4. Discussion
Long-term application of morphine and other opioid narcotics induces addiction composing physical and psychological dependence. In the present study, physical dependence developed following bi-daily subcutaneous injections of morphine in mice for 7 consecutive days, and application of naloxone induced withdrawal signs including shakes, jumps, genital licks, fecal excretion and body weight loss. In contrast, bi-daily subcutaneous injections of BAA up to 300 μg/kg/day for 7 days did not induce any physical dependence, consistent with the previous finding in which daily subcutaneous BAA did not induce jumping responses following nalorphine challenge (Tang et al., 1986). In addition, single subcutaneous injection of BAA alleviated naloxone-induced withdrawal signs in morphine physical dependence mice. In doses ranging from 30 to 300 μg/kg in morphine dependence mice, BAA caused a dose-related inhibition of abrupt withdrawal signs, typically shakes and body weight loss with ED50 values of 74.4 and 105.8 μg/kg respectively. Consistently, the BAA analog lappaconitine was reported to alleviate morphine and cocaine physical dependence (Qu and Qu, 1994). On the other respect, daily subcutaneous injections of morphine but not BAA (300 μg/kg/day) for 5 consecutive days induced remarkable CPP acquisition with high conditioning scores. Single subcutaneous injection of BAA (300 μg/kg) entirely abolished morphine-induced CPP acquisition. All these results indicate that BAA does not induce withdrawal signs and CPP acquisition, moreover, it markedly alleviates morphine-induced both physical and psychological dependence.
Opiates such as morphine and heroin, act at the mesolimbic dopamine pathway projecting from ventral tegmental (VTA) to NAc (Patyal et al., 2012). Opiates drugs effectively stimulate dopamine release in NAc within 1 hour after administering intracerebroventricularly or locally into VTA (Murphy et al., 1996; Sebastian et al., 2016). Hippocampal input to the shell of NAc (NAcSh) is important for driving NAc activity which may play a role in reward behaviors (LeGates et al., 2018). Thus, it can be seen that NAc and hippocampus, located around the cerebroventricular area, are important sites for drugs to produce addiction. Dynorphin A regulates the activity of dopamine neurons by acting on κ-opioid receptors in mesolimbic, NAcSh , prefrontal cortex and VTA that have been implicated in drug abuse liability (Meshul and McGinty, 2000; Volkow et al., 2009). Extracellular dopamine levels in NAc are increased in mice lacking κ-opioid receptors, revealing that the dynorphin/κ-opioid receptor system inhibits dopamine neurotransmission and counteracts the reward of opioid drugs of abuse (Chefer et al., 2005; Shippenberg et al., 2007b). Hippocampal CA1 has been demonstrated in reward-association of learning and memory especially in the CPP model (Fanselow and Dong, 2010; Riahi et al., 2013). The intra-CA1 administration of D1/D2 receptor antagonists SCH23390 and sulpiride, and microinjection of the κ-opioid receptor agonist U50,488H into NAcSh, resulted in a decrease in morphine conditioning scores (Assar et al., 2016). In this study, we explored whether the dynorphin A/κ-opioid receptor system in NAc and hippocampus was closely associated with BAA-attenuated morphine physical dependence and psychological dependence. Subcutaneous injection of BAA in morphine-treated mice stimulated the expression of dynorphin A in NAc and hippocampus at 1 hour after injection, which was in agreement with the time-course of its anti-addictive effects. The results are parallel to the previous findings in which intrathecal and subcutaneous injection of BAA, bullatine A and lappaconitine stimulated the expression of dynorphin but not β-endorphin in the spinal cord (Li et al., 2016a; Li et al., 2016b).
The notion is further supported by the following intervention injections through the cerebroventricular route which is located around NAc and hippocampus. It was previously reported that single intravenous injection of dynorphin A attenuated withdrawal symptoms of morphine physical dependence (Takemori et al., 1993). The present study further demonstrated that intracerebroventricular injection of the dynorphin A antiserum totally eliminated systemic BBA-inhibited morphine withdrawal signs and CPP acquisition. In addition, the highly specific κ-opioid receptor antagonist GNTI, given intracerebroventricularly, also entirely eliminated systemic BBA-inhibited morphine physical and psychological dependence. The results are in agreement with the previous publications that the κ-opioid receptor agonist salvinorin A punished self-administration of cocaine and remifentanil in monkeys (Freeman et al., 2014), and that addition of the κ-opioid receptor agonist U69,593 to fentanyl produced a proportion-dependent decrease of fentanyl self-administration in rats (Negus et al., 2008). In agreement, GNTI and other specific κ-opioid receptor antagonist nor-BNI (but not the μ- or δ-opioid receptor antagonist) blocked the antinociceptive effects of BAA (Li et al., 2016b; Huang et al., 2020b), bullatine A (Huang et al., 2016) and lappaconitine (Sun et al., 2018) in the rodent models of neuropathic pain, bone cancer pain, inflammatory pain and visceral pain.
