Age, Sex, and Dominance-Related Mushroom Body Plasticity in the Paperwasp Mischocyttarus Mastigophorus

Age, Sex, and Dominance-Related Mushroom Body Plasticity in the Paperwasp Mischocyttarus mastigophorus

Yamile Molina, Sean O’Donnell Animal Behavior Program, Department of Psychology, University of Washington, Seattle, Washington 98195

Received 1 November 2007; revised 23 January 2008; accepted 24 January 2008

ABSTRACT: Social Hymenoptera are important age. We show that MB development differs between the models for analyzing functional brain plasticity. sexes. Males, who frequently depart to seek mating These insects provide the opportunity to learn how opportunities, have larger MB calyx collars (which individuals’ social roles are related to ﬂexible investment receive optic input) than females. In contrast, females in different brain regions. We assessed how age, sex, and have augmented predominantly antenna-innervated MB individual behavior inﬂuence brain development in calyx lips, which may be useful for nestmate recognition a primitively eusocial paper wasp, Mischocyttarus and interactions on the nest. Sex differences in MB devel- mastigophorus. Previous research in other species has opment increased with age. After accounting for age and demonstrated experience-dependent changes in central sex effects, social aggression was positively correlated and primary sensory centers in the brain. The mushroom with MB calyx volume for both sexes. We found little evi- body (MB) calyx is a central processing region involved dence for relationships among sex, age, or behavior and in sensory integration, learning and memory and may be the volumes of peripheral sensory processing structures. particularly relevant to social behavior. We extend We discuss the implications of gender- and age-related earlier cross-sectional studies of female brain/behavior effects on brain volume in relation to male and female associations by measuring sex- and age-related differen- life history and reproductive success. ' 2008 Wiley Periodi- ces in MB calyx volume, and by quantifying optic lobe cals, Inc.* Develop Neurobiol 68: 950–959, 2008 and antennal lobe development. Age did predict MB Keywords: mushroom bodies; age; sex differences; development: calyx neuropils increased in volume with dominance behavior; social insects

INTRODUCTION demands placed on each region (Schu¨rmann, 1987; Barton et al., 1995; Hampton and Shettleworth, 1996; Both between species, and among individuals within Ratcliffe et al., 2006; Lindenfors et al., 2007; but species, relative investment in tissue from different see Healy and Rowe, 2007). Tradeoffs in tissue brain regions often corresponds to the cognitive investment among functionally distinct brain regions occur in a wide array of species (Barton et al., 1995; Correspondence to: Y. Molina ([email protected]). Smulders et al., 2000; Gronenberg, 2001; White and Contract grant sponsor: National Science Foundation. McDonald, 2002). In insects and related arthropods, Contract grant sponsor: NSF; contract grant number: IBN mushroom bodies (MB) are receiving increasing 0347315. attention as a brain region that exhibits both evolu- ' 2008 Wiley Periodicals, Inc. *This article is a US Government work and, as such, is in the public domain in the United States of tionary lability and individual functional plasticity America. (Farris, 2005; Fahrbach, 2006). The MB are paired Published online 24 March 2008 in Wiley InterScience (www. neuropils within the forebrain of many arthropods. interscience.wiley.com). DOI 10.1002/dneu.20633 MB neurons contribute to cognitive processes that

