Synthesis and In Vitro Characterization of Selective Cannabinoid CB2 Receptor Agonists: Biological Evaluation against Neuroblastoma Cancer Cells

1,8-naphthyridine-3-carboxamide structures were previously identified as a promising scaffold from which to obtain CB2R agonists with anticancer and anti-inflammatory activity. This work describes the synthesis and functional characterization of new 1,8-naphthyridin-2(1H)-one-3-carboxamides with high affinity and selectivity for CB2R. The new compounds were able to pharmacologically modulate the cAMP response without modulating CB2R-dependent β-arrestin2 recruitment. These structures were also evaluated for their anti-cancer activity against SH-SY5Y and SK-N-BE cells. They were able to reduce the cell viability of both neuroblastoma cancer cell lines with micromolar potency (IC50 of FG158a = 11.8 μM and FG160a = 13.2 μM in SH-SY5Y cells) by a CB2R-mediated mechanism. Finally, in SH-SY5Y cells one of the newly synthesized compounds, FG158a, was able to modulate ERK1/2 expression by a CB2R-mediated effect, thus suggesting that this signaling pathway might be involved in its potential anti-cancer effect.


Introduction
Neuroblastoma (NB) is the most common solid extracranial tumor among children characterized by a severe mortality rate [1]. NB begins from immature nerve cells of the sympathetic nervous system and can lead to a wide range of clinical outcomes from spontaneous regression to incurable progression, with resistant therapy and poor prognosis [2,3]. Therefore, it is necessary to find novel therapeutic compounds that can achieve effective and efficient results against neuroblastomas.
Cannabinoid receptor 2 (CB2R), together with cannabinoid receptor 1 (CB1R), belongs to the endocannabinoid system (ECS), a relatively recently discovered physiological system known to play a fundamental role in establishing and maintaining human health [4]. CB1R is abundant in the central nervous system (CNS), where it plays a well-established role in regulating neuronal excitability [5]. In contrast, CB2R is predominantly expressed at the peripheral level and is involved in the anti-inflammatory effects of cannabinoids without causing psychotropic adverse effects, which are mainly associated with the stimulation of CB1R [6]. CB2R continues to be of interest because of its involvement in several pathological conditions including cancers. CB2R is reported to be upregulated in various tumor tissues, such as melanoma [7], bladder cancer [8], breast cancer [9], colon cancer [10], and hepatocellular carcinoma [11]. Many CB2R agonists have exhibited an antitumor effect in several cancers [12,13].
In a recent study reported in the literature, additional therapeutic targets against high-risk neuroblastoma were defined using an integrative data analysis combined with experimental evaluation in cell lines from patient-derived xenografts. In this work, CB2R was identified as one of the most promising targets for the treatment of high-risk neuroblastoma [14].
Molecules 2022, 27, x FOR PEER REVIEW 2 of 13 effects of cannabinoids without causing psychotropic adverse effects, which are mainly associated with the stimulation of CB1R [6]. CB2R continues to be of interest because of its involvement in several pathological conditions including cancers. CB2R is reported to be upregulated in various tumor tissues, such as melanoma [7], bladder cancer [8], breast cancer [9], colon cancer [10], and hepatocellular carcinoma [11]. Many CB2R agonists have exhibited an antitumor effect in several cancers [12,13].
In a recent study reported in the literature, additional therapeutic targets against high-risk neuroblastoma were defined using an integrative data analysis combined with experimental evaluation in cell lines from patient-derived xenografts. In this work, CB2R was identified as one of the most promising targets for the treatment of high-risk neuroblastoma [14].
The present work describes the synthesis of the new 1,8-naphthyridin-2(1H)-one-3carboxamide derivatives FG158a, FG160a, and FG161a ( Figure 2). The new compounds are structurally similar to the selective CB2R agonist N1-hydroxypentyl derivative LV62 previously reported by us [26] but with the hydroxyl group being replaced by a bromine atom (FG158a), a chlorine atom (FG160a), or an azido group (FG161a) (Figure 2). The effectiveness of the novel 1,8-naphthyridine-3-carboxamide derivatives and of compound LV62 were determined against neuroblastoma cells. Several 1,8-Naphthyridine derivatives have gained the attention of researchers for their anticancer properties and SAR studies on their efficacy as antitumor active compounds have been reported [21]. Furthermore, many other heterocyclic compounds such as indole, furano, isoxazole derivatives characterized by a carboxamide substituent have shown interesting anti-proliferative activity against different cancer cell lines [22][23][24][25].

