The cannabinoid-induced tetrad is a preclinical model to test whether compounds produce cannabinoid-like behavioral effects, suggesting that they might be CB1 agonists. However, to definitively establish CB1 receptor mediation, these behavioral findings must be corroborated with binding assays and functional assays. This model assesses four specific phenotypic effects of CB1 agonists: hypolocomotion, hypothermia, analgesia and catalepsy [34]. In this study, we slightly modified the classical cannabinoid tetrad assay, originally developed by B.R. Martin [35] to evaluate three indazole-carboxamide SCs, namely 5F-ADB, MDMB-4en-PINACA and ADB-4en-PINACA with Δ⁹-THC as a positive control [34]. The results showed that all three SCs induced the typical tetrad effects. They significantly decreased locomotion, body temperature and pain sensitivity, and impaired the capacity to initiate movement in mice. The potency ranking was 5F-ADB > MDMB-4en-PINACA > ADB-4en-PINACA > Δ⁹-THC, with approximately a 375-, 107- and 25-fold higher potency than Δ⁹-THC, respectively, which is consistent with previous reports [14,26,27]. The results highlight the unpredictable effects and potential for severe adverse reactions. Notably, 5F-ADB was the most potent, likely due to its 5-fluoropentyl tail enhancing blood–brain barrier penetration and CB1 binding affinity, as supported by prior work on fluorinated psychoactives [36].

Motor impairment is one of the main behavioral effects observed after the systemic administration of cannabinoid receptor agonists, posing a significant threat to physical health and public safety [37]. Our study demonstrated that MDMB-4en-PINACA (0.01 mg/kg and 0.03 mg/kg) and Δ⁹-THC (1 mg/kg) significantly increased the locomotor activity, and was significantly inhibited as the doses increased, which demonstrated the biphasic modulation effect of locomotor activity in mice. We confirmed the previous findings of the dose-dependent motor stimulating and inhibiting effects of the cannabinoids receptor agonist [9,38]. But, this result was inconsistent with the recent literature which found that the administration of MDMB-4en-PINACA elicited an immediate hypolocomotive response [31]. The data are also inconsistent with earlier reports showing that 5F-ADB produced short-acting locomotor depression (ED₅₀ = 1.1 mg/kg) without prior stimulation [17]. This discrepancy may reflect differences in dosing regimens or animal models.

We observed notable differences in the relative potency of tetrad effects across the tested compounds. Δ9-THC and ADB-4en-PINACA were 3–5 times and 2–3 times more potent in suppressing locomotion than in producing hypothermia, analgesia or catalepsy. In contrast, the potencies of hypothermia induced by MDMB-4en-PINACA and 5F-ADB were 3–4 times and 2 times lower, respectively, than those of locomotion suppression, analgesia and catalepsy. 5F-ADB exhibited the lowest hypothermia potency among all tetrad effects. This might be related to differences in the key brain regions and mechanisms of tetrad effects induced by these drugs, as well as differences in chemical structures. For instance, subtle variations in side chain modifications or core scaffolds may alter receptor binding kinetics or regional brain distribution, leading to the selective amplification or dampening of specific effects. Further studies mapping CNS activation patterns could clarify these mechanistic distinctions.

Case reports have outlined that the three SCs may possess high abuse potential [21,32,39], but there is rare direct evidence supporting this speculation, especially in vivo data. In the present study, we systematically evaluated the abuse potential of the three SCs using both the CPP paradigm and the precipitated withdrawal paradigm.

