Ethanol exposure during the third trimester equivalent does not affect GABAA or AMPA receptor-mediated spontaneous synaptic transmission in rat CA3 pyramidal neurons

Background Ethanol exposure during the rodent equivalent to the 3rd trimester of human pregnancy (i.e., first 1–2 weeks of neonatal life) has been shown to produce structural and functional alterations in the CA3 hippocampal sub-region, which is involved in associative memory. Synaptic plasticity mechanisms dependent on retrograde release of brain-derived neurotrophic factor (BDNF) driven by activation of L-type voltage-gated Ca2+ channels (L-VGCCs) are thought to play a role in stabilization of both GABAergic and glutamatergic synapses in CA3 pyramidal neurons. We previously showed that ethanol exposure during the first week of life blocks BDNF/L-VGCC-dependent long-term potentiation of GABAA receptor-mediated synaptic transmission in these neurons. Here, we tested whether this effect is associated with lasting alterations in GABAergic and glutamatergic transmission. Methods Rats were exposed to air or ethanol for 3 h/day between postnatal days three and five in vapor inhalation chambers, a paradigm that produces peak serum ethanol levels near 0.3 g/dl. Whole-cell patch-clamp electrophysiological recordings of spontaneous inhibitory and excitatory postsynaptic currents (sIPSCs and sEPSCs, respectively) were obtained from CA3 pyramidal neurons in coronal brain slices prepared at postnatal days 13–17. Results Ethanol exposure did not significantly affect the frequency, amplitude, rise-time and half-width of either sIPSCs or sEPSCs. Conclusions We show that an ethanol exposure paradigm known to inhibit synaptic plasticity mechanisms that may participate in the stabilization of GABAergic and glutamatergic synapses in CA3 pyramidal neurons does not produce lasting functional alterations in these synapses, suggesting that compensatory mechanisms restored the balance of excitatory and inhibitory synaptic transmission.

