- Email: [email protected]
Triosephosphates are toxic to superoxide dismutase-deficient Escherichia coli
Biochimica et Biophysica Acta 1622 (2003) 128 – 132 www.bba-direct.com
Triosephosphates are toxic to superoxide dismutase-deficient Escherichia coli Ludmil Benov *, Anees F. Beema, Fatima Sequeira Department of Biochemistry, Faculty of Medicine, Kuwait University, P.O. Box 24923 Safat, 13110, Kuwait Received 6 February 2003; received in revised form 2 June 2003; accepted 18 June 2003
Abstract Increase in the production of triosephosphates has been considered an important factor leading to diabetic complications. It might be expected that like the other short chain monosaccharides, triosephosphates autoxidize producing superoxide radical and a,h-diketones. Since superoxide can also initiate the oxidation of short chain sugars, free radical chain reactions are possible. If such reactions occur in vivo, triosephosphates would be more deleterious to cells lacking superoxide dismutase (SOD) than to normal cells. Here we demonstrate that triosephosphates kill a SOD-deficient Escherichia coli mutant much more than the parental, SOD-proficient strain. The effect is oxygendependent and is partially suppressed by aminoguanidine. Increased production of superoxide and diketones appeared to be the cause of triosephosphates toxicity. D 2003 Elsevier B.V. All rights reserved. Keywords: Dihydroxyacetone phosphate; D-Glyceraldehyde-3-phospate; Superoxide dismutase; Triosephosphates autoxidation; Methylglyoxal; Aminoguanidine
1. Introduction Dihydroxyacetone phosphate (DHAP) and D-glyceraldehyde-3-phospate (GA3P) are glycolytic intermediates whose steady state concentrations normally are very low. Increase of the triosephosphates production is found in diabetes and seems to be among the causes of diabetic complications [1,2]. One reason for the low intracellular concentrations of these glycolytic intermediates seems to be their intrinsic instability. DHAP and GA3P undergo spontaneous degradation to methylglyoxal (MG) [3,4], a compound that is highly toxic[5]. It might be expected that like the other short chain monosaccharides, triosephosphates autoxidize in air, producing superoxide radical and a corresponding diketone [6]. Since superoxide can also initiate the oxidation of such compounds, a free radical chain oxidation is possible [7]. Thus, oxidative stress would enhance triosephosphates degradation, and as a consequence, formation of toxic diketones. On the other hand, a
* Corresponding author. Tel.: +965-531-9489; fax: +965-533-8908. E-mail address: [email protected] (L. Benov). 0304-4165/03/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0304-4165(03)00134-X
rise in triosephosphates concentration might lead to increased production of superoxide. We demonstrate herewith that superoxide dismutase (SOD)-deficient Escherichia coli lose viability if incubated with triosephosphates. The reason appeared to be cell damage resulting from increased production of superoxide and dicarbonyls.
2. Materials and methods DL-Glyceraldehyde-3-phosphate (50 mg/ml in water) and DHAP, dilithium salt, were obtained form Sigma. Yeast extract, Bacto-tryptone and casamino acids were from Difco. LB medium contained 10 g Bacto-tryptone, 5 g yeast extract, and 10 g NaCl per liter and was adjusted to pH 7.0 with f 1.5 g of K2HPO4. M9CA medium consisted of minimal A salts [8], 0.2% casamino acids, 0.2% glucose, 3 mg pantothenate, and 5 mg of thiamine per liter. The strains of E. coli used were as follows: GC4468 = parent; QC1799 = GC4468 DsodA3, DsodB-kan [9]. Strains were grown overnight at 37 jC, with shaking in air, in LB medium containing 50 Ag/ml kanamycin where indicated. The overnight cultures were diluted 200-fold
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
into M9CA medium not containing antibiotics, and were grown to A600 nm = f 0.5. The cells were washed twice and resuspended to a density of A600 nm = 0.1 in M9 salts containing 0.2% glucose. Freshly prepared filter-sterilized GA3P or DHAP solutions were added and suspensions were incubated for 2 h before enumeration of surviving cells. At the end of the incubation, A600 nm was measured again and suspensions were suitably diluted and plated on LB plates for counting colonies. Plates were incubated aerobically at 37 jC for 24 – 48 h prior to counting. For the anaerobic experiments, cell suspensions were preincubated 15 min in sealed tubes full to the top. Before being sealed, and after additions, the cultures were bubbled with Ar for 5 min. These cells were plated on LB agar containing 0.2% glucose and plates were incubated anaerobically. For assaying glyceraldehyde-3-phosphate dehydrogenase (GA3PD), cells incubated with GA3P or DHAP as described above, were washed two times in ice-cold 50 mM Tris –HCl, pH 7.5, resuspended in the same buffer and lysed in a French press. The extracts were clarified by centrifugation and then assayed for protein [10], and GA3PD [11]. The GA3PD assay was performed in Tris –HCl, pH 7.5, and cysteine was omitted from the assay mixture to avoid reactivation of the enzyme. MG was assayed by using 1,2-diaminobenzene as derivatizing reagent, according to the procedure of Cordeiro and Ponces Freire [12]. HPLC analysis was performed in a Shimadzu 10ADM VP liquid chromatograph equipped with CLASS-VP Chromatography data system software. The column was Shim-pack VP-ODC 4.6 250 with a STR-2 ODS-2 precolumn. The mobile phase was 40% (v/v) 25 mM ammonium formate buffer, pH 3.4, and 60% (v/v) methanol. A sample volume of 100 Al was injected. Mutagenesis was monitored by assaying the frequency of thymine-negative (Thy ) mutants. Thy mutants are resistant to the drug trimethoprim and can be selected from a Thy+ population [13]. The assay was performed as described by Farr et al. [14]. For counting Thy mutants, aliquots of the cultures were plated directly on LB plates containing thymine (50 Ag/ml) and trimethoprim (15 Ag/ ml). For enumeration of cells, cultures were suitably diluted and plated on LB plates containing thymine, but not trimethoprim. Experiments were repeated at least three times. Figures represent mean F S.E.
