|Home | About | Journals | Submit | Contact Us | Français|
Herein, we report that dihydrolipoic acid (DHLA) and lipoic acid (LA) plus lipoamide dehydrogenase and NADH denitrosate S-nitrosocaspase 3 (CASP-SNO). In HepG2 cells, S-nitrosocysteine ethyl ester (SNCEE) impeded the activity of CASP-SH, while a subsequent incubation of the cells in SNCEE-free medium resulted in endogenous denitrosation and reactivation of CASP-SH. The latter process was inhibited in thioredoxin reductase-deficient HepG2 cells, in which, however, LA markedly reactivated CASP-SH. The data obtained are discussed with focus on low molecular mass dithiols that mimic the activity of thioredoxin in reactions of protein S-denitrosation.
Caspases are a family of cysteine proteases that play an essential role in the signaling cascade leading to apoptosis. Apoptosis, or programmed cell death, is distinguished from lytic or necrotic cell death by a number of biochemical and structural events. Apogenic signals trigger specific signaling pathways, including activation of proteases, which are followed by the appearance of specific morphologic changes such as condensation of nuclei and cytoplasm, blebbing of cytoplasmic membranes, and finally fragmentation into apoptotic bodies that are phagocytosed by neighboring cells (1). Apoptosis is important to physiologic processes such as cell selection in development and immunologic responses (2), control of organ size in maturation and regeneration (3), and normal cell turnover throughout the organism (4). Dysregulated apoptosis may contribute to pathologic states such as autoimmune disease (5) and malignancy (6). Upon exposure to a proapoptotic signal, zymogen forms of caspases constitutively present in cells are proteolytically cleaved and activated. Initiator caspases such as caspase 8, 9, and 10 can cleave other caspases, while executioner caspases, including caspases 3 (CASP-SH), 6, and 7, cleave death substrates (7).
Nitric oxide (NO), produced either extracellularly by low molecular mass (LMM) compounds or intracellularly by nitric oxide synthases, impedes CASP-SH activity via reactions of S-nitrosation (8–10). CASP-SH can undergo poly-S-nitrosation, whereby all SNO functions in its p12 subunit are denitrosated by GSH except for a single SNO group (9). Since the latter was not observed in a mutant form of CASP-SH lacking the active site cysteine, Zach et al. proposed that in cells NO nitrosates this cysteine to form S-nitrosocaspase 3 (CASP-SNO) that is resistant to reduction by GSH (9). However, denitrosation of CASP-SNO back to CASP-SH with reconstitution of its proteolytic activity is catalyzed in cellular cytosol by thioredoxin type 1 (Trxn; (11, 12); scheme 1). In mitochondria, thioredoxin type 2 has been found to mediate denitrosation of mitochondria-associated CASP-SH, a process required for caspase-3 activation (13). Trxn contains a -Cys-Gly-Pro-Cys- motif that is essential for its enzyme activity. Cysteines 32 and 35 in the active site of Trxn(SH)2 can reduce SNO functions in substrate proteins with concomitant disulfide ring closure to Trxn(S)2 and release of nitroxyl (HNO; (12, 14, 15)). In turn, Trxn(S)2 is converted back to Trxn(SH)2 by the NADPH-dependent thioredoxin reductase (TrxnR; (12, 16)).
Sustained production of endogenous NO has been shown to decrease both the expression and the activities of Trxn and thioredoxin reductase (TrxnR; (17–19). This suggests that shifts in the rates of S-nitrosation and denitrosation of CASP-SH following changes in either NO production or activity of the Trxn/TrxnR/NADPH system will modulate CASP-SH-dependent apoptotic pathways. Within the scope of this mechanism, we were interested to verify whether LMM dithiols can mimic the activity of Trxn toward CASP-SNO. To this end, we have focused on dihydrolipoic acid (DHLA; 6,8-dimercaptooctanoic acid), which is the reduced form of lipoic acid (LA; 5-[1,2]dithiolan-3-yl-pentanoic acid), a cyclic disulfide that is an essential prosthetic group in the dihydrolipoyl transacetylase component of the •-ketoacid dehydrogenase complex in mitochondria. In cells, LA (Scheme 2, 1) is reduced to DHLA (2) by lipoamide dehydrogenase (LD) with consumption of NADH; in addition, reduction of LA to DHLA is catalyzed by NADPH-dependent reductases (including TrxnR; (20, 21)). Herein, we present experimental evidence that DHLA fully denitrosates CASP-SNO in chemical systems and regenerates the activity of CASP-SH in TrxnR-deficient HepG2 cells exposed to NO.
