Nicotinic acetylcholine receptors: upregulation, age-related effects, and associations with drug use

Abstract

Nicotinic acetylcholine receptors are ligand-gated ion channels that exogenously bind nicotine. Nicotine produces rewarding effects by interacting with these receptors in the brain’s reward system. Unlike other receptors, chronic stimulation by an agonist induces an upregulation of receptor number that is not due to increased gene expression in adults; while upregulation also occurs during development and adolescence there have been some opposing findings regarding a change in corresponding gene expression. These receptors have also been well studied with regard to human genetic associations and, based on evidence suggesting shared genetic liabilities between substance use disorders, numerous studies have pointed to a role for this system in comorbid drug use. This review will focus on upregulation of these receptors in adulthood, adolescence, and development, as well as the findings from human genetic association studies which point to different roles for these receptors in risk for initiation and continuation of drug use.

Background and Significance

Drug use

Although alcohol and tobacco use are legal, they contribute to severe and widespread problems. Worldwide, 3.3 million people die each year due to the harmful use of alcohol, representing 5.9% of worldwide deaths. Furthermore, 5.1% of the global burden of disease and injury is attributed to alcohol and recent causal relationships have been established between harmful drinking and occurrence of infectious diseases such as tuberculosis and HIV/AIDS (WHO). As of July 2015, tobacco was estimated to kill up to half of its users (WHO). In the U.S. alone, 1 in 5 deaths are attributable to smoking (CDC), and an additional 6.8 million people suffer from a serious illness caused by smoking (CDC).

Over the years spanning 2005 and 2010 between 3.4 and 6.6% of the adult population aged 15–64 used illicit drugs. Roughly 10 to 13% of these users subsequently developed drug dependence and/or a drug use disorder with high prevalence rates of serious disorders such as HIV, hepatitis C, and hepatitis B. Illicit drug use is responsible for approximately 1 in every 100 deaths among adults (UNODC). In America illicit drug use is increasing; in 2012 9.2% of the population aged 12 or older had used an illicit drug or abused a psychotherapeutic medication in the past month. Finally, 52% of new drug users are under 18, illustrating the importance of studying these behaviors during development since most people use drugs for the first time in their teenage years (NIDA).

Evidence for shared genetic influences between different classes of drugs

Epidemiological and familial studies have shown that comorbidity among substance use disorders (SUDs, i.e. meeting abuse or dependence criteria for more than one legal or illegal drug) is high (Bierut et al., 1998, Kapusta et al., 2007, Kendler et al., 1997, Kessler et al., 1997, Merikangas et al., 1998, Pickens et al., 1995). Converging evidence from twin studies highlights the importance of genetic factors on SUDs with estimates of heritability ranging from 0.30–0.70 (Agrawal & Lynskey, 2008). Furthermore, although genetic factors specific to each substance have been identified, research has indicated that a common genetic factor underlies much of the variation in SUDs in adults (Agrawal et al., 2004, Kendler et al., 2003, Palmer et al., 2015, Palmer et al., 2012, True et al., 1999a, True et al., 1999b, Tsuang et al., 2001, Xian et al., 2008). Although work by Kendler and colleagues has implicated two underlying genetic factors with separate influences on licit and illicit drugs, these factors where shown to be highly correlated (r = 0.82) (Kendler et al., 2007). These results point to a common mechanism in the development of SUDs (Vanyukov et al., 2003).

Similar estimates have been seen for SUDs in adolescence, indicating an underlying genetic liability for substance use (Hopfer et al., 2003). Problem use has been shown to be more heritable than initiation or regular use in adolescents (Rhee et al., 2003) and twin analyses have shown significant genetic correlations for problem use across substances (Young et al., 2006). Substance use is a developmental problem that increases linearly with age (Young et al., 2002) and common genetic factors have been suggested to be particularly important for early onset SUDs that emerge in late adolescence and early adulthood (Iacono et al., 2008, Palmer et al., 2009). Similar to findings in adults, a study by Rhee and colleagues suggested two hypotheses for the comorbidity between alcohol and illicit drug dependence in adolescents: a single general liability, or two highly correlated separate liabilities (Rhee et al., 2006). Finally, tobacco has been shown to pose the greatest substance-specific risk for developing subsequent use problems (Palmer et al., 2009) and as such the remainder of this review will focus specifically on the effects of tobacco and the receptors to which it binds in the brain.

Nicotinic acetylcholine receptors

Physiology

Although there are many compounds in tobacco smoke, nicotine is considered to be the major addictive component of tobacco smoke (Gunby, 1988, Rose, 2006, Stolerman & Jarvis, 1995). Nicotine binds muscle and neuronal nicotinic acetylcholine receptors (nAChRs, encoded by the CHRN genes), members of a family of ligand-gated ion channels, in the peripheral and central nervous systems (Gotti et al., 2009). They are pentameric receptors composed of various subunits ranging from alpha1-alpha10 (α1-α10) and beta1-beta4 (β1-β4) (discussed in more detail below) clustered around a central ion pore. Upon stimulation by nicotine, the five subunits undergo a conformational change causing the central pore to open, allowing extracellular ions to enter the cell. Nicotine is able to cross the blood-brain barrier and bind neuronal nAChRs in the brain (Clarke, 1987), producing rewarding effects by interacting with nAChRs in the brain’s reward system (Changeux, 2010).

Muscle nicotinic receptors

Muscle nAChRs all have the same pentameric stoichiometry of (α1)₂β1δγ (fetal-type) or (α1)₂β1δγ (adult-type) (Le Novere et al., 2002). Although primarily known to be expressed in muscle tissue, previous work has suggested that subunits of these receptors are also present in mammalian ciliary ganglia (Pugh et al., 1995), vestibular and cochlear hair cells (Scheffer et al., 2007), as well as brain regions such as the cortex, hippocampus, and cerebellum (Ghedini et al., 2010). In addition, a splice variant of the α1 subunit was shown to expressed in numerous locations including brain, kidney, heart, liver, lung, and thymus in humans (Talib et al., 1993).

Neuronal nicotinic receptors

Neuronal nAChRs are comprised of combinations of the α2-α10 and β2-β4 subunits with variable stoichiometry that are expressed in the central and peripheral nervous systems, located both presynaptically and postsynaptically (Millar & Gotti, 2009). All of the subunits are expressed in the mammalian central nervous system with the exception of the α8 subunit, which is only expressed in chicks (Britto et al., 1992, Schoepfer et al., 1990). Neuronal nAChRs can assemble in two ways; they can form homomeric receptors in which all five subunits are the same (α7, α9), or they can form heteromeric receptors in which the five subunits that make up the functional receptor include different α and β subunit combinations, which yield different pharmacological properties.

In the mammalian system, α7 and α9 subunits form homomeric receptors when expressed as isolated subunits in cells (Couturier et al., 1990, Elgoyhen et al., 1994, Schoepfer et al., 1990). The α10 subunit, although it cannot form a functional homomeric receptor, is most closely related to the α9 subunit and forms a functional receptor when coexpressed with the α9 subunit (Elgoyhen et al., 1994, Lustig et al., 2001, Sgard et al., 2002). In the central nervous system, the α9 and α10 subunits have been primarily identified in the inner ear cells, particularly the cochlear and vestibular hair cells (Elgoyhen et al., 1994, Elgoyhen et al., 2001, Luo et al., 1998, Lustig et al., 1999, Lustig et al., 2001, Vetter et al., 1999), as well as the pituitary (Sgard et al., 2002). The α7 subunit, however, is widely expressed throughout the mammalian central nervous system (reviewed in (Millar & Gotti, 2009)).

