1. No category

Report
Dual Effect of Wasp Queen Pheromone in Regulating
Insect Sociality
Highlights
Authors
d
Specific queen pheromones induce sterility in social insect
workers
Cintia A. Oi, Annette Van Oystaeyen, ...,
Jelle S. van Zweden, Tom Wenseleers
d
One of these queen pheromones is shown to also signal egg
maternity
Correspondence
d
d
The pheromone enables workers to recognize or ‘‘police’’
eggs laid by cheater workers
This shows that queen pheromones regulate insect sociality
in several distinct ways
Oi et al., 2015, Current Biology 25, 1–3
June 15, 2015 ª2015 Elsevier Ltd All rights reserved
http://dx.doi.org/10.1016/j.cub.2015.04.040
[email protected] (C.A.O.),
[email protected] (T.W.)
In Brief
Social insect queens use pheromones to
stop the workers from reproducing. Oi
et al. show that in wasps, one of these
sterility-inducing pheromones is also
used by the queen to mark her eggs and
enable the workers to recognize and
‘‘police’’ eggs laid by other workers. This
shows that queen pheromones regulate
insect sociality in several distinct ways.
Please cite this article in press as: Oi et al., Dual Effect of Wasp Queen Pheromone in Regulating Insect Sociality, Current Biology (2015), http://
dx.doi.org/10.1016/j.cub.2015.04.040
Current Biology
Report
Dual Effect of Wasp Queen Pheromone
in Regulating Insect Sociality
Cintia A. Oi,1,* Annette Van Oystaeyen,1 Ricardo Caliari Oliveira,1 Jocelyn G. Millar,2 Kevin J. Verstrepen,3,4
Jelle S. van Zweden,1 and Tom Wenseleers1,*
1Laboratory
of Socioecology and Social Evolution, KU Leuven, Naamsestraat 59, 3000 Leuven, Belgium
of Entomology and Department of Chemistry, University of California, Riverside, Riverside, CA 92521, USA
3VIB Laboratory for Systems Biology, KU Leuven, Gaston Geenslaan 1, 3001 Leuven, Belgium
4CMPG Laboratory for Genetics and Genomics, KU Leuven, Gaston Geenslaan 1, 3001 Leuven, Belgium
*Correspondence: [email protected] (C.A.O.), [email protected] (T.W.)
http://dx.doi.org/10.1016/j.cub.2015.04.040
2Department
SUMMARY
Eusocial insects exhibit a remarkable reproductive
division of labor between queens and largely sterile
workers [1, 2]. Recently, it was shown that queens
of diverse groups of social insects employ specific,
evolutionarily conserved cuticular hydrocarbons to
signal their presence and inhibit worker reproduction
[3]. Workers also recognize and discriminate between eggs laid by the queen and those laid by
workers, with the latter being destroyed by workers
in a process known as ‘‘policing’’ [4, 5]. Worker
policing represents a classic example of a conflictreducing mechanism, in which the reproductive
monopoly of the queen is maintained through the
selective destruction of worker-laid eggs [5, 6]. However, the exact signals used in worker policing have
thus far remained elusive [5, 7]. Here, we show that
in the common wasp, Vespula vulgaris, the pheromone that signals egg maternity and enables the
workers to selectively destroy worker-laid eggs is in
fact the same as one of the sterility-inducing queen
signals that we identified earlier [3]. These results
imply that queen pheromones regulate insect sociality in two distinct and complementary ways, i.e., by
signaling the queen’s presence and inhibiting worker
reproduction, and by facilitating the recognition and
policing of worker-laid eggs.
RESULTS AND DISCUSSION
Even though social insects may appear to work together in total
harmony, social insect colonies are in fact rife with conflict [1, 5].
