Physiological Entomology (2004) 29, 464–471
Epicuticular lipids and fertility in primitively social wasps (Hymenoptera Stenogastrinae) S. TURILLAZZI1,2, M. F. SLEDGE1, L. DAPPORTO1, M. LANDI1, D. FANELLI1, L. FONDELLI1, P. ZANETTI1 and F . R . D A N I 2 1
Dipartimento di Biologia Animale e Genetica and 2Centro Interdipartimentale di Spettrometria di Massa, Universita` di Firenze, Firenze, Italy Abstract. Lipid cuticular profiles of females of four species belonging to three
different genera of stenogastrine wasps are examined in connection with reproductive potential (relative ovarian development). Cuticular lipids may not only represent the cues for nestmate discrimination (already behaviourally ascertained in three of the same species), but also allow discrimination of fertile and non fertile individuals. Comparisons with more socially evolved insects are reported and discussed. Key words. Chemical communication, epicuticular lipids, fertility, social wasps, Stenogastrinae.
Introduction The outer layer of the insect exoskeleton comprises a mixture of lipids that covers the entire body. This layer offers an efficient protection against the loss of water and penetration of microorganisms, and has been one of the major contributing factors in the evolutionary success of the insects. The main components of the lipid mixture are long-chained hydrocarbons, mainly alkanes, alkenes and methyl-branched alkanes. Epicuticular hydrocarbons have secondarily acquired communicative roles in many insects. In solitary insects, they may function as contact pheromones enabling recognition of sexual partners or as kairomones when they make the recognition of a host by parasitoids possible (Howard, 1993). In social insects (mainly ants, bees, wasps and termites), cuticular lipids are involved in a number of communicative functions, especially in inter- and intracolonial communication. The proportions of cuticular lipid components have been shown to be both species and colony-specific, furnishing potential cues to recognize nestmates and discriminate alien individuals and social parasites (Howard, 1993; Smith & Breed,
Correspondence: Stefano Turillazzi, Dipartimento di Biologia Animale e Genetica, Universita` di Firenze, via Romana 17, 50125, Firenze, Italy. Tel.: þ39 055 2288218; fax: þ39 055 222565; e-mail:
[email protected]
464
1995; references in Vandermeer et al., 1998). Recent research has highlighted several additional roles possibly played by cuticular compounds within colonies. One of the most important is the signalling of reproductive status by fertile individuals. Queens and egg-laying individuals are predicted to produce signals of their presence and reproductive capacity (e.g. ovarian activity) (Keller & Nonacs, 1993), thus allowing workers, in response, to forego reproduction and reap inclusive fitness benefits by rearing offspring produced by the queen. In several ants (Peeters et al., 1999; Liebig et al., 2000), bees (Ayasse et al., 1995) and wasps (Sledge et al., 2001), variation in the proportions of one or more cuticular hydrocarbons may convey information regarding an individual’s reproductive capabilities. In social wasps, nestmate recognition has been well studied in the genus Polistes (Vespidae, Polistinae), where many facets of this phenomenon are understood, and the involvement of cuticular lipids in nestmate recognition is demonstrated (Singer & Espelie, 1992; Dani et al., 1996; Lorenzi et al., 1997). Nestmate recognition based on cuticular hydrocarbons has also been demonstrated in some species of the more advanced social subfamily of the Vespidae, the Vespinae (e.g. Vespa crabro, Ruther et al., 1998). However, little is known about the most primitively social subfamily of the Vespidae, the Stenogastrinae. The Stenogastrinae, or hover wasps, represent a taxon of approximately one hundred species in seven genera inhabiting the forests of South-east Asia from South India to New Guinea
#
2004 The Royal Entomological Society
Epicuticular lipids and fertility in primitively social wasps and from Vietnam to Indonesia (Carpenter & Starr, 2000). The largest genera are represented by Parischnogaster, Liostenogaster and Eustenogaster, which are all distributed in the Oriental Region. These wasps form very small societies (usually less than 10 females) in nests with notably varied architecture (Turillazzi, 1991). Colonies often have more than one egg-layer and, in a few species, dominance hierarchies are formed amongst females (Hansell et al., 1982; Turillazzi & Pardi, 1982; Hansell, 1983). In other species, eusocial traits are evident only in particular stages of the colonial cycle, usually when recently emerged daughters remain and help their mother in nest defence, rearing of larvae and foraging before they become reproductively mature and leave (Turillazzi & Hansell, 1991). Thus, they exhibit features more characteristic of vertebrate societies (with helpers) than eusocial insects (Turillazzi, 1989; Brockmann, 1997), although important differences can be observed in the relatively long developmental time of the wasps with respect to the average adult life (Field et al., 1998). These wasps have been proposed as important models for the study of the factors favouring the origin of social behaviour in insects (Yoshikawa et al., 1969; WestEberhard, 1978; Turillazzi & Pardi, 1982), and several of their characteristics have been highlighted as important in promoting or limiting their social evolution (e.g. nest material, Hansell, 1987; larval rearing, Turillazzi, 1989; ensured fitness returns and long developmental times, Field et al., 2000). Stenogastrine wasps also comprise