new insights into the roles of juvenile hormone and ecdysteroids in honey bee reproduction
TRANSCRIPT
Journal of Insect Physiology 59 (2013) 655–661
Contents lists available at SciVerse ScienceDirect
Journal of Insect Physiology
journal homepage: www.elsevier .com/ locate/ j insphys
New insights into the roles of juvenile hormone and ecdysteroids inhoney bee reproduction
0022-1910/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.jinsphys.2013.04.006
⇑ Corresponding author. Tel.: +49 3303 293830; fax: +49 3303 293840.E-mail addresses: [email protected] (J. Wegener), [email protected]
(Z.Y. Huang), [email protected] (M.W. Lorenz), [email protected] (J.I. Lorenz), [email protected] (K. Bienefeld).
Jakob Wegener a,⇑, Zachary Y. Huang b, Matthias W. Lorenz c, Judith I. Lorenz c, Kaspar Bienefeld a
a Institute for Bee Research, Friedrich-Engels-Str. 32, D-16540 Hohen Neuendorf, Germanyb Department of Entomology, Michigan State University, East Lansing, MI 48824, USAc Department of Animal Ecology 1, University of Bayreuth, D-95440 Bayreuth, Germany
a r t i c l e i n f o
Article history:Received 18 December 2012Received in revised form 16 April 2013Accepted 18 April 2013Available online 28 April 2013
Keywords:Juvenile hormoneMakisteroneSpermathecal glandVitellogeninDimorphismCaste
a b s t r a c t
In workers of the Western honeybee, Apis mellifera, juvenile hormone (JH) and ecdysteroids regulatemany aspects of age polyphenism. Here we investigated whether these derived functions in workers havedeveloped by an uncoupling of endocrine mechanisms in adult queens and workers, or whether parallelscan be found between the roles of the two hormones in both castes. We looked at yolk protein metabo-lism as a process central to the physiology of both queens and workers, and at sperm storage as a featureof the queen alone. Queens of differing fertility status (virgin, virgin but CO2-treated, inseminated, freshlylaying and 1–2 years-old) were compared regarding vitellogenin (Vg), JH and ecdysteroid-titers in theirhemolymph, as well as ovarian yolk protein and spermathecal gland composition. Our results showedthat hormone titres were unrelated to the composition of spermathecal glands. JH-concentrations inthe hemolymph were low in the groups of queens characterized by yolk uptake into the ovaries, and highin pre-vitellogenic queens or animals that were forced to interrupt egg-laying by caging. Ecdysteroid-concentrations were higher in untreated virgins than after insemination or during egg-laying. They werenot affected by the caging of queens. These patterns of hormone changes were parallel to those knownfrom worker bees. Together, these findings suggest a conserved role for JH as repressor of vitellogeninuptake into tissues, and for ecdysteroids in preparing tissues for this process. An involvement of thetwo hormones in the regulation of sperm storage seems unlikely. Our results add to the view that JHand ecdysteroids act similarly on the yolk protein metabolism of both castes of A. mellifera. This mayimply that it was the biochemical versatility of Vg rather than that of hormonal regulatory circuits thatallowed for the functional separation of the two castes.
� 2013 Elsevier Ltd. All rights reserved.
1. Introduction
In most insect species studied so far, including several socialHymenoptera, JH and ecdysteroids are involved in ovary activation,vitellogenin (Vg) synthesis and/or oogenesis in adult females (re-viewed in Hagedorn and Kunkel, 1979; Raikhel et al., 2005; Hartf-elder and Emlen, 2005; Gilbert, 2012). In the honey bee (Apismellifera) however, JH and ecdysteroids are important elementsfor behavioral regulation in the quasi-sterile worker caste (Amdamet al., 2003, 2005; Nelson et al., 2007; Paul et al., 2005). Hemo-lymph concentrations of JH are low in hive bees and high in forag-ers (Huang et al., 1994). JH interacts with Vg to regulate longevityand behavioral maturation (Amdam et al., 2003, 2005; Nelsonet al., 2007). Although a minimum concentration of the hormone
appears to be required for initiating Vg synthesis (Rutz andLüscher, 1974; Hartfelder and Emlen, 2005), injections of largerdoses of JH or analogs lead to the dropping of Vg titers in the hemo-lymph (Rutz et al., 1976; Pinto et al., 2000) and to the early onset offoraging (Jaycox et al., 1974). Allatectomized workers show re-duced flight performance and metabolic rates, indicating addi-tional and maybe more fundamental roles of this hormone(Sullivan et al., 2003). The role of ecdysteroids in worker behaviorhas only recently been recognized. The ovaries are a likely sourceof ecdysteroids or at least participate in their synthesis in workers(Amdam et al., 2010; Yamazaki et al., 2011), and individuals withlarger or supernumerary ovaries have been shown to forage earlierthan individuals with smaller ones (Amdam et al., 2010; Wanget al., 2010). The expression of genes involved in ecdysteroid sig-naling in the mushroom bodies of worker brains also suggests arole of these hormones in behavioral development (Paul et al.,2005; Yamazaki et al., 2011).
The discovery of these new roles of JH and ecdysteroids in hon-ey bee workers leads to the question of how they are reconciled
656 J. Wegener et al. / Journal of Insect Physiology 59 (2013) 655–661
with the persistent need for regulation of reproductive processes inqueens. Results on JH titers in queens are inconsistent. While Fluriet al. (1981) found no relationship with mating status, Fahrbachet al. (1995) reported higher titers in newly emerged and bankedqueens than in sexually mature virgins and actively laying queens.Both in vivo and in vitro analyses suggest a loss of the typical role ofJH as a stimulant for Vg synthesis (reviewed in Engels and Imper-atriz-Fonseca, 1990; Hartfelder and Emlen, 2005), similar to thesituation in workers. Corona et al. (2007) suggested that JH mayinteract with Vg and elements of the insulin signaling pathwayto regulate queen longevity.
