Diesel Exhaust Particles Selectively Induce Both Proinflammatory Cytokines and Mucin Production in Cornea and Conjunctiva Human Cell Lines

Abstract

Purpose.: To evaluate the effect of diesel exhaust particles (DEP) on the viability, proliferation, apoptosis, secretion of cytokines (IL-6, IL-8, and TNF-α), and mucin gene transcription (MUC1, MUC5AC, and MUC16) in human epithelial cells of the cornea (HCLE) and conjunctiva (IOBA-NHC).

Methods.: HCLE and IOBA-NHC cells were incubated with DEP (10–500 μg/mL) for 24 hours. Cell proliferation was evaluated by the 3-(4,5-dimethylthiazol-2-yl)–2,5-diphenyltetrazolium bromide assay. Apoptotic cells were measured by an annexin V–FITC and propidium iodide kit for flow cytometry. Proinflammatory cytokines were determined by an ELISA kit. Mucin gene transcription was quantified by real-time PCR.

Results.: DEP significantly decreased the viability, proliferation, and secretion of IL-8, but increased the secretion of IL-6 on both HCLE and IOBA-NHC cell lines in a dose-dependent manner. Neither cornea nor conjunctiva cells incubated with DEP released TNF-α. DEP induced a significant increase in the percentage of apoptotic cells in IOBA-NHC, whereas no changes were observed in HCLE. Finally, DEP significantly decreased the transcription levels of MUC1 and MUC16 in HCLE, but increased the transcription levels of MUC1, MUC5AC, and MUC16 in IOBA-NHC.

Conclusions.: These findings suggest that human corneal and conjunctival epithelial cells incubated with DEP showed cytotoxicity and an inflammatory response mediated by IL-6, not by TNF-α or IL-8. Also, the decrease in mucin expression in the cornea cells might leave exposed areas in the cornea for contact with DEP. Finally, the increase in mucin expression in the conjunctiva cells might be involved at least in the clearance of DEP to protect the ocular epithelium.

Spanish Abstract

Introduction

Diesel exhaust particles (DEP) are one of the major forms of particulate matter in urban air pollution. On the ocular mucosa, the tear film is the only interface that separates the epithelial cells (cornea and conjunctiva) from the environment. Thus, DEP is practically in contact with it. 

Numerous papers have addressed the effects of DEP on the respiratory mucosa. ^1–6 Acute and chronic inhalation of DEP in healthy individuals and in those with preexisting respiratory disease results in respiratory toxicity with consequent development of lung edema, infiltration of polymorphonuclear leukocytes, production of proinflammatory cytokines, and reactive oxygen species (ROS). ^7–10 However, few studies have shown the effects of DEP on the ocular mucosa. ^11–13  

Inhabitants of cities are exposed to high levels of air pollutants, including DEP, and often present irritation, burning, foreign body sensation, hyperemia, and itching in the eye. ¹⁴ They also show a reduced time of rupture of the tear film and slight damage to the surface layer of the cornea. ^15,16 Moreover, air pollutants exacerbate the symptoms and signs present in patients with dry eye or other chronic diseases of the conjunctiva. ¹⁷  

Different cell surface-associated mucins have been detected in cornea and conjunctiva epithelia (MUC1, MUC4, and MUC16). Conjunctival epithelium also expresses the small, soluble MUC7. Conjunctival goblet cells produce the gel-forming mucin MUC5AC. Mucins contribute differently to the protection of the ocular surface against allergens, pathogens, extracellular molecules, abrasive stress, and drying. ¹⁸ We hypothesize that mucin gene expression in cornea and/or conjunctiva exposed to air pollutant could be modified to carry out the clearance of particulate matter. 

The aim of this study was to evaluate in vitro the effect of DEP for 24 hours on corneal (HCLE) and conjunctival (IOBA-NHC) human epithelial cells. We analyzed cell viability, cell death, secretion of proinflammatory cytokines, and mucin gene expression. 

