Mice Lacking Inositol 1,4,5-Trisphosphate Receptors Exhibit Dry Eye

Tear secretion is important as it supplies water to the ocular surface and keeps eyes moist. Both the parasympathetic and sympathetic pathways contribute to tear secretion. Although intracellular Ca2+ elevation in the acinar cells of lacrimal glands is a crucial event for tear secretion in both the pathways, the Ca2+ channel, which is responsible for the Ca2+ elevation in the sympathetic pathway, has not been sufficiently analyzed. In this study, we examined tear secretion in mice lacking the inositol 1,4,5-trisphosphate receptor (IP3R) types 2 and 3 (Itpr2−/−;Itpr3−/−double-knockout mice). We found that tear secretion in both the parasympathetic and sympathetic pathways was abolished in Itpr2−/−;Itpr3−/− mice. Intracellular Ca2+ elevation in lacrimal acinar cells after acetylcholine and epinephrine stimulation was abolished in Itpr2−/−;Itpr3−/− mice. Consequently, Itpr2−/−;Itpr3−/− mice exhibited keratoconjunctival alteration and corneal epithelial barrier disruption. Inflammatory cell infiltration into the lacrimal glands and elevation of serum autoantibodies, a representative marker for Sjögren’s syndrome (SS) in humans, were also detected in older Itpr2−/−;Itpr3−/− mice. These results suggested that IP3Rs are essential for tear secretion in both parasympathetic and sympathetic pathways and that Itpr2−/−;Itpr3−/− mice could be a new dry eye mouse model with symptoms that mimic those of SS.


Introduction
Because tears keep the cornea and conjunctiva continuously moist, and a reduction in tear volume results in dry eyes (e.g. keratoconjunctivitis sicca), investigation of the regulatory mechanisms underlying tear secretion is crucial for understanding the pathology of ocular systems and for the development of new treatments for dry eyes.
Tear secretion from the lacrimal glands is regulated by two types of nerves: parasympathetic and sympathetic. The activation of parasympathetic and sympathetic nerves predominantly releases the neurotransmitters acetylcholine (Ach) and norepinephrine, respectively [1,2]. Upon binding to muscarinic acetylcholine receptors, Ach activates phospholipase C and produces inositol 1,4,5-trisphosphate (IP 3 ), which in turn triggers intracellular Ca 2+ release through the IP 3 receptor (IP 3 R) from the endoplasmic reticulum (ER) in lacrimal gland acinar cells [1]. Stimulation of the aand b-adrenergic receptors by norepinephrine also induces Ca 2+ release from internal stores [1,2]. However, in contrast to the established role of IP 3 Rs in the cholinergic pathway, the Ca 2+ channels that contribute to Ca 2+ elevation in the sympathetic pathway are still obscure. It was reported that the activation of a1adrenergic receptor, a predominant type of adrenergic receptor in lacrimal glands, increases intracellular Ca 2+ without IP 3 production, and cyclic ADP-ribose is thought to be involved in the Ca 2+ increase via the ryanodine receptor-another Ca 2+ channel on the ER [2][3][4][5].

Ethics Statement
All animal procedures in this study were approved by the Animal Experimental Committees at the Institutes of Physical and Chemical Research (RIKEN) -Research Center for Brain Science Institute (BSI) (Permit Number: H25-2-202). All efforts were made to minimize animal suffering. Mice [6] were housed on a 12 h light-dark cycle, with the dark cycle occurring from 8:00 P.M. to 8:00 A.M in a specific pathogen-free environment of the Laboratory Animal Facility of the RIKEN Brain Science Institute. In all experimental groups, mice were used at 6-40 weeks of age and 50% were female. Tear collection from mouse eyes was performed under anesthesia with intraperitoneal injection of ketamine and xylazine.

Measurement of Tear Secretion
The mice were anesthetized by intraperitoneal injection of 36 mg/kg ketamine (Daiichi Sankyo, Tokyo, Japan) and 16 mg/ kg xylazine (Bayer Healthcare, Leverkusen, Germany). Tear production was stimulated by intraperitoneal injection of 3 mg/ kg pilocarpine (Santen, Osaka, Japan) or 1 mg/kg epinephrine at 1 min after the anesthesia. Tears were collected for 15 min and the volume was calculated every 5 min during the 15-min duration using 0.5-mL capillary microglass tubes (Drummond, PA, USA). After the measurement, the mice were sacrificed, and the lacrimal glands were extirpated. Then, the lacrimal gland weights were measured, and the mean values were calculated to obtain the average lacrimal gland weight of the mice. The tear secretion volume was adjusted for the weight of the each lacrimal gland.

