A tool for nuclear imaging of the SARS-CoV-2 entry receptor: molecular model and preclinical development of ACE2-selective radiopeptides

Three‐dimensional molecular model of the ACE2/DX600 complex

The 3D-structure of the DX600 peptide in water has been solved by NMR [31]. The resulting conformational ensemble showed that the constraints imposed by the intramolecular disulfide bridge leads to the formation of a central closed structured loop, while the C and N termini have a predominantly disordered structure. The structured loop contains a consensus inhibitory motif (CxPxRxxPWxxC) [25] that is likely to interact with the buried catalytic residues at the core of the catalytic domain of ACE2. The structure of human ACE2 bound to the small-molecular-weight inhibitor MLN-4760 [32] was used as a template to model the complex with the DX600 peptide using PyMOL (The PyMOL Molecular Graphics System, Version 2.5.2, Schrödinger, LLC). This model was subsequently refined using the Haddock v2.4 software [33]. Finally, an N-α-acetyl-lysine residue coupled to a gallium-coordinating HBED-CC chelator on the ε-amino group was fused to Gly1 in DX600 using the crystal structure of HBED-CC bound to the nickel-binding periplasmic protein as a template.

ACE2- and ACE-targeting peptides

The ACE2-targeting peptides were obtained as custom synthesis from piCHEM (Research & Development GmbH, Raaba-Grambach, Austria). The cyclic DX600 peptides were functionalized with an N-α-acetyl-lysine residue at the N terminus. Subsequently, a DOTA, NODAGA or HBED-CC chelator, respectively, was coupled to the N terminus at the ε-amino group present on the lysine sidechain to obtain DOTA-DX600, NODAGA-DX600 and HBED-CC-DX600 (Fig. 1; Additional file 1: Table S1). The chemical identity and purity (> 95%) of the purchased peptides were confirmed by HR-MS (MALDI-TOF) and analytical HPLC, respectively (Additional file 1: Fig. S1, Table S2). The unmodified DX600 peptide, herein referred to as cyclo-DX600 (piCHEM, Research & Development GmbH, Raaba-Grambach, Austria), was used for in vitro experiments to block ACE2.

An ACE-targeting agent was designed based on the BPP9a peptide as a lead structure [34]. In brief, the immobilized and side chain-protected BPP9a peptide was produced by standard Fmoc solid-phase peptide chemistry. Afterward, the peptide was functionalized with a spacer constituted of two consecutive 6-aminohexanoic residues at the N terminus. The resulting N-terminal amino group was coupled with a DOTA-tris(tBu) ester. Acid-catalyzed cleavage from the resin led to complete removal of the protecting groups (Additional file 1: Scheme S1). The crude peptide was purified using semipreparative HPLC, resulting in moderate yields (19%) of the desired DOTA-BPP9a (Fig. 1). The chemical identity and purity of the peptide were confirmed by HRMS and HPLC analyses, respectively (Additional file 1: Fig. S2).

Post-purification of gallium-67

Gallium-67 (no-carrier-added [67Ga]GaCl3 in ~ 0.1 M HCl) was purchased from Curium Netherlands B.V., the Netherlands, via b.e.imaging GmbH (Switzerland). A post-delivery process of the purchased gallium-67 was necessary to enable radiolabeling at the indicated molar activities. This was carried out according to a previously published method [35]. In brief, other trivalent transition metals were selectively reduced by addition of TiCl3 and, subsequently, the desired [67Ga]GaCl3 was isolated by absorption chromatography performed on Amberchrom CG-161M resin. An extraction resin was used for concentration of the [67Ga]GaCl3 in a small volume of dilute HCl.

Preparation of the radiopeptides and stability of the radiopeptide formulation

The purified [67Ga]GaCl3 in HCl (~ 0.1 M) was added to a sodium acetate solution (0.5 M, pH 8) to obtain a buffered system of pH ~ 4. After adding the respective peptide (DOTA-DX600, NODAGA-DX600, HBED-CC-DX600 or DOTA-BPP9a) in a quantity to obtain the desired molar activity (5–20 MBq/nmol), the reaction mixture was incubated at 95 °C for 15 min. Quality control was performed using a Merck Hitachi LaChrom HPLC system (Additional file 1). The radiopeptides were used directly for in vitro and in vivo studies without further purification steps. The stability of the radiopeptides (20 MBq/nmol) formulated in 0.9% NaCl (10 MBq/100 µL) was assessed at room temperature (Additional file 1). The highest molar activity used in these studies was 20 MBq/nmol which means that one of 73–74 molecules was labeled with gallium-67. Nevertheless, the labeling efficiency of the DX600-based peptides was assessed using the highest possible molar activity that would still allow for the preparation of the DX600-based radiopeptide at a radiochemical purity of > 95% (Additional file 1).

