The cannabis preparations marijuana and hashish are the most popular and most commonly used illicit drugs worldwide. These are derived from the female plant of Cannabis sativa L (family Cannabinaceae). Marijuana is prepared from the dried flowering tops and leaves; hashish consists of dried cannabis resin and compressed flowers (Ashton, 2001[4]). Cannabinoids are a group of C21 terpenophenolic compounds uniquely produced by Cannabis sativa plant. The primary psychoactive constituent is Δ⁹-tetrahydrocannabinol (THC) (Mechoulam and Gaoni, 1967[40]). Other plant cannabinoids include Δ⁸-THC, cannabinol and cannabidiol (Adams and Martin, 1996[3]).

Memory or the retention of learned information is fundamental to human beings. Cannabis use causes working and short-term memory deficits in humans as well as in experimental animals (Solowij and Battisti, 2008[71]; Fadda et al., 2004[20]). This effect applies to both short and long-term use of cannabis (Solowij and Battisti, 2008[71]; Solowij et al., 2011[72]); the impairments may persist well beyond the period of intoxication, and recovery of functions might take weeks following abstinence (Pope et al., 2001[57]) or persist for longer time (Solowij et al., 2002[73]). In chronic users, cannabis might impair the ability to learn and remember new information. Early onset, long-term use and higher frequency of use are seen as risk factors for cognitive impairments (Harvey et al., 2007[27]). Cannabinoids exert their effects by interaction with specific endogenous cannabinoid receptors. Two G protein-coupled cannabinoid receptor subtypes have been cloned: CB1 receptor which is widely distributed throughout the brain, but particularly in the cerebral cortex, hippocampus, cerebellum, thalamus and basal ganglia (the brain regions involved in cognition, memory, reward, pain perception, and motor coordination) and CB2 receptor which is mainly expressed on immune cells, but also in central nervous system. These receptors also respond to endogenous ligands, the endocannabinoids such as anandamide and 2-arachidonoylglycerol, which are arachidonic acid derivatives. Cannabinoid CB1 receptors were identified in the hippocampus, the brain region most closely associated with memory (Pertwee and Ross, 2002[55]; Pertwee, 2005[53]; Svíženská et al., 2008[75]). The precise nature of memory deficits in cannabis users and their neural substrates still require further research (Solowij and Battisti, 2008[71]).

In the treatment of memory disorders whether occurring as a part of normal aging or due to a pathological process e.g., Alzheimer's disease, drugs such as piracetam, vinpocetine, Ginkgo biloba or donepezil which improve learning and memory are frequently prescribed (McDaniel et al., 2003[39]). Piracetam was the first of the so called "nootropics", a term introduced by Giurgea (1973[23]) to indicate this category of drugs that enhance memory, facilitate learning and protect memory processes against conditions which tend to disrupt them. The drug is a pyrrilodine derivative (2-oxo-1-pyrrolidine acetamide), a cyclic derivative of gamma-aminobutyric acid, which has been shown to facilitate learning and preserve memory from disruption under different experimental conditions (Gouliaev and Senning, 1994[24]; He et al., 2008[28]). Piracetam enhances recovery from post-stroke aphasia (Kessler et al., 2000[35]), improves cognitive performance in the elderly (Waegemans et al., 2002[83]) and in patients undergoing bypass surgery (Holinski et al., 2008[30]). The drug possesses rheological, blood flow promoting effects (Müller et al., 1999[45]) and mitochondrial membrane stabilizing properties (Keil et al., 2006[34]; Leuner et al., 2010[37]).

