Cardiovascular Research 1998 39(1):155-164; doi:10.1016/S0008-6363(98)00037-6
© 1998 by European Society of Cardiology
Copyright © 1998, European Society of Cardiology
The mouse with trisomy 16 as a model of human hearts with common atrioventricular junction
Robert H Andersona,*,
Sandra Webbb and
Nigel A Brownb
aSection of Paediatrics, National Heart and Lung Institute, Royal Brompton Campus Imperial College School of Medicine, London SW3 6LY, UK
bDepartment of Anatomy and Developmental Biology, St. George's Hospital Medical School, London, UK
* Corresponding author. Tel.: +44 (171) 351 8751; Fax: +44 (171) 351 8230; E-mail: r.anderson@ic.ac.uk
Received 30 September 1997; accepted 12 January 1998
 |
Abstract
|
|---|
Objective: To establish if the mouse with trisomy 16 is a suitable
animal model with which to elucidate the development of a common
atrioventricular junction.
Methods: The junctional morphologies
in the normal human heart and those with a common atrioventricular
junction are compared and contrasted. These are then related
to observations made in normal mice and those with trisomy 16.
So as better to understand development, a full description is
given first of the normal atrioventricular junctions. Developmental
implications are discussed because failure of fusion of the
endocardial cushions cannot account for all the anomalies found
in RXR alpha knockout, and in iv/iv mice.
Results: Mice with
trisomy 16 showed evidence of deficiencies of atrioventicular
septation and possessed a common atrioventricular junction,
but the valvar orifices were not balanced between the ventricles
as is the case in humans. Whilst some mice showed affinities
with human tricuspid atresia, other cardiac malformations in
the mice had no counterparts in human cardiac pathology. In
humans both "partial" and "complete" forms of "atrioventricular
canal malformations" share a basically common muscular junctional
morphology, the differences being due exclusively to the way
the bridging leaflets are fused to each other and/or the septum.
Conclusions: It is simplistic to use the mouse with trisomy
16 as a model for cardiac abnormalities seen in humans. A spectrum
more comparable to humans is found in RXR knockout mice. Study
of the iv/iv mouse may help elucidate the genetic steps involved
in normal and abnormal atrioventricular septation.
KEYWORDS Experimental; Heart; Pathophysiology; Congenital defects; Histopathology; Morphogenesis; Trisomy 16 mouse; Embryology
|
1 Introduction
|
|---|
The mechanics underscoring the formation of hearts with common
atrioventricular junction have long fascinated both embryologists
and clinicians
[1]. It has often been assumed that the characteristic
malformation which lacks the normal atrioventricular septal
structures
[2]develops because of a problem with fusion of the
atrioventricular endocardial cushions—hence the old title
of "endocardial cushion defect"
[3, 4]. The difficulty in proving
this concept, however, reflects the fact that, although it could
be argued that we are now more certain of the extent of the
structures derived from the cushions
[5], we do not yet know
whether the cushions themselves were defective in the abnormal
hearts, nor the extent of failure of fusion needed to produce
an abnormality. This uncertainty, in turn, reflects the limited
opportunities for comparing the development of the cushions
in normal and abnormal human hearts
[5, 6]. Because of this,
there is great need for animal models of deficient atrioventricular
septation. At first, it was thought that the mouse with trisomy
16 would provide such a model, and hence elucidate the likely
mechanisms producing abnormal cardiac development in humans
with trisomy 21
[7, 8]. Our own studies, however, showed marked
differences in the nature of the hearts with deficient septation
as seen in the mouse with trisomy 16 when compared with the
typical forms of atrioventricular septal defect and common atrioventricular
junction encountered in humans
[9]. This probably reflects the
fact that the chromosome 16 in the mouse also carries genes
which, in the human, are carried on the chromosome 22
[10].
Indeed, it is surely not coincidental that our results from
the mouse with trisomy 16 revealed abnormal development of the
outflow tracts, lesions known to be produced in humans with
microdeletions on chromosome 22 (so-called "Catch-22"
[11]).
Because of this, there is the need for other animal models.
In this respect, use of transgenic animals, including gene knockouts,
will be of great value if it can be shown that the lesions produced
are more akin to the human anomalies. Such a claim was recently
made for the RXR alpha knockout
[12], it being argued that the
malformations seen were directly comparable with the spectrum
of abnormal atrioventricular septation encountered in humans
with "atrioventricular canal malformations". The arrangement
of the normal atrioventricular junctions depicted in this study,
however, bears little resemblance to our understanding of the
salient anatomy
[13]. Any study hoping to clarify the mode of
development of cardiac malformations must surely be based upon
a correct appreciation of the underlying morphology. This means
that it is essential to recognize the phenotypic differences
between hearts having normal atrioventricular junctions and
those with deficient atrioventricular septation, over and above
the obvious differences in septal morphology. This feature was
lacking in the otherwise outstanding study of Gruber et al.
