A New Model of Experimental Autoimmune Keratoconjunctivitis Sicca (KCS) Induced in Lewis Rat by the Autoantigen Klk1b22

Abstract

purpose. This study was designed to generate an inducible autoimmune model of keratoconjunctivitis sicca (KCS) for study of pathogenesis of the disease.

methods. Lewis rats were immunized with a mixture of lacrimal and salivary gland extract or recombinant mouse protein kallikrein 1b22 (Klk1b22) emulsified in complete Freund’s adjuvant (CFA). For disease induction by adoptive transfer of primed cells, donor rats were received with T-cell blasts. KCS were observed by either clinical signs or histology.

results. The autoantigen Klk1b22, isolated from the lacrimal and salivary glands, readily induced Sjögren’s syndrome (SS)-like KCS in the recipients. The diseased animals presented the clinical and pathologic symptoms that resemble related human disease. Most immunized rats showed an increase, then a decrease in tear volume, together with corneal opacity and ocular lesions. Histologic examination revealed that the rats displayed the cardinal signs of primary SS-like KCS, including marked lymphocytic infiltration of the lacrimal and salivary glands and destruction of the acinar cells. Immunofluorescence studies showed that both CD8⁺ and CD4⁺ T cells were heavily infiltrated, with the former cells predominant in the damaged ducts. Finally, adoptive transfer of Klk1b22-reactive T cells induced more severe disease with earlier onset.

conclusions. Klk1b22 is an autoantigen for inducing an experimental SS-like KCS in Lewis rats. The availability of this new and reproducible rat model should provide a new and needed tool for studying the pathogenesis of SS.

Sjögren’s syndrome (SS) is a common autoimmune disease in which the infiltration and destruction of the salivary and lacrimal glands leads to xerostomia (dry mouth) and keratoconjunctivitis sicca (KCS; dry eyes). This disease affects approximately 0.5% to 2.7% of the general population, and most commonly occurs in middle-aged and older women. ¹ SS is classified as either primary or secondary. In primary (p)SS, autoimmune destruction is limited to the exocrine glands, whereas secondary SS is frequently associated with other autoimmune diseases, such as lupus erythematosus, rheumatoid arthritis, and scleroderma. pSS presents with varying degrees of severity, which may arise through different combinations of genetic and environmental factors. ² The immunologic basis for the disease pathogenesis is not clear. Given that clinical signs associated with human SS often appear late in disease progression, a suitable animal model is needed to assess the early events and underlying mechanisms responsible for the glandular hypofunction in patients. 

Several mouse models of SS have been reported. ³ Most of these are spontaneously developed disease occurring in genetically defective mice, such as NOD, NFS/sld, MRL/lpr, NZB/W F1, IQI/Jic, and Aly/aly mice ^{4 5 6 7 8 9} ; transgenic mice overexpressing immunoregulatory genes, such as HTLV-1 tax, BAFF, IL-10, and IL-14a ^{10 11 12 13} ; and knockout mice deficient of TGF-β1, aromatase, ID3, ^{14 15 16} and class 1A phosphoinositide-3-kinase in T cells. ¹⁷ Inducible models of SS were also reported, such as those induced by immunization with salivary gland extracts ¹⁸ or lacrimal gland extracts and autoantigens ^{19 20} or carbonic anhydrase II. ²¹ These models have been used to study the immune mechanisms of SS; however, the mechanisms that lead to activation of autoreactive T cells and disease development remain largely unclear. 

Here we demonstrate that immunization with the protein kallikrein 1b22 (Klk1b22) and transfer of Klk1b22-stimulated T cells in naïve rats could induce SS-like symptoms. The immunized rats showed altered tear production (first increased followed by decreased tear volume), inflammation in the cornea, and blepharitis beginning on day 10 post-immunization, marked lymphocytic infiltration of the lacrimal glands and damaged acinar cells and ducts, and elevated titers of anti-SS antigen B (SSB) antibody. Immunohistochemical studies showed a large number of infiltrations of CD8⁺ and CD4⁺ T cells in the lacrimal glands. Our results indicate that Klk1b22 is an autoantigen that can initiate both autoreactive T-cell activation and autoantibody production, leading to organ-specific autoimmunity. This established rat model should provide us with a needed tool for studying the pathogenesis and therapeutic intervention of pSS. 

