Introduction

Lewy bodies (LB) are protein inclusions containing disaggregated oligomers of many cellular proteins. The German neurologist named Friederich Lewy[1] was the first physician-scientist to describe the abnormal protein deposits in 1912 in people with paralysis agitans and, later on, Parkinson disease. Dystrophic neurites (LNs) are precursors of LB and they can contain deposits of ubiquitin (Ub) and alpha-synuclein (a-Syn), formally becoming LB and accumulating in synaptic terminals and axonal processes. A-Syn, encoded by SNCA, is a protein found in presynaptic terminals and thought to have an important neurotransmission communication between neurons and neurotransmitter vesicle trafficking.[2][3][4] Ubiquitin is a protein involved in changing proteins biochemically to target them to degradation and autophagy.[5] The Ub is usually at the core and neurofilaments at the outermost layer. Per recent animal model studies, a mutated form of a-Syn (A30P) leads to disaggregation, transcriptional deregulation, and silencing of DNA; it also leads to the disruption of Golgi and endoplasmic reticulum, and finally it may also play a role in collagen gene transcription, a basement membrane protein that maintains the integrity of dopaminergic neurons. These cellular processes go awry in dementia with Lewy bodies (DLB) and Parkinson disease (PD).[6] These aberrant oligomers are also present in many other neurodegenerative disorders including, Alzheimer disease (AD), striatonigral degeneration, olivopontocerebellar atrophy, and pantothenate kinase-associated neurodegeneration (Hallervorden Spatz disease), but it is unknown if they share the same pathogenesis. This article focuses on DLB and PD dementia as they are the most common forms of dementia besides Alzheimer disease. The worldwide incidence of DLB is about 1% to 2% (approximately 3 to 4 per 100,000 person-years) on patients older than 65 years and about 5% of all dementia cases in those over the age of 75.[7][8] For PD dementia, approximately 10% of a PD population will develop dementia annually with an overall prevalence of 31.1% in patients between 65 to 90 years old. The data concerning age at disease or dementia onset are highly variable in the literature.

The disaggregation of alpha-synuclein, Ub, and LN are a pathognomonic neuropathological finding on autopsy in patients with PD, and DLB, found in many regions within the central nervous system including the cerebral cortex, limbic cortex, substantia nigra, hippocampus. Structurally, they are single, multiple, or polymorphic. It characterizes them into two major groups: classical or cortical LB. The classifications are based on morphology, cellular components, and location within the brain. Although Lewy Bodies are implicated in many of these disorders, it is debatable whether they are a consequence of cellular damage and dysfunction within the neuro integrity or a cause itself of the clinicopathological symptoms seen in patients. This article briefly covers the different anatomical, clinical and biochemical pathologies involved with Lewy body formation. The goal is not to provide an exhaustive description of LBs, rather a snapshot that can be useful for the everyday clinician interested in the topic.

Anatomical Pathology

Based on a semi-quantitative assessment of LBs in a large autopsy series, the Braak staging was proposed in 2003[9][10] to characterize LB pathology lesion progression in the brain. It attempts to centralize and provide a systematic framework of the progression of the disorder using neuroanatomy LB lesion density and progressive dissemination starting from the brainstem and ending in the neocortex. Per this categorization, the lesions begin in the lower brainstem and involve the intermediate reticular zone, anterior olfactory nucleus, and nucleus basalis membrane (NBM), with midbrain regions being preserved (stage 1). It continues progressing into the caudal raphe nuclei, gigantocellular reticular nucleus, and coeruleus-subcoeruleus complex (stage 2). Professionals consider these stages asymptomatic or pre-symptomatic and may explain early non-motor (autonomic and olfactory) symptoms that characterize disorders such as PD and DLB, preceding much of the motor and sensory dysfunctions. During stage 3, the inclusions will stain in the locus coeruleus (LC), amygdala, nuclei of the basal forebrain, and posterolateral and posteromedial substantia nigra compacta (SNc), with relative preservation of the cortex. Finally, we can characterize Stage 4 by the further damage to the amygdala, temporal limbic, and neocortex; it correlates these with symptomatic stages of a disease like PD. Once the abnormal protein inclusions reach stages 5 and 6, it is spread through the cortex with devastating effects on complex sensory association cortex regions and prefrontal and temporal areas (e.g., hippocampus) involved in executive brain functions, and learning and memory. Different neurotransmitters subtypes compose the anatomical structures and a neuron-to-neuron network of communication (e.g. norepinephrine, acetylcholine, serotonin, among others) according to regional specific brain structures affected, adding complexity to the possible variable clinical syndromes that characterize neurodegenerative disorders such as PD and DLB.

