Chemicals

(2S,3S)-Desethylreboxetine was purchased from ABX (Radeberg, Germany). Methanol was HPLC grade from Merck. Noradrenaline, 3,4-dihydroxybenzylamine, and isoproterenol were purchased from Merck, citric acid monohydrate and sodium acetate from Kanto Chemical (Tokyo, Japan), ethylenediaminetetraacetic acid from Dojindo (Kumamoto, Japan), and sodium octane sulfonate from FUJIFILM Wako Pure Chemical (Osaka, Japan). All other chemicals were of analytical grade and were used without further pretreatment. Stock solutions of noradrenaline, 3,4-dihydroxybenzylamine and isoproterenol were prepared separately at a concentration of 1 mg/mL in 0.2 M perchloric acid at -80 °C.

Animals

F344/NSlc rats (n = 15) (male, 8 weeks old, Japan SLC, Inc., Hamamatsu, Japan) were housed in groups of two and maintained on a 12-h light/dark cycle (lights on at 8 AM) with unlimited access to food and water. These rats were allowed to acclimate to their new environment for at least 1 week before being used in experiments. The animals used here were maintained and handled in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals and our institutional guidelines.

Neurotoxin treatments

Rats were treated twice with a neurotoxin one week apart to degenerate noradrenergic nerves and used for PET measurements 1–2 weeks after the last administration. First, rats were pretreated with the 5-HT reuptake blocker fluoxetine (10 mg/kg, i.p.) to protect serotonergic terminals. DSP-4 (5 mg/kg in 5 rats or 50 mg/kg in 6 rats, i.p., Sigma-Aldrich, Saint Louis, MO, USA) or saline in four rats (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) was administered 30 min after pretreatment.

Preparation of [11C]MRB

Radioactive 11C was generated by the 14N(p, α)11C nuclear reaction using a cyclotron. [11C]CH3I was prepared from [11C]CO2 according to the conventional method. The 11C methylation of (2S,3S)-Desethylreboxetine to [11C]MRB, HPLC purification and formulation were performed using an automated synthesis system (CUPID C-11-BII, Sumitomo Heavy Industries, Tokyo, Japan). Thus, obtained [11C]CH3I was trapped in 250 µL of anhydrous DMF containing 0.5 mg of (2S,3S)-Desethylreboxetine (1.75 µmol) and 8.5 µL of 5 M NaOH (42.5 µmol) at − 15 °C to − 20 °C, and then the reaction mixture was heated to 100 °C for 10 min. The radioactive mixture containing [11C]MRB was diluted with 1 mL of HPLC mobile phase and transferred to a column (10 mm I.D. × 250 mm, CAPCELL PAK C18, SHISEIDO, Tokyo, Japan) attached to an HPLC system (JACSO, Tokyo, Japan). Elution with 30:70 v/v CH3CN/0.2 M ammonium formate in sterile water at a flow rate of 5 mL/min yielded a radioactive fraction corresponding to pure [11C]MRB (retention time: 9.3 min). The fraction was collected in a rotary evaporator and evaporated to dryness at approximately 90 °C under reduced pressure. The residue was dissolved in 3 mL of sterile tween-saline and filtered through a 0.22 µm MillexⓇ-GV filter (Merck Millipore Ltd., USA). At the end of the synthesis, 0.5–1.3 GBq of [11C]MRB was obtained with a molar activity of 34–102 GBq/µmol.

Brain PET scans

All rats (n = 15) (male, 10–12 weeks old, 226 ± 26 g) underwent [11C]MRB PET brain imaging. Rats were scanned for 60 min with our small animal PET scanner (FX3200, TriFoil Imaging, CA, USA) after injection of [11C]MRB (24.3–66.8 MBq, 0.34–4.1 nmol) via the tail vein under isoflurane anesthesia (~ 2.0%). All PET images were reconstructed using the three-dimensional ordered subset expectation maximization method (4 subsets and 20 iterations; voxel size: 0.6 × 0.5 × 0.5 mm with a resolution of 0.92 mm full width at half maximum at the center of the view). The time frames were 1 min × 10 frames, 2 min × 5 frames and 4 min × 10 frames.

PET data analysis

Noradrenaline transporter densities in the hippocampus were estimated from dynamic PET data calculating the outcome measure, non-displaceable binding potential (BPND). In brief, summed PET images (Frame #1–10) were manually coregistered with a rat brain MR template, then the same rigid transformation was applied to the corresponding dynamic PET images. Then, tentative R1 (relative blood flow) parametric images were estimated using MRTM2 with the globus pallidus as the reference region. Second, the tentative R1 images were manually coregistered with a rat brain MR template, then the revised rigid transformation was applied to the corresponding dynamic PET images to extract time-activity curves in the hippocampus and globus pallidus. Then, the multilinear reference tissue model 2 [17] was applied using the globus pallidus as the reference tissue (t* = 0.5 min, k2' = 0.02 min−1) to estimate BPND.

We use the tentative R1 images because R1 reflects distribution of blood flow in the brain, allowing more accurate coregistarion by R1’s clearer outline of brain than does the summed early phase images. We chose the globus pallidus as the reference region based on the following reasons. First, the region was reported to express minimal amount of noradrenaline transporters in rat brain [18]. Second the region showed the lowest time-activity curves in our studies. Third, the region appeared not to be affected by spillover effect from the extra brain high radioactivity. The k2' value was calculated from the averaged k2' values estimated using Ichise’s multilinear reference tissue model [17] in the whole brain of four rats treated with 0 mg of DSP4. All PET data analyses were performed in PMOD 4.3 (PMOD Technologies LLC, Zurich, Switzerland).

