The dual role of S-nitrosoglutathione (GSNO) during thymocyte apoptosis

ISSN 0898-6568/96 $15.00 SSDI 0898-6568(95)02051-9

Cell. Signal. Vol. 8, No. 3, pp. 173-177, 1996 Copyright © 1996 Elsevier Science Inc. ELSEVIER

The Dual Role of S-nitrosoglutathione (GSNO) During Thymocyte Apoptosis Katrin Sandau and Bernhard Br~ine* UNIVERSITYOF ERLANGEN-NORNBERG,FACULTYOF MEDICINE,

DEPARTMENTOF MEDICINEIV-ExPERIMENTALDIVISION,91054 ERLANGEN,GERMANY

ABSTRACT. Nitric oxide (NO) is known to regulate redox-sensitive signalling pathways in physiology and pathophysiology. Depending on its concentration, the NO-releasing compound S-nitrosoglutathione (GSNO) causes negative and positive regulation of thymocyte apoptosis. At levels below 0.6 mM, GSNO produces deoxyribonucleic acid (DNA) laddering, which is inhibited by activation ofprotein kinase C (PKC), cycloheximide treatment, and calcium chelation. Higher concentrations of the NO donor (1-2 mM) suppress thymocyte apoptosis initiated by the classical agonist dexamethasone. Inhibition of apoptosis by NO is analogous to the action of the thiol-blocking compound N-ethylmaleimide (NEM) and the glutathione-S-transferase substrate 1-chloro-2,4-dinitrobenzene (CDNB). Inhibition of apoptosis results from thiol modification of critical proteins in response to NO treatment. Depending on the concentration, GSNO can be involved either in toxic or in protective signalling in thymocyte biology. CELLSIGNAL8;3:173-177, 1996. KEY WORDS. Nitric oxide, Thymocytes, Apoptosis, Redox regulation, Thiol groups

INTRODUCTION

of regulatory genes, the critical biochemical events during apoptosis remain unclear. Oxidative stress, referring to the generation of free radicals, treatment with peroxides, or generation of reactive oxygen species, induces apoptosis. In contrast, antioxidants and Bcl-2 largely confer resistance to physiological cell death [11-13]. Furthermore, it has been demonstrated that oxidation of cellular sulfhydryl groups initiates the cellular suicide program [14], while targeting of a critical thiol group, which is essential for apoptosis, inhibits deoxyribonucleic acid (DNA) cleavage [15]. Oxidative reactions and thiol modification are feasible for NO, since the chemistry of NO is determined by its given biological milieu. With the notion that the N O donor S-nitrosoglutathione (GSNO) gives rise to radical formation after homolytic cleavage of the nitrogen-sulfur bond, but also produces a thioltargeting nitrosonium ion (NO +) after heterolytic decomposition, we studied the effect of GSNO in rat thymocytes, a wellcharacterized model system for apoptosis [ 16]. Our results imply a dual role of GSNO during thymocyte apoptosis.

Nitric oxide (NO) is endogenously produced by a family of constitutive or inducible NO-synthase (NOS) enzymes [1], or is released from compounds generally termed N O donors. These agents are valuable tools for studying the physiology, pathophysiology, and pharmacology of NO, irrespective of NOS involvement [2]. In biological systems, oxygen, superoxide, and transition metals are prime targets of NO, while the cytotoxic capacity of N O comprises interactions with thiol-containing or redox-sensitive, metal-containing proteins [3]. Mechanistically, the generation of peroxynitrite (ONOO-), inhibition of Fe-S enzymes, deregulation of poly adenosine diphosphate (ADP)ribosyl transferase, and energy depletion is considered a likely scenario for cell death [4, 5]. Although in some systems N O effectivelycauses cell death by necrosis, in others the progressive intra- or extracellular generation of NO initiates apoptosis, accompanied by p53 expression [6]. Apoptosis is a morphologically distinct form of cell death common to most multicellular organisms. It is characterized by cell shrinkage, membrane blebbing, chromatin condensation, and deoxyribonucleic acid (DNA) fragmentation [7]. Despite its initiation by a variety of stimuli [8-10], and the identification

