A circadian clock regulates efflux by the blood-brain barrier in mice and human cells

Abstract

The blood-brain barrier (BBB) is critical for neural function. We report here circadian regulation of the BBB in mammals. Efflux of xenobiotics by the BBB oscillates in mice, with highest levels during the active phase and lowest during the resting phase. This oscillation is abrogated in circadian clock mutants. To elucidate mechanisms of circadian regulation, we profiled the transcriptome of brain endothelial cells; interestingly, we detected limited circadian regulation of transcription, with no evident oscillations in efflux transporters. We recapitulated the cycling of xenobiotic efflux using a human microvascular endothelial cell line to find that the molecular clock drives cycling of intracellular magnesium through transcriptional regulation of TRPM7, which appears to contribute to the rhythm in efflux. Our findings suggest that considering circadian regulation may be important when therapeutically targeting efflux transporter substrates to the CNS.

Fig. 1. Efflux of ABCB1 substrate from the brain oscillates over the course of the day.

a Schematic of experiment. The ABCB1 substrate RHB was intravenously injected via jugular vein into mice. Mice were allowed 90 min to recover and brains and sera were collected. Fluorescence was read at ex540/em590nm using a plate reader. b Brain RHB levels are regulated by ABCB1-mediated efflux. RHB or RHB and the ABCB1-inhibitor tariquidar was intravenously injected into WT mice. Individual mice are shown with triangle markers and means ± SEM are shown (n = 6; 2 independent experiments). c RHB injection was performed on WT mice at indicated time points. Individual points are shown with triangle markers and lines represent the mean. pCycle value was calculated by JTKCycle analysis (n = 19; 3 independent experiments). d Absence of vascular leakage in both day and night. Evans Blue was intravenously injected into mice untreated or treated with LPS 24 h prior. Animals were sacrificed and brains and liver were collected after 30 min. Evans Blue was extracted from tissue and amount was measured at absorbance 620 nm. Individual data points and means are shown. n = 8 control, 6 LPS-treated; 2 independent experiments. p-value was determined by Student’s T-test.

Fig. 2. A circadian clock in the BBB regulates brain permeability of RHB.

RHB was intravenously injected via jugular vein into mice. Mice were allowed 90 min to recover and brains and sera were collected. Fluorescence was read at ex540/em590nm using a plate reader. a Clock ablation in endothelial cells disrupts the brain permeability rhythm to RHB. RHB was injected into control (n = 18; 4 independent experiments) or endothelial-specific Bmal1-deficient (n = 31; 4 independent experiments) mice at indicated time points. b Global circadian disruption ablates brain permeability rhythm to RHB in mice. RHB was injected into Cry1/2 het (n = 24; 3 independent experiments) and Cry1/2 DKO (n = 23; 4 independent experiments) mice at indicated time points. Individual mice are shown with triangle markers, and lines represent the mean. pCycle values were calculated by JTKCycle analysis.

Fig. 3. Cycling of circadian clock genes and a highly expressed magnesium transporter in brain endothelial cells.

Control or endothelial-cell specific Bmal1-deficient mice were collected at ZT2, ZT6, ZT10, ZT14, ZT18, and ZT22 (n = 12; 6 time points, 2 independent experiments); brains were dissected and endothelial cells were isolated by FACS using CD31 antibody. RNA was extracted and sequenced with HiSeq. a Few transcripts cycle in BBB endothelial cells. Heatmap of cycling transcripts of control animals (Meta2d, q < 0.6) is organized by peak and trough with corresponding transcripts of Bmal1-deficient endothelial cells. b Expression of core circadian genes shows robust rhythms in mouse BBB. Transcripts of controls (black) are rhythmic (pCycle <0.001) while Bmal1-deficient endothelial cells (red) are not. pCycle values were calculated by Meta2d analysis adjusted for multiple comparisons. c Trpm7 is highly expressed in BBB endothelium. Expression of known magnesium transporters in controls (n = 12; 2 independent experiments) are shown as means ± SEM. d Trpm7 shows rhythmic expression in brain endothelial cells. qPCR of sorted brain endothelial cells (n = 24; 6 time points, 4 independent experiments) were probed for Nr1d1 and Trpm7 expression across the circadian day normalized to GAPDH. pCycle values were calculated by JTKCycle analysis.

Fig. 4. Cyclic efflux in human BECs is likely driven by intracellular magnesium oscillations.

