Juvenile hormone titers and caste differentiation in the damp-wood termite Hodotermopsis sjostedti (Isoptera, Termopsidae) Richard Cornette 1 , Hiroki Gotoh, Shigeyuki Koshikawa 2 , Toru Miura Laboratory of Ecological Genetics, Graduate School of Environmental Science, Hokkaido University, Sapporo 060-0810, Japan article info Article history: Received 1 October 2007 Received in revised form 17 April 2008 Accepted 18 April 2008 Keywords: Termite JH Pyriproxyfen LC-MS Caste differentiation abstract Termites are social insects, presenting morphologically distinct castes, performing specific tasks in the colony. The developmental processes underlying caste differentiation are mainly controlled by juvenile hormone (JH). Although many fragmentary data support this fact, there was no comparative work on JH titers during the caste differentiation processes. In this study, JH titer variation was investigated using a liquid chromatography-mass spectrometry (LC-MS) quantification method in all castes of the Japanese damp-wood termite Hodotermopsis sjostedti, especially focusing on the soldier caste differentiation pathway, which was induced by treatment with a JH analog. Hemolymph JH titers fluctuated between 20 and 720 pg/ml. A peak of JH was observed during molting events for the pseudergate stationary molt and presoldier differentiation, but this peak was absent prior to the imaginal molt. Soldier caste differentiation was generally associated with high JH titers and nymph to alate differentiation with low JH titers. However, JH titer rose in females during alate maturation, probably in relation to vitellogenesis. In comparison, JH titer was surprisingly low in neotenics. On the basis of these results in both natural and artificial conditions, the current model for JH action on termite caste differentiation is discussed and re-appraised. & 2008 Elsevier Ltd. All rights reserved. 1. Introduction Insect societies are complex structures emerging from the cooperation of a large number of individuals. This phenomenon of cooperation constitutes an important factor in the evident ecological success of social insects. A key for efficient cooperation is the division of labor, with conspicuous castes that accomplish different behavioral tasks (Wilson, 1971). Division of labor is often accompanied by morphological specializations in relation to specific tasks. This distinct polymorphism of social insect castes is subject to environmental factors and results from differential gene expression, it can consequently be considered as a case of polyphenism (Evans and Wheeler, 1999; Miura et al., 1999; Miura, 2001, 2004; Scharf et al., 2003). All members of the society present a similar genetic background, but in response to environmental factors or pheromones, the hormonal status of each individual will be modified and induce gene expression, leading to a specific developmental pathway (Noirot, 1991; Nijhout, 1999; Evans and Wheeler, 2001). In this scenario, hormonal regulations play a central role for caste differentiation, and juvenile hormone (JH) was actually shown to control caste ontogeny in several social insects (Hartfelder, 2000; Hartfelder and Emlen, 2005). For the implications of JH control in social insect polyphenism, the case of the highly dimorphic ant genus Pheidole is exemplified. In these ants, high JH levels during embryonic development induce the differentiation of queen caste, whereas a JH sensitive period exists during post-embryonic development, which will determine the differentiation into workers or alternatively into huge soldiers with a heavily armored head (Nijhout, 1994; Sameshima et al., 2004). Among social insects, termites present one of the most complex and conspicuous caste systems, probably because termites are hemimetabolous insects and most of the castes are derived from immature stages (Thorne, 1996; Roisin, 2000). Many termite species also exhibit a very flexible caste differentiation system, as it is the case for the Japanese damp-wood termite Hodotermopsis sjostedti, which was used in the present study. In this species, caste differentiation pathway is linear. The seventh larval instar assume the main part of the work in the society, and such individuals were consequently considered as false workers or pseudergates (Grasse ´ and Noirot, 1947). Pseudergates are totipo- tent and can differentiate into soldiers through the transient stage presoldier, into alate imagoes through a single nymphal stage, or into neotenics (reproductive castes not derived from winged forms) as well (Miura et al., 2000, 2004). Stationary molts of ARTICLE IN PRESS Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/jinsphys Journal of Insect Physiology 0022-1910/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jinsphys.2008.04.017 Corresponding author. Tel./fax: +8111706 4524. E-mail address: [email protected] (T. Miura). 1 Present address: Anhydrobiosis Research Unit, National Institute of Agro- biological Science, 1–2 Owashi, Tsukuba, Ibaraki 305-8634, Japan. 2 Present address: Laboratory of Molecular Biology, University of Wisconsin Madison, 1525 Linden Drive, Madison, Wisconsin 53706, USA. Journal of Insect Physiology 54 (2008) 922– 930
Termites are social insects, presenting morphologically distinct castes, performing specific tasks in the
colony. The developmental processes underlying caste differentiation are mainly controlled by juvenile
hormone (JH). Although many fragmentary data support this fact, there was no comparative work on JH
titers during the caste differentiation processes. In this study, JH titer variation was investigated using a
liquid chromatography-mass spectrometry (LC-MS) quantification method in all castes of the Japanese
damp-wood termite Hodotermopsis sjostedti, especially focusing on the soldier caste differentiation
pathway, which was induced by treatment with a JH analog. Hemolymph JH titers fluctuated between
20 and 720 pg/ml. A peak of JH was observed during molting events for the pseudergate stationary molt
and presoldier differentiation, but this peak was absent prior to the imaginal molt. Soldier caste
differentiation was generally associated with high JH titers and nymph to alate differentiation with low
JH titers. However, JH titer rose in females during alate maturation, probably in relation to
vitellogenesis. In comparison, JH titer was surprisingly low in neotenics. On the basis of these results
in both natural and artificial conditions, the current model for JH action on termite caste differentiation
is discussed and re-appraised.
& 2008 Elsevier Ltd. All rights reserved.
1. Introduction
Insect societies are complex structures emerging from thecooperation of a large number of individuals. This phenomenonof cooperation constitutes an important factor in the evidentecological success of social insects. A key for efficient cooperationis the division of labor, with conspicuous castes that accomplishdifferent behavioral tasks (Wilson, 1971). Division of labor is oftenaccompanied by morphological specializations in relation tospecific tasks. This distinct polymorphism of social insect castesis subject to environmental factors and results from differentialgene expression, it can consequently be considered as a case ofpolyphenism (Evans and Wheeler, 1999; Miura et al., 1999; Miura,2001, 2004; Scharf et al., 2003). All members of the societypresent a similar genetic background, but in response toenvironmental factors or pheromones, the hormonal status ofeach individual will be modified and induce gene expression,leading to a specific developmental pathway (Noirot, 1991;Nijhout, 1999; Evans and Wheeler, 2001). In this scenario,
ll rights reserved.
.
