bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1 Re-emergence and diversification of a specialised antennal lobe morphology

2 in ithomiine butterflies 3 4 Authors: 5 Billy J Morris1*, Antoine Couto1,2, Asli Aydin3, Stephen H Montgomery2*.

6 7 Affiliations:

1 8 Dept. of Zoology, University of Cambridge, Downing Street, Cambridge, CB2 3EJ 9 2 School of Biological Sciences, University of Bristol, 24 Tyndall Avenue, Bristol, BS8 1TQ 10 3 School of Medicine, Koc University, Rumelifeneri Yolu 34450 Sarıyer / Istanbul, Turkey

11 12 * corresponding authors: 13 BJM: [email protected] 14 SHM: [email protected] 15 16 Abstract 17 How an organism’s sensory system functions is central to how it navigates its environment and 18 meets the behavioural challenges associated with survival and reproduction. Comparing sensory 19 systems across can reveal how facets of behaviour and ecology promote adaptive shifts 20 in the relative importance of certain environmental cues. The olfactory system is prominent 21 model for investigating how ecological factors impact sensory reception and processing. Notably 22 work in led to the discovery of vastly expanded structures, termed a macroglomerular 23 complex (MGC), within the primary olfactory processing centre. These structures typically process 24 pheromonal cues and provide a classic example of how variation in size can influence the 25 functional processing of sensory cues. Though prevalent across moths, the MGC was lost during 26 the early evolution of butterflies, consistent with evidence that courtship initiation in butterflies is 27 primarily reliant upon visual cues, rather than long distance olfactory signals like pheromones. 28 However, a MGC has recently been reported to be present in a species of ithomiine, Godryis 29 zavaleta, suggesting this once lost neural adaptation has re-emerged in this clade. Here, we show 30 that MGC’s, or MGC-like morphologies, are indeed widely distributed across the ithomiine tribe, 31 and vary in both structure and the prevalence of sexual dimorphism. Based on patterns of 32 variation across species with different chemical ecologies, we suggest that this structure is 33 involved in the processing of both plant and pheromonal cues, of interlinked chemical constitution, 34 and has evolved in conjunction with the increased importance and diversification of plant derived 35 chemicals cues in ithomiines.

1 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

36 Key words: 37 Neural adaptation, chemical ecology, , olfaction, pheromones, sexual signalling. 38 39 Introduction 40 An organism’s sensory system is its interface with the rest of the world, the link between its internal 41 and external environments. The manner in which sensory systems vary can reveal how different 42 species are attuned to different cues, the association between cues and behaviour, and how 43 behavioural variation maps to the evolution of sensory systems. Lepidopterans have often been 44 used as models to investigate how ecological variability affects the evolution of olfactory systems 45 (Carlsson et al. 2011; Bisch-Knaden et al. 2012; Carlsson et al. 2013; Namiki et al. 2014; van Dijk 46 et al. 2017), and how the central brain processes olfactory information (Kanzaki et al. 1989; Anton 47 and Hansson, 1994; Tabuchi et al. 2013). This includes classic work characterising pheromones, 48 the olfactory response to these chemical cues, and the manner in which the pheromone 49 processing system evolves (Butenandt et al. 1959; Klun and Maini, 1979; Namiki et al. 2014). 50 Studies on lepidopteran sensory systems have provided crucial insights into how sensory systems 51 function, how separate strands of information are processed and integrated within the brain, and 52 the relationship between sensory systems and ecological variables (Couto et al. 2020). 53 As in all , the primary olfactory processing structure within the lepidopteran brain is 54 the antennal lobe (AL). The AL is composed of a collection of functional and morphological units, 55 termed glomeruli. Each glomerulus is a synapse dense region composed of the axon terminals of 56 antennal sensory neurons that typically express the same olfactory receptor (Vosshall et al. 2000), 57 local interneurons that refine the olfactory message, and projection neurons that convey 58 information to higher brain centres. Odorants elicit activity across a range of olfactory receptors, 59 and associated glomeruli, encoding the odorant identity through the combinatorial activation of 60 glomeruli (Joerges et al. 1997; Galizia et al. 1999; Carlsson et al. 2002; Wang et al. 2003; Hallem 61 and Carlson, 2006; Zube et al. 2007). Despite their ecological diversity, within Lepidoptera the 62 antennal lobe is relatively consistent in its structure, being made up of ~60-80 glomeruli (Rospars, 63 1983; Berg et al. 2002; Kazawa et al. 2009; Heinze and Reppert, 2012; Montgomery and Ott, 64 2015; Montgomery et al. 2016; Zhao et al. 2016). However, in moths a prominent morphologically 65 distinct sub-cluster of glomeruli occur at the base of the antennal nerve (Bretschneider, 1924; 66 Matsumoto and Hildebrand, 1981; Koontz and Schneider, 1987). This glomerular cluster is termed 67 a Macroglomerlar complex (MGC) and is composed of enlarged, ‘macro’ glomeruli (MG), and 68 smaller, associated glomeruli, termed ‘satellite’ glomeruli. These glomeruli often display an 69 extreme degree of sexual dimorphism, being vastly enlarged in males relative to females

2 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

70 (Matsumoto and Hildebrand, 1981; Koontz and Schneider, 1987). Though first identified in moths, 71 MGs/MGCs have subsequently been observed in a diverse range of insects, including in 72 Blattodea, Diptera, Hymenoptera, and Lepidoptera (Chambille et al. 1980; Couto et al. 2016; 73 Kelber et al. 2009; Ibba et al. 2010; Keubler et al. 2010). MGCs are typically involved in processing 74 pheromonal cues detected by the antennal sensilla, where their corresponding olfactory receptors 75 are highly expressed in a greater number of sensory neurons, providing heightened sensitivity 76 (Warner et al. 2007; Miura et al. 2009). MGs responsive to host plant related cues have also been 77 reported (Ibba et al. 2010), suggesting they reflect an efficient way of increasing sensitivity to 78 biologically important odours to each species. MGs are therefore classic examples of how 79 neuropil size reflects functional performance, as variation in their volume is generally associated 80 with variation in sensitivity to their corresponding odour (Gronenberg and Hölldobler, 1999; El 81 Jundi et al. 2009b; Warner et al. 2007; Miura et al. 2009). Furthermore, MGC structure and 82 composition is variable across closely related species, suggesting they may co-evolve adaptively 83 with species specific ecological and behavioural traits (Kondoh et al. 2003; Namiki et al. 2014; 84 Bastin et al. 2018). 85 While MGCs are ubiquitous in moths (Rospars and Hildebrand, 2000; Huetteroth and 86 Schachtner, 2005; El Jundi et al., 2009b; Løfaldli et al., 2010; Yan et al. 2019), including diurnal 87 species (Stöckl et al. 2016), they are absent in several phylogenetically disparate butterflies 88 (Rospars, 1983; Carlson et al. 2011; Heinze and Reppert, 2012; Montgomery et al. 2016) 89 suggesting they were lost at the origin of Papilionoidea. This has been interpreted as reflecting 90 an increased reliance on visual cues, and the decreased importance of long-distance chemical 91 signalling in butterfly mating behaviours (Rospars, 1983; Rutowski, 1991; Andersson et al. 2007). 92 However, this view is being revisited. Evidence is accumulating that pheromone cues function in 93 interspecific discrimination, sexual attraction and discrimination, and expediate female 94 acceptance in courtship in a range of butterflies (Andersson et al. 2007; Constanzo and Monteiro, 95 2007; Schulz et al. 2007; Mérot et al. 2015; Chengzhe et al. 2017; Darragh et al. 2017). 96 One diverse tribe of diurnal butterflies with particular reliance on olfactory cues are the 97 Ithomiini. Ithomiines utilise derivatives of pyrrolizidine alkaloids (PAs) for both chemical defence 98 and intraspecific communication (Pliske, 1975, Pliske et al. 1976; Brown, 1984). PAs are 99 sequestered from particular species of plants at the adult stage, with males being significantly 100 more attracted and motivated by these resources (Pliske, 1975; Brown 1984). Males provision 101 eggs with PAs through the spermatophore, providing chemical protection to the egg and larvae 102 (Brown 1984). PA derived pheromones are secreted from ‘hair pencils’, specialised, elongated 103 cells found on the dorsal surface of the androconial gland on the forewing (Schulz et al.1988).

