| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Oregon Hearing Research Center/Vollum Institute, Oregon Health and Science University, Portland, Oregon; and 2Institute of Experimental Medicine Academy of Sciences of the Czech Republic, 142 20 Prague 4, Czech Republic
Submitted 5 August 2004; accepted in final form 18 September 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
), the rate of transmitter clearance from the cleft (Draguhn and Heinemann 1996), numbers of inputs (Kim and Kandler 2003), and the postsynaptic receptor subunit composition (Bosman et al. 2002; Hollrigel and Soltesz 1997; Tia et al. 1996). In immature neurons, high [Cl–]i causes GABA and glycine to be depolarizing, and this may result in an interesting interplay between inhibitory and excitatory processes (Ben-Ari 2002; Ben-Ari et al. 1989; Ehrlich et al. 1999; Ganguly et al. 2001; Obata et al. 1978). Moreover, young inhibitory pathways of brain stem auditory nuclei are predominantly GABAergic, switching with time to glycinergic transmission (Kotak et al. 1998; Nabekura et al. 2004). It is therefore of interest to determine how the complex refinement of inhibitory input occurs and how it relates to the simultaneous maturation of excitatory systems.
We sought to monitor the multiple processes underlying inhibitory development in a preparation in which there is parallel information available about the growth of excitatory synapses and associated ion channels. Numerous studies have explored the development of the calyx of Held, a giant excitatory terminal of the medial nucleus of the trapezoid body (MNTB), and showed that excitatory signaling mechanisms become faster and stronger with age (von Gersdorff and Borst 2002). By the onset of hearing, specialized Ca2+ channels (Iwasaki and Takahashi 1998), K+ channels (Dodson et al. 2002; Elezgarai et al. 2003; Ishikawa et al. 2003; Wang et al. 1998), and AMPA receptors (Joshi et al. 2004) are expressed. In addition, the vesicle pool size increases, and release probability decreases (Iwasaki and Takahashi 2001; Taschenberger et al. 2002). These properties allow the mature calyx to process high-frequency signals with precision and fidelity. In contrast, little is known about the maturation of transmitter systems in the MNTB that inhibit calyceal responses.
Recently, we reported the presence of a powerful glycinergic system capable of shunting the excitatory signals originating from the calyx of Held (Awatramani et al. 2004). In this study, we characterized the maturation of both glycinergic and GABAergic inputs to MNTB neurons and tested their functional properties. Our results complement previous anatomical studies (Adams and Mugnaini 1990; Campos et al. 2001; Piechotta et al. 2001; Roberts and Ribak 1987) and show that glycinergic inhibition becomes larger and faster with age and acquires the ability to follow high-frequency trains of stimuli. At intermediate ages, some axonal inputs are both GABAergic and glycinergic. Most striking was that the vigorous maturation of glycinergic inhibition, which overwhelms GABAergic transmission, is markedly delayed with respect to the development of excitation and the onset of hearing.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Coronal slices of brain stem were prepared from 5- to 27-day-old Wistar rats as previously described (Turecek and Trussell 2001). In brief, animals were decapitated, the brain stem was dissected and secured in a chamber with cyanoacrylate glue, and 200- to 300-µm-thick sections were cut with a vibratome (VT1000S, Leica, Deerfield, IL). Slices were immediately transferred to an incubation chamber containing a warm (37°C) extracellular solution composed of (in mM) 125 NaCl, 25 glucose, 2.5 KCl, 1 MgCl2, 2 CaCl2, 1.25 NaH2PO4, 25 NaHCO3, 0.4 ascorbic acid, 3 myo-inositol, and 2 sodium pyruvate, bubbled with 5% CO2-95% O2, for 1 h, after which the chamber was brought passively to room temperature. Recordings were typically obtained within 4 h of slicing.
Whole cell recordings
When ready for use, slices were transferred to a recording chamber and bathed with extracellular solution (21–22°C or 36–37°C) through a gravity-fed perfusion system, at about 3 ml/min (bath volume, 1.5 ml). Data in Figs. 9–11 were obtained at 36–37°C. MNTB neurons were viewed using a Zeiss Axioskop FS microscope equipped with differential interference contrast optics and a 63x (Achroplan, Zeiss) water-immersion objective. Whole cell voltage-clamp recordings were made from MNTB neurons with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA). Signals were filtered at 5–10 kHz and sampled at 20–100 kHz. For recording-evoked inhibitory postsynaptic currents (IPSCs), glass pipettes (1–2 M
) were filled with internal solution containing (in mM) 150 CsCl, 10 TEA-Cl, 5 EGTA, 1 MgCl2, 10 HEPES, 2 ATP, 0.3 GTP, 10 phosphocreatine, and 2 QX 314 (
320 mOsM), and pH adjusted to 7.3 with CsOH. The current traces were not corrected for a 2-mV junction potential. In some initial experiments, CsMeSO3 or CsF were used in place of CsCl (and currents were recorded at 0 mV). The small differences in decay kinetics of the IPSCs measured at 0 and –70 mV were insignificant compared with changes that occurred with age, and therefore data obtained with these different electrode solutions were grouped together. For current-clamp experiments, pipettes were filled with (in mM) 140 K-gluconate, 5 KCl, 1 MgCl2, 10 HEPES, 0.05 EGTA, 2 ATP, 0.3 GTP, and 10 phosphocreatine (pH 7.3). Voltage signals were corrected off-line for a 14-mV junction potential. The series resistance usually was <3–7 M a few minutes after whole cell mode was established, and usually increased to a stable value between 7 and 15 M within 10 min of recording. This was compensated on-line by 70–90% (lag, 10–20 µs).
