Age-related GABA_A receptor subunit changes in rat auditory cortex

Abstract

Auditory cortex (AI) shows age-related decreases in pre-synaptic markers for GABA and degraded AI neuronal response properties. Prior studies find age-related increases in spontaneous and driven activity, decreased spectral and directional sensitivity, and impaired novelty detection. The present study examined expression of GABA_A receptor (GABA_AR) subunit message, protein and quantitative GABA_AR binding in young, middle-aged, and aged rat AI, with comparisons to adjoining parietal cortex. Significant loss of GABA_AR α₁ subunit message across AI layers was observed in middle-aged and aged rats while α₁ subunit protein levels declined in layers II and III. Age-related increases in GABA_AR α₃ subunit message and protein levels were observed in certain AI layers. GABA_AR subunits, including β₁, β₂, γ₁, γ_2s, and γ_2L, primarily, but not exclusively, showed age-related declines at the message and protein levels. The ability of GABA to modulate [³H]TBOB binding in the chloride channel showed age-related decreases in peak binding and changes in desensitization kinetics. Collectively, age-related changes in GABA_AR subunit composition would alter the magnitude and temporal properties of inhibitory synaptic transmission and could underpin observed age-related functional changes seen in the elderly.

1. Introduction

Age-related functional changes in a number of sensory systems are strongly suggestive of a loss of normal adult inhibitory amino acid neurotransmission (Angelotti and Macdonald, 1993; Belelli et al., 2005; Burianova et al., 2009; for review Canlon et al., 2010; Caspary et al., 2008; Gutierrez et al., 1997; Lloyd et al., 1990; Maksay and Ticku, 1985; Malherbe et al., 1990; Mendelson and Rajan, 2011; Olsen et al., 1990; Pinto et al., 2010; Sakurai et al., 1994; Suta et al., 2011; Syka, 2002; Winer, 1992; Wisden et al., 1992; Ymer et al., 1990). In part, these changes in central inhibition likely reflect a compensatory age-related response to decreased peripheral sensory input, reflecting homeostatic plasticity (Noreña, 2011; Oliver et al., 2011; Richardson et al., 2011; Turrigiano and Nelson, 2004). Compensatory age-related changes can result in decreased markers of normal functional inhibition in auditory and visual cortices (Hua et al., 2006; Hughes et al., 2010; Leventhal et al., 2003; Liang et al., 2008; Schmidt et al., 2010). Even when presented with suprathreshold acoustic stimuli, many middle-aged and elderly humans show decreased speech understanding, impaired sound localization, and loss in the ability to extract novel or salient signals from a complex acoustic background (Anderson et al., 2012; Dubno et al., 1984; Fitzgibbons and Gordon-Salant, 1994,2010; Fogerty et al., 2010; Lui and Mendelson, 2003; Ostroff et al., 2003; Pichora-Fuller et al., 2007; Schneider et al., 1994; Snell, 1997; Strouse et al., 1998; Suta et al., 2011; Tremblay et al., 2002, 2003).

Recent electrophysiologic studies in rat and primate auditory cortex (AI) find age-related increases in spontaneous and sound-evoked discharge rates as well as less precise directional sensitivity, loss of spectral precision, and impaired novelty detection (de Villers-Sidani et al., 2010; Hughes et al., 2010; Juarez-Salinas et al., 2010; Martin Del Campo, et al., 2012). Therefore, both human and animal studies suggest that changes in sound processing are consistent with a hypothesis of an age-related loss of normal adult functional inhibition.

Pre-synaptic markers for GABA, including GABA levels and levels of the GABA synthetic enzyme glutamic acid decarboxylase (GAD) are down-regulated across aged AI in humans and animal models of aging (Burianova et al., 2009; de Villers-Sidani et al., 2010; Ling et al., 2005; McGeer and McGeer, 1980). Few studies have examined GABA_A receptor (GABA_AR) markers across the layers of AI (Pirker et al., 2000; Wisden et al., 1992; Yu et al., 2006). The impact of aging on the subunit makeup of GABA_ARs across AI layers is the focus of the present study. The adjoining parietal cortex was used for comparison. GABA_ARs exist as pentameric subunit complexes which can be allosterically modulated by numerous pharmacological agents (Rabow et al., 1995 ; Sieghart, 1992a, b, c, d; Sieghart, 1995; Sieghart et al., 1992; Wafford et al., 1993; Yu et al., 2006). Molecular cloning has revealed 6-α, 4-β, 3-γ, 1-δ, 1-ε, 1-π, 1-θ, and 3-ρ GABA_A receptor subunits (Olsen and Sieghart, 2008,2009; Rabow et al., 1995; Rudolph et al., 2001; Sieghart, 1995; Wafford and Ebert, 2006). GABA_AR subunit constructs exhibit specific regional and likely cortical layer specific distributions (Pirker et al., 2000; Wisden et al., 1992; Yu et al., 2006). Altered GABA_AR subunit composition/stoichiometry, potentially in response to age-related presynaptic changes, would impact GABA_AR mediated inhibitory function. Age-related subunit changes would alter the magnitude and temporal precision of inhibitory currents, in turn degrading sensory processing (Angelotti and Macdonald, 1993; Ducic et al., 1995; Macdonald and Olsen, 1994; Takesian et al., 2012; Wafford et al., 1993). The present study examined GABA_AR subunit message, protein and quantitative GABA_AR binding of selective GABA_A ligands in young, middle-aged, and aged rat auditory cortex.

2. Materials and methods

2.1. Animals

Young-adult (4–6 months), middle-aged (20–22 months) and aged (30–32 months) male Fischer Brown Norway (FBN) rats were obtained from Harlan Sprague-Dawley, Inc. (Indianapolis, IN). All experiments were carried out under animal use protocols approved by the Southern Illinois University School of Medicine Laboratory Animal Care and Use Committee. Age-related hair cell loss and age-related threshold shifts for this strain have been previously described (Turner and Caspary, 2005; Wang et al., 2009a).

2.2. Sampling criteria for rat auditory cortex

Sections were collected through the center of AI from an area at Bregma −4.80 mm (Plate 39) to Bregma −4.16 (Plate 36) (Paxinos and Watson, 1998) identified by measuring 2.25 mm dorsal from the rhinal fissure. The present study used criteria adapted from Winer (1992) and Games and Winer (1988) to define data collection areas sampled from layers II–VI of FBN rat AI. A detailed algorithm for measures used to identify AI layers II–VI was derived from Winer (1992) and Games and Winer (1988) and is detailed in Ling et al. (2005).

