American Journal of Medical Genetics 125A:285–289 (2004) Limb Deficiency Defects, MSX1, and Exposure to Tobacco Smoke Suzan L. Carmichael, 1 * Gary M. Shaw, 1 Wei Yang, 1 Edward J. Lammer, 2 Huiping Zhu, 3 and Richard H. Finnell 3 1 March of Dimes Birth Defect Foundation/California Department of Health Services, California Birth Defects Monitoring Program, Oakland, California 2 Children’s Hospital Research Institute, Oakland, California 3 Institute of Biosciences and Technology, Texas A&M University System Health Science Center, Houston, Texas There is increasing evidence from epidemio- logic studies that genetic susceptibilities may modify the teratogenic effects of smok- ing. A previous study suggested that mater- nal smoking in the presence of a dinucleotide repeat polymorphism for MSX1 produced an almost fivefold increased risk for limb anomalies, providing evidence for a gene– environment interaction. The current study examined this potential interaction, using case-control data with several methodologic improvements, including a larger sample size and more detailed information on cigar- ette smoke exposures. Cases (n ¼ 92) were ascertained from pregnancies ending in 1987–1989, and controls (n ¼ 180) were ran- domly selected from eligible liveborn infants. In telephone interviews, women reported smoking behaviors during the month before pregnancy through the end of the first trime- ster. Odds ratios (OR) for maternal and paternal smoking ranged from 1.0 to 1.4, risk estimates were imprecise; for example, the OR for maternal smoking 20 cigarettes per day, versus none, was 1.3 (95% confidence interval (CI) 0.5–3.4). Relative to the homo- zygous wildtype, the OR was 1.5 (95% CI 0.7 – 3.5) for the homozygous variant genotype and 0.8 (95% CI 0.5 – 1.4) for the heterozygous variant genotype. There was no evidence that maternal smoking or both parents smoking, in combination with a susceptible MSX1 genotype, conferred an additional increase in risk of limb defects. This study did not find a gene – environment interaction between maternal smoking, infant MSX1 CA repeat polymorphism, and risk of limb defi- ciency defects. This finding contrasts with results of a previous study, which provided initial evidence for such an interaction. Seve- ral important methodological differences may have contributed to the differences in findings between the two studies. ß 2003 Wiley-Liss, Inc. KEY WORDS: limb deficiency defects; MSX1; homeobox genes; smoking; pregnancy; conge- nital malformations INTRODUCTION Congenital limb anomalies, which affect approxi- mately 5 per 10,000 births [Stoll et al., 1992], are associated with considerable physical disability and medical care costs. Although the etiology of limb anomalies is poorly understood, maternal smoking has emerged as a risk factor in several studies; reported increases in risk range from approximately 30–100% [Aro et al., 1984; Kallen, 1989; Czeizel et al., 1994; Wasserman et al., 1996; Martinez-Frias et al., 1999]. There is increasing evidence from epidemiologic studies that genetic susceptibilities may modify the teratogenic effects of smoking [Hwang et al., 1995, 1998; Shaw et al., 1996; Romitti et al., 1999; Honein et al., 2000]. It has been proposed that vascular disruptive events or toxic exposures associated with smoking may interfere with the function of variant gene products [Hwang et al., 1998]. Grant sponsor: Centers for Disease Control and Prevention (Center of Excellence Award No. U50/CCU913241); Grant sponsor: US Environmental Protection Agency’s Science to Achieve Results Program; Grant number: 82829201; Grant sponsor: Cigarette and Tobacco Surtax Fund of the California Tobacco-Related Disease Research Program, University of Cali- fornia; Grant number: 1RT466. *Correspondence to: Dr. Suzan L. Carmichael, California Birth Defects Monitoring Program, 1917 Fifth St., Berkeley, CA 94710. E-mail: sca@cbdmp.org Received 19 February 2003; Accepted 21 July 2003 DOI 10.1002/ajmg.a.20517 ß 2003 Wiley-Liss, Inc.