The sinus node was considered as a pacemaker of the heart by Keith and Flack (1907), and there was considerable variation in its shape, size, and exact location in different species of animals. In previous studies, a number of descriptions about the anatomy and histology of sinus node existed for the different species. In human (James, 1961b; Anderson and Ho, 1998), bovine (James, 1965), camel (Ghazi and Tadjalli, 1996), horse (Bishop and Cole, 1967), sheep (Caesar et al., 1958), pig (Opthof et al., 1987), monkey (Alings et al., 1990), rabbit (Bleeker et al., 1980), guinea-pig (Opthof et al., 1985), and cat (Opthof et al., 1986), the sinus node was located in the right atrium at the junction of the crista terminalis with venous tissue—the cranial and posterior vena cava, and the intercaval region between the two great veins. The sinus node occupied the entire thickness between the endocardium and epicardium as in the rabbit (Bleeker et al., 1980), guinea-pig (Opthof et al., 1985), and monkey (Alings et al., 1990). However, in pig (Opthof et al., 1987) and dog (Woods et al., 1976), there was a layer of atrial muscle between the sinus node and endocardium.
A characteristic feature of the sinus node was much connective tissue, mainly collagen and fibroblasts (De Mazière et al., 1992). In human and dog (James et al., 1966), camel (Ghazi and Tadjalli, 1996), pig (Opthof et al., 1987), and monkey (Alings et al., 1990), in the center of sinus node, there were characteristic the pacemaker cells (P cells). In rabbit (Bleeker et al., 1980; Masson-Pevet et al., 1984), human and dog (James et al., 1966; Oosthoek et al., 1993), the transitional cells (T cells) extended several millimeters from the center to periphery of the sinus node where they met the working myocardium. The T cells could be isolated from the rabbit sinus node (Verheijck et al., 1998). Isolated cells could also have various morphologies: Denyer and Brown (1990) and Verheijck et al. (1998) classified rabbit sinus node cells into elongated spindle, spindle, and spider cells—spider cells, unlike elongated spindle and spindle cells, have multiple cytoplasmic projections.
The sinus node artery in many kinds of species has been demonstrated, including human (Verhaeghe and Van Der Hauwaert, 1967; Anderson et al., 1979; DiDio et al., 1995; Kawashima and Sasaki, 2003), camel (Ghazi and Tadjalli, 1996), bovine (James, 1965), horse (Bishop and Cole, 1967), sheep (Yalcin et al., 2004), dog (Izumisawa et al., 1994), and rabbit (James, 1967).
Yak is one of the most important breed living in the Qinghai-Tibetan plateau. Approximately 14 million yaks are found in China, accounting for 90% of the total number of yaks in the world. However, no information is yet available on the morphological details of sinus node and its artery in yaks. This study was undertaken to deal with the morphology of the sinus node and sinus node artery of yak to make basic information for comparison with other mammalians.
MATERIALS AND METHODS
The sinus node was studied in 40 yaks. The sinus node artery of yak was determined by Acrylonitrile Butadiene Styrene (ABS) casting of the coronary arteries followed by hydrochloric acid corrosion in 50 yaks, by postmortem coronary angiography in 50 yaks, and by histological methods in 20 yaks. The anatomic feature of yak heart was observed in 20 adult yaks. All animals were killed by exsanguinations via the abdominal aorta in slaughter house. The hearts were obtained immediately after the subjects were killed.
To show the sinus node artery, this study used the injection-corrosion casting technique and angiography, separately. The coronary orifices were cannulated, then three different concentrations of ABS solutions (10% ABS in acetone, 15% ABS in acetone, and 20% ABS in acetone) and 25% BaSO4 were injected by hand under controlled pressure of 60–90 mmHg, respectively. Then the heart that was injected with ABS solution was placed in 30% hydrochloric acid to corrupt the soft tissue, and then flushed with a fine jet of tap water to remove the residue from the cast. For a final cleaning and removal of the residue, the cast was taken photos. Angiograms of the intact heart which was injected with 25% BaSO4 were obtained using a mobile high frequency X-ray generator (Model KX-50vet, Shanghai, China) by taking roentgenograms in different directions. The distance was 70 cm, exposure time 10 sec, kV being set at 40, and mA at 32.
