Fundam. Clin. Phatmacol., 3 (1989) 67-78 0 Elsevier, Paris

67

COMPARATIVE HEMODYNAMIC STUDY OF A NEW AMINOSTEROID : LND-623 WITH ITS 2W-ISOMER: LND-369, AND WITH DIGOXIN

AND DIGOXIGENIN-RHAMNOSIDE IN THE ANESTHETIZED DOG

J. WEISSENBURGER, J. HECKLE l , M. BIOUR, J.M. POIRIER, F.X. JARREAU*, P. JAILLON and G. CHEYMOL

Laboratoire de Pharmacologie, Facultk de Mkdecine Saint-Antoine. 75571 Park Cedex 12, France

(Received 2-7-1987; accepted 6-9-1988)

Summary - Hemodynamic effects of LND-623, a new aminosteroid lacking the C17 lactone ring and the C14 hydroxyl group common to the natural glycosides, were studied in the pentobarbital-anesthetized dog and compared to those of its 2Ocr-isomer LND-369 and of digoxin and digoxigenin-rhamnoside (DRh). Twenty-four mongrel dogs were divided into 4 groups. Group I received either LND-623 or saline on study day 1 and the other drug or saline 1 wk later. Saline was replaced by digoxin in group 11, digoxigenin- rhamnoside in group 111, and LND-369 in group IV. All drugs except LND-369 were infused as 3.10-9 mol-kg-l-min-' over 20 minutes. LND-369 was infused at twice the dose.

LND-623 increased left ventricular dP/dt for at least 3 h with a peak at end-infusion or 15 min later, accompanied by a transient vasopressor effect. LND-369 induced, at twice the dose, an inotropic effect of comparable magnitude but of shorter duration. Inversely, it provoked a more marked and prolonged vasopressor effect than its 2Op-isomer, LND-623. Maximal digoxin inotropic effect occurred later but was of comparable magnitude to that induced by LND-623. Its vasopressor effects reached a plateau rapidly and remained sustained until min 200. Digoxigenin-rhamnoside inotropic but not vasopressor effects are weaker than those of LND-623.

It is concluded that LND-623, although lacking the most common structural features of the natural cardiac glycosides, provoked rapid and sustained inotropic activities with transient vasopressor effects. These time-course effects differ from digoxin, and these differences are unrelated to their sugar-moiety characteristics. LND-623 inotropic effect is twice as potent as its 2Ocr-isomer.

hernodynamic I pentobarbital I dog I LND-623 I dlgoxin

Introduction

Various chemical modifications of cardiac glycosides have been carried out with the aim of enhancing the margins of safety of these drugs. Results have shown that : (1) the unsaturated C17-lactone ring (Brown et al., 1984; Thomas et al., 1974), (2) the C14 hydroxyl group (Naidoo el al., 1974; Fullerton et al., 1979), (3 ) the sugar moie- ty (Brown and Thomas, 1983) and the spatial configuration of the lateral chain (Thomas et al., 1979) can make a significant contribution to digitalis-like activity affecting both potency and time-dependent properties. Two new aminosteroids, LND-623 (3P-rhamnosyloxy 14-P-amino 5 0 (H) pregnan 2q-01) and its 2Ckt-isomer LND-369 synthetized from 5P-pregnane, possess neither the C17 lactone ring, replaced by a short open-chain, nor the C14 hydroxyl group, replaced by an amine group (see

68 J. WEISSENBURGER et al.

Table I). Nevertheless, preliminary screening studies have suggested that LND-623 could possess inotropic properties comparable to those of digoxin (Jarreau et al., 1984; Swinghedauw et al., 1983). These latter drugs differ also by their sugar moieties (tridigitoxoside vs. rhamnoside). Since Brown and Thomas (1983) have shown that tridigitoxoside-digitoxigenin is less potent than rhamnoside-digitoxigenin, it remains to be explored how far rhamnoside substitution of the sugar moiety of digoxin con- tributes to the relative potency of LND-623 and digoxin. Furthermore, cardiac steroids may alter vascular tone and therefore influence the outcome of therapy. So far, only cardiac effects of LND-623 have been reported. These cardiac effects require evalu- ation under controlled conditions with respect to cardiac loads. On the other hand, since this new cardioactive steroid presents a very atypical chemical structure of the lateral group - in relation to cardiac steroids - it may be useful to explore the phar- macodynamic effects of modification of the spatial configuration of this lateral chain by means of a 20a-OH-isomerism.

