Implementing preventive strategies would represent a major improvement in schizophrenia outcomes. Given that only 30–40% of ultra-high risk individuals convert to full psychotic state (Gee and Cannon, 2011), the scarce clinical investigations of preventive treatments face important ethical and technical challenges (Lambert et al., 2016). Pre-clinical studies, therefore, play a fundamental role, since they allow investigating early interventions with the clear distinction between subjects who will develop behavioral abnormalities from those who will not. Nonetheless, few pre-clinical studies have addressed this important clinical issue, with the limitation of not evaluating potential side effects or focusing only on psychosis-like behaviors (Piontkewitz et al., 2009, 2011; Meyer et al., 2010; Herrmann et al., 2014; Peres et al., 2016a).

In this scenario, CBD emerges as a promising intervention: in addition to displaying an antipsychotic profile, CBD does not induce the motor side effects commonly associated with the use of antipsychotic drugs (Zuardi et al., 1991; Leweke et al., 2012; Dos-Santos-Pereira et al., 2016). On contrary, CBD can attenuate the development of parkinsonian and/or dyskinetic motor behaviors induced by several agents (El-Alfy et al., 2010; Gomes et al., 2013; Peres et al., 2016b; Sonego et al., 2016). Supported by these data, we first investigated CBD’s preventive potential in the poly I:C neurodevelopmental model of schizophrenia. We presented, for the first time, evidence of CBD’s preventive action (Peres et al., 2016a). Nevertheless, we were able to evaluate CBD’s effect only on hyperlocomotion. Here, using the SHR strain, we were able to confirm CBD’s beneficial effect on hyperlocomotion and extend it to other schizophrenia-like behavioral abnormalities: the deficits in PPI and in contextual fear conditioning.

Interestingly, SHRs – mainly young animals – have been suggested as an animal model to study attention deficit/ hyperactivity disorder (ADHD), based on theirs impulsivity and hyperactivity (Sagvolden et al., 2005). Nonetheless, in these studies, authors use Wistar Kyoto rats (WKYs) as the control strain, contrary to our works using WRs. WKYs might be an inappropriate control, considering that they display inactivity and depressive-like behavior (Overstreet, 2012). Moreover, the pharmacological validity of SHRs as a model for ADHD is controversial: studies show that the administration of the drugs used to treat ADHD, such as methylphenidate or amphetamine, is not able to improve SHRs ADHD-like behaviors, or even potentiates them (Amini et al., 2004; van den Bergh et al., 2006; Yang et al., 2006; Bizot et al., 2007; Barron et al., 2009; Calzavara et al., 2011). These concerns notwithstanding, a comorbidity between schizophrenia and adult ADHD has been described (Donev et al., 2011), and at high-risk young relatives of schizophrenia patients may display comorbid ADHD (Keshavan et al., 2003; Oner and Munir, 2005). In addition, schizophrenia patients or high-risk individuals with ADHD perform worse in neuropsychological tests, when compared to those without ADHD (Keshavan et al., 2003; Oner and Munir, 2005; Donev et al., 2011). Of relevance for this work, diagnosis of ADHD in peri-adolescence is associated with the emergence of schizophrenia symptomatology later in life (Kim-Cohen et al., 2003; Hennig et al., 2017). In this scenario, our data may also indicate that CBD displays a preventive potential for high-risk individuals with ADHD as a comorbidity or for those that develop schizophrenia and display a previous history of ADHD.

Increased locomotor activity is linked to augmented dopamine transmission in the mesolimbic system, and therefore is used to model the positive symptoms of schizophrenia (Lipska and Weinberger, 2000). Peripubertal administration of haloperidol, clozapine, risperidone or fluoxetine prevents the emergence of increased locomotor activity after the administration of psychostimulant drugs in the poly I:C model (Piontkewitz et al., 2009, 2011; Meyer et al., 2010), suggesting that these compounds could prevent the positive symptoms. Nevertheless, some of these early treatments had detrimental effects on the response to psychostimulant drugs of control animals (Meyer et al., 2010; Piontkewitz et al., 2011).

