- Email: [email protected]
Detection of Oxidative DNA Damage to Ischemic Reperfused Rat Hearts by 8-Hydroxydeoxyguanosine Formation
J Mol Cell Cardiol 30, 1939–1944 (1998) Article No. mc980752
Detection of Oxidative DNA Damage to Ischemic Reperfused Rat Hearts by 8-Hydroxydeoxyguanosine Formation Gerald A. Cordis1, Gautam Maulik1, Debasis Bagchi2, Walter Riedel3 and Dipak K. Das1 1
University of Connecticut School of Medicine, Farmington, CT, USA; 2Creighton University School of Pharmacy, Omaha, NE, USA; 3Max-Planck Institut, Bad Nauheim, Germany (Received 6 April 1998, accepted in revised form 5 June 1998) G. A. C, G. M, D. B, W. R D. K. D. Detection of Oxidative DNA Damage to Ischemic Reperfused Rat Hearts by 8-Hydroxydeoxyguanosine Formation. Journal of Molecular and Cellular Cardiology (1998) 30, 1939–1944. Reactive oxygen species that are generated in the ischemic heart upon reperfusion, play a significant role in the pathogenesis of reperfusion injury. Although DNA is a well known target for free radical attack, little attention has been paid to the injury of DNA molecules associated with ischemia and reperfusion. In this study, the formation of 8-hydroxydeoxyguanosine (8-OHDG), a product of hydroxyl radical (OH.)-DNA interaction, was monitored in the post-ischemic myocardium. A simple high performance liquid chromatography (HPLC), with uv detection, detected pmol levels of 8-OHDG in the pre-ischemic heart which increased steadily and progressively as a function of reperfusion time. A similar rise in 8-OHDG was noticed when isolated hearts were perfused with a OH.-generating system. Corroborating with the increased 8-OHDG formation, increased amount of creatine kinase was released from the coronary effluent indicating increased tissue injury. The formation of 8-OHDG was completely blocked when hearts were preperfused with oxygen-free-radical scavenger, 1,3dimethyl-2-thiourea (DMTU) which also significantly reduced the appearance of CK in the coronary effluent, suggesting that oxidative DNA damage play a role in the pathophysiology of ischemic reperfusion injury. 1998 Academic Press
K W: DNA; 8-OHDG; Heart; Ischemia/reperfusion; Oxidative stress; Oxygen free radicals.
Introduction Numerous reports exist in the literature to support the notion that reperfusion of ischemic myocardium is associated with the generation of oxygen-derived free radicals, which play a significant role in reperfusion injury (Baker et al., 1988; Powell and Hall, 1990; Kukreja and Hess, 1993; Das and Maulik, 1994; Kramer et al., 1994). Among the oxygen free radicals, hydroxyl radical (OH.) is believed to be the most detrimental causative agent for the pathogenesis of reperfusion injury. Virtually every biomolecule is a potential target for OH. radical attack. Both in vitro (Prasad and Das, 1989)
and in vivo (Meerson et al., 1982; Das et al., 1987) studies indicate the formation of lipid peroxidation products in the ischemic reperfused myocardium, suggesting the breakdown of sarcolemmal phospholipids by the action of OH.. Although DNA is also a potential target for free radicals, except for ischemia/reperfusion-induced cleavage and apoptosis (Maulik et al., 1998), little evidence exists to indicate DNA damage in the ischemic reperfused hearts. In this study, we sought to determine the presence of 8-OHDG in the DNA from hearts undergoing ischemia and reperfusion. The results of our study demonstrate a steady and progressive rise in the
Please address all correspondence to: Dipak K. Das, Cardiovascular Division, Department of Surgery, University of Connecticut, School of Medicine, Farmington, CT 06030-1110, USA.
0022–2828/98/101939+06 $30.00/0
1998 Academic Press
1940
G. A. Cordis et al.
Group I
75 min perfusion with KHB buffer only
Group II
15' P
Group III 15' DMTU
30' I
30' R
30' I
30' R
Group IV
15' P
30' perfusion with OH*
Group V
15' DMTU
30' perfusion with OH*
Figure 1 Schemata of the experimental protocol.
8-OHDG formation as a function of the duration of reperfusion time. The formation of 8-OHDG was completely blocked by preperfusing the hearts in the presence of oxygen free radical scavenger, 1,3dimethyl-2-thiourea (DMTU), suggesting that reperfusion of ischemic myocardium causes oxidative DNA damage.
Materials and Methods Isolated rat heart preparation Sprague–Dawley rats weighing about 300 g were anesthetized with pentobarbital (80 mg/kg, i.p.). After intravenous administration of heparin (500 IU/kg), the chests were opened, and the hearts were rapidly excised and mounted on a non-recirculating Langendorff perfusion apparatus (Maulik et al., 1993). Retrograde perfusion was established at a pressure of 100 cm H2O with an oxygenated normothermic Krebs–Henseleit bicarbonate (KHB) buffer with the following ion concentrations (in m): 118.0 NaCl, 24.0 NaHCO3, 4.7 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.7 CaCl2 and 10.0 glucose. The KHB buffer had been previously equilibrated with 95% O2/5% CO2, pH 7.4 at 37°C. The hearts were randomly divided into five groups: the first group of hearts were perfused for 75 min with buffer alone; the second and the fourth group of hearts were perfused with buffer alone for 15 min, and then the second group was subjected to 30 min of ischemia followed by 30 min of reperfusion, while the fourth group was perfused with a OH.-generating system (100 l xanthine, 8 mU/ml xanthine oxidase, 100 l FeCl3 and 100 l EDTA) for 60 min; the third and fifth group of hearts were perfused with buffer in the presence of 100 l DMTU (a hydroxyl radical scavenger) for 15 min, and then the third group was subjected to 30 min of ischemia followed by 30 min of reperfusion, while the fifth group was perfused for 60 min with a OH.generating system. The schemata of the protocol is shown in Figure 1. Coronary perfusates were
collected before ischemia (baseline) and during reperfusion for subsequent determination of creatine kinase (CK) activity, a presumptive marker for cellular damage.
