Introduction
The evolutionary success of insects in terrestrial environments is indisputably related to their ability to conserve water (Garwood and Edgecombe, Reference Garwood and Edgecombe2011). One of the factors that contribute to this physiological trait is the presence of a cuticle that protects them against desiccation. The insect cuticle is a structure composed of several layers, among which the most efficient against water loss is the lipid layer associated with the epicuticle (Hadley, Reference Hadley1980). It might contain saturated or unsaturated hydrocarbons, branched or not, free fatty acids, aldehydes, alcohols, esters, ketones, glycerides, and sterols (Blomquist, Reference Blomquist, Blomquist and Bagnères2010).
Among the cuticular lipids, hydrocarbons deserve special attention because they are often present in high concentrations and play important roles in intra- and interspecific communication in insects (Lockey, Reference Lockey1991; Singer, Reference Singer1998; Howard and Blomquist, Reference Howard and Blomquist2005). Cuticular hydrocarbons (CHCs) in insects are present as a complex mixture of molecules, often characteristic of each species (Bagnères and Wicker-Thomas, Reference Bagnères, Wicker-Thomas, Blomquist and Bagnères2010). Within the same species, the CHC composition may vary between sexes (Steiger and Stökl, Reference Steiger and Stökl2014) and in relation to different life cycle stages (Lockey, Reference Lockey1991; Howard and Blomquist, Reference Howard and Blomquist2005; Thomas and Simmons, Reference Thomas and Simmons2008). Other factors such as the insects' diet (Liang and Silverman, Reference Liang and Silverman2000; Ingleby, Reference Ingleby2015), reproductive stage (Scott et al., Reference Scott, Madjid and Orians2008; Caselli et al., Reference Caselli, Favaro, Petacchi, Valicenti and Angeli2023), different populations of the same species (Hirai et al., Reference Hirai, Akino, Wakamura and Arakaki2008), and different colonies of social insects (Jutsum et al., Reference Jutsum, Saunders and Cherrett1979; Morel et al., Reference Morel, Vander Meer and Lavine1988; Dahbi et al., Reference Dahbi, Cerdá, Hefetz and Lenoir1996) as well as different castes in the same colony (Wagner et al., Reference Wagner, Brown, Broun, Cuevas, Moses, Chao and Gordon1998; Greene and Gordon, Reference Greene and Gordon2003) can alter the CHC composition both quanti- and qualitatively.
Regarding their structure, CHCs have carbon chains ranging from 11 to 43 atoms, both even and odd numbered. They are distributed into three main classes: n-alkanes, olefins, and methyl-branched alkanes. Olefins and methyl-branched alkanes usually occur as isomeric mixtures (Lockey, Reference Lockey1991; Blomquist, Reference Blomquist, Blomquist and Bagnères2010). The most common olefins in insects are alkenes, but compounds with two or, less frequently, more unsaturations may also be present. The position of the double-bonds in the carbon chain is of key relevance to their biological activities and is usually experimentally determined through derivatisation of the unsaturated compounds, followed by analysis by GC-MS (Lockey, Reference Lockey1988; Blomquist, Reference Blomquist, Blomquist and Bagnères2010).
Due to their taxon-specific character, the chemical profiles of insects' cuticles can be used to assist in species identification along with other more conventional taxonomic approaches (i.e., morphological characters and molecular biology) (Kather and Martin, Reference Kather and Martin2012). In fact, their efficiency as a chemotaxonomic tool in insect research has been demonstrated in identifying cryptic species (Soon et al., Reference Soon, Castillo-Cajas, Johansson, Paukkunen, Rosa and Ødegaard2021) and initial trends in reproductive isolation among populations (Hartke et al., Reference Hartke, Sprenger, Sahm, Winterberg, Orivel, Baur, Beuerle, Schmitt, Feldmeyer and Menzel2019).
