Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1998 Nov 1;512(Pt 3):809–817. doi: 10.1111/j.1469-7793.1998.809bd.x

Persistent depolarizing action of GABA in rat Cajal-Retzius cells

Jean-Marc Mienville 1
PMCID: PMC2231241  PMID: 9769423

Abstract

  1. To characterize membrane properties that might be relevant to the function and fate of Cajal-Retzius (CR) cells, the pharmacological and physiological effects of GABA acting at GABAA receptors were studied in CR cells from embryonic (E18) and postnatal (P11–13) slices of rat neocortex.

  2. From the embryonic to the postnatal stage, GABA-induced maximum current almost tripled, the EC50 increased from 38 to 74 μm, and the Hill number increased from 1.4 to 1.9. Muscimol-elicited currents were qualitatively and quantitatively similar to those produced by GABA.

  3. GABA-induced changes in the amplitude of large-conductance Ca2+-activated K+ channel current recorded on-cell from E18 CR cells were consistent with depolarization.

  4. GABA-mediated depolarization of embryonic and postnatal CR cells was studied directly with perforated-patch recording techniques. Ten micromolar and 1 mm GABA, respectively, depolarized E18 CR cells to −27 ± 1 and −25 ± 3 mV. These same concentrations of GABA depolarized P11 CR cells to −36 ± 3 and −23 ± 3 mV.

  5. In postnatal cortex, GABA (100 μm) increased the firing rate of CR cells 7.3-fold. By contrast, the firing of hippocampal pyramidal cells from slices of the same age (P12) was totally and reversibly blocked by GABA.

  6. These experiments suggest that contrary to its postnatal inhibitory shift observed in other cells, the depolarizing effect of GABA remains in CR cells from E18 until their virtual disappearance.

First described over a century ago (Ramón y Cajal, 1891), Cajal-Retzius (CR) cells have recently reawakened a sharp interest with the discovery that they synthesize and secrete a protein, Reelin, which controls neuronal migration and final position during neocortical lamination (D'Arcangelo et al. 1995; Ogawa et al. 1995; Rakic & Caviness, 1995). Notwithstanding this pivotal role, little information is available on the basic physiological and pharmacological properties of CR cells (Kim et al. 1995; Hestrin & Armstrong, 1996; Zhou & Hablitz, 1996; Mienville & Barker, 1997; Schwartz et al. 1998). Specifically, nothing is known about potential stimuli or factors that may regulate Reelin secretion. This particular information would be very useful to better understand regulatory processes operative in neocorticogenesis, and the potential role that abnormalities of neuronal migration may play in the aetiology of neuropsychiatric disorders (Kalus et al. 1997). Another still unexplained feature of CR cells is their disappearance from cortical Layer I, which in the rat occurs after the second week postnatal. To explain this disappearance, it is still unclear whether CR cells undergo morphological changes, become diluted in a rapidly expanding cortex, or simply die (see Marín-Padilla, 1998, for review). Recently, the hypothesis of cellular death has received strong experimental support (Derer & Derer, 1992; Del Rio et al. 1996).

It is well established, in many neuronal populations, that during early stages of development GABA, acting on GABAA receptors, depolarizes the plasma membrane due to a Cl− equilibrium potential that is more positive than the resting potential. This phenomenon has been particularly well described for rat hippocampal cells in which, after the first week postnatal, the effect of GABA switches from depolarizing to hyperpolarizing, apparently owing to a more efficient Cl− extrusion (Ben-Ari et al. 1997). This switch, together with the differentiation of glutamatergic transmission, eventually allows for functional maturation of hippocampal circuits.

The present results provide the first electrophysiological characterization of GABA responses in embryonic and postnatal CR cells. Moreover, they indicate that in late postnatal cortex, CR cells fail to promote the switch to a hyperpolarizing GABA effect, which may entail important consequences for the physiology and fate of these cells.

