Selection for reproduction under short photoperiods changes ...The incidence of reproductive...

RESEARCH ARTICLE

Selection for reproduction under short photoperiods changesdiapause-associated traits and induces widespread genomicdivergenceHannele Kauranen1Dagger Johanna Kinnunen1 Anna-Lotta Hiillos1 Pekka Lankinen2 David Hopkins1R Axel W Wiberg3 Michael G Ritchie3 and Anneli Hoikkala1

ABSTRACTThe incidence of reproductive diapause is a critical aspect of lifehistory in overwintering insects from temperate regions Much hasbeen learned about the timing physiology and genetics of diapausein a range of insects but how themultiple changes involved in this andother photoperiodically regulated traits are inter-related is not wellunderstood We performed quasinatural selection on reproductionunder short photoperiods in a northern fly species Drosophilamontana to trace the effects of photoperiodic selection on traitsregulated by the photoperiodic timer andor by a circadian clocksystem Selection changed several traits associatedwith reproductivediapause including the critical day length for diapause (CDL) thefrequency of diapausing females under photoperiods that deviatefrom daily 24 h cycles and cold tolerance towards the phenotypestypical of lower latitudes However selection had no effect on theperiod of free-running locomotor activity rhythm regulated by thecircadian clock in fly brain At a genomic level selection inducedextensive divergence from the control line in 16 gene clusters involvedin signal transduction membrane properties immunologlobulins anddevelopment These changes resembled those detected betweenlatitudinally divergent D montana populations in the wild and involvedSNP divergence associated with several genes linked with diapauseinduction Overall our study shows that photoperiodic selection forreproduction under short photoperiods affects diapause-associatedtraits without disrupting the central clock network generating circadianrhythms in fly locomotor activity

KEY WORDS Reproductive diapause Critical day length Coldtolerance Genome analyses Photoperiodic timer Circadian clock

INTRODUCTIONChanges in day length often act as token stimuli which helporganisms to anticipate the forthcoming cold season (Tauber et al1986) In northern arthropod species with a facultative reproductivediapause emerging females enter reproductive diapause when theday length decreases below a critical point The timing of diapause

described as a critical day length (CDL) critical night length (CNL)or critical photoperiod is under strong local selection pressure largelybecause of restriction in the length of the growing period and stressfuloverwintering conditions Several insect species have been found toshow latitudinal variation in this trait (eg Bradshaw and Lounibos1977 Lankinen 1986a Schmidt et al 2005 Tyukmaeva et al2011) as well as in the frequency of diapausing individuals (Schmidtet al 2005) cold tolerance (Božicevic et al 2016 Soumlrensen et al2016) and the period and damping rate of their free-running rhythms(Allemand and David 1976 Lankinen 1986a Pittendrigh andTakamura 1989) However the extent to which these multipleresponses are correlated or independent is still poorly understood

The circadian clock systems of most multicellular organismsconsist of at least one specialized pacemaker oscillator (the centralclock) that responds to environmental signals and coordinatesrhythmic output in peripheral oscillators In Drosophilamelanogaster the peripheral oscillators in different tissues canalso be directly entrained by light (Bell-Pedersen et al 2005) anddaily (circadian) rhythms such as locomotor activity and eclosionare controlled by different oscillators which may differ inperiod length and light sensitivity (Engelmann and Mack 1978)Furthermore circadian rhythms may differ in the peptidergic circuitthat links the clock to motor outputs modulating these rhythms(King et al 2017 King and Sehgal 2018) Photoperiodic control ofseasonal (circannual) rhythms is even more complicated as theserhythms can be regulated either by a circadian clock network or aphotoperiod timer andor by their interaction (Hut et al 2013Denlinger et al 2017) The photoperiodic timer helps organisms todistinguish short photoperiods from long ones and the traits itregulates are typically onoff traits such as diapause response ormigration (Saunders 1981 Bradshaw and Holzapfel 2010)Genetic and physiological mechanisms behind photoperiodic timemeasurement are still largely unknown some models stressing theirindependence from and others their integration with the circadianclock In hourglass-type models photoperiodic time measurementis suggested to function independently and be based on theaccumulation of a hypothetical chemical substance during thedark period so that a photoperiodic response is induced after acertain number of short-night cycles have accumulated in thecounter system (Lees 1973 1986) A circadian-based model forphotoperiodic time measurement originally proposed by Buumlnning(1936) has been developed into several robust models involvingone or more circadian oscillators that may be coupled to each otherand which may show a certain degree of damping (see Vaz Nunesand Saunders 1999) An hourglass-like clock and the dampedoscillator model could in fact be based on the same clockmechanism and differ only in the damping coefficient of theoscillator(s) concerned (Saunders 2005)Received 3 May 2019 Accepted 4 September 2019

1Department of Biological and Environmental Science University of Jyvaskyla40014 Jyvaskyla Finland 2Department of Biology University of Oulu 90014 OuluFinland 3School of Biology Dyers Brae House University of St AndrewsSt Andrews KY16 9TH UKPresent address Department of Environmental Sciences Evolutionary BiologyUniversity of Basel Basel CH-4051 Switzerland

DaggerAuthor for correspondence (hannelekauranen22gmailcom)

HK 0000-0001-8792-4983 JK 0000-0003-3443-5507 DH 0000-0001-7752-0141 RAWW 0000-0002-8074-8670 MGR 0000-0001-7913-8675

1

copy 2019 Published by The Company of Biologists Ltd | Journal of Experimental Biology (2019) 222 jeb205831 doi101242jeb205831

Journal

ofEx

perim

entalB

iology

Insect photoperiodic responses involve a sequence of events fromphotoreception and the measurement of night or day length to theaccumulation of a hypothetical diapause titer in a countermechanismand the downstream regulation of events leading to release orretention of neurohormones regulating diapause (Saunders 2014)Both the measurement of successive photoperiods and theiraccumulation by the counter mechanism are likely to functionunder the circadian system (Saunders 2012) while the downstreamcascading system governing circadian behaviors and photoperiodicresponses may be completely different (Goto 2013) Denlinger et al(2017) suggested that rapid adaptive response in traits regulated byphotoperiodic timer could occur without disrupting daily circadianorganization even if one or a few circadian clock genes functionpleiotropically as a phase reference point for the photoperiodictimer Circadian clock output genes whose dysregulation disruptsbehavioral rhythms without affecting oscillations in the molecularclock itself have been identified in several species (Omura et al2016 King et al 2017 King and Sehgal 2018) It is also becomingobvious that neurotransmitters and neuropeptides play a central rolein regulating the production andor release of hormones involved indiapause induction (Sim and Denlinger 2013 Naumlssel et al 2013Andreatta et al 2018) Reproductive diapause can further trigger ahormonal cascade which induces changes in metabolism (Kubraket al 2014) increases cold tolerance (Vesala et al 2012Wallingford et al 2016) and reduces insectsrsquo mating drive andpheromone production (Ala-Honkola et al 2018) However there isstill debate on whether changes in CDL and other photoperiodicallyregulated traits occur through local selection directed on thecircadian clock system (Pittendrigh and Takamura 1987) ordirectly on the traits involved (Bradshaw et al 2003)Quasinatural selection where the study organisms are transferred

into altered environmental conditions over successive generationsoffers an efficient way to examine the role of environmental factorsin trait evolution and to study trait correlations and trade-offs (Fry2003) For example Agrawal (2000) performed such selectioninducing a host-shift in a polyphagous spider mite Tetranychusurticae and found evidence of a correlation between mitesrsquoperformance on a novel host and their host-plant preference Thelsquoevolve and resequencersquo approach goes one step further by allowingevolutionary responses to selection to be examined at the genomelevel to identify diverged genomic regions and functional pathwaysas well as localizing genes diverging under the altered selectionregime (Turner et al 2011 Burke 2012 Wiberg et al 2017)Martins et al (2014) used this technique and exposed Drosophilamelanogaster flies to Drosophila C virus for 20 generationsAfter selection they pool-sequenced treated and control lines andfound two genome regions containing several candidate genesconnected with an increased survival of the flies under differentviral treatmentsWe have used quasinatural selection combined with the lsquoevolve

and resequencersquo approach to find out whether selection forreproduction under short photoperiods changes CDL and othertraits associated with reproductive diapause and whether it affectsthese traits directly or by modifying the action of the circadian clocknetwork The selection experiment was performed on a northern flyspeciesDrosophila montana where the photoperiod is a key cue forthe induction of adult reproductive diapause This species showsrobust latitudinal clines in CDL (Tyukmaeva et al 2011 Lankinenet al 2013 Venera Tyukmaeva PL JK HK and AHunpublished data) and in the frequency of diapausing femalesunder lightdark (LD) cycles that deviate markedly from 24 hphotoperiodic cycles in NandandashHamner (NndashH) experiment (PL

unpublished results) D montana is one of the most cold-tolerantDrosophila species (Kellermann et al 2012) and its cold-tolerancehas been found to show variation between latitudinally andaltitudinally distinct populations (Vesala and Hoikkala 2011RAWW unpublished results) as well as seasonal photoperiodiccold acclimation (Vesala et al 2012) We hypothesize that if ourselection affects CDL directly and not through changes in the centralcircadian clock network it should affect CDL and other diapause-associated traits but it should not disrupt fly locomotor activityrhythm which is regulated by the central circadian clock (King et al2017) We also expect the genetic divergence of selection and controllines to resemble divergence detected among latitudinally distinct wildDmontana populations (RAWW unpublished data) and to includeSNP variation associated with genes important in the photoperiodiccontrol of CDL and other traits which diverge during selection

MATERIALS AND METHODSQuasinatural selection for reproduction under short-day(late summer) conditionsWe established a genetically variable base population ofDrosophilamontana Stone Griffen and Patterson 1941 by collecting 102fertilized females from a northern Oulanka (Finland 664degN)population in summer 2013 Species identification was performedby sequencing part of the mtDNACOI region of one progeny fly perfemale as described in Simon et al (1994) using forward andreverse primers COI_1F 5prime-ATCTATCGCCTAAACTTCAGCC-3primeand COI_1R 5prime-ACTTCAGGGTGACCAAAAAATC-3prime F3progenies of wild-caught D montana females (about 10 virginsexually mature females and males per progeny) were transferredinto a population cage containing eight bottles with malt medium(Lakovaara 1969) The flies were allowed to mate and produceprogeny in the cage for another 3 generations before startingselection (Fig 1A) During selection the flies were maintained inmalt vials in wooden light-insulated cabinets illuminated withcompact fluorescent lamps (LedEnergie) with light intensity of600ndash1100 lx during the photophase in 17plusmn1degC Length of the lightperiod was decreased by 05 h in each generation in the chamberscontaining selection line replicates until the incidence of femalediapause exceeded 50 This helped us to keep the strength ofselection close to 50 (keeping the flies continuously in the sameday length would have lowered the diapause percentage and reducedselection strength in each generation Roff 1994) Control linereplicates were maintained through the experiment in constant 24 hlight (LL) mimicking the light condition at Oulanka during fliesrsquobreeding season in early summer