2.3. Quantitative in situ hybridization

Eighteen FBN rats (6 young-adult, 6 middle-aged and 6 aged) were decapitated, brains rapidly removed, rinsed in ice-cold phosphate-buffer saline (PBS) (pH 7.4, DEPC-treated), frozen in powdered dry ice, and stored at −80°C. Serial transverse sections (16μm) through AI were cut using a cryostat (Leica CM1850 Microsystems Nussloch GmbH, Nussloch, Germany) set at −18°C. Sections were thaw-mounted onto Superfrost/Plus slides (Thermo Fisher Scientific, Pittsburgh, PA, USA) at approximately the same position, two sections per slide, and stored at −20°C (<48 hours) until processed for in situ hybridization.

GABA_A subunit probe preparation: Nine 40–48 mer oligonucleotide probes were synthesized and purified by Sigma Genosys (Woodlands, TX). Sequences selected were based upon published sequences: α₁ (Khrestchatisky et al., 1989); α₂ (Pritchett and Seeburg, 1990); α₃ (Malherbe et al., 1990); α₄ (Wisden et al., 1991); β_1–3 (Ymer et al., 1989); γ₁, γ_2s, _& γ_2L (Ymer et al., 1990). Procedures for oligonucleotide probe end-labeling, in situ hybridization steps and data analysis are as described in (Ling et al., 2005) and were modified from Milbrandt et al. (1997). In brief, five picoMoles of oligonucleotide probes in 50μl labeling mixture were 3′ end-labeled for 10 min at 37°C with 0.5 μM of ³⁵S-deoxyadenosine triphosphate (dATP) (PerkinElmer Inc., Downers Grove, IL, USA) using terminal deoxynucleotidyl transferase (16units/μl) (Fisher Scientific, Pittsburgh, PA, USA). The reaction was halted by addition of 50 μl of TE buffer. Ten mg/ml tRNA was added to enhance recovery of the labeled probe. Labeled probes were extracted using a phenol/chloroform. After hybridization and post-washing steps, slides were dried and dipped in NTB-2 photographic emulsion (VWR, West Chester, PA, USA) and stored in the dark at 4°C for 4 weeks. Exposed sections were then developed, fixed, and counterstained with Thionin for cell identification. Adjacent sections were used as controls for specificity. Competitive blocking of labeled oligonucleotides using excess concentrations (50-fold) of unlabeled oligonucleotide and incubation with labeled sense oligonucleotides were used as controls. Detailed hybridization procedure, consistency and quality control was described in Ling et al. (2005).

Quantitative analysis of hybridization labeling: Images were captured using a CoolSnap monochrome digital camera connected to an MCID-Elite 6.0 imaging system (InterFocus Imaging Ltd., Cambridge, England) with 40× objective. Accumulation of silver grains over neuronal cell bodies was interpreted as hybridization of the probe to its corresponding mRNAs (Fig. 1). The identity of sections was concealed/blinded to insure unbiased quantification. The counting parameters such as threshold, light intensity, and counting area were maintained consistently throughout the counting procedure for a particular subunit. Only grains within the neuronal perimeter were counted by the automated counting system-MCID Elite 6.0 (InterFocus Imaging Ltd., Cambridge, England) over cells distinguishable from adjacent cells and showing a visible nucleus. Quantitative comparisons were made only within a given subunit probe not across probes. With two sections per animal, two different fields, of fixed size, from each of the layers (II–VI) of AI in each section were digitized and grain counts, neuronal number, size, and area recorded. Background labeling measurements were obtained from three random areas located off the tissue sections. Somatic area and number of grains over the somata of at least 10 cells in each of the layers (II–VI) of AI in each section were measured. Data were collected as grain density (number of grains/100μm² of cell area) and corrected by subtraction of nonspecific hybridization for each layer/subunit/age group. Analysis of Variance (ANOVA) was used to determine if differences in background-adjusted mean grain density was attributed to the treatment variables. Tests subsequent to the ANOVA were carried out using the Bonferroni procedure to control overall type I error rate (Ling et al., 2005).

Figure 1.

Distribution of GABA_AR α₁ and α₃ mRNA in layer III neurons of AI from young and aged FBN rats. Clusters of silver grains represent hybridization of transcripts of GABA_AR α₁ (A&B) and α₃ (C&D) with ³⁵S-labeled selective oligonucleotide probes. Reduction of silver grains in aged AI neurons of layer III (B), when compared with that in neurons from young adult rat AI (A), indicates the age-related loss of GABA_AR α₁ mRNA. In contrast, GABA_AR α₃ mRNA shows increased in aged layer III neurons (D). Scale bar=10 μm.

2.4. Quantitative immunohistochemistry

The methods used for quantitative immunohistochemistry were based on those published by Ling et al. (2005).

Tissue preparation for immunohistochemistry: FBN rats were anesthetized with a mixture of ketamine (105 mg/kg body wt. i.p.) and xylazine (7 mg/kg body wt. i.p.), and transcardially perfused with 150 ml of physiological saline containing 0.1% of sodium nitrite, followed by 1 L of fixative containing 4% paraformaldehyde in Sorenson’s K-Na phosphate buffer (pH 7.4). Brains were removed, post-fixed for 1 h in the same solution, washed in 0.1 M PBS for 30 min and immersed overnight in PBS containing 20% sucrose. Cryoprotected tissues were stored at −80°C.

Immunohistochemistry: FBN rats used in GABA_AR α₁ and β₁ studies were 4 young, 4 middle-aged and 4 aged, and in GABA_AR α₃ and β₂ studies there were 6 young, 6 middle-aged and 6 aged rats. Serial transverse sections through AI were cryostat sectioned at 30 μm and collected as free-floating sections in ice-cold 0.1 M PBS. The sections were rinsed in PBS, transferred to blocking solution (1.5% normal serum and 5% non-fat dry milk in PBS) for 30 min and incubated at room temperature in primary antibodies for 1 h and then at 4°C overnigh t with agitation. Polyclonal goat anti-GABA_AR α₁ and β_1–2 (1:150) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Polyclonal rabbit anti-GABA_AR α₃ (1:500) was obtained from Alomone Labs (Jerusalem, Israel). After rinsing in PBS, sections were processed using Vectastain ABC kits (Vector Laboratories, Burlingame, CA, USA). The labeling was visualized using diaminobenzidine (DAB, Sigma-Aldrich, St. Louis, MO, USA) and DAB reaction time was fixed at 2 min. Sections were then mounted onto the Superfrost/Plus slides. When possible, specificity of primary antibodies was tested by pre-incubation with the control peptide antigens. Secondary antibodies were controlled in cells processed as described above but in the absence of primary antibodies.