129
Fig. 1. Triosephosphates induce mutations and kill the SOD-deficient E. coli. Panel A, mutagenesis—Overnight LB cultures of GC4468 (SODproficient) and QC1799 (SOD-deficient) were diluted 200-fold in M9CA medium and were grown aerobically at 37 jC and 200 rpm to a density of A600 nm f 0.2 – 0.3. At this point, GA3P or DHAP were added and the cells were kept for 2 h on the shaker. After the incubation, the cells were diluted and plated on LB plates containing thymine for assessing cell number or aliquots were plated without dilution on LB plates containing thymine and trimethoprim for counting Thy mutants. Panel B, loss of viability—Mid log cultures of parental and SOD-deficient E. coli grown in M9CA medium were diluted to A600 nm = 0.1 and were incubated 2 h with 5.0 mM GA3P or DHAP, or with 0.5 mM MG. After the incubation, suspensions were diluted and plated on LB agar for counting colonies.
the Thy mutants in the parental culture, while for the sodAsodB culture, the increase was 4 –6-fold. At higher concentration (5.0 mM), GA3P and DHAP caused a loss of viability of E. coli (Fig. 1B). Both compounds were more toxic for the sodAsodB than for the SOD-competent parental strain. Since the mutant differs from the parent only by the absence of cytoplasmic superoxide dismutases, these results suggest that superoxide plays a key role in triosephosphatesinduced cell damage.
3. Results 3.1. Effect of oxygen GA3P and DHAP are mutagens and kill the SODdeficient strain-Incubation with 2.0 mM GA3P or DHAP for 2 h increased the number of the Thy mutations in both parental and sodAsodB strains (Fig. 1A). There was a significant difference, however, between the two strains. Thus, GA3P and DHAP caused less than 2-fold increase of
If the main reason for the higher sensitivity of the sodAsodB strain toward GA3P and DHAP is higher superoxide steady state concentration, then anaerobiosis should diminish triosephosphates toxicity. As seen, incubating the cells with GA3P or DHAP in anaerobic atmo-
130
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
sphere significantly increased the viability of the sodAsodB cells (Fig. 2). Triosephosphates, like other short chain sugars, autoxidize in air producing O2S and a dicarbonyl compound, hydroxypyruvaldehydephosphate [6]. Dicarbonyls can modify proteins and nucleic acids and are toxic and mutagenic [5,15 –17]. Methylglyoxal (pyruvaldehyde), an analogue of hydroxypyruvaldehydephosphate, causes loss of viability at a concentration 10 times lower than that of GA3P or DHAP (Fig. 1B). In this case, however, no significant difference between the sensitivity of the parental and the sodAsodB strain was observed. 3.2. Accumulation of dicarbonyls Superoxide has been shown to serve as both an initiator and a propagator of the short chain sugars autoxidation [7]. One possible explanation for the role of the O2S in triosephosphates toxicity is that superoxide promotes the conversion of triosephosphates to diketones, which actually kill the cells. Analysis of the intracellular dicarbonyl content revealed that after incubation with GA3P or DHAP, the SOD-deficient cells contained more MG than did parental cells (Fig. 3). Technical difficulties did not allow us to determine the intracellular content of hydroxypyruvaldehydephosphate, the immediate product of triosephosphates oxidative degradation. 3.3. Glyceraldehyde-3-phosphate dehydrogenase MG is produced during the non-oxidative decomposition of GA3P and DHAP [3], and therefore its production should not be directly affected by the intracellular [O2S ]. One way superoxide might increase MG production, however, is by inactivating GA3PD [18,19]. This would prevent the me-
Fig. 2. Effect of oxygen on GA3P and DHAP toxicity. Mid log suspensions of parental and SOD-deficient E. coli were diluted to A600 nm = 0.1 and were incubated 2 h aerobically or anaerobically with 5.0 mM GA3P or DHAP. After the incubation, suspensions were diluted and plated on LB agar for counting colonies. Bars: (1) parental, aerobic; (2) parental, anaerobic; (3) sodAsodB, aerobic; (4) sodAsodB, anaerobic.