In the case of ecdysteroids, one study (Robinson et al., 1991)showed a positive correlation between fertility and hormone titers,while another (Hartfelder et al., 2002) failed to detect this relation-ship and came to the conclusion that in higher eusocial bees,ecdysteroids may have lost their role in reproduction.
Our first aim was therefore to resolve the discrepancies of ear-lier studies regarding the phenotypic link between hemolymph ti-ters of JH and ecdysteroids on one hand and Vg and queenreproductive status on the other. The second aim was to use thisknowledge regulatory mechanisms in the queen caste to look forphysiological patterns that are similar to those described for theworker caste. Our third aim was to investigate possible relation-ships between the two hormones and sperm storage by A. melliferaqueens. The capacity for long-term storage of male gametes tendsto be correlated with the level of sociality in several groups of antsas well as bees (Ito and Ohkawara, 1994; Martins and Serrao,2002). While non-social species typically store sperm for periodsof hours to weeks, eusocial species store male gametes for monthsor even years (Gobin et al., 2006). It can therefore be argued thatlong-term sperm storage in the Hymenoptera likely presents anadaptation to sociality in the queen caste. Attraction of sperm tothe spermatheca, their conservation and reactivation for fertiliza-tion are thought to be mediated by different classes of secretionsfrom the spermathecal glands (Ruttner and Koeniger, 1971; Lenskyand Schindler, 1967; Koeniger, 1970; Verma and Shuel, 1973; Dal-lai, 1975; Collins et al., 2006; Wegener et al., 2013). As sperm stor-age is functionally closely linked to mating and oogenesis, wehypothesized that the classical insect gonadotropins JH and ecdys-teroids may be involved in the regulation of spermathecal glandsas well. Hence, we looked for associations between JH and ecdy-steroid titers and changes in the chemical composition of the sper-mathecal glands. As far as we know, endocrine control of thespermathecal glands has never been studies in any insect before.
2. Materials and methods
2.1. Queens
Queens of the subspecies A. mellifera carnica were reared instrong, queenless colonies according to standard techniques (Rutt-ner, 1976). All were daughters of the same, multiply mated mother.Shortly after emergence, they were transferred into mating nucleiof approximately 1500–2000 workers. The entrances of the hiveswere equipped with queen excluder grids to prevent natural mat-ing. When the queens were 10–11 days old, all but 5 were narco-tized with CO2 for seven minutes and reintroduced into theirnuclei. Eighteen to 24 h later, all but 7 of the previously CO2-trea-ted queens were artificially inseminated with semen of mixed ori-gin (7 ll/queen) following standard procedures (Ruttner, 1983).Insemination involved a second CO2-narcosis of 3–7 min. Anotherset of 10 of the inseminated queens (12–13 days old, 24 h afterinsemination) were caught from their nuclei and samples for thedetermination of Vg and hormone titers were taken as describedbelow. The uninseminated, unnarcotized as well as the uninsemi-
nated but CO2-treated queens were sampled in the same way.The remaining 9 queens were allowed to initiate egg-laying insidetheir nuclei, which they did after 4–10 days. On the 24th to 25th dayafter emergence, they were caught and samples were taken fromthem as well. In addition to the queens reared especially for thisexperiment, two outgroups were included in the analyses. Bothconsisted of 1–2 year-old, fertile A. mellifera carnica-queens takenfrom colonies of various sizes. These queens were of diverse genet-ic origins, although all belonged to the same subspecies, A. melliferacarnica. In one group, hemolymph and ovary samples were takendirectly (<10 min) after removal from the colonies. In a secondgroup, queens were prevented from egg-laying for 4–10 h by cag-ing in wire mesh cages (3 � 4 � 0.8 cm) before sampling. Theywere accompanied by 4–6 workers. The cages were supplied witha drop of honey and kept at room temperature in the dark. Alto-gether, the following groups of queens were compared:
(A) 12–13 days post-emergence, no CO2-treatment, no insemi-nation (n = 5).
(B) 12–13 days post-emergence, 24-36 h after CO2-treatment(n = 7).
(C) 12–13 days post-emergence, 24 h after insemination(n = 10).
(D) 24–25 days post-emergence, shortly after onset of egg-lay-ing (n = 9).
(E) 1–2 years old, actively egg-laying (n = 9).(F) 1–2 years old, caged for 4–10 h before dissection (n = 9).
2.2. Sample preparation
Queens were cold-anaesthetized on ice. For the measurement ofhormone and Vg-titers, two samples of hemolymph (1–3 and 5–6 ll) were collected into glass capillaries from small incisions inthe intersegmental membrane between the last two abdominalsegments. The larger sample was diluted in 500 ll ice-cold metha-nol and later used for the measurement of JH. The smaller sample,used for Vg and ecdysteroid quantification, was expelled into500 ll of ice-cold Tris-buffer (0.05 mol/l TRIS, 0.16 mol/l NaCl,0.1 mmol/l phenylmethylsulfonyl fluoride, pH 8.5). This samplewas further divided into two subsamples: one percent was usedfor the estimation of Vg and 99% was used for the estimation ofecdysteroids.
The queens were decapitated and their ovaries placed into500 ll of Tris-buffer for the measurement of ovarian Vg/vitellin.The spermathecae together with the adhering spermathecal glandswere placed on a microscope slide. The glands were removed witha pair of forceps and frozen in 300 ll of TRIS-buffer.