To help minimize variability, unrelated to treatments, immunostaining and measurements were carried out in parallel groups, with tissue from one young, one middle-aged and one aged animal processed at the same time. Sections were blinded so age groups were unknown to the observer. Flat-field correction was performed prior to digitizing images and held consistent across each group. Digital images of immuno- processed sections were captured at an objective magnification of 40× as described above. Two fields from each AI layer (LII–LVI) per section and two to four sections per animal were analyzed. Relative optical density (ROD) measurements, which are proportional to immunostaining intensity, were measured from all positively stained neurons encountered across layers II–VI of AI. All ROD measurements were corrected by subtracting background values obtained from the measurements of immunonegative cells in layer V. Only neurons with intact soma outlines and discernible nuclei and nucleoli were measured. All data were expressed as means ± S.D. of ROD.

2.5. Quantitative receptor binding autoradiography

RO15-4513 binds both “wild type” (2α₁2β₂γ₂) and non-“wild type” GABA_ARs. Functionally, RO15-4513 acts as a partial inverse agonist at γ₂ subunit containing GABA_ARs (Korpi et al., 2002; Luddens and Wisden, 1991; Wisden et al., 1991), while acting as an agonist at α₄ and α₆ subunits-containing GABA_ARs (Hadingham et al., 1996; Knoflach et al., 1996; Linden et al., 2011; Wafford et al., 1996). Binding of the GABA_AR radioligand t-butylbicycloorthobenzoate (TBOB), can be modulated by varying concentrations of GABA, and has been used in picrotoxin ligand binding assays for studying GABA_AR pharmacology and receptor diversity (Lloyd et al., 1990; Maksay and Ticku, 1985; Olsen et al., 1990; Sakurai et al., 1994). [³H]RO15-4513 saturation analysis based on previous studies was used in order to reveal differences in kd or Bmax (Braestrup et al., 1983; Niddam et al., 1987; Ruano et al., 1993). Concentrations (1, 3, 5, 8, 10, 15nM) of [³H]RO15-4513 (20Ci/mmol, Perkinelmer Inc., San Jose, CA, USA) were added to the incubation buffer and 100μM of flumazenil was added as a displacer. Modulation of [³H]TBOB binding was carried out with increasing concentrations of GABA from 10nm to 5μm (Milbrandt and Caspary, 1995). Autoradiograms were generated by apposing slides to a phosphor screen, and the screen was then scanned using Cyclone phosphor system (Perkinelmer Inc., San Jose, CA, USA). Images were collected at 600 DPI and analyzed using OptiQuant image analysis software. The superficial layers II–IV were grouped into one box (38% cortical thickness), with layer V (26% cortical thickness) and layer VI (22% cortical thickness) (Games and Winer, 1988) windowed using one box for each layer. Digital light units (DLU) were converted into fmol/mg protein using a standard curve generated from the co-exposed ¹⁴C-embedded plastic standards (American Radioactive Chemicals, St. Louis, MO, USA) (Pan et al., 1983).

3. Results

The impact of aging on hair cell loss and auditory thresholds (23dB parallel shift) across frequency have been previously described for the FBN rat aging model used in the present study (Turner and Caspary, 2005; Wang et al., 2009b).

3.1. Age-related GABA_AR subunit message changes

GABA_AR subunit message and protein levels were obtained from primary auditory cortex (AI) and parietal cortex (PtA) in sections from young, middle-aged and aged FBN rats. Automated, non-stereological collection of in situ hybridization data provided information regarding neuronal number, size, and area. Consistent with the visual cortical findings of Peters et al. (1983) no significant age-related changes in neuronal number, neuronal size, and neuronal area were observed across AI layers.

Table 1 summarizes results of age-related GABA_AR subunit message changes across AI layers for all subunits examined (α_1,2,3,5; β_1–3, γ_1, γ_2s, γ_2L). Age-related subunit message levels for young-adult, middle-aged, and aged rats are presented both as raw grain counts over neurons across AI layers II–VI and as percent change from young- adult for middle-aged, and aged AI (Table 1; p values less than 0.05, unless otherwise stated).

Table 1.