Materials and Methods

Cells

The human corneal epithelium cell line (HCLE) was kindly provided by Gipson, ¹⁹ and the normal human conjunctival epithelium cell line (IOBA-NHC) was kindly provided by Diebold. ²⁰ HCLE cells were grown in serum-free keratinocyte media (K-sfm) supplemented with 0.2 ng/mL epidermal growth factor (EGF), 25 μg/mL bovine pituitary extract (Gibco-Life Technologies, Carlsbad, CA), 0.4 mM CaCl₂ (Biopack; Zárate, Buenos Aires, Argentina), 50 U/mL penicillin, 50 μg/mL streptomycin, and 2 μg/mL amphotericin B (Gibco-Life Technologies) in a humid atmosphere of 37°C with 5% CO₂. HCLE cells were switched at approximately half confluence to a 1:1 mixture of K-sfm and Dulbecco's modified Eagle's medium–Nutrient Mixture F-12 (Ham) (1:1) with glutamine, high glucose, and 15 mM HEPES (DMEM-F12; Gibco-Life Technologies) for 24 hours to achieve confluence. IOBA-NHC were grown in DMEM-F12 supplemented with 10% fetal bovine serum (Internegocios, Mercedes, Buenos Aires, Argentina), 2 ng/mL EGF, 5 μg/mL hydrocortisone (Sigma, St. Louis, MO), 1 μg/mL bovine pancreas insulin (Sigma), 50 U/mL penicillin, 50 μg/mL streptomycin, and 2 mg/mL amphotericin B in a humid atmosphere of 37°C with 5% CO₂. 

Diesel Exhaust Particles

DEP from diesel motor combustion were kindly provided by Saldiva. ²¹ First, a 500 μg/mL DEP solution was made by weighing DEP, adding DMEM-F12, and sonicating in an ultrasonic bath (Ultrasonic Cleaner; Testlab, Bernal Oeste, Buenos Aires, Argentina) for 15 minutes. Then, 10, 50, and 100 μg/mL DEP solutions were made from the DEP 500 μg/mL solution and brought to the final volume with DMEM-F12. Finally, all DEP solutions (10–500 μg/mL) were sonicated for 15 minutes. 

HCLE and IOBA-NHC Incubated With DEP

Corneal or conjunctival cell monolayers were incubated with DEP (10, 50, 100, and 500 μg/mL) or with DMEM-F12 (control cells) for 24 hours. Then, cell monolayers were washed twice with phosphate-buffered saline (PBS) 1×, pH = 7.4. HCLE cells were removed with 0.05% trypsin-EDTA (Gibco-Life Technologies), and IOBA-NHC cells were removed with 0.25% trypsin-EDTA (Sigma) to perform the assays. 

Cell Viability Determination

The cell suspensions from each condition were mixed with an equal amount of 0.4% Trypan blue (Gibco-Life Technologies). The resulting suspension was placed in a Neubauer chamber for counting dead (blue) and live (colorless) cells under light microscopy. 

MTT Assay

The 3-(4,5-dimethylthiazol-2-yl)−2,5-diphenyltetrazolium bromide (MTT) that is yellow and soluble is reduced by active mitochondria in living cells to formazan salts (violet and insoluble). For that reason, MTT is used to measure cell proliferation. HCLE and IOBA-NHC cells were plated at 3 × 10⁴ cells per well in a 24-well tissue culture plate and grown for 24 hours prior to treatment. The culture medium was removed, and DEP (0, 10, 50, 100, and 500 μg/mL) solutions made in culture medium were added. Then at 24 hours, the DEP solutions were removed, and each well was washed twice with 1× PBS, pH = 7.4. Culture medium with 0.4 mg/mL MTT (Sigma) was added to each well (final volume = 500 μL). Plates were incubated for 3 hours in a humid atmosphere of 37°C with 5% CO₂. Intracellular formazan products were lysed by the addition of 250 μL 10% sodium dodecyl sulfate (Sigma), and incubation was carried out overnight at room temperature in the dark. Finally, the lysate of each well was homogenized, transferred to a tube, and centrifuged at 10,000g for 10 minutes in order to eliminate DEP. The absorbance of each supernatant was recorded at 550 nm using a plate reader (iMark Microplate Absorbance Reader; Bio-Rad, Hercules, CA). 