Distribution coefficients of the radiopeptides

The distribution coefficients (logD values) of [67Ga]Ga-DOTA-DX600, [67Ga]Ga-NODAGA-DX600, [67Ga]Ga-HBED-CC-DX600 and [67Ga]Ga-DOTA-BPP9a were determined by the shake flask method. An aliquot of the respective radiopeptides (0.5 MBq; 25 pmol, 25 µL) was added to a vial containing a mixture of phosphate buffered saline (PBS) pH 7.4 (1475 µL) and n-octanol (1500 µL). After vortexing vigorously for 1 min, the vials were centrifuged (560 rcf; 6 min) to obtain phase separation. The quantity of activity in the organic and aqueous phases was determined in a γ-counter (PerkinElmer, Wallac Wizard 1480). The distribution coefficients were expressed as the logarithm of the ratio of counts per minute (cpm) measured in the n-octanol phase to the cpm measured in the PBS phase and indicated as the average of three independent measurements (± standard deviation, SD), each performed with five replicates.

Blood plasma protein-binding properties

The radiopeptides (0.3 MBq; 15 pmol) were incubated in human or murine blood plasma (150 μL) for 30 min at 37 °C followed by cooling on ice, addition of ice-cold PBS (150 μL) and transfer of the samples to amicon filter devices (Amicon Ultra, 0.5 mL, Merck Millipore; cutoff 10 kDa). The filter ensured the retention of the radiopeptides bound to mouse plasma proteins or human plasma proteins during filtration (14,000 rcf, 30 min at 4 °C). Afterward, the filter inserts were inverted and centrifuged again (200 rcf, 3 min) to recover the protein-bound fraction of the radiopeptides. The activities of the free radiopeptides in the filtrate, as well as in the filter insert and the plasma protein-bound radiopeptides (Abound), were counted for activity separately in a γ-counter. The total activity (Atotal) was determined as the sum of the activities in the filtrate, filter insert, and protein-bound fractions. The percentage of radiopeptide bound to mouse serum albumin (MSA) and human serum albumin (HSA) was calculated as Abound/Atotal*100. The results were presented as average ± SD of three independent experiments (Additional file 1).

Cell culture

HEK-293 cells transfected with ACE2 or ACE, referred to as HEK-ACE2 and HEK-ACE, respectively, were obtained from Innoprot (Innovative Technologies in Biological Systems S.L. Bizkaia, Spain). The cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with nonessential amino acids, fetal calf serum and antibiotics. Hygromycin B was used to maintain the expression of ACE2 and ACE, respectively. The HEK cells were cultured under standard conditions at 37 °C and 5% CO2 and subcultured using PBS/EDTA and trypsin.

Validation of ACE2 and ACE expression in HEK cells

Expression of ACE2 or ACE on the respective HEK cell lines was verified using immunofluorescence microscopy. HEK-ACE and HEK-ACE2 cells grown on coverslips were fixed using paraformaldehyde solution (4%). After rinsing with PBS, the cells were incubated with the blocking solution [5% bovine serum albumin (BSA)] containing Triton X-100 in PBS. After removal of the blocking solution, mouse anti-ACE2 antibody (1:200 diluted; Santa Cruz Biotechnology, Inc. sc-390851) and mouse anti-ACE (1:50 diluted; Santa Cruz Biotechnology, Inc. sc-23908) in 1% BSA were added, followed by incubation at 4 °C overnight. The cells were rinsed with PBS before incubation with equine anti-mouse secondary antibody conjugated with AlexaFluor® 488 (1:500 dilution in 1% BSA; abcam ab150105) for 1 h protected from light. The samples were stained with DyLight™ 554 phalloidin and 4′,6-diamidino-2-phenylindole (DAPI) to visualize F-actin of the cytoskeleton and nuclei, respectively, followed by rinsing with PBS and mounting onto glass slides using ProLong Gold Antifade mounting agent. Fluorescence images were acquired using a confocal microscope (Leica Stellaris 5). The images were analyzed using the ImageJ software version 1.52d. ACE2 and ACE were visualized on the HEK-ACE2 and HEK-ACE cell line, respectively (Fig. 2). The expression of ACE2 and ACE on the respective HEK cell lines was also confirmed by Western blot (Additional file 1).

