Tacrolimus interacts with voriconazole to reduce the severity of fungal keratitis by suppressing IFN-related inflammatory responses and concomitant FK506 and voriconazole treatment suppresses fungal keratitis

Isolation and culture of bone marrow–derived neutrophils, macrophages, and intraepithelial lymphocytes

Bone marrow–derived neutrophils (denoted here as neutrophils) and intraepithelial lymphocytes (denoted here as lymphocytes) were isolated from 8-week-old healthy female C57BL/6 mice, as described by Khajah and Andonegui [27] and Montufar-Solis and Klein [28]. To isolate neutrophils, marrow cells were flushed from the femur and tibia with ice-cold Roswell Park Memorial Institute (RPMI) 1640 medium (Invitrogen, Carlsbad, CA) and then rinsed and treated with erythrocyte lysis buffer to remove the red blood cells. Resuspended cells were overlaid on discontinuous Percoll density gradients (52%, 64%, and 72%) and centrifuged at 400 ×g for 30 min at room temperature, after which neutrophils were isolated from the bottom layer (64–72%). For macrophages, after the resuspended cells were acquired, wash medium was used to collect the cells in a sterile 50-ml conical centrifuge tube on ice, which was then centrifuged at 500 ×g for 10 min at room temperature. The cell pellet was resuspended in macrophage complete medium which consists of RPMI 1640 medium supplemented with 20% fetal bovine serum (FBS; cat. 12657-029; Gibco, Carlsbad CA), 30% L929-cell conditioned medium (LCCM), a source of macrophage colony stimulating factor [29], 100 U/ml penicillin, 100 µg/ml streptomycin, and 2 mM L-glutamine. Cells were seeded in Optilux Petri dishes (BD Biosciences, San Jose CA) and incubated at 37 °C in a 5% CO₂ atmosphere. After 4 days, an extra 10 ml of medium was added to each plate, and the cells were incubated for an additional 3 days. The supernatants was discarded, and the attached cells were washed with 10 ml of sterile PBS (1X; 120 mM NaCl, 20 mM KCl, 10 mM NaPO₄, 5 mM KPO₄, pH 7.4) after differentiation [30].

For lymphocyte isolation, the small intestine tissues were removed, and Peyer’s patches were excised. The tissues were then flushed of fecal material, opened longitudinally, and cut into 3- to 4-mm pieces in RPMI 1640 supplemented with fetal calf serum (10% v/v; Nutricell, Campinas, SP, Brazil) and gentamicin at 20 mg/ml (GE Healthcare Biosciences, Pittsburgh, PA). The tissue fragments were rinsed several times in Ca²⁺/Mg²⁺-free PBS, transferred to Ca²⁺/Mg²⁺-free PBS containing 5 mM EDTA (Sigma-Aldrich, St. Louis, MO) and 2 mM dithiothreitol, and shaken in a warm bath at 37 °C for 30 min. The cell suspensions were subsequently filtered through 20-ml syringe barrels containing wetted nylon wool, centrifuged, suspended in 3 ml of 40% isotonic Percoll (Sigma-Aldrich), layered on top of 70% isotonic Percoll (4 ml), and centrifuged for 30 min at 400 ×g. Lymphocytes were then recovered from the Percoll interface and washed with centrifugation in supplemented RPMI 1640 medium.

Cell culture and stimulation

The neutrophils and lymphocytes were cultured in RPMI 1640 supplemented with 10% FBS, 1% penicillin-streptomycin, and 5% 2 mM L-glutamine on a 12-well culture plate, which the macrophages culture media added one more component, 5% LCCM. Trypan Blue (Invitrogen, Carlsbad, CA) exclusion indicated that the cell viability was approximately 85%. A microscopy evaluation demonstrated that more than 85% of the isolated cells were neutrophils, macrophages, or lymphocytes. The A6(1) mouse corneal epithelial cell line was a gift from Dr. Peggy Zelenka (National Eye Institute/NIH) and was cultured following her protocol [31]. In A6 (1) mouse corneal epithelial cell line, twelve short tandem repeat (STR) loci were amplified using the commercially available EX20 Kit from Beijing Microread Genetics Co., Ltd. (Beijing, China, Appendix 1). The cell line sample was processed using the 3730xl DNA Analyzer (Applied Biosystems®). Data were analyzed using GeneMapper® ID-X v1.2 software (Applied Biosystems, Beijing, China). Appropriate positive and negative controls were run and confirmed for the sample submitted. The STR analyses are presented in Appendix 1.

