Collection of and field experiments with all cerambycid species described below, were conducted under SISBIO permit #46395-1 and #46395-2 from the Brazilian Ministry of the Environment. Adult A. distinctus and A. electus of both sexes (mating status unknown) were caught with panel traps (black corrugated plastic) constructed as described previously (Silva et al., 2016a,b). Internal surfaces of traps were coated with a 50% aqueous dilution of Fluon® (Insect-a-Slip, BioQuip Products Inc., Rancho Dominguez, CA, USA) to render traps more slippery and hence increase trapping efficiency (Graham et al., 2010). In order to collect live beetles, traps were modified by fitting 5-l plastic collection jars to the bottom of the traps, with holes punched in the jar bottoms for drainage. We hung traps from inverted L-shaped PVC pipe hangers with the trap bases ~50 cm above ground. Traps were baited with absolute ethanol (100 ml) dispensed from a clear low-density polyethylene press-seal sachet (10 × 14 cm, 80 μm wall thickness, Talge-Qualidade e Segurança, Balneário Camboriú, SC, Brazil). Lures were suspended in the open central slot of the panel traps. We deployed four traps on a farm located in Valentim Gentil, SP, Brazil (20°22′17.7″S, 50°04′46.6″W), which contains fragments of Cerrado or “Brazilian Savanna” (sensu Goodland, 1971) mixed with pastureland. Because we had captured adults of A. distinctus and A. electus at the same site with light traps in a previous study intended for other purposes, we inferred that these species were probably nocturnal. Therefore, traps were checked for beetles daily soon after sunrise from 20 September to 10 October 2015. Twelve adults (seven females and five males) of A. distinctus and nine adults (five females and four males) of A. electus were trapped. Beetles were sexed by the length of the antennae relative to the body; antennae are markedly longer than the length of the body in males but not females (Gahan and Arrow, 1903). The sexes were held separately in 0.2–l plastic containers and shipped overnight to the Laboratory of Chemical Ecology and Insect Behavior at the University of São Paulo, Piracicaba, SP, Brazil (~400 km distance). There, beetles were allowed at least 48 h to acclimate under laboratory conditions (23 ± 2°C, 60 ± 10% RH, ~5,000 lux light intensity, and light:dark regime of 12:12 h) prior to being used for collection of headspace volatiles. Beetles were provided with 10% sucrose solution as nourishment, in glass vials (7.5 cm long × 1 cm id) plugged with cotton rolls.

Headspace volatiles were collected from individual beetles, or from pairs of the same species and sex, in custom made 0.5–l cylindrical glass chambers. A vial containing 10% sucrose solution provided nutrition. Collectors consisted of glass pipettes (8.5 cm long × 0.5 cm id) containing 150 mg of 80/100 mesh HayeSep® Q (Supelco, Bellefonte, PA, USA), a generic adsorbent for trapping volatile organic compounds such as pheromones, held in place with glass wool plugs. The collector was fastened to the chamber outlet with a Teflon screw cap. Chambers were flushed with charcoal-filtered air at a flow rate of ~300 ml/min. Beetle volatiles were collected continuously for 48 h under the environmental conditions described above, and individual beetles of both sexes were aerated 1–3 times. Aerations of chambers with sugar feeders but without beetles were run in parallel as controls for system contaminants. Volatiles were rinsed from collectors into silanized glass vials using four successive 0.25-ml aliquots of methylene chloride collected into one vial. The resulting extracts were stored at −30°C until analysis.

Aeration extracts were first analyzed by gas chromatography-flame ionization detection (GC-FID) with a Shimadzu GC-2010 gas chromatograph (Shimadzu Corp., Kyoto, Japan) fitted with an HP5-MS capillary column (30 m × 0.25 mm i.d. × 0.25 μm film; Agilent Scientific, Santa Clara, CA, USA). One microliter aliquots were injected in splitless mode with an injector temperature of 250°C and helium as carrier gas with a linear velocity of 25 cm/sec, and with the purge valve opened after 1 min. The GC oven was programmed from 35°C for 1 min, increased to 40°C at 2°C/min (hold for 1 min), and then increased to 250°C at 10°C/min (hold for 10 min). Ratios of compounds were estimated from the integrated peak areas, and have not been corrected for the FID relative response factors. Extracts which contained sex-specific compounds were shipped to the University of California, Riverside, USA, where they were re-analyzed in splitless mode with an Agilent 7820A GC coupled to an Agilent 5977E mass selective detector (GC-MS), using an HP-5 column (same dimensions as above), and a temperature program of 40°C/1 min, 10°C/min to 280°C (hold for 10 min). Helium was used as carrier gas, with a linear velocity of 37 cm/s. An aliquot was also analyzed on a midpolarity DB-17 column (25 m × 0.20 mm i.d., 0.33 μm film; J&W Scientific, Folsom, CA, USA) using the same temperature program as above. Tentative identifications of compounds were confirmed by matching their retention times and mass spectra with those of authentic standards. The absolute configuration of the insect-produced 3-hydroxyhexan-2-one was determined by analyses of extract aliquots on a Hewlett-Packard 5890 GC fitted with a Cyclodex-B GC column (30 m × 0.25 mm, 0.25 μm film, J&W Scientific) programming from 50°C/1 min, 5°C/min to 200°C, and with injector and detector temperatures of 100 and 200°C, respectively. Configurations were confirmed by coinjection of extracts with authentic standards of the two enantiomers (Hanks and Millar, 2016).