The rest of the neuronal nAChR subunits form heteromeric receptors consisting of α and β subunits (Anand et al., 1991, Boulter et al., 1987, Deneris et al., 1988, Wada et al., 1989). For heteromeric receptors, subunit composition varies, as does assembly; furthermore, the variable stoichiometry confers differences in calcium permeability and agonist and antagonist sensitivity between different receptors (Luetje & Patrick, 1991) (reviewed in (Millar & Gotti, 2009)). Studies have shown that α2-α4 and β2 and β4 subunits can form functional receptors when expressed as a pair-wise combination of one α subunit and one β subunit; functional expression has been observed for α2β4, α3β2, α3β4, α4β2, and α4β4 combinations (Duvoisin et al., 1989, Papke et al., 1989). The β3 and α5 subunits cannot form functional nAChRs unless expressed in combination with another α and β subunit pair, as neither of these possesses a binding site for an agonist or antagonist. Electrophysiological approaches have been used to distinguish triple pair subunits from pair subunits and subsequently identified receptors composed of the α3β2α5 (Gerzanich et al., 1998, Wang et al., 1996), α3β4α5 (Fucile et al., 1997, Gerzanich et al., 1998, Wang et al., 1996), and α4β2α5 (Ramirez-Latorre et al., 1996) subunits. In addition, the β3 subunit has been shown to co-assemble into a functional receptor composed of α3β3β4 subunits in oocytes (Groot-Kormelink et al., 1998), and may promote expression and stability of human α6 subunits in transfected cell lines (Tumkosit et al., 2006). Expression studies have also demonstrated that the α6 subunit assembles into functional triplet receptors, including α6β3β4 (Kuryatov et al., 2000, Tumkosit et al., 2006), as well as α3α6β4 and α3α6β2 (Fucile et al., 1998, Kuryatov et al., 2000) subunit combinations. Lastly, α6 and β2 combinations have also been observed, particularly as α4α6β2β3 and α6β2β3 receptors (Salminen et al., 2004). While neuronal nAChRs are widespread throughout the central and peripheral nervous systems, α4β2 (Anand et al., 1991, Flores et al., 1992, Wada et al., 1989) receptors are the most abundant heteromeric receptors in the brain and have the highest affinity for nicotine (Wonnacott, 1990).

Upregulation of nicotinic acetylcholine receptors

The response of nicotinic receptors during chronic exposure to agonists has been referred to as a paradox (Wonnacott, 1990), since they do not behave as other receptors during chronic exposure to an agonist. Note that an agonist binds and activates a receptor whereas an antagonist binds and blocks receptor activation. Numerous studies have shown that chronic exposure of a receptor to an antagonist typically leads to upregulation, or an increased number of receptors, while chronic exposure of a receptor to an agonist causes downregulation, or a decreased number of receptors (Creese & Sibley, 1981, Wonnacott, 1990). However, this phenomenon is not observed with some subtypes of nAChRs, most notably, the α4β2 nAChR. Upon extended exposure of nAChRs to an agonist, α4β2 receptor numbers are upregulated. It is important to clarify that for the purposes of this discussion, upregulation refers to an increase in receptor numbers (e.g. not upregulation of mRNA), and the studies mentioned below have focused exclusively on measuring protein levels to assess the number of nAChRs.

Early studies on the brains from rats (Schwartz & Kellar, 1983) and mice (Marks et al., 1983) repeatedly exposed to nicotine showed an increase in [³H]acetylcholine and (−)-[³H]nicotine binding, two radiolabeled nAChR ligands that predominantly measure α4β2 nAChRs. Schwartz and Kellar treated rats with 2 mg/kg of nicotine twice a day for 10 days and observed an increase in [³H]acetylcholine binding in cerebral cortex of rats (Schwartz & Kellar, 1983). Similarly, Marks et al. (1983) chronically infused mice with 5 mg/kg/hr nicotine for 8–10 days (mice are less sensitive to the effects of nicotine than rats are (Matta et al., 2007), accounting for the difference in nicotine treatments in these two experiments) and reported an increase in (−)-[³H]nicotine binding in the cortex, midbrain, hindbrain, hippocampus, and hypothalamus, but not striatum (Marks et al., 1983). These results were confirmed in human post mortem brains in 1988. By measuring (−)-[³H]nicotine binding in smokers compared to nonsmokers, Benwell and colleagues demonstrated that smokers showed higher levels of (−)-[³H]nicotine binding in most grey areas including the hippocampus, cortex, gyrus rectus and median raphe nucleus, with the exception of the medulla oblongata which showed no changes in (−)-[³H]nicotine binding. Importantly, in the areas where smoking was associated with increased (−)-[³H]nicotine binding, radiolabeled ligand binding increased by 50–100% (Benwell et al., 1988). In all three of the above studies, increased radiolabeled ligand binding was confirmed to be due to an increase in receptor number, rather than an increase in binding affinity (Benwell et al., 1988, Marks et al., 1983, Schwartz & Kellar, 1983). Thus, under chronic nicotine exposure, α4β2 nAChRs undergo a nicotine-induced upregulation of receptor numbers at the membrane.

Subsequent studies provided evidence that when nicotine exposure is ceased, decreased binding of (−)-[³H]nicotine occurs after 7–10 days in mice (Marks et al., 1985), 15–20 days in rats (Collins et al., 1990), and between 21 days and 2 months in humans (Breese et al., 1997, Mamede et al., 2007). In humans, nAChR upregulation was additionally shown to be dose-dependent (Breese et al., 1997, Perry et al., 1999). Gopalakrishnan et al. (1997) detected a significant increase in [³H]cytisine binding upon exposure to nicotine at concentrations as low as 100 nM, and a maximal 15-fold increase in binding at concentrations of 10 μM of nicotine in the human embryonic kidney cell line HEK293 that expressed heterologous α4β2 nAChRs (Gopalakrishnan et al., 1997). This is particularly interesting because smokers have an average serum concentration of 100–200 nM nicotine after smoking (Benowitz, 1996, Henningfield et al., 1993), suggesting that nAChR upregulation could occur in smokers in between, as well as during, periods of smoking.