METHODS

Tissue preparation

In situ patch-clamp methods (Edwards et al. 1989) were used as previously described (Mienville & Barker, 1997). Pregnant Fisher rats were killed by inhalation of a rising concentration of CO2, and embryonic day (E) 18 fetuses or postnatal day (P) 11–13 pups were rapidly decapitated, in compliance with guidelines set out by the National Institutes of Health. After removal of the cerebrum in cold (4°C) phosphate-buffered saline (Dulbecco's PBS with Ca2+ and Mg2+; Quality Biological, Gaithersburg, MD, USA), 200–300 μm-thick coronal slices of intermediate cortex, some of which contained the hippocampus, were cut on a TPI 1000 vibratome (St Louis, MO, USA), and placed in the same medium as that used for recording (both at room temperature, 23–25°C). This medium consisted of (mm): 120 NaCl, 5 KCl, 1.25 NaH2PO4, 2 MgCl2, 2 CaCl2, 26 NaHCO3, 1 sodium pyruvate and 10 dextrose; supplemented with essential amino acids and minimum essential medium vitamins (Life Technologies), and bubbled with 95 % O2-5 % CO2 to yield a pH of 7.4. CR cells were identified chiefly on anatomical bases, according to: (1) their location on the external border of Layer I; (2) fusiform or ovoid, mostly bipolar morphology; and (3) somatic and neuritic horizontal orientation (Ramón y Cajal, 1891; Zhou & Hablitz, 1996; Mienville & Barker, 1997). It has been shown, in this laboratory, that cells displaying these features are Reelin positive (Pesold et al. 1998). CR cells could be easily differentiated from another type of cell present in Layer I, which had a round soma and no visible processes under Hoffman modulation contrast optics. Surprisingly, these cells had a noticeably lower input resistance than CR cells, which suggests that they may correspond to the neurogliaform type. Indeed, neurogliaform cells possess an extensive axonal arbor that is only visible after dye loading (Hestrin & Armstrong, 1996; Schwartz et al. 1998). A further criterion of identification was the fact that presumed neurogliaform cells had a resting potential of ≤ −60 mV, whereas CR cells consistently had a relatively low resting potential (∼-40 mV) at all ages studied. This finding, however, is at variance with both the study of Zhou & Hablitz (1996), who found a low resting potential for all Layer I cells, and that of Hestrin & Armstrong (1996) which indicated a resting potential of −64 mV for CR cells. Nevertheless, the virtual disappearance of ‘CR-like’ cells at P13 is a further clue to their identity. In fact, the major difficulty of finding such cells at that stage was an incentive to include P11–12 experiments. Incidently, the fact that many more CR cells can be found at these earlier stages, especially at P11, suggests a fairly abrupt disappearance. Unless specified, ‘postnatal’ in the present paper thus refers to P11–13.

Drug application and patch-clamp recording

The drug delivery system consisted of an eight-line bundle connected to three-way solenoid valves (Neptune Research, W. Caldwell, NJ, USA) fed by a peristaltic pump, and converging onto a 100 μm-diameter (i.d.) micromanifold (ALA Scientific Instruments, Westbury, NY, USA). The tip of the manifold was placed ∼200 μm from the cell and solution was ejected at ∼170 mm s−1. The ‘on’ and ‘off’ rates of open-tip currents measured by switching from a control solution to the same solution diluted 50 % with water had a time constant of 37.5 ± 4.4 ms (n = 10 trials). These considerations may seem relevant to the relatively slow activation kinetics displayed by GABA and muscimol-induced currents (Figs 1 and 2), as transient mixing of the test solution with control solution may occur at the tip of the manifold (dead volume < 1 μl). However, such slow kinetics are probably due, for the most part, to the diffusional barrier intrinsic to the slice preparation (Haas et al. 1993). Since recording was performed at various depths irrespective of the age of the preparation, this should not invalidate comparisons of dose-response curve parameters between embryonic and postnatal cells. In absolute terms, the maximum response might be underestimated because peak currents may reflect a mixture of receptor states, some being activated and others desensitized. If present, such an error must be minimal, however, because inactivation was much slower than activation. For example, fitting the exponential portions of the current evoked by 3 mm GABA in Fig. 1C yielded time constants of 0.2 and 1.3 s for activation and inactivation, respectively.

Figure 1. Pharmacological characterization of GABAA receptor-mediated currents in embryonic and postnatal CR cells.

