Cultured adult rabbit pacemaker cells for gene transfer studies

Lakatta, Edward

Abstract

In order to genetically manipulate key proteins involved in the autonomic regulation process, we had to develop a technique for the culture of rabbit Sinoatrial node cells, as its impossible to do so in freshly isolated SANC. We have been able to obtain stable adult rabbit cultured SANC (c-SANC), to characterize their properties, and have successfully overexpressed proteins in c-SANC via adenovirus directed acute gene-transfer technique. Our results show that on the first day of primary SANC culture, most of the cells tend to spread out and could stay alive for up to 8 days, a period which would allow us to introduce exogenous proteins into c-SANC. By immuno-staining, we detected essential proteins involved in autonomic regulation in c-SANC, including type 2 sarcoplasmic reticulum Ca2+ release channel, i.e. type 2 ryanodine receptors (RyR2), L-type Ca2+ channel, hyperpolarization-activated cyclic nucleotide-gated channel 4, phospholamban (PLB), Sarco/Endoplasmic Reticulum Ca2+-ATPase 2a and Sodium-Calcium exchanger. At 34 plus/minus 0.5 degrees C, c-SANC generate spontaneous, rhythmic action potentials (AP), but at a level (1.35 plus/minus 0.02 Hz, n=804, over 2 to 8 days into culture) is roughly 50% of that of f-SANC (2.79 plus/minus 0.04 Hz, n=203, p<0.001). Although both c-and f-SANC generate rhythmic APs and AP triggered global Ca2+ release transients, the rhythmicity of c-SANC AP or Ca2+ transient is less robust than that of f-SANC, as indicated by a lower rhythmicity index of the autocorrelation function in c-SANC versus f-SANC (p<0.001). By comparing the rhythmicity index of c-SANC and f-SANC at control or different treatments, the rhythmicity of AP is positively correlated with the AP firing rate (r2=0.75). Spontaneous Local Ca2+ Releases (LCR) period are increased in c-SANC, and are correlated with the decay time of AP triggered global Ca2+ release transients in both cell types, but with an increased variability in c-SANC vs. f-SANC. It is well documented that the peptide inhibitor of protein kinase A (PKA), PKI, can dramatically reduce or stop the beating rate of f-SANC. We hypothesized that the relatively low beating rate of c-SANC, is possibly due to the down-regulated PKA signaling in the cultured cells. Indeed, non-specific PDE inhibitor IBMX (100 microMolar, 10min) increases the AP firing rate of c-SANC to a similar maximum to the treatment of f-SANC. Furthermore, acute stimulation of beta-adrenergic receptors with 1 microMolar isoproterenol (ISO) for 10 min accelerates AP and Ca2+-transient kinetics, reduces the LCR period, and accelerates the AP firing rate to a similar maximum in c-SANC (3.34 plus/minus 0.05 Hz, n=150) and f-SANC (3.55 plus/minus 0.06 Hz, n=126). In addition, we observed that the phosphorylation level of RyR2, which is indexed by the fluorescence density of phosphorylated RyR2 at Ser2809 normalized by its own total RyR2 fluorescence density, is substantially lower in c-SANC (1.32 plus/minus 0.06, n=47) than in f-SANC (1.66 plus/minus 0.15, n=24, p<0.01). While acute ISO stimulation raises the RyR2 phosphorylaiton at Ser2809 to a similar level in both cell types, PKI treatment reduces the phosphorylation level. More specifically, the phosphorylation level of PLB at Ser16, a PKA specific site, is also significantly lower in c-SANC than f-SANC. Similarly, ISO acute stimulation increases and PKA inhibition by PKI decreases PLB phosphorylation at Ser16 in both cultured and freshly isolated SANC, supporting the interpretation that PKA signaling is down-regulated in cultured SANC compared with freshly isolated SANC. What is the mechanism underlying the PKA down-regulation in cultured pacemaker cells? Based upon the above data and the fact that the activation of pertussis toxin (PTX)-sensitive Gi signaling is involved in the beating rate reduction of f-SANC, we measured the protein expression level of type 2 regulator of G protein signaling (RGS2), which functions as a powerful negative regulator of PTX-sensitive Gi signaling. As we expected, the protein level, indexed by the immuno-labeling density along the cell membrane, is substantially lower in 2 day cultured SANC (149.9 plus/minus 4.0, n=100) than in f-SANC (201.9 plus/minus 6.0, n=88, p<0.001). 2 hours incubation of 1 microMolar ISO enhances the staining density of RGS2 and PKI completely inhibits ISOs effect. Functionally, over-expression of RGS2 via adenovirus directed acute gene-transfer technique increases the spontaneous beating rate of cultured SANC from 1.35 plus/minus 0.05 Hz (n=91) to 1.86 plus/minus 0.05 Hz (n=50, p<0.001), which is 66% of f-SANCs AP firing ate. This effect is not because of adenovirus infection, as introducing the green fluorescent protein (GFP) into c-SANC via the same technique, does not affect the cell beating rate, and there is no correlation between AP firing rate and GFP expression level. To our surprise, over-expression of RGS2 in c-SANC increased the phosphorylation level of PLB at Ser16 but not the phosphorylation level of RyR2 at Ser2809. Furthermore, when cultured SANC were treated with 0.4micrograms/ml PTX overnight, the spontaneous beating rate is boosted to 2.38 plus/minus 0.11 Hz (n=45), 85% of f-SANCcs AP firing rate. Partial rescue of c-SANCs AP firing rate by PTX treatment or RGS2 overexpression indicate that a reduction in PKA-dependent Ca2+-cycling protein phosphorylation that is Gi-dependent is involved in prolongation of LCR period and reduced spontaneous AP firing rate of c-SANC, and that this deficit can be reversed by pharmacologic or genetic manipulation. In summary, we have defined important characteristics of a cultured rabbit SANC model, in which altered cAMP/PKA modulation that develops in culture, reduces the AP firing rate and rhythmicity. Specifically, a Gi-dependent PTX-sensitive reduction in PKA-dependent Ca2+-cycling protein phosphorylation, likely due, in part at least, to reduced RGS signaling, increases the variability of LCR period and prolongs the average LCR period, and increases the variability of AP cycle length and reduces the average spontaneous AP firing rate. The altered phenotypes of c-SANC, and its rescue by the maneuvers employed provide additional support for the coupled-clock hypothesis of pacemaker cells. Specifically, culture conditions, via activation of Gi signaling, interfere with coupled-clock function by interfering with SR Ca2+ cycling and also interfering with its coupling to sarcolemma ion channels, leading to alteration in automaticity and rhythmicity. This culture model will enable future studies in which genetic manipulation of adult pacemaker cells can be employed to glean novel mechanistic insights into adult pacemaker cell function.