Pretreatment With TLR2 and TLR4 Ligand Modulates Innate Immunity in Corneal Fibroblasts Challenged With Aspergillus fumigatus

Abstract

Purpose.: To study the innate immunity in telomerase-immortalized human stroma fibroblasts (THSFs) challenged with Aspergillus fumigatus hyphae after copretreatment with TLR2 and TLR4 ligand.

Methods.: THSFs were pretreated with different concentrations of zymosan and/or lipopolysaccharide (LPS) at different time periods, and challenged with high-dose Aspergillus fumigatus hyphae. The gene expression and protein secretion of inflammatory cytokines (TNF-α, IL-6, and IL-8) were detected by RT-PCR and ELISA. The effects of stimulation or pretreatment of TLR ligands on proliferation of THSFs were measured by MTT analysis.

Results.: In the certain concentration range, pretreatment of THSFs with zymosan suppressed gene expression of inflammatory cytokines (TNF-α and IL-6). Copretreatment with zymosan and LPS suppressed gene expression and protein secretion more strongly compared with pretreatment with zymosan or LPS alone. Zymosan and/or LPS pretreatment suppressed lethal effect of A. fumigatus to THSFs in a certain period.

Conclusions.: Pretreatment of THSFs with TLR2-specific ligand zymosan results in a state of A. fumigatus hyphae tolerance. Copretreatment with TLR2 and -4 ligands (zymosan and LPS) leads to a stronger state of A. fumigatus hyphae tolerance, and suppresses the lethal effect of A. fumigatus .

Chinese Abstract

Introduction

Fungal keratitis is a serious suppurative vision-threatening disease, caused by fungi after corneal epithelium damage. In recent years, there has been an obvious increase of its incidence in China, and the prognosis of the disease is poor. ¹ Corneal ulcers caused by corneal inflammation response are an important cause of blindness. Thus, clarification of the mechanism of corneal fungal infection is important for the treatment and prevention of fungal keratitis. ² The positive rate of fungal culture in the normal conjunctiva sac is 2% to 25%, which will not cause eye disease. Only when corneal epithelial cells as the first barrier to resistance of pathogenic infection are damaged, or the barrier function is abated such as by trauma or dysbacteriosis caused by long-term use of antibiotics, ³ will the pathogenic fungi invade the corneal epithelium and stroma layer. ^4–6 This suggests that corneal stromal cells as the second cell barrier of the cornea participate in the host innate immune response. ^5,7,8  

Recent studies have suggested that the Toll-like receptors (TLRs) family, as pathogen-associated molecular patterns (PAMPs), played important roles in the process of corneal innate immune response to pathogenic microorganism invasion. Previous studies in our laboratory first showed the existence of TLR1-9 in corneal epithelial cells. ^9,10 Our recent studies found that corneal TLR4-MyD88-NF-κB and TLR4-ERK1/2 pathways induced the secretion of cytokines in human corneal epithelium cell lines challenged by Acanthamoeba, ¹¹ and that TLR2, -4, and -5 are major pattern recognition receptors that defend corneal epithelial cells against fungi by induction of proinflammatory cytokines and chemokines, such as IL-1β, IL-6, and IL-10. ^11–14  

In the infected environment, the different components of fungi stimulate multiple TLR signaling pathways to play their roles, and the interactions between signal pathways have an important impact on the host inflammatory response. ¹⁵ Although TLR immune response pathways and patterns of TLRs are different, previous study has shown that different TLRs have impacts on each other in the inflammatory response. ¹⁶ For example, activation of TLR4 + TLR7 leads to a greater production of proinflammatory cytokines and chemokines than either receptor alone. ¹⁷ Wimmer et al. ¹⁸ found that activation of TLR4 or TLR9 induced tolerance to each other. 

Although there are at least 13 subtypes of the TLR family, TLR2 and TLR4 play especially important roles in recognition and respond to invading fungi. ^15,19–22 TLR2 primarily responds to zymosan and lipoprotein from fungal and bacterial cell walls, whereas TLR4 recognizes lipopolysaccharide (LPS) from the cell walls of Gram-negative bacteria. Previous studies in vitro reported that exposure of human corneal epithelial cells to Aspergillus fumigatus and Fusarium resulted in an upregulation of TLR2 and TLR4, and the release of IL-1β, IL-6, IL-8, and IL-10. ^12,13,21,22 Other studies have shown that pretreatment with a small dose of TLR2 or TLR4 specific ligand induces self-tolerance or cross-tolerance of host cells. ^16,23,24 Innate immunity of fungal keratitis is caused by stimulation of TLRs from fungal components, so study of a single ligand cannot explore the mechanism of corneal resistance to fungal infection very well. To explore the impact of TLR2 tolerance and TLR4 tolerance to innate immunity in telomerase-immortalized human stroma fibroblasts (THSFs) challenged with A. fumigatus hyphae, we copretreated the cells with zymosan and LPS, and subsequently detected proinflammatory cytokines and chemokines. 