3 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

104 The expression of PA-derived pheromones is believed to represent an honest signal of male 105 quality and facilitates mating receptivity in females (Boppré, 1978). Male pheromones have been 106 shown to serve a variety of functions across ithomines, acting purely as an attractant in some 107 species, and as both a short-range female-attractant and long-range male-repellent in others 108 (Pliske, 1975). Consistent with these differences, pheromone blends vary qualitatively and 109 quantitatively across the tribe (Edgar et al. 1976; Brown, 1984; Brown 1987; Schulz 1988; Trigo 110 et al. 1994; Trigo et al. 1996; Schulz et al. 2004; Stamm et al. 2019), and male mating strategies 111 range from establishing and defending territories, to gregarious leks, or aggressive ‘take downs’ 112 of females (Pliske, 1975). 113 The strong sexual dimorphism in adult attraction to PA sources, and the utilisation of these 114 chemicals for both chemical defence and pheromones, suggests that olfactory adaptations to 115 detect these compounds have had particular importance in ithomiine evolution. Indeed, the first 116 sexually dimorphic MGC recorded in butterflies was recently described in an ithomine, Godyris 117 zavaleta (Montgomery and Ott, 2015). This observation was surprising given evidence MGCs 118 were lost early in butterfly evolution, with even species within the relatively closely related 119 subfamily lacking MGCs (Heinze and Reppert, 2012). This strongly suggests the 120 independent, convergent evolution of a once lost neural adaptation in this tribe, a striking example 121 of a reversal in a phylogenetic trend. However, it remains unclear how prevalent MGCs are across 122 ithomiines, whether this structure functions solely for the location of PA sources, or whether it is 123 used for long-distance pheromone detection (Montgomery and Ott, 2015). Here, we utilise a 124 comparative approach, sampling a further twelve, diverse species of ithomiine, to investigate 125 these questions. 126 127 Methods 128 129 i) 130 Specimens were collected in the Estación Científica Yasuní, in the Parque Nacional Yasuní, 131 Orellana Province, Ecuador, during November-December 2011, and September-October 2012, 132 under permit 0033-FAU-MAE-DPO-PNY and exported under permits 001-FAU-MAE-DPO-PNY 133 and 006-EXP-CIEN-FAU-DPO-PNY. Permits were obtained from Parque Nacional Yasuní, 134 Ministerio Del Ambiente, La Dirección Provincial de Orellana. Species representing 12 genera, 135 excluding Godyris, were selected on the basis of phylogenetic distribution and available sample 136 size, and represent 8 of the 10 ithomiini subtribes (Table S1). Dissection and fixation of specimens 137 were performed at the Estación Científica Yasuní. Brains were exposed by removing a section of

4 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

138 head carapace under HEPES-buffered saline (HBS; 150 mM NaCL; 5mM KCL; 5 mM CaCl2; 25 139 mM sucrose; 10mM HEPES; ph 7.4), before being fixed with zinc formaldehyde solution (ZnFA;

140 0.25% [18.4 mM] ZnCl2; 0.788% [135mM] NaCl; 1.2% [35mM] sucrose; 1% formaldehyde) for 16- 141 20 hours. Extraneous head tissue was then removed, and brains were washed three times in 142 HBS. Samples were transferred to 80% methanol/20% DMSO for at least two hours, then stored 143 in 100% methanol. Samples were kept at room temperature until return to the UK, then transferred 144 to -20°C. 145 146 ii) Immunohistochemistry 147 Samples were rehydrated using serial Tris buffer-methanol solutions (90%, 70%, 50%, 30% and

148 0%), each for 10 minutes. Brains were then incubated for two hours in NGS-PBSd (5% Normal 149 Goat Serum, 1% DMSO, 94% 0.1M PBS), before being exposed to anti-SYNORF1 (Antibody 150 3C11; Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA, RRID:

151 AB_2315424; Buchner, 1996) in solution with PBSd-NGS, at a ratio of [1:30], for 3.5 days at 4°C.

152 Non-bound antibody was removed by washing with PBSd (1% DMSO, 99% 0.1M PBS) three 153 times. Goat anti-mouse secondary antibody, Cy2-conjugated (Jackson ImmunoResearch; Cat

154 No. 115–225‐146, RRID: AB_2307343, West Grove, PA), was then applied at [1:100] in PBSd- 155 NGS for 2.5 days at 4°C. Samples were imbued with glycerol through graded exposure in 0.1M 156 Tris buffer (1%, 2%, 4% each for two hours, and 8%, 15%, 30%, 60%, 70%, 80%, each for 1 157 hour), and full dehydrated by washing with 100% ethanol (3 x 30 minutes). Methyl Salicylate was 158 then underlaid, and the brains allowed to sink. This was repeated twice before transfer to storage 159 vials of methyl salicylate. 160 161 iii) Confocal imaging and Image Segmentation 162 Samples were mounted in methyl salicylate between two cover slips, either side of a hole bored 163 through an aluminium slide. Mechanitis and Ithomina were imaged on a Leica SP5 microscope 164 using a 10x 0.4NA objective. All other species were scanned on an Olympus IX3-SSU using a 165 10x 0.4NA objective. As we do not compare raw volumes across species, the use of different 166 microscopes does not affect our subsequent analyses. For each individual, a single stack was 167 taken encompassing the whole AL with a z-step of 1µm between each optical section and a x-y 168 resolution of 1024x1024 pixels. Consistent light detection was ensured by adjusting the laser 169 intensity and gain with depth. To correct for the artefactual shortening of the z-dimension of 170 images due to the air objective lenses, a correction factor of 1.52 was applied to the z-dimension 171 of the image stacks (Heinze and Reppert, 2012; Montgomery and Ott, 2015). Image segmentation