950 Mushroom Body Development in Wasps 951 include multisensory integration, associative learning, 1995; Ehmer et al., 2001). However, specific cogni- and spatial memory (Mizunami et al., 1998; Lozano et tive demands may vary with an animal’s role within a al., 2001; Davis, 2005; Farris and Roberts, 2005). The group or colony. In social insects, different reproduc- MB provide excellent opportunities to test associative castes have disparate life histories. In addition to tions between cognitive function and neuroanatomy in rates of egg laying, queens and workers differ dra- part because of their unique anatomical organization. matically in task performance, and in how much time Within the MB, the intrinsic neurons, known as Ken- they spend away from their nests. Castes often exhibit yon cells, innervate paired cup-like structures known anatomically distinct brain organization (Julian and as the calyces. The Kenyon cell bodies form a layer Gronenberg, 2002; Hannaford et al., 2003; Groh surrounding the anatomically distinct calyces. The et al., 2006; Molina and O’Donnell 2007). Within the Kenyon cell dendrites project into the calyces, where worker castes, individual variation in task perform- they arborize and receive input from afferent sensory ance corresponds to MB development as well neurons (Mobbs, 1982; Ehmer and Hoy, 2000; Gro- (Withers et al., 1993; O’Donnell et al., 2004; Kuhn- nenberg, 2001). MB calycal volume reflects the extent Buhlmann and Ru¨diger, 2006). of Kenyon cell dendrite arborization, as well as the Leaving the nest to forage is often associated with number of synapses the Kenyon cells make with increased MB calyx development, but recent studies antennal and optic lobe neurons (Farris et al., 2001; on primitively social paper wasps suggest that social Ehmer and Gronenberg, 2004; Seid et al., 2005; Fahr- dominance can also be an important factor in MB de- bach, 2006). Functionally, MB calyx volume is asso- velopment. Socially dominant Mischocyttarus masti- ciated with cognitive processing challenges from vari- gophorus and Polistes instabilis workers have signifi- ous stimuli and complex behaviors, such as feeding cantly larger MB calyces than subordinate workers strategies, task performance, and social interactions (Molina and O’Donnell, 2007; O’Donnell et al., (Withers et al., 1993; Farris and Roberts, 2005; 2007). In both cases, social dominance was more im- Molina and O’Donnell, 2007; O’Donnell et al., 2007). portant than foraging in predicting MB calyx devel- Within the MB calyx, anatomical subregions are opment. However, worker age was not measured in functionally distinct. In eusocial Hymenoptera that these studies. Worker age may mediate or even regu- have been studied to date, the lip of the MB calyx is late the relationship between behavior and neural de- innervated by the antennal lobe, and the size of this velopment (Fahrbach et al., 1998; Gronenberg et al., MB subregion may reflect the relative importance of 1996; Farris et al., 2001; Ku¨hn-Bu¨hlmann and chemosensory and/or tactile input and processing. Wehner, 2006). Here, we extend the previous cross- The collar of the MB calyx receives mainly visual sectional studies by quantifying age effects on MB ca- input in eusocial Hymenoptera. The basal ring is lyx development in the paper wasp M. mastigophorus. composed of two modality-specific zones, one which We also present the first comparison of male and receives visual input and the other olfactory informa- female MB development in paper wasps. Mischocytta- tion (Gronenberg, 2001). At the individual level, vari- rus mastigophorus males are particularly relevant to ation in the size of MB calyx subregions may reflect the analysis of age and dominance relationships with the modality-specific cognitive challenges an insect MB development. The males remain on their natal encounters (Gronenberg, 2001). Species and sex com- nests substantially longer than many social Hymenop- parative data on social Hymenoptera and other insects tera males, and they are highly unusual because they support this idea: the relative size of MB calyx are dominant over their female nest mates (O’Donnell, lip and collar regions covary as predicted with differ- 1999). Social dominance may function to increase ences in reliance on visual versus antennal input males’ access to nutrients and therefore longevity: the (Gronenberg and Ho¨lldobler, 1999; Gronenberg, males take a substantial portion of the food from 2001; Hannaford et al., 2003; Groh et al., 2006; incoming foragers. Male M. mastigophorus do not Molina and O’Donnell, 2007). bring food to their nests, but they do depart and return Several lines of evidence suggest that investment to their nests several times daily (O’Donnell 1999). in MB tissue increases with sociality in insects. Pre- Male behavior away from the nest is poorly known, liminary comparative studies suggested that enlarged but they are likely to be seeking mating opportunities MB evolved in social species, relative to solitary spe- (Litte, 1981; Beani, 1996; O’Donnell, 1999). cies, in several lineages (Howse, 1974; Farris and The development of more peripheral brain regions Strausfeld, 2003; but see Ehmer and Hoy, 2000; also covaries with age, experience, and sex (Heisenberg Farris and Roberts, 2005). In some species, MB et al., 1995; Barth et al., 1997; Brockmann and Bruck- development is greater for members of social groups, ner, 1999; Sigg et al., 1997; Julian and Gronenberg, relative to solitary individuals (Heisenberg et al., 2002; Hannaford et al., 2003; Ehmer and Gronenberg,