Isolation of DNA To isolate DNA from the hearts, experiments were terminated at different time points, hearts rapidly frozen in liquid N2, and stored at −70°C for subsequent DNA isolation. At a later date, hearts (1 g) were homogenized in 10 ml solution containing 1% SDS and 1 m EDTA using a Polytron homogenizer, and the homogenates incubated at 38°C for 30 min with proteinase K (500 lg/ml) (Gupta, 1984). 0.5 ml of 1 Tris-HCl buffer, pH 7.4 was added to the homogenate, and the resulting solution was extracted with 1 volume each of phenol, and a mixture of 1:1 (v/v) phenol and Sevag (chloroform/ isoamyl alcohol 24:1, v/v). The phases were separated by centrifugation, and the aqueous phases collected and pooled. DNA was precipitated with 0.1 vol of ethanol at −20°C. After centrifugation and a 70% ethanol rinse, the DNA was dissolved in 2 ml of a solution containing 1.5 m NaCl, 150 l Na-citrate, 1 m EDTA. The solution was incubated for 30 min at 37°C with RNase T1 (50 units/ml) and RNase A (100 lg/ml). The DNA solutions were extracted with Sevag and precipitated with NaCl and ethanol (Gupta, 1984).
DNA hydrolysis The DNA pellets were dissolved in 0.5 ml of 20 m sodium acetate, pH 4.8, and incubated for 30 min at 37°C with 63 mg of Nuclease P1. Fifty ll of 1 Tris-HCl, pH 7.4, was added, and the nucleotides incubated for 60 min at 37°C with 6.3 units of E. coli alkaline phosphatase, as described previously (Das and Engelman, 1990).
High performance liquid chromatography (HPLC) of bases Twenty-five ll of the filtered aliquot of the hydrolysed DNA was injected onto a C18 Beckman Ultrasphere column (3 lm particle size, 7.5 cm× 4.6 mm) equipped with a Waters HPLC chromatograph containing Millenium software and the 996 photodiode array detector. The sample was run isocratically for 10 min using a mobile phase containing 4% acetonitrile–0.1% acetic acid. The
1941
DG
0.40
8-OHDG
0.30
DA T
0.20 0.10
0.5
1.5
2.5
3.5
4.5
Wa vel en
300
gth
250
(nm )
0.00
5000
Peak area ( MV*s)
DC
Absorbance (V)
8-OHDG Formation in Ischemic Reperfused Heart
Results HPLC detection of 8-OHDG As shown in Figure 2, HPLC chromatograms of the authentic standards for deoxycytosine (DC), deoxyguanosine (DG), deoxyadenosine (DA), thymidine (T) and 8-hydroxydeoxyguanosine-OH. adduct (8-OHDG) showed retention times of 1.3, 2.7, 3.2, 3.5 and 3.8 min, respectively. The wavelength scans show that all the bases have wavelength maximum in the 250–280 nm range, including 8OHDG; however, only 8-OHDG has a wavelength maximum above 280 nm, specifically at 297 nm. Only DC and 8-OHDG have significant response at 297 nm. A 25 ll injection of 1.25–25 lmol of 8OHDG standard was chromatographed with the areas determined at each concentration both at 254 and at 297 nm and graphed as depicted in Figure 3. Both standard curves were linear with r values of 0.9870 and 0.9810, respectively. The response factor of 1.015284×10−9 at 254 nm is only slightly lower than the response factor of 1.216572×10−9 at 297 nm. A very low amount of 8-OHDG was detected in the normal heart (Fig. 4). The amount of 8-OHDG
2000 1000
1000
250 500 750 Concentration of 8-OHDG (µ M)
Figure 3 Standard curves for 8-OHDG at 254 nm (Χ) and 297 nm (Μ). A 25-ll injection of 1.25–25 mmol of 8-OHDG standard was chromatographed as described in Materials and Methods. The area in mV∗s was determined at the selected wavelength using photodiode array detection.
8-OHDG formed (% increase over baseline)
DNA–OH. conjugate, 8-OHDG, (both the extracted and authentic standard) was detected at a retention time of 3.8 min using a wavelength of 297 nm. Purity of 8-OHDG was further determined by the uv scan of the authentic standard with comparison to the rat heart 8-OHDG.
3000
0 50 100
Time (min)
Figure 2 A 3-dimensional scan of the HPLC chromatogram of deoxycytosine (DC), deoxyguanosine (DG), deoxyadenosine (DA), thymidine (T) and 8-OHDG standards. A 25-ll injection of 2.5 nmol of each standard was chromatographed as described in Materials and Methods. A 3-dimensional scan was obtained using the M-996 photodiode array detector and Millenium software.
4000
240 220 200 180 *
160 * 140
* *
120 100 80
BL
30 I
10 R
20 R
30 R
Reperfusion (min)
Figure 4 Effects of DMTU on ischemia/reperfusion-induced production of 8-OHDG in rat heart. Isolated rat hearts were perfused in the presence (Ε) or absence (Χ) of 100 l for 15 min followed by 30 min of reperfusion. 8-OHDG was estimated in the DNA of the hearts as described in Materials and Methods. Results are expressed as means±... for six hearts in each group. Each assay was run in duplicate. ∗P<0.05 compared to control.
increased significantly after 20 min of reperfusion following 30 min of ischemia and maintained up to 30 min of reperfusion. About 1547±121 pmol/ mg DNA of 8-OHDG was formed after 20 min of reperfusion as compared to 1008±98 pmol/mg DNA of 8-OHDG formed at the baseline level (Table 1). The amount of 8-OHDG did not increase after subsequent reperfusion. DMTU almost abolished the amount of 8-OHDG, suggesting that the formation of 8-OHDG is mediated by oxygen-free radicals.
1942
G. A. Cordis et al. Table 1 Production of 8-OHDG in the ischemic reperfused heart. Experimental condition
8-OHDG formed (pmol/mg DNA)
Baseline 30 min Ischemia 30 min ischemia+10 min reperfusion 30 min ischemia+20 min reperfusion 30 min ischemia+30 min reperfusion
−DMTU
+DMTU
1008±98 1079±103 857±87 1547±121∗ 1500±118∗
— 118±26 70±14 65±12 184±27
∗ P<0.05 compared to baseline.
