J. Endocrinol. Invest. 15.79-84,1992
Doxorubicin cardiotoxicity is associated with alterations of plasma levels of atrial natriuretic factor N. Bernardini*, C. Agen*, S. Favilla**, R. Danesi*, and M. Dei Tacca* *Istituto di Farmacologia Medica, and **Clinica Medica I, Universita di Pisa, Pisa, Italy and 37.9 ± 4.1 pg/ml, 180 and 360th minute, respectively). Rats treated with multiple doses of doxorubicin showed a significant increase in plasma ANF levels at the 21st (88.3 ± 7.7 pg/ml) and 31st day (61.0 ± 14.3 pg/ml) of the study compared to vehicle-treated animals at the same time points (41.8 ± 8.0 and 26.5 ± 7.2 pg/ml, respectively). At the 42nd day plasma ANF concentration in doxorubicin-treated rats was not significantly different fram vehicle-treated rats. In both studies ANF level changes occurred in the setting of acute or chronic doxorubicin-induced cardiac damage, as evidenced by alterations of hemodynamic and ECG parameters, These results suggest that the significant alterations in plasma ANF levels during doxorubicin treatment represent a new aspect of cardiac toxicity of the drug and might contribute to doxorubicin-induced heart damage.
ABSTRACT. In order to determine the involvement of the atrial natriuretic factor (ANF) in a model of drug-induced cardiomyopathy, the effects of a single or repeated doses of doxorubicin on plasma ANF levels were examined. Female Wistar rats were treated with doxorubicin at two different schedules: a single 10 mg/kg iv dose or multiple 3 mg/kg iv doses once a week for 3 weeks; control groups were given vehicle (isotonic saline, 0.9% NaCI) intravenously. ANF was assayed in plasma by a specific and sensitive radioimmunoassay method and cardiac function was evaluated by monitoring of ECG and hemodynamic parameters. In the doxorubicin single-dose study plasma ANF values were measured during a period of 6 hours after dosing and were found to be significantly decreased at the 180th (12.5 ± 2.9 pg/ml) and 360th minute (19.4 ± 1.2 pg/ml) after dosing, compared to vehicle-treated animals (35.1 ± 5.7 INTRODUCTION
(8) found a marked reduction of specific atrial granules, severe macrovacuolations and myofibrillar loss in cardiac tissue from rats given doxorubicin, while Bellini et al. (5) demonstrated in the dog that the drug affects both the number and the morphology of specific atrial granules containing ANF. On the basis of the association between doxorubicin-induced cardiotoxicity and alterations of specific atrial granules, the present study investigates in the rat, a reliable animal model of drug-induced cardiotoxicity (9), the changes in plasma ANF levels after a single or repeated treatment with doxorubicin, The results demonstrate that significant alterations in plasma ANF levels occur in cardiomyopathic animals.
Mammalian cardiocytes are endowed with specific secretory granules containing the atrial nutriuretic factor (ANF), a polypeptidic hormone intimately involved in fluid, electrolyte and blood pressure homeostasis (1). The turnover of ANF is abnormal in cardiovascular disorders such as hypertension (2), congestive heart failure (3) and cardiomyopathies (4,5). Plasma ANF levels increase in different animal models of congestive heart failure which develops after myocardial infarction (6) or hereditary cardiomyopathy (7) Anthracycline antibiotics constitute a major class of chemotherapeutic agents for the treatment of cancer. Among these doxorubicin possesses a very large spectrum of antitumor activity but a dose-dependent cardiotoxicity can occur in patients, limiting the use of the drug during repeated courses of treatment In rats treated with doxorubicin, Solcia et al.
MATERIALS AND METHODS Animals and drugs Adult female Wistar rats, weighing 180 ± 10 g, were housed in temperature-controlled rooms at 22 ± 2 C, 55-65% relative humidity, on a 12-h lighting cycle. The animals were observed for 10 days before entering the study and baseline ECGs were recorded to exclude any cardiac abnormality,
Key-words.· ANF, doxorubicin, cardiotoxicity, rat. Correspondence: Prof. Mano Dei Tacca, MD., Istltuto di Farmacologia Medica, Universita di Pisa, Via Rama, 55, 1-56126 Pisa, Italy.
Received March 14, 1991, aeeepted Oetober 31,1991.
79
N. Bernardini, C. Agen, S. Favil/a, et a/.
Ooxorubicin multiple-dose study Animals were randomly divided into 2 groups of 15 rats each and weekly treated iv with either doxorubicin 3 mg/kg or vehicle (isotonic saline, 0 9% NaCI) once a week for 3 weeks. Doxorubicin dose was chosen on the basis of previous report (9) and has been demonstrated to induce a severe cardiotoxicity and low mortality in Wistar female rats. ECG (lead 11) was recorded weekly until the end of the study (6th week); ORS complex and SexT segment (all expressed in msec), and T wave (!-lV) were measured as previously described (9). At the 21 st, 31 st, and 42nd day of the study, 5 rats were sacrificed at each time point and blood was collected and processed as reported below.
