Neuroinflammation in Parkinson’s Disease: Triggers,Mechanisms,and Immunotherapies

α-Synuclein Aggregation and Neuroinflammation

Microglia are important innate immune cells that participate in the central nervous system’s first line of defense and function by mediating phagocytosis and inflammatory responses to maintain homeostasis. The aggregation of misfolded α-syn can activate microglia (Volpicelli-Daley and Brundin 2018; Zhang and others 2005). Overactivated microglia can activate other innate immune components in the central nervous system and induce peripheral adaptive immune responses, resulting in excessive damage to neurons (Harms and others 2013; Liddelow and others 2017; Waisman and Johann 2018). This phenomenon was originally identified in cell culture and animal models of PD. In a mouse model of PD in which human α-syn was overexpressed by a recombinant adeno-associated virus vector, serotype 2, it has been found that there was a prominent increase of CD68-positive microglia four weeks after injection of the recombinant adeno-associated virus vector, serotype 2 (AAV2), and human α-syn into the substantia nigra of mice (Theodore and others 2008). In in vitro cultured BV-2 microglia, treatment of extracellular α-syn can induce a positive NF-κB response and the production of pro-inflammatory cytokines (Couch and others 2011). In addition, a recent study confirmed the presence of α-syn-reactive T cells and showed that they were most abundant immediately after diagnosis of PD (Lindestam Arlehamn and others 2020). In 2019, a study demonstrated that pathologic α-syn trigged robust inflammatory activation of human monocytes from PD patients (Grozdanov and others 2019). Research on the molecular mechanism of microglial activation has made the following progress. (1) Membrane receptor pathway: α-syn can specifically bind to the TLR2, TLR4, and CD36 receptors on the surface of the microglial membrane, which activates the receptor-mediated inflammation signal pathway (Béraud and others 2011; Fellner and others 2013; Su and others 2008). (2) Intracellular signal transduction pathway: α-syn can increase the content of STAT3 and activate the JAK/STAT signal transduction pathway in microglia, which can induce the expression of MHC-II molecules and inflammatory genes (Qin and others 2016; Sarkar and others 2020). (3) Inflammasome pathway: α-syn can activate the NLRP3 inflammasome through the Fyn molecular pathway, leading to increased IL-1β release (Panicker and others 2019). (4 RNA binding protein pathway: α-syn can induce the expression of the RNA-binding protein Caspase8 and increase the release of IL-1β (Sarkar and others 2020).

Genetic Factors and Neuroinflammation

Familial forms of PD account for up to 5% to 10% of cases and usually result in early-onset disease. Common PD-linked genes include LRRK2, SNCA, VPS35, PRKN, DJI, PINK1, and PARK7, among which LRRK2, VPS35, PRKN, and PINK1 play important roles in the immune system (Deng and others 2018). LRRK2 is not only a common gene in autosomal dominant PD but is related to sporadic PD (Foo and others 2016). LRRK2 is highly expressed in peripheral blood mononuclear cells (Hongge and others 2015). Previous studies have found that the LRRK2 affected microglial function in different ways. By transcriptomics analysis, LRRK2 knock-out microglia cells show attenuated induction of mitochondrial SOD2 in response to α-syn pro-formed fibrils, indicating that microglial LRRK2 may contribute to the pathogenesis of PD via altered oxidative stress signaling (Russo and others 2019). In a Mn exposure animal model, activated microglia can release multiple inflammatory cytokines and the expression of LRRK2 is upregulated. Furthermore, inhibition of LRRK2 can decrease the expression of inflammatory cytokines and recover autophagic function of microglia (Chen and others 2018). In addition, recent studies have proposed a role for LRRK2 in vesicular trafficking, playing a crucial role in the autophagy/lysosomal pathway, and essential for normal lysosomal function (Cookson 2016). LRRK2 can combine with the intracellular sorting protein VPS35 to regulate the phosphorylation of RabGTPases, which can mediate the phagocytosis, extracellular secretion, and autophagy functions of immune cells by regulating vesicle transport (Beilina and others 2014; Steger and others 2016). In turn, autophagy influences inflammation through the transport of degradable material to the lysosome (Kluss and others 2019). Therefore, LRRK2 mutations may lead to inflammatory reactions by influencing intracellular trafficking (Henry and others 2015). PRKN and PINK1 are also common genes in PD. A study has demonstrated that acute and chronic mitochondrial stress in vivo leads to a STING-meditated type I interferon response in mice absence of parkin or PINK1. Additionally, elevated cytokine levels in serum of asymptomatic PRKN mutation carriers have been found, indicating that parkin and PINK1 prevent inflammation by clearing damaged mitochondria (Sliter and others 2018). The co-expression of parkin and PINK1 can inhibit antigen presentation of MHC-I molecules, which can activate CD8+ T cells. Mutation of both genes can lead to excessive activation of immune cells (Matheoud and others 2016). The above studies reveal that LRRK2, PRKN, and PINK1 play important roles in the immune system.