Place conditioning is a classic assay to evaluate the rewarding effect in vivo, which has revealed the psychological dependence of drugs. Many compounds such as NM-2201, SDB-005 and Δ⁹-THC have shown a biphasic effect, which could induce CPP at low doses and induce CPA at high doses [40,41,42,43]. In the present study, we concluded that the three SCs and Δ⁹-THC all possessed a rewarding effect and aversive effect that induced significant CPP and CPA in a dose-dependent manner by intraperitoneal injection. This phenomenon may relate to the release of dopamine in the nucleus accumbens shell (NAc shell), as demonstrated by the inverted U-shaped DA release patterns following SC administration [44]. However, the exact CPA mechanisms remain unclear, as high-dose aversion could represent either true drug aversion or motor impairment—a distinction requiring future studies with operant assays or microdialysis during conditioning. In contrast, higher doses that induce CPA could lead to nonspecific locomotor suppression (as seen in our tetrad data). Notably, the minimal effective CPP doses were substantially lower than those required for locomotor suppression: 5F-ADB induced CPP at a dose of 0.003 mg/kg, which is 1/10 of the dose which suppressed locomotion significantly. MDMB-4en-PINACA and ADB-4en-PINACA induced CPP at a dose of 0.03 mg/kg and 0.1 mg/kg, respectively, which is 1/3 of the dose which could suppress locomotion significantly. Compared with the dose of Δ⁹-THC used to induce CPP (3 mg/kg), 5F-ADB showed the highest potency to induce CPP, followed by MDMB-4en-PINACA and ADB-4en-PINACA. It indicated that the rewarding effects of 5F-ADB, MDMB-4en-PINACA and ADB-4en-PINACA were approximately 1000 times, 100 times and 30 times that of Δ⁹-THC, respectively. Although the potency ratios may differ from in vitro data reported previously, in which the EC₅₀s were approximately 290 times (5F-ADB), 27 times (MDMB-4en-PINACA) and 10 times (ADB-4en-PINACA) that of Δ9-THC, respectively, the rank order was consistent [14,26,27]. The differences in the ratios also suggested that the potency obtained from in vitro experiments sometimes cannot fully reflect the in vivo effects of drugs. Meanwhile, we can speculate that the three SCs all have high abuse potential. However, in our study, Δ⁹-THC elicited CPP at 3 mg/kg and CPA at 100 mg/kg, which is higher than that reported before, where CPP doses were approximately 0.1–0.3 mg/kg and CPA doses were 1 mg/kg and above [45]. This may be related to the different types of experimental animals used and different administration method. It is worth noting that the observed CPA may be due to either a drug aversion effect or a locomotor suppression effect (as demonstrated in our tetrad assay (Figure 2)). The reasons for this phenomenon need further research. Despite the relatively modest CPP magnitudes (~100 s), the statistical significance of low-dose preference, combined with the dose-dependent shift to aversion, supports the biological relevance of these findings. While stronger effects (200–300 s) are typical for highly reinforcing drugs (e.g., opioids), cannabinoids often produce subtler behavioral responses, as seen in prior studies [46,47]. The inverted U-shaped dose–response curve, coupled with the dissociation between rewarding and locomotor-suppressive doses, underscores the high abuse potential of these SCs. Future studies could refine the conditioning protocols to enhance the sensitivity to cannabinoid-specific reward phenotypes.

Withdrawal signs were induced by the synthetic cannabinoids and the physical withdrawal syndrome closely resembled that seen in cannabis dependence [48]. The CB1 receptor antagonist rimonabant precipitates cannabinoid withdrawal symptoms in rodents mainly including head twitches, face rubbing, scratching, wet dog shakes, piloerection and front paw tremors [49,50,51]. In our withdrawal assay, we selected two common and easily observable signs—paw tremors and head twitches—as the primary withdrawal measures for the experiment. The results showed that repeated administration for 5 days in a dose-increasing manner, 5F-ADB, MDMB-4en-PINACA and ADB-4en-PINACA could induce paw tremors and head twitches in mice after injecting rimonabant (Figure 4). Compared with the negative group, 5F-ADB caused the most paw tremors, followed by MDMB-4en-PINACA and ADB-4en-PINACA. For head twitches, 5F-ADB and MDMB-4en-PINACA showed comparable frequencies, both more than that of ADB-4en-PINACA. Overall, 5F-ADB caused the strongest physical withdrawal symptoms, while ADB-4en-PINACA was the weakest one. Meanwhile, the initial dose of 5F-ADB in the withdrawal experiment is lower than that of MDMB-4en-PINACA, ADB-4en-PINACA and Δ⁹-THC. These results suggest that the physical dependence potency ranking was 5F-ADB > MDMB-4en-PINACA > ADB-4en-PINACA > Δ⁹-THC, like in the tetrad. It should be pointed out that rimonabant alone also caused head twitches, as seen in the vehicle–rimonabant group (Veh + Rim). This phenomenon has been reported in the literature [52]. As a potent CB1 receptor antagonist, rimonabant can precipitate mild withdrawal-like symptoms even in drug-naive animals by blocking basal endocannabinoid tone. Importantly, our data demonstrated that the Drug + Rim groups showed greater symptom frequencies than the Veh + Rim controls (p < 0.05). This phenomenon suggested that, while rimonabant alone produced baseline effects, the amplified responses in SC-pretreated animals specifically reflect cannabinoid dependence. The Veh + Rim data serve as a crucial baseline for distinguishing pharmacological withdrawal from nonspecific effects.

