SEARCH

SEARCH BY CITATION

Keywords:

  • heart development;
  • common pulmonary vein;
  • atrial septation;
  • Xenopus laevis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The heart of lung-breathing vertebrates normally shows an asymmetric arrangement of its venoatrial connections along the left-right (L-R) body axis. The systemic venous tributaries empty into the right atrium while the pulmonary venous tributaries empty into the left atrium. The ways by which this asymmetry evolves from the originally symmetrically arranged embryonic venous heart pole are poorly defined. Here we document the development of the venous heart pole in Xenopus laevis (stages 40–46). We show that, prior to the appearance of the mouth of the common pulmonary vein (MCPV), the systemic venous tributaries empty into a bilaterally symmetric chamber (sinus venosus) that is demarcated from the developing atriums by a circular ridge of tissue (sinu-atrial ridge). A solitary MCPV appears during stage 41. From the time point of its first appearance onwards, the MCPV lies cranial to the sinu-atrial ridge and to the left of the developing interatrial septum and body midline. L-R lineage analysis shows that the interatrial septum and MCPV both derive from the left body half. The CPV, therefore, opens from the beginning into the future left atrium. The definitive venoatrial connections are established by the formation of a septal complex that divides the lumen of the venous heart pole into systemic and pulmonary venous flow pathways. This complex arises from the anlage of the interatrial septum and the left half of the sinu-atrial ridge. Developmental Dynamics 240:1518–1527, 2011. © 2011 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The venous pole of the mature heart of lung-breathing vertebrates normally shows remarkable asymmetries with respect to the left-right (L-R) body axis. The systemic venous tributaries, for example, normally empty exclusively into the right atrium while the pulmonary venous tributaries normally empty exclusively into the left atrium. This asymmetry is of vital importance for lung-breathing vertebrates since it contributes to the separation of oxygenated and de-oxygenated blood. It might, thus, be surprising to note that, despite more than 200 years of research on the development of the vertebrate heart, we have currently no commonly accepted concept about the way in which the initially almost symmetrically arranged venous pole of the embryonic heart normally becomes transformed into the asymmetric arrangement of the mature heart. Especially the way in which the pulmonary venous tributaries normally become connected exclusively to the left atrium has been a matter of some debate during the past 100 years.

The pulmonary vasculature arises from an endothelial vascular plexus surrounding the embryonic foregut (Brown,1913; Rammos et al.,1990; DeRuiter et al.,1993). This plexus initially has no direct connections to the developing heart but drains its blood into the umbilical, vitelline, and cardinal venous systems. Later in development, the pulmonary portion of the vascular foregut plexus becomes directly connected to the venous heart pole via a solitary vein stem, which is called the common pulmonary vein (CPV). In human beings and some other mammals, the CPV is only transiently present since it becomes incorporated into the left atrium until its four tributaries empty separately into the left atrium.

Three aspects of CPV development are differently described in the literature. The first is the mechanism of formation of this blood vessel. In the vast majority of studies, the CPV is described as originating from an endothelial sprout that grows out from the dorsal wall of the venous heart pole and reaches the pulmonary portion of the vascular foregut plexus via the dorsal mesocardium (Flint,1906; Fedorow,1910; Neill,1956; Van Praagh and Corsini,1969; Los,1978; Kutsche and Van Mierop,1988; Webb et al.,1998,2000,2001; Männer and Merkel,2007). In a few studies, on the other hand, descriptions are given that suggest that the CPV may derive from endothelial strands, vacuoles, or channels that form in situ within the dorsal mesocardium (Brown,1913; Robertson,1914; Davies and MacConaill,1937; Bertens et al.,2010).

The second unsolved aspect of CPV development is the initial position of its mouth with respect to the systemic venous system and the developing atriums. The venous pole of embryonic vertebrate hearts is traditionally said to consist of two morphological sub-components: (1) a confluence chamber for the systemic veins, called the “sinus venosus”; and (2) the initially undivided “primitive atrium,” which lies downstream from the sinus venosus. The classical concept of CPV formation sees the initial position of the future mouth of the CPV (MCPV) already within the dorsal wall of the primitive atrium. Its later confinement to the left atrial cavity is explained mainly by the formation of the primary interatrial septum (Born,1889; Neill,1956; Van Praagh and Corsini,1969; Goor and Lillehei,1975; Los,1978; Webb et al.,1998,2000,2001; Soufan et al.,2004; Anderson et al.,2006; Männer and Merkel,2007; Bertens et al.,2010; Sizarov et al.,2010). On the other hand, there are data, which seem to suggest that the initial position of the MCPV is within the dorsal wall of the sinus venosus (Goette,1875; His,1887; Flint,1906; Hochstetter,1906; Fedorow,1910; Brown,1913; Buell,1922; Dor et al.,1987; Kutsche and Van Mierop,1988; DeRuiter et al.,1995; Tasaka et al.,1996; Blom et al.,2001). The connection of the CPV to the developing left atrium is explained by a positional shift of its mouth combined with the formation of the primary interatrial septum. It should be noted here that some researchers have presented data that seem to cast doubt on the existence of the sinus venosus as a discrete morphological segment in the embryonic heart of higher vertebrates. The systemic and pulmonary venous tributaries are both said to empty directly into the primitive atrium from the onset of atrial development (Webb et al.,1998,2000; Soufan et al.,2004; Anderson et al.,2006).

