As humans, we continuously seek out things that can produce pleasurable feelings. One of the few ways to attain this feeling of euphoria or pleasure is by consumption of chemical substances. However, continuous consumption of these substances can lead to development of dependence towards them and this is more commonly known as addiction. Currently, the cost of social and economic impacts due to drugs of abuse addiction sums up to more than $740 billion annually (National Institute on Drug Abuse, 2017). Despite this, the clear neurophysiological mechanisms underlying development and progression of addiction is still unknown.
Addiction can be defined as repeated self-administration of alcohol or other drugs (AOD’s) despite knowledge of adverse medical and social consequences and attempts to abstain from AOD use (Robert & Koob, 1997). Initial intake of drug may be due influenced by genetic, psychosocial or environmental factors however, subsequent doses is most probably caused by action on drug on the brain to induce the drug-seeking behaviour. Addiction is comprised of three stages which are ‘binge/intoxication’, ‘withdrawal/negative affect’, and ‘preoccupation/anticipation’ as seen in Figure 1 (Koob &Volkow, 2010).
Figure 1. The Three stages of addiction and brain areas each stage associates with (Herman & Roberto, 2015)
‘Binge/intoxication’ refers to the consistent intake of drug after initial dose which may cause decrease in dopamine release after each intake due to sensitization. This will then lead to ‘withdrawal/negative affect’ stage where absence of drug will cause a decrease in dopamine causing anxious and restless feelings resulting in the craving or ‘preoccupation/anticipation’ stage. Two major factors known to modulate these behavioural changes are reinforcement where a stimulus increases the chance of response and neuroadaptation, the process by which neuronal structures change in response to drug exposure. Modulation of these factors motivates initial response to a drug and formation of long-term craving however, relapse is thought to be caused by permanent neuroadaptations that will cause discomfort during withdrawal (Robert & Koob, 1997).
To this day, treatments for addiction hasn’t been particularly successful and one of the most common problem is relapse. As stated previously, this is thought to be caused by neuroadaptations therefore highlighting the importance of understanding the underlying mechanisms for successful treatments. In this review article, we provide current breakthroughs on the mechanisms leading to neuroadaptations in nucleus accumbens (NAc), a key region in addiction studies.
Drug-Induced Structural Plasticity in the Nucleus Accumbens and its Input Regions
Drugs has been known to induce structural plasticity of dendrites since 1997 (Robinson & Kolb, 2004; Russo et al., 2009; Dietz et al., 2009; Russo et al. 2010). Since then, researches on various drugs of abuse have shown to induce a structural change in the brain’s reward circuitry such as opiates decreasing number of NAc medium spiny neurons (MSN) in contrast to stimulants which increases NAc MSN numbers. Early withdrawal after exposure to chronic cocaine induces expression of N-methyl-D-aspartate (NMDA) glutamate receptors at MSN surface causing silent synapse formation and long-term depression (LTD). Prolonged withdrawal will cause retraction of the NMDA receptors to be replaces by α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) glutamate receptors changing structure of the spine to become mushroom shaped and promoting long-term potentiation (LTP). This change however can be reversed easily by a challenge dose of cocaine (Figure 2) (Russo et al.,2010).
Figure 2. Cocaine-induced synaptic and structural plasticity (Russo et al., 2010)
At a molecular level, this is caused by regulation of actin cytoskeleton via scaffolding proteins as well as GTPases and they are activated by transcription factor ΔFosB and cyclic AMP response element binding protein (CREB) which induces spinogenesis (Heiman et al., 2008; Kim et al., 2009). On the other hand, ΔFosB also regulates cyclin dependent kinase 5 (Cdk5) (Kumar et al., 2005) as well as nuclear factor κB (NFκB) (Russo et al., 2009) where both molecules play a role in cocaine-induced spine formation showing the major role of ΔFosB in cocaine-induced structural changes. The only paradox here is that both opiates and cocaine induce similar behavioural phenotypes (Russo et al., 2010) as well as drug administration and withdrawal symptoms although they have completely opposite effect on the NAc MSNs. A few possible hypothesises include changes in synaptic plasticity having a bidirectional property where a change in both directions result in similar behavioural phenotypes, decrease in neuronal complexity in one brain area is compensated by a strengthening of another area or another pathway is induced by opiates which suppresses spinogenesis (Lüscher et al.,2008; Ikemoto et al., 2007; Belin et al., 2008). Further studies are needed to confirm these.
