Endocardial cushion development and heart loop architecture in the trisomy 16 mouse

DEVELOPMENTAL DYNAMICS 206:301309 (1996)

Endocardia1 Cushion Development and Heart Loop Architecture in the Trisomy 16 Mouse SANDRA WEBB, ROBERT H. ANDERSON, AND NIGEL A. BROWN Department of Anatomy and Developmental Biology, St. George’s Hospital Medccal School, London, SWl7 ORE, United Kingdom (S.W., N.A.B.); Department of Paediatrics, National Heart and Lung Institute, London, S W 3 6LY, United Kingdom (S.W., R.H.A.)

ABSTRACT Murine trisomy 16 (Ts16) is a model for Down’s syndrome and has close to a 100% incidence of atrioventricular septal defects (AVSDs). These have been proposed to result from abnormal development of the endocardial cushions, but the mechanisms are unknown. We aim to identify the initial defects in Ts16 hearts, both to characterise the pathogenesis of AVSDs and as a first step in the search for molecular mechanisms. In 38 litters from an Rb(l1.16)2H/Rb(16.17)7Bnr x C57BU6J cross, which was examined on days 10 and 11 of gestation, 28.4% of embryos were trisomic. Trisomic embryos were uniformly retarded compared to their normal litter mates, having on average 3.3 fewer somite pairs. All further comparisons were made between embryos of the same somitic stage. Twenty-one trisomic and 21 normal embryos of between 15 and 43 somites were seri- ally sectioned, and stereomorphometric methods were used to reconstruct the volumes of the endocardial cushions and to count their number of mesenchymal cells. There were fewer cells in Ts16 superior and inferior cushions. In contrast, the volumes of trisomic cushions were significantly greater than normal. Thus, cell density was markedly lower in trisomic cushions. Importantly, the volumes of the cushions in trisomic embryos were already greater than normal at the 18 somite stage, prior to the invasion of cushions by mesenchymal cells. The architecture of Ts16 heart tubes in 1625 somite embryos was subtly abnormal. This was reflected in the angle between the axis of the atrioventricular canal and the first pharyngeal cleft, which was significantly larger in trisomic hearts and showed a different relationship to somite stage when compared to normal embryos. These observations suggest that the primary cardiac defect in Ts16 mice may be localised to the myocardium, thus influencing the shape of the heart tube, with changes in the mesenchymal population of the endocardial cushions being later events. Whether AVSDs arise from one or both of these abnormalities remains to be estab- lished. o ISM Wiley-Liss, Inc.

Key words: Trisomy 16, Heart, Development, En- docardial cushion, Atrioventricular septal defect

0 1996 WILEY-LISS. INC

INTRODUCTION The membranous atrioventricular septum ties to-

gether the central structures of the mammalian heart: the atrioventricular valves, the crest of the muscular ventricular septum, and the leading edge of the atrial septum. In addition, because the tricuspid valve is set below the mitral valve, part of the atrioventricular septum is also a muscular structure that partitions the right atrium from the left ventricle. This segment is formed by the overlapping margins of the muscular atrial and ventricular septa. In humans, a characteristic congenital malformation is produced by disjunc- tion of these structures, known variously as atrioventricular septal defect (AVSD), endocardial cushion defect, or atrioventricular canal malformation. The lesion is characterised by absence of atrioventricular septal structures, “scooping” of the ventricular septum, and presence of a common atrioventricular junction, which, in most cases, is guarded by a common valvular orifice with unique superior and inferior bridging leaflets (Anderson et al., 1991).

The development of the various parts of the atrioventricular septum is not well understood, although the early steps are being identified. When the heart tube forms, its organisation has been likened to that of an extended basement membrane (Kitten et al., 1987). It consists of an endothelial lining separated from an outer layer of myocardial cells by an extracellular matrix, the so-called cardiac jelly (Davis, 1924). In the regions of the future outflow tract and atrioventricular canal, endocardial cushions develop as especially thick- ened regions of cardiac jelly that occlude the lumen. The heart at this stage beats in a peristaltic fashion, and the opposing atrioventricular endocardial cushions, filled with the pliable extracellular matrix, act as primitive valves to prevent retrograde flow of blood.

Initially, the extracellular matrix of the endocardial cushions is acellular. Some of the epithelial endocardial cells that form the luminal face of the cushions then undergo a morphological transformation into a migrating mesenchymal cell population, which moves

Received October 23, 1995; accepted January 29, 1996. Address reprint requestskorrespondence to Sandra Webb, Depart-

ment of Anatomy and Developmental Biology, St. George’s Hospital Medical School, Cranmer Terrace, London, SW17 ORE, United King- dom.