Apoptosis Evaluation

The number of dead cells was determined by annexin V–FITC and propidium iodide (PI) kit for flow cytometry (BD Biosciences, Franklin Lakes, NJ). Cells at a density of 2.5 × 10⁵ were placed in cytometer tubes, washed twice with cold 1× PBS, and resuspended in 100 μL 1× annexin V binding buffer. Then, 5 μL annexin V–FITC and 5 μL IP were added to each tube. The tubes were gently vortexed and incubated for 15 minutes in the dark. Finally, 400 μL 1× annexin V binding buffer was added to each tube, and the flow cytometric analysis was performed (FACSCalibur; BD Biosciences) (argon laser, 488 nm). The values obtained were analyzed by Cyflogic 1.2.1 (CyFlo Ltd., Turku, Finland). Annexin V binds to negatively charged phospholipid surfaces with a higher specificity for phosphatidylserine (PS) than most other phospholipids. PS is an internal plasma membrane phospholipid that in the early apoptotic cascade is exposed on the outer layer of the plasma membrane, before the cells become permeable to PI. Therefore, early apoptotic cells are annexin V–FITC positive and PI negative, and cells that are in late apoptosis or already dead (necrosis) are both annexin V–FITC and PI positive. Viable cells are annexin V–FITC and PI negative. 

Proinflammatory Cytokine Determination

The culture media from each experimental condition were used to determine cytokine secretion. In order to eliminate the particles, the media were centrifuged at 10,000g for 10 minutes at 4°C. Particle-free supernatants were used to determine the secretion of TNF-α, IL-6, and IL-8 using ELISA kits (BD Biosciences) following the manufacturer's technical specifications. 

Mucin Gene Expression Evaluation

Total RNA was isolated from 2 × 10⁶ cells from each experimental condition with a commercial column kit (RNeasy Mini Kit; Qiagen, Duesseldorf, Germany). Then the cDNA was synthesized from 5 μg total RNA per sample. Random hexamer primers were added to each sample, followed by heating at 70°C for 10 minutes. After a brief centrifugation, a mixture containing PCR buffer, MgCl₂, Deoxynucleotide Triphosphates (dNTPs), and dithiothreitol was added to every sample, and incubation at room temperature was performed for 5 minutes. Reverse transcription was performed by adding 1 μL reverse transcriptase (SuperScript II Reverse Transcriptase; Invitrogen-Life Technologies, Carlsbad, CA) to the mixture and incubating 10 minutes at 25°C, 50 minutes at 42°C, and 15 minutes at 70°C. 

Both RNA and cDNA were quantified in a ND-1000 Spectrophotometer (NanoDrop Technologies Inc., Wilmington, DE), and their purity was confirmed by the absorbance ratio 260:280 of all samples that were between 1.8 and 2. 

First, conventional RT-PCR experiments were performed to confirm the expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), MUC1, and MUC16 when amplifying HCLE and IOBA-NHC cDNA with the primers used in this study. Also, the expression of MUC5AC was confirmed on IOBA-NHC cDNA (Table). PCR products were electrophoresed in 1.5% agarose gels containing ethidium bromide. 

Table

 

Primers Used in Conventional and Real-time PCR

Gen	Sense Primer	Antisense Primer	Size, pb
MUC123	5′-GTGCCCCCTAGCAGTACCG-3′	5′-GACGTGCCCCTACAAGTTGG-3′	123
MUC5AC23	5′-TCCACCATATACCGCCACAGA-3′	5′-TGGACCGACAGTCACTGTCAAC-3′	103
MUC1624	5′-GCCTCTACCTTAACGGTTACAATGAA-3′	5′-GGTACCCCATGGCTGTTGTG-3′	114
GAPDH23	5′-GAAGGTGAAGGTCGGAGTC-3′	5′-GAAGATGGTGATGGGATTTC-3′	226

