1α,25-Dihydroxyvitamin D₃ enhances cerebral clearance of human amyloid-β peptide(1-40) from mouse brain across the blood-brain barrier

Male C57BL/6 mice (8-10 weeks old) were purchased from Japan SLC (Hamamatsu, Japan). All experiments were approved by the Animal Care Committee of the Graduate School of Pharmaceutical Sciences, Tohoku University.

Monoiodinated, non-oxidized and lyophilized [¹²⁵I]hAβ(1-40) labeled from hAβ(1-40) (Biosource, CA, USA) was from PerkinElmer Life Sciences (Boston, MA, USA) (2,200 Ci/mmol). [³H]Dextran (100 mCi/mg) was obtained from American Radiolabeled Chemicals, Inc. (St. Louis, MO, USA). 1,25(OH)₂D3 was obtained from Biomol Research Laboratories (Plymouth Meeting, PA, USA). Forskolin and xylazine hydrochloride were purchased from Sigma (St Louis, MO, USA). U0126 was obtained from Calbiochem (Merck, Darmstadt, Germany). Ketaral 50 (ketamine hydrochloride) was purchased from Sankyo Co. (Tokyo, Japan). All other chemicals were analytical grade commercial products.

The mice were injected i.p. with 100 μL of 1,25(OH)₂D3 at a dose of 0.2, 1 or 5 μg dissolved in H₂O containing 5% ethanol, or forskolin at a dose of 2.3 or 23 μg dissolved in 100 μL of H₂O containing 25% dimethyl sulfoxide. Control mice were injected i.p. with 100 μL of each vehicle solution. After an indicated time (24, 48, or 72 h), the in vivo brain elimination experiments were performed using the intracerebral microinjection technique [25]. Briefly, mice were anesthetized with an intramuscular injection of xylazine (1.22 mg/kg) and ketamine (125 mg/kg), and placed in a stereotaxic frame (SRS-6; Narishige, Tokyo, Japan) that determines the coordinates of the mouse brain coinciding with the secondary somatosensory cortex 2 (S2) region. A small hole was made 3.8 mm lateral to the bregma, and a fine injection needle fitted to a 5.0 μl microsyringe (Hamilton, Reno, NE, USA) was advanced to a depth of 2.5 mm. The solution of [¹²⁵I]hAβ(1-40) was freshly prepared from lyophilized [¹²⁵I]hAβ(1-40) immediately before each experiment. Lyophilized [¹²⁵I]hAβ(1-40) was solubilized in H₂O and then added to an equal volume of 2-fold concentration of extracellular fluid (ECF) buffer (244 mM NaCl, 50 mM NaHCO₃, 6 mM KCl, 2.8 mM CaCl₂, 2.4 mM MgSO₄, 0.8 mM K₂HPO₄, 20 mM D-glucose, and 20 mM HEPES, pH 7.4). Gel filtration analysis confirmed that a single peak of monomer [¹²⁵I]hAβ(1-40) was detected, and 98.3% of the radioactivity was recovered, suggesting that release of free [¹²⁵I] was negligible [26]. The applied solution (0.3 μl) containing [¹²⁵I]hAβ(1-40) and [³H]dextran in an ECF buffer was administered into the S2 region over a period of 30 sec. After microinjection, the microsyringe was left in place for 4 min to minimize any backflow. At 5, 30 or 60 min after microinjection, ipsilateral (left) and contralateral (right) cerebrum and cerebellum were excised and dissolved in 2.0 ml 2 M NaOH at 60°C for 1 h. The ¹²⁵I and ³H radioactivities of the samples were measured in a γ-counter (ART300, Aloka, Tokyo, Japan) for 3 min and a liquid scintillation counter (TRI-CARB2050CA, Packard Instruments, Meriden, CT, USA) for 5 min, respectively. The BEI was defined by Equation (1), and the percentage of substrate remaining in the ipsilateral cerebrum (100-BEI) was determined using Equation (2):

The apparent elimination rate constant (k_el) was determined from the slope given by fitting a semilogarithmic plot of (100-BEI) versus time, using the nonlinear least-squares regression analysis program MULTI [27].

Endogenous soluble Aβ(1-40) in mouse brain extracts was quantified according to the previously-reported method [28]. Twenty-four h after 1,25(OH)₂D3 (1 μg/mouse) or vehicle injection i.p., mice were sacrificed and the brains were rapidly isolated. The cerebellum and olfactory bulbs were discarded, and the remaining brains were snap-frozen in liquid nitrogen. To extract soluble endogenous full-length mouse Aβ(1-40), the frozen brains were homogenized in 10 volumes of ice-cold Tris buffer (50 mM Tris-HCl, 250 mM sucrose and protease inhibitor cocktail (Sigma, St. Louis, MO, USA)). Samples were mixed in diethylamine (DEA) to yield 0.4% concentration and centrifuged at 100,000 × g for 45 min at 4°C. The resultant supernatant was neutralized in 1/10 volume of 0.5 M Tris-HCl (pH 6.8) and used for analysis. Endogenous full-length mouse Aβ(1-40) levels were determined using an ELISA (IBL, Gunma, Japan) according to the manufacturer's protocol.