Developmental Neurobiology 952 Molina and O’Donnell

2004). Variation in mushroom body volume may Behavioral Data Collection depend on changes in these primary sensory processing We observed each colony every 3 days, except during occa- centers, such as the optic and antennal lobes (Gronen- sional heavy rainfall. The order of observing colonies was ¨ berg and Holldobler, 1999). We examine this possibil- not altered. For the first 2 days of each 3-day observation ity by measuring both peripheral and central processing cycle, behavioral data were collected on two colonies per structures that are innervated by two sensory organs: day (day 1: colonies A and B; day 2: colonies C and D). Be- the compound eyes and the antennae. havioral data were collected continuously for 3 h in the By measuring brain development in individually morning (1.5 h per colony), between 0700 and 1000 h local marked, known-age M. mastigophorus wasps, we time, and for 2 h in the afternoon (1 h per colony), between addressed the following questions: (1) Do males and 1200 and 1400 h. On the third day, behavioral data were females differ in the volumes of targeted brain collected on one colony (colony G) in the morning for 1.5 h regions, particularly in their investment in modality- between 0700 and 1000 h local time and for 1 h in the after- specific sensory processing brain regions? (2) Does noon, between 1200 and 1400 h. Colonies were observed on a total of 11–14 days each over the course of 35–42 age predict brain development, and are age effects days. sex-specific? (3) Does individual variation in behav- We collected behavioral data using a portable tape cas- ior predict brain development after accounting for sette recorder while sitting 0.5 m from the nest, facing the age effects, and are behavioral effects sex-specific? cell openings. All occurrences of the following acts were We were particularly interested in foraging and domi- noted: arrivals and departures from the nest, taking resour- nance interactions. (4) For females, does reproductive ces from incoming foragers, giving aggression (chasing and physiology (ovary development) further predict brain biting nest mates), and receiving aggression (see Itoˆ, 1985 development? (5) Do environmental and ecological and O’Donnell, 1998, 1999 for descriptions of Mischocytta- covariates influence both peripheral and central proc- rus behavior). The timing of arrivals and departures was essing? To answer this, we compared sex, age, and also noted and used to calculate time spent on nest. Queens behavioral influences on optic (OL) and antennal were identified based on observations of egg laying; one female per colony laid eggs during the study. lobes (AL) versus the corresponding MB calyx regions – the collar and lip, respectively. Collection of Subjects and Ovary Dissections MATERIALS AND METHODS Nests and all adult wasps present were collected after dark on the evening of the last behavioral observation day. Study Site and Subject Colonies Wasps were individually stored in an aldehyde-based fixa- tive (Prefer, Anatech, Ltd.). The ovaries were dissected Data were collected from 27 July to 7 August, 2006 from from each female’s gaster and photographed under a dis- five post-worker emergence Mischocyttarus mastigophorus secting microscope at 103 magnification using a digital colonies. The nests were observed in situ in Monteverde, camera. The maximum length and width of the two largest Costa Rica (108180N; 848490W; O’Donnell 1998, 1999; oocytes were measured from the digital photographs using O’Donnell et al., 2007). Observation nests were located on Adobe Photoshop software. The measurements were con- the eves of buildings. Behavioral data were collected on verted to mm by photographing and measuring a stage mi- marked wasps whose date of emergence, and therefore age, crometer. Because oocytes are roughly elliptical, oocyte was known. All resident adult wasps on the nests were indi- area was calculated as: p 3 [1/2] length 3 [1/2] width. We vidually marked with paint pens 1–2 days before observa- used the mean area of the two largest oocytes as an index of tions began, and these were excluded from further analyses. ovary development (Keeping, 2000; Markiewicz and To identify known-age subjects, the observation colo- O’Donnell, 2001; Keeping, 2002; Foster et al., 2004). nies were surveyed daily. Newly emerged wasps (total n ¼ 57, n ¼ 10–14 per subject nest) were captured, anesthetized with ether, marked with paint pens, and returned to their Histology and Neuroanatomical nests. To increase our sample size of age-known wasps, we Measurements also introduced adults (n ¼ 33; n ¼ 5–7 per subject nest) that emerged from source combs taken from other nearby We selected a subset of the known-age individuals as sub- colonies. Subjects from these combs were marked on the jects for neuroanatomical analyses. Subjects from each col- day of adult emergence and introduced onto observation ony were chosen to vary widely in age (range: 2–45 days nests. As with other social Hymenoptera, newly-emerged old) as well as behavior (total n ¼ 34: 25 females, 9 males; M. mastigophorus females are accepted into foreign colo- mean ¼ 5 wasps/colony). We particularly targeted similarly nies (Bell et al., 1974; Jeanne et al., 1988; O’Donnell and aged individuals with different behavioral roles, such as Jeanne, 1993; S. O’D., personal observation). subordinate foragers versus dominant nest wasps.

