Developmental Changes in Electrophysiological Properties and Synaptic Transmission in Rat Intracardiac Ganglion Neurons Katrina Rimmer and Alexander A. Harper
J Neurophysiol 95:3543-3552, 2006. First published 12 April 2006; doi: 10.1152/jn.01220.2005 You might find this additional info useful... This article cites 29 articles, 14 of which you can access for free at: http://jn.physiology.org/content/95/6/3543.full#ref-list-1 This article has been cited by 5 other HighWire-hosted articles: http://jn.physiology.org/content/95/6/3543#cited-by Updated information and services including high resolution figures, can be found at: http://jn.physiology.org/content/95/6/3543.full
This information is current as of June 2, 2013.
Journal of Neurophysiology publishes original articles on the function of the nervous system. It is published 12 times a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2006 by the American Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at http://www.the-aps.org/.
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
Additional material and information about Journal of Neurophysiology can be found at: http://www.the-aps.org/publications/jn
J Neurophysiol 95: 3543–3552, 2006. First published April 12, 2006; doi:10.1152/jn.01220.2005.
Developmental Changes in Electrophysiological Properties and Synaptic Transmission in Rat Intracardiac Ganglion Neurons Katrina Rimmer and Alexander A. Harper Division of Molecular Physiology, School of Life Sciences, University of Dundee, Dundee, Scotland, United Kingdom Submitted 18 November 2005; accepted in final form 27 February 2006
Intrinsic cardiac ganglia (ICG) are interconnected clusters of neurons located throughout the atrial epicardium and interatrial septum. They are innervated by the vagus nerve and send their projections to discrete regions of the heart. The final pattern of discharge in these neurons regulates chronotropic, dromotropic, and inotropic elements of cardiac function (Adams and Cuevas 2004). Traditionally the intrinsic cardiac nervous system was held as a simple relay for parasympathetic inputs to cardiac end effectors. However, recent work has clearly indicated that in the adult these peripheral ganglia are complex neural networks with multiple neuronal subtypes and are capable of complex reflex control of regional cardiac function (Ardell 2004). Autonomic control of the heart changes during early postnatal development in mammals. Direct vagal nerve stimulation studies reveal significant decreases with postnatal age in the intrin-
sic heart rate and parasympathetic control of the heart (Gootman 2004). In the rat, the parasympathetic response is established before birth, whereas functional sympathetic innervation of the heart is not present until 2 wk of age (Robinson 1996). Understanding the electrophysiological specialization of the neuronal elements and the strength and spatial distribution of interneuronal coupling within the living ganglion is a prerequisite to understanding its normal function. Appreciation of the innervation of ICG will lead to greater insight into the dysfunction and failure of this autonomic pathway, such as the diminished parasympathetic tone that is a hallmark of cardiovascular disease and aging (Lagi et al. 1999). Both the electrical properties and synaptic responses of ICG neurons must be characterized to assess changes in neuronal integration within the intrinsic cardiac nervous system that occur during development. ICGs of adult animals contain a heterogeneous population of neurons, identified on the basis of their electrical and synaptic properties (Edwards et al. 1995; Selyanko 1992; Selyanko and Skok 1992). However, it is not known when the specification of neuronal subtypes occurs. The intrinsic properties of a neuron are not static, but can alter during development by the up- or downregulation of expression of ion channels. In studies using rat dissociated ICG neurons the expression and characteristics of the hyperpolarization-activated (Ih) and inward rectifier K⫹ [IK(IR)] currents (Hogg et al. 2001), ATP-sensitive K⫹ conductance [IK(ATP)] (Hogg and Adams 2001), and ␥-aminobutyric acid type A (GABAA) receptor channels (Fischer et al. 2005) all show changes during postnatal development. There is little information regarding postnatal development of electrical characteristics in parasympathetic ganglia, although there is a reasonable cohort of data for sympathetic ganglia, where in many cases the functional phenotype changes with development. For example, Anderson et al. (2001) examined the pre- and postnatal electrophysiological properties of sympathetic neurons. This study concluded that much of the diversity of electrophysiological characteristics in the mature animal is acquired during fetal development. Parasympathetic regulation of cardiac performance originates in the CNS and traverses the parasympathetic ganglia. Synaptic activation of the neurons in the cardiac ganglia thus needs to be considered as part of this pathway from the brain to the heart. Several animal models have been used for studying synaptic responses in the mature, adult ICG: guinea pigs (Edwards et al. 1995); dogs (Bibevski et al. 2000; Xi-Moy et
Address for reprint requests and other correspondence: A. A. Harper, Division of Molecular Physiology, School of Life Sciences, University of Dundee, Dundee DD1 5EH, Scotland, UK (E-mail:
[email protected]).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
INTRODUCTION
www.jn.org
0022-3077/06 $8.00 Copyright © 2006 The American Physiological Society
3543
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
Rimmer, Katrina and Alexander A. Harper. Developmental changes in electrophysiological properties and synaptic transmission in rat intracardiac ganglion neurons. J Neurophysiol 95: 3543–3552, 2006. First published April 12, 2006; doi:10.1152/jn.01220.2005. We charted postnatal changes in the intrinsic electrophysiological properties and synaptic responses of rat intrinsic cardiac ganglion (ICG) neurons. We developed a whole-mount ganglion preparation of the excised right atrial ganglion plexus. Using intracellular recordings and nerve stimulation we tested the hypothesis that substantial transformations in the intrinsic electrical characteristics and synaptic transmission accompany postnatal development. Membrane potential (Em) did not change but time constant () and cell capacitance increased with postnatal development. Accordingly, input resistance (Rin) decreased but specific membrane resistance (Rm) increased postnatally. Comparison of the somatic active membrane properties revealed significant changes in electrical phenotype. All neonatal neurons had somatic action potentials (APs) with small overshoots and small afterhyperpolarizations (AHPs). Adult neurons had somatic APs with large overshoots and large AHP amplitudes. The range of AHP duration was larger in adults than in neonates. The AP characteristics of juvenile neurons resembled those of adults, with the exception of AHP duration, which fell midway between neonate and adult values. Phasic, multiply adapting, and tonic evoked discharge activities were recorded from ICG neurons. Most neurons displayed phasic discharge at each developmental stage. All neurons received excitatory synaptic inputs from the vagus or interganglionic nerve trunk(s), the strength of which did not change significantly with postnatal age. The changes in the electrophysiological properties of the postganglionic neuron suggest that increased complexity of parasympathetic regulation of cardiac function accompanies postnatal development.
