Variants in calcium voltage-gated channel subunit Alpha1 C-gene (CACNA1C) are associated with sleep latency in infants

RESEARCH ARTICLE Va ian s in calcium ol age-ga ed channel subuni Alpha1 C-gene (CACNA1C) a e associa ed wi h sleep la ency in in an s Ka i Kan oja ¨ i 1,2 , Johanna Liuhanen 1,2 , Ou i Saa enpa ¨a ¨-Heikkila ¨ 3 , Anna-Liisa Sa omaa 4 , Anneli Kyllia ¨inen 5 , Pi jo Po ¨lkki 6 , Julia Jaa ela 1,2 , Auli Toi ola 1,2 , Lili Milani 7 , Sa i- Leena Himanen 4,8 , Ta ja Po kka-Heiskanen 9 , Juulia Paa onen 10 , Tiina Paunio 1,2 * 1Genomics and Bioma ke s Uni , Na ional Ins i u e o Heal h and Wel a e, Helsinki, Finland, 2Depa men o Psychia y, Uni e si y o Helsinki and Helsinki Uni e si y Cen al Hospi al, Helsinki, Finland, 3Human In o ma ion P ocessing–labo a o y, Uni e si y o Tampe e, Tampe e, Finland, 4Depa men o Clinical Neu ophysiology, Tampe e Uni e si y Hospi al, Medical Imaging Cen e and Hospi al Pha macy, Pi kanmaa Hospi al Dis ic , Tampe e, Finland, 5School o Social Sciences and Humani ies/Psychology, Uni e si y o Tampe e, Tampe e, Finland, 6Depa men o Social Sciences, Uni e si y o Eas e n Finland, Kuopio, Finland, 7The Es onian Genome Cen e , Uni e si y o Ta u, Ta u, Es onia, 8Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Tampe e, Finland, 9Ins i u e o Biomedicine/Physiology, Uni e si y o Helsinki, Helsinki, Finland, 10 Child and Adolescen Men al Heal h, Na ional Ins i u e o Heal h and Wel a e, Helsinki, Finland *[email p o ec ed] Abs ac Gene ic a ian s in CACNA1C (calcium ol age-ga ed channel subuni alpha1 C) a e associ- a ed wi h bipola diso de and schizoph enia whe e sleep dis u bances a e common. In an expe imen al model, Cacna1c has been ound o modula e he elec ophysiological a chi ec- u e o sleep. The e a e s ong gene ic in luences o consolida ion o sleep in in ancy, bu only a ew s udies ha e hus a esea ched he gene ic ac o s unde lying he p ocess. We hypo hesized ha gene ic a ian s in CACNA1C a ec he egula ion o sleep in ea ly de el- opmen . Se en a ian s ha we e ea lie associa ed (genome-wide signi ican ly) wi h psy- chia ic diso de s a CACNA1C we e selec ed o analyses. The s udy sample consis s o 1086 in an s (520 gi ls and 566 boys) om he Finnish CHILD-SLEEP bi h coho (geno- yped by Illumina In inium PsychA ay BeadChip). Sleep leng h, la ency, and nigh ly awa- kenings we e epo ed by he pa en s o he in an s wi h a home-deli e ed ques ionnai e a 8 mon hs o age. The gene ic in luence o CACNA1C a ian s on sleep in in an s was exam- ined by using PLINK so wa e. Th ee o he examined CACNA1C a ian s, s4765913, s4765914, and s2239063, we e associa ed wi h sleep la ency (pe mu ed P<0.05). The e was no signi ican associa ion be ween s udied a ian s and nigh awakenings o sleep du a ion. CACNA1C a ian s o psychia ic diso de s we e ound o be associa ed wi h long sleep la ency among 8-mon h-old in an s. I emains o be cla i ied whe he he indings e e o de ec i e egula ion o sleep, o o dis ac ibili y o sleep unde ex e nal in luences. PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 1 / 12 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Ci a ion: Kan oja¨ i K, Liuhanen J, Saa enpa¨a¨- Heikkila¨O, Sa omaa A-L, Kyllia¨inen A, Po¨lkki P, e al. (2017) Va ian s in calcium ol age-ga ed channel subuni Alpha1 C-gene (CACNA1C) a e associa ed wi h sleep la ency in in an s. PLoS ONE 12(8): e0180652. h ps://doi.o g/10.1371/jou nal. pone.0180652 Edi o : Thomas H. J. Bu ne, Uni e si y o Queensland, AUSTRALIA Recei ed: Janua y 26, 2017 Accep ed: June 19, 2017 Published: Augus 9, 2017 Copy igh : ©2017 Kan oja¨ i e al. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y S a emen : The e hical app o al limi s he indi idual-le el da a a ailabili y om CHILD-SLEEP coho , and p ohibi s he au ho s om making da a se publicly a ailable. Da a a e a ailable om he co esponding au ho (Tiina Paunio) upon e hical app o al om he E hical Commi ee o Pi kanmaa Hospi al Dis ic (sec e a y Ki si Ko honen a ki si.ko honen@pshp. i). In oduc ion CACNA1C encodes he alpha subuni o he L- ype ol age-dependen calcium channel Ca 1.2, which is highly exp essed in hippocampus, ce eb al co ex, and ce ebellum [1]. Un il now, genome-wide associa ion s udies ha e de ec ed se en a ian s a CACNA1C which a e associa ed wi h psychia ic diso de s. The Psychia ic GWAS Conso ium Bipola Diso de Wo king G oup [2] epo ed genome-wide, signi ican associa ion be ween CACNA1C a ian s4765913 and bipola diso de . They also showed, oge he wi h he Psychia ic Genomic Conso ium (PGC), ha associa ion was s onge be ween s4765913 and a combined sample o schizoph enia and bipola diso de han wi h schizoph enia alone [2]. The Schizoph enia Psychia ic Genome-Wide Associa ion S udy (GWAS) Conso ium [3] conduc ed a mega- analysis o samples consis ing o schizoph enia and bipola diso de which eached a genome- wide signi ican associa ion wi h a ian s4765905. This associa ion was eplica ed in wo s udies [4,5]. The C oss-Diso de G oup o he Psychia ic Genomics Conso ium [6] pe o med GWAS o samples o schizoph enia, bipola diso de , ASD, a en ion de ici -hype ac i i y diso de (ADHD), and majo dep essi e diso de (MDD). Va ian s1024582 a CACNA1C was associ- a ed wi h c oss-diso de a he genome-wide signi icance le el when all i e diso de s we e included. In model selec ion analyses, a ian s4765914 was associa ed wi h sample consis ing o bipola diso de , MDD and schizoph enia [6]. The Schizoph enia Wo king G oup o he Psychia ic Genomics Conso ium [7] combined all a ailable schizoph enia samples and iden- i ied 108 loci ha me genome-wide signi icance in schizoph enia. One o he loci was a CACNA1C, wi h signi ican associa ion o a ian s s2007044 and s2239063. Ano he s udies obse ed genome-wide signi ican associa ion be ween a ian s1006737 and schizoph enia [8] and a combined sample o bipola diso de and schizoph enia [9]. The s udies o human b ain imaging ha e sugges ed ha s1006737 may a ec s uc u es and unc ions o b ain in schizoph enia, such as co ical whi e ma e in eg i y [10–12]. Allelic a ia ion o s1006737 also has an impac on egional g ay ma e olume in heal hy indi iduals [13]. Fu he mo e isk allele AA o schizoph enia inc eases L- ype ol age-ga ed calcium channel cu en den- si y and le els o CACNA1C mRNA in induced human neu ons [14]. S udies in adul s ha e esol ed ha a ian s in CACNA1C a e associa ed wi h sleep diso de and sleep ai s, including na colepsy [15], sleep la ency [16], and sleep quali y [16,17]. Mos o he associa ed a ian s a e loca ed in he hi d in on o CACNA1C, nea he se en a ian s ela ed o psychia ic ai s. None o hese sleep ela ed a ian s me he genome-wide signi i- cance le el. Cacna1c has been ound o modula e he elec ophysiological a chi ec u e o sleep in mice. In a s udy by Kuma and colleagues [18], haploinsu iciency o Cacna1c educed EEG spec al gamma powe du ing wake and REM sleep, which migh indica e lowe ed ale ness du ing wake ulness and educed co ical ac i a ion du ing REM sleep [18]. In addi ion he e o- zygous Cacna1c mice demons a ed lowe REM sleep ebound a e sleep dep i a ion [18]. Posi i e symp oms o schizoph enia a e ela ed o phenomenological and neu obiological ea- u es o REM sleep [19]. The mal unc ioning REM-sleep p ocesses o he e ozygous CACNA1C knockou mice esemble he impai ed sleep egula ion obse ed in schizoph enia [18]. Sleep dis u bances a e common in bipola diso de [20] and schizoph enia [21]. In child en and adolescen s, sleep p oblems a e highly-p e alen in andem wi h psychia ic diso de s, such as ADHD, anxie y, mood diso de s, and ASD [22]. Insu icien sleep has nega i e impac on in an ´s neu obeha io al and cogni i e unc ions and heal h [23]. Sho sleep du a ion, p olonged sleep onse and equen nigh awakenings