secreting NK cells promote induction of dry eye disease - CiteSeerX

6 downloads 0 Views 1MB Size Report
USA, Richmond, VA, USA) was prepared according to the instructions of the manufacture, and ..... kinds: dendritic cells and NK cells take center stage. Nat. Rev.
Article

Interferon-␥-secreting NK cells promote induction of dry eye disease Yihe Chen,1 Sunil K. Chauhan, Daniel R. Saban, Zahra Sadrai, Andre Okanobo, and Reza Dana Schepens Eye Research Institute, Massachusetts Eye and Ear Infirmary, Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts, USA RECEIVED NOVEMBER 12, 2010; REVISED FEBRUARY 14, 2011; ACCEPTED MARCH 1, 2011. DOI: 10.1189/jlb.1110611

ABSTRACT NK cells have been increasingly reported to be an important effector in autoimmune diseases. However, nothing is known in this regard in DED, the most common eye pathology, which is characterized by sustained inflammation on the ocular surface. In the present study, we have examined the profile of NK cells on the ocular surface as well as in the draining lymphoid tissues during the development of this disease. Our data demonstrate activated NK cells during the disease-induction phase. Moreover, in vivo depletion of NK cells in mice results in reduced disease severity and diminished proinflammatory cytokines. Furthermore, we show that NK cells are also able to modulate the maturation of APCs, which is correlated with IFN-␥ from NK cells. Together, our findings provide new in vivo evidence that IFN-␥-secreting NK cells can promote induction of DED via direct target tissue damage and indirect influence on the priming phase of an adaptive immune response in secondary lymphoid tissue. J. Leukoc. Biol. 89: 965–972; 2011.

Introduction DED is one of the most common ocular disorders, affecting millions of people in the United States alone, with a prevalence nearly two times higher in women than in men [1, 2]. It has become clear that a complex immune and inflammatory process is involved in the pathogenesis of DED. Although the precise immunopathogenic mechanisms of DED remain unknown, there has been a growing body of evidence demonstrating the presence of proinflammatory cytokines IL-1, TNF-␣, IL-6, and IL-8 [3– 6], as well as T lymphocytes, especially CD4⫹ T cells, in the ocular surface tissues [7, 8]. Recent studies from our lab [9, 10] have further revealed that the generation of autoreactive Th1 and Th17 cells in the regional lymphoid tissues is critical for sustaining the ocular surface inflammation in DED.

Abbreviations: CEC⫽controlled environment chamber, CT⫽comparative threshold, DED⫽dry eye disease, EAE⫽experimental autoimmune encephalomyelitis

0741-5400/11/0089-965 © Society for Leukocyte Biology

Despite these findings, few studies to date have focused on the immune responses occurring at the very beginning (induction phase) of DED. A better understanding of this initial induction phase is essential for a more thorough understanding of the immunopathogenesis of this disease, including the subsequent disease progression and maintenance phases involving adaptive T cell responses. NK cells are large granular lymphocytes that do not express TCRs or BCRs. They have been viewed traditionally as effector cells that are capable of killing infected and transformed cells via release of cytokines and cytolytic activity without preconditioning [11]. In addition, increasing evidence has emerged regarding the important role of NK cells in the pathogenesis of autoimmune diseases. However, studies of NK cell function in vivo by NK cell depletion have resulted in contradictory findings in several animal models of autoimmune disease. Winkler-Pickett et al. [12] have shown diminished clinical disease and decreased adaptive responses in NK cell-depleted EAE mice, an animal model of multiple sclerosis. In another study using experimental autoimmune myasthenia gravis, Shi et al. [13] have reported that depletion of NK cells before priming reduces the severity of disease. Similar protective effects of NK cell depletion were also observed in animal models of type 1 diabetes [14]. In contrast, several investigators have reported NK cells to have a suppressive function in different autoimmune disease models such as lpr/lpr mice, which display a phenotype similar to patients with systemic lupus erythematosus [15], and even in EAE [16, 17] and type 1 diabetes [18] models. Here, using a well-characterized mouse model of DED [19], we provide new insights into the role of NK cells in the immunopathogenesis of DED by testing the hypothesis that early NK cell responses promote the initiation of DED through secretion of IFN-␥, as well as facilitating maturation of APCs in regional lymphoid tissue.

1. Correspondence: Schepens Eye Research Institute, 20 Staniford St., Boston, MA 02114, USA. E-mail: [email protected]

Volume 89, June 2011

Journal of Leukocyte Biology 965

MATERIALS AND METHODS

Mouse model of DED Six- to 8-week-old female C57BL/6 mice (Charles River Laboratories, Wilmington, MA, USA) were used in this study. All animal experiments were approved by the Institutional Animal Care and Use Committee and adhered to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research. DED was developed by placement of mice in a CEC with a relative humidity below 30%, airflow of 15 L/min, and a constant temperature of 21°C–23°C for up to 7 consecutive days [19]. To maximize ocular dryness, the mice in the CEC also received topical application of 1% atropine sulfate eye drops (Falcon Pharmaceuticals, Fort Worth, TX, USA), twice daily for the first 48 h, and s.c., 0.1 mL injections of 5 mg/mL scopolamine hydrobromide (SigmaAldrich, St. Louis, MO, USA), three times daily (9 AM, 1 PM, and 5 PM), on their dorsal surface for the duration of the CEC exposure [20]. Ageand sex-matched mice maintained in the standard environment were used as normal controls.