All reagents used were purchased from Sigma Chemical Co. (St. Louis, MO). Human RhoA protein was purchased from Cytoskeleton, Inc. (Denver, CO). The solutions used in the experiments were prepared in deionized and Chelex-100-treated water or potassium phosphate buffer.
S-nitrosation of the ethyl ester of cysteine (HS-Cys-OEt) was carried out with nitrosooxy ethane (C2H5ONO) as described previously (22). Briefly, HS-Cys-OEt (0.5 g) was dissolved in methanol (5 mL) containing C2H5ONO (0.5 mL; b.p. 13 °C) and the reaction solutions was incubated for 30 min in ice. Thereafter, the solvent and the remaining C2H5ONO were rotor-evaporated (2 mm Hg) at room temperature and the nitrosothiol formed was recrystallized from methanol. The purity of the nitrosothiols was assessed by UV spectrophotometry (in methanol, •(343)SNCEE = 1019 M−1cm−1 and •(544)SNCEE = 36 M−1cm−1 (22). During the preparation of SNCEE, care must be exercised in handling solutions containing C2H5ONO, as inhalation of its vapor may cause severe headache and heart excitation. The preparation must therefore be conducted in an efficient fume cupboard.
S-nitrosation of CASP-SH (3 •M) was performed with GSNO (0.3 mM) at 20 °C for 30 min in 0.1 M phosphate buffer containing 0.2 mM EDTA. Thereafter, CASP-SNO was separated from GSH and the excess of GSNO via ultrafiltration (3 kDa Vivaspin cut-off filter; 30 min at 12,000g), which included 4 washing cycles with 0.2 mL of the reaction buffer.
S-nitrosocaspase 3 was quantified following its Cu+-catalyzed breakdown to CASP-SH and NO with concomitant chemiluminescence measurements of the latter in the gas-phase using a Sievers Nitric Oxide analyzer (NOA ™ 280i; Boulder, CO) (11). The purge vessel of the NO analyzer was filled with 5 mL of 0.1 M phosphate buffer (pH 7.4; 20 °C; gas carrier, He). In the reaction vessel, a steady-state concentration of Cu+ was maintained by a large excess of ascorbic acid over CuCl2 (50 mM vs. 0.2 mM). Thus, multiple injections of aliquots (5 •L) containing CASP-SNO could be made without any significant loss of analytical sensitivity. Under these experimental conditions, NaNO2 (up to 0.1 mM) does not interfere with the analysis of CASP-SNO (11).
In cell lysates, CASP-SH activity was measured fluorometrically on a LS50B Perkin Elmer spectrofluorimeter (•ex = 380 nm; (•em = 420 nm; excitation/emission slit, 5 nm) using 0.15 mM Asp-Glu-Val-Asp-7-amido-4-methyl-coumarin as a substrate (Sigma, Inc. St. Louis, MO).
Human hepatoma (HepG2) cells were cultured in the Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 100 U/mL penicillin, 100 µg/mL streptomycin, and 2 mM l-glutamine in a humidified atmosphere in 5% CO2 at 37°C.
CASP-SH activation in HepG2 cells - Cells (2 × 106 cells/flask) were grown in T75 flasks for 24 hours and then incubated for 4.5 hours with medium (control) or with medium containing TNF-• (15 nM) and cycloheximide (40 µM; Sigma; St. Louis, MO; (23)).Thereafter, the cells were washed with PBS (3 × 10 mL) and incubated for 15 min at 37 °C with standard incubation medium containing SNCEE. Whole-cell lysates for analysis of CASP-SH were harvested by repeated freeze and thaw cycles followed by centrifugation at 15,000g for 15 min at 4°C.
