Chronic spinal cord electrical stimulation protects against 6-hydroxydopamine lesions

Abstract

Although L-dopa continues to be the gold standard for treating motor symptoms of Parkinson's disease (PD), it presents long-term complications. Deep brain stimulation is effective, but only a small percentage of idiopathic PD patients are eligible. Based on results in animal models and a handful of patients, dorsal column stimulation (DCS) has been proposed as a potential therapy for PD. To date, the long-term effects of DCS in animal models have not been quantified. Here, we report that DCS applied twice a week in rats treated with bilateral 6-OHDA striatal infusions led to a significant improvement in symptoms. DCS-treated rats exhibited a higher density of dopaminergic innervation in the striatum and higher neuronal cell count in the substantia nigra pars compacta compared to a control group. These results suggest that DCS has a chronic therapeutical and neuroprotective effect, increasing its potential as a new clinical option for treating PD patients.

Figure 1. DCS improves weight recovery of 6-OHDA lesioned rats.

(a) Time course of the experiment. Numbers indicate experimental weeks from onset at time 0 (6-OHDA lesion). Rats in 6-OHDA + DCS group sustained epidural DCS electrode implantation (week −1) 1 week before the 6-OHDA lesion (week 0). Rats in 6-OHDA and sham control groups underwent only bilateral lesion procedure; 6-OHDA group (total 52.5 μg 6-OHDA), sham control group (only vehicle solution). Timing of motor assessment and 30 minute DCS sessions is illustrated by grey and black upward arrows respectively. Six weeks post lesion (week 6), brains were collected and processed for immunohistochemistry (IHC). (b) Changes in body weight after bilateral intrastriatal 6-OHDA lesion with or without DCS treatment. Lesioned, non-treated rats (n = 8) suffered sustained weight loss with little-to-none recovery. Lesioned rats with DCS treatment (n = 7, 30 min, 333 Hz continuous DCS during 30 min twice a week, starting 7th day, black arrow) recovered body weight significantly faster than non-treated rats (p < 0.0001, two-way repeated measure ANOVA). *:p < 0.05(day 11), **:p < 0.01(days 12–14), ***:p < 0.001(days 15–42), Bonferroni multiple comparisons. c) Maximum weight loss is significantly larger for non-treated rats compared to DCS treated rats. *:p < 0.05, Mann-Whitney test. d) DCS treatment reverses the trend of weight loss almost immediately, while non-treated rats continue to lose weight till week 5 as shown by weight change relative to week 1, which is significantly higher for treated rats as compared to non-treated. *:p < 0.05, **:p < 0.01, ***:p < 0.001, Mann-Whitney test. (e) DCS treatment results in accelerated weight recovery. Rate of weight gain is significantly higher in treated rats compared to non-treated. *:p < 0.05, Mann-Whitney test, a.u. (arbitrary unit). All error bars are s.e.m.

Figure 2. DCS restores motor functions in PD rats.

(a) Changes in rat posture (measured as major axis length, greater length implying better posture) with or without DCS treatment. Lesioned rats develop crouched posture resulting in shorter major axis length. DCS treatment restores posture significantly faster than non-treated rats [groups × weeks interaction: p < 0.05, two-way repeated measure ANOVA, *:p < 0.05 at week 3.5 and 4.5 between DCS treated (n = 6) and non-treated rats (n = 8), n.s.: DCS treated rats were not significantly different from controls (n = 4) from week 2, Bonferroni multiple comparisons]. (b) Distance travelled and (c) average speed during a 30 min open field session was not significantly different between the groups (groups × weeks interaction: p < 0.0001 for distance, groups × weeks interaction: p < 0.01 for speed, but no differences on post-hoc analysis for both). (d) Spectral power of the acceleration vector in frequency range 0.5–4.75 Hz (indicating jerky non-smooth locomotion) was significantly higher in lesioned rats than DCS treated and controls towards the end [p < 0.05, two-way repeated measure ANOVA, *:p < 0.05 (6-OHDA + DCS compared to 6-OHDA), +:p < 0.05 (6-OHDA compared to controls), Bonferroni multiple comparisons]. (e) Stride length measured at week 4 was significantly higher for DCS treated rats as compared to non-treated (p < 0.01, one-way ANOVA, *:p < 0.05, **:p < 0.01-Tukey's multiple comparison test). All error bars are s.e.m.