Macrophages, neutrophils, lymphocytes, and A6(1) cells were divided into three groups at a seeding density of 1 × 10⁶ cells/well in 12-well plates: Group I was the control group (which received PBS), group II received 10 mg/ml zymosan (Invitrogen) for 8 h, and group III received 10 mg/ml zymosan for 24 h. The concentration and administration duration of zymosan were based on a previous study [26].

Aspergillus fumigatus spore preparation

The strain of Aspergillus fumigatus used in this investigation was AS 3.0772, and it was purchased from the China General Microbiological Culture Collection Center, Beijing, China. The yeast strain was grown on Sabouraud dextrose agar (Difco, Detroit, MI) at 30 °C for 4 days. Spores were harvested and washed in sterile PBS and then diluted to a concentration of 1 × 10⁵ colony-forming units (CFUs)/ml (resulting in an A. fumigatus spore preparation, or AFSP).

Mice and murine model of Aspergillus fumigatus fungal keratitis

Six- to eight-week-old inbred female C57BL/6 mice were purchased from the Guangdong Provincial Centre for Animal Research (Guangzhou, China). The mice were housed in a standard animal facility with a controlled temperature (22–24 °C) and photoperiod (12 h:12 h light-dark cycle) and were given free access to food and water. The animal experiments complied with the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research (Guangzhou, China; approval ID: 20120308) and The Institutional Review Board of the Zhongshan Ophthalmic Center. The mice were anaesthetized intraperitoneally with xylazine (1.9 mg/ml) and ketamine (37.5 mg/ml), and every effort was made to minimize suffering. Each experiment was repeated three times.

The right eye of each mouse was punctured using a 30-gauge needle to form a tunnel in the corneal stroma. Next, a 33-gauge Hamilton syringe was inserted through the tunnel, and 2 μl of a 1 × 10⁵ conidia solution was injected into the corneal stroma [32]. The eyelids were then rubbed together for several seconds to distribute the inoculum evenly over the corneal surface. The A. fumigatus–infected mouse corneas exhibited obvious corneal edema, with stromal infiltration 1 day post-infection. The mice were examined using a slit-lamp microscope (Carl Zeiss Meditec, Dublin, CA), and those that did not meet this standard were excluded.

Mice in which fungal keratitis was successfully induced were randomly divided into four groups: group I, the PBS group (n=6/group/time); group II, the voriconazole group (n=6/group/time); group III, the FK506 group (n=6/group/time); and group IV, the voriconazole plus 0.05% FK506 group (n=6/group/time). Group I received PBS topically four times per day for 7 consecutive days, and group II received 10 µg/ml voriconazole topically four times per day for 7 consecutive days. The voriconazole powder (99.9% purity) was obtained directly from Pfizer Central Research (VFEND; Pfizer, New York, NY) and was suspended in 0.3% Noble agar to a concentration of 1 mg/ml. Group III received 0.05% FK506 solution topically four times per day for 7 consecutive days. Finally, group IV was subconjunctivally injected with 5 μl of a 0.05% FK506 solution 24 h before infection, after which the mice with infected eyes received solutions containing 0.05% FK506 and 10 µg/ml voriconazole topically four times per day for 7 consecutive days; the applications of the two eye drops were separated by 5 min. The 0.05% FK506 solution was supplied by Zhongshan Ophthalmic Center, Guangzhou, China and was prepared as described previously [33]. FK506 (Fujisawa Pharmaceutical co Ltd, Osaka, Japan) was obtained in pure powder form and reconstituted with ethanol to obtain a concentration of 5%. The solutions of Tween-80 (0.01 ml), polyethylene hydrogenated castor oil (0.03 g), and CMC (0.3 g), which were previously prepared in sterile water, respectively, were added to 1 ml of the FK506 solution. Then glycol (0.015 ml), sodium citrate (0.05 g), and merthiolate (0.002 g) were added. Adequate sterile water was replenished to reach the volume of 100 ml, which made the final concentration of FK506 to 0.05%. The suspensions were mixed vigorously by a magnetic stirrer. The formulation was a mixture of normal saline, hydrogenated castor oil (20% w/v), Tween-80 (5.5% v/v), hydroxymethyl cellulose (0.3% w/v), glycerol (8% v/v), and thiomersalate (0.002% w/v).

