Seasonal gene expression kinetics between diapause phases in Drosophila virilis group species and overwintering differences between diapausing and non-diapausing females

1 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 www.na u e.com/scien i ic epo s Seasonal gene exp ession kine ics be ween diapause phases in D osophila i ilis g oup species and o e win e ing di e ences be ween diapausing and non-diapausing emales Tiina S. Salminen1,2, Lau a Vesala1,2, As a Laiho3, Mikko Me isalo1, Anneli Hoikkala1 & Maa ia Kanka e1 Mos no he n insec species expe ience a pe iod o de elopmen al a es , diapause, which enables hem o su i e o e he win e and pos pone ep oduc ion un il a o able condi ions. We s udied he iming o ep oduc i e diapause and i s long- e m e ec s on he cold ole ance o D osophila mon ana, D. li o alis and D. ezoana emales in seasonally a ying en i onmen al condi ions. A he same ime we aced exp ession le els o 219 genes in D. mon ana using a cus om-made mic oa ay. We show ha he seasonal swi ch o ep oduc i e diapause occu s o e a sho ime pe iod, and ha o e win e ing in ep oduc i e diapause has long-las ing e ec s on cold ole ance. Some genes, such as Hsc70, Jon25Bi and pe iod, we e up egula ed h oughou he diapause, while o he s, including egucalcin, couch po a o and Tho , we e up egula ed only a i s speci ic phases. Some o he exp ession pa e ns induced du ing he sensi i e s age, when he emales ei he en e diapause o no , emained induced ega dless o he la e condi ions. qPCR analyses con i med he indings o he mic oa ay analysis in D. mon ana and e ealed simila gene exp ession changes in D. li o alis and D. ezoana. The p esen s udy helps o achie e a be e unde s anding o he gene ic egula ion o diapause and o he plas ici y o seasonal esponses in gene al. Seasonally changing en i onmen al condi ions p esen a ious kinds o challenges o species li ing a high la i udes, se ing equi emen s o he plas ici y o li e-his o y ai s impo an o ep oduc ion and su i al. Acco dingly, many no he n species om insec s o e eb a es a e able o o ecas he o h- coming ad e se season on he basis o annual changes in en i onmen al cues and o use his in o ma ion o op imally ime hei de elopmen and ep oduc ion. While biochemical and physiological esponses can usually be e e sed o e a sho ime scale, he ai s showing de elopmen al plas ici y, like ep oduc- i e diapause, end o equi e a longe ime o be e mina ed1. In pho ope iodically induced adul ep o- duc i e diapause, emales’ o a ian de elopmen is hal ed o a p e- i ellogenic s age du ing la e summe , con inuing only when he days become longe and wa me in sp ing. This kind o win e diapause is 1Uni e si y o Jy askyla, Depa men o Biological and En i onmen al Science, P.O. Box 35, FI-40014, Uni e si y o Jy äskylä, Finland. 2BioMediTech, Uni e si y o Tampe e, Bioka u 6, F1-33014 Finland. 3Finnish DNA Mic oa ay Cen e, Bioin o ma ics eam, Tu ku Cen e o Bio echnology, Tykis öka u 6, FI-20521 Tu ku, Finland. Co espondence and eques s o ma e ials should be add essed o T.S.S. (email: [email p o ec ed]) o o M.K. ([email p o ec ed]) ecei ed: 27 Janua y 2015 Accep ed: 07 May 2015 Published: 11 June 2015 OPEN www.na u e.com/scien i ic epo s/ 2 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 usually accompanied by physiological, mo phological and beha iou al changes ha allow insec s o cope wi h a ious kinds o s esses du ing he cold pe iod2. T acing he exp ession kine ics, i.e. changes in he exp ession pa e ns o genes associa ed wi h diapause, does no only o e a unique oppo uni y o gain in o ma ion on he gene ic egula ion o he diapause i sel , bu also o e s p elimina y s eps o iden i y some o he candida e genes unde lying he adap a ion o no he n en i onmen s. Daily and seasonal changes in li e-his o y ai s impo an in adap a ion o changing en i onmen al condi ions ha e usually been s udied a mo e o less cons an ligh :da k and/o empe a u e cycles in he labo a o y3,4. While s udies like hese a e ex emely help ul o de ec ing esponses o speci ic en i on- men al cues and acing linkages be ween pheno ypic changes and gene exp ession, hey may no e eal he comple e oles o daily and seasonal hy hms in wild popula ions. In s udies ha in ol e simul ane- ous changes in daily empe a u e and ligh ing g adien s, he en ainmen s caused by hese ac o s ha e been ound o ein o ce each o he and na ow down e.g. he iming o ac i i y peaks o insec s5,6. Fo example, Vanin e al.7 s udied he ac i i y hy hms