Table 2 Production of 8-OHDG in heart during perfusion with OH.-generating system.
8-OHDG formed (% increase over baseline)
250
200
150
* *
*
10
20
100
50
Time of perfusion with OH.generating system
8-OHDG formed (pmol/mg DNA) −DMTU
+DMTU
Baseline 10 min 20 min 30 min
1008±98 2109±193∗ 2367±208∗ 2076±187∗
— 62±16 125±25 107±29
∗ P<0.05 compared to baseline.
BL
30
Perfusion (min)
260
Figure 5 Effects of DMTU on OH -induced production of 8-OHDG in rat heart. Isolated rat hearts were perfused in the presence (Ε) or absence (Χ) of 100 l nm DMTU for 15 min followed by 30 min perfusion with the OH.generating system. 8-OHDG was estimated in the DNA obtained from the hearts as described in Materials and Methods. Results are expressed as means±... for six hearts in each group. Each assay was run in duplicate. ∗P<0.05 compared to control.
To confirm the role of oxygen-derived free radicals in 8-OHDG formation, isolated rat hearts were perfused with a OH.-generating system. As shown in Figure 5, the amount of 8-OHDG increased by twofold after 10 min of perfusing the hearts with OH.generating system. This increased 8-OHDG level was also maintained up to 30 min of perfusion with OH.-generating system. Amount of 8-OHDG formed was 2367±208 and 2076±187 pmol/mg DNA after 20 and 30 min of OH.-perfusion, respectively, compared to 1008±98 pmol/mg DNA at the baseline control (Table 2). The appearance of 8-OHDG was almost completely inhibited in the hearts preperfused with DMTU, demonstrating the role of OH. in 8-OHDG formation. Hearts preperfused with OH. scavenger, DMTU provided significant protection from the free radical injury as evidenced by the reduced CK release from the heart. CK release increased progressively as a
Creatine kinase release (% increase over baseline)
.
240 220 200 180
*
160
*
140 *
120 100
BL
10 R
20 R
30 R
Reperfusion (min)
Figure 6 Effects of DMTU on ischemia/reperfusion and OH.-induced production of creatine kinase in hearts. Isolated rat hearts were perfused in the presence (Ε) or absence (Χ) of 100 l DMTU for 15 min followed by 30 min ischemia and 30 min reperfusion. Creatine kinase (CK) was estimated in the coronary effluents as described in Materials and Methods. Results are expressed as means±... for six hearts in each group. Each assay was run in duplicate. ∗P<0.05 compared to control.
function of the duration of reperfusion time (Fig. 6). Perfusing hearts with OH.-generating system caused progressive increase in CK release from the coronary effluent. DMTU significantly reduced the amount of CK released from the heart, indicating
8-OHDG Formation in Ischemic Reperfused Heart
that at least a part of the ischemic reperfusion injury is due to the production of free radicals in the heart.
Discussion The generation of oxygen-derived free radicals play a crucial role in the pathogenesis of myocardial ischemic reperfusion injury. A significant number of evidences exist to support that reactive oxygen species are formed at the onset of reperfusion (Ferrari et al., 1989). The presence of OH. in the ischemic reperfused myocardium was confirmed using electron spin resonance spectroscopy (ESR) as well as high-performance liquid chromatography (HPLC) (Tosaki et al., 1993). The increased production of oxygen-free radicals in conjunction with the decreased activity of the antioxidant defense is believed to be an important mediator for myocardial reperfusion injury (Das et al., 1986). Virtually every biomolecule including lipids, proteins, DNA and RNA is the potential target for free radical attack (Floyd and Schneider, 1990). A number of previous studies demonstrated increased appearance of lipid peroxidation products associated with the reperfusion of ischemic myocardium (Meerson et al., 1982; Das et al., 1987). Recently, using a very sensitive method, we have clearly demonstrated that the amount of MDA production is significantly increased during the reperfusion following a brief period of ischemia (Floyd and Schneider, 1990; Cordis, 1995). Although DNA represents a potential target for free radicals, no report is available to indicate the involvement of DNA damage in reperfusion injury. Damage of DNA by oxygen free radicals results in the production of a large number of lesions which can be grouped into strand breaks and base modification products. At least three modified bases, 8-hydroxyguanine, 5-hydroxymethyluracil, and thymine glycol, are formed when OH. attacks a DNA molecule (Floyd and Schneider, 1990). Following an oxidative insult on DNA, guanine bases rapidly undergo 8-hydroxylation followed by the repairment of the damaged DNA by exonucleases. In this study, we have clearly detected the appearance of 8-OHDG in the DNA of the ischemic reperfused myocardium. As shown in Table 1, the amount of 8-OHDG steadily increased with the progression of reperfusion time. A similar trend in the appearance of 8-OHDG in the DNA was observed when the isolated hearts were perfused with a OH.generating system (Table 2). In the present study, the appearance of 8-OHDG was almost completely
1943
inhibited in the hearts preperfused with DMTU, a OH. scavenger, suggesting that the appearance of this product is indeed due to the oxidative insult on DNA. 8-OHDG is a specific product formed by the attack of OH. on DNA. It is formed when OH. interacts with a guanine molecule at the C8 position (Kasai et al., 1989). In normal cells, damaged sequences of DNA are rapidly repaired through a process of excision which is catalysed by exonucleases. If for any reason, DNA is left unrepaired, it may lead to mutagenicity (Floyd and Schneider, 1990). 8-OHDG may appear due to any imbalance between OH. generation, antioxidant defense, and repair of damaged sequences of DNA. 8-OHDG appears to be a sensitive and integral marker of oxidative damage to DNA. Several methods to quantitate 8-OHDG are available in the literature. These include gas chromatographymass spectroscopy (GC-MS) (Dizdaroglu, 1985), HPLC detection with electrochemical detector (Floyd and Schneider, 1990) as well as radioimmunoassays (Musarrat and Wani, 1994). Other investigators measuring 8-OHDG with HPLC have used electrochemical detection mainly for its increased sensitivity. However, electrochemical detection has certain limitations. For example, the response depends on the condition of working electrode which varies during the same work day, not to mention day to day. Therefore, standards must be run frequently to account for the change in response. Also, the electrochemical response decreases exponentially as it approaches the lower range of detection limit. In this report, we describe for the first time a very simple HPLC technique using uv detection. The samples can be run isocratically for 10 min using a mobile phase containing 4% acetonitrile and 0.1% acetic acid. The 8-OHDG is readily detected at a retention time of 3.8 min using 297-nm wavelength. The uv scans of 8-OHDG showed a maximum at 297 nm at which the other deoxynucleotides have negligible response. Therefore, even though a significant amount of thymidine is present, it does not swamp out the 8-OHDG at 297 nm as it does at 254 and 260 nm. Since this HPLC procedure uses an isocratic gradient where all the bases are chromatographed in 10 min, and since the column does not need any preconditioning or washing, the method becomes an extremely rapid and effort-free procedure for routine use. In summary, the results of this study provided evidence for the first time that the reperfusion of ischemic myocardium results in the DNA modification producing 8-OHDG. The formation of