Materials and Methods

Animals

Female Lewis rats, aged 6 to 8 weeks, obtained from Harlan-Sprague–Dawley (Indianapolis, IN), or female C57BL/6 mice, aged 6 to 10 weeks, obtained from Jackson Laboratory (Bar Harbor, ME) were housed and maintained in the animal facilities of the University of Louisville. The treatment of the animals conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Induction of Experimental Autoimmune SS

Lewis rats were immunized subcutaneously (SC) with 200 μL of an emulsion containing 500 μg of lacrimal and salivary gland extract or 200 μg of recombinant mouse Klk1b22 and 500 μg of Mycobacterium tuberculosis H37Ra (Difco, Detroit, MI) in incomplete Freund’s adjuvant (Sigma, St. Louis, MO), distributed over six spots on the tail base and flank. A single dose of 200 ng of pertussis toxin (PTX) was injected intraperitoneally (IP) on the same day as the antigen. 

For adoptive transfer of SS by transfer of antigen-specific T cells, nylon wool-purified T cells prepared from the draining lymph nodes along the injection sites including popliteal, superficial inguinal, axillary, and lateral axillary lymph nodes and spleen of rats immunized 12 days previously with gland extract, fraction VII purified from the extract, or Klk1b22 were cultured with 20 μg mL⁻¹ of the test antigen (Ag). After 2 days, the T cells were injected IP in 0.5 mL of PBS into naïve Lewis recipients (5 × 10⁶ cells/rat). 

Clinical Evaluation of SS

Animals immunized with Ag, or after adoptive transfer of Ag-specific T cells, were examined twice a week for clinical signs of SS, such as tear volume changes (Schirmer’s test), ^{22 23} cornea epithelial lesions (superficial punctate keratitis [SPK]), ^{24 25} or inflammation of the eyelids (blepharitis). ²⁶ All the tests were performed within 10 minutes after anesthetization by the cocktail containing ketamine HCl (75 to 100 mg/kg; Lloyd Laboratories, Shenandoah, IA) and xylazine HCl (10 mg/Kg; Hospira Inc., Lake Forest, IL) in PBS. 

Schirmer’s Test.

A 1 mm × 7 mm strip of filter paper was used to measure the tears produced over a period of 2 minutes. The strip was placed at the junction of the middle and lateral third of the lower eye lid with 1 mm of the strip inside the eyelid. The test was performed in the morning under ambient light and in the same location. The results of the test were classed as abnormal if, after subtraction of the 1 mm inside the eyelid, there was less than 1 mm or more than 4 mm wetting of the filter paper in 2 minutes. 

Superficial Punctate Keratitis (SPK).

The staining test was performed in the rats after anesthesia by applying 1% fluorescein sodium to the surface of the eye and removing excess dye by washing with PBS, leaving only dye bound to corneal abnormalities. These were visualized under cobalt blue light directed on the eye through a slit-lamp. Fluorescein staining was graded according to the area and density of the stained lesions, as reported previously. ^{24 25} The area of SPK was graded from A0 to A3 and the density from D0 to D3. The grading was the area (A0–A3) multiplied by the density (D0–D3). Grade I is score 1 to 3; Grade II, score 4 to 6; Grade III, score 7 to 9 (Table 1) . 

Blepharitis.

Blepharitis was diagnosed by the appearance of crusty debris (in the lashes or the corner of the eyes or on the lids) and erythematous eyelid margins. 

Disease was graded on a scale of 0 (no disease) to 4 (maximum disease) depending on the presence of these three clinical signs (Table 2) . 

Pathologic Evaluation of SS

Inflammation and damage to the lacrimal and salivary glands were confirmed by histopathology. Whole tissues were fixed in 10% buffered formalin (pH 7.4) for 24 hours and transferred to 70% ethanol until processed. The fixed and dehydrated tissue was embedded in paraffin, and 5 μm sections were cut and stained with hematoxylin and eosin. Lacrimal gland sections were graded according to previously published methods, ²⁷ and modified (Table 3) . 

Antigen Extraction

Total proteins were extracted from lacrimal glands and salivary glands of B6 mice by homogenization and sonication following the method of a total protein extraction kit (BioChain Institute, Inc., Hayward, CA). Briefly, the glands were isolated from PBS perfused mice, cut into small pieces, and incubated for 5 minutes on ice with 1.0 mL per gram tissue of protease inhibitors in PBS (Sigma). The tissues were then sonicated on ice for 3 times of 10 seconds each using a sonic dismembrator (model 100; Fisher, Pittsburgh, PA) and the sonication centrifuged at 13,000g at 4°C for 20 minutes. The supernatant was collected and the protein concentration determined using Bradford reagent (Sigma). 