Clinical Pathology

Some scholars and scientists are skeptical that the Braak staging system can accurately correlate neuroanatomical LB density/dissemination with predictable clinical symptoms. The staging is based on retrospective clinical pathologic studies in younger-onset PD patients with about 6% to 40% of the cases not following the proposed caudo-rostral progression pattern of pathology.[11] For example, there was sparing of medullary nuclei in 7% to 8.3% of clinically manifested PD cases with alpha-synuclein inclusions in the midbrain and cortex corresponding to Braak stages 4 and 5, whereas mild parkinsonian symptoms were already observed in stages 2 and 3. Additionally, mixed pathologies with variable Alzheimer-type and DLB pathology can further complicate the precision of the Braak stages. Unfortunately, there is no clear relationship between the Braak LB stage and the clinical severity of PD, retrospective autopsy series in 30% to 55% of elderly subjects with widespread Lewy-related pathology (Braak stages 5 and 6) reported no definite neuropsychiatric symptoms, suggesting considerable cerebral compensatory mechanisms. Almost half of the LB-positive cases during the Braak based studies were not classifiable. The etiology of these deviations from the current staging schemes of alpha-synuclein pathology in PD and DLB, its relation to the onset of classical parkinsonian symptoms, and lack of definite clinical deficits despite widespread alpha-synuclein pathology required a better classification system.[12]

Beach and colleagues (2009)[13] created a modified criterion that correlates nigrostriatal degeneration, cognitive impairment, and motor dysfunction with many Lewy Body associated disorders. This classification was more specific by combining anatomical inclusion findings with biochemical enzymatic activity of TH, an enzyme upstream of L-dopa production and a precursor of dopamine production in the brain. It contains the following anatomical pathology staging: I, olfactory bulb only; IIa, brainstem predominant; IIb, limbic predominant; III, brainstem and limbic; and IV, neocortical. This study established a multi-factorial correlation between the deterioration of striatal TH concentration, Mini-Mental State Examination score, and Unified Parkinson's Disease Rating Scale part III. The authors found significant correlations between these measures and LB-a-Syn pathology.

The above-mentioned complexity is exemplified in patients with incidental Lewy body disease (ILBD), or LBs found in autopsy with no documented clinical parkinsonism.[14][15] The distribution of LBs in ILBD is similar to the PD, but the lesions spare the limbic cortex with, on average, lower Braak scores, compared to diagnosed PD cases. In the latter, LBs are found in most brain regions with significantly higher Braak PD stage. Additionally, decreased TH immunoreactivity was shown in the striatum and epicardial nerve fibers of ILBD, but to a lesser extent than PD.[15][16] However, about 5% to 20% of asymptomatic patients on autopsy have shown abundant Lewy pathology with a distribution pattern similar to that seen in PD, with substantia nigra relatively well preserved. This challenges the classical Braak brainstem to cortex distribution and suggests more of a multiregional disease progression from the onset. There are prevalent outlying cases that do not conform to any staging scheme, with varying distributions of LBs in patients likely greater than expected for PD but may also reflect the different cohorts examined different technical issues or incidental Lewy body disease rather than overt PD.  Ultimately, the Beach modified staging showed a promising correlation between anatomical pathology findings, biochemical markers, and clinical ancillary testing; however, it requires validation at a larger scale.

Biochemical and Genetic Pathology

The biochemical pathways involved in the formation of LB, specifically alpha-synuclein, are associated with protein misaggregation and impaired protein cleavage. [17] Wild-type alpha-synuclein is prone to form oligomeric and prefibrillar structures that alter lysosomal trafficking and mitochondrial function. Specifically, in the mitochondria, early soluble oligomeric forms of a-Syn interfere with complex I in the electron transport chain, generating reactive oxygen species. This can disrupt high energy production for functional neurotransmitter communication between dopaminergic neurons in the brain. Some scientists consider LBs to be markers of ongoing neuronal damage, other as a sequela of harmless end products of sequestration of toxic molecules that may or may not be directly involved in apoptosis or cell death. There is an inverse correlation between cell viability and the amount of a-Syn aggregates in the cells. Future studies will determine whether LBs inclusions are upstream or downstream of apoptosis or cell death.

