In vivo Imaging of Cannabinoid Type 2 Receptors: Functional and Structural Alterations in Mouse Model of Cerebral Ischemia by PET and MRI

Purpose Stroke is one of the most prevalent vascular diseases. Non-invasive molecular imaging methods have the potential to provide critical insights into the temporal dynamics and follow alterations of receptor expression and metabolism in ischemic stroke. The aim of this study was to assess the cannabinoid type 2 receptor (CB2R) levels in transient middle cerebral artery occlusion (tMCAO) mouse models at subacute stage using positron emission tomography (PET) with our novel tracer [18F]RoSMA-18-d6 and structural imaging by magnetic resonance imaging (MRI). Procedures Our recently developed CB2R PET tracer [18F]RoSMA-18-d6 was used for imaging neuroinflammation at 24 h after reperfusion in tMCAO mice. The RNA expression levels of CB2R and other inflammatory markers were analyzed by quantitative real-time polymerase chain reaction using brain tissues from tMCAO (1 h occlusion) and sham-operated mice. [18F]fluorodeoxyglucose (FDG) was included for evaluation of the cerebral metabolic rate of glucose (CMRglc). In addition, diffusion-weighted imaging and T2-weighted imaging were performed for anatomical reference and delineating the lesion in tMCAO mice. Results mRNA expressions of inflammatory markers TNF-α, Iba1, MMP9 and GFAP, CNR2 were increased to 1.3–2.5 fold at 24 h after reperfusion in the ipsilateral compared to contralateral hemisphere of tMCAO mice, while mRNA expression of the neuronal marker MAP-2 was markedly reduced to ca. 50 %. Reduced [18F]FDG uptake was observed in the ischemic striatum of tMCAO mouse brain at 24 h after reperfusion. Although higher activity of [18F]RoSMA-18-d6 in ex vivo biodistribution studies and higher standard uptake value ratio (SUVR) were detected in the ischemic ipsilateral compared to contralateral striatum in tMCAO mice, the in vivo specificity of [18F]RoSMA-18-d6 was confirmed only in the CB2R-rich spleen. Conclusions This study revealed an increased [18F]RoSMA-18-d6 measure of CB2R and a reduced [18F]FDG measure of CMRglc in the ischemic striatum of tMCAO mice at subacute stage. [18F]RoSMA-18-d6 might be a promising PET tracer for detecting CB2R alterations in animal models of neuroinflammation without neuronal loss. Supplementary Information The online version contains supplementary material available at 10.1007/s11307-021-01655-4.

The aim of the current study was to evaluate the novel CB 2 R tracer, [ 18 F]RoSMA-18-d6, in the transient middle cerebral artery occlusion (tMCAO) mouse models of focal cerebral ischemia [64][65][66][67][68][69][70] using microPET. In addition, [ 18 F] FDG was included for evaluation of the cerebral metabolic rate of glucose (CMRglc) and as a readily available radiotracer benchmark for neuroinflammation [71,72]. Diffusion-weighted imaging (DWI) and T 2 -weighted imaging were performed for anatomical reference and for delineating the lesion in tMCAO mice. [ 18 F]RoSMA-18-d6 was synthesized by nucleophilic substitution of the tosylate precursor with [ 18 F]KF/Kryptofix222 in acetonitrile [63]. The crude product was purified by reverse phase semi-preparative high-performance liquid chromatography and formulated with 5 % ethanol in water for intravenous injection and for biological evaluations. In a typical experiment, a moderate radiochemical yield of ~ 12 % (decay corrected) was achieved with a radiochemical purity > 99 %. The molar activities ranged from 156 to 194 GBq/μmol at the end of synthesis. The identity of the final product was confirmed by comparison with the HPLC retention time of the non-radioactive reference compound by co-injection. [ 18 F] FDG was obtained from a routine clinical production from the University Hospital Zurich, Switzerland.

