Apoptosis Gene Expression and Death Receptor Signaling in Mitomycin-C–Treated Human Tenon Capsule Fibroblasts

Abstract

purpose. To examine the effect of mitomycin-C on the expression of apoptosis genes in human Tenon capsule fibroblasts and to evaluate whether death receptor signaling modulates mitomycin-C cytotoxicity.

methods. Bcl-2, Bax, Bcl-x, Fas (CD95) and tumor necrosis factor (TNF) receptor expression was determined by flow cytometry in control and mitomycin-C–treated Tenon fibroblasts. Fibroblast death was quantified using a lactate dehydrogenase release assay. The effect of Fas and TNF-receptor signaling was evaluated using Fas-specific antibodies and soluble TNF-α.

results. Tenon fibroblasts constitutively express Bcl-2, Bax, and Bcl-x in culture. Mitomycin-C (0.4 mg/mL) induced a small but consistent increase in the expression of all three proteins. Tenon fibroblasts express low levels of Fas but are resistant to the effects of Fas-receptor ligation. Mitomycin-C (0.01–1.0 mg/mL) led to a significant increase in Fas expression at all concentrations tested (P < 0.01). Pretreatment with mitomycin-C (0.4 mg/mL) rendered fibroblasts susceptible to agonistic anti-Fas monoclonal IgM antibodies (50–500 ng/mL) and led to a further 50% reduction in viable fibroblasts at 48 hours, compared with mitomycin-C alone (P < 0.05). Antibodies that block the Fas receptor did not inhibit mitomycin-C–induced apoptosis.

conclusions. Mitomycin-C alters apoptosis gene expression and primes fibroblasts to the effects of Fas receptor ligation. Factors other than the level of Fas receptor expression modulate the response to Fas receptor signaling. Determining the signals that regulate fibroblast apoptosis may help to refine therapeutic strategies for switching off the subconjunctival healing response and maintaining intraocular pressure control.

Mitomycin-C has a profound inhibitory effect on the scarring response after glaucoma filtration surgery. Estimates suggest that intraoperative mitomycin-C is used in at least 50% of primary trabeculectomies in the United States and Japan. ¹ The benefits derived from the antiscarring activity have, however, been tempered by a possible increase in postoperative complications, including chronic hypotony, bleb leak, and endophthalmitis. ^{2 3} Despite widespread clinical use, the exact mechanisms by which single short applications of this chemotherapeutic agent mediate a prolonged antiscarring effect remain uncertain. 

The ability of single, short, intraoperative applications of mitomycin-C to induce long-term inhibition of fibroblast proliferation is well established. ^{4 5} Histologic analysis of mitomycin-C–treated tissue, however, has revealed largely acellular bleb tissue, ^{6 7} suggesting that treatment not only inhibits fibroblast proliferation, but also induces fibroblast death. We have shown previously that clinically relevant treatments with mitomycin-C in vitro induce Tenon capsule fibroblast death by apoptosis. ⁸ Fibroblast apoptosis heralds the termination of a physiological wound-healing response by converting active granulation tissue into relatively inactive scar. ⁹ It is therefore possible that mitomycin-C mediates its long-term inhibition by inducing fibroblast apoptosis and prematurely switching off the healing response. ⁸  

The Bcl-2 family of proteins are located on the mitochondrial membrane and either promote (Bax, Bak, Bad) or inhibit apoptosis (Bcl-2, Bcl-X_L) through regulating release of cytochrome c from the mitochondria into the cytoplasm. ^{10 11} The Tumor necrosis family of proteins, including the tumor necrosis factor (TNF) receptors and Fas (CD95/APO-1) are located on the plasma membrane. Receptor activation triggers apoptosis in susceptible cells. Fas is an important member of the TNF family, in that receptor ligation mediates apoptotic death in a wide range of physiological and pathologic processes. The role of Fas in chemotherapy-induced apoptosis remains controversial. A number of cytotoxic drugs including mitomycin-C and 5-fluorouracil have been shown to upregulate the expression of cell surface Fas in tumor cell lines. ^{12 13 14} In addition, manipulating Fas receptor signaling may modulate chemotherapy-induced apoptosis. ^{15 16} The role of Fas receptor signaling with respect to mitomycin-C–mediated apoptosis in fibroblasts has not previously been reported. 