Cyclocephaline beetles (Melolonthidae, Cyclocephalini) comprise a diverse group of essentially Neotropical Scarabaeoidea, currently encompassing ~500 spp. (Moore et al., Reference Moore, Cave and Branham2018a). Some have rhizophagous larvae, commonly referred to as white grubs, which are widely known as pests in crops, lawns, and reforestation areas (Moore et al., Reference Moore, Cave and Branham2018b). The adults of nine of the 14 currently acknowledged genera within the tribe are highly specialised pollinators and florivores that are attracted to their hosts by floral fragrances (Maia et al., Reference Maia, Dötterl, Kaiser, Silberbauer-Gottsberger, Teichert, Gibernau, do Amaral Ferraz Navarro, Schlindwein and Gottsberger2012, Reference Maia, Santos, Gonçalves, Navarro and Nuñez-Avellaneda2018; Moore et al., Reference Moore, Cave and Branham2018b). The functional details behind specific sexual recognition in these insects have only been superficially assessed (Nóbrega et al., Reference Nóbrega, Maia, Lima, Felix, Souza and Pontes2022), but it is generally agreed that close-range signals (i.e., visual, tactile, and/or chemical) might be involved.
The taxonomy of cyclocephaline beetles at the species level is primarily dependent on male genitalia characters (Endrödi, Reference Endrödi1985; Moore et al., Reference Moore, Cave and Branham2018b). Therefore, it requires precise dissection, assembly, and analysis of anatomical structures, making classification challenging or even unfeasible if only females are available. Various species of the most speciose genus in the tribe, Cyclocephala (> 350 spp.), display multiple levels of polymorphism (Ratcliffe et al., Reference Ratcliffe, Cave and Cano2013, Reference Ratcliffe, Cave and Paucar-Cabrera2020), whereas others form morphologically indistinguishable complexes with cryptic species (Neita-Moreno, Reference Neita-Moreno2021). While an integrative DNA barcoding framework for the tribe Cyclocephalini has not yet been examined (Moore et al., Reference Moore, Cave and Branham2018b; Ratcliffe et al., Reference Ratcliffe, Cave and Paucar-Cabrera2020), alternative accessible tools for species differentiation could advance research not only in taxonomy but also in ecology and crop management (Lyal et al., Reference Lyal, Kirk, Smith and Smith2008; Lagomarsino and Frost, Reference Lagomarsino and Frost2020).
In the present study, we investigated the cuticular chemical profiles of six species within the genus Cyclocephala and two distinct populations of one species within the genus Erioscelis. We aimed to address the following questions: (i) Is it possible to discriminate between species based on their cuticular chemical profiles?; (ii) Are there any sex-related differences in these profiles?
Materials and methods
Investigated species and collection sites
Cyclocephala celata Dechambre, 1980: This species is found from the northeastern Atlantic Coast of Brazil across dry seasonal forests of South America to Bolivia and Paraguay (Endrödi, Reference Endrödi1985; Moore et al., Reference Moore, Cave and Branham2018a). Adults are medium-sized, measuring about 15–17 mm in length. Known pollinators of Annonaceae and Araceae (Maia et al., Reference Maia, Dötterl, Kaiser, Silberbauer-Gottsberger, Teichert, Gibernau, do Amaral Ferraz Navarro, Schlindwein and Gottsberger2012; Parizotto and Grossi, Reference Parizotto and Grossi2019). Specimens used in the present study were collected in inflorescences of Philodendron acutatum Schott (Araceae) between March and June 2014 in the municipality of Igarassu, state of Pernambuco (7°49′S, 35°02′W; ca. 110 m.a.s.l.) (refer to Maia et al., Reference Maia, Schlindwein, Navarro and Gibernau2010 for details).
Cyclocephala cearae Höhne, 1923: This species is endemic to northeastern Brazil where it is found from the coastal Atlantic Forest to the Caatinga (Endrödi, Reference Endrödi1985; Moore et al., Reference Moore, Cave and Branham2018a). Adults are medium-sized, measuring about 17–18 mm in length. Known pollinators of Araceae (Maia et al., Reference Maia, Gibernau, Carvalho, Gonçalves and Schlindwein2013). Specimens for the present study were collected in inflorescences of Taccarum ulei Eng. & K. Krause between March and June 2014 in the municipality of Igarassu, state of Pernambuco (7°47′S, 34°55′W; ca. 80 m.a.s.l.) (refer to Maia et al., Reference Maia, Gibernau, Carvalho, Gonçalves and Schlindwein2013 for details).