DED score Corneal fluorescein staining was used as a clinical evaluation tool for DED severity. Fluorescein (Sigma-Aldrich; 1 ␮l 2.5%) was applied into the lateral conjunctival sac of the mice, and after 3 min, corneas were examined with a slit lamp biomicroscope under cobalt blue light. Punctate staining was recorded in a masked manner with the standard National Eye Institute grading system of 0 –3 for each of the five areas of the cornea— central, superior, inferior, nasal, and temporal [21].

Real-time PCR Conjunctiva and submandibular and cervical draining LNs from mice were removed, frozen in TRIzol威 reagent (Invitrogen, Carlsbad, CA, USA), and stored at – 80°C until used. Total RNA was isolated with the RNeasy威 micro kit (Qiagen, Valencia, CA, USA), according to the manufacturer’s recommendations, and reverse-transcribed using the SuperScriptTM III kit (Invitrogen). Real-time PCR was performed using TaqMan威 Universal PCR master mix and predesigned primers for NK1.1 (Mm00824341_m1), IFN-␥ (Mm00801778_m1), TNF-␣ (Mm99999068_m1), and GAPDH (Mm99999915_g1; Applied Biosystems, Foster City, CA, USA) in an ABI Prism威 7900HT sequence detection system (Applied Biosystems). The GAPDH gene was used as an endogenous control for each reaction. The results of quantitative PCR were analyzed by the CT method, in which the target change ⫽ 2–⌬⌬CT. The results were normalized by the CT value of GAPDH, and the mean CT of relative mRNA level in the normal, untreated group or non-NK-depleted (control) DED group was used as the calibrator.

Flow cytometry analysis Single-cell suspensions were prepared from conjunctiva by collagenase digestion. Briefly, conjunctivae were removed and cut into small fragments, followed by digestion with 2 mg/mL collagenase type IV (Sigma-Aldrich) and 0.05 mg/mL DNase I (Roche, Basel, Switzerland) for 1 h at 37°C with agitation. The suspension was then triturated through a 30-gauge needle to homogenize the remaining tissue and filtered through a 70-␮m cell strainer (BD Biosciences, Bedford, MA, USA). Trypan blue exclusion assay confirmed cell viability. Cells were then double-stained with PE-conjugated anti-NK1.1 and allophycocyanin-conjugated anti-TCR-␤ (eBioscience, San Diego, CA, USA). Single-cell suspensions were prepared from draining LNs using a 70-␮m cell strainer. Cells were then quadruple-stained with the following antibodies: FITC-conjugated anti-CD11b, PE-Cy7-conjugated antiCD11c, PE-conjugated anti-I-Ab, and Alexa Fluor 647-conjugated anti-CD80 or FITC-conjugated anti-CD11b, PE-Cy7-conjugated anti-CD11c, PE-conjugated anti-I-Ab, and Alexa Fluor 647-conjugated anti-CD86 (BioLegend, San Diego, CA, USA). Intracellular IFN-␥ production by NK cells was assessed, as described previously [22], with some modifications. Briefly, 106 total LN cells in 100 ␮l medium were stimulated in 96-well flat-bottom plates with

966 Journal of Leukocyte Biology

Volume 89, June 2011

50 ng/mL PMA and 500 ng/mL ionomycin (both from Sigma-Aldrich) for 6 h at 37°C and 5% CO2 in the presence of GolgiStopTM (4 ␮l/6 mL cell culture, BD Biosciences) to inhibit cytokine secretion. The cells were stained for PE-conjugated anti-NK1.1 and allophycocyanin-conjugated antiTCR-␤. After fixed with intracellular fixation buffer, cells were permeabilized with permeabilization buffer and stained for intracellular IFN-␥. Control samples were stained with appropriate isotype-matched control antibodies. All of the antibodies and staining buffers were purchased from eBioscience. Stained cells were analyzed on an LSR II flow cytometer (BD Biosciences), and then all data were saved for subsequent analysis using Summit v4.3 software (Dako Colorado, Fort Collins, CO, USA).