a e associa ed wi h social-emo ional p oblems in oddle s [24]. The e a e s ong gene ic in luences o consolida ion o sleep in in ancy [25], bu li le is known ega ding how gene ic a ia ion a ec s sleep in ea ly CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 2 / 12 Funding: This s udy was unded by he Academy o Finland (g an s 134880; www.aka. i/skidi-kids, 253346 o TP) and Gyllenbe g ounda ion ( o TP). The polysomnog aphy s udy was inancially suppo ed by Compe i i e S a e Resea ch Financing o he Expe Responsibili y a ea o Tampe e Uni e si y Hospi al (G an s 9P013, 9R007, 9S007). The unde s had no ole in s udy design, da a collec ion and analysis, decision o publish, o p epa a ion o he manusc ip . Compe ing in e es s: The au ho s ha e decla ed ha no compe ing in e es s exis . childhood. Touche e e al. 2013 epo ed ha nigh ime sleep du a ion is s ongly in luenced by gene ic ac o s [25]. Thei s udy comp ised 995 wins and sleep du a ion was measu ed a 6, 18, 30 and 48 mon hs o age. They obse ed s ong gene ic in luences a consolida ed nigh ime sleep du a ion a 6, 30 and 48 mon hs. In con as sha ed en i onmen al in luences explained a la ge p opo ion o a iance in day ime sleep du a ion [25]. Fi s GWAS o sleep du a ion in child en de ec ed genome-wide signi ican associa ion a ch omosome 11 bu his was no eplica ed in independen samples [26]. Child en we e 2–14 yea s old. No GWAS o sleep ai s in in an s has been published ye . In his s udy we a emp ed o de e mine whe he a ian s in CACNA1C a e associa ed wi h al e ed sleep pa ame e s in Finnish in an s. We we e in e es ed in his ques ion because 1) a i- an s in his gene ha e been associa ed wi h psychia ic disease, especially bipola diso de and schizoph enia, 2) bipola diso de and schizoph enia a e associa ed wi h a ious sleep dis u - bances, 3) he Cacna1c channel has been implica ed in sleep-wake egula ion, and 4) he e a e s ong gene ic in luences o sleep consolida ion in in ancy. We explo ed he da a om 1089 Finnish in an s ha unde wen genome wide geno yping and om which in an sleep ques ion- nai e da a was a ailable. A small subse o hese babies had also unde gone polysomnog aphy. To assess insu icien and dis u bed sleep we examined sleep du a ion, nigh ly awakenings, sleep la ency, WASO, SEI and REM/NREM a io in hese in an s. We chose o e alua e o se en gene a ian s ha ha e been shown o be associa ed genome-wide signi ican ly wi h bipo- la diso de and schizoph enia. The p ima y inding was ha h ee o he SNPs we e associa ed wi h p olonged sleep la ency in a s a is ically signi ican manne . None we e associa ed wi h changes in o al sleep ime, o nigh ime awakenings. Ma e ials and me hods The s udy sample, CHILD-SLEEP [27], is a Finnish bi h coho collec ed om Pi kanmaa Hospi al Dis ic , comp ising 1643 in an s bo n Ap il 2011−Feb ua y 2013, and hei pa en s. The ocus o CHILD-SLEEP is on he ole o ea ly sleep and ci cadian hy hm in gene al popu- la ion. In his s udy, he pa en al ques ionnai e da a ela ed o he sleep o child en and suc- cess ully geno yped DNA samples we e a ailable om 1086 babies (520 gi ls and 566 boys) who we e 8 mon hs old. The e hical app o al o CHILD-SLEEP was ob ained om he E hical Commi ee o Pi kanmaa Hospi al Dis ic (R11032/9.3.2011). The w i en in o med consen s we e ob ained om he pa en s. Families we e in o med o hei igh s o e mina e hei pa - icipa ion in he s udy a any ime du ing da a collec ion. The e alua ion o sleep o he in an s in his s udy is based on pa en al ques ionnai es ISQ ( he in an sleep ques ionnai e) and BISQ (a b ie sc eening ques ionnai e o in an sleep p oblems) [28,29]. ISQ and BISQ a e bo h eliable and alid measu es o in an sleep [30].The selec ed pheno ypes o his s udy we e o al sleep ime (TST), he numbe o nigh awakenings and sleep la ency a he age o eigh mon hs. To assess hese sleep pa e ns o he in an