Materials and Methods

Cell Line

The THSFs were provided by Professor Fu-Shin Yu (Wayne State University, Detroit, Michigan). 

Culture of THSFs

THSFs were maintained in Dulbecco's modified Eagle's medium (DMEM; Invitrogen Life Technologies, Carlsbad, CA) with 10% fetal bovine serum (Gibco, Carlsbad, CA) in a humidified 5% CO₂ incubator at 37°C. The cells were seeded into six-well plates at a density of 4 × 10⁵ cells per well and cultured in normal growth medium. THSFs were cultured in serum-free DMEM for 16 hours before treatment. 

Strain and Preparation of Hyphae

The A. fumigatus strain, provided by the China Centre for Type Culture Collection, was cultured in Sabouraud fluid media for 24 hours at 37°C on a shaking table with a rotation speed of 200 rpm. Under these conditions, abundant conidia were produced. To collect hyphal fragments, a medium that contained conidia, was added to Sabouraud fluid media, at a final concentration of 10⁸ microorganisms/mL and cultured at 26°C for 18 hours on a shaking table with a rotation speed of 500 rpm. The tubes were centrifuged at 1500g for 10 minutes after incubation. The pellet, almost exclusively containing hyphae, was washed twice and subsequently resuspended in phosphate-buffered saline (PBS). The pellet was killed by heat treatment at 56°C for 60 minutes. Nonviable hyphal fragments were centrifuged, resuspended, and kept frozen at −80°C until use. 

Tolerance Induction

THSFs were pretreated with different concentrations (10,000–10 ng/mL) of the TLR2 special ligand, zymosan ( Saccharomyces cerevisiae ; Sigma Chemical Co., St. Louis, MO), for 24 hours at 37°C. The cells were washed twice with serum-free medium after incubation and then restimulated with A. fumigatus hyphae (10⁶ CFUs/mL) for 4 hours. Total RNA was extracted from THSFs for the measurement of cytokines TNF-α and IL-6 by RT-PCR. 

Our previous study showed that pretreatment of THSFs with low-dose LPS results in diminished production of cytokines IL-6 and IL-8. ⁴ So the experiment presented in the following text was designed to investigate the effect of two TLR ligands in a single pretreatment and copretreatment. THSFs were copretreated with the optimum concentration of zymosan and LPS ( Escherichia coli ; Sigma Chemical Co.) for various periods (in hours: 6, 12, 18, 24, and 30) at 37°C. The cells were washed twice with serum-free medium after incubation and then restimulated with A. fumigatus hyphae (10⁶ CFUs/mL) for 4 hours as previously described. The culture media and cells were harvested at the indicated time points for measurement of cytokines (TNF-α, IL-6, and IL-8), mRNA, and protein levels. 

Real-Time RT-PCR

THSFs were pretreated with zymosan and/or LPS, and then restimulated with A. fumigatus hyphae as described above for measurement of cytokine levels. After extracting from THSFs by using TRIzol reagent (Invitrogen, Life Technologies), total RNA was reverse transcribed using a first-strand cDNA synthesis kit according to the manufacturer's directions (Toyobo Co., Ltd., Shanghai, China). The volume of RT-PCR was 10 μL, containing 1 μL of cDNA, 5 μL of mix (SYBR Green PCR Master Mix; Toyobo Co., Ltd.), 0.5 μL of each primer and 3 μL of diethylpyrocarbonate-treated water. The real-time primers are listed in Table 1. Real-time PCR reactions were performed in triplicate on a sequence detection system (ABI Prism 7000; Life Technologies/Applied Biosystems, Inc., Foster City, CA). Samples were amplified with TNF-α, IL-6, IL-8, and GAPDH primers for determination of the initial relative quantity of cDNA in each sample, and all PCR products were normalized to that amount of GAPDH. 

Table 1

 

Primers for Real-Time Polymerase Chain Reaction

Target Gene	Forward Sequence (5′–3′)	Reverse Sequence (5′–3′)
TNF-α	ATGTGGAACTGGCAGAGGAG	ACGAGCAGGAATGAGAAGAGG
IL-6	AAAGAGGCACTGGCAGAAAACA	CAGGGGTGGTTATTGCATCT
IL-8	CACCGGAAGGAACCATCTCA	GGAAGGCTGCCAAGAGAGC
GAPDH	TGAACGGGAAGCTCACTGG	TCCACCACCCTGTTGCTGTA

ELISA Measurement of Cytokines

The concentrations of TNF-α, IL-6, and IL-8 in cell culture supernatants were determined by ELISA according to the manufacturer's instructions. The measurement was performed according to the manufacturer's instructions (KYM Co., Ltd., Beijing, China). All results of samples were normalized by comparing against a standard curve and performed in triplicate. 

