Arch Toxicol (1990) 6 4 : 6 2 3 - 6 3 8

Archives of

Toxicology 9 Springer-Verlag 1990

Effects of calcium channel blockers on the development of early rat postimplantation embryos in culture G a b r i e l e Stein, M i t h i l e s h K u n m r S r i v a s t a v a * , H a n s - J o a c h i m M e r k e r , a n d D i e t h e r N e u b e r t Institut ftir Toxikologie und Embryopharmakologie, Freie Universidk Berlin, Garystrasse 5, 1000 Berlin 33 Received July 17, 1989/Accepted June 19, 1990

A b s t r a c t . Rat embryos (9.5-day-old) were cultured for

48 h in the presence of nifedipine (NIF), nimodipine (NIM), nitrendipine (NIT), gallopamil HCI (GAL), verapamil HC1 (VER) and diltiazem HC1 (DIL). The effects on growth and morphogenetic differentiation in vitro were monitored. Dose-response relationships were evaluated, including an assessment of the "no-observed-effect-level" (NOEL) or the "lowest-observed-effect-level" (LOEL), and the lowest concentration tested inducing abnormalities in 100% of the embryos ("100% EL"). The morphological alterations observed at the highest concentrations were very similar for all six drugs. The abnormalities concerned yolk sac circulation and morphology, as well as heartbeat, the morphology of the heart, head, neural tube, or forelimbs, and the shape of the embryo. The abnormal embryos were also growth retarded (decrease in protein content and crown-rump length). Interference with calcium channel functions seems to represent an interesting model for studying a special kind of abnormal prenatal development, especially the differentiation of certain mesenchymal structures. The concentration ranges between NOELs and 100% ELs were found to be: NIM = 0.1 - 1 pg/ml; NIT and VER = 1-10 ktg/ml; DIL = 1 - 3 0 ktg/ml, and LOELs100%ELs were: GAL = 1- 10 p.g/ml; NIF = 10-30 gg/ml. K e y w o r d s : Rat whole-embryo culture - In vitro - Cal-

cium channel blockers - Nifedipine - Nimodipine Nitrendipine - Gallopamil - Verapamil - Diltiazem

Abbreviations: Nifedipine = NIF; Nimodipine = NIM; Nitrendipine = NIT; Gallopamil HC1 = GAL; Verapamil HC1 = VER; Diltiazem HCI = DIL 100%EL = lowest dose tested inducing abnormalities in 100% of the embryos * On leave of absence from: Industrial Toxicology Research Centre, Lucknow, India Offprint requests to: D. Neubert

Introduction

Calcium channel blockers are widely prescribed today for therapeutic purposes and some have been recommended for use during pregnancy for defined indications: e. g. VER for treating fetal paroxysmal tachycardia (Wolff et al. 1980; Lilja et al. 1984; Truccone and Mariona 1985), VER and NIF during tocolytic therapy (Strigl et al. 1980; Read and Wellby 1986), and NIF, NIT and VER for treatment of hypertension (Orlandi et al 1986; Constantine et al. 1987; Laragh 1987; Pipkin and Lawrence 1987). These therapeutic recommendations all concern treatment late in pregnancy after organogenesis has already been completed. For most of the calcium channel blockers data published on reproductive and developmental toxicity are scarce, and most of the information is in the form of unpublished reports presented to regulatory agencies. At least two of these agents, NIF and DIL, have been shown in animal experiments to be able to induce teratogenic effects and to increase the incidence of embryolethality (e. g. Ariyuki 1975; Fukunishi et al. 1980; Cabov and Palka 1984; Yoshida et al. 1988). We report here on results of comparative studies of various calcium channel blockers using the whole-embryo culture system in which adverse effects on early organogenesis can be evaluated. The aim of our studies was to: - evaluate whether these agents exhibit typical adverse effects on embryonic development, - analyse whether the observed effects are basically the same for all the calcium channel blockers, and - establish quantitative relationships, allowing a comparative assessment of the potency of these agents to induce abnormal development in vitro.

With the results obtained in vitro it may be possible to correlate these data with effects observed in studies on reproductive toxicity in vivo. The relevance of such in vitro data for a risk assessment in pregnant women exposed therapeutically to these agents and the principle problems connected with such extrapolations will be discussed.

624

Materials and m e t h o d s

Animal maintenance Wistar rats (Bor: Wisw/spf, TNO; Fa. Winkelmann, Borchen, FRG) were kept under spf conditions at a constant day/night cycle (light from 9:00 a. m. to 9:00 p. m.). They received AltrominR 1324 pellet feed and tap water ad libitum. One male was caged with three females for mating (period from 6:00 to 8:00 a. m.); 7:00 a.m. was regarded as hour 0 of embryonic age if sperm were detected in the vaginal smears.

Whole-embryo culture Rat embryos were cultured for 48 h according to the procedure described by Klug et al. (1985). The embryos (1 to 4 somite stage) were explanted aseptically on day 9.5 of gestation (228-232 h after mating). Three to four embryos were cultivated at 38.8•176 in a 50 ml tightly sealed culture bottle containing 6 ml heat-inactivated and sterile-filtered bovine serum and 1 ml Tyrode's buffer enriched with 525 p.g L-methionine/ml buffer, using a roller system. The gas phase consisted of: 10% 02, 5% CO2, 85% N2 during the first 36 h, and 50% 02, 5% CO2, 45% N2 during the final 12 h of the culture period. No antibiotics were added to the culture medium. Embryos were evaluated using the following criteria: number of somites, crown-rump length, yolk sac diameter, protein content, type and frequency of abnormalities, and a score indicating the development of the various organ anlagen. Furthermore, histological examinations were performed on serial sections. To calculate the score, individual morphologically visible organ anlagen of the embryo such as shape, neural tube, head, eye, ear, heart, forelimb, hindlimb, tail and blood were evaluated. The sum of these values gives the score (a high score indicates a more advanced developmental stage). For statistical analysis a Mann-Whitney test was used to assess significant differences in the variables measured as compared to controls. In addition to the evaluation procedure according to Klug et al. (1985), the yolk sac morphology was routinely assessed. Each concentration was tested in at least two different experimental series performed on different days. Embryos from five to seven dams were used for each tested level. The water-soluble substances diltiazem HCI (Gi3decke, Freiburg, FRG), verapamil HC1 and gallopamil HC1 (Knoll AG, Ludwigshafen, FRG) were added to the culture medium after dissolution in Tyrode's buffer. Nifedipine, nimodipine and nitrendipine (Bayer AG, Wuppertal, FRG) were dissolved in DMSO, and 5 gl of the solution were added to 7 ml culture medium using a 10 gl Hamilton syringe. The solutions were made up freshly immediately before each experiment. Care was taken not to expose the light-sensitive compounds (NIF, NIM and NIT) to normal light, therefore all work was performed under yellow light (Osram L40W-62).

Histology Light microscopy. The embryos were fixed in Bouin's solution and embedded in ParaplastR. Subsequently, 5-10 I-tm thick serial sections were cut and stained with hematoxylin/eosin.

Electronmicroscopy.The embryo-yolk sac unit was fixed in tannic acid (0.5%)-glutaraldehyde (2%) in phosphate buffer (0.1 M, pH 7.2), embedded in Epon R, sectioned using a Reichert microtome and contrasted with uranyl acetate/lead citrate. The evaluation was performed using a Zeiss EMI0 transmission electron microscope.

Results

Dose-response relationships A n o v e r v i e w o f the results o b t a i n e d with the w a t e r - s o l u b l e c o m p o u n d s is g i v e n in T a b l e 1; the results for the w a t e r - i n s o l u b l e substances are listed in T a b l e 2.

