Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Reperfusion of ischemic myocardium is known to be associated with a variety of ventricular arrhythmias and myocardial dysfunction that can lead to severe cardiac impairment and cell death (Manning et al., 1984; Pogwizd and Corr, 1986; Forman et al., 1990; Yamada et al., 1990). Lipid peroxidation of myocardial cell membranes by reactive oxygen species (ROS) such as superoxide anion (O₂), hydrogen peroxide (H₂O₂), hydroxyl radical (^·OH), and singlet oxygen (¹O₂) has been implicated as a potential mechanism for these deleterious effects (Kako, 1987; Halliwell et al., 1992; Esterbauer et al., 1993). Over the years, a variety of therapeutic approaches to protect the myocardium against these oxidants have been investigated. Although the formation of ROS is known to occur after the reperfusion of ischemic organs (Zweier et al., 1987, 1989), questions remain regarding the most effective therapeutic approach to prevent the detrimental effects of this oxidant injury. Prior studies have evaluated the efficacy of antioxidative enzymes such as superoxide dismutase or catalase or specific inhibitors of ROS-generating enzymes such as allopurinol or oxypurinol (Jolly et al., 1984; Manning et al., 1984; Pryzyklenk and Kloner, 1986). However, variable protection was observed with these enzymes, and this may be due to their inability to access intracellular region or limited efficacy against only one type of oxidant. Thus, there has been a need to develop antioxidant drugs that are both readily internalized and able to scavenge a range of ROS (Black et al., 1994; Kilgore et al., 1994).

Small-molecular-weight, stable nitroxides have been shown to have potential therapeutic values in a variety of disease processes, including myocardial reperfusion injury (Gelvan et al., 1991; Samuni et al., 1991; Mohsen et al., 1995; Zhang et al., 1998a), trauma (Zhang et al., 1998b), ulcerative colitis and mucosal injury (Karmeli et al., 1995), radioprotection (Hahn et al., 1992), leukocyte-endothelial cell adhesion (Russel et al., 1998), and doxorubicin (Adriamycin)-induced cardiotoxicity (Monti et al., 1996). The protective effects of nitroxides have been attributed to antioxidative processes, which include: 1) superoxide dismutase-mimicking activity (Krishna et al., 1992, 1996a), 2) induction of catalase-like activity in hemeproteins (Krishna et al., 1996b), and 3) radical scavenging. In addition, the nitroxides are cell permeable, making it possible to provide both intracellular and extracellular protection against oxidative stress. Recent studies demonstrated that nitroxides inhibit lipid peroxidation (Cighetti et al., 1997), protect against H₂O₂-induced cytotoxicity in Chinese hamster cells (Mitchell et al., 1991) and cultured cardiomyocytes (Samuni et al., 1991; Mohsen et al., 1995), and prevent postischemic reperfusion injury in the isolated heart (Gelvan et al., 1991). In biological tissues, the nitroxides are reduced to hydroxylamine form, and it has been well established that these two forms of nitroxide coexist in tissues. The hydroxylamine has also been shown to protect isolated cardiomyocytes against ROS-mediated injury, possibly due to a mechanism different from that of nitroxides (Zhang et al., 1998b).

A group of pyrroline-based compounds (Fig. 1) have been shown to possess class I antiarrhythmic activity (Hankovszky et al., 1986; Krishna et al., 1998; Xue et al., 1998). The amino compound N-[(2,2,5,5-tetramethyl-3-pyrroline-3-carboxamido)propyl]phthalimide (TPC-NH) is oxidized to the hydroxylamine (TPC-NOH) and nitroxide (TPC-NO) in mice (Twomey et al., 1997). The TPC-NH and TPC-NOH are diamagnetic, whereas the TPC-NO is a stable paramagnetic molecule and can be directly detected by electron paramagnetic resonance (EPR) spectroscopy. Also recently, it has been reported that the TPC-NH suppresses various canine ventricular arrhythmias (Xue et al., 1998). We provide direct evidence for the protective effects of TPC-NH and its oxidation metabolites in an isolated rat heart model. We demonstrate that these compounds exhibit markedly enhanced protection against reperfusion injury, presumably due to a combination of antioxidative and antiarrhythmic mechanisms.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Structure of TPC-NH and its oxidation products: TPC-NOH and TPC-NO. In biological tissues, TPC-NH is oxidized to TPC-NO, which undergoes reversible one-electron reduction to TPC-NOH. TPC-NO is a free radical and can be measured directly using EPR spectroscopy.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals

The drug TPC-NH (amine) and its oxidation metabolites TPC-NOH (hydroxylamine) and TPC-NO (nitroxide) were synthesized as previously reported (Hankovszky et al., 1986). The components of the modified Krebs' buffer solution, lactate dehydrogenase (LDH) diagnostic kit, lidocaine, xanthine (X), and xanthine oxidase (XO) were purchased from Sigma Chemical Co. (St. Louis, MO).

Isolated Heart Perfusion

Female Sprague-Dawley retired-breeder rats (weight, 300 ± 30 g; Harlan Company, Boston, MA) were used. All experiments were carefully conducted in compliance with the National Institutes of Health guidelines for the use of laboratory animals. After complete anesthesia (65 mg/kg pentobarbital i.p.), the heart was excised and the ascending aorta was rapidly cannulated. Retrograde perfusion was initiated according to the method of Langendorf at a constant pressure of 80 mm Hg using modified Krebs-bicarbonate-buffered perfusate containing 17 mM glucose, 120 mM NaCl, 25 mM NaHCO₃, 5.9 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgCl₂, and 0.5 mM EDTA. All perfusate solutions were routinely filtered through two 1.2-µm Millipore filters and bubbled with 95% O₂/5% CO₂ gas mixture at 37°C. A side arm in the perfusion line located just proximal to the aortic cannula allowed infusion of the drug solutions. The drugs were infused at a dilution of 1:20, with respect to the coronary flow (CF), using a Harvard Apparatus infusion pump. Contractile functions of the heart were measured using a fluid-filled latex balloon inserted into the left ventricular cavity, through the atrioventricular valve. The balloon was connected via a hydraulic line to a Spectramed P23XL pressure transducer with pressures amplified to a Gould four-channel strip chart recorder as well as to a personal computer equipped with MacLab data acquisition software. The balloon volume was adjusted to achieve an initial left ventricular end-diastolic pressure (LVEDP) of 10 ± 2 mm Hg, and all subsequent measurements were performed at the same balloon volume. The CF was measured with a T106 Transonic small animal flowmeter just proximal to the aortic cannulation.

Studies were performed in four groups: 1) control, 2) TPC-NH, 3) TPC-NOH, and 4) TPC-NO, wherein the drug or saline control was infused for 1 min before ischemia and during the initial 5 min of reperfusion. Studies were also performed on two additional groups wherein the TPC-NH was infused: 5) only during preischemia, or 6) only during the first 5 min of reperfusion. Studies were performed with at least seven hearts per group.

Myocyte Preparation

Adult rat ventricular myocytes were isolated according to an enzymatic technique. Briefly, 2- to 4-month-old Sprague-Dawley rats were anesthetized with pentobarbital. The hearts were quickly removed and retrogradely perfused with a low Ca²⁺-, collagenase-, and protease-containing bicarbonate buffer at 37°C. The perfusion was terminated when the heart tissue became soft. The ventricles were cut off, and the cardiac myocytes were mechanically desegregated. Myocytes were then rinsed in a bicarbonate solution and finally resuspended in HEPES buffer containing 1.0 mM Ca²⁺. The myocytes were then used in cell cytotoxicity studies.

LDH Assay

Cell toxicity studies were performed on isolated ventricular myocytes. The release of LDH after cell membrane damage due to exposure to X/XO was measured. The isolated myocytes were divided into four groups: 1) control, 2) cells + TPC-NH, 3) cells + X/XO, and 4) cells + X/XO + TPC-NH. Six experiments were performed in each group. The samples were mixed with 170 µM NADH and 0.1 M phosphate buffer (pH 7.5) containing 600 µM sodium pyruvate. The mixture was immediately transferred to a 1-ml quartz cuvette, and absorbance was recorded at 340 nm at 30-s intervals for 10 min using a Hewlett-Packard 8452A diode-array spectrophotometer. The activity of LDH was calculated from the rate of NADH oxidation and expressed as LDH U/ml of cell suspension.