302 WEBB ET AL.

into the cushions (Patten et al., 1948; Markwald et al., 1990a). This transformation is a response to inductive signals from the myocardium that appear to be re- stricted to the atrioventricular and outflow regions. Some of the molecules secreted by the myocardium into the extracellular matrix have been identified, as have some of the genes that are regulated in the endocardium (Krug et al., 1985, 1987; Markwald et al., 1990a; Huang et al., 1995). The further form-shaping changes in the cushions are not well understood, and, although it is clear that the endocardial cushions play a funda- mental role in the septation of the heart, their exact material contribution to the formation of septa and valvular leaflets remains unknown.

The trisomy 16 (Ts16) mouse is a valuable model for studies of atrioventricular development, because, uniquely among the mouse trisomies, it has a very high incidence of AVSDs (Pexieder et al., 1981). Ts16 is also considered a model for Down’s syndrome (Epstein, 1985; Epstein et al., 1985; Reeves et al., 1986; Gearhart et al., 1986) because of the genetic synteny between human chromosome 21 and mouse chromosome 16 (Nadeau, 1989; Lyon and Kirby, 1995). The obligate Down’s region maps to mouse chromosome 16, although a small telomeric portion of human chromosome 21 maps to mouse chromosomes 10 and 17. Con- genital heart defects are found in approximately 40% of live births in Down’s syndrome, and most are AVSDs. This contrasts with congenital heart defects in the gen- eral population, where only 2% present as AVSDs (Fer- encz et al., 1989). The obligate Down’s region appears to be implicated in heart defects (Korenberg et al., 1990, 1992), although this is based on data from a small number of affected individuals.

We have been using Ts16 to investigate systematically the development of the atrioventricular endocardial cushions in comparison to normal hearts. In this study, the aim was to identify the initial morphological changes in Ts16 hearts. It is generally thought that failure of the endocardial cushions to fuse results in the characteristic lesion, although lack of fusion could be a consequence rather than a cause. Therefore, we have studied the hearts of Ts16 and normal embryos before and during the period in which the cushions are populated with mesenchymal cells. We find that there are fewer cells in trisomic cushions but, more importantly, that the cushions differ in size and shape at an earlier stage, and the architecture of the whole heart tube is abnormal in Ts16 embryos.

RESULTS Developmental Data

We examined 38 litters collected on days 10 and 11 of gestation. On average, 28.4% of embryos in each litter were trisomic (range 0-100%). To compare the developmental stages of trisomic and disomic embryos, we calculated the mean number of somites for embryos of the two karyotypes in each litter. The mean difference was 3.3 somites, with the trisomic embryos having

fewer than their normal disomic litter mates (P < 0.001; paired t test). The increase in somites over time was compared between trisomic and disomic embryos. There was no significant difference in the rate of somitic development between the normal and trisomic groups over days 10 and 11 of gestation [normals: 10.5 k 2.3 (S.E.) somited24 hrs; trisomics: 11.6 2 2.5 somited24 hrsl. These data suggest that trisomic embryos were already developmentally delayed at the start of day 10 and did not catch up significantly over this 2-day period. The retardation of trisomic embryos appeared to be uniform, with no overt evidence of asyn- chrony between individual tissues or organs. In all further parts of this study, comparisons between trisomic and disomic morphologies were made for embryos with the same number of somites but not necessarily from the same litter.

Stereological Morphology Study For serial sectioning and morphometry, we selected

at random 21 disomic and 21 trisomic embryos from the 38 litters, covering the range of 15-43 somites. This included the period when the endocardial cushions be- gan to be populated by endocardially derived mesenchymal cells. The calculated cushion volumes, cell numbers, and densities for all 42 individual embryos are shown in Figure 1. In addition, the values for embryos of 20 somites (day 10, at the start of cellularisa- tion) and 35 somites (day 11, with well-populated cushions) are given in Table 1, and representative sections at these stages are shown in Figure 2. Endocardial cushion volume. The volumes of

both the superior and inferior cushions were greater in Ts16 embryos than in disomic controls. It is striking that, of the 24 embryos examined with 18-31 somites, only one trisomic embryo (25 somite) did not have a cushion volume greater than normal (Fig. lA), although there was some overlap between the two popu- lations at the earliest and latest stages studied. Anal- ysis of the whole data set by regression, with trisomy and the number of somites as predictor variables, showed that the difference between trisomic and disomic embryos was statistically highly significant (P c 0.001 and P < 0.002 in the superior and inferior cushions, respectively; Fig. 1A). The superior cushion was significantly larger than the inferior cushion in both the disomic and trisomic embryos (P < 0.001 for both normal and trisomic groups; t test). The magnitude of the disparity between the superior and inferior cushions was greater in the trisomic than in the normal embryos (P < 0.001; t test).

