Improvement of Lacrimal Function by Topical Application of CyA in Murine Models of Sjögren’s Syndrome

Abstract

purpose. The object of this study was to evaluate improvement of lacrimal gland (LG) function after topical cyclosporin A (CyA).

methods. Topical CyA (0.01% and 0.1%) was applied to two mouse models of Sjögren’s syndrome, the NFS/sld after thymectomy and the nonobese diabetic (NOD) mouse, and the functional integrity of the lacrimal gland was evaluated by measuring basal and stimulated tear secretion and its histologic integrity by examining it for apoptosis and lymphocyte invasion.

results. After treatment with CyA at 0.1% in the NFS/sld mice, tear function increased, and there was a decrease in lymphocyte infiltration of the LG and a decrease in apoptotic figures among the acinar cells. In the NOD mice, tear function also improved, but there was no associated decrease in lymphocyte infiltration. However, the expression of Fas ligand (FasL) in NOD mice by infiltrating lymphocytes was suppressed with 0.1% CyA eye drops.

conclusions. CyA appears to improve tear secretion in mouse models of Sjögren’s syndrome by preventing lymphocyte-induced apoptosis of acinar cells. In one model this was achieved by preventing lymphocyte infiltration and in the other by reducing expression of FasL expression on infiltrating lymphocytes.

Sjögren’s syndrome (SS) is characterized by three symptoms: dry eye, dry mouth, and arthritis, as originally described by Sjögren in 1933. ¹ Lymphocytes infiltrate the lacrimal gland (LG) and salivary gland (SG) in SS, and the lymphocyte infiltration is associated with destruction and dysfunction of the glands. ^{2 3} The destruction of the glands is the primary cause of the severe dry eye and dry mouth that are patients’ major symptoms. ⁴ Because the LGs are seriously affected, patients cannot produce reflex tears, resulting in severe squamous metaplasia of the ocular surface epithelium. ^{5 6} Although the mechanism of destruction of the glands is unclear, a contribution by lymphocytes is highly suspected. We have reported increased expression of Fas, Fas ligand (FasL), granzyme A, and perforin in LGs of patients with SS. ^{7 8 9} FasL is a member of the TNF family and is expressed on activated lymphocytes. FasL induces target cells to undergo apoptosis. ¹⁰ These findings, suggest that the infiltrating lymphocytes may induce apoptosis of duct and acinar cells in SS through a Fas–FasL interaction. 

Although topical cyclosporin A (CyA) has long been discussed for the treatment of dry eye in the field of ophthalmology, and clinical application is now under serious consideration, ^{11 12 13} there is no definite evidence that CyA can improve dry eye. More specifically, tear production by patients with dry eye, as measured by the Schirmer test, has failed to show any increase in response to CyA. ^{14 15} Nevertheless, CyA has been shown to have a therapeutic effect in a dog model of dry eye, ¹¹ and we recently reported that topical CyA prevented lymphocyte infiltration in the NFS/sld SS mouse model. ¹⁶ The degree to which lymphocyte infiltration is responsible for lacrimal gland dysfunction is still unclear, because gland dysfunction has been shown in diabetes in the absence of lymphocyte infiltration. ^{17 18}  

There is still controversy about which animal model of SS is optimal in exhibiting the characteristics of human SS, such as decreased tears, lymphocyte infiltration of the LGs, and ocular surface squamous metaplasia. We have reported that topical CyA can prevent lymphocyte infiltration of the LG in NFS/sld mice after thymectomy at 3 days after birth. ¹⁶ We used this animal model because it is associated with lymphocyte infiltration of only the salivary and lacrimal glands, which causes dry eye and dry mouth. Haneji et al. ¹⁹ isolated the autoantigen from this mouse model, and the autoantibody has been detected in patients with SS as well as in animals. Because autoantibodies are thought to be important in the pathogenesis of SS, ²⁰ this animal model is considered a good model of SS. Thus, in our previous study we showed that CyA can prevent lymphocyte infiltration in one mouse model of SS. However, the question remains whether CyA can effectively improve and lead to functional improvement of the glands after massive lymphocyte infiltration. Furthermore, it remains unknown whether this effect of CyA is specific to NFS/sld mice alone. We therefore decided to use both the nonobese diabetic (NOD) mouse and the NFS/sld mouse in this experiment. The NOD mouse was originally considered a model of nonobese diabetes and has recently been extensively used to study pancreatic islet B-cell destruction and of exocrine autoimmune disorders. ^{21 22 23 24} Although diabetes developed only in the females, lymphocyte infiltration of the LG developed in the male mice. ^{25 26 27} We have confirmed that the NOD mouse also exhibits decreased tear production. ²⁸  

