Assembly of GABAA receptors composed of alpha1 and beta2 subunits in both cultured neurons and fibroblasts

Abstract

GABAA receptors are believed to be pentameric hetero-oligomers, which can be constructed from six subunits (alpha, beta, gamma, delta, epsilon, and rho) with multiple members, generating a large potential for receptor heterogeneity. The mechanisms used by neurons to control the assembly of these receptors, however, remain unresolved. Using Semliki Forest virus expression we have analyzed the assembly of 9E10 epitope-tagged receptors comprising alpha1 and beta2 subunits in baby hamster kidney cells and cultured superior cervical ganglia neurons. Homomeric subunits were retained within the endoplasmic reticulum, whereas heteromeric receptors were able to access the cell surface in both cell types. Sucrose density gradient fractionation demonstrated that the homomeric subunits were incapable of oligomerization, exhibiting 5 S sedimentation coefficients. Pulse-chase analysis revealed that homomers were degraded, with half-lives of approximately 2 hr for both the alpha1((9E10)) and beta2((9E10)) subunits. Oligomerization of the alpha1((9E10)) and beta2((9E10)) subunits was evident, as demonstrated by the formation of a stable 9 S complex, but this process seemed inefficient. Interestingly the appearance of cell surface receptors was slow, lagging up to 6 hr after the formation of the 9 S receptor complex. Using metabolic labeling a ratio of alpha1((9E10)):beta2((9E10)) of 1:1 was found in this 9 S fraction. Together the results suggest that GABAA receptor assembly occurs by similar mechanisms in both cell types, with retention in the endoplasmic reticulum featuring as a major control mechanism to prevent unassembled receptor subunits accessing the cell surface.

Fig. 1.

Biochemical characterization of the GABA_A receptor α1^(9E10) and β2^(9E10) subunits produced on Semliki Forest virus infection of BHK cells. BHK cells were infected with either the α1^(9E10) (A) or β2^(9E10) (B) subunit virus. At defined periods (between 0.5 and 24 hr as indicated) after infection the cells were labeled for 30 min with 500 μCi/ml [³⁵S]methionine. Infected cells were then lysed, and labeled proteins were resolved by SDS-PAGE followed by fluorography. The subunits were also immunoprecipitated from infected cells labeled with [³⁵S]methionine and treated with (+) or without (−) tunicamycin (5 μg/ml), as shown in C and D for the α1^(9E10) and β2^(9E10)subunits, respectively. The migration of protein standards (Bio-Rad) is indicated.

Fig. 2.

Surface expression of GABA_Areceptor subunits in BHK cells. BHK cells were infected with either single α1^(9E10) or β2^(9E10)virus or co-infected with both, and expression was then determined by immunofluorescence with (+) and without (−) permeabilization 20 hr after infection. Images were then collected by confocal microscopy.A, α1^(9E10), 9E10 antibody (+);B, β2^(9E10), 9E10 antibody (+); C, α1^(9E10), 9E10 antibody (−); D, α1^(9E10)and β2^(9E10), 9E10 antibody (−). Images were collected from cells infected with both subunit viruses and co-stained using a rabbit polyclonal antibody against the α1 subunit and a monoclonal antibody (BD17) against the β2 subunit coupled to fluorescein and rhodamine secondary antibodies, respectively. The images collected for the two channels are shown in E andF for the α1^(9E10) and β2^(9E10) subunits, respectively.

Fig. 3.

Electrophysiological properties of GABA_A receptors produced in BHK cells. A, Whole-cell recordings form BHK cells infected with Semliki Forest viruses expressing α1^(9E10)β2^(9E10). Homomeric α1^(9E10) and β2^(9E10)subunits were insensitive to GABA, and β2^(9E10)subunits were also insensitive to pentobarbital (PB). The duration of the ligand application is indicated by solid lines. Holding potential, −50mV. B, GABA (•, □, and ▪) and pentobarbital (▵) equilibrium concentration responses for BHK cells expressing α1^(9E10)β2^(9E10) or single subunits. data were fitted byI/I_max = [1/(1 + {[A]/EC₅₀}ⁿ)], whereI and I_max represent GABA-activated and maximally activated current, n is the Hill coefficient, and A is the GABA concentration. EC₅₀ = 1.47 ± 0.12 μm;n = 1.0 ± 0.08. Data points indicate mean ± SEM.

