Subchronic effects of ligands of cannabinoid receptors on learning and memory processes of olfactory bulbectomized rats

The brain endocannabinoid system has been shown to play a role in many physiological processes, including mood, learning and memory. It is also involved in the pathogenesis of anxiety, depression, mood disorders, as well as neurodegenerative disorders, although the exact mechanisms by which cannabinoid receptors interfere in these disorders are not well established. The aim of the present study was to evaluate the effects of cannabinoid ligands HU‐210 (CB1 receptor agonist) and SR 141716A (CB1 receptor antagonist) on learning and memory processes of rats with depressive – like state, induced by bilateral olfactory bulbectomy. The bilateral olfactory bulbectomy (OBX) is a validated model of depression, which can be used also as an animal model of Alzheimer’s disease. We found that the subchronic treatment of OBX rats with HU 210 and SR 141716A exerted modulatory effect on rat’s performance in both active avoidance (shuttle box) and passive avoidance (step through) tests. HU 210 ameliorated the memory deficits of OBX rats; however, the scores of the sham‐operated controls had not been reached. SR 141716A modified the avoidance performance in OBX rats and showed a memory enhancing effect in the sham‐operated rats. Our findings suggest that CB1 receptors might be involved in avoidance learning and memory acquisition in OBX rats.


INTRODUCTION
Endocannabinoid system (ECS) plays role in many physiological processes, including mood, learning and memory. It is also involved in the pathogenesis of anxiety and mood disorders, as well as neurodegenerative disorders (Hill and Gorzalka, 2009;Ranieri et al., 2016). ECS consists of the endogenous cannabinoids (endocannabinoids), cannabinoid receptors and the enzymes that synthesize and degrade endocannabinoids. The cannabinoid receptors (CB1 and CB2) belong to the class of G-protein coupled membrane receptors. CB1 receptors are expressed mainly in the central nervous system with highest density being detected in the basal ganglia, hippocampus, cerebellum, prefrontal cortex, amygdala (Mackie, 2005). The endogenous cannabinoids act as retrograde modulators of neurotransmission being released on demand from the postsynaptic neurons and binding to the presynaptically located CB1 receptors (Freund et al., 2003;De Petrocellis and Di Marzo, 2009). The manipulation of the endocannabinoid signaling often produces contrasting findings concerning cognition and emotions, which can be related to the fact that CB1 receptors are expressed at stimulatory (glutamatergic) and inhibitory (GABAergic) synapses (Ruehle et al., 2012).
The role of the cannabinoid receptors in learning and memory processes has been intensively studied, but is not yet fully understood. There are numerous reports about in general, memory impairing effects of cannabinoid agonists and memory enhancing effects of cannabinoid antagonists (Zanettini et al., 2011;Kruk-Slomka et al., 2017). However, this is a simplification, as the effects may differ depending on the dose, behavioral methods used, experimental design, etc. There is an intensive research on modulation of the components of ECS in attempt to develop pharmacological therapy for treatment of anxiety and depressive disorders. Studies also have provided evidence that the ECS has neuro-protective properties and might be a target in neurodegenerative diseases (Ranieri et al., 2016).
The bilateral olfactory bulbectomy (OBX) is a validated model of depression. The removal of bulbi olfactorii produces a syndrome of behavioral, neurochemical, neuroendocrine, immune, etc. alterations, that resembles human depressive disorder (Kelly et al., 1997;Song and Leonard, 2005). OBX causes long lasting neurodegenerative changes in many brain areas and therefore, it is suggested that it might be a valuable tool in the study of neurodegenerative disorders like Alzheimer's disease (AD) (Douma et al., 2011;Borre et al., 2012). OBX can be used as an animal model of AD because it produces some effects similar to AD-related symptoms and pathology (Aleksandrov et al., 2004;Yehuda and Rabinovitz, 2013;Franks et al., 2015). Behavioral abnormalities developed by the OBX rodents include hyperlocomotion, memory disturbances, reduced sexual activity, aggressive behavior, hyperemotionality. OBX rats display deficits in passive-avoidance test (Kelly et al., 1997) and impaired response in the two-way active avoidance task (Gomita et al., 1984;Archer et al., 1984). While the activation of the CB1 receptor by agonists has been found to predominantly improve the depressive-like state in animal models (Segev et al., 2014;Kruk-Slomka et al., 2015;Haj-Mirzaian et al., 2017), the data on the role of CB1 antagonists are contradictory. Previously, we have demonstrated impaired performance of OBX rats in both active and passive avoidance tests (Tashev et al., 2010) and modulatory effects of acutely i.c.v applied CB1 ligands on the performance of OBX rats in these tests (Marinov et al., 2013).
Based on the results of our acute treatment experiments, the aim on the present study was to evaluate the subchronic effects of CB1 agonist HU 210 and CB1 antagonist SR 141716A applied i.c.v. for 7 days on the background of developed depressive-like state, on learning and memory performance of OBX rats, tested in two-way active avoidance (TWAA) and passive avoidance (PA) paradigms.

