2-Arylpropionic Acid Pyrazolamides as Cannabinoid CB2 Receptor Inverse Agonists Endowed with Anti-Inflammatory Properties

Among the most recent proposals regarding the mechanism of action of dipyrone, the modulation of cannabinoid receptors CB1 and CB2 appears to be a promising hypothesis. In this context, the present work describes a series of five novel pyrazolamides (7–11) designed as molecular hybrids of dipyrone metabolites and NSAIDs, such as ibuprofen and flurbiprofen. Target compounds were obtained in good overall yields (50–80%) by classical amide coupling between 4-aminoantipyrine and arylacetic or arylpropionic acids, followed in some cases by N-methylation of the amide group. The compounds presented good physicochemical properties in addition to stability to chemical (pH 2 and 7.4) and enzymatic (plasma esterases) hydrolysis and showed medium to high gastrointestinal and BBB permeabilities in the PAMPA assay. When subjected to functional testing on CB1- or CB2-transfected cells, compounds demonstrated an inverse agonist profile on CB2 receptors and the further characterization of compound LASSBio-2265 (11) revealed moderate binding affinity to CB2 receptor (Ki = 16 µM) with an EC50 = 0.36 µM (Emax = 63%). LASSBio-2265 (11) (at 1, 3, and 10 mg/kg p.o.) was investigated in the formalin test in mice and a remarkable analgesic activity in the late inflammatory phase was observed, suggesting it could be promising for the treatment of pain syndromes associated with chronic inflammatory diseases.