Immunohistochemical staining For whole-mount immunofluorescence-conjunctival staining, freshly excised conjunctivae were washed in PBS and fixed in acetone for 15 min. For cross-sectional staining of the conjunctiva, whole eyeballs were excised, frozen in optimal cutting temperature compound, cut into 7 ␮m frozen sections, and fixed in acetone for 15 min at room temperature. The immunostaining was performed as described previously [23, 24]. The following primary antibodies were used: FITC-conjugated antiNK1.1 (eBioscience; 1:100) and purified hamster anti-CD3e (BD Biosciences; 1:200). The secondary antibody used was the Cy3-conjugated goat anti-Armenian hamster antibody (Jackson Laboratories, Bar Harbor, ME, USA; 1:800). Slides were studied under a confocal laser-scanning microscope (Leica TCS-SP5, Lasertechnik, Heidelberg, Germany).

ELISAs For protein extraction, the conjunctivae were harvested and stored in cold, sterile PBS containing protease inhibitors (Sigma-Aldrich) at –80°C until used. The samples were homogenized on ice and centrifuged. The supernatant was assayed for levels of IFN-␥ with commercial ELISA kits (eBioscience).

In vivo NK cell depletion and IFN-␥ neutralization For NK cell depletion, lyophilized rabbit antiasialo GM1 (Wako Chemicals USA, Richmond, VA, USA) was prepared according to the instructions of the manufacture, and 50 ␮l was injected to mice i.p., 2 days before and 2 days after CEC exposure. For IFN-␥ neutralization, lyophilized antimouse IFN-␥ mAb (R&D Systems, Minneapolis, MN, USA) was reconstituted in sterile saline according to the instructions of the manufacturer and then was administered i.p. to mice at a dose of 100 ␮g at Day 1 and 50 ␮g at Day 2 of CEC exposure.

Statistical analyses An unpaired, two-tailed Student’s t test was used, and differences were considered significant at P ⬍ 0.05.

RESULTS

Early infiltration of NK cells in the conjunctiva of DED mice To determine whether NK cells are involved in the pathogenesis of DED, we first examined mRNA expression of NK1.1, a relatively specific NK cell surface marker (also expressed by NKT cells), on the ocular surface during the DED induction period up to 7 days. A significant increase of NK1.1 mRNA copy number was observed in the conjunctiva during the first 2 days (Fig. 1). Therefore, the conjunctivae from mice induced for DED for 2 days were examined further for NK cells using immunohistochemical staining. In the cross-section (Fig. 2A) and conjunctival whole mount (Fig. 2B) samples, NK1.1⫹CD3e– NK cells (excluding NK1.1⫹CD3e⫹ NKT cells)

www.jleukbio.org

Chen et al. Role of NK cells in dry eye disease

tiasialo GM1 antibody 2 days before (Day –2) and 2 days after (Day 2), placing the mice into the CEC to remove NK cells for the whole experimental duration. Control mice were treated with irrelevant rabbit Ig. Corneal fluorescein staining was performed at baseline (Day 0), Day 2, and Day 5. NK cell-depleted mice exhibited significantly reduced corneal fluorescein staining scores at Days 2 and 5, as compared with non-NK-depleted mice (Fig. 4A). Next, we performed real-time PCR for two key inflammatory cytokines, IFN-␥ and TNF-␣, mRNA levels on conjunctival tissues from NK cell-depleted and non-NK-depleted mice with Figure 1. Increased conjunctival NK1.1 expression in the very early phase of DED. Relative NK1.1 mRNA levels normalized to GAPDH were determined by real-time PCR at different time-points during the 7-day CEC exposure. Data represent mean ⫾ sem of three mice at each time-point; *P ⬍ 0.05.

were located in the conjunctival stroma, and more NK cells were observed in DED compared with normal controls. Quantitative analysis of NK cells in the conjunctiva by flow cytometry showed that the frequency of NK1.1⫹TCR␤– NK cells (excluding NK1.1⫹TCR␤⫹ NKT cells) in DED mice (6.41%) was more than three times higher than that in normal mice (1.91%; Fig. 2C).

Increased IFN-␥-secreting NK cells in the draining LNs during the initiation of disease As previous evidence indicates regional draining LNs as the primary site where autoreactive T cell responses are initiated and expanded in DED [9, 10], we tested whether NK cells in the draining LNs are involved in DED pathogenesis and examined the kinetics of NK cells in the LNs from mice exposed to CEC for up to 7 days. We did not find any significant change in the frequency of NK1.1⫹TCR␤– NK cells (range: 0.41– 0.63%) in the 7-day CEC exposure duration compared with that in normal mice (Fig. 3A and B). Subsequently, we detected the functional status of NK cells in LNs via triple-staining with fluorochrome-conjugated anti-NK1.1, anti-TCR-␤, and anti-IFN-␥ antibodies. The percentage of IFN-␥-secreting NK cells showed a significant increase (twofold) during the first 3 days (with an average of 30%) but then declined to and maintained at a level similar to that in normal mice (with an average of 15%) for the remaining days (Fig. 3C and D).