3544
K. RIMMER AND A. A. HARPER
METHODS
Preparation Wistar rats (Harlan UK, Oxon, UK) were used at three stages of postnatal development: neonates (P2–P5), juveniles (P14 –P21), and young nonpregnant female adults (ⱖ6 wk, 150 –220 g). Rats were killed by stunning and cervical dislocation, in accordance with current UK Home Office guidelines. The heart and lungs were quickly excised and placed in ice-cold Krebs solution. The right atrial ganglion plexus and underlying myocardium were isolated from the dorsal surface of the atria. The whole-mount preparation was pinned out in the recording chamber (35-mm petri dish lined with Sylgard No. 184 [Dow-Corning, Midland, MI], approximately 1.0 ml volume) and superfused with bicarbonate-buffered Krebs solution (approximately 2 ml/min), gassed with 95% O2-5% CO2 to pH 7.4. The temperature was controlled by a Peltier heating device (Medical systems PDMI-2 microincubator) to 35°C, monitored by an independent thermistor probe in the recording chamber (Yellow Springs Instruments, Yellow Springs, OH). The tissue was left to resuscitate in these conditions for a period of ⱖ30 min before commencing recording. ICG neurons were visualized using differential interference contrast optics on a fixed-stage microscope. Recordings were normally made from the largest ganglion, located at the junction of the right superior vena cava and right atrium, which regulates the sinoatrial node (Sampaio et al. 2003).
Electrophysiological recordings Intracellular current-clamp recordings were made from postganglionic ICG somata using sharp glass microelectrodes (GC120F; HarJ Neurophysiol • VOL
vard Apparatus, Edenbridge, UK) with resistances of approximately 120 M⍀ when filled with 0.5 M KCl. Membrane voltage responses were recorded with a conventional bridge amplifier (Axoclamp 2A, Axon Instruments, Union City, CA). Voltage signals were filtered at 20 kHz (Frequency Devices 902), digitized at 50 kHz, and transferred to a Pentium 4 computer using an A/D converter (Micro 1401 MKII interface, CED, Cambridge, UK) and Spike2 software (CED). Two pulse protocols were used: short depolarizing pulses (ⱕ3 ms in duration) to directly evoke single action potentials and 200 ms long hyperpolarizing and depolarizing pulses to measure passive properties and discharge characteristics, respectively. For short pulses, action potential parameters measured were overshoot and AHP duration to 50 and 80% recovery (AHP50, AHP80), and AHP amplitude using a Spike2 script. Input resistance (Rin) was calculated from the steady state of the voltage response to small hyperpolarizing long-current pulses (聿 ⫺0.1 nA), and time constant () was measured by fitting 20 – 80% of the rising phase at 聿 ⫺100 pA with a single exponential function using Origin software (Microcal, version 6, Northampton, MA). Discharge activity was classified as being phasic, multiply adapting, or tonic on application of a current pulse at twice threshold intensity.
Nerve stimulation Branches of the vagus and interganglionic nerve trunks were stimulated using a suction electrode connected to a constant-voltage isolated stimulator (Digitimer DS2, Herts, UK). Nerve trunks were stimulated using stimulus pulses of 0.02 to 0.2 ms width, 5 to 50 V amplitude.
Solutions and pharmacological agents Krebs solution contained (in mM): 118 NaCl, 25 NaHCO3, 1.13 NaH2PO4, 4.7 KCl, 1.8 CaCl2, 1.3 MgCl2, glucose 11.1 (Smith et al. 2001), and all reagents were of analytical grade. Apamin was purchased from Alomone (Jerusalem, Israel). Data are presented as the means ⫾ SD of the number of observations indicated, and were compared using ANOVA (Tukey–Kramer multiple comparisons test), paired and unpaired t-tests, and 2 tests (GraphPad InStat software, Version 3.00, GraphPad Software, San Diego, CA) as indicated in the text. RESULTS
All results presented are from ICG neurons with a resting membrane potential of at least ⫺40 mV and overshooting somatic action potentials. Recordings were stable for ⱖ10 min before taking readings. Many neurons in the intact ICG exhibited spontaneous excitatory postsynaptic potential (EPSP) activity, and occasionally spontaneous action potentials (APs) were recorded that were overshooting. The EPSPs normally had amplitudes at least an order of magnitude greater than recording noise and had a fast rising phase and a falling phase similar to the time constant of the neuron. Such activity was prevalent in the neonatal ICG with nearly 60% of neurons (18/31) displaying spontaneous events, decreasing with postnatal age to nearly 53% (23/43) in juveniles and 33% (14/42) in adults. Cd2⫹ (100 M) inhibited spontaneous EPSPs and APs in neurons of two adults, two juveniles, and one neonatal. In contrast to previous reports (Edwards et al. 1995; Selyanko 1992) no instances of spontaneous sustained rhythmical action potential discharge, akin to pacemaker-like activity, were recorded. Sustained repetitive discharge could be evoked in some neurons on injection of small depolarizing currents (data not shown).