5 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

172 was performed in Amira 5.4.1 (ThermoFisher Scientific; RID: SCR_007353). Volumes of 173 evaluated areas were exported using the measure statistics tool. Volumes for surface models 174 were plotted using the nat R package (Bates et al. 2020). Putative MGCs were identified on the 175 basis of internal fibrous structure and location at the base of the antennal nerve, as described in 176 Montgomery and Ott (2015). To formally examine the presence of MGs, each individual 177 glomerulus was segmented in two focal male ALs for each species. In Methona 4 males were 178 analysed in this way to confirm the apparent lack of both MGs and MGCs (see results). Males 179 were chosen for this initial assessment as the MGs are larger in Godyris males (Montgomery and 180 Ott, 2015), as is commonly observed across moths (Rospars and Hildebrand, 1992; Huetteroth 181 and Schachtner, 2005; El Jundi et al., 2009b; Løfaldli et al., 2010; Yan et al. 2019). Subsequent 182 visual inspection of females did not reveal any glomeruli specifically enlarged in females. Three 183 tests were used to evaluate whether MGs were present in focal individuals, based on 184 methodologies used in previous publications (Montgomery and Ott, 2015; Keubler et al, 2010; 185 Kelber et al. 2009). As the method used by Keubler et al (2009) was seen to be the most 186 conservative (see Supplementary File) we focus on these results in the main text. Under this 187 method, a glomerulus is considered to be a MG if its volume is greater than the 90th percentile of 188 glomeruli volumes plus k times the difference between the 10th and 90th percentiles (Kuebler et 189 al. 2010). We discriminated between MGs and normal glomeruli using a threshold k value of 1.5, 190 which defines a moderate outlier (Sachs, 1988). MGs were considered to be present if they 191 passed the discrimination threshold in both fully segmented males. We note that our definition of 192 a MG is dependent upon the volumetric distribution of the other glomeruli, meaning that expansion 193 of non-MGC glomeruli may obscure the presence of glomeruli expanded to a similar degree as 194 MGs in other genera. We therefore provide the glomeruli size distributions for all species Figures 195 S2-14. 196 For subsequent evaluations of sexual dimorphism we segmented: i) all glomeruli 197 comprising the putative MGC, including both MGs and closely associated satellite glomeruli 198 identified based on their physical position and internal structure, ii) the combined volume of all 199 glomeruli, iii) total AL volume including glomeruli and the internal antennal lobe hub. Data for 200 Godyris zavalata was taken from Montgomery and Ott (2015) but images were checked for 201 consistency with newly obtained data. 202 203 iv) Statistical analyses 204 In each species, linear models were used to test whether sexual dimorphism is present in the 205 MGC glomeruli using the total volume of non-MGC glomeruli, calculated by subtracting the

6 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

206 volume of MGC glomeruli from the total glomerular volume, as an additional factor to control for 207 overall size. We also examined sexual dimorphism in total glomerular volume minus MGC 208 volumes, correcting for the AL hub volume, a region of low synaptic density that is composed of 209 tracts interlinking glomeruli and higher brain areas, to account for size variation. Normality of the 210 residual errors was assessed using a Shapiro-Wilks test, and equal variance a Breusch-Pagan 211 test. Data was seen to be normal in all cases. Where variance of errors was not equal we 212 reassessed the data with robust error linear models. These corroborated the results of the linear 213 models in all cases. Multiple testing was accounted for using a sequential Bonferroni correction 214 (Benhamini and Hochberg, 1995). As sample size is limited in some species, we also report the 215 Hedge’s g effect size (Hedges and Olkin, 1985) for all statistical tests. 216 217 Results 218 219 i) Ithomiines vary in the presence and form of the macro-glomerular complex 220 Ithomiine ALs are similar in structure to other Lepidoptera, glomeruli (averaging between 64-74 221 glomeruli; Table S2) positioned around a central fibrous region, the ‘AL hub’, which contains 222 projections between glomeruli and downstream targets (Huetteroth and Schachtner, 2005; El 223 Jundi et al., 2009b; Heinze and Reppert, 2012; Montgomery and Ott, 2015). MGs that passed our 224 statistical threshold were observed in Melinaea, Mechanitis, Forbestra, Ithomia, Hypothryris, 225 Pseudoscada, and Hypoleria (Figure 1). These structures were all components of a multi- 226 glomerular complex (MGC) located at the dorsal base of the antennal nerve, with the exception 227 of P. florula where we observe one MG within the MGC, and a second on the dorsomedial AL 228 surface. Two MGs within the putative MGC cluster are observed in Mechanitis and Melinaea 229 (Table S2, Figure S3-S4). In species lacking MGs, with the exception of Methona, we still identify 230 a glomerular complex at the dorsal base of the antennal nerve that is distinct from other glomeruli 231 in the remaining four species (Figure 1; Figure S2-4, 6-14). Despite the variable presence of a 232 formally determined MG within this structure, we refer to it as an MGC throughout as we 233 hypothesise it has a homologous function across species, with differential expansion of individual 234 glomeruli. 235 These MGCs occur in a corresponding position to that observed in Godyris, but the degree 236 to which this structure is raised with respect to the AL surface is variable. Oleriina species exhibit 237 a more homogenous glomerular surface than those of other subtribes, whereas the MGC is 238 particularly pronounced in species of Dircennina and Godyridina (Figure S11-14). The 239 composition of the MGC is also highly variable across the tribe. In the basal Mechanitina, the

7 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

240 MGC contains two glomeruli. In representatives of Melinaeina, Ithomiina, Napeogenina, and 241 Olerinia we see a single additional satellite glomerulus. Despite being within Olerinia, Hyposcada 242 does not share this satellite glomeruli, and the two observed MGC glomeruli are of reduced size. 243 The MGC observed in Dircennina and Godyridina have acquired further satellite glomeruli, with 244 their MGCs containing 4 glomeruli, or 5 in the case of Pseudoscada. The most morphologically 245 complex MGC is observed in Pseudoscada, with one MG and four satellite glomeruli. 246 247 ii) Variation in sexual dimorphism in antennal lobe structures 248 Within the MGC, we also observe varying levels of sexual dimorphism across ithomiines. The 249 size of at least one glomerulus is sexually dimorphic, being of greater size in males, in Mechanitis, 250 Forbestra, Hypothyris, Pseudoscada, and Godyris (Table 1a). This is limited to a single 251 glomerulus in all genera with the exception of Pseudoscada, which has three sexually dimorphic 252 glomeruli (Table 1a; Figure 2). After multiple test correction, we do not observe dimorphism in any 253 MGC glomeruli of Melinaea, Ithomia, Napeogenes, Hyposcada, Oleria, Calithomia, and Hypoleria 254 (Table 1a). However, we note that in many non-significant tests we observe effect sizes that are 255 comparable with those observed in sexually dimorphic genera (Table 1a) (Cohen, 1988). This 256 may suggest that with increased sample size some of these genera would also display significant 257 levels of sexual dimorphism. The total volume of non-MGC glomeruli were seen to be sexually 258 monomorphic, with the exception of Hyposcada (Table 1b). 259 260 Discussion 261 262 Within Lepidoptera, significant variation has been observed in the volume of olfactory processing 263 areas. In moths, males have vastly expanded glomeruli in the antennal lobe (MGs/MGCs) that 264 are typically sexually dimorphic and responsive to pheromones (Bretschneider, 1924; Matsumoto 265 and Hildebrand, 1981; Koontz and Schneider, 1987). Here, we present evidence for the 266 widespread distribution of MGCs across the Ithomiini, the only tribe of butterflies currently known 267 to possess this kind of olfactory specialisation (Montgomery and Ott, 2015). We provide evidence 268 that, having secondarily re-evolved this AL specialisation, ithomiine MGC have diversified, and 269 are variable across species in both composition and the degree of sexual dimorphism. 270 Two hypotheses have been proposed to explain why ithomiines have secondarily acquired 271 a trait lost at the base of butterflies. First, ithomiine MGCs could reflect heightened sensitivity of 272 males to PA resources, increasing their foraging efficiency (Montgomery and Ott, 2015). PAs are 273 plant-derived chemicals utilised for several key fitness traits; chemical defence, pheromone

