Chronic oral administration of P. gingivalis induces microglial activation and degeneration of dopaminergic neurons possibly through increase in gut permeability and peripheral IL-17A in LRRK2 R1441G mice

Background The R1441G mutation in the leucine-rich repeat kinase 2 (LRRK2) gene results in late-onset Parkinson’s disease (PD). Peripheral inammation and gut microbiota are closely associated with the pathogenesis of PD. Chronic periodontitis is a common type of peripheral inammation, which is associated with PD. Porphyromonas gingivalis (Pg), the most common bacterium causing chronic periodontitis, can cause alteration of gut microbiota. It is not known whether Pg-induced dysbiosis plays a role in the pathophysiology of PD. Methods


Abstract Background
The R1441G mutation in the leucine-rich repeat kinase 2 (LRRK2) gene results in late-onset Parkinson's disease (PD). Peripheral in ammation and gut microbiota are closely associated with the pathogenesis of PD. Chronic periodontitis is a common type of peripheral in ammation, which is associated with PD. Porphyromonas gingivalis (Pg), the most common bacterium causing chronic periodontitis, can cause alteration of gut microbiota. It is not known whether Pg-induced dysbiosis plays a role in the pathophysiology of PD.

Methods
In this study, live Pg were orally administrated to animals, three times a week for one month. Pg-derived lipopolysaccharide (LPS) was used to stimulate peripheral blood mononuclear cells in vitro. The effects of oral Pg administration on the gut and brain were evaluated through behaviors, morphology, and cytokine expression.

Results
Dopaminergic neurons in the substantia nigra were reduced and activated microglial cells were increased in R1441G mice given oral Pg. In addition, an increase in mRNA expression of tumor necrosis factor (TNFα) and interleukin-1 β (IL-1β) as well as protein level of α-synuclein together with a decrease in zonula occludens-1 (Zo-1) were detected in the colon in Pg-treated R1441G mice. Furthermore, serum interleukin-17A (IL-17A) and brain IL-17 receptor A (IL-17RA) were increased in Pg-treated R1441G mice.

Conclusions
These ndings suggest that LRRK2 causes gut leakage and further mediates peripheral IL-17A response in Pg-treated animals. We, thus, put forward the hypothesis that IL-17A in the serum may result in activation of the IL-17A-IL-17RA axis that aggravates dysfunction of dopaminergic neurons and provokes microglial activation in LRRK2 R1441G mice.

Background
Parkinson's disease (PD) is the second most common neurodegenerative disease that results in a progressive movement disorder characterized by slowness, rigidity, gait di culty, and rest tremors [1]. Degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNpc) is one of the pathological hallmarks of PD [2][3]. Although the exact cause of PD remains poorly understood, it is generally believed that complex interactions between genetic and environmental factors contribute its development.
Leucine-rich repeat kinase 2 (LRRK2) mutants are the most common genetic factors in the pathogenesis of PD [4]. Substantial evidence suggests that mutant LRRK2 strongly activates brain immune cells, which in turn mediate neurodegeneration through neuroin ammation [5][6]. Interestingly, LRRK2 has been also linked to several systemic in ammatory diseases, such as in ammatory bowel disease and leprosy [7][8].
Activation of LRRK2, however, has been reported to induce opposite effects in the brain and the periphery. For example, activation of LRRK2 protects against infection in the gut, but causes neurodegeneration in the brain [9][10].
Recently, chronic systemic in ammatory diseases have been linked to the risk of developing PD. Periodontal disease is a common chronic in ammatory disease and is associated with PD [11][12][13].
The relationship between intestinal function disorder and PD has attracted much attention [16][17]. Until now, the link between the two diseases was based only on motor disturbances caused by PD, which could lead to progression of periodontal disease [12][13]. However, whether periodontal disease can have an in uence on initiation and progression of PD through the intestinal pathway and the underlying mechanism remain unclear.
Recent studies suggest that peripheral lymphocytes may play a central role in the pathophysiology of PD [18]. For example, interleukin-17A (IL-17A) level was signi cantly increased in the serum of patients with PD [19][20][21]. Furthermore, IL-17A could induce human induced pluripotent stem cell-derived midbrain neuronal cell death, possibly through IL-17 receptor A (IL-17RA) [19][20]. IL-17A is mainly driven by Th17 lymphocytes. Interestingly, Th17 cells have been linked to several immune-related diseases, including periodontal disease [22][23]. Th17 cells are also essential for normal defense against gut pathogens [24]. Therefore, we hypothesized that oral Pg might induce peripheral in ammatory responses leading to degeneration of dopaminergic neurons through the gut in LRRK2 R1441G mice.

