Differentiation (1992) 51 : 105-1 12 Ontogeny, Neoplasia and Differentiation Therapy

0 Springer-Verlag 1992

Early heart development in the chick embryo: effects of isotretinoin on cell proliferation, a-actin synthesis, and development of contractions Darrell J. Wiens, Tamara K. Mann, Dan E. Fedderson, W. Kimryn Rathmell, and Barbara H. Franck Department of Biology, University of Northern Iowa, Cedar Falls, IA 50614, USA Accepted in revised form May 27, 1992

Abstract. Isotretinoin is a potent retinoic acid used in the treatment of skin disorders. Though very effective, it is teratogenic if administered during pregnancy, and its teratogenic effect may be related to the normal activity of retinoids as signalling molecules in the embryo. Although its exact mechanism of action is unknown, it has been suggested that it causes its characteristic pattern of defects that includes heart defects, by inhibiting the migration of neural crest cells. However, other effects on cells are known. We studied early cardiac cell proliferation using incorporation of bromodeoxyuridine (BrdU) and detection with a monoclonal anti-BrdU. Proliferation in heart tissue of whole embryo cultures was inhibited in medium with l o p 6M isotretinoin to 62% of the control level in myocardium. We studied its effects in culture on precardiac explant development in the absence of the neural crests. Culture of precardiac mesodermal-endodermal explants revealed that development of heart vesicles from the mesoderm was little affected, but the development of heartbeat was inhibited depending on dose in the to M range. The effect on development of contractions was augmented in the presence of serum; it could be duplicated by all-transretinoic acid, and it was reversible. Synthesis of the aactin isotype, analyzed by isoelectric focusing, was found to be inhibited or delayed. The results suggest multiple effects of retinoids on growth, morphogenesis, and differentiation of early cardiac tissue, and are discussed in relation to the potential role of retinoids in early embryogenesis.

Introduction The heart appears early in avian development when paired regions of lateral plate mesoderm migrate anteriorly and medially to the position of heart formation near the anterior intestinal portal of the embryo [38]. Correspondence t o: D.J. Wiens

Here a subgroup of mesodermal cells dissociates from the layer to form paired endocardia1 tubes that shortly fuse. Pre-myocardial mesoderm then surrounds these to create a single, linear heart tube at 35 h of development (Hamburger-Hamilton stage 10 [ 161). Although myofibrillogenesis is underway, cell proliferation continues, providing growth until the time of hatching [ 12, 551. Myogenesis in the early myocardium is characterized by elevation of a-actin synthesis during stage 8 [51], which is synthesized initially from smooth muscle a-actin transcripts, and later from sarcomeric a-actin transcripts [40, 541. Filament bundles first appear at stage 9, followed by accumulation of organized myofibrils [27, 531 and the onset of contractions at stage 10. Isotretinoin is a retinoic acid (RA) analog that has been used effectively to treat severe cystic acne and other chronic dermatoses [351. RAs and megadoses of vitamin A have long been known to be teratogenic in laboratory animals (for reviews, see [39, 421). Therefore isotretinoin was labeled a category X medication, contraindicated during pregnancy. Despite warnings, inadvertant use during pregnancy has resulted in many reports of induced defects [23]. These are characteristically craniofacial, thymic, and cardiac [5]. Because of this pattern, a common mechanism to explain the cause was proposed: that neural crest cell migration is inhibited in the presence of isotretinoin, preventing neural crest cell contribution to tissues forming in these regions [23, 361. There is evidence that neural crest cells contribute to the aorticopulmonary septum of the heart [4, 19, 201, and that mesenchymal cell migration can be inhibited by retinoids in vitro [32, 37, 45, 481. However, because other teratogenic agents and conditions result in similar malformations, accentuation of cell death has also been proposed as a common teratogenic mechanism [3, 461. Mechanisms of teratogenesis other than these may be involved. Retinoids can both stimulate and inhibit cell growth [15, 29, 39, 411, and it is well established that RAs profoundly influence differentiation (for reviews, see [39, 531) and the metabolism of connective tissue extracellular matrix (ECM) [1,14, 17,341. In addi-

106

tion, the teratogenic effects of retinoids may be related to the proposed normal activity of retinoic acid as a signalling molecule during development. There is evidence for such a role in limb development [47, 49, 501, limb regeneration [7, 181, and pattern formation of the central nervous system and embryonic axis 1113, 26, 29, 431. Cytoplasmic retinoid binding proteins that can limit the free concentration of retinoic acid (CRABP) and retinol (CRBP) are arrayed in gradients in chick limb bud and central nervous system cells [24, 25, 311, and there are, as well, temporal and regional differences in the expression pattern of retinoic acid nuclear receptors [30, 441. Because of the demonstrated multiple cellular effects of retinoids, and because of their involvement in many early developmental events, it is important to investigate whether isotretinoin perturbs early aspects of heart development, particularly if these effects can be studied in the absence of neural crest cells. Therefore we have utilized a precardiac mesodermal-endodermal explant culture system [52] to examine the effect of isotretinoin on cell proliferation, development of heartbeat, and synthesis of muscle-specific actin during the earliest period of cardiac morphogenesis.