The present study showed that an increase in tear production (functional improvement) was obtained by application of 0.1% and 0.01% CyA eyedrops three times a day even after massive lymphocytic infiltration had developed. Different findings were observed in regard to lymphocyte infiltration in the two different SS models: NFS/sld mice and NOD mice. In NFS/sld mice the lymphocyte infiltration improved after topical CyA, whereas it persisted in NOD mice. Nevertheless, topical CyA prevented acinar cell apoptosis in both animal models, which may be the reason for the functional improvement. We also investigated the possible mechanism of the effect of CyA and found that it suppressed lymphocyte adhesion to cultured LGs from the NFS/sld mouse and that it suppressed FasL expression on infiltrating lymphocytes in the LGs of NOD mice. The results of our study clearly suggest the therapeutic potential of CyA for the treatment of dry eye associated with SS. 

Methods

Animals

Female NFS/N strain mice carrying the mutant gene sld were maintained in our pathogen-free mouse colony (Department of Pathology, Tokushima University School of Dentistry) and given food and water ad libitum. Thymectomy was performed on day 3 after birth as previously described. ^{29 30} Sixty female mice were investigated in this study. 

Eight-week-old male NOD mice were obtained from Japan CLEA (Tokyo, Japan) and maintained under standard conditions in the animal facilities of Tokyo Dental College, Ichikawa General Hospital. Sixty NOD male mice were used in this study. All investigations adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Functional Assay: Measurement of Tear Production

Basal tear secretion in mice anesthetized with pentobarbital sodium (65 mg/kg; Abbott Laboratories, North Chicago, IL) was measured by the cotton thread test (Showa, Tokyo, Japan). Cotton thread was placed under the lower lid of each eye near the medial canthus for 5 minutes, and the length of wet thread was measured. The measurements of tear volume were performed four times during the 30-minute period of anesthesia from two data points in each animal. Basal secretion was calculated as mean tear volume/mean body weight. Weight was shown not to vary between the groups (P > 0.1). 

We developed a reliable method of measuring tear production in mice with commercially available cotton thread. We measured tear production over 30 minutes and calculated the mean value as basal tearing. Because stimulated tearing is also an important function of the LGs, we caused stimulated tear production by injecting pilocarpine and measured the response by the cotton thread test, as described. To measure pilocarpine-stimulated tear secretion, animals were intraperitoneally injected with 0.05 mg of pilocarpine (Wako, Osaka, Japan) in saline per 100 g body weight under pentobarbital sodium anesthesia (65 mg/kg). Stimulated tearing started within 5 minutes after stimulation and peaked between 10 and 20 minutes. We used the value between 10 and 20 minutes as the measurement of reflex tearing. Stimulated tear production was measured in the same manner as basal secretion, and the average of the second and third measurements (10 minutes after application of pilocarpine and 20 minutes after application) was divided by body weight and calculated as stimulated tear secretion. Data are expressed as means ± SEM of 6 to 12 eyes. 

Ocular Surface Evaluation

The ocular surface was evaluated 1 day before measuring tear secretion. Under pentobarbital sodium anesthesia (35 mg/kg), 1 ml of 1.0% sodium fluorescein in saline was applied to the eyes, followed by washing with saline. The cornea was then examined for fluorescein staining under slit lamp and scored on the following scale: 0, no staining; 0.5, slight punctate staining; 1, punctate staining of more than one quarter of the cornea; 2, presence of a corneal epithelial defect; 3, corneal epithelial defect covering more than one eighth of its surface area. Data are expressed as means ± SEM of 6 to 12 eyes. 

