Neurturin-Deficient Mice Develop Dry Eye and ... - Semantic Scholar

Neurturin-Deficient Mice Develop Dry Eye and Keratoconjunctivitis Sicca Xiu Jun Song,1,2 De-Quan Li,1 William Farley,1 Li Hui Luo,1 Robert O. Heuckeroth,3 Jeffrey Milbrandt,4 and Stephen C. Pflugfelder1 PURPOSE. Neurturin has been identified as a neurotrophic factor for parasympathetic neurons. Neurturin-deficient (NRTN⫺/⫺) mice have defective parasympathetic innervation of their lacrimal glands. This study was conducted to evaluate tear function and ocular surface phenotype in NRTN⫺/⫺ mice. METHODS. Determined by tail genomic DNA PCR, 25 NRTN⫺/⫺ mice and 17 neurturin-normal (NRTN⫹/⫹) mice aged 6 weeks to 4 months were evaluated. Aqueous tear production, tear fluorescein clearance and corneal sensation were serially measured. Corneal permeability to AlexaFluor dextran (AFD; Molecular Probes, Eugene, OR) was measured by a fluorometric assay at 485 nm excitation and 530 nm emission. Histology was evaluated in PAS-stained sections. Mucin and HLA class II (IA) antigen were assessed by immunofluorescent staining. Tear IL-1␤ was measured by ELISA, and tear matrix metalloproteinase (MMP)-9 by zymography. Gene expression in the corneal epithelia was analyzed by semiquantitative RT-PCR. RESULTS. In comparison to that in age-matched NRTN⫹/⫹ mice, aqueous tear production, tear fluorescein clearance, and corneal sensation were significantly reduced in NRTN⫺/⫺ mice, whereas corneal permeability to AFD was significantly increased. Immunoreactive MUC-4 and -5AC mucin and goblet cell density (P ⬍ 0.001) in the conjunctiva of NRTN⫺/⫺ mice were lower than in NRTN⫹/⫹ mice. The expression of MUC-1 and -4 mRNA by the corneal epithelium was reduced in NRTN⫺/⫺ mice. There were a significantly greater number of IA antigen-positive conjunctival epithelial cells in NRTN⫺/⫺ mice than NRTN⫹/⫹ mice. Tear fluid IL-1␤ and MMP-9 concentrations and the expression of IL-1␤, TNF-␣, macrophage in-

From the 1Ocular Surface Center, Cullen Eye Institute, Department of Ophthalmology, Baylor College of Medicine, Houston, Texas; the Departments of 3Pediatrics and Molecular Biology and Pharmacology and 4Pathology and Internal Medicine, Washington University, St. Louis, Missouri; and the 2Third Hospital of the Hebei Medical University, Shijiazhuang, China. Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2002. Supported in part by National Eye Institute Grant EY11915 (SCP), an unrestricted grant from Research to Prevent Blindness, the Oshman Foundation, the William Stamps Farish Fund, National Institute of Diabetes and Digestive and Kidney (NIDDK) diseases Grant R01DK57038 (ROH), and the Washington University Digestive Diseases Research Core Center Pilot/Feasibility Program (supported by NIDDK Grant DK52574). Submitted for publication December 20, 2002; revised April 30, 2003; accepted June 2, 2003. Disclosure: X.J. Song, None; D.-Q. Li, None; W. Farley, None; L.H. Luo, None; R.O. Heuckeroth, None; J. Milbrandt, None; S.C. Pflugfelder, 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. Corresponding author: Stephen C. Pflugfelder, Ocular Surface Center, Cullen Eye Institute, Department of Ophthalmology, Baylor College of Medicine, 6565 Fannin Street, NC-205, Houston, TX 77030; [email protected]. Investigative Ophthalmology & Visual Science, October 2003, Vol. 44, No. 10 Copyright © Association for Research in Vision and Ophthalmology

flammatory protein (MIP)-2, cytokine-induced neutrophil chemoattractant (KC), and MMP-9 mRNA by the corneal epithelia were significantly increased in NRTN⫺/⫺ mice, compared with NRTN⫹/⫹ mice. CONCLUSIONS. Neurturin-deficient mice show phenotypic changes and ocular surface inflammation that mimic human keratoconjunctivitis sicca. This model supports the importance of a functional ocular surface-central nervous system-lacrimal gland sensory-autonomic neural network in maintaining ocular surface health and homeostasis. (Invest Ophthalmol Vis Sci. 2003;44:4223– 4229) DOI:10.1167/iovs.02-1319