Classical and cortical LBs share immunochemical and biochemical characteristics. The major biochemical components include a-Syn, ubiquitin (Ub), and phosphorylated Ub associated with multiple other substances, including structural fibrillary elements, a-Syn-binding proteins, proteins implicated in the Ub-proteasome system, synphilin-1, aggresome- and mitochondria-related proteins, and cytoskeletal, cytosolic, and cellular response proteins, among others.

LBs have a distinct central Parkin-positive and Ub-positive domain with a-Syn in the periphery. Aggregation of a-Syn, synphilin and Parkin within LBs suggests that parkin plays a role in ubiquitination and modification of a-Syn. Tyrosine hydrolase (TH) and choline-acetyl transferase (ChAT) are co-localized in cortical LBs; brainstem LBs have TH and ChAT immunoreactivity in the core, surrounded by a peripheral rim of a-Syn, suggesting that they may disrupt cholinergic and catecholaminergic transmitter production.[18]  LBs and pale bodies are immunoreactive for autophagic proteins p62 and NBR1 which may sequester the soluble oligomeric a-Syn into inclusions.[19][20] They further contain 14-3-3 proteins that are involved in signal transduction pathways and interact with a-Syn and torsin A, a protein that may serve as a chaperone for misfolded proteins that require degradation.[21] A-Syn aggregates in LBs are refractory to clearance and may contribute to increased death of aggregate-bearing cells. Proteomic analysis of cortical LBs has revealed 200+ metabolites and proteins related to multiple or unknown functions.[22] In substantia nigra proteomic study, they identified 204 proteins displayed significant expression level changes in PD patients versus controls, implicated in pathogenic processes including mitochondrial dysfunction, oxidative stress, or cytoskeleton impairment.[23]

In summary, the Ub-proteasome system and the autophagy pathway help our brain clear toxic forms of misaggregated proteins, including helping with a-Syn turnover. Dysfunctional proteolytic pathways may result in the accumulation of toxic forms of a-Syn. Ubiquitinated proteins in LBs may be a manifestation of attempts by the cellular machinery to eliminate damaged cellular components and delay the onset of neuronal degeneration. Some evidence showing this is the high incidence of mitochondrial DNA deletion in LB-positive neurons in PD brain compared to LB-negative neurons and controls, suggesting increased mitochondrial damage in LB-positive neurons.[24][25] These deletions trigger neuroprotective mechanisms that at one point may become overwhelmed. However, how this process works in the mitochondria, specifically how soluble converts to insoluble aggregates in neurons is poorly understood and requires further research.

Morphology

The classical LB are spherical hyaline eosinophilic cytoplasmic inclusions (8 to 30 micrometer-diameter) comprising concentric lamellar bands surrounded by a pale-stained halo of 10 nm of wide radiating fibrils. Cortical LBs are difficult to find under the microscope, chemically eosinophilic, rounded, angular, without a halo. Additionally, cortical LBs are disorganized, granulofibrillary structures composed of 10- to 27-nm wide filaments, without a central core. They concentrate in lower cortical layers, with recorded accumulation in the hippocampus, insular cortex, amygdala, and sometimes cingulate gyrus. Similar lesions, rounded areas of granular, pale-staining eosinophilic material displacing neuromelanin (NM) in brainstem neurons are precursors of LBs.  As described, the alpha-synuclein is the best marker to detect LBs. Antibodies that recognize N-terminal epitopes (synucleins 505, 506, and 514) selectively detect a-Syn.[26]

Mechanisms

The etiology and pathogenesis of synucleinopathies are complex with proposed interactions between gene transcription and expression and environmental insults that lead to cellular dysfunction in vulnerable neurons in the central nervous system. Dysfunctional regulation, processing, cleavage, and degradation of abnormal cytoskeletal and endovesical (Golgi and endovascular reticulum) proteins within the cells are thought to interfere with mitochondrial dynamics, membrane integrity, proteosome, and lysosomal pathways. Genetic susceptibility may determine susceptibility, for example, free radicals generated through disruption of complex I activity, increasing the nuclear or mitochondrial genomic deficit. Environmental interactions factors such as the metabolism of exogenous drugs can generate endogenous neurotoxins that can also trigger the generation of reactive oxygen species that disrupt calcium homeostasis and trigger cell death.  Neuronal pathways are susceptible to oxidative processes, and therefore, become a prime target to these insults. Specifically, reactive oxygen species can trigger excitotoxicity by increasing glutamate in the synapses, triggering neuroinflammatory microglial processes, and most likely, a combination of all these complex processes. Interestingly, Lewy bodies are like other aggregated proteins such as neurofibrillary tangles (NFT) in AD and/or Pick bodies in terms of being present as markers of neurodegenerative processes. What exact analogous mechanism they have in common to disturb cellular activity is complex and is currently under investigation.[27][28][29][30]