In recent years, genome-wide association studies (GWAS) have found that polymorphisms of certain gene loci increase the susceptibility of sporadic PD. In 2017, a GWAS identified 17 novel PD loci and found that several of newly identified PD risk genes play a role in lysosomal biology and autophagy (Chang and others 2017). To further provide the most comprehensive survey of genetic risk within PD, a GWAS in 2019 identified 90 independent common genetic risk factors for PD, nearly doubling the number of known risk variants. Additionally, it also identified intracranial and putaminal volume as potential future PD biomarkers, and cognitive performance as a PD risk factor (Nalls and others 2019). Most gene loci are related to autoimmunity, and polymorphisms of HLA alleles are closely related to the risk of PD (Aliseychik and others 2018; Foo and others 2017). It has been reported that polymorphisms at multiple single nucleotide sites in the HLA class II gene locus increase the genetic susceptibility of PD, including the rs3129882, rs75855844, rs2395163, rs660895, and rs4248166 sites (Wissemann and others 2013). A study in 2010 found that the rs3129882 locus polymorphism was associated with the risk of PD. Located in the noncoding region of the HLA-DRA gene, this polymorphism can regulate expression of the HLA-DR and HLA-DQ genes (Hamza and others 2010). Meta-analysis of genome-wide sequencing studies showed that a polymorphism at the rs75855844 locus in the HLA-DRB5 gene was associated with PD (Nalls and others 2011). Another meta-analysis in 2012 reported that the rs660895 polymorphism in the HLA-DRB1 gene was a protective factor for PD (Ahmed and others 2012). MHC-II molecules, encoded by HLA genes, are expressed on the surface of a variety of activated immune cells and recognize epitopes, which can induce adaptive immune responses by antigen presentation (Aliseychik and others 2018). The replacement of single nucleotides in the coding region of the HLA-DR, HLA-DQ, and HLA-DP genes can cause dysfunction of the MHC-II molecules (Hammer and others 1993). HLA class II gene expression depends on the regulatory sequence of the noncoding region, which may affect the expression of multiple structural genes at the same time (Choi and others 2011). Therefore, changes in the coding and non-coding regions of HLA class II genes affect the function of MHC-II molecules. MHC-II molecules in the central nervous system are mainly derived from microglia. Astrocytes and vascular endothelial cells also show low levels of expression (Hayes and others 1987). Activated microglia express MHC-II molecules, which can induce adaptive immune cascades and damage neurons by activating T cells (Depboylu and others 2012; Williams and others 2018). The above-mentioned studies show that polymorphisms in the HLA gene locus can regulate the expression of MHC-II molecules on the surface of microglia and induce a neuroinflammatory response in PD.

Mitochondrial Dysfunction and Neuroinflammation

Mitochondrial dysfunction and elevated oxidative stress levels have been detected in the SN of PD patients and animal models (Chiu and others 2019; Licker and others 2014; Mishra and others 2019). In LRRK2(G2019S) mutant patient tissue, mitochondrial membrane potential and total intracellular ATP levels decreased, and mitochondrial elongation and interconnectivity increased, indicating that LRRK2 mutant may affect mitochondrial function and morphology (Mortiboys and others 2010). DJ1 protein exists in the mitochondrial matrix and intermembrane cavity. Deletion of DJ1 leads to the disruption of mitochondrial dynamics and cell apoptosis (Irrcher and others 2010). PINK1 mutation can disrupt mitochondrial morphology and function (Clark and others 2006). Aggregation of α-syn can inhibit the function of proteasome complex 1, disrupt the calcium balance in the mitochondrial matrix, promote mitochondrial division, inhibit mitochondrial fusion, and eventually cause mitochondrial dysfunction (Devi and others 2008; Kamp and others 2010; Luth and others 2014). After mitochondrial function is compromised, a large number of reactive oxygen species, mitochondrial fragments, and mitochondrial DNA (mtDNA) fragments are produced. Reactive oxygen species are a major pro-inflammatory stimulus via the activation of nuclear factor κB (NF-κB; Schreck and others 1991). mtDNA fragments can also trigger inflammation by interacting with toll-like receptors (TLRs), nucleotide-binding oligomerization domain (NOD)-like receptor family pyrin domain containing 3 (NLRP3) inflammasomes, and the cytosolic cyclic GMP/AMP synthase (cGAS)-stimulator of interferon genes (STING) DNA-sensing system (Zhang and others 2010; Zhou and others 2011). Once these substances are released, they may act as damage-associated molecular patterns (DAMPs), triggering an innate immune inflammatory response by binding to danger signal receptors, in particular, mtDNA fragments (Grazioli and Pugin 2018). Mitochondria are the only organelles that contain their own genetic material. The proteins encoded by mtDNA form four proteasome complexes located on the inner membrane of mitochondria, which are responsible for transferring electrons and synthesizing ATP. Damage to mtDNA and proteasome complexes can cause mitochondrial dysfunction. The incidence of mtDNA mutations, deletions, cloning, and rearrangements in PD is significantly higher than in normal controls, indicating that the mtDNA in PD is damaged (Bender and others 2006; Bury and others 2017). Therefore, mitochondrial dysfunction can induce a neuroinflammatory response.

