Characterization of the cardiac phenotype in ... - Wiley Online Library

DEVELOPMENTAL DYNAMICS 237:426 – 435, 2008

PATTERNS & PHENOTYPES

Characterization of the Cardiac Phenotype in Neonatal Ts65Dn Mice Austin D. Williams,1 Corey H. Mjaatvedt,2 and Clara S. Moore1*

The Ts65Dn mouse is the most-studied of murine models for Down syndrome. Homology between the triplicated murine genes and those on human chromosome 21 correlates with shared anomalies of Ts65Dn mice and Down syndrome patients, including congenital heart defects. Lethality is associated with inheritance of the T65Dn chromosome, and anomalies such as right aortic arch with Kommerell’s diverticulum and interrupted aortic arch were found in trisomic neonates. The incidence of gross vascular abnormalities was 17% in the trisomic population. Histological analyses revealed interventricular septal defects and broad foramen ovale, while immunohistochemistry showed abnormal muscle composition in the cardiac valves of trisomic neonates. These findings confirm that the gene imbalance present in Ts65Dn disrupts crucial pathways during cardiac development. The candidate genes for congenital heart defects that are among the 104 triplicated genes in Ts65Dn mice are, therefore, implicated in the dysregulation of normal cardiogenic pathways in this model. Developmental Dynamics 237:426 – 435, 2008. © 2007 Wiley-Liss, Inc. Key words: development; organogenesis; transgenic approaches; gene expression; morphogenetic processes Accepted 25 October 2007

INTRODUCTION Down syndrome (DS) is a complex genetic disorder caused by the triplication of human chromosome 21 (Hsa 21) and has been estimated to occur in 1 out of 733 live births in all ethnic groups (Centers for Disease Control and Prevention, 2006). The neurological and physical abnormalities typically observed in DS patients vary not only in prevalence but also in severity. It was not until 1959 when Lejeune attributed these phenotypes to trisomy 21 that clinicians and researchers began to understand the genetic basis for DS developmental abnormalities (Lejeune et al., 1959). DS patients have recognizable features due to consistent

craniofacial alterations in all affected individuals: flattened nasal bones and persistent epicanthic folds, as well as smaller brains with decreased neuronal density (Keeling et al., 1997, Pinter et al., 2001). Other components of bone and cartilage are variably affected (Keeling et al., 1997), while congenital heart defects occur in approximately 50% of DS neonates (Freeman et al., 1998). Defects include both intracardiac anomalies and abnormal patterning of vasculature. Intracardiac anomalies range from the most common atrioventricular septal defects (Haak et al., 2002) to the less common abnormal tricuspid valvular function (Faiola et al., 2005). Com-

1

mon vascular anomalies include patent ductus arteriosus and tetralogy of Fallot (Spahis and Wilson, 1999) and aberrant right subclavian artery (Chaoui et al., 2005). The range of phenotypes in DS individuals indicates that gene dosage errors alter developmental pathways in the genesis of disparate organ systems, including the heart and vasculature. Several murine models have been developed to simulate the genotypephenotype correlation in DS. Exact DS mouse models are difficult to create because mouse orthologs to Hsa 21 genes are dispersed among Mus musculus chromosomes (Mmu) 10, 16 and 17 (Reeves et al., 2001). Gene mapping studies that identified syntenic

Biology Department, Franklin and Marshall College, Lancaster, Pennsylvania Department of Cell Biology and Anatomy, Medical University of South Carolina, Charleston, South Carolina Grant sponsor: National Heart, Lung, and Blood Institute; Grant numbers: R15 HL081099 and R01 HL66231. *Correspondence to: Clara S. Moore, P.O. Box 3003, Biology Dept., Franklin and Marshall College, Lancaster, PA 17604-3003. E-mail: [email protected] 2

DOI 10.1002/dvdy.21416 Published online 26 December 2007 in Wiley InterScience (www.interscience.wiley.com).

© 2007 Wiley-Liss, Inc.

CARDIAC PHENOTYPES IN TS65DN MICE 427

regions between the mouse and human genomes prompted many models with triplication of distal Mmu 16, the chromosome with the greatest region syntenic to Hsa 21, to be created. The results have been genetically and phenotypically variable. Several murine DS models require elaborate mating schemes to obtain progeny because of embryonic lethality, others have variable mosaicism, and the maintenance of germline trisomy is difficult in others (Moore and Roper, 2007). The most commonly used of these models is the Ts(1716)65Dn mouse (abbreviated as Ts65Dn) created at The Jackson Laboratories in 1990 (Davisson et al., 1990). In Ts65Dn, the distal region of Mmu 16 is attached to centromeric Mmu 17, effectively triplicating orthologs to 104 of 231 coding genes (Olson et al., 2004) or 136 of a total 364 possible reading frames on Hsa 21 (Gardiner et al., 2003) in mice inheriting the T65Dn marker chromosome. Many studies have strongly correlated Ts65Dn phenotypes with the features observed in DS individuals: immunological alterations (Paz-Miguel et al., 1999), cerebellar abnormalities (Baxter et al., 2000; Lorenzi and Reeves, 2006), impairments in learning and memory (Reeves et al., 1995; Galdzicki and Siarey, 2003), and differences in craniofacial skeleton structure (Richtsmeier et al., 2000). Cardiovascular anomalies in Ts65Dn neonates were first reported in a screening study in which right aortic arch was accompanied by septal defects in 8.3% of Ts65Dn cadavers suffering neonatal mortality (Moore, 2006). The intracardiac anomalies are analogous to congenital heart defects typical of DS patients, 80% of which are interruptions in the atrioventricular or ventricular septum (Freeman et al., 1998). Aortic arch anomalies, however, are often associated with the Hsa 22 genes from the DiGeorge region (Marino et al., 1999), orthologs of which are not triplicated in this model. Other recently created DS models demonstrate cardiac abnormalities during development. A transchromosomic mouse (Tc1) inheriting nearly all of Hsa 21 displays high rates of interventricular septal defects as septation occurs at embryonic day (E) 14.5 (O’Doherty et al., 2005). A mouse