The purpose of this study was to determine whether Tenon capsule fibroblasts express apoptosis related gene products in culture, ascertain the effect of mitomycin-C treatment on gene expression, and investigate the effect of modulating Fas receptor signaling in mitomycin-C–treated fibroblasts. 

Methods

Establishing a Human Tenon Capsule Fibroblast Line

A primary cell line of human Tenon capsule fibroblasts was established from explanted subconjunctival Tenon capsule isolated during glaucoma filtration surgery and propagated in culture as described previously. ¹⁷ The tenets of the Declaration of Helsinki were observed and institutional human experimentation committee approval was granted. Explanted tissue was anchored onto the bottom of a six-well plate (BD Biosciences, San Jose, CA) with a sterile coverslip and overlaid with Dulbecco modified Eagle’s medium (DMEM; Sigma, Poole, UK). All medium used was supplemented with l-glutamine (2 mM) penicillin (100,000 U/L; both from Gibco, Uxbridge, UK), and fetal calf serum (FCS; 10% of final volume; Gibco). Once the monolayers had reached confluence (at approximately 2 weeks), the fibroblasts were passaged and cultured in 175-cm² tissue culture flasks. 

Fas (CD 95/APO-1)-Sensitive Jurkat T-Cells

Fas sensitive human leukemic lymphoblastic Jurkat T-cells (J6) were provided as a kind gift of Stan Wickeramesinghe (Department of Hematology, Royal Free Hospital, London, UK). Jurkat T-cells were maintained in RPMI and 10% FCS and passaged with a split ratio of 1:20. 

Antimetabolite Treatment

Tenon fibroblasts were seeded at concentration of 10,000 fibroblasts per well into 24-well plates (BD Biosciences) and incubated overnight. The fibroblast monolayers were then washed and covered with single applications of mitomycin-C (Kyowa Hakko Kogyo, Ltd., Tokyo, Japan), as described previously. ⁴ Unless otherwise stated, the treatment time for all experiments was 5 minutes. Antimetabolites were diluted in phosphate-buffered saline (PBS). Control fibroblasts were treated with a 5-minute application of PBS. After treatment, the monolayers were washed immediately three times with 500 μL PBS and incubated in growth medium. 

Analysis of Cell Death

Cell Morphology. The number of viable fibroblasts was evaluated as previously described, ¹⁸ by counting the number of attached fibroblasts at ×40 magnification using phase-contrast microscopy (model OM2; Olympus, Tokyo, Japan). To quantify the proportion of cells displaying apoptotic morphology, cytospins of attached cells and cells in supernatant were prepared as previously described. ⁸ Apoptotic cells were easily identified on the basis of nuclear morphology when examined by light microscopy. The percentage of apoptosis was calculated from the ratio of apoptotic to viable fibroblasts from randomly selected fields. At least 300 cells were counted per cytospin at ×40 magnification. 

Lactate Dehydrogenase (LDH) Release Assays.

A lactate dehydrogenase release assay was also used to quantify cell death. This commercially available assay quantifies lactate dehydrogenase released by dead cells into the supernatant. Lactate dehydrogenase is a stable cytoplasmic enzyme present in all cells. It is rapidly released after damage to the plasma membrane. Lactate dehydrogenase catalyzes the reduction of a colorless tetrazolium salt to colored formazan, which absorbs a broad spectrum of light with maximum absorbance at approximately 492 nm. Absorbance was measured with a spectrophotometric microtiter plate reader (Titertek Plus; MTX Lab Systems, Vienna, VA). Lactate dehydrogenase is present in serum and can therefore be used only to assay cell death in serum-free conditions. 