Chemistry
Synthesis of Compounds FG158a, FG160a, and FG161a The synthetic route to obtain the new 1,8-naphthyridin-2(1H)-on-3-carboxamide derivatives FG158a, FG160a, and FG161a is outlined in Scheme 1. N-(4methylcyclohexyl)-1,8-naphthyridin-2(1H)-on-3-carboxamide, prepared as previously reported in the literature [17], was subjected to a N-alkylation reaction by treatment with cesium fluoride in anhydrous DMF at room temperature for 1 h and then with 1,5dibromopentane or 1-bromo-5-chloropentane at 30 °C for 24 h to yield the desired 1,8naphthyridin-2-one derivatives FG158a and FG160a. The synthesis of the azide derivative FG161a included an additional step starting from compound FG158a with sodium azide at 60 °C for 12 h. Each crude mixture was purified by flash chromatography to produce FG158a, FG160a, and FG161a; in addition, the separation of the cis and trans isomers of each compound was achieved.  (Table 1). These results showed that the replacement of the hydroxy group of the hydroxypentyl substituent in position 1 of LV62 by the bromine, chloro, or azido group made it possible to maintain high affinity and selectivity for CB2R. This finding is in accordance with the result obtained previously by replacement of the hydroxy group of LV62 by a fluorine atom [26].

[ 3 H]CP55,940 Binding Assays
The binding affinities (K i values) of the novel derivatives FG158a, FG160a, and FG161a and the compound LV62 [26] as a mixture of cis and trans isomers ( ∼ = 1:1) were evaluated at hCB1R and hCB2R using a [ 3 H]CP55,940 radioligand displacement assay and membranes derived from Chinese hamster ovary (CHO) cells stably expressing either receptor. The results are summarized in Table 1. The non-selective orthosteric CBR ligand CP55,490 was used as a reference compound. At hCB2R, the new compounds FG158a, FG160a, and FG161a fully displaced [ 3 H]CP55,940, proving that they are high-affinity CB2R ligands (K i values from 45 nM to 16.5 nM) analogously to the reference compound LV62 (K i = 37 nM) ( Table 1). At hCB1R, the new compounds FG158a, FG160a, FG161a, and LV62 were not able to displace [ 3 H]CP55,940, indicating no affinity of these ligands for hCB1R, and hence high CB2R selectivity (Table 1). These results showed that the replacement of the hydroxy group of the hydroxypentyl substituent in position 1 of LV62 by the bromine, chloro, or azido group made it possible to maintain high affinity and selectivity for CB2R. This finding is in accordance with the result obtained previously by replacement of the hydroxy group of LV62 by a fluorine atom [26]. to recruit β-arrestin2, since G protein-coupled receptors also interact with β-arrestin, which facilitates receptor internalization, recycling, degradation, and signaling. Interestingly, the trend relating the potency and efficacy values for compounds FG158a, FG160a, and FG161a suggests that these were not able to modulate β-arrestin2 recruitment (EC50 > 10,000 nM) ( Figure 3A; Table 1). In contrast, compound LV62 produced an increase in the β-arrestin2 recruitment ( Figure 3B; Table 1).