1944
G. A. Cordis et al.
8-OHDG not only can serve as a sensitive marker for the development of oxidative stress associated with ischemia and reperfusion but also for other oxidative stress mediated heart diseases such as cardiomyopathy and congestive heart failure.
Acknowledgements This study was supported in part by National Heart, Lung, and Blood Institute Grants NIH HL 22559 and HL-33889.
References A S, D HH, 1992. Isolation of a malonaldehyde-deoxyguanosine adduct from rat liver DNA. Free Rad Biol Med 13: 695–699. B JE, F CC, O GN, K B, 1988. Myocardial ischemia and reperfusion: direct evidence for free radical generation by electron spin resonance spectroscopy. Proc Natl Acad Sci USA 85: 2786–2788. B K, F I. 1981. DNA strand scission by enzymatically generated oxygen radicals. Arch Biochem Biophys 206: 414–419. B RK, MB A, L J, K FJ, 1995. Oxidative damage to DNA in patients with cystic fibrosis. Free Rad Biol Med 18: 801–806. C GA, M N, D DK, 1995. Detection of oxidative stress in heart by estimating the dinitrophenylhydrazine derivative of malonaldehyde. J Mol Cell Cardiol 27: 1645–1653. D DK, E RM, 1990. Mechanism of free radical generation during reperfusion of ischemic myocardium. In: Das DK and Essman WB (eds). Oxygen Radicals: Systemic Events and Disease Processes. Basel: Karger, 97–121. D DK, M N, 1994. Evaluation of antioxidant effectiveness in ischemia reperfusion tissue injury methods. Methods Enzymol 233: 601–610. D DK, E RM, R JA, B RH, O H, L S, 1986. Pathophysiology of superoxide radical as potential mediator of reperfusion injury in pig heart. Basic Res Cardiol 81: 155–166. D DK, E RM, F D, O H, B RH, 1987. Developmental profiles of protective mechanisms of heart against peroxidative injury. Basic Res Cardiol 82: 36–43. D M. 1985. Formation of an 8-hydroxyguanine monkey in deoxyribonucleic acid on c-irradiation in aqueous solution. Biochemistry 24: 4476–4481. F RA, S JE, 1990. Hydroxy free radical damage to DNA. In: Vigo-Pelfrey C (ed.). Membrane Lipid Oxidation, vol 1. Boca Raton, Florida: CRC Press. F R, C S, B GM, C E, P E,
G G, A A, 1989. Oxygen free radicalmediated heart injury in animal models and during bypass surgery in humans. Ann NY Acad Sci 570: 237–253. F CG, S MK, P JW, D P, A BN, 1990. Oxidative damage to DNA during aging: 8hydroxy-2′-deoxyguanosine in rat organ DNA and urine. Proc Natl Acad Sci 87: 4533–4537. G RC, 1984. Nonrandom binding of the carcinogen N-hydroxy-2-acetylaminoflurene to repetitive sequences of rat liver DNA in vitro. Proc Natl Acad Sci USA 81: 6943–6947. H B, A OI, DNA . FEBS Lett 281: 9–19. K H, C PF, K Y, N S, O A, T H, 1989. Formation of 8-hydroxyguanine moiety and evidence for its repair. Carcinogenesis 7: 1849–1851. K JH, M V, W WB, 1994. Lipid peroxidation-derived free radical production and postischemic myocardial reperfusion injury. Ann NY Acad Sci 723: 181–196. K RK, H ML, 1993. Oxygen radicals, neutrophilderived oxidants, and myocardial reperfusion injury. In: Das DK (ed.). Pathophysiology of Reperfusion Injury. Boca Raton, Florida: CRC Press, 221–241. M N, D DK, G M, C GA, A N, D N, 1993. Reduction of myocardial ischemic reperfusion injury by sialylated glycosphingolipids, gangliosides. J Cardiovasc Pharmacol 22: 74–81. M N, Y T, D DK, 1998. Oxidative stress developed during the reperfusion of ischemic myocardium induces apoptosis. Free Rad Biol Med 24: 869–875. M RZ, K VE, L YP, B LM, A YV, 1982. The role of lipid peroxidation in pathogenesis of ischemic damage and antioxidant protection of the heart. Basic Res Cardiol 77: 465–470. M P, P A, 1989. Free radicals and myocardial protection: a surgical viewpoint. Ann Thoracic Surg 47: 939–944. M J, W AA, 1994. Quantitative immunoanalysis of promutagenic 8-hydroxy-2′-deoxyguanosine in oxidized DNA. Carcinogenesis 15: 2037–2043. P SR, H D, 1990. Use of salicylate as a probe for OH. formation in isolated ischemic rat hearts. Free Rad Biol Med 9: 133–141. P MR, D DK, 1989. Effect of oxygen-derived free radicals and oxidants on the degradation in vitro of membrane phospholipids. Free Rad Res Commun 7: 381–388. T A, B D, H D, P T, C GA, D DK, 1993. Comparisons of ESR and HPLC methods for the detection of hydroxyl radicals in ischemic/reperfused hearts. A relationship between the genesis of oxygen-free radicals and reperfusion-induced arrhythmias. Biochem Pharmacol 45: 961–969. T TD, 1987. Chemical “snapshots” of DNA: using the hydroxyl radical to study the structure of DNA and DNA-protein complexes. Trends Biochem Sci 12: 297–300.
