The floral illusion: A parasitic beetle mimics the scent of flowers to attract bees

15 Animals are not known to biosynthesize floral signals to manipulate pollinators, although such 16 mimicry could profoundly shape plant-pollinator interactions. Larvae of t he poisonous European 17 blister beetle Meloe proscarabaeus parasitize multiple solitary bee species, yet the mechanism 18 enabling host attraction has remained unresolved. Here we show that these larvae lure bees by 19 emitting a bouquet of volatile compounds that closely resembles floral scent. Chemical analyses 20 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 2 reveal a complex blend of monoterpenoids derived from ( S)-linalool, a ubiquitous floral volatile. 21 Behavioral assays demonstrate that these compounds function as floral cues, eliciting attraction in 22 bees. Transcriptomic and functional analyses identify cytochrome P450 enzymes that oxidize (S)-23 linalool, demonstrating that larvae biosynthesize these plant-like volatiles de novo. Together, these 24 findings broaden the scope of interkingdom chemical mimicry and uncover a strik ing form of 25 sensory deception in which an insect chemically assumes the signal identity of a flower, revealing 26 that animals can evolve biosynthetic pathways to exploit plant–pollinator communication. 27 28 Main 29 Mimicry, broadly defined as deceptive resemblance, is a widespread evolutionary strategy utilized 30 across the tree of life 1. Since Bates’ seminal observations of visual mimicry among Amazonian 31 butterflies in 1862 2, the concept has broadened to include decep tion across all major sensory 32 modalities, including visual 3-5, tactile6, acoustic7, and chemical8-12. Although many studies focus 33 on mimicry in predator -prey or host -parasite interactions 13-15, deceptive resemblance also plays 34 key roles in other interspecific relationships. One striking example is phoresy, where one organism 35 (the phoront) uses another (the host) for transport to essential resources 16. Phoronts frequently 36 evolve morphological, behavioral, and/or chemical traits that imitate benign or mutualistic species, 37 deceiving hosts into providing transport 8, 17 -19. Moreover, phoresy has been proposed as an 38 evolutionary precursor to parasitism in certain systems 16, and complete dependence on a host for 39 dispersal can further drive the emergence of obligate parasitic relationships8. 40 Blister beetles (Coleoptera: Meloidae) exemplify the intersection of mimicry, phoresy, and 41 parasitism. Most species parasitize solitary bees 20. In the subfamily Meloinae, gravid females 42 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 3 deposit large egg clutches underground; the eclosed larvae (triungula) are highly mobile and either 43 climb nearby vegetation to await hosts (Fig. 1A)8, 21, 22 or, in non-phoretic species, actively search 44 for bee nests on the soil surface 23. These larvae commonly target solitary bees, rapidly attaching 45 to them, and are then transferred to the bee nest. Once in the nest, triungula dismount, consume 46 the bee egg and provisions, and later complete several distinct developmental stages before 47 emerging as adults the following year24. 48 Host-attraction strategies in kleptoparasitic phoretic Meloe species vary and often exploit host -49 specific sensory cues. For instance, Meloe franciscanus triungula in North America aggregate on 50 vegetation and emit a blend of female bee sex pheromones to specifically attract male Habropoda 51 spp.8, 22. Alternatively, M. strigulosus larvae station themselves on flowers , where they intercept 52 foraging pollinators21. In contrast, triungula of the European Black oil beetle M. proscarabaeus 53 typically form conspicuous orange aggregations on grasses and exhibit little host specificity, 54 suggesting a generalist phoretic strategy 25. This behavioral divergence led us to hypo thesize that 55 M. proscarabaeus triungula use an alternative volatile cue to attract a broad range of pollinator 56 hosts. Here, we demonstrate that M. proscarabaeus larvae emit a complex bouquet of floral-scent 57 monoterpenoids to deceive foraging bees. This dis covery reveals a previously undescribed form 58 of interkingdom chemical mimicry and expands the conceptual framework of sensory deception 59 and phoretic host attraction. 