model that produces a triplication of the entire Hsa 21 homologous segment of Mmu16 has recently been reported (Li et al., 2007). These mice inherit a third copy of an additional 12 known Hsa 21 orthologs not present on T65Dn. At E 18.5, the Dp(16)1Yu/⫹ embryos recapitulate several DS phenotypes, such as annular pancreas and cardiovascular anomalies in 37% of the embryos. The complex steps in cardiogenesis are among the most important in mammalian development. Primitive structures are required to cooperatively supply blood to a developing embryo while simultaneously completing differentiation into mature cardiac structures. The primitive heart tube folds and becomes septated by formation and differentiation of the endocardial cushions, a region of thickened cardiac jelly and mesenchymal cells. The endocardial cushions in the atrioventricular region fuse, effectively separating the open canal into the left and right sides. The endocardial cushions serve as primitive valves, and eventually elongate to form valve leaflets (Armstrong and Bischoff, 2004). Septation is not completed until the spiral aortico-pulmonary septum meets the atrioventricular endocardial cushions and muscular interventricular septum. The cushions form the membranous part of the interventricular septum, and complete the closure between the left and right ventricles. In the atria, septation begins as a crescent-shaped endocardial fold grows from the cranial-dorsal region of the presumptive atria. This curtain-like septum, the septum primum, fuses with the endocardial cushions in the atrioventricular region. The resulting hole at the septum-cushion interface is the ostium primum. As the septum continues to proliferate and the ostium primum disappears, apoptosis in the upper part of the septum primum forms another opening called the ostium secundum. A second crescentshaped septum, the septum secundum, forms to the right of the septum primum. The ovular opening formed by the two septa, the foramen ovale, provides unique interatrial communication in fetal circulation in order to bypass the nonfunctional lungs. As intracardial fluid flow dynamics change after birth, the septum secundum

fuses with the septum primum, and blood flow between atria is blocked. Abnormalities in formation or closure of the foramen ovale constitute interatrial septal defects. The lack of appropriate migration, fusion, and differentiation of endocardial cushions defines the pathogenesis of a variety of valvular and septal defects (Gittenberger-de Groot et al., 2005). A number of genetic factors, as well as gene overexpression like that of DS, can disrupt normal cardiac development (Marino and Diglio, 2000). There is an increased mortality rate among infants born with cardiovascular malformations (Tanner et al., 2005), and in a population-based study congenital heart defects were found to be the main cause of death in DS children less then ten years of age (Yang et al., 2002). Surgical treatment greatly increased the rate of survival of those with congenital heart defects (Yang et al., 2002), so the predisposition to fatal cardiovascular anomalies is an important consideration in evaluating survival and prognosis of DS individuals. Ts65Dn is a useful and informative murine model for DS. The well-defined region of triplication makes genotype-phenotype comparison between DS patients and Ts65Dn mice worthwhile in identifying candidate genes, or genes likely to contribute to the observed phenotypes. Analysis of DS phenotypes that are recapitulated in Ts65Dn mice, as well as those that are not present, is crucial in comparing results to gene mapping studies in DS (Reeves et al., 2001), as well as phenotypes seen in Dp(16)1Yu/⫹ (Li et al., 2007) and other murine models. Specifically, candidate genes DSCR1 (Rothermel et al., 2000) and DSCAM (Barlow et al., 2001) have been proposed to contribute to cardiac abnormalities in DS when overexpression occurs. These genes are triplicated in both Ts65Dn and Dp(16)1Yu/⫹ mice (Li et al., 2007), and the correlation of cardiac phenotypes would support the hypothesis of the genes’ roles in the genesis of cardiac anomalies. Dosage imbalance results in altered gene expression levels in many different tissues in a DS mouse model like Ts65Dn (Kahlem et al., 2004). An abnormal copy number of Hsa 21 genes DSCR1 and DYRK1a has been shown to mis-

428 WILLIAMS ET AL.

regulate the NFATc transcription factor (Arron et al., 2006). Thus, developmental defects may be due to upregulating and downregulating key signaling molecules, genes on Hsa 21 and those with which the triplicated genes interact. The current study adds to a previous anatomical screening study (Moore, 2006) with gross and histological characterization of the cardiovascular phenotypes of Ts65Dn neonates. We expected to detect a measurable rate of cardiovascular defects in the Ts65Dn model because of anomalies described in Down syndrome individuals and other Down syndrome mouse models. We have also collected genotypic data and analyzed the postnatal and potential prenatal loss observed in the Ts65Dn colony. Here we present a comprehensive analysis of the cardiovascular phenotypes observed in multiple complete litters from our Ts65Dn colony. We expand upon analysis of gross anatomy with histological and immunohistochemical staining for the detection of intracardiac defects at the molecular level. We present evidence that trisomy-related dysregulation produces cardiovascular defects and lethality in this Down syndrome model.

RESULTS Loss of Ts65Dn Progeny Within the Ts65Dn colony, the survival of progeny born to trisomic mothers is limited (Table 1). In the colony at large, 41% of litters demonstrate complete loss of all pups due to maternal neglect or failure to feed, regardless of whether the pups were T65Dn carriers, and all progeny survived in only 26% of litters. In the remaining 33% of the litters, there was a partial loss of some, but not all, progeny. The transmission rate of the T65Dn chromosome at birth in litters in which partial loss was noted was 49.7%, not significantly different than expected Mendelian ratios (n ⫽ 189, ␹2 ⫽ 0.005, P ⫽ 0.942). However, at weaning, the transmission rate dropped to 36.2% in these litters, which is significantly different than the expected 50% (n ⫽ 116, ␹2 ⫽ 8.828, P ⫽ 0.003). The same trend of selective loss is noted when analyzing transmission

TABLE 1. T65Dn Marker Chromosome Transmission Rate in Offspring of Trisomic Mothersa Number of Litter survival

Pups

Litters

Percent Ts65Dn

No loss Selective loss Birth Weaning Complete loss

124

28

36.3*

189 116 170

36 36 38

49.7 36.2* 45.9

Overall Birth Weaning

491 241

110 65

44.7* 34.0**

a

Litters were categorized as experiencing no loss, selective loss, or complete loss before weaning and genotype data from litters were pooled. Eight singleton litters that were excluded from the categorical analysis were included in the overall calculation. Pearson’s chi-squared tests were used to test the null hypothesis that there is no difference between measured transmission rates and expected Mendelian ratios. *P ⬍ 0.05; **P ⬍ 0.001.

rate data from the population collectively. Trisomic pups represented 44.7% of the population at birth, but only 34.0% at weaning. Yet the transmission rate was significantly different from 50% at birth (n ⫽ 491, ␹2 ⫽ 5.721, P ⫽ 0.017) and weaning (n ⫽ 241, ␹2 ⫽ 18.627, P ⬍ 0.001). Three neonates out of 104 collected during this study were identified as stillborn based on the lack of removal from extraembryonic tissue and the identification of uninflated lungs upon dissection. All three pups were confirmed to be T65Dn carriers. Finally, excluding litters resulting in complete loss, 37.4% of trisomic pups (n ⫽ 139) and 12.1% of euploid pups (n ⫽ 174) died within 48 hr of birth.

