β-Hydroxybutyrate preferentially enhances neuron over astrocyte respiration while signaling cellular quiescence

Abstract

While ketone bodies support overall brain energy metabolism, it is increasingly clear specific brain cell types respond differently to ketone body availability. Here, we characterized how SH-SY5Y neuroblastoma cell, primary neuron, and primary astrocyte bioenergetics and nutrient sensing pathways respond to β-hydroxybutyrate (βOHB). SH-SY5Y cells and primary neurons, but not astrocytes, exposed to βOHB increased respiration and decreased PI3K-Akt-mTOR signaling. Despite increased carbon availability and respiration, SH-SY5Y cells treated with βOHB reduced their overall metabolic activity and cell cycling rate. Levels of the quiescence-regulating Yamanaka factors increased to a broader extent in SH-SY5Y cells and primary neurons. We propose a βOHB-induced increase in neuron respiration, accompanied by activation of quiescence associated pathways, could alleviate bioenergetic stress and limit cell senescence. This in turn could potentially benefit conditions, including brain aging and neurodegenerative diseases, that feature bioenergetic decline and cell senescence.

Keywords: Aging; Bioenergetics; Ketones; Mitochondria; Quiescence; Senescence.

Figure 1.. Chronic exposure to βOHB increases respiration and shifts ADP/ATP and NAD⁺/NADH ratios in SH-SY5Y cells.

A) Shifts in OCR and ECAR become more pronounced with longer exposure to βOHB in vitro. B) Both acute and chronic exposure to βOHB increased basal OCR. C) Chronic exposure to βOHB reduced basal ECAR. D) OCR associated with ATP production was increased both acutely and chronically with βOHB. E) Coupling efficiency was reduced by βOHB treatment. F) ADP/ATP ratio decreased with chronic exposure to βOHB. G) NAD⁺/NADH ratio increased with chronic βOHB treatment (≥ 7 days in vitro). n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 of post-hoc Tukey’s analysis between groups respectively. Black bars represent mean ±SEM.

Figure 2.. βOHB reduces mitochondrial membrane potential in SH-SY5Y cells.

A) Representative images of TMRE labeling of SH-SY5Y cells under different conditions with Hoescht nuclear stain. B) βOHB reduced mitochondrial membrane potential with chronic treatment as measured by TMRE intensity. C) Acute and chronic βOHB exposure reduced JC1 measured mitochondrial membrane potential. D) Incorporation of NAO into mitochondria is reduced at 24 hours and chronically upon exposure to βOHB. n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 of post-hoc Tukey’s analysis between groups respectively. Black bars represent mean ±SEM.

Figure 3.. Chronic treatment with βOHB increases autophagic capacity without altering flux and leads to increases in mitochondrial mass.

A) Western blot images of autophagy flux samples from vehicle and chronic βOHB samples. B) βOHB increased the total amount of both LC3B lipidated and unlipidated forms under baseline and bafilomycin treated conditions but does not increase the delta compared to untreated samples. C) βOHB induced an increase in the mitophagy adaptor protein NDP52 that matched changes in LC3B. D) βOHB increased levels of mitophagy protein BNIP3 that was observable only with bafilomycin treatment. E) βOHB produced no significant change in the amount of mitophagy protein Nix/BNIP3L regardless of bafilomycin treatment. F) mtDNA copy number as measured by qPCR of fold change of copies of MT-ND1 in isolated gDNA relative to nuclear encoded copies of β2M. n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 of post-hoc Tukey’s analysis between groups respectively. Black bars represent mean ±SEM.

Figure 4.. βOHB reduces SH-SY5Y metabolic activity and growth rate without increasing cell death.

A) βOHB reduced cell metabolic activity as measured by MTT assay at both 24-hour and chronic timepoints. B) βOHB did not induce SH-SY5Y cell death. Camptothecin positive and negative HeLa cells were included as apoptosis positive and negative controls respectively. C) βOHB slowed cell growth rate that became readily apparent 3 days following cell seeding. Doubling time increased from 17 hours to 33 hours. D) Treatment with βOHB did not reveal a change in the distribution of cells in different phases of the cell cycle under asynchronous growth conditions. E) Treatment with βOHB along with cell synchronization by double thymidine block and release revealed an accumulation of cells in the G0–G1 phase indicating a bottleneck through the G1-S transition at the 18-hour post release timepoint. n = 12 per group; HeLa controls n = 3, 3. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 of post-hoc Tukey’s analysis between groups respectively. Black bars represent mean ±SEM.

