Download Structure and Function of the Developing Zebrafish Heart

THE ANATOMICAL RECORD 260:148 –157 (2000)
Structure and Function of the
Developing Zebrafish Heart
NORMAN HU,1* DAVID SEDMERA,2 H. JOSEPH YOST,1,3 AND
EDWARD B. CLARK1,4
1
University of Utah, Department of Pediatrics, Salt Lake City, Utah 84132
2
Medical University of South Carolina, Department of Cell Biology and Anatomy,
Charleston, South Carolina 29425
3
University of Utah, Department of Oncological Sciences, Huntsman Cancer
Institute, Salt Lake City, Utah 84112
4
Primary Children’s Medical Center, Salt Lake City, Utah 84113
ABSTRACT
The combination of optical clarity and large scale of mutants makes the
zebrafish vital for developmental biologists. However, there is no comprehensive reference of morphology and function for this animal. Since study of
gene expression must be integrated with structure and function, we undertook a longitudinal study to define the cardiac morphology and physiology of
the developing zebrafish. Our studies included 48-hr, 5-day, 2-week,
4-week, and 3-month post-fertilization zebrafish. We measured ventricular
and body wet weights, and performed morphologic analysis on the heart
with H&E and MF-20 antibody sections. Ventricular and dorsal aortic
pressures were measured with a servonull system. Ventricular and body
weight increased geometrically with development, but at different rates.
Ventricle-to-body ratio decreased from 0.11 at 48-hr to 0.02 in adult. The
heart is partitioned into sinus venosus, atrium, ventricle, and bulbus arteriosus as identified by the constriction between the segments at 48-hr.
Valves were formed at 5-day post-fertilization. Until maturity, the atrium
showed extensive pectinate muscles, and the atrial wall increased to two to
three cell layers. The ventricular wall and the compact layer increased to
three to four cell layers, while the extent and complexity in trabeculation
continued. Further thickening of the heart wall was mainly by increase in
cell size. The bulbus arteriosus had similar characteristics to the myocardium in early stages, but lost the MF-20 positive staining, and transitioned
to smooth muscle layer. All pressures increased geometrically with development, and were linearly related to stage-specific values for body weight
(P ⬍ 0.05). These data define the parameters of normal cardiac morphology
and ventricular function in the developing zebrafish. Anat Rec 260:
148 –157, 2000. © 2000 Wiley-Liss, Inc.
Key words: Zebrafish; ventricular weight; cardiac morphology;
cardiac development; myocardial growth; ventricular blood pressure; dorsal aortic blood pressure
The zebrafish (Danio rerio) is a major new model system
for cardiovascular development and genetics. It presents a
prototypic vertebrate heart with only a single atrium and
ventricle. The molecular mechanisms governing its patterning appear to be similar to those in more complex
hearts of higher vertebrates (Weinstein and Fishman,
1996; Fishman et al., 1997).
The zebrafish heart consists of four chambers (sinus
venosus, atrium, ventricle, and bulbus arteriosus) connected in series. It pumps desaturated venous blood to the
©
2000 WILEY-LISS, INC.
ventral aorta leading to the gill arches where oxygenation
occurs, and from where it is distributed to the rest of the
Grant sponsor: NIH; Grant number: HL50582.
*Correspondence to: Norman Hu, University of Utah, Department of Pediatrics, 2B465 SOM, 50 N. Medical Drive, Salt Lake
City, UT 84132. E-mail: [email protected]
Received 29 March 2000; Accepted 3 July 2000
ZEBRAFISH HEART DEVELOPMENT
body. While form and function of the cardiovascular system have been studied in the zebrafish prior to hatching
(Lee et al., 1994; Stainier et al., 1993, 1996), morphological description of its anatomy and the hemodynamic function, especially of development after the earliest stages of
its formation, is not available. Despite its apparent simplicity, the zebrafish heart shares a common structural
scenario with an avian or mammalian heart, and, as such,
can serve as a model system for various experimental
studies (Stainier et al., 1996; Weinstein and Fishman,
1996; Warren and Fishman, 1998).
The cardiovascular system appears when needs for oxygen and nutrition cannot be met by diffusion alone, because of the volume or increased metabolic rate of an
organism (Burggren and Pinder, 1991; Pelster and Burggren, 1996). As such, during the course of embryogenesis,
the heart is the first definitive organ to develop and become functional, as any later survival depends on its
proper function (Clark, 1990). However, there is a period
when the heart is already functional, but not yet essential
in the early developmental stages. For instance, the first
heart beat in the chick embryo occurs before the connections of circulation are completed, and development is
possible for several hours without full blood circulation
(Manasek, 1970). In the zebrafish embryo, this period
actually lasts for several days because of the small size
and relatively low metabolism. This enables analysis of
mutants with compromised or no cardiac function for a
considerable period of time (Stainier and Fishman, 1992;
Stainier et al., 1996). In contrast, a similar phenotype in
the mouse would result in early lethality and embryo
reabsorption.
