INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

POSTNATAL EVOLUTION of sympathetic innervation occurs concurrently with a decrease in action potential duration (APD) in rat and dog ventricular muscle (32). We have previously demonstrated that daily injection of newborn rats for 10 days with nerve growth factor (NGF) or NGF antibody (Ab), respectively, accelerates or delays cardiac sympathetic innervation (29). In this earlier study, alteration in the time course of cardiac sympathetic innervation was confirmed using tissue catecholamine determinations, tyrosine hydroxylase assays, and the release of norepinephrine by tyramine infusion. With this method, we previously found that acceleration of the time course of innervation consistently shortens the electrocardiographic Q-T interval and ventricular APD (29, 36).

K⁺ carries the major outward currents that determine repolarization, thereby controlling the duration of the action potential. The progressive shortening of APD in rat heart during postnatal development (26) has been ascribed to evolution of the transient outward K⁺ current (I_to) and the inwardly rectifying K⁺ current (I_K1; see Refs. 37 and 38). In addition, K⁺ currents are important targets for the actions of many neurohumors and intracellular second messengers that modulate repolarization (18). Therefore, we used protocols for sympathetic modulation similar to those employed previously (29) to determine the relationship between sympathetic innervation and the electrocardiogram and action potential and the K⁺ currents I_to and I_K1 in rat epicardial myocytes.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Treatment of rats. The method of injection with NGF, placebo, or Ab was described by Malfatto et al. (29). Littermate Wistar rats born in our animal care facility were randomly separated into three groups (3-4 animals/group per litter). Each rat was injected subcutaneously from days 0 through days 13-15 of life. One group was treated with 1.0 µg/day NGF on days 0-5, 1.5 µg/day on days 6-10, and 2 µg/day on days 11-15. A second group was administered 10 µl/day NGF Ab on days 0-5, 15 µl/day on days 6-10, and 20 µl/day on days 11-15. The third group received normal saline (the solvent for NGF and Ab) as a placebo. All solutions were administered in a final volume of 100 µl. NGF and Ab were purchased from Collaborative Research (Bedford, MA). Electrocardiograms of the animals in the three groups were recorded on days 0, 3, 6, 9, and 11, as reported in previous studies, and manifested comparable results (data not shown; see Refs. 29 and 36).

Preparation of myocytes. Epicardial myocytes were isolated on days 13-15 using a method modified after Liu et al. (28) . On the day of study, two to three rats were narcotized with CO₂ and decapitated, and the hearts were rapidly excised. Only the apical two-thirds of the ventricles were used to avoid contamination by atrial cells and the great vessels. The ventricles were cut open, and the endocardium and septum were removed. The remaining epicardium was washed three times in a solution (dissociation solution) containing the following (in mM): 110 NaCl, 5.4 KCl, 4 NaHCO₃, 1.6 MgCl₂ · 6H₂O, 1.8 NaH₂PO₄, 20 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), 5 glucose, 4 L-glutamine, and 10 taurine. The epicardium was then chopped into small pieces that were bathed in the dissociation solution, also containing 3 mg/ml trypsin (Sigma Chemical, St. Louis, MO) and 5 mg/ml bovine albumin (Sigma), and placed in a Gyrotory water bath at 37°C for 15 min. The first incubation solution was discarded to remove connective tissue and debris. There were four successive changes of enzyme solution under these conditions, and the tissue was triturated with a Pasteur pipette after each incubation. After the remaining tissue pieces were removed, the enzyme solution was collected and centrifuged. The cells were then resuspended using the dissociation solution containing 5 mg/ml of bovine albumin. All data are based on studies of at least three to four separate litters of animals. Only rod-shaped, quiescent cells were used for electrophysiological experiments. The cells from the three treatment groups were morphologically similar at the light microscopic level, although the Ab-treated cells had a smaller mean membrane capacitance (see RESULTS).

Recording techniques. Epicardial myocytes were transferred to a 0.5-ml Lucite bath on the stage of an inverted microscope (Nikon Diaphot; Nikon Instrument, Tokyo, Japan). The cells were allowed to settle onto a glass coverslip coated with poly-L-lysine to facilitate adhesion of the cells to the glass. The cells were superfused at a rate of 2 ml/min with Tyrode solution (see below) for action potential recording. The standard gigaohm seal, whole cell recording method (17) was used to measure the action potentials and the ionic currents. The voltage-current clamp circuit was provided by an Axopatch 1C patch-clamp amplifier (Axon Instruments, Foster City, CA). Patch electrodes were made using borosilicate glass capillary tubes and a micropipette puller (model P 80/PC; Sutter Instrument, Novato, CA). The pipettes had a resistance of 3-4 M. Data were sampled using pCLAMP software (5.5) and a TL1 analog-to-digital interface (Axon Instrument) and were stored on disk using a Dell computer. Data were analyzed using pCLAMP 6.0. The experiments were performed at 31°C.

