Further validation to support clinical translation of [¹⁸F]FTC-146 for imaging sigma-1 receptors

Unless otherwise stated, chemicals were purchased from commercial sources and used without further purification. BD1047 for mice studies was purchased from Sigma Aldrich. [¹⁹F]FTC-146 (99.1 % chemical purity) reference compound was provided by Albany Molecular Research Inc. (Albany, NY). Radiochemistry and semi-preparative high performance liquid chromatography (HPLC) was performed using a TRACERlab FX-FN module (GE Healthcare) with an ancillary HPLC pump (Dionex P680) and UV detector (KNAUER K-2001). Analytical HPLC was performed on a quaternary pump (Agilent 1200 series) equipped with an autosampler, a photodiode array UV detector, and a model 105S single-channel radiation detector (Carroll & Ramsey Associates). Radiotracer doses for animal studies were measured using a dose calibrator (Capintec, CRC-15 PET). Positron emission tomography/computed tomography (PET/CT) imaging of mice was performed using microPET/CT hybrid scanner (Siemens Inveon). All PET data were reconstructed using a 3-dimensional ordered subsets expectation maximization algorithm (16 subsets, 4 iterations) with scatter and dead time correction, and a matrix size of 128 × 128 × 159. Attenuation correction was applied to dataset from CT image.

In vitro binding assays for VAChT were conducted with human VAChT stably expressed in PC12^A123.7 cells at about 50 pmol/mg of crude extract. No significant amounts of S1Rs and S2Rs were present. The VAChT radioligand used was 5 nM (−)-[³H]vesamicol (24.3 Ci/mmol; 899 GBq/mmol), and the assay was conducted at final concentrations of 10⁻¹¹ M to 10⁻⁴ or 10⁻⁵ M of S1R-targeting compounds [21]. Nonspecific binding was determined with 5 μM (±)-vesamicol. Unlabeled (−)-vesamicol was used as an external standard (K _i ~ 15 nM) and the sealed mixtures were allowed to equilibrate at 23 °C for 20 h. Duplicate data were averaged and fitted by regression of a rectangular hyperbola to estimate the K _i values of the novel compounds (Additional file 1: Figure S1).

Oprs1 mutant (+/−) OprsGt (IRESBetageo) 33Lex litters on a C57BL/6 J × 129 s/SvEv mixed background were purchased from the Mutant Mouse Regional Resource Center, UC Davis, CA, from which homozygous WT (Sigma-1 receptor +/+) and S1R-KO (Sigma-1 receptor −/−) mice were obtained through in-house breeding. All mice were maintained on a 4 % fat diet (Harland Teklad, 8604 M/R) and subjected to standard light cycles (12 h/12 h light–dark). Stanford University Institutional Animal Care and Use Committee approved all experimental procedures involving animals.

Mice were bred onsite, and tail samples were collected from mice litters and sent to Transnetyx Inc to determine genotype (Cordova, TN, USA). Mice were always re-genotyped at the conclusion of each experiment.

[¹⁸F]FTC-146 was synthesized via aliphatic nucleophilic substitution (¹⁸F/tosylate exchange) using TRACERlab FX-FN as previously described [19]. Briefly, tosylate precursor solution (2 mg in 1 mL anhydrous DMSO) was added into azeotropically dried ¹⁸F/K₂₂₂/K₂CO₃ complex, heated to 150 °C for 15 min, and then the crude product was purified on semi-prep HPLC. The [¹⁸F]FTC-146 HPLC fraction was formulated in saline containing no more than 10 % ethanol.