Developmental Neurobiology Mushroom Body Development in Wasps 953

Head capsules were removed and embedded in plastic the structure of interest (Mayhew, 1982). Quantification resin for histological sectioning (Embed 812, Electron Mi- was performed on every other section. Volume of each struc- croscopy Sciences). We made 15 lm thick coronal sec- ture was calculated as a product of the sum of points counted tions using a rotary microtome and steel histology knives. over sections, the area per point (MB: 0.0032 mm2; Sections were mounted on gelatin-coated slides and AL: 0.0030 mm2, OL: 0.0058) and the distance between sec- stained with tolouidine blue. The head capsule sections tion planes (30 lm). were photographed through the103 primary objective of a Neuroanatomical measurements were made blind to the compound light microscope using a digital camera. We subject’s behavior. We measured volume of the following collected MB counts on all subjects, but were only able to brain regions for one MB hemisphere per adult wasp: the collect optic lobe counts for 30 wasps (hence abbreviated Kenyon cell body region (henceforth abbreviated, Kcb OL; 21 females, 9 males) and antennal lobe counts for 27 region), and two subregions of the calyx neuropil: the lip wasps (hence abbreviated AL; 19 females, 8 males). To and the collar + basal ring (Mobbs, 1982; Ehmer and Hoy, quantify the area of targeted brain regions in the sections, 2000; O’Donnell et al., 2007). The collar and basal ring we overlaid a digital square grid (grid resolution depended were grouped because boundaries between these subdivi- on the structure being measured) atop the photomicro- sions were often ambiguous, whereas boundaries between graph in Adobe Photoshop, and counted the number of the lip and collar were always distinct. For each individual, intersection points that fell within the OL (medulla and we calculated relative volume ratios (e.g., calyx: Kcb lobula), AL, the MB calyx (lip and collar + basal ring), region ratio), because body size may influence absolute vol- and MB Kenyon cell body region (Ehmer and Hoy, 2000; umetric measurements (Wehner et al., 2007). We also Molina and O’Donnell, 2007; O’Donnell et al., 2004, examined OL and AL volumes as ratios (OL: OL+AL 2007 Fig. 1). The grid size was chosen such that at least regions). 125 points were counted per structure in each subject. Ev- ery other section was quantified moving caudally after quantifying the rostral-most section containing tissue from Statistical Analyses We used General Linear Models/multiple regression methods to assess the relationships of age, sex and behavior with mushroom body volume. Behavioral data that did not meet assumptions of parametric tests (e.g., unequal variances) were square-root transformed (Pedhauzer, 1982). Unless otherwise noted, all significant relationships shown are the results of partial correlation analysis – significant relationships that hold after accounting for the effects of the other predictor variables. As in previous studies (Withers et al., 1993; Molina and O’Donnell, 2007; O’Donnell et al., 2004, 2007), we used volume ratios as response variables – total calyx: Kcb region, lip: Kcb region, collar + basal ring: Kcb region, OL: OL+AL regions, and AL: OL + AL regions. On the basis of MB innervation patterns in other social Hymenoptera, we assumed that MB calyx lip volume would reflect capacity for processing antennal inputs (chemosensory and/or tactile), while MB collar volume would reflect visual processing capacity (Gronenberg, 2001). All statistical models included colony identity to con- trol for colony effects, unless otherwise noted. First, we performed partial correlations among peripheral sensory centers and MB calycal subregions. We then calculated multiple regression analyses describing sex differences in absolute volumes and relative volume ratios of the brain structures. Next, we used multiple regression analyses to examine relationships between behavioral covariates and volume data. If behavioral covariates were significantly Figure 1 Coronal sections of mushroom bodies, optic associated with brain size, we reported t-values, after lobe, and antennal lobe in one hemisphere of a female Mis- accounting for other predictor variables (i.e. colony iden- chocyttarus mastigophorus wasp. Mushroom body Kenyon tity, other behaviors). If not significant, we report overall cell bodies (Kcb), calycal subregions (l, lip and c + br, F-values for behavioral covariates, after accounting for collar plus basal ring), antennal lobe (al) and optic lobe other predictor variables. For behaviors, we used total fre- subregions (me, medulla and lo, lobula) are labeled. quencies of key behavioral acts: foraging (females) or

