Intracardiac Blood Flow Patterns Related to the Yolk Sac Circulation of the Chick Embryo
Abstract Intracardiac flow patterns during heart development were studied by injection of india ink into the yolk sac circulation of chick embryos at Hamburger-Hamilton stages 10 to 17. We injected india ink into a small venule or capillary, carefully preventing application of overpressure to the vascular system, and recorded the intracardiac route by video. From stage 12 onward, blood flow was laminar, and separate intracardiac currents were visualized. The yolk sac was divided into a left and a right half. Blood coursed through each half in concentric loops, ranging from the marginal sinus to the sinus venosus. This parallel array persisted within the heart. Bilateral to the embryo, two lateral regions arose that extended wedgelike within each half, resulting in six equally sized yolk sac regions at stage 16. The process of heart looping was not accompanied by a change in flow pattern. However, developmental changes of the yolk sac circulation were reflected in alteration of the intracardiac flow pattern. From stage 16 onward, the intracardiac flow pattern was no longer determined by the left- or right-hand side of the yolk sac but by bilateral anterior, lateral, and posterior regions of the yolk sac. Blood from the lateral regions of the yolk sac was preferentially distributed to the head. The results show that in preseptation stages a relatively stable flow pattern is present. We suggest that alterations in blood flow could influence the process of normal heart development.
The effect of blood flow on heart development might be studied by manipulations at both the arterial and the venous pole of the heart. To explain anomalies caused by these changes in blood flow, extensive knowledge of normal flow patterns during embryonic development is necessary. The description of normal intracardiac flow patterns is rather contradictory.1 2 3 4 In early experiments, glass models of the heart were used in which each sinus horn was connected to a stream of fluid. Looping of the glass heart tube made the streams spiral around each other for 270°.1 5 6 By direct observation of moving blood cells in the embryonic heart, two spiral blood cell streams, separated from each other and the heart wall by cell-free areas, were described.1 2 7 8 9 An important role was attributed to these spiral streams in the formation of the atrial, ventricular, and aortopulmonary septa. It was hypothesized that septa arose between the separate streams because of lesser internal pressure.2 7 8 10 11 Furthermore, septation defects were explained as a result of abnormal spiral streams.6 12
In other experiments, dye indicators were injected into the cardiovascular system for the study of intracardiac flow patterns. Leyhane3 injected methylene blue into the main trunk of either the right or the left vitelline vein and visualized two currents in chick embryos of stages 17 to 19. Rychter and Lemez13 showed the distribution of blood from each vitelline vein through the pharyngeal arch arteries but did not describe intracardiac flow patterns. Yoshida et al4 also described two intracardiac routes that were complementary in two periods, namely stages 14 to 18 and 19 to 22. Although the injections in the study by Leyhane cover stages similar to those in the study by Yoshida et al, the intracardiac routes found by Leyhane from the omphalomesenteric vein to the atrioventricular canal were contrary to the results found by Yoshida et al, whereas the part from ventricle to truncus was similar. Only the experiments by Yoshida et al revealed a distribution of dye to just one side of the pharyngeal arch arteries. None of the experiments with the dye indicators, however, demonstrated spiral streams of 270°.3 4 13
Although experiments are described in which blood from the yolk sac circulation is visualized through the heart or pharyngeal arch arteries,3 4 13 none were performed by injection of dye into the smallest venules of particular yolk sac regions. A short pilot study on the distribution of blood from the yolk sac through the embryonic heart indicated that a large vein consists of a large number of currents resulting from the confluence of a number of capillaries. Therefore, direct injection of dye into a large vessel results in a great variance in labeling of intracardiac routes.
In the present study, normal blood flow patterns were examined by the india ink injection technique. India ink was injected into tributaries of the vitelline system, and the intracardiac flow pattern was studied. The relation between the continuously changing yolk sac vessels, which might lead to a varying flow pattern through the heart, and heart development was investigated.
