INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IN THE VENTRICLE, the expression of L-type (I_Ca,L) and T-type calcium current (I_Ca,T) changes during ontogenesis. In most mammalian species studied, the current density of I_Ca,L increases during development (10, 17, 24, 25, 32; see also 9). The results on I_Ca,T are not definitive, with studies (17, 31, 46) in the rat ventricle reporting a postnatal decrease in current density and studies in the rabbit ventricle finding either a developmental increase (44) or an absence of the current at all ages (36). In addition, Nuss and Marban (35) recorded strong I_Ca,T in newborn mice, whereas no current was found in fetal mice (10).

There is no information about the developmental changes of calcium currents in sinus node cells (SNC). In the adult central sinus node, which has little or no functional fast sodium current (I_Na) (2, 11), I_Ca,L contributes not only to plateau potential as in the ventricle but also to action potential upstroke and perhaps late diastole (47). I_Ca,T is more pronounced than in ventricular myocytes and may participate in impulse initiation (13, 20). Because of the specific functional role of calcium currents in SNC, any developmental change in their biophysical characteristics could have significant impact on impulse initiation and automaticity.

The importance of studying these changes is emphasized by the recent finding that SNC from newborn rabbits have a prominent TTX-sensitive fast I_Na (2). In that study (2), TTX markedly prolonged the cycle length of newborn cells but did not change it in adult cells, demonstrating that I_Na plays a key role in impulse initiation in the newborn sinus node. Additional data indicate that I_Na directly contributes to diastolic depolarization in newborn SNC (3). The observation that newborn (but not adult) SNC exhibited pronounced slowing of spontaneous rate after block of I_Na suggests that, during development, another current (or currents) functionally substitutes for I_Na with respect to impulse initiation. The present study tested the hypothesis that I_Ca,L and/or I_Ca,T serve this purpose. Accordingly, one would expect an age-dependent increase in calcium current density and/or changes in biophysical characteristics resulting in increased contribution to impulse initiation. Unexpectedly, we did not find any developmental increase but rather a decrease in I_Ca,L current density with age. However, the activation curve of I_Ca,L was shifted negative in adult cells, whereas the inactivation curve was shifted positive; these opposite shifts should result in increased "window" current in adult cells compared with newborn cells. The wider window current, and in particular the more negative position of the activation curve, can increase inward current during late diastole and reduce the threshold potential for impulse initiation by I_Ca,L in adult SNC. An additional mechanism to increase the role of I_Ca,L in action potential initiation may be the faster recovery from inactivation found in adult SNC in this study.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SNC were prepared using a previously described method (12). Newborn (3-10 days old) and adult (41-48 days old) female rabbits were anesthetized with a mixture of ketamine (60 mg/kg) and xylazine (4.6 mg/kg) and then euthanized in accordance with protocols approved by the Animal Care and Use Committee of Columbia University. The heart was removed, and the sinus node was isolated and cut into four to six strips. Strips were rinsed for 5 min in a glass tube with a solution containing (in mM) 140 NaCl, 5.4 KCl, 0.5 MgCl₂, 1.2 KH₂PO₄, 5 HEPES-NaOH, 50 taurine, and 5.5 glucose (pH 6.9) and then triturated for 15-20 min in 5 ml of the same solution including enzymes and 200 µM CaCl₂. The amount of the following enzymes varied depending on the individual activity of samples and the age of the rabbits: 1.8-6 mg (600-2,000 units) collagenase (Worthington Biochem), 0.32-0.65 mg (1.5-3 units) protease (Sigma), and 0.05-0.10 mg (4.75-9.5 units) elastase (Sigma). Strips were rinsed for 5 min in K⁺-rich solution (see composition in Ref. 47) and triturated in a fresh 5-ml volume of this solution to disperse cells mechanically. Both enzymatic and mechanical digestions were performed at 37°C. During the following 20 min, the concentrations of Na⁺ and Ca²⁺ were gradually increased while the concentration of K⁺ was decreased to achieve final concentrations of 100, 1.3, and 35 mM, respectively. Cells were kept in this solution at 4-6°C and used in experiments 1-8 h after complete isolation.