To determine the clinical scores on day 1, day 3, day 5, and day 7 post-infection, all infected corneas were imaged to visualize disease progression. The two observers were blinded to the test groups, and the observers scored the same image in turn and performed mutual checks. The area and depth of corneal opacity and surface regularity were evaluated with a scoring system, in which each item has a score of up to 4 to grade the severity of corneal damage. Ocular disease was graded using clinical scores ranging from 0 to 12 according to the scoring system developed by Wu and Wilhelmus [32]. A total score of 5 or lower indicated mild keratitis, a score of 6 to 9 (including 9) represented moderate keratitis, and a score of 9 to 12 indicated severe keratitis. The animals were euthanized at the indicated end points of the experiments with cervical dislocation under isoflurane anesthesia.

Real-time PCR

With a cataract knife, the cornea was perforated at the limbus, making an opening large enough to insert the tip of a pair of corneal scissors. Under an operating microscope, each whole cornea of each mouse was then carefully cut off through the corneal limbus with the corneal scissors. The total RNA in each cornea was extracted using an RNeasy Micro Kit (Qiagen Sciences, Germantown, MD) according to the manufacturer’s protocol. Additionally, the total RNA in primary neutrophils, lymphocytes, and macrophages and A6(1) cells was extracted using TRIzol reagent (Invitrogen) according to the manufacturer’s protocol. The RNA was quantified using a NanoDrop 2000C spectrophotometer (Thermo Scientific, West Palm Beach, FL) and reverse transcribed into cDNA using a kit (Fermentas, St. Leon-Rot, Germany). The primer sequences used for IFNs (IFN-α, IFN-β, and IFN-γ), inflammatory cytokines (IL-1α, IL-6, IL-12, and IL-17), and β-actin are listed in Table 1. Quantitative real-time PCR was performed using a CFX96 Real-Time PCR System (Bio-Rad, Hercules, CA) and the SYBR Green Master Mix (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions. The data were analyzed according to the comparative Ct (ΔΔCT) method and were normalized to the expression of β-actin in each sample. In the real-time PCR, the thermal cycle was programmed for 30 s at 98 °C for initial denaturation, followed by 35 cycles of 10 s at 98 °C for denaturation, 10 s at 59°C for annealing, 10 s at 72 °C for extension, and 1 min at 72 °C for the final extension. The melting curves and gel electrophoresis of the end products were obtained to confirm the specificities of the PCR reactions.

Western blotting

For corneal expression of IFNs, whole corneas (n=6/group/time) were collected and pooled from normal uninfected and infected C57BL/6 eyes at day 1, day 3, and day 5 post-infection. For detection of zymosan-infected macrophages, neutrophils, lymphocytes, and A6(1) cells, the cultured cells were washed three times with ice-cold PBS and then treated with lysis buffer. The specific steps of protein extraction and analysis of the protein levels were performed according to Chen et al.’s (2013) study [34]. Protein extraction was specifically performed using the Pierce BCA Protein Assay Kit (Thermo Scientific, West Palm Beach, FL). In total, 40 µg protein was loaded and separated by 12% (for IFN-β) or 10% (for IL-1α) sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) was accomplished, followed by transfer onto a polyvinylidene difluoride (PVDF) membrane (Invitrogen). The membrane was then blocked for 2 h with 5% (for IFN-β) or 3%(for IL-1α) bovine serum albumin (BSA) in Tris-buffered saline containing 0.1% Tween-20 and incubated with a monoclonal rat anti-mouse IFN-β antibody (Abcam, Cambridge, UK) or a polyclonal goat anti-mouse IL-1α antibody (R&D Systems, Minneapolis, MN) overnight at 4 °C. After the membrane was washed, it was incubated with an anti-rat or peroxidase-conjugated secondary antibody (Abcam, Cambridge, UK) for 2 h. Finally, the blots were detected using the Odyssey Infrared Imaging System (LI-COR Biosciences, Frederick, MD) according to the manufacturer’s protocol. For normalization, within the same membrane, β-actin was used as a loading control.

