Male Sprague-Dawly rats purchased from Samtako Bio (Kyung Gi-Do, Korea) were housed under a 12/12 h light/dark schedule with free access to food and water until used. Animals were grouped by postnatal day (P), as follows: infantile (P_7–9 and P_14–16), juvenile (P_21–23 and P_28–30), adolescence (P_36–37), and young adulthood (P_42–44 and P_49–51). Brains were rapidly extracted for electrophysiological recordings or Western blotting from animals anesthetized with ketamine and xylazine (80 mg/kg and 12 mg/kg, i.p., respectively). Animals in early infantile stage (P_7–9) were euthanized by decapitation without anesthesia. All animal experimentation was conducted in compliance with the policies of Chungnam National University regarding the use and care of animals.

Patch-clamp recordings were obtained in acutely prepared coronal hippocampal slices from male rats, as described previously [618]. Briefly, slices were perfused with artificial cerebrospinal fluid (aCSF; in mM: NaCl 126, KCl 2.5, MgSO₄ 1, NaHCO₃ 26, NaH₂PO₄ 1.25, glucose 20, ascorbic acid 0.4, CaCl₂ 1, pyruvic acid 2; pH 7.3~7.4; saturated with 95%O₂–5%CO₂) at a ~3 ml/min flow. Recordings were obtained at 32℃ using an Axopatch 200B amplifier (Axon Instruments, Foster City, CA). The series resistance was motored throughout the experiments. Neurons localized in the outer half of the granule cell layer were selected to minimize the effects of neurogenesis [19]. Patch pipettes were filled with a high Cl⁻ containing solution (in mM): KCl 140, HEPES 10, Mg²⁺ATP 5, MgCl₂ 0.9, and EGTA 10. Current output was filtered at 2 kHz and digitized at 10 kHz (Digidata 1322A, pClamp 9 software, Axon Instruments). I_tonic was defined as the difference between the holding current (I_holding) before and after application of the GABA_A receptor blocker bicuculline (20 µM). Drugs were added to the perfusing aCSF solution at known concentrations. All drugs except NO-711 (Tocris, UK) were purchased from Sigma-Aldrich (St. Louis, MO, USA).

All proteins from the dissected dentate gyrus (DGs) were lysed with 1× passive lysis buffer (Cell Signaling Technology, Danvers, MA, USA) and quantified using a Coomassie Protein assay kit (Bio-Rad, Hercules, CA, USA). Approximately 50 µg of protein was electrophoresed on a 10% sodium dodecyl sulfate polyacrylamide gel (SDS–PAGE) and transferred onto nitrocellulose membranes. The blots were blocked with 1× Tris buffered saline (TBS)-Tween 20 containing 3% bovine serum albumin (BSA) +2% heparan sulfate (HS) for 1 h at room temperature (5% TTBS; Gibco, USA). The blots were then incubated at 4℃ with primary antibodies against GABA_AR α₅ subunit, GAT-1, and GAT-3 (1:1,000; Millipore, USA) in 5% TTBS, respectively. The next day, the blots were incubated with a horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (1:2,000; Santa Cruz Biotechnology, USA). An enhanced chemiluminescence detection kit (ECL; Pierce, USA) was used to visualize antibody binding, and the intensity of the bands was measured using Image J software 1.42q (NIH, USA).

Numerical data are presented as means±standard error of the mean (SEM). Student's t-tests and analysis of variance, followed by post-hoc tests, were used as appropriate.

Altered expression of neuronal and glial GATs could contribute to I_tonic changes by altering the extracellular GABA concentration at different ages. To understand the age-dependent changes in GAT expression, we directly compared the expression of GAT-1 and GAT-3 in the DGs at different ages. Western blot analysis showed that the expression of GAT-1 and GAT-3 in the DGs gradually increased at P₇, P₁₄, and P₂₁, and thereafter stabilized at P₂₈, P₃₅, and P₄₂ (Fig. 1).

Changes in the expression and/or relative contribution of GAT-1 and/or GAT-3 may cause I_tonic changes at different ages. To discriminate the functional role of neuronal and glial GATs in I_tonic, we measured and compared the I_tonic of DGGCs in the presence of selective GAT-1 and GAT-3 blockers at different ages.

Bath application of NO-711 (5 µM), a selective GAT-1 blocker [20], failed to induce consistent changes in I_holding at P_6–8, while it induced a similar inward shift in I_holding (I_NO-711) at P_13–15 (21.8±4.4 pA, n=7), P_20–22 (20.8±6.1 pA, n=6), P_27–29 (21.4±3.6 pA, n=5), and P_34–36 (27.3±3.4 pA, n=7) (Fig. 2A). In a subset of experiments, we measured I_NO-711 in the presence of 1 µM GABA to confirm that GAT-1 activity facilitates I_tonic at the early infantile age. GABA (1 µM)-induced I_holding shift were decreased with age (P_6–8, 16.3±4.2 pA, P_20–22, 7.6±1.7 and P_34–36, 3.3±1.2 pA, n=5~6). Additional application of NO-711 (GABA+NO-711) caused a significant inward shift in I_holding even at P_6–8, which was still much smaller than at P_20–22 and P_34–36 (p<0.01 in both cases; Fig. 2B).