8 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

274 synthesis, and nuptial gifts (Pliske, 1975, Pliske et al. 1976; Brown, 1984). Scarcity of PA 275 resources, or competition amongst males, may have increased the selective pressure for 276 sensitivity to PAs and led to expanded processing regions. This hypothesis would predict a 277 consistent pattern of MG presence and sexual dimorphism across all species with this ecological 278 strategy. However, our results are not consistent with this prediction. Both the presence of an MG 279 and the presence and degree of sexual dimorphism vary across ithomiine genera strongly 280 attracted to PAs. This implies a lack of direct association between PA use, or detection, and 281 MGCs. Despite its variability, a MGC, or at least a positionally homologous glomerular cluster, is 282 identifiable in all members of the tribe, with the exception of Methona. The absence of an MGC in 283 Methona may instead be consistent with an indirect relationship between the use of PAs and the 284 presence of an MGC, as this genus does not strongly rely on PAs for chemical defence, nuptial 285 gifts or pheromone precursors (Brown, 1987; McClure et al. 2019). In addition, representatives of 286 subtribes with the most morphologically pronounces MGCs (Godyriina, Ithomiia) have particularly 287 strong male responses to PA baits (Pliske 1975, Brown 1984). Despite this, the lack of ubiquitous 288 sexual dimorphism and inconsistency of MGC structure argue against a singular role for the MGC 289 in PA foraging. Hence, the structural diversification of the ithomiine MGC may relate to how, rather 290 than merely if, PAs are used in sexual communication. 291 Indeed, the second hypothesis for the origin of this structure suggests a role in pheromone 292 processing (Montgomery and Ott, 2015). Across ithomiines, male pheromones are used for the 293 attraction of females, with some lineages additionally utilising pheromones in male territorial 294 defence (Pliske 1975). In many of these species, males are repulsed by lactones, which are 295 present in the pheromone blend of phylogenetically disparate ithomiines (Schulz et al. 2004). 296 Genera whose pheromone blends contain lactones (Melinaea, Ithomia, Hypothyris, Calithomia, 297 Pseudoscada, Godyris, and Hypoleria (Schulz et al. 2004)) all have additional glomeruli 298 incorporated into their MGCs. Whilst expanded MGCs are also observed in Napeogenes and 299 Oleria, which lack pheromonal lactones, these genera also have novel pheromone blends, 300 utilising compounds that are not observed in other ithomiines (Shultz et al. 2004; Stamm et al. 301 2019). The differentiation of pheromone repertoires in these species may suggest that OR 302 duplication events have enabled the utilisation of additional pheromones, that were either latent 303 or thereafter derived. In addition, in genera where the expression of PA-derived pheromones is 304 reduced, such as Hyposcada (Schulz et al. 2004; Stamm et al. 2019), we observe a reduction in 305 the volume and/or number of MGC glomeruli. These features suggest a significant role of 306 pheromone usage in the function of the ithomiine MGC. The position of the ithomiine MGC is also 307 comparable to the putative pheromone processing cluster of glomeruli in D. plexxipus, which

9 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

308 shares a similar fibrous structure (Heinze and Reppert, 2011; Montgomery and Ott, 2015). The 309 important role that PAs play in the ecology of Danaid species may suggest that this glomerulus 310 was ancestrally sensitive to PA-derived chemicals in Danaids and ithomiines, and was elaborated 311 in ithomiines (Brown 1987, Schulz et al. 1988). 312 While our data is consistent with a role for the MGC in sexual communication, sexual 313 dimorphism is variable across the tribe, showing no clear phylogenetic pattern. The lack of 314 consistent sexual dimorphism may reflect similar reliance of each sex on these cues, but in 315 differing contexts and/or differing valances. PA-derived pheromones have been hypothesised to 316 represent honest signals of male quality (Trigo et al. 1994), which may be salient cues for both 317 reproductively receptive females, and males during male-male competition. In this case sexual 318 dimorphism could be explained by inhibitory adaptation, in which chronic exposure to odorants 319 results in the expansion of responsive glomeruli through increased innervation by inhibitory AL 320 local neurons (Devaud et al. 2001; Sachse et al. 2007; Anton et al. 2015). Male ithomiines are 321 strongly attracted to PA resources, with sensitivity to these compounds essential, despite being 322 a source of these volatile compounds themselves (Brown, 1984; Trigo et al. 1996). Male-biased 323 sexual dimorphism may therefore be a mechanism to overcome chronic self-exposure to PAs. 324 This hypothesis would predict varying degrees of sexual dimorphism depending on both the 325 amount of pheromone utilised by males, and the similarity of compounds in the male-pheromone

326 blend and PAs used for chemical defence. Further, we note that as our samples are derived from 327 wild populations and some of our volumetric estimates show high levels of intraspecific variation, 328 which could be consistent with a plastic response to differential exposure to PAs/PA-derivatives 329 in the adult or juvenile environment. 330 While neither hypothesis solely explains the emergence and full diversity of ithomiine 331 MGCs, the combination of selection associated with finding and utilising PAs for pheromonal 332 compounds likely explains both for the origin and diversification of this structure, respectively. 333 However, the mechanism behind this neural innovation is unclear. Butterfly pheromones are 334 processed by glomeruli that are also responsive to plant odours(Larsdotter-Mellström et al. 2016), 335 unlike moths, in which pheromone signals are processed by a separate subset of glomeruli from 336 other odours (Kanzaki and Shibuya, 1983; Christensen and Hildebrand, 1987; Hansson et al. 337 1991). The evolution of a MGC in the evolutionary context in which the pheromone response 338 pathway is integrated into general odour processing raises several questions: i) did the MGC 339 originate from existing glomeruli that were sensitive to plant emitted PAs, becoming specialised 340 to detect PA-derived pheromones? ii) were MGC glomeruli, and their associated olfactory 341 receptors, produced by co-option of existing receptors or duplication of PA-sensitive receptors?

10 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

342 iii) do pheromonal responses remain integrated with general odours during olfactory processing 343 or have parallel pathways evolved in ithomiines? 344 We suggest that these questions can be answered through a combination of functional 345 assays and phylogenetics. For example, the response of MGs to pheromone extracts and/or plant 346 PAs can be assessed using electrophysiology or in vivo Ca2+ imaging, in ithomiines and in danaids 347 comparing putatively homologous glomeruli. Similar approaches can be used to evoke odour 348 maps in response to different stimuli (Galizia et al. 2000; Carlsson et al. 2011) which would 349 demonstrate whether pheromones are encoded in combination with general odours, as observed 350 in other butterflies (Larsdotter-Mellström et al. 2016), or separately, as seen in moths (Boeckh et 351 al. 1965). Finally, identifying ORs that are functionally linked to MGCs through analyses of their 352 sex-biased expression in species showing glomerular dimorphism, combined with phylogenetic 353 studies to test homology of these loci with plant-detecting or pheromone-detecting ORs in other 354 Danaiids and more broadly across Lepidopterans, would provide a test of the ancestral function 355 of this pathway. 356 In conclusion, we have identified a highly dynamic cluster of glomeruli in the ithomiine tribe 357 that includes significantly expanded MGs. To date, ithomiines are the only tribe of butterflies 358 reported to have acquired this AL specialisation, strongly implying it has evolved secondarily. We 359 show that ithomiine MGCs are variable in presence, composition, and degree of sexual 360 dimorphism. Our data supports the hypothesis that foraging for plant derived chemical defence 361 may be the ancestral source of selection pressure favouring the evolution of MGCs in this tribe, 362 with their subsequent elaboration associated with the diversification of ithomiine pheromonal 363 cues. This line of communication could be particularly important in ithomiines as they form multi- 364 species mimicry rings, with ecological convergence within co-mimetic species (Elias et al. 2008) 365 potentially rendering long range visual mating cues less reliable. 366 367 Acknowledgements: 368 We thank Álvaro Barragán, Emilia Moreno, PabloJarrín, and David Lasso from the Estación 369 Científica Yasuní and Pontificia Universidad Católica del Ecuador, and Maria Arévalo from the 370 Parque Nacional Yasuní Ministerio Del Ambiente for assistance with collection and exportation 371 permits, and Francisco Ramlrez Castro for assistance in the field in 2011. We also thank Swidbert 372 Ott for advice early on in this project, and Matt Wayland and the Imaging Facility at the Dept of 373 Zoology, University of Cambridge for confocal support. This work was funded by a Royal 374 Commission for the Great Exhibition Research Fellowship, a Royal Society Research Grant and 375 a NERC IRF (NE/N014936/1) to SHM.