Materials and Methods
Fertile White Leghorn eggs (Gallus domesticus) were incubated at 37°C and 60% to 70% relative humidity. The chick embryos were staged according to the age-determination criteria of Hamburger and Hamilton.14 Eighty-six embryos between stages 10 and 17 were carefully removed from their yolks; care was taken that the vitelline circulation was not damaged. Each embryo was placed in an agar-covered plastic Petri dish immersed in Locke solution (0.94% NaCl, 0.0045% KCl, and 0.004% CaCl2) at 37°C. Preliminary studies indicated that heart rate was not markedly changed after embryo removal. Moreover, careful microscopic observation did not reveal alterations in cardiac performance. We took utmost care to keep the embryos in optimal physiological conditions. A glass needle (tip diameter, 2 to 12 μm), produced by means of a horizontal pipette puller (Bachofer 462), was connected by a pressure-insensitive oil-filled tube to a Hamilton syringe, allowing careful and controlled administration of minute amounts of india ink (1:5 diluted in Locke solution and filtered through a 0.2-μm Millipore filter (Schleicher & Schuell, FP 030/3) to remove larger particles. The amount of solution was <0.002 μL, and administration lasted ≈30 seconds. This amount is expected to have no effect on hemodynamics.15
To immobilize the embryo and maintain an unobstructed vitelline circulation, the yolk sac was carefully stretched and adhered to the agar. India ink was injected into a capillary or small venule with a diameter of 10 to 75 μm. To avoid disturbing laminar flow patterns as observed under the microscope from the flow of erythrocytes, only slight manual pressure was applied to the microinjection system. The heart pulsations appeared to exert enough suction to empty the injection needle.
A limited number of embryos were injected in ovo to check the possibility of alterations in flow caused by removal of the embryo. These experiments in stages 16 and 17 showed that flow patterns in ovo and ex ovo were identical, proving that removal was not detrimental to the observed pattern.
The route of the india ink through the blood vessels and the heart was recorded with a video camera (Sony, DXC-151P) and a U-matic videocassette recorder (Sony, VO-5630). The flow was described at the following segments of the heart: sinus venosus, atria, atrioventricular canal, ventricle, conotruncus, and pharyngeal arch arteries. For the description of the intracardiac route from a ventral view, embryos older than stage 14 were turned in a position that allowed visualization of the complete heart. Performed carefully, this manipulation did not result in obstruction of the main vitelline vessels, which could have resulted in disturbance of the normal flow pattern. Moreover, some embryos were kept left side up to reveal the left-hand side of the pharyngeal arch arteries. For each segment of the heart it was determined whether the inner, central, or outer curvature of the heart was followed. Moreover, for each segment of the heart it was determined whether a certain current was localized ventrally, centrally, or dorsally. In “Results” this information is added in parentheses after the first description: eg, “blood coursed through the outer curvature of the ventricle and outflow tract (dorsal).”
Finally, micrographs were made of representative intracardiac currents from stills of the videotape by means of a computerized analyzer and scanner (Sony, Color Video Printer Mavigraph).
Development of the Yolk Sac Veins
The vascularization of the yolk sac is related to the development of the chick embryo and develops accordingly in an anteroposterior direction. At stages 10 to 13, the embryonic heart is tubular and starts looping. Blood is drained via the right and left anterior vitelline veins (Fig 1a⇓). From stages 14 to 15 onward, two lateral vitelline veins develop (Fig 2a⇓). At stage 16 the heart is completely looped. The proximal part of the posterior vitelline vein has developed. Blood is drained via the right and left anterior vitelline veins, the right and left lateral vitelline veins, and the posterior vitelline vein (Fig 3a⇓). At stage 17 the right anterior vitelline vein has disappeared, leaving one anterior vein, two lateral veins, and one posterior vitelline vein, the latter now completely developed (Fig 4a⇓).
Stages 10 to 13
Although heart contractions were seen from stage 10 onward, laminar blood flow was first noticed at stage 12. Fig 1⇑ represents a compilation of the intracardiac routes seen after injection of india ink into small vessels of either the right or left half of the yolk sac or two small lateral regions bilateral to the embryo.
Blood from the right half of the yolk sac drained via the right anterior vitelline vein. It coursed through the outer curvature of the heart and took the right first pair of pharyngeal arch arteries and the lateral part of the right dorsal aorta (Fig 5a⇓). Blood from the left half of the yolk sac drained via the left anterior vitelline vein. Within the embryo, the course was the mirror image of the route of blood from the right half of the yolk sac (Fig 5a⇓).
