Parkinson’s disease (PD) is an age-related, multifactorial disorder, resulting in the demise of dopaminergic neurons in the substantia nigra pars compacta (SNpc) of the midbrain, which subsequently leads to motor difficulties, tremor, and postural instability in affected individuals (Pang et al., 2019). While there is a genetic component to the disease, with 10% of all cases carrying a mutation in one of the causal PD genes, 90% of patients are deemed idiopathic.

The majority of studies published to date describe molecular mechanisms centered around α-synuclein aggregation, mitochondrial dysfunction, dysregulated autophagy flux, and neuroinflammation as the underlying causes of PD (Wang et al., 2015; Maiti et al., 2017). Interestingly, all of these processes are also affected by aging, which leads to functional decline, both at the physiological and molecular level. Thus, it is not surprising that aging is considered a major risk factor for the development of PD (Jin et al., 2020).

Additionally, “inflammaging” is a novel term coined to define basal, low-level inflammation during adult life that, with time, turns into a destructive, pathological process. On the one hand, lower levels of inflammation are considered to have a positive outcome on the overall cellular state. On the other hand, during prolonged inflammation, beneficial mechanisms of defense start to wear off while damaging insults increase. This phenomenon might explain why seemingly low-grade inflammatory occurrences can have a significant negative impact on health in older individuals (Calabrese et al., 2018).

Inflammation is one of the hallmarks of PD and it is propagated mostly through microglia cells, which are responsible for the innate immune defense of the brain. Early brain tissue studies showed an upregulation of microglial cells in the SNpc and higher expression of human major histocompatibility complex class II (MHC-II) molecules, while in human serum and cerebrospinal fluid (CSF), increased concentrations of cytokines such as IL-1β, IL-6, TNF-α, IL-2, IL-18, and, IL-10 were detected (Nagatsu and Sawada, 2005; Brodacki et al., 2008; Long-Smith et al., 2009; Collins et al., 2012; Wang et al., 2015; Badanjak et al., 2021). In line with these results, our own immunohistochemistry and single-nuclei transcriptomic analyses in postmortem midbrain tissue revealed an increase in abundance and a decrease in the complexity of microglia in IPD tissue, suggestive of an activated state. Moreover, patient microglia presented a disease-specific gene expression signature, indicating a significant role of these cells in the pathogenesis of the movement disorder (Smajić et al., 2020). Also, most recently, genes of the IFN-γ signaling pathway were found to be dysregulated in IPD patients (Magalhaes et al., 2021).

One of the most commonly implicated inflammatory pathways in PD is the inflammasome pathway (Chao et al., 2014; Sebastian-Valverde and Pasinetti, 2020; Yan et al., 2020). The NOD-, LRR-, and pyrin domain-containing protein 3 (NLRP3) is by far the most studied inflammasome, the main function of which is the clearance of pathogens. NLRP3 is a cytosolic sensor of intracellular and extracellular stimuli such as damage-associated and pathogen-associated molecular patterns (DAMPs and PAMPs, respectively). Two signals are necessary to fully activate this pathway, a priming signal and an activation signal. The priming signal is characterized by the upregulation of IL1B and NLRP3 expression, while the secondary signal is characterized by the release of mature cytokines (Swanson et al., 2019). Underlining the relevance of the NLRP3 inflammasome, IL-1β has been associated with disease pathogenesis in multiple PD biomarker studies (Koprich et al., 2008; Su et al., 2008; Nakahira et al., 2011; Gillardon et al., 2012; Pike et al., 2021).

In our current study, we investigated inflammation markers in different models of IPD. First, we explored available RNAseq transcriptomic datasets from postmortem midbrain tissues to assess the expression of key cytokines in IPD. Next, we differentiated microglia from iPSC from IPD and control donors to test whether these cells can mirror the phenotypes observed in the brain. We detected elevated levels of IL1B and IL10 in whole tissue or single cell RNAseq datasets from IPD nigral or midbrain sections. Indicative of the fidelity of iPSC-derived cellular PD models, both IL1B and IL10 were also upregulated in patient microglia upon lipopolysaccharide (LPS) treatment. This coincided with increased protein abundance of NLRP3 in these cells, further implicating the inflammasome in the pathogenesis of PD.