Fig. 4.

Differential sedimentation of the α1^(9E10) subunit dependent on coexpression with the β2^(9E10) subunit. Cells infected with the α1^(9E10) (A) or co-infected with both the α1(9E10) and β2^(9E10)(B) subunit viruses were lysed and subjected to sucrose density gradient fractionation 16 hr after infection. Gradient fractions were separated by SDS-PAGE, and the α1^(9E10) subunit was detected via Western blotting using 9E10 antibody (A) or a rabbit polyclonal antibody against the α1 subunit (B). Sedimentation coefficients for the α1^(9E10)subunit were determined with reference to proteins with known sedimentation: BSA (4.3 S), aldolase (7.4 S), and catalase (11.2 S).

Fig. 5.

Pulse–chase analysis of GABA_Areceptor assembly. Cells co-infected with both subunit viruses were labeled with 100 μCi/ml [³⁵S]methionine 2 hr after infection for 1 hr and chased for 0 (A), 6 (B), or 20 (C) hr excess cold methionine. The cells were then lysed and subjected to sucrose density gradient fractionation. Gradient fractions were then immunoprecipitated with 9E10 antibody, and the α1^(9E10) and β2^(9E10)subunits were resolved by SDS-PAGE. Sedimentation coefficients for the α1^(9E10) and β2^(9E10)subunits were determined as described in Figure 4.

Fig. 6.

Quantification of receptor assembly. The levels of incorporated methionine in α1^(9E10)(○) and β2^(9E10) (•) subunits were quantified in gradient fractions for cells chased for 0 (A), 6 (B), and 20 (C) hr using a Bio-Rad phosphorimager. Background was subtracted using the same volume that was used to integrate the subunit signals.

Fig. 7.

Degradation of single GABA_A receptor subunits expressed in BHK cells. BHK cells were infected with Semliki Forest viruses producing either the α1^(9E10)(A) or β2^(9E10)(B) subunit. Two hours after infection the cells were labeled with [³⁵S]methionine (400 μCi/ml) for 20 min and then chased for differing periods. The cells were then lysed, and receptor subunits were isolated by immunoprecipitation with 9E10 antibody. Precipitated material was then subjected to SDS-PAGE, and the subunit levels were quantified using a Bio-Rad phosphorimager.

Fig. 8.

Time dependence of surface expression of receptors composed of α1^(9E10) and β2^(9E10) subunits. Co-infected cells were fixed at differing periods after infection and processed simultaneously for fluorescence with 9E10 antibody without permeabilization. Images were collected from cells, and the average intensity of staining is shown. Data were collected from at least five cells at each time point.

Fig. 9.

Subunit ratio of GABA_A receptors composed of α1^(9E10) and β2^(9E10) subunits. A, BHK cells expressing both receptor subunits were pulse-labeled with [³⁵S]methionine (200 μCi/ml) for 2 hr; 3 hr after infection, the cells were then chased for 12 hr with excess cold methionine. Cell lysates were then subjected to sucrose density gradient fractionation, and the fractions corresponding to the 9 S peak were pooled and immunoprecipitated with 9E10 antibody. Receptor subunits were then resolved by SDS-PAGE and quantified using a Bio-Rad phosphorimager. After correction for methionine content (α1 = 9; β2 = 15) a ratio of 0.9 was found for the α1^(9E10):β2^(9E10) subunits in the experiment shown. B, Labeled cells were exposed to 9E10 antibody at 4°C for 30 min. The cell surface receptor population was then isolated via immunoprecipitation with protein G in the presence of excess 9E10 peptide and resolved by SDS-PAGE, and subunit levels were quantified. A ratio of 1.1 was found for the α1^(9E10):β2^(9E10) subunits in the experiment shown.

Fig. 10.

Infection of cultured SCG neurons with Semliki Forest viruses expressing GABA_A receptor subunits. Cultured SCG neurons were infected after 3 d in culture, and receptor expression was determined by immunofluorescence using 9E10 antibody 12 hr after infection with (+) or without (−) permeabilization. Images were then collected using confocal microscopy from cells infected with A, α1^(9E10) (+); B, β2^(9E10) (+); C, α1^(9E10) (−); and D, α1^(9E10) and β2^(9E10) (−).