In a follow-up experiment, one set of traps baited with racemic 3-hydroxyhexan-2-one, 1-(1H-pyrrol-2-yl)-1,2-propanedione, or a 2:1 blend of the two (see below for trap and lure details) were deployed at the experimental site on 22 October 2016 to capture more beetles alive for collection and analysis of headspace volatiles, as described above. Traps were checked for beetles daily for 2 weeks. In addition to A. distinctus and A. electus, these traps yielded at least five individuals of each sex of the cerambycine species Ambonus interrogationis (Blanchard), Amorupi fulvoterminata (Berg), Chrysoprasis aurigena (Germar), Itaclytus olivaceus (Laporte & Gory), Neoclytus pusillus (Laporte & Gory), Orthostoma abdominale (Gyllenhal), Sphaerion inerme White, Stizocera consobrina Gounelle, and Uncieburia nigricans (Gounelle) (Table 1). We analyzed the sex-specific volatiles produced by these species, except for A. fulvoterminata and S. consobrina, because none of the captured specimens of those species survived long enough to be used for headspace collections. For the remaining species, at least five aeration extracts were made and analyzed for each sex and species.

Voucher specimens of all species have been deposited in the collection of the museum in the Department of Entomology and Acarology, University of São Paulo, Piracicaba, SP, Brazil.

Racemic 3-hydroxyhexan-2-one (henceforth “C6-ketol”) was purchased from Bedoukian Research (Danforth, CT, USA), 1-(1H-pyrrol-2-yl)-1,2-propanedione (henceforth “pyrrole”) was synthesized as described by Zou et al. (2016), and 3-methylthiopropan-1-ol (henceforth “thiopropanol”) was purchased from Oakwood Chemicals (Estill, SC, USA).

We field-tested the reconstructed blends of male-specific volatiles (see Section Results) for each of the two Ambonus target species at the site where the beetles originally had been collected (see above) from 16 September to 28 October 2016. We used black flight-intercept panel traps coated with Fluon® as described above, but with trap basins filled with a few centimeters of water to which were added a few drops of detergent and ~25 g of NaCl to kill and preserve captured beetles. Lures consisted of resealable polyethylene sachets (Bagettes® model 14770, 5 × 7.5 cm, 50 μm wall thickness; Cousin Corp., Largo, FL, USA) containing a cotton dental wick. Each lure was loaded with 1 ml of an isopropanol solution of the synthesized pheromone candidates. Treatments (mg of compound/lure) were randomly assigned to traps, as follows: (1) C6-ketol (50 mg); (2) pyrrole (25 mg); (3) C6-ketol (50 mg) + pyrrole (25 mg); (4) C6-ketol (50 mg) + pyrrole (25 mg) + thiopropanol (4 mg); and (5) control (1 ml of neat isopropanol). Lures were replaced every 2 weeks. Doses and lure replacement times were chosen based on our empirical results from testing cerambycid aggregation-sex pheromones over the past decade using these sachets as release devices. Traps were placed ~15 m apart in four blocks spaced ~30 m apart, and each treatment was represented once in each block of traps. Traps were monitored every 2–3 d (total of 19 count dates), with treatments being rotated within blocks at each count to control for positional effects.

Differences between treatment means, blocked by location and date, were tested separately for each species using the nonparametric Friedman's Test (PROC FREQ, option CMH; SAS Institute, 2011) because data violated homoscedasticity assumptions of ANOVA (Sokal and Rohlf, 1995). Replicates were defined by block and by collection date. Replicate blocks in which no specimens of the species in question had been caught in any trap were dropped from analyses. In recognition of the multiple statistical tests of treatment effects (N = 11 analyses; see Section Results), significance levels were adjusted to α = 0.005 according to the Bonferroni procedure (Quinn and Keough, 2002). Where a significant overall Friedman's test was obtained, pairs of treatment means were compared using the nonparametric Dunn-Nemenyi multiple comparison test, which limits Type I errors to acceptable levels (Zar, 2010; Elliott and Hynan, 2011). Proportions of females trapped were compared to a nominal proportion of 0.5 with 95% Clopper-Pearson exact confidence intervals (Newcombe, 1998).