Assessment of Cell Death

Cell death of isolated myocytes exposed to superoxide radicals was assessed using the Trypan blue staining technique: 0.2 ml of the cell suspension, 0.3 ml of Hanks' balanced salt solution, and 0.5 ml of 0.4% Trypan blue solution were mixed together and allowed to stand for 15 min. A small amount of this mixture was then transferred to hemocytometer chamber and viewed under a microscope. All the myocytes in the central 1-mm square and four 1-mm corner squares were counted for staining. The myocytes were divided into four groups: control, exposure to TPC-NH alone, exposure to X/XO in the presence of catalase (500 U/ml), and exposure to X/X + TPC-NH in the presence of catalase. The number of viable cells in each group was counted and compared with the control group.

EPR Spectroscopy

In Vitro EPR Characterization of Oxidation Products of TPC-NH. The effect of a variety of oxidants including superoxide (O), hydrogen peroxide (H₂O₂), singlet oxygen (¹O₂), ferryl (Fe⁴⁺), and alkylperoxide (ROO^·) on TPC-NH was studied using EPR spectroscopy. X (0.5 mM) and XO (0.02 U/ml) in aerobic phosphate buffer at pH 7.4 containing 0.1 mM diethylenetriaminepentaacetic acid (DTPA) were used to generate O₂ radicals. Catalase (500 U/ml) was included to scavenge hydrogen peroxide in the solution. Singlet oxygen was generated by the photoexcitation of rose bengal (Kukreja et al., 1991). The ferryl species were generated using horse myoglobin and hydrogen peroxide (Krishna et al., 1996b). 2,2'-Azobis-2-amidinopropane dihydrochloride (AAPH; 25 mM) was used in aerobic solutions at 37°C to generate ROO^· radicals (Niki, 1990). EPR measurements were carried out using X-band (9.78 GHz) with a TM₁₁₀ flat cell.

Measurements of TPC-NO in Whole Heart. EPR spectroscopy measurements on the whole intact heart were performed using an L-band EPR spectrometer with a reentrant resonator as described previously (Kuppusamy et al., 1995). After infusion of the drugs, the hearts were subjected to no-flow global ischemia. The hearts were quickly removed from the perfusion setup, washed, and placed inside the EPR sample cavity. Serial EPR spectra were acquired to continuously monitor the concentration of nitroxide (TPC-NO) in the heart for the next 30 min of ischemic duration. The heart was maintained at 37°C with a gentle flow of humidified warm air around the heart.

EPR Spectroscopic Measurement of TPC-NO from Heart Effluents. After the infusion of drugs, the coronary effluents were collected in 20-s aliquots during the preischemic and reperfusion periods. Samples were immediately frozen in liquid nitrogen and stored at 77 K until EPR measurements. The samples were thawed and EPR spectra were recorded at room temperature using a flat cell with IBM-Bruker ER 300 spectrometer operating at X-band with a TM₁₁₀ cavity. The spectrometer settings were: modulation frequency of 100 kHz, modulation amplitude of 0.5 G, microwave power of 20 mW, and microwave frequency of 9.78 GHz. Spectra were acquired as a sum of 10 scans with 30 s/scan sweep time. Spectral acquisition, analysis, and quantification were performed as described previously (Kuppusamy et al., 1995).

