Exploring the canthariphilous species of the Tolfa Mountains (Latium, Central Italy): taxonomic diversity, phenology, and putative sources of cantharidin

Abstract Canthariphilous species are those arthropods attracted to cantharidin (CTD), a defensive compound produced by two beetle families (Meloidae and Oedemeridae). Although several species are known to be attracted to CTD, canthariphily was recently discovered in new species, suggesting that the list of canthariphilous species is still far from being complete. A systematic sampling focused to detect canthariphilous species has never been performed in Italy. The present research provides a list of seven canthariphilous species (Diptera: Ceratopogonidae, Anthomyiidae; Coleoptera: Anthicidae) from the Tolfa Mountains (Latium, Central Italy) resulting from a one-year sampling with CTD-baited and control traps. New species (Atrichopogon atriscapulus and A. tolfensis) were found to be attracted to CTD, and other species, already known as canthariphilous, were recorded for the first time in the Italian fauna (A. atriscapulus and A. meloesugans). A new scenario about the ecological significance of CTD in the sexual selection of canthariphilous species was speculated in A. meloesugans. Finally, a list of CTD-producing species occurring in the sample area was provided to suggest putative natural CTD sources.

The ecological significance of CTD in most of canthariphilous species is still unexplored or poorly documented.It has been proposed that some species might confuse CTD with analogous compounds driving aggregation or food searching (Dettner 1997;Tallamy et al. 1999;Hashimoto & Hayashi 2014).According to this hypothesis, the attraction to CTD might be merely accidental in some species, excluding an adaptive significance of this terpene in their ecology.Differently, other canthariphilous species can sequester CTD from producing beetles, and possibly use it as a deterrent against predators and parasites.Species able of sequestering CTD have been observed in some families of Coleoptera (Anthicidae, Cleridae, and Pyrochroidae) (Schütz & Dettner 1992;Frenzel & Dettner 1994;Holz et al. 1994;Eisner et al. 1996aEisner et al. , 1996b;;Molfini et al. 2022) and Diptera (Anthomyiidae and Ceratopogonidae) (Frenzel & Dettner 1994).CTD sequestering has been also speculated in other Coleoptera (Cantharidae, Cerambycidae, Chrysomelidae, and Melyridae) (Islami & Nikbakhtzadeh 2009) and in Hemiptera (Miridae) (Ramírez et al. 2021).In particular, some species of Anthicidae, Pyrochroidae and Ceratopogonidae can transfer the ingested CTD to the eggs during oviposition, implying a role of CTD in offspring defence (Schütz & Dettner 1992;Frenzel & Dettner 1994;Holz et al. 1994;Eisner et al. 1996aEisner et al. , 1996b)).
Although experimentally demonstrated in only a few species, it is generally assumed that males of several Anthicidae and Pyrochroidae have peculiar glands that allow them to present ingested CTD to females during courtship, thus being advantaged in sexual competition (Schütz & Dettner 1992;Holz et al. 1994;Eisner et al. 1996a).In some species of Anthicidae [especially in the tribe of Microhoriini (Anthicinae) and in the genus Notoxus Geoffroy, 1762 (Notoxinae)], these glands are localised in specialised elytral notches (also present in some genera of the basal blister beetle subfamily Eleticinae) (Abdullah 1965;Selander 1966;Bologna 1991;Kejval & Chandler 2020), while in Pyrochroidae these are localised within a peculiar cranial apparatus occurring in several genera of Pyrochroinae (e.g., Young 2019).
Some canthariphilous species are known to be distributed in the Italian peninsula, but a systematic sampling with CTD-baited traps to detect canthariphilous species has never been performed thus far in Italy.This work aims contributing to assess the taxonomic diversity and phenology of the Italian canthariphilous fauna through a one-year sampling in a natural area where the presence of CTD producers (blister beetles and false blister beetles) is well ascertained: the EU Special Protection Area "Comprensorio Tolfetano-Cerite-Manziate" (IT6030005) in the province of Rome (Tolfa Mountains, Tolfa, Rome, Latium, Italy).

Material and methods
Sampling method involved two pairs of funnel traps (Horiuchi et al. 2018) 1).
Each pair of traps consisted in one CTD-baited trap (0.5 ml of a 10 −2 M solution of synthetic CTD in acetone) and one control trap (0.5 ml of acetone) approximately 2 m apart (Hashimoto & Hayashi 2014).The distance between pairs at each site was about 30 m.In each site, one pair was placed at the ground level, while the other was suspended at about 1.5 m.Traps were active for 24 hours every two weeks from June 3 rd , 2020, to June 15 th , 2021, for a total of 24 surveys.After each sampling session, traps were capped with cotton wool and stored at −20°C to euthanize the sampled specimens.Specimens were then preserved in 70% ethanol for species identification.Binomial test implemented in R (R Core Team 2021) was used to test both the attraction to CTD of each taxon and differences in attraction between sexes.Only taxa with p-value < 0.05 or sampled more than three times in CTDbaited traps (without record in control traps) were identified at the species level.Difference in CTD attraction between sexes was tested only in species suspected to acquire CTD for offspring-defence and courtship (Hashimoto & Hayashi 2014).The material is deposited in the personal collections of the authors which identified the species (Nardi G., Anthicidae; Szadziewski R., Ceratopogonidae; Bologna M.A., Anthomyiidae and unidentified specimens).The diversity of canthariphilous species was described with the Shannon index H (Shannon & Weaver 1949) and the evenness index E (Pielou

548
M. Molfini et al. 1975) calculated as in Hashimoto and Hayashi (2014).Faunistic records of blister beetles and false blister beetles, obtained from over 40 years of entomological samplings on the Tolfa Mountains (deposited in the collection of Bologna M.A.), were used to compile a list of CTD-producing species that might serve as CTD sources for canthariphilous species in the area (Table I).