Although the majority of PD cases suffer from the idiopathic form of the movement disorder, the cause of neurodegeneration in these individuals has not been extensively investigated. Genome wide association studies (GWAS) identified over 40 PD risk loci, the majority of which overlaps with known autosomal dominant PD genes, most notably SNCA and LRRK2, while other studies revealed the presence of heterozygous variants in autosomal recessively inherited PD genes (Simón-Sánchez et al., 2009; Nalls et al., 2014; Zeng et al., 2018; Germer et al., 2019; Lai et al., 2020). The main difficulty scientists are facing when studying IPD is the heterogeneous nature of the disease, which is further exacerbated by a plethora of environmental and epigenetic influences.

Inflammation has been considered a hallmark of PD since the late 1980’s, when an upregulation of reactive microglia was first seen in patient brain tissue samples (McGeer et al., 1988). Positron emission tomography (PET) imaging dyes, which allow visualizing activated microglia in vivo over time, have been tested as biomarkers for disease progression (Gerhard et al., 2006; Terada et al., 2016; Roussakis and Piccini, 2018). Unfortunately, however, the findings from PET studies have not been successfully transferred to the clinic and the exact molecular mechanisms triggering neuroinflammation in PD currently remain elusive.

Thus, in this study, we investigated patient-derived microglia to explore the inflammatory component of IPD. First, we made use of publicly available transcriptomic data from nigral postmortem tissue to assess the expression of different cytokines in IPD patients compared to age-matched controls. Previous reports established certain secreted cytokines as reliable biomarkers in serum and plasma of PD patients, among them IL-1β, IL-18, IL-6, IL-10, IL-8, TNF-α (Nagatsu et al., 2000; Brodacki et al., 2008; Qin et al., 2016). In line with these studies, our meta-analysis of published brain SNpc case-control transcriptomics datasets indicated elevated levels of IL10 and IL1B in IPD patients.

To explore the cellular origin of this upregulation in IPD, we made use of our previously generated snRNAseq dataset from midbrain IPD and control tissue. Multiple cytokines, including IL10 and IL1B, were predominantly expressed in microglia. This is in accordance with microglia acting as the main player of the immune system in the CNS. Moreover, in the same published dataset, we observed an increase in the microglia number as well as morphological alterations, indicative of an activated state, in IPD patients (Smajić et al., 2020). Next, to corroborate our findings from homogenized SN tissue, we investigated whether any of the aforementioned cytokines show an IPD-specific expression pattern in microglia. While we observed an increase in the expression of all investigated candidates in the patient compared to control cells, the levels of IL10, IL18, and IL1B were the most abundant.

Albeit informative, these postmortem results may be confounded by the fact that they only represent the molecular situation during the latest stage of the disease. Thus, to study inflammatory phenotypes and pathways in an in vitro IPD model, we differentiated microglia from control and patient-derived iPSCs using a published protocol (van Wilgenburg et al., 2013; Haenseler et al., 2017). While cultured IPD microglia did not show altered morphology (data not shown), we observed elevated phagocytosis in these cells indicative of overactive immune function. Phagocytosis is an integral part of microglial homeostatic function, and is not only involved in the recognition of self and non-self threats, but also in the engulfment of synaptic elements and the pruning process. Furthermore, enhanced and uncontrolled clearance is contributing to synaptic degeneration. Indeed, multiple PD studies showed loss of presynaptic terminals and synaptic changes in PD patient compared to control brains (Delva et al., 2020; Matuskey et al., 2020). It is also worth noting that disrupting the phagocytic ability of mouse glia was sufficient to rescue the neuronal degeneration phenotype observed in these animals after LPS injection (Bodea et al., 2014). Since we observed the differences in phagocytosis already at basal level, one may speculate that the genetic background in IPD glia contributes to the development of the disease. However, it is still unclear whether overactive, defective or perturbed uptake triggers PD pathogenesis (Janda et al., 2018).