In generation F6 200 females and males were separated within3 days of emergence When sexually mature (sim3 weeks old) theflies were allowed to mate in malt vials for 48 h in groups of 10 fliesof each sex They were then transferred into fresh malt vials forfurther mating and egg laying and into new vials 3 times at 3 dayintervals to prevent larval crowding Half of the egg vials weretransferred into 19 h5 h lightdark (LD) (selection line CDL of thebase population) and the other half kept in LL (control line) beforethe flies of next generation flies started to emerge (Fig 1B) Thesame procedure was repeated in generation 7 in respectivephotoperiods using 100 sexually mature females and males perline to establish 3 replicates for both lines (Fig 1C) In generationsF8ndashF15 selection was continued with 100 females and 50 males perreplicate for selection line replicates (half of the females wereexpected to enter diapause) and 50 females and 50 males for controlline replicates (Fig 1D) In each generation the flies were allowedto mate for 48 h in the same LD conditions in which they had

2

RESEARCH ARTICLE Journal of Experimental Biology (2019) 222 jeb205831 doi101242jeb205831

Journal

ofEx

perim

entalB

iology

emerged after which they were transferred into malt vials for egglaying as described above The egg vials of selection line replicateswere transferred in each generation into 05 h shorter photoperiods(in generation F8 progeny flies were kept in the same LD conditionsas their parents because there were few flies but this wascompensated for by transferring the progenies into 1 h shorterphotoperiods in the next generation)

The proportion of diapausing females was calculated in eachgeneration by dissecting 77ndash100 females for each selection linereplicate and about 50 females for the control line replicates afterprogeny production Selection was completed in generation F15where the frequency of diapausing females increased above 50 inall selection line replicates After finishing active selection theselection line replicates were maintained in LD 159 (the final LD

Population cage established from F3 progenies of 102 wild-caught and transferred from 19degC (LL) to 17degC (LL) before the flies of F4generation started to emerge

200 and 200 mated and allowed to lay eggs Egg vials divided into LD 195 and LL conditions (17degC)bull CDL

100 and 100 emerged inLD 195 allowed to mateand lay eggs to establish 3selection line replicates

bull Locomotor activity rhythm

100 and 50 per replicatetransferred in each generation into a shorter LD and allowed to reproduce as long as female diapause remained lt50

50 and 50 per replicateallowed to reproduce in LL(female diapause closeto 0 in all generations)

Selection line maintained in LD 159and control line in LL (16degC) for 8ndash10months in mixed generationsbull CDLbull NndashH experiments

Eggs of F15 generation transferred to LLin both linesbull F16 Genome samplesbull F17 CDLbull F18 Cold tolerance and acclimationbull F19 Locomotor activity rhythm

100 and 100 emerged in LLallowed to mate and lay eggs to establish 3 control line replicates


F3ndashF5

A

B

C

D

E

F6

F7

F8ndashF15

Selection line Control line

Fig 1 Procedure for quasinatural selection for reproduction under shorter day length complemented with phenotypic and genotypic assays inDrosophila montana The experiment was started by establishing a genetically variable base population (A) dividing the flies into two lighting conditionsLD 195 and LL (B) and establishing three replicates for each line (C) The flies of the selection line were transferred into 05 h shorter day length when thefemale diapause percentage decreased below 50 in all replicates of this line while control line flies were maintained in LL throughout the experiment (D)Phenotypic assays were performed before selection (generations F6 and F7 BC) 2ndash4 generations after active selection (generations F17ndashF19) and aftermaintaining the flies for 8ndash10 months under diapause-inducing conditions (selection line) or in LL (control line E) Generations shown on the right side andorwithin the boxes refer to adult generation and fly numbers to one replicate of each line (both lines had three replicates from generation F8)

3


Journal

ofEx

perim

entalB

iology

conditions used in the selection experiment) and the control linereplicates in LL at 16degC in mixed generations

Phenotypic assaysTo determine whether selection for reproduction under shortphotoperiods induced changes in CDL and other photoperiodicallycontrolled traits and whether it also affected traits thought to beregulated by circadian clocks we assayed several phenotypic traits atone or more stages of selection experiment The first set ofphenotypic assays including CDL and female locomotor activityrhythm was performed in the base population (Fig 1B) CDL coldtolerance and locomotor activity rhythm were analyzed for allselection and control line replicates 2ndash4 generations after selectionhad finished (Fig 1E) Prior to performing the latter assays all linereplicates were maintained in constant light (LL) for two to threegenerations to exclude possible nongenetic components of linedifferences according to the protocols of quasinatural selection(Fry 2003) Generations where the assays were performed are givenin Fig 1The frequency of diapausing females under photoperiodic cycles

that deviate from daily 24 h cycles in NndashH experiments was studiedusing the flies that had been maintained in LD 159 (selection linereplicates) and LL (control line replicates) in mixed generations in16degC for 8 to 10 months after finishing active selection Theseexperiments as well as the last CDLmeasurements were performedin the University of Oulu (all other experiments were done in theUniversity of Jyvaumlskylauml)

Methods and regimes for fly phenotypingCritical day length for diapause induction (CDL) was estimatedfrom the photoperiodic response curves as the point where 50 offemales enter diapause Emerged flies were maintained underdifferent LD cycles for 21 days in malt vials in similar cabinets asthe ones used for fly maintenance during selection (compactLedEnergie fluorescent lamps with light intensity of 600-1100 luxduring the photophase) After this the reproductive stage of thefemales was determined by dissecting their abdomen and recordingwhether their ovaries contained at least one fully developed egg orwhether they had entered diapause (characterized by small ovarieswith no egg yolk Tyukmaeva et al 2011) The photoperiodicresponse curve of the base population females was plotted for theLDs 204 195 186 177 and 168 in 17plusmn1degC (N=59ndash95 females

LD exact sample sizes in Table 1) Respective curves for the controland selection line replicates after the experiment were drawn in LDs213 204 195 186 177 168 159 1410 and 1311 in 17plusmn1degC(N=69ndash97 femalesLD more detailed information about the samplesize can be found in Table 1) in generation F17

CDL of selection and control line replicates was measured asecond time 8 to 10 months after finishing active selection using thesame protocol and similar cabinets Here the cabinets wereilluminated with one white fluorescent lampchamber (9 WMegaman Germany) with light intensity of 300-1000 lux andthe photoperiodic response curves were drawn in LDs 195 186177 and 168 for the selection line replicates and in LDs 222 213204 195 and 186 for the control line replicates in 16plusmn03degC(N=131ndash507 femalesLD exact sample sizes in Table 1)

The NndashH experiment involves sets of experiments where a shortlight period (10ndash12 h) is usually coupled with varying lengths ofdark periods so that the total period of the LD cycle varies betweenexperiments (Nanda and Hamner 1958) If the photoperiodic timemeasurement relies on the function of circadian clock(s) the peakamplitude of insect photoperiodic responses should fluctuate in asim24 h rhythm reflecting the period of the circadian oscillator Basedon our earlier NndashH experiment (Kauranen et al 2013) we knowthat D montana females do not show cycling amplitude peaks intheir diapause response under changing night lengths and in thepresent study our main interest was focused on the effects ofselection on femalesrsquo overall tendency to enter diapause under theseconditions NndashH experiments were performed on selection andcontrol line replicates which had been maintained in LD 159 andLL respectively in mixed generations in 17degC for 4 to 5 generationsafter finishing active selection The study consisted of five separateexperiments performed in LD 126 1212 1224 1236 or 1272(choice of these photoperiods was based on a previous NndashHexperiment Kauranen et al 2013) In addition a sample of flieswas maintained in constant darkness (DD) to see if the femalesrequire light input to enter diapause Experimental flies (87ndash388femaleseach LD cycle or DD specific information about thesample sizes can be found in Table 2) were transferred intothe climate chambers at the pupal stage in bottles containing maltmedium so that the bottles containing flies of one selection andone control line replicate were always in the same chamberCabinets were illuminated with compact fluorescent lamps (9 WMegaman Germany) with light intensity of 300ndash1000 lx during the

Table 1 Number of the studied flies in each LD conditionwhenmeasuring CDL for base population and for selection and control line replicates bothin generation F17 and 8ndash10 months after finishing active selection

Replicate

LD

222 213 204 195 186 177 168 159 1410 1311

Base population F17 78 94 95 80 59Selection 1 86 89 90 84 81 94 89 87 76

2 85 69 83 86 87 81 86 86 893 90 81 86 93 78 74 82 87 89

Control 1 92 86 76 96 78 82 82 82 882 80 81 81 86 81 77 83 81 863 91 97 94 88 85 86 87 81 83

8ndash10 months after finishing selectionSelection 1 138 223 303 313

2 143 282 332 2473 131 507 494 355

Control 1 296 319 289 257 1812 479 338 491 335 3313 340 171 313 187 172

4


Journal

ofEx

perim

entalB

iology

photophase at 16plusmn03degC Flies were kept in the chambers for21plusmn2 days until they reached maturity and then transferred into afreezer until the diapause state of all females was determined asexplained aboveCold tolerance of D montana flies increases in response to a

decreasing photoperiod in late summer and autumn (photoperiodiccold acclimation Vesala et al 2012) and to take this phenomenoninto account we measured the cold tolerance of selection andcontrol line for the flies reared in LL and in LD 195 177 159 and1311 Parental females were allowed to lay eggs in the given LDconditions in successive 3 h periods and the emerging flies werecollected at 12 h intervals which allowed us to collect materialevenly along the modal distributions of fliesrsquo pre-adult developmenttime of each replicate Each of the five samples per line replicateconsisted of 16 males and 16 females which were used in theexperiments at the age of 20ndash27 days Female reproductive stagewas determined after the experiment as before Fliesrsquo cold tolerancewas measured using the critical thermal minimum (CTmin) method(Overgaard et al 2011) the flies were transferred into a water-glycol bath in sealed vials in 16degC after which temperature wasdecreased by 05degC minminus1 The temperature when the flies enteredchill coma (ie were no longer able to stand on their feet) wasrecorded as their CTmin The total number of experiments was 64and the males and females from different selection regimes and LDswere randomly assignedLocomotor activity rhythm of the flies was traced in Trikinetics

Drosophila High Resolution Activity Monitors (Waltham MAUSA) (N=27ndash32 females and males per replicate see Table 3) with9 consolidated infrared beams per tube Flies were collected 1ndash2

days after emergence and entrained under LD 195 for 6 days afterwhich their free-running rhythmicity was measured in LL for 8 daysat 17degC (see Kauranen et al 2012)

Genome scans50 females of each selection and control replicate were frozen ingeneration F16 and stored at minus20degC DNA was extracted from10 females at a time and the extractions pooled into 6 samples of50 flies (one for each replicate of the selection and control line) DNAextractionwasmade using CTAB solution (Applichem BioChemicaGermany) with Proteinase K (Qiagen Germany) treatment followedby phenolndashchloroformndashisoamyl alcohol purification with RNAaseA(Qiagen Germany) treatment DNA concentration was checkedusing a Qubit spectrophotometer (Thermo Fisher Scientific MAUSA) and the purity with TapeStation (Agilent Technologies SantaClara CA USA) Extracted DNA samples were sequenced at theEdinburgh genomics (Scotland UK) facility using Illumina HiSeq Xplatform with 350 bp insert size and by 150 bp pair-end sequencingachieving a coverage of about 200times