The third unsolved aspect of CPV development is the initial position of its mouth with respect to the L-R body axis. Several authors report that the MCPV is originally located in the body midline and, therefore, is initially connected to the left as well as to the right half of the venous heart pole (Brown,1913; Davies and MacConaill,1937; Auër,1948; Van Praagh and Corsini,1969; Bliss and Hutchins,1995; Webb et al.,1998; Wessels et al.,2000; Jongbloed et al.,2004). Its connection to the left atrium is explained by the formation of the primary interatrial septum, which is said to form along the right side of the MCPV. On the other hand, we can find several studies in which the initial position of the MCPV was already found within the left half of the venous heart pole (His,1887; Born,1889; Flint,1906; Hochstetter,1906; Fedorow,1910; Robertson,1914; Neill,1956; Goor and Lillehei,1975; Los,1978; Tasaka et al.,1996). An interesting variant of the latter theme may exist in lizard, alligator, and chick embryos. Here some researchers have found a bilateral pair of endothelial evaginations at the dorsal wall of the venous heart pole of which the right one disappears while the left one grows out into the dorsal mesocardium to form the CPV (Fedorow,1910; Dor et al.,1987; Kutsche and Van Mierop,1988; Männer and Merkel,2007).

While much of the recent research activities on pulmonary vein development have focused on the question of whether the CPV is initially connected to the sinus venosus or to the primitive atrium (DeRuiter et al.,1995; Tasaka et al.,1996; Blom et al.,2001; Webb et al.,1998;2000;2001; Soufan et al.,2004; Anderson et al.,2006), only little attention has been paid to the questions of whether and how the topogenesis of the MCPV is mechanistically linked to the specification of the L-R body axis. This is surprising since it is a well-known fact that human congenital cardiac malformations with bilaterally symmetric arrangement of the atrial appendages, so-called left and right atrial isomerism, are frequently associated with abnormal connections of the systemic and pulmonary venous tributaries (Uemura et al., 1995; Smith et al.,2006). During the past few years, genetic mouse models for left as well as right atrial isomerism have been used to analyze the role of L-R axis specification in the development of the venous heart pole (Liu et al.,2002; Anderson et al.,2004; Weninger et al.,2005; Hildreth et al.,2009). Analyses of Pitx2c-deficient mice, for example, have shown that this transcription factor, which normally is expressed only in the left half of the heart tube (Franco et al.,2000; Campione et al.,2001), suppresses a default program of sinus node formation in the left atrium (Mommersteeg et al.,2007). Unfortunately, up to now, analyses of such mutant mice have not clarified how the topogenesis of the MCPV is mechanistically linked to the L-R axis specification. During the past 15 years, studies on non-mammalian model organisms such as the chick, the frog Xenopus laevis, and the zebrafish have revolutionized our knowledge about the developmental basis of the specification of the L-R body axis (for recent reviews see Ray and Belmonte,2008; Blum et al.,2009; Vandenberg and Levin, 2010). We, therefore, think that it might be helpful if the current research on the development of the venous heart pole was not focused primarily on mammalian models but would additionally include studies on non-mammalian models. The frog X. laevis is a well-established model organism in the study of early vertebrate heart development (Lohr and Yost,2000; Mohun et al.,2000; Warkman and Krieg,2006) as well as in the study of L-R axis specification (Ramsdell,2005,2006; Blum et al.,2009). X. laevis is a lung-breathing vertebrate that survives experimental perturbations of L-R axis specification up to late stages of organogenesis and, therefore, may be well suited to experimental studies exploring the role of L-R axis specification in the development of the venous heart pole. Unfortunately, however, information on the development of its venous heart pole is lacking in the scientific literature up to now. In the present study, we have analyzed the development of the venous heart pole in X. laevis, for the first time. We have focused our study on the documentation of the development of the systemic and pulmonary venoatrial connections, thereby aiming to answer the following questions: (1) Does the CPV of X. laevis arise from an endothelial sprout growing out from the dorsal wall of the venous heart pole or does it derive from an endothelial channel forming in situ within the dorsal mesocardium? (2) Does a sinus venous exist as a discrete morphological segment in the embryonic X. laevis heart? (3) Is the CPV of X. laevis initially connected to the sinus venosus or to the primitive atrium? (4) Is the MCPV of the X. laevis heart an original midline structure, which would mean that its wall is formed in equal parts by material from the left as well as the right body halves, or is the MCPV derived from the left-body half? With this study, we introduce X. laevis as a promising model organism for future studies aimed at exploring the role of L-R axis specification in the development of the venous heart pole.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The basic morphological design from which the definitive morphology of the venous heart pole evolves is achieved at stages 40/41 (Fig. 1). At these stages, the venous pole of the X. laevis heart is seen as a bilaterally almost symmetric cavity consisting of two sub-components: (1) the sinus venosus, which lies at the bottom of the pericardial cavity and is connected with the systemic venous tributaries; and (2) the primitive atrium, which lies cranial (downstream) to the sinus venosus. The border between the sinus venosus and the primitive atrium is internally marked by a circular ridge of tissue and is externally marked by a corresponding circular furrow. In the following, these anatomical landmarks will be called the sinu-atrial ridge and sinu-atrial furrow, respectively. The plane defined by these circular landmarks shows a striking dorso-caudal tilt with respect to the transverse body plane (Fig. 1C). In contrast to reptilian, avian, and mammalian embryos, a dorsal mesocardium is found only at the dorsal wall of the sinus venosus and not at the dorsal wall of the primitive atrium (Fig. 1C). This peculiar finding corresponds to earlier observations made in embryos of the frog Rana temporaria and the newt Triton taeniatus (Fedorow,1910) and, therefore, might be a specific feature of amphibian hearts.

