Distinct physiological mechanisms underlie altered glycinergic synaptic transmission in the murine mutants spastic, spasmodic, and oscillator

Abstract

Spastic (spa), spasmodic (spd), and oscillator (ot) mice have naturally occurring glycine receptor (GlyR) mutations, which manifest as motor deficits and an exaggerated "startle response." Using whole-cell recording in hypoglossal motoneurons, we compared the physiological mechanisms by which each mutation alters GlyR function. Mean glycinergic miniature IPSC (mIPSC) amplitude and frequency were dramatically reduced (>50%) compared with controls for each mutant. mIPSC decay times were unchanged in spa/spa (4.5 +/- 0.3 vs 4.7 +/- 0.2 ms), reduced in spd/spd (2.7 +/- 0.2 vs 4.7 +/- 0.2 ms), and increased in ot/ot (12.3 +/- 1.2 vs 4.8 +/- 0.2 ms). Thus, in spastic, GlyRs are functionally normal but reduced in number, whereas in spasmodic, GlyR kinetics is faster. The oscillator mutation results in complete absence of alpha1-containing GlyRs; however, some non-alpha1-containing GlyRs persist at synapses. Fluctuation analysis of membrane current, induced by glycine application to outside-out patches, showed that mean single-channel conductance was increased in spa/spa (64.2 +/- 4.9 vs 36.1 +/- 1.4 pS), but unchanged in spd/spd (32.4 +/- 2.1 vs 35.3 +/- 2.1 pS). GlyR-mediated whole-cell currents in spa/spa exhibited increased picrotoxin sensitivity (27 vs 71% block for 100 microM), indicating alpha1 homomeric GlyR expression. The picrotoxin sensitivity of evoked glycinergic IPSCs and conductance of synaptic GlyRs, as determined by nonstationary variance analysis, were identical for spa/spa and controls. Together, these findings show the three mutations disrupt GlyR-mediated inhibition via different physiological mechanisms, and the spastic mutation results in "compensatory" alpha1 homomeric GlyRs at extrasynaptic loci.

Figure 1.

Methods used to quantify glycinergic mIPSCs in hypoglossal motoneurons. A, shows 9 s of pharmacologically isolated mIPSCs (holding potential, –70 mV) from a wild-type mouse, recorded in the presence of TTX (1 μm), CNQX (10 μm), and bicuculline (10 μm). B, The addition of strychnine (1 μm) completely abolished the large inward currents in A, confirming that the mIPSCs recorded under these conditions are mediated by GlyRs. C, Examples of captured glycinergic mIPSCs (from A; aligned to rising phase onset) showing their highly variable amplitudes. D, Plot of peak mIPSC amplitudes (650 events) during the course of an experiment (cell illustrated in A), showing the stability of the response. E, Relationship between 10 and 90% rise time and amplitude for mIPSCs shown in D (r² = 0.06). F, Averaged glycinergic mIPSC from the data in A (average of 100 events). The averaged mIPSC is best fit by a single decay time constant (4.5 ms; open circles). G, An electrically evoked IPSC (average of 10 responses; holding potential, –70 mV), recorded under the same conditions as in A, after stimulation (0.6 ms; 4 V; 0.1 Hz) with a bipolar electrode placed in the reticular formation. The evoked response is completely abolished by the addition of 1 μm strychnine. H, Voltage–current relationship for the glycinergic IPSC shown in G. The reversal potential was near 0 mV, the calculated Nernst reversal potential for equivalent chloride concentrations across the neuronal membrane.

Figure 2.

Properties of glycinergic mIPSCs in control (spa/+) and spa/spa mice. A, Three consecutive 5 s traces showing glycinergic mIPSCs recorded in a control mouse. Recordings were made in the presence of TTX (1 μm), CNQX (10 μm), and bicuculline (10 μm) at a holding potential of –70 mV. B, Glycinergic mIPSCs recorded in a spa/spa mouse under conditions identical to A. Note the dramatic reduction in mIPSC amplitude and frequency. C, Overlapping amplitude histograms for control (thin outline) and spa/spa (thick lines) constructed from 888 and 290 events, respectively. The inset shows aligned average mIPSCs for control (thin trace) and spa/spa (thick trace). D, Averaged mIPSCs from control and spa/spa mice (inset in C) scaled to the same peak amplitude. Note that the traces completely overlap.

Figure 3.

