Parkinson's disease: Before the motor symptoms and beyond

Abstract

The understanding of the biology of Parkinson's disease (PD) has advanced rapidly over the past 3 decades. In particular, the early pathological changes described by Braak et al. and the awareness of extensive and clinically relevant premotor manifestations are of diagnostic and therapeutic importance.

We review those manifestations and their contribution to the clarification of the pathophysiologic processes of PD, and discuss the implications for treatment of the disease.

Introduction

For almost two centuries the clinical description and diagnostic criteria of Parkinson's disease (PD) rested exclusively on the motor manifestations of the disease. The first notion by James Parkinson [1] that “the senses and intellect remain uninjured” has been accepted unchallenged for 150 years; depression, when present, was thought to be reactive, and autonomic changes were thought to be either age-related or drug-induced. However, levodopa, as a very effective drug, could suppress or eliminate many if not all the three foremost parkinsonian features of tremor, rigidity and bradykinesia, allowing other manifestations to come to the forefront, including cognitive, affective, autonomic and sensory ones.

In parallel to these clinical observations, data on the pathology of PD also accumulated. The first observations of depigmentation of the substantia nigra, loss of dopaminergic neurons and deposition of Lewy bodies (LB) in surviving cells, correlated with the dopaminergic loss. However, gradually, additional observations were collected. Among the first was the loss of cholinergic neurons in the nucleus basalis of Meynert, which was associated with cognitive impairment, similar to that observed in Alzheimer's disease (AD) [2]. In addition to the cognitive deficit, abnormalities of the autonomic system became recognized, particularly constipation, which was attributed to deposition of LB-like structures in the autonomic neurons in the plexuses of Auerbach, the first demonstration that PD is not strictly confined to the central nervous system [3], [4].

The breakthrough came through a combination of illuminating discoveries, each being insufficient by itself to change the global picture.

Epidemiological data accumulated to demonstrate that not only constipation, but also olfactory loss, depression, a variety of non-specific pain syndromes and sleep disorders, actually antedated the first clinical motor manifestation of PD [5]. This was complemented by the identification of the chemical constitution of LB. The discovery of ubiquitin which led to awarding the Nobel Prize in Chemistry in 2004 to Hershko, Ciechanover and Rose, was important, although the non-specific function of ubiquitin in the protein disposal pathway was only a first step. Very soon after, the discovery of α-synuclein as an integral component of LB had a great impact, because genetic studies showed that mutations, duplications or triplications of the α-synuclein gene led to a phenotype clinically indistinguishable from sporadic PD [6], [7]. Thus, the hypothesis of α-synuclein as a central player in the cellular neuronal decay has been accepted unopposed.

Modern biological techniques allowed the production of anti-α-synuclein antibodies, which could stain intracellular α-synuclein deposits in LB. But it was the meticulous work of Heiko Braak and colleagues which really extended and expanded the understanding. Using semi-thin brain sections Braak et al. were able to show that these antibodies stained not only classical LB but also α-synuclein deposits in axons and dendrites [8], [9]. The importance of these deposits lies in the fact that although the relevance of LB to neurodegeneration has always been debated, it was hard to accept that nerve terminals loaded with α-synuclein deposits could function normally.

Furthermore, Braak et al. demonstrated a very orderly progression of Lewy pathology throughout the brain, in which the substantia nigra lesions were never the earliest to appear. These were preceded by Lewy neurites in the olfactory bulb and the dorsal motor nucleus of the vagus nerve (DMNX), and then progressing rostrally along the brain stem, affecting the locus coeruleus and raphe nuclei before hitting the midbrain substantia nigra. Further upward extension of the pathologic process involves the cortex. The cerebellum is also affected by this process [10].

The changes of the DMNX are believed to result in constipation (and possibly other autonomic manifestations), while the locus coeruleus and raphe nuclei changes were thought to underlie the depression and sleep disturbances.

Braak et al.'s data were generally well accepted, although some non-consenting voices have been heard [11]. However, two important interpretations of the Braak hypothesis must be discussed. Although constipation and olfactory changes are common premotor manifestations of PD [12], [13], they definitely do not appear in all cases, and the same is true for depression and sleep disorders. This obvious observation by itself does not negate the Braak hypothesis, which is qualitative rather than quantitative. The theory does not imply that a given stage will only begin once lower stage structures have been completely eliminated. Rather, it just says that some DMNX neurons, not all, will be affected at an early stage of the disease. Consequently, depression or sleep associated features may be skipped, if the number of neurons affected by synucleinopathy in the relevant brain-stem structures is below a given threshold, and similarly even the typical motor manifestation of PD may not necessarily appear prior to more rostral changes.

Secondly, the Braak hypothesis was interpreted to suggest that the fact that the pathological changes appear sequentially implies a causative process, i.e., that the DMNX changes inflict an upward change, perhaps by spread of the pathology trans-synaptically. In fact, Braak et al. later suggested that the pathological process of PD may actually begin in the periphery, e.g. the gastro-intestinal system, where a non-identified toxin could have been absorbed from the alimentary tract which inflicted damage to local neurons which, in turn, damaged the DMNX, etc. [14].

There is no direct proof of this theory, and since correlation can never be a proof of causation, additional evidence must be presented before the theory can be accepted. One alternative interpretation may be that different neuronal populations (and different neurons in a given population) vary in their susceptibility to the pathologic process of α-synuclein deposition, which occurs independently in each.