Article Figures & Data
Figures
- Figure 4.
Concentration-response profiles for direct 3α5αP effects (n = 6) (A, B), potentiation of 0.5 μm GABA responses (n = 4) (C, D), and direct GABA responses (n = 4) (E, F). The shading of traces represents increasing concentrations of 3α5αP (A, C) or GABA (E). In C, the lightest trace represents the response to GABA alone, from which potentiation was calculated (GABA response is represented as zero, and potentiation was calculated as described in Materials and Methods). Peak amplitudes of responses were measured for summary plots. For all summary plots, the solid lines represent fits to the Hill equation, axn/(bn + xn), where a is the maximum response, b is the half-maximum concentration (EC50), x is the potentiator concentration, and n is the Hill coefficient. Fit parameters are given in the text for B and D. For GABA responses, the EC50 value was 9.1 μm, and the Hill coefficient was 1.8.
- Figure 7.
Cyclodextrin effects on 3α5αP potentiation at recombinant receptors. A, Example of the effect of γ-cyclodextrin (γ-CD) on 3α5αP potentiation of GABA responses. B, Summary of the effects of α-,β-, andγ-cyclodextrin on 3α5αP responses. Concentrations were fixed at 500 μm cyclodextrin and 0.5 μm 3α5αP (n = 4 cells). C, 3α5αP concentration-response relationship obtained in the presence and absence of 500 μm γ-cyclodextrin. GABA was constant at 2 μm (n = 4-8 at each concentration). Solid lines represent a fit to the Hill equation with the Hill coefficient constrained to ≤2. Note the lack of effect of γ-cyclodextrin at saturating 3α5αP concentrations (fits predicted a maximum potentiation of 9.2 and 10.6 for absence and presence of γ-cyclodextrin, respectively). The EC50 value for 3α5αP potentiation shifted from 1.1 to 9.2 μm. D, No effect of γ-cyclodextrin on responses to GABA alone (n = 6) or to potentiation by lorazepam (Lzp; 1 μm; n = 4) and pentobarbital (Ptb; 100 μm; n = 4) was observed.
- Figure 1.
3α5αP currents from solitary glutamatergic hippocampal islands. A, Example IPSC elicited from -50mV. The current is a result of subtracting one sweep in the presence of 25 μm bicuculline from a sweep in the absence of bicuculline. B, Example EPSC elicited from another solitary neuron from -50 mV. The subtraction was performed with 1 μm NBQX (2,3-Dioxo-6-nitro-1,2,3,4-tetrahydrobenzo[f]quinoxaline-7-sulfonoamide). The pipette solution for both recordings contained cesium methanesulfonate. C, Current gated from a holding potential of -20 mV by 300 nm 3α5αP in the presence and absence of bicuculline (25 μm). Bicuculline was present throughout the trace. Note that in this and subsequent figures, there is a slight lag between the switch in solutions, denoted by the top trace, and onset or offset of the measured current. Partly, this lagresults from a combination of exchange time in the deadspace of the perfuser tip and distance of the perfuser from the recorded cell. After accounting for this lag, liquid junction potential exchanges suggested an exchange time much faster (∼100 msec) than the rise and decay of the 3α5αP-gated currents.
- Figure 3.
A steroid antagonist completely blocks responses to 3α5αP alone. All responses are from the same cell. A, Response to 300 nm 3α5αP in a solitary neuron. B, Response to the coapplication of 300 nm 3α5αP with 10 μm 17PA. Note the complete loss of current. C, A 5 sec application of 0.5 μm GABA was followed with a 30 sec coapplication of GABA plus 300 nm 3α5αP. The dotted line denotes the amplitude of the GABA current at the end of the 5 sec pre-application of GABA. D, Same protocol as C, except that 10 μm 17PA was present during the coapplication phase of the experiment. Note that 17PA incompletely blocked the potentiation.
- Figure 5.