Cardiovascular Structure Cages with trisomic dams mated to euploid studs were monitored to isolate neonatal cadavers. Non-necrotic and non-cannabalized samples were dissected, blood samples collected, and the thoracic vasculature visualized. In 52 neonates for which genotyping identified the presence of trisomy, 15.3% showed some type of cardiovascular abnormality, including right aortic arch with Kommerell’s diverticulum, persistent truncus arteriosus (Fig. 1B), right aortic arch with right-sided ductus arteriosus (Fig. 1C), and aberrant right subclavian artery. Right aortic arch and aberrant

right subclavian were also identified in several Ts65Dn mice that survived to adulthood (data not shown). To determine the incidence of cardiovascular defects in the trisomic population, rather than only in dying neonates, a subset of seven litters of pups (total n ⫽ 38) was selected for complete histological analysis. The T65Dn transmission rate ranged from 25 to 55% in these litters, with an overall transmission rate of 47.4% in the subpopulation. In the 7 complete litters analyzed (18 trisomic and 20 euploid neonates), we found aortic arch defects in 17% of the trisomic pups. Two neonates had right-sided aortic arch (Fig. 1D) and another had interrupted aortic arch Type B. The aortic arch interruption was between the left carotid and the left subclavian arteries (Fig. 1E). In this phenotype, the aorta gives rise to the left and right common carotid and right subclavian arteries, while the dorsal aorta is connected to the pulmonary trunk by an arterial duct that forms the aortic arch and supplies the left subclavian artery. The right aortic arches of the two trisomic neonates were accompanied by Kommerell’s diverticulum, which is apparent with a dorsal view of the isolated heart and lungs (Fig. 1F). In this vascular pattern, the left-sided ductus arteriosus, branching from the pulmonary trunk, meets the right aortic arch dorsally, forming a vascular ring around the


Fig. 1. Aortic arch anomalies in Ts65Dn neonates. Frontal views of (A) the normal thoracic vessels and left aortic arch of a euploid neonate and (B) persistent truncus arteriosus (PTA, white arrowhead), (C) abnormal right aortic arch with right-sided ductus arteriosus, (D) right aortic arch with left-sided ductus, and (E) interrupted aortic arch in trisomic neonates. Dorsal view (F) of trisomic heart and lungs from D with the descending right aortic arch and Kommerell’s diverticulum (KD, white arrowhead) attached to the dorsal aorta. All scale bars ⫽ 1 mm. AoA, aortic arch; DAo, dorsal aorta; LA, left atrium; LCC, left common carotid artery; LSC, left subclavian artery; PT, pulmonary trunk; RA, right atrium; RCC, right common carotid artery; RSC, right subclavian artery.

thoracic region of the esophagus and trachea. Additionally, the left subclavian artery rises from the dorsal portion of the ductus arteriosus. Though the ventral ductus arteriosus closes normally after birth, the dorsal portion of the vessel, Kommerell’s diverticulum, remains open in order to maintain blood flow from the dorsal aorta to the aberrant left subclavian artery. The dorsal aortas of these pups descended caudally to the left of the vertebrae. One pup with aortic arch anomalies was stillborn and the other two died of natural causes within 48 hr of birth. Histological analyses of all neonates within the subset of seven litters identified intracardiac defects in the atrial and ventricular septa. Measurements of foramen ovale area for all hearts collected (except three that could not be analyzed due to the non-frontal an-

gles at which they were embedded) demonstrated abnormalities in the interatrial septa of the trisomics. Trisomic pups had a significantly larger foramen ovale than euploid pups (F1,20 ⫽ 13.785, P ⫽ 0.001). Significance is maintained in further analysis of foramen ovale area in litters from which all pups were euthanized (F1,12 ⫽ 4.875, P ⫽ 0.047). Interestingly, the foramen ovale of one trisomic neonate with interrupted aortic arch was three times larger than the mean of the population (Fig. 2). Interventricular septal defects have been observed in other Ts65Dn neonates; therefore, serial heart sections from the progeny of the seven selected litters were examined for interventricular septal defects with light microscopy. The interventricular septum of one neonate with interrupted aortic arch had defects in the aortic outflow

Fig. 2. Boxplot of foramen ovale area. Foramen ovale (FO) area was measured in euploid and Ts65Dn neonatal hearts isolated within 48 hr of birth. FO area was significantly greater in trisomic pups that in euploid (F1,20 ⫽ 13.785, P ⫽ 0.001). One obvious outlier was a trisomic pup with interrupted aortic arch (asterisk).

region (Fig. 3A) and more dorsally in the muscular ventricular region (Fig. 3B) that allowed communication with the right ventricle.

430 WILLIAMS ET AL.

An immunohistological study of neonates from the initial screen (Moore, 2006) revealed that a trisomic pup with a right aortic arch and an interventricular septal defect had abnormal deposition of muscle elements in the tricuspid valves (Fig. 3D,F). Analysis of the subpopulation of neonates demonstrated that multiple trisomic pups demonstrate intracardiac anomalies in protein localization. Immunohistological studies in nine Ts65Dn and six euploid pups show regions of abnormal sarcomeric proteins in the hearts of two trisomic neonates (Fig. 3E), specifically in the tips of tricuspid valves. Alpha-sarcomeric myosin-positive cells were present at the valvepapillary connection in both trisomic and euploid hearts; however, staining was present in a greater area of the valve leaflets in Ts65Dn hearts.