To investigate the functional significance of changes in GAT-3 expression, a selective GAT-3 blocker, SNAP-5114, was used [14]. Because SNAP-5114 (100 and 300 µM) alone failed to elicit changes in I_holding, we sequentially applied NO-711 and NO-711+SNAP-5114. In the presence of NO-711, 300 µM SNAP-5114 induced a significant inward shift in I_holding (I_SNAP-5114) at P_6–8, P_20–22, and P_34–36. I_SNAP-5114 was significantly smaller at P_6–8 than at P_20–22, and P_34–36 (p<0.05 in both cases; Fig. 2C). Even in the presence of NO-711, 100 µM SNAP-5114 did not induce significant changes in I_holding at all tested age groups (Fig. 2C). Thus, in a subset of experiments, we assessed I_SNAP-5114 in the presence of 1 µM GABA at different ages. GABA induced significant I_holding changes at P_6–9 (22.6±3.5 pA, n=6), while it induced minimal changes in I_holding at P_20–22 (1.3±1.0 pA, n=5) and P_34–36 (1.9±0.9 pA, n=5). An additional 100 µM SNAP-5114 efficiently induced an inward shift in I_holding at P_6–9, while it failed to cause significant changes in I_holding at P_20–22 and P_34–36 in the presence of 1 µM GABA (Fig. 2D). These results support the idea that GAT3 activity is apparent with an increased extracellular GABA concentration in the hippocampus [16].

To investigate the combined role of GAT-1 and GAT-3 in developmental I_tonic changes, we measured and compared the I_tonic of DGGCs in the presence of the nonselective GAT blocker, nipecotic acid (NPA), at different ages.

In contrast to selective GAT-1 or GAT-3 blockade, NPA (100 µM) induced a significant inward shift in the I_holding of DGGCs (I_NPA), blocked by BIC, at all tested age groups. Interestingly, I_NPA gradually decreased during the infantile and juvenile periods (P_6–8, P_13–15, P_20–22, and P_27–29), and stabilized in adolescence and young adulthood (P_34–36 and P_41–43) (Fig. 3A and B).

We hypothesize that the age-dependent I_NPA change may mirror the functional maturation of GABA_ARs rather than the upregulation of GAT expression in preadolescence, and that the large I_NPA may be due to a high ambient GABA level. Therefore, we directly compared INPA inhibition of L-655,708, an inverse agonist at the benzodiazepine binding site of α₅-GABA_ARs [21]. The bath application of L-655,708 (5 µM) efficiently blocked I_NPA at all tested ages (p<0.01 in all cases; Fig. 3A). Interestingly, in agreement with gradual I_NPA attenuation with age, L-655,708-sensitive I_NPA also gradually decreased during preadolescence (Fig. 3B). As a result, the portion of L-655,708-sensitive total I_NPA did not differ by age.

To understand the functional changes in GABA_ARs according to age, we directly compared I_tonic activated by exogenous GABA (5 µM) and their sensitivity to L-655,708 at different ages.

I_tonic gradually decreases as infants mature into adolescence. The large I_NPA that characterized infantile periods (P_6–8 and P_13–15) gradually decreased at the juvenile (P_20–22, and P_27–29) and adolescent (P_34–36) stages, and thereafter stabilized in young adults (P_41–43) (Fig. 4A and B). The bath application of L-655,708 (5 µM) partially blocked I_tonic in the presence of GABA at all tested ages (p<0.01 in all cases; Fig. 4A). In agreement with I_tonic attenuation, L-655,708-sensitive I_tonic gradually decreased in preadolescence (Fig. 4A and B). We observed a tendency for the portion of L-655,708-sensitive total I_tonic to decrease with age, although this did not reach statistical significance.

In further experiments, we directly compared the expression of the GABA_AR α₅ subunit in DGs in the different age groups (Fig. 4C and D). Although we detected very low level of GABA_ARs α₅ subunit immune reactivity at P₇ and P₁₄ in DGs, Western blot analysis showed that the expression of the GABA_ARs α₅ subunit gradually decreased in the juvenile (P₂₁ and P₂₈) and adolescent (P₃₅) periods, and thereafter stabilized in the young adults (P₄₂). The degree of GABA_ARs α₅ subunit expression was not further changed at P₄₉ and P₅₆ (data not shown).