As expected, because at baseline human brains have different levels of nicotine binding in different regions of the brain (Court & Clementi, 1995), the magnitude of increase in nicotine binding sites is not universal among all areas of the brain (Benwell et al., 1988, Perry et al., 1999). Regions of the thalamus appear most resistant to nicotine-induced upregulation of nAChR numbers in rodent models (Kellar et al., 1989, Pauly et al., 1991). In humans, the medulla oblongata, as well as different layers of the cortex and hippocampus are resistant to, or undergo nAChR upregulation at variable rates (Benwell et al., 1988, Perry et al., 1999). Given that nAChR subunits show variable expression in different regions of the brain, this suggests that individual neuronal nicotinic receptor subtypes may undergo differential upregulation according to location in the brain. In addition to upregulation of α4β2 subunit-containing receptors (Flores et al., 1992, Zhang et al., 1995), α3, albeit requiring a higher dose of nicotine, as well as α7, to a lesser extent, subunits are also upregulated (Olale et al., 1997, Peng et al., 1997, Rasmussen & Perry, 2006, Wang et al., 1996). However, not all nAChR subunits undergo nicotine-induced upregulation. For example, it has been reported that α4β2 nAChRs that also contain the α5 subunit (α4β2α5 nAChRs) do not upregulate in response to nicotine treatment (Mao et al., 2008). Autoradiographic studies have demonstrated that chronic nicotine exposure decreases the number of α6-containing receptors (Lai et al., 2005, Marks et al., 2014, Mugnaini et al., 2006, Perry et al., 2007). However, in the presence of β3 subunits this phenomenon is not observed and the α6-containing receptors do not downregulate, suggesting that one role of β3 is to inhibit nicotine-induced downregulation (Perry et al., 2007). Although the typical serum concentration of nicotine in smokers has little effect on muscle nAChRs (Lindstrom, 1996), upregulation of ganglionic α3β4 and α7 subunits, as well as the muscle type nAChR subunits α1, β1, γ, and δ, has been observed on the cell surface after a 48 hour exposure to nicotine at concentrations as low as 1μM (Ke et al., 1998). These findings confirm results from two previous studies on muscle nAChRs (Luther et al., 1989, Siegel & Lukas, 1988). However, earlier studies found that nAChRs in muscle cells are downregulated 40% by chronic exposure to nicotinic agonists (Appel et al., 1981, Gardner & Fambrough, 1979, Noble et al., 1978) at concentrations ranging from 10⁻³ M to 10 μM (for a review see (Lindstrom, 1996)). Ke et al. (1998) also found that different concentrations of nicotine upregulate different receptor subunits to varying degrees. For example, α3β4 subunits were more prone to upregulation than muscle nAChRs.

Subunit mRNA is unchanged in response to nicotine

Surprisingly, nicotinic receptor subunit mRNA is unchanged in response to nicotine. Early studies by Marks and colleagues showed that while there is an upregulation in α4β2 nAChR binding levels, mRNA for these nAChR subunits as well as for the α2, α3, α5 and β4 subunits remains steady under a nicotine dose of 4 mg/kg/hr (similar to the 5 mg/kg/hr dose originally shown to produce nAChR upregulation in the mouse (Marks et al., 1983)) (Marks et al., 1992). These results were replicated in subsequent studies performed in mouse cell lines (Bencherif et al., 1995, Peng et al., 1994, Zhang et al., 1995), human cell lines (Peng et al., 1997) and rat brain (Bencherif et al., 1995), and are consistent under nicotine concentrations ranging from 10 nM to 5 μM. Ke et al. (1998) expanded the aforementioned work in cultured human cell lines and reported that changes in nAChR numbers are not attributable to changes in CHRNA1, CHRNA3, CHRNA5, CHRNA7, CHRNB2, CHRNB4, CHRND, or CHRNG mRNA levels under exposure to 1 mM nicotine (Ke et al., 1998). It is important to keep in mind that many of the cell lines utilized do not endogenously express nAChRs. Thus, nAChR expression is not controlled by the endogenous regulatory sequences in each cell line, which could confound the results but also clearly demonstrates that native regulatory sequences are not required for upregulation. Although there have been many studies on nAChR subunit mRNA in human brains, these studies have primarily focused on nAChR expression changes in Parkinson’s and Alzheimer’s Diseases, or in aging brains, and to date only one study was conducted to examine the correlation between nAChR subunit protein and mRNA in smokers. In 2003, Mousavi et al. found that protein levels of the α4 and α7 nAChR subunits were significantly increased in the temporal cortex of smokers compared to those of nonsmokers. As in the rodent and cell line studies, there were no differences in CHRNA4 and CHRNA7 mRNA in the temporal cortex of smokers as compared to those of nonsmokers (Mousavi et al., 2003). Thus, all current evidence strongly indicates that the nicotine-induced upregulation of nAChR numbers is independent of transcriptional events.

Upregulation in development

Maternal smoking presents a large public health concern (Mathews, 2001) and research has indicated prenatal nicotine exposure can lead to altered behavior, potentially through interactions with the nAChRs. Numerous studies have found evidence for nAChR (particularly the α4β2-containing and α7 receptors) upregulation in rodents exposed to nicotine during prenatal (Hagino & Lee, 1985, Navarro et al., 1989, Popke et al., 1997, Slotkin et al., 1987, Tizabi & Perry, 2000, Tizabi et al., 1997, van de Kamp & Collins, 1994), neonatal (Eriksson et al., 2000, Huang & Winzer-Serhan, 2006, Miao et al., 1998) and postnatal (Narayanan et al., 2002, Nunes-Freitas et al., 2011) development. Nicotine exposure in each of these studies differed with each experimental objective. The majority of the studies looking at prenatal nicotine exposure used dosages around 6 or 9 mg/kg/day in rats and 2 mg/kg/hr in mice, which is roughly in agreement with what was used previously. However, Hagino and Lee’s 1985 study used a slightly lower dose of 175 μg/0.9μl/hr in rats with similar findings (Hagino & Lee, 1985). The doses used were much more varied in neonatal mice, yet the results were nevertheless consistent. Postnatal rats were treated with 2, 4, or 6 mg/kg/day nicotine, similar to the dose used by Schwartz and Kellar (Schwartz & Kellar, 1983). One study found evidence for downregulation of α7 receptors in rats after exposure to 9 mg/kg/day nicotine during gestation (Tizabi et al., 2000). Concurrent prenatal and postnatal chronic nicotine administration has also induced receptor upregulation in monkeys (Slotkin et al., 2002). Interestingly, nAChR upregulation during rodent fetal development was more long-lasting than in the adult, implying nicotine has some action beyond immediate receptor upregulation (Slotkin et al., 1987, van de Kamp & Collins, 1994). However, the nicotine doses were slightly different than those used in the first studies showing upregulation; 2 mg/kg/hr verses 5 mg/kg/hr nicotine (Marks et al., 1983) in mice and 6 mg/kg/day verses 4 mg/kg/day (Schwartz & Kellar, 1983) in rats. Despite the fact that nAChR upregulation after nicotine exposure in utero does not persist into adolescence or adulthood, prenatal nicotine exposure may, nonetheless, alter the response to nicotine in adolescence or adulthood in rodents (Abreu-Villaca et al., 2004, Gold et al., 2009, Slotkin et al., 2007). Unlike in adulthood, pups born from dams exposed to 2–4 mg/kg/day nicotine have increased nAChR mRNA expression, particularly that of the α7, α4, and β2 subunits (Frank et al., 2001, Lv et al., 2008, Shacka & Robinson, 1998), as well as the α2 subunit (Lv et al., 2008), immediately after nicotine administration is discontinued. Subsequent work has suggested that adolescent rats exposed to 2 mg/kg/day nicotine during gestation show corresponding decreases in nAChR number, as measured by [¹²⁵I]-epibatidine, and expression of several nAChR subunit mRNAs (Chen et al., 2005). Neonatal nicotine exposure is more similar to nicotine exposure in adult animals in that no change in mRNA expression of the receptors is seen across a range of nicotine concentrations (Huang & Winzer-Serhan, 2006, Miao et al., 1998). Evidence for both receptor upregulation and subunit mRNA expression has also been seen in primary cultures from prenatal human brain exposed to 10 or 100 μM nicotine (Hellstrom-Lindahl et al., 2001). Using a mouse model, work by Picciotto and colleagues have demonstrated that the effects of nicotine exposure during development on later behaviors are attributable to the neuropharmacological effects of nicotine itself, not the effect of nicotine on maternal behaviors (Heath et al., 2010a). Furthermore, the nAChRs are suggested to mediate this effect. Mice developmentally exposed to nicotine though their drinking water (200 μg/ml nicotine) display hypersensitive passive avoidance behavior in adulthood; this behavior is likely driven by α4β2 and α4β2α5 nAChRs on corticothalamic neurons. Specifically, expression of the β2 and α5 subunits is necessary for this behavior (Heath et al., 2010b). The studies above present clear evidence that nicotine exposure during early development acts on and affects both nAChR receptor number and mRNA expression. Although outside the scope of this review, the effect of prenatal nicotine exposure on the central nervous system has been reviewed at length (Dwyer et al., 2009, Pauly & Slotkin, 2008, Winzer-Serhan, 2008). Collectively, long term and widespread consequences of such exposures have been proposed to play a role in various neurobehavioral and physiological disorders (Abbott & Winzer-Serhan, 2012).