For light microscopy, the cranial vena cava (CVC) and adjacent atrium of the heart of yak were excised. Regions of the tissue were dissected as shown in Fig. 1. The blocks were carefully fixed in 4% neutral paraformaldehyde phosphate buffer (pH 7.3), then embedded, and serially sectioned with the rotary microtome set at 4 μm. Every 10th section was mounted and stained with hematoxylin and eosin in conjunction with the combined method of Weigert's Van Giesson and Masson trichrome to examine their structure. Extra sections were prepared as required.
For immunohistochemical study, the sections were labeled for connexin43 by streptavidin/peroxidase complex immunostaining technique. Antisera were obtained from Abcam, UK. After being dewaxed and taken to graded alcohol, the sections were pretreated with 3% hydrogen peroxide for 10 min and with normal goat serum for 15 min. The optimum dilution of the antisera was applied at 1:3,000 and incubated for 2 hr at 37°C in a moist chamber, after which sections were washed in phosphate buffered saline (PBS, pH 7.2–7.4) three times. A biotinylated anti-rabbit secondary antibody was applied at a dilution of 1:200 for 10 min. Streptavidin-horseradish peroxidase was applied to the sections for 10 min, after which they were again washed in PBS three times. Reaction products were formed with 3,3-diaminobenzidine tetrahydrochloride. After rinsing, sections were lightly counterstained with Delafield's hematoxylin, dehydrated, and mounted. Light microscopic observations and section images acquirement were performed using Olympus DP71 Light Microscope (including DP control and Image-Pro Express, Japan).
For transmission electron microscopy, tissue samples were taken from the midportion of the sulcus terminalis of yak heart immediately after slaughter, cut into pieces of approximately 1 mm3 immersed in 2.5% glutaraldehyde in neutral phosphate buffer (pH 7.3, at 4°C) and postfixed in 1% osmium tetroxide for 2 hr. The tissue was dehydrated in graded ethanol and embedded in Epon812. Orientation of the block was achieved by examining 1 μm sections stained with Toluidine blue. Thin sections (60–80 nm) were cut on a Leica EM-UC6 ultramicrotome, mounted on G200 grids, stained with alcoholic uranyl acetate and lead citrate, and viewed with a JEOL 1230 electron microscope at 120 kV.
The Yak Heart
The weight of the yak heart varied in relationship to body dimensions, ranged 750–1,950 g, average 1,134 g. The size of yak heart and the thickness of ventricular wall were shown in Table 1. The wall of the left ventricle was much thicker than that of the right ventricle.
Table 1. Anatomic parameters of the heart in adult yak (20 specimens)
Anatomic parameters of the heart surface
Circumferential length of the heart (cm)
Long axis of the heart (cm)
The distance from the end right ventricle to the apex (cm)
31.94 ± 2.49
14.22 ± 1.90
2.50 ± 0.53
Thickness of right ventricular wall
Coronary sulcus (cm)
Apex of heart (cm)
0.93 ± 0.62
1.10 ± 0.16
0.68 ± 0.20
Thickness of left ventricular wall
Coronary sulcus (cm)
Apex of heart (cm)
2.08 ± 0.17
2.97 ± 0.19
1.87 ± 0.30
Thickness of interventricular septum
Coronary sulcus (cm)
Apex of heart (cm)
2.13 ± 0.29
2.68 ± 0.13
1.90 ± 0.38
The right and left atria and right and left ventricles had distinctive anatomic features. The anatomic right ventricle was characterized as follows: fine moderator band crossed the chamber and connected the posterior and anterior papillary muscle; coarse trabecular muscles at the apical and inflow portion of the chamber. The anatomic left ventricle has the following features: two well-developed papillary muscles (anterolateral and posteromedial); fine trabecular muscles at the apical portion of the chamber; two or more cords of left ventricle was present 100%; the mitral had a highly developed and coordinated group of structures, chordae tendineae extended from the ventricular surfaces of both leaflets and attached to both papillary muscles.