s hei pa en s we e asked he ollowing ques ions: “How many hou s does you child sleep a nigh (19 pm - 07am)?” (Sleep du a ion a nigh , BISQ10) and “How many hou s does you child sleep a day ime (07am– 19pm)?” (Sleep du a ion a day ime, BISQ11). BISQ10 and BISQ11 we e agg ega ed o o al sleep ime. The in o ma ion o nigh awakenings (ISQ5) was col- lec ed wi h a ques ion “How many imes a nigh (be ween 24.00 and 06.00), you baby will usually wake up and need eassu ance?” (0 = Does no wake up a all, 1 = Once a nigh , 2 = Two imes a nigh , 3 = Th ee imes a nigh , 4 = Fou imes a nigh , 5 = Fi e o mo e imes a nigh ). The ques ion o Sleep la ency (ISQ1) was “How long does i usually ake o se le you baby o sleep?” (1 = Less han 10 minu es, 2 = 10–20 minu es, 3 = 20–30 minu es, 4 = 30–40 minu es, 5 = 40–50 minu es, 6 = 50–60 minu es, 7 = one hou o mo e). To al sleep CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 3 / 12 ime and nigh awakenings we e no mally dis ibu ed. Because sleep la ency o babies was skewed, we dicho omized he a iable based on he mean o he sleep la ency in his s udy (Table 1) and ea lie sleep s udies in child en [24,31] as less han 20 minu es and mo e han 20 minu es. A subcoho o he child en Child Sleep babies unde wen an ambula o y o e -nigh PSG in h ee di e en ages (1, 8, and 24 mon hs o age) as desc ibed in de ail ea lie [32]. A he age o eigh mon hs, 72 in an s we e eco ded. The eco dings we e s a ed a he amilies’ homes close o he babies’ usual bed ime. Howe e , he objec i e sleep la ency could no be measu ed in he PSGs, as he e ening ou ines a ied subs an ially. O en babies ell asleep while ea ing and he eal “ligh s o ” ime could no be de ined. The leng h o he eco dings a ied depend- ing on he habi ual leng h o o e -nigh sleep o he babies. Fo he pu pose o his s udy, nine hou s o eco ding we e analyzed, beginning a he sleep onse . The sleep s ages we e sco ed acco ding o he es ablished ules [33]. The sleep pa ame- e s chosen o his s udy we e WASO (wake a e sleep onse ), SEI (sleep e iciency) and R/ NREM ( a io be ween REM and NREM sleep) (Table 1). WASO was he ime o wake ulness (in minu es) du ing he nine hou s o eco ding. SEI was he o al sleep ime di ided by he nine hou s o eco ding. R/NREM a io was he amoun o s age R sleep di ided by he amoun o NREM sleep du ing he nine hou s [32]. Umbilical co d blood sample was d awn om each newbo n. DNA was ex ac ed acco ding o s anda d p ocedu es. DNA samples we e geno yped wi h Illumina In inium PsychA ay BeadChip a Es onian Genome Cen e and quali y con ol (QC) was pe o med wi h PLINK (h p://pngu.mgh.ha a d.edu/~pu cell/plink/). Ma ke s we e emo ed o missingness (>5%), Ha dy-Weinbe g equilib ium (p- alue <1 x 10 −6 ), and low mino allele equency (<0.01). Indi iduals we e checked o missing geno ypes (>5%), ela edness (iden ical by descen calcula ion, PI_HAT>0.2) and popula ion s a i ica ion (mul idimensional scaling). A e QC, pheno ype and high quali y geno ype da a a 8 mon hs was a ailable o 1086 (pa en al ques ionnai es) and 63 (PSG) babies. Geno yped da a was impu ed wi h IMPUTE 2 [34] agains Finnish WGS (whole-genome sequencing) [35] and 1000 genomes [36] (Phase 3, eleased a Feb ua y, 2013) e e ence panels. Se en SNPs (single nucleo ide polymo phism) ea lie associa ed o psychia ic ai s a CACNA1C wi h genome wide signi icance we e selec ed o analyses (Table 2). The associa ion be ween se en SNPs and quan i a i e sleep measu es we e es ed wi h linea (TST and nigh ly awakenings) and logis ic (sleep la ency) eg ession analyses implemen ed wi h PLINK. Table 1. Desc ip ion on s udy a iables. Va iable N Range Mean P a TST 1086 9–17 h 13.3 h 0.67 Nigh awakenings 1086 0 - 5 imes 2 imes a nigh 0.64 Sleep la ency 1086 <10 min - 1 h 10–20 min 0.17 WASO 63 24–120 min 67.2 min 0.09 SEI 63 76–96% 86.7% 0.62 R_NREM 63 0.30–0.73 0.49 0.16 Family a mosphe e b 1086 7–43 13.5 0.81 B eas eeding 1086 - 44.8% (b eas ed) 0.96 Illness 1086 - 12.4% (had illness) 0.82 a p- alue o - es o he di e ence be ween boys and gi ls b Small sco e e lec s be e a mosphe e h ps://doi.o g/10.1371/jou nal.pone.0180652. 