Gut Microbiome Dysbiosis and Neuroinflammation

The gut microbiome (GM) plays a critical role in microglia maturation, differentiation, and function. In 2015, a study showed that colonization with a complex microbiota modulated microglia activation and maturation status. But in the absence of intestinal microbes, the innate immune response of microglia was diminished. In germ-free (GF) mice, supplementation with short-chain fatty acids, microbial metabolites could rescue impaired microglia maturation. So gut microbiota and metabolites are crucial for microglia maturation and activation (Erny and others 2015). Gut microbiome dysbiosis, chronic inflammation of the intestinal mucosa, and accumulation of α-syn in the enteric nervous system have been confirmed in PD patients (Kim and others 2019; Lin and others 2019; Parashar and Udayabanu 2017; Schwiertz and others 2018). The mucosal and fecal microbial community of PD patients is different than control subjects, showing proinflammatory dysbiosis (Keshavarzian and others 2015). Similarly, a study in 2020 found compositional and metabolic alterations in the Parkinson’s microbiota (Cirstea and others 2020). In the α-syn-overexpressing mouse mode, colonization with microbiota from PD patients induces enhanced motor dysfunction compared to microbiota transplants from healthy human donors. Administration of specific microbial metabolites to germ-free mice promotes neuroinflammation and motor symptoms. So signals from gut microbes are required for the neuroinflammatory responses in a model of PD (Sampson and others 2016). Therefore, gut microbiome dysbiosis can induce a neuroinflammatory response (see Figure 1).

OPEN IN VIEWER

Figure 1.

Activation of microglia. Misfolded α-syn specifically binds to membrane receptors on the surface of microglia, which activates the NLRP3 inflammasome through the Fyn molecular, leading to increased IL-1β release. Polymorphisms in the HLA class II gene locus affect the function of MHC-II molecules and antigen presentation. Mutations in PRKN, PINK1, and α-syn cause mitochondrial dysfunction. Damaged mitochondria produces reactive oxygen species (ROS) and mitochondrial DNA (mtDNA) fragments, activating the NLRP3 inflammasome. Microbial metabolites produced by gut microbiome modulate microglia activation and maturation.

Central Innate Immune Cells

Excessive activation of microglia has been detected by PET imaging and autopsy in PD patients (Gerhard and others 2006; McGeer and others 1988). Activated microglia damage neurons through the following mechanisms: (1) ingesting damaged neurons and degrading them directly, leading to neuronal death (Morsch and others 2015; Neniskyte and others 2014); (2) secreting a large number of proinflammatory factors, such as IL-1β, TNF, and IFN, which amplify local inflammation (Panicker and others 2019); (3) transforming astrocytes into a toxic phenotype, damaging neurons (Yun and others 2018); (4) expressing MHC-II molecules for antigen presentation, activating T cells, and inducing adaptive immune responses to damage neurons (Williams and others 2018); and (5) phagocytizing aggregated α-syn and promoting its spread to healthy neurons (Guo and others 2020; Jimenez-Ferrer and others 2021). Normal astrocytes can secrete a variety of neurotrophic factors to protect neurons from damage and promote the repair of damaged neurons (Bylicky and others 2018). IL-1β, TNF, and complement C1q secreted by activated microglia can transform astrocytes into toxic phenotypes, causing them to lose their neuroprotective effects and inducing damage to neurons (Shao and others 2013). In the brain, mast cells (MCs) reside on the brain side of the BBB and in the leptomeninges (Florenzano and Bentivoglio 2000; Guo and others 2020). MCs are considered first responders in the neuroinflammation-initiating events and are able to magnify immune responses in the brain (Wang and others 2020). The cross-talk between neurons, glial cells, and mast cells, which is regulated by neuroinflammatory cytokines and chemokines, is involved in the neurodegeneration in PD (Kempuraj and others 2018). Activated mast cells are able to degranulate, releasing vasoactive mediators and matrix degrading molecules such as proteases, which leads to increased permeability of the BBB and helps the migration of peripherally activated immune components to the central nervous system (Kempuraj and others 2020; Wang and others 2020; see Figure 2).