, National Institute of Agro-
-8634, Japan.
ogy, University of Wisconsin
06, USA.
hormonal regulations play a central role for caste differentiation,and juvenile hormone (JH) was actually shown to control casteontogeny in several social insects (Hartfelder, 2000; Hartfelderand Emlen, 2005). For the implications of JH control in socialinsect polyphenism, the case of the highly dimorphic ant genusPheidole is exemplified. In these ants, high JH levels duringembryonic development induce the differentiation of queen caste,whereas a JH sensitive period exists during post-embryonicdevelopment, which will determine the differentiation intoworkers or alternatively into huge soldiers with a heavily armoredhead (Nijhout, 1994; Sameshima et al., 2004).
Among social insects, termites present one of the mostcomplex and conspicuous caste systems, probably becausetermites are hemimetabolous insects and most of the castes arederived from immature stages (Thorne, 1996; Roisin, 2000). Manytermite species also exhibit a very flexible caste differentiationsystem, as it is the case for the Japanese damp-wood termiteHodotermopsis sjostedti, which was used in the present study. Inthis species, caste differentiation pathway is linear. The seventhlarval instar assume the main part of the work in the society, andsuch individuals were consequently considered as false workers orpseudergates (Grasse and Noirot, 1947). Pseudergates are totipo-tent and can differentiate into soldiers through the transient stagepresoldier, into alate imagoes through a single nymphal stage, orinto neotenics (reproductive castes not derived from wingedforms) as well (Miura et al., 2000, 2004). Stationary molts of
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930 923
pseudergates into pseudergates and regressive molts fromdifferentiated nymphs to undifferentiated pseudergates have alsobeen reported (Koshikawa et al., 2001).
As supported by the various previous studies, JH plays acentral role also in termite caste determination. Early experi-ments showed that implantation of active corpora allata(the JH-producing endocrine glands) from adult cockroaches orsexually functional termites into pseudergates of Kalotermes
flavicollis could induce soldier differentiation, whereas corporaallata from pseudergates or nymph did not produce this effect(Luscher, 1958; Lebrun, 1967). On the basis of Luscher’s (1958)experiments, Nijhout and Wheeler (1982) proposed a model fortermite caste differentiation, in which three JH-sensitive periodswere defined during the intermolt of the pseudergate instar forsuccessively sexual characters, non-sexual adult characters andsoldier characters (Fig. 1). According to this model, a high JH titerthrough these three sensitive periods would induce soldierdifferentiation, whereas alate production would appear in acontext of continuously low JH titer. Alternatively, a high JH titerat first and then lower during the last period would induce thepseudergate molt, and the opposite pattern the differentiation ofneotenics (Fig. 1).
Other supporting evidences for this model are the experi-mental data using various juvenile hormone analogs (JHA) toinduce soldier caste differentiation after the treatment onpseudergates or workers (Hrdy and Krecek, 1972; Howard andHaverty, 1979; Lelis and Everaerts, 1993). Such JHA treatment hasalso been shown to induce soldier differentiation in H. sjostedti
(Ogino et al., 1993; Koshikawa et al., 2005; Cornette et al., 2006).Interestingly, the JHA treatment of nymphs at various timingsduring the nymphal intermolt period induced intercastes withdifferent degrees of soldier characters mixed with adult characters(Miura et al., 2003).
Until recently, only sparse information was available about JHtiter in termites and for a long time JH production has been justinferred from data on the size of corpora allata (Noirot, 1969;Noirot and Bordereau, 1991). However, with the development ofsuitable techniques, the actual JH production in termites has beeninvestigated with various methods, such as radiochemical assayand GC-MS (Greenberg and Tobe, 1985; Park and Raina, 2004;Brent et al., 2005). The general features emerging from all theseinvestigation are low JH levels in workers or pseudergates andsoldiers, and high JH levels in presoldiers and in the maturingqueens. Various reports indicate that JH levels rise during the
Fig. 1. Model of Nijhout and Wheeler for the JH-mediated control of caste
determination in termites. Three JH-sensitive periods were postulated for sexual
characters, non-sexual adult characters and soldier characters, respectively (black
bars). Each curve represents a hypothetical profile of JH titer changes during the
course of the pseudergate instar, leading to differentiation of the indicated caste.
(Redrawn from Nijhout and Wheeler, 1982).
soldier differentiation process (Park and Raina, 2004; Liu et al.,2005; Mao et al., 2005). These data, associated to inductionexperiments with JHA, consequently support the need for high JHlevels in the soldier differentiation component of the Nijhout andWheeler (1982) model. However, no direct data on JH levelsduring the differentiation process of pseudergates and reproduc-tives is available to validate the model.
In the present study, the adaptation of a direct and rapidquantification method using liquid chromatography-mass spec-trometry (LC-MS) allowed the investigation of hemolymph JH titerin every caste of H. sjostedti. Induction of the presoldier molt byJHA treatment has also allowed us to monitor the intrinsic JH titerduring the course of the developmental transition to presoldier,and these data were compared to natural changes in JH titerduring different developmental transitions. Our results fit themodel of Nijhout and Wheeler (1982) for soldier and alate castedifferentiation, but we found an inverse pattern for JH titer duringpseudergate differentiation.
2. Materials and methods
2.1. Insects
Colonies of H. sjostedti were sampled from decayed fallen treesin the evergreen forests of Yakushima Island, Kagoshima pre-fecture, Japan, in May 2005 and May 2006. They were kept in thelaboratory as stock colonies in plastic containers at approximately25 1C under constant darkness. The rotten wood containing thecolonies was supplemented with damp pine tree woodcuts, thusallowing colonies to be kept in the laboratory for several years.Summer individuals (pseudergates) were collected within 2months after colony sampling, until the swarming period. Winterindividuals were collected between 7 and 10 months or between18 and 22 months after colony sampling.
2.2. Hormonal treatment and induction of presoldier differentiation
Pseudergates were isolated in groups of 10 individuals in Petridishes and fed with filter paper. They were treated with a topicalapplication of 5 mg pyriproxyfen (juvenile hormone analog;Sumitomo chemical, Osaka, Japan) diluted in 5 ml acetone perindividual. Presoldier molting was normally induced 2 weeks aftertreatment (Ogino et al., 1993; Cornette et al., 2006). Somepseudergates were also treated with a topical application of30mg JH III (Sigma, St Louis, MO, USA) diluted in 5ml acetone perindividual. However, even at this concentration, JH III did notinduce the differentiation to presoldiers.
