Abstract
The nuclear transcription factor-κB (NF-κB) and free radicals are known to be involved in apoptosis. We studied the effects of a series of di-aryl-substituted pyrazole NF-κB inhibitors including tepoxalin on tumor necrosis factor α (TNFα)-induced apoptosis in murine fibrosarcoma WEHI 164 cells. We found that potent inhibitors of NF-κB were also effective in attenuating apoptosis. WEHI 164 cells that had been dually treated with tepoxalin and the antioxidant pyrrolidine dithiocarbamate (PDTC) were significantly protected from TNFα-induced killing. To study the role of free radicals in mediating TNFα-induced apoptosis, stable WEHI 164 cells overexpressing Bcl-2, an antioxidant protein, were generated. These cells were protected from TNFα-induced apoptosis and neither tepoxalin nor PDTC provided further significant protection. These results suggest that Bcl-2, PDTC, and tepoxalin may attenuate apoptosis in this system by affecting the same signaling pathway or converging pathways. Because tepoxalin suppresses the release of free radicals, PDTC scavenges free radicals and Bcl-2 is an antioxidant protein, free radicals are among the key mediators of this TNF-induced killing event. Tepoxalin and antioxidants may be useful in developing new therapeutics for treating neurodegenerative diseases, autoimmune deficiency syndrome, and ischemia-reperfusion injuries.
Apoptosis is known to be involved in embryonic development and maintenance of hemostasis in adult tissues, reflecting a crucial balance between cell death and cell proliferation (Korsmeyer, 1995; Vaux and Strasser, 1996). This is exemplified during thymic ontogeny by positive and negative selection processes that allow selection of functionally active immune cells that are directed at foreign antigens. Likewise, the nervous system is continuously being fine-tuned by selecting for those neurons that respond to specific nerve growth factors that are produced in different environments.
To study the complex signaling pathways leading to apoptosis, both biological and chemical inhibitors have been used. Biological inhibitors such as the proto-oncogene Bcl-2 (Kroemer, 1998) has provided clues into the various mechanisms whereby cells can be protected from apoptosis. Chemicals including inhibitors of interleukin-1β-converting enzyme (Nemeth et al., 1997), kinases (Coleman et al., 1997), proteases (Higuchi et al., 1995), and antioxidants (Huang et al., 1998), have also been shown to attenuate apoptosis. Thus, oxidative stress and proteolysis may be critically involved in apoptosis (Buttke and Sandstrom, 1994; Wood and Youle, 1994; Powis et al., 1995). In particular, stimulation of the 55 kD tumor necrosis factor α (TNFα) receptor results in a rapid rise in the level of intracellular free radicals whereas overexpression of antioxidant proteins such as Bcl-2, superoxide dismutase, catalase, or glutathione peroxidase attenuates apoptosis (Hirose et al., 1993;Sandstrom et al., 1994; Vaux and Strasser, 1996; Kroemer, 1998).
Several chemical inhibitors of apoptosis including glucocorticoid analogs and inhibitors of phospholipase A2, cyclooxygenase (CO), or lipoxygenase (LO) only showed partial inhibition of apoptosis (Suffys et al., 1987; Hockenbery et al., 1993; Metodiewa et al., 1998). We have previously reported a series of di-aryl-substituted pyrazole compounds, originally identified as dual CO/5-LO inhibitors (Argentieri et al., 1994) to be potent inhibitors of nuclear factor-κB (NF-κB) mediated transactivation (Kazmi et al., 1995). This series of compounds, including tepoxalin, also blocked free radical generation by inhibiting the peroxidase (PO) function of prostaglandin synthase 1 (PGHS1; Tam et al., 1995). Here we show that tepoxalin, together with a free radical scavenging antioxidant such as pyrrolidine dithiocarbamate (PDTC), effectively reduced TNFα-induced apoptosis in the murine fibrosarcoma WEHI 164.
Materials and Methods
General chemicals were obtained from Sigma (St. Louis, MO) and cell culture reagents were obtained from Gibco BRL (Gaithersburg, MD).
Establishment of WEHI 164 Cells Overexpressing Murine Bcl-2.