Animals
All animal procedures were performed according to the Guide for the Care and Use of Laboratory Animals of Sun Yat-sen University (Guangzhou, China). All animals were housed in a speci c pathogen-free facility with a 12:12 h light/dark cycle, ad libitum food and water. In this study, 3-to 4-month-old FVB/NJ, and FVB/N-Tg (LRRK2*R1441G)135Cjli/J mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA) and crossed in the Guangdong Laboratory Animals Monitoring Institute (Guangzhou, China). At one month, all littermates were genotyped. Genotyping was done by polymerase chain reaction (PCR) of tail DNA using a protocol from Jackson Laboratory. A total of 40 mice were used in this study and assigned to four groups: FVB/N + carboxymethyl cellulose (F + C), FVB/N + Pg (F + Pg), R1441G + C, and R1441G + Pg.

Pg cultures and administration
Pg was cultured in broth (Brian Heart Infusion, L-cysteine hydrochloride monohydrate, yeast extract, and chloroproto-ferriheme, Sigma-Aldrich, St. Louis, Missouri, USA). After that, Pg was placed in an anaerobic container for 48 h at 37°C. A total of 10 9 colony-forming units of live Pg was suspended in 0.1 ml phosphate-buffered saline (PBS) with 2% carboxymethyl cellulose (CMC) (Sigma-Aldrich), and given to each mouse by gavage three times a week for about a month, as described previously [14][15]. The control group was administered 0.1 ml PBS with 2% CMC without Pg. After administration, all mice were allowed to eat and drink ad libitum.

Rotarod test
Animals were placed on an accelerating rotarod (Xin Ruan, Shanghai, China) with an accelerated speed of 4-40 rpm for 5 min, and the latency to fall was recorded each time. Animals were tested three times a day for three consecutive days, allowing for two days of training and acclimatization. A resting time of at least 30 min was given between trials. The results are presented as the average of the three times.

Open eld
Animals were placed in the chamber (45 × 45 × 45 cm) with a video camera (Xin Ruan, Shanghai, China). Every mouse was carefully placed in the center of the chamber and allowed to freely explore the chamber.
Animals were tested for two consecutive days, allowing for one day of training and acclimatization. The movement of mice was lmed and analyzed automatically for 10 min.

Statistical analysis
Data were analyzed with one-way ANOVA with Tukey's multiple comparisons test using GraphPad Prism 6.0 software. The results are expressed as the mean ± SEM. Statistical signi cance was set at P < 0.05.

Oral Pg induced dopaminergic neuronal degeneration in the SNpc of mutant LRRK2 mice
To examine whether oral Pg can induce dopaminergic neuronal degeneration, Pg was administrated orally to FVBN mice and LRRK2 R1441G transgenic mice for a month. Immuno uorescence TH staining was used to examine loss of SNpc dopaminergic cells. We found that there was a signi cant loss of TH + neurons in the SNpc in R1441G mice, but not in FVBN mice (Fig. 1a). Confocal immuno uorescence imaging revealed active caspase-3 in the cytoplasm and cell nucleus of TH + SNpc dopaminergic neurons of LRRK2 R1441G mice, but not FVBN mice (Fig. 1b). Immunoblot further con rmed that protein level of cleaved active caspase-3 was greatly increased in the SN of LRRK2 R1441G mice after Pg treatment compared to FVBN mice (Fig. 1c). In addition, LRRK2 R1441G mice exhibited a signi cant reduction in the immuno uorescence intensity of SNpc MAP2 + dendrite (Fig. 1d), which was accompanied by the reduction in MAP2 + protein level (Fig. 1e). In contrast, the immuno uorescence intensity of SNpc MAP2 + dendrite was not signi cantly altered in FVBN + Pg. To further determine whether in ammation-induced neuronal loss is mutation speci c, Pg was administered to mice overexpressing human wild-type LRRK2 (WT-OX). We found that there were no signi cant differences in TH + number and expression levels of MAP2 + protein between Pg-treated WT-OX and WT-OX mice (Supplementary Figs. 1a, b), thereby suggesting that oral Pg-induced neurodegeneration was mutant LRRK2-dependent. Meanwhile, there were no signi cant differences in the rotarod and open eld tests among these groups ( Supplementary  Fig. 1c).