Nicotine also produces various effects on the nAChRs when administered in adolescence. Upregulation of α4β2-containing and α7 receptors has been seen in rats administered 2–6 mg/kg/day nicotine in adolescence and periadolescence (Doura et al., 2008), particularly in the midbrain, hippocampus, and cortex (Abreu-Villaca et al., 2003, Trauth et al., 1999). Transient upregulation of α7 receptors in rats after exposure of 2 or 6 mg/kg/day nicotine has been seen also in the striatum (Slotkin et al., 2004). Compared to adults, adolescent upregulation is more persistent (up to 1 month post-treatment), has been seen at doses of nicotine as low as 0.6 mg/kg/day in rats, (Abreu-Villaca et al., 2003), and is more robust, showing increased binding in the midbrain and striatum compared to adults (Levin et al., 2007). Consistent with studies that treated adult rats with 3 or 6 mg/kg/day nicotine (Mugnaini et al., 2006, Perry et al., 2007), downregulation of α6-containing receptors was observed immediately in adolescent rats after discontinuing chronic nicotine treatment of roughly 3 mg/kg/day (Doura et al., 2008). Finally, similar to prenatal exposure, adolescent rat exposure to 6 mg/kg/day nicotine has been shown to alter the activity of the cholinergic system in adulthood (Slotkin et al., 2008). For example, one study found an increase in the mRNA expression of the α5, α6, and β2 subunits of adult rats that self-administered nicotine (roughly 20 μl of 0.04 mg/kg for 1 hour) during adolescence (Adriani et al., 2003).

Proposed mechanisms of upregulation

Many studies have been conducted to explore the relation between upregulation of nAChRs and nicotine exposure. Nicotine can pass through the plasma membrane of a cell and has been found in high concentration in cell organelles, as measured in the submaxillary glands of rats by Putney and Borzelleca in 1971 (Putney & Borzelleca, 1971). There are several major models for how nAChR upregulation occurs, some of which are not necessarily mutually exclusive. These include 1) migration from a pool of pre-assembled receptors in the cell, 2) alteration of translation rates via two 2^nd messenger pathways, 3) change in stoichiometry of cell surface receptors, 4) increased intracellular assembly of nAChRs in combination with decreased turnover of surface receptors, 5) increased exocytic trafficking of nAChRs to the cell surface, 6) increased subunit maturation and assembly in the endoplasmic reticulum (ER), possibly through the molecular chaperoning of nAChRs by nicotine, and 7) inhibition of subunit degradation leading to new receptors from recycled subunits (Figure 1). It will be important to determine how these different mechanisms might be integrated during development and possibly modified by exposure to nicotine. It is also important to note that the vast majority of studies examining the mechanism of upregulation have focused on α4β2 nAChRs.

Figure 1.

Cellular processes proposed to play a role in nicotine-induced nAChR upregulation. ER: endoplasmic reticulum.

Early work by Bencherif et al. (1995) suggested that the increase in α4β2 nAChR numbers comes from a pool of assembled receptors within the cell as opposed to altered receptor turnover based on the observation that exposure to 1 μM nicotine had no effect on the degradation of nAChRs in an M10 mouse cell line. They hypothesized that the increase in nicotine binding sites is due a migration of intracellular nAChRs to the surface of the cell and observed that nAChR upregulation had an upward limit of 150%, suggesting the intracellular nAChR reservoir is finite. The authors hypothesized that the intracellular reservoir could exist as a result of several different mechanisms and set the stage for subsequent research (Bencherif et al., 1995).

Another theory is that two 2^nd messenger pathways influence nAChR upregulation, possibly through altering translation rates. A series of three in vitro studies by Gopalakrishnan et al. (1997) showed that cells treated with 1 μM nicotine and cyclohexamide, a protein synthesis inhibitor, were resistant to nicotine-induced α4β2 nAChR upregulation, cells treated with 1 μM nicotine and the second messenger analog dibutyl cyclic adenosine monophosphate (cAMP) showed enhanced nicotine-induced nAChR upregulation, and cells treated with 1 μM nicotine and protein kinase C (PKC) inhibitors failed to produce nicotine-induced nAChR upregulation. This supports the theory that the nicotine-induced nAChR upregulation could be due to two second messenger pathways, one involving PKC and one involving cAMP, either individually or in conjunction with one another (Gopalakrishnan et al., 1997). A potential role for PKC in α4β2 expression was supported by a study by Nashmi et al. (2003), demonstrating increased trafficking of α4β2 nAChRs to the plasma membrane after the activation of PKC in vitro (Nashmi et al., 2003).

In 2005, Vallejo et al. (2005) proposed that upregulation of α4β2 nAChRs is due to a change in ligand affinity of receptors already present on the cell surface. Using a biotinylation assay they determined that nicotine exposure did not alter the ratio of intracellular to surface receptors. The authors experiments also suggested that nAChR upregulation is independent of intracellular receptor trafficking and concluded that what is termed upregulation is the result of receptors changing from a resting state to an active and high affinity state following chronic nicotine exposure (Vallejo et al., 2005). However, subsequent work has provided evidence contradicting this theory. Marks et al. (2011) showed that mice chronically treated with 1, 2, or 4 mg/kg/hr nicotine show a dose-dependent increase in antibody binding to the α4 and β2 nAChR subunits. This work strongly suggests that nicotine-induced nAChR upregulation reflects increased levels of nAChR protein (Marks et al., 2011) and not a conformational change in existing nAChRs enabling them to bind ligands with higher affinity. It is important to note that these studies had vastly different experimental protocols; Vallejo et al. (2005) used a human cell line with heterologous expression of α4β2 nAChRs exposed to 10 μM nicotine (Vallejo et al., 2005) whereas Marks et al. (2011) exposed mice up to 4 mg/kg/hr nicotine (Marks et al., 2011).