001 CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 4 / 12 Co a ia es used in he analyses we e gende , i s h ee p incipals componen s o GWAS, am- ily a mosphe e [27], b eas eeding (dicho omized a iable: b eas eeding/b eas eeding + o - mula o o mula), and babies’ illnesses (milk alle gy, o he alle gy, colic, in ec ions, e lux, congeni al hea disease, inna e de elopmen al diso de , neu ological diso de , g ow h e a - da ion, o he illness). Babies´ illnesses we e used as a dich omized a iable: no illness / 1 o mo e illness. Desc ip ions o he s udy a iables a e p esen ed in Table 1. Associa ions we e co ec ed o mul iple es ing by he max (T) pe mu a ion in PLINK wi h 10 000 pe mu a ions pe SNP. The c i e ion o signi icance was se a pe mu a ion p <0.05. Because o mul iple co ela ed pheno ypes Bon e oni co ec ion was no used in his s udy. I is likely o be o e ly conse a i e as he analyses a e no independen o each o he . Powe calcula ions we e pe o med wi h he Gene ic Powe Calcula o (h p://pngu.mgh. ha a d.edu/~pu cell/gpc/) [37] assuming an addi i e model (TST and nigh awakenings) o case-con ol o h eshold-selec ed quan i a i e ai (sleep la ency), 1% quan i a i e ai locus (QTL) a iance o addi i e gene ic a iance and pe ec linkage disequilib ium be ween QTL and he ma ke s. The e was 85% s a is ical powe o de ec a ian s wi h signi icance le el o 0.05 when polymo phisms wi h allele equencies om 0.2 o 0.3 we e analyzed. Resul s All se en a ian s included in he analyses o his s udy lie wi hin 167 kb egion o he CACNA1C in on h ee (Fig 1). LD-s uc u e o he s udied a ian s is p esen ed in Fig 2. The e a e wo LD- blocks in he egion. The i s egion comp ises he a ian s s2007044, s1006737, s4765905, and s1024582 (D‘0.95). Ano he LD-block consis s o SNPs s4765913, s4765914, and s2239063 (D‘0.81). The esul s o he associa ion analyses a e shown in Table 3. Be a alues indica e he di ec- ion o he e ec o he mino allele. Empi ical p- alue 1 (EMP1) is a poin -wise p- alue om 10 000 pe mu a ions and empi ical p- alue 2 (EMP2) means co ec ed empi ical p- alue o e all s udied SNPs. In he p ima y analyses we used only sex as a co a ia e. Va ian s s4765913 (mino allele A), s4765914 (mino allele T), and s2239063 (majo allele A) we e associa ed Table 2. SNPs associa ed wi h psychia ic ai s in genome-wide signi ican le el in CACNA1C. SNP Pheno ype Re e ence s2007044 Schizoph enia [7] s1006737 bipola diso de , schizoph enia [8,9] s4765905 bipola diso de , Schizoph enia [3–5] s1024582 bipola diso de , schizoph enia [6] s4765913 bipola diso de , schizoph enia [2] s4765914 bipola diso de , MDD, schizoph enia [6] s2239063 Schizoph enia [7] h ps://doi.o g/10.1371/jou nal.pone.0180652. 002 Fig 1. Loca ion o s udied SNPs a CACNA1C.Se en SNPs associa ed wi h psychia ic diso de s a e loca ed in in on h ee o CACNA1C gene. h ps://doi.o g/10.1371/jou nal.pone.0180652.g001 CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 5 / 12 wi h longe sleep la ency (pe mu ed P <.05, EMP1) in addi i e model. When pe mu ed 10 000 imes o e all 7 a ian s (EMP2) he a ian s s4765914 and s2239063 emained signi i- can . In dominan model he s onges associa ion was de ec ed be ween a ian s2239063 and sleep la ency (be a = −0.4023, P = 0.0037, EMP1 = 0.0036, EMP2 = 0.017). In ecessi e model a ian s4765913 showed he s onges associa ion o sleep la ency (be a = 0.8076, P = 0.003487, EMP1 = 0.0025, EMP2 = 0.0167). The e was no signi ican associa ion be ween s udied a ian s and nigh awakenings o sleep du a ion. When he co a ia es amily a mo- sphe e, b eas eeding, babies’ illnesses and gene ic p incipal componen s we e