The aim of our study was thus to investigate hemodynamic properties of LND-623 and to compare them with those of its 20a-isomer LND-369 and of digoxin and digoxigenin-rhamnoside.

Materials and Methods

General procedures Twenty-four mongrel dogs (9-24 kg) were anesthetized with sodium pentobarbital (25 mg/kg IV), with supplemental doses being given as needed. After intubation they were ventilated mechanically with room air (Palmer respirator) and tidal volume was adjusted to maintain the expiratory CO, concentration around 5% (Beckman LB2). Body temperature, measured in the pulmonary artery (see below), was maintained by means of a heating table (Braun Apparatus type 2016). Heart rate (HR) was recorded from the continuously monitored DII electrocardiogram lead. A polyethylene tube was inserted into the central aorta via a femoral artery and connected to a Statham P23 dB transducer for measuring systolic (SBP) and diastolic (DBP) blood pressures. Mean blood pressure (MBP) was calculated by means of the formula: MBP = DBP + (SBP - DBP)/3. A 7F Millar transducer-tip catheter was placed into the left ventricle via the carotid artery for measuring the left ventricular (LV) end-diastolic pressure

Tableau I. Main structural characteristics of digoxin, LND compounds and DRh

D w Sugar moiety C14 C17 lateral chain

Digoxin DRh

LND-623 LND-369

Trigitoxoside OH Rhamnoside OH

lactone ring lactone ring

Rhamnoside NH2 ZY-OH, open short-chain Rhamnoside NH2 20cr-OH. open short-chain

HEMODYNAMIC STUDY OF LND-623 IN DOG 69

(LVEDP) and peak rate of rise in LV pressure (dP/dt). Pressure recordings were made at 50 or 125 mm/s with a Beckman Dynograph RC multichannel recorder. A Swan-Ganz catheter (either 5F pediatric or 7F 93A095 : 15 ern between tip and injectable port) was inserted into the pulmonary artery via the jugular vein and connected to a cardiac output (CO) computer (Edwards 9520A). CO was obtained by thermal dilution of 5 ml of physiologic saline at room temperature injected into the right atrium. CO was measured successively 5 times : the 2 extreme values were discarded and the 3 remaining results averaged. This average was noted as the cardiac output. Total peripheral resistance (TPR) was calculated from the usual formula:

TPR (dynes.se~.cm-~ = 80+ (MBP)/CO.

Protocol Each dog received LND-623 and a second drug which was saline solution (N = 6), digoxin gU = a), digoxigenin-rhamnoside (N = 6)' or LND-369 (N = 6). Dogs were randomly allocated to one drug on study day 1 and the other 1 wk later. Control values for the hemodynamic variables described above were made 45-60 min after anesthesia was started. Either saline or drug was then administered IV over 20 min. Hernodynamic variables were determined im- mediately, then every 30 min until 200 min after the infusion was started.

Drugs All drugs except LND-369 were infused as 3.10-9mol.kg-1.min-1 (i.e. 47 pg/kg digoxin, 29 pg/kg LND-623, and 3 1 pg/kg digoxigenin-rhamnoside total administered doses, respect- ively. Twice the infusion rate was used for LND-369 (total dose, 58 pg/kg). The corresponding flow rate was 1 ml/min for 0.9% saline administration. A 0.25 mg/ml stock solution of the LND compounds or digoxigenin-rhamnoside (DRh) was prepared daily if needed. DRh was dissolved in a vehicle of ethanol (53%), glycerol (36%), and water, and the LND compounds in water and a few drops of 0.2 N HCl. Commercial digoxin solution (Digoxin@ batch nos. 187 and 191) was used also as stock solution. For each drug an appropriate amount of the stock solution was diluted in water to achieve the desired final concentration. All drugs were sup- plied by Laboratoires Nativelle (Longjumeau, France).

