Development and evaluation of deuterated [18F]JHU94620 isotopologues for the non-invasive assessment of the cannabinoid type 2 receptor in brain

Abstract

Background: The cannabinoid type 2 receptors (CB2R) represent a target of increasing importance in neuroimaging due to its upregulation under various neuropathological conditions. Previous evaluation of [¹⁸F]JHU94620 for the non-invasive assessment of the CB2R availability by positron emission tomography (PET) revealed favourable binding properties and brain uptake, however rapid metabolism, and generation of brain-penetrating radiometabolites have been its main limitations. To reduce the bias of CB2R quantification by blood-brain barrier (BBB)-penetrating radiometabolites, we aimed to improve the metabolic stability by developing -d₄ and -d₈ deuterated isotopologues of [¹⁸F]JHU94620.

Results: The deuterated [¹⁸F]JHU94620 isotopologues showed improved metabolic stability avoiding the accumulation of BBB-penetrating radiometabolites in the brain over time. CB2R-specific binding with K_D values in the low nanomolar range was determined across species. Dynamic PET studies revealed a CB2R-specific and reversible uptake of [¹⁸F]JHU94620-d₈ in the spleen and to a local hCB2R(D80N) protein overexpression in the striatal region in rats.

Conclusion: These results support further investigations of [¹⁸F]JHU94620-d₈ in pathological models and tissues with a CB2R overexpression as a prerequisite for clinical translation.

Keywords: Cannabinoid type 2 receptor; Neuroinflammation; Positron emission tomography; [18F]JHU94620.

Scheme 1

Synthesis of JHU94620-d₄

Scheme 2

Synthesis of JHU94620-d₈

Scheme 3

General synthesis scheme for [¹⁸F]JHU94620-d_x; x = 0, 4 or 8

Fig. 1

Development of [¹⁸F]JHU94620-d₈ as well as its radiometabolite analysis in vivo in rodents and in vitro using liver microsome assays. a The structure of [¹⁸F]JHU94620, its deuterated analogues and representative radio-HPLC chromatograms (method A) of [¹⁸F]JHU94620 (upper HPLC chromatogram) and [¹⁸F]JHU94620-d₈ (lower HPLC chromatogram) obtained from the brain of CD-1 mice 30 min p.i.; b parent fraction of the radiotracer determined in plasma, brain and spleen in CD-1 mice at 30 min (n = 7, 10 and 4, respectively, mean ± SD); c in Wistar rats over time (n = 3, each timepoint; mean ± SD); and d derived from human liver microsome (HLM, n = 1) and rat liver microsome assay (RLM, n = 1) assay, where the curves represent the exponential fits (c and d)

Fig. 2

In vitro binding studies of [¹⁸F]JHU94620-d₈ to the mice, rat, pig and human CB2R. a haematoxylin and eosin staining (H&E) of cryosections of mouse, rat and pig spleen, and autoradiograms provided in quantilumen per pixel (QL px⁻¹) of total binding and blocking with 10 µM of GW405833 or self-blocking. Additionally, a magnified section of the pig spleen (square) is shown; b exemplary autoradiograms of rat spleen cryosections incubated with the radioligand or—metabolites derived from a 60 min human liver microsomal digestion (HLM_60min); and c determination of the equilibrium dissociation constant (K_D) by homologous displacement studies using cell membranes of CHO-cells overexpressing the human CB2R (hCB2R) or cryosections of rat and mouse spleens in autoradiography studies (means ± SD, n = 3)

Fig. 3

Dynamic PET study to investigate the CB2R-specific uptake of [¹⁸F]JHU94620-d₈ into the spleen of rat over time. a Exemplary maximal intensity projections (MIPs) of T1-weighted MR, averaged PET frames from 0 to 30 min after administration of the radioligand under baseline and pre-blocking with GW405833, and merged MIPs (left to right), whereby the dotted circle marks the region of the spleen (magnified region in PET at the right); b Time activity curves of the mean standardized uptake value (SUV_mean ± SD) of the left ventricle (blood) and spleen; and c the SUV ratio (SUVr ± SD) of the spleen normalized to blood under baseline (Vehicle) and after pre-blocking with 1.5 mg kg⁻¹ bodyweight GW405833 or JHU94620-d₈ (n = 3)

Fig. 4

PET imaging of [¹⁸F]JHU94620-d₈ in a rat model with a local hCB2R overexpression in the right hemisphere of the brain. a Representative coronal planes of merged MR and PET images of averaged time frames of the control group and displacement group (5mg kg⁻¹ GW405833 injected 20 min after the start of the PET acquisition); b and c showing the corresponding time-activity curves (TACs) of the right (target region, hCB2R D80N) and the reference regions (contralateral and cerebellar region) expressed in mean standardized uptake values (SUV_mean), as well as the normalized TACs to the reference regions (SUVr); d HE-staining and autoradiograms (in vitro) of a 10 µm brain section showing the total binding of the radiotracer, co-incubation with 10 µM GW405833 or reference compound (target region, hCB2R D80N, circled); and e homologues displacement study (autoradiography) for the estimation of the binding affinity of the radiotracer (n = 3)

Fig. 5

Estimation of the total volume distribution V_T and the binding potential BP_ND of [¹⁸F]JHU94620-d₈ in the brain of rats with a local overexpression of the hCB2R(D80N) in the right hemisphere. a exemplary parametric V_T map in mL ccm⁻¹ (two-tissue compartment modelling) of a coronal head plane; b comparison of the V_T in different brain regions (n = 3);cexemplary parametric BP_ND map (SRTM); and d comparison of the BP_ND in different brain regions (n = 3)