A 50- to 100-µs, 5- to 50-V pulse generated through an isolated stimulus unit (AMPI Iso-flex) and delivered via an electrode filled with extracellular solution was used to stimulate inhibitory axons. The placement and the stimulus intensity were optimized to obtain the largest responses. IPSCs were recorded in the presence of 10 µM 6,7-dinitroquinoxaline-2, 3-dione (DNQX; Tocris) and 100 µM (±)-2-amino-5-phosphonopentanoic acid (AP5; RBI). To elicit miniature IPSCs in young animals (P5–P7), 50 mM K+ containing extracellular solution (adjusted to 340 mOsm) was applied for 1–10 s in the presence of DNQX, AP-5, and TTX until spontaneous events became apparent. Other drugs were added to the perfusate, as indicated: 0.3–0.5 µM strychnine hydrochloride (Sigma), 10 µM SR-95531 (Tocris), 10 µM zolpidem (Tocris), and 0.5 µM TTX (Alomone). Data were collected 3 min after wash-in of the drug. In some experiments, GABA and glycine were applied by pressure ejection (Picospritzer II; <1 psi; 5–10 µm diam of mouth of an application pipette; distance from cell: 50–60 µm).
For perforated-patch recordings, the tip of the patch pipette was first suction-filled with solution containing (in mM) 140 KCl, 3 MgCl2, 10 HEPES, and 5 EGTA (pH 7.3 with KOH) and backfilled with the same solution that additionally contained gramicidin (10–50 µg /ml). After obtaining a gigaohm seal, perforation was monitored using a 10-mV step depolarization, until the access resistance stabilized to between 10 and 30 M (within 15–45 min). GABA or glycine was briefly (100–200 ms) applied every 30–60 s, 500 ms after the voltage was stepped to a specified potential. Changing the order of the voltage steps did not affect the EGABA/glycine, suggesting that Cl– loading during the protocols was insignificant (Ehrlich et al. 1999).
Conductance-clamp experiments
Simulated excitatory and inhibitory conductances (EPSGs and IPSGs) were injected with an SM-1 amplifier (Cambridge Conductance). The current response to voltage from the patch amplifier has a 10–90% rise time of 290 ns. EPSG and IPSG waveforms were generated based on a 100-Hz train of EPSCs (Erev = 0) and IPSCs (EGABA = –50 mV) measured in P5–P7 rats. At these ages, the contribution of N-methyl-D-aspartate (NMDA) conductance is variable (Leao and von Gersdorff 2002). Here, for simplicity, it was omitted in the simulation of excitatory conductances.
Data analysis
Evoked IPSCs were analyzed in Clampfit 9.0 (Axon Instruments). Spontaneous miniature IPSCs (mIPSCs) were detected using a sliding template procedure (Axograph 4.0). The threshold for detection was set low, and noise that met trigger specifications was rejected on visual inspection. Aligned and baselined mIPSCs were averaged, and the decays were fit by single or double exponential function (based on the improvement of the summed square error): 
,where D(t) is the decay of the mIPSC as a function of time (t); A1 + A2 = A are constants; and 
fast and slow are fast and slow decay time constants, respectively. In some cases, adding the second exponent did not significantly decrease the SSE, and A2 was set at to 0. The weighted decay time constant was calculated as wd = (A1 x fast + A2 x slow)/(A1 + A2). To assess the relative GABA and glycine components of mixed IPSCs, individual events were baselined and fit by the function: D(t) = IGLYRDGLYR(t) + IGABARDGABAR(t), where DGLYR(t) and DGABAR(t) are averaged decays of GABAergic and glycinergic mIPSCs measured in the presence of strychnine and SR-95531, respectively. IGLYR and IGABAR were varied to obtain the best fit of the mIPSCs. Using this method, most of the events could be categorized as GABAergic, glycinergic, or mixed; some (20% in P8–P12 and 32% in P15–P22 rats) were too small to categorize. Results are expressed as mean ± SE (except as noted), and the significance was determined by linear regression or by implementing a Student's t-test (with significance indicated by P < 0.05).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
IPSCs were evoked at room temperature with a glass stimulating electrode filled with extracellular solution, placed in close proximity (30–100 µM) to the cell of interest. In young rats (P5–P7), IPSCs were quite small (peak amplitude, 143 ± 41 pA). However, IPSCs had significantly grown in amplitude (618 ± 94 pA) by the onset of hearing (P11–P12; Fig. 1A), a period by which the development of excitatory transmission is reported to have reached a near-mature state (Iwasaki and Takahashi 1998; Taschenberger and von Gersdorff 2000; Taschenberger et al. 2002). Surprisingly, the size of IPSCs continued to increase beyond P12, as shown in Fig. 1, A and B. Considering all rats >P20, the IPSC was 6.2 ± 1.1 nA (n = 21), and in P24–P25 rats alone, they were 12.3 ± 3.7 nA (n = 4; see Awatramani et al. 2004).