Lactate dehydrogenase was measured in the supernatant of mitomycin-C–treated Tenon capsule fibroblasts, using a cytotoxicity detection kit (LDH; Roche Molecular Biochemicals, Philadelphia, PA), according to the manufacturer’s guidelines. Briefly, human Tenon capsule fibroblasts previously seeded into 24- or 48-well plates (BD Biosciences) were treated with a single 5-minute application of mitomycin-C or 5-fluorouracil, as described earlier; washed in PBS; and incubated in 400 μL phenol red–free DMEM with 1% bovine serum albumin (Sigma). After 48 hours’ incubation, 100 μL of supernatant was extracted from each well and placed into separate wells of a 96-well plate (BD Biosciences). Catalyst solution (100 μL at 37°C) was added to each well, followed by 15 minutes’ incubation. Absorbance was measured by a microtiter plate reader with a 490- to 492-nm filter. Background absorbance was measured with wells containing phenol red–free DMEM only. Low-absorbance control cultures contained untreated Tenon capsule fibroblasts and high-absorbance control cultures equating to 100% apoptosis were obtained by lysing fibroblasts with 100 μL 0.1% Triton-X added to 300 μL DMEM. The percentage of apoptosis was calculated by  

\[\mathrm{\%\ apoptosis\ {=}\ (experimental\ cells\ -\ low-absorbance\ control\ cells)/}\]

 

\[\mathrm{(high-absorbance\ control\ cells\ -\ low-absorbance\ control\ cells).}\]

 

Flow Cytometry.

Flow cytometric analysis was performed (FACStar-plus; BD Biosciences) with a 100-mW 488-nm argon laser light source. Light was filtered with an FL-1 filter at 520 ± 20 nm or an FL-2 filter at 580 ± 20 nm. Acquisition and analysis was performed on computer (LYSIS II software; BD Biosciences). 

Cell Preparation.

For fibroblast staining, unfixed monolayer cells were collected by trypsinization and collected together with cells in supernatant and wash solutions. The fibroblast suspension was centrifuged at 1000 rpm for 5 minutes and resuspended in 100 μL PBS. Non–contact-dependent cells were washed, centrifuged, and resuspended in 100 μL PBS. 

Fas and Bcl-2 Expression by Direct Immunofluorescence.

This single-step staining procedure with fluorochrome-conjugated antibody was used when possible. Anti Fas-FITC (UB2; 5 μL at a concentration of 0.5 μg/mL; Immunotech, Marseille, France) and anti-Bcl-2-FITC (5 μL at a concentration of 0.5 μg/mL; Dako, Glostrup, Denmark) was added to 100,000 cells in 100 μL PBS, mixed and then incubated for 15 minutes at room temperature. The cells were then washed and fixed in 100 μL 1% paraformaldehyde and stored at 4°C in the dark. 

Bax and Bcl-x Detection by Indirect Immunofluorescence.

In this two-step procedure, cells were labeled with the unconjugated first-layer anti-Bax IgG2 (Immunotech) and anti-Bcl-x_l/s polyclonal antibodies (Santa Cruz Biotechnology, Santa Cruz, CA), washed, and subsequently labeled with a second-layer fluorochrome-conjugated monoclonal antibody directed against the first layer. The primary antibody was added for 30 minutes at room temperature. In experiments in which a rabbit polyclonal primary layer was used, background fluorescence was reduced by a 15-minute incubation in RPMI containing FCS (10%). After application of the first layer, fibroblasts were washed and resuspended in 100 μL PBS. Five microliters of appropriate FITC- or phycoerythrin (PE)-conjugated second-layer antibody diluted to 0.5 μg/mL in PBS was added for 30 minutes at room temperature. The cells were then washed, fixed in 100μ L 1% paraformaldehyde, and stored at 4°C in the dark. 

Cytoplasmic Staining.

Direct or indirect staining techniques were used. Cells (100,000) suspended in 100 μL PBS were incubated in a fixative (Permeafix; Ortho Diagnostic Systems, Amersham, UK) diluted 1:1 in water for 45 minutes at room temperature. The cells were then washed and resuspended in 100 μL PBS and labeled with 5 μL FITC- or PE-conjugated antibody, or an equivalent volume of unconjugated first-layer antibody for 30 minutes at room temperature. Cells were then washed and, when indicated, resuspended in 100 μL PBS and 5 μL FITC- or PE-conjugated second-layer antibody for 30 minutes at room temperature. The cells were again washed and fixed in 100 μL 2% paraformaldehyde in PBS and stored at 4°C. 