A case in point is the queen-worker conflict over male production, observed in many species, whereby rogue workers will surreptitiously lay unfertilized eggs, destined to become males,
instead of working for the benefit of the colony [5]. To keep
such transgressing workers at bay, many species have evolved
a ‘‘policing’’ system in which eggs laid by workers are selectively
detected and cannibalized [5, 8]. First discovered in the honeybee a quarter of a century ago [4], this phenomenon of worker
policing has since been reported in more than a dozen social insect species, including bees, ants, and wasps [8]. It has long
been hypothesized that the workers’ ability to discriminate between queen-laid and worker-laid eggs might be aided by a
pheromone mark left on the queen’s eggs [5]. Such a queen
egg-marking pheromone should be evolutionarily stable
because the pheromone would benefit both the queen, by
increasing her share of the colony’s reproduction, and the
workers, by reducing the chance of them accidentally destroying
the queen’s eggs and thereby reducing their inclusive fitness
[5, 8–10]. However, despite decades of research, no social insect queen egg-marking pheromone has been unambiguously
identified [5, 7, 11].
Cuticular hydrocarbons provide a waxy protective layer on the
cuticles of all insect life stages, including eggs [12, 13]. The profiles of cuticular chemicals are known to show pronounced differences between castes [3, 11] and have acquired important
signaling functions in social insects, including as queen pheromones [3]. We hypothesized that the same hydrocarbons that
function as queen pheromones might also play a role in signaling
egg maternity. That is, based on evolutionary parsimony, we predicted that the co-option and repurposing of an existing maternal
signal might be easier than evolving an entirely new signaling
system from scratch. To test this hypothesis, we carried out experiments with the common wasp, Vespula vulgaris (Figure 1A),
from which we recently identified several sterility-inducing queen
pheromones [3], and in which the workers, as in the honeybee,
police each other’s eggs [14]. To shortlist possible egg-marking
pheromones, we first carried out a detailed comparison of the
hydrocarbon profiles of queen-laid and worker-laid eggs ([15];
Figure S1 and Supplemental Experimental Procedures). Our
analysis showed that two compounds were consistently more
abundant on the surface of queen-laid eggs than on the surface
of worker-laid eggs, namely 3-methylnonacosane (3-MeC29),
which we earlier found to act as a sterility-inducing queen signal
[3], and 3-methylheptacosane (3-MeC27), which shows weaker
caste differences on the cuticle but is highly characteristic of
queen-laid eggs (Figure S1).
Subsequently, we collected six colonies of the common wasp,
from which we isolated about half of the worker force from their
mother queen to stimulate egg laying. We then treated workerlaid eggs with synthetic versions of the queen-characteristic
pheromones 3-MeC27 and 3-MeC29, either individually or as a
blend (‘‘mix,’’ Figure 1B), dissolved in acetone. These treatments
Current Biology 25, 1–3, June 15, 2015 ª2015 Elsevier Ltd All rights reserved 1
Please cite this article in press as: Oi et al., Dual Effect of Wasp Queen Pheromone in Regulating Insect Sociality, Current Biology (2015), http://
dx.doi.org/10.1016/j.cub.2015.04.040
A
B
Figure 1. Experimental Setup Used to Test the Role of Queen Pheromones in Signaling Egg Maternity during Worker Policing in the
Common Wasp
A
B
Figure 2. A Sterility-Inducing Queen Pheromone Also Protects the
Queen’s Eggs against Worker Policing in the Common Wasp
(A) To test the hypothesis that hydrocarbon queen pheromones, which
stop workers from reproducing [3], could also play a role in signaling queen
egg maternity, we performed experiments with the common wasp Vespula
vulgaris. As in most social insects, common wasp queens have a near
monopoly over reproduction [14]. This reproductive monopoly is enforced
partly via a mechanism of ‘‘worker policing,’’ whereby workers selectively
detect and destroy eggs laid by transgressing workers [14].
(B) In bioassays, we determined whether treatment with an acetone solution of
the hydrocarbon 3-methylnonacosane (3-MeC29), one of the main sterilityinducing queen signals in this species [3]; 3-methylheptacosane (3-MeC27),
which is characteristic of queen-laid eggs; or a mix of both caused worker-laid
eggs to become more ‘‘queen-like’’ and thus acceptable to the workers upon
reintroduction into their mother colony.
(A) Treatment of worker-laid eggs with acetone solutions of the queen egg
characteristic compounds 3-MeC29 and 3-MeC27, either as single compounds
(‘‘single’’) or as a blend of the two compounds (‘‘mix’’), caused their relative
abundance to significantly increase relative to the control and to approach
those found naturally on queen-laid eggs (mean ± SEM, Tukey’s post hoc
tests, n = 18 eggs each per egg type and treatment from N = 6 colonies each;
Table S1). Compound concentrations were compared separately but significance levels are presented only once, given that the results were the same for
both compounds (***p < 0.001; NS, not significant).
(B) Our bioassays confirmed that 3-MeC29 also signals egg maternity, because
the percentage of eggs that were policed was significantly reduced when
3-MeC29 was applied onto worker-laid eggs compared to the solvent-only
control (mean ± 95% confidence level, binomial mixed model, Table S1; N = 6
replicate colonies, n = 625–724 eggs per treatment; ***p < 0.001; NS, not
significant).
rendered worker-laid eggs chemically more similar to queen-laid
eggs (Figure 2A; Table S1), and our hypothesis was that application of these compounds should therefore reduce the rate at
which these eggs were policed. By reintroducing the chemically
manipulated eggs into the mother colony (Figure 1B), we
confirmed this hypothesis. Specifically, the bioassays confirmed
that 3-MeC29 was strongly correlated with signaling egg maternity because it significantly reduced the rate at which workerlaid eggs were policed compared to solvent-treated controls
(Figure 2B; Table S1). By contrast, 3-MeC27, which is present
in significantly higher amounts on queen-laid eggs than on
worker-laid eggs, did not inhibit policing, and the blend of both
compounds was not significantly different from 3-MeC29 used
alone. The fact that applying the compound 3-MeC29 onto
worker-laid eggs only partially protected them against policing,
however, indicates that our treatments were only partially effective at mimicking the complete profiles of queen-laid eggs. This
could be explained either by small concentration mismatches
(Figure 2A) or by the fact that worker-laid eggs also contain
some specific cues that give away their maternal origin, and
which we could not experimentally remove. Indeed, our chemical analysis reveals several short-chain mono- and dimethyl
2 Current Biology 25, 1–3, June 15, 2015 ª2015 Elsevier Ltd All rights reserved
Please cite this article in press as: Oi et al., Dual Effect of Wasp Queen Pheromone in Regulating Insect Sociality, Current Biology (2015), http://
dx.doi.org/10.1016/j.cub.2015.04.040
alkanes that are specific to worker-laid eggs (Figure S1) and
might be produced as some intrinsic side effect of worker-specific physiological processes [16]. Hence, 3-MeC29 forms an
important part of the queen signal, but an additional contribution
of other queen- or worker-specific compounds cannot be
entirely excluded. In fact, a recent study has shown that in
ants, a hydrocarbon fertility signal is maximally effective only
when perceived within its full chemical context [17]. It is likely
that the same is true for the queen egg-marking signal that we
have documented here.
Overall, our study is the first to unambiguously identify a social
insect queen egg-marking pheromone used in worker policing.
Furthermore, our results show that the egg-marking pheromone
appears to be the same as one of the main sterility-inducing queen
signals in this species [3]. From an evolutionary perspective, these
findings make sense because a queen pheromone has all the
necessary attributes to be a marker of maternal origin and could
easily be co-opted into a valuable secondary function as an
egg-marking pheromone. A dual function of queen pheromones
in inducing sterility and egg marking had been suggested previously for volatile queen pheromone components in the termite
Reticulitermes speratus [18, 19], as well as for non-volatile hydrocarbons in the ants Camponotus floridanus [20] and Pachycondyla
inversa [11]. In the latter two cases, particular long-chain hydrocarbons are highly specific for queens and their eggs, but a dual
function in communication and regulation of sociality was not
confirmed with defined blends of pure compounds, causing the
functional roles of these compounds to remain speculative in
these cases. Nonetheless, these results together with ours clearly
suggest that hydrocarbon queen pheromones have a central role
in regulating the spectacular reproductive division of labor of
diverse lineages of hymenopteran social insects.