NK cell depletion leads to reduced severity of DED as a result of absence of IFN-␥ To assess further the role of NK cells in the development of DED, we induced DED in the absence of NK cells by administering antiasialo GM1 antibody. We monitored NK1.1⫹TCR␤– NK cells in the peripheral blood and draining LNs 2 days and 5 days after antibody injection to determine whether the antibody treatment could deplete NK cells efficiently. In addition, we established the specificity of antiasialo GM1 antibody by confirming a preferential depletion of NK cells without markedly reducing other potentially affected cell populations, such as NKT cells, T cells, macrophages, or DCs (data not shown). Accordingly, we depleted NK cells in vivo by administering an-

www.jleukbio.org

Figure 2. Higher NK cell infiltration in the conjunctiva of mice in the induction phase of DED. NK cells were examined in the ocular surface qualitatively by immunohistochemical staining with NK1.1 and CD3e antibodies (A and B) and quantitatively, by flow cytometry analysis with NK1.1 and TCR-␤ antibodies (C). (A) Representative confocal images of eye cross-sections stained for NK1.1 (green), CD3e (red), and nucleus (blue) in normal controls (NL) and dry eye mice with 2 days induction (DED). NK cells, identified as NK1.1⫹CD3e– cells, were located in the conjunctiva only. Cor, Cornea; Conj, conjunctiva. (B) Representative confocal images of whole-mount conjunctiva showing more NK1.1⫹CD3e– NK cells (marked with arrows) present in the conjunctival stroma of DED than that of normal controls. (C) Flow cytometry analysis of conjunctiva from normal controls and DED, with 10 conjunctivae pooled in each group. NK cells are identified as the NK1.1⫹TCR␤– population. ISO, PE-conjugated mouse Ig2a ␬ allophycocyanin-conjugated Armenian hamster IgG; FSC, forward-scatter; SSC, side-scatter.

Volume 89, June 2011

Journal of Leukocyte Biology 967

Figure 3. Increased IFN-␥-secreting NK cells in draining LNs in the very early phase of DED. Regional draining LNs were harvested for frequency analysis of NK1.1⫹TCR␤– NK cells (A and B) and IFN-␥-secreting NK cells (C and D) at different time-points during the 7-day CEC exposure. (A and C) Representative experiments at Day 1 are shown. The indicated percentages of NK cells as a proportion of total LN cells and IFN-␥⫹ NK cells as a proportion of total NK cells were measured in normal and DED mice. For intracellular IFN-␥ expression by NK cells, total LN cells were stimulated with PMA and ionomycin for 6 h, and the IFN-␥⫹ NK cells were determined by comparing with unstimulated cells. (B and D) Kinetic plots of the frequencies of NK cells and IFN-␥⫹ NK cells in draining LNs. Data are the mean ⫾ sem of three mice at each time-point; *P ⬍ 0.01; **P ⬍ 0.005.

2 days induction. Both cytokines were reduced significantly in NK cell-depleted mice (Fig. 4B), and the reduction of IFN-␥ was also confirmed by ELISA at protein level (Fig. 4C). To further explore the role of IFN-␥ in the development of DED, the mice were treated with IFN-␥-neutralizing antibody or sterile saline as controls for the first 2 days of CEC exposure. As shown in Fig. 4D, neutralization of IFN-␥ resulted in the same reduction of corneal fluorescein staining as NK cell depletion.

NK cell depletion leads to inhibition of maturation of APCs Given the important roles of APCs in the activation and differentiation of T cells in the draining LNs, we hypothesized that NK cells may positively regulate the maturation of APCs. To assess this possibility, we evaluated the frequencies of CD11b⫹ and CD11c⫹ cells, as well as the frequencies and levels of expression of MHC-II, CD80, and CD86 on CD11b⫹ and CD11c⫹ cells in the draining LNs from non-NK-depleted and NK cell-depleted mice at Day 2, which is a representative timepoint when abundant IFN-␥ is secreted by NK cells as shown in Fig. 3. No significant change in the frequencies of CD11b⫹ (range: 5.95–12.34%) or CD11c⫹ cells (range: 4.43– 8.19%) in the draining LNs was observed (data not shown). However, compared with normal mice, non-NK-depleted DED mice showed 968 Journal of Leukocyte Biology

Volume 89, June 2011

a significant increase in the frequencies of MHC-II⫹CD11b⫹, CD86⫹CD11b⫹, CD80⫹CD11c⫹, and CD86⫹CD11c⫹ cells, as well as in the expression intensities of MHC-II and CD80 on CD11b⫹ cells, and MHC-II on CD11c⫹ cells (Fig. 5A and B). In contrast, NK cell-depleted mice demonstrated no significant increase in the frequency of mature APC populations or up-regulated expression of those maturation markers on CD11b⫹ or CD11c⫹ cells (Fig. 5A and B). Meanwhile, the IFN-␥ mRNA level in draining LNs from NK cell-depleted mice was decreased dramatically compared with that from control DED mice (Fig. 5C).