Understanding the structure–activity relationship (SAR) is crucial for predicting the pharmacological effects of new analogs, evaluating the abuse potential of structural analogs and providing important reference information for drug control departments. In this study, we sought to elucidate the structure-activity relationships according to the results of in vivo experiments. The three SCs tested in this study all belong to the indazole-carboxamide family. They share the same indazole “core”, carboxamide “neck” (linker) and differences in the “head” and “tail” groups. Based on the analysis of the present results, we concluded that modifications to the head and tail regions led to some differences in potency. The name of 5F-ADB (5F-MDMB-PINACA) is derived from its structural features: a fluoro moiety at position 5 of the pentyl chain (5F), a dimethyl methyl butanoate (tert-leucinate) head (MDMB), a pentyl tail (P), an indazole core (INA) and a carboxamide linker (CA). Structurally, MDMB-4en-PINACA contains a pent-4-ene moiety on a pentyl tail, i.e., it differs in the tail (5-fluoropentyl) from 5F-ADB. Halogen substitution enhances lipophilicity, facilitating blood–brain barrier penetration. Canazza et al. demonstrated that fluorination can increase the power and/or effectiveness of SCs [53]. The present study showed that 5F-ADB is more potent in inducing tetrad effects, CPP and precipitation of withdrawal compared with MDMB-4en-PINACA, probably due to its fluorination, which may contribute to the rapid penetration across the blood–brain barrier into the central nervous system. Gillis et al. [54] reported that fluorination is widely recognized to improve both receptor binding affinity (2–5 fold potentiation) and metabolic stability in psychoactive compounds, supporting our observation that 5F-ADB (average ED₅₀ = 0.04 mg/kg) was 3.5 times more potent than MDMB-4en-PINACA (average ED₅₀ = 0.14 mg/kg) in producing tetrad effects. In summary, the observed in vivo potencies reflect a combination of two key factors: (1) differential CB1 receptor binding affinities, as demonstrated in our cited in vitro studies [14,26,27], (2) compound-specific pharmacokinetic profiles, including blood–brain barrier penetration and metabolic stability. For instance, the superior potency of 5F-ADB likely stems from both its high CB1 affinity and enhanced brain bioavailability due to fluorination [53,54]. In addition, ADB-4en-PINACA has the same tail as MDMB-4en-PINACA, featuring an L-tert-leucinamide rather than a tert-leucinate “head”. In this study, it was observed that MDMB-4en-PINACA had stronger tetrad effects, rewarding effect and precipitation of withdrawal than ADB-4en-PINACA. We concluded that a tert-leucinate moiety in the “head” leads to a higher potency than the corresponding L-tert-leucinamide head group. The deduction was consistent with a previous study with the same structural analogs ADB-4en-PINACA and MDMB-4en-PINACA, in which the conclusion was reached that the presence of the L-tert-leucinate head group led to more potent and sustained effects than the L-tert-leucinamide head group [31]. These findings will help us study the structure–activity relationships of SCs from the perspective of animal experiments.