SUPPLEMENTAL INFORMATION
Supplemental Information includes one figure, one table, and Supplemental
Experimental Procedures and can be found with this article online at http://
dx.doi.org/10.1016/j.cub.2015.04.040.
AUTHOR CONTRIBUTIONS
C.A.O., J.S.v.Z., and T.W. designed and contributed to all aspects of this
study. A.V.O. first conceived the experiments and provided some of the
bioassay results. R.O.C. helped to collect nests and perform bioassays.
J.G.M. synthetized the methyl-branched hydrocarbons. K.J.V. provided the
GC-MS instrument used to analyze the chemical profiles of eggs. C.A.O.,
J.S.v.Z., and T.W. wrote the manuscript.
ACKNOWLEDGMENTS
This study was supported by Research Foundation Flanders grant G.0A51.15
and KU Leuven Centre of Excellence grant PF/2010/007 (to T.W.), postdoctoral fellowship 12Q7615N of the Research Foundation Flanders (to
J.S.v.Z.), CNPq-Brazil scholarships 201959/2012-7 and 238127/2012-5 (to
C.A.O. and R.C.O.), and Hatch project CA-R*ENT-5181-H (to J.G.M.). We
are grateful to A. Vollet-Neto and A. Vandoren for assistance with fieldwork.
Received: March 7, 2015
Revised: April 20, 2015
Accepted: April 20, 2015
Published: May 7, 2015
REFERENCES
1. Bourke, A.F.G. (2011). Principles of Social Evolution. (Oxford University
Press).
2. Ratnieks, F.L.W., Foster, K.R., and Wenseleers, T. (2011). Darwin’s special
difficulty: the evolution of ‘‘neuter insects’’ and current theory. Behav.
Ecol. Sociobiol. 65, 481–492.
3. Van Oystaeyen, A., Oliveira, R.C., Holman, L., van Zweden, J.S., Romero,
C., Oi, C.A., d’Ettorre, P., Khalesi, M., Billen, J., Wa¨ckers, F., et al. (2014).
Conserved class of queen pheromones stops social insect workers from
reproducing. Science 343, 287–290.
4. Ratnieks, F.L.W., and Visscher, P.K. (1989). Worker policing in the honeybee. Nature 342, 796–797.
5. Ratnieks, F.L.W., Foster, K.R., and Wenseleers, T. (2006). Conflict resolution in insect societies. Annu. Rev. Entomol. 51, 581–608.
6. Ratnieks, F.L.W., and Wenseleers, T. (2008). Altruism in insect societies
and beyond: voluntary or enforced? Trends Ecol. Evol. 23, 45–52.
7. Martin, S.J., Chaˆline, N., Ratnieks, F.L.W., and Jones, G.R. (2005).
Searching for the egg-marking signal in honeybees. J. Negative Results
2, 1–9.
8. Wenseleers, T., and Ratnieks, F.L.W. (2006). Comparative analysis of
worker reproduction and policing in eusocial hymenoptera supports relatedness theory. Am. Nat. 168, E163–E179.
9. Ratnieks, F. (1988). Reproductive harmony via mutual policing by workers
in eusocial Hymenoptera. Am. Nat. 132, 217–236.
10. Wenseleers, T., Helantera¨, H., Alves, D.A., Duen˜ez-Guzma´n, E., and
Pamilo, P. (2013). Towards greater realism in inclusive fitness models:
the case of worker reproduction in insect societies. Biol. Lett. 9, 20130334.
11. van Zweden, J.S., Heinze, J., Boomsma, J.J., and d’Ettorre, P. (2009). Ant
queen egg-marking signals: matching deceptive laboratory simplicity with
natural complexity. PLoS ONE 4, e4718.
12. Blomquist, G.J., and Bagne`res, A.G., eds. (2010). Insect Hydrocarbons:
Biology, Biochemistry, and Chemical Ecology (Cambridge University
Press).
13. Blum, M.S., and Hilker, M. (2002). Chemical protection of insect eggs. In
Chemoecology of Insect Eggs and Egg Deposition, M. Hilker, and T.
Meiners, eds. (Blackwell), pp. 61–90.
14. Foster, K.R., and Ratnieks, F.L. (2001). Convergent evolution of worker
policing by egg eating in the honeybee and common wasp. Proc. Biol.
Sci. 268, 169–174.
15. Bonckaert, W., Drijfhout, F.P., d’Ettorre, P., Billen, J., and Wenseleers, T.
(2012). Hydrocarbon signatures of egg maternity, caste membership and
reproductive status in the common wasp. J. Chem. Ecol. 38, 42–51.
16. Oi, C.A., van Zweden, J.S., Oliveira, R.C., Van Oystaeyen, A., Nascimento,
F.S., and Wenseleers, T. (2015). The origin and evolution of social insect
queen pheromones: Novel hypotheses and outstanding problems.
BioEssays. Published online April 27, 2015. http://dx.doi.org/10.1002/
bies.201400180.
17. Smith, A.A., Millar, J.G., and Suarez, A.V. (2015). A social insect fertility
signal is dependent on chemical context. Biol. Lett. 11, 20140947.
18. Matsuura, K., Himuro, C., Yokoi, T., Yamamoto, Y., Vargo, E.L., and Keller,
L. (2010). Identification of a pheromone regulating caste differentiation in
termites. Proc. Natl. Acad. Sci. USA 107, 12963–12968.
19. Matsuura, K. (2012). Multifunctional queen pheromone and maintenance
of reproductive harmony in termite colonies. J. Chem. Ecol. 38, 746–754.
20. Endler, A., Liebig, J., Schmitt, T., Parker, J.E., Jones, G.R., Schreier, P.,
and Ho¨lldobler, B. (2004). Surface hydrocarbons of queen eggs regulate
worker reproduction in a social insect. Proc. Natl. Acad. Sci. USA 101,
2945–2950.
Current Biology 25, 1–3, June 15, 2015 ª2015 Elsevier Ltd All rights reserved 3
Current Biology
Supplemental Information
Dual Effect of Wasp Queen Pheromone
in Regulating Insect Sociality
Cintia A. Oi, Annette Van Oystaeyen, Ricardo Caliari Oliveira, Jocelyn G. Millar,
Kevin J. Verstrepen, Jelle S. van Zweden, and Tom Wenseleers
Queen vs worker-laid eggs
QLE vs WLE
Queen vs worker cuticle
FD
1.8
2.1
1.2
1.3
1.2
1.6
1.6
1.9
1.4
1.3
1.3
1.1
1.2
1.1
1.1
1.0
1.1
-1.0
-1.1
-1.0
-1.1
-1.1
-1.2
-1.2
-1.2
-1.4
-1.3
-1.3
-1.2
-1.2
-1.1
-1.2
-1.2
-1.3
-1.7
-1.6
-1.4
-1.6
-1.6
-1.4
-2.1
-1.9
-1.5
-2.0
-1.8
-1.5
-2.0
-2.0
-2.1
3-MeC27
3-MeC29
C25
C26
C27
7-MeC27
C29:1
11,17-, 13,17-, 15,19-diMeC31
C31:1
5,9-diMeC29
11-, 13-, 15-MeC31
C28
11-, 13-, 15-MeC29
10-, 12-, 14-, 16-MeC28
C27:1 (+isomer)
4-MeC26
7-MeC29
x,y-diMeC30
5-MeC27
3-MeC25
11-, 13-MeC27
4,x-diMeC28
11,15-diMeC27
C25:1 (+isomer)
C24
4,x-diMeC26
4-MeC28
C23
C31
5,x-diMeC27
C30
10-, 12-, 14-MeC26
C29
7-MeC25
4-MeC24
4,x-diMeC24
3,x-diMeC27
6-MeC24
3,x-diMeC25
5-MeC25
5,x -diMeC23
3,x-diMeC23
5,x-diMeC25
3-MeC23
10-,12-,14-MeC24
11-, 13-, 15-MeC25
7-MeC23
5-MeC23
9-, 11-MeC23
1
2
3
4
5
6
Colony
2
4
5
padj
0.0075
0.045
0.047
0.045
0.045
0.045
0.034
0.034
0.034
0.034
0.034
0.030
0.030
0.030
0.012
Q vs W
FD
1.4
7.4
-1.1
1.6
3.6
-1.1
1.2
5.5
1.4
-2.0
1.5
3.2
1.0
-1.5
-1.2
-1.9
1.2
1.2
-1.3
-2.8
-2.0
-1.4
-2.1
-2.0
-2.3
-2.6
1.8
-3.6
3.3
-1.9
-1.2
-2.6
5.2
-3.4
-3.8
-2.8
-2.4
-3.4
-2.8
-3.3
-2.6
-3.3
-2.5
-4.9
-3.7
-3.2
-4.0
-4.0
-3.9
padj
0.012
0.0054
0.0025
0.0059
0.0025
0.039
0.019
0.029
0.039
0.037
0.039
0.039
0.0069
0.039
0.032
0.012
0.012
0.034
0.0057
0.032
0.031
0.021
0.034
0.023
0.039
0.031
0.049
0.039
0.039
0.029
0.034
0.034
0.032
0.032
0.034
6
Fold difference
-8
-4
-2
0
2
4
8
Supplemental Information Supplemental Figure S1 Figure S1. Queen‐characteristic hydrocarbons in the common wasp Vespula vulgaris (related to Fig. 1B). Based on a reanalysis of the data from our earlier study [S1], it can be seen that only two hydrocarbons, 3‐methylheptacosane (3‐MeC27) and 3‐methylnonacosane (3‐MeC29), are much more abundant on the surface of queen‐laid eggs (QLE) than worker‐laid eggs (WLE) (fold difference in relative peak area > 1.5 and FDR corrected p‐level < 0.05, highlighted in red in the QLE vs. WLE columns). One of these compounds, 3‐MeC29, was earlier shown to act as a queen signal that stops workers from reproducing [S2], together with the queen‐characteristic linear alkanes C27, C28 and C29 (highlighted in red in the Q vs. W columns). Analyses of the cuticle are based on the comparison of relative peak areas, and for egg extracts are based on the average relative peak areas of 5 to 10 QLEs and WLEs each per colony from a total of 6 colonies, whereas for cuticle extracts they are based on the average relative peak areas of 10 workers per colony as well as that of their mother queen from a total of 4 colonies. Subsequently, colony‐average Log2 transformed relative peak areas of the 49 hydrocarbon peaks present were compared using paired t‐tests, after which significance levels were FDR corrected using the Benjamini‐Hochberg procedure. Further methodological details can be found in ref. [S1]. (FD = fold difference in relative peak areas, padj = Benjamini‐Hochberg FDR corrected p values, QLE and WLE = hexane extract of queen‐laid and worker‐laid eggs, Q and W = cuticular hexane extract of mature egg‐
laying queens and nonreproductive workers, Cx and Cx:1 = linear alkanes and alkenes with chain length x, y‐MeCx = monomethyl branched alkanes, y,z‐diMeCx = dimethyl branched alkanes) Supplemental Table S1. Significance of the effect of treating worker‐laid eggs with queen‐
characteristic hydrocarbons and of the treatments on egg policing rates. Application of synthetic versions of the queen‐characteristic compounds 3‐methylheptacosane (3‐MeC27, A) and 3‐methylnonacosane (3‐MeC29, B), or a mix of both, onto worker‐laid eggs (WLE) resulted in relative peak areas of 3‐MeC27 and 3‐MeC29 that were significantly higher than in the control WLEs, and non‐significantly different from the levels observed in queen‐laid eggs (QLE). Significance levels are based on Tukey posthoc tests carried out on a mixed model in which the relative peak area of each of the two compounds were coded as the dependent variable and colony and treatment were coded as a random and fixed factors, respectively (***: p < 0.001, n = 18 eggs each per egg type and treatment condition from 6 replicate colonies each). Policing bioassays (C) confirm that treatment of eggs with the queen pheromone 3‐MeC29 [S2], either alone or in a mix with 3‐MeC27, significantly reduced the odds for eggs to be policed compared to the control. Egg policing rates also showed a significant positive association with the number of workers present in each discriminator colony, but this was not confounded with treatment due to the paired experimental design that we used. Significance levels are based on a binomial mixed model in which discriminator colony and treatment were coded as random and fixed factors and the number of cells in each test comb and the number of workers in the discriminator colony were included as continuous covariates. (***: p < 0.001, odds ratio = odds of worker‐laid eggs being policed relative to control eggs) A) Effect of 3‐MeC27 treatment 3‐MeC27 vs Control WLE Mix vs Control WLE QLE vs Control WLE 3‐MeC27 vs Mix 3‐MeC27 vs QLE Mix vs QLE Coefficient SE 0.017 0.017 0.024 0.00027 ‐0.0073 ‐0.0076 0.0043
0.0043
0.0061
0.0040
0.0066
0.0066