Because of its small size, the functional study of zebrafish heart is a challenge, and physiological data are not
easily accessible. Furthermore, there is no comprehensive
reference of morphology and cardiovascular physiology for
this animal. Studies of gene expression and characterization of mutants with abnormal heart development and
function must be integrated with comprehensive knowledge of normal structure and function. We, therefore, undertook a longitudinal study to define the cardiac morphology and physiology of the developing zebrafish.
MATERIALS AND METHODS
Experimental Stages
We studied 48-hr (hatching), 5-day (early larva), 2-week
(late larva), 4-week (young adult), and 3-month (adult)
post-fertilization zebrafish. Staging was based on external
landmarks including somite number, external features,
and cardiac morphology (Kimmel et al., 1995). These selected stages represented embryo hatching, completion of
the heart septation, and maturation of the zebrafish heart.
The fertilized eggs were from normal matings. The zygotes
were incubated in the instant ocean solution (0.075 g/L) at
28.5°C (Westerfield, 1995). Hatched embryos were transferred to the 2 liter tanks for further maturation.
Body and Ventricular Wet Weights
The heart was removed from zebrafish euthanized with
10 g/Kg tricaine. The body of the zebrafish was blotted to
remove excess water, and weighed on an OHAUS balance
(OHAUS Scale Corp., Florham Park, NJ) accurate to ⫾10
␮g. The ventricle was weighed after the bulbus arteriosus
and the atrium were trimmed off. We weighed six to ten
149
groups of pooled samples at each stage. For stages prior to
4-weeks post-fertilization, 20 ventricles were pooled for
each measure of wet weight. Five hearts were pooled for
each measure for stages beyond 4-weeks post-fertilization.
Morphological Analysis
Zebrafish were fixed with 80% ethanol-20% DMSO for
24 hr, and then stored in 70% ethanol until processing.
Adult hearts were perfused with PBS followed by 4%
paraformaldehyde in PBS, immersed for 2 hr in the same
fixative, and stored in PBS with 0.1% sodium azide. For
histology, specimens were dehydrated through an ascending ethanol series, and embedded in paraffin in transverse, frontal, or sagittal orientation. Serial sections were
cut and stained with hematoxylin-eosin.
Selected sections were also stained with MF-20 monoclonal antibody (Developmental Studies Hybridoma Bank,
University of Iowa, Iowa City, IA), which stained myosin
in skeletal and cardiac muscle. As a secondary antibody,
we used Alexa 488 coupled goat-anti-mouse (Molecular
Probes, Eugene, OR), and the nuclei were counterstained
with propidium iodide (Sigma, St. Louis, MO). Images
were captured digitally on a Leica DMLB microscope fitted with an epifluorescence attachment and Hamamatsu
C5810 color camera (Lieca, Deerfield, IL), and further
processed using Adobe Photoshop software (Adobe, San
Jose, CA). Adult hearts, in addition to routine histology,
were embedded in acrylamide, and sectioned in different
planes into 200 micron acrylamide slabs (Germroth et al.,
1995). The slabs were stained in-gel with rhodamine-conjugated phalloidin for actin. The slabs were examined on a
confocal microscope (Zeiss Axiophot with BioRad MRC
1024).
Ventricular and Aortic Pressure Measures
Zebrafish were anesthetized with tricaine methanesulfonate (MS222) in instant ocean solution at a dose of 0.40
g/Kg. The fish were transferred to a shallow concave
trough of a custom-built glass chamber (Radnoti, Monrovia, CA). The trough was filled with instant ocean solution, and the temperature of the solution was monitored
by a thermistor, and maintained by 15 L/min flow of 30°C
water through the hollow wall of the glass chamber. We
measured ventricular and dorsal aortic blood pressures
(n ⱖ 6 for each age) with a servonull pressure system
(900A, WP Instruments, Sarasota, FL) at stages beyond
48 hr. It was not technically feasible to record waveforms
at 48-hours post-fertilization zebrafish. The probe was a
2–5 micron diameter tip glass micropipette filled with 1M
NaCl. The cannula was zeroed at the level of the embryo/
larva before and after pressure measurements. The servonull measured pressure is linear (y ⫽ 0.995⫻ ⫺0.23)
and highly correlated (r ⫽ 0.99, S.E.E. ⫽ 0.11 mmHg)
when compared to a standing water column over the range
of 0 to 30 mmHg (Clark and Hu, 1982, 1990; Hu and
Clark, 1989; Clark 1990). Analog pressure waveforms
were digitally sampled at 2 ms intervals by an analog to
digital conversion board (LabView, National Instruments,
Austin, TX), and stored for processing on the Jaz cartridge
(Iomega, Roy, UT).