Properties of glycinergic mIPSCs in control (spd/+) and spd/spd mice. A, Three consecutive 5 s traces showing glycinergic mIPSCs recorded in a control mouse. Recordings were made in the presence of TTX (1 μm), CNQX (10 μm), and bicuculline (10 μm) at a holding potential of –70 mV. B, Glycinergic mIPSCs recorded in a spd/spd mouse under identical conditions to A. Note the reduction in mIPSC amplitude and frequency. C, Overlapping amplitude histograms for control (thin outline) and spd/spd (thick lines) constructed from 893 and 228 events, respectively. The inset shows aligned average mIPSCs for control (thin trace) and spd/spd (thick trace). D, Averaged mIPSCs from control and spd/spd mice (inset in C) scaled to the same peak amplitude. The mIPSC from spd/spd (heavy trace) has a significantly faster decay than the control mIPSC.

Figure 4.

Properties of glycinergic mIPSCs in control (ot/+) and ot/ot mice. A, Consecutive 5 s traces showing glycinergic mIPSCs recorded from a control mouse. Recordings were made in the presence of TTX (1 μm), CNQX (10 μm), and bicuculline (10 μm) at a holding potential of –70 mV. B, Glycinergic mIPSCs recorded in an ot/ot mouse under identical conditions to A. Note the dramatic decrease in mIPSC amplitude and frequency. Some small strychnine-sensitive mIPSCs can, however, still be resolved (asterisks). C, Overlapping mIPSC amplitude histograms for a control (thin outline) and ot/ot (thick lines) mouse constructed from 992 and 185 events, respectively. The inset shows aligned average mIPSCs for control (thin trace) and ot/ot (thick trace). D, Averaged mIPSCs from control and ot/ot mice (inset in C), overlapped and scaled to the same peak amplitude. The mIPSC from ot/ot (heavy trace) has a significantly slower decay than the control mIPSC.

Figure 5.

Noise analysis on excised patches from neurons in spastic and spasmodic mice. A, Membrane current recordings (256 ms epochs) obtained in outside-out patches excised from the somata of hypoglossal motoneurons in a control (spa/+) animal (holding potential, –50 mV). Note the clear increase in membrane noise (bottom vs top trace) after the addition of 25 μm glycine to the perfusate. B, Membrane current recorded in a patch from a spa/spa animal in ACSF and after the addition of glycine. Membrane noise increases after the addition of glycine. C, Current variance analysis for membrane currents recorded in patches from a spa/+ (gray circles) and spa/spa (black squares) animal. Variance was calculated over 256 ms epochs during the rising phase of the glycine-activated current as described in Materials and Methods. The solid lines are linear fits to the current variance data, and the slope of this relationship predicts mean single-channel conductance, which is greater in the spastic animal (96 vs 33 pS). D, Identical analysis to C for the spasmodic mutation. The slope of linear fits to the initial part of the current variance plot is similar in patches from spd/+ (gray circles) and spd/spd animals (black squares).

Figure 6.

Picrotoxin sensitivity of glycine-mediated whole-cell currents in the spastic mutant. A, Picrotoxin sensitivity (50 μm) of glycine-mediated whole-cell currents. Left three traces (black lines) show whole-cell currents recorded in a hypoglossal motoneuron (holding potential, –70 mV) from a control (spa/+) animal in the presence of TTX (1 μm), CNQX (10 μm), and bicuculline (10 μm). The traces show sequential responses (30 s intervals) to 3 s applications of glycine, glycine plus 50 μm picrotoxin, and then glycine. Results for a similar experiment to A in a spa/spa motoneuron are shown in next sequence of three traces (gray lines). Glycine-mediated whole-cell currents in spa/spa neurons are more sensitive to 50 μm picrotoxin. B, Sensitivity of glycine-mediated whole-cell currents to 100 μm picrotoxin. Experiments are presented in the same manner as in A. The picrotoxin sensitivity of glycine-mediated currents is again much greater in spa/spa versus control neurons.

Figure 7.

Picrotoxin sensitivity of evoked IPSCs in the spastic mutant. A, Evoked IPSCs (average of 10 records) recorded in ACSF, ACSF plus picrotoxin (50 μm), and after wash back to ACSF. Data are from a control (spa/+) mouse. B, Summary of the effect of picrotoxin (50 μm) on the normalized amplitude of evoked IPSCs in control and spa/spa neurons (n = 5 and 4, respectively). Neurons in control and spa/spa animals showed identical picrotoxin sensitivity. Error bars indicate SE. C, Mean current variance plots for mIPSCs from a control and spa/spa neuron. The mean single-channel currents for these neurons were calculated as 3.1 and 3.4 pA, respectively (for details, see Materials and Methods and Results). Mean single-channel current/conductance was not significantly different for controls versus spa/spa animals.