EFFECT OF SODIUM 2,3-DIMERCAPTOPROPANE-1-SULPHONATE (DMPS) ON CHRONIC DAUNORUBICIN TOXICITY IN RABBITS: COMPARISON WITH DEXRAZOXANE

dithiol iron DMPS Summary: A possible protective action of DMPS (a dithiol chelating agent) against chronic daunorubicin toxicity in rabbits in comparison with dexrazoxane was investigated. The rabbits were divided into five groups: control (saline, 1 ml/kg i.v.), daunorubicin (3 mg/kg i.v.), DMPS (50 mg/kg i.v.); the remaining two groups were pre-treated either with dexrazoxane (60 mg/kg i.p.) or DMPS (50 mg/kg i.v.) 30 min before administration of daunorubicin (3 mg/kg i.v.). Drugs were given once a week for 10 weeks. Routine biochemical parameters were determined in weeks 1, 5 and 11. In the 11 th week, invasive haemodynamic parameters were measured, then the rabbits underwent autopsy, cardiac tissue was examined by light microscopy and scored semiquantitatively. The contents of calcium, potassium, magnesium, iron and selenium were measured in the left heart ventricle. DMPS administered alone was well tolerated and did not cause any major signs of toxicity. It decreased the cardiac content of calcium, but did not affect the iron concentration. In contrast to dexrazoxane, DMPS pre-treatment did not prevent the decline in body weight in weeks 8–11 caused by daunorubicin, actually worsened mortality (26.7% vs 40.0%), did not ameliorate daunorubicin-induced nephrotic syndrome, and did not prevent the occurrence of the severe myocardial lesions. Unlike dexrazoxane, a lack of protective effect of DMPS against chronic daunorubicin toxicity in rabbits was demonstrated. The underlying cause may consist in the fact that DMPS does not efficiently chelate tissue iron and thus may not prevent the formation of oxygen free radicals.

sible protective action of DMPS against chronic daunorubicin toxicities, primarily against cardiotoxicity, in rabbits. The action of DMPS was compared with that of a reference agent -dexrazoxane.

Material and methods
Animals. Medium size Chinchilla male rabbits (weight range 2.80-3.30 kg at the beginning of the experiment) were used. The experiments complied with the "European Convention for the Protection of Vertebrate Animals Used for Experimental and other Scientific Purposes" (1) and were under the supervision of the Ethical Committee of the Faculty of Medicine in Hradec Králové.
Invasive heamodynamic measurement. At the end of the experiment the rabbits were anaesthetized with sodium pentobarbitone (25 mg/kg i.v.). Polyethylene catheters (1.6/2.4 mm; filled with heparinized saline, 10 IU/ml) were introduced into the right femoral artery and through the left carotid artery into the left cardiac ventricle, respectively. The catheters were connected to Gold Statham P 23 ID pressure transducers (USA). The mean arterial blood pressure, left ventricular pressure, and left ventricular differential pressure (dP/dt max ) were recorded following a 15 min equilibrium period after an instrumentation with the use of a polygraph Biomedica C6b (Italy) and a differentiator (VÚFB Prague, Czech Republic).
Histological examination. Tissue blocks of transversally sectioned heart ventricles (the region under the atria) were fixed by immersion in 10% buffered formalin. Paraffin sections (7 µm) were regularly stained with Masson's blue trichrome. A 5-point scale of morphological changes was applied for light microscopic semiquantitative evaluation of myocardial lesions (12).
Myocardial elements content assay. Calcium, potassium, magnesium, iron and selenium content were measured in samples of the dried myocardial tissue after microwave digestion with nitric acid and hydrogen peroxide. Calcium, magnesium, and selenium were determined by atomic absorption spectrophotometry using an Unicam Sollar analyser 959 (USA). Potassium and iron were measured photometrically using an Eppendorf Efox 5053 (Germany) and Hitachi 717 (Japan), respectively.
Statistical analysis. The data are expressed as the mean ± SEM for n observations. The homogeneity of variances between groups was tested by F test and the appropriate unpaired t-test was used for the following comparisons: daunorubicin and DMPS groups were compared to the control group; "treated" groups (i.e. dexrazoxane + daunorubicin and DMPS + daunorubicin) were compared with the daunorubicin group. Changes in data within groups (i.e. the 5 th and 11 th weeks in comparison with the 1 st week) were tested by the paired t-test. A χ 2 test was used to determine the significance of differences in the severity of cardiomyopathy scores between groups. P < 0.05 was used as the level of statistical significance.

