(3S,6E)-Nerolidol-mediated rendezvous of Cyclocephalini beetles, Cyclocephala paraguayensis, in bottle gourd flowers

Cyclocephalini beetles of genus Cyclocephala (Coleoptera: Melolonthidae: Dynastinae) use flowers of some plants as food, shelter, and mating stands. However, little is known about floral scent chemistry involved in this interaction. Here we show that a sesquiterpene alcohol mediates attraction of Cyclocephala paraguayensis Arrow on bottle gourd flowers, Lagenaria siceraria (Curcubitaceae). Both males and females started to aggregate on flowers at twilight; after that, mating began and remained for the entire night. The major constituent of the airborne volatiles from L. siceraria was fully characterized as (3S,6E)-nerolidol, which elicited electroantennographic responses on male and female antennae. Field bioassays showed that traps baited with the natural stereoisomer or a mix of nerolidol isomers captured significantly more males and females of C. paraguayensis than control traps. Analysis from the gut content of these Cyclocephalini beetles showed the presence of pollen, suggesting that they also use bottle gourd flowers for their nourishment. Taken together, these results suggest that (3S,6E)-nerolidol plays an essential role in the reproductive behavior of C. paraguayensis by eliciting aggregation, mating, and feeding.


Introduction
Floral scents have been reported as important assets for the visitation of beetles of the tribe Cyclocephalini (Melolonthidae: Dynastinae) [1][2][3][4]. The interaction of beetles and flowers may result in mutualistic pollination typically by the cantharophilous floral syndrome of basal angiosperms, whose flowers are usually bisexual, protogynous, very scented, and thermogenic [4,5]. The Cyclocephalini beetles are primarily attracted by strong odors released by flowers as chemical cues to indicate a rendezvous site for food, shelter, and mating stands [4,6,7].
Cyclocephala paraguayensis Arrow, is a Neotropical species that is frequently found on eudicot plant species [14,15]. Here, we observed this beetle visiting flowers of Lagenaria siceraria (Molina) Standle (Curcubitaceae), which incidentally represents a new host flower record for the beetle family Melolonthidae. This plant is originated from Africa and nowadays is spread worldwide [16]; it produces white flowers with separate sexes in the same plant (monoecious) but can assume andromonoecious form, when hermaphrodite flowers are present [17]. Anthesis occurs throughout the night, at which time the lepidopteran pollinators, specially hawkmoths, visit the flowers [18].
We surmised that the volatile components of bottle gourd flowers mediate aggregative behavior of C. paraguayensis for feeding and mating. We studied the floral scent chemistry of L. siceraria flowers and showed that (3S,6E)-nerolidol plays an important role in the reproductive behavior of C. paraguayensis by eliciting aggregation, mating, and feeding.

Sexual and feeding behavior of insects
The behavior of C. paraguayensis visitation on L. siceraria flowers was observed under field conditions at night for four hours (7:00 p.m to 11:00 p.m. Seventy-five percent of bottle gourd flowers harbored more than one mating couple, thus, suggesting that aggregation on bottle gourd flowers created mating opportunities for the beetles. These Cyclocephalini beetles usually remained inside the floral receptacles until the next morning (Fig 1A-B). Bottle gourd flowers presented gnawing marks or even major damages resulted from beetles feeding (Fig 1E-F). Of note, pollens of these flowers were found in the gut of beetles (80% females; 50% males, n = 10) when dissected (Fig 1G-H).

Floral scent identification
Gas chromatography linked to electroantennographic detection (GC-EAD) analysis of airborne volatiles collected from L. siceraria showed the presence of a prominent, antennally active peak at 20.5 min (Fig 2). The prominent compound was tentatively identified by GC-MS as a nerolidol isomer, which led us to conduct cis and trans identification followed by chiral analysis.
The EAD active compound was then identified as transnerolidol (i.e., (6E)-nerolidol, IR = 1568) by comparing the mass spectra and retention indexes of the natural product and authentic standards (Fig 3a-f). Lastly, chiral analysis using nerolidol standards led to the full characterization of the natural product as (3S,6E)-nerolidol (Fig 3g).