important models for studying the role of mechanisms, such as kin recognition, that constitute a requisite in social evolution, as well as social communication in general. Very few studies have been conducted in this regard. Using behavioural observations and experiments, Cervo et al. (1996) first demonstrated that nestmate recognition occurred in colonies of three species, whereas subsequent studies have shown that nestmate discrimination is probably linked to variation in cuticular hydrocarbon proportions (Zanetti et al., 2001; Cervo et al., 2002; Destri et al., 2002). The role of exocrine glands in these wasps has also been highlighted. The Dufour’s gland secretion, used for brood rearing and nest defence (Turillazzi, 1991), contains many compounds found on the cuticle (Keegans et al., 1992; Sledge et al., 2000) and could be used as a template for newly emerged adults when recognizing nestmates (Cervo et al., 2002). The present study investigates whether cuticular compounds of several stenogastrine species act as cues not only for nestmate recognition, but also for signalling of physiological status of females belonging to the same colony. A number of colonies are analysed from two species of stenogastrine wasps collected in Peninsular Malaysia: Eustenogaster fraterna (Bingham) and Liostenogaster vechti (Turillazzi), and data collected previously on other two species, Parischnogaster striatula (du Buysson) and Liostenogaster flavolineata (Cameron), are re-examined, including data on the Dufour’s gland secretion. Chemical fertility signals found in primitive social wasps are discussed and a role of such signals in evolution of these societies is presented. #
465
Materials and methods Species studied Eustenogaster fraterna is one of the largest hover wasps. It forms colonies composed of a maximum of six to seven females (Turillazzi & Gerace, 1992; Francescato et al., 2002; Landi et al., 2003) on a nest protected by a flask-like envelope. Liostenogaster vechti forms colonies with up to seven females on a bracket-like or ringed nest (Turillazzi, 1990). Colonies of this species are usually found in large clusters (up to more than 600 nests) in sheltered places or on buildings. For this species, Cervo et al. (1996) reported the presence of an efficient nestmate recognition mechanism. Parischnogaster striatula forms colonies composed of a maximum of five females and builds spiral shaped nests. Previous studies of this species demonstrated the presence of nestmate recognition (Zanetti et al., 2001), despite earlier data indicating frequent exchanges of individuals among colonies (Yoshikawa et al., 1969; but see also Hansell, 1982). A large percentage of females within colonies often possess developed ovaries (Turillazzi, 1987). Liostenogaster flavolineata is one of the best-known species in the group. It builds mud comb nests up to more than 100 cells, sometimes in huge aggregates (Hansell et al., 1982). Its biology is characterized by a very long larval development (on average, more than 100 days). Nestmate recognition has been studied by Cervo et al. (1996, 2002).
Material examined Eustenogaster fraterna adult females (n ¼ 49) belonging to 18 colonies were collected at dusk in 1999 at Bukit Fraser (1000–1500 m above sea level) in Pahang state, Malaysia. Cuticular lipids were extracted from live individuals using the method described by Turillazzi et al. (1998). The system consists of using commercially available cotton wool cleaned in hexane for 10 min (and dried), and then rubbed gently on the cuticle of either the abdomen or thorax of individual wasps. To facilitate rubbing on the cuticle, cotton wool pieces (0.5 0.5 cm) were wrapped around toothpicks (cleaned in hexane for 10 min and dried). The cotton wool used for each wasp was then placed individually in a 2-mL vial and stored at room temperature. The wasps were then killed by freezing and stored in alcohol for future dissection. The wasps, used also for biomolecular studies, were dissected in the laboratory at Rice University (Houston, Texas) and methods and data on their ovarian development can be found in Landi et al. (2003); an index of ovarian development was calculated taking into account the total number of oocytes and the number of mature or nearly mature eggs present in the ovaries (which enabled categorization of females into potential egg layers and non-egg layers). Liostenogaster vechti adult females (n ¼ 45), belonging to 11 colonies were collected at dusk in November 2000 at Bukit Fraser from a single cluster of nests found on the
2004 The Royal Entomological Society, Physiological Entomology, 29, 464–471
466 S. Turillazzi et al. wall of a brick building located in the forest. Cuticular lipids were extracted from live individuals as described above (Turillazzi et al., 1998). Animals were then killed by freezing and stored in alcohol for future dissection. In Italy, all females were dissected and an index of ovarian development calculated as the average length of the six largest oocytes in the ovaries. The females were also divided into potential egg-layers and non-egg layers according to the development of the ovaries. Spermathecae were also examined to establish whether females were mated.
Liostenogaster flavolineata adult females (n ¼ 39) were collected in 1995 and 1996 at Genting Tea Estate (Genting Sempah, Pahang State). Females were frozen and dissected soon thereafter and ovary development was scored in three categories (threadlike or small, partially developed and fully developed). Methods and equipment used for the analysis of these samples have been reported previously by Cervo et al. (2002).