Figure 1

Open in a new tab
A, bicuculline (0.1 mm) completely and reversibly inhibits outward currents evoked by 1 s applications of 100 μm GABA (indicated by bars below the traces). B and C, selected traces obtained with various concentrations of GABA applied for 3 s as indicated by the bars above the traces. D, concentration-current curves for E18 (○) and P11 (□) CR cells. Data were fitted with an equation of the form:
graphic file with name tjp0512-0809-mu1.jpg
where I is current amplitude and nH is the Hill number. See text for experimental conditions and parameter values.

Figure 2. Muscimol-induced currents in embryonic and postnatal CR cells.

Figure 2

Open in a new tab

Top traces show representative currents obtained at E18 and P11 with three concentrations of muscimol applied for 3 s, as indicated by the bars above the traces (CsCl pipette solution; −60 mV holding potential). The lower diagram summarizes data obtained in 8 E18 (□) and 17 P11 (×) CR cells.

Pipette solutions contained KCl (147 mm), CsCl (147 mm), or KCl + potassium gluconate (10 + 137 mm) as main salts. These solutions were buffered with Hepes (10 mm) and an appropriate amount of KOH or CsOH to yield a pH of 7.2. Both KCl and CsCl solutions were used for GABA dose-response studies at E18, yielding similar results. Given the lack of effect of baclofen (3/3 cells at both E18 and P13) and the complete abolition of the current by bicuculline (Fig. 1A), which argues against the presence of GABAB receptors in CR cells, these results were pooled. Perforated-patch experiments were carried out with the cationic pore-forming antibiotics amphotericin B and gramicidin as previously described (Kyrosis & Reichling, 1995). Experiments with amphotericin B were performed within 10 min as required to avoid drifts in chloride reversal potential (Kyrosis & Reichling, 1995). For all perforated-patch and cell-attached experiments, as well as some whole-cell experiments, potassium gluconate pipette solution was used, in which case a 10 mV junction potential was taken into account.

Currents were voltage clamped and amplified with a List EPC7 unit. Whole-cell currents were sampled on-line via a Digidata 1200B interface, and analysed with the pCLAMP 6 suite of programs (Axon Instruments). Single-channel currents and current-clamp recordings were stored on videotape with an Instrutech VR100B unit (Great Neck, NY, USA). Single-channel analysis was done with the PAT program of the SES package (courtesy of John Dempster, Strathclyde University, UK). Single-channel current amplitude was measured with readout cursors and occasionally checked with double or triple Gaussian fitting of amplitude histograms. Open probability (Po) was estimated both with transition detection methods and from amplitude histograms. Also part of SES, Windows WCP was used to measure cell capacitance as previously described (Mienville & Barker, 1997). Data are presented as means ± s.e.m. and were compared with Student's t test.

RESULTS

Voltage-clamp study of GABAA receptors in CR cells

In E18 and P13 CR cells, GABA-evoked outward currents recorded with potassium gluconate-containing pipettes at a holding potential of −20 mV were completely abolished by bicuculline (0.1 mm), indicating the exclusive involvement of GABAA receptors (Fig. 1A). Similar results were obtained in four of four E18 and seven of seven P11–13 CR cells. Currents evoked by various concentrations of GABA in E18 and P11 CR cells are shown in Fig. 1B and C, respectively. Under these conditions of symmetrical [Cl−], the current reversed near 0 mV (not shown) and thus was inward at −60 mV. The inactivation rate appeared to increase with GABA concentration. GABA dose-response curves were obtained by fitting a Michaelis-Menten equation to the peak current values measured for each concentration of agonist (Fig. 1D). The maximum current recorded in postnatal cells was nearly 3 times larger than that measured in embryonic cells. Since there was no significant difference in capacitance between embryonic and postnatal CR cells (14.1 ± 0.7 pF (n = 31) vs. 15 ± 0.9 pF (n = 26), respectively; P = 0.4), this suggests an increase in the density of GABAA receptors. On the other hand, the EC50 (38 ± 6 μm at E18 (n = 21) vs. 74 ± 8 μm at P11 (n = 12); P = 0.001) and Hill number (1.4 ± 0.1 at E18 vs. 1.9 ± 0.1 at P11; P = 0.006) were significantly different.