D

ry eye is a common condition that affects 10% of the population between the ages of 30 and 60 years, increasing in prevalence to 15% of the population aged 65 years or more.1,2 Dry eye results from dysfunction of the integrated ocular surface-secretory glandular functional unit. This may result from disease of the sensory afferent nerves innervating the ocular surface, the autonomic efferent nerves innervating the tear-secreting glands, or the tear-secreting glands themselves. Dysfunction of the integrated functional unit leads to an unstable tear film, ocular surface inflammation, and epithelial disease, termed keratoconjunctivitis sicca (KCS).3–5 The pathogenesis of the tear film instability and KCS in dry eye is not well understood. A dry eye animal model with dysfunction of the integrated functional unit mimicking human dry eye disease would be a useful tool for investigating these mechanisms. Animal models of dry eye disease have been created by various means, including surgical removal of the tear-producing glands, ocular surface desiccation by mechanically inhibiting blinking, and pharmacologic inhibition of tear secretion.6 –9 We have reported that inhibiting tear secretion in mice by systemic administration of the muscarinic cholinergic antagonist scopolamine produces KCS that mimics human dry eye disease.9 This model is labor intensive and difficult to continue for prolonged periods; however, it suggests that mice with parasympathetic autonomic nerve dysfunction due to disease or gene deletion may be relevant models for the study of dry eye. Neurturin is a member of the transforming growth factor-␤ family that functions as a neurotrophic factor for several classes of neurons.10,11 Neurturin acts through a two-component receptor system consisting of the ligand-specific GFR␣-2 receptor and the common receptor tyrosine kinase c-Ret.12 Neurturin is essential for the development of specific postganglionic parasympathetic neurons. It has also been shown to support the development and maintenance of cutaneous trigeminal sensory nerves. Neurturin-deficient (NRTN⫺/⫺) mice generated by homologous recombination are viable and fertile, but have defects in their autonomic, enteric, and sensory nervous systems.13 Parasympathetic innervation of the lacrimal gland and submandibular salivary glands is dramatically reduced in NRTN⫺/⫺ mice, as is the number of GFRa2-expressing neurons in the trigeminal ganglion.11,13 The purpose of this study was to evaluate tear function and ocular surface phenotype in NRTN⫺/⫺ mice. Markers of human KCS (corneal epithelial permeability, conjunctival goblet cell density, and ocular sur4223

4224

Song et al.

IOVS, October 2003, Vol. 44, No. 10 surface, beginning 15 mm away from the cornea. The distance in millimeters where a blink or withdrawal was observed was recorded.

Corneal Permeability to AlexaFluor Dextran

FIGURE 1. Genotype determination by PCR of mouse tail genomic DNA by using specific neurturin gene primer pairs. The PCR products were analyzed by 1.5% agarose gel electrophoresis. A 330-bp band was obtained from wild-type mice (⫹/⫹), a 200 bp band from neurturin gene knockout mice (⫺/⫺), and both bands from the heterozygotes (⫹/⫺).

face inflammation) in neurturin-deficient and wild-type mice were investigated.

Materials and Methods Animals Neurturin gene– deficient mice were generated by homologous recombination.13 All studies were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Twenty-five neurturin gene knockout mice (NRTN⫺/⫺) and 17 wildtype control mice (NRTN⫹/⫹) were used in the study. Genotypes of 1-month-old pups from heterozygous parents were tested by using genomic DNA isolated from their tails (Genomic DNA Isolation kit; Sigma-Aldrich, St. Louis, MO). PCR amplification was performed on a thermal cycler (DNA Thermal Cycler 480; GeneAmp PCR kit; Applied Biosystems, Foster City, CA) with the neurturin gene primer pair 4743: CCGACGCGGTGGAGCTTCGAGAACTT and 4742: AAGGACACCTCGTCCTCATAGGCCGT, resulting in a 330-bp fragment from wildtype mice (NRTN⫹/⫹), and the gene primer pair 4743 and 4744: GAGATCAGCAGCCTCTGTTCCACATAC yielding a 200-bp fragment from homozygotes (NRTN⫺/⫺). Both fragments were obtained from the heterozygotes (NRTN⫹/⫺; Fig. 1).

Aqueous Tear Production Aqueous tear production was measured with a phenol-red–impregnated cotton thread (Zone-Quick, Oasis, CA), which was placed in the lateral canthus for 20 seconds. Wetting of the thread was measured on a millimeter scale.

Fluorescein Clearance Test The fluorescein clearance test is a measure of total tear production, tear spread, and tear drainage.14 This test was performed by instilling 1 ␮L 2% sodium fluorescein into the conjunctival sac. After 15 minutes, 1 ␮L of phosphate-buffered saline (PBS) was instilled, and the fluorescein-stained tear fluid was collected atraumatically from the lateral tear meniscus for 20 seconds under a surgical microscope by using a porous 1 ⫻ 8-mm polyester rod (America Filtrona Co., Richmond, VA). The rod was placed into a 10-␮L micropipette tip within a 1.5-mL tube, which was then centrifuged for 5 minutes after addition of 99 ␮L PBS directly onto the polyester rod. The solution was transferred to a single well in a 96-well plate. The fluorescein concentration was measured with a fluorophotometer (CytoFluor II; Perseptive Biosystems, Framingham, MA), using 485-nm excitation and 530-nm emission filters.

Measurement of Corneal Sensation A stream of CO2 gas at a pressure of 1 psi through a 1-mm diameter plastic catheter tip was delivered perpendicularly to the corneas of mice. The tip of the catheter was slowly advanced toward the corneal

The corneal uptake of 10-kDa AlexaFluor dextran (AFD; Molecular Probes, Eugene, OR) was measured by using a modification of a previously reported technique.9 Briefly, 1 ␮L of 0.3% AFD was instilled onto the ocular surface 15 minutes before death. Excised corneas were rinsed four times with 200 ␮L balanced salt solution (BSS; Alcon Lab, Inc., Fort Worth, TX) and placed in 200 ␮L BSS. The solution containing the corneal tissue was protected from light and placed on an orbital shaker. The concentration of eluted AFD was measured with 485-nm excitation and 530-nm emission filters at 10, 20, and 60 minutes on a fluorophotometer (CytoFluor II; Perseptive Biosystems).