OPEN IN VIEWER

Figure 2.

The interaction of central innate immune cells. IL-1β, TNF, and complement C1q secreted by activated microglia transform neuroprotective astrocytes into neurotoxic phenotypes, inducing neuronal death. Activated mast cells are able to degranulate and release vasoactive mediators and matrix degrading molecules, such as histamine, TNF-α, and proteases, which lead to increased permeability of the blood-brain barrier.

Peripheral Immune Cells

The number of monocytes expressing MHC-II molecules and activated T cells in the CSF of PD patients is significantly increased (Fiszer and others 1994; Schröder and others 2018). In an animal model overexpressing α-syn, a large number of CCR2+ monocytes migrated to the SN region of the brain, and the deletion of CCR2 prevented their migration and subsequent death of dopaminergic neurons (Harms and others 2018). In autopsy samples of PD patients, the number of CD4+T cells and CD8+T cells in the SN is significantly increased, especially in the vicinity of blood vessels and dopaminergic neurons expressing neuromelatonin (Brochard and others 2009; Sommer and others 2018). The above studies have revealed that monocytes and T cells in the peripheral blood of PD can migrate to specific areas of the brain through the damaged BBB during acute and chronic inflammation, indicating that peripheral immune cells are also involved in the pathogenesis of PD.

Monocyte

Monocytes are peripheral myeloid-lineage cells that function like microglia. A recent study showed that soluble CD163, a monocyte/macrophage-specific biomarker, increases in the CSF of PD patients, suggesting increasing monocytic activation. Interestingly, in vitro studies suggested that α-syn can induce macrophage activation and sCD163 shedding, while sCD163 might enhance α-syn uptake by myeloid cells and participate in the clearance of α-syn (Nissen and others 2020). In addition, monocytes that migrate into the brain tissue also show an antigen-presenting effect by expressing MHC-II molecules, which can activate T cells together with microglia and induce neuroinflammatory responses (Witoelar and others 2017). Mutations in glucocerebrosidase (GBA1) impair the activity of the encoded lysosomal lipid metabolism enzyme (GCase), which are linked to an increased risk of PD. GCase is highly expressed in monocytes. In the PD patients even without GBA1 mutations, GCase activity is also significantly reduced in monocytes (Atashrazm and others 2018). In 2019, a study revealed that PD involves monocytic changes in blood, and these cells show reduced viability and are unresponsive to specific stimuli (Nissen and others 2019).

T Cell

CD4+ T cells that migrate into the brain can proliferate and differentiate into Th1 and Th17 cells after being activated by MHC-II molecules (Iba and others 2020; Rostami and others 2020). Th1 cells secrete proinflammatory factors, including IFN-γ, TNF-α, and IL-2. These proinflammatory factors can positively activate microglia and expand the central innate immune response, indirectly damaging neurons (Iba and others 2020). Th17 cells are a subgroup of T cells that have been studied extensively in recent years. Th17 cells can directly damage neurons, although the specific mechanism is unknown. In the peripheral blood of PD patients and autopsy, the number of Th17 cells is significantly increased, indicating that Th17 cells are involved in the pathogenesis of PD (Sommer and others 2018). In MPTP-induced animal models, we found that Th17 cells migrated from the periphery to the brain tissue. In vitro culture of midbrain neurons induced and differentiated by PD pluripotent stem cells revealed that IL-17A levels and neuronal death were significantly increased after adding autologous Th17 cells (Liu and others 2017a), indicating that Th17 cells may damage neurons by producing IL-17A. CD8+ T cells show cytotoxic effects and induce apoptosis. CD8+ T cells in brain tissue can be activated by MHC-I molecules on the surface of dopaminergic neurons, which directly leads to neuronal death (Cebrián and others 2014).

