Length polymorphism in the Period 3 gene is associated with sleepiness and maladaptive circadian phase in night-shift workers CHRISTOPHER L. DRAKE 1 , REN BELCHER 1,4 , RYAN HOWARD 1,5 , THOMAS ROTH 1 , ALBERT M. LEVIN 2 and VALENTINA GUMENYUK 1,3 1 Sleep Disorders and Research Center, Henry Ford Hospital, Detroit, MI, USA, 2 Department of Public Health Sciences, Henry Ford Hospital, Detroit, MI, USA, 3 Magnetoencephalography Laboratory, Meadowlands Hospital, Secaucus, NJ, USA, 4 Pritzker School of Medicine, University of Chicago, Chicago, IL, USA, and 5 University of Michigan Medical School, Ann Arbor, MI, USA Keywords circadian rhythms, DLMO, Fatigue, MSLT, PER 3, shift work disorder Correspondence Christopher L. Drake, PhD, Sleep Disorders and Research Center, Henry Ford Hospital, 2799 W. Grand Boulevard, CFP-3, Detroit, MI 48202, USA. Tel.: 313-916-4455; fax: 313-916-5167; e-mail: cdrake1@hfhs.org Accepted in revised form 25 October 2014; received 26 May 2014 DOI: 10.1111/jsr.12264 SUMMARY The objective of the current study was to determine if night-shift workers carrying the five-repeat variant of the Period 3 gene show elevated levels of nocturnal sleepiness and earlier circadian phase compared with homozygotes for the four-repeat allele. Twenty-four permanent night-shift workers were randomly selected from a larger study. Participants took part in an observational laboratory protocol including an overnight multiple sleep latency test and half-hourly saliva collection for calculation of dim-light melatonin onset. Period 3 –/5 shift workers had significantly lower multiple sleep latency test during overnight work hours compared with Period 3 4/4 workers (3.52 Æ 23.44 min versus 10.39 Æ 6.41 min, P = 0.003). We observed no significant difference in sleepiness during early morning hours following acute sleep deprivation. Long-allele carriers indicated significantly higher sleepiness on the Epworth Sleep- iness Scale administered at 17:00 hours (12.08 Æ 2.55 versus 8.00 Æ 1.94, P < 0.001). We observed a significantly earlier melatonin onset in Period 3 –/5 individuals compared with Period 3 4/4 shift workers (20:44 Æ 6:37 versus 02:46 Æ 4:58, P = 0.021). Regression analysis suggests that Period 3 genotype independently predicts sleepiness even after controlling for variations in circadian phase, but we were unable to link Period 3 to circadian phase when controlling for sleepiness. Period 3 –/5 shift workers showed both subjective and objective sleepiness in the pathological range, while their Period 3 4/4 counterparts showed sleep- iness within normal limits. Period 3 –/5 night workers also show a mean circadian phase 6 h earlier (i.e. less adapted) than Period 3 4/4 workers. Because Period 3 –/5 workers have maladaptive circadian phase as well as pathological levels of sleepiness, they may be at greater risk for occupational and automotive accidents. We interpret these findings as a call for future research on the role of Period 3 in sleepiness and circadian phase, especially as they relate to night work. INTRODUCTION A variable number tandem repeat (VNTR) polymorphism of the Period 3 (PER3) gene has been associated with individual differences in several sleep and circadian functions (Dijk and Archer, 2010; Goel et al., 2009; Groeger et al., 2008; Rupp et al., 2013; Viola et al., 2007). Carriers of the less common five-repeat allele (PER3 4/5 or PER3 5/5 ; here- after noted PER3 –/5 ) make up about 35–40% of the popu- lation (Ciarleglio et al., 2008), and PER3 5/5 individuals are more susceptible to cognitive decrements following sleep deprivation, particularly at the circadian nadir (~05:00 hours; Viola et al., 2007). A recent study comparing PER3 geno- types showed that PER3 5/5 individuals had higher indicators ª 2014 European Sleep Research Society 1 J Sleep Res. (2014) Regular Research Paper