Proliferation of THSFs Stimulated by TLR Ligands

THSFs were seeded into 96-well plates at a density of 1.5 × 10⁴ cells per well, and cultured in normal growth medium for 24 hours. THSFs were cultured in serum-free DMEM for 24 hours before treatment. Then the cells were cultured in 1000 ng/mL zymosan, 10 ng/mL LPS, 100 ng/mL zymosan + 10 ng/mL LPS, or 10⁶ CFUs/mL A. fumigatus for various periods (in hours: 12, 24, and 36) at 37°C. Next, the culture medium was changed to DMEM containing 20 μL of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) solution (Amresco Life Science Research, Ltd., Solon, OH), and the cells continued to be cultured for 4 hours. The medium was then changed to DMSO, and the optical density (OD) values were examined using a UV spectrophotometer. Each treatment was repeated three times and negative controls were treated with serum-free DMEM. 

Proliferation of THSFs Pretreated by TLR Ligands

THSFs were cultured as above, after which the cells were pretreated in DMEM containing TLR ligands (1000 ng/mL zymosan, 10 ng/mL LPS, or 100 ng/mL zymosan + 10 ng/mL LPS) for various periods (in hours: 12, 24, and 36) at 37°C. The cells were restimulated with A. fumigatus hyphae (10⁶ CFUs/mL) for 4 hours. Then, DMEM containing 20 μL of MTT solution was added after THSFs were washed twice with serum-free medium, and the cells were subsequently cultured for 4 hours. The medium was then exchanged for DMSO, and the OD values were examined using a UV spectrophotometer. Each treatment was repeated three times and negative controls were treated with serum-free DMEM. 

Statistical Analysis

All results were expressed as mean ± SD and were repeated in at least three independent experiments. Statistical analysis was determined by one-way ANOVA and Student's t-test using a commercial software program (SPSS, version 13.0; SPSS, Inc., Chicago, IL). Differences were considered statistically significant at P < 0.05. 

Results

Effect of LPS Pretreatment on the Production of Proinflammatory Cytokines and Chemokines Challenged by A. fumigatus in THSFs

To prove whether zymosan induces tolerance to A. fumigatus in THSFs, we retreated THSFs with different concentrations (10,000–10 ng/mL) for 12 hours and rechallenged them with A. fumigatus hyphae for 4 hours. Comparisons of TNF-α and IL-6 gene expression were analyzed among groups cultured for various zymosan concentrations (Fig. 1). 

Figure 1

 

Pretreatment with zymosan-induced tolerance of THSFs to A. fumigatus hyphae. Low expression of TNF-α (A) and IL-6 (B) was observed. The THSFs were cultured with zymosan of 10,000 to 10 ng/mL for12 hours, and a blank control group cultured with serum-free medium. The cells were washed twice, and subsequently stimulated with A. fumigatus hyphae for 4 hours. Gene expression of TNF-α and IL-6 was assayed by RT-PCR. The gene expression levels were normalized to GAPDH and presented as a relative ratio. Bars: mean ± SD of three independent experiments. *P < 0.05; **P < 0.001 compared with control group without zymosan pretreatment. ##P < 0.001 compared with group cultured without zymosan or A. fumigatus .

Our study showed that pretreatment with zymosan significantly impaired gene expression of TNF-α and IL-6 in THSFs after being challenged by A. fumigatus hyphae (Fig. 1). The cells pretreated with 1000 ng/mL zymosan for 8 hours reduced TNF-α and IL-6 secretion compared with untreated controls (61% and 53%, respectively); however, pretreatment with 10,000 mg/mL zymosan had almost no effect on inhibition of inflammatory cytokine (TNF-α and IL-6) expression. Pretreatment with zymosan can induce tolerance of THSFs to A. fumigatus , and the inhibition effect is related to the concentration in a certain range. 

Effect of Copretreatment With Zymosan and LPS on the Production of Proinflammatory Cytokines and Chemokines Challenged With A. fumigatus in THSFs

We copretreated THSFs with zymosan (1000 or 100 ng/mL) and LPS (10 ng/mL) for various periods (in hours: 6, 12, 18, or 24), and subsequently stimulated them with A. fumigatus hyphae for 4 hours to determine whether copretreatment with zymosan and LPS induces tolerance to A. fumigatus in THSFs (Fig. 2, Table 2). 

Figure 2

 

Copretreatment with zymosan and LPS-induced tolerance of THSFs to A. fumigatus . THSFs were cultured with 1000 ng/mL Zym + 10 ng/mL LPS, 100 ng/mL Zym + 10 ng/mL LPS, or DMEM for various periods (in hours: 6, 12, 18, 24, or 30) before the addition of A. fumigatus hyphae. The cells were assayed by real-time RT-PCR to assess mRNA expression of TNF-α (A), IL-6 (C), and IL-8 (E). The gene expression levels were normalized to GAPDH and presented as a relative ratio. Total cell supernatants were prepared after A. fumigatus hyphae challenge and analyzed for TNF-α (B), IL-6 (D), and IL-8 by ELISA (F) (n = 3). Bars: mean ± SD of three independent experiments. *P < 0.05; **P < 0.001 compared with control group without copretreatment. #P < 0.05; ##P < 0.001 compared with group copretreated with 1000 ng/mL Zym + 10 ng/mL LPS.