Reads were trimmed to remove remaining adapter sequences andlow-quality bases using Trimmomatic v032 (Bolger et al 2014)Because of a general drop-off in DNA quality at the end of reads allreads were first cropped to 145 bp and then clipped from the leadingand trailing edges if base quality fell below 20 Reads were alsoclipped if the base quality fell below 20 in a sliding window of 5 bpand finally any reads shorter than 100 bp were removed Reads weremapped to the D montana reference genome (Parker et al 2018)using bwa mem (v077) (Li 2013) Samtools (v13) (Li et al2009) was used to keep only properly paired reads and reads with amapping quality greater than 30 (Schloumltterer et al 2014)Additionally regions around indels were locally re-aligned withPicard v 2141 (Broad Institute httpbroadinstitutegithubiopicard) and GATK (McKenna et al 2010 DePristo et al 2011)Finally PCR duplicates were removed with samtools

SNPs were called first with samtools mpileup and a newlydeveloped heuristic SNP caller PoolSNP (Kapun et al 2018preprint) The minimum base quality to consider a read to be usedfor base calling was 15 An allele was considered if the numberof reads supporting it was gt4 across all populations SNPs werecalled if the coverage was greater than 33 and less than the 95thpercentile of coverage (as calculated for each contig separately) In

Table 2 Number of the studied flies in each LD condition and inconstant darkness (DD) in NandandashHamner experiment

LD

DD126 1212 1224 1236 1272

Selection 1 250 130 250 359 293 1552 280 161 193 270 372 2233 248 221 239 314 368 341

Control 1 167 133 228 257 353 2692 371 160 388 198 159 3283 252 87 242 224 318 221

Table 3 Locomotor activity rhythm parameters for the base population and the selection and control line replicates

Replicate N

LD (195) LL

R () Period (h) Power R () Period (h) Power

Base population 57 100 239plusmn004 2055plusmn133 737 261plusmn04 530plusmn67

Selection 1 32 875 241plusmn01 1118plusmn111 719 262plusmn05 532plusmn54 32 938 241plusmn02 940plusmn93 563 256plusmn06 586plusmn94

2 22 864 240plusmn01 1659plusmn163 772 259plusmn06 1147plusmn18 25 88 238plusmn01 1556plusmn198 72 249plusmn03 93plusmn77


Control 1 32 719 236plusmn03 1418plusmn119 467 245plusmn10 705plusmn85 32 656 239plusmn02 821plusmn13 580 261plusmn07 94plusmn21



N number of individuals tested R percentage of rhythmic flies LD lightndashdark cycle used in entrained conditions LL constant light Period length of the intrinsicday of the flies in hours Power of periodogram test was defined as the amplitude of the peak only for the rhythmic flies from LombndashScargle periodogram withsignificance level Plt0 05 Values are meansplusmnsem

5


Journal

ofEx

perim

entalB

iology

total 4297404 SNPs were called for further analysis Furthermore1312706 SNPs could be placed on 4 autosomal linkage groupsand the X-chromosome using an available linkage map (Parkeret al 2018)

Statistical analysisAll statistical analyses were performed with R 351 (httpscranr-projectorg) For all regression type models mentioned below modelselection was done by assessing the best-fitting models with Akaikeinformation criterion using the AIC function and log-likelihoodtests using the anova function in R

Critical day length for diapause induction (CDL)For the analysis of the CDL responses the data were made binarywith values of 0 for diapausing female and 1 for non-diapausingfemale To determine the CDL for the base population and theselection and control line replicates the photoperiodic responsecurves (femalesrsquo diapause percentages under different LDs) wereanalyzed by a dose response model using the drm function from thedrc R package (Ritz et al 2015) This package fits a range of modelswith sigmoid or biphasic distributions and here we are predictingthe proportion of diapausing females as the response that occurs perchange (ie lsquodosersquo) in hours of light in the LD cycle The best-fittingmodel was chosen using a lack-of-fit test and AIC scores using themselect and modelFit functions (Ritz et al 2015) This lack-of-fittest uses an approximate F-test to compare dose-response modelswith different parameters to a more general ANOVA model to seewhich is the best fitting (Bates and Watts 1988) This model wasfitted with a three-parameterWeibull function (coded in the functionas lsquofct=W23rsquo in R) where upper limit of the model is set equal to 1

NandandashHamner experimentDifferences in the percentage of diapausing females in the selectionand control line replicates under different photoperiods (12 h light+612 24 36 or 72 h dark) or constant darkness (DD)was tested for eachLD or DD period with a generalised linear mixed model (GLMM)with a bimodal distribution and a logit link function Here selectionregime (selection or control) was used as a fixed factor and replicateas a random factor As the six LDDD cycles were tested separatelyP-values for comparisons between selection and control lines werecorrected using Bonferroni correction for multiple testing

Cold toleranceDifferences in the CTmin values of the flies were tested using a linearmixed model (LMM) and fitted using maximum likelihood (ML)and lmer function in R Selection regime (selection or controlline) LD (expressed as hours of light) and fly reproductive type(male diapausing female and non-diapausing female) and theirinteractions were used as fixed factors and replicate as a randomfactor For interpretation this model was fitted to a type II ANOVAwith P values computed with aWald test and interpretation aided byinteraction plots produced from the LMM model

Locomotor activity rhythmThe locomotor activity data were analyzed with the ActogramJprogram (Schmid et al 2011 available at httpsimagejnetActogramJ) to trace the rhythmicity of the flies and to determine thelength of the period (τ) ie the length of the flyrsquos intrinsic day in LLRhythmicity of the flies was determined from femalesrsquo daily activityrhythm profiles by displaying the data as double-plotted actograms(48 h plots) both under entraining (LD) and free-running conditions(LL) and using the LombndashScargle periodogram method with

a significance level of 005 The power of the LombndashScargleperiodogram analysis was defined as the amplitude of the peak onlyfor the rhythmic flies from LombndashScargle periodogram withsignificance level Plt005 Flies that did not survive throughoutthe experiment (both entrained and free-running conditions) wereexcluded from the analysis For more detailed information on thelocomotor activity analysis see Kauranen et al (2012)

The data were made binary with values of 0 for arrhythmic fliesand 1 for rhythmic flies to determine whether selection had affectedfliesrsquo ability to retain their activity rhythm in constant light (LL)A fly was determined to be rhythmic if the periodogram analysisdetected significant periodicity in its activity rhythm measuredacross several consecutive days (the significance level in the LombndashScargle periodogram analysis was adjusted to 005) Only the fliespossessing a clear entraining rhythm in 195 LD were used in theanalysis We analyzed the instance of rhythmicity with generalizedlinear models (GLM) with bimodal distribution and a logit linkfunction Here selection treatment (baseline selection or control)and sex were used as fixed factors As there were no significantdifferences between the models of control and selection line datawith or without replicates (likelihood ratio test df=1 χ2=122P=026) or sex (likelihood ratio test df=1 χ2=25 P=026) wedropped replicate and sex from the model and compared theselection and control lines to the base line directly To examinewhether selection had affected the length of the period of fliesrsquoactivity rhythm the difference in the length of the period betweenLD and LL treatment was first calculated for each fly The data werethen analyzed with a linear mixed model (LMM) using restrictedmaximum likelihood estimation (REML) where selection regime(selection or control) and sex were used as fixed factors andreplicate as a random factor Only flies possessing a clear entrainingrhythm in 195 LD were used in analysis

Identifying consistent allele frequency differences betweenselection and control lines and detecting annotated geneclusters and genes with divergent SNPsSNPs with a consistent allele frequency difference betweentreatment groups across replicates were identified with GLMsusing a quasibinomial error distribution (Wiberg et al 2017) Themodel structure was

y treatment thorn e eth1THORN

where y is read count of the alternative alleles in each samplelsquotreatmentrsquo is the experimental evolution treatment for each sampleand e is a quasibinomially distributed error term P-values wereconverted to q-values using the qvalue R package (Storey et al2015) A q-value cutoff of 005 was chosen to correct the FDR formultiple testing The closest genes to each SNP were identified byusing the closestBed tool from BEDtools (v2270) (Quinlan andHall 2010) and theD montana reference genome annotation (Parkeret al 2018) A SNP was associated with a gene if it lay within 10 kbup- or downstream of the gene start and end coordinates

We also performed a transcription factor (TF) binding motifenrichment analysis using the AME tool (with default options) inthe MEME suite and the Fly Factor Survey (httpmccbumassmededuffs) database of TF binding motifs (Bailey et al 2009 Buskeet al 2010) We used the regions 30 bp up- and downstream of topSNPs as input

Finallywe ran functional annotation clusteringwithDAVID (v68)(Huang et al 2009ab) using the Drosophila virilis annotation as abackground set Additionally phenotypic enrichment analysis was

6


Journal

ofEx

perim

entalB

iology

performed with DroPhEA (Weng and Liao 2011) to test forenrichment of known phenotypes within the phenotype levels 2through 7

RESULTSQuasinatural selection for reproduction under short-dayconditions reveals genetic variation and plasticity in CDLThe percentage of diapausing females varied over the courseof selection between 22 and 53 in selection line replicates(it only remained below this in generation F8) while in controlline replicates respective percentages varied between 0 and 8Selection was stopped in generation F15 when more than 50 of thefemales of each replicate entered diapause under 159 LD CDLs ofthe base population and the selection and control line replicates weredetermined from the photoperiodic response curves as the pointwhere 50 of the females enter diapause (Fig 2) In the basepopulation CDL was 19447 LD and two generations after theend of selection (generation F17) the mean CDL over the threeselection line replicates was 17565 LD and the three control linereplicates 18753 LD CDL of the selection line was significantlyshorter than those of the base population (182 h Est=122se=006 t=2045 Plt0001) or the control line (127 h Est=119se=003 t=3609 Plt0001) while the difference between the basepopulation and the control line was not significant (055 hEst=003 se=005 t=066 P=0509) Selection had no effect onfemale diapause propensity under short-day conditions (day lengthsof 16 h or shorter) where the diapause percentages were close to100 in both linesPhotoperiodic response curves and CDLs of the control and

selection line replicates were measured again after 8ndash10 months ofrelaxed selection prior to performing NndashH experiments Here theCDLs of the three selection line replicates were 17565 17466and 177 LD and for the three control line replicates 20535

20238 and 20238 LD with a strongly significant difference inthe mean dose response of two lines (Est=262 se=007 t=367Plt0001)

Selection decreases female diapause propensity underexotic photoperiods towards the southern phenotypeNndashH experiments revealed a clear connection between CDL andfemalesrsquo diapause frequency under the photoperiods that deviate from24 h cycles femalesrsquo tendency to enter diapause was significantlylower in selection line replicates (short CDL) than in control linereplicates (long CDL) in all unnatural LDs andDD (Plt0001) but notin 1212 LD conditions (Table 4 Fig 3) The difference wasparticularly large in the shortest scotophase (126 LD) where CNL(6 h) was shorter than the CNL of the selection line replicates incritical photoperiod (CDL 177 CNL 63 h) D montana femaleswith a long northern-type CDL have also been found to enterdiapause under photoperiods that deviate from 24 h cycles in NndashHexperiments more frequently than those having a short southern-typeCDL in wild clinal populations (PL unpublished data)

Cold tolerance is affected by photoperiod reproductive typeand marginally by selection regimeCold tolerance of the selection and control line males and thenon-diapausing females of selection line increased (CTmin valuedecreased) towards shorter day lengths while the diapausingfemales of both lines showed no changes in their cold tolerance andthe non-diapausing females of control line even showed a slightdecrease in this trait (Fig 4) Control line males had about 05degClower CTmin values than the selection line males in all LDs and thecold tolerance of diapausing females in all LDs was as low as that ofthe males of the same line under the shortest photoperiod (Fig 4)The replicates explained only 377 of variance in fliesrsquo cold-tolerance Changes in CTmin values were significant across both of