The siRNA directed against TrxnR was 5'-AGACCACGUUACUUGGGCAdTdT-3' and the control was a scrambled sequence (5'-AGGCAAAUCACGGUGUCCUdT dT-3') that does not match any sequence in the GenBank human database for >16 nt ((24); Dharmacon RNA Technologies; Chicago, IL). Approximately 2 × 105 HepG2 cells were plated per well in a six-well plate. The following day, siRNA were transfected with lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), with 30 pmoles siRNA per well (25), according to the manufacturer’s recommendations. Transfection with the same amount of nonspecific siRNA was performed as control. The cells were harvested and analyzed at 24, 48 and 72 hours after transfection for cell viability, activity, and level of TrxnR protein.
Cells were disrupted by three consecutive freeze and thaw cycles and then centrifuged at 15,000 g for 15 min at 4 °C to remove membrane fractions. Equal amounts of protein (20–30 µg) were resolved by SDS PAGE (10%) and transferred on to a nitrocellulose membrane. The membrane was blocked in 5% (w/v) dried milk powder in TBS (Tris-buffered saline) with (v/v) 0.01% Tween-20 at room temperature and was incubated with anti-TrxR1 (Rabbit IgG, Upstate) (1:500 dilution) overnight in 5% (w/v) BSA in TBS with 0.01% (v/v) Tween-20. Secondary antibody (HRP-conjugated anti-rabbit IgG; 1:1,000 dilution; Pierce; Rockforf, IL) was then added to the membrane for 1 hour at room temperature, washed, and thereafter the membrane was exposed to HRP substrate (Pierce SuperSignal West Pico Chemiluminescent Substrate) and visualized by chemiluminescence on autoradiography film.
TrxnR activity was determined in a coupled assay with E. coli Trxn (10 •M) and 5,5•-Dithiobis(2-nitrobenzoic acid) (DTNB) as described in ref . (11). One unit of TrxnR activity was defined as the formation of 74 •moles of 5-mercapto-2-nitro-benzoic acid (1 absorbance unit at 412 nm; •412 = 13500 M−1.cm−1) per min per mL at pH 7.0 at 25 °C.
Results are given as mean ± S. E. (n = 3 – 6).
LMM compounds containing vicinal dithiols such as threo-1,4-dimercapto-2,3-butanediol (Cleland’s reagent; DTT) and DHLA react with RSNO to form RSH and disulfides with generation of HNO, NH3, NH2OH and NO. In this reaction, the ratio between end products depends on the presence of nitrate anion and the degree of S-nitrosation of the parent dithiols. Mono-S-nitroso thiols tend to release HNO (12, 14, 26), whereas di-S-nitrosothiols undergo cyclization to disulfides with release of NO (14, 27).
At mM concentrations, DTT and both LA and DHLA have been shown to cause toxicity via induction of reductive stress (28) and activation of CASP-SH, respectively ((29) and the references therein). However, pharmacological experiments are often carried out with 10 mg of LA per kg of body weight (30, 31), whereas therapeutic results in humans are achieved at plasma concentrations of 10–20 •g/mL (~50–100 µM) LA (31, 32). Hence, we were interested to verify whether CASP-SNO can be denitrosated by pharmacological concentrations DHLA (Scheme 2).
Purified CASP-SNO, which contains 5 SNO functions per mol of protein (12), exhibited insignificant proteolytic activity and did not decompose to any significant extent in the presence of LA or LD plus NADH (Figure 1a,c). However, both DHLA and the complete LA/LD/NADH system denitrosated CASP-SNO (Fig. 1b,d) with concomitant reconstitution of the enzyme activity of CASP-SH (Fig. 2). Notably, CASP-SNO was fully denitrosated by DHLA. In contrast to LA, a series of LMM disulfides have been shown to inhibit CASP-SH via formation of protein mixed disulfides (33, 34).