Recording solutions. The action potentials were recorded in Tyrode solution containing (in mM) 131 NaCl, 1.8 NaH₂PO₄, 18 NaHCO₃, 5.5 dextrose, 0.5 MgCl₂ · 6H₂O, 4 KCl, and 2.7 CaCl₂ and were gassed with 95% O₂-5% CO₂. The recording pipette solution contained (in mM) 120 potassium aspartate, 30 KCl, 1 MgCl₂ · 6H₂O, 5 MgATP, 10 HEPES, and 5 glucose. I_to was measured in a bath solution (5) containing (in mM) 144 N-methyl-D-glucamine chloride, 5.4 KCl, 1 MgCl₂ · 6H₂O, 2.5 CaCl₂, 10 HEPES, and 0.3 CdCl₂. N-methyl-D-glucamine was used to replace Na⁺ in the bath solution, and CdCl₂ was employed to block Ca²⁺ current. The pipettes were filled with a solution containing (in mM) 140 KCl, 1 MgCl₂ · 6H₂O, 5 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 5 MgATP, 5 Na₂-creatine phosphate, 0.2 GTP, and 10 HEPES. In voltage-clamp experiments, 4-aminopyridine (4-AP) at a concentration of 2 mM was used to define I_to as the 4-AP-sensitive current (5). In the action potential studies, a lower concentration of 0.5 mM was employed to avoid potential nonspecific actions of 4-AP (27).

The I_K1 measurements were carried out in a bath solution containing (in mM) 140 NaCl, 5.4 KCl, 2 MgCl₂ · 6H₂O, 10 glucose, 10 HEPES, 1.8 CaCl₂, and 0.3 CdCl₂. In some experiments, BaCl₂ (2 mM) was dissolved in this solution to block I_K1. The composition of pipette solution for I_K1 was as follows (in mM): 120 potassium aspartate, 30 KCl, 1 MgCl₂ · 6H₂O, 5 MgATP, 11 EGTA, 1 CaCl₂, and 10 HEPES. All solutions were filtered (0.22 µm) before use. pH was adjusted to 7.2 with KOH for pipette solutions and 7.4 for bath solutions with HCl. None of the data have been corrected for liquid junction potential, which was determined to be 8 ± 0.5 mV (n = 4) and 4 ± 0.4 mV (n = 4) for I_K1 and I_to solutions, respectively.

Data analysis and curve fitting. I_to and I_K1 were expressed as current density. Curves were compared using nested analysis of variance unless otherwise indicated in the text, and individual data points were subsequently compared using a Bonferroni t-test for multiple comparisons. The significance level was determined at P < 0.05. In addition, the inactivation curves for I_to in the three treatment groups were fit by the Boltzmann equation, I = (I_max  I_min)/{1 + exp[(V  V_1/2)/S]} + I_min, where I_max and I_min are maximal and minimal currents, respectively, I is the current expressed as a function of the prepulse voltage V, V_1/2 is the voltage at which 50% of the current is inactivated, and S is the slope factor. This allowed calculation and comparison of the mean midpoint of the inactivation curve (V_1/2) and the steepness of the curve (S) between the treatment groups.

Preparation of RNA. The lower two-thirds of the left and right ventricle after 15 days of treatment with NGF, Ab, or placebo was rapidly dissected out and quick-frozen in liquid N₂. Tissue samples were then homogenized in guanidinium thiocyanate. Total RNA was prepared by pelleting the homogenate over a CsCl step gradient. All RNA samples were quantified by spectrophotometric analysis.

Ribonculease protection assay. DNA templates for the preparation of RNA probes were prepared as described previously (10, 11). In all cases, a significant amount of nonhybridizing sequence (50 bp) was included in the probe to allow easy distinction between the probe and the specific protected band. The specificity of the assay was such that there was no evidence for unwanted cross-reaction between any probe and another nonspecific K⁺ channel transcript.