[12]. In this review, therefore, we compare and contrast the
junctional morphologies in the normal human heart, and in human
hearts exhibiting deficient atrioventricular septation in the
setting of a common atrioventricular junction ("endocardial
cushion defect"). We then contrast these findings with observations
made in normal mice, and in the mouse with trisomy 16. We also
make reference to the malformations found in the mice with deficiency
of RXR alpha
[12], and to the mouse with the iv/iv mutation
[14]. In this way, we hope to focus attention on the underlying
anatomic features which require recognition by those seeking
to establish the processes of normal and abnormal development.
|
2 The normal atrioventricular junctions
|
|---|
The atrioventricular junctions are the area of the heart where
the atrial myocardium is inserted into the base of the ventricular
mass. These areas serve not only to provide the hinge for the
atrioventricular valves, but also to insulate the two muscle
masses one from the other, except at the site of penetration
of the atrioventricular conduction axis. In the normal heart,
both in man (
Figs. 1 and 2
), and in the mouse (
Fig. 3), there
are two such junctions, one on the right side surrounding the
orifice of the tricuspid valve, and the other on the left around
the mitral valve. There is then a very limited septal area between
them. On the right side, the right atrial myocardium is contiguous
with the entirety of the base of the right ventricle, and with
the muscular ventricular septum, apart from the area occupied
by the membranous part of the septum. The outflow tract of the
right ventricle is discrete and free-standing, being formed
by the tubular muscular subpulmonary infundibulum which supports
the entirety of the pulmonary valvar leaflets. On the left side,
both the mitral and aortic valves are incorporated within the
ventricular base (
Fig. 1). Because of this, there is very limited
contact between left atrial myocardium and the muscular ventricular
septum, with the muscular atrioventricular junction being confined
to the postero–inferior margin of the ventricular base
(
Fig. 2). Antero–superiorly, the orifice of the aortic
valve is interposed between the mitral valvar orifice and the
septum (
Fig. 1). This arrangement leads to complexity in the
formation of the fibrous septal area, which separates the right
side of the heart from the subaortic outflow tract. This so-called
membranous septum is divided into atrioventricular and interventricular
components by the attachment of the hinge of the septal leaflet
of the tricuspid valve on its right side. There then seems to
be a more extensive septal area posteriorly, where the hinge
of the tricuspid valve is off-set relative to that of the mitral
valve. Dissection of the relationships of atrial and ventricular
walls in this area, nonetheless, reveals that an extension of
the fibro–fatty tissue occupying the postero–inferior
atrioventricular groove interposes between the atrial and ventricular
musculatures, which lie side-by-side in the apparently septal
area. The area, therefore, represents a muscular atrioventricular
sandwich. In this analogy, the fibro–fatty tissue represents
the meat in the bread slices of the muscular sandwich. The entire
composite is not a true intracardiac atrioventricular septum
if a septum is defined on the basis of representing that part
of the wall which can be removed without exiting from the cavities
of the heart
[15]. This fact is of importance when considering
the mechanics of atrioventricular septation. The formation of
this area, along with incorporation of the coronary sinus in
the left atrioventricular groove, depends upon differential
growth and moulding of tissues subsequent to the completion
of active septation. Involved in such differential growth is
another important area often considered septal, namely the tissue
between the orifices of the inferior caval vein and the coronary
sinus. Dubbed the sinus septum, this area is no more than the
infolded wall of the systemic venous sinus between the venous
orifices. It can be likened to the crotch in a pair of trousers.