Purification of the Antigen

The supernatant from the extraction (8 mg of protein in 2 mL) was applied to a 2.6 × 60 cm gel filtration column (HiPrep Sephacryl S-200 column; Pharmacia, Piscataway, NJ) previously equilibrated with running buffer (0.05 M sodium phosphate, 0.15 M NaCl [pH 7.2]). The column was eluted with running buffer at a rate of 0.4 mL/min, and 0.4 mL samples collected in vials. The protein concentration in each vial was measured using Bradford reagent (Sigma). Ten fractions were made, based on the protein concentration curve. 

SDS–PAGE Electrophoresis and Western Blotting

SDS–PAGE electrophoresis was carried out using a 12% Bis–Tris gel at a constant voltage of 200 V for 45 minutes at room temperature and the gel stained for 2 hours with Coomassie brilliant blue G-250. 

For Western blotting, the proteins on the gel were transferred to a nitrocellulose filter at a constant voltage of 30 V for 1 hour. The filter was then blocked for 1 hour at room temperature with 5% nonfat milk and incubated for 2 hours at room temperature with serum samples (diluted 1:200) from diseased rats or patients, followed by reaction with horseradish peroxidase (HRP)-conjugated anti-rat or human IgG (diluted 1:10,000) (Bio-Rad) and visualization with a system for quantitative analysis (ECL system; Amersham Biosciences, Piscataway, NJ). 

Trypsin Digestion and Mass Spectrometry Analysis

To obtain peptides for mass spectrometry analysis, protein bands were excised from the SDS-PAGE gel and digested with trypsin by a modification of the method of Jensen et al. ²⁸ The tryptic peptides were applied by a thin film-spotting procedure for MALDI-MS analysis using α-cyanohydroxycinnamic acid as the matrix on stainless steel targets, as described by Jensen et al. ²⁸ Mass spectral data were obtained using a mass spectrometer (TOF-Spec 2E; MicroMass Communications, Cary, NJ) and a 337 nm N₂ laser at 20% to 35% power in reflector mode. Spectral data were obtained by averaging 10 spectra, each of which was the composite of 10 laser firings. Mass axis calibrations were accomplished using peaks from tryptic auto-hydrolysis ²⁹ . The peptide masses obtained by MALDI-MS analysis were used to search the National Center for Biotechnology Information database (NCBI; www.matrixscience.com) to identify the proteins from which they originated. A MOWSE score more than 71 indicated a significant match, ensuring the probability of the match not being a random event. 

Generation of Recombinant Mouse Klk1b22 Protein

Using salivary cDNA from B6 mice as the template, the open reading frame encoding the full-length of mouse Klk1b22 was amplified by PCR using the primers 5′-GGA TCCATGAGGTTCCTGATCCTGTTCC-3′ and 5′-CTCGAGGGGGTTTTTGGCCATAGTGTCT-3′. The PCR products were digested with BamHI and XhoI and ligated into the BamHI/XhoI sites of pET41 (Novagen, San Diego, CA). Expression of the recombinant protein fused to glutathione S-transferase U (GST) was induced in Escherichia coli BL21 by incubation with 1 mM isopropyl-β-D-thiogalactoside. Inclusion bodies were extracted using bacterial protein extraction reagent (Pierce, Rockford, IL) and solubilized in 8 M urea, and the protein was purified by gel permeation chromatography on a gel filtration column (Superdex 200HR; GE Healthcare Life Sciences, Piscataway, NJ) equilibrated with 8 M urea and examined by SDS-PAGE gel electrophoresis and Western blotting. 

Immunohistochemical Staining

Purified primary antibodies against rat CD3⁺, CD4⁺, CD8⁺, or B cells (OX12) and appropriate biotinylated second antibodies were purchased from Vector Laboratories (Burlingame, CA). Paraffin-embedded tissue slides were deparaffinized and rehydrated with xylene and 100%, 95%, and 80% ethanol. Endogenous peroxidase activity was quenched for 10 minutes at RT with 0.15% hydrogen peroxide solution. After antigen retrieval with a citrate-buffered solution in a boiling water bath, the tissue was blocked in 5% goat serum for 1 hour at RT. The slides were incubated overnight at 4°C with primary antibodies, for 1 hour at RT with biotinylated second antibodies and for 45 minutes at RT with HRP-labeled streptavidin (R&D, Minneapolis, MN). Bound antibody was visualized using diaminobenzidine (DAB; DAB Substrate Kit; Vector Laboratories, Burlingame, CA) and the slides were counterstained with methyl green solution (Methyl Green Nuclear Counterstain; Sigma). 