DISCUSSION We have shown here that NK cells participate in the development of DED via an early and rapid response to desiccating stress. By NK cell depletion, we provide evidence that NK cells influence disease course by secretion of IFN-␥, an important inflammatory cytokine involved in DED [25, 26], and interaction with APC, a key cellular component bridging innate and adaptive immune responses. Accumulating evidence indicates that NK cells can affect the development of autoimmunity through immunoregulatory or immunopathogenic mechanisms in different animal models. However, little is known so far about whether NK cells have a role in the induction of DED, the most common ophthalmic

www.jleukbio.org

Chen et al. Role of NK cells in dry eye disease

Figure 4. Effect of NK cell depletion in reducing disease severity and levels of related inflammatory cytokines and effect of IFN-␥ neutralization in ameliorating disease. (A) Corneal fluorescein staining, a clinical measurement of DED severity, was performed in NK cell-depleted (treated with antiasialo GM1) and non-NK-depleted (treated with control rabbit Ig) mice at Days 0, 2, and 5. The staining was graded between 0 (absence) and 3 (maximal staining) for each of five areas of cornea, the sum of which was calculated as the final score. (B) Relative mRNA expressions of IFN-␥ and TNF-␣ in conjunctiva from non-NK-depleted (control) and NK cell-depleted DED mice were detected at Day 2 by real-time PCR. Results are expressed as the relative amount of mRNA normalized to GAPDH. (C) IFN-␥ protein level in conjunctiva from normal, non-NK-depleted DED (Non-NK-dep. DE), and NK cell-depleted DED (NK-dep. DE) mice was measured at Day 2 by ELISA. DE, Dry eye. (D) Corneal fluorescein staining was performed in IFN-␥-blocked (treated with IFN-␥-neutralizing antibody) and non-IFN-␥-blocked (treated with saline) mice at Days 0, 2, and 5. The same grading method was used. Data shown represent the mean ⫾ sem of a single experiment (three to four mice/group) out of three performed; *P ⬍ 0.05; **P ⬍ 0.01.

pathology. In the present study, we have demonstrated early infiltration of NK cells in the ocular surface, specifically in the conjunctival stroma, where T cells have been found [7]. The fact that there is an over threefold increase of the NK cell population in conjunctiva suggests that peripheral circulating NK cells are rapidly recruited and accumulated in the conjunctiva during the initial disease course. These cells are poised for an early response to desiccating stress by releasing inflammatory cytokines such as IFN-␥, supported by increased IFN-␥ level in DED (Fig. 4B and C), or by other mechanisms such as cytotoxity. Meanwhile, significantly higher numbers of IFN-␥-secreting NK cells were revealed in the regional draining LNs at the same early time-frame, indicating that NK cells are not only able to cause direct damage to the ocular surface but are also capable of influencing disease development through action on other immune cells in the secondary lymphoid compartment where autoreactive T effectors are generated [9, 10]. In addition, it is noteworthy that although the proportion of NK cells in DED conjunctiva (⬃6.4%, Fig. 2) is much higher than that in the LNs (⬃0.5%, Fig. 3), the absolute number of NK cells as well as NK1.1 transcript level in DED conjunctiva [⬃1280/conjunctiva (calculated by 2⫻104 as average total conjunctival cell number) and ⬃70 copies/106 copies of GAPDH mRNA (Fig. 1), respectively] is less than that in the LNs [⬃5000/conjunctiva (calculated by 1⫻106 as average total LN cell number) and ⬃1494 copies/106 copies of GAPDH mRNA (data not shown), respectively]. Elimination of NK cells results in a significant reduction in disease severity evidenced by corneal fluorescein staining, suggesting a disease-promoting effect of early NK cell responses in the initiation of DED. To date, few markers or genes have been described that are absolutely NK cell-spe-

www.jleukbio.org

cific (many overlap with subsets of T cells or NKT cells), which has made NK cells difficult to target by knockout technology or by specific antibody depletions [27]. One concern is that asialo GM1 is not restricted to NK cells, and it is present on some cytotoxic T cells as well [28]; therefore, we also examined the effect of this antibody on different cell populations in adult mice but found no marked decrease in T, NKT, CD11b⫹, or CD11c⫹ cells. In addition, PK136 anti-NK1.1 antibody, the other most popular antibody for the study of NK cell function in vivo, cannot be used in this study, as it can deplete NK and NKT cells, and NKT cells are also able to rapidly produce high levels of cytokines, including IFN-␥, upon stimulation [29], which would further confound the interpretation of our results. With the attenuation of the disease, decreased proinflammatory cytokine levels have been observed on the ocular surface in NK cell-depleted mice. The reduction of IFN-␥ and TNF-␣ can be related to depletion of NK cells, which are able to produce both cytokines; however, this does not exclude the possibility that the dysfunction of other cells involved in DED has been corrected to some extent by NK cell depletion. For example, the decline of TNF-␣ may be attributed in part to the reduced damage of corneal and conjunctival epithelia [30] by NK cells, which may be confirmed by further studies. In addition, neutralization of IFN-␥ during the induction of DED has led to reduced disease as well, which suggests that the abundant IFN-␥ from early activated NK cells is critical to the pathogenesis of DED. The finding that significantly higher numbers of activated NK cells are present in the draining LNs provides evidence for the supporting roles of NK cells in DED, not only in the target organ but also in the secondary lymphoid tissues, Volume 89, June 2011