and accompanying changes in he exp ession le els o speci ic clock genes in D. melanogas e lies in he wild. In e es ingly, hei s udy shows ha se e al key labo a o y-based assump ions abou ci cadian beha iou in his species, including he an icipa ion o ligh s on and o ansi ions, he midday sies a, and he dominance o ligh s imuli o e empe a u e, a e no suppo ed by obse a ions in na u ally luc ua ing condi ions7. Seasonal hy hms ha e been s udied in na u al su oundings less o en han daily hy hms e en hough hese kinds o s udies would p o ide aluable in o ma ion abou adap a ion o seasonali y. Vesala e al.8 s udied he cold ole ance o D. mon ana lies by mimicking he seasonal changes in day leng h and empe a u e o he lies’ collec ion si e, and showed an inc ease in he cold ole ance o he lies owa ds he win e o be accompanied by changes a speci ic me aboli e le els. This kind o s udy scheme can also be applied o s udy he en i onmen al ac o s e oking acul a i e ep oduc i e diapause, as well as o ace pheno ypic and ansc ip ional changes du ing di e en phases o diapause, which should be unde s ood as a p ocess a he han a s a us9. In a acul a i e ep oduc i e diapause, he diapause induc ion usually akes place a a p e-diapause s age as a esponse o sho ening day leng h and dec eas- ing empe a u e. Du ing he ini ia ion phase, he in ensi y o ep oduc i e diapause quickly eaches i s maximum and is accompanied by a apid dec ease in he insec s’ me abolic a e, while he main enance phase is cha ac e ized by main aining he de elopmen al a es and a ela i ely low me abolic a e. The e mina ion phase is cha ac e ized by he s a o he sexual ma u a ion9. The s udies by Rineha e al.10 and Ragland e al.11 on he lesh ly, Sa cophaga c assipalpis, ha e elegan ly shown how speci ic genes a e up- o down- egula ed in diapausing emales compa ed o non-diapausing ones du ing di e en phases o diapause. Howe e , hese s udies we e pe o med in cons an condi ions in he labo a o y, and hey lack he in o ma ion abou gene exp ession kine ics when he lies shi om one diapause phase o ano he in changing en i onmen al condi ions. In he p esen s udy we ha e aced he e ec s o seasonally changing condi ions on he de elopmen ( ep oduc i e diapause s. sexual ma u a ion) and cold ole ance o he emales o h ee sympa ic no h- e n D. i ilis g oup species (D. mon ana, D. li o alis and D. ezoana), and simul aneously s udied hei gene exp ession kine ics. Females o all hese species en e diapause when he day leng h dec eases below a c i ical poin , and can emain in his s a e o up o nine mon hs12–14. Howe e , hese species di e in hei seasonal li e cycles15 and geog aphical dis ibu ion16. In his s udy, we aimed o ind answe s o he ollowing ques ions: (i) when does he onse o diapause occu in an en i onmen showing g adual changes bo h in day leng h and empe a u e, (ii) does he o e win e ing s a e o he emales a ec hei i ness in e ms o cold ole ance a e he diapause has e mina ed, and (iii) how do he gene exp ession pa e ns o he s udied se o 219 genes a y du ing he ini ia ion, main enance and e mina ion phases o diapause in di e en seasonal en i onmen s. Resul s and discussion Rea ing he emales o h ee no he n D osophila i ilis g oup species in a clima e chambe mimicking seasonal changes a hei home si e (Fig. 1) e ealed se e al in e es ing phenomena linked wi h dia- pause and seasonal adap a ion. Fi s , he seasonal swi ch o ep oduc i e diapause occu ed in all h ee species a app oxima ely he same day leng h as has been ound o occu in D. mon ana in cons an ligh :da k cycle and empe a u e condi ions in he labo a o y14,17 (Fig.2). Second, in D. mon ana and D. li o alis he o e win e ing s a e o he emales appea ed o a ec hei cold ole ance e en a e he diapause had e mina ed (Fig. 3). Thi d, changes in he exp ession le el o he s udied genes be ween he ini ia ion, main enance and e mina ion phases o diapause, as well as be ween he emales o e win- e ing a diapause o non-diapause s a e, showed ha he up egula ion o mos o he s udied genes was phase speci ic, while a ew genes ha we e up egula ed du ing he ea ly adul hood emained up egula ed h oughou he emales’ li espan (Table1, Supplemen a y Table 1, Fig.4). Seasonal swi ch o ep oduc i e diapause occu s wi hin a na ow ime pe iod. Seasonal im- ing o ep oduc i e diapause was de e mined by checking he o a ian de elopmen s age o h ee weeks old D. mon ana, D. li o alis and D. ezoana emales, which had been ans e ed in o he clima e chambe