Neurobiological and Neurophysiological Mechanisms Underlying Nicotine Seeking and Smoking Relapse

Approximately 16% of American adults smoke traditional cigarettes, and the prevalence of noncombustible tobacco use is rising dramatically. The use of electronic cigarettes (e-cigarettes) has tripled in middle- and high-school students in the United States in the last 4 years [1, 2]. Tobacco smoking remains the leading cause of preventable death in the United States, accounting for approximately 1 out of every 5 deaths annually [3]. On average, smokers die 13.5 years earlier than nonsmokers [4] and cigarette smoking-related illnesses cost more than USD 300 billion annually in the United States (i.e., USD 156 billion in lost productivity and USD 170 billion in health care expenditures) [5]. While smoking cessation at any age significantly increases life expectancy [6], only 3% of smokers are able to quit successfully on their own [7]. The majority of smokers who relapse do so during the first week of a quit attempt [8, 9]. Unfortunately, current FDA-approved medications for smoking cessation have modest efficacy and are associated with high rates of relapse [10]. Thus, there is a critical need to develop novel and more effective smoking cessation pharmacotherapies. Successful development of novel smoking cessation medications requires translational studies that advance and accelerate preclinical findings into pilot clinical trials [11, 12]. Preclinical studies utilizing rodent models of addiction-like behaviors are critical towards identifying molecular substrates that could serve as targets for new smoking cessation pharmacotherapies. Indeed, this drug discovery process can be guided by an improved understanding of the neurobiology that mediates smoking relapse.

Nicotine is the principal psychoactive chemical in tobacco that mediates the reinforcing effects of cigarette smoking and other forms of tobacco [13-17]. Therefore, the vast majority of biomedical research investigating the neurobiological and neurophysiological bases of smoking cessation focuses primarily on nicotine – although more recent studies have begun to investigate the role of other tobacco constituents and byproducts [18-23]. Intravenous drug self-administration in laboratory animals and humans is a behavioral paradigm used commonly to measure the reinforcing efficacy of nicotine [24, 25]. Nicotine self-administration has the highest degree of face validity of all animal models of nicotine addiction, primarily because it mimics voluntary tobacco consumption in humans [25]. Therefore, the drug self-administration model is a critical component of key studies examining the neural mechanisms underlying voluntary nicotine taking and seeking in rodents [24, 26, for review see,27, 28]. While nicotine self-administration studies in human subjects typically involve cigarette-smoking behavior, the majority of nicotine self-administration studies in rodents entail intravenous administration immediately following an operant response (i.e., lever press or nose poke). Intravenous self-administration is used primarily because this route simulates the rapid rise in arterial nicotine and rapid distribution of nicotine into the brain that occurs via the typical pulmonary route of exposure in humans [29]. Plasma nicotine levels are similar in animals self-administering nicotine and human tobacco smokers, further validating this animal model of nicotine addiction [30]. In addition, the predictive validity associated with altered nicotine self-administration and the effectiveness of smoking cessation medications appears to be relatively high [11]. For example, all three FDA-approved treatments for smoking cessation (i.e., varenicline, nicotine replacement therapies, and bupropion) decrease nicotine self-administration in rodents [31-33]. Finally, it is thought that nicotine self-administration has good construct validity as an emerging literature indicates that similar changes in dopaminergic, cholinergic, GABAergic, glutamatergic, serotonergic, and cannabinoid systems mediate the reinforcing effects of nicotine in both humans and laboratory animals [25].

Smoking relapse is typically modeled in animals using the nicotine self-administration/extinction/reinstatement paradigm. Briefly, nicotine seeking can be reinstated after a period of nicotine self-administration and the subsequent extinction of the drug-reinforced behavior. The same stimuli that precipitate smoking relapse (i.e., stress, exposure to drug-associated stimuli and re-exposure to the drug itself) during periods of abstinence in humans can be used to reinstate nicotine-seeking behavior in rodents [34-36]. For example, following extinction of nicotine self-administration, systemic injections of relatively low doses of nicotine and/or cues previously paired with nicotine taking reinstate operant responding in the absence of nicotine reinforcement in rodents [37, 38]. The validity of the nicotine reinstatement model as an in vivo medication screen appears promising for smoking relapse [39]. Indeed, varenicline, an FDA-approved smoking cessation pharmacotherapy, attenuates the reinstatement of nicotine seeking in rats [40, 41] as well as smoking relapse in humans [42]. However, despite robust effects on nicotine reinstatement in rats, the efficacy of smoking cessation pharmacotherapies (i.e., varenicline and bupropion) in humans remains modest. This is likely due to the complexity of smoking behavior and relapse in humans as well as an incomplete understanding of the neurobiological mechanisms underlying relapse. It is important to note that the reinstatement paradigm, which is used to model relapse in human smokers, does not require nicotine dependence, as measured by somatic signs of physical withdrawal. Therefore, it is possible that some of the neuroadaptations that underlie smoking relapse are distinct from the mechanisms mediating physical withdrawal from nicotine. This presents a distinct advantage of preclinical research aimed at identifying specific neuroadaptations that promote or facilitate nicotine-seeking behavior and smoking relapse. However, the lack of physical dependence also highlights a potential limitation, as nicotine seeking in humans is likely due to both avoidance of withdrawal syndrome and relapse mechanisms. Thus, this is a consideration of preclinical studies of nicotine seeking that requires additional attention.

To better understand the mechanisms mediating nicotine addiction-like behaviors, including relapse, several researchers have investigated nicotine-mediated changes in neurophysiology and plasticity using in vitro biochemical, electrochemical, and electrophysiological preparations [43-47]. The goal of this article is to review an emerging literature examining the neurotransmitter systems and the plasticity-dependent neurobiological and neurophysiological mechanisms underlying nicotine-seeking behavior in rodents, particularly as it relates to relapse in human smokers. We will also highlight important areas of consideration for future investigations. A more complete understanding of how nicotine-induced neuroadaptations alter neural circuits to produce nicotine-seeking behavior could lead to the development of novel, targeted pharmacotherapies for smoking cessation.

Nicotine reinforcement is mediated through different neurotransmitter systems, and many investigations of nicotine-seeking behavior (in animal models) and relapse behavior (in human smokers) have focused on neurotransmitter mechanisms of drug-taking and drug-seeking behavior. The reinforcing effects of nicotine are mediated, in large part, by activation of neuronal nicotinic acetylcholine receptors (nAChRs) in the mesoaccumbens dopamine pathway [26, 48, 49]. Indeed, systemic administration of nicotine increases extracellular dopamine levels in the nucleus accumbens (NAc) [50-52], specifically in the shell subregion [53]. Moreover, direct infusions of nicotine into the ventral tegmental area (VTA) [52, 54] and NAc [54, 55] also increase dopamine release in the accumbens. Furthermore, acute and chronic nicotine exposures produce dynamic changes in dopamine levels in the NAc that are associated with nicotine reinforcement [53, 56]. The ability of nicotine to activate mesoaccumbens dopamine projections relies on β2-containing nAChRs expressed on VTA dopamine neurons [57] and α7* nAChRs (* indicates the possible presence of additional, unidentified subunit[s]) expressed on glutamatergic afferents in the VTA [43]. There is clear evidence that neuronal α4β2* nAChRs (* indicates the possible presence of additional subunits) play a critical role in nicotine reinforcement [26, 58, 59], which also supports the utility of the ongoing development of novel smoking cessation pharmacotherapies targeting this receptor population. It is also clear that glutamate transmission in the brain plays a critical role in nicotine addiction [60-62]. Specifically, nicotine exposure increases extracellular glutamate levels in the VTA [63] and promotes excitatory tone in the VTA, both of which can enhance dopamine cell firing to facilitate nicotine reinforcement. Given this literature, many studies of nicotine seeking and smoking relapse have focused on dopaminergic, cholinergic, and glutamatergic mechanisms [36, 60]. However, more recent studies reveal novel roles for other neurotransmitter and neuropeptide mechanisms. In this section, we will review how these systems contribute to nicotine seeking in rats, and possibly to relapse behavior in humans.

