The Joumal of Erpcrintental Biology" 3t1-1.1659-1666 (2(Xll ) 1659 Pnnted in Crert Brrtrin {OThe Comprn! of Bioiogists Linrtcd l(X) I JEBl.l l8
RAPID COLD-HARDENINGOF DROSOPHILAMELAIVOGASTER (DIPTERA: DROSOPHILIDAE)DURING ECOLOGICALLY BASEDTHERMOPERIODIC CYCLES
JONATHAN D. KELTYX ANDRICHARD E. LEE JN Departmento.f Zoology,212 Pearson Hall, Mictni Universir:^,Oxfnrd, OH 45056, USA *Presentaddress:TheUniversityoiChicago.Departnentof OrganismalBiotogyandAnatomy. l027E.57thStreet.Chicago, IL60637,USA (e-mail: j-kelty @ uchicago.edu)
Accepted 7 Februart'; published ort WWW 5 April 2001
summary In contrast to most studies of rapid cold-hardening, in critical thermal minima (CI*i") decreasedby 1.9"C. Cold which abrupt transfers to low temperatures are used to hardiness increased with the number of thermoperiods to induce an acclimatory response.the primary objectives of which flies were exposed; i.e. flies exposed to six this study were to determine (i) whether rapid cold- thermoperiods were more cold-tolerant than those exposed hardening was induced during the cooling phase of to two. Endogenous levels of Hsp70 and carbohydrate an ecologically based thermoperiod, (ii) whether the cryoprotectants were unchanged in rapidly cold-hardened protection afforded vvas Iost during warming or adults compared rvith controls held at a constant 23 "C. ln contributed to increased cold-toleranceduring subsequent nature, rapid cold-hardening probably affords subtle cycles and (iii) whether the major thermally inducible benefits during short-term cooling, such as allorving D. stress protein (Hsp70) or carbohl'drate cryoprotectants melanogasterto remain active at lower temperatures than contributed to the protection afforded by rapid cold- they otherwise could. hardening. During the cooling phaseof a single ecologically based thermoperiod, the tolerance of Drosopltila melanogastertoth at-7"C increasedfrom 5+5 96survival Kev w,ords:acclimation. cold shock, cold tolerance. crvoprotectant. to 62.5t7.3 7o (means + s.E.\,l.,l{=40-60), lvhile their Hsp70.heat-shock protein. Drosophila nrclarutgctster.
Introduction In contrastto most studiesof acciimation.which examine heniolymphincreases b), approximately300% (Chen et al.. phenornenainduced over prolongedexposures lasting days. 1987).On a time scalesimilar to that of rapidcolcl-hardening rveeksor evenmonths. a numberof studiesduring the past l5 (i.e. u'ithin minutes),l-righ temperatures elicit the svnthesisof years have be_gunto examinerelatively lapid processesof heat-shockproteins (Burton et al.. 1988:Goto and Kimura. acciimationto low and high temperature(for reviews,see 1998:Feder and Hotmann.1999). These proteins presumably Federand Krebs. 1997:Deniinger and Lee, 1998:Denlinger protectagainst thermal injury b;"acting as rnolecular chaperones. and Yocum. 1998). Upon exposureto rnoderatelylow minimizing the aggre-sationof non-native proteins and temperature.diverse insect speciesexhibit a rapid cold- promotingthe degradationand removalof both non-naiiveand hardeningresponse. ,,vhich within periodsof minutesto hours aggregatedproteins from thecell (Federand Hofmann.1999). enhancestheir cold-tolerance(Chen et al.. 1987;Lee et al.. Within an organism. a varietl, of therntal protective 1987:Coulson and Bale. 1990;Coulson and Bale. 1992:Kelty mechanismsmay act in svnchronv to prevent cellular and Lee. 1999).For example,prior exposureof housefly darnage(Feder and Hofmann. 1999).Substantial evidence (fuIttscttdoryestica) pupae to 0'C lor 90min increasedtheir denonstratesthat sugarsand poiyols botli serve protective sun'ival lblloiving treatmentlor 2 h at -7 "C from 0 7oto above rolesat lorvtemperatures and heip to stabilizenrenrbrane lipids 80Vc (Coulsonand Bale. 1990). Similarly, insectscan andcellular proteins at high temperature(Kim andLec. 1993: acclimaterapidly to irightemperature (Milkman. 1963: Levins. Ramoset al.. 1997).Siniilarlv. exposure to low temperaturc 1969:Chen et al.. 1991;Krebs and Feder. 1997). inducesthe explessionof stresspl-oteins, but onlv uponreturn The phl,siologicalrnechanisms underlvrng rapid acciimation to liighertenlperatures. and high tempcrilturcsi.rock rncrelses to iow and high temperatureremarn poorlv understood.ln the toleranceof insectsto lorv ternperatul'cs(Burton ct al.. prepu'ationfbr rvinter,insects accurnuiate low-molecular-mass 1988:jopiin et rl.. 1990;Nunamaker et rl.. 1996:Yiangou et polyirvdriculcohols and sugarslDenlinger aud Lee. 1998). al.. 1997:Coto and Kirnura.l9c)8; Goto ct al.. 1998t.Thus. Sirrrilarly. during lapid cold-hardeningin Surt:ttltltuqu commolrelenreuts rnav underliea toicrancco1' lou, and high cra,;sipolpis.the givceroi concentrationwithin the pupal Iemperatufes. 1660 J. D. KslrY ,rNoR. E. LeEJn