To facilitate c o m p a r i s o n , the drug c o n c e n t r a t i o n s are d i v i d e d into three c a t e g o r i e s (Table 3): 1. the N O E L ( n o - o b s e r v e d - e f f e c t - l e v e l ) ; c o n c e r n i n g all tested variables o f the e m b r y o c o m p a r e d to controls; 2. the L O E L ( l o w e s t - o b s e r v e d - e f f e c t - l e v e l ) ; growth (often the m o s t sensitive v a r i a b l e ) is a l r e a d y definitely affected, but all m o r p h o g e n e t i c differentiations are not or o n l y m i l d l y affected and the f r e q u e n c y o f abnormalities o n l y slightly increased. U s u a l l y up to 10% abnormalities are a c c e p t e d as " n o r m a l " u n d e r our e x p e r i m e n t a l conditions. 3. 100%EL, i.e. the l o w e s t concentration tested at w h i c h 100% o f the e m b r y o s show a b n o r m a l i t i e s a n d the d e v e l o p m e n t a l and g r o w t h v a r i a b l e s are s e v e r e l y affected.

Macroscopic evaluations To facilitate an o v e r v i e w o f the results the data are discussed a c c o r d i n g to the effect c a t e g o r i e s observed.

Concentrations: I O0%EL (morphology of embryos and yolk sacs). T h e types o f a b n o r m a l i t i e s o b s e r v e d w h e n testing all six c a l c i u m channel b l o c k e r s with the in vitro syst e m w e r e v e r y similar: t y p i c a l l y , the y o l k sac circulation was a l m o s t a b o l i s h e d , and p a t h o l o g i c a l c h a n g e s o f the heart, head, neural tube, and f o r e l i m b s were seen or the shape o f the e m b r y o was altered. S o m e t i m e s the heart beat was no l o n g e r detectable. U s u a l l y the somites were not affected, s h o w i n g clear b o u n d a r i e s and a r e g u l a r arrangem e n t (Figs. 1, 2 and 3). In m o s t y o l k sacs the b r a n c h - l i k e pattern o f the m a i n v e s s e l s was not e s t a b l i s h e d and no circulation was visible. T h e vessels w e r e s w o l l e n to different extents and f o r m e d a n a s t o m o s e s and c i r c u l a r patterns. T h e m o r p h o l o g i c a l app e a r a n c e o f the y o l k sac surface r e s e m b l e d a " g o l f ball" (Fig. 3.1). A d d i t i o n a l l y , s o m e y o l k sacs were flattened, but not c o l l a p s e d (Fig. 3.2). T h r e e different p a t h o l o g i c a l p h e n o m e n a c o u l d b e obs e r v e d for the heart: the heart tube s h o w e d an additional b u l g e in the region o f the p r o s p e c t i v e left ventricle (Fig. 3.7); the heart tube itself was swollen (Figs. 3.5 and 3.6), and the p e r i c a r d i u m was s w o l l e n (Fig. 2.3 b). A l l neural tubes w e r e c l o s e d (an open c a u d a l neuroporus was o n l y present in three e m b r y o s ) , but their walls s o m e t i m e s s h o w e d a w a v y a p p e a r a n c e . This d i d not necessarily include both sides or the w h o l e length o f the tube (Fig. 3.4). Some affected embryos showed a typical abnormality o f the head: the t e l e n c e p h a l i c v e s i c l e s p r o t r u d e d on both sides in front o f the f o r m e r p r o s e n c e p h a l i c region (Fig. 3.5). A d d i t i o n a l l y , s m a l l and s o m e w h a t irregularly s h a p e d h e a d s occurred. Often size differences b e t w e e n the fight a n d left forel i m b were o b s e r v e d (Fig. 3.3). O n l y obvious size differences w e r e c o n s i d e r e d a b n o r m a l . S i n c e the o b s e r v e d a b n o r m a l i t i e s w e r e rather c o m p l e x , a survey o f the results is g i v e n in T a b l e 4 w h i c h s h o w s the t y p e and f r e q u e n c y o f the o b s e r v e d c l e a r - c u t a b n o r m a l i t i e s and other o b s e r v a t i o n s unusual for d a y 11.5 control e m b r y o s . T h e latter w e r e r e g a r d e d as retardations.

625 Table 1. Calcium channel antagonists: verapamil HCI - gallopamil HCI - diltiazem HCI

Control

Yolk sac diameter (ram)

Crownrump (mm)

4.62 4.44 4.29

3.87 3.72 3.54

Somites (n = )

Protein (gg/E)

27

211 194 167 n = 32

27 26

n = 37 Verapami 1HC1 1 gg/ml n=19

4.62 4.44

3.84 3.66

28 196

26 n=18

159 n=14

27

163 135"* 115 n= 16

3.48

VerapamilHCI 3 gg/ml n=22

4.92 4.68** 4.37

3.60 3.51"* 3.18

VerapamilHCI 10 ~tg/ml n=21

4.20 4.08** 3.99

2.46 2.40** 2.28

19 18"* 17

Gallopamil -HCI 1 ~tg/ml n=20

4.68 4.56 4.28

3.60 3.44** 3.20

26

Gallopamil -HC1 10/ag/ml n=22 DiltiazemHCI 1 gg/ml

4.88 4.62

26 25 n=21

27 25 n= 19

2.46 2.31"*

4.38

2.16

4.68 4.47 4.32

3.84 3.75 3.66

4.62 4.38 4.17

3.54 3.30** 2.97

DiltiazemHC1 30 gg/ml n=20

4.37 4.08**

3.92

2.63 2.46**

2.34

24 22** 21

15 34

15

48

17

257

2x 2 100

9• 3

10 x 4 Ix5 0

38 35

19 34

19

74

25

70**

23**

63 n= 18

17 x 0 3x 2

20 x 0

36

17"*

14

2x 1 3x 2 12 x 3 4x4

38 37

206 n=15 172 133"* 121 n= 16

100

19 • 0 2x I 1x4

38 36

22

27 25* 24 n=20

1x 3

34

20**

26

18 x 0

14

63

243**

5

37 36

57**

28

37 • 0

36

19

27

0

38

16"*

n=20 DiltiazemHC1 l0 gg/ml n=21

157 134"* 125 n= 16

Abn/E

35

37

76 69** 61 n= 18

%Abn

38 37

218

27

4.32

Score

19

17 x 0 2x l 1 x2 1• 1x 1

100

8x 2

3 x3 8x4

For the first 5 columns the boldface numbers represent the median value, the numbers at the top and bottom show the third and the first quartile. * = significantly different from controls at p <0.05 ** = significantly different from controls at p <0.01 %Abn = frequency of abnormalities in % Abn/E= frequency of abnormalities per embryo (i. e. 7 x I = 7 embryos with l abnormality)

Concentrations: LOEL (morphology of embryos and yolk sacs). A f u n c t i o n i n g y o l k sac circulation is easily r e c o g n i z a b l e in control e m b r y o s with thick vessels, transparent tissue and with the reflection o f light in the f l o w i n g bloo d cells. In contrast, detection o f circulation is difficult in y o l k sacs w i t h s m a l le r vessels or with not so transparent tissues. F o r this r e a s o n circulation was s o m e t i m e s difficult to detect but the o v e r a ll yolk sac appearance c o u l d still not be r e g a r d e d as abnormal. T a b l e 5 s u m m a r i z e s the results on the appearance o f the y o l k sacs at concentrations corresponding to a L O E L . A t the c o n c e n t r a t i o n s c o r r e s p o n d i n g to a L O E L the variations o b s e r v e d m o s t l y c o n c e r n e d deviations in shape ( i n c o m p l e t e flexions). In addition, single abnormalities o f

the heart, head or e x t r e m i t i e s w e r e r e c o r d e d within the groups o f a p p r o x i m a t e l y 20 e v a l u a t e d e m b r y o s .