Histopathology and Electron Microscopy
For histopathology, the extracted lacrimal glands and conjunctiva were embedded in an optimal cutting temperature compound (Sakura Finetechnical, Tokyo, Japan). Frozen sections (5-mm thick) of the lacrimal glands or the conjunctiva were fixed with 10% formalin neutral buffer solution (Wako, Osaka, Japan) and stained with hematoxylin and eosin or with the periodic acid-Schiff reagent. For electron microscopic observation, a portion of the lacrimal glands was fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer overnight and was post-fixed with 1.0% osmic acid in 0.1 M cacodylate buffer. The specimens were dehydrated with ethanol and embedded in epoxy resin. The ultra-thin sections (80 nm) were double-stained with uranyl acetate and lead citrate, and were examined under a transmission electron microscope (1200 EXII; JEOL, Tokyo, Japan).

Measurement of Acinar Cell Area of the Lacrimal Glands
For quantitative analysis, hematoxylin/eosin (HE)-stained sections of the lacrimal glands from wild-type and Itpr2 2/ 2 ;Itpr3 2/2 mice were used. The lacrimal acinar cell area was measured as reported previously [10].

Measurement of Intracellular Ca 2+ Concentration in Lacrimal Gland Cell Suspensions
Following deep anesthesia by the intraperitoneal injection of 60 mg/kg nembutal (Dainippon Sumitomo Pharma, Osaka, Japan), the mice were sacrificed. Subsequently, the exorbital lacrimal glands were immediately removed, placed in cold balanced salt solution (BSS) containing 115 mM NaCl, 5.4 mM KCl, 2 mM Ca 2+ , 1 mM Mg 2+ , 20 mM Hepes, and 10 mM glucose (pH7.4), and rapidly minced under exposure to 2 mg/mL collagenase type 2 (Worthington, Malvern, PA, USA) in BSA. The material was then digested for 10 min at 37uC with 2 mg/mL of collagenase type 2 in BSS, the suspension being gently passed through a pipette several times. After the digestion, 1 mL of BSS was added to the preparation and then centrifuged at 1006g for 3 min. The pellet was rinsed in 1 mL BSS and centrifuged in order to collect the lacrimal gland cells.
The isolated lacrimal gland cell preparation was incubated in 5 mM fura-2 AM (Dojindo)/BSS for 45 min at room temperature, rinsed twice, resuspended in 500 mL of BSS, and stored at 4uC. For the two-dimensional measurement of Ca 2+ changes, a 75-mL sample of fura-2-loaded lacrimal gland cells was dispersed on a Cell-Tak (BD Biosciences, Bedford, MA, USA)-coated glass coverslip that formed the bottom of the recording chamber, mounted on the stage of an inverted fluorescein microscope (IX70, Olympus, Tokyo, Japan), and perfused with BSS at a rate of 2 mL/min at room temperature. Excitation of fura-2 was performed every 5 s by alternate illumination with 340 and 380 nm light. The resultant fluorescence (510-550 nm; F340/ F380) was imaged using an objective lens (UPlanApo 20x/340, Olympus) and a silicon-intensified target camera to obtain pseudocolored images of F340/F380, and stored in a personal computer using the ARGUS50/CA software (Hamamatsu Photonics, Shizuoka, Japan). The peak amplitude Ca 2+ responses (R, delta Fura-2 ratio 340/380) were expressed as the averaged amplitude from 0-50 sec was equal to zero.

Real Time RT-PCR
Total RNA was extracted from cells in the lacrimal glands of the mice using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Complementary DNA was produced from total RNA using Superscript VILO TM Master Mix (Invitrogen). Quantitative real-time PCR was performed using the StepOne-Plus Real Time PCR system (Applied Biosystems) with Fast Advanced Master Mix (Applied Biosystems) and the predesigned primers for tumor necrosis factor alpha (TNF-a), interleukin-6 (IL-6), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) [TaqMan Gene Expression Assay (TNF-a: Mm00443258-m1, IL-6: Mm00446190-m1, and GAPDH: Mm99999915-g1)]. The mRNA levels were evaluated by the DDCT method, and normalized to GAPDH mRNA.

Enzyme-linked Immunosorbent Assay (ELISA) for Immunoglobulins and Auto-antibodies
The amounts of mouse immunoglobulins and auto-antibodies in sera from wild-type and Itpr2 2/2 ;Itpr3 2/2 mice were analyzed by ELISA. For the detection of antibodies to SS-A antigens, the mouse sera were diluted 1:100 and analyzed using mouse anti-SS-A IgG ELISA kits (Alpha Diagnostics, San Antonio, TX, USA).