View larger version (121K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 1 The short axis section of the normal human heart is viewed from beneath. It shows the separate atrioventricular junctions for the right and left ventricles, but emphasizes the wedged position of the aortic valve between the mitral valve and the septum. The dotted line illustrates the area of aortic–mitral fibrous continuity. Compare with Fig. 5.
|
|
View larger version (47K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 2 This diagram compares the short axis morphology of the normal heart and hearts with common atrioventricular junction (upper panel) with their "four chamber" equivalents along the lines x–x (lower panels). AVSD=atrioventricular septal defect.
|
|
View larger version (195K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 3 The short axis of a mouse heart close to term demonstrated by scanning electron microscopy. As in the human heart, the subaortic outlet is wedged between the aortic leaflet of the mitral valve (asterisk) and the septum.
|
|
|
3 Significance of anatomy to development
|
|---|
Appreciation of the structure of the normal atrioventricular
junctions as described above helps in the understanding of the
contributions made to the definitive cardiac structures from
the atrioventricular endocardial cushions. It is generally accepted
that the superior and inferior atrioventricular cushions, along
with the outflow cushions, contribute to the membranous septum
[5, 16]. The tissues produced by fusion of the cushions then
provide the scaffold for incorporation of the aortic outflow
tract within the left ventricle, eventually being seen as the
area of aortic–mitral valvar fibrous continuity. When
the atrioventricular membranous septum is intact, therefore,
and the aortic outflow tract is normally incorporated in the
left ventricle, it is difficult logically to invoke failure
of fusion of the cushions as a developmental mechanism. This
is the morphologic situation pertaining in so-called "isolated"
cleft of an otherwise normally structured mitral valve (
Fig. 4).
Yet some do contend that this anomaly is a "forme fruste" of
atrioventricular canal malformation
[17], and some of the RXR
alpha knockout mice exhibit this lesion
[12]. The junctional
phenotype of the hearts with "isolated" mitral valvar clefts,
nonetheless, is separate right and left atrioventricular junctions
[18]. This normal phenotypic arrangement of the atrioventricular
junctions is fundamentally different from that found in other
hearts often considered to have a "cleft" in the mitral valve,
namely those with a space between the left ventricular component
of the bridging leaflets in the setting of a common atrioventricular
junction (true "atrioventricular canal malformations"—see
below). If the "clefts" seen in the two phenotypically dissimilar
lesions are both the consequence of failure of fusion of the
atrioventricular endocardial cushions
[12], then the failure
of fusion has to be far less severe in the hearts containing
the "isolated" cleft of the mitral valve. As we have established,
the very presence of normal atrioventricular junctional morphology
is indicative that the initial steps in development have proceeded
normally. Only in this way is it possible to account for the
normal formation of the fibrous atrioventricular septum, and
normal formation of the area of aortic–mitral valvar fibrous
continuity. It is almost as if the "isolated" cleft represents
subsequent breakdown of the normally constructed seam in the
aortic leaflet of the mitral valve.
View larger version (141K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 4 This abnormal human heart, with a perimembranous ventricular septal defect closed by a patch, is viewed from the left ventricle. The aorta is normally positioned within the ventricle, with normal continuity between its valve and the mitral valve. The leading edge of the aortic leaflet of the mitral valve, however, shows a minimal cleft (arrow). The heart has normal atrioventricular septation, and is fundamentally different from those with deficient atrioventricular septation.
|
|
|
4 Human hearts with common atrioventricular junction
|
|---|
All the evidence from the study of congenitally malformed human
hearts indicates that atrioventricular junctional arrangement,
atrioventricular valvar structure, and septal anatomy, are mutually
independent morphologic features. Thus, human malformations
characterised by presence of a common atrioventricular junction
can be found when the atrial chambers are connected to only
one ventricle (double inlet via a common atrioventricular orifice),
or when each atrium is joined via the common junction to its
own ventricle. Although those with each atrium connected to
a separate ventricle usually have an extensive atrioventricular
septa defect, examples can sometimes be found with intact septal
structures
[19, 20]. Recognition of the independence of these
three features is crucial if the ongoing discussion is properly
to be appreciated. It is the arrangement with each atrium connected
to its own ventricle, but through a common muscular atrioventricular
junction (
Figs. 5 and 6), which is traditionally described as
an "endocardial cushion defect", or an "atrioventricular canal
malformation". The entity exists because of complete lack of
formation of the normal fibrous atrioventricular septum, and
failure of formation of the postero–inferior atrioventricular
muscular sandwich. Two anatomic prototypes are generally recognised,
the so-called complete and partial variants. In terms of atrioventricular
junctional morphology, however, the arrangement in the two prototypes
is identical, but fundamentally different from the normal phenotype.