60 61 Larvae emit floral monoterpenoids 62 To characterize volatile emissions from M. proscarabaeus triungula, adult beetles were collected 63 in early spring (February–April 2024–25) from Jena, Thuringia, Germany, and maintained under 64 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 4 controlled conditions for mating and oviposition ( Supplementary Fig. 1). After approximately 65 three weeks, egg clutches hatched synchronously, and aggregating triungula were harvested for 66 volatile analysis (Supplementary Fig. 2 ). Headspace solid -phase microextraction (HS -SPME) 67 coupled with achiral gas chromatography -electron ionization -mass spectrometry (GC -EI-MS) 68 revealed a complex volatile profile, comprising more than thirty C19–C27 saturated and unsaturated 69 long-chain hydrocarbons (t r 20.95–29.22 min; Fig. 1B ; Supplementary Fig. 3 ), similar to 70 compounds previously repo rted in M. franciscanus larvae22. However, in addition to these 71 hydrocarbons, we detected several structurally distinct monoterpenoids (t r 9.83–14.05 min; Figs. 72 1C and D), including cis- and trans- diastereoisomers of linalool oxide (furanoid) (1), linalool (2), 73 multiple stereoisomers of lilac aldehyde ( 3), linalool oxide (pyranoid) ( 4), linalool-6,7-epoxide 74 (5), and lilac alcohol ( 6), and 8 -oxolinalool (7) and 8 -hydroxylinalool (8). Compound identities 75 (1–8) were confirmed by comparison with synthesized reference standards (Supplementary Fig. 76 4). 77 78 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 5 79 Fig. 1 Lifecycle and volatile emissions of Meloe proscarabaeus triungula. (A) Lifecycle of the 80 toxic European Black oil beetle M. proscarabaeus. After emerging and mating in the spring (i), 81 gravid females dig underground chambers (ii –iii) and oviposit thousands of yellow eggs (iv). 82 Triungula hatch after several weeks and form conspicuous orange aggregations on vegetation (v), 83 where they await con tact with a solitary bee host that inadvertently carries them to their nest to 84 complete their development to adulthood (vi). ( B) Total ion current (TIC) chromatogram of 85 triungulin volatile profile, obtained via HS-SPME collection and achiral GC-EI-MS analysis. (C) 86 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 6 Alongside long -chain hydrocarbons (t r 20.95–29.22 min), M. proscarabaeus triungula emit a 87 complex bouquet of floral-scent monoterpenoids (tr 9.83–14.05 min, 1–8). Peak identification: 1a 88 and 1b) (5S)-linalool oxide (furanoid)‡, 2) (S)-linalool, 3a–c) (5S)-lilac aldehyde‡, 4a and 4b) (6S)-89 linalool oxide (pyranoid)‡, 5a and 5b) (3S)-linalool-6,7-epoxide‡, 6a–c) (5S)-lilac alcohol‡, 7) (S)-90 8-oxolinalool, 8) (S)-8-hydroxylinalool. *Background; ‡Mixture of stereoisomers. (D) Structural 91 and stereochemical assignments of identified monoterpene volatiles were confirmed by 92 comparison with synthetic standards using achiral and chiral GC-EI-MS. 93 94 Chiral GC -EI-MS analysis revealed that M. proscarabaeus triungula exclusively emit the ( S)- 95 enantiomer of linalool ((S)-2) (Supplementary Figs. 5 and 6) along with a range of (S)-2-derived 96 metabolites bearing conserved stereochemistry at the C-3 position (Fig. 1D, Supplementary Figs. 97 S7–S20). Absolute configurations were verified by comparison with synthetic reference standards 98 derived from rac-, ( R)-, and ( S)-2. In total, triungula emitted 17 structurally distinct 99 monoterpenoids (1–8), all of which are known floral volatiles widespread among a ngiosperms26, 100 27 such as Berberis vulgaris, Prunus spp., and Salix spp.28-31, which serve as the main food source 101 for pollinating bees in the spring, when M. proscarabaeus triungula also emerge. However, except 102 for (S)-232 and linalool oxide 133, none of these monoterpenoids have been previously reported to 103 be produced in an insect. 