Histological examination

Eyes from different groups at each time point were fixed in 4% formaldehyde in 0.075 M phosphate buffer for 24 h, dehydrated in increasing concentrations of ethanol (70–99%), and embedded in paraffin (Merck, Darmstadt, Germany) at 60 °C. Sections 5 mm in thickness were then cut and rinsed with deionized water at 45 °C; single sections were mounted on SuperFrost Plus glass slides (Menzel-Glaser, Braunschweig, Germany). The slides were subsequently dried at 60 °C for 1 h. Hematoxylin and eosin (H&E) staining was then performed to determine the degrees of corneal edema, inflammatory cell infiltration, and corneal irregularity.

Statistical analyses

The clinical scores are reported as the mean ± standard error of the mean (SEM), and the one-way ANOVA test was used to determine the statistical significance of the differences among the control, voriconazole, and voriconazole plus FK506 groups, as well as the differences in the RNA expression levels of inflammatory cytokines in the in vivo and in vitro studies. Differences were considered statistically significant at a p value of less than 0.05.

Zymosan increased type I and II IFN expression levels in neutrophils, lymphocytes, macrophages, and A6(1) cells

To test the potential of zymosan, a component of the fungal cell wall, in stimulating the expression of IFN transcripts, we stimulated neutrophils (Figure 1A), lymphocytes (Figure 1B), macrophages (Figure 1C), and A6(1) cells (Figure 1D) with zymosan. The mRNA expression levels of IFN-α, IFN-β, and IFN-γ were highest at 8 h after stimulation (all p<0.05) and then decreased rapidly at 24 h compared with the levels at 8 h (all p<0.05), regardless of cell type. The protein levels of one of the IFNs, IFN-β, were also tested in neutrophils, lymphocytes, macrophages, and A6(1) cells after stimulation with zymosan. The western blot data showed that IFN-β expression was statistically significantly increased at 8 h and then gradually decreased up to 24 h after infection in these four cell types (p<0.05; Figure 1E,F). We concluded that zymosan stimulation increased type I and II IFN RNA and protein expression levels in neutrophils, lymphocytes, and macrophages.

Zymosan enhanced inflammatory responses in neutrophils, lymphocytes, macrophages, and A6(1) cells

To ascertain the in vitro function of zymosan in neutrophils, lymphocytes, macrophages, and A6(1) cells, the levels of selected inflammatory cytokines downstream of type I and II IFNs were evaluated with real-time PCR and western blotting. In neutrophils, the mRNA expression level of the inflammatory cytokine IL-1α increased dramatically, by approximately 200-fold, at 8 h after zymosan stimulation (p<0.001) and then decreased slightly at 24 h (p<0.05) compared with the level in the unstimulated control (Figure 2A). The mRNA levels of IL-6, IL-12, and IL-17 were also increased at 8 h by approximately 10-, 35-, and 55-fold (all p<0.001), respectively, and then were reduced to 5-, 18- and 25-fold, respectively, their prestimulation levels compared with the levels in the unstimulated control at 24 h (all p<0.01; Figure 2A). In lymphocytes, zymosan enhanced the mRNA expression level of IL-1α by approximately 2,000-fold at 8 h (p<0.001) and by 1,000-fold at 24 h (p<0.001; Figure 2B). The mRNA expression levels of IL-6, IL-12, and IL-17 were elevated in a similar manner as in the neutrophils at 8 h and 24 h (Figure 2B). In macrophages, the mRNA expression levels of all of the inflammatory cytokines (IL-1α, IL-6, IL-12, and IL-17; Figure 2C) showed changes similar to those observed in the neutrophils and lymphocytes. Finally, in A6(1) cells, the mRNA expression levels of all inflammatory cytokines peaked at 8 h and then gradually and statistically significantly decreased up to 24 h (Figure 2D). Additionally, the pro- and mature protein levels of IL-1α were checked in the neutrophils, lymphocytes, macrophages, and A6(1) cells, and the proform and the mature form were elevated increasingly to 8 h and then decreased up to 24 h after infection with zymosan statistically (p<0.05), in which the expression level of the proform was higher than that of the mature form (Figure 2E,F).