be ween July and Augus (see Fig.1). The c i ical day leng h o diapause induc ion (CDL; 50% o emales en e diapause) was ound o be LD 19:5 (co esponding o July 30 h) in all s udy species; a e his da e he p opo ion o diapausing emales inc eased o nea ly 100% wi hin one week (Fig.2). In p e ious www.na u e.com/scien i ic epo s/ 3 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 s udies he CDLs o D. mon ana s ains om Oulanka popula ion ha e been ound o a y be ween LD 18.5:5.5 and LD 19:5 a 16 °C14,17. These indings suppo each o he and con i m he a gumen by Lumme12 and Wa abe13 ha he ep oduc i e diapause o D. i ilis g oup lies is egula ed mainly by pho ope iodic cues. Females’ ep oduc i e s a e du ing o e win e ing a ec s hei cold ole ance a e diapause has been e mina ed. The chill coma eco e y imes (CCRTs) o 250-day-old D. mon ana, D. li o a- lis and D. ezoana emales we e measu ed 30 days a e he win e pe iod had ended (Fig.1), when also he emales ha had o e win e ed in diapause had ma u e o a ies. The eco e y imes o he emales a ied signi ican ly bo h among he species (ANOVA: F6,237 = 2.183, P < 0.05) and be ween he emales’ o e - win e ing s a es (ANOVA: F1,237 = 5.477, P < 0.05). D. mon ana and D. ezoana emales eco e ed equally as (in a e age 6–7 min.; Tukey HSD, P = 0.077; Fig.3), while D. li o alis emales eco e ed mo e slowly (in 13–16 minu es) han D. mon ana (Tukey HSD, P < 0.001) o D. ezoana (Tukey HSD, P < 0.001). All h ee species a e ound in no he n Scandina ia, whe e he win e s a e cold, bu D. li o alis is dis ibu ed also in cen al and sou he n Eu ope16,18. The la ge dis ibu ion a ea o D. li o alis migh ha e caused ade-o s in he cold and hea ole ance o he lies, leading o a lowe le el o cold ole ance19. Howe e , i should be kep in mind ha low empe a u e causes se e al ypes o inju ies in insec s, and acco dingly, di e en ypes o p o ec i e mechanisms a e needed20. CCRT es used he e desc ibes mainly ole ance o di ec inju ies ha a e caused by se e e sho - e m cold exposu e20, while he ole ance o indi ec chilling inju ies ha is caused by milde cold exposu e o e longe pe iod may be equally o e en mo e i al21. O e all, he CCRTs o he lies o all s udy species we e qui e sho , indica ing high cold ole ance compa ed o hose o mos o he D osophila species22. Kelle man e al.21 showed (using C i ical The mal Minimum me hod) ha in genus D osophila he high cold ole ance is mos ly es ic ed o wo species g oups; D. i ilis (including he species used in his s udy) and D. obscu a. Vesala and Hoikkala23 ha e shown ha diapausing D. mon ana lies eco e as e om chill coma han non-diapausing lies i he lies a e no acclima ed o cold, and Vesala e al.24 ha e demons a ed ha he CCRT o diapausing emales a e sho es be ween la e au umn and ea ly sp ing. Ou s udy shows ha he emales’ o e win e ing s a e has e en longe -las ing e ec s on hei cold ole ance. We also ound signi - ican in e ac ion be ween he species and he o e win e ing s a e o he emales (ANOVA: F2,237 = 4.688, Figu e 1. Ligh :da k cycles and empe a u e condi ions in he en i onmen al chambe and sample collec ion poin s. The g ey a ea (scale on le Y-axis) shows seasonal changes in he day leng h and he lines (scale on he igh Y-axis) espec i e changes in empe a u e du ing he day (dashed line), dawn/ dusk in sp ing ( hicke dashed line) and he nigh (solid line). The ligh :da k cycles and he da es indica e when he emales we e ans e ed in o he chambe and when he di e en age ND (non-diapause) and D (diapause) emales we e collec ed. In addi ion o he mic oa ay s ain (175OJ8, D. mon ana), wo addi ional D. mon ana s ains and h ee D. li o alis and D. ezoana s ains we e collec ed o de ec ing he onse o ep oduc i e diapause, and o qPCR (samples 150D and 150 ND) and CCRT expe imen s (samples 250D and 250ND). www.na u e.com/scien i ic epo s/ 4 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 P < 0.01): while D. mon ana and D. li o alis emales ha had o e win e ed in diapause eco e ed as e om chill coma han he emales ha had o e win e ed wi h ma u e o a ies in (F1,97 = 10.388, P < 0.01 and F1,87 = 4.316, P < 0.01, espec i ely), he e we e no signi ican di e ences be ween he wo o e win- e ing ypes in D. ezoana (F1,112 = 1.117, P = 0.293, Fig.3). Inc eased cold ole ance o D. mon ana and D. li o alis emales ha o e win e ed in diapause could be due o ha some diapause- ela ed mechanisms a e s ill ‘swi ched on’ a e he diapause has e mina ed and/o ha hese emales a e in gene al in