Doxorubicin hydrochloride was obtained by Farmitalia-Carlo Erba (Milano, Italy). The drug was dissolved in vehicle (isotonic saline, 0.9% NaCI) and solutions were freshly prepared and protected from the light prior to each administration. Chemicals used throughout the study were of analytical grade. Ooxorubicin single-dose study Animals were randomly divided into 2 groups of 25 rats each and anesthetized by urethane 1 g/kg ip. The trachea was cannulated to assure free airways and the animals were allowed to breathe room air. Under these conditions urethane anesthesia has been shown to be suitable for cardiovascular investigation (10, 11). A polyethylene catheter with a metal tip was placed in the common carotid artery and tied in place to monitor mean arterial blood pressure (MABP, mmHg) by means of apressure transducer (Statham model P231D) connected to an APC channel of a Battaglia-Rangoni ESO 600 polygraph (Battaglia-Rangoni, Bologna, Italy). The maximun rate of rise of systemic arterial blood pressure (SA dP/dt max , mmHg/sec), was taken as an indirect index of left ventricular contractility (12), and was measured by differentiating the pressure signal by means of an analog device (AO/DP/NS channel, Battaglia-Rangoni, Bologna, Italy) (13). Lead II electrocardiogram was recorded as previously reported (9); the time interval between the apex of both S wave and T wave (SexT segment, msec), the interval between the onset of R wave and the end of S wave (ORS complex, msec), the amplitude of T wave (!-lV) and heart rate (beats/min) were measured and taken as cardiotoxicity endpoints. After the animals had been in stable hemodynamic conditions, they were treated with a single dose of doxorubicin 10 mg/kg iv or vehicle (isotonic saline, 0.9% NaCI) iv. The infusion rate was 0.2 ml/min and the volume injected was 1.0 ml/kg. The doses administered corresponded to the lethai dose at stabilization for 50% of treated animals, LD 5o , for Wistar female rat. Animals sacrificed at 180 and 360 min after doxorubicin or vehicle injection were not subjected to ECG and hemodynamic recording. At 15, 30, 60, 180, 360 min after doxorubicin infusion, 1 ml of carotid blood was taken from 5 animals from each group, and processed as reported below. The use of 5 animals at each time point was necessary to avoid hypovolemia due to blood drawing which could modify ANF release from atrial cardiocytes. An additional group of 5 animals was treated with isotonic saline, as reported above, and blood was immediately drawn to measure basal (time 0) plasma ANF levels.
Plasma sampling and ANF assay Blood was drawn from carotid artery and collected in precooled tubes containing EDTA 1.5 mg/mi and aprotinin 1000 U/ml (Trasylol®, Bayer, Milano, Italy). Sam pies were immediately centrifuged at 4 C (4,000 rpm for 20 min) and plasma was stored at -70 C for a maximum of 8 weeks. ANF assay was performed as previously described (14). Briefly, ANF was extracted from plasma sampies (0.5 ml) with a SepPak C 18 column (Waters Associates, Milford, MA, USA) and measured by radioimmunoassay usirlg a commercially available kit (Peninsula Labs., Inc., Belmont, CA, USA). The assay had amid-range of 20 pg/100 !-ll and the lowest detectable dose 01 ANF was 6 pg/ml. The intra- and interassay variation coefficients were 12.5% and 15.0%, respectively. Statistical analysis Statistical analysis was performed with a NWA STATPAKTM sofware (Northwest Analytical, Inc, Portland, OR, USA) and Honeywell XP computer. Presented data are the mean ± SE of 5 observations. The eflects of treatment and time on ECG and hemodynamic parameters and plasma ANF levels were analyzed by use of two-way analysis of variance (ANOVA) followed by the Tukey test. A p value less than 0.05 was considered significant.
RESULTS Ooxorubicin single-dose study In urethane-anesthetized rats, the volume and infusion rate did not interfere with the basal ANF plasma values which were 35.1 ± 2.2 pg/ml. These values are in agreement with data obtained by Tikkanen et al. (6) in similar experimental conditions. Plasma ANF values in doxorubicin-treated rats were significantly decreased during the 180th (12.5 ± 2.9 pg/ml) and the 360th minute (19.4 ± 1.2 pg/ml) compared
80
ANF and doxorubicin cardiotoxicity
E
Cl
S: c
(Fig. 2); an increase in both MABP (significant at 15-60 min compared to baseline and vehicle-injected animals) and heart rate (significant at 10 min compared to baseline and vehicle-injected rats) were also observed (data not shown). No changes in ORS complex and T wave were recorded.
50 40
0
...
~
'EQ)
30
"'-vehicle (NaCI 0.9%)
0
c 0
0 LI..
Doxorubicin multiple-dose study Rats treated with doxorubicin showed a significant increase in plasma ANF levels at the 21 st (883 ± 7.7 pg/ml) and 31 st day (61.0 ± 14.3 pg/ml) after the first doxorubicin injection compared to baseline (35.1 ± 2.2 pg/ml) and values from vehicle-injected animals at the same time points (41.8 ± 8.0 and 26.5 ± 7.2 pg/ml, respectively, Figure 3). During the last week of the study plasma ANF levels declined (42.0 ± 14.3 pg/ml) and were not significantly different fram baseline and contral values (Fig. 3). The occurrence of a severe degree of cardiomyopathy was documented in rats given doxorubicin by the progressive enlargement of the SexT segment wh ich was the earliest ECG change; this widening became significant at the 21 st day and continued to increase throughout the study (Fig. 4). A flattening of T wave (significant at 28-42 days compared to baseline, data not shown), indicating the occurrence of alterations of the repolarization phase of the heart and a ORS complex enlargeme nt (significant at 28-35 days compared to baseline and vehicle-injected animals, data not shown), indicating a prolongation of the depolarization phase of myocardium, completed the pattern of doxorubicin-induced ECG abnormalities.
20
Z
oe:(
111
E cn (11
10
c::
0 60
0
120
180
240
300
360
Time after treatment (min) Fig. 1 - Plasma ANF levels (pg/ml) in animals given doxorubiein (OXR) 10 mg/kg or vehicle iv. Eaeh point represents the mean of 5 rats ± SE (verlieal bars). 'p < 0..0.5 vs basal va lues (time 0.) and values measured at the same time point in animals given vehiele.
to baseline (35.1 ± 2.2 pg/ml) and vehicle-injected rats (35.1 ± 5.7 and 37.9 ± 4.1 pg/ml at the 180th and 360th minute, respectively, Fig. 1). The early occurrence of acute cardiac toxicity in rats given doxorubicin was documented by ECG alterations, the most severe of which was a progressive widening of the SexT segment, indicating a prolongation of the repolarization phase of the myocardium (Fig. 2). Significant changes were recorded in SA dP/dt max , starting 5 min after the infusion
4
U CI)
.s
3
E
2
cn
cCI)
Cl
U CI) ~
*
Cl
~
400
_E
~ *
*
*
~
cn
~ 200 c( cn
c
.!:
CI)
Cl
600
.s
t; cn
c ca .c
OXR 10 mglkg
J:
E
CI)
800
CI)
0
Cl
c ca .c
U
U
-1
0
15
30
45
60
0
-200 0
15
Time after treatment (min)
81
30
45
60
Fig. 2 - Changes in Sa T segment duration (msee, left) and SA dP/dtmax (mmHg/see, right) in animals given doxorubiein (OXR) 10 mg/kg or vehiele iv. Eaeh point represents the mean of 5 rats ± SE (vertieal bars). *p < 0..0.5 vs basal values (time 0.) and values measured at the same time point in animals given vehicle. Mean baseline values ± SE were: SaT segment. 13.2 ± 1.9 msee; SA dP/dl max , 2930. 7 ± 55.3 mmHg/sec.