Vinpocetine (vinpocetine-ethyl apovincaminate), a synthetic derivative of the alkaloid vincamine from periwinkle (Vinca minor) is phosphodiesterase (PDE) inhibitor, selective for PDE1 (van Staveren et al., 2001[81]) and a blocker of voltage-gated Na+ channels (Sitges et al., 2006[69]) which improves learning and memory deficits in rodents (DeNoble, 1987[ref:13]). The drug is widely used to improve the cognitive functions of patients with cerebrovascular disease and chronic cerebral hypoperfusion (Horváth, 2001[ref:32]) and to decrease the risk of transient ischemic attacks and strokes in patients with chronic cerebrovascular insufficiency (Valikovics, 2007[ref:80]). Vinpocetine is a potent vasodilator at the cerebral vascular bed, increasing cerebral blood flow and regional cerebral glucose uptake (Vas et al., 2002[82]).

Standardized extracts from the leaves of Ginkgo biloba contains 24 % ginkgo-flavone glycosides and 6 % terpenoids (ginkgolides, bilobalide). Ginkgo biloba extracts are widely used to improve cognition and memory in cerebral insufficiency and mild cognitive impairment (Bäurle et al., 2009[5]). The extract also showed a beneficial effect in reducing the total and negative symptoms of chronic schizophrenia when used as an add-on therapy to antipsychotic medication (Singh et al., 2010[68]). Ginkgo biloba prevented stress- and corticosterone-induced impairments of spatial memory and reduced neuronal loss in hippocampus (Takuma et al., 2007[76]; Walesiuk and Braszko, 2009[84]). Ginkgo biloba also showed neuroprotective and therapeutic effects in experimental cerebral ischemia (Saleem et al., 2008[64]) and attenuated the toxin-induced neurodegeneration of the nigrastriatal pathway (Rojas et al., 2008[60]). The extract has been shown to possess vasodilator and blood flow enhancing (Chung et al., 1999[10]) as well as antioxidant and free radical scavenging properties (Rojas et al., 2008[60]).

The main neurochemical deficit seen in Alzheimer's disease is reduced acetylcholine content and choline acetyltransferase activity (the enzyme synthesizing acetylcholine) in the nucleus basalis of Meynert and the hippocampus (Whitehouse et al., 1982[85]). Acetylcholinesterase inhibitors (AChEI) e.g., donepezil are thus being used for the symptomatic treatment of Alzheimer's disease to enhance cholinergic neurotransmission indirectly, by inhibiting the enzyme which hydrolyses acetylcholine (Musiał et al., 2007[48]).

The aims of the present study were therefore to: (1) study the behavioral alterations associated with long-term administration of cannabis preparations on spatial learning and memory; (2) investigate the effect of memory enhancing drugs piracetam, vinpocetine, Ginkgo biloba and the acetylcholine esterase inhibitor donepezil for their possible modulation of the cannabis-induced memory alterations; (3) detect alterations in the level of the brain neurotransmitters dopamine, serotonin (5-HT) and noradrenaline; (4) assess the effect of cannabis administration on oxidative stress markers in brain since oxidative stress has been implicated in memory impairment (Clausen et al., 2010[11]). In addition, the effect of cannabis on liver oxidative stress and liver enzymes under the different experimental conditions was studied. A total extract from Cannabis sativa was used based on the fact that the effect of the whole plant which is abused by humans differs from that of THC in view of its content of other cannabinoids, terpenoids and flavonoids (Russo and McPartland, 2003[63]).