Ischemia-Reperfusion Protocol

After an equilibrium period of 15 min to allow for functional stabilization, baseline values of CF, LVEDP, left ventricular systolic pressure (LVSP), and heart rate (HR) were measured. The hearts were then subjected to a 1-min, preischemic, controlled infusion of drug/saline, at 5% of the CF rate, to achieve a final concentration of 50 µM in the perfusate. Subsequently, the hearts were subjected to 30 min of global, no-flow ischemia, followed by 45 min of reperfusion. During the first 5 min of reperfusion, the hearts were also subjected to a controlled infusion of the drugs at the same concentration and flow rate as in the preischemic state. After the first 5 min, reperfusion was continued with the Krebs' buffer alone for the remaining 40 min. Serial hemodynamic measurements were performed during the period of reperfusion. Coronary effluents were collected, in 1.5-ml aliquots, during the preischemic infusion period. The coronary effluent was also collected, in 20-s intervals, for the first 2 min of reperfusion and after this for a total reperfusion period of 45 min. The aliquots containing the effluent was immediately frozen in liquid nitrogen to preserve the drug and its oxidative metabolites until EPR measurements.

Statistical Analysis

Data are presented as mean ± S.E. Comparisons between groups were made with a one-way ANOVA designed for repeated measures. A P value of <.05 was considered statistically significant

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Oxygen Radical-Mediated Reactions. To delineate the efficiency of TPC-NH in scavenging oxygen-derived species, we used EPR spectroscopy to monitor the generation of TPC-NO, which is an oxidized product of TPC-NH. A solution of TPC-NH in aerated phosphate buffer (0.1 M, pH 7.4) containing 0.1 mM DTPA, 5 mM X, and 0.02 U/ml XO did not show any EPR signal of TPC-NO, suggesting that superoxide has no effect on this compound. Also, no significant effect was observed with ferryl myoglobin (50 µM). The reaction of alkylperoxyl radicals with TPC-NH was examined using an alkylperoxyl radical-generating compound, AAPH (Polysciences, Warrington, PA). The AAPH at ambient temperatures decomposes to produce alkyl radicals, which under aerobic conditions react with oxygen to produce alkyl peroxyl radicals. A time-dependent oxidation of TPC-NH to TPC-NO was observed in presence of 25 mM AAPH under aerobic conditions. Figure 2 shows the EPR spectra obtained after 15 min of incubation. The data suggest that TPC-NH can scavenge alkylperoxyl radicals and potentially inhibit lipid peroxidation reactions.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   X-band (9.78 GHz) EPR spectra of TPC-NO derived from alkylperoxyl radical-mediated oxidation of TPC-NH. EPR spectra of TPC-NO were measured from 1 mM TPC-NH in PBS containing 0.1 mM DTPA solution aerobically incubated at 37°C without (a) and with (b) the alkylperoxyl radical-generating system AAPH (25 mM) for 15 min. The TPC-NO showed a triplet spectrum with 14N hyperfine coupling constant of 16.05 G. Spectral acquisition parameters were modulation amplitude of 0.5 G, microwave power of 10 mW, and acquisition time of 5 min. The strong triplet signal suggests that TPC-NH is oxidized to TPC-NO by alkylperoxyl (ROO·) radicals.

In Vitro Measurement of Protection against Oxidative Cellular Injury. It has been reported that both hydroxylamine and nitroxide forms protect cardiomyocytes against oxidative cellular injury (Zhang et al., 1998a). This was shown by measuring LDH release from isolated cardiomyocytes exposed to a flux of superoxide radicals. To investigate whether the TPC-NH is also efficacious in protecting cardiomyocytes against superoxide-mediated oxidative cellular injury, in vitro cellular studies were performed. Freshly isolated adult rat left ventricular cardiomyocytes were exposed to a steady flux of superoxide (O₂), generated enzymatically by the aerobic X/XO reaction in the presence of catalase (500 U/ml) to scavenge hydrogen peroxide. The experiment used 1 × 10⁶ myocytes/ml with 0.5 mM X and 16 mU of XO at pH 7.4. The cellular damage caused by the superoxide radicals was estimated by measuring the leakage of cytoplasmic LDH. Figure 3 shows the amount of LDH leakage from cardiomyocytes subjected to 30 min of oxidative stress in the presence and absence of TPC-NH. Myocytes in the absence of TPC-NH released 236 ± 5 U/ml LDH, whereas in the presence of TPC-NH, the release of LDH was 114 ± 4 U/ml. This shows that TPC-NH exerted more than 50% protection of myocytes from oxidative damage caused by ROS.