N. Bernardini, C. Agen, S. Favi//a, el al.
characteristic of which was an early and progressive increase in the SexT segment, indicating the prolongation of the repolarization phase of the myocardium (9). This ECG change is peculiar to doxorubicin cardiotoxicity and is used to monitor the development of drug-induced cardiotoxicity and to assess its severity (9). Drug infusion induced a rapid, significant depression in the SA dP/dtmax followed by a sustained increase. The short-term impairment in SA dP/dtmax might be due, at least in part, to a direct inhibitory effect of doxorubicin on cardiac function. Rather than a direct drug-induced effect of doxorubicin on cardiac contractility, the increase in SA dP/dt max , observed fram the 30th to the 60th minute after drug infusion, is likely to be the consequence of the complex interaction at the cardiovascular level of the impressive release of vasoactive substances, including catecholamines, histamine and prostaglandins (15) by doxorubicin. During the development of acute doxorubicin cardiotoxicity, the plasma ANF concentration significantly decreased 180 and 360 minutes after drug infusion. In the prese nt study we did not measure the right atrial filling pressure; however, in previous experiments, we have observed that doxorubicin acutely changes the right atrial filling pressure but this change is usually modest and a sustained decrease in this parameter, which could account for a reduced release of ANF from cardiocytes, was never observed. Therefore, a possible reason for the reduction of plasma ANF levels might be a drug-induced alteration of the exocytotic process (16, 17), by which ANF is released from cardiocytes into the bloodstream (17, 18). Doxorubicin-induced alteration of fluidity of cytomembranes (16) and impairment of Ca 2 + movements thraugh cell membranes (19) could alter the exocytosis of ANF granules, a Ca2 +-dependent process (20). The increase in granule density acutely observed in dogs after a single dose of doxorubicin (5) agrees with the present data and might be dependent on an inverse relationship between atrial granule density and plasma ANF levels (18). During the multiple-dose study a significant rise in plasma ANF occurred at the 21 st and 31 st day; this finding was concomitant with the development of SexT segment widening, the most sensitive index of drug-induced cardiomyopathy, indicating a drug-induced prolongation of the repolarization phase of myocardium (9). Together, these results may be interpreted as showing an increased rate of ANF secretion when ventricular function is chronically impaired (21, 22). The increase in ANF plasma levels after repeated doses of doxorubicin are likely to be secondary to drug-induced heart dam-
100 ~
E
0,
S:
80
i...
60
r::; 0
'E GI U
r::; 0
40
u
u.
z
Cl
cu E
20
Treatments
t/l
~
(!I
c:
0
7
0
14
21
28
35
42
Time after first treatment (days) Fig. 3 - Plasma ANF levels (pg/ml) in anima/s given doxorubicin (OXR) 3 mg/kg/week iv for 3 weeks or vehicle iv (arrows). Each point rcpresents the mean of 5 rats± SE (vertical bars) *p < 0.0.5 vs basal values (time o.) and values measured at the same time point in animals given vehic!e.
DISCUSSION The present study provides the first evidence that doxorubicin-induced acute and chranic cardiotoxicity is associated with significant changes in plasma ANF levels. The acute cardiac toxicity of doxorubicin was shown by the occurrence of ECG abnormalities, the most
7
Cl GI
DXR 3 mg/kg \
6
*
t/l
-
.§. r::; GI
E
CI
5
*
Treatments
j
4
GI
t/l
~
3
.S
2
fJ)
GI
CI C
(!I
.c:: U
Vehicle (NaCI 0.9%)
0
0
7
14
21
28
35
42
Time after first treatment (days) Fig. 4 - Changes in SaT segment duration (msec) in animals given doxorubicin (OXR) 3 mg/kg/week iv for 3 weeks or vehic!e Iv (arrows). Each point represents the mean of 5 rats ± SE (vertical bars). *p < 0..0.5 vs basal values (time o.) and values measured at the same time point in anima/s given vehicle.
82
ANF and doxorubicin cardiotoxicity
age, as demonstrated by other authors in experimental (6, 23) and clinical heart disease (3, 24, 25). At the end of the multiple-dose study plasma ANF levels dropped and were not significantly different from those of contra I rats. In this phase hearts were severely damaged by doxorubicin, as showed by the increase in SaT segment; therefore the "inappropriate" reduction in plasma ANF levels is most likely to be due to an impairment of ANF synthesis and/or processing since doxorubicin significantly reduces protein synthesis in chronically damaged heart (26). In conclusion, the present data indicate that doxorubicin induces significant changes in plasma ANF levels of rats; this may be due to the direct action of the drug or to the impaired cardiac function. Regardless the primary cause of the alteration of plasma ANF levels, these changes may contribute to the cardiovascular abnormalitiescharacteristlc of doxorubicin toxicity. Further investigations are necessary to determine which step of ANF synthesis or release represents the target of doxorubicin action.
6.
7.
8.
9.
10.
ACKNOWLEDGMENTS This study was supported by grants from the Italian Association for Cancer Research (A.I.R.C.) and the National Research Council (C.N.R.) special project "Interazioni reciproche tra sistema endocrino. nervoso e immunitario". This work was performed under the technical assistance and collaboration 01 Mr. B. Stacchini and P Duranti.
11.
12.