Table 2.

 

mRNA Expression and Protein Secretion of Inflammatory Cytokines After Copretreatment With Zymosan and LPS (Mean ± SD)

	6 h	12 h	18 h	24 h	30 h
TNF-α
RT-PCR (×10−5)
1000 ng/mL Zym + 10 ng/mL LPS	8.58 ± 0.0045	7.36 ± 0.15	9.15 ± 0.14	10.17 ± 0.46	11.23 ± 0.67
100 ng/mL Zym + 10 ng/mL LPS	4.16 ± 0.18**##	1.75 ± 0.067**##	2.39 ± 0.084**##	4.92 ± 0.057**##	8.32 ± 0.47
DMEM– A. fumigatus	9.99 ± 0.63	10.37 ± 0.21	11.12 ± 0.35	10.24 ± 0.49	10.13 ± 0.50
DMEM–DMEM	1.11 ± 0.057	0.94 ± 0.035	1.05 ± 0.035	1.14 ± 0.068	1.28 ± 0.055
ELISA (pg/ml)
1000 ng/mL Zym + 10 ng/mL LPS	11.64 ± 0.52	10.05 ± 0.18	13.43 ± 0.081	15.15 ± 0.21	18.39 ± 0.41
100 ng/mL Zym + 10 ng/mL LPS	6.24 ± 0.28**##	4.53 ± 0.21**##	6.061 ± 0.067**##	8.97 ± 0.14**##	15.14 ± 0.23
DMEM– A. fumigatus	15.11 ± 0.33	16.16 ± 0.23	15.46 ± 0.18	15.2 ± 0.34	15.73 ± 0.93
DMEM–DMEM	3.1 ± 0.14	3.52 ± 0.37	2.91 ± 0.33	3.26 ± 0.37	3.38 ± 0.53
IL-6
RT-PCR (×10−3)
1000 ng/mL Zym + 10 ng/mL LPS	9.72 ± 0.45	7.65 ± 0.13	8.71 ± 0.13	9.16 ± 0.74	9.73 ± 0.46
100 ng/mL Zym + 10 ng/mL LPS	5.47 ± 0.16**##	2.19 ± 0.2**##	2.76 ± 0.032**##	5.66 ± 0.21**##	8.68 ± 0.057
DMEM– A. fumigatus	9.99 ± 0.017	9.86 ± 0.15	9.9 ± 0.17	9.66 ± 0.57	9.9 ± 0.17
DMEM–DMEM	2.93 ± 0.34	3.29 ± 0.12	2.57 ± 0.21	2.52 ± 0.35	3.13 ± 0.25
ELISA (pg/mL)
1000 ng/mL Zym + 10 ng/mL LPS	21.08 ± 0.75	17.66 ± 0.14	24.69 ± 0.13	25.35 ± 0.41	29.33 ± 0.42
100 ng/mL Zym + 10 ng/mL LPS	15.11 ± 0.27**##	12.85 ± 0.069**##	14.27 ± 0.074**##	19.2 ± 0.15**##	26.73 ± 0.29
DMEM– A. fumigatus	28.40 ± 0.54	28.53 ± 0.84	29.24 ± 1.45	27.45 ± 0.66	29.12 ± 0.96
DMEM–DMEM	3.38 ± 0.53	11.14 ± 0.21	10.88 ± 0.28	10.06 ± 0.27	11.05 ± 0.47
IL-8
RT-PCR (×10−3)
1000 ng/mL Zym + 10 ng/mL LPS	8.77 ± 0.27	7.64 ± 0.23	8.61 ± 0.15	8.61 ± 0.15	9.63 ± 0.46
100 ng/mL Zym + 10 ng/mL LPS	5.14 ± 0.12**##	2.27 ± 0.079**##	3.15 ± 0.071**##	6.37 ± 0.18**##	8.69 ± 0.16
DMEM– A. fumigatus	9.93 ± 0.075	9.81 ± 0.15	9.46 ± 0.56	9.71 ± 0.5	9.89 ± 0.19
DMEM–DMEM	3.13 ± 0.25	1.68 ± 0.09	1.61 ± 0.051	1.56 ± 0.065	1.58 ± 0.23
ELISA (pg/mL)
1000 ng/mL Zym + 10 ng/mL LPS	203.9 ± 1.9	210.73 ± 4.99	263.05 ± 4.42	293.04 ± 3.86	293.21 ± 2.23
100 ng/mL Zym + 10 ng/mL LPS	152.57 ± 2.61**##	127.33 ± 1.85**##	141.8 ± 1.96**##	217.72 ± 4.24**##	281.58 ± 2.19
DMEM– A. fumigatus	284.57 ± 1.64	286.26 ± 5.99	286.86 ± 3.01	287.12 ± 3.85	285.70 ± 1.83
DMEM–DMEM	113.63 ± 1.38	110.77 ± 2.8	107.81 ± 7.07	108.78 ± 2.05	112.23 ± 2.59

*   P < 0.05; **P < 0.001 compared with control group without copretreatment.