Statistical analysis Chemical analysis Cuticular compounds were extracted from the cotton wool using solvents as follows: 300 mL (E. fraterna) or 500 mL (L. vechti) of pentane were added to the vials and shaken for 20 min in an ultrasonic bath. The cotton wool was then removed with forceps and the pentane evaporated to dryness under a nitrogen stream. The extracts were then resuspended in 10 mL heptane. Similar procedures were performed on blank pieces of cotton wool for control purposes and to identify possible contaminants. Sample analyses were performed using a Hewlett Packard 5890 A gas chromatograph coupled to an HP 5971 A (Palo Alto, California) mass selective detector (using 70eV electron impact ionization) or a Varian Saturn 2000 gas chromatography–mass spectroscopy (GC-MS) ion trap (Walnut Creek, California) in chemical ionization mode with acetonitrile as reagent gas (Moneti et al., 1997). For the Saturn instrument, ion trap temperature was set at 140 C with an ionization time of 2 ms and reaction time at 40 ms. A fused silica capillary column coated with cross-linked methyl-silicone (Rtx-5MS, 30 m 0.25 mm 0.5 mm) was used in both machines. The injector port and transfer line were set at 280 C. The carrier gas was helium. The temperature programme was: from 70 C to 150 C at a rate of 30 C min1; after 5 min at 150 C, the temperature was raised to 320 C at 5 C min1. The volume injected was 2 mL (E. fraterna) or 3 mL (L. vechti). Position of the double bond in the alkenes was determined only for E. fraterna using the method described previously by Moneti et al. (1997). Parischnogaster striatula adult females (n ¼ 20) belonging to seven colonies were collected in 1996 at Genting Tea Estate (Genting Sempah, Pahang State). From these females, both the cuticular lipids and the Dufour’s gland secretion was analysed by GC-MS. Treatment and analyses of the samples are described in the study by Zanetti et al. (2000, 2001). Position of the double bond in the alkenes was determined using the method described by Moneti et al. (1997). These females were classified into three groups according to their ovarian development: class 1 corresponded to females with completely undeveloped or only slightly developed ovaries; class 2 corresponded to females with ovaries containing partially developed oocytes or less than three completely developed oocytes; class 3 corresponded to females whose ovaries contained more than three completely developed oocytes. #
All statistical analyses were performed using SPSS1 11.0 for Windows (Chicago, Illinois). For analysis of lipid profiles, the relative percentage areas of each peak (representing one or, sometimes, more cuticular compounds) were first calculated in each female. Stepwise Discriminant Analysis was used to determine whether the predefined groups could be discriminated on the basis of the cuticular lipid composition of the individuals constituting the groups. Wilks’ lambda significance and the percentage of correct assignments were used to evaluate the validity of the discriminant function. No transformation of the data was performed to allow for a more immediate understanding of the relevance of the different compounds in the statistical analysis. In fact, even if ‘the linear discriminant function requires that the predictor variables have a multivariate normal distribution, the function has been shown to perform fairly well in a variety of other situations’ (Norusis, 1992, p. 2). Non-parametric Mann–Whitney U-tests were used to compare percentages of various compounds of individuals with different ovarian development and Spearman correlations to test for a correlation between ovarian development and compound concentrations.
Results Identified compounds of the cuticular layer Table 1 details the cuticular compounds found in E. fraterna and L. vechti. In both species, the cuticular mixture is characterized by a relatively low number of compounds compared with Polistes wasps and Parischnogaster mellyi (Beani et al., 2002) analysed after a similar extraction procedure and in similar gas chromatographic conditions. In both species, there was a high number of unsaturated hydrocarbons compared with other classes of hydrocarbons. In E. fraterna, no methyl-branched alkanes were found, as already reported for P. striatula (Zanetti et al., 2001), whereas several monomethyl alkanes were found in L. vechti. When comparing hydrocarbon chain length, E. fraterna possessed hydrocarbons only up to 29 carbon atoms, as in P. striatula (Zanetti et al., 2000) and L. flavolineata(Cervo et al., 2002), whereas L. vechti possessed generally longer chain compounds (up to 33 carbon atoms), similar to those Beani et al. (2002) found in P. mellyi.
2004 The Royal Entomological Society, Physiological Entomology, 29, 464–471
Epicuticular lipids and fertility in primitively social wasps Table 1. Compounds found on the cuticle of Eustenogaster fraterna (n ¼ 49 females from 18 colonies collected at Bukit Fraser) and Liostenogaster vechti (n ¼ 45 females from 11 colonies collected at Bukit Fraser). Eustenogaster fraterna
Liostenogaster vechti
n-C18 n-C19 n-C20 9-C21:1 7-C21:1 n-C21 9-C22:1 n-C22 9-C23:1 C20-OH n-C23 9-C24:1 n-C24 9-C25:1 n-C25 n-C27 n-C29