Topical Cyclosporin A Application

In the topical application study, CyA dissolved in 2 μl of ophthalmic solution (0.01% and 0.1% CyA eye drops; Santen, Osaka, Japan) or vehicle only was applied 3 times a day to both eyes with a micropipette. We had already confirmed that topical CyA could reach the LG and was effective for prevention of the disease when administered to NFS/sld mice at the age of 4 weeks, before lymphocyte infiltration occurred. ¹⁶ To investigate its therapeutic effect, we initiated CyA therapy at a later stage. Because massive lymphocyte infiltration becomes established at approximately 8 weeks after birth, application was begun 6 days a week in NFS/sld mice at 10 weeks of age. Animals were killed at age 16 weeks (after 6 weeks of application: n = 6 for vehicle, 0.01%, and 0.1%), at age 20 weeks (10 weeks of application: n = 6 for vehicle, 0.01%, and 0.1%), and at age 24 weeks (14 weeks of application: n = 6 for vehicle, 0.01%, and 0.1%). NFS/sld mice (n = 6) were used as starting point control animals and killed at the age of 10 weeks. 

CyA was applied to NOD mice in the same manner, beginning at 10 weeks of age when the LG infiltration was already severe. Animals were killed at age 14 weeks (after 4 weeks of application: n = 6 for vehicle, 0.01%, and 0.1%), at age 18 weeks (after 8 weeks of application: n = 6 for vehicle, 0.01%, and 0.1%) and at age 22 weeks (after 12 weeks of application: n = 6 for vehicle, 0.01%, and 0.1%). The NOD mice (n = 6) were used as starting point control animals and killed at the age of 10 weeks. 

Histologic Examination and Grading of Inflammatory Lesions

LGs were removed from the mice, and one half of each LG specimen was fixed with 10 mM phosphate-buffered saline (PBS) containing 4% formaldehyde (pH 7.2) and prepared for histologic examination. The sections were stained with hematoxylin and eosin and used for TdT-dUTP terminal nick-end labeling (TUNEL) assay. Histologic grading of the inflammatory lesions was performed according to the method proposed by White and Casarett. ³¹ Histologic evaluation of LGs was performed in a blind manner, and one tissue section of each LG was evaluated. The other half of each LG was embedded in optimal cutting temperature compound (OCT; Sakura, Tokyo, Japan), frozen in liquid nitrogen, sectioned at 8 μm on a cryostat, and stained as outlined in a later section. 

Detection of Apoptotic Cells

We performed in situ DNA nick-end labeling to detect apoptotic cells. ³² Paraffin-embedded tissue sections (4 μm) were deparaffinized by 3 washes in xylene (Koso, Tokyo, Japan), followed by 3 minutes of successive washes in 100%, 90%, and 70% ethanol and distilled water. Slides were equilibrated in PBS before deproteinization with proteinase K for 30 minutes at 37°C. DNA fragmentation was detected by using a kit (Mebstain; MBL, Nagoya, Japan) that incorporates biotin-conjugated dUTP into DNA by TdT. 

The reaction was allowed to continue at 37°C for 1 hour. Sections were covered with avidin-conjugated fluorescein for 30 minutes at 37°C and then counterstained by 0.5 μg/ml propidium iodide (Wako) for 15 minutes at 4°C. Negative controls consisted of tissue sections incubated as described earlier but without addition of TdT. The thymus of BALB/c mice was used as a positive control. 

TUNEL staining of NFS/sld mice of the LG was performed at the age of 24 weeks (after 14 weeks of application, six animals) and of the NOD mice at the age of 18 weeks (after 8 weeks of application, three to four animals). One tissue section per animal was used. 