Cyclocephala paraguayensis Arrow, 1903: This species is widely distributed in South America, found north from the Colombian Orinoquia to the Chacoan Argentina and Uruguay in the south (Endrödi, Reference Endrödi1985; Moore et al., Reference Moore, Cave and Branham2018a). Adults are relatively small, measuring about 11–13 mm in length. Specialised florivores associated with cacti and cultivated plants (Moore and Jameson, Reference Moore and Jameson2013). Specimens used in the present study were collected with light traps installed in a sugarcane field between March and June 2014 in the municipality of Igarassu, state of Pernambuco (7°49′S, 35°02′W; ca. 110 m.a.s.l.) (refer to Maia et al., Reference Maia, Schlindwein, Navarro and Gibernau2010 for details).
Cyclocephala distincta Burmeister, 1847: This species is widely distributed in South America, found north from Colombia and the Guianas to the southern Brazilian state of Santa Catarina (Endrödi, Reference Endrödi1985; Moore et al., Reference Moore, Cave and Branham2018a). Adults are small, measuring about 9–10 mm in length. Specialised florivores associated with palms (Moore and Jameson, Reference Moore and Jameson2013; Maia et al., Reference Maia, Santos, Gonçalves, Navarro and Nuñez-Avellaneda2018). Specimens used in the present study were collected with scent-baited traps (e.g., 2-isopropyl-3-methoxypyrazine see Maia et al., Reference Maia, Santos, Gonçalves, Navarro and Nuñez-Avellaneda2018 for details) installed in a sugarcane field between December 2016 and March-April 2017 in the municipality of Igarassu, state of Pernambuco (7°49′S, 35°02′W; ca. 110 m.a.s.l.) (refer to Maia et al., Reference Maia, Schlindwein, Navarro and Gibernau2010 for details).
Cyclocephala forsteri Endrödi, 1963: This species is widely distributed in South America, found north from the Colombian Orinoquia to the southern Brazilian state of Santa Catarina (Moore et al., Reference Moore, Cave and Branham2018a). Immatures are rhizophages associated with soy crops (Santos and Ávila, Reference Santos and Ávila2007). Adults are large-sized beetles, measuring about 20–22 mm in length. Specialised florivores associated with palms (Oliveira and Ávila, Reference Oliveira and Ávila2011; Maia et al., Reference Maia, Reis, Navarro, Aristone, Colombo, Carreño‐Barrera, Núñez‐Avellaneda and Santos2020). Specimens used in the present study were collected with scent-baited traps (e.g., 4-methylanisole + 2-isopropyl-3-methoxypyrazine + 2-sec-butyl-3-methoxypyrazine) installed in a rural area ca. 30 km southeast from the municipality of Campo Grande, state of Mato Grosso do Sul (20°37′S 54°31′W; ca. 590 m.a.s.l.), between October and November 2015 (refer to Maia et al., Reference Maia, Reis, Navarro, Aristone, Colombo, Carreño‐Barrera, Núñez‐Avellaneda and Santos2020 for details).
Cyclocephala ohausiana Höhne, 1923: This species is native to the Cerrado savanna in central Brazil, as well as its transitional zone with the Atlantic Forest (Endrödi, Reference Endrödi1985; Moore et al., Reference Moore, Cave and Branham2018a). Adults are medium-sized beetles, measuring about 15–17 mm in length. Pollinators of Annonaceae (Moore and Jameson, Reference Moore and Jameson2013; Costa et al., Reference Costa, Silva, Paulino-Neto and Pereira2017). Specimens used in the present study were collected in flowers of Annona coriacea Mart. (Annonaceae) between October and November 2015 in a rural area ca. 30 km southeast from the municipality of Campo Grande, state of Mato Grosso do Sul (20°37′S 54°31′W; ca. 590 m.a.s.l.) (refer to Maia et al., Reference Maia, Reis, Navarro, Aristone, Colombo, Carreño‐Barrera, Núñez‐Avellaneda and Santos2020 for details).