Statistical Analysis
All summarized data were expressed as means 6 SEM. Statistical significance was calculated by unpaired Student's t-test or Mann-Whitney U-test. A p value less than 5% was considered statistically significant.

Results
Itpr2 2/2 ;Itpr3 2/2 Mice had Severe Defects in Tear Secretion Via Both Cholinergic and Adrenergic Receptor Pathways We have previously reported that IP 3 R2 and IP 3 R3 play crucial roles in secretions from salivary, pancreatic, and nasal glands [6,7]. However, the subtypes of IP 3 R expressed in lacrimal glands and their contribution to tear secretion remain unknown. To analyze the role of IP 3 Rs in lacrimal glands, we measured tear flow in mice deficient in IP 3 Rs (Fig. 1A). Since the body weight and lacrimal gland weight were different between wild-type and mutant mice (Figs. 1B, 1C), the tear volume was normalized against lacrimal gland weight. After the intraperitoneal administration of pilocarpine, a cholinergic receptor agonist, wild-type mice shed a large volume of tears in a time-dependent manner (Fig. 1D, E). Tear secretion in Itpr2 2/2 mice was comparable with that in wild-type mice, while Itpr3 2/2 mice shed more tears than the wild-type mice. In contrast, tear secretion was abolished in Itpr2 2/2 ;Itpr3 2/ 2 mice (Fig. 1D).
We also examined the contributions of IP 3 Rs in tear secretion via the sympathetic pathway. As shown in Fig. 1F, tear flow by intraperitoneal administration of epinephrine was clearly observed in wild-type mice, but not in Itpr2 2/2 ;Itpr3 2/2 mice. These results suggest that IP 3 R2 and IP 3 R3 are the predominant subtypes of Acetylcholine-and Epinephrine-induced Ca 2+ Signals are Abolished in Itpr2 2/2 ;Itpr3 2/2 Lacrimal Acinar Cells We next examined the expression level of each IP 3 R subtype in the lacrimal glands. We found that all three types of IP 3 Rs were expressed in mouse lacrimal glands ( Fig. 2A). No bands were detected with anti-Pan-IP 3 R antibodies in the Itpr2 2/2 ;Itpr3 2/2 lacrimal gland lysates ( Fig. 2A). In addition, IP 3 Rs were detected by anti-Pan-IP 3 R antibodies in lacrimal gland lysates from Itpr2 2/ 2 but not in Itpr3 2/2 mice (Fig. 2B), suggesting that IP 3 R3 exhibits the highest expression level among the three subtypes. Immunohistochemical studies using the anti-IP 3 R3 antibody revealed that IP 3 R3 is localized at the restricted region near the apical membranes in the acinar cells where endocrine secretion occurs (Fig. 2C). IP 3 R3 fluorescein staining was not detectable in Itpr3 2/2 mice (Fig. 2C).
Ca 2+ transients were clearly observed in response to acetylcholine (Ach) in wild-type lacrimal gland acinar cells in a dose-dependent manner (Fig. 2D). The Itpr2 2/2 and Itpr3 2/2 acinar cells showed Ca 2+ responses that were comparable to those of the wild-type cells, except that the Itpr3 2/2 cells exhibited relatively rather long-lasting Ca 2+ signals with decreased peak amplitudes, especially at 3.0 mM Ach (Figs. 2D, 2E). These long-lasting Ca 2+ signals were likely due to the nature of the residual IP 3 R2, which has the highest affinity for IP 3 among the three types of IP 3 Rs, and might explain the larger amount of tear secretion in Itpr3 2/2 mice (Fig. 1D). In contrast, Ach-induced Ca 2+ transients were diminished in the Itpr2 2/2 ;Itpr3 2/2 acinar cells (Figs. 2D, 2E).
