Background
Synucleinopathy is a collective term for neurodegenerative diseases such as Parkinson’s disease (PD), dementia with Lewy bodies (DLB), and multiple system atrophy (MSA). Initial evidence for a cell-non-autonomous mechanism underlying synucleinopathies came from the discovery of Lewy pathology in grafted cells in post mortem brains of PD patients [
1,
2]. Therefore, prion-like transmission of αS lesions was proposed and the concept of seeded aggregation – a nucleation-dependent process similar to the one described for prions and amyloid-beta (Aβ) – was suggested [
3,
4]. This concept gained further support by in vitro findings showing that the exogenous application of αS fibrils induced Lewy body-like pathology in cultured neurons [
5‐
8]. Similarly, induction and spreading of αS lesions was reported in mice using recombinant αS preformed fibrils (αS pff) or brain homogenates containing aggregated αS [
9‐
15].
Induction and propagation of proteopathic lesions are largely dependent on the interaction between the seed and the host environment [
16]. Spreading of αS lesions occurs along neural pathways and cell-to-cell transfer is cell-type dependent and promoted by neural activity [
11,
17‐
19]. Thus, seed propagation is best studied in a living environment that closely mimics the adult or aged brain. While mice have been instrumental in the past to study such prion-like propagation of pathogenic seeds [
16], mouse models are time-consuming and costly, and experimental manipulations to mechanistically understand propagation of seeds are challenging.
Studying propagation of pathogenic seeds in humans is even more intricate. Human stem cell-derived organoids and 3D-culture systems have been developed to mimic the human brain environment [
20‐
22], however, they remain controversial regarding their purity, maturity, and cell subtype-identity. Grafting of human stem cell derived tissue into adult mice is another exciting new development [
23,
24] but the influence of the murine host on the human transplant needs to be addressed further.
To study α-synuclein (αS) inclusion formation, we induced seeding in long-term murine hippocampal slice cultures obtained from postnatal mouse brain, which preserves the complex cellular brain microenvironment. We found inclusions in both neurons and microglia with spreading of the lesions along neural pathways. Both induction and propagation could be blocked by a human αS seed-recognizing antibody. To foster translation, progression of synucleinopathy and associated neurodegeneration in these cultures was monitored by assessing neurofilament light chain (NfL) protein levels in the culture medium, a biomarker used in preclinical animal studies as well as in clinical settings. We then applied the initial findings from the murine cultures to resection-derived adult human brain tissue cultures and succeeded in inducing αS inclusions in a true adult human brain environment.
Methods
Mice
Wildtype C57BL/6 J, heterozygous Thy1-h[A53T]αS transgenic (tg) [
25], and Snca
−/− (C57BL/6-Snca
tm1Rosl) [
26] pups were used for preparation of HSCs. Thy1-h[A53T]αS tg mice overexpress human mutant (A53T) αS under control of a Thy1-promoter element. Experimental procedures were carried out in accordance with the veterinary office regulations of Baden-Wuerttemberg (Germany) and were approved by the local Animal Care and Use Committees.
Preparation of mouse hippocampal slice cultures (HSCs)
HSCs were prepared from pups at postnatal day 4–6 (P4–6) according to previously published protocols [
27,
28]. After decapitation, brains of the pups were aseptically removed, the hippocampi were dissected and cut perpendicular to the longitudinal axis into 350 μm sections with a tissue chopper. Carefully selected intact hippocampal sections were transferred into petri dishes containing an ice-cold buffer solution (minimum essential medium (MEM) supplemented with 2 mM GlutaMAX™ at pH 7.3). Three sections were placed onto a humidified porous polyethylene (PTFE) membrane insert (Merck Millipore, PICM0RG50) and into a 6-well plate with 1.2 ml culture medium (20% heat-inactivated horse serum in 1x MEM complemented with GlutaMax™ (1 mM), ascorbic acid (0.00125%), insulin (1 μg/ml), CaCl
2 (1 mM), MgSO
4 (2 mM) and D-glucose (13 mM) adjusted to pH 7.3) per well. HSCs were kept at 37 °C in humidified CO
2-enriched atmosphere. The culture medium was completely changed three times per week.