C23:1 n-C23 n-C24 C25:1 n-C25 11-,13-meC25 n-C26 13-,12-meC26 C27:1 n-C27 13-,11-meC27 C28:1 n-C28 14-,13-,12-meC28 C29:1 n-C29 15-,13-meC29 C30:1 n-C30 15-, 14-,13 -meC30 C31:1 isomer A C31:1 isomer B n-C31 15-,13-meC31 n-C32 n-C33
Cuticular lipids and reproductive status In E. fraterna, ovarian development was negatively correlated with 7-C25:1 (Table 2) and potential egg-layers possessed significantly less of this compound than females with poor or undeveloped ovaries (Table 3). By contrast, n-C23 was found to be more concentrated on egg-layers than on non-egg-layers (Table 3). Stepwise discriminant analysis correctly classified 31 (88.6%) out of 35 individuals into potential egg-layers or non-egg-layers using the variables: 7-C25:1, n-C23, 9-C22:1, n-C26 and n-C24 (Table 4). A second analysis correctly classified 29 (81.8%) out of 35 individuals
467
into mated and unmated females using the variables 7-C25:1 and 9-C23:1 (Wilks’ lambda ¼ 0.720, P < 0.01). In L. vechti, ovarian development was positively correlated with one isomer of C31:1 (Table 2) and potential egg-layers had significantly more of this hydrocarbon on the cuticle than females with undeveloped ovaries (Table 3). Stepwise discriminant analysis correctly classified 29 (67.4%) out of 43 individuals into potential egg layers and non-egg layers and the same isomer of C31:1 was the only compound entered in the function (Table 4). Entering all the variables together in the discriminant analysis resulted in the percentage of individuals correctly assigned being raised to 93%. In P. striatula, a negative correlation was found between ovarian development and concentrations of several alkenes (7-C21:1, 7-C23:1, 9-C24:1, 7-C25:1) and a positive correlation for several alkanes (n-C26, n-C27, n-C28 and n-C29) (Table 2). Stepwise discriminant analysis discriminated between females with undeveloped ovaries (class 1) and females with mature or almost mature oocytes (class 2 and 3) in 95% of the cases (Table 4). The compounds selected in the discriminant analysis were: n-C27, 9-C25:1, 6,9-C23:2, 7-C24:1. n-C27 and had significantly higher concentrations in the egg-laying females, similar to other three alkanes (Table 3), whereas the concentration of 9-C25:1 was higher in the non-egg-laying females, as were those of five other alkenes (Table 3). A significant, negative correlation was also found between ovary development and three alkene concentrations for the composition of the Dufour’s gland secretion (7-C23:1, Spearman rho ¼ 0.497, P ¼ 0.026; 9-C23:1, Spearman rho ¼ 0.763, P < 0.0001; 9-C27:1 Spearman rho ¼ 0.451, P ¼ 0.046), but a positive correlation was found for another alkene (7-C24:1, Spearman rho ¼ 0.493, P ¼ 0.027). Also for the Dufour’s gland, the obtained discriminant analysis functions correctly separated females with undeveloped ovaries and females with mature or almost mature oocytes in 95% of cases (Wilks’ lambda ¼ 0.295, d.f. ¼ 3, P < 0.0001). The compounds selected in the discriminant analysis were: 9-C23:1, 9-C29:1; 1-docosanol. 9-C23:1 was present in a significantly higher concentration in the non-egg-laying females, as were the other four components (9-C23:1, U ¼ 6, P < 0.001; 6,9-C23:2, U ¼ 23, P ¼ 0.046; 7-C23, U ¼ 18, P ¼ 0.016; 9-C25, U ¼ 20, P ¼ 0.025; 9-C27, U ¼ 20, P ¼ 0.025), whereas 1-octadecanol was more prevalent in the egg-laying females (U ¼ 23, P ¼ 0.046).
Table 2. The percentages of compounds that are correlated (negatively and positively) with ovarian development in the four species examined.
Correlation with ovaries
Eustenogaster fraterna (n ¼ 35)
Negative
7-C25:1**
Positive
Liostenogaster vechti (n ¼ 43)
C31:1*
Spearman rho non-parametric correlation test: *P < 0.05, **P < 0.01.
#
2004 The Royal Entomological Society, Physiological Entomology, 29, 464–471
Liostenogaster flavolineata (n ¼ 39)
Parischnogaster striatula (n ¼ 20)
7-C21:1*
7-C21:1* 7-C23:1* 9-C24:1* 7-C25:1* n-C26* n-C27* n-C28** n-C29**
C20-OH* n-C23* n-C24** n-C25*
468 S. Turillazzi et al. Table 3. Epicuticular compounds with concentrations significantly higher in egg-laying than non-egg-laying females in colonies belonging to the four species examined. Eustenogaster fraterna
Liostenogaster vechti
Liostenogaster flavolineata
Parischnogaster striatula
Egg laying
n ¼ 23 n-C23*
n ¼ 19 C31:1*
Non-egg-laying
n ¼ 12 7-C25:1**
n ¼ 24
n ¼ 21 C20-OH* n-C23* n-C24** n-C25* n ¼ 18 7-C21:1*
n ¼ 11 n-C26* n-C27* n-C28* n-C29* n¼9 7-C21:1* 7-C22:1* 7-C23:1* 9-C25:1* 7-C25:1* 9-C27:1*
Mann–Whitney U-test: *P < 0.05; **P < 0.01.
Table 4. Percentages of individuals correctly classified in the two classes (undeveloped and developed) of ovarian development by a stepwise discriminant analysis (SDA) on the basis of their epicuticular compound composition, and lists of compounds selected in the analysis for each species.
% of corrected classified individuals Wilk’s lambda P Compounds used by the SDA
Eustenogaster fraterna (n ¼ 35)
Liostenogaster vechti (n ¼ 43)
Liostenogaster flavolineata (n ¼ 39)
Parischnogaster striatula (n ¼ 20)
88.6% *** 7-C25:1 n-C23 9-C22:1 n-C26 n-C24
67.4% * C31:1
66.7% ** 7-C21:1
95.0% *** n-C27 9-C25:1 6,14-C23:2 7-C24:1
Wilk’s Lambda P: *P < 0.05; **P < 0.01; ***P < 0.001.