Several authorities recognise lesions intermediate between the
complete and partial forms, although using different criterions
to make these distinctions
[21, 22]. Be that as it may, the
difference between the two major groups, and the difference
in the "intermediate" variants, reflects the arrangement of
the valvar leaflets rather than the junctional morphology, specifically
the interrelationships of the bridging leaflets of the effectively
common atrioventricular valve, either to each other, or to the
atrial and ventricular septal structures
[2]. To stress once
more, the arrangement of the muscular atrioventricular junctions
is comparable within the overall group (
Figs. 5–7), and
is fundamentally different from normal hearts in all respects
(
Figs. 1 and 2
). In essence, the abnormal hearts are characterised
by a common area of junction between the atrial and ventricular
muscle masses (
Fig. 5), as opposed to the separate right and
left junctions seen in normal hearts, and all hearts with normal
atrioventricular septation (
Fig. 1), including those with isolated
cleft of an otherwise normally constructed mitral valve (
Fig. 4).
View larger version (141K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 6 This "four chamber section" (compare with Fig. 2—right lower panel) shows a human heart with an atrioventricular septal defect guarded by a common valve. The septal defect is bounded by the atrial (asterisk) and ventricular (star) septums.
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 7 These two human hearts both have deficient atrioventricular septation with a common atrioventricular junction. In the upper specimen (a), there is a common atrioventricular valvar orifice, with discrete and separate bridging leaflets crossing the septum. This is the so-called "complete" defect. The second heart (b) has separate valvar orifices for the right and left ventricles because a tongue of valvar tissue joins together the bridging leaflets. Although often said to be a "partial" defect, or an ostium primum atrial septal defect, the dissections show that the junctional morphology is directly comparable in the two hearts.
|
|
If the bridging leaflets themselves are separate from one another
in hearts with a common atrioventricular junction (
Fig. 7a),
then there is a common valvar orifice. If, in contrast, the
bridging leaflets in hearts with a common atrioventricular junction
are joined to each other along the plane of the ventricular
septum, then there are separate atrioventricular valvar orifices
for the right and left ventricles (
Fig. 7b). The level of shunting
across the septal deficiency, which is almost always to be found
at the centre of the common junction, is then dependent on the
relationship between the bridging leaflets and the edges of
the atrial and ventricular septums (
Fig. 8). Still further anatomical
variability can be seen according to the way the common junction
is shared between the ventricles, but almost always there is
a balanced arrangement. In some instances, human hearts may
be associated with overriding of the aorta as part of tetralogy
of Fallot
[23], or can be found with double outlet from the
right ventricle. In the majority of cases, despite the common
junction, the aortic outflow tract is incorporated within the
left ventricle. When compared with the normal heart, however,
the left ventricular outflow tract is narrowed, and the aortic
valvar leaflets are in fibrous continuity only with the superior
bridging leaflet (compare
Figs. 1 and 5). Furthermore, the left
atrioventricular valve, be it guarding a separate valvar orifice
for the left ventricle within the common junction, or the left
half of a common valve, has three leaflets (
Figs. 5 and 7).
In this arrangement, the space between the left ventricular
components of the bridging leaflets represents a zone of apposition
between them, and hence is part of the common atrioventricular
orifice. This space bears no resemblance, neither anatomically
nor functionally, to the "isolated" cleft of an otherwise normally
structured aortic leaflet of the mitral valve found in hearts
with separate right and left atrioventricular junctions (compare
Figs. 4 and 7). The two entities are phenotypically different
even if they may be linked developmentally on the basis of different
degrees of failure of fusion of the atrioventricular endocardial
cushions
[12].
|
5 Cardiac anomalies in the mouse with trisomy 16
|
|---|
All the hearts we studied from mice with trisomy 16
[9], and
those previously described by others
[7, 8]showed evidence of
deficient atrioventricular septation, and possessed a common
atrioventricular junction. The morphology seen, however, was
markedly different from the typical human patterns illustrated
above. This is because, although there is always an "ostium
primum" defect (
Fig. 9a), in none of the animals produced in
our breeding protocol
[9]did we find examples in which the common
atrioventricular junction was balanced between the right and
left ventricles, and in none was the aorta committed entirely
to the left ventricle. Instead, we found that, in half our animals,
the common atrioventricular junction was exclusively connected
within the left ventricle, and was guarded by a valve with three
leaflets (
Fig. 9b). In these hearts with exclusive left ventricular
connection of the junction, despite the presence of an "ostium
primum" defect, the floor of the right atrium was entirely muscular,
showing affinities with the typical arrangement of tricuspid
atresia as seen in humans. In the remaining animals, the common
junction was connected in part to the right ventricle (
Fig. 10a),
but never to sufficient extent to produce a balanced arrangement.
The ventriculo–arterial connections were abnormal in all
the trisomic mice, with either overriding of the aortic valve
(
Fig. 10b) or double outlet right ventricle in the majority.