Methods Precardiac explant culture. Fertile, White-Leghorn chicken eggs were purchased from Hy-Vac Labs (Gowrie, Iowa, USA). Eggs were incubated at 38" C for 30 h to obtain embryos of stages 6-8. Embryos were removed using filter paper rings and placed ventral side up in Earle's balanced salt solution (EBSS) for microdissection. Using fine glass needles, the precardiac mesoderm regions together with adjacent endoderm were removed as rectangles from the right and left sides. The explants were transferred to Falcon Primaria culture dishes having a medium-soaked substratum of egg albumin and agarose, according to the methods of DeHaan [8] and Wiens et al. [52]. The medium consisted of medium 199 (Sigma) with 1YOantibiotic/antimycotic (Sigma), supplemented (in designated experiments) with 2% fetal bovine serum and 4% horse serum (Sigma). Isotretinoin (Sigma) or retinoic acid (Sigma) was added from a 6 mg/ml stock solution in dimethylsulfoxide (DMSO) to achieve concentrations of 0.25-20.0 pg/nil (8.3 x lo-' to 6.7 x M ) . Control dishes received an equal volume of DMSO only. The explants were cultured as pairs with one serving as the control, the other treated, and were oriented in the dishes with the anterior end identified. In some experiments whole embryos, adhered between two filter paper rings, were cultured using the same conditions. After placing explants or embryos on the substratum, medium was removed so as to flatten but not recess the tissues on the medium-soaked substratum. Dishes were incubated 24 h at 38" C in an atmosphere of 5% C 0 2 in air and 100% humidity. The culture dishes were protected from light as much as possible. Following the culture period, explants were examined for the presence of heart vesicles and contractions. The position of the vesicle within the explant was designated as anterior, middle, or posterior. Spontaneous contractions were noted, and if not present, the explants were mechanically stimulated with a fine glass probe. If they still failed to contract, a negative result was recorded. In order to assess the toxic effect of isotretinoin, control and 3 pg/ml isotretinoin treated explants that had been incubated in culture overnight were tested for cell death using the lysosomotropic vital stain Nile blue sulfate. Living explants were incubated for 30 min at 38" C in EBSS containing Nilc bluc sulfate (Harleco,

Philadelphia) diluted 1 : 10,000. They were then rinsed i n EBSS and examined for the degree of cellular uptake of the dye. Assay of cell proliferation. Stage 6-8 chick embryos were removed from eggs incubated 30 h using filter paper rings and were cultured upon a medium-soaked agarose-albumin substratum overnight in the presence or absence of 1 pg/ml isotretinoin (3.3 x Mj. Following incubation the nucleotide base analog bromodeoxyuridine (BrdU) was added to the culture medium for 90 min during which it was incorporated into DNA synthesizing cell nuclei. Embryos were then rinsed, fixed in 3.7% formaldehyde, dehydrated in an ethanol series, embedded in paraffin, and sectioned across or sagittally at 6 pm. Sections mounted on polylysine coated slides were deparaffinized, treated for 30 min with 1 N HCI, rinsed in phosphate buffered saline (PBS), and incubated with a monoclonal antibody to BrdU (Amersham). Antibody binding was detected using a peroxidase conjugated secondary antibody and hydrogen peroxide together with color intensifier (Amersham). The sections were counterstained with eosin and mounted with GVA Mount (Zymed Laboratories, San Francisco). The intensely stained nuclei of the endocardium and myocardium were enumerated as a proportion of total nuclei in the sections, and the differences in average proportions were tested for significance with the student's t-test. Actin biosynthesis assessment. Isotretinoin-treated and control explants were cultured as described, but in Corning tissue culture treated 96-well plates (Corning Glass Works, Corning, N.Y.) without the substratum. Under these conditions development is the same [52]. Approximately 100 pCi [35S]-methionine(100-200 Cil mmole; Amersham) was applied in a volume of 10 pl to the explant before incubation overnight to label newly-synthesized proteins. Matched control and treated explants were then rinsed 3 x in EBSS and prepared for isoelectric focusing (IEF) in polyacrylamide slab gels. The samples contained 20% f2% trichloroacetic acid (TCA) precipitable counts as determined by collection of 5% TCA precipitates on Millipore filters (0.2 pm), and there was no significant difference between control and isotretiuoin (3 pg/ml) treated samples. The methods used for IEF were those of O'Farrell [ 3 3 ] , as modified for slab-gels and the resolution of actin isotypes in developing cardiac tissue by Wiens and Spooner [51]. The dried gels were exposed to Kodak X-Omat RP film with intensifying screens at -80" C for 1&20 days. The autoradiograms were scanned in an ISCO 1312 Gel Scanner with UA5 recorder (Lincoln, Neb., USA).