Contextual fear conditioning is a memory task with an emotional component. In the SHR strain, the contextual fear conditioning deficit is specifically reversed by the administration of antipsychotic drugs (Calzavara et al., 2009), being suggested as a model for the cognitive symptoms of schizophrenia. Contextual fear conditioning deficits are also seen in other animal models for schizophrenia (Duffy et al., 2010; Gill et al., 2017; Latusz et al., 2017). Our findings therefore point to beneficial effects of CBD on the cognitive impairments and are in accordance with a recent work showing that prolonged treatment with CBD improves the deficits on recognition and working memory in the poly I:C model (Osborne et al., 2017). It is noteworthy that Osborne et al. (2017) performed the treatment with CBD in a later stage (late adolescence/adulthood) and assessed the behaviors during treatment. Regarding preventive treatments, administration of antipsychotic drugs are able to prevent deficit on latent inhibition, a cognitive abnormality, in the poly I:C model (Piontkewitz et al., 2009, 2011; Meyer et al., 2010). On the other hand, adolescent administration of olanzapine impairs working memory in rats (Milstein et al., 2013).

PPI is a task used to assess the functioning of sensorimotor gating – a physiological process that filters sensory information, preventing overload and cognitive fragmentation (Powell et al., 2012). PPI deficits are seen in schizophrenia patients (Braff et al., 2001) and in several animal models for this disorder (Swerdlow et al., 2008). Data show that PPI deficits are associated with impaired functional status and with the presence of positive symptoms (Swerdlow et al., 2006; Kumari et al., 2008). Peripubertal administration of clozapine, but not haloperidol, prevents the emergence of PPI impairments in the poly I:C model (Meyer et al., 2010). In fact, early treatment with haloperidol promotes PPI deficits in the control offspring (Meyer et al., 2010). Here, we demonstrate, for the first time, that peripubertal treatment with CBD improves PPI deficits in an animal model for schizophrenia, without having detrimental effects on control animals.

Peripubertal treatment with CBD did not improve the reduced social interaction displayed by the SHRs. Indeed, while acute administration of CBD restores SHRs’ deficits on PPI and on contextual fear conditioning (Levin et al., 2012, 2014), it does not increase social interaction (Almeida et al., 2013). Beneficial effects of prolonged treatment with CBD on social interaction are seen in the poly I:C model (Osborne et al., 2017), however, in the presence of CBD administered during late adolescence/adulthood and using a higher dose (10 mg/kg, twice daily). The doses used here were chosen based on our previous work showing preventive effect (CBD was administered from days 30–60 and the beneficial effect on behavior was observed in adulthood) of 1 mg/kg CBD in the poly I:C model (Peres et al., 2016a). We chose to evaluate here a range of doses, considering that CBD effects usually follow an inverted U-shaped dose-response curve (Guimaraes et al., 1990; Nazario et al., 2015). In accordance, all beneficial effects of CBD were seen at the lowest dose (0.5 mg/kg). Nonetheless, we cannot disregard that using a higher dose of CBD could result on improvement of the social interaction deficit, as seen by Osborne et al. (2017).

In view of the low conversion rate and the risks of a peripubertal treatment, the search for preventive approaches must consider their safety. Peripubertal treatment with CBD increased social interaction as well as contextual fear conditioning response in WR control animals. Both behaviors can be modulated in this strain by acute administration of anxiolytic or anxiogenic compounds (File and Seth, 2003; Calzavara et al., 2009). Having in order that CBD also presents an anxiolytic profile, its long-term effect on WR strain (observed in the absence of the drug – 1 month after the end of the treatment) could be the result of a modification in anxiety levels. Nevertheless, social interaction and fear conditioning respond on opposite directions to anxiety drugs (contrary to what is seen here). It must be highlighted that CBD did not modify any other behavior in the control strain, indicating that it does not promote detrimental behavioral side effects. These data point to a safe use of chronic treatment with CBD during peri-adolescence in the individuals that will not develop schizophrenia in the adulthood.