11 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

376 377 References: 378 Andersson J, Borg-Karlson A-K, Vongwanich N, Wiklund C. 2007 Male Sex Pheromone 379 Release and Female Mate Choice in a Butterfly. J. Exp. Biol. 210, 964-970. 380 (doi:10.1242/jeb.02726) 381 382 Anton S, Hansson BS. 1994 Central Processing of Sex Pheromone, Host Odour, and 383 Oviposition Deterrent Information by Interneurons in the Antennal Lobe of the Female 384 Spodoptera littoralis (Lepidoptera: Noctuidae). J. Comp. Neurol. 350, 199-214. 385 (doi:10.1002/cne.903500205) 386 387 Anton S, Chabaud MA, Schmidt-Büsser D, Gadenne B, Iqbal J, Juchaux M, List O, Gaertner C, 388 Devaud JM. 2015 Brief Sensory Experience Differentially Affects the Volume of Olfactory Brain 389 Centres in a Moth. Cell Tissue Res. 364, 59-65. (doi:10.1007/s00441-015-2299-0) 390 391 Bastin F, Couto A, Larcher V, Phiancharoen M, Koeniger G, Koeniger N, Sando J-C. 2018 392 Marked Interspecific Differences in the Neuroanatomy of the Male Olfactory System of Honey 393 Bees (Genus Apis). J. Comp. Neurol. 526, 3020-3034. (doi:10.1002/cne.24513) 394 395 Bates AS, Manton J D, Jagannathan SR, Costa M, Schlegel P, Rohlfing T, Jefferis GSXE. 2020 396 The natverse, a Versatile Toolbox for Combining and Analysing Neuroanatomical Data. eLife 9, 397 e53350. (doi:10.7554/eLife.53350) 398 399 Berg BG, Galizia CG, Brandt R, Mustaparta H. 2002 Digital Atlases of the Antennal Lobe in Two 400 Species of Tobacco Budworm Moths, the Oriental Helicoverpa assulta (Male) and the American 401 Heliothis virescens (Male and Female). J. Comp. Neurol. 446, 123-134. 402 (doi:10.1002/cne.10180) 403 404 Benjamini Y, Hochberg Y. 1995 Controlling the False Discovery Rate: A Practival and Powerful 405 Approach to Multiple Testing. J. R. Stat. Soc. B 57, 289-300. (doi:10.1111/j.2517- 406 6161.1995.tb02031.x) 407 408 Bisch-Knaden S, Carlsson MA, Sugimoto Y, Schubert M, Mißbach C, Sachse S, Hansson BS. 409 2012 Olfactory Coding in Five Moth Species from Two Families. J. Exp. Biol. 215, 1542-1551. 410 (doi:10.1242/jeb.068064) 411 412 Boeckh J, Kaissling KE. Schneider D, 1965 Insect Olfactory Receptors. Cold Spring Harb. 413 Symp. Quant. Biol. 30, 263–280. (doi:10.1101/sqb.1965.030.01.028) 414 415 Boppré M. 1978 Chemical Communication, Plant Relationships and Mimicry in the Evolution of 416 Danaid Butterflies. Entomol. Exp. Appl. 24, 64-77. (doi:10.1111/j.1570-7458.1978.tb02782.x) 417 418 Bretschneider. F. 1924 Über die Gehire des Eichenspinners und des Seidenspinners 419 (Lasiocampa quercus L. und Bombyx mori). Jena. 2. Naturwissenschaften 60, 563-578.

12 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

420 421 Brower AVZ, Willmott K, Silva-Brandão KL, Garzón-Orduña IJ, Freitas AVL. 2014 Phylogenetic 422 Relationships of Ithomiine Butterflies (Lepidoptera: : Danainae) as Implied by 423 Combined Morphological and Molecular Data. System. Biodivers. 12, 133-147. 424 (doi:10.1080/14772000.2014.899650) 425 426 Brown KS. 1984 Chemical Ecology of Dehydropyrolizidine Alkaloids in Adult Ithomiinae 427 (Lepidoptera: Nymphalidae). Rev. Bras. Biol. 44, 435-460. 428 429 Brown Jr. K. 1987 Chemistry at the Solanaceae/Ithomiinae Interface. Ann. Missouri Bot. 74, 430 359-397. (doi:10.2307/2399406) 431 432 Butenandt VA, Beckman R, Stamm D, Hecker E. 1959 Uber den sexsuallockstoff des 433 eidenspinners Bombyx mori. Reindarstellung und konstitution. Z. Naturforschg. 14, 283–28. 434 435 Carlsson MA, Galizia CG, Hansson BS. 2002 Spatial Representations of Odours in the Antennal 436 Lobe of the Moth Spodoptera littoralis (Lepidoptera: Noctuidae). Chem. Senses 27, 231-244. 437 (doi:10.1093/chemse/27.3.231) 438 439 Carlsson MA, Bisch-Knaden S, Schäpers A, Mozuraitis R, Hansson BS, Janz N. 2011 Odour 440 Maps in the Brains of Butterflies with Divergent Host-Plant Preferences. PLoS One 6, e24025. 441 (doi:10.1371/journal.pone.0024025) 442 443 Carlsson MA, Schäpers A, Nässel DR, Janz N. 2013 Organization of the Olfactory System of 444 Nymphalidae Butterflies. Chem. Senses 38, 355-367. (doi:10.1093/chemse/bjt008) 445 446 Chambille I, Rospars JP, Masson C. 1980 The Deutocerebrum of the Cockroach Blaberus 447 craniifer Burm: Spatial Organization of the Sensory Glomeruli. J. Neurobiol. 11, 135–157. 448 (doi:10.1093/chemse/bjt008) 449 450 Chazot N, Willmott K, Lamas G, Freitas A, Piron-prunier P, Arias C, Mallet J, De-silva DL, Elias 451 M. 2019 Renewed Diversification Following Miocene Landscape Turnover in a Neotropical 452 Butterfly Radiation. Glob. Ecol. Biogeogr. 28, 1118-1132. (doi:10.1111/geb.12919) 453 454 Chengzhe L, Wang H, Chen X, Yao J, Shi L, Zhou C. 2017 Role of Visual and Olfactory Cues in 455 Sex Recognition in Butterfly Cethosia cyane cyane. Sci. Rep. 7: 5033. (doi:10.1038/s41598- 456 017-04721-6) 457 458 Christensen TA, Hildebrand JG. 1987 Male-Specific, Sex Pheromone Selectinve Projection 459 Neurons in the Antennal Lobes of the Moth Manduca sexta. J. Comp. Physiol. 160, 553-569. 460 (doi:10.1007/BF00611929) 461 462 Cohen J. 1988 Statistical Power Analysis for the Behavioral Sciences. New York, NY: 463 Routledge Academic.