Statistical Analysis
For physiologic data, at least five to seven consecutive
cycles were analyzed for each fish. Quantitative data are
150
HU ET AL.
Fig. 1. A: Body and ventricular wet weight during zebrafish development. There is almost a six-fold increase in the body wet weight at
4-week post-fertilization zebrafish, while the ventricular wet weight retains a three-fold increase during this period of development. Less than
2% of the ventricular wet weight is accountable for the total weight in the
mature adult (3-months post-fertilization). B: Log-log plot of ventricular
versus body wet weight shows linearity and significant correlation during
development.
presented as means ⫾ standard error of the mean. We
performed regression analysis, with a statistically significant level defined as a P value of less than 5% (P ⬍ 0.05).
RESULTS
Ventricular and Body Wet Weights
Both ventricular and body wet weights increased with
development. Increase in ventricular wet weight was relatively less than body weight of the matched stages (Fig.
1A). The relative weight of the ventricle decreased as the
body weight increased from the embryonic stage to the
young adult. At 48-hr post-fertilization, the ventricle accounted for 10.5% of the total body weight, but decreased
to 1.8% in the 3-months post-fertilization adult. Log ventricular wet weight correlated linearly with log body wet
weight of the zebrafish (Fig. 1B).
Morphologic Analysis
At 48-hr post-fertilization, the zebrafish heart consisted
of a smooth-walled tube partitioned into all four segments
with the definite structure of sinus venosus, atrium, ventricle, and bulbus arteriosus. Each part was externally
identified by a constriction between the segments. The
myocardium was about one cell thick in each segment,
except for ventricle which had two to three cell layers (Fig.
2). The heart was lined with endocardium, separated from
the myocardium by a layer of cardiac jelly in the atrium
and bulbus arteriosus. The ventricle and sinus venosus
did not have a cardiac jelly layer. There were no valves
separating the segments at this stage. The heart was
enclosed in the pericardial cavity with paired pericardial
muscles running in caudo-cranial direction, and covered
with epicardium.
At 5-days post-fertilization, the ventricle had developed
extensive trabeculation while the inner surface of the
atrium was still smooth (Fig. 3). The atrial wall and the
trabeculae were one- to two-cell layers thick. The outer
ventricular myocardial layer was about two to three cells
thick. The bulbus arteriosus had similar characteristics to
myocardium with positive MF-20 staining. The valves between the atrium and the ventricle, and between the ventricle and the bulbus arteriosus had two cusps, with the
leaflets having a mesenchymal, cushion-like appearance.
At 2-weeks post-fertilization, the pectinate muscles in
the atrium that were barely discernible in the previous
stage, became more prominent. Ventricular morphology
was very similar to the previous stage. The bulbus arteriosus was still positive for MF-20 (Fig. 4). This immunostaining was more prominent in the ventricle than in the
atrium, but had a relatively weaker signal when compared
to the neighboring skeletal muscle. The valves separating
the ventricle from the atrium had almost reached maturity, with the leaflets notably thinner than the previous
stages.
At 4-weeks post-fertilization, the trabeculation in the
ventricle was more complex, but similar to the compact
layer, which never exceeded three-cell thick. The MF-20
staining was more intense in the trabeculae than in the
outer compact layer. The pectinate muscles in the atrium
were more extensive when compared to the earlier stages.
However, the atrial wall was still only one- to two-cell
layers thick. The bulbus arteriosus lost its MF-20 positive
staining (Fig. 5), and showed a smooth muscle phenotype
when observed under the standard HE staining. The cellularity of the valves increased, and each individual leaflet
showed thickening when compared to the previously described stage.
In the 3-months post-fertilization adult fish, the constrictions between individual segments were more prominent, giving the heart chambers a less rounded appearance and contributing to the overall characteristic shape
(Fig. 6). The atrium showed an extensive network of pectinate muscles, which were heavily branched. The atrial
wall increased to two- to three-cell layers thick, as did the
pectinate muscles. Further thickening of these structures
relative to younger stages was clearly due to increase in
cell size, seen in H&E stained sections. The ventricle
developed a thicker outer compact myocardial layer, as
seen by staining for actin, with the epicardial vessels
running in the subepicardium and penetrating into it (Fig.