Materials

*Author to whom correspondenceshould be addressed. Abbreviations: GSNO-S-nitrosoglutathione;EGTA-ethylene glycol bis (13-aminoethylether)-N,N,N',N'-tetraaceticacid;BAPTA-AM- 1,2-bis-(2-aminophenoxy)ethane-N,N,N',N'tetraaceticacid tetra(acetoxymethyt)ester; NEMN-ethylmaleimide;CDNB- 1-chloro-2,4-dinitrobenzene;TPA-12-O-tetradecanoylphorbol-13-acetate,GSH-reduced glutathione. Received27 August; and accepted6 November 1995.

Dexamethasone, 12-O-tetradecanoylphorbol-13-acetate (TPA), cycloheximide, ethylene glycol bis([~ aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), N-ethylmaleimide, and diphenylamine were purchased from Sigma, Deisenhofen, Germany. 1,2-Bis-(2-aminophenoxy)ethane-N,N,N',N',-tetraacetic acid tetra (acetoxymethyl) ester) (BAPTA-AM), and 1-chloro-2,4-

MATERIALS AND METHODS

174 dinitrobenzene came from Calbiochem (Bad Soden, Germany. RPMI 1640 culture medium, glutamine, medium supplements, fetal calf serum (FCS), and agarose were obtained from Gibco, (Berlin, Germany). All other chemicals were of the highest grade of purity commercially available.

Preparation of thymocytes Suspension ofthymocytes from 3-wk-old male Sprague-Dawley rats was achieved as described previously [ 17]. Thymocytes (107 cells/ml) were suspended in RPMI 1640 medium supplemented with 2 mM glutamine, 100 U / m L penicillin, I00 ~tg/mLstreptomycin, and 10% heat-inactivated FCS. Cells were kept at 37°C in a humidified incubator in an atmosphere of 5% CO2 (for overnight incubations). Short-term incubation was done in a water bath at 37°C.

GSNO synthesis S-nitrosoglutathione was synthesized as described previously [18]. Briefly, glutathione was dissolved in 0.625 N HCI at 4°C to a final concentration of 625 mM. An equimolar amount of NaNO2 was added and the mixture was stirred for 40 min. After the addition of 2.5 vol of acetone, stirring was continued for another 20 rain, followed by filtration of the precipitate. GSNO was washed once with 80% acetone, twice with 100% acetone, and finally thrice with diethyl ether, and dried under vacuum. Freshly synthesized GSNO was characterized by highperformance liquid chromatographic (HPLC) analysis and ultraviolet (UV) spectroscopy.

Quantitation of DNA fragmentation DNA fragmentation was assayed as reported [17]. Briefly, following incubations, cells were centrifuged, resuspended in 250 gL TE buffer (I0 mM Tris-HC1 and 1 mM ethylenediamine tetraacetic acid [EDTA], pH 8.0), and lysed by adding 250 IxL cold lysis buffer containing 2 mM EDTA, 0.5% (v/v) Triton X-100, and 5 mM Tris-HCl, pH 8.0. Samples were allowed to lyse for 30 min at 4°C prior to centrifugation (15 min at 14,000 g) to separate intact chromatin (pellet) from DNA fragments (supernatant). Pellets were resuspended in 500 ~tLTE buffer and the DNA contents of pellets and supernatants were measured using the diphenylamine reagent.

DNA agarose gel.electrophoresis Thymocytes were lysed and centrifuged as described in order to separate DNA fragments from intact chromatin. DNA fragments in the supernatants were precipitated overnight with 1 mL 100% ethanol and 50 btL 5 M NaCI at - 2 0 ° C . After centrifugation (14,000 g for 15 min), pellets were incubated in 500 ~tLTE buffer containing 100 ~tg/mLribonuclease A (RNase A) at 37°C for 30 min, extracted with phenol/chloroform/ isoamylalcohol (25:24:1), and again precipitated overnight at - 20°C. DNA fragments were electrophoretically separated on a 1% agarose gel and visualized by ultraviolet fluorescence after staining of the agarose gel with ethidium bromide (1 ~tg/mL).