REFERENCES 1. Needleman P., Blaine E.H., Greenwald J.E., Michener M.L .. Saper C.B., Stockman P.T., Tolunay H.E. The biochemical pharmacology of atrial peptides. Annu. Rev. Pharm. Tox. 29: 23,1989. 2. Taylor MA, Ragsdale NV, Carlos BA, Ayres R., Gear A.R, Phil D. Atrial natriuretic factor in essential hypertension. Life Sci. 44: 603, 1989. 3. Tikkanen 1., Fyhrquist F., Metsarinne K., Leidenius R. Plasma atrial natriuretic peptide in cardiac disease and during infusion in healthy volunteers. Lancet 2. 66, 1985.
13.
14.
15.
4. Saito Y., Nakao K., Arai H, Sugawara A, Morii N., Yamanda T., Itoh H, Shiono S., Mukoyama M., Obata K., Yause H., Ohkubo H., Nakanishi S., Imura H. Atrial natriuretic peptide (ANP) in human ventricle increased gene expression of ANP in dilated cardiomyopathy. Biochem. Biophys. Res. Commun. 148.211,1987. 5. Bellini 0, Danesi R., Bernardini N, Cardini G., Marzilli M, Dei Tacca M. Doxorubicin affects both the number and the mor-
16.
17.
18.
83
phology of specific atrial granules in dog heart. AnticancGrRes. 8:599, 1988. Tikkanen T., Tikkanen I, Fyhrquist F. Elevated plasma atrial natriuretic peptide in rats with myocardial inlarcts. Life Sci. 40.' 659, 1986. Franch HA, Callahan L.T., Blaine E.H. Plasma and atrial content of atrial natriuretic factor in cardiomyopathic hamster. Life Sci. 39. 1151,1986. Solcia E., Ballerini L., Bellini 0., Magrini U., Bmtazzoli C., Tosana G., Sala L., Balconi F., Rallo F. Cardiomyopathy of doxorubicin in experimental animals. Factors affecting the severity, distribution and evolution 01 myocardial lesions. Tumori 67. 461, 1981. Danesi R, Dei Tacca M., Soldani G. Measurement 01 the SexT segment as the most reliable electrocardiogram parameter for the assessment 01 adriamycin-Induced cardiotoxicity in the rat J. Pharmacol. Methods 16:251,1986. Oe Wildt D.J, Oe Jong Y., Hillen F.C, Steerenberg PA, Van Hoesel Q.G.C.M. Cardiovascular effects of doxorubicin-induced toxicity in the intact Lou/MWsl rat and in isolated heart preparations. J. Pharmacol. Exp. Ther. 235.' 234,1985. Maggi CA, Meli A Suitability of urethane anesthesia for physiopharmacological investigations in various systems. Part 2: cardiovascular system. Experientia 42: 292, 1986. Chan SHH, Ong B.T. A simple experimental index for the evaluation 01 inotropic responses. J. Pharmacol. Methods 18.23,1987. Danesi R, Bernardini N., Marchetti A, Bernardini M.C, Dei Tacca M. Acute and chronic cardiac toxicity 01 4' -deoxy -4' iodo-doxorubicin in rats. Cancer Chemother. Pharmacol. 25: 326, 1990. Pedrinelli R., Panarace G., Spessot M. Taddei S., Favilla S., Graziadei L., Lucarini A, Salvetti A Low-dose atrlal natriuretic lactor in primary aldosteronism: renal, hemodynamic, and vascular effects. Hypertension 14.156,1989. Bristow M.R. Anthracycline cardiotoxicity. In: Bristow M.R. (Ed.), Drug-induced heart disease. Elsevier, Amsterdam, 1980, p. 191. Demant E:.J.F., Wassermann K. Doxorubicin induced alterations in lipid metabolism of cultured myocardial cells. Biochem. Pharmacol. 34.1741,1985. Baker PF., Knight D.E. Exocytosis: control by calcium and other factors. Br. Med. Bull. 42. 399,1986. Skepper J.N., Navaratnam V., Martensz ND.
N Bernardini, C Agen, S, Fa villa, et al.
19,
20.
21.
22.
Effects 01 expansion 01 blood volume and bilateral vagotomy on specilic heart granules and release 01 atrial natriuretic peptide in the rat Cell Tissue Res, 258, 211, 1989, Monti E, Piccinini F, Favalli L, Villani F, Role 01 the last-exchanging calcium compartment in the early cardiotoxicity 01 anthracycline analogs. Biochem. Pharmacol. 32· 3303, 1983. von Harsdorl R, Lang R, Fullerton M, Smith AL., Woodcock EA Right atrial dilatation increases inositol-(1, 4, 5) trisphosphate accumulation. Implication lor the control 01 atrial natriuretic peptide release. FEBS Lett. 233. 201, 1988. Wanless R.B., Anad IS, Poole-Wilson PA, Harris P An experimental model 01 chronic cardiac lailure using adriamycin in the rabbit: central haemodynamics and regional blood Ilow. Cardlovasc. Res. 21. 7, 1987. Rankin AJ. Mechanism lor the release 01 atrial natriuretic peptide. Can. J. Physiol. Pharmacol. 65: 1673,1987.
23. Wilkins M.R., Settle SL, Stockmann PT Needleman P. Maximizing the natriuretic ellect 01 endogenous atriopeptin in a rat model 01 heart lailure. Proc. Natl. Acad. Sci USA 87. 6465, 1990. 24. Burnett J.C., Kao pe, Hu O.C., Heser wo .. Heublein 0., Granger JP, Opgenorth T.J., Reeder G.S. Atrial natriuretic peptide elevation in congestlve heart fallure in the human. Science 231.' 1145, 1986. 25. Hirata Y, Hishii M, Matsuoka H, Sugimoto T, lizuka M., Uchida Y, Serizawa T., Sato H., Kohmoto 0., Mochizuki T., Sugimoto T, Miyata A, Kangawa K .. Matsuo H. Plasma concentration 01 ex-human atrial natriuretic polypeptide and cyclic GMP in patients with heart disease. Am. HeartJ. 113.1463,1987. 26. Sazuka Y., Tanizawa H, Takino Y. Ellect 01 adriamycin on ONA, RNA and protein biosynthesis in mouse tissues, in connection with ItS cardiotoxicity. Jpn. J. Cancer Res. 80: 1000,1989.