Journal of Leukocyte Biology 969

Figure 5. Inhibition of maturation of APCs in draining LNs of NK cell-depleted DED mice. (A and B) Draining LNs were evaluated for the expression of MHC-II, CD80, and CD86 on CD11b⫹ (A) and CD11c⫹ (B) cells at Day 2 by flow cytometry analysis. Representative histograms are shown on the left. The indicated frequency (above gating bars) and mean fluorescence intensity (MFI; below gating bars) were measured in normal, non-NK-depleted DED, and NK cell-depleted DED mice. Frequency and intensity of each maturation marker expression in CD11b⫹ and CD11c⫹ cell populations were calculated, respectively, and shown in bar graphs on the right. Data represent mean ⫾ sem of three to four mice. (C) Relative mRNA expressions of IFN-␥ in draining LNs from non-NK-depleted and NK cell-depleted DED mice were detected at Day 2 by real-time PCR. Results are expressed as the relative amount of mRNA normalized to GAPDH. Data shown represent the mean ⫾ sem of a single experiment (three to four mice/group) out of three performed; *P ⬍ 0.05; **P ⬍ 0.01.

970 Journal of Leukocyte Biology

Volume 89, June 2011

www.jleukbio.org

Chen et al. Role of NK cells in dry eye disease

where NK cells can act as mediators between innate and adaptive immunity. Of particular interest in this regard is the capacity of NK cells to modulate the function of APCs. Evaluation of changes in APC subpopulations, CD11b⫹ and CD11c⫹ cells, in NK cell-depleted DED mice by our studies has shown that acquisition of MHC-II, CD80, and CD86 on APCs is inhibited, which suggests that NK cells promote maturation of APCs in the pathophysiological conditions of DED. This could be a mechanism by which NK cells promote induction of DED, as maturation of APCs is critical to the priming of naı¨ve T cells in the primary immune response [31]. Current data have demonstrated activated T cells, especially Th17 cells, in the draining LNs of DED mice and that these cells are resistant to regulatory T cellmediated suppression [10]. Furthermore, Th17 cells have also been established as the dominant pathogenic effectors in DED by recent in vivo IL-17 neutralization studies [6, 10]. Although the precise mechanisms of the generation of autoreactive T cells in DED are still not clear, it has been proposed that stimulation signals from mature APCs are necessary for the differentiation and proliferation of Th1 and Th17 cells in draining LNs [32]. The increased levels of mature APCs in the draining LNs of non-NK-depleted DED mice observed in the present study provide additional support for this hypothesis. It has been documented in vitro that NK cells can eliminate immature DCs (CD11c⫹ subpopulation) by cytotoxic lysis or induce the maturation of DCs under appropriate conditions where IFN-␥ is required [33]. However, no obvious difference in the frequency of DCs is observed in our animal model, suggesting that induction of maturation, rather than immature DC elimination, is the main effect of NK cells on DCs. In addition, in our model, activated NK cells in draining LNs secrete significantly more IFN-␥, and this is corroborated by our finding that depletion of NK cells confers a significant reduction in LN IFN-␥ levels. Taken together, these data clearly support a critical role for IFN-␥ in the maturation of APCs. Furthermore, the ratio of NK cells to DCs is thought to be one of the factors that determines the outcome of their interaction: the low ratio of NK cells to DCs (1:5) favors DC maturation, whereas the high ratio (5:1) results in DC elimination [34]. This is consistent with our animal model, in which, the ratio of NK cells to DCs in draining LNs is ⬃1:10 (⬃0.5% of NK cells vs. ⬃5% of DCs). Collectively, the current data indicate that NK cells influence the outcome of DED immunopathology by secretion of IFN-␥ and promotion of APC maturation. Further exploration of the effect of early NK cell responses on later development of adaptive T cell-mediated immune responses in DED will help to understand further the immunopathogensis of this common ocular disease.

AUTHORSHIP Y.C. designed the study, performed most experiments, analyzed data, and wrote the manuscript; S.K.C. contributed to

www.jleukbio.org

the study design, data analysis, and manuscript writing; D.R.S. helped in flow cytometry experiments and data analysis; Z.S. helped in immunohistochemistry experiments; A.O. helped in creation of the disease animal model; and R.D. assisted in study design, data analysis, and manuscript writing.