Preparation and treatment of human slice cultures
Approval (#338/2014BO2) of the ethics committee of the University of Tuebingen as well as written informed consent was obtained from all patients and allowed spare tissue from resective brain surgery to be used for human slice cultures. Human neocortical slice cultures were prepared according to a previously published protocol [
29]. Access tissue from temporal lobe was obtained from patients who had undergone resective brain surgery (patient 1: 61 years old, female, epilepsy due to ganglioglioma; patient 2: 49 years old, male, epilepsy due to cortical tubers; patient 3: 22 years old, male, epilepsy due to hippocampus sclerosis). Tissue was transported from the surgery room to the laboratory in oxygenated (95% O
2 and 5% CO
2) ice-cold artificial cerebrospinal fluid (aCSF; 110 mM C
5H
14ClNO, 26 mM NaHCO
3, 10 mM D-glucose, 11.6 mM Na-ascorbate, 7 mM MgCl
2, 3.1 mM Na-pyruvate, 2.5 mM KCl, 1.25 mM NaH
2PO
4, and 0.5 CaCl
2). The tissue was cut into 250 μm slices perpendicular to the cortical surface using a vibratome. Afterwards, the slices were kept in aCSF equilibrated with carbogen for 30 min at room temperature (RT) before they were transferred onto culture membrane inserts (Merck Millipore, PICM0RG50) in a 6-well plate for cultivation. For the first hour following the slicing procedure, the slices were cultured in 1.5 ml NSC media (48% DMEM/F-12 (Life Technologies), 48% Neurobasal (Life Technologies), 1x N-2 (Capricorn Scientific), 1x B-27 (Capricorn Scientific), 1x Glutamax (Life Technologies), 1x NEAA (Life Technologies) + 20 mM HEPES. From then on, human slice cultures were grown in human cerebrospinal fluid (hCSF) obtained from normal pressure hydrocephalus patients via lumbar puncture [
29]. Approval of the ethics committee of the University of Tuebingen as well as written informed consent from all patients was obtained (#338/2016A). The hCSF was collected, pooled and centrifuged at 2,000 x g at 4 °C for 10 min. The supernatant was kept at − 80 °C and thawed at RT before changing the medium. Cultures were kept in 1.5 ml hCSF at 37 °C in humidified CO
2-enriched atmosphere. The hCSF was completely changed three times per week. For αS overexpression, AAV1/2-CMV/CBA-human-A53T-α-synuclein-WPRE-BGH-polyA virus (AAV-hA53T-αS) (Vigene Biosciences, titer of 5 × 10
12) was injected evenly spaced, once per 5 mm
2 of the slice, using a picospritzer (PDES-O2DX/NPI electronics, Tamm, Germany).
Brain-derived αS seeds and αS pre-formed fibrils
For brain-derived seeds, brain homogenates were prepared as previously described [
13]. In short, fresh frozen pooled brainstems from spontaneously ill male and female Thy1-h[A53T]αS tg mice (7–8 months old,
n = 3 brainstems per homogenate) and corresponding non-tg controls (20 months old,
n = 2) were used. Homogenisation at 10% (w/v) was done in sterile PBS (Pre-cellys, 4 × 10 s at 5500 rpm), followed by vortexing and centrifugation at 3,000 x g for 5 min at 4 °C. The supernatants were collected, aliquoted and stored at −80 °C until use.
