Major depressive disorder is one of the world's greatest public health problems, with a lifetime incidence in the United States of more than 12% in men and 20% in women. The etiology of the disease is not fully understood, but it is currently accepted that major depressive disorder is caused by the cumulative impact of genetic background, adverse events in childhood, and ongoing or recent stress (Nestler and Carlezon 2006). A comprehensive theory for the neurobiological basis of depression emerging in recent years is the neurotrophic hypothesis, suggesting that a reduction in neuroplasticity and the expression of brain derived neurotrophic factor (BDNF) could contribute to depression and that antidepressants mediate their therapeutic benefit, in part, by increasing levels of this factor in the hippocampus (Russo and Nestler 2013).
Depressive disorders are often accompanied by substance abuse disorder, with about one-third of patients with major depression diagnosed with comorbid substance use disorders (Davis et al. 2008). In laboratory animals, depressive-like symptoms have been observed following chronic drug use and withdrawal (Anraku et al. 2001; Perrine et al. 2008), whereas enhancement of addictive behaviors was found in some animal models of depression (Holmes et al. 2002; Lin et al. 2002). Several theories attempted to explain this comorbidity between drug exposure and depression (Davis et al. 2008). For example, the self-medication hypothesis suggests that psychiatric patients, and especially depressed patients, use drugs to alleviate the symptoms of their illness (Markou et al. 1998). On the other hand, it has also been suggested that substance abuse may increase vulnerability to depression through behavioral and neurophysiological alterations (Chinet et al. 2006).
These theories, however, usually refer to extensive drug use, which is characterized by addiction, and is accompanied by effects of withdrawal (Volkow et al. 2011). For this reason, it is difficult to dissociate the ‘pure’ effects of mere exposure to drugs from the effects of full-scale addiction and especially withdrawal. In addition, much less attention has been given to the effects of drug use in lower scales, what is usually termed ‘recreational drug use’ (Albertson 2013). This pattern of drug use is growing in number and is becoming an international problem, even without leading to drug addiction and dependence (Wood et al. 2009). The few studies that examined the effects of subchronic drug exposure on depressive characteristics have focused mainly on the effects of withdrawal during the first 24–48 h following drug secession (Barr et al. 1999; Harrison et al. 2001). Long-term effects of subchronic drug exposure were shown primarily in complex learned behaviors (Klein et al. 2007; Bassareo et al. 2013) or under restricted conditions of limited access to reward (Avena and Hoebel 2003), and these studies usually did not examine the neurobiological mechanisms related to depression.
One such mechanism, which might mediate the effects of drug exposure on mood is their effect on BDNF expression (Graham et al. 2007), which appears to play a key role in the pathophysiology of depression (Russo and Nestler 2013). For example, reduction in BDNF protein levels in the hippocampus was observed in both postmortem analysis of depressed patients and in animal models of depression (Duman and Monteggia 2006). On the other hand, increased BDNF levels (or BDNF signaling) was observed in the pre-limbic cortex (PLC) (Taliaz et al. 2013), ventral tegmental area (Christoffel et al. 2011), and the Nucleus Accumbens (NAc) (Eisch et al. 2003). In addition, BDNF itself has been shown to up-regulate GluA1, subunit of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor (Narisawa-Saito et al. 1999), which has been shown to be altered both in association with depression (Martinez-Turrillas et al. 2002) and drug use (Carlezon and Nestler 2002), making it another target of interest relating to drug effects on mood.
In laboratory animals, subchronic administration of both cocaine (Ujike et al. 1996; Velazquez-Sanchez et al. 2010) and heroin (Shabat-Simon et al. 2008) cause a gradual increase in the psychomotor effect of the drug (i.e., psychomotor sensitization). Here, we used a similar exposure protocol to investigate the long-term effects of subchronic exposure to these drugs on characteristics of depression. Sprague–Dawly rats received daily drug injections (10 days for cocaine, 5 days for heroin, once a day), and then tested in a series of commonly used depression-like behavioral tests, representing a range of depressive symptoms. Following behavioral testing, we measured BDNF and GluA1 protein levels in the PLC, NAc, and in the dorsal and ventral Hippocampus regions. Finally, in order to test whether drug exposure aggravates vulnerability to depression, we conducted an additional experiment, in which animals were administered with the same subchronic regimen of cocaine or heroin, and then challenged with the chronic mild stress (CMS) procedure (Willner et al. 1992). This procedure is known to induce anhedonia (especially as measured by reduction in sucrose preference) (Willner 1997), whereas this study examined whether this behavioral effect can be blocked or enhanced by subchronic exposure to cocaine or heroin.