104 105 Floral-scent mimicry attracts bees 106 Several monoterpenoids emitted by M. proscarabaeus triungula—including (S)-linalool ((S)-2) 107 and its derivatives (1 and 3–8)—are common floral volatiles known to attract pollinators, including 108 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 7 solitary wild bees 28-31, 34. We hypothesized that by production and emission of these volatiles, 109 triungula closely mimic a floral scent that , in turn, can lure a br oad range of phoretic hosts. 110 Although M. proscarabaeus larvae primarily parasitize ground -nesting solitary bees 20, triungula 111 have also been recovered from nests of social bee species 20, 35, suggesting that volatile emission 112 by these larvae may also mediate phoretic interactions with these unsuitable social insect hosts that 113 can eject the larvae from the nest after transmission 36. Notably, the observation of M. 114 proscarabaeus triungula on both suitable and unsuitable bee hosts is consistent with a general 115 attraction strategy. 116 To test whether M. proscarabaeus triungulin monoterpenoids attract solitary bees, we first 117 conducted dual -choice olfactometer assays with polylectic Osmia bicornis , a commercially 118 available wild bee species (Fig. 2A; Supplementary Fig. 21). Live triungula, emitting a complex 119 blend of monoterpenoids and long-chain hydrocarbons attracted both male and female O. bicornis 120 (Fig. 2B). Similarly, volatile extracts obtained via dynamic headspace volatile (DHV) collection—121 which contained markedly reduced amounts of long -chain hydrocarbons —also attracted both 122 sexes of O. bicornis (Fig. 2C ). Notably, DH V collection captured the full triun gulin 123 monoterpenoid profile, whereas parallel solvent-based extractions failed to recover these volatiles 124 (Supplementary Fig. 22). 125 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 8 126 Fig. 2 Triungulin -derived monoterpenoids attract solitary and social bees. (A) Dual-choice 127 Y-tube behavioral assays were performed with solitary (Osmia bicornis and Colletes similis) and 128 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 9 social (Apis mellifera and Bombus terrestris) bee species. (B) Both male and female O. bicornis 129 were attracted to live triungula. ( C) Volatile extracts from aggregating triungula, obtai ned via 130 dynamic headspace volatile (DHV) collection, also attracted male and female O. bicornis. (D) A 131 synthetic blend of (S)-linalool ((S)-2)-derived monoterpenoids (1–8) elicited significant attraction 132 from female C. similis, A. mellifera, B. terrestris, and both sexes of O. bicornis. (E) Behavioral 133 assays with individual monoterpenoids showed sex -specific responses in O. bicornis. Control 134 assays with wheat in both Y -tube arms confirmed no side bias. All bees, except C. similis, were 135 flower-naïve. Numbers in bars indicate absolute choices. χ² test: ***p < 0.001, **p < 0.01, *p < 136 0.05, NS = not significant. §Mixture of stereoisomers; †Geranylacetone; tentatively identified using 137 the NIST MS-Library v. 3.0 (2023). 138 139 As previously noted, M. franciscanus larvae attract male solitary bees by emitting a blend of long-140 chain hydrocarbons that mimic female bee sex pheromones 22. To determine whether triungulin -141 emitted monoterpenoids can independently elicit attraction—without the influence of co-occurring 142 long-chain hydrocarbons—we exposed male and female O. bicornis to a synthetic blend of ( S)-143 linalool derivatives (compounds 1–8; Fig. 2C ; Supplementary Fig. 23A ) lacking these 144 hydrocarbons. Both sexes were significantly attracted to this blend ( χ² test, p < 0.05). In contrast, 145 only males, but not females , responded to an analogous blend of synthetic ( R)-linalool-derived 146 monoterpenoids ( Supplementary Fig. 23B ), highlighting the importance of the natural 147 stereoisomeric composition of the triungulin volatile bouquet. 