Neuroinflammation in Parkinson Disease

There is evidence for a role of neuroinflammation in Parkinson disease (PD) that begins with α-synuclein (α-syn) aggregation, involves microglial activation and the innate and adaptive immune system, and is required for neurotoxicity and cell loss to occur. The term neuroinflammation is used frequently, but there is no universally accepted definition. We define neuroinflammation as a reactive response of immune cells and the mediators they produce (eg, cytokines, chemokines, reactive oxygen species, and other secondary messengers), which lead to inflammation of neural tissue.¹ PD is characterized by a loss of dopaminergic (DA) neurons in the substantia nigra, marked by pathologically misfolded α-syn and Lewy bodies, in neural cell bodies.² It has been hypothesized that the α-syn pathology begins in the gut, travels through the vagus nerve, and enters the central nervous system (CNS) at the dorsal motor nucleus, although this has since been debated.³ Brain tissue of patients with PD who underwent fetal nigral transplantation showed increased cytoplasmic α-syn in the transplanted tissue, supporting the theory that α-syn may propagate between cells similarly to prions.⁴

Neuroinflammation has been implicated from the earliest disease pathology, including promoting α-syn aggregation, ultimately causing DA cell loss. Recent studies suggest dysregulated innate and adaptive immune responses are triggered by misfolded and aggregated α-syn, and other factors (eg, microbiota) also play a critical role in triggering neuroinflammation. Further aggregation of α-syn and eventual DA loss caused by neuroinflammation result in symptoms and signs of PD.⁵ Considering this potential etiology and pathophysiology, there are multiple immunotherapies under investigation for treating patients with PD.⁶ In this article, we review the basics of the immune system and evidence of neuroinflammation in PD disease pathology. In an accompanying article, we review therapeutic investigations aimed at modifying the mechanisms leading to neuroinflammation and disease progression in PD.

Basics of the Immune System

The immune system is often divided into the innate and adaptive immune system (Figure),^5,7 which are not mutually exclusive systems but rather, are different phases of a coordinated defense system. Components and pathogen specificity differ in the innate and adaptive immune systems.

Innate Immune System

Cells of the innate immune system include mast cells, monocytes and macrophages, neutrophils, natural killer (NK) cells, and others. Toll-like receptors (TLRs) on circulating cells of the innate immune system, such as macrophages, serve as a rapid first line of defense in recognizing pathogens. TLR are also found inside the cells in endosomes to detect intracellular pathogens or signals from dying cells. Different types of TLRs recognize different types of microbial components and respond by changing signaling in the innate immune system to regulate the cell response to different pathogens.

In the brain, microglial cells are the predominant immune cell, functioning as macrophage cells, scavenging the brain for damaged neurons and infectious entities. Microglial cells communicate with astrocytes, which regulate homeostasis in a healthy brain and respond to neuronal injury.⁸ There are also brain macrophages that are distinct from microglial and other immune cells in the brain. Infiltrating immune cells from the periphery, such as macrophages and monocytes, are also involved in cross talk with microglia further augmenting immune-mediated neuroinflammation.⁵

Adaptive Immune System

The adaptive immune system is comprised of lymphocytes (ie, T and B cells) and antigen-presenting cells (ie, macrophages and dendritic cells). There are different subtypes of T cells, including helper T cells (ie, Th1, Th2, Th17), regulatory T cells (T_regs), and cytotoxic T cells. Th1 and Th17 cells are considered proinflammatory, and Th2 and T_regs are generally characterized as anti-inflammatory. Each T-cell subtype produces a different set of cytokines. Although Th cells are named for a ‘helper’ function, they have the propensity to cause inflammation and tissue damage. Activated cytotoxic cells can directly kill other cells. Disruption of typically orderly cross talk amongst immune cell types can lead to a dysregulated immune response and proinflammatory state, which is seen in PD.⁵

Major histone complexes (MHC) displayed on cell surfaces are vital in presenting peptides to T cells. In humans, MHCs are encoded by a broad variety of human leukocyte antigen (HLA) genes. Lysosomal peptides, which contain peptides that come from outside of a cell, attach to and are presented by MHCII. Only dendritic cells, B cells, microglia, and macrophages have MHCII antigens on their cell surfaces. In contrast, MHCIs are located on all nucleated cells, and present peptides that were degraded in proteasomes, meaning these peptides came from within the cell, such as viral antigen particles. Helper T cells can release cytokines and activate B cells, causing the B cells to differentiate into plasma cells that make antibodies. In turn, antibodies neutralize pathogens, activate mast cells and complement, and direct NK cells.