Introduction
The pathophysiology of ischemic stroke is complex and associated with a myriad of cellular and molecular pathways. The severe reduction in cerebral blood flow (CBF) initiates a cascade of hemodynamic, vascular and inflammatory processes in a time-dependent manner in the supplied brain territory, and subsequent defensive response for repair related to lesion expansion and containment. Irreversible tissue damage occurs in the core of the ischemic area, while neurons in the ischemic penumbra face excitotoxicity, peri-infarct polarizations, inflammation, and apoptosis, leading to a secondary tissue damage and expansion of the lesion if reperfusion cannot be restored within an early time frame [1][2][3]. Neuroinflammation post stroke has been an important therapeutic target. Anti-inflammatory, immunomodulatory treatments and microglia-targeted therapy were evaluated in clinical stroke trials [4][5][6]. Thus, there is a need for imaging the regional neuroinflammatory pattern for understanding disease mechanism and for therapeutic monitoring.
Cannabinoid type 2 receptors (CB 2 R) are mainly expressed by immune cells including monocytes and macrophages. In the brain, CB 2 Rs are primarily found on microglia and have low expression levels under physiological conditions [1,3,33]. Upregulation of brain CB 2 R expression is reported under acute inflammation such as ischemic stroke, and related to lesion extension in the penumbra and subsequent functional recovery [34]. Treatment with CB 2 R agonists has (TTC) histology staining for validation 24 h or 48 h after reperfusion. Animals were housed in ventilated cages inside a temperature-controlled room, under a 12-h dark/light cycle. Pelleted food (3437PXL15, CARGILL) and water were provided ad libitum. Paper tissue and red Tecniplast mouse house® (Tecniplast, Milan, Italy) shelters were placed in cages as environmental enrichments. All experiments were performed in accordance with the Swiss Federal Act on Animal Protection and were approved by the Cantonal Veterinary Office Zurich (permit number: ZH018/14 and ZH264/16).
Surgeries for tMCAO and sham-operation were performed using standard-operating procedures as described before [73,74]. Anesthesia was initiated by using 3 % isoflurane (Abbott, Cham, Switzerland) in a 1:4 oxygen/air mixture and maintained at 2 %. Before the surgical procedure, a local analgesic (Lidocaine, 0.5 %, 7 mg/kg, Sintectica S.A., Switzerland) was administered subcutaneously (s.c.). Temperature was kept constant at 36.5 ± 0.5 °C with a feedback controlled warming pad system. All surgical procedures were performed in 15-30 min. After surgery, buprenorphine was administered as s.c. injection (Temgesic, 0.1 mg/kg b.w.), and at 4 h after reperfusion and supplied thereafter via drinking water (1 mL/32 mL of drinking water) until 24 h or 48 h. Animals received softened chow in a weighing boat on the cage floor to encourage eating. tMCAO animals were excluded from the study if they met one of the following criteria: Bederson testing was performed 2 h post-reperfusion. Bederson score of 0, no reflow after filament removal, and premature death.

mRNA Isolation, Reverse-Transcription Reaction, and Real-Time Polymerase Chain Reaction
Brain hemispheres of C57BL/6 mouse and tMCAO mice at 24 h and 48 h post reperfusion were used for total mRNA isolation according to the protocols of the Isol-RNA Lysis Reagent (5 PRIME, Gaithersburg, USA) and the bead-milling TissueLyser system (Qiagen, Hilden, Germany). Quanti-Tect® Reverse Transcription Kit (Qiagen, Hilden, Germany) was used to generate cDNA. The primers (Microsynth, Balgach, Switzerland) used for the quantitative polymerase chain reaction (qPCR) are summarized in Supplementary Table 1. Quantitation of CNR2, ionized calcium binding adaptor molecule 1 (Iba1), tumor necrosis factor (TNF-a), matrix metallopeptidase 9 (MMP9), glial fibrillary acidic protein (GFAP), and microtubule-associated protein 2 (MAP-2) mRNA expression was performed with the DyNAmo™ Flash SYBR® Green qPCR Kit (Thermo Scientific, Runcorn, UK) using a 7900 HT Fast Real-Time PCR System (Applied Biosystems, Carlsbad, USA). The amplification signals were detected in real-time, which permitted accurate quantification of the amounts of the initial RNA template during 40 cycles according to the manufacturer's protocol. All reactions were performed in duplicates and in two independent runs. Quantitative analysis was performed using the SDS Software (v2.4) and a previously described 2 − ΔΔCt quantification method [75]. The specificity of the PCR products of each run was determined and verified with the SDS dissociation curve analysis feature.