The Sinus Node
The sinus node of yak was generally located at the junction between the CVC and the free wall of right atrium (Fig. 1), lay less than 0.5 mm beneath the epicardium of sulcus terminalis, and separated from that surface by a thin layer of loose connective tissue (Fig. 2). There were no working myocardium between sinus node and epicardium (Figs. 2, 3). In some cases, the sinus node was located above the crista terminalis, in the wall of CVC (Fig. 4A). The sinus node was oblong or spindle with long axis paralleled to crista terminalis. Between the sinus node and endocardium there was usually some right atrial myocardium (Fig. 3), but the sinus node sometimes extended directly to endocardium (Fig. 2).
The sinus node had an extensive framework of collagen and cells with different diameters showing poor staining properties were clearly distinguishable within the sinus node (Fig. 2). The sinus nodal cells were arranged in the collagen framework, in comparison to the more regular arrangement of atrial myocardium, and which contained a large perinuclear clear zone typical of most specialized cardiac tissue.
Two main groups of cells were distinguished within the sinus node.
The P cells were small and faintly staining, and had a perinuclear clear zone identical to the appearance, which tend to occur in grape-like clusters with each cell shaped much like a grape (Figs. 5, 6). Most of the central portion of sinus node was composed of clusters of P cells. No intercalated discs were visible by light microscopy.
The external surface of P cells was smooth, with noticeable absence of the marginal scalloping typical of T cells. This was most likely related to the paucity of myofibrils. The P cells had a central nuclear with a generally circular (Fig. 7), sometimes oval profile. The chromatin was fine and there were usually one or more nucleoli visible.
In contrast to T cells (Figs. 8 and 9) and working myocardium (Fig. 10), P cells were relatively poor in myofibrils (Figs. 7, 11, 12 and 13). Most myofibrils in P cells did not attach to the plasma membrane. The orientation of myofibrils were random in all directions, rather than in a single axis, as they have in working myocardium parallel to the longitudinal axis of cell. The form of Z lines was often irregular at the junction of the myofibrils. This arrangement was characteristic for the P cells. In some P cells, the myofibrils were spirally arranged, thereby they were cut both longitudinally and transversally.
The P cells contained less and appeared smaller mitochondria than working myocardium. The mitochondria of P cells were tended to be located in the central zone of cytoplasm and randomly distributed, and their profiles showed great variation in sections. The internal appearance of P cell mitochondria was strikingly different from that of working myocardium, exhibiting much simpler internal structure with very few cristae mitochondriales and little intercristal matrix (Fig. 11).
Intercellular junctions of P cells were almost exclusively by direct apposition of plasma membranes and there were no intercalated discs and few intercellular junctional complexes such as desmosomes or zonula adherens and gap junctions (Fig. 12A). The desmosomes and the zonula adherens were dispersed, but they were also found concentrated at some places (Figs. 12B, 13). Gap junctions were much less frequent in the P cells than in working myocardium and were conspicuously small when present (Fig. 12C).
Many pinocytic vesicles invaginated from the plasma membranes in P cells and appeared more numerous than in T cells or working myocardium (Fig. 13).
The T cells were elongated cells with characteristics intermediate between P cells and working myocardium. They were longer and slender, their cytoplasm contained more myofibrils and hence appeared darker than P cells (Figs. 5 and 6). The T cells not only intermingled between P cells but predominated at the margins of sinus node where they joined working myocardium. The cross-striations were observed in T cells by light microscopy (Fig. 14).
The T cells had fully scalloped membranes and well-developed intercalated discs (Fig. 8). As more myofibrils appeared in T cells, they tended to orient parallel to each other and inserted at the plasma membrane, where desmosomes appeared in concomitantly increasing number to form intercalated discs (Fig. 8). The increasing numbers, more organized distribution, and greater internal complexity of the mitochondria paralleled the increasing number of myofibrils (Fig. 9). Some T cells appeared inhomogeneous internally, with one half resembling a P cell and the other resembling working myocardium (Fig. 15). Such cells were in areas between P cells and working myocardium. The T cells made contact with both P cells and working myocardium as well as with each other. The junctions among the T cells were composed of plasma membrane apposition with numerous desmosomes (Fig. 16).
Within the center of sinus node there were syncytial cells, which were attached to collagen fibers and continuous with the P cells (Fig. 17).
Nerve Fibers and Ganglion
Nerve fibers and ganglion were present at the periphery of sinus node. The ganglion contained a few neuronal cell bodies and nerve fiber bundles, and surrounded by a capsule of the connective tissue (Fig. 18).