In the present study the effect of cannabis extract with known Δ⁹-THC content on working memory and on oxidative stress in brain and liver of mice was investigated. The repeated daily administration of the extract at doses corresponding to 5, 10 and 20 mg Δ⁹-THC/kg was associated with impaired learning and memory when tested in the water maze test. Mice given cannabis spent more time to locate the hidden platform than their untreated counterparts. The effect was most evident with the initial administration of the extract and maintained throughout the study. There was a clear dose-related response during the three trials of the test. Scopolamine, an anticholinergic drug, impaired memory performance in the water maze test which is in accordance with other studies (Smith et al., 2002[70]). Mice treated with scopolamine and cannabis needed significantly more time to locate the hidden platform than the scopolamine only-treated group, suggesting exacerbation of the cognitive impairing effects of scopolamine by cannabis. Studies have shown that cannabis preparations or their primary psychoactive constituent Δ⁹-THC affect memory processing in humans and in experimental animals. Prose recall and source memory were poorer in daily users with high THC levels in hair (Morgan et al., 2011[42]). In healthy volunteers, Δ⁹-THC (2.5, and 5 mg) given intravenously disrupted immediate and delayed word recall, impaired performance tests of distractibility, verbal fluency and working memory (D'Souza et al., 2004[15]). Similarly, orally given Δ⁹-THC led to acute impairment of attentional functioning and working memory (Roser et al., 2008[ref:61]). Impaired memory performance has been reported in mice after the oral administration of hashish extract in dosages corresponding to 1, 5, and 10 mg Δ⁹-THC/kg (Frischknecht et al., 1985[22]) and after inhalation of marijuana smoke (Niyuhire et al., 2007[49]) and in rats following intraperitoneal injection of Δ⁹-THC-rich extracts (2 or 5 mg/kg) (Fadda et al., 2004[20]). Cannabidiol, another constituent of the plant Cannabis sativa, that does not possess psychoactive properties, appeared to antagonize the memory impairing effect of Δ⁹-THC in the same extract (Fadda et al., 2004[20]), though not the working memory deficits induced by scopolamine and dizocilpine (Fadda et al., 2006[21]). The cannabis induced memory alterations are likely to be mediated by Δ⁹-THC-inducecd activation of cannabinoid CB1 receptor in brain. Delta (9)-THC-rich extracts impaired memory performance in rats (Fadda et al., 2004[20]), while Δ⁹-THC-induced spatial memory impairment was reversed by cannabinoid CB(1) receptor antagonist, suggesting that the effect of Δ⁹-THC is mediated through cannabinoid CB(1) receptors (Egashira et al., 2012[18]). Studies also suggested the involvement of prefrontal dopamine receptors (Rodrigues et al., 2011[59]) as well as μ- and κ-opioid receptors (Egashira et al., 2012[18]) in Δ⁹-THC-induced disruption of spatial working memory. Cannabis effect on learning and memory processes probably involves depolarization-induced suppression of excitatory mechanism in the CA1 area of hippocampus (Ebrahimpour et al., 2010[17]).