Proliferation Assay

T cells from Lewis rats immunized with lacrimal and salivary glands extract, Fraction VII, or recombinant Klk1b22 were prepared and seeded at a density of 4 × 10⁵ cells/well in 96-well plates and cultured at 37°C for 72 hours in a total volume of 200 μL of medium with or without the test Ag in the presence of irradiated syngeneic spleen APCs (1 × 10⁵), and [³H]thymidine incorporation during the last 8 hours was assessed using a microplate scintillation counter (Packard Instruments, Meriden, CT). 

Anti–La-SSB ELISA

Serum samples (100 μL; 1:100 dilution) from naïve and diseased rats or from healthy subjects and SS patients were added to wells pre-coated with 100 μL of human SSB Ag (300 ng; Biospecific, Emeryville, CA), followed by reaction with HRP-conjugated anti-rat or human IgG (1:10,000 dilution; Sigma). 

Results

Induction of Lacrimal and Salivary Gland Inflammation by Immunization with Protein Extracts Derived from Mouse Lacrimal and Salivary Glands

To identify the possible autoantigen that might induce experimental SS in Lewis rats, we prepared the tissue extracts of lacrimal and salivary glands from C57BL/6 mice, emulsified them in CFA, and immunized the emulsion to naïve Lewis rats. The immunized animals were then monitored for clinical signs of keratoconjunctivitis sicca (KCS), including alteration of tear secretion, superficial punctuate keratitis (SPK), and blepharitis twice a week. As shown in Figure 1A , the animals (n = 20) showed altered tear secretion approximately 10 days post-immunization. In the early phase of the disease (first 30 days post-immunization), the immunized rats appeared to have significantly increased tear secretion; reduced tear secretion became apparent after 30 days post-immunization. On the other hand, the incidence of SPK was significantly increased after day 35 post-immunization (Fig. 1B) . Blepharitis was not a common sign in these animals (Fig. 1C) . Nearly 70% of the immunized rats developed one, two, or all three of these indicators. Histologic analysis revealed various degrees of lymphocytic infiltration and destruction of the lacrimal glands in rats with clinical signs (Fig. 2 , Table 3 ). The diseased tissues showed multiple lymphocytic paraductal foci, destruction of the acinar structures, and hypertrophy of the ductal structures in the lacrimal glands (Fig. 2A) . The percentage of tissues of each grade at days 7 to 30 and 30 to 120 post-immunization are also summarized in Figure 2B . A point worthy of note is that the degree of tissue damage did not always parallel to the clinical scores (data not shown). Immunohistochemical staining of the lacrimal glands revealed that, in earlier and milder disease, CD4⁺, CD8⁺ T cells, and OX-12+ B cells were all detected among the infiltrating cells of the inflamed tissues; however, in severe cases, CD8⁺ T cells became apparently dominant (Fig. 3A) . At day 24 post-immunization, when tear production was above normal, CD3⁺ cells were observed in the conjunctiva, but not in the cornea (Fig. 3B) . Approximately 70% of the lacrimal tissues obtained from the immunized rats had inflammatory infiltrations, among which a few cases lacked clinical symptoms. No visible pathology was detected in other organs, such as the lung, liver, and spleen; only two out of 11 kidneys examined showed mild abnormalities (grade I; Fig. 3B ). No immunohistological changes occurred in other ocular tissues such as choroid and retina (data not shown). 

Identification of Pathogenic Autoantigen in the Tissue Extracts

To identify the pathogenic autoantigen in the tissue extract of the salivary and lacrimal gland, the protein extracts of the lacrimal and salivary glands were fractioned by a gel filtration column. Ten crude protein fractions were prepared, based on the protein size (Fig. 4A) . When each fraction was tested for their ability to induce proliferative response by T cells from immunized rats, a strong response to the protein in fraction VII was observed (Figs. 4B and 4C) . Moreover, immunization to naïve rats with proteins in the fraction VII (100 μg of protein) induced SS in 12 of 16 rats (Table 4) . The clinical and histopathological inflammation observed in these animals was similar to that induced by the extract (Figs. 1 and 2) . In contrast, none of the rats injected with other fractions developed SS (Table 4) . 