In summary, in PD patients, central innate immune cells activated by α-syn and damaged neurons can not only directly damage neurons and promote the spread of α-syn but also activate T cells that migrate from the periphery to the center through antigen presentation. Activated T cells can release proinflammatory factors to positively activate central innate immune cells. Injured neurons and increased α-syn levels in the process of neuroinflammation can also positively activate central innate immune cells. It is apparent that after the immune system is activated, two positive feedback loops are formed in PD. One includes central innate immune cells and peripheral adaptive immune cells, and the other includes central innate immune cells and damaged neurons. These two positive feedback loops can cause the neuroinflammation to be in an excessive and long-lasting activation state in PD, which leads to the continuous death of dopaminergic neurons and other vulnerable neurons, promoting the progression of PD (see Figure 3).

OPEN IN VIEWER

Figure 3.

The formation of two positive feedback loops. Activated microglia express MHC-II molecules, and then activate CD4+ T cells by antigen presentation. After being activated, CD4+ T cells proliferate and differentiate into Th1 cells and Th17 cells. Th1 cells secrete proinflammatory factors, which positively activate microglia and expand the central innate immune response. Activated microglia secrete a large number of reactive oxygen species to damage neurons. Damaged neurons produce cell debris and release misfolded α-syn, which can positively activate microglia. TH17 cells secrete IL-17 and express intercellular adhesion molecule LFA-1. CD8+ T cells can be activated by MHC-I molecules on the surface of neurons. Both can directly damage neurons.

The Interaction between the Central and Peripheral Immune Systems

Due to the existence of the BBB, the central nervous system has long been considered a place for the body’s immune escape, and immune research on the central nervous system is limited to the central innate immune cells. In 2012, a study found a glial lymphatic vascular pathway in the brain of rodents. This pathway is composed of three anatomical structures, including the periarterial space, the perivenous space, and the brain tissue space, which finally drain to the cervical lymph nodes (Iliff and others 2012). Subsequent studies on CSF imaging and PET imaging have also confirmed the existence of this pathway in the human brain, known as the meningeal lymphatic system (de Leon and others 2017; Ringstad and others 2017). This anatomical structure provides a pathway for the central and peripheral immune systems to communicate with each other. In addition, chronic inflammation in the central nervous system can damage the integrity of the BBB and promote peripheral immune cells and inflammatory factors to migrate to the central nervous system (Kortekaas and others 2005; Louveau and others 2015). As mentioned above, the migration of peripheral immune cells to the central nervous system has been demonstrated in PD patients. The possible mechanisms are as follows. (1) Mast cells interaction with glial cells and neurons during neuroinflammation has been demonstrated in PD animal models (Kempuraj and others 2018). Activated mast cells can release preformed mediators such as histamine, TNF-α, and proteases. Proteases can degrade tight junction proteins and extracellular matrix components, influencing BBB integrity (Wang and others 2020). TNF-α can downregulate the expression of interendothelial adherens and tight junction proteins, leading to elevation of paracellular permeability (Rochfort and others 2014). (2) The formation of new blood vessels leads to increased permeability of the BBB. The levels of vascular growth factors, such as vascular endothelial growth factor (VEGF) and placental growth factor (PIGF), increase in the CSF of PD patients, which are positively correlated with the levels of proinflammatory factors secreted by microglia (Janelidze and others 2015). As shown by PD autopsy, the level of VEGF, the number of vascular endothelial cell nuclei, and new blood vessels in the substantia nigra increased (Faucheux and others 1999; Wada and others 2006). In 2015, a study showed a higher CSF/plasma albumin ratio in PD patients, indicating dysfunction of the BBB. Furthermore, the researchers observed a positive correlation between the level of VEGF and the CSF/plasma albumin ratio, indicating that VEGF might be a potent trigger of vascular leakage (Janelidze and others 2015). Therefore, chronic inflammation in the central nervous system may promote the formation of new blood vessels by increasing levels of angiogenesis factors, leading to increased permeability of the BBB. The meningeal lymphatic system and the damaged BBB provide pathways for the interaction between the central and peripheral immune systems. So the central and peripheral immune systems, to some extent, are connected, but studies on whether neuroinflammation in the central and peripheral nervous system in PD are consistent are yet to provide sufficient results.