Scanning electron microscopy studies had previously shown that adult males of A. distinctus have sex-specific pores lying within indentations on the anterior corners of the prothoracic tergum (Ray et al., 2006). Such pores are known to be associated with glands that produce aggregation-sex pheromones (sensu Cardé, 2014) in other species in the Cerambycinae (Ray et al., 2006). To confirm the presence of the male-specific pores in A. distinctus, and check for the pores in A. electus, photographs of the pronota of pinned males and females were taken with a Leica Digital DFC450 camera on a Leica MDG41 stereomicroscope (Leica Microsystems, Heerbrugg, Switzerland). Close-up images were taken at ~130X magnification.

GC-FID and GC-MS analyses of odors collected from males of A. distinctus and A. electus revealed three and two peaks, respectively, that were absent in the analogous collections from females (Figure 1). Compounds 1 and 3 were tentatively identified as previously known cerambycid pheromone components, i.e., C6-ketol and the pyrrole, respectively, from their mass spectra, and the identifications were confirmed by matching their retention times and mass spectra with those of authentic standards. Further analyses of the extracts from each species on a chiral stationary phase Cyclodex B GC column determined that males of both species emitted only (R)-3-hydroxyhexan-2-one. Compound 2, detected only in extracts from A. distinctus males, was tentatively identified as 3-methylthiopropan-1-ol by matching its mass spectrum to one in the National Institute of Standards (NIS) mass spectral database installed on the GC-MS, and the identification was confirmed with an authentic standard, as described above. For A. distinctus, the ratio of (R)-3-hydroxyketone: 3-methylthiopropanol: pyrrole was (mean ± SE) 100: 3.6 ± 1.0: 3.4 ± 0.7 (N = 8) in the aeration extracts. For A. electus, (R)-3-hydroxyketone and the pyrrole were present in a ratio of 100: 5.7 ± 0.7 (N = 7).

Analyses of headspace volatiles of males of the non-target species determined that A. interrogationis, I. olivaceus, N. pusillus, and O. abdominale all produced only (R)-3-hydroxyhexan-2-one, with no trace of the pyrrole or thiopropanol. Whereas volatiles collected from males of C. aurigena, S. inerme, and U. nigricans all contained sex-specific compounds (identification of the new structures is in progress), none of these species produced the C6-ketol or the pyrrole.

In total, 264 adults of A. distinctus and 276 adults of A. electus were caught in the field bioassay that tested the putative pheromone components. Adults of A. distinctus were attracted in significant numbers only to traps baited with the full three-component pheromone blend (C6-ketol + pyrrole + thiopropanol (Figure 2A). The proportion of females was not significantly different from 0.5 (56% female; 95% Clopper–Pearson exact confidence interval: 0.50–0.62; P = 0.067). Similarly, adults of A. electus were significantly more attracted by their pheromone blend (i.e., C6-ketol and pyrrole) than to individual components and controls (Figure 2B), and the sex ratio of captured beetles was female-biased (65% female; 95% Clopper–Pearson exact confidence interval: 0.57–0.72; P < 0.001). Although this mean was ~50% higher than that of the blend containing thiopropanol, the difference between the 2-component and the 3-component blend was not statistically significant (Figure 2B).

Another 900 beetles of nine species in five tribes of the Cerambycinae were captured during the course of the field bioassays, among them the congener A. interrogationis (Table 1). Adults of this species were significantly and equally attracted by the treatments containing the C6-ketol alone or the two blends containing the C6-ketol as a component, indicating that the C6-ketol was necessary and sufficient to achieve attraction (Table 1). Similarly, I. olivaceus, N. pusillus, and O. abdominale appeared to be attracted by the C6-ketol or blends containing the C6-ketol (Table 1).

In contrast, adults of C. aurigena were most attracted to the two blends containing the C6-ketol and the pyrrole, with the two component blend of the C6-ketol and pyrrole being significantly more attractive than either component alone (Table 1). This was unexpected because neither of these compounds were among those detected in analyses of headspace volatiles from male C. aurigena (see above). In addition, adults of A. fulvoterminata and S. consobrina were attracted only to traps baited with the pyrrole alone, with no beetles being caught in traps baited with blends of the pyrrole with the C6-ketol, or with the C6-ketol and thiopropanol. Finally, there were no statistically discernable treatment effects for U. nigricans and S. inerme, despite there being totals of 96 and 50 beetles caught by baited traps compared to 0 and 1 in controls, respectively.

Photographic examination of the prothoraces of beetles confirmed that males of A. distinctus had the presumed pheromone pores on the tergum, and also demonstrated the presence of analogous male-specific pores on the tergum of male A. electus (Figure 3).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.