Introduction
The modulation of the endocannabinoid system has received great focus among therapeutic possibilities for treating pain processes [1]. In vivo studies have demonstrated the influence of cannabinoid receptors in mediating nociception [2][3][4][5][6][7][8][9][10]. However, the promotion of central effects attributed to CB 1 receptor activation, such as hypoactivity, hypothermia, and catalepsy [11,12] which can translate into psychoactivity in humans remains of concern. The use of selective CB 2 agonists has emerged as alternative, as these compounds attenuate neuropathic pain without the characteristic side effects of CB 1 activation [13].
The importance of CB 2 receptors in the modulation of nociception is related to their expression in cells of both the immune and the peripheral nervous system and the relationship of endocannabinoids with decreased inflammation has been demonstrated in several inflammatory pain models [14][15][16][17]. Activation of the endocannabinoid system has also been associated with decreased leukocyte recruitment and increased production of anti-inflammatory cytokines [18]. relationship of endocannabinoids with decreased inflammation has been demonstrated in several inflammatory pain models [14][15][16][17]. Activation of the endocannabinoid system has also been associated with decreased leukocyte recruitment and increased production of anti-inflammatory cytokines [18].
Dipyrone (1) is an analgesic and antipyretic drug with spasmolytic action and is widely used in Brazil, Latin America, Germany, Spain, and some African countries [19]. In vivo studies suggest its analgesic effect depends on the activation of cannabinoid receptors [20][21][22], as selective CB1 antagonists but not CB2 antagonists [23] block the dipyrone-associated anti-hyperalgesia. It is suggested that activation of cannabinoid receptors results from the action of arachidonic acid (4)-conjugated dipyrone metabolites, as also proposed for paracetamol [20]. Two major pyrazolamine metabolites are detected after administration of dipyrone to humans: 4-methylaminoantipyrine (2), formed after chemical hydrolysis, and 4-aminoantipyrine (3), as product of cytochrome P450 (CYP3A4) enzymatic N-demethylation ( Figure 1) [21]. The biosynthesis of arachidonylpyrazolamide metabolites could result from the action of fatty acid amide hydrolase (FAAH) [22] and these metabolites, henceforward termed (5) and (6), have demonstrated micromolar affinity for CB1 and CB2 receptors, although their intrinsic efficacy profiles have not yet been characterized [20].  The present work describes new structural analogs of arachidonylpyrazolamides, (5) and (6), which exploit the replacement of the arachidonate subunit (A, Figure 2) for a substituted arylacetic or arylpropionic subunit present in some nonsteroidal anti-inflammatory drugs (NSAIDs), especially those belonging to the profen class, due to their well-known relationship with the metabolism of endocannabinoids [24,25]. The present work describes new structural analogs of arachidonylpyrazolamides, (5) and (6), which exploit the replacement of the arachidonate subunit (A, Figure 2) for a substituted arylacetic or arylpropionic subunit present in some nonsteroidal antiinflammatory drugs (NSAIDs), especially those belonging to the profen class, due to their well-known relationship with the metabolism of endocannabinoids [24,25]. The use of arylacetic and arylpropionic NSAIDs in the design of the new compounds (7)(8)(9)(10)(11)(12) aimed to mimic the hydrophobic interactions of the alkenyl subunit of arachidonic acid (4) with CB1 and CB2, as previously identified in cyclo-oxygenase (COX) enzymes [26]. Therefore, ibuprofen (13) and flurbiprofen (14) were selected as representative NSAIDs for the arylpropionic (profen) class and felbinac (15), an arylacetic derivative structurally related to (14), in order to investigate the influence of the methyl group in the benzylic position. Once synthesized, the target compounds were evaluated regarding their in vitro pharmacokinetic profiles and then characterized for their in vitro intrinsic efficacy. Moreover, the best compound was submitted to a classical in vivo model to confirm its potential as an analgesic drug candidate that acts as a modulator of cannabinoid receptors for the treatment of acute and chronic pain states. The use of arylacetic and arylpropionic NSAIDs in the design of the new compounds (7)(8)(9)(10)(11)(12) aimed to mimic the hydrophobic interactions of the alkenyl subunit of arachidonic acid (4) with CB 1 and CB 2 , as previously identified in cyclo-oxygenase (COX) enzymes [26]. Therefore, ibuprofen (13) and flurbiprofen (14) were selected as representative NSAIDs for the arylpropionic (profen) class and felbinac (15), an arylacetic derivative structurally related to (14), in order to investigate the influence of the methyl group in the benzylic position. Once synthesized, the target compounds were evaluated regarding their in vitro pharmacokinetic profiles and then characterized for their in vitro intrinsic efficacy. Moreover, the best compound was submitted to a classical in vivo model to confirm its potential as an analgesic drug candidate that acts as a modulator of cannabinoid receptors for the treatment of acute and chronic pain states.
Next, pyrazolamides (7-9) were submitted to an N-methylation step after treatment with sodium hydride in THF [28], followed by the addition of methyl iodide, as illustrated in Scheme 2. By using these conditions, pyrazolamides 10 and 11 derived from ibuprofen and flurbiprofen, respectively, were prepared in 75% and 79% yields. On the other hand, the expected N-methyl carboxamide (12) derived from 9 could not be obtained using this methodology, and instead, the bismethylated derivative (16) was formed in a 56% yield (Scheme 2). This behavior was associated to the similar pKa values between the amide and the α-carbonyl hydrogens of 9 [29,30], favoring deprotonation on both sites and bisalkylation to provide compound 16. Pyrazolamides 7 and 8 were not susceptible to alkylation at the benzylic position due to the steric effect produced by the methyl group originally present. Scheme 1. Synthesis of unsubstituted NSAID-pyrazolamide derivatives (7-9).
Next, pyrazolamides (7-9) were submitted to an N-methylation step after treatment with sodium hydride in THF [28], followed by the addition of methyl iodide, as illustrated in Scheme 2. By using these conditions, pyrazolamides 10 and 11 derived from ibuprofen and flurbiprofen, respectively, were prepared in 75% and 79% yields. On the other hand, the expected N-methyl carboxamide (12) derived from 9 could not be obtained using this methodology, and instead, the bismethylated derivative (16) was formed in a 56% yield (Scheme 2). This behavior was associated to the similar pKa values between the amide and the α-carbonyl hydrogens of 9 [29,30], favoring deprotonation on both sites and bisalkylation to provide compound 16. Pyrazolamides 7 and 8 were not susceptible to alkylation at the benzylic position due to the steric effect produced by the methyl group originally present.
Attempts to circumvent this problem by changing the base to potassium carbonate in acetone were not successful, and the only product obtained after 48 h was pyrazolamide 16 in low yields. In addition, the use of another strategy that foresaw the initial methylation of Boc-protected 3, followed by deprotection and coupling with 15, was also unsuccessful due to the low reactivity of N-Boc-pyrazolamide when subjected to alkylation (see Supplementary Material). Therefore, considering that the main motivation for the synthesis of 12 was to evaluate how the absence of the methyl group in the benzylic position could influence the modulation of cannabinoid receptors, we decided to continue the investigation with only the five initially planned pyrazolamides (7)(8)(9)(10)(11).
Compounds 7-11 were fully characterized using spectroscopic techniques (see Supplementary Material), and their purity was determined by reverse-phase HPLC analysis and high-resolution mass spectrometry to be greater than 95%, which was considered adequate for the next step of investigating their drug-like properties and biological actions on cannabinoid receptors. Attempts to circumvent this problem by changing the base to potassium carbonate in acetone were not successful, and the only product obtained after 48 h was pyrazolamide 16 in low yields. In addition, the use of another strategy that foresaw the initial methylation of Boc-protected 3, followed by deprotection and coupling with 15, was also unsuccessful due to the low reactivity of N-Boc-pyrazolamide when subjected to alkylation (see Supplementary Material). Therefore, considering that the main motivation for the synthesis of 12 was to evaluate how the absence of the methyl group in the benzylic position could influence the modulation of cannabinoid receptors, we decided to continue the investigation with only the five initially planned pyrazolamides (7)(8)(9)(10)(11).
Compounds 7-11 were fully characterized using spectroscopic techniques (see Supplementary Material), and their purity was determined by reverse-phase HPLC analysis and high-resolution mass spectrometry to be greater than 95%, which was considered adequate for the next step of investigating their drug-like properties and biological actions on cannabinoid receptors.