Erioscelis emarginata (Mannerheim, 1829): This species is found from the coastal Atlantic Forest west to the Cerrado in Brazil, from the Chocoan Argentina to Ecuador. Adults are large-sized beetles, measuring about 21–24 mm in length. Known pollinators of Araceae (Moore and Jameson, Reference Moore and Jameson2013). Specimens used in the present study were collected in inflorescences of Philodendron bipinnatifidum Schott ex Endl. and P. mello-barretoanum R. Burle-Marx ex G.M. Barroso at two distinct locations in central Brazil. The first in the municipality of Taguatinga, Federal District, in December 2013 (15°51′S, 48°01′W; ca. 1120 m.a.s.l.). The second in the municipality of Nova Odessa, state of São Paulo, in October 2014 (22°46′S, 47°18′W; ca. 580 m.a.s.l) (refer to Barros et al., Reference Barros, Astúa, Grossi, Iannuzzi and Maia2020 for details). We considered the two collection sites as separate populations of E. emarginata, hereby referred to as ‘Taguatinga’ and ‘Nova Odessa’.
Upon collection the beetles were placed in semi-transparent plastic containers (30 × 21 × 21 cm) with perforated lids and filled with sieved and moistened pot soil. The containers were maintained at a temperature of 27 ± 2°C, relative humidity of 70 ± 5%, and a photoperiod of 14 hours of light and 10 hours of darkness for 24 to 96 hours before processing (as described below). The beetles were provided with fresh apple slices for food, which were available ad libitum.
Species identifications were conducted by Professor Paschoal Coelho Grossi, curator of the Entomological Collection at the Federal Rural University of Pernambuco (CERPE) and an expert in Dynastinae taxonomy. Voucher specimens for each species have also been deposited at CERPE.
Sampling and analysis of cuticular extracts
Female and male specimens of the seven investigated species, including two distinct populations of E. emarginata, were placed either individually or in batches of 10 (for C. distincta and C. paraguayensis) into clean 5 mL borosilicate glass vials (Sigma-Aldrich, USA). Rapid gradual freezing (−5 to −20°C; 60 min) was used to euthanise the beetles, which were then rinsed with deionised H2O to remove soil and food particles adhered to the cuticle surface. Subsequently, they were placed on clean absorbent paper to remove excess water, and dried using a stream of N 2. After drying, the specimens were put in clean borosilicate glass 5 mL vials, containing 1 ml (individual) or 2.5 ml (batch) of n-hexane (HPLC grade, bidistilled) and swirled gently for one minute. Preliminary results showed that a short-lasting extraction time minimises the interference from non-cuticular lipids, eliminating the need for a chromatographic column to purify the eluates.
The crude eluates were finally filtered using a borosilicate glass wool plug (Sigma-Aldrich, USA) packed inside a borosilicate glass Pasteur pipette into 2.0 ml (individual) or 4.0 ml (batch) chromatography vials. The extracted eluates from the batches of 10 individuals were additionally concentrated under a laminar N 2 flow to a final volume of 0.5 ml. All samples were stored under −24°C for 5 to 160 days until further processing.
The samples were analysed by gas chromatography coupled to mass spectrometry (GC-MS) using an Agilent 5975C Series GC/MSD quadrupole system (Agilent Technologies, Palo Alto, USA) equipped with an HP-5 ms non-polar column (Agilent J&W; 30 m × 0.25 mm i.d., 0.25 μm film thickness). An aliquot of 1.0 μl of each sample was injected in splitless mode. The GC oven temperature was set at 60°C for 1 min, then increased at a rate of 15°C min−1 up to 260°C, and kept for 20 min. The carrier gas (He) flow was kept constant at 1 ml min−1. MS source and quadrupole temperatures were set respectively at 230 and 150°C, and the mass spectra were recorded at 70 eV (EI mode) with a scanning speed of 0.5 scan/s from m/z 35 to 550. The compounds were identified by comparing their mass spectra and retention times to those of authentic standards available in the MassFinder 4, NIST20, Wiley Registry™ 9th Edition, and MACE_R4 (Schulz and Möllerke, Reference Schulz and Möllerke2022) reference libraries, incorporated in the software Agilent MSD Productivity ChemStation. Identification of double-bond positions in 7- and 9-alkenes was achieved through sample derivatisation (Francis and Veland, Reference Francis and Veland1987). The peak areas in the chromatograms were integrated to obtain the total ionic signal, and their values were used to determine the relative proportions of each compound. A homologous series of linear alkanes (C 9 – C 40) was used to determine linear retention indices (RIs) (van Den Dool and Kratz, Reference van Den Dool and Kratz1963) and provide a means for comparison with data available in the literature. Unassigned compounds corresponding to < 1% of the total individual peak area in at least one of the analysed samples were pooled.