7). The compact layer was three or four cells thick, while
the trabeculae were about two cell layers. As in the
atrium, these structures were substantially thicker than
the earlier described stages due to the increased cell size.
Bulbus arteriosus was composed of a thick layer of smooth
ZEBRAFISH HEART DEVELOPMENT
Fig. 2. Transverse sections through a 48-hr post-fertilization zebrafish heart. A: The yolk sac is most prominent at the caudal heart
region. Aⴕ: High power view of the atrial wall shows the epicardium (Ep),
and the single cell layer myocardium (My) separated from the endocardium (En) by acellular cardiac jelly (CJ). B: Inner surface of the entire
heart is smooth, and the chambers can be distinguished only on the
151
basis of their relative position in the serial sections. Bⴕ: High power view
of the ventricle shows the myocardium reaches two cell layers in some
areas. Hematoxylin-eosin staining. A, atrium; BA, bulbus arteriosus; 夝,
pericardial muscles; V, ventricle. Scale bar ⫽ 100 microns in A, B; 10
microns in A⬘, B⬘.
152
HU ET AL.
Fig. 3. Transverse (A,B) and sagittal (C) sections through a 5-days
post-fertilization zebrafish heart. A: Caudal cross-section shows the
atrium (filled with blood) and the ventricle with prominent trabeculation.
Aⴕ: High power view of the atrial wall with its single cell layer myocardium
(My) covered by epicardium (Ep), and the first forming pectinate muscle
(pm) protruding into the lumen. B: Cranial section of the same specimen
shows the myocardial character of the bulbus arteriosus, and the cushion-like character of the developing valves. Bⴕ: High power view of the
ventricular wall structure. The compact layer (Co) is two cell layers, while
the trabeculae (Tr) are single-cell thick. Arrowheads point to the ventriculo-bulbar valve leaflets. C: All chambers are encountered on the
sagittal section. Note the spouting of the trabeculation in the outer
curvature of the ventricle (partly filled with blood). Boxed areas show the
character of the ventriculo-bulbar (arrowheads, Cⴕ) and atrioventricular
(arrows, Cⴖ) valves. Hematoxylin-eosin staining. A, atrium; BA, bulbus
arteriosus; 夝, pericardial muscles; V, ventricle. Scale bar ⫽ 100 microns
in A–C; 10 microns in A⬘–C⬘, C⬙.
muscle, into which invaginated deep pockets of the endocardium. The valves attached directly to the myocardium
of the atrioventricular canal or the ventricular outflow.
The leaflets were thinned towards their edges, and were
entirely free without any support of the papillary muscles.
0.05 for each pressure curve). Ventricular peak systolic
pressure and end-diastolic pressure increased from 0.47 ⫾
0.09 mmHg and 0.08 ⫾ 0.07 mmHg, respectively at 5-days
post-fertilization to 2.49 ⫾ 0.25 mmHg and 0.27 ⫾ 0.10
mmHg, respectively, at 3-months post-fertilization adult
(Fig. 9). Ventricular peak systole was consistently higher
than the aortic systolic pressures throughout the developing stages, indicating a pressure gradient between the
ventricle and dorsal aorta of each stage. All peak systolic
and diastolic pressures in both ventricle and dorsal aorta
were linearly related to stage-specific values for embryo
weight (r ⱖ 0.962, P ⬍ 0.05 for each; Fig. 9). Cycle length
decreased with zebrafish development, and ranged from
571.2 ⫾ 34.5 ms at 5-days post-fertilization to 392.3 ⫾
27.8 ms at 3-months post-fertilization adult (Table 1).
Each of the cardiac time intervals (diastole and systole)
shortened during cardiac development.
Ventricular and Aortic Pressures
The ventricular pressure in the developing zebrafish
had a distinct systolic component, and two-components of
diastolic phase consisting of an early diastolic slow pressure rise and an end-diastolic accentuation reflecting the
atrial contraction (Fig. 8). The aortic pressure waveform
had also a systolic and diastolic portion, and contained a
dicrotic notch consistent with wave reflection from the
peripheral vascular system. Both ventricular systolic and
diastolic pressures and aortic peak pressure increased
geometrically with the developing stage (r ⱖ 0.7267, P ⬍
ZEBRAFISH HEART DEVELOPMENT
Fig. 4. Anti-myosin MF-20 staining (green) on the transverse sections
of a 2-weeks post-fertilization zebrafish heart. A: Section through the
caudal portion of the heart (compare with Fig. 3A) shows little progression from the previous stage. Note that the strongest signal is in the
ventricular trabeculations (Tr), while the staining in the atrium is faint.
