Striatal spine plasticity in Parkinson's disease: pathological or not?

Abstract

Parkinson's disease (PD) is characterized by a dramatic loss of dopamine that underlies complex structural and functional changes in striatal projection neurons. A key alteration that has been reported in various rodent models and PD patients is a significant reduction in striatal dendritic spine density. Our recent findings indicate that striatal spine loss is also a prominent feature of parkinsonism in MPTP-treated monkeys. In these animals, striatal spine plasticity is tightly linked with the degree of striatal dopamine denervation. It affects predominantly the sensorimotor striatal territory (i.e. the post-commissural putamen) and targets both direct and indirect striatofugal neurons. However, electron microscopic 3D reconstruction studies demonstrate that the remaining spines in the dopamine-denervated striatum of parkinsonian monkeys undergo major morphological and ultrastructural changes characteristic of increased synaptic efficacy. Although both corticostriatal and thalamostriatal glutamatergic afferents display such plastic changes, the ultrastructural features of pre- and post-synaptic elements at these synapses are consistent with a higher strength of corticostriatal synapses over thalamic inputs in both normal and pathological conditions. Thus, striatal projection neurons and their glutamatergic afferents are endowed with a high degree of structural and functional plasticity. In parkinsonism, the striatal dopamine denervation induces major spine loss on medium spiny neurons and generates a significant remodeling of corticostriatal and thalamostriatal glutamatergic synapses, consistent with increased synaptic transmission. Future studies are needed to further characterize the mechanisms underlying striatal spine plasticity, and determine if it represents a pathological feature or compensatory process of PD.

Fig. 1. Loss of striatal spines in MPTP-treated monkeys

(A–F) Examples of Golgi-impregnated medium spiny neurons in the striatum of a control (A–C) and a MPTP-treated (D–F) monkey to illustrate the dramatic reduction in dendritic spines in the MPTP-treated animal compared with control. The boxed areas in B and E are shown at higher magnification in C and F, respectively. (G) Quantitative measurements of the density of dendritic spines in various striatal regions of control versus MPTP-treated monkeys. Note the significant reduction in spine density throughout the whole striatum. Abbreviations: Pre: Pre-commissural; Com: Commissural; Post: Post-commissural. (H–K) Significant reduction in the density of total striatal spines (I), or in D1-immunoreactive (D1-IR, J) and D1-immunonegative (K) spines in the striatum of MPTP-treated monkeys. The density values along the Y axis are number of spines/µm² of striatal tissue. (H) depicts an example of labeled (LSp) and unlabeled (USp) spines in D1-immunostained striatal tissue of a control monkey. An unlabeled dendrite (D) and terminal (T) are also depicted in the neuropil. Scale bars: A: 25 µm (valid for D); C and F: 5 µm; E: 5 µm (valid for B); H: 0.5 µm. (See Villalba et al., 2009 [11] for more details).

Fig. 2. 3D reconstruction of dendritic spines in the monkey striatum

(A, B) Electron micrographs showing a vGluT1- (A) and a vGluT2- (B) immunoreactive axon terminal (T) forming asymmetric synapses with the heads (H) of dendritic spines in the striatum of a control monkey. The neck of the spine (N) is also labeled in these micrographs. (A.1–B.1) 3D reconstruction of the two labeled terminals and their postsynaptic targets shown in A and B. In A.2 the axo-spinous complex has been rotated to better illustrate the entire extent of the synaptic junction. (C) Quantitative measurements of the volumes of spines and terminals (in µm³) and the surface areas of postsynaptic densities (PSD; in µm²) at axo-spinous synapses formed by vGLuT1- or vGluT2-immunoreactive boutons. The total number of reconstructed terminals in each group is indicated in parentheses. Scale bars: 1 µm.