Results

The effect of isotretinoin on cell proliferation Whole embryos cultured overnight as controls or treated with 1 pg/ml isotretinoin underwent development from preheart stages 6, 7, or 8 to embryos of stages 11 or 12 that possessed primitive heart tubes. The heart tissues were assessed for cell proliferation by determining the proportion of total nuclei in sagittal and cross sections that had incorporated the nucleotide analog BrdU. These nuclei were detected with monoclonal anti-BrdU and a peroxidase-conjugated secondary antibody. They stained an intense blue-black color and were easily identified (Fig. 1). Table 1 displays the mean proportions of stained nuclei. The 6% inhibition resulting from isotretinoin treatment in the myocardium was significant. Even though the difference we observed in endocardium was greater (8% inhibition), it was not significant. The much lower number of endocardia1 cells visible in each section, resulted in more variability (Table 1). The

107

Table 1. Effect of isotretinoin on embryonic heart cell proliferation Heart Tissue

Myocardium Endocardium

Mean Percentage of Stained Nuclei per Section Control

1 pg/ml Isotretinoin

16% f.5 Yo * n = 19 sections 20% f.89” n = 15 sections

10%f6%b n = 11 sections 12% & 14% n = 11 sections

a The number of immunostained nuclei/the number of total nuclei visible in the tissue x 100%. Embryos were cultured overnight in control or 1 pg/ml isotretinoin containing medium and then incubated 3 h further in the same medium containing bromodeoxyuridine (BrdU). They were then fixed, embedded, and immunostained with a monoclonal antibody to BrdU. Nuclei were counted from 56 sections representing 4 experimental trials Significantly different control, P < 0.05

greater average proportion of stained nuclei found in the endocardium compared to myocardium was not significant. Stained cells were not observed to be more frequent in any region of the developing hearts when viewed in sagittal or cross section. The effect of retinoids on the development of heart vesicle form and contractions, and vesicle orientation

To study the effect of isotretinoin on development in the absence of contributions from neural crest cells, the paired precardiac mesodermal regions with adjacent endoderm were removed from stage 6-8 embryos and cultured as explants for 24 h. Previous studies have demonstrated the fidelity of this culture system as a model of in vivo development [9, 521. 1. Development of vesicleform and contractions. Isotretinoin treatment yielded differing results when the explant cultures were examined for the development of heart vesicles and the ability to contract (summarized in Fig. 3 and Table2). The control member of each pair of explants invariably developed a vesicle that was either spontaneously contracting or would contract if mechanically stimulated with a glass probe. But (in terms of the percentage of explants that developed heart contractions) the explants cultured in the presence of isotretinoin exhibited a dose-dependent inhibition of the ability to contract, even though a vesicle still formed and general morphology looked comparable to that of controls (Fig. 2). At the higher doses some loss of vesicle forming ability was found but even at 6 pg/ml, 37/39 explants had formed vesicle structure (Fig. 3). Complete inhibition of development of contractions was observed at 20 pg/ml. There was no strong correlation of stage (HH stages 6, 7, or 8) at the time of explantation with inhibition of contraction, although the earlier stages were somewhat more sensitive. Incubation of treated explants that did not beat for an additional 24 h period (total of 48 h) did not result in initiation of contractions. Incu-

Fig. 1. Light micrographs of control (A) and isotretinoin-treated (B) chick embryos showing cross-sections through the heart, and stained with anti-bromodeoxyuridine (BrdU) to detect cell proliferation. The endocardium ( e ) and myocardium (m)can be seen separated by the cardiac jelly. Nuclei synthesizing DNA are stained an intense black among other nuclei. Although the thickness of the myocardial layer varied in the sections from 1 to 3 or 4 cells thick, no general difference was found between control and isotretinoin-treated embryos. In this control section (A), the proportion of stained nuclei is approximately 22%. In this section of embryo treated with 1 pg/ml isotretinoin (B), the proportion of stained nuclei is approximately 15%. Bar, 15 pm

bation of explants treated 16-24 h with 3 pg/ml isotretinoin, which did not contract, for an additional 24 h in control medium did result in development of contractions in most cases (Table 2). Similarly, control explants that were contracting after 16 h, incubated for an additional 8 h in 3 pg/ml isotretinoin were still able to contract. Thus the inhibition of contractions by isotretinoin was reversible, but once contractions had developed, they could not be inhibited. The staining of living explants with the vital dye Nile blue sulfate was carried out to check for a cytotoxic effect. In both control explants and in explants treated with 3 pg/ml isotretinoin, only a few stained cells were found at the edges. There was no apparent difference between control and treated explants in the number of cells which had taken up the dye.