B: Section through the cranial part of the heart shows positive MF-20
staining in the wall of the bulbus arteriosus. The signal in the pericardial
153
muscle (indicated by “夽”) is over-saturated to show the pattern in the
heart. C,D: High power views from sections adjacent to A and B,
respectively. Tissue autofluorescence (red) reveals the atrioventricular
(arrows) and ventriculo-bulbar (arrowheads) valves. A, atrium; BA, bulbus arteriosus; V, ventricle. Scale bar ⫽ 100 microns in A, B; 10 microns
in C, D.)
154
HU ET AL.
Fig. 5. Frontal (A) and sagittal (B) section through a 4-weeks zebrafish heart. A: Note the mature, bicuspid character of the ventriculobulbar valve (arrowheads), as well as prominent pectinate muscles (pm).
B: MF-20 staining (green) shows, as well as does the hematoxylin-eosin
staining in A, the regression of the myocardial phenotype of the bulbus
arteriosus. Note that the staining is stronger in the neighboring skeletal
muscle (indicated by “夽”) than in the ventricular trabeculations. Nuclei
(red) are counterstained with propidium iodide. A, atrium; BA, bulbus
arteriosus; V, ventricle. Scale bar ⫽ 100 microns.
DISCUSSION
These data are the first multidimensional analysis of
cardiac structure and function in the developing zebrafish.
Although the zebrafish heart develops rapidly in the early
embryonic stages, there is only a three-fold increase in the
ventricular weight in the young adult (4-weeks post-fertilization) when compared to a 24-hr (hatching) post-fertilization zebrafish. The linear log-log ventricular-to-body
weight ratio is similar to the heart-to-body ratio noted in
birds and mammals (Rakusan, 1984; Clark et al., 1986).
Relative ventricular weight is greatest in newborn, but
declines as the animal matures (Grande and Taylor,
1965). The relatively greater heart weight early in development accommodates the circulatory requirement of the
zebrafish. Vascular resistance is initially higher in early
heart development, but decreases as new resistance units
(vessels) are added to the circulation (Clark and Hu,
1982).
The early differentiation of the individual segments is
best demonstrated by specific antibody staining (Stainier
and Fishman, 1992; Fishman and Stainier, 1994), as the
usual morphological markers (trabeculations) are not developed yet. Similar to embryonic chick heart tube before
stage 16, the 48-hr post-fertilization zebrafish has no
valves, and forward flow of blood is maintained by apposition of the cardiac cushions composed mainly of cardiac
jelly. The formation of trabeculae in the ventricle occurs
during the first week, between 72 and 120-hr post-fertilization (http://zfish.uoregon.edu). The pectinate muscles in
the atrium follow closely the development of ventricular
trabeculation, being first apparent at 5-days post-fertilization and more prominent at 4-weeks post-fertilization.
The valves between the individual compartments appear
at 5-days post-fertilization, and mature in structure by
2-weeks post-fertilization.
The bulbus arteriosus has initially a myocardial character, similar to the situation in embryonic higher vertebrates, or adult elasmobranchs and teleosts (Rychter and
Rychterova, 1981; Van Mierop and Kutsche, 1984; Icardo
et al, 1999). The transition towards smooth muscle is
essentially completed by 4-weeks post-fertilization. There
is still substantial growth and thickening of this structure
later on, with development of deep-in-pockets of the endocardium. The bulbus arteriosus functions as a capacitor,
maintaining continuous blood flow into the gills arches.
Fine branches of epicardial vessels are strongly suggestive
of coronary arteries that supply nutrients to the wall from
the outside, so that it is not dependent on diffusion from
the lumen.
Initially, the myocardium is only one to two cells thick,
and does not exceed four cell layers in the outer compact
zone of the ventricle at 4-weeks post-fertilization. Eventually, in the adult, the compact layer is still composed of
only four cell layers, suggesting that there has been no
trabecular compaction as seen in fetal higher vertebrates.
The finding of a vascularized compact layer is surprising
in such a small fish, as it was believed that its presence is
associated in fish with larger body size and active metabolism (Ostadal et al., 1975; Santer and Greer Walker,
1980). We suggest a detailed re-examination of adult fish
reported to “lack” a compact layer, since both the compact
layer and epicardial vessels (coronary artery) are, in our
experience, easy to overlook. Probably, the only purely
“venous heart” can be found in invertebrates, such as
oyster, where extensive trabeculations are present in both
atrium and ventricle.