thumbnail image

Figure 1. A, B: These SEM micrographs, viewed frontally, show the dorsal wall of the venous heart pole from a stage-40/41 Xenopus laevis embryo. The venous pole is seen as a bilaterally almost symmetric cavity consisting of two sub-components: (1) the sinus venosus and (2) the primitive atrium. The border between the sinus venosus and the primitive atrium is internally marked by a circular ridge of tissue (dotted line in B) and is externally marked by a corresponding circular furrow (arrows in B). These anatomical landmarks are called the sinu-atrial ridge and sinu-atrial furrow, respectively. C: Sagittal histological section through the heart of a stage-41 Xenopus laevis embryo. Arrowheads point to the ventral and dorsal halves of the sinu-atrial ridge. Note that the plane defined by the sinu-atrial ridge shows a dorso-caudal tilt with respect to the transverse body plane. Note also that the dorsal mesocardium is present only at the level of the sinus venosus. This finding corresponds to earlier observations made in embryos of the frog Rana temporaria and the newt Triton taeniatus (Fedorow,1910). A, primitive atrium; L, liver; LS, left sinus horn; O, outflow tract; P, pulmonary anlage; RS, right sinus horn; SV, sinus venosus; V, primitive ventricle.

Download figure to PowerPoint

Up to the early stage 41, there are no visible signs for the anlage of the interatrial septum. Thus, the primitive atrium is initially seen as a common, undivided chamber whose left and right halves are both connected to the systemic venous system via the left and right halves of the sinus venosus, respectively. Morphological evidence of the presence of single or bilaterally paired endothelial evaginations representing the anlage of the MCPV is also lacking before stage 41.

At stage 41, a single slit-shaped opening representing the MCPV was found at the dorsal wall of the venous heart pole in 50% of the embryos (4/8) examined by SEM (Fig. 2). At stage 42 and older, the MCPV was found in every embryo examined (see Figs. 5A–C and 6A–C). Histological sections show that, before the appearance of this opening, a single blind-ending vessel representing the CPV is found within the mesenchyme of the dorsal wall of the sinus venosus close to the embryonic body midline (Fig. 3). At the time point of its first appearance at stages 41 or 42, the MCPV lies to the left of the embryonic body midline and cranial to the sinu-atrial ridge and, therefore, opens into the left half of the primitive atrium (Figs. 2A–D and 4A, D). At this stage, the MCPV is caudally guarded by a tiny endocardial fold whose free edge follows the course of the sinu-atrial ridge (Fig. 2A–D). In the following text, this fold will be called the sinu-pulmonary fold since it separates the MCPV from the sinus venosus. On the left side of the MCPV, the free edge of the sinu-pulmonary fold is continuous with the relatively sharp left-hand portion of the sinu-atrial ridge while on the right side of the MCPV the free edge of the sinu-pulmonary fold is continuous with the relatively shallow right-hand portion of the sinu-atrial ridge. Also at this location, the right-hand end of the sinu-pulmonary fold abuts on the caudal end of a shallow, almost vertically oriented ridge of tissue, which represents the anlage of the interatrial septum (Fig. 2A–D). This ridge of tissue lies near to the embryonic body midline and will be called the interatrial ridge in the following text. Histological sections show that the interatrial ridge forms as a consequence of striking differences in the structure of the dorsal walls of the future left and right atriums. In the dorsal wall of the future right atrium, a thick layer of extracellular matrix (cardiac jelly) is interposed between the myocardium and endocardium while, in the dorsal wall of the left atrium, endocardium and myocardium are closely apposed to each other (see Fig. 6).