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
al. 1993); pigs (Smith 1999); and rats, adult (Selyanko and Skok 1992) and neonatal (Seabrook et al. 1990). Nerve-evoked responses were blocked by hexamethonium and so are mediated by nicotinic acetylcholine receptors (Bibevski et al. 2000; Selyanko and Skok 1992). However, in some neonatal rat ICG neurons synaptic transmission was resistant to blockade by the nicotinic antagonist mecamylamine, suggesting a noncholinergic component (Seabrook et al. 1990). We developed a whole-mount rat cardiac ganglia preparation that allows electrical properties and synaptic responses of neurons to be analyzed. We studied three time points during postnatal development: neonates (2–5 days), juveniles (14 –21 days), and adults (ⱖ6 wk). The juvenile time window was chosen to coincide with the development of functional sympathetic control of cardiac regulation. The postnatal gap between parasympathetic and sympathetic development affords a valuable model for studying integration of chemical signaling within ganglia. We hypothesize that generation of the several classes of neurons reported for the mature ICG occurs during postnatal development, as observed in dissociated ICG neurons (Hogg et al. 2001). However, results from dissociated neurons are not always in accordance with those from whole-mount preparations; for example, P2X receptor channels are not expressed in whole-mount postganglionic submandibular (Smith et al. 2001) and sympathetic (Inokuchi and McLachlan 1995) neurons but are present in dissociated neurons from these ganglia. If there are any alterations in electrophysiological characteristics and synaptic transmission, the next step is to determine whether they are simply associated with postnatal increases in neuronal size or whether they follow functional sympathetic innervation. Preliminary aspects of these results were previously reported (Rimmer and Harper 2004).
DEVELOPMENTAL CHANGES IN INTRACARDIAC GANGLION NEURONS
Postnatal developmental changes in postganglionic membrane properties
The active properties of adult ICG neurons were distinct from those of neonatal neurons (Table 2). Representative action potentials evoked by short (⫹0.3 nA) current pulses, from a neonatal and two individual adult neurons, are shown in Fig. 2A. The AP overshoot (OS) in neonatal neurons (approximately 9 mV) was significantly less than that from juveniles and adults (approximately 18 mV). The AHP after the action potential was characterized by its amplitude and decay time. Action potentials in neonatal ICG neurons had small amplitude AHPs, ranging from 5.8 to 14.6 mV, with a mean value of 10.3 mV. Adult AHP amplitude was significantly greater compared with the neonatal AHP, with a mean value of 17.7 ⫾ 3.1 mV (Fig. 2A and Table 2). Several ways of measuring AHP duration have been described for ICG neurons, such as time to 50% recovery, AHP50 (Edwards et al. 1995), or total AHP duration (Selyanko 1992; Smith 1999). To afford a comparison with these studies we measured both the duration of AHP50 and 80% recovery (AHP80). Arguably, the latter can be measured more accurately than time to full recovery. Scatterplots are shown for AHP50 and AHP80 values at each stage of development (Fig. 2B). ACTIVE PROPERTIES.
TABLE
Neurons from ICG in neonates had short AHPs, for which the mean duration of AHP50 was 11.3 ms and the mean AHP80 was 25.0 ms. The corresponding maximum duration was ⬍17.5 and ⬍50 ms, respectively (Fig. 2B). Juvenile and adult neurons exhibited a greater range of AHP80 values: 16.2–139 and 15.3–163 ms, respectively. The incidence of neurons with AHP80 values ⬎50 ms increases with development: 10/43 juvenile and 22/41 adult neurons had an AHP80 ⬎50 ms. No neurons recorded possessed slow AHPs (IsAHP), which have a duration of 1–2 s, reported first for myenteric neurons (reviewed in Sah and Faber 2002). Histograms were plotted of adult AHP50 and AHP80 values (Fig. 2C) to show the frequency distribution for each. Table 2 presents the mean AHP50 and AHP80 data for each stage of development. The mean values for juvenile AHP50 and AHP80 were approximately midway between that for neonates and adults (Fig. 2B). Juvenile data were separated into sets: P14 –P15, P16 –P19, and P20 –P21 days, to ascertain whether increased functional sympathetic development has any impact on active and passive membrane properties, because the sympathetic nervous system begins to appear at P14 and is fully functional by P21 (Robinson 1996). With the exception of AHP amplitude, there was no significant difference in any of the membrane properties measured, suggesting that active properties have attained nearadult values by P14. We hypothesized that the increase in AHP50 and AHP80 durations results from increasing expression of small Ca2⫹activated K⫹ channels (SKCa), whose current underlies the AHP (Sah and Faber 2002). To test this, 100 nM apamin—a selective SK channel blocker—was applied to four adult and four neonatal neurons (Fig. 3A). The AHP50 and AHP80 values in control conditions and in the presence of apamin are plotted in Fig. 3B. Apamin reduced the mean AHP50 values from 17.0 ⫾ 2.3 to 11.6 ⫾ 0.8 ms (P ⬍ 0.05) in neonates and from 31.5 ⫾ 13.0 to 10.8 ⫾ 1.0 ms (P ⬍ 0.05) in adults. The AHP80 was also shortened: 43.2 ⫾ 4.8 to 20.6 ⫾ 2.7 ms (P ⬍ 0.001) in neonates and 62.7 ⫾ 14.4 to 20.2 ⫾ 2.1 ms in adults (P ⬍ 0.001, paired t-test). Thus apamin reduces the AHP to nearly the same duration in neonates and adults. Evoked discharge To characterize firing discharge, 200 ms depolarizing current pulses, ⫹0.1 to ⫹0.5 nA, were applied. Neurons were split into three groupings on the basis of firing properties at twice threshold stimulus intensity: phasic, multiply adapting and tonic; an example of each is shown in Fig. 4, A, B, and C for neonates, juveniles, and adults, respectively. The number of
1. Postnatal changes in the passive electrical properties of ICG neurons
Em, mV Rin, M⍀
, ms C, pF Rm, k⍀ cm2
Neonate
Juvenile
Adult
P Values