Concentrations: NOEL (morphology of embryos and yolk sacs). At the c o n c e n t r a t i o n d e s i g n a t e d as N O E L the m o r p h o l o g y o f the y o l k sac was always normal. T h e abnormalities f o u n d in t w o e m b r y o s w e r e (a) after 1 g g V E R / m l : " s q u i r r e l - s h a p e d " (defined as an e m b r y o in w h i c h the caudal and cranial parts o f the neural tube are fused and the e m b r y o has failed to rotate; in these e m b r y o s several o r g an anlagen w e r e always affected) and (b) after 0.1/,tg N I M / m I : a head abnormality.

626 Table 2. Calcium channel antagonists: nifedipine, nitrendipine, nimodipine

Control

Yolk sacdiameter (ram)

Crownrump (mm)

4.76 4.47 4.25

3.86 3.72 3.48

27

4.62 4.41 4.20

3.80 3.54 3.36

27

Somites (n = ) 28 26

n = 26 4.50 Nifedipine 10 I.tg/ml n = 25

4.32

3.60 3.42*

26 n = 23

115 n =20 63 57** 48 n=19

3.15

4.71 4.50 4.29

2.34 2.28** 2.10

20 18"* 17

Nitrendipine 1 lag/ml n=20

4.56 4.50 4.38

3.77 3.66 3.60

27

Nitrendipine 10 ~tg/ml n=22

5.16 4.89** 4.68

2.28 2.19"* 2.04

20 19"* 18

Nimodipine 0.1 ~tg/ml n=23

4.56*

Nimodipine 1 I.tg/ml n=19

4.80*

4.74

4.98 4.38

62 53** 52 n=21

26x0

20

20xO 4xl lx4

34

4xl 6x2 7x3 4x4

26 21"* 19

1110

38 38* 36

0

20xO

23 21"* 19

100

llx3 lx4 3x5

4

22 x 0 lxl

7x2

38 38

167 n = 18

0

38 36*

246 191

26

2.58 2.16

180 n = 15

28 27

3.48

2.28**

212 194

26

4.02 3.66

4.44

28

26x0

36

165 135"*

4.11

Nifedipine 30 p.g/ml n=21

158 n = 23

28

0

38 37

26

Abn/E

36

204 180

%Abn

38 37

154 n=24

28 25

Score

213 175

n=26 DMSOControl

Protein (~g/E)

36

21

60

27

20**

51"*

24**

17 n=18

31 n=15

22

95

1xO 4xl lOx2 4x3

For the first 5 columns the boldface numbers represent the median value, the numbers at the top and bottom show the third and the first quartile. * = significantly different from controls at p <0.05 ** = significantly different from controls at p <0.01 %Abn = frequency of abnormalities in % Abn/E = frequency of abnormalities per embryo (i. e. 7 • 1 = 7 embryos with 1 abnormality)

Table 3. Concentrations of the various calcium channel antagonists lead-

Micromorphological evaluations

ing to the corresponding effect categories NOEL flag/ml) Nifedipine Nitrendipine Nimodipine Verapamil HC1 Gallopamil HCI Diltiazem HCI

1 0.1 1 1

LOEL (lag/ml)

100%EL (lag/ml)

10

30 10 1 10 10 30

3 1 10

The molecular weights of the six compounds tested are rather similar, ranging from 346 to 521 (i. e. 1 ~g/ml corresponds to 1.9 ~M GAL, 2 ~tM VER, 2.2 p.M DIL, 2.4 ~tM NIM, 2.8 ~tM NIT, 2.9 ~tM NIF)

Light microscopic findings. T o e m b r y o s a p p e a r i n g normal u p o n

e l u c i d a t e w h e t h e r the macroscopic inspection m i g h t s h o w s i g n s o f a b n o r m a l d e v e l o p m e n t at the m i c r o s c o p i c l e v e l , e m b r y o s e x p o s e d to the v a r i o u s c a l c i u m channel b l o c k e r s at c o n c e n t r a t i o n s c o r r e s p o n d i n g to the N O E L and L O E L w e r e s e c t i o n e d s e r i a l l y and i n v e s t i g a t e d histologically. N e i t h e r c o n t r o l s n o r e m b r y o s treated w i t h c o n c e n t r a tions c o r r e s p o n d i n g to the N O E L o r L O E L r e v e a l e d any h i s t o l o g i c a l a b n o r m a l i t i e s w h e n c o m p a r e d to d a y 11.5 e m b r y o s in v i v o . T h e e m b r y o s e x p o s e d to c o n c e n t r a t i o n s c o r r e s p o n d i n g to the 1 0 0 % E L o f the six d r u g s s h o w e d s i m i l a r h i s t o l o g i c a l d e v i a t i o n s . H o w e v e r , the e x t e n t to w h i c h the d i f f e r e n t reg i o n s o f the e m b r y o s w e r e a f f e c t e d w a s s u b s t a n c e d e p e n dent. G e n e r a l l y , a less d e n s e l y p a c k e d e m b r y o n i c m e s e n c h y m e w a s o b s e r v e d . O n o c c a s i o n this led to u n u s u a l l y

627

Fig. 1. Rat embryos after 48 h of culture in the presence of water-soluble compounds: 1 = Control; 2 a = 1 lag GAL/ml, 2 b = 10 lag GAL/ml; 3 a = 1 lag VER/ml, 3 b = 3 lag VER/ml, 3 c = 10 lag VER/ml; 4 a = 1 lag DIL/ml, 4 b = 10 lag DIL/ml, 4 c = 30 lag DIL/ml. Pictures 2 b, 3 c and 4 c show abnormal embryos exposed to the 100%EL category; 2 a, 3 b, and 4 b embryos of the LOEL category and 3 a and 4 a the embryos of the NOEL category large intra-ernbryonic cavities. Necroses occurred in individual embryonic parts, sometimes accompanied by the described gross morphological abnormalities. Judged by light microscopy these embryos were not considered "dead", since generalized necroses or nuclear pyknoses as clear-cut signs of cell death were not present (Figs. 4 - 9 ) . During normal development of the embryonic heart the outer one-layered myoepithelium grows in a cone-shaped fashion into the cardiac jelly (Fig. 5 a). At the same time, individual cells that have detached from the endothelium migrate into the cardiac jelly. Invasions from the myocardium, as well as from the endothelium, were missing in the swollen cardiac tubes in the ventricle and truncus regions of the VER- and GAL-treated embryos. This led to an arrangement of the cardiac tube that consisted of the one-layered myoepithelium, an almost cell-free cardiac jelly, and the endothelium (Figs. 5 b and 5 d). Death of the contractile cells is less likely, since in most cases necroses,

pyknoses, vacuolisations or similar indications of cell lesion could not be detected by light microscopy. Although morphological investigations did not show any swelling of the cardiac tube in the DIL-, NIF-, NIT- and NIM-treated groups, a varying degree of inhibition of migration was revealed by light microscopic investigation (Fig. 5 c). Necroses also occurred to a varying degree within the neural tube after exposure to all six substances. The frequency of necroses apparently correlated with the frequency of wavy wails of the neural tube, i.e. single necroses were observed in the presence of VER and DIL, whereas addition of NIF, NIT and NIM to the medium caused many necroses (Figs. 6b, c, d). Also in the head region, large cavities due to the almost complete absence of mesodermal cells were seen. The visible changes in the shape of the head were possibly due to the lack of a mesoderm which is normally responsible for the shape (Figs. 5b, c, d; 6a, b).