108

Fig. 2. Light micrograph of cardiac explants cultured overnight showing vesicle morphology. Control explants (A) developed beating vesicles ( 0 ) that were most often at the anterior end but sometimes in the middle. Explants cultured in medium with isotretinoin or retinoic acid usually formed vesicles that did not contract. Retin-

Table 2. The effect of isotretinoin and retinoic acid on the development of heartbeat in precardiac explants in culture"

oic acid present at 1 pg,/ml (B and C) had a greater effect on vesicle morphogenesis than isotretinoin at equivalent concentrations, resulting in a majority of vesicles located in the middle (B, u ) , or in no vesicle formation at all (C). Bur, 75 pm

Treatment

Experiments with serum Controls' Is0 tretinoin Isotretinoin Retinoic Acid Isotretinoin Isotretinoin Isotretinoin Experiments without serum Controls Isotretinoin Isotretinoin Retinoic Acid Is0 tretinoin Tsotretinoin Isotretinoin Reversal Experiments Controls reincubated in isotretinoin Treated, reincubated as controls

Concentration (Pgiml)

-

Beating explants/ total explants

Inhibition (percentage

0.25 1.oo 1.oo 3.00 6.00 20.0

101/101 10120 6/19 411 3 5118 1/25 OjlO

0 50*11 68111 69k13 72111 72+ 9 100

0.25 1.oo 1.oo 3 .OO 6.00 20.0

92/92 12/11 16/20 5118 8/23 211 4 016

0 29i11 201 9 72+11 65k10 861 9 100

3.00

911 3

31

3.00

9112

25

+ sd

b,

Culture period was 24 h sd =standard deviation of binomial distribution Control explants were cultured as matched pairs in each experiment, and the total number of all controls is recorded Explants were cultured as controls for 24 h as described, then transferred to 3 pg/ml isotretinoin in medium and cultured 24 h further a

109

A 100

80

[-

60

Heartbeat Vesicle

r

40

formed

20

0 0

.25

1.0

3.0

6.0

1.o RA

20

IsotreIinoin Retinoid Concentration (ug/ml)

B 100

Anterior

Middle

Posterlor

0

Control

RA

Control

RA

No Serum Serum Fig. 3. Development of heart form and function in precardiac explants. Precardiac mesoderm-endoderm regions were removed from stage 6-8 embryos and cultured as explant pairs, one side as a control and the other in the presence of the indicated concentration of isotretinoin or all-trans-retinoic acid (RA). After 24 h the explants were observed for the presence of a heart vesicle and heart contractions. The control explants are all included in one group here. A Development of heart vesicles and contractions. The number of explants in each group (reading from left to right) is 152, 37, 39, 41, 39, 16, and 31 (RA). All-trans-retinoic acid strongly inhibited vesicle formation whereas isotretinoin did not. B Orientation of vesicle position. Control and RA-treated (1 pg/ml) vesicles cultured in the absence or presence of serum were scored according to anterior, middle, or posterior position (legend). The groups had 38 (control, no serum), 18 (RA-treated, no serum), and 13 (control, serum), and 13 (RA-treated, serum) explants each. RA diminished the number of anterior vesicles, especially in cultures without serum

2. The eflect of serum. Because the delivery of retinoids to cells might be modulated by carrier proteins that recognize cell surface receptors, explant cultures were incubated in medium that did or did not contain serum. The results are summarized in Table 2. All control explants acquired heart contractions, and isotretinoin inhibited the percentage that acquired contractions at all doses tested. However isotretinoin was much more effective at the two lower doses of 0.25 and 1.00 pg/ml when serum was present. The difference was greatest a t 1.00 pg/ml (Table 2). At higher doses the presence of serum had little effect.

3. The ejyect of retinoic acid. In order to compare the effects of isotretinoin on development of heart contractions with those of the putative natural morphogen retinoic acid, precardiac explants cultured in medium containing 1 pg/ml RA were examined for the ability to form heart vesicles and to contract. RA proved to be an equally or more potent inhibitor than isotretinoin (Fig. 3 ) . If the experimental data are divided into groups of explants cultured with and without serum (Table 2), this difference is revealed to be the result of much stronger inhibition by RA in the absence of serum. Whereas isotretinoin’s inhibitory effect on development of heartbeat is weaker without serum present, that of RA remains strong.

4 . Orientation of heart vesicle development. Because the precardiac mesoderm migrates anteriorly across the adjacent endoderm, the explants were oriented in culture with the anterior ends identified. The mesoderm moves toward this end to form a tube-like vesicle. The endoderm, however, does not remain flat and passive but contracts so that a somewhat variable and ambiguous morphology emerges. But with the observation of many explant cultures, general trends were noted in controls, and at 5 doses of isotretinoin (data not shown). When treated with isotretinoin, the proportion of anterior vesicles was somewhat diminished. However the effect, as revealed by the data, was not dose-dependent and the number forming posterior vesicles was not consistently affected. These results were the same whether or not serum was present in the culture medium. Thus isotretinoin does not appear to alter directional orientation of the precardiac mesoderm migration as seen in this system. Treatment of explant cultures with retinoic acid, however, did alter heart vesicle morphogenesis (Figs. 2 and 3). Explants treated with 1 pg/ml RA in serum-free medium formed vesicle structure in only 11 of 18 cases (61 YO, compared to 97% of isotretinoin-treated explants at this dose). Among these most were middle (73%) whereas the control counterparts all formed vesicles, most of which were anterior. If serum was present the development of vesicle morphology was better, and more frequently anterior, but the effect of RA was similar (Fig. 3B). Thus RA appears to perturb the orientation of precardiac mesodermal migration and the development of heart vesicle form, particularly in the absence of serum. 5 . Whole embryo cultures. In order to assess the effect