|
wd of 53.2 ± 9.0 ms (n = 13; range, 15.9–122.8 ms). By the onset of hearing, IPSCs were significantly faster (P11–P12; average wd was 13.1 ± 3.1 ms; n = 7) and less variable (range, 7.1–23.4 ms). This trend continued over the following 2 wk, as depicted in Fig. 1Bii. By P24–P25, the IPSCs were extremely fast and had mean wd = 3.9 ± 0.5 ms. Thus inhibitory inputs experience an apparently continuous development in size and shape over the first month after birth. As shown below, this transformation is a reflection of changes in the types of receptors and transmitters mediating the IPSC and in the temporal coordination of transmitter release. Age-dependent contributions of GABA and glycine receptors to the IPSCs
To determine whether the observed change acceleration of the evoked IPSCs was due to change in receptor types, we assessed the contribution of GABAA and glycine receptors to the IPSC, using the antagonists SR-95531 (10 µM) and strychnine (300–500 nM), respectively. Responses in young animals (P5–P7) were more sensitive to SR-95531 than to strychnine (Fig. 2). In this age group, SR-95531 blocked IPSCs by 87.3 ± 2.4% (n = 5; Fig. 2B) of the evoked IPSC, while strychnine decreased responses by only 23.1 ± 3.5% (n = 4). Considering the weak antagonist action of strychnine on GABAA receptors (Jonas et al. 1998), it is likely that the small and slow IPSCs observed in young rats were predominantly mediated by GABAA receptors. These data also indicate that the small effect of strychnine on IPSCs in the youngest age group does not reflect the presence of strychnine-resistant glycine receptors (Kungel and Friauf 1997).
|
wd = 24 ± 4 ms; n = 4; data not shown) could be evoked, indicating that weak GABAergic inputs persist in more mature MNTB. Thus within a span of 1 wk, there is a proliferation of glycinergic inputs, which outweighs the initial GABAergic ones.
Next, GABA and glycine receptor expression were examined by testing the sensitivity of MNTB neurons to the respective agonists. The conductance elicited by puffs of saturating concentrations of GABA (1 mM) was similar in young and old rats (Fig. 3; P > 0.05). Interestingly, GABA-evoked responses were >10 times larger than the synaptically evoked GABAergic responses described below. If we assume that maximal stimuli recruited a significant fraction of input fibers, these data would suggest that a large fraction of GABAA receptors may be nonsynaptic. In contrast, sensitivity to glycine (1 mM) increased by two orders of magnitude between ages P5–P7 and P13–P15 (Fig. 3; 0.001 vs. 0.20 µS; P < 0.001). This increase in glycine sensitivity is consistent with previous studies of Kungel and Friauf (1997) and corresponds well to the observed age-dependent increase in the contribution of glycine receptors to the IPSCs.
|
To determine how changes in the properties of transmitter responses at single synaptic sites contributed to the observed developmental changes in the evoked IPSCs, we recorded mIPSCs in the presence of TTX and antagonists of glutamate receptors (see METHODS). Under these conditions, all mIPSCs are either GABAergic or glycinergic, because no synaptic activity was seen following addition of both 300–500 nM strychnine and 10 µM SR-95531 (data not shown; n = 6 cells from rats P10–P18). We first examined the properties of GABAA receptor–mediated mIPSCs in the presence of 300–500 nM strychnine (Fig. 4) and found that they became briefer with age. To show the age-dependent acceleration of GABAergic mIPSCs, the average mIPSCs recorded in 32 cells are each plotted in Fig. 4A. Decreases in the fast and slow decay time constants (r = –0.42, P < 0.05 for fast; r = –0.39, P < 0.05 for slow) in combination with an increase in the contribution of the fast (r = 0.58, P < 0.0001) resulted in briefer mIPSCs (Fig. 4B). Aside from the kinetics, an age-dependent decrease in the average peak amplitude of GABA mIPSCs was also observed (Fig. 4B; r = –0.36, P < 0.05). Prior to onset of glycinergic transmission, mIPSCs averaged 81 ± 22 pA (P5–P7, n = 7 cells). In P9–P12, mIPSCs were 46 ± 10 pA (n = 7 cells), and P20–P25 mIPSCs were 40 ± 7 pA (n = 4 cells). Hence, GABAergic mIPSC became briefer but smaller in amplitude, consistent with pharmacological changes in the evoked IPSC.