During experiments, cells were placed in a heated (35°C) bath and superfused with a physiological saline solution of the following composition (in mM): 140 NaCl, 5.4 KCl, 2 CaCl₂, 1 MgCl₂, 5 HEPES, and 10 glucose (pH 7.4). Only single, spindle-shaped, spontaneously beating cells were chosen for electrophysiological measurements. Membrane currents were recorded by the whole cell or perforated patch techniques using an IBM-compatible computer equipped with pCLAMP software (versions 6.0.3 or 8; Axon Instruments), a DigiData 1200 series interface (Axon Instruments), and an Axopatch 1C amplifier (Axon instruments). Borosilicate glass pipettes (Sutter Instrument) were filled with a solution of the following composition (in mM): 90 aspartic acid, 10 NaCl, 100 CsOH, 30 CsCl, 2 MgCl₂, 5 EGTA, 2 CaCl₂, 10 HEPES, 2 ATP Na₂, and 0.1 GTPNa₂; pH 7.2 (pCa = 7). Pipette resistance was 3-4 M.

The whole cell configuration of the patch clamp was used to study I_Ca,T. After a seal was formed, the membrane was ruptured, and capacitance currents were compensated; the cell was superfused with a low-Na⁺ solution of the following composition (in mM): 50 NaCl, 5 CsCl, 90 triethylammonium chloride, 0.5 MgCl₂, 10 HEPES, 2 CaCl₂, 5.5 glucose, and 0.01 TTX (pH 7.4). After I_Ca,L rundown had ceased (see RESULTS), the current-voltage (I-V) relationship was determined from a holding potential of 80 mV using two protocols. The first protocol included a 50-ms step to 50 mV to inactivate I_Ca,T and then 200-ms steps to potentials ranging from 70 to 50 mV (increment, 10 mV). The second protocol was the same except that the step to 50 mV was omitted. I_Ca,T was found by subtracting currents obtained with the first protocol from those obtained with the second one. Linear leakage and capacitative transients were corrected by proportional subtraction of mean current recorded during steps from 80 to 60 mV (n = 10 records).

To study the steady-state inactivation of I_Ca,T, the potential was held at 40 mV, and a 600-ms conditioning pulse (to voltages from 100 to 40 mV, 10-mV steps) was applied followed by a 100-ms pulse to 30 mV. In these experiments, 10 µM nifedipine (Sigma) was added to the control low-Na⁺ solution to suppress I_Ca,L.

To study I_Ca,L, the perforated patch technique was employed by adding 100-200 µg/ml amphotericin B (Sigma) to the pipette solution. Two to ten minutes after the seal was formed, when series resistance was reduced to 15-20 M, the normal superfusate was exchanged for one containing (in mM) 135 NaCl, 10 CsCl, 1.8 CaCl₂, 1 MgCl₂, 5 HEPES, 10 glucose, and 0.003 TTX. With the use of the perforated patch, the rundown of I_Ca,L in either adult or newborn SNC was only 4% when checked every 5 s using 100-ms pulses to 20 mV from holding potential of 50 mV over a period of 4 min (data not shown). To construct I-V curves, 200-ms steps ranging from 45 to 55 mV were applied from a holding potential of 50 mV in the absence and in the presence of 10 µM nifedipine. I_Ca,L was measured as the nifedipine-sensitive current.

The perforated patch and solutions described above also were used to study steady-state inactivation of I_Ca,L and recovery of I_Ca,L from inactivation. In this case, 100 µM NiCl₂ and 10 µM TTX were added to the cesium-containing solution. To study inactivation of I_Ca,L, we used a 1,000-ms conditioning pulse (to voltages from 80 to 10 mV, 10-mV steps) followed by a 2-ms step to 50 mV before the 300-ms test pulse to 10 mV. Holding potential was 50 mV. Currents were measured relative to the peak current at the test potential after the conditioning potential of 80 mV (I_max). Linear leakage and capacitative transients were eliminated by a standard subtraction method.