Secondly, real-time PCR was performed on the Rotor Gene (Qiagen). From the total reverse transcription volume of 20 μL, 1 μL was used for each real-time PCR. Real-time PCR was performed in triplicate in a total volume of 25 μL. Syber green fluorophore was used as marker of the amplified products (Rotor-Gene SYBR Green RT-PCR Kit; Qiagen). GAPDH was used as housekeeping gene to correct for differences in the amount of total RNA added to each reaction. A nontemplate control was included in all the experiments performed with real-time PCR to evaluate DNA contamination of the reagents used for amplification. None of the experiments resulted in a nontemplate control-positive signal, indicating that there was no DNA contamination in the RNA used for the assays. The threshold cycles (C_T values) obtained were analyzed according to the model presented by Pfaffl. ²²  

Statistical Analysis

The results presented correspond to three independent experiments performed in triplicate. A one-way ANOVA test (α = 0.05) and Dunnett's test (α = 0.05) as a post hoc test were used for statistical analysis in SPSS 17.0 (Sun Microsystems, Inc., Santa Clara, CA). 

Results

Viability, Proliferation, Apoptosis, and Cell Death

Viability and proliferation on both cell lines (HCLE and IOBA-NHC) decreased with DEP in a dose–response manner (Fig. 1). The values obtained by Trypan blue exclusion were consistent with the values obtained by MTT assay. Significant differences (P ≤ 0.01) were found in both parameters (viability and proliferation) in both cell lines incubated with 50, 100, and 500 μg/mL DEP compared to controls. 

Figure 1

 

Box plots of viability (% of viable cells by Trypan blue exclusion) and proliferation (total cell number by Trypan blue exclusion and absorbance measured by MTT assay) in HCLE and IOBA-NHC incubated with DEP for 24 hours. *P < 0.05 versus control cells (not incubated with DEP) by Dunnett's post hoc test.

HCLE cells incubated with 100 μg/mL DEP showed no significant differences in the percentage of viability, apoptosis, and dead cells compared to control cells (Fig. 2). In contrast, the percentage of early apoptotic cells (Fig. 2, bottom right quadrant, 22.44 ± 2.7%) and late apoptotic and necrotic cells (Fig. 2, top right quadrant, 16.97 ± 1.9%) of IOBA-NHC cells incubated with 100 μg/mL DEP significantly (P = 0.037) increased compared to that in control cells. In addition, the percentage of viable cells (Fig. 2, top left quadrant, 65.6 ± 2.6%) significantly (P < 0.01) decreased in IOBA-NHC cells incubated with 100 μg/mL DEP compared to control cells. This was also consistent with the viability estimated by Trypan blue exclusion. 

Figure 2

 

Dot plots of HCLE and IOBA-NHC analyzed by flow cytometry with 488 nm laser. Annexin V–FITC was detected with a 550 nm filter and PI with a 650 nm filter. Viable cells were annexin V–FITC and PI negative. Early apoptotic cells were annexin V–FITC positive and PI negative. Late apoptotic and necrotic cells were both annexin V–FITC and PI positive. *P < 0.05 versus control cells (not incubated with DEP) by Dunnett's post hoc test.

Proinflammatory Cytokine Secretion

Both cell lines showed an increase in the release of IL-6 and a decrease in the release of IL-8 (Fig. 3). For HCLE, the increase of IL-6 was significant for all DEP concentrations used when compared to controls: control versus 10 μg/mL DEP (P = 0.037), control versus 50 μg/mL DEP (P = 0.041), control versus 100 μg/mL DEP (P = 0.010), and control versus 500 μg/mL DEP (P = 0.028). IL-8 secretion significantly decreased (P < 0.001) in a dose–response manner when compared to that in controls. For IOBA-NHC, the increase of IL-6 was significant (P = 0.017) when cells were incubated with 500 μg/mL DEP, while the release of IL-8 significantly decreased when cells were treated with DEP concentrations equal to or higher than 100 μg/mL (100 μg/mL, P = 0.002) and 500 μg/mL (P < 0.001). No release of TNF-α was observed for both corneal and conjunctival cells. 