1410 168 186

LD cycle

204 1410 168 186 204

02

0

Pro

porti

on o

f dia

paus

ing

fem

ales

04 Estimated CDL

06

08

10 Base populationReplicatesAverage

0

02

04

06

08

10

A B

Fig 2 Photoperiodic response curves (PPRCs) for the base population (N=59ndash95) and for theDmontana selection and control line replicates and theirmeans in generation F17 (N=69ndash98 per replicate) Lines are dose-response modeled predictions Exact sample sizes for each time pointreplicate are shown inTable 1 Arrows indicate the estimated CDLs of base population (19447 LD) and the average critical day lengths (CDL) of selection (A 17565 LD) and controllines (B 18753 LD)

7


Journal

ofEx

perim

entalB

iology

the LD conditions and between the fly reproductive types (malesand non-diapausing and diapausing females Table 5) while thedifference between the selection conditions remained on theborderline of significance (Table 5) Furthermore there werestrongly significant interactions between fly reproductive type andLD and selection regime but the three-way interaction was notsignificant (Table 5) Photoperiodic cold acclimation was most clearin males where the cold tolerance increased by an estimated 015degCper 1 h decrease in photoperiod

Selection increases the proportion of rhythmic flies but hasno effect on fliesrsquo free-running rhythmThe locomotor activity rhythm of the flies was studied under bothentraining (195 LD) and constant light (LL) conditions the latterrepresenting fliesrsquo free-running rhythmicity In 195 LD all basepopulation females were rhythmic (males were not studied) whilein selection and control line replicates the percentage of rhythmicflies varied 864ndash100 and 639ndash867 respectively (Table 3) InLL the selection line flies showed higher free-running rhythmicitythan the control line flies (GLM Est= 062 se=028 z= 222P=0027) while the differences between the base population andthe selection (GLM Est=030 se=032 z=09 P=0367) or thecontrol line (GLM Est=minus025 se=032 z=minus075 P=0452) werenot significant (see Table 3)

Rhythmic flies possessed a clear evening-activity peak in 195LD and most flies also maintained this peak in LL However thelength of the free-running period (τ) in LL did not show significantdifferences between the selection and control lines (LMMEst=minus753 se=2439 t=minus0309 P=077) or between the sexes(LMM Est=1141 se=2354 t=048 P=063)

Selection led to genetic divergence throughout the genomeSelection and control line replicates showed significant differences(q value lt005) in allele frequencies of 4496 SNPs (hereafter calledlsquodivergent SNPsrsquo) 1445 of which could be localized to the mainchromosome arms of D montana using the existing anchoredgenome A Manhattan plot of these SNPs (Fig 5) indicates thatinstead of being randomly distributed across the main chromosomesthey are enriched on chromosomes 3 and 4 (Table S1 χ2=63552df=4 Plt001) This pattern holds when using the linkage grouplength as a proportion of the total genome length or the proportion ofSNPs out of all SNPs as the expected proportions of divergent SNPsThus there seems to be an excess of divergent SNPs on the 3rd and4th chromosomes (Table S1)

Allele frequencies across the control and selection lines in thetwo treatments showed a combination of large allele frequencydifferences as well as smaller differences with a very consistenteffect between the lines The regions around top SNPs wereenriched for a single TF binding motif Adult enhancer factor 1(Aef1) Additionally top SNPs lay within 10 kb of 1756 (Table S2)D montana gene models which have an identifiable ortholog inD virilis These were tested for functional annotation clusteringwith DAVID (v68) (Huang et al 2009ab) using the D virilisannotation as a background set This found 100 functional clusters16 of which were significantly enriched (Table S3) Most of theseclusters consisted of factors linked with signaling cluster 1 involvedfactors linked with membranetransmembrane changes clusters 411 and 14 with ion transport clusters 9 10 and 16 with signaltransduction and clusters 3 6 and 8 with neuropeptideneurotransmitter signaling The remaining clusters involvedfactors affecting mainly embryogenesis differentiation andgrowth cluster 2 involved immunoglobulins cluster 5 epidermalgrowth factor cluster 7 homeobox clusters 12 and 15 protein kinaseand tyrosine activity and cluster 13 transcription regulation

Table 4 General linear mixed model results for NandandashHamnerexperiment differences in the percentage of diapausing females in theselection and control line replicates under different photoperiods orconstant darkness (DD)

χ2 df P-value

LD 126 9140 1 lt0001LD 1212 0 1 1LD 1224 1442 1 lt0001LD 1236 1975 1 lt0001LD 1272 4323800 1 lt0001DD 8817 1 lt0001

Selection regimes (selection or control) were used as fixed factors andreplicate as a random factor P-values for comparisons between control andselected lines for the five LD and one DD cycles were corrected usingBonferroni P values correction Residuals df=3 Significant P-values areshown in bold

126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD

Fig 3 The percentage of diapausing females in selection and control lines in different photoperiodsConditions were 12 h photophases followed by 6 1224 36 or 72 h scotophases as well as total darkness (DD) Curves follow the mean value of the three replicates of each line The number of flies per LD cyclereplicate can be found in Table 2

8


Journal

ofEx

perim

entalB

iology

Genes located within 10 kb of significant SNPs are listed inTable S4 and the list of most-divergent 100 genes in Table S5Comparing the gene list with genes known to be involvedin photoreceptionsignal transduction circadian rhythms andneurohormonal control of reproductive diapause revealed severalSNP-associated genes playing a role at different steps of diapauseinduction and CDL determination A large number of genesassociated with diverged SNPs were involved in photoreception(pinta Cpn w st Arr1 crb cry and Pld) andor ion channelfunction (TrapA1 Trpm Trpγ norpA stops Shab brv3 dysc andslob) Several genes were also involved in time measurement andactivity rhythms through the circadian clock system (eg Shab dysccry sgg nf1 Fmr1 dh31 Lkr SIFaR OambDATDopR1DopR2HmgrAstA-R2 and Inr) Furthermore at least 12 of the genes were Gprotein-coupled receptors that are active in neuropeptide signalingandor neurotransmitter (mainly octopamine and dopamine)synthesis or reception (Oamb Octbeta3R Ple Ddc DAT DopR1DopR2 Hmgrc AstA-R2 Inr Pdk1 Ide sNPF-R slob GABA-B-R1 GABA-B-R2 SIFaR and Tor) Several of the latter genesas well as Inr Pi3K59F jhamt dpp and Jheh1 are known to affectfly development and diapause through insulin signaling and

20-hydroxyecdysone and on juvenile hormone metabolism Moreinformation on the function of these as well as some of the keyreferences is given in Table S6

DISCUSSIONThe correct timing of reproductive diapause increases femalesurvival rate during the unfavorable season and their progenyproduction in the following warm period and thus it is highlyadaptive especially in temperate regions (Hahn and Denlinger2007) Evolution of longer CDLs at higher latitudes has beensuggested to result from selection from longer summer days (photicenvironment) on the circadian clock system (Pittendrigh andTakamura 1987) or directly on the critical photoperiodCDL(Bradshaw et al 2003) In our study divergence of CDL and otherdiapause-associated traits between selection and control linesoccurred without disrupting circadian rhythm in fly locomotoractivity which gives support to the latter view

Selection changes traits associated with reproductivediapause without altering circadian rhythm in fly locomotoractivityDiapause propensity and CDL have been shown to exhibit highgenetic variation within populations and thus these traits can beexpected to be rapidly altered by selection (Tauber et al 1986Lankinen et al 2013) Our earlier studies have revealed about 2 hdifference between the CDLs of the most northern (67degN) andsouthern (62degN)D montana populations in Finland as well as highvariation (up to 3 h) within the Oulanka (664degN) population used inthe present study (Tyukmaeva et al 2011 Lankinen et al 2013)CDL of selection line replicates decreased by 118 h on averagecompared with control line replicates over 9 generations ofselection which corresponds to 25ndash5 deg change on a latitudinalscale (Danilevsky et al 1970 Tyukmaeva et al 2011) It also

ndash10

Males

SelectionControl

Diapausing females Non-diapausing females

ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C

Fig 4 Interaction plot for the linear mixed model comparing how selection regime fly reproductive type and LD affect cold tolerance Selection versuscontrol results in (A) male (B) diapausing female and (C) non-diapausing female flies Lines show predicted mean CTmin values measured in LL and LD 195177 159 and 1311 and error bars show modeled se

Table 5 ANOVA results from LMM of CTmin

Value χ2 df P-value

LD 4208 1 lt0001Selection regime 377 1 0052Reproductive type 1542 2 lt0001LDtimesSelection regime 018 1 0675LDtimesReproductive type 2776 2 lt0001Selection regimetimesReproductive type 814 2 0017Selection regimetimesReproductive typetimesLD 218 2 0336

Significant P-values are shown in bold Residual df=964 Reproductive typeincludes males and non-diapausing and diapausing females

9


Journal

ofEx

perim

entalB

iology

suggests that at the fliesrsquo home site (664degN) the base populationand control line females would enter diapause about 2 weeks earlierthan the selection line females (July 31st and August 15threspectively) Similar shifts towards shorter (southern type)critical photoperiod have been detected in the wild as a responseto climate warming for example in Wyeomyia smithii mosquitoes(Bradshaw and Holzapfel 2001) and Phratora vulgatissima leafbeetles (Dalin 2011)CDLs of the selection and control line replicates were measured

for a second time after maintaining the flies in mixed generations inLD 159 (selection line) and LL (control line) at 16degC for 4 to 5generations Here the mean CDL of the selection line replicates(176) was about the same as that measured in generation F17 (177)while that of the control line replicates increased from 189to 205 h The second set of CDLs was measured at a temperaturesim1degC lower than the first set and thus the two measurements arenot strictly comparable (CDL of D montana becomes longerin lower temperatures V Tyukmaeva personal communication)A plausible explanation for why the effect of decreased temperatureis evident in the CDL of the control line but not in that of theselection line is that in the latter line the temperature effect has beencompensated by continued selection for shorter CDL in LD 159Oikarinen and Lumme (1979) have previously succeeded inshortening CDL of Drosophila littoralis females by 11ndash18 h bymaintaining the flies for seven generations in LD 186 which showsthat this kind of selection can be quite effective

In the NndashH experiment circadian clock oscillations are suggestedto be involved in the photoperiodic time measurement if the insectsrsquoshort- or long-day responses show amplitude peaks under LD cycleswhose period (T ) is close to 24 h and its multiples (Nanda andHamner 1958) We have previously shown thatD montana femalesmeasure night rather than day length for diapause timing and thattheir diapause frequency shows a peak only when Twas close to 24 h(12 h light+12 h dark Kauranen et al 2013) In the present studyselection line females showed a clear diapause peak in LD 1212(T=24 h) while the control line females with a long lsquonorthernrsquo CDLentered diapause at a high frequency under all photoperiods studiedAll these findings can be explained by the damping version of thecircadian external coincidence model (Lewis and Saunders 1987) ifthe damping of circadian oscillator(s) increases towards North (notethat in the present study free-running locomotor activity rhythmdampened more quickly in the lsquonorthern typersquo control than in thelsquosouthern typersquo selection line) In this model females are expected toenter diapausewhen the night length exceeds their CNL (24 h CDL)which could also explain why the diapause frequency of theselection line females (CNL 65) but not that of the control linefemales (CNL 53) dropped drastically in LD 126 (CNLs arecalculated from the CDLs measured in Oulu University during thesame time period and temperature as the NndashH experiments) Linedifferences in diapause frequency in DD resembles a latitudinal clinedetected in the diapause frequency of D littoralis females in thiscondition (Lumme and Oikarinen 1977)