Sandau and Brune

Glutathione determination Thymocytes were centrifuged (14,000 g for 15 s) and lysed for 15 rain in 100 ~tL 5% sulfosalicylic acid. Reduced glutathione was separated by centrifugation (10,000 g for 10 min) and analyzed by reversed-phase HPLC following precolumn derivatization with the sulfhydryl-reactive dye monobromobimane. Sixty microliters of the supernatant containing reduced glutathione was incubated at room temperature in the clark with 25 btL 12.5 % N-ethylmorpholine, supplemented with 2 mM monobromobimane. After 15 min, 10 ~tL of 0.37 MHCIO4 was added and a 10-1*L aliquot of the solution was analyzed by HPLC separation, as described [19]. A 150 × 4.6-mm column packed with 3-1~m particles of nucleosil-ODS (Cls) was used at ambient temperature, at a flow rate of 1.6 mL/min, for the HPLC separation. Elution solvent A consisted of 9% acetonitrile and 0.25% acetic acid in distilled water, pH 3.9; solvent B was 0.25% acetic acid, pH 3.9. The elution profile was as follows: 0-2 min, 10% solvent B isocratic; 2-7 min, 10% solvent B, linear gradient. Following separation, the column was washed for 5 min with 75% acetonitrile to remove late-eluting fluorescent material. Elution of the monobromobimane derivative of reduced glutathione (GSH) was at 5.5 min. A fluorescence detector (Merck-Hitachi F 1100 Merck, Darmstadt, Germany), operated at an excitation wavelength of 380 nm and an emission wavelength of 480 nm, was used for analyte quantitation.

Statistical analysis Statistical analysis was performed with the paired Student's t-test. Values statistically different from controls (P > 0.05) are marked by asterisks (*). RESULTS AND DISCUSSION NO-donor-induced apoptosis in rat thymocytes was quantified using the diphenylamine assay. The 20% spontaneous fragmentation observed during a 16-h incubation was in good agreement with literature data. GSNO up to 0.8 mM caused further fragmentation to values around 45% (Fig. 1A). With higher GSNO concentrations (0.8 mM-2 mM) the apoptotic rate declined to values below initial rates. This is not only obvious with the diphenylamine assay, but also became apparent when investigating DNA ladder formation (Fig. 1C). Notably, at higher concentrations of GSNO, no necrotic marker (i.e., trypan blue uptake) was measurable. Two other NO donors, sodium nitroprusside (SNP) and S-nitroso-N-acetylpenicillamine (SNAP), have been successfully used to cause thymocyte apoptosis, with maximal fragmentation of about 30% and 35%, respectively (data not shown). However, the specific bell-shaped dose-response curve for DNA fragmentation was unique for GSNO. Internucleosomal DNA cleavage (DN A laddering) is considered a hallmark of apoptosis. We investigated DNA laddering in response to 0.6 mM GSNO and dexamethasone, a classical thymocyte apoptosis inducing agent (Fig. 1B). In controls, DNA ladder formation was virtually absent, while dexamethasone and GSNO produced substantial

Nitric Oxide and Thymocytes A

O

175

50

40

//

\i

30

20 z

lO

I

I

I

I

I

0.0

0.5

1.0

1.5

2.0

GSN0

[mM]

e FIGURE 1. GSNO- and dexamethasone-induced DNA fragmentation in rat thymocytes. (A) GSNO-induced thymocyte DNA fragmentation. Thymocytes (10 v ceUs/mL) were incubated for 16 h with different GSNO concentrations. Fragmentation (mean values _+ S.E.M. of at least 3 separate experiments) was assayed with the diphenylamine assay. For details, see Materials and Methods. (B) DNA laddering in thymocytes. Thymocytes (2 x 10 ¢ cells/mL) were incubated with 100 nM dexamethasone (Dex) or 0.6 mM GSNO for 16 h, followed by DNA fragmentation analysis. Results are representative of 3 similar assays. Experimental details are given in the Materials and Methods section. (C) Low- and high-degree GSNO-induced DNA laddering in thymocytes. Thymocytes were incubated with 0.6 and 2 mM GSNO for 16 h, followed by agarose gel fragmentation analysis. Details are the same as in (B).