2.3. Hemolymph collection and JH extraction
Individuals from each natural caste, premolting nymphswith swollen wing buds (just before imaginal molt) and whitepresoldiers (taken within 1 day after molting) were collected fromstock colonies. Premolting pseudergates engaged in stationarymolting (white with flattened abdomen) and pseudergates takenjust after the stationary molt (with still unsclerotized cuticle)were collected from old stock colonies (more than 6 months aftersampling in the field), which produced no more new soldiers. Weconfirmed the fate of such individuals randomly sampled fromthose stock colonies and all of them underwent a stationary molt(n ¼ 20). JHA-treated pseudergates were taken at 24 h, 3, 7, 10, and14 days after the treatment and JHA-induced presoldiers, takenon the day of ecdysis or 7 days later, were also sampled. Finally, JHIII-treated pseudergates were collected at 24 h and 7 days after
Fig. 2. Mass spectra of standard JHIII in methanol (A) and of intrinsic JHIII from a pseudergate hemolymph sample (B). Ions detected after JHIII ionization were [M+Na]+,
[M+H]+, [M�CH2O]+ and [M�OH]+ at m/z 289, 267, 235 and 249, respectively.
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930924
treatment, in order to evaluate the rate of exogenous JH IIIdegradation in comparison to the rate of JHA degradation.
Termites were anaesthetized on ice and hemolymph wascollected from a section at the base of the legs, with a glass Pasteurpipette. Pooled aliquots of 60ml hemolymph were stored into glassvials at �80 1C until use. Hemolymph from several individuals hadto be pooled, as the amount from a single individual was very small.For accurate measurement of individual hemolymph volumes, eachindividual was weighted after anaesthesia and the collectedhemolymph volume was determined with a micropipette. The JHextraction protocol was modified from Westerlund and Hoffmann(2004). Briefly, each 60ml aliquot of hemolymph was blown into540ml methanol/isooctane (1:1, v/v) containing 30 ng fenoxycarb(Wako chemical Co., Ltd., Japan) as an internal standard. Themixture was vortexed for 20 s and allowed to stand at roomtemperature for 30 min, before centrifugation at 8500g for 15 min.The isooctane phase was transferred into a new glass vial, themethanol phase was vortexed and centrifuged at 10,000g for 30 minand then combined with the isooctane phase. The resulting mixturewas stored at �20 1C or concentrated down to 20ml and transferredto an autosampler vial for immediate analysis.
Five ml from each 20ml concentrated sample was separated on a150�2 mm2 C18 reverse-phased column (YMC-Pack Pro C-18.5 mm, YMC Co. Ltd. Japan) protected by a guard column(YMC-Pack Pro, sphere ODS, YMC Co. Ltd. Japan) with gradientelution of water/methanol (0–15 min 80–100% methanol,15–20 min 100% methanol) at a flow rate of 0.2 ml/min, utilizingan Agilent 1100 HPLC system with autosampler.
MS analysis was performed by electrospray ionization (ESI) inthe positive mode on a microTOF-HS (Brucker Daltonik GmbH,Bremen, Germany) under the following conditions: the electro-spray capillary was set at 4.5 kV and the dry temperature was200 1C. The nitrogen pressure was 1.2 Bar for the nebulizer and thedrying gas nitrogen flow-rate was 4 L/min. Ionization of standardJH III generated [M�CH2O]+, [M�OH]+, [M+H]+ and [M+Na]+ ions(Fig. 2A). In hemolymph samples, [M+H]+ and [M+Na]+ were themajor ions observed (Fig. 2B). Quantification of JH III, fenoxycarband pyriproxyfen was therefore performed by monitoring the[M+H]+ and [M+Na]+ ions. A calibration curve of JH III (Sigma,St Louis, MO, USA) was made, containing the same concentrationof the internal standard fenoxycarb as in hemolymph samples. JHIII titer from each sample was then calculated after analysis of thechromatogram data on the QuantAnalysis software (BrukerDaltonics K.K., Yokohama, Japan). Under these conditions, thedetection limit estimated for standard JH III samples in methanolafter direct analysis was 5 pg/injection. For the confirmation ofthis limit, we measured samples of 5 pg/injection and actuallyobtained the measurements of 14.574.5 pg/injection (mean7s.d.,n ¼ 9). These measurements overestimated JH concentration nearthe detection limit, but these figures were far below the range ofJH content in our samples (750–9000 pg/injection).
3. Results
3.1. Hemolymph volume and caste differentiation
During the collection of hemolymph for JH extraction, amarked variation in the hemolymph volume was observed among
ARTICLE IN PRESS
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930 925
castes. Hemolymph was expressed from the body by pressing onthe abdomen until fat body appeared from the wound. Conse-quently, we consider that the major part of free hemolymphvolume was obtained from each individual. The approximatevolume per individual deduced from pooled hemolymph collec-tions was relatively low, between 2 and 4ml in pseudergates,soldiers and alates (Fig. 3A). In order to validate the hemolymphvolumes (Fig. 3A), we performed more accurate individualmeasurements, and similar values were obtained (Fig. 3C). JHAtreatment did not affect this volume during the first few days. Incomparison, a clear augmentation of hemolymph volume, up to10–14ml per individual was observed during the molting period.Since only a slight diminution of individual body weight wasobserved during this period (Fig. 3B), such an increase ofhemolymph volume was not correlated to body weight. Hemo-lymph volume rose during the 5 days prior to the molt and wasstill high on the day following ecdysis, except for a drop of 2–3 ml,probably corresponding to the loss of molting fluid, was observed.
Fig. 3. Variations of hemolymphatic volume and body weight in relation to the
caste or to the developmental stage. (A) Estimated volume of hemolymph per
individual deduced from the total volume of hemolymph collected for each caste.
The number above each bar corresponds to the total number of individual bled to
obtain the value. (B) Changes in the individual body weight during induced soldier
differentiation. Error bars: standard deviation. Total number of measurements is
indicated over each error bar. (C) Validation of the values obtained by
approximation in pooled samples (A): The volume of hemolymph was measured
for each individual in pseudergates and JHA-treated insects prior to the presoldier
molt. In all cases, the mean value is presented with error bars representing
standard deviation and the total number of individual measurements is indicated
over each error bar. PE: pseudergates; before SM: white individuals taken within 5
days before stationary molt; after SM: individuals taken within 1 day after
stationary molt; JHA24h-14d: JHA-treated individuals taken from 24 h to 14 days
after treatment; PS 0d: natural presoldiers taken within 1 day after molt; PS:
natural presoldiers; S: soldiers; N: nymphs; pmN: premolting nymphs taken
within 1 day before imaginal molt; A: alates; Neo: neotenics.
This rise of hemolymph volume during the molt events wasobserved in stationary molts and during soldier or alate castedifferentiation as well. Although not engaged in any immediatemolting process, nymphs presented an unusually high volume ofhemolymph. It was also the case for presoldiers and neotenics,with a hemolymph volume more than two-fold of the valueobserved in pseudergates.