As noted above, nicotine effects on the mesolimbic reward system, and on dopamine transmission in particular, are known to mediate nicotine reinforcement. There are five dopamine receptor subtypes classified as D1-like (D1 and D5) or D2-like (D2, D3, and D4), based on sequence homology and pharmacology [64-66]. Systemic administration of either a D1-like or D2-like dopamine receptor antagonist is sufficient to reduce cue-induced reinstatement of nicotine seeking [67]. While these findings indicate that dopamine receptors play an important role in nicotine seeking, they do not identify roles for individual dopamine receptor subtypes in the reinstatement of nicotine seeking. Importantly, however, human genetics studies have identified single nucleotide polymorphisms in the dopamine D4 receptor gene that correlate with cue reactivity in human smokers [68-70]. The systemic administration of a dopamine D4 receptor antagonist attenuates both cue- and nicotine-induced reinstatement of drug seeking [71]. These findings suggest that dopamine D4 receptors could serve as potential targets for novel smoking cessation pharmacotherapies. In addition to these dopamine D4 receptor-mediated effects, systemic administration of selective dopamine D3 receptor antagonists reduces drug-seeking behavior elicited by either an acute priming injection of nicotine [72, 73] or re-exposure to conditioned stimuli that were paired with nicotine taking prior to withdrawal [74]. Repeated systemic nicotine administration produces locomotor sensitization, which is associated with increased expression of dopamine D3, but not D1 and D2, receptor subtypes in the NAc shell [75]. Dopamine D3 receptors are expressed at high levels in the NAc shell, specifically in medium spiny GABAergic projection neurons [76-78]. However, it is not clear if similar D3 mechanisms in the accumbens mediate nicotine seeking during abstinence. Recent evidence indicates that pharmacological blockade of dopamine D3 receptors in the basolateral amygdala (BLA) and lateral habenula, but not the NAc, regulate the reinstatement of nicotine-seeking behavior [74]. Thus, future studies are required to identify the precise role of dopamine receptor subtypes in neurocircuits known to mediate nicotine seeking.

Endogenous fatty acid amides function as ligands of alpha-type peroxisome proliferator-activated receptors (PPAR-α) and are metabolized by fatty acid amide hydrolase (FAAH). Inhibition of FAAH prevents nicotine-induced elevations in extracellular dopamine levels in the NAc shell and reduces nicotine-induced excitation of VTA dopamine neurons [79, 80]. Moreover, pharmacological inhibition of FAAH attenuates nicotine seeking, and these effects are due, in part, to activation of PPAR-α [80, 81]. Consistent with the effects of FAAH inhibitors, PPAR-α agonists decrease nicotine-induced excitation of VTA dopamine neurons, thereby blocking nicotine-induced dopamine release in the NAc shell, and attenuate nicotine- and cue-induced reinstatement of nicotine-seeking behavior [82, 83]. These data indicate that activating PPAR-α is sufficient to prevent nicotine seeking by decreasing dopaminergic tone.

Several pieces of evidence point to a strong role for glutamatergic mechanisms underlying nicotine-seeking behavior. For example, extracellular glutamate levels in the NAc increase following cue-induced nicotine reinstatement [84]. Additionally, nicotine-seeking behavior is associated with changes in the cystine-glutamate exchanger, a heterodimeric protein complex that regulates extrasynaptic glutamate levels [85]. Specifically, withdrawal following voluntary nicotine taking reduces expression of the catalytic subunit (xCT) of the cystine-glutamate exchanger in the VTA and NAc [86]. Together, these results suggest that reversing or restoring nicotine-induced decreases in extracellular glutamate levels in the brain may prevent the reinstatement of nicotine-seeking behavior. However, strategies aimed at increasing extracellular glutamate levels in the brain during nicotine withdrawal have produced mixed effects on nicotine-seeking behavior. For example, systemic administration of N-acetylcysteine, a cystine pro-drug that increases the activity of the xCT resulting in increased extracellular glutamate levels, has been shown to attenuate [87] or have no effect [88] on the reinstatement of nicotine seeking. With regard to the clinical literature, N-acetylcysteine does not produce a therapeutically robust response in human smokers. For example, N-acetylcysteine decreases self-reported number of cigarettes smoked per day but does not produce concurrent reductions in breath carbon monoxide levels and fails to prevent relapse in human smokers [86, 89]. Thus, the efficacy of N-acetylcysteine in reducing nicotine seeking in both rodents and humans appears weak at best.

At the receptor level, withdrawal following repeated nicotine taking is associated with increased glutamate receptor expression. Specifically, nicotine withdrawal increased protein expression of AMPA receptor GluA1 subunits and NMDA receptor GluN2A and GluN2B subunits, as well as reduced protein expression of the glutamate transporter-1 in the NAc core [84, 86]. Pharmacological inhibition of GluN2A- and GluN2B-containing NMDA receptors in the NAc core decreases cue-induced reinstatement of nicotine seeking [84]. Interestingly, a nonselective NMDA receptor antagonist increases cue-induced nicotine seeking when infused directly into the NAc core [90]. It is hypothesized that this increase in drug seeking when NMDA receptors are inhibited is due, in part, to activation of non-NMDA glutamate receptors (e.g., mGlu5 receptors) following cue-induced increases in extracellular glutamate levels in the NAc core [90]. Consistent with this hypothesis, administration of the mGlu5 receptor antagonist MPEP attenuates cue-induced nicotine-seeking behavior [91]. Interestingly, systemic administration of the mGlu1 receptor antagonist EMQMCM also attenuates cue- and nicotine-induced reinstatement of nicotine seeking [92]. Due to adverse effects of AMPA and NMDA receptor antagonists in humans, attention has focused on developing type I mGlu (mGlu1) receptor antagonists for smoking cessation [36]. In addition to these mGlu1 effects, repeated nicotine exposure and subsequent withdrawal reduce mGlu2/3 receptor function in the VTA and NAc [93]. Furthermore, systemic administration of the mGlu2/3 receptor agonist LY379268 attenuates cue-induced reinstatement of nicotine seeking [93]. However, mGlu2/3 receptor agonists are nonselective and tolerance develops with repeated administration [93]. For these reasons, focus has shifted to selective positive allosteric modulators (PAMs) of mGlu2 receptors. Recent studies show that systemic administration of a PAM of mGlu2 receptors reduces nicotine- and cue-induced reinstatement of nicotine-seeking behavior, and that these effects are associated with reduced dopamine release in the NAc [94, 95]. Based on these studies, a phase II clinical trial examining the efficacy of the mGlu2 receptor PAM AZD8529 in human smokers is currently underway [96], with preliminary results forthcoming.