Oral Pg increased microglial activation in the SNpc of mutant LRRK2 mice
Over-activation of microglia has been linked to neurodegeneration in PD [25][26]. In the present study, there was a signi cant increase in the number of activated Iba1-positive microglia in the SNpc in R1441G Discussion LRRK2 is highly expressed in immune cells and mutation of LRRK2 has been linked to both intestinal in ammatory disease and PD [5,[29][30]. In this study, we investigated the contribution of oral Pg to the mice compared to FVBN and WT-OX mice, one month following treatment with oral Pg (Fig. 2a,  Supplementary Fig. 2a).

Oral Pg increased LRRK2 activation in the SN of R1441G mice
Mutant LRRK2 has been implicated in neuronal cell death and microglial in ammatory response of SNpc [5,10]. In this study, both LRRK2 and LRRK2 p935 were signi cantly increased in the SN of Pg-treated R1441G mice compared to Pg-treated FVBN mice (Fig. 3a, b). Although LRRK2 protein expression was also increased in the SN of Pg-treated WT-OX mice, LRRK2 p935 was not altered in WT-OX after Pg treatment (Supplementary Fig. 2c). Double immuno uorescence staining using anti-LRKK2, anti-TH, and anti-Iba1 was performed to visualize the co-localization of LRRK2 in SNpc dopaminergic neurons and microglia. Consistent with western blots, the immunosignal of LRRK2 was evident in Pg-treated R1441G mice. In addition, LRRK2 was partially co-localized with TH + neurons and Iba1 + microglia (Figs. 3c, d ).

Mutant LRRK2 exacerbated Pg-induced peripheral IL-17A secretion and IL-17RA upregulation
IL-17A has been linked to the activation of LRRK2 [27]. We rst examined the serum levels of IL-17A in animals receiving either CMC or Pg. There was no signi cant difference in IL-17A between FVBN and R1441G mice following CMC treatment (Fig. 4a). However, serum IL-17A was signi cantly increased in R1441G mice compared to FVBN mice following oral administration of Pg (Fig. 4a). Consistently, IL-17A was signi cantly increased in the supernatant of PBMCs from R1441G mice stimulated by Pg-derived LPS compared with FVBN mice (Fig. 4b). However, IL-17A protein level remained unchanged in the SN (Supplementary Fig. 2d). Instead, IL-17RA protein level was elevated in the SN of R1441G mice with Pg compared with FVBN mice (Fig. 4c). In addition, IL-17RA was co-localized with TH in the SN of R1441G mice with Pg (Fig. 4d)