3.2. JH titer during natural caste differentiation
JH III was the only JH homolog detected in the hemolymph ofH. sjostedti and the hemolymph JH titer fluctuated between 20 and720 pg/ml. Newly collected colonies in May contained a largeproportion of soldiers and sometimes nymphs. Reared in labora-tory, these colonies produced alates in July, reflecting a swarmingduring summer in natural conditions. The JH titer of pseudergatesfrom such colonies was about 3437130 pg/ml (mean7s.d., n ¼ 7)(Fig. 4). In nymphs, the JH titer was lower and subsequently itdropped as low as 57724 pg/ml (mean7s.d., n ¼ 6) in premoltingnymphs with swollen wing buds (within 1 day prior to ecdysis).Several days after the molt to alates (with darkened and completelysclerotized cuticle), the JH titer of males was back to a value similarto that of nymphs, with 197730 pg/ml (mean7s.d., n ¼ 3).However, female alates exhibited a strong JH production and theirJH titer reached 599756 pg/ml (mean7s.d., n ¼ 3). In comparison,the JH titer of mature neotenics was about 85760 pg/ml(mean7s.d., n ¼ 6) in males and females as well (Fig. 4). Thisvalue was surprisingly low, since investigated neotenic individualswere engaged in reproductive tasks and females were laying eggs.
In pseudergates, collected from old stock colonies in winter,the mean JH titer was about 175794 pg/ml (mean7s.d., n ¼ 6),which is nearly the half of JH titer observed in pseudergates fromnewly collected colonies in May (Fig. 5). Stationary molts wereinvestigated in such colonies, which were no longer producingnew presoldiers. Individuals engaged into the process of sta-tionary molt were recognized by their white and flattenedabdomen, indicating that they had undergone gut purge. Suchindividuals would complete ecdysis within 5 days (Cornette et al.,2007). In those premolting individuals, the JH titer was relativelyhigh with 507774 pg/ml (mean7s.d., n ¼ 5) (Fig. 5). Within 1 dayafter ecdysis, the JH titer dropped to 2907150 pg/ml (mean7s.d.,n ¼ 6), which was a value still higher than the JH titer observed inpseudergates.
3.3. JH titer during induced and natural soldier caste differentiation
In comparison to natural pseudergates from stock colonies,intrinsic JH titer declined sharply after JHA treatment, to reach amean of 41719 pg/ml (mean7s.d., n ¼ 6) 3 days after treatment(Fig. 5). Subsequently, JH production rose again and JH titerexceeded the value of pseudergates to reach a peak at4307220 pg/ml (mean7s.d., n ¼ 4) on day 14 after treatment,just prior to the presoldier molt. In JHA-induced presoldiers, theJH titer dropped rapidly after the ecdysis to a value similar to thatobserved in pseudergates, and this state was maintained at leastfor the first week of presoldier life. In comparison, the drop of JHtiter after ecdysis was less accentuated in naturally formedpresoldiers and was rapidly followed by a new increase, up to5067225 pg/ml (mean7s.d., n ¼ 3) (Fig. 5). This increase of JHtiter in natural presoldiers occurred between the second day afterecdysis and 5 days before the soldier molt, when the white andflattened abdomen indicated that gut purge was finished. Finally,in differentiated mature soldiers, JH titer fluctuated around2977180 pg/ml (mean7s.d., n ¼ 5), which was a relatively highvalue compared to pseudergates.
Fig. 4. JH titers during the differentiation of reproductive castes. Abbreviations are
the same as in Fig. 3; except Neo (~,#): neotenic females and males, respectively;
A (~,#): alate females and males, respectively. JH titers are shown separately for
males and females in alates and neotenics. Error bars: standard deviation. Total
number of measurements is indicated over each error bar (one measurement
corresponds to a pool of 60 ml of hemolymph).
Fig. 5. JH titers observed during stationary molt and JHA-induced soldier caste
differentiation. Abbreviations are the same as in Fig. 3. JHA PS 0d-7d: presoldiers
obtained by JHA treatment and taken on the day of molt or 7 days after molt; PS
0d: natural presoldiers taken within 1 day after molt. Error bars: standard
deviation. Total number of measurements is indicated over each error bar (one
measurement corresponds to a pool of 60 ml of hemolymph).
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930926
3.4. Clearance of exogenous hormonal compounds from hemolymph
The LC-MS method allowed us also to monitor the variation inconcentration of exogenous compounds in the hemolymph,during artificial induction of soldier differentiation. Twenty-fourhours after JHA treatment, the mean JHA (pyriproxyfen) concen-tration detected into the hemolymph was 13.474 ng/ml(mean7s.d., n ¼ 4) (Fig. 6A). The maximal measured JHAconcentration was 19 ng/ml and the proportion of JHA incorpo-rated into the hemolymph represented consequently less than1.5% of the amount of JHA topically applied on each individual.However, a larger proportion of JHA may also be sequestered inother tissues. Nevertheless, JHA concentration was at first much
higher than physiological levels of the natural JH and it decreasedrapidly during the few days following treatment, before reaching aplateau at around 2 ng/ml, from day 7 to day 10 after treatment.Prior to the presoldier molt, on day 14 after treatment, JHAconcentration dropped to 252787 pg/ml (mean7s.d., n ¼ 4),which was however, a value still comparable to the normal levelof intrinsic JH. This late drop of JHA concentration may be partiallyexplained by the coincidental augmentation of hemolymphvolume. After the presoldier molt, JHA was completely clearedfrom the hemolymph and its concentration fell to below the limitof detection.
The effect of ectopic JH III treatment on hemolymph JH titerwas also investigated (Fig. 6B). Twenty-four hours after topicalapplication of JH III, the mean value of JH titer (intrinsic JHIII+exogenous JH III) was over 8 ng/ml, a value greatly exceedingthe normal physiological concentration of JH. However, this largeamount of exogenous JH was rapidly degraded within 7 days aftertreatment, JH titer then dropped to 75 pg/ml, which was very low,compared to the physiological JH titer naturally occurring inpseudergates.
4. Discussion
4.1. JH titer in H. sjostedti
JH III was the only form of juvenile hormone detected inH. sjostedti, and this corresponds with the results obtained fromother termite species in previous studies (Noirot and Bordereau,1991; Park and Raina, 2004; Yagi et al., 2005; Brent et al., 2005). InH. sjostedti, measured JH titers were higher than the valuesobtained in other species by different methods. This is inaccordance with the results of Westerlund (2004), who foundthat analyses with the LC-MS method yielded significantly higherJH titers than GC-MS for the same samples. However, H. sjostedti
JH titers were of the same order of concentration, compared toresults for the pea aphid (Westerlund and Hoffmann, 2004), andto JH titers obtained by radioimmunoassay in the termiteReticulitermes flavipes (Okot-Kotber et al., 1993) or by the GC-MSmethod in Zootermopsis angusticollis (Brent et al., 2005). Althoughobtained from whole body extracts, JH titers measured by GC-MSin the subterranean termite, Coptotermes formosanus, (Park andRaina, 2004; Liu et al., 2005; Mao et al., 2005) were actually lowerthan in H. sjostedti.