84
Doxorubicin cardiotoxicity is associated with alterations of plasma levels of atrial natriuretic factor N. Bernardini*, C. Agen*, S. Favilla**, R. Danesi*, and M. Dei Tacca* *Istituto di Farmacologia Medica, and **Clinica Medica I, Universita di Pisa, Pisa, Italy and 37.9 ± 4.1 pg/ml, 180 and 360th minute, respectively). Rats treated with multiple doses of doxorubicin showed a significant increase in plasma ANF levels at the 21st (88.3 ± 7.7 pg/ml) and 31st day (61.0 ± 14.3 pg/ml) of the study compared to vehicle-treated animals at the same time points (41.8 ± 8.0 and 26.5 ± 7.2 pg/ml, respectively). At the 42nd day plasma ANF concentration in doxorubicin-treated rats was not significantly different fram vehicle-treated rats. In both studies ANF level changes occurred in the setting of acute or chronic doxorubicin-induced cardiac damage, as evidenced by alterations of hemodynamic and ECG parameters, These results suggest that the significant alterations in plasma ANF levels during doxorubicin treatment represent a new aspect of cardiac toxicity of the drug and might contribute to doxorubicin-induced heart damage.
ABSTRACT. In order to determine the involvement of the atrial natriuretic factor (ANF) in a model of drug-induced cardiomyopathy, the effects of a single or repeated doses of doxorubicin on plasma ANF levels were examined. Female Wistar rats were treated with doxorubicin at two different schedules: a single 10 mg/kg iv dose or multiple 3 mg/kg iv doses once a week for 3 weeks; control groups were given vehicle (isotonic saline, 0.9% NaCI) intravenously. ANF was assayed in plasma by a specific and sensitive radioimmunoassay method and cardiac function was evaluated by monitoring of ECG and hemodynamic parameters. In the doxorubicin single-dose study plasma ANF values were measured during a period of 6 hours after dosing and were found to be significantly decreased at the 180th (12.5 ± 2.9 pg/ml) and 360th minute (19.4 ± 1.2 pg/ml) after dosing, compared to vehicle-treated animals (35.1 ± 5.7 INTRODUCTION
(8) found a marked reduction of specific atrial granules, severe macrovacuolations and myofibrillar loss in cardiac tissue from rats given doxorubicin, while Bellini et al. (5) demonstrated in the dog that the drug affects both the number and the morphology of specific atrial granules containing ANF. On the basis of the association between doxorubicin-induced cardiotoxicity and alterations of specific atrial granules, the present study investigates in the rat, a reliable animal model of drug-induced cardiotoxicity (9), the changes in plasma ANF levels after a single or repeated treatment with doxorubicin, The results demonstrate that significant alterations in plasma ANF levels occur in cardiomyopathic animals.
Mammalian cardiocytes are endowed with specific secretory granules containing the atrial nutriuretic factor (ANF), a polypeptidic hormone intimately involved in fluid, electrolyte and blood pressure homeostasis (1). The turnover of ANF is abnormal in cardiovascular disorders such as hypertension (2), congestive heart failure (3) and cardiomyopathies (4,5). Plasma ANF levels increase in different animal models of congestive heart failure which develops after myocardial infarction (6) or hereditary cardiomyopathy (7) Anthracycline antibiotics constitute a major class of chemotherapeutic agents for the treatment of cancer. Among these doxorubicin possesses a very large spectrum of antitumor activity but a dose-dependent cardiotoxicity can occur in patients, limiting the use of the drug during repeated courses of treatment In rats treated with doxorubicin, Solcia et al.
MATERIALS AND METHODS Animals and drugs Adult female Wistar rats, weighing 180 ± 10 g, were housed in temperature-controlled rooms at 22 ± 2 C, 55-65% relative humidity, on a 12-h lighting cycle. The animals were observed for 10 days before entering the study and baseline ECGs were recorded to exclude any cardiac abnormality,
Key-words.· ANF, doxorubicin, cardiotoxicity, rat. Correspondence: Prof. Mano Dei Tacca, MD., Istltuto di Farmacologia Medica, Universita di Pisa, Via Rama, 55, 1-56126 Pisa, Italy.
Received March 14, 1991, aeeepted Oetober 31,1991.
79
N. Bernardini, C. Agen, S. Favil/a, et a/.
Ooxorubicin multiple-dose study Animals were randomly divided into 2 groups of 15 rats each and weekly treated iv with either doxorubicin 3 mg/kg or vehicle (isotonic saline, 0 9% NaCI) once a week for 3 weeks. Doxorubicin dose was chosen on the basis of previous report (9) and has been demonstrated to induce a severe cardiotoxicity and low mortality in Wistar female rats. ECG (lead 11) was recorded weekly until the end of the study (6th week); ORS complex and SexT segment (all expressed in msec), and T wave (!-lV) were measured as previously described (9). At the 21 st, 31 st, and 42nd day of the study, 5 rats were sacrificed at each time point and blood was collected and processed as reported below.