Statistics All values were reported as means f SE. Drug- and time-induced variations of hernodynamic measurements were studied for each drug or for saline by means of a two-way analysis of variance followed by Student's t-test for paired series if differences between means were discovered. Differences between overall hernodynamic effects of the drugs were tested by com- parisons of the area under the curve (AUC) for each parameter (analysis of variance followed by Student's t-test). A P value of 0.05 or less was considered statistically significant.

Results

Figs. 1 and 2 summarize the mean increase from basal values after each drug. Detailed results of these increases (mean f SE) are reported Table 111.

Basal values and control group Mean hemodynamic basal values are reported Table 11. Basal HR of the digoxin group was lower than those of the other groups. Basal dP/dt of the LND-369 group was

800

600

400

200

0

-200

J. WEISSENBURGER et al.

-m- -0-

-c

+ +

0 30 60 YO 120 150 180 210

30 1 20 1

L o 0

-10

LND-623 saline digoxin DRh LND-369

0 30 60 YO 120 150 180 210 mh~tes

Fig. 1. Mean variations in maximal rate of rise of ventricular pressure and mean blood pressure after drug or saline infusions. Except for LND-369, equimolecular doses were administered. LND-369 was administered at twice the dose.

higher than those of the LND-623 and digoxin groups. After saline infusion, LVEDP decreased significantly at min 80 and then increased significantly after min 170. After hour 2, saline (or time) induced a significant increase in TPR (Table 111).

Heart rate LND-623, digoxin, DRh, and LND-369 produced no consistent effects on HR. In 5 out of 24 dogs, transient ventricular tachycardias occurred at the end of, or for 15 min after LND-623 infusion, but were no longer present after the next 15 min.

HEMODYNAMIC STUDY OF LND-623 IN DOG 71

-3 -

3

m

0

0 30 60 90 120 150 180 210

1

* LND-623 4 saline + digoxin + DRh 4 LND-369

0 30 60 90 120 150 180 210 minutes

Fig. 2. Mean variations in LVEDP and calculated TPR after drug or saline infusions. Except for LND-369, equimolecular doses were administered. LND-369 was administered at twice the dose.

Blood pressure LND-623 increased MBP only during the infusion and the following 15 min. LND-369 increased MBP significantly throughout the study with a maximal effect at 140 min. Digoxin induced similar time-dependent properties but these were more marked. DRh provoked only weak, nonsignificant pressor effects (Fig. 1, Table 111). The overall pressor effects of LND-623 differed significantly from those of digoxin and DRh (Table IV).

72 J. WEISSENBURGER et al.

Table 11. Basal hemodynamic values (to) recorded before drug or saline infusions.

Saline LND-623 Digoxin DRh LND-369

HR bpm 134 k8.2 138 k4.3 111 k8.2. 136 f 8.3 142 rt4.6 MBP mmHg 131 k8.0 132 k4.0 128 k8.0 133 k10 129 k7.0 TPR dyneswc. 5.3k0.8 5.2k0.4 6.4k0.8 6.3k 0.7 4.6rt0.4

LVEDP mmHg 8 kl .1 8 k0.7 6.3k0.9 8.2k 1.1 6.4k1.2 dP/dt mmHg-l0-3/s 2.3k0.2 2.3kO.l 1.8k0.1 2 . l f 0.3 2.9k0.2. co l/min 2.2k0.3 2.3k0.2 1.7k0.2 1.8k 0.2 2.4rt0.2

cm-5.104

Comparisons with LND-623 are statistically significant at Pe0.05. HR, heart rate; MBP, mean blood pressure; TPR, total peripheral resistance; LVEDP, left ventricular end diastolic pressure; dP/dt, maximal rate of rise of left ventricular pressure; CO, cardiac output; DRh, digoxin-rhamnoside. Values are expressed as mean f SE. (J. Weissenburger. Hemodynamic of LND 623.)

Cardiac contractility LV dP/dt increased rapidly during LND-623 infusion. Maximal increase was observed either at the end of, or for 15 rnin after the infusion. LV dP/dt then decreased slow- ly, remaining significantly elevated throughout the entire experiment. LND-369 in- creased LV dP/dt between 20 and 110 rnin after dosing, with a maximal effect at 20 min. The maximal inotropic effect of digoxin developed 1 h after the end of the infusion and was no longer present 3 h later. DRh increased LV dP/dt between 20 and 80 rnin after dosing with a maximal effect at 80 min (Fig. 1). The overall effect of LND-623 differed significantly from saline and DRh and was comparable to those of digoxin and of the double-dosed LND-369 (Table IV).