In vivo MRI
Data were acquired at 24 h after reperfusion on a 7 T Bruker Pharmascan (Bruker BioSpin GmbH, Germany), equipped with a volume resonator operating in quadrature mode for excitation and a four-element phased-array surface coil for signal reception and operated by Paravision 6.0 (Bruker Bio-Spin) [73,[76][77][78]. Mice were anesthetized with an initial dose of 4 % isoflurane in oxygen/air (200:800 ml/min) and maintained at 1.5 % isoflurane in oxygen/air (100:400 ml/ min). Body temperature was monitored with a rectal temperature probe (MLT415, ADInstruments) and kept at 36.5 ± 0.5 °C using a warm water circuit integrated into the animal support (Bruker BioSpin GmbH, Germany). T 2 -weighted MR images were obtained using a spin echo sequence (TurboRARE) with an echo time 3 ms, repetition time 6 ms, 100 averages, slice thickness 1 mm, field-ofview 2.56 cm × 1.28 cm, matrix size 256 × 128, giving an in-plane resolution of 100 μm × 100 μm. For DWI, a fourshot spin echo-echo planar imaging sequence with an echo time = 28 ms, repetition time = 3000 [76,77], acquired with a field-of-view of 3.3 cm × 2 cm and a matrix size of 128 × 128, resulting in a nominal voxel size of 258 μm × 156 μm. Diffusion-encoding was applied in the x-, y-, and z-directions with b values of 100, 200, 400, 600, 800, and 1000 s/mm 2 , respectively, acquisition time 3 min 48 s. The ischemic lesion was determined as an area of significant reduction of the apparent diffusion coefficient (ADC) value compared with the unaffected contralateral side [79]. On T 2 -weighted images, the lesion was determined as an area of hyperintensities compared with the contralateral side.

In vivo microPET Studies
MicroPET/CT scans were performed at 24 h after reperfusion with a calibrated SuperArgus µPET/CT scanner (Sedecal, Madrid, Spain) with an axial field-of-view of 4.8 cm and a spatial resolution of 1.6-1.7 mm (full width at half maximum). tMCAO and the sham-operated C57BL/6 J mice were anesthetized with ca. 2.5 % isoflurane in oxygen/air (1:1) during tracer injection and the whole scan time period. The formulated radioligand solution ([ 18 F]FDG: 9.9-11 MBq or [ 18 F]RoSMA-18-d6: 7.2-13 MBq) was administered via tail vein injection, and mice were dynamically scanned for 60 min. For blocking experiments, 1.5 mg/kg GW405833 (Fig. 3a) was dissolved in a vehicle of 2 % Cremophor (v/v), 10 % ethanol (v/v), and 88 % water for injection (v/v) and injected together with [ 18 F]RoSMA-18-d6. Body temperature was monitored by a rectal probe and kept at 37 °C by a heated air stream (37 °C). The anesthesia depth was measured by the respiratory frequency (SA Instruments, Inc., Stony Brook, USA). µPET acquisitions were combined with CT for anatomical orientation and attenuation correction. The obtained data were reconstructed in user-defined time frames with a voxel size of 0.3875 × 0.3875 × 0.775 mm 3 as previously described [80].

Triphenyltetrazolium Chloride (TTC) Staining
To assess the ischemic lesion severity in the brain of tMCAO mice and to validate the absence of lesion in the shamoperated mice, staining with TTC staining was performed. After measurements mice were euthanized, their brains were removed and 1-mm thick brain slices were obtained with a brain matrix. Slices were incubated in a 2.5 % TTC solution (Sigma-Aldrich, Switzerland) in PBS at 37 °C for 3 min. Photographs of the brain sections were taken. Edema-corrected lesion volumes were quantified as described [81].

Biodistribution Studies in the Mouse Brain
After PET/CT scanning of tMCAO mice at 24 h after reperfusion with [ 18 F]RoSMA-18-d6, animals were sacrificed at 70 min post injection by decapitation. The spleen and brain regions of ischemic ipsilateral area and contralateral hemisphere were collected for analysis with a gamma counter. The accumulated radioactivities in the different tissues were expressed as percent normalized injected dose per gram of tissue normalized to 20 g body weight of the animals (norm. percentage injected dose per gram tissue ( % ID/g tissue)).