Changes of GABA_A Receptor Subunits mRNA levels in AI of FBN rats

		Young	Middle	% from Young	Aged	% from Young
α1	LII	5.9 ± 1.9	4.5 ± 2.2	−23.73*	3.6 ± 1.9	−38.98*
	LIII	5.4 ± 1.8	4.3 ± 2.2	−20.37*	3.2 ± 1.7	−40.74*
	LIV	5.9 ± 2.0	4.8 ± 2.3	−18.64*	4.0 ± 1.7	−32.20*
	LV	5.7 ± 1.9	4.6 ± 2.2	−19.30*	3.9 ± 1.7	−31.58*
	LVI	6.8 ± 2.2	5.4 ± 2.2	−20.59*	3.9 ± 2.0	−42.65*
α2	LII	3.8 ± 1.9	3.9 ± 1.9	2.63	3.6 ± 1.6	−5.26
	LIII	3.3 ± 1.4	3.8 ± 1.8	15.15*	3.1 ± 1.4	−6.06
	LIV	3.7 ± 1.6	3.9 ± 2.1	5.41	3.8 ± 2.0	2.70
	LV	3.4 ± 1.4	4.0 ± 1.7	17.65*	3.3 ± 1.4	−2.94
	LVI	4.1 ± 2.0	5.0 ± 1.8	21.95*	3.6 ± 1.5	−12.20*
α3	LII	6.5 ± 2.4	7.1 ± 3.2	9.23	8.5 ± 2.6	30.77*
	LIII	6.0 ± 1.8	6.8 ± 3.0	13.33*	7.8 ± 2.5	30.0*
	LIV	7.2 ± 2.5	7.6 ± 3.2	5.56	8.0 ± 2.6	11.11
	LV	6.7 ± 2.0	7.2 ± 2.9	7.46	7.4 ± 2.7	10.45*
	LVI	7.9 ± 2.6	7.6 ± 3.0	−3.80	7.7 ± 2.1	−2.53
α5	LII	7.2 ± 2.2	6.7 ± 2.4	−6.94	6.7 ± 2.4	−6.94
	LIII	6.0 ± 1.8	6.2 ± 2.5	3.33	6.0 ± 2.0	0
	LIV	6.0 ± 2.4	6.6 ± 2.7	10.0	6.2 ± 2.6	3.33
	LV	5.6 ± 2.0	5.6 ± 2.0	0	6.0 ± 2.2	7.14
	LVI	6.4 ± 2.3	6.3 ± 2.3	−1.56	6.7 ± 2.5	4.69
β1	LII	5.5 ± 2.1	5.4 ± 2.3	−1.82	4.3 ± 1.9	−21.82*
	LIII	5.2 ± 1.8	5.3 ± 2.6	1.92	3.9 ± 1.6	−25.0*
	LIV	5.4 ± 2.1	5.5 ± 2.2	1.85	4.6 ± 2.0	−14.81*
	LV	5.2 ± 2.0	5.7 ± 2.0	9.62*	4.5 ± 1.7	−13.46*
	LVI	6.2 ± 2.4	5.6 ± 2.2	−9.68*	4.9 ± 2.1	−20.97*
β2	LII	4.4 ± 2.0	4.3 ± 1.9	−2.27	3.1 ± 1.5	−29.55*
	LIII	4.1 ± 1.6	3.8 ± 1.7	−7.32	2.8 ± 1.6	−31.71*
	LIV	4.1 ± 1.9	4.2 ± 1.7	2.44	3.4 ± 1.6	−17.07*
	LV	4.4 ± 1.7	4.2 ± 1.7	−4.55	3.4 ± 1.4	−22.73*
	LVI	5.5 ± 2.0	5.0 ± 2.2	−9.09	4.3 ± 1.9	−21.82*
β3	LII	7.5 ± 2.8	8.0 ± 2.3	6.67	6.1 ± 2.5	−18.67*
	LIII	6.9 ± 2.5	7.3 ± 2.5	5.80	5.6 ± 2.5	−18.84*
	LIV	7.5 ± 2.7	7.6 ± 2.4	1.33	6.4 ± 2.6	−14.67*
	LV	7.1 ± 2.7	7.3 ± 1.7	2.82	6.6 ± 2.7	−7.04
	LVI	8.3 ± 2.8	8.2 ± 2.6	−1.20	7.9 ± 3.7	−4.82
γ1	LII	9.5 ± 3.3	8.6 ± 3.4	−9.47*	7.6 ± 2.7	−20.0*
	LIII	9.2 ± 3.3	8.4 ± 3.6	−8.70*	7.2 ± 2.1	−21.74*
	LIV	9.9 ± 3.4	8.6 ± 3.5	−13.13*	7.5 ± 2.4	−24.24*
	LV	9.9 ± 3.3	9.6 ± 3.1	−3.03	8.6 ± 2.1	−13.13*
	LVI	11.0 ± 3.3	10.3 ± 3.5	−6.36*	9.2 ± 2.5	−16.36*
γ2S	LII	6.0 ± 1.5	6.2 ± 1.8	3.33	5.9 ± 1.7	−1.67
	LIII	5.6 ± 1.5	5.8 ± 1.9	3.57	5.2 ± 1.6	−7.14
	LIV	5.7 ± 1.6	6.3 ± 2.1	10.53*	5.6 ± 1.5	−1.75
	LV	5.4 ± 1.2	5.9 ± 1.8	9.26*	5.7 ± 1.4	5.56
	LVI	5.8 ± 1.7	6.3 ± 2.0	8.62	6.0 ± 1.9	3.45
γ2L	LII	5.3 ± 1.9	4.2 ± 2.3	−20.75*	3.2 ± 1.7	−39.62*
	LIII	4.9 ± 1.6	4.1 ± 2.0	−16.33*	3.1 ± 1.9	−36.73*
	LIV	5.4 ± 1.6	4.2 ± 2.1	−22.22*	3.7 ± 2.0	−31.48*
	LV	5.1 ± 1.8	4.6 ± 2.0	−9.80*	3.9 ± 1.9	−23.53*
	LVI	5.7 ± 1.6	4.9 ± 2.1	−14.04*	4.4 ± 2.0	−22.81*

The data represent means ± SD (number of grains/100 μm ).

^*Significant difference between the means of young vs. aged groups and young vs. middle-aged (p<.05).

Aging and GABA_AR subunit α_1,2,3,5 message changes

Significant step-wise age-related decreases in GABA_AR α₁ subunit message were observed across layers of AI and PtA (Tables 1 and 2). Images from in situ hybridization show an age-related reduction in number of silver grains, representing GABA_AR α₁ subunit message over AI layer III neurons (Figs. 1A, B). The age-related α₁ subunit message loss was significant across all layers (p<0.01) for young vs. aged with percent reductions between 30 and 42 percent (Table 1, Fig. 2A). Middle-aged animals showed GABA_AR α₁ subunit message levels that were intermediate between young and aged α₁ subunit message levels (Table 1, Fig. 2A). Age-related changes in PtA GABA_AR α₁ subunit message displayed a similar staircase aging pattern to that seen in AI (Table 2, Fig. 2A).

Table 2.

Changes of GABA_A Receptor Subunits mRNA levels in PtA of FBN rats

		Young	Middle	% from Young	Aged	% from Young
α1	LII	7.1 ± 2.1	4.3 ± 2.3	−39.44*	3.7 ± 1.6	−47.89*
	LIII	7.0 ± 2.0	4.6 ± 2.0	−34.29*	3.8 ± 1.7	−45.71*
	LIV	7.7 ± 2.1	4.9 ± 2.3	−36.36*	3.9 ± 1.5	−49.35*
	LV	6.4 ± 1.7	4.9 ± 2.1	−23.44*	3.8 ± 1.5	−40.36*
	LVI	7.1 ± 2.1	5.1 ± 2.3	−28.17*	4.0 ± 1.4	−43.66*
α3	LII	6.8 ± 2.1	6.9 ± 2.5	1.47	7.3 ± 1.8	7.35
	LIII	6.9 ± 2.3	6.7 ± 2.4	−2.90	8.0 ± 2.1	15.94*
	LIV	6.7 ± 2.1	6.8 ± 2.3	1.49	7.5 ± 2.3	11.94*
	LV	7.0 ± 2.1	6.6 ± 2.4	−5.71	6.6 ± 1.9	−5.71
	LVI	7.6 ± 2.7	7.7 ± 2.7	1.32	7.3 ± 2.6	−3.95
β2	LII	5.1 ± 1.8	4.2 ± 1.8	−17.65*	3.7 ± 1.5	−27.45*
	LIII	4.5 ± 2.1	4.2 ± 1.7	−6.67	3.3 ± 1.4	−26.67*
	LIV	4.9 ± 1.8	4.5 ± 1.9	−8.16	4.1 ± 1.7	−16.33*
	LV	4.7 ± 1.8	3.9 ± 1.5	−17.02*	3.7 ± 1.5	−21.28*
	LVI	5.4 ± 1.9	4.8 ± 1.7	−11.11*	4.2 ± 1.6	−22.22*
γ2L	LII	4.6 ± 1.9	4.6 ± 1.7	−0	3.9 ± 1.7	−15.22*
	LIII	4.4 ± 1.8	4.3 ± 1.6	−2.27	4.0 ± 2.1	−9.09
	LIV	4.8 ± 1.7	4.5 ± 1.7	−6.25	4.2 ± 2.0	−12.5
	LV	4.5 ± 1.4	4.3 ± 1.8	−4.44	3.8 ± 2.0	−15.56*
	LVI	5.0 ± 1.7	5.1 ± 1.9	2.0	4.4 ± 2.0	−12.0

