Utilizing associational resistance for biocontrol: impacted by temperature, supported by indirect defence

Himanene al. BMC Ecol (2015) 15:16 DOI 10.1186/s12898-015-0048-6 RESEARCH ARTICLE U ilizing associa ional esis ance o biocon ol: impac ed by empe a u e, suppo ed byindi ec de ence Sa i J Himanen1*, Thuy Nga T Bui2, Mengis u M Maja2 and Ja mo K Holopainen2 Abs ac Backg ound: Associa ional he bi o e esis ance is po en ia ed by neighbou ing he e ogenic plan species ha impac a ocal plan ’s a ac ion o he bi o es o he damage ha hey cause. One mechanism o con e associa ional esis ance is belie ed o be exposu e o neighbou -emi ed ola iles, he ecei e s o which ange om in a- and in e speci ic neighbou plan s o highe - ophic-le el insec s. In p e ious s udies he passi e adso p ion o neighbou - emi ed semi ola iles has been epo ed, bu li le is known ega ding he mechanisms and ecological consequences on he ecei e plan and i s associa ed bio a. To u ilize ola ile-based associa ional esis ance o ag icul u al applica- ions, i is impe a i e o know i s e ec i eness unde a ying diu nal empe a u es and whe he he bi o e na u al enemies, p o iding biological con ol, a e impac ed. Mimicking a ying diu nal empe a u es in a labo a o y se -up, we assessed how he i ophic model sys em B assica ole acea a . i alica (b occoli)–Plu ella xylos ella (c uci e special- is he bi o e)–Co esia es alis (endopa asi oid o P. xylos ella) is in luenced by exposu e o he na u al semi ola ile emi e plan Rhododend on omen osum Ha maja. Resul s: Rhododend on omen osum-exposed B. ole acea was less suscep ible o P. xylos ella o iposi ion a bo h nigh - ime (12°C) and day- ime (22°C) empe a u es and less a ou ed and damaged by P. xylos ella la ae a 12°C. Exposu e did no in e e e wi h indi ec de ence, i.e. a ac ion o he na u al enemy C. es alis on hos -damaged, R. omen osum-exposed B. ole acea unde 22°C, while he e was a educ ion in a ac ion (ma ginal p e e ence owa ds hos -damaged B. ole acea) unde 12°C. Conclusions: The abili y o R. omen osum exposu e o ende associa ional esis ance agains an ag icul u ally impo - an B assica he bi o e P. xylos ella wi hou se e ely comp omising he specialis pa asi oid C. es alis hos loca ion encou ages u he s udies on he po en ial o using his na u ally abundan plan o biocon ol. The gene ali y o ou inding on empe a u e as a po en ial egula ing mechanism o he e icacy o semi ola ile emi e -based associa ional esis ance owa ds specialis pes la al damage should be u he s udied in na u al and ag icul u al associa ions. Ou s udy emphasizes he need o de elop echniques o compa e ola iles a he lea e sus ai in e ace and associa e hei appea ance and ecological ole wi h imes o ac i i y and le el o specialisa ion o he bi o es and hei na u al enemies. Keywo ds: Associa ional esis ance, He bi o y, Hos loca ion, Indi ec de ence, Pa asi oids, Semi ola iles, Tempe a u e © 2015 Himanen e al. This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/ publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Backg ound Associa ional esis ance, he p o ec ion o a ocal plan om he bi o y ia neighbou ing a he e ogenic species [1, 2], is an example o how plan –plan in e ac ions Open Access *Co espondence: [email p o ec ed] 1 Na u al Resou ces Ins i u e Finland (Luke), Managemen and P oduc ion o Renewable Resou ces, P oduc ion Sys ems, Lönn o inka u 5, FI-50100 Mikkeli, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle migh a ec plan de ence agains bio ic a acke s. Such neighbou -media ed esis ance is belie ed o ope - a e ia mul iple s a egies: he “ esou ce concen a ion hypo hesis” is based on lowe hos plan densi y and educed accessibili y o hos plan s by a acking he - bi o es in mixed communi ies [3], and he “enemies hypo hesis” p edic s ha na u al enemy abundance is inc eased when ood sou ces and shel e a e p o ided by neighbou s, he eby educing he bi o y [4]. The Page 2 o 12 Himanene al. BMC Ecol (2015) 15:16 “semiochemical di e si y hypo hesis” elies on ola iles dis u bing he bi o e ol ac ion, which is used in hos loca ion and ep oduc ion [1]. Ai bo ne plan ola iles a e impo an in ochemical cues o associa ed spe- cialis species bio a in pa icula [5]. The hos loca ion by insec s [6] and he ac i a ion o a plan ’s inducible de ences o a ac na u al enemies o he bi o e a ack- e s [7] o p o ide indi ec de ence a e pa ly egula ed by ola ile signals. The complexi y o ola ile signal- ling in mul i ophic sys ems is illus a ed by nume ous ecen indings sugges ing no el mechanisms o ola- ile-media ed in e ac ions. Such examples include he passi e adso p ion and e- elease o neighbou s’ de en- si e ola iles wi h semi ola ile cha ac e is ics, i.e. low apo iza ion a e o he gaseous phase, on he su ace o in e speci ic oliage [8]. Associa ional esis ance as media ed by plan ola iles has ecen ly been demons a ed o a numbe o associa- ions [8–11]. Associa ional esis ance in