View larger version (11K): [in this window] [in a new window]   Fig. 3.   Bar diagram depicting ROS-mediated cellular injury in adult ventricular myocytes. Cellular injury was assessed by measuring the amount of LDH released from myocytes subjected to oxidative stress for 30 min at room temperature by the addition of X/XO, in both the presence and absence of TPC-NH. Data are expressed as mean ± S.E. in LDH U/ml.

In Vivo Measurement of TPC-NH Metabolites in the Intact Heart. Isolated rat hearts were subjected to control perfusion, followed by infusion of 100 µM TPC-NH, TPC-NOH, or TPC-NO for 1 min. The hearts were then subjected to no-flow global ischemia and quickly transferred to the L-band EPR resonator. The EPR signal of TPC-NO was monitored continuously for up to 30 min. Typical EPR spectra of TPC-NO are shown in Fig. 4. At the end of 30 min of ischemia, hearts were homogenized and treated with 10 mM ferricyanide to selectively oxidize TPC-NOH to TPC-NO, and their EPR spectra were measured. The spectra of TPC-NO from hearts treated with TPC-NH, TPC-NOH, and TPC-NO are shown in Fig. 4, a-c, respectively, whereas the spectra after treatment with ferricyanide are shown in Fig. 4, e-g. A characteristic triplet spectrum with ¹⁴N hyperfine-splitting constants of 15.5 and 16.4 G was observed. The unequal coupling constants observed in the low-frequency (1.2 GHz) EPR spectra are due to the breakdown of the high-field approximation at 427 G used in this experiment. The spectra of ferricyanide-treated heart tissues correspond to the sum of TPC-NOH and TPC-NO at the end of 30 min of ischemia. Hearts treated with TPC-NH did not show any nitroxide (Fig. 4a) or hydroxylamine (Fig. 4d) at the end of 30 min of ischemia. This may suggest that TPC-NH was not metabolized to nitroxide. On the other hand, hearts treated with TPC-NOH was observed to show a small amount of TPC-NO (Fig. 4b), but at the end of 30 min of ischemia, the ischemic heart consisted entirely of TPC-NOH (Fig. 4, b and e). Hearts loaded with TPC-NO showed reduction in TPC-NO at least partially to its hydroxylamine (Fig. 4, c and f). At the end of 30 min of ischemia, it was observed that the entire amount of TPC-NO was reduced to TPC-NOH (Fig. 4e). The results thus suggest that at the end of 30 min of ischemia, the TPC-NH was not oxidized, whereas TPC-NOH and TPC-NO remained in the tissue mostly in the reduced form TPC-NOH.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   L-band (1.2 GHz) in vivo EPR spectra of TPC-NO from the heart. Rat hearts were infused with 100 µM solutions of TPC-NH, TPC-NOH, or TPC-NO for 1 min and subjected to no-flow global ischemia. The EPR spectra of TPC-NO from whole intact hearts were measured continuously for 30 min. Typical spectra acquired at 10 min of ischemia from hearts infused with TPC-NH (a), TPC-NOH (b), TPC-NO (c) are shown. At the end of 30-min ischemia, hearts were homogenized, and the homogenates were treated with 10 mM ferricyanide to convert TPC-NOH to TPC-NO. Right, spectra were obtained from the homogenates of hearts treated with TPC-NH (d), TPC-NOH (e), or TPC-NO (f). The TPC-NO showed a three-line spectrum with 14N hyperfine splitting 15.5 and 16.4 G. Spectral acquisition parameters were frequency of 1.2 GHz, modulation amplitude of 1.0 G, and microwave power of 5 mW.