2.3. Quantification of JH, ecdysteroids and Vg
The samples intended for JH-measurement were shipped on dryice from Germany to Michigan (USA). JH III was extracted and JHtiters were estimated according to well-established proceduresused in honey bees (Huang et al., 1994; Jassim et al., 2000; Sloneet al., 2012). Liquid scintillation counting was performed using aTricarb2100TR (Packard), which gave the radioactivity in disinte-grations per min for each sample. A standard curve with variousamounts (0, 3, 10, 30, 100, 300, 1000, 3000, and 10,000 pg) of stan-dard JH-III (Sigma) was obtained each day that samples were mea-sured, following the protocol of Huang and Robinson (1996). Eachsample was determined in duplicates. The amount of JH in thesamples was corrected by dividing by two, because the racemicmixture of JH standard overestimates the 10R JH in biological sam-ples by a factor of two.
The samples intended for the quantification of Vg and ecdyster-oids were transported within Germany from Berlin to Bayreuth on
]lµ/gµ[nineg olleti v
hpmylo
meH
200
150
100
50
0
yravorepnilletiV
]stnelaviuqe -ni negolletivg
m[
40
30
20
10
0]lµ/gp[
enomr oh
eline vujhp
mylome
H
80
60
40
20
0
sdioretsydcehp
mylo]lµ/.uq e-A
en oretsikg
150
100
50
A
B
C
D
a
aa
a
a
a
a
aa
b
b
b
a a
b
c
bc
a
a
ab
bc abc
cc
J. Wegener et al. / Journal of Insect Physiology 59 (2013) 655–661 657
dry ice. The hemolymph-buffer-mixture was transferred from theoriginal reaction tubes that had been used for hemolymph sam-pling into 6.5 ml polypropylene tubes (Sarstedt, Nümbrecht, Ger-many). The original tubes were rinsed twice with 500 ll ofmethanol each by vigorous vortexing and a 5 min treatment inan ultrasonic bath. Rinsing solutions were combined with thehemolymph-buffer-mixture and 3 ml isooctane was added fordefatting. After vortexing and treatment in the ultrasonic bath,samples were centrifuged (5500g, 4 �C, 10 min). After discardingthe top phase (isooctane containing the lipids), the remainingmethanol-buffer mixture was transferred into a new 6.5 ml tube,mixed with 2 ml of water and concentrated to a volume of500 ll in a Speed Vac. The following fractionation of ecdysteroidsby solid phase separation (to obtain free ecdysteroids) as well astheir quantification by radioimmuno assay were performed as de-scribed in Lorenz et al. (1997) with the following modifications:DBL-1 antiserum, which displays a high cross-reactivity for makis-terone A, was used. The standard curve was obtained by usingmakisterone A (AG Scientific, USA) as the non-labeled standard(0–10,000 pg).
Vg/vitellin in queen hemolymph and ovaries was determinedby ELISA as in Wegener et al. (2009a). Ovarian tissue was homog-enized in the TRIS-buffer for 35 s with a fast-spinning motorizedpestle. Each sample was measured in two dilutions, 2000- and10,000-fold. Protein measured in the ovaries is probably a mixtureof Vg and vitellin. In the honey bee, the two are known to be elec-trophoretically indistinguishable (Engels et al., 1990). It is knownfrom previous studies that the antibody used shows affinity to bothVg and vitellin (Wegener et al., 2009a, 2010). As only Vg was usedfor the standard curves, results are given in Vg-equivalents.
2.4. Determination of the chemical composition of the spermathecalglands
Spermathecal glands were homogenized by use of a hand-heldmotorized pestle. The amounts of free carbohydrate, glycogen, li-pid, and protein were measured according to Lorenz (2003) withthe modifications described in Lorenz (2007). In addition, to in-crease the sensitivity of the method, which was necessary due tothe small size of the spermathecal glands, the solvent volumes ofthe procedure were down-scaled by a factor of 5.
2.5. Statistical analysis
IBM-SPSS 19 was used for data analysis. Differences betweenthe treatment groups were analyzed using the Kruskal-Wallis non-parametric ANOVA followed by stepwise step-down multiple test-ing (SPSS, 2010) to identify homogenous groups of treatments.
meH
p[
0
Fertility status of queens
,sraey2-1
cage
d,sraey2-1
e litr ef
fo tesnoretfag niyal-gge
re tfah4 2
noi tanimesni
2O
Cretfah42
nigriv
Fig. 1. Ecdysteroids and Juvenile hormone in hemolymph and vitellin/vitellogeninin hemolymph and ovaries of queens of different reproductive status. Bars showmeans ± SE. A total of 49 queens were used for the analyses. Different letters on topof bars indicate significant differences between treatments (P < 0.05).
3. Results
Fig. 1 summarizes the measurements of hormones and Vg. Vgwas present already in the hemolymph of virgins not treated withCO2. Hemolymph protein titers did not differ significantly amongdifferent treatments (Fig. 1A, n = 48; df = 5; Z = 1.83; P = 0.87),whereas ovarian Vg/vitellin varied strongly (n = 48; df = 5;Z = 26.9; P < 0.001). Average quantities were low (<4 mg Vg-equiv-alent/ovary) in untreated, CO2-treated and freshly inseminatedanimals. They were much higher (17–28 mg/ovary) in fertile ani-mals (treatments D–F). Among the groups of fertile queens, ageor caging did not significantly affect the quantities of the two pro-teins in the ovaries (P > 0.05). Levels of JH also differed significantlyamong the treatments (Fig. 1C, n = 49; df = 5; Z = 33.9; P < 0.001).They were moderate in untreated virgins and remained so afterCO2-treatment. They were lower in freshly inseminated queens,
and even lower in queens that had initiated oviposition (treat-ments C and D, P < 0.05 in both cases). JH-titers significantlyincreased if queens were caged before sampling (P < 0.05).