Mice were scanned with a dedicated small animal 7 Tesla Varian Magnex Scientific MR scanner with custom-designed pulse sequences and radiofrequency (RF) coils using standard methods. The scanner consisted of a superconducting magnet (Magnex Scientific) with 7.0 Tesla field strength, a gradient insert (Resonance Research, Inc.) with clear bore size of 9 cm (770 mT/m, 2500 T/m/s), a General Electric (GE) console, and Copley 266 amplifiers. For each MR scan, mice were anesthetized using isoflurane gas (2.0–3.0 % for induction and 1.5–2.5 % for maintenance). Body temperature was continually monitored and maintained within the normal range using a temperature sensor. Respiration rate and oxygen saturation were also monitored throughout scan. Coronal brain images were acquired using T2-weighted fast spin echo sequences (TE/TR 58.5 ms/4000 ms) using 9 NEX, a 256 × 256 matrix, 3 cm field of view, slice thickness of 500 μm, and a total imaging time of 19 min. Mouse brain MRI data was used to provide an anatomical reference frame for PET/CT imaging.

Mice were anesthetized using isoflurane inhalation (2.0–3.0 % for induction and 1.5–2.5 % for maintenance, isofluorane in oxygen at a flow rate of 2 L/min). Following tail vein cannulation, each catheterized mouse was placed in a custom-made 4 × 4 mouse bed. All mice were kept warm using an infrared warming pad (Kent Scientific) placed under and around the bed while under isoflurane anesthesia in the microPET/CT scanner (pixel size 200 μm). A 5-min CT scan was performed just prior to each PET scan. Acquisition of dynamic PET data (4 × 15 s, 4 × 60 s, 11 × 300 s) for WT (n = 4) and S1R-KO (n = 4) mice commenced just prior to intravenous administration of [¹⁸F]FTC-146 (45–130 μCi; 1.67–4.81 MBq), followed by 50 μL sterile heparinized saline to flush cannula, and yielded a total of 19 frames over 60 min. Blocking studies involved pre-treating WT (n = 4) and S1R-KO (n = 4) mice with BD1047 (1 mg/kg, Sigma Aldrich) 10 min prior to radioligand administration. PET, CT, and MR images were imported into Inveon Research Workspace 4.0 (Siemens) and co-registered using the automatic affine registration. Tracer activity over time was measured in regions of interest in the PET image (mean activity in the region) by using each brain MR image as an anatomical guide (Additional file 2: Figure S2). Regions of interest were selected prior to the commencement of this study, and included previously reported S1R-rich brain structures (i.e., caudate putamen, cerebellum, cortex, and hippocampus) [22].

WT (n = 5) and S1R-KO mice (n = 5) were administered [¹⁸F]FTC-146 (60–90 μCi; 2.2-3.3 MBq) via tail vein injection. Blood samples were collected via cardiac puncture 25 min following tracer administration, and each mouse was perfused with saline (10–20 mL) prior to sacrificing and removing organs of interest (i.e., brain, heart, kidney, liver, lung, muscle, pancreas, small intestine, spleen, stomach). The radioactivity in weighed organs was assessed in automated gamma counter (Cobra II; Packard) and decay corrected to the time of radiotracer injection using diluted aliquots of the initial administered dose as standards. Another set of WT mice (n = 4) were used to evaluate the ex vivo stability of [¹⁸F]FTC-146 in brain, liver, kidney, small intestine, and stomach homogenates at 25 min after tracer administration (using previously reported methods [20]).

Coronal brain sections from WT (n = 3), heterozygous (n = 1), and S1R-KO (n = 4) mice were harvested 30 min post-injection of [¹⁸F]FTC-146 (200–300 μCi; 7.4–11.1 MBq). After perfusing each mouse with saline (15–20 mL), the brain was removed and divided into two sagittal halves. Half of each brain was embedded in optimal cutting temperature (OCT) compound (Tissue-Tek) and frozen on dry ice for ex vivo autoradiography, while the other half was fixed in 4 % paraformaldehyde (PFA) for immunohistochemistry. Muscle samples from front leg, known to contain negligible levels of S1R, were also dissected and embedded alongside corresponding brain tissue to be used as an internal reference region for each mouse in order to normalize ex vivo autoradiography data. Coronal brain and muscle sections (20 μm thick) were cut using a cryostat microtome HM500 (Microm), mounted on microscope slides (Fisherbrand SuperfrostTM Plus Microscope Slides), air-dried for at least 10 min, and then exposed to ¹⁸F-sensitive storage phosphor screens (Perkin Elmer) overnight at −20 °C. Each phosphor screen was scanned using a Typhoon 9410 Variable Mode Imager (Amersham Biosciences) and image data were visualized and quantified using ImageJ (Image processing and analysis software in Java, version 1.45 s). Anatomy of brain sections was confirmed by Nissl staining (cresyl violet acetate; Sigma Aldrich) using standard techniques.

