Bacterial products initiation of alpha-synuclein pathology: an in vitro study Authors: Octavian Costin Ioghen, Gisela Gaina, Ioana Lambrescu, Emilia Manole, Sevinci Pop, Teodora Maria Niculescu, Oana Mosoia, Laura Cristina Ceafalan, Bogdan Ovidiu Popescu Published: 2024 ABSTRACT Parkinson’s Disease (PD) is a prevalent and escalating neurodegenerative disorder with significant societal implications. Despite being considered a proteinopathy, in which the aggregation of α-synuclein is the main pathological change, the intricacies of PD initiation remain elusive. Recent evidence suggests a potential link between gut microbiota and PD initiation, emphasizing the need to explore the effects of microbiota-derived molecules on neuronal cells. In this study, we exposed dopaminergic-differentiated SH-SY5Y cells to microbial molecules such as lipopolysaccharide (LPS), rhamnolipid, curli CsgA and phenol soluble modulin α-1 (PSMα1). We assessed cellular viability, cytotoxicity, growth curves and α-synuclein levels by performing MTS, LDH, real-time impedance readings, qRT-PCR and Western Blot assays respectively. Statistical analysis revealed that rhamnolipid exhibited concentration-dependent effects, reducing viability and inducing cytotoxicity at higher concentrations, increasing α-synuclein mRNA and protein levels with negative effects on cell morphology and adhesion. Furthermore, LPS exposure also increased α-synuclein levels. Curli CsgA and PSMα-1 showed minimal or no changes. Our findings suggest that microbiota-derived molecules, particularly rhamnolipid and LPS, impact dopaminergic neurons by increasing α-synuclein levels. This study highlights the potential involvement of gut microbiota in initiating the upregulation of α-synuclein that may further initiate PD, indicating the complex interplay between microbiota and neuronal cells. Introduction Parkinson’s disease (PD) is the second most common neurodegenerative disease affecting the population, having not only a clinical impact on the lives of patients but also a resounding effect on society1. Significantly concerning is the increasing rate of Parkinson’s disease (PD) occurrence in the population—the most rapid among neurological disorders, estimated to double in prevalence by 20401. Reducing these numbers may be possible by preventive measures that target the initiation and progression of PD pathogenesis1,2. An efficient neuroprotective treatment does not yet exist due to the unresolved puzzle that is the biomolecular picture of PD2. Thus, deciphering the intimate mechanism of PD pathogenesis is the key to overcoming PD morbidity. PD is currently considered a proteinopathy3. The aggregation of α-synuclein leads to neuronal loss, especially of the dopaminergic neurons from substantia nigra pars compacta, which are responsible for parkinsonism4,5. Alpha-synuclein is a protein with conformational plasticity and, in cases of sporadic Parkinson’s disease where the specific circumstances are not yet fully understood, has the ability to convert from an unfolded form or α-helical tetramer to a β-sheet structure that forms oligomeric nuclei4,5. These nuclei favour the binding of other α-synuclein monomers and assemble into fibrillar aggregates4,6. Amyloid fibrils accumulate, and concurrently, proteins such as ubiquitin, parkin, microtubular proteins and mitochondria-related proteins are associated with these conglomerates7,8. They are visualized as specific eosinophilic inclusions called Lewy bodies, which represent the pathological hallmark of Parkinson’s disease (PD)8. Originally, the perspective on the spatio-temporal distribution of Parkinson’s disease (PD) pathology was that from the onset of the disease, the Lewy bodies and neuronal death simultaneously occur in the substantia nigra pars compacta and other important anatomical regions affected by the disease9. In 2003, Heiko Braak et al. published a study on autopsies that challenged this initial view and proposed a pathological staging10. Their study described that the medulla oblongata and olfactory bulbs are the primary sites of the lesion and that a rostral spread occurs with a successive involvement of the midbrain, basal forebrain, and in the end, the cortex10. The mechanism of rostral progression of pathological changes, affecting neighbouring structures has been described as a prion-like spread of conformationally changed α-synuclein11–14.Seemingly, the trans-synaptic transmission induces an alteration in the native α-synuclein that favours the initiation of aggregation, and eventually, the altered protein is further transmitted to the next neuron, probably via exosomes, endocytosis or tunnelling nanotubes15–17. Subsequent studies revealed that Lewy bodies appear early in the enteric nervous system, suggesting that this is a significant site for initiating α-synuclein aggregation and the protein is further disseminated in the central nervous system (CNS) via the vagus nerve18–20. Moreover, Hawkes et al. proposed the dual hit hypothesis, which suggests that the process of α-synuclein aggregation is initiated by an unidentified pathogen that enters the respiratory route and is subsequently swallowed in the gastrointestinal system21,22. Although a pathogen has not been identified, this hypothesis points out the notable role of the gastrointestinal tract and nasal cavity in the initiation of α-synuclein aggregation. Further, a special emphasis falls on the microbiota and its metabolites. The role of microbiota gained interest when metagenomics studies reported a different pattern of gut microbiota profile in PD patients compared to controls23. Enterobacteriaceae, which are Gram-negative bacteria that contain lipopolysaccharide (LPS), are increased in PD patients, while there is a decrease in the number of Prevotellaceae, which are short-chain fatty acids (SCFA)-producing bacteria23. These results suggest that the balance between the pro-inflammatory LPS and the anti-inflammatory SCFA in the gastrointestinal tract is disturbed in PD patients23. However, it is currently premature to assert a causative relationship between the onset of PD and