There was a relatively fast-expanding region in each half of the yolk sac, situated just caudal to the sinus venosus (compare Figs 1⇑, 2⇑, and 3⇑). This region encloses a capillary plexus that will give rise to the main stems of the future vitelline arteries. Blood that left the embryo via the capillary plexus coursed either through this small lateral region (★, ∗ in Fig 1⇑), whereafter it coursed centrally through the heart, or coursed more caudally, resulting in a route along the inner curvature (•) or the outer curvature (−) of the heart. Blood from this small lateral region was first visualized after a second passage from either the right or the left half of the yolk sac. From stage 13 onward it was technically possible to make injections directly into these regions. Blood entered the heart via the right or left sinus venosus and coursed centrally through the heart, the medial part of each dorsal aorta, and finally through the caudal part of the yolk sac. The route was continued via preferential long parallel loops to the anterior vitelline veins or the marginal sinus.
At stage 13 most experiments showed ink through both right and left pharyngeal arch arteries.
Stages 14 to 15
Fig 2⇑ shows a compilation of the intracardiac routes after injection into either the left or the right half of the yolk sac.
Although heart looping has progressed considerably, blood from the right half of the yolk sac (−) followed the same intracardiac route as at stage 13. It still drained via the right anterior vitelline vein, subsequently coursing through the inner curvature of the sinus venosus, atria, and atrioventricular canal and crossing in the ventricle to the outer curvature of the conotruncus (dorsal). Blood from the left half of the yolk sac (•) drained via the left anterior vitelline vein. By comparing this stage with the former stage we predicted that blood from the left anterior vitelline vein would course through the outer curvature of the primitive atrium and atrioventricular canal and through the inner curvature of the conotruncus (ventral). However, most experiments showed a route similar to that of the blood from the right half of the yolk sac (−) (Fig 5b⇑). The two smaller lateral regions extending wedgelike in each half of the yolk sac deserve special attention. Blood from these regions (★, ∗) coursed centrally through the sinus venosus, atria, atrioventricular canal, ventricle, and conotruncus.
All currents coursed through both first pharyngeal arch arteries, which formed the main pathway to the dorsal aorta. Blood from the left lateral yolk sac region (★) went preferentially to the head (3 cases of 3), whereas blood from the right lateral yolk sac region (∗) did not (only 1 case of 4).
In none of the experiments did we observe ink flowing through the left side of the atrioventricular canal, the future mitral ostium. This route was only taken during the second passage of blood and is of intraembryonic origin.
At stage 16 there was no longer a difference in flow pattern between blood from the right or the left side of the yolk sac. Instead, an anteroposterior difference was observed. Fig 3⇑ represents a compilation of the intracardiac routes after injection into the anterior, lateral, or posterior yolk sac regions.
Blood from the anterior regions of the yolk sac (−) entered the sinus venosus via the right or left anterior vitelline veins. Blood coursed through the inner curvature of the sinus venosus, atria, and atrioventricular canal and crossed in the ventricle to the outer curvature of the conotruncus (dorsal). In both early and late stage 16 embryos, the route continued via the second and third pharyngeal arch arteries.
Blood from the lateral regions of the yolk sac (★, ∗) entered the sinus venosus either via the proximal part of the anterior vitelline veins or the lateral vitelline veins and coursed centrally through the atria, atrioventricular canal (dorsal), ventricle, and conotruncus (ventral) (Fig 5c⇑). In the early stage 16 embryo, the route coursed cranially to the head via the first pharyngeal arch artery. Blood coursed along the posterior face of the first and second pharyngeal arch arteries to the dorsal aorta. In the late stage 16 embryo, the route continued to the head via the cranial part of either the first or the second pharyngeal arch artery. The route was continued in a caudal direction via the second pharyngeal arch artery. Blood from the lateral yolk sac regions coursed preferentially to the head of the embryo in 7 cases of 9.
Blood from the posterior regions of the yolk sac (•) entered the sinus venosus either via the proximal part of the lateral vitelline veins or the posterior vitelline vein and coursed centrally through the sinus venosus, atria, and atrioventricular canal (central) and crossed in the ventricle to the outer curvature (ventral) of the conotruncus (Fig 5d⇑). The route was continued caudally via either the first or the second pharyngeal arch artery. Sometimes (1 of 6 cases) this current went via the first pharyngeal arch artery to the head.
In none of the experiments did we observe india ink flowing through the future mitral ostium, as in stages 14 and 15.
Fig 4⇑ shows a compilation of the intracardiac routes after injection into the anterior, lateral, or posterior yolk sac regions.