of isotretinoin on heart development in the embryonic tissue environment, whole embryos adhered to filter paper rings were cultured overnight under the same conditions as explants, using 1 or 20 pg/ml isotretinoin, or 1 pg/ml retinoic acid. In embryos treated at 1 pg/ml and occasionally in control embryos, various malformations of the heart, somites, head, and neural tube were observed, however no inhibition of the development of heart contractions was found. In these embryos, whether the primitive hearts were normal or malformed, they were able to contract. Whole embryos cultured at 20 pg/

110

ml isotretinoin, however, did show an inhibition of development that depended on stage (data not shown). Eight embryos at stage 9 or 10 developed normally in culture, with well-formed, beating hearts. But of nine embryos at stages 5, 6, 7, or 8, only five formed hearts in culture, all showing defects in form, and only four had any beating tissue. One embryo formed a complete heart that would not beat at all. Thus the stages earlier than stage 9 were very sensitive to this high dose of isotretinoin, but the sensitivity was overcome by stage 9. The effect of Isotretinoin on actin isotype synthesis Isoelectric focusing gels were used to separate proteins labeled in vitro with [35S]-methionine during heart development in precardiac explants. Figure 4 displays an autoradiogram produced from a dried and exposed gel that shows the results of one experiment in which four pairs of explants were analyzed. In each explant the actins were the most abundant among the labeled proteins resolved. In control explants the muscle-specific a-

1

2

3

4

5

6

7

8

Fig. 4. Isoelectric focusing gel autoradiogram showing control and isotretinoin-treated, 35S-methionine-labeled, cardiac explant sonicates. The position of the 2-,P-, and y-actins is indicated. Control explants are represented in lanes 1, 3, 5, and 7; treated explants in lanes 2, 4, 6, and 8. Lanes 1 and 2 contain a matched pair of explants cultured from a Hamburger-Hamilton stage 8- embryo; lanes 3 and 4 from a stage 8'; lanes 5 and 6 from a stage 6- ; and lunes 7 and 8 from a stage 6 + embryo. The culture period was approximately 16 h. The autoradiogram demonstrates reduction or delay in synthesis of cc-actin

Table 3. Effect of Isotretinoin on actin synthetic ratios in the embryonic chick heart

isoform was always synthesized in addition to the nonmuscle p- and y-isoforms, and was a prominent band. But in explants treated with 3 pg/ml isotretinoin, a-actin was usually found to be absent or much less prominent (Fig. 4). The autoradiogram shown in Fig. 4 and those produced from several experiments were scanned with a densitometer to quantitate the ratios of actin isotype synthesis. The areas under the peaks produced by the densitometer corresponding to a-, p-, and y-actin isotype bands were calculated for each explant and their proportions of the total actin were averaged. Results derived from the autoradiogram shown in Fig. 2 are presented in Table 3. These data indicated a slight inhibition or delay (1 6% difference) in the appearance of muscle-specific a-actin compared to controls.

Discussion The results presented show that isotretinoin exerts specific, previously unidentified effects on early heart development that are not connected to its effect on migrating neural crest cells. These effects include a decrease of cell proliferation, a reduction of the proportion of cultured precardiac explants that develop contractions, and an inhibition or delay of the expected rise in synthetic ratio of a-actin. We also found that at low treatment doses, the effect of isotretinoin is augmented when serum is present in the culture medium. Similar but more pronounced effects were seen with all-trans-retinoic acid. Isotretinoin significantly inhibited cell proliferation in the early myocardium, thus jeopardizing its early growth and the embryonic circulatory capacity. Its average measured effect in endocardium was even greater, raising the possibility of inadequate valve development. However the effect was not significantly different from control as there was large sample variation. Recent evidence suggests that the effect of retinoic acid on cell growth in a variety of normal and transformed cell types is mediated through junctional communication, possibly by preventing the diffusion of growth factors from cell to cell via gap junctions [28], or by altering ion channel activity [ 6 ] .Our data are consistent with this hypothesis. The inhibition of the development of contractions in explant cultures depended on concentration in the lo-' to l o p 5M range. As assessed by Nile blue sulfate dye uptake, and by the healthy appearance and undiminished ability to form vesicles and synthesize proteins observed in treated explants, this arrest of development was not due to cytotoxicity. We also found that all-trans-

Control

actin isotype proportion"