The shape of the endocardial cushions was different in the trisomic embryos compared to normal embryos, and this may explain the larger volumes of the cushions. On day 10, the cushions in trisomic embryos were elongated and tapered farther into the atrial compo- nent of the heart, particularly with respect to the superior cushion (compare to 20-somite embryos in Fig. 2A,B). In some embryos at day 11, this was still the

ENDOCARDIAL CUSHION DEVELOPMENT IN THE Ts16 MOUSE 303

Fig. 1. Endocardial cushion volumes, cell numbers and cell densities for 21 normal (triangles) and 21 trisomy 16 (squares) embryos on days 10 and 11 of gestation in relation to numbers of somites. P values refer to

regression analyses comparing normal and trisomic embryos. A: Cush- ion volumes. B: Numbers of cells within the cushions. C: Cell density (number of cells/unit volume) within the cushions.

304 WEBB ET AL.

TABLE 1. Endocardia1 Cushion Volumes and Cell Numbers in Trisomic and Normal Embryos of 20 Somites (Day 10) and 35 Somites (Day 11)

Cushion volume Number o f Cushion cel l density ( ~ 1 0 ~ pm3) cushion cells (cells/pm3)

Number o f Karyotype somites Superior In fer ior Superior In fer ior Superior In fer ior

Norma l 20 3.3 2.3 18 78 5.5 34.0 21 2.5 2.3 36 36 14.5 15.8

Trisomic 20 4.7 4.7 6 60 1.3 12.7 20 4.9 4.2 6 24 1.2 5.8

Norma l 34 8.8 7.4 1,206 1,368 137.0 184.1 35 4.4 4.6 1,236 1,536 282.6 334.9

Trisomic 34 8.2 6.4 942 1,404 114.7 221.0 35 6.4 4.7 1,002 1,242 155.9 262.0

Fig. 2. Micrographs of representative sections through the atrioventricular canal of a normal day 10 embryo of 20 somites (A), a trisomic day 10 embryo of 20 somites (B), a normal day 11 embryo of 34 somites (C), and a trisomic day 11 embryo of 34 somites (D) . The endocardial cushions at 20 somites have few or no cells, and the trisomic cushions (6) are longer and taper farther into the primary atrium compared to normal (A). At 34 somites, the cushions of the normal embryo (C) show an even

case, but, in many, the difference in shape was not apparent (see Fig. 2C,D). By day 1 2 , l day later than in this study, normal cushions had changed markedly to a more cuboidal shape, whereas we have observed that tapering persists in trisomic embryos, resulting in longer, slender cushions (data not shown).

Total numbers of cells. The endocardial cushions

distribution of mesenchymal cells, with a slightly higher density in the inferior cushion. The cushions of the trisomic embryo of 34 somites (D) demonstrate a reduction in the number of mesenchymal cells in comparison to the normal disomic embryo. The myocardial layer in the region of the atrioventricular canal may be thinned and altered in morphology in the trisomic embryos. Scale bar = 50 pm.

of all embryos at the 15-somite stage were acellular, and very few cells were present by 20 somites (Table 1). Over the whole period studied, there were fewer cells in trisomic cushions. By regression analysis, as described above, the number of cells in superior cushions was statistically different in trisomic and disomic embryos (P < 0.011, whereas the comparison for inferior cush-


ions was just nonsignificant (P = 0.055; Fig. 1B). There were more cells in inferior than in superior cushions for both normal and trisomic embryos (P < 0.001; t test), but the magnitude of this disparity was not different in trisomic and disomic embryos. The differences in cellularity between trisomic and disomic and superior and inferior cushions are shown in Figure 2C,D.

Density of cells. Given the described differences in volumes and total numbers of cells, it follows that the density of mesenchymal cells (number of cells per unit volume) was markedly reduced in both endocardial cushions in the trisomic group (P < 0.001 for both superior and inferior cushions; Fig. 1C). In both trisomic and disomic animals, the density of cells in the inferior cushion was greater than in the superior cushion (P < 0.001; t test). With regard to cell numbers, this disparity was not different between the two groups.