⫺46.6 ⫾ 4.9 (32) 161.6 ⫾ 60.7 (32)
⫺44.7 ⫾ 4.0 (43) 127.7 ⫾ 52.5 (40)
⫺47.2 ⫾ 6.4 (44) 118.2 ⫾ 52.6 (44)
4.2 ⫾ 2.0 (32) 27.7 ⫾ 10.7 (32)
5.8 ⫾ 3.3 (40) 47.9 ⫾ 23.4 (40)
6.7 ⫾ 3.8 (44) 62.6 ⫾ 30.7 (44)
4.2 ⫾ 2.0 (32)
5.8 ⫾ 3.3 (40)
6.7 ⫾ 3.8 (44)
ns n vs. a** n vs. j* n vs. a** n vs. a*** n vs. j** j vs. a* n vs. a***
Values are means ⫾ SD; number of neurons in parentheses. ANOVA was used to calculate P values: *P ⬍ 0.05; ** P ⬍ 0.01; and *** P ⬍ 0.005; ns, not significant. Rin values calculated for current steps ⫺100 pA or less. Rm values calculated from Rin ⫻ cell capacitance, assuming 1 pF ⫽ 100 m2. J Neurophysiol • VOL
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
Membrane potential (Em) was similar: approximately ⫺45 mV at the three postnatal stages sampled. Neurons with an Em of ⬎ ⫺60 mV were normally inexcitable and were excluded from this analysis because it was unclear whether these were neurons or glial cells. These acceptance criteria may explain why the Em values reported in this study are slightly more positive compared with other data from adult rat ICG neurons (e.g., Selyanko 1992; Xi-Moy and Dun 1995). There were changes in input resistance (Rin) and time constant () of ICG neurons during postnatal development (Table 1 and Fig. 1B). In neonatal ICG neurons the input resistance (Rin) was significantly greater than that of juveniles and adults. Time constant () and calculated cell capacitance (Cin) increased with age, mirroring the increase in neuron size (Fig. 1B). This is depicted in the photomicrographs shown in Fig. 1A. A decrease in input resistance would be expected to accompany increased cell size, as indicated by cell capacitance. Interestingly, the Rin values for juvenile and adult ICG neurons were similar. Membrane resistance (Rm) values were calculated from Rin ⫻ cell capacitance (assuming 1 pF ⫽ 100 m2) and increased significantly from neonates to adults (see Table 1). This is in accordance with a previous study of dissociated ICG neurons (Hogg et al. 2001). PASSIVE PROPERTIES.
3545
3546
K. RIMMER AND A. A. HARPER
Synaptic responses
FIG. 1. A: intrinsic cardiac neurons in a living neonatal (i), juvenile (ii), and adult (iii) sinoatrial node ganglion preparation viewed with differential interference contrast optics. B: scatterplots illustrating postnatal change in (i) input resistance (Rin), (ii) time constant (), and (iii) cell capacitance (Cin). Line denotes mean value. There are significant changes in all 3 parameters from neonates to adults, which are associated with the increase in neuron size, seen in A. *P ⬍ 0.05; **P ⬍ 0.01; ***P ⬍ 0.005; ****P ⬍ 0.001.
J Neurophysiol • VOL
Electrical stimulation of branches of the vagus (preganglionic) or interganglionic nerve trunks was performed using rectangular constant voltage pulses (width 0.02– 0.2 ms; 5–50 V). The stimulus duration and amplitude required to evoke a synaptic response were normally greater for neonates than for juvenile and adult ICG neurons. This presumably reflects the caliber and state of myelination of the efferent axons, because the final steps of association between axons and Schwann cells occur in late embryonic and early postnatal stages (Jessen and Mirsky 2005). Two stimulus protocols were used: single stimuli presented at frequencies of ⱕ0.2 Hz (Fig. 7) and trains of 20 twice-threshold intensity stimuli at 5, 10, 20, and 50 Hz with a 30 s intertrain interval (Fig. 8A). The latter protocol was also used to ascertain the following frequency for direct, intrasomatic stimulation and to correlate these data with those of synaptic transmission (Fig. 8B). Synaptically evoked postsynaptic events in adult (n ⫽ 5) and neonatal (n ⫽ 5) ganglia were reversibly blocked by nominally Ca2⫹ free, high Mg2⫹ (6 mM) Krebs solution (Fig. 6A), as previously demonstrated for adult rat intracardiac ganglia (Selyanko and Skok 1992). Superfusion of 0 Ca2⫹/high Mg2⫹ Krebs solution produced a dramatic depolarization (9.9 ⫾ 6.3 mV, n ⫽ 5) in neurons from neonatal ganglia, and less so in adult neurons (3.6 ⫾ 2.8 mV, n ⫽ 5). In addition to preganglionic axons, it has been reported that the vagus and ICG nerve connectives contain postganglionic axons (Edwards et al. 1995). Stimulating the vagus may evoke APs in these axons that will be conducted antidromically to ICG neurons. In three instances putative antidromic APs were recorded, indicated by very short latencies, no apparent EPSP, and discharge of all-or-nothing somatic APs (Selyanko and Skok 1992). To test whether these action potentials were antidromic, the wide-spectrum Ca2⫹ channel blocker cadmium
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
action potentials fired at twice-threshold current intensity is plotted for each neuron in Fig. 4B: 77% of neonatal neurons were phasic, 10% multiply adapting, and 13% were tonic. Juvenile neurons exhibited phasic firing in 70% of observed cases; multiply adapting, 15%; and those remaining, 15% tonic. For adults, 76% were phasic neurons, 12% multiply adapting, and 12% tonic. Application of apamin (100 nM) changed the firing properties from phasic to tonic in the neonatal (n ⫽ 4) and adult (n ⫽ 3) neurons studied (data not shown). Hyperpolarizing current pulses can induce time-dependent rectification (TDR), held as the signature of the H-current (Pape 1996). Such behavior was observed in most neurons at all stages of development. The extent of TDR was quantified by measuring steady-state voltage response to a hyperpolarizing current pulse to approximately ⫺100 ⫾ 10 mV and expressing this as a percentage of peak membrane potential excursion. The values were 96 ⫾ 3% (n ⫽ 32), 94 ⫾ 3% (n ⫽ 42), and 94 ⫾ 3.2% (n ⫽ 44) for ICG neurons of neonates, juveniles, and adults, respectively. The adult and juvenile values were significantly different from those of neonates (P values of ⬍0.01 and ⬍0.05, respectively; ANOVA). Examples of voltage traces and current–voltage plots displaying the range of inward rectification from each developmental stage are shown in Fig. 5.