Doxorubicin hydrochloride was obtained by Farmitalia-Carlo Erba (Milano, Italy). The drug was dissolved in vehicle (isotonic saline, 0.9% NaCI) and solutions were freshly prepared and protected from the light prior to each administration. Chemicals used throughout the study were of analytical grade. Ooxorubicin single-dose study Animals were randomly divided into 2 groups of 25 rats each and anesthetized by urethane 1 g/kg ip. The trachea was cannulated to assure free airways and the animals were allowed to breathe room air. Under these conditions urethane anesthesia has been shown to be suitable for cardiovascular investigation (10, 11). A polyethylene catheter with a metal tip was placed in the common carotid artery and tied in place to monitor mean arterial blood pressure (MABP, mmHg) by means of apressure transducer (Statham model P231D) connected to an APC channel of a Battaglia-Rangoni ESO 600 polygraph (Battaglia-Rangoni, Bologna, Italy). The maximun rate of rise of systemic arterial blood pressure (SA dP/dt max , mmHg/sec), was taken as an indirect index of left ventricular contractility (12), and was measured by differentiating the pressure signal by means of an analog device (AO/DP/NS channel, Battaglia-Rangoni, Bologna, Italy) (13). Lead II electrocardiogram was recorded as previously reported (9); the time interval between the apex of both S wave and T wave (SexT segment, msec), the interval between the onset of R wave and the end of S wave (ORS complex, msec), the amplitude of T wave (!-lV) and heart rate (beats/min) were measured and taken as cardiotoxicity endpoints. After the animals had been in stable hemodynamic conditions, they were treated with a single dose of doxorubicin 10 mg/kg iv or vehicle (isotonic saline, 0.9% NaCI) iv. The infusion rate was 0.2 ml/min and the volume injected was 1.0 ml/kg. The doses administered corresponded to the lethai dose at stabilization for 50% of treated animals, LD 5o , for Wistar female rat. Animals sacrificed at 180 and 360 min after doxorubicin or vehicle injection were not subjected to ECG and hemodynamic recording. At 15, 30, 60, 180, 360 min after doxorubicin infusion, 1 ml of carotid blood was taken from 5 animals from each group, and processed as reported below. The use of 5 animals at each time point was necessary to avoid hypovolemia due to blood drawing which could modify ANF release from atrial cardiocytes. An additional group of 5 animals was treated with isotonic saline, as reported above, and blood was immediately drawn to measure basal (time 0) plasma ANF levels.
Plasma sampling and ANF assay Blood was drawn from carotid artery and collected in precooled tubes containing EDTA 1.5 mg/mi and aprotinin 1000 U/ml (Trasylol®, Bayer, Milano, Italy). Sam pies were immediately centrifuged at 4 C (4,000 rpm for 20 min) and plasma was stored at -70 C for a maximum of 8 weeks. ANF assay was performed as previously described (14). Briefly, ANF was extracted from plasma sampies (0.5 ml) with a SepPak C 18 column (Waters Associates, Milford, MA, USA) and measured by radioimmunoassay usirlg a commercially available kit (Peninsula Labs., Inc., Belmont, CA, USA). The assay had amid-range of 20 pg/100 !-ll and the lowest detectable dose 01 ANF was 6 pg/ml. The intra- and interassay variation coefficients were 12.5% and 15.0%, respectively. Statistical analysis Statistical analysis was performed with a NWA STATPAKTM sofware (Northwest Analytical, Inc, Portland, OR, USA) and Honeywell XP computer. Presented data are the mean ± SE of 5 observations. The eflects of treatment and time on ECG and hemodynamic parameters and plasma ANF levels were analyzed by use of two-way analysis of variance (ANOVA) followed by the Tukey test. A p value less than 0.05 was considered significant.
RESULTS Ooxorubicin single-dose study In urethane-anesthetized rats, the volume and infusion rate did not interfere with the basal ANF plasma values which were 35.1 ± 2.2 pg/ml. These values are in agreement with data obtained by Tikkanen et al. (6) in similar experimental conditions. Plasma ANF values in doxorubicin-treated rats were significantly decreased during the 180th (12.5 ± 2.9 pg/ml) and the 360th minute (19.4 ± 1.2 pg/ml) compared
80
ANF and doxorubicin cardiotoxicity
E
Cl
S: c
(Fig. 2); an increase in both MABP (significant at 15-60 min compared to baseline and vehicle-injected animals) and heart rate (significant at 10 min compared to baseline and vehicle-injected rats) were also observed (data not shown). No changes in ORS complex and T wave were recorded.
50 40
0
...
~
'EQ)
30
"'-vehicle (NaCI 0.9%)
0
c 0
0 LI..
Doxorubicin multiple-dose study Rats treated with doxorubicin showed a significant increase in plasma ANF levels at the 21 st (883 ± 7.7 pg/ml) and 31 st day (61.0 ± 14.3 pg/ml) after the first doxorubicin injection compared to baseline (35.1 ± 2.2 pg/ml) and values from vehicle-injected animals at the same time points (41.8 ± 8.0 and 26.5 ± 7.2 pg/ml, respectively, Figure 3). During the last week of the study plasma ANF levels declined (42.0 ± 14.3 pg/ml) and were not significantly different fram baseline and contral values (Fig. 3). The occurrence of a severe degree of cardiomyopathy was documented in rats given doxorubicin by the progressive enlargement of the SexT segment wh ich was the earliest ECG change; this widening became significant at the 21 st day and continued to increase throughout the study (Fig. 4). A flattening of T wave (significant at 28-42 days compared to baseline, data not shown), indicating the occurrence of alterations of the repolarization phase of the heart and a ORS complex enlargeme nt (significant at 28-35 days compared to baseline and vehicle-injected animals, data not shown), indicating a prolongation of the depolarization phase of myocardium, completed the pattern of doxorubicin-induced ECG abnormalities.
20
Z
oe:(
111
E cn (11
10
c::
0 60
0
120
180
240
300
360
Time after treatment (min) Fig. 1 - Plasma ANF levels (pg/ml) in animals given doxorubiein (OXR) 10 mg/kg or vehicle iv. Eaeh point represents the mean of 5 rats ± SE (verlieal bars). 'p < 0..0.5 vs basal va lues (time 0.) and values measured at the same time point in animals given vehiele.
to baseline (35.1 ± 2.2 pg/ml) and vehicle-injected rats (35.1 ± 5.7 and 37.9 ± 4.1 pg/ml at the 180th and 360th minute, respectively, Fig. 1). The early occurrence of acute cardiac toxicity in rats given doxorubicin was documented by ECG alterations, the most severe of which was a progressive widening of the SexT segment, indicating a prolongation of the repolarization phase of the myocardium (Fig. 2). Significant changes were recorded in SA dP/dt max , starting 5 min after the infusion
4
U CI)
.s
3
E
2
cn
cCI)
Cl
U CI) ~
*
Cl
~
400
_E
~ *
*
*
~
cn
~ 200 c( cn
c
.!:
CI)
Cl
600
.s
t; cn
c ca .c
OXR 10 mglkg
J:
E
CI)
800
CI)
0
Cl
c ca .c
U
U
-1
0
15
30
45
60
0
-200 0
15
Time after treatment (min)
81
30
45
60
Fig. 2 - Changes in Sa T segment duration (msee, left) and SA dP/dtmax (mmHg/see, right) in animals given doxorubiein (OXR) 10 mg/kg or vehiele iv. Eaeh point represents the mean of 5 rats ± SE (vertieal bars). *p < 0..0.5 vs basal values (time 0.) and values measured at the same time point in animals given vehicle. Mean baseline values ± SE were: SaT segment. 13.2 ± 1.9 msee; SA dP/dl max , 2930. 7 ± 55.3 mmHg/sec.