Expression in
E. coli, purification and quality control of human recombinant monomeric wt αS was done as previously described [
30]. For fibril formation, soluble wt αS was incubated in Tris-HCl buffer (50 mM Tris-HCl, pH 7.5, 150 mM KCl) at 37 °C under continuous shaking for 5 days and formation of fibrils was assessed with Thioflavin T. The fibrils were quality checked by transmission electron microscopy after negative staining before and after fragmentation as reported previously [
30,
31]. In addition, the fibrils limited proteolytic pattern was also assessed as previously reported [
30,
31]. Fragmentation was achieved by sonication using a sonotrode (sonication for 20 min, 0.5 s pulses; Sonicator UIS250V, equipped with VialTweeter, Hielscher Ultrasound Technology, Germany). The fibrils were imaged using a Jeol 1400 transmission electron microscope following their adsorption onto carbon-coated 200 mesh grids and negative staining with 1% uranyl acetate. The images were recorded using a Gatan Orius CCD camera (Gatan Inc., Pleasanton, CA, USA) (Supplementary Fig.
1). The average size of the fibrils after fragmentation (47 ± 5 nm) was derived from length distribution measurements and their average molecular weight (16,200 kDa) was derived from analytical ultracentrifugation sedimentation velocity measurements. The fibrils (350 μM) were aliquoted (6 μl per tube), flash frozen in liquid nitrogen and stored at − 80 °C. Fibrils were labeled with NHS-ester ATTO-550 as previously described [
32].
Seeding of the cultures
Murine HSCs were kept for 10 days in vitro (DIV-10) without any experimental treatment. At DIV-10, 1 μl of αS pff (350 μM or dilutions thereof) or brainstem homogenate of Thy1-h[A53T]αS tg or wt mice was pipetted on top of each culture. Human slice cultures were kept until DIV-3 without any experimental treatment. At DIV-3, 1 μl of αS pff (35 μM) was pipetted on top of the culture.
Local injection of αS pff into HSCs was performed into cornu ammonis 3 (CA3) at DIV-10. CA3 was identified as the region next to dentate gyrus (DG), which can be observed by light microscopy (dense cell layer resembling a horseshoe). Microinjection pipette (Science Products GmbH, GB150TF-10) was pulled using a micropipette puller (Sutter instruments, Model P-97; settings: 1 cycle, heat = 520, pull = 50, velocity = 50, time = 250). To immediately visualise the injection, 0.3 μl FastGreen dye (Carl Roth) was added to a 6 μl aliquot of ATTO-550-labelled or unlabelled αS pff (350 μM), upon loading the pipette. The loaded pipette was then inserted into the holder of a picospritzer (PDES-O2DX/ NPI electronics, Tamm, Germany), and the very end of the tip was carefully broken under visual guidance with sterile forceps, until a pressure pulse of 10 ms was able to release a small droplet of αS pff from the tip. Post-injection microscopic examination of the broken tip revealed an opening size of 25–40 μm. HSCs in their 6-well-plates were placed under the light microscope, and the pipette was carefully inserted into CA3. With a pressure pulse of 10 to 20 ms (10 ms for larger tip openings, 20 ms for smaller tip openings), a small volume of αS pff was injected. After having injected all slices, (approximately 1 h later) the medium was changed.
Antibody treatment
The antibody HLu-3 is a human IgG1-recognizing αS. The epitope of the antibody was determined to be amino acid 113–115 of human αS, using arrays of overlapping linear peptides at Pepscan (Pepscan Zuidersluisweg 28,243 RC Lelystad, The Netherlands). The affinity to human, mouse and cynomolgus αS monomers is determined to be 31 nM by surface plasmon resonance (BIAcore® 3000). HLu-3 has approximately a 200-fold avidity shift in the binding to fibrillated forms of human αS (determined by competition ELISA). HLu-3 has no cross-reactivity to beta- & gamma-synuclein. Reference item used in the study was a negative isotype-matched control, anti-HIV-1 gp120 human IgG1 antibody (b12).