|
fast of the mIPSCs decreased apparently exponentially as a function of age (Fig. 5B; r = –0.90, P < 0.0001), suggesting that the functional properties of these synapses begin to transform as soon as the synapses form. In addition, the relative contribution of this fast component also significantly increased (r = 0.46, P < 0.01). In the oldest group of animals (P26–P27), fast had reached values as fast as 1.5 ms (see Awatramani et al. 2004), but still had not yet stabilized at this age. Aside from the kinetics, an age-dependent increase in the average peak amplitude of glycine mIPSCs was also observed (Fig. 5; r = 0.57, P < 0.001). At P9–P12, mIPSCs averaged 124 ± 48 pA (n = 6 cells). In P20–P25 rats, the mIPSCs were significantly larger (275 ± 35 pA, n = 14 cells), and individual events sometimes exceeded 1 nA. Previous studies indicate that the large mIPSCs of the MNTB are not sensitive to agents that affected intracellular calcium and internal stores (Lim et al. 2003) and thus probably do not reflect multivesicular release. Hence, an increased number of receptors at single synaptic sites may contribute to the larger glycinergic mIPSCs, consistent with the increased expression of glycine receptors observed in MNTB neurons over development (Fig. 3). Other factors such as receptor occupancy could also contribute to larger mIPSCs. However, it should be noted that the observed increase in the mIPSC is not sufficient to explain the much larger increase in the evoked IPSCs (Fig. 1).
|
). To categorize the mIPSCs, the average decays of pharmacologically isolated GABAergic and glycinergic events were scaled to obtain the best fit for individual events (see METHODS). The amplitude of the glycinergic component was plotted against the GABAergic component (Fig. 6). In Fig. 6Aiii, open circles show the amplitude of fast and slow components in pharmacologically isolated glycinergic events, and filled symbols show data for GABAergic events. Dashed lines delineate ±2 SD for the mean amplitudes of each component and divide the data into four quadrants. In the absence of GABA/glycine antagonists, of 1,075 events (from 10 cells) analyzed, 35% fell below the horizontal line, indicating that they were GABAergic, and 30% fell to the left of the vertical line, indicating that they were glycinergic (Fig. 6Aii). However, about 15% were in the upper-right quadrant, indicating that they were biphasic. Hence, at P8–P11, GABA and glycine may be co-released at single synaptic sites.
|
), only a few dual-component mIPSCs were evident. Figure 6Bi shows normalized mIPSCs measured in a P21 rat in the presence of zolpidem. In this cell, the GABA and glycinergic events fell into two distinct groups with little overlap. Similarly, in a population of mIPSCs collected from eight cells (900 events), 26% were clearly categorized as GABAergic and 41% as glycinergic, but <1% of the events had both GABA and glycine components (Fig. 6Bii). Hence, co-transmission via GABAA and glycine receptors at single synaptic sites is a transient phenomenon in the developing MNTB. IPSCs evoked by minimal stimulation
As noted above, increases in the average mIPSCs size (Fig. 5B) could not fully account for the much larger developmental changes in the evoked IPSCs (Fig. 1). An increase in the number of inputs or in the strength of the each input could underlie the observed increase in the IPSC. Here we assess the output of single axons.
In P5–P7 rats, the amplitude of the IPSC (Fig. 1; peak amplitude 143 ± 41 pA) was similar to that of the mIPSCs (81 ± 22 pA, n = 7), suggesting that the strength of single axonal inputs was small. However, by P9–P12, there was a large increase in the amplitude of the evoked IPSC (511 ± 96 pA, n = 10). To examine the transmitter release from single axons, the intensity of the stimulus was adjusted such that the number of failures was >60%. Under these conditions, the probability of the IPSCs being generated by multiple axons was low. The average IPSC evoked by minimal stimulation in P9–P12 rats was 248 ± 34 pA (excluding failures; n = 12). In rats older than P15, the average amplitude of the peak IPSC (2034 ± 503 pA; n = 26) had increased, with individual inputs generating as much as 8 nA. Figure 7B plots the amplitude of single-axon responses as a function of age, and a linear fit had a slope r = 0.52 (P < 0.005). These data showed that the strength of single axons greatly increases over this time period, reflecting both increases in quantal size and probably in synapse number.
|
wd > 40 ms) resembling GABAergic IPSCs, whereas 46% had fast kinetics (wd < 11ms) and were probably glycinergic. However, 38% of the responses had distinct biphasic decay kinetics. Moreover, at these ages, inputs with different decay rates could innervate the same cell. This is shown in Fig. 7Ai, where the average inhibitory responses for three different stimulating positions are shown. Note that biphasic responses (Fig. 7Ai, black trace) did not arise from averaging GABAergic and glycinergic responses. Every IPSC elicited by stimulating that axon had biphasic decays similar to the depicted average IPSC. These dual kinetic IPSCs probably arose from axons that released transmitters which activated both GABAA and glycine receptors. This conclusion is reinforced by the finding that the slow component was increased in the presence of 10 µM zolpidem and blocked by SR95531 (Fig. 8, B and C), while the fast component was sensitive to strychnine (Fig. 8D). In rats >P15, 88% of the single-axon evoked IPSCs were fast, whereas 8% were mixed. Only in 4% of the cases were slow IPSC observed; these were small in amplitude and resembled IPSCs measured in strychnine. Hence, co-transmission of GABA and glycine is prominent only at intermediate ages and is detected only in a subgroup of inhibitory axons.