The murine fibrosarcoma cell line, WEHI-164, was obtained from American Type Culture Collection (Rockville, MD). The cells were maintained in RPMI 1640 medium containing 10% fetal calf serum (FCS) and antibiotics at 37°C in 5% CO2. Full length murine Bcl-2cDNA (Veis et al., 1993) was inserted into the mammalian expression vector pcDNA3 (InVitrogen, Carlsbad, CA). The recombinant plasmid was introduced into wild-type WEHI 164 cells using the Superfect transfection reagent (Qiagen, Valencia, CA) following the procedures as described by the vendor. Stable transfectants were selected in culture medium supplemented with 1 mg/ml Geneticin (Gibco BRL, Gaithersburg, MD). Individual clones were obtained by standard limiting dilution. Overexpression of Bcl-2 in these clones was detected by Western blot analysis of whole cell lysates prepared from 2 × 105 cells. Blots were first probed with 1 mg/ml hamster anti-mouse Bcl-2 (PharMingen, San Diego, CA) and then with horseradish peroxidase-conjugated goat anti-hamster IgG (Jackson Laboratory, Bar Harbor, ME). Chemiluminescence detection was performed using an ECL kit (Amersham Pharmacia, Piscataway, NJ). Clones were then tested for their resistance to murine TNFα (R&D Systems, Minneapolis, MN).
Enzyme-Linked Immunosorbent Assay (ELISA) for DNA Fragmentation.
DNA fragmentation was measured using a cell death detection ELISA kit (Boehringer Mannheim, Indianapolis, IN) according to vendor’s specifications. Briefly, WEHI 164 cells (5 × 105) were incubated at 37°C for 5 h with 100 pg/ml TNFα in the presence of inhibitors. Cells were washed, lysed, and the nuclei were removed by centrifugation at 100g for 10 min. The resulting supernatant was diluted 1:10 and incubated in 96-well plates coated with anti-histone monoclonal antibody for 90 min. The wells were washed and incubated for an additional 90 min with anti-DNA antibody conjugated with peroxidase. Nucleosome fragments were detected using ABTS substrate solution and the optical density read at 410 nm on a ThermoMax plate reader (Molecular Devices, Sunnyvale, CA).
Apoptotic Cell Morphology.
WEHI 164 cells were incubated for 5 h in the presence of 100 pg/ml TNFα and inhibitors. Photographs were then taken using TMX100 film (Eastman Kodak, Rochester, NY) in a WILD Photoautomat MPS55 camera system attached to a Leitz Labovert FS inverted microscope.
51Chromium Release Assay.
TNFα-mediated apoptosis was quantified by measuring 51Cr released from labeled WEHI 164 cells. WEHI 164 cells (1 × 106) were labeled with 100 μCi sodium (51Cr) chromate (Amersham Pharmacia, Piscataway, NJ) for 1 h at 37°C. After washing twice with PBS to remove excess radioactivity, labeled cells were resuspended in RPMI 1640 medium containing 10% FCS at a concentration of 5 × 104 cells/ml. Cells (5 × 103) were dispensed into each well of a round bottom 96-well plate. TNFα and inhibitors were added to the appropriate wells. The cells were incubated for 18 h at 37°C after which cell-free supernatants were collected and analyzed by gamma counting in a LKB Clini-gamma counter (EG&G Wallac, Gaithersburg, MD). 51Cr release was calculated as a percent of the 51Cr release in test wells minus background wells with identical drug treatments but in the absence of TNFα divided by total 51Cr uptake.
MTT Assay.
Quantification of TNFα-mediated cytotoxicity was also performed using a CELLTITRE 96 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay kit (Promega, Madison, WI) involving the reduction of tetrazolium dye to formazan by mitochondrial enzymes. Briefly, WEHI 164 cells (2 × 104/well) in RPMI-1640 medium containing 10% FCS were plated in flat bottom 96-well plates in the presence of increasing concentrations of TNFα and in the presence or absence of inhibitors. The cells were incubated for 18 h at 37°C after which 15 μl of MTT dye was added to each well. After a 4-h incubation at 37°C, 100 μl of solubilization/stop solution was added and incubated at 37°C for an additional 1 h. The optical density was read at 570 nm and percent viability was calculated as a percentage of the OD570 in wells with identical drug treatments but in the absence of TNFα.
Electrophoretic Mobility Shift Assay.
After 30 min of pretreatment with the appropriate agents and 1-h stimulation with 100 pg/ml TNFα, WEHI 164 cells (2 × 107) were washed twice in ice-cold PBS and nuclear extracts were prepared as described previously (Kazmi et al., 1995) DNA binding was carried out in a total volume of 20 μl using 5 μg of nuclear protein in a reaction mixture consisting of 10 mM Tris-HCl, pH 7.0, 2 mM EDTA, 1 mM DTT, 0.05% NP-40, 0.25 mg/ml BSA, 0.1 mg/ml poly-deoxyinosine deoxycytosine (Gibco BRL, Gaithersburg, MD) and approximately 2.5 ng/ml32P-labeled oligonucleotides. The reaction mixtures were incubated for 20 min at room temperature and then subjected to electrophoresis through a 4% polyacrylamide gel. Electrophoresis was carried out in 1× Tris-borate-EDTA buffer at 170 mV, 30 mA for 4 h. Gels were dried and autoradiographed for 12 to 16 h at −70°C.