The data represent means±SD (number of grains/100 μm²).

^*Significant difference between the means of young vs. aged groups and young vs. middle-aged (Mid, p<.05).

Figure 2.

Age-related changes of GABA_AR subunit α₁, α₃ and β₂ message and proteins in FBN rat AI and PtA. Bar graphs represent the percentage changes of middle-aged (n = 6) and aged (n = 6) from the young (red x-axis, n = 6) for the message and protein levels of GABA_AR α₁ (A&B), α₃ (C&D) and β₂(E&F). Error bars represent the standard error of the means. (*: p<.05, Yng = Young).

While significant age-related declines in α₁ subunit message were observed across layers of AI and PtA, an apparent compensatory age-related increase in α₃ message level was observed for a subset of layers in AI and PtA (Tables 1 and 2; Fig. 2C). GABA_AR α₃ subunit message levels showed significant (p<.001) age-related increases over neurons in supragranular AI layers II–III, and output layer V with trends toward GABA_AR α₃ subunit message level increases in AI layer IV (Table 1, Fig. 2C). Aged AI layers II and III showed age-related α₃ subunit message level increases near 30% (Fig. 2C) when compared to young layers. Similar, but more modest pattern for GABA_AR α₃ subunit message increases were observed in layers III and IV of PtA (Table 2, Fig. 2C).

No consistent age-related changes were observed for α₂ or α₅ GABA_AR subunit message over neurons in AI layers II–VI (Table 1). However, GABA_AR α₂ subunit showed significant increases in middle-age, (AI layers II, V and VI) before returning to near young-adult levels in aged rat AI (Table 1).

Aging and GABA_AR β_1–3 subunit message changes

Significant age-related β_1–3 GABA_AR subunit message losses were seen across all AI layers (Table1, Fig. 2E). β_1–2 subunits showed changes between 17 and 34 percent across all layers of aged AI when compared to young animals (Table 1, Fig. 2E). AI β_1–2 subunit changes in middle-aged animals were generally not significantly different from young-adult levels, with the exception of β₁ subunit changes in AI layers V and VI (Table 1). GABA_AR β₃ subunit message showed significant age-related reductions in AI layers II–IV and non-significant changes in AI of middle aged animals (Table 1).

Aging and GABA_AR subunit γ₁, γ_2s and γ_2L message changes

Significant age-related decreases were seen for γ₁ and γ_2L GABA_AR subunit message levels while no age-related γ_2s subunit message changes were observed for aged rat AI (Table 1). Only γ_2L GABA_AR subunit message levels were examined in PtA. Age-related changes for γ_2L GABA_AR subunit message in PtA were smaller than those observed for AI and were significant only in LII and LV of aged PtA (Table2). Unfortunately, selective antibodies with adequate signal to noise ratios were not available to allow for quantitative immunohistochemistry of γ₁, γ_2s and γ_2L GABA_AR subunit proteins.

3.2. Aging and GABA_AR subunit protein changes: α_1&3

Densitometric immunohistochemical studies were used to assess GABA_AR subunit protein levels over individual neurons across the layers of AI and PtA for GABA_AR α subunits which showed significant age-related message changes. Age-related protein changes focused on the α_1&3 GABA_AR subunits, from cortical neurons in AI and PtA (Table 3). Confocal images showed age-related loss of α₁ GABA_AR subunit fluorescence (red) and the apparent compensatory age-related increase in α₃ subunit protein (green) (Fig. 3). Age-related protein changes were, for the most part smaller than, but consistent with α_1&3 subunit message changes (Fig. 2 & Table 1). Significant age-related reductions in α₁ GABA_AR subunit protein levels (10%–17%) were seen across AI layers reaching significance in superficial layers II and III (Fig. 2B, Table 3). Consistent with the observed increase of α₃ subunit message, there was an age-related up-regulation of α₃ GABA_AR subunit protein in AI. GABA_AR α₃ subunit increases ranged between 2% and 13% reaching significance for neurons in layers II–III and layer V (Fig. 2D, Table 3). Similar age-related α_1&3 subunit message and protein changes were observed in adjoining parietal cortex (Tables 2 and 3). Age-related increases in α₃ subunit protein in PtA exceeded changes observed in AI α₃ subunit protein. However, the general pattern of age-related α_1&3 subunit changes was similar across the two cortical areas.

Table 3.

Comparison of Age-related Changes of GABA_A Receptor Subunit Protein Levels in A1 and PtA