a be e condi ion in sp ing han hose o e win e ing wi h ma u e o a ies. Gene exp ession kine ics du ing di e en phases o diapause. Gene exp ession changes ha occu du ing diapause ha e been commonly s udied by compa ing he exp ession le els o diapaus- ing and non-diapausing emales main ained in cons an diapause inducing o p e en ing ligh :da k cycles25–27. Mo eo e , mos o he ea lie s udies ha e ocused on gene exp ession changes occu ing du - ing a speci ic phase o diapause26,28, while only some s udies ha e co e ed he whole diapause pe iod11. We aced he gene exp ession kine ics du ing di e en phases o diapause and also be ween diapausing (D) and non-diapausing (ND) emales. Onse /ini ia ion phase. Ini ia ion phase o diapause was s udied by compa ing he gene exp ession pa e ns o 7-day-old emales ha we e des ined o diapause and emales ha we e a he ea ly s age o sexual ma u a ion (7D s. 7ND; Fig.1; Supplemen a y Table 1.). The gene showing highes up egula- ion in young diapause-des ined emales was CG9747, a gene wi h acyl-CoA Del a11-desa u ase ac i i y ha is in ol ed in lipid me abolism and oxida ion- educ ion p ocesses29. O he genes up egula ed in diapause-des ined emales included he cy oskele on modi ica ion gene Ac in 42A (Ac 42A), ci cadian clock gene pe iod (pe ) and couch po a o (cpo), which has been connec ed o diapause in D. melano- gas e 30 and in Culex pipiens31, as well as a calcium-binding p o ein egucalcin (homolog o D osophila cold acclima ion gene, Dca) (Fig.4). All hese genes, excep Ac 42A, ha e been linked wi h he main e- nance phase o diapause o wi h cold ole ance in D. mon ana in ou ea lie s udies24,26,32,33. To disen angle which genes a e up egula ed du ing he onse phase o diapause in con as o he main enance phase, we compa ed 7-day-old diapausing emales o 50-day-old diapausing emales. This compa ison b ough up mul iple genes connec ed e.g. o pho o ansduc ion (Supplemen a y Table 1.) Up egula ion o pho o asduc ion genes highligh s he gene ic basis o he esponse o diapause inducing day leng hs in no he n D osophila species. Among hese genes nei he inac i a ion no a e po en ial D (ninaD) showed he highes up egula ion in 7-day-old emales and ano he highly exp essed gene was Jonah 25Bi (Jon25Bi), which is known o be in ol ed in p o eolysis and o ha e endopep idase ac i i y29. Also Ac 42A and Myosin hea y chain (Mhc) genes we e up egula ed in young diapausing emales when compa ed o olde emales. Main enance phase. As he lies aged and p oceeded om he onse phase o he main enance phase, we obse ed mode a ely low gene exp ession changes (FC < 2) in mos o he s udied genes. A com- pa ison be ween he 50 (50D) and he 7 (7D) days old diapausing emales e ealed up egula ion in ou ibosomal genes, se e al hea shock and s ess ela ed genes, as well as genes linked o locomo ion Figu e 2. Onse o ep oduc i e diapause. Seasonal swi ch o ep oduc i e diapause in A) D. mon ana (s ains 3OL8, 26OL8 and 175OJ8), B) D. li o alis (s ains 202OJ8, 219J8 and 280OJ8) and C) D. ezoana (s ains 67OJ8, 124OJ8 and 143OJ8) emales (o de o he s ains as lis ed he e). The p opo ion o diapausing emales inc eased sha ply in all h ee species in pho ope iods sho e han 20:4LD. The c i ical day leng h (CDL), whe e he p opo ion o he emales en e ing o ep oduc i e diapause exceeded 50%, was 19:5LD in all species and s ains. www.na u e.com/scien i ic epo s/ 5 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 and ci cadian hy hms (Supplemen a y Table 1). As he lies aged u he and he main enance phase was p olonged o he o e win e ing phase (150D) (cons an da kness and + 4 °C o ou mon hs), he highes exp ession change was ound in Dopa deca boxylase (Ddc), which ca alyzes he inal s ep in he syn hesis o neu o ansmi e s dopamine and se o onin34. These so called biogenic amines ha e been connec ed o mul iple phenomena in D osophila sp., including s ess esis ance35 and longe i y36. O he up egula ed genes in 150D emales included se e al hea shock genes, o which many a e known o be highly up egula ed in se e al insec species du ing diapause10, and genes ha a e connec ed o ci cadian clock ( ille ( i) and ui less ( u)). Te mina ion phase. The e mina ion o he ep oduc i e diapause began a he end o he a i icial yea , du ing sp ing condi ions. 5–10 emales we e emo ed om he chambe a di e en ime poin s in he sp ing and hei o a ies we e checked o obse e he ini ia ion o he e mina ion phase (da a no shown). Du ing he sp ing, he 220-day-old emales