Non-stationary discharge patterns in motor cortex under subthalamic nucleus deep brain stimulation

Abstract

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) directly modulates the basal ganglia (BG), but how such stimulation impacts the cortex upstream is largely unknown. There is evidence of cortical activation in 6-hydroxydopamine (OHDA)-lesioned rodents and facilitation of motor evoked potentials in Parkinson's disease (PD) patients, but the impact of the DBS settings on the cortical activity in normal vs. Parkinsonian conditions is still debated. We use point process models to analyze non-stationary activation patterns and inter-neuronal dependencies in the motor and sensory cortices of two non-human primates during STN DBS. These features are enhanced after treatment with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), which causes a consistent PD-like motor impairment, while high-frequency (HF) DBS (i.e., ≥100 Hz) strongly reduces the short-term patterns (period: 3-7 ms) both before and after MPTP treatment, and elicits a short-latency post-stimulus activation. Low-frequency DBS (i.e., ≤50 Hz), instead, has negligible effects on the non-stationary features. Finally, by using tools from the information theory [i.e., receiver operating characteristic (ROC) curve and information rate (IR)], we show that the predictive power of these models is dependent on the DBS settings, i.e., the probability of spiking of the cortical neurons (which is captured by the point process models) is significantly conditioned on the timely delivery of the DBS input. This dependency increases with the DBS frequency and is significantly larger for high- vs. low-frequency DBS. Overall, the selective suppression of non-stationary features and the increased modulation of the spike probability suggest that HF STN DBS enhances the neuronal activation in motor and sensory cortices, presumably because of reinforcement mechanisms, which perhaps involve the overlap between feedback antidromic and feed-forward orthodromic responses along the BG-thalamo-cortical loop.

Keywords: Parkinson's disease; deep brain stimulation; information rate; motor cortex; point processes; receiver operating characteristic curve; spike trains.

Figure 1

Histology. (A,B) Reconstruction of the location of the DBS lead and microelectrode recording sites for animal A (A) and B (B). Sections are labeled from I to VIII (A) and from I to III (B), respectively. The number on top of each section is the distance anterior to intra-aural line. (C) Example of histological section from which the lead location is determined (animal A, section VIII). (D,E) Example of section through the ipsilateral (D) and contralateral (E) substantia nigra pars compacta (SNpc) and ventral anterior thalamus (VTA) stained with TH antibodies after unilateral intracarotid injection of MPTP (animal A). See Results for explanation of labels A–E.

Figure 2

Evaluation of the fitting procedure for a target neuron in M1 cortex (animal A, normal) during 130 Hz DBS. (A) The predicted spiking probability was computed from the estimated model parameters, the spike train of the target neuron and the spike train of another neuron in the same ensemble. (B) Spike train of the target neuron in the same period. (C) Kolmogorov-Smirnov (KS) plot after time rescaling of the estimated model on the validation data. Grey lines are 95% confidence bounds. (D) ROC curve for the target neuron (TP = true-positive; FP = false-positive) on validation data. The color code in the legend also applies to the AUC values.

Figure 3

Response to the DBS pulse in animal A (A,B) and B (C,D) in normal condition. First row (top): post-stimulus time raster of a neuron in M1 (A,C) and S1 (B,D) cortex during 100 Hz STN DBS. Second row: post-stimulus time histogram (PSTH) normalized to the pre-DBS neuronal activity for the neurons whose raster is depicted in the top row. Grey dashed lines indicate significance levels (±1.96). Third row (bottom): percentage of neurons with z-score >1.96 during the inter-stimulus time interval in M1 (A,C) and S1 (B,D) cortex for various DBS frequencies. Legend in (A) also applies to (B–D).

Figure 4

Poisson factor and recurrent activation patterns (RPs) in M1 (A–C) and S1 (D–F) cortex in normal (black bars) and MPTP (grey bars) conditions (animal A), both at rest (0 Hz) and under DBS (50, 100, 130 Hz). (A,D) Population-mean value of the Poisson factor e^σ (mean ± S.E.M.). (B,E) Percentage of neurons with RFPs (3–7 ms period). (C,F) Percentage of neurons with long-term patterns (30–50 ms period). Asterisks indicate significant differences under DBS vs. no DBS (in A,D: t-test, p < 0.05; in B,C,E,F: χ²-test, p < 0.05). Diamonds indicate significant difference in MPTP vs. normal conditions (in A,D: t-test, p < 0.05: in B,C,E,F: χ²-test, p < 0.05).

Figure 5

Population-mean value of the point process model parameters for M1 cortex in normal (solid line) and MPTP (dashed line) conditions (animal A). (A–D) Mean value of e^β_r, r = 1, …, 18. (E–H) Mean value of e^δ_h, h = 1, …, 18. (I–K) Mean value of e^γ_ν, ν = 1, …, 8. Parameters are depicted vs. the history bins for a generic time t. Parameters in the first column were estimated at rest (0 Hz). Parameters in the following columns were estimated under 50, 100, 130 Hz STN DBS, respectively.

Figure 6

Temporal ensemble dependencies (EDs) in M1 (A,B) and S1 (C,D) cortex in normal and MPTP conditions (animal A), both at rest (0 Hz) and during DBS (50, 100, 130 Hz). (A,C) Percentage of neurons with FEDs (3–7 ms period). (B,D) Percentage of neurons with long-term dependencies (30–50 ms period). Color code, asterisks and diamonds are as in Figure 4.