148 Considering that O. bicornis is likely a n unsuitable host for M. proscarabaeus triungula—as it 149 typically nests in pre-existing cavities in wood, hollow stems, loess, clay, or masonry37—we next 150 tested the synthetic (S)-2-derived blend in dual-choice assays with wild-caught females of Colletes 151 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 10 similis, a ground -nesting solitary bee, and with workers of the eusocial generalists Bombus 152 terrestris and Apis mellifera (Fig. 2D). These assays confirmed the attraction of C. similis and 153 additionally revealed that triungula can also attract unsuitable social bee hosts, a capacity that may 154 enhance their phoretic dispersal. For instance, M. strigulosus triungula have been observed to 155 detach from unsuitable hosts during inter -floral movements21. Such a “hop-on, hop-off” strategy 156 likely increases dispersal efficiency and may similarly occur in M. proscarabaeus. 157 While the sex pheromones of A. mellifera and B. terrestris have been described previously as long-158 chain hydrocarbon derivatives38, 39 and do not include monoterpenoids, certain colletid bees40, such 159 as Colletes cunicularius 32, use (S)-2 as a mate attractant . However, we could not detect 160 monoterpenoids in either sex of C. similis or O. bicornis (Supplementary Fig. 24). Together with 161 the observation that triungulin monoterpenoids attract several distantly related bee species, this 162 supports the conclusion that bee attraction to triungulin monoterpenoids is based on floral scent 163 mimicry rather than the imitation of bee sex pheromones. In addition, the bright orange coloration 164 and shape of the larval aggregates may also visually mimic floral signals , a possibility further 165 supported by the contrasting appearances of M. franciscanus and closely related M. violaceus41 166 triungula, which are markedly darker8, 20. Indeed, increasing evidence indicates that mimics exploit 167 multiple sensory channels simultaneously, enhancing the effectiveness of the deception42. 168 We next investigated each compound ( 1–8) individually in dual -choice assays using naïve male 169 and female O. bicornis (Fig. 2D). All monoterpenoids except (3 S)-linalool-6,7-epoxide ((3S)-5) 170 and (5S)-lilac alcohol ((5S)-6) elicited significant attraction in males (χ² test, 0.001 > p 0.05). However, (5S)-lilac 173 aldehyde ((5 S)-3), (6 S)-linalool oxide (pyranoid) ((6 S-4), (5 S-6), and ( S)-8-oxolinalool (( S)-7) 174 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 11 were significantly attractive to females (0.01 > p < 0.05). The observed attraction of female bees 175 to the monoterpenoid blend as well as several of the individual components support the hypothesis 176 that mimicry of floral scent in M. proscarabaeus larvae can bypass male bees and directly lure 177 female hosts, unlike triungula of M. franciscanus who only attract intermediary male Habropoda 178 hosts and then have to move to females during copulation in order to reach the nest . The direct 179 attraction of females by M. proscarabaeus larvae likely enhances their chances of nest entry and 180 successful development. 181 Considering that M. proscarabaeus triungula usually aggregate on vegetation but sometimes climb 182 to reach flowers, we hypothesized that larval monoterpenoids have a dual rol e: facilitating both 183 host attraction and larval aggregation. Indeed, olfactometric as says showed that triungula were 184 significantly attracted to a synthetic blend of ( S)-linalool-derived monoterpenoids ( 1–8), 185 supporting a dual role for these volatiles in bee attraction and larval clustering ( Supplementary 186 Fig. 25 , Supplementary Table 1 ). This result suggests an intriguing evolutionary scenario in 187 which triungula likely originally relied on floral volatiles as cues to locate flowers, where they 188 could await pollinating insect hosts. The emission of such scents by the larvae may have initiall y 189 merely amplified the natural floral signal. However, ultimately, this also led to the aggregation of 190 triungula, allowing clustered larvae to generate a substantially stronger bouquet than solitary 191 individuals and thereby reduc e their dependence on actual flowers. In