Neuroinflammation and PD Pathology

α-syn Aggregation

Some researchers have proposed that inflammation may trigger the development of α-syn pathology.⁹ The precise function of α-syn is unclear, and it can be phosphorylated and nitrated, which contributes to misfolding.¹⁰ Pathologic misfolding of α-syn is associated with microglial activation. When microglia are activated, they release free oxygen radicals and toxic proinflammatory cytokines, such as interleukin (IL)-1Β and tumor necrosis factor (TNF)-α, leading to neurotoxicity. More pathogenic variants of misfolded α-syn can be taken up by neurons and cause misfolding of endogenous α-syn.^11,12 Debate continues regarding whether α-syn aggregation is initiated in the gut or brain or both.

Microglial Activation

Microglia can be activated by misfolded α-syn and by other triggers, such as proinflammatory bacterial products. Oligomeric α-syn is an agonist for TLR2 and TLR4, pattern recognition receptors that detect pathogens and are integral in initiating both innate and adaptive immune responses.^13,14 Blocking α-syn and TLR2 binding with an antiTLR antibody prevents α-syn aggregation in neurons and astroglia as well as neuroinflammation, neurodegeneration, and behavioral deficits in a mouse model of PD.¹⁵

T-cell Activation and Dysregulation

T-cell−mediated dysfunction also plays an active role in neuronal cell death that depends on MHCII on microglia, used to present antigen to T cells. Mice missing MHCII are protected from α-syn-related toxicities.^16,17 Mice lacking functional T cells or with severe combined immunodeficiency disease (lacking both T and B cells) are protected against parkinsonism caused by neurotoxic 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine (MPTP)-induced neurodegeneration.² In humans with PD, T cells that recognize α-syn peptides have been found,¹⁸ with higher reactivity to α-syn than in people without PD.¹⁹ Stored blood from 1 individual who developed PD showed that α-syn−specific T cells were present 10 years before symptom development and diagnosis, which was which was similar to what was observed in a cohort analysis, with higher α-syn-specific T-cell reactivity in those who were more recently diagnosed with PD.¹⁹ The exact significance of α-syn-specific T cells is unknown. Other investigators have found decreased T_regs in people with PD. T_regs prevent inappropriate helper T-cell imbalances that can lead to a proinflammatory state and are thus critical for maintaining immune homeostasis.²⁰ The proinflammatory Th17 cell linked to autoimmunity, has also been suggested as having a role in PD pathogenesis. Th17 cells produce the proinflammatory cytokine, IL17A, which has a number of downstream effects, including promoting differentiation of B cells into plasma cells that produce antibodies.²¹

B Cells and Immunoglobulins

B cells play an important role in adaptive immunity and may also contribute to PD pathogenesis. T helper cells stimulate B cells to differentiate into plasma cells, which make immunoglobulins—seen in areas of neurodegeneration in brain tissue from people with PD. Over 20 years ago, purified immunoglobulin from people with PD that were injected into rat substantia nigra led to DA cell death, persistent perivascular inflammation, and microglia infiltration.²² In a small cohort of patients with PD, 12% had elevated antineuronal antibodies compared with 0% in healthy controls.²³ A number of autoantibodies (Table) have been associated with parkinsonism. Patients with these antibodies may respond to immunotherapy (eg, steroids, intravenous immunoglobulin [IVIG], and rituximab).^24,25 Although parkinsonian phenotypes of people with these antibodies are thought to differ from sporadic PD, the prevalence of these specific antibodies in the general population of patients diagnosed with sporadic PD is unknown.

Genes and Neuroinflammation

Changes in multiple immune function-related genes (eg, phosphatase and tensin homolog deleted on chromosome 10[PTEN]-induced kinase 1 [PINK1], DJ-1, leucine-rich repeat protein kinase-2 [LRRK2], and HLA-DR) increase the risk of developing PD through mechanisms involving neuroinflammation.²⁶PINK1 is a rare gene associated with early-onset PD that may link environmental risk factors, genetic predisposition, and autoimmunity via a mitochondria-related autoimmune mechanism.²⁷ Mice with homozygous PINK1 mutation (knock-out mice) do not show signs of parkinsonism alone, but researchers were able to show that infection with a specific gram-negative bacteria led to signs of parkinsonism in the mice. These PINK1-deficient mice displayed mitochondrial antigens that led to the development of a clone of CD8⁺ killer T cells that could cross the blood-brain barrier. When cocultured with DA neurons, the mitochondrial-specific autoreactive T cells led to neuronal death, suggesting PINK1 may serve an important function modulating mitochondrial antigen presentation to immune cells. When PINK1 is absent (eg, in the case of an inherited genetic mutation) and the mouse is challenged with a bacterial infection, those antigens initiate a T-cell mediated autoimmune reaction against DA cells in the brain. DA neurons may be particularly susceptible to this autoimmune attack because of their high concentration of MHCI (the complex that presents antigens to T cells and other immune cells).² This study demonstrates a model in which a gastrointestinal bacteria may lead to autoimmunity and symptoms of parkinsonism.