Detection of Oxidative DNA Damage to Ischemic Reperfused Rat Hearts by 8-Hydroxydeoxyguanosine Formation Gerald A. Cordis1, Gautam Maulik1, Debasis Bagchi2, Walter Riedel3 and Dipak K. Das1 1
University of Connecticut School of Medicine, Farmington, CT, USA; 2Creighton University School of Pharmacy, Omaha, NE, USA; 3Max-Planck Institut, Bad Nauheim, Germany (Received 6 April 1998, accepted in revised form 5 June 1998) G. A. C, G. M, D. B, W. R D. K. D. Detection of Oxidative DNA Damage to Ischemic Reperfused Rat Hearts by 8-Hydroxydeoxyguanosine Formation. Journal of Molecular and Cellular Cardiology (1998) 30, 1939–1944. Reactive oxygen species that are generated in the ischemic heart upon reperfusion, play a significant role in the pathogenesis of reperfusion injury. Although DNA is a well known target for free radical attack, little attention has been paid to the injury of DNA molecules associated with ischemia and reperfusion. In this study, the formation of 8-hydroxydeoxyguanosine (8-OHDG), a product of hydroxyl radical (OH.)-DNA interaction, was monitored in the post-ischemic myocardium. A simple high performance liquid chromatography (HPLC), with uv detection, detected pmol levels of 8-OHDG in the pre-ischemic heart which increased steadily and progressively as a function of reperfusion time. A similar rise in 8-OHDG was noticed when isolated hearts were perfused with a OH.-generating system. Corroborating with the increased 8-OHDG formation, increased amount of creatine kinase was released from the coronary effluent indicating increased tissue injury. The formation of 8-OHDG was completely blocked when hearts were preperfused with oxygen-free-radical scavenger, 1,3dimethyl-2-thiourea (DMTU) which also significantly reduced the appearance of CK in the coronary effluent, suggesting that oxidative DNA damage play a role in the pathophysiology of ischemic reperfusion injury. 1998 Academic Press
K W: DNA; 8-OHDG; Heart; Ischemia/reperfusion; Oxidative stress; Oxygen free radicals.
Introduction Numerous reports exist in the literature to support the notion that reperfusion of ischemic myocardium is associated with the generation of oxygen-derived free radicals, which play a significant role in reperfusion injury (Baker et al., 1988; Powell and Hall, 1990; Kukreja and Hess, 1993; Das and Maulik, 1994; Kramer et al., 1994). Among the oxygen free radicals, hydroxyl radical (OH.) is believed to be the most detrimental causative agent for the pathogenesis of reperfusion injury. Virtually every biomolecule is a potential target for OH. radical attack. Both in vitro (Prasad and Das, 1989)
and in vivo (Meerson et al., 1982; Das et al., 1987) studies indicate the formation of lipid peroxidation products in the ischemic reperfused myocardium, suggesting the breakdown of sarcolemmal phospholipids by the action of OH.. Although DNA is also a potential target for free radicals, except for ischemia/reperfusion-induced cleavage and apoptosis (Maulik et al., 1998), little evidence exists to indicate DNA damage in the ischemic reperfused hearts. In this study, we sought to determine the presence of 8-OHDG in the DNA from hearts undergoing ischemia and reperfusion. The results of our study demonstrate a steady and progressive rise in the
Please address all correspondence to: Dipak K. Das, Cardiovascular Division, Department of Surgery, University of Connecticut, School of Medicine, Farmington, CT 06030-1110, USA.
0022–2828/98/101939+06 $30.00/0
1998 Academic Press
1940
G. A. Cordis et al.
Group I
75 min perfusion with KHB buffer only
Group II
15' P
Group III 15' DMTU
30' I
30' R
30' I
30' R
Group IV
15' P
30' perfusion with OH*
Group V
15' DMTU
30' perfusion with OH*
Figure 1 Schemata of the experimental protocol.
8-OHDG formation as a function of the duration of reperfusion time. The formation of 8-OHDG was completely blocked by preperfusing the hearts in the presence of oxygen free radical scavenger, 1,3dimethyl-2-thiourea (DMTU), suggesting that reperfusion of ischemic myocardium causes oxidative DNA damage.
Materials and Methods Isolated rat heart preparation Sprague–Dawley rats weighing about 300 g were anesthetized with pentobarbital (80 mg/kg, i.p.). After intravenous administration of heparin (500 IU/kg), the chests were opened, and the hearts were rapidly excised and mounted on a non-recirculating Langendorff perfusion apparatus (Maulik et al., 1993). Retrograde perfusion was established at a pressure of 100 cm H2O with an oxygenated normothermic Krebs–Henseleit bicarbonate (KHB) buffer with the following ion concentrations (in m): 118.0 NaCl, 24.0 NaHCO3, 4.7 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.7 CaCl2 and 10.0 glucose. The KHB buffer had been previously equilibrated with 95% O2/5% CO2, pH 7.4 at 37°C. The hearts were randomly divided into five groups: the first group of hearts were perfused for 75 min with buffer alone; the second and the fourth group of hearts were perfused with buffer alone for 15 min, and then the second group was subjected to 30 min of ischemia followed by 30 min of reperfusion, while the fourth group was perfused with a OH.-generating system (100 l xanthine, 8 mU/ml xanthine oxidase, 100 l FeCl3 and 100 l EDTA) for 60 min; the third and fifth group of hearts were perfused with buffer in the presence of 100 l DMTU (a hydroxyl radical scavenger) for 15 min, and then the third group was subjected to 30 min of ischemia followed by 30 min of reperfusion, while the fifth group was perfused for 60 min with a OH.generating system. The schemata of the protocol is shown in Figure 1. Coronary perfusates were
collected before ischemia (baseline) and during reperfusion for subsequent determination of creatine kinase (CK) activity, a presumptive marker for cellular damage.