ACKNOWLEDGMENTS This study was supported by National Institutes of Health EY019098 and EY20889 and the Research to Prevent Blindness Lew R. Wasserman Merit Award (R.D.).

REFERENCES

1. Schaumberg, D. A., Sullivan, D. A., Buring, J. E., Dana, M. R. (2003) Prevalence of dry eye syndrome among US women. Am. J. Ophthalmol. 136, 318 –326. 2. Schaumberg, D. A., Dana, R., Buring, J. E., Sullivan, D. A. (2009) Prevalence of dry eye disease among US men: estimates from the Physicians' Health Studies. Arch. Ophthalmol. 127, 763–768. 3. Solomon, A., Dursun, D., Liu, Z., Xie, Y., Macri, A., Pflugfelder, S. C. (2001) Pro- and anti-inflammatory forms of interleukin-1 in the tear fluid and conjunctiva of patients with dry-eye disease. Invest. Ophthalmol. Vis. Sci. 42, 2283–2292. 4. Massingale, M. L., Li, X., Vallabhajosyula, M., Chen, D., Wei, Y., Asbell, P. A. (2009) Analysis of inflammatory cytokines in the tears of dry eye patients. Cornea 28, 1023–1027. 5. Enrı´quez-de-Salamanca, A., Castellanos, E., Stern, M. E., Ferna´ndez, I., Carren ˜ o, E., Garcı´a-Va´zquez, C., Herreras, J. M., Calonge, M. (2010) Tear cytokine and chemokine analysis and clinical correlations in evaporativetype dry eye disease. Mol. Vis. 16, 862– 873. 6. De Paiva, C. S., Chotikavanich, S., Pangelinan, S. B., Pitcher III, J. D., Fang, B., Zheng, X., Ma, P., Farley, W. J., Siemasko, K. F., Niederkorn, J. Y., Stern, M. E., Li, D. Q., Pflugfelder, S. C. (2009) IL-17 disrupts corneal barrier following desiccating stress. Mucosal Immunol. 2, 243– 253. 7. Stern, M. E., Gao, J., Schwalb, T. A., Ngo, M., Tieu, D. D., Chan, C. C., Reis, B. L., Whitcup, S. M., Thompson, D., Smith, J. A. (2002) Conjunctival T cell subpopulations in Sjo¨gren’s and non-Sjo¨gren’s patients with dry eye. Invest. Ophthalmol. Vis. Sci. 43, 2609 –2614. 8. Niederkorn, J. Y., Stern, M. E., Pflugfelder, S. C., De Paiva, C. S., Corrales, R. M., Gao, J., Siemasko, K. (2006) Desiccating stress induces T cell-mediated Sjo¨gren’s syndrome-like lacrimal keratoconjunctivitis. J. Immunol. 176, 3950 –3957. 9. El Annan, J., Chauhan, S. K., Ecoiffier, T., Zhang, Q., Saban, D. R., Dana, R. (2009) Characterization of effector T cells in dry eye disease. Invest. Ophthalmol. Vis. Sci. 50, 3802–3807. 10. Chauhan, S. K., El Annan, J., Ecoiffier, T., Goyal, S., Zhang, Q., Saban, D. R., Dana, R. (2009) Autoimmunity in dry eye is due to resistance of Th17 to Treg suppression. J. Immunol. 182, 1247–1252. 11. Andoniou, C. E., Coudert, J. D., Degli-Esposti, M. A. (2008) Killers and beyond: NK-cell-mediated control of immune responses. Eur. J. Immunol. 38, 2938 –2942. 12. Winkler-Pickett, R., Young, H. A., Cherry, J. M., Diehl, J., Wine, J., Back, T., Bere, W. E., Mason, A. T., Ortaldo, J. R. (2008) In vivo regulation of experimental autoimmune encephalomyelitis by NK cells: alteration of primary adaptive responses. J. Immunol. 180, 4495– 4506. 13. Shi, F. D., Wang, H. B., Li, H., Hong, S., Taniguchi, M., Link, H., Van Kaer, L., Ljunggren, H. G. (2000) Natural killer cells determine the outcome of B cell-mediated autoimmunity. Nat. Immunol. 1, 245–251. 14. Poirot, L., Benoist, C., Mathis, D. (2004) Natural killer cells distinguish innocuous and destructive forms of pancreatic islet autoimmunity. Proc. Natl. Acad. Sci. USA 101, 8102– 8107. 15. Takeda, K., Dennert, G. (1993) The development of autoimmunity in C57BL/6 lpr mice correlates with the disappearance of natural killer type 1-positive cells: evidence for their suppressive action on bone marrow stem cell proliferation, B cell immunoglobulin secretion, and autoimmune symptoms. J. Exp. Med. 177, 155–164. 16. Zhang, B., Yamamura, T., Kondo, T., Fujiwara, M., Tabira, T. (1997) Regulation of experimental autoimmune encephalomyelitis by natural killer (NK) cells. J. Exp. Med. 186, 1677–1687. 17. Xu, W., Fazekas, G., Hara, H., Tabira, T. (2005) Mechanism of natural killer (NK) cell regulatory role in experimental autoimmune encephalomyelitis. J. Neuroimmunol. 163, 24 –30. 18. Lee, I. F., Qin, H., Trudeau, J., Dutz, J., Tan, R. (2004) Regulation of autoimmune diabetes by complete Freund's adjuvant is mediated by NK cells. J. Immunol. 172, 937–942.