In L. flavolineata, four compounds (C20-OH; n-C23; n-C24; n-C25) were positively correlated with ovarian development, whereas one (7-C21:1) was negatively correlated (Table 2). The four compounds positively correlated were significantly more concentrated in females with fully and partially developed ovaries than in females with threadlike or small ovaries (Table 3). By contrast, 7-C21:1, negatively correlated with ovarian development, and another alkene (9-C22:1) was more concentrated in females with small ovaries than in females with developed ovaries (Table 3). 7-C21:1 was the only compound entered in the stepwise discriminant function (Table 4) in the discriminant analysis between females with small and developed ovaries. The calculated function correctly assigned only 66.7% of the females. When all the variables were entered together, 97.4% of the females were correctly assigned (Wilks’ lamba ¼ 0.201; P ¼ 0.034). A similar percentage (94.9%) was obtained when the three categories of ovarian development were considered in a discriminant analysis entering all the variables together (Fig. 1). The Wilks’ lambda was statistically significant considering both the functions together (Wilks’ lambda ¼ 0.057; P ¼ 0.041). When only females with small and fully developed ovaries were considered, 78.1% of females were correctly assigned in the stepwise discriminant analysis (Wilks’ lambda ¼ 0.649, P ¼ 0.006). #
Discussion Similar to the other stenogastrines studied, the cuticular lipids of E. fraterna and L. vechti are relatively simple when compared with those of Polistes wasps. Alkenes are among the principal components in all the species, whereas this class of compounds is not quantitatively important in Polistes (Espelie et al., 1990; Singer et al., 1998; Sledge et al., 2001). By contrast, methyl alkanes, which are the most numerous compounds in Polistes wasps, have been reported only for L. vechti as monomethyl alkanes and, for P. mellyi, both as mono- and di-methyl alkanes (Beani et al., 2002). Aliphatic primary alcohols, only seldom found in Polistes wasps, have been found in E. fraterna, L. flavolineata and P. striatula. In the latter species, they are more concentrated in the Dufour’s gland than on the cuticle (Zanetti et al., 2000). When compared with Polistes species, differences between congeneric species of Liostenogaster and of Parischnogaster are much more evident. Although differences among Polistes species are quantitative rather than qualitative, qualitative differences can be observed between P. striatula (Zanetti et al., 2000) and P. mellyi (Beani et al., 2002) (alcohols present in the first species and absent in the second, many methyl alkanes present in the second species and absent in the first), and between L. flavolineata (Cervo
2004 The Royal Entomological Society, Physiological Entomology, 29, 464–471
Epicuticular lipids and fertility in primitively social wasps 4
Discriminant function 2
3
2
1
0
–1
–2
–3 –6
–4
–2
0
2
4
6
Discriminant function 1 Group centroids
Partially developed
Fully developed
Small ovaries
Fig. 1. Summary plot of stepwise discriminant analysis on cuticular compounds of females (n ¼ 39) of Liostenogaster flavolineata showing groups of individuals with different ovarian development (Wilk’s Lambda and P associated with Wilk’s lambda considering the two functions: Wilk’s Lambda ¼ 0.06, P ¼ 0.041; considering the second function only, Wilk’s lambda ¼ 0.30, P < 0.05; percent of correctly classified individuals 94.9%).
et al., 2002) and L. vechti (current study) (alcohols present in the first species and absent in the second, methyl alkanes present in the second species and absent in the first). In all four species examined, the cuticular lipid profile of fertile individuals differed in the levels of one or more compounds from that of nonfertile individuals. Moreover, in all four species, the concentration of one or more compounds was correlated with the degree of ovarian development. Parischnogaster striatula was the species with the highest number of compounds differing in the concentration and being correlated with ovarian development. For this species, separation by discriminant analysis of fertile and nonfertile individuals was almost complete (95%); at the other extreme, the discrimination falls to approximately 65% when a stepwise discriminant analysis is applied to Liostenogaster, but rises to more than 90% when all the variables are used in the analysis. Curiously, the worse discrimination was between colonies of the two species belonging to the genus Liostenogaster and between the species that build unenveloped nests, as if the recognition of an individual based on chemical cues is more important in the species with nests where visual recognition of an individual is probably impossible. #
469
The discriminating compounds differ between all the species; at least one or more alkenes are correlated with ovarian development (and also various alkanes in P. striatula), and fertile and nonfertile individuals differed for the abundance of the same compounds on their cuticle. Significant differences in percentages for linear alkanes and alkenes have been found between sterile and fertile individuals in the vespine wasp Vespa crabro (Butts et al., 1991) where queens are characterized by higher percentages of n-C24 and n-C27 and by lower percentages of C28:1 and 3-meC23 compared with workers. In the ant Harpegnathos saltator, mated egglaying queens and workers (¼ gamergates) are characterized by the presence of 13,23-dimeC37 and higher percentages of long-chained hydrocarbons (Liebig et al., 2000). In the queenless ant Dinoponera quadriceps, gamergates (¼ mated workers) and dominant females have higher percentages of 9-C31:1 than subordinate females and the quantity of this compound is linked with ovarian activity (Monnin et al., 1998; Peeters et al., 1999). In the social wasp Polistes dominulus, Bonavita-Cougourdan et al. (1991) demonstrated that foundresses can be distinguished from their offspring for the relative proportions of some alkanes and methyl alkanes, whereas Sledge et al. (2001) reported statistical differences for various cuticular compounds between alphas and subordinate foundresses in multiple-foundress foundations in a different population of the same species. This may be a general phenomenon also in the stenogastrine wasps, and may represent an ancestral situation in the relationship between ovarian activity and proportions of cuticular hydrocarbons. In P. striatula, significant correlations are also seen between ovarian development and three alkenes (9-C23:1, 7-C23:1, 9-C27:1) present in the secretion of the Dufour’s gland. This secretion (placed on-eggs when laid and, on small larvae, acting as a medium where adults deposit food for the larvae) is quite similar to the epicuticular layer (Zanetti et al., 2000; Beani et al., 2002; Cervo et al., 2002), and may function as a template that newly emerged adults use in chemical nestmate recognition, or as a mark deposited by the egg-laying individuals on their just laid eggs. A similarity is also seen between ant-guard and the Dufour’s gland secretion composition, as reported in Sledge et al. (2000) and cuticular hydrocarbons, as reported here, of Eustenogaster fraterna). Dominance hierarchies are rare in hover wasps and, when present, are relatively mild compared with other social wasps (P. nigricans serrei, Turillazzi & Pardi, 1982; P. mellyi, Hansell, 1982; L. flavolineata, Hansell et al., 1982; P. jacobsoni, Turillazzi, 1988). Despite this, given the simple structure of hover wasp societies and the reduced number of females within colonies, it is unclear why strong aggressive behaviours are absent during social interactions among individuals. In other social species where dominance hierarchies are formed, aggressive behaviours decrease or cease after establishment of the hierarchy (Peeters, 1993), suggesting that a switch in discriminant signalling occurs. Individuals with developed ovaries represent a consistent percentage in colonial populations of various stenogastrine species, and this may be a sign of primitive social organization
2004 The Royal Entomological Society, Physiological Entomology, 29, 464–471
470 S. Turillazzi et al. although this is highly conjectural. These wasps have the same chemical cues, and this could form the basis for a fertility signal to be used in intracolonial communication in the absence of physical displays of the type performed by Polistes wasps. This requires confirmation through behavioural tests, although differences in the present study that occur mainly in classes of cuticular hydrocarbons (alkenes), which are thought to be especially important in chemical communication (Espelie et al., 1990; Breed, 1998; Dani et al., 2001), are suggestive. In conclusion, the cuticular chemical profile characterizes fertile and nonfertile females in the colonies of at least two species of primitive hover wasp. The difference in epicuticular lipid layer composition could be due to factors influencing the metabolism of these compounds or directly to the acquisition of particular substances from the environment, which are linked to differences in behaviour. Newly emerged individuals often remain on nest before foraging, whereas egg-laying females readily stop this activity when substituted by younger ones. Hormonal production by endocrine glands, and by ovaries themselves, can influence the secretion of particular cuticular compounds as demonstrated in solitary and social insects. Although these differences may not necessarily be connected to chemical communication (such as nestmate discrimination and fertility signalling), cuticular compounds represent possible information sources regarding the identity and status of members within colonies. Establishing the extent to which insects use information from cuticular hydrocarbons represents a challenge for future research. Acknowledgements We thank Dr Angelo Fortunato (Dipartimento Biologia Animale e Genetica, Universita` di Firenze, Italy) and Dr Rosli Hashim (Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia) for their assistance in Malaysia, and Mr Henry Barlow, who welcomed us to his property. The study was supported by funds from the Italian MURST (COFIN 1999 and 2001), University of Firenze and the Italian CNR, as well as through the research network ‘Social evolution’ of the Universities of Copenhagen, Firenze, Keele, Sheffield, Uppsala, Wu¨rzburg and the ETH of Zu¨rich financed by the European commission via the Training and Mobility of Researchers (TMR) programme.
References Ayasse, M., Marlovits, T.C., Tengo¨, J. et al. (1995) Are there pheromonal dominance signals in the bumblebee Bombus hypnorum L. (Hymenoptera, Apidae). Apidologie, 26, 163–180. Beani, L., Sledge, M.F., Maiani, S. et al. (2002) Behavioural and chemical analysis of scent-marking in the lek system of a hoverwasp (Vespidae, Stenogastrinae). Insectes Sociaux, 49, 191–299. Bonavita-Cougourdan, A., The´raulaz, G., Bagne`res, A.G. et al. (1991) Cuticular hydrocarbons, social organization and ovarian development in a polistine wasp: Polistes dominulus Christ. Comparative Biochemistry and Physiology B, 100, 667–680.