Antibodies Against Connexin43
In longitudinally sectioned of the working myocardium of yak, antibodies against connexin43 revealed a localized punctate staining pattern at the position of intercalated disks. In addition, connexin43 could be detected at the lateral sides of myocardium, albeit to a lesser extent (Fig. 19). This distribution was described as being typical for myocardial gap junctions and showed that specific gap junction labeling was achieved and that the epitopes to which the antibodies were directed were conserved in yak.
The sinus nodal cells, both the P cells and T cells, were stained by anti-connexin43 antibodies. The cells of the central part of sinus node of yak did not reveal reaction with anti-connexin43 (Fig. 4C). In the direction of CVC and right atrium, where P cell clusters connected with T cells, the junctions became more similar to intercalated discs in their fine structural organization and contained increasing numbers and sizes of gap junctions, a mixture of bundles of myocardium that were negative or positive for connexin43 was present (Fig. 4B, D).
The Sinus Node Artery
The sinus node artery originated from trunk of left coronary artery in five of 100 cases (5%) (Fig. 20), from left circumflex branch in 93 of 100 cases (93%) (Figs. 21 and 22). Although the origins were different, they always traversed the epicardial fat in the atrioventricular groove, coursing upward and ran along the anterior wall of left atrium, passing under the left atrial appendage to reach the region of junction between CVC and right atrium. Occasionally, the sinus node artery arose from right coronary artery in two of 100 cases (2%), which ran along the anterior wall of right atrium. As it reached the anterior interatrial sulcus, it turned upward to supply the sinus node. The total course of sinus node artery arising from left coronary artery was rather longer than the course from right coronary artery.19
In all hearts, the sinus node artery was the largest atrial coronary branch. The diameter of sinus node artery was 3 mm from the origin. Numerous small branches left the sinus node artery to terminate in the atrial wall or to anastomose with other atrial coronary arterioles (Fig. 22). Large anastomoses, however, were rare. It is remarkable that in most preparations, the diameter of sinus node artery tapered very little as its course was followed through the sinus node.
The sinus node artery of yak was unlike that of the human which located at the margin of sinus node (Figs. 2 and 3). The intima showed the usual arrangement which found in medium-sized arteries had clearly internal elastic membrane (Fig. 23). The media were composed of an inner circular and outer longitudinal smooth muscle layer. The muscle bundles of the outer longitudinal layer were arranged in a segmental pattern of origin and insertion separated by elastic and collagen fibers. Elastic and collagen septa extended from the adventitia to surround the outer longitudinal muscle bundles (Fig. 23). These septa became continuous with the elastic framework which surrounded the inner circular muscle layer (Fig. 23). The adventitia was composed of elastic and collagen fibers (Fig. 23). In the peripheral portion of the adventitia, large nerve bundle was found running in a longitudinal direction (Figs. 3 and 24). As the artery proceeded to enter the sinus node laterally, it gradually gave rise to several small branches distributed throughout the sinus node (Fig. 24).
The sinus node of yak was mostly located at the junction between CVC and right atrium at the sulcus terminalis. In some cases, the sinus node of yak was situated in the wall of CVC. The cranial margin of sinus node parted from the sulcus terminalis was 1.35 mm, and in broad contacted with the cells of CVC wall. The opposite border lay under the endocardium and in broad contacted with the cells of the crista terminalis. Agreement with that frequently met with in rabbit (Bojsen-Moller and Tranum-Jensen, 1972). The sinus node of yak also contained a large amount of dense collagen frame. The location of sinus node and the density of collagen fibers in sinus node were variables in different species, as shown in Table 2.
Table 2. Comparisons of the sinus node in different species
At the junction of the midportion of sinus intercavarum and terminal crest
Less collagen fibers
The histologic appearance of P cells in yak corresponded to that in other species. They were smaller, paler staining and much more interwoven than working myocardium, and had a perinuclear clear zone. Similar observations as previously reported in human (James, 1961b), horse (Bishop and Cole, 1967), camel (Ghazi and Tadjalli, 1996), and dog (James, 1962).