Fig. 3. Accumulation of MG. All conditions are as in Fig. 1B except that after the incubation the cells were washed, lysed, and cell-free extracts were assayed for MG.
tabolism of triosephosphates via the glycolytic pathway and would lead to formation of MG. We found that after incubation with GA3P or DHAP, the SOD-deficient mutant contained about half the GA3PD activity of the parental cells incubated at the same conditions (0.066 F 0.019 Amol/ min/mg protein in the sodAsodB against 0.139 F 0.026 Amol/min/mg protein in the parental strain). 3.4. Effect of aminoguanidine If triosephosphates toxicity is entirely due to production of dicarbonyls, then scavenging of dicarbonyls with aminoguanidine should prevent triosephosphates-induced cell death. Fig. 4 shows that in the DHAP-treated parental strain 100 AM aminoguanidine increased the survival rate to f 90% of the control, and 500 AM provided practically full protection. In the sodAsodB cells, 500 AM aminoguanidine doubled the survival rate, but nevertheless could not raise it above 65%. Almost the same results were
Fig. 4. Effect of aminoguanidine. All conditions are as in Fig. 1B. Aminoguanidine was added to final concentrations of 100 and 500 AM and DHAP was 5.0 mM.
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
obtained about the effect of aminoguanidine on GA3Ptreated cells (not shown).
4. Discussion Excessive production of advanced glycation end products (AGE) appears to be important in the pathogenesis of diabetic complications. Recent data show that hyperglycemia increases intracellular AGE primarily by increasing the concentration of AGE-forming MG [20]. It is generally accepted that MG is formed by non-oxidative mechanism, primarily by beta-elimination of phosphate from triosephosphate intermediates in anaerobic glycolysis [4,5]. In diabet e s , h y p e rg l y c e m i a - i n d u c e d o v e r p r o d u c t i o n o f mitochondrial superoxide reversibly inhibits GA3PD, which increases triosephosphate levels [21,22]. Here we demonstrate that in cells lacking SOD, increase of triosephosphates concentration has deleterious consequences. Because their carbonyl groups cannot be blocked by cyclization, triosephosphates are susceptible to autoxidation yielding superoxide and hydroxypyruvaldehydephosphate [6]. As a consequence, GA3PD is inactivated, leading to accumulation of another toxic dicarbonyl compound, methylglyoxal. Each of these products, superoxide and dicarbonyls, contributes to triosephosphates-induced cell damage. Superoxide, through the Fenton reaction, promotes the formation of hydroxyl radicals, which indiscriminately attack all cellular components. Dicarbonyls can interact with, and modify many different compounds in the cell including proteins and nucleic acids [5,23], thus inactivating enzymes and causing mutations [5,24]. Since superoxide is both an initiator and a product of sugars autoxidation, a vicious cycle might be envisaged in which overproduction of superoxide increases triosephosphates levels and degradation, which in turn causes more superoxide production and release of toxic dicarbonyls. It thus appears that among the important functions of SOD is to prevent the conversion of unstable glycolytic intermediates into highly reactive, toxic compounds.
Acknowledgements This work was supported by grant MB031 from Kuwait University. We are grateful to Danielle Touati (Institute Jacques Monod, CNRS, University Paris, France) for providing the E. coli strains used herein.
References [1] P.J. Thornalley, I. Jahan, R. Ng, Suppression of the accumulation of triosephosphates and increased formation of methylglyoxal in human red blood cells during hyperglycaemia by thiamine in vitro, J. Biochem. 129 (2001) 543 – 549.