DISCUSSION In this study, we confirmed selective loss of Ts65Dn neonates. Previous work has shown that half of all embryos from trisomic mothers tested midway through gestation carried the T65Dn marker chromosome, but the transmission rate is slightly reduced by birth and continues to drop to 36% by weaning (Roper et al., 2006). The 20% rate of transmission reported by The Jackson Laboratory (Liu et al., 2003) also differs from the expected 1:1 ratio based on random segregation of the marker chromosome in female meiotic events. These reduced rates indicate selective loss of trisomic neonates as previously reported (Moore, 2006). Following genotyping of the pups that died postnatally and their weaned sibs, it is clear that most selective loss is a function of genotype. In litters with selective loss, our results parallel those of others (Roper et al., 2006): half of the neonates are T65Dn carriers at birth, but selective loss of trisomic progeny decreases the percent of trisomic weanlings to almost one third (Table 1). These results suggest that inheritance of the T65Dn marker chromosome predisposes neonates to phenotypes resulting in mortality within 48 hr of birth. In contrast, in litters in which there was no postnatal lethality, the transmission rate was significantly less than 50%, suggesting prenatal loss of trisomic fe-

Fig. 3. Histology and anti-muscle immunoreactivity in neonatal heart tissue. Hematoxylin and eosin staining of trisomic neonatal heart sections indicate interventricular septal defects (white arrowheads) in (A) the subarterial segment of the left ventricle and (B) the more dorsal and caudal muscular region. Fluorescent immunohistochemistry on frontal sections of (C) euploid and (D–F) Ts65Dn neonatal hearts sectioned at 8 ␮m with anti-sarcomeric myosin (MF20). C: Normal wild type heart in which cardiac muscle (green) is detected in the ventricular walls and interventricular septum, counterstained with propidium iodide (red). D: Heart of Ts65Dn mouse with right aortic arch and interventricular septal defect (white arrowhead) and muscularized valves (green). F: Boxed area in D is magnified to show immunoreactivity (white arrowheads) in mural leaflet (ML) of the tricuspid valve of trisomic neonate. E: Tricuspid valve of trisomic neonate without gross defects in which muscle is detected at the tip of the leaflet (white arrowheads) visualized using epifluorescent microscopy. Detection of muscle in adjacent regions (black arrowheads) indicates areas of connection to papillary muscle. Scale bars in A and B ⫽ 0.1 mm. IVS, interventricular septum; LA, left atrium; LV, left ventricle; ML, mural leaflet; RA, right atrium; RV, right ventricle; SL, septal leaflet; VW, ventricular wall.


tuses. The litters that result in complete loss due to maternal neglect have an intermediate rate of transmission. This group presumably is composed of litters that would have resulted in no loss or selective loss if given appropriate maternal care. We hypothesize that trisomic fetuses with the most severely compromising phenotypes die before birth, resulting in litters with a significantly decreased rate of T65Dn chromosome transmission and no postnatal lethality. Moreover, trisomic fetuses with less serious phenotypes survive until after birth, but die of trisomy-related problems shortly after birth. This hypothesis matches the selective loss of trisomic neonates we observe by postnatal day 2. Continuing to measure the transmission rate in embryological work at different stages of development may elucidate at which developmental stages Ts65Dn embryos are being lost. Assessment of cardiovascular phenotypes and their severity may also contribute to the understanding of why some trisomic progeny die prenatally and others postnatally. We have found that the trisomic dams in our colony are poor caretakers. Nearly half of all dams neglect their first litter, resulting in complete loss (n ⫽ 51 mothers). Additionally, the umbilical cord and extraembryonic tissues of several living pups were still attached upon collection, regardless of genotype, indicating that mothers are not selectively abandoning trisomic progeny. The failure of some trisomic pups to feed, and identification of air within the stomach of other trisomic pups, suggests that these neonates have gastrointestinal complications or difficulty nursing. These findings correlate to problems seen in DS infants. Mothers of 57% of DS infants report that their infants had feeding problems (Yang et al., 2002), and congenital intestinal obstructions are more prevalent among DS patients than the population-at-large (Delebar et al., 2006). Gastrointestinal defects such as annular pancreas and malrotation of the intestine have been identified in Dp(16)1Yu/⫹ embryos (Li et al., 2007), but not in Ts65Dn mice. The defects we have described in the great vessels (Fig. 1) are reminiscent of the defects seen in DS. The right subclavian artery arises directly

from the dorsal aorta as the fourth cephalic vessel in about one third of DS fetuses (Chaoui et al., 2005). Several other vascular anomalies have been identified in DS patients less than two years of age, with patent ductus arteriosus, or the failure of the neonatal vessel to close after birth, present in 8% and tetralogy of Fallot in 3% of the children (Yang et al., 2002). Identification of aberrant right subclavian artery and patent truncus arteriosus due to failure of outflow tract septation are similar to defects seen in DS. Although some similarities do exist, the cardiac phenotypes of DS patients and Ts65Dn mice differ a great deal. Specifically, interrupted aortic arch (Type B) and the vascular ring around the trachea and esophagus of multiple trisomic neonates with aortic arch anomalies are not phenotypes typically associated with DS. In both DS patients and Ts65Dn mice, however, we hypothesize that disruption of cardiogenic pathways is caused by misregulation of the genes on Hsa 21 (or orthologs on Mmu16) and the genes with which they interact. The value in comparing the phenotypes of this mouse model to DS is in the fact that not all Hsa 21 genes are triplicated in Ts65Dn. By comparing the similarities and differences in the genotype-phenotype correlation, we may be led to a more thorough understanding of the regulation of heart development, and the specific genes on Hsa 21 involved in different aspects of cardiogenesis. We also hypothesize that the dysmorphology seen in the vasculature of Ts65Dn neonates arises in the complex remodeling of the aortic arch system. Two mechanisms may contribute to the genesis of the anomalies. First, several studies have described that altered levels of apoptosis in the pharyngeal arch arteries result in aortic arch anomalies (Waller et al., 2000; Molin et al., 2002). We have found a reduced level of apoptosis in the atrioventricular and outflow tract regions of Ts65Dn embryos (unpublished data), suggesting that cardiovascular anomalies arise due to abnormal cell death. A second potential mechanism lies in Buselmaier’s “arch-rival” hypothesis (Buselmaier et al., 1991), in which changes in intracardiac fluid flow dynamics due to septal defects