20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125

Fig 5 Manhattan plot of ndashlog10(q-values) for the SNPs on themain chromosomal arms ofDmontana selection and control line replicates Significantlydivergent SNPs are shown in red The horizontal dashed red line gives the 005 q-value threshold

10


Journal

ofEx

perim

entalB

iology

D montana flies have been found to show some variation in theircold tolerance between populations from different latitudes andaltitudes (Vesala and Hoikkala 2011 RAWW unpublisheddata) as well as notable photoperiodic cold acclimation towards thecold season (Vesala et al 2012) In the present study selectiondecreased the cold tolerance of males and diapausing females in allphotoperiods examined while the results for non-diapausingfemales were less clear Male cold tolerance increased linearlytowards shorter photoperiods while that of the diapausing femalesshowed practically no changes under different LDs which suggeststhat female cold tolerance is phenotypically associated withdiapause while male cold tolerance shows clearer photoperiodiccold acclimationD montanamales are known to enter reproductivediapause (become reproductively inactive) under the same daylengths that induce diapause in females (Ala-Honkola et al 2018)but the present study shows that this does not increase their coldtolerance Vesala and Hoikkala (2011) have previously shown thatthe strength of association between female diapause and coldtolerance varies between D montana populations and Vesala et al(2012) have demonstrated that the females of a D montana strainwhich lacks diapause are less cold tolerant than those of the otherstrains of this species Furthermore Teets and Denlinger (2013)have argued that seasonal cold adaptation can be a component ofoverwintering diapause but the two processes can also occurindependently and Pegoraro et al (2014) have suggested thatdifferences between diapause and cold tolerance phenotypes mayrepresent two separate photoperiodic circuits that use differentgenetic networksDecrease in CDL and other diapause-associated traits in the

selection lines was not accompanied by changes in the period of flyfree-running locomotor activity rhythm which shows that the traitsassociated with reproductive diapause can occur without disturbingmechanisms of the circadian clock Another species of the D virilisgroup D littoralis has been found to show correlated latitudinalvariation in CDL and fliesrsquo circadian eclosion rhythm (Lankinen1986b) but in this species the traits also diverged in a selectionexperiment (Lankinen and Forsman 2006) Lack of correlationbetween the circadian clock-regulated traits (pupal eclosion egghatch and oviposition rhythms) and photoperiodic responses hasbeen verified also in the pink boll worm moth Pectinophoragossypiella (Pittendrigh and Minis 1971) and between circadianeclosion rhythm and CDL inDrosophila auraria (Pittendrigh et al1984 Pittendrigh and Takamura 1987) Furthermore Bradshawet al (2003) succeeded in changing both the critical photoperiodand the amplitude but not the period of diapause responserhythmicity of W smithii mosquitoes under extended days in anNndashH selection experiment

What do genome analyses tell us about the divergencebetween the control and selection linesExperimental evolution combined with genome sequencing hasproved to be an effective approach to study the connection betweenphenotypic responses and genetic changes during selection (Tobleret al 2013 Schloumltterer et al 2014 Graves et al 2017) In total4496 SNPs show a significant difference between the control anddiapause selection lines 1445 of which could be placed on linkagegroups There was a significant deviation from a randomdistribution of SNPs It seems likely that the number of SNPsinvolved could be overestimated because of linkage and hitchhikingeffects Unfortunately little is known of the recombination map inD montana In a functional enrichment analysis we detected16 functional gene clusters that diverged between selection and

control line replicates 12 of these clusters contained factorsinvolved in signal transduction including membranes ionchannels and neuropeptideneurotransmitter signaling Changesin the function of genes within these clusters could affect signaltransduction at any level from photoreception and the passage ofvisual and other signals in fly brain to the regulation of varioushormones through neurotransmitters The remaining 4 clustersinvolved factors affecting embryogenesis differentiation andgrowth Notably many of above-mentioned gene clustersespecially those in the first group of clusters have been shown toalso differ between wild populations of this species (Parker et al2018 RAWW et al unpublished data)

A transcription factor motif enrichment analysis found that amotif of the transcription factor Adult enhancer factor 1 (Aef1) wasover-represented in the regions surrounding top SNPs InD melanogaster Aef1 is a transcription factor that is a repressorof Adh in the adult fat body (Falb and Maniatis 1992) Aef1 is alsoan inhibitor of yolk protein 1 and 2 in males (Tarone et al 2012) andseems to influence olfactory behavior (Tunstall et al 2012)

Recent studies in insects have discovered genes and genenetworks involved in different steps of diapause inductionincluding photoreception and transduction (Fowler and Montell2013 Kistenpfenning et al 2018) circadian regulation (Zhanget al 2010 Dubowy and Sehgal 2017 Liang et al 2017)and neurohormonal control of reproductive diapause (Naumlssel andWinther 2010 Diniz et al 2017) Furthermore Bryon et al(2013) and Zhao et al (2017) detected upregulation of severalgenes encoding G protein-coupled receptors especially those foroctopamine neuropeptide F proctolin and tachykinin during earlydiapause in the two-spotted spider mite T urticae In our studyseveral genes associated with SNPs diverging between selection andcontrol line replicates were involved in photoreception ion channelfunction andor in the function and output of the circadian clocksystem The list of genes also included G protein-coupled receptorsthat function in neuropeptide signaling andor neurotransmittersynthesis or reception Finally several genes functioned in insulinsignaling and on 20-hydroxyecdysone and juvenile hormonepathways that play a central role in diapause induction Moreinformation on the function of implicated genes as well as keyreferences are provided in Table S5 Of course further studies arerequired to examine the functional involvement of any of these genesin photoperiodic adaptation but our results confirm that changes inCDL involve extensive genome-wide changes in genes involvedin signaling

ConclusionsPhotoperiodic reproductive diapause has evolved multiple timesin arthropod species including within the genus Drosophila(Hand et al 2016) and the interaction between PTM and circadianclock also seems to vary between species (Meuti and Denlinger2013) D montana is a northern Drosophila species with a robustreproductive diapause unimodal daily activity rhythm and an abilityto maintain free-running locomotor activity rhythm in constant lightbut not in constant darkness (Kauranen et al 2012) This and otherD virilis group species also differ from a more southern speciesD melanogaster at the neuronal level for example in the expressionpattern of the blue light photopigment cryptochrome (CRY) and theneuropeptide pigment-dispersing factor (PDF) (Kauranen et al2012 Hermann et al 2013) D virilis group species (includingD montana) probably lost CRY and PDF expression in their long(l-LNvs) and small (s-LNvs) ventrolateral clock neuronsrespectively but gained PDF expression in their central brain

11


Journal

ofEx

perim

entalB

iology

under selection from environments subject to major photoperiodicchanges throughout the year with extremely long photoperiods insummer (Menegazzi et al 2017 Beauchamp et al 2018) Beer et al(2017) give Acyrthosiphon pisum aphids and northern D virilisgroup species D montana Drosophila ezoana and D littoralis asexamples of the species exhibiting weak circadian rhythms androbust seasonal responses to shortening photoperiod Weak clockshave been suggested to be generally more plastic than strong clockswhich may partly explain why they can easily adapt to extremechanges in day length that prevail in the North (Abraham et al2010) Even though anticipating daily reoccurring events is anadvantage an oscillator that drives strong circadian rhythms over along time may be potentially less flexible in resetting to seasonalchanges (Beer et al 2017)Short selection experiments in the laboratory differ in many

fundamental ways from the long-term selection occurring in thewild but nevertheless our study revealed important aspects of lifehistory evolution First it showed that adaptive divergence of CDLand other diapause-related traits can occur without disturbingcircadian clock-regulated locomotor activity rhythm and that CDLalso shows plasticity for reproduction under short-day conditionsSecond resequencing the selection and control line replicatesrevealed widespread genomic divergence including significantdivergence in SNPs located in or near several genes important forphotoreception circadian rhythms signal transduction aminergicsignaling and hormone secretion It would be interesting to findout how these genes function under different LDs and temperaturesfor example through alternative splicing and whether they showlatitudinal variation in D montana populations Such informationwould help us to trace the steps leading to changes in CDL and otherphotoperiodically regulated traits in the wild and to understand howadaptation to new or changing environmental conditions occurs at thegenetic level As Bergland et al (2014) have stated environmentalheterogeneity can promote the long-term persistence of functionalpolymorphisms within populations that can fuel a fast directionaladaptive response in the future

AcknowledgementsGenome sequencing was performed at Edinburgh genomics (Scotland UK) We aregrateful to Professor David S Saunders for his valuable comments on an earlierversion of this manuscript

Competing interestsThe authors declare no competing or financial interests

Author contributionsConceptualization MGR AH Methodology HK JK A-LH PL RAWWMGR AH Software DH RAWW MGR Validation HK RAWWMGR Formal analysis DH RAWW MGR AH Investigation HK JKA-LH PL RAWW AH Resources HK JK A-LH PL AH Datacuration HK DH RAWW MGR AH Writing - original draft HK AHWriting - review amp editing HK DH RAWW MGR AH Visualization HKDH RAWW AH Supervision HK AH Project administration HK AHFunding acquisition AH

FundingThework has been supported by the Academyof Finland toAH (project 267244) andNatural Environment Research Council (NERC) funding (NEJ0208181 to MGRNEL5018521 to RAWW)

Data availabilityRaw reads have been submitted with the NCBI short read archive under BioProjectID PRJNA575616

Supplementary informationSupplementary information available online athttpjebbiologistsorglookupdoi101242jeb205831supplemental

ReferencesAbraham U Granada A E Westermark P O Heine M Kramer A and

Herzel H (2010) Coupling governs entrainment range of circadian clocks MolSyst Biol 6 438 doi101038msb201092

Agrawal A A (2000) Host-range evolution adaptation and trade-offs in fitness ofmites on alternative hosts Ecology 81 500-508 doi1018900012-9658(2000)081[0500HREAAT]20CO2

Ala-Honkola O Kauranen H Tyukmaeva V Boetz F A Hoikkala ASchmitt T (2018) Diapause affects cuticular hydrocarbon composition andmating behavior of both sexes in Drosophila montana Insect Sci doi1011111744-791712639

Allemand R and David J R (1976) The circadian rhythm of oviposition inDrosophila melanogaster a genetic latitudinal cline in wild populationsExperientia 32 1403-1405 doi101007BF01937401

Andreatta G Kyriacou C P Flatt T and Costa R (2018) Aminergic signalingcontrols ovarian dormancy in Drosophila Sci Rep 8 2030 doi101038s41598-018-20407-z

Bailey T L Boden M Buske F A Frith M Grant C E Clement L Ren JLi W W and Noble W S (2009) MEME Suite tools for motif discovery andsearching Nucleic Acids Res 37 202-208 doi101093nargkp335

Bates D M and Watts D G (1988) Nonlinear Regression Analysis and itsApplications New York USA Wiley amp Sons