131
[2] Y. Hamada, J. Nakamura, K. Naruse, T. Komori, K. Kato, Y. Kasuya, R. Nagai, S. Horiuchi, N. Hotta, Epalrestat, an aldose reductase ihibitor, reduces the levels of Nepsilon-(carboxymethyl)lysine protein adducts and their precursors in erythrocytes from diabetic patients, Diabetes Care 23 (2000) 1539 – 1544. [3] S.A. Phillips, P.J. Thornalley, The formation of methylglyoxal from triose phosphates. Investigation using a specific assay for methylglyoxal, Eur. J. Biochem. 212 (1993) 101 – 105. [4] J.P. Richard, Mechanism for the formation of methylglyoxal from triosephosphates, Biochem. Soc. Trans. 21 (1993) 549 – 553. [5] P.J. Thornalley, Pharmacology of methylglyoxal: formation, modification of proteins and nucleic acids, and enzymatic detoxification—a role in pathogenesis and antiproliferative chemotherapy, Gen. Pharmacol. 27 (1996) 565 – 573. [6] P. Thornalley, S. Wolff, J. Crabbe, A. Stern, The autoxidation of glyceraldehyde and other simple monosaccharides under physiological conditions catalysed by buffer ions, Biochim. Biophys. Acta 797 (1984) 276 – 287. [7] T. Mashino, I. Fridovich, Superoxide radical initiates the autoxidation of dihydroxyacetone, Arch. Biochem. Biophys. 254 (1987) 547 – 551. [8] T. Maniatis, E.F. Fritsch, J. Sambrook, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor, New York, 1982. [9] D. Touati, M. Jacques, B. Tardat, L. Bouchard, S. Despied, Lethal oxidative damage and mutagenesis are generated by iron in delta for mutants of Escherichia coli: protective role of superoxide dismutase, J. Bacteriol. 177 (1995) 2305 – 2314. [10] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folinphenol reagent, J. Biol. Chem. 193 (1951) 265 – 275. [11] R.E. Amelunxen, D.O. Carr, Glyceraldehyde-3-phosphate dehydrogenase from rabbit muscle-1, Methods Enzymol. 41 (1975) 264 – 267. [12] C. Cordeiro, A. Ponces Freire, Methylglyoxal assay in cells as 2methylquinoxaline using 1,2-diaminobenzene as derivatizing reagent, Anal. Biochem. 234 (1996) 221 – 224. [13] J.H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1972. [14] S.B. Farr, R. D’Ari, D. Touati, Oxygen-dependent mutagenesis in Escherichia coli lacking superoxide dismutase, Proc. Natl. Acad. Sci. U. S. A. 83 1986, pp. 8268 – 8272. [15] J.D. Tucker, R.T. Taylor, M.L. Christensen, C.L. Strout, M.L. Hanna, A.V. Carrano, Cytogenetic response to 1,2-dicarbonyls and hydrogen peroxide in Chinese hamster ovary AUXB1 cells and human peripheral lymphocytes, Mutat. Res. 224 (1989) 269 – 279. [16] N. Murata-Kamiya, H. Kaji, H. Kasai, Types of mutations induced by glyoxal, a major oxidative DNA-damage product, in Salmonella typhimurium, Mutat. Res. 377 (1997) 13 – 16. [17] N. Murata-Kamiya, H. Kamiya, H. Kaji, H. Kasai, Methylglyoxal induces G:C to C:G and G:C to T:A transversions in the supF gene on a shuttle vector plasmid replicated in mammalian cells, Mutat. Res. 468 (2000) 173 – 182. [18] D.A. Armstrong, J.D. Buchanan, Reactions of O .2, H2O2 and other oxidants with sulfhydryl enzymes, Photochem. Photobiol. 28 (1978) 743 – 755. [19] J.D. Buchanan, D.A. Armstrong, The radiolysis of glyceraldehyde-3phosphate dehydrogenase, Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 33 (1978) 409 – 418. [20] M. Shinohara, P.J. Thornalley, I. Giardino, P. Beisswenger, S.R. Thorpe, J. Onorato, M. Brownlee, Overexpression of glyoxalase-I in bovine endothelial cells inhibits intracellular advanced glycation endproduct formation and prevents hyperglycemia-induced increases in macromolecular endocytosis, J. Clin. Invest. 101 (1998) 1142 – 1147. [21] R.J. Knight, K.F. Kofoed, H.R. Schelbert, D.B. Buxton, Inhibition of glyceraldehyde-3-phosphate dehydrogenase in post-ischaemic myocardium, Cardiovasc. Res. 32 (1996) 1016 – 1023. [22] X.L. Du, D. Edelstein, L. Rossetti, I.G. Fantus, H. Goldberg, F. Ziyadeh, J. Wu, M. Brownlee, Hyperglycemia-induced mitochon-
132
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
drial superoxide overproduction activates the hexosamine pathway and induces plasminogen activator inhibitor-1 expression by increasing Sp1 glycosylation, Proc. Natl. Acad. Sci. U. S. A. 97 2000, pp. 12222 – 12226. [23] M.A. Glomb, V.M. Monnier, Mechanism of protein modification by
glyoxal and glycolaldehyde, reactive intermediates of the Maillard reaction, J. Biol. Chem. 270 (1995) 10017 – 10026. [24] G. Leoncini, M. Maresca, A. Bonsignore, The effect of methylglyoxal on the glycolytic enzymes, FEBS Lett. 117 (1980) 17 – 18.