- Top of page
- Materials and methods
- Acknowledgments and conflict of interest disclosure
The impact of drug use on mood was previously studied in the context of chronic addiction and withdrawal [for example see (D'Souza and Markou 2010)], without much emphasis on the hazardous long-lasting effects of short-term drug exposure and recreational drug use. In the current study, we assessed the long-term influence of subchronic exposure to cocaine or heroin on established behavioral paradigms representing different aspects of depressive-like behavior. A subchronic regimen was chosen as opposed to chronic prolonged regimen (Chartoff et al. 2012), which is known to promote depressive symptoms at least during the acute withdrawal period, as well as lasting neuroadaptations (Koob and Le Moal 2001).
The behavioral assessments performed during 2 weeks after subchronic exposure to heroin (5 days) or cocaine (10 days) in the present study revealed various depressive-like outcomes. The significant reduction in home-cage activity of the cocaine pre-treated rats compared to saline pre-treated rats during the dark phase of the day can represent psychomotor retardation, a known depressive symptom (Pulvirenti and Koob 1993). In a previous study, a similar experiment involving 9 days of cocaine administration caused a disruption in circadian feeding behavior (Giorgetti and Zhdanova 2000). Such reduction in activity, however, was not apparent in the heroin pre-treated rats in our experiment. A possible explanation for this dichotomy might be that cocaine can affect circadian rhythm-related gene expression (Uz et al. 2005) and related behaviors (McClung et al. 2005) following a single or a subchronic administration, whereas similar effects of heroin and other opiates were only observed following chronic administration or during withdrawal in human addicts [e.g., (Chen et al. 2006; Li et al. 2009)].
On the other hand, only rats in the heroin group displayed longer periods of immobility during the forced swim test, which represents behavioral despair, another hallmark feature of depression (McArthur and Borsini 2006). This result is in line with previous data, demonstrating the influence of opioid receptors knockdown on behavior during the forced swim test in mice (Filliol et al. 2000). Another study had shown elevations in immobility time even following a shorter exposure to cocaine for 5 days (Filip et al. 2006); however, in that study a lower dose was used and behavior was examined only a few days after the final drug administration.
The sucrose preference test did not show any difference between drug and saline pre-treated groups (unless exposed to the chronic mild stress procedure), suggesting that subchronic exposure to heroin or cocaine by itself does not induce anhedonic-like behavior. Indeed, numerous studies have demonstrated the anhedonic effects of cocaine and heroin, although they relate primarily to states of prolonged addiction or during acute withdrawal (Markou and Koob 1991; Dalley et al. 2005; Zijlstra et al. 2009). On the contrary, some studies demonstrated an opposite effect, where repeated subchronic exposure to cocaine resulted in higher salience given to sucrose reward (Klein et al. 2007) and similar exposure to amphetamine generated higher sucrose consumption (Avena and Hoebel 2003). However, these findings were obtained in the context of either complex learning paradigms and/or limited access to sucrose, and therefore might be the result of a compulsive reaction to a stressful condition (Koob 2009).