causes symmetrical pharyngeal arch arteries to compete for priority until the inappropriate arch regresses. As previously described in chick development (Hogers et al., 1997; Broekhuizen et al., 1999), changes in blood flow caused by altering extracardiac fluid flow, which necessarily changes intracardiac fluid dynamics, disrupt aortic arch development. Thus, embryos with intracardiac defects would exhibit changes in intracardiac hemodynamics, potentially leading to abnormal patterning of the aortic arch. Of the nearly half of DS patients with congenital heart defects, 80% of these individuals present with interruptions in the atrioventricular or ventricular septum (Freeman et al., 1998). Ventricular septal defects, such as those shown in Figure 3, are easy to identify histologically by visually scanning serial sections of the heart. Atrial septal defects, however, are more difficult to identify in neonates because of the normal presence of the foramen ovale. We see a significant difference in foramen ovale area when comparing trisomic to euploid neonates (Fig. 2). These findings lead to two hypotheses: (1) trisomic neonates die before their euploid sibs, allowing more time for the closing of foramen ovales of euploid pups after birth; and (2) trisomic neonates exhibit developmental delays when compared to their euploid sibs. Subsequent analysis of euthanized litters, in which neonates are presumably age-matched, maintains significance in foramen ovale area. This finding supports hypothesis two. However, there is attrition in the trend, which lends support to the first hypothesis. The broad foramen ovales of pups with aortic arch anomalies indicate the presence of atrial septal defects, supporting previous findings of septal defects in conjunction with aortic arch anomalies in Ts65Dn cadavers (Moore, 2006). These findings also support the “arch-rival” hypothesis (Buselmaier et al., 1991). Differences in blood flow in the atrial region, and across the interventricular septum in one case, potentially affect the maintenance and regression of key pharyngeal arch arteries. We have identified regions of inappropriate valvular muscle composition in the hearts of trisomic neonates with and without gross vascular defects due

432 WILLIAMS ET AL.

to altered endocardial cushion differentiation. The tricuspid valves in multiple trisomic pups in this study showed abnormal expression of sarcomeric components. These findings indicate that Ts65Dn progeny may have valvular dysfunction due to protein misexpression that is identifiable by anomalies at the molecular level, but not upon gross examination. These cardiac insufficiencies may contribute to the early and selective mortality of Ts65Dn progeny. Trisomy could predispose fetuses to inappropriate migration or proliferation, altered apoptosis, or disrupted differentiation of endocardial cushion cells during cardiogenesis. Defects in heart morphogenesis have been observed with an increase or a reduction of cushion mesenchyme in TGF␤2 null (Bartram et al., 2001) and trisomy 16 mice (Webb et al., 1999; Waller et al., 2000), respectively. We hypothesize a mechanism whereby decreased apoptosis in the atrioventricular region leads to a population of cells being inappropriately maintained in this key region during cardiogenesis. Thus, a lack of normal apoptosis (Sharma et al., 2004) and subsequent differentiation of cells may also affect the leaflets of the valves. The cells that normally undergo apoptosis do not have a well-defined fate, and may differentiate into anomalous cell types, including muscle. Altered apoptosis and myocardialization during differentiation of endocardial cushions has been demonstrated in TGF-␤2 knockout mice (Bartram et al., 2001). An abnormal increase of cushion cell fate into the myocardial lineage may reduce the critical mass of cushion mesenchyme needed to properly form or maintain valve leaflets resulting in a structural or functional defect. Disruptions in other cellular signaling pathways and in extracellular matrix remodeling could also contribute to the abnormal differentiation resulting in inappropriate myoprotein expression in the atrioventricular cushion mesenchyme. Although other mouse models also show valvular abnormalities, the mechanisms by which these defects arise have not yet been elucidated. Cleft mitral valves were noted in Dp(16)1Yu/⫹ embryos, which inherit a triplication of the 116 known Hsa 21 orthologs (Li et al., 2007). Failure of

elongation of the valve leaflet was seen in earlier embryonic stages in mice with disruptions in NFATc signaling due to overexpression of the Hsa 21 genes DSCR1 and DYRK1a (Arron et al., 2006). These genes are triplicated in Ts65Dn and Dp(16)1Yu/⫹ mice, thus the NFATc pathway could be important in the valvular defects seen in these DS models. However, these two genes are also triplicated in other segmental trisomy 16 mice, Ts1Cje (Sago et al., 1998) and Ts2Cje (Villar et al., 2005), which reportedly suffer neither perinatal lethality nor cardiovascular defects. Therefore, in addition to NFATc pathway disruptions, an extra copy of genes proximal to the trisomic region of these mice (proximal to SOD1) predisposes mouse embryos to the valvular, outflow tract, and aortic arch defects observed in Dp(16)1Yu/⫹ mice and described here in Ts65Dn. Other candidate genes triplicated in Ts65Dn and Dp(16)1Yu/⫹ mice, but not in Ts1Cje and Ts2Cje include ADAMTS-1 and ADAMTS-5. These proteases are involved in extracellular matrix molecule degradation remodeling, including cleavage of the proteoglycan versican variants (Sandy et al., 2001; Russell et al., 2003). Versican is required for normal pre-valvular leaflet and septa formation in the mouse and alterations in versican expression can alter the myocardial phenotype in developing hearts (Yamamura et al., 1997; Kern et al., 2006, 2007). With the variability in cardiovascular phenotypes seen in the Ts65Dn mouse, it is likely that elevated expression of several of the candidate genes disrupting multiple pathways in differentiation is requisite to production of a detectable phenotype.

rate transmission rate and incidence data. The incidence of cardiac anomalies in Ts65Dn mice is greater than in some segmental trisomy models, but less than the incidence reported for DS and in other trisomic DS models (Moore and Roper, 2007; Delabar et al., 2006). It is intriguing that the Dp(16)1Yu/⫹ model, with only 12 more Hsa 21 orthologs triplicated than in our model, has double the incidence of defects (Li et al., 2007), suggesting genes just outside the region triplicated in Ts65Dn may greatly increase the incidence and severity of cardiac phenotypes. The Dp(16)1Yu/⫹ incidence data are based on late embryonic stages, while our results focus on neonates. There is a possibility, supported by our transmission data, that some prenatal loss of affected trisomics occurs in Ts65Dn pregnancies, effectively reducing the incidence of defects at birth compared to embryonic stages. To understand the cardiovascular anomalies of this DS mouse model requires extensive embryological analysis with key stages in murine development to examine the spatial and temporal dynamics of cardiogenesis. In order to gain a complete understanding of the origin of the cardiovascular defects described here, anatomical and molecular studies of vessel and septal formation must be employed. The identification of additional candidate genes and regions of protein misexpression will add another dimension to this understanding. By characterizing the vascular and intracardiac defects in Ts65Dn neonates and continuing with embryological work, we will contribute meaningfully to the understanding of gene imbalance and its impact on cardiac development.