Subjects
The experiments were carried out on male Wistar rats (200-220 g at the time of surgery). For each behavioral task (locomotor activity, shuttle box, step through), the animals were randomly divided into 6 major experimental groups (7 animals in each group). Group I -sham-operated (sham) and group II -OBX (controls) received saline; group III and IV (sham) as well as group V and VI (OBX) received HU 210 or SR141716A respectively. The rats were housed in polypropylene boxes with free access to food and water. The animals were maintained in a constant temperature environment (22 ± 2°C) on a 12 h light/dark cycle (lights on at 6:00 a.m.). The behavioral experiments were carried out between 9:00 a.m. and 1:00 p.m. After the testing procedure, the rats were returned to their home cages.

Ethical statement
The experiments were performed according to the "Principles of laboratory animal care" (NIH publication no. 85-23, revised 1985) and the rules of the Ethics Committee of the Institute of Neurobiology, Bulgarian Academy of Sciences. All efforts were made to minimize animal suffering and reduce the number of animals used in the study.

Surgical procedures
OBX was performed according to the method described by Kelly et al. (1997). Animals were anesthetized with Calypsol (50 mg/kg i.p.) and placed in a stereotaxic apparatus. The surgical procedure involved drilling two burr holes 2 mm in diameter at the points 8 mm anterior to bregma and 2 mm from the midline on its both sides (coordinates of bulbi olfactorii were detected according to the stereotaxic atlas of Pellegrino and Cushman (1967). The bulbs were aspirated with a stainless needle attached to a water pump. The implantation of the cannula into the right lateral ventricle was performed 7 days after OBX surgery, as previously described (Marinov et al., 2013). The sham operation included the same procedures as for the olfactory bulbectomy, without the removal of the olfactory bulbs. During the 7 day recovery period, the rats were handled daily.

Drugs
HU 210 and SR 141716A (Tocris Bioscience) were dissolved ex tempore in a 1:19 solution of dimethyl sulfoxide/0.9% saline. One µl of drug solution HU-210 (0.5%) or SR 141716A (0.3%) with pH=7.4 were infused i.c.v. through an injection cannula connected by poly-ethylene tubing with a constant rate microsyringe (Hamilton, Reno, NV, USA) over a period of 1 min. The injection cannula was left in place for additional 30 s. After the surgery for cannula implantation, the animals were allowed to recover for 7 days. During the recovery period, the rats were handled daily. HU 210 (5 μg/1 μl), SR 141716A (3 μg/1 μl) or saline (1 μl) were infused i.c.v. for 7 consecutive days to the respective group, starting on the 15 th day after OBX procedure.
The behavioral tests were carried out 21 days after the surgery for the removal of the olfactory bulbs. Before each test, the rats were placed for 5 min in the respective test cage for 3 consecutive days to acclimatize to it (Fig. 1).

Locomotor activity
Locomotor activity was recorded in an Opto Varimex apparatus (Columbus Instruments, USA). The experimental chamber was 50 cm × 50 cm × 25 cm. The apparatus records the number of photo beam interruptions during the movements of the animal. It provides counting of the number of horizontal (ambulation) and vertical movements (rearing) in arbitrary units (AU). The information obtained was recorded automatically for 30 min. The rats were placed in the central quadrant of the activity monitor, 5 min after the microinjection of the drugs.