Ribonuclease (RNase) protection assays were performed as described previously (10). For each sample point, 10 µg of total RNA were used in the assay. A probe for the rat cyclophilin gene (9) was included in the hybridization as an internal control to confirm that the sample was not lost or degraded during the assay. Five micrograms of yeast tRNA were used as a negative control to test for the presence of probe self-protection bands. RNA expression was quantified directly from dried RNase protection gels using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

APD and membrane potential measurements. Current-clamp experiments were carried out using dissociated ventricular epicardial myocytes from rats treated with NGF, Ab, or placebo. The cells were stimulated via the patch electrode using suprathreshold constant-current pulses at 0.2 Hz. Figure 1 shows action potential records from three individual cells treated, respectively, with Ab, placebo, or NGF. The table in Fig. 1 shows the data from the three sets of cells. This limited number of action potentials was sufficient to confirm consistency with our previously reported results from intact myocardium (29, 36), i.e., hearts from rats treated with Ab have significantly longer APD at 90% repolarization (APD₉₀) compared with NGF-treated animals, and results from placebo-treated rats are intermediate. Although the decrease that we report in maximum diastolic potential in Ab-treated rats is not significant with this small n value (Fig. 1, P > 0.05), it is of the same magnitude as that described by us previously in a larger series of animals, in which the change was significant (36).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Action potentials recorded from epicardial myocytes of rats treated with nerve growth factor (NGF) antibody, placebo, or NGF. Action potentials in single cells were evoked at 5-s intervals by a suprathreshold depolarizing current pulse delivered through the recording pipette. Action potential duration at 50% (APD50) and 90% (APD90) repolarization and maximum diastolic potential (MDP) are displayed in the table. APD90 in antibody-treated epicardial myocytes is significantly longer than the NGF group. The placebo group is intermediate. Data obtained from 4 litters.

I_to. To test the role of I_to in determining APD in young rats, we studied the effect of the I_to blocker 4-AP (0.5 mM) on APD of myocytes from rats treated with placebo. We limited the concentration used in these action potential studies because we had previously demonstrated the importance of nonspecific effects of high 4-AP doses on APD in a prior study (27). Because 0.5 mM 4-AP may not fully block I_to, these experiments provide a lower limit on the contribution of I_to to APD. Figure 2 demonstrates that 0.5 mM 4-AP reversibly prolonged the APD by 26% (P < 0.05), suggesting that I_to is critically involved in action potential repolarization.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Effect of 4-aminopyridine (4-AP) on action potential duration in epicardial myocytes treated with placebo. Left: action potential recorded from a single epicardial myocyte before, during, and after administration of 0.5 mM 4-AP. Action potentials were evoked as described in Fig. 1. Right: table of APD90 values from a group of cells before, during, and after 4-AP. APD90 was significantly increased by 0.5 mM 4-AP, and this effect was readily reversed on washout. Data are expressed as means ± SE. * P < 0.05 cf. control (paired Student's t-test); n value is 5 for control and 4-AP and 3 for washout.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Sensitivity of transient outward current (Ito) to 4-AP in the different treatment conditions. Top: records from 2 cells of animals treated with antibody or NGF. Cells were depolarized to a test potential of +50 mV from a holding potential of 80 mV. , Control, i.e., before application of 4-AP; , effect of 2 mM 4-AP. Ito was almost completely blocked by 4-AP. Bottom: data from groups of cells are expressed as means ± SE at the +50 mV test potential. Percentage of inhibition was used to compare the extent of blockade by 4-AP in the 3 treatment groups.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Density change of Ito in epicardial myocytes from rats treated with antibody, placebo, and NGF. Cells were depolarized to various potentials from a holding potential of 80 mV at a rate of 0.1 Hz. Current (I)-voltage (V) relationship was plotted after the currents were normalized to their capacitance. Density was decreased in the myocytes treated with antibody compared with those treated with NGF at potentials positive to +40 mV. * P < 0.05, antibody group vs. NGF group. Placebo group did not differ from the treated groups. Normalized I-V relations for NGF and antibody (AB) are plotted in inset and superimpose, indicating that the voltage dependence for Ito activation was not shifted. Current at +60 mV was used as maximal current (Imax).

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Effects of NGF and antibody on steady-state inactivation of Ito. Protocol used is shown in inset. A 500-ms prepulse to a range of potentials was followed by a 300-ms test pulse to a fixed potential to elicit Ito. These paired pulses were delivered at 0.1 Hz from a holding potential of 80 mV. Amplitude measurements were normalized to the value obtained at the most negative potential in each experimental condition. Data from groups of cells are expressed as means ± SE, and the curves are fit by the Boltzmann equation. Voltage at which 50% of the current is inactivated (V1/2) and slope factor were calculated from the equation described in METHODS. V1/2 of the inactivation curve from the cells treated with antibody was shifted 13 mV when compared with NGF and placebo (* P < 0.05).