Immunohistochemistry

OCT compound–embedded tissue sections (8 μm) were fixed with acetone and washed with 50 mm PBS (pH 7.4). Slides were blocked with 10% normal goat serum for 20 minutes and incubated for 30 minutes with rabbit polyclonal antibodies to FasL (N2; Santa Cruz Biochemistry, Santa Cruz, CA). The first antibodies were detected with Oregon Green 514-conjugated goat anti-rabbit IgG antibody (Molecular Probes, Leiden, The Netherlands). After washing with PBS, the sections stained with immunofluorescein were visualized with a fluorescence microscope (ACAS 570; Meridian Instruments, Okemos, MI). ⁸ The following microscope settings were used: wavelength = 488 nm, dichroic filter = 510 nm, step size = 1.0 μm, laser power = 200 mW, and scan strength = 10%. Background FasL fluorescence was measured by treating the liver of BALB/c mice with the primary and secondary antibodies, because FasL is not expressed in the liver of BALB/c mice. The average fluorescence density was determined by measuring five fields scanned in the LG by the microscope. FasL immunostaining by the LG of NOD mice was performed at 18 weeks of age (after 8 weeks of application, three to four animals). One tissue section per animal was used. 

Adhesion Assay

Six intact female NFS/N strain mice (5–6 weeks of age) were used. The methods of culturing thymocytes and lacrimal gland epithelial cells have been described. ³³ Mouse lacrimal gland epithelial cells were prepared as previously described. ^{33 34} Briefly, the lacrimal glands were collected from five female NFS/sld mice (3 weeks of age), decapsulated, minced into 1-mm² pieces, washed with Hanks’ balanced salt solution (HBSS) without Ca²⁺ and Mg²⁺, and placed in a 60-dish culture plate containing HBSS with 0.76 μg/ml EDTA, 4.9μ g/ml l-ascorbic acid, and 4.9 μg/ml reduced glutathione, ³⁵ and 20,000 lymphocytes were added to the cultured epithelial cells of each lacrimal gland. LG acinar cells pretreated with or without CyA were incubated with interleukin (IL)-2 (50 U/ml) and concanavalin A (ConA; 5 μg/ml) for 1 hour. After 10 hours, the wells were washed once with serum-free RPMI and Giemsa stained, and lymphocytes adhering to the lacrimal gland epithelial cells were counted under a microscope. The results are presented as mean percentage of adhesion ± SE of triplicate determinations in each experiment. CyA was not toxic to the lymphocytes or the lacrimal gland epithelial cells during the 10-hour incubation period under these conditions (data not shown). 

Systemic Evaluation

The thyroid, pancreas, bronchus, lung, kidney, and liver were examined histologically for lymphocyte infiltration and possible side effects of topical use of CyA. Lymphocyte infiltration was studied, and the general condition of the animals was carefully monitored. 

Statistical Analysis

The corneal fluorescein staining and LG lymphocyte infiltration scores were analyzed by the Kruskal–Wallis test followed by the Steel multiple comparison test. Basal and stimulated tear secretions were analyzed by analysis of variance followed by the parametric Dunnett multiple comparison test. The intensity of FasL fluorescein staining in the LG was analyzed by Student’s unpaired t-test. 

Results

Functional Improvement of Acinar Cells

The mean basal and stimulated tearing values at 10 weeks of age of NFS/sld mice were 0.21 ± 0.01 mm/5 minutes per gram body weight and 0.32 ± 0.03 mm/5 minutes per gram body weight (Figs. 1A 1B ). Both basal and reflex tear production were increased in the 0.1% CyA group at the age of 16 weeks (after 6 weeks of treatment, P < 0.01). At the age of 24 weeks (after 14 weeks of treatment), basal tearing had increased significantly (P < 0.01) in both the 0.1% and 0.01% CyA groups. Although the therapeutic effect of CyA on stimulated tearing was slightly weaker, topical CyA application improved LG function in NFS/sld mice. 

In the NOD mouse, basal secretion started to increase at the age of 18 weeks (after 8 weeks of treatment) in the 0.1% CyA group (P < 0.01), and the maximal effect was observed at the age of 22 weeks (after 12 weeks of treatment), with a significant increase in both the 0.1% (P < 0.01) and 0.01% (P < 0.05) CyA groups. The effect of CyA on stimulated tearing was slightly less, and the only significant increase was observed at the ages of 14 weeks (after 4 weeks of treatment) and 18 weeks (after 8 weeks of treatment) in the 0.1% CyA group (P < 0.01) and at the age of 14 weeks (after 4 weeks of treatment) in the 0.01% CyA group (Figs. 2A 2B ; P < 0.05). No significant increase was observed at the age of 22 weeks (after 12 weeks of treatment) in either the 0.1% or the 0.01% CyA group. 