Angle of the Atrioventricular Junction Subjective comparison of trisomic and disomic heart

tubes suggested a significant difference in overall architecture. To examine this systematically, a second series of embryos was collected from nine litters. The mean proportion of trisomic embryos (28.9%) was consistent with the first series, as was the developmental retardation (mean difference of 2.2 somites). To quan- tify the arrangement of the heart tube relative to the body, the position of the atrioventricular canal compared to the first pharyngeal cleft was measured as the angle between them (Fig. 3). This angle was measured in 32 normal and 13 trisomic embryos and analysed by regression, with trisomy and the number of somites as predictor variables. There was a significant effect of trisomy (P < 0.001), with an estimated difference between the two groups of 15.5" (95% confidence interval 9.3-21.7'). There was also a significant (P = 0.03) interaction between trisomy and somitic number. This shows that the relationship between the measured angle and the number of somites was different in disomic and trisomic mice, the angle decreasing with increasing somites for disomic mice but increasing for trisomics (Fig. 4A). For disomic mice that were analysed separately, the decrease in angle was statistically significant (P = 0.03), but the increase in trisomic mice was not significant (P = 0.2), probably because of the fewer embryos in this group.

To evaluate whether the heart angle and the relationship with somites observed in the disomic group was consistent with other normal embryos, we measured the angle between the atrioventricular canal and the first pharyngeal cleft in 53 embryos from a C57BL16J x C57BL/6J cross (Fig. 4B). The absolute angle and the decrease with increasing somites were consistent with the disomic series of embryos.

DISCUSSION We have demonstrated that the atrioventricular en-

docardial cushions in trisomic mice are significantly larger in volume than those in disomic controls, par-

ticularly over the 18-31 somite phase. This is in contrast to previous reports describing the cushions to be hypoplastic in Ts16 embryos (Miyabara, 1990a,b). The earlier reports, however, were based on gross observa- tion and did not involve any quantitative measure- ments. We have shown that, although the cushions in trisomic mice on the day 10 of gestation have a greater volume, they are elongated and taper markedly into the atrium. This morphology persists in the trisomic mice, whereas, by day 12 in normal hearts, the cushions become more cuboidal. The slender and tapered appearance could well have been misinterpreted as a reduction in size in previous studies. A new study (Hilt- gen et al., 1996), in consonance with our findings, reports endocardial cushion volumes larger on average in Ts16 on days 11 and 12 of gestation. Hiltgen et al., however, did not observe any statistical significant difference, probably because of their experimental design (see below).

The numbers of cells in trisomic cushions were reduced; therefore, coupled with the increased volume, the cellular density was markedly less than in normal embryos. Significantly, we have also demonstrated that the architecture of the heart tube is abnormal in trisomic embryos. Although we have no direct evidence for the causal relationships between the changes in cushion volume, cellularity, and heart tube morphology, the temporal sequence is informative. The increased volume of the endocardial cushions and abnormalities in the heart tube are present a t a stage (about 18-20 somites) when there are few, if any, cells in the cushions of normal or trisomic embryos. It is unlikely, then, that any defect in production, migration, or function of cells within the cushions is the cause of the defects in volume or shape, although the converse relationship is plausible. Because the signals from the myocardium that induce the endocardium to transform into mesenchymal cells must traverse the cushion, it is possible that an increase in cushion volume could di- minish the signal.

The abnormal cushion volume and the change in heart tube shape both point towards the myocardium as the site of the primary defect in Ts16. It is the myocardium that secretes the extracellular matrix, which fills the endocardial cushions (Markwald et al., 1990a,b). In addition, the myocardium, as the major cellular layer, is presumably the primary determinant of the shape of the heart tube, as demonstrated in the zebrafish by Stainier et al. (1995). We have observed an apparent thinning of the myocardium in Ts16 hearts, but whether this is a consistent difference awaits ob- jective confirmation. There are some indications that the extracellular matrix of Ts16 cushions is not normal. Staining of the extracellular matrix with alcian blue at pH 2.5 showed a regular network of filaments with fine granules. Migrating mesenchymal cells appeared to be distributed along the network in normal mice, whereas, in Ts16 cushions, the network was ir- regular or missing in parts (Miyabara, 1990a). In ad-

306 WEBB ET AL.

Fig. 3. Measurement of the heart angle in day 10, 28-somite embryos. A: Normal. B: Trisomic (left lateral view). The drawings show the positions of the reference line, which is drawn through the first pharyngeal cleft, and the test line, which is drawn parallel to the luminal face of the

superior endocardial cushion. The angle was measured at the intersec- tion of the two lines. This trisomic embryo has a particularly pronounced abnormality of the heart tube architecture, whereas most were more sub- tle.