thumbnail image

Figure 2. A–D: These SEM micrographs, viewed frontally, show the topography of the MCPV within the dorsal wall of the venous heart pole at the time point of its first appearance at stage 41. A,B: Low-magnification views showing the whole venous pole. C,D: Higher magnification views of the border between the sinus venosus and primitive atrium. At the time point of its first appearance, the MCPV (arrow) is seen as a slit-shaped opening. This opening is caudally guarded by a tiny endocardial fold, which we call “the sinu-pulmonary fold.” The free edge of the sinu-pulmonary fold (+++) follows the course of the sinu-atrial ridge (…). The right-hand end of the sinu-pulmonary fold is continuous not only with the right-hand portion of the sinu-atrial ridge but additionally abuts on the caudal end of the vertically oriented anlage of the interatrial septum (ooo), which we call the “interatrial ridge.” The MCPV, therefore, lies within the dorsal wall of the future left atrium. Abbreviations as in Figure 1.

Download figure to PowerPoint

thumbnail image

Figure 3. This series of subsequent histological sections (transverse plane) through the sinus venosus of a stage-41 X. laevis embryo shows a blind-ending CPV (*) within the dorsal mesocardium and dorsal wall of the sinus venosus. Sections are presented in caudal-cranial order (thickness is 7 μm). Note that the CPV has no open connection with the lumen of the venous heart pole. FG, foregut; other abbreviations as defined in Figure 1.

Download figure to PowerPoint

thumbnail image

Figure 4. These SEM micrographs, viewed frontally, show the topography of the MCPV (arrow) within the dorsal wall of the venous heart pole at subsequent stages of development (A, D: stage 41; B, E: stage 42; C, F: stage 46). A–C: The position of the MCPV in relation to the sinus venosus and the developing left and right atriums. D–F: The position of the MCPV in relation to the body midline (indicated by white lines). Note that from the time point of its first appearance at stage 41 and onwards, the MCPV lies to the left of the embryonic body midline and within the dorsal wall of the developing left atrium. Note also that the interatrial septum lies within the median body plane. This corresponds to the situation found in the adult X. laevis heart (De Graaf,1957). Abbreviations as defined in Figures 1 and 3.

Download figure to PowerPoint

thumbnail image

Figure 5. These frontal and cranial SEM views into the opened atriums of X. laevis hearts show subsequent steps in the development of the interatrial septum (A, D: stage 42; B, E: stage 43; C, F: stage 46). A, D: At stage 42, the free edge of the sinu-pulmonary fold (+ + +) shows a slightly oblique course. Its right-hand end is directly continuous with the interatrial ridge (***) while its left-hand end is continuous with the left-hand portion of the sinu-atrial ridge. During stages 43 (B, E) to 46 (C, F), the sinu-pulmonary fold more and more forms a unit with the rapidly growing interatrial ridge and, thereby, contributes to the formation of the interatrial septum. The left-hand end of the “interatrial ridge–sinu-pulmonary fold unit” grows along the left-hand portion of the sinu-atrial ridge towards the dorsal atrio-ventricular endocardial cushion. This not only leads to the separation of the left and right atrium but, additionally, leads to the separation of the left atrium from the sinus venosus. As the final consequence, the sinus venosus opens exclusively into the right atrium while the common pulmonary vein opens exclusively into the left atrium (arrows point to the MCPV). DAV, dorsal atrio-ventricular endocardial cushion; other abbreviations as defined Figures 1 and 3.

Download figure to PowerPoint

At stage 42, the free edge of the sinu-pulmonary fold no longer shows its initially horizontal course but rather a slightly oblique course (Fig. 5A,D). This is because its right-hand end is no longer continuous with the right-hand portion of the sinu-atrial ridge, but has now become directly continuous with the interatrial ridge (Fig. 5A,D). The left-hand end of the sinu-pulmonary fold, in contrast, is still continuous with the left-hand portion of the sinu-atrial ridge (Fig. 5A,D).

thumbnail image

Figure 6. This transverse histological section through the heart of a stage-43 X. laevis embryo shows the difference in the structures of the dorsal walls of the developing left and right atriums leading to the presence of the interatrial ridge (arrow). In the dorsal wall of the right atrium, a thick layer of extracellular matrix (cardiac jelly) is interposed between the myocardium and endocardium while, in the dorsal wall of the left atrium, endocardium and myocardium are closely apposed to each other. Abbreviations as defined in Figures 1 and 3.