DEVELOPMENTAL CHANGES IN INTRACARDIAC GANGLION NEURONS TABLE
3547
2. Action potential and AHP properties of ICG neurons from neonatal, juvenile, and adult rats Neonate
Juvenile
Adult
P Values
APos, mV AHP, mV
9.4 ⫾ 5.6 (31) 10.4 ⫾ 2.6 (31)
17.9 ⫾ 7.6 (43) 15.7 ⫾ 3.0 (43)
19.3 ⫾ 7.1 (41) 17.7 ⫾ 3.1 (41)
AHP50, ms
11.2 ⫾ 2.8 (31)
19.2 ⫾ 2.8 (43)
31.3 ⫾ 23.5 (41)
AHP80, ms
25.0 ⫾ 8.4 (31)
43.0 ⫾ 23.9 (43)
60.6 ⫾ 33.0 (41)
n vs. a, j*** n vs. a, j*** j vs. a** n vs. a*** j vs. a** n vs. a*** n vs. j** j vs. a**
Values are means ⫾ SD; number of neurons in parentheses. ANOVA was used to calculate P values: ** P ⬍ 0.01 and *** P ⬍ 0.005.
Single-stimuli responses Synaptic responses were subjectively classified into three groups based on the synaptic response and AP waveform: Group 1: weak, where nerve stimulation evokes a subthreshold EPSP. Group 2: secure, where AP arises late in the EPSP when present or can occur without an underlying EPSP. Group 3: strong, where AP arises early during the EPSP. These synaptic
A
(ii)
Examples of Em recordings in response to trains of stimuli delivered to the nerve trunk, at 5, 10, 20, and 50 Hz, are presented for a neonatal, juvenile, and adult ICG neuron in Fig. 8A. The ability to faithfully follow frequencies from indirect
C (i)
AHP50 = 13.7 ms AHP80 = 29.3 ms
AHP50 (ms)
0 mV
Trains of stimuli
B
I +0.3 nA
(i)
responses are shown in Fig. 7A and the relative proportion of each group at the three stages of development is shown in Fig. 7B. There is no significant change in synaptic strength between neonates, juveniles, and adults (2 test). Group 2 responses that occurred without an EPSP, although they appear similar, differ from antidromic APs in several ways. When the stimulus was reduced, a subthreshold EPSP was revealed in the orthodromic APs but not the antidromic APs. The antidromic responses recorded could faithfully follow frequencies of ⱖ50 Hz, whereas orthodromic could not.
100
100
50
50
0
AHP50 = 8.4 ms AHP80 = 26.0 ms
0
Neonatal
Juvenile
Adult
15
AHP50 = 63.4 ms AHP80 = 109.8 ms
AHP80 (ms)
(ii)
0 mV
10
5
0
No of adult neurons
0 mV
(iii)
150
(i)
150
200
200
(ii)
150
150
100
100
50
20 ms 50 ms
50 mV
50
0
0
Neonatal
Juvenile
Adult
15
10
5
0
No of adult neurons
FIG. 2. A: examples of somatic action potentials (APs) recorded from a neonate (i) and adult intracardiac neurons, one with a short afterhyperpolarization (AHP80, ii), the other with a long AHP80 (iii), evoked with brief ⫹0.3-nA, 3-ms current pulses. Top time calibration bar refers to A(i) and (ii); bottom calibration bar refers to A(iii). B: scatterplots showing the distribution of AHP50 (i) and AHP80 (ii) recovery values of neonatal, juvenile, and adult intrinsic cardiac ganglion (ICG) neurons. C: histograms showing the AHP50 (i) and AHP80 (ii) durations for adult ICG neurons.
J Neurophysiol • VOL
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
(100 M Cd2⫹), which blocks synaptic transmission, was added to the superfusing Krebs solution (Fig. 6B). Cd2⫹ did not block antidromic conduction, but did reduce AP overshoot and AHP amplitude and duration, presumably as a result of the failure of activation of Ca2⫹-activated K⫹ channels. Em did not change after addition of Cd2⫹ in neurons of the three adults, two juveniles, and one neonatal studied (grouped data were ⫺45.4 ⫾ 3.7 mV in control to ⫺44.6 ⫾ 5.0 mV in the presence of Cd2⫹).
3548
K. RIMMER AND A. A. HARPER
A
B
Neonates
Neonate (ii)
100 nM apamin
60
40
40
0 mV
0 mV
60
AHP50 (ms)
Control
(ii)
AHP50 (ms)
(i)
20
Control
20
0
Adult (iii)
Control
(iv)
100 nM apamin
0
Apamin
(iii)
Control
Apamin
(iv)
100
0 mV
Adults
(i)
100
0 mV
AHP80 (ms)
80
60
60
40
20 ms
50 mV
40
20
20
0
50 ms
0
Control
Apamin
Control
Apamin
3. A: somatic AP traces produced by injection of a brief ⫹0.3-nA pulse from a neonatal (i, ii) and adult neuron (iii, iv) in control, and after addition 100 nM apamin. Top time calibration bar refers to A(i, ii); bottom bar refers to A(iii, iv). B: AHP50 (i, ii) and AHP80 values (iii, iv) in control and in the presence of apamin plotted for neonates (i) and (iii) and adults (ii) and (iv). FIG.