Moreover, Itpr2 2/2 ;Itpr3 2/2 acinar cells exhibited no epinephrine-induced Ca 2+ transients (Fig. 2F, G). The diminished Ca 2+ signals in the Itpr2 2/2 ;Itpr3 2/2 acinar cells on epinephrine stimulation was not due to the depletion of Ca 2+ stores, because cyclopiazonic acid (CPA), a Ca 2+ pump inhibitor, induced a considerable Ca 2+ leak from the endoplasmic reticulum (Fig. 2F). These results suggest that IP 3 R2 and IP 3 R3 are essential for Ca 2+ signals in both the sympathetic and parasympathetic pathways.  We carefully checked the ocular surfaces of Itpr2 2/2 ;Itpr3 2/2 mice. A significant amount of debris was observed on the corneal surfaces in Itpr2 2/2 ;Itpr3 2/2 mice (Fig. 3A). Abnormalities of the conjunctival surface bound to abundant mucin complex were observed in Itpr2 2/2 ;Itpr3 2/2 mice (Fig. 3B). A reduction in the number of goblet cells, a common feature of dry eye patients, was also observed in Itpr2 2/2 ;Itpr3 2/2 mice. In addition, Itpr2 2/ 2 ;Itpr3 2/2 mice showed increased corneal fluorescein staining at 6 weeks (Figs. 3C, D), which indicates corneal epithelial barrier disruption in these mutant mice. This was not due to the abnormal development of the corneal surface, because no significant difference was observed in corneal staining between the ocular surfaces of wild-type and Itpr2 2/2 ;Itpr3 2/2 mice at 3 weeks after birth, immediately after the mice opened their eyes (data not shown). Moreover, Itpr2 2/2 ;Itpr3 2/2 mice showed increased blink rates because of insufficient tear flow on the ocular surface (Fig. 3E). Atrophy of the Lacrimal Glands in Itpr2 2/2 ;Itpr3 2/2 Mice We next performed histological analysis of the lacrimal gland tissues, and found atrophy of the lacrimal gland acinar units with marked lymphocytic infiltration in Itpr2 2/2 ;Itpr3 2/2 mice more than 10 weeks of age (Fig. 4A). Electron micrographs also demonstrated the distinct morphology of acinar cells between wild-type and Itpr2 2/2 ;Itpr3 2/2 mice. Secretory vesicles were located near the acinar lumen side and the well-developed endoplasmic reticulum (ER) structure was clearly observed in the cytoplasm near the apical side of the wild-type lacrimal acinar cells (Fig. 4B). In the Itpr2 2/2 ;Itpr3 2/2 acinar cells, however, an excessive number of secretory vesicles accumulated and distributed in the cytoplasm, making it difficult to detect the ER in the cytoplasm (Fig. 4B). We also found that the Itpr2 2/2 ;Itpr3 2/2 acinar cells seemed to be smaller than wild-type acinar cells. The lacrimal acinar cell area in Itpr2 2/2 ;Itpr3 2/2 mice was approximately 40% smaller than that in wild-type mice (Fig. 4C).
Inflammation of the Lacrimal Glands in Itpr2 2/2 ;Itpr3 2/2 Mice To further explore the infiltration state of the lacrimal glands in Itpr2 2/2 ;Itpr3 2/2 mice, we classified the inflammatory infiltrates by using several lymphocyte markers (leukocyte; CD45, macrophage; F4/80, T-cell; CD4 and CD8, B-cell; CD19). We found that CD45-positive inflammatory mononuclear cells infiltrated the lacrimal glands in Itpr2 2/2 ;Itpr3 2/2 mice at 10 weeks (Fig. 5A, left panel, white arrow heads). These CD45-positive cells were located in the interstitial space around the lacrimal gland acinar cells. Macrophages and activated T-cells were the major inflammatory cells at 10 weeks (Fig. 5A); however, the population of infiltrating cells changed thereafter, and many B cells were detected at 40 weeks (Fig. 5A, right panel, arrow). We also checked the inflammatory environment of the lacrimal glands by evaluating the levels of pro-inflammatory cytokines. We found that the expression levels of pro-inflammatory cytokines such as TNF-a and IL-6 were significantly increased in the lacrimal glands in Itpr2 2/2 ;Itpr3 2/2 mice ( Fig. 5B and C). Itpr2 2/2 ;Itpr3 2/2 Mice Present Autoantibodies against Ribonucleoprotein SSA We finally examined the concentrations of immunoglobulins and autoantibodies against ribonucleoprotein SSA, one of the most commonly detected autoantibodies in patients with SS, in the serum of Itpr2 2/2 ;Itpr3 2/2 mice. As shown in Fig. 6A, we found that the concentration of immunoglobulin was significantly higher in Itpr2 2/2 ;Itpr3 2/2 mice than in wild-type mice. Moreover, the levels of autoantibodies against SSA were significantly higher in Itpr2 2/2 ;Itpr3 2/2 mice compared to wild-type mice at 10 weeks, when the infiltrates were observed (Fig. 6B).