Fig. 4. Heart angles of disomic and trisomic embryos in relation to somite stage. A: Thirty-two disomic (triangles) and 13 trisomic (squares) embryos from nine litters on day 10. B: Fifty-three normal C57BU6J embryos on day 10. Best-fit regression lines are shown.

dition, there appeared to be less fibronectin-immuno- reactive material in Ts16 cushions. In contrast, no difference was observed in the rate of proliferation of mesenchymal cells between trisomic and normal mice, as determined by labelling with bromodeoxyuridine (Miyabara, 1990a).

What is the relationship between the development of AVSDs and cushion size, cell numbers, or heart tube shape? Others have argued that abnormal cushion shape and cellularity may prevent normal fusion of the cushions and, hence, lead to AVSD (Miyabara, 1990a; Hiltgen et al., 1996). We have confirmed that the normal pronounced change in cushion shape between days 10 and 12 of gestation does not occur in Ts16 hearts. A recent study suggests that the mesenchymal cells iso-

lated from Ts16 hearts have a different interaction with extracellular matrix components compared to cells from normal hearts (Carver, personal communi- cation). Although the cushions clearly play a role in the development of heart defects in Ts16, we would argue that misalignment of the heart tube is also of funda- mental importance. Others have previously noted abnormal ventriculoarterial connections in Ts16 hearts at term (Miyabara, 1990a), and we have recently com- pleted a systematic study (Webb et al., unpublished data). We find that in none of Ts16 hearts at full term is the aorta exclusively connected to the left ventricle. This is additional evidence supporting the notion that these defects, and perhaps the AVSDs, are a result of early misalignment of the components of the looped heart tube.

Another interpretation of the outflow tract defects and of the thymic hypoplasia, which is also seen in Ts16 fetuses, is that these relate to abnormalities in migration of the neural crest. Neural crest cells from the occipital region migrate to the outflow tract of the heart in avian and mammalian embryos (Kirby et al., 1983; Kirby and Waldo, 1990; Fukiishi and Morriss- Kay, 1992; Serbedzija et al., 1992). The outflow tract defects in Ts16 mice at term are reminiscent of those seen in DiGeorge syndrome in the human, which are thought to be related to abnormal migration of the neural crest origin. Mouse chromosome 16 also has some genetic synteny with human chromosome 22qll (Lyon and Kirby, 1995), a region that is implicated in Di- George syndrome (Carey et al., 1992; Driscoll et al., 1992; Wilson et al., 1992).

In the light of our finding that the volume of the cushions is increased in Ts16 mice, it is interesting that the membranous atrioventricular septum may be enlarged in humans with trisomy 21. Rosenquist et al. (1974) reported that Down’s syndrome patients without septa1 defects have a significant enlargement of the membranous atrioventricular septum when compared to normal hearts. It is also worth noting that our finding of consistently more cells in the inferior than the superior cushion in mice is exactly paralleled in the chick, where, at every developmental stage examined, the inferior cushion contained more cells than the superior cushion (Wienecke et al., 1995).

We have shown that Ts16 embryos, in common with other mouse trisomies, are smaller than their normal litter mates. We have quantified this delay to be an average of 3.3 somites on days 10 and 11 of gestation. It has been suggested that altered placental morphology in trisomic mice may be responsible for the delay in development (Bersu et al., 1989; Kornguth et al., 1992). Although we found that Ts16 embryos were already retarded on day 10, we have no information on the earliest stage at which delay can be detected, which could even be prior to implantation. It is clear, however, that a difference in stage of 3.3 somites can influence accurate comparison of heart morphology during early organogenesis, and this may be a complication in pre-

308 WEBB ET AL.

vious studies of Ts16 development (Miyabara, 1990a,b; Hiltgen et al., 1996).