Two-way active avoidance test
The animals were trained in a shuttle-box apparatus according to the method of Buresova and Bures (1983) and modified by Petkov et al. (1993). The shuttle box apparatus is a box (50 × 29 × 21 cm) separated into two compartments by a wall with a U-shaped gate. Light (21 W) was used as a conditioned stimulus (CS); the un-conditioned stimulus (US) was electric non-scrambling current (20-30 V, 0.5 mA, AC, 50 Hz) delivered through the grid floor for 12 s. The CS preceded the onset of the US by 9 s and continued during the action of the US. Prior to avoidance training the rats were given a 6-min session of exposure to light stimuli in the shuttle box (alternating 9 s light and 9 s interval) for adaptation. During the first training session each rat was trained with 50 trials; 24 hours later was carried out the second training session. At the retention test (24 h after the 2 nd training session), the light stimulus was applied for 9 s and was followed by a 1 s electric shock.
Each trial began with a 9 s light stimulus, followed by a 12 s shock. One trial lasted 21 s, the inter-trial interval was 9 s. Each trial started in the compartment where the rat was located at the end of the inter-trial interval. If the rat crossed the barrier to the opposite compartment within 9 s after the onset of light stimulus, the stimulus was terminated and no shock was delivered (avoidance response). A crossing response escape during shock terminated the stimuli. If the rat failed to cross during the entire trial (inadequate reaction), the light and the shock was terminated. The data from each shuttle box were fed into an analogue-to-digital converter, coupled to a computer with appropriate software. The number of avoidances was analyzed as a primary measure of learning and memory in the active avoidance task. The drugs were microinjected 5 min prior to the first and second training session.

Passive avoidance test
The Ugo-Bassile avoidance instrument is a cage divided into two compartments. In the training trial, the rat is placed in the brightly lit compartment and it must learn to remain there and not to escape to the preferred dark compartment to avoid a mild foot shock. Once the rat had entered the dark compartment, the door was closed and an electrical shock (0.3-0.35 mA for 3 s) was delivered through the grid floor. One training trial and two retention tests were conducted according to the method described by (Gozzani and Izquierdo, 1976). Retention tests (no shocks) were performed 3 h and 24 h after the training trial. The animals were placed in the light compartment, and step-through latency was estimated by measuring the length of time (latent time) for the rat to move to the dark compartment. A maximum latency of 180 seconds was used as a criterion for learning. The drugs were microinjected i.c.v. 5 min prior to the training trial.

Verification
Following termination of the experiments and immediately prior sacrifice, the rats were injected with 1 ml of 2% fast green dye through injection cannula into the right lateral ventricle. The cannula placement was checked visually after dissection. If cannula was successfully placed in a ventricle, the dye was distributed throughout the ventricular system and the liquor was colored. The verification of the olfactory bulbectomy was done by visual inspection. Rats with incomplete (<80%) removal of the olfactory bulbs were excluded from the statistical analysis. Thus, 4 rats were discarded.

Statistical analysis
Оne-way ANOVA was used to process the data obtained for total number of horizontal and for vertical movements during the whole 30-min period of observation.
Two-way repeated measures ANOVA were used to analyze the data obtained for number of avoidances (shuttle-box) between subject factors: treatment (six levels: sham-saline, sham-HU, sham-SR, OBX-saline, OBX-HU and OBX-SR) and time (three levels: 1 st day, 2 nd day and 3 rd day retention test -24 h after the 2 nd training day).
Two-way repeated measures ANOVA were used to process the data obtained for latent time (step through) between subject factors: treatment (six levels: sham-saline, sham-HU, sham-SR, OBX-saline, OBX-HU and OBX-SR) and time (3 rd hour and 24 th hour).
ANOVA data were further analysed by post hoc Student-Newman-Keuls (SNK) test where appropriate. Analysis of the passive avoidance data (step through) was performed using χ2 tests.

Effects of HU-210 and SR 141716A in sham-operated rats
Separate one-way ANOVA analysis on the total number of horizontal or vertical movements for the whole 30-min period of observation showed a significant effect for factor drug (F 2,20 =87.597; P≤0.0001) for the number of horizontal movements, and respectively for the number of vertical movements: (F 2,20 =51.877; P≤0.0001). The microinjections of HU 210 significantly decreased the total number of horizontal (P≤0.0001), and vertical (P≤0.0001) movements as compared to saline-treated controls, while SR 141716A increased the locomotor activity (P≤0.0001; P≤0.0001 respectively) as compared to the controls ( Fig. 2A-B).