Results and discussion
Baited and control traps collected 2,659 and 55 specimens, respectively, with seven species significantly attracted to CTD-baited traps (binomial test p-value < 0.001) (see Supplementary Material: Table S1).
Canthariphilous species were identified in Diptera (Anthomyiidae and Ceratopogonidae) and Coleoptera (Anthicidae), as discussed below (Figure 2).The diversity (H = 1.05) and evenness (E = 0.37) indices, calculated from the pooled frequencies of all canthariphilous species across all months, were consistent with those observed in a previous study conducted in central Japan (1.09 and 0.36, respectively; Hashimoto & Hayashi 2014).Values of H and E for each sampling session are shown in Table I.

Diptera, Anthomyiidae
The 79% of specimens collected in CTD-baited traps were identified as belonging to the Anthomyia pluvialis complex (Diptera: Anthomyiidae) (Michelsen 1980) (N = 2,114) (Figure 2), evidencing a conspicuous attraction of this taxon to CTD (only seven individuals were collected in control traps).However, although A. pluvialis is well known to be attracted to CTD (e.g., Görnitz 1937;Dettner 1997;Hemp & Dettner 2001) and individuals were observed lapping the body surface of death blister beetles (Bologna & Havelka 1985), CTD seems not having a role in offspring-defence and the ecological significance of canthariphily in this species remains unknown (Frenzel & Dettner 1994;Dettner 1997).The abundance of A. pluvialis strongly influenced the diversity indices of the canthariphilous community, and values increased to H = 2.23 and E = 0.86 when the taxon was excluded from the analysis.
Adult females of some Atrichopogon spp.are known to feed on the haemolymph of both Meloidae and Oedemeridae, from which they sequester CTD (e.g., Wirth 1980;Frenzel & Dettner 1994;Szadziewski et al. 2007;Ciliberti et al. 2020;Hashimoto & Tateno 2022).Although males have reduced biting mouthparts, some evidence suggests that they can sequester CTD, but it is unclear how they obtain the compound (Frenzel & Dettner 1994).It has been proposed that males could intake CTD from the exuded haemolymph of blister beetles and false blister beetles (autohaemorrhaging is a defensive strategy adopted by several species of CTD producers; Fratini et al. 2021) or from the liquid faeces of female biting midges (Szadziewski & Elżbieta 2022).
As supposed for other species, females of A. meloesugans could not be particularly attracted to CTD itself, but other unknown signals, alone or in combination with CTD, might be involved in triggering the attraction to CTD-producing species (Dettner 1997;Molfini et al. 2022).Furthermore, the attraction of males to pure-CTD could suggest a mating preference for females that have ingested CTD from beetles.This might indicate a sexual selection of males towards such females.If confirmed, this hypothesis could represent a reversal of typical sex roles observed in other canthariphilous Figure 2. Total number of individuals collected for each species in CTD-baited traps during the one-year sampling on the Tolfa Mountains.Logarithmic scale (log 10 ) is used for individuals of Anthomyia pluvialis (Anthomyiidae, Diptera) due to the broad range of variation.Male/female ratio is given for species belonging to Ceratopogonidae (Diptera) and Anthicidae (Coleoptera).Sampling dates are coloured to highlight the seasons (pink, spring; green, summer; yellow, autumn; grey, winter).Binomial test was used to test the occurrence in CTD-baited traps vs. control traps and differences in attraction between males and females.X-axis, sampling date (24 hours); y-axis, number of individuals; M, males (blue); F, females (purple); NS, p-value > 0.05; asterisks (***), p-value < 0.001.insects, where sexual selection is driven by females towards males that have ingested CTD (see paragraph Coleoptera, Anthicidae) (Schütz & Dettner 1992;Holz et al. 1994;Eisner et al. 1996aEisner et al. , 1996b)).Additionally, it is possible that CTD acts as a kairomone, attracting males to CTD-producing species where females can also be found (Dettner 1997;Hashimoto & Hayashi 2016).
It is worth noting that interspecific interactions with CTD-producing beetles have never been documented in A. albiscapulus, A. atriscapulus and A. tolfensis, thus it is not possible to discern if these species confuse CTD with analogous compounds, or if this terpene plays an actual role in their ecology (Dettner 1997;Tallamy et al. 1999;Hashimoto & Hayashi 2014).
Adults of Ceratopogonidae were active for several months on the Tolfa Mountains (Figure 2), suggesting a wide spectrum of putative CTD-producing species as CTD source (Table I).In particular, adults of A. albiscapulus were active all over the year except for the late spring; A. atriscapulus was mainly collected during the winter season, with an activity period ranging from early autumn to early spring; A. tolfensis was collected during the summer-autumn period; and A. meloesugans showed its activity peak in the early spring, with two smaller peaks in both the early autumn and winter.The three peaks of activity observed in A. meloesugans might suggest interspecific interactions with several co-occurring Meloe species with different phenology (e.g., M. proscarabaeus Linnaeus, 1758, M. tuccia Rossi, 1792, and M. erythrocnemus Pallas, 1782 in the spring; M. autumnalis Olivier, 1792, M. mediterraneus Müller, 1925 in the autumn; M. mediterraneus Müller, 1925, M. ganglbaueri Apfelbeck, 1905, M. baudii Leoni, 1907 in the winter) (Bologna 1988(Bologna , 1991)).
This research also added two new species to the Italian fauna (A.atriscapulus and A. meloesugans), since only A. albiscapulus was already recorded in the Italian peninsula (Borkent et al. 2013) and A. tolfensis was newly described based on specimens from this sampling, currently representing an Italian endemite (Szadziewski et al. 2022).So far, in Europe, A. atriscapulus was exclusively reported in Poland and Lithuania (Szadziewski et al. 2007;Borkent et al. 2013) and A. meloesugans in Poland and Netherlands (Ciliberti et al. 2020), so this finding represents the third European record for both species.