628

Fig. 2. Rat embryos after 48 h of culture in the presence of the water-insoluble compounds: I = DMSO-Control; 2 a = 10 I.tgNIF/ml, 2 b = 30 ~tg NIF/ml; 3a --- 1 ~tg NIT/ml, 3b= lO~tgNIT/ml; 4a=O.1 ~tg NIM/ml, 4 b = 1 ~tg NIM/ml. Pictures 3 a and 4 a show embryosexposed to the NOEL category,2 a of the LOEL category and 2 b, 3 b and 4 b of the 100%ELcategory.Arrow indicates heart abnormality(swollenpericardium) Since generally sagittal longitudinal sections were prepared for better orientation in the histological investigations, both forelimbs could not always be analyzed in comparison. Thus, the sectional plane was not always optimum to allow a clear-cut statement on the differences in limb size between the fight and left side. Histologically the evaluated limbs were either normal or they showed the described mesenchymal cavities (Fig. 4 d, 7 a). Since these cavities occurred quite often, this pathological effect may have been responsible for the disturbances in limb development. Histologically, the somites did not show any lesions. Mesenchymal portions were also missing in the yolk sac. Light microscopic inspection showed large lacunae. Electron microscopy revealed that these were bordered by vascular endothelium. The lacunae appeared instead of normal vessels which are embedded in the mesoderm between the yolk sac epithelium and the mesothelium (Figs. 4b, c, e, 6a). Electron microscopic findings. The yolk sacs of the explants exposed to 10 ~g NIT/ml were also investigated electron microscopically. The following abnormalities were found: - the first abnormality concerned the cylindrical cells of the yolk sac epithelium. Almost all types of inclusions, such as electron-dense granules, were missing, and in the

cells that still contained inclusions there were decreases in number and size (Figs. 8 b, 9 b). - The second abnormality concerned the mesenchyreal portions of the yolk sac wall. The ultrastructural observations confirmed the light microscopic findings of large cavities between the yolk sac epithelium and the mesothelium of the exocoelomic cavity due to the absence of the mesenchymal cells normally seen surrounding the vessels. The typical endothelial cells were still present. Although not studied in detail, we assume that the electron microscopical appearance of the tissues exposed to the other five drugs will be identical or at least very similar.

Discussion Can our data help evaluate an embryofeto-toxic risk assessment with relevance to man? What is the relevance of the results obtained in our studies for interpreting results obtained from in vivo studies (segment II) in other laboratories? Pharmacokinetic data will have to be considered in order to answer both of these questions. The present study illustrates the difficulties that we are faced with when attempting to interpret the significance and applicability of data from in vitro studies in developmental toxicology for human risk assessment.

629

Fig. 3. Abnormalities of embryoand yolk sac after exposure to 1, 3, 4 and 5 = 10 lag VER/ml; 2 = I pg NIM/ml; 6 = 10 lag GAIdml; 7 = 30 lag NIF/ml; 8 = 10 lag Dlldml. For explanation of the types of abnormalities see legend to Table 4, The black arrows in 3 indicate the size difference in the right and left hand side of the embryo(limb abnormalitytype Q). The black arrows in 4 indicate a wavy neural wall (= neural tube abnormalitytype N). The black arrows in 5 indicate the head abnormality type O

Possible mode of action and primary target The six calcium channel blockers tested belong in three different chemical classes: nifedipine (NIF), nimodipine (NIM) and nitrendipine (NIT) are 1,4-dihydropyridines, - gallopamil HC1 (GAL) and verapamil HC1 (VER) are phenylalkylamines, and - diltiazem HC1 (DIL) is a 1,5-benzothiazepine. -

The aim of these comparative in vitro studies was to investigate whether (a) each drug may give rise to distinct, compound-specific effects on embryonic development, (b) the induction of specific effects is related to one of the basic chemical structures, or (c) there is a common ability to influence embryonic development perhaps due to the calcium channel antagonistic properties of all of these agents. The results presented here permit us to conclude that despite differences in chemical structure all the calcium channel blockers exhibit an identical (or at least very similar) effect on embryonic development in vitro, particularly

on the yolk sac. The differences observed with the various agents were merely quantitative. The swelling of the heart tube and the absence of heartbeat were only observed after exposure to VER and GAL, but since the histological examinations showed similarities with the other four drugs, this could also just be an especially pronounced effect. These findings suggest a common mechanism of action for the six calcium channel blockers for inducing disturbances on embryonic development. Since the observed adverse effects occurred independently of the chemical structure of the agents tested, we conclude that these alterations are probably due to the calcium channel blocking properties. This provides the opportunity to study the possible mode of action with more sophisticated biochemical techniques. As with many embryofeto-toxic agents a steep concentration-response curve (with less than a factor of 10 between the NOEL and the 100%EL) was observed. Although it is obvious from the results of our studies that the yolk sac function (circulation as well as endocytotic properties) was affected, it is difficult to decide from the information now available whether the effects on the

630 Table 4. Type and frequency of abnormalities and observations unusual for day 11.5: effects of calcium channel antagonists at the lowest tested concentration level that induced 100% abnormalities Yolk sac

Heart

21xA (100%)

3 x D+E+F 6 x D+E 6 x D+F 5xD lxE (100%)

12xA 10xB (100%)

17 x E 3xE+F 2xE+D (100%)

6x A 13 x B lxI (95%)

13xD 5 x D+F 1 x D+E+F lxE (100%)

VerapamilHC1 10 gg/ml n=21

GallopamilHC1 10 I.tg/ml n=22

DiltiazemHCI 30 gg/ml n=20

Nifedipine 30 gg/ml n=21

Nitrendipine 10 gg/ml n=22 Nimodipine 1 gg/ml n = 19

9xA 11 x B lxC (100%)

9xD 1 x D+H 1 x D+E+F 4xII (52%)

19 x A 3xB (100%)

16xD 1x E 1 x D+F+G 2xII (82%)

15 x A 3xB (95%)

4x D 1 x D+F 3xII (26%)

Heart beat

Shape

8x:(38%)

16xK lxlIl (76%)

2Ix:(95%)

11 x K 11 x I I I (50%)

Neural tube Head

3 xQ 8xV 2xN (10%)

18xO (86%)

13xN (59%)

2xO 18 x P 1 xO+P IxV (95%) 16 x O + P 2xP

9x K 8 x III

Ear

Eye

Tail

Somites

21 x V I 2xW

(14%) 8xQ 1 x Q+R 7xV

22 x VI

(41%) 7• 2xQ+R

20xVI

lxT

1 xlV

5 xV

1 xU

(45%)

(90%)

(45%)

(10%)

3x K 16xIII (14%)

15xN (71%)

3• 11 x P 6xlV (67%)

lxQ 1 xQ+R 12 x V I I (10%)

1 xK+M 21xN (95%)

13xlV (36%)

8xP 12xV (32%)

+

+

Forelimb

20xVI 1•

8• (38%)

7xQ

22xVl 5xS (23%)

2xK 2xL +

17 x l I I (23%)

6xO 10 x P

+

14 x l I l

15 x N (80%)

1 xlV (63%)

