Rapid target-specific remodeling of fast-spiking inhibitory circuits after loss of dopamine

Abstract

In Parkinson's disease (PD), dopamine depletion alters neuronal activity in the direct and indirect pathways and leads to increased synchrony in the basal ganglia network. However, the origins of these changes remain elusive. Because GABAergic interneurons regulate activity of projection neurons and promote neuronal synchrony, we recorded from pairs of striatal fast-spiking (FS) interneurons and direct- or indirect-pathway MSNs after dopamine depletion with 6-OHDA. Synaptic properties of FS-MSN connections remained similar, yet within 3 days of dopamine depletion, individual FS cells doubled their connectivity to indirect-pathway MSNs, whereas connections to direct-pathway MSNs remained unchanged. A model of the striatal microcircuit revealed that such increases in FS innervation were effective at enhancing synchrony within targeted cell populations. These data suggest that after dopamine depletion, rapid target-specific microcircuit organization in the striatum may lead to increased synchrony of indirect-pathway MSNs that contributes to pathological network oscillations and motor symptoms of PD.

Figure 1. Changes in FS-MSNs connections after dopamine depletion

A. Probabilities of finding synaptic connections between presynaptic FS interneurons and postsynaptic D1 and D2 MSNs in striatum of saline- or 6-OHDA-injected mice. The number of connections per number of attempts is listed for each condition. In saline-injected mice, FS interneurons were more likely to be connected to D1 MSNs compared to D2 MSNs (p = 0.05) but in 6-OHDA-injected mice, FS interneurons were more likely to be connected to D2 MSNs compared to D1 MSNs (p = 0.04). *p = 0.05, **p = 0.04, ***p = 0.004. B. Probabilities of finding synaptic connections for FS-D1 MSN pairs either 3 days or 1 week after surgeries. Connection probability was not significantly different at either time point. C. Same as (B) but for FS-D2 MSN pairs. At both 3 days and 1 week after 6-OHDA injection, the probabilities of finding synaptic connections onto D2 MSNs were significantly increased compared to those measured in saline-injected mice (*p = 0.04 at 3 days and 0.009 at 1 week). D. Example of a paired recording where action potentials in a presynaptic FS interneuron elicit uIPSCs in a postsynaptic MSN. Individual trials (n = 10) are shown in grey with the average overlaid in black. The example is a recording from a 6-OHDA-injected mouse and the postsynaptic cell is a D1 MSN. E. Amplitudes of uIPSCs recorded in D1 and D2 MSNs in saline- and 6-OHDA-injected mice. There was no significant difference in uIPSC amplitudes across conditions. Circles represent data from individual cells and bars are the average for each population (saline_D1 = 267 ± 273 pA; saline_D2 = 328 ± 406 pA; 6-OHDA_D1 = 322 ± 470 pA; 6-OHDA_D2 = 566 ± 781 pA). F. Example of a paired recording in which a train of 10 action potentials (20 Hz) in a presynaptic FS interneuron elicits uIPSCs in a postsynaptic MSN. The example is a recording from a 6-OHDA injected mouse and the postsynaptic cell is a D2 MSN. G. Synaptic depression of FS-MSN synapses measured at 10, 20, 50, and 100 Hz in saline- and 6-OHDA-injected mice. Synaptic depression was calculated as the amplitude of uIPSC_8–10 / uIPSC₁. Error bars are sem. Synaptic depression was similar onto D1 (n = 9, saline; n = 9, 6-OHDA) and D2 MSNs (n = 8 saline; n = 8, 6-OHDA)) and was not changed by dopamine depletion.

Figure 2. FS-MSN synapses are not silenced by tonic dopamine levels in the slice

A. Bar graph comparing the probability of finding synaptic connections between presynaptic FS interneurons and postsynaptic D1 or D2 MSNs in control slices vs. slices incubated in a cocktail of D1 and D2 dopamine receptor antagonists (5 µM SCH23390 and 10 µM sulpiride, respectively). B. Amplitudes of uIPSCs recorded in D1 and D2 MSNs in control slices and from slices perfused with dopamine receptor antagonists. There was no significant difference in uIPSC amplitudes across conditions. Circles represent data from individual cells and bars are the average for each populations. C. Example of a paired recording in which a train of 10 action potentials (20 Hz) in a presynaptic FS interneuron elicits uIPSCs in a postsynaptic MSN. The example is from a recording done in the presence of dopamine receptor antagonists. The postsynaptic cell is a D1 MSN. D. Synaptic depression of FS-MSN synapses measured at 10, 20, 50, and 100 Hz in control slices and slices perfused with dopamine receptor antagonists. Synaptic depression was calculated as the amplitude of uIPSC_8–10 / uIPSC₁. Error bars are sem. Synaptic depression was similar onto D1 and D2 MSNs and was not changed by application of dopamine receptor antagonists.