#   P < 0.05; ##P < 0.001 compared with group copretreated with 1000 ng/mL Zym + 10 ng/mL LPS.

We demonstrated that copretreatment of THSFs with zymosan and LPS resulted in impaired production of TNF-α, IL-6, and IL-8 in response to a secondary A. fumigatus hyphae challenge. Interestingly, THSFs copretreated with 100 ng/mL zymosan and 10 ng/mL LPS had lower TNF-α, IL-6, and IL-8 gene expression and secretion compared with groups copretreated with 1000 ng/mL zymosan and 10 ng/mL LPS when rechallenged with A. fumigatus hyphae. The inhibition was observed 6 hours after copretreatment with zymosan and LPS by reducing TNF-α (Fig. 2A), IL-6 (Fig. 2C), and IL-8 (Fig. 2E) gene expression. Proinflammatory cytokines and chemokines gene expression stimulated by A. fumigatus hyphae were significantly reduced after 12 hours of copretreatment. The decrease of proinflammatory cytokines and chemokines gene expression were still statistically significant after 24 hours of copretreatment with zymosan and LPS, but were not influenced after 30 hours. Furthermore, we tested the effect of copretreatment on cytokine secretion by ELISA. The cytokine levels were consistent with trends observed in gene expression of TNF-α, IL-6, and IL-8 (Figs. 2B, 2D, 2F). 

In another part of the experiment, we compared the impaired proinflammatory cytokines and chemokines production of the copretreatment with the single ligand pretreatment. THSFs were cultured with 100 ng/mL zymosan + 10 ng/mL LPS, 1000 ng/mL zymosan, or 10 ng/mL LPS for various periods (in hours: 6, 12, 18, 24, or 30) before a secondary A. fumigatus hyphae challenge for 4 hours. It was observed that inhibition of gene expression after copretreatment with zymosan and LPS was stronger than that after zymosan or LPS pretreatment for all time points (in hours: 6, 12, 18, 24, and 30) in the production of TNF-α, IL-6, and IL-8 (Fig. 3, Table 3); 12 hours seems to be “the best period” (pretreatment in the period can inhibit the production of inflammatory cytokines in the maximal degree) to preexpose, and the same phenomenon in time-dependent experiments was observed, no matter that the cells were cultured with single ligand or mixed ligands. Our studies showed that copretreatment with zymosan and LPS induces tolerance to A. fumigatus hyphae in THSFs, and the dosage of 100 ng/mL zymosan + 10 ng/mL LPS is appropriate (pretreatment in the dosage can inhibit the production of inflammatory factor in the maximal degree) to induce tolerance, better than any other dosages of mixed ligands or the single ligand. 

Figure 3

 

Comparison of impaired proinflammatory cytokines between copretreatment with zymosan + LPS and single ligand pretreatments (zymosan or LPS). THSFs were cultured with 100 ng/mL Zym + 10 ng/mL LPS, 1000 ng/mL Zym, or 10 ng/mL LPS for various periods (in hours: 6, 12, 18, or 24). The cells were washed twice, and challenged again with A. fumigatus hyphae for 4 hours. THSFs were assayed by real-time RT-PCR to assess mRNA expression of TNF-α (A), IL-6 (C), and IL-8 (E). The gene expression levels were normalized to GAPDH and presented as a relative ratio. Total cell supernatants were prepared after A. fumigatus high challenge and analyzed for TNF-α (B), IL-6 (D), and IL-8 by ELISA (F) by ELISA (n = 3). Bars: mean ± SD of three independent experiments. *P < 0.05; **P < 0.001 compared with group cultured with zymosan. #P < 0.05; ##P < 0.001 compared with group cultured with LPS.

Table 3.

 

Comparison of Impaired Proinflammatory Cytokines Between Copretreatment With Zymosan + LPS and Single Ligand Pretreatments (Mean ± SD)