One concern regarding these differences was the relatively slow activation kinetics of GABA-induced currents (see Methods). The slow onset of the responses, particularly at low agonist concentrations, might suggest that the actual concentration reaching the cells was affected by secondary mechanisms such as GABA uptake. In that case, postnatal development of the uptake mechanism might account for the differences in EC50 and maximum response. To address this question, voltage-clamp experiments were performed with muscimol, a potent GABAA receptor agonist which is a poor substrate for the GABA uptake system. The top traces of Fig. 2 clearly show that muscimol-induced currents displayed kinetics similar to those observed when GABA was used. This result suggests that the responses of CR cells to GABA are not affected by uptake; rather, their kinetics probably depend on the diffusional barrier imposed by tissue surrounding the cells. Moreover, maximum responses obtained with saturating concentrations of muscimol (cf. the similar current amplitudes at 0.1 and 0.3 mm in Fig. 2) were comparable to those obtained with GABA (Fig. 1D). Muscimol-induced maximum current was ∼2.4 times larger in postnatal than in embryonic CR cells.

BK channels as sensors of GABA-induced membrane depolarization

In order to investigate the physiological action of GABA, it is necesssary to preserve intracellular Cl− (Cli−), a condition not satisfied by whole-cell recording. Cell-attached recording of large-conductance Ca2+-activated K+ (BK) channels used as sensors of membrane potential (Vm) may provide a first assessment of relative changes in Vm. E18 CR cells, it turns out, consistently display spontaneous BK channel activity, which was enhanced when the intracellular [Ca2+] was increased with ionomycin (Fig. 3A). Under control conditions, Po of BK channels was 2.4 × 10−3 ± 1.3 × 10−3; it was raised to 0.25 ± 0.09 (n = 6) by addition of 10 μm ionomycin, an ∼100-fold increase. Recovery from ionomycin was either complete (4 patches) or absent (1 patch); Fig. 3A represents an intermediate case. Channel conductance was estimated by measuring current amplitude at various potentials, which yielded a linear slope of 161 pS, and a reversal potential, −40 mV, equal to the CR cell resting potential (Fig. 3B).

Figure 3. Characterization of BK channel currents in embryonic CR cells.

Figure 3

Open in a new tab

A, cell-attached recording showing single-channel activation by local application of 10 μm ionomycin onto an E18 CR cell; the bar indicates the period of application. In this case, open probability increased from 0.002 under control conditions to 0.11 after ionomycin application. B, single-channel current amplitude (i) vs. pipette potential (Vp). Data from 12 cell-attached patches recorded with potassium gluconate pipette solution are fitted with a linear regression curve.

Advantage was taken of the frequent occurrence of BK channels in E18 CR cells to investigate Vm changes in response to GABA. Thus, membrane depolarization should decrease the amplitude of inward BK channel currents, whereas the opposite (increased amplitude) should occur with outward currents. As this is exactly what happened during GABA application (Fig. 4), the simplest explanation is that GABA depolarizes ‘intact’ CR cells in which [Cl−]i has not been altered by whole-cell dialysis (note that GABA applied to the cell cannot reach and affect BK channels isolated by the cell-attached pipette). Since BK channel i-V relationships can be considered as linear (Fig. 3B), results obtained at different pipette potentials were pooled, indicating a 3.4 ± 0.6 pA change in single-channel current in the presence of 100 μm GABA (n = 6). With a 161 pS single-channel conductance, it can be predicted that 100 μm GABA depolarizes E18 CR cells by ∼21 mV. Considering the magnitude of current generated by 100 μm GABA (Fig. 1B and D), and the range of input resistance (0.5–5 GΩ) measured in CR cells (Zhou & Hablitz, 1996), it is safe to consider this value as maximal depolarization (see below).

Figure 4. Modulation of BK channel amplitude by GABA (100 μm) via changes in membrane potential of E18 CR cells.

