Basal activation of astrocytic Nrf2 in neuronal culture media: Challenges and implications for neuron-astrocyte modelling

Abstract

As a gatekeeper of antioxidant and anti-inflammatory cell protection, the transcription factor Nrf2 is a promising therapeutic target for several neurodegenerative diseases, leading to the development of Nrf2 activators targeting Keap1-dependent and independent regulatory mechanisms. Astrocytes play a crucial role in regulating neuronal physiology in health and disease, including Nrf2 neuroprotective responses. As neurons require specific conditions for their differentiation and maintenance, most 2D and 3D co-culture systems use medias that are compatible with neuronal differentiation and function, but also ensure astrocyte survival. Few studies, however, assess the molecular adaptations of astrocytes to changes from astrocyte maintenance medias alone, and their subsequent effects on neurons which may represent technical rather than physiological responses. Our findings show that while Nrf2 can be effectively activated by the Keap1-Nrf2 protein-protein interaction disruptor 18e, and classical Nrf2 activators dimethylfumarate and CDDO-Me, in human primary cortical astrocyte monocultures, their efficacy is lost in LUHMES neuron-astrocyte co-cultures. Further investigation revealed that the Advanced DMEM/F12-based LUHMES differentiation media maximally induced basal Nrf2 activity in astrocytes alone, compared to astrocyte maintenance media, thus preventing pharmacological activation. Although Neurobasal slightly activated basal Nrf2, this was not significant and did not prevent further activation by dimethylfumarate, suggesting that this media has less impact on astrocytic Nrf2 activity relative to Advanced DMEM/F12. As Nrf2 is a key regulator of oxidative damage and neuroinflammation, modelling these common features of neurodegenerative diseases may be confounded by environments that maximally activate basal Nrf2. Our findings thus suggest caution in media selection for neuron-astrocyte co-culture in disease modelling and therapeutic Nrf2 activator discovery, and suggest use of Neurobasal over Advanced DMEM/F12 medias for this purpose.

Keywords: Nrf2; Nuclear factor erythroid 2-related factor 2; astrocytes; neuronal modelling.

Figure 1.

Confirming efficacy of Nrf2 activators in human astrocyte monocultures. (a) Activity of NQO1, as measured by simultaneous reduction of its’ substrate menadione and water-soluble tetrazolium salt-1, was used as a read-out of Nrf2 activity. Optimum total protein concentration to measure NQO1 activity in the linear phase was determined using samples from untreated human astrocyte cultures. Baseline absorbance, independent from NQO1 activity, was measured in the presence of the NQO1 inhibitor dicoumarol, and OD values subtracted from all data-points presented (n = 11, three independent experiments). (b) Human astrocytes were treated with pharmacological Nrf2 activators 18e (10 µM), DMF (40 µM) and CDDO-Me (10 nM) for 30-32 h, in ScienCell complete human astrocyte media, as well as 0.1% DMSO vehicle control (n = 8-10, three independent experiments), and Nrf2 activation determined via NQO1 assay using 2 µg total protein per sample. All data are presented as means ± SEM. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. **p < 0.01 and ***p < 0.001. Values derived from each independent vial of cells are denoted by colour and shape.

Figure 2.

Measuring efficacy of pharmacological Nrf2 activators in LUHMES neuron-astrocyte co-cultures. (a) LUHMES neurons were seeded onto a layer of primary human astrocytes at day two of the co-culture protocol, then fixed and immunostained at day six post-differentiation for β-III tubulin (green) and the astrocyte-specific marker ALDH1 L1 (red). Nuclei were labelled by DNA staining with DAPI (blue). (I) Neurons alone (II) Human astrocyte culturesalone. (II) Nuclei alone. (IV) Merged image of A-C. (V) No antibody, negative control. (VI) Cells immunostained with goat anti-mouse 488 and goat anti-rabbit 594 secondary antibodies alone. Images were captured using a ZEISS LSM 800 confocal microscope with 63x oil lens. Scale bars, 10 µm (b) Effects of Nrf2 activators on NQO1 activity in LUHMES neuron-astrocyte co-cultures maintained in LUHMES differentiation media, following treatment with 18e (10 µM), DMF (40 µM), CDDO-Me (10 nM), and 0.1% DMSO vehicle control, for 25-32 h. (n = 6, three independent experiments). Values derived from each independent vial of cells are denoted by colour and shape.

Figure 3.

Measuring effects of Advanced DMEM/F12-based LUHMES differentiation media on Nrf2 activity in human astrocyte cultures. (a) Kinetic analysis of NQO1 activity. Human astrocyte cultures were treated with LUHMES differentiation media (Advanced DMEM/F12), either alone (untreated control), or in the presence of non-targeting (NT) siRNA or Nfe2I2 siRNAs, and compared to cells maintained in complete astrocyte media. Data are presented as the mean absorbance at 440 nm from three independent experiments, each performed in duplicate. (b) NQO1 activity in human astrocytes following 5-min reactions. Data are presented as mean absorbance at 440 nm/µg total protein ± SEM (n = 6, three independent experiments). (c) Nfe2l2 mRNA expression in response to siRNA SMARTpools. Human astrocyte cultures were treated with non-targeting siRNA or Nfe2I2 siRNAs and compared to untreated controls, all prepared in complete astrocyte media. Nfe2l2 gene expression was normalised to the geometric mean of GAPDH and RPLP0 housekeepers, and data are presented as mean fold change ± SEM. (n = 3, performed in triplicate, three independent experiments). Analysis performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Values derived from each independent vial of cells are denoted by colour and shape.

Figure 4.

Investigating the effects of Advanced DMEM/F12 LUHMES differentiation media components, glucose and tetracycline, on Nrf2 activity in astrocytes. (a) Kinetic analysis of NQO1 activity. Human astrocyte cultures were treated with Advanced DMEM/F12 LUHMES differentiation media, with or without tetracycline (10 µg/mL) or astrocyte media containing 3.15 g/L glucose for 32 h and compared to untreated control cells maintained in complete astrocyte media at 1 g/L glucose. Data are presented as the mean absorbance at 440 nm of three independent experiments, each performed in triplicate. (b & c) End-point analysis of NQO1 activity, after 5-min reaction following tetracycline and glucose treatments (B) or addition of neuronal media supplements, N2, B27 or removal of astrocyte media supplements AGS and FBS (C). Data are presented as mean absorbance at 440 nm/µg total protein ± SEM (n = 6-9, three independent experiments). Analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *p < 0.05. Values derived from each independent vial of cells are denoted by colour and shape.

Figure 5.

Investigating the effects of Neurobasal medium on Nrf2 activity in human astrocytes. NQO1 activity in human astrocyte cultures treated with neurobasal media for 32 h via (a) Kinetic measurement and (b) at end-point after 5-min reactions. (c) End-point measurement of NQO1 activity in astrocytes following treatment with 40 µM DMF in either complete astrocyte or Neurobasal medias for 28 h. Data are presented as mean absorbance/ug total protein ± SEM (n = 5–9, three independent experiments). Analysis in (B) was performed using an unpaired two-tailed t-test and in C using One-way ANOVA followed by Tukey’s post hoc analyses *p < 0.05. Values derived from each independent vial of cells are denoted by colour and shape.