13 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

464 465 Constanzo K, Monteiro A. 2007 The Use of Chemical and Visual Cues in Female Choice in the 466 Butterfly Bicyclus anynana. Proc. R. Soc. B 274, 845-851. (doi:10.1098/rspb.2006.3729) 467 468 Couto A, Lapeyre B, Thiéry D, Sandoz J-C. 2016. Olfactory Pathways of the Hornet Vespa 469 velutina: New Insights into the Evolution of the Hymenopteran Antennal Lobe. J. Comp. Neurol. 470 524, 2335-2359. (doi:10.1002/cne.23975) 471 472 Couto A, Wainwright JB, Morris BJ, Montgomery SH. 2020 Linking Ecological Specialisation to 473 Adaptions in Butterfly Brains and Sensory Systems. Curr. Opin. Insect. Sci. In Press. 474 (doi:10.1016/j.cois.2020.09.002) 475 476 Darragh K, Vanjari S, Mann F, Gonzalez-Rojas MF, Morrison CR, Salazar C, Pardo-Diaz C, 477 Merrill RM, McMillan WO, Schulz S, Jiggins CD. 2017. Male Sex Pheromone Components in 478 Heliconius Butterflies Released by the Androconia Affect Female Choice. PeerJ 5: e3953 479 (doi:10.7717/peerj.3953) 480 481 Devaud JM, Acebes A, Ferrús A. 2001 Odor Exposure Causes Central adaptation and 482 Morphological Changes in Selected Olfactory Glomeruli in Drosophila. J. Neurosci. 21, 6274- 483 6282. (doi:10.1523/JNEUROSCI.21-16-06274.2001) 484 485 Edgar J, Culvenor CCJ, Pliske.TE. 1976 Isolation of a Lactone, Structurally Related to the 486 Esterifiying Acids of Pyrrolizidine Alkaloids, from the Costal Fringes of Male Ithomiinae. J. 487 Chem. Ecol. 2, 263-270. (doi:10.1007/BF00988276) 488 489 Elias M, Gompert Z, Jiggins C, Willmott K. 2008 Mutualistic Interactions Drive Ecological Niche 490 Convergence in a Diverse Butterfly Community. PLoS Biol. 6, e300. 491 (doi:10.1371/journal.pbio.0060300) 492 493 El Jundi B, Huetteroth W, Kurylas A E, Schachtner J. 2009 Anisometric brain dimorphism 494 revisited: implementation of a volumetric 3D standard brain in Manduca sexta. J. Comp. Neurol. 495 517, 210–225. (doi:10.1002/cne.22150) 496 497 Galizia CG, Sachse S, Rappert A, Menzel R. 1999 The Glomerular Code for Odor 498 Representation is Species Specific in the Honeybee Apis mellifera. Nat. Neurosci 2, 473-478. 499 (doi:10.1038/8144) 500 501 Galizia CG, Sachse S, Mustaparta H, 2000 Calcium Responses to Pheromones and Plant 502 Odours in the Antennal Lobe of the Male and Female Moth Heliothis virescens. J. Comp. 503 Physiol. A 186, 1049-1063. (doi:10.1007/s003590000156) 504 505 Gronenberg W, Hölldobler B. 1999 Morphological Representation of Visual and Antennal 506 Information in the Ant Brain. J. Comp. Neurol. 412, 229-240. (doi:10.1002/(SICI)1096- 507 9861(19990920)412:2<229::AID-CNE4>3.0.CO;2-E)

14 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

508 509 Hallem EA, Carlson JR, 2006 Coding of Odors by a Receptor Repertoire. Cell 125, 143-160. 510 (doi:10.1016/j.cell.2006.01.050) 511 512 Hansson BS, Christensen TA, Hildebrand JG. 1991 Functionally Distinct Subdivisions of the 513 Macroglomerular Complex in the Antennal Lobe of the Male Sphinx Moth. J. Comp. Neurol. 312, 514 264-278. (doi:10.1002/cne.903120209) 515 516 Hedges LV, Olkin I. 1985 Statistical Methods for Meta-Analysis. San Diego, CA: Academic 517 Press. 518 519 Heinze S, Reppert SM. 2012 Anatomical Basis of Sun Compass Navigation 1: the General 520 Layout of the Monarch Butterfly Brain. J. Comp. Neurol. 520, 1599-1628. 521 (doi:10.1002/cne.23054) 522 523 Huetteroth W, Schachtner J. 2005 Standard Three-Dimensional Glomeruli of the Manduca sexta 524 Antennal Lobe: a Tool to Study both Developmental and Adult Neuronal Plasticity. Cell Tissue 525 Res. 319, 513–524. (doi:10.1007/s00441-004-1016-1) 526 527 Ibba R, Angioy AM Hansson BS, Dekker T. 2010 Macroglomeruli for Fruit Odors Change Blend 528 Preference in Drosophila. Sci. Nat. 97, 1059-1066.(doi:10.1007/s00114-010-0727-2) 529 530 Joerges J, Küttner A, Galizia CG, Menzel R. 1997 Representations of Odours and Odour 531 Mixtures Visualized in the Honeybee Brain. Nature 387, 285-288. (doi:10.1038/387285a0) 532 533 Kanzaki R, Shibuya T. 1983 Olfactory Neural Pathway and Sexual Pheromone Responses in 534 the Deutocerebrum of the Male Silkworm Moth, Bombyx mori (Lepidoptera: Bombycidae). Appl. 535 Entomol. Zool. 18, 131-133. (doi:10.1303/aez.18.131) 536 537 Kanzaki R, Arba EA, Strausfeld NJ, Hildebrand JG. 1989 Physiology and Morphology of 538 Projection Neurons in the Antennal Lobe of the Male Moth Manduca sexta. J. Comp. Physiol. A 539 165, 427-453. (doi:10.1007/BF00611233) 540 541 Kazawa T, Namiki S, Fukushima R, Terada M, Soo K, Kanzaki R. 2009 Constancy and 542 Variability of Glomerular Organisation in the Antennal Lobe of the Silkmoth. Cell Tissue Res. 543 336, 119-136. (doi:10.1007/s00441-009-0756-3) 544 545 Kelber C, Rössler W, Roces F, Kleineidam CJ. 2009 The Antennal Lobes of Fungus-Growing 546 Ants (Attini): Neuroanatomical Traits and Evolutionary Trends. Brain Behav. Evol. 73, 273–284. 547 (doi:10.1159/000230672) 548 549 Klun JA, Maini S. 1979 Genetic basis of an insect chemical communication system: The 550 European corn borer. Environ. Entomol. 8, 423-426. (doi:10.1093/ee/8.3.423) 551