EXPERIMENTAL PROCEDURES Generation of Trisomic Embryos

Ts16 embryos were generated by crossing males dou- bly heterozygous for two Robertsonian translocations [Rb(11.16)2H/Rb(16.17)7Bnrl with C57BL/6J females (a normal, all acrocentric strain). This produces litters that are a mix of disomic (normal), monosomic 16 (which die before day lo), and Ts16 embryos. After mating overnight, the presence of a copulation plug the next morning was taken as evidence of successful mating, and this was designated as day 1 of gestation. Dams were killed by cervical dislocation on days 10 and 11 of gestation, and conceptuses were explanted into Minimal Essential Medium prewarmed to 37°C. Under a binocular dissecting microscope, the extraem- bryonic membranes were removed, and the visceral yolk sacs were reserved for karyotyping. Embryos were examined for gross morphology, and the number of pairs of somites were counted. Embryos were then im- mediately fixed by immersion in 2% glutaraldehyde and 1% formaldehyde buffered with 0.05 M sodium cac- odylate, pH 7.4 (adjusted to 330 mosm with NaCl), at 4°C overnight.

Karyotyping-Acetic Acid Disaggregation Method

Visceral yolk sacs were placed in a hypotonic aqueous solution of 1% trisodium citrate for 30 min at room temperature followed by overnight fixation at 4°C in freshly prepared methano1:glacial acetic acid (3:l). Af- ter fixation, the membranes were disaggregated by placing in 60% acetic acid (aqueous solution) in an em- bryological watch glass for 5 min at room-temperature. Metaphase spreads were prepared from 100 pl aliquots of the disaggregated samples and were stained with 10% Giemsa’s stain (modified formula) in phosphate- buffered saline, pH 7.4. The metaphase spreads were examined for the presence of metacentric (Robertso- nian) chromosomes; disomic mice have one metacentric chromosome (a total of 40 chromosome arms), whereas trisomic embryos have two metacentric chromosomes (a total of 41 chromosome arms).

Serial Sectioning After fixation, embryos were rinsed, dehydrated, em-

bedded in paraffin wax, and serial sectioned longitu- dinally at a nominal thickness of 5 pm by using a ro- tary microtome. Sections were dewaxed, rehydrated, stained with Harris’s haematoxylin, and counter- stained with 1% eosin (aqueous solution). The sections were dehydrated and coverslipped by using DPX mounting medium.

Morphometric Analysis A total of 42 embryos were examined, 21 disomic and

21 trisomic. For each embryo, every sixth section and

its adjacent section were examined throughout the area containing the atrioventricular endocardial cushions. The position of the first section sampled was cho- sen by random number, R, taken from a random number table, where 1 5 R 5 6, followed by every sixth section through the endocardial cushions. Micrographs of the sampled sections were taken using a Zeiss D-7082 transmitted-light stereo photomicroscope.

Particles of an arbitrary size can be counted unambiguously by comparing two parallel adjacent sections, one designated the reference section and the other designated the look-up section. The chance of sampling any given particle in a single section is pro- portional to its size, with large particles having a greater chance of being sampled than smaller ones. If particles are counted in two adjacent sections, then the probability that a particle is sampled in one section but not in the parallel section is the same for both large and small particles (see in appendix A in Gundersen, 1986). To avoid missing particles in the space between the two facing planes of a pair of adjacent sections, they must be separated by a distance smaller than the minimum particle height, h (Sterio, 1984).

The size of the nuclei in the migrating mesenchymal cells of the endocardial cushions was found to be greater than 6 Fm, when measured with an eyepiece micrometer that was calibrated with a stage micrometer. The optimal dissector height for this study, therefore, was taken to be 5 pm. Mesenchymal cell nuclei that were seen in the first (reference) but not in the second (look-up) section were counted. Mesenchymal cell nuclei that were seen in both sections of the dissector pair or that were present in the look-up section, but not in the reference section, were disregarded.

Endocardia1 cushion volume. The volume of the endocardial cushion tissue was estimated by using Cavalieri’s principle, which states that the volume of an object is equal to the sum of the area of its sections multiplied by the mean section thickness. Because the entire endocardial cushions were sectioned for each sample, one may select a known and fixed fraction of sections for the estimation of endocardial cushion area (Pakkenberg and Gundersen, 1988). Again, for this study, every sixth section was selected. The cushion area was calculated by video-capture of the micrograph image and analysis by using the Freelance software package (Sight Systems, Foster Findley Associates Ltd.).

Angle of the Atrioventricular Junction A series of 13 trisomic and 32 normal embryos were

collected from nine litters on day 10 of gestation, and the yolk sacs were karyotyped. The embryos were pho- tographed, and the angle of the atrioventricular junction was measured. A reference line was drawn through the first pharyngeal cleft, and a second line was drawn through the atrioventricular canal parallel to the luminal face of the superior cushion. The angle formed by the test line bisecting the reference line was

Cell counts.


measured (Fig. 3). Regression analysis was applied to the results. To confirm the trend seen in the disomic embryos, the angle of the atrioventricular canal, relative to a reference line drawn through the first pharyngeal cleft, was measured in a further 53 embryos from a C57BL/6J x C57BL/6J cross (the maternal back- ground strain).