had jus ini ia ed he e mina ion phase o ep o- duc i e diapause (220D) and we e s ill a he p e- i ellogenic s age. The highes exp ession di e ences in 220-day-old emales compa ed o 150-day-old diapausing emales we e de ec ed in he hea shock gene Hsc70, ollowed by he Phosphoglucona e mu ase (Pgm), inac i a ion no a e po en ial D (InaD) and A es in 2 (A 2) genes which a e all linked o pho o ansduc ion. This could indica e he impo ance o pho ope iodic cues du ing he e mina ion phase o he ep oduc i e diapause, al hough i has been hypo hesized ha empe a u e would be he main cue in he e mina ion o diapause13. A he age o 250 days du ing he sp ing condi ions, emales’ diapause was ully e mina ed and all he emales had ma u e o a ies. Compa ison be ween samples 250D and 220D e ealed d as ic exp ession changes in many genes ha had no been de ec ed du ing o he diapause phases. This is no su p ising as diapause e mina ion induces o a ian ma u a ion and se e al physiological changes. Due o he la ge amoun o di e en ially exp essed genes be ween hese wo o se phases, only pa o he da a will be dis- cussed he e, and a e no included in he Supplemen a y Table 1. Genes showing he highes up egula ion in 250-day-old i ellogenic emales, when compa ed o 220-day-old p e- i ellogenic emales, included Py oline 5-ca boxyla e educ ase (P5c : FC 315.7), in ol ed in p oline biosyn hesis, and h ee hea shock genes (Hsp26, Hsp70/Hsp90 o ganizing p o ein homolog (Hop) and Hsp23), and Jon25Bi. En i onmen al condi ions expe ienced du ing ea ly adul hood a ec gene exp ession le els la e in li e. Salminen and Hoikkala37 ha e shown ha he sensi i e pe iod o he onse o ep oduc- i e diapause in D. mon ana is a e eclosion, and i s leng h is a ec ed by empe a u e and i is induced by pho ope iodic cues. The en i onmen al cues expe ienced du ing he ea ly s ages o li e can po en- ially cause i e e sible changes in de elopmen bo h wi hin and be ween gene a ions38. In he e, we ha e Figu e 3. Chill coma eco e y imes o o e win e ed emales. Chill coma eco e y imes (CCRT; seconds ± SE) o 250-day-old D. mon ana, D. li o alis and D. ezoana emales ha had o e win e ed in he clima e chambe in ep oduc i e diapause ( illed ci cles) o in non-diapausing s a e (open ci cles). A he ime o es ing diapause was e mina ed and all emales had ully de eloped o a ies. www.na u e.com/scien i ic epo s/ 6 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 s udied he e ec o ea ly s age cues on wi hin gene a ion gene exp ession pa e ns be ween diapausing and non-diapausing emales. Compa isons be ween emales ha o e win e ed ei he in diapause (150D) o wi h ma u e o a ies (150ND) enabled us o iden i y gene exp ession di e ences be ween he emales ha we e o he same age and ha had been main ained in he same condi ions a e he i s 10 days o hei li e ( he i s 10 days a e eclosion we e spen ei he in diapause inducing o diapause inhibi ing condi ions). Dissec ion o he o a ies showed ha emales o he 150D g oup we e in diapause while he emales o he 150ND g oup had no laid hei eggs o abso bed he yolk om he eggs du ing he o e win e ing pe iod (i.e. hey s ill had ully de eloped o a ies). Compa ison o he gene exp ession pa e ns be ween 150D and 150ND emales (Table1) e ealed up egula ion in se e al genes in diapausing emales, including Ca alase (Ca ), Ac 42A, pe , Mhc, and Tho (Fig.4), sugges ing ha hese gene ha e an impo an ole in o e - win e ing diapausing emales. In e es ingly, many o he genes up egula ed in he ea lie phases o diapause we e also up egula ed in 150D emales when compa ed o 150ND emales, e en hough hese wo g oups had been main ained in he same condi ions o mo e han 140 days and had been collec ed om he chambe du ing he same seasonal ime poin (o e win e ing; + 4 °C and cons an da kness). The same end was also seen in he non-diapausing 150-day-old emales, which showed a high esemblance o he younge non-diapausing emales (da a no shown). These indings show ha he swi ch be o e sexual ma u a ion o ep oduc- i e diapause du ing he i s 10 days o hei li e leads o he es ablishmen o semi-pe manen gene exp ession p o iles ha pe sis h oughou adul hood. The indings also emphasize he impo ance o Gene symbol Ini ia ion 7D / 7ND Ini ia ion 7D / 50D Main enance 50D / 7D O e win e ing 150D / 50D O e win e ing 150D / 150ND Te mina ion 1 220D / 150D Te mina ion 2 250D / 220D Molecula unc ion/biological p ocess/ p o ein unc ion Ac 42A 2.5** 6.4*** — — 2.5** — — s uc u al cons i uen o cy