As a tirst stepin determiningthe ecological signiliclnce of the capacityfor rapidrcclimation to low or high temperature. we recentlydemonstrated that rvhenD. tnelutogttJterwere cooledat slower.and thus more natural. rates they exhibitedI signiticantly greater cold-rolerancethan did flies cooled (Kelty 1999). When cooied at at higher fates and Lee. i) ecologicallyrelevant rates (e.g.0.05"Cmin-r). flies not only exhibitedhigher rates of survivalat sub-zerotemperatures than = - IO their more rapidly cooled counterparts,but rapid cold- ! hardening provided them with proiection at I I "C. a temperaturethat flies llre more likely to encounterin nature thanthe near 0 "C exposuresused to elicitrapid cold-hardening in most previous studies.Furthermore. the fact that D. ntelurLogastercooled at slowerrates entered a stateof cold torpor (i.e. reachedtheir criticalthermai minimum or CTmin) et lower temperaturesthan those cooled at higher rates l6:00 l9:00 2l:00 0l:00 04:00 07:00 l0:00 ll:00 l6:00 provides strong evidencethat rapid cold-hardeningcould Time of day (h) beneht this speciesat the temperatul'esthev are likely to Fig. 1. The 2,1h thermopenodto rvhich Dntsctl.thilrtrnelunoqaster encoullterin nature. were exposed. Arrows representthe points at rvhich flies were the presentstudy. we investigatedrapid acclimationto In removedand their cold-toierance(assessed as survival atter I h at low temperaturesin D. melunogtLslerduring an ecologicallv -7'C and as CZ'nin)or heat-tolerance(assessed as sunival alter I h based thermoperiod. The cooiing phase of a single at i8 or l9"C) determined.The upper bar shows the l2h: l2h thermoperiodinduced substantialprotection against cold- light:darkphotoperiod. with the hlled region representingthe dark shock injurv. much of which r.vasretained during the rvarmin-Q period.CGriu, critical thermalminimum. phase of the thermoperiod.The increasedcold-tolerance appearedto be not only a tunction of the phase of the thermoperiodfiom which aduitswere removed. but was also shadedD. meltuutgctslerhabitat at lliami University's Ecology dependenton the numberot thennoperiodsto r.vhichthey had Research Center (Oxlbrd, OH. USA) during the spring, been exposed;flies exposedto multiplethermopenods were sul1rmerand autumn of 1996. These data lvere used to program substantiallvmore tolerantof low temperaturesthan flies an incubator to cycle beiween 9 " and l3'C ttver a period of exposedto a singlethermoperiod. 2-lh tFig. l); similar thermoperiodslrequently occulred during mid-spring and mid-autumn. The incubator was set to il photoperiodof l2h:i2h L:D. with the lights turned on at Nlaterials and methods 09:00h. lnsectreurttt? We first determinedwhether D. rneltmo,gasrarcold-hardened Drosoplilo rneltutogaster(Oregon-R strain) were reared rapidlv during the cooling phase of the programmed 'C undera long-davphotoperiod (15 h:9 h L:D) ct 23 in halt- thermoperiodand whether hardening rvas lost during warming. pint milk bottles containing DrosopliLtt medium (corn Flies rvere lightly anesthetizedrvith COt and transt'enedin rneal.molasses, yeast. agar) as tood and es a substratumtbr groupsof l0 into giassculture tubes (12x75mmt contlining oviposition.Nervll, emergecl adults were removed tiom bottles approximately0.3 ml of Drosophila medium and a t'ew -qrains daily and transt'erredto tresh medium on which they rvere of live velst.These flies were then allowed to recoverat 23'C allor.vedto teedand lay eg-rstbr 9 days.Within 8 h of eclosion' lor at lelst 6h. Afier recovery,flies rvere ti) kept at l3 "C tbr the llies rvere lightly anesthetizedwith CO:, segregatedby up to 6 davs,(ii) transt-enedto the thermoperiod(beginning at genderrnd transf'enedinto liesh bottlescontaining Drosopltila 23"C) for 0-6 days.(iii) keptat 16'C lbr up to 6 daysor (iv) rnedium.This enableclus to studv vilgin flies and cxamine rnaintainedat l3'C for txe clr hve additionaldays then potentialdit'ferences in the capacitvoi Inale and t-emaieD. subjectedto all or partof thethermoperiod. As outlinedbelow. tnelutoqttster-to cold-hardenrapidlv. Because age ma1'eft'ect the thernraltolerances of D. melunsester subjectedto the this capacitytCzcjka and Lce. 1990).all tliesrvere I davsold thermoperiodwere assessedat the times and temperatures lt the l'reginningtrf e'achexperiment. cxcept where the effect irrdicatedin Fig. 1.The thermaltolerances of D. rnelunoqasrer oC of agervas examined. heldat constantl6 or l3 wereassessed everv 1l-24h. 