Furthermore, in many of the animals there was failure of formation
of the subpulmonary infundibulum, so that the leaflets of both
arterial valves were at the same level, or else there was a
common arterial trunk. This arrangement is comparable to that
seen in humans with common arterial trunk, or else when there
is a so-called doubly committed and juxtaarterial ventricular
septal defect. These various combinations seen in the mice with
trisomy 16, therefore, do have their counterparts in human cardiac
pathology, but the abnormalities of the ventriculo–arterial
junctions are found only exceedingly rarely in the setting of
human hearts with common atrioventricular junction and deficient
atrioventricular septation ("atrioventricular canal malformations").
View larger version (94K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 9 This scanning electron micrograph (a) of a mouse with trisomy 16, sacrificed on the 18th embryonic day, shows an "ostium primum" defect and a right ventricle which receives only a small part of the common atrioventricular junction, most of which is connected within the dominant left ventricle. In this second mouse with trisomy 16, sacrificed on the 17th embryonic day and displayed by scanning electron microscopy (b), the common atrioventricular junction is connected exclusively within the dominant left ventricle. The left atrioventricular valve has a trifoliate arrangement, although only the superior and inferior leaflets remain in this dissection. The aortic valve is overriding the septum (asterisk).
|
|
View larger version (97K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 10 In this mouse with trisomy 16, sacrificed on the 18th embryonic day and revealed by dissection and scanning electron microscopy (a), the common atrioventricular junction has been displayed by removing the atrial myocardium. The junction is connected in part to the right ventricle, but not in the balanced fashion typical for humans (see Fig. 6). Note the side-by-side arterial trunks, which arise exclusively from the right ventricle. The second mouse with trisomy 16, also sacrificed on the 18th embryonic day (b), has the atrioventricular valve connected in part to the right ventricle in association with overriding of the aortic valve. Note the muscular outlet septum (asterisk).
|
|
|
6 Developmental implications
|
|---|
How do the anatomic arrangements described above help us as
we seek to determine the mechanisms producing so-called "atrioventricular
canal malformations"? The phenotype of the human anomaly characterised
by deficient atrioventricular septation is the common atrioventricular
junction, irrespective of whether the valvar leaflets guarding
the junction are arranged to produce a common orifice or separate
valvar orifices for the right and left ventricles (the so-called
"complete" and "partial" variants). The arrangement of the bridging
leaflets in both these variants is remarkably similar to the
disposition of the atrioventricular endocardial cushions during
early stages of embryological development. If it is presumed
that the cushions do, indeed, form the scaffold of the leaflets,
then note must be taken of the fact that exactly the same junctional
arrangement of the atrial and ventricular musculatures is seen
when the bridging leaflets are fused to each other and to the
crest of the ventricular septum (ostium primum defect—
Fig. 7b)
as when they are discrete and separate structures with a space
between their inferior surface and the ventricular septum (complete
defect—
Fig. 7a). It is difficult to reconcile both these
facts with the presumption that the overall anomaly is dependent
upon failure of fusion of the atrioventricular endocardial cushions.
In terms of muscular junctional morphology, there is no spectrum
of anatomic severity. As we have demonstrated, the atrioventricular
junction is just as common in the "partial" as in the presumed
"complete" variant. Both are guarded by a basically common valve
with leaflets bridging the ventricular septum, even if divided
into right and left components in the supposedly partial form.
All these hearts are different phenotypically from normal hearts,
which have separate muscular junctions surrounding the mitral
and tricuspid valves, with a fibrous atrioventricular septum
interposed between the right side and the subaortic outflow
component of the left ventricle. It is this latter phenotype
which is found in hearts with an "isolated" cleft of the mitral
valve. The phenotypic differences, in total, represent an "all-or-none"
phenomenon. The spectrum of severity reflects the arrangement
of the leaflets themselves, involving the manner of their formation
rather than their junctional attachments.