The TM-BBB4 cell line, established from transgenic mice harboring the temperature-sensitive SV40 large T-antigen gene [24], was used in this study. TM-BBB4 cells were cultured at 33°C in Dulbecco's modified Eagle's medium (DMEM; Nissui Pharmaceutical Co., Tokyo, Japan), supplemented with 20 mM NaHCO₃, 2 mM L-glutamine, 15 ng/ml endothelial cell growth factor, 100 U/ml benzyl penicillin, 100 mg/ml streptomycin sulfate, and 10% fetal bovine serum (Moregate, Bulimba, Australia) in an atmosphere of 95% air and 5% CO₂.

Isolated mouse brain capillaries were prepared as described previously [29]. Total RNA was prepared from isolated mouse brain capillaries, TM-BBB4 cells and mouse brain using TRIzol reagent (Life Technologies, Grand Island, NY, USA) and RNeasy plus kit (Qiagen, Tokyo, Japan) according to the manufacturer's protocols. Single-stranded cDNA was prepared from 1 μg of total RNA by means of reverse transcription (DNA Ligation Kit, Takara, Osaka, Japan) using oligo dT primer. The PCR was performed with GeneAmp (PCR system 9700, Perkin-Elmer, Norwalk, CT, USA) using specific primers through 1 cycle of 94°C for 2 min, and 35 cycles of 94 °C for 30 s, 55°C for 30 s and 72°C for 1 min and finally 72°C for a further 5 min. The sequences of the primers were as follows: sense primer 5'-CGA TGC CCA CCA CAA GAC CTAC-3' and antisense primer 5'-CAG CAT GGA GAG CGG AGA CAG-3' for vitamin D receptor (GenBank Accession Number, NM_009504), sense primer 5'-TTT GAG ACC TTC AAC ACCC C-3' and antisense primer 5'-ATA GCT CTT CTC CAG GGA GG-3' for β-actin (GenBank Accession Number, NM_031144). The RT-PCR of each sample RNA in the absence of reverse transcriptase was used as a negative control. The RT-PCR products were separated by electrophoresis on an agarose gel in the presence of ethidium bromide (0.6 μg/mL) and visualized using an imager (EPIPRO 7000; Aisin, Aichi, Japan).

TM-BBB4 cells were seeded on 24-well plates (Becton Dickinson, Bedford, MA, USA) at a density of 1.0 × 10⁵ cells/2 cm²/well and cultured for 24 h. The cells were treated with the indicated concentrations of forskolin and U0126 or vehicle for 24 h, and then internalization of [¹²⁵I]hAβ(1-40) into TM-BBB4 cells was examined as previously described [26]. Briefly, the extent of TM-BBB4 cell-[¹²⁵I]hAβ(1-40) (0.15 μCi/200 μl in ECF buffer) binding was measured for 3 min at 37°C. After the indicated times, binding was terminated by removing the solution, and the cells were washed three times with 1 ml ice-cold ECF buffer. The cells were then incubated with 1 ml ice-cold acetate-barbital buffer (28 mM CH₃COONa, 120 mM NaCl, 20 mM barbital sodium (pH 3.0) and 360 mOsm/kg) for 20 min at 4°C to remove [¹²⁵I]hAβ(1-40) bound to the cell surface. After incubation, the buffer was then recovered (acid-soluble fraction) and the cells were subsequently washed three times with ice-cold acetate-barbital buffer.

Acid-resistant binding represents the amount of internalized [¹²⁵I]hAβ(1-40) in the TM-BBB4 cells. The cells were solubilized with 200 μl of 5 M NaOH overnight and neutralized with 200 μl of 5 M HCl. The ¹²⁵I radioactivity was measured using a γ-counter (ART310, Aloka, Tokyo, Japan). The protein content of the cultured cells was measured using a DC protein assay kit (Bio-Rad, Hercules, CA, USA) with bovine serum albumin as a standard.

To examine the effect of 1,25(OH)₂D3 treatment on brain-to-blood [¹²⁵I]hAβ(1-40) efflux transport across the BBB, the time profile of the percentage of [¹²⁵I]hAβ(1-40) remaining in the ipsilateral cerebrum was examined at 24 h after i.p. administration of a single dose of 1,25(OH)₂D3 (1 μg/mouse) (Figure 1). [¹²⁵I]hAβ(1-40) was eliminated from the brain in a time-dependent manner after the administration of either 1,25(OH)₂D3 or vehicle (Figure 1A). The percentage of [¹²⁵I]hAβ(1-40) remaining in 1,25(OH)₂D3-treated mouse brain was significantly lower than that in the vehicle-treatment group at 60 min. The apparent elimination rate constant across the BBB of [¹²⁵]hAβ(1-40) determined from the slope of the (100-BEI (%)) time profile was 2.38 × 10^-2 ± 0.09 × 10^-2 min^-1 (mean ± SD) in the 1,25(OH)₂D3-treated group, which was significantly greater (1.29-fold) than that of the vehicle-treated group (1.85 × 10^-2 ± 0.07 × 10^-2 min^-1, mean ± SD) (Figure 1B). The initial percentage of [¹²⁵I]hAβ(1-40) remaining in the mouse brain treated with 1,25(OH)₂D3 was 118% (Figure 1A), which is not significantly different from that of the vehicle-treated group (114%), indicating that the distribution volume of [¹²⁵I]hAβ(1-40) in the brain was not affected by 1,25(OH)₂D3 treatment. This suggests that 1,25(OH)₂D3-treatment did not affect hAβ(1-40) binding to brain parenchymal cells in mouse brain.