'f he rmtt p er i ttd e.rpe ri tn ent s Sun'ivuLut lott' untl li.qlt remperatures To deterntrnewhat thermoperioda tiuit lly is likelv to As an index of overall cold-hardiness.we assessedthe experiencein nature.a computerizeddata-accltrisition svstern caplcitv of D. rnelanogusterto survive I h tlf exposure to (CarnpbellScicntiflc. modei CRl0) rvas used to record -7 "C. We chosethis temperalurebecause preliminarv data tcrnperatureeverv hour ut rhe leaf litterisorlinterface ln I inciicateclthat. in the absenceof cold-hardening,it produced .siffi 'r{i' ?nzt
$;* Cold-hurdeningofDrosophila melanosaster l66l ff ,1,, nearly100 d/o ntortality (Kelty _i" asa resultof non-freezinginjury extractto be assayedwas evaporated to drvnessat 45 "C under andLee. 1999).Tubes of flieswere inverted in test iil _slass tubes nltrogen gas. and then resuspendedin water before being partially(>90Vc) submerged in a refrigeratedbath (Neslab. injectedonto a Dionex PA-l column. Upon elution fiom -7'C. model RTE140)set at Inversionprevented entrapment the column. sugarsand pol1,olswere detectedby pulsed of Uiesin themediun'r and the initiation of inoculativefreezing amperometry. by frozen rnedium. The temperaturein each tube was monitored on a chart recorderusing a copper{onstantan Detectiottof Hsp70 bl S_e.lelectrophoresis tntl v'estenr thermocouplelvhose tip was positionedagainst rhe inner DIoffilrg surfaceof the foam pluggingthe tube. which rs wherethe flies D. tnelanoga.rrerwere allocated to fir'e groups of three. fell whenincapacitated bl,chilling. Exposures were rimed from Theseffies were kept ar 23 "C. subjecredto 37'C for 90min, the point when the averagetemperature amongst tubes reached removed from the thermoperiodat the points indicatedin -6.9"C (within 10min of rransferro the barh).Sun'ival was Fig. 1, or held at 0"C for lh. After eachtrearmenr. flies were assessed24h afier the endof the sub-zeroexposure period as immediatelysnap-frozen in liquid nitrogen.then storedat the percentageof fliesper tube able to right themselvesand -80'C until assavedfor theheat-shock protein Hsp70. Soluble walk. proteinswere extractedb1, homogenizing each group of flies irr ice-cold27o (wlv) completeprotease inhibitor (Boehringer Assessmentof critical tlterntalntinirnu Mannheim 1691198) in phosphate-bufferedsaline. The -10-60 To assessClmin, -sroups of flies were transferredinto resultinghomogenate was centriluged for 30min at l4 000revs jacketed a -9lasscolumn. the temperarureof which was min-1.and the supernatantwas [ecovered. Protein content was controlled by circulating fluid fiom a programmable de{erminedusing a BCA assay(Pielce Biochemicals). proteins retiigeratedbath. After l0mrn, the temperaturewas decreased were separatedby electrophoresingl0 ;rg of prorein in r. from 23"C at l.0"Cmin Fliesunable to clingto surfacesin each lane of a 107c Tris-HCl sodium dodecvl sulfhre the column (i.e. thosecooled ro rheir Clmin) fell into glass poiyacry,ls6ide(SDS-PAGE) gei. Followingtheir separation 'C culture tubes. which were chansed everr, 0.1 tall in by SDS-PAGE. proteinswere transfenedelectrophoreticallv temperature.To prevent flies from crawling ro a position ontoa PVDF membranethat was then blocked u,ith phosphate- sufficientlyclose to rhe openend of the tube rhat they could buff'eredsaline containing 107. non-farpowdered miik. The fly out and escape,the inner surf-aceof each tube was coated membranewas washedsuccessively in solutionscontaining with Fluon. Although D. ntelttnogctsterof both genders primary antibody specific fbr inducible Drosophila Hsp70 exhibited ne-gativegeotropism, males exhibited _qreater (7.FB.a gift h'omSusan Linquist. Universitv of Chicago).rhen iocomotor acti\/it)/ and often becameprematurelv trapped in peroxidase-conjugated-soat anti-rat IgG secondarl'antiboclv the collectinstubes. therebv biasing our estimationof Clmin. (JacksonImmunoResearcir Laboratories. no. I l2-035-003). For thisreason. only lemaleswere used fbr thisdetermination. Bound antigen was detectedby chemiluminescenctusins Because of the small size of D. n.reLanogastet-,body the SuperSignalCL-HRP substratesystenl accordin_qro temperaturewas approximatedas the air temperaturewithin the manufacturer'sinstructions (Pierce Biochemical no. the column(Huey et al.. 1992). 3.+080).