Klk1b22, the SS Immunogenic Ag

To determine the disease-inducing molecule in the fraction VII, we compared protein composition between fraction VII and fraction I. SDS-PAGE analysis showed that all but three of the protein bands are indistinguishable between the pathogenic fraction VII and fraction I. The approximate molecular weights of these bands were 16 to 36 kDa (Fig. 5A) . Subsequently, Western blotting (Fig. 5B)identified a protein with a molecular mass at 30 kDa which reacted with serum samples from SS rats, but not with samples from naïve rats and rats with experimental autoimmune uveitis (EAU). Peptide mass fingerprinting identified the protein as Klk1b22 with 51% protein coverage (Fig. 5C) . 

To confirm that Klk1b22 was the Ag responsible for SS, a recombinant fusion protein consisting of mouse Klk1b22 fused to GST was prepared (Fig. 6A) . We then tested the ability of GST-Klk1b22 to induce disease. Animals were immunized with gland extract, GST-Klk1b22, or GST inclusion. The GST-K1k1b22–immunized rats, but not the GST-immunized rats, developed disease clinically and histologically identical with that induced by fraction VII (data not shown). T cells from the GST-Klk1b22–immunized rats responded strongly to GST-Klk1b22, less to gland extract, and not at all to GST or other unrelated Ags, such as the interphotoreceptor retinoid-binding protein (IRBP) peptide 1177–1191 (R16) used to induce EAU (Fig. 6B) . 

To further test whether Klk1b22-reactive T cells could induce disease, we harvested lymph nodes and spleen from fusion protein–injected rats at 10 days post-immunization, purified T cells using a nylon-wool column, stimulated them with GST-Klk1b22 for 2 days, and injected the blast activated T cells (5 × 10⁶/rat) into naïve Lewis rats. As shown in Table 5 , severe SS was induced. Disease started earlier than with immunization (day 6 compared to day 10) and the histopathological features in these animals were more severe than those induced by immunization (score 3 compared to 2). 

Autoantibodies in SS Rats

Autoantibodies to the La/SSB ribonucleoprotein are commonly found in patients with Sjögren’s syndrome. Since the homology between human and rat is more than 90%, we used human SSB as coating antigen in an ELISA. Analysis of sera (1:100 dilution) from rats with SS induced by gland extract, fraction VII, or Klk1b22-GST showed a significantly higher titer of anti-SSB antibodies than samples from naïve and EAU rats (Fig. 7) , the titer being highest in the rats immunized with Klk1b22-GST. 

Discussion

Despite the high prevalence of SS in humans, the disease pathogenesis is poorly understood and treatment options are mainly palliative. Spontaneous disease models have provided some understanding of disease pathogenesis. Unfortunately, autoantigens that are responsible for the disease remained elusive and immune cells that caused the disease have not yet been characterized. 

By immunization with an extract of mouse salivary or lacrimal glands, we have successfully induced SS in the Lewis rat. Furthermore, success of inducing SS by adoptive transfer of T cells from the immunized rats confirmed the existence of pathogenic protein in the salivary and lacrimal glands. By screening the cellular response using an in vitro T-cell proliferation assay and disease-inducing ability, we have revealed that the Klk1b22 of molecular weight of 30KDa is the major responsible autoantigen. 

It is not unusual in experimental autoimmune models, especially actively induced models (antigen-immunized) in contrast to adoptively transferred models with sensitized lymphocytes, to find that not all the immunized animals develop autoimmune disease (e.g., 25% of the rats immunized with fraction VII had mild or no SS). The variable incidence of disease may be attributed to the fact that the number of autoreactive T cells induced by immunization may differ from host to host—that is, biological variation. In fact, adoptive transfer studies usually demonstrate more consistent results, since T cells from multiple rats are injected into one recipient. 