Lymphocyte Infiltration of the LG

In NFS/sld mice, the grade of lymphocyte infiltration at the age of 10 weeks (before treatment) was 3.0 ± 0.5, indicating massive lymphocyte infiltration of the LG. At the ages of 16 weeks and 20 weeks (after 6 weeks and 10 weeks of treatment), there were no significant differences between the vehicle group and the CyA group, although a slight tendency for decreased lymphocyte infiltration was observed in the treatment groups (Fig. 3A ). At the age of 24 weeks (after 14 weeks of treatment), lymphocyte infiltration in the 0.1% CyA group was 1.8 ± 0.3, lower than in the vehicle group (2.8 ± 0.4; P < 0.01; Figs. 3A 3B 3C ). The results were dose dependent. The lymphocyte infiltration decreased after 14 weeks of treatment with 0.1% CyA eye drops. 

In the NOD mice, the grade of lymphocyte infiltration was 3.0 ± 0.3 at the age of 10 weeks (before treatment), showing establishment of the LG disease. There were no differences between the 0.1% CyA and 0.01% CyA treatment groups and the vehicle group at any time during the observation period (Figs. 4A 4B 4C ). The CyA eye drops had no effect on lymphocyte infiltration in the NOD mice. 

Ocular Surface Evaluation

The ocular surface of the NFS/sld and NOD mice showed very occasional staining. The ocular surface score varied from 0.1 ± 0.1 to 0.3 ± 0.1, and there were no significant differences during the observation period. The ocular surface of these animal models was unaffected by the decreased tear production. 

Prevention of Acinar Cell Apoptosis

Apoptosis of LG duct–acinar cells was analyzed in specimens from 24-week-old NFS/sld mice and 18-week-old NOD mice. Only LG acinar cells stained positively by the TUNEL method, and no lymphocytes were stained by TUNEL. There were TUNEL-positive cells in the vehicle groups of both models (Figs. 5A 6A ), whereas hardly any staining was observed in the LG of the 0.1% CyA groups (Figs. 5B 6B) . The TUNEL-positive cells were acinar cells, not lymphocytes, and 0.1% CyA effectively decreased the number of apoptotic acinar cells, whether lymphocyte infiltration had occurred or not. 

Suppression of Lymphocyte Adhesion

To determine the mechanism by which CyA decreased the inflammatory lesions in the LGs of the NFS/sld mice, we investigated the effect of CyA on lymphocyte adhesion to cultured LG epithelial cells from NFS/sld mice. CyA dose-dependently decreased lymphocyte adhesion to the LG epithelial cells of NFS/sld mice (Fig. 7) . 

Suppression of FasL Expression in Infiltrating Lymphocytes

In NOD mice, tear production increased and acinar cell apoptosis decreased in the CyA treatment groups, even when massive lymphocyte infiltration was present. To determine whether killing by infiltrating lymphocytes was suppressed in the CyA groups, we measured lymphocyte FasL expression. In the vehicle groups, FasL was expressed in many infiltrating lymphocytes, whereas treatment with CyA was associated with a dramatic decrease in expression of FasL on infiltrating lymphocytes (Fig. 8A 8B ). When the fluorescein staining was measured under the microscope, the average FasL-positive signal was 137 ± 27 in the vehicle groups, as opposed to only 37 ± 9 in the CyA-treated groups (P < 0.01, Fig. 9 ). 

Systemic Effect of CyA

No increased lymphocyte infiltration was detected in the thyroid, pancreas, bronchi, lung, kidney, or liver of either NFS/sld or NOD mice treated with CyA. No systemic side effects were observed in either NFS/sld or NOD mice, except for the ex-pected lymphocyte infiltration of other organs in the NOD mice. 

Discussion

It was very important to confirm increased tear secretion in response to CyA, because lymphocyte infiltration does not necessarily correlate with LG dysfunction. ^{17 18} Acinar cells can be abnormal without lymphocyte infiltration in certain mouse models of autoimmune diseases, including NOD and NFS/sld mice. ^{19 36} In this study we showed that tear secretion increased in both NOD and NFS/sld mice after topical use of CyA, confirming that CyA can lead to functional improvement in these animal models. However, there was no functional improvement in stimulated secretion in 22-week-old NOD mice (Fig. 2B) , suggesting that the effect of CyA may be limited when the disease has progressed. This should be further investigated in future experiments. 