Conclusions

EXPERIMENTAL PROCEDURES

This comprehensive study describes both vascular and intracardiac defects in 17% of Ts65Dn neonates. The strengths of this study include the use of complete and euthanized litters. This allowed us to prepare heart tissue for histology such that samples were more easily contrasted with one another. We were also able to make comparisons of age-matched pups for foramen ovale measurements. Finally, the use of complete litters assured complete and accu-

Animals Ts65Dn mice were developed at The Jackson Laboratory (Bar Harbor, ME) by an induced Robertsonian translocation of Mmu16C3.3-4 to the centromere of Mmu17A2 (Davisson et al., 1990). B6EiC3Sn a/A-Ts(1716)65Dn founder female mice were mated to B6Ei ⫻ C3Sn F1 male mice to establish a breeding colony. Trisomic fe-


males produced in the colony were mated to F1 euploid males to maintain the inbred backcross necessary for maintenance of T65Dn marker chromosome transmission. Surviving pups were weaned at 3 to 4 weeks of age. The animals were housed with a 12-hr light/dark cycle, controlled temperature and humidity, and water and food available ad libitum. All animal care protocols were approved by the Franklin and Marshall College Institutional Animal Care and Use Committee. For histological analysis, seven complete litters of pups (total n ⫽ 38) were collected from the Ts65Dn colony and analyzed. The pups of three of the litters all died within 48 hr of birth due to maternal neglect; one euploid pup had undergone partial parental cannibalization, so the heart was not recovered. The pups of four litters were collected and euthanized within 24 hr of birth.

Genotyping Mice were genotyped using interphase fluorescence in situ hybridization on blood cells as previously described (Moore et al., 1999). Modifications include the use of a Hybaid OmniSlide Thermal Cycler for denaturing and probe annealing, followed by one wash in 2⫻ SSC at 68°C.

Transmission Rate We inspected mouse cages at least twice daily to prevent parental cannibalization after neonatal death. Data from litters from which all pups and weanlings were genotyped were pooled at birth (n ⫽ 110 litters, 491 pups) and at weaning (n ⫽ 65 litters, 241 weanlings) and the T65Dn marker chromosome transmission rate was calculated. We categorized litters into three groups according to survival patterns: litters in which there was no postnatal loss, litters in which all pups died prior to weaning, and litters in which there was partial or selective loss prior to weaning. We excluded singleton litters, which could not be categorized as selective loss, from categorical analysis. Data were analyzed to determine if transmission frequencies differed from Mendelian ratios based on normal segregation of maternal chromosomes during meio-

sis. Pearson’s chi-square tests were performed on genotype data to test the null hypothesis that there was no difference between measured transmission rates and expected Mendelian ratios.

sured the section in which the opening was at its maximum width. Based on the elliptical shape of the opening, foramen ovale area was calculated using the formula A ⫽ ␲ab ⫽ ␲

Dissection Living pups were euthanized with gaseous 2-bromo-2-chloro-1,1,1-trifluoroethane (halothane; Sigma) exposure for 20 min. Using watchmaker’s forceps, we opened the thoracic cavity and removed the ribcage and thymus. The great vessels and heart were examined for anomalies and imaged. We flushed the hearts with 1⫻ PBS (Sigma) and cardiac relaxation buffer (0.1 mM EGTA, 30 mM KCl, 0.2 mM verapamil, and 10 U/mL heparin in Hank’s balanced salt solution; Cambrex Bio Science, Inc.) in order to clear the chambers of blood for histological preps. After morphological analysis, the heart and lungs were removed simultaneously, fixed in freshly prepared PBS-buffered 3% paraformaldehyde for 2 hr, and stored in 1⫻ PBS until preparation for histology.

Histology Isolated and fixed organs were dehydrated through an ethanol series into xylene and embedded in Paraplast XTRA tissue embedding medium (Fisher Scientific) with the ventral surface of the heart down. Embedded organs were stored at 4°C until use. The embedded hearts were serially sectioned at 8 ␮m with a rotary microtome. The frontal sections were dewaxed with xylene and rehydrated through an ethanol series in preparation for histological staining or fluorescent immunohistochemistry. For histology, tissues were stained in Harris hematoxylin solution (Protocol, Fisher Scientific) and counterstained in eosin (Protocol, Fisher) and mounted in Histomount X in xylenes (Accra Lab, Inc.).

Analysis of Foramen Ovale Area Slides of each heart were examined serially: the ventral and dorsal edges of the foramen ovale were noted. Using Image J freeware, we also mea-

冉

ventral ⫺ dorsal thickness 2 ⫻

冉

冊冊

width . 2

Comparisons of foramen ovale area were performed using one-way analysis of variance (ANOVA) to test the null hypotheses that there was no difference in foramen ovale area when comparing Ts65Dn and euploid pups.

Fluorescent Immunohistochemistry Primary anti-mouse monoclonal antisarcomeric myosin antibody (MF20) was visualized with a secondary ALEXA 488 goat anti-mouse IgG antibody (Molecular Probes). Rehydrated tissues were placed in metal racks and treated with Antigen Unmasking Solution (Vector) according to the manufacturer’s protocol. Slides were blocked [3% normal goat serum (Sigma), 1% bovine serum albumin (BSA, Sigma) in 1⫻ PBS] for 1 hr, incubated overnight at 4°C with primary antibody [(1:200) in blocking buffer, washed (1% BSA in PBS)], incubated in a humid chamber with secondary antibody (1:200) for 2 hr at room temperature. Slides were washed and mounted in fluorescent mounting medium (0.3% n-propyl gallate, 20% 4⫻ PBS, 80% glycerol) and viewed with confocal or epifluorescent microscopy. The MF20 antibody developed by D.A. Fischman was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained at the University of Iowa, Department of Biological Sciences, Iowa City, Iowa, 52242.

ACKNOWLEDGMENTS We thank Christine Kern for assistance with confocal microscopic imaging and Mark Olson for thoughtful discussions. We extend our gratitude to Regina Toto and David Janerich for technical assistance in genotyping samples. This research was supported by National Heart, Lung, and Blood

434 WILLIAMS ET AL.

Institute R15 HL081099 to C.S.M. and R01 HL66231 to C.H.M., a Merck Undergraduate Research Fellowship, and the Franklin & Marshall College William and Lucille Hackman Scholars Program.