This point is crucial, since Gruber et al. [12]seemingly analysed their important series of RXR alpha knockout mice as if the junctional phenotypes were identical in all their animals (their Figure 2). Yet their illustrations show clearly the presence of a common atrioventricular junction in some animals (their Figure 3i), and others with a cleft of an aortic leaflet of an otherwise normally structured mitral valve (their Figure 5f). If, as they do, it is argued that failure of fusion of the atrioventricular endocardial cushions is responsible for all the abnormal hearts, then some other complex event must underscore the formation of the common atrioventricular junction. Very likely, the abnormality involves also the musculature of the atrioventricular canal, shown to have an abnormal architecture in the mouse with trisomy 16 [24]. The iv/iv mouse is another model which includes a common atrioventricular junction as one of its parts [14], and in which looping is known to be abnormal. Study of this model may help elucidate the mechanisms involved. In this respect, note should also be taken of the fact that our study of the trisomic mouse showed that from an early stage the cushions in the abnormal animals were of larger volume than normal, but with a lower density of cells. When judged on the basis of somitic counts, overall temporal development was delayed in the trisomic animals. All of these are important facts for those seeking the genes which govern the abnormal development. The entirety of the atrioventricular junction requires attention, and not just the cushions. Particular attention should be directed to the fundamental differences in morphology already emphasized between the hearts with a common atrioventricular junction and those with normal atrioventricular junctions (Table 1).
During development, there is certainly a stage at which the
disposition of the normal cushions reflects the arrangement
of the bridging leaflets seen in hearts with common atrioventricular
junction and atrioventricular septal defect. At the stage at
which these features are comparable, nonetheless, the entirety
of the outflow segment of the normal heart is still supported
by the developing right ventricle. In human hearts with common
atrioventricular junction and atrioventricular septal defect,
the subaortic outflow tract is usually incorporated within the
left ventricle, albeit in narrowed form when compared to the
normal arrangement. In the trisomic mice with deficient atrioventricular
septation, we never encountered a heart with the aorta appropriately
connected within the left ventricle. The mechanisms of transfer
of the aorta from the developing outlet segment to the definitive
left ventricle represent another unsolved embryologic conundrum.
Proper attention to the sequence of development of the mice
with trisomy 16, taking note of morphological and genetic events
and comparing them to normal events, could well clarify these
processes, as well as the equally problematic mechanism of transfer
of the right atrioventricular orifice appropriately to the developing
right ventricle. Thus, half of the mice with trisomy 16 have
the entire atrioventricular junction connected within the left
ventricle, but in the presence of an atrioventricular septal
defect. Univentricular connection to a dominant left ventricle
is the characteristic feature of the human malformation known
typically as tricuspid atresia
[25], but the human lesion is
only very rarely found with an "ostium primum" defect
[26],
this being the dominant feature of the trisomic mice. The difference
between classical tricuspid atresia with or without an ostium
primum defect lies in the formation of the atrioventricular
septum. An atrioventricular membranous septum is present in
the hearts characterised by absence of the right atrioventricular
connection, and there is normal mitral–aortic valvar fibrous
continuity in such hearts. Absence of the atrioventricular septal
structures is the key feature of the hearts with common atrioventricular
junction. Here again, recent observations recapitulating old
information confirm that attention needs to be focused on structures
other than just the atrioventricular endocardial cushions. As
long ago as 1880, Wilhelm His the elder pointed to the significance
of a contribution from the extracardiac tissues to the vestibule
of the tricuspid valve
[27]. His dubbed this structure the "spina
vestibuli". Tasaka et al.
[28]resurrected interest in this structure,
or rather the lack of it, in the formation of the primary atrial
septum. Our findings endorse the validity of their suggestion,
and point further to the importance of appropriate incorporation
of the systemic venous sinus within the developing right atrium.
It is the growth into the heart of the vestibular spine along
the right margin of the pulmonary venous portal, along with
correct rotation of the horns of the systemic venous sinus,
which make possible the development of the atrioventricular
muscular sandwich between the off-set hinges of the leaflets
of the mitral and tricuspid valves. Growth of the sandwich,
however, is a much later development than closure of the embryonic
interventricular communication by the membranous septum. Formation
of the sandwich cannot occur until there has been appropriate
separation of the atrioventricular junctions. It is usually
associated with incorporation of the subaortic outflow tract
within the left ventricle. Those seeking to establish perturbations
in the formation and sequencing of genes need to be aware of
the temporal sequence of these events, and that the formation
of the valvar leaflets occurs in part by a process of delamination
of inlet myocardium
[29]. Note needs also to be taken of the
fundamental difference in junctional morphology which is the
consequence of failure of formation of the atrioventricular
septal structures. Careful study of the mouse with trisomy 16
(
Table 1), along with the RXR alpha knockout and the iv/iv mutant,
should provide answers to many of these questions, even if not
all the animals develop in a fashion comparable to humans with
deficient atrioventricular septation. Providing that note is
taken of the differences in junctional morphology between hearts
with intact and deficient atrioventricular septal structures,
many more answers will be provided by these models. Significantly,
the models also show abnormalities in septation of the arterial
segment of the heart, and in formation of the subpulmonary infundibulum.