To further validate our findings from postmortem tissue, we analyzed different inflammasome components in the iPSC-derived microglia cultures. IPD microglia were more reactive after priming with LPS, as indicated by enhanced expressions of IL1B and IL10, and higher mRNA and protein levels of NLRP3 compared to treated control cells. Higher levels of IL1B and NLRP3 in IPD microglia indicate a stronger priming step, which is necessary for downstream inflammasome activation and immune response. While we are the first to show NLRP3 dysregulation in iPSC-derived IPD microglia, our results are in agreement with findings from genetic PD models. Specifically, α-synuclein fibrils were shown to induce NLRP3 activation, and loss of the PD-associated protein Parkin triggered the release of mitoDAMPs into the cytosol, which in turn activated the NLRP3 inflammasome in mice (Zhong et al., 2016; Gordon et al., 2018; Ji et al., 2020; Pike et al., 2021). Moreover, NLRP3 was shown to regulate IL10 levels in mice macrophages, with IL10 production being decreased in NLRP3^–/– mice (Gurung et al., 2015; Kobayashi et al., 2016). This is consistent with our observation of IL10 and NLRP3 co-regulation. In line with a biomarker study in serum, IPD patients had higher levels of IL-10 compared to healthy individuals (Rentzos et al., 2009). Furthermore, while the relationship of IL-1β and IL-10 has not been extensively studied in the context of PD, there are reports showing that, under inflammatory conditions, IL-10 selectively inhibits the release of IL-1β (Sun et al., 2019). Patient-derived microglia will be a useful model to explore the molecular mechanisms linking IL-10 and IL-1β in PD in more detail.

Further of interest with regard to inflammation in genetic but also IPD is the kinase LRRK2. Affected individuals harboring mutations in LRRK2 closely mirror the clinical picture of IPD patients (Tolosa et al., 2020) with kinase activity dysregulation being a shared feature of both forms of the disease. Due to its high abundance in immune cells, researchers have speculated that LRRK2 may be crucially involved in the regulation of neuroinflammatory processes (Gardet et al., 2010; Di Maio et al., 2018; Fyfe, 2018). Studies investigating LRRK2 expression in IPD brain tissue showed a significant downregulation in dopaminergic neurons, which may contribute to the pathology of the movement disorder (Simunovic et al., 2009; Sharma et al., 2011; Yılmazer et al., 2021). In agreement with these reports, we detected significantly reduced LRRK2 expression and a trend toward lower LRRK2 protein abundance in IPD patient microglia at baseline. Inflammatory insults exacerbated this phenotype, leading to a further reduction in LRRK2 protein levels in the IPD patient-derived cells. However, the exact pathways connecting LRRK2 downregulation to microglia dysfunction in IPD warrant further investigation.

Taken together, inspired by published biomarker studies, we investigated inflammatory phenotypes in different models of IPD. In both, nigral and midbrain RNAseq datasets, we observed a disease-specific upregulation of IL10 and IL1B. Furthermore, from our postmortem single-cell results, we derived that this overexpression predominantly stems from microglia. Next, to test whether we could reproduce this phenotype in a dish, we generated iPSC-derived IPD microglia. Further implicating IL10 and IL1B, in IPD, the expression of these cytokines was also enhanced in patient microglia upon LPS treatment. Finally, we identified an upregulation of NLRP3 on RNA and protein level, corroborating our findings concerning IL10 and IL1B. However, in light of the variability of sporadic PD, our results from a small sample may only be representative for a subset of IPD cases, warranting validation studies in larger cohorts. Moreover, while our study highlights the relevance of microglia in IPD, further experiments will be needed to decipher the exact pathways triggering neuroinflammation in sporadic PD patients.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/geo/, GSE157783; https://www.ncbi.nlm.nih.gov/geo/, GSE8397.

Patients gave written and informed consent. The study was approved by the Comité National d’Ethique de Recherche Luxembourg (CNER, vote 201411/05 V1.3).

SAC provided training in iPSC-derived microglia. KB, PM, SD, and SLP collected the data. KB, PM, SS, L-CT, PMAA, and EG performed the analysis. KB, PM, SS, PMAA, CV, and AG wrote the manuscript, which was reviewed by all authors. CV and AG conceived the study. ND, TR, and JCS contributed to the establishment of fibroblast cultures and iPSC generation from patient and control cells. AG acquired funding for the study and was in charge of direction and planning of the study.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.