To study the recovery from inactivation, a two-pulse protocol was used. The holding potential was 50 mV; both pulses were from 50 to 20 mV and of 100-ms duration. The interval between the pulses was increased from 10 to 2,800 ms. The protocol was run twice in the absence and presence of nifedipine, and I_Ca,L was measured as the nifedipine-sensitive current.

For the voltage-dependent activation and inactivation kinetics of I_Ca,T and I_Ca,L, we assumed a single-exponential relation, as employed previously in the sinoatrial node (15, 20, 48). Activation and inactivation were analyzed according to Boltzmann fitting analysis, and recovery from inactivation was according to exponential analysis (Microcal Origin, version 6.0). Data are presented as mean ± SE. Statistical significance was determined by ANOVA or Student's t-test as appropriate. A value of P < 0.05 was regarded as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

T-Type Calcium Current

Current density. After the patch was ruptured, the cell was adapted to the new conditions for several minutes, during which time the rundown of I_Ca,L was estimated. The rundown of I_Ca,L during whole cell patch-clamp recording is a well-known phenomenon (see, for example, Ref. 5). The rundown of I_Ca,T was negligible or lacking (16, 34, 41) . Moreover, Wetzel et al. (44) found in rabbit ventricular myocytes that with rundown of I_Ca,L, I_Ca,T became more prominent. However, if I_Ca,L rundown has not stabilized before the measurement of I_Ca,T the value for I_Ca,T found as a difference current may be distorted. To estimate the dynamics of I_Ca,L rundown, 200-ms pulses to 20 mV preceded by short steps to 50 mV from a holding potential of 80 mV were repeated every 5 s. Generally, the current stabilized in 2 min at ~70% of the initial magnitude (data not shown). After this time, two I-V protocols were run as described in METHODS. A typical result is shown in Fig. 1, which depicts the family of current traces with and without the conditioning step to 50 mV (Fig. 1, A and B) as well as the calculated difference current traces (offset for clarity; Fig. 1C). At both ages studied, the I-V curves constructed from the difference current had a threshold between 60 and 50 mV and a peak at 20 mV (Fig. 2). The peak current in cells from newborn and adult animals was 4.4 ± 0.7 (n = 6) and 4.3 ± 0.7 pA/pF (n = 5), respectively. No significant difference between I_Ca,T I-V curves constructed for these two ages was found (two-way ANOVA, P = 0.49). Nickel, which is reported to preferentially inhibit I_Ca,T in SNC (20), was used at a concentration of 100 µM and reduced peak current density at 20 mV by 80 ± 4% and 65 ± 9% in newborn and adult animals, respectively (data not shown). The I-V data were used to construct activation curves for adult and newborn SNC.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Experimental protocol used to record T-type calcium current (ICa,T). A and B: two sets of typical traces of Ca2+ currents consecutively recorded from the same sinus node cell (SNC). A: currents obtained by stepping from a holding potential of 80 mV to the range of 70 to 50 mV. These currents contained both ICa,T and L-type calcium current (ICa,L). B: currents obtained by the same steps after a 50-ms step to 50 mV to inactivate ICa,T. Corresponding wave form configurations are shown above each set of traces. C: difference current (defined as ICa,T) obtained by subtracting B from A; only the initial portion of current records are shown, with capacitance transients cropped, and every other trace was omitted for clarity. The transient current at the start of the prepulse to 50 mV (B) largely represents uncompensated capacitance, which has been corrected for in the difference traces of C.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Absence of developmental changes in ICa,T density and activation. A: peak current-voltage (I-V) relation for ICa,T in adult and newborn SNC. Currents were recorded by the difference method illustrated in Fig.1; n = 6 for newborn SNC and 5 for adult SNC. B: activation curves were constructed using the reversal potential estimated from the control I-V curves in A (see text). Curves were fitted with a Boltzmann function (lines); no significant differences between the two age groups were found in either the midpoint (V1/2) or inverse slope factor.