Figure 2A shows the dose-dependence of 1,25(OH)₂D3's effect on the percentage of [¹²⁵I]hAβ(1-40) remaining in the ipsilateral cerebrum at 60 min after microinjection; this time point was chosen because the percentage of [¹²⁵I]hAβ(1-40) remaining decreased time-dependently up to 60 min (Figure 1A, open circles). Twenty-four h after single i.p. administration of 1,25(OH)₂D3 (0.2, 1 and 5 μg/mouse), the percentage of [¹²⁵]hAβ(1-40) remaining was decreased by 12.9%, 25.4% and 19.5%, respectively, compared with the vehicle-treated control (Figure 2A).

The effect of time after 1,25(OH)₂D3 administration was examined at the amount of 1 μg/mouse, which exhibited the greatest effect in Figure 2A. As shown in Figure 2B, a significant reduction of the percentage of [¹²⁵]hAβ(1-40) remaining was observed at 24 h after i.p. administration of a single dose of 1,25(OH)₂D3, compared with the vehicle-treated control, while no significant reduction was detected at 48 or 72 h after the i.p. administration. No significant decrease was observed in body weight or residual amount of [³H]dextran, which is an impermeable marker, in mouse brain up to 72 h after the administration (data not shown), suggesting that a single dose of 1,25(OH)₂D3 did not cause significant toxicity or disruption of BBB integrity.

1,25(OH)₂D3 exerts its biological effects by transcriptional activation of target genes via binding to the VDR. RT-PCR analysis was performed to determine VDR expression in mouse brain, mouse isolated brain capillaries and TM-BBB4 cells [24]. A band of the size predicted for VDR was obtained in each case (Figure 4), suggesting that VDR is expressed in mouse brain capillary endothelial cells.

1,25(OH)₂D3 also exerts non-genomic biological effects via a complex signaling process, which has been reported to involve increased levels of cyclic AMP (cAMP) [30] and activation of the mitogen-activated protein kinase kinase (MEK) pathway [31, 32]. Since forskolin increases the level of cAMP following activation of adenylyl cyclase in vitro, the effect of forskolin on the remaining percentage of [¹²⁵I]hAβ(1-40) in the ipsilateral cerebrum was examined (Figure 5). Twenty-four h after i.p. administration of single-dose forskolin (2.3 or 23 μg/mouse), the percentage of [¹²⁵I]hAβ(1-40) remaining in mouse brain was decreased by 16.4% in either case compared with the vehicle-treated group at 60 min after intracerebral [¹²⁵I]hAβ(1-40) microinjection.

To determine whether MEK is involved in the forskolin-induced enhancement of [¹²⁵I]hAβ(1-40) elimination from the brain, the effect of U0126, an inhibitor of MEK activation, on [¹²⁵I]hAβ(1-40) internalization into TM-BBB4 cells was assessed. In the present study, [¹²⁵I]hAβ(1-40) bound to the cell surface was washed off with acetate-barbital buffer (pH 3.0) after the uptake, and the acid-resistant binding represents the amount of internalized [¹²⁵I]hAβ(1-40) in TM-BBB4 cells. Pre-treatment with forskolin (10 μM) significantly increased the acid-resistant binding of [¹²⁵I]hAβ(1-40) by TM-BBB4 cells (Figure 6A), indicating that [¹²⁵I]hAβ(1-40) internalization into TM-BBB4 cells was enhanced by forskolin treatment (Figure 6A). As shown in Figure 6B, U0126 (25 μM) suppressed the forskolin-induced enhancement of [¹²⁵I]hAβ(1-40) acid-resistant binding. Pre-treatment with U0126 alone significantly reduced the acid-resistant binding of [¹²⁵I]hAβ(1-40) by 11.2% compared with the control. This reduction could be due to inhibition of basal MEK activation and perhaps in part due to a toxic effect of U0126, since the protein content of the cells after the uptake was reduced by 19.8% by U0126 treatment (89.8 ± 6.0 μg protein/well and 72.0 ± 7.0 μg protein/well, respectively, mean ± SEM, n = 4). The acid-resistant binding of [¹²⁵I]hAβ(1-40) was not significantly different between the group pre-treated with U0126 alone and that treated with forskolin and U0126. The protein content of the cells was not also affected significantly (72.0 ± 7.0 μg protein/well for U0126 only and 66.4 ± 3.9 μg protein/well for forskolin and U0126, mean ± SEM, n = 4). This result suggests that MEK activation is involved in the enhancement of Aβ(1-40) internalization by forskolin in TM-BBB4 cells.