gene al has been u ilised o decades in ag icul u e in he o m o mixed c opping wi h he aim o educing pes damage [12, 13]. Only in app oxima ely 50% o cases, has pes p o ec- ion by neighbou s been e ec i e in mixed cul u es [14]. Speci ic a oma ic plan ola iles ha e o en been es ed wi h success o he pu pose o epelling o masking hos cues (i.e., ol ac o y camou lage) owa ds he bi o es in a ious ypes o ag oecosys ems (e.g., [15, 16]), al hough he ac i e ole o he chemical componen has also been doub ed, and opinions ha e been aised in a ou o companion c ops ac ing as pu e physical ba ie s [17]. Va ia ion in he en i onmen al condi ions and speci ic- i y by he s udied plan -he bi o e sys em can complica e e ealing any gene al unde lying mechanisms p o iding associa ional esis ance alone o in conce wi h o he simul aneously ac ing in e ac ions [9]. Much emains o be e ealed ega ding mul i ophic in e ac ions ha a e media ed by ola iles, pa icula ly on hei en i onmen- al pe sis ence [18–21], po en ial ade-o s be ween he many ecological e ec s o which hey con ibu e [22], and hei co-e olu ion and adap i e alue [23]. The u i- liza ion o associa ional esis ance o biocon ol in he o m o ap c ops [24, 25], epellen o hos odou -mask- ing companion plan s [16], in e c opping [15], o syn- he ic compounds eleased om dispense s [26] migh also bene i g ea ly om an imp o ed unde s anding o he mechanisms egula ing ola ile-media ed associa- ional esis ance. Unde s anding he concu en oles and in e play be ween di e en ola ile-media ed in e ac ions (wi h he same compound changes po en ially impac ing mul iple in e ac ions simul aneously) is essen ial when aiming o u ilize ola iles in pes p o ec ion. Fo exam- ple, a ade-o migh exis i he adso bed compounds, which a e emi ed by he neighbou and con e associa- ional esis ance, in e e e wi h he a ac ion o na u al enemies o supp ess he bi o e p essu e [22]. Thus, when designing biocon ol applica ions based on ola iles, impac s on bo h he he bi o es [16] and hei na u- al enemies [27–29] and, op imally, he pe sis ence o impac s unde abio ic a ia ion as encoun e ed in ield en i onmen s should be conside ed. Neighbou ola- iles migh a iably impac na u al enemies, as ecen ly shown o ladybi ds on po a o exposed o onion ola iles: TMTT was an a ac an , whe eas (E)-ne olidol ac ed as a epellen [30]. I he neighbou -emi ed compounds a e semi ola ile in na u e, he impac migh be e en s onge o mo e p olonged. The s abili y and unc ional dis ance o associa ional esis ance is ano he puzzling issue also when conside - ing ai bo ne plan ola iles in ield condi ions [21, 31]. Because insec ol ac ion is belie ed o be ema kably sensi i e [6], small changes in ola ile signals may e eal o disguise a g ea deal o in o ma ion, making ola ile- based de ences ulne able o in e e ence [19]. Associa- ional esis ance occu s in na u e ac oss a ange o abio ic a ia ion, including diu nally and seasonally a ying empe a u e and high p ecipi a ion as well as d ough episodes and changes in a mosphe ic cons i uen s, such as ca bon dioxide. O hese, he p e ailing empe a u e is well known o i s ole in he con ol o di usion and ola ili y o ola ile compounds and hei longe i y in he a mosphe e [32, 33]. The majo C5 ola ile isop ene is syn hesized and emi ed in g ea e amoun s unde inc easing empe a u es [34]. Fo o he plan ola iles, such as mono e penes and sesqui e penes, he empe a- u e dependence o emission is less ob ious. Bo h physi- ological and physicochemical egula ion akes place in plan oliage and a he lea -a mosphe e in e ace [35]. In addi ion, he abili y o s o e hese compounds o hei p ecu so s in specialized lea s uc u es (e.g. sec e- o y ca i ies o ichomes) and hei inducibili y upon abio ic and bio ic s esses c ea es a iance om di ec esponsi eness o inc easing empe a u e o modelling hei species-speci ic emission po en ial [20, 35]. In he case o sesqui e penoids, he e m semi ola ile has been used o desc ibe hei less ola ile na u e and s ickiness o su aces [36, 37]. Fo such compounds, empe a u e in luences hei e- elease om su aces [37]. Despi e he ole o empe a u e in ola ile emissions, li le is known ega ding how empe a u e in luences he ac ual in o - ma ion-ca ying capaci y o plan ola iles; o example, could associa ional esis ance as media ed by ola iles be subs an ially imp o ed o hampe ed by empe a u e changes? B assica plan s hos a ple ho a o economically damag- ing he bi o es [12, 38]. Thus, we chose B assica ole acea Page 3 o 12 Himanene al. BMC Ecol (2015) 15:16 a . i alica (b occoli, B. ole acea om he e on)–Plu ella xylos ella–Co esia es alis (Haliday) as he ag icul u al i ophic model sys em o be s udied along wi h he well-de ended pe ennial e e g een sh ub Rhododen- d on omen osum Ha maja (p e iously Ledum palus e L.). Rhododend on omen osum is na i e o no he n la i udes and emi s speci ic semi ola iles [palus ol (C15H26O), ledol (C15H26O) and