Activation of I_Ca,T. Activation curves were constructed according to the equation d(V) = g(V)/g_max = [I/(V_rev  V)]/g_max, where d(V) is channel activation, g(V) is conductance, g_max is maximal conductance, V_rev is the reversal potential, and V is the test potential. The linear portion of the positive limb of the I-V curve was extrapolated to obtain V_rev, which was 36 and 37 mV for newborn and adult SNC, respectively. As shown in Fig. 2B, the activation curves are similar at both ages, with no significant difference in either the midpoint (V_1/2; 33 ± 2 mV, n = 6, for newborn SNC and 35 ± 2 mV, n = 5, for adult SNC, P = 0.483) or inverse slope factor (7.1 ± 0.6 and 5.6 ± 0.7 mV, respectively, P = 0.135).

Steady-state inactivation. Steady-state inactivation of I_Ca,T was expressed as f(V) = I(V)/I_max, where I is peak current at the test potential after any given conditioning potential and I_max is the peak current after a conditioning potential of 100 mV (Fig. 3). We found no developmental differences in the availability of I_Ca,T (V_1/2: 65 ± 2.1 mV, n = 6, for newborn SNC and 65.1 ± 2.4 mV, n = 6, for adult SNC, P = 0.943; inverse slope factor: 9.7 ± 0.7 for newborn SNC and 10.5 ± 0.7 mV for adult SNC, P = 0.464). In both adult and newborn SNC, I_Ca,T was inactivated by ~20% at 80 mV (Fig. 3C), indicating that the current density in the I-V relationship of I_Ca,T in SNC held at 80 mV was slightly but equivalently underestimated at both ages.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Absence of developmental difference in steady-state inactivation of ICa,T. A: waveform configuration with conditioning pulses varying from 100 to 40 mV and a test pulse to 30 mV. B: steady-state inactivation curves obtained in the presence of 10 µM nifedipine to suppress ICa,L; n = 6 for both age groups. Curves were fit with Boltzmann functions (lines).

L-Type Calcium Current

Current density. Representative current traces in the absence and presence of nifedipine (10 µM) and the calculated difference current are shown in Fig. 4, A-D, for an adult cell. The I-V relations of the current in the two age groups are shown in Fig. 4E. Both curves were bell shaped but peaked at different voltages: approximately 0 mV for newborn and approximately 5 mV for adult SNC. Peak current density in newborn SNC (17.58 ± 2.46 pA/pF, n = 15) was significantly (P = 0.049) greater than that in adult SNC (12.15 ± 1.42 pA/pF, n = 8). According to two-way ANOVA, the two I-V curves are significantly different (P < 0.01).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Age-dependent differences in ICa,L density. A: voltage protocol employed. B and C: original current traces recorded using voltage-clamp protocol shown in A for control currents (B) and currents from the same adult SNC after administration of 10 µM nifedipine (C). D: nifedipine-sensitive difference current (ICa,L) obtained by subtracting C from B. E: peak I-V relation for ICa,L in newborn SNC (n = 14-15) and adult SNC (n = 8-10); the curves differed significantly (two-way ANOVA). F: voltage dependence of the time of half decay of ICa,L (T1/2); the curves do not differ (two-way ANOVA).

Kinetics of decay. Figure 4F shows the voltage dependence of half-decay time (T_1/2) of I_Ca,L in newborn and adult SNC. The two curves have a similar shape, with the fastest T_1/2 in the range of 20 to 0 mV. Two-way ANOVA revealed no significant difference (P = 0.739) between the curves.

Activation of I_Ca,L. For I_Ca,L, V_rev was 55 mV for both newborn and adult. As shown in Fig. 5, the I_Ca,L activation curve for adult SNC was parallel to that for newborn SNC (inverse slope factor: 6.9 ± 0.2, n = 8-10, and 7.3 ± 0.1 mV, n = 14-15, respectively), but the curve for adult SNC was shifted to more negative potentials by ~5 mV (V_1/2: 22.3 ± 1.4 mV in the adult and 17.3 ± 0.8 mV in the newborn, P = 0.001).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Effect of age on activation and availability of ICa,L. Activation (d) and steady-state inactivation (f) curves for ICa,L and their Boltzmann fits (lines) in newborn and adult SNC are shown. The activation curve for adult SNC shifted negative, whereas the inactivation curve shifted positive, relative to the corresponding curves for newborn SNC, resulting in a wider "window" current in adult SNC. V1/2 for both activation and inactivation are significantly different between the two age groups (see text).