The Different Immune Inflammatory Responses between the Central and Peripheral Nervous Systems

PD is a degenerative disease of the central nervous system, and CSF is the specimen that best reflects the neuroinflammation changes in PD. However, clinical CSF specimens are difficult to obtain. Previous studies have implied that central and peripheral immunity are interconnected. Currently, most studies on neuroinflammation in PD use peripheral blood as the specimen. A recent meta-analysis reports increased IL-1β, TNF-α, IL-6, and IL-10 in the blood of patients with PD, whereas in CSF, IL-1β, IL-6, transforming growth factor (TGF)-β1, and C-reactive protein (CRP) are elevated (King and Thomas 2017). In order to verify whether neuroinflammation in the central and peripheral nervous system of PD patients is consistent, a study in 2020 tested the levels of inflammatory factors in the CSF and peripheral blood of the same research subject and found that the values of five inflammatory factors were different. The level of IL-8 was higher in the CSF than in peripheral blood, while the levels of IFN-γ, IL-6, IL-10, and TNF-α were lower in the CSF than in peripheral blood (Wijeyekoon and others 2020). A subsequent study measured 40 immune-related serum and CSF biomarkers. In CSF, VCAM-1 (vascular cell adhesion molecule 1), ICAM-1 (intercellular adhesion molecule 1), IL-8, SAA (serum amyloid A), PIGF (placental growth factor), IL-15, and VEGF-D (vascular endothelial growth factor-D) increased in patients with PD. In serum of PD patients, VCAM-1 increased, whereas VEGF-C decreased (Nissen and others 2020). These changes are different in CSF versus serum, supporting a distinctive immune profile in the brain and periphery. The possible reasons are as follows: (1) The central and peripheral immune cells are different. Microglia are unique, innate immune cells in the central nervous system, which play an important role in the neuroinflammatory response, but they do not exist in the periphery. (2) The migration of peripheral immune cells to the central nervous system is selective, not random, penetration. The migration of peripheral monocytes and T cells to the central nervous system is completed under the co-induction of chemokines, vascular endothelial cell adhesion molecules, and MHC-II molecules, mainly appearing around the blood vessels and damaged neurons in the SN (Cose and others 2006). (3) The central and peripheral immune microenvironments are different. The activation of immune cells is completed under the induction of the immune microenvironment. Different immune microenvironments can lead to different immune inflammatory response states. Peripheral immune cells can reach the organs of the whole body via blood circulation. Metabolites, damaged cells, and foreign antigens of each tissue can activate peripheral immune cells, while the BBB in the central nervous system can block peripheral antigens from entering the center, protecting the central immune cells from being affected (Becher and others 2017). Therefore, the existence of the BBB causes the immune microenvironment to differ between the central and peripheral nervous systems, resulting in inconsistent inflammatory responses.

Immunotherapy Targeting α-Syn

The aggregation of α-syn has become a potential treatment target. A large number of monoclonal antibodies that specifically bind to different epitopes of α-syn have been developed and tested in animal models (Games and others 2014; Kim and others 2018). In 2014, an experiment on a transgenic mouse model of PD found that after being injected with a monoclonal antibody that targets the C-terminus of α-syn, the level of C-terminus truncation in the mouse decreased significantly and the motor symptoms were improved (Games and others 2014). With the support of preclinical experimental data, in 2018, the first clinical trial of an intravenous monoclonal antibody for the treatment of PD was initiated, whose main purpose was to evaluate the safety and tolerability of the drug. This trial included 80 early PD patients who were injected with different doses of PRX002/RG7935 (α-syn monoclonal antibody). Each dose showed good safety and tolerability, and the amount of α-syn monomer in the peripheral blood of subjects was significantly reduced (Jankovic and others 2018). BIIB054 is a monoclonal antibody that targets the N-terminus of α-syn. In the initial clinical trial, 18 PD patients were enrolled. The 16-week follow-up results showed that the drug was well tolerated and no serious adverse events were observed. The second phase of this study is currently undergoing a multicenter clinical trial (Brys and others 2019; Weihofen and others 2019). MEDI1341 is a monoclonal antibody that targets the C-terminus of α-syn. It is currently under research and clinical trial data have not yet been reported (Jankovic 2019). All of the above clinical trials involve passive immunization, and active immunization methods are also being explored. In two transgenic animal models of synuclein disease, inoculation with AFFITOPEs (short peptides) can induce the production of anti-α-syn antibodies. These antibodies can bind to the aggregated α-syn and improve the animal’s behavioral performance (Mandler and others 2014; Mandler and others 2015), but currently there are no reports of active immunization in PD animal models.