To assess the potential of DHLA to denitrosate CASP-SNO in intact cells, we have conducted experiments with control and TrxnR-deficient HepG2 cells. In cells, TrxnR was silenced using siRNA, while release of CASP-SH in the cellular cytosol was attained with TNF- and cycloheximide (11). In contrast to hepatocytes, HepG2 cells are deficient in some phase I enzymes (35–37), including alcohol dehydrogenase class III (ADH; also referenced as GSNO reductase), which catalyses the denitrosation of GSNO (38). Hence, control and TrxnR-deficient HepG2 cells offered the advantage to study Trxn-catalyzed reactions of S-denitrosation without interference of ADH.
Incubation of HepG2 cells for 72 hours with siRNA (30 nM) caused a ~ 50% decrease in the activity of TrxnR (Figure 3A) without affecting the levels of intracellular GSH to any significant extent (data not shown). Western blot analysis established that, in siRNA-treated cells, the protein levels of TrxnR had markedly decreased (Figure 3A, inset). For S-nitrosation of thiols, cells were incubated with SNCEE, which readily crosses cell membranes and trans-S-nitrosates intracellular proteins (11, 12, 22). While at concentrations of 0.2 – 0.8 mM SNCEE impedes the activity of TrxnR and causes toxicity in HepG2 cells (11), experiments aimed to assess the nitrosative inactivation of CASP-SH were carried out with 50 µM SNCEE.
Exposure of HepG2 cells to cycloheximide and TNF- resulted in a dose- and time-dependent activation of CASP-SH, whereby maximal activity of the protease was attained after and incubation for 4.5 hours with 15 nM TNF- (Fig. 3B). Thereafter, control and TrxnR-deficient HepG2 cells were incubated for 15 min with 50 µM SNCEE. The cells were then washed with PBS and the activity of CASP-SH was assessed either immediately (time for preparation of reaction solutions for spectral analysis, 5 min) or after an additional incubation for 60 min at 37 °C in SNCEE-free incubation medium.
The rationale for this experimental design was based on previously established kinetics of denitrosation of CASP-SNO by the Trxn/TrxnR/NADPH system (11), with the hypothesis that time-dependent increases in CASP-SH activity would reflect the rates of endogenous reactions of S-denitrosation. Treatment of control HepG2 cells with SNCEE led to ~ 80 % (5 min) and 5 % (60 min in SNCEE-free incubation medium) inhibition of CASP-SH (Figure 4A). The regeneration of CASP-SH activity was insignificant in TrxnR-deficient cells (Figure 4B), which suggests the requirement for Trxn catalysis in this process. However, substitution of SNCEE with LA followed by incubation for 60 min led to a marked reactivation of CASP-SH activity in TrxnR-deficient HepG2 cells (Fig. 4B), presumably via intracellular reduction of LA to DHLA and reaction of the latter with CASP-SNO. Notably, CASP-SH activity was reconstituted more efficiently by 50 •M LA than by 5 mM DTT (Fig. 4B). In this set of experiments, qualitatively the same effects were observed when SNCEE was substituted for S-nitroso-N-acetyl-DL-penicillamine (SNAP; incubation time, 1 hour; data not shown).
The data presented herein provide a proof of concept that LMM dithiols can mimic the activity of Trxn in reactions of protein S-denitrosation. Several disease states have been associated with both increased S-nitrosation of proteins and modulation of the homeostasis of Trxn. Examples of such diseases are liver steatosis and cirrhosis (39–41), rheumatoid arthritis (42–45), bronchopulmonary dysplasia (46–48), and asthma (49, 50). Hence, DHLA and structurally similar LMM disulfides/dithiols could be instrumental in assessments of the involvement and specific roles of reactions of S-nitrosation. It could be further hypothesized that denitrosation of S-nitrosothiols by LMM dithiols may be used to counteract the toxicity of NO and drugs that act as S-nitrosating agents, such as glyceryl trinitrate, nitrosooxyethane and nitrosoaspirine (22, 51–53). This is an experimentally verifiable hypothesis for which support has not previously been generated. Hence, further studies are needed to assess the structure/activity relationship of the reduction of LMM disulfides to dithiols by NADH-and NADPH-dependent reductases, as well as to define the factors that affect the rates of protein S-denitrosation by LMM dithiols.
This work is supported by NIH grant GM044100 and Walter Reed Army Institute of Research grants W81XWH-09-P-0631 and W81XWH-06-1-0247.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.