Clinicopathologic Correlations

This section focuses on describing the clinical symptoms and pathophysiological findings that distinguish DLB with PD dementia, as those are the most common diseases associated with Lewy body deposition. The section also notes promising biomarkers for the early diagnosis of DLB or PD. DLB is the second most common type of progressive dementia after Alzheimer disease dementia. DLB predominantly shows greater cognitive, psychiatric, and autonomic deficits than PD, which includes severe and rapid fluctuations in verbal memory, frontal executive function deficits, attention, spontaneous visual hallucinations, delusions, orthostatic hypotension, prominent falls/gait disturbances, and more rapidly progressive cognitive decline compared to motor symptoms. Similarities with PD include the development of motor symptoms including rigidity, akinesia, mood disturbances, REM sleep behavior disorder, language, and visuospatial and constructive deficits, and cognitive impairment (develops later in PD patients). Resting tremors are more common with Parkinson disease than DLB. The typical rigidity, tremors, akinesia, and postural instability precede the cognitive deficits in PD dementia.

PET and postmortem studies have revealed more pronounced cortical atrophy, elevated cortical and limbic Lewy body pathologies, higher Abeta and tau loads in cortex and striatum in DLB compared to PD dementia. Specifically, gray matter cortical atrophy more frequent and more severe in DLB. White matter hyperintensities in the temporal lobe are more severe and more frequent in DLB. There is also a difference in functional connectivity and corticostriatal disruption with PD dementia being characterized by frontal cortical disruption and DLB with more parietal and occipital disruption. Additionally, amyloid and tau signal seems to be more prevalent in DLB on PET imaging and CSF profiles. Functional differences in advanced nuclear imaging are seen in DLB compared to PD dementia with decreased DAT binding in the caudate related to functional impairment. Cerebrospinal fluid (CSF) alpha-synuclein oligomers are increased in PD dementia in comparison to DLB. Overall, PET studies have failed to show any differences in cortical and striatal cholinergic and dopaminergic deficits.

Clinical management of both DLB and PD dementia includes cholinesterase inhibitors like rivastigmine. Atypical antipsychotic agents like quetiapine may improve psychosis in PDD/DLB, but the evidence for this is poor and adverse effects from such therapy are common and may be severe. Non-pharmacological interventions can also be effective but require further study. Currently, no disease-modifying therapies are available, although current efforts are on the way to find diagnostic markers that could identify these diseases sooner in hopes to start treatment sooner and prolong the disease progression. Many biomarkers including CSF oligomeric A-syn, DJ-1, lysosomal enzymes (e.g., glucocerebrosidase) show promise as a diagnostic biomarker of PD, but require the standardization of available assay, homogenous protocols, and surpass other methodological challenges to reach a large scale validation that can be reliable for clinical use.[31][32][33][34][35]

Clinical Significance

Although Lewy body pathologies are intriguing and may provide a clue of the pathophysiology of DLB, PD dementia, and other related neurodegenerative disorders, it is discovered on autopsy and therefore not useful for routine clinical diagnosis. DLB and PD are important dementia syndromes that overlap in many clinical features, genetics, neuropathology, and management. They are considered subtypes of an alpha-synuclein-associated disease spectrum (Lewy body diseases), from the incidental Lewy body disease and non-demented Parkinson disease to PDD, DLB, and DLB with Alzheimer disease at the most severe end. The clinical symptoms in relation to the different cognitive, psychiatric, and motor domains should be sufficient for the diagnosis. A rigorous history and physical exam, with ancillary testing including imaging and nuclear studies that like DAT, CSF profiles, and tau and amyloid pathology can be useful for diagnostic purposes but does not change medical management of these disorders. The interesting phenomenon of various biochemical markers that accumulate in DLB and PD dementia including alpha-synuclein, ubiquitin, tau, and amyloid provides evidence of the multi-factorial and complex processes that become dysfunctional in chronic and progressive neurodegenerative disease such as DLB and PD dementia. The pathophysiological threshold to develop clinical symptoms when these cellular and biochemical processes are disrupted is elusive. Although this anatomical neuropathological classification above mentioned is useful, it does not entirely predict the diagnosis, progression, and prognosis in all the synucleopathies. Further research is needed to develop relevant biomarkers and therapeutic targets that can delay the progression of diseases such as DLB and PD.