Upon examination of BDNF protein levels in the brains of the animals following behavioral procedures, we noticed partial alterations that are parallel to findings from established animal models of depression. In the present study, increases in BDNF levels were found in the pre-limbic cortices of animals from both heroin and cocaine groups when compared with the saline group. Increases in BDNF levels in the PLC have been shown previously even following a single injection of cocaine, morphine, and methamphetamines (Le Foll et al. 2005), however these effects were transient, and did not persist for weeks as demonstrated here. Similar increase in BDNF in this subregion of the prefrontal cortex was observed in several animal models of depression (Lee et al. 2006; Fanous et al. 2010; Taliaz et al. 2013). In addition, we found a decrease in BDNF levels in the dorsal Hippocampus of the heroin pre-treated animals when compared with saline pre-treated animals, whereas a slightly smaller decrease observed in the cocaine group was not statistically significant. This result seems to be in contrast to the effect shown by Filip et al. (2006) where 5 days of cocaine induced an increase in BDNF mRNA in the dorsal hippocampus. However, it should be noted that cocaine has been shown to produce opposing effects on BDNF mRNA and protein levels (Fumagalli et al. 2007), suggesting post-translational alterations. Most importantly, such a decrease in dorsal Hippocampus BDNF levels has been shown in numerous models of depression and is the key feature in the ‘neurotrophic hypothesis of depression’ (Krishnan and Nestler 2010). Moreover, similar subchronic administration of morphine resulted in reduction in hippocampal neurogenesis in rats (Eisch et al. 2000), in almost the same degree of severity as was caused by long-term heroin self-administration. Reduction in hippocampal neurogenesis has been suggested to be involved in the pathophysiology of depression, as antidepressant treatments increase the levels of neurogenesis in this region both in human (Boldrini et al. 2012) and in laboratory animals (Malberg et al. 2000), with BDNF serving as a major factor in the regulation of this process (Taliaz et al. 2010).
The lack of differences in GluA1 levels analyzed, despite the BDNF alterations and the regulatory effect of BDNF on GluA1 expression (Fortin et al. 2012), suggests potential compensatory mechanisms and that subchronic drug exposure was not sufficient to induce long-term alterations in glutamate-related plasticity in these regions. Another possible explanation is that changes occurring in GluA1 levels in response to subchronic drug administration may be specific to synaptic compartments (Boudreau et al. 2007; Schumann and Yaka 2009), whereas we analyzed whole tissue homogenates that could have masked such effects. Further research is needed to decipher the synapse-specific effects of subchronic drug administration on glutamatergic receptors. Dopaminergic receptors may also be involved in the behavioral effects observed in the current study, as several lines of evidence associate major depression with a state of reduced dopamine (DA) transmission, including compensatory up-regulation of D2 receptors (Dunlop and Nemeroff 2007). As a result of the substantial role of the DA system in abused drug effects (Koob and Volkow 2010), it should be interesting to examine the possible changes in expression of D1 and D2 receptors in relation to the depression-like behaviors induced by subchronic cocaine or heroin exposure.
Taken together, results from the first experiment indicate a partial depressive-like effect of subchronic drug exposure, expressed both in terms of behavior and brain neurochemistry. The main feature of depression that was not affected by the drugs was anhedonia, represented by the sucrose preference test. We therefore designed an additional experiment, in which animals were exposed to subchronic cocaine or heroin, as in the first experiment, and then challenged by the 4-week chronic mild stress paradigm, a known model of environmentally-induced anhedonia (Willner et al. 1992). Indeed, chronic stress induced significant reductions in sucrose preference expressed by both cocaine and heroin pre-treated groups when compared with the saline pre-treated group. Furthermore, the reductions in sucrose preference were accompanied by reductions in BDNF protein levels in the dHC, a neurochemical alteration which embodies the key feature in the ‘neurotrophic hypothesis of depression’, as stated above. These reductions suggest that drug subchronic exposure had instigated greater sensitivity to chronic stress and its depressive-like effects, both behavioral and neurochemical. These lasting effects of drugs combined with stress extend numerous studies which have demonstrated how drugs of abuse exhort enhanced effects under conditions of acute or chronic stress [for example (Shaham and Stewart 1994) and (Ahmed and Koob 1997)]. It should be noted, however, that any drug effects may be dose-specific and/or alter as a function of time. In this study, we have monitored lasting effects (during 3 weeks) after subchronic daily exposure for one dose per drug, and with a measurement of each behavioral and neurochemical parameter at a single time-point. Therefore, further research is needed to determine the wide-range effects of these drugs, using different doses and monitoring the time-course of each behavioral and neurochemical effect.
In conclusion, our study had presented evidence that subchronic daily exposure to a standard dose of either cocaine or heroin can precipitate certain aspects of depression in laboratory rats. In addition, such exposure can generate higher vulnerability to depression in general, as demonstrated by greater anhedonic-like response to chronic stress. These results demonstrate the severe lasting effects of ‘pure’ drug exposure, which do not require prolonged drug use or the emergence of withdrawal symptoms. This also highlights the importance that should be given to prevention and education amongst recreational and small-scale drug users, in addition to ‘heavy’ drug addicts.