15 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

552 Kondoh Y, Kaneshiro KY, Kimura K, Yamamoto D. 2003 Evolution of Sexual Dimorphism in the 553 Olfactory Brain of Hawaiian Drosophila. Proc. R. Soc. B 270, 1005-1013. 554 (doi:10.1098/rspb.2003.2331) 555 556 Koontz MA, Schneider D. 1987 Sexual Dimorphism in Neuronal Projections from the Antennae 557 of Silk Moths (Bombyx mori, Antheraea polyphemus) and the Gypsy Moth (Lymantria dispar) 558 Cell Tissue Res. 249, 39-50. (doi:10.1007/BF00215416) 559 560 Kreiger J, Grosse-Wilde E, Gohl T, Dewer YME, Raming K, Breer H. 2004 Genes Encoding 561 Candidate Pheromone Receptors in a Moth (Heliothis virescens). PNAS 101, 11845-11850. 562 (doi:10.1073/pnas.0403052101) 563 564 Kuebler LS, Kelber C, Kleineidam CJ. 2010 Distinct Antennal Lobe Phenotypes in the Leaf- 565 Cutting Ant (Atta vollenweideri). J. Comp. Neurol. 518, 352-365. (doi:10.1002/cne.22217) 566 567 Larsdotter-Mellström H, Eriksson K, Liblikas I, Wiklund C, Borg-Karlson AK, Nylin S, Janz N, 568 Carlsson MA. 2016 It’s All in the Mix: Blend-Specific Behavioural Response to a Sexual 569 Pheromone in a Butterfly. Front. Physiol. 7, 68. (doi:10.3389/fphys.2016.00068) 570 571 572 Løfaldli BB, Kvello P, Mustaparta H. 2010 Integration of the Antennal Lobe Glomeruli and Three 573 Projection Neurons in the Standard Brain Atlas of the Moth Heliothis virescens. Front. Syst. 574 Neurosci. 4, 5. (doi:10.3389/neuro.06.005.2010) 575 576 Matsumoto SG, Hildebrand JG. 1981 Olfactory Mechanisms in the Moth Manduca sexta: 577 Response Characteristics and Morphology of Central Neurons in the Antennal Lobe. Proc. R. 578 Soc. B 213, 249-277. (doi:10.1098/rspb.1981.0066) 579 580 Mérot C, Frérot B, Leppik E, Joron M. 2015 Beyond Magic Traits: Multimodal Matings Cues in 581 Heliconius Butterflies. Evol. 69, 2891-2904. (doi:10.1111/evo.12789) 582 583 Miura N, Nakagawa T, Tatsuki S, Touhara K, Ishikawa Y. 2009 A Male-Specific Odourant 584 Receptor Conserved Through the Evolution of Sex Pheromones in Ostrinia Moth Species. Int. J. 585 Biol. Sci. 5, 319-330. (doi:10.7150/ijbs.5.319) 586 587 Montgomery SH, and Ott SR. 2015 Brain Composition in Godryis zavaleta, a Diurnal Butterfly, 588 Reflects an Increased Reliance on Olfactory Information. J. Comp. Neurol. 523, 869-891. 589 (doi:10.1002/cne.23711) 590 591 Montgomery SH, Merrill RM, Ott SR. 2016 Brain Composition in Heliconius Butterflies, 592 Posteclosion Growth and Experience-Dependent Neuropil Plasticity. J. Comp. Neurol. 524, 593 1747-1769. (doi:10.1002/cne.23993) 594

16 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

595 Namiki S, Daimon T, Iwatsuki C, Shimada T, Kanzaki R. 2014 Antennal Lobe Organization and 596 Pheromone Usage in Bombycid Moths. Biol. Lett. 10, 20140096. 597 (doi.org/10.1098/rsbl.2014.0096) 598 599 Pliske TE. 1975a Courtship Behaviour of the Monarch Butterfly, Danaus plexippus L.. Ann. 600 Entomol. Soc. Am. 68, 143-151. (doi:10.1093/aesa/68.1.143) 601 602 Pliske TE. 1975b Attraction of Lepidoptera to Plants Containing Pyrrolizidine Alkaloids. Environ. 603 Entomol. 4, 455-473. (doi:10.1093/ee/4.3.455) 604 605 Pliske TE, Edgar JA, Culvenor CCJ. 1976 The chemical basis of Attraction of Ithomiine 606 Butterflies to Plants Containing Pyrrolizidine Alkaloids. J. Chem. Ecol. 2, 255–262. 607 (doi:10.1007/BF00988275) 608 609 Rospars JP. 1983 Invariance and Sex-Specific Variations of the Glomerular Organization in the 610 Antennal Lobes of a Moth, Mamestra brassicae, and a butterfly, Pieris brassicae. J. Comp. 611 Neurol. 220, 80-96. (doi:10.1002/cne.902200108) 612 613 Rospars JP, Hildebrand JG. 2000 Sexually Dimorphic and Isomorphic Glomeruli in the Antennal 614 Lobes of the Sphinx Moth Manduca sexta. Chem. Senses 25, 119-129. 615 (doi:10.1093/chemse/25.2.119) 616 617 Rutowski RL. 1991 The Evolution of Male Mate-Locating Behaviour in Butterflies. Am. Nat. 138, 618 1121-1139. (doi:10.1007/BF01054526) 619 620 Sachs L. 1988 Statistische Methoden: Planung und Auswertung. Berlin: Springer. 621 622 Saches S, Rueckert E, Keller A, Okada R, Tanake NK, Ito K, Vosshall LB. 2007 Activity- 623 Dependent Plasticity in an Olfactory Circuit. Neuron 56, 838-850. 624 (doi:10.1016/j.neuron.2007.10.035) 625 626 Shulz S, Francke W, Edgar J, Schneider D. 1988 Volatile Compounds from Androconial Organs 627 of Danaine and Ithomiine Butterflies. Z. Naturforsch. 43, 99-104. (doi:10.1515/znc-1988-1-219) 628 629 Schulz S, Beccaloni G, Brown KS, Boppré M, Freitas AVL, Ockendels P, Trigo JR. 2004 630 Semiochemicals Derived from pyrrolizidine Alkaloids in Male Ithomiine Butterflies (Lepidoptera: 631 nymphalidae: Ithomiinae). Biochem. Syst. Ecol. 32, 699-713. (doi:10.1016/j.bse.2003.12.004) 632 633 Schulz S, Estada C, Yildizhan S, Boppré M, Gilbert LE. 2007 An Antiaphrodisiac in Heliconius 634 melpomene Butterflies. J. Chem. Ecol. 34, 82-93. (doi:0.1007/s10886-007-9393-z) 635 636 Stamm P, Mann F, McClure M, Elias M, Schulz S. 2018 Chemistry of the Androconial Secretion 637 of the Ithomiine Butterfly, Oleria onega. J. Chem. Ecol. 45, 768-778. (doi:10.1007/s10886-019- 638 01100-5)

17 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

639 640 Stöckl A, Heinze.S, Charalabidis A, El Jundi B, Warrant E, Kelber A. 2016 Differential 641 Investment in Visual and Olfactory Brain Areas Reflects Behavioural Choices in Hawk moths. 642 Sci. Rep. 6: 26041. (doi:10.1038/srep26041) 643 644 Tabuchi M, Sakurai T, Mitsuno H, Namiki S, Minegishi R, Shiotsuki T, Uchino K, Sezutsu H, 645 Tamura T, Haupt SS, Nakatano K, Kanzaki R. 2013 Pheromone Responsiveness Threshold 646 Depends on Temporal integration by Antennal Lobe Projection neurons. PNAS 110, 15455- 647 15460. (doi:10.1073/pnas.1313707110) 648 649 Trigo JR, Barata LES, Brown KS. 1994 Stereochemical Inversion of Pyrrolizidine Alkaloids by 650 Mechanitis polymnia, (Lepidoptera: Nymphalidae: Ithomiinae): Specificity and Evolutionary 651 Significance. J. Chem. Ecol. 20, 2883-2899. (doi:10.1007/BF02098396) 652 653 Trigo JR, Brown KS, Henriques SA, Euclides L, Barata S. 1996 Qualitative Patterns of 654 Pyrrolizidine Alkaloids in Ithomiinae Butterflies. Biochem. Syst. Ecol. 24, 181-188. 655 (doi:10.1016/0305-1978(96)00008-7) 656 657 Van Dijk LJA, Janz N, Schäpers A, Gamberale-Stille G, Carlsson MA. 2017 Experience- 658 Dependent Mushroom Body Plasticity in Butterflies: Consequences of Search Complexity and 659 Host Range. Proc. R. Soc. B 284, 20171594. (doi:10.1098/rspb.2017.1594) 660 661 Wang JW, Wong AM, Flores J, Vosshall LB, Axel R. 2003 Two-Photon Calcium Imaging 662 Reveals an Odor-Evoked Map of Activity in the Fly Brain. Cell 112, 271-282. 663 (doi:10.1016/S0092-8674(03)00004-7) 664 665 Warner KW, Anderson AR, Trowell SC, Theilmann DA, Robertson HM, Newcomb RD. 2007 666 Female-biased Expression of Odourant Receptor Genes in the Adult Antennae of the Silkworm, 667 Bombyx mori. Insect Mol. Biol. 16, 107-119. (doi:10.1111/j.1365-2583.2007.00708.x) 668 669 Yan X, Wang Z, Xie J, Deng C, Sun X, Hao C. 2019 Glomerular Organization of the Antennal 670 Lobes of the Diamondback Moth, Putella xylostella L.. Front. Neuroanat. 13, 4. 671 (doi:10.3389/fnana.2019.00004) 672 673 Zhao X-C, Chen Q-Y, Guo P, Xie G-Y, Tang Q-B, Guo X-R, Berg BG. 2016 Glomerular 674 Identification in the Antennal Lobe of the Male Moth Helicoverpa armigera, J. Comp. Neurol. 675 524, 2993-3013. (doi:10.1002/cne.24003) 676 677 Zube C, Kleineidam CJ, Kirschner S, Neef J, Rössler W. 2007 Organization of the Olfactory 678 Pathway and Odor Processing in the Antennal Lobe of the Ant Camponotus floridanus. J. 679 Comp. Neurol. 506, 425-441.(doi:10.1002/cne.21548) 680 681 682