I_K1. The method for determination of I_K1 is depicted in Fig. 6A. The voltage was stepped from 100 to 0 mV from a holding potential of 40 mV, evoking a time-dependent inwardly rectifying current that decayed to a sustained, apparently steady-state level at the end of the 250-ms pulse. This current was completely blocked by 2 mM Ba²⁺. It manifests significant inward current at negative potentials and marked inward rectification at potentials positive to 80 mV. A small negative slope was observed in some of the cells, which was also reported by Wahler (37). Characteristics such as Ba²⁺ sensitivity, inward rectification, inactivation of inward current, and negative slope indicate that the current is indeed I_K1 (Fig. 6A).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Determination of peak and steady-state inward rectifier K+ current (IK1) in an epicardial myocyte. A: current records before (control) and in the presence of 2 mM Ba2+. Difference records, i.e., Ba2+-sensitive current, are at right. Peak currents were measured at the beginning of the pulse, and the steady-state I-V relationship was measured at the end of the pulse Vh, holding potential. B: I-V relationship of peak IK1 in the 3 groups. C: I-V relationship of steady-state IK1 in the 3 groups. Currents were normalized to their capacitance. Myocytes in antibody-treated group had the smallest density of both peak and steady-state IK1 when compared with NGF (* P < 0.05). Placebo group did not differ from either treated group.

Predicted relation between repolarization of action potential and I_to and I_K1 current density. The current density of I_to in Ab-treated rats is ~70% of that in NGF-treated rats, as shown in the I_to I-V relationship curves (Fig. 4). The difference between these two groups is greatest as potentials become more positive and only reaches statistical significance at +40 mV (Fig. 4), a voltage not within the physiological range. A similar consideration applies to the I_K1 data. Nonetheless, there are significant differences in repolarization of action potentials between the two treatment groups (Fig. 1). These observations led us to measure the slope of action potential repolarization at two voltages, 0 and 40 mV, to represent ranges at which I_to and I_K1, respectively, would be expected to contribute to repolarizing current. The location of these voltages during action potential repolarization can be appreciated by examining Fig. 2, which displays a voltage axis to the left of the action potentials. If the action potentials generated are membrane action potentials, then the current flowing across the cell membrane at any time can be calculated according to the equation I = CdV/dt, where C is capacitance; therefore, I/C = dV/dt. I/C is the current density and the unit of measurement is pA/pF, and dV/dt is the slope of action potential repolarization. The predicted current density (calculated from the action potential repolarization) around 0 mV is 3.15 pA/pF (n = 5) in NGF-treated rats and 1.92 pA/pF (n = 4) in Ab-treated rats. Therefore, the predicted current density around 0 mV in Ab-treated rats is 61% that of NGF-treated rats. The actual I_to current densities at 0 mV obtained from our studies are 2.01 (n = 16) and 1.47 pA/pF (n = 14) in NGF- and Ab-treated rats, respectively, indicating that the Ab-treated group has an I_to density that is 73% of the NGF group. It has been reported that Cd²⁺ (100 µM) shifts the availability of I_to positive by 10 mV (1). Because Cd²⁺ was present in our recording solution, we therefore also calculated I_to at +10 mV. The ratio of Ab and NGF is 77% (2.66 vs. 3.44 pA/pF). The good agreement between the current measurements and slope calculation suggests that the difference in actual I_to density in these two groups probably is a significant contributor to the observed difference in early action potential repolarization.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In our previous work (29), we found that 1) norepinephrine level in heart is highest in NGF-treated rats and lowest in Ab-treated rats when compared with placebo; 2) tyrosine hydroxylase-positive material in the ventricles of NGF-treated animals is greater than in placebo-treated animals; and 3) positive chronotropy in response to tyramine is greatest in NGF-treated animals and smallest in Ab-treated animals. These data, together, support our contention that NGF treatment increases and Ab attenuates the functional sympathetic innervation of the heart. The similarity in APD measurements in our previous and present results confirm that treatment with Ab in these animals is sufficient to retard the decrease in APD that is seen with normal development and with NGF treatment. In our earlier work, we also noted that the membrane potential of Ab-treated rats was lower than that of placebo- or NGF-treated rats. Both the lower membrane potential and the long Q-T interval seen may be explained, at least in part, by the differences we have found in the density of I_K1 and I_to.