fragmentation, with G S N O being less active than dexamethasone. GSNO-activating pathways were probed for a role of protein kinase C (PKC), m acromolecular synthesis, and calcium C A 2÷. Activation of PKC by 100 nM TPA drastically reduced GSNO-induced apoptosis (Table 1). Whereas PKC-mediated protection for agonists such as dexamethasone is a subject of controversy [ 16], our experiments imply reduced fr agmentation due to PKC activation in response to GSNO. In accord with findings reported in the literature [16], we found that inhibitors of macromolecular synthesis suppress thymocyte apoptosis. As indicated in Table 1, a concentration of l0 ~tM cycloheximide reduced G S N O (0.6 mM)-induced D N A fragmentation to much lower values. This substantiates the role of protein synthesis during thymocyte apoptosis. Calcium is a key cation because internucleosomal cleavage appears to be Ca2+/Mg2+-dependent. Various reports describe chelation of extracellular Ca 2÷ by EGTA and buffering of intracellular Ca 2÷ by BAPTA-AM as preventing apoptosis [16]. A1-

though a recent study [20] questioned an early increase in intracellular Ca 2+ for glucocorticoid-activated apoptosis in thymocytes, our results corroborate a role for C a 2+ in the late or terminal D N A degradation pathway in response to GSNO. Data given in Table 1 indicate that fragmentation by G S N O is reduced in the presence of 5 mM E G T A / 5 0 liM BAPTA.AM. In aggregate, NO-induced thymocyte apoptosis is suppressed by activation of PKC, calcium chelation, and inhibition of protein biosynthesis. In seeking an explanation for the bell-shaped dependency of D N A fragmentation on G S N O concentration we concentrated on thiol modification. Thiol modification is relevant to N O signalling [3] as it is to D N A cleavage [14, 15]. Two established mechanisms were employed: G S H depletion and thiol alkylation. The glutathione-S-transferase substrate 1-chloro-2,4dinitrobenzene (CDNB) either depletes GSH or alkylates thioredoxin reductase [21], whereas N-ethylmaleimide (NEM) alkylates sulfydryl groups. D N A fragmentation induced by

Sandau and Brfine

176 TABLE 1. Signalling pathways in relation to dexamethasoneand S-nitrosoglutathione-induced thymocyte apoptosis Control

Dex 100 nM

100

GSNO 0.6 mM

80 + + + +

TPA, 100 nM CHX, 10 ~tM EGTA, 5 mM BAPTA-AM, 50 ~tM

20% 24% 29% 18%

± ± ± ±

4% 3% 6% 5%

77% 80% 42% 17%

± ± ± -+

6% 5% 2%* 5%*

44% 20% 26% 2%

± ± ± ±

6% 3%* 3%* 1.5%*

TABLE 2. Sulfhydryl modulation during dexamethasone- and S-nitrosoglutathione-induced thymocyte apoptosis Control

Dex 100 nM

GSNO 0.6 mM

20% _+ 4% 13% .+ 5% 19% _+ 2%

77% -+ 8% 45% _+ 8%* 13% _+ 8%*

44% .+ 6% 27% .+ 2%* 17% .+ 8%*

Thymocytes (107 cells/mL) were treated with 100 nM dexamethasone (Dex) or 0.6 mM S-nitrosoglutathione (GSNO) in combination with 5 I~M 1-chloro2,4-dinitrobenzene (CDNB) or 10 BM N-ethylmaleimide (NEM) for 16 h. Fragmentation was quantitated with the diphenylamine assay. Results are mean values ± S.D. from at least 3 separate experiments. Asterisks (*) mark statistically significantly different values.