Figure 3

 

Box plots of IL-6 and IL-8 measured by ELISA in HCLE and IOBA-NHC incubated with DEP for 24 hours. *P < 0.05 versus control cells (not incubated with DEP) by Dunnett's post hoc test.

Mucin Expression

A unique band corresponding to the predicted size for GAPDH, MUC1, MUC16, and MUC5AC was obtained after 40 cycles of cDNA amplification (Fig. 4). This result confirms the use of the primers chosen in the present study. 

Figure 4

 

Conventional PCR products. The photos show a unique band corresponding to the predicted size for GAPDH, MUC1, MUC5AC, and MUC16. PCR products were electrophoresed in 1.5% agarose gels containing ethidium bromide.

HCLE incubated with DEP showed a decrease in the expression of the two mucins (MUC1 and MUC16) evaluated. In contrast, IOBA-NHC incubated with DEP showed an increase in the expression of the three mucins (MUC1, MUC16, and MUC5AC) evaluated (Fig. 5). For HCLE, the decrease in the expression of mucins was statistically significant for 10 and 100 μg/mL DEP for MUC1 (P = 0.016, P = 0.002) and MUC16 (P = 0.020, P = 0.003), respectively. For IOBA-NHC, the increased expression of mucins was statistically significant when cells were exposed to 100 μg/mL DEP (MUC1, P < 0.001; MUC5AC, P = 0.049; MUC16, P < 0.001). 

Figure 5

 

Box plots of ratio fold changes in the expression of the mucins evaluated by real-time PCR in HCLE and IOBA-NHC incubated with DEP for 24 hours. *P < 0.05 versus control cells (not incubated with DEP) by Dunnett's post hoc test.

Discussion

In this study, in human corneal and conjunctival epithelial cells in vitro, we decided to evaluate the effect of DEP concentrations (10, 50, 100, and 500 μg/mL) similar to those analyzed in epithelial cells of human respiratory mucosa in vitro. ^25,26 Also, the only study published on DEP in conjunctiva epithelial cells in vitro used a concentration of DEP of 100 μg/mL because this concentration is almost equivalent to the maximum concentration in air observed in the Japanese environment since 1995. ¹¹  

In human corneal and conjunctival epithelial cells incubated with DEP, we observed a dose–response decrease in viability and proliferation. Also, in human conjunctival epithelial cells incubated with 100 μg/mL DEP, we found increases in both early and late apoptotic cells. This result is consistent with the cytotoxicity reported in human bronchial epithelial cells incubated with DEP at the same concentrations used in this study. ²⁶  

Studies in airway human epithelial cells and recently published work on conjunctiva human epithelial cells incubated with DEP for 24 hours have shown an increase in the mRNA levels and secretion of IL-6 and IL-8 and no release of TNF-α. ^11,25,26 In congruence with these results, we found that DEP increased the secretion of IL-6 on both corneal and conjunctival cells in a dose-dependent manner and did not induce secretion of TNF-α. This suggests that IL-6 is implicated in the inflammatory response of the ocular surface reported in individuals living in urban areas. ¹⁵  

In contrast with the results on respiratory mucosa and human conjunctiva, we found that DEP decreased the secretion of IL-8 on both corneal and conjunctival cells in a dose-dependent manner. We have noticed that mRNA levels of IL-8 are increasingly upregulated by 100 μg/mL DEP in human bronchial epithelial cells ²⁶ and human conjunctiva epithelial cells, ¹¹ but these levels of mRNA are not consistent with the levels of IL-8 protein measured. Also, Totlandsdal and coworkers ²⁶ affirmed that the relative increase of DEP-induced release was more pronounced for IL-6 than for IL-8, with the mRNA levels of IL-6 minor compared to the mRNA levels of IL-8. We detected the protein secretory leukoprotease inhibitor (Slpi) in human cornea and conjunctiva epithelial cells incubated with 100 μg/mL DEP for 24 hours (data not shown). Slpi suppress the expression of neutrophil elastase. ²⁷ Thus, one possible hypothesis to explain the high levels of IL-8 mRNA compared with levels of IL-8 protein is that suppression of neutrophil elastase by Slpi blocked the expression of IL-8. On the other hand, IL-8 is an interleukin that has chemotactic activity for polymorphonuclear neutrophils (PMNs). Inflammatory infiltrate with predominance of PMNs was shown in ex vivo lung tissue exposed to DEP. ^28,29 In contrast, a previous study from our group showed that exposure to urban air pollutants is unable to induce infiltration of PMNs in human conjunctiva. ³⁰ Thus, the decrease in IL-8 that we found has a relationship with our previous findings in humans. 