Control Experiments

Control Cells.

Bcl-2 and TNF receptor family gene expression is well established in phytohemagglutinin (PHA)-activated IL-2 dependent T-cell lines. ¹⁹ These were used a positive control cultures for apoptosis gene expression. 

Control Antibody.

For indirect immunofluorescence, an isotype control irrelevant first-layer antibody was used, followed by the standard protocol for the conjugated second-layer antibody. For one-step labeling with FITC- or PE-conjugated antibodies, an irrelevant directly conjugated isotype control was used (e.g., anti-CD 45 RO-FITC, Dako) for fibroblast staining. 

Analysis

At least 5000 cells were analyzed in each specimen. When possible, triplicate specimens were prepared per treatment group. Analysis was performed on viable cells unless stated otherwise. Viable cells were gated according to forward- and side-scatter profiles. Markers were set on histograms plotting fluorescence intensity against cell counts on computer (WinMDI software version 1.3.3, Windows 3.1 Multiple Document Interface Flow Cytometry application; developed by Joe Trotter, The Scripps Research Institute, La Jolla, CA). Protein expression was determined from overlays created to show fluorescence intensity profiles obtained from isotype-stained control samples and samples with maximum positive fluorescence. 

Results

Flow Cytometry

Forward- versus side-scatter profiles for mitomycin-C and PBS-treated fibroblasts are shown in Figure 1 . Forward scatter increased with cell size (x-axis), and side scatter increased with cellular granularity (y-axis). Apoptotic cells were characteristically smaller and more granular than viable cells and therefore fall to the left and above viable cells on a scatterplot. Mitomycin-C (0.4 mg/mL for 5 minutes) led to a reduction of viable fibroblasts and an accumulation of smaller, more granular apoptotic fibroblasts. The scatter profile changes obtained for PHA-activated IL-2–dependent T-lymphocytes deprived of IL-2 for 48 hours showed a similar accumulation of smaller, more granular apoptotic cells. 

Analysis of background fluorescence intensity with isotype control antibody identified a small increase in background fluorescence in mitomycin-C–treated fibroblasts, compared with that in PBS-treated fibroblasts. The increase in median fluorescence was small, usually less than 10 intensity units (range 2–10, data not shown). Isotype control antibody staining was performed on PBS- and mitomycin-C–treated fibroblasts in all experiments, to ensure that the difference in nonspecific staining did not exceed this range. 

Bcl-2 Family Expression

Tenon fibroblasts constitutively expressed Bcl-2 in culture. The median fluorescence intensity (MFI ± SD, expressed in intensity units) obtained from three separate experiments performed in triplicate with anti-Bcl-2 FITC was 28.1 ± 2.9 compared with 7.9 ± 4.4, obtained with an FITC-conjugated IgG isotype control antibody (statistically significant at P = 0.004, paired t-test). To determine the effect of mitomycin-C on Bcl-2 expression, fibroblasts were examined 48 hours after a single 5-minute application of mitomycin-C (0.4 mg/mL) or PBS (control). Mitomycin-C led to a small but consistent increase in Bcl-2 expression. MFI for three experiments performed in triplicate was 33.1 ± 5.8, compared with PBS-treated control samples (23.1 ± 2.7; statistically significant at P = 0.045, paired t-test; Fig. 2 , top panel). 

Tenon capsule fibroblasts also expressed Bax and Bcl-x constitutively in culture. The MFI for Bax and Bcl-x expression appeared more varied than Bcl-2 and did not consistently reach statistical significance. In general, a small increase in Bax and Bcl-x expression occurred at 48 hours after mitomycin-C treatment compared with control PBS treatment. The MFIs and profiles obtained from a representative experiment are shown in Figure 2 (bottom two panels). Positive staining was obtained in the T-cell control cultures with all three antibodies used (data not shown). 