60

40 Z N

dexamethasone and G S N O was significantly reduced by coincubation with 5 ~tM CDNB or 10 ~tM NEM (Table 2). At these concentrations of the thiol-modifying agents, a direct cell-damaging effect as measured by trypan blue exclusion can be eliminated. CDNB and NEM largely reversed or totally inhibited D N A fragmentation without causing cell necrosis. Comparably, GSNO-induced D N A laddering was abolished by CDNB (data not shown). These results are in accord with inhibitory data for Cd 2÷, Hg 2÷, dichloroisocoumarin, and NEM in studies of high-molecular-weight D N A fragmentation in isolated rat liver nuclei in response to Mg 2÷ and Ca 2÷ [15]. Thiol modification and inhibition of dexamethasone-induced apoptosis can be envisioned, since D N A binding of the activated glucocorticoid receptor requires reduced cysteine residues [22]. Our data suggest a thiol-group requirement for receptor independent (i.e., NO-mediated) apoptotic signalling in thymocytes. It seems likely that critical thiol groups are involved in a distal step leading to endonuclease activation or suppression of the enzyme itself. Establishing CDNB and NEM as potent inhibitors of GSNO- and dexamethasone-initiated thymocyte apoptosis, we concentrated on the suppressive potency at elevated GSNO concentrations. Considering the potential action of GSNO on intracellular thiols [3, 23], we measured the level of reduced GSH after NO-donor addition. In control thymocytes we measured 1620 _+ 250 pmol GSH/107 cells (mean + S.D., n = 3).

\\\\\"

\\\\~

O .a.-q

qD

Thymocytes (107 cells/ml) were co-stimulated for 16 h with dexamethasone (Dex, 100 nM), GSNO (0.6 mM), or vehicle in the presence or absence of 100 nM TPA, 10 gM cycloheximide (CHX), and a combination of 5 mM EGTA plus 50 ~tM BAPTA-AM. D N A fragmentation was quantitated with the diphenylamine assay. Results are mean values -+ S.D. from at least 3 different experiments. Statistically significantly different values are marked by asterisks (*).

+ CDNB, 5 ~tM + NEM, 10 p.M

-

20

N

0

FIGURE 2. Inhibition ofdexamethasone-induced DNA cleavage by GSNO. Thymocytes (10 7 ceUs/mL)were incubated for 16 h with dexamethasone (Dex; 100 nM) and combinations of dexametha. sone with 0.6 mM and 2 mM GSNO, respectively. DNA fragmentation was quantitated with the diphenylamine assay. Details are given in Figure 1. Results are expressed as mean values _+S.E.M., n~3.

With 0.6 mM GSNO, this decreased to 630 + 120 pmol G S H / 107 cells, while 2 mM G S N O further reduced this to 210 + 80 pmol GSH/107 cells (mean _+ S.D., n = 3). Because our results suggested thymocyte thiol modification in response to GSNO, we addressed the co-stimulatory efficacy of dexamethasone and GSNO. Figure 2 shows D N A fragmentation of around 80% induced by dexamethasone. With the addition of GSNO, the dexamethasone response was sequentially suppressed. By analogy with CDNB and NEM, GSNO down-regulates dexamethasone-elicited responses. In addition to the initiation of apoptosis, the modulation of apoptosis involves an additional NO-regulatory pathway, and must be considered for NOproducing cellular systems. The demonstration of such paradoxical actions is characteristic of systems sensitive to redox regulation. This is exemplified by neurons, in which N O can be either toxic or protective, depending on the chemistry it undergoes [24]. NO+-driven S-nitrosylation will inhibit the N-methyl-D-aspartate (NMDA) receptor, while the formation

Nitric Oxide and Thymocytes of peroxynitrite will damage neurons. By analogy, N O is a potent inducer of macrophage apoptosis [6], whereas B-cell apoptosis is inhibited by a still controversial mechanism [25, 26]. Redox modulation may affect apoptosis in a more general way, considering transcription factors such as AP-1 and NF-KB, which are activated during thymocyte apoptosis [27]. Whereas redox events are associated with initial activation, massive oxidation or thiol blocking may hinder their D N A binding and transcriptional activation [28]. Opposing actions of N O may be relevant for thymocyte biology. T h e protective effect of N O against thymocyte apoptosis can be reproduced by treatment of the cells with thiol-modulating agents. T h e N O + chemistry related to S-nitrosylation is an established mechanism for enzyme inhibition and may participate in protection. In contrast, induction of D N A fragmentation may be associated with reactive nitrogen oxide formed during the reaction of N O with oxygen or superoxide. We would like to thank the Deutsche Forschungsgemeinschafi and the European Community for their support.