#
Breed, M.D. (1998) Chemical cues in kin recognition: criteria for identification, experimental approaches, and the honey bee as an example. Pheromone Communication in Social Insects (ed. by R. K. Vandermeer, M. D. Breed. M. L. Winston and K. E. Espelie), pp. 57–78. Westview Press, Boulder, Colorado. Brockmann, H.J. (1997) Cooperative breeding in wasps and vertebrates: the role of ecological constraints. The Evolution of Social Behavior in Insects and Arachnids (ed. by J. C. Choe and B. J. Crespi), pp. 347–371. Cambridge University Press, U.K. Butts, D.P., Espelie, K.E. & Hermann, H.R. (1991) Cuticular hydrocarbons of four species of social wasps in the subfamily Vespinae: Vespa crabro L., Dolichovespula maculata (L.), Vespula squamosa (Drury), and Vespula maculifrons (Buysson). Comparative Biochemistry and Physiology B, 99, 87–91. Carpenter, J.M. & Starr, C.K. (2000) A new genus of hover wasps from South east Asia (Hymenoptera: Vespidae; Stenogastrinae). American Museum Novitates, 3291, 1–12. Cervo, R., Dani, F.R. & Turillazzi, S. (1996) Nestmate recognition in three species of stenogastrine wasps (Hymenoptera Vespidae). Behavioral Ecology and Sociobiology, 39, 311–316. Cervo, R., Dani, F.R., Zanetti, P. et al. (2002) Chemical nestmate recognition in a stenogastrine wasp, Liostenogaster flavolineata (Hymenoptera Vespidae). Ethology, Ecology and Evolution, 14, 351–363. Dani, F.R., Fratini, S. & Turillazzi, S. (1996) Behavioural evidence for the involvement of Dufour’s gland secretion in nestmate recognition in the social wasp Polistes dominulus (Hymenoptera: Vespidae). Behavioral Ecology and Sociobiology, 38, 311–319. Dani, F.R., Jones, G.R., Destri, S. et al. (2001) Deciphering the recognition signature within the cuticular chemical profile of paper wasps. Animal Behaviour, 62, 165–171. Destri, S., Dapporto, L., Fanelli, D. et al. (2002) Preliminary investigations on the role of cuticular hydrocarbons in nestmate recognition of Parischnogaster mellyi (Hymenoptera Vespidae). Insect Social Life, 4, 145–150. Espelie, K.E., Wenzel, J.W. & Chang, G. (1990) Surface lipids of social wasp Polistes metricus Say and its nest and nest pedicel and their relation to nestmate recognition. Journal of Chemical Ecology, 16, 2229–2241. Field, J.P., Foster, W., Shreeves, G. & Sumner, S. (1998) Ecological constraints on independent nesting in facultatively eusocial hover wasps. Proceedings of the Royal Society of London, Series B, 265, 973–977. Field, J.P., Shreeves, G., Sumner, S. & Casiraghi, M. (2000) Insurance-based advantage to helpers in a tropical hover wasp. Nature, 404, 869–871. Francescato, E., Massolo, A., Landi, M. et al. (2002) Colony membership, division of labour and genetic relatedness among females of colonies of Eustenogaster fraterna (Hymenoptera, Vespidae, Stenogastrinae). Journal of Insect Behavior, 15, 153–170. Hansell, M.H. (1982) Colony membership in the wasp Parischnogaster striatula (Stenogastrinae). Animal Behaviour, 30, 1258–1259. Hansell, M.H. (1983) Social behaviour and colony size in the wasp Parischnogaster mellyi (Saussure) Stenogastrinae (Hymenoptera Vespidae). Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, Series C. Biological and Medical Sciences, 86, 167–178. Hansell, M.H. (1987) Nest building as a facilitating and limiting factor in the evolution of eusociality in the Hymenoptera. Oxford Surveys in Evolutionary Biology, vol. 4 (ed. by P. Harvey, H. and Partridge, L. ), pp. 155–181. Oxford University Press, U.K.
2004 The Royal Entomological Society, Physiological Entomology, 29, 464–471
Epicuticular lipids and fertility in primitively social wasps Hansell, M.H. Samuel, C. & Furtado, J.I. (1982) Liostenogaster flavolineata: social life in the small colonies of an Asian tropical wasp. The Biology of Social Insects (ed. by M. D. Breed, C. D. Michener and H. E. Evans), pp. 192–195. Westview, Boulder, Colorado. Howard & R.W. (1993) Cuticular hydrocarbons and communication. Insect Lipids: Chemistry, Biochemistry and Biology (ed. by D. W. Stanley-Samuelson and P. Nelson), pp. 179–226. University of Nebraska, Lincoln, Nebraska. Keegans, S.J. Morgan, E.D. Jackson, B.D. et al. (1992) The chemical nature and origin of the abdominal substance of the social wasp Liostenogaster flavolineata (Vespidae, Stenogastrinae). Biology and Evolution of Social Insects (ed. by J. Billen), pp. 217–221. Leuven University Press, Belgium. Keller, L. & Nonacs, P. (1993) The role of queen pheromones in social insects: queen control or queen signal? Animal Behaviour, 45, 787–794. Landi, M., Strassmann, J., Queller, D. & Turillazzi, S. (2003) Low relatedness and high rates of queen turnover in the stenogastrine wasp, Eustenogaster fraterna. Insectes Sociaux, 50, 262–267. Liebig, J., Peeters, C., Oldham, N.J. et al. (2000) Are variations in cuticular hydrocarbons of queens and workers a reliable signal of fertility in the ant Harpegnathos saltator? Proceedings of the National Academy of Sciences of the U.S.A., 97, 4124–4131. Lorenzi, M.C., Bagneres, A.G., Clement, J.L. & Turillazzi, S. (1997) Polistes