Fas and TNF-Receptor Expression

Tenon fibroblasts expressed low levels of Fas constitutively in culture. The MFI obtained with anti-Fas FITC was 14.7 ± 3.4 compared with isotype control IgG FITC (7.9 ± 1.6; Fig. 3 ; statistically significant at P = 0.004, paired t-test). Fas receptor expression increased in a dose-dependent manner after 5-minute exposures to mitomycin-C in doses ranging between 0.01 and 1.0 mg/mL. The increase in Fas expression was statistically significant at all concentrations tested, compared with PBS-treated control samples (P < 0.01 for all concentrations, ANOVA with Bonferroni post-test correction; Figs. 4A 4B ). Peak expression was seen after exposure to 0.4 mg/mL (Fig. 4C) , with an increase in mean expression of approximately 250%. Fas expression decreased at 1.0 mg/mL. 

Tenon fibroblasts did not express the TNF receptor constitutively in culture. There was no difference in MFI obtained with anti-TNF receptor antibody (7.5 ± 0.4, Fig. 5B ) compared with isotype control (7.4 ± 0.4). Furthermore, TNR receptor expression was not induced 48 hours after mitomycin-C (0.4 mg/mL). Positive labeling in PHA-activated T-cells confirmed a functional antibody (Fig. 5A) . 

Fas Receptor Signaling in Mitomycin-C–Treated Fibroblasts

To examine the biological efficacy of anti-Fas antibodies, we used a Fas-sensitive Jurkat T-cell line, which undergoes apoptosis after Fas receptor ligation. ²⁰ Anti-Fas IgM (CH11; Upstate Biotechnology, Lake Placid, NY) induced significant Jurkat T-cell apoptosis compared with the IgM isotype control antibody (data not shown). 

To determine the effect of Fas receptor activation in mitomycin-C–treated Tenon fibroblasts, medium containing anti-Fas IgM (CH11) was added immediately after single 5-minute applications of mitomycin-C (0.4 mg/mL) or PBS in the control fibroblasts. The number of viable fibroblasts was determined by phase-contrast microscopy. Mitomycin-C alone reduced the number of fibroblasts compared with saline-treated control cells by inhibiting proliferation and inducing cell death (Fig. 6A 6B) . In the mitomycin-C–treated fibroblasts, anti-Fas IgM (CH11) led to a median reduction of 43.6% (range, 33.5%–71.1%) in viable fibroblasts compared with mitomycin-C alone (Fig. 6B) . The reduction in fibroblasts was statistically significant for 50 and 500 ng/mL (P < 0.05; Mann-Whitney test). Because antibody-induced killing may have been a consequence of complement-mediated lysis, we repeated the experiment in serum-free conditions with an isotype control antibody that also binds to the fibroblast surface (anti-fibroblast surface antigen IB10; Sigma). In these experiments cell death was measured using the lactate dehydrogenase release assay. Anti-Fas IgM (CH11) led to an approximate 50% increase in fibroblast death compared with mitomycin-C treatment alone (P < 0.001, Mann-Whitney). The isotype control antibody, however, had no effect compared with mitomycin-C alone (Fig. 6C) . A low but consistent increase in cell death was observed in PBS-treated fibroblasts incubated with anti-Fas IgM (<5% at 48 hours). 

Previous dose–response experiments showed that 5-minute applications of mitomycin-C at concentrations of 0.1 mg/mL or less induce a low level of fibroblast apoptosis. ⁸ We were interested in determining the effect of Fas signaling after sublethal treatments with mitomycin-C. Tenon fibroblasts were exposed to mitomycin-C (0.1 mg/mL) for 5 minutes, washed, and incubated with various concentrations of anti-Fas IgM (50 ng/mL) for 48 hours. Anti-Fas IgM did not significantly increase fibroblast death (Fig. 6D , P > 0.5; Mann-Whitney). 

In further support of the nonparticipation of TNF-α, mitomycin-C–treated fibroblasts were incubated in growth medium containing soluble TNF-α. TNF-α had no effect on mitomycin-C–induced fibroblast death compared with mitomycin-C treatment alone (Fig. 5C) . 