Of the important findings in the present studies is the effect of cannabis on neurotransmitter levels in brain. It is evident that the repeated administration of cannabis extract increased brain levels of serotonin, noradrenaline and dopamine. Other studies have shown increased dopamine and noradrenaline release in rodent brain by THC in several regions of the brain, including striatal, nucleus accumbens and prefrontal areas (Muntoni et al., 2006[46]; Robledo et al., 2007[58]). Moreover, increased release of norepinephrine in the rat frontal cortex was observed after systemic administration of the synthetic cannabinoid agonist WIN 55,212-2 (Oropeza et al., 2005[50]). In healthy human subjects, THC induced dopamine release in the human striatum (Bossong et al., 2009[7]). As for 5HT, Δ⁹-THC attenuated methylenedioxymethamphetamine (MDMA)-induced decreases in 5-HT levels and in serotonin transporter (SERT) binding in the frontal cortex, parietal cortex, and striatum (Shen et al., 2011[66]) while stimulating CB-1 decreased the effect of citalopram on increasing serotonin levels in the prefrontal cortex (Kleijn et al., 2011[36]). In contrast, rats injected with the synthetic cannabinoid HU210 exhibited increased 5HT1A receptor density and mRNA expression in the CA1 region of the hippocampus (Zavitsanou et al., 2010[88]). These effects of cannabis on brain monoamines could account for the cognitive and attention deficits and anxiety reactions seen in cannabis users. Cannabis sativa, however, contains many constituents including over 70 different cannabinoids (Hollister, 1988[31]). Thus cannabis can result in different effects from those of Δ⁹-THC alone. For example, cannabigerol a non-psychoactive constituent behaved as a potent alpha (2) adrenoceptor agonist and a 5HT(1A) antagonist (Cascio et al., 2010[9]). Meanwhile, cannabidiol, another non-psychoactive increased extracellular dopamine levels in nucleus accumbens (Murillo-Rodríguez et al., 2011[47]). While Δ⁹-THC acts as a CB1 and CB2 receptor partial agonist, Δ⁹-tetrahydrocannabivarin, behaves either as a CB1 antagonist or, at higher doses, as a CB1 receptor agonist (Pertwee, 2008[54]). The present study has also provided evidence that the nootropic drugs piracetam, vinpocetine, Ginkgo biloba or donepezil were capable of modulating neurotransmitter levels in mice treated with cannabis. Serotonin was decreased by all drugs. Noradrenaline was normalized by vinpocetine and donepezil though increased by piracetam and Ginkgo biloba. Meanwhile, piracetam, vinpocetine, Ginkgo biloba decreased the cannabis-induced increments in dopamine. Studies have shown that most nootropics influencing cognitive mechanisms affect neurotransmitters levels in brain. Vinpocetine through selective blockade of voltage-sensitive presynaptic Na⁺ channels inhibits the transporter-mediated release of monoamine neurotransmitters. The drug impairs the vesicular storage of dopamine as well (Trejo et al., 2001[77]). Ginkgo biloba has been shown to increase central dopamine, noradrenaline (Yoshitake et al., 2010[87]) and 5-HT (Blecharz-Klin et al., 2009[6]) levels. Piracetam increases dopamine in cortex and striatum (Wustmann et al., 1990[86]; Stancheva and Alova, 1994[74]). In one study, piracetam abolished the amnestic effect of 6-hydroxydopamine and restored to control values the noradrenaline level in the frontal cortex and hippocampus (Gouliaev and Senning, 1994[24]). Donepezil administration was associated with a significantly increased release of dopamine (Liang and Tang, 2006[38]), noradrenaline (Shearman et al., 2006[65]) in cortex or hippocampus, but decreased extracellular serotonin levels (Shearman et al., 2006[65]).

The effect of cannabis on brain oxidative stress, an imbalance between free radicals generation and antioxidant defense mechanisms is important in view of the evidence linking cellular damage arising from increased oxidative stress to neuronal degeneration and decline in cognitive function associated with normal aging or caused by different pathological states (Dröge, 2003[14]; Zhou et al., 2008[89]; Halliwell, 2006[26]). The present study suggests that cannabis extract alters the oxidative balance in the brain. This however, appears to be in favor of reducing lipid peroxidation. Malondialdehyde an index of lipid peroxidation activity (Gutteridge, 1995[25]), is decreased and there was significant increase of GSH, an important antioxidant defense system, especially with the highest dose of cannabis examined. Nitric oxide in brain is also decreased by cannabis administration. The significance of this finding is yet to be determined. Nitric oxide is an important signaling molecule involved in neurotransmission and in maintaining vascular tone via its vasodilator properties. Nitric oxide can be also detrimental to neural tissue if generated in excess by inflammatory cytokines due to the action of inducible nitric oxide synthase (Moncada et al., 1991[41]). In mice treated with the anticholinergic drug scopolamine so as to produce memory deficits, lipid peroxidation, though not significantly increased by scopolamine is decreased to normal value by the highest dose of cannabis. Reduced glutathione which is decreased by scopolamine is restored by cannabis administration. The extract also lessened the increase in brain nitric oxide by scopolamine. Still brain glucose decreased after cannabis administration under these conditions. These findings confirm observations in the saline-treated mice. These results are intriguing in view of the studies reporting neuroprotective effects for certain cannabinoids under experimental substances (Pazos et al., 2012[52]). These antioxidant properties of the cannabis extract can be explained by the fact that cannabis extract is not merely Δ⁹-THC, but rather a mixture of over 600 different chemical compounds. Cannabinoids, a group of C21 terpenophenolic compounds uniquely produced by Cannabis sativa plant are considered to be the main biologically active constituents of the Cannabis sativa plant, of which currently at least 70 are known (Brenneisen, 2006[8]). Other cannabinoids such as cannabidiol (CBD), cannabinol (CBN) and tetrahydrocannabivarin (THCV) can result in different effects from those of Δ⁹-THC alone (Pertwee, 2008[54]). Some cannabinoids were found to exert neuroprotective effects (Pazos et al., 2012[52]). The beneficial effects of cannabis extract on brain lipid peroxidation can also be ascribed to the flavonoids it contains. One important finding, however, was the decreased brain glucose by cannabis administration, thereby, impairing cerebral energy metabolism. This impairment of brain energetics, can explain the effect of the extract on memory function. The oxidative status of hepatic tissue was also examined. In contrast to the effect of repeated cannabis extract on oxidative stress in brain, the extract did no affect lipid peroxidation or reduced glutathione levels in the liver tissue.