Corneal human epithelial cells incubated with DEP showed a decrease in the expression of MUC1 and MUC16. Also, we found that DEP increased the secretion of IL-6. In congruence with these results, the study by Albertsmeyer et al. established a decrease in the expression of MUC1 and MUC16 on HCLE when these cells were stimulated with IL-6 cytokine. ³¹ We do not know if the decrease of cell surface-associated mucins (MUC1, MUC16) is due to a direct effect of DEP or an indirect effect of DEP through early stimulation of IL-6. Cell surface-associated mucins, as well as MUC1 and MUC16, can contribute to the maintenance of the mucosal barrier integrity, preventing the penetration of extracellular molecules onto ocular surface epithelia. ¹⁸ We suggest that the decreased expression in MUC1 and MUC16 may produce a smooth glycocalyx at the epithelia–tear film interface, not preventing the penetration of extracellular molecules onto ocular surface epithelia. The smooth glycocalyx leaves areas of the corneal epithelium most exposed to DEP and could explain the decrease in the time of rupture of the tear film and the symptoms experienced by normal individuals who live and work in cities with high air pollution levels. ^15,30  

Conjunctival human epithelial cells incubated with DEP showed an increase in the expression of MUC1, MUC5AC, and MUC16. These results are related to our previous studies in humans demonstrating a significant increase in goblet cells and mucosubstancia in the epithelial conjunctiva. ³⁰ Gel-forming mucins as well as MUC5AC have the capability to trap allergens and debris in order to facilitate their clearance from mucosal surfaces. ¹⁸ We suggest that the increase in the expression of MUC1, MUC5AC, and MUC16 might produce an adaptive process that contributes to protecting conjunctiva from adverse effects of air pollution and also contributes to the clearance of DEP on the ocular surface. 

This study shows the effects of exposure to DEP for 24 hours on corneal and conjunctival human epithelial cells. In these conditions, DEP causes a dose-dependent decrease in cell viability, proliferation, and secretion of IL-8 and an increase in the secretion of IL-6 on corneal and conjunctival human epithelial cells. DEP had no effect on the secretion of TNF-α. Finally, DEP decreased the transcription levels of MUC1 and MUC16 on corneal human epithelial cells, but increased the transcription levels of MUC1, MUC5AC, and MUC16 on conjunctival human epithelial cells. These findings suggest that epithelial cells from the ocular surface could elicit different adaptive mechanisms when exposed to DEP. 

It is noteworthy that these results were obtained in epithelial monolayers of cornea and conjunctiva. The human corneal epithelium is stratified; their columnar cells are attached to the basement membrane through hemidesmosomes. The human conjunctival epithelium is multilayered columnar. An epithelial monolayer incubated with DEP is not the same as stratified epithelium exposed to DEP. 

Further studies on the molecular mechanisms whereby DEP could stimulate an adaptive response on the ocular surface will contribute to the understanding of ocular symptoms that healthy individuals may present and the complications that patients with eye diseases such as dry eye may suffer in cities with air pollution. 

Supplementary Materials

Spanish Abstract 

Acknowledgments

The authors thank Graciela Zaccagnini and Nadia Orona for their professional assistance. 

Supported by Agencia Nacional de Promoción Científica y Tecnológica, Subsidio PICT 2007-02252, Argentina; and Fundação de Amparo a Pesquisa do Estado de São Paulo, Edital 18, Brazil. 

Disclosure: J. Tau, None; P. Novaes, None; M. Matsuda, None; D.R. Tasat, None; P.H. Saldiva, None; A. Berra, None