We used the main compounds for comparative statistical analyses (see below), selected through an arbitrary threshold of ≥ 2% relative concentration (total individual peak area) in at least one of the analysed samples.
Statistical analyses
To assess differences in the cuticular lipid profiles of the seven investigated species, we first subjected the compositional chemical data (relative proportion of compounds in the samples) to a centred log ratio (clr) transformation with a 0.001 threshold to minimise misinterpretations caused by a constrained data set (as reviewed by Brückner and Heethoff, Reference Brückner and Heethoff2017). We then used the clr-transformed data to generate a two-way paired group (UPGMA) hierarchical clustering and heatmap using the Euclidean similarity index. This approach allowed us to track the influence of individual chemical compounds on the species' dendrogram.
We conducted a global permutational multivariate analyses of variance (PERMANOVA; Pseudo-F value) for the discrimination of groups/clusters recovered in the two-way UPGMA and a permutational analysis of multivariate dispersions (PERMDISP; F value) to elucidate the dispersion of the PERMANOVA.
Additionally, we ran pairwise PERMANOVA tests with the raw compositional chemical data (Bray-Curtis dissimilarity index) to assess differences between sexes (‘female’ vs ‘male’, except for C. paraguayensis), and populations (‘Taguatinga’ vs ‘Nova Odessa’, for E. emarginata only). We also conducted similarity percentages tests (SIMPER) with the raw compositional chemical data to determine the contribution of individual compounds to differences in chemical profiles among species, as well as between genera, sexes, and populations, emphasising quantitative trends. All statistical analyses were performed using PAST v.4.11 (Hammer et al., Reference Hammer, Harper and Ryan2001). Figures and graphs were edited using Adobe Illustrator 2023 (Adobe Systems Inc.).
Results
We identified 74 compounds in the solvent cuticular extracts of adult Cyclocephala spp. (n = 50 samples, six spp.) and E. emarginata (n = 51 samples, two populations). Linear alkanes (C 21 to C 34), particularly odd-chain C 21 – C 29, and unsaturated hydrocarbons, especially (Z)-7 and (Z)-9 odd-chain C 21 – C 35 alkenes, were the most well-represented cuticular lipids in all analysed samples, ranging from 94.7 to 33.2% and 66.2 to 5.1% relative composition, respectively. Fatty acyls (mostly acids and esters) and methyl-branched alkanes corresponded from null to 12.0 and 10.2%, respectively (table 1).
Average relative amounts (%) of cuticular lipids ordered per chemical class in hexane extracts of female (♀) and male (♂) adult specimens of Cyclocephala spp. and Erioscelis emarginata (Melolonthidae. Dynastinae. Cyclocephalini).
Individuals from two populations were analysed for E. emarginata (see Materials and methods for details). Linear retention indices (RIs) were calculated based on a homologous series of linear alkanes (C9 – C40) (van Den Dool and Kratz, Reference van Den Dool and Kratz1963) and provide a means for comparison with data available in the literature. The identities of (Z)-7 and (Z)-9-alkenes were confirmed by DMDS derivatisation and comparison with RI data in mass spectral libraries (see Materials and methods for details). Compounds in bold lettering were identified in > 5% average relative concentration in at least one species and or population. The 10 compounds marked with an asterisk (*) are the main contributors to overall interspecific dissimilarity (see Results for details). Only compounds accounting for at least 1% average amount in any species are presented. Unassigned and identified compounds below this threshold were pooled as ‘Other compounds’ and their total relative percentages added. Also refer to Table Data S1 for more details.