nerve stimuli increased with postnatal development, with 11/17 (65%) adult and 6/17 (35%) juvenile neurons being able to follow frequencies of up to 50 Hz. Synaptically evoked responses recorded from neurons in neonatal ICG exhibit a wide range of behavior. One neuron was unable to follow 0.2 Hz, another capable of faithfully following frequencies of up to 0.5 Hz (data not shown), whereas 3/8 could follow 20 Hz and only 1/6 was able to follow 50 Hz. Figure 8B graphically depicts the average number of successful APs during a 20 AP train for both indirect synaptic responses and direct somatic-evoked discharge at each stage of development. DISCUSSION
This is the first study to directly examine the electrophysiological properties and synaptic transmission during postnatal maturation of a mammalian intracardiac ganglion. This investigation has made two novel findings. First we have demonstrated that there are clear changes in the passive and active properties of the ICG neurons between neonates, juveniles, and adults. Second, we have characterized synaptic transmission at the three stages of development and established that there is no significant change in synaptic transmission with postnatal development. These experiments were performed at a temperature, divalent ion concentration, and pH buffering system relevant to in vivo conditions. By contrast, the only study to investigate these properties for the neonatal ICG (Seabrook et al. 1990) was conducted at room temperature and with high divalent ion concentrations. Therefore direct comparisons with studies of adult ICG neurons (e.g., Selyanko 1992) cannot be made. Data from dissociated neonatal and adult ICG neurons indicate that many passive and active membrane properties are temperature J Neurophysiol • VOL
dependent, such as input resistance and patterns of evoked firing frequency (Cuevas et al. 1997). The first aim of this study was to document changes in passive and active membrane properties with respect to postnatal development. Input resistance and time constant altered with age, as did neuron size, gauged by cell capacitance with the juvenile window being approximately midway on the growth curve. The specific membrane resistance increased, in good agreement with a previous report for dissociated neonatal and adult rat ICG neurons (Hogg et al. 2001). The decrease in specific membrane resistance accompanying postnatal development may be attributed to changes in the expression of Ih and IM. Ih density in dissociated adult ICG neurons is approximately 50% of its value in neonatal neurons (Hogg et al. 2001). In the whole-mount preparation, application of the wide-spectrum K⫹ channel blocker Ba2⫹ (1 mM), which blocks the M-current, results in a greater decrease in slope conductance in neonatal compared with adult ICG neurons (Rimmer and Harper 2004). Active membrane properties of neonates and adults were likewise distinct. ICG neurons in neonates possess APs with small overshoots and shallow and short-duration AHPs. In contrast, adult neurons have large overshoots and AHP amplitudes and many neurons have AHP80 values ⬎50 ms, which is not observed in neonates. Because neonates have short AHPs, they can faithfully follow higher frequencies evoked by trains of direct, intrasomatic, depolarizing currents than juvenile and adult ICG neurons. The mean juvenile AHP duration values were midway between those for neonates and adults. This behavior is distinct from the AP overshoot and AHP amplitude, which were all marginally less than adult values. Therefore membrane properties of juvenile ICG neurons are distinct from neonates but
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
AHP80 (ms)
80
DEVELOPMENTAL CHANGES IN INTRACARDIAC GANGLION NEURONS
B Juveniles
Neonates (i) 0 mV
C
D Number of neurons
A
Adults
(i)
(i)
0 mV
0 mV
20
Neonates 15
10
5
0
0 mV
0 mV
(ii)
0 mV
0 mV
(iii)
(iii)
0 mV
0 mV
0
1
2
3
20
Number of neurons
6
7
8
9
10
7
8
9
10
7
8
9
10
Juveniles
10
5
0
1
2
3
4
5
6
Adults
20
200 ms
5
15
0
50 mV
4
15
10
5
0
0
1
2
3
4
5
6
Number of action potentials fired FIG. 4. Voltage traces in response to a 200 ms, ⫹0.2 nA, twice-threshold, depolarizing current pulses from neonate (A), juvenile (B), and adult neurons (C). A–C (i) illustrates examples of discharge activity classified as phasic; A–C (ii) gives examples of multiply adapting firing; and A–C (iii) show discharge activity classified as tonic. D: frequency histograms depicting the number of APs fired in response to a 200 ms twice-threshold depolarizing current pulse for neonates, juveniles, and adults.
FIG. 5. Representative voltage traces, in response to long depolarizing (⫹0.2 nA) and hyperpolarizing (⫺0.1 to ⫺0.4 nA) 200-ms current pulses and the corresponding current–voltage plots from neonatal [A(i–iii)], juvenile [B(i–iii)], and adult [C(i–iii)] intracardiac neurons. Range of evoked discharge and time-dependent inward rectification are shown. (E) denotes steady-state measurement and (●) denotes peak hyperpolarizing from the voltage excursion measurements.
J Neurophysiol • VOL
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
(iii)
(ii)
Number of neurons
(ii)
3549
3550
K. RIMMER AND A. A. HARPER
B
2⫹ 2⫹ FIG. 6. A: action of superfusing nominally zero Ca /high (6 mM) Mg Krebs solution on synaptic transmission in ICG neurons in a neonatal (i, ii) and adult (iii, iv) ICG ganglion. Em in (ii) was reset to native values recorded in normal Krebs solution. B: Cd2⫹ (100 M) blocks synaptic transmission in a P18 juvenile ICG neuron (i, ii) but does not prevent antidromic conduction in an adult neuron (iii, iv).