HLu-3 or control antibody were either mixed with human αS pff (1:1 in PBS to obtain a final concentration of 35 μM) and drop seeded on top of each HSC at DIV-10, or added to the medium. For the latter, antibodies were added to pre-warmed culture medium upon medium change (final concentration of 350 nM). Depending on experimental scheme, antibody-supplemented medium was used 7 days prior to the application of αS pff seeds (-7 days post-injection, dpi) or 1 h after αS pff seeds (0 dpi). Thereafter, the cultures were treated with antibody-supplemented medium with every medium change.
Histological analysis of cultures
Cultures were fixed with 4% paraformaldehyde (PFA) in PBS at pH 7.4 for 2 h (HSCs) or for 24 h (human slice cultures). After fixation, HSCs were rinsed 3 times with 0.1 M PBS for 10 min and stored in PBS at 4 °C for up to 1 month until sectioning. Human slice cultures were rinsed 3 times with PBS supplemented with 0.2% TritonX-100 (PBST) and incubated with PBST overnight for permeabilization. The membrane carrying the fixed HSCs was cut out and mounted onto a planar agar block. With a vibratome (Leica VT 1000S Vibratome, Leica Bio-systems), the cultures were sliced into 50 μm sections. Typically, 5–6 intact sections per HSC were obtained and collected in PBS to be stained within 1 week. Human slice cultures were stained unprocessed.
Antigen retrieval was performed by heating the sections in 10 mM citrate buffer (1.8 mM citric acid, 8.2 mM trisodium citrate, pH 6.0) at 90 °C for 35 min (HSCs) or overnight at 4 °C and subsequently 30 s at 90 °C (human slice cultures). Sections were blocked with 5% NGS (HSCs, 2 h) or 1% NGS (human slice cultures, overnight at 4 °C) and 0.3% PBST. For detection of αS phosphorylated at Ser-129, a rabbit monoclonal pS129 antibody (Abcam, EP1536Y, Cat# ab51253, 1:1,000) was used. For microglia detection rabbit monoclonal iba1 (Wako Chemicals GmbH, Cat# 019–19,741, 1:250) and for neuronal staining and structure mouse monoclonal NeuN antibody (Millipore GmbH, Cat# MAB377, 1:500) and chicken anti-MAP2 (Abcam, Cat# ab5392, 1:500) were used. Following Alexa-fluorophore-conjugated secondary antibodies were applied in a concentration of 1:250: goat-anti-rabbit Alexa-568 (Thermo Fisher, Cat# A11011); goat-anti-mouse Alexa-488 (Thermo Fisher, Cat# A11001); goat-anti-mouse Alexa-568 (Invitrogen, Cat# A11004); goat-anti-rabbit Alexa-633 (Thermo Fisher, Cat# A21070); goat-anti-chicken Alexa-488 (Invitrogen, Cat# A21467). DAPI counterstaining was performed at a concentration of 1:500.
For staining with amyloid binding dyes pentamer formyl thiophene acetic acid (pFTAA) [
33,
34] and thioflavin S (ThioS), sections were incubated for 1 h with either freshly prepared pFTAA (1.5 mM in de-ionized water, used at 1:500 in PBS) or ThioS (Sigma-Aldrich, Cat# T1892; 1% w/v ThioS in milliQ H
2O). ThioS-stained sections were washed 2 x in 70% EtOH and for 10 min. Slices were transferred on glass slides and coverslipped with Dako Fluorescence mounting medium (Biozol Diagnostika, Cat# S3023).
Sections were analyzed using an Axioplan2 imaging microscope (Zeiss, Jena, Germany) and digitised with an AxioCam HRm black and white camera (Zeiss) using AxioVision 4.8 software (Zeiss). With a Plan Neofluar 10x/0.50 objective lens (Zeiss), 16-bit RGB mosaics of the whole culture were obtained with a resolution of 170 pixels / μm. High resolution images were acquired using a Zeiss LSM 510 META (Axiovert 200 M) confocal microscope with an oil immersion 40 × /1.3 or 63 × /1.4 Plan Apochromat objective and LSM software 4.2 (Carl Zeiss). Sequential excitation of fluorophores was performed using lasers with the wavelength 405 nm (DAPI), 488 nm (Alexa-488 coupled secondary antibodies, ThioS, pFTAA), 543 nm (Alexa-568 coupled secondary antibodies, pFTAA), and 633 nm (Alexa-633 coupled antibodies).