fact, M. proscarabaeus 192 larvae typically emerge in the spring near bee nests in areas that have few ground-flowering plants 193 and where early-flowering shrubs and trees, which act as main food source for pollinating bee s, 194 are often several hundred meters away. Thus, flower mimicry may have helped M. proscarabaeus 195 to occupy new ecological niches. Remarkably , this host -attraction strategy also targets a 196 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 12 fundamental behavioral mechanism essential for the survival of pollinating insects, limiting their 197 potential for rapid coevolutionary escape. 198 199 Triungula biosynthesize monoterpenoids 200 Following eclosion, larvae of M. proscarabaeus consistently emit floral-scent monoterpenoids 1–201 8. However, it remained unclear whether these volatiles were sequestered, maternally transferred, 202 or biosynthesized de novo. However, as larvae do not feed on plants following eclosion 20, the 203 sequestration of floral monoterpenoids seems rather unlikely. T o determine if larval 204 monoterpenoids 1–8 were maternally derived, we analyzed the volatile profiles of adult females 205 and their eggs, using HS -SPME GC-EI-MS (Supplementary Figs. 26 and 27). However, apart 206 from the emission of (S)-linalool ((S)-2) by eggs (Supplementary Fig. 28), we could detect no 207 other floral monoterpenoids in eggs or adults. This result, together with the conserved ( S) 208 stereochemistry at the linalool C -3 position among compounds 1 and 3–8, led us to hypothesize 209 that triungula biosynthesize these monoterpenoids from (S)-2 (Fig. 3A). 210 211 212 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 13 213 Fig. 3 Triungula biosynthesize a range of ( S)-linalool-derived floral monoterpenoids. (A) 214 Proposed biosynthetic pathway of floral-scent volatiles in triungula showing early P450-mediated 215 oxidation of central precursor ( S)-linalool ((S)-2) to epoxide (3 S)-5 and diol ( S)-8. Spontaneous 216 ring-closure of labile epoxide (3S)-5 subsequently affords linalool oxides (5S)-1 and (6S)-4, while 217 further oxidation of ( S)-8 furnishes intermediate (S)-7 that cyclizes to lilac aldehyde (5 S)-3. 218 Enzymatic reduction of aldehyde (5S)-3 generates alcohol (5S)-6. (B) Differential gene expression 219 analysis identified a cytochrome P450 reductase (MpRed) and P450 -encoding transcripts 220 upregulated in larvae relative to adult females. Relative (Rel.) expression values are based on reads 221 per kilobase of transcript per million reads mapped (RPKM) values obtained by RNA -seq (for 222 absolute RPKM values, see Supplementary T able 4 ). ( C) Functional characterization of 223 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 14 CYP347BT1 in S. cerevisiae expressing either CYP347BT1 and MpRed or MpRed alone in the 224 presence of (S)-2. Reaction products were analyzed using GC-EI-MS. Peak identification: 1a and 225 1b) (5 S)-linalool oxide (furanoid), 2) ( S)-linalool, 4) (6 S)-linalool oxide (pyranoid), 5) (3 S)-226 linalool-6,7-epoxide. Note: Only traces of (6 S)-4 and (3 S)-5 were detected. *Background. ( D) 227 Functional characterization of CYP345BZ1 in S. cerevisiae expressing either CYP345BZ1 and 228 MpRed or MpRed alone in the presence of (S)-2. Reaction products were analyzed using GC-EI-229 MS. Peak identification: 8) (S)-8-hydroxylinalool. 230 231 As compounds 2 and 5, and 3, 7, and 8 are known intermediates in the biosynthesis of linalool 232 oxides (1 and 4) and lilac alcohol ( 6), respectively, we hypothesized that two cytochrome P450 233 (P450) enzymes generate structural diversity within the triungulin monoterpenoid bouquet by 234 oxidizing (S)-linalool ((S)-2) to (S)-8 and (3S)-5. To test this, we extracted RNA from aggregating 235 triungula and adult females, and generated a de novo transcriptome assembly. Differential gene 236 expression analysis identified 15 transcripts annotated as P450 or P450 -like, with elevated 237 expression in triungula but not in adult females, along with a putative P450 reductase ( MpRed) 238 (Fig. 3B). 