akin to adults. This indicates that evolution of the diversity of the electrophysiological properties in adult ICG neurons is not simply linked to growth of the neurons. Furthermore, the properties of juvenile ICG neurons do not change significantly during the P14 –P21 time frame, inclining us to the view that the development of functional sympathetic innervation is not responsible for these changes. The identity of the mechanism(s) underpinning the differentiation of neuron types is clearly an important area for future investigation. Developmental regulation of Ca2⫹-activated K⫹ channels (KCa) could underlie the age-dependent changes in AHP properties. The signaling pathways regulating developmental changes in the properties and expression of KCa channels in embryonic chick parasympathetic ganglion neurons has been the subject of a series of elegant studies (Cameron and Dryer 2000; Cameron et al. 1998, 1999; Dryer 1998). A similar analysis is required for mammalian ICG neurons. We can infer that an increase in SKCa channel expression underpins the changes in AHP duration that accompany development, from the results of experiments using the SK channel blocker apamin. Apamin reduced both adult and neonatal AHP50 and AHP80 durations to almost the same value. Apamin also switched the discharge characteristics from phasic to tonic in neonates and adults, implicating a role for SK channels in control of firing. In contrast, apamin produced no change in firing discharge in dissociated neonatal rat intracardiac neurons (Cuevas et al. 1997). An increase in SKCa current may also contribute to the small change in distribution of evoked firing J Neurophysiol • VOL
FIG. 7. A: examples of synaptically evoked APs recorded from juvenile ICG neurons illustrating the different strengths of synaptic transmission. B: histograms showing the relative proportion of each synaptic response group for neonates (n ⫽ 20), juveniles (n ⫽ 25), and adults (n ⫽ 22),; differences not significant (2 test).
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
from neonates to adults (Fig. 4B). Discharge activity is primarily controlled by the activity of the M-channel in dissociated neonatal neurons (Cuevas et al. 1997). Differential densities of the M-current may also contribute to this behavior. The presence of TDR evoked by hyperpolarizing current pulses is held as the signature of Ih (Pape 1996). There was an increase in TDR with postnatal development, which is apparently inconsistent with the results of Hogg et al. (2001). Other voltage-sensitive currents operating in this voltage range could be contributing to the membrane potential trajectory. Clearly, voltage-clamp studies in the intact ganglion are required to resolve this apparent discrepancy. Spontaneous synaptic and action potential discharge was recorded from a population of neurons, as reported previously (Edwards et al. 1995; Xi-Moy et al. 1993). There are no reports of spontaneous EPSPs, action potential discharge, or pacemaker firing in cultured or dissociated ICG neurons (Adams and Cuevas 2004). The spontaneous activity recorded in this study was blocked by application of 100 M Cd2⫹, indicating such activity is synaptically evoked rather than originating from damage arising from the impalement.
A
DEVELOPMENTAL CHANGES IN INTRACARDIAC GANGLION NEURONS
3551
A
B
The second component of the study charted synaptic transmission from the neonate to the adult ganglion. Several types of postsynaptic responses were recorded from ICG neurons in response to a single presynaptic stimulus. These were divided into three classes: weak, secure, and strong based on the ability to fire an AP and the presence and timing of the secondary EPSP in relation to the action potential. A similar scheme was adopted for rat submandibular neurons (Smith et al. 2000). Proportions of these synaptic responses recorded from ICG neurons at different stages of postnatal development were not significantly different. Maximum action potential following frequency was higher for indirect, synaptically evoked events than for those generated by brief intrasomatic depolarizing current pulses, presumably because of increased conductance resulting from the underlying EPSP. The test frequency range (5–50 Hz) examined in the present study was used in previous investigations to stimulate the cardiac vagus in the rat (Jones et al. 1998) and was recorded during efferent reflex discharge in the cat (Kunze 1972), and thus can be regarded as being physiologically relevant. The ability of synaptically evoked action potentials to follow high-frequency repetitive stimulation shows modest improvement with postnatal development.
mammalian ICG. This has several implications, among which is how changes within the neural circuit during postnatal development affect heart function. It is becoming clear that local neural circuits may be important in setting up arrhythmic substrates. There are numerous clinical case reports of cardiac brady-arrhythmias arising from activity in the CNS and traversing the parasympathetic ganglia, such as during or after epileptic seizures (Rugg-Gunn et al. 2004). ICGs need to be considered as part of the pathway that sets up conditions for such arrhythmias. Therefore further analysis of the circuitry and synaptic transmission within intracardiac neurons and interactions with sympathetic and central nervous systems should promote further understanding regarding control of the heart during development and disease. For example, several studies have reported abnormalities in ganglionic transmission in the chronic rapid ventricular pacing model of heart failure (Bibevski and Dunlap 1999, 2004). ACKNOWLEDGMENTS
We thank D. J. Adams for constructive criticism of a draft of the manuscript and S. McMullan for the CED Spike Script for analysis of action potential characteristics. GRANTS
Functional significance This work has sought to increase knowledge regarding somatic membrane properties and synaptic transmission within J Neurophysiol • VOL
This work was supported by a project grant from the British Heart Foundation to A. A. Harper. K. Rimmer is a Biotechnology and Biological Sciences Research Council, United Kingdom Research Student.
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
FIG. 8. A: examples of postsynaptic events evoked by trains of nerve stimuli. A(i)–(iii) show trains of 20 stimuli at 5, 10, 20, and 50 Hz for a neonate, juvenile, and adult neuron. Note the different timescale bars for each applied frequency. B: depression of successful APs recorded during repetitive stimuli delivered to either the preganglionic nerve trunk (indirect stimulation, i) or short depolarizing current pulses injected into the soma (direct stimulation, ii). Ratio of the number of APs to the number of stimuli was used to give an index of the frequency dependency of synaptic transmission (i) and evoked somatic APs (ii). Number of successful somatic APs evoked decreased as the frequency of a train of 20 stimuli increased. Numbers of neurons are given in parentheses.