Developmental Neurobiology 954 Molina and O’Donnell nest departure (males), giving and receiving aggression, and taking resources from foragers. We estimated individuals’ total frequencies of performance of each behavioral act over the course of the study by extrapolating from their observed rates of performance. We summed all observed behavioral acts for each individual and we adjusted these values to correct for unequal observation effort (data collection time per nest). This measure of behavioral performance assumes that the effects of behavioral acts on the wasp’s neural development are cumula- tive and additive. We first analyzed behavioral links to brain size in females and males separately, after accounting for age and colony identity. For females, we also examined the relationship between ovary development and neuroanatomy. Finally, we analyzed all subjects to- gether to describe age, sex, and behavioral effects on brain volume. Figure 2 Bar graphs depicting mean + SE of total calyx, lip, and collar volume ratios (relative to the Kenyon cell body region) for male and female Mischocyttarus mastigophorus wasps. Females and males did not differ in total calycal volumes relative to the Kenyon cell bodies, but we RESULTS found sex-specific calycal organization. Females had larger MB lips than males and males had larger MB collars. Aster- Associations Among Primary and Central isks indicate p < 0.05. Processing Regions Female Brain Development: Volumes of MB calyx lips and collars were Individual Differences positively associated (absolute volume: r ¼ 0.63, df ¼ 28, p < 0.001; relative to Kcb regions: r ¼ For females, giving aggression was the only signifi- 0.61, df ¼ 28, p < 0.001). Absolute OL and AL vol- cant behavioral covariate of total calyx development umes did not correlate with each other (r ¼ 0.22, df (calyx: Kcb region; t15 ¼ 2.16, p ¼ 0.05). This asso- ¼ 21, p ¼ 0.31) nor with respective MB subregions ciation was stronger for the collar + basal ring: Kcb (AL, lip: r ¼0.18, df ¼ 21, p ¼ 0.42; OL, collar: r region ratios (t15 ¼ 2.43, p ¼ 0.03) than the lip: Kcb ¼0.18, df ¼ 24, p ¼ 0.38). Relative AL volume region ratios (t15 ¼ 1.55, p ¼ 0.14). Females with and lip: Kcb region ratios were non- higher rates of aggression had larger calyces relative significantly negatively associated (r ¼0.39, df ¼ to the Kcb region, and enlarged collars in particular. 21, p ¼ 0.07). We did not find significant associations between behavioral covariates and relative OL and AL volumes (overall F4,9 ¼ 0.24, p ¼ 0.91). Ovary development was significantly positively associated with total ca- Sex Differences in Brain Development lyx: Kcb region (r ¼ 0.49, df ¼ 18, p ¼ 0.02) and with the development of both calycal subregions The sexes did not differ in total MB calycal volume (lip:Kcb region: r ¼ 0.56, df ¼ 18, p ¼ 0.01; collar + (absolute volume: t ¼ 1.62, p ¼ 0.12; relative to 28 basal ring: Kcb region: r ¼ 0.45, df ¼ 18, p ¼ 0.05), Kcb region: t ¼ 0.99, p ¼ 0.33), but calycal organi- 28 but not with optic or antennal lobes (r ¼ 0.06, df ¼ zation differed between males and females (Fig. 2). 11, p ¼ 0.86). Thus, for females, both behavioral Females had larger MB calycal lips (lip: Kcb region: dominance and ovary development were positively t ¼2.16, p ¼ 0.04; lip: collar + basal ring: t ¼ 28 28 linked to relative total MB calyx size. 6.09, p < 0.0001), while males had larger collar + basal rings relative to the Kcb region (t28 ¼ 2.20, p ¼ 0.04). Absolute lip and collar + basal ring volumes Male Brain Development: paralleled relative volumetric measurements (lip: t28 Individual Differences ¼1.40, p ¼ 0.17; collar: t28 ¼ 2.60, p ¼ 0.01). Vol- umes of the primary sensory centers did not differ Among males, none of the behavioral covariates was between sexes (absolute: OL: t25 ¼0.04, p ¼ 0.97; significantly associated with the overall calyx: Kcb AL: t22 ¼ 0.08, p ¼ 0.94; relative OL and AL: t22 ¼ region ratio (overall F4,3 ¼ 0.44, p ¼ 0.78), nor with 0.10, p ¼ 0.92). collar + basal ring: Kcb region ratios (F4,3 ¼ 3.88, Developmental Neurobiology Mushroom Body Development in Wasps 955

After controlling for colony, sex and age effects, giving aggression was the only signiﬁcant behavioral predictor of calyx: Kcb region ratios (t22 ¼ 3.24, p ¼ 0.004; Fig. 4). This relationship held for both calycal subregions, relative to the Kcb region (lip: t22 ¼ 2.45, p ¼ 0.02; collar + basal ring: t22 ¼ 3.49, p ¼ 0.005). Thus, wasps which direct more aggression to nest mates have enlarged calyces.