D et e nn i ttot io n oJ ccu' b o h\, d rat e cn, op r ot e c t a n t s Dcttcttutalysis Hirh-perfbrmanceliquid chromatography(as describedby Data rvereanalvzed usins analysisof variance(ANOVA). Hendrixand Wei. 1994)was usedto determinervhich, if any. When comparingsurvival rates. data rvere iirst transfbrmed by carbohvdrateswere synthesizedby D. ntelattogLt.tterduring takins the arcsineand squareroot of tlre obsen,edsurvival rapid cold-hardenin-e.Under light CO: anesthesia.groups of proportions.Treatment dilfelences were consider.ed significant -10-85mg) 90-150 ttotaiing 2-da1,-oldadulrs were weiehed at P<0.0-5.Values are reporled as mealts+ s.l.M. andplaced in glassculture tubes. which werethen sealedrvith foam rubber.Flies were allowedto recoverai 13'C lbr 6h. and were then eitherkept at 23'C or transferredto 16.9 or Results 1'C tbr 2h. Flies were then preservedin I ml of 807o non- Rttp icl r: o I d - Itu rd eri ttg tlu ri tts utt t oI o,q i c a Ih' ba s e tI .e denaturedethanol. Carbohydrateswere extractedtionr D. llle rnloperrod tnclutogastersamples in three steps (809b non-denatured We first dererminedu,hether rapid cold-lrarcleningrvas ethanoiat 80"C tbr 20rnin.507cethanol ar 80'C fbr l5rlin. inducedduring the cooling phaseo1' an ccolosicallvbased and80% ethanolat 80"C lbr l5min). Charcoalrvas aclded to thermoperiodand rvhetherthe protectionit proviciedr.vas lost eachextract to removernost lipid. all antinoacids and most. if duringwarning. To do so.D. tnclutto.qu.rrrrwere subjected to (Hendrix not all. color and Peelen.1987). The charcoalr.vas all or partof sucha rhenroperioci.and their capacitv to survi\,e then oC removedbv passinethe extracts through a slass-hberfilter I h of exposureto -7 rvasthen asscsscd lTabie I ). .\s llies paper(Whatntan GF/A) andtwo iaversof extra-fineglass-tiber cooledliom l3 to l6"C at an averarerare o1' l.-{.Ch-1. rhcir paper(Whatrnan GF/F). Becauseethanol interf'eres rvitl-r the survivalalter sub-zerotreatntent increased srgniiicantlv l'rrtrn detectionof potentialcrvoprotectants, rhe portioll o1-caclr 5+5% to 19.lt6..-l7o (P=0.002!. Then. rrsrl.)ev cooied il.om
L--. 1662 J. D. Krlrv ,rxoR. E. LEeJn 'fable I . Elfects of'repetLtetlthermul ctclinil on tlrc CT,,iu rt.l 80 -+- I daysthen cycled Drosophila melanogal;teruntL on tlteir ubilin' ttt sun'it'e I lt -!- 3 daysthen c-u-cled -7oC exposure to -*- i davsthen cvcled ___ _60 Day of Time Temperature Survival Clnin a adultIii'e (h) r"C) (7a) ('C) = 1 l7:00 23 5.0*i.0 7.9t0.1 -Z10 l l2:00 l6 29.2x6.3 6.1t0.I - J ()9:00 9 62.5+1.3 6.0+0.1 ll:30 t6 62.5t9.6 6.lr0.l 220
-l l7:00 23 18.3t8.9 6.9r0.1
-l l2:00 t6 60.0*5.3 6.7t0.t -+ 09:00 9 66.'7t9.2 5.3t0.1 -l l2:30 l6 78.3t6.0 6.1t0.I + l7:00 t3 70.0t5.3 5.6t0.t 17:00 12.00 09:00 Il:00 17:00 1 l7:00 l3 83.3t9.2 6.8t0.1 Timeof dayt h ) 1 ll:00 t6 90.0r1.6 6.5*0.i Fig.3. Effectsof ageon thecapacrty of Drosnltltiltttneleuogoster to 8 09:00 9 86.7t6.I 5.8t0.1 cold-l-rardenrapidly during an ecologicall;-based thermoperiod. I l2:30 l6 93.3t.19 NA Overall.flies cycled from the besinning of the'econd day of adult 6.9t0.I 8 i 7:0() l3 9-1.it3.3 litb weresignificantly better able to surviveI h of exposr:reto -7 "C thanwere flies that began cycling at an oiderage (P<0.0001). Each CT,r,n,criticrl themai mrnrtrrum. pointrepresents the mean percentage survivai : s.E.