Field evaluation of synthetic nerolidol
In our first field trial, a total of six adults of Cyclocephalini beetle, C.
In the second field experiment, we evaluated the response of adults of C.
paraguayensis to nerolidol-baited traps in two different locations: Cassilândia, State of Mato Grosso do Sul and Piracicaba, State of São Paulo. In both locations, traps baited with nerolidol (a mix of isomers) captured significantly more beetles than control traps.
In Cassilândia, 280 adults of C. paraguayensis were captured by nerolidol-baited traps during the tests, with no beetles captured in control traps (two-tailed Exact Binomial test; P < 0.001; Fig 4A). Although total captures in Piracicaba (n= 32) were lower than in Cassilândia, nerolidol-baited traps captured significantly more beetles than control traps (two-tailed Exact Binomial test; P < 0.001). Interestingly, the sex ratio of captured beetles did not differ significantly from the 1:1 ratio in both Cassilândia (51.4% females;  In the third field tests, we compared captures of C. paraguayensis in traps baited with nerolidol (a mix of isomers), (3S,6E)-nerolidol, and control traps.
Throughout these experiments, a total of 1,146 beetles were captured in nerolidol traps (i.e., traps baited with either enantiopure nerolidol, or a mix of isomers). By contrast, only two individuals (males) were caught in control traps. Although treatment and control were significantly different (Q2,30 = 20.002; P < 0.001; Fig 4B), there was no statistical difference in catches by the mix of isomers or (3S,6E)-nerolidol ( Fig 4B). We, therefore, concluded that the other isomers were neither attractants nor inhibitors. Of note, the sex ratio of the captured beetles in the two nerolidol treatments were somewhat female-biased (mix of isomers = 54.4% female, 95% Clopper-Pearson exact confidence interval: 0.509-0.578; P = 0.012; and (3S,6E)-nerolidol = 58.4% female, 0.544-0.622; P < 0.0001). However, as we did not measure the sex ratio of the field populations while conducting bioassays, it is not possible to assert that traps captured significantly more female than male beetles. Our results suggest that (3S,6E)-nerolidol mediates aggregation of the Cyclocephalini beetle, C. paraguayensis, on bottle gourd flowers, L. siceraria. We showed that this sesquiterpene alcohol represents the major VOC emitted by the floral scents of L. siceraria, and field tests demonstrated that nerolidol-baited traps captured significantly more beetles than control traps. Surprisingly, field tests showed captures in traps baited with racemic nerolidol (a mix of four isomers) did not differ significantly from those in traps having the enantiopure nerolidol as lure. Thus, the three other nonnatural isomers of nerolidol are neither attractant nor deterrent.

Discussion
We, therefore, concluded that C. paraguayensis rendezvous in L. siceraria is mediated by nerolidol. There is no evidence in the literature indicating that chemical communication in C. paraguayensis involves sex pheromones. If it does, it is unlikely that they need a long-range aggregation-sex pheromone (sensu Cardé 2014) [19] as nerolidol mediates aggregation of both sexes. It is conceivable, however, that C. paraguayensis utilizes a hitherto unknown short-range, or contact sex pheromone for species and/or sex recognition.
In this study we showed that C. paraguayensis also uses bottle gourd flowers of L. siceraria for nourishment given that pollen grains were found in the gut epithelium of males and females. The energy-rich pollen grains may increase the fitness of these Cyclocephalini beetles for mating and oviposition [4]. Our findings do not provide evidence for a mutualistic insect-plant relationship, as previously suggested for other anthophilous cyclocephalines [10,12]. The floral biology traits of curcubits seem to be more suitable for pollination by bees [20], and flowers are particularly successful pollinated by hawkmoths [18]. Additionally, C. paraguayensis visit caused damage to L. siceraria flowers, although the effects of these damages on the development of fruits remain to be investigated.

Chemical analysis
The EAD active compound in the VOCs was identified by gas chromatography Academy of Sciences of the Czech Republic). All compounds solution in hexane (40 ppm) were injected and detected at 250 ºC. Initial oven temperature was 85 ºC for 1 min and then increased to 190 ºC at 2 ºC/min. For both chiral identification and field assays, we used the (E)-nerolidol from Sigma-Aldrich, which was previously determined to be comprised of almost 100% (3S,6E)-nerolidol [21], and also confirmed in our GC analysis (Fig 3d).

Field bioassays of synthetic flower volatiles
We conducted three independent field trials. In the first tests, we mimicked the

Statistical Analysis
Data from experiments 1 and 2 were analyzed with the two-tailed Exact Binomial test to verify whether the ratio of beetles attracted to nerolidol and control differed significantly from the expected 1:1 ratio at 5% probability [22]. The test was performed from a spreadsheet provided by http://www.biostathandbook.com/exactgof.html (accessed January 17, 2019).
Differences between treatment means from experiment 3 were analyzed using the nonparametric Friedman's Test (PROC FREQ, option CMH; SAS Institute, 2011), because data did not pass the assumptions of ANOVA [23]. Replicates were defined by block and collection dates. Replicates with no capture of C. paraguayensis in any trap were excluded from analyses. Once an overall significance was detected by the Friedman's test, pairs of means were compared using the REGWQ multiple range test, which controls for type I experiment-wise error rate [24]. For all bioassays, the sex ratio of attracted adults of C. paraguayensis was compared to a nominal proportion of 0.5 with 95% Clopper-Pearson Exact Confidence Intervals [25].