Figure 5.. βOHB reduces PI3K-Akt-mTOR pathway activation, increases histone acetylation, and induces transcription of quiescence regulating Yamanaka factors in SH-SY5Y cells.

A) Western blot images of PI3K-Akt-mTOR pathway enzymes and their regulatory post-translational modification sites along with acetylation status of histone 3. B) Treatment with βOHB reduced Akt phosphorylation at both Ser473 and Thr308 activation sites as well as reduced phosphorylation of the mTOR downstream target p70 S6 Kinase at Thr389. Total protein levels of Akt were increased by βOHB treatment. C) Histone 3 acetylation increased at lysine 9. D) βOHB increased mRNA levels of Yamanaka factors Oct4 (POU5F1) and Sox2, but only demonstrated non-significant trends in myc. Klf4 transcripts were not detected in the SH-SY5Y cells under any treatment condition. n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 respectively. Black bars represent mean ±SEM.

Figure 6.. In SH-SY5Y cells βOHB reduced phosphorylation of the AMPK downstream target ACC1, altered lipid and cholesterol homeostasis, and increased PDHE1α phosphorylation.

A) Western blot images demonstrating changes in phosphorylation of ACC1 at Ser79. B) Chronic βOHB exposure reduced phosphorylation of ACC1 at the inhibitory Ser79 site regulated by activated AMPK. C) Fluorescent imaging of green BODIPY 493/503 labeled lipid droplets at 20x magnification with Hoescht 33342 nuclear staining. D) Quantification of lipid droplet number per cell indicate βOHB treatment increased the total number of lipid droplets. E) Chronic exposure to βOHB reduced cellular triglyceride content. F) Fluorescent images of SH-SY5Y cells labeled with the cholesterol reporter filipin III in blue and nuclear stain NucGreen at 20x magnification. G) βOHB reduced cholesterol content in SH-SY5Y cells. H) Western blot images demonstrating changes in PDHE1α subunit phosphorylation. I) βOHB increases phosphorylation of PDHE1α at the inhibitory Ser293 site. n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 respectively. Black bars represent mean ±SEM.

Figure 7.. In rat primary neurons βOHB reduced PI3K-Akt-mTOR activation, increased histone acetylation, and increased protein levels of Oct4 but did not influence primary rat astrocytes in the same manner.

A) Fluorescent imaging showing enrichment of primary rat neurons in culture, as indicated by MAP2 and GFAP staining. Hoechst 33342 was used to visualize nuclei. B) Fluorescent imaging showing enrichment of primary rat astrocytes in culture. C-D) Western blot images of protein changes in primary rat neuron and astrocyte whole cell lysates. E) In neurons βOHB reduced phosphorylation of Akt in a manner that recapitulated SH-SY5Y cells while simultaneously increasing the amount of histone acetylation and Oct4 protein. F) In astrocytes βOHB reduced S473 Akt phosphorylation while increasing T308 phosphorylation. βOHB did not increase histone acetylation as observed in SH-SY5Y and rat neurons. Vehicle treated rat astrocytes demonstrated only a single band of Oct4 protein whereas βOHB treated astrocytes showed an increased density of the original band plus a second band at a slightly higher molecular weight. n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 respectively. Black bars represent mean ±SEM.

Figure 8.. βOHB increased respiration in primary rat neurons in a manner that reflected its effects on SH-SY5Y cells.

A-C) βOHB increased respiration with chronic exposure. Unlike SH-SY5Y cells, in primary neurons chronic βOHB increased the basal ECAR. D) Chronic βOHB (≥ 7 DIV) increased respiration associated with ATP production. E-F) OCR attributed to proton leak was increased by βOHB, but any association with reduced coupling efficiency was not statistically significant. G-H) Maximum respiration was increased by βOHB, but no increase in spare capacity was observed with chronic culturing conditions. I) Non-mitochondrial oxygen consumption was increased by βOHB. n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 respectively. Black bars represent mean ±SEM.

Figure 9.. βOHB does not alter basal metabolic fluxes in primary astrocytes but does enhance spare capacity.

A-C) Chronic βOHB (≥ 7 DIV) did not alter primary astrocyte basal OCR or ECAR. D) βOHB did not alter OCR associated with ATP production in astrocytes. E-F) βOHB reduced leak rate while increasing coupling percentage in astrocytes. G-H) βOHB increased both maximum and spare respiratory capacity in astrocytes. I) βOHB did not alter the astrocyte non-mitochondrial oxygen consumption rate. n = 12 per group. *, **, ***, **** correspond to p-values < 0.05, 0.01, 0.001, and 0.0001 respectively. Black bars represent mean ±SEM.