239 Each candidate P450 was individually co-expressed with MpRed in Saccharomyces cerevisiae and 240 screened for biosynthetic activity in vivo in the presence of ( S)-2. One candidate , designated 241 CYP347BT1 according to P450 nomenclature, epoxidized (S)-2 to primarily yield linalool oxide 242 (furanoid) stereoisomers, (2 S,5S)- and (2R,5S)-1, along with trace amounts of (3 S)-linalool-6,7-243 epoxide ((3S)-5) and linalool oxide (pyranoid) diastereomers (6S)-4 (Fig. 3C). Another candidate 244 designated CYP345BZ1, catalyzed the terminal allylic C-H hydroxylation of (S)-2 to produce (S)-245 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 15 8-hydroxylinalool (S)-8 (Fig. 3D). Together, these results confirm that M. proscarabaeus triungula 246 independently biosynthesize floral-scent monoterpenoids. 247 Consistent with in vivo assays, HS -SPME and DH V collection extracts of triungula contained 248 relatively low amounts of (3 S)-5 and (6S)-4, compared to linalool oxide (5 S)-1 (Fig. 1B and C). 249 Furthermore, GC -EI-MS analysis indicated that (3 S)-5 undergoes both 5 - and 6 -exo-tet 250 intramolecular cyclization to form (5 S)-1 and (6 S)-4, respectively, during analysis 251 (Supplementary Fig. 29), congruent with the spontaneous degradation of synthetic (5S)-5 to (5S)-252 1 and (6S)-4 over time (Supplementary Fig. 30). The formation of both (2 R,5S)- and (2S,5S)-1 253 stereoisomers from ( S)-2 in the presence of CYP347BT1 and MpRed further suggests that 254 CYP347BT1-mediated epoxidation proceeds non -stereoselectively, initially generating labile 255 intermediates (3S,6S)- and (3S,6R)-5 from (S)-2. 256 257

258 Our results reveal a hitherto undescribed form of interkingdom aggressive chemical mimicry in 259 which M. proscarabaeus triungula collectively emit a complex bouquet of floral -scent 260 monoterpenoids to lure solitary and social bees. Behavioral assays show that these volatiles act as 261 floral-scent mimics rather than pheromone analogs, enabling attraction of male and female solitary 262 bees, promoting s uccessful phoretic dispersal , and coordinating larval aggregation . We identif y 263 two P450 enzymes, CYP347BT1 and CYP345BZ1, that generate committed biosynthetic 264 precursors ((3S)-5 and (S)-8) responsible for the suite of (S)-linalool-derived monoterpenoids (1, 265 3, 4, 6, and 7). Together, these findings broaden the evolutionary framework of mimicry by 266 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 16 demonstrating how an insect can independently produce plant-like signals to manipulate inter- and 267 intraspecific interactions. 268 269 Data availability: 270 All data required to evaluate the conclusions proposed herein are present in the Manuscript or 271 associated Supplementary Information. Genes characterized in this study are deposited in the 272 NCBI GenBank with the following accession number s: MpRed ( PX569140), CYP347BT1 273 (PX569141), and CYP345BZ1 (PX569142) and s equencing data are available at 274 https://doi.org/10.17617/3.7TXB4J. 275 276

277 We thank Daniel Veit and Angela Lehmann (MPI-ICE) for the design and construction of volatile 278 collection and behavioral bioassay apparatuses, Sarah Heinicke (MPI-ICE) for assistance with GC-279 EI-MS data acquisition, Prof. David Nelson (UTHSC) for kindly annotating P450 genes, 280 CYP345BZ1 and CYP347BT1, and Drs Klaus Gase, Mohamed Omar Kamileen, Maite Colinas, 281 Song Wu (MPI-ICE), and Hannah M. Rowland (MPI-ICE/UOL) for helpful discussion. 282 283 Funding 284 This work was supported by funding from the Max Planck Society. 285 286 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 17 Author contributions 287 Conceptualization: R.M.A., M.Ka., S.E.OC., and T.G.K.; Data curation: R.M.A., H.V., and 288 T.G.K.; Formal analysis: R.M.A., D.K., H.V., and T.G.K.; Funding acquisition: S.E.OC.; 289 Investigation: R.M.A., D.K., H.V., K.L., and A.D.; Methodology: R.M.A., D.K., M.Ku., and 290 T.G.K.; Project administration: R.M.A. and T.G.K.; Resources: S.E.OC.; Supervision: S.E.OC. 291 and T.G.K.; Validation: R.M.A., D.K., K.L., and T.G.K.; Visualization: R.M.A. and T.G.K.; 292 Writing – original draft: R.M.A. and T.G.K.; Writing – review and editing: R.M.A., D.K., H.V., 293 M.Ka., S.E.OC., and T.G.K. 294 295 Corresponding authors: 296 Correspondence and requests for materials should be addressed to [email protected] and 297 [email protected] 298 299 Ethics declarations 300 Competing interest: 301 The authors declare that they have no competing interests. 302 303 304 305 306 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 18