Quantification of immunohistochemical stainings
For quantification of pS129- and ThioS-positive inclusions in HSCs, whole culture mosaic images were acquired on an Axioplan2 imaging microscope as described above. Cultures with sectioning artefacts or cultures that were injected in the wrong site (e.g. into CA1 instead of CA3) were excluded from the analysis with FIJI ImageJ (version 2.1.0/1.53c). Images were blinded, colour channels were split, background was subtracted (rolling ball radius 50 pixels), and the intensity threshold was manually adjusted. To exclude unspecific staining of ThioS, the particle size of signal in the green channel was limited to 20–200 μm2. On each mosaic, the percentage of pS129- and ThioS-positive signal over the whole culture was calculated.
To quantify percentage of pS129 and ThioS-signal in the injection site and other hippocampal subregions, images were blinded, regions of interest (ROIs) for CA3, CA2, CA1, DG and subiculum were selected based on the brain map of the Allan Mouse Brain Atlas (2004) (P56, coronal) and stored in FIJI before channels were split. Background was subtracted and intensity threshold for pS129 or ThioS was adjusted for the whole culture. For each ROI, the percentage of pS129- and ThioS- positive area over the respective ROI area was calculated.
Heatmaps
To demonstrate spreading through hippocampal subregions, heatmaps were prepared by illustrating the mean percentage of pS129- or ThioS area obtained in the ROI analysis (see “
Quantification of immunohistochemical stainings”) by a colour code. Using Microsoft Excel (v.16), mean values were assigned to a colour based on a 3-colour-scale from white (#FFFFFF) via yellow (#FDBF2D) to red (#BE0712). The colours were identified using Adobe Photoshop CS5 (v.12) and applied to the respective area of a hippocampal map in Adobe Illustrator CS5 (v.15).
Biochemical analysis of cultures
Slice culture homogenates were prepared from treated or untreated cultures. Cultures were removed from the membrane, pooled (n = 16) and immediately frozen on dry ice and stored at − 80 °C until use. Frozen slice cultures were homogenised with a syringe in 160 μl sterile PBS, aliquoted and stored at − 80 °C until further use. For immunoassays (Western blotting) homogenates in PBS were shifted to high salt (HS) buffer (50 mM Tris-HCl pH 7.5, 750 mM NaCl, 5 mM EDTA, 1% phosphatase and protease inhibitor cocktails). 100 μL of homogenate were incubated on ice in N-lauroylsarcosyl (Sigma, Cat# 61747, Saint-Quentin-Fallavier, France) at a final concentration of 10%, and were left on ice for 15 min before they were loaded on a 10% sucrose cushion and ultracentrifuged at 186,000 x g for 1 h 10 min at 4 °C. The supernatant was collected, and the pellet was resuspended in sample buffer, and sample buffer was also added to the supernatant. Proteins were separated in a 4–12% SDS NuPage Gel and electroblotted onto Amersham nitrocellulose membranes (VWR International Merck Eurolab, Cat# 10600001). Membranes were incubated with 0.4% PFA for 30 min, and then saturated with 5% dry milk in 0.1% PBS-Tween20 (0.1%). Monoclonal rabbit antibody against pS129 (AbCam, Cat# ab51253) at 1:1,000 dilution to detect phosphorylated αS species, monoclonal mouse antibody against αS (BD Transduction Laboratories, Cat# 610786) at 1:1,000 dilution to detect total αS, and rabbit β-actin antibody (AbCam, Cat# ab8227) as a loading control were used. Membranes were then incubated for 1 h with anti-rabbit antibody or anti-mouse antibody at 1:20,000 for 1 h at RT. Samples were visualized with chemiluminescence using SuperSignal West Dura Extended or Pico (both Thermo Scientific).