]sdnalg foriap/gµ[n egocyl
G
8
6
4
2
0
]sdnalgforiap/gµ[setardyhobrac
eerF
8
6
4
2
0
A
B
a
a
a
aa
a
ab
b
b
a
b b
658 J. Wegener et al. / Journal of Insect Physiology 59 (2013) 655–661
Hemolymph titers of ecdysteroids varied among treatment groups(n = 47; df = 5; Z = 16.83; P = 0.005; Fig. 1D). They were higher inuntreated virgins than in old, fertile queens regardless of caging(P < 0.05), while animals from the other treatment groups showedintermediate values.
The ovaries of all caged queens contained yellowish bodies to-wards the posterior end of the ovarioles (Fig. 2), indicative of folli-cles in the process of degeneration (Gauthier et al., 2011) orresorption (Hitchcock, 1956; Flanders, 1959). They did not containany mature eggs as were found in the ovaries of most layingqueens.
Fig. 3 shows the results of the analyses of spermathecal glands.Glycogen titers did not vary according to queen reproductive status(Fig. 3A, n = 49; df = 5; Z = 9.1; P = 0.078). Free carbohydrate con-tents did vary (Fig. 3B, n = 49; df = 5; Z = 15.2; P = 0.01), showinga peak in freshly inseminated animals and dropping sharply afterthe onset of egg-laying. Lipids also varied in quantity (Fig. 3C,n = 49; df = 5; Z = 22.0; P = 0.001), with lower amounts found inCO2-treated and freshly inseminated animals than in all othertreatment groups (P < 0.05). Protein titers in the glands displayedthe clearest variations between the treatments (Fig. 3D, n = 49;df = 5; Z = 47.1; P < 0.001). They dropped slightly after insemina-tion, but then increased again after the onset of egg-laying(P < 0.05). They were still higher in 1–2 year-old, fertile animals,and slightly dropped if animals were caged before sampling(P < 0.05).
]sdn algforiap/gµ[dipiL
13
10
8
5
3
0
ni s]sdnalgfori ap/
20
15
C
D
ab
aa
ab
b
ab
c
4. Discussion
4.1. A caste-transgressing role for JH in A. mellifera?
Juvenile hormone is a stimulant of reproductive dominance, Vgsynthesis and/or oogenesis in several groups of social Hymenop-tera, including Vespidae (reviewed by Strambi, 1990), Halictidae(Bell, 1973; Smith et al., 2013), and Apidae (reviewed in Hartfelder,2000). Even in the closely related bumble bee Bombus terrestris, JHtiters are positively linked to ovary development and social statusof queenless workers (Röseler, 1977; Röseler and Röseler, 1988;Bloch et al., 2000a). In adult A. mellifera however, JH has long beenthought to have lost its classical role as stimulant of gonadic activ-ity in favor of a new role as ‘‘behavioral pacemaker’’ (Robinsonet al., 1991; Robinson and Vargo, 1997; Hartfelder, 2000; Engelset al., 1990; Pinto et al., 2000). This view has been changed by
Fig. 2. Degeneration/resorption of follicles in caged queens. Stereomicroscopicimage of the dissected abdomen of a fertile queen bee after 6 h of caging. Arrowsindicate yellowish bodies representing follicles in the process of resorption/degeneration.
etorPtne laviuqe-ASB
gµ[
10
5
0
Fertility status of queens
,sraey2-1
d egac
,sraey2-1
elitref
fo tesno retfagniya l-gge
r etfah42
noit animesni
2O
Cretfah42
nig riv
aab
b
ab
d
Fig. 3. Chemical composition of spermathecal glands in queens of differentreproductive status. Bars show means ± SE. A total of 49 queens were used forthe analyses. Different letters on top of bars indicate significant differences betweentreatments groups (P < 0.05). No significant differences between treatment groupswere observed regarding glycogen content.
the finding that JH and ecdysteroids do participate in the initiationof Vg-synthesis, but that this event has shifted in time from theadult to the late pupal stage (Barchuk et al., 2002).
J. Wegener et al. / Journal of Insect Physiology 59 (2013) 655–661 659
The pattern of JH-concentrations we observed in the hemo-lymph of adult queens is generally in good accordance with thedata of Robinson et al. (1991) and Fahrbach et al. (1995). In con-trast, Fluri et al. (1981) detected no differences in JH-titers be-tween virgin, non-reproductive and mated, laying queens. Thismay be due to the fact that in their experiment, all queens werecaged for >1 day before sampling. Results from our caged animalssuggest that this procedure is likely to have influenced theoutcome.
We show that in adult queens, changes in hemolymph titers ofJH coincide with changes in the reproductive status. JH-titers werefound to be low in females that are engaged in active oogenesis,regardless of their chronological age and mating status, and highin queens where this is not the case. Caging induced a sudden stopof oogenesis, as evidenced by the occurrence of degenerating/partly resorbed follicles or oocytes and the absence of mature eggs.This stop coincided with a sharp rise in JH-concentrations. Doubletreatment with CO2, which is known to lead to the initiation of Vgsynthesis and oogenesis in honeybee queens (Mackensen, 1947;Engels and Ramamurty, 1976), also led to a drop in JH titers. Tosummarize, JH concentrations were low in the hemolymph of ani-mals with starting or ongoing vitellogenesis, and high in previtell-ogenic animals or animals that showed signs of yolk resorption. Asimilar negative relationship between JH and Vg uptake into tis-sues can also be found in workers, where hemolymph concentra-tions of JH are low in nurses and winter bees, both of which takeup Vg into their hypopharyngeal glands, as well as in individualswith vitellogenic ovaries. Foragers, in which neither ovaries norhypopharyngeal glands are known to take up Vg, have higherhemolymph titers of JH (Fluri et al., 1982; Huang et al., 1994). To-gether, these findings strongly suggest a role of JH as suppressor ofVg-synthesis and/or uptake in both castes of A. mellifera. Treatmentof young workers with the JH-analog pyriproxyfen leads to thesuppression of Vg synthesis (Pinto et al., 2000), which seems tosuggest that the effect of JH lies at the level of synthesis. Coronaet al. (2007) showed that expression of the Vg-gene in queen headsis reduced by administration of a high dose of another JH-analog,methoprene, or of an unspecified dose of JH III. On the other hand,the same authors show that abdominal concentrations of Vg-mRNA are highest in queens shortly after emergence, when JH-ti-ters are known to peak as well (Fahrbach et al., 1995). An earlierstudy found that injections of JH III prevent synthesis only at highdoses, while low doses stimulate Vg production (Rutz et al., 1976).Engels et al. (1990) reported that application of JH III to queens orincubation of queen fat bodies with this hormone did not affect Vgsynthesis. The fact that in the untreated virgin queens of our study,high titers of JH coincided with the presence of relatively high lev-els of Vg in the hemolymph also seems to contradict a universalrole for JH as suppressor of yolk protein synthesis. Therefore, wesuggest that the primary role of JH in Vg metabolism is thesuppression of uptake by target tissues rather than the suppressionof synthesis.