Effects of HU-210 and SR 141716A in OBX rats
Three weeks after the OBX procedure, OBX rats showed hyperlocomotion, increased number of move-289 Acta Neurobiol Exp 2020, 80: 286-296 ments in the Opto Varimex apparatus. Separated one way ANOVA on the total number of horizontal and vertical movements of OBX rats demonstrated a significant effect for bulbectomy factor (respectively F 1,13 =93.120, P≤0.00001; F 1,13 =66.305, P≤0.0001). Post hoc SNK comparisons demonstrated that the total number of horizontal (P≤0.0001), and vertical (P≤0.0001) movements was higher as compared to sham-operated controls ( Fig. 2A-B) ANOVA of the total number horizontal and vertical movements after infusion of HU 210 and SR 141716A in OBX rats showed a significant effect for factor drug (respectively F 2,20 =107.006, P≤0.00001; F 2,20 =48.923, P≤0.0001).
The SNK test showed that HU-210 significantly decreased the number of both horizontal (P≤0.0001) and vertical (P≤0.0001) movements as compared to the OBX saline-treated rats. The horizontal activity was not significantly different in comparison to the sham-operated rats (P-NS), the number of vertical movements was lower (P≤0.001) ( Fig. 2A-B).
Post hoc SNK comparisons showed that HU 210 impaired the performance of sham-operated rats by decreasing the number of avoidances on the 1 st day (P≤0.04); 2 nd day (P≤0.01) and аt the retention test (P≤0.006) as compared to the sham-saline controls (Fig. 3A-С).
SR 141716A microinjected in the sham-operated rats increased significantly the number of avoidances on the 1 st day (P≤0.04), 2 nd day (P≤0.03) and at the retention test (P≤0.04) as compared to the saline-treated controls (Fig. 3A-С).
The bilateral removal of bulbi olfactorii impaired the performance of OBX rats in the shuttle box. The SNK test showed significantly lower number of avoidances as compared to the sham-controls on the 1 st training day (P≤0.0001), the 2 nd training day (P≤0.0001) and on the retention test (P≤0.0001) (Fig. 3A-С).
The post hoc test of the effects of the CB1 ligands on the number of avoidances showed that HU 210 ameliorated the memory deficits of OBX rats, demonstrated by an increased number of avoidances on the 1 st day (P≤0.02); 2 nd day P≤0.005 and the retention test (P≤0.003) 290 Acta Neurobiol Exp 2020, 80: 286-296 as compared to the OBX-saline controls. However, the number of avoidances on the two training trials and at the retention test remained lower (respectively P≤0.02; P≤0.005, P≤0.003) as compared to the sham-operated rats (Fig. 3A-С).
SR 141716A revealed a memory deteriorating effect in OBX rats by worsening the memory deficits, induced by the bulbectomy. The CB1 antagonist decreased the number of avoidances in the ТWАА test on the 1 st , 2 nd training day and at the retention test as compared to both OBX saline-treated controls (P≤0.05; (P≤0.05; P≤0.01) and sham-operated rats (P≤0.0001; P≤0.0001; P≤0.0001) (Fig. 3A-С).
HU 210 microinjected in the sham-operated rats reduced the latent time at 3 rd hour (P≤0.01) and 24 th hour (P≤0.006) as compared to the sham-operated rats and decreased the percentage of rats, reaching the learning criteria at the retention tests -on 3 rd hour (χ 2 =2.800, P≤0.05) and 24 th hour (χ 2 =4.667, P≤0.02) (Fig. 4A-B; Table I). SR 141716A increased the latent time of the sham-operated rats on the 3 rd hour only (P≤0.05), with no effect on the 24 th hour (P=NS). As compared to the sham-saline rats, where 57% reached learning criteria on the 3 rd hour), SR 141716A increased the percent-age of rats to 71%, while on the 24 th hour there was no significant difference (the percentage remained 71%, same as the one for the saline-treated controls) (Fig. 4A-B; Table I).
The bilateral removal of rat bulbi olfactorii produced a significant decrease in the latent time of OBX rats on 3 rd h (P≤0.0001) and on 24 th h after training (P≤0.0001) as compared to the sham-operated controls. The percentage of OBX rats, that did not reach the learning criteria diminished to 0% in both retention tests (P≤0.001) and was lower on the 3 rd h (χ 2 =4.000; P≤0.05) and 24 th h (χ 2 =6.002; P≤0.02) as compared to the sham controls ( Fig. 4A-B; Table I).
The post hoc test of the effects of HU 210 revealed аn increased latent time of the OBX rats on the 3 rd hour (P≤0.03) and the 24 th h (P≤0.04). The number of OBX rats reaching the learning criteria on 24 th hour (χ 2 =2.333, P≤0.05) was higher as compared to the OBX-saline controls. HU 210-treated OBX rats had a decreased latent time on 3 rd and 24 th hour (P≤0.005 and P≤0.01 respectively); the number of rats reaching learning criteria also was lower on 3 rd (χ 2 =2.800, P=NS) and 24 th hour (χ 2 =2.571, Р≤0.05) as compared to the sham-operated rats. (Fig. 4A-B; Table I).
SR 141716A decreased the latent time of OBX rats on 3 rd hour (Р≤0.05); on 24 th h (Р≤0.01) increased it, and did not affect the number of rats, reaching the learning criteria at the 3 rd and 24 th hour as compared to the OBX-saline controls. The comparison of the effects of SR 141716A in OBX rats to the sham-saline group showed a decrease of the latent time on 3 rd hour (P≤0.001), 24 th hour (P≤0.001) and a decrease of the number of rats, reaching learning criteria on 3 rd (χ 2 =5.600, Р≤0.02) and 24 th hour (χ 2 =7.778, Р≤0.01) (Fig. 4A-B; Table I).