Triosephosphates are toxic to superoxide dismutase-deficient Escherichia coli Ludmil Benov *, Anees F. Beema, Fatima Sequeira Department of Biochemistry, Faculty of Medicine, Kuwait University, P.O. Box 24923 Safat, 13110, Kuwait Received 6 February 2003; received in revised form 2 June 2003; accepted 18 June 2003
Abstract Increase in the production of triosephosphates has been considered an important factor leading to diabetic complications. It might be expected that like the other short chain monosaccharides, triosephosphates autoxidize producing superoxide radical and a,h-diketones. Since superoxide can also initiate the oxidation of short chain sugars, free radical chain reactions are possible. If such reactions occur in vivo, triosephosphates would be more deleterious to cells lacking superoxide dismutase (SOD) than to normal cells. Here we demonstrate that triosephosphates kill a SOD-deficient Escherichia coli mutant much more than the parental, SOD-proficient strain. The effect is oxygendependent and is partially suppressed by aminoguanidine. Increased production of superoxide and diketones appeared to be the cause of triosephosphates toxicity. D 2003 Elsevier B.V. All rights reserved. Keywords: Dihydroxyacetone phosphate; D-Glyceraldehyde-3-phospate; Superoxide dismutase; Triosephosphates autoxidation; Methylglyoxal; Aminoguanidine
1. Introduction Dihydroxyacetone phosphate (DHAP) and D-glyceraldehyde-3-phospate (GA3P) are glycolytic intermediates whose steady state concentrations normally are very low. Increase of the triosephosphates production is found in diabetes and seems to be among the causes of diabetic complications [1,2]. One reason for the low intracellular concentrations of these glycolytic intermediates seems to be their intrinsic instability. DHAP and GA3P undergo spontaneous degradation to methylglyoxal (MG) [3,4], a compound that is highly toxic[5]. It might be expected that like the other short chain monosaccharides, triosephosphates autoxidize in air, producing superoxide radical and a corresponding diketone [6]. Since superoxide can also initiate the oxidation of such compounds, a free radical chain oxidation is possible [7]. Thus, oxidative stress would enhance triosephosphates degradation, and as a consequence, formation of toxic diketones. On the other hand, a
* Corresponding author. Tel.: +965-531-9489; fax: +965-533-8908. E-mail address: [email protected] (L. Benov). 0304-4165/03/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0304-4165(03)00134-X
rise in triosephosphates concentration might lead to increased production of superoxide. We demonstrate herewith that superoxide dismutase (SOD)-deficient Escherichia coli lose viability if incubated with triosephosphates. The reason appeared to be cell damage resulting from increased production of superoxide and dicarbonyls.
2. Materials and methods DL-Glyceraldehyde-3-phosphate (50 mg/ml in water) and DHAP, dilithium salt, were obtained form Sigma. Yeast extract, Bacto-tryptone and casamino acids were from Difco. LB medium contained 10 g Bacto-tryptone, 5 g yeast extract, and 10 g NaCl per liter and was adjusted to pH 7.0 with f 1.5 g of K2HPO4. M9CA medium consisted of minimal A salts [8], 0.2% casamino acids, 0.2% glucose, 3 mg pantothenate, and 5 mg of thiamine per liter. The strains of E. coli used were as follows: GC4468 = parent; QC1799 = GC4468 DsodA3, DsodB-kan [9]. Strains were grown overnight at 37 jC, with shaking in air, in LB medium containing 50 Ag/ml kanamycin where indicated. The overnight cultures were diluted 200-fold
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
into M9CA medium not containing antibiotics, and were grown to A600 nm = f 0.5. The cells were washed twice and resuspended to a density of A600 nm = 0.1 in M9 salts containing 0.2% glucose. Freshly prepared filter-sterilized GA3P or DHAP solutions were added and suspensions were incubated for 2 h before enumeration of surviving cells. At the end of the incubation, A600 nm was measured again and suspensions were suitably diluted and plated on LB plates for counting colonies. Plates were incubated aerobically at 37 jC for 24 – 48 h prior to counting. For the anaerobic experiments, cell suspensions were preincubated 15 min in sealed tubes full to the top. Before being sealed, and after additions, the cultures were bubbled with Ar for 5 min. These cells were plated on LB agar containing 0.2% glucose and plates were incubated anaerobically. For assaying glyceraldehyde-3-phosphate dehydrogenase (GA3PD), cells incubated with GA3P or DHAP as described above, were washed two times in ice-cold 50 mM Tris –HCl, pH 7.5, resuspended in the same buffer and lysed in a French press. The extracts were clarified by centrifugation and then assayed for protein [10], and GA3PD [11]. The GA3PD assay was performed in Tris –HCl, pH 7.5, and cysteine was omitted from the assay mixture to avoid reactivation of the enzyme. MG was assayed by using 1,2-diaminobenzene as derivatizing reagent, according to the procedure of Cordeiro and Ponces Freire [12]. HPLC analysis was performed in a Shimadzu 10ADM VP liquid chromatograph equipped with CLASS-VP Chromatography data system software. The column was Shim-pack VP-ODC 4.6 250 with a STR-2 ODS-2 precolumn. The mobile phase was 40% (v/v) 25 mM ammonium formate buffer, pH 3.4, and 60% (v/v) methanol. A sample volume of 100 Al was injected. Mutagenesis was monitored by assaying the frequency of thymine-negative (Thy ) mutants. Thy mutants are resistant to the drug trimethoprim and can be selected from a Thy+ population [13]. The assay was performed as described by Farr et al. [14]. For counting Thy mutants, aliquots of the cultures were plated directly on LB plates containing thymine (50 Ag/ml) and trimethoprim (15 Ag/ ml). For enumeration of cells, cultures were suitably diluted and plated on LB plates containing thymine, but not trimethoprim. Experiments were repeated at least three times. Figures represent mean F S.E.