3 kg/ml Isotretinoin

2

P

Y

CI

B

Y

40+3%

15il%

45+2%

23+8Yob

21&7%

56+11%

Areas under peaks from autoradiogram bands scanned by densitometry were calculated, and the proportion of total actin area (all 3 bands) for each isotype was calculated as percent and averaged for 4 pairs of explant sonicates Significantly different from control, PI 0.05

a

111

retinoic acid used at 1 pg/ml caused the same level of inhibition as isotretinoin at 3 pg/ml in serum-free cultures. It is possible that isotretinoin is converted by the tissues to retinoic acid [21], or that it is recognized by the same receptors. Accompanying the failure of heartbeat development, isotretinoin reduced the proportion of cl-actin synthesis as a fraction of total actin synthesis in explant cultures. Although this is apparently a smooth muscle isoform [40, 541, it is a marker for cardiac myocyte differentiation, first detectable in precardiac cells between stages 8 and 9, and subsequently increasing in synthetic ratio [51]. Taken together, the effects observed on cell proliferation and development of contractions are consistent with the interpretation that these retinoids alter junctional communication, a requirement for conduction of excitatory impulses through the cardiac cells [22] as well as for diffusion of growth and differentiation factors. Gap junction communication is now known to be of major importance in chick limb bud patterning [2]. Yet contractions developed in whole embryo cultures despite the presence of retinoid at the same concentration. This paradox may be rectified if adjacent tissues are viewed as being capable of protecting the precardiac cells from excess retinoid. This could be accomplished if retinoid binding proteins in adjacent tissue cells metabolize and inactivate retinoid, or serve as a reservoir for the excess, preventing its diffusion or transport into the heart-forming regions where its effects on junctional development or other developmental events would occur. Indeed, a reservoir role for retinoid binding proteins has been suggested [24, 251. If this is the case, then higher concentrations of retinoid should be expected to inhibit heartbeat development in whole embryo cultures by overcomming the reservoir capacity. In fact we found that treatment with 20 pg/ml did prevent most embryos from developing heartbeat. Although the effects of isotretinoin on heart cell proliferation and differentiation that we have found may be direct ones, they may also result from disruption of gap junctional communication, and either action could be mediated through retinoid binding proteins and the nuclear receptors. The increase in inhibition by isotretinoin of heartbeat development found when explants were cultured in medium containing serum might be interpreted as the result of the higher concentration of retinoid that exists when the retinoid already present in serum (approximately lo-' M in human serum [lo, 111) is added to the experimental concentrations of isotretinoin. At 0.25 pg/ml isotretinoin, with no serum present, the proportion of heartbeat development was inhibited by 29%. However the inhibition at this dose in the presence of serum was 50% ; thus the augmentation of inhibition by serum was 21 YO.Therefore control cultures would be expected to show 21 YOinhibition with serum present. However they showed no inhibition. At the concentration of 1 pg/ml isotretinoin, the augmentation of inhibition by serum was a considerably higher 48%. It seems unlikely that the augmentation of inhibition by serum is due to the added retinoid concentration of serum. A more likely interpretation for the effect of serum is that serum proteins for which the precardiac cells have a receptor(s) arc enhancing the transport or binding of

the retinoids in the cells, increasing their effect. In this case, augmentation of the inhibitory effect over that found in no-serum cultures would be predicted as the dose of retinoid is increased until the serum carrier proteins are saturated, after which the effect would depend only on dose. This interpretation is consistent with the data. The saturating concentration of retinoid for the serum proteins would appear to be about 3 pg/ml ( l o p 5 M), as indicated by the reduction of the difference between inhibition found in the presence and absence of serum at this and higher doses. As all-trans-retinoic acid at the same concentration was equally inhibitory with or without serum, it apparently requires no serum protein for transport or binding to cells in this concentration range. In conclusion, we have presented evidence that isotretinoin, 13-cis-retinoic acid, has multiple effects on growth and differentiation of cardiac myocytes, including an inhibition of cell proliferation, development of heart contractions, and a-actin synthetic ratio. Its effect on the development of heartbeat is augmented in the presence of serum, and can be duplicated by all-transretinoic acid, a likely natural morphogen. These findings provide some insight into the mode of action of retinoids as teratogens, and as endogenous developmental agents. It will be important to examine their effects on other characteristics of cardiac development, particularly the production of extracellular matrix, the assembly of myofibrils, and the formation of cell junctions. Acknowledgements. This work was supported by the American Heart Association, Iowa affiliate grant IA-90-G-40, and by Summer Research Fellowships from the graduate college of the University of Northern Iowa.