GATs embedded in axon terminal membranes and/or astrocyte plasma membranes regulate ambient GABA levels. In many neural systems, GAT-1 antagonists alone result in a smaller I_tonic versus that observed when both GAT-1 and GAT-3 are blocked [2223]. This was explained by the involvement of both GAT-1 and GAT-3 transporters in regulating extracellular GABA concentrations around neurons. Similarly, I_NPA was much larger than I_NO-711 in infantile and juvenile DGGCs in the present study. Interpretation of this difference is complicated because, as opposed to NO-711, NPA is a GAT substrate. In addition, NPA could result in heteroexchange for GABA by GATs [24]. However, given that the concentrations of NPA and NO-711 used in this study were, respectively, about five and ten times that of the IC₅₀ used for GAT-1 blockade [14], a simple interpretation could be that NPA blocked GAT-3 more efficiently than NO-711 during the infantile stage. However, our results showed that the GAT-1 blocker alone, and the GAT-1 blocker with additional application of SNAP-5114 (100 µM), resulted in a similar average I_tonic of 20 pA in DGGCs after the late infantile periods; these results contradict the idea that GAT-3 actively contributed to the I_tonic of DGGCs. In the present study, SNAP-5114 facilitated I_tonic at the concentration of 300 µM, which was near to the IC₅₀ for GAT-1 blockade [14], further confounding the role of GAT-3 in the I_tonic of DGGCs. Combined with the fact that both GAT-1 and GAT-3 expression increased until the late juvenile periods, these results appear to support a major role of GAT-1 in the GABA uptake regulating I_tonic in the adult hippocampus [16]. In general, our results suggest that GAT-1 and GAT-3 play a primary and adjunctive role, respectively, in regulating I_tonic of DGGCs in preadolescece.

As with other neurotransmitter transporters, GATs can also act in reverse mode and thus release GABA from cells. Indeed, GABA can be secreted from cells by the reversed transport direction of GATs, particularly during early postnatal stages [2526]. Thus, it is possible that GABA release via reversed GAT activity is integral in maintaining GABA levels that activate I_tonic [27] in the infantile and early juvenile stages. However, our finding that GAT-1 and GAT-3 blockers always enhanced the I_tonic of DGGCs suggest that the two transporters operate synergistically to promote GABA uptake, which was seen at all tested ages in our experiments.

Both the α₅ and d subunit are key mediating components of the I_tonic of DGGCs [28]. Alterations in GABA concentrations affect the relative contribution of specific GABA_ARs to I_tonic as different receptor populations are recruited [8]. In the present study, BIC uncovered basal I_tonic shown by the outward I_holding shift went over the initial level, especially when the inward shift in I_holding by exogenous GABA and GAT blockers was less than ~30 pA. α₅–GABA_ARs contribute to I_tonic when ambient GABA concentrations increase, while at low ambient GABA concentrations the activation of δ–GABA_ARs predominates [9]. In the present study, I_NPA was larger than I_NO-711 during preadolescence, which could be explained by an ambient GABA concentration sufficient to recruit additional α₅–GABA_ARs in the presence of NPA, but not in NO-711. Our results showed that the portion of L-655,708-sensitive I_NPA ranged from ~27% to ~32%; this is consistent with previous findings showing that I_tonic is mediated by α₅–GABA_ARs and is responsible for ~29% of the total I_tonic in DGGCs [28]. Thus, α₅–GABA_ARs mediated I_NPA at all tested ages. Combined with the results whereby I_NPA and GABA_AR α₅ subunit expression gradually decreased with age, our results suggest that the age-dependent I_NPA decrease mirrors the functional decrease of α₅–GABA_ARs rather than changes in GATs activity, during preadolescence.

However, there may be an as-yet undiscovered, non α₅–containing GABA_ARs responsible for the large I_NPA observed during preadolescences. Indeed, in the present study, GABA_AR α₅ subunit immunoreactivity was not detectable in DGs at infantile stages. It is also notable that the small amplitude of I_NO-711 prevented us from comparing the sensitivity of I_NPA and that of I_NO-711 to L-655,708 in DGGCs. Regarding GABA_ARs activated in the presence of NPA, it is also noteworthy that NPA can directly activate GABA_AR-like channels [29]. However, to the best of our knowledge, there is no information on the composition of GABA_ARs directly activated by NPA. However, it is still of interest that GABA_AR α₅ subunit expression gradually decreased with the postnatal development in various brain regions.

Overall, our results showed that GATs blockades elevated ambient GABA level sufficiently to harmonize with α₅-GABA_ARs, resulting in an age-dependent I_tonic decrease in preadolescent brains. Combined with the fact that GABA_AR α₅ subunit expression in the hippocampus is closely related to learning and memory in young adults [3031], selective pharmacological modulators, such as α₅-GABA_AR selective inverse agonists, may be effective in increasing cognitive performance in memory disorders [32]. Future studies are warranted to elucidate the pathophysiology of α₅-GABA_ARs generating I_tonic combined with GAT blockades in the developing brains of preadolescents.