Beauchamp M Bertolini E Deppisch P Steubing J Menegazzi P andHelfrich-Forster C (2018) Closely related fruit fly species living at differentlatitudes diverge in their circadian clock anatomy and rhythmic behavior J BiolRhythms 33 602-613 doi1011770748730418798096

Beer K Joschinski J Sastre1 A A Krauss J and Helfrich-Forster C(2017) A damping circadian clock drives weak oscillations in metabolism andlocomotor activity of aphids (Acyrthosiphon pisum) Sci Rep 7 14906 doi101038s41598-017-15014-3

Bell-Pedersen D Cassone V M Earnest D J Golden S S Hardin P EThomas T L and Zoran M J (2005) Circadian rhythms from multipleoscillators lessons from diverse organisms Nat Rev Genet 6 544-556 doi101038nrg1633

Bergland A O Behrman E L OrsquoBrien K R Schmidt P S and Petrov D A(2014) Genomic evidence of rapid and stable adaptive oscillations over seasonaltime scales in Drosophila PLoS Genet 10 e1004775 doi101371journalpgen1004775

Bolger A M Lohse M andUsadel B (2014) Trimmomatic a flexible trimmer forIllumina sequence data Bioinformatics 30 2114-2120 doi101093bioinformaticsbtu170

Bozicevic V Hutter S Stephan W and Wollstein A (2016) Populationgenetic evidence for cold adaptation in European Drosophila melanogasterpopulations Mol Ecol 25 1175-1191 doi101111mec13464

Bradshaw W E and Holzapfel C M (2001) Genetic shift in photoperiodicresponse correlated with global warming Proc Natl Acad Sci USA 9814509-14511 doi101073pnas241391498

BradshawW E andHolzapfel C M (2010) What season is it anyway Circadiantracking vs photoperiodic anticipation in insects J Biol Rhythms 25 155-165doi1011770748730410365656

Bradshaw W E and Lounibos L P (1977) Evolution of dormancy and itsphotoperiodic control in pitcher plant mosquitoes Evolution 31 546-567 doi101111j1558-56461977tb01044x

Bradshaw W E Quebodeaux M C and Holzapfel C M (2003) Circadianrhythmicity and photoperiodism in the pitcher-plant mosquito adaptive responseto the photic environment or correlated response to the seasonal environmentAm Nat 161 735-748 doi101086374344

Bryon A Wybouw N Dermauw W Tirry L and Van Leeuwen T (2013)Genome wide gene-expression analysis of facultative reproductive diapause inthe two-spotted spider mite Tetranychus urticae BMC Genomics 14 815-834doi1011861471-2164-14-815

Bunning E (1936) Die endogene tagesrhythmik als grundlage derphotoperiodischen reaction Ber Dtsch Bot Ges 54 590-607

Burke M K (2012) How does adaptation sweep through the genome Insightsfrom long-term selection experiments Proc R Sci B 279 5029-5038 doi101098rspb20120799

Buske F A Boden M Bauer D C and Bailey T L (2010) Assigning roles toDNA regulatory motifs using comparative genomics Bioinformatics 26 860-866doi101093bioinformaticsbtq049

Dalin P (2011) Diapause induction and termination in a commonly univoltine leafbeetle (Phratora vulgatissima) Insect Sci 18 443-450 doi101111j1744-7917201101417x

Danilevsky A S Goryshin N I and Tyschenkov V P (1970) Biologicalrhythms in arthropods Annu Rev Entomol 15 201-244 doi101146annureven15010170001221

Denlinger D L Hahn D A Merlin C Holzapfel C M and Bradshaw W E(2017) Keeping time without a spine what can the insect clock teach us aboutseasonal adaptation Phil Trans R Soc B 372 20160257 doi101098rstb20160257

12


Journal

ofEx

perim

entalB

iology

DePristo M A Banks E Poplin R Garimella K V Maguire J R Hartl CPhilippakis A A del Angel G Rivas M A Hanna M et al (2011) Aframework for variation discovery and genotyping using next-generation DNAsequencing data Nature Genet 43 491-498 doi101038ng806

Diniz D F A Ribeiro de Albuquerque C M Oliva L O Varjal de Melo-Santos M A and Ayres C F A (2017) Diapause and quiescence dormancymechanisms that contribute to the geographical expansion of mosquitoes andtheir evolutionary success Parasit Vectors 10 310 doi101186s13071-017-2235-0

Dubowy C and Sehgal A (2017) Circadian rhythms and sleep in Drosophilamelanogaster Genetics 205 1373-1397 doi101534genetics115185157

Engelmann W and Mack J (1978) Different oscillators control the circadianrhythm of eclosion and activity in Drosophila J Comp Physiol 127 229-237doi101007BF01350113

Falb D and Maniatis T (1992) Drosophila transcriptional regulation repressorprotein that binds specifically to negative control elements in fat body enhancersMol Cell Biol 12 4093-4103 doi101128MCB1294093

Fowler M A and Montell C (2013) Drosophila TRP channels and animalbehavior Life Sci 92 394-403 doi101016jlfs201207029

Fry J D (2003) Detecting ecological trade-offs using selection experimentsEcology 84 1672-1678 doi1018900012-9658(2003)084[1672DETUSE]20CO2

Goto S G (2013) Roles of circadian clock genes in insect photoperiodismEntomol Sci 16 1-16 doi101111ens12000

Graves J L Hertweck K L Phillips M A Han M V Gabral L G BarterT T Greer L F Burke M K Mueller L D andRose M R (2017) Genomicsof parallel experimental evolution in Drosophila Mol Biol Evol 34 831-842doi101093molbevmsw282

Hahn D A and Denlinger D L (2007) Meeting the energetic demands of insectdiapause nutrient storage and utilization J Insect Physiol 53 760-773 doi101016jjinsphys200703018

Hand S C Denlinger D L Podrabsky J E andRoy R (2016) Mechanisms ofanimal diapause recent developments from nematodes crustaceans insectsand fish Am J Physiol 11 R1193-R1211 doi101152ajpregu002502015

Hermann C Saccon R Senthilan P R Domnik L Dircksen H Yoshii Tand Helfrich-Forster C (2013) The circadian clock network in the brain ofdifferent Drosophila species J Comp Neurol 521 367-388 doi101002cne23178

Huang D W Sherman B T and Lempicki R A (2009a) Systematic andintegrative analysis of large gene lists using DAVID bioinformatics resources NatProtoc 4 44-57 doi101038nprot2008211

Huang D W Sherman B T and Lempicki R A (2009b) Bioinformaticsenrichment tools paths toward the comprehensive functional analysis of largegene lists Nucleic Acids Res 37 1-13 doi101093nargkn923

Hut R A Paolucci S Dor R Kyriacou C P and Daan S (2013) Latitudinalclines an evolutionary view on biological rhythms Proc R Soc B 28020130433 doi101098rspb20130433

Kapun M Aduriz M G B Staubach F Vieira J Obbard D Coubert CStabelli O R Kankare M Haudry AWiberg R AW et al (2018) Genomicanalysis of European Drosophila melanogaster populations on a dense spatialscale reveals longitudinal population structure and continent-wide selectionbioRxiv doi101101313759

Kauranen H Menegazzi P Costa R Helfrich-Forster C Kankainen A andHoikkala A (2012) Flies in the north locomotor behavior and clock neuronorganization of Drosophila montana J Biol Rhythms 27 377-387 doi1011770748730412455916

Kauranen H Tyukmaeva V and Hoikkala A (2013) Involvement of circadianoscillation(s) in the photoperiodic time measurement and the induction ofreproductive diapause in a northern Drosophila species J Insect Physiol 59662-666 doi101016jjinsphys201304007

Kauranen H Kinnunen J Hopkins D and Hoikkala A (2018) Direct andcorrelated responses to bi-directional selection on pre-adult development time inDrosophila montana J Insect Physiol 116 77-89 doi101016jjinsphys201904004

Kellermann V Loeschcke V Hoffmann A A Kristensen T N Flojgaard CDavid J R Svenning J-C and Overgaard J (2012) Phylogeneticconstraints in key functional traits behind speciesrsquo climate niches patterns ofdesiccation and cold resistance across 95 Drosophila species Evolution 663377-3389 doi101111j1558-5646201201685x

King A N and Sehgal A (2018) Molecular and circuit mechanisms mediatingcircadian clock output in the Drosophila brain Eur J Neurosci doi101111ejn14092

King A N Barber A F Smith A E Dreyer A P Sitaraman D NitabachM N Cavanaugh D J and Sehgal A (2017) A Peptidergic circuit links thecircadian clock to locomotor activity Curr Biol 27 1915-1927 doi101016jcub201705089

Kistenpfenning C Nakayama M Nihara R Tomioka K Helfrich-Forster Cand Yoshii T (2018) ATug-of-war between cryptochrome and the visual systemallows the adaptation of evening activity to long photoperiods in Drosophilamelanogaster J Biol Rhythms 33 24-34 doi1011770748730417738612

Kubrak O I Kucerova L Theopold U and Nassel D R (2014) The sleepingbeauty how reproductive diapause affects hormone signaling metabolismimmune response and somatic maintenance in Drosophila melanogaster PLoSONE 9 e113051 doi101371journalpone0113051

Lakovaara S (1969) Malt as a culture medium for Drosophila species DrosophilaInformation Service 44 128

Lankinen P (1986a) Geographical variation in circadian eclosion rhythm andphotoperiodic adult diapause in Drosophila littoralis J Comp Physiol 159123-142 doi101007BF00612503

Lankinen P (1986b) Genetic correlation between circadian eclosion rhythm andphotoperiodic diapause in Drosophila littoralis J Biol Rhythms 1 101-118doi101177074873048600100202

Lankinen P and Forsman P (2006) Independence of genetic geographicalvariation between photoperiodic diapause circadian eclosion rhythm and Thr-Glyrepeat region of the period gene inDrosophila littoralis J Biol Rhythms 21 3-12doi1011770748730405283418

Lankinen P Tyukmaeva V I and Hoikkala A (2013) Northern Drosophilamontana flies show variation both within and between cline populations in thecritical day length evoking reproductive diapause J Insect Phys 59 745-751doi101016jjinsphys201305006

Lees A D (1973) Photoperiodic time measurement in the aphid Megoura viciaeJ Insect Physiol 19 2279-2316 doi1010160022-1910(73)90237-0

Lees A D (1986) Some effects of temperature on the hour glass photoperiod timerin the aphid Megoura viciae J Insect Physiol 32 79-89 doi1010160022-1910(86)90160-5

Li H (2013) Aligning sequence reads clone sequences and assembly contigs withBWA-MEM arXiv 13033997v1 [q-bioGN] httpsarxivorgabs13033997

Li H Handsaker B Wysoker A Fennell T Ruan J Homer N Marth GAbecasis G Durbin R and 1000 Genome Project Data ProcessingSubgroup (2009) The sequence alignmentmap format and SAMtoolsBioinformatics 25 2078-2079 doi101093bioinformaticsbtp352

Liang X Holy T E and Taghert P H (2017) Series of suppressive signalswithin the Drosophila circadian neural circuit generates sequential daily outputsNeuron 94 1173-1189e4 doi101016jneuron201705007

Lewis R D and Saunders D S (1987) A damped circadian oscillator model of aninsect photoperiodic clock I Description of themodel based on a feedback controlsystem J Theor Biol 1 47-59 doi101016S0022-5193(87)80030-9

Lumme J and Oikarinen A (1977) The genetic basis of the geographicallyvariable photoperiodic diapause in Drosophila littoralis Hereditas 86 129-142doi101111j1601-52231977tb01221x

Martins N E Faria V G Nolte V Schlotterer C Teixeira L Sucena E andMagalhaes S (2014) Host adaptation to viruses relies on few genes withdifferent cross-resistance properties Proc Nat Acad Sci USA 111 5938-5943doi101073pnas1400378111

McKenna A Hanna M Banks E Sivachenko A Cibulskis K KernytskyA Garimella K Altshuler D Gabriel S Daly M et al (2010) The Genomeanalysis toolkit a MapReduce framework for analyzing next-generation DNAsequencing data Genome Res 20 1297-1303 doi101101gr107524110

Menegazzi P Benetta E D Beauchamp M Schlichting M Steffan-Deweter I and Helfrich-Forster C (2017) Adaptation of circadian neuronalnetwork to photoperiod in high-latitude European Drosophilids Curr Biol 27833-839 doi101016jcub201701036

Meuti M E and Denlinger D L (2013) Evolutionary links between circadianclocks and photoperiodic diapause in insects Integr Comp Biol 53 131-143doi101093icbict023

Nanda K K and Hamner K C (1958) Studies on the nature of the endogenousrhythm affecting photoperiodic response of Biloxi soybean Bot Gaz 120 14-25doi101086335992

Nassel D R and Winther A M (2010) Drosophila neuropeptides in regulation ofphysiology and behavior Prog Neurobiol 92 42-104 doi101016jpneurobio201004010

Nassel D R Kubrak O I Liu Y Luo J and Lushchak O V (2013) Factorsthat regulate insulin producing cells and their output in Drosophila Front Physiol4 252 doi103389fphys201300252

Oikarinen A and Lumme J (1979) Selection against photoperiodic reproductivediapause inDrosophila littoralis Hereditas 90 119-125 doi101111j1601-52231979tb01299x

Omura S Numata H and Goto S G (2016) Circadian clock regulatesphotoperiodic responses governed by distinct output pathways in the bean bugRiptortus pedestris Biol Rhythm Research 47 937-945 doi1010800929101620161212515

Overgaard J Hoffmann A A and Kristensen T N (2011) Assessingpopulation and environmental effects on thermal resistance in Drosophilamelanogaster using ecologically relevant assays J Therm Biol 36 409-416doi101016jjtherbio201107005

Parker D J Wiberg R A W Trivedi U Tyukmaeva V I Gharbi K ButlinR K Hoikkala A Kankare M andRitchie M G (2018) Inter and intraspecificgenomic divergence in Drosophila montana shows evidence for cold adaptationGenome Biol Evol 10 2086-2101 doi101093gbeevy147

13


Journal

ofEx

perim

entalB

iology

Pegoraro M Gesto J S Kyriacou C P and Tauber E (2014) Role forcircadian clock genes in seasonal timing testing the Bunning hypothesis PLoSGenet 10 e1004603 doi101371journalpgen1004603

Pittendrigh C S and Minis D H (1971) The photoperiodic time measurement inPectinophora gossypiella and its relation to the circadian system in that species InBiochronometry (ed M Menaker) pp 212-250 Washington DC Natl Acad Sci

Pittendrigh C S and Takamura T (1987) Temperature dependence andevolutionary adjustment of critical night length in insect photoperiodism ProcNatl Acad Sci USA 84 7169-7173 doi101073pnas84207169

Pittendrigh C S and Takamura T (1989) Latitudinal clines in the properties of acircadian pacemaker J Biol Rhythms 4 217-235 doi101177074873048900400209

Pittendrigh C S Elliot J and Takamura T (1984) The circadian component inphotoperiodic induction In Photoperiodic Regulation of Insect and MolluscanHormones (ed R Porter and G M Collins) pp 26-41 London UK PitmanPublishing Ltd

Quinlan A R and Hall I M (2010) BEDTools a flexible suite of utilities forcomparing genomic features Bioinformatics 26 841-842 doi101093bioinformaticsbtq033

Ritz C Baty F Streibig J C and Gerhard D (2015) Dose-response analysisusing R PLoS ONE 10 e0146021 doi101371journalpone0146021

Roff D (1994) The evolution of dimorphic traits predicting the genetic correlationbetween environments Genetics 136 395-401

Saunders D S (1981) Insect photoperiodism the clock and the counter PhysiolEntomol 6 99-116 doi101111j1365-30321981tb00264x

Saunders D S (2005) Erwin Bunning and Tony Lees two giants of chronobiologyand the problem of time measurement in insect photoperiodism J Insect Physiol5 599-608 doi101016jjinsphys200412002

Saunders D S (2012) Insect photoperiodism seeing the light Physiol Entomol37 207-218 doi101111j1365-3032201200837x

Saunders D S (2014) Insect photoperiodism effects of temperature on theinduction of insect diapause and diverse roles for the circadian system in thephotoperiodic response Entomol Sci 17 25-40 doi101111ens12059

Schmid B Helfrich-Forster C and Yoshii T (2011) A new ImageJ plug-inldquoActogramJrdquo for chronobiological analyses J Biol Rhythms 26 464-467 doi1011770748730411414264

Schmidt P S Matzkin L Ippolito M and Eanes W F (2005) Geographicvariation in diapause incidence life-history traits and climatic adaptation inDrosophila melanogaster Evolution 59 1721-1732 doi101111j0014-38202005tb01821x

Schlotterer C Tobler R Kofler R and Nolte V (2014) Sequencing pools ofindividuals-mining genome-wide polymorphism data without big funding NatRev Genet 15 749-763 doi101038nrg3803

Sim C and Denlinger D L (2013) Insulin signaling and the regulation of insectdiapause Front Physiol 4 189 doi103389fphys201300189

Simon C Frati F Beckenbach A Crespi B Lu H and Flook P (1994)Evolution weighting and phylogenetic utility of mitochondrial gene sequencesand a compilation of conserved polymerase chain reaction primersAnn EntomolSoc Am 87 651-701 doi101093aesa876651

Sorensen J G Kristensen T N and Overgaard J (2016) Evolutionary andecological patterns of thermal acclimation capacity inDrosophila is it important forkeeping up with climate changeCurr Opin Insect Sci 17 98-104 doi101016jcois201608003

Storey J D Bass A J Dabney A and Robinson D (2015) qvalue Q-valueestimation for false discovery rate control R Package version 280 httpgithubcomjdstoreyqvalue

Tarone A M McIntyre L M Harshman L G and Nuzhdin S V (2012)Genetic variation in the Yolk protein expression network of Drosophilamelanogaster Sex-biased negative correlations with longevity Heredity 109226-234 doi101038hdy201234

Tauber M J Tauber S and Masaki C A (1986) Seasonal Adaptations ofInsects New York USA Oxford University Press

Teets N M and Denlinger D L (2013) Physiological mechanisms of seasonaland rapid cold-hardening in insects Physiol Entomol 38 105-116 doi101111phen12019

Tobler R Franssen S U Kofler R Orozco-terWengel P Nolte VHermisson J and Schlotterer C (2013) Massive habitat-specific genomicresponse in Drosophila melanogaster populations during experimental evolutionin hot and cold environments Mol Biol Evol 31 364-375 doi101093molbevmst205

Turner T L Steward A D Fields A T Rice W R and Tarone A M (2011)Population-based resequencing of experimentally evolved populations revealsthe genetic basis of body size variation in Drosophila melanogaster PLoS Genet7 e10001336 doi101371journalpgen1001336

Tunstall N E Herr A de Bruyne M andWarr C G (2012) A screen for genesexpressed in the olfactory organs of Drosophila melanogaster identifies genesinvolved in olfactory behaviour PLoS ONE 7 e35641 doi101371journalpone0035641

Tyukmaeva V I Salminen T S Kankare M Knott K E and Hoikkala A(2011) Adaptation to a seasonally varying environment a strong latitudinal clinein reproductive diapause combined with high gene flow in Drosophila montanaEcol Evol 2 160-168 doi101002ece314

Vaz Nunes M and Saunders D S (1999) Photoperiodic time measurement ininsects a review of clock models J Biol Rhythms 14 84-104 doi101177074873099129000470

Vesala L and Hoikkala A (2011) Effects of photoperiodically inducedreproductive diapause and cold hardening on the cold tolerance of Drosophilamontana J Insect Physiol 57 46-51 doi101016jjinsphys201009007

Vesala L Salminen T S Kankare M and Hoikkala A (2012) Photoperiodicregulation of cold tolerance and expression levels of regulacin gene in Drosophilamontana J Insect Physiol 58 704-709 doi101016jjinsphys201202004

Wallingford A K Lee J C and Loeb G M (2016) The influence of temperatureand photoperiod on the reproductive diapause and cold tolerance of spottedwingdrosophila Drosophila suzukii Entomol Exp Appl 159 327-337 doi101111eea12443

Weng M-P and Liao B-Y (2011) DroPhEA Drosophila phenotype enrichmentanalysis for insect functional genomics Bioinformatics 27 3218-3219 doi101093bioinformaticsbtr530

Wiberg R A W Gaggiotti O E Morrissey M B and Ritchie M G (2017)Identifying consistent allele frequency differences in studies of stratifiedpopulations Methods Ecol Evol 8 1899-1909 doi1011112041-210X12810

Zhang Y Liu Y Bilodeau-Wentworth D Hardin P E and Emery P (2010)Light and temperature control the contribution of specific DN1 neurons toDrosophila circadian behavior Curr Biol 20 600-605 doi101016jcub201002044

Zhao J-Y Zhao X-T Sun J-T Zou L-F Yang S-X Han X Zhu W-CYin Q and Hong X-Y (2017) Transcriptome and proteome analyses revealcomplex mechanisms of reproductive diapause in the two-spotted spider miteTetranychus urticae Insect Mol Biol 26 215-232 doi101111imb12286