Histology and Immunofluorescent Staining The whole eyeball together with the eyelids and conjunctiva was embedded in a mixture of 75% (vol/vol) OCT compound (Sakura Finetek USA, Inc., Torrance, CA) and 25% (vol/vol) aqueous mounting medium (Immu-Mount; Thermo-Shandon, Pittsburgh, PA), and then flash frozen in liquid nitrogen. Sections (10 ␮m thick) were cut and stained with periodic acid-Schiff (PAS) reagent. For immunofluorescent staining, sections were fixed with 100% methanol at 4°C for 10 minutes and blocked with 5% normal goat serum in PBS for 30 minutes. The primary antibody was applied for 1 hour at room temperature. These goat polyclonal antibodies were reactive with MUC-4 ASGP2 C-terminal peptide (C-pep; a gift from Kermit Carraway, University of Miami, Miami, FL), MUC-5AC (a gift from Marcia Jumblatt, University of Louisville, Louisville, KY) or HLA class II antigen (IA; Pharmingen, San Diego, CA). After a wash with PBS, the secondary antibody (AlexaFluor-488 conjugate; 1:100 dilution; Molecular Probes) was applied for 1 hour in a dark incubation chamber. After a wash with PBS, antifade medium (Gel-Mount; Fisher, Atlanta, GA) containing 1 ␮g/mL Hoechst 33342 dye and a coverslip were applied. Sections were examined and photographed with an epifluorescence microscope (Eclipse 400; Nikon, Tokyo, Japan) and a digital camera (model DMX 1200; Nikon).

Tear Collection PBS (1.5 ␮L) containing 0.1% bovine serum albumin (BSA) was instilled into the conjunctival sac. The tear fluid and buffer were collected with a 1-␮L volume glass capillary tube (Drummond Scientific Co., Broomhall, PA) by capillary action from the tear meniscus in the lateral canthus. The tear solution was stored at ⫺80°C until zymography and ELISA were performed.

IL-1␤ ELISA and Gelatin Zymography The IL-1␤ concentration in tear samples was assayed with ELISA (Quantikine M Murine ELISA kit; R&D Systems, Minneapolis, MN), according to the manufacturer’s protocol. The level of gelatinolytic enzymes in the tear fluid was measured by SDS-PAGE gelatin zymography, according to a previously reported method.15 A 2-␮L tear sample (from both eyes of each mouse) was added to SDS-PAGE sample buffer and fractionated by electrophoresis on an 8% polyacrylamide gel containing gelatin (0.5 mg/mL). The gels were soaked in 0.25% Triton X-100 for 30 minutes at room temperature to remove the SDS and incubated in a digestion buffer containing 50 mM Tris-HCl, 150 mM NaCl, 10 mM CaCl2, 2 ␮M ZnSO4, 0.01% Brij-35, and 5 mM phenylmethylsulfonyl fluoride (PMSF), a serine protease inhibitor, at 37°C overnight, to allow proteinase digestion of its substrate. The gels were rinsed in distilled water and stained with 0.25% Coomassie brilliant blue R-250 in 40% isopropanol for 2 hours and destained with 7% acetic acid.

RNA Isolation and Semiquantitative RT-PCR Total RNA from the corneal epithelium was isolated by the acid guanidium thiocyanate-phenol-chloroform extraction method,16 and

KCS in Neurturin-Deficient Mice

IOVS, October 2003, Vol. 44, No. 10

4225

TABLE 1. Mouse Primer Sequences Used for RT-PCR Gene IL-1␤ TNF-␣ MIP-2 KC MMP-9 MUC-1 MUC-4 GAPDH

GenBank Accession M15131 M11731 X53798 J04596 NM_013599 NM_013605 AF218265 M32599

Sense Primer

Antisense Primer

PCR Product (bp)

TGAGCTGAAAGCTCTCCACC TCAGCCTCTTCTCATTCCTG AGTGAACTGCGCTGTCAATGC GCTGGGATTCACCTCAAGAACA ATCCAGTTTGGTGTCGCGGAGC CTACTACCAAGAACTGAAGAG CCATTCATCACTGGAAGGTC GCCAAGGTCATCCATGACAAC

CTGATGTACCAGTTGGGGAA TGAAGAGAACCTGGGAGTAG TCCTTTCCAGGTCAGTTAGC CCTTCTACTAGCACAGTGGTTG GAAGGGGAAGACGCACAGCT CTCATAGGATGGTAGGTGTCC AGCAGCCGCAGAAGTGAAAGA GTCCACCACCCTGTTGCTGTA

297 333 211 455 552 423 525 498

stored at ⫺80°C before use. The gene expression was analyzed by reverse transcription–polymerase chain reaction (RT-PCR)16 with a housekeeping gene, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), as an internal control. In brief, first-strand cDNAs were synthesized from 0.5 ␮g of total RNA with murine leukemia virus (MuLV) reverse transcriptase. PCR amplification of the first-strand cDNAs was performed with specific primer pairs for murine cDNA of matrix metalloproteinase (MMP)-9, IL-1␤, TNF-␣, macrophage inflammatory protein (MIP)-2, cytokine-induced neutrophil chemoattractant (KC), MUC-1, MUC-5AC, or GAPDH (Table 1). Semiquantitative RT-PCR was achieved by terminating reactions at intervals of 24, 28, 32, 36, and 40 cycles for each primer pair to ensure that the PCR products formed were within the linear portion of the amplification curve.