Heart Functional Data. To evaluate the cardioprotective action of TPC-NH and its oxidation metabolites, isolated rat hearts were perfused with a 50-µM concentration of each compound during 1 min before ischemia and during the first 5 min of reperfusion. LVSP, LVEDP, HR, and CF were measured preischemia and during 45 min of reperfusion. In all the hearts, the preischemic LVEDP value was adjusted to 10 ± 2 mm Hg. The other preischemic baseline values were LVSP, 160 ± 42 mm Hg; HR, 320 ± 24 beats/min; and CF, 16 ± 5 ml/min. The left ventricular developed pressure (LVDP) was computed as the difference between LVSP and LVEDP. The rate-pressure product (RPP) was obtained as a product of HR and LVDP. The data obtained during reperfusion were expressed as a percentage of their corresponding preischemic baseline values. Figure 5 shows the recovery of LVDP, RPP, and CF as a function of reperfusion time. The percentage recovery of LVDP in the hearts infused with TPC-NOH or TPC-NO was significantly higher (P < .01) than that of control. At the end of 45 min of reperfusion, the LVDP of hearts treated with TPC-NOH and TPC-NO showed 30 ± 4 and 21 ± 1% of recovery, respectively, compared with 11 ± 1% for control (Fig. 6). The recovery of LVDP in TPC-NH-treated hearts was even more significant (40 ± 3%; P < .001) than its oxidative metabolites. Also, similar recoveries were observed with respect to RPP values. The RPP values at the end of 45 min of reperfusion were 38 ± 3, 31 ± 3, and 22 ± 2% for hearts treated with TPC-NH, TPC-NOH, and TPC-NO, respectively, compared with 11 ± 1% for control hearts. The recovery of CF in treated hearts was not different (P > .05) from that of control hearts (Fig. 5), suggesting that the three compounds did not alter the CF.

View larger version (35K): [in this window] [in a new window]   Fig. 5.   Recovery of contractile functions in isolated hearts subjected to ischemia and reperfusion. Perfused rat hearts were subjected to 30 min of global no-flow ischemia, followed by 45-min reperfusion. Hearts were infused with 50 µM TPC-NH, TPC-NOH, TPC-NO, or buffer during 1 min before ischemia and during the first 5 min of reperfusion. During reperfusion, LVDP, RPP (an index of mechanical function), and CF were measured and expressed as the percentage of the corresponding preischemic baseline values. , hearts perfused with only Krebs-bicarbonate buffer (n = 12); , hearts infused with TPC-NH (n = 7); , hearts infused with TPC-NOH (n = 7); , hearts infused with TPC-NO (n = 7). Data are plotted as mean ± S.E.

View larger version (57K): [in this window] [in a new window]   Fig. 6.   Recovery of contractile functions at 45 min of reperfusion in isolated hearts subjected to 30 min of global ischemia. Hearts were infused with 50 µM TPC-NH, TPC-NOH, TPC-NO, lidocaine, or Krebs' buffer during 1 min before ischemia and during the first 5 min of reperfusion. TPC-NH was also administered either only during the preischemic infusion [TPC-NH(PRE)] or only during the 5-min reperfusion period [TPC-NH(POST)]. The contractile functions are expressed as a percentage of the preischemic baseline. Data are plotted as mean ± S.E.

EPR Spectroscopy of Heart Effluents. Because it was observed that TPC-NH was capable of scavenging hydrogen peroxide (in the presence of trace metals), singlet oxygen, and alkylperoxyl radicals to form the nitroxide radical TPC-NO, the effluents from reperfused hearts were collected and examined using EPR spectroscopy. Heart effluents were collected for every 20 s during the 45 min of reperfusion and immediately frozen in liquid nitrogen. The frozen samples were later thawed, and their nitroxide content was measured and quantified as described in Materials and Methods. Figure 7 shows the intensity of TPC-NO signal as a function of reperfusion time. The signal intensity showed an increase during the first 20 to 40 s, and it gradually decreased thereafter. Hearts were also perfused with 50 µM TPC-NOH, effluents collected and analyzed (data not shown). The signal intensity showed a similar increase during the first 20 to 40 s, and it gradually decreased thereafter.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Intensity of the TPC-NO signal, measured by X-band EPR spectroscopy, from the coronary effluents of the heart. Rat hearts were subjected to 30 min of global no-flow ischemia and reperfused for 45 min. The coronary effluents were collected for the first 15 min of reperfusion and measured for the EPR signal of TPC-NO. , from hearts treated with TPC-NH for 1 min before the onset of ischemia; , from hearts treated with TPC-NH only during the first 5 min of reperfusion. Results are expressed as mean ± S.E. Hearts pretreated with TPC-NH showed a significant increase in the TPC-NO signal intensity during the first 1 min of reperfusion.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The TPC-NH is a five-member sterically hindered pyrroline-based compound characterized by three structural groups: an aromatic ring that anchors the drug into the lipophilic alkyl chains of membrane phospholipids, an amino group that ionizes at pH 8 to 9, and an interconnecting chain that joins the aromatic ring and the amino group and also has substituents that are capable of hydrogen bonding. In biological tissues, TPC-NH is converted to TPC-NO, which in turn is bioreduced to give the corresponding hydroxylamine TPC-NOH. This interconversion gives the potential of tissues treated with TPC-NH to have all three forms of the compound coexisting at any time during the treatment period.