The current study investigated the effect of piracetam, vinpocetine, Ginkgo biloba extract or donepezil on the memory impairment induced by cannabis extract in the water maze test. These drugs are widely prescribed to enhance memory function due to mild cognitive impairment or Alzheimer's disease (McDaniel et al., 2003[39]). Piracetam, vinpocetine, Ginkgo biloba or donepezil co-administered with cannabis resulted in significantly shorter latencies compared with mice treated with only cannabis, which indicated improved learning and memory or in other words improvement of the cognitive impairing effects of Cannabis sativa by these drugs. This effect of cannabis decreased by memory enhancing drugs with piracetam resulting in the most-shorter latency compared with the cannabis. Biochemically, cannabis altered the oxidative status of the brain with decreased MDA, increased GSH, but decreased nitric oxide and glucose. In cannabis-treated rats, the level of GSH in brain showed further increase after vinpocetine and donepezil and was markedly elevated after Ginkgo biloba. Piracetam restored the decrease in glucose and nitric oxide by cannabis. Of the memory enhancing drugs used in an attempt to counteract the effect of cannabis on memory performance, piracetam resulted in the most-shorter latency compared with the cannabis. Piracetam resulted in significantly shorter latency compared with vinpocetine over the 4 weeks of the study. The effect of piracetam was more evident during the second and third trials of the test. Moreover, only piracetam significantly increased brain glucose compared with the cannabis-treated group. The drug has been reported to reverse regional depressions in glucose metabolism in the rat hippocampus after scopolamine treatment (Piercey et al., 1987[56]). When given to patients with Alzheimer's disease, piracetam significantly improved regional glucose metabolism in most cortical areas (Heiss et al., 1988[29]). Interestingly, GSH is increased by donepezil, vinpocetine and Ginkgo biloba. In other studies, vinpocetine and piracetam increased GSH in different brain areas (Abdel-Salam et al., 2011[2]). In vitro, vinpocetine displayed scavenging activity at human therapeutic serum concentration (Horvath et al., 2002[33]). Studies indicated an antioxidant effect for Ginkgo biloba extracts. In mice, Ginkgo biloba attenuated 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced neurodegeneration of the nigrastriatal pathway, most likely due to an inhibitory effect against oxidative stress (Rojas et al., 2008[60]). In vitro, Ginkgo biloba was able to block Aβ (1-42)-induced cell apoptosis, reactive oxygen species accumulation and mitochondrial dysfunction (Shi et al., 2009[67]).

In summary, this study examined cannabis induced memory deficits in the Morris water maze, using mice. Cannabis was found to impair performance as measured by the time taken to locate a submerged platform compared to controls. The positive control scopolamine also slowed performance and added to the cannabis deficit when the two agents were tested together. The impaired memory performance is likely to involve decreased brain glucose availability as well as alterations in brain monoamine neurotransmitter levels. The cannabis induced performance deficit was attenuated by the memory enhancing drugs, with piracetam resulting in the most-shorter latency compared with the cannabis and restoring the cannabis-induced decrease in brain glucose and nitric oxide.