Fig. 2figure 2

Immunofluorescence microscopy images of A HEK-ACE2 and B HEK-ACE cells. Nuclei were stained with DAPI (blue); F-actin of the cytoskeleton was stained with fluorescent phalloidin (red); ACE2 and ACE were stained with a fluorophore-conjugated secondary antibody (green) after incubation with anti-ACE2 and anti-ACE antibodies; The merged images show the overlay of all stainings. The scale bar corresponds to 20 μm

Cell uptake of the radiopeptides

The uptake and internalization of [67Ga]Ga-DOTA-DX600, [67Ga]Ga-NODAGA-DX600, [67Ga]Ga-HBED-CC-DX600 and [67Ga]Ga-DOTA-BPP9a were determined using HEK-ACE2 and HEK-ACE cells. The cells were seeded in poly-D-lysine-coated 12-well plates allowing cell adhesion and growth overnight. After rinsing the cells with PBS, they were incubated with the respective radiopeptide (25 µL, 1.9 pmol, 38 kBq). In some cell samples, the radiopeptides were co-incubated with excess of cyclo-DX600 (2 µM) or lisinopril (4 µM) to block ACE2 or ACE, respectively. After incubation of the cells for 1 h or 3 h, they were rinsed with PBS or acidic stripping buffer (glycine buffer with NaCl 0.9%, pH 2.8) to determine the total uptake and internalization of the radiopeptides, respectively. Cell samples were lysed using NaOH (1 M, 1 mL) before counting the activity in a γ-counter. The results were expressed as a percentage of total added activity and normalized to an average content of ~ 0.3 mg protein per well.

ACE2- and ACE-binding affinity of the radiopeptides

In order to determine the KD values of the DX600-based radiopeptides and [67Ga]Ga-DOTA-BPP9a, HEK-ACE2 and HEK-ACE cells were seeded in 48-well-plates allowing adhesion and growth overnight. The cell samples were incubated with increasing concentrations (1‒2000 nM) of the radiopeptides (5 MBq/nmol) on ice. Some of the cell samples were co-incubated with an excess of cyclo-DX600 or lisinopril (20 µM, 10 nmol) to block ACE2 or ACE, respectively, enabling the determination of non-specific binding (Additional file 1). After 1 h incubation, the cells were rinsed with PBS and lysed with NaOH (1 M, 600 µL) to count the activity in a γ-counter. The KD values were determined by plotting specific binding (total binding minus unspecific binding) against the molar concentration of the added radiopeptide. A nonlinear regression analysis was performed using GraphPad Prism software (version 8.3.1).

In vitro plasma stability of the radiopeptides

The 67Ga-labeled DX600-radiopeptides (20 MBq/nmol) were incubated in human and mouse plasma (10 MBq/200 µL), respectively, at 37 °C. Aliquots were taken at variable time points up to 24 h for analysis using reversed phase thin layer chromatography (TLC) with citrate buffer (0.1 M, pH 5.5) as a mobile phase. Images of the TLC plates were obtained using a storage phosphor system (Cyclone Plus, PerkinElmer) and quantified using the OptiQuant software (version 5.0, Bright Instrument Co Ltd., PerkinElmer™) (Supplementary Information).

In vivo studies—mouse model

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. In particular, all animal experiments were carried out according to the guidelines of the Swiss Regulations for Animal Welfare. Preclinical studies have been ethically approved by the Cantonal Committee of Animal Experimentation and permitted by the responsible cantonal authorities (License N° 75743). Five-week-old female CD1/nude mice and female immunocompetent FVB mice were obtained from Charles River Laboratories (Sulzfeld, Germany) and fed with standard rodent chow ad libitum. CD1/nude were subcutaneously inoculated with HEK-ACE2 cells (6–8 × 106 cells in 100 µL PBS) on the right shoulder and approximately one week later with HEK-ACE (4 × 106 cells in 100 µL PBS) on the left shoulder. Imaging and biodistribution studies were performed 2–4 weeks later when the xenografts reached a size of 100–300 mm3.