For NF-κB binding, oligonucleotides derived from the NF-κB sequence of the murine κ chain gene were used (Kazmi et al., 1995):
Results
Effects of Tepoxalin and Analogs on TNFα-Induced Apoptosis.
The WEHI 164 fibrosarcoma cell line is extremely sensitive to TNFα, even in the picogram/milliliter range. TNFα induces apoptosis, as measured by both 51Cr release and cell viability, in almost 100% of the WEHI cells over a 24-h incubation period. Tepoxalin, when added simultaneously with TNFα, suppressed apoptosis (Fig.1). We have demonstrated previously that tepoxalin is a potent inhibitor of NF-κB. Thus, a series of tepoxalin analogs were evaluated for their ability to block both TNFα-induced NF-κB transactivation and apoptosis. Table1 shows that there is a direct correlation between the suppression of NF-κB activation and apoptosis. Compounds that are the most potent in suppressing NF-κB activation are also the most effective in blocking apoptosis.
We have previously shown that tepoxalin inhibits both the CO and PO functions of PGHS1 (Tam et al., 1995). Although the CO function of PGHS1 is the rate-limiting step in the generation of prostaglandins, the PO is involved in both prostaglandin synthesis and free radical generation. Tepoxalin, but not other PGHS1 inhibitors such as naproxen or indomethacine (which inhibits only the CO function), is an inhibitor of free radical generation (Tam et al., 1995) and the subsequent NF-κB transactivation (Kazmi et al., 1995). Structure-activity relationship studies also indicate that, although the di-aryl-substituted pyrazole structure of tepoxalin is responsible for inhibiting prostaglandin synthesis, the hydroxamic acid moiety is important for suppressing both the LO activity and free radical generation (Tam et al., 1995). In this report, we have evaluated additional analogs of tepoxalin. Clearly the hydroxamic side chain is important for blocking apoptosis because removal of the hydroxamic acid by carboxylic acid (RWJ-20142, Table 1), acetic acid, or hydroxylamine (data not shown) totally abrogated the suppressive effects on apoptosis. A chlorophenyl modification on the hydroxamic side chain (RWJ-23696) enhanced the ability to suppress both NF-κB and apoptosis by ∼2-fold, possibly by restricting the rotation of the side chain. Interestingly, when the chlorophenyl group was substituted with a biphenyl (RWJ-24159), instead of enhancement, complete loss of activity was observed. A thiocarbamate analog (RWJ-25856) also conferred protection from apoptosis and inhibited NF-κB with slightly improved efficacy over the other analogs. Pure CO inhibitors such as naproxen or LO inhibitors such as zileuton had no effect on apoptosis or NF-κB transactivation (IC50 > 50 μM, data not shown). Moreover, tepoxalin, RWJ-25856, and RWJ-23696 are also potent dual inhibitors of both the CO and PO functions of the PGHS1 whereas RWJ 20142 is a pure CO inhibitor (Tam et al., 1995).
Tepoxalin Synergizes with PDTC in Inhibiting NF-κB and Apoptosis.
It has been shown previously that antioxidants such as PDTC and N-acetyl cysteine inhibit TNFα-induced NF-κB (Schreck et al., 1992). Because tepoxalin has minimal antioxidant properties (data not shown), the possible additive or synergistic interaction of tepoxalin with the antioxidant PDTC was evaluated by the dual administration of drugs in cultures treated with TNFα for 18 h. In this system, >90% of the cells were committed to apoptosis within 6 to 8 h after TNFα treatment, only a small fraction between 5 to 8% of the cells died of necrosis in the 18-h treatment period. Tepoxalin combined with a suboptimal dose of PDTC (10 μM) provided >90% protection from TNFα-induced apoptosis in WEHI 164 cells (Fig. 1). This combination of drugs did not affect cell viability when administered to the cells without TNFα treatment (<5% cell death after 18 h). Furthermore, the drugs must be given to the cells either before or immediately after TNFα administration to achieve maximal efficacy. The protective effect is comparable to that of an optimal dose of 100 μM PDTC (data not shown). However, increased toxicity was observed in cells treated with 100 μM of PDTC for 16 h, indicating that the use of PDTC at high doses may be undesirable. Electromobility gel shift studies using nuclear extracts from TNFα-treated WEHI 164 cells and NF-κB specific oligonucleotides, indicated that although tepoxalin alone reduced NF-κB binding, almost all NF-κB activity was abolished upon dual administration with PDTC (Fig. 2). Similar protective effects against TNF-α-induced apoptosis and NF-κB activity could be obtained when N-acetyl cysteine [10 μM] substituted for PDTC and used in combination with tepoxalin; however, increased cellular toxicity (35% cell death after 18 h) was observed (data not shown). Other antioxidants such as vitamin E and ascorbic acid failed to display any protective effects against cell death in this system.