4.2. Functions of ecdysteroids in honey bee queens – similarities to theworker caste
In B. terrestris, the role of ecdysteroids in reproduction is lessclear than for JH, although higher concentrations in the hemo-lymph are linked with dominance status in queenless workers(Bloch et al., 2000b; Geva et al., 2005). In adults of this species,the ovaries are an important source of ecdysteroids (Geva et al.,2005). In workers of A. mellifera however, only the early steps ofecdysteroid synthesis seem to be located in the ovary, whereasthe later stages mainly take place in the brain and fat body(Yamazaki et al., 2011).
Earlier studies regarding ecdysteroids in queen honey bees havegiven inconsistent results. Robinson et al. (1991) found higher ti-ters in laying queens than in functionally sterile or laying workers,whereas Hartfelder et al. (2002) found no differences betweennewly-emerged and actively laying queens, or between queensand workers of any reproductive status. Here we found higher ti-ters in sexually mature virgin than in laying queens. While thesefindings are insufficient to resolve the discrepancies between thetwo earlier reports, they show that elevated ecdysteroid titers inqueens are a transient state, explaining why different studies,using animals of differing reproductive status, may come to seem-ingly contradictive results.
Ecdysteroids in workers bees are known to promote generalprotein synthesis by the fat body of queens (Engels et al., 1990),and titers are higher in reproductive animals of both castes thanthey are in non-reproductive workers (Robinson et al., 1991). Onthe other hand, there is no continuous relationship between indi-vidual levels of ecdysteroids and development of either ovariesor hypopharyngeal glands in workers (Hartfelder et al., 2002;Wegener et al., 2009b). Instead there is a peak of hemolymph ecdy-steroid concentration on day three of adult life (Hartfelder et al.,2002), i.e. shortly before the onset of secretory activity of the hyp-opharyngeal glands, while titers are low during the phase of inten-sive secretion of the glands (approximately days 4–14). Inqueenless worker bees, high titers have been found in individualswhose ovaries were on the brink of vitellogenesis (Wegeneret al., 2009b), while individuals with undeveloped or vitellogenicovaries had lower titers (Hartfelder et al., 2002; Wegener et al.,2009b). In the present study, we found high ecdysteroid levels inqueens with pre-vitellogenic ovaries, as evidenced by a low Vg/vitellin-content, and lower levels in animals with vitellogenic ova-ries. Taken together, these results seem to suggest a role for ecdys-teroids in preparing tissues for Vg uptake. Robinson et al. (1991)showed that the ecdysteroid-levels measured in laying queensare still much higher than those present in queenright nursesand foragers. This could mean that in addition to its proposed rolein preparing organs for Vg uptake, sustained levels of ecdysteroidsare required for oogenesis. This hypothesis is supported by resultsof Paul et al. (2005), who detected transcripts of an ecdysteroid-dependent transcription regulator gene, AmE74, in pre-vitellogenicand vitellogenic egg chambers of actively laying queens. In vitello-genic egg chambers, the transcripts are found in the follicle cells,which in A. mellifera are known to take up Vg and channel it tothe surface of the oocyte (Fleig, 1995). However, no transcripts ofAmE74 were detected in the heads of active nurse bees, so thisgene is either not or only transiently involved in Vg-uptake bythe hypopharyngeal glands.
4.3. Role of JH and ecdysteroids in the regulation of the spermathecalglands
Sociality can evolve independently of long-term sperm storage,and not all sperm-storing species are social. Nevertheless, thecapacity to store male gametes in great numbers for prolongedperiods of time is a characterizing feature of the queens of socialHymenoptera (Gobin et al., 2006), indicating that it may be adap-tive within the context of hymenopteran sociality. Sperm storageis not a static state, but a process, involving the steps of attractingsperm to the site of storage, keeping them alive, and activatingthem for fertilization. In A. mellifera, the spermathecal glands arethought to be involved in all of these steps, and different classesof substances from the glands have been suggested to play a roleat different stages of the process (Lensky and Schindler, 1967; Koe-niger, 1970; Verma and Shuel, 1973; Dallai, 1975; Collins et al.,2006; Baer et al., 2009; Wegener et al., 2013). Given that spermstorage is a process and that it has to be coordinated with other
660 J. Wegener et al. / Journal of Insect Physiology 59 (2013) 655–661
processes like mating and oogenesis, it likely needs regulation, andthe classical insect gonadotropins JH and ecdysteroids seemed pos-sible candidate signal molecules. However, the fluctuations ob-served in the chemical composition of the glands did not followthe same pattern as those of the two groups of hormones, so a di-rect control of the glandular activity by JH or ecdysteroids seemsunlikely. Instead, some of our results could shed a light on possiblefunctions of different classes of substances during sperm storage.The fact that free carbohydrates tended to accumulate in theglands after CO2-treatment and insemination and then sharplydropped after the onset of egg-laying may indicate that these sub-stances could be involved in the transfer of sperm into the sperma-theca. Energy generation in freshly ejaculated honey bee sperm isknown to be sugar-driven (Verma, 1978), while sperm stored in-side the theca are thought to catabolize lipids (Verma and Shuel,1973), which would be in accordance with the observed rise in li-pid contents in the glands of newly-laying queens. Dallai (1975)observed lipid droplets in the glands of mated, but not of virginhoneybee queens. Interestingly, strong differences in gland compo-sition were observed between queens that had recently begun ovi-position and animals that had been fertile for >1 year. This changewas most marked in the case of protein. Enzymes from the sper-mathecal glands are known to support energy metabolism ofstored sperm and provide protection against oxidative stress (Col-lins et al., 2006; Baer et al., 2009). Our results suggest that theimportance of this enzymatic support system grows as spermato-zoa increase in age.