At stage 17, blood from the anterior regions (−) of the yolk sac entered the sinus venosus via the (left) anterior vitelline vein and a minor part via the lateral vitelline veins and coursed through the inner curvature of the sinus venosus, atria, atrioventricular canal (ventral), ventricle (ventral), and the conotruncus (dorsal) and took the second and third pharyngeal arch arteries to the dorsal aorta (Fig 5e⇑). Compared with stage 16, this route has shifted from a position adjacent to the inner curvature of the inflow portion to a position adjacent to both inflow and outflow portions of the loop in stage 17. Blood from the lateral regions of the yolk sac (★, ∗) entered the sinus venosus via either the lateral vitelline veins or the proximal part of the posterior vitelline vein and coursed centrally through the sinus venosus and atria and then the atrioventricular canal (dorsal), ventricle, and conotruncus (central to ventral). This route continued in a cranial direction through the first and second pharyngeal arch arteries, whereas the second (caudal part) and third pharyngeal arch arteries were used for a caudal direction. Therefore, blood from the lateral yolk sac regions went preferentially to the head of the embryo (7 cases of 8).
Blood from the posterior regions of the yolk sac (•) entered the sinus venosus either via the proximal part of the lateral vitelline veins or the posterior vitelline vein and coursed through the inner curvature of the sinus venosus and through the atria and then followed either a central course (dorsally) or an outer curvature through the atrioventricular canal, ventricle, and conotruncus (Fig 5f⇑). The route was continued in a caudal direction via the second and third pharyngeal arch arteries. Compared with stage 16, there was an additional current through the future mitral ostium and the outer curvature of the ventricle. Either one of the two currents or both currents were observed in embryos.
At every stage a few experiments did not follow the general pattern. When the injected vessel exceeded a diameter of ≈100 μm, various routes could be observed, because larger vessels contain more currents. Also, when injections were made in the border zone between two yolk sac regions, the route of the adjacent region was sometimes observed. Finally, in some cases the flow pattern of a specific region was observed that would be expected for that region in the next developmental stage. From stage 16 onward, yolk sac regions were a more stable factor for intracardiac routes than were major vitelline veins. This was extremely apparent in some experiments in which india ink was injected into a capillary of a specific yolk sac region and then returned to the heart via two different vitelline veins. For instance, ink injected into the right lateral yolk sac region returned via both the right lateral vitelline vein and the posterior vitelline vein, whereafter the two laminar currents united in the sinus venosus and followed an intracardiac route in accordance with the lateral region.
Laminar flow depends on the relations between velocity, viscosity, and tube diameter.16 17 A minimal velocity is necessary to start laminar flow.9 16 The first heartbeats occur at stage 9.18 Before stage 12, blood flow is irregular with occasional backflow and mixing, and laminar flow can be observed from stage 12 onward. In the early embryo the values of velocity and viscosity of blood will never exceed the critical value for turbulence, which is expressed as a Reynolds number of >970±80.5 9 16 Blood flow remains laminar even in a pumping heart that contains irregularities in the inner wall such as trabeculae, endocardial cushions, and developing septa. Because blood flow is laminar in the stages studied, possible patterns can be investigated.
Because alterations of intracardiac flow patterns occur at the same time as major developmental changes of the yolk sac circulation, this changing yolk sac circulation could be responsible for the changing flow pattern. At stages 12 to 13, the left- and right-hand regions of the yolk sac differ in distribution of blood, leading to a pattern of parallel currents in the heart, and this left/right separation is continued in the pharyngeal arch arteries. The yolk sac develops in a cranial-to-caudal direction, resulting in a disappearance of the difference between left and right regions from anterior to posterior, eventually leading to six yolk sac regions. At stage 16 there is no longer a difference in distribution pattern between the left and right sides of the yolk sac. The development of the posterior vitelline vein results in a new additional intracardiac route. At stage 17, the right anterior vitelline vein has disappeared, resulting in a shift of the intracardiac route from the anterior yolk sac regions to a complete inner curvature of the heart. With further development of the posterior vitelline vein, the posterior intracardiac route changes too, resulting sometimes in an additional current through the outer curvature of the ventricle.
From stages 14 to 16 it is not possible to visualize blood from any yolk sac region coursing through the left side of the atrioventricular canal, the future mitral ostium. This means that exclusively blood from the embryo, which has a rather low oxygen tension,19 flows through the future mitral ostium. At stage 17, blood from the posterior regions of the yolk sac courses through the left side of the atrioventricular canal. The atrioventricular canal, which is originally circular, has to become mediolaterally oval before atrioventricular endocardial cushions can develop.20 The transition of low-oxygen blood to high-oxygen blood might interfere with differential growth of the atrioventricular canal, which results in an oval shape.