Several studies have provided evidence that upregulation of α4β2 nAChRs is due to mechanisms that increase assembly of intracellular nAChRs and decrease the rate of turnover of surface nAChRs (Kuryatov et al., 2005, Peng et al., 1994, Wang et al., 1998). In a mouse fibroblast cell line, expressing α4 and β2 subunits, exposure of the protein synthesis inhibitor cyclohexamide with 5 μM nicotine showed that (−)-[³H]nicotine binding remained higher compared to control cells for a longer time, indicating that nAChR degradation was prolonged (Peng et al., 1994). This work was replicated with the α3 and β2 subunits at a higher nicotine concentration of 100 μM nicotine in HEK cells, and additionally found a very small increase of β2 subunits within the cell despite the 5 fold increase in α3β2 receptors on the cell surface, suggesting the rate of assembly of the complete nicotinic receptor increases with exposure to nicotine (Wang et al., 1998). Finally, Kuryatov et al (2005) proposed that nicotine binds to the acetylcholine binding site of partially assembled α4β2 nAChRs in the lumen of the ER where a plethora of unassembled nAChR subunits reside. Binding of nicotine to an (α4β2)₂ tetramer in the ER would then cause a conformational change that would increase the rate of addition of a 5^th subunit to produce a mature nAChR, thus accounting for nicotine-induced nAChR upregulation (Kuryatov et al., 2005).

Another theory is that nAChR upregulation is a consequence of increased nAChR exocytic trafficking to the cell surface. The authors disrupted the secretory pathway with brefeldin A, a compound that disrupts trafficking from the Golgi apparatus, and found that cells exposed to brefeldin A alone show a decrease in nAChR surface numbers and cells exposed to both brefeldin A and nicotine showed no upregulation of surface nAChRs. This result, however, was not seen when the authors measured intracellular nAChRs (Darsow et al., 2005). These results indicated that transport through the secretory pathway is necessary for upregulation of surface nAChRs but there is likely another mechanism responsible for increasing the intracellular concentration of nAChRs. Although a stark contrast to previous findings, this study used a concentration of 500 nM (Darsow et al., 2005), much smaller than the 1–100 μM concentrations of nicotine utilized in the studies described above (Kuryatov et al., 2005, Peng et al., 1994, Wang et al., 1998).

Similar to some of the work described above, several studies have proposed nAChR upregulation is caused by increased subunit maturation and receptor assembly in the ER (Harkness & Millar, 2002, Nashmi et al., 2003, Sallette et al., 2005). Evidence for this theory has come from various studies using transfected α4 and β2 subunits. In one such study, the amount of total β2 subunit protein, but not α4 subunit protein, folded into correct conformation was influenced by both co-assembled partner subunits (the α4 subunit) and by chronic nicotine treatment of 10 μM, suggesting that nAChR upregulation was not due to increase in total protein but how readily the subunits mature and fold in the ER (Harkness & Millar, 2002). Subsequent studies concluded that increased assembly of nAChRs occurs in the soma of neurons after which these nAChRs must be subsequently trafficked to the dendritic processes where there is a larger overall concentration of nAChRs, and that 1 μM nicotine exposure increases the assembly of nAChRs (Nashmi et al., 2003). Additional evidence supporting this theory showed that α4 and β2 subunits undergo complex oligosaccharide glycosylation, a modification known to only occur in the ER and the Golgi apparatus, before they become heteromeric complexes and, furthermore, these glycosylations were found on all pentameric complexes, indicating that the receptor must be completely assembled into a 5-subunit pentameric complex in order to exit the ER and Golgi. Micromolar levels of nicotine exposure were shown to increase the amount of complex oligosaccharides on nAChR subunits (Sallette et al., 2005). In the same vein, a theory has been proposed to account for increased subunit maturation and receptor assembly after nicotine exposure, termed “inside-out pharmacology.” In short, the main posits of this theory are: nicotine and certain other nicotinic ligands enter the ER or Golgi where they bind and stabilize newly forming nAChR pentamers; this “pharmacological chaperoning” of nAChRs by nicotine results in increased trafficking of nAChRs through the secretory pathway; and, once at the plasma membrane, nAChRs remain stabilized by nicotine thus reducing receptor turnover (Henderson & Lester, 2015). In summary, these studies demonstrate several important findings: nAChR subunits undergo maturation and subsequently are assembled into pentamers in the ER of the cell soma and translocated to the cell surface; this process could escalate upon exposure to nanomolar amounts of nicotine; and this phenomenon might in fact be driven by the pharmacological chaperoning of nAChRs by nicotine (Harkness & Millar, 2002, Henderson & Lester, 2015, Nashmi et al., 2003, Sallette et al., 2005).

Yet another theory poses that subunit degradation is blocked and subunits are consequently recycled to form new nicotinic receptors (Christianson & Green, 2004, Ficklin et al., 2005, Rezvani et al., 2007). Treatment with proteasomal inhibitors has been shown to increase nAChR numbers and assembly in several cells lines (Christianson & Green, 2004). Ubiquilin-1, a ubiquitin-like protein that can interact with both the proteasome and ubiquitin ligases in protein degradation, co-immunoprecipitated with the α3 subunit and reduced the number of surface nAChRs when transfected into cells. While neurons not exposed to nicotine and injected with a ubiquilin-1 lentivirus showed no changes in α3-containing nAChR levels, nicotine-induced upregulation of α3-containing nAChRs was abated in neurons exposed to 100 μM nicotine after injection with a ubiquilin-1 lentivirus (Ficklin et al., 2005). Another study showed that nicotine itself can act as a partial proteasome inhibitor: in the prefrontal cortex of mice treated with 0.5 mg/kg of nicotine every 6 hours for 1 day, nicotine caused an increase in ubiquitinated proteins, including the α7 subunit suggesting reduced proteosomal activity. This inhibition of proteosomal activity was not due to a decrease in expression of the proteasome, as nicotine was found to increase expression of proteasomal subunits (Rezvani et al., 2007). Subsequent work by the same group focused on UBXN2A, a protein containing an ubiquitin-like domain that binds to ubiquitin. This protein was also observed to bind to and alter to α3 subunit expression: over expression of UBXN2A resulted in an increase in α3β2 nAChRs, and expression of UBXN2A was found to decrease levels of α3 subunit ubiquitination (Rezvani et al., 2009).

While the above studies have supported various theories of nicotine-induced nAChR upregulation, many of them are not mutually exclusive. It is important to note that there are receptor-specific assembly folding factors that fold individual nAChR subunits present in some mammalian cells lines and not others (Sweileh et al., 2000), and different nAChR upregulation mechanisms may be impacted by cell line specific folding factors. Recently, work by Govind et al. (2012) has proposed that nAChR upregulation is not due to a single process as suggested by some of the above studies, but multiple processes occurring at different rates. By measuring the kinetics of upregulation in cells exposed to 10 μM nicotine, the authors surmised that upregulation occurred at two different rates: an initial fast component of upregulation that saturated after 4 hours, during which binding of conformation-dependent antibodies occurred suggesting the fast component of upregulation results from nicotine-induced conformational changes as previously reported (Vallejo et al., 2005); and a slower component that increased with continued nicotine exposure, during which proteasomal inhibition produced even higher levels of subunits in the cell and the authors concluded was due to decreased proteasome-dependent subunit degradation and increased subunit assembly in line with previous work (Harkness & Millar, 2002, Nashmi et al., 2003, Rezvani et al., 2007, Sallette et al., 2005). These results suggest that there are many mechanisms, each acting at a different rate, that contribute to nAChR upregulation (Govind et al., 2012). Although the findings of several of the studies presented above directly contrast those of other studies, it is important to consider differences in nicotine exposure and experimental procedure as possible explanations. It is possible these mechanisms are differentially regulated during different developmental stages, under different conditions of nicotine exposure, and with additional variation contributed by genetic differences (discussed below).