129
Fig. 1. Triosephosphates induce mutations and kill the SOD-deficient E. coli. Panel A, mutagenesis—Overnight LB cultures of GC4468 (SODproficient) and QC1799 (SOD-deficient) were diluted 200-fold in M9CA medium and were grown aerobically at 37 jC and 200 rpm to a density of A600 nm f 0.2 – 0.3. At this point, GA3P or DHAP were added and the cells were kept for 2 h on the shaker. After the incubation, the cells were diluted and plated on LB plates containing thymine for assessing cell number or aliquots were plated without dilution on LB plates containing thymine and trimethoprim for counting Thy mutants. Panel B, loss of viability—Mid log cultures of parental and SOD-deficient E. coli grown in M9CA medium were diluted to A600 nm = 0.1 and were incubated 2 h with 5.0 mM GA3P or DHAP, or with 0.5 mM MG. After the incubation, suspensions were diluted and plated on LB agar for counting colonies.
the Thy mutants in the parental culture, while for the sodAsodB culture, the increase was 4 –6-fold. At higher concentration (5.0 mM), GA3P and DHAP caused a loss of viability of E. coli (Fig. 1B). Both compounds were more toxic for the sodAsodB than for the SOD-competent parental strain. Since the mutant differs from the parent only by the absence of cytoplasmic superoxide dismutases, these results suggest that superoxide plays a key role in triosephosphatesinduced cell damage.
3. Results 3.1. Effect of oxygen GA3P and DHAP are mutagens and kill the SODdeficient strain-Incubation with 2.0 mM GA3P or DHAP for 2 h increased the number of the Thy mutations in both parental and sodAsodB strains (Fig. 1A). There was a significant difference, however, between the two strains. Thus, GA3P and DHAP caused less than 2-fold increase of
If the main reason for the higher sensitivity of the sodAsodB strain toward GA3P and DHAP is higher superoxide steady state concentration, then anaerobiosis should diminish triosephosphates toxicity. As seen, incubating the cells with GA3P or DHAP in anaerobic atmo-
130
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
sphere significantly increased the viability of the sodAsodB cells (Fig. 2). Triosephosphates, like other short chain sugars, autoxidize in air producing O2S and a dicarbonyl compound, hydroxypyruvaldehydephosphate [6]. Dicarbonyls can modify proteins and nucleic acids and are toxic and mutagenic [5,15 –17]. Methylglyoxal (pyruvaldehyde), an analogue of hydroxypyruvaldehydephosphate, causes loss of viability at a concentration 10 times lower than that of GA3P or DHAP (Fig. 1B). In this case, however, no significant difference between the sensitivity of the parental and the sodAsodB strain was observed. 3.2. Accumulation of dicarbonyls Superoxide has been shown to serve as both an initiator and a propagator of the short chain sugars autoxidation [7]. One possible explanation for the role of the O2S in triosephosphates toxicity is that superoxide promotes the conversion of triosephosphates to diketones, which actually kill the cells. Analysis of the intracellular dicarbonyl content revealed that after incubation with GA3P or DHAP, the SOD-deficient cells contained more MG than did parental cells (Fig. 3). Technical difficulties did not allow us to determine the intracellular content of hydroxypyruvaldehydephosphate, the immediate product of triosephosphates oxidative degradation. 3.3. Glyceraldehyde-3-phosphate dehydrogenase MG is produced during the non-oxidative decomposition of GA3P and DHAP [3], and therefore its production should not be directly affected by the intracellular [O2S ]. One way superoxide might increase MG production, however, is by inactivating GA3PD [18,19]. This would prevent the me-
Fig. 2. Effect of oxygen on GA3P and DHAP toxicity. Mid log suspensions of parental and SOD-deficient E. coli were diluted to A600 nm = 0.1 and were incubated 2 h aerobically or anaerobically with 5.0 mM GA3P or DHAP. After the incubation, suspensions were diluted and plated on LB agar for counting colonies. Bars: (1) parental, aerobic; (2) parental, anaerobic; (3) sodAsodB, aerobic; (4) sodAsodB, anaerobic.