Steady-state inactivation. As apparent in Fig. 5, the threshold for inactivation was approximately 60 mV for both the newborn and adult; at 40 mV, 30% and 15% of I_Ca,L was inactivated in newborn and adult SNC, respectively. Boltzmann curves best fitting the data from newborn and adult SNC were parallel (inverse slope factors: 6.6 ± 0.3, n = 7, and 7.1 ± 0.4 mV, n = 7, respectively) but contrary to the results on activation, the curve for adult SNC shifted to more positive potentials; the shift was significant (V_1/2: 33.4 ± 1.4 and 28.3 ± 1.7 for newborn and adult SNC, respectively, P = 0.038).

Recovery from inactivation. The results of these experiments are presented in Fig. 6, with individual current traces for a 10-day-old SNC shown in Fig. 6A and the mean time dependence of recovery plotted in Fig. 6B. In newborn SNC, I_Ca,L recovered more slowly than in adult SNC. The time course of recovery was best described by a biexponential function, the time constant () for the fast component being significantly different (112.2 ± 13.4 ms, n = 6, and 63.1 ± 8.5 ms, n = 5, for newborn and adult SNC, respectively, P = 0.017).  of the slow component also tended to be larger in newborn than adult SNC (2,538 ± 1,476 vs. 958 ± 254 ms), but the difference was not significant (P = 0.363).

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Effect of age on recovery from inactivation of ICa,L. A: two-pulse protocol and typical traces of nifedipine-sensitive current for a 10-day-old SNC. Only the first 7 of 12 traces are shown. B: time dependence of recovery from inactivation. Curves were fit by a biexponential function (lines), and the time constant for the faster component was significantly different (see text). Inset, expansion of early portion of curve.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We studied the functional expression of two types of calcium channels in SNC to assess their possible role in action potential generation in young and adult hearts. The results show no differences in current density, activation, or availability of I_Ca,T in newborn SNC compared with adult SNC. On the contrary, significant developmental changes were found in I_Ca,L.

In this study, I_Ca,T was identified as a difference between current containing both I_Ca,T and I_Ca,L and current containing only I_Ca,L. As expected for a specific dissection of the I_Ca,T component, the difference current had fast activation and decay and sensitivity to nickel. As previously mentioned, adult rabbit SNC have very little or no fast I_Na (2). The current identified as I_Ca,T was not the TTX-sensitive I_Na expressed in newborn SNC, because 1) the difference current in newborn SNC had the same properties as in adult SNC, 2) 100 µM nickel does not inhibit the fast I_Na of newborn SNC (2), and 3) the superfused solution contained reduced Na⁺ and 30 µM TTX, a concentration three orders of magnitude greater than the IC₅₀ for I_Na block in newborn SNC (26 nM) (4).

In our study, no significant age-related differences were found in I_Ca,T density, activation, and availability. These results differ from those in several studies of rabbit ventricular myocytes. In one study (36), no I_Ca,T was found in both newborn and adult rabbit ventricular myocytes. In another study (44), I_Ca,T was detected and reduced in neonatal compared with adult cells; however, a detailed analysis of activation and inactivation relations was not included. In the case of rabbit SNC, further studies are required to determine which subunits of the T-type calcium channel are expressed developmentally. It was recently shown (6) that the adult mouse sinus node has transcripts for two T-type calcium channel isoforms, Cav3.1 and Cav3.2 (_1G- and _1H-subunits, respectively; see Ref. 23), with a dominant expression of Cav3.1. The pattern of expression of the isoforms in the newborn sinus node is not known. However, the two isoforms do differ in their sensitivity to nickel when tested in heterologous expression systems, with Cav3.2 being the more sensitive (29). Given that we did not observe a significant difference either in the percent block by 100 µM nickel or any biophysical difference in current characteristics in newborn SNC versus adult SNC, there is no reason to propose the presence of a different isoform expression pattern in the newborn.