	6 h	12 h	18 h	24 h	30 h
TNF-α
RT-PCR (×10−5)
100 ng/mL Zym + 10 ng/mL LPS	4.16 ± 0.18**##	1.75 ± 0.067**##	2.39 ± 0.084**##	4.92 ± 0.057**##	8.32 ± 0.47**##
1000 ng/mL Zym	5.63 ± 0.072	3.63 ± 0.3	4.78 ± 0.39	9.16 ± 0.099	9.79 ± 0.23
10 ng/mL LPS	6.2 ± 0.13	4.52 ± 0.14	5.23 ± 0.11	8.43 ± 0.11	12.77 ± 0.67
DMEM– A. fumigatus	9.99 ± 0.63	10.37 ± 0.21	11.12 ± 0.35	10.24 ± 0.49	10.13 ± 0.50
ELISA (pg/mL)
100 ng/mL Zym + 10 ng/mL LPS	6.24 ± 0.28**##	4.53 ± 0.21**##	6.061 ± 0.067**##	8.97 ± 0.14**##	15.14 ± 0.23**##
1000 ng/mL Zym	9.26 ± 0.3	6.28 ± 0.28	8.42 ± 0.093	14.42 ± 0.23	15.64 ± 0.12
10 ng/mL LPS	10.98 ± 0.22	7.38 ± 0.12	9.50 ± 0.14	12.27 ± 0.3	17.13 ± 0.3
DMEM– A. fumigatus	15.11 ± 0.33	16.16 ± 0.23	15.46 ± 0.18	15.2 ± 0.34	15.73 ± 0.93
IL-6
RT-PCR (×10−3)
100 ng/mL Zym + 10 ng/mL LPS	5.47 ± 0.16*##	2.19 ± 0.2**##	2.76 ± 0.032**##	5.66 ± 0.21**##	8.68 ± 0.057*##
1000 ng/mL Zym	6.22 ± 0.072	3.93 ± 0.045	5.11 ± 0.17	8.71 ± 0.13	10.17 ± 0.29
10 ng/mL LPS	7.56 ± 0.28	4.47 ± 0.036	5.49 ± 0.16	8.16 ± 0.19	19.67 ± 0.58
DMEM– A. fumigatus	9.99 ± 0.017	9.86 ± 0.15	9.9 ± 0.17	9.66 ± 0.57	9.9 ± 0.17
ELISA (pg/mL)
100 ng/mL Zym + 10 ng/mL LPS	15.11 ± 0.27*##	12.85 ± 0.069**##	14.27 ± 0.074**##	19.2 ± 0.15**##	26.73 ± 0.29*##
1000 ng/mL Zym	17.49 ± 0.37	15.67 ± 0.15	18.18 ± 0.33	23.76 ± 0.14	27.15 ± 0.27
10 ng/mL LPS	19.98 ± 0.66	16.06 ± 0.26	19.05 ± 0.15	22.56 ± 0.13	32.26 ± 1.15
DMEM– A. fumigatus	28.40 ± 0.54	28.53 ± 0.84	29.24 ± 1.45	27.45 ± 0.66	29.12 ± 0.96
IL-8
RT-PCR (×10−3)
100 ng/mL Zym + 10 ng/mL LPS	5.14 ± 0.12**##	2.27 ± 0.079**##	3.15 ± 0.071**##	5.18 ± 0.15**##	8.69 ± 0.16**##
1000 ng/mL Zym	6.26 ± 0.04	3.5 ± 0.11	7.01 ± 0.18	8.59 ± 0.19	8.94 ± 0.14
10 ng/mL LPS	6.46 ± 0.04	4.92 ± 0.06	5.25 ± 0.2	7.43 ± 0.33	9.97 ± 0.052
DMEM– A. fumigatus	9.93 ± 0.075	9.81 ± 0.15	9.46 ± 0.56	9.71 ± 0.5	9.89 ± 0.19
ELISA (pg/mL)
100 ng/mL Zym + 10 ng/mL LPS	152.57 ± 2.61**##	127.33 ± 1.85**##	141.8 ± 1.96**##	217.72 ± 4.24**##	281.58 ± 2.19
1000 ng/mL Zym	191.38 ± 1.35	175.74 ± 3.17	226.3 ± 5.72	275.22 ± 3.15	311.08 ± 2.19
10 ng/mL LPS	220.18 ± 1.08	187.59 ± 2.28	211.11 ± 6.71	263.52 ± 3.1	322.98 ± 2.67
DMEM– A. fumigatus	284.57 ± 1.64	286.26 ± 5.99	286.86 ± 3.01	287.12 ± 3.85	285.70 ± 1.83

*   P < 0.05; **P < 0.001 compared with group cultured with zymosan.

#   P < 0.05; ## P < 0.001 compared with group cultured with LPS.

Effect of Treatment and Pretreatment of TLR Ligands on the Proliferation of THSFs

MTT analysis showed that treatment with zymosan, LPS, or cotreatment with both TLR ligands did not suppress proliferation of THSFs in a certain period (Fig. 4; Table 4). Although OD values of groups stimulated with TLR ligand(s) decreased, there were no significant differences compared with control group after 12 and 24 hours stimulation. However, OD values decreased after 36-hour stimulation of TLR ligands. The analysis also showed that stimulation of high-dosage A. fumigatus hyphae decreased the proliferation of THSFs for all periods. Our study suggested that low-dosage zymosan and/or LPS did not influence proliferation of THSFs significantly in a certain period. 

Figure 4

 

Effect of treatment of TLR ligands on the proliferation of THSFs. THSFs were stimulated with 1000 ng/mL Zym, 10 ng/mL LPS, 100 ng/mL Zym + 10 ng/mL LPS, or 10⁶ CFUs/mL A. fumigatus for various periods (in hours: 12, 24, and 36), with serum-free DMEM as control. The experiment was repeated for three times.

Table 4.