		Auditory Cortex (A1)					Parietal Cortex (PtA)
		Young	Middle	% from young	Aged	% from Young	Young	Middle	% from Young	Aged	% from Young
α1	LII	0.090±0.035	0.090±0.046	0	0.075±0.038	−16.67*	0.073±0.026	0.072±0.031	−1.37	0.065±0.036	−10.96
	LIII	0.089±0.034	0.095±0.045	6.74	0.074±0.035	−16.85*	0.071±0.027	0.070±0.032	−1.41	0.069±0.039	−2.82
	LIV	0.081±0.033	0.093±0.048	14.81	0.068±0.030	−16.05	0.071±0.027	0.068±0.029	−4.23	0.069±0.039	−2.82
	LV	0.079±0.031	0.087±0.045	10.13	0.071±0.035	−10.13	0.078±0.029	0.075±0.033	−3.85	0.061±0.040	−21.79*
	LVI	0.078±0.032	0.089±0.044	14.10	0.066±0.035	−15.38	0.075±0.029	0.075±0.036	0	0.061±0.034	−18.67*
α3	LII	0.136±0.034	0.141±0.030	3.68	0.152±0.039	11.76*	0.103±0.020	0.117±0.039	13.59*	0.110±0.030	6.80
	LIII	0.134±0.041	0.138±0.034	2.98	0.152±0.034	13.43*	0.097±0.025	0.122±0.035	25.77*	0.126±0.037	29.90*
	LIV	0.122±0.036	0.116±0.032	−4.92	0.125±0.031	2.46	0.093±0.025	0.118±0.030	26.88*	0.109±0.034	17.20*
	LV	0.125±0.036	0.131±0.036	4.8	0.136±0.031	8.8*	0.099±0.028	0.125±0.030	26.26*	0.114±0.024	15.15*
	LVI	0.096±0.027	0.113±0.028	17.71*	0.104±0.027	8.33	0.085±0.026	0.111±0.024	30.59*	0.095±0.021	11.76*
β1	LII	0.178±0.037	0.152±0.043	−14.61*	0.163±0.044	−8.43	0.104±0.043	0.125±0.055	20.19*	0.110±0.049	5.77
	LIII	0.159±0.042	0.139±0.037	−12.58*	0.157±0.033	−1.26	0.114±0.045	0.118±0.038	3.51	0.119±0.050	4.39
	LIV	0.141±0.039	0.117±0.031	−17.02*	0.138±0.037	−2.13	0.097±0.046	0.112±0.030	15.46	0.107±0.040	10.31
	LV	0.125±0.054	0.128±0.036	2.4	0.123±0.038	−1.6	0.093±0.041	0.120±0.036	29.03*	0.109±0.057	17.20
	LVI	0.118±0.041	0.098±0.024	−16.95*	0.117±0.042	−0.85	0.090±0.044	0.092±0.029	2.22	0.099±0.053	10.0
β2	LII	0.128±0.026	0.124±0.044	−3.13	0.105±0.034	−17.97*	0.080±0.041	0.089±0.027	11.25	0.095±0.041	18.75*
	LIII	0.122±0.024	0.112±0.038	−8.20	0.100±0.030	−18.03*	0.078±0.028	0.092±0.035	17.95*	0.093±0.035	19.23*
	LIV	0.114±0.024	0.096±0.035	−15.79*	0.093±0.031	−18.42*	0.067±0.021	0.086±0.025	28.36*	0.080±0.036	19.40*
	LV	0.107±0.026	0.103±0.028	−3.74	0.096±0.030	−10.28*	0.085±0.033	0.099±0.030	16.47*	0.085±0.038	0
	LVI	0.093±0.028	0.088±0.025	−5.38	0.085±0.026	−8.60*	0.083±0.023	0.093±0.027	12.05*	0.076±0.030	−8.4

Figure 3.

Confocal images of immuno double-labeling using GABAAR α₁ (red) and α₃ (green) in AI from young adult and aged FBN rat. A clearly increased α₃ immuno-positive staining is seen in the aged AI when compared to the young AI. Scale bar = 120μm.

Aging and GABA_AR β_1&2 subunit protein changes

Neuronal protein levels were obtained for β_1&2 GABA_AR subunits but not for the β₃ subunit due to the lack of availability of a specific antibody. Consistent with age-related declines in β₂ subunit message across AI layers, GABA_AR β₂ subunit protein levels declined significantly across AI layers (Table 3, Fig. 2F). Non-significant decreases were also observed for β₂ GABA_AR subunit protein in middle-aged AI (Table 3; Fig. 2F). GABA_AR β₂ subunit protein increases in PtA were in sharp contrast to what was observed for β₂ GABA_AR subunit protein in neighboring AI and in contrast to β ₂ GABA_AR subunit message level decreases observed for PtA and AI (Table 3). Significant increases for β₂ GABA_AR subunit protein were observed across most PtA layers for middle-aged and aged animals when compared to young-adult PtA (Table 3; Fig. 2F).

In contrast to age-related decreases for β₂ subunit protein, β₁ GABA_AR subunit protein decreased significantly only in middle-aged AI with no significant changes observed for aged AI compared to young-adult AI (Table 3). In addition, there were significant increases in β₁ subunit protein levels in LII and LV of middle-aged PtA but no significant changes in aged PtA (Table 3).

3.3. Age-related pharmacological changes of GABA_A receptors in AI

Groups of young (4–6 months), middle-aged (20–24 months ), and aged (30–34 months) FBN rats were used to further examine the impact of aging on intact mature GABA_ARs and the ability of GABA to modulate ligand binding at the picrotoxin binding site in the GABA_AR pore of AI neurons (Milbrandt, et al., 1996). Figure 4 shows higher levels of RO15-4513 binding in the superficial layers of AI. RO15-4513 is thought to be sensitive to the identity of α and γ GABA_AR subunits (Luddens and Wisden, 1991). A significant age-related loss of [³H]RO15-4513 GABA_AR binding sites (Bmax) was found across all layers of aged AI (p<0.017, n=8, 8, 8,) while kd values were unaltered (Table 4). In contrast to [³H]RO15-4513 binding, [³H]TBOB binding was highest in the deep layers of AI (Fig. 5). In this assay, in the absence of GABA, GABA_ARs are closed and no [³H]TBOB binding could occur in the chloride channel of either young or aged AI GABA_ARs (Fig. 5). With increasing concentrations of GABA (10nM-5μM), AI neuronal GABA_AR chloride channels were activated/opened providing access for [³H]TBOB binding at picrotoxin sites. At higher GABA concentrations, GABA_ARs became desensitized and the binding curve began to approximate baseline at the highest concentration of GABA (5μM). This cycle of events was significantly altered in aged AI. Figure 5 shows the age-related change in the [³H]TBOB binding curve between young-adult and aged AI layer VI as GABA levels were increased. The observed age-related changes in RO15-4513 binding and TBOB modulation supported subunit message and protein data which indicated an age-related change in the makeup and stoichiometry of GABA_ARs across the layers of aged AI.

Figure 4.

[³H]RO15-4513 GABA_AR binding in AI of young and aged FBN rats. A significant age-related loss of GABA_A receptor binding can be seen across all layers of AI at the concentrations of 5 and 8 nM of ³[H] RO15-4513. Highest binding levels were observed in the superficial layers of AI.

Table 4.