307 1. Font, E. Mimicry, camouflage and perceptual exploitation: The evolution of deception in 308 nature. Biosemiotics. 12, 7–24 (2019). 309 2. Bates, H.W. XXXII. Contributions to an insect fauna of the Amazon valley. Lepidoptera: 310 Heliconidæ. Trans. Linn. Soc. London. 23, 495–566 (1862). 311 3. Fathinia, B., Rastegar-Pouyani, N., E., Todehdehghan, F., Amiri, F. Avian deception using an 312 elaborate caudal lure in Pseudocerastes urarachnoides (Serpentes: Viperidae). Amphibi. 313 Reptil. 36, 223–231 (2015). 314 4. Drummond, H., Gordon, E.R. Luring in the neonate Alligator snapping turtle ( Macroclemys 315 temminckii): Description and experimental analysis. Z. Tierpsychol. 50, 136–152 (1979). 316 5. Lloyd, J.E. Aggressive mimicry in Photuris: Firefly femmes fatales. Science. 149, 653–654 317 (1965). 318 6. Wignall, A.E., Taylor, P.W. Assassin bug uses aggressive mimicry to lure spider prey. Proc. 319 Roy. Soc. B: Biol. Sci. 278, 1427–1433 (2011). 320 7. de Oliveira Calleia, F., Rohe, F., Gordo, M. Hunting strategy of the Margay (Leopardus wiedii) 321 to attract the wild Pied tamarin (Saguinus bicolor). Neotrop. Primates. 16, 32–34 (2009). 322 8. Saul-Gershenz, L.S., Millar, J.G. Phoretic nest parasites use sexual deception to obtain 323 transport to their host's nest. Proc. Nat. Acad. Sci. U. S. A. 103, 14039–14044 (2006). 324 9. Eberhard, W.G. Aggressive chemical mimicry by a Bolas spider. Science. 198, 1173–1175 325 (1977). 326 10. Mochizuki, K. Olfactory floral mimicry of injured ants mediates the attraction of 327 kleptoparasitic fly pollinators. Curr. Biol. 35, 5097–5105 (2025). 328 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 19 11. Maran, T. Mimicry and meaning: Structure and semiotics of biological mimicry (Springer, 329 2017). 330 12. Ayasse, M., Schiestl , F.P., Paulus , H.F., Ibarra , F., Francke , W. Pollinator attraction in a 331 sexually deceptive orchid by means of unconventional chemicals. Proc. Roy. Soc. London, Ser. 332 B, Biol. Sci. 270, 517–522 (2003). 333 13. Taylor, C.H., et al. Mapping the adaptive landscape of Bates ian mimicry using 3D -printed 334 stimuli. Nature. 644, 706–713 (2025). 335 14. Jamie, G.A., et al. Multimodal mimicry of hosts in a radiation of parasitic finches. Evolution. 336 74, 2526–2538 (2020). 337 15. Igic, B., et al. A shared chemical basis of avian host–parasite egg colour mimicry. Proc. Roy. 338 Soc. B, Biol. Sci. 279, 1068–1076 (2012). 339 16. Bartlow, A.W., Agosta , S.J. Phoresy in animals: Review and synthesis of a common but 340 understudied mode of dispersal. Biol. Rev. 96, 223–246 (2021). 341 17. Hafernik, J., Saul -Gershenz, L. Beetle larvae cooperate to mimic bees. Nature. 405, 35–36 342 (2000). 343 18. von Beeren, C., Tishechkin, A.K. Nymphister kronaueri von Beeren & Tishechkin sp. nov., an 344 army ant-associated beetle species (Coleoptera: Histeridae: Haeteriinae) with an exceptional 345 mechanism of phoresy. BMC Zool. 2, 3 (2017). 346 19. Warburg, S., Zvik, Y., Gavish -Regev, E. Hitching a ride on a scorpion: the first record of 347 phoresy of a myrmecophile pseudoscorpion on a myrmecophile scorpion. Arachnol. Lett. 66, 348 34–37 (2023). 349 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 20 20. Lückmann, J., Niehuis, M. Die Ölkäfer in Rheinland -Pfalz und im Saarland: Verbreitung, 350 Phänologie, Ökologie, Situation und Schutz (Gesellschaft für Naturschutz und Ornithologi e 351 Rheinland-Pfalz e. V., 2009). 352 21. Pinto, J.D., Westcott, R.L., Stouthamer, R., Rugman-Jones, P.F. Phoretic relationships of the 353 Blister beetle Meloe (Meloe) strigulosus Mannerheim (Coleoptera: Meloidae) from a coastal 354 dune habitat in Oregon. Trans. Am. Entomol. Soc. 146, 549–576 (2020). 