Transgenic technology has potentially solved many of the immunological difficulties of using pig organs to support life in the human recipient. Nevertheless, other problems still remain. Knowledge of cardiac anatomy of the pig (Sus scrofa) is limited despite the general acceptance in the literature that it is similar to that of man. A qualitative analysis of porcine and human cardiac anatomy was achieved by gross examination and dissection of hearts with macrophotography. The porcine organ had a classic ' Valentine heart ' shape, reflecting its location within the thorax and to the orientation of the pig's body (unguligrade stance). The human heart, in contrast, was trapezoidal in silhouette, reflecting man's orthograde posture. The morphologically right atrium of the pig was characterised by the tubular shape of its appendage (a feature observed on the left in the human heart). The porcine superior and inferior caval veins opened into the atrium at right angles to one another, whereas in man the orifices were directly in line. A prominent left azygous vein (comparable to the much reduced left superior caval or oblique vein in man) entered on the left side of the pig heart and drained via the coronary sinus. The porcine left atrium received only 2 pulmonary veins, whereas 4 orifices were generally observed in man. The sweep between the inlet and outlet components of the porcine right ventricle was less marked than in man, and a prominent muscular moderator band was situated in a much higher position within the porcine right ventricle compared with that of man. The apical components of both porcine ventricles possessed very coarse trabeculations, much broader than those observed in the human ventricles. In general, aortic-mitral fibrous continuity was reduced in the outlet component of the porcine left ventricle, with approximately two-thirds of the aortic valve being supported by left ventricular musculature. Several potentially significant differences exist between porcine and human hearts. It is important that these differences are considered as the arguments continue concerning the use of transgenic pig hearts for xenotransplantation.

Background Cardiac conduction is influenced by peptidergic mechanisms as well as classic neurotransmitters. The distribution of peptide-containing nerves has not been well defined.Methods and Results Immunofluorescence and histochemical techniques were used to visualize the innervation of the human conduction system and to distinguish nerve subpopulations according to their peptide and enzyme content. Nerve fibers and fascicles displaying immunoreactivity for protein gene product 9.5 (PGP 9.5) were more numerous in the sinus and atrioventricular nodes than in the penetrating bundle, bundle branches, and adjacent myocardium. The relative density of innervation was greater in the central region of the sinus node than in the peripheral regions. Nerve densities were also higher in the transitional region of the atrioventricular node compared with its compact region. Acetylcholinesterase (AChE)-positive nerves were the main subtype identified in the sinus and atrioventricular nodes, representing half to two thirds of the stained area occu-

The immunological problems of pig hearts supporting life in human recipients have potentially been solved by transgenic technology. Nevertheless, other problems still remain. Autonomic innervation is important for the control of cardiac dynamics and there is evidence suggesting that some neurons remain intact after transplantation. Previous studies in the human heart have established regional differences in both general autonomic innervation and in its component neural subpopulations. Such studies are lacking in the pig heart. Quantitative immunohistochemical and histochemical techniques were used to demonstrate the pattern of innervation in pig hearts (Sus scrofa). Gradients of immunoreactivity for the general neural marker protein gene product 9.5 were observed both within and between the endocardial, myocardial and epicardial plexuses throughout the 4 cardiac chambers. An extensive ganglionated plexus was observed in the epicardial tissues and, to a lesser extent, in the myocardial tissues. The predominant neural subpopulation displayed acetylcholinesterase activity, throughout the endocardium, myocardium and epicardium. These nerves showed a right to left gradient in density in the endocardial plexus, which was not observed in either the myocardial or epicardial plexuses. A large proportion of nerves in the ganglionated plexus of the atrial epicardial tissues displayed AChE activity, together with their cell bodies. Tyrosine hydroxylase (TH)-immunoreactive nerves were the next most prominent subpopulation throughout the heart. THimmunoreactive cell bodies were observed in the atrial ganglionated plexuses. Endocardial TH-and NPYimmunoreactive nerves also displayed a right to left gradient in density, whereas in the epicardial tissues they showed a ventricular to atrial gradient. Calcitonin gene-related peptide (CGRP)-immunoreactive nerves were the most abundant peptide-containing subpopulation after those possessing NPY immunoreactivity. They were most abundant in the epicardial tissues of the ventricles. Several important differences were observed between the innervation of the pig heart compared with the human heart. These differences may have implications for the function of donor transgenic pig hearts within human recipients.