alterations in gut microbiota. This can be attributed to several variables including prodromal symptoms of PD, such as constipation, and the influence of medications like anticholinergics and COMT inhibitors, which possess the potential to alter the gut microbiota profile24,25. Hence, there is a need to specifically study the interplay between the cells in the intestinal wall and the molecules produced by the altered PD-gut microbiota. Given the presence of dopaminergic neurones in the enteric nervous system and the ability of gut microbiota metabolites to disrupt the intestinal barrier, enter the bloodstream, disrupt the blood–brain barrier, and interact with dopaminergic neurones in the substantia nigra, a crucial step in understanding the role of gut microbiota in PD pathogenesis is to assess the impact of microbial compounds on dopaminergic neurones26–28. In this in vitro study, we treated the dopaminergic-differentiated SH-SY5Y human neuroblastoma cell line with several microbiota-derived molecules such as LPS, rhamnolipid, curli CsgA, and phenol soluble modulin alpha-1 (PSMα1) and we assessed cellular viability, growth curves and α-synuclein levels. The rationale for choosing these bacterial products is based on their known roles in PD and related research. LPS is a major component of the Gram-negative bacteria, which are increased in PD patients’ gut microbiota, it is known to induce inflammation and permeability changes in the blood–brain barrier, facilitating the interaction of gut-derived molecules with CNS neurons23,29. Additionally, LPS has been shown to increase α-synuclein aggregation in vitro and in vivo models, making it a relevant molecule for studying PD pathogenesis30,31. Rhamnolipid is an amphiphilic molecule produced by Pseudomonas aeruginosa, that it has been shown to induce α-synuclein fibrillation in vitro, suggesting its potential involvement in PD pathogenesis32. Curli CsgA is a major component of Escherichia coli biofilms; curli fibers have structural similarities to human amyloids and have certain roles in the promotion of α-synuclein aggregation33–35. PSMα1, produced by Staphylococcus aureus, is involved in biofilm formation and it has the potential to catalyze α-synuclein aggregation36,37. Results SH-SY5Y dopaminergic differentiation To establish a dopaminergic phenotype, SH-SY5Y cells were subjected to a differentiation protocol, resulting in significant morphological changes indicative of neuronal differentiation. The human neuroblastoma cell line SH-SY5Y was initially cultured in standard growth conditions until they reached sufficient confluence. For differentiation, the cells were first pre-treated and then exposed to a specialized differentiation medium supplemented with retinoic acid and phorbol-12-myristate-13-acetate (PMA). Morphological changes were monitored throughout the differentiation process using phase contrast microscopy. Initially, undifferentiated SH-SY5Y cells displayed clusters of variously shaped neuronal bodies, primarily round and ovoid, with short neurites and few synapses (Fig. 1A). By the end of the differentiation protocol, the cells exhibited a uniform distribution with significant morphological changes, including pyramidal and stellate neuronal bodies, numerous and elongated neurites, and an increased number of synapses (Fig. 1C). These changes indicated successful differentiation into dopaminergic-like neurons.Fig. 1Stepwise graphical representation of SH-SY5Y dopaminergic differentiation protocol with corresponding phase contrast microscopy images (40× objective). In undifferentiated cells (A), clusters of cells, various shapes of neuronal bodies, especially round and ovoid, short neurites, and few synapses are noticed. At the end of the differentiation protocol (C), significant changes such as uniform distribution of the cells, pyramidal and stellate neuronal bodies, numerous and long neurites, and an increase in the number of synapses are observed. Stepwise graphical representation of SH-SY5Y dopaminergic differentiation protocol with corresponding phase contrast microscopy images (40× objective). In undifferentiated cells (A), clusters of cells, various shapes of neuronal bodies, especially round and ovoid, short neurites, and few synapses are noticed. At the end of the differentiation protocol (C), significant changes such as uniform distribution of the cells, pyramidal and stellate neuronal bodies, numerous and long neurites, and an increase in the number of synapses are observed. In order to confirm the dopaminergic differentiation of SH-SY5Y cells, we conducted a Western Blot analysis specific to tyrosine hydroxylase (TH) and D2 Dopamine Receptor (D2DR) as indicators of dopaminergic differentiation. The findings validate the successful differentiation of SH-SY5Y cells into a dopaminergic phenotype, as seen by the elevated levels of dopaminergic markers in differentiated SH-SY5Y in comparison to undifferentiated cells. (Supplementary Fig. 1, Supplementary Fig. 2). MTS viability and LDH cytotoxicity assays after 48h of treatment with microbiota-derived molecules The effect of rhamnolipid on the viability and cytotoxicity of dopaminergic-differentiated SH-SY5Y cells after 48 h of treatment was assessed in a concentration range from 1.25 to 320 µg/ml in ½ dilution series (Fig. 2A). Between 1.25 and 80 µg/ml no decrease in MTS value was seen. However, starting from 160 µg/ml, a steep loss of cell viability was noticed, with 3.82% ± 1 viability at 160 µg/ml (p < 0.001) and 2% ± 1.5 viability at 320 µg/ml (p < 0.001). In the cytotoxicity assays there was a lower LDH release at concentrations between 1.25 and 40 µg/ml (between 18.52% and 22.50%) compared to the control (31.62 ± 2.2%). Starting from 80 µg/ml, a gradual increase in cytotoxicity was observed, with values of 52.54 ± 4.5% (p < 0.001) at 80 µg/ml and slightly above 100% at 160 and 320 µg/ml (101.96 ± 5.2% and 102.64 ± 5.5% respectively, p < 0.001).Fig. 2MTS and LDH assays were performed on dopaminergic-differentiated SH-SY5Y cells 48 h after treatment with rhamnolipid (A), lipopolysaccharide (B), curli CsgA (C), PSMα1 (D) and rotenone (E), used as a positive control for the upregulation of α-synuclein expression. The data were expressed as the mean ± SEM of two independent experiments in triplicate (n = 6). Statistical analyses used fold-change calculation as fold increase to control followed by statistical significance testing (One-Way ANOVA followed by Dunnett’s Test), *p < 0.05, **p < 0.01, ***p < 0.001. MTS and LDH assays were performed on dopaminergic-differentiated SH-SY5Y cells 48 h after treatment with rhamnolipid (A), lipopolysaccharide (B), curli CsgA (C), PSMα1 (D) and rotenone (E), used as a positive control for the upregulation of α-synuclein expression. The data were expressed as the mean ± SEM of two independent experiments in triplicate (n = 6). Statistical analyses used fold-change calculation as fold increase to control followed by statistical significance testing (One-Way ANOVA followed by Dunnett’s Test), *p < 0.05, **p < 0.01, ***p < 0.001. 48 h of LPS treatment in a concentration range from 0.25 to 64 µg/ml in ½ serial dilutions did not determine a statistically significant decrease in cell viability or an increase in cell cytotoxicity (Fig. 2B). Nevertheless, minimal changes were found, such as a discrete decrease of viability at higher concentrations, 32 and 64 µg/ml respectively (93.52 ± 2% and 90.25 ± 2.8% respectively), and a slightly increase of cytotoxicity between 8 and 64 µg/ml (27.66 ± 3.4% for 8 µg/ml, 30.25 ± 3.6% for 16 µg/ml, 28.02 ± 3.5% for 32 µg/ml and 30.47 ± 4.7% for 64 µg/ml) compared to the control (17.66 ± 1.3%). The MTS and LDH assays conducted on dopaminergic-differentiated SH-SY5Y cells after a 48 h treatment with concentrations from 0.25 to 32 µg/ml of curli CsgA revealed a decrease in viability to 90.08 ± 1.2% and a small increase of cytotoxicity to 33.30 ± 2.4% at 32 µg/ml compared to the 29.83 ± 1.8% value of the control, but without statistical significance (Fig. 2C). The same range of concentrations as in the curli CsgA treatment, between 0.25 and 32 µg/ml, were assessed in PSMα1 treatment, and the same effect was noticed at 32 µg/ml, diminished viability (87.67 ± 4.1%) and a small increase in cytotoxicity (44.41 ± 0.7%) compared to the control (40.7 ± 2.7%) (Fig. 2D). The cytotoxicity and viability after treatment with rotenone, which was used as a positive control for the neuronal upregulation of α-synuclein, were also assessed (Fig. 2E). Although the MTS and LDH assays after rotenone treatment on SH-SY5Y cells were already described in the literature, only a few studies evaluated the effect on dopaminergic-differentiated SH-SY5Y cells. A general decrease in viability at around 80% between 0.3 and 5 µM and a descending curve starting from the concentration of 10 µM (72.38 ± 3.7% with p = 0.05 for 10 µM, 40.63 ± 1.7% with p < 0.001 for 20 µM and 22.95 ± 5.1% with p < 0.002 for 40 µM) were noticed. Cytotoxicity significantly increased at 20 µM (60.97 ± 4.8%, p < 0.001) and 40 µM (79.47 ± 9.4%, p < 0.001) compared to the control (36.05 ± 1.1%). Real-time surface impedance monitoring of the dopaminergic differentiated SH-SY5Y cells during 120 h of treatment with microbiota-derived molecules During the differentiation, the halt of proliferation was confirmed by showing the lack of change in Cell Index represented as a horizontal curve on the graphic in comparison with the ascending curve of the proliferative SH-SY5Y cells (Fig. 3). Hence, knowing that the proliferation is absent, any change in the impedance readings and, secondarily any modification of the Cell Index curve would be a consequence of an impairment in cellular adhesion and of changes in morphology.Fig. 3Real-time cell monitoring of SH-SY5Y cells in the proliferative state using growth medium (supplemented with 10% fetal bovine serum) vs. SH-SY5Y cells that follow the dopaminergic differentiation protocol. The continuous increase in cell index in the proliferative SH-SY5Y cells reflects cell proliferation. In contrast the horizontal curve without significant cell index changes in the cells that follow the dopaminergic differentiation protocol reflects the stop of proliferation. The spikes after 72 h represent the pause of readings while changing the medium. In the cells that follow the dopaminergic differentiation protocol, a differentiation medium was added from the beginning to prevent the detachment of cells from the E-16 plates when changing the medium after 24 h of incubation (excluding day 0 presented in the differentiation protocol). To more accurately depict variations in the cell index throughout the course of the cell monitoring period, normalisation was not carried out. Representative image of triplicate wells. Real-time cell monitoring of SH-SY5Y cells in the proliferative state using growth medium (supplemented with 10% fetal bovine serum) vs. SH-SY5Y cells that follow the dopaminergic differentiation protocol. The continuous increase in cell index in the proliferative SH-SY5Y cells reflects cell proliferation. In contrast the horizontal curve without significant cell index changes in the cells that follow the dopaminergic differentiation protocol reflects the stop of proliferation. The spikes after 72 h represent the pause of readings while changing the medium. In the cells that follow the dopaminergic differentiation protocol, a differentiation medium was added from the beginning to prevent the detachment of cells from the E-16 plates when changing the medium after 24 h of incubation (excluding day 0 presented in the differentiation protocol). To more accurately depict variations in the cell index throughout the course of the cell monitoring period, normalisation was not carried out. Representative image of triplicate wells. Treatment with microbial compounds concentrations was chosen based on the end-point assays performed previously, MTS viability and LDH cytotoxicity assays, and the effects were monitored for 120 h. The treatment with rhamnolipid was performed in three concentrations: 10, 40, and 80 µg/ml (Fig. 3A). The 10 µg/ml concentration did not induce any modification in the Cell Index, having an identical curve