3 x V

18x VIII 1 xVII

Letters indicate the different abnormalities within one variable. R o m a n numerals indicate variables (mostly retardations). Arabic numerals indicate the frequency of an observation. In brackets the frequency of abnormalities within one variable is given in per cent Abnormalities: Yolk sac: A = overall form is flattened; swollen vessels; main vessels do not form branch-like pattern; no circulation to be observed (Fig. 3.2) / B = form of yolk sac round; swollen vessels like A (Fig. 3.1); C = yolk sac surface ruffled; no vessels or circulation. I = Thin main vessels developed, no circulation to be observed Heart: D = bulge in the region of prospective left ventricle (Fig. 3.7) / E = swollen cardiac tube (Fig. 3.6) / F = swollen pericardium (Fig. 2.3 b) / G = slight constriction in the region of the atrium c o m m u n e / H = swollen atrium commune. II = stricture between atrium c o m m u n e and prospective left ventricle still missing

Heart-beat: : - = no heart beat detectable Shape: K = kinking in lower rump region (Fig. 3.8) / L = kinking of the tail / M = kinking in region of rhombencephalon. III = incomplete flexion, telencephalon not yet approaching heart Neural tube: N = walls of neural tube show wavy appearance (Fig. 3.4). (Note: all neural tubes in this study were closed) Head: O = telencephalic vesicles have grown beyond the former prosencephalic region (Fig. 3.5) / P = other abnormalities, including very small heads, irregular shapes, etc. IV = fissura telodiencephalica still missing Forelimb: Q = left limb >>>>>right limb (Fig. 3.3) / R = swollen limb. V = limbs still missing Ear: VI = otic anlage still forms a pit / VII = closed vesicle with recessus dorsalis Eye: S = irregular borderlines of eye vesicle Tail: T = Bleb at tail tip / U = kinking in tail region Somites: W = irregular arrangement

T a b l e 5. Yolk sac morphology and function at LOEL concentrations

embryo should be considered direct effects or are the indirect result of disturbed yolk sac function. The breakdown of the yolk sac circulation must drastically affect embryonic development. However, according to the histological examinations the adverse effects observed within the tissues of the embryo proper cannot only be considered "unspecific". The very pronounced effect on the embryonic mesenchyme appears to be the result of a direct effect of the drugs.

Nifedipine Verapamil HCI Gallopamil HCI Diltiazem HCI

10 gg/ml 3 gg/ml 1 gg/ml 10 gg/ml

Normal

Deviating

Abnormal

91% 20% 9% 33%

9% 80% 82% 67%

0% 0% 9% 0%

In the specimens summarized under deviating, circulation was not clearly detectable, due to rather thin main vessels and a not very transparent tissue color but it could not be categorized as abnormal

631

Fig. 4. Rat e m b r y o s after 48 h o f culture: (a) = Control: Sagittal section s o m e w h a t lateral to the m e d i a n plane, h = heart anlage; m = m e s e n c e p h a l o n ; n = neural tube; p = p r o s e n c e p h a l o n ; r = r h o m b e n cephalon; * = otic vesicle; arrow = somites; arrowhead = first branchial arch. (b) = 10 I.tg G A L / m I : 9 = a b s e n c e of n o r m a l sized v e s s e l s in the yolk sac w a l l ; s m a l l a r r o w = the outer contour of the head is d e f o r m e d ; a r r o w h e a d = e m p t y s p a c e s in the m e s e n c h y m e ; a r r o w = p r o s e n c e p h a l o n ; h = heart anlage; e = e c t o p l a c e n t a l cone w i t h placenta anlage. (c) = 10 ~ g N I T / m l : * = Large lacunae b e t w e e n the y o l k sac e p i t h e l i u m and the m e s o t h e l i u m . A r r o w = p r o s e n c e p h a l o n ; h = heart anlage. (d) = 30 lag D I L / m l : a r r o w = open optic vesicle; 9 = e m p t y spaces in the m e s e n c h y m e of the l i m b bud; c = c o e l o m i c cavities; r = r h o m b e n e c e p h a l o n . (e) = 30 lag NIF/ml: * = large lacunae in the y o l k sac wall; h = heart anlage; n = neural tube; p = prosencephalon

Comparison with in vivo embryotoxicity studies in the rat

The histological inspection indicates that the embryos cannot be regarded as dead. However, it is extremely unlikely that embryos that did not establish a functioning circulation on gestational day 11.5 would survive under in vivo conditions. Therefore, the present observations may explain the increase in resorptions recorded in the in vivo studies with some of the calcium channel blocking drugs. The alterations observed in the embryo proper at all effective concentrations always correlated with drastic morphological and probably functional changes within the yolk sac. For this reason, our data do not provide good evidence for a possible direct teratogenic action of this class of agents at concentrations not affecting the circulation. However, additional teratogenic effects (possibly induced through effects on the mesenchyme) for single substances of this group are still possible. Published data on the embryotoxic potential of calcium channel blockers in vivo are only available for NIF and DIL. Besides an increased rate of embryomortality, NIF induced teratogenic effects in the cardiovascular system (Cabov and Palka 1984), defects of the finger skeleton, i.e. hyperphalangy (Yoshida et al. 1988), short limbs, oligodactyly, short tail, and hematomas (Fukunishi et al. 1980). Some of these defects could be induced by single oral

administrations of 20-150 mg NIF/kg body wt on 1 day between days 10 and 16 of pregnancy. For NIF additional data from the drug manufacturer in the form of a report to the regulatory agencies (information given to physicians), point to a teratogenic potential of this substance and it is mentioned that disturbances in postnatal development occur at daily doses of 30 mg NIF/kg body wt or higher. Reproductive toxicity has furthermore been studied with DIL in mice, rats and rabbits (Ariyuki 1975) using i.p. injections. While treatment of mice during days 7 - ! 2 of pregnancy induced embryomortality only (with concentrations of or exceeding 12.5 mg DIL/kg body wt) various teratogenic effects were observed subsequent to single doses of 25 or 50 mg DIL/kg body wt given on days 9 or 13. In rats (treated during days 9 - 14 of pregnancy) 80 mg DIL/kg body wt was found to be teratogenic, as was the case following a single dose on day 13 or 14; the resorption rate was clearly increased at this dose. Studies with rabbits revealed an increased resorption rate (70%) at 12.5 mg DIL/kg body wt (when applied on days 7-16) and an apparent increase in abnormalities in the few (approximately 20) surviving fetuses. Unpublished data from studies on developmental toxicity represent the only (but not readily available) information for the other calcium channel blockers. Summaries of

632

Fig. 5. Rat embryos with and without treatment after a 48 h culture period: (a) = Control: o = otic vesicle; * = first branchial arch; a = part of the atrium and v = ventricle of the heart anlage with the myoepithelium proliferating in a cone-shaped fashion into the cardiac jelly (arrow head). Myoepithelium and endothelium are linked. (b) = 10 lag GAL/mI: arrow = empty spaces in the mesenchyme of the head; p = prosencephalon. No thickening of the myoepithelium in the ventricular region (v) of the heart anlage. * = clear-cut dilatation of the space between myoepithelium and endothelium. Missing linkage between the two layers. (c) = I pg NIM/ml: * = large empty spaces in the region of the mesenchyme of the head. Missing thickening of the myoepithelium in the region of the ventricle anlage (v); a = atrium; o = optic process; r = rhombencephalon. (d) = 10 lag VER/ml: * = empty spaces in the region of the mesenchyme of the head. Distinct swellings in the region of the heart anlage (v). o = transition of the eye process into the prosencephalon; r = rhombencephalon

Table 6. Segment II studies in rat Substance

Smallest dose showing: (in mg/kg body wt per day) Teratogenic effects

Increased embryofetal mortality

NIF NIT NIM

30 a n.t.e, at 100 b n.t.e, at 100 a§

10a? no info 100a?