a Human BEC contain a circadian clock. A stable line was established from hCMEC/D3 cells containing Per2-dLuc. Cells were synchronized with a 30 min pulse of dexamethasone and bioluminescent counts were measured over 5 days with a luminometer (LumiCycle 32). Data were analyzed with LumiCycle software and representative detrended plot is shown. b Verapamil inhibits RH123 efflux. Suspended cells were incubated with RH123 with or without the ABCB1-inhibitor verapamil on ice for 15 min. Excess RH123 was removed and half of the culture was incubated at 37 °C for 30 min to allow for optimal efflux conditions while the rest remained on ice. The amount of intracellular RH123 was determined by flow cytometry. Representative histograms of each treatment condition are shown. c Efflux of RH123 from BBB cells is rhythmic. Cells were synchronized with dexamethasone and efflux assay was performed at the indicated time point. The percentage of RH123 fluorescence effluxed in 30 min comparing the level of fluorescence in cultures with or without 37 °C incubation is shown as means ± SEM. Cells were incubated with vehicle (n = 81; 9 time points, 9 experiments) or the ABCB1-inhibitor verapamil (n = 45; 9 time points, 5 experiments). pCycle values were calculated by JTKCycle analysis. d Chelating magnesium reduces ratio of bound to free MagFura2. hCMEC/D3 cells were incubated with MagFura2-AM and indicated doses of EDTA-AM. MagFura was measured at ex330/em490nm (free) ex369/em490nm (bound). Means of normalized fluorescence of bound Magfura2 (F_bound) over the fluorescence of free Magfura2 (F_free) ± SEM are shown. n = 4, representative of 2 independent experiments. One-way repeat measures mixed effect analysis with Dunnett’s multiple comparisons test was used to compare each experimental group to control. e Intracellular magnesium levels oscillate in phase with efflux cycles. hCMEC/D3 cells were incubated with intracellular magnesium indicator Magfura2-AM at the indicated time points after dexamethasone synchronization and measured at ex330/em490 (free) ex369/em490 (bound) using a plate reader. Means of normalized fluorescence of bound Magfura2 (F_bound) over the fluorescence of free Magfura2 (F_free) ± SEM are shown (n = 50; 10 time points; 5 independent experiments.). pCycle values were calculated by JTKCycle analysis. f Reducing intracellular magnesium inhibits efflux. Verapamil and/or EDTA was added to hCMEC/D3 cells and RH123 was measured by flow cytometry. The percent of RH123 of efflux was normalized to control (n = 5–8 from 4 independent experiments). One-way ANOVA was used to compare all experimental groups to control with Dunnett’s multiple comparisons test.

Fig. 5. TRPM7 is a clock-controlled gene in human brain endothelial cell culture.

a, b BMAL1 and TRPM7 cycle in phase following dexamethasone synchronization. hCMEC/D3 mRNA was extracted at indicated time points post-synchronization. Real-time PCR analysis was performed for BMAL1 and TRPM7. Data are shown as means ± SEM (n = 40; 10 time points from 4 independent experiments.) pCycle values were calculated by JTKCycle analysis. c BMAL1 binds TRPM7 e-box site. TRPM7 gDNA from hCMEC/D3 cells were immunoprecipitated with either IgG or BMAL1 antibody. Primers for PER2, TRPM7 e-boxes were assessed by real-time PCR. Data are shown as mean fold change of sites binding to BMAL1 normalized to IgG (n = 6 plates from 3 independent experiments). d mRNA from hCMEC/D3 cultures was extracted at indicated time points post-synchronization. Real-time PCR analysis was performed for ABCB1. Data are shown as means ± SEM (n = 40 plates; 10 time points; 4 independent experiments). pCycle values were calculated by JTKCycle analysis. e, f Protein levels of TRPM7, but not ABCB1 oscillate in hCMEC/D3 cultures. Cell lines were synchronized with a pulse of dexamethasone. Cell lysates were collected between 12 and 48 h later. Lysates were blotted with antibodies against TRPM7 or ABCB1 and ACTIN. Representative immunoblot of TRPM7 (n = 4; 4 independent experiments) or ABCB1 (n = 3; 3 independent experiments) and ACTIN and means ± SEM of quantifications are shown. Quantification of blots was performed in ImageJ. P-values were determined using one-way ANOVA to compare among time points.

Fig. 6. Knockdown of TRPM7 reduces and dampens rhythms of intracellular Mg²⁺ and xenobiotic efflux.

Stable hCMEC/D3 cell lines containing scrambled siRNA and GFP or TRPM7 siRNA and GFP were made with Lentiviral transduction. Stable lines were synchronized with dexamethasone and assayed at the indicated time points. a Knockdown of TRPM7 results in reduced Mg²⁺ and loss of cycling. Generated cell lines were incubated with intracellular magnesium indicator Magfura2-AM at the indicated time points after dexamethasone synchronization and measured at ex330/em490 (free) ex369/em490 (bound) using a plate reader. Means of normalized fluorescence of bound Magfura2 (F_bound) over the fluorescence of free Magfura2 (F_free) ± SEM are shown (n = 50; 10 time points, 5 independent experiments). p < 0.0001 comparing control to TRPM7 siRNA cell line by paired T-test, showing less intracellular Mg²⁺ in the siTRPM7 cell line across time points. pCycle values were calculated by JTKCycle analysis. b Knockdown of TRPM7 reduces xenobiotic efflux and dampens rhythms. Cells were synchronized with dexamethasone and RHB efflux assay was performed at the indicated time point. Control or TRPM7 siRNA-treated suspended cells were incubated with RHB on ice for 15 min. Excess RHB was removed and half of the culture was incubated at 37 °C for 30 min to allow for optimal efflux conditions while the rest remained on ice. The amount of intracellular RHB was determined by flow cytometry. The percentage of RH123 fluorescence effluxed in 30 min comparing the level of fluorescence in cultures with or without 37 °C incubation is shown as means ± SEM (n = 40; 10 time points, 5 independent experiments. pCycle values were calculated by JTKCycle analysis. p = 0.0085 comparing control to TRPM7 siRNA cell line by paired T-test, showing reduced RHB efflux in the siTRPM7 cell line across time points.

Fig. 7. Model of clock regulation of xenobiotic efflux at the BBB.

Core circadian clock transcription factors BMAL1 and CLOCK activate transcription of Trpm7. Increased TRPM7 allows for higher intracellular free magnesium, which regulate the activity of xenobiotic transporters. Higher transporter activity decreases the level of xenobiotics in the brain. In the absence of BMAL1/CLOCK, TRPM7 and therefore free intracellular magnesium is low. Lower levels of magnesium are associated with decrease the activity of the xenobiotic transporters, thus xenobiotics are retained in the brain.