Figure 4

Open in a new tab

Upper panels show inward (left) and outward (right) currents obtained at positive or negative pipette potentials (Vp; note the different amplitude calibrations). The positions of the Control and GABA labels indicate the closed state. Lower panels show normalized amplitude histograms obtained from the corresponding patch current in control and in 100 μm GABA (current was sampled over a longer period than the records displayed). Data were fitted with Gaussian curves shown as thin (control) or thick (GABA) dotted lines. The numbers on the right-hand graph indicate peak amplitudes; circled numbers correspond to GABA exposure. Note that activity is greater at depolarized potentials (negative Vp; cf. the two-level activity), consistent with the voltage sensitivity of BK channels. Further, in addition to its effect on amplitude, GABA increases open probability, visible especially in the right panels, in a manner consistent with its depolarizing effect.

Current-clamp study of GABA-induced depolarization of embryonic and postnatal CR cells

Since in postnatal CR cells BK channel activity was much less frequent, and in order to obtain an absolute value for GABA-mediated depolarization, the perforated-patch method was chosen to directly monitor Vm without altering [Cl−]i (Kyrosis & Reichling, 1995). Assuming a minimum input resistance of 1 GΩ for E18 CR cells, it was predicted that 10 μm GABA (see Fig. 1B and D) would be sufficient to reach maximal depolarization. Figure 5Aa, Ab and Ba shows that this prediction was verified. Ten micromolar GABA depolarized E18 CR cells to −27 ± 1 mV, which is not different (P = 0.2, paired t test) from the −25 ± 3 mV potential reached with 1 mm GABA, a concentration that saturates receptors (Fig. 1D). The only difference was the rate at which Vm returned to resting potential (Fig. 5Aa and Ab), being reasonably fast with 10 μm GABA but taking 38 ± 3 s with 1 mm GABA. Certainly, voltage responses obtained with 1 mm GABA did not inactivate in the same manner as current responses (Fig. 1). This is due to the fact that even a small amount of current remaining during inactivation is sufficient to keep cells depolarized. The fact that neither current (Figs 1 and 2) nor voltage responses elicited with high agonist concentrations deactivated within a reasonable time despite the switch to control solution is probably due to agonist trapping within the slice. In postnatal CR cells, there was a noticeable difference between the effects of 10 μm and 1 mm GABA (Fig. 5Ac, Ad and Bb). The 10 μm agonist concentration did not depolarize cells to the same extent as 1 mm (P = 0.003). This may be due to a lower affinity of postnatal cells, as suggested by their higher EC50 (see above), a lower input resistance (Zhou & Hablitz, 1996), or both. Recovery from 1 mm GABA-induced depolarization was also somewhat faster than in E18 cells (Fig. 5Ad), taking 27 ± 4 s (P = 0.06). After breaking the patch to reach the whole-cell recording configuration, subsequent application of GABA always resulted in hyperpolarization of both embryonic and postnatal cells (not shown), consistent with the low chloride content of potassium gluconate pipette solution (see Methods).

Figure 5. GABA-mediated depolarization of embryonic and postnatal CR cells.

Figure 5

Open in a new tab

A, perforated-patch current-clamp recordings showing the effects of GABA on membrane potential (Vm) at E18 and P11. GABA was applied for 3 s (indicated by bars below the traces) at the concentrations indicated. a and b, amphotericin B perforation; c and d, gramicidin perforation. B, summary of the magnitude of depolarization elicited by two concentrations of GABA in E18 (a; n = 3) and P11 (b; n = 4) CR cells. The bottom of each box represents resting potential, the top corresponds to the potential reached by the cell in the presence of GABA.

Physiological consequences of GABAergic depolarization

Whereas spontaneous spiking activity was absent in E18 CR cells, it was rather common in postnatal cells. Inasmuch as firing can be recorded on-cell, such an occurrence again was opportune because it provided a more physiological test of the action of GABA while also leaving [Cl−]i unaltered. Consistent with its depolarizing action, GABA increased the firing rate of postnatal CR cells 7.3 ± 3.8-fold (n = 5). Figure 6A and C shows an example of such an activation induced in a P12 CR cell. Depending on the application time of GABA, increased firing was sometimes followed by accommodation and silencing of the cell (not shown). That this effect was through depolarization is supported by the concomitant decrease in spike amplitude that probably corresponds to a lesser driving force for sodium ions (Fig. 6A). In order to verify whether the depolarizing effect of GABA was cell specific, and to rule out any artifact (through mechanical stimulation of the cell by the drug application, for example), on-cell firing was recorded in P12 hippocampal pyramidal cells. In these cells and at this postnatal stage, activation of GABAA receptors has been shown to result in membrane hyperpolarization (Ben-Ari et al. 1997). Indeed, application of GABA onto P12 hippocampal cells brought firing to a halt (Fig. 6B and D) in five of five cells. Moreover, this effect appeared to be due to hyperpolarization as indicated by the increased amplitude of some of the remaining spikes (arrowhead in Fig. 6B), which is consistent with an increased driving force for sodium ions.