CB2R Functional Activity: cAMP and β-Arrestin2 Assays
Following the initial [ 3 H]CP55,940 binding assessment, FG158a, FG160a, FG161a, and LV62 (as a mixture of cis and trans isomers) were evaluated for their functional activity to modulate the Gα i/o -dependent inhibition of forskolin (FSK)-stimulated cAMP accumulation and β-arrestin2 recruitment in CHO cells stably expressing hCB2R. The non-selective orthosteric CBR ligand CP55,490 was used as a reference compound. For the cAMP inhibition assay, cells were treated with 10 µM FSK and synthetized compounds for 90 min to assess compound concentration-dependent activity. The results showed that LV62 displayed the highest potency (34 (3.9-220)) nM and efficacy (110 ± 8.8). (Table 1). FG160a and FG161a were able to modulate the cAMP response but their potency and efficacy were lower than those of the compound LV62 (Table 1). Conversely, compound FG158 proved not to affect this signaling pathway ( Figure 3A; Table 1).

Investigation of Toxicity Levels against the Neuroblastoma Cells
The cytotoxic anticancer activity of CBR agonists has been suggested by several studies [12,13]. The selective CB2R agonist JWH-133 inhibits SH-SY5Y cell proliferation, presumably via a CB2R-independent mechanism [27]. We tested the anti-proliferative activity of the novel CB2R agonists FG158a, FG160a, FG161a, and LV62 in neuroblastoma cell lines (Figure 4). The compounds were assayed as a mixture of cis and trans isomers. We observed that in SH-SY5Y cells after 24 h ( Figure 4A), 48 h ( Figure 4B), and 72 h ( Figure 4C) of treatment, FG158a, FG160a, FG161a, and LV62 inhibited cell viability. After 72 h of treatment, the effect was higher and could also be observed at lower concentrations. The inhibitory effect was confirmed in another neuroblastoma cell line, SK-N-BE cells. After 72 h of treatment, the compounds inhibited cell viability ( Figure 4D).
To address a CB2R-mediated mechanism, we tested FG158a, LV62, and FG160a at inhibitory concentrations in SH-SY5Y cells pre-incubated with the CB2R antagonist SR144528. After 72h of treatment, we observed a reversion in cell viability at 20 and 40 μM of FG158a and FG160a in the presence of SR144528. The inhibitory effect of LV62 was reverted only at 40 μM ( Figure 5). These data suggest that CB2R agonists can arrest neuroblastoma cell viability and that their effect is mediated by CB2R.  To address a CB2R-mediated mechanism, we tested FG158a, LV62, and FG160a at inhibitory concentrations in SH-SY5Y cells pre-incubated with the CB2R antagonist SR144528. After 72 h of treatment, we observed a reversion in cell viability at 20 and 40 µM of FG158a and FG160a in the presence of SR144528. The inhibitory effect of LV62 was reverted only at 40 µM ( Figure 5). These data suggest that CB2R agonists can arrest neuroblastoma cell viability and that their effect is mediated by CB2R.

Effect of FG158a on ERK1/2
Among the CB2R agonists tested, we selected FG158a as it provides the best doseresponse curve in SH-SY5Y cells ( Figure 4B). The effect of CB2R agonists on the ERK pathway has been previously detected in other tumor models such as a breast cancer model [28]. The dysregulation of the ERK/MAPK pathway is crucial in neuroblastoma. In neuroblastoma primary cells, the ERK/MAPK pathway has been observed to be highly activated [29]. We addressed the effect of FG158a on MAPK kinase activation by the evaluation of its component ERK1/2 in the presence and in the absence of the antagonist SR144528 ( Figure 6). We observed that 20 µM FG158a inhibited ERK1/2 expression with respect to the untreated control cells, whereas the antagonist SR144528 reversed the inhibitory effect of the agonist. These data demonstrate the activity of FG158a on ERK1/2 levels and that the effect is mediated by CB2R. The statistical significance was calculated with two-way ANOVA using Dunnett's multiple comparisons test.