Download figure to PowerPoint

During stages 43 to 46, the course of the free edge of the sinu-pulmonary fold shifts from a slightly oblique to an almost vertical direction (Fig. 5B,C,E,F). The sinu-pulmonary fold more and more forms a unit with the rapidly growing interatrial ridge and, thereby, contributes to the formation of the interatrial septum. The left-hand end of the “interatrial ridge–sinu-pulmonary fold unit” grows along the left-hand portion of the sinu-atrial ridge towards the dorsal atrio-ventricular endocardial cushion. Consequently, the formation and growth of the “interatrial ridge–sinu-pulmonary fold unit” not only leads to the separation of the left from the right atrium but, additionally, leads to the separation of the left atrium from the sinus venosus. As the final consequence, the sinus venosus opens exclusively into the right atrium while the CPV empties exclusively into the left atrium (Fig. 5C,F). A further consequence of the growth of the sinu-pulmonary fold is the transformation of the MCPV from an initially slit-shaped structure into a funnel-shaped structure connecting the CPV with the left atrium (Fig. 7A–C). This pulmonary funnel may correspond to the so-called “pulmonary recess” in the mature heart of the frog Rana tigrina (Sharma,1957).

thumbnail image

Figure 7. A–C: These sagittal histological sections through the hearts of X. laevis embryos show subsequent steps in the development of the sinu-pulmonary fold and MCPV (A, stage 41; B, stage 43; C, stage 46). A: At stage 41, the sinu-pulmonary fold separates the CPV (*) from the sinus venosus. Its free edge (red arrowhead) forms a small portion of the sinu-atrial ridge whose ventral part is marked by the green arrowhead (blue arrowhead points to the dorsal part of the sinu-atrial furrow). B, C: During subsequent stages of development, the sinu-pulmonary fold grows towards the dorsal AV-cushion and, thereby, forms a septum that separates the sinus venosus from the left atrium (area between the red arrowheads). The MCPV becomes a funnel-shaped structure that lies dorsal to the sinus venosus and connects the CPV (*) with the left atrium. AV, atrio-ventricular canal; other abbreviations as defined previously in figure legends.

Download figure to PowerPoint

The facts that, at the time point of its first appearance and onwards, the MCPV lies to the left of the embryonic body midline (Fig. 4D–F) and to the left of the future interatrial septum (Figs. 2A–D and 5D–F) suggest that the MCPV of X. laevis embryos may derive from the left body half as suggested by previous data from amphibian, reptilian, and avian embryos (Hochstetter,1906; Fedorow,1910; Dor et al.,1987; Kutsche and Van Mierop,1988; Männer and Merkel,2007). Due to its close proximity to the embryonic body midline, however, we could not clearly rule out the possibility that the MCPV initially forms as a “midline structure” as suggested by data from mammalian embryos (Brown,1913; Davies and MacConaill,1937; Auër,1948; Van Praagh and Corsini,1969; Bliss and Hutchins,1995; Webb et al.,1998; Wessels et al.,2000; Jongbloed et al.,2004). An ideal solitary midline structure should be a bilaterally symmetric structure that is built up in equal parts by material from the left as well as the right body halves. To clarify whether the MCPV is of bilateral or of unilateral left side origin, we have carried out L-R lineage tracing experiments by injecting dextran-conjugated fluorescent dyes into the blastomeres at the two-cell stage. The results of these experiments show that the interatrial septum and the walls of the MCPV derive from the left body half (Fig. 8A–C).

thumbnail image

Figure 8. A–C: These transverse histological sections show the left-right (L-R) lineage composition of the heart of a stage-46 X. laevis embryo. Sections are presented in cranio-caudal order. Left and right lineages were labelled with Oregon Green 488-conjugated dextran (green) and Alexa 647-conjugated dextran (red), respectively. Green labelling of the interatrial septum (arrowhead) and the wall of the MCPV (*) indicate their origin from the left lineage. Abbreviations as defined previously in other figure legends.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

In the present study, we have analyzed, for the first time, the development of the venous pole of the heart in the frog X. laevis. We have documented several aspects of the development of the venoatrial connections, which will be discussed in the following paragraphs.

Mechanism of CPV Formation

The first aspect is the mechanism of formation of the common pulmonary vein (CPV). In the majority of reports on previously analyzed species, the CPV was said to arise from an endothelial sprout, sometimes called the “pulmonary pit,” that grows out from the dorsal wall of the venous heart pole and reaches the developing intra-pulmonary vasculature via the dorsal mesocardium (Flint,1906; Fedorow,1910; Neill,1956; Van Praagh and Corsini,1969; Los,1978; Kutsche and Van Mierop,1988; Webb et al.,1998,2000,2001; Männer and Merkel,2007). In the present study, we did not find any morphological evidence for such an endothelial evagination before the appearance of the definitive MCPV at stage 41. Instead, a single blind-ending endothelial channel (the future CPV) was found within the mesenchyme of the dorsal wall of the sinus venosus close to the embryonic body midline (Fig. 3). This suggests that in X. laevis, the CPV does not form by sprouting (angiogenesis) from the endothelial lining of the venous heart pole but rather forms in situ within the mesenchyme (vasculogenesis) of the dorsal wall of the sinus venosus and dorsal mesocardium. It might be possible that this feature is specific for X. laevis. Data from lungfishes, cats, humans, and turtles, however, suggest that in situ formation of blood vessels (vasculogenesis) might contribute to CPV formation to a much larger extent than previously thought (Brown,1913; Robertson,1914; Davies and MacConaill,1937; Bertens et al.,2010).