\'l.oi 6-18tubes each Survilal data lre meansI s.e.ll. sun'ival of 6-18 tubes erchconteining 10 rdullflies. containrngl0 flies. Critical thermalminimum values;ire rreanst s.e.lt. tbr -10-60 flies. Cold-tolerancecontinued to increaseduring subsequent NA. not sampled. thermoperiodsover the next 6 days (Fig.2). By the time D. melturc,qasterhad cooied to l6'C duringthe coolingphase of -7'C l6 to 9oC (averagerate 0.6"Ch-l), theircapacity to survire the secondthermoperiod. their survival alter t h at sub-zero treatment again increased significantly (to (60.0t5.8%) wassignificantly greater than that of fliescooled 62.5t1.-\%: P<0.000I when comparedwith either the 13'C to l6"C during the first thermoperiod(Table 1. P=i).0033). or l6'C sroups). Although their cold-hardinessdecreased Furthermore.the averasesurvival of flies testedduring all or significantl;r as they were warmed at an average rilie of partof 2 daysof cycling(60.3t.1.7 7c) ,,vas significantly qreater 1.75'Ch-l fiom 9 to l3 "C (P=0.026).they retained much of than that ol flies tested during all or part of the first the cold-hardinessthat had accrueddunng the thermoperiod thermoperiod(35.0t5.07c. P=0.000,1. Fig. 2). Overall. flies r-1S.3t8.9oc t. subjected to all or part of a sixth thermoperiod were signiiicantly more cold-tolerantthan those cycled tbr I (P<0.0001)or r davs (P=0.00011.exhibiting rn average survivalrate of 88.3t3.07o fbllowins I h ttf exposureto -7 "C tFig.2). When kept at a constant 13 "C. the capacity of D. melonogasterto cold-hardendecreased rapidly with age tFig.3). Fliescvcled beginning when thev rverei daysold oC were srgnificrntlyless able to sur'"'iveI h at -7 than were their counterpartsthat enteredthe thermoperiodwhen 2 days old LP<0.0001).If keptat 13" tbr 7 daysbetbre cycling. all aduitsrvere killed bv I h ol exposureto -7'C. regardlessof the point lt rvhich thev had been translerrediiorn the l-t) thermoperiodtFig. 3). Consecutivecvcle numbcr To determinervhelher gender atfected the capacityof D. melanogu.sterto cold-hardenrapidiy. rvecompared the survival of repeatedthernoperiodic cvciins, lbr up to 7 davs. Fig. i. Eftects of maies and t'emalestransf'ened 3t various tin.resduring two on the cold-tolerirnceof Drosophilu rnelutogusterrtssessed as consecutivcthermoperiods to -7'C tbr I h (Fig. -1).As with survivri ltier I h ilt -7'C. Each colurnn fepresentsthe meln the cxperimentalpopuiation es a rvhole. the capacitv of pcrcentngesurvival + s.lr.!1.of -15-60tubes oi I0 flieslveraged over exposureto -7'C increased Irn cntire therrnoperiodicc-vcie. Different letters lbove the coiumns each gender to survive a I h indicate that survival ievels are sisnillcantlvdifferent ti'onl erch signiiicantlybetween the first lnd the sectlndthermoperiod otherxt P<0.(XX)1. (P<0.000 I . l'or both genders). However. rnales were Cold-lnrdeningof Drosophilamelano_easrer I 663 r00 ABCDEFGH
.><
= OU
=40
17:00 9:00 17:00 9:00 l7:00 Timeof da1,(h1 -i. Fig.5. A representativewestem blor showing Fig. Effectsof genderon thecapacity of Dro,sophilamelnno,gaste r theexpressron of. the rnducible70kDa heat-shock protein (Hsp70. to cold-hardenrapidly during two successiveecoloeically based inclicatedby theanow.) in heat-stressedDrosophih nrclru.tosusler therrnoperiods.overall. mareswere better able to cold-harde' taduitswere exposedto rapidly. -i7"C for 90min:lane A1. Hsp70 rvas not as reflectedin their capacitvro survive I h at _7 "C detectedin solubleprotein extractsl'rom D. melonoguster (F=21.71.P<0.0001). Each poinr representsthe meanpercentace keptat 0"C fbr lh (laneCi or fiom thoseof adultssampled at an1' survival+ s.E.t\4.of 3-9 tubeseach conrainins t0 aduitflies. timepoint in thethermopelod shown in Fig.I (lanesD-H). No proteinwas loaclecl rnto lane B.