biglumis bimaculatus epicuticular hydrocarbons and nestmate recognition (Hymenoptera, Vespidae). Insectes Sociaux, 44, 123–138. Moneti, G., Pieraccini, G., Dani, F.R. et al. (1997) Ion–molecule reactions of ionic species from acetonitrile with unsaturated hydrocarbons for the identification of the double-bond position using an ion trap. Journal of Mass Spectrometry, 32, 1371–1373. Monnin, T., Malosse, C. & Peeters, C. (1998) Solid-phase microextraction and cuticular hydrocarbon differences related to reproductive activity in queenless ant Dinoponera quadriceps. Journal of Chemical Ecology, 24, 473–490. Norusis, M.J. (1992) SPSS/PCþ Professional Statistic, version 5.0 SPSS Inc., Chicago, Illinois. Peeters, C. (1993) Monogyny and polygyny in ponerine ants with or without queens. Queen Number and Sociality in Insects (ed. by L. Keller), pp. 234–261. Oxford University Press, U.K. Peeters, C., Monnin, T. & Malosse, G. (1999) Cuticular hydrocarbons correlated with reproductive status in a queenless ant. Proceedings of the Royal Society of London, Series B, 266, 1323–1327. Ruther, J., Sieben, S. & Schricker, B. (1998) Role of cuticular lipids in nestmate recognition of the European hornet Vespa crabro L. (Hymenoptera, Vespidae). Insectes Sociaux, 45, 169–179. Singer, T.L. & Espelie, K.E. (1992) Social wasps use nest paper hydrocarbons for nestmate recognition. Animal Behaviour, 44, 63–68. Singer, T.L., Espelie, K.E. & Gamboa, G.J. (1998) Nest and nestmate discrimination in independent-founding paper wasps. Pheromone Communication in Social Insects: Ants, Bees, Wasps, and Termites (ed. by R. K. Vander Meer, M. D. Breed, M. Winston and K. Espelie), pp. 104–125. Westview Press, Boulder, Colorado. Sledge, M., Boscaro, F. & Turillazzi, S. (2001) Cuticular hydrocarbons and reproductive status in the social wasp Polistes dominulus. Behavioral Ecology Sociobiology, 49, 401–409. Sledge, M.F., Fortunato, A., Turillazzi, S. et al. (2000) Use of Dufour’s gland secretion in nest defence and brood nutrition by
#
471
hover wasps (Hymenoptera, Stenogastinae). Journal of Insect Physiology, 46, 753–761. Smith, B.H. & Breed, M.D. (1995) The chemical basis for nestmate recognition and mate discrimination in social insects. Chemical Ecology of Insects II (ed. by R. T. Carde´ and W. J. Bell), pp. 287–317. Chapman & Hall, New York, New York. Turillazzi, S. (1987) Social biology of the Stenogastrinae: temporary dynamic reproductive strategy in the wet Tropics. Chemistry and Biology of Social Insects (ed. by J. Eder and H. Rembold), pp. 381–382. Verlag Peperny, Germany. Turillazzi, S. (1988) Social biology of Parischnogaster jacobsoni (Du Buysson) (Hymenoptera Stenogastrinae). Insectes Sociaux, 35, 133–143. Turillazzi, S. (1989) The origin and evolution of social life in the Stenogastrinae (Hymenoptera Vespidae). Journal of Insect Behavior, 2, 649–661. Turillazzi, S. (1990) Social biology of Liostenogaster vechti Turillazzi 1988 (Hymenoptera Stenogastrinae). Tropical Zoology, 3, 68–87. Turillazzi, S. (1991) The Stenogastrinae. The Social Biology of Wasps (ed. by K. Ross and R. Matthews), pp. 74–98. Cornell University Press, Ithaca, New York. Turillazzi, S. & Gerace, L. (1992) Size and characteristics of colonies of Eustenogaster fraterna (Hymenoptera, Vespidae, Stenogastrinae). Ethology, Ecology and Evolution, Special Issue, N2, 65–67. Turillazzi, S. & Hansell, M.H. (1991) Biology and social behaviour of three species of Anischnogaster (Vespidae, Stenogastrinae) in Papua New Guinea. Insectes Sociaux, 38, 423–437. Turillazzi, S. & Pardi, L. (1982) Social behavior of Parischnogaster nigricans serrei (Du Buysson) (Hymenoptera Stenogastrinae) in Java. Annals of the Entomological Society of America, 75, 657– 664. Turillazzi, S., Sledge, M.F. & Moneti, G. (1998) Use of a simple method for sampling cuticular hydrocarbons from live social wasps. Ethology, Ecology and Evolution, 10, 293–297. Vandermeer, R.K., Breed, M.D., Winston, M.L. & Espelie, K.E. (eds) (1998) Pheromone Communication in Social Insects. Westview Press, Boulder, Colorado. West-Eberhard, M.J. (1978) Polygyny and the evolution of social behavior in wasps. Journal of the Kansas Entomological Society, 51, 832–856. Yoshikawa, K., Ohgushi, R. & Sakagami, S.F. (1969) Preliminary report on entomology of the Osaka City University, 5th Scientific Expedition to Southeast Asia 1966. With descriptions of two new genera of stenogasterine [sic] wasps by J. Van der Vecht. Nature Life Southeast Asia, 6, 153–182. Zanetti, P., Dani, F.R., Moneti, G. et al. (2000) Chemical nature of the Dufour gland secretion in the stenogastrine wasp Parischnogaster striatula du Buysson (Hymenoptera: Vespidae). Insect Social Life, 3, 47–52. Zanetti, P., Dani, F.R., Destri, S. et al. (2001) Nestmate recognition in Parischnogaster striatula (Hymenoptera Stenogastrinae), visual and olfactory recognition cues. Journal of Insect Physiology, 47, 1013–1020.
Accepted 28 July 2004
2004 The Royal Entomological Society, Physiological Entomology, 29, 464–471