Results
General toxicity. Treatment with DMPS did not prevent a decline in body weight caused by daunorubicin administration starting in week 9 ( Fig. 1). Mortality encountered in the daunorubicin group achieved 26.7%, DMPS co-administration actually worsened it to 40.0%. In contrast, no mortality was observed in rabbits pre-treated with dexrazoxane. Growth curve displayed nearly the same shape as in the control group.
Biochemical parameters. Important changes in biochemical parameters are summarized in Table 1. These changes, characteristic for the nephrotic syndrome induced by daunorubicin, were partly reduced by dexrazoxane, but not by DMPS co-administration. Other measured parameters either did not show any consistent changes or the changes were difficult to interpret (e.g. changes in CK activity were affected by repeated i.m. injections). c -compared to the control group (P < 0.05)

Tab. 2:
Parameters of the invasive haemodynamic measurement at the end of the experiment (week 11).  Note: values not measured due to interference between fractions of albumin and α 1 -globulin compared to: * -initial value, c -the control group, d -the daunorubicin group (P < 0.05) Invasive haemodynamic measurement. Daunorubicin significantly depressed left ventricular contractility indicated by a decrease in dP/dt max (Table 2). DMPS itself did not change this parameter. Pre-treatment with both dexrazoxane and DMPS tended to attenuate depressive effect of daunorubicin. None of the treatment regimens changed significantly the mean blood pressure.
Postmortem examination. Marked signs of congestion -ascites, pleural effusion and hydropericardium -were observed in 26.7% of rabbits treated with daunorubicin and in 20.0% of animals pre-treated with DMPS. In other groups of rabbits no clear signs of congestion were visible.
Histopathology. The evaluation of myocardial damage is summarised in Table 3. The normal appearance of the myocardial tissue of control rabbits is shown in Fig. 2. Daunorubicin caused the disperse toxic damage of the rabbit myocardium followed by proliferation of fibrotic tissue as a reparative process (Fig. 3). The pre-treatment with dexrazoxane prevented the occurrence of this severe damage (Fig. 4) in comparison with DMPS pre-treatment, where the results were similar to or worse than the results in daunorubicin group (Fig. 5). DMPS itself did not cause any significant changes in the myocardium, i.e. the appearance of the heart tissue resembled the normal stage.
Myocardial content of elements. Daunorubicin caused a significant increase in the content of calcium and a significant decrease in potassium and magnesium concentration. Cardiac levels of iron and selenium were not significantly changed (Table 4). DMPS decreased calcium concentration by more than one third as compared to the controls, tended to decrease potassium and magnesium concentration. The concentration of iron and selenium was not significantly affected. Dexrazoxane pre-treatment prevented "calcium overload" induced by daunorubicin, other changes in cardiac elements concentration induced by daunorubicin were not significantly affected, though some tendency to a decrease in iron and selenium concentration was observed. DMPS co-administration also prevented an increase in calcium concentration induced by daunorubicin; levels of other elements were not practically changed by DMPS pre-treatment.

Discussion
Using the rabbit model of chronic anthracycline cardiotoxicity, which has been widely used in the studies of various protectants (6,9,12,16,17,40,41), the present authors studied a possible protective action of DMPS against daunorubicin toxicity, especially cardiotoxicity. The selected dose of DMPS, i.e. 50 mg/kg i.v., is comparable (molar basis) with the usual dexrazoxane doses in fixed combinations with anthracyclines (weight ratio 20:1) (23).
General toxicity, postmortem examination. Daunorubicin caused a significant decrease in body weight, especially in the last three weeks of the experiment (Fig. 1). Unlike DMPS, dexrazoxane pre-treatment normalized the growth curve of rabbits. Similarly, mortality induced by daunorubicin (26.7%) was prevented by dexrazoxane. This is consistent with the results of other experimental studies (16,19). In contrast, DMPS co-administration actually worsened mortality induced by daunorubicin (40.0%). Unlike DMPS, dexrazoxane prevented the occurrence of fluid accumulation in all pre-treated animals.
Biochemical parameters. The most marked changes in biochemical parameters demonstrated a nephrotoxic action of daunorubicin. Pathogenesis of anthracycline nephrotoxicity is not clear as yet. Most studies support a role of oxidative stress and lipid peroxidation (10,31,32,42); some authors, however, emphasize other mechanisms, e.g. interference with DNA metabolism (5). Partial protection against daunorubicin nephrotoxicity by dexrazoxane noted in the present study suggests a certain role of oxygen free radicals in its pathogenesis. In contrast, DMPS failed to prove any nephroprotective action. In blood, DMPS is rapidly oxidized to disulfide forms (30). In the kidney, disulfides may be reduced to the parent DMPS involving a glutathione (GSH)-disulfide exchange reaction (35). This may ensure the intracellular chelating activity of reduced DMPS, but at the expense of the GSH content and subsequent attenuation of antioxidant mechanisms involving GSH. Moreover, iron chelating activity of DMPS in renal tissue was not demonstrated in rats given DMPS in a cumulative dose 1.26 g/kg i.p. during 3 week interval (38).
Invasive haemodynamic measurement. Chronic daunorubicin cardiomyopathy manifested by a significant decrease in dP/dt max -a parameter of left ventricular contractility (26). DMPS alone did not change this parameter. Both dexrazoxane and DMPS pre-treatment partially non-significantly diminished a decrease in cardiac contractility induced by daunorubicin. Daunorubicin-induced decrease in cardiac contractility was not accompanied by a significant decrease in the mean blood pressure, possibly by a counteracting activation of vasoconstrictor mechanisms (3).
Histopathology. In compliance with some previous studies (17,18,40) conspicuous differences in the extent and severity of the myocardial damage were found mainly between daunorubicin and dexrazoxane-daunorubicin groups. Though the differences between the pre-treated groups and the daunorubicin group were not statistically significant (Table 3), unlike DMPS, dexrazoxane prevented the severe myocardial damage. In comparison with the above-mentioned studies (17,18,40), presence of many necrotic cells accompanied by mononuclear infiltrate (macrophages and scattered lymphocytes), and followed by proliferation of the fibrotic tissue were characteristics of daunorubicin-induced cardiomyopathy in our long-lasting (11 weeks) experiment.
Myocardial content of elements. A significant increase in myocardial calcium concentration (calcium overload) caused by daunorubicin observed in this study has been a welldocumented feature of anthracycline cardiotoxicity. This may result from the block of calcium-release channels in sarcoplasmic reticulum and subsequent accumulation of calcium (37), and from the inhibition of transport systems regulating calcium movement (33). Calcium overload induced by daunorubicin was prevented by pre-treatment with both dexrazoxane and DMPS. However, the underlying mechanisms of the preventive action seem to be different for individual drugs. By binding intracellular iron dexrazoxane decreases harmful cellular biochemical effects induced by anthracyclines, i.e. oxygen-derived free radicals production (14,34,44), lipid peroxidation and subsequent damage of cellular membrane systems whose integrity is essential for cellular calcium homeostasis (46). Lesser extent of myofibrilar damage was seen also in this study as dexrazoxane prevented occurrence of the severe myocardial damage, i.e. stage 3 and 4 (necrosis of the cells) ( Table 3, Fig.  4). In contrast, DMPS prevented calcium overload produced by daunorubicin possibly by direct chelation of calcium ions (Table 4). On the other hand, the present study did not prove any iron chelating action of DMPS, which was proposed hypothetically (22) and demonstrated in in vitro experiments (24). However, in in vivo studies, DMPS, used even in higher doses than in the present study, did not markedly change iron content in various tissues including the cardiac one (36,38). A tendency to an increase in selenium concentration in daunorubicin group may possibly reflect an up-regulation of antioxidant enzyme gene expression and an increase in glutathione peroxidase (GSH-Px) activity in response to oxidative stress. This has been demonstrated in various models of anthracycline cardiotoxicity (21,47). GSH-Px has been shown to serve as a major metabolic form of selenium against oxidative stress (7). Unlike DMPS, pre-treatment with dexrazoxane partially (non-significantly) attenuated an increase in selenium concentration. This may reflect partial reduction of oxidative stress induced by daunorubicin.

Conclusions
Data obtained in the present study suggest that DMPS does not appear to be an efficient iron chelator in vivo, and as the result, in a dose used in this study it did not demonstrate an effective action against daunorubicin toxicity in rabbits.