NAc can be sub-divided into two regions: the core and the shell. Both regions have different input and output projections (Zahm, 2000) and thought to play different roles in reward pathway (Ito et al., 2004). Recent studies have also reported on different dendritic compartments specifically the proximal and distal (Spruston, 2008). Cocaine regulation of dendritic spines can only be observed in thin, highly motile spine (Kasai et al., 2010) which were thought to be relevant to learning (Moser et al., 1994; Dumitriu et al., 2010) and addiction (Shen et al., 2009; LaPlant et al., 2010). Cocaine exposure caused an increase in spine density in the shell region but a decrease in proximal MSNs in the core region which is seen to be far more enduring (Dumitriu et al., 2012). This enduring change in core reciprocates the idea that shell is involved in addiction development while core in the learning of the addiction or long-term potentiation (Di Chiara, 2002; Ito et al., 2004; Meredith et al., 2008). In a study done by Kourrich and Thomas (2009), however, showed an increase in core MSNs and a decrease in shell MSNs raising the possibility that spine regulation may be compensating the changes in MSNs or spine regulation may be causing a homeostatic tuning of MSNs excitability. Few studies showed homeostatic increase in MSNs excitability following spine downregulation (Azdad et al., 2009; Ishikawa et al., 2009; Huang et al., 2011) supporting the latter but the clear association between both processes is still unknown. A possible mechanism behind the selective downregulation of core MSNs could be dopamine since cocaine withdrawal decreases dopamine levels (Parsons et al., 1991; Baker et al., 2003). Further support to this could be from the higher convergence of the dual glutamatergic and dopaminergic pathways in the core (Zahm and Brog, 1992).
Other than that, a study has found that there is also structural plasticity in input regions to the nucleus accumbens. These inputs include the ventral tegmental area (VTA) which is thought to be important for rewarding stimuli, ventral hippocampus (vPHC) for encoding contextual information, basolateral amygdala (BLA) for relaying emotional context and medial prefrontal cortex (mPFC) providing operational value (Nestler, 2004, Russo & Nestler, 2013). There are two types of medium spiny neurons in the nucleus accumbens specifically dopamine receptor-1-expressing (D1-MSN) and dopamine-receptor-2-expressing (D2-MSN) where D1-MSN is responsible for rewarding stimulation compared to aversive in D2-MSN (Lobo et al., 2010). After cocaine exposure, there was an increase in spine density in BLA and vHPC neurons firing to D1-MSN (Barrientos et al., 2018; Russo et al., 2010) and a decrease in spine projection in mPFC. Since BLA encodes emotional context, the initial increase in spines after exposure may be necessary in forming affective response to cocaine. On the other hand, spines in vHPC only increase during challenge cocaine after prolonged withdrawal to allow enough time to strengthen VHPC-NAc pathway to provide contextual representation of drug seeking and decrease in spines in mPFC may reflect the loss of top-down control in addicted individuals (Barrientos et al., 2018).
Regulation of Pathways by Different Sub-Regions of the Nucleus Accumbens
NAc receive dopamine (DA) projections from the ventral tegmental area (VTA) (Björklund and Dunnett, 2007, Ikemoto, 2007, Morales and Margolis, 2017) and this pathway play a major role in motivated behaviours, reinforcement learning and reward processing (Hamid et al., 2016; Salamone and Correa, 2012; Schultz, 2016; Watabe-Uchida et al., 2017). Like any other process, there are negative feedback pathways to balance the projections and prevent overexpression of DA. This arise from various structures (Matsui et al., 2014) but recent studies show that NAc is the main source of this inhibitory input (GABAergic input) (Beier et al., 2015; Watabe-Uchida et al., 2012). There were few conflicting results on this with studies suggesting inputs from NAc to VTA to be disinhibiting (Bocklisch et al., 2013; Chuhma et al., 2011; Xia et al., 2011) and a recent study addressing that NAc synapse onto VTA GABA as well as DA neurons via GABA-A receptor (GABAAR) and GABA-B receptor (GABABR) respectively (Edward et al., 2017). This, however, also projects a different result compared to the study done by Paladini in 1999 where inhibitory responses from the striatum to DA neurons were blocked by GABA-A antagonist hinting at pathway mediated by GABA-A instead.
In this particular study by Hongbin et al. in 2017, the shell component of the NAc is further subdivided into medial shell (NAcMed) and lateral shell (NAcLat). D1-MSN in the NAcMed is found inhibiting NAcMed-projecting DA neurons via GABAAR while NAcLat-projecting DA neurons via GABABR. D1-MSNs in the NAcLat, on the other hand, projects onto VTA GABA to result in disinhibition of NAcLat-projecting DA neurons (Figure 3).
Figure 3. Feedback Loops between Sub-regions of Nucleus Accumbens and Dopamine Neurons in the Ventral Tegmental Area (Hongbin et al., 2018)
This provides a clear indication of specific sub-region and its role in inhibition suggesting that previous controversial studies may be due to absence of sub-region specificity. VTA GABAAR activation will cause an increase in DA transmission to NAcLat due to disinhibition of NAcLat-projecting DA neurons as well as decrease in projections to NAcMed due to direct suppression which agrees with Edward et al. (2017) where he stated NAc inhibition of VTA GABA and DA neurons via GABAAR and GABABR respectively. Therefore, NAc plays an important role in removal of tonic inhibition of DA neurons from VTA GABA which could then result in activation of the rewards pathway (Paladini and Roeper, 2014, Watabe-Uchida et al., 2017). Further research also shows projections from NAc shell to VTA to be relapse promoting while projections to lateral hypothalamus to be extinction promoting (Gibson et al., 2018) once again highlighting the importance of this pathway in addiction and the possibility of preventing relapse in future treatments.
Inhibitory Role of Glycinergic Transmission in Nucleus Accumbens
GABAergic transmission is widely known as the main inhibitory transmission neurotransmitter especially in the central nervous system. That brings us to the major role it plays in the rewarding pathway specifically in the NAc. Recent study (Muñoz et al., 2018), however, has found presence of synaptic and non-synaptic glycinergic transmission in the NAc and is thought to have an inhibitory role in the rewards system due to its inhibitory role in other brain regions causing problems in sensory processing (Buckwalter et al., 1994; Mülhardt et al., 1994; Ryan et al., 1994) as well as seizure-like symptoms (Koch et al., 1996; Rees et al., 2001). The glycinergic transmission in NAc, unlike its transmission in other brain regions, is sensitive to propofol but resistant to ethanol suggesting ethanol being mediated by non-synaptic glycine receptors (Maguire et al., 2014; Förstera et al., 2017). Its input to D1-MSNs suggest its involvement in the reward-related learning and this can be backed up by the importance of D1-MSNs in maintenance of self-administration of Propofol (Lian et al., 2013) as well as increase in ΔFosB in NAc by Propofol doses (Xiong et al., 2011).
In conclusion, there are many complex mechanisms underlying neuroadaptations of the NAc which contributes to addiction. Further research is required not only in the NAc but also pathways in the limbic system, the main reward system as well as on different drug types to get a better understanding of the neuronal mechanisms underlying these neuroadaptations.
...(download the rest of the essay above)