Drug-like Properties of Pyrazolamides 7-11
After full characterization of compounds 7-11, we proceeded to determine their physicochemical and in vitro pharmacokinetic properties, given their strategic relevance for developing novel orally available drug candidates.

Drug-like Properties of Pyrazolamides 7-11
After full characterization of compounds 7-11, we proceeded to determine their physicochemical and in vitro pharmacokinetic properties, given their strategic relevance for developing novel orally available drug candidates.
Although the results suggested elevated absorption of 7-11, clearance rates are also of concern for ideal bioavailability. Therefore, we also investigated the stability of 7-11 against enzymatic hydrolysis, as amide groups are potentially labile to plasma carboxylesterases [39] (Table 1). Flurbiprofen-derived pyrazolamides (8 and 11) showed increased stability against enzyme hydrolysis than corresponding ibuprofen-derived pyrazolamides (7 and 10), possibly by the steric effect of the bulkier and less flexible biphenyl system compared to 4-isobutylphenyl in the latter. The loss of ring coplanarity also contributes to the greater plasma stability of both 8 and 11, as the absence of the ortho-effect in 9 favors its metabolism, as shown in Table 1.
Furthermore, considering that the compounds developed in the present work have amide functional groups, the stabilities of the compounds were evaluated at gastric (pH 2) ( Figure 3A) and plasmatic pH (7.4) ( Figure 3B) to assess their susceptibility to pH-dependent chemical hydrolysis. After 240 min, derivatives 7-11 showed recovery rates of 70-100%, indicating the high chemical stability of these compounds at both pH values [40]. As expected, tertiary pyrazolamides 10 and 11 proved to be more resistant to hydrolysis than the corresponding secondary amides. (a) Thermodynamic solubility determined by UV spectroscopy; (b) Determined by UV method after n-octanol/buffer partitioning at pH 7.4; (c) Soybean L-α-phosphatidylcholine was used as artificial lipid membrane in the wells of the donor plate; (d) Porcine brain lipid was used as artificial lipid membrane in the wells of the donor plate; (e) Determined after 240 min.
The membrane permeability of 7-11 measured using a gastrointestinal tract (GI) parallel artificial membrane permeability assay [35] (PAMPA) ranged from 2.5-15.3 ×10 −6 cm/s, which indicated excellent absorption rates (>60%) after oral administration of these substances (Table 1). In addition, considering that cannabinoid receptors are also expressed in the central nervous system [36], the PAMPA-BBB assay was also performed to determine the potential of these compounds to cross the blood-brain barrier [37,38] (Table  1). As expected, all the pyrazolamides had a permeability rate compatible with accessing the CNS (CNS+).
Although the results suggested elevated absorption of 7-11, clearance rates are also of concern for ideal bioavailability. Therefore, we also investigated the stability of 7-11 against enzymatic hydrolysis, as amide groups are potentially labile to plasma carboxylesterases [39] (Table 1). Flurbiprofen-derived pyrazolamides (8 and 11) showed increased stability against enzyme hydrolysis than corresponding ibuprofen-derived pyrazolamides (7 and 10), possibly by the steric effect of the bulkier and less flexible biphenyl system compared to 4-isobutylphenyl in the latter. The loss of ring coplanarity also contributes to the greater plasma stability of both 8 and 11, as the absence of the ortho-effect in 9 favors its metabolism, as shown in Table 1.
Furthermore, considering that the compounds developed in the present work have amide functional groups, the stabilities of the compounds were evaluated at gastric (pH 2) ( Figure 3A) and plasmatic pH (7.4) ( Figure 3B) to assess their susceptibility to pH-dependent chemical hydrolysis. After 240 min, derivatives 7-11 showed recovery rates of 70-100%, indicating the high chemical stability of these compounds at both pH values [40]. As expected, tertiary pyrazolamides 10 and 11 proved to be more resistant to hydrolysis than the corresponding secondary amides. Therefore, both absorption rates and chemical and plasma stability of 7-11 supported the possibility of their use by oral route, with flurbiprofen-derived pyrazolamides 8 and 11 presenting the best physicochemical (solubility, lipophilicity) and in vitro pharmacokinetic profile (permeability and stability) in comparison to the other analogs evaluated.