 

Effect of Treatment of TLR Ligands on the Proliferation of THSFs (Mean ± SD)

Treatment	12 h	24 h	36 h
1000 ng/mL Zym	0.246 ± 0.016	0.338 ± 0.02	0.419 ± 0.024**
10 ng/mL LPS	0.226 ± 0.087	0.338 ± 0.028	0.33 ± 0.009**
100 ng/mL Zym + 10 ng/mL LPS	0.233 ± 0.015	0.325 ± 0.006	0.418 ± 0.013**
A. fumigatus	0.189 ± 0.03*	0.202 ± 0.08**	0.18 ± 0.035**
DMEM	0.273 ± 0.013	0.382 ± 0.044	0.571 ± 0.007

*   P < 0.05; **P < 0.001 compared with group cultured with DMEM.

MTT analysis also showed that pretreatment with zymosan, LPS, or cotreatment with both TLR ligands suppressed the lethal effect of A. fumigatus to THSFs in a certain period (Fig. 5; Table 5). OD values of THSFs pretreated with TLR ligand(s) for 12 hours decreased compared with the control group. However, OD values of groups pretreated with TLR ligand(s) for 24 hours did not decrease compared with control group, and were higher compared with the group cultured with DMEM– A. fumigatus . Although OD values of THSFs pretreated with 1000 ng/mL zymosan and 10 ng/mL LPS for 36 hours decreased, proliferation of the group pretreated with mixed ligands was not influenced. The study suggested that pretreatment of TLR ligands may protect THSFs from A. fumigatus invasion in a certain period, and copretreatment with appropriate concentration of zymosan and LPS may have a better effect. 

Figure 5

 

Effect of pretreatment of TLR ligands on the proliferation of THSFs. THSFs were pretreated with 1000 ng/mL Zym, 10 ng/mL LPS, and 100 ng/mL Zym + 10 ng/mL LPS for various periods (in hours: 12, 24, and 36), and then stimulated THSFs with 10⁶ CFUs/mL A. fumigatus for 4 hours. A blank control group was cultured with serum-free medium. The experiment was repeated for three times.

Table 5.

 

Effect of Pretreatment of TLR Ligands on the Proliferation of THSFs (Mean ± SD)

Pretreatment	12 h	24 h	36 h
1000 ng/mL Zym– A. fumigatus	0.198 ± 0.014**	0.31 ± 0.014#	0.332 ± 0.007**
10 ng/mL LPS– A. fumigatus	0.202 ± 0.018*	0.0347 ± 0.023#	0.35 ± 0.026**
100 ng/mL Zym + 10 ng/mL LPS– A. fumigatus	0.207 ± 0.015*	0.354 ± 0.035##	0.429 ± 0.023*##
DMEM– A. fumigatus	0.207 ± 0.015*	0.230 ± 0.012**	0.311 ± 0.023**
DMEM–DMEM	0.262 ± 0.026	0.343 ± 0.036	0.512 ± 0.027

*   P < 0.05; **P < 0.001 compared with group cultured with DMEM.

#   P < 0.05; ##P < 0.001 compared with group cultured with DMEM– A. fumigatus .

Discussion

In the present study, we pretreated THSFs with zymosan, which led to inhibition of gene expression of proinflammatory cytokines and chemokines (TNF-α and IL-6) after restimulation with A. fumigatus hyphae. Furthermore, we copretreated THSFs with zymosan and LPS, and less secretion of TNF-α, IL-6, and IL-8 after restimulation with A. fumigatus hyphae was shown above. These demonstrated that the preexposure to mixed ligands had a stronger effect on inhibition of production of proinflammatory cytokines and chemokines than any single ligand. According to the MTT assay, irrespective of the cotreatment with zymosan and LPS or stimulation separately, proliferation of THSFs was not influenced in a certain period (12–24 hours). Simultaneously, pretreatment with zymosan and/or LPS decreased the proliferation inhibition caused by A. fumigatus hyphae. 

Keratocytes play a substantial role in the regulation of inflammation in the infected cornea. The corneal epithelium is an efficient physical barrier to infections and the first line of defense against microbial pathogens. ²⁵ Interestingly, the human corneal epithelium efficiently expresses some essential components of the TLR signaling pathway, such as MD-2, which is crucial to the LPS–TLR4 signaling pathway. ²⁶ These suggest that the corneal epithelium can tolerate the normal flora of the ocular surface. Once the barrier is destroyed, the invading microbial pathogens will cause an initial immune response induced by TLRs. Previous study showed that keratocytes are actively involved in the ocular immune response to pathogenic microorganisms under TLRs' triggering and secreted proinflammatory cytokines and chemokines, IL-6, and TNF-α. ²⁷ Other studies in our laboratory have shown that TLR2 and TLR4 are involved in the development of fungal keratitis. ^4,15,18 Zhang et al. ⁴ successfully suppressed excessive antifungal inflammation of THSFs caused by induction of TLR4-tolerance, whereas gene expression of the antimicrobial peptide was not affected. The present study showed that preexposure of THSFs to TLR ligands leads to tolerance, and may prove that THSFs are a bridge between innate immunity and adaptive immunity. 