Biodistribution studies in xenografted mice

Biodistribution studies in xenograft-bearing mice were performed 2–4 weeks after cell inoculation. Mice were intravenously injected with [67Ga]Ga-DOTA-DX600, [67Ga]Ga-NODAGA-DX600, [67Ga]Ga-HBED-CC-DX600 or [67Ga]Ga-DOTA-BPP9a (3 MBq, 0.5 nmol, 100 µL) diluted in NaCl 0.9% containing 0.05% BSA. The mice were sacrificed 3 h after administration of the radiopeptides. Selected tissues and organs were collected, weighed and counted using a γ-counter. The results were listed as a percentage of the injected activity per gram of tissue mass (% IA/g), using counts of a defined volume of the original injection solution measured simultaneously, resulting in decay-corrected values. Statistical analysis was performed by applying a one‐way ANOVA test with a Tukey’s multiple comparisons post‐test (GraphPad Prism software, version 8.3.1).

SPECT/CT imaging studies

SPECT/CT imaging was performed using a four-head, multiplexing, multipinhole small-animal SPECT camera (NanoSPECT/CT™, Mediso Medical Imaging Systems, Budapest, Hungary). Each head was outfitted with a tungsten-based aperture of nine 1.4 mm-diameter pinholes and a thickness of 10 mm. CT scans of 7.5 min duration were followed by SPECT scans of ~ 50 min performed 1 h, 3 h and 24 h after injection of the respective radiopeptide (10 MBq, 0.5 nmol, 100 μL), diluted in NaCl 0.9% containing 0.05% BSA. During the scan, mice were anesthetized with a mixture of isoflurane and oxygen. The images were acquired using Nucline software (version 1.02, Mediso Ltd., Budapest, Hungary). The real-time CT reconstruction used a cone-beam-filtered backprojection. The reconstruction of SPECT data was performed with the HiSPECT software (version 1.4.3049, Scivis GmbH, Göttingen, Germany) using γ-energies of 93.20 keV (± 10%), 184.60 keV (± 10%) and 300.00 keV (± 10%) for gallium-67. All images were prepared using the VivoQuant post-processing software (version 3.0, inviCRO Imaging Services and Software, Boston, US). A Gauss post-reconstruction filter (full width at half maximum, 1 mm) was applied, and the scale of activity for gallium-67 was set as indicated on the SPECT/CT images. In order to estimate the wash-out of the DX600-based radiopeptides from the HEK-ACE2 xenografts and kidneys over time, the absolute amount of activity was determined in these tissues using the VivoQuant quantification tool. The accumulated activity (non-decay-corrected value in MBq) in the xenografts and kidneys, respectively, obtained at the 1 h p.i. timepoint was set as 100%. The percentage of activity retained in these tissues was determined at 3 h and 24 h p.i. based on the absolute activity determined at these timepoints.

In vivo stability of the radiopeptides

In order to investigate the stability of the radiopeptides, the 67Ga-labeled DX600 peptides (20 MBq/nmol) were tested in immunocompetent FVB mice (10 MBq in 100 μL per mouse). Before sacrificing, blood was collected from each mouse 1 h after injection of the radiopeptides. After centrifugation, the blood plasma was analyzed using reversed phase TLC with citrate buffer (0.1 M, pH 5.5) as a mobile phase. The kidneys were collected and processed to investigate the presence of intact radiopeptides and potential metabolites using the same TLC system. The analysis of the chromatograms was performed as described for the in vitro plasma stability (Additional file 1).

In vitro autoradiography

Autoradiography studies were performed on frozen tissue sections of HEK-ACE2 and HEK-ACE xenografts. The sections were exposed to [67Ga]Ga-HBED-CC-DX600 (0.1–20 MBq/nmol; 0.5 MBq/mL) in Tris–HCl buffer (170 mM, pH 7.6, with 5 mM MgCl2) with 1% (w/v) BSA, with molar activities ranging from 0.1 to 20 MBq/nmol, for 60 min at room temperature. Cyclo-DX600 (10 µM) was added to block ACE2 binding. After incubation, the tissue sections were rinsed with buffer and MilliQ-water and air-dried. Images were obtained using a storage phosphor system (Cyclone Plus, PerkinElmer) and quantified using the OptiQuant software (version 5.0, Bright Instrument Co Ltd., PerkinElmer™). The specific binding to ACE2 and ACE was calculated by subtracting the signal intensity measured as digital light unit (DLU)/mm2 on the slides incubated with excess of cyclo-DX600 from the DLU/mm2 signal of the sections incubated with only the radiopeptide. The specific binding of the radiopeptide at various molar activities (0.1–20 MBq/nmol) was expressed as a percentage of the signal reached at 20 MBq/nmol which was set as 100%. The results are presented as average ± SD of three independent experiments, each performed with two replicates of HEK-ACE2 xenograft tissue and one replicate of HEK-ACE xenograft tissue.

留言 (0)

沒有登入
gif