VER GAL

n.t.e, at 3 x 20 a n.t.e, at 2 x 7.5 d 2 x 4.75 d

3 x 20 a

DIL

25 e,f 80~.g

12,5e,f,g,h

a Drug information; b Hoffman 1984; c Schltiter 1986; d personal communication Knoll AG; e Ariyuki 1975 Studies with: f = mice; g -- rats; h = rabbits no info = no information available n.t.e, at 100 = no teratogenic effects recorded in segment II studies in rats with oral doses up to 100 mg/kg body wt per day

these results (segment II studies in the rat, oral administration) are accessible from the basic scientific information provided with the marketing of these drugs. It is difficult to judge whether all of these studies were performed according to todays' standards (Table 6).

Comparison of in vitro data with plasma levels in the rat Pharmacokinetic data have to be considered in order to compare the results obtained in our in vitro studies with the situation probably existing in vivo. Plasma concentrations achieved in vivo under defined exposure/treatment conditions are the most reasonable for comparison with in vitro culture medium concentrations. A comparison of our in vitro data with the results of segment II studies is difficult, since no pharmacokinetic studies have been concurrently performed with such experiments, and data on plasma levels have to be taken from other studies conducted under different experimental conditions. Furthermore, comparing only plasma levels of the

633

Fig. 6. Rat embryos alter 48 h of culture: (a) = 10 lag VER/mI: * = empty spaces in the mesenchyme of the head. Dilatation of the space between yolk sac epithelium and mesothelium of the yolk sac wall (y). p = prosencephalon; m = mesencephalon. (b) = 10 lag NIT/mI:prosencephalon (p) with eye process (arrow); numerous necroses in the neuroepithelium; * = large empty spaces in the mesenchyme. (c) = 10 lag GAL/mI:huge cavities (*) next to the neural tube (n), which is slightly deformed. Some cell debris in the cavity (arrow head). (d) = 10/ag VER/mI:like Fig. 6c. o = deformed otic vesicle

parent drug will neglect other important pharmacokinetic aspects such as protein-binding, or the presence of different isomers with different biological activities, or the role which metabolites may play for the induction of embryotoxic effects in vivo, and concentrations in maternal plasma may not reflect the actual concentrations reached in the embryo. Therefore, only rough estimates can be made. Some of the pharmacokinetic data obtained in rats after oral administration are compiled in Table 7. The following tentative conclusions may be drawn: one might assume that the in vivo teratogenic dose (see above) of 30 mg NIF/kg body wt could produce plasma peak levels that at least approach the in vitro LOEL of 10000 ng/ml. The in vivo application of 3 • 20 mg VER/kg body wt per day caused increased prenatal mortality in rats and led to plasma concentrations

about two to three times lower than the in vitro LOEL of about 3000 ng/ml. For GAL, even the repeated in vivo application of 2 • 7.5 mg GAL/kg body wt per day apparently does not lead to plasma concentrations equivalent to the in vitro LOEL of 1000 ng/ml. After in vivo application of 100 mg NIT/kg body wt plasma levels might approach concentrations effective in vitro, but there is no information about in vivo effects from studies on reproductive toxicity. There are no corresponding pharmacokinetic data in the rat available for NIM or DIL. These very rough estimates indicate that in the segment II studies in rats plasma concentrations may have been reached that are able to interfere with embryonic development in vitro. This may explain the embryotoxic effects seen, e.g. with NIF.

634

Fig. 7. Rat embryos after 48 h of culture: (a) = 30 jag DIL/ml: oblique section of an embryo with neural tube (n). The limb bud anlagen with thickened epithelium (arrow) and large empty spaces (*) in the mesenchyme. (b) = 3 ~g VER/ml: Vena cava (*) with defined endothelial proliferations (arrow)

Comparison of in vitro data with plasma levels in human therapy For human therapy the m a x i m u m recommended daily doses (MRDD) in mg are: NIF (6 x 20 p . o . / 2 x 40 RR p.o./ 30/24 h infusion); NIT ( 2 x 2 0 p.o.); N I M ( 4 x 6 0 p . o . / 4 8 / 2 4 h infusion); VER ( 4 x 120/ 3 x 160/ 2 x 2 4 0 RR p.o./ 100/24 h infusion); G A L ( 4 x 5 0 p.o.); DIL ( 4 x 9 0 RR/ 3 x 1 2 0 p.o./ 300/24 h infusion). Levels reached under steady-state conditions during oral therapy or after constant infusion may be expected to mimic best an in vitro situation with the drug present at the same concentration over the entire culture period. Unfortunately, such data for man are only available for some of the investigated agents. For this reason rough estimates can only be made using other available data, less convenient for risk assessment (cf Table 7). These data suggest that for NIM (the drug leading to abnormal development in vitro at the lowest concentration) plasma levels achieved in man are very close to a N O E L

for the in vitro effect. As far as suitable data are available for the other compounds, the therapeutic plasma concentrations may be assumed to be at least one order of magnitude from the concentrations which interfere with embryonic development in vitro, when one regards applications producing rather constant plasma levels. Single individual peak levels may lead to a difference of less than one order of magnitude.

Unsolved problems and drawbacks when attempting extrapolations from in vitro data There are a number of problems which could not be adequately dealt with in the present study. It is, for example, generally accepted that drug effects are better related to the free (unbound) rather than to the total plasma concentration. In order to evaluate data obtained with different drugs or to predict effects from in vitro to in vivo the free concentrations of the drugs or the tissue levels should be corn-

635

Fig. 8. Yolk sac epithelium after a 48 h culture period: (a) = Control: Numerous electron-dense inclusions (arrow) in the apical cellular region (proteinresorbing apparatus). (b) = 10 ~tg NIT/ml. Missing inclusions, some lipid droplets (*). m = mesenchymal cell

pared. However, in the case of the calcium channel blockers there are several factors which complicate such comparisons or extrapolations. These include:

Protein binding. The six calcium channel antagonists studied here are all extensively bound to human plasma proteins: NIF 99% (Otto and Lesko 1986), NIT 98% (Eichelbaum et al. 1986), NIM 99% (drug information), VER about 90% (Keefe et al. 1981; McGowan et al. 1983), GAL 93% (Rutledge and Pieper 1987), DIL about 80% (Morselli et al. 1979; Bloedow et al. 1980; Kwong et al. 1985).

Concentration dependence. The degree of protein binding has been shown to be concentration dependent: e. g. protein binding of 98% is documented for NIF in human serum in the presence of 200 ng/ml, but of only 92% with 20000 ng/ml (Rosenkranz et al. 1974), and for DIL, binding to rat plasma proteins, a decrease in protein binding has been found between the concentrations of 300 and 30000 ng/ml from 84% to 66%, respectively (Nakamura et al. 1987).

and rats in vivo with that in cow serum in vitro. Species differences in protein binding have been documented for NIF at least between man and dog (Rosenkranz et al. 1974), but not between rat and man (cited in Downing and Hollingsworth 1988), and for DIL, i.e. with 300 ng/ml, between rat plasma (84%) and human plasma (63%) and human serum albumin (40%) and bovine serum albumin (66%) (Nakamura et al. 1987). Also for VER species differences seem to occur, since binding to rat plasma proteins (Todd and Abernethy 1986, 1987) is about 98%, but to human plasma protein only about 90%. No data are available for bovine serum.

pH dependence. Differences in protein binding due to various pH values may also be of significance for comparative assessments since the pH of the culture medium shifts during the culture period from about 7.8 to 7.0. A reduction in protein binding is described for a decrease in pH in the case of NIF (Rosenkranz et al. 1974) and GAL (90% with pH = 7.0/95% with pH = 8.0; Rutledge and Szlaczky 1988).