The right and left blood flows described by Bremer1 and Jaffee2 7 8 as spiraling around each other in the sinus venosus and in the outflow tract are not observed. Because we used a dye indicator, as did Rychter and Lemez,13 Leyhane,3 and Yoshida et al,4 it is possible to follow exactly the complete intracardiac route without the optical inaccuracy of watching moving blood cells.1 2 7 8 Leyhane and Yoshida et al made injections in large vitelline veins. We learned that this will result in great variance of results, which might explain why the intracardiac routes they found were only partially similar.
From stage 13 onward, injections made in either left or right yolk sac regions result in ink flow through both pharyngeal arch arteries. This is in contradiction to the flow pattern described by Yoshida et al,4 in which only two main bloodstreams were described, one exclusively through the right-hand side of the pharyngeal arch system and the other through the left-hand side. Like Rychter and Lemez,13 we observe a distribution through the pharyngeal arch arteries more delicate than only a left and right current.
The filling of capillaries in the head of the embryo with black india ink is evident. Therefore, it is possible to study preferential flow to the head of the embryo. From stage 14 onward, india ink injected into the lateral regions of the yolk sac goes preferentially to the head. The most cranial part of the first pair of pharyngeal arch arteries is used for transporting blood to the head. With increasing age, the contribution of the second pair of pharyngeal arch arteries to transportation of blood specifically to the head increases. The fact that blood from the lateral yolk sac regions has a specific distribution to the head could be of interest for long-term manipulation of specific veins that drain mainly the lateral part of the yolk sac, perhaps leading to congenital malformations of the head or face. Orts Llorca et al21 found that cauterization of the left lateral vitelline vein at stage 16 resulted in left microphthalmias.
Our experiments show that blood from specific yolk sac regions follow a specific stage-dependent intracardiac route. The development of the vitelline vessels is related to the intracardiac flow pattern. We assume that this change in intracardiac flow pattern is probably important for normal heart development (authors’ unpublished data, 1995). Interfering with the normal development of intracardiac flow by manipulating the venous inflow of the heart may result in abnormal heart development. In vitro development of the straight heart tube revealed that the first looping process of the heart occurs irrespective of blood flow.22 23 However, cardiac jelly proliferates without limit and obstructs the whole cardiac lumen in the absence of blood flow,21 indicating the need for blood flow for the precise molding of the cardiac jelly into cardiac cushions. Future studies will deal with the role of specific intracardiac currents in the development of the heart by changing particular flow patterns after ligating specific vitelline veins.
This work was supported by grant 900-516-096 of the Netherlands Heart Foundation. Bas Blankevoort is acknowledged for preparing the illustrations.
- Received August 4, 1994.
- Accepted December 29, 1994.
- © 1995 American Heart Association, Inc.
Leyhane JC. Visualization of blood streams in the developing chick heart. Anat Rec. 1969;163:312-313.
Yoshida H, Manasek F, Arcilla RA. Intracardiac flow patterns in early embryonic life: a reexamination. Circ Res. 1983;53:363-371.
Romhányi G. Über die Rolle Haemodynamischer Factoren im Normalen und Pathologischen Entwicklungsvorgang des Herzens. Acta Morphol Hung. 1952;2:297-312.
DeVries PA, Saunders JB de JM. Development of the ventricles and spiral outflow tract in the human heart. Contrib Embryol. 1962;37:87-114.
Jaffee OC. Hemodynamic analysis of experimentally produced cardiac malformations. Anat Rec. 1966;154:509. Abstract.
Wagman AJ, Hu N, Clark EB. Effect of changes in circulating blood volume on cardiac output and arterial and ventricular blood pressure in the stage 18, 24, and 29 chick embryo. Circ Res. 1990;67:187-192.
Coulter NA, Pappenheimer JR. Development of turbulence in flowing blood. Am J Physiol. 1949;159:401-408.
Little RC, Little WC. Hemodynamics. In: Physiology of the Heart and Circulation. Chicago, Ill: Year Book Medical Publishers, Inc; 1989:219-234.
Steding G, Seidl W. Contribution to the development of the heart, part I: normal development. Thorac Cardiovasc Surg. 1980;20:386-409.