N. Bernardini, C. Agen, S. Favi//a, el al.
characteristic of which was an early and progressive increase in the SexT segment, indicating the prolongation of the repolarization phase of the myocardium (9). This ECG change is peculiar to doxorubicin cardiotoxicity and is used to monitor the development of drug-induced cardiotoxicity and to assess its severity (9). Drug infusion induced a rapid, significant depression in the SA dP/dtmax followed by a sustained increase. The short-term impairment in SA dP/dtmax might be due, at least in part, to a direct inhibitory effect of doxorubicin on cardiac function. Rather than a direct drug-induced effect of doxorubicin on cardiac contractility, the increase in SA dP/dt max , observed fram the 30th to the 60th minute after drug infusion, is likely to be the consequence of the complex interaction at the cardiovascular level of the impressive release of vasoactive substances, including catecholamines, histamine and prostaglandins (15) by doxorubicin. During the development of acute doxorubicin cardiotoxicity, the plasma ANF concentration significantly decreased 180 and 360 minutes after drug infusion. In the prese nt study we did not measure the right atrial filling pressure; however, in previous experiments, we have observed that doxorubicin acutely changes the right atrial filling pressure but this change is usually modest and a sustained decrease in this parameter, which could account for a reduced release of ANF from cardiocytes, was never observed. Therefore, a possible reason for the reduction of plasma ANF levels might be a drug-induced alteration of the exocytotic process (16, 17), by which ANF is released from cardiocytes into the bloodstream (17, 18). Doxorubicin-induced alteration of fluidity of cytomembranes (16) and impairment of Ca 2 + movements thraugh cell membranes (19) could alter the exocytosis of ANF granules, a Ca2 +-dependent process (20). The increase in granule density acutely observed in dogs after a single dose of doxorubicin (5) agrees with the present data and might be dependent on an inverse relationship between atrial granule density and plasma ANF levels (18). During the multiple-dose study a significant rise in plasma ANF occurred at the 21 st and 31 st day; this finding was concomitant with the development of SexT segment widening, the most sensitive index of drug-induced cardiomyopathy, indicating a drug-induced prolongation of the repolarization phase of myocardium (9). Together, these results may be interpreted as showing an increased rate of ANF secretion when ventricular function is chronically impaired (21, 22). The increase in ANF plasma levels after repeated doses of doxorubicin are likely to be secondary to drug-induced heart dam-
100 ~
E
0,
S:
80
i...
60
r::; 0
'E GI U
r::; 0
40
u
u.
z
Cl
cu E
20
Treatments
t/l
~
(!I
c:
0
7
0
14
21
28
35
42
Time after first treatment (days) Fig. 3 - Plasma ANF levels (pg/ml) in anima/s given doxorubicin (OXR) 3 mg/kg/week iv for 3 weeks or vehicle iv (arrows). Each point rcpresents the mean of 5 rats± SE (vertical bars) *p < 0.0.5 vs basal values (time o.) and values measured at the same time point in animals given vehic!e.
DISCUSSION The present study provides the first evidence that doxorubicin-induced acute and chranic cardiotoxicity is associated with significant changes in plasma ANF levels. The acute cardiac toxicity of doxorubicin was shown by the occurrence of ECG abnormalities, the most
7
Cl GI
DXR 3 mg/kg \
6
*
t/l
-
.§. r::; GI
E
CI
5
*
Treatments
j
4
GI
t/l
~
3
.S
2
fJ)
GI
CI C
(!I
.c:: U
Vehicle (NaCI 0.9%)
0
0
7
14
21
28
35
42
Time after first treatment (days) Fig. 4 - Changes in SaT segment duration (msec) in animals given doxorubicin (OXR) 3 mg/kg/week iv for 3 weeks or vehic!e Iv (arrows). Each point represents the mean of 5 rats ± SE (vertical bars). *p < 0..0.5 vs basal values (time o.) and values measured at the same time point in anima/s given vehicle.
82
ANF and doxorubicin cardiotoxicity
age, as demonstrated by other authors in experimental (6, 23) and clinical heart disease (3, 24, 25). At the end of the multiple-dose study plasma ANF levels dropped and were not significantly different from those of contra I rats. In this phase hearts were severely damaged by doxorubicin, as showed by the increase in SaT segment; therefore the "inappropriate" reduction in plasma ANF levels is most likely to be due to an impairment of ANF synthesis and/or processing since doxorubicin significantly reduces protein synthesis in chronically damaged heart (26). In conclusion, the present data indicate that doxorubicin induces significant changes in plasma ANF levels of rats; this may be due to the direct action of the drug or to the impaired cardiac function. Regardless the primary cause of the alteration of plasma ANF levels, these changes may contribute to the cardiovascular abnormalitiescharacteristlc of doxorubicin toxicity. Further investigations are necessary to determine which step of ANF synthesis or release represents the target of doxorubicin action.
6.
7.
8.
9.
10.
ACKNOWLEDGMENTS This study was supported by grants from the Italian Association for Cancer Research (A.I.R.C.) and the National Research Council (C.N.R.) special project "Interazioni reciproche tra sistema endocrino. nervoso e immunitario". This work was performed under the technical assistance and collaboration 01 Mr. B. Stacchini and P Duranti.
11.
12.