The ability of nitroxyl radicals to inhibit in vitro lipid peroxidation was recently reported (Cighetti et al., 1997). Lipid peroxidation induced by Fenton's reagent in liver microsomes and egg phosphotidylcholine was found to be inhibited by stable lipophilic steroidal nitroxide radical. The steroidal nitroxide was shown to function as a chain-breaking antioxidant. It was further observed that the inhibition was comparable to that of -tocopherol, which is a lipid-soluble antioxidant known to work in fatty areas such as the lipids, suggesting that a good affinity for cell membranes increases the lipid peroxidation inhibitory effect. Although the mechanism of inhibition of lipid peroxidation in cell membranes by nitroxides is due to the termination of lipid peroxyl radical cascade, their intracellular protection against oxidative damage has been ascribed to the ability of nitroxides to oxidize the reduced metals, such as iron and copper. The oxidation of reduced metal ions will preempt the Fenton reaction and prevent the formation of secondary ^·OH radicals. It is also possible that nitroxides offer cellular protection through reaction with intracellular O₂ and secondary radicals such as R^·, RO^·, or ROO^·, terminating the propagation of radical chain reactions.

In most of the research aimed at investigation of the therapeutic potential (antioxidant, antiarrhythmic, and radiation protection) of the nitroxides, the role of hydroxylamine form has not been investigated in detail. The bioreduction product of 4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl, or Tempol (hydroxylamine), also showed significant protection of cardiomyocytes against cell membrane damage induced by X/XO (Zhang et al., 1998b). The antioxidative activity of the hydroxylamine was found to be greater than that of the nitroxide itself (Zhang et al., 1998a). The mechanism of protection by the hydroxylamine form has been suggested to be due to detoxification of lipid radicals involving a hydrogen atom abstraction reaction leading to chain termination and production of nonradical species:

(1)

Because of the equilibrium between nitroxide and hydroxylamine in tissues, their concentrations in tissues are continuously replenished. This interconversion enables them to act in a catalytic mode, contrary to common antioxidants, which operate in a stoichiometric mode.

The EPR measurements (Fig. 4) of tissue TPC-NOH, and TPC-NO from hearts with 1-min preischemic infusion of TPC-NH, TPC-NOH, and TPC-NO suggest that at the end of 30 min ischemia, that is, at the beginning of reperfusion, the tissue consisted only of TPC-NH in the case of TPC-NH-infused hearts or TPC-NOH in the case of hearts infused with the other compounds. Reintroduction of oxygen during the reperfusion phase causes oxidative conversion of TPC-NH and TPC-NOH to TPC-NO, which undergoes subsequent reversible bioreduction to TPC-NOH, thus maintaining a steady-state concentration of all the three compounds in the reperfused tissue. Furthermore, preadministration followed by 30 min of ischemic duration results in complete internalization of TPC-NH and its metabolites in cells. Because all three forms coexist at any time in tissue and because their antioxidant action is catalytic in nature, a comprehensive "sweep" of toxic oxygen radicals can occur, leading to a more enhanced protection compared with the presence of any single compound. The in vivo administration of TPC-NH provides three different antioxidants in a steady-state concentration in tissues.