Statistical Analysis Depending on the normality of the data distribution, the t-test or Mann-Whitney test was used for statistical comparison of assay results between groups.

Corneal Sensation Corneal sensation was assessed in both groups of 10-week-old mice. The distance of the CO2 stream from the mouse cornea that triggered a blink reflex was significantly less in NRTN⫺/⫺ mice (mean ⫽ 2.40 ⫾ 1.13 cm; n ⫽ 34 eyes) than in NRTN⫹/⫹ mice (mean ⫽ 7.62 ⫾ 2.0 cm, n ⫽ 28 eyes; P ⬍ 0.001; Mann-Whitney test).

Corneal Epithelial Permeability to AFD Decreased corneal epithelial barrier function is a key feature of human KCS. Corneal epithelial barrier function was assessed by measuring corneal permeability to AFD. The corneas of 10-week-old NRTN⫺/⫺ mice were more permeable to AFD (36.11 ⫾ 5.1 U, n ⫽ 9) than age related NRTN⫹/⫹ mice (22.6 ⫾ 4.1 U, n ⫽ 9; P ⬍ 0.05 by the Mann-Whitney test). This finding suggests that the corneal epithelial barrier function is altered in NRTN⫺/⫺ mice.

Histology and Epithelial Mucin Expression

RESULTS Aqueous Tear Production and Clearance Aqueous tear production was reduced in NRTN⫺/⫺ mice (n ⫽ 25) compared with age-matched NRTN⫹/⫹ mice (n ⫽ 17). The difference did not reach statistical significance at 6 weeks (P ⫽ 0.126), but did at 8 weeks and in older mice (P ⬍ 0.05– 0.01, Fig. 2). Similarly, NRTN⫺/⫺ mice from 8 weeks forward had significantly delayed tear fluorescein clearance compared with age-matched NRTN⫹/⫹ mice (P ⬍ 0.001, Fig. 3).

FIGURE 2. Comparison of the tear production between wild-type (⫹/⫹) and neurturin gene knockout (⫺/⫺) mice at different ages (6 –20 weeks). Data show mean and SD (error bars). *P ⬍ 0.05, ** ⬍ 0.01; Mann-Whitney test.

The corneal epithelium of 10-week-old NRTN⫺/⫺ mice was noted to be markedly thickened, approximately eight epithelial cell layers thick compared with five epithelial cell layers in wild-type mice (Fig. 4, top). The corneal epithelial cells in NRTN⫺/⫺ mice had a more basal cell phenotype and often contained PAS-positive granules. RT-PCR showed that the expression of MUC-1 and -4 mRNA by the corneal epithelium was lower in NRTN⫺/⫺ than in NRTN⫹/⫹ mice (Fig. 5). Differences in mucin expression were observed in the conjunctiva. Immunofluorescent staining for the cell membrane mucin MUC-4 was much less in the bulbar and tarsal conjunc-

FIGURE 3. Comparison of tear fluorescein clearance between wildtype (⫹/⫹) and neurturin gene knockout (⫺/⫺) mice at different ages (6 –20 weeks). Data show mean and SD (error bars) fluorescence units. **P ⬍ 0.001; Mann-Whitney test.

4226

Song et al.

FIGURE 4. Representative PAS-stained sections for corneal and conjunctival histology of NRTN⫹/⫹ and NTRN⫺/⫺ mice. K, cornea, BCj, bulbar conjunctiva, TCj, tarsal conjunctiva.

tiva of NRTN⫺/⫺ mice than in NRTN⫹/⫹ mice (Fig. 6). Conjunctival goblet cell density was measured over 100-mm lengths on the tarsal and bulbar conjunctiva (Fig. 4, bottom). The number of goblet cells in the bulbar (0.68 ⫾ 0.25) and tarsal (10.25 ⫾ 2.3) conjunctiva of NRTN⫺/⫺ mice was significantly lower than in the bulbar (10.5 ⫾ 3.7) and tarsal (51.6 ⫾ 2.1) conjunctiva of NRTN⫹/⫹ mice (n ⫽ 3, P ⬍ 0.001 for both sites; Mann-Whitney test). This was supported by reduced expression of the goblet cell mucin MUC-5AC by the NRTN⫺/⫺ mice (Fig. 7). Leukocytes were observed in the epithelium and surface of the tarsal conjunctiva in many sections of NRTN⫺/⫺ mice, but rarely in the NRTN⫹/⫹ mice. These findings suggest that NRTN⫺/⫺ mice undergo conjunctival phenotypic changes similar to those in human KCS.


FIGURE 6. Upper panel: representative immunofluorescent staining for MUC-4 in the conjunctiva of NRTN⫹/⫹ and NRTN⫺/⫺ mice. Lower panel: conjunctival sections of respective mouse strains counterstained with Hoechst 33342 dye. Abbreviations are as in Fig 4.