Fig. 3. Accumulation of MG. All conditions are as in Fig. 1B except that after the incubation the cells were washed, lysed, and cell-free extracts were assayed for MG.
tabolism of triosephosphates via the glycolytic pathway and would lead to formation of MG. We found that after incubation with GA3P or DHAP, the SOD-deficient mutant contained about half the GA3PD activity of the parental cells incubated at the same conditions (0.066 F 0.019 Amol/ min/mg protein in the sodAsodB against 0.139 F 0.026 Amol/min/mg protein in the parental strain). 3.4. Effect of aminoguanidine If triosephosphates toxicity is entirely due to production of dicarbonyls, then scavenging of dicarbonyls with aminoguanidine should prevent triosephosphates-induced cell death. Fig. 4 shows that in the DHAP-treated parental strain 100 AM aminoguanidine increased the survival rate to f 90% of the control, and 500 AM provided practically full protection. In the sodAsodB cells, 500 AM aminoguanidine doubled the survival rate, but nevertheless could not raise it above 65%. Almost the same results were
Fig. 4. Effect of aminoguanidine. All conditions are as in Fig. 1B. Aminoguanidine was added to final concentrations of 100 and 500 AM and DHAP was 5.0 mM.
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
obtained about the effect of aminoguanidine on GA3Ptreated cells (not shown).
4. Discussion Excessive production of advanced glycation end products (AGE) appears to be important in the pathogenesis of diabetic complications. Recent data show that hyperglycemia increases intracellular AGE primarily by increasing the concentration of AGE-forming MG [20]. It is generally accepted that MG is formed by non-oxidative mechanism, primarily by beta-elimination of phosphate from triosephosphate intermediates in anaerobic glycolysis [4,5]. In diabet e s , h y p e rg l y c e m i a - i n d u c e d o v e r p r o d u c t i o n o f mitochondrial superoxide reversibly inhibits GA3PD, which increases triosephosphate levels [21,22]. Here we demonstrate that in cells lacking SOD, increase of triosephosphates concentration has deleterious consequences. Because their carbonyl groups cannot be blocked by cyclization, triosephosphates are susceptible to autoxidation yielding superoxide and hydroxypyruvaldehydephosphate [6]. As a consequence, GA3PD is inactivated, leading to accumulation of another toxic dicarbonyl compound, methylglyoxal. Each of these products, superoxide and dicarbonyls, contributes to triosephosphates-induced cell damage. Superoxide, through the Fenton reaction, promotes the formation of hydroxyl radicals, which indiscriminately attack all cellular components. Dicarbonyls can interact with, and modify many different compounds in the cell including proteins and nucleic acids [5,23], thus inactivating enzymes and causing mutations [5,24]. Since superoxide is both an initiator and a product of sugars autoxidation, a vicious cycle might be envisaged in which overproduction of superoxide increases triosephosphates levels and degradation, which in turn causes more superoxide production and release of toxic dicarbonyls. It thus appears that among the important functions of SOD is to prevent the conversion of unstable glycolytic intermediates into highly reactive, toxic compounds.
Acknowledgements This work was supported by grant MB031 from Kuwait University. We are grateful to Danielle Touati (Institute Jacques Monod, CNRS, University Paris, France) for providing the E. coli strains used herein.
References [1] P.J. Thornalley, I. Jahan, R. Ng, Suppression of the accumulation of triosephosphates and increased formation of methylglyoxal in human red blood cells during hyperglycaemia by thiamine in vitro, J. Biochem. 129 (2001) 543 – 549.