References 1. Abergel RP, Meeker CA, Oikarinen H, Oikarinen AI, Uitto J (1985) Retinoid modulation of connective tissue metabolism in keloid fibroblast cultures. Arch Dermatol 121 :632-635 2. Allen F, Tickle C, Warner A (1990) The role of gap junctions in patterning of the chick limb bud. Development 108 :623-634 3. Alles AJ, Sulik KK (1990) Retinoic acid-induced spina bifida: evidence for a pathogenetic mechanism. Development 108:7381 4. Beall AC, Rosenquist T H (1990) Smooth muscle cells of neural crest origin form the aorticopulmonary septum in the avian embryo. Anat Rec 226: 360-366 5 . Benke PJ (1984) The isotretinoin teratogen syndrome. JAMA 251 3257-3269 6. Bosma M, Sidell N (1988) Retinoic acid inhibits Ca2+ currents and cell proliferation in a B-lymphocyte cell line. J Cell Physiol 135:317-323 7. Crawford K, Stocum D L (1988) Retinoic acid proximalizes level-specific properties responsible for intercalary regeneration in axolotl limbs. Development 104:703-712 8. DeHaan RL (1963) Organization of the cardiogenic plate in the early chick embryo. Acta Embryo1 Morphol Exp 6:26-38 9. DeHaan RL (1964) Cell interactions and oriented movements during development. J Exp Zoo1 157: 127-138 10. De Leenheer AP, Lambert WE, Claeys I (1982) All-trans-retinoic acid: measurement of reference values in human sera by high performance liquid chromatography. J Lipid Res 23: 13621367 11. De Ruyter MG, Lambert WE, DeLeenheer AP (1979) Retinoic acid : An endogenous compound of human blood. Unequivocal

112 demonstration of endogenous retinoic acid in normal physiological conditions. Anal Biochem 98 :402-409 12. Doyle CM, Zak R, Fischman DA (1973) DNA synthesis and DNA polymerase in chick hearts. J Cell Biol 59:85a 13. Durston AJ, Timmermans JPM, Hage WJ, Hendriks HFJ, Vries NJ, Heideveld M, Nieuwkoop PD (1989) Retinoic acid causes an anteroposterior transformation in the developing central nervous system. Nature 340: 140-144 14. Edward M, Mackie RM (1989) Retinoic acid-induced inhibition of lung colonization and changes in the synthesis and properties of glycosaminoglycans of metastatic B16 melanoma cells. J Cell Sci 94: 537-543 15. Goulding EH, Pratt RM (1986) Isotretinoin teratogenicity in whole embryo culture. J Craniofac Genet Dev Biol 6:99-112 16. Hamburger V, Hamilton HL (1951) A series of normal stages in the development of the chick embryo. J Morphol 88:49-92 17. Kato S, De Luca LM (1987) Retinoic acid modulates attachment of mouse fibroblasts to laminin substrates. Exp Cell Res 1731450-462 18. Keeble S, Maden M (1989) The relationship among retinoid structure, affinity for retinoic acid-binding protein, and ability to respecify pattern in the regenerating axolotl limb. Dev Biol 132:2 6 3 4 19. Kirby ML, Gale TF, Stewart DE (1983) Neural crest cells contribute to normal aorticopulmonary septation. Science 220:1059-1061 20. Kirby ML, Turnage KL, Hayes BM (1985) Characterization of conotruncal anomalies following ablation of “cardiac” neural crest. Anat Rec 21 3 :87-93 21. Klug S, Lewandowski C, Wildi L, Neubert D (1989) All-transretinoic acid and 13-cis-retinoic acid in the rat whole-embryo culture : abnormal development due to the all-irans isomer. Arch Toxicol 63 :440-444 22. Komuro HA, Hirota A, Yada T, Sakai T, Fujii S, Kamino K (1985) Effects of calcium on electrical propagation in early embryonic precontractile heart as revealed by multiple-site recordings of action potentials. J Gen Physiol 85: 365-382 23. Lammer EJ, Chen DT, Hoar RM, Agnish ND, Benke PJ, Braun JT, Curry CJ, Fernhoff PM, Grix AW, Lott IT, Richard JM, Sun SC (1985) Retinoic acid embryopathy. N Engl J Med 313(14):837-841 24. Maden M, Summerbell D (1986) Retinoic acid-binding protein in the chick limb bud : identification at developmental stages and binding affinities of various retinoids. J Embryo1 Exp Morph97:239-250 25. Maden M, Ong DE, Summerbell D, Chytil F (1988) Spatial distribution of cellular protein binding to retinoic acid in the chick limb bud. Nature 335:733-735 26. Maden M, Ong DE, Summerbell D, Chytil F (1989) The role of retinoid-binding proteins in the generation of pattern in the developing limb, the regenerating limb and the nervous system. Development 1989 Supplement: 109-1 19 27. Manasek FJ (1968) Embryonic development of the heart. I. A light and electron microscopic study of myocardial development in the early chick embryo. J Morph 125:329-366 28. Mehta P, Bertram J, Loewenstein W (1989) The actions of retinoids on cellular growth correlate with their actions on gap junctional communication. J Cell Biol 108: 1053-1065 29. Mitrani E, Shimoni Y (1989) Retinoic acid inhibits growth in agarose of early chick embryonic cells and may be involved in regulation of axis formation. Development 107:275-280 30. Momoi T, Kitamoto T, Seno H, Momoi M (1988) The distribution of cellular retinoic acid binding protein (CRABP) in the central nervous system of the chick embryo during development. Proc Japan Acad 64(B) :294-297 31. Momoi T, Momoi M, Kumagai H, Utsumi H, Miyagawa-Tomita S, Takao A (1989) The expression of retinoic acid receptor in the chick limb bud during early developmental stages. Proc Japan Acad Ser B 65: 191-194 32. Morriss GM (1975) Abnormal cell migration as a possible factor in the genesis of vitamin-A-induced craniofacial anomalies.