DISCUSSION
The olfactory bulbectomy is a validated model of depression. Apart from being used primarily as a model for examining the effects of new antidepressant drugs, OBX is also considered a suitable model to study neurodegenerative disorders, such as Alzheimer's disease. OBX model can be used to evaluate the relationship between the abnormal neuronal signaling and the induced cognitive changes. The removal of olfactory bulbs in rodents induces behavioral changes including increased locomotor activity in novel environment, hyperemotionality, memory deficits, etc. Tuite,1981, Kelly et al., 1997).
In the present study we aimed to investigate the role of CB1 receptors in avoidance learning and memory of rats with OBX model by using CB1 receptor agonist HU 210 and CB1 receptor antagonist SR 141716A. We revealed modulatory effects of the CB1 ligands, administered i.c.v. for 7 consecutive days to OBX rats with developed depressive-like state, on аctive and passive avoidance behavior. HU 210 improved partially the learning and memory deficits of OBX rats in both tasks, demonstrated by an increased number of avoidances (shuttle box), prolonged latent time and increased number of rats, reaching learning criteria (step through). However, the scores of the sham-operated controls had not been reached. The CB1 antagonist SR 141716A impaired the performance of OBX rats in the shuttle box test only, and showed inconsistent effects in the passive avoidance test. In the sham-operated group, HU 210 induced significant learning and memory disturbances, while the SR 141716A administration enhanced avoidance retention in both tasks.
The locomotor activity of rats was evaluated in an Opto Varimex apparatus. HU 210 normalized locomo-tor hyperactivity of the OBX rats, while SR 141716A exacerbated the depressive-like symptoms by further increasing the locomotor activity. In the sham-operated rats the CB1 ligands showed the same modulatory effects: HU 210 inhibited locomotion, while SR 141716A increased it.
As far as drug-or lesion-induced alterations in locomotor activity are suggested to confound mostly passive avoidance response (Ögren and Stiedl, 2015), we used active and passive avoidance paradigms for evaluation of the effects of CB1 ligands. In the passive-avoidance task (step through) the performance depends on the suppression of the innate drive to enter a dark compartment and rats with decreased locomotor activity would stay longer time over the illuminated platform before moving to the dark compartment (extended latent time).
An interesting finding of our study was that the CB1 ligands produced opposite effects on acquisition and retention of memory in sham and OBX rats, while the locomotor effects remained unchanged in both groups. Thus, HU 210 inhibited locomotion (sham and OBX rats), deteriorated memory (sham) and ameliorated memory deficits (OBX), while SR 141716A increased locomotor activity (sham and OBX rats), enhanced memory (sham), and worsened performance in the active avoidance test (OBX).
The fact that CB1 ligands exerted unidirectional locomotor effects in both sham and OBX rats but oppositely affected the avoidance behavior of the two groups, allows for the assumption that the altered locomotor activity does not interfere significantly with the observed learning and memory effects.
The two avoidance tests, used by us, share similarities, concerning the involvement of major structures of the limbic system and some cortical and brainstem 292 Acta Neurobiol Exp 2020, 80: 286-296 areas. In addition, each of the two tasks is associated with specific brain areas. Lesion studies have shown that passive avoidance test in rodents requires an intact corticohippocampal circuit (Burwell et al., 2014), while active avoidance is a conditioned task which includes amygdala and requires intact amygdala projections to the ventral striatum (Ramirez et al., 2015). Passive avoidance is mostly fear-motivated, inhibitory conditioning, while active avoidance is an example of operant conditioning. The deficit in passive avoidance task is a consistent finding in OBX model (Sieck, 1972;Kelly et al., 1997), which has been interpreted as an abnormal defensive freezing capacity of the OBX rat (Primeaux and Holmes, 1999). However, the data about the rat's performance in the active avoidance task are contradictory, no significant changes, improved or impaired performance have been observed (Gomita et al., 1984;Archer et al., 1984, Grecksch et al.,1997Tashev et al., 2010). Yamamoto et al. (1997) suggested that attention deficit is involved in the memory impairment of OBX rats tested in 3-lever operant task. The effects of the CB1 agonist on the avoidance behavior are similar to the ones previously reported upon an acute i.c.v injection, where HU 210 improved the scores of the OBX rats in both avoidance tasks and impaired the performance of the sham group (Marinov et al., 2013). Unlike the acute treatment, where SR 141716A enhanced the avoidance learning of sham rats in TWAA only, the subchronic treatment enhanced memory of the sham group (both tasks) and showed a tendency to exacerbate memory deficits of the OBX rats in TWAA.
Data are contradictory regarding possible influence of the rat's anxiety-like behavior on the avoidance responses. Two strains of mice with different level of anxiety were tested in a PA paradigm and the C57BL/6J strain with lower anxiety showed longer retention latencies than the more anxious DBA/2J mice (Baarendse et al., 2008). In a previous study we demonstrated that acutely i.c.v. injected HU 210 alleviated the OBX-induced anxiety, while SR 141716A failed to produce significant effects (Marinov et al., 2019). Therefore, we cannot exclude the possibility that the anxiety-modulating effects of the CB1 ligands might affect the avoidance performance.