The cuticular chemical profiles significantly differed among the seven investigated species, as evidenced by PERMANOVA (Pseudo-F = 70.04, P < 0.0001; pairwise Pseudo-F ≥ 13.83, P < 0.01) and PERMDISP (F = 67.12, P < 0.0001; pairwise F ≥ 18.31, P < 0.01). Overall, the qualitative and quantitative differences in the relative concentrations of ten compounds alone (four alkanes, six (Z)-alkenes) explained 85.6% of the dissimilarity among species.
The only species in which the cuticular chemical profiles significantly differed between sexes was C. ohausiana (Pseudo-F = 64.52, P < 0.0001), with an overall average dissimilarity of 65.2% that was mostly explained by differential occurrence and relative proportions of C 21 and C 23 (Z)-alkenes and linear alkanes, as well as a heneicosadiene of unknown structure (RI 2072) (refer to table 1 for details).
In the UPGMA dendrogram and heatmap, we can visually observe a distinct separation of all examined taxa into discrete clades (fig. 1). Notably, we identified two sister clades of E. emarginata within a larger group that encompasses all Cyclocephala species. It is worth highlighting that these two E. emarginata clades did not correspond to the ‘Nova Odessa’ and ‘Taguatinga’ populations we investigated, as the largest clade included samples from both populations (fig. 1). Nevertheless, a pairwise PERMANOVA test provided statistical support for the differentiation between these two populations (Pseudo-F = 61.14, P < 0.0001).
Paired group (UPGMA) two-way dendrogram and heat map of the cuticular chemical profiles of Cyclocephala spp. and Erioscelis emarginata (Melolonthidae, Dynastinae, Cyclocephalini) using centred log ratio (clr)-transformed compositional chemical data (relative proportion of compounds in the samples). Each column correspond to one sample (see Materials and methods for details) and each row to a cuticular lipid identified by GC-MS analysis. Arrows mark the rows of the compounds that contributed most to clade formation in a similarity percentages test (SIMPER) with the raw compositional chemical data. Only compounds accounting for at least 2% average amount in any species (35 in total) were used for the analysis (see Materials and methods for details).
Discussion
The cuticular lipid profiles of the seven Cyclocephalini species investigated in our study exhibited high relative concentrations of odd, very long-chain (>C 20; VLC) n-alkanes and (Z)-alkenes. These compounds are widely found in the epicuticles of various insect groups, including Scarabaeoid beetles (Blomquist, Reference Blomquist, Blomquist and Bagnères2010). Interestingly, the identities and relative proportions of these compounds were highly effective in explaining the well-supported species groupings observed in our study. In fact, a mere ten compounds (out of 77 identified in total) accounted for over 85% of the dissimilarity among species.
In a chemotaxonomic analysis of Mediterranean earth-boring dung beetles (Geotrupidae), Niogret et al. (Reference Niogret, Felix, Nicot and Lumaret2019) identified the presence of n-alkanes and (Z)-alkenes in the cuticular lipid profiles of all 12 species studied. The relative concentrations of these compounds varied, ranging from 4.8 to 17.3% for n-alkanes and 0.9 to 14.0% for (Z)-alkenes. Similar findings were reported by Fletcher et al. (Reference Fletcher, Allsopp, McGrath, Chow, Gallagher, Hull, Cribb, Moore and Kitching2008) in a study involving six Australian species of Scarabaeidae, where all species except one exhibited the presence of n-alkanes or (Z)-alkenes in their cuticular profiles. However, unlike our study, both investigations found that methyl-branched alkanes were also significantly abundant, accounting for anywhere between 12 and 83% of the relative concentration of cuticular lipids.