In Vitro Activity Profiles of Pyrazolamides 7-11
The agonist profile of pyrazolamides 7-11 towards cannabinoid receptors was investigated by measurement of cyclic AMP (cAMP) concentrations in CHO cells transfected with either human CB 1 or CB 2 . The assay is based on receptor coupling to G proteins of the Gi/o family, whose activation results in adenylate cyclase inhibition and a decrease in intracellular cAMP [41][42][43]. When evaluated in CB 1 -expressing cells, compounds 7-11 did not significantly altered the intracellular cAMP at either 1 µM or 10 µM [44], as the reduction in cAMP concentration did not exceed 25% of the maximum response obtained . Results indicate a concentration-dependent weak response and, although no structure-activity relationship was evidenced, these findings demonstrate that introduction of the methyl group in 10 and 11 did not result in a significant change in the activity profile of these compounds against CB 1 receptors compared to 7 and 8, respectively.
The agonist profile of pyrazolamides 7-11 towards cannabinoid receptors was investigated by measurement of cyclic AMP (cAMP) concentrations in CHO cells transfected with either human CB1 or CB2. The assay is based on receptor coupling to G proteins of the Gi/o family, whose activation results in adenylate cyclase inhibition and a decrease in intracellular cAMP [41][42][43]. When evaluated in CB1-expressing cells, compounds 7-11 did not significantly altered the intracellular cAMP at either 1 µM or 10 µM [44], as the reduction in cAMP concentration did not exceed 25% of the maximum response obtained with control agonist CP-55,940 ( Figure 4 and Supplementary material). Results indicate a concentration-dependent weak response and, although no structure-activity relationship was evidenced, these findings demonstrate that introduction of the methyl group in 10 and 11 did not result in a significant change in the activity profile of these compounds against CB1 receptors compared to 7 and 8, respectively. On another hand, when evaluated in cells expressing human CB2 receptors, compounds 7-11 showed a significant increase in cAMP concentrations when tested at 10 µM [44] (Figure 5). This result was in marked contrast with the response to 100 nM WIN 55212-2, which completely abolished the production of cAMP by forskolin, indicating a concentration-dependent inverse agonist profile for 7-11. Additionally, LASSBio-2265 (11) showed an important effect starting from 1 µM, while no effect was observed in CB1-expressing cells. On another hand, when evaluated in cells expressing human CB 2 receptors, compounds 7-11 showed a significant increase in cAMP concentrations when tested at 10 µM [44] ( Figure 5). This result was in marked contrast with the response to 100 nM WIN 55212-2, which completely abolished the production of cAMP by forskolin, indicating a concentrationdependent inverse agonist profile for 7-11. Additionally, LASSBio-2265 (11) showed an important effect starting from 1 µM, while no effect was observed in CB 1 -expressing cells. Considering the ensemble of results, LASSBio-2265 (11) was selected for further investigation to confirm the hypothesis of CB2 inverse agonism. A concentration-response curve for the cAMP increase induced by compound 11 was assessed using chemiluminescent detection in CHO-K1 CNR2 Gi cells stimulated with 8 µM (EC20) forskolin, validated against full agonist CP 55,940 and inverse agonist SR 144,528 [45]. The functional status of the CB2 receptor was monitored by cAMP quantification in relation to controls and it was Considering the ensemble of results, LASSBio-2265 (11) was selected for further investigation to confirm the hypothesis of CB 2 inverse agonism. A concentration-response curve for the cAMP increase induced by compound 11 was assessed using chemiluminescent  [45]. The functional status of the CB 2 receptor was monitored by cAMP quantification in relation to controls and it was possible to observe an increase in cAMP by DiscoverX HitHunter cAMP XS+ assay. The obtained data confirmed that LASSBio-2265 (11) is a partial inverse agonist of human CB 2 receptors, with an EC 50 of 0.369 ± 0.03 µM and an efficacy of 63% of the maximum effect.
Moreover, the binding affinity of LASSBio-2265 (11) to human CB 2 receptors was also studied to further confirm if the inverse agonist activity of LASSBio-2265 (11) was indeed product of an interaction with CB 2 receptors [46]. Competitive binding was evaluated in membranes from CB 2 -expressing CHO cells using [ 3 H]-WIN55212-2 as radioligand probe. The results indicated that 11 shows a moderate affinity for CB 2 receptors, with Ki = 16.0 ± 3 µM. However, the affinity obtained for compound 11 is of similar magnitude to those found by Rogosch and coworkers for compounds 5 and 6, as illustrated in Table 2. Potency and affinity values were determined from in vitro assays performed in duplicate. Table 2. CB 2 receptor binding affinity of LASSBio-2265 (11) and arachidonyl-pyrazolamides (5) and (6).