Lymphocyte infiltration in NFS/sld mice was reduced by topical CyA which was started after the establishment of lymphocyte infiltration. Thus, CyA can be used to treat as well as prevent the disease. We showed that CyA could decrease lymphocyte adhesion to LG cells in vitro, suggesting a possible mechanism of action of CyA. CyA inhibits expression of intercellular adhesion molecule (ICAM)-1 in skin. ³⁷  

The drawback to our study is that we did not investigate the adhesion molecule of NOD mice and FasL expression of lymphocytes in NFS/sld mice. Without these controls, we cannot say definitively that the two mechanisms are present in both animals. There may be some overlapping mechanism for the improvement of tear function. However, it is our observation that FasL expression is suppressed in lymphocyte infiltration after CyA treatment in NOD mice. 

It was surprising that although CyA improved LG function in NOD mice, there was no decrease in lymphocyte infiltration. This resembles Mikulicz’s disease, in which lacrimal secretion is unaffected despite the massive lymphocyte infiltration. ⁸ In SS, the acinar cells undergo apoptosis and stop functioning, whereas in Mikulicz’s disease they do not. The LGs of NOD mice treated with CyA were similar to the LGs in Mikulicz’s disease. We have demonstrated by TUNEL staining that the acinar cells of the LGs of both NFS/sld and NOD mice show decreased apoptosis. CyA is known to inactivate T cells and suppress FasL expression. ³⁸ Our study showed that FasL expression is suppressed in lymphocytes infiltrating the LG after CyA treatment. Thus, the mechanism of CyA action may be prevention of apoptosis by suppression of expression of FasL, even though the lymphocyte infiltration itself was not prevented. This resembles insulitis without diabetes in MRL-lpr/lpr animals, in which the Fas–FasL system does not function. ³⁹  

It is interesting that CyA prevents lymphocyte infiltration in one animal model of SS (NFS/sld) and suppresses FasL expression by infiltrating lymphocytes in the other model (NOD). Because both animals showed functional improvement, CyA may be indicated for the clinical treatment of dry eye. However, our observations provide an important warning for human clinical application of CyA. Although the fundamental mechanism of action of CyA is T-cell suppression, the outcome is not always similar. When CyA was given to NFS/sld mice systemically, lymphocyte infiltration in the LGs and SGs was aggravated, and there was increased lymphocyte infiltration of the lung, kidney, and pancreas. ¹⁶ Thus, CyA may have a negative effect at certain concentrations in certain animals because of different mechanisms. This possibility is particularly important because CyA may be used for long periods of time. 

In summary, functional improvement of the LG was achieved by the topical application of 0.1% CyA in two mouse models. The mechanism of action of CyA seems to be prevention of acinar cell apoptosis through either prevention of lymphocyte infiltration or suppression of FasL expression on infiltrating lymphocytes. The mechanism of CyA’s effects should be investigated before clinical application of CyA, with careful attention to unexpected side effects. 

 Supported by grants from Oral Health Science Center, Tokyo Dental College; Santen Pharmaceutical Co., Ltd.; and the Medical School Faculty and Alumni Grants of Keio University Medical Science Fund.

 Submitted for publication November 30, 1999; revised May 16 and September 20, 2000; accepted September 29, 2000.

 Commercial relationships policy: F (KT, HF); N (all others).

 

Figure 1.

 

(A) Effects of CyA eye drops on basal tear secretion in the NFS/sld mouse. Each point represents the mean ± SE in 10 to 12 eyes. **P < 0.01, *P < 0.05 versus vehicle (parametric Dunnett multiple comparison test). The highest secretion occurred in the 0.1% CyA group. (B) Effects of CyA eye drops on pilocarpine-stimulated tear secretion in the NFS/sld mouse. Each point represents the mean ± SE of 10 to 12 eyes. **P < 0.01 versus vehicle (parametric Dunnett multiple comparison test). The highest secretion is in the 0.1% CyA group. (•) Vehicle; (⋄) 0.01% CyA; (○) 0.1% CyA.

