 months) of AQP5+/+ mice to 4.63 ± 0.46 (1 month) and 5.29 ± 1.25 (6 months) of AQP5−/− mice (n = 6 per group; Figure 5B). Downregulation of α-SMA was further confirmed by western blot in AQP5−/− mice (p<0.05; n = 3 per group; Figure 5C,D).
TEM images confirmed that the rough ER was abundant and orderly, with many ribosomes attached to its surface in the lacrimal glands of the quercetin group. There was a lot of ER retained, and more floccules were found in the reticulum pool in the lacrimal gland of the DMSO group (Figure 6A). More importantly, the intraperitoneal injection of quercetin significantly attenuated the expression of GRP78, CHOP, Caspase12, and Bax (p<0.005, n = 3 per group; Figure 6B,C). In contrast, the expression of Bcl-2 increased significantly, reaching 1 ± 0.001 for the DMSO group and 1.21 ± 0.076 for the quercetin group (Figure 6B,C). Lipid accumulation slightly decreased, with Oil Red O staining density observed at 14,118.8 ± 1,307.11 for the DMSO group and 3,484.38 ± 198.71 for the quercetin group (Figure 6D,E). Real-time PCR confirmed that the mRNA levels of PPARα, CPT1α (p<0.05, n = 3 per group), and CPT2 (p<0.005, n = 3 per group) were downregulated in the lacrimal glands of the quercetin group compared with the DMSO group (Figure 6F).
Compared with the DMSO group, the quercetin group showed less inflammatory infiltration in the lacrimal epithelial cells (Figure 7Ai,B). Instead, numbers of inflammatory cells, such as neutrophils, macrophages, and CD4+ T cells, fell significantly with quercetin treatment (Figure 7Aii-iv,B). Immunofluorescence staining and western blot analysis showed that quercetin had no significant effect on the expression levels of α-SMA (Figure 7C,D). Proinflammatory factors IL-6, IL-1β, MCP-1, and TNF-α were downregulated; CXCL2, a chemokine, was also significantly downregulated in the lacrimal glands of the quercetin group (n = 3 per group; Figure 7E). The tear volume increased from 6.625 ± 2.501 in the DMSO group to 13.938 ± 3.746 in the quercetin group (p<0.05, n = 6 per group; Figure 7F). Under a slit lamp, it was found that quercetin partially alleviated punctate epithelial defects in the corneas of AQP5−/− mice (Figure 7G).
The tear film’s aqueous layer is mainly secreted by the lacrimal gland. Many factors related to tears and the ocular surface can cause dry eye, usually accompanied by increased osmolarity of the tear film and ocular surface inflammation. Dry eyes can cause eye discomfort, visual impairment, tear film instability, and potential damage to the ocular surface [36]. It is known that the popularization of air conditioning, frequent usage of visual display terminals, and increase in wearers of soft contact lenses may all be associated with a rise in the incidence of dry eye [37].
Abnormal AQP5 trafficking in SS may contribute to decreased lacrimation and dry eye in patients [38]. A missense mutation in AQP5 (p.G308A) probably resulted in abnormal membrane insertion or ineffective trafficking. Rats with this mutation show extremely low water secretion from their salivary glands [39]. In our previous study, we found that AQP5−/− mice naturally develop dry eye symptoms. However, the mechanism by which the loss of AQP5 protein induces dry eye in mice remains unclear. To explore the exact mechanism of the decrease in tear secretion caused by AQP5 deficiency, we analyzed its ultrastructure. Rough ER in the lacrimal gland cells of AQP5−/− mice was slightly dilated and mostly short, and ribosomes were attached to it without shedding.
Prolonged or severe ER stress affects the function and structure of proteins and lipids [40,41]. ER stress–induced inflammation in the lacrimal glands plays an important role in the pathogenesis of dry eye syndrome or SS [42]. Thus, ER stress can potentially serve as a target for treating dry eye. Our results indicated that the lacrimal glands of AQP5−/− mice undergo more severe ER stresses than normal lacrimal glands do. AQP5 deficiency increased the levels of CHOP, GRP78, and Bax while decreasing Bcl-2 expression. CHOP is an ER stress–specific nuclear transcription factor. GRP78, a calcium ion binding molecule located in the ER, shows marker upregulation in ER chaperones in response to ER stress [43]. In contrast, Bcl-2 is a cellular protein that inhibits apoptosis. ER stress induces Caspase12-mediated apoptotic cell death [44]. Our western blot results showed a high expression of cleaved Caspase12 in AQP5−/− mice. However, low expression of the cleaved Caspase12 protein was observed in the normal lacrimal glands.
In dry eye, ER stress induces abnormal lipid metabolism and various inflammatory and apoptotic pathways, leading to lacrimal gland cell death [45]. We found that lipid droplets accumulated along with changes of PPARα,CPT1α, and CPT2 in AQP5−/− mice. We also found that ER stress activated inflammatory signaling pathways based on increases in inflammatory cell infiltration. Expression levels of proinflammatory factors, such as IL-6, IL-1β, MCP-1, and TNF-α significantly increased in the lacrimal glands of 1- and 6-month-old AQP5−/− mice compared with the age-matched AQP5+/+ mice. It is presumed that ER stress induces lipid deposition in acinar cells and the surrounding microenvironment. ER stress also activated inflammation, which induced acinar cell apoptosis, as reported previously [45].
Decreased aqueous tear secretion could develop as a consequence of several pathological changes in AQP5−/− mice. First, lipid deposition induced by ER stress compromises cell function [46,47]. Second, infiltration of inflammatory cells disrupts acinar cells [48,49]. Third, the inflammatory cytokines secreted by infiltrated inflammatory cells directly reduce aqueous secretion of acinar cells [50]. Finally, MECs exert pressure on the acinus and promote tear secretion [51,52]. Our results indicate that the MEC structure became disordered and disrupted in AQP5−/− mice. These pathological changes are likely to account for how AQP5 deficiency reduces aqueous tear secretion.
Quercetin is a natural product with antioxidative properties [53,54]. It has been shown to inhibit mitochondrial dysfunction and inflammation in osteoarthritis animal models [55,56]. Moreover, it has been found that 0.5% quercetin eye drops aid in increasing tear volumes. Quercetin treatment can also increase the density of goblet cells [28]. Our results indicated that quercetin alleviated ER stress by attenuating the expression of GRP78, CHOP, Caspase12, and Bax. Our findings revealed that quercetin alleviated lipid accumulation in AQP5−/− mice via downregulated expression of CPT1α and CPT2. We also found that quercetin reduced inflammatory cell infiltration and proinflammatory factor expression.
In summary, AQP5 deficiency induced severe ER stress and inhibited fatty acid clearance in the lacrimal gland. Losses in structural integrity and declines in fluid secretion are consistent with proinflammatory cytokine upregulation and inflammatory cell infiltration. Quercetin indeed reduced inflammation and lipid deposition, while it partially ameliorated declines in lacrimal gland tear secretion. These results suggest that quercetin may provide therapeutic benefits in dry eye induced by AQP5 deficiency.
This work was supported by the National Natural Science Foundation of China (No. 81970782, No. 81600721), Shandong Provincial Natural Science Foundation (No. ZR2018MH016), Qingdao Postdoctoral Application Research Project (No. 40518060071), China Postdoctoral Science Foundation (No. 2017M612211).