L VEDP A slight, significant decrease in LVEDP was observed after LND-623 infusion throughout the entire study (Fig. 2, Table 111). No significant effect on LVEDP was observed after digoxin, DRh, or LND-369 infusion. Significant differences between the mean AUC of each drug-induced effect were observed only between LND-623 and saline (Table IV).

Cardiac output CO increased slightly between SO and 140 rnin after LND-623, but this effect was significant only at min 50. CO decreased significantly between rnin 110 and rnin 200 after DRh administration. The other drugs or saline had no significant effect on CO. The overall decrease in CO observed after LND-369 and DRh was significantly dif- ferent from the effect of LND-623 (Table IV).

HEMODYNAMIC STUDY OF LND-623 IN DOG 73

Table 111. Changes in hemodynamic values after drug or saline infusions.

time Saline LND-623 Digoxin DRh LND-369 (min)

20 50 80

HR 110 @pm) 140

170 200

20 50

MBP 80 (mrnHg) 110

140 170 200

20 50

LVEDP 80 (mmHg) 110

140 170 200

20 50

TPR 80 (dynes-s ec... 110 ... cm-5,10-3 140

170 200

20 50

dP/dt 80 (mmHg/s) 110

140 170 200

20 50 80

co 110 (l/min) 140

170 200

-3.Of5.0 -2.0f4.5

5.7 i 4 .9 1.3 i3 .3

- 2.3 f 5.7 - 4.3 i6 .1 -0.2f7.6

5.3f3.0 1.7f3.5 0.7 f 6.0 5.7 f 5.0 6.2 i 6.6 7.7 f 6.2 9.7i6.1

- 0.2 f 0.7 - 0.5 i 0.4 -2.0f0.Sb - 0.4 f 0.7

1.1 f0.7 1.4 f0.4b 1.0f0.6

-0.150.4 -0.1 f 0 . 3

O . O i 0 . 5 0.2 f 0.4 1.1 f0.4. 1.3 i 0.5' 1.3 i 0.5'

17f42 -40*%

19 f 54 19 f 54

- 4 2 i 144 -52f133 -%f 124

0.0 f 0.2 O.OfO.1 0.1 f 0 . 2 0.0 f 0.2

-0.2f0.1 -0.2f0.1 - 0.2f 0.1

-3.Of4.9 3.0f4.7

-2.Oi4.9 -7.Ort5.0 -6.Of 3.9 -6.Of4.3 -4.Of4.1

21.9 f 2 J b -2.8f4.9 -4.9i3.8 - 0.2 f 2.9

4 .6 i 3.0 6.8i3.1 8.6f2.6'

- 1.9f0.6b -2.1 d ~ 0 . 6 ~ - 2.8 f 0.5b - 2.4 f O S b - 1.9 f 0.6b - 1.8f0.7b - l.5f0.7b

1.1f0.2b - 0.5 f 0.4 - 0.4 f 0.4 -0.1 f 0 . 4

0.2 f 0.4 0.6 i 0.4 0.8f0.3'

6985 112b 772 f 1 36b 607 It 1 16b 528 f Wb 537 f 1 lob 523 i 122b 429f llSb

o.oio. l 0.3iO.l' 0.2 f 0 . l O.lfO.1 O.lfO.1 0 . O f 0.1

-0.1 i O . 1

1l.Oi 8.9 - 8.0f 3.7

7 . 0 f 12.2 1O.Of 15.5 3.0f7.3 l.Of2.0

-6.0i6.5

18.7 f 5.5b 19.2f 5.5b 23.0f6.Sb 24.7 i 7.6b 26.7 f 7.4b 16.5 i 8.9 13.8f4.8

0 . O i 0.7 -0.3 f 0.1 - 0 . 6 i 1.9

0.Of 2.1 0.2 f 2.5 1.5i2.0 2.5 f2 .2

0.4 i 0.4 1.1f0.7 0 . 7 f 1.0 1.2f0.9 1.5 f 0 . 6 1.3 f0 .5 0.8 f 0.4

4 5 0 i 131b 512f76b 748 f 1 22b 649 f sob

708 f 104b 432 f 145' 361 i 159

0.1 fO.1 0.0 f 0.1 0.1 f 0 . 2 O.lfO.1 0.0*0.1

-0.lfO.l - 0.1 f 0.1

-7.Of6.5 - 1.0+6.5

5 . 0 f 11.0 -3.Oi7.8 -9.Oi 10.6 -8.Oi7.3 - 12.0f8.6

8.5 f 2.4 7.7 f 3.4

14.5 i 3.7 11.2i 6.4 7.5 f 5.9

10.0f6.5 9.2 i 6.8

- 1.0f0.9 - 1.3f 1.0 - 1.0f0.9 - 1.2f0.8 - 1 . 2 i 1.6

0.6 f I .9 1.0i2.2

0.6 f 0.2' 0.5 f 0.3 0.4i0.6 1 .9 i 0.5' 1.9f 1.0 2 .6 i 1.1. 2.5fl.O'

236 f 93a 286 f 78. 568 f 79b 265 f 118 62 f 145

151 f 6 8 35 f42

-0.1io.o -0.1fO.1

O.Of0.2 - 0.3 i 0.2' - 0.3 f 0.2 - 0.4 i 0.2' - 0.4 i 0.2'

- 15.0f8.0 -6.Oi8.0 - 9.0f 10.0 -6.Ort11.0 - 6.0 i 14.0 - 2.0 i 14.0 -4.0f 18.0

13.5f4.2b 14.0i 3.Sb 15.5f4.Sb 21.2f 3.8b 21.7 f 5.1. 23.3 f 4 . P 18.2 f6.1'

-2.1 f O . 5 - 1.8i0.8 - 1.6f1.1 - 1.4f 1.6 - 1.8f 1.2 -0.6f 1.5 -0.3i1.8

0.7 i 0.3 1.2f0.5 1.3f0.6 1.4 i 0.7 1.8 f 0.7' 1.9 i 0.7' 1.4 f 0.7.

739 f 1 17b 512f 116b 413 f 108b 493 i 178' 384 f 238 279 i 200 195f218

0.0 i 0.1 - 0.2 f 0.2 -0.2f0.2 - 0.2 f 0.2 - 0.3 f 0.2 - 0.3 f 0.2 -0.1i0.2

Abbreviations, see Table I. 'Drug and time' effects are statistically significant at: a P< 0.05 ; bP<O.Ol.

74 J. WEISSENBURGER et al.

Table IV. Areas under the curves of 'drug and time'-induced changes in hernodynamic values (mean f SE).

Saline LND-623 Digoxin DRh LND-369

HR -2.2f7.8 -6.1f6.7 1.5f12 -11*15 -13k7.3 rninsbpm- 10-

MBP 9.9f 8.7 7.8f5.2 39k9.0" 18f6.5 33 f 8' min-rnrnHg.10-2

TPR 89 f 61 31f59 196f114" 272f98b 270f llOb min-dynes-sec . . . -10 -6cm - 5

LVEDP - 0.2 f 0.4 -4 f0 .9 0.9f2.8' - 1 f 1.7 -2.6* 1.9 min-mrnHg.

dP/dt -3*8.0b 115*20 106f15 45 f 13b 82 f 21. minmmHg. 10-3/s

min.l/min co - 12* 17 17f 19 4 f 15 - 43 f 27' -36f30"

Abbreviations, see Table 11.

Hernodynamic of LND 623.) Comparisons with LND-623 are statistically significant at a P<O.OS, or PcO.01. (J. Weissenburger:

TPR TPR increased markedly and transiently during the infusion of LND-623, returning to below basal values up to 80 min and then increasing again after min 170. LND-369 increased TPR after min 50 and DRh from min 110. Similar variations were observed after digoxin, but these never reached significance (Fig. 2, Table 111). Between-drugs comparisons showed significant differences between the overall effect of LND-623 and those of digoxin, DRh, or LND-369 (Table IV).