added o he analyses, he esul s o sleep la ency emained signi ican . When gi ls and boys we e analyzed sepa a ely, associa ion o a ian s4765914 wi h sleep la ency was s a is ically signi ican in boys (be a = 0.4331, P = 0.008, EMP1 = 0.008, EMP2 = 0.037) bu no in gi ls in addi i e model. In dominan model s4765914 (mino allele T) sho ened TST in boys (be a = -0.248, P = 0.0275, EMP1 = 0.0309, EMP2 = 0.1162). The e was no signi ican asso- cia ion be ween sex and CACNA1C a ian s in nigh awakenings. We also pe o med haplo ype analyses o sleep la ency, TST and nigh awakenings. They we e in line wi h p ima y analyses and se e al haplo ypes we e associa ed wi h sleep la ency (P <0.05). In logis ic eg ession analyses wo haplo ypes we e signi ican a e 10 000 pe mu- a ion o e all se en SNPs (EMP2 <0.05). O hem equencies o haplo ype (CC) comp ising Fig 2. LD s uc u e o he SNPs analyzed a CACNA1C.Ma ke s wi h linkage disequilib ium (0< 2 1) a e shown in ed h ough pale ligh pink (colo in ensi y dec eases wi h dec easing 2 alue). h ps://doi.o g/10.1371/jou nal.pone.0180652.g002 CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 6 / 12 a ian s s4765914 and s2239063 and TCC o a ian s s4765913, s4765914 and s2239063 di e ed signi ican ly be ween in an s wi h sho e sus long la ency (P = 0.001581, P = 00169, espec i ely). Haplo ypes comp ising p o ec i e alleles o psychia ic diso de s we e mo e common in sho la ency (haplo ype CC: eq sho /long la ency = 0.3156/0.243, haplo ype TCC: eq sho /long la ency = 0.3172/0.245). The e was no signi ican associa ion be ween any o he haplo ypes and TST o nigh awakenings. We analyzed he associa ion o polysomnog aphy (PSG) a iables in a sample o 63 babies a 8 mon hs o age. The e was no signi ican co ela ion be ween subjec i e measu e o nigh ly awakenings (ISQ5) and objec i e measu es WASO o SEI. The esul s o PSG analyses a e shown in Table 4. The e was no signi ican associa ion be ween 7 a ian s and he PSG a iables. Discussion In his s udy, some o he a ian s ela ed o psychia ic ai s showed nominally signi ican associa ion wi h sleep. In he analyses o he whole sample a ian s s4765913, s4765914, and Table 3. The esul s o associa ion analyses be ween CACNA1C a ian s and TST, nigh awakenings and sleep la ency, N = 1086. SNP TST Nigh awakenings Sleep la ency be a P EMP1 a EMP2 b be a P EMP1 a EMP2 b be a P EMP1 a EMP2 b s2007044*-0.02258 0.6871 0.6825 0.9935 -0.04966 0.4547 0.4558 0.8971 0.06581 0.5299 0.5212 0.9486 s1006737 -0.02226 0.6996 0.7002 0.9945 -0.07104 0.3011 0.2945 0.733 0.05984 0.5821 0.5683 0.9696 s4765905 -0.02166 0.7092 0.7135 0.9952 -0.05912 0.3922 0.3802 0.8402 0.04947 0.6511 0.6379 0.9863 s1024582 -0.01756 0.7545 0.7571 0.9983 -0.03266 0.6229 0.6201 0.98 0.02755 0.7936 0.787 0.9993 s4765913 -0.04492 0.4643 0.4686 0.9069 0.04481 0.5405 0.5384 0.951 0.2561 0.02332 0.0237 0.08759 s4765914 -0.05098 0.4365 0.4339 0.8845 0.1299 0.09763 0.09774 0.324 0.3026 0.01115 0.0106 0.0423 s2239063*-0.01161 0.8368 0.8384 0.9998 0.02467 0.7116 0.7183 0.9943 -0.2767 0.01138 0.011 0.0426 Addi i e model, adjus ed wi h sex *impu ed a Empi ical p- alue 1 (EMP1) = poin -wise p- alue om 10,000 pe mu a ions b Empi ical p- alue 2 (EMP2) = co ec ed empi ical p- alue by max (T) pe mu a ions, TST = o al sleep ime. h ps://doi.o g/10.1371/jou nal.pone.0180652. 