The relatively large density of I_Ca,T in adult SNC and the sensitivity of SNC action potentials to nickel (which blocks I_Ca,T in SNC) led to the suggestion that I_Ca,T contributes to the later two-thirds of diastolic depolarization and plays an important role in impulse initiation in SNC (20). The present study did not directly address this issue. However, the observation that I_Ca,T is functionally mature in the newborn SNC and yet cells at this age exhibit slow or no spontaneous rate in the absence of I_Na (2) suggests that I_Ca,T is not sufficient to assure normal impulse initiation. The lack of developmental changes in I_Ca,T found in our experiments also makes I_Ca,T an unlikely candidate as a current substituting for I_Na in the adult SNC. In terms of the specific biophysical characteristics of I_Ca,T in SNC, the activation data reported here agree with those reported by two other groups (20, 30). The inactivation relation in our study was more shallow than that reported by Hagiwara et al. (20) but comparable with that of Lei et al. (30).

The following differences in I_Ca,L were found in this study for adult SNC versus newborn SNC: 1) current density decreased, 2) activation curve shifted to the left, 3) steady-state inactivation (availability) curve shifted to the right, and 4) recovery from inactivation became faster. Despite the greater current density in the newborn, there was no difference in the time course of inactivation. Because this time course includes a calcium-dependent component, this suggests that the local calcium concentration at the inner surface of the channel may not differ with age. This might argue against either a greater calcium influx per channel in the neonate or a significant channel clustering. However, one should remember that studies (21, 39, 40) in adult SNC have provided evidence of a contribution of intracellular calcium regulation to automaticity and that there is evidence in the ventricle (but not yet in the sinus node) for developmental regulation of calcium regulatory mechanisms (18). Therefore, the possibility of developmental differences in calcium homeostasis and/or its contribution to automaticity in SNC merits future consideration.

The opposite developmental shift in activation and inactivation curves results in a wider window current in adult SNC, potentially providing more current in the voltage range of diastolic depolarization. However, this is opposed by a smaller current density in adult SNC. To estimate the current available at window voltages, we calculated I_Ca,L density multiplied by corresponding values of I_Ca,L availability (Fig. 7). This current represents the size of I_Ca,L that would be available in steady-state conditions. Available I_Ca,L was greater in adult SNC than in newborn SNC at voltages from 45 to 5 mV. Voltages corresponding to the diastolic depolarization range (50 to 35 mV) are of particular interest. At 40 mV, the calculated available current in adult SNC was 1.9 times that in newborn SNC (1.38 and 0.74 pA/pF, respectively). Verheijck et al. (42) suggested that I_Ca,L activates at more negative potentials than those reported here in adult SNC and thus could contribute to the entire range of diastolic depolarization.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   ICa,L available at steady-state conditions (window current) in newborn and adult SNC. Values were found as a product of availability of ICa,L (see Fig. 5) and current density (see Fig. 4E) at the same voltages.

In agreement with our results on rabbit SNC, Wetzel et al. (45) have shown that, in adult rabbit ventricular myocytes, I_Ca,L inactivates at more positive potentials than in fetal and neonatal myocytes, resulting in a wider window current and suggesting that increasing availability of I_Ca,L may be a common feature of developing rabbit cardiac cells. On the other hand, Osaka and Joyner (36) found no difference in activation or inactivation of I_Ca,L in ventricular myocytes in newborn versus adult rabbits.