In vitro measurement of noradrenaline transporter levels

The protein levels of the norepinephrine transporter in the hippocampus of eight rats were determined by Western blotting. These samples were homogenized in RIPA buffer (FUJIFILM Wako Pure Chemical Corporation, Japan) with Protease and phosphatase inhibitor cocktail (cOmplete™ and PhosSTOP™; Roche Diagnostics GmbH, Germany). After centrifugation at 1000 × g, the protein levels of the supernatants were measured using the protein assay kit (TaKaRa BCA Protein Assay kit; Takara Bio, Inc., Japan). Samples containing 4 μg of protein were loaded onto 10% SDS–polyacrylamide gels for electrophoresis. Protein bands in the gels were transferred to polyvinylidene difluoride membranes by electroblotting. The membranes were incubated overnight at 4 °C with respective primary antibodies. These antibodies include a polyclonal antibody against rabbit noradrenaline transporter (3 μg/mL; Merck Millipore Ltd., USA) or a polyclonal antibody against mouse β-actin (1:5000 dilution; Proteintech, USA). The membranes were then further incubated with secondary antibodies (peroxidase-conjugated goat anti-rabbit IgG, 1:3000 dilution; Proteintech, USA, and peroxidase-conjugated goat anti-mouse IgG, 1:3000 dilution; Abcam, USA). Immunoreactive bands were visualized by enhanced chemiluminescence (Thermo Scientific, USA). Bands were detected using ImageQuant LAS-4000 (GE Healthcare, USA). Bands were evaluated by densitometry using ImageJ software (National Institute of Health, Bethesda, USA). OD values of noradrenaline transporter signals were compared and normalized to the maximum intensity in the control group.

In vitro measurement of noradrenaline levels

The level of noradrenaline in the hippocampus of all the rats was measured by using an HPLC equipped with electrochemical detection (ECD). From the rat brains removed immediately after PET imaging, the hippocampi were sectioned and weighed, then frozen in liquid nitrogen and ground into powder using a grinding mill (SK-100, Tokken, Chiba, Japan). The ground samples were then homogenized in ice-cooled 0.2 M perchloric acid containing 0.2 µg/mL DHBA and IPT using an ultrasonic homogenizer (VP-050N, TAITEC, Tokyo, Japan). The suspensions were left on ice for 30 min, then centrifuged at 12,000 rpm for 15 min at 4 °C, and the supernatants were filtered through a 0.45 µm Cosmonice Filter W (Nacalai Tesque, Kyoto, Japan). Protein levels were measured using a BCA protein assay kit (Takara Bio Inc., Shiga, Japan). Monoamine levels in the samples and noradrenaline standard were measured using HPLC-ECD (ECD-700; Eicom, Kyoto, Japan) equipped with a reversed-phase column (EICOMPAK SC-50DS, 3.0 mm × 150 mm; Eicom, Kyoto, Japan) with mobile phase (3.10 g/L sodium acetate 8.84 g/L citric acid monohydrate, 220 mg/L SOS, 5 mg/L EDTA-2Na, and 20% methanol) pumped at a flow rate of 0.4 mL/min.

Hematoxylin and eosin staining and immunohistochemistry

Brain sections including the pons were subjected to hematoxylin and eosin staining and immunohistochemistry of dopamine β-hydroxylase to confirm degeneration in the locus coeruleus by DSP-4 treatment. For hematoxylin and eosin staining, frozen tissue brain sections were stained with Mayer’s hematoxylin solution (FUJIFILM Wako Pure Chemical Corporation, Japan) for 5 min at r.t. and then rinsed with water. They were stained with Eosin Y (FUJIFILM Wako Pure Chemical Corporation, Japan) for 3 min at r.t. and then rinsed with water. They were dehydrated through an increasing gradient of ethanol, 70%, 90% and 99%, for 5 min each. They were cleared in xylene three times for 3 min each before mounting with Mount-Quick (COSMO BIO Co., LTD., Japan). Images were captured with a microscope (BZ-X800, Keyence, Japan).

For immunohistochemistry, frozen brain sections were immobilized with 10% formalin neutral buffer solution (FUJIFILM Wako Pure Chemical Corporation, Japan) at 4 °C for 1.5 h. They were then treated with 0.3% Triton-X in PBS, 1% BSA at r.t. for 1.0 h. They were incubated with Blocking One Histo (NACALAI TESQUE, INC., Japan) for 10 min at r.t. to block non-specific staining, followed by incubation in a chamber at 4 °C overnight with the primary antibody (anti-dopamine β-hydroxylase antibody, clone 4F10.2, MAB308, Merck, Germany). After washing with PBS-T, they were incubated with the secondary antibodies (goat anti-mouse IgG (H + L) highly cross-absorbed secondary antibody, Alexa Fluor 546, A-11030, Thermo Fisher SCIECTIFIC K.K., Japan) at r.t. for 1 h. They were covered with VECTASHIELD Hard Set Mounting Medium with DAPI (Vector Laboratories, Inc., CA, United States). Images were captured with a fluorescence microscope (BZ-X800, Keyence, Japan).

Statistics

Results are expressed as mean ± SD of replicate experiments. Multiple group comparisons were performed by one-way analysis of variance followed by Dunnett’s post hoc multiple comparison test. Statistical analyses were performed using Prism v9 software (GraphPad Software Inc., La Jolla, CA). Differences were considered significant at p < 0.05.