These combinations of abnormalities are rarely encountered in
humans with deficient atrioventricular septation, although frequently
seen in humans as abnormalities in their own right.
|
7 Conclusions
|
|---|
The mouse with trisomy 16 was proposed initially as a model
for the cardiac abnormalities seen in humans with trisomy 21,
notably atrioventricular septal defect with common atrioventricular
junction. It is now clear that this concept was simplistic.
Although the trisomic mice exhibit deficient atrioventricular
septation, and develop a common atrioventricular junction, the
anatomic prototypes are markedly different from those encountered
in the human situation. A spectrum much more comparable to the
human abnormalities is found in mice with knockout of the genes
encoding the RXR alpha receptor. This latter model also includes
animals with so-called "isolated" cleft of the normally positioned
mitral valve, providing a paradoxical situation where comparable
genotypes produce remarkably different phenotypes. This points
to the crucial interaction of both genetic and epigenetic factors
in the formation of congenital malformations. Careful study
of both models, along with mice having the iv/iv mutation, which
also incorporates a common atrioventricular junction as part
of the abnormal cardiac phenotype, should help elucidate the
genetic as opposed to the epigenic steps involved in normal
and abnormal atrioventricular septation. Such studies need to
look beyond the endocardial cushions, taking note also of the
atrioventricular junctional myocardium, the contributions of
the extracardiac tissues through the vestibular spine, and the
significance of appropriate incorporation of the systemic venous
sinus within the developing right atrium. Study of the mouse
models should also help establish the mechanics of, and genes
involved in, transfer of the right atrioventricular orifice
to the developing right ventricle and the aorta to the developing
left ventricle, as well as clarifying the mode of formation
of the free-standing muscular subpulmonary infundibulum.
Time for primary review 28 days.
|
Acknowledgements
|
|---|
Some of the photographs of the human hearts were taken at Children's
Hospital of Pittsburgh, PA, USA, and we thank Drs. James R.
Zuberbuhler and William C. Neches for permission to reproduce
them. The work was supported by grants from the British Heart
Foundation.
|
References
|
|---|
- Van Mierop LHS. Pathology and pathogenesis of endocardial cushion defects. Surgical implications. In: Davila JC, editor. Second Henry Ford Hospital International Symposium on Cardiac Surgery. New York: Appleton–Century–Crofts, 1977:201–207.
- Becker A.E, Anderson R.H. Atrioventricular septal defects. What's in a name? J Thorac Cardiovasc Surg (1982) 83:461–469.[Web of Science][Medline]
- Campbell M, Missen G.A.J. Endocardial cushion defects: common atrioventricular canal and ostium primum. Br Heart J (1957) 19:403–418.[Free Full Text]
- Van Mierop L.H.S, Alley R.D, Kausel H.W, Stranahan A. The anatomy and embryology of endocardial cushion defects. J Thorac Cardiovasc Surg (1962) 43:71–83.[Web of Science][Medline]
- de Jong S, Viràgh S, Moorman A.F.M. Cardiac development a morphologically integrated molecular approach. Cardiol Young (1997) 7:228–240.[Web of Science]
- Wenink A.C.G, Zevallos J.C. Developmental aspects of atrioventricular septal defects. Int J Cardiol (1988) 18:65–78.[CrossRef][Web of Science][Medline]
- Miyabara S, Gropp A, Winking H. Trisomy 16 in the mouse fetus associated with generalized edema and cardiovascular and urinary tract anomalies. Teratology (1982) 25:369–380.[CrossRef][Web of Science][Medline]
- Miyabara S, Sugihara H, Yonemitsu N, Yun K. Comparative study of the phenotypic expression of mice trisomy 16 by different female strains: attempt at an animal model for human trisomy 21. Cong Anom (1984) 24:283–292.