Immunoassay for total αS measurements in brain homogenate
For αS measurements, brain homogenates were extracted as follows: aliquots were thawed on ice, mixed 1:3.2 with cold formic acid (FA) (minimum 96% purity; Sigma, St. Louis, MO, USA), sonicated for 35 s at 4 °C, and spun at 25,000 g at 4 C for 1 h. The supernatant was equilibrated (1:20) in neutralization buffer (1 M Tris base, 0.5 M Na2HPO4, 0.05% NaN3).
Concentrations of human αS in brain homogenates were determined with an electrochemiluminescence-linked immunoassay using the MSD Human α-Synuclein Kit (Meso Scale Discovery, Gaithersburg, MD, USA), or by Single Molecule Array (Simoa) technology using the Simoa™ Human Alpha-Synuclein Discovery Kit (Quanterix, Billerica, MA, USA) according to manufacturer’s instructions. FA-soluble brain homogenates were diluted up to 1:10,000 in Diluent 35 (Meso Scale Discovery) or 1:100 in Alpha-Synuclein Sample Diluent (Simoa) before the measurement, and analyzed in duplicates on a Mesoscale Sector Imager 6000 or a Simoa HD-1 Analyzer. MSD DISCOVERY WORKBENCH software 3.0 or Simoa Software Version 1.5 for HD-1 Analyzer was used for data analysis. Internal reference samples were used as controls on every plate.
Immunoassay for aggregated αS measurements in brain homogenate and pre-formed fibrils
αS aggregates were measured using a HTRF-FRET assay developed by Cisbio (#6FASYPEG, Cisbio). αS pff and tg brain homogenate were serially diluted and a HTRF-FRET signal measured on a PHERAstar (BMG LABTECH) using 337 nm laser excitation, simultaneous dual emission 665 nm / 620 nm and HTRF technology. Data is reported as 665 nm / 620 nm x 10,000. αS pff needed to be diluted > 50,000 fold to be on the proper side of the hook effect of the assay, whereas the tg brain homogenate did not show hook effect issues at any dilutions. For αS pff, a dilution of 204,800-fold resulted in a signal of approximately 20,000, whereas the tg brain homogenate only was diluted 25-fold to reach a similar aggregation level. The difference in dilutions (e.g. 204,800-fold vs 25-fold) to reach a similar aggregation level was used to estimate the relative αS aggregation level per volume of sample. Results over several dilutions were combined to reach the final result of 8,000 x more aggregates in αS pff relative to tg brain homogenate per volume.
NfL immunoassay
Culture medium was collected, aliquoted, and kept at −80 °C until use. NfL concentrations were determined by Single Molecule Array (Simoa) technology using the highly sensitive Simoa™ NF-Light Advantage Kit (Quanterix, Billerica, MA, USA) according to manufacturer’s instructions [
35]. Medium samples were pre-diluted 1:10 or 1:50 in NF-Light sample diluent and measured in duplicates on a Simoa HD-1 Analyzer (Quanterix). Internal reference samples were used as controls on every plate. Detection limit of NfL was 0.038 pg/ml.
Culture thickness
While sectioning the freshly fixed cultures with a vibratome (see above), the amount of 50 μm sections was assessed. Although the first and the last section sometimes were not exactly 50 μm, the amount of sections × 50 μm equals roughly the culture thickness at the time of fixation.