oskele on A 1 — — — — 3.1** — — opsin binding, pho o ecep o cell main enance Ca — — — — 2.9** — — ca alase ac i i y CG6785 — — 9.4*** 6.1*** 4.0** — — s ea oyl-CoA 9-desa u ase ac i i y CG8630 — — — — 3.6** — — ol age-ga ed po assium channel ac i i y CG15531 — — — — 9.3*— — esponse o hea CG17928 — — — — 10.8*** — — s ea oyl-CoA 9-desa u ase ac i i y D oj2 — — — 3.2*** 3.2*** — — Wn -p o ein binding, signal ansduc ion d p — — 2.2** — 2.2** — 2.7*** locomo o dy — — — 2.1*2.6** — — imaginal disc-de i ed wing mo phogenesis Eip71CD — 4.2*** — — 2.9*— 3.3*** de e mina ion o adul li espan u — — — 3.0*** 2.2** — — mul i-o ganism ep oduc i e p ocess, ma ing Gale 2.8** 2.5*** — — 2.9*** — — UDP-glucose 4-epime ase ac i i y Hsc70 — — 3.4*** 3.3*** 2.9*** 13.0** —hea shock chape onin-binding ken — — — — 2.5*** — — emale analia de elopmen LanA — — 2.4*** — 5.0*** — — immune esponse; g ow h; locomo o y beha io Mhc — 3.9*** — — 2.2** — — ac in-dependen ATPase ac i i y ninaD — 28.4*** — — 3.1*— — sca enge ecep o ac i i y, pho o ansduc ion pe 2.1** — 2.7*** 2.1*2.2*— — phosphoglyce a e mu ase ac i i y so — — — — 2.0** — — chape one binding, p o ein lipida ion Tho — — 2.3** 2.5** 2.4** — — immune esponse, de e mina ion o adul li espan i — — — 4.0*** 2.1*— — ci cadian hy hm Table 1. Gene exp ession old changes be ween di e en diapause phases and seasons. Lis o genes ha showed signi ican exp ession changes wi h Fold Change (FC) ≥ 2, when he 150-day-old diapausing and non-diapausing emales we e compa ed. Table also shows he o he diapause phases, whe e he same genes we e up egula ed. Molecula unc ion and/o biological p ocess and /o p o ein unc ion a e acco ding o he Flybase e sion 3_2013. Signi icance le els: *P < 0.05 **P < 0.01 ***P < 0.001. www.na u e.com/scien i ic epo s/ 7 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 he co ec iming o diapause and he long las ing e ec s o he induced de elopmen al pa hways la e in li e (Table1, Fig.4). Gene exp ession changes be ween di e en phases o diapause. Ou s udy pinpoin s a ious genes whose unc ion is connec ed o speci ic phases o diapause, bu also hose genes ha a e up eg- ula ed h oughou he diapause. F om he 219 genes on he mic oa ay, Ac 42A, Mhc, Jon25Bi, Tho , pe , CG6785, d ape (d p ), Ecdysone-induced p o ein 28/29kD (Eip71CD) and Hsc70 (Table1), s ayed up egula ed du ing di e en phases o diapause (Fig.4). Ac in and myosin p o eins a e in ol ed in he s uc u al cons i uen o cy oskele on and may be in ol ed in he modi ica ion o he insec s’ cy oskele on du ing ea ly adul hood a he han in diapause egula ion pe se. Howe e , ac in has been ound o be highly abundan du ing he ea ly diapause in Nasonia i ipennis la ae39 and Culex pipiens adul s40. In ou s udy, bo h Ac 42A and Mhc genes we e up egula ed in young emales (7D) du ing he onse s age o diapause, as well as in o e win e ing dia- pausing emales (150D), when compa ed o non-diapausing o e win e ing emales (150ND). Jon25Bi Figu e 4. Seasonal gene exp ession kine ics. Seasonal gene exp ession kine ics is shown as means and s anda d e o o means (SEM) o he hyb idiza ion in ensi y alues among he mic oa ay p obes wi h he highes in ensi ies o a gi en gene. Diapausing emales a e ma ked wi h illed ci cles, non-diapausing emales wi h open ci cles, and he 250-day-old emales ha had o e win e ed in ep oduc i e diapause, bu had al eady e mina ed ep oduc i e diapause and de eloped o a ies, a e ma ked sepa a ely wi h g ay. www.na u e.com/scien i ic epo s/ 8 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 also known as Se ine p o ease 4, has a se ine- ype endopep idase ac i i y. Up egula ion o Jonah-genes has been de ec ed in D. melanogas e immedia ely a e diapause, du ing egg de elopmen , and he gene has been sugges ed o play a ole in pos -diapause egula ion o diges i e ac i i y41, as well as in he signal- ansduc ion ac i a ing Toll-pa hway in immune esponse42. In ou s udy, Jon25Bi was up egula ed h oughou he onse , main enance and e mina ion pe iods o diapause (Table 1). Tho (synonyms: elF4E binding p o ein, d4E-BP), on he o he hand, was up egula ed du ing he main enance (50D) and o e win e ing (150D) s ages o diapause. In D osophila, Tho is egula ed a he ansc ip ional le el by a o khead ansc ip ion ac o Foxo43,44, which is a downs eam signaling componen o he insulin/IGF1 pa hway45. Shu down o he insulin-signaling pa hway, leading o Foxo ac i a ion, has been sugges ed o play a ole in he induc ion o ep oduc i e