bundle in the CA3 sub-region [5,6]. A reduction in 3 H-vinlidene kainate binding sites was demonstrated in the CA3 stratum lucidum of the ventral hippocampus from 45 day-old rat offspring exposed to ethanol during fetal development [7]. Using electron microscopy, Tanaka et al [8] showed that prenatal ethanol exposure decreases the number of synapses in the CA3 sub-region at gestational day 21. Studies suggest that exposure during periods equivalent to the human 3 rd trimester of pregnancy can have even more significant effects on this hippocampal sub-region. West and Hamre [9] reported that exposure to ethanol between postnatal day (P) 1 and P10 was associated with the presence of aberrant intra-pyramidal and infra-pyramidal mossy fibers across the CA3 sub-region. Binge-like ethanol exposure during P4-P10 (but not gestational days [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20] decreased the number and density of pyramidal cells in this sub-region [10,11]. A similar finding was reported by Miki et al [12] who detected a reduction in CA3 pyramidal neuron number in rats exposed to ethanol between P10 and P15. However, it is noteworthy that studies using both guinea pigs and rats have failed to detect alterations in the number of pyramidal neurons in this hippocampal sub-region [13,14]. Therefore, several studies have investigated the possibility that developmental ethanol exposure impairs the function of CA3 neurons rather than affecting their morphology. An electrophysiological study with 50-70 day-old offspring from rats exposed to ethanol throughout gestation reported a reduction in the frequency of high potassiuminduced epileptiform bursts in the CA3 stratum radiatum [15]. Galindo et al [16] found that acute ethanol exposure increased network-driven giant depolarizing potentials in CA3 pyramidal neurons from neonatal rats, an effect that is likely a consequence of increased GABA A receptormediated excitatory synaptic transmission. It was also demonstrated that these immature neurons do not develop tolerance to this effect of ethanol [17]. Acute and repeated ethanol exposure between P2 and P6 was shown to inhibit brain-derived neurotrophic factor (BDNF)-and L-type voltage-gated Ca 2+ channel (L-VGCC)-dependent long-term potentiation of GABA A receptor-mediated spontaneous postsynaptic currents in CA3 pyramidal neurons [18]. BDNF/L-VGCC-dependent plasticity mechanisms are thought to play a role in the stabilization of both GABAergic and glutamatergic synapses in developing hippocampal neurons [19][20][21]. Based on these findings, we hypothesized that the ethanol-induced alterations of BDNF/L-VGCC-dependent synaptic plasticity result in a persistent reduction in both GABAergic and glutamatergic synaptic currents in CA3 pyramidal neurons. To test this hypothesis, we exposed neonatal rats to ethanol from P3-P5 and measured GABA A receptor-and AMPA receptordependent spontaneous postsynaptic currents at P13-P17 using patch-clamp slice electrophysiological techniques.
We evaluated the impact of this ethanol exposure paradigm on GABA A receptor-mediated spontaneous inhibitory postsynaptic currents (sIPSCs). Slice electrophysiological recordings were performed at P13-17. Fig. 1a shows examples of sIPSCs recorded from CA3 pyramidal neurons in slices from control and ethanol treated rats. Neither the frequency of these events (U = 20; P = 0.39 by Mann-Whitney test) nor their amplitude (U = 26; P = 0.86 by Mann-Whitney test) were significantly different between treatment groups (Fig. 1b-c). Likewise, the half-width of the events was not significantly affected by ethanol exposure (t(12) = 0.96; P = 0.35 by t-test; Fig. 1d) nor was the rise time (control = 2.04 ± 0.42 ms, n = 8; ethanol = 1.71 ± 0.53 ms, n = 6; t(12) = 0.48; P = 0.63 by t-test).
We then assessed the impact of P3-5 ethanol exposure on AMPA receptor mediated sEPSCs. Slice electrophysiological recordings were also performed at P13-17. Sample sEPSC recordings are shown in Fig. 2a. Neither the frequency of these events (t(14) = 1.35; P = 0.19 by ttest) nor their amplitude (t(14) = 0.76; P = 0.45 by t-test) were significantly different between treatment groups (Fig. 2b-c). Moreover, neither the sEPSC half-width (t(14) = 0.78; P = 0.44 by t-test; Fig. 2d) nor the rise time (control = 1.82 ± 0.07 ms, n = 8; ethanol = 1.71 ± 0.11 ms, n = 8, t(14) = 0.77; P = 0.45 by t-test) were significantly affected by ethanol exposure. Spontaneous EPSCs triggered by glutamate release from mossy fibers have been demonstrated to have amplitudes ≥100 pA [23]. Therefore, we analyzed the effect of ethanol exposure on these large events. As shown in Table 1, ethanol exposure did not significantly affect the frequency, amplitude, risetime or half-width of these events. Small events with an amplitude ≤100 pA were not significantly affected either. The percent of events with amplitudes ≥100 pA was 17.5 ± 4.6 % for the control group and 17.2 ± 2.5 % for the ethanol group (t(14) = 0.063; P = 0.95 by unpaired t-test; n = 8).