S1R immunostaining was performed using free-floating 40-μm-thick coronal brain sections from WT (n = 3) and S1R-KO (n = 4) mice. Sections were incubated in a 1 % H₂O₂, 50 % TBS (pH = 7.4)/MeOH solution for 30 min to quench the endogenous peroxidase activity and subsequently blocked with 10 % normal goat serum (NGS; Vector Laboratories) in TBST (1 % Triton X-100, Sigma Aldrich) for 1 h to reduce nonspecific staining and to permeabilize the tissue. Sections were then incubated with S1R specific primary antibody (1:400, [23]) containing 5 % NGS and TBST overnight at room temperature. Following this, the sections were incubated with biotinylated anti-rabbit secondary antibody 1:200 (Vector Laboratories) in 5 % NGS and TBST for 1 h at room temperature and then with avidin-biotin complex (diluted 1:1000 in TBS, Vector Laboratories) for 90 min at room temperature. Finally, sections were incubated with 3,3′-diaminobenzidine (DAB; Sigma Aldrich) for 10 min, mounted on slides, dehydrated, and then cover-slipped with Permount (Sigma Aldrich). Images of the immunostained sections were taken with a Nanozoomer (Hamamatsu), and images were quantified using ImageJ.

Mice (WT, female n = 3; male n = 3) were imaged for 2 h using Inveon PET/CT following a high-resolution CT (pixel size 103.2 μm). An additional two 10-min static PET/CT scans at 4 and 6 h were performed on these mice that were allowed to be awake between scans. ROIs were drawn over the brain, thyroid, heart, liver, lung, gallbladder, spleen, kidneys, urinary bladder, bone, muscle, and testicle/uterus of the mice by using the image analysis software IRW (Siemens) with the help of the CT images. The %ID/g associated with each ROI was extracted and underwent an animal-human biokinetic extrapolation using the percent kg/g method. Source organ residence times were calculated using a standard quantitation platform organ level internal dose assessment (OLINDA) utilizing a bi-exponential model without bladder voidance. The projected human radiation doses were then computed for male and female phantoms using the source organ residence times.

This study was conducted by SoBran Bioscience (Baltimore, MD). It consisted of one test article treatment group of ten male and ten female Sprague–Dawley rats dosed with [¹⁹F]FTC-146 at 0.069 mg/kg, equivalent to 250× the anticipated clinical dose mass. An additional group of ten males and ten females received the vehicle, 10 % ethanol in sterile saline, and served as the control. All rats received a dose volume of 5 mL/kg. The rats were dosed intravenously once on study day 1. Five male and five female rats from each group were bled on study day 3 and the remaining rats were bled on study day 15. All animals were euthanized and necropsied following blood collection. Parameters evaluated for test article effect included survival, clinical observations, body weight, body weight gain, clinical pathology, gross pathology, organ weights, and microscopic pathology.

Results are presented as mean value ± standard deviation (STDV) for radiochemistry, PET/CT imaging, ex vivo biodistribution and stability and dosimetry studies; mean value ± standard error of the mean (SEM) is shown for ex vivo autoradiography and immunohistochemistry studies. Unpaired t tests (one tail) were used to compare PET/autoradiography/IHC data from WT with S1R-KO mice. Correlations between autoradiography mean signal intensity and S1R staining were determined by Pearson correlation test. All statistical analyses were performed using Microsoft Excel (version 14.0.7) or GraphPad Prism software (version 6.0c; GraphPad). A p value of less than 0.05 was considered as statistically significant.