Figure 3. DCS protects nigrostriatal dopaminergic system.

(a) Representative immunostaining for tyrosine hydroxylase (TH) in striatum (DAB stain). Note the higher dopaminergic innervation in the striatum (CPu, Acb) of DCS treated rat as compared to non-treated, scale bar = 1 mm, CC = Corpus Callosum, LV = Lateral Ventricle, CPu = Caudate Putamen, Acb = Nucleus Accumbens. 6-OHDA lesion caused a 67% decrease of striatal TH levels (measured by contrast index), with respect to the sham control group, while treatment of the 6-OHDA lesioned rats with DCS resulted only in a 35% decrease, (top panel, 3c). The difference between the TH levels of 6-OHDA (n = 6) and 6-OHDA + DCS (n = 6) groups was significant (*:p < 0.05, Bonferroni's Multiple Comparison, (b) Representative immunostaining for tyrosine hydroxylase in substantia nigra pars compacta (SNc), scale bar = 500 um, SNc = substantia nigra pars compacta, VTA = ventral tegmental area. 6-OHDA lesion resulted in a severe loss of TH immunoreactivity (measured by contrast index) in the SNc. 6-OHDA rats showed 74% loss in TH CI as compared to sham controls while DCS treated rats showed only 44%. There was significant difference between the TH levels of 6-OHDA (n = 6) and 6-OHDA + DCS (n = 6) groups in the SNc (*:p < 0.05, Bonferroni's Multiple Comparison, middle panel, 3c. Dopaminergic neuronal cell loss in SNc (expressed as % of neuronal count in sham controls) was significantly higher in 6-OHDA (93.32 ± 1.62, n = 8) rats as compared to 6-OHDA + DCS (87.43 ± 1.95, n = 6) rats (*:p < 0.05, t-test, 2 tailed). All error bars are s.e.m.

Figure 4. Neuroprotective effect of DCS on weight and PD symptoms.

(a) There was significant correlation between weight and major axis length (measure of posture) throughout the experiment (Spearman test, p < 0.0001). (b) Body weight had a significant negative correlation with spectral power of acceleration vector (indicating jerky non-smooth locomotion) – Spearman test, p < 0.0001. (c) Spectral power of acceleration vector and speed were significantly correlated throughout the experimental period (Spearman test, p < 0.0001). Stride length measured at week 4 had significant correlation with major axis length ((d), Spearman test, p < 0.01) and body weight ((e), Spearman test, p < 0.01) indicating that recovery of body weight was related to an overall improvement in motor symptoms.

Figure 5. DCS reverts weight, behavioral, and cellular parameters back to normality.

(a) Principal component analysis (PCA) was performed using 9 rats having all the data variables at a single time point at the end of the experiment [weight (day 42); axis length, acceleration power, distance and speed at week 5.5; stride (week 4); TH immunoreactivity of striatum and SNc and SNc cell count). Representation of the individual rat data points, using the first three PCs on a 3D space, shows that the rats tend to cluster according to their experimental group. Agglomerative cluster analysis of the data (dotted ellipsoids) correctly identified the 6-OHDA rats (red dots), but failed to separate the control (blue squares) and the 6-OHDA + DCS (green triangles) rats. (b) PCA was performed with four motor parameters [speed, distance, non-smooth gait (acceleration power), and posture (axis length)] for all time points and the first two PC scores of all groups were averaged for each time point and plotted against time post lesion. The resulting graph shows the progression of the groups throughout the 6 weeks of the experiment. While the three experimental groups start close to each other (week 0.5–1.5), the 6-OHDA and 6-OHDA + DCS groups split from control after week 2. By the end of the experimental time, the 6-OHDA + DCS group joins the control group (c) Euclidean distances of the average PCs of each group shows that the motor parameters of the 6-OHDA + DCS group started to drift from the untreated 6-OHDA group after week 1.5 and became close to control parameters after week 4.