ledene (C15H24)] ha a e capable o being adso bed o neighbou ing oliage [8]; hus, his species ep esen s a s ong semi ola ile emi - e ha is abundan in no he n en i onmen s, making i in e es ing o es o i s po en ial o p o iding associa- ional esis ance o use in biocon ol. The diamond-back mo h P. xylos ella L. (Lepidop e a: Yponomeu idae) is a majo ag icul u al pes o B assica wi h a global dis ibu- ion [38]. The mo h mass mig a es by wind cu en s o no he n egions and is a acked by nume ous p eda- o s and pa asi oids, o which he la al pa asi oids, e.g., Co esia sp. and Diadegma sp. pa asi oid wasps, a e he mos e ec i e o limi ing mo h popula ion g ow h [39]. Co esia es alis (Haliday) (Hymenop e a: B aconidae), a specialis endopa asi oid pa asi izing all o he la al ins a s o P. xylos ella, uses plan -emi ed ola iles in hos sea ching [40]. Al hough we we e in e es ed in ad ancing knowledge on he ecological mechanism and empe a u e sensi i i y o associa ional esis ance, ou p ima y aim was o assess he po en ial o using R. omen osum o biocon ol in his s udy. The speci ic objec i es o ou s udy we e o assess (1) whe he associa ional esis ance owa ds P. xylos ella is con e ed by exposu e o R. omen osum in B. ole acea; (2) whe he R. omen osum-media ed asso- cia ional esis ance is e ec i e bo h a nigh - ime and day- ime empe a u es (12 and 22°C upon es ing); (3) whe he exposu e o R. omen osum in e e es wi h indi- ec de ence, i.e., he a ac ion o he C. es alis pa a- si oid by hos -induced ola iles, o B. ole acea; and (4) whe he he impac o R. omen osum exposu e on indi- ec de ence is a ec ed by empe a u e (12 e sus 22°C) upon exposu e. Expe imen al se -ups o he s udy a e shown in Figu e1 (see “Me hods” o de ails). Resul s Ve i ying he adso p ion o R. omen osum ola iles on oneighbou ing B. ole acea We i s e i ied ha R. omen osum (RT om he e on) exposu e esul ed in semi ola iles being adso bed and e- eleased om B. ole acea oliage, as p e iously epo ed o bi ch oliage [8]. RT b anch emissions we e domina ed by he sesqui e pene alcohol palus ol, which comp ised 43% o he o al ola ile emissions (Fig- u e2a). The o he main compounds ha we e de ec ed we e β-my cene, a omadend ene and ledol (wi h 34, 10 and 9% o he o al emission, espec i ely). Palus ol was de ec ed om emission p o ile o RT-exposed B. ole acea a bo h exposu e empe a u es: 6 and 22°C (Figu e2b). RT-o igina ing ledol and ledene we e emi ed only om RT-exposed B. ole acea plan s ha we e exposed a 22°C. No RT-speci ic compounds we e emi ed om he co - esponding con ol B. ole acea plan s (Figu e2b). The e we e no s a is ically signi ican di e ences be ween con ol and RT-exposed B. ole acea in he emissions o ola iles de ec ed om bo h ea men s, i.e., α-pinene, δ-ca ene and 1,8-cineole (Figu e2b, P>0.05). Responses o P. xylos ella oR. omen osum‑exposed B. ole acea O iposi ion by P. xylos ella was lowe on RT-exposed han con ol B. ole acea plan s (Figu e3, F1,36=13.5, P<0.001). A 22°C, he e we e 38% and, a 12°C, 36% less o iposi ion on RT-exposed plan s. A highe num- be o eggs we e laid when he es ing empe a u e du - ing o iposi ion was 22°C compa ed o 12°C (Figu e3, F1,36=131.6, P<0.001). Plu ella xylos ella la ae p e e ed o eed on he con- ol a he han he RT-exposed B. ole acea lea es a 12°C (Figu e4a, P< 0.05). This p e e ence was main- ained h oughou all o he obse a ion ime poin s om 30min o 24h a e elease. When he exposu e and es ing ook place a 22°C, no signi ican di e ence in he choice o he la ae be ween he con ol and he RT-exposed B. ole acea lea es was de ec ed a any o he obse a ion ime poin s (Figu e4b, P>0.05). The lea a ea ha was consumed by P. xylos ella la - ae was highe in he con ol han in he RT-exposed B. ole acea lea es a 12°C (Figu e5a, =5.75, d =41, P< 0.001). The e was no di e ence in he amoun o lea damage be ween he ea men s a 22°C ( =1.61, d =38, P=0.116). The numbe o eeding holes on RT- exposed B. ole acea was smalle han ha on con ol B. ole acea lea es a 12°C (Figu e5b, =−3.26, d =41, P=0.002), whe eas no di e ence be ween ea men s was ound when he es ook place a 22°C ( =−0.95, d =39, P=0.347). Hos loca ion byC. es alis: ole o R. omen osum exposu e Co esia es alis emales showed no p e e ence o in ac B. ole acea o e RT ola iles (Figu e6, χ2=0.27, P=0.602), and only 33% o he pa asi oids made a choice in his beha iou al es compa ison. The unc ioning o he B. ole acea–P. xylos ella–C. es alis i ophic in e - ac ion was e i ied by he g ea e a ac ion o C. es alis emales owa ds hos -damaged a he han in ac B. ole - acea ola iles (χ2=3.90, P=0.048). The pa asi oids also p e e ed hos -damaged, RT-exposed B. ole acea o e in ac plan s when he exposu e ook place a day- ime Page 4 o 12 Himanene al. BMC Ecol (2015) 15:16 empe a u e (22°C) (χ2=4.83, P=0.028). No p e e ence by C. es alis was obse ed be ween hos -damaged and hos -damaged RT-exposed B. ole acea a his empe a- u e (χ2=1.11, P=0.292). The pa asi oid choosing a es we e g ea e han 60% in all h ee assays wi h hos -dam- aged B. ole acea as an odou sou ce. When using plan s ha we e exposed o RT a 12°C, C. es alis p e e ed hos -damaged RT-exposed B. ole acea plan s o in ac RT-exposed plan s wi h a ma ginal s a- is ical signi icance (Figu e6, χ2=3.81, P=0.067). The pa asi oid had no p e e ence o hos -damaged o e hos - damaged RT-exposed B. ole acea (χ2=0.24, P=0.743). Discussion Ou esul s e ealed ha RT exposu e educed he o i- posi ion and la al eeding o a key pes he bi o e o c u- ci e ous plan s: he diamondback mo h P. xylos ella on B. ole acea. A colde empe a u e (12°C) upon exposu e was in eg al o he inc ease in la al esis ance. In addi- ion, indi ec de ence, e alua ed he e as he a ac ion o he specialis pa asi oid C. es alis by hos -induced ola- iles, emained pe sis en despi e RT exposu e, al hough indica ions o a educed esponse we e ound when he exposu e ook place a a colde empe a u e (12°C). Associa ional esis ance iaR. omen osum: mechanisms o ac ion andpo en ial o biocon ol The p o ision o associa ional esis ance by RT ola iles, as p e iously sugges ed o a na u al neighbou (Be ula sp.) [8], was con i med in his s udy on a B assica c op species and owa ds a damaging global pes : he dia- mondback mo h. In ex ac s o RT, wi h ledene, ledol and palus ol o ming he majo i y o he composi ion, epel- lence has been epo ed owa ds a ious a h opods [41, 42]. Ou inding ha exposu e ia ai alone was su icien o ende inc eased esis ance owa ds bo h adul and 2. He bi o e esponses 1. Ve i ying adso p ion o R. omen osum ola iles on o B. ole acea 3. Pa asi oid esponses R. omen osum ola iles R. omen osum – exposed B. ole acea ola iles Con ol B. ole acea ola iles Exposu e o 24h a 6°C and 22°C Vola iles collec ed a 22°C Exposu e o 8 h a 22°C ollowed by 16 h a 6°C Exposu e o 48h a 12°C and 22°C Co esia es alis a acon es ed a 22°C Combina ions es ed o 22°C exposu es: Combina ions es ed o 12°C exposu es: Expe imen al se -ups o he s udy RT CC Analysis o Plu ella xylos ella o iposi ion O iposi ion o 24h a 12°C and 22°C P. xylos ella la al choice and eeding Exposu e o 24 h a 12°C and 22°C C RT La al es ing o 24h s s s C+Px C RT RT+Px C RT s RT s s a 12°C and 22°C Figu e1 Expe imen al se -ups o he s udy. The s udy consis ed o h ee pa s es ing plan , he bi o e and pa asi oid esponses owa ds R. omen- osum (RT) exposu e. Tempe a u es used in each pa a ied acco ding o he esponse o be es ed, and we e ep esen a i e o bo eal en i on- men a ia ion in ea ly summe nigh - ime and day- ime empe a u es. In Pa 1 ( ola ile analysis), a 6°C nigh - ime minimum empe a u e e sus day- ime 22°C was used in o de o e eal whe he a empe a u e-dependen adso p ion–deso p ion p ocess o RT semi ola iles occu s on B. ole acea. He bi o e o iposi ion in Pa 2 was compa ed a a e age day- ime (22°C) e sus a e age nigh - ime (12°C) empe a u es, wi h equal ini ial RT exposu e (day- ime 22°C ollowed by minimum nigh - ime empe a u e o 6°C). La al choice and eeding was es ed a s abile 12 and 22°C em- pe a u es o allow compa ison o a 24 h pe iod o eeding a nigh - ime e sus day- ime empe a u es along wi h empe a u e- ep esen a i e RT exposu es. In Pa 3, he day-ac i e pa asi oids we e es ed a 22°C, wi h he p io RT exposu e conduc ed a a e age day- ime (22°C) and nigh - ime (12°C) empe a u es. Px = P. xylos ella. Pho os o P. xylos ella and C. es alis: Ja mo Holopainen. Page 5 o 12 Himanene al. BMC Ecol (2015) 15:16 la al s ages o P. xylos ella, wi h no need o he physical applica ion o he compounds, sugges s ha RT could be a pes -supp essi e in e c op o B assica. Ou s udy is no able o de e mine whe he he inc eased esis ance ac s as an ol ac o y, as an con ac cue o bo h. Fu he mo e, he mode o ac ion o RT exposu e leading o associa ional esis ance on B. ole a- cea also emains an open ques ion: he obse ed impac s could s em om RT ola iles ha a e adso bed on o and e- eleased om B. ole acea lea es o om B. ole a- cea ola iles induced by RT exposu e. Because B. ole a- cea in insic ola ile emission was low and he e we e no indica ions o o he compounds being al e ed in he p o ile a e RT exposu e han he RT-speci ic semi ola- iles, which can be adso bed on o oliage [8], he la e Figu e2 Vola ile emission om Rhododend on omen osum (a) and con ol and R. omen osum-exposed B assica ole acea (b). The mean ± SEM emission o indi idual ola ile compounds de ec ed a e shown. B assica ole acea plan s we e exposed o R. omen osum o 24 h a 6 o 22°C, and ola ile emissions we e collec ed pos -exposu e a 22°C. n = 3 in (a) and n = 4 in (b) o bo h exposu e empe a u es. 0 50 100 150 200 250 12°C 22°C O iposion pe plan Tempe a u e Con ol RT-exposed P ea <0.001 P emp<0.001 P ea x emp=0.620 Figu e3 O iposi ion by Plu ella xylos ella on con ol and R. omen osum-exposed B assica ole acea a 12 and 22°C. The alues ep esen he mean numbe o P. xylos ella eggs pe plan ± SEM based on h ee independen expe imen s (n = 4 in each expe imen ). P alues o he main e ec s o ea men and empe a u e and hei in e ac ion a e shown. 