Conversely, the trabeculae stay only about two cells
thick, depending entirely on nutrition and oxygenation
from the ventricular lumen. A higher power view of adult
ventricular trabeculation shows substantial thickening in
comparison with earlier stages. However, a closer look
reveals that this may be due to cardiomyocyte hypertrophy, as the number of cell layers stays the same. More
intense anti-myosin staining in both trabeculae and the
pectinate muscles of the atria suggests more advanced
differentiation, similar to the embryonic chick. It supports
the hypothesis that the higher level of stress in the inner
side of the heart wall conditions and differentiation (Sedmera et al., 2000). Unlike the trabeculations of the higher
vertebrates, both atrial and ventricular trabeculations of
the zebrafish have more strut-like character, and are more
uniform without apparent regional differences (apical vs.
basal, left vs. right). Similarly, there is no indication of
formation of the papillary muscles. The organization of
the myocardium is, in part, determined by the hemody-
ZEBRAFISH HEART DEVELOPMENT
155
Fig. 7. Projection of seven confocal sections (1 micron apart) through
the ventricular wall of an adult (3-months post-fertilization) zebrafish. The
epicardial vessel (arrow) penetrates into the compact myocardium. Rhodamine-phalloidin staining for actin. Co, compact myocardium; Tr, trabeculae. Scale bar ⫽ 50 microns.
Fig. 8. Typical example of a ventricular and dorsal aortic pressure
waveform of a 5-days post-fertilization zebrafish. d, diastolic component; s, systolic component; 1, ventricular peak systolic pressure; 2,
ventricular end-diastolic pressure. Arrow indicates the dicrotic notch in
the dorsal aortic pressure.
Fig. 6. Sagittal (A–C) and frontal (D) sections through an adult (3months post-fertilization) zebrafish heart. A: All four chambers have their
unique characteristic shape and morphology. Note the abundant pectinate
muscles in the atrium. B: High power view of the atrium. While some areas
of the myocardium are single-cell thick, the pectinate muscles (pm) can
reach to three cell layers. Ep, epicardium. C: High power view of the
ventricular wall. Increase in cell size is apparent from comparison with
sections in earlier stages at the same magnification (100⫻). Compact myocardium (Co) reaches up to four cell layers. The trabeculae (Tr) are considerably thicker than at the earlier developing stages. Hematoxylin-eosin
staining. D: MF-20 (green) staining shows the continuity between the atrial
and ventricular myocardium, and loses the MF-20 staining in the bulbus
arteriosus and the valves (arrowhead points to ventriculo-bulbar valve, and
arrow to atrioventricular valve). Nuclei (red) are counterstained with propidium iodide. A, atrium; BA, bulbus arteriosus; SV, sinus venosus; V,
ventricle. Scale bars ⫽ 100 microns in A, D; and 10 microns in B, C.
namic forces interacting with the developing ventricular
wall.
The contour of the ventricular and aortic pressure
curves is markedly similar to that recorded from the avian
and mammalian heart. Pressure curves of the early developing stages are also similar to the mature heart. The
distinct early diastolic pressure suggests that the heart is
a valve-like structure, which prevents retrograde flow
during development. The accentuation in the end-diastolic
pressure is due to atrial systolic augmentation of ventricular filling. Ventricular peak pressure exceeds dorsal aortic peak pressure throughout the developing stages. There
are numerous resistance points that could create such a
pressure gradient; for instance, the resistance to flow
156
HU ET AL.
Fig. 9. Development of ventricular and dorsal aortic peak systolic
and end-diastolic pressures in the zebrafish. All pressures are linearly
related to age-specific values for body weight.
TABLE 1. Cycle length of the developing zebrafish
(express as mean ⴞ SEM)
Cycle Length (ms)
5-days post-fertilization
2-weeks post-fertilization
4-weeks post-fertilization
3-months post-fertilization
571.2 ⫾ 34.5
488.7 ⫾ 48.2
463.0 ⫾ 44.7
392.3 ⫾ 27.8
across the developing cushions between the ventricle and
the bulbus arteriosus (Hu and Keller, 1995). In addition,
the derivatives of four pairs of branchial arteries from the
ventral aorta could enhance the pressure drop in the dorsal aorta. There is a three-fold increase in ventricular
end-diastolic pressure during development, which defines
a decrease in ventricular compliance as the heart wall
thickens.
