Proxy gene-by-environment Mendelian randomization study confirms a causal effect of maternal smoking on offspring birthweight, but little evidence of long-term influences on offspring health

Abstract

20Objective To validate a novel proxy gene-by-environment (G×E) Mendelian randomization (MR) 21 approach by replicating the previously established effect of maternal smoking heaviness in 22 pregnancy on offspring birthweight, and then use GxE MR to investigate the effect of smoking 23 heaviness in pregnancy on offspring health outcomes in later life and grandchild's birthweight. 24Design A proxy G×E MR using participants' genotype (i.e. rs16969968 in CHRNA5) as a proxy for their 25 mother's genotype. 26 Setting UK Biobank. 27Participants 289,684 white British men and women aged 40-69 in UK Biobank. 28Main outcome measures Participants' birthweight and later life outcomes (height, body mass index, 29 lung function, asthma, blood pressure, age at menarche, years of education, fluid intelligence score, 30 depression/anxiety, happiness), and birthweight of female participants' first child. 31Results In our proof of principle analysis, each additional smoking-increasing allele was associated 32 with a 0.018 (95% confidence interval (CI): -0.026, -0.009) kg lower birthweight in the "maternal 33 smoking during pregnancy" stratum, but no meaningful effect (-0.002kg; 95% CI: -0.008, 0.003) in 34 the "maternal non-smoking during pregnancy" stratum (interaction P-value=0.004). We found little 35 evidence of an effect of maternal smoking heaviness on participants' later life outcomes. We found 36 the differences in associations of rs16969968 with grandchild's birthweight between grandmothers 37 who did versus did not smoke were heterogeneous (interaction P-value=0.042) among female 38 participants who did (-0.020kg per allele; 95% CI: -0.044, 0.003) versus did not (0.007kg per allele; 39 95% CI: -0.005, 0.020) smoke in pregnancy. 40Conclusions Our study demonstrated how offspring genotype can be used to proxy for mothers' 41 genotype in G×E MR. We confirmed the previously established causal effect of maternal smoking on 42 offspring birthweight but found little evidence of an effect on long-term health outcomes in the 43 WHAT IS ALREADY KNOWN TO THIS TOPIC 47Heavier maternal smoking in pregnancy causes lower offspring birthweight 48Maternal smoking in pregnancy is also associated with offspring outcomes in later life and 49 grandchild's birthweight, but it is not known whether these associations are causal 50Understanding the transgenerational causal effects of maternal smoking heaviness in pregnancy is 51 important to inform public health policies 52 WHAT THIS STUDY ADDS 53The proxy gene-by-environment Mendelian randomization approach can be used to explore 54 maternal effects on offspring phenotypes when maternal genetic information is unavailable 55The approach confirmed the causal effect of smoking on offspring birthweight. 56Maternal smoking status in pregnancy modulates the effect of grandmother's smoking heaviness in 57 pregnancy on grandchild's birthweight, highlighting the importance of smoking cessation before 58 pregnancy in each generation 59Outcomes in participants (G1) 132

How humans can contribute to Mendelian randomization analyses

“…MR studies are now extending this design to estimate causal effects of phenotypes on offspring outcomes, for example demonstrating that maternal smoking does lower offspring birthweight, but has little influence on other offspring health outcomes. 33 In this issue of The International Journal of Epidemiolgy , Evans et al . 34 develop a novel approach to inferring the effects of parental exposures on offspring outcomes within a structural equation modelling (SEM) framework.…”