It could be argued that a long-term modulation of anxiety by CBD might be related to its beneficial effects seen in the behavioral abnormalities displayed by the SHR strain. Nevertheless, the fear conditioning deficit of SHRs is improved by antipsychotic but not by anxiolytic or anxiogenic drugs (Calzavara et al., 2009). Conversely, anxiolytic drugs increase social interaction, an effect not seen in the present study. Additionally, PPI does not seem to be modulate by acute, chronic or deprivation of treatment with benzodiazepines (Rigdon and Viik, 1991; De Ross et al., 2008).

Still concerning the safety of a long-term treatment with CBD during peri-adolescence, we performed a screening of potential metabolic and motor side effects. Previous studies show that CBD does not modify physiological parameters, nor affects embryonic development or vitality of non-tumor cell lines (Bergamaschi et al., 2011; Iffland and Grotenhermen, 2017). CBD seems to be well tolerated, with the most reported side effects being changes on appetite, diarrhea, and tiredness (Bergamaschi et al., 2011; Iffland and Grotenhermen, 2017; Boggs et al., 2018; McGuire et al., 2018). Moreover, CBD has been reported to display a safe, or even beneficial, profile for the development of motor side effects (Peres et al., 2018).

Accordingly, here CBD did not alter animals’ body weight gain, serum levels of triglycerides or glycemia. Also, catalepsy and oral movements – motor side effects commonly seen with chronic antipsychotic use – were not induced. This is the first study to demonstrate that prolonged treatment with CBD during development does not promote detrimental motor or metabolic side effects in rodents. However, further studies are needed to warrant the safety of a treatment with CBD during a young age. It must be also considered that CBD can inactivate or induce cytochrome P450 isozymes, depending on the drug regimen, which may lead to important drug interactions (Bergamaschi et al., 2011; Iffland and Grotenhermen, 2017).

With respect to the antipsychotic profile of CBD in the SHR model, our previous works show that CBD restores the SHRs impairments in contextual fear conditioning and PPI (Levin et al., 2012, 2014), but does not modify SHRs locomotor activity or social interaction (Almeida et al., 2013). These data indicate that acute administration of CBD on adulthood ameliorates only SHRs cognitive deficits. In this respect, CBD differs from both typical and atypical antipsychotic drugs. Here, peripubertal treatment with CBD displayed beneficial effects on a broader spectrum of behaviors, considering that in addition to restore/prevent the emergence of cognitive impairments, CBD prevented hyperlocomotion. Concerning side effects, neither the motor nor the metabolic alterations – related to the chronic use of typical and atypical antipsychotics, respectively – were seen.

Cannabidiol treatment began at 30 days of age, when SHRs already display diminished social interaction, but do not display increased locomotor activity or impaired PPI (Niigaki et al., unpublished). Therefore, CBD was able to halt the emergence of hyperlocomotion and PPI deficits in the SHR strain. The underlying mechanisms whereby CBD exerts these preventive beneficial effects are likely multifaceted, given that CBD presents several molecular targets (Ibeas Bih et al., 2015). Our data suggest involvement of the serotoninergic system. Here, treatment with CBD increased the ratio 5-HIAA/serotonin on PND 61 and enhanced 5-HIAA levels on PND 90 in prefrontal cortex, suggesting an increased turnover of serotonin. In accordance, Linge et al. (2016) demonstrated that treatment with CBD for 14 days increases the serotonin release in the ventral-medial prefrontal cortex after a challenge dose of CBD. This effect might be related to the 5-HT_1A agonism proposed for CBD (Russo et al., 2005) because the 5-HT_1A antagonist WAY100635 prevents the increased serotonin release induced by acute administration of CBD (Linge et al., 2016). Interestingly, the administration of WAY100635 attenuates the increase in prefrontal cortex dopamine levels induced by atypical antipsychotic drugs with either low or high affinity for 5-HT_1A (Ichikawa et al., 2001, 2002). This increase in prefrontal dopamine levels is suggested to be relevant to the beneficial effects of these agents (McCreary and Newman-Tancredi, 2015). As a future perspective, investigating the role of 5-HT_1A receptors on the preventive effects of CBD by using 5-HT_1A antagonists would be of great relevance.