Klks are a family of trypsin-like or chymotrypsin-like secreted serine proteases generally classified into two major categories: one in blood, called plasma Klk, and another secreted mainly by glandular organs called tissue Klk, in which there are 15 subfamily members in human, 25 in mouse, and 13 in rat. ³⁰ Their general function is to convert inactive kininogen to active kinins, which regulate the local blood pressure, blood flow, and electrolyte balance. ³¹ Recently, evidence has been obtained that suggests that the kallikrein cascade plays a significant role in the initiation and maintenance of inflammatory responses. ^{32 33} In the spontaneous mouse model of SS in IQI/Jic mice, tissue Klk-13 has been identified as an autoantigen. ³⁴ The autoantibodies in the sera from these mice (>12 weeks of age) with SS bind to Klk-1 and Klk-13. Klk-13, but not Klk-1, induces a proliferative response of splenic T cells from IQI/Jic mice from the age of 4 weeks onward. In addition, markedly enhanced Klk-13 expression is observed in the salivary glands of these mice. These results indicate that Klk-13 acts as an autoantigen and may increase the number and activity of T cells responsive to organs expressing Klk-13. Additionally, saliva from SS patients has been reported to contain significantly increased levels of Klks considered to be derived from affected duct epithelial cells. ^{35 36} Our preliminary data also revealed that antibodies in the sera of SS rats and SS patients bind to recombinant mouse KLk1b22 on immunoblot (data not shown). Because of the high homology in the amino acid sequences among these Klk members, ^{37 38} they may share the common antigenic domain to initiate T or B cell-mediated autoimmune reactions. 

Previous studies have shown that multiple antigens may be involved in the pathogenesis of this disease, such as Ro antigen, ³⁹ La antigen, ⁴⁰ and α-fodrin, which was present exclusively in epithelial duct cells of the salivary glands in NOD mice but not in control BALB/c mice. ⁴¹ The KLK1b22 protein apparently differs from the previously characterized ones, in that Ro (SS-A) is a 60-kDa protein and α-fodrin, a molecular weight of 120-KD. 

The success of disease induction by Klk1b22-reactive T cells emphasizes the role of T cells in disease initiation and progression, in which both CD4⁺ and CD8⁺ T-cell subsets are involved, since both were observed in the inflamed glands (Fig. 3A) . Large numbers of CD8⁺ T cells were found around the ducts of the lacrimal glands, and in their epithelial lining. Further examination of whether CD8⁺ T cells initiate epithelial and acinar cell death or accelerate cellar atrophy caused by CD4⁺ T cells will help to identify potential therapeutic target(s) to prevent or halt disease progression. In addition, the role of effector CD4⁺ T-cell subsets (Th1, Th2, and Th17) in inflammation will also need to be studied by examining patterns of cytokine production. 

It remains to be determined whether the anti-Klk1b22 autoantibodies are involved in the pathogenesis of disease. The occurrence of B cells in the inflamed glands and autoantibodies in diseased rats may partially account for the fact that the degree of local destruction within exocrine glands does not always correlate with secretory dysfunction. ⁴² Glandular dysfunction does not arise solely from destruction associated with lymphocytic infiltration, suggesting other mechanisms are involved, including autoantibodies and cytokines. 

We have shown in Figure 1that more than 25% of immunized rats developed severe corneal epitheliopathy as early as day 7, yet tear production was actually increased at that time point, a common clinical observation in patients with corneal epitheliopathy. This may reflect autoimmune inflammation of the corneal or conjunctival epithelium, or compositional changes of tears because of dysfunction of other tear secreting cells (e.g., cells in the meibomian or harderian glands, or the conjunctival goblet cells). CD3⁺ T cells were detected in the conjunctiva at day 24 post-immunization, when tear production is above normal, indicating that primary inflammation might occur in the conjunctiva in direct contact with the cornea. In addition, Klk1b22 in the epithelial cells of these glands may be the target for autoimmune-associated inflammation, with secondary plugging of the tear glands along the eyelid margins leading to blepharitis. The localization of Klk1b22 in these tissues by direct staining with anti-Klk1b22 antibodies may provide direct evidence for the observations. 

Disease in our model mainly affected the salivary and lacrimal glands, with mild penetration in the kidney (Fig. 3C) . Although Klks are detected in various organs, they are expressed most abundantly in the duct epithelial cells in the salivary glands, ^{31 43} where lymphocytic infiltration initially appears in IQI/Jic mice ⁴⁴ and in our model. This suggests that the involvement of the target organs correlates with the level of the Ag present. A kinetic study using our model is required to understand whether an autoimmune reaction against Klks plays a critical role in disease progression from periductal infiltrations within salivary glands to systemic autoimmunity targeting multiple organs. 