Isolation of DNA To isolate DNA from the hearts, experiments were terminated at different time points, hearts rapidly frozen in liquid N2, and stored at −70°C for subsequent DNA isolation. At a later date, hearts (1 g) were homogenized in 10 ml solution containing 1% SDS and 1 m EDTA using a Polytron homogenizer, and the homogenates incubated at 38°C for 30 min with proteinase K (500 lg/ml) (Gupta, 1984). 0.5 ml of 1 Tris-HCl buffer, pH 7.4 was added to the homogenate, and the resulting solution was extracted with 1 volume each of phenol, and a mixture of 1:1 (v/v) phenol and Sevag (chloroform/ isoamyl alcohol 24:1, v/v). The phases were separated by centrifugation, and the aqueous phases collected and pooled. DNA was precipitated with 0.1 vol of ethanol at −20°C. After centrifugation and a 70% ethanol rinse, the DNA was dissolved in 2 ml of a solution containing 1.5 m NaCl, 150 l Na-citrate, 1 m EDTA. The solution was incubated for 30 min at 37°C with RNase T1 (50 units/ml) and RNase A (100 lg/ml). The DNA solutions were extracted with Sevag and precipitated with NaCl and ethanol (Gupta, 1984).
DNA hydrolysis The DNA pellets were dissolved in 0.5 ml of 20 m sodium acetate, pH 4.8, and incubated for 30 min at 37°C with 63 mg of Nuclease P1. Fifty ll of 1 Tris-HCl, pH 7.4, was added, and the nucleotides incubated for 60 min at 37°C with 6.3 units of E. coli alkaline phosphatase, as described previously (Das and Engelman, 1990).
High performance liquid chromatography (HPLC) of bases Twenty-five ll of the filtered aliquot of the hydrolysed DNA was injected onto a C18 Beckman Ultrasphere column (3 lm particle size, 7.5 cm× 4.6 mm) equipped with a Waters HPLC chromatograph containing Millenium software and the 996 photodiode array detector. The sample was run isocratically for 10 min using a mobile phase containing 4% acetonitrile–0.1% acetic acid. The
1941
DG
0.40
8-OHDG
0.30
DA T
0.20 0.10
0.5
1.5
2.5
3.5
4.5
Wa vel en
300
gth
250
(nm )
0.00
5000
Peak area ( MV*s)
DC
Absorbance (V)
8-OHDG Formation in Ischemic Reperfused Heart
Results HPLC detection of 8-OHDG As shown in Figure 2, HPLC chromatograms of the authentic standards for deoxycytosine (DC), deoxyguanosine (DG), deoxyadenosine (DA), thymidine (T) and 8-hydroxydeoxyguanosine-OH. adduct (8-OHDG) showed retention times of 1.3, 2.7, 3.2, 3.5 and 3.8 min, respectively. The wavelength scans show that all the bases have wavelength maximum in the 250–280 nm range, including 8OHDG; however, only 8-OHDG has a wavelength maximum above 280 nm, specifically at 297 nm. Only DC and 8-OHDG have significant response at 297 nm. A 25 ll injection of 1.25–25 lmol of 8OHDG standard was chromatographed with the areas determined at each concentration both at 254 and at 297 nm and graphed as depicted in Figure 3. Both standard curves were linear with r values of 0.9870 and 0.9810, respectively. The response factor of 1.015284×10−9 at 254 nm is only slightly lower than the response factor of 1.216572×10−9 at 297 nm. A very low amount of 8-OHDG was detected in the normal heart (Fig. 4). The amount of 8-OHDG
2000 1000
1000
250 500 750 Concentration of 8-OHDG (µ M)
Figure 3 Standard curves for 8-OHDG at 254 nm (Χ) and 297 nm (Μ). A 25-ll injection of 1.25–25 mmol of 8-OHDG standard was chromatographed as described in Materials and Methods. The area in mV∗s was determined at the selected wavelength using photodiode array detection.
8-OHDG formed (% increase over baseline)
DNA–OH. conjugate, 8-OHDG, (both the extracted and authentic standard) was detected at a retention time of 3.8 min using a wavelength of 297 nm. Purity of 8-OHDG was further determined by the uv scan of the authentic standard with comparison to the rat heart 8-OHDG.
3000
0 50 100
Time (min)
Figure 2 A 3-dimensional scan of the HPLC chromatogram of deoxycytosine (DC), deoxyguanosine (DG), deoxyadenosine (DA), thymidine (T) and 8-OHDG standards. A 25-ll injection of 2.5 nmol of each standard was chromatographed as described in Materials and Methods. A 3-dimensional scan was obtained using the M-996 photodiode array detector and Millenium software.
4000
240 220 200 180 *
160 * 140
* *
120 100 80
BL
30 I
10 R
20 R
30 R
Reperfusion (min)
Figure 4 Effects of DMTU on ischemia/reperfusion-induced production of 8-OHDG in rat heart. Isolated rat hearts were perfused in the presence (Ε) or absence (Χ) of 100 l for 15 min followed by 30 min of reperfusion. 8-OHDG was estimated in the DNA of the hearts as described in Materials and Methods. Results are expressed as means±... for six hearts in each group. Each assay was run in duplicate. ∗P<0.05 compared to control.
increased significantly after 20 min of reperfusion following 30 min of ischemia and maintained up to 30 min of reperfusion. About 1547±121 pmol/ mg DNA of 8-OHDG was formed after 20 min of reperfusion as compared to 1008±98 pmol/mg DNA of 8-OHDG formed at the baseline level (Table 1). The amount of 8-OHDG did not increase after subsequent reperfusion. DMTU almost abolished the amount of 8-OHDG, suggesting that the formation of 8-OHDG is mediated by oxygen-free radicals.
1942
G. A. Cordis et al. Table 1 Production of 8-OHDG in the ischemic reperfused heart. Experimental condition
8-OHDG formed (pmol/mg DNA)
Baseline 30 min Ischemia 30 min ischemia+10 min reperfusion 30 min ischemia+20 min reperfusion 30 min ischemia+30 min reperfusion
−DMTU
+DMTU
1008±98 1079±103 857±87 1547±121∗ 1500±118∗
— 118±26 70±14 65±12 184±27
∗ P<0.05 compared to baseline.
Table 2 Production of 8-OHDG in heart during perfusion with OH.-generating system.