Changes in the shape of the embryonic heart during
passive filling reflect a balance of passive myocardial properties and filling forces, which can influence cardiac development (Icardo, 1989; Hu et al., 1991). The intraventricular pressure was negative at the termination of
isometric relaxation in all stages. While the exact atrioventricular pressure differential is not known for the zebrafish heart, a negative ventricular pressure at the onset
of ventricular filling is consistent with the presence of
ventricular diastolic suction (Robinson et al., 1986). Diastole begins after the ventricle reaches maximum systolic
elastance, and ventricular relaxation is associated with a
rapid fall in intraventricular pressure, which is followed
by ventricular filling (Sagawa et al., 1988). The embryonic
zebrafish heart fills under extremely low pressure (less
than 1 mmHg), and initial filling is likely assisted by
diastolic suction. The subsequent extent of ventricular
filling is also influenced by the material properties of the
heart, particularly by passive stiffness. Passive myocardial stiffness may change with remodeling of the myocardium that includes the increases in trabeculation and wall
thickness during development.
The decline in early diastolic pressure to near zero implies that wall stress changes markedly during the cardiac
cycle. Changing wall stress is likely an important extrinsic
factor in the regulation of myocardial formation. Wall
stress can be calculated using a variety of parameters
including myocardial properties, geometry, and internal
pressure. The geometry in the adult zebrafish heart could
be actually more complex than in the mature avian or
mammalian ventricle. The myocardial wall in the zebrafish ventricle has a thin compact myocardial layer and
varying trabecular density, length, and porosity (Sedmera
et al., 2000). While there have been estimates of average
wall stress in the early embryonic chick heart (Leatherbury et al., 1990), indices such as myocardial properties,
accurate geometry of trabecular orientation and density of
the developing zebrafish remain to be evaluated. At
present, we know that physiologic function responds to
changes in structure (strains), and probably changes in
structural surface forces (shears).
These detailed morphological and physiological descriptions of the developing zebrafish heart demonstrate that
its cardiovascular development are comparable to avian
and mammalian. Similar to cardiogenesis of higher vertebrates, the zebrafish develops an extensive trabeculation
in the early stage, which persists until adulthood. Epicardial vessels supply the outer compact layer of the ventricle. The bulbus arteriosus has initially a myocardial character, but it transforms into a smooth muscle structure.
The percentage of ventricular-to-body weight ratio decreases, while heart rate, ventricular, and dorsal aortic
blood pressures increase with development. These data
are essential to provide the framework for molecular
study, and a baseline for future characterization of cardiac
mutants in the zebrafish.
ACKNOWLEDGMENTS
We thank Jeffrey Essner, PhD, Patricia SolesRosenthal, and Chiffvon Stanley for their skillful technical assistance. This study was supported by an Innovative
Research Grant from Primary Children’s Medical Center
Foundation, Salt Lake City, Utah, and National Institutes
of Health (to D.S.). H.J.Y. is an Established Investigator of
the American Heart Association. MF20 antibody was obtained from the Developmental Studies Hybridoma Bank
developed under the auspices of the NICHD, and maintained by the University of Iowa, Department of Biological
Sciences, Iowa City, IA.
LITERATURE CITED
Burggren WW, Pinder AW 1991. Ontogeny of cardiovascular and
respiratory physiology in lower vertebrates. Annu Rev Physiol 53:
107–153.
Clark EB, Hu N 1982. Developmental hemodynamic changes in the
chick embryo from stage 18 to 27. Circ Res 51:810 – 815.
Clark EB 1990. Growth, morphogenesis and function: The dynamics
of cardiac development. In: Moller JH, Neal W, Lock J, editors.
Fetal, neonatal and infant heart disease. New York: AppletonCentury-Crofts, NY. p 3–23.
Clark EB, Hu N 1990. Hemodynamics of the developing cardiovascular system. Ann NY Acad Sci 588:41– 47.
Clark EB, Hu N, Dummett JL, Vandekieft GK, Olson C, Tomanek R
1986. Ventricular function and morphology in the chick embryo
stage 18 to 29. Am J Physiol 250:H407–H413.
Fishman MC, Stainier DY 1994. Cardiovascular development. Prospects for a genetic approach. Circ Res 74:757–763.
ZEBRAFISH HEART DEVELOPMENT
Fishman MC, Stainier DY, Breitbart RE, Westerfield M 1997.
Zebrafish: Genetic and embryological methods in a transparent
vertebrate embryo. Meth Cell Biol 52:67– 82.
Germroth PG, Gourdie RG, Thompson RP 1995. Confocal microscopy
of thick sections from acrylamide gel embedded embryos. Microsc
Res Tech 30:513–520.
Grande F, Taylor HL 1965. Adaptive changes in the heart, vessels,
and patterns of control under chronically high loads. In: Handbook
of physiology. Circulation. Washington, DC: Am Physiol Soc, sec 2,
vol III, chap. 74. p 2615–2678.