Effect of Tepoxalin and PDTC on DNA Fragmentation and Cell Morphology.
Five hours after treatment with TNFα and inhibitors, WEHI 164 cells were examined for both DNA fragmentation and cytoarchitectural patterns. Although untreated WEHI 164 cells exhibited normal cellular morphology (Fig. 3a), TNFα-treated cells exhibited traits typical of apoptosis. Blebs and membranous apoptotic bodies, cell shrinkage, and condensation of nucleolus were evident in about 50% of the cells (Fig. 3b). Tepoxalin or a low dose of PDTC, when administered separately, only resulted in a slight reduction of detectable apoptotic cells (Fig. 3, c and d). When tepoxalin and PDTC were administered together, a dramatic reduction of apoptotic cells was observed (Fig. 3e).
DNA fragmentation, as assayed by cell death detection ELISA, which provides a semiquantitative assay for apoptosis, demonstrated similar results (Fig. 4). Considerable protection from apoptosis was observed when tepoxalin and PDTC were dually administered whereas only partial protection was detected when the drugs were given separately.
Bcl-2-Expressing WEHI 164 Clone Is Resistant to TNFα-Induced Apoptosis.
WEHI 164 cells were stably transfected with murine Bcl-2. Transfectants that survive Geneticin selection were subcloned and several clonal lines were further evaluated. Western blots probed with anti-murine Bcl-2 indicated that these lines expressed very high levels of Bcl-2 compared with the wild-type or the mock transfected cells (Fig.5a). These WEHI 164 cells that over-expressed Bcl-2 were about 500-fold more resistant to TNFα-induced killing than the wild-type or mock transfected cells. Interestingly, neither tepoxalin nor PDTC alone could confer significant additional protection (Fig. 5b). In agreement with a previous report (Herrmann et al., 1997), we also failed to detect any significant changes in the NF-κB activity of these Bcl-2transfectants when compared with the wild-type controls (data not shown).
Discussion
Murine fibrosarcoma WEHI 164 cells have been shown to be highly sensitive to human and murine monocyte-macrophage-induced killing (Herberman and Holden, 1978). Although macrophages can mediate the extracellular killing of nucleated cells via diverse mechanisms, it has been demonstrated that WEHI 164 cells are killed by a monocyte toxin that can be neutralized by antibodies to TNFα (Bersani et al., 1986). The mechanisms involved in such TNFα-mediated apoptosis include free radicals (Brekke et al., 1994), caspases, and proteolysis (Kumar et al., 1997; Jaeschke et al., 1998).
We have previously shown that tepoxalin, a di-aryl-substituted pyrazole with a hydroxamic side chain, inhibits NF-κB transactivation (Kazmi et al., 1995). The suppressive effect of tepoxalin on NF-κB was further demonstrated by its ability to inhibit the transcription of NF-κB-regulated genes such as IL-2 (Zhou et al., 1994), IL-6 (Kazmi et al., 1995), IL-8, or adhesion molecules (Lee et al., 1996). The hydroxamate portion of tepoxalin is essential for NF-κB inhibition because removal of the hydroxamic acid leaving the carboxylic acid abolished all inhibitory activity on NF-κB-mediated transactivation (Kazmi et al., 1995). Using tepoxalin and its structurally related analogs, we attempted to study their possible effects on apoptosis and identify an equivalent biological process that these chemicals may be mimicking.
Tepoxalin and two of its structural analogs, RWJ-23696 and RWJ-25856, potently blocked apoptosis. Previously, these compounds had been shown to block free radical release by inhibiting the PO function of PGHS1. Both RWJ-20142 and RWJ-24159, inhibitors of the CO function of PGHS1, potently blocked prostaglandin E2 production upon TNFα treatment without affecting either apoptosis or NF-κB binding (Munroe et al., 1995). These data suggest that a correlation between free radical generation by the PGHS1 and apoptosis clearly exists. It has also been shown that, although butylated hydroxyanisole effectively prevented TNFα-induced apoptosis of WEHI 164 cells, butylated hydroxytoluene did not, despite the fact that both are equally potent antioxidants (Brekke et al., 1994). This discrepancy may be a result of the fact that butylated hydroxyanisole, but not butylated hydroxytoluene, is an inhibitor of the PO activity of PGHS1 as shown earlier (Vanderhoek and Lands, 1973).