Figure 6. Opposite effects of GABA on firing of postnatal CR and hippocampal cells.

Figure 6

Open in a new tab

Spontaneous firing was recorded in the cell-attached configuration and was increased by a 1 s application of 100 μm GABA (double arrows) in a P12 CR cell (A and C), or completely and reversibly inhibited in a P12 hippocampal cell (B and D). The frequency diagrams shown in C and D correspond, respectively, to the records displayed in A and B.

discussion

The present results demonstrate the early expression and large increase in density of GABAA receptors in, respectively, embryonic and late postnatal CR cells. The differences in EC50 and Hill number between embryonic and postnatal CR cells are suggestive of a possible maturation of these receptors, perhaps due to selective subunit assembly (Sigel et al. 1990). The key finding of this study is that the physiological effect of GABA on CR cells appears to be depolarizing throughout the life of these transient cells (assuming that this is also the case prior to E18). A different situation appears to prevail in other systems, for example in hippocampal cells (Ben-Ari et al. 1997). Early on in those cells, GABA, through GABAA receptors, depolarizes the plasma membrane, and may thereby provide the condition necessary for activation of NMDA receptors. One week after birth, this activation would be provided by the emergence of AMPA receptors, while GABAA receptors take on their adult inhibitory function. It is unclear, at this point, what could be the nature of the inhibitory input, if any, to CR cells. As mentioned earlier, baclofen appears devoid of effect in these cells (including during perforated-patch recording), which would tend to rule out a GABAB receptor-mediated inhibition. Moreover, a consistent observation during voltage-clamp recording of CR cells from E18 to P13 is their lack of spontaneous postsynaptic currents (PSCs). This is in striking contrast with the robust excitatory and inhibitory (including GABAB receptor-mediated) PSCs displayed by their neighbouring Layer II pyramidal neurons in postnatal slices (author's unpublished observations). Kim et al. (1995) were able to evoke synaptic responses in CR cells apparently only when the cortical plate was stimulated. On the other hand, postnatal but not embryonic CR cells do display spontaneous firing (Fig. 6A). Since both have low resting potentials (Fig. 5) and neither display spontaneous PSCs, the most likely cause of this pacemaker type of firing seen in older cells is maturation of their intrinsic membrane properties (Zhou & Hablitz, 1996).

As it is often postulated that CR cells are GABAergic (Marín-Padilla, 1998; Pesold et al. 1998), the physiological significance of the depolarizing effects of GABA may instead have to be conceived in terms of a paracrine or autocrine action of the transmitter. Since low GABA concentrations may fully depolarize embryonic CR cells (Fig. 5Aa and Ba), steady activation of GABAA receptors from very early stages, coupled with inefficient Cli− extrusion (Ben-Ari et al. 1997), may result in redistribution of chloride ions across the membrane, which would explain the persistence of the depolarizing action of GABA. Indeed, the ∼-25 mV potential recorded upon exposure to a saturating concentration of GABA (Fig. 5) suggests a [Cl−]i of ∼50 mm. If one considers depolarization (and subsequent Ca2+ entry) as a widespread stimulus for secretion (Penner & Neher, 1988), GABA may constitute an autocrine signal for Reelin secretion. It may be argued that the effect of GABA is automatically counteracted by activation of BK channels (Fig. 4). This is unlikely, however, because at or near resting potential BK channel activity is low (Fig. 3A), probably owing to a low [Ca2+]i, and GABA application only modestly increases Po (Fig. 4, left panels). In other words, at low [Ca2+]i, GABA-mediated depolarization from ∼-40 to ∼-25 mV probably corresponds to the foot of the BK channel activation curve. (Conversely, at a pipette potential of −110 mV (Fig. 4, right panels), a 15 mV Vm depolarization corresponds to a change in patch potential from +70 to +85 mV, which may lie within the steep portion of the curve.) Moreover, the experiments depicted in Fig. 5 clearly demonstrate unimpeded depolarization by GABA.