131
[2] Y. Hamada, J. Nakamura, K. Naruse, T. Komori, K. Kato, Y. Kasuya, R. Nagai, S. Horiuchi, N. Hotta, Epalrestat, an aldose reductase ihibitor, reduces the levels of Nepsilon-(carboxymethyl)lysine protein adducts and their precursors in erythrocytes from diabetic patients, Diabetes Care 23 (2000) 1539 – 1544. [3] S.A. Phillips, P.J. Thornalley, The formation of methylglyoxal from triose phosphates. Investigation using a specific assay for methylglyoxal, Eur. J. Biochem. 212 (1993) 101 – 105. [4] J.P. Richard, Mechanism for the formation of methylglyoxal from triosephosphates, Biochem. Soc. Trans. 21 (1993) 549 – 553. [5] P.J. Thornalley, Pharmacology of methylglyoxal: formation, modification of proteins and nucleic acids, and enzymatic detoxification—a role in pathogenesis and antiproliferative chemotherapy, Gen. Pharmacol. 27 (1996) 565 – 573. [6] P. Thornalley, S. Wolff, J. Crabbe, A. Stern, The autoxidation of glyceraldehyde and other simple monosaccharides under physiological conditions catalysed by buffer ions, Biochim. Biophys. Acta 797 (1984) 276 – 287. [7] T. Mashino, I. Fridovich, Superoxide radical initiates the autoxidation of dihydroxyacetone, Arch. Biochem. Biophys. 254 (1987) 547 – 551. [8] T. Maniatis, E.F. Fritsch, J. Sambrook, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor, New York, 1982. [9] D. Touati, M. Jacques, B. Tardat, L. Bouchard, S. Despied, Lethal oxidative damage and mutagenesis are generated by iron in delta for mutants of Escherichia coli: protective role of superoxide dismutase, J. Bacteriol. 177 (1995) 2305 – 2314. [10] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folinphenol reagent, J. Biol. Chem. 193 (1951) 265 – 275. [11] R.E. Amelunxen, D.O. Carr, Glyceraldehyde-3-phosphate dehydrogenase from rabbit muscle-1, Methods Enzymol. 41 (1975) 264 – 267. [12] C. Cordeiro, A. Ponces Freire, Methylglyoxal assay in cells as 2methylquinoxaline using 1,2-diaminobenzene as derivatizing reagent, Anal. Biochem. 234 (1996) 221 – 224. [13] J.H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1972. [14] S.B. Farr, R. D’Ari, D. Touati, Oxygen-dependent mutagenesis in Escherichia coli lacking superoxide dismutase, Proc. Natl. Acad. Sci. U. S. A. 83 1986, pp. 8268 – 8272. [15] J.D. Tucker, R.T. Taylor, M.L. Christensen, C.L. Strout, M.L. Hanna, A.V. Carrano, Cytogenetic response to 1,2-dicarbonyls and hydrogen peroxide in Chinese hamster ovary AUXB1 cells and human peripheral lymphocytes, Mutat. Res. 224 (1989) 269 – 279. [16] N. Murata-Kamiya, H. Kaji, H. Kasai, Types of mutations induced by glyoxal, a major oxidative DNA-damage product, in Salmonella typhimurium, Mutat. Res. 377 (1997) 13 – 16. [17] N. Murata-Kamiya, H. Kamiya, H. Kaji, H. Kasai, Methylglyoxal induces G:C to C:G and G:C to T:A transversions in the supF gene on a shuttle vector plasmid replicated in mammalian cells, Mutat. Res. 468 (2000) 173 – 182. [18] D.A. Armstrong, J.D. Buchanan, Reactions of O .2, H2O2 and other oxidants with sulfhydryl enzymes, Photochem. Photobiol. 28 (1978) 743 – 755. [19] J.D. Buchanan, D.A. Armstrong, The radiolysis of glyceraldehyde-3phosphate dehydrogenase, Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 33 (1978) 409 – 418. [20] M. Shinohara, P.J. Thornalley, I. Giardino, P. Beisswenger, S.R. Thorpe, J. Onorato, M. Brownlee, Overexpression of glyoxalase-I in bovine endothelial cells inhibits intracellular advanced glycation endproduct formation and prevents hyperglycemia-induced increases in macromolecular endocytosis, J. Clin. Invest. 101 (1998) 1142 – 1147. [21] R.J. Knight, K.F. Kofoed, H.R. Schelbert, D.B. Buxton, Inhibition of glyceraldehyde-3-phosphate dehydrogenase in post-ischaemic myocardium, Cardiovasc. Res. 32 (1996) 1016 – 1023. [22] X.L. Du, D. Edelstein, L. Rossetti, I.G. Fantus, H. Goldberg, F. Ziyadeh, J. Wu, M. Brownlee, Hyperglycemia-induced mitochon-
132
L. Benov et al. / Biochimica et Biophysica Acta 1622 (2003) 128–132
drial superoxide overproduction activates the hexosamine pathway and induces plasminogen activator inhibitor-1 expression by increasing Sp1 glycosylation, Proc. Natl. Acad. Sci. U. S. A. 97 2000, pp. 12222 – 12226. [23] M.A. Glomb, V.M. Monnier, Mechanism of protein modification by
glyoxal and glycolaldehyde, reactive intermediates of the Maillard reaction, J. Biol. Chem. 270 (1995) 10017 – 10026. [24] G. Leoncini, M. Maresca, A. Bonsignore, The effect of methylglyoxal on the glycolytic enzymes, FEBS Lett. 117 (1980) 17 – 18.