|
Besides the kinetics of the IPSCs, a critical parameter that determines the efficacy of inhibition is the Cl– concentration gradient across the cell membrane. We next examined whether the Cl– gradient changed as inhibition switched from a GABA to a glycinergic system. Gramicidin perforated-patch recordings were obtained from MNTB neurons before (P5–P7) and after (P13–P15) the onset of hearing. In the immature neurons, brief application of GABA was found to depolarize the membrane potential as shown in Fig. 9Ai. GABA responses were measured at several potentials, and an I-V plot was constructed to determine EGABA (Fig. 9, Aii and C). The average EGABA measured in P5–P7 rats was –50 ± 5 mV, significantly more depolarized than the resting potential (–67 ± 3 mV, P < 0.05; n = 4, Fig. 9D). In contrast, glycine hyperpolarized the membrane potential in the older group of animals (Fig. 9, B and C). The Eglycine was determined to be –80 ± 4 mV, which was significantly more negative compared with the resting potential measured at these ages (–70 ± 3 mV, P < 0.05; n = 6). Assuming that both transmitters activate Cl– channels of similar selectivity, these results indicate a negative shift in the ECl– and are consistent with recent findings in lateral superior olive (LSO) neurons (Balakrishnan et al. 2003; Ehrlich et al. 1999; Kakazu et al. 1999).
|
To assess the functional role of the GABAergic (in 500 nM strychnine) and glycinergic transmission, we measured the properties of evoked IPSCs at physiological temperature (36–37°C) and at high stimulus rates. In P5–P7 rats, the amplitude of GABAergic IPSCs was small (0.4 ± 0.2 nA, n = 4) and had slow decay kinetics (wd = 41 ± 14 ms). When the stimulating frequency was increased to 100 Hz, IPSCs summated, and a maximum peak current (2.4 ± 0.5 nA) was reached by the 10th stimulus (Fig. 10A, left inset). During the course of the train responses, release became asynchronous, and often failures to release after the stimulus were observed (in Fig. 10A, failure indicated by asterisk). Interestingly, these responses were followed by prolonged asynchronous release, which continued for several hundred milliseconds after the cessation of the last stimulus (Fig. 10A). To quantify the decay of responses after train stimuli, the peak amplitude after the last stimulus was normalized, and the area of the "tail current" was measured. After 20 stimuli at 100 Hz, asynchronous responses were found to decay in 396 ± 24 ms (Fig. 10, A and D). In older rats (P16–P20), GABAergic IPSCs (measured in strychnine) were similar in amplitude (0.2 ± 0.05 nA, n = 6, P = 0.36) but were significantly briefer (wd = 7 ± 2 ms, P < 0.01), consistent with the changes in quantal current described above. When stimulated at 100 Hz, responses were more phasic than in younger rats, as shown in Fig. 10A (right panel inset). After the train, responses decayed in 86 ± 17 ms, significantly faster those in young rats, but still much longer than the response to single shocks (Fig. 10D). When the frequency of the train stimulus was increased to 500 Hz, GABAergic axons failed to respond to every stimulus. Thus GABAergic responses remain small in amplitude throughout development and cannot follow high-frequency trains of stimuli. Their temporal response seems suited to mediate a tonic inhibition.
|
). However, even after facilitation and summation, GABAergic responses never approached the absolute current levels of the glycinergic inputs. Effectiveness of depolarizing GABAergic conductance in immature neurons
When EGABA is positive to the resting potential, as in the younger rats, GABA or glycine will depolarize the resting membrane potential. To test the effect of depolarizing GABA in the immature MNTB (P5–P7), we first injected simulated conductances into MNTB neurons. These conductances (IPSGs) were based on 100-Hz trains of IPSCs (maximum peak conductance, 40 nS) and EGABA was set to –50 mV, as determined in the previous experiments. Although "GABAergic" IPSGs significantly depolarized cells to an average of –61 ± 1 mV (Erest = –75 ± 3 mV; n = 5 cells), never were these responses sufficiently large to trigger spikes (Fig. 11A). In contrast, injection of EPSGs (with peak value of 50 nS and reversal potential of 0 mV) modeled on trains of EPSCs recorded separately reliably evoked action potentials as shown in Fig. 11B. Moreover, EPSGs triggered spikes reliably even when they were preceded by IPSGs (Fig. 11C). These results suggest that depolarizing GABAergic inputs are probably too weak to inhibit calyceal transmission.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
GABAergic transmission apparently begins early in the life of the MNTB, and it is possible that it first develops in parallel with excitation. However, glycinergic inputs appear after a delay, and this may coincide with morphological refinement of the calyx. During the first 2 wk after birth, the immature calyx loses postsynaptic coverage and develops its mature finger-like structure (Kandler and Friauf 1993). It may be that new inhibitory synapses are best able to form contacts after postsynaptic territory is made available. Thus the morphological maturation of the calyx itself could play a role in the development of glycinergic inputs.