with the control. At 40 µg/ml, a decrease in Cell Index started to appear after 40 h, with a normalized Cell Index (nCI) of 87.46% since the beginning of treatment at 72 h, 66.39% at 96 h, and 44.02% at 120 h. At 80 µg/ml, a concentration where there was a cytotoxic effect in LDH assay, a severe impairment of the impedance measurements was observed. An initial increase in nCI of up to 118.53% in the first 10 h, followed by an abrupt decrease (nCI = 80.14% at 24h, nCI = 15.97% at 48h) indicates a critical change in cell adhesion. Similar to the MTS and LDH assay results, LPS did not induce changes of impedance, the cells keeping a nCI around 1 throughout the 120 h monitoring process (Fig. 4B).Fig. 4Real-time cellular impedance monitoring of dopaminergic-differentiated SH-SY5Y cells treated with microbiota-derived molecules. Based on the MTS and LDH end-point assay results, several concentrations were chosen to evaluate their effects for 120 h, such as 10 µg/ml, 40 µg/ml, and 80 µg/ml of rhamnolipid (A), 1 µg/ml, 32 µg/ml and 64 µg/ml of LPS (B), 16 µg/ml of curli CsgA (C) and 16 µg/ml of PSMα1 (D). As in the previous experiments, rotenone was chosen as a positive control in concentrations of 0.1 µM, 0.3 µM, and 10 µM. The cell index was normalized to 1 immediately after treatment addition. The impedance measurements presented in the images are from the moment of treatment addition, while the dopaminergic differentiation was not shown here. The images are representative of two independent experiments, each performed in duplicate wells (n = 4). Real-time cellular impedance monitoring of dopaminergic-differentiated SH-SY5Y cells treated with microbiota-derived molecules. Based on the MTS and LDH end-point assay results, several concentrations were chosen to evaluate their effects for 120 h, such as 10 µg/ml, 40 µg/ml, and 80 µg/ml of rhamnolipid (A), 1 µg/ml, 32 µg/ml and 64 µg/ml of LPS (B), 16 µg/ml of curli CsgA (C) and 16 µg/ml of PSMα1 (D). As in the previous experiments, rotenone was chosen as a positive control in concentrations of 0.1 µM, 0.3 µM, and 10 µM. The cell index was normalized to 1 immediately after treatment addition. The impedance measurements presented in the images are from the moment of treatment addition, while the dopaminergic differentiation was not shown here. The images are representative of two independent experiments, each performed in duplicate wells (n = 4). Curli CsgA 16 µg/ml produced a slight and slow decrease of the Cell Index Curve, with a nCI of 91.02% at 48 h and 76.55% at 120h, while 16 µg/ml of PSMα1 maintained a nCI around the value of 1 with a similar curve compared to the control (Fig. 4C,D). Although rotenone was evaluated at 0.1, 0.3, and 10 µM, with a 100-fold increase from the lowest to the highest concentration, the graphical representation of CI was similar for all three concentrations showing a continuous decrease, with a nCI of around 85% at 48 h and around 45% at 120 h (Fig. 4E). Nevertheless, at the highest concentration (10 µM), an initial increase of nCI of up to 120.37% in the first 10 h was noticed, in a similar way as with rhamnolipid 80 µg/ml. α-synuclein mRNA levels after 48h of treatment with microbiota-derived molecules qRT-PCR was performed on dopaminergic-differentiated SH-SY5Y cells after 48h of treatment with microbial compounds to measure the mRNA levels of α-synuclein (Fig. 5). Rotenone 0.3 µM was used as a positive control of α-synuclein overexpression, and the results showed a doubling (p < 0.001) of mRNA levels in comparison to the control.Fig. 5QRT-PCR for α-synuclein mRNA performed 48 h after treatment with bacterial compounds (curli CsgA 16 µg/ml, PSMα1 16 µg/ml, LPS 1 µg/ml, 32 µg/ml and 64 µg/ml, rhamnolipid 10 µg/ml, 40 µg/ml and 80 ug/ml) on dopaminergic-differentiated SH-SY5Y cells. Rotenone was used as a positive control for α-synuclein overexpression. The data are shown as a fold increase to control. The data were normalized to GAPDH. Data are expressed as the mean ± SEM of three independent experiments carried out at least in duplicate. Statistical analyses used fold-change calculation as fold increase to control followed by statistical significance testing (for Rha 40 µg/ml and Rha 80 µg/ml which were non-parametric data we used Kruskal–Wallis test and Dunn’s multiple comparison test, while for the other parametric data we used One-Way ANOVA followed by Dunnett’s Test) *p < 0.05, **p < 0.01, ***p < 0.001. QRT-PCR for α-synuclein mRNA performed 48 h after treatment with bacterial compounds (curli CsgA 16 µg/ml, PSMα1 16 µg/ml, LPS 1 µg/ml, 32 µg/ml and 64 µg/ml, rhamnolipid 10 µg/ml, 40 µg/ml and 80 ug/ml) on dopaminergic-differentiated SH-SY5Y cells. Rotenone was used as a positive control for α-synuclein overexpression. The data are shown as a fold increase to control. The data were normalized to GAPDH. Data are expressed as the mean ± SEM of three independent experiments carried out at least in duplicate. Statistical analyses used fold-change calculation as fold increase to control followed by statistical significance testing (for Rha 40 µg/ml and Rha 80 µg/ml which were non-parametric data we used Kruskal–Wallis test and Dunn’s multiple comparison test, while for the other parametric data we used One-Way ANOVA followed by Dunnett’s Test) *p < 0.05, **p < 0.01, ***p < 0.001. Rhamnolipid was evaluated at 10, 40, and 80 µg/ml, and a concentration-dependent increase of α-synuclein mRNA was noticed. At 10 µg/ml, no change appeared, while at 40 µg/ml there was a 1.77-fold increase (p = 0.007). At 80 µg/ml, the increase in mRNA exceeded the increase from the rotenone treatment, having a 2.30-fold increase (p = 0.001). LPS concentrations induced a discrete dose-dependent increase of mRNA mean levels but without statistical significance. At 1 µg/ml, levels were similar to the control. At 32 µg/ml an increase of 11.15%, while at 64 µg/ml an increase of 30.25% was noticed. 16 µg/ml of curli CsgA and PSMα1 did not modify the mRNA levels of α-synuclein in dopaminergic-differentiated SH-SY5Y cells. Intracellular α-synuclein levels after 48h of treatment with microbiota-derived molecules Western Blot was performed to assess the intracellular α-synuclein protein levels in cells treated for 48 h with the selected microbial compounds (Fig. 6). 