Nicotinic acetylcholine receptors and human genetic association studies

Drug use phenotypes

There have been numerous associations between the CHRN genes and various drug phenotypes (Table 1), particularly those relating to nicotine. It is worth noting that although other ethnicities have been used and show association with these genes and drug phenotypes, the majority of these associations have been seen in European American (EA) samples. Many of these associations lie within a cluster of genes on chromosome 15q25: CHRNA3, CHRNB4, and CHRNA5. This cluster of genes has been the most consistently replicated association with cigarettes per day (CPD) (Berrettini et al., 2008, Cannon et al., 2014, Caporaso et al., 2009, Furberg et al., 2010, Gabrielsen et al., 2013, Liu et al., 2010, Saccone et al., 2010a, Sarginson et al., 2011, Sorice et al., 2011, Stevens et al., 2008, Thorgeirsson et al., 2010) and nicotine dependence (ND) (Baker et al., 2009, Bierut et al., 2007, Bierut et al., 2008, Broms et al., 2012, Chen et al., 2009, Erlich et al., 2010, Haller et al., 2012, Maes et al., 2011, Saccone et al., 2009a, Saccone et al., 2009b, Sherva et al., 2010, Spitz et al., 2008, Thorgeirsson et al., 2008, Weiss et al., 2008, Wessel et al., 2010), but more recently the genes for other nAChR subunits have been associated with smoking and other drug phenotypes. CHRNB3-A6 on chromosome 8p11 has reached genome-wide significance levels for association with CPD (Thorgeirsson et al., 2010) and ND (Bierut et al., 2007, Culverhouse et al., 2014, Haller et al., 2012, Hoft et al., 2009b, Rice et al., 2012, Saccone et al., 2009a, Saccone et al., 2010b, Saccone et al., 2007, Wang et al., 2014). In addition, evidence for association between common variants in other CHRN genes (CHRND-G, CHRNB1, CHRNA10, CHRNA4, and CHRNB2) and CPD and ND have been reported, although less replicated (Chen et al., 2013, Han et al., 2011, Kamens et al., 2013, Li et al., 2005, Saccone et al., 2009a, Saccone et al., 2010b).

Table 1.

Association of nAChR markers with phenotypes for human drug use.

Phenotype	Gene(s)	Sample(s)	Reference
NDa	CHRNA4	Families from Anhui Province, China	(Feng et al., 2004)
ADb	CHRNA4	Unrelated Korean males	(Kim et al., 2004)
ND	CHRNA4	Subjects of European or African ancestry from southern US (TN, MS, or AK)	(Li et al., 2005)
Subjective responses to alcohol and tobacco, past 6 month use alcohol	CHRNA4, CHRNB2	CADDc	(Ehringer et al., 2007)
ND	CHRNB3, CHRNA5	COGENDd	(Bierut et al., 2007, Saccone et al., 2007)
Age of initiation of alcohol and tobacco	CHRNA3-B4	CADD, NYSe	(Schlaepfer et al., 2008)
Early subjective response to tobacco	CHRNB3-A6	CADD, Add Healthf	(Zeiger et al., 2008)
CPDg	CHRNA5-A3	Three European populations	(Berrettini et al., 2008)
AD	CHRNA5-A3	COGAh	(Wang et al., 2009)
ND	CHRNA5-A3	>11,000 Icelandic and European samples	(Thorgeirsson et al., 2008)
“Pleasurable buzz” during early experimentation with smoking	CHRNA5	American Caucasians and African Americans	(Sherva et al., 2008)
ND	CHRNA5-A3-B4	COGA	(Bierut et al., 2008)
Age-dependent ND	CHRNA5-A3-B4	Three American cohorts	(Weiss et al., 2008)
ND	CHRNB3-A6	NYS	(Hoft et al., 2009b)
ND	CHRNA5-A3-B4, CHRNB3-A6, CHRND-G	COGEND	(Saccone et al., 2009a)
Cocaine dependence	CHRNA5	FSCDi, COGA	(Grucza et al., 2008)
Heavy smoking	CHRNA5-A3-B4	American Cancer Society CPS-II Cohort, and the CPS-II Nutrition Cohort	(Stevens et al., 2008)
Level of response to alcohol	CHRNA5-A3-B4	San Diego Sibling Pair	(Joslyn et al., 2008)
ND	CHRNA5-A3-B4	Two Texas samples	(Spitz et al., 2008)
Smoking cessation	CHRNB2	Pharmacogenetic trial of bupropion for smoking cessation	(Conti et al., 2008)
Smoking cessation	CHRNA5, CHRNA2	Pharmacogenetic trial of bupropion for smoking cessation	(Heitjan et al., 2008)
Smoking cessation during pregnancy	CHRNA5-A3-B4	European pregnant women	(Freathy et al., 2009)
CPD	CHRNA5-A3	Cancer Genetic Markers of Susceptibility	(Caporaso et al., 2009)
Alcohol consumption	CHRNB3-A6	NYS Family Study	(Hoft et al., 2009a)
ND	CHRNA5-A3-B4	Utah and Wisconsin cohorts	(Baker et al., 2009)
ND	CHRNA5-A3-B4	African Americans subjects from COGEND	(Saccone et al., 2009b)
ND, AD	CHRNA5-A3	Virginia Adult Twin Study	(Chen et al., 2009)
Heavy alcohol use	CHRNA6	Spanish population	(Landgren et al., 2009)
CPD	CHRNA3	Tobacco and Genetics Consortium, ENGAGEj, Ox-GSKk	(Furberg et al., 2010)
ND	CHRNB3-A6, CHRNA4, CHRNB1, CHRNA10, CHRND-G	COGEND	(Saccone et al., 2010b)
CPD	CHRNA5-A3-B4, CHRNB3-A6	ENGAGE	(Thorgeirsson et al., 2010)
ND in treatment-seeking smokers	Neuronal CHRN genes	Group Health	(Wessel et al., 2010)
Smoking initiation, smoking cessation	CHRNA5-A3-B4	Korean subjects	(Li et al., 2010)
Dizziness to tobacco	CHRNB3, CHRNA10	COGEND	(Ehringer et al., 2010)
Smoking quantity	CHRNA5-A3-B4	Os-GSK	(Liu et al., 2010)
Smoking quantity	CHRNA5-A3-B4	CGASP l	(Saccone et al., 2010a)
ND, cocaine dependence, AD	CHRNA5-A3-B4	Five American cohorts (Uconn, Yale, MUSC, UPENN, McLean)	(Sherva et al., 2010)
Opioid dependence severity, ND severity	CHRNA5-A3-B4	Outpatients in treatment for opioid dependence	(Erlich et al., 2010)
Heavy smoking	CHRNA5-A3-B4	Three German Cohorts (KORA, NCOOP, ESTHER)	(Winterer et al., 2010)
ND	CHRNA5	Yale University Transdisciplinary Tobacco Use Research Center	(Lori et al., 2011)
Subjective response to nicotine	CHRNB2	Daily smokers from the Denver/Boulder area	(Hoft et al., 2011)
Smoking quantity, response to smoking cessation therapy	CHRNA5-A3-B4	Two studies, MT1 and MT2	(Sarginson et al., 2011)
CPD, ND	CHRNA4	Five American cohorts (Uconn, Yale, MUSC, UPENN, McLean)	(Han et al., 2011)
ND	CHRNA5-A3	Virginia Twin Registry	(Maes et al., 2011)
Smoking quantity	CHRNA5-A3-B4	Three Italian populations	(Sorice et al., 2011)
ND	CHRNA4	Five American cohorts (Uconn, Yale, MUSC, UPENN, McLean)	(Xie et al., 2011)
ND	CHRNA5-A3-B4, CHRNB3-A6	COGEND	(Haller et al., 2012)
ND, CPD	CHRNB3	SAGEm	(Rice et al., 2012)
ND, regular drinking	CHRNA5-A3-B4	Finnish Twin Cohort	(Broms et al., 2012)
Objective measures of tobacco exposure	CHRNA5-A3	Six European studies	(Munafo et al., 2012)
Onset of habitual smoking	CHRNA5-A3-B4	COGA	(Kapoor et al., 2012)
Genetic vulnerability to smoking in early-onset smokers	CHRNA5	Meta-Analysis	(Hartz et al., 2012)
Smoking quantity	CHRNA5-A3	STOMPn Genetics Consortium	(David et al., 2012)
General substance use initiation	CHRNA5-A3-B4	Add Health	(Lubke et al., 2012)
Pack years of smoking	CHRNA5-A3	Chinese Han population of chronic obstructive pulmonary disease patients	(Zhou et al., 2012)
ND	CHRNA5-A3-B4, CHRNB3-A6	COGEND	(Chen et al., 2012a)
CPD, age of smoking cessation, response to pharmacologic therapy	CHRNA5-A3-B4	Atherosclerosis Risk in Communities, University of Wisconsin Transdisciplinary Tobacco Use Research Center	(Chen et al., 2012b)
ND	CHRNA4, CHRNB2	Japanese	(Chen et al., 2013)
ND	CHRNA4, CHRNB2	CADD, GADDo	(Kamens et al., 2013)
Alcohol use	CHRNA5-A3-B4	National FINRISK Study, Health 2000 Survey	(Hallfors et al., 2013)
Response to smoking cessation therapies	CHRNA5-A3-B4	Eight clinical trials for smoking cessation	(Bergen et al., 2013)
Smoking quantity, snus consumption	CHRNA5-A3-B4	Nord-Trøndelag Health Study	(Gabrielsen et al., 2013)
ND	CHRNB3	Subjects of European, African, and Asian ancestry from MSTCCp, SAGE, CGEMSq, and KAREr studies	(Cui et al., 2013)
Nicotine intake	CHRNA5-A3-B4	Alaskan natives	(Zhu et al., 2013)
Onset of regular smoking	CHRNA5-A3-B4	Meta-Analysis	(Stephens et al., 2013)
Cocaine use disorder	CHRNB3	SAGE	(Sadler et al., 2014)
CPD	CHRNA5-A3-B4, CHRNB3-A6, CHRNA2	SECASPs	(Cannon et al., 2014)
Frequency of binge drinking	CHRNA4	SECASP	(Coon et al., 2014)
ND	CHRNA2, CHRNA6	European and African Americans from TN, AR, MI, and MS; SAGE	(Wang et al., 2014)
CPD	CHRNB4	COGEND	(Haller et al., 2014b)
Opioid dependence and withdrawal	CHRNA3	SAGE	(Muldoon et al., 2014)
Dizziness at smoking initiation	CHRNA3, CHRNA4, CHRNA6, CHRNA7	Nicotine Dependence in Teens	(Pedneault et al., 2014)
Smoking cessation	CHRNA5-A3-B4	Two smoking cessation trials in African Americans: the Nicotine Gum Study, and Bupropion Study	(Zhu et al., 2014)
Use of smokeless tobacco	CHRNA5-A3	Two studies conducted in India: the International multicenter oral cancer study, and the Mumbai study	(Anantharaman et al., 2014)
Severe withdrawal in a subgroup of smokers with higher lifetime prevalence of depression	CHRNA4	Treatment-seeking smokers from 5 Hungarian cessation centers	(Lazary et al., 2014)
COPD, lung cancer, heavy smoking (pack years)	CHRNA7	Southern and eastern Han Chinese population	(Yang et al., 2015)
AD and cocaine dependence	CHRNA3, CHRNB3	COGA	(Haller et al., 2014a)
ND	CHRNB3-A6	COGA, COGEND, FSCD	(Culverhouse et al., 2014)
CPD	CHRNA5, CHRNB4, CHRNA6	Virginia Twin Study on Adolescent and Behavioral Development	(Clark et al., 2015)
ND	CHRNA5	Meta-analysis	(Olfson et al., 2015)