Position of the MCPV With Respect to the Systemic Venous Tributaries and the Primitive Atrium

Much of the recent research activities on pulmonary vein development have focused on the question of whether the CPV is initially connected to a confluence chamber for the systemic venous tributaries, traditionally called the “sinus venosus,” or to the primitive atrium (DeRuiter et al.,1995; Tasaka et al.,1996; Blom et al.,2001; Webb et al.,1998,2000,2001; Soufan et al.,2004; Anderson et al.,2006; Männer and Merkel,2007). In this context, whether a venous confluence chamber really exists as a discrete morphological segment in the heart of higher vertebrate embryos before the appearance of the MCPV has been questioned (Webb et al.,1998,2000; Soufan et al.,2004; Anderson et al.,2006). In the present study, we have found that prior to the appearance of the MCPV, the systemic venous tributaries of X. laevis embryos empty into a bilaterally almost symmetric chamber that is internally demarcated from the developing atriums by a circular ridge of tissue (Fig. 1). The venous pole of the embryonic X. laevis heart obviously has a morphologically identifiable confluence chamber for the systemic venous tributaries. We have, therefore, named this chamber the “sinus venosus” and named its internal border with the primitive atrium the “sinu-atrial ridge.” We have, furthermore, found that, from the time point of its first appearance onwards, the MCPV lies cranial (downstream) to the sinu-atrial ridge (Fig. 2) and, therefore, is directly connected to the atrial heart segment from the beginning. This finding corresponds to previous reports on avian (chick) and mammalian (mouse, human) embryos (Webb et al.,1998,2000,2001; Soufan et al.,2004; Anderson et al.,2006; Männer and Merkel,2007) but differs from all previous reports on amphibian embryos in which the initial position of the MCPV was seen within the dorsal wall of the cranialmost portion of the sinus venosus (Goette,1875; Hochstetter,1906; Fedorow,1910). How can this discrepancy be explained? A general problem in the assignment of the MCPV to either the sinus venosus or the primitive atrium is its position close to the sinu-atrial border. This opens a window for divergent interpretations of the morphological situation if the sinu-atrial border (sinu-atrial ridge) cannot be clearly identified in the specimens under investigation. Previous studies on frog and other amphibian embryos were primarily carried out on 2-dimensional histological sections (Goette,1875; Hochstetter,1906; Fedorow,1910) and not on 3-dimensional specimens as in the present study. Histological sections, especially those in the transverse body plane, do not facilitate an easy identification of the sinu-atrial ridge. We, therefore, speculate that the previous assignments of the MCPV to the dorsal wall of the cranialmost portion of the sinus venosus might be “artifacts of 2-dimensional histology.” This speculation is, furthermore, based on the fact that the plane defined by the sinu-atrial ridge shows a striking dorso-caudal tilt with respect to the transverse body plane of X. laevis embryos (Fig. 1C). As a consequence, transverse histological sections through the cranial portion of the sinus venosus will not only show the lumen of the sinus venosus but will also show the caudalmost portion of the dorsal wall of the primitive atrium. This situation may be interpreted as evidence that the caudalmost portion of the dorsal atrial wall belongs to the sinus venosus. Divergent interpretations of the morphological situation may also explain different descriptions of pulmonary vein development in higher vertebrates, where some authors have also seen the initial position of the MCPV within the dorsal wall of the sinus venosus (His,1887; Flint,1906; Hochstetter,1906; Fedorow,1910; Brown,1913; Buell,1922; Dor et al.,1987; Kutsche and Van Mierop,1988; DeRuiter et al.,1995; Tasaka et al.,1996; Blom et al.,2001).