REFERENCES 1. Needleman P., Blaine E.H., Greenwald J.E., Michener M.L .. Saper C.B., Stockman P.T., Tolunay H.E. The biochemical pharmacology of atrial peptides. Annu. Rev. Pharm. Tox. 29: 23,1989. 2. Taylor MA, Ragsdale NV, Carlos BA, Ayres R., Gear A.R, Phil D. Atrial natriuretic factor in essential hypertension. Life Sci. 44: 603, 1989. 3. Tikkanen 1., Fyhrquist F., Metsarinne K., Leidenius R. Plasma atrial natriuretic peptide in cardiac disease and during infusion in healthy volunteers. Lancet 2. 66, 1985.
13.
14.
15.
4. Saito Y., Nakao K., Arai H, Sugawara A, Morii N., Yamanda T., Itoh H, Shiono S., Mukoyama M., Obata K., Yause H., Ohkubo H., Nakanishi S., Imura H. Atrial natriuretic peptide (ANP) in human ventricle increased gene expression of ANP in dilated cardiomyopathy. Biochem. Biophys. Res. Commun. 148.211,1987. 5. Bellini 0, Danesi R., Bernardini N, Cardini G., Marzilli M, Dei Tacca M. Doxorubicin affects both the number and the mor-
16.
17.
18.
83
phology of specific atrial granules in dog heart. AnticancGrRes. 8:599, 1988. Tikkanen T., Tikkanen I, Fyhrquist F. Elevated plasma atrial natriuretic peptide in rats with myocardial inlarcts. Life Sci. 40.' 659, 1986. Franch HA, Callahan L.T., Blaine E.H. Plasma and atrial content of atrial natriuretic factor in cardiomyopathic hamster. Life Sci. 39. 1151,1986. Solcia E., Ballerini L., Bellini 0., Magrini U., Bmtazzoli C., Tosana G., Sala L., Balconi F., Rallo F. Cardiomyopathy of doxorubicin in experimental animals. Factors affecting the severity, distribution and evolution 01 myocardial lesions. Tumori 67. 461, 1981. Danesi R, Dei Tacca M., Soldani G. Measurement 01 the SexT segment as the most reliable electrocardiogram parameter for the assessment 01 adriamycin-Induced cardiotoxicity in the rat J. Pharmacol. Methods 16:251,1986. Oe Wildt D.J, Oe Jong Y., Hillen F.C, Steerenberg PA, Van Hoesel Q.G.C.M. Cardiovascular effects of doxorubicin-induced toxicity in the intact Lou/MWsl rat and in isolated heart preparations. J. Pharmacol. Exp. Ther. 235.' 234,1985. Maggi CA, Meli A Suitability of urethane anesthesia for physiopharmacological investigations in various systems. Part 2: cardiovascular system. Experientia 42: 292, 1986. Chan SHH, Ong B.T. A simple experimental index for the evaluation 01 inotropic responses. J. Pharmacol. Methods 18.23,1987. Danesi R, Bernardini N., Marchetti A, Bernardini M.C, Dei Tacca M. Acute and chronic cardiac toxicity 01 4' -deoxy -4' iodo-doxorubicin in rats. Cancer Chemother. Pharmacol. 25: 326, 1990. Pedrinelli R., Panarace G., Spessot M. Taddei S., Favilla S., Graziadei L., Lucarini A, Salvetti A Low-dose atrlal natriuretic lactor in primary aldosteronism: renal, hemodynamic, and vascular effects. Hypertension 14.156,1989. Bristow M.R. Anthracycline cardiotoxicity. In: Bristow M.R. (Ed.), Drug-induced heart disease. Elsevier, Amsterdam, 1980, p. 191. Demant E:.J.F., Wassermann K. Doxorubicin induced alterations in lipid metabolism of cultured myocardial cells. Biochem. Pharmacol. 34.1741,1985. Baker PF., Knight D.E. Exocytosis: control by calcium and other factors. Br. Med. Bull. 42. 399,1986. Skepper J.N., Navaratnam V., Martensz ND.
N Bernardini, C Agen, S, Fa villa, et al.
19,
20.
21.
22.
Effects 01 expansion 01 blood volume and bilateral vagotomy on specilic heart granules and release 01 atrial natriuretic peptide in the rat Cell Tissue Res, 258, 211, 1989, Monti E, Piccinini F, Favalli L, Villani F, Role 01 the last-exchanging calcium compartment in the early cardiotoxicity 01 anthracycline analogs. Biochem. Pharmacol. 32· 3303, 1983. von Harsdorl R, Lang R, Fullerton M, Smith AL., Woodcock EA Right atrial dilatation increases inositol-(1, 4, 5) trisphosphate accumulation. Implication lor the control 01 atrial natriuretic peptide release. FEBS Lett. 233. 201, 1988. Wanless R.B., Anad IS, Poole-Wilson PA, Harris P An experimental model 01 chronic cardiac lailure using adriamycin in the rabbit: central haemodynamics and regional blood Ilow. Cardlovasc. Res. 21. 7, 1987. Rankin AJ. Mechanism lor the release 01 atrial natriuretic peptide. Can. J. Physiol. Pharmacol. 65: 1673,1987.
23. Wilkins M.R., Settle SL, Stockmann PT Needleman P. Maximizing the natriuretic ellect 01 endogenous atriopeptin in a rat model 01 heart lailure. Proc. Natl. Acad. Sci USA 87. 6465, 1990. 24. Burnett J.C., Kao pe, Hu O.C., Heser wo .. Heublein 0., Granger JP, Opgenorth T.J., Reeder G.S. Atrial natriuretic peptide elevation in congestlve heart fallure in the human. Science 231.' 1145, 1986. 25. Hirata Y, Hishii M, Matsuoka H, Sugimoto T, lizuka M., Uchida Y, Serizawa T., Sato H., Kohmoto 0., Mochizuki T., Sugimoto T, Miyata A, Kangawa K .. Matsuo H. Plasma concentration 01 ex-human atrial natriuretic polypeptide and cyclic GMP in patients with heart disease. Am. HeartJ. 113.1463,1987. 26. Sazuka Y., Tanizawa H, Takino Y. Ellect 01 adriamycin on ONA, RNA and protein biosynthesis in mouse tissues, in connection with ItS cardiotoxicity. Jpn. J. Cancer Res. 80: 1000,1989.
84