0.3 μM of rotenone induced a 1.44-fold increase (p = 0.01) of intracellular α-synuclein compared to the control.Fig. 6Western Blot for α-synuclein performed 48 h after treatment with bacterial compounds (curli CsgA 16 µg/ml, PSMα1 16 µg/ml, LPS 1 µg/ml, 32 µg/ml and 64 µg/ml, rhamnolipid 10 µg/ml, 40 µg/ml and 80 µg/ml) on dopaminergic-differentiated SH-SY5Y cells. The blots were cut prior to hybridisation with antibodies due to technical reasons to ensure they fit the incubation boxes. Rotenone 0.3 µM was used as a positive control for α-synuclein overexpression. Controls were established as treatment with the vehicle of the stock solution of the bacterial compounds (PBS, DMSO, PBS + 33% glycerol). Quantification of the α-synuclein levels is represented in (A) and immunoblots in (B) (three different blots are inserted in the same picture, each of the blots with the vehicle control of the treatments). The data of intracellular α-synuclein protein were normalized to GAPDH. The data are shown as fold increase to control, each of the controls is considered = 1.0, which explains why in (A) there is only one control bar and in (B) there are three control samples (CTRL Rha LPS, CTRL CsgA PSM, CTRL Rot). The data are shown as a fold increase to control. Data are expressed as the mean ± SEM of four independent experiments carried out in duplicate (for curli CsgA 16 µg/ml and PSMα1 16 µg/ml – three independent experiments carried out in duplicate). Statistical analyses used fold-change calculation as fold increase to control followed by statistical significance testing (One-Way ANOVA followed by Dunnett’s Test), *p < 0.05, **p < 0.01, ***p < 0.001. Western Blot for α-synuclein performed 48 h after treatment with bacterial compounds (curli CsgA 16 µg/ml, PSMα1 16 µg/ml, LPS 1 µg/ml, 32 µg/ml and 64 µg/ml, rhamnolipid 10 µg/ml, 40 µg/ml and 80 µg/ml) on dopaminergic-differentiated SH-SY5Y cells. The blots were cut prior to hybridisation with antibodies due to technical reasons to ensure they fit the incubation boxes. Rotenone 0.3 µM was used as a positive control for α-synuclein overexpression. Controls were established as treatment with the vehicle of the stock solution of the bacterial compounds (PBS, DMSO, PBS + 33% glycerol). Quantification of the α-synuclein levels is represented in (A) and immunoblots in (B) (three different blots are inserted in the same picture, each of the blots with the vehicle control of the treatments). The data of intracellular α-synuclein protein were normalized to GAPDH. The data are shown as fold increase to control, each of the controls is considered = 1.0, which explains why in (A) there is only one control bar and in (B) there are three control samples (CTRL Rha LPS, CTRL CsgA PSM, CTRL Rot). The data are shown as a fold increase to control. Data are expressed as the mean ± SEM of four independent experiments carried out in duplicate (for curli CsgA 16 µg/ml and PSMα1 16 µg/ml – three independent experiments carried out in duplicate). Statistical analyses used fold-change calculation as fold increase to control followed by statistical significance testing (One-Way ANOVA followed by Dunnett’s Test), *p < 0.05, **p < 0.01, ***p < 0.001. Rhamnolipid 10 μg/ml did not cause a significant increase, while the higher concentrations did, above the levels induced by the positive control treatment with rotenone. Rhamnolipid at 40 µg/ml caused a 1.60-fold increase (p < 0.001), and 80 µg/ml induced an 1.48-fold increase (p = 0.003) compared to the control. Intracellular α-synuclein levels were also increased in the neuronal cells treated with different LPS concentrations. LPS 1 µg/ml caused a mean increase of 1.28 times compared to the control but without statistical significance. Concentrations of 32 µg/ml and 64 µg/ml induced a 1.39-fold increase, respectively 1.36-fold increase (p = 0.02, respectively p = 0.04) compared to the control. Comparable to qRT-PCR results, curli CsgA at 16 µg/ml and PSMα1 at 16 µg/ml did not significantly modify the intracellular α-synuclein levels. Discussions While numerous neuronal in vitro models use the SH-SY5Y human neuroblastoma cell line in the proliferative state, in this study, we used dopaminergic-differentiated SH-SY5Y cells, as they exhibit a closer resemblance to human dopaminergic neurons38–41. Only by observing the morphological differences between the two states of SH-SY5Y cells (proliferative and differentiated) (Fig. 6), one can conclude that the differentiated cells represent a much more accurate in vitro model. Regarding the measurement of α-synuclein levels, although SH-SY5Y is a simple neuronal in vitro model, a significant advantage of this study is the use of a cell line that was not genetically modified to overexpress α-synuclein or to express an altered protein. Thus, any results of an increase in α-synuclein levels appear on a native and basal state of expression and bring valuable information regarding the pathogenesis of PD initiation10,23. To gain a better understanding of the role of gut microbiota in the pathogenesis of PD, it is imperative to look into how exposure to these microbial compounds affects dopaminergic neurones. This is because the enteric nervous system contains dopaminergic neurones, and metabolites produced by gut microbiota have the potential to compromise the intestinal barrier, enter the bloodstream, and breach the blood–brain barrier to interact with the dopaminergic neurones in the substantia nigra26,27. As previously stated, the discovery that PD pathological changes appear in the gastrointestinal tract has brought the gut microbiota into focus10,23,42,43. Hence, to unravel the precise role of the microbes in the initiation of PD, it is crucial to actively search for molecules produced by microbiota that can induce the overexpression or aggregation of α-synuclein. Bacterial amphiphiles, such as rhamnolipid and LPS, and functional amyloids, such as curli and PSMα1, are