Aggregation of alpha-synuclein in enteric neurons does not impact function in vitro

Abstract

Recent evidence implicates a gut-first pathogenesis in the enteric nervous system (ENS) within a portion of PD patients, yet in vitro investigations have primarily focused on the central nervous system. Here, the preformed fibril (PFF) PD model is applied with co-administered groups of butyrate and lipopolysaccharide to model the effects of the local gut microbiome. Significant PFF uptake and retention occur in isolated rat enteric neurons compared to untreated controls resulting in increasing immunostained aggregate conformation-specific, alpha-synuclein (a-Syn) average intensity between 6 µg PFF and untreated controls. Cortical neurons significantly retain PFFs with an increase in aggregated a-Syn average intensity within all dosages. Differences in growth cone morphology but not dynamics in PFF-treated ENS cultures occur. Electrophysiological recordings via a microelectrode array (MEA) indicate no overall difference in spontaneous spike rate. However, only untreated controls respond to PD-relevant dopamine stimulus, while 1 µg PFF and control populations respond to stimulus with ENS-abundant acetylcholine. Finally, no differences in substance P levels-correlated with PD and neurodegeneration-are observed. Overall, these findings suggest the ENS retains PFF dosage absent acute loss in function, however, does experience changes in growth cone morphology and dopamine-stimulated activity.

Figure 1

Schematic representation of the experimental design. (A) Illustration of the theorized preformed fibrils (PFF) toxic progression from the introduction of PFFs to uptake into neurons and development of further aggregation from endogenous α-synuclein (a-Syn). (B) Timeline of experimental design. Cultures were fed every 2–3 days throughout the experimental timeline. (C) Representative images of neuron populations (cortical and enteric) isolated for experimental procedures illustrated linking to anatomical location. Images are merged (Red = beta-3 tubulin, Blue = DAPI). Scale = 100 µm. Figure created with BioRender.

Figure 2

Live-cell imaging quantification of preformed fibril (PFF) intensity throughout experiments. (A) Average intensity of images across culture period. Box-cox transformed multilevel model accounting for experimental replicate and well with Tukey p-value adjustment used. N = 3–6, m = 9–21, images = 15–31. (B) Average intensity of images on day 21 of culture. Kruskal–Wallis test and Wilcoxon rank-sum multiple comparisons with Benjamini–Hochberg p-value adjustment utilized. N = 3–6, m = 8–20, images = 13–29. (C) Representative images of cultures over time between groups (orange = PFF fluorescent tag). Day 7 background signal is attributed to free-floating PFF after addition. (*p < 0.05, **p < 0.01, ***p < 0.001). Scale = 100 µm. Pixel intensity in uint8 (values range from 0 to 255). Each dot represents an image, error bars = SEM.

Figure 3

Representative images of neuron populations following 21 days of culture. ENS cultures were co-administered with LPS and/or butyrate either (A) alone (control), or (B) with PFFs. (C) Untreated cortical neurons. (D) PFF-dosed cortical neurons. (red = beta 3-tubulin (B3T), blue = DAPI, green = aggregated α-synuclein (a-Syn), orange = preformed fibril fluorescent tag). Scale = 20 µm.

Figure 4

Quantification of preformed fibril (PFF) tag and aggregated α-synuclein (a-Syn) intensity throughout experiments. 2.5 mM of Butyrate (But) and 5 µg/mL lipopolysaccharide (LPS) were added with every feeding following PFF administration. The average intensity of (A) PFF and (B) aggregated a-Syn within ENS neuron morphology. Sample size ENS: N = 3–7, m = 9–18, images = 60–123. The average intensity of (C) PFF and (D) aggregated a-Syn within cortical neuron morphology. Sample size cortex: N = 4–8, m = 12–27, images = 57–131. Pixel intensity in uint8 (values range from 0 to 255). Each dot represents an image, and error bars = SEM. Statistical analysis was conducted using a 6th root transformation multilevel model adjusting for experimental replicate and well (A) or experimental replicate alone (B–D) with Tukey p-value adjustment (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 5

Growth cone quantitative assessment of ENS cultures observed on day 14 of culture. (A) Growth cone representative diagram. Figure made with Biorender. (B) Representative brightfield image of a growth cone. Day 14 measurement of (C) filopodia count, (D) filopodial activity (sum of extensions and retraction events per time passed), (E) lamellipodium area, (F) filopodial activity normalized per filopodia. Statistical analyses were conducted with either logarithmic (C,E) or 4th root (D,F) transformation and either general linear model (D,E) or multilevel model adjusting for experimental replicate and well (F) or replicate alone (C). N = 4, m = 11–12, growth cones = 43–54. Each dot represents a growth cone. Error bars = SEM. (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 6

Functional assessment of neurons using microelectrode arrays (MEA). (A) Representative waveform (B) Schematic made in Biorender. (C) Merged image of neurons near 30 µm diameter electrodes (black). Orange = PFF tag. Scale = 100 µm. (D) Representative histogram of detected action potential rate as spikes per second from spontaneous activity recorded on day 20 (upper 7% of values cropped). Color indicates experimental group (red = control, green = 1 µg PFF, blue = 4 µg PFF). (E) Representative raster plot of spontaneous spike activity over time (seconds). (F) Estimated means of spike count (5-min period) by stimulation and PFF dosage (method 1). Error bar = SEM. N = 1–4, number of MEAs = 1–4, wells = 2–11, number of electrodes = 10–50. (N > 1 outside day 28, 4 µg PFF. Zero-inflated, negative binomial generalized multilevel model adjusting for culture day and nested random effects for experimental replicate and electrode with Tukey p-adjustment was used. (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 7

Substance P ELISA quantification of Day 21 spent medium demonstrates no significant differences between groups. N = 4–7. Error bars = SEM.