- Webb S, Brown N.A, Anderson R.H. Cardiac morphology at late fetal stages in the mouse with trisomy 16: consequences for different formation of the atrioventricular junction when compared to humans with trisomy 21. Cardiovasc Res (1997) 34:515–524.[Abstract/Free Full Text]
- Nadeau J. Maps of linkage and synteny homologies between mouse and man. Trends Genet (1989) 5:82–86.[CrossRef][Web of Science][Medline]
- Wilson D.I, Burn J, Scambler P, Goodship J. Di George syndrome: part of CATCH 22. J Med Genet (1993) 30:852–856.[Abstract/Free Full Text]
- Gruber P.J, Kubalak S.W, Pexieder T, Sucov H.M, Evans R.M, Chein K.R. RXRa deficiency confers genetic susceptibility for aortic sac, conotruncal, atrioventricular cushion, and ventricular muscle defects in mice. J Clin Invest (1997) 98:1332–1343.[CrossRef][Web of Science]
- Webb S, Brown N.A, Anderson R.H. The structure of the mouse heart in late fetal stages. Anat Embryol (1996) 194:37–47.[Medline]
- Seo J.-W, Brown N.A, Anderson R.H. Abnormal laterality and congenital anomalies. Relations of visceral and cardiac morphologies in the iv/iv mouse. Circulation (1992) 86:642–650.[Abstract/Free Full Text]
- Anderson R.H, Brown N.A. The anatomy of the heart revisited. Anat Rec (1996) 246:1–7.[CrossRef][Medline]
- Odgers P.N.B. The development of the pars membranacea septi in the human heart. J Anat (1938) 72:247–259.[Web of Science][Medline]
- Di Segni E, Edwards J.E. Cleft anterior leaflet of the mitral valve with intact septa. A study of 20 cases. Am J Cardiol (1983) 51:919–926.[CrossRef][Web of Science][Medline]
- Sigfùsson G, Ettedgui J.A, Silverman N.H, Anderson R.H. Is a cleft in the anterior leaflet of an otherwise normal mitral valve an atrioventricular canal malformation? J Am Coll Cardiol (1995) 26:508–515.[Abstract]
- Silverman N.H, Ho S.Y, Anderson R.H, Smith A, Wilkinson J.L. Atrioventricular septal defect with intact atrial and ventricular septal structures. Int J Cardiol (1984) 5:562–572.
- Ho S.Y, Russell G, Gerlis L.M. Atrioventricular septal defect with intact septal structures in a 74-year-old. Int J Cardiol (1990) 26:371–373.[CrossRef][Web of Science][Medline]
- Wakai C.S, Edwards J.E. Developmental and pathologic considerations in persistent common atrioventricular canal. Proc Staff Meet Mayo Clin (1956) 31:487–500.[Medline]
- Bharati S, Lev M, McAllister H.A.J, Kirklin J.W. Surgical anatomy of the atrioventricular valve in the intermediate type of common atrioventricular orifice. J Thorac Cardiovasc Surg (1980) 79:884–889.[Abstract]
- Delius R.E, Kumar R.V, Elliott M.J, Stark J, de Leval M.R. Atrioventricular septal defect and tetralogy of Fallot: a 15 year experience. Eur J Cardiothorac Surg. (1997) 12:171–176.[Abstract]
- Webb S, Anderson R.H, Brown N.A. Endocardial cushion development and heart loop architecture in the trisomy 16 mouse. Dev Dyn (1996) 206:301–309.[CrossRef][Web of Science][Medline]
- Orie J.D, Anderson C, Ettedgui J.A, Zuberhuler J.R, Anderson R.H. Echocardiographic–anatomical correlations in tricuspid atresia. J Am Coll Cardiol (1995) 26:750–758.[Abstract]
- Van Praagh S, Vangi V, Hee Sul J, Metras D, Parness I, Castaneda A.R. Tricuspid atresia or severe stenosis with partial common atrioventricular canal: anatomic data, clinical profile and surgical considerations. J Am Coll Cardiol (1991) 17:932–943.[Abstract]
- His W. Die Area interposita, die Eustachi'sch Klappe und die Spina vestibuli. In: Anatomie menschlicher Emryonen. Leipzig: Verlag von F.C.W. Vogel, 1880:146–152.
- Tasaka H, Krug E.L, Markwald R.R. Origin of the pulmonary venous orifice in the mouse and its relation to the morphogenesis of the sinus venosus, extracardiac mesenchyme (spina vestibuli), and atrium. Anat Rec (1996) 246:107–113.[CrossRef][Medline]
- Lamers W.H, Virágh S, Wessels A, Moorman A.F.M, Anderson R.H. Formation of the tricuspid valve in the human heart. Circulation (1995) 91:111–121.[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
|
|
 
D Papadimitriou, T Xanthos, I Dontas, P Lelovas, and D Perrea
The use of mice and rats as animal models for cardiopulmonary resuscitation research
Lab Anim,
July 1, 2008;
42(3):
265 - 276.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
1 Introduction