4.4. Implications for the physiology of A. mellifera caste dimorphism
West-Eberhard (1996) suggested that in eusocial species, therole of regulatory mechanisms derived from those of solitaryancestors is modulated by the nutritional status of individuals,resulting in a splitting between (well-nourished) queens and (lesswell nourished) nurses (‘‘split function’’-hypothesis). We herepresent an example of how functions of regulatory circuits inqueens and workers are split regarding their effects on the life his-tory of individuals, while remaining the same at a fundamentalphysiological level. Whether (or to which degree) these circuitsare based on those of solitary ancestors cannot be deduced fromour findings. Like in many solitary species (reviewed in Hartfelder,2000), JH and ecdysteroids in A. mellifera queens seem to be in-volved in reproductive processes. On the other hand, the negativeassociation between Vg uptake and JH is clearly atypical of the sit-uation in the Hymenoptera as a whole (reviewed in Robinson andVargo, 1997).
At present, the caste-transgressing roles for JH and ecdysteroidsregarding Vg uptake we propose here are mostly based on correla-tions only and have to be verified by an inductive approach. If con-firmed, they would add to the view that the supposed loss offunction of the two groups of hormones in honey bee reproductionhas to be reconsidered.
Acknowledgements
We thank Anja Rogge for excellent apicultural assistance. Thiswork was supported by funds of the German Ministry for Food,Agriculture and Consumer Security (BMELV) through the interme-diary of the Federal Office for Agriculture and Food (BLE), withinthe framework of the program for innovation. Some of the equip-ment used was financed through the European Fund for RegionalDevelopment (EFRE).
References
Amdam, G.V., Norberg, K., Hagen, A., Omholt, S.W., 2003. Social exploitation ofvitellogenin. Proceedings of the National Academy of Sciences of the UnitedStates of America 100, 1799–1802.
Amdam, G.V., Aase, A.L.T.O., Seehuus, S.C., Fondrk, M.K., Norberg, K., Hartfelder, K.,2005. Social reversal of immunosenescence in honey bee workers. ExperimentalGerontology 40, 939–947.
Amdam, G., Page, R.E., Fondrk, M., Brent, C., 2010. Hormone response tobidirectional selection on social behavior. Evolution and Development 12,428–436.
Baer, B., Heazlewood, J.L., Taylor, N.L., Eubel, H., Millar, A.H., 2009. The seminal fluidproteome of the honeybee Apis mellifera. Proteomics 9, 2085–2097.
Barchuk, A.R., Bitondi, M.M.G., Simões, Z.L.P., 2002. Effects of juvenile hormone andecdysone on the timing of vitellogenin appearance in hemolymph of queen andworker pupae of Apis mellifera. Journal of Insect Science 2, 1 (8 pp.).
Bell, W.J., 1973. Factors controlling initiation of vitellogenesis in a primitively socialbee, Lasioglossum zephyrum (Hymenoptera: Halictidae). Insectes Sociaux 20,253–260.
Bloch, G., Borst, D.W., Huang, Z.Y., Robinson, G.E., Cnaani, J., Hefetz, A., 2000a.Juvenile hormone titers, juvenile hormone biosynthesis, ovarian developmentand social environment in Bombus terrestris. Journal of Insect Physiology 46, 47–57.
Bloch, G., Hefetz, A., Hartfelder, K., 2000b. Ecdysteroid titer, ovary status, anddominance in adult worker and queen bumble bees (Bombus terrestris). Journalof Insect Physiology 46, 1033–1040.
Collins, A.M., Caperna, T.J., Williams, V., Garrett, W.M., Evans, J.D., 2006. Proteomicanalyses of male contributions to honey bee sperm storage and mating. InsectMolecular Biology 15, 541–549.
Corona, M., Velarde, R.A., Remolina, S.R., Moran-Lauter, A., Wang, Y., Hughes, K.A.,Robinson, G.E., 2007. Vitellogenin, juvenile hormone, insulin signaling, andqueen honey bee longevity. Proceedings of the Academy of Sciences of theUnited States of America 104, 7128–7133.
Dallai, R., 1975. Fine structure of the spermatheca of Apis mellifera. Journal of InsectPhysiology 21, 89–109.
Engels, W., Ramamurty, P.S., 1976. Initiation of oögenesis in allatectomised virginhoneybee queens by carbon dioxide treatment. Journal of Insect Physiology 22,1427–1432.
Engels, W., Imperatriz-Fonseca, V.L., 1990. Caste development, reproductivestrategies, and control of fertility in honey bees and stingless bees. In: Engels,W. (Ed.), Social Insects: An Evolutionary Approach to Castes and Reproduction.Springer, Berlin, pp. 167–230.
Engels, W., Kaatz, H., Zillikens, A., Simoes, Z.L.P., Trube, A., Braun, R., Dittrich, F.,1990. Honey bee reproduction: vitellogenin and caste-specific regulation offertility. In: Hoshi, M., Yamashita, O. (Eds.), Advances in InvertebrateReproduction, Elsevier, vol. 5. London, New York, pp. 495–502.