With respect to nAChR subtypes, a growing literature continues to delineate a role for α4β2* and α7* nAChRs in nicotine-seeking behavior and suggests that these two subpopulations of nAChRs may play differential roles in drug- versus cue-primed reinstatement of nicotine seeking. Indeed, varenicline, currently the best-in-class treatment for smoking cessation, was developed based on its action as a partial agonist at α4β2* nAChRs [97]. Interestingly, varenicline appears to have differential effects on the reinstatement of nicotine seeking depending upon the stimulus used to elicit drug seeking. For example, varenicline attenuates nicotine seeking following an acute priming injection of nicotine [40]. However, varenicline has no effect on cue-induced reinstatement of nicotine seeking [40] and even enhances cue-induced nicotine seeking at high doses [98]. These results suggest that α4β2* nAChRs mediate nicotine primed, but not cue-induced, reinstatement of drug seeking. This is further supported by a study showing that cue-induced reinstatement of nicotine seeking is attenuated following pretreatment with an α7*, but not α4β2*, nAChR antagonist [99]. Interestingly, chronic nicotine exposure promotes formation of α7* nAChR-NMDA receptor heterodimeric protein complexes in the hippocampus, and blocking this interaction attenuates cue-induced nicotine seeking [100]. Taken together, these findings indicate that nicotine priming-induced reinstatement of drug seeking is mediated by α4β2* nAChRs and that cue-induced reinstatement of nicotine seeking is mediated by α7* nAChRs.

The role of other nAChR subtypes in nicotine seeking is less clear. In light of varenicline’s modest efficacy in promoting long-term abstinence [11], more recent studies have focused on newer subtype-specific nAChR agonists. Systemic administration of the α7* nAChR agonist ABT-107, and the α4β2*/α6β2* agonist ABT-089 attenuates the reinstatement of nicotine seeking indicating an important role for α7* nAChRs and α4-, α6-, and β2-containing nAChRs, respectively, in this behavioral response [101]. A more recent study found that systemic administration of the α3β4* nAChR ligand AT-1001 decreases nicotine priming-induced reinstatement of drug seeking [102]. This emerging literature expands our understanding of the role of nAChR subtypes in nicotine seeking and identifies potential new targets for novel smoking cessation pharmacotherapies.

In addition to agonists of α4β2* nAChRs, PAMs of α4β2* nAChRs may also represent potential smoking cessation medications. PAMs bind nAChRs at allosteric sites that are distinct from the orthosteric binding site for nicotine. By enhancing receptor activity, as well as the probability of nicotine-induced channel opening, PAMs can substantially increase and prolong α4β2* nAChR responses to nicotine from tobacco smoke [103]. Based on this unique mechanism, PAMs of α4β2* nAChRs may reduce the amount of nicotine consumed in daily smokers and reduce the likelihood that a smoking lapse may lead to relapse. α4β2* nAChRs can assemble into two stoichiometrically and functionally different protein complexes characterized by the α4:β2 subunit ratio. The 2(α4)3(β2) and 3(α4)2(β2) subtypes represent channels with high or low sensitivity, respectively, to acetylcholine [104, 105]. Nicotine and varenicline bind to and activate both high- and low-sensitivity populations of α4β2 nAChRs [106]. While the relevance of these findings to the therapeutic effects of nicotine and varenicline is unknown, it raises the intriguing possibility that compounds targeting high- or low-sensitivity α4β2 nAChRs may be more efficacious anti-smoking medications [107]. Recently, NS9283, a PAM that binds selectively to low-sensitivity 3(α4)2(β2) nAChRs, was shown to attenuate nicotine seeking when administered systemically to rats [108]. These findings indicate that stoichiometry-selective PAMs of nAChRs, specifically 3(α4)2(β2) nAChR subtypes, may prevent smoking relapse.

Given nicotine’s demonstrated actions at endogenous nicotine acetylcholine receptors (nAChRs), activation of nAChRs has appropriately been a large focus of nicotine-mediated behavioral investigations. However, it is also apparent that altering acetylcholine levels is sufficient to alter nicotine-mediated behaviors – particularly nicotine-seeking behavior. Several studies have been performed with acetylcholinesterase inhibitors (AChEIs), which increase extracellular levels of acetylcholine in the brain and augment cholinergic transmission through inhibition of acetylcholinesterase, a catabolic enzyme responsible for metabolizing acetylcholine in the synapse [109, 110]. Systemic administration of the AChEIs galantamine [37] and donepezil [111] attenuates nicotine seeking in rats, suggesting that enhanced cholinergic tone is sufficient to reduce nicotine consumption and smoking relapse. The translational implications of these findings are supported by a small clinical trial showing a significant reduction in total daily cigarettes smoked and the subjective measures of smoking in cigarette smokers treated with galantamine [112]. Moreover, systemic galantamine administration improves cognitive performance in rodents and humans during nicotine withdrawal [113, 114]. Given that cognitive deficits during nicotine withdrawal predict smoking relapse [115, 116], these results suggest that AChEIs may improve cognitive performance in abstinent smokers and prevent smoking relapse. Taken together, these studies suggest that AChEIs could be re-purposed as pharmacotherapies for smoking cessation [117].

GABA is the canonical inhibitory neurotransmitter and an increasing number of studies are delineating GABAergic regulation of reward-related activity [60, 118, 119]. With respect to nicotine-seeking behavior, systemic administration of the GABA_B receptor agonist CGP44532 has been shown to attenuate cue-induced reinstatement of nicotine-seeking behavior [120]. Consistent with these effects, the GABA_B receptor agonist baclofen decreases nicotine priming-induced reinstatement of drug seeking [121]. While baclofen also reduces total cigarettes smoked per day and tobacco craving in human smokers [122], compliance is limited by adverse effects including cognitive impairments, sedation, muscle relaxation and tolerance [123]. In contrast, PAMs of GABA_B receptors have fewer adverse effects than full agonists. PAMs of GABA_B receptors potentiate GABA signaling only in brain regions and at times when endogenous GABA is released [124]. Systemic administration of BH177, a PAM of GABA_B receptors, attenuates cue-induced reinstatement of drug seeking in rats [125]. Collectively, these findings indicate an important role for GABA_B receptors in nicotine seeking and suggest that the efficacy of PAMs of GABA_B receptors in smoking cessation should be explored further.

In addition to the effects of pharmacological agents of GABA receptors on nicotine seeking, recent studies have begun to examine the effects of chronic nicotine exposure and subsequent withdrawal on the endogenous GABA system. Cue-induced reinstatement of nicotine seeking is associated with increased expression of the GABA_A receptor subunits α1 and γ2 in synaptic membrane fractions from the medial prefrontal cortex (mPFC) [126]. Furthermore, inhibiting surface expression of GABA_A receptors in the dorsal, but not ventral, mPFC increases nicotine seeking, suggesting a region-specific role of mPFC GABA_A receptors in relapse [126]. Consistent with these findings, infusions of muscimol, a GABA_A receptor agonist, into the mPFC attenuates cue-induced reinstatement of nicotine seeking [126]. Thus, it has been proposed that increased GABA_A receptor signaling in the mPFC may represent a compensatory response to reduce nicotine seeking [126].