significantlymore tolerantof sub_zeroexposure than were Denlin-gerand Lee. 1998).Althou-uh Hsp70 fenralesrP<0.000 rvasexpressed in | t. positive controls(flies exposed to 37.C 1br90 mrn;. nonewas detectedbv westernbrot anarysis of protein EJfectoJ ctn ec'ologicalh. basetl rhermolteriod exrractsfrom fries on CT,,i, kept at 23"C. thosesampled ar eachpoint labeledin Fis. I or The temperatureat which an orqanism enters a state of thoseexposed to 0'C (Fig.5). torpor(i.e. its CZn,,n)is oftenused as an indexof the etfbctof thermaiacciimation on bel.ravioralfunction (David et al.. i99g: Hori and Kimura. 1998;Kelt1, and Lee. 1999).To determine Discussion whetherrapid cold-hardeninecould benefit D. meranosctsrer Leeet al. (Leeet al., i987) ibund at temperatures thatrapid cold_hardening. that they would be likely ro encounrer.in inducedby brief exposureto 0'c. ailowecia varietvof insecis nature,we determinedwhether their Clrnin changedduring to survtveexposures to otherwise the lethalsub_zet,o remperatures. thermoperiod. During the first thermoperiod, as O. Simiiarll,.Meats (Meats, 1913) demonsrrared rhar when cooled tnelrutosastet-cooled fiom 23 to 9 "C. their Clmin decreased gradually(as quickly as 1"Cmin-l).a species significantlvfrom 7.9+0.2 .C of tephrititifruir to 6.0+0.1 (p<0.0001).Howevsl, t11,I.Dacus . 'roni') oC. exhibiteda lower cold torpor remperature in contrastto survival after I h at _7 Clmtn remained thanwhen cooledabruptly. Botl.r l_ee ct al. (Lee unchangedduring er al.. l9g7) subsequenrrhermoperiods (Table I ). and Meats (Meats. 1973) hl,pothesizedthat thc capacitvto acclimaterapidlv to low temperarurei i.e. to Mecltctnismsof rapid cold-lrurclening:rol e of.c.arbohtclrate colcl-harclen rapidly) allows the cold-toleranceoi- ct1'ol) rote ctatlts an organlsrl to track shon-term changesin eni,ironmentlltentpe-r.arure,. suclt cs in a numberof insects.increased cold_tolerance. rncludine thoseoccurrin,e during rrarureldiurnuj that at'forded eotrlinq.Althoueh the by rapid cold-hardenins,is correlaredwirh tir! rlrost recentor'these studies rvas publislrctl tlvcr l decad-ereo. productionof crvoprotectivesubstances such as polvhvdric little is known about rhe ecoiogicalreler.,ance of rapid coict_ alcoholsand sugars(Chen et al.. l9g7: t_ee.tOStl. Higtr_ hardening.In thepresent studv. our primarvobjecrives perfbrnance liquid chromatography wereto was therefore used to determinewhether exposurc determiner, to tlte cooling phase of an whether carbohvdrate cryoprotecrants were ecologicallvbased thermopeilod woukl indrlcc raprclcokl_ synthesizeddurinc the rapid cold_iiardening processrn D. lrarderringin D. nrclcuto.qo.\tcr ntelanogaster. lnd rvhether the protection A numberof carbohydrateswere fbund in this atfordedr.vould be retainedor losttiuring rhe q,arntinq phasc. species.includin-e glycerol. trehalose, liucrose and other We found that the presumed. colcl-toleranccol D. ntclanoquster but undetermined.carbohl,drates. Howeyer. the rncreasedsirnificanrly durinq rhe cooiingphase oi l single concentrationof cuchremained approximatelv the same oC in flies thermoperiodbegun on thc seconciciav ot keprat .3 or cooled -l .C adult lilc. Durrng to 9. or I fbr I h. tltis phase oi'the cvcle. the clpacitl, ,.t1,D. rnelutroqu.