80 60 40 20 020406 08 0 30 min 1h 2h 4h 24h % P. xylos ella la ae Con ol RT-exposed 12°C ** ** ** ** ** 80 60 40 20 020406 08 0 30 min 1h 2h 4h 24h % P. xylos ella la ae Con ol RT-exposed 22°C a b Figu e4 Plu ella xylos ella la al choice owa ds con ol e sus R. omen osum-exposed B assica ole acea a 12 and 22°C. The pe cen - age dis ibu ion (mean ± SEM) o la ae in choice es s pe o med a 12°C (a) and 22°C (b) is shown. Each eplica e includes he choices (posi ioning on ei he lea ) o i e indi idual la ae. n = 20–21 in bo h o he independen eplica e expe imen s o which he esul s a e summa ized he e. S a is ically signi ican di e ences in la al choice a e indica ed by as e isks (**P < 0.01). Page 6 o 12 Himanene al. BMC Ecol (2015) 15:16 hypo hesis is no suppo ed. The e o e, RT ola iles a e likely candida es o passi ely p o iding inc eased esis - ance. These pe sis en compounds, which a e emi ed in la ge amoun s by RT, could disguise hos loca ion, o ipo- si ion o eeding s imula ion cues, such as glucosinola e b eak-down p oduc s o he specialis pes P. xylos ella [39]. Al e na i ely, hese compounds migh cause di ec epellence [41, 42]. Thus, ollow-up labo a o y wo k wi h syn he ic palus ol, ledol and ledene is needed o e eal he ac ing mechanism o RT-exposu e-media ed asso- cia ional esis ance. The po en ial impo ance o RT semi ola iles is u he suppo ed by he obse ed empe a u e dependence o he esis ance owa ds la al eeding: a 12°C, he RT spe- ci ic semi ola iles a e s uck on su aces. In [37], an e o o es ablish a i icial b occoli lea es, i.e., a wax laye on a mic oscope slide, o de ec ing adso p ion o induced ola iles, showed how induced sesqui e penes in pa - icula we e adso bed on o he wax and hen e- eleased upon inc easing empe a u e. As epicu icula wax is he su ace encoun e ing RT ola iles unde ae ial exposu e, hei mode o ac ion could in ol e an in e ac ion wi h wax chemicals. Plu ella xylos ella p e e s o o iposi on glossy e sus waxy lea es, and high epicu icula wax bloom makes cabbage mo e esis an o he pes [43]. The chemical composi ion o he wax a ies be ween he ypes, i.e., he p opo ion o alkanes and ke ones is lowe , while ha o a y acids and i e penoids is la ge on glossy wax ype lea es [44]. In addi ion, also P. xylos ella la ae a e impac ed by he wax p ope ies: hey spend mo e ime sea ching and less ime eeding on glossy cab- bages [44]. Thus, one hypo hesis o he mechanism o he pa ly empe a u e-dependen RT-media ed esis - ance in ou s udy is he chemical composi ional changes ha a e induced by RT ola iles in B. ole acea lea wax. In biocon ol, companion c ops wi h epo ed impac s on ag icul u ally impo an he bi o es ha e o en included a oma ic- ola ile-emi ing he bs [45, 46], indi- ca ing ha he po en ial semi ola ile cha ac e is ics ha hei main ola iles possess migh be mo e gene ally impo an o he e ec i eness o associa ional esis ance. Howe e , he e ec i eness o he b in e c opping impac - ing he bi o e abundance a ies [12, 46]. Addi ional a i- abili y complica ing unde s anding o he ole o ola iles o associa ional esis ance is c ea ed by he di e en ial beha iou al pa e ns ha a e likely obse ed o gene al- is e sus specialis he bi o es [12]. In addi ion, he e is ample a ia ion by species in u ilising ola iles in hos loca ion [6] in compa ison o o he cues, such as pheno- ypic di e ences ac ing as physical ba ie s o isual cues [17, 47]. Ou inding ha empe a u e impac s he e ec- i eness o associa ional esis ance emphasizes he need o include a mo e de ailed analysis o changes in ola ile emissions unde na u al empe a u e a ia ions conside - ing hei p ope ies a ec ing hei ola ili y o adso p ion on o oliage [36, 37]. The ola ile p o ile o a plan species is a gene ically egula ed p ope y ha is impo an o be assessed when de eloping ag icul u al applica ions based on associa ional esis ance [15, 16]. Pe sis en indi ec de ence: does high ol ac o y speci ici y a uppe ophic le els p o ec agains in e e ence? Indi ec de ence, assessed he e o he a ac ion o a spe- cialis pa asi oid C. es alis o hos -damaged B. ole acea ola iles, was no in e e ed wi h by RT exposu e a 22°C. The a ac ion was sligh ly educed ollowing he 12°C exposu e, indica ing ha RT exposu e migh become mo e meaning ul a mo ning and la e e ening hou s when he empe a u e dec eases. The g ea e endency o he s udied he bi o e o be in luenced by RT expo- su e a he han i s na u al enemy is suppo ed by a ield s udy in which he e was no di e ence in pa asi ism, bu 0 0.5 1 1.5 2 2.5 12°C 22°C A ea ed (cm2) Con ol RT-exposed *** 0 2 4 6 8 10 12 14 12°C 22°C Numbe o eeding holes Con ol RT-exposed ** b a Figu e5 Lea a ea ed and eeding holes on con ol and R. omen- osum-exposed B assica ole acea. The amoun o lea a ea ed a 12°C and a 22°C (a) and he numbe o eeding holes made a 12°C and a 22°C (b) (mean ± SEM) by i e Plu ella xylos ella la ae o e 24 h on con ol and R. omen osum (RT)-exposed B. ole acea lea es a e shown. n = 20–21 in bo h o he independen eplica e expe imen s o which he esul s a e summa ized he e. S a is ically signi ican di e ences be ween ea men s a e indica ed by as e isks (**P < 0.01, ***P < 0.001). Page 7 o 12 Himanene al. BMC Ecol (2015) 15:16 densi ies o la al and pupal s ages o P. xylos ella we e educed in he mixed c opping o cabbage and co iande [47]. The abili y o a non-hos scen o ac owa ds bo h a he bi o e and i s na u al enemy, has been epo ed o a Yeheb (Co deauxia edulis) plan ex ac [48] which a he same ime a ac ed C. es alis and epelled P. xylos ella. We ound no such a ac ing e ec om RT owa ds C. es alis ha could ha e led o syne gy in P. xylos ella biocon ol. To ou knowledge, he passi e adso p ion o he e ospe- ci ic species emi ed ola iles, and he esul ing associa- ional esis ance has no been conside ed in he con ex o dis u bing indi ec de ence ea lie . Howe e , he abun- dance and o aging e iciency o pa asi oids ha e been s udied in complex e sus less complex en i onmen s [49–51], wi h chemical complexi y hypo hesized o be one in luence o a h opod abundance [50]. The pa asi- oid Diadegma semiclausum en e ed a Sinapis–B assica in e c op as e han a Ho deum–B assica in e c op, bu ook mo e ime o ind hos he e [49], indica ing ha species composi ion is meaning ul, e en hough he co- c op ola iles we e no analysed, pe mi ing a compa i- son o ola ile-media ed and o he cues. In [51], as he habi a complexi y inc eased, s onge hos cues o C. glome a a hos inding we e concluded o be equi ed. In addi ion, in [52], C. glome a a, wi h a wide hos ange compa ed o ha o he mo e specialised Co esia ubec- ula, was less e ec i e in hos inding in mixed cul u es o B. ole acea and po a o. In a labo a o y s udy wi h ola ile in e e ence in oduced by isop ene, ano he pa asi oid- speci ic esponse was ound: isop ene in e e ed wi h he hos loca ion in D. semiclausum bu no in C. ubecula [22]. The mixing o non-hos he bi o e-induced cab- bage ola iles and hos -induced ola iles o bean did no in e e e wi h he a ac ion o specialis p eda o y mi es o p ey [53]. In [54], non-hos ola iles we e e i ied o be sensed by a p edaceous bee le, sugges ing he abili y o non-hos ola iles in pa icula o hampe i ophic in e ac ions. I could be ha mo e specialised pa asi- oids, such as C. es alis in ou s udy, a e less sensi i e o dis u bance han he bi o e na u al enemies wi h a wide hos ange and hus a highe endency o adap a ion. Thus, i will be in e es ing o es in he u u e whe he he specializa ion o he i ophic in e ac ion plays a key ole in he sensi i i y o neighbou plan semi ola ile in e e ence. Tempe a u e asa media o o associa ional esis ance Tempe a u e a ies g ea ly in na u e and in ag oecosys- ems, diu nally, seasonally and egionally, impac ing plan phenology and insec ac i i y and, hus, hei in e ac ions [55]. Nigh - ime empe a u es o en all close o 10°C in 100755025025 50 75 100 % C. es alis pa asi oids n= 100 B. ole acea n=50 In ac n=100 In ac n=100 Hos -damaged n=80 RT-exposed a 12°C n=60 Hos -damaged a 12°C RT Hos -damaged Hos -damaged + RT-exposed Hos -damaged + RT-exposed Hos -damaged + RT-exposed a 12°C Hos -damaged+ RT-exposed a 12°C * * 18 15 10 21 11 24 32 41 22 37 17 21 Figu e6 O ien a ion o pa asi oid Co esia es alis in Y- ube ol ac ome e assays. The selec ion o he indi idually es ed pa asi oids (% o pa a- si oids making a selec ion) owa ds ei he o he odou sou ces in a wo-a m ol ac ome e is shown. In ac , hos -damaged and R. omen osum (RT)-exposed e e o ea men s o B. ole acea plan s. Hos -damaged plan s expe ienced 48 h o Plu ella xylos ella L. la al eeding and RT-exposed plan s neighbou ed a po ed R. omen osum plan o 48 h be o e es ing a he empe a u e shown. The o al numbe o pa asi oids es ed (n) and numbe o indi iduals making a selec ion owa ds he odou sou ces a e also epo ed. S a is ically signi ican di e ences in he selec ions a e indica ed by as e isks (*P < 0.05). Page 8 o 12 Himanene al. BMC Ecol (2015) 15:16 bo eal egions ( he na u al habi a s o RT) in he sum- me . Thus, noc u nal and day-ac i e he bi o es migh be a iably a ec ed by exposu e o RT semi ola iles and hei p esence in ai e sus oliage. Plu ella xylos ella mo hs a e nigh -ac i e, and hei o iposi ion peaks a dusk, bu ligh du ing nigh (as ypical o bo eal summe ) does no inhibi o iposi ion [39]. Inc easing empe a u e has ens he du a ion o he li e cycle o P. xylos ella as well as i s o iposi ion and la al eeding ac i i y [39]. The epo ed equal educ ion in o iposi ion on RT-exposed plan s unde bo h s udied empe a u es sugges s a diu nally pe - sis en inc ease in esis ance owa ds P. xylos ella. O ipo- si ion is ag onomically impo an because i de e mines he ini ial accep ance o he hos plan . Plu ella xylos