Primary cortical cell tri-culture to study effects of amyloid-β on microglia function and neuroinflammatory response

Abstract

Background: Microglia play a critical role in neurodegenerative disorders, such as Alzheimer's disease, where alterations in microglial function may result in pathogenic amyloid-β (Aβ) accumulation, chronic neuroinflammation, and deleterious effects on neuronal function. However, studying these complex factors in vivo, where numerous confounding processes exist, is challenging, and until recently, in vitro models have not allowed sustained culture of critical cell types in the same culture.

Objective: We employed a rat primary tri-culture (neurons, astrocytes, and microglia) model and compared it to co-culture (neurons and astrocytes) and mono-culture (microglia) to study microglial function (i.e., motility and Aβ clearance) and proteomic response to exogenous Aβ.

Methods: The cultures were exposed to fluorescently-labeled Aβ (FITC-Aβ) particles for varying durations. Epifluorescence microscopy images were analyzed to quantify the number of FITC-Aβ particles and assess cytomorphological features. Cytokine profiles from conditioned media were obtained. Live-cell imaging was employed to extract microglia motility parameters.

Results: FITC-Aβ particles were more effectively cleared in the tri-culture compared to the co-culture. This was attributed to microglia engulfing FITC-Aβ particles, as confirmed via epifluorescence and confocal microscopy. FITC-Aβ treatment significantly increased microglia size, but had no significant effect on neuronal surface coverage or astrocyte size. Upon FITC-Aβ treatment, there was a significant increase in proinflammatory cytokines in tri-culture, but not in co-culture. Aβ treatment altered microglia motility evident as a swarming-like motion.

Conclusions: The results suggest that neuron-astrocyte-microglia interactions influence microglia function and highlight the utility of the tri-culture model for studies of neuroinflammation, neurodegeneration, and cell-cell communication.

Keywords: Alzheimer's disease; amyloid-beta; cell motility; cytokine profile; live cell imaging; microglia; neural cell culture; neuroinflammation; phagocytosis.

Figure 1.

Aβ clearance analysis. Representative epifluorescence images displaying microglia (Iba1, yellow), green FITC-Aβ particles, and counterstained nuclei (DAPI, blue) in (A) co-culture and (B) tri-culture after a 96-h exposure to 1 μM FITC-Aβ initiated at DIV 10. Scale bar = 100 μm. (C) A confocal microscopy image of the tri-culture displays colocalization of green FITC-Aβ with microglia. The cultures were stained to identify neurons βIII-tubulin, red), microglia (Iba1, white), and nuclei (DAPI, blue) Scale bar = 50 μm. (D) Change in number of particles per FOV area and (E) size distribution of FITC-Aβ particles in the co-culture and tri-culture (mean ± SD, n = 3 biological replicates with at least 3 wells or slides per biological replicate with 5 FOV per well/slide). (F) Number of particles (normalized to the number of particles in the initial FOV) of FITC-Aβ particles in the tri-culture was monitored every 4 h up until 52 h. **** p < 0.0001 as determined by using two-way ANOVA with post hoc Tukey’s test.

Figure 2.

Cell morphological analysis. (A, B) Representative epifluorescence images of co-cultures and tri-cultures after a 96-h exposure to 1 μM FITC-Aβ initiated at DIV 10. The cultures were stained to identify neurons (βIII-tubulin), microglia (Iba1), and astrocytes (GFAP). The scale bar is 100 μm in all images except for 20 μm in the inset images displaying representative single microglia. (C) Percent neuronal surface coverage. (D) Microglia size (E) Astrocyte size. These endpoints were quantified and compared using two-way ANOVA with post hoc Tukey’s test for neurons and astrocytes and Welch’s t-test for microglia size. The plots display the mean ± SD, n = 3. (individual data points represents the three biological replicates with at least 3 wells or slides per biological replicate with 5 fields-of-view per well/slide). * p < 0.05.

Figure 3.

Proteomic analysis of conditioned media. (A) The cytokine heatmap displays the logarithm of relative fluorescence intensity. Only cytokines that resulted in a two-fold or higher change of the raw fluorescence intensity for any pairwise comparison between the four conditions (i.e., co-culture with/without Aβ and tri-culture with/without Aβ) are included. A full heatmap can be found in (Supplemental Figure 5). Hierarchical cluster analysis revealed three major groups of cytokines that increased in the co-culture (green), increased in the tri-culture (red), and increased only in Aβ-treated tri-culture (blue). Log-scale relative fluorescence or concentration of (B) pro-/anti-inflammatory cytokines and (C) chemokines that play a role in microglia motility. All plots display mean ± SD (n = 3 biological replicates with at least 3 wells per biological replicate pooled into one sample). Distinct letters above the bars indicate statistically-significant groups compared to the other groups, as determined with a two-way ANOVA with post hoc Tukey’s test, where p < 0.05 is considered statistically-significant.

Figure 4.

Microglia motility analysis. Mono- and tri-cultures with or without FITC-Aβ treatment were measured every 5 min up to 4 h. The composite cell trajectory of microglia in (A) mono-cultures and (B) tri-cultures, with initial position of all cells translated to zero coordinate for normalization (plotted n = 50 cells from one biological replicate with 5 fields of view from multiple wells). The cells were incubated with Aβ for 1 h and the time-lapse images were taken for 4 h. (E) Euclidean distance, (F) microglia speed, and (G) average directionality ratio in mono- and tri-cultures with or without FITC-Aβ treatment (mean ± SD, n = 3). Two-way ANOVA with post hoc Tukey’s test was used to compare the groups, where *p < 0.05, **p < 0.01, ***p < 0.001 show levels of statistical significance while “ns” indicate no statistical significance.