Fas Receptor Blockade after Mitomycin-C

To evaluate the efficacy of the Fas-blocking antibody (anti-Fas IgG; M3; Immunotech), Jurkat T-cells were pretreated with the blocking antibody for 30 minutes before the addition of Fas-activating anti-Fas IgM (CH11). Jurkat T-cell death was determined by cell morphology on cytospin preparations. Anti-Fas (M3) inhibited apoptosis (P < 0.01, Mann-Whitney), whereas the isotype control (M33, Immunotech), which binds to the Fas receptor but has no effect on signal transduction, had no effect on anti-Fas IgM–induced apoptosis (Fig. 7A) . To determine whether Fas receptor blockade affects mitomycin-C–induced apoptosis, fibroblasts were incubated in anti-Fas IgG (M3) at various concentrations for 30 minutes before treatment with mitomycin-C (0.4 mg/mL), as previously described. ²¹ After treatment, the monolayers were irrigated and reincubated in Fas-blocking IgG (M3) at the same concentrations. Apoptosis was measured after 48 hours, using the lactate dehydrogenase release assay. Anti-Fas M3 had no effect on mitomycin-C–induced apoptosis at the concentrations tested (Fig. 7B) . 

Discussion

The data presented in this article reveal that Tenon fibroblasts express apoptosis-related gene products constitutively in vitro. The expression of these genes, in particular the cell surface death receptor Fas (CD 95), are altered by clinically relevant treatments of mitomycin-C. Activating the Fas receptor augmented the cytotoxic effects of mitomycin-C but only at concentrations of mitomycin-C that, per se, induce significant fibroblast death. Unlike some forms of chemotherapy-induced apoptosis in tumors, mitomycin-C–induced fibroblast apoptosis was not inhibited by Fas receptor blockade, suggesting that mitomycin-C can induce apoptosis by means independent of Fas receptor signaling. 

Apoptosis is an active gene-directed mode of cell death that plays a critical role in the maintenance of homeostasis. It is activated by numerous triggers, including irreparable internal damage, deprivation of external survival factors, and conflicts in cell-cycle signals that both promote and inhibit cell division. In addition, in mammalian cells a receptor-based mechanism of “instructive” apoptosis has evolved, in which ligation of cell surface death receptors enables an organism to eliminate specific cells. Death receptors play a role in diverse processes, such as T-cell-mediated cytotoxicity, ²³ peripheral T-cell tolerance, ²⁰ and immune privilege, in which Fas-ligand expression by uveal tissue induces apoptosis in activated lymphocytes. ²⁴ The wound-healing response leads to a large expansion in the number of fibroblast that is restored at the end of the healing response by fibroblast apoptosis. ^{9 25 26} The mechanisms that regulate fibroblast death remain poorly understood. 

The Bcl-2 family of proteins regulate the susceptibility of a cell to apoptotic stimuli. The relative ratio of pro- to anti-apoptotic protein dimers regulates cytochrome c release and therefore sets the apoptotic threshold. ²⁷ Tenon fibroblasts constitutively express Bcl-2, Bcl-x, and Bax in culture. Mitomycin-C led to a small increase in protein expression of all the Bcl-2 family members tested, without an obvious shift to a proapoptotic phenotype. This was surprising, in light of the fact that these cells had been exposed to a potent apoptotic signal. 