In Vivo Activity of LASSBio-2265 (11) in the Formalin-Induced Licking Response in Mice
Selective inverse agonists of CB 2 receptors, JTE-907 [47,48], SMM-189 [49], and Sch.414319 [50], have been indicated in the literature as beneficial agents for the treatment of diverse pathological conditions associated with inflammatory processes. Cascio and co-workers [51] described the antinociceptive activity of CB 2 receptor inverse agonists in the late phase of the classical formalin test, which induces a biphasic stereotypical nocifensive behaviour. Therefore, the analgesic profile of LASSBio-2265 (11) was evaluated in the same model ( Figure 6), which measures nociceptive behavior after subcutaneous (s.c.) injection of dilute formalin (1.25% in saline, 30 µL) into one of the two hind paws of mice [52]. This assay presents two temporally distinct phases. The first (early) phase evokes nociception through the direct effect of formalin on the acute activation of pain-sensing C fibers at the peripheral endings of sensory neurons involved in pain transmission. After an interval of 10-15 min, a second, inflammatory-driven phase (late phase) of sustained pain behavior appears, in which sensory fiber activity is accompanied by inflammation and central sensitization [53].
In the first phase, only morphine (2.5 mg/kg) promoted significant analgesia compared to vehicle (5% DMSO in 0.9% NaCl), acetylsalicylic acid (ASA, 200 mg/kg) and LASSBio-2265 (11) at 1, 3 and 10 mg/kg. In contrast, during the second phase, LASSBio-2265 (11) proved to be comparable to both morphine and ASA, with dose-dependent progressive effect, thus proving to be promising for the treatment of conditions that lead to inflammatory pain. Naïve animals were evaluated in two phases of assay without any treatment ( Figure 6). LASSBio-2265 (11) at 1, 3 and 10 mg/kg. In contrast, during the second phase, LASSBio-2265 (11) proved to be comparable to both morphine and ASA, with dose-dependent progressive effect, thus proving to be promising for the treatment of conditions that lead to inflammatory pain. Naïve animals were evaluated in two phases of assay without any treatment ( Figure 6). . The results are presented as the mean ± SD (n = 7 mice per group). Statistical significance was calculated by one-way ANOVA followed by Tukey's test. * p < 0.05 compared with vehicle-treated mice.