Bmax and kd of RO15-4513 Saturation Analysis

Layer	Bmax (fmol/mg protein)			Kd (nM)
	Young	Middle	Aged	Young	Middle	Aged
LII/LIII/LIV	618±44.7	550±25.3*	564±34.7*	1.03±0.40	0.80±0.22	0.99±0.33
LV	545±26.2	485±16.9*	510±26.9*	0.81±0.23	0.63±0.15	0.81±0.26
LVI	505±22.3	465±15.4*	472±21.8*	0.75±0.21	0.56±0.14	0.64±0.20

^*Significance from the young (p < 0.05).

Figure 5.

GABA modulation of ³[H]TBOB binding in layer VI of young and aged FBN rat AI. Increasing concentrations of GABA (10nM-5μM) were added to the TBOB assay. A significant age-related loss of TBOB binding was observed throughout aged neocortex when compared to young AI (the error bars represent S.E.M.). The age-related shift in the GABA dose-response curve in layer VI of FBN rat AI suggests a change in GABA’s ability to activate/open aged GABA_A receptors (n = 4 young and 4 aged).

4. Discussion

The present findings of age-related GABA_AR subunit and pharmacologic changes strongly support previous neurochemical, human psychophysical, and animal physiologic studies suggesting dysfunctional inhibitory processing of acoustic information in aged auditory cortex.

4.1. Age-related GABA_AR subunit changes and discordance between message and protein changes

The present findings in AI for GABA_AR subunit mRNA levels are consistent with cortical changes described by Gutierrez et al. (1997) showing substantial age-related changes for α1, β_2/3, γ₂ GABA_AR subunit messages in other brain areas. In addition, the present study found age-related loss of GABA_AR β₁γ₁ subunit message across AI layers (Table 1) Changes specific to the GABA_AR α₁, the wild-type α subunit message were across AI layers and greater than 30%. GABA_AR α₁ findings were consistent with previous studies of GABA_AR α₁ subunit message changes in aging neocortex, where Mhatre et al. (1992) described an 86% age-related decrease and Gutierrez et al. (1997) found a 29% reduction in the neocortex. In contrast to previous studies which did not observe significant age-related cortical subunit protein changes (Gutierrez, et al., 1996, 1997; Rissman, et al., 2007; Yu, et al., 2006), the present study finds α_1&3 GABA_AR subunit protein changes in AI consistent with observed age-related message changes. However, the present study did find substantial quantitative and qualitative discordance between GABA_AR subunit message changes and protein changes as previously noted by Gutierrez et al. (1997,1996), and Wang et al. (2009b). With the notable exception observed for GABA_AR β subunit changes, the present age-related protein findings were qualitatively consistent with, but more modest than, corresponding subunit message changes in both AI and PtA. Relatively smaller age-related percent changes for protein expression, compared to subunit message changes, may reflect post-translational compensatory mechanisms or perhaps a less sensitive method for assessing cellular protein levels relative to cellular message levels. Contrary to this latter possibility, there were examples of some age-related protein changes which exceeded age-related subunit message changes. One example found that age-related GABA_AR α₃ subunit protein changes in PtA were greater than corresponding GABA_AR α₃ message changes (Tables 1 and 3).

The most striking example of message/protein discordance with aging found significantly increased GABA_AR β_1&2 subunit proteins in PtA in the face of dramatically decreased β_1&2 subunit message levels (Figs. 2E and 2F; Tables 1 and 3). It is important to understand how aging affects both expression and post-translational processing of GABA_AR subunits. The presence of age-related discordance between subunit message and protein is emblematic of post-translational age-related changes. These findings may reflect robust compensatory post-translational aging mechanisms which will require further study. It is unlikely that these findings are due to experimental error since all measurements were blinded and age-related measures of β_1&2 subunit message and protein levels were carried out in different animals, while comparisons between AI and PtA were carried out in the same animals. Immunolabeling over neocortex was not observed to be uneven over PtA and AI, which are adjacent structures. Categorically, similar age-related changes between subunit message and protein have been described for glycine receptor subunits in the aging dorsal cochlear nucleus (Wang, et al., 2009b).

4. 2. GABA_AR α₁ and α₃ subunit protein changes

The present study examined a subset of GABA_AR subunit proteins partially limited by the availability of high quality subunit antibodies for certain GABA_AR subunits. As noted above, significant age-related GABA_AR subunit protein decreases occurred across layers in AI for α₁ & β₂ with many age-related protein changes approaching 20% (Fig. 2B and F, Table 3). PtA displayed similar GABA_AR subunit protein changes with aging with the GABA_AR β_1&2 subunit protein exceptions noted above. Perhaps as a compensatory change for the profound decrease in the GABA_AR α₁ subunit protein, GABA_AR α₃ subunit proteins tended to increase with age in both AI and PtA. These changes were significant for LII, III, and V in AI and all but LII in PtA. The mechanism for, and the significance of, these age-related compensatory subunit changes are unknown at the present time. A recent aging study in human visual cortex examined GABA_AR related protein changes and reported an age-related trend toward increased GABA_AR α₃ subunit protein between 20–80 years of age (Pinto et al., 2010). The age-related down-regulation of α₁ message and protein and the layer selective age-related up-regulation of α₃ message and protein are suggestive of a reverse of compensatory changes seen in development and other models of GABA deafferentation (Caspary et al., 2008). Caspary et al. (1990) found an age-related reduction in GABA release in inferior colliculus of aged F344 rats. As reviewed above, models of sensory aging are suggestive of altered GABA inhibitory neurotransmission. The present findings and similar studies of the aged inferior colliculus suggest that this loss of GABA tone is at least in part a result of plastic changes in GABA_AR subunit composition (Caspary et al., 2008). Developmental and expression GABA_AR subunit studies suggest that observed subunit changes are consistent with smaller peak evoked IPSCs having longer-slower time-constants, perhaps in an effort to compensate for the loss of inhibitory input (Bosman et al., 2002; Juttner et al., 2001; Wafford et al., 1993). Evidence for age-related compensatory GABA_AR subunit changes have been described for inferior colliculus (Caspary et al., 1999) and are implicit in a number of other studies (Rissman et al., 2007; Zhou et al., 2011).