diapause in Culex pipiens46. The peak in he exp ession le el o Tho in diapausing D. mon ana emales du ing he main enance and o e win e ing pe iod may hus be a e lec ion o inc eased ac i i y o Foxo. Tho ansc ip ion has also been connec ed o s a - a ion and oxida i e s ess in D osophila47,48 and we ha e ound i o be up egula ed du ing cold accli- ma ion in non-diapausing D. mon ana emales24. The ci cadian clock gene pe iod (pe ) has been ound o show d ama ic changes in i s exp ession le el in he p esence o absence o a he mope iod and/o pho ope iod49,50. In e es ingly, pe was up egula ed h oughou he li e-cycle o diapausing D. mon ana emales including he o e win e ing pe iod (150D emales in cons an empe a u e and da kness). Ikeno e al.51 ha e de ec ed he in ol emen o pe in diapause induc ion also in bean bug, Rip o us pedes is, whe e silencing o pe h ough RNAi caused he bugs o a e diapause e en when hey we e aised in a diapause-inducing pho ope iod. Consis ency o gene exp ession pa e ns be ween di e en D. mon ana s ains and sis e species. Gene exp ession pa e ns o His3.3A, egucalcin, cpo and Tho genes we e in es iga ed wi h a qPCR me hod using 150-day-old diapausing and non-diapausing D. mon ana emales om he s ain ha was used in he mic oa ay assay and wo addi ional s ains. In addi ion, he exp ession o he same genes was obse ed om h ee s ains o D. li o alis and D. ezoana (Fig. 5, Table 3 in supple- men a y ma e ial). Fo hese s udies one up egula ed (Tho ) and one down egula ed (His3.3A) gene (in 150-day-old diapausing lies) we e selec ed, as well as wo genes (cpo and egucalcin) ha did no show signi ican di e ences be ween diapausing and non-diapausing emales in he mic oa ay s udy, bu ha e been connec ed o diapause in ou p e ious s udies in D. mon ana26,32,33. qPCR esul s showed simila up- and down egula ion as obse ed in he mic oa ay assay o all he genes s udied, and species compa isons indica ed a consis ency in gene exp ession be ween hem; he only excep ion was cpo, which showed opposi e pa e ns o gene exp ession in D. mon ana and D. ezoana, compa ed o one D. li o alis s ain (Fig.5). The highes o e all exp ession alues we e obse ed in His3.3A, wi h signi ican down egula ion in diapausing emales in all h ee species, con i ming he obse a ion in he D. mon ana mic oa ay s udy. His one genes ha e been ound o be down egula ed du ing diapause also in N. i ipennis39 and Caeno habdi is elegans52 and he exp ession o hese genes is known o be connec ed wi h DNA eplica ion and p o ein ansla ion53. Among he o he h ee genes, Tho was signi ican ly up egula ed in diapausing D. mon ana emales in he mic oa ay assay, while in he qPCR analysis he exp ession changes eached signi ican le els in only one D. mon ana s ain and in one o he wo D. ezoana and D. li o alis s ains (Fig.5). In he mic oa ay assay, he a iance in he exp ession o egucalcin gene be ween he h ee biological eplica es o he 150ND emales was so la ge ha i is ha d o say whe he he gene was up- o down egula ed in compa ison o 150D emales. In he qPCR s udy, howe e , egucalcin was signi ican ly down egula ed in 150D emales in all he h ee species. Finally, cpo showed a sligh , bu non-signi ican up egula ion in 150D emales in he mic oa ay assay, while in qPCR a signi ican up egula ion was e iden in all he D. mon ana s ains and in one D. ezoana s ain, and su p isingly a signi ican down egula ion in one D. li o alis s ain. Conclusions Gene exp ession and pheno ypic changes induced by pho ope iodic and/o empe a u e cues play an impo an ole in adap a ion o seasonally a ying en i onmen s and a e a widesp ead phenomenon among empe a e species. Howe e , modula ion o gene exp ession, which is one o he main mecha- nisms leading o plas ici y in li e-his o y ai s, has only seldom been s udied in condi ions esembling hose in he wild. We de e mined he onse o ep oduc i e diapause in h ee no he n sympa ic D. i ilis g oup species, showing ha he seasonal ime window o en e ing o diapause is almos iden ical be ween hese sympa ic species, indica ing s ong adap a ion in he iming o ep oduc i e diapause. When he gene exp ession kine ics was s udied be ween he di e en phases o diapause in D. mon ana, he ele a ed ole o he pho o ansduc ion genes in he onse phase indica ed he impo an ole o he pho ope iodic cues on he co ec esponses a he physiological le el. Mos impo an ly, we also show ha some o he s udied genes whose up egula ion was induced a he onse phase o diapause, we e up egula