Discussion
In this study, we have shown that the type 2 and type 3 IP 3 Rs are predominantly expressed in lacrimal glands and that IP 3 Rs are essential for tear secretion via both the sympathetic and parasympathetic signaling pathways. We also found that Ca 2+ signals in response to epinephric as well as cholinergic receptors were diminished in Itpr2 2/2 ;Itpr3 2/2 lacrimal gland cells. The lack of tear flow resulted in increased eye blink rates, and the corneal surface and conjunctiva were severely damaged in Itpr2 2/ 2 ;Itpr3 2/2 mice. As the mutant mice aged, Itpr2 2/2 ;Itpr3 2/2 mice displayed atrophy and infiltration of lacrimal glands as well as the production of autoantibodies against SSA in the sera, which are clinical features observed in human SS [11,12]. Thus, our Itpr2 2/ 2 ;Itpr3 2/2 mice constitute a novel dry eye mouse model with an SS-like phenotype.
It is well known that norepinephrine released from sympathetic nerves predominantly activates a1-adrenergic receptors and induces Ca 2+ elevation in lacrimal acinar cells [13]. However, in contrast to the established role of IP 3 R in Ca 2+ elevation induced by parasympathetic stimuli, the Ca 2+ channels that are responsible for cytosolic Ca 2+ elevation triggered by a-adrenergic stimuli are not clearly identified in lacrimal acinar cells. Several previous studies suggested a role for ryanodine receptors in Ca 2+ elevation in lacrimal glands by norpinephrine [3]. Our study clearly demonstrated that IP 3 Rs contribute significantly to adrenergic tear secretion as well as cholinergic tear secretion in vivo. Ca 2+ transients triggered by epinephrine were diminished in Itpr2 2/ 2 ;Itpr3 2/2 lacrimal gland acinar cells. These results suggest that Ca 2+ release from IP 3 Rs is a crucial event in both cholinergic and adrenergic signal transduction in lacrimal glands, which underlies the lack of tear secretion, resulting in the abnormal ocular surface seen in Itpr2 2/2 ;Itpr3 2/2 mice.
It is an important observation that Itpr2 2/2 ;Itpr3 2/2 mice developed only corneal and conjunctival injuries at 6 weeks of age and showed lacrimal gland infiltrations only after 10 weeks of age. Thus, ocular surface disturbance seems to occur prior to lymphocyte infiltration into the lacrimal glands in Itpr2 2/ 2 ;Itpr3 2/2 mice. Together with the previous finding that the desiccating stress of the ocular surface induces lacrimal gland inflammation and infiltration [14], corneal surface and conjunctival injuries caused by long-lasting dysfunction of lacrimal acinar cells may lead to the activation of antigen-presenting cells [15] and the subsequent breakdown of self-tolerance against endogenous epitopes shared among lacrimal gland units. Further studies are necessary for a clear understanding of the mechanism of infiltration in the lacrimal glands, which might contribute to the pathogenesis of SS in humans.
In conclusion, we have demonstrated that IP 3 R2 and IP 3 R3 play a central role in tear secretion and maintenance of the lacrimal glands. Our data indicate that Ca 2+ release from IP 3 Rs in lacrimal gland acinar cells is essential for sympathetic as well as cholinergic tear secretion. Together with the defect in saliva secretion observed in our previous study [6], the diversified symptoms of Itpr2 2/2 ;Itpr3 2/2 mice including lacrimal gland inflammatory foci, ocular surface disruption, and the production of autoantibodies against SSA fulfill the criteria for a diagnosis of SS, established by the American-European Consensus Group [16]. We believe that Itpr2 2/2 ;Itpr3 2/2 mice will be a useful tool for the analysis of pathological mechanisms and for the development of new treatment strategies for SS. Figure 6. Elevation of serum immunoglobulins and autoantibodies to SS-A antigens in Itpr2 2/2 ;Itpr3 2/2 mice. (A) Serum levels of immunoglobulins. Serum samples were collected from 8-week-old wild-type and Itpr2 2/2 ;Itpr3 2/2 mice. Serum levels of IgG, IgA, IgG1, IgG2a, and IgG3 were measured by ELISA. (B) Serum levels of autoantibodies in wild-type (6 weeks, n = 10; 10-35 weeks, n = 11) and Itpr2 2/2 ;Itpr3 2/2 (6 weeks, n = 8; 10-35 weeks, n = 7) mice. Serum levels of autoantibodies to SS-A antigens. Bars show the means. All data are presented as means 6 SEM. Student's t-test, *P,0.05. All experiments were performed at least three times, and representative data are shown. doi:10.1371/journal.pone.0099205.g006