One SNP linked to two diseases—addiction and cancer: A Double Whammy? Nicotine addiction and lung cancer susceptibility

A new genome-wide association study provides compelling evidence that a polymorphic variation linked to a nicotinic receptor subunit cluster contributes to increased risk of becoming addicted to nicotine—presumably through the expression of these receptors in brain regions implicated in reward—and of suffering from lung cancer through the expression of these receptors in epithelial cells in the lung.

Hype, hope and hoops

Almost 90 years ago, James Ewing (1866–1943), a physician founder of the American Cancer Society, proposed that heredity is likely to contribute to the etiology of cancer through the indirect transmission of enhanced liability.¹ It took us many decades to fully grasp the meaning of this concept, which encompasses not only the primary role of oncogenic genes, but also a much broader set of interacting genetic, environmental and behavioral processes. It is hard to imagine a better example of Ewing's ‘enhance liability factor’ than the case of a single-nucleotide polymorphism working at the crossroads between facilitating behaviors conducive to nicotine dependence and creating a permissive environment for the development of lung cancer. Yet, this is precisely the conclusion we are invited to draw from a recent report² that illustrates the ability of genome-wide association studies (GWAS) to detect genetic contributions that increase not only the risk of complex bio-behavioral disorders such as addiction, but also the risk of cancer.

The evidence that heredity modulates the risk of nicotine addiction,³ led to the expectation that genetic advances would help us identify the genetic basis of substance use disorders in general and of nicotine addiction in particular. Unfortunately, the full potential of genome scans took longer to crystallize than expected, mainly due to the poor rate of replicability among the initial studies. But improvements in genome-wide association strategies kept eroding the barriers for establishing reliable connections (for example, small size effects, ill-defined endophenotypes, multiple test penalty, cost and so on).

The deCODE project offers a preview of things to come.² In this study the genomes of close to 11 000 Icelandic smokers were interrogated with a platform displaying over 306 000 single-nucleotide polymorphisms. The report shows that a cluster containing the α3, α5 and β4 gene subunits of the nicotinic acetylcholine receptor (nAChR) had a strong effect on nicotine dependence and the risk of lung cancer and peripheral arterial disease. Moreover, the frequency of the variant increases with addiction severity (addiction here is used interchangeably with the term dependence as per DSM IV).⁴

The findings are remarkable for two reasons. First, they converge with independently gathered data that had identified nonsynonymous variations in the same gene cluster as significantly associated with nicotine dependence.^5,6 Second, they are also consistent with two independent genome-wide association studies that point to the same locus as a modulator of susceptibility for lung cancer.^7,8 In addition, there may be an element of added interest in that the α3 and α5 genes partially overlap at their 3′ ends in a tail-to-tail configuration (Figure 1a). As these genes are coexpressed in some areas of the central nervous system (for example, striatum, habenula) and periphery (for example, bronchial epithelial cells) (Figure 1b), their natural antisense transcripts could give rise to RNA–RNA duplexes with potential regulatory functions.⁹

Figure 1.

Convergent evidence has established that a polymorphism in the α3/α5/β4 nicotinic receptor subunits cluster is associated with increased risk of nicotine dependence (ND) and of two of its most devastating medical consequences, lung cancer (LC) and peripheral arterial disease (PAD)—solid arrow path on right side. We do not know at this time whether the structural link between the members of this nicotinic acetylcholine receptor (nAChR) subunit cluster (a) plays a role in the association; whether the polymorphism in this cluster can directly increase the sensitivity of the target tissues to tobacco smoke and/or nicotine toxicity—dashed arrow connecting (a and b); or whether all the increase in lung cancer and peripheral arterial disease can be indirectly ascribed to the behavioral effects of the α3/α5/β4 variant on smoking quantity and risk of nicotine addiction (c). (The approximate distributions of the α3 (red), α5 (blue) and β4 (gray) nAChR subunits are schematically depicted in the coronal section of a mouse brain shown on the right side of (b).)

A ‘double whammy’ effect?