14


Journal

ofEx

perim

entalB

iology

Insect photoperiodic responses involve a sequence of events fromphotoreception and the measurement of night or day length to theaccumulation of a hypothetical diapause titer in a countermechanismand the downstream regulation of events leading to release orretention of neurohormones regulating diapause (Saunders 2014)Both the measurement of successive photoperiods and theiraccumulation by the counter mechanism are likely to functionunder the circadian system (Saunders 2012) while the downstreamcascading system governing circadian behaviors and photoperiodicresponses may be completely different (Goto 2013) Denlinger et al(2017) suggested that rapid adaptive response in traits regulated byphotoperiodic timer could occur without disrupting daily circadianorganization even if one or a few circadian clock genes functionpleiotropically as a phase reference point for the photoperiodictimer Circadian clock output genes whose dysregulation disruptsbehavioral rhythms without affecting oscillations in the molecularclock itself have been identified in several species (Omura et al2016 King et al 2017 King and Sehgal 2018) It is also becomingobvious that neurotransmitters and neuropeptides play a central rolein regulating the production andor release of hormones involved indiapause induction (Sim and Denlinger 2013 Naumlssel et al 2013Andreatta et al 2018) Reproductive diapause can further trigger ahormonal cascade which induces changes in metabolism (Kubraket al 2014) increases cold tolerance (Vesala et al 2012Wallingford et al 2016) and reduces insectsrsquo mating drive andpheromone production (Ala-Honkola et al 2018) However there isstill debate on whether changes in CDL and other photoperiodicallyregulated traits occur through local selection directed on thecircadian clock system (Pittendrigh and Takamura 1987) ordirectly on the traits involved (Bradshaw et al 2003)Quasinatural selection where the study organisms are transferred

into altered environmental conditions over successive generationsoffers an efficient way to examine the role of environmental factorsin trait evolution and to study trait correlations and trade-offs (Fry2003) For example Agrawal (2000) performed such selectioninducing a host-shift in a polyphagous spider mite Tetranychusurticae and found evidence of a correlation between mitesrsquoperformance on a novel host and their host-plant preference Thelsquoevolve and resequencersquo approach goes one step further by allowingevolutionary responses to selection to be examined at the genomelevel to identify diverged genomic regions and functional pathwaysas well as localizing genes diverging under the altered selectionregime (Turner et al 2011 Burke 2012 Wiberg et al 2017)Martins et al (2014) used this technique and exposed Drosophilamelanogaster flies to Drosophila C virus for 20 generationsAfter selection they pool-sequenced treated and control lines andfound two genome regions containing several candidate genesconnected with an increased survival of the flies under differentviral treatmentsWe have used quasinatural selection combined with the lsquoevolve

and resequencersquo approach to find out whether selection forreproduction under short photoperiods changes CDL and othertraits associated with reproductive diapause and whether it affectsthese traits directly or by modifying the action of the circadian clocknetwork The selection experiment was performed on a northern flyspeciesDrosophila montana where the photoperiod is a key cue forthe induction of adult reproductive diapause This species showsrobust latitudinal clines in CDL (Tyukmaeva et al 2011 Lankinenet al 2013 Venera Tyukmaeva PL JK HK and AHunpublished data) and in the frequency of diapausing femalesunder lightdark (LD) cycles that deviate markedly from 24 hphotoperiodic cycles in NandandashHamner (NndashH) experiment (PL

unpublished results) D montana is one of the most cold-tolerantDrosophila species (Kellermann et al 2012) and its cold-tolerancehas been found to show variation between latitudinally andaltitudinally distinct populations (Vesala and Hoikkala 2011RAWW unpublished results) as well as seasonal photoperiodiccold acclimation (Vesala et al 2012) We hypothesize that if ourselection affects CDL directly and not through changes in the centralcircadian clock network it should affect CDL and other diapause-associated traits but it should not disrupt fly locomotor activityrhythm which is regulated by the central circadian clock (King et al2017) We also expect the genetic divergence of selection and controllines to resemble divergence detected among latitudinally distinct wildDmontana populations (RAWW unpublished data) and to includeSNP variation associated with genes important in the photoperiodiccontrol of CDL and other traits which diverge during selection

MATERIALS AND METHODSQuasinatural selection for reproduction under short-day(late summer) conditionsWe established a genetically variable base population ofDrosophilamontana Stone Griffen and Patterson 1941 by collecting 102fertilized females from a northern Oulanka (Finland 664degN)population in summer 2013 Species identification was performedby sequencing part of the mtDNACOI region of one progeny fly perfemale as described in Simon et al (1994) using forward andreverse primers COI_1F 5prime-ATCTATCGCCTAAACTTCAGCC-3primeand COI_1R 5prime-ACTTCAGGGTGACCAAAAAATC-3prime F3progenies of wild-caught D montana females (about 10 virginsexually mature females and males per progeny) were transferredinto a population cage containing eight bottles with malt medium(Lakovaara 1969) The flies were allowed to mate and produceprogeny in the cage for another 3 generations before startingselection (Fig 1A) During selection the flies were maintained inmalt vials in wooden light-insulated cabinets illuminated withcompact fluorescent lamps (LedEnergie) with light intensity of600ndash1100 lx during the photophase in 17plusmn1degC Length of the lightperiod was decreased by 05 h in each generation in the chamberscontaining selection line replicates until the incidence of femalediapause exceeded 50 This helped us to keep the strength ofselection close to 50 (keeping the flies continuously in the sameday length would have lowered the diapause percentage and reducedselection strength in each generation Roff 1994) Control linereplicates were maintained through the experiment in constant 24 hlight (LL) mimicking the light condition at Oulanka during fliesrsquobreeding season in early summer

In generation F6 200 females and males were separated within3 days of emergence When sexually mature (sim3 weeks old) theflies were allowed to mate in malt vials for 48 h in groups of 10 fliesof each sex They were then transferred into fresh malt vials forfurther mating and egg laying and into new vials 3 times at 3 dayintervals to prevent larval crowding Half of the egg vials weretransferred into 19 h5 h lightdark (LD) (selection line CDL of thebase population) and the other half kept in LL (control line) beforethe flies of next generation flies started to emerge (Fig 1B) Thesame procedure was repeated in generation 7 in respectivephotoperiods using 100 sexually mature females and males perline to establish 3 replicates for both lines (Fig 1C) In generationsF8ndashF15 selection was continued with 100 females and 50 males perreplicate for selection line replicates (half of the females wereexpected to enter diapause) and 50 females and 50 males for controlline replicates (Fig 1D) In each generation the flies were allowedto mate for 48 h in the same LD conditions in which they had

2


Journal

ofEx

perim

entalB

iology













F3ndashF5

A

B

C

D

E

F6

F7

F8ndashF15



3


Journal

ofEx

perim

entalB

iology









Replicate

LD

222 213 204 195 186 177 168 159 1410 1311


2 85 69 83 86 87 81 86 86 893 90 81 86 93 78 74 82 87 89

Control 1 92 86 76 96 78 82 82 82 882 80 81 81 86 81 77 83 81 863 91 97 94 88 85 86 87 81 83


2 143 282 332 2473 131 507 494 355

Control 1 296 319 289 257 1812 479 338 491 335 3313 340 171 313 187 172

4


Journal

ofEx

perim

entalB

iology









LD

DD126 1212 1224 1236 1272

Selection 1 250 130 250 359 293 1552 280 161 193 270 372 2233 248 221 239 314 368 341

Control 1 167 133 228 257 353 2692 371 160 388 198 159 3283 252 87 242 224 318 221


Replicate N

LD (195) LL










5


Journal

ofEx

perim

entalB

iology














6


Journal

ofEx

perim

entalB

iology







1410 168 186

LD cycle

204 1410 168 186 204

02

0

Pro

porti

on o

f dia

paus

ing

fem

ales

04 Estimated CDL

06

08


0

02

04

06

08

10

A B


7


Journal

ofEx

perim

entalB

iology







χ2 df P-value



126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD


8


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology













F3ndashF5

A

B

C

D

E

F6

F7

F8ndashF15



3


Journal

ofEx

perim

entalB

iology









Replicate

LD

222 213 204 195 186 177 168 159 1410 1311


2 85 69 83 86 87 81 86 86 893 90 81 86 93 78 74 82 87 89

Control 1 92 86 76 96 78 82 82 82 882 80 81 81 86 81 77 83 81 863 91 97 94 88 85 86 87 81 83


2 143 282 332 2473 131 507 494 355

Control 1 296 319 289 257 1812 479 338 491 335 3313 340 171 313 187 172

4


Journal

ofEx

perim

entalB

iology









LD

DD126 1212 1224 1236 1272

Selection 1 250 130 250 359 293 1552 280 161 193 270 372 2233 248 221 239 314 368 341

Control 1 167 133 228 257 353 2692 371 160 388 198 159 3283 252 87 242 224 318 221


Replicate N

LD (195) LL










5


Journal

ofEx

perim

entalB

iology














6


Journal

ofEx

perim

entalB

iology







1410 168 186

LD cycle

204 1410 168 186 204

02

0

Pro

porti

on o

f dia

paus

ing

fem

ales

04 Estimated CDL

06

08


0

02

04

06

08

10

A B


7


Journal

ofEx

perim

entalB

iology







χ2 df P-value



126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD


8


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology









Replicate

LD

222 213 204 195 186 177 168 159 1410 1311


2 85 69 83 86 87 81 86 86 893 90 81 86 93 78 74 82 87 89

Control 1 92 86 76 96 78 82 82 82 882 80 81 81 86 81 77 83 81 863 91 97 94 88 85 86 87 81 83


2 143 282 332 2473 131 507 494 355

Control 1 296 319 289 257 1812 479 338 491 335 3313 340 171 313 187 172

4


Journal

ofEx

perim

entalB

iology









LD

DD126 1212 1224 1236 1272

Selection 1 250 130 250 359 293 1552 280 161 193 270 372 2233 248 221 239 314 368 341

Control 1 167 133 228 257 353 2692 371 160 388 198 159 3283 252 87 242 224 318 221


Replicate N

LD (195) LL










5


Journal

ofEx

perim

entalB

iology














6


Journal

ofEx

perim

entalB

iology







1410 168 186

LD cycle

204 1410 168 186 204

02

0

Pro

porti

on o

f dia

paus

ing

fem

ales

04 Estimated CDL

06

08


0

02

04

06

08

10

A B


7


Journal

ofEx

perim

entalB

iology







χ2 df P-value



126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD


8


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology









LD

DD126 1212 1224 1236 1272

Selection 1 250 130 250 359 293 1552 280 161 193 270 372 2233 248 221 239 314 368 341

Control 1 167 133 228 257 353 2692 371 160 388 198 159 3283 252 87 242 224 318 221


Replicate N

LD (195) LL










5


Journal

ofEx

perim

entalB

iology














6


Journal

ofEx

perim

entalB

iology







1410 168 186

LD cycle

204 1410 168 186 204

02

0

Pro

porti

on o

f dia

paus

ing

fem

ales

04 Estimated CDL

06

08


0

02

04

06

08

10

A B


7


Journal

ofEx

perim

entalB

iology







χ2 df P-value



126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD


8


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology














6


Journal

ofEx

perim

entalB

iology







1410 168 186

LD cycle

204 1410 168 186 204

02

0

Pro

porti

on o

f dia

paus

ing

fem

ales

04 Estimated CDL

06

08


0

02

04

06

08

10

A B


7


Journal

ofEx

perim

entalB

iology







χ2 df P-value



126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD


8


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology







1410 168 186

LD cycle

204 1410 168 186 204

02

0

Pro

porti

on o

f dia

paus

ing

fem

ales

04 Estimated CDL

06

08


0

02

04

06

08

10

A B


7


Journal

ofEx

perim

entalB

iology







χ2 df P-value



126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD


8


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology







χ2 df P-value



126

0

02

04

06

08

10

1212 1218 1224LD cycle

Fem

ales

in d

iapa

use

()

1242

SelectionControl

DD


8


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology





ndash10

Males

SelectionControl


ndash15

ndash20

ndash25

LL 195 177 159 1311 LL 195 177LD cycle

CT m

in (deg

C)

159 1311 LL 195 177 159 1311

A B C






9


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology




20

15

10

05

20

15

2

3

4

5

X

10

05

20

15

10

05log 1

0 (q

-val

ue)

20

15

10

05

20

15

10

05

0 25 5Position (Mb)

75 10 125


10


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology








11


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology


































12


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology
















































13


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology



































14


Journal

ofEx

perim

entalB

iology

Selection for reproduction under short photoperiods changes ...The incidence of reproductive...

Documents

Transcript of Selection for reproduction under short photoperiods changes ...The incidence of reproductive...