microbial molecules produced by the gut microbiota that raised the attention of PD researchers over time33,36,44,45. Bacterial amphiphiles are molecules with both hydrophobic and hydrophilic components produced by the gut bacteria and act as biosurfactants in order to promote the survival and growth of the biofilms44. Their amphiphilic nature indicates that each of the compounds has a critical micelle concentration (CMC) above which micelle formation is induced44,46. The CMC may be relevant in the context of PD pathogenesis because if proteins prone to aggregation are sequestered inside micelles, the amyloid formation is rapidly induced32,44. Rhamnolipid is a bacterial amphiphilic molecule produced by Pseudomonas aeruginosa that may colonize the gastrointestinal tract and also the nasal cavity47,48. There is a lack of evaluation of the rhamnolipid effects in PD pathogenesis, and direct studies on rhamnolipid levels in PD are limited. An article published in 2018 studied the fibrillation kinetics between rhamnolipid and α-synuclein and showed that rhamnolipid micelles induced the formation of α-synuclein fibrillates32. The importance of assessing the existence of a possible link between rhamnolipid and PD is amplified by the fact that rhamnolipid has the potential to be used on a large scale as a biological alternative for household detergents or cosmetic products that replace chemical surfactants49,50. Hence, our study indicates a concentration-dependent effect on neuronal viability, cytotoxicity, cell adhesion, and morphology. Moreover, we showed a concentration-dependent α-synuclein overexpression in dopaminergic neurons, which exceeds the rotenone-induced upregulation. LPS is the endotoxin of Gram-negative bacteria. It has a strong proinflammatory effect induced by the stimulation of Toll-like receptor 4 (TLR-4) and is an amphiphilic molecule that enhances the formation of biofilms51,52. Numerous studies highlight the inflammatory effect of the LPS at the level of the intestinal barrier, causing permeabilization53. Although permeabilization enables the interaction between microbiota compounds and other cells in the intestinal wall, such as neurones from the enteric plexus, it also promotes the transport of microbiota metabolites (including LPS) into the bloodstream. These metabolites can then reach a permeabilized blood–brain barrier, allowing direct contact with CNS dopaminergic neurones54. Similarly to rhamnolipid-exposure, the in vitro exposure of α-synuclein to LPS initiates the aggregation pathway55. LPS was previously assessed as a compound to use for PD in vitro and in vivo models (with direct CNS or systemic administration), especially for the models that evaluate the neuroinflammation and the interaction between microglia and neurons, knowing that the dopaminergic neurotoxicity secondary to LPS exposure is dependent on the microglial activation56. Recent studies showed an altered intracellular α-synuclein and an increase of the intracellular α-synuclein levels in a mouse enteroendocrine cell line after exposure to LPS, suggesting the possible role of LPS in initiating PD pathology in the gastrointestinal tract45,57. Research has shown that LPS exposure induces neuroinflammation in SH-SY5Y cells, reducing cell viability30. Using MTT assays, significant decreases in cell viability were observed post-LPS treatment, indicating cytotoxic effects30. The study also found that LPS-stimulated microglia-derived factors exacerbate neuronal cell death in co-culture systems30. Another study, conducted by Ling et al., investigated the impact of LPS on SH-SY5Y cell viability, testing concentrations from 0.1 to 100 µg/mL. The results, obtained using MTT assays, showed that LPS did not significantly affect SH-SY5Y cell viability, indicating that LPS alone does not induce cytotoxicity in these cells31. However, the study found that conditioned medium from LPS-stimulated BV2 microglial cells significantly reduced SH-SY5Y cell viability31. This highlights that the cytotoxic effects of LPS on SH-SY5Y cells are mediated through microglial activation rather than a direct effect of LPS on the neuronal cells themselves31. In our study, we found minimal changes in cell viability and cytotoxicity after LPS treatment, which is supported by the study by Ling et al. Our study showed interesting results after neuronal treatment with LPS, such as a lack of change in the cell viability, cytotoxicity and cell adhesion as compared to control, regardless of the concentrations, but an increase in α-synuclein levels. This pattern differed from those observed in the treatment with the other compounds. Functional amyloids are proteins produced by bacteria and have a role in the construction of biofilms58. The proteins have a β-sheet structure that favours self-aggregation and leads to the formation of amyloids that represent the bricks of the biofilm59. Besides self-aggregation, functional amyloids can induce cross-seeding of human proteins prone to aggregation59. Thus, exposure of α-synuclein to the environmental gut functional amyloids can initiate PD pathological changes, especially in altered α-synuclein prone to aggregate34. Curli protein, especially with its major subunit component CsgA and PSMα1, are functional amyloids produced by Escherichia coli and Staphylococcus aureus respectively, whose role in the aggregation of α-synuclein and in the pathology of Parkinson’s disease started to be investigated in recent studies33–36. Several studies suggest that both Escherichia coli and Staphylococcus aureus are found at increased levels in PD patients, and their presence may contribute to the inflammatory and neurodegenerative processes associated with the disease60,61. Moreover, increased staining for Escherichia coli in the intestinal mucosa also correlates with increased intestinal permeability in PD patients61. The structural similarities between curli fibrils and human pathological amyloids were highlighted, suggesting that curli promote the aggregation process of human amyloids59,62. A study revealed that curli fibers produced by Escherichia coli accelerate the