However, consideration of the amplitudes of IPSCs and mIPSCs indicates that the inhibitory input is structurally rather modest, considering its ability to inhibit calyceal transmission (Awatramani et al. 2004). In P5–P7 rats, transmission is predominantly GABAergic, and the maximal evoked IPSC (143 pA) is only slightly larger than mIPSCs (81 pA), indicating a small number of weak inputs. At P9–P12, maximal IPSCs were 511 pA, while single-axon inputs averaged 248 pA. GABAergic and glycinergic mIPSCs were roughly equally present, and overall, averaged 85 pA. Thus there were about two axonal inputs with quantal contents of about three. By P20–P25, where glycine dominated transmission, peak IPSCs were 6.2 nA, minimal stimulation averaged 1.9 nA, and the quanta averaged 275 pA. These numbers suggest that, even at this age, there were still only about three glycinergic inputs, each with a quantal content of about seven. These estimates may err for several reasons. For example, it was probably not possible to activate all fibers to a given cell. If synapses had low release probabilities, the quantal content may greatly underestimate the number of release sites. Nevertheless, it seems apparent that glycinergic inputs are far fewer than the >600 glutamatergic sites in each calyx of Held (Satzler et al. 2002; Taschenberger et al. 2002). This mismatch of synapse number for excitation and inhibition is apparently compensated at the level of the quantum: for similar driving forces, mean calyceal mEPSCs average 33 pA (Sahara and Takahashi 2001), while glycinergic mIPSCs events are more than eight times larger.
Shift from GABAergic to glycinergic transmission in the MNTB
A shift from GABA to glycinergic transmission has been reported in spinal cord (Gao and Ziskind-Conhaim 1995; Gao et al. 1998; Keller et al. 2001) and certain auditory nuclei (Korada and Schwartz 1999; Kotak et al. 1998; Nabekura et al. 2004; Turecek and Trussell 2002). Recently, Nabekura et al. (2004) reported that, in the lateral superior olive (LSO), this shift might occur at the level of single presynaptic sites, i.e., individual boutons switch from releasing GABA to glycine. In that study, the proportion of dual-component mIPSCs dropped by 17% between P7 and P17, the relative amplitude of the GABA component of these events fell by 32%; because GABA immunostaining of vesicles also dropped, the decline in the relative contribution of GABA to inhibition was attributed to reduced GABA release. Our physiological observations of a GABA-glycine switch differ in several respects from the events in the LSO. First, a complete elimination of dual-component events was observed. This may be because we tracked dual transmission in rats as old as P21. Second, the absolute strength of glycinergic transmission and glycine sensitivity in MNTB increased markedly with time, while GABAergic function was relatively constant. Therefore comparisons between the proportions of the two types of IPSC will appear as a decline in GABAergic transmission. Third, we found that not all axonal inputs exhibited dual transmission. It remains unclear if axons producing dual component IPSCs represent GABAergic axons that transition to a glycinergic phenotype. It is possible alternatively that the youngest glycinergic inputs initially release both transmitters, but switch to a glycine-only state quickly enough that only a fraction of the dual-transmitting fibers can be detected. Finally, it may be that the switch is mediated postsynaptically. While GABA sensitivity did not change in MNTB (or in LSO; Nabekura et al. 2004), a redistribution of GABA receptors away from subsynaptic sites might still occur.
Glycine and GABA receptor subunit expression in the auditory brain stem
A switch from "fetal" 
-2 to adult-like -1 and -3 subunits during the first 2–3 postnatal wk is a hallmark of glycine receptor development (Akagi et al. 1991; Becker et al. 1988; Malosio et al. 1991; Singer et al. 1998; Watanabe and Akagi 1995). However, in the MNTB of newborn rats, the expression of the -2 receptor subunit is weak (Piechotta et al. 2001; Sato et al. 1995) and only ceases later in development (8–10 wk; Sato et al. 1995), whereas the -1 receptor subunit expression rapidly increases during the first 3 postnatal wk (Friauf et al. 1997; Piechotta et al. 2001), but does not peak until 8 wk after birth (Sato et al. 1995).
The delayed maturation of the expression of glycine receptor subunits matches the late development of glycinergic responses reported here. For example, in the youngest rats (<P8), the small conductance change in response to exogenous glycine, along with the absence glycinergic IPSC and mIPSCs, corresponds with the diffuse, weak expression of -1 and -2 receptor subunits (Piechotta et al. 2001; Sato et al. 1995). In the following weeks, as -1 receptor expression increases, so too does IPSC amplitude, frequency, and decay rate. Moreover, as -1 subunit expression continues to rise (Sato et al. 1995), we note that, at the oldest ages at we could record (P27), it seems that the amplitude and decay rates of glycinergic IPSCs have not yet stabilized developmentally. The close correspondence of the physiological maturation of glycinergic transmission with the expression of the -1 subunit of the glycine receptor suggests an involvement for this subunit in mediating glycinergic transmission. However, strong expression of -3 glycine receptor subunits in the adult MNTB (Sato et al. 1995) suggests that multiple -subunits may participate in synaptic transmission.
