In this study we propose to test the hypothesis that a SNP influencing smoking addiction will reduce the likelihood of giving up smoking just before or in pregnancy, and in turn affect fetal growth and gestational age. We would also like to test whether maternal genotype has longer term effects on childhood metabolic and growth outcomes, through "programming" to an adverse intra-uterine environment. Finally if, available , the effects of maternal genotype on paternal smoking behaviour and paternal genotype on paternal smoking will also be interesting to test.

Decode Genetics reported at a Keystone Meeting (Santa Fe, Genetics of complex traits) in March 2008 that a SNP, rs1051730, in the CHRNA3 (cholinergic receptor, nicotinic, alpha) gene is associated with the number of cigarettes smoked per day within smokers. The nimor allele frequency varies from 0.30 in light smokers (1-10 cigarettes per day) to 0.40 in people who smoke greater than 31 cigarettes per day. The odds ratio for being a heavy smoker compared to a light smoker (1.41) is greater than that for being a smoker at all (1.17) indicating the association is about addiction to nicotine rather than taking up smoking in the first place. The SNP is associated with lung cancer and peripheral artery disease in the direction expected, although with stronger effects than predicted from the simple SNP vs smoking quantity and smoking quantity vs disease associations.

Specific hypotheses:

1. Genotype predicts smoking behavior before during and after pregnancy.

2. Maternal genotype influences offspring fetal growth and gestational age.

3. Maternal genotype infleunces offspring growth and metabolism in childhood.

4. Offspring genotype influences fetal growth and gestational age (independently of maternal genotype - fetal genotype may alter in utero sensitivity to the toxic effects of nicotine).

5. If paternal DNA becomes available during the course of the project we would be keen to test fathers as well to test the hypothesis that the SNP alters expectant fathers' ability to give up smoking when their partners are pregnant. (Indeed without paternal DNA we could still test whether maternal genotype influenced paternal behaviour - although without paternal genotype this test will be less powerful)

* Specific ALSPAC phenotypes being considered:

To do this we would like to genotype (at Kbioscience) all ~20,000 ALSPAC samples. We will need the following phenotypes to test our hypotheses:

1. Full details of maternal smoking details before, during and after pregnancy,

2. Birth weight, length and head circumference. Gestational age as offspring primary outcomes

3. Growth measures in childhood (height, weight and BMI aged 7-11)

4. Covariates of birth weight to check if genotype is acting through them: maternal age, maternal BMI, smoking , parity, twin status to exclude non-singletons, ethnicity as genotype frequency may alter with ethnic origin and confound analyses.

Genotyping of one SNP will only use 5-10ng of DNA per sample based on the current Kbioscience techniques.