The abnormalities seen in the OBX rat are considered to be a result of neurodegenerative changes and disrupted neuronal connections between the olfactory bulbs and other brain regions. Olfactory bulbectomy causes neuronal degeneration and dysfunction in the projection areas to the cortex, hippocampus, amygdala, locus coeruleus (Kelly et al., 1997, Morales Medina et al., 2017. Abnormalities in many neurotransmitter systems in the brain have been reported such as nor-adrenergic, serotonergic, cholinergic, glutamate, etc. (Song and Leonard, 2005).
Significant alterations of the ECS in several rat brain structures have been detected in the OBX model. The levels of endogenous cannabinoids were changed in prefrontal cortex, hippocampus, striatum and nucleus accumbens; the CB1 receptor expression was lower in the hippocampus, dorsal striatum, and nucleus accumbens and the CB2 receptor expression decreased in the prefrontal cortex and hippocampus (Smaga et al., 2017). Eisenstein et al. (2009) provided evidence for dysfunction of ECS in olfactory bulbectomized rats which may have an impact on the functional activity of other neurotransmitter systems. Hyperlocomotion is a major OBX-induced depressive-like behavioral symptom (van Riezen and Leonard, 1990), which has been attributed to the hyperdopaminergic transmission (Masini et al., 2004). Findings suggest that dysregulation in the ECS (inability to efficiently modulate brain dopaminergic neurotransmission) is implicated in the hyperactive locomotor response induced by OBX (Eisenstein et al., 2010).
Our results support the accumulating data about the antidepressant and memory improving effects due to enhanced activity of endocannabinoid system in different experimental animal models accompanied by deficits in cognitive processes (McLaughlin andGobbi, 2012, Segev et al., 2014;Kruk-Slomka et al., 2015;Haj-Mirzaian et al., 2017). The effects of the subchronically i.c.v applied CB1 ligands in the PA task are in accordance with the reported involvement of CB1 receptors in memory-related processes in the inhibitory avoidance test in mice. Acutely injected oleamide, a CB1 receptor agonist, showed antidepressant-like effect of in the forsed swim test, while AM 251, a CB1 receptor antagonist, did not provoke any effect in this test (Kruk-Slomka and Biala, 2016).
The neuroprotective properties of the cannabinoids in models of neurodegenerative disorders have been confirmed by numerous studies during last decade and may account for the improved performance in the avoidance tests upon subchronic treatment with HU 210 observed by us. The neuroprotective potential involves ECS and its numerous targets by activation of CB1 and CB2 receptors in different types of neurons and glial cells (Fernández-Ruiz et al., 2015. Experimental data suggest that fear may be involved in the maintenance of avoidance, although fear and avoidance may not be always correlated (Rachman and Hodgson, 1974;Krypotos et al., 2015). Passive avoidance is a hippocampal dependent task, based on associative (emotional) learning similar to contextual fear conditioning (Ögren and Stiedl, 2010;Burwell et al., 2014). Regarding the effects of the CB1 ligands in the sham-operated rats, our findings are in line with the reports about the established role of hippocampal CB1 receptors in the memory disruptive effects of systemically administered cannabinoids. It was demonstrated that the injection of CB1/CB2 receptor agonist WIN55, 212-2 and the inhibitor of endocannabinoid reuptake and breakdown AM404 in the hippocampal CA1 area facilitated the extinction of fear related inhibitory avoidance, while the CB1 receptor antagonist AM251 impaired it (Abush and Akirav, 2010). Wise et al. (2009) reported that the intrahippocampal administration of rimonabant (CB1 receptor antagonist), prevented the memory disruptive effects of systemically administered Δ9-THC in the radial arm maze. In addition, we support the findings that manipulation of CB1 transmission affects classically conditioned fear (Chhatwal et al., 2005;Pamplona et al., 2008).
In our study we used SR 141716A, which is selective CB1 antagonist that shows properties of inverse agonist (Pertwee, 2005) that can exhibit agonist activity on constitutively active receptors. SR 141716A affected negatively the performance of OBX rats in the TWAA only, and improved learning and long-term memory of sham rats in both tasks. Active avoidance is a type of Pavlovian fear conditioning paradigm. The amygdala and its projection to the ventral striatum are essential for aversively motivated activity (Kapp et al., 1984;Ramirez et al., 2015). Discrete lesions of the amygdala produced deficits in active avoidance (Werka et al., 1978, Choi et al., 2010. It could be suggested that the effects of CB1 ligands on the acquisition and memory retention in TWAA are linked to the facilitated extinction of aversive memories through the selective inhibitory effects of endocannabinoids on local inhibitory networks in the amygdala (Marsicano et al., 2002). In this regard, CB1 inverse agonist, but not neutral antagonist has been shown to enhance the retention of contextual fear conditioning in rats (Sink et al., 2010).
Recently, a meta-analysis was conducted on the effects of CB1 receptor agonists, antagonists, and negative allosteric modulators on memory (Borgan et al., 2019). It revealed that CB1 receptor agonists when administered acutely, but not chronically, impair spatial and non-spatial memory and the effects were inversely associated with dose. No effects of acutely administered CB1R antagonists have been reported and there were no data about chronic administration of the drugs. Our data confirm the established detrimental effect of CB1 ligands on memory, and can contribute to the understanding how pharmacological manipulation of CB1 receptors influences cognitive performance.