The cell surface death receptors Fas and the TNF receptor trigger apoptosis after binding their respective ligands. Increased death receptor expression primes cells to stimuli that signal through these receptors. In our experiments, although Tenon fibroblasts constitutively expressed Fas in culture, they did not undergo apoptosis after Fas receptor stimulation. Fas expression increased after mitomycin-C in a dose-dependent manner, up to a concentration of 0.4 mg/mL. Expression was decreased after treatment with 1.0 mg/mL, possibly reflecting the extremely potent apoptotic stimulus and the inability of severely disrupted fibroblasts to maintain protein production. Fas receptor ligation promotes mitomycin-C–induced fibroblast death, confirming that the upregulated receptor is functional. The observation that lower concentrations of mitomycin-C also upregulated Fas expression is exciting, because it suggests that the growth-arrested fibroblasts can be primed to secondary signals that activate the Fas receptor. This would have potential clinical value, because wound-healing responses slowed by the lower, safer intraoperative doses of mitomycin-C may be terminated by subsequent Fas receptor ligation. We found that Fas receptor ligation did not induce significant levels of apoptosis in fibroblasts treated with the lower sublethal concentrations of mitomycin-C. Regulatory mechanisms other than Fas receptor expression must therefore influence the outcome of Fas signaling. A potential regulatory candidate is the Fas ligand inhibitory protein (Flip), ²⁸ which inhibits downstream of the receptor. Our findings are in accord with others that increased Fas receptor expression does not necessarily render cells susceptible to Fas ligand–mediated apoptosis. ²⁸  

Fas receptor blocking antibody (M3) did not inhibit mitomycin-C–induced apoptosis. We excluded the possibility that the antibody does not block Fas signal transduction by demonstrating that anti-Fas IgM–induced apoptosis in Jurkat T-cells was inhibited by the M3 antibody. There is conflicting evidence of the role of Fas in tumor cell death after exposure to chemotherapeutic agents. Fas signaling has been shown to be essential for bleomycin-induced apoptosis in hepatoma cells, ²¹ as well as doxorubicin-induced apoptosis in human leukemia T-cell lines. ¹⁵ The absolute requirement for Fas signaling in all chemotherapy-induced apoptosis has recently been brought to question, however, with evidence that a number of chemotherapeutic agents including doxorubicin can induce Fas-independent apoptosis. ²⁹ Our findings contribute to this, in that Tenon fibroblast apoptosis induced by mitomycin-C can proceed independently of Fas. The presence of a number of possible pathways that can induce apoptosis after severe cell damage has obvious selection advantage. 

Accurate control of subconjunctival scar formation remains an elusive goal in our ability to achieve predictable lowering of intraocular pressure after glaucoma filtration surgery. The potent ability of single, short applications of mitomycin-C to induce apoptosis in fibroblasts may account for the long-term inhibition in scar formation. Devitalizing bleb tissue may also contribute to postoperative complications, however, such as bleb leak and susceptibility to infection. An understanding of the mechanisms that regulate fibroblast death in normal, abnormal, or drug-modified wound-healing responses should permit tighter control of the scarring response and optimize the effect of filtration surgery. In particular, the ability to prime a fibroblast to subsequent death signals provides an attractive approach to switching off the response once sufficient healing has occurred and a target intraocular pressure has been achieved. 

 This work was undertaken by investigators who receive support from the National Health Service (NHS) Executive. The views expressed in the publication are those of the authors and not necessarily of the NHS Executive.

 Supported by Wellcome Trust Grants 045202 (JGC) and 1055183 (LHC) and the Royal National Institute for the Blind (JTD).

 Submitted for publication March 14, 2001; revised August 23, 2001; accepted September 10, 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.

 

Forward- versus side-scatter profiles for human Tenon fibroblasts (A) and an IL-2–dependent T-cell line (B). (A) Forward- (x-axis) and side-scatter (y-axis) profiles for fibroblasts 48 hours after treatment with PBS (left) or mitomycin-C (right). (B) The T-cell profiles are shown for cells in the presence of IL-2 (2 ng/mL; left) or after 48 hours of IL-2 deprivation (right). Region gates were set for live (R1) and apoptotic (R2) cells. Data are the percentage of counts within the gated regions from representative plots derived from one of three separate experiments performed in triplicate.

Figure 2.

 

Bcl-2, Bcl-x, and Bax expression in mitomycin-C–treated (gray filled histograms) and control PBS-treated (open histogram, black solid line) Tenon fibroblasts 48 hours after treatment. The fluorescence intensity is shown on a log scale (x-axis) and represents the level of protein expression. Open histogram, dashed lines: plots obtained with isotype control antibodies for mitomycin-C (gray) and PBS (black) treated fibroblasts. Data are MFIs from representative histograms.

Figure 3.