Furthermore, conspicuous variations of the estimatedamount of hemolymph per individual were observed in thepresent study and such a phenomenon was also described in acrustacean species (Ziegler et al., 2000). Although the generalhydration level of the insects and the bleeding method mayaffect the volume of hemolymph obtained, a clear increase ofhemolymph volume was evident in H. sjostedi a few daysbefore the molt to every caste, when intensive mobilization ofstorage proteins was described (Cornette et al., 2007). The gutcontents may be one important source of water for such avolume increase that occurs simultaneously with the gut purge(Cornette et al., 2007). As a consequence, hemolymph volume is asignificant factor that should be taken into account for thedetermination of JH titer by indirect methods. JH titer is generallyconsidered as the balance between synthesis by the corporaallata, degradation by esterases, epoxide hydrolases or P450s andby sequestration by proteins such as hexamerins (Hartfelder,2000; Lafont, 2000; Zhou et al., 2006). It is important to take thelatter factors into account, but neglecting the variations ofhemolymph volume may also lead to some overestimation ofJH titer during molting events.
Fig. 6. Variation of exogenous compounds in the hemolymph after topical application on pseudergates. (A) Pyriproxyfen (JHA) titer. (B) JH III titer, including both intrinsic
and externally applied JH III. Abbreviations are the same as in Fig. 3. Error bars: standard deviation. Total number of measurements is indicated over each error bar (one
measurement corresponds to a pool of 60 ml of hemolymph).
Fig. 7. Models of JH level patterns for each caste differentiation pathway, inferred from the data of JH titer quantification. (A) Pattern of JHA and intrinsic JH titers during
induced soldier caste differentiation. (B) JH level pattern proposed for natural soldier caste differentiation: constantly high JH level is required for soldier differentiation. (C)
JH titer changes during alate caste differentiation: constantly low JH level is required for alate differentiation and then the maturation of female imagoes is accompanied
with JH production. (D) Evolution of JH level during stationary molt: low JH level in the intermolt followed by a peak of JH at the time of ecdysis should induce stationary
molt. Abbreviations are the same as in Fig. 3.
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930 927
4.2. JH titer and caste differentiation
A general feature of hemolymph JH titer in H. sjostedti is a peakthat appears before molting. This peak of JH was observed duringinduced soldier differentiation, stationary molt and is probablyalso present during the other larval molts. Interestingly, in thecase of JHA treatment, there is a time lag between the peak of JHtiter and the size of the corpora allata, which reached a maximumearlier, on day 7 after JHA treatment (Cornette et al., unpublishedobservation). During induced soldier differentiation, a negativeeffect of JHA was first observed on the intrinsic JH titer, which fellrapidly (Fig. 7A). Such a negative effect of exogenous JH or JHA onintrinsic JH titer was observed previously in other insects(Anspaugh and Roe, 2005; Chen et al., 2005). In spite of thisinhibitory effect of JHA on endogenous JH titer, the JH titersubsequently rose prior to the presoldier molt and this later peakof JH may thus be considered as an independent component of thephysiological program leading to the molt.
The model that we propose for JH titer changes duringH. sjostedti soldier caste differentiation in natural conditions isillustrated in Fig. 7B. Soldier differentiation is traditionallythought to occur when JH titer is high, because JH treatment orimplantation of active corpora allata can induce this differentia-tion in various termite species (Nijhout and Wheeler, 1982; Noirotand Bordereau, 1991). Furthermore, soldier production inH. sjostedti was maximal before swarming in summer (Matsumotoand Hirono, 1985), when the mean JH titer of pseudergates wasvery high. Incidentally, JH titer of pseudergates exhibited obviousvariation, with a higher level in summer (Fig. 4) and a lower levelin stock colonies in winter (Fig. 5). Similarly, seasonal variation ofJH titers was observed in C. formosanus with a peak in summer,coinciding with elevated seasonal temperatures and high soldierproduction (Liu et al., 2005). Consequently pseudergates with ahigh JH titer throughout their intermolt may be more inclined todifferentiate into presoldiers (Fig. 7B), whereas pseudergates withlower JH titer during the intermolt would undergo stationary molt
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930928
(Fig. 7D). However, stationary molts require a higher JH titersubsequently during the premolting period (Fig. 7D). The basicdifference between soldier differentiation and pseudergate sta-tionary molting would consequently be found earlier in theintermolt, when the JH titer is probably higher than a putativethreshold only in individuals destined to undergo presoldiermolt. Such a physiological pattern would occur preferentially insummer (high JH titers in general), or may be mimicked by thejuvenoid effect of JHA.
In natural conditions, corpora allata activity and JH titer wereshown to be high in presoldiers from different species (Grasse,1982; Park and Raina, 2004). In H. sjostedti, the JH titer ofpresoldiers was also high, after a slight drop just following themolt (Fig. 7B). In terminally differentiated soldiers, JH titer isgenerally equivalent to that observed in workers or pseudergates(Grasse, 1982). However, H. sjostedti soldiers retained a relativelyhigh JH titer even in winter, when pseudergates JH titers havedropped (Fig. 5). This phenomenon may be linked to the increasedvolume of corpora allata observed in soldiers (Cornette et al.,unpublished observation). Similarly, JH titer was found to behigher in soldiers than in workers in C. formosanus (Park andRaina, 2004; Liu et al., 2005).
Alate differentiation appeared to require low JH titers through-out the developmental process from the nymphal instar and wascharacterized by the absence of a JH peak at the imaginalmolt (Fig. 7C). The disappearance of JH prior to metamorphosisis a general feature in holometabolous insects (Nijhout, 1994),and a drop in JH titer before imaginal molt was also observedin hemimetabolous insects such as cockroaches (Lanzrein et al.,1985a). After the imaginal molt, female alates before swarmingexhibited a conspicuous elevation of their JH titer, which wasnot observed in males. This difference in JH titer betweensexes was not observed in other castes or intermediary stages.However, the high JH titer of female alates was probablyrelated to the onset of vitellogenesis. In comparison, the elevationof JH titer in the queens of other termite species was observedlater, after colony foundation (Noirot and Bordereau, 1991; Brentet al., 2005).
Interestingly, some nymphs did not differentiate into alatesand underwent a regressive molt to pseudergates (Koshikawaet al., 2001). Such a phenomenon is relatively common in lowertermites as well as wing bud mutilation of nymphs by conspecifics(Roisin, 2000). Since JH titer was demonstrated to increase instressed honey bees (Lin et al., 2004) and particularly afteraggressive behavior in burying beetles (Scott, 2006), it is possiblethat nymphs of H. sjostedti with a low JH titer during the onset ofthe molt (Fig. 7C) may suddenly increase their JH titer in responseto aggression from conspecifics. The resultant JH profile wouldmimic the pattern for a stationary molt (Fig. 7D), thus inducingregressive molting in the pseudergates. In such a scenario,individual termites could be acting directly on the JH titer oftheir conspecifics by aggressive interactions, and thus avoidingexcessive dispersal in response to environmental or nutritionalfactors.