Given the critical involvement of increased mesolimbic dopamine signaling in nicotine reinforcement and behavior, preclinical studies have begun to study endocannabinoid regulation of both the dopamine system and nicotine-seeking behavior. Indeed, studies have focused on cannabinoid-1 (CB1) receptors due to their ability to regulate VTA dopamine cell firing and dopamine release in the NAc [127, 128]. In addition, systemic administration of CB1 receptor antagonists and inverse agonists attenuates the reinstatement of nicotine-seeking behavior [81, 129-132]. Consistent with these effects, activation of CB1 receptors promotes nicotine seeking during withdrawal [133]. In contrast to CB1 antagonist effects, CB2 receptor antagonists do not affect reinstatement of nicotine seeking [133]. Overall, these findings highlight a role for the endocannabinoid system, specifically CB1 receptors, in the reinstatement of nicotine seeking.

There are two endogenous ligands for the cannabinoid receptors: anandamide (AEA) and 2-arachidonoylglycerol. Pharmacological inhibition of FAAH, the primary catabolic enzyme that breaks down AEA, attenuates the reinstatement of nicotine seeking, reduces nicotine-induced excitation of VTA dopamine neurons and decreases nicotine-induced increases in extracellular dopamine levels in the NAc shell of rats [79-81]. Consistent with these effects, FAAH inhibitors also decrease nicotine reinstatement in non-human primates [134]. Inhibition of FAAH increases levels of the endocannabinoid AEA as well as the noncannabinoid fatty acid ethanolamides oleoylethanolamide (OEA) and palmitoylethanolamide (PEA) (see the section Dopamine above for a discussion of OEA and PEA in nicotine seeking) [135, 136]. The effects of FAAH inhibitors on nicotine seeking are mediated, in part, by activation of CB1 receptors [134, 137]. Consistent with these effects, systemic administration of AM404, an inhibitor of the AEA transporter, reduces nicotine-induced elevations in extracellular dopamine levels in the NAc shell and attenuates nicotine-seeking behavior, effects likely due to increased levels of AEA in the brain [138-140]. Thus, novel pharmacotherapies targeting FAAH/CB1 receptor signaling may represent promising smoking cessation medications.

In addition to nicotine and nicotine-associated cues, stress exposure can also elicit nicotine seeking during withdrawal. The effects of stress on drug seeking are mediated by numerous neurotransmitter and neuromodulator systems including norepinephrine transmission [141]. Systemic administration of the α2 adrenergic receptor agonist clonidine attenuates stress-induced reinstatement of nicotine seeking [142]. These effects are due, in part, to activation of α2 adrenergic receptors in the central nucleus of the amygdala (CeA) [143]. Systemic administration of the α1 adrenergic receptor antagonist prazosin attenuates nicotine- and cue-induced reinstatement of nicotine seeking as well as nicotine-induced release of dopamine in the NAc shell [144]. These studies suggest that α1 and α2 adrenergic receptor agonists may reduce smoking relapse in human smokers. While no studies have examined the efficacy of α1 adrenergic receptor agonists in human smokers, α2 adrenergic receptor agonists have been shown to reduce stress- and cue-induced nicotine craving in human smokers [145-147]. β-Adrenergic receptors have also been shown to play a role in nicotine seeking. For example, administration of the nonselective β-adrenergic receptor antagonist propranolol transiently reduces nicotine reinstatement [148]. Consistent with these effects, propranolol decreases nicotine craving in human smokers [149]. These exciting translational studies suggest that α2 adrenergic receptor agonists and β-adrenergic receptor antagonists could be efficacious smoking cessation medications.

In addition to the norepinephrine system, central serotonin signaling has also been examined in relapse models. Systemic administration of the 5-HT2A receptor antagonist M100907 attenuates nicotine seeking and these effects may be due to reduced NAc dopamine release following a priming injection of nicotine and/or re-exposure to conditioned stimuli [150]. Consistent with these effects, activation of 5-HT2A receptors is associated with increased release of dopamine in the NAc and prefrontal cortex [151-153]. While 5-HT2A receptor antagonists on their own do not affect dopamine levels, these compounds attenuate psychostimulant-induced increases in extracellular dopamine levels in the brain [154, 155]. Recent studies have also examined the efficacy of serotonin (5-HT) 2C receptor agonists in the nicotine reinstatement model. Similar to 5-HT2A receptor antagonists, 5-HT2C receptor agonists attenuate nicotine seeking in rats [150]. Thus, selective ligands of 5-HT receptor subtypes have been proposed as novel smoking cessation medications based on their ability to regulate dopamine transmission in the brain [156].

Nicotine interacts with many neuropeptide systems to promote reinforcement and reinstatement (for review see [157]). For example, dysregulation of dynorphin/kappa opioid receptor signaling in the brain may contribute to stress-induced reinstatement of drug seeking [158, 159]. A recent study showed that systemic administration of the selective kappa opioid receptor antagonist nor-binaltorphimine attenuates stress-, but not cue-, induced reinstatement of nicotine seeking [160]. Interestingly, systemic administration of the mixed mu/kappa opioid receptor antagonist naltrexone reduces cue-induced reinstatement of nicotine seeking [161]. Taken together, these results suggest that opioid receptor subtypes may play differential roles in nicotine seeking with mu opioid receptors involved in cue-induced nicotine seeking and kappa opioid receptors involved in stress-induced nicotine seeking.

Stress-induced reinstatement of nicotine seeking is also mediated by corticotropin-releasing factor 1 (CRF1) receptors. For example, intracerebroventricular infusions of the nonselective CRF1/CRF2 receptor antagonist D-Phe CRF_(12–41) attenuates stress-induced reinstatement of nicotine seeking [142]. These effects appear to be due to pharmacological inhibition of CRF1 receptors as selective antagonists of CRF1, but not CRF2, receptors decrease stress-induced reinstatement of nicotine seeking [162-164].

Increased hypocretin transmission mediates the reinforcing effects of nicotine [165] and may produce a state of hyperarousal that promotes drug seeking [166]. In agreement, an infusion of hypocretin directly into the ventricles reinstates nicotine seeking [163]. Moreover, systemic administration of the hypocretin-1 receptor antagonist SB334867 reduces cue-induced nicotine seeking and these effects are due, in part, to increased protein kinase C signaling in the NAc [167]. In contrast, the role of hypocretin-2 receptors in nicotine seeking is not clear. One study found no effect of the hypocretin-2 receptor antagonist TCSOX229 on cue-induced nicotine seeking [167]. However, another study showed that systemic administration of the hypocretin-2 receptor antagonist 2-SORA attenuates cue-, but not nicotine-, induced reinstatement of drug seeking [168]. Thus, these preliminary studies suggest that hypocretin transmission may play an important role in cue-induced nicotine seeking. In addition to kappa opioid, CRF and hypocretin systems, there is emerging evidence that other neuropeptide systems may play an important role in nicotine seeking [157]. For example, systemic administration of the melanocortin 4 receptor antagonists HS014 and HS024 attenuates stress-induced reinstatement of nicotine seeking [169]. Taken together, these findings clearly identify a role for neuropeptide systems in nicotine reinstatement and suggest the neuropeptide-based therapies may prevent stress-induced nicotine seeking.

Recent studies have begun to examine the role of metabolic factors in regulating nicotine-mediated behaviors [170]. For example, glucagon-like peptide-1 (GLP-1), an incretin hormone, has been shown to regulate both food and drug reinforcement [171]. With regard to nicotine addiction, systemic administration of a GLP-1 receptor agonist attenuates nicotine-induced increases in extracellular dopamine levels in the NAc and reduces the rewarding properties of nicotine [172]. GLP-1 receptor agonists also reduce nicotine self-administration indicating that activation of GLP-1 receptors is sufficient to reduce the reinforcing efficacy of nicotine [173]. The role of GLP-1 signaling in nicotine seeking, however, is unclear. From a translational perspective, it will be important to determine the efficacy of GLP-1 receptor agonists in reducing nicotine withdrawal-mediated behaviors including the reinstatement of nicotine seeking to support clinical trials of GLP-1 receptor agonists in human smokers.

As noted in the section Neurotransmitter Mechanisms Regulating Nicotine Seeking above, nAChRs in the mesoaccumbens dopamine system are known to play a critical role in nicotine-taking and -seeking behaviors [57, 58, 174, 175]. Activation of nAChRs in other brain regions such as the amygdala, hippocampus, and PFC are also known to impact nicotine taking, particularly in the context of reward learning and cognition [176, 177, as reviewed in,178]. Over the last 20 years, seminal studies from different laboratories have revealed important insights into the nAChR mechanisms underlying nicotine-induced plasticity. Early electrophysiology studies focused on nicotine-mediated synaptic plasticity in the hippocampus and in the VTA [43, 45, 179-182]. In subsequent years, electrochemical investigations also revealed plasticity-related mechanisms through which nicotine enhances phasic dopamine signaling within the NAc [46, 47, 183]. In many of these investigations, the authors often suggested that nicotine-induced plasticity in specific brain regions could mediate nicotine reinforcement, nicotine-mediated reward learning, and nicotine-seeking behavior. Many of the electrophysiological and electrochemical investigations have been performed using acute, single nicotine exposure to provide key insights into the mechanisms underlying these phenomena. Of particular note, the use of in vitro slice application of nicotine doses that mimic nicotine exposure levels in plasma of human smokers has facilitated significant advancements in our understanding of nAChR plasticity [184]. Increasingly, these in vitro investigations have begun to use repeated nicotine exposure, a model that more closely mirrors repeated exposure in human smokers, especially with respect to the mechanisms underlying nicotine withdrawal-induced behavioral deficits [185, 186]. Indeed, the mechanisms underlying nicotine-mediated plasticity may differ from those mediating acute nicotine effects. In this section, we will review key findings on nicotine-mediated plasticity and discuss the implications of such findings on our understanding of nicotine-seeking behavior and smoking relapse. We will also highlight areas where additional study can provide greater understanding of these neurobiological mechanisms and can potentially provide new insights for novel therapeutic interventions.

As noted above, VTA nAChRs are necessary and sufficient for nicotine reinforcement, as measured by nicotine self-administration in rodents [58, 175, 187, 188]. Thus, several slice electrophysiological studies have examined the mechanisms through which VTA nAChRs mediate neuronal activity and signaling in the mesolimbic dopamine system. Dopamine neurons in the VTA exhibit two types of firing patterns distinguished as tonic (3–7 Hz) or burst (15–20 Hz) firing [189-191]. Burst firing of VTA dopamine neurons in particular, is known to play a critical role in drug-taking and drug-seeking behaviors [192-194]. Previous work revealed that even in the absence of nicotine, β2* nAChRs are required for dopamine burst firing, whereas spontaneous burst firing is still observed in mutant mice lacking α7 nAChRs [57]. The presence of nicotine has been shown to enhance burst firing of VTA dopamine neurons [57, 195, 196], an effect that is absent in mutant mice that do not express the β2 nAChR subunit [57]. The role of α7 nAChRs in nicotine-induced burst firing, however, appears to be more complex. While there are no overall differences in nicotine-induced burst firing of VTA dopamine neurons between α7 nAChR knockout mice and wild-type controls, the emergence of both inhibitory (in ½ of tested neurons) and enhanced (in ½ of tested neurons) dopamine responses to nicotine are seen in α7 nAChR knock-out mice [57]. Therefore, it has been proposed that β2* nAChRs play a role in regulating dopamine neuronal excitation, while α7* nAChRs play a role in fine-tuning dopamine neuronal responses [57]. However, important questions regarding the role of nAChR subtype-specific regulation of VTA dopamine cell firing in nicotine-seeking behavior remains to be tested.

In addition to activation of nAChRs, nicotine exposure causes desensitization of nAChRs, which may in part explain acute tolerance to nicotine [197, 198]. Notably, α7* and β2* nAChRs desensitize at different rates [184]. Specifically, acute nicotine exposure in slice, at concentrations thought to persist in the brain hours after cigarette smoking (20–80 nM) enhances VTA dopamine burst firing by preferentially desensitizing β2* nAChRs on GABAergic interneurons, and not affecting (or limited desensitization of) presynaptic α7* nAChRs on glutamatergic afferents [43, 182, 184, 199]. This differential desensitization of nAChRs results in decreased GABAergic inhibition of dopamine neurons and increased glutamatergic drive onto dopamine neurons, resulting in increased VTA dopamine neuron firing. A relatively recent study demonstrated that β2* nAChRs on GABAergic interneurons are required for intracranial nicotine self-administration [200]. The role of desensitization of these specific receptors in nicotine seeking remains to be tested. A more recent body of work has demonstrated that α6* nAChRs on VTA neurons are more resistant to desensitization, allowing for prolonged dopamine cell depolarization by nicotine [44]. However, the role of α6* nAChRs is complicated by evidence that this receptor population is also expressed on GABAergic afferents in apposition to VTA dopamine neurons [199]. Further understanding of the precise role of nAChR subtypes and the desensitization kinetics of these receptor subpopulations in response to acute and chronic nicotine exposure in nicotine-seeking behavior is still needed.

It is important to note that the concentrations of nicotine used in many of the slice electrophysiology experiments described above were selected because they fall within the range typically found in the plasma of human smokers (50–500 nM) [184, 201]. We also note that the concentrations of plasma nicotine in a human smoker are likely much higher than nicotine concentrations in the brain [22, 201-203]. For example, the concentration of nicotine in the rodent brain is approximately 0.3% of the plasma nicotine levels measured in rats following nicotine self-administration [22]. Thus, it is possible that the nicotine concentrations used in some slice experiments are higher than would be identified in brain tissue of human smokers and of rodents self-administering nicotine. Due the potential nicotine dose-related issues indicated above, an ongoing consideration for the field is to identify mechanistic insights from nicotine electrophysiological studies in VTA slices and translate them to nicotine-related behavior. Electrophysiological investigations in rodents that have undergone abstinence after nicotine self-administration, and re-exposure to nicotine and nicotine-related cues, are needed to understand the role of region-specific nAChR activation and desensitization in nicotine seeking.

Electrophysiological experiments in which an animal is first exposed to acute or repeated nicotine can offer unique insights into the consequences of such exposure on both synaptic plasticity and nicotine-related behavior. Both acute and chronic nicotine administration can alter nAChR expression profiles in the VTA. However, the chronicity and route of administration, as well as dose, likely have differential impacts on neuroadaptations, as these factors drastically impact nicotine-seeking behaviors. Broadly, nAChRs are upregulated in the VTA following chronic and self-administered nicotine in rats [204-206]. The nAChR upregulation observed in human smokers is long lasting, even during smoking abstinence [207] and is thought to be a compensatory change in response to the nAChR desensitization described above. Incubation of nicotine in cultured VTA cells results in the upregulation of α4β2* nAChRs [208]. Likewise, chronic subcutaneous nicotine exposure in rats increases α4β2* and α7* nAChR binding sites in the VTA [209]. Nicotine-mediated upregulation of nAChRs is likely to influence behavior, as functional upregulation of α4* nAChRs in VTA GABAergic neurons has been shown to increase sensitivity to nicotine reward, measured by conditioned place preference [210]. While upregulation of nAChRs impacts reward-related behaviors, the role of nAChR upregulation on nicotine seeking is not well studied. It has been proposed that persistent upregulation of nAChRs during nicotine abstinence may promote craving, resulting in relapse [205]. However, additional work is needed to more fully test this hypothesis.

In addition to altering nAChR expression, nicotine can also alter glutamate receptor profiles in the VTA, thereby impacting synaptic strength. For instance, an acute nicotine injection followed by a period of abstinence increases the AMPA/NMDA ratio in VTA dopamine neurons [45]. More specifically, nicotine application in brain slices increases AMPA/NMDA ratio, which occurs via α7* but not α4β2* nAChR mechanisms [199]. Extensions of this published work evaluating AMPA/NMDA ratio in VTA slice from rats following self-administration of nicotine, and during abstinence following self-administration, will provide further understanding of the impact of glutamatergic synaptic plasticity on nicotine-seeking behaviors. Indeed, increased synaptic strength, via altered nAChR and glutamate receptor-mediated plasticity, may contribute to both the maintenance of nicotine-taking behavior and to nicotine-seeking behavior.

The NAc receives direct dopaminergic projections from the VTA. Many of the VTA nAChR mechanisms noted in the previous section also regulate NAc dopamine signaling, which is known to play a critical role in reward-related behavior. However, the role of NAc nAChRs in mediating nicotine reinforcement was initially unclear, given that blockade of NAc nAChRs failed to produce an attenuation of nicotine self-administration [183]. However, more recent work has revealed complex effects of nicotine and NAc nAChRs on dopamine signaling, which provides some clarity for the previously observed lack of NAc nAChR effects on nicotine self-administration. The current evidence suggests that nicotine modulates dopamine release in the NAc, and in the dorsal striatum, in a frequency-dependent manner via accumbal nAChR desensitization [46, 47]. Specifically, nicotine decreases dopamine release evoked by tonic-like firing activity, but maintains, and in some instances increases, dopamine release during phasic-like frequency stimulation – thereby acting as a frequency filter for dopamine release [46, 47, 211, 212]. This frequency gating is dependent on presynaptic β2* [47], α6* [213-215], α4* [214], and β3* nAChRs [216, 217]. Mechanistically, it has been proposed that this frequency filtering of tonic and phasic mediated dopamine release could underlie the encoding of salient, nicotine-associated stimuli [46, 47]. This has important behavioral implications, given the impact of both nicotine-paired cues and the encoding of such cues in guiding reward- or drug-seeking behavior [47, 218]. However, whether this nicotine-mediated frequency filtering directly contributes to cue-induced reinstatement remains untested.

Approximately 50% of nAChRs expressed on dopamine terminals in the striatum are of the α6α4β3β2* subtype [219]. Evidence suggests that the α6* nAChR is important for striatal dopamine release and nicotine-related behaviors [220, 221]. Indeed, intra-NAc shell infusion of α-conotoxin-MII, an α6* nAChR antagonist, reduced motivation to self-administer intravenous nicotine infusions in combination with associated cues [222]. To our knowledge, further parsing of whether α6* nAChRs mediate nicotine seeking has not been conducted, and this presents an interesting opportunity for future work. In addition to the effects of direct nicotine exposure, withdrawal from nicotine delivered via drinking water has been shown to increase basal dopamine concentrations in the NAc shell and to decrease phasic and tonic dopamine release, but with increased contrast between tonic and phasic release [223]. This change was accompanied by decreased control of β2* nAChRs over dopamine release in the NAc, as inhibition of β2* nAChRs was ineffective in attenuating phasic dopamine release in mice withdrawn from nicotine exposure [223]. Extending these studies to investigations of accumbal dopamine release during nicotine abstinence or withdrawal and subsequent cued reinstatement would provide further understanding of the role of nAChR subtypes in the mechanisms by which conditioned cues facilitate relapse.

In addition to the aforementioned effects of nicotine on NAc dopamine levels, nicotine also increases extracellular glutamate levels in the NAc [224, 225]. Nicotine-induced glutamate release can occur independently of dopamine-related mechanisms, as dopamine receptor antagonism in the NAc and denervation of dopamine efferents to the NAc had no effect on glutamate release stimulated by nicotine [225]. More recent work has shown that glutamate release in the NAc sensitizes over repeated nicotine exposures when delivered intravenously [226]. Additionally, chronic nicotine injections, at doses that result in locomotor sensitization, have been shown to increase dendritic length and spine density in the NAc shell [227], which could contribute to changes in glutamate release. These changes in dendrites and spines may underlie long-lasting behavioral sensitization to chronic nicotine exposure and could be important for nicotine-seeking behaviors. Nicotine has also been shown to regulate glutamate receptor expression. Nicotine self-administration may result in increased NR2B NMDA subunit expression in NAc shell, although these changes did not reach significance [228]. In contrast, nicotine exposure has been shown to downregulate NR2B-containing NMDA subunits on nAChR-expressing dopaminergic terminals in NAc synaptosomes [229]. Moreover, nicotine self-administration is associated with decreased mGlu2/3 receptor expression in the NAc shell [93]. Additional changes in the cysteine/glutamate exchange and glial glutamate exchange systems, important regulators of glutamate release, have also been observed after nicotine exposure [84, 86, 88]. For instance, acute activation of the cystine/glutamate exchange (xCT) system with N-acetylcysteine attenuates cue-induced nicotine reinstatement [87], effects that are mediated by mGluR 2/3 [88], although results describing the effective dose of N-acetylcysteine to suppress reinstatement are mixed. Importantly, N-acetylcysteine has been shown to reduce daily cigarettes smoked in smokers [86], especially if combined with varenicline [89]. Work in human smokers testing the efficacy of N-acetylcysteine on relapse and smoking cue-related tasks is warranted. Withdrawal from nicotine self-administration also reduces expression of xCT and glutamate transporter-1 in the NAc [86]. Interestingly, reductions of xCT are not observed during withdrawal if chronic nicotine is delivered via minipump. This could indicate that nicotine-induced reduction in xCT requires phasic daily nicotine exposure, as opposed to constant nicotine exposure [86]. However, it is also possible that the xCT reduction may act as an interoceptive cue predicting the availability of nicotine. This remains to be tested.

The above nicotine-induced changes in glutamate receptor expression are consistent with the glutamate hypothesis of addiction, that drug-induced alterations in glutamate homeostasis to the NAc are critical in controlling drug-seeking behavior [62]. One body of work in particular has demonstrated that reinstatement of nicotine seeking is mediated by glutamatergic plasticity in the NAc core [84]. This work specifically demonstrated that a cue-induced reinstatement test increased extracellular glutamate, spine head diameter, and AMPA/NMDA ratio in the NAc. Withdrawal from nicotine self-administration resulted in increased expression of NR2A and NR2B subunit protein expression in the NAc core, while pharmacological inhibition of these subunits attenuated cue-induced reinstatement. These results suggest that rapid glutamatergic plasticity in the accumbens contributes to nicotine-seeking behavior elicited by re-exposure to conditioned cues previously paired with nicotine taking.

One facilitator of relapse is that a drug-paired context or cue serves as a memory for the drug itself and promotes drug-seeking behavior. Therefore, nicotine-induced plasticity in regions that regulate memory consolidation and learning may also mediate nicotine-seeking behavior. Nicotine has been shown to both improve working memory and to enhance hippocampal-dependent learning [230]. In contrast, chronic nicotine administration, when delivered at high doses, enhances hippocampal neurogenesis, the growth of new neurons, specifically in the dentate gyrus region of the hippocampus [231-233]. Hippocampal neurogenesis has been observed after daily 21-h long-access sessions, but not daily 1-h short-access sessions, of nicotine self-administration [234]. In contrast, short access to nicotine self-administration has been reported to increase cell death and decrease neurogenesis in the dentate gyrus [235]. However, the relationship between changes in neurogenesis and nicotine-seeking behavior and cognition are not known. As hippocampal neurogenesis is suggested to play a role in learning and memory [236], one hypothesis is that nicotine-induced increases in neurogenesis may facilitate learned associations between nicotine and paired environmental cues, and thus facilitate cue-induced nicotine seeking during abstinence.

Nicotine induces hippocampal long-term potentiation (LTP) in vitro and in vivo, which is thought to be a neurophysiological correlate of learning and memory [237]. Specifically, acute and chronic nicotine exposure produces LTP in the CA1 [238] and dentate gyrus [239] regions of the hippocampus. In addition, in vitro activation of nAChRs, using local acetylcholine application, potentiates excitatory activity via presynaptic and postsynaptic nAChR mechanisms in hippocampal slices [240]. Systemic experimenter-administered nicotine is also sufficient to increase synaptic LTP in the dentate gyrus [241]. This systemic nicotine-induced LTP can be blocked by dopamine D1-receptor blockade, indicating that nicotine-induced hippocampal plasticity requires dopaminergic signaling [241]. This mechanism may be important for nicotine seeking, as dopamine D1 receptor blockade also blocks nicotine-conditioned place preference, which is dependent on nicotine-paired context and/or cues [241]. In the hippocampus, nAChRs are densely expressed on inhibitory interneurons [242], and nicotine-induced hippocampal LTP is at least in part dependent on local disinhibition of granule cells [243]. Nicotine-mediated hippocampal plasticity and its role in learning and plasticity have been discussed extensively [230, 244, 245].

The amygdala is associated with drug taking, drug reward, and relapse. While not as extensively studied as the hippocampus, the amygdala also plays an important role in the neurocircuitry mediating nicotine-taking and -seeking behaviors [246]. Like other brain regions discussed, nicotine induces synaptic plasticity in the amygdala. Chronic nicotine administered via drinking water in mice facilitates LTP in the amygdala, a process that depends on activation of α7* and β2* nAChRs [247]. Further, nicotine-induced LTP in the amygdala is long-lasting, up to 72 h following nicotine cessation. Given that these changes remain following cessation of nicotine exposure, it is possible that they are important in mediating nicotine-seeking behavior during abstinence, although this has not been tested. Recent work has also established the basolateral amygdala (BLA) and central amygdala (CeA) as important subregions mediating nicotine seeking and craving. Stress-induced reinstatement of nicotine-seeking behavior is attenuated by activation of α2 adrenergic receptors in the CeA [143]. Similarly, blockade of dopamine D3 receptors in the BLA attenuates cue-induced reinstatement of nicotine seeking [248]. These findings suggest that CeA α2 adrenergic receptors and BLA dopamine D3 receptors differentially mediate reinstatement of nicotine seeking elicited by stress and cues. Further, inhibition of nicotine-associated neuronal ensembles, using Daun02 in the BLA [249] and CeA [250], reduced drug-seeking behavior, indicating the importance of these regions in nicotine seeking, and perhaps in nicotine craving. Future research on these neuronal circuits will provide more insights into the role of the amygdala in nicotine-seeking behaviors.

The prefrontal cortex (PFC) is known to mediate drug reinforcement, extinction learning, and drug-seeking behavior. In the PFC, plasticity depends on correlative activity at pre- and postsynaptic sites, referred to as spike-timing-dependent plasticity [251, 252]. Nicotine increases the threshold for this spike-timing-dependent plasticity in medial PFC [253], while also increasing glutamatergic drive to these same regions via activation of presynaptic nAChRs [254, 255]. Both β2* and α7* nAChRs appear to be important in modulating glutamate release, as measured using PFC synaptosomes, though via different mechanisms [256]. PFC β2* nAChRs regulate glutamate release through voltage-gated calcium channels, whereas α7* nAChRs regulate glutamate release through intracellular calcium stores [256]. Chronic noncontingent nicotine administration results in the upregulation of β2* and downregulation of GluA1 and GluA2/3 in the PFC [257]. Likewise, chronic nicotine enhances NMDA currents in PFC [258]. Investigating the impact of self-administered nicotine, and its abstinence, on glutamate and nicotinic receptor expression in the PFC is an important next step to these experiments, as important differences have been noted between contingent (i.e., self-administration) and noncontingent procedures [28, 259]. Little attention has focused on the effects of withdrawal and abstinence from nicotine on plasticity or receptor function in the PFC and this should be a focus of future research questions to better understand the mechanisms underlying nicotine seeking.

An ongoing challenge for the nicotine and tobacco fields is the need to identify and develop novel, efficacious smoking cessation pharmacotherapies, especially given the modest efficacy of current tobacco cessation methods and the high rates of smoking relapse. As we have emphasized throughout this review, the identification of novel therapeutic targets can be strongly informed and guided by preclinical investigations in rodent models of nicotine addiction, including the reinstatement model of smoking relapse. Indeed, much of our understanding of the dopaminergic, glutamatergic, and cholinergic mechanisms underlying nicotine seeking and relapse behavior has been delineated from research in rodent models [67, 71, 87, 91, 92, 102]. Mechanistically, different neurotransmitter systems also interact to influence nicotine-seeking behavior and smoking relapse. As we have described, previous studies have already examined cholinergic and endocannabinoid influences on the dopamine system and how such processes mediate relapse behavior [46, 47, 128, 129, 132]. In many of the preclinical studies to date, investigators have demonstrated the efficacy of regulating extracellular neurotransmitter levels or manipulating specific neurotransmitter receptors with pharmacological antagonists or agonists (both full and partial agonists) to alter cue and/ or nicotine-primed reinstatement of drug seeking in rodents. More recently, preclinical studies have begun to screen the efficacy of PAMs of nAChRs in attenuating nicotine seeking [108]. By enhancing endogenous tone of neurotransmitter levels in the brain, PAMs may have greater efficacy than full and partial agonists in reducing nicotine seeking and produce less adverse effects. PAMs of α4β2* nAChRs and GABA_B receptors have recently been shown to attenuate nicotine reinstatement, results that support pharmacological approaches towards increasing endogenous acetylcholine and GABA tone, respectively, at these receptor populations during smoking cessation [108, 125]. Moreover, as the role of specific nAChR subtypes in nicotine seeking becomes clearer, stoichiometry-selective compounds targeting specific populations of nAChRs may represent efficacious smoking cessation therapies. Consistent with this strategy, a recent study showed that a PAM of nAChRs containing three α4 subunits and two β2 subunits reduces nicotine seeking in rats [108]. From a translational perspective, these preclinical studies identify novel pharmacological approaches that attenuate nicotine-seeking behavior in rodent models, and may also be used to decrease relapse behavior in human smokers.

In addition to cue- and nicotine-induced reinstatement studies, and their therapeutic implications for smoking relapse, the field will greatly benefit from additional investigations of neurotransmitter mechanisms underlying stress-mediated relapse behavior. Preclinical studies thus far revealed important roles for the norepinephrine, endocannabinoid, and neuropeptide systems (including opiate receptor and CRF mechanisms) in stress-primed reinstatement [142, 150, 160, 260]. Another critical consideration for ongoing research will be a closer examination of gender differences and their influence on stress-mediated relapse. Indeed, emerging data from multiple studies continue to provide evidence for gender-specific differences in successful quit attempts in smokers [19] as well as sex-specific effects regarding the role of stress in nicotine use and relapse in preclinical models and in humans [261-263]. Evidence from smokers suggests that men are more sensitive to the pharmacological effects of nicotine, where women are more sensitive to smoking-related cues [264, 265], and these gender differences have important implications for smoking relapse and cessation strategies [266]. It is intriguing to think that sex-specific mechanisms underlying stress-mediated relapse and cue reactivity may be differentially targeted by novel smoking cessation pharmacotherapies to reduce relapse in male and female smokers.

In addition to these behavioral studies, significant mechanistic insight has been gained through in vitro investigations of the neuroadaptations associated with nicotine exposure. These observed neuroadaptations include nicotine-mediated upregulation of subtype-specific nAChRs [204-206], nicotine-induced synaptic plasticity [43, 181, 239, 247], nicotine-mediated frequency filtering of dopamine signaling [46, 47], nicotine-induced alterations in NAc dendritic spine length and density [227], and nicotine-mediated hippocampal neurogenesis [231-233]. The behavioral implications of these findings are often interpreted based on the brain regions in which they are identified. Specifically, nicotine-mediated neuroadaptations in specific loci are thought to play a role in nicotine reinforcement (VTA and NAc), nicotine-related learning (PFC and hippocampus), nicotine craving (VTA, NAc, amygdala, and PFC), and cue-mediated nicotine-taking and nicotine-seeking behaviors (NAc, hippocampus, and amygdala). A critical factor in these mechanistic studies is a consideration of whether the in vitro experiments were performed using behaviorally relevant nicotine doses. Admittedly, it is challenging to translate relevant nicotine doses from a behavioral in vivo experiment to physiologically relevant nicotine doses in an in vitro slice experiment. In particular, we note that nicotine concentrations used in some in vitro slice experiments may be higher than would be observed in the brain of a rodent self-administering nicotine. In some previous in vitro slices studies, lower nicotine concentrations (20–80 nM) have also been used as an estimate of nicotine concentrations that may exist in the brains of human smokers hours after cigarette smoking [184]. We note that the challenges associated with in vitro nicotine dose will continue to be an important consideration for future studies. As in vitro studies continue to be performed, we also argue that an effective approach for gaining more direct mechanistic insight into nicotine seeking will be the examination of nicotine-induced neuroadaptations in rodents, or in rodent brain slices, after nicotine self-administration and subsequent abstinence. Similarly, important insights can be gained from performing in vitro slice studies in tissue following stress-, drug-, and cue-primed reinstatement of nicotine seeking. Unfortunately, there is a paucity of literature investigating synaptic plasticity that occurs during the reinstatement of nicotine seeking. Clearly, these experiments are not trivial. Thus, such approaches will benefit from, and likely require, collaborations between in vivo- and in vitro- focused nicotine researchers. This presents an exciting opportunity for the nicotine and tobacco field that has the potential to greatly enhance our understanding of nicotine seeking and craving-induced relapse behavior.

Here, we have presented both in vivo and in vitro preclinical investigations and their importance for guiding and informing clinical research aimed at providing improved pharmacotherapies to prevent nicotine relapse. However, one challenge in translating preclinical findings to clinical investigations is the possibility that individual differences in subpopulations of tobacco users might lead to differential neuroadaptations, and therefore differential efficacy of smoking cessation pharmacotherapies [266, 267]. Therefore, these are aspects that must also be considered in the experimental design of preclinical investigations, to better inform clinical investigation (e.g., how might age or sex moderate reinstatement due to stress, drug, or environmental cues).

Another highly relevant consideration is the changing landscape of tobacco use. While rates of smoking traditional cigarettes continue to decline, use of noncombustible products (electronic cigarettes; snus) is increasing [268]. Thus, one remaining question is whether subpopulations of tobacco product users have different neurophysiological mechanisms driving nicotine seeking. Thus, researchers should keep the tobacco product population, modeled in their studies, in mind. This could provide important information for cessation treatment across populations of tobacco users.

Developing the next generation of smoking cessation medications depends on advancing our understanding of the neurobiological and neurophysiological mechanisms underlying nicotine seeking and smoking relapse. Here, we have highlighted important strategies that could be used to develop novel smoking cessation medications. Identifying more efficacious pharmacotherapies to treat smoking relapse are needed to further decrease tobacco-related morbidity and mortality.

L.E.R. and N.A.A. were supported by the Yale Tobacco Center of Regulatory Science (TCORS) grant P50 DA036151 from the National Institute on Drug Abuse (NIDA) and the Federal Drug Administration (FDA) Center for Tobacco Products (CTP). H.D.S. was supported by grants DA037897 and DA039393 from the NIDA. The content is solely the responsibility of the authors and does not necessarily represent the official view of the NIH or the FDA.

The authors do not have any conflicts of interest to declare.