Although VLC-alkanes are commonly found in the epicuticles of insects (Blomquist, Reference Blomquist, Blomquist and Bagnères2010) and have been identified as pheromonal components in various taxa (El-Sayed, Reference El-Sayed2023), their precise role in chemical communication among beetles remains poorly understood (Howard and Blomquist, Reference Howard and Blomquist2005). It is widely recognised that the molecular discriminative features of VLC-alkanes are relatively limited, being primarily based on carbon-chain length (van Zweden and d'Ettorre, Reference van Zweden, d'Ettorre, Blomquist and Bagnères2010). The variation in VLC-alkane proportions observed across different insect species, populations, sexes, and castes, nonetheless, provides support for their influence on behavioural modulation (van Zweden and d'Ettorre, Reference van Zweden, d'Ettorre, Blomquist and Bagnères2010).
On the other hand, (Z)-alkenes and methyl-branched alkanes possess greater potential as informative semiochemicals based on their molecular structure. This is attributed to factors such as the position of the methyl group or double-bond, and stereoisomerism (van Zweden and d'Ettorre, Reference van Zweden, d'Ettorre, Blomquist and Bagnères2010). In the case of beetles, (Z)-7 and (Z)-9-alkenes with odd-chain lengths ranging from C 23 to C 29 have been found to elicit contact-based sexually-oriented behaviour in various species of Cerambycidae (as reviewed by Ginzel, Reference Ginzel, Blomquist and Bagnères2010). Therefore, it is plausible that the presence of these compounds in the cuticular lipid profiles of Scarabaeiodea species could serve similar functions. The same can be said for methyl-branched alkanes, which can act as highly specific constituents of sex or aggregation pheromones due to their variability and chiral nature (van Zweden and d'Ettorre, Reference van Zweden, d'Ettorre, Blomquist and Bagnères2010). While our current findings are derived from a limited sampling effort, we propose further exploration of the potential involvement of unsaturated and methyl-branched alkanes in contact-based chemical communication among cyclocephaline beetles. In this context, it is also worth highlighting the alkadienes present in significant relative proportions in the cuticular lipid profiles of two of the studied species, namely C. ohausiana and C. paraguayensis.
We were surprised to find that sex-related differences were only evident in one of the taxa investigated in our study. This finding is intriguing considering the highly aggregative behaviour observed in flower-loving species of cyclocephaline beetles, often comprised by hundreds (Gottsberger and Silberbauer-Gottsberger, Reference Gottsberger and Silberbauer-Gottsberger1991; Scariot et al., Reference Scariot, Lleras and Hay1991), even thousands (Henderson, Reference Henderson1986; Núñez, Reference Núñez Avellaneda2014), of individuals. It is also worth noting that these aggregations regularly consist of multiple species coexisting (Beach, Reference Beach1982; Young, Reference Young1986; Maia et al., Reference Maia, Gibernau, Carvalho, Gonçalves and Schlindwein2013; Costa et al., Reference Costa, Silva, Paulino-Neto and Pereira2017), a context that has led to speculation about the potential role of contact or short-range chemical communication in sex recognition among these beetles (Nóbrega et al., Reference Nóbrega, Maia, Lima, Felix, Souza and Pontes2022). Such communication mechanisms could serve to prevent both same-sex pairing and interspecific copulation.
Indirect evidence of intraspecific contact chemical communication in cyclocephaline beetles has been obtained through controlled rearing experiments involving various species of Cyclocephala (Rodrigues et al., Reference Rodrigues, Nogueira, Echeverria and Oliveira2010; Nogueira et al., Reference Nogueira, Rodrigues and Tiago2013; Souza et al., Reference Souza, Maia, Schlindwein, Albuquerque and Iannuzzi2014). These experiments have demonstrated successful copulation and egg fertilisation occurring in small aggregations or even in one-on-one pairings, indicating at least some level of sex differentiation. Notably, Nóbrega et al. (Reference Nóbrega, Maia, Lima, Felix, Souza and Pontes2022) propose that male C. distincta are capable of distinguishing conspecific females upon contact, and that specific innate female traits trigger male sexual behaviour. This suggests that contact-based chemical cues play a crucial role in facilitating reproductive interactions within the species.
It is important to note that when examining trap collections and directly assessing insects within flowers, a consistent observation is that the female-to-male sex ratios of cyclocephaline beetles typically range from 1:1 to 2:1 (Beach, Reference Beach1982; Albuquerque et al., Reference Albuquerque, Grossi and Iannuzzi2016; Maia et al., Reference Maia, Reis, Navarro, Aristone, Colombo, Carreño‐Barrera, Núñez‐Avellaneda and Santos2020). This finding indicates that, on average, males within species-specific aggregations achieve a copulation success rate of approximately 50 to 75% from their random mating attempts.