Fahrbach, S., Giray, T., Robinson, G.E., 1995. Volume changes in the mushroombodies of adult honey bee queens. Neurobiology of Learning and Memory 63,181–191.
Flanders, S.E., 1959. Embryonic starvation. An explanation of the defectivehoneybee egg. Journal of Economic Entomology 52, 166–167.
Fleig, R., 1995. Role of the follicle cells for yolk uptake in ovarian follicles of thehoney bee Apis mellifera L. (Hymenoptera: Apidae). International Journal ofInsect Morphology and Embryology 24, 427–433.
Fluri, P., Sabatini, A.G., Vecchi, M., Wille, H., 1981. Blood juvenile hormone, proteinand vitellogenin titres in laying and non-laying queen honeybees. Journal ofApicultural Research 20, 221–225.
Fluri, P., Lüscher, M., Wille, H., Gerig, L., 1982. Changes of weight of the pharyngealgland and haemolymph titres of juvenile hormone, protein and vitellogenin inworker honey bees. Journal of Insect Physiology 28, 61–68.
Gauthier, L., Ravallec, M., Tournaire, M., Cousserans, F., Bergoin, M., Dainat, B.,Miranda, J.D., 2011. Viruses associated with ovarian degeneration in Apismellifera L. queens. PLoS One 6, e16217, doi: 16210.11371/journal.pone.0016217.
Geva, S., Hartfelder, K., Bloch, G., 2005. Reproductive division of labor, dominance,and ecdysteroid levels in hemolymph and ovary of the bumble bee Bombusterrestris. Journal of Insect Physiology 51, 811–823.
Gilbert, L.E., 2012. Insect Endocrinology. Academic Press, London, Waltham, SanDiego.
Gobin, B., Ito, F., Peeters, C., Billen, J., 2006. Queen-worker differences inspermatheca reservoir of phylogenetically basal ants. Cell and Tissue Research326, 169–178.
Hagedorn, H.H., Kunkel, J.G., 1979. Vitellogenin and vitellin in insects. AnnualReview of Entomology 24, 475–505.
Hartfelder, K., 2000. Insect juvenile hormone: from ‘status quo’ to high society.Brazilian Journal of Medical and Biological Research 33, 157–177.
Hartfelder, K., Bitondi, M.M.G., Santana, W.C., Simoes, Z.L.P., 2002. Ecdysteroid titerand reproduction in queens and workers of the honey bee and of a stinglessbee: loss of ecdysteroid function at increasing levels of sociality? InsectBiochemistry and Molecular Biology 32, 211–216.
Hartfelder, K., Emlen, D.J., 2005. Endocrine control of insect polymorphism. In:Gilbert, L.I., Iatrou, K., Gill, S.S. (Eds.), Comprehensive Insect Molecular Science,vol. 3. Elsevier, Oxford, pp. 651–703.
Hitchcock, J.D., 1956. Honey bee queens whose eggs all fail to hatch. Journal ofEconomic Entomology 49, 11–14.
J. Wegener et al. / Journal of Insect Physiology 59 (2013) 655–661 661
Huang, Z.-Y., Robinson, G., Borst, D., 1994. Physiological correlates of division oflabor among similarly aged honey bees. Journal of Comparative Physiology A174, 731–739.
Huang, Z.-Y., Robinson, G.E., 1996. Regulation of honey bee division of labor bycolony age demography. Behavioral Ecology and Sociobiology 39, 147–158.
Ito, F., Ohkawara, K., 1994. Spermatheca size differentiation between queens andworkers in primitive ants – relationship with reproductive structure of colonies.Naturwissenschaften 81, 138–140.
Jassim, O., Huang, Z.Y., Robinson, G.E., 2000. Juvenile hormone profiles of workerhoney bees during normal and accelerated behavioral development. Journal ofInsect Physiology 46, 243–249.
Jaycox, E.R., Skowronek, W., Guynn, G., 1974. Behavioral changes in workerhoneybees (Apis mellifica L.) induced by injections of a juvenile hormonemimic. Annals of the Entomological Society of America 67, 529–535.
Koeniger, G., 1970. Bedeutung der Tracheenhülle und der Anhangdrüse derSpermatheka für die Befruchtungsfähigkeit der Spermatozoen derBienenkönigin (Apis mellifica L.). Apidologie 1, 55–71.
Lensky, Y., Schindler, H., 1967. Motility and reversible inactivation of honeybeespermatozoa in vivo and in vitro. Annales de l’Abeille 10, 5–16.
Lorenz, J.I., Lorenz, M.W., Hoffmann, K.H., 1997. Factors regulating juvenilehormone and ecdysteroid biosynthesis in Gryllus bimaculatus (Ensifera:Gryllidae). European Journal of Entomology 94, 369–379.
Lorenz, M.W., 2003. Adipokinetic hormone inhibits the formation of energy storesand egg production in the cricket Gryllus bimaculatus. ComparativeBiochemistry and Physiology B Comparative Biochemistry 136, 197–206.
Lorenz, M.W., 2007. Oogenesis-flight syndrome in crickets: age-dependent eggproduction, flight performance, and biochemical composition of the flightmuscles in adult female Gryllus bimaculatus. Journal of Insect Physiology 53,819–832.
Mackensen, O., 1947. Effect of CO2 on initial oviposition of artificially inseminatedand virgin queen bees. Journal of Economic Entomology 40, 344–349.
Martins, G.F., Serrao, J.E., 2002. A comparative study of the spermatheca in bees.Sociobiology 40, 711–720.
Nelson, C.M., Ihle, K.E., Fondrk, M.K., Page, R.E., Amdam, G.V., 2007. The genevitellogenin has multiple coordinating effects on social organization. PlosBiology 5, 673–677.