Intracerebral injection of αS tg mice
Male and female 3–4-month-old Thy1-h[A53T]αS tg mice were anesthetized using a mixture of ketamine (100 mg/kg body weight) and xylazine (10 mg/kg body weight) in saline. 2.5 μl of brain homogenate (see above) was then infused bilaterally into the dorsal hippocampus (AP − 2.5 mm, L ± 2.0 mm, DV − 1.8 mm) by stereotactic injection. Injection speed was 1.25 μl/min, and the needle was slowly removed after being kept in place for an additional 2 min. The surgical area was cleaned with sterile saline and the incision was sutured. Mice were kept under infrared light for warmth and monitored until recovery from anaesthesia.
After incubation periods of up to 30 days, mice were perfused for 5 min with ice-cold PBS. Brains were removed and immersion-fixed in 4% PFA in 0.1 M PB at pH 7.4 for 48 h, and then placed in 30% sucrose in PBS for 48 h. Brains were frozen in 2-methylbutane, cooled with dry ice, and then serially cut into 25 μm sagittal sections using a freezing-sliding microtome. The sections were collected in cryoprotectant (35% ethylene glycol, 25% glycerol in PBS) and stored at − 20 °C until use. Immunohistochemical staining of αS inclusions was done as described for HSCs (see above).
Statistical analysis
All statistics were performed using PRISM software (GraphPad v.9). Data were tested for normality using the Shapiro-Wilk test. If the groups passed the normality test and only two groups were compared, an unpaired two-tailed t-test was applied. If the two groups did not pass the normality test, the non-parametric Mann-Whitney test was used for comparison. To estimate the influence of two different variables, a two-way-ANOVA was performed. If ANOVA revealed significant effects, post hoc Bonferroni’s multiple comparisons test was used. The mean and standard error of the mean (SEM) are reported for each experimental group.
Discussion
Neurodegenerative diseases are characterised by proteopathic lesions which develop many years, if not decades before the first symptoms and lesions at death are likely to be different from the ones driving the disease [
16]. Therefore, understanding the critical early phase of seeding and finding biomarkers to monitor disease progression require disease relevant models. Here we present αS lesion development in a mouse hippocampal slice culture model and find that microglial inclusions and NfL release follow neuronal αS inclusions. Moreover, we show that NfL release and spreading of αS lesions can be inhibited by a human αS antibody. Finally, we demonstrate that αS lesions can also be induced in adult human brain slice cultures.
Previous efforts to develop organotypic murine slice cultures of synucleinopathies mainly used viral overexpression of αS [
42]. In this study, we expand these endeavours and show that one-time seeding of brain-derived or synthetic αS seeds is sufficient for the induction and subsequent spreading of αS lesions in murine hippocampal slice cultures. Neuronal αS inclusions in the cultures appear mature and their biochemical properties (phosphorylation of serine 129 and sarkosyl insolubility) similar to previously observed in cultures [
43,
44], as wells as αS-transgenic or wildtype mice in which the lesions have been induced by seeded aggregation [
9‐
12]. The present findings of slower αS lesions induction and overall lower burden of pathology in wt compared to tg cultures are in line with observations in mouse models [
9,
10] and can be linked to enhanced vulnerability of neuronal populations with high αS expression levels [
45,
46]. The morphology of the induced neuronal αS lesions in the cultures was dependent on the αS seed type (αS-tg mouse brain-derived or synthetic αS pff), again, similar to what has been described for seeded induction in mice [
9‐
11,
47‐
49]. These observations are consistent with prion-like templated propagation [
4] and structural differences between pff and brain-derived human αS seeds [
50]. In the slice cultures as well as in mice, highly concentrated αS pff are more seeding active compared to the tg mouse brain material. Hence, when comparing the specific activity of αS pff with tg brain-derived aggregates (seed activity per αS molecules), brain-derived αS seeds appear to be more potent in seeding aggregation consistent with other proteopathic seeds [
16].
The finding that αS lesions in murine slice cultures are positive for the amyloid-binding dye pFTAA is in line with recent data in humans using pFTAA and other luminescent-conjugated oligothiophenes [
51,
52]. This not only confirms the amyloid nature of the inclusions in the cultures, but it also opens up the possibility to use these fluorescent dyes for live imaging of αS lesions and for spectral discrimination of αS conformers [
52‐
54].