The present investigation demonstrates that all the three compounds, administered separately, are capable of protecting against ischemia-reperfusion injury in isolated rat hearts. However, the addition of TPC-NH for 1 min before ischemia is significantly more protective than the addition of TPC-NH at the start of reperfusion. The observations that the oxidative metabolites are not made during ischemia and that TPC-NH added before ischemia may be still present during reflow may suggest that much of the protective effect of TPC-NH is not due to metabolism or to the oxidative metabolites. The protective effect of TPC-NH against postischemic injury could be due to: 1) reduction in the severity of ischemic damage incurred by the deprivation of oxygen (ischemic protection), 2) a direct antiarrhythmic activity providing protection against arrhythmia-induced damage (antiarrhythmic protection), or 3) scavenging of toxic oxidative species that cause cardiac damage during reperfusion (antioxidant protection). The observation that hearts pretreated (1 min, preischemic) with TPC-NH showed a better recovery of contractile function compared with post-treated (5 min, beginning reperfusion) hearts (P < .05) suggests that there is both ischemia- and reperfusion-associated protection. The possibility that the protection could be due to the ability of TPC-NH to function as an antiarrhythmic agent was considered. Because the functional recovery of hearts treated with lidocaine was not as prominent as that of hearts treated with TPC-NH, it is clear that there are additional pathways that have an important role in its marked myocardial protection. It is seen from Fig. 6 that there is a marked decrease in the CF in hearts treated with lidocaine, suggesting that lidocaine is vasoactive. The vascular effect of lidocaine in rat cremaster muscle preparations has been reported (Johns et al., 1985). A biphasic response to increasing concentrations of lidocaine was observed. Progressive constriction was noted with lidocaine concentrations from 1 to 1000 µg/ml, whereas vasodilation was observed for 10 mg/ml lidocaine. In our experiments, lidocaine was used at a concentration of about 100 µg/ml, so the decrease in CF seen in our hearts treated with lidocaine is consistent with prior literature.

The particular design of TPC-NH and its metabolites having a lipophilic aromatic end and a hydrophilic amino group at the other end separated by alkyl chain imparts these compounds with the ability to detoxify oxygen radicals in the lipid-rich membrane as well as water-rich cytosolic areas. These types of compounds have been shown to have much higher solubility in the aqueous/lipid membrane interface (Subczynski et al., 1998). This may result in enhanced membrane activity leading to blockage of cardiac sodium and calcium channels, thus offering ischemic protection. In addition, TPC-NH may also protect Na⁺,K⁺-ATPase function of cardiac myocytes against ischemia and reperfusion-induced inactivation. Our results show that TPC-NH is capable of providing membrane stabilization by localized site-targeted detoxification of ROS that are generated during reperfusion. Thus, it appears that the compound is capable of providing both ischemic and postischemic myocardial protection against reperfusion injury.

The TPC-NH is able to take up oxygen from the most reactive oxygen radicals to form nitroxide. This is similar to vitamin E in that it scavenges reactive oxygen intermediates (oxy radicals and also nonradicals) to form a nitroxide radical (Fig. 8). This radical can be reduced by ascorbic acid to the labile diamagnetic N-hydroxyl compound (TPC-NOH). The semidehydroascorbate radical formed in this process is reduced by cellular thiols. The advantage of this process is that the TPC-NOH is a nontoxic intermediate, which is sufficiently reactive to reduce/scavenge reactive oxygen intermediates by transferring its hydrogen.

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Schematic of the action of TPC-NH in the prevention of oxidant injury. TPC-NH is similar to vitamin E in scavenging reactive oxygen intermediates to form a nitroxide radical (TPC-NO). This radical can be reduced by ascorbic acid to the diamagnetic N-hydroxyl compound (TPC-NOH). The semidehydroascorbate radical formed in this process is reduced by cellular thiols. The TPC-NOH is a nontoxic intermediate, which is sufficiently reactive to reduce/scavenge reactive oxygen intermediates by transferring its hydrogen atom.

In summary, the pyrroline-based antiarrhythmic nitroxide precursor compound TPC-NH is metabolized in vivo to form the corresponding hydroxylamine and nitroxide derivatives and offers the potential of membrane-targeted antioxidant action against myocardial postischemic reperfusion injury.