IA Expression in the Conjunctiva Aberrant HLA class II (IA) antigen expression by the conjunctival epithelium has been reported to increase in human KCS. Scattered stromal cells and epithelial dendritic cells in the conjunctiva of NTRN⫹/⫹ mice were immunofluorescent for IA antigen. In contrast, numerous IA-positive conjunctival epithelial cells (Fig. 8, arrow) were observed in NTRN⫺/⫺ mice. This finding indicates there is immune activation of the conjunctival epithelium in NRTN⫺/⫺ mice.

Tear Fluid IL-1␤ ELISA and Gelatin Zymography Elevated concentrations of the proinflammatory cytokine IL-1␤ and the protease MMP-9 have been detected in the tear fluid of humans with KCS. IL-1␤ and MMP-9 were detected in tear fluid of wild-type and NRTN⫺/⫺ mice by ELISA and gelatin zymography, respectively. Compared with that in NRTN⫹/⫹ mice, the IL-1␤ concentration in tear fluid of NRTN⫺/⫺ mice was slightly

FIGURE 5. Representative semiquantitative RT-PCR showing decreased expression of (A) MUC-1 and (B) MUC-4 mRNA by corneal epithelia in NRTN⫺/⫺ mice, compared with age-matched NRTN⫹/⫹ mice, with (C) GAPDH mRNA as an internal control.

FIGURE 7. Upper panel: representative immunofluorescent staining for MUC-5AC in the conjunctiva of NRTN⫹/⫹ and NRTN⫺/⫺ mice. Lower panel: conjunctival sections of respective mouse strains counterstained with Hoechst 33342 dye. Abbreviations are as in Figure 4.



4227

FIGURE 10. A representative zymogram showing MMP-9 protein in the tear fluid of NRTN⫹/⫹ and NRTN⫺/⫺ mice aged 8 to 12 weeks.

FIGURE 8. Upper panel: representative immunofluorescent staining for IA in the conjunctiva of NRTN⫹/⫹ and NRTN⫺/⫺ mice. Lower panel: conjunctival sections of respective mouse strains counterstained with Hoechst 33342 dye. Abbreviations as in Figure 4.

higher in 2-month-old mice (44.74 ⫾ 9.94 pg/mL vs. 57.79 ⫾ 14.93 pg/mL, n ⫽ 10, P ⬎ 0.05), and was dramatically increased in 4-month-old mice (29.83 ⫾ 16.16 pg/mL vs. 189.57 ⫾ 67.10 pg/mL, n ⫽ 10, P ⬍ 0.05 by the t-test, Fig. 9). Murine MMP-9 (105 kDa digestion band) was observed in the tear fluid of 88% (22/25) of NRTN⫺/⫺ mice aged 8 to 12 weeks, compared with a weak MMP-9 band in the tear fluid of 29.4% (5/17) of NRTN⫹/⫹ mice at similar ages (Fig. 10).

mRNA Expression of IL-1␤, TNF-␣, MIP-2, KC, and MMP-9 by the Corneal Epithelium The expression of RNA encoding the inflammatory cytokines IL-1␤ and TNF-␣, the chemokines MIP-2 and KC, and the protease MMP-9 was evaluated in the corneal epithelia of wildtype and NRTN⫺/⫺ mice, according to the results of semiquantitative RT-PCR of pooled total RNA from six to eight corneal

FIGURE 9. IL-1␤ concentration in tear fluid of NRTN⫺/⫺ and NRTN⫹/⫹ mice aged 2 (n ⫽ 10) and 4 (n ⫽ 10) months and all ages (n ⫽ 20) by ELISA. Data are the mean ⫾ SD.

epithelia in each group of mice aged 12 to 16 weeks. The corneal epithelia of NRTN⫹/⫹ mice showed a low level of expression of MMP-9 mRNA, whereas IL-1␤, TNF-␣, MIP-2, and KC mRNA were barely detectable. The corneal epithelium of NRTN⫺/⫺ mice expressed higher levels of MMP-9, IL-1␤, TNF-␣, MIP-2, and KC mRNAs than did wild-type mice (Fig. 11). These findings indicate that the expression of these inflammatory cytokines and chemokines was dramatically stimulated in the corneal epithelia of the NRTN⫺/⫺ mice.

DISCUSSION In this study we report a naturally occurring and permanent dry eye model in genetically manipulated neurturin-deficient mice. These mice show significantly decreased aqueous tear production and tear fluorescein clearance. Accompanying altered tear function are altered corneal epithelial barrier function, reduced mucin production, and ocular surface inflammation, all phenotypic changes that are observed in human KCS. Based on these findings, we believe that neurturin-deficient mice provide an exciting novel model for the study of the pathogenesis and natural history of dry eye disease. It is recognized that the ocular surface, the tear secreting glands, the central nervous system and their interconnecting reflex neural pathways function as an integrated functional unit.17,18 Tear flow is engendered through stimulation of ocular surface, eyelid, and nasal mucosal trigeminal sensory afferent nerves.19 –21 These nerves synapse in the pons with efferent parasympathetic nerves that innervate the lacrimal glands.22–24 Disease or damage of the afferent or efferent arms of the integrated ocular surface lacrimal functional unit are common causes of dry eye in humans. Herpes virus infections and surgical amputation of afferent trigeminal nerves in the

FIGURE 11. Representative semiquantitative RT-PCR showing increased expression of IL-1␤, TNF-␣, MIP-2, KC, and MMP-9 mRNA by corneal epithelia in NRTN⫺/⫺ mice, compared with age-matched NRTN⫹/⫹ mice, with GAPDH mRNA as an internal control.