Data Analysis and Statistics
Images were processed and analyzed using PMOD 4.2 software (PMOD Technologies Ltd., Zurich, Switzerland). The time − activity curves were deduced from specific volume-ofinterest that were defined based on a mouse MRI T 2 -weighted image template [82]. Radioactivity is presented as standardized uptake value (SUV) (decay-corrected radioactivity per cm 3 divided by the injected dose per gram body weight). [ 18 F] RoSMA-18-d6 SUVR was calculated by using the midbrain in the corresponding hemisphere as reference brain region. For [ 18 F]FDG PET, regional SUV was calculated. Two-way ANOVA with Sidak post hoc analysis was used for comparison between groups (Graphpad Prism 9.0, CA, USA).

Results
Increased expression of inflammation makers and neuronal damage after focal cerebral ischemia in tMCAO mice mRNA levels were measured to address the question whether mouse non-ischemic and ischemic hemispheres differ in their expression levels of CNR2 and other inflammatory genes. CNR2 mRNA expression was increased to around 1.3-fold after 24 h reperfusion and at 48 h in the ipsilateral comparing to contralateral hemisphere (Fig. 1a). Similarly, 1.5-2.5fold increases were observed in the mRNA expression of inflammatory markers including TNF-α, Iba1, MMP9, and GFAP at 24 h and 48 h after reperfusion in the ipsilateral compared to contralateral brain region (Fig. 1b-e). MAP-2 Values represent mean ± standard deviation. Expression levels were quantified by qPCR relative to β-actin. *p < 0.05, ***p < 0.001, ****p < 0.0001. expression has been shown to be a reliable marker of neurons that undergo cell death [83,84]. The neuron-specific MAP-2 expression was markedly reduced in the ipsilateral compared to contralateral hemisphere at 24 h and 48 h after reperfusion (Fig. 1f). As similar CNR2 mRNA expression was observed at 24 h and 48 h, our studies were performed at early time point of 24 h after reperfusion for investigating the functional, structural, and molecular changes in the following experiments.

Reduced Cerebral Glucose Metabolism and Structural MRI Lesion Following tMCAO
Reduced [ 18 F]FDG uptake was observed in the presumed MCA territory of the ipsilateral hemisphere in tMCAO mice, while there was no difference in [ 18 F]FDG uptake between hemispheres in sham-operated mice (Fig. 2a). SUVs were significantly lower in the ipsilateral striatum in tMCAO compared to the contralateral side and compared to the same region in sham-operated mice 1.8 vs 1.4 (Fig. 2b). There were no differences in [ 18 F]FDG uptake in the cortex and cerebellum between the ipsilateral and contralateral hemisphere in tMCAO mice and sham-operated mice. T 2 -weighted MRI and DWI imaging were performed in tMCAO and sham-operated animals at 24 h after reperfusion (Fig. 2c). The lesions in the ipsilateral side in the striatum and cortex were visible as areas of decreased values on the ADC maps calculated from DWI and as areas of increased intensities on the T 2 -weighted MR images at 24 h after reperfusion following 1 h tMCAO (Figs. 2c-d). Ischemic lesions in the tMCAO were also seen ex vivo as white areas, while viable tissue appeared red in TTC stained brain sections (Fig. 2e). Homogenous deep red color was observed across both hemispheres in sham-operated mice, verifying the absence of any lesion. The hemispheric lesion volumes in tMCAO mice were 42.8 ± 10.2 % (mean ± standard deviation).