In contrast to glycine receptors, less is known about the age-dependence of the GABA receptor subunit expression. Surprisingly, in the adult MNTB, neurons express a "slow" GABAA receptor containing the 3 subunit (Campos et al. 2001). Consistent with these findings, decay kinetics of GABAergic mIPSCs were relatively slow (d 20 ms) compared with those synapses where the "fast" (d 10 ms) 1 subunits predominate (Bosman et al. 2002; Hollrigel and Soltesz 1997; Vicini et al. 2001). Hence, in the MNTB, glycine receptors are used in fast signaling pathways, whereas GABAergic systems may mediate tonic inhibition through slower GABAA receptors.
It is striking that the developmental changes we describe for IPSCs in MNTB are paralleled by a switch from GABAA to glycine receptors on the calyx of Held (Turecek and Trussell 2002). There, however, GABAA receptors disappear almost completely by P12, after the emergence of calyceal glycine receptors. It is tempting to speculate that presynaptic glycine receptor expression is coordinated with the development of glycine boutons, which provide for their activation (Turecek and Trussell 2001). However, it is less clear how presynaptic GABA receptors get activated, since postsynaptic GABAergic transmission is always weak in the MNTB; moreover, we have not been able to show spillover of GABA from boutons to calyceal receptors (R. Turecek, unpublished observations). It remains possible that calyceal GABA receptors respond to graded changes in ambient levels of GABA, a source of transmitter that would not have been detected in our experiments.
Development of synchronous release
Asynchronous release is thought to arise from factors including the accumulation of Ca2+ in the presynaptic terminal (Atluri and Regehr 1998; Goda and Stevens 1994; Rahamimoff and Yaari 1973), facilitation of release probability, and the balance between depletion and recovery rates of vesicles available for immediate release (Lu and Trussell 2000; Otsu et al. 2004). During development, Ca2+ dynamics in the calyx changes due to a variety of factors including acceleration of the presynaptic action potentials, changes in Ca2+ channels, extrusion rates, and buffer capacities (Chuhma et al. 2001; Iwasaki and Takahashi 1998; Lohmann and Friauf 1996). The marked decrease in IPSC decay rates after train stimulation is indicative of change in presynaptic Ca2+ dynamics. However, the improvement in synchrony seen with GABA IPSCs in older rats is still small compared with the extremely well-timed release of glycine following high-frequency presynaptic stimuli. Thus the two classes of terminal appear to employ distinct modes of transmission.
Depolarizing GABAergic transmission in young MNTBs
Besides the slower decay time and smaller amplitude of GABAergic IPSCs in week-old rats, we also found that the synaptic potentials were depolarizing, rather than hyperpolarizing as in older rats. Moreover, the kinetics of exocytosis was strikingly different, such that in the youngest rats, IPSCs were incapable of entraining to modest stimulus rates (100 Hz). Indeed, these responses were marked by a barrage of small IPSCs that continued long after the stimuli were terminated. We showed using conductance clamp that these features of the time course of trains of IPSCs and the driving force for the current combined to generate a plateau depolarizing during synaptic activation. In studies of the chick, it was observed that, even in relatively mature auditory brain stem, depolarizing, asynchronous release of GABA results in an effective depolarizing block of excitation (Lu and Trussell 2001; Monsivais et al. 2000). This is clearly not the case in young MNTB, since calyceal responses could not be shunted by the small GABA IPSC. Nor was the GABAergic depolarization by itself capable of eliciting action potentials postsynaptically. While these features indicate little electrical consequence of GABAergic transmission, it remains possible that they serve to activate voltage-gated Ca2+ channels; indeed a stable plateau depolarization would be optimal for this purpose. Such a mechanism could have a developmental signaling function, perhaps in triggering the synthesis and clustering of glycine receptors (Kirsch and Betz 1998).
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: L. O. Trussell, Oregon Hearing Research Center/Vollum Inst., Mail Code L-335A, 3181 SW Sam Jackson Park Rd., Portland, OR 97239 (E-mail: trussell{at}ohsu.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Akagi H, Hirai K, and Hishinuma F. Cloning of a glycine receptor subtype expressed in rat brain and spinal cord during a specific period of neuronal development. FEBS Lett 281: 160–166, 1991.[CrossRef][ISI][Medline]
Akagi H and Miledi R. Heterogeneity of glycine receptors and their messenger RNAs in rat brain and spinal cord. Science 242: 270–273, 1988.
Atluri PP and Regehr WG. Delayed release of neurotransmitter from cerebellar granule cells. J Neurosci 18: 8214–8227, 1998.
Awatramani GB, Turecek R, and Trussell LO. Inhibitory control at a synaptic relay. J Neurosci 24: 2643–2647, 2004.