Immunotherapy Targeting Microglia

Previous studies have suggested that microglia can differentiate into different subtypes after activation. The M1 subtype is a proinflammatory phenotype that can secrete a large number of proinflammatory factors. The M2 subtype is an anti-inflammatory phenotype that can secrete anti-inflammatory factors and promote the repair of damaged neurons (Orihuela and others 2016). In PD patients, the differentiation of microglia into the M1 subtype increases (Du and others 2017). The oversimplified “polarization” designation of microglia is an attempt to simplify data interpretation. Of note is that microglial polarization has not been established by research findings. Now, the microglial polarization paradigm hinders rather than aids research progress and should be discarded (Ransohoff 2016). Current research is focusing on the characterization of activated microglia for discovering new drug target.

TLR2 and TLR4 are membrane receptors expressed on the surface of activated microglia, which can mediate proinflammatory responses (Hughes and others 2019; Kim and others 2013). Candesartan cilexetil can inhibit the expression of TLR2 and TLR4, reversing the proinflammatory phenotype of microglia (Daniele and others 2015; Dasu and others 2009). The antibiotic rifampin and its autoxidation product rifampicin quinone can inhibit the proinflammatory response mediated by TLR2 (Acuña and others 2019); TAK-242 and RSLA are small molecule antagonists of TLR4, which can inhibit the activation of TLR4 and reduce the production of reactive oxygen species and proinflammatory factors by microglia in vitro (Hughes and others 2019). However, studies on these drugs are limited to in vitro cell culture, and clinical trials have not yet started. CB2 receptors are membrane receptors of the endocannabinoid system (Navarrete and others 2018). The activation of CB2 receptors on the surface of microglia can inhibit the production of proinflammatory factors (Javed and others 2016). In MPTP-induced PD animal models, after administration of JWH133 (selective CB2 receptor agonist), the levels of reactive oxygen species and proinflammatory factors produced by microglia are reduced and the migration of peripheral immune cells to the central nervous system is reduced (Chung and others 2016). Therefore, CB2 receptors may become an effective target for inhibiting microglial activation. The JAK/STAT and NF-κB pathways in microglia are classic inflammatory signal transduction pathways that can maintain the M1 phenotype after activation (Hu and Ivashkiv 2009; Takeda and Akira 2004). Tanshinone-I and α-asarone can selectively inhibit the NF-κB pathway. In animal models of PD, administration of these two drugs can inhibit the differentiation of the M1 subtype and reduce the levels of proinflammatory factors (Kim and others 2015; Wang and others 2015). The activation of histamine receptor 4 (H4R) and the NLRP3 inflammasome promotes the secretion of proinflammatory factors by microglia. In animal models of PD, injection of JNJ7777120 (an H4R antagonist) into the lateral ventricle and administration of MCC950 (a small molecule antagonist of NLRP3) can inhibit the activation of microglia, reduce damage to dopaminergic neurons, and improve animal behavior (Gordon and others 2018; Zhou and others 2019). Vitamin D is a common biological agent. Vitamin D can reduce the secretion of proinflammatory factors and increase the production of anti-inflammatory factors, including IL-10 and TGF-β, which is conducive to the maintenance of the M2 phenotype. These effects have been demonstrated in animal models of PD (Kim and others 2006). Peroxisome proliferation-activated receptors (PPARs) are nuclear receptors that can inhibit the activation of microglia, reduce the neuroinflammatory response, and exert a neuroprotective effect (Agarwal and others 2017). Pioglitazone and rosiglitazone are PPAR-γ agonists, and their anti-inflammatory effects have been confirmed in animal models of PD (Dehmer and others 2004; Swanson and others 2011). However, the results of clinical trials are controversial (Brakedal and others 2017; Brauer and others 2015; Connolly and others 2015). The above-mentioned studies show that there are a number of molecular targets that can regulate the activation of microglia, but more clinical trials are needed to explore these effects.

Immunotherapy Targeting T Cells

T cells are the main cells of the peripheral adaptive immune response. They are activated in PD and migrate to the central nervous system to participate in neuroinflammation-mediated neuronal damage (Brochard and others 2009). The differentiation of subpopulations of T cells in PD is altered. The ratio of regulatory T cells (Treg) reduces (Baba and others 2005), which leads to excessive activation of T cells. Therefore, regulating the differentiation of T cells may delay the progression of PD.