REFERENCES Armstrong EJ, Bischoff J. 2004. Heart valve development: endothelial cell signaling and differentiation. Circ Res 95: 459 –470. Arron JR, Winslow MM, Polleri A, Chang C, Wu H, Gao X, Neilson JR, Chen L, Heit JJ, Kim SK, Yamasaki N, Miyakawa T, Francke U, Graef IA, Crabtree GR. 2006. NFAT dysregulation by increased dosage of DSCR1 and DYRK1A on chromosome 21. Nature 441: 595–600. Barlow GM, Chen XN, Shi ZY, Lyons GE, Kurnit DM, Celle L, Spinner NB, Zackai E, Pettenati MJ, Van Riper AJ, Vekemans MJ, Mjaatvedt CH, Korenberg JR. 2001. Down syndrome congenital heart disease: a narrowed region and a candidate gene. Genet Med 3:91–101. Bartram U, Molin DG, Wisse LJ, Mohamad A, Sanford LP, Doetschman T, Speer CP, Poelmann RE, Gittenberger-de Groot AC. 2001. Double-outlet right ventricle and overriding tricuspid valve reflect disturbances of looping, myocardialization, endocardial cushion differentiation, and apoptosis in TGF-beta(2)-knockout mice. Circulation 103: 2745–2752. Baxter LL, Moran TH, Richtsmeier JT, Troncoso J, Reeves RH. 2000. Discovery and genetic localization of Down syndrome cerebellar phenotypes using the Ts65Dn mouse. Hum Mol Genet 9:195– 202. Broekhuizen MLA, Hogers B, DeRuiter MC, Poelmann RE, Gittenberger-de Groot AC, Wladimiroff JW. 1999. Altered hemodynamics in chick embryos after extraembryonic venous obstruction. Ultrasound Obstet Gynecol 13:437–445. Buselmaier W, Bacchus C, Stertz H. 1991. Genesis and systematization of cardiovascular anomalies in murine trisomy 16. In: Epstein CJ, editor, The morphogenesis of Down syndrome. New York: Wiley-Liss, Inc. p. 203–214. Centers for Disease Control and Prevention. 2006. Improved National Prevalence Estimates for 18 selected Major Birth Defects: United States, 1999 –2001. Morb Mortal Wkly Rep 54:1301–1305. Chaoui R, Heling KS, Sarioglu N, Schwabe M, Dankof A, Bollmann R. 2005. Aberrant right subclavian artery as a new cardiac sign in second- and third trimester fetuses with Down syndrome. Am J Obstet Gynecol 192:257–263. Davisson MT, Schmidt C, Akeson EC. 1990. Segmental trisomy of murine chromosome 16: a new model system for studying Down syndrome. Prog Clin Biol Res 360:263–280.

Delebar JM, Aflalo-Rattenbach R, Creáu N. 2006. Developmental defects in trisomy 21 and mouse models. Sci World J 1:1945–1964. Faiola S, Tsoi E, Huggon IC, Allan LD, Nicolaides KH. 2005. Likelihood ratio for trisomy 21 in fetuses with tricuspid regurgitation at the 11 to 13⫹6-week scan. Ultrasound Obstet Gynecol 26:22–27. Freeman SB, Taft LF, Dooley KJ, Allran K, Sherman SL, Hassold TJ, Khoury MJ, Saker DM. 1998. Population-based study of congenital heart defects in Down syndrome. Am J Med Gen 80:213–217. Galdzicki Z, Siarey RJ. 2003. Understanding mental retardation in Down syndrome using trisomy 16 mouse models. Genes Brain Behavior 2:167–178. Gardiner K, Fortna A, Bechtel L, Davisson MT. 2003. Mouse models of Down syndrome: how useful can they be? Comparison of the gene content of human chromosome 21 with orthologous mouse genomic regions. Gene 318:137–47. Gittenberger-de Groot AC, Bartelings MM, Deruiter MC, Poelmann RE. 2005. Basics of cardiac development for the understanding of congenital heart malformations. Ped Res 57:169 –176. Haak MC, Bartelings MM, Gittenberger-de Groot AC, van Vugt JM. 2002. Cardiac malformations in first-trimester fetuses with increased nuchal translucency: ultrasound diagnosis and postmortem morphology. Ultrasound Obstet Gynecol 20:14 –21. Hogers B, DeRuiter MC, Gittenberger-de Groot AC, Poelmann RE. 1997. Unilateral vitelline vein ligation alters intracardiac blood flow patterns and morphogenesis in the chick embryo. Circulation Res 80:473–481. Kahlem P, Sultan M, Herwig R, Steinfath M, Balzereit D, Eppens B, Saran NG, Pletcher MT, South ST, Stetten G, Lehrach H, Reeves RH, Yaspo ML. 2004. Transcript level alterations reflect gene dosage effects across multiple tissues in a mouse model of Down syndrome. Genome Res 14:1258 –1267. Keeling JW, Hansen BF, Kjaer I. 1997. Pattern of malformations in the axial skeleton in human trisomy 21 fetuses. Am J Med Genet 68:466 –471. Kern CB, Twal WO, Mjaatvedt CH, Fairey SE, Toole BP, Iruela-Arispe ML, Argraves WS. 2006. Proteolytic cleavage of versican during cardiac cushion morphogenesis. Dev Dyn 235:2238 –2247. Kern CB, Norris RA, Thompson RP, Argraves WS, Fairey SE, Reyes L, Hoffman S, Markwald RR, Mjaatvedt CH. 2007. Versican proteolysis mediates myocardial regression during outflow tract development. Dev Dyn 236:671–683. Lejeune J, Gautier M, Turpin R. 1959. Etudes des chromosomes somatiques de neuf enfant mongoliens. [Studies of the somatic chromosomes of nine mongoloid infants.] CR Acad Sci 248:1721–1722. Li Z, Yu T, Morishima M, Pao A, LaDuca J, Conroy J, Nowak N, Matsui S, Shiraishi I, Yu YE. 2007. Duplication of the entire 22.9 Mb human chromosome 21 syntenic