Although tepoxalin is a potent inhibitor of free radical generated by PGHS1, coadministration with a suboptimal dose of the antioxidant PDTC dramatically amplifies its ability to attenuate apoptosis. Clearly, the PO of the PGHS1 is not the only source of free radical production and addition of a free radical scavenger such as PDTC enhances the efficacy of protection. However, because almost 100% protection can be achieved by dual administration of tepoxalin and PDTC, this suggests that free radicals are the predominant molecules leading to cell death in TNFα-induced apoptosis in WEHI 164 cells.
WEHI 164 cells stably transfected with the murine proto-oncogeneBcl-2 were highly resistant to TNFα-induced apoptosis. Almost 1000-fold higher concentrations of TNFα were required to achieve the same extent of apoptosis when comparing the overexpressingBcl-2 cells with the wild-type cells. Interestingly, although the combined regimen further protected the Bcl-2 transfectants, neither tepoxalin alone nor PDTC alone offered any additional protection. Thus it appears that the sites of intervention for tepoxalin, PDTC, or Bcl-2 may be on different foci along converging pathways or a common vertical signal transduction pathway. Tissue distribution studies indicate that Bcl-2 is present on mitochondrial membranes where most of the oxidases involved in electron transport reside, and also on the endoplasmic reticulum and nuclear membrane where PGHS1 and the other isoform prostaglandin synthase 2 (PGHS2) are exclusively located. In wild-type WEHI 164 cells, Bcl-2 levels are very low and free radicals produced by peroxidases such as PGHS1 and PGHS2 could trigger the apoptotic signal (Munroe and Lau, 1995). This signal can be blocked by tepoxalin, which inhibits PGHS1 peroxidase and even more effectively when administered together with another antioxidant to scavenge any remaining free radicals. The recent reports that showed both Bcl-xL or Bcl-2 can exert an antiapoptotic function in cells by affecting caspases activation (Jaeschke et al., 1998; Srinivasan et al., 1998) and that Bcl-2 suppresses apoptosis in prostatic carcinoma cells could be attributed to disruption of the NF-κB pathway (Herrmann et al., 1997) provides an alternate mechanism to explain how antioxidants may protect against apoptosis.
The relationship between NF-κB and apoptosis, however, remains obscure. It has been demonstrated that NF-κB inhibited TNF-induced apoptosis (Van Antwerp et al., 1998) and that inhibition of NF-κB nuclear translocation enhanced apoptotic killing (Wang et al., 1996). Interestingly, we demonstrated that tepoxalin and its analogs, which are inhibitors of NF-κB, potentiate the effects of PDTC, in attenuating apoptosis in the WEHI 164 system. These discrepancies suggest that in different biological systems, apoptosis may be governed by distinct signaling pathways, or more likely, in the WEHI 164 system, NF-κB induction and apoptosis are separate pathways being triggered by a common initiating sequence of events. The free radical generated from sources such as PGHS1 represents one such event. Nuclear translocation of p65, NF-κB transactivation or IκB degradation studies would better define the exact role of NF-κB in this apoptosis system. The combination of tepoxalin and antioxidants may be useful in designing new therapies for treating autoimmune deficiency syndrome (Ameisen et al., 1995) and chronic neurodegenerative disorders (Mattson, 1998; Akama et al., 1998) where specific cell types are lost by apoptosis via different mechanisms.
Footnotes
-
Send reprint requests to: Dr. Daniel H.S. Lee, Rm 345, Research Building, The R.W. Johnson Pharmaceutical Research Institute, McKean & Welsh Roads, Spring House, PA 19477-0776. E-mail:dlee{at}prius.jnj.com
- Abbreviations:
- NF-κB
- nuclear factor-κB
- TNFα
- tumor necrosis factor α
- PDTC
- pyrrolidine dithiocarbamate
- CO
- cyclooxygenase
- LO
- lipoxygenase
- PO
- peroxidase
- PGHS1
- prostaglandin synthase 1
- MTT
- 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
- ELISA
- enzyme-linked immunosorbent assay
- FCS
- fetal calf serum
- Received September 10, 1998.
- Accepted January 8, 1999.
- The American Society for Pharmacology and Experimental Therapeutics