It must be noticed that the beneficial behavioral effects of CBD were observed 1 month after the interruption of treatment. The development of schizophrenia is strongly associated with changes in the neurodevelopmental course (Piper et al., 2012). Therefore, the preventive effects of CBD observed here are probably the results of a beneficial action on the neurodevelopmental process. In accordance, multiple studies describe CBD’s antioxidant, anti-inflammatory and neuroprotective actions. In pigs, CBD prevents the neuronal damage and the increases in excitotoxicity, oxidative stress and inflammation induced by a hypoxic-ischemic brain injury protocol (Pazos et al., 2013). The antioxidant effects of CBD are also seen in rat models of binge alcohol consumption, sepsis, epilepsy, Huntington’s disease, Parkinson’s disease, and mania (Hamelink et al., 2005; Cassol et al., 2010; Campos et al., 2017). CBD has also been reported to increase (Magen et al., 2010; Reus et al., 2011; Barichello et al., 2012; Campos et al., 2015; Mori et al., 2017; Sales et al., 2018) or decrease (ElBatsh et al., 2012) BDNF levels in rodents’ brains. Here, prolonged peripubertal treatment with CBD did not modify the BDNF levels in prefrontal cortex or striatum. In agreement with our findings, other authors report no changes on BDNF levels in prefrontal cortex (Reus et al., 2011; Campos et al., 2012, 2015) or striatum (ElBatsh et al., 2012) after repeated treatment with CBD (7 or 14 days). Moreover, in some studies CBD-induced alterations in BDNF levels are not associated with its behavioral effect (Magen et al., 2010; Reus et al., 2011). The search for neuroprotective targets other than BDNF that might underlie the beneficial CBD’s preventive effect is promising and currently under investigation by our group.

Another mechanism of action that might be involved in CBD’s preventive effects is the increase in the levels of endocannabinoids. Data suggest that anandamide, the main endocannabinoid, may display a protective role in schizophrenia: anandamide levels are inversely correlated with the psychotic symptoms in patients (Giuffrida et al., 2004). Moreover, Leweke et al. (2012) reported that treatment with CBD in schizophrenia patients results in a significant clinical improvement, which is positively associated with increase in anandamide levels. As a future perspective, we aim to quantify anandamide levels on brain regions of WRs and SHRs treated with CBD during peri-adolescence.

One important limitation of our work is that we evaluated the preventive effects of CBD only in male animals. This choice was based on the fact that the behavioral and pharmacological validation of the SHR strain as an animal model for schizophrenia was performed only with males (Calzavara et al., 2009, 2011; Levin et al., 2011). Considering that, in humans, females display a late age of schizophrenia onset (Donoghue et al., 2014), seem to require lower doses of antipsychotic drugs and to display higher incidence of side effects (Lange et al., 2017), the evaluation of pharmacological strategies in females is of extreme relevance. We are currently starting a behavioral and pharmacological characterization of adult SHR females to extend the validation of the model to female rats and allow future investigations.

In summary, the present study reinforces and extends the beneficial and safe preventive effects of peripubertal treatment with CBD on halting the emergence of behavioral abnormalities that mimic the positive and cognitive symptoms of schizophrenia. The translation of the present data to the clinical context indicates the possibility of diminishing the severe suffering of individuals that develop schizophrenia. Strengthening the relevance of our findings, while the positive symptoms of schizophrenia are devastating but treatable by antipsychotic drugs (with all the risk of developing significant side effects), the cognitive deficits associated with the disorder do not benefit from any current clinical treatment. Our innovative findings further support the notion that CBD has neuroprotective effects and may be a potential compound to prevent schizophrenia. Future double-blind, long-term, controlled clinical trials with large samples of people at high risk for psychosis would be necessary and opportune to further investigate these possibilities.