aggregation of α-synuclein, a hallmark of Parkinson’s disease33. Mono-colonization with curli-producing Escherichia coli worsened motor impairments and gastrointestinal issues in α-synuclein-overexpressing mice, enhancing α-synuclein aggregation and inflammation in the gut and brain33. A recent study showed that exposure to bacterial curli combined with fiber-deprived diet accelerate PD progression in a PD mouse model63. In our chosen range of concentrations, we noticed a discrete change in neuronal viability and cytotoxicity at the highest concentration in both treatments (32 µg/ml). We also observed a tendency to decrease adhesion in the curli CsgA treatment. Apart from these, no other changes were noticed, and there was no increase in α-synuclein mRNA or intracellular protein levels. These remarks do not exclude the possible role of inducing intraneuronal α-synuclein aggregation, which was not evaluated in this study. Although we did not assess the intracellular α-synuclein aggregation, the lessons learned from the genetic causes of PD, such as α-synuclein duplication or triplication, show us that elevated intracellular levels undeniably promote the initiation of aggregation, with the commencement of the process being only a matter of time64–66. Methods Chemicals and reagents Retinoic acid (cat.no. R2625), Phorbol 12-Myristate 13-Acetate (PMA) (cat.no. P8139) and rotenone (cat.no. R2625) were purchased from Sigma-Aldrich (Saint Louis, MO, USA) and the stock solutions were obtained by diluting in Dimethyl sulfoxide (DMSO). Lipopolysaccharide (LPS) from Escherichia. coli O111:B4, γ-irradiated, BioXtra, suitable for cell culture (cat.no. L4391) were purchased from Sigma-Aldrich (Saint Louis, MO, USA) and the stock solution was obtained by diluting in phosphate buffer solution (PBS). Rhamnolipid 90% (cat.no. R90) was purchased from AGAE Technologies (Corvallis, OR, USA) and the stock solution was obtained by diluting in PBS. Recombinant Staphylococcus aureus PSMα1 (cat.no. CSB- EP314123FLF-1) and Recombinant Escherichia coli O157:H7 Major curlin subunit (CsgA) (cat.no. CSB-YP821944EOD-1) were purchased from Cusabio Technology LLC (Houston, TX, USA) and the stock solution was obtained by diluting in PBS with 33% glycerol. Cell culture, differentiation protocol and treatment Human neuroblastoma cell line SH-SY5Y was obtained from CLS Cell Line Service (Eppelheim, Germany). For the proliferative state, cells were grown in T75 filter vent flasks. The cell medium used was DMEM with 4500 mg/L glucose, L-glutamine, and 1% antibiotic–antimycotic, supplemented with 10% FBS, and cells were incubated in a dark humidified atmosphere of 5% CO2 at 37 °C. The medium was replaced every 2–3 days, and the cells were subcultured when reaching 90% confluence. The cells were used in experiments below passage 18 since the cell line was supplied by the manufacturer. For the dopaminergic differentiation, we used the protocol previously described in the literature (Fig. 1)67–70. For the experiments that measure mRNA levels of α-synuclein by RT-PCR and intracellular α-synuclein protein levels by Western Blot, cells were seeded in 6 well plates. At day 0, the cells were seeded at a density of 8 × 104 cells/ cm2 in a cell medium supplemented with 10% FBS. After 24 h, the medium was replaced with a differentiation medium that consisted of DMEM with 4500 mg/L glucose, L-glutamine, and 1% antibiotic–antimycotic, but supplemented with a lower concentration of FBS (3%), and 10 μM retinoic acid. After 72 h, the medium was replaced with differentiation medium, supplemented with 80 nM of phorbole-myristate-acetate (PMA) for another 72 h, after which the medium was changed, and the cells were treated for 48 h with microbial-derived molecules according to the experimental design. Cell morphology was constantly investigated by phase contrast microscopy, and NISelements software was used to capture images (Fig. 1). MTS viability and LDH cytotoxicity assays Viability and cytotoxicity assays were assessed on dopaminergic-differentiated SH-SY5Y cells in 96-well plates. Cells were seeded at a density of 4 × 104 cells/well and followed the dopaminergic differentiation protocol. Cells were then incubated with 100 µL of ½ serial dilutions of the selected microbial-derived molecules. After 48 h, viability and cytotoxicity tests were assessed based on two independent experiments, each one in triplicate. The viability was evaluated by MTS (3-(4,5-dimethyl thiazol-2-yl)5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) (CellTiter 96® AQueous One Solution Reagent, Promega). 20 µL of MTS + Phenazine methosulfate (PMS) One Solution Reagent were added in each well, and the plates were incubated for 2 h, according to manufacturer instructions. The optical density (OD) of the plates was read at a microplate reader at an absorbance of 490 nm. The absorbance of medium-only wells was measured in triplicates for each experiment. Controls were established in triplicate for each experiment with dopaminergic-differentiated SH-SY5Y cells incubated with the solvent of the treatment (vehicle-only controls). Viability was calculated as % of vehicle-only control. After the OD of the medium-only wells was subtracted from each result, the following formula was used: % of viability = (sample OD/vehicle-only control OD) × 100. The cytotoxicity was evaluated by lactate dehydrogenase (LDH) release (CytoTox 96® Non-Radioactive Cytotoxicity Assay, Promega). After the 48 h treatment incubation, 50 µL of the supernatant were transferred to fresh 96-well plates and 50 µL of the Cytotoxicity Reagent were added. The plates were incubated for 30 min at room temperature in the dark. After 30 min, the Stop Solution was added, and the OD of the plates was read by a microplate reader at an absorbance of 490 nm. Vehicle-only and positive controls were established, and the absorbance of medium-only wells was measured. The positive control (LDH max) represents the maximum LDH release after lysing the dopaminergic-differentiated neurons with 9% Triton X solution for 45 min. Cytotoxicity was calculated as % of the positive