^aND: Nicotine Dependence

^bAD: Alcohol Dependence

^cCADD: Center on Antisocial Drug Dependence

^dCOGEND: Collaborative Study on the Genetics of Nicotine Dependence

^eNYS: National Youth Survey

^fAdd Health: National Longitudinal Study of Adolescent Health

^gCPD: Cigarettes per Day

^hCOGA: Collaborative Study on the Genetics of Alcoholism

^IFSCD: Family Study of Cocaine Dependence

^jENGAGE: European Network of Genomic and Genetic Epidemiology

^kOx-GSK: Oxford-GlaxoSmithKline

^lCGASP: Consortium for the Genetic Analysis of Smoking Phenotypes

^mSAGE: Study of Addiction: Genetics and Environment

ⁿSTOMP: Study of Tobacco in Minority Populations

^oGADD: Genetics of Antisocial Drug Dependence

^pMSTCC: Mid-South Tobacco Case Control Study

^qCGEMS: Genome Wide Scan of Lung Cancer and Smoking

^rKARE: Korea Association Resource study

^sSECASP: Social and Emotional Contexts of Adolescent Smoking Patterns

Work by Saccone and colleagues have confirmed at least three independent genetic signals in the 15q25 cluster (Saccone et al., 2010a). Twin studies have estimated that ND and CPD are roughly 56–72% and 51–61% heritable in men and women, respectively (Broms et al., 2006, Kendler et al., 1999, Lessov et al., 2004, True et al., 1999b, Vink et al., 2004); however, the most well-replicated of these associations accounts for a very small proportion of the variance in smoking (roughly 1 cigarette per day) (Berrettini & Doyle, 2012, Saccone et al., 2010b). This single nucleotide polymorphism (SNP), rs16969968 (a nonsynonymous SNP residing in CHRNA5, also often tagged by rs1051730), has been consistently associated with ND and smoking behavior (Baker et al., 2009, Berrettini et al., 2008, Bierut et al., 2007, Chen et al., 2012a, Chen et al., 2009, Furberg et al., 2010, Hartz et al., 2012, Liu et al., 2010, Saccone et al., 2010a, Saccone et al., 2007, Sherva et al., 2010, Stevens et al., 2008, Thorgeirsson et al., 2008, Thorgeirsson et al., 2010, Weiss et al., 2008, Winterer et al., 2010). This signal has also been associated with alcohol dependence (AD) (Wang et al., 2009), cocaine dependence (CD) (Grucza et al., 2008, Sherva et al., 2010), opioid dependence (Sherva et al., 2010), and opioid dependence severity (Erlich et al., 2010). Interestingly, studies in EAs have shown that the rs16969968 risk allele for ND is a protective allele for AD (Chen et al., 2009) and CD (Grucza et al., 2008). Furthermore, the finding with CD has been replicated in another EA sample, as well as an African American sample (Sherva et al., 2010). Although the biological mechanisms for the bi-directional effect are unknown, one explanation might be due to the contribution of nAChRs to both excitatory and inhibitory effects on dopaminergic reward pathways (Grucza et al., 2008).