8-OHDG formed (% increase over baseline)
250
200
150
* *
*
10
20
100
50
Time of perfusion with OH.generating system
8-OHDG formed (pmol/mg DNA) −DMTU
+DMTU
Baseline 10 min 20 min 30 min
1008±98 2109±193∗ 2367±208∗ 2076±187∗
— 62±16 125±25 107±29
∗ P<0.05 compared to baseline.
BL
30
Perfusion (min)
260
Figure 5 Effects of DMTU on OH -induced production of 8-OHDG in rat heart. Isolated rat hearts were perfused in the presence (Ε) or absence (Χ) of 100 l nm DMTU for 15 min followed by 30 min perfusion with the OH.generating system. 8-OHDG was estimated in the DNA obtained from the hearts as described in Materials and Methods. Results are expressed as means±... for six hearts in each group. Each assay was run in duplicate. ∗P<0.05 compared to control.
To confirm the role of oxygen-derived free radicals in 8-OHDG formation, isolated rat hearts were perfused with a OH.-generating system. As shown in Figure 5, the amount of 8-OHDG increased by twofold after 10 min of perfusing the hearts with OH.generating system. This increased 8-OHDG level was also maintained up to 30 min of perfusion with OH.-generating system. Amount of 8-OHDG formed was 2367±208 and 2076±187 pmol/mg DNA after 20 and 30 min of OH.-perfusion, respectively, compared to 1008±98 pmol/mg DNA at the baseline control (Table 2). The appearance of 8-OHDG was almost completely inhibited in the hearts preperfused with DMTU, demonstrating the role of OH. in 8-OHDG formation. Hearts preperfused with OH. scavenger, DMTU provided significant protection from the free radical injury as evidenced by the reduced CK release from the heart. CK release increased progressively as a
Creatine kinase release (% increase over baseline)
.
240 220 200 180
*
160
*
140 *
120 100
BL
10 R
20 R
30 R
Reperfusion (min)
Figure 6 Effects of DMTU on ischemia/reperfusion and OH.-induced production of creatine kinase in hearts. Isolated rat hearts were perfused in the presence (Ε) or absence (Χ) of 100 l DMTU for 15 min followed by 30 min ischemia and 30 min reperfusion. Creatine kinase (CK) was estimated in the coronary effluents as described in Materials and Methods. Results are expressed as means±... for six hearts in each group. Each assay was run in duplicate. ∗P<0.05 compared to control.
function of the duration of reperfusion time (Fig. 6). Perfusing hearts with OH.-generating system caused progressive increase in CK release from the coronary effluent. DMTU significantly reduced the amount of CK released from the heart, indicating
8-OHDG Formation in Ischemic Reperfused Heart
that at least a part of the ischemic reperfusion injury is due to the production of free radicals in the heart.
Discussion The generation of oxygen-derived free radicals play a crucial role in the pathogenesis of myocardial ischemic reperfusion injury. A significant number of evidences exist to support that reactive oxygen species are formed at the onset of reperfusion (Ferrari et al., 1989). The presence of OH. in the ischemic reperfused myocardium was confirmed using electron spin resonance spectroscopy (ESR) as well as high-performance liquid chromatography (HPLC) (Tosaki et al., 1993). The increased production of oxygen-free radicals in conjunction with the decreased activity of the antioxidant defense is believed to be an important mediator for myocardial reperfusion injury (Das et al., 1986). Virtually every biomolecule including lipids, proteins, DNA and RNA is the potential target for free radical attack (Floyd and Schneider, 1990). A number of previous studies demonstrated increased appearance of lipid peroxidation products associated with the reperfusion of ischemic myocardium (Meerson et al., 1982; Das et al., 1987). Recently, using a very sensitive method, we have clearly demonstrated that the amount of MDA production is significantly increased during the reperfusion following a brief period of ischemia (Floyd and Schneider, 1990; Cordis, 1995). Although DNA represents a potential target for free radicals, no report is available to indicate the involvement of DNA damage in reperfusion injury. Damage of DNA by oxygen free radicals results in the production of a large number of lesions which can be grouped into strand breaks and base modification products. At least three modified bases, 8-hydroxyguanine, 5-hydroxymethyluracil, and thymine glycol, are formed when OH. attacks a DNA molecule (Floyd and Schneider, 1990). Following an oxidative insult on DNA, guanine bases rapidly undergo 8-hydroxylation followed by the repairment of the damaged DNA by exonucleases. In this study, we have clearly detected the appearance of 8-OHDG in the DNA of the ischemic reperfused myocardium. As shown in Table 1, the amount of 8-OHDG steadily increased with the progression of reperfusion time. A similar trend in the appearance of 8-OHDG in the DNA was observed when the isolated hearts were perfused with a OH.generating system (Table 2). In the present study, the appearance of 8-OHDG was almost completely
1943
inhibited in the hearts preperfused with DMTU, a OH. scavenger, suggesting that the appearance of this product is indeed due to the oxidative insult on DNA. 8-OHDG is a specific product formed by the attack of OH. on DNA. It is formed when OH. interacts with a guanine molecule at the C8 position (Kasai et al., 1989). In normal cells, damaged sequences of DNA are rapidly repaired through a process of excision which is catalysed by exonucleases. If for any reason, DNA is left unrepaired, it may lead to mutagenicity (Floyd and Schneider, 1990). 8-OHDG may appear due to any imbalance between OH. generation, antioxidant defense, and repair of damaged sequences of DNA. 8-OHDG appears to be a sensitive and integral marker of oxidative damage to DNA. Several methods to quantitate 8-OHDG are available in the literature. These include gas chromatographymass spectroscopy (GC-MS) (Dizdaroglu, 1985), HPLC detection with electrochemical detector (Floyd and Schneider, 1990) as well as radioimmunoassays (Musarrat and Wani, 1994). Other investigators measuring 8-OHDG with HPLC have used electrochemical detection mainly for its increased sensitivity. However, electrochemical detection has certain limitations. For example, the response depends on the condition of working electrode which varies during the same work day, not to mention day to day. Therefore, standards must be run frequently to account for the change in response. Also, the electrochemical response decreases exponentially as it approaches the lower range of detection limit. In this report, we describe for the first time a very simple HPLC technique using uv detection. The samples can be run isocratically for 10 min using a mobile phase containing 4% acetonitrile and 0.1% acetic acid. The 8-OHDG is readily detected at a retention time of 3.8 min using 297-nm wavelength. The uv scans of 8-OHDG showed a maximum at 297 nm at which the other deoxynucleotides have negligible response. Therefore, even though a significant amount of thymidine is present, it does not swamp out the 8-OHDG at 297 nm as it does at 254 and 260 nm. Since this HPLC procedure uses an isocratic gradient where all the bases are chromatographed in 10 min, and since the column does not need any preconditioning or washing, the method becomes an extremely rapid and effort-free procedure for routine use. In summary, the results of this study provided evidence for the first time that the reperfusion of ischemic myocardium results in the DNA modification producing 8-OHDG. The formation of
