The sinoatrial (SA) node, the pacemaker of the heart, is heterogeneous. Functionally, the SA node can be divided into two areas, the centre and the periphery. The centre normally acts as the leading pacemaker site, whereas the periphery normally serves as the conduction pathway from the centre to the atrial muscle. The SA node is heterogeneous in many respects. (i) Protein expression: in the rabbit, whereas both the periphery and the centre express neurofilament165 but not atrial natriuretic peptide (unlike atrial muscle which expresses atrial natriuretic peptide but not neurofilament165), the periphery but not the centre expresses connexin43 (Cx43) (Dobrzynski et al. 2002). (ii) Electrophysiology: we have evidence of a gradient in cell type from the periphery to the centre in the rabbit in terms of action potential shape, densities of ionic currents and expression of connexins (responsible for electrical coupling) (Boyett et al. 2000). The possible significance of this is revealed by a mathematical model of a string of 50 SA node cells connected to a string of 50 atrial cells. In the model, the central SA node cells, with their characteristic electrophysiology and low coupling conductance, are only able to show pacemaker activity and drive the atrial cells, which are more hyperpolarized and have a high coupling conductance, when they are separated from the atrial cells by peripheral cells, which have intermediate electrophysiological characteristics and perhaps an intermediate coupling conductance. (iii) Ca²⁺ handling: for the rabbit, evidence suggests that the density of the L-type Ca²⁺ current is less in central than peripheral cells (Musa et al. 2002). Consistent with this, immunolabelling of Cav1.2 protein is less in central than peripheral cells (Musa et al. 2002). As well as a decrease in the expression of the Ca²⁺ channel, there are decreases in the expression of other Ca²⁺ handling proteins from the periphery to the centre in the rabbit: Na⁺-Ca²⁺ exchanger, RYR2 and SERCA2 (Musa et al. 2002). There is evidence that the decrease in the density of the Ca²⁺ handling proteins from the periphery to the centre affects Ca²⁺ handling in the rabbit SA node: there is evidence that the intracellular Ca²⁺ transient is smaller and slower, the Ca²⁺ content of the sarcoplasmic reticulum (SR) is less and the contribution of the SR to the Ca²⁺ transient is less in central than peripheral cells (Lancaster et al. 2001, 2002). Pacing is regulated by the intracellular Ca²⁺ transient (e.g. Rigg & Terrar, 1996). Although it has been reported that the abolition of the intracellular Ca²⁺ transient by a sufficiently high concentration of ryanodine (30 mM) can abolish pacing in rabbit SA node cells (Bogdanov et al. 2001), we believe that the regulation is more modest: in the intact SA node of the rabbit, exposure to 30 mM ryanodine for 1 h decreased the spontaneous rate by 20 ± 3 % (n = 4). The putative regional difference in the Ca²⁺ transient above is expected to result in a regional difference in the intracellular Ca²⁺ regulation of pacing and there is evidence of this for the rabbit SA node: 2 mM ryanodine significantly slowed pacing of putative peripheral cells by 23 ± 8 % (n = 20), whereas it had no significant effect on the pacing of putative central cells (n = 18).

During ageing, there is a decline in SA node function (a decrease in spontaneous activity and a slowing of SA node conduction) and this may be related to the SA node heterogeneity. In part, it could be the result of a further loss of connexin43 (Cx43) and Ca²⁺ channels from the centre of the SA node: in the SA node centre of the guinea-pig (as in rabbit – see above) there are regions lacking immunolabelling of Cx43 and Cav1.2 proteins and these regions increased in size from 3.5 ± 0.6 mm² (n = 5) in 1 month animals to 47.7 ± 2.0 mm² (n = 5) in 38 month animals in the case of Cx43 (Jones et al. 2001), and from 2.1 ± 0.1 mm² (n = 4) in 1 month animals to 18.7 ± 2.3 mm² (n = 3) in 38 month animals in the case of Cav1.2.