Through the electron microscopic examination of the P cells of yak, we found P cells contained few, irregularly oriented myofibrils. Mitochondria in P cells were also randomly distributed within the cytoplasm. The size and shape of mitochondria, with simple internal structure, were more variable than that in working myocardium. The findings were agreement with the observations of the previous study in human and dog (James et al., 1966), horse (Bishop and Cole, 1967), rabbit (Trautwein and Uchizono, 1963), monkey (Colborn and Carsey, 1972; Viragh and Porte, 1973), and rat (Ayettey et al., 1988). In sheep, however, the mitochondria contained many well-developed cristae (Caesar et al., 1958).
The T cells of yak had increasing numbers of myofibrils and mitochondria which gradually oriented in a fashion similar to working myocardium, which was similar to that of guinea-pig (Opthof et al., 1985). Some T cells of yak appeared inhomogeneous internally, with one half resembling a P cell and the other resembling working myocardium, agreement with the findings of James et al. (1966).
Nerve fibers and ganglions were only present at the periphery of sinus node of yak. This was similar to that seen in most mammalian species, including dog (James, 1962) and bovine (James, 1965). While, in horse (Bishop and Cole, 1967), human (James, 1961b), and cat (Ghazi et al., 1998), the nerve fibers were within the sinus node, but the ganglia were not seen within the central portion of sinus node.
The cells of the central part of sinus node of yak did not reveal reaction with anti-connexin43. The absence of immunologically detectable cellular concentrations of connexin43 in the sinus node seemed to be a general phenomenon, as it was found in rat (van Kempen et al., 1991; Gourdie et al., 1992), human, and bovine (Oosthoek et al., 1993). Gap junctions were detected by electron microscopy in the central part of sinus node of yak, similar to that observed in some species, such as monkey (Colborn and Carsey, 1972; Viragh and Porte, 1973), guinea-pig (Opthof et al., 1985), and rat (Ayettey et al., 1988). However, compared with atrial myocardium, the gap junctions in sinus node of yak were smaller in size and fewer in number, making them below detection level of the immunohistochemical staining methods. The findings were agreement with the observation of the previous study in human and bovine (Oosthoek et al., 1993).
The sinus node artery of yak originated from left coronary artery in 98%, from right coronary artery in 2%. There were variables in different species, as shown in Table 3. None of them had mentioned a sinus node artery arising directly from the trunk of left coronary artery although Bokeriya et al. (1984) and James (1961a) reported the sinus node artery of human arose from the trunk of left coronary artery in 25.95% and 2.22%, respectively. Yalcin et al. (2004) observed the sinus node artery of sheep originated from the trunk of left coronary artery in 17%. In yak, we observed that the sinus node artery originated from the trunk of left coronary artery in 5% of specimens.
Table 3. Percentage of the origin of sinus node artery in different species
The sinus node artery of yak which was located at the periphery of sinus node, entered the node laterally and gave rise to several small branches distributed throughout the sinus node. It was surrounded by framework of collagen fibers and the sinus nodal cells. The locations of the artery were various in different species. In human (James, 1961b; He et al.,1991; Berdajs et al., 2003; Futami et al., 2003), dog (James, 1962), and camel (Ghazi and Tadjalli, 1996), the sinus node artery was a centrally located artery, ran through the inside of sinus node, and interlaced the collagen and elastic fibers that were irregularly organized around the central artery. In rabbit, the sinus node was not organized around the central artery but contained only arterial branches of a size appropriate for a nutrient function alone (James, 1967). In bovine, a number of branches of appropriate size served the nutrient function, but the sinus node was distinctly not organized around any one of these branches (James, 1965). In horse, a large sinus node artery was consistently present in the sinus node (Bishop and Cole, 1967).
The sinus node artery in yak had an internal elastic membrane throughout its course, the present observation was in agreement with that made in camel (Ghazi and Tadjalli, 1996), whereas in human, at some points within the sinus node, the artery did not possess an internal elastic membrane (James, 1961b). In the peripheral portion of the adventitia of the sinus node artery in yak, large nerve bundle was found running in a longitudinal direction. This was similar to that of human (Ryback and Mizeres, 1965).
In summary, the sinus node of yak was basically similar to those of other domestic animals. This included its location, internal structure, and the sinus nodal cells. The blood supply of yak sinus node differed from that in other species, the sinus node artery of yak originated almost exclusively from left coronary artery and is located at the periphery of sinus node.