Species differences. Such differences have to be considered

Isomers. NIT, NIM, VER and GAL are administered as a

since we want to compare free concentrations in humans

racemic mixture. Since normal measuring methods for

636

Fig. 9. Apical part of the yolk sac epithelium after 48 h of culture. Tannic acid fixation: (a) = Control: (arrows) Numerous tannic acid-positive vesicles and tubuli (coated vesicles). (b) = 10 gg NIT/ml: numerous coated pits, but only a few invaginations (arrows)

Table 7. Synopsis of the concentrations with the six calcium channel antagonists. Culture medium concentrations in vitro versus plasma concentrations in vivo in rat segment II studies and under therapeutic conditions in man. Concentrations (ng/ml) NIF

NIT

NIM

VER

GAL

DIL

1 000

1 000 3000 10 0 0 0

1 000 10 000

1 000 10000 30 000

17-72 l

119-1372

In vitro NOEL LOEL 100%EL

1 000 10 000 30 0 0 0

10 0 0 0

In vivo at m a x i m u m recommended h u m a n dose at lower than m a x i m u m recommended

9

100

?(63 - 800) ( 167 - 4094)

9 (50-90)(3613~

(35 - 2805 )

687 ( 4 0 - 125)

40 - 606

human dose in segment H study in rats

>2500 a

>4006

t Drug information, range of mean steady-state plasma concentrations (max. recommended dose) after p.o. or infusion (for similar data see R~imsch et al. 1984) 2 Biihler et al. 1984, after 2 x 2 4 0 m g (sustained release formula) 3 Frishmann et al. 1982, after 3 x 160 mg. VER plasma levels after infusion are lower than the levels obtained after p. o. (Reiter et al. 1982) 4 Hung et al. 1988, after 3 x 120 mg. During infusion about 200 ng/ml are reached (drug information) 5 Cmax after single dose application of 10 mg p.o. (Kleinbloesem et al. 1987; Donnelly et al. 1988; Menke et al. 1988)

9

1000~ - 1500

391N - 390

6 Steady-state concentration after 20 mg (Kirch et al. 1984; R~imsch and Sommer 1984) 7 Stieren et al. 1983, after 3 x 50 m g The plasma levels of a Ct=3o min = 2500; b C m a x = 400; c C m a x = 10001 5 0 0 ; d C m a x = 250 were reached after a.b a single dose of 5 mg/kg body wt of NIF and NIT; c repeated doses of 3 x 20 mg VER/kg body wt per day, and d repeated doses of 2 x 7.5 m g GAL/kg body wt per day a Patzschke et al. 1975; b Krause et al. 1988; c,d personal communication, Knoll AG

637

concentration determination do not discriminate between the different optical isomers, a comparison of in vivo and in vitro levels is difficult if the isomers possess different biological properties and if the ratio of the enantiomers changes in vivo. It has been shown that the (-)-isomers of NIM and NIT (Bellemann et al. 1982) and VER (Eichelbaum et al. 1984) are biologically more effective than the (+)-isomer. Furthermore, in the case of VER it is known that the ratio of (+)/(-)-isomer changes in vivo because the (+)-isomer is metabolically more stable (Eichelbaum et al. 1984).

Concluding remarks The concentrations of the calcium channel antagonists exhibiting an adverse effect in the whole-embryo culture appear high (IaM range) when compared with data published (e. g. Bolger et al. 1983; Godfraind and Wibo 1985; Qar et al. 1988) for receptor binding (nM range). However, up till now, no clue on the possible concentrations reached at the targets under our in vitro conditions exists. This study clearly indicates the difficulties we are faced with at the present time when attempting to extrapolate results from in vitro studies to the situation possibly existing in man (or even in experimental animals in vivo). In order to integrate data obtained with in vitro techniques into risk assessment studies, new strategies have to be established. This especially concerns the planning of specially designed pharmacokinetic studies to provide relevant information for the interpretation of in vitro data. Acknowledgements. The fellowship granted to Mithilesh Kumar Srivastava by the German Academic Exchange Programm (DAAD) making his stay in Berlin possible is thankfully acknowledged. The calcium channel blockers were generously provided by Bayer AG, Knoll AG, and GOdecke AG. We appreciate the many fruitful discussions with the Knoll AG and the advice given by the Bayer AG with respect to the handling of the light-sensitive substances. These studies were supported by grants from the Bundesministerium fiir Forschung und Technologie and the Deutsche Forschungsgemeinschaft. Jane Klein-Friedrich is gratefully acknowledged for her help in preparing the manuscript.

References Ariyuki F (1975) Effects of diltizem hydrochloride on embryonic development - species differences in the susceptibility and stage specificity in mice, rats and rabbits. Folio Anat Jap 52: 103- 117 Bellemann P, Ferry D, L/ibbecke F, Glossmann H (1982) (3H)-nimodipine and (3H)-nitrendipine as to61s to directly identify the sites of action of 1,4-dihydropyridine calcium antagonists in guinea-pig tissues. Arzneimittelforschung 32:361 -363 Bloedow DC, Piepho RW, Nies AS, Gal J (1980) Serum binding of diltiazem in humans. Pharmacologist 22:192 Bolger GT, Gengo P, Klockowski R, Luchowski E, Siegel H, Janis RA, Triggle AM, Triggle DJ (1983) Characterization of binding of the Ca++ channel antagonist, [3H]-nitrendipine, to guinea-pig ileal smooth muscle. J Pharmacol Exp Ther 225:291 -309 Bfihler V, Einig H, Stieren B, Hollman M (1984) Pharmacokinetics of verapamil sustained release formulation with multiple applications. Naunyn-Schmiedeberg's Arch Pharmacol [Suppl] 325:R86 Cabov AN, Palka E (1984) Some effects of Cordipin R (nifedipine) administered during pregnancy in the rat. Teratology 29:21 a