Not incompatible with this scenario is another potentially important role for GABA-mediated depolarization. Since cortical CR cells disappear around the second postnatal week in the rat, failure to efficiently extrude Cli−, coupled with an increased GABAA receptor density, might be responsible for their death. The latter may occur through excessive depolarization by GABA, and subsequent overactivation of NMDA receptors - which is known to be lethal during certain critical periods (McDonald & Johnston, 1990) - by ambient glutamate (Sah et al. 1989). This idea is not inconsistent with the robust [Ca2+]i and electrophysiological responses elicited in CR cells by NMDA receptor activation (Schwartz et al. 1998; Mienville, 1998).

In summary, the present data provide evidence for an unusual postnatal persistence of the depolarizing action of GABA in a cell population that plays an important role in neocorticogenesis. Inasmuch as this action may be predicted to affect, and perhaps regulate, the function and life cycle of CR cells, these perspectives provide a novel focus for future investigations of the role of GABA in neuronal migration (Behar et al. 1996).

Acknowledgments

I wish to thank Dr Stefano Vicini for suggesting the BK channel experiment, and Dr Erminio Costa for stimulating discussions.

References

  1. Behar TN, Li Y-X, Tran HT, Ma W, Dunlap V, Scott C, Barker JL. GABA stimulates chemotaxis and chemokinesis of embryonic cortical neurons via calcium-dependent mechanisms. Journal of Neuroscience. 1996;16:1808–1818. doi: 10.1523/JNEUROSCI.16-05-01808.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ben-Ari Y, Khazipov R, Leinekugel X, Caillard O, Gaiarsa J-L. GABAA, NMDA and AMPA receptors: a developmentally regulated ‘ménage à trois’. Trends in Neurosciences. 1997;20:523–529. doi: 10.1016/s0166-2236(97)01147-8. [DOI] [PubMed] [Google Scholar]
  3. D'Arcangeloapos;arcangelo G, Miao GG, Chen S-C, Soares HD, Morgan JI, Curran T. A protein related to extracellular matrix proteins deleted in the mouse mutant reeler. Nature. 1995;374:719–723. doi: 10.1038/374719a0. 10.1038/374719a0. [DOI] [PubMed] [Google Scholar]
  4. Del rio JA, Heimrich B, Super H, Borrell V, Frotscher V, Soriano E. Differential survival of Cajal-Retzius cells in organotypic cultures of hippocampus and neocortex. Journal of Neuroscience. 1996;16:6896–6907. doi: 10.1523/JNEUROSCI.16-21-06896.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Derer P, Derer M. Development of Cajal-Retzius cells in vivo and in vitro. In: Sharma SC, Goffinet AM, editors. Development of the Central Nervous System in Vertebrates. New York: Plenum Press; 1992. pp. 113–129. [Google Scholar]
  6. Edwards FA, Konnerth A, Sakmann B, Takahashi T. A thin slice preparation for patch clamp recordings from neurones of the mammalian central nervous system. Pflügers Archiv. 1989;414:600–612. doi: 10.1007/BF00580998. [DOI] [PubMed] [Google Scholar]
  7. Haas HL, Härtterauml;rtter S, Hermes B, Hiemke C. Bath perfusion of slices with drugs. In: Kettenmann H, Grantyn R, editors. Practical Electrophysiological Methods: A Guide for In Vitro Studies in Vertebrate Neurobiology. New York: Wiley-Liss; 1993. pp. 129–131. [Google Scholar]
  8. Hestrin S, Armstrong WE. Morphology and physiology of cortical neurons in layer I. Journal of Neuroscience. 1996;16:5290–5300. doi: 10.1523/JNEUROSCI.16-17-05290.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Kalus P, Senitz D, Beckmann H. Cortical layer I changes in schizophrenia: a marker for impaired brain development? Journal of Neural Transmission. 1997;104:549–559. doi: 10.1007/BF01277671. [DOI] [PubMed] [Google Scholar]