\tcrLo sLlrvlve exposure to a sub-zero tentperaturerncrelscd M ec I tut i sn,s o.t'-rap i d t.o !tl -I n rtl cn i n,q: H sp 70 srsnificantly,r.vhile their CI,n11decreascii Althoughthe sisnilicanrlr,.Tlre inducti.nof stressproteins at higli rempefatLu.e dcurcuse ('frn,,r irr tlratrve olrst.r-rcd orgy..j1 1l1cl-nropgl 111,11.. has bee' cxtensi'elvstuclied. rittre is known oi wherherrhev c),cles(2. I"C) rvassirnilar lo thattrbsen,ed lre involvedin pr.er,iousl-v(Hori low temperaturetolerance (titr ,, t-euie*.,.. and Kimura,l99tt) in Dr..r.1thiiutrttpc:.i.f r.tt.r.'ccii'tated rcr 1664 J.D. Kelrv eNoR. E. LEsJn l-5"C fbr t6 days(2.3 "C) lnd to our previousfinding that the The absence of discernible carbohydratecryoprotectant 'C Cfrninof D. meltmogu.rrercooled at 0.L min-l was 1.6"C synthesisby D. tnelattogasterin thepresent studv extends our lowerthan that of f'licscooled at 1.0'Cmin-l (Keltyand Lee. previous finding that glycerol was not synthesizedby this 1999).Taken together. the survival and CZ6;6 data sug-eest that speciesduring rapid cold-hardening (Kelty and Lee. 1999). rapid cold-hardeningoccurs during thermoperiodic cycling in However, this findin-e contrasts rvith data tiom S. nature. c'rttssipolpis,in which rapidcold-hardening is correlatedwith The increasesin cold hardinessthat we observedprobablv a threefoldincrease in hemolymphglvcerol concentration representa conservativeestimate of the capacity of D. (Chenet a1.,1987: Lee. l99l). tneLcutogusterto cold-hardenrapidly in nature.Populations .\dditional mechanismsmav pertieipetein iupid cold- oi D. melcnngusterkept in constant conditions exhibit hardening,including other stressproteins. For instance.in quantitativegenetically based ditferences. such as thermal plants. a group of small hearshockproteins is correlated range,body size and f-ecundity,trom thosekept underother rvith the acquisitionof cold-tolerance(Sabehat et al.. 1998). conditions(Cavicchi et al.. 1985:lvlorin et al.. 1997).The line Another possibilityis that. during rapid cold-hardening,the we usedin this study(Oregon-R1 has been held in a thermally compositionof lipid membranesle.g. ceil membranes, constant envlronment(approrimately 23 "C) for nearly a endoplasmicreticulum) is alteredsuch that thesemembranes century and probably representsa more stenothetmal maintaintheir fluidityover a widerrange of temperaturesthan populationthan thosetound in nature.As such.the ranseof would otherrvisebe the case.Such homeoviscous adaptation temperaturesto which this populationis able to acclimateis hasbeen described in a varietvof invertebratesand vertebrates probiibly namower than that of naturai temperate-zone during both long- and short-termacclirnation (Carey and populrtions(Somero et al.. 1996). Hazel,1989; Hazel, 1995: Ohtsu et al., 1998). Although prolongedexposure to moderatelylow or high Much remainsto be learnedabout the mechanistic basis and temperatureotten increasesthe cold- or heat-tolerancean ecologicalrelevance of rapidcold-hardening. Nonetheiess. our organism.respectively (e.g. Levins. 1969;Hori and Kimura. previousreport (Kelty andLee. 1999)and clata presented here 1998). animalsin nature are exposedto temperaturesthat indicate that, in nature.shon-term cooiing such as occurs lluctuateon both a diumai anda seasonalbasis. We fbundthat during natural thermoperiodsinduces rapid increasesin cold- much of the cold-tolerancegained bv D. nteltmoga.rrerdurinu tolerance.The importanceof this phenomenonin natureis an initialthermoperiod remained during the wanr-ringphase of probably not to increasethe capacityof D. rnelctnogctsterto the cycle.Funhernrore. the level of protectionincleased rvith surviveexposure to sub-zerotemperatures on a diurnal basis the number of cycles experienced.Exposure to fluctuating since rarelv, if ever. would a temperate-zonefly encounter temperaturesis an eifectivemeans of lcclimating insectsto temperaturesranging fiom l3 to below0oC over thecourse of Iow temperature(e.g. Hanec and Beck. 1960:Anderson and a singleday. Rather. this processrnay benefit the or-eanismby Hanvood. 1966; Horwath and Duman. 