Finally, the intriguing case of neotenics should be mentioned,because both sexes presented a very low JH titer, although femaleswere functionally developed and should be engaged in vitellogen-esis, which requires elevated JH levels (Nijhout, 1994). This is alsoin contradiction to histological data, which showed that thecorpora allata from neotenics were hypertrophied and apparentlyactive (Cornette et al., unpublished observation). Cyclic vitello-genesis could explain a temporarily low JH titer, but it is hard tobelieve that all investigated individuals were physiologicallysynchronized. However, disparities between JH synthesis andthe actual hemolymph JH titer were also observed in a cockroach(Treiblmayr et al., 2006). Another possible interpretation is that
corpora allata may be involved in the secretion of compoundsdifferent from JH.
4.3. Ectopic juvenoids and soldier differentiation
JHA treatment induced high proportions of soldier differentia-tion (up to 90% of presoldier formation). It is commonly acceptedthat the action of JHA is linked to its juvenoid effect: theexogenous JHA mimics a high JH titer throughout the threesensitive periods leading to soldier differentiation (Nijhout andWheeler, 1982; Noirot and Bordereau, 1991). Pyriproxyfen (JHA)was shown to exhibit general juvenoid effect (Wilson, 2004) andbecause of its atypical structure, may not be affected by the majorJH degrading enzymes (Zhang et al., 1998). However, thehemolymph JHA titer decreased markedy after treatment ofpseudergates with pyriproxyfen. Enzymes specialized in thedegradation of xenobiotic compounds, such as P450s, may beresponsible for this clearance of JHA from hemolymph (Zhang etal., 1998; Feyereisen, 1999). Another possible mechanism for JHAclearance is excretion through the Malpighian tubules, especiallyduring molts, which have a cleansing effect by their contributionto the excretion of metabolic products during this period (Grasse,1982). This phenomenon may explain the abrupt disappearanceof JHA from hemolymph, observed after the molt to presoldiers(Fig. 6).
In contrast to the level of JHA, which was present inhemolymph until day 14 after treatment, ectopic JH III wasrapidly degraded and this suggests a strong activity of JHdegrading enzymes, JH esterase, JH epoxide hydrolase or cyto-chrome P450 (Gilbert et al., 2000). Such a strong JH degradingactivity could explain why JH III failed to induce soldierdifferentiation in H. sjostedti (Cornette et al., 2006).
In addition to its juvenoid effect, JHA was also suggested toexhibit a prothoracicotropic effect (Wanyonyi, 1974; Nijhout,1994). In relation to this, JHA treatment induced hypertrophy ofthe prothoracic gland (Cornette et al., unpublished observation).In the present study, the ecdysone titer was also investigated byLC-MS, but the amounts of ecdysone were close to the limit ofdetection and did not allow accurate quantification. Nevertheless,the peak of ecdysone prior to the presoldier molt was detected, aswell as an early peak between 24 h and 3 days after JHA treatment(data not shown). Such an early peak of ecdysteroids was alsodetected during soldier differentiation of Macrotermes michaelseni
(Okot-Kotber, 1983) and this early peak was influenced by JHapplication (Lanzrein et al., 1985b). Consequently, the inductionof soldier differentiation by JHA is probably the result of acombination of its juvenoid effect with an ecdysiotropic action.
4.4. General features of the physiology in termite
caste differentiation
In the present study, JH titer was investigated in all castes andsome transitional stages of one termite species. These resultsprovide new information to discuss the model for JH-mediatedcontrol of caste determination in termites, proposed by Nijhoutand Wheeler (1982). Firstly, initiation of soldier caste differentia-tion is associated with high JH titers throughout all JH sensitiveperiods of the pseudergate instar and subsequently in presoldiers.Our results and many other studies (see for review Noirot andBordereau, 1991) support this statement. Secondly, alate differ-entiation requires a constantly low JH titer. The present dataconfirm that JH titer was low throughout the nymphal instar andin male alates, however hormone levels were not investigatedprior to the molt from pseudergate to nymph and we can onlyspeculate the absence of JH peak at the timing of the molt, which
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930 929
would discriminate JH titer pattern of nymphal molt fromstationary molt. In practice, pseudergates destined to differentiateinto nymphs or neotenics may not be morphologically distin-guishable and their fate was not predictable in H. sjostedti, thus JHtiters could not be quantified in such individuals.
Concerning the stationary molt, in the model proposed byNijhout and Wheeler (1982), JH titer was proposed to fall off in thesecond half of the pseudergate stage, and to reach a low levelduring the last JH sensitive period corresponding to soldiercharacters. On the contrary, the present study shows an increaseof JH titer in the last portion of pseudergate stage, prior tothe molt. Consequently, the timing of JH sensitive periods inH. sjostedti may be different from that deduced in Nijhout andWheeler (1982) model for K. flavicollis. We have also discussed aputative peak of ecdysone between 24 h and 3 days after JHAtreatment. This timing is correlated with a switch in geneexpression (Cornette et al., 2006) and also with the induction ofstorage protein intake into the fat body, which is an ecdsyteroid-dependent phenomenon (Cornette et al., 2007). Such an earlypeak of ecdysteroids was also observed under natural conditions(Okot-Kotber, 1983). Consequently, we suggest the existence of acommitment peak of ecdysteroids around 12 days prior to themolt, which would induce soldier differentiation if combined witha high JH titer. According to Nijhout and Wheeler’s (1982) model,the JH-sensitive periods for sexual characters and non-sexualadult characters would occur prior to the sensitive period forsoldier characters cited above. Unfortunately, this time windowwas not investigated in H. sjostedti and this portion of the modelremains untested.
In conclusion, the results presented here corroborate Nijhoutand Wheeler’s (1982) model concerning soldier differentiation,but do not allow any definitive statement for alate determination.Concerning the stationary molt, our data do fit the modelonly if the peak of JH preceding ecdysis in pseudergates isconsidered as following the three JH-sensitive periods that arementioned. Further investigations on ecdysteroid titers, butalso on JH titers in other termite species will hopefully help toresolve this matter.