Effect of FG158a on ERK1/2
Among the CB2R agonists tested, we selected FG158a as it provides the best doseresponse curve in SH-SY5Y cells ( Figure 4B). The effect of CB2R agonists on the ERK pathway has been previously detected in other tumor models such as a breast cancer model [28]. The dysregulation of the ERK/MAPK pathway is crucial in neuroblastoma. In neuroblastoma primary cells, the ERK/MAPK pathway has been observed to be highly activated  [29]. We addressed the effect of FG158a on MAPK kinase activation by the evaluation of its component ERK1/2 in the presence and in the absence of the antagonist SR144528 (Figure 6). We observed that 20 μM FG158a inhibited ERK1/2 expression with respect to the untreated control cells, whereas the antagonist SR144528 reversed the inhibitory effect of the agonist. These data demonstrate the activity of FG158a on ERK1/2 levels and that the effect is mediated by CB2R.

Chemistry
Commercially available reagents were purchased from Merk Life Science (Milano, Italy) or Fluorochem (Glossop, UK) and used without purification. 1 H NMR and 13 C NMR were recorded at 400 and 100 MHz, respectively, on a Bruker AVANCE IIITM 400

Chemistry
Commercially available reagents were purchased from Merk Life Science (Milano, Italy) or Fluorochem (Glossop, UK) and used without purification. 1 H NMR and 13 C NMR were recorded at 400 and 100 MHz, respectively, on a Bruker AVANCE IIITM 400 spectrometer (Mannheim, Germany). Chemical shift (δ) is reported in parts per million related to the residual solvent signal, while coupling constants (J) are expressed in Hertz (Hz). Organic solutions were dried over anhydrous Na 2 SO 4 . Evaporation was carried out in vacuo using a rotating evaporator. Silica gel flash chromatography was performed using silica gel 60 Å (0.040-0.063 mm; Merck Life Science S.r.l., Milano, Italy). Reactions were monitored by TLC on Kieselgel 60 F254 (Merck Life Science S.r.l., Milano, Italy) with detection by UV light (λ = 254 nm). All reactions involving air-or moisture-sensitive reagents were performed under nitrogen atmosphere using anhydrous solvents. Melting points were determined on a Kofler hot-stage apparatus and are uncorrected. Elemental analyses were performed in our analytical laboratory and agreed with theoretical values to within ± 0.4%.

Radioligand Displacement Assay
We have previously described radioligand displacement assays in detail [30]. Briefly, 0.7 nM [ 3 H]CP55,940, tris binding buffer (50 mM Tris-HCl, 50 mM Tri-base, 0.1% BSA, pH 7.4), and 50 µg hCB1R or hCB2R cell membranes were used to a total assay volume of 500 µL per reaction at 37 • C for 60 min. Binding reactions were terminated by ice-cold tris binding buffer and subsequent vacuum filtration (24-well sampling manifold (Brandel cell harvester; Brandel Inc., Gaithersburg, MD, USA)) using Brandel GF/B filters soaked in wash buffer at 4 • C for at least 24 h. Filters were washed with 1.2 mL of tris-binding buffer six times, then oven-dried for 60 min before being incubated in 3 mL of scintillation fluid (Ultima Gold XR, PerkinElmer, Seer Green, Buckinghamshire, UK). Radioactivity was quantified by liquid scintillation spectrometry. All compounds were first prepared at 10 mM solutions in DMSO. The final vehicle concentration for assays was 0.1% DMSO. All assays were performed in duplicate.

HitHunter ® cAMP Assay
The quantification of FSK-stimulated cAMP accumulation using the DiscoveRx HitHunter assay (DiscoveRx, Eurofins, Fremont, CA, USA) has been described in detail previously elsewhere [30]. Briefly, 20,000 cells/well were plated in low-volume 96-well plates and incubated overnight in Opti-MEM with 1% FBS at 37 • C and 5% CO 2 . The next day, the media were replaced with cell assay buffer (DiscoveRx). Immediately after this, the cells were simultaneously treated with 10 µM FSK and ligands for 90 min. Subsequent additions of cAMP antibody solution and cAMP working detection solutions followed the manufac-turer's directions (DiscoveRx) and cells were stored for 60 min at room temperature. cAMP solution A (DiscoveRx) was added according to the manufacturer's directions followed by another 60 min room temperature incubation prior to chemiluminescence being measured on a Cytation5 plate reader (top read, gain 200, integration time 10,000 ms).