Paul, R.K., Takeuchi, H., Matsuo, Y., Kubo, T., 2005. Gene expression of ecdysteroid-regulated gene E74 of the honeybee in ovary and brain. Insect MolecularBiology 14, 9–15.
Pinto, L.Z., Bitondi, M.M., Simoes, Z.L.P., 2000. Inhibition of vitellogenin synthesis inApis mellifera workers by juvenile hormone analogue pyriproxyfen. Journal ofInsect Physiology 46, 153–160.
Raikhel, A.S., Brown, M.R., Belles, X., 2005. Hormonal control of reproductiveprocesses. In: Gilbert, L.I. (Ed.), Comprehensive Molecular Insect Science, vol. 3.Elsevier Pergamon, Oxford, pp. 433–491.
Robinson, G.E., Strambi, C., Strambi, A., Feldlaufer, M.F., 1991. Comparison ofjuvenile hormone and ecdysteroid haemolymph titers in adult worker andqueen honey bees (Apis mellifera). Journal of Insect Physiology 37, 929–937.
Robinson, G.E., Vargo, E.L., 1997. Juvenile hormone in adult eusocial hymenoptera:Gonadotropin and behavioral pacemaker. Archives of Insect Biochemistry andPhysiology 35, 559–583.
Röseler, P.F., 1977. Juvenile hormone control of oögenesis in Bumblebee workers,Bombus terrestris. Journal of Insect Physiology 23, 985–992.
Röseler, P.F., Röseler, I., 1988. Influence of juvenile hormone on fat body metabolismin ovariectomized queens of the bumblebee, Bombus terrestris. InsectBiochemistry 18, 557–563.
Ruttner, F., Koeniger, G., 1971. Filling of spermatheca of honeybee queen – activemigration or passive transport of spermatozoa. Zeitschrift für VergleichendePhysiologie 72, 411–422.
Ruttner, F., 1976. The Instrumental Insemination of the Queen Bee, 2nd ed.Apimondia, Bukarest.
Ruttner, F., 1983. Queen Rearing – Biological Basis and Technical Instructions, 2nded. Apimondia, Bucharest.
Rutz, W., Lüscher, L., 1974. The occurrence of vitellogenin in workers and queens ofApis mellifica and the possibility of its transmission to the queen. Journal ofInsect Physiology 20, 897–909.
Rutz, W., Gerig, L., Wille, H., Lüscher, M., 1976. The function of juvenile hormone inadult worker honey bees, Apis mellifera. Journal of Insect Physiology 22, 1485–1491.
Slone, J.D., Stout, T.L., Huang, Z.Y., Schneider, S.S., 2012. The influence of dronephysical condition on the likelihood of receiving vibration signals from workerhoney bees, Apis mellifera. Insectes Sociaux 59, 101–107.
Smith, A.R., Kapheim, K.M., Pérez-Ortega, B., Brent, C.S., Wcislo, W.T., 2013. Juvenilehormone levels reflect social opportunities in the facultatively eusocial sweatbee Megalopa genalis (Hymenoptera: Halictidae). Hormones and Behavior 63, 1–4.
SPSS, 2010. IBM SPSS Statistics 19 core system user’s guide, IBM-SPSS.Strambi, A., 1990. Physiology and reproduction in social wasps. In: Engels, W. (Ed.),
Social Insects – an Evolutionary Approach to Caste and Reproduction. Springer,Berlin, pp. 59–75.
Sullivan, J., Fahrbach, S., Harrison, J., Capaldi, E., Fewell, J., Robinson, G., 2003.Juvenile hormone and division of labor in honeybee colonies: effects ofallatectomy on flight behavior and metabolism. Journal of ExperimentalBiology 206, 2287–2296.
Verma, L.R., Shuel, R.W., 1973. Respiratory metabolism of the semen of the honey-bee, Apis mellifera. Journal of Insect Physiology 19, 97–103.
Verma, L.R., 1978. Biology of honeybee spermatozoa 2. Effect of washing and sugarson respiration as measured by the cartesian diver technique. Journal ofApicultural Research 17, 176–181.
Wang, Y.J., Kaftanoglu, O., Siegel, A., Page, R.E., Amdam, G., 2010. Surgicallyincreased ovarian mass in the honey bee confirms link between reproductivephysiology and worker behavior. Journal of Insect Physiology 56, 1816–1824.
Wegener, J., Lorenz, M.W., Bienefeld, K., 2009a. Physiological consequences ofprolonged nursing in the honey bee. Insectes Sociaux 56, 85–93.
Wegener, J., Huang, Z.Y., Lorenz, M.W., Bienefeld, K., 2009b. Regulation ofhypopharyngeal gland activity and oogenesis in honey bee (Apis mellifera)workers. Journal of Insect Physiology 55, 716–725.
Wegener, J., Lorenz, M.W., Bienefeld, K., 2010. Differences between queen- andworker-laid eggs of the honey bee (Apis mellifera). Apidologie 41, 116–126.
Wegener, J., Zschörnig, K., Onischke, K., Fuchs, B., Schiller, J., Müller, K., 2013.Conservation of honey bee (Apis mellifera) sperm phospholipids during storagein the bee queen – a TLC/MALDI-TOF MS study. Experimental Gerontology 48,213–222.
West-Eberhard, M.J., 1996. Wasp societies as microcosms for the study ofdevelopment and evolution. In: Turillazzi, S., West-Eberhard, M.J. (Eds.),Natural History and Evolution of Paper Wasps. Oxford Press, New York, pp.290–317.
Yamazaki, Y., Kiuchi, M., Takeuchi, H., Kubo, T., 2011. Ecdysteroid biosynthesis inworkers of the European honeybee Apis mellifera L. Insect Biochemistry andMolecular Biology 41, 283–293.