Increased [ 18 F]RoSMA-18-d6 Retention in the Striatum After tMCAO
To analyze the distribution of [ 18 F]RoSMA-18-d6 in tMCAO mice brain, dynamic µPET/CT scans were performed at 24 h after reperfusion. The standard uptake values (SUVs) of [ 18 F] RoSMA-18-d6 did not reveal significant difference in various brain regions of tMCAO mice ( Supplementary Fig. 1). However, we found a reduced uptake at early time frame (1-3 min) and a similar uptake after 7 min in the ipsilateral side compared to that of the contralateral side (Fig. 3c, and whole head images in Supplementary Fig. 2). Thus, to exclude the perfusion influence, we averaged the brain signals from 21-61 min and selected the midbrain as the reference region. Higher [ 18 F]RoSMA-18-d6 SUVR was observed in the ischemic ipsilateral striatum compared to the contralateral striatum (two-way ANOVA with Sidak multiple comparison correction, 0.97 ± 0.02 vs 0.87 ± 0.06, p = 0.0274), but not in other brain regions such as cortex (Figs. 3d, e). The increased signals at ischemic ipsilateral striatum, however, could not be blocked by the selective CB 2 R agonist GW405833 (Fig. 3e).
At the end of the in vivo experiments, we dissected the mice to verify the activity accumulation and specificity of [ 18 F]RoSMA-18-d6 in the spleen and different brain regions with a gamma counter. In line with the results obtained from the averaged SUVRs in the tMCAO mouse brain, the radioactivity in the ipsilateral side was indeed significantly higher than that of the contralateral hemisphere (0.037 ± 0.007 vs 0.026 ± 0.003, n = 5 each group), but no blockade effect was seen under blocking conditions (Fig. 4a). As expected, radioactivity in the CB 2 R-rich spleen was much higher than the brain and 58 % of the signals was blocked by co-injection of CB 2 R specific ligand GW405833, demonstrating specific target engagement of [ 18 F]RoSMA-18-d6 in vivo (Fig. 4b).

