Figure 1.
Interaction of rEag1 N- and C-termini with 14-3-3θ.
(A) Schematic representation of (top) the structural topology of the rEag1 channel and (bottom) the rEag1 GST-N207 and GST-C0 fusion proteins. (B) Yeast two-hybrid assay. cDNA encoding rEag1-N207 or C0 segment was fused to the coding sequence for LexA DNA binding domain and subcloned into the pGilda vector. cDNA for the B42 transcriptional activation domain alone (Empty) or in combination with 14-3-3θ was subcloned into the pJG4-5 vector. Yeasts co-transformed with the pGilda- and the pJG4-5-based plasmids were streaked on leucine-lacking plates. (C) GST pull-down assay of in vitro translated 14-3-3θ. Pull-down products were immunoblotted with the anti-14-3-3θ antibody. Indicated to the left are the molecular weight markers (in kDa). (D,E) Cell lysates prepared from HEK293T cells expressing myc-14-3-3θ were used for GST pull-down assay with GST or the fusion protein GST-N207/GST-C0. (Left panels) Coomassie blue staining of the GST proteins. (Right panels) Immunoblotting of pull-down products with the anti-myc antibody. Input volume was 5% of that of the cell lysates for pull-down.
Figure 2.
Isoform specificity of 14-3-3 binding with rEag1 N- and C-termini.
(A) GST pull-down assay of cell lysates from HEK293T cells transfected with various myc-tagged 14-3-3 isoforms. (Left panel) Coomassie blue staining of the GST proteins. (Right panel) Immunoblotting of pull-down products with the anti-myc antibody. Input volume shown at the bottom corresponds to 5% of the total cell lysates for pull-down. (B) Quantification of the pull-down efficiency of different 14-3-3 isoforms. The protein band intensities of individual myc-14-3-3 isoforms affinity precipitated by GST-N207 or GST-C0 in (A) were divided by those of cognate total inputs, thereby minimizing the potential bias conferred by the variation in protein expression among different 14-3-3 isoforms. Densitometric scans of immunoblots were obtained from three independent experiments.
Figure 3.
The contribution of PAS and CNBHD to rEag1 interaction with 14-3-3θ.
GST pull-down assays of rEag1 N-terminal and C-terminal GST fusion proteins containing specific structural domains. (Upper panels) Schematic representation of the rEag1 N-terminal (A) or C-terminal (B,C) GST fusion proteins. (Lower left panels) Coomassie blue staining of the GST proteins. (Lower right panels) Immunoblotting of pull-down products with the anti-myc antibody.
Figure 4.
Phosphorylation-independent interaction of rEag1 with 14-3-3θ.
(A) Co-immunoprecipitation of myc-14-3-3θ and rEag1 proteins. (Left panel) rEag1/rEag2 was co-expressed with an empty vector (−) or myc-tagged 14-3-3θ (+) in HEK293T cells. Cell lysates were immunoprecipitated (IP) by using the anti-myc antibody, followed by immunoblotting (WB) with the anti-myc or the anti-rEag1/rEag2 antibody. The protein bands corresponding to rEag1/rEag2 and 14-3-3θ are highlighted with arrow and arrowhead, respectively. (Right panel) Cell lysates from myc-14-3-3θ only or co-expression of rEag1 and myc-14-3-3θ were immunoprecipitated by using the anti-rEag1 antibody. Input volumes correspond to 5% of the total cell lysates used for immunoprecipitation. These co-immunoprecipitation data are representative of three to five independent experiments. (B) rEag1 was co-expressed with an empty vector or myc-tagged 14-3-3θ in HEK293T cells. 24 hrs after transfection, indicated cells were subject to 1-hr treatment with 1 µM okadaic acid or staurosporine. (Upper panel) Total cell lysates were immunoblotted with the anti-Akt (total Akt) or anti-phosphorylated Akt (pAkt) antibodies to monitor the cellular phosphorylation status. β-actin was run as a loading control. (Lower panel) Cell lysates were immunoprecipitated (IP) by using the anti-myc antibody, followed by immunoblotting (WB) with the anti-myc or the anti-rEag1 antibody. (C) Quantification of (upper panel) the Akt phosphorylation level (pAkt/Akt) and (lower panel) the co-immunoprecipitation (CO-IP) efficiency of 14-3-3θ and rEag1. The CO-IP efficiency was determined by the ratio of the protein band intensities of immunoprecipitated rEag1 to those of cognate total inputs. The mean values were subsequently normalized with respect to that of the no-treatment control of 14-3-3θ/rEag1 co-expression. Densitometric scans of immunoblots were obtained from three independent experiments. Asterisk denotes a significant difference from the no-treatment control of 14-3-3θ/rEag1 co-expression (*, t-test: p<0.05).
Figure 5.
Endogenous expression of 14-3-3θ and rEag1 in neurons.
(A) Co-immunoprecipitation of 14-3-3θ and rEag1. Detergent solubilized proteins from the lysates of rat forebrain were immunoprecipitated (IP) with the anti-14-3-3θ (upper panel) or the anti-rEag1 antibody (lower panel), followed by immunoblotting (WB) analyses with the anti-14-3-3θ or the anti-rEag1 antibody. The non-immune mouse or rabbit IgG was used in parallel as negative control. Input volumes correspond to 5% of the total cell lysates used for immunoprecipitation. The arrowhead and arrow refers to the protein bands of 14-3-3θ and rEag1, respectively. (B) Immunofluorescence staining of rEag1 (left panels) and 14-3-3θ (middle panels) in cultured hippocampal neurons. The area highlighted in the white boxes is viewed under a higher magnification (I, II). Arrows label the sites of co-localization of 14-3-3θ and rEag1 (right panels), which displayed significant punctuate patterns over a wide region along the neurites. Scale bar, 25 µm. These co-immunoprecipitation and immunofluorescence data are representative of four to seven independent experiments.
Figure 6.
Localization of 14-3-3θ and rEag1 in synaptosomal and PSD fractions.
(A) Subcellular fractionation separated rat brains into multiple fractions: homogenate (H), soluble fraction (S1), crude membrane fraction (P2), synaptosomal fraction (SPM), and two postsynaptic density (PSD) preparations (PSD I: one Triton X-100 wash; PSD II: two Triton X-100 washes), all of which were subject to immunoblotting analyses with the indicated antibodies. 25 µg and 5 µg refer to the amount of total protein loaded in each lane. (B) Quantitative analyses of protein abundance in different subcellular fractions. Densitometric scans of immunoblots were obtained from three to five independent experiments. Data were presented as normalized values with respect to cognate protein expression levels in the homogenate (H) fraction.
Figure 7.
The effect of 14-3-3θ over-expression on rEag1 K+ currents.
(A) (Left panel) Representative K+ currents recorded from HEK293T cells expressing rEag1 in the absence or presence of 14-3-3θ. HEK293T cells were co-transfected with the cDNAs for rEag1 and myc-vector or myc-14-3-3θ in the molar ratio of 1∶5. The holding potential was −90 mV. The pulse protocol comprised 300-ms depolarizing test pulses ranging from −90 to +50 mV, with 10-mV increments. (Right panel) Normalized mean K+ current density (at +40 mV) of rEag1 channels in the absence or presence of myc-14-3-3θ. The numbers in the parentheses refer to the number of cells analyzed, and the asterisk denotes significant difference from the rEag1 control (*, t-test: p<0.05). (B) (Left panel) Representative K+ currents recorded from oocytes expressing rEag1 in the absence or presence of 14-3-3θ. The molar ratio for cRNA co-injection was 1∶5 and 1∶10 for 14-3-3θ and Kvβ1, respectively. The pulse protocol was identical to that described in (A). (Right panel) Normalized mean K+ current density (at +40 mV) of rEag1 channels in the absence or presence of 14-3-3θ. (C) Biophysical properties of rEag1 channels in the absence (open circles) or presence (filled diamonds) of 14-3-3θ. The voltage-dependant curves for steady-state activation (upper left panel), activation kinetics (upper right panel), deactivation kinetics (lower left panel), and non-superimposable Cole-Moore shift (lower right panel) were analyzed as described previously [17]. Data were collected from recordings performed in oocytes.
Figure 8.
Reversal of the 14-3-3θ suppression of rEag1 K+ currents by the 14-3-3 antagonist difopein.
(A) GST pull-down assay of the cell lysates prepared from HEK293T cells over-expressing the YFP vector, YFP-difopein, or YFP-R18 mutant. Pull-down products were detected by immunoblotting with the anti-14-3-3θ antibody. Compared to the vector control (lane 1), introduction of difopein (lane 2) resulted in a 75% and 64% reduction in the amount of 14-3-3θ pull-down by the GST-C0 and GST-N207 fusion proteins, respectively. In contrast, no significant difference was observed in the presence of the inactive mutant control (lane 3). (B) Normalized mean K+ current density recorded from HEK293 cells stably expressing rEag1 channels. As indicated, these stable cell lines were subject to transient transfection with various cDNA constructs. The mean current density at +40 mV for each co-expression condition was normalized with respect to that of the co-expression of rEag1 and difopein. The numbers in the parentheses refer to the number of cells analyzed, and the asterisk denotes significant difference from the rEag1-difopein co-expression control (*, t-test: p<0.05).
Figure 9.
Lack of effect of 14-3-3θ over-expression on the total and surface expression of rEag1 protein.
(A) Representative result of surface biotinylation experiments. Intact HEK293T cells were biontylinated on ice and thereafter solubilized. (Surface) Cell lysates were pulled down with streptavidin agarose beads, followed by immunoblotting with the anti-rEag1 antibody. (Input) Cell lysates were directly employed for immunoblotting analyses. Input represents 5% of the total protein used for streptavidin pull-down. Also shown at the bottom are the corresponding β-actin expression levels for each lane. The specificity of the biotinylation procedure was verified by the absence of β-actin bands in the surface fraction. (B) Quantification of total and surface expression of rEag1 in the absence or presence of 14-3-3θ over-expression. The total protein density (top panel) was determined as the ratio of input signal to the cognate β-actin signal. The surface expression efficiency (bottom panel) was expressed as the ratio of surface signal to the corresponding total protein density. The mean values were subsequently normalized with respect to that of vector control. Densitometric scans of immunoblots were obtained from three independent experiments.
Figure 10.
Estimation of the single channel conductance of rEag1 channel under different co-expression paradigms.
rEag1 K+ channels were co-expressed with either the control vector, difopein, or 14-3-3θ in HEK293 cells. (A) Comparison of the mean current density (pA/pF) at +40 mV: rEag1+vector, 111.2±9.5; rEag1+difopein, 126.9±13.2; rEag1+14-3-3θ, 76.8±5.8. The numbers in the parentheses refer to the number of cells analyzed, and the asterisk denotes significant difference from the rEag1+vector control (*, t-test: p<0.05). (B) Representative non-stationary fluctuation analysis of rEag1 K+ currents. (Top row) Whole cell current traces were evoked by 300-ms depolarizations from −90 mV to +40 mV, which were used to generate the ensemble mean (second row) and variance (third row). (Bottom row) The mean-variance plot (during the 300-ms test pulse, with the linear capacitative component eliminated)(open circles) was fit with a parabolic function (solid curve) to estimate the single channel current (0.9 pA). (C) Box plot presentation of the distribution of the estimated single channel conductance. Mean values (pS): rEag1+vector, 7.5±0.9; rEag1+difopein, 7.0±2.3; rEag1+14-3-3θ, 8.0±1.1. No significant difference was found among the three co-expression conditions (One-way ANOVA: p>0.05).