Cumulative evidence indicates that microglia in brain particularly in NAc, VTA and hippocampus are influenced and activated by chronic exposure to abuse drugs such as opiates and alcohol (Miguel-Hidalgo et al., 2002; Hutchinson et al., 2009; Coller and Hutchinson, 2012). Drug abuse activates microglia and produces a large number of inflammatory factors, which affect synapse reconstruction, chemical changes in neural signal transduction and phagocytosis of apoptotic neurons, and ultimately regulate the dopamine reward signaling pathway and enhance drug dependence and addiction (Kovacs, 2012; Garaschuk and Verkhratsky, 2019). On the other hand, recent studies have shown that microglia have an alternative activation state or protective state, which activates anti-inflammatory cascades and exhibits neuroprotection and antinociception (Hu et al., 2012; Fan et al., 2015; Wu et al., 2017; Wu et al., 2018). Our present study provides additional evidence that BAA stimulates microglia to express dynorphin A to attenuate drug dependence and addiction. Subcutaneous injection of BAA stimulated dynorphin A expression only in microglia and not in astrocytes and neurons in NAcSh and hippocampal CA1, identified by immunofluorescence staining of dynorphin A with microglial cellular marker Iba-1, astrocytic marker GFAP and neuronal marker NeuN. The results are in agreement with the previous finding in which injection of the BAA analog bullatine A specifically stimulated microglial (but not astrocytic or neural) expression of dynorphin A in the spinal cords of neuropathic rats (Huang et al., 2016). We further demonstrated that intracerebroventricular injection of the microglial activation inhibitor minocycline entirely blocked systemic BAA-inhibited morphine-induced physical and psychological dependence. Consistently, the antinociceptive effects of BAA and its analogs bullatine A and lappaconitine were also blocked by intrathecal injection of minocycline in the rat models of pain hypersensitivity (Huang et al., 2016; Sun et al., 2018a; Huang et al., 2020b). These results highlight that stimulation of microglia expresses and releases dynorphin A, in contrast to proinflammatory cytokines and neurotrophins, and inhibits morphine physical and psychological dependence and pain hypersensitivity.
Drug addiction is a chronically and relapsing disorder characterized by compulsive seeking and taking drugs regardless of the adverse effects it could be cause (Shippenberg et al., 2007b). The main treatment approach for drug addiction is detoxification and relieves withdrawal symptoms, and methadone substitution is most commonly used. This treatment, however, is just provided to inpatients in well-equipped management institutions due to its own physical dependence (Kampman and Jarvis, 2015; Tran et al., 2017). Clonidine is a non-opioid detoxifying agent and diminishes withdrawal symptoms via activation of adrenergic α2-receptors, and is superior to other treatments since it is not addictive and does not produce euphoria, although the side-effect of postural hypotension limits its clinical use (Wilson and DiGeorge, 1993). On the other respect, psychotherapy for addiction should be emphasized conceptually. However, only a few patients are unfortunately willing to undergo it, and many patients are still carving for addictive substances after improving withdrawal symptoms, which eventually leads to relapse (Woody et al., 1983; Arevalo et al., 2008). Thus, the approaches at present to solve opioid addiction especially psychological dependence are limited. Our current study provides a solid pharmacological base for BAA to alleviate both morphine physical and psychological dependence at the animal level. In addition, oral administration of the BAA analog lappaconitine over 6 days reduced or eliminated withdrawal symptoms such as yawing, shedding tears, chilliness, mydriasis and restlessness in drug addict patients whose duration of heroin or opium addiction ranged from 1 to 4 years (Qu and Qu, 1994). Taken together, all these results indicate that BAA is a unique and promising clinical development candidate to treat opioid addiction, especially for psychological dependence. The data also suggest that targeting microglial expression and secretion of dynorphin A is a potential approach for the opioid drug addiction treatment.