The current that we identified as I_to was recorded in Na⁺-free solution and in the presence of Ca²⁺ channel blockade, with 0.3 mM CdCl₂. It was an outward current with a time-dependent inactivation consistent with an earlier description of I_to (23). The external application of 2 mM 4-AP rapidly reduced the peak amplitude of the inactivating component of the outward current without significantly affecting the sustained current component, as observed in other studies (5, 6). The 4-AP sensitivity of this current further confirms that it is I_to. That I_to is the major repolarizing current responsible for the alteration of APD after injection of NGF or Ab to the neonatal rat is suggested by the following: 1) there is a lower I_to density in Ab-treated rats than NGF-treated animals; 2) the magnitude of reduction at 0 mV is consistent with the change in repolarization slope at this voltage; 3) APD of epicardial myocytes from rats treated with placebo is prolonged by low concentrations of 4-AP, which preferentially blocks I_to; and 4) I_to plays an important role in the shortening of APD in rat heart during development (38).

That I_to density increases severalfold during the postnatal development of heart has been reported in freshly dissociated rat ventricular myocytes (38), cultured rat ventricular myocytes (25), and human atrial myocytes (8). Jeck and Boyden (22) reported that neonatal canine ventricular myocytes completely lack a definable I_to until ~2 mo of age. Hence, the majority of earlier studies are consistent with a developmental increase in I_to, the timing of which roughly parallels that of sympathetic innervation. This is consistent with our results with NGF and Ab, although we did not see the severalfold change in I_to density reported developmentally by others. This may be due to the fact that NGF and its Ab modulate sympathetic development only to a degree and/or that sympathetic innervation, in its own right, may be a modulator rather than the prime determinant of I_to changes during development.

No shift of I_to activation was observed by us. However, the I_to inactivation curve shifted toward more positive potentials in the Ab-treated groups, which is in contrast to reports of no change in inactivation during development in human atrial (16) and rabbit ventricular myocytes (33). This could arise from species differences and/or regional variations in the ventricular wall (2, 7) and/or a nonspecific action of Ab. However, this shift is unlikely to impact on the contribution of I_to to action potential repolarization, since, in all treatment groups, the inactivation was fully relieved at the resting potential.

That the density of peak and steady-state I_K1 was smaller in the Ab-treated animals than in animals treated with NGF suggests a role for sympathetic innervation in the evolution of the current. This finding is consistent with reports that I_K1 increases developmentally in embryonic chick ventricular myocytes (24), young rabbit ventricular myocytes (21), and freshly isolated rat ventricular cells between neonatal days 3 and 10 and then remains constant through adulthood (30, 37). Such developmental changes may reflect a variety of influences on the current, including that of the sympathetic nervous system. The changes seen in I_K1, in turn, could reflect differences in channel number, conductance, and/or open probability (30, 37).

Changes in I_K1 impact on the action potential by influencing the resting potential (35) and the final phase of repolarization (14, 19). Hence, an increase in I_K1 can shorten APD and hyperpolarize the membrane. Although we did not observe a significant change in maximum diastolic potential in our three treatment groups, the membrane potential was lower in the Ab than in the placebo or NGF groups. In an earlier report, we also found a lower membrane potential in the Ab group, consistent with the present observation of reduced I_K1 (36). In this respect, the magnitude of membrane potential difference was comparable to that in the present study, but the n was larger, and the result was statistically significant.

Our data clearly suggest that the densities of I_to and I_K1 are lower in rats treated with NGF Ab than those treated with NGF for the first 15 days of life and that the reduced densities of I_to and I_K1 may contribute to the prolongation of APD. Our investigation of K⁺ channel transcript abundance in hearts from the three treatment groups showed no significant differences in message levels. Given the modest difference in current seen among the three groups, this result is not surprising. The change in transcript level may have been below our limit of resolution. Alternatively, it may be that innervation or NGF regulates current level by affecting the channel posttranscriptionally or otherwise modifying current density.

Finally, although we have focused on I_to and I_K1, this does not rule out additional changes in other currents during development and cardiac innervation, which thereby impact on action potential configuration. For example, it has been reported that L-type Ca²⁺ current density increases developmentally in the rat ventricle (15), and nerve-muscle coculture experiments suggest that sympathetic innervation may play a role (31). However, due to additional developmental changes in channel kinetics (as measured in rabbit), the contribution of the current during normal activity may not increase and may even decrease with age (39). Another current worthy of future consideration is the Na⁺/Ca²⁺ exchange current, which has been reported to contribute an inward current that would slow terminal repolarization in the rat ventricle (34). Furthermore, the message levels of the Na⁺/Ca²⁺ exchanger decrease developmentally in both rat and rabbit as the sarcoplasmic reticulum develops (4), which would be consistent with a developmental reduction in the slowing of repolarization and thus a shorter APD in the adult ventricle. However, this developmental decrease in message level of the exchanger has been attributed to the postnatal surge in thyroid hormone level (3) rather than the ontogeny of sympathetic innervation.