We propose to test the hypothesis that genetic variation at two loci will reduce the likelihood of stopping smoking during the perinatal period. We expect these effects to be independent and will not investigate them in combination.

The first SNP (rs1051730) is located within the CHRNA3/A5 gene cluster, which encodes the nicotinic receptor alpha-3 and alpha-5 sub-units. Itwas originally reported to be associated with nicotine dependence in 2008 (Thorgeirsson et al., 2008), and this association has subsequently replicated in most (but not all) studies published since. It has also been investigated in relation to smoking cessation (e.g., Freathy et al., 2009). The National Institute of Drug Abuse (NIDA) has formed a working group to investigate the strength of evidence for the association of SNPs in chromosome 15 (including rs1051730) with smoking cessation. We propose to share the ALSPAC data published by Freathy and colleagues with this collaborative initiative, given the importance of including large, high-quality data sets of this kind.

The second SNP (rs4680) is located within the COMT gene, which encodes the catechol-O-methyltransferase enzyme, responsible for the degradation of extracellular dopamine, in particular in the prefrontal cortex. It has been widely investigated across a range of psychiatric phenotypes. We have recently reported an association between this SNP and smoking cessation in two clinical trials (Munafo et al., 2008; Johnstone et al., 2007), but subsequent attempts at replication have been inconsistent. One study found evidence of association in two independent samples of females only (Collila et al., 2005). Given evidence that oestrogen modulates COMT gene transcription (e.g., Jiang et al., 2003), it is possible that any effects of COMT genotype may be stronger in females.

Additional data on maternal depression over the same period are requested in order to explore whether any genetic associations with smoking cessation also predict change in depression symptom score, given the uncertainty regarding the direction of causation between cigarette smoking and depression.

Specific hypotheses:

1. The rs1051730 T allele will be associated with a reduced likelihood of smoking cessation;

2. The rs4680 G allele will be associated with a reduced likelihood of smoking cessation.

ALSPAC phenotypes required:

1. Maternal smoking status before, during and after pregnancy;

2. Maternal daily cigarette consumption and years smoked;

3. Maternal depression symptoms (EPDS) before, during and after pregnancy;

4. Maternal ancestry, age, educational attainment.

ALSPAC genotypes required:

1. rs1051730;

2. rs4680.