Altered Ca2+ responses and antioxidant properties in Friedreich's ataxia-like cerebellar astrocytes

Abstract

Friedreich's ataxia (FRDA) is a neurodegenerative disorder characterized by severe neurological signs, affecting the peripheral and central nervous system, caused by reduced frataxin protein (FXN) levels. Although several studies have highlighted cellular dysfunctions in neurons, there is limited information on the effects of FXN depletion in astrocytes and on the potential non-cell autonomous mechanisms affecting neurons in FRDA. In this study, we generated a model of FRDA cerebellar astrocytes to unveil phenotypic alterations that might contribute to cerebellar atrophy. We treated primary cerebellar astrocytes with an RNA interference-based approach, to achieve a reduction of FXN comparable to that observed in individuals with FRDA. These FRDA-like astrocytes display some typical features of the disease, such as an increase of oxidative stress and a depletion of glutathione content. Moreover, FRDA-like astrocytes exhibit decreased Ca2+ responses to purinergic stimuli. Our findings shed light on cellular changes caused by FXN downregulation in cerebellar astrocytes, likely impairing their complex interaction with neurons. The potentially impaired ability to provide neuronal cells with glutathione or to release neuromodulators in a Ca2+-dependent manner could affect neuronal function, contributing to neurodegeneration.

Keywords: Ca2+ signalling; Cerebellar astrocytes; Friedreich's ataxia; Mitochondria; Oxidative stress.

Fig. 1.

Frataxin knockdown in cerebellar astrocytes. (A) Representation of the pLKO.1-TRC cloning vector encoding one of three different shRNAs (either scrambled, sh535 or sh380) and the enhanced green fluorescent protein (EGFP). The expression of shRNAs is driven by the hU6 promoter. Ampicillin resistance was used for plasmid selection. (B) The graph shows the percentage of lentivirally transduced astrocytes assessed by high-throughput microscopy; n=10,107 cells for scrambled, 9771 for sh535, and 8263 for sh380, from five biological replicates; each dot represents the mean percentage of transduced cells in a single culture well. (C) Western blot of protein lysates derived from wild-type (wt), scrambled-, sh535- and sh380-transduced astrocytes, immunostained with antibodies detecting FXN (15 kDa). GFP (25 kDa) was used to assess the transduction efficiency; α-tubulin (50 kDa) was used as a loading control. (D) FXN protein levels, normalized to α-tubulin levels, in scrambled-, sh535-, and sh380-transduced astrocytes, compared to wt astrocytes. Data are expressed as mean±s.e.m.; n=6; *P<0.05, **P<0.01 (Kruskal–Wallis test). (E) Fxn mRNA levels, normalized to β-actin, in scrambled- and sh380-transduced astrocytes, determined by RT-qPCR compared to wt. Data are expressed as mean±s.e.m.; n=3 *P<0.05 (Brown–Forsythe and Welch one-way ANOVA tests with multiple comparisons).

Fig. 2.

Increased ROS levels and impaired antioxidant defences in FRDA-like astrocytes. (A) Quantification of ROS production, analysed at the single-cell level by high-throughput microscopy, and evaluated by CellROX fluorescence in sh380- compared to scrambled-transduced cells. Each dot represents the average of an individual culture well, from three biological replicates. Data are expressed as mean±s.e.m. *P<0.05 (Mann–Whitney test). (B) Measurement of GSH content, estimated in single cells by mBCl fluorescence, in scrambled and sh380-transduced astrocytes. Each dot represents the fluorescence levels recorded at the plateau phase (20 min after mBCl administration) in each individual cell. Data from three biological replicates, are expressed as mean±s.e.m.; n=90 cells for scrambled, 179 for sh380; ***P<0.001. (C) Fura-2 fluorescence kinetics, analysed at the Ca²⁺-insensitive excitation wavelength of 356 nm, upon administration of Fe²⁺ (100 μM) in the presence of pyrithione, an iron ionophore (20 μM), in scrambled- and sh380-transduced astrocytes (blue and red, respectively). Fura-2 basal fluorescence (a) was quenched by Fe²⁺ entry induced by Fe²⁺/Pyr administration (b); the oxidation of Fe²⁺ to Fe³⁺ eventually caused fluorescence recovery (c). Data from three biological replicates are expressed as mean±s.e.m.; n=74 cells for scrambled, 81 for sh380. (D) Percentage of fura-2 fluorescence recovery in scrambled- and sh380-transduced astrocytes. Data correspond to the ratio of fluorescence recovery to fluorescence quenching [(c−b)/(b−a)]. Each dot represents the fluorescence of a single cell. Data from three biological replicates are expressed as mean±s.e.m.; n=74 cells for scrambled, and 81 for sh380; ***P<0.001. Statistical analysis in B and D was performed with a linear mixed effect model with nested random effects on optical microscopy fields within the experiment. A.U., arbitrary units.

Fig. 3.

Mitochondrial membrane potential is not affected in FRDA-like astrocytes. (A) Representative images from three biological replicates (n=229 high-throughput microscopy fields for scrambled and 222 for sh380) of mitochondria loaded with TMRM, from scrambled- and sh380-transduced astrocytes. Scale bars: 50 μm. (B,C) Scrambled- and sh380-transduced astrocytes loaded with TMRM were analysed by high-throughput microscopy. After the first run of acquisition (B), astrocytes were treated with FCCP (4 μM) for 10 min and the fluorescence acquisition was repeated on the same cells (C). In C, the fluorescence decrease caused by FCCP treatment was normalized to the untreated condition. Each dot represents the fluorescence of a single cell. Data from three biological replicates are expressed as mean±s.e.m.; n=229 high-throughput microscopy (HTM) fields for scrambled, and 222 for sh380; ***P<0.001 (linear mixed effect model with nested random effects on optical microscopy fields within the experiment).

Fig. 4.

Decreased Ca²⁺ responses in FRDA-like astrocytes. (A) Peaks of Ca²⁺ response following acute ATP administration (100 µM) in scrambled- and sh380-transduced astrocytes. Each dot represents the maximum peak scored in each single cell. Data from five biological replicates are expressed as mean±s.e.m.; n=237 cells for scrambled, and 292 for sh380; ****P<0.0001. (B,C) [Ca²⁺]_i increase caused by Ca²⁺ store depletion following thapsigargin (TG) treatment. In B, representative kinetics of [Ca²⁺]_i elevation upon TG administration (1 µM). In C, TG-mediated depletion of Ca²⁺ stores. Each dot in C represents a single cell. Data, from three biological replicates are expressed as mean±s.e.m.; n=101 cells for scrambled and 87 for sh380; ****P<0.0001. (D) Peaks of Ca²⁺ response following ATP administration (100 µM) in cerebellar astrocytes treated for 24 h with Fe³⁺ (administered as ferric ammonium citrate, FAC, 50 μM). Each dot represents the maximum fluorescence peak scored in each individual cell. Data from five biological replicates are expressed as mean±s.e.m.; n=313 cells for scrambled, 237 for sh380; ****P<0.0001. Statistical analysis in A, C and D was performed with a linear mixed effect model with nested random effects on optical microscopy fields within the experiment.