4228

Song et al.

cornea cause dry eye.21 Efferent parasympathetic dysfunction has been observed in patients with rheumatoid arthritis who have secondary Sjo ¨ gren syndrome and may be due to circulating autoantibodies to the glandular M3 muscarinic acetylcholine receptors.25,26 Neurturin-deficient mice have defective autonomic nerves, including defective lacrimal gland innervation.13 Immunostaining studies revealed that parasympathetics nerve fibers that were readily identifiable within the lacrimal glands of wild-type mouse were almost completely absent in the glands of NRTN⫺/⫺ mice.13 Reduced tear production as a result of defective lacrimal gland autonomic innervation is therefore a likely cause of the aqueous tear deficiency that develops in NRTN⫺/⫺ mice. Another potential cause is reduced afferent stimulation of tear production and tear clearance.27,28 Compared with wild-type mice, NRTN⫺/⫺ mice were observed to have significantly reduced corneal sensitivity to a stream of CO2 gas that has been reported to stimulate polymodal sensory receptors in the cornea.29 These defects in neural pathways connecting the lacrimal functional unit make neurturin-deficient mice a physiologically relevant animal model of dry eye. Dry eye produces ocular irritation and ocular surface disease, termed KCS, which results in blurred and fluctuating vision and an increased risk of sight-threatening corneal infection and ulceration.5,30,31 The histologic features of KCS are abnormal proliferation and differentiation of the ocular surface epithelium with decreased density of conjunctival goblet cells and decreased production of mucus by the ocular surface epithelium.32,33 These cellular changes are accompanied by altered corneal epithelial barrier function that is detected clinically as increased corneal uptake of fluorescein dye.34,35 The corneal epithelium of NRTN⫺/⫺ mice was observed to be thicker than that in NRTN⫹/⫹ mice and it showed intracellular accumulation of PAS-positive glycoproteins rather than the homogeneous staining pattern in the apical cells of wild-type mice. These findings suggest that there is abnormal differentiation of the corneal epithelium in these mice. Accompanying these histologic findings was a significantly greater corneal epithelial permeability to 10-kDa AFD in NRTN⫺/⫺ mice than in NRTN⫹/⫹ mice, indicating altered corneal epithelial barrier function. An abnormal PAS staining pattern and dramatically reduced conjunctival goblet cell density were observed on the tarsal and bulbar conjunctiva of NRTN⫺/⫺ mice, compared with NRTN⫹/⫹ mice. Cytological studies have shown that decreased conjunctival goblet cell density is a hallmark of KCS in humans.32,36 Conjunctival goblet cells are the main source of the gel-forming mucin MUC-5AC, which lubricates and protects the ocular surface.37–39 Our results showed that immunodetectable MUC-4 and MUC-5AC in the conjunctival epithelium (Figs. 6, 7) and MUC-1 and -4 mRNA in the corneal epithelium (Fig. 5) were markedly reduced in NRTN⫺/⫺ mice. These findings indicate that the NRTN⫺/⫺ mouse exhibits many of the histologic changes that occur in human KCS. There is increasing recognition that ocular surface inflammation may play a critical role in the pathogenesis of KCS. Increased expression of several inflammatory mediators has been identified on the ocular surface of eyes with human KCS. These include increased concentrations of proinflammatory cytokines in the conjunctival epithelium and tear fluid18,40 – 43 and increased concentration and activity of proteases, such as plasmin and MMP-9 in the tear fluid.41,44 The proinflammatory cytokines IL-1 and TNF-␣ are important mediators of inflammation and immunity.45,46 IL-1 is a potent inducer of other inflammatory cytokines, such as IL-6 and TNF-␣, and of chemokines, such as IL-8.47 In mice, IL-1 and TNF-␣ induce the key chemoattractants, KC and MIP-2.48,49 MIP-2 is a homologue of human IL-8 that promotes leukocyte recruitment.50 –52 IL-1 and TNF-␣ also stimulate the production of MMP enzymes by epi-

IOVS, October 2003, Vol. 44, No. 10 thelial and inflammatory cells.42,53 Our results showed that the concentrations of IL-1␤ and MMP-9 proteins in tear fluid were significantly increased, and the mRNA expression of IL-1␤, TNF-␣, and MMP-9, as well as chemokines MIP-2 and KC, by the corneal epithelia was dramatically upregulated in NRTN⫺/⫺ mice, compared with NRNT⫹/⫹ mice. The increase in these soluble inflammatory mediators was accompanied by increased leukocyte infiltration of the conjunctival epithelium in NRTN⫺/⫺ mice. These findings indicate that the increased inflammatory mediators in our mouse model parallel those in human dry eye. In conclusion, neurturin-deficient mice exhibit a phenotype of ocular surface disease and inflammation that mimics human KCS. This model supports the importance of a functional ocular surface-central nervous system-lacrimal gland sensory-autonomic neural network for maintenance of ocular surface health and homeostasis. This model may be a useful tool for studying the mechanism of human dry eye disease.