ACKNOWLEDGMENTS The authors thank Ian Harragan for his expert tech-

nical assistance, Ceri Davies for advice on morphometry, and Martin Bland for the statistical analyses. This study was financed by the British Heart Foundation and the U.K. Medical Research Council.

REFERENCES Anderson, R.H., Baker, E.J., Ho, S.Y., Rigby, M.L., and Ebels, T. (1991) The morphology and diagnosis of atrioventricular septa1 defects. Cardiol. Young 1:290-305.

Bersu, E.T., Mossman, H.W., and Kornguth, S.E. (1989) Altered placental morphology associated with murine trisomy 16 and murine trisomy 19. Teratology 40:513-523.

Carey, A.H., Kelly, D., Halford, S., Wadey, R., Wilson, D., Goodship, J., Burn, J . , Paul, T., Sharkey, A., Dumanski, J., et al. (1992) Mo- lecular genetic study of the frequency of monosomy 22qll in Di- George syndrome. Am. J . Human Genet. 51:964-970.

Davis, C.L. (1924) The cardiac jelly of the chick embryo. Anat. Rec. 27:201-202.

Driscoll, D.A., Budarf, M.L., and Emanuel, B.S. (1992) A genetic eti- ology for DiGeorge syndrome: Consistent deletions and microdele- tions of 22qll. Am. J . Human Genet. 50:924-933.

Epstein, C.J. (1985) Mouse monosomies and trisomies as experimental systems for studying mammalian aneuploidy. Trends Genet. 1:129-134.

Epstein, C.J., Cox, D.R., and Epstein, L.B. (1985) Mouse trisomy 16: An animal model of human trisomy 21 (Down syndrome). Ann. N.Y. Acad. Sci. 450:157-168.

Ferencz, C., Neill, C.A., Boughman, J.A., Rubin, J.D., Brenner, J.I., and Perry, L.W. (1989) Congenital cardiovascular malformations associated with chromosome abnormalities: An epidemiologic study. J . Pediatr. 114:79-86.

Fukiishi, Y., and Morriss-Kay, G.M. (1992) Migration of cranial neural crest cells to the pharyngeal arches and heart in rat embryos. Cell Tissue Res. 268:l-8.

Gearhart, J.D., Davisson, M.T., and Oster-Granite, M.L. (1986) Au- tosomal aneuploidy in mice: Generation and developmental conse- quences. Brain Res. Bull. 16:789-801.

Gundersen, H.J. (1986) Stereology of arbitrary particles. A review of unbiased number and size estimators and the presentation of some new ones, in memory of William R. Thompson. J . Microsc. 143:3-45.

Hiltgen, G.G., Markwald, R.R., and Litke, L.L. (1996) Morphoge- neticic alterations during endocardial cushion development in the trisomy 16 (Down syndrome) mouse. Pediatr. Cardiol. 17:21-30.

Huang, J.X., Potts, J.D., Vincent, E.B., Weeks, D.L., and Runyan, R.B. (1995) Mechanisms of cell transformation in the embryonic heart. Ann. N.Y. Acad. Sci. 752:317-330.

Kirby, M.L., and Waldo, K.L. (1990) Role of neural crest in congenital heart disease. Circulation 82:332-340.

Kirby, M.L., Gale, T.F., and Stewart, D.E. (1983) Neural crest cells contribute to normal aorticopulmonary septation. Science 220: 1059-1061.

Kitten, G.T., Markwald, R.R., and Bolender, D.L. (1987) Distribution of basement membrane antigens in cryopreserved early embryonic hearts. Anat. Rec. 217:379-390.

Korenberg, J.R., Kawashima, H., Pulst, S.-M., Ikeuchi, T., Oga- sawara, N., Yamamoto, K., Schonberg, S.A., West, R., Allen, L., Magenis, E., Ikawa, K., Taniguchi, N., and Epstein, C.J. (1990)

Molecular definition of a region of chromosome 21 that causes fea- tures of the Down syndrome phenotype. Am. J . Human Genet. 47: 236-246.

Korenberg, J.R., Bradley, C., and Disteche, C.M. (1992) Down syndrome: Molecular mapping of the congenital heart disease and duo- denal stenosis. Am. J. Human Genet. 50:294-302.