Discussion

Results of our study have shown that LND-623 : (0 increased cardiac contractility for more than 3 h ; ( i i ) induced a very transient increase in arterial blood pressure and TPR; (iii) decreased LVEDP slightly; and (iv) had no effect on HR. Results also showed that LND-623 is as potent as digoxin and twice as potent as its 20a-OH- isomer, LND-369, on cardiac contractility and for peripheral vascular effects. In some dogs infusion of LND-623 induced transient ventricular arrhythmias. In contrast to digoxin, the maximal inotropic effect of LND compounds occurs rapidly. The substitution of the tridigitoxoside of digoxin by a rhamnoside does not explain the differences between the effects of LND-623 and digoxin.

HEMODYNAMIC STUDY OF LND-623 IN DOG 75

After h 2, a significant increase in TPR, mainly related to a decrease in cardiac output, was observed in the control group. Similar variations in TPR and CO were described by Priano et al. (1969) after prolonged pentobarbital anesthesia leading to respiratory acidosis. However, in our study, ventilation was artificial and tidal volume was adjusted throughout the entire experiment to maintain the expiratory CO, concentration around 5%, in order to compensate for this respiratory acidosis. Another explanation could be that prolonged artificial ventilation leads to perspiration-induced fluid losses, which may be compensated for by the increase in TPR. Whatever the explanation for that lack of stability in the preparation may be, observations made after h 2 must be analyzed cautiously.

No heart rate changes throughout the study were observed after any drug or saline infusion. Similar results were obtained elsewhere with either LND-623 (Jarreau et al., 1984) or digoxin. Inability to confirm the well known slight bradycardic effect of digoxin is commonly observed in the pentobarbital-anesthetized dog model (Cook et al., 1980; Jarreau et al., 1984). This is usually explained by an autonomic dysregula- tion (i. e. ( i ) pentobarbital-induced vagolytic effects ; and (ii) stress-related sympathetic alterations observed in untrained dogs).

In the present study LVEDP tended to decrease and blood pressure to increase after any drug. These tendencies cannot explain the increase of LV dP/dt (Mason et al., 1971). Therefore, we consider that the observed increase in LV dP/dt reflects a primary effect of the LND compounds digoxin and DRh on cardiac contractility. The delayed maximal inotropic effect and the rapid and sustained increase in blood pressure after digoxin are comparable with results of previous studies (Deutscher et al., 1972; Songu-Mize et al., 1983). We found the magnitude of the maximal inotropic effects of LND-623 and digoxin to be comparable. This result differs only slightly from that reported by Jarreau et al. (1984), who found the effective dose-ranges of these drugs to be comparable, but that LND-623 presented a significantly higher maxi- mal effect. This divergence may result from protocol differences only. Their study used 3 dogs with continuous infusion of pentobarbitone. Pentobarbitone was shown to decrease cardiac contractility progressively (Priano et al., 1969). Pharmacodynamic interaction of the anesthetic and cardiotropic drugs should result in a reduction mainly in their late inotropic effects. As the maximal digoxin effect occurs about 1 h later than that of LND-623, the use of a constant-infusion anesthesia introduced a bias in the comparisons between these maximal effects. On the other hand, the discon- tinuous administration of pentobarbital used in our study may increase intra-individual hemodynamic variations and therefore mask a difference between drugs by reduced reproducibility of the measurements.

On the other hand, differences in mean basal values of both heart rate and LV dP/dt were observed in our study, although the number of dogs included for each drug or saline study was twice that reported by Jarreau et al. (1984). Therefore, the slight differences between these studies can easily be explained by between-group dif- ferences in cardiac performance.

Although the inotropic action of LND-623 persisted throughout the experiment,

76 J. WEISSENBURGER et al.

as with digoxin, its vasopressor effect was no longer apparent 15 min after the end of the infusion. Such a discrepancy between duration of vascular effects and direct cardiac effects is observed with numerous glycosides and is usually explained by their heterogeneous mode of action on the vascular bed (direct effect, neurohumoral secretory effect, and so on). The vascular effects of cardiac glycosides are also known to be dependent on their rate of administration. Ouabain administered by a 10-s in- fusion increased blood pressure to a much greater extent than the equivalent dose administered by a 15-min infusion (DeMots et al., 1978). The fact that the blood pressure effect of LND-623 is rapidly reversible could be explained, therefore, by its pharmacokinetic particularities on the vascular bed or neurohumoral secretory effectors.