Coleoptera, Anthicidae
Three species of Anthicidae [Microhoria fasciata fasciata (Chevrolat, 1834), M. terminata (Schmidt, 1842), and Anthelephila pedestris (Rossi, 1790)], were collected exclusively in CTD-baited traps (Figure 2).Only males of M. f. fasciata and M. terminata were significantly attracted (p < 0.001), while the number of A. pedestris individuals (only four females) was too low to statistically assess an attraction.Although in literature both sexes of A. pedestris have been found to be attracted to CTD (Schütz & Dettner 1992), we collected only few females without a statistical significance, suggesting a lower attraction in this species than in the two Microhoria Chevrolat, 1877 species.However, broadly, these results confirmed previous observations, indeed canthariphily is commonly present in Microhoriini (which includes Microhoria), with only males attracted to CTD, and rarely observed in Formicomini (which includes Anthelephila Hope, 1833) with both sexes attracted (Hemp & Dettner 2001;Kejval & Chandler 2020).
Noteworthy, males of Microhoriini have modified elytral apex that contain putative CTD glands showed to females during courtship (Schütz & Dettner 1992;Kejval & Chandler 2020).Since in this group only males were attracted, our results are coherent with the hypothesis that CTD acts as a selective agent that increases male mating success in M. f. fasciata and M. terminata (Schütz & Dettner 1992;Kejval & Chandler 2020).Contrarily, males of A. pedestris have simple elytral apex and lower attraction to CTD, suggesting that this terpene likely does not play a key role in sexual selection of this species.
Despite their attraction to CTD, it is still unclear from which CTD-producing species M. f. fasciata and M. terminata sequester this compound.As far as we know, only M. f. fasciata was reported feeding on the blister beetle Meloe violaceus (Marsham, 1802) (Bucciarelli 1976), while M. terminata was only observed attracted to CTD (Hemp & Dettner 2001).According to our sampling in Central Italy, adults of Microhoria species have different phenology (Figure 2) which might result in different natural sources of CTD (Table I).

Conclusions
This research represents the first systematic sampling specifically aimed at collecting and assessing the phenology of canthariphilous species in Italy.Our results were in line with previous findings, highlighting the presence of canthariphilous species in Anthomyiidae, Ceratopogonidae and Anthicidae and adding details on their phenology, with new records for the Italian fauna.
Much remains to be understood concerning the attraction to CTD of canthariphilous species, and why some of them seem not to be attracted to the pure compound (Molfini et al. 2022).Further faunistic samplings using additional traps to CTDbaited traps (e.g., traps baited with CTDproducing species or with CTD analogous produced by plants), together with behavioural and chemical analyses, could allow to better understand pattern of attraction of canthariphilous species, their hosts, the ecological role of CTD, and lead to the discovery of canthariphily in new taxa of arthropods.

Figure 1 .
Figure 1.Sampling sites on the Tolfa Mountains (Rome, Latium, Italy).Orange dots indicate the location of the four sites (A, B, C, D).Each yellow dot indicates a pair of funnel traps (CTD-baited trap and control trap), two pairs of traps for each site were set (about 30 m apart).All the images are north-oriented.Satellite images were taken from Google Earth Pro.

Table I .
Checklist of Meloidae and Oedemeridae (in alphabetical order) recorded from Tolfa Mountains and close area (Latium, Italy) with the roughly phenology of adults.Species are proposed as putative natural sources of CTD for canthariphilous species.Diversity (H) and evenness (E) indices for canthariphilous species are reported at the top of the table for each sampling date.Table is divided and coloured as in Figure2to facilitate comparison.