DISCUSSION

We found evidence for both age-related and age-inde- Figure 3 Partial regression plot illustrating the relation- pendent factors that correspond to brain development, ship between age and mean calyx: Kenyon cell body region ratios, after accounting for colony identity, for male and and specifically MB development, in M. mastigopho- female Mischocyttarus mastigophorus wasps. Both sexes rus. The ratio of total MB calyx volume to the adja- experienced an age-dependent calycal expansion, and sex cent Kenyon cell body region increased with both age differences in calyx size increased with age, as shown by and with the frequency of dominating nest mates. We regression lines. Closed circles and solid regression lines also found neural investment differences between represent females, open circles and dashed regression lines sexes, which may reflect disparate life histories. Sex males. differences within the M. mastigophorus MB were specific to regions that are innervated by distinct sen- p ¼ 0.15). However, male lip: Kcb region ratios were sory modality inputs (Gronenberg, 1999, 2001; Groh ¼ negatively associated with rates of departures (t3 et al., 2006). Females had larger MB calyx lips, ¼ ¼ 6.48, p 0.007) and receiving aggression (t3 which receive primarily antennal input. Males had 5,63, p ¼ 0.01), and positively related to rates of larger collar + basal rings, which receive primarily ¼ ¼ giving aggression (t3 5.89, p 0.01). We did not visual input from the optic lobes. This disparate orga- find any significant behavioral links to optic nor nization suggests that the sexes differ in their reliance ¼ ¼ antennal lobes (overall F4,2 2.88, p 0.27). Thus, on these sensory modalities (Gronenberg, 2001). Fur- males with high rates of giving aggression and low thermore, the magnitude of sex differences in MB de- rates of receiving aggression and departures also had velopment increased with age. larger relative MB lips. Olfactory and tactile information, sensed primarily via the antennae, may be especially useful for on-nest

Age, Sex, and Brain Development Pooling the sexes, age was a significant positive predictor of total calyx: Kcb region ratios (t28 ¼ 2.65, p ¼ 0.01; Fig. 3). We also found positive associations between age and both calycal subregions relative to the Kcb region (lip: t28 ¼ 3.85, p ¼ 0.001; collar + basal ring: t28 ¼ 1.88, p ¼ 0.07). Lip: collar ratios were not associated with age (t28 ¼ 0.59, p ¼ 0.56), indicating that there was an age-dependent volume expansion across the entire calyx. Age was not associated with relative optic lobe or antennal lobe volumes (t22 ¼ 0.30, p ¼ 0.77). There was a significant (age X sex) interaction for Figure 4 Partial regression plot portraying the relation- calyx: Kcb region ratios. As the wasps aged, sex dif- ship between rates of giving aggression and calyx: Kenyon ferences increased (ANCOVA heterogeneity of cell bodies ratios, after accounting for colony identity, age, ¼ ¼ slopes test: F1,26 9.72, p 0.004, Fig. 3). This was sex, and other behavioral covariates. Wasps with high rates especially strong for the collar + basal ring subregion of giving aggression also possessed enlarged calyces, rela- (t26 ¼ 12.74, p ¼ 0.001). tive to the Kcb region. Developmental Neurobiology 956 Molina and O’Donnell social interactions (Bergman and Moore, 2003; Cas- increased social interactions beyond mating and lek- sill, 2003; Ozaki et al., 2005). Nestmate wasps king opportunities (O’Donnell, 1999). engage in repeated, reciprocal interactions that may We did not find significant associations of sex, require recognizing other individuals, as well as asso- age, or behavior with the development of the periph- ciating individuals with their relative dominance eral processing regions in the brain. To the extent that rank. These social cognition tasks are likely to have a volume changes are meaningful, this suggests that strong pheromonal basis (Trianni et al., 2004; central MB processing is the major plastic response Howard and Blomquist, 2005; Leonhardt et al., 2007; to cognitive challenges in adults in this species. This Steiger et al., 2007). Previous work on Polistes insta- is in contrast to some insect species, in which sex, bilis also indicates the importance of olfaction: age, and environmental factors influence OL and AL socially dominant females had larger MB calyx lips size (Heisenberg et al., 1995; Barth et al., 1997; Sigg (Molina and O’Donnell, 2007). Similarly, the number et al., 1997; Julian and Gronenberg, 2002). Environ- of olfactory synaptic complexes in the MB calyx of mental and ecological factors may still influence OL honey bee queens increases with age (Groh et al., and AL function as well as their degree of connectiv- 2006). ity to the MB, even without volume changes. For Mating behavior of M. mastigophorus is unknown, instance, age influences the number of glomeruli acti- but we assume that males departing from the nest are vated by specific odors in bees (Wang et al., 2005). seeking mating opportunities (O’Donnell, 1999). Off Further study closely connecting sex, age, and behav- nest, Polistes and Mischocyttarus males patrol for ior to peripheral neural organization is needed. mates (Litte, 1981; Keeping et al., 1986) or form mat- Studies on experience-dependent changes in MB ing leks (Litte, 1979; Beani and Calloni, 1991; Beani development have often focused on the onset and fre- 1996). Males may require greater visual processing quency of spatially demanding tasks such as foraging capabilities than females when searching for mating (Farris, 2005; Fahrbach, 2006). Foraging is heavily opportunities. This may be particularly true for Mis- age-dependent in most advanced eusocial species, chocyttarus males, as they orient visually to conspe- including ants, wasps, and bees (Withers et al., 1993; cifics and even females from other physically similar Gronenberg et al., 1996; O’Donnell et al., 2004). species (Litte, 1979; Keeping et al., 1986). Males of Even so, there are task-related increases in MB vol- some other social insects that use vision in mate ume, after accounting for age effects (Gronenberg searching show greater optic lobe development rela- et al., 1996; Fahrbach et al., 1998; Farris et al., 2001). tive to females (Gronenberg, 1999; Ehmer and Gro- Though there is weak if any age-regulated task per- nenberg, 2004; but see Hannaford et al., 2003). After formance in primitively social species (Jeanne, controlling for age, socially dominant females, which 1991), rates of social interactions do increase with remain on the nest the majority of the time, have age in M. mastigophorus (Molina and O’Donnell, larger collars than subordinate foragers, suggesting unpublished data). Although we found age-related that vision does remain important for females’ on- changes in MB development, behavioral differences nest interactions. Polistes fuscatus paper wasps rec- explained additional variation in calyx volume. Even ognize nest mates using facial and abdominal mark- after accounting for age effects, social aggression to ings (Tibbetts, 2002, 2004). It is unknown whether nest mates was positively linked to MB development Mischocyttarus wasps use such cues when directing in both sexes, and ovary development was positively aggression to nestmates. linked to MB calyx volume in females. Augmented Mischocyttarus mastigophorus males possess sur- neural tissue that is dedicated to complex cognitive prisingly large MB calyces, in contrast to MB found processing may become important when wasps begin in other social hymenopteran males. Though males in participating in nestmate interactions. For instance, some other species have larger peripheral processing associative memory may improve the likelihood of regions, such as enlarged optic and antennal lobes, direct benefits (i.e. access to resources) when interact- females in more advanced eusocial species tend to ing with nestmates that vary in dominance rank. have larger MB calyces (Schu¨rmann, 1987; Brock- Males and females with greater MB volume may mann and Bruckner, 1997; Gronenberg and Ho¨lldo- obtain resources such as food at greater rates, and bler, 1999; Hannaford et al., 2003; Ehmer and Gro- hence increase their own capacity for reproduction, if nenberg, 2004). Mischocyttarus mastigophorus may they can discriminate and give aggression to individ- require more flexible cognitive strategies than males uals that either have food or will likely forage. in other species. These males may require cognitive Enhanced central processing may also be helpful if machinery more similar to females, because of their dominance interactions regulate colony activity, as longer lives, extended tenure on natal nests, and has been found among females in primitively eusocial