Figure 3. Anatomical changes in FS interneurons after dopamine depletion

A. Brightfield image of a biocytin-filled FS interneuron from a saline-injected mouse. B–C. Higher magnification images of axons and dendrites for the boxed regions ’1’ and ‘2’ in A. Axons were distinguished from dendrites based on their narrower diameter and beaded appearance. D. Reconstructions of biocytin-filled FS interneurons from the striatum of saline-injected (left, same cell as in A) or 6-OHDA-injected (right) mice. Axons are grey and dendrites are black. E. Example of a reconstructed neuron (from a saline-injected mouse) within the radial grid used for a Sholl analysis. For display purposes, every other circle is omitted, so radii increase in 25 µm intervals. For the actual analysis, radii increased in 12.5 µm intervals. F. Bar graph showing significant increase (*p = 0.04) in total axon length of FS interneurons in 6-OHDA-injected mice relative to saline-injected mice. G. Bar graph showing no significant change in total dendrite length of FS interneurons in 6-OHDA-injected mice relative to saline-injected mice. H. Bar graph displaying the maximum radius at which axon crossings were detected. The maximum radius (a measure of axonal area) was not significantly different between saline- and 6-OHDA-injected mice. I. Graph showing the average number of grid crossings as a function of distance from the soma by axons of FS interneurons in saline-injected and 6-OHDA-injected mice. The number of crossings was significantly greater in 6-OHDA-injected mice compared to saline-injected mice (*p = 0.04).

Figure 4. Target-specific increase in inhibitory terminals from FS interneuron after dopamine depletion

A–B. Confocal images showing immunostains against the inhibitory presynaptic marker vGAT, and the FS interneuron marker, PV in sections from saline (A) and 6-OHDA-injected mice (B). All images were pseudocolored to better visualize vGAT/PV colocalization. Scale bars are 5 µm. C. Bar graph showing the percentage of vGAT-labeled pixels that were also labeled by PV. The colocalization of these two markers was significantly increased in 6-OHDA-injected mice relative to saline-injected mice, suggesting an increase in inhibitory terminals from FS interneurons. *p < 0.0001. D–E. Confocal image of basket-like synapses around the soma of D2 MSNs in sections from saline- (D) or 6-OHDA-injected mice (E). Scale bars are 5 µm. F. Bar graph showing an increase in the number of vGAT terminals that were co-labeled with PV around the somas of D2 MSN in 6-OHDA-injected mice compared to saline-injected mice. *p = 0.003. G–H. Confocal image of basket-like synapses around the soma of D1 MSNs in sections from saline- (G) or 6-OHDA-injected mice (H). Scale bars are 5 µm. I. Bar graph showing no change in the number of vGAT terminals that were co-labeled with PV around the somas of D1 MSNs in 6-OHDA injected mice compared to saline-injected mice.

Figure 5. Inhibitory mini frequencies measured in D1 and D2 MSNs

A. Example traces of mIPSCs recorded from D1 MSNs in saline-injected (left) or 6-OHDA-injected (right) mice. mIPSCs were recorded in the presence of 5 µM NBQX and 1 µM TTX. MSNs were voltage clamped at −80 mV. Due to high [Cl⁻] in the internal, E_Cl = 0 mV and mIPSCs are inward. B. Cumulative probability plot of mIPSC amplitude onto D1 MSNs in saline-injected and 6-OHDA-injected mice. C. Cumulative probability plot of mIPSC inter-event interval (isi) recorded in D1 MSNs in saline-injected and 6-OHDA-injected mice. D. Example traces of mIPSCs recorded from D2 MSNs in saline-injected (left) or 6-OHDA-injected (right) mice. E. Cumulative probability plot of mIPSC amplitude onto D2 MSNs in saline-injected and 6-OHDA-injected mice. F. Cumulative probability plot of mIPSC isi recorded in D1 MSNs in saline-injected and 6-OHDA-injected mice. The mIPSC isi was significantly shorter in D2 MSNs recorded in 6-OHDA-injected mice relative to saline-injected mice (*p < 0.0001, Kolmogorov-Smirnov test). G. Bar graph summarizing mIPSC amplitudes for the population of D1 and D2 MSNs recorded in saline-injected and 6-OHDA-injected mice. Circles represent data from individual cells and bars are the average for each population. Amplitudes were not significantly different across conditions. D1_saline = 46.5 ± 5.6 pA, n = 14; D2_saline = 49.0 ± 8.1 pA, n = 14; D1_6-OHDA 45.7 ± 3.9 pA, n = 15; D2_6-OHDA = 48.1 ± 6.4 pA, n = 14. H. Bar graph summarizing mIPSC frequencies for the population of D1 and D2 MSNs recorded in saline-injected and 6-OHDA-injected mice. mIPSC frequencies were significantly higher onto D2 MSNs in 6-OHDA-injected mice compared to saline-injected mice (*p = 0.04; **p = 0.02; ***p = 0.007). D1_saline = 9.0 ± 2.8 Hz, n = 14; D2_saline = 6.6 ± 2.5 Hz, n = 14; D1_6-OHDA 8.6 ± 2.7 Hz, n = 15; D2_6-OHDA = 11.3 ± 3.8 Hz, n = 14.

Figure 6. Increased innervation of D2 MSNs by fast-spiking interneurons promotes synchrony in a model of the striatal circuit

A. Diagram of synaptic connections in the model and the connection probability of each. Loops indicate lateral connections onto the same MSN subtype, not autapses. The conductance of all FS synapses was set to 1 nS and the conductance of all MSN synapses was set to 0.2 nS. B. Raster plot showing the firing of each neuron in the control network in response to input, simulated by a Gaussian noise function with a standard deviation of 100 pA. C. Normalized average cross-correlogram for the population of D1 MSNs and D2 MSNs. The normalized cross-correlogram was computed from data rebinned into 5 ms and converted to a z-score. D. Raster plot showing the firing of each neuron in the dopamine-depleted network in response to input, simulated by a Gaussian noise function with a standard deviation of 100 pA. E. Normalized average cross-correlogram for the population of D1 and D2 MSNs. Data was binned into 5 ms bins and converted to a z-score. Note the marked increased in synchrony apparent in the population of D2 MSNs.