Along the same lines, parkin (PRKN) gene mutations, which are the most common cause of autosomal recessive early-onset PD, play a role in neuroinflammation and autoimmunity. Parkin is thought to help clear damaged mitochondria and to have an anti-inflammatory role during intracellular bacterial infections.²⁸ Infection by bacteria that have the cell surface molecule lipopolysaccharide (LPS) activates inflammasomes. Similarly, when α-syn is recognized by microglia, inflammasomes are activated by the microglia. Inflammasomes can trigger an inflammatory cascade leading to proinflammatory cytokine release, activate the adaptive immune system, and lead to dopamine cell death in PD patients. A particular type of inflammasome, NOD-, LRR- and pyrin domain-containing protein 3(NLRP3), inhibits the protective role of parkin. Inflammasome activation plays an important role in a number of other autoimmune diseases, and there is mounting evidence of its role in PD pathogenesis.²⁹

Preclinical data suggests that environmental factors may influence susceptibility to develop PD in at-risk populations through neuroinflammation. One of these factors, as mentioned, includes the presence of gram-negative bacteria in the gut. Gram-negative bacteria have an outer shell consisting partly of LPS , which can stimulate microglia to release neurotoxic factors.³⁰ Moreover, bacteria and LPS may translocate through a leaky gut in PD to trigger an inflammatory response and may worsen disease phenotype.^31,32 For example, germ-free mice displayed worse motor dysfunction when infected with microbiota of patients with PD.³³ As mentioned, patients with Crohn disease and ulcerative colitis have inflammatory bowel disease, and these conditions are associated with leaky gut.^34,35

Some variants of LRRK2 mutation are associated with increased risk of Crohn disease and PD.^36,37LRRK2 gene mutation is among the most common causes of autosomal dominantly inherited PD, and is implicated in neuronal and systemic inflammation. Immune cells express high levels of the LRRK2 enzyme, implicated in fighting bacterial infections as well as regulating the resident microbiota community in the gut.³⁶ These data further suggest that disruption of gut microbiota community (“dysbiosis”) might be a plausible trigger/enabler for neuroimmune/inflammatory state in PD. However, the pathophysiology directly showing LRRK2 gene mutation and the progression to PD is less well understod.^38-40

Cytokines

Several inflammatory cytokines, including IL-1Β and IL-6, have also been found to be elevated in the substantia nigra pars compacta and cerebrospinal fluid (CSF) of people with PD. There is also increasing preclinical evidence suggesting the proinflammatory cytokine, IL-17A, which is released by Th17 cells, may play a role in neuronal cell death in PD.⁴¹ This mechanism of autoimmunity has been demonstrated in patients with psoriasis, who on average have a 38% increased risk of developing PD,³⁴ and in patients with other immune conditions such as multiple sclerosis, inflammatory bowel disease, rheumatoid arthritis, and cardiovascular disease.⁴² Although less well studied than other markers of inflammation in PD, IL-17A is a pro-inflammatory cytokine that has been shown to contribute to neuronal cell death in a human induced pluripotent stem cell-based model of PD. Other peripheral markers of proinflammatory cytokines are present in higher levels in the blood of patients with PD compared with healthy controls, including IFNγ, TNFα, and interleukins IL1Β, IL2, IL6, RANTES, and IL10.^43-45

Conclusion

These studies show there are a number of complex mechanisms in which neuroinflammation is implicated in the pathogenesis of PD. Both the innate and adaptive immune systems are implicated. Activation of microglia and neuroinflammation by α-syn, and T cells, B cells, and immunoglobulins have all been found in brains from people who had PD. Mutations in multiple genes regulating the ability of mitochondria to present antigens are implicated in PD. Autoantibodies can cause parkinsonism, and some autoimmune diseases confer a higher risk for PD. Taken together, these data suggest pathogenesis of PD may be immune-mediated.

Key Points

• Neuroinflammation plays a vital role in causing neurotoxicity in patients with PD.
• Neuroinflammation involves, but is not limited to microglial activation, activation of the adaptive immune system, cytokine release.
• Genes and the environment play an important role in modulating and activating the immune system and contribute to neuroinflammation.
• Many potential targets to prevent neuroinflammation are being investigated (see Immunomodulators for Parkinson Disease in this issue).

Acknowledgments

We would like to thank Alan Landay, PhD at Rush University and the Parkinson’s Foundation for his generous contributions to editing this manuscript.