It will be important next to try and tease apart the relationships between α3/α5/β4 variants on one hand and the etiology of nicotine dependence and of lung cancer on the other. It is reasonable to propose that nicotine dependence is partly responsible for the observed smoking associated pathology (Figure 1c), and that the increased risk of lung cancer is a good example of gene by environment interaction. This notion is supported by the behavioral phenotype of α5 subunit knock-out mice, which display very low sensitivity to nicotine-induced behaviors and seizures¹⁰ and fail to display the withdrawal signs normally seen after discontinuation of chronic nicotine.¹¹ Taken together, these findings suggest that α5 plays an important role in the short- and long-term psychoactive effects of nicotine.

However, the hypothesis of a ‘double whammy’ scenario is supported by the fact that the risk of lung cancer that can be attributed to the α3/α5/β4 variant is higher than what one could explain by the variant's effect on smoking quantity.² This suggests a direct link between α3/α5/β4 and lung cancer mediated by non-neuronal nAChRs, which are engaged peripherally in functions beyond neurotransmission. Consistent with this notion, nicotinic receptor antagonists can block the nitrosa-mine-dependent malignant transformation of respiratory cells in vitro.¹² In addition, tobacco smoke and nicotine can both mediate the stepwise overexpression of nAChR subtypes, which leads to increased Ca^{+ +} permeability in exposed cells. Thus, a switch in the nAChR composition (involving α3 and α5 subunits, among others) could change receptor function, leading to profound pathologic effects in cells exposed to nicotine.¹³ To investigate this hypothesis further, it may now be interesting to revisit the early evidence suggesting a role for nAChRs in the regulation of cell-to-cell communications, adhesion and motility.¹⁴

Therefore, α3/α5/β4-linked variations could contribute to increased risk of nicotine dependence and to lung cancer, independently and on two levels, (a) by increasing the number of cigarettes smoked and the likelihood of nicotine dependence and (b) by inserting themselves right into the pathophysiologic cascade that leads to lung cancer (Figure 1b). As the authors of the study are careful to point out, reaching a conclusion as to the actual weight of each of these pathways will require studies that take into account other aspects of smoking behavior (for example, age of initiation, puff duration, and depth of inhalation) that may in fact contribute to much of the remaining risk.

Clinical and research implications

During his address on ‘The causal and formal genesis of cancer’ at the London Cancer Conference in 1928, Dr Ewing remarked that ‘one may hardly aim to eliminate the tobacco habit, but cancer propaganda should emphasize the danger signs that go with it.’ Fortunately, the message has been heard loud and clear; an ongoing massive education campaign can be credited with dramatic reductions in smoking prevalence. On the other hand, Dr Ewing's pessimism was equally justified: despite major advances in the last half century, cigarette smoking remains the leading cause of preventable disease and mortality in the United States, causing some five million premature deaths around the world every year. It is apparent that the success of our antismoking campaign efforts may have reached a plateau and that the acceleration of scientific discovery offers the best chances of spurring further progress in reducing tobacco-related diseases.

High quality genetic risk information similar to that provided by the deCODE study, opens up exciting new avenues for investigating the functional role of specific nAChR subunits on central nervous system circuit plasticity and on the physiology of peripheral tissues affected by smoking. Studies on the role of α3 and α5 nicotinic receptor subunits in the primate habenula-interpeduncular tract,^15,16 for example, represent a natural path for further inquiry, as this axis is involved in a variety of brain functions and behaviors that include reward phenomena, anxiety and depression.¹⁷ Thus, the connection provides fertile grounds for future investigations into the establishment and maintenance of smoking behaviors from the vantage point of the self-medication theory of (nicotine) addiction.

Equally important, the recent data suggest the α3/α5/β4 subunit cluster may harbor a promising and much needed new target for the development of next generation, nonclassical pharmacotherapeutics designed to treat not only nicotine addictions but also lung cancer, which may involve the dysregulation of epithelial and other α3/α5/β4-expressing cells.

In summary, research that takes advantage of genome-wide association studies is likely to enhance our understanding of the neurobiological processes underlying vulnerability for nicotine addiction and its medical consequences, and help us achieve further reductions in this deadly behavior.