This study was prompted by the prospect of transgenic pigs providing donor hearts for transplantation in human recipients. Autonomic innervation is important for the control of cardiac dynamics, especially in the conduction system. Our objective was to assess the relative distribution of autonomic nerves in the pig heart, focusing initially on the conduction system but addressing also the myocardium, endocardium and epicardium (see Crick et al. 1999). Quantitative immunohistochemical and histochemical techniques were adopted. All regions of the conduction system possessed a significantly higher relative density of the total neural population immunoreactive for the general neuronal marker protein gene product 9.5 (PGP 9.5) than did the adjacent myocardium. A similar density of PGP 9.5-immunoreactive innervation was observed between the sinus node, the transitional region of the atrioventricular node, and the penetrating atrioventricular bundle. A differential pattern of PGP 9.5-immunoreactive innervation was present within the atrioventricular node and between the components of the ventricular conduction tissues, the latter being formed by an intricate network of Purkinje fibres. Numerous ganglion cell bodies were present in the peripheral regions of the sinus node, in the tissues of the atrioventricular groove, and even in the interstices of the compact atrioventricular node. Acetylcholinesterase (AChE)-containing nerves were the dominant subpopulation observed, representing 60-70 % of the total pattern of innervation in the nodal tissues and penetrating atrioventricular bundle. Tyrosine hydroxylase (TH)-immunoreactive nerves were the next most abundant neural subpopulation, representing 37 % of the total pattern of innervation in the compact atrioventricular node compared with 25 % in the transitional nodal region. A minor population of ganglion cell bodies within the atrioventricular nodal region displayed TH immunoreactivity. The dominant peptidergic nerve supply possessed immunoreactivity for neuropeptide Y (NPY), which displayed a similar pattern of distribution to that of TH-immunoreactive nerve fibres. Calcitonin gene-related peptide (CGRP)-immunoreactive nerves represented 8-9 % of the total innervation of the nodal tissues and penetrating atrioventricular bundle, increasing to 14-19 % in the bundle branches. Somatostatin-immunoreactive nerve fibres were relatively sparse (4-13 % of total innervation) and were most abundant in the nodes, especially the compact atrioventricular node. The total pattern of innervation of the porcine conduction system was relatively homogeneous. A substantial proportion of nerve fibres innervating the nodal tissues could be traced to intracardiac ganglia indicative of an extensive intrinsic supply. The innervation of the atrioventricular node and ventricular conduction tissues was similar to that observed in the bovine heart, but markedly different to that of the human heart. It is important that we are aware of these findings in view of the future use of transgenic ...

Background The aim of the present investigation was to determine the relative distribution of autonomic and sensory nerves in the cardiac conduction tissues of calves. Methods A quantitative immunohistochemical and histochemical technique was adopted. Results Immunoreactivity to the general neuronal marker protein gene product 9.5 (PGP 9.5) demonstrated that all regions of the conduction system possessed a higher relative density of total nerves when compared with the surrounding myocardial tissues. Unlike myocardial innervation, the conduction system did not display an atrial‐to‐ventricular gradient in nerve density. PGP 9.5‐immunoreactive nerve trunks and varicose nerve fibres were more numerous in the transitional atrioventricular node and the penetrating atrioventricular bundle than in either the sinus node, compact atrioventricular node, or bundle branches. The Purkinje network of the ventricular conduction tissues possessed a rich supply of PGP 9.5‐immunoreactive nerve trunks and varicose nerve fibres. Acetylcholinesterase (AChE)‐positive nerves were the main subtype identified in the sinus and atrioventricular nodes and in the ventricular conduction tissues, representing 50–80% of the area occupied by PGP 9.5‐immunoreactive nerves. The compact atrioventricular node possessed AChE‐positive and tyrosine hydroxylase (TH)‐immunoreactive nerves in similar proportions (45%), although, in general, TH‐immunoreactive nerves had a lower relative nerve density than AChE‐positive nerves. Neuropeptide Y (NPY)‐immunoreactive nerves represented the main peptide‐containing subpopulation and occurred throughout the conduction system, displaying a similar pattern of distribution and relative density to those demonstrating TH immunoreactivity. Nerve fibres immunoreactive for somatostatin, vasoactive intestinal polypeptide, substance P, and calcitonin gene‐related peptide formed relatively minor subpopulations. Conclusions The general innervation of the bovine conduction tissues exhibits significant regional variation. Throughout all regions of the conduction system, AChE‐positive nerve represented the dominant subtype when compared with TH‐ and NPY‐immunoreactive nerves. The distribution and relative density of nerve subtypes in the nodal tissues of the bovine conduction system are similar to those observed in man, whereas differences were observed in other regions, such as the atrioventricular bundle and bundle branches. This finding must be considered by those making interspecies comparisons. © 1996 Wiley‐Liss, Inc.