B) Effect of 3‐MeC29 treatment Coefficient 3‐MeC29 vs Control WLE Mix vs Control WLE QLE vs Control WLE 3‐MeC29 vs Mix 3‐MeC29 vs QLE Mix vs QLE C) Effect on egg policing rates (Control) 3‐MeC27 3‐MeC29 Mix Nr. of cells Nr. of workers 0.0087 0.0072 0.012 0.0015 ‐0.0036 ‐0.0050 Coefficient ‐2.22 ‐0.19 ‐0.44 ‐0.44 0.0012 0.0067 SE 0.0017
0.0017
0.0023
0.0016
0.0024
0.0024
SE 0.41 0.12 0.12 0.12 0.0027
0.0014
Odds ratio 0.82 0.65 0.64 z value p‐value 3.85 3.79 3.97 0.067 ‐1.12 ‐1.16 < 0.001 < 0.001 < 0.001 1.00 0.67 0.65 *** *** *** z value p‐value 5.05 < 0.001 *** 4.18 < 0.001 *** 5.34 < 0.001 *** 0.94 0.78 ‐1.46 0.46 ‐2.06 0.16 z value p‐value ‐5.40 ‐1.60 ‐3.61 ‐3.55 0.45 4.83 6.9 x 10‐8 0.11 0.0003 *** 0.0004 *** 0.65 1.3 x 10‐6 *** Supplemental methods Identification of putative egg‐marking pheromones To identify putative queen egg‐marking pheromones in the common wasp Vespula vulgaris, we reanalyzed the raw hydrocarbon profile data of queen‐laid eggs (QLEs) and worker‐laid eggs (WLEs) as reported on in our earlier study [S1]. Comparison of the average profiles of 5 to 10 WLEs and QLEs from a total of 6 colonies (total n QLE=44, total n WLE=39) demonstrates that only two hydrocarbons, 3‐methylheptacosane (3‐MeC27) and 3‐methylnonacosane (3‐MeC29), were consistently much more abundant on the surface QLEs than WLEs (avg. fold difference in relative peak area > 1.5 and FDR corrected p‐level < 0.05, based on paired t‐tests on Log2 transformed relative peak areas, Fig. S1). One of these compounds, 3‐MeC29, was earlier shown to act as a queen signal that stops workers from reproducing [S2], together with the queen‐
characteristic linear alkanes C27, C28 and C29 (Fig. S1). Based on this, we hypothesized that 3‐
MeC29 could act as a maternal marker in two different contexts, and play a role both as a sterility‐inducing queen signal as well as a queen egg‐marking pheromone used to signal maternity and stop workers from accidentally policing the queen’s eggs. Policing bioassays To test the bioactivity of the putative egg‐marking pheromones 3‐MeC29 and 3‐MeC27 we treated worker‐laid eggs with an acetone solution of synthetic versions of the two compounds, either alone or in a mix of both, to make them more queen‐like (Fig. 1B) and tested if this reduced the rate at which these eggs were policed compared to a solvent‐only sham treated control. This was done by pipetting 5 µl of a 0.15 µg/µL (for 3‐MeC29) or 0.35 µg/µL (for 3‐MeC27) acetone solution of either compound, or a mix of both, onto each worker‐laid egg (Fig. 1B). These concentrations were chosen to optimally mimic the concentration of these compounds present naturally on queen‐laid eggs. Fig. 2A and Tables S1A and S1B show that this treatment was indeed effective, as the application of 3‐MeC27 and 3‐MeC29, or a mix of both, onto worker‐laid eggs resulted in relative peak areas of 3‐MeC27 (Table S1 A) and 3‐MeC29 (Table S1 B) that were significantly higher than in the control eggs, and non‐significantly different from the levels observed in queen‐laid eggs (GLMM with colony as random factor). Subsequently, the acetone was allowed to evaporate for 10 min and the test combs filled with worker‐laid eggs exposed to our different queen pheromone treatments were placed on a wire mesh support inside a discriminator colony to check egg policing rates (Fig. 1B). A comb containing brood and the mother queen was placed on second mesh above the test comb, after which the workers were added (ca. 200‐300 individuals). The wire mesh acted as a queen excluder, as it had a size that enabled the workers, but not the queen, to pass through. In