It is likely that within the range of interactions between cyclocephaline beetles and their host plants, spanning from highly specialised (monophily) to more generalised (polyphily), species with a narrower preference (oligophilous) are increasingly dependent on floral scents to ensure their reproductive success. This scenario seems to be evident in diverse coastal forest ecosystems of French Guiana, where several co-flowering species of Araceae (such as Philodendron, Dieffenbachia, and Montrichardia) are each exclusively visited by a unique species or subset of species (Gibernau et al., Reference Gibernau, Barabé, Cerdan and Dejean1999, Reference Gibernau, Barabé, Labat, Cerdan and Dejean2003; Gibernau and Barabé, Reference Gibernau and Barabé2002; Gibernau, Reference Gibernau2015). This intricate interplay between specific floral hosts and distinct beetle species underscores the importance of chemical signals in facilitating successful reproductive interactions.
A contrasting scenario unfolds in the interactions involving large-flowered Annona spp., specifically A. coriacea, in the Cerrado savanna ecosystems of central Brazil. These nocturnally blooming flowers attract at least seven potentially syntopic species of Cyclocephala, including C. ohausiana (Costa et al., Reference Costa, Silva, Paulino-Neto and Pereira2017). Within the floral chambers of Annona spp., multiple congeneric species often aggregate together, creating a complex ecological milieu. This coexistence of closely related species may impose selective pressures that drive the development of more efficient mechanisms for contact-based sex discrimination. It is conceivable that these selective pressures contribute to the observed significant sex differences in the cuticular lipid profiles of C. ohausiana.
Our research findings underscore the significance of cuticular lipid profiles as robust tools for distinguishing among specific flower-loving cyclocephaline beetle species. These findings hold particular relevance in the identification of cryptic species groups, notably within the Cyclocephala genus, such as the black species documented by Neita-Moreno (Reference Neita-Moreno2021), as well as the ‘sexpunctata’ and ‘latericia’ species complexes (Moore, Reference Moore2011; Santos, Reference Santos2014), among others. Moreover, our study emphasises the imperative need for further research to establish appropriate comparative methodologies, enabling the investigation of variations in contact communication among cyclocephaline beetle species in diverse ecological conditions. Such research is pivotal in comprehending the spectrum of interactions involving cyclocephaline beetles and their floral hosts, ranging from monophily to polyphily. The chemotaxonomic characteristics of lipid profiles in cyclocephaline beetles, coupled with the potential involvement of CHCs in intraspecific chemical communication, highlight prospective applications in the realms of rapid taxonomic assignment and the targeted management of pest species. As previously illustrated by Potter and Haynes (Reference Potter and Haynes1993), non-polar solvent extracts derived from female southern masked chafers (Cyclocephala borealis Arrow, 1911) trigger sex-specific male behaviour. This intriguing phenomenon can be leveraged in strategies encompassing selective baiting, mass captures, or sexual disruption techniques.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0007485323000664
Acknowledgements
The authors express their gratitude to Paschoal Coelho Grossi for identifying the beetles. Additionally, we extend our thanks to Eduardo Gomes Gonçalves and Rafaella de Lucena Nóbrega for their assistance in collecting E. emarginata specimens in Taguatinga and C. distincta specimens in Igarassu, respectively. We would also like to acknowledge the invaluable support of Sílvio Columbelli and Wesley Nantes during fieldwork in Mato Grosso do Sul. ACDM is sincerely grateful for the hospitality extended by Harri Lorenzi, founder of the Plantarum Institute in Nova Odessa. ACDM and GKNS were partially supported by scholarships from the Coordenação de Aperfeiçoamento de Nível Superior (CAPES). DMAF acknowledges the financial support provided by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, 306351/2022-6) and FACEPE (APQ-0443-1.06/15). The author(s) received no direct funding to support the research, authorship, and/or publication of this article.
Competing interests
None.