 

Fas expression in untreated Tenon capsule fibroblasts. Fibroblasts were labeled with anti-Fas FITC or an FITC-conjugated isotype control (IgG) antibody. (A) The fluorescence profile obtained with anti-Fas (filled) and isotype control antibody (dotted line). Tenon capsule fibroblasts constitutively express low levels of Fas. (B) The MFIs were derived from six separate experiments performed in triplicate.

Figure 4.

 

The effect of mitomycin-C concentration on fibroblast Fas expression. Fas expression was assessed in viable (gated) fibroblasts 48 hours after single 5-minute applications of mitomycin-C at the concentrations shown. Fibroblasts were labeled with anti-FITC–conjugated isotype control (A) or Fas-FITC (B). Fluorescence intensity is plotted against mitomycin-C dose (C). Data are the MFIs ± SD derived from a representative of three separate experiments performed in triplicate.

Figure 5.

 

TNF receptor expression and signaling. (A) PHA-activated T-cells maintained in IL-2 (2 ng/mL) were labeled with anti-TNF receptor (filled histogram) or isotype control antibody (open histogram), which served as a positive control. (B) Tenon fibroblasts harvested 48 hours after treatment with mitomycin-C (0.4 mg/mL; filled histogram) or PBS (open histogram, solid gray line) were labeled with anti-TNF receptor or isotype control antibody (dotted lines), followed by an FITC-conjugated second layer. The histograms are representative of one of three separate experiments performed in triplicate. (C) Tenon fibroblasts were treated with mitomycin-C (0.4 mg/mL) for 5 minutes and incubated with TNF-α at the concentrations shown. Fibroblast death was quantified with the lactate dehydrogenase release assay. Data are from a representative of two separate experiments ± SD performed with four replicates per group.

Figure 6.

 

The effect of anti-Fas-IgM on mitomycin-C–treated fibroblasts. (A) Tenon capsule fibroblasts were treated with a 5-minute application of mitomycin-C (0.4 mg/mL) or PBS and incubated for 48 hours with anti-Fas IgM at the concentrations shown. The number of viable fibroblasts was counted from five randomly selected fields at× 40 magnification. Anti-Fas IgM induced reduction in the number of viable fibroblasts in mitomycin-C–treated but not PBS-treated fibroblasts. (B) Data from the same experiment with the number of fibroblasts expressed as a percentage of control (same treatment but no antibody). This representation negates the effect of continued proliferation in PBS-treated fibroblasts. (C) The experiment was repeated in serum-free conditions with an isotype control antibody, which allowed cell death to be measured using lactate dehydrogenase release assays. Fas-activating antibody augmented mitomycin-C–induced cell death. Isotype control antibody had no effect. (D) Tenon fibroblasts treated with a lower concentration of mitomycin-C (0.1 mg/mL) were exposed to different concentrations of anti-Fas IgM. Apoptosis was quantified by measuring lactate dehydrogenase activity in the supernatant. Anti-Fas IgM had no effect on Tenon capsule fibroblasts apoptosis after application of the lower concentration of mitomycin-C. Data are the mean ± SD from a representative of two (A) and three (B) separate experiments performed in quadruplicate.

Figure 7.

 

The effect of Fas-blocking antibody (M3) on Jurkat apoptosis and mitomycin-C–induced fibroblast apoptosis. (A) Jurkat T-cells pretreated with blocking antibody M3 or nonblocking isotype control M33 (both at 10 μg/mL) for 30 minutes before the addition of 100 ng/mL activating anti-Fas IgM (CH11). Cell death was measured with the lactate dehydrogenase release assay Anti-Fas M3 blocked CH11-mediated apoptosis. (B) Tenon fibroblasts were incubated for 30 minutes in medium containing M3 at concentrations shown or in M33 (10 μg/mL) before and after treatment with mitomycin-C (0.4 mg/mL for 5 minutes). Fibroblast death was measured after 48 hours with the lactate dehydrogenase release assay. Blocking antibody M3 did not inhibit mitomycin-C–induced apoptosis. Data are the mean ± SD from a representative of two independent experiments performed in quadruplicate.