4.3. Age-related receptor binding changes reflect altered GABA_A subunit content of functional receptors

Subunit composition/stoichiometry can dramatically affect receptor pharmacology and channel function (Angelotti and Macdonald, 1993; Caspary et al., 1999; Ducic et al., 1995; Macdonald and Olsen, 1994; Rudolph et al., 2001; Sigel et al., 1990; Wafford et al., 1993). Age-related changes in GABA_AR subunit constructs would impact the pharmacology of GABA_ARs (Ebert et al., 1994; Wafford et al., 1993). Age-related changes in GABA_AR binding were previously reported for non-auditory cortical structures and hippocampus (Concas et al., 1988; Erdo and Wolff, 1989; Mhatre and Ticku, 1992; Ruano et al., 1992). Previous receptor binding studies have shown age- related changes in GABA_AR pharmacology of inferior colliculus using several subunit selective radiolabeled GABA_AR ligands (Milbrandt et al., 1994,1996). RO15-4513 is thought to differentially bind GABA_AR constructs containing different α and γ GABA_AR subunits (Ebert et al., 1994; Wafford et al., 1993). The literature is not definitive on the binding properties of the BDZ inverse agonist RO15-4513 but strongly suggests a preference for binding constructs containing α₅>α₁ (Lingford-Hughes et al., 2002). The present study found significant RO15-4513 binding in the upper layers of AI in agreement with Pirker et al. (2000) description of moderate levels of α₅ subunit containing GABA_ARs in neocortical layers IV and high levels of α₁ GABA_ARs in supragranular layers of the neocortex. Data from the present study finds reduced RO15-4513 binding in aged AI compared to young-adult AI. This age-related change likely reflects the observed α_x and γ_x subunit changes or decreased numbers of functionally assembled and inserted GABA_ARs due to age-related changes in trafficking and or anchoring proteins (Wang et al., 2009a).

The present findings (Figure 5) are consistent with previous studies showing age-related loss in the ability of GABA to modulate binding at the picrotoxin site with age (Erdo and Wolff, 1989; Mhatre and Ticku, 1992; Milbrandt et al., 1996). TBOB selectively binds to convulsant sites associated with the chloride channel (Olsen et al., 1990). The differential ability of GABA to modulate the binding of picrotoxin analogs, such as TBOB and TBPS, to the picrotoxin site within the chloride channel of different GABA_A constructs were examined (Im et al., 1994). These authors found that maximal enhancement of TBPS binding by GABA (opening of channels to allow binding) in cloned rat GABA_AR subtypes varied with the isoforms (153 ± 10, 438 ± 16 and 139 ± 29% for α₁β₂, α₃β₂, α₆ β₂, respectively). The present binding study did not allow us to accurately discriminate individual AI layer changes but the highest levels of TBOB binding was found in infragranular layers of AI.

4.4. Age-related changes of GABA neurotransmission and inhibitory function in AI

An increasing number of studies in AI describe significant age-related losses of presynaptic markers for GABA and functional changes indicative of a loss of normal adult GABAergic function. Glutamic acid decarboxylase (GAD), the primary synthesizing enzyme for GABA, is significantly decreased in rat AI (Burianova et al., 2009; Ling et al., 2005) and parallels a significant decrease in the number and optical density of parvalbumin labeled neurons in rat AI (de Villers-Sidani et al., 2010; Martin Del Campo et al., 2012; Ouda et al., 2008). Human auditory cortex shows age-related decreased levels of markers for normal adult GABA function (McGeer and McGeer, 1976; Pinto et al., 2010). Functional loss of adult primate GABAergic function has been described in visual cortex (Betts et al., 2005; Leventhal et al., 2003; Schmolesky et al., 2000). Age-related changes suggestive of a loss of normal young adult GABAergic function in AI have been recently reviewed (Caspary et al., 2008; Mendelson and Rajan, 2011). Young and aged FBN rats show a fairly parallel (15–20dB) age-related threshold increase across frequencies (Caspary et al., 2005; Wang et al., 2009b). Recent cortical electrophysiology studies are strongly suggestive of an age-related loss of inhibitory function resulting in increased spontaneous and driven activity in the upper layers of rat and primate primary (AI) and secondary auditory cortex (Hughes et al., 2010; Juarez-Salinas et al., 2010). Age-related loss in the ability to localize sound in space in primate AI and secondary auditory cortex and a negative impact on novelty detection in rats can be directly related to inhibitory changes. We would have preferred if the upper layers showed greater age-related changes in subunit changes than the deeper layers since it appears that some of the greater changes occur in the layers with the highest levels of GABA_A receptors (Prieto et al., 1994). However, the apical dendrites extending up to LI–III have their cell bodies in lV and V which may well confound the relative distribution of age-related changes because of the somatic expression of the subunit markers show somatic expression although receptors may be in the supragranular layers. Behavioral and evoked potential studies are strongly suggestive of reduced temporal processing in aged rat central auditory pathway (Suta et al., 2011). Rat studies describe age-related losses in the ability to detect novel sounds in AI, which was partially reversed by training, resulting in up-regulated parvalbumin labeling in aged animals (de Villers-Sidani et al., 2010).

In support of the present observations, a recent study by Schmidt et al. (2010) found significant age-related decreases in paired-pulse inhibition in both auditory and parietal cortices. In contrast to the present findings and those of Gutierrez et al. (1997), this study described GABA_AR a₁ subunit protein increases with aging in parietal cortex for a subset of rats (Schmidt et al., 2010). Studies in the primate visual cortex described age-related changes in the visual receptive fields system (Juarez-Salinas et al., 2010; Leventhal et al., 2003) while Juarez-Salinas et al. (2010) described degraded spatial tuning in unit responses from AI and secondary auditory cortical areas. Collectively, the present findings are in agreement with the studies reviewed above, by showing significant age-related loss of wild-type (α₁β₂γ₂) GABA_AR markers commonly found in young-adult AI. In situ hybridization and immunohistochemistry data showed age-related changes of mRNA and protein within the individual GABA_AR subunits, while receptor binding study revealed the pharmacological changes when these subunit proteins assembled as functional GABA_A receptors. Loss of wild-type GABA_ARs and their replacement by GABA_AR constructs with different subunit combinations would be expected to show slower inhibitory response kinetics and lower peak currents impairing the ability to reliably process temporally demanding stimuli (Richardson et al., 2012; Wafford et al., 1993). It is likely that these age-related changes in normal GABA_A receptor function could impact speech understanding in a subset of the human elderly population.

Abbreviations

AI

auditory cortex

DAB

diaminobenzidine

dATP

deoxyadenosine triphosphate

GABA_AR

GABA_A receptor

GAD

GABA synthetic enzyme glutamic acid decarboxylase

IC

inferior Colliculus

PtA

parietal cortex

TBOB

t-butylbicycloorthobenzoate