In our study, that peak effect of LND-623 was sometimes accompanied by rapidly reversible arrhythmias. Toxic effects of cardiac steroids, like their vasopressor ef- fects, have been shown to be dependent on the rate of infusion. Cook et al. (1980) and Mudge et al. (1978) studied the toxicity of a cardiac steroid using the same doses administered by continuous infusion for the former and successive bolus injections for the latter. Only repeated bolus injections were shown to result in ventricular tachycardia. These discrepancies were interpreted as resulting from higher intermit- tent plasma levels after bolus injections. As LND-623 possesses a more rapid effect on LV dP/dt than digoxin, one could speculate that the myocardial availability of LND-623 should be higher than that of digoxin using the same rate of infusion and should result in a higher frequency of arrhythmias. This pharmacokinetic hypothesis could explain the faster onset of inotropic action and the more frequent occurrence of ventricular arrhythmias with LND-623 than with digoxin. An interesting observa- tion is the rapid reversibility of these cardiac arrhythmias usually encountered with cardiac steroids presenting a brief duration of effect, such as acetylstrophanthidin, but not with long-lasting inotropic agents such as digoxin (Klein et al., 1971). An explanation could be found in the Na,K-ATPase-isoenzyme studies. Recent research on dog heart has identified low- and high-affinity receptors. High-affinity receptors are related mainly to inotropism and low affinity receptors mainly to toxicity (Maix- ent et al., 1987). Consequently, the cardiac effects of LND-623 could be explained if this drug possessed a fast binding rate to both the high- and the low-affinity recep- tors, but a much faster dissociation rate to the low- than to the high-affinity recep- tor. Further studies are necessary to confirm these speculations.

The spatial configuration of the lateral chain has been shown to contribute to the time-dependent properties of digitalis. Actodigin has been shown to be less potent than its ‘isomer’, digitoxigenin-glucose (Cummings and Beaulieu, 1977). LND-369 maximal inotropic effects were, in our study, comparable to those of LND-623, but the latter provoked effects of shorter duration. The sustained vasopressor effect of the 20a-isomer differed from the effects of the 2q-isomer. The differences in kinetics of the inotropic and the vasopressor properties of these 2 isomers remain unexplained.

Although LND-623 lacks the structural characteristics of digitalic derivatives, its inotropic, vasopressor, and toxic effects are those of all classical compounds of this

HEMODYNAMIC STUDY OF LND-623 IN DOG 77

type, including minimal effects on cardiac output. Thus one could question the im- portance of the sugar moiety in the activity of LND-623. The nature of the sugar moiety was shown to participate in the duration of action and in the relative potency of the cardiac glycosides (Cook ef af., 1980; Fullerton ef al., 1980). Our study of digoxigenin-rhamnoside has demonstrated that such a substitution does not provide digoxin with the time-course characteristics of LND-623, in particular the maximal inotropic effect which was, contrary to LND-623, delayed until long after the end of the infusion. Apart from their sugar moieties, the chemical differences between LND-623 and digoxin include the C14 and the C17 modifications. The C14 amine group constitutes the main structural difference between LND-623 genin and pregnane- dJol, the most common progesterone metabolite which, to our knowledge, has never been shown to present any significant cardiac effect. However, at least in the case of actodigin, the sugar addition may contribute unexpectedly to the steroid's activity (Fullerton ef al., 1980) and the C20 and C3 isomerism of LND-623, compared to pregnane-diol, could also be of some importance. A recent unpublished study by Jar- reau has concluded the inactivity of pregnane 3p, 2W-diol rhamnoside (personal com- munication) and, in light of our results, shows the importance of the C14 aminogroup.

In conclusion, the present study has shown that LND-623, although lacking the most common structural features of the natural cardiac glycosides, provokes rapid onset and sustained inotropic activities with transient vasopressor effects. This new aminosteroid is twice as potent as its 20a-isomer: LND-369, and presents a time- course of effect that differs from that of digoxin, with this difference being unrelated to their sugar moiety characteristics.

