The Role of Langerhans Cells in Pseudomonas aeruginosa Infection

Abstract

purpose. Previous experimental studies have shown that extended-wear contact lens usage results in a centripetal migration of Langerhans cells from the conjunctiva into the central cornea. To test the consequences of this, Langerhans cells were induced into the cornea before Pseudomonas aeruginosa infection in BALB/c mice that are normally resistant (the cornea heals) and in C57BL/6 mice that are susceptible (the cornea perforates) to bacterial challenge.

methods. Mean clinical scores, slit lamp examination, adenosine diphosphatase (ADPase), and acid phosphatase staining as well as immunostaining with DEC-205, B7-1, CD4, and interleukin-2 receptor (IL-2R) antibodies and histopathologic, RT-PCR, and delayed-type hypersensitivity (DTH) analyses were used to examine the effects on bacterial disease after polystyrene bead induction of Langerhans cells into the cornea before bacterial challenge.

results. No difference in disease response was observed in bead- versus sham-treated C57BL/6 mice after bacterial infection; however, significant differences leading to corneal perforation were seen in BALB/c mice that included an increased number of Langerhans cells in the central cornea at 1 and 6 days after infection, an increased number of B7-1⁺ (mature) Langerhans cells at 6 days after infection, CD4⁺ and IL-2R⁺ T cells at 5 days after infection, enhanced DTH, and increased mRNA levels for IFN-γ in cornea and cervical lymph nodes. Alternately, levels of IL-4 were significantly higher in the cornea and cervical lymph nodes of sham- versus bead-treated animals.

conclusions. These data provide evidence that Langerhans cells are critical in the innate immune response to P. aeruginosa and provide new information regarding the mechanisms governing resistance versus susceptibility to bacterial infection with this opportunistic pathogen.

The cornea, unlike many other tissues, has no professional antigen-presenting cells that are able to take up, process, and present antigen to cells of the immune system. ¹ In the adjacent conjunctiva, however, resident cells (Langerhans cells) are present that constitutively express major histocompatibility class II antigen. Although Langerhans cells do not normally reside in the central cornea, ¹ almost any stimulus, ^{2 3 4 5} including extended-wear contact lens usage, ⁶ results in the centripetal migration of the cells from the adjacent conjunctival epithelium into the cornea. Langerhans cells in the conjunctival tissue are immature cells that have limited antigen-presenting functions, but after ocular herpetic ^{2 7} Pseudomonas aeruginosa, ⁸ or Acanthamoeba ⁹ infection, maturation and migration of these cells are enhanced. 

The consequences of Langerhans cells in the central cornea may serve to prime the cornea to respond more rapidly and severely to insults and to enhance immune responsiveness. ^{3 7 9 10 11} For example, previous studies have shown that the presence of Langerhans cells in central cornea before exposure to parasite challenge augmented antigen presentation and was beneficial in preventing development of Acanthamoeba keratitis. ⁹ In contrast, no study has similarly investigated the consequences of Langerhans cell induction into the cornea before bacterial challenge with P. aeruginosa; this was the purpose of the current study. 

Our studies provide direct evidence that the presence of Langerhans cells in the central cornea of resistant BALB/c mice before bacterial challenge resulted in significant differences in disease outcome, as reflected by increased numbers of Langerhans cells in cornea that are mature, inflammation of the stroma with a predominant macrophage accompanied by a T-cell infiltrate, changes in transcript levels of IFN-γ and IL-4, and a shift from resistance to the susceptible phenotype. 

Materials and Methods

Mice

Female BALB/c and C57BL/6 (B6) mice were purchased from the Jackson Laboratory (Bar Harbor, ME) at 8 weeks of age and maintained in a domestic animal facility. Mice were housed in accordance with the National Institutes of Health guidelines and all procedures conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Bacterial Preparation

P. aeruginosa strain 19660 was purchased from the American Type Culture Collection (ATCC, Rockville, MD). The stock culture was maintained on peptone tryptic soy broth (PTSB) ⁸ slants (PTSB solidified with 1.7% agar; Difco Laboratories, Detroit, MI) at 4°C and fresh slants prepared every 2 weeks. Cultures were grown in PTSB at 37°C on a rotary shaker at 150 rpm for 18 hours to an approximate optical density of 1.6 at 540 nm, centrifuged at 6000g for 10 minutes at 15°C, washed, and resuspended in 5 mL sterile saline to a concentration of 1.0 × 10⁸ colony-forming units (CFU)/μL. For infection, two dilutions (1/10 each) were made in sterile saline for a final concentration of 1.0 × 10⁶ CFU/μL. ⁸  