Pg-treatment increased the accumulation of α-synuclein in neurons of the colon and induced activation of LRRK2
Emerging evidence suggests that α-synuclein accumulates in neurons of the gut prior to the brain in PD [17,28]. Although histological analysis revealed normal morphology of the colon and small intestine in all groups ( Supplementary Fig. 3a), we found that the α-synuclein in the myenteric plexus of the colon was higher in Pg-treated R1441G mice than in control mice (Fig. 5a). There was no detectable α-synuclein in the brain and small intestine (data not shown). Furthermore, immunoblot analysis demonstrated a signi cant increase in LRRK2 and LRRK2 935 protein levels in the colon of Pg-treated R1441G mice compared to each of the other three groups of mice. Besides, oral administration of Pg led to a signi cant increase in mRNA expression of TNF-a, IL-1β, and Zo-1 in the colon of R1441G mice, but not FVBN mice. pathogenesis of mutant LRRK2-PD in LRRK2 (R1441G) mice. Although oral Pg induced a mild in ammatory response in the intestine, it caused a signi cant loss of dopaminergic neurons and profound microglial activation in the SNpc. In addition, oral Pg resulted in an IL-17A immune response in the periphery and upregulation of IL-17RA protein level of dopaminergic neurons, thereby suggesting that the interaction between IL-17A with IL-17RA may be responsible for neurodegeneration and neuroin ammation in LRRK2-associated PD. Furthermore, these oral Pg-mediated harmful effects are accompanied by an increase in LRRK2 935 expression, which suggests a critical role of LRRK2 kinase in Pg-induced neuropathogenesis in LRRK2-associated PD.
Systemic in ammation has been shown to induce dopaminergic neuronal death through activation of LRRK2 [6]. We consistently found that dopaminergic degeneration was evident in R1441G animals following Pg administration. This event is mediated by aberrant LRRK2 kinase, as evidenced by an increase in LRRK2 935 expression in both the brain and colon. Interestingly, Pg-induced expression of LRRK2 935 led to profound activation of microglia, whereas the gut morphology remained intact in R1441G mice. Thus, it is intriguing how oral Pg induces neuroin ammation in the brain. In PD, αsynuclein is considered to play a pivotal role in brain-gut-microbiota axis interactions [28,31]. Gut in ammation induces expression of α-synuclein and the latter travels along with the vague nerve to initiate the process of α-synuclein misfolding in the brain, which leads to neuroin ammation [32]. Previously, LRRK2 activity has been shown to enhance expression of α-synuclein [33][34]. Pg-mediated LRRK2 activation consistently induced expression of α-synuclein in the colon of R1441G mice. However, α-synuclein levels in the brain were not signi cantly different between FVBN and R1441G mice, thereby suggesting that α-synuclein may not be responsible for activation of microglia. This is in contrast to a recent observation that injection of α-synuclein brils in the gut mediates the spread of pathologic αsynuclein in the brain via the vagus nerve [35]. Use of different animal models may explain the discrepancy between our study and the previous study. Indeed, α-synuclein brils can effectively seed the formation of Lewy body-like inclusions due to their high aggregation propensity. Alternatively, gutmediated systemic in ammation may induce brain in ammation via circulating cytokines [32,36]. In the periphery, it has been reported that activation of LRRK2 in monocytes and macrophages modulates proin ammatory responses, such as IL-1β and TNF-a, which in turn lead to loss of the epithelial barrier function [5,9,[37][38]. We consistently found that oral Pg induced LRRK2 activation and reduced epithelial barrier protein Zo-1. Consequently, loss of Zo-1 may evoke the release of Pg-derived LPS into the blood stream. Elevated IL-17A level has been reported in the serum from patients with PD [19][20]. Emerging evidence indicates that IL-17A can induce neuroin ammation in animal models and PD patients [39][40][41][42]. IL-17A level was consistently increased in the serum of R1441G mice with oral Pg, as well as Pg-derived LPS-stimulated PBMC from R1441G mice. Peripheral IL-17A has been reported to disrupt and cross the blood-brain-barrier [42]. However, neither IL-17A nor Th17 cells could be detected in the brain. Thus, peripheral IL-17A may be responsible for neurodegeneration. IL-17RA is required for the biological activity of IL-17A [43]. In our study, we found that IL-17RA was increased in the dopaminergic neurons of the SN. Several studies have shown that IL-17 could trigger neuronal death through IL-17RA [19]. Thus, it is likely that serum IL-17A may mediate neuronal death through interaction with IL-17RA in Pg-induced R1441G mice. Interestingly, IL-17A has been reported to mediate dopaminergic neurons degeneration via IL-17RA in microglia in a previous study [41]. However, we did not detect any expression of IL-17RA in microglia, although reactive microglia were evident in Pg-induced R1441G mice. The differing results may be due to the use of different models in the experiments.
Microglia in a state of heightened reactivity have a vital role in PD [44]. Evidence suggests that mutant LRRK2 may enhance microglial process outgrowth and in ammatory response leading to chronic damage of dopaminergic neurons [45-46]. Our study consistently showed a signi cant increase in Iba1 number in R1441G LRRK2 mice. Besides, our study showed LRRK2 co-localization in microglia and neurons. We, therefore, put forward the hypothesis that activation of the IL-17A-IL-17RA axis aggravates dysfunction of dopaminergic neurons and induces microglial activation in R1441G LRRK2 animals.

Conclusions
Our results indicate that activation of LRRK2 may have an important role in gut barrier leakage and IL-17A production, and subsequently trigger neuronal death through the IL-17A-IL-17RA axis. In conclusion, our results elucidate the role of the brain-gut axis in the pathophysiology of Parkinson's disease.

Availability of data and materials
All data generated or analysed during this study are included in this published article [and its supplementary information les].