Materials
Commercial reagents were obtained from Sigma-Aldrich (St. Louis, MI, USA). Melting points were determined by differential scanning calorimetry with a Shimadzu calorimeter (Model DSC-60). Fourier-transformed infrared (FTIR) spectra were obtained with ThermoScientific spectrometer (Model Nicolet iS10). 1 H-and 13 C-NMR spectra were obtained with VARIAN 400-MR and 500-MR spectrometers. High-performance liquid chro- . The results are presented as the mean ± SD (n = 7 mice per group). Statistical significance was calculated by one-way ANOVA followed by Tukey's test. * p < 0.05 compared with vehicle-treated mice.

Materials
Commercial reagents were obtained from Sigma-Aldrich (St. Louis, MI, USA). Melting points were determined by differential scanning calorimetry with a Shimadzu calorimeter (Model DSC-60). Fourier-transformed infrared (FTIR) spectra were obtained with Thermo-Scientific spectrometer (Model Nicolet iS10). 1 H-and 13 C-NMR spectra were obtained with VARIAN 400-MR and 500-MR spectrometers. High-performance liquid chromatography with detection by photodiode arrays (HPLC-PDA) was performed with Shimadzu LC20AD apparatus using a Kromasil 100-5C18 column (4.6 mm × 250 mm) and SPD-M20A detector (Diode Variety). Ultraviolet-visible measurements were performed in a Femto scanning spectrophotometer (Model 800XI). High-resolution mass spectrometry (Orbitrap-HRMS) analysis was performed using a QExactive Hybrid Quadrupole Orbitrap Mass Spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) with electrospray ionization (ESI) using solutions of the compounds (1 µg/mL) prepared in a 7:3 ratio of water:methanol fortified with 0.1% formic acid and 5 mM ammonium formate.

Aqueous Solubility
The test was performed based on the absorbance obtained by ultraviolet spectroscopy. The scanning wavelength was determined according to the highest absorbance observed for each compound. Saturated aqueous solution of the analyzed compounds were stirred for 4 h at 37 • C, and then filtered through a 0.45 µm PVDF filter and transferred to a quartz cuvette (10 mm optical path) for reading. Solubility was determined by linear regression based on the equation of the line [31].

Determination of the Distribution Coefficient at pH 7.4 (log D 7.4 )
The assay was performed based on the shake-flask method and the absorbance readings by ultraviolet spectroscopy at maximum absorbance wavelength. To a Falcon tube with a capacity of 15 mL, 5 mL of buffer solution at pH 7.4 and 5 mL of 1-octanol were added. Then, the tested compounds were added at defined concentrations of approximately 10-15 µM, and the mixtures were homogenized via vigorous manual shaking. Next, the Falcon tubes were stirred for 4 h at 37 • C, and the experiment was carried out in triplicate for each compound. Subsequently, each sample was placed in a separatory funnel, and the collected aqueous phase was filtered through a 0.45 µm PVDF filter and transferred to a quartz cuvette (10 mm optical path) for UV/Vis reading. The distribution coefficient of each compound at pH 7.4 was determined by linear regression using the straight-line equation [33].