Comparison of GABA-gated and 3α5αP-gated currents. A, Comparison of responses to 5 sec applications of concentrations calculated to be 20% of maximum for GABA and 3α5αP. Note that the slower lag for drug washout compared with wash-in was a result of a slower rate of perfusion (and therefore exchange time) for the saline wash. The slower saline wash was used to maintain cell viability for longer periods of time. B, Comparison of responses in the same cell as in A to 60 sec applications of the same concentrations of agonists. C, D, Direct gating has slower onset than potentiation. Comparison of direct gating onset (C) with potentiation onset (D) in the same cell at the same concentration of 3α5αP (100 nm). In D, GABA was used at 0.5 μm. Solid lines represent single-exponential fits to the development of the steroid-induced current or current potentiation. In this example, the time constant of current development was 328 sec, and the time constant of potentiation development was 13 sec. E, Summary of onset time constants obtained from four neurons in which direct gating and potentiation were directly compared.
- Figure 2.
Incomplete block of 3α5αP-gated currents by gabazine. A, Example trace from an experiment using 3α5αP (300 nm) alone and plus gabazine (10 μm). B, Example trace from the same cell showing the effect of gabazine on GABA (0.5 μm) currents potentiated by 300 nm 3α5αP. Note the more complete block compared with the administration of 3α5αP alone. C, Summary of experiments comparing the degree of gabazine block of directly gated currents with potentiated currents as in A and B. n = 4 cells in which both potentiation and gating were measured. The difference between the degree of block was significant (p < 0.01) by paired t test.
- Figure 6.
No effect of pertussis toxin on responses of glutamatergic hippocampal neurons to 3α5αP. A, B, Control neuron, the synaptic depression induced by 0.1 μm 5′-N-ethylcarboxamidoadenosine (NECA), and the response of the same neuron to a 30 sec application of 300 nm 3α5αP. C, D, Neuron treated with pertussis toxin. Note the lack of NECA effect on EPSC but robust 3α5αP current. E, Summary of the effect of overnight pertussistoxin treatment on the NECA effect on EPSC amplitudes (n = 9 control and 9 pertussis-treated cells). Five control and four pertussis toxin-treated neurons were recorded using a K+-based pipette solution to account for the possibility that Cs+ loading occluded NECA-induced synaptic depression. There was no consistent difference among Cs+-loaded cells and K+-loaded cells, so results were pooled for the summary. F, Summary of pertussistoxin effects on amplitude of 3α5αP currents, measured at the end of a 30 sec pulse of 300 nm 3α5αP(n = 4 control neurons and 5 pertussis toxin-treated neurons).
- Figure 8.
γ-Cyclodextrin effects on direct gating offset kinetics in hippocampal neurons. A, Example showing speeding of the offset of 3α5αP (300nm) current in a hippocampal neuron by 500 μm γ-cyclodextrin. Note that although slow saline washes were used for some experiments (see Fig. 4), the solution exchange times for saline and cyclodextrin solutions in the present experiment were matched. B, Summary of results from four neurons treated as in A. Deactivation was fit with a single-exponential function, the average time constant of which is plotted (*p = 0.01). C-E, Comparison of brief γ-cyclodextrin and bicuculline effects on deactivation of directly gated currents. All data in C-E are from a single neuron. Protocols are as indicated. Concentrations of drugs were 300 nm 3α5αP, 500 μm γ-cyclodextrin, and 25 μm bicuculline. F, Control experiment for the speed of γ-cyclodextrin washout. The experimental protocol is as indicated, with GABA at 0.5 μm, 3α5αP at 0.1 μm, and γ-cyclodextrin at 500 μm. Note that 3α5αP was co-applied with GABA and γ-cyclodextrin was co-applied with GABA and 3α5αP where indicated. Note that the redevelopment of potentiated current after γ-cyclodextrin exposure was similar to initial onset of potentiation suggesting that γ-cyclodextrin washout is not rate limiting.
- Figure 9.
The effect of direct gating on neuronal excitability. A-D, Solitary glutamate neurons were treated as indicated. Cells were exposed to drugs for 40 sec before a 400 msec direct depolarizing current injection (300 pA for this cell; top traces) to elicit action potentials (bottom traces). The calibration bar in C applies to A-D. E, Summary of the number of action potentials elicited under the labeled conditions in five solitary neurons. *p < 0.05 significantly different from control and from steroid plus 17PA.