References 1. Schein OD, Munoz B, Tielsch JM, Bandeen-Roche K, West S. Prevalence of dry eye among the elderly. Am J Ophthalmol. 1997;124:723–728. 2. Hikichi T, Yoshida A, Fukui Y. Prevalence of dry eye in Japanese eye centers. Graefes Arch Clin Exp Ophthalmol. 1995;233:555– 558. 3. Bron AJ, Mengher LS. The ocular surface in keratoconjunctivitis sicca. Eye. 1989;3:428 – 437. 4. Fox RI, Robinson C, Curd J. Sjo ¨ gren’s syndrome: proposed criteria for classification. Arthritis Rheum. 1994;29:577–583. 5. Lemp MA. Evaluation and differential diagnosis of keratoconjunctivitis sicca. J Rheumatol Suppl. 2000;61:11–14. 6. Fujihara T, Nagano T, Nakamura M, Shirasawa E. Lactoferrin suppresses loss of corneal epithelial integrity in a rabbit short-term dry eye model. J Ocul Pharmacol Ther. 1998;14:99 –107. 7. Fujihara T, Murakami T, Fujita H, Nakamura M, Nakata K. Improvement of corneal barrier function by the P2Y(2) agonist INS365 in a rat dry eye model. Invest Ophthalmol Vis Sci. 2001;42:96 –100. 8. Burgalassi S, Panichi L, Chetoni P, Saettone MF, Boldrini E. Development of a simple dry eye model in the albino rabbit and evaluation of some tear substitutes. Ophthalmic Res. 1999;31:229 –235. 9. Dursun D, Wang M, Monroy D, et al. A mouse model of keratoconjunctivitis sicca. Invest Ophthalmol Vis Sci. 2002;43:632– 638. 10. Golden JP, DeMaro JA, Osborne PA, Milbrandt J, Johnson EM Jr. Expression of neurturin, GDNF, and GDNF family-receptor mRNA in the developing and mature mouse. Exp Neurol. 1999;158:504 – 528. 11. Rossi J, Luukko K, Poteryaev D, et al. Retarded growth and deficits in the enteric and parasympathetic nervous system in mice lacking GFR alpha2, a functional neurturin receptor. Neuron. 1999;22: 243–252. 12. Pezeshki G, Franke B, Engele J. Evidence for a ligand-specific signaling through GFRalpha-1, but not GFRalpha-2, in the absence of Ret. J Neurosci Res. 2001;66:390 –395. 13. Heuckeroth RO, Enomoto H, Grider JR, et al. Gene targeting reveals a critical role for neurturin in the development and maintenance of enteric, sensory, and parasympathetic neurons. Neuron. 1999;22:253–263. 14. Afonso AA, Monroy D, Stern ME, Feuer WJ, Tseng SC, Pflugfelder SC. Correlation of tear fluorescein clearance and Schirmer test scores with ocular irritation symptoms. Ophthalmology. 1999; 106:803– 810. 15. Sobrin L, Liu Z, Monroy DC, et al. Regulation of MMP-9 activity in human tear fluid and corneal epithelial culture supernatant. Invest Ophthalmol Vis Sci. 2000;41:1703–1709. 16. Li DQ, Tseng SC. Three patterns of cytokine expression potentially involved in epithelial-fibroblast interactions of human ocular surface. J Cell Physiol. 1995;163:61–79. 17. Mathers WD. Why the eye becomes dry: a cornea and lacrimal gland feedback model. CLAO J. 2000;26:159 –165.

IOVS, October 2003, Vol. 44, No. 10 18. Stern ME, Beuerman RW, Fox RI, Gao J, Mircheff AK, Pflugfelder SC. The pathology of dry eye: the interaction between the ocular surface and lacrimal glands. Cornea. 1998;17:584 –589. 19. Jordan A, Baum J. Basic tear flow: does it exist? Ophthalmology. 1980;87:920 –930. 20. Pflugfelder SC. Tear fluid influence on the ocular surface. Adv Exp Med Biol. 1998;438:611– 617. 21. Gupta A, Heigle T, Pflugfelder SC. Nasolacrimal stimulation of aqueous tear production. Cornea. 1997;16:645– 648. 22. Ehinger B. Ocular and orbital vegetative nerves. Acta Physiol Scand Suppl. 1966;268:1–35. 23. Yasui T, Karita K, Izumi H, Tamai M. Correlation between vasodilation and secretion in the lacrimal gland elicited by stimulation of the cornea and facial nerve root of the cat. Invest Ophthalmol Vis Sci. 1997;38:2476 –2482. 24. Nikkinen A, Uusitalo H, Lehtosalo JI, Palkama A. Distribution of adrenergic nerves in the lacrimal glands of guinea-pig and rat. Exp Eye Res. 1985;40:751–756. 25. Barendregt PJ, van der Heijde GL, Breedveld FC, Markusse HM. Parasympathetic dysfunction in rheumatoid arthritis patients with ocular dryness. Ann Rheum Dis. 1996;55:612– 615. 26. Bacman S, Berra A, Sterin-Borda L, Borda E. Muscarinic acetylcholine receptor antibodies as a new marker of dry eye Sjo ¨ gren syndrome. Invest Ophthalmol Vis Sci. 2001;42:321–327. 27. Collins M, Seeto R, Campbell L, Ross M. Blinking and corneal sensitivity. Acta Ophthalmol (Copenh). 1989;67:525–531. 28. Toda I, Asano-Kato N, Komai-Hori Y, Tsubota K. Dry eye after laser in situ keratomileusis. Am J Ophthalmol. 2001;132:1–7. 29. Belmonte C, Acosta MC, Schmelz M, Gallar J. Measurement of corneal sensitivity to mechanical and chemical stimulation with a CO2 esthesiometer. Invest Ophthalmol Vis Sci. 1999;40:513–519. 30. Nelson JD. Diagnosis of keratoconjunctivitis sicca. Int Ophthalmol Clin. 1994;34:37–56. 31. Pflugfelder SC. Differential diagnosis of dry eye conditions. Adv Dent Res. 1996;10:9 –12. 32. Pflugfelder SC, Tseng SCG, Yoshino K, Monroy D, Felix C, Reis BL. Correlation of goblet cell density and mucosal epithelial membrane mucin expression with rose bengal staining in patients with ocular irritation. Ophthalmology. 1997;104:223–235. 33. Pflugfelder SC. Advances in the diagnosis and management of keratoconjunctivitis sicca. Curr Opin Ophthalmol. 1998;9:50 –53. 34. Gobbels M, Spitznas M. Corneal epithelial permeability of dry eyes before and after treatment with artificial tears. Ophthalmology. 1992;99:873– 878. 35. Yokoi N, Kinoshita S. Clinical evaluation of corneal epithelial barrier function with the slit-lamp fluorophotometer. Cornea. 1995;14:485– 489. 36. Nelson JD, Wright JC. Conjunctival goblet cell densities in ocular surface disorders. Arch Ophthalmol. 1984;102:1049 –1051. 37. Gipson IK, Inatomi T. Cellular origin of mucins of the ocular surface tear film. Adv Exp Med Biol. 1998;438:221–227.