Position of the MCPV With Respect to the Left and Right Body Halves

In the majority of the previously published reports on mammalian embryos, the anlage of the MCPV was described as a midline structure that is initially connected to both halves of the venous heart pole but later becomes confined to the left atrial cavity by the formation of the primary interatrial septum (Brown,1913; Davies and MacConaill,1937; Auër,1948; Van Praagh and Corsini,1969; Bliss and Hutchins,1995; Webb et al.,1998; Wessels et al.,2000; Jongbloed et al.,2004). In the majority of the previously published reports on non-mammalian embryos, on the other hand, the initial position of the MCPV was already seen within the left half of the venous heart pole (Hochstetter,1906; Fedorow,1910; Robertson,1914; Dor et al.,1987; Kutsche and Van Mierop,1988; Männer and Merkel,2007). Our scanning electron microscopic data from X. laevis embryos show that, in this species, the MCPV forms to the left of the embryonic body midline (Fig. 4D–F) and to the left of the future interatrial septum (Figs. 2A–D and 5D–F). We have additionally carried out L-R lineage tracing experiments, which have shown that the interatrial septum and the wall of the MCPV both derive from the left body half (Fig. 8). These data are in accord with the above-mentioned data from other non-mammalian species and with a previous L-R lineage analysis, which has shown that the interatrial septum of the X. laevis heart derives from the left body half (Ramsdell et al.,2006). Our present findings thus show that in X. laevis, the MCPV is of unilateral left side origin. It is tempting to speculate that the same holds true for other non-mammalian species in which the initial position of the MCPV has also been seen within the left half of the venous heart pole. How can this situation be related to the situation in mammalian embryos in which the MCPV is frequently described as an original midline structure? One possible explanation for the divergent findings may be species-specific differences in the topogenesis of the MCPV. An ideal solitary midline structure should be built up in equal parts by material from the left as well as right body half. Midline structures, therefore, are of a bilaterally symmetric nature. It is conceivable that, in the majority of previously analyzed non-mammalian species, the MCPV may have been formed only within the left half of a bilaterally symmetric “MCPV forming field” and, therefore, been of left side origin. In most of the previously analyzed mammalian species, on the other hand, the MCPV may have been formed within the whole “MCPV forming field” and, therefore, been of bilateral (midline) origin. This hypothesis may also explain previous findings in lizards (Lacerta agilis), alligators (Alligator mississippiensis) and chicks (White leghorn). In these species, some researchers found a bilateral pair of endothelial evaginations at the dorsal wall of the embryonic venous heart pole of which the right one disappeared while the left one grew out into the dorsal mesocardium to form the MCPV (Fedorow,1910; Dor et al.,1987; Kutsche and Van Mierop,1988; Männer and Merkel,2007). Future studies on both mammalian as well as non-mammalian species are needed, firstly, to test our above-mentioned hypothesis and, secondly, to clarify how the topogenesis of the MCPV is mechanistically linked to the L-R axis specification. Our present data provide the basis for the interpretation of future experimental studies on X. laevis.

Establishment of the Structural Separation of the Pulmonary and Systemic Venous Flow Pathways at the Level of the Venous Heart Pole

Correct positioning of the MCPV within the dorsal wall of the venous heart pole only sets the scene for the later confinement of the pulmonary venous return to the left atrial cavity. It does not lead per se to the physiologically important structural separation of the pulmonary and systemic venous flow pathways, which is normally found at the venous pole of the mature heart. To establish the mature pattern of venoatrial connections, the developing left atrial cavity must remain connected to the MCPV but must lose its original communications with the sinus venosus and right atrial cavity. The developing right atrial cavity, on the other hand, must remain connected to the systemic venous tributaries but must lose its original communication with the future left atrial cavity and MCPV. In the present study, we have found that, in X. laevis, this is accomplished by the formation of a septal complex that divides the lumen of the embryonic venous heart pole into systemic and pulmonary venous flow pathways (Figs. 4, 5). This septal complex is formed by two anlagen, which we have named the “interatrial ridge” and the “sinu-pulmonary fold.” The interatrial ridge is a vertical ridge of tissue that grows from the dorsal atrial wall into the lumen of the primitive atrium so that the future left and right atrial cavities become separated from each other. It is of left side origin (Ramsdell et al.,2006; our present data) and seems to correspond to the anlagen of the interatrial septum of birds and primary interatrial septum of mammals, both of which have recently been identified as structures of left side origin (Franco et al.,2000; Wessels et al.,2000; Campione et al.,2001; Männer and Merkel,2007). The sinu-pulmonary fold is an endocardial fold that belongs to the left half of the sinu-atrial ridge. At the time point of its first appearance at stage 41, it only separates the MCPV from the sinus venosus. During subsequent developmental stages, however, it merges with the interatrial ridge and grows along the left half of the sinu-atrial ridge towards the dorsal wall of the atrio-ventricular canal so that the whole left atrial cavity becomes separated from the sinus venosus (Fig. 5). These findings correspond to previous descriptions of atrial development in amphibian embryos (Hochstetter,1906; Fedorow,1910). They deviate in part, however, from previous descriptions of atrial development in mammalian embryos where the separation of the left atrial cavity from the systemic venous system is explained by “rotation” or “reorientation” of the systemic venous tributaries rather than by septation (Webb at al.,1998; Anderson et al., 2003;2004). Species differences might explain this discrepancy.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Embryos

Pigmented X. laevis embryos were obtained from the laboratory of Prof. Tomas Pieler, Department of Developmental Biochemistry, GZMB, Georg-August-University of Göttingen. Embryos were incubated until they had reached developmental stages 39 to 46 according to Nieuwkoop and Faber (1967).