1944
G. A. Cordis et al.
8-OHDG not only can serve as a sensitive marker for the development of oxidative stress associated with ischemia and reperfusion but also for other oxidative stress mediated heart diseases such as cardiomyopathy and congestive heart failure.
Acknowledgements This study was supported in part by National Heart, Lung, and Blood Institute Grants NIH HL 22559 and HL-33889.
References A S, D HH, 1992. Isolation of a malonaldehyde-deoxyguanosine adduct from rat liver DNA. Free Rad Biol Med 13: 695–699. B JE, F CC, O GN, K B, 1988. Myocardial ischemia and reperfusion: direct evidence for free radical generation by electron spin resonance spectroscopy. Proc Natl Acad Sci USA 85: 2786–2788. B K, F I. 1981. DNA strand scission by enzymatically generated oxygen radicals. Arch Biochem Biophys 206: 414–419. B RK, MB A, L J, K FJ, 1995. Oxidative damage to DNA in patients with cystic fibrosis. Free Rad Biol Med 18: 801–806. C GA, M N, D DK, 1995. Detection of oxidative stress in heart by estimating the dinitrophenylhydrazine derivative of malonaldehyde. J Mol Cell Cardiol 27: 1645–1653. D DK, E RM, 1990. Mechanism of free radical generation during reperfusion of ischemic myocardium. In: Das DK and Essman WB (eds). Oxygen Radicals: Systemic Events and Disease Processes. Basel: Karger, 97–121. D DK, M N, 1994. Evaluation of antioxidant effectiveness in ischemia reperfusion tissue injury methods. Methods Enzymol 233: 601–610. D DK, E RM, R JA, B RH, O H, L S, 1986. Pathophysiology of superoxide radical as potential mediator of reperfusion injury in pig heart. Basic Res Cardiol 81: 155–166. D DK, E RM, F D, O H, B RH, 1987. Developmental profiles of protective mechanisms of heart against peroxidative injury. Basic Res Cardiol 82: 36–43. D M. 1985. Formation of an 8-hydroxyguanine monkey in deoxyribonucleic acid on c-irradiation in aqueous solution. Biochemistry 24: 4476–4481. F RA, S JE, 1990. Hydroxy free radical damage to DNA. In: Vigo-Pelfrey C (ed.). Membrane Lipid Oxidation, vol 1. Boca Raton, Florida: CRC Press. F R, C S, B GM, C E, P E,
G G, A A, 1989. Oxygen free radicalmediated heart injury in animal models and during bypass surgery in humans. Ann NY Acad Sci 570: 237–253. F CG, S MK, P JW, D P, A BN, 1990. Oxidative damage to DNA during aging: 8hydroxy-2′-deoxyguanosine in rat organ DNA and urine. Proc Natl Acad Sci 87: 4533–4537. G RC, 1984. Nonrandom binding of the carcinogen N-hydroxy-2-acetylaminoflurene to repetitive sequences of rat liver DNA in vitro. Proc Natl Acad Sci USA 81: 6943–6947. H B, A OI, DNA . FEBS Lett 281: 9–19. K H, C PF, K Y, N S, O A, T H, 1989. Formation of 8-hydroxyguanine moiety and evidence for its repair. Carcinogenesis 7: 1849–1851. K JH, M V, W WB, 1994. Lipid peroxidation-derived free radical production and postischemic myocardial reperfusion injury. Ann NY Acad Sci 723: 181–196. K RK, H ML, 1993. Oxygen radicals, neutrophilderived oxidants, and myocardial reperfusion injury. In: Das DK (ed.). Pathophysiology of Reperfusion Injury. Boca Raton, Florida: CRC Press, 221–241. M N, D DK, G M, C GA, A N, D N, 1993. Reduction of myocardial ischemic reperfusion injury by sialylated glycosphingolipids, gangliosides. J Cardiovasc Pharmacol 22: 74–81. M N, Y T, D DK, 1998. Oxidative stress developed during the reperfusion of ischemic myocardium induces apoptosis. Free Rad Biol Med 24: 869–875. M RZ, K VE, L YP, B LM, A YV, 1982. The role of lipid peroxidation in pathogenesis of ischemic damage and antioxidant protection of the heart. Basic Res Cardiol 77: 465–470. M P, P A, 1989. Free radicals and myocardial protection: a surgical viewpoint. Ann Thoracic Surg 47: 939–944. M J, W AA, 1994. Quantitative immunoanalysis of promutagenic 8-hydroxy-2′-deoxyguanosine in oxidized DNA. Carcinogenesis 15: 2037–2043. P SR, H D, 1990. Use of salicylate as a probe for OH. formation in isolated ischemic rat hearts. Free Rad Biol Med 9: 133–141. P MR, D DK, 1989. Effect of oxygen-derived free radicals and oxidants on the degradation in vitro of membrane phospholipids. Free Rad Res Commun 7: 381–388. T A, B D, H D, P T, C GA, D DK, 1993. Comparisons of ESR and HPLC methods for the detection of hydroxyl radicals in ischemic/reperfused hearts. A relationship between the genesis of oxygen-free radicals and reperfusion-induced arrhythmias. Biochem Pharmacol 45: 961–969. T TD, 1987. Chemical “snapshots” of DNA: using the hydroxyl radical to study the structure of DNA and DNA-protein complexes. Trends Biochem Sci 12: 297–300.