control. After the OD of the medium-only wells was subtracted from each result, the following formula was used: % of cytotoxicity = (sample OD/ positive control OD) × 100. Real-time cells monitoring by impedance readings Real-time monitoring of dopaminergic-differentiated SH-SY5Y cells under treatment was assessed by measuring cellular impedance using the xCELLigence platform (Agilent Technologies, Santa Clara, CA, USA). This system allows real-time cell analysis (RTCA) of the adhesion properties of cells by detection of variations in electrical impedance that are transformed to a dimensionless parameter, Cell Index (CI). The CI values registered for each sample provide information about cell number, proliferation rate, size/shape, and attachment quality71. Cells were seeded at a density of 3 × 104 cells/well on E-16 plates and observed during differentiation. Then, the dopaminergic neurons were treated with the aforementioned compounds at the concentrations established by the end-point experiments (MTS and LDH assays). When the cell’s media was changed, the readings were paused. The CI values registered every 30 min were normalized and processed using RTCA Software 2. Real-time quantitative PCR Total RNA was extracted using a PureLink® RNA Mini Kit Invitrogen (ThermoScientific, USA) according to the manufacturer’s instructions. The purity and integrity of RNA were evaluated using a nano-spectrophotometer NanoDrop 2000 (ThermoScientific, USA). 1000ng of RNA was retro-transcribed into cDNA using the SuperScript IV VILO Master Mix (ThermoScientific, USA) at the following conditions: 10 min at 37°C followed by 10 min at 25°C and 50 min at 30°C after adding SuperScript Master Mix. The reaction was terminated at 85°C for 5 min. cDNAs were stored at -20°C until the time of use. The RT-qPCR assays were performed with the SYBR™ Green PCR Master Mix. 1 µl of the 10 × cDNA dilution was used for amplification with 0.3 µM forward and reverse primers for alpha-synuclein. For qualitative detection of α-synuclein on QuantStudio™ 7 Flex Real-Time PCR System (Applied Biosystems), the program for PCR step was: 2 min at 50°C, one cycle; 10 min at 95°C, one cycle; 15 s at 95°C, 60 s at 60°C, 40 cycles. The 2-ΔΔCT method evaluated relative gene expression, with GADPH as the normalisation reference gene. Western blot The cells were lysed, and the cell culture was washed with ice-cold PBS. Following that the cells were detached with a cell scraper in RIPA cell lysis buffer with a 1% fresh protease inhibitor cocktail (Halt Protease Inhibitor Cocktail 78,429, Thermo Scientific). The mixture was pipetted into microcentrifuge tubes and incubated for 1 h on ice. The lysate was then clarified by spinning for 10 min at 12,000 RPM at 4°C. The supernatant was transferred to a fresh tube on ice and aliquoted to be stored at -20°C. For SDS-PAGE electrophoresis, the cell lysate was treated with 4× Laemmli buffer (Thermo Scientific), and the proteins were separated on a 15% acrylamide gel. The transfer was made on a nitrocellulose membrane at 80V ct., 90 min. Before 5% BSA blocking, the membrane was treated with a solution of 4% paraformaldehyde and 0.1% glutaraldehyde to block detachment of the target protein and improve the detection of alpha-synuclein (according to the method of Sasaki et al.—Sasaki, A., Arawaka, S., Sato, H. et al.72). The protein was detected with rabbit anti- α-synuclein primary monoclonal antibody (Recombinant anti-α-synuclein antibody [EPR20535], ab212184, Abcam), overnight at 4°C , followed by incubation with goat anti-rabbit secondary antibody (sc-2004, Santa Cruz), for 1 h at room temperature. The chemiluminescent method (ECL Substrate Kit (High Sensitivity) ab133406) allowed the visualization of proteins using a CCD camera (UVP ChemiDoc-It Imager). The normalization of α-synuclein was made with a structural protein, GAPDH. The images were analysed by densitometry to evaluate the relative amount of protein staining and the results were quantified in terms of optical density using ImageJ software. Statistical analysis Data are represented as mean ± standard error of mean (SEM) and were statistically analysed using GraphPad Prism v 9.5. The first step was to assess the normal distribution of the data, and Shapiro–Wilk test and Kolmogorov–Smirnov test were performed. Data were compared to the control group, and p < 0.05 was considered statistically significant. For parametric data one-way ANOVA and Dunnett’s multiple comparisons test were performed, while for non-parametric data Kruskal–Wallis test and Dunn’s multiple comparison test were performed. The results of the tests were represented on the graphics as *p < 0.05, **p < 0.01, ***p < 0.001. Conclusion Our in vitro study offers significant insights into the potential mechanisms by which microbiota-derived molecules, particularly rhamnolipid and LPS, contribute to PD) pathogenesis. While it does not definitively resolve the complexities surrounding the initiation of PD pathology within the gastrointestinal tract, it underscores the critical role these microbial compounds may play. Specifically, our findings suggest that dopaminergic-differentiated SH-SY5Y neurons serve as a viable model for both enteric and central nervous system neurons, reflecting the dual pathways through which these toxins can exert their effects. The upregulation of α-synuclein observed in response to bacterial amphiphiles introduces a novel mechanism of microbiota-induced PD pathogenesis, augmenting the established understanding of direct α-synuclein aggregation. This highlights the importance of further investigating the permeability changes in both the intestinal and blood–brain barriers, which could facilitate the transit of these microbial molecules from the gut to the brain. Our study strongly advocates for deeper exploration into the gut-brain axis and the role of microbiota in neurodegenerative diseases, emphasizing the need for comprehensive strategies to mitigate these microbial influences and potentially delay or prevent the onset of PD. Supplementary Information Supplementary Information. Supplementary Information.