A growing body of research has pinpointed the importance of neuroinflammation as an essential contributor to the pathogenesis of synucleinopathies [
55]. As recently reported for αS tg mice [
36], we also found in the cultures pFTAA- and ThioS-positive (but largely pS129-negative) inclusions in microglia. However, microglial inclusions developed with a delay of 2–3 weeks compared to the neuronal inclusions. The local appearance of the microglia inclusions was always linked to the neuronal inclusions both in magnitude and location. Intriguingly, targeting αS seeds with an αS antibody blocked both, neuronal and microglial inclusions. Although final proof of the nature and conformation of the microglia inclusions remains to be established, the present observations are consistent with the view that the microglial inclusions contain αS of neuronal origins [
36] and that microglia are somehow involved in the spreading of αS lesions in brain [
56].
Modelling proteopathic lesions in slices with a brain-like environment bears the advantage that the cell-to-cell transfer of αS aggregates between interconnected regions may occur in a similar way to that in vivo [
16,
57]. Immunotherapy targeting aggregated αS is a vigorously pursued therapeutic strategy, although its mechanism of action remains to be further investigated. Since it is not expected that antibodies readily enter intact neurons, αS antibodies that neutralise αS seeds are thought to capture the seeds extracellularly when transferred from cell to cell [
58‐
62]. The observation that an αS antibody was capable of reducing the number of αS inclusions in DG and CA1 where the αS lesions spread, but less so in the CA3 injection area, suggests that this human antibody can inhibit the spreading of αS seeds. Although, the current results do not provide a conclusive explanation for the mechanisms involved, our results demonstrate the utility of this culture system for studying αS-targeting disease-modifying drugs.
Early biomarkers that indicate the initial development of proteopathic lesions before the occurrence of clinical symptoms are essential for early diagnosis and disease progression monitoring. An increase of NfL in CSF and blood has been observed in αS-tg mice as well as in human synucleinopathies [
38,
63]. The present findings that αS lesions and associated neurodegeneration in murine slice cultures can be monitored by measuring NfL levels in the culture medium reveals the culture system as an important tool for translational research. NfL measurements in cultures provide even some advantages over animal models (i.e. NfL in CSF) since longitudinal measurements are possible and changes of NfL levels can be directly related to lesions as well as to therapeutic efforts.
While murine slice cultures now allow to study the formation, spreading, and targeting of the inclusions in a brain-like environment, it is important to acknowledge their limitations given that these cultures are derived from postnatal brain and therefore lack the aspect of aging. Furthermore, recent transcriptome studies revealed differences in disease-relevant genes between mouse and human cell populations [
64‐
68]. Thus, the here described proof-of-principle translation of the slice culture model from mouse to human appears a first step forward and will allow to confirm findings derived from mouse cultures in a true aged human brain environment.
Human long-term slice cultures have only recently been developed [
29,
69]. The cultures used in the present study were derived from three individuals (22–61 years old) of both sexes. Since this small sample size does not allow any conclusions about the effects of age and gender, such parameters could be addressed in the future by increasing the donor sample size. In particular, it will be interesting to study whether αS lesion induction is more prominent in aged donors compared to younger donors. Other open questions are whether longer culturing of the human cultures will allow the induction of αS lesions without additional overexpression of A53T-αS and whether microglia inclusions will also develop at prolonged culturing.
The induction of αS lesions in adult human brain cultures is an exciting step forward. However, limitations remain and need to be overcome before the human culture model can be used in a routine lab environment. The availability of suited resection tissue is limited and long-term cultivation of human cultures appears more difficult compared to the murine cultures. In addition, human CSF is needed to nurture the human cultures whose availability is also limited. Thus, at present, the human slice culture model is best used to confirm findings from the mouse cultures in a true aged human brain environment.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.