355 22. Saul-Gershenz, L., Millar, J.G., McElfresh, J.S., Williams, N.M. Deceptive signals and 356 behaviors of a cleptoparasitic beetle show local adaptation to different host bee species. Proc. 357 Natl. Acad. Sci. 115, 9756–9760 (2018). 358 23. López-Estrada, E.K., et al. Mitogenomics and hidden -trait models reveal the role of phoresy 359 and host shifts in the diversification of parasitoid blister beetles (Coleoptera: Meloidae). Mol. 360 Ecol. 31, 2453–2474 (2022). 361 24. Clausen, C.P. Phoresy among entomophagous insects. Annu. Rev. Entomol. 21, 343 –368 362 (1976). 363 25. Klausnitzer, B. Beobachtungen zur Lebensweise von Meloe proscarabaeus Linnaeus, 1758 364 (Coleoptera: Meloidae). Gredleriana. 5, 209–216 (2005). 365 26. Knudsen, J.T., Eriksson, R., Gershenzon, J., Ståhl, B. Diversity and distribution of floral scent. 366 Bot Rev. 72, 1–120 (2006). 367 27. Dötterl, S., Gershenzon, J. Chemistry, biosynthesis and biology of floral volatiles: Roles in 368 pollination and other functions. Nat. Prod. Rep. 40, 1901–1937 (2023). 369 28. Naef, A., Roy, B.A., Kaiser, R., Honegger, R. Insect -mediated reproduction of systemic 370 infections by Puccinia arrhenatheri on Berberis vulgaris. New Phytol. 154, 717–730 (2002). 371 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 21 29. Radulović, N.S., Đorđević, A.S., Zlatković, B.K., Palić, R.M. GC-MS analyses of flower ether 372 extracts of Prunus domestica L. and Prunus padus L. (Rosaceae). Chem. Pap. 63, 377–384 373 (2009). 374 30. Füssel, U., Dötterl, S., Jürgens, A., Aas., G. Inter- and intraspecific variation in floral scent in 375 the genus Salix and its implication for pollination. J. Chem. Ecol. 33, 749–765 (2007). 376 31. Tollsten, L., Knudsen, J.T. Floral scent in dioecious Salix (Salicaceae)—a cue determining the 377 pollination system? Pl. Syst. Evol. 182, 229–237 (1992). 378 32. Borg-Karlson, A-K., et al. (S)-(+)-Linalool, a mate attractant pheromone component in the bee 379 Colletes cunicularius. J. Chem. Ecol. 29, 1–14 (2003). 380 33. Li, S., Guo, J., Li, H., Hao, D. Involvement of a novel cytochrome P450 CYP6HX3 from a 381 specialist herbivore, Pagiophloeus tsushimanus, in the metabolism of host -plant terpenoids. 382 Pestic. Biochem. Physiol. 210, 106366 (2025). 383 34. Dötterl, S., et al. Linalool and lilac aldehyde/alcohol in f lower scents: Electrophysiological 384 detection of lilac aldehyde stereoisomers by a moth. J. Chromatogr. A. 1113, 231–238 (2006). 385 35. Bologna, M.A. Fauna d'Italia: Coleoptera Meloidae, 28 (Calderini, 1991). 386 36. Radoi, I., et al . First report on the effect of Meloe spp. larvae invasion on Apis mellifera 387 carpathica bees in some apiaries in Romania. 2020, 1776–1780. 388 37. Ostap-Chec, M., Kierat , J., Kuszewska , K., Woyciechowski , M. Red mason bee ( Osmia 389 bicornis) thermal preferences for nest sites and their effects on offspring survival. Apidologie. 390 52, 707–719 (2021). 391 38. Coppée, A., et al . Age -dependent attractivity of males’ sexual pheromones in Bombus 392 terrestris (L.) [Hymenoptera, Apidae]. Chemoecology. 21, 75–82 (2011). 393 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint 22 39. Brockmann, A., Dietz, D., Spaethe, J., Tautz, J. Beyond 9 -ODA: Sex pheromone 394 communication in the European honey bee Apis mellifera L. J. Chem. Ecol. 32, 657 –667 395 (2006). 396 40. Bergström, G., Tengö, J. Linalool in mandibular gland secretion of Colletes bees 397 (Hymenoptera: Apoidea). J. Chem. Ecol. 4, 437–449 (1978). 398 41. Sánchez-Vialas, A., et al. Phylogeny of Meloini blister beetles (Coleoptera, Meloidae) and 399 patterns of island colonization in the Western Palaearctic. Zool. Scr. 50, 358–375 (2021). 400 42. Jamie, G.A., et al. The past and future of mimicry research. Nat. Ecol. Evol. 9, 1081–1085 401 (2025). 402 .CC-BY 4.0 International licenseavailable under a (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 The copyright holder for this preprintthis version posted January 15, 2026. ; https://doi.org/10.64898/2026.01.15.699641doi: bioRxiv preprint