In: Neubert D, Merker HJ (eds) New approaches to the evaluation of abnormal embryonic development. Thieme, Stuttgart, pp 678-687 33. O’Farrell PH (1975) High-resolution two-dimensional electrophoresis of proteins. J Biol Chem 250:4007-4021 34. Oikarinen A, Vuorio T, Makela J, Vuorio E (1989) 13-Cis retinoic acid and dexamethasone modulate the gene expression of epidermal growth factor receptor and fibroblast proteoglycan 40 core protein in human skin fibroblasts. Acta Derm Venereol (Stockh) 69:466-469 35. Peck GL, Olsen TG, Yoder FW, Styrauss JS, Downing DT, Pandya M, Butkus D, Arnaud-Battandier J (1979) Prolonged remissions of cystic and conglobate acne with 13-cis-retinoic acid. N Engl J Med 300(7): 329-333 36. Poswillo D (1975) The pathogenesis of the Treacher Collins syndrome (mandibulofacial dystosis). Br J Oral Surg 13: 1-26 37. Pratt RM, Goulding EH, Abbott BD (1987) Retinoic acid inhibits migration of cranial neural crest cells in the cultured mouse embryo. J Craniofac Genet Dev Biol4: 205-218 38. Rawles ME (1943) The heart-forming areas of the early chick blastoderm. Physiol Zoo1 16: 22-42 39. Roberts AB, Sporn MB (1984) Cellular biology and biochemistry of the retinoids. In: Roberts AB, Sporn MB, Goodman DS (eds) The Retinoids, Vol 2. Academic Press, pp 209-286 40. Ruzika DL, Schwartz RJ (1988) Sequential activation of a-actin genes during avian cardiogenesis : vascular smooth muscle ractin gene transcripts mark the onset of cardiomyocyte differentiation. J Cell Biol 107:2575-2586 41. Schroder EW, Black PH (1980) Retinoids; tumor preventers or tumor enhancers? J Natl Cancer Inst 65 :671-675 42. Shenefelt RE (1972) Morphogenesis of malformations in hamsters caused by retinoic acid: Relation to dose and stage at treatment. Teratology 5 : 103-1 18 43. Sive HL, Draper BW, Harland RM, Weintraub H (1990) Identification of a retinoic acid-sensitive period during primary axis formation in Xenopus laevis. Genes Dev 4: 932-942 44. Smith SM, Eichele G (1991) Temporal and regional differences in the expression pattern of distinct retinoic acid receptor-p transcripts in the chick embryo. Development 111:245-252 45. Smith-Thomas L, Lott I, Bronner-Fraser M (1987) Effects of isotretinoin on the behavior of neural crest cells in vitro. Dev Biol 123:276-280 46. Sulik KK, Cook CS, Webster WS (1988) Teratogens and craniofacial malformations : relationships to cell death. Development 103 Supplement: 213-232 47. Thaller C , Eichele G (1987) Identification and spatial distribution of retinoids in the developing chick limb bud. Nature 327: 625-628 48. Thorogood P, Smith L, Nicol A, McGinty R, Garrod D (1982) Effects of vitamin A on the behavior of migratory neural crest cells in vitro. J Cell Sci 57: 331-350 49. Tickle C, Alberts BM, Wolpert L, Lee J (1982) Local application of retinoic acid to the anterior margin of the limb bud mimics the action of the polarizing region. Nature 296: 564565 50. Tickle C, Lee J, Eichele G (1985) A quantitative analysis of the effect of all-trans-retinoic acid on the pattern of chick wing development. Dev Biol 109:82-95 51. Wiens D, Spooner BS (1983) Actin isotype biosynthetic transitions in early cardiac organogenesis. Eur J Cell Biol 30 :60-66 52. Wiens D, Sullins M, Spooner B (1984) Precardiac mesoderm differentiation in vitro : actin-isotype synthetic transitions, myofibrillogenesis, initiation of heartbeat, and the possible involvement of collagen. Differentiation 28 :62-72 53. Wolf G (1984) Multiple functions of vitamin A. Physiol Rev 64:873-937 54. Woodcock-Mitchell J, Mitchell JJ, Low RB, Kieny M, Sengel P, Rubbia L, Skalli 0 ,Jackson B, Gabbiani G (1988) cc-Smooth muscle actin is transiently expressed in embryonic rat cardiac and skeletal muscles. Differentiation 39: 161-166 55. Zak R (1973) Cell Proliferation during cardiac growth. Am J Cardiol 3 1 :21 1--219