Discussion
This study assessed the utility of CB 2 R PET tracer [ 18 F] RoSMA-18-d6 for imaging tMCAO mouse at subacute stage, concomitant with decreased CMRglc levels and formation of a structural lesion. Previous PET imaging of stroke animal models led to inconclusive results. In a rat model of photothrombotic stroke at 24 h after surgery, increased [ 11 C]NE40 (CB 2 R tracer) uptake and unvaried [ 11 C]PK11195 (TSPO tracer) uptake were reported [85]. In another study, [ 11 C] NE40 uptake did not show any difference in the same rat model of photothrombotic stroke [86]. Moreover, reduced [ 11 C]A836339 (CB 2 R tracer) uptake was reported in a focal tMCAO rat model over 1-28 days after occlusion [56]. Possible reasons for these different observations include the time point of assessment, different methods for inducing acute stroke (transient or permanent ischemia) resulting in variations of ischemic severity and levels of inflammatory cell expression [36]. CB 2 R has negligible expression in the mouse brain and is mainly expressed in the spleen under physiological conditions [39,45,[64][65][66][67][68][69]87]. Under neuroinflammatory conditions, CB 2 R is upregulated in activated microglial cells. In this study, we used quantitative real-time polymerase chain reaction to measure gene expression levels of CNR2, TNFα, Iba1, MMP9, GFAP, and MAP-2 at 24 h and 48 h. All tested inflammatory markers displayed increased mRNA levels in the ipsilateral brain hemisphere, in agreement with the reported findings in tMCAO mouse model [34,38,88,89]. In line with the increased CNR2 gene expression levels, significantly higher [ 18 F]RoSMA-18-d6 SUVR (standard uptake value ratio) was observed in striatum at ipsilateral vs contralateral under baseline conditions in our PET studies. The 50 % reduction of the neuronal marker MAP-2 indicated neuronal damage.
The dynamic µPET scan using [ 18 F]RoSMA-18-d6 indicated a reduced perfusion in the lesion brain regions at the first time frame of 1-3 min. This is probably due to the changes of microvascular response (no-reflow phenomenon) and the reduction in neuronal activity. Taking the midbrain as the reference region, the ratios of SUV averaged from 21 to 61 min revealed increased [ 18 F]RoSMA-18-d6 SUVR in the ipsilateral ischemic striatum compared to that of the contralateral side. Our ex vivo bio-distribution studies confirmed the difference of the radioactivity distribution in the left and right brain hemisphere. The in vivo specificity of [ 18 F]RoSMA-18-d6 towards CB 2 R is evidenced by a 58 % reduction in radioactivity in the mouse spleen under blockade conditions in ex vivo biodistribution studies. Underlying reasons for the lack of specificity of [ 18 F]RoSMA-18-d6 in the mouse brain may include (1) the increased tracer availability in the blood induced by blocking the CB 2 R peripheral targets in the presence of the blocker GW405833; and (2) the relatively low brain uptake of our CB 2 R-selective radioligand [ 18 F] RoSMA-18-d6 in the mouse brain resulted in undetectable changes of radiosignals under baseline and blockade conditions. Notably, the time-activity curves of [ 18 F]RoSMA-18-d6 in tMCAO mouse brain showed remarkably higher initial brain uptake under blockade conditions than the baseline in both sides of the mouse brain ( Supplementary  Fig. 1), indicating the influence of blocking CB 2 R target In the spleen about 58 % of the [ 18 F] RoSMA-18-d6 binding (norm % ID/g tissue) was blocked. No significant blocking was observed in the brain. Data are presented as the mean of the percentage of injected dose per gram tissue normalized to 20 g body weight; mean ± standard deviation. % ID/g: percentage injected dose per gram. *p < 0.05, **p < 0.01.
in the peripheral organs on the availability of radiotracer concentrations in the blood. In our previous studies with Wistar rats, the spleen uptake of [ 18 F]RoSMA-18-d6 was blocked by nearly 90 %. The higher specific binding of [ 18 F] RoSMA-18-d6 in rat spleen is likely due to the increased available targets. Govaerts et al. reported higher binding sites B max in the rat spleen (0.71 ± 0.02 pmol/mg protein) compared to that in the mouse spleen (0.31 ± 0.03 pmol/mg protein) by using [ 3 H]CP55940 binding assay [90]. In the mouse P-glycoprotein assay, compound RoSMA-18 exhibits an efflux ratio (ER) of 4.6, which partly explains the low initial brain uptake in mice. Compared to other CB 2 R radioligand developed in our group, [ 18 F]ROSMA-18-d6 has the highest specificity (86 %) towards CB 2 R based on the ex vivo biodistribution results in the rat spleen and was therefore selected for in vivo studies.
We observed that [ 18 F]FDG measure of CMRglc was reduced in the ischemic areas, i.e., ipsilateral striatum of the tMCAO mice at 24 h after reperfusion. The reduced CMRglc was reported in many earlier studies in disease animal models and in stroke patients, [91][92][93][94], masking CMRglc reduction of neuronal tissue in the brain. At an extended time points of the recovery stage from days 4 to 40, an increased CMRglc level was reported in the ischemic regions due to the increased consumption from inflammatory cells along with microglial activation [91,95,96].
There are several limitations in the current study. (1) As there is no reliable specific CB 2 R antibody, we did not include immunohistochemical staining for CB 2 R protein distribution in the mouse brain. The qPCR measures of CNR2 mRNA level provided an alternative readout, but do not provide spatial distribution of cerebral CB 2 R expression. (2) Due to the logistic barrier, MRI and µPET/ CT scans were performed with different cohorts of animals. Nevertheless, standard operating procedures for the surgery were used. (3) Our in vivo data with tMCAO mice were collected at 24 h after surgery, longitudinal imaging of tMCAO mice with [ 18 F]RoSMA-18-d6 along with structural and functional readout will provide further insight into the spatio-temporal dynamics of CB 2 R expression in the brain. Moreover, [ 18 F]FDG-PET was used as a benchmark for neuroinflammation in the current study. Further study using TSPO tracer to assess the glial activation along with CB 2 R tracer will provide valuable insights.

Conclusion
Our newly developed CB 2 R PET tracer, [ 18 F]RoSMA-18-d6, revealed limited utility to image neuroinflammation in the ischemic ipsilateral of the tMCAO mice at 24 h after reperfusion. Although lesion regions in tMCAO mouse brain could be followed by the ratios of averaged SUVs from 21 to 61 min with midbrain as the reference region, the in vivo specificity of [ 18 F]RoSMA-18-d6 was confirmed only in the CB 2 R-rich spleen. Different neuroinflammatory animal models which have comparable neuronal numbers in the lesion regions are recommended for evaluation of CB 2 R in further PET imaging studies.