1986).For instance. producingmore subtleetfects during less severecooling to Hanecand Beck (Hanecand Beck, 1960)tbund tl-rat larvae of temperaturesthat this speciesfiequently encounters. A good the Europeancorn borer (Ostrinitt rubilttLis) subjectedto a exampleof suchan effect is the rapid increasein Clmin that thermoperiod(12h at 5 "C alternatedrvith 12h at 15'C) tbr 3 occurswhen D. rneLanogasterJre cooled gradually rather than weeksrvere as tolerantof a -18h exposureto -20'C as those abruptly(Nleats. 1973: Kelty andLee, i999). If t-uturestudies acclimatedct r constant5'C for the sameperiod. employ gradualcooling at ecologicallyrelevant rates, then As in the presentstudy. the capacity to acclimateto low or even more subtle etl-ectsat multiple physiologicalleveis high temperaturesotten decreaseswith age (Rose. 199lt. may be identifiedin flies cooled to even less severelow Czajkaand Lee (Czajkaand Lee. 1990)tbund that. etter the temperatures. lifth dai, of adult lif-e.the capacitl,of D. tnelano,gorrerto cold- l.rardenrapidiy at 0'C decreasedsigniticantlv. Dahlgaard et al. We thank l number clf individuiils r.vhocontributed to the (Dahlgaardet al.. 1995)demonstrated that a 75 min exposure researchpresented herein. Don Hendrixconducted the assays to 36.-5"C increasedthe capacityof voungD. tneltuto.gctsterto tbr carbohydratecryoprotectants. Scott Shreve assistedin surviveexposure to -10.7"C muchmore than it did thatof older a number of thermoperiodexpedments. Special thanks to flies.Although fiequently observed. it remainsunclelr why the Ilartin Feder tbr comments on an earlv drati of the capacitl,of D. tnelunogosterto tcclirnateto low or high nranuscript.This researchwas supportedby grantstiorn NSF temperaturedecreases with lge. (numbersIBN 0090204and IBN-9728573) to R.E.L. Neitlier Hsp70 nor carbohvdrate cryoprotectants contribr,rtedto the capacityof D. tnelano,gusterto cold-harden rapidly.The absenceof Hsp70expression duling coolingis References c()nsistentwith thefinding that neither the rRNA codingfor .\nderson,.\. W. and Harrvood.R. F. ( 1966).Cold tolerance tn Drcsoplila Hsp70nor that codinglor u 13 kDa stressprotein adultt'enraie CLLl.e.r tur.vilis tCoquillet). '\4osquittt Ner'.r 26, l-7. expressedby Surcophaqut'russipolpis in responseto chilling Burton,V., llitchelt,Il. K.. Young,P. and Petersen. N. S. i1988). is expressedby D. rnelunoga:iterrt anv time during the Heat shock protection against cold stress oi Dro.tophila thermoperiodtested (J. Rinehart.personal cotnrnunication). meluutquster. ,\lol. Cell. 8ir.,1.ll, -155{-)-155j. Cold-hardeningofDrosophiia melanogaster 1665
Carey, C. and Hazel.J. R. (1989).Diurnal variationin membrane responsesand temperature adaptations in subtropicaland temperate lipid conrpositionof Sonorandesert teleosts. J. E.rp.Bittl. 147. speciesof Dro.s
Dro.sophiluspecies rcquire cold tolerancc:Qualitative changes of heat-shockproteins at lorv temperatures.PIuttt Plnsiol. Ll7. phospholipids.Eur. .1.Bioclrcm. 252. 608-61 l. 65t-658. Ilamos, ,\.. Raven. N. D. H., Sharp, R. J., Bartolucci, S., Rossi. Somero,G. N., Dahlotl E. and Lin. J. J. (1996).Stenotherms and lI., Cannio, R., Lebbink,,J..vanDeroost,.f., deVos. W. NI. and eurytherms:iVlechanisms establishing thermal optima lnd tolerance Santos,H. ( 1997).Stabilization of cnzvmesaglinst thermal stress ranges.In .\ninul:; urul Tenperuture(.ed. I. Johnstonand A. rrndl'reeze-tlrving bv ntunnosylglycerate. Appl. Ettt'.lvlicrobiol. 63. Bennctt).pp. 53-77. Cambndge:C;rrnbridge Universitv Press. t020--1025. Yiangou, NI., Tsapogas, P., \ikolaidis. N. and Scouras, Z. G. Rose, NI. R. (19911.Evolutiot'Lun' Biolttgv of Ageing.New York: ( 1997).Helt shockgene expression during recovery after transient Oxford UniversitvPress. cold shock in Dro,sophiluuurario (Diptera: Drosophilidae). Sabehat,A., Lurie, S. and Weiss,D. (1998).Expresston of small Ctrttbios92. 9l-98.