References I. Marletta M. A. (1994) Cell 78, 927-930. 2. Moncada S., Palmer R. M. J. and I-Iiggs E. A. (1991) Pharmacol. Rev. 43, 109-142. 3. Stamler J. S. (1994) Cell 78, 931-936. 4. Zhang J., Dawson V. L., Dawson T. M. and Snyder S. H. (1994) Science 263, 687-689. 5. Kolb H. and Kr6ncke K-D. (1993) Diabetes Rev. 1, 116-126. 6. Met~mer U. K., Ankarcrona M., Nicotera P. and Brfine B. (1994) FEBS Lett. 355, 23-26. 7. Arends M. J. and Wyllie A. H. (1991) Int. Rev. Exp. Pathol. 32, 223-254. 8. Schwartzman R. A. and Cidlowski J. A. (1993) Endoc. Rev. 14, 133-151. 9. McConkey D. J., Nicotera P. and Orrenius S. (1994) Immunol. Rev. 142, 343-363.

177 10. Corcoran G. B., Fix L., Jones D. P., Moslen M. T., Nicotera P., Oberhammer F. A. and Buttyan R. (1994) Toxicol. Appl. Pharmacol. 128, 169-181. 11. Wolfe J. T., Ross D. and Cohen G. M. (1994) FEBS Lett. 352, 58-62. 12. Slater A. F. G., Nobel S. I., Maellaro E., Bustamante J., Kimland M. and Orrenius S. (1995) Biochem. J. 306, 771-778. 13. Buttke T. M. and Sandstrom P. A. (1994) Immunol. Today 15, 7-10. 14. Sato N., Iwata S., Nakamura K., Hori T., Mori K. and Yodoi J. (1995) J. Immunol. 154, 3194-3203. 15. Cain K., Inayat-Hussain S. H., Kokileva L. and Cohen G. M. (1995) FEBS Lett. 358, 255-261. 16. Kizaki H. and Tadakuma T. (1993) Microbiol. Immunol. 37,917925. 17. McConkey D. J., Nicotera P., Hartzell P., Bellomo G., Wyllie A. H. and Orrenius S. (1989) Arch. Biochem. Biophys. 269, 365-370. 18. Hart T. W. (1985) Tetrahedron Lett. 26, 2013-2016. 19. Svardal A. M., Mansoor M. and Ueland P. M. (1990) Anal. Biochem. 184, 338-346. 20. Iseki R., Kudo Y. and lwata M. (1993) J. Immunol. 151, 51985207. 21. Arner E. S. J., Bj6rnstedt M. and Holmgren A. (1995) J. Biol. Chem. 270, 3479-3482. 22. Esposito F., Cuccovillo F., Morra F. and Cimino F. (199 5) Biochim. Biophys. Acta 1260, 308-314. 23. Clancy R. M., Levartovsky D., Leszcynska-Piziak J., Yegudin J. and Abramson S. B. (1994) Proc. Natl. Acad. Sci. USA 91, 36803684. 24. Lipton S. A., Choi Y. B., Pan Z. H., Lei S. Z., Chen H. S., Sucher N. J., Loscalzo J., Singl D. J. and Stamler J. S. (1993) Nature 364, 626-632. 25. Mannick J. B., Asano K., Izumi K., Kieff E. and Stamler J. S. (1994) Cell 79, 1137-1146. 26. Genaro A. M., Hortelano S., Alvarez A., Martinez-A. C. and Bosca L. (1995) J. Clin. Invest. 95, 1884-1890. 27. Sikora E., Grassilli E., Radziszewska E., Bellesia E., Barbieri D. and Franceschi C. (1993) Biochem. Biophys. Res. Commun. 197, 709-715. 28. Dr6ge W., Schulze-Osthoff K., Mihm S., Galter D., Schenk H., Eck H-P., Roth S. and Gm~inder H. (1994) FASEB J. 8, 11311138.