In summary, a new 30 kDa protein Klk1b22 has been characterized as and autoantigen that induces SS in Lewis rats. Transfer of Klk1b22-specific T cells induced SS with the cardinal features of human pSS with an earlier onset and higher intensity than in immunized rats. This model provides a useful tool for studying the immune mechanisms involved in the disease and for developing new therapies and diagnostic tests. 

²  These authors contributed equally to this work and should therefore be regarded as equivalent authors.

 Supported in part by NIH Grants NEI EY12974, EY14599; Vision Research Infrastructure Development (R24 EY015636); a Research to Prevent Blindness (RPB) career development award (HS); an RPB Robert E. McCormick scholar award (MMJ); the Commonwealth of Kentucky Research Challenge Trust Fund (HJK); and an unrestricted grant from RPB, New York, NY.

 Submitted for publication February 28, 2008; revised June 5, August 11, and October 15, 2008; accepted March 10, 2009.

 Disclosure: G. Jiang, None; Y. Ke, None; D. Sun, None; H. Li, None; M. Ihnen, None; M.M. Jumblatt, None; G. Foulks, None; Y. Wang, None; Y. Bian, None; H.J. Kaplan, None; H. Shao, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

The Grade of SPK

Corneal Surface Area	Density of Damaged Lesions
A0 No punctate staining	D0 No punctate staining
A1 Less than one-third	D1 Sparse density
A2 One-third to two-thirds	D2 Moderate density
A3 More than two-thirds	D3 High density with overlapping lesion

Table 2.

 

Clinical Evaluation of SS

Score	Description
		Tear Secretion*	SPK, †	Blepharitis
0	Normal	0–3	No
I	Abnormal	0–3	No
II	Normal or abnormal	4–6	No
III	Normal or abnormal	<6	Yes
IV	Total loss	9	No or yes

*  Schirmer’s test was performed under anesthesia in the morning. A result >4 mm or <1 mm in 2 min was classed as abnormal.

†  The SPK score was evaluated by staining the cornea with fluorescein and graded.

Table 3.

 

Histological Evaluation of SS

Score	Description
0	Acinar cell border is well defined; extracellular space is apparent; no nuclear drop-out; no infiltration of cells; no acinar atrophy
I	Acinar cell border is less defined; extracellular space is diminished; partial nuclear drop-out; blood vessel and small duct dilation
II	Cellular infiltration localized to small ducts and blood vessels as well as paraductal acinar cells
III	Acinar cell atrophy primarily localized to paraductal acinar cells
IV	Focal area of severe tissue damage characterized by acinar cell atrophy and extensive cellular infiltration and small duct dilation

Figure 1.

 

Lewis rats immunized with a mixed mouse salivary and lacrimal gland extract develop SS. Kinetics of changes in tears using Schirmer’s test (A) and the incidence of SPK (B) in the immunized animals (n = 20 in both). (C) Presence of SPK on the cornea of a naïve (left) and an immunized (center) rat detected by fluorescent dye, and (right) blepharitis—inflammation around the eyes. The values shown in (A) are means ± SD. *P < 0.05 and **P < 0.01, compared with naïve control.

Figure 2.

 

The range of histopathological changes of the lacrimal glands in immunized rats. (A) Hematoxylin/eosin (H&E) staining of the lacrimal glands from naïve and immunized rats at day 35 post-immunization (magnification, ×10 left; ×40 right). The disease was scored as follows: grade 0: well defined acinar cell border (▴), apparent extracellular space, no nuclear drop-out, no infiltration of cells, no acinar atrophy; grade I: acinar cell border is less defined (▴), extracellular space is diminished, partial nuclear drop-out (↑), blood vessel and small duct dilation (∗); grade II: cellular infiltration localized to small ducts (▴) and blood vessels (↑) as well as paraductal acinar cells (*); grade III: acinar cell atrophy primarily localized to paraductal acinar cells (▴); grade IV: focal area of severe tissue damage characterized by acinar cell atrophy (▴) and extensive cellular infiltration (↑) and small duct dilation (*). (B) The percentage of tissues of each grade at days 7 to 30 and days 30 to 120 post-immunization.

Figure 3.