Preparation and Fixation of the Embryos for Scanning Electron Microscopy (SEM)

Preparation and fixation of the embryos was performed as follows: embryos anaesthesized with 1.3-amino-benzoate methane sulphonic acid salt (Sigma, St. Louis, MO) were transferred to a Petri dish filled with Locke's solution. The bottom of the Petri dish was covered with a layer of wax. Insect needles were inserted into the mouths and caudal trunks of the embryos to physically fix them to the bottom of the Petri dish. The pericardial cavity was then opened using electrolytically sharpened tungsten needles and the still beating hearts were perfused with Locke's solution (via a micropipette inserted into the ventricular cavity) until all visible signs of blood were removed from the heart and great vessels. To fix hearts in standardized dilation, a calcium-free Locke's solution of 20 mmol/l manganese chloride was used for final perfusion (Asami,1979). MnCl2 causes a cardiac arrest in a general dilation by blocking the calcium channels. After cardiac arrest, the embryos were externally rinsed (via a micropipette) with a 25% solution of glutaraldehyde. These treatments led to a rapid chemical prefixation of the hearts in the dilated state (Männer,1992). Final fixation of the embryos was carried out according to established protocols (Männer et al.,1996).

The fixed specimens were dehydrated in the usual manner and dried by the critical point method. The dried specimens were mounted on aluminium taps with conducting silver and sputtered with gold-palladium to a layer of about 40 nm (“Cool” Sputtering System Type E 5100, Polaron Equipment, Wafford, UK). They were examined and photographed with a Zeiss DSM 960 scanning electron microscope. Examinations of the specimens were performed stepwise alternately with micro-dissections using electrolytically sharpened tungsten needles. As the first step, the internal aspects of the developing atriums were analyzed after removal of the cardiac segments downstream to the developing atriums. As the second step, the internal aspects of the sinus venous, sinu-atrial border, and atriums were analyzed after removal of the liver and the ventral wall of the sinus venosus.

Histology and Conventional Light Microscopy

For histological examinations by conventional light microscopy, embryos were collected and fixed in Bouin's solution at stages 39–46. After stepwise dehydration in ethanol, they were cleared with methyl benzoate and embedded in Paraplast (Sherwood; St. Louis, MO). Serial sections (frontal, horizontal, or sagittal plane) were cut at 7 μm, transferred to albumin/glycerin-coated glass slides, and stained with haematoxylin and eosin according to Harris. Hearts of these embryos were not perfused with Locke's solution prior to fixation but were fixed in a dilated state caused by external rinsing with a calcium-free Locke's solution of 20 mmol/l manganese chloride.

Left-Right Cell Lineage Labeling, Histology, and Confocal Imaging

Left-right (L-R) cell lineage labeling was performed as described previously (Jahr et al.2008): eggs from pigmented X. laevis were fertilized and dejellied with 2.5% cysteine. For L-R cell lineage labeling, we used the dextran-conjugated fluorescence dyes Oregon Green 488 and Alexa fluor 647 (Ramsdell et al.,2005,2006). Oregon Green 488-conjugated dextran (10,000 MW; lysine fixable; Invitrogen, Carlsbad, CA) and Alexa fluor 647-conjugated dextran (10,000 MW; anionic fixable; Invitrogen) were each diluted to 2.5 mg/ml in RNAse-free water. At the two-cell stage, one blastomere was pressure injected into the marginal zone with 4 nl of Oregon Green 488-dextran while the other was pressure injected with 4 nl of Alexa fluor 647-dextran.

Embryos were collected and fixed in a 4% solution of paraformaldehyde at stages 41–46. Subsequent to fixation, embryos were screened for appropriate targeting of the dyes to only one side of the embryo. Only embryos showing distinct L-R hemi-labeling were then processed for histological sectioning as described above. Serial sections (frontal, horizontal, or sagittal plane) were cut at 7 μm and transferred to albumin/glycerin-coated glass slides. Slides were coverslipped with DPX mounting medium.

Histological sections were examined and images were captured using a Leica TCS SP2 Confocal System mounted on a Leica DM IRE 2 inverted microscope. The dextran-conjugated fluorophores were excited using either 496 nm (Argon-Krypton laser) for Oregon Green 488 or 633 nm (HeNe laser) for Alexa fluor 647. Pseudo-colored images (green for Oregon Green 488 and red for Alexa fluor 647) were imported into Adobe Photoshop for superimposition and adjustment for contrast and brightness.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Tomas Pieler and Katharina Damianitsch for providing the embryos and for their help in carrying out the microinjections of dextran-conjugated fluorescent dyes in their laboratory. We also thank Bernd Püschel for his help in confocal laser scanning microscopy, Kirsten Falk-Stietenroth and Hannes Sydow for technical and photographical assistance, and Cyrilla Maelicke for correcting the English manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES