Neuropeptide systems and new treatments for nicotine addiction

Abstract

RATIONALE

The mildly euphoric and cognitive enhancing effects of nicotine play a role in the initiation of smoking, while dysphoria and anxiety associated with smoking cessation contribute to relapse. After the acute withdrawal phase, smoking cues, a few cigarettes (i.e., lapse), and stressors can cause relapse. Human and animal studies have shown that neuropeptides play a critical role in nicotine addiction.

OBJECTIVES

The goal of this paper is to describe the role of neuropeptide systems in the initiation of nicotine intake, nicotine withdrawal, and the reinstatement of extinguished nicotine seeking.

RESULTS

The reviewed studies indicate that several drugs that target neuropeptide systems diminish the rewarding effects of nicotine by preventing the activation of dopaminergic systems. Other peptide-based drugs diminish the hyperactivity of brain stress systems and diminish withdrawal-associated symptom severity. Blockade of hypocretin-1 and nociceptin receptors and stimulation of galanin and neurotensin receptors diminishes the rewarding effects of nicotine. Both corticotropin-releasing factor type 1 and kappa-opioid receptor antagonists diminish dysphoria and anxiety-like behavior associated with nicotine withdrawal and inhibit stress-induced reinstatement of nicotine seeking. Furthermore, blockade of vasopressin 1b receptors diminishes dysphoria during nicotine withdrawal and melanocortin 4 receptor blockade prevents stress-induced reinstatement of nicotine seeking. The role of neuropeptide systems in nicotine-primed and cue-induced reinstatement is largely unexplored, but there is evidence for a role of hypocretin-1 receptors in cue-induced reinstatement of nicotine seeking.

CONCLUSION

Drugs that target neuropeptide systems might decrease the euphoric effects of smoking and improve relapse rates by diminishing withdrawal symptoms and improving stress resilience.

1. Introduction

With 1 billion tobacco users, nicotine is one of the most widely abused drugs in the world (WHO 2011). Smoking increases the risk for numerous diseases that decrease the quality of life or lead to premature death including cancer, chronic obstructive pulmonary disease, heart disease, dementia and Alzheimer’s (Ott et al. 1998; Pirie et al. 2013; Postma et al. 2015; Thun et al. 2013). In the 20^th century alone, 100 million people died as a result of tobacco smoke exposure (WHO 2015). Smoking rates in Western countries are on the decline and about 80% of tobacco users now live in low and middle income countries with limited access to smoking cessation treatments (WHO 2015).

The positive reinforcing effects of nicotine facilitate the initiation of smoking (Finkenauer et al. 2009). Smoking induces mild euphoria, relaxation, and enhances cognitive function in people who are not yet tolerant to the effects of nicotine (Ague 1973; Benowitz 1988; Wesnes and Warburton 1983). Nicotine is the main component of tobacco smoke that leads to the development of tobacco addiction. This is supported by animal studies and the widespread use of e-cigarettes by people who have never smoked conventional cigarettes (Corey et al. 2013; Corrigall and Coen 1989; Johnston et al. 2015). The e-cigarettes deliver nicotine but only small amounts of other tobacco smoke ingredients (Famele et al. 2014; Pellegrino et al. 2012).

A brief period of nicotine intake can lead to changes in brain networks that cause compulsive drug taking (Adermark et al. 2016; Bruijnzeel 2012). Smoking cessation leads to negative affective symptoms such as dysphoria, anxiety, and impaired cognitive function (Hughes et al. 1991; Hughes and Hatsukami 1986). The negative affective symptoms increase the risk for relapse (Bruijnzeel and Gold 2005). Studies with e-cigarettes and other nicotine delivery products (e.g., gum, patch, nasal spray) show that nicotine decreases the desire to use tobacco and diminishes withdrawal (Dawkins et al. 2012; Harrell et al. 2015; Shiffman et al. 2006). Thus, nicotine is the main component in tobacco smoke that prevents people from decreasing tobacco use and quitting smoking (Bardo et al. 1999; Stolerman and Jarvis 1995). It should be noted that conventional cigarettes are more reinforcing than e-cigarettes because there are compounds in tobacco smoke that are rewarding on their own and enhance the rewarding properties of nicotine. For example, acetaldehyde is formed during the combustion of tobacco and is self-administered by rodents, induces conditioned place preference (CPP), and potentiates the rewarding effects of nicotine (Belluzzi et al. 2005; Brown et al. 1979; Myers et al. 1982; Smith et al. 1984). Other tobacco ingredients such as sugar and menthol are added to tobacco to mask the harsh taste of tobacco smoke and diminish the irritating effects of smoke in the respiratory system (Talhout et al. 2006; Willis et al. 2011). Thus, although nicotine is the main psychoactive compound in conventional cigarettes and e-cigarettes there are significant differences between these products (for an overview see, Fagerstrom et al. 2015; Grana et al. 2014; Talhout et al. 2007).

Without smoking cessation treatment, the majority of people relapse within the first week of abstinence when withdrawal signs are most severe (Hughes et al. 2004; Jarvis 2004). The relapse rate is extremely high and without pharmacotherapy only about 5% of smokers are able to maintain abstinence for 12 months (Hughes et al. 2004). Smoking induces adaptations in the brain that contribute to craving for nicotine, withdrawal, and relapse. This review focuses on the role of neuropeptide systems in nicotine addiction. However, nicotine also leads to changes in the expression of nicotinic acetylcholine receptors (nAChRs) which may contribute to relapse. Tobacco use in humans and chronic nicotine or tobacco smoke exposure in rodents leads to an upregulation of nAChRs in the brain (Benwell et al. 1988; Marks et al. 1983; Small et al. 2010; Staley et al. 2006). In active smokers, the nAChRs are in a desensitized state but cessation of nicotine use leads a reactivation of these receptors which contributes to withdrawal and relapse (Dani and Heinemann 1996).

The US Food and Drug Administration (FDA) has approved several smoking cessation drugs that diminish withdrawal and improve the relapse rate. Approved treatments include nicotine-based treatments such as nicotine gum and patches. In addition, the FDA has approved the monoamine reuptake inhibitor bupropion and the α4β2 nAChR partial agonist varenicline for smoking cessation (Hurst et al. 2013). Interestingly, a recent animal study suggests that the co-administration of bupropion and varenicline might be more effective in decreasing nicotine self-administration than either drug alone (Hall et al. 2015). Another novel approach to prevent relapse to smoking might be the use of acetylcholinesterase inhibitors. These drugs enhance cholinergic transmission and have received FDA approval for the treatment of Alzheimer’s disease (Birks 2006). Recent studies suggest that acetylcholinesterase inhibitors (e.g., galantamine and donepezil) attenuate the self-administration of nicotine in rodents and decrease smoking in humans (Ashare et al. 2016; Hopkins et al. 2012; Kimmey et al. 2014).

There are not yet any FDA-approved peptide-based treatments for smoking cessation, but there is rapid progress in this area of research. The FDA has already approved 60 peptide-based treatments for other disorders and 140 peptide-based treatments are currently in clinical trials (Fosgerau and Hoffmann 2015; Kaspar and Reichert 2013). The brain bioavailability of most orally administered or injected peptides is relatively poor because of enzymatic breakdown and poor blood brain barrier penetration. However, the brain availability of peptides can be improved by masking enzyme cleavage sites (i.e., decreasing breakdown) and increasing lipophilicity to facilitate blood brain barrier passage (Egleton and Davis 2005). Furthermore, small molecule non-peptide drugs have been developed that can be orally administered and target neuropeptide receptors in the brain (Bailey et al. 2011). Finally, clinical studies have shown that neuropeptides can be administered intranasally and affect human behavior (Yatawara et al. 2015). For example, oxytocin nasal spray improves social interactions in children with autism (Yatawara et al. 2015).

There is strong evidence for a role of brain peptide systems in nicotine addiction. Preclinical studies have provided evidence for a role of CRF, dynorphin, and hypocretin-1 in nicotine addiction (See Table 1 for an overview). The goal of this review is to provide an overview of the role of neuropeptide systems in the positive reinforcing effects of nicotine, nicotine withdrawal, and the reinstatement of nicotine seeking. In most animal studies, experimental drugs are administered once immediately before the animals are tested. Therefore, details about the treatment regimen are not reported unless the drug is administered chronically. The reviewed studies indicate that neuropeptides contribute to the development and maintenance of nicotine addiction, and that different peptides play a role in each stage of the addiction cycle.

Table 1.

Role of neuropeptides in nicotine addiction

Target	Drug	Test	Route and Dose	Effect	Reference
Acute reward
Hypocretin 1 rec. blockade	SB-334867	IVSA	ip, acute, 1–4 mg/kg	+	(Hollander et al. 2008)
Galanin rec. stimulation	Galnon	CPP	ip, acute, 0.01–0.2 mg/kg	+	(Jackson et al. 2011)
Nociceptin rec. blockade	SB612111	IVSA	ip, acute, 0.3–3 mg/kg	+	(Cippitelli et al. 2016)
Neurotensin rec. stimulation	NT69L	IVSA	ip, acute, 1 mg/kg	+	(Boules et al. 2011)
Ghrelin rec. stimulation	JMV2959	CPP	ip, acute, 6 mg/kg	+	(Jerlhag and Engel 2011)
Somatic withdrawal signs
NPY rec. stimulation	NPY	Count signs	icv, acute, 1–16 μg	+	(Rylkova et al. 2008)
Y1 rec. stimulation	[d-His26]-NPY	Count signs	icv, acute, 1–16 μg	+	(Rylkova et al. 2008)
Oxytocin rec. stimulation	Oxytocin	Count signs	ip, acute, 0.06–1 mg/kg	+	(Manbeck et al. 2014)
Kappa-opioid rec. blockade	LY2456302	Count signs	po, acute, 1–10 mg/kg	+	(Jackson et al. 2015)
Dysphoria
CRF rec. blockade	D-Phe CRF(12–41)	ICSS	icv, acute, 1–20 μg	+	(Bruijnzeel et al. 2007)
CRF rec. blockade	D-Phe CRF(12–41)	ICSS	CeA, Nacc, acute, 5–500 ng per side, bilateral	+	(Marcinkiewcz et al. 2009)
CRF1 rec. blockade	R278995/CRA0450	ICSS	icv, acute, 1–20 μg	+	(Bruijnzeel et al. 2009)
V1b rec. blockade	SSR149415	ICSS	icv, chronic, 6 days, 0.5 μg/day	+	(Qi et al. 2015a)
Kappa-opioid rec. blockade	LY2456302	CPA	po, acute, 1 and 3 mg/kg	+	(Jackson et al. 2015)
CRF overexpression	AAV2/5-CRF	ICSS	CeA, chronic	+	(Qi et al. 2014)
CRF overexpression	AAV2/5-CRF	ICSS	BNST, chronic	−	(Qi et al. 2016)
CRF2 rec. blockade	Astressin-2B	ICSS	icv, acute, 1–20 μg	−	(Bruijnzeel et al. 2009)
NPY rec. stimulation	NPY	ICSS	icv, acute, 1–16 μg	−	(Rylkova et al. 2008)
Y1 rec. stimulation	[d-His26]-NPY	ICSS	icv, acute, 1–16 μg	−	(Rylkova et al. 2008)
Oxytocin rec. stimulation	Oxytocin	ICSS	ip, acute, 0.06–2 mg/kg	−	(Manbeck et al. 2014)
V1b rec. blockade	SSR149415	ICSS	icv, acute, 0.1–2 μg	−	(Qi et al. 2015a)
CRF rec. blockade	D-Phe CRF(12–41)	ICSS	BNST, acute, 5–500 ng per side, bilateral	−	(Marcinkiewcz et al. 2009)
Anxiety
NRG3 pathway disruption	Afatinib	NIH, marble-burying test	systemic, chronic, 10 days, 10–20 mg/kg	+	(Turner et al. 2014)
CRF1 rec. blockade	MPZP	EPM	sc, acute, 20 mg/kg	+	(Cohen et al. 2015)
CRF1 rec. blockade	Antalarmin	EPM, marble burying test	IPN, acute, 1 μg/side, unilateral	+	(Zhao-Shea et al. 2015)
Y2 rec. blockade	JNJ-31020028	social interacti on test	ip, chronic, 7 days, 20 mg/kg	+	(Aydin et al. 2011)
Kappa-opioid rec. blockade	JDTic	EPM	sc, acute, 8 mg/kg	+	(Jackson et al. 2010)
Kappa-opioid rec. blockade	LY2456302	EPM	po, acute, 1–10 mg/kg	+	(Jackson et al. 2015)
CRF2 rec. blockade	Antisauvagine-30	EPM	IPN, acute, 2 μg/side, unilateral	−	(Zhao-Shea et al. 2015)
Cue-induced reinstatement
Hypocretin 1 rec. blockade	SB334867	IVSA	ip, acute, 5 and 10 mg/kg	+	(Plaza-Zabala et al. 2013)
Hypocretin 2 rec. blockade	TCSOX229	IVSA	ip, acute, 5 and 10 mg/kg	−	(Plaza-Zabala et al. 2013)
Nicotine-primed reinstatement
Kappa-opioid rec. blockade	nor-BNI	CPP	sc, acute, 10 mg/kg	−	(Jackson et al. 2013)
Stress-induced reinstatement
CRF rec. blockade	D-Phe CRF(12–41)	IVSA	icv, acute, 1–20 μg	+	(Zislis et al. 2007)
CRF1 rec. blockade	R278995/CRA0450	IVSA	icv, acute, 1–20 μg	+	(Bruijnzeel et al. 2009)
CRF1 rec. blockade	Antalarmin	IVSA	sc, acute, 30 mg/kg	+	(Plaza-Zabala et al. 2010)
MC4 rec blockade	HS014 and HS024	IVSA	icv, acute, 1–10 μg (HS014), 1 and 3 μg (HS024)	+	(Qi et al. 2015b)
Hypocretin 1 rec. blockade	SB334867	IVSA	ip, acute, 5 and 10 mg/kg	+	(Plaza-Zabala et al. 2010)
Kappa-opioid rec. stimulation	nor-BNI	CPP	sc, acute, 10 mg/kg	+	(Jackson et al. 2013)
CRF2 rec. blockade	Astressin-2B	IVSA	icv, acute, 1–20 μg	−	(Bruijnzeel et al. 2009)

The plus (+) signs indicate that the treatment diminished the rewarding effects of nicotine, withdrawal, and relapse. Minus (−) signs indicate that the treatment was ineffective. Abbreviations: BNST, bed nucleus of the stria terminalis; CeA, central nucleus of the amygdala; CPA, conditioned place aversion; CPP, conditioned place preference; EPM, elevated plus maze test; ICSS, intracranial self-stimulation; icv, intracerebroventricular; ip, intraperitoneal; IPN, interpeduncular nucleus; IVSA, intravenous self-administration; Nacc, nucleus accumbens; NIH, novelty-induced hypophagia; po, oral; rec, receptor; sc, subcutaneous.

2. Genetic variation in hypocretin and galanin receptor genes affects risk for smoking

Genome-wide association (GWA) studies continue to provide important insight into the neuronal mechanisms underlying smoking in humans. Some of the first GWA studies showed that variations in the sequence of α3, α5, and β4 nAChR subunit genes affects the risk for developing and maintaining a nicotine addiction (Berrettini et al. 2008; Chen et al. 2009; Liu et al. 2010; Wen et al. 2016). More recent genetic studies revealed that variability in the α4 nAChR subunit gene affects the risk for developing nicotine addiction in humans (Hancock et al. 2015; Thorgeirsson et al. 2016).

Interestingly, GWA studies also point to a role for neuropeptide systems in nicotine addiction. One study, in which only Japanese subjects were included, reported that an SNP (rs2653349) in the hypocretin-2 receptor gene (HCRTR2) affects the severity of nicotine dependence. Smokers with the A/G genotype of the rs2653349 SNP were more likely to have severe nicotine dependence than smokers with the G/G genotype (Nishizawa et al. 2015). Interestingly, carriers of the A/G genotype were also more likely to use methamphetamine at a young age than carriers of the G/G genotype. This suggests that this rs2653349 SNP may predispose people for drug use. A study with US smokers (Yale University Transdisciplinary Tobacco Use Research Center) provides strong evidence for variation in the galanin-1 receptor (GALR1) gene in the development of nicotine addiction. Cubells and colleagues showed that an SNP in the GALR1 gene (rs2717162) increased self-reported craving during a quit attempt (Lori et al. 2011). The same rs2717162 SNP decreases the effectiveness of bupropion as a smoking cessation aid (Gold et al. 2012). Smokers who carried at least one minor C allele and were treated with bupropion were more likely to report severe cravings and less likely to maintain abstinence than smokers with TT alleles. Another study reported that variations in the GALR1 gene affect the likelihood of heavy smoking (Jackson et al. 2011). In conclusion, these studies indicate that variations in nAChR genes and neuropeptide receptor genes (GALR1 and HCRTR2) affect the likelihood of developing a tobacco addiction, the effectiveness of smoking cessation aids, and the risk for relapse.

3. Neuropeptides and the rewarding effects of nicotine

3.1 CRF and the rewarding effects of nicotine

The activation of nAChRs plays a critical role in the self-administration of nicotine and nicotine-induced CPP (Corrigall et al. 1994; Walters et al. 2006; Watkins et al. 1999). After the development of dependence, the endogenous release of CRF and the resulting negative mood state can drive nicotine intake (i.e., negative reinforcement). Corticotropin-releasing factor is a neuropeptide and is expressed in several brain sites including the paraventricular nucleus of the hypothalamus, central nucleus of the amygdala (CeA), bed nucleus of the stria terminalis (BNST), ventral tegmental area (VTA), and locus coeruleus (LC)(Grieder et al. 2014; Swanson et al. 1983; Vale et al. 1981). Hypothalamic CRF neurons project to the median eminence and mediate the release of ACTH into the periphery. Extrahypothalamic CRF orchestrates behavioral and autonomic responses to stressors (Koob and Heinrichs 1999; Nijsen et al. 2001), and many of these effects are independent of CRF’s effects on the hypothalamic–pituitary–adrenal (HPA) axis (Eaves et al. 1985; Sutton et al. 1982). Two CRF receptors have been discovered, namely the CRF₁ and CRF₂ receptor (Chen et al. 1993; Lovenberg et al. 1995; Perrin et al. 1993). Corticotropin-releasing factor serves as endogenous ligand for the CRF₁ receptor and urocortin 2 and 3 are ligands for the CRF₂ receptor (Lewis et al. 2001). Urocortin 1 binds with a slightly higher affinity to the CRF₁ than the CRF₂ receptor (Lewis et al., 2001). Drug withdrawal-induced behavioral and physiological changes are predominantly mediated via the activation of CRF₁ receptors (Bruijnzeel and Gold 2005; Koob 1999; Steckler and Holsboer 1999). The activation of CRF₂ receptors counteracts the effects of stress responses and CRF₁ receptor activation (Bale and Chen 2012).

The great majority of the nicotine/CRF studies investigated the role of CRF in withdrawal and relapse, but several studies have also investigated the role of CRF in the rewarding effects of nicotine. These studies mostly suggest that CRF increases nicotine intake in dependent, but not in non-dependent, animals. Rats with extended access to nicotine develop dependence and display increased nicotine intake after a period of abstinence (George et al. 2007). Blockade of CRF₁ receptors with MPZP prevents this increase in nicotine intake, but does not affect nicotine intake in control rats with limited access to nicotine (George et al. 2007). Chronic nicotine intake increases CRF levels in the VTA and this could contribute to high levels of nicotine intake in dependent animals (Grieder et al. 2014). This is supported by the observation that downregulation of CRF mRNA in the VTA decreases nicotine self-administration in an extended access paradigm (21 h/day access) (Grieder et al. 2014). Decreasing CRF levels in the VTA does not affect nicotine intake in rats with limited access to nicotine. On a similar note, blockade of CRF₁ receptors does not affect nicotine-induced CPP but prevents stress-induced potentiation of nicotine-induced CPP (Brielmaier et al. 2012). In conclusion, these studies indicate that CRF₁ receptor antagonists decrease nicotine intake in dependent animals. Furthermore, stressors potentiate the reinforcing properties of nicotine and this is prevented by CRF₁ receptor antagonists. Corticotropin-releasing factor type 1 receptor antagonists may help to reduce smoking in dependent smokers and in non-dependent smokers who use tobacco to cope with stressful situations.

3.2 Other peptides than CRF and rewarding effects of nicotine

Corticotropin releasing factor mainly plays a role in nicotine intake in dependent animals, but numerous other neuropeptides regulate nicotine intake in non-dependent animals. Kenny and colleagues have provided insight into the role of hypocretin transmission in nicotine intake (Hollander et al. 2008). Their studies showed that the hypocretin-1 receptor antagonist SB-334867 decreases nicotine self-administration in rats with limited access (1 h) while not affecting responding for food pellets. Pretreatment with the hypocretin-1 receptor antagonist also prevented the nicotine-induced lowering of brain reward thresholds in the intracranial self-stimulation (ICSS) paradigm. A decrease in brain reward thresholds is indicative of a potentiation of brain reward function (i.e., euphoria) (Der-Avakian and Markou 2012a). Therefore, this indicates that hypocretin-1 receptor antagonists decrease the acute rewarding effects of nicotine. Hypocretin fibers and hypocretin-1 receptors have been detected in the insular cortex. Blockade of the hypocretin-1 receptors in this brain site attenuates the self-administration of nicotine (Hollander et al. 2008). Therefore, activation of hypocretin-1 receptors in the insular cortex may play a role in the initiation of smoking.

Several other neuropeptides have been implicated in the rewarding effects of nicotine including nociceptin/orphanin FQ, neurotensin, galanin, and ghrelin (Boules et al. 2011; Cippitelli et al. 2016; Jerlhag and Engel 2011). Nociceptin mediates its effects in the brain by acting upon the nociceptin receptor. Stimulation of the nociceptin receptor with AT-202 increases nicotine self-administration and blockade of this receptor with SB612111 decreases nicotine self-administration in rats (Cippitelli et al. 2016). Similar effects were observed in dependent and non-dependent rats. Nociceptin is a member of the opioid receptor family and blockade of this receptor prevents drug-induced dopamine release in the nucleus accumbens (Nacc)(Lutfy et al. 2001; Vazquez-DeRose et al. 2013). Dopamine release in the NAcc plays a critical role in the reinforcing properties of nicotine and therefore nociceptin receptor antagonists may decrease nicotine intake by inhibiting the nicotine-induced increase in dopamine signaling (Corrigall and Coen 1991).

Neurotensin is another neuropeptide that could affect the rewarding properties of nicotine by modulating dopamine release. The neurotensin receptor agonist NT69L reduces nicotine self-administration in rats and inhibits nicotine-evoked dopamine release in the NAcc (Boules et al. 2011; Liang et al. 2008). However, NT69L also inhibits the nicotine-induced increase in norepinephrine levels in the prefrontal cortex (PFC)(Liang et al. 2008). Norepinephrine release in the PFC plays a critical role in the reinforcing properties of drugs of abuse (Ventura et al. 2003). Therefore, neurotensin agonists may decrease nicotine intake by modulating dopamine signaling in the NAcc and norepinephrine signaling in the PFC.

Ghrelin is produced in high levels by the stomach during fasting and its levels decrease after food intake (Perello and Dickson 2015). Ghrelin is also produced by neurons in the arcuate hypothalamic nucleus (Kojima et al. 1999; Mondal et al. 2005). The release of ghrelin increases food intake and enhances the rewarding properties of food (Perello and Dickson 2015). It has been suggested that ghrelin modulates the activity of reward systems and thereby regulates food intake. Blockade of ghrelin receptors attenuates nicotine-induced CPP and nicotine-induced dopamine release (Jerlhag and Engel 2011). Therefore, ghrelin receptor blockade could potentially decrease the positive reinforcing effects of smoking.

Galanin positive neurons have been identified in the bed nucleus of stria terminalis (BNST), central amygdala (CeA), medial septum, thalamus, hypothalamus, dorsal raphe nucleus, and brain stem areas (Melander et al. 1986). Galanin plays a role in a wide range of biological functions including energy homeostasis, reproduction, arousal, sleep, and cognition (Gundlach 2002; Lang et al. 2015). Several studies with smokers indicate that galanin plays a role in nicotine craving and withdrawal (Jackson et al. 2011; Lori et al. 2011). Preclinical studies suggest that galanin also mediates the rewarding effects of nicotine. Picciotto and colleagues used galanin knock-out mice to study the role of galanin in nicotine reward (Neugebauer et al. 2011). Nicotine induced CPP in the galanin knock-out mice and the wild-type mice but higher doses were needed to induce CPP in the knock-out mice. This suggests that galanin knock-out mice are less sensitive to the rewarding effects of nicotine and that galanin signaling contributes to nicotine reward. The findings of this study are not in line with a study that used a pharmacological approach to investigate the role of galanin in the rewarding effects of nicotine. In this study, the galanin receptor agonist galnon blocked the rewarding properties of nicotine in the CPP test (Jackson et al. 2011). It is somewhat surprising that a study with KO mice suggests that galanin enhances the rewarding effects of nicotine and that a pharmacologic study suggests that galanin receptor activation inhibits nicotine reward. It has been suggested that these differences are due to compensatory mechanisms in the KO mice or due to strain differences (Jackson et al. 2011). It might also have been possible that that galnon affects nAChR signaling and diminishes the rewarding effects of nicotine via this off-target effect (Jackson et al. 2011).

In conclusion, the present studies indicate that several neuropeptides play a role in the rewarding effects of nicotine. There is strong evidence that hypocretin-1 and nociceptin receptor blockade diminishes the rewarding effects of nicotine. Stimulation of neurotensin and galanin receptors also attenuates the rewarding effects of nicotine. Therefore, these studies suggest that modulating hypocretin-1, nociceptin, neurotensin, and galanin systems may decrease smoking in people in whom smoking is mainly driven by positive reinforcement processes.

4. Dysphoria and nicotine withdrawal

During the last decades, a wide range of animal models have been developed to investigate the dysphoria associated with smoking cessation. In most of these studies, animals were rendered nicotine dependent using intermittent injection protocols or minipumps that continuously deliver nicotine (Epping-Jordan et al. 1998; Malin 2001). Nicotine withdrawal can be observed after the administration of non-selective nAChR antagonists (precipitated withdrawal) and after the cessation of nicotine administration (spontaneous withdrawal)(Epping-Jordan et al. 1998). Spontaneous withdrawal has greater face validity than precipitated withdrawal as it more closely models human smoking cessation. The precipitated withdrawal method has, however, some significant advantages over the spontaneous withdrawal method. One of the advantages of the precipitated withdrawal method is that nAChR blockade induces withdrawal rapidly and withdrawal signs subside quickly because of the short half-life of nAChR antagonists (Debruyne et al. 2003). Furthermore, the magnitude and duration of precipitated withdrawal symptoms are consistent across sessions and therefore several doses of an experimental drugs can be tested in each animal (Bruijnzeel et al. 2012; Skjei and Markou 2003).

Cessation of chronic nicotine administration induces a stress response, increases corticosterone levels, and decreases operant responding for food pellets, which might be indicative of a negative mood state (Benwell and Balfour 1979; Corrigall et al. 1989; LeSage et al. 2006). Cessation of nicotine administration also leads to immobility in the forced swim test, which might be indicative of a depressive-like state (Mannucci et al. 2006; Roni and Rahman 2014). This increase in immobility in the forced swim test has been observed for up to 60 days after the cessation of nicotine administration (Mannucci et al. 2006). The conditioned place aversion (CPA) procedure has also been used to assess the aversive state associated with precipitated nicotine withdrawal. In this test, the negative motivational properties of drug withdrawal are paired with a neutral test environment. After pairing, the previously neutral environment acts as a conditioned stimulus and induces avoidance behavior (Tzschentke 1998).

The nAChR receptor antagonist mecamylamine induces place aversion in rats chronically treated with nicotine but not in control rats (Suzuki et al. 1996). The ICSS procedure has also been used to investigate the negative mood state associated with drug withdrawal (Bruijnzeel and Markou 2004; Epping-Jordan et al. 1998). In the discrete trial ICSS procedure the animals are prepared with electrodes in the lateral hypothalamus (LH)/medial forebrain bundle and when placed in the operant chamber they can self-stimulate their brain reward system. By systematically changing the intensity of the electrical current and determining the lowest current that supports self-stimulation, the state of the brain reward system can be assessed (Der-Avakian and Markou 2012b). The acute administration of drugs of abuse increases the sensitivity to the electrical stimuli and lowers brain reward thresholds, which is indicative of a potentiation of brain reward function. In contrast, cessation of chronic drug administration leads to an increase in ICSS thresholds, which is indicative of a negative mood state. The administration of nAChR antagonists to nicotine dependent animals or cessation of chronic nicotine administration leads to elevations in brain reward thresholds (Epping-Jordan et al. 1998; Johnson et al. 2008). One of the main advantages of the ICSS procedure is that it provides a quantitative measure of the emotional state of the brain reward system during withdrawal. The ICSS procedure is a well validated test to assess the dysphoria associated with nicotine withdrawal. Elevations in brain reward thresholds are not exclusive to nicotine withdrawal but are also observed during withdrawal from other drugs including opioids, alcohol, and amphetamine (Bruijnzeel et al. 2006; Cryan et al. 2003b; Schulteis et al. 1995).

The FDA has approved two non-nicotine treatments for smoking cessation. Both bupropion and varenicline have been shown to prevent the elevations in brain reward thresholds associated with nicotine withdrawal (Cryan et al. 2003a; Igari et al. 2013). Therefore, this animal model has face and predictive validity for identifying treatments that diminish dysphoria associated with smoking cessation.

In addition to dysphoria, many animal studies have investigated somatic signs associated with precipitated and spontaneous nicotine withdrawal (Damaj et al. 2003; Malin et al. 1992; Watkins et al. 2000). Somatic withdrawal signs in rodents include teeth chattering, chews, gasps, writhes, ptosis, and head and body shakes (Malin et al. 1992). These signs are often recorded for 5–10 minutes after the rats or mice are placed in an observation chamber. Both nicotine and the smoking cessation drug bupropion diminish somatic withdrawal signs (Cryan et al. 2003a; Malin et al. 1992). It should be noted that somatic withdrawal signs in humans are mild and that relapse to smoking is mainly driven by dysphoria and craving (Bruijnzeel 2012; Hughes 2007; Koob and Volkow 2016). Furthermore, drugs that diminish affective signs might not diminish somatic signs and vice versa (Bruijnzeel et al. 2010). Therefore, this suggests that somatic withdrawal signs have only weak predictive value for identifying treatments that diminish affective withdrawal signs and craving in humans.

4.1 CRF and dysphoria associated with nicotine withdrawal

In our laboratory, we have extensively explored the role of CRF in the dysphoria associated with nicotine withdrawal. In one of our first experiments, we investigated if blockade of CRF receptors with the CRF₁/CRF₂ receptor antagonist D-Phe CRF_(12–41) diminishes the elevations in brain thresholds associated with precipitated nicotine withdrawal (Bruijnzeel et al. 2007). Central (icv) administration of D-Phe CRF_(12–41) diminished mecamylamine-induced elevations in brain reward thresholds. In a follow-up experiment, the role of CRF₁ and CRF₂ receptors in the dysphoria associated with precipitated nicotine withdrawal was investigated (Bruijnzeel et al. 2009). The highly selective CRF₁ receptor antagonist R278995/CRA0450 was used to block CRF₁ receptors and CRF₂ receptors were blocked with astressin-2B. Blockade of CRF₁, but not CRF₂, receptors, prevented the elevations in brain reward thresholds associated with precipitated nicotine withdrawal. These studies suggest that CRF₁ receptor activation at least partly mediates the dysphoria associated with smoking cessation. It was then investigated which brain areas play a role in the dysphoria associated with nicotine withdrawal (Marcinkiewcz et al. 2009). Administration of D-Phe CRF_(12–41) into the CeA and Nacc shell prevented the elevations in brain reward thresholds in the nicotine dependent rats. In contrast, blockade of CRF₁/CRF₂ receptors in the BNST was ineffective. To investigate if CRF₁ receptors in the CeA play a role in nicotine withdrawal, we conducted a study with the selective CRF₁ receptor antagonist R278995/CRA0450 (Bruijnzeel et al. 2012). This study showed that blockade of CRF₁ receptors in the CeA diminishes the dysphoria associated with nicotine withdrawal.

To further explore the role of CRF in the CeA in the regulation of mood states and withdrawal, the effect of the overexpression of CRF in the CeA on the dysphoria associated nicotine withdrawal was investigated. Corticotropin-releasing factor was overexpressed in the CeA using an adeno-associated virus (AAV), with AAV2 terminal repeats and AAV5 capsids (AAV2/5), that selectively transduces neurons in the brain (Burger et al. 2004). The rats were trained on the ICSS paradigm and when the thresholds were stable they received viral vectors that delivered GFP or CRF and brain reward thresholds were assessed for 4 weeks. Previous studies suggest that increased CRF release in the CeA contributes to dysphoria (Bruijnzeel et al. 2012; Marcinkiewcz et al. 2009). However, chronic overexpression of CRF in the CeA did not affect brain reward thresholds. It was then investigated if the overexpression of CRF in the CeA affects the dysphoria associated with nicotine withdrawal (Qi et al. 2014). Precipitated and spontaneous withdrawal led to large elevations in brain reward thresholds, which was diminished in rats that overexpressed CRF. At the end of the experiment, we investigated the effect of the administration of the AAV-CRF vector on CRF, CRF₁ receptor, and CRF₂ receptor levels. Administration of the AAV-CRF vector led to an increase in CRF levels, a decrease in CRF₁ receptor levels, and an increase CRF₂ receptors levels. Therefore, the overexpression of CRF in the CeA may have diminished the dysphoria associated with nicotine withdrawal by downregulating CRF₁ receptors. However, there is evidence that stimulation of CRF₂ receptors has anti-stress and antidepressant-like effects (Chen et al. 2006; Tanaka and Telegdy 2008). Therefore, the upregulation of CRF₂ receptors may have contributed to preventing withdrawal in the nicotine withdrawing rats. In a recent study we also investigated the AAV2/5 mediated overexpression of CRF in the BNST on the dysphoria associated with nicotine withdrawal (Qi et al. 2016). The overexpression of CRF in the BNST did not affect baseline brain reward thresholds, but prevented the elevations in brain reward thresholds associated with precipitated and spontaneous withdrawal. The overexpression of CRF increased CRF₁ and CRF₂ receptor levels in the BNST and increased the CRF₂/CRF₁ receptor ratio. This suggests that the overexpression of CRF in the BNST diminishes the dysphoria during nicotine withdrawal by inducing an increase in the CRF₂/CRF₁ receptor ratio.

Considering the strong evidence that CRF mediates the dysphoria associated with nicotine withdrawal, clinical studies may investigate whether CRF₁ receptor blockade diminishes the acute dysphoria associated with smoking cessation. In conclusion, these animal studies indicate that the dysphoria associated with nicotine withdrawal is at least partly mediated by the activation of CRF₁ receptors.

4.2 Other peptides than CRF and dysphoria associated with nicotine withdrawal

Although more than one hundred neuropeptides have been discovered, the role of only a few neuropeptides in nicotine withdrawal have been investigated (Bruijnzeel 2012; Zhang et al. 2014). In our laboratory, we investigated the effects of NPY on somatic and affective nicotine withdrawal signs (Rylkova et al. 2008). Neuropeptide Y is widely expressed throughout the brain and low levels of NPY have been associated with depressive-disorders (De Quidt and Emson 1986a; de Quidt and Emson 1986b; Redrobe et al. 2002). Central administration of NPY decreases somatic morphine and alcohol withdrawal signs (Woldbye et al. 1998; Woldbye et al. 2002). In our nicotine study, NPY attenuated the somatic signs associated with both precipitated and spontaneous nicotine withdrawal. The specific Y1 receptor agonist [d-His²⁶]-NPY decreased abdominal constrictions during precipitated withdrawal and decreased overall somatic signs during spontaneous withdrawal. We also investigated the effect of NPY and the Y1 receptor agonist [d-His²⁶]-NPY on the dysphoria associated with precipitated nicotine withdrawal. Neuropeptide Y or [d-His²⁶]-NPY did not affect the elevations in brain reward thresholds associated with precipitated nicotine withdrawal. High doses of NPY or [d-His²⁶]-NPY actually increased brain reward thresholds in the control rats. This increase in brain reward thresholds was blocked by the Y1 receptor antagonist BIBP-3226. These studies indicate that Y1 receptor agonists may diminish some of the somatic signs associated with smoking cessation, but these compounds do not diminish the dysphoria associated with smoking cessation and high doses might induce impairments in reward function. Harris and colleagues found something similar with oxytocin (Manbeck et al. 2014). They showed that oxytocin diminishes the somatic withdrawal signs associated with precipitated nicotine withdrawal but did not affect the elevations in brain rewards thresholds. Similar to NPY, oxytocin elevated the brain reward thresholds of the control rats.

There is strong evidence from animal studies that blockade of vasopressin 1b (V1b) receptors diminishes the effects of stress and has antidepressant-like effects (Iijima and Chaki 2007; Iijima et al. 2014). Therefore, in our laboratory we investigated the effect of a V1b receptor antagonist on the dysphoria associated with nicotine withdrawal (Qi et al. 2015a). Acute administration of the V1b receptor antagonist SSR149415 slightly diminished the elevations in ICSS thresholds associated with precipitated withdrawal. Chronic (6 days) administration of SSR149415 was more effective and completely prevented the elevations in ICSS thresholds associated with precipitated nicotine withdrawal. Interestingly, a recent clinical study reported that chronic treatment with the V1b receptor antagonist ABT-436 decreases smoking in patients with alcoholism (Ryan et al. 2016). Future studies may investigate if V1b receptor blockade diminishes the dysphoria associated with smoking cessation and improves the relapse rate.

To our knowledge, only the acute effects of NPY and oxytocin on nicotine withdrawal have been investigated. Considering the fact that the V1b receptor antagonist was more effective after chronic treatment it might be worthwhile to explore whether chronic treatment with NPY or oxytocin is also more effective. The kappa-opioid receptor system has been shown to play a critical role in regulating mood states (Bruchas et al. 2010; Bruijnzeel 2009). Blockade of kappa-opioid receptors with LY2456302 also diminishes nicotine-withdrawal induced CPA and thus suggesting that it diminishes the negative emotional state associated with nicotine withdrawal (Jackson et al. 2015).

In conclusion, the reviewed studies indicate that blockade of kappa-opioid receptors and chronic blockade of V1b receptors diminishes the dysphoria associated with nicotine withdrawal. Neuropeptide Y and oxytocin diminish somatic signs associated with nicotine withdrawal but do not diminish the dysphoria associated with nicotine withdrawal. These studies would suggest that kappa-opioid and V1b receptor antagonists decrease the dysphoria in people who try to quit smoking.

4.3 Norepinephrine and dysphoria associated with nicotine withdrawal

Norepinephrine plays a critical role in regulating the release of neuropeptides in the brain. For example, norepinephrine stimulates the release of CRF in the PVN and other brain sites (Dunn et al. 2004). Two noradrenergic cell groups have been located in the mammalian brain (Dahlström and Fuxe 1964). Noradrenergic neurons in the LC give rise to the dorsal noradrenergic bundle. The LC provides most of the norepinephrine input to the forebrain areas and activation of the LC enhances attention, arousal, and learning and memory (Aston-Jones 2005). Noradrenergic cell groups in the lateral tegmentum give rise to the ventral noradrenergic bundle. These cells innervate the hypothalamus, septum, and subcomponents of the extended amygdala such as the CeA and BNST (Moore and Card 1984).

We investigated the effects of the α1-adrenoceptor antagonist prazosin, the α2-adrenoceptor agonist clonidine, and the β₁/β₂-adrenoceptor antagonist propranolol on the dysphoria and somatic signs associated with precipitated nicotine withdrawal (Bruijnzeel et al. 2010). These drugs inhibit noradrenergic transmission in the brain by blocking α1 or β-adrenergic receptors or stimulating presynaptic α2-adrenergic receptors. We found that α1-adrenoceptor blockade (prazosin), but not α2-adrenoceptor activation (clonidine) or β₁/β₂-adrenoceptor blockade (propranolol), attenuates the elevations in brain reward thresholds associated with precipitated nicotine withdrawal. In contrast, α2-adrenoceptor activation and β1/β2-adrenoceptor blockade, but not α1-adrenoceptor blockade, decreased somatic nicotine withdrawal signs. This pattern of results suggests that only α1-adrenoceptor blockade diminishes the dysphoria associated with nicotine withdrawal. Dysphoria and drug craving are the main causes of relapse to drug use (Bruijnzeel 2012; Bruijnzeel and Gold 2005). Therefore, drugs that only decrease somatic signs may not prevent relapse. In conclusion, the reviewed studies suggest that drugs that block α1-adrenoceptors may prevent dysphoria in people trying to quit smoking and thereby decrease the risk for relapse.

5. Nicotine withdrawal and anxiety

Smoking cessation leads to an increase in anxiety and this contributes to relapse to smoking (Bruijnzeel 2012). Smokers are more likely to have an anxiety disorder than non-smokers and smokers with an anxiety disorder are less motivated to quit (Johnson et al. 2000; Zvolensky et al. 2007). The smoking cessation drug varenicline diminishes anxiety in humans after quitting smoking and this may improve the relapse rate (Foulds et al. 2013). In contrast, there is no evidence that the smoking cessation treatment bupropion alleviates anxiety after smoking cessation (Shiffman et al. 2000).

Several neuropeptides have been implicated in nicotine withdrawal-induced anxiety-like behavior including CRF, neuregulin 3 (NRG3), NPY, and dynorphin (Aydin et al. 2011; Cohen et al. 2015; Jackson et al. 2015; Turner et al. 2014). Some of the first studies with CRF showed that it increases anxiety-like behavior and more recent studies showed that it contributes to anxiety-like behavior after cessation of nicotine administration (Cohen et al. 2015; Dunn and File 1987; Swerdlow et al. 1986). Cohen and colleagues reported that rats with extended access to nicotine (21 h/day, 4 days/week) display an increase in anxiety-like behavior in week 10, 72 h after the last self-administration session (Cohen et al. 2015). This increase in anxiety-like behavior was blockade by systemic and intra-CeA administration of the CRF₁ receptor antagonist MPZP.

In an elegant study, Turner and colleagues showed that nicotine withdrawal leads to an increase in NRG3 levels in the hippocampus (Turner et al. 2014). It was then investigated if nicotine withdrawal leads to anxiety-like behavior in NRG3^ska mice. These mice have a mutation in the NRG3 gene, which leads to low levels of NRG3 and NRG3 levels do no increase after cessation of nicotine administration (Howard et al. 2005; Turner et al. 2014). Nicotine withdrawal-induced anxiety-like behavior is diminished in the NRG3^ska mice. Furthermore, chronic blockade of the NRG3 receptor signaling pathway with Afatinib (BIBW-2992) decreases nicotine withdrawal-induced anxiety-like behavior. Therefore, this suggests that nicotine withdrawal leads to an increase in NRG3 signaling which contributes to anxiety-like behavior.

The habenula-interpeduncular system is a brain network that has been implicated in the regulation of emotional states and drug addiction (Hikosaka 2010). The medial habenula receives input from the septum and diagonal band of broca and is via the fasciculus retroflexus connected to the interpeduncular nucleus (IPN). The IPN connects to serotonergic neurons in the raphe nuclei. Both the habenula and the IPN play a role in the development of nicotine dependence. Administration of nAChR antagonists into the habenula and the IPN induces somatic withdrawal signs (Salas et al. 2009). Furthermore, nAChRs in the habenula regulate anxiety-like behavior. Mice that express hypersensitive α6/α4β2 nAChRs in the habenula are more anxious than control mice (Pang et al. 2016). Tapper and colleagues investigated the role of CRF in the habenula-IPN circuit in anxiety-like behavior during nicotine withdrawal. They showed that CRF release in the IPN mediates anxiety-like behavior. Administration of CRF in the IPN increased anxiety-like behavior and blockade of CRF₁ receptors in this brain site prevented nicotine-withdrawal induced anxiety-like behavior. Decreasing CRF levels in the VTA with shRNAs decreased nicotine-withdrawal induced activation of the IPN and also anxiety-like behavior. This suggests that there is a CRF projection from the VTA to the IPN that increases anxiety-like behavior during nicotine withdrawal (Zhao-Shea et al. 2015).

Neuropeptide Y stimulates food intake and blunts stress responses (Reichmann and Holzer 2016). The effects of NPY are opposite to those of CRF. Corticotropin-releasing factor increases anxiety-like behavior while NPY and Y1 agonists decrease anxiety-like behavior (Heilig et al. 1989; Sajdyk et al. 1999). The Y2 receptor is located presynaptically and stimulation of the Y2 receptor decreases NPY release and increases anxiety-like behavior (Colmers et al. 1991; King et al. 1999; Nakajima et al. 1998). Blockade of Y2 receptors with JNJ-31020028 decreases anxiety-like behavior in alcohol withdrawing rats (Cippitelli et al. 2011). Chronic treatment with the same Y2 receptor antagonist also diminishes nicotine withdrawal-induced anxiety-like behavior in the social interaction test (Aydin et al. 2011). To our knowledge, the effect of NPY or Y1 receptor agonists on nicotine withdrawal-induced anxiety-like behavior has not been investigated. However, based on the above discussed findings it is predicted that NPY or Y1 receptor agonists also decrease anxiety-like behavior in nicotine withdrawing rats.

There is extensive evidence for a role of kappa-opioid receptor signaling in anxiety-like behavior. Kappa-opioid receptor agonists increase anxiety-like behavior and antagonists decrease anxiety-like behavior (Bruchas et al. 2010; Bruijnzeel 2009). Nicotine withdrawal leads to changes in dynorphin levels, which is the endogenous ligand for the kappa-opioid receptor (Chavkin et al. 1982). Cessation of nicotine administration leads to a decrease in dynorphin levels and an increase in prodynorphin mRNA levels in the Nacc (Isola et al. 2008). This suggests that nicotine withdrawal leads to an increase in the release and production of dynorphin. Blockade of dynorphin receptors in the brain with the long-action kappa opioid receptor antagonist JDTic or with the short acting antagonist LY2456302 decreases anxiety-like behavior in nicotine withdrawing mice (Jackson et al. 2010; Jackson et al. 2015).

6. Reinstatement of Nicotine Seeking

The great majority of people who try to quit smoking relapse during the first year. There are several factors that increase the risk for relapse. First of all, dysphoria leads to craving and thereby increases the risk for relapse. Also, cues associated with drug use and small amounts of drugs of abuse can lead to craving and relapse. Animal models have been developed to study the role of stress, cues, and drugs in relapse to smoking (Shaham et al. 2003). One of the most widely used animal model to study relapse to smoking is the rat nicotine self-administration paradigm. Rats are trained to self-administer nicotine and after about 2-weeks nicotine seeking is extinguished by replacing nicotine with saline. Operant responding can be reinstated by footshock stress, administration of nicotine, or cues associated with the self-administration of nicotine (Buczek et al. 1999; Chiamulera et al. 1996; LeSage et al. 2004; Liu et al. 2006). Several studies have investigated the effects of the nAChR antagonist mecamylamine and the FDA approved smoking cessation drugs varenicline and bupropion in these animal models. Blockade of nAChRs with mecamylamine dose dependently attenuates cue-induced reinstatement of nicotine seeking (Liu et al. 2007). Very low doses of mecamylamine, which did not affect responding for food or nicotine, decreased cue-induced reinstatement of nicotine seeking. This suggests that relatively low doses of nAChR antagonists could be used to diminish cue-induced craving and prevent relapse in smokers. The smoking cessation drug varenicline inhibits reinstatement of operant responding induced by nicotine-prime and a combination of nicotine-prime and cues (O’Connor et al. 2010; Swalve et al. 2016). In contrast, bupropion potentiates cue-induced reinstatement of nicotine seeking (Liu et al. 2008). This might be due to the fact that bupropion inhibits the re-uptake of dopamine and thereby enhances the effect of the cue (Shaham et al. 2003; Warner and Shoaib 2005). This is supported by the observation that dopamine receptor blockade diminishes cue-induced reinstatement of drug-seeking (Crombag et al. 2002; Liu and Weiss 2002).

The CPP procedure has also been used to investigate the neuronal mechanisms that mediate the reinstatement of nicotine seeking. Nicotine-prime reinstates extinguished CPP and this is blocked by the nAChR antagonist mecamylamine and the smoking cessation drugs varenicline and bupropion (Biala and Budzynska 2006; Biala et al. 2010; Budzynska and Biala 2011). Preclinical studies point to a role for the serotonergic system in the reinstatement of nicotine seeking. Both the 5-HT_2C receptor agonist Ro60-0175 and the 5-HT_2A receptor antagonist M100907 decrease nicotine-primed and cue-induced reinstatement of nicotine seeking (Fletcher et al. 2012).

The role of neuropeptide systems in the reinstatement of nicotine seeking is somewhat underinvestigated but there is evidence for a role of CRF, kappa-opioid, hypocretin-1, and MC4 receptors (Plaza-Zabala et al. 2013; Qi et al. 2015b; Zislis et al. 2007). Studies from our laboratory suggest that stress-induced reinstatement of nicotine seeking is at least partly mediated by the activation of CRF₁ receptors. Central (icv) administration of the non-specific CRF₁/CRF₂ receptor antagonist D-Phe CRF_(12–41) prevents stress-induced reinstatement of nicotine seeking (Zislis et al. 2007). Furthermore, blockade of CRF₁ receptors with R278995/CRA045, but not blockade of CRF₂ receptors with astressin-2B, prevents stress-induced reinstatement of nicotine seeking in rats (Bruijnzeel et al. 2009). Blockade of CRF₁ receptors also prevents stress-induced reinstatement of nicotine seeking in the mouse (Plaza-Zabala et al. 2010).

In our laboratory, we investigated the role of the MC4 receptor in stress-induced reinstatement of nicotine seeking (Qi et al. 2015b). The endogenous ligand for the MC4 receptor is the pro-opiomelanocortin (POMC)-derived peptide alpha-melanocyte stimulating hormone (α-MSH)(Adan et al. 2006; Gantz et al. 1993). Stimulation of the MC4 receptor increases anxiety and depressive-like behavior and therefore it was predicted that blockade of the MC4 receptor would diminish the effects of stress in the reinstatement procedure (Chaki et al. 2003; Kokare et al. 2010; Serova et al. 2013). Indeed, we found that the MC4 receptor antagonists HS014 and HS024 prevented stress-induced reinstatement of nicotine seeking. The MC4 receptor antagonists did not affect responding for food pellets, which suggests that MC4 receptor blockade did not have sedative effects.

The hypocretins (hypocretin-1 and hypocretin-2) are hypothalamic neuropeptides that are derived from a common precursor protein called prepro-hypocretin (de Lecea et al. 1998). The hypocretins are only produced in the lateral and posterior hypothalamus and it has been estimated that there are only about eleven hundred hypocretin neurons in the rat brain (Date et al. 1999; Peyron et al. 1998). Despite the fact that there are only a small number of hypocretin neurons, they project to a wide range of brain sites and regulate a variety of behaviors including feeding and sleep (Chemelli et al. 1999; Sakurai et al. 1998). The hypocretins also play a role in stress responses and drug addiction (Bruijnzeel 2012). Berrendoro and colleagues thoroughly investigated the role of the hypocretin receptors in the reinstatement of nicotine seeking. They showed that hypocretin-1 reinstates extinguished nicotine seeking in a CRF₁ receptor independent manner (Plaza-Zabala et al. 2010). In the same study it was shown that blockade of CRF₁ receptors prevents stress-induced reinstatement of nicotine seeking, but hypocretin-1 receptor blockade was ineffective. This indicates that both the hypocretin and CRF system plays a role in relapse, but only the CRF system plays a role in stress-induced relapse. In a follow-up study, the role of the hypocretin system in cue-induced reinstatement of nicotine was investigated. Blockade of hypocretin-1 receptors with SB334867, but not blockade of hypocretin-2 receptors with TCSOX229, attenuated cue-induced reinstatement of nicotine-seeking (Plaza-Zabala et al. 2013). Taken together, these studies indicate that stress-induced reinstatement of nicotine seeking is mediated via CRF₁ receptors and cue-induced reinstatement is mediated via hypocretin-1 receptors. We are not aware of any studies that investigated the role of CRF in cue-induced reinstatement of nicotine seeking. However, a recent study showed that blockade of CRF₁ receptors in the insular cortex reduces cue-induced reinstatement of cocaine seeking (Cosme et al. 2015). Because there is strong overlap in the neuronal mechanisms that mediate the reinstatement of nicotine and cocaine seeking, it might be possible that CRF₁ receptors also play a role in cue-induced reinstatement of nicotine seeking (Cosme et al. 2015).

Neuropeptide S (NPS) is produced by a small group of neurons that is located adjacent to the LC in the brainstem (Clark et al. 2011; Pape et al. 2010; Xu et al. 2004). Neuropeptide S increases locomotor activity, arousal, and wakefulness and decreases anxiety-like behavior (Pape et al. 2010; Reinscheid and Xu 2005; Xu et al. 2004). Nicotine has been shown to increase NPS and NPS receptor levels in the brainstem but otherwise little is known about the role of NPS in nicotine addiction (Lage et al. 2007). Interestingly, several studies with other drugs have shown that NPS facilitates the reinstatement of drug seeking and NPS antagonists decrease drug intake and the reinstatement of drug seeking (Kallupi et al. 2010; Pañeda et al. 2009; Schmoutz et al. 2012; Thorsell et al. 2013). Therefore, these studies warrant exploration of the role of NPS in nicotine intake and the reinstatement of nicotine seeking.

The kappa-opioid receptor system plays a critical role in stress-responses and anxiety-like behavior. To our knowledge, the role of kappa-opioid receptors in the reinstatement of extinguished nicotine self-administration has not been investigated. However, one study investigated the role of kappa-opioid receptors in rats in which CPP was extinguished. In this study, the kappa-opioid receptor antagonist nor-BNI blocked stress-induced reinstatement of CPP but not nicotine-prime induced reinstatement (Jackson et al. 2013).

In conclusion, the reviewed studies indicate the nAChR antagonist mecamylamine and the smoking cessation drug varenicline diminish nicotine-prime and cue-induced reinstatement of nicotine seeking. The effects of bupropion on reinstatement has been poorly evaluated but one study suggests that bupropion potentiates cue-induced reinstatement of nicotine seeking. The reviewed studies indicate that neuropeptide systems are a viable target for drugs to prevent relapse to smoking. There is strong evidence that blockade of CRF₁ and kappa-opioid receptors diminishes stress-induced reinstatement of nicotine seeking. Blockade of the hypocretin-1 receptor prevents cue-induced reinstatement of nicotine seeking. These studies suggest that neuropeptide based treatments could be used to decrease the risk for relapse.

7. Neuronal activation of peptide systems

The transcription factor Fos has been widely used to study the effects of stressors and drugs on neuronal activity in the brain (Bruijnzeel et al. 1999; Morgan and Curran 1995). Numerous studies have investigated the brain sites that are being activated by low and rewarding doses of nicotine. These studies show that noncontingent administration of nicotine, nicotine self-administration, and place preference training with nicotine leads to an increase in Fos expression in brain sites that signal reward function (Kiba and Jayaraman 1994; Pagliusi et al. 1996; Pascual et al. 2009). Acute nicotine administration increases Fos positive cells in the NAcc, caudate putamen, and PFC and this is diminished by non-selective blockade of nAChRs (Kiba and Jayaraman 1994). These findings are in line with a pharmacological fMRI study in which the effect of nicotine on brain activity (BOLD signal) was investigated (Bruijnzeel et al. 2015). Nicotine administration increased the BOLD signal in the NAcc, dorsal striatum, amygdala, prefrontal cortical areas, and the motor and sensory cortex. Pretreatment with a nAChR antagonist blocked the effects of nicotine on the BOLD signal. These histological and imaging studies indicate that nicotine increases the activity of brain sites that play a role in reward signaling and cognition.

Few studies have investigated whether there is a difference in the brain sites being activated by brief versus prolonged nicotine self-administration. However, a recent study investigated Fos expression in the brain after 10 versus 47 days of nicotine self-administration (Clemens et al. 2014). Interestingly, both self-administration protocols induced Fos expression in brain areas that have been associated with reward function such as the NAcc core and shell, VTA, CeA, and basolateral amygdala (BLA). However, prolonged access to nicotine also leads to Fos expression in the dorsomedial and dorsolateral striatum and the substantia nigra. These brain sites have been associated with habitual and compulsive drug use (Belin and Everitt 2008; Pierce and Vanderschuren 2010). Thus, this suggests that both acute and prolonged nicotine use leads to the activation of the brain reward system. Brain networks that regulate habitual behaviors are only activated after prolonged drug use.

There is strong evidence that cessation of chronic nicotine administration leads to the activation of anti-reward systems which induces dysphoria and anxiety-like behavior (Koob and Volkow 2016). However, despite overwhelming evidence for a role of neuropeptide systems in nicotine withdrawal there is only sparse evidence from Fos studies that nicotine withdrawal activates brain sites. One of the first studies that investigated Fos expression in rats undergoing withdrawal showed that withdrawal leads to an increase in Fos expression in the CeA but not in other brain sites (Panagis et al. 2000). Although pharmacological studies suggest that withdrawal leads to the activation of brain stress systems, several Fos studies suggest that nicotine withdrawal leads to decreased activity in brain sites (Balerio et al. 2004; Varani et al. 2012). In a recent study, it was shown that nicotine withdrawal led to a decrease in the number of Fos positive cells in the BNST, BLA, and dentate gyrus. Nicotine withdrawal did not affect Fos expression in a wide range of other brain sites including the NAcc, cingulate cortex, caudate putamen, CA1 and CA3 region of the hippocampus, and medial habenular nucleus (Varani et al. 2012).

Although there are a large number of studies that investigated the effects of nicotine and nicotine withdrawal on Fos expression in the brain, only a few colocalization studies have been conducted to investigate which neuropeptide systems are activated or inhibited by nicotine. Muschamp and colleagues investigated the role of hypocretin neurons in the LH in nicotine withdrawal (Simmons et al. 2016). Hypocretin neurons project from the LH to the VTA and hypocretin release in the VTA increases the firing of dopaminergic neurons (Korotkova et al. 2003). Interestingly, acute nicotine administration increases Fos expression in hypocretin neurons in the LH and nicotine withdrawal leads to a decrease in Fos expression in these neurons (Fadel and Burk 2010; Simmons et al. 2016). Overall, this suggests that a decrease in activity in hypocretin neurons in the LH contributes to the hypo-dopaminergic state in nicotine withdrawing animals (Hildebrand et al. 1998).

Although the majority of studies suggest that nicotine withdrawal does not affect Fos expression or decreases Fos expression, one study suggested that nicotine withdrawal can increase Fos expression in peptide neurons (Plaza-Zabala et al. 2012). In this study, nicotine withdrawal did not affect Fos expression in a wide range of brain sites including the CeA, BLA, NAcc, medial septum, and hippocampus. However, nicotine withdrawal increased Fos expression in hypocretin neurons in the LH and the perifornical and dorsomedial hypothalamus (PFA/DMH). In addition, withdrawal increased Fos expression in CRF neurons in the PVN. A retrograde tracing study showed that the hypocretin neurons project from the LH and PFA/DMH to the PVN. Interestingly, systemic administration of the hypocretin-1 receptor antagonist decreased somatic withdrawal signs and decreased Fos expression in the PFA/DMH and in CRF neurons in the PVN. This suggests that activation of hypocretin neurons that project from the PFA/DMH to CRF neurons in the PVN increases somatic nicotine withdrawal signs.

In conclusion, these studies indicate that acute nicotine administration increases Fos expression in the reward system and many other brain sites. Nicotine withdrawal does not affect Fos expression in most brain sites but affects Fos expression in some brain sites that regulate reward function and stress responses. Nicotine withdrawal might decrease Fos expression in the reward system (LH) and increase Fos expression in some brain sites (CeA, PVN, PFA/DMH) that contribute to stress responses during nicotine withdrawal. It is also interesting to note that nicotine withdrawal seems to decrease Fos expression in hypothalamic hypocretin neurons that project to the VTA (Reward) and increases Fos expression in hypocretin neurons that project to the PVN (anti-reward/stress). Overall, the reviewed studies suggest that nicotine withdrawal decreases activity in brain sites that signal reward and increases activity in brain sites that mediate stress responses and negative affective states.

8. Conclusion

In this overview article we discuss the role of neuropeptides in the development and maintenance of nicotine addiction. The reviewed studies indicate that different neuropeptide systems play a role in each stage of the addiction cycle. It should be noted that not all the discussed neuropeptides have been evaluated in all stages of the addiction cycle. Therefore, it cannot be ruled out that some neuropeptides have more widespread effects than is currently believed.

It should also be noted that the great majority of the nicotine studies have been conducted with animals that self-administered nicotine under a limited access schedule (1h/day) or received nicotine non-contingently via injections or minipumps. In contrast, humans often smoke for most of the day and carefully titrate the amount of nicotine self-administered (Benowitz 2009). Recently, animal models have been developed that more closely model human smoking. One prime example of this is the extended access nicotine self-administration procedure. In this model, rats have extended-access (21–23h) to nicotine for 1–4 days which is then followed by several days of abstinence (Cohen et al. 2012; Flores et al. 2016). This model leads to high levels of nicotine intake and might better model human smoking than limited access models or the non-contingent administration of nicotine. These new models may help to identify novel mechanisms that contribute to the development and maintenance of compulsive smoking.

The rewarding effects of nicotine are pivotal for the acquisition of smoking and the dysphoria associated with smoking cessation increases the risk for relapse. After the acute withdrawal phase, stressors, a small amount of nicotine (i.e., lapse), and cues contribute to relapse. The reviewed studies indicate that blockade of hypocretin-1 and nociception receptors, and stimulation of galanin, neurotensin, and ghrelin receptors diminishes the rewarding effects of nicotine (Boules et al. 2011; Cippitelli et al. 2016; Hollander et al. 2008; Jackson et al. 2011; Jerlhag and Engel 2011). Overall, a different group of neuropeptides plays a role in nicotine withdrawal. Stimulation of NPY and oxytocin receptors and blockade of kappa-opioid receptors diminishes somatic nicotine withdrawal signs, but stimulation of NPY and oxytocin receptors does not diminish the dysphoria associated with nicotine withdrawal (Manbeck et al. 2014; Rylkova et al. 2008). The dysphoria associated with nicotine withdrawal is greatly diminished by CRF₁ and kappa-opioid receptor blockade and chronic V1b receptor blockade (Bruijnzeel et al. 2009; Qi et al. 2015a). Blockade of the CRF₁ and kappa-opioid receptors also diminishes anxiety-like behavior associated with nicotine withdrawal and stress-induced reinstatement of nicotine seeking (Bruijnzeel et al. 2009; Cohen et al. 2015; Jackson et al. 2015). Considering the fact that cues and smoking lapses increase the risk for relapse to smoking, it is somewhat surprising that only a few studies have investigated the role of neuropeptides in nicotine-primed and cue-induced reinstatement. One study reported that blockade of hypocretin-1 receptors diminishes cue-induced reinstatement of nicotine seeking (Plaza-Zabala et al. 2013). Thus, blocking the hypocretin-1 receptor could diminish the acute rewarding effects of smoking and provide protection against relapse. In conclusion, the reviewed studies indicate that neuropeptides play a critical role in the acquisition of tobacco smoking, withdrawal, and relapse. Therefore, peptide-based treatments could improve smoking rates by preventing the transition to high levels of smoking, diminishing withdrawal signs, and preventing stress- and cue-induced urges to smoke.