Many of these genes have been associated also with alcohol use. CHRNA5-A3-B4 has been associated with AD (Chen et al., 2009, Sherva et al., 2010, Wang et al., 2009), alcohol use (Broms et al., 2012, Hallfors et al., 2013), age of initiation of alcohol use (Schlaepfer et al., 2008), and level of response to alcohol (Joslyn et al., 2008). CHRNB3-A6 have been associated with alcohol consumption (Hoft et al., 2009a) and heavy alcohol use (Landgren et al., 2009). Finally, CHRNA4 has been associated with AD (Kim et al., 2004) and frequency of binge drinking (Coon et al., 2014). Both CHRNA4 and CHRNB2 have been associated with subjective response to alcohol (Ehringer et al., 2007).

More recently, several groups have used Next Generation sequencing approaches to comprehensively assess the coding regions of CHRN genes. In particular, past evidence for common variant associations in CHRNA4 have been limited. However, sequencing studies have revealed this gene has been technically more challenging to sequence, suggesting there may be genomic features in this region that have impeded traditional SNP assays. In a sample of EAs participating in a smoking cessation trial, Wessel et al. (2010) found evidence for association with ND between common and rare variants in CHRNA5 and CHRNB2, and for rare variants in CHRNA4 (Wessel et al., 2010). At the time, the research team was able to sequence the other CHRN genes using 454 technology, but needed to use Sanger sequencing for CHRNB2 and CHRNA4. In 2011, Xie et al. reported a protective effect of rare alleles in CHRNA4 for ND in a small EA sample (Xie et al., 2011). Some of these variants were shown to lead to significant differences in nAChR expression, subcellular distribution, and sensitivity to nicotine-induced receptor upregulation in HEK 293 cells and Xenopus laevis oocyte studies (Mcclure-Begley et al., 2014). In addition, rare variants in CHRNB4 have been associated with ND (Haller et al., 2012), and rare variants in CHRNB3 and CHRNA3 have been associated with AD and CD (Haller et al., 2014a). More recently, rare variants in CHRNA5 and CHRNB4 (but not CHRNA3) were associated with CPD in an EA sample, as was CHRNA6 (but not CHRNB3) (Clark et al., 2015). Finally, evidence for rare variant associations in both EA and African American samples in the CHRNA5 gene were reported for ND (Olfson et al., 2015). Collectively, the emerging data from rare variant studies suggests there are many different ways in which genetic variation in the CHRN genes may lead to risk for increased smoking. It will be important to draw on these databases of naturally occurring rare variants to perform follow-up functional molecular work and complementary animal studies in order to understand the molecular and pharmacological mechanisms for different kinds of variants.

In addition, recent pharmacogenetic studies have demonstrated that risk haplotype at CHRNA5-A3-B4 is highly predictive of a person’s response to treatment with Varenicline (a nAChR agonist) (Bergen et al., 2013, Chen & Bierut, 2013). To put this in terms of number needed to treat (NNT) for one person to successfully quit smoking, the NNT is 4 among individuals who carry the risk haplotype in the region, whereas among those with the low-risk haplotype the NNT >1000. Thus the collective work of many who have studied the human CHRN genes over the last 10 years is beginning to bear fruit with regard to potential future clinical application.

Age-related effects

Although rs16969968 has been robustly associated with ND and CPD as described above, recent research has indicated that this signal is not associated with age of onset of regular smoking or age of smoking initiation (Furberg et al., 2010, Hartz et al., 2012, Stephens et al., 2013). While two studies, including a large-scale meta-analysis, have found that age of onset of smoking modifies the association between rs16969968 and smoking severity in that that the risk allele for rs16969968 is associated with severity of heavy smoking and ND in smokers who began smoking before age 16 (Hartz et al., 2012, Weiss et al., 2008), research has primarily shown no association between rs16969968 and age of onset of smoking (Hartz et al., 2012, Stephens et al., 2013) and age of initiation of smoking (Furberg et al., 2010, Stephens et al., 2013, Thorgeirsson et al., 2010).

A recent meta-analysis examined two signals independent of rs16969968 (rs578776 and rs1948/rs684513) and found evidence for association with age of onset of smoking but not age of initiation of smoking (Stephens et al., 2013). Although this did not replicate previous associations with rs1948 and rs684513 and early age of tobacco initiation in two samples (Schlaepfer et al., 2008), it is in agreement with two previous genome-wide association studies (Furberg et al., 2010, Thorgeirsson et al., 2010) and one candidate gene study (Winterer et al., 2010) in finding no association with age of smoking initiation. Previous associations have been identified between rs684513 and age of onset of smoking (Broms et al., 2012), heavy smoking (Winterer et al., 2010), and ND (Broms et al., 2012, Lori et al., 2011), and between rs1948 and the tolerance factor of the Nicotine Dependence Syndrome Scale (Broms et al., 2012). In the meta-analysis by Stephens and colleagues in 2013 rs578776 had the strongest association with age of onset of smoking (Stephens et al., 2013), replicating previous work (Broms et al., 2012). This SNP has previously been associated with ND (Broms et al., 2012, Saccone et al., 2010a, Saccone et al., 2009a, Saccone et al., 2009b, Weiss et al., 2008) and heavy smoking (Stevens et al., 2008). Although the evidence for signals in the CHRN gene cluster on chromosome 15q25 remains mixed for association with age of initiation, studies suggest that some of the signals in the cluster that affect onset of smoking may be different from those affecting dependence and heavy smoking.

Conclusion

Chronic exposure by an agonist induces an upregulation of nAChRs that is not due to increased gene expression in adults. Numerous mechanisms have been proposed to be responsible for this phenomenon: increased nAChR subunit maturation and folding, increased receptor trafficking, and decreased subunit degradation, possibly through the pharmacological chaperoning of nAChRs by nicotine; and increased translation and 2^nd messenger signaling. While upregulation also occurs during prenatal, neonatal, and postnatal developmental periods, as well as adolescence, there have been conflicting results regarding a change in corresponding gene expression. However, there is strong evidence that nicotine exposure during these influential periods of development contributes to later changes in the cholinergic system. Results from human genetics studies have provided also strong indication that variation with the CHRN genes is associated with SUDs. Most of the human genetic studies on the nAChRs however have focused on specific associations with nicotine behaviors and have shown several well-replicated signals that are associated with smoking quantity and ND. However, recent work has demonstrated that the most well-replicated of these associations with smoking levels and ND is not associated with age of smoking initiation or age of regular smoking. These differences highlight the overall complexity of molecular genetics in context of similarly heterogeneous phenotypes. As human geneticists continue to search for genes associated with drug behaviors, it is clear that careful attention to the phenotype, and developmental stage of subjects, will be an important consideration for obtaining a global view of how genetics influence behavior over the life-course. Although this review has focused on the nAChRs, other genes are likely to contribute at unique stages of development in concert with nicotine and other drug exposures to modify risk for SUDs.