4229

38. Inatomi T, Spurr-Michaud S, Tisdale AS, Zhan Q, Feldman ST, Gipson IK. Expression of secretory mucin genes by human conjunctival epithelia. Invest Ophthalmol Vis Sci. 1996;37:1684 – 1692. 39. Pflugfelder SC, Solomon A, Stern ME. The diagnosis and management of dry eye: a twenty-five-year review. Cornea. 2000;19:644 – 649. 40. Barton K, Nava A, Monroy DC, Pflugfelder SC. Cytokines and tear function in ocular surface disease. Adv Exp Med Biol. 1998;438: 461– 469. 41. Afonso AA, Sobrin L, Monroy DC, Selzer M, Lokeshwar B, Pflugfelder SC. Tear fluid gelatinase B activity correlates with IL-1␣ concentration and fluorescein clearance in ocular rosacea. Invest Ophthalmol Vis Sci. 1999;40:2506 –2512. 42. Solomon A, Dursun D, Liu Z, Xie Y, Macri A, Pflugfelder SC. Proand anti-inflammatory forms of interleukin-1 in the tear fluid and conjunctiva of patients with dry-eye disease. Invest Ophthalmol Vis Sci. 2001;42:2283–2292. 43. Tishler M, Yaron I, Geyer O, Shirazi I, Naftaliev E, Yaron M. Elevated tear interleukin-6 levels in patients with Sjogren syndrome. Ophthalmology. 1998;105:2327–2329. 44. Virtanen T, Konttinen YT, Honkanen N, Harkonen M, Tervo T. Tear fluid plasmin activity of dry eye patients with Sjo ¨ gren’s syndrome. Acta Ophthalmol Scand. 1997;75:137–141. 45. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood. 1996;87:2095–2147. 46. Dinarello CA, Wolff SM. The role of interleukin-1 in disease. N Engl J Med. 1993;328:106 –113. 47. Selvan RS, Kapadia HB, Platt JL. Complement-induced expression of chemokine genes in endothelium: regulation by IL-1-dependent and -independent mechanisms. J Immunol. 1998;161:4388 – 4395. 48. Calkins CM, Bensard DD, Shames BD, et al. IL-1 regulates in vivo C-X-C chemokine induction and neutrophil sequestration following endotoxemia. J Endotoxin Res. 2002;8:59 – 67. 49. Tessier PA, Naccache PH, Clark-Lewis I, Gladue RP, Neote KS, McColl SR. Chemokine networks in vivo: involvement of C-X-C and C-C chemokines in neutrophil extravasation in vivo in response to TNF-alpha. J Immunol. 1997;159:3595–3602. 50. Xue ML, Thakur A, Willcox M. Gene expression of pro-inflammatory cytokines and chemokines in mouse eye infected with Pseudomonas aeruginosa. Clin Exp Ophthalmol. 2002;30:196 – 199. 51. Kernacki KA, Barrett RP, Hobden JA, Hazlett LD. Macrophage inflammatory protein-2 is a mediator of polymorphonuclear neutrophil influx in ocular bacterial infection. J Immunol. 2000;164: 1037–1045. 52. Yan XT, Tumpey TM, Kunkel SL, Oakes JE, Lausch RN. Role of MIP-2 in neutrophil migration and tissue injury in the herpes simplex virus-1-infected cornea. Invest Ophthalmol Vis Sci. 1998; 39:1854 –1862. 53. Li DQ, Lokeshwar BL, Solomon A, Monroy D, Ji Z, Pflugfelder SC. Regulation of MMP-9 production by human corneal epithelial cells. Exp Eye Res. 2001;73:449 – 459.