 

Lymphocytic infiltration of the lacrimal glands in SS rats. (A) Lacrimal glands of normal rats (left) and rats with mild SS (center) or severe SS (right) were stained with H&E (top row) or HRP-conjugated antibodies against CD4⁺ T cells (second row), CD8⁺ T cells (third row), or OX12 (B cells; bottom row). (B) Immunohistochemical staining of CD3⁺ T cells (solid arrows, Bc) in the conjunctiva and cornea of naive (Ba) and day 24 post-immunized (Bb, Bc) Lewis rats. (Bc) Higher magnification of the inset in (Bb). CJ, conjunctiva; CN, cornea. (C) Summary of the pathologic examination of various organs in the immunized rats at day 35 post-immunization. ^aGrade I.

Figure 4.

 

Preliminary identification of the Ag in lacrimal gland extract using a proliferation assay. (A) The lacrimal gland extract was applied to a HiPrep 26/60 Sephacryl S-200 column and 10 fractions taken, based on the protein sizes. (B, C) T cells from the spleen of rats immunized with a mixed lacrimal and salivary gland extract were prepared at day 14 post-immunization and stimulated with 20 μg/mL of protein from each fraction (B) or selected tubes (C) in the presence of irradiated syngeneic antigen-presenting cells, and proliferation measured using the [³H]-thymidine incorporation assay.

Table 4.

 

Pathogenicity of Protein Fractions Prepared from Gel Filtration

Ag	Ag Dose (μg)	Incidence*	Severity, †
						Mild	Severe
Fraction I	100	0/16	0	0
Fraction VI	100	0/16	0	0
Fraction VII	100	12/16	4	8
Fraction VIII	100	0/16	0	0

*  Incidence of SS is expressed as positive/total lacrimal glands following pathological examination.

†  Severity of inflammation on histopathological examination is grouped as mild (1+ to 2+) or severe (3+ to 4+).

Figure 5.

 

Antigenicity of proteins in fraction VII. (A) Fraction I and fraction VII used for immunization were subjected to SDS-PAGE, followed by staining with Coomassie brilliant blue. M, molecular weight markers; F1, fraction I; F7, fraction VII. (B) Sera from naïve rats and rats with EAU or SS were tested for reactivity with proteins in Fraction VII by immunoblotting. W, crude extract; F7, fraction VII. The arrows in the blot indicate the 30 kDa antigen in the lacrimal gland extract specifically reactive with sera from SS rats, but not from naïve and EAU rats. (C) Identification of Klk1b22 as the immunogen for SS. Peptide mass fingerprint analysis of a 30-KDa protein matched nine peptide masses with Klk1b22. The total coverage was 51% of the protein. The peptides covered in Klk1b22 are shown in bold letters.

Figure 6.

 

Preparation and purification of Klk1b22 fusion protein and its reactivity with antibodies and T cells from SS rats. (A) Purified GST inclusion (lane 2) and GST-Klk1b22 (lane 3) were electrophoresed in SDS-PAGE gels and stained with Coomassie Blue; lane 1 is molecular weight standards. (B) A serum sample from an SS rat binds to GST-Klk1b22 (lane 2), but not GST inclusion (lane 1) on immunoblot; similar results were obtained with more than 10 samples. (C) Proliferation of T cells from rats immunized with GST-Klk1b22. Naïve rats were immunized with GST-Klk1b22, then T cells were harvested at day 10 post-immunization and cultured with medium (no) or 20 μg/mL of R16, GST, mixed gland extract, or GST-KlKb22 in the presence of irradiated antigen-presenting cells. Proliferation was measured using the [³H]-thymidine incorporation assay.

Table 5.

 

Comparison of the Disease Induced by Ag Immunization and Transfer of Ag-Reactive T Cells

	Incidence*	Onset, †	Duration	Severity, ‡
Immunization	23/30	10 ± 2	>60 d	2.0 ± 0.5
Adoptive transfer	27/30	6 ± 1	>60 d	3.0 ± 1.0

*  Incidence of SS is expressed as positive/total lacrimal glands following pathological examination.

†  Days after immunization.

‡  Mean ± SD of the pathological score.

Figure 7.

 

Autoantibodies against La-SSB in SS rats. Sera (1:100 dilution) from naïve rats or from rats immunized 14 days previously with R16 (uveitogenic Ag for EAU), mixed lacrimal and salivary gland extract, fraction VII, or GST-Klk1b22 were tested for La-SSB autoantibody titer by ELISA.

The authors thank Tom Barkas for editorial assistance and John Gamel for the histologic evaluation.