Acknowledgments
We are grateful to Prof. Tatsufumi Okino, Takaya Kisugi andHiroshi Matsuura for providing access to LC-MS facilities and fortheir help during analyses. We thank also Kiyoto Maekawa, YukiIshikawa and Asano Ishikawa for their help during the field andlaboratory studies on termites. We are finally grateful to theorganizing committee of JH9 congress in York, UK. Sumitomochemicals (Osaka, Japan) generously provided the juvenilehormone analog. This work was supported by postdoctoralfellowships from the Inoue Foundation for Science, and byGrants-in-Aid for Scientific Research (Nos. 15687001, 17770012,18047002 and 18370007) from the Ministry of Education, Culture,Sports, Science and Technology of Japan.
References
Anspaugh, D.D., Roe, R.M., 2005. Regulation of JH epoxide hydrolase versus JHesterase activity in the cabbage looper, Trichoplusia ni, by juvenile hormoneand xenobiotics. Journal of Insect Physiology 51, 523–535.
Brent, C.S., Schal, C., Vargo, E.L., 2005. Endocrine changes in maturing primaryqueens of Zootermopsis angusticollis. Journal of Insect Physiology 51,1200–1209.
Chen, Y., Chen, S., Chang, H., Sun, G., Kou, R., 2005. Effect of exogenous juvenilehormone III and precocene II on agonistic behavior and the corpora allata invitro activity in the male lobster cockroach Nauphoeta cinerea (Dictyoptera:Blaberidae, Oxyhaloinae). Zoological Studies 44, 409–416.
Cornette, R., Koshikawa, S., Hojo, M., Matsumoto, T., Miura, T., 2006. Caste-specificcytochrome P450 in the damp-wood termite Hodotermopsis sjostedti (Isoptera,Termopsidae). Insect Molecular Biology 15, 235–244.
Cornette, R., Matsumoto, T., Miura, T., 2007. Histological analysis on fat bodydevelopment and molting events during soldier differentiation in the damp-wood termite Hodotermopsis sjostedti (Isoptera, Termopsidae). ZoologicalScience 24, 1066–1074.
Evans, J.D., Wheeler, D.E., 1999. Differential gene expression between developingqueens and workers in the honey bee, Apis mellifera. Proceedings of theNational Academy of Sciences of the United States of America 96, 5575–5580.
Evans, J.D., Wheeler, D.E., 2001. Gene expression and the evolution of insectpolyphenisms. BioEssays 23, 62–68.
Feyereisen, R., 1999. Insect P450 enzymes. Annual Review of Entomology 44,507–533.
Gilbert, L.I., Granger, N.A., Roe, R.M., 2000. The juvenile hormones: historical factsand speculations on future research directions. Insect Biochemistry andMolecular Biology 30, 617–644.
Grasse, P.P., 1982. Termitologia: Anatomie, Physiologie, Reproduction des Termites.Masson, Paris, New York, 676pp.
Grasse, P.P., Noirot, C., 1947. Le polymorphisme social du termite a cou jaune(Calotermes flavicollis F.). Les faux-ouvriers ou pseudergates et les muesregressives. Comptes Rendus de l’Academie des Sciences 224, 219–221.
Greenberg, S., Tobe, S.S., 1985. Adaptation of a radiochemical assay for juvenilehormone biosynthesis to study caste differentiation in a primitive termite.Journal of Insect Physiology 31, 347–352.
Hartfelder, K., 2000. Insect juvenile hormone: from ‘‘status quo’’ to high society.Brazilian Journal of Medical and Biological Research 33, 157–177.
Hartfelder, K., Emlen, D.J., 2005. Endocrine control of insect polyphenism. In:Gilbert, L.I., Iatrou, K., Gill, S.S. (Eds.), Comprehensive Molecular Insect Science,Vol. 3. Elsevier, Oxford, pp. 651–703.
Howard, R.W., Haverty, M.I., 1979. Termites and juvenile hormone analogues:a review of methodology and observed effects. Sociobiology 4, 269–278.
Hrdy, I., Krecek, J., 1972. Development of superfluous soldiers induced by juvenilehormone analogues in the termite, Reticulitermes lucifugus santonensis. InsectesSociaux 19, 105–109.
Koshikawa, S., Matsumoto, T., Miura, T., 2001. Regressive molt in the Japanesedamp-wood termite Hodotermopsis japonica (Isoptera: Termopsidae). Socio-biology 38, 495–500.
Koshikawa, S., Cornette, R., Hojo, M., Maekawa, K., Matsumoto, T., Miura, T., 2005.Screening of genes expressed in developing mandibles during soldierdifferentiation in the termite Hodotermopsis sjostedti. FEBS Letters 579,1365–1370.
Lafont, R., 2000. Understanding insect endocrine systems: molecular approaches.Entomologia Experimentalis et Applicata 97, 123–136.
Lanzrein, B., Gentinetta, V., Abegglen, H., Baker, F., Miller, C., Schooley, D., 1985a.Titers of ecdysone, 20-hydroxyecdysone and juvenile hormone III throughoutthe life cycle of a hemimetabolous insect, the ovoviviparous cockroachNauphoeta cinerea. Experientia 41, 913–917.
Lanzrein, B., Gentinetta, V., Fehr, R., 1985b. Titres of juvenile hormone andecdysteroids in reproductive and eggs of Macrotermes michaelseni: relation tocaste determination? In: Watson, J., Okot-Kotber, B.M., Noirot, C. (Eds.), CasteDifferentiation in Social Insects. Pergamon Press, Oxford, pp. 307–327.
Lebrun, D., 1967. La determination des castes du termite a cou jaune (Calotermesflavicollis Fabr.). Bulletin de Biologie Franco-Belge 101, 139–217.
Lelis, A., Everaerts, C., 1993. Effects of juvenile hormone analogues upon soldierdifferentiation in the termite Reticulitermes santonensis (Rhinotermitidae,heterotermitinae). Journal of Morphology 217, 239–261.
Lin, H., Dusset, C., Huang, Z., 2004. Short-term changes in juvenile hormone titersin honey bee workers due to stress. Apidologie 35, 319–327.
Liu, Y., Henderson, G., Mao, L., Laine, R.A., 2005. Seasonal variation of juvenilehormone titers of the formosan subterranean termite, Coptotermes formosanus(Rhinotermitidae). Environmental Entomology 34, 557–562.
Luscher, M., 1958. Experimentelle Erzeugung von Soldaten bei der TermiteKalotermes flavicollis. Naturwissenschaften 45, 69–70.
Mao, L., Henderson, G., Liu, Y., Laine, R.A., 2005. Formosan subterranean termite(Isoptera: Rhinotermitidae) soldiers regulate juvenile hormone levels andcaste differentiation in workers. Annals of the Entomological Society ofAmerica 98, 340–345.
Matsumoto, T., Hirono, Y., 1985. On the caste composition of a primitive termiteHodotermopsis japonicus Holmgren (Isoptera, Termopsidae). Scientific Papers ofthe College of Arts and Sciences—The University of Tokyo 35, 211–216.