Kornguth, S., Bersu, E., Anderson, M., and Markley, J . (1992) Cor- relation of increased levels of class I MHC H-2Kk in the placenta of murine trisomy 16 conceptuses with structural abnormalities re- vealed by magnetic resonance microscopy. Teratology 45383-391.

Krug, E.L., Runyan, R.B., and Markwald, R.R. (1985) Protein extracts from early embryonic hearts initiate cardiac endothelial cytodiffer- entiation. Dev. Biol. 112:414-426.

Krug, E.L., Mjaatvedt, C.H., and Markwald, R.R. (1987) Extracellular matrix from embryonic myocardium elicits an early morphogenetic event in cardiac endothelial differentiation. Dev. Biol. 120:348-355.

Lyon, M.F., and Kirby, M.C. (1995) Mouse chromosome atlas. Mouse Genome 93:23-66.

Markwald, R.R., Mjaatvedt, C.H., and Krug, E.L. (1990a) Induction of endocardial cushion tissue formation by adheron-like, molecular complexes derived from the myocardial basement membrane. In: “Developmental Cardiology: Morphogenesis and Function,” Clark, E.B., and Takao, A. (eds). New York: Futura Publishing, pp. 191- 204.

Markwald, R.R., Mjaatvedt, C.H., Krug, E.L., and Sinning, A.R. (1990b) Inductive interactions in heart development. Role of cardiac adherons in cushion tissue formation. Ann. N.Y. Acad. Sci. 588:13- 25.

Miyabara, S. (1990a) Cardiovascular malformations of mouse trisomy 16: Pathogenetic evaluation as an animal model for human trisomy 21. In: “Developmental Cardiology: Morphogenesis and Function,” Clark, E.B., and Takao, A. (eds). New York: Futura Publishing, pp. 409-430.

Miyabara, S. (1990b) Animal model for congenital anomalies induced by chromosome aberrations. Congen. Anom. 30:49-68.

Nadeau, J. (1989) Maps of linkage and synteny homologies between mouse and man. Trends Genet. 582-86.

Pakkenberg, B., and Gundersen, H.J.G. (1988) Total number of neu- rons and glial cells in human brain nuclei estimated by the dissector and the fractionator. J . Microsc. 15O:l-20.

Patten, B.M., Kramer, T.C., and Barry, A. (1948) Valvular action in the embryonic chick heart by localized apposition of endocardial masses. Anat. Rec. 102:299-311.

Pexieder, T., Miyabara, S., and Gropp, A. (1981) Congenital heart disease in experimental (fetal) mouse trisomies: Incidence. In: “Mechanisms of Cardiac Morphogenesis and Teratogenesis,” Pex- ieder, T. (ed). New York Raven Press, pp. 389-399.

Reeves, R.H., Gearhart, J.D., and Littlefield, J.W. (1986) Genetic basis for a mouse model of Down syndrome. Brain Res. Bull. 16803-814.

Rosenquist, G.C., Sweeney, L.J., Amsel, J., and McAllister, H.A. (1974) Enlargement of the membranous ventricular septum: An internal stigma of Down’s syndrome. J . Pediatr. 85490-493.

Serbedzija, G.N., Bronner-Fraser, M., and Fraser, S.E. (1992) Vital dye analysis of cranial neural crest cell migration in the mouse embryo. Development 116:297-307.

Stainier, D.Y.R., Weinstein, B.M., Detrich, H.W. 111, Zon. L.I., and Fishman, M.C. (1995) Cloche, an early acting zebrafish gene, is required by both the endothelial and hematopoietic lineages. De- velopment 121:3141-3150.

Sterio, D.C. (1984) The unbiased estimation of number and sizes of arbitrary particles using the dissector. J . Microsc. 134:127-136.

Wienecke, M.M., Thompson, R.P., and Germroth, P.G. (1995) Asym- metry of the inferior and superior atrioventricular endocardial cushions. In: “Developmental Mechanisms of Heart Disease,” Clark, E.B., Markwald, R.R., and Takao, A. (eds). New York: Fu- tura Publishing, pp. 309-314.

Wilson, D.I., Cross, I.E., Goodship, J.A., Brown, J., Scambler, P.J., Bain, H.H., Taylor, J.F., Walsh, K., Bankier, A,, Burn, J., et al. (1992) A prospective cytogenetic study of 36 cases of DiGeorge syndrome. Am. J. Human Genet. 51:957-963.

Endocardial cushion development and heart loop architecture in the trisomy 16 mouse

Documents

Transcript of Endocardial cushion development and heart loop architecture in the trisomy 16 mouse