PathHunter ® CB1R β-Arrestin2 Assay
The quantification of β-arrestin2 recruitment using the DiscoveRx PathHunter assay has been described by our group previously and is summarized here. [30]. Twenty thousand cells/well were grown overnight in Opti-MEM containing 1% FBS at 37 • C and 5% CO 2 in low-volume 96 well plates. The next day, cells were treated with compounds for 90 min at 37 • C. Detection solution was added to cells according to the manufacturer's directions (DiscoveRx) and cells were stored for 60 min at room temperature. After this incubation, chemiluminescence was quantified on a Cytation5 plate reader (top read, gain 200, integration time 10,000 ms).

Sulforhodamine B (SRB) Assay
Cells seeded for 24 h in triplicate in 96-well plates at a density of 8000 cells/well were left to adhere overnight. CB2R agonists were added to the culture in triplicate at concentrations ranging from 2.5-40 µM. After 24 h, 48 h, and 72 h, cells were fixed, shaking for 2 h at 4 • C with 50% v/v trichloroacetic acid and washed with distilled water. Cells left overnight to dry were stained with 0.4% w/v SRB in 1% v/v acetic acid (at room temperature for 30 min on a shaker) and then washed with 1% acetic acid to allow for the removal of the unbound dye. TRIS-HCl 10 mM, pH 7.4 was added to the cells and absorbance was measured at 495 nm as previously shown [32,33] using a Glomax ® Discover Microplate Reader (Promega, Madison, WI, USA). In combinatory assays, similar procedures were adopted and the CB2R antagonist SR144528 (1 µM) was added 1 h before the CB2 agonists (10, 20 and 40 µM) to SHSY5Y cells that were harvested as described above after 72 h of incubation.

Statistical Analysis
[ 3 H]CP55,940 radioligand competition binding data are provided as % change from maximal 3 H bound (i.e., 100%). Data for HitHunter cAMP and PathHunter β-arrestin2 data are shown as % of maximal CP55,940 response (i.e., 100%). Estimates of K i , EC 50 , and E max were determined using non-linear regression three-parameters (GraphPad, Prism, v. 9.0). Data are mean with 95% confidence interval (C.I.) (EC 50 ) or mean ± SEM, n = 3-6 indepen-dent experiments performed in duplicate. Statistical analyses were by nonoverlapping C.I. or two-way ANOVA followed by Bonferroni's post hoc test. * p < 0.05 relative to CP55,940 within assay. The details of statistical analyses performed for SRB assays and densitometry analysis is reported in the figure legends.

Conclusions
Beyond the well-documented effects of CB2R agonists in inflammatory and neurodegenerative diseases, CB2R has been recently identified as a novel target in xenograft models of neuroblastoma, suggesting its promising role in high-risk neuroblastoma. Additionally, the anti-proliferative activity in neuroblastoma cell lines of the gold standard CB2 agonist JWH-133 has been reported. Here, we explored the anti-cancer activity of novel CB2R agonists addressing their inhibitory effects in neuroblastoma cell lines. We observed that the arrest of cell viability by these novel compounds was mediated by the CB2R. The ERK/MAPK signaling pathway is known to play a role in neuroblastoma, contributing to the progression toward a pro-tumor cell phenotype, and is also involved in resistance to chemotherapy. In SH-SY5Y cells, we observed that FG158a was able to modulate ERK1/2 expression and that the effect is mediated by CB2R, thus suggesting that the potential anti-cancer effect of this compound might involve this signaling pathway.
The advantage of the CB2R-selective compounds described here might be the documented anti-inflammatory activity of CB2R agonists. It is well known that inflammation can promote cancer and alteration of the tumor microenvironment contributing to tumor progression, inhibition of apoptosis, induction of angiogenesis, and resistance to chemotherapeutics. The crucial role of CB2R in neurodegeneration and in various types of tumors, due to its involvement in the (neuro)-inflammatory processes, makes highly selective CB2R agonists compelling compounds against neuroblastoma.