The TLR family has at least 13 identified subtypes. ^28–30 Multiple TLR pathways are activated by the different components of fungi, and the interplay between TLR signaling pathways has been reported and may have important effects on host inflammatory responses. ¹⁵ It has been shown that TLR2 and TLR4 play crucial roles in the recognition and respond to A. fumigatus . ^15,31 Our study demonstrated that copretreatment with zymosan (100 ng/mL) and LPS (10 ng/mL) induced tolerance to A. fumigatus hyphae in THSFs, and the inhibitory effect of copretreatment on the production of proinflammatory cytokines and chemokines was stronger than that with zymosan or LPS alone. To our knowledge, this is the first study about the effect of cotolerance of TLR2 and TLR4 on A. fumigatus infection in THSFs. Additionally, we observed time-dependent and dosage-dependent differences in the inhibitory effect of not only zymosan pretreatment, but also copretreatment on cytokine production induced by A. fumigatus in a certain range. We presumed that the copretreatment for a certain time period and the dosage may induce a critical activation status once the corneal stroma is infected by fungi, the innate immune response would be more effectively and rapidly activated. In the present study, it was found that compared with 1000 ng/mL zymosan + 10 ng/mL LPS, the inhibitory effect of copretreatment with 100 ng/mL zymosan + 10 ng/mL LPS was stronger regardless of the periods of copretreatment (in hours: 6, 12, 18, or 24), which suggested that combination of most effective dosage of zymosan and LPS was sufficient to induce tolerance. 

Other studies showed the existence of synergy and cross-tolerance between TLR2 and TLR4 in mononuclear macrophages and dendritic cells. ^32,33 Hence, we may speculate that synergistic and cross-tolerance effects exist between TLR2 and TLR4 in fungal keratitis. There is a common signaling pathway or protein linking TLR2 and TLR4 agonists, which can influence the transduction mechanisms of both receptors in THSFs, although the mechanism is so complicated that there is no unified conclusion. Bagchi et al. ¹⁶ concluded that these synergistic effects are observed mainly when the activation of a TLR relies on MyD88 for signaling, whereas another one is capable of MyD88-independent signaling, such as TLR2 and TLR4. Other studies, however, showed that both receptors, which exhibit synergistic and cross-tolerance effects, rely on MyD88, and the mechanism is based on the downstream protein, such as NF-κB or JNK. ^34,35  

MTT assay in our study demonstrated that stimulation of low-dose TLR ligands did not influence proliferation of THSFs significantly in a certain period (12 and 24 hours). Although nontoxicity could not be declared, these data suggested that appropriate dosage of TLR ligand(s) was low-toxic to THSFs. Proliferation of THSFs copretreated with 100 ng/mL zymosan and 10 ng/mL LPS for 24 and 36 hours was better compared with cells cultured in DMEM– A. fumigatus . Moreover, copretreatment with zymosan (100 ng/mL) and LPS (10 ng/mL) had a better effect in suppressing the lethal effect of A. fumigatus and protect THSFs (24 and 36 hours). These suggested that pretreatment and coprretreatment with TLR ligands protected THSFs from A. fumigatus invasion. 

Although corneal inflammation is important for host defense against invading fungi, an excessive host inflammatory response can cause tissue damage, resulting in loss of corneal transparency and even blindness. Thus, the key to prevent and control blindness caused by corneal fungal infection is to regulate corneal antifungal infection ability. Our study expounded the inhibitory effect of copretreatment with TLR ligand in corneal fibroblasts challenged with A. fumigatus on the production of proinflammatory cytokines and chemokines and low-toxic to THSFs, and our previous study showed that immune tolerance mediated by the TLR-specific ligand will not affect normal corneal PMN migration. ⁴ All of them suggested that TLR ligand may be used to treat or prevent an excessive inflammatory response and control corneal infection. However, the main limitation of the study is that the TLR ligand may cause an inflammatory reaction before tolerance is induced. Thus, the concentration of the ligand for treatment must be adequate to induce tolerance to A. fumigatus , but not cause an excessive inflammatory response. Mixed specific ligands of TLR2 and TLR4 caused milder inflammatory response and induced stronger tolerance to A. fumigatus compared with the TLR2- or TLR4-specific ligand alone. Mixed ligands of zymosan and LPS may provide a better choice to induce tolerance to A. fumigatus because of its evident suppression of excessive host inflammatory response and protectiveness for keratocytes. Therefore, our study suggests that the proper concentration of mixed zymosan and LPS may be used to prevent corneal fungal infection and control an excessive inflammatory response of fungal keratitis. 

Supplementary Materials

pdf S1 

Acknowledgments

The authors thank Fu-Shin Yu of Wayne State University (Detroit, Michigan) for providing THSFs, and Edward C. Mignot, Asad Muhammad, and Leyi Wang of Shandong University (People's Republic of China) for linguistic advice. 

Supported by National Natural Science Foundation of China Grants 30571997 and 30872807. 

Disclosure: Y. Li, None; H. Yang, None; X. Wu, None 

Footnotes

 YL and HY contributed equally to the work presented here and should therefore be regarded as equivalent authors.