Strengths and limitations
A strength of the study was that we examined the subchroinic effects of CB1 receptor agonist and antagonist, on avoidance performance in rats with a OBX model. Data are lacking about the effects of subchronic and chronic administration of CB1 antagonists on non-spatial memory in rodents.
Our research has some limitations which have to be considered. We used a single dose of the drugs and it was applied prior to task. The experimentation with different drug doses, route of administration, оr treatment protocols (i.e. injection of drugs after training) would give more light to the modulatory effects of the CB1 receptor ligands on learning and memory processes in depressive-like states. The comparison of the effects of CB1 antagonist/inverse agonist to a CB1 neutral antagonist may be beneficial for evaluation of the effects of CB1 receptor blockade. Studies with chronic administration of CB1 agonist are also needed, as it looks promising to bring full recovery of the memory disturbances in OBX rats.

CONCLUSION
Our study demonstrated that the subchronic i.c.v. administration of CB1 antagonist HU 210 attenuated the depressive-like state in an OBX rat model by normalizing the locomotor activity and partially improving the memory disturbances, induced by the olfactory bulbectomy. The CB1 antagonist SR 141716A modified the avoidance performance in OBX rats and showed a significant memory enhancing effect in the sham-operated rats. The results point to a possible involvement of CB1 receptors in the learning and memory deficits of OBX rats, tested in active and passive avoidance paradigms. Future studies would add to our understanding of the contribution of the CB1 receptors to the mechanisms of memory impairment in depression and Alzheimer's disease.