References

Brown L., Erdmann E. & Thomas R. (1983) Digitalis structure-activity relationship analysis. Conclusions from indirect binding studies with cardiac (Na', K+)-ATPase. Eiochem. Pharm. 32, 2767-2774

Brown L. & Thomas R. (1983) Comparison of the inotropic potencies of some synthetic and natural oc- curing cardiac glycosides using isolated left atrium of guinea pig. Ameim. Forsrh. Drug Res. 33,814-817

Cook L.S., Caldwell R.W. & Nash C.B. (1980) Changes in cardiac glycosides activity produced by aminosugar addition. J. Pharmacol. Exp. Ther. 215, 198-204

Cummings J.R. & Beaulieu G. (1977) Positive inotropic and antiarrhythmic actions of actodigin in dogs. Arch. Inl. Pharmacodyn. 228, 92-98

DeMots M., Rahimtoola S.H., McAnulty J.H. & Porter G.A. (1978) Effects of ouabain on coronary and systemic vascular resistance and myocardial oxygen consumption in patients without heart failure. Am. .I. Cardiol. 41, 88-93

Deutscher R.N., Harrison D.C. & Goldman R.H. (1972) The relation between myocardial 3H-digoxin concentration and its hemodynamic effects. Am. J. Cardiol. 29, 47-55

Fullerton D.S., Rohner D.C., From A.H.L. & Ahmed K. (1979) Digitalis genin activity: Side-group. Car- bony1 oxygen position is a major determinant. Science 205, 917-919

Fullerton D.S., Yoshioka K., Rohrer D.C., From A.H.L. & Ahmed K. (1980) A crystallographic, con- formational energy and biological study of actodigin (AY-22.241) and its genin. Mol. Pharrnacol. 17,

Jarreau F.X., Koenig J.J. & Fenard S. (1984) A new aminosteroid: LND-623. Eur. Heart. J. 5,309-314 43-51

78 J. WEISSENBURGER et al.

Klein M., Nejad N.S., Lown B., Hagemeijer F. & Barr I. (1971) Correlation of the electrical and mechanical changes in the dog heart during progressive digitalization. Circ. Res. 29, 635-645

Mason D.T., Braunwald E., Cove11 J.W., Sonnenblick E.H. & Ross J. Jr. (1971) Assessment of cardiac contractility : the relation between the rate of pressure rise and ventricular pressure during isovolumic systole. Circulation 44, 47-58

Maixent J.M., Charlemagne D., De la Chapelle B. & Lelikre L.G. (1987) Two Na, K-ATPase Isoen- zymes in canine cardiac myocytes. J. Biol. Chem. 262. 12458-12462

Mudge G.H. Jr., Lloyd B.L., Breenblatt D.J. & Smith T.W. (1978) Inotropic and toxic effects of a polar cardiac glycoside derivative in the dog. Circ. Res. 43, 847-854

Naidoo B.K., Witty T.R., Remers W.A. & Besch H.R. (1974) Cardiotonic Steroids I: Importance of 14j-hydroxy group in digitoxigenin. J. Pharmac. Sciences 63. 1391-1394

Priano L.L., Traber D.L. & Wilson R.D. (1%9) Barbiturate anesthesia: An abnormal physiologic situa- tion. J. Pharmacol. Exp. Ther. 165, 126-135

Songu-Mize E., Nash C.B., Elam J.T. & Caldwell R.W. (1983) Metabolism and cardiac actions of a polar aminocardenolide and digoxin in the conscious dog. J. Pharmacol. Exp. Ther. 226, 65-70

Swinghedauw B., Jarreau F.X., Nittenberg A., Mouas C., Preteseille M. & Lelievre L. (1983) Effect of two new glycosides on contractility and on sarcolemmal Na+-K+ ATPase in rate heart. J. Mol. Cell. Cardiol. 15 (Suppl. 2), 55 (abstr.)

Thomas R., Boutagy J. & Gelbart A. (1974) Cardenolid analogs. V. Cardiotonic activity of semisynthetic analogs of digitoxigenin. J. Pharmacol. Exp. Ther. 131, 219-231

Thomas R., Men J., Pitts B.J.R. & Schwartz A. (1979) Cardenolid analogs. An explanation for the unusual properties of AY 22241. Eur. J. Pharmacol. 53, 227-237