In addition to steady-state inactivation and the threshold of activation, the kinetics of recovery from inactivation may further limit the contribution of I_Ca,L to impulse initiation in newborn SNC. We found that, in adult SNC, I_Ca,L recovers significantly faster than in newborn SNC ( of the fast component of recovery was 63.1 ± 8.5 ms in adult SNC vs. 112 ± 13.4 ms in newborn SNC). I_Ca,L can effectively recover from inactivation beginning with the end of action potential repolarization. On the basis of a study (22) in newborn and adult intact sinus nodes of rabbits, the time from the end of repolarization to the point at which diastolic depolarization reaches action potential threshold can be estimated to be 150 ms at physiological temperature. If recovery kinetics during diastolic depolarization mirror the data shown in Fig. 6 (which is based on a fixed potential of 50 mV), then one would predict that I_Ca,L recovers to 60% and 75% of maximal values in newborn and adult SNC, respectively. Thus, in the newborn, the negative shift of the steady-state inactivation relation (Fig. 5) combined with the slower recovery from inactivation in the newborn are likely to result in less I_Ca,L available in late diastole, whereas the positive shift of the activation relation will require depolarization to less negative voltages to achieve threshold for I_Ca,L activation.

There are several possible explanations for the observed changes in parameters of I_Ca,L with development. There could be an age-dependent isoform switch of one or more channel subunits. Four isoforms have been described for the pore-forming ₁-subunit of L-type calcium channels, among them _1C (cardiac form) and _1D (neuroendocrine form) (23) as well as splice variants (14). Bohn et al. (6) demonstrated that the adult mouse sinus node has clearly detectable mRNAs for Cav1.2 (_1C-subunit) as well as a low level of Cav1.3 (_1D-subunit). As noted by these authors, Cav1.3 may be of functional importance in SNC because mice lacking _1D have abnormalities in sinus rhythm (38). If rabbit SNC also have more than one type of ₁-subunit, it would be important to assess the relative prevalence of the isoforms as a function of development. One also must consider alterations in the relative abundance or isoform expression of other channel subunits. Furthermore, changes in membrane localization and/or cytoskeletal interaction are reported to impact on the functional expression of currents, including Na⁺ (8), K⁺ (7, 33), and Ca²⁺ currents (28, 37).

In addition, changes in calcium channel function may be secondary to the developmental changes of other calcium signaling components and/or their interaction with calcium channels. Recently, it has been shown that inhibition of Ca²⁺/calmodullin-dependent protein kinase II (CaMKII) in adult rabbit SNC shifted the I_Ca,L availability curve negative by 11 mV (43) and significantly prolonged recovery from inactivation. The data suggest the existence of strong control of steady-state inactivation of I_Ca,L by CaMKII. In the rat heart, the predominant expression of the functionally important -subunit of CaMKII switches from the ₄-isoform in fetal and neonatal hearts to the ₃-isoform in adult hearts (19), which, in theory, can provide a basis for functional modification of calcium current. Furthermore, other signaling cascades that regulate the basal phosphorylation level of calcium channels are known to be subject to developmental changes in the rabbit ventricle (26, 27), and a similar age-dependent regulation could occur in the sinus node. In this regard, it is worth remembering that I_Ca,L was measured with the perforated patch technique. Thus the intracellular environment was minimally disrupted, and measured differences in ionic currents could in part be secondary to differences in the intracellular environment (e.g., divalent cation concentration) as a function of age.

In conclusion, we found significant developmental changes in I_Ca,L but not I_Ca,T in rabbit SNC that could contribute to the deficiencies in neonatal automaticity after block of I_Na. There are changes in activation, steady-state inactivation, and recovery from inactivation of I_Ca,L, resulting in the likely higher availability of this current in adult SNC during late diastole at an age when I_Na no longer significantly contributes to SNC diastolic depolarization (3) or impulse initiation. Finally, it should be noted that this does not rule out additional differences in other currents in the sinus node as a function of age. A previous study (1) reported that the pacemaker current exhibited greater current density with no change in the voltage dependence in the neonate, thus arguing against a deficiency of that current in the newborn. However, the role of developmental changes in potassium currents in the sinus node remains unexplored. In addition, it is known that the adult sinoatrial node exhibits regional heterogeneity with respect to ion channel expression and function (48); the extent of heterogeneity within the neonatal sinoatrial node has not been fully explored.