Constantine G, Beevers DG, Reynolds AL, Luesley DM (1987) Nifedipine as a second line antihypertensive drug in pregnancy. Br J Obstet Gynaecol 94:1136- 1142 Donnelly R, Elliott HL, Meredith PA, Kelmann AW, Reid JL (1988) Nifedipine: individual responses and concentration-effect relationships. Hypertension 12:443-450 Downing SJ, Hollingsworth M (1988) Nifedipine kinetics in the rat and relationship between its serum concentrations and uterine and cardiovascular effects. Br J Pharmacol 9 5 : 2 3 - 3 2 Eichelbaum M, Mikus G, Vogelgesang B (1984) Pharmacokinetics of (+)-, (-)- and ( + )-verapamil after intravenous administration. Br J Clin Pharmacol 17:453--458 Eichelbaum M, Fischer C, Heuer B, Mikus G (1986) Pharmakokinetik und Metabolismus von Nitrendipin. In: Distler A (ed) Calcium-Antagonisten in der Hochdrucktherapie. Europ~iisches BayotensinSymposium 1985, Schattauer, Stuttgart, New York, pp 35-42 Frishman W, Kirsten E, Klein M, Pine M, Johnson SM, Hillis LD, Packer M, Kates R (1982) Clinical relevance of verapamil plasma levels in stable angina pectoris. Am J Cardiol 50: 1180- 1184 Fukunishi K, Yokoi Y, Yoshida H, Nose T (1980) Effects of nifedipine on rat fetuses. Med Consult New Remed 17:2245-2256 (Japanese) Godfraind T, Wibo M (1985) Subcellular localization of [3H]-nitrendipine binding sites in guinea-pig ileal smooth muscle. BrJ Pharmacol 85:335-340 Hoffman K (1984) Toxicological studies with nitrendipine. In: Scriabine A e t al. (eds) Nitrendipine. Urban & Schwarzenberg, Baltimore, Munich, pp 2 5 - 32 Hung J, Hackett PL, Stephen PF, Gordon SPF, llett KF (1988) Pharmacokinetics of diltiazem in patients with unstable angina pectoris. Clin Pharmacol Ther 43:466-470 Keefe DL, Yee YG, Kates RE (1981) Verapamil protein binding in patients and in normal subjects. Clin Pharmacol Ther 2 9 : 2 1 - 2 6 Kirch W, Hutt H J, Heidemann H, Rfimsch K, Janisch HD, Ohnhaus EE (1984) Drug interactions with nitrendipine. J Cardiovasc Pharmacol 6: $982- $985 Kleinbloesem CH, van Brummelen P, Breimer DD (1987) Nifedipine. Relationship between pharmacokinetics and pharmacodynamics. Clin Pharmacokinet 12:12-29 Klug S, Lewandowski C, Neubert D (1985) Modification and standardization of the culture of early postimplantation embryos for toxicological studies. Arch Toxicol 58:84-88 Krause HP, Ahr HJ, Beermann D, Siefert HM, Suwelack D, Weber H (1988) The pharmacokinetics of nitrendipine. I. Absorption, plasma concentrations, and excretion after single administration of [~4C]nitrendipine to rats and dogs. Arzneimittelforschung 38: 1593- 1599 Kwong TC, Sparks JD, Sparks CE (1985) Lipoprotein and protein binding of the calcium channel blocker diltiazem. Proc Soc Exp Biol Med 178:313-316 Laragh JH (1987) Round table discussions. J Cardiovasc Pharmacol [Suppl] 10:203-212 Lilja H, Karlsson K, Lindecrantz K, Sabel KG (1984) Treatment of intrauterine supraventricular tachycardia with digoxin and verapamil. J Perinat Med 12:151 - 154 McGowan FX, Reiter MJ, Pritchett ELC, Shand DG (1983) Verapamil plasma binding: relationship to alpha-l-acid glycoprotein and drug efficacy. Clin Pharmacol Ther 33:485-490 Menke G, Kausch M, Rietbrock N (1988) Eine Model|untersuchung zur Bioaquivalenz von gel6stem Nifedipin aus Weichgelantinekapseln. Arzneimittelforschung 38: 300- 304 Morselli PL, Rovei V, Mitchard M, Durand A, Gomeni R, Larribaud J (1979) Pharmacokinetics and metabolism of diltiazem in man (observations on healthy volunteers and angina pectoris patients). In: Bing RJ (ed) New drug therapy with a calcium antagonist. Diltiazem Hakone Symposium 1978, Excerpta Medica, Amsterdam, Princeton, pp 152- 167 Nakamura S, Suzuki T, Sugawara Y, Usuki S, Ito Y, Kume T, Yoshikawa M, Endo H, Ohashi M, Harigaya S (1987) Metabolic fate of diltiazem. Distribution, excretion and protein binding in rat and dog. Arzneimittelforschung 37: 1244- 1252

638 Otto J, Lesko LJ (1986) Protein binding of nifedipine. J Pharmacol 38: 399-400 Orlandi C, Marlettini MG, Cassani A, Trabatti M, Agostini D, Salomone T, Testoni N (1986) Treatment of hypertension during pregnancy with the calcium antagonist verapamil. Curr Ther Res 39: 884- 893 Patzschke K, Schlossmann K, Horster FA (1975) Nifedipin (AdalatR) Pharmakokinetik und Biotransformation. Ther Ber 47:197-203 Pipkin FB, Lawrence MR (1987) Preliminary studies on the cardiovascular effects of nitrendipine in human pregnancy. J Cardiovasc Pharmacol 9:$286-$289 Qar J, Barhanin J, Romey G, Henning R, Lerch U, Oekonomopulos R, Urbvach H, Lazdunski M (1988) A novel high affinity class of Ca 2+ channel blockers. Mol Pharmaco133: 363-369 Riirnsch KD, Sommer J (1984) Pharmacokinetics and metabolism of nitrendipine. In: Scriabine A, Vanov S, Deck K (eds) Nitrendipine. Urban & Schwarzenberg, Baltimore, Munich, pp 409-421 R~nsch KD, Graefe KH, Sommer J (1984) Pharmacokinetics and metabolism of nimodipine. In: Betz E, Deck K, Hoffmeister F (eds) Nimodipine. Pharmacological and clinical properties. Schattauer Vetlag, Stuttgart, New York, pp 147-159 Read MD, Wellby DE (1986) The use of a calcium antagonist (nifedipine) to suppress preterm labour. Br J Obstet Gynaecol 93: 933-937 Reiter M J, Shand DG, Aanonsen LM, Wagoner R, McCarthy E, Pritchett ELC (1982) Pharmacokinetics of verapamil: experience with a sustained intravenous infusion regimen. Am J Cardiol 50:716-721 Rosenkranz H, Schlossmann K, Scholtan W (1974) Die Bindung von 4-(26-Nitrophenyl)-2,6-dimethyl-l,4-dihydropyridin-3.5-dicarbons~iuredimethylester (Nifedipin) sowie von anderen koronarwirksa-

men Stoffen an die Eiweil3k6rper des Serums. Arzneimittelforschung 24:455-466 Rutledge DR, Pieper JA (1987) Gallopamil binding to human serum proteins. Eur J Clin Pharmacol 33:375 -380 Rutledge DR, Szlaezky M (1988) The effect ofpH on gallopamil protein binding. Pharm Res 5:619-620 Schliiter G (1986) Toxicological investigations with nimodipine. Arzneimittelforschung 36: 1733-1735 Stieren B, Biihler V, Hege HG, Hollmann M, Neuss H, Schlepper M, Weymann J (1983) Pharmakokinetic und Metabolismus yon Gallopamil. In: Kaltenbach M, Hopf R (eds) Gallopamil. Springer-Verlag, Berlin, Heidelberg, New York, Tokyo, pp 90-96 Strigl R, Gastroph G, Hege HG, D6ring P, Mehring W (1980) Nachweis yon Verapamil im mfitterlichen und fetalen Blut des Menschen. Geburtsh u Frauenheilk 40:496-499 Todd EL, Abernethy DR (1986) Pharmacokinetics and dynamics of ( +)-verapamil in lean and obese zucker rats. J Pharmacol Exp Ther 238:642-647 Todd EL, Abernethy DR (1987) Physiological phannacokinetics and pharmacodynamics of (~e)-verapamil in female rats. Biopharm Drug Disp 8:285-297 WolffF, Breuker KH, Schlensker KH, Bolte A (1980) Prenatal diagnosis and therapy of fetal heart rate anomalies: with a contribution on the placental transfer of verapamil. J Perinat Med 8:203 -208 Trnccone N, Mariona F (1985) Intrauterine conversion of fetal supraventricular tachycardia with combination of digoxin and verapamil. Pediatr Pharmacol 5: 149-153 Yoshida T, Kanamori S, Hasegawa Y (1988) Hyperphalangeal bones induced in rat pups by maternal treatment with nifedipine. Toxicol Lett 40: 127-132