  10. Kim HG, Fox D, Connors BW. Properties of excitatory synaptic events in neurons of primary somatosensory cortex of neonatal rats. Cerebral Cortex. 1995;5:148–157. doi: 10.1093/cercor/5.2.148. [DOI] [PubMed] [Google Scholar]
  11. Kyrosis A, Reichling DB. Perforated-patch recording with gramicidin avoids artifactual changes in intracellular chloride concentration. Journal of Neuroscience Methods. 1995;57:27–35. doi: 10.1016/0165-0270(94)00116-x. [DOI] [PubMed] [Google Scholar]
  12. McDonald JW, Johnston MV. Physiological and pathophysiological roles of excitatory amino acids during central nervous system development. Brain Research Reviews. 1990;15:41–70. doi: 10.1016/0165-0173(90)90011-c. [DOI] [PubMed] [Google Scholar]
  13. Mararíniacute;n-Padilla M. Cajal-Retzius cells and the development of the neocortex. Trends in Neurosciences. 1998;21:64–71. doi: 10.1016/s0166-2236(97)01164-8. [DOI] [PubMed] [Google Scholar]
  14. Mienville J-M. Low resting potential and upregulation of NMDA receptors: a possible mechanism for Cajal-Retzius cell death? Society for Neuroscience Abstracts. 1998;24:274. doi: 10.1523/JNEUROSCI.19-05-01636.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Mienville J-M, Barker JL. Potassium current expression during prenatal corticogenesis in the rat. Neuroscience. 1997;81:163–172. doi: 10.1016/s0306-4522(97)00171-1. [DOI] [PubMed] [Google Scholar]
  16. Ogawa M, Miyata T, Nakajima K, Yagyu K, Seike M, Ikenaka K, Yamamoto H, Mikoshiba K. The reeler gene-associated antigen on Cajal-Retzius neurons is a crucial molecule for laminar organization of cortical neurons. Neuron. 1995;14:899–912. doi: 10.1016/0896-6273(95)90329-1. [DOI] [PubMed] [Google Scholar]
  17. Penner R, Neher E. The role of calcium in stimulus- secretion coupling in excitable and non-excitable cells. Journal of Experimental Biology. 1988;139:329–345. doi: 10.1242/jeb.139.1.329. [DOI] [PubMed] [Google Scholar]
  18. Pesold C, Impagnatiello F, Pisu MG, Uzunov DP, Costa E, Guidotti A, Caruncho HJ. Reelin is preferentially expressed in neurons synthesizing γ-aminobutyric acid in cortex and hippocampus of adult rats. Proceedings of the National Academy of Sciences of the USA. 1998;95:3221–3226. doi: 10.1073/pnas.95.6.3221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Rakic P, Caviness VS., Jr Cortical development: view from neurological mutants two decades later. Neuron. 1995;14:1101–1104. doi: 10.1016/0896-6273(95)90258-9. [DOI] [PubMed] [Google Scholar]
  20. Ramamón y Cajaloacute;n y cajal S. Sur la structure de l'écorce cérébrale de quelques mammifères. Cellule. 1891;7:125–176. [Google Scholar]
  21. Sah P, Hestrin S, Nicoll RA. Tonic activation of NMDA receptors by ambient glutamate enhances excitability of neurons. Science. 1989;246:815–818. doi: 10.1126/science.2573153. [DOI] [PubMed] [Google Scholar]
  22. Schwartz TH, Rabinowitz D, Unni V, Kumar VS, Smetters DK, Tsiola A, Yuste R. Networks of coactive neurons in developing layer 1. Neuron. 1998;20:541–542. doi: 10.1016/s0896-6273(00)80993-9. [DOI] [PubMed] [Google Scholar]
  23. Sigel E, Baur R, Trube G, Mohler H, Malherbe P. The effect of subunit composition of rat brain GABAA receptors on channel function. Neuron. 1990;5:703–711. doi: 10.1016/0896-6273(90)90224-4. [DOI] [PubMed] [Google Scholar]
  24. Zhou FM, Hablitz JJ. Postnatal development of membrane properties of layer I neurons in rat neocortex. Journal of Neuroscience. 1996;16:1131–1139. doi: 10.1523/JNEUROSCI.16-03-01131.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES