|
Figure 1. Dose response curves for ATP inhibition reveal no effect of tandem linkage on ATP sensitivity and suggest coassembly of S1-WT with S2-ND. (A) Fractional currents remaining in the presence of various concentrations of cytosolic ATP for channels formed by SUR1+Kir6.2, S1-WT, S1-ND (black circles, triangles, and diamonds), SUR2A+Kir6.2, S2-WT and S2-ND (blue circles, triangles, and diamonds). The lines are fits to the Hill equation of the data obtained for the different channel groups. The values of Ki and the Hill coefficient (n) are listed. The Ki and n of SUR1+Kir6.2 and S1-WT, or of SUR2A+Kir6.2 and S2-WT, are indistinguishable. (B) Fractional currents remaining in the presence of various concentrations of cytosolic ATP in patches excised from oocytes injected with an equal molar amount of S1-WT and S2-ND cRNA (red symbols). The data and fit lines for homomeric S1-WT (black) and S2-ND (blue) channels are replotted from A for comparison. The red solid line is the best fit to the S1-WT+S2-ND coexpression data assuming that S1-WT and S2-ND do not coassemble; i.e., a mixture of the two Hill functions obtained for homomeric S1-WT (black dashed line) and S2-ND (blue dotted line) was fitted by leaving the fractional amplitude (a1) of the S1-WT component as the sole free parameter. This fit returned a1 = 0.68 ± 0.14, but clearly failed to describe the data. The red dotted line is the fit to the same data with a1 fixed to 0.5.
|
|
Figure 2. Coassembly with S1-WT reduces the basal whole-oocyte current and increases the glibenclamide-sensitive current of S2-ND. (A) Time courses of representative whole-cell currents recorded from oocytes expressing S1-WT (0.5 ng; black trace), S2-ND (1.5 ng; blue trace), and S1-WT+S2-ND (0.5 ng and 1.5 ng, respectively; red trace). Currents were recorded at −80 mV by TEVC (see Materials and methods). 3 mM sodium azide (Az), 340 μM diazoxide (Dzx), 10 μM glibenclamide (Gb), and 3 mM BaCl2 were applied to the bath solution as indicated by the labeled horizontal bars. Approximately 1-min stationary sections (−//−) were omitted from the S1-WT (black) and S2-ND (blue) traces to align the timing of all drug applications for all three traces. The red dotted line is the sum of the black and blue traces and illustrates the whole-cell current expected for an oocyte coexpressing S1-WT and S2-ND assuming that the two subunits do not coassemble. The inset shows the entire S1-WT current trace at an expanded current scale. (B) Bar chart summary of the mean basal (IBasal), azide-activated (IAz), diazoxide-activated (IDzx), and glibenclamide-sensitive (IGb) current components, as well as of the total currents (ITotal) in oocytes expressing S1-WT, S2-ND, and S1-WT+S2-ND.
|
|
Figure 3. Dissection of the voltage-dependent and voltage-independent components of spermine block for S1-WT channels. (A) Family of normalized, leak-subtracted ramp I/V curves obtained in a patch containing S1-WT channels in the absence of (black line), or in the presence of 5 (red line), 20 (cyan line), 100 (blue line), and 500 μM (green line) spermine. (B) Family of grel/V curves obtained from the I/V curves shown in A (solid symbols, color coding as in A). Dotted and solid lines represent fits to the individual grel/V curves of simple Boltzmann functions without (dotted lines) or with (solid lines) an added constant “leak.” (C–F) Plots of the fit parameters Vh (C), Vs (D), and leak (F), obtained by fitting individual grel/V curves with Eq. 2, and of the ratio −Vh/Vs (E), as a function of spermine concentration. Color coding is the same as in A. The black line in E was obtained by linear regression, that in F by a nonlinear least-squares fit to Eq. 7. Obtained parameters are presented in the panels.
|
|
Figure 4. Spermine block of homotetrameric S1-WT, S1-ND, and S2-ND channels characterized by global fitting. (A, C, and E) Families of normalized ramp I/V curves obtained in patches containing S1-WT (A), S1-ND (C), and S2-ND (E) channels; in the absence of (black lines), or in the presence of 5 (red lines), 20 (cyan lines), 100 (blue lines), and 500 μM (green lines) spermine. Except for S2-ND, the ramp I/V curves were leak subtracted. (B, D, and F) Families of grel/V curves obtained from the I/V curves shown in (A, C, and E) (solid symbols, color coding as in A, C, and E). Families of solid black lines represent global fits to the ensembles of four grel/V curves. Obtained fit parameters were K0 = 3.19 mM, ρ = 0.67, nH = 0.86, Mleak = 0.67, Kleak = 121 μM for S1-WT (B); K0 = 2.31 μM, ρ = 3.34, nH = 1.13 for S1-ND (D); and K0 = 4.69 μM, ρ = 3.05, nH = 1.26 for S2-ND (F).
|
|
Figure 5. Spermine block of channels resulting from coexpression of a 1:1 molar ratio of S1-WT and S1-ND subunits reveals random coassembly. (A) Family of normalized, leak-subtracted ramp I/V curves, obtained in a patch from an oocyte injected with a 1:1 molar ratio of S1-WT+S1-ND cRNA; in the absence of (black line), or in the presence of 5 (red line), 20 (cyan line), 100 (blue line), and 1,000 μM (pink line) spermine. (B) Family of grel/V curves obtained from the I/V curves shown in A (solid symbols, color coding as in A). The family of solid black lines represents a 10-free parameter global fit to the ensemble of the four grel/V curves, assuming random coassembly. Obtained fit parameters were p = 0.46, K01 = 7.02 μM, ρ1 = 2.83, nH1 = 0.97, K02 = 247 μM, ρ2 = 2.47, nH2 = 0.77, K03 = 2.30 mM, ρ3 = 0.87, nH3 = 0.77. The parameters for components 0 and 4 were fixed to the values shown in Table I. The family of dotted black lines represents the best global fit assuming no coassembly between S1-WT and S1-ND subunits. (C) Visual display of average fit parameters for components i = 0…4. Numbers in the top panel identify components.
|
|
Figure 6. Spermine block of channels resulting from coexpression of a 1:3 or 3:1 molar ratio of S1-WT and S1-ND subunits verifies the global fitting procedure. (A and C) Families of normalized, leak-subtracted ramp I/V curves obtained in patches from oocytes injected with a 1:3 (A) or 3:1 (C) molar ratio of S1-WT+S1-ND cRNA; in the absence of (black lines) or in the presence of 5 (red lines), 20 (cyan lines), 100 (blue lines), and 500 μM (green lines) spermine. (B and D) Families of grel/V curves obtained from the I/V curves shown in A and C (solid symbols, color coding as in A and C). Families of solid black lines represent global fits to the ensembles of four grel/V curves assuming random coassembly, but allowing for only a single free parameter (p). All other parameters were fixed to the values shown in Table I. Obtained fit parameters were p = 0.28 in B and p = 0.76 in D.
|
|
Figure 7. Spermine block of channels resulting from coexpression of a 1:1 molar ratio of S1-WT and S2-ND subunits reveals near-random coassembly. (A) Family of normalized, leak-subtracted ramp I/V curves, obtained in a patch from an oocyte injected with a 1:1 molar ratio of S1-WT+S2-ND cRNA; in the absence of (black line), or in the presence of 5 (red line), 20 (cyan line), 100 (blue line), and 500 μM (green line) spermine. (B) Family of grel/V curves obtained from the I/V curves shown in (A) (solid symbols, color coding as in A). The family of solid black lines represents a 13-free parameter global fit to the ensemble of the four grel/V curves, allowing for nonrandom coassembly. Obtained fit parameters were a1 = 0.306, a2 = 0.261, a3 = 0.244, a4 = 0.188, K01 = 172 μM, ρ1 = 2.84, nH1 = 0.95, K02 = 2.49 mM, ρ2 = 1.28, nH2 = 0.95, K03 = 2.82 mM, ρ3 = 0.78, nH3 = 0.95. The parameters for components 0 and 4 were fixed to the values shown in Table I. The family of dotted black lines represents the best global fit assuming no coassembly between S1-WT and S2-ND subunits. (C) Visual display of average fit parameters for components i = 0…4. Numbers in the top panel indicate the identities of the components. (D) Average fractional amplitudes ai (top) and normalized fractional amplitudes (bottom) obtained for components i = 0…4. For normalization, a p value was first calculated using Eq. 10; each ai value was then divided by the fractional amplitude expected for a binomial distribution generated by that particular p (Eq. 9).
|
|
Figure 8. Spermine block of channels resulting from coexpression of a 1:3 or 3:1 molar ratio of S1-WT and S2-ND subunits verifies the global fitting procedure. (A and D) Families of normalized, leak-subtracted ramp I/V curves obtained in patches from oocytes injected with a 1:3 (A) or 3:1 (D) molar ratio of S1-WT+S2-ND cRNA; in the absence of (black lines), or in the presence of 5 (red lines), 20 (cyan lines), 100 (blue lines), and 500 μM (green lines) spermine. (B and E) Families of grel/V curves obtained from the I/V curves shown in A and D (solid symbols, color coding as in A and C). Families of solid black lines represent global fits to the ensembles of four grel/V curves allowing for only four free parameters (a1…a4). All other parameters were fixed to the values shown in Table I. Obtained fit parameters were a1 = 0.590, a2 = 0.022, a3 = 0.065, a4 = 0.003 in B; and a1 = 0.091, a2 = 0.170, a3 = 0.359, a4 = 0.363 in E. The p values calculated from ai using Eq. 10 were p = 0.21 in B, and p = 0.74 in E.
|
|
Figure 9. Dose responses for ATP inhibition of channels obtained by coexpression of free SUR1, SUR2A, and Kir6.2 suggest the coassembly of free SUR1 with free SUR2A. (A) Representative current traces obtained in inside-out patches from oocytes expressing SUR1+Kir6.2 (black trace), SUR2A+Kir6.2 (blue trace), or SUR1+SUR2A+Kir6.2 (SUR1:SUR2A=1:1; red trace). The membrane was held at −100 mV and ATP at various concentrations was applied to the bath solution as indicated. (B) Fractional currents remaining in the presence of various concentrations of cytosolic ATP in patches excised from oocytes expressing SUR1+Kir6.2 (black symbols), SUR2A+Kir6.2 (blue symbols), or SUR1+SUR2A+ Kir6.2 (red symbols). Black and blue solid lines show fits to the Hill equation of the data for SUR1+Kir6.2 and SUR2A+Kir6.2, respectively. Fit parameters are listed Fig. 1 A. Red dotted and solid lines illustrate predicted curves for SUR1+SUR2A+Kir6.2 assuming that SUR1 does not coassemble with SUR2A. The red dotted line is the simple arithmetic average of the two Hill functions obtained for SUR1+Kir6.2 (black line) and SUR2A+Kir6.2 (blue line), expected if SUR1 and SUR2A protein was expressed at an equal molar ratio but did not coassemble. The red solid line is the best fit to the SUR1+SUR2A+Kir6.2 data by a mixture of the two parent Hill functions, with the fractional amplitude of channels formed from SUR1+Kir6.2, a1, left as a free parameter. This fit returned a1 = 0.69 ± 0.06, but still failed to describe the data.
|
|
Figure 10. Dose responses for tolbutamide inhibition of channels obtained by coexpression of free SUR1, SUR2A, and Kir6.2 suggest the coassembly of free SUR1 with free SUR2A. (A) Representative current traces obtained in inside-out patches from oocytes expressing SUR1+Kir6.2 (black trace), SUR2A+Kir6.2 (blue trace), or SUR1+SUR2A+Kir6.2 (SUR1:SUR2A=1:1; red trace). The membrane was held at −100 mV and tolbutamide at various concentrations was applied to the bath solution as indicated. The green line indicates the application of 2.5 mM ATP. The biphasic current increase during the washout of ATP is a perfusion artifact. (B) Fractional currents remaining in the presence of various concentrations of cytosolic tolbutamide in patches excised from oocytes expressing SUR1+Kir6.2 (black symbols), SUR2A+Kir6.2 (blue symbols), or SUR1+SUR2A+Kir6.2 (red symbols). The solid black line is a fit of Eq. 1 to the SUR1+Kir6.2 data, yielding fit parameters Ki1 = 14 ± 4 μM, h1 = 0.83 ± 0.15, Ki2 = 2.4 ± 0.2 mM, h2 = 2.8 ± 0.9, and L=0.51 ± 0.04. The blue solid line is a fit to the Hill equation of the data for SUR2A+Kir6.2, yielding fit parameters Ki = 7.19 ± 0.06 mM, n = 1.74 ± 0.02. Red dotted and solid lines illustrate predicted curves for SUR1+SUR2A+Kir6.2 assuming that SUR1 does not coassemble with SUR2A. The red dotted line is the simple arithmetic average of the two fit functions obtained for SUR1+Kir6.2 (black line) and SUR2A+Kir6.2 (blue line), expected if SUR1 and SUR2A protein was produced at an equal molar ratio but did not coassemble. The red solid line is the best fit to the SUR1+SUR2A+Kir6.2 data by a mixture of the two parent functions, with the fractional amplitude of channels formed from SUR1+Kir6.2, a1, left as a free parameter. This fit returned a1 = 0.81 ± 0.07, but still failed to describe the data.
|