003 Table 4. Polysomnog aphy esul s, N = 63. SNP WASO SEI REM/NREM be a P EMP1 a EMP2 b be a P EMP1 a EMP2 b be a P EMP1 a EMP2 b s2007044*2.304 0.6337 0.6343 0.9726 -0.4188 0.6665 0.6698 0.9829 -0.0001271 0.995 0.9956 1 s1006737 -3.049 0.5525 0.5487 0.9365 0.5307 0.6072 0.6158 0.9638 -0.004783 0.8249 0.8305 0.9988 s4765905 -2.874 0.5799 0.5765 0.9505 0.5024 0.6307 0.6395 0.9724 -0.00398 0.8549 0.8549 0.9995 s1024582 -3.054 0.5241 0.5321 0.9193 0.3914 0.6849 0.6919 0.9867 -0.009561 0.6486 0.6526 0.9754 s4765913 -1.09 0.8247 0.8245 0.999 0.1462 0.8825 0.8845 0.9997 0.005384 0.8056 0.8096 0.9984 s4765914 -0.2365 0.9631 0.9644 1 -0.08532 0.9337 0.9304 1 0.008536 0.7147 0.7171 0.9905 s2239063*-8.536 0.1379 0.1395 0.3816 1.785 0.1223 0.1252 0.3471 -0.03394 0.1514 0.15 0.4028 WASO = wake a e sleep onse , SEI = sleep e iciency, REM = apid eye mo emen sleep, NREM = non- apid eye mo emen sleep, addi i e model, adjus ed wi h sex, h ee i s p incipal componen s o GWAS, home a mosphe e, b eas eeding and babies’ illnesses *impu ed a Empi ical p- alue 1 (EMP1) = poin -wise p- alue om 10,000 pe mu a ions b Empi ical p- alue 2 (EMP2) = co ec ed empi ical p- alue by max (T) pe mu a ions. h ps://doi.o g/10.1371/jou nal.pone.0180652. 004 CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 7 / 12 s2239063 we e associa ed wi h sleep la ency. The mino allele A o s4765913 was associa ed wi h longe sleep la ency. Psychia ic GWAS Conso ium Bipola Diso de Wo king G oup [2] epo ed ha allele A is associa ed wi h bipola diso de (P = 1.52x10 -8 ) and wi h combined samples o bipola diso de and schizoph enia (P = 7.7x10 -8 ). In ou s udy, he mino allele T o s4765914 was associa ed wi h longe ime o sleep la ency in in an s. C oss-Diso de o he Psychia ic Genomics Conso ium [6] epo ed ha his same allele is associa ed wi h bipola diso de , majo dep essi e diso de , and schizoph enia. T allele is also associa ed wi h amyg- dala s uc u e and unc ion in adolescen s; in he s udy o Sumne and colleagues [38] homo- zygous ca ie s o T allele exhibi ed smalle amygdala olume compa ed o indi iduals wi h homozygous majo C allele [38]. Reduced amygdala olume has been obse ed in bipola dis- o de in adul s and you h in ea lie s udies [39,40]. The e is a complex in e play be ween sleep and emo ions; sleep dep i a ion impai s he connec i i y be ween amygdala and p e on al co ex, which ha e a di ec impac on indi idual´s abili y o egula e emo ions [41]. Finally, in ou s udy he majo allele A o s2239063 was associa ed wi h longe sleep la ency in 8-mon h-old babies. This same allele is associa ed wi h schizoph enia a he genome-wide sig- ni icance le el (P = 1.93e -8 ) [7]. Long sleep la ency is common in schizoph enia [21] and bipola diso de [20]. Conside ing haplo ype analyses we obse ed ha equencies o haplo ypes which comp ised he p o ec i e alleles o psychia ic ai s, CC ( s4765914 and s2239063) and TCC ( s4765913, s4765914 and s2239063) we e signi ican ly mo e equen wi hin babies wi h sho sleep la ency ime compa ed o in an s wi h long la ency. When we analyzed boys and gi ls sepa a ely, he psychia ic isk allele T o s4765914 was signi ican ly associa ed wi h p olonged sleep la ency and sho ened TST in boys bu no in gi ls. In case his inding is no conside ed as lack o powe due o small sample size o s udied g oups bu ue biological gende di e ence, i sugges s ha male gende may be gene ically ulne able o sleep dis u bances. In ea lie s udies he e we e sex-speci ic di e ences in in lu- ences o CACNA1C a ian s on mood diso de s [42] and unc ional eco e y om episodes o schizoph enia [43]. This could be explained by ho monal di e ences. Es ogen di ec ly po en- ia es neu onal L- ype Ca 2+ channels [44] and inhibi s Ca 2+ in lux h ough L- ype ol age- ga ed CA 2+ channels [45]. Va ian s in CACNA1C a e associa ed wi h sleep la ency and quali y o sleep in adul s [16,17]. In ou s udy we obse ed ha some o he CACNA1C a ian s o psychia ic diso de s we e ound o be associa ed wi h sleep la ency in babies a 8 mon hs-o -age. These a ian s we e no signi ican ly associa ed wi h sleep du a ion o nigh ly awakenings. Ou esul s o sleep du a ion in in an s a e in line wi h ea lie s udies whe e no genome-wide signi ican associa ion o sleep du a ion a CACNA1C has been epo ed [16,26,46–50]. GWAS o nigh ly awakenings has no been published in adul s. Mo e s udies a e needed o see i CAC- NA1C egula es nigh ly