Figure 2.

 

(A) Effects of CyA eyedrops on basal tear secretion in the NOD mouse. Each point represents the mean ± SE of 6 to 10 eyes.** P < 0.01, *P < 0.05 versus vehicle (parametric Dunnett multiple comparison test). The highest secretion is in the 0.1% CyA group. (B) Effects of CyA eye drops on pilocarpine-stimulated tear secretion in the NOD mouse. Each point represents the mean ± SE of 6 to 10 eyes. **P < 0.01, *P < 0.05 versus vehicle (parametric Dunnett multiple comparison test). The highest secretion was in the 0.1% CyA group, which at 18 weeks showed significant difference in volume, whereas by 22 weeks, there was no significant difference in volume. (•) Vehicle; (⋄) 0.01% CyA; (○) 0.1% CyA.

Figure 3.

 

(A) Effect of CyA on inflammatory lesions in the LGs of the NFS/sld mouse. Inflammatory lesions were graded according to the method of White and Casarett. ³¹ Each point represents the mean ± SE of five to six glands.** P < 0.01 versus vehicle (nonparametric Dunnett multiple comparison test). The grade was lowest in the mice treated with 0.1% CyA. (•) Vehicle; (⋄) 0.01% CyA; (○) 0.1% CyA. (B) Histopathology of the LG inflammatory lesions in the NFS/sld mouse at the age of 24 weeks (vehicle) showed massive lymphocyte infiltration. (C) Histopathology of the LG inflammatory lesions in the NFS/sld mouse at the age of 24 weeks (0.1% CyA group) showed dramatically decreased lymphocyte infiltration. Bar, 100 μm.

Figure 4.

 

(A) Effect of CyA on inflammatory lesions in the LGs of the NOD mouse. The inflammatory lesions were graded according to the method of White and Casarett. ³¹ Each point represents the mean ± SE of five to six glands. No changes were observed at any time during the study period. (•) Vehicle; (⋄) 0.01% CyA; (○) 0.1% CyA. (B) Histopathology of the LG inflammatory lesions in NOD mice at the age of 18 weeks (vehicle) showed massive lymphocyte infiltration. (C) Histopathology of the LG inflammatory lesions in NOD mice at the age of 18 weeks (0.1% CyA group) showed no decrease in lymphocyte infiltration. Bar, 100 μm.

Figure 5.

 

Effect of CyA on apoptotic cells in the LGs of NFS/sld mice. (A) TUNEL staining of an LG of NFS/sld mouse at the age of 24 weeks (vehicle). Some cells were clearly stained and showed apoptotic signs. (B) TUNEL staining of an LG of NFS/sld mouse at the age of 24 weeks (0.1% CyA). No cells showed apoptosis. Bar, 50 μm.

Figure 6.

 

Effect of CyA on apoptotic cells in the LGs of NOD mice. (A) TUNEL staining of an LG of a NOD mouse at the age of 18 weeks (vehicle). Some cells were clearly stained and showed apoptosis. (B) TUNEL staining of an LG of a NOD mouse at the age of 18 weeks (0.1% CyA). No cells showed apoptosis.

Figure 7.

 

Effect of CyA on lymphocyte adhesion to the LG epithelial cells of NFS/sld mice. CyA decreased lymphocyte adhesion to LG epithelial cells in a dose-dependent fashion. Each value represents the mean ± SE of triplicate determinations in each experiment.

Figure 8.

 

Effect of CyA on expression of FasL in the LGs of NOD mice. (A) Vehicle group. Infiltrating lymphocytes showed strong fluorescein staining, suggesting upregulation of FasL expression. (B) CyA (0.1%)-treated group. Infiltrating lymphocytes did not show fluorescein staining, suggesting downregulation of FasL expression.

Figure 9.

 

Comparison of FasL expression in the LGs of NOD mice treated with vehicle or CyA for 8 weeks (at the age of 18 weeks). Each value represents the mean ± SE of three to four mice. Statistical comparison by unpaired Student’s t-test of NOD mice treated with vehicle or 0.1% CyA (**P < 0.01).