Miura, T., 2001. Morphogenesis and gene expression in the soldier-castedifferentiation of termites. Insectes Sociaux 48, 216–223.
Miura, T., 2004. Proximate mechanisms and evolution of caste polyphenism insocial insects: from sociality to genes. Ecological Research 19, 141–148.
Miura, T., Kamikouchi, A., Sawata, M., Takeuchi, H., Natori, S., Kubo, T., Matsumoto,T., 1999. Soldier caste-specific gene expression in the mandibular glands ofHodotermopsis japonica (Isoptera: Termopsidae). Proceedings of the NationalAcademy of Sciences of the United States of America 96, 13874–13879.
Miura, T., Hirono, Y., Machida, M., Kitade, O., Matsumoto, T., 2000. Castedevelopmental system of the Japanese damp-wood termite Hodotermopsisjaponica (Isoptera: Termopsidae). Ecological Research 15, 83–92.
Miura, T., Koshikawa, S., Matsumoto, T., 2003. Winged presoldiers induced by ajuvenile hormone analog in Zootermopsis nevadensis: implications for plasticityand evolution of caste differentiation in termites. Journal of Morphology 257,22–32.
R. Cornette et al. / Journal of Insect Physiology 54 (2008) 922–930930
Miura, T., Koshikawa, S., Machida, M., Matsumoto, T., 2004. Comparative studies onalate wing formation in two related species of rotten-wood termites:Hodotermopsis sjostedti and Zootermopsis nevadensis (Isoptera: Termopsidae).Insectes Sociaux 51, 247–252.
Nijhout, H.F., 1994. Insect Hormones. Princeton University Press, Princeton, 267pp.Nijhout, H.F., 1999. Control mechanisms of polyphenic development in insects.
BioScience 49, 181–192.Nijhout, H.F., Wheeler, D.E., 1982. Juvenile hormone and the physiological basis of
insect polymorphisms. The Quarterly Review of Biology 57, 109–133.Noirot, C., 1969. Glands and secretions. In: Krishna, K., Weesner, F.M. (Eds.), Biology
of Termites, Vol. 1. Academic Press, New York, pp. 89–123.Noirot, C., 1991. Caste differentiation in Isoptera: basic features, role of
pheromones. Ethology Ecology and Evolution 1, 3–7 (special issue) .Noirot, C., Bordereau, C., 1991. Termite polymorphism and morphogenetic
hormones. In: Gupta, A.P. (Ed.), Morphogenetic Hormones of Arthropods.Roles in Histogenesis, Organogenesis and Morphogenesis. Rutgers UniversityPress, New Brunswick, pp. 293–324.
Ogino, K., Hirono, Y., Matsumoto, T., Ishikawa, H., 1993. Juvenile hormone analogue,S-31183, causes a high level induction of presoldier differentiation in theJapanese damp-wood termite. Zoological Science 10, 361–366.
Okot-Kotber, B.M., 1983. Ecdysteroid levels associated with epidermal eventsduring worker and soldier differentiation in Macrotermes michaelseni (Isoptera:Macrotermitinae). General and Comparative Endocrinology 52, 409–417.
Okot-Kotber, B.M., Prestwich, G.D., Strambi, A., Strambi, C., 1993. Changes inmorphogenetic hormone titers in isolated workers of the termite Reticulitermesflavipes (Kollar). General and Comparative Endocrinology 90, 290–295.
Park, Y.I., Raina, A.K., 2004. Juvenile hormone III titers and regulation of soldiercaste in Coptotermes formosanus (Isoptera: Rhinotermitidae). Journal of InsectPhysiology 50, 561–566.
Roisin, Y., 2000. Diversity and evolution of caste patterns. In: Abe, T., Bignell, D.E.,Higashi, M. (Eds.), Termites: Evolution, Society, Symbioses, Ecology. KluwerAcademic Publishers, Dortrecht, pp. 95–119.
Sameshima, S., Miura, T., Matsumoto, T., 2004. Wing disc development duringcaste differentiation in the ant Pheidole megacephala (Hymenoptera: Formici-dae). Evolution and Development 6, 336–341.
Scharf, M.E., Wu-Scharf, D., Pittendrigh, B.R., Bennett, G.W., 2003. Caste- and deve-lopment-associated gene expression in a lower termite. Genome Biology 4, R62.
Scott, M., 2006. Resource defense and juvenile hormone: The ‘‘challengehypothesis’’ extended to insects. Hormones and Behavior 49, 276–281.
Juvenile hormone titer versus juvenile hormone synthesis in female nymphsand adults of the German cockroach, Blattella germanica. Journal of InsectScience 6, 47.
Wanyonyi, K., 1974. The influence of juvenile hormone analogue ZR 512 (ZOECON)on caste development in Zootermopsis nevadensis (Hagen) (Isoptera). InsectesSociaux 21, 35–44.
Westerlund, S.A., 2004. Measuring juvenile hormone and ecdysteroid titers ininsect haemolymph simultaneously by LC-MS: the basis for determiningthe effectiveness of plant-derived alkaloids as insect growth regulators. M.Sc.Thesis, University of Bayreuth, Bayreuth, p. 102.
Westerlund, S.A., Hoffmann, K.H., 2004. Rapid quantification of juvenile hormonesand their metabolites in insect haemolymph by liquid chromatography-mass spectrometry (LC-MS). Analytical and Bioanalytical Chemistry 379,540–543.
Wilson, T., 2004. The molecular site of action of juvenile hormone and juvenilehormone insecticides during metamorphosis: how these compounds killinsects. Journal of Insect Physiology 50, 111–121.
Yagi, K.J., Kwok, R., Chan, K.K., Setter, R.R., Myles, T.G., Tobe, S.S., Stay, B., 2005. Phe-Gly-Leu-amide allatostatin in the termite Reticulitermes flavipes: content inbrain and corpus allatum and effect on juvenile hormone synthesis. Journal ofInsect Physiology 51, 357–365.
Zhang, L., Kasai, S., Shono, T., 1998. In vitro metabolism of pyriproxyfen bymicrosomes from susceptible and resistant housefly larvae. Archives of InsectBiochemistry and Physiology 37, 215–224.
Zhou, X., Tarver, M.R., Bennett, G.W., Oi, F.M., Scharf, M.E., 2006. Two hexameringenes from the termite Reticulitermes flavipes: sequence, expression, andproposed functions in caste regulation. Gene 376, 47–58.
Ziegler, A., Grospietsch, T., Carefoot, T.H., Danko, J.P., Zimmer, M., Zerbst-Boroffka, I.,Pennings, S.C., 2000. Hemolymph ion composition and volume changes in thesupralittoral isopod Ligia pallasii Brandt, during molt. Journal of ComparativePhysiology B 170, 329–336.