Treg cells can release anti-inflammatory factors IL-10 and TGF-β, inhibit antigen presentation, and induce immune tolerance, which inhibit T cell-mediated adaptive immune response (Benner and others 2004; Reynolds and others 2009). In PD patients, a decrease in the number of Treg cells leads to excessive activation of T cells. Therefore, inducing the differentiation of Treg subgroups may have a neuroprotective effect. GM-CSF is a granulocyte macrophage colony-stimulating factor. An appropriate dose of GM-CSF can increase the number of Treg cells and induce T cell immune tolerance (Bhattacharya and others 2015). The anti-inflammatory effect has been confirmed in animal models of PD (Kosloski and others 2013; Schutt and others 2018), and the safety and effectiveness have also been confirmed in clinical trials (Gendelman and others 2017). Anti-CD3mAbs and the neuropeptide vasoactive intestinal peptide (VIP) can also induce the differentiation of Treg subgroups and reduce neuroinflammation (Delgado and others 2005; Penaranda and others 2011). Both have shown expected effects in the study of PD animal models (Kuhn and Weiner 2016; Olson and others 2016), but currently there are no clinical trial reports. CD4+ T cells can be activated by antigens presented by MHC-II molecules and differentiate into TH1 effector cells to exert a proinflammatory effect. However, this activation process requires the binding of costimulatory molecules CD28 (expressed on the surface of CD4+ T cells) and CD80/CD86 (expressed on the surface of antigen-presenting cells), and blocking the combination of both can inhibit the activation of CD4+ T cells (Greenfield and others 1998; Linsley and others 1992; Prasad and others 1994). Abatacept and belatacept are synthetic CTLA-4 analogs that can bind to CD80/CD86 molecules. With higher affinity than CD28, they can inhibit the activation of CD4+ T cells (Korhonen and Moilanen 2009; Larsen and others 2005; Linsley and others 1991). These two drugs have shown expected effects in the treatment of autoimmune diseases and organ transplantation (Goldzweig and Hashkes 2011; Kirk and others 1997; Vincenti and others 2010), and they may also be effective in immunotherapy for PD. TH17 cells belong to a subset of effector T cells and have neurotoxic effects. TH17 cells can secrete IL-17, which causes neuronal damage by combining with IL-17R expressed on the surface of neurons. In addition, it can express the intercellular adhesion molecule LFA-1, which binds to ICAM-1 to damage neurons (Liu and others 2017a; Sommer and others 2018). Blocking the LFA-1 molecule on the surface of TH17 cells or IL-17R, ICAM-1 molecules on the surface of neurons may reduce neuronal damage. TH17 cells are expected to become a potential target for PD immunotherapy (see Table 1).

OPEN IN VIEWER

Table 1.

Studies of Immunotherapy in Parkinson Disease.

Target	Compound	Mechanism	Phase	Findings
α-Synuclein	PRX002/RG7935	Targeting a carboxy-terminal epitope of α-synuclein	I	Completed; safe and tolerable, robust binding to peripheral α-synuclein
	BIIB054	Targeting amino terminus of α-synuclein	II	Completed; safe and tolerable in healthy controls and patients with PD
	MEDI1341	Decreasing extracellular α-synuclein level	I	Recruiting
Microglia	Candesartan cilexetil	Inhibiting the expression of TLR2 and TLR4	Preclinical	Reversing the pro-inflammatory phenotype of microglia
	TAK-242/RSLA	Small molecule antagonists of TLR4	Preclinical	Attenuating oxidative stress, TNF-α production by microglia
	JWH133	Selective CB2 receptor agonist	Preclinical	Reducing production of proinflammatory cytokines by the microglia
	Tanshinone I	Inhibiting the activation of NF-κB signaling	Preclinical	Reserving the expression of anti-inflammatory markers, preventing dopaminergic neurodegeneration
	JNJ7777120	Histamine-4 receptor antagonist	Preclinical	Suppressing activation of microglia, reducing Lewy body-like neuropathology, improving motor behaviors in PD models
	MCC950	A small-molecule inhibitor of NLRP3 inflammasome	Preclinical	Preventing dopaminergic degeneration and α-syn pathology, and alleviating motor behaviors in PD models
T cells	GM-CSF	A cytokine that act as anti-inflammatory modulators depending on the dose	Ib	In clinical trials, GM-CSF was safe and well-tolerated, motor functions of PD patients were improved. A Phase Ib clinical trial is ongoing.
	LBT3627	A selective VIPR2 agonist	Preclinical	Creating an anti-inflammatory micro-environment, altering Th1/Th17 cytokine responses in animal models of PD
	Belatacept	Blocking the ligation of CD28 with CD80/CD86	Preclinical	It has been approved in the USA for use in kidney transplantation, but it has not been reported in the treatment of PD