awakenings. The limi a ions o his s udy we e he c oss-sec ional app oach o he esea ch and he ac ha we lacked eplica ion da a. Fu he , he powe o de ec isk a ian s o sleep dis u bances wi h polysomnog aphy was limi ed because he amoun o objec i e da a was qui e small. The objec i e sleep la ency o he in an s could no be measu ed because he eco ding was no s a ed a bed ime. This is why we chose pa en al ques ionnai es as a measu e o he sleep la ency. The e a e also se e al o he candida e genes o sleep dis u bances like ecen ly de ec ed RBFOX3 in me a-analyses o sleep la ency wi h la ge da ase [51]. In ou s udy we ocused CACNA1C because o i s cen al ole in sleep and psychia ic p oblems. Powe o GWAS was limi ed because he sample size was ela i ely small bu me a-analyses wi h o he childhood coho s will be pe o med la e . The esul s o o he candida e genes o sleep dis u bances in CHILD-SLEEP p ojec will be epo ed elsewhe e. CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 8 / 12 None heless, o e all CACNA1C emains a po en ial candida e gene o sleep dis u bances and he a ian s may a ec sleep al eady in ea ly childhood. Collec ion o CHILD-SLEEP da a is con inuing and we will ollow he sleep and men al heal h o hese child en. Unde s anding how he gene ic a ian s in luence in sleep/wake egula ion al eady in in ancy may ha e an impac on helping he amilies o in e e e in in an s´ sleep p oblems a ea ly s age. In a long e m he g owing knowledge o gene ics could also be used as basis o de elopmen o he a- pies and ea ly diagnos ics o sleep and neu opsychological diso de s. Acknowledgmen s This s udy was unded by he Academy o Finland (g an s 134880; h p://www.aka. i/skidi- kids, 253346 o TP) and Gyllenbe g ounda ion ( o TP). The polysomnog aphy s udy was inancially suppo ed by Compe i i e S a e Resea ch Financing o he Expe Responsibili y a ea o Tampe e Uni e si y Hospi al (G an s 9P013, 9R007, 9S007). We acknowledge An i- Pekka Sa in and Samuli Ripa i o e e ence panels and ad ices in impu a ion. Ma ja-Rii a Rau iainen and An i-Jussi A ¨mma¨la¨a e acknowledged o help in geno ype da a quali y con- ol. The au ho s wish o acknowledge CSC–IT Cen e o Science, Finland, o compu a ional esou ces. Au ho Con ibu ions Concep ualiza ion: KK TP. Funding acquisi ion: TP. In es iga ion: KK JL OSH ALS AK PP JJ AT SLH TPH JP TP. Me hodology: KK AK SLH TP. P ojec adminis a ion: TP. Resou ces: AT LM. Supe ision: TP. Visualiza ion: KK. W i ing – o iginal d a : KK. W i ing – e iew & edi ing: KK OSH ASL AK PP ASH TPH JP TP. Re e ences 1. Be ge SM, Ba sch D. The ole o L- ype ol age-ga ed calcium channels Ca 1. 2 and Ca 1. 3 in no - mal and pa hological b ain unc ion. Cell Tissue Res. 2014; 357: 463–476. h ps://doi.o g/10.1007/ s00441-014-1936-3 PMID: 24996399 2. Psychia ic GWAS Conso ium Bipola Diso de Wo king G oup. La ge-scale genome-wide associa ion analysis o bipola diso de iden i ies a new suscep ibili y locus nea ODZ4. Na Gene . 2011; 43: 977– 983. h ps://doi.o g/10.1038/ng.943 PMID: 21926972 3. Schizoph enia Psychia ic Genome-Wide Associa ion S udy (GWAS) Conso ium. Genome-wide asso- cia ion s udy iden i ies i e new schizoph enia loci. Na Gene . 2011; 43: 969–976. h ps://doi.o g/10. 1038/ng.940 PMID: 21926974 4. Hamshe e ML, Wal e s JTR, Smi h R, Richa ds A, G een E, G oze a D, e al. Genome-wide signi ican associa ions in schizoph enia o ITIH3/4, CACNA1C and SDCCAG8, and ex ensi e eplica ion o asso- cia ions epo ed by he Schizoph enia PGC. Mol Psychia y. 2013; 18: 708–712. h ps://doi.o g/10. 1038/mp.2012.67 PMID: 22614287 5. Takahashi S, Gla SJ, Uchiyama M, Fa aone SV, Tsuang MT. Me a-analysis o da a om he Psychia - ic Genomics Conso ium and addi ional samples suppo s associa ion o CACNA1C wi h isk o CACNA1C a ian s a ec sleep la ency in in an s PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0180652 Augus 9, 2017 9 / 12