ella does no o iposi on non-hos plan s as i elies on a a i- e y o chemical and physical hos ecogni ion cues [56]; RT exposu e seems o in e e e wi h one o se e al o hese. Ou esul on esis ance owa ds P. xylos ella la al eeding eme ging a a lowe empe a u e is less bene icial o he plan as P. xylos ella eeding inc eases by empe a- u e [39]. Howe e , he inding is in e es ing wi h ega ds o hypo hesizing a egula ing ole o empe a u e in semi ola ile-media ed associa ional esis ance. The em- pe a u e dependence o he ola iles ac ing migh o e one hypo he ical explana ion o why a oma ic he bs ha a e ich in sesqui e penes ha e o en shown success, bu also ailu es in p o iding esis ance in mixed c opping [12, 14–16]. To e eal whe he his is he case, he lea -ai in e ace dynamics o he speci ic ola iles ac ing should be s udied in de ail unde empe a u e egimes. Ano he conside a ion in he long- e m e ec i eness o associa ional esis ance as con e ed by he neighbou ’s ola iles is ha he non-s a ic na u e o he signal, medi- a ed by empe a u e, migh also help slow bo h he adap- a ion o he bi o es o he scen [57] and he e olu ion o b eaking hos plan esis ance by he bi o es. Tempe a- u e con ol ende s he signal mos e ec i e a ce ain ime pe iods only, pa ly esembling he mode o unc- ion o inducible de ences [23]. Bo h mechanisms migh be ecologically and e en e olu iona ily suppo ed by he dynamic na u e o signal appea ance in addi ion o sa ing esou ces om he ecei e . The o e all e ec o empe a u e a ia ions on he plan -he bi o e dynamics in a mul i ophic sys em depends on he compa a i e esponses o he bi o es and hei na u al enemies. The empe a u e sensi i i y o di - e en ophic-le el insec s a ies [58]; pa asi oids and p eda o s a e gene ally assumed o be mo e esponsi e han hei p ey, as hey ely mo e on mobili y in p ey sea ching and a e physiologically mo e esponsi e o empe a u e. Upon empe a u e changes, he bi o es, on he o he hand, ace bo om-up changes in plan hos quali y, show di ec physiological and ep oduc ional esponses [55], plus a e con on ed by al e ed op-down con ol om na u al enemies [59]. The pa asi oid s udied he e, C. es alis, o iposi s upon pho ophase wi h op i- mal empe a u e o pa asi ism a 20–35°C [60]. Thus, ou inding ha he hos loca ion by C. es alis was no in e e ed wi h by RT exposu e a 22°C sugges s ha using RT o P. xylos ella biocon ol should no comp o- mise he abili y o C. es alis o loca e hos on B. ole acea. The sligh ly educed a ac ion owa ds plan s ha we e exposed o RT a 12°C sugges s, howe e , ha exposu e o RT migh a ec he in e ac ion a low empe a u es such as du ing ea ly mo ning. Conclusions The abili y o RT exposu e, ea lie disco e ed o ende associa ional esis ance in na u e, o in luence B assica he bi o e o iposi ion and la al eeding encou ages u - he es ing o i s po en ial o biocon ol. The gene ali y o ou inding on he egula o y ole o empe a u e on associa ional esis ance should be es ed in he u u e on na u al and ag icul u al plan sys ems a ying in plan ola ile composi ion (acknowledging he di e en ial ole o semi ola ile e sus mo e ola ile compounds). Addi ionally, he le el o specialisa ion and he use o he isual, ol ac o y and chemo ecep o cues o a ge ed he bi o es should be s udied oge he o dis inguish he ole o ola ile-based and o he neighbou ing po en ia ed de ences. Finally, ou s udy emphasizes ha o de elop- ing biocon ol applica ions based on associa ional esis - ance, mo e a en ion needs o be paid o he compa ibili y o plan de ences wi h he bi o e diu nal beha iou and imes o ac i i y as well as ha o hei na u al enemies. Me hods Plan s, he bi o es andna u al enemies B assica ole acea a . i alica (b occoli) c . Lucky seed- lings we e g own om seed in 0.66 l po s (4 plan s in each, mix u e 2:1:1 o e ilized compos , NPK 100:30:200mgl−1: B2 pea NPK 110:40:220mgl−1: sand) in a g eenhouse ( empe a u e 20/16°C day/nigh ) and used a app oxima ely 3weeks o age o expe imen s. R. omen osum (RT) b anches ha we e collec ed om a na u al habi a in Suonenjoki, Finland (62°38′42.683″N, 27°3′55.383″E) and s o ed a 6°C p io o use i neces- sa y, we e used in he exposu e ea men s. Po ed g een- house-g own RT plan s (seed o igin: na u al habi a in Neulaniemi, Kuopio, g own in 3:1 pea :sand) o ci ca 20 cm in heigh we e used in he Y- ube ol ac ome e es s. The P. xylos ella la ae and adul s and C. es alis pa asi- oids ha we e used in he expe imen s o igina ed om labo a o y popula ions main ained on B. ole acea a . i alica (20–25°C empe a u e and 16L:8D pho ope iod) a he Uni e si y o Eas e n Finland. Page 9 o 12 Himanene al. BMC Ecol (2015) 15:16 Vola ile collec ion andanalysis Fo exposu e, ou B. ole acea po s (wi h ou 21-days-old plan s/po ) we e enclosed in insec cages (33×33×60cm, wi h wo sides co e ed wi h a ab ic mesh) wi h 15 cu hibe na ing RT b anches (collec ed igh be o e he expe imen s om he na u al habi a ) ha had hei s ems in wa e - illed 250-ml decan e bo - les. Con ol cages had an equal se -up bu hos ed no RT b anches. The cages we e placed a 6°C (cold oom) and 22°C (a labo a o y ume hood) unde low ligh (app oxi- ma ely 150µmol m−2s−1). The colde nigh empe a- u e was used o mimic minimum bo eal ea ly summe empe a u es. A e 24h, he plan s we e emo ed om exposu e, and he ola iles we e collec ed a 22°C using a dynamic bag enclosu e me hod. The RT b anch emis- sions we e collec ed om h ee b anches a 22°C o de e mine hei cha ac e is ic emission p o ile. Fo ola ile collec ion, he en i e shoo o a i e-lea s age b occoli plan o RT b anch was enclosed inside p e-cleaned (hea ing o 1h a 120°C) mul ipu pose cook- ing bags (polye hylene e eph hala e (PET), 25×55cm in size, Look, Te inex L d, UK), and he bag opening was ied a ound he base o he s em wi h a piece o hin ga - den wi e. A e he enclosu e o he plan , ai ha had been il e ed h ough cha coal and a MnO2 sc ubbe we e pumped in o bags h ough an opening in one o he op co ne s o he PET bags a a low a e o 600mlmin−1 o 10min o lush ou esidual con aminan s. The sam- pling line was inse ed in o ano he co ne o he bag and ied wi h a wi e. A e ixing he lines, he in low a e was educed o 300mlmin−1, and he ola ile emissions we e sampled in s ainless s eel ubes ha we e illed wi h 150mg o Tenax TA adso ben (Supelco, Belle on e, PA USA) o 30min a he a e o abou 200mlmin−1 wi h a acuum pump (Thomas 5002 12V DC). An equal ligh le el (app oxima ely 300µmolm−1s−1) was p o ided by placing wo lamps (Li al Shu le Plus, Li al Oy, Sipoo, Finland wi h Os am Delux F, 24W luo escen lamps, Os am AG, Munich, Ge many) on bo h sides o each plan . Sample ubes we e sealed wi h Te lon-coa ed b ass caps immedia ely a e collec ion and s o ed in he e ig- e a o un il analysis. The plan biomass was o en-d ied (60°C) o de e mine he d y weigh s o he plan s. The plan ola ile samples we e analysed by a gas ch oma og aph-mass spec ome e (GC ype 6890, MSD 5973: Hewle Packa d; Wilming on, DE, USA) as desc ibed in [61]. The apped compounds we e deso bed wi h a he mal deso p ion uni (Pe kin-Elme ATD400 Au oma ic The mal Deso p ion sys em) a 250°C o 10min, c yo ocused a −30°C and injec ed on o a HP-5 capilla y column (50m×0.2mm i.d.×0.5μm ilm hickness, Hewle -Packa d) wi h helium as a ca ie gas. The o en empe a u e p og amme was held a 40°C o 1min and hen inc eased o 210°C a a a e o 5°C min−1 and inally o 250°C a a a e o 20°C min−1. The com- pounds we e iden i ied and quan i ied by compa ing he spec a o ex e nal s anda ds o a ailable compounds and he Wiley lib a y (John Wiley & Sons, L d, Chiches- e , UK). The emission a es we e calcula ed as emission pe d y weigh (DW) pe hou . O iposi ion byP. xylos ella mo hs RT exposu e was iden ical in he se -up as used o he plan s o ola ile analysis. He e, he exposu e was main- ained (as well as he co esponding con ol plan s wi h no RT b anches inside cages) in ume hoods a 22°C o he p e ious day (8h) ollowed by 16h a 6°C (cold oom, ep esen ing ypical ea ly summe nigh - ime minimum empe a u e) un il he s a o he o iposi ion expe i- men . All o he plan s we e equally exposed o assu e equal ini ial amoun s o po en ially adso bed semi ola- iles by exposu e, e ealing solely day- ime e sus nigh - ime empe a u e di e ences o o iposi ion. The o iposi ion in con ol and RT-exposed B. ole a- cea was expe imen ed in 2.6m3 empe a u e-con olled g ow h chambe s (desc ibed in [62]) a empe a u es o 12 and 22°C (mimicking a e age bo eal nigh - ime and day- ime empe a u es, espec i ely) and a 22L:2D pho- ope iod wi h app oxima ely 250µmolm−2s−1 ligh le el. One con ol and one RT-exposed B. ole acea po we e enclosed in each o he ou eplica e insec cages unde bo h o he empe a u e ea men s. Thi y P. xylos ella mo hs we e andomly selec ed by he expec a ion o he 50:50 emale:male a io (popula ion well s abilised o assu e ha bo h sexes we e p esen and allowing ma - ing as in na u e) [48] and eleased in o each cage. A e 24h, he B. ole acea plan s we e emo ed om cages and s o ed a 6°C un il egg coun ing. Th ee independen ep- lica e expe imen s we e ca ied ou . Hos choice and eeding byP. xylos ella la ae Fo he bi o e choice es s, he exposu e o B. ole acea lea es o RT ola iles was conduc ed in 1.5-l glass con- aine s holding wo 30-ml glass ials app oxima ely 3cm apa , as in [8]. Ten o ele en independen sys ems we e buil pe eplica e expe imen ( wo independen epli- ca e expe imen s we e conduc ed o bo h empe a u es), each ha ing wo B. ole acea lea es in one ial and ei he an RT b anch (RT-exposu e ea men ) o a B. ole acea lea (con ol) in he o he ial. The exposu es and he ol- lowing choice es s ook place o 24h o each pa in a labo a o y ume hood (22°C) and a cold oom (12°C), mimicking a e age bo eal ea ly summe day- ime and nigh - ime empe a u es, espec i ely. Fo choice es s, he pe ioles o B. ole acea lea es (one each om con ol and exposed sys em se -ups)