Chemical Stability
To a 2 mL microtube, 2 µL of a 5 mM concentrated solution of the analyzed compound (solubilized in DMSO) and 248 µL of acid buffer (0.2 M potassium chloride and 0.2 M HCl; pH = 2) or neutral (dibasic phosphate; pH = 7.4) were added. The experiment was carried out in triplicate. After vortexing, the mixture was placed in a water bath at 37 • C under vigorous stirring for 0, 30, 60, 120, and 240 min. After each reaction time, 248 µL of basic buffer (phosphate buffer, pH = 7.4) was added to neutralize the pH of the medium in the experiments using acidic buffer (pH = 2). Extraction of the compound was carried out by adding 1 mL of acetonitrile. The entire contents of the microtube were vortexed vigorously, and the container was placed in a freezer (−20 • C) to freeze the aqueous phase. Then, the organic phase was separated, filtered through an HPLC filter with a 0.45 µm PVDF membrane of 25 mm in diameter, and analyzed by HPLC-PDA (mobile phase: acetonitrile:water 60:40). The injection volume of the samples was 20 µL at a flow rate of 1.0 mL/min. Quantitation was performed at 272 nm for compounds (7), (8), and (10), 247 nm for compound (8) and LASSBio-2265 (11), and 253 nm for compound (9)

Agonist Activity toward CB1 and CB2 Receptors
The cellular functional assay used to evaluate the agonist activity of planned pyrazolamides (7)(8)(9)(10)(11)  and coworkers [44]. Briefly, pyrazolamides 7-11 were assayed at concentrations of 1 µM and 10 µM to evaluate their ability to change intracellular cAMP in CHO cells transfected with human CB 1 and CB 2 receptors [44], for 30 or 10 min at 37 • C after stimulation with 10 µM forskolin, respectively. Negative and blank controls were performed in triplicate with and without the addition of forskolin, respectively. Positive controls used were CB agonists CP-55,940 (at 1 nM) and WIN 55,212-2 (at 100 nM).
3.4.2. Inverse Agonist Activity toward CB2 Receptors [45] The cellular functional assay used to evaluate the inverse agonist activity of LASSBio-2265 (11) toward the cannabinoid CB 2 receptor was carried out at Eurofins DiscoveryX Corporation USA (https://www.discoverx.com/home) under the study number FR095-0025341-Q (accessed on 02 June 2021) using the GPCR Biosensor Assay. Cell line cAMP Hunter CHO-K1 CNR2 Gi was seeded at a total volume of 20 µL into white-walled, 384-well microplates and incubated at 37 • C prior to testing in the presence of 8 µM (EC 20 ) forskolin. The media was aspirated from the cells and replaced with 15 µL a 2:1 solution of Hanks' balanced salt solution (HBSS)/10 mM Hepes:cAMP XS+ Ab reagent. Intermediate dilution of the sample stocks was performed to generate 4× samples in assay buffer containing 32 µM forskolin. Then, 5 µL of each 4× sample was added to the cells and incubated at 37 • C for 60 min. The final assay vehicle concentration was 1%. After incubation, the assay signal was generated by incubation with 20 µL cAMP XS+ ED/CL lysis cocktail for 1 h followed by incubation with 20 µL cAMP XS+ EA reagent for 3 h at room temperature. After signal generation, the microplates were read with a PerkinElmer Envision instrument for chemiluminescent signal detection. Compound activity was analyzed using the CBIS data analysis suite (ChemInnovation, San Diego, CA, USA). Percent activity was calculated using the following formula: % Inverse Agonist Activity = 100% × ((mean RLU of test sample − mean RLU of EC 20 forskolin)/(mean RLU of forskolin positive control − mean RLU of EC 20 control)).
Data obtained from duplicate experiments were fit to concentration-response curves by non-linear regression of the Hill equation. Curve-fit parameters and standard errors were calculated using GraphPad Prism 6.0 (Hercules, CA, USA).

Human CB2 (Agonist Radioligand) Receptor Binding Assay
Assays of binding to CB 2 receptors were performed at Eurofins Discovery France (https://www.eurofinsdiscoveryservices.com/) under the study number 100057099 (accessed on 3 June 2021) using the protocol previously published by Munro, Thomas, and Abu-Shaar [46]. Human recombinant CB 2 receptors were expressed in CHO cells, and binding was performed with [ 3 H]-WIN 55212-2, which is a nonselective agonist radioligand. The analysis was performed using software developed at Cerep (Hill software, Santa Barbara, CA, USA) and SigmaPlot 4.0 for Windows ( © 1997 by SPSS Inc., Chicago, IL, USA).
Data obtained from duplicate experiments were fit to concentration-response curves by non-linear regression of the Hill equation. Curve-fit parameters and standard errors were calculated using GraphPad Prism 6.0 (Hercules, CA, USA).