Polystyrene Beads

To induce centripetal Langerhans cell migration into the cornea of both groups of mice, sterile polystyrene microspheres (1 μm; Polysciences, Inc., Warrington, PA) were washed in 70% ethanol and Dulbecco’s (D)-PBS (Sigma, St. Louis, MO) and 7 μL applied to the wounded cornea (described later) as reported before. ¹² Mice (n = 5/group per time point) were killed (5, 7, and 10 days later) and ocular Langerhans cells stained, as described later, to determine the peak time of their migration into the central cornea. Control mice (n = 5/group per time point) were similarly sham treated with PBS only. As reported before, ¹² maximum Langerhans cell migration was seen at approximately 1 week after bead placement (data not shown). 

Infection and Ocular Response

BALB/c and B6 mice (n = 5/group per treatment) were challenged with P. aeruginosa at 7 days after bead placement. Mice were anesthetized and the left cornea scarified with a needle (25-gauge 5/8; Becton Dickinson, Rutherford, NJ), as described before. ^{13 14} A 5-μL aliquot containing a 1.0 × 10⁶-CFU/μL suspension of bacteria was applied to the wounded cornea and ocular disease evaluated daily until 7 or 10 days after infection (PI) Corneal disease was graded as described before ¹⁵ : 0, clear or slight opacity, partially or fully covering the pupil; +1, slight opacity, fully covering the anterior segment; +2, dense opacity, partially or fully covering the pupil; +3, dense opacity, covering the entire anterior segment; and +4, corneal perforation or phthisis. Mean clinical scores were calculated by summation of the scores for each group (n = 5/group per time point) of mice divided by the number of mice scored at each time point. 

Preparation, Staining, and Quantitation of Langerhans Cells

To compare the number of Langerhans cells in the cornea after ocular infection, BALB/c mice (with or without cells induced into the cornea) were challenged with viable P. aeruginosa, as described above. At PI days 1, 4, and 6, epithelia were collected (n = 5/group per time point) and stained with adenosine diphosphatase (ADP) as described. ^{6 16} Briefly, eyes were placed in 0.02 M EDTA-PBS buffer (pH 7.2) at 37°C for 3 hours. Corneal and conjunctival epithelia were separated from the stroma, fixed in cacodylate-buffered formaldehyde for 20 minutes at 4°C, and washed four times for 10 minutes each with cold 0.1 M cacodylate buffer. Epithelial sheets were incubated in ADPase substrate or in dihydroxyphenyalanine (DOPA)-oxidase (negative control)–containing ADPase buffer, 2% lead nitrate, and ADPase (5 mg/mL; Sigma) for 15 minutes at 37°C. Sheets were washed four times for 10 minutes each with Tris-maleate buffer (pH 7.2), developed for 5 minutes in a 1:10 ammonium sulfide solution, washed again three times for 10 minutes each with buffer, transferred to a glass slide, mounted in glycerol, and coverslipped. Representative areas were photographed, the magnification increased to ×200, and the Langerhans cells counted. ⁶ No positively stained cells were detected in control DOPA-oxidase–treated tissues. Experiments were repeated once similarly. Representative data typical of a single experiment are shown. 

Dual-Antibody Staining

To determine whether there was a difference in the maturation of Langerhans cells in cornea after bead versus sham treatment, monoclonal antibodies (mAbs) DEC-205 (clone NLDC-145, ATCC) specific for Langerhans cells or B7-1 (clone 16-10A1; PharMingen, San Diego, CA), for B7-1 costimulatory molecule were used. A rat anti-human HLA-DR5 (clone SFR3-DR5, IgG2b; ATCC) served as a nonspecific control mAb, as reported before. ⁸ Epithelia from BALB/c mice (n = 5/group per treatment) were processed at PI days 4 and 6 by fixing in 2% paraformaldehyde-0.1 M cacodylate buffer at 4°C for 15 minutes. Sheets were washed four times at 4°C in 0.1 M cacodylate buffer, for 10 minutes each, and nonspecific binding was blocked with 0.1 M PBS containing 1% BSA (Sigma). Sheets were incubated in DEC-205 mAb (5.6 mg/mL) at 4°C overnight, washed in PBS-BSA, and incubated for 1 hour at 37°C in a 1:100 dilution of FITC-conjugated goat anti-rat IgG (Jackson ImmunoResearch, West Grove, PA). For B7-1 staining, the same sheets were washed for 15 minutes in PBS-BSA, placed in a blocking solution containing a 1:10 dilution of rat IgG (Jackson ImmunoResearch) for 1 hour, and incubated for 2 hours at 37°C with biotin-conjugated B7-1 mAb (1:25). Three PBS-BSA washes (10 minutes each) followed before incubation in streptavidin-conjugated rhodamine (3 μg/mL; Jackson ImmunoResearch) at 37°C for 1 hour. Sheets were mounted on glass slides in a modified mounting medium containing 10 mg p-phenylenediamine, 0.01 M PBS, and glycerol, to prolong fluorescence. Representative areas were similarly photographed under a microscope (Axiophot; Carl Zeiss, Thornwood, NY), identical FITC- and rhodamine-stained fields digitized with a digital camera (SPOT; Diagnostic Instruments, Sterling Heights, MI) and images overlaid using image management software (MetaMorph; Universal Imaging Corp., West Chester, PA). The overlaid images were printed, and dual-labeled cells were quantitated on at least eight fields (each field at ×200) for each group of mice, as described, and the number of positive cells expressed as the mean ± SEM. The experiment was repeated three times. Representative data from a single experiment are shown. 

Histopathology

Eyes were enucleated (n = 3/group per time point) at PI days 5 and 7, rinsed in PBS, and fixed in 1% osmium tetroxide, 2.5% glutaraldehyde, and 0.2 M Sorenson phosphate buffer (pH 7.4; 1:1:1) at 4°C for 3 hours. Specimens were dehydrated in graded ethanols and embedded in plastic, as described. ¹⁷ Thick (1.5-μm) and thin (60–75-nm) sections were cut, stained, observed, and photographed using either of two microscopes (Axiophot; Zeiss for thick sections; and transmission electron microscope model 1010; JEOL, Tokyo, Japan, for thin sections). 

Acid Phosphatase Staining

Infected eyes of bead- versus sham-treated BALB/c mice (n= 5/group) were collected at PI day 5, embedded in optimal cutting temperature (OCT) compound and snap frozen in liquid nitrogen, as described before. ^{8 17} Sections (10 μm) were collected on poly-l-lysine–coated slides (Polysciences). For staining, ¹⁸ slides were fixed in cold acetone for 5 minutes, air dried, incubated with naphthol AS-BI phosphate, washed in distilled water, and coverslipped with mounting medium (Accu-mount; Baxter Scientific, Deerfield, IL). Positively reacting cells appeared pink to red. 

Immunostaining

Infected eyes of BALB/c mice, with or without Langerhans cells induced into the cornea before infection (n = 2/group per time point), were enucleated at PI day 7. Eyes were embedded in OCT and snap frozen as described. ^{8 17} Sections were cut and collected as for histopathology and incubated for 1 hour with primary mAbs specific for CD4 (rat IgG2a, clone H129.19, 1:10), and CD25 (IL-2R, rat IgM, clone 7D4, 1:50 dilution; PharMingen). Sections were incubated with 0.3% hydrogen peroxide for 30 minutes to block endogenous peroxidase activity and for 1 hour with a biotinylated secondary antibody, anti-rat IgG2a (CD4, 1:25) or anti-rat IgM (IL-2R, 1:100) (PharMingen). Horseradish peroxidase–conjugated avidin (1:25 or 1:100; Zymed, San Francisco, CA) was incubated with the sections for 30 minutes before adding 3,3′-diaminobenzidine tetrahydrochloride (Pierce, Rockford, IL) for 10 to 15 minutes. Control sections were incubated similarly using HLA-DR5, a nonspecific mAb, as described. ⁸ The experiment was repeated once similarly, and data representative of a single experiment are shown. 

Reverse Transcription–Polymerase Chain Reaction

Infected corneas and ipsilateral draining cervical lymph nodes (CLNs) were removed from bead- and sham-treated BALB/c mice (n= 5/group) at PI day 5, frozen in liquid nitrogen and stored at− 70°C. Frozen samples were homogenized in RNA extraction agent (STAT-60; Tel-Test, Friendsville, TX), and total RNA was isolated per the manufacturer’s instruction. Total RNA (50 ng) was reversed transcribed using random primers (Gibco BRL, Grand Island, NY) and reverse transcriptase (Sensiscript Qiagen, Valencia, CA) in the presence of 10 U inhibitor (RNase; Promega, Madison, WI). Amplification of cDNA was conducted with Taq polymerase (Gibco BRL), and specific primers for IFN-γ, IL-4, and β-actin in a thermal cycler (GeneMate; ISC BioExpress, Kaysville, UT). The cycling conditions were 94°C for 45 seconds, 59°C for 30 seconds, and 72°C for 1 minute for 35 cycles, with a final extension at 72°C for 10 minutes. The primers used were 5′-TGCATCTTGGCTTTGCAGCTCTTCCTCATGGC-3′ (sense) and 5′-TGGACCTGTGGGTTGTTGACCTCAAACTTGGC-3′ (antisense) for IFN-γ, 5′-GGGGGGATTTGTTAGCATCTCTTG-3′ (sense) and 5′-CACTCTCTGTGGTGTTCTTCGTTGC-3′ (antisense) for IL-4, and 5′-GTGGGCCGCTCTAGGCACCAA-3′ (sense) and 5′-CTCTTTGATGTCACGCACGATTTC-3′ (antisense) for β-actin, which yielded amplified products of 364, 262, and 539 bp, respectively. Control RT-PCR without reverse transcriptase during RT was performed to confirm that there was no DNA contamination in the total RNA samples. Twenty microliters of final PCR products was analyzed by electrophoresis with 1.2% agarose gels stained with ethidium bromide. The bands were visualized under UV transillumination and quantitated using an image analysis system (AlphaImager 2000 Documentation & Analysis; Alpha Innotech Corp., San Leandro, CA). Integrated density values (IDVs) for the IFN-γ and IL-4 PCR products were corrected for the amount of β-actin on each sample. Representative data from one of two replicate similar experiments are shown. Data are expressed as the mean IDV of three PCR samples from five separate mice. 

Delayed-Type Hypersensitivity Assay

For this assay, BALB/c mice with or without Langerhans cells in the central cornea before bacterial challenge (n = 5/group) were infected as described earlier. At PI day 5, 2 × 10⁷ CFU of heat-killed P. aeruginosa (10 μL in 0.01 M PBS) was injected subcutaneously into the ear pinna ipsilateral to the infected eye. PBS was injected into the contralateral ear as a control. Ear thickness was measured just before injection and at 24 and 48 hours after challenge, using an engineer’s micrometer. ⁸ Delayed-type hypersensitivity (DTH) was calculated as follows: (24- or 48-hour measurement minus 0-hour measurement)_{antigen-challenged ear} minus (24- or 48-hour measurement minus 0-hour measurement)_{PBS-challenged ear}. ¹⁹ The experiment was repeated once similarly, and data representative of one experiment are shown. 

Statistical Analysis

An unpaired two-tailed Student’s t-test was used to determine the significance of the mean clinical scores, B7-1/DEC-205 dual staining of Langerhans cells, and DTH assays. A P ≤ 0.05 confidence interval was used to determine the level of significance. A nonparametric test (Mann-Whitney) also was performed on the mean clinical score data and provided similar statistical significance values at the P ≤ 0.05 confidence interval (data not shown). 

Results

Ocular Response to Infection

Maximal numbers of Langerhans cells were present in the bead- versus sham-treated corneas of both BALB/c and B6 mice at 7 days after treatment (data not shown). At this time, mice were infected and ocular disease graded to evaluate the progression and outcome of bacterial infection. Figures 1A and 1B show the mean clinical scores for B6 and BALB/c experimental groups (each mouse strain with and without bead treatment) from PI days 1 through 7 or PI day 10 day, respectively. No significant differences were seen in bead- versus sham-treated B6 mice after bacterial infection. The disease progressed from slight opacity on PI day 1 to perforation by PI day 7 in both groups, regardless of bead treatment. Because the corneas of B6 mice had all perforated by PI day 7, the experiment was terminated and the animals killed. In contrast, in bead- versus sham-treated BALB/c mice, significantly worsening disease was observed at PI days 3 through 10, with corneal perforation evident at PI days 7 through 10 (P = 0.611, P = 0.0334, P = 0.0137, P = 0.0012, P = 0.0001 at PI days 1, 3, 5, 7, and 10, respectively). Representative eyes from BALB/c mice treated with or without beads photographed with a slit lamp at PI day 10 (Fig. 2) . The cornea of a sham-treated mouse (Fig. 2A) appears near normal, whereas the cornea of a bead-treated animal is more opaque and near perforation (Fig. 2B) . 

ADPase Staining

To further examine the disparate response of BALB/c mice to infection after bead treatment, the number of Langerhans cells in the central cornea after infection was quantitated in BALB/c mice treated with or without beads at PI days 1, 4, and 6 (Fig. 3) . Although statistically significant, only a slight increase in the number of Langerhans cells was observed in the cornea of bead- versus sham-treated mice at PI days 1 and 6, whereas similar numbers of cells were detected at PI day 4 in both groups (P = 0.0001, P = 0.6788, and P = 0.0002 at PI days 1, 4, and 6, respectively). 

DEC-205 and B7-1 Staining

Because of the importance of Langerhans cell maturation in antigen presentation, the number of cells in the central cornea expressing the B7-1 maturation marker was also determined in BALB/c mice at PI days 4 and 6, in mice with or without bead treatment, by using a dual-labeling antibody technique (Table 1) . The number of dual-labeled cells expressing both the murine Langerhans cell-specific DEC-205 marker and the B7-1 marker was not significantly different at PI day 4 (P = 0.1038). However, at PI day 6, although the number of Langerhans cells stained by the DEC-205 marker was similar in the two groups, a significant increase in the number of B7-1⁺ cells was seen in the bead- versus sham-treated animals (P = 0.0001). 

Histopathology

Corneas from the bead- versus sham-treated BALB/c mice also were examined histopathologically at PI days 5 and 7, and marked differences in the disease response were documented (Fig. 4) . At PI day 5, epithelial thinning and stromal inflammation were obvious in both the bead- and sham-treated eyes. However, most striking was the predominately mononuclear rather than the routinely expected polymorphonuclear neutrophil (PMN) cell infiltrate that characterized the bead- versus sham-treated corneal stroma (Figs. 4A 4B 4C 4D) . By PI day 7, the cornea of bead-treated mice was devoid of epithelium, the stroma was almost completely disorganized, and mononuclear cells remained as the predominate cellular infiltrate. Free bacteria were scattered in the remnants of the stroma and in the anterior chamber, with corneal perforation obvious in some of the mice (Fig. 4E) . In contrast, in mice in which the eye was sham treated, the epithelium was intact, cell infiltrate (mainly PMNs) had diminished, the stroma was less edematous, stromal cells and collagen bundles as well as the endothelium were intact, and few free bacteria were observed (Fig. 4F) . To confirm the disparate nature of the cellular infiltrate, transmission electron microscopy was performed and revealed that in the bead-treated eye, the infiltrated cells were large and mononuclear when compared with smaller sized, multilobed PMNs characteristic of the stroma of the sham-treated eye (Figs. 5A 5B) . Acid phosphatase staining was used to corroborate the transmission EM data and authenticate that the mononuclear cells were macrophages. Positively stained cells in a bead- versus sham-treated eye are shown at PI day 5 (Figs. 6A 6B) . Many acid phosphatase-positive red-colored cells characterized the central corneal stroma of the bead- versus sham-treated cornea. Although no positively stained cells were found in this area in the sham-treated cornea (Fig. 6B) , a few cells were seen peripherally in the cornea and adjacent conjunctiva, respectively (data not shown). 

Immunostaining

Because only the susceptible disease response has been characterized by an infiltration of CD4⁺ T cells that are of the T-helper (Th)1 type subset, ^{13 14} we next tested the cornea of bead- versus sham-treated mice to determine whether these cells also were present in the bead-treated cornea. Figures 7A and 7B illustrates positively stained CD4⁺ T cells in the cornea of bead- but not sham-treated mice at PI day 5. The cells were activated as indicated by positive staining with IL-2R antibody (Fig. 7C) . Figure 7D shows the negative staining pattern for the anti-HLA-DR5 irrelevant antibody used as a control for the staining procedure. 

Reverse Transcription–Polymerase Chain Reaction

Because we had previously shown that a Th1 type T-cell response correlated with increased levels of IFN-γ in a susceptible mouse strain ^{8 13} but had no information regarding this cytokine in BALB/c mice, we used RT-PCR to test bead- versus sham-treated mice for mRNA transcript levels of IFN-γ in cornea and draining CLNs at PI day 5. These results are shown in Figure 8 as a representative agarose gel and a graph plotting the band IDV. Bead-treated mice had significantly elevated levels of IFN-γ mRNA transcripts in both the draining CLNs and cornea (P = 0.01 and P = 0.02, respectively). We next tested whether there were differences in the levels of an anti-inflammatory cytokine, IL-4, in the two groups (Fig. 9) . Bead-treated mice had significantly decreased mRNA transcript levels for IL-4 in both CLN and cornea when compared with sham-treated mice (P = 0.0003 and P = 0.0008, respectively) at PI day 5. We also tested the responses of the two groups to challenge injection into the ear pinna of heat-killed bacterial antigen at PI day 5 by measuring DTH. This assay was a systemic measure of T-cell responsiveness to Pseudomonas antigen challenge (Fig. 10) . DTH was measured at 24 and 48 hours after antigen challenge, and the level was significantly elevated in bead- versus sham-treated mice at both times (P = 0.015 and P = 0.004, respectively). 

Discussion

Langerhans cells are powerful antigen-presenting dendritic cells that are localized in many mucosal sites, including the eye. ^{11 20} The functional role of these important cells in bacterial, ⁸ viral, ⁷ and parasitic ⁹ infection, particularly in the eye, has lately received increased attention. In this regard, previous studies have postulated that the presence of these cells in the cornea, as has been documented in experimental extended-wear contact lens usage ⁶ could function to prime the eye to a more rapid response to insult. The present study tested the effects of Langerhans cells induced into the cornea before bacterial infection on the inflammatory response after P. aeruginosa challenge. The results provide substantive evidence that extended-wear contact lens users, when exposed to P. aeruginosa, are at greater risk of development of bacterial ulcers than non–lens wearers. Specifically, results of this study have provided evidence that, in susceptible strains of mice, such as B6, ¹³ induction of Langerhans cells into the central cornea before infection does not alter the outcome of disease—that is, corneal perforation. Because there is no change in disease response phenotype, B6 mice were not tested further in this system. In contrast, in BALB/c, a Th2-responsive mouse strain ¹⁴ that is normally resistant to P. aeruginosa challenge, ¹⁴ induction of Langerhans cells into the central cornea before infection was disastrous, converting the resistant to a susceptible phenotype. After bead application and infection in BALB/c mice, increased numbers of ADPase-positive Langerhans cells were detected in the central cornea when compared with sham-treated mice. Perhaps more importantly, a significantly greater number of Langerhans cells in the cornea of bead- versus sham-treated mice expressed B7-1. Expression of this molecule, a member of the B7 family of costimulatory molecules, has been shown to enhance the capacity of Langerhans cells to present antigen to CD4⁺ T cells. ^{21 22} In addition, using antibody neutralization and gene-knockout mice, it has recently been shown that B7-CD28 costimulation is critical in generation of the susceptible response to P. aeruginosa infection in B6 mice. ⁸  

The role of the Langerhans cell in other ocular models has been tested after induction of the cell into cornea before infection. The presence of Langerhans cells in cornea before infection with Acanthamoeba augments antigen presentation and is beneficial in the prevention and development of Acanthamoeba keratitis. ⁹ In direct contrast to the beneficial effect of Langerhans cells in the cornea, in immunopathologic diseases such as herpes keratitis, involving antigen presentation to T cells during the inductive phase of an immune response, Langerhans cells have been shown to migrate from the eye to draining CLNs, ⁷ where they present antigen to naïve CD4⁺ T cells, which then migrate back to the eye and further augment inflammation. This inflammatory response in the eye may be a double-edged sword, serving to eradicate the pathogen, but often promoting the destruction of host ocular tissue and loss of vision. 

The corneas of bead- versus sham-treated mice also were examined histopathologically. Unexpectedly, the inflammatory infiltrate was dramatically altered in bead- versus sham-treated animals. Instead of the expected PMN cell infiltrate, ^{17 23} bead-treated mice evidenced a predominately mononuclear stromal infiltrate. That the cells were mononuclear was corroborated by transmission electron microscopy, and acid phosphatase staining ¹⁸ confirmed that the cells were macrophages. 

The role of macrophages in initiation and regulation of inflammation has been studied in P. aeruginosa–induced pneumonia in mice. ²⁴ After aerosol depletion of macrophages with clodronate disodium, depleted mice showed low mortality within 24 hours after infection, but high mortality at a later time, in contrast with nondepleted mice. The results suggest that depletion of macrophages may have beneficial early effects, but late deleterious effects on lung injury and survival in cases of Pseudomonas-induced pneumonia. In contrast, in the case of infection with other Gram-negative bacteria, macrophage activation and cytokine production are a well-documented, endotoxin lipopolysaccharide (LPS)-mediated pathway. ²⁵ LPS, the major component of the outer membrane of Gram-negative bacteria is a strong modulator of immune responses and activates macrophages to synthesize a variety of inflammatory cytokines that are responsible for much of the pathologic response observed in Gram-negative sepsis. Although the current report did not specifically test the role of LPS, the latter overall paradigm agrees well with the data presented for bead- versus sham-treated BALB/c mice and suggests that excessive numbers of macrophages and their soluble mediators in cornea contribute to corneal perforation. 

In viral herpetic diseases, macrophages have been shown to play an important role in restricting the growth of herpes simplex virus (HSV)-1 after corneal infection. ¹⁸ It was concluded that the cells were required for development of an acquired immune response, presumably by functioning in antigen processing and presentation. However, the data in that report did not support the attractive hypothesis that macrophages are major participants in innate immunity to HSV. In contrast, in Acanthamoeba keratitis, depletion of macrophages with dichloromethylene diphosphonate–containing liposomes profoundly exacerbated keratitis in treated hamsters and led this group to conclude that in parasitic infection of the cornea, macrophages probably act as a first line of defense and eliminate a significant number of Acanthamoeba trophozoites. ²⁶ In contrast, it is the PMNs that have been shown to be necessary for effective clearance of P. aeruginosa from the infected cornea in mice, ²⁷ and undoubtedly this cell also may provide a source of cytokines and/or chemokines that contribute to bacterial clearance. 

Although PMNs are often regarded as the major cell type infiltrating the cornea of P. aeruginosa–infected experimental models of disease ²⁷ or patients, ^{28 29} a role for CD4⁺ T cells has been established as a susceptibility factor in Th1-responder strains of mice, ^{13 14} and their presence in cornea has been correlated with corneal perforation and upregulation of inflammatory cytokines such as IFN-γ. ^{8 13} Because of the characteristic association of CD4⁺, IL-2R⁺ (activated) Th1 type T cells in the cornea with the susceptible response in B6 mice, ^{8 13} we next asked whether in conversion of the resistant BALB/c mouse to the susceptible phenotype, we would also detect an ingress of T cells into the cornea, never seen in BALB/c mice after P. aeruginosa infection. ¹⁴ The detection of activated CD4⁺ T cells in the bead-treated cornea correlates well with our hypothesis that the presence of this cell in cornea elicits a T-cell–mediated pathogenetic response that serves to enhance corneal destruction and perforation. ¹³  

Further, to confirm that the cells were activated, we not only tested for IL-2R⁺ cells in the cornea but also used RT-PCR to test for IFN-γ, a Th1 type cytokine. IFN-γ was present and mRNA transcripts were elevated in the cornea and CLNs of bead- versus sham-treated mice. We also confirmed systemically by DTH assay that cell-mediated responses were upregulated in the bead-treated mouse. From these data, we predict that Langerhans cells may also be involved in recruitment of T cells into cornea. In support of this contention, Langerhans cells isolated from mouse epidermis were found to constitutively express mRNA for MIP-1α ^{30 31} and also to secrete MIP-1γ, a novel CC chemokine, suggesting that in skin, Langerhans cells may be active in recruitment of T cells before activation. ³²  

Because of the potential anti-inflammatory effect of IL-4, ³³ we also tested for transcript levels of this cytokine in both the cornea and draining CLNs of sham- versus bead-treated BALB/c mice. The results provide evidence that mRNA levels for IL-4 were significantly higher in the sham- versus bead-treated corneas and CLNs. We hypothesize, but have not yet tested, that this presence and amount of IL-4 balances IFN-γ levels in the sham-treated BALB/c mouse cornea and that this may be essential for the resistance phenotype. Alternately, IL-4 may stimulate phagocytosis of PMNs or potentiate PMN degranulation and respiratory burst, ³⁴ hastening bacterial clearance and deterring PMN persistence in cornea. ²³  

 Supported by National Eye Institute Grants R01-EY02986 and P30-EY04068.

 Submitted for publication June 1, 2001; revised August 22, 2001; accepted September 19, 2001.

 Commercial relationships policy: N.

 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.

 

Figure 1.

 

Mean clinical scores of ocular disease in mice with Langerhans cells induced into the cornea before infection with P. aeruginosa. (A) No significant difference in disease response was observed between bead- and sham-treated B6 mice at PI days 1, 3, 5, and 7 (P = 0.5517, P = 1.000, P= 0.1089, and P = 1.000, respectively). (B) Bead treatment of BALB/c mice resulted in a significantly more severe disease response at PI days 3, 5, 7, and 10 when compared with sham-treated mice (P = 0.611, P = 0.0334, P= 0.0137, P = 0.0012, and P = 0.0001 at PI days 1, 3, 5, 7, and 10, respectively).

Figure 2.

 

Slit lamp photomicrographs of ocular disease in sham- versus bead-treated BALB/c mice at PI day 10. Only faint opacity was seen in the sham-treated eye (A), whereas, increased opacity is evident in the eye of the bead-treated mouse (B). Magnification, ×7.

Figure 3.

 

Quantitation of Langerhans cells (LC) in bead- versus sham-treated infected BALB/c mice after ADPase staining. The number of Langerhans cells in central cornea differed significantly in bead- versus sham-treated groups at PI days 1 and 6 (P = 0.0001, P = 0.6788, P = 0.0002, at PI days 1, 4, and 6, respectively).

Table 1.

 

Dual Labeling with DEC-205 and B7-1 at PI Days 4 and 6 in BALB/c Mouse Corneas

Labeled Cells	−Beads	+Beads	Significance
PI day 4*
DEC-205	473	479
Dual	143	159	P = 0.1038
PI day 6, †
DEC-205	486	482
Dual	99	125	P = 0.0001, ‡

*  −Beads n = 8 fields; +beads n = 10 fields

^†  −Beads n = 10 fields; +beads n = 8 fields

^‡  The total number of labeled cells counted are shown. Numbers of dual-labeled cells were significantly different at PI day 6, but not at PI day 4.

Figure 4.

 

Histopathology of bead- versus sham-treated BALB/c mouse corneas at PI days 5 and 7. Inflammatory cells are shown in the stroma of bead- (A) and sham-treated (B) corneas at PI day 5. At higher magnification, large mononuclear cells predominated in the stroma of the bead-treated mouse cornea (C) in contrast to a predominantly PMN infiltrate in the cornea of sham-treated mice (D). By PI day 7, the corneal stroma of bead-treated mice was completely disorganized (E), whereas the cornea of sham-treated mice was re-epithelialized, with minimal stromal edema and inflammatory cell infiltrate (F). Magnification, (A, B, E, F) ×35; (C, D) ×275.

Figure 5.

 

Transmission electron micrographs of the central corneal stroma of bead- versus sham-treated BALB/c mice. Numerous mononuclear versus PMN cells are illustrated in the corneal stroma of (A) bead- versus (B) sham-treated mice at PI day 5. Magnification,× 850.

Figure 6.

 

Acid phosphatase staining in the cornea of bead- versus sham-treated BALB/c mice at PI day 5. Clusters of acid phosphatase–positive reddish cells were seen in the stroma of (A) bead- but not (B) sham-treated mice. Magnification,× 325.

Figure 7.

 

T-cell immunostaining in the cornea of bead- versus sham-treated BALB/c mice at PI day 5. (A) Numerous CD4⁺ T cells were localized just below the epithelium in the cornea of bead-treated mice at PI day 5. (B) In contrast, no T cells were observed in the corneas of sham-treated mice. (C) IL-2R staining showed that a small percentage of the CD4⁺ cells in corneas of bead-treated mice were activated. (D) A negative control sample stained with an isotype-matched, nonspecific anti-HLA-DR5 mAb. Magnification,× 325.

Figure 8.

 

Representative agarose gel shows IFN-γ expression in cornea (top right) and CLNs (top left) of bead- versus sham-treated BALB/c mice at PI day 5. β-Actin served as the control. In the graph plotting the band IDV (bottom), data are expressed as the mean IDV ± SEM of three PCR samples from five separate mice in each group. P = 0.01 and 0.02 in CLN and cornea, respectively.

Figure 9.

 

Representative agarose gel shows IL-4 expression in cornea (top right) and CLN (top left) of bead- versus sham-treated BALB/c mice at PI day 5. β-Actin served as the control. Data are expressed as the mean IDV ± SEM of three PCR samples from five separate mice in each group. P = 0.0003 and 0.0008 in CLN and cornea, respectively.

Figure 10.

 

P. aeruginosa–specific DTH in bead- versus sham-treated BALB/c mice. When challenged with heat-killed bacterial antigen at PI day 6, bead- versus sham-treated mice had a significantly greater response at 24 and 48 hours after challenge (P = 0.015 and P = 0.004, respectively).