Type I and II IFN expression levels were elevated in the fungal keratitis mouse model

To determine the expression levels of type I and II IFNs, a fungal keratitis mouse model was established, in which corneal swelling, corneal ulcers, and corneal infiltration gradually advanced from day 1 to day 5 post-infection and then decreased on day 7 (Figure 3A), after which the fungal keratitis gradually healed. The clinical score peaked on day 5, with a score of 9, compared with scores of 4 and 6 on day 1 and day 3, respectively; the clinical score then decreased to 7 on day 7 (all p<0.05; Figure 3B). The mRNA levels of type I IFNs, including IFN-α/β (Figure 3C,D), and the type II IFN IFN-γ (Figure 3E) revealed a similar trend as in the clinical score, with the levels of all three IFNs increasing in a time-dependent manner and peaking on day 5 post-infection, followed by a reduction on day 7 (all p<0.001).

FK506 and voriconazole lessened inflammatory responses in fungal keratitis

Given that the mRNA expression levels of type I and type II IFNs and downstream inflammatory cytokines were statistically significantly upregulated in vitro and in vivo during fungal infection, we suppressed fungal activity alone or in conjunction with immune response inhibition to further examine the impact of fungal infection on the expression of type I and II IFNs. In particular, FK506 and voriconazole were used to suppress immune responses and fungal activity, respectively.

The administration of voriconazole and FK506 was helpful for inhibition of immune responses, and slit-lamp images showed the slightest corneal opacity and surface regularity on day 1 to day 7 among the four groups, whereas the corneal inflammation seemed to be more severe on day 5 in the voriconazole or FK506 group. The voriconazole group and the FK506 group exhibited similar inflammatory severity, and the control group had the most serious inflammation (Figure 4A). The clinical scores (Figure 4B) were lowest in the voriconazole plus FK506 group (p<0.05), followed by the voriconazole group and the FK506 group (both p<0.05), with no statistically significant difference between the two groups (p>0.05), and the control group had the highest score (p<0.05). Furthermore, H&E staining confirmed that the corneas in the voriconazole plus FK506 group showed the mildest pathological changes, with the least inflammatory infiltration in the stroma and anterior chamber on day 5 post-infection. The control group exhibited the most serious corneal pathological changes, with the greatest inflammatory infiltration in the stroma and anterior chamber on day 5 post-infection, when the voriconazole plus FK506 group demonstrated slightly reduced corneal infiltration and relatively slight inflammation compared with the control group (Figure 4C).

FK506 and voriconazole inhibited the expression levels of type I and II IFNs in fungal keratitis

Notably, as shown in Figure 5, the levels of type I (IFN-α/β) and II (IFN-γ) IFNs gradually increased following the initial infection, peaked on day 5, and then declined on day 7. FK506 combined with voriconazole suppressed the mRNA expression levels of type I and II IFNs obviously, as did voriconazole alone and FK506 alone. The voriconazole plus FK506 group exhibited statistically significantly lower mRNA expression levels of type I and II IFNs (all p<0.05; Figure 5A–C), followed by the voriconazole group and the FK506 group, which did not statistically significantly differ for any IFN (all p>0.05). The control group had the highest mRNA expression levels of type I and II IFNs (Figure 5A–C). Furthermore, the protein levels of IFN-β were examined, and the results indicated that the control group had the highest protein expression levels of IFN-β and that the voriconazole plus FK506 group had the lowest levels (Figure 5D), while there was no obvious differences between the voriconazole group and the FK506 group shown in Figure 5E (p>0.05).

FK506 and voriconazole regulated the inflammatory responses downstream of type I and II IFNs in fungal keratitis

To further explore the inflammatory responses regulated by FK506 and voriconazole, the inflammatory cytokine mRNA levels were measured in the control, voriconazole, and voriconazole plus FK506 groups. Compared with the control and voriconazole groups, the voriconazole plus FK506 group exhibited statistically significantly decreased expression levels of inflammatory cytokines, including IL-1α (Figure 6A, p<0.05), IL-6 (Figure 6B, p<0.05), IL-12 (Figure 6C, p<0.05), and IL-17 (Figure 6D, p<0.05), at day 1, day 3, day 5, and day 7 post-infection. However, the voriconazole group and the FK506 group had no statistically significant differences in the expression levels of IL-1α, IL-6, IL-12, and IL-17. The proform and the mature form of protein expression levels of IL-1α were also explored, revealing a similar tendency as the mRNA expression levels, and the protein levels of the proform of IL-1α were higher than those of the mature form. The voriconazole plus FK506 group exhibited the lowest protein levels of IL-1α, followed by the voriconazole group and the FK506 group, and the control group had the highest protein levels of IL-1α statistically (p<0.05; Figure 6E,F).