Hu N, Clark EB 1989. Hemodynamics of the stage 12 to stage 29 chick
embryo. Circ Res 65:1665–1670.
Hu N, Keller BB 1995. Relationship of simultaneous atrial and ventricular pressures in stage 16-27 chick embryos. Am J Physiol
269:H1359 –1362.
Hu N, Connuck DM, Keller BB, Clark EB 1991. Diastolic filling
characteristics in the stage 12 to 27 chick embryo ventricle. Pediat
Res 29:334 –337.
Icardo JM 1989. Endocardial cell arrangement: Role of hemodynamics. Anat Rec 225:150 –155.
Icardo, JM, Colvee E, Cerra MC, Tota B 1999. Bulbus arteriosus of the
Antarctic teleosts. II. The red-blooded Trematomus bernacchii.
Anat Rec 256:116 –126.
Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF 1995.
Stages of embryonic development of the zebrafish. Dev Dyn 203:
253–310.
Leatherbury L, Braden DS, Tomita H, Gauldin HE, Jackson WF
1990. Hemodynamic changes: Wall stresses and pressure gradients in neural crest-ablated chick embryos. Ann NY Acad Sci
588:305–313.
Lee RK, Stainier DY, Weinstein BM, Fishman MC 1994. Cardiovascular development in the zebrafish. II. Endocardial progenitors are
sequestered within the heart field. Development 120:3361–3366.
Manasek FJ 1970. Histogenesis of the embryonic myocardium. Am J
Cardiol 25:149 –168.
Ostadal B, Schiebler TH, Rychter Z 1975. Relations between development of the capillary wall and myoarchitecture of the rat heart. Adv
Exp Med Biol 53:375–388.
Pelster B, Burggren WW 1996. Disruption of hemoglobin oxygen
transport does not impact oxygen-dependent physiological processes in developing embryos of zebrafish (Danio rerio). Circ Res
79:358 –362.
157
Rakusan K 1984. Cardiac growth, maturation and aging. In: Zak R,
editor. Growth of the heart in health and disease. New York: Raven.
p 131–163.
Robinson TF, Factor SM, Sommenblick EH 1986. The heart as a
suction pump. Sci Am 254:84 –91.
Rychter Z, Rychterova V 1981. Angio- and myoarchitecture of the
heart wall under normal and experimentally changed conditions.
In: Pexieder T, editor. Perspectives in cardiovascular research, Vol.
5, mechanisms of cardiac morphogenesis and teratogenesis. New
York: Raven. p 431– 452.
Sagawa K, Maughan L, Suga H, Sunagawa K 1988. Historical overview. In: Cardiac contraction and the pressure-volume relationship.
New York: Oxford Press. p 8 –28.
Santer RM, Greer Walker MG 1980. Morphological studies on the ventricle of teleost and elasmobranch hearts. J Zool 190:259–272.
Sedmera D, Pexieder T, Vuillemin M, Thompson RP, Anderson RH
2000. Developmental patterning of the myocardium. Anat Rec 258:
319 –337.
Stainier DY, Fishman MC 1992. Patterning the zebrafish heart tube:
Acquisition of anteroposterior polarity. Dev Biol 153:91–101.
Stainier DYR, Fishman MC 1994. The zebrafish as a model system to
study cardiovascular development. Trends Cardiovasc Med 4:207–
212
Stainier DYR, Fouquet B, Chen JN, Warren KS, Weinstein BM,
Meiler SE, Mohideen MAPK, Neuhass SCF, Solnica-Krezel L,
Schier AF, Zwartkruis F, Stemple DL, Malicki J, Driever W, Fishman MC 1996. Mutations affecting the formation and function of
the cardiovascular system in the zebrafish embryo. Development
123:285–292.
Stainier DYR, Lee RK, Fishman MC 1993. Cardiovascular development in the zebrafish. II. Myocardial fate map and heart tube
formation. Development 119:31– 40.
Van Mierop LHS, Kutsche LM. 1984. Comparative anatomy and
embryology of the ventricles and arterial pole of the vertebrate
heart. In: Nora JJ, Takao A, editors. Congenital heart disease:
Causes and processes. New York: Futura. p 459 – 479.
Warren KS, Fishman MC 1998. “Physiological genomics”: Mutant
screens in zebrafish. Am J Physiol 275:H1–H7.
Weinstein BM, Fishman MC 1996. Cardiovascular morphogenesis in
zebrafish. Cardiovas Res 31:E17–E24.
Westerfield M 1995. The zebrafish book: A guide for the laboratory use
of zebrafish (Danio rerio). Eugene, OR: University of Oregon Press.