Discussion
Our ethanol exposure paradigm did not significantly affect the properties of sIPSCs, a finding that does not support the hypothesis that ethanol exposure during this critical period, when CA3 pyramidal neurons express BDNF/L-VGCC-dependent synaptic plasticity, produces a lasting impairment in GABA A receptor-dependent synaptic transmission. We did not observe changes in sIPSC frequency, amplitude, rise-time or half-width, indicating that ethanol exposure neither affects spontaneous GABA release nor the function of postsynaptic GABA A receptors. These findings suggest that ethanol did not have an overall effect on the function of GABAergic interneurons that provide inhibitory inputs to the soma and proximal dendrites of CA3 pyramidal neurons. Due to space-clamp limitations, GABAergic inputs to distal dendritic sites cannot be reliably measured using somatic patch-clamp electrodes [24]. Therefore, we cannot discount the possibility that these were affected by ethanol exposure. This issue could be addressed in the future using current-clamp recordings of spontaneous inhibitory postsynaptic potentials or voltage-clamp experiments of IPSCs evoked by electrical stimulation of distal inputs. As mentioned above, BDNF/L-VGCC-dependent long-term potentiation of GABAergic transmission is thought to play a role in stabilizing GABA A receptor-containing synapses in the soma and/or proximal dendrites of CA3 pyramidal neurons [19]. Therefore, it is somewhat surprising that ethanol-induced inhibition of this plasticity mechanism did not affect the properties of GABA A -dependent sIPSCs [18]. A potential explanation for the lack of a lasting effect is that the action of ethanol is transient, occurring only while this agent is present during the 3 h/day exposure on P3-P5 (and several hours/day post-exposure while ethanol is being metabolized) and reversing at some point between this period and the time at which the electrophysiological recordings were acquired (P13-P17). We previously demonstrated that ethanol vapor exposure from P2-P16 causes a significantly delay in the acquisition of action potentialindependent GABA A receptor mediated postsynaptic currents in CA3 pyramidal neurons [25], suggesting that repeated ethanol exposure encompassing a longer period of development could alter plasticity mechanisms involved in the maturation of GABAergic synapses. Out studies also failed to demonstrate an effect of ethanol exposure on the frequency, amplitude, rise-time or half-width of AMPA receptor-mediated sEPSCs, including large events likely mediated by glutamate release from mossy fibers. These results suggest that ethanol did not have lasting effects on glutamatergic transmission at the pre-or post-synaptic levels. The lack of an effect of ethanol on large sEPSCs is unexpected given that BDNF/L-VGCC-dependent synaptic plasticity has been shown to strengthen mossy fiber-to-CA3 pyramidal neuron synapses during postnatal development [20]. Moreover, we have previously demonstrated that acute ethanol exposure inhibits glutamate release and AMPA receptor-mediated EPSCs in slices from P3-P10 rats [26] and it would have been expected for these acute effects to cause lasting compensatory pre-and/or post-synaptic alterations in glutamatergic transmission [27][28][29]. As was the case for GABAergic transmission, it is possible that neurons were able to counteract the effects of ethanol, leading to a normalization of glutamatergic transmission at the time of our electrophysiological experiments. For instance, neurons could have engaged alternative pathways involved in synapse stabilization, including upregulation of modulatory neurotransmitter systems (e.g., serotonin) [30,31].

Conclusions
We demonstrate here that a 3 rd trimester-equivalent ethanol exposure paradigm that significantly impairs synaptic plasticity mechanisms thought to be involved in the stabilization of GABAergic and glutamatergic synapses in CA3 pyramidal neurons does not produce lasting alterations in spontaneous synaptic transmission mediated by GABA A and AMPA receptors. These findings suggest that neonatal CA3 pyramidal neurons deploy compensatory mechanisms in response to this exposure paradigm that ultimately restore the balance of excitatory and inhibitory synaptic transmission. It should be emphasized, however, that even if the effects of our P3-P5 exposure paradigm are transient, these could still alter the trajectory of developmental processes other than the refinement of GABAergic and glutamatergic synaptic transmission [32] and this issue should be investigated in the future.
Software, San Diego, CA). Data were initially analyzed with the Kolmogorov-Smirnov normality test and if they followed a normal distribution, they were subsequently analyzed using parametric tests (t-tests or two-way ANOVA). If data did not pass the normality test, they were analyzed using two-tailed Mann-Whitney tests. Data are presented as mean ± SEM. The unit of determination is the average of results obtained with slices from a single animal (1-3 cells/animal; each cell from a different slice).

Competing interests
The authors declare that they have no competing interests.
Authors' contributions BCB provided intellectual contribution, exposed the animals to ethanol vapor, performed the electrophysiological recordings, performed data analysis, and edited the manuscript. CFV contributed intellectually, assisted with the vapor chamber exposure paradigm, performed data analyses, interpreted the data, and wrote/edited the manuscript. Both authors read and approved the final manuscript.