Balakrishnan V, Becker M, Lohrke S, Nothwang HG, Guresir E, and Friauf E. Expression and function of chloride transporters during development of inhibitory neurotransmission in the auditory brainstem. J Neurosci 23: 4134–4145, 2003.
Becker CM, Hoch W, and Betz H. Glycine receptor heterogeneity in rat spinal cord during postnatal development. EMBO J 7: 3717–3726, 1988.[ISI][Medline]
Ben-Ari Y. Excitatory actions of gaba during development: the nature of the nurture. Nat Rev Neurosci 3: 728–739, 2002.[CrossRef][ISI][Medline]
Ben-Ari Y, Cherubini E, Corradetti R, and Gaiarsa JL. Giant synaptic potentials in immature rat CA3 hippocampal neurones. J Physiol 416: 303–325, 1989.
Bosman LW, Rosahl TW, and Brussaard AB. Neonatal development of the rat visual cortex: synaptic function of GABAA receptor alpha subunits. J Physiol 545: 169–181, 2002.
Campos ML, de Cabo C, Wisden W, Juiz JM, and Merlo D. Expression of GABA(A) receptor subunits in rat brainstem auditory pathways: cochlear nuclei, superior olivary complex and nucleus of the lateral lemniscus. Neuroscience 102: 625–638, 2001.[CrossRef][ISI][Medline]
Chuhma N, Koyano K, and Ohmori H. Synchronisation of neurotransmitter release during postnatal development in a calyceal presynaptic terminal of rat. J Physiol 530: 93–104, 2001.
Dodson PD, Barker MC, and Forsythe ID. Two heteromeric Kv1 potassium channels differentially regulate action potential firing. J Neurosci 22: 6953–6961, 2002.
Draguhn A and Heinemann U. Different mechanisms regulate IPSC kinetics in early postnatal and juvenile hippocampal granule cells. J Neurophysiol 76: 3983–3993, 1996.
Ehrlich I, Lohrke S, and Friauf E. Shift from depolarizing to hyperpolarizing glycine action in rat auditory neurones is due to age-dependent Cl- regulation. J Physiol 520: 121–137, 1999.
Elezgarai I, Diez J, Puente N, Azkue JJ, Benitez R, Bilbao A, Knopfel T, Donate-Oliver F, and Grandes P. Subcellular localization of the voltage-dependent potassium channel Kv3.1b in postnatal and adult rat medial nucleus of the trapezoid body. Neuroscience 118: 889–898, 2003.[CrossRef][ISI][Medline]
Friauf E, Hammerschmidt B, and Kirsch J. Development of adult-type inhibitory glycine receptors in the central auditory system of rats. J Comp Neurol 385: 117–134, 1997.[CrossRef][ISI][Medline]
Ganguly K, Schinder AF, Wong ST, and Poo M. GABA itself promotes the developmental switch of neuronal GABAergic responses from excitation to inhibition. Cell 105: 521–532, 2001.[CrossRef][ISI][Medline]
Gao BX, Cheng G, and Ziskind-Conhaim L. Development of spontaneous synaptic transmission in the rat spinal cord. J Neurophysiol 79: 2277–2287, 1998.
Gao BX and Ziskind-Conhaim L. Development of glycine- and GABA-gated currents in rat spinal motoneurons. J Neurophysiol 74: 113–121, 1995.
Goda Y and Stevens CF. Two components of transmitter release at a central synapse. Proc Natl Acad Sci USA 91: 12942–12946, 1994.
Hollrigel GS and Soltesz I. Slow kinetics of miniature IPSCs during early postnatal development in granule cells of the dentate gyrus. J Neurosci 17: 5119–5128, 1997.
Ishikawa T, Nakamura Y, Saitoh N, Li WB, Iwasaki S, and Takahashi T. Distinct roles of Kv1 and Kv3 potassium channels at the calyx of Held presynaptic terminal. J Neurosci 23: 10445–10453, 2003.
Iwasaki S and Takahashi T. Developmental changes in calcium channel types mediating synaptic transmission in rat auditory brainstem. J Physiol 509: 419–423, 1998.
Iwasaki S and Takahashi T. Developmental regulation of transmitter release at the calyx of Held in rat auditory brainstem. J Physiol 534: 861–871, 2001.
Jonas P, Bischofberger J, and Sandkuhler J. Corelease of two fast neurotransmitters at a central synapse. Science 281: 419–424, 1998.
Joshi I, Shoknalla S, Titus P, and Wang LY. The role of AMPA receptor gating in the development of high-fidelity neurotransmission at the calyx of Held synapse. J Neurosci 24: 183–196, 2004.
Juttner R, Meier J, and Grantyn R. Slow IPSC kinetics, low levels of alpha1 subunit expression and paired-pulse depression are distinct properties of neonatal inhibitory GABAergic synaptic connections in the mouse superior colliculus. Eur J Neurosci 13: 2088–2098, 2001.[CrossRef][ISI]
) and strychnine (
) is significantly more depolarized than resting membrane potential (
) is more negative than resting membrane potential in P13–P15 rats (P < 0.05; n = 6). Temperature was held at 36–37°C.