this way, the queen was confined to the upper compartment and prevented from depositing new eggs in the lower test comb during the experiment. The two combs were placed on strips of cardboard to provide an air space between the wire mesh and to allow the workers free access to all the cells. The nest box was connected to a foraging box in which food and water (mealworms, Apifonda sugar paste and a moist cotton ball) were provided ad libitum in falcon tubes. At the start of each experiment, a cell map was drawn of the test comb in which the location of all the worker‐laid eggs inside the cells as well as the treatments they were exposed to were indicated. Then, 24 hours later, we recorded the number of eggs that were policed by the workers in the different treatments (cf. [S3]) and also counted the total number of workers that were present inside the discriminator colony (average number in the end of the experiment: 271, SD = 115, N=6 replicates). Combs with fresh worker‐
laid eggs were obtained from queenless colonies containing ca. 200‐300 workers each, kept in the same type of nest box but provided with just a single empty comb for egg deposition. Ca. 10 days after queen removal, these workers started to lay eggs [S3], thereby enabling us to collect nearly full combs of worker‐laid eggs a couple of days later (i.e. after ca 14 days). Colonies were placed in a temperature controlled climate room at 30 C. The total dimension of the nest plus foraging box was 36 x 15 x 15 cm (width x depth x height). The box was covered with a perspex lid and provided with a ventilation mesh, and the nest box was kept in the dark using a black plastic cover. All experiments were replicated six times, using six common wasp colonies collected in the vicinity of Leuven (Belgium) in July and August 2014, which were split in two parts, to serve as a source of worker‐laid eggs (for the part without a queen) or as a discriminator colony (for the part with the mother queen). Synthesis of 3‐MeC29 and 3‐MeC27 was performed as described in [S2]. Egg policing rates were compared across the different treatments using a binomial mixed model with logit link function using lme4 v. 1.1‐8 in R v. 3.1.1. In this model, discriminator colony and treatment were coded as random and fixed factors, respectively, whereas the number of cells in each test comb and the total number of workers present in the discriminator colony were included as continuous covariates. Statistical significance of the difference relative to the solvent control was assessed using Wald z tests. Overdispersion was checked for, but found not to be significant. Final results were expressed in terms of the estimated mean percentage of eggs eaten in the different treatments and control, and were calculated using the effects package, together with asymptotic 95% confidence limits. Supplemental references S1. S2. S3. Bonckaert, W., Drijfhout, F.P., d'Ettorre, P., Billen, J., and Wenseleers, T. (2012). Hydrocarbon signatures of egg maternity, caste membership and reproductive status in the common wasp. J. Chem. Ecol. 38, 42‐51. Van Oystaeyen, A., Oliveira, R.C., Holman, L., Van Zweden, J.S., Romero, C., Oi, C.A., d'Ettorre, P., Khalesi, M., Billen, J., Wäckers, F., et al. (2014). Conserved class of queen pheromones stops social insect workers from reproducing. Science 287, 287‐290. Foster, K.R., and Ratnieks, F.L. (2001). Convergent evolution of worker policing by egg eating in the honeybee and common wasp. Proc. R. Soc. Lond. B 268, 169‐174.