Developmental Neurobiology Mushroom Body Development in Wasps 957 species like M. mastigophorus (see Reeve, 1991 for Davis RL. 2005. Olfactory memory formation in Drosoph- review). A common physiological mechanism, such ila: From molecular to systems neuronscience. Ann Rev as ecdysteroid titers, may both activate ovary devel- Neurosci 28:275–302. opment and promote an increase in calycal volume. Ehmer B, Gronenberg W. 2004. Mushroom body volumes This linking of brain development and reproductive and visual interneurons in ants: Comparisons between sexes and castes. J Comp Neurol 469:198–213. physiology is plausible, given that hormonally-regu- Ehmer B, Hoy R. 2000. Mushroom bodies of vespid wasps. lated genes (E74) have recently been found both in J Comp Neurol 416:93–100. honeybee worker MB interneurons and queen abdo- Ehmer B, Reeve HK, Hoy RR. 2001. Comparison of brain mens (Paul et al., 2005). The patterns that we docu- volumes between single and multiple foundresses in the mented support the hypothesis that behavior associ- paper wasp Polistes dominulus. Brain Behav Evol 57: ated with direct reproduction poses important cogni- 161–168. tive challenges to primitively social paper wasps. Fahrbach SE. 2006. Organization of the mushroom bodies of the insect brain. Ann Rev Entomol 51:209–232. Thanks to Anjali Kumar and several anonymous Fahrbach SE, Moore D, Capaldi EA, Farris SM, Robinson reviewers for comments on earlier drafts. Research was con- GE. 1998. Experience-expectant plasticity in the mush- ducted under permits from the Ministry of the Environment room bodies of the honeybee. Learn Mem 5:115–123. and Energy, Republic of Costa Rica, Scientific Passport Farris SM. 2005. Evolution of insect mushroom bodies: Old #0387, and in accordance with the laws of the Republic of clues, new insights. Athropod Struct Dev 34:211–234. Costa Rica. Any opinions, findings, and the conclusions or Farris SM, Roberts NS. 2005. Coevolution of generalist recommendations, are those of the authors, and do not neces- feeding ecologies and gyrencephalic mushroom bodies in sarily reflect the views of the National Science Foundation. insects. Proc Natl Acad Sci USA 102:17394–17399. Thanks to the Organization for Tropical Studies for ongoing Farris SM, Robinson GE, Fahrbach SE. 2001. Experience- research support, including assistance in obtaining permits. and age-related outgrowth of intrinsic neurons in the mushroom bodies of the adult worker honeybee. J Neuro- sci 21:6395–6404. Farris SM, Strausfeld NJ. 2003. A unique mushroom body REFERENCES substructure common to basal cockroaches and to ter- mites. J Comp Neurol 456:305–320. Barth M, Hirsch HVB, Meinertzhagen IA, Heisenberg M. Foster RL, Brunskill A, Verdirame D, O’Donnell S. 2004. 1997. Experience-dependent developmental plasticity in Reproductive physiology, dominance interactions, and the optic lobe of Drosophila melanogaster. J Neurosci division of labor among bumble bee workers. Physiol 17:1493–1504. Entomol 29:327–334. Barton RA, Purvis A, Harvey PH. 1995. Evolutionary radi- Groh C, Ahrens D, Rossler W. 2006. Environment- and ation of visual and olfactory brain systems in primates, age-dependent plasticity of synaptic complexes in the bats and insectivores. Philos Trans R Soc London B Biol mushroom bodies of honey bee queens. Brain Behav Sci 348:381–392. Evol 68:1–14. Beani L. 1996. Leklike courtship in paper wasps: \A pro- Gronenberg W. 1999. Modality-specific segregation of longed, delicate, and troublesome affair." In: Turillazzi, input to ant mushroom bodies. Brain Behav Evol 54:85– S, West-Eberhard, MJ, editors. The Natural History and 95. Evolution of Paper Wasps. Oxford: Oxford University Gronenberg W. 2001. Subdivisions of the hymenopteran Press, pp 113–125. mushroom body calyces by their afferent supply. J Comp Beani L, Calloni C. 1991. Leg tegumental glands and male Neurol 435:474–489. rubbing behavior at leks in Polistes dominulus (Hyme- Gronenberg W, Heeren S, Ho¨lldobler B. 1996. Age-depend- noptera: Vespidae). J Insect Behav 4:449–462. ent and task-related morphological changes in the brain Bell WJ, Breed MD, Richards KW, Michener CD. 1974. and the mushroom bodies of the ant Camponotus Social, stimulatory and motivational factors involved in floridanus. J Exp Biol 199:2011–2019. intraspecific nest defense of a primitively eusocial halic- Gronenberg W, Ho¨lldobler B. 1999. Morphologic represen- tine bee. J Comp Physiol 93:1432–1451. tation of visual and antennal information in the ant brain. Bergman DA, Moore PA. 2003. Prolonged exposure to J Comp Neurol 412:229–240. social odors alters subsequent social interactions in cray- Hampton RR, Shettleworth SJ. 1996. Hippocampal lesions fish. Integrative Comp Biol 43:1070. impair memory for location but not color in passerine Brockmann A, Bruckner D. 1999. Dimorphic antennal sys- birds. Behav Neurosci 110:831–835. tems in gyandromorphic honey bees. Apis mellifera 1 Hannaford S, Foster RL, Gibson J, Sattler A. 2003. Effects (Hymenoptera: Apidae) Int J Insect Morphol Embryol of age and caste on the brain and mushroom bodies of 28:53–60. the bumblebee. Bombus huntii. Integrative Comp Biol Cassill D. 2003. Rules of supply and demand regulate 43:1035. recruitment to food in an ant society. Behav Ecol Socio- Healy SD, Rowe C. 2007. A critique of comparative studies biol 54:441–450. of brain size. Proc Roy Soc Biol Sci 274:453–464.