Animals
Swiss Webster mice (20-25 g, 30-40 days old), from both sexes were donated by the Instituto Vital Brazil (Niteroi, Rio de Janeiro, Brazil). The mice were maintained in a room with a 12 h light-dark cycle at 22 ± 2 • C and 60% to 80% humidity, with food and water provided ad libitum. The mice were acclimatized to the laboratory conditions for at least 1 h before each test onset and were used only once throughout the experiments. All protocols were conducted in accordance with the Guidelines on Ethical Standards for Investigation of Experimental Pain in Mice (Zimmermann, 1983) and followed the principles and guidelines adopted by the National Council for the Control of Animal Experimentation (CONCEA) approved by the Ethical Committee for Animal Research (CEUA, approval in 30 April 2019, receiving the number 31/19). All experimental protocols were performed during the light phase. The animal numbers per group were kept to a minimum, and at the end of each experiment, the mice were euthanized by ketamine/xylazine overdose.

Formalin-Induced Licking Behavior
This assay was performed as described by Hunskaar et al. [53] and adapted by Gomes et al. [58]. This model was characterized by a response that occurred in two phases. The first phase (acute neurogenic pain) occurred during the first 5 min after the intraplantar injection of formalin, and the second phase (inflammatory pain) occurred 15 to 30 min postinjection. Mice were pretreated with oral doses of LASSBio-2265 (11, at 1, 3 or 10 mg/kg, p.o.), morphine (2.5 mg/kg, i.p.), acetylsalicylic acid (ASA, 200 mg/kg, p.o.), or the vehicle (same amount of DMSO used to solubilize the substances + NaCl 0.9%) 60 min before the administration of formalin. The naïve group was composed of animals that did not receive any oral treatment. Then, mice (n = 7 per group) received 20 µL of formalin (2.5% v/v) in the dorsal surface of the left hind paw. The time that the mice spent licking the injected paw was immediately recorded. The amount of DMSO used did not affect or interfere per se with the experimental model. The final volume administered to mice was 0.1 mL. The protocol was carried out by a blinded experimenter that did not know the drug treatment conditions and groups.

Conclusions
NSAID-pyrazolamides 7-11 were obtained in good yields from classical synthetic methodologies. All compounds presented adequate aqueous solubility, lipophilicity, permeability, and stability to chemical and enzymatic hydrolysis for the development of orally administered drug candidates. Compounds 7-11 also showed high permeability through artificial membranes mimicking the blood-brain barrier, as cannabinoid receptors are widely distributed in the central nervous system. Their stability against carboxylesterases also reinforces that observed effects are not resulting from the effect of precursors, i.e., 4-aminoantipyrine and NSAID. The chemical stability of pyrazolamides 7-11 indicated that the compounds are highly stable and are not susceptible to pH-dependent hydrolysis, unlike dipyrone.
The in vitro experiments performed for characterization of the intrinsic efficacy of LASSBio-2265 (11) in CB 2 receptors showed a moderate ligand profile (Ki = 16 µM) with EC 50 = 0.36 µM (Emax = 63%) in CHO cells. For in vivo studies, antinociceptive effects of 11 were demonstrated by decreased pain behaviour in the formalin test in mice at 3 mg/kg, suggesting that it could be a promising candidate for the treatment of pain syndromes associated with chronic inflammatory diseases.  Table S1: Absorbed fraction of drugs used as standard and pyrazolamides (7)(8)(9)(10)(11) determined in the PAMPA-GTI model; Table S2: Permeability of drugs used as standard, and pyrazolamides (7)(8)(9)(10)(11) determined in the PAMPA-BBB model; Table S3: Evaluation the agonist effect against the CB1 receptors of the pyrazolamides (7-11); Table S4: Evaluation the agonist effect against