region on mouse chromosome 16 causes cardiovascular and gastrointestinal abnormalities. Hum Mol Genet 16:1359 – 1366. Liu DP, Schmidt C, Billings T, Davisson MT. 2003. Quantitative PCR genotyping assay for the Ts65Dn mouse model of Down syndrome. Biotechniques 35:1170 – 1178. Lorenzi HA, Reeves RH. 2006. Hippocampal hypocellularity in the Ts65Dn mouse originates early in development. Brain Res 1104:153–159. Marino B, Digilio MC. 2000. Congenital heart disease and genetic syndromes: specific correlation between cardiac phenotype and genotype. Cardiovasc Pathol 9:303–15. Marino B, Digilio MC, Toscano A, Giannotti A, Dallapiccola B. 1999. Congenital heart defects in patients with DiGeorge/ velocardiofacial syndrome and del22q11. Genet Couns 10:25–33. Molin DG, DeRuiter MC, Wisse LJ, Azhar M, Doetschman T, Poelmann RE, Gittenberger-de Groot AC. 2002. Altered apoptosis pattern during pharyngeal arch artery remodeling is associated with aortic arch malformations in Tgfbeta2 knock-out mice. Cardiovasc Res 56:312– 322. Moore CS. 2006. Postnatal lethality and cardiac anomalies in the Ts65Dn Down Syndrome mouse model. Mamm Genome 17:1005–1012. Moore CS, Roper RJ. 2007. The power of comparative and developmental studies for mouse models of Down syndrome. Mamm Genome 18:431–443. Moore CS, Lee JS, Birren B, Stetten G, Baxter LL, Reeves RH. 1999. Integration of cytogenetic with recombinational and physical maps of mouse chromosome 16. Genomics 59:1–5. O’Doherty A, Ruf S, Mulligan C, Hildreth V, Errington ML, Cooke S, Sesay A, Modino S, Vanes L, Hernandez D, Linehan JM, Sharpe PT, Brandner S, Bliss TVP, Henderson DJ, Nizetic D, Tybulewicz VLJ, Fisher EMC. 2005. An aneuploid mouse strain carrying human chromosome 21 with Down syndrome phenotypes. Science 309:2033–2037. Olson LE, Richtsmeier JT, Leszl J, Reeves,RH. 2004. A chromosome 21 critical region does not cause specific Down syndrome phenotypes. Science 306:687– 690. Paz-Miguel JE, Flores R, Sańchez-Velasco P, Ocejo-Vinyals G, de Diego JE, de Rego JL, Leyva-Cobiań F. 1999. Reactive oxygen intermediates during programmed cell death induced in the thymus of the Ts(1716)65Dn mouse, a murine model for human Down syndrome. J Immun 163:5399 –5410. Pinter JD, Eliez S, Schitt JE, Capone GT, Reiss AL. 2001. Neuroanatomy of Down’s syndrome: a high-resolution MRI study. Am J Psychiatry 158:1659 –1665. Reeves RH, Irving NG, Mora TH, Wohn A, Kitt C, Sisodia SS, Schmidt C, Bronson RT, Davisson MT. 1995. A mouse model for Down syndrome exhibits learning


and behavior deficits. Nature Genet 11: 177–184. Reeves, RH, Baxter, LL, Richtsmeier, JT. 2001. Too much of a good thing: mechanisms of gene action in Down syndrome. Trends Genet 17:83–88. Richtsmeier JT, Baxter LL, Reeves RH. 2000. Parallels of craniofacial maldevelopment in Down syndrome and Ts65Dn mice. Dev Dyn 217:137–145. Roper RJ, St. John HK, Philip J, Reeves RH. 2006. Perinatal loss of Ts65Dn “Down syndrome” mice. Genetics 172: 437–443. Rothermel B, Vega RB, Yang J, Wu H, Bassel-Duby R, Williams RS. 2000. A protein encoded within the Down syndrome critical region is enriched in striated muscles and inhibits calcineurin signaling. J Biol Chem 275:8719 –8725. Russell DL, Doyle KM, Ochsner SA, Sandy JD, Richards JS. 2003. Processing and localization of ADAMTS-1 and proteolytic cleavage of versican during cumulus matrix expansion and ovulation. J Biol Chem 278:42330 –42339. Sago H, Carlson EJ, Smith DJ, Kilbridge J, Rubin EM, Mobley WC, Epstein CJ, and Huang TT. 1998. Ts1Cje, a partial tri-

somy 16 mouse model for Down syndrome, exhibits learning and behavioral abnormalities. Proc Natl Acad Sci USA 95:6256 –6261. Sandy JD, Westling J, Kenagy RD, IruelaArispe ML, Verscharen C, RodriguezMazaneque JC, Zimmermann DR, Lemire JM, Fischer JW, Wight TN, Clowes AW. 2001. Versican V1 proteolysis in human aorta in vivo occurs at the Glu441-Ala442 bond, a site that is cleaved by recombinant ADAMTS-1 and ADAMTS-4. J Biol Chem 276:13372– 13378. Sharma PR, Anderson RH, Copp AJ, Henderson DJ. 2004. Spationtemporal analysis of programmed cell death during mouse cardiac septation. Anat Rec 277A: 355–369. Spahis JK, Wilson GN. 1999. Down syndrome: perinatal complications and counseling experiences in 216 patients. Am J Med Genet 89:96 –99. Tanner K. Sabrine NM, Wren C. 2005. Cardiovascular malformations among preterm infants. Pediatrics 116: E833– E838. Villar AJ, Belichenko PV, Gillespie AM, Kozy HM, Mobley WC, and Epstein

CJ. 2005. Identification and characterization of a new Down syndrome model, Ts[Rb(12.1716)]2Cje, resulting from a spontaneous Robertsonian fusion between T(1716)65Dn and mouse chromosome 12. Mamm Genome 16:79 – 90. Waller BR, McQuinn T, Phelps AL, Markwald RR, Lo CW, Thompson RP, Wessels A. 2000. Conotruncal anomalies in the trisomy 16 mouse: an immunohistochemical analysis with emphasis on the involvement of the neural crest. Anat Rec 260:279 –293. Webb S, Anderson RH, Lamers WH, Brown NA. 1999. Mechanisms of deficient cardiac septation in the mouse with trisomy 16. Circ Res 84:897–905. Yamamura H, Zhang M, Markwald R, Mjaatvedt C. 1997. A heart segmental defect in the anterior-posterior axis of a transgenic mutant mouse. Dev Biol 186: 58 –72. Yang Q, Rasmussen SA, Friedman JM. 2002. Mortality associated with Down’s syndrome in the USA from 1983 to 1997: a population-based study. Lancet 359: 1019 –1025.