Developmental Neurobiology 958 Molina and O’Donnell

Heisenberg M, Heusipp M, Wanke C. 1995. Structural duction in eusocial wasps. J Comp Physiol A 187:327– plasticity in the Drosophila brain. J Neurosci 15:1951– 333. 1960. Mayhew TM. 1982. A review of recent advances in stereol- Howard RW, Blomquist GJ. 2005. Ecological, behavioral, ogy for quantifying neural structure. J Neurocytol 21: and biochemical aspects of insect hydrocarbons. Ann 313–328. Rev Entomol 50:371–393. Mizunami M, Weibrecht JM, Strausfeld NJ. 1998. Mush- Howse PE. 1974. Design and function of the insect brain. room bodies of hte cockroach: Their participation in In: Browne LB, editor. Experimental Analysis of Insect place memory. J Comp Neurol 402:520–537. Behavior. Berlin: Springer, pp 180–195. Mobbs PG. 1982. The brain of the honey bee, Apis melli- Itoˆ Y. 1985. Social behavior and social structure of neotrop- fera L. The connections and spatial organization of the ical paper wasps Mischocyttarus angulatus Richards and mushroom bodies. Philos Trans R Soc London B Biol M. basimacula (Cameron). J Ethol 2:17–29. Sci 298:309–354. Jeanne RL. 1991. Polyethism. In: Ross KG, matthews RW, Molina Y, O’Donnell S. 2007. Mushroom body volume is editors. The Social Biology of Wasps. Ithaca, NY: Com- related to social aggression and ovary development in the stock Publishing Associates, pp 389–425. paperwasp Polistes instabilis. Brain Behav Evol 70:137– Jeanne RL, Downing HA, Post DC. 1988. Age polyethism 144. and individual variation in Polybia occienetalis,an O’Donnell S. 1998. Dominance and polyethism in the euso- advanced eusocial wasp. In: Jeanne RL, editor. Inter- cial wasp Mischocyttarus mastigophorus (Hymenoptera, individual Behavioral Variability in Social Insects. Vespidae). Behav Ecol Sociobiol 43:327–331. Boulder: Westview Press, pp 323–357. O’Donnell S. 1999. The function of male dominance in the Julian GE, Gronenberg W. 2002. Reduction of brain vol- eusocial wasp mischocyttarus mastigophorus (Hymenop- ume correlates with behavioral changes in queen ants. tera. Vespidae). Ethology 105:273–282. Brain Behav Evol 60:152–164. O’Donnell S, Donlan NA, Jones TA. 2004. Mushroom Keeping MG. 2000. Morphological-physiological variabil- body structural plasticity is associated with temporal ity and differentiation of reproductive roles among foun- polyethism in eusocial wasp workers. Neurosci Lett 356: dresses of the primitively eusocial wasp. Belongaster 159–162. petiolata (Deeger) (Hymenoptera: Vespidae). Insectes O’Donnell S, Donlan NA, Jones TA. 2007. Organizational Soc 47:147–154. and dominance-associated differences in mushroom body Keeping MG. 2002. Reproductive and worker castes in the organization in the paper wasp, Mischocyttarus mastigo- primitive eusocial wasp Belongaster petiolata (DeGeer) phorus. J Neurobiol 67:39–46. (Hymenoptera: Vespidae): Evidence for pre-imaginal O’Donnell S, Jeanne RL. 1993. Methroprene accelerates determination. J Insect Physiol 48:867–879. age polyethism in workers of a social wasp (Polybia Keeping MG, Lipschitz D, Crewe RM. 1986. Chemical occidentalis). Physiol Entomol 18:189–194. mate recognition and release of male sexual behavior in Ozaki M, Wada-Katsumata A, Fujikawa K, Iwasaki M, Polybine wasp, Belonogaster petiolata (Hymenoptera: Yokohari F, Satoji Y, Nisimura T, et al. 2005. Ant nest- Vespidae) J Chem Ecol 12:773–779. mate and non-nestmate discrimination by a chemosen- Ku¨hn-Bu¨hlmann S, Ru¨diger W. 2006. Age-dependent and sory sensillum. Science 309:311–314. task-related volume changes in the mushroom bodies of Paul RK, Takeuchi H, Matsuo Y, Kubo T. 2005. Gene visually guided desert ants, Cataglyphis bicolor. J Neuro- expression of ecdysteroid-regulated gene E74 of the hon- biol 66:511–521. eybee in ovary and brain. Insect Mol Biol 14:9–15. Leonhardt SD, Brandstaetter AS, Kleineidam CJ. 2007. Pedhauzer EJ. 1982. Multiple Regression in Behavioral Reformation process of the neuronal template for nest- Research. USA: Thomson learning, Inc. mate-recognition cues in the carpeter ant Camponotus Ratcliffe JM, Fenton MB, Shettleworth SJ. 2006. Behav- floridanus. J Comp Physiol A 193:993–1000. ioral flexibility positively correlated with relative brain Lindenfors P, Nunn CL, Barton RA. 2007. Primative brain volume in predatory bats. Brain Behav Evol 67:165–176. architecture and selection in relation to sex. BMC Biol 5: Reeve HK. 1991. Polistes. In: Ross KG, matthews RW, edi- Art no. 20. tors. The Social Biology of Wasps. Ithaca, NY: Com- Litte M. 1979. Mischocyttarus flavitarsis in Arizona: social stock Publishing Associates, pp 99–148. and nesting biology of a polistine wasp. Z Tierpsychol Schu¨rmann FW. 1987. Histology and ultrastructure of the 50:282–284. onychophoran brain. In: Gupta, AP, editor. Arthropod Litte M. 1981. Social biology of the polistine wasp Mischo- Brain: Its Evolution, Structure and Functions. New York: cyttarus labiatus: Survival in a Columbian rain forest. Wiley, pp 159–180. Smith Cont Zool 327:1–27. Seid MA, Harris KM, Traniello JFA. 2005. Age-related Lozano VC, Armengaud C, Gautheir M. 2001. Memory changes in the number and organization of synapses in impairment induced by cholinergic antagonists injected the lip region of the mushroom bodies in the ant Pheidole into the mushroom bodies of the honeybee. J Comp dnetata. J Comp Neurol 488:269–277. Physiol A 187:249–254. Sigg D, Thompson CM, Mercer AR. 1997. Activity-de- Markiewicz DA, O’Donnell S. 2001. Social dominance, pendent changes to the brain and behavior of the honey task performance and nutrition: Implications for repro- bee, Apis mellifera. J Neurosci 17:7148–7156.

Developmental Neurobiology Mushroom Body Development in Wasps 959

Smulders TV, Shiftlett MW, Sperling AJ, DeVoogd TJ. Trianni V, Labella TH, Dorigo M. 2004. Evolution of direct 2000. Seasonal changes in neuron numbers in the hippo- communication for a swarm-bot performing hole avoid- campal formation of a food-hoarding bird: The black- ance. Lect Notes Comp Sci 3172:130–142. capped chickadee. J Neurobiol 44:414–422. Wang SP, Zhang SW, Sato K, Srinivasan MV. 2005. Matu- Steiger S, Peschke K, Francke W, Muller JK. 2007. The smell ration of odor representation in the honeybee antennal of parents: Breeding status inﬂuences cuticular hydrocar- lobe. J Insect Physiol 51:1244–1254. bon pattern in the burying beetle Nicrophorus vespilloides. Wehner R, Fukushi T, Isler K. 2007. On being small: Brain Proc R Soc London B Biol Sci 274:2211–2220. allometry in ants. Brain Behav Evol 69:220–228. Tibbetts EA. 2002. Visual signals of individual identity in White NM, McDonald RJ. 2002. Multiple parallel memory the wasp Polistes fuscatus. Proc R Soc London B Biol systems in the brain of the rat. Neurobiol Learn Mem Sci 269:1423–1428. 77:125–184. Tibbetts EA. 2004. Complex social behaviour can select for Withers GS, Fahrbach SE, Robinson GE. 1993. Selective variability in visual features: A case study in Polistes neuroanatomical plasticity and division of labour in the wasps. Proc R Soc London B Biol Sci 271:1955–1960. honeybee. Nature 364:238–240.

Developmental Neurobiology