Volume 89, June 2011

Journal of Leukocyte Biology 971

19. Barabino, S., Shen, L., Chen, L., Rashid, S., Rolando, M., Dana, M. R. (2005) The controlled-environment chamber: a new mouse model of dry eye. Invest. Ophthalmol. Vis. Sci. 46, 2766 –2771. 20. Goyal, S., Chauhan, S. K., Zhang, Q., Dana, R. (2009) Amelioration of murine dry eye disease by topical antagonist to chemokine receptor 2. Arch. Ophthalmol. 127, 882– 887. 21. Lemp, M. A. (1995) Report of the National Eye Institute/Industry workshop on clinical trials in dry eyes. CLAO J. 21, 221–232. 22. Shi, Y., Ling, B., Zhou, Y., Gao, T., Feng, D., Xiao, M., Feng, L. (2007) Interferon-␥ expression in natural killer cells and natural killer T cells is suppressed in early pregnancy. Cell. Mol. Immunol. 4, 389 – 394. 23. Rashid, S., Jin, Y., Ecoiffier, T., Barabino, S., Schaumberg, D. A., Dana, M. R. (2008) Topical ␻-3 and ␻-6 fatty acids for treatment of dry eye. Arch. Ophthalmol. 126, 219 –225. 24. Ecoiffier, T., El Annan, J., Rashid, S., Schaumberg, D., Dana, R. (2008) Modulation of integrin ␣4␤1 (VLA-4) in dry eye disease. Arch. Ophthalmol. 126, 1695–1699. 25. De Paiva, C. S., Villarreal, A. L., Corrales, R. M., Rahman, H. T., Chang, V. Y., Farley, W. J., Stern, M. E., Niederkorn, J. Y., Li, D. Q., Pflugfelder, S. C. (2007) Dry eye-induced conjunctival epithelial squamous metaplasia is modulated by interferon-␥. Invest. Ophthalmol. Vis. Sci. 48, 2553–2560. 26. Albertsmeyer, A. C., Kakkassery, V., Spurr-Michaud, S., Beeks, O., Gipson, I. K. (2010) Effect of pro-inflammatory mediators on membrane-associated mucins expressed by human ocular surface epithelial cells. Exp. Eye Res. 90, 444 – 451.

972 Journal of Leukocyte Biology

Volume 89, June 2011

27. Shi, F. D., Van Kaer, L. (2006) Reciprocal regulation between natural killer cells and autoreactive T cells. Nat. Rev. Immunol. 6, 751–760. 28. Ehl, S., Nuesch, R., Tanaka, T., Myasaka, M., Hengartner, H., Zinkernagel, R. (1996) A comparison of efficacy and specificity of three NK depleting antibodies. J. Immunol. Methods 199, 149 –153. 29. Godfrey, D. I., Hammond, K. J. L., Poulton, L. D., Smyth, M. J., Baxter, A. G. (2000) NKT cells: facts, functions and fallacies. Immunol. Today 21, 573–583. 30. Luo, L., Li, D. Q., Doshi, A., Farley, W., Corrales, R. M., Pflugfelder, S. C. (2004) Experimental dry eye stimulates production of inflammatory cytokines and MMP-9 and activates MAPK signaling pathways on the ocular surface. Invest. Ophthalmol. Vis. Sci. 45, 4293– 4301. 31. Huq, S., Liu, Y., Benichou, G., Dana, M. R. (2004) Relevance of the direct pathway of sensitization in corneal transplantation is dictated by the graft bed microenvironment. J. Immunol. 173, 4464 – 4469. 32. Chauhan, S. K., Dana, R. (2009) Role of Th17 cells in the immunopathogenesis of dry eye disease. Mucosal Immunol. 2, 375–376. 33. Degli-Esposti, M. A., Smyth, M. J. (2005) Close encounters of different kinds: dendritic cells and NK cells take center stage. Nat. Rev. Immunol. 5, 112–124. 34. Piccioli, D., Sbrana, S., Melandri, E., Valiante, N. M. (2002) Contact-dependent stimulation and inhibition of dendritic cells by natural killer cells. J. Exp. Med. 195, 335–341.

KEY WORDS: immunopathogenesis 䡠 antigen-presenting cells 䡠 ocular surface

www.jleukbio.org