3552
K. RIMMER AND A. A. HARPER
REFERENCES
J Neurophysiol • VOL
95 • JUNE 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by guest on June 2, 2013
Adams DJ and Cuevas J. Electrophysiological properties of intrinsic cardiac neurons. In: Basic and Clinical Neurocardiology, edited by Armour JA and Ardell JL. New York: Oxford Univ. Press, 2004, p. 1– 60. Anderson RL, Jobling P, and Gibbins IL. Development of electrophysiological and morphological diversity in autonomic neurons. J Neurophysiol 86: 1237–1251, 2001. Ardell JL. Intrathoracic neuronal regulation of cardiac function. In: Basic and Clinical Neurocardiology, edited by Armour JA and Ardell JL. New York: Oxford Univ. Press, 2004, p. 118 –152. Bibevski S and Dunlap ME. Ganglionic mechanisms contribute to diminished vagal control in heart failure. Circulation 99: 2958 –2963, 1999. Bibevski S and Dunlap ME. Prevention of diminished parasympathetic control of the heart in experimental heart failure. Am J Physiol 287: H1780 –H1785, 2004. Bibevski S, Zhou Y, McIntosh JM, Zigmond RE, and Dunlap ME. Functional nicotinic acetylcholine receptors that mediate ganglionic transmission in cardiac parasympathetic neurons. J Neurosci 20: 5076 –5082, 2000. Cameron JS, Dryer L, and Dryer SE. Regulation of neuronal K⫹ currents by target-derived factors: opposing actions of two different isoforms of TGFbeta. Development 126: 4157– 4164, 1999. Cameron JS and Dryer SE. BK-Type K(Ca) channels in two parasympathetic cell types: differences in kinetic properties and developmental expression. J Neurophysiol 84: 2767–2776, 2000. Cameron JS, Lhuillier L, Subramony P, and Dryer SE. Developmental regulation of neuronal K⫹ channels by target-derived TGF beta in vivo and in vitro. Neuron 21: 1045–1053, 1998. Cuevas J, Harper AA, Trequattrini C, and Adams DJ. Passive and active membrane properties of isolated rat intracardiac neurons: regulation by Hand M-currents. J Neurophysiol 78: 1890 –1902, 1997. Dryer SE. Role of cell– cell interactions in the developmental regulation of Ca2⫹-activated K⫹ currents in vertebrate neurons. J Neurobiol 37: 23–36, 1998. Edwards FR, Hirst GD, Klemm MF, and Steele PA. Different types of ganglion cell in the cardiac plexus of guinea-pigs. J Physiol 486: 453– 471, 1995. Fischer H, Harper AA, Anderson CR, and Adams DJ. Developmental changes in expression of GABAA receptor-channels in rat intrinsic cardiac ganglion neurones. J Physiol 564: 465– 474, 2005. Gootman PM. Ontogeny of the cardiac nervous system. In: Basic and Clinical Neurocardiology, edited by Armour JA and Ardell JL. New York: Oxford Univ. Press, 2004, p. 252–271. Hogg RC and Adams DJ. An ATP-sensitive K⫹ conductance in dissociated neurones from adult rat intracardiac ganglia. J Physiol 534: 713–720, 2001. Hogg RC, Harper AA, and Adams DJ. Developmental changes in hyperpolarization-activated currents Ih and IK(IR) in isolated rat intracardiac neurons. J Neurophysiol 86: 312–320, 2001.
Inokuchi I and McLachlan EM. Lack of evidence for P2X-purinoceptor involvement in fast synaptic responses in intact sympathetic ganglia isolated from guinea-pigs. Neuroscience 69: 651– 659, 1995. Jessen KR and Mirsky R. The origin and development of glial cells in peripheral nerves. Nat Rev Neurosci 6: 671– 682, 2005. Jones JF, Wang Y, and Jordan D. Activity of C fibre cardiac vagal efferents in anaesthetized cats and rats. J Physiol 507: 869 – 880, 1998. Kunze DL. Reflex discharge patterns of cardiac vagal efferent fibres. J Physiol 222: 1–15, 1972. Lagi A, Cencetti S, Fattorini L, and Tamburini C. Age-related changes of cardiac parasympathetic modulation after vasovagal syncope. Am J Cardiol 83: 977–980, 1999. Pape HC. Queer current and pacemaker: the hyperpolarization-activated cation current in neurons. Annu Rev Physiol 58: 299 –327, 1996. Rimmer K and Harper AA. Developmental changes in the passive and active membrane properties of rat intracardiac neurons in situ: the effects of Ba2⫹ and Cs⫹ (Abstract). J Physiol 557P: PC62, 2004. Robinson RB. Autonomic receptor– effector coupling during post-natal development. Cardiovasc Res 31, Spec. No.: E68 –E76, 1996. Rugg-Gunn FJ, Simister RJ, Squirrell M, Holdright DR, and Duncan JS. Cardiac arrhythmias in focal epilepsy: a prospective long-term study. Lancet 364: 2212–2219, 2004. Sah P and Faber ES. Channels underlying neuronal calcium-activated potassium currents. Prog Neurobiol 66: 345–353, 2002. Sampaio KN, Mauad H, Spyer KM, and Ford TW. Differential chronotropic and dromotropic responses to focal stimulation of cardiac vagal ganglia in the rat. Exp Physiol 88: 315–327, 2003. Seabrook GR, Fieber LA, and Adams DJ. Neurotransmission in neonatal rat cardiac ganglion in situ Am J Physiol 259: H997–H1005, 1990. Selyanko AA. Membrane properties and firing characteristics of rat cardiac neurones in vitro. J Auton Nerv Syst 39: 181–189, 1992. Selyanko AA and Skok VI. Synaptic transmission in rat cardiac neurones. J Auton Nerv Syst 39: 191–199, 1992. Smith AB, Hansen MA, Liu DM, and Adams DJ. Pre- and postsynaptic actions of ATP on neurotransmission in rat submandibular ganglia. Neuroscience 107: 283–291, 2001. Smith AB, Motin L, Lavidis NA, and Adams DJ. Calcium channels controlling acetylcholine release from preganglionic nerve terminals in rat autonomic ganglia. Neuroscience 95: 1121–1127, 2000. Smith FM. Extrinsic inputs to intrinsic neurons in the porcine heart in vitro. Am J Physiol 45: R455–R467, 1999. Xi-Moy SX and Dun NJ. Potassium currents in adult rat intracardiac neurons. J Physiol 486: 15–31, 1995. Xi-Moy SX, Randall WC, and Wurster RD. Nicotinic and muscarinic synaptic transmission in canine intracardiac ganglion cells innervating the sinoatrial node. J Auton Nerv Syst 42: 201–213, 1993.