18 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

683 Figures: 684 Figure One:

685 686 Figure One: A) Ithomiine phylogeny illustrating variation in PA usage and summarizing results on MGC 687 morphology. The phylogeny is based upon Chazot et al. 2019, and displays ithomiini subtribes and 688 representative genera (Browers et al. 2014) with associated PA usage and putative MGC data. Subtribes 689 where MGC was not evaluated are shown in grey. HP denotes presence of absence of hairpencils, and the 690 sex where these are presence is indicated by F/M for female and male. PA use is categorised into three 691 primary categories, use of PAs for defense, PAs as pheromones, and whether these PA pheromones 692 contain lactones, Def,, Phero., and Lact., respectively. MGC information is given for number of MGs and 693 total MGC glomeruli (MG) and whether there is evidence of sexual dimorphism in any MGC glomeruli (SD). 694 B) Volume rendering of synapsin (3C11) immunofluorescence depicting brain anatomy, highlighting the 695 position and morphology of MGC highlighted in colour, in Oleria gunilla B’) and B’’) Napeogenes larina. C- 696 H) Surface model of full glomeruli segmentations, an example of anti‐synapsin immunofluorescence in an

19 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

697 antennal lobe confocal sections, and the distribution of glomerular volumes in illustrative genera. In 698 glomerular distribution plots, the discrimination threshold above which a glomerulus is considered a MG is 699 indicated in black, and glomeruli within the putative MGCs highlighted in red. The genera shown are 700 Methona, Forbestra, Ithomia, Hypothyris, Pseudoscada, and Godyris, C-H) respectively. For full results see 701 Figure S2-14. 702 703 Figure Two:

704 705 Figure Two: Varying levels of sexual dimorphism observed in MGC glomeruli volumes across selected 706 ithomiine genera showing (A-E) Forbestra (♀5, ♂12), Ithomia (♀11, ♂9), Callithomia (♀5, ♂6), 707 Pseudoscada (♀11, ♂7), Godyris (♀8, ♂8), respectively. Females are shown in blue, males are shown in 708 red. Significance denoted by: *<0.05, **<0.01, ***<0.001. For full results see Figure S2-14. 709 710 711 712 713 714

20 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

715 Tables: 716 Table 1a: Statistical tests for MGC glomeruli sexual dimorphism, evaluating the influence of sex 717 with the total non-MGC glomeruli volume as a second factor. Significant results are highlighted 718 in bold. 719 Genera Structure Linear Model (Structure ~ Sex + All Glomeruli) Estimate Std. Error t value* p value Bonferroni p value Hedge’s g* Melinaea MGC1 49267 25269 1.950 0.099 1.702 MGC2 12290 10126 1.214 0.270 1.059 MGC3 51579 17799 2.898 0.027 0.081 2.529

Mechanitis MGC1 30930 14994 2.063 0.069 1.346 MGC2 43309 12909 3.355 0.008 0.017 2.190

Forbestra MGC1 28997 11792 2.459 0.028 0.056 1.369 MGC2 48043 16962 2.832 0.013 0.027 1.577

Ithomia MGC1 5042 11732 0.430 0.673 0.201 MGC2 4730 3098 1.527 0.145 0.713 MGC3 11941 6044 1.976 0.065 0.922

Hypothyris MGC1 46094 15897 2.900 0.011 0.033 1.513 MGC2 -2678 8762 -0.306 0.764 -0.159 MGC3 7646 10458 0.731 0.476 0.381

Napeogenes MGC1 -6120 12716 -0.481 0.638 -0.268 MGC2 28951 21363 1.355 0.197 0.754 MGC3 1878 12216 0.154 0.880 0.086

Oleria MGC1 22264 17922 1.242 0.254 0.775 MGC2 15645 13485 1.160 0.284 0.724 MGC3 18143 15229 1.191 0.272 0.744

Hyposcada MGC1 29946 28744 1.042 0.338 0.650 MGC2 -533 14175 -0.038 0.971 -0.023

21 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Calithomia MGC1 3470 7307 0.475 0.648 0.308 MGC2 311 11999 0.026 0.980 0.017 MGC3 6356 5675 1.12 0.295 0.727 MGC4 2691 6805 0.395 0.703 0.257 Peudoscada MGC1 31996 5733 5.582 <0.001 <0.001 2.815 MGC2 2488 2473 1.006 0.330 0.507 MGC3 5870 1542 3.807 0.002 0.010 1.920 MGC4 2737 2083 1.314 0.209 0.663 MGC5 22187 4648 4.774 <0.001 0.001 2.408

Hypoleria MGC1 4508 6862 0.657 0.523 0.356 MGC2 12863 12080 1.065 0.306 0.577 MGC3 -908 6499 -0.140 0.891 -0.076 MGC4 8911 5547 1.607 0.132 0.870

Godryis MGC1 40294 12560 3.198 0.007 0.028 1.677 MGC2 -4044 6453 -0.627 0.542 -0.329 MGC3 9095 10606 0.858 0.407 0.450 MGC4 49099 8686 5.653 <0.001 <0.001 2.965 720 721 * negative values are in the direction of females, positive values in the direction of males 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740

22 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.13.336206; this version posted October 13, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

741 Table 2: Results of Statistical tests AL sexual dimorphism, evaluating the influence of sex with 742 the AL-hub volume as a second factor. Significant results highlighted in bold. 743 Genera Linear Model (AL ~ Sex + AL Hub) Estimate Std. Error t value* p value Hedge’s g* Melinaea -22946 844708 -0.027 0.979 -0.024

Mechanitis 238652 487445 0.490 0.636 0.320

Forbestra 361659 278020 1.301 0.214 0.724

Ithomia 73192 246454 0.297 0.770 0.139

Hypothyris -104491 473039 -0.221 0.828 -0.115

Napeogenes 157416 316589 0.497 0.627 0.277

Oleria -63231 229058 -0.276 0.790 -0.172

Hyposcada -1377952 315420 -4.369 0.005 -2.727

Calithomia -450012 275856 -1.631 0.141 -1.059

Peudoscada 67279 149040 0.451 0.658 0.228

Hypoleria -481089 376968 -1.276 0.224 -0.691

Godryis 57206 279056 0.205 0.841 0.108 744 745 * negative values are in the direction of females, positive values in the direction of males

23