ed h oughou he li espan o he emales when compa ed o non-diapausing emales expe iencing he same en i onmen al condi ions. These indings highligh he impo ance o en e ing ep oduc i e diapause a a co ec seasonal ime and i s long las ing e ec s on he lies’ li e-his o y ai s, and enable us o plan and ca y ou new s udies on lies’ diapause and cold ole ance e.g. on he whole ansc ip ome le el. www.na u e.com/scien i ic epo s/ 9 Scien i ic RepoR s | 5:11197 | DOi: 10.1038/s ep11197 Ma e ials and me hods S udy species and s ains. The s udy was pe o med using adul emales om h ee iso emale lines o D osophila mon ana (3OL8, 26OL8 and 175OJ8), D. li o alis (202OJ8, 218OJ8 and 280OJ8) and D. ezoana (67OJ8, 124OJ8 and 143OJ8). The mic oa ay expe imen was pe o med using D. mon ana s ain 175OJ8, while he pheno ypic s udies (swi ch o diapause and changes in cold ole ance) and qPCR we e pe o med wi h he emales o all nine s ains. The s ains we e es ablished om he p ogenies o emales collec ed om Oulanka (Finland; 66 °N and 29 °E) in summe 2008 and main ained in he labo a o y unde diapause-p e en ing condi ions (con inuous ligh , 19 ± 1 °C, 65% humidi y) since hei es ablish- men . Expe imen s we e s a ed wi hin one yea o he s ains’ es ablishmen . Mimicking en i onmen al condi ions in a clima e chambe . D. mon ana, D. li o alis and D. ezoana emales we e collec ed om he main enance s ock bo les wi hin one day o eclosion and ans- e ed in o he clima e chambe (Sanyo MLR-351H, Sanyo, CA, USA), which had been p og ammed o mimic daily and seasonal changes in day leng h and empe a u e om ea ly summe o la e sp ing a he lies’ home si e in no he n Finland. T ans e ing young emales in o he chambe in ea ly and la e summe condi ions enabled us o ob ain bo h non-diapausing and diapausing emales, which we e hen ea ed in he chambe o up o 9 mon hs (Fig.1). Humidi y was kep a 60% h oughou he a i icial yea , excep du ing he win e pe iod when i was dec eased o 30%. Daily cycles in pho ope iod and empe a u e we e changed weekly a he same pace as hese changes occu in na u e; he only excep ion was he o e win e ing pe iod du ing which he lies we e main ained in cons an da kness and a + 4 °C o ou mon hs. This pe iod was kep sho e and milde han he win e in he wild, which enabled us o keep bo h diapausing (D) and non-diapausing (ND) emales ali e o e he win e pe iod. (Fig.1). Seasonal iming o ep oduc i e diapause. The day leng h a which D. mon ana, D. li o alis and D. ezoana emales en e ep oduc i e diapause was aced by ans e ing eshly eme ged emales o h ee iso emale s ains (50–80 emales pe s ain) pe species in o he chambe du ing six ime poin s ep esen ing he condi ions be ween 11 h o July (LD 22:2) and 23 d o Augus (LD 16:8) in he wild. The o a ian de elopmen al s age o he emales was checked 21 days a e he emales had been ans e ed o he chambe (Fig.1), using a p o ocol desc ibed in Tyukmae a e al.14. Cold ole ance o o e win e ed D. mon ana, D. li o alis and D. ezoana emales. The e ec s o o e win e ing on he cold ole ance o 250-day-old emale lies ha had spen he win e ei he in diapause (250D) o wi h ma u e o a ies (250ND), was es ima ed by measu ing hei chill coma eco e y imes (CCRT)54. Samples o 250ND emales o hese expe imen s we e ob ained by main aining a g oup o eshly eme ged emales om each s ain o en days a e hei eclosion in cons an ligh and a 19 °C o induce o a ian de elopmen , and ans e ing hem in o he chambe in ea ly Sep embe condi ions (3 d o Sep embe ). Samples o 250D emales we e ob ained by ans e ing a se o emales ha had eme ged on he same day as he abo e-men ioned ND emales and ans e ing hem immedia ely in o he chambe (23 d o Augus ) in diapause inducing condi ions (Fig.1). The cold ole ance o o e win e ed ND and D Figu e 5. Gene exp ession le els be ween he h ee s udied D osophila species. Mic oa ay esul s we e alida ed om he D. mon ana s ain 175OJ8 wi h qPCR. Beside he alida ion, exp ession le els o ou candida e genes we e compa ed be ween 150-day-old diapausing and non-diapausing emales in wo addi ional D. mon ana (A), h ee D. li o alis (B) and h ee D. ezoana (C) iso emale s ains. Fold change di e ences o he His one 3.3A and Tho genes in D. mon ana s ain 175OJ8 used in he mic oa ay s udy a e shown in D. mon ana igu e. The wo o he genes, cpo and egucalcin, did no show signi ican exp ession changes in he mic oa ay s udy. Signi icance le els oge he wi h old change alues a e gi en in he supplemen a y able 2 o all he compa isons.