Identification and field evaluation of the sex pheromone of a Brazilian population of Spodoptera cosmioides

The objective of this work was to identify the sex pheromone of Spodoptera cosmioides and to evaluate whether there is pheromone cross‐attraction in Spodoptera sp. (Lepidoptera: Noctuidae). Spodoptera cosmioides gland extracts were analyzed by GC‐FID and GC‐MS. Wind tunnel and electrophysiology experiments were conducted to evaluate the role of gland compounds. In the field, different pheromone traps were tested: S. frugiperda commercial lure; (9Z)‐9‐tetradecenyl acetate (Z9‐14:OAc) and (9Z,12E)‐9,12‐tetradecadienyl acetate (Z9,E12‐14:OAc) trap; two females of S. cosmioides trap; and hexane control trap. Four acetates were identified in the S. cosmioides female gland extracts as Z9‐14:OAc, Z9,E12‐14:OAc, (11Z)‐11‐hexadecenyl acetate (Z11‐16:OAc) and hexadecyl acetate (16:OAc), but only the first two acetates induced electrophysiological responses from S. cosmioides male antennae. In wind tunnel experiments, S. cosmioides and S. frugiperda males responded more strongly to conspecific blends; however, there was some cross‐attraction, as 47% males of S. frugiperda and 25% males of S. cosmioides responded to heterospecific blends. In field experiments, S. frugiperda and S. cosmioides showed the same response pattern as observed in the wind tunnel bioassays. In summary, the sex pheromone components of S. cosmioides are Z9‐14:OAc and Z9,E12‐14OAc; they are important for conferring species specificity, and there is pheromone‐mediated cross attraction between S. frugiperda and S. cosmioides.


Introduction
The black armyworm, Spodoptera cosmioides Walker (Lepidoptera: Noctuidae), is a moth that occurs in tropical South America (Silvain & Lalanne-Cassou, 1997).In Brazil, it has been registered on more than 30 different crops, including soybean, maize, cotton, coffee, onion, and sunflower, causing severe damage that leads to substantial losses of production (Nagoshi, 2009;Lima et al., 2015).Due to its economic importance, several studies on the biology, rearing methods, biological control, and economic damage have been conducted for the population of S. cosmioides in Brazil (Bavaresco et al., 2004;Pomari et al., 2013).
Spodoptera cosmioides was considered a synonym for Spodoptera latifascia Walker (Lepidoptera: Noctuidae), but previous studies of French Guiana populations showed important differences in morphology, physiology, and sex pheromone composition (Silvain & Lalane-Cassou, 1997;Lalanne-Cassou et al., 1999).The sex pheromone for French Guiana populations was identified as a blend, consisting of two components, (9Z)-9-tetradecenyl acetate (Z9-14:OAc) and (9Z,12E)-9,12-tetradecadienyl acetate (Z9,E12-14:OAc) (Teixeira et al., 1989;Monti et al., 1995;Lalanne-Cassou et al., 1999).Studies have shown that there is great variability in the sex pheromone composition in Lepidoptera of different populations, even for strains that share the same habitat (El-Sayed et al., 2003).For instance, corn and rice strains of S. frugiperda (J.E.Smith) (Lepidoptera:Noctuidae) that inhabit the same geographic area have different calling time and sex pheromone blends (Groot et al., 2008;Unbehend et al., 2013).If this phenomenon is extended to other species of Spodoptera, there is a possibility that the population of S. cosmioides from French Guiana emits a different sex pheromone blend compared to the populations from central Brazil.
The use of sex pheromones combined with other control measures, such as biological control, could be an important strategy to manage Spodoptera spp. in Brazil, and to minimize the large amount of insecticides currently used to control Spodoptera populations in crop fields (Moscardi et al., 2012).The sex pheromones of S. eridania (Cramer), S. exigua (Hubner), S. frugiperda, and S. praefica (Grote) are commercially available (Meagher et al., 2008).The identification of S. cosmioides sex pheromone composition could be a relevant tool for its management.In addition, it is well known that some insect pheromones, when tested in field conditions, show cross-attraction to related species; this phenomenon has been reported for Hemiptera (Endo et al., 2006;Tillman et al., 2010), Coleoptera and Lepidoptera, including some species of Spodoptera (Mitchell & Tumlison, 1994;Meagher et al., 2008).In many regions of Brazil, populations of S. cosmioides are found coexisting with populations of S. frugiperda and S. eridania (Teodoro et al., 2013), thus, the possibility for identifying an attractive pheromone blend for different species could help to develop a multitarget monitoring technique.
The objective of this work was to identify the sex pheromone blend of S. cosmioides from Brazil, and to evaluate whether there was cross-attraction with S. frugiperda.
Pupae were sexed and placed inside 3 L plastic containers.After emergence, male and female moths were kept separately.Adults were fed with a sugar solution comprising 1 L water, 50 g honey, 50 g sugar, 1 mg Nipagin, and 1 mg ascorbic acid (Schmidt et al., 2001).For bioassays, all insects were 1 to 3-days-old.The insects were reared in climate chambers (Lab-Line, Mellrose Park, Ill, USA) on a 12:12 light: dark reverse photoperiod at 25 o C and 65% relative humidity.
Sex pheromone glands were excised from 2-4-day-old virgin calling females.Glands were forced to extrusion by gently pressuring the tip of the abdomen and were excised using small spring scissors model 15003-08 (FST, Vancouver, CO, Canada).Five to eight glands were placed in a 0.5 mL conical vial containing 100 µL of hexane, and were extracted at room temperature for 20 min.The extracts were filtered, using glass wool to remove solid particles, and pre-concentrated to 50 μL under a pure N 2 flow.The extracts (N=6) were stored at -20ºC until use.
The gland extracts were analysed by GC (Agilent 7890A, DB-5MS column, 60 m x 0.32 mm ID, 0.25 μm film, Supelco, Bellefonte, PA, USA), with the oven temperature maintained at 50ºC for 2 min, then increased at 5ºC min -1 to 180ºC for 0.1 min, followed by a gradual increase of 10ºC min -1 to 250ºC for 20 min.The column effluent was analysed with a flame ionization detector (FID) at 270ºC.One microliter of each sample was injected in splitless mode with helium as carrier gas.The samples were also analysed using a DB-WAX column and subjected to the same temperature program and flow conditions, to calculate the retention index (RI) of each compound.The data were collected with EZChrom Elite software and handled using Excel (Microsoft Office 2007, Microsoft Corporation, USA).For compound identification, selected extracts were analysed using an Agilent 5975C quadrupole mass spectrometer equipped with a DB-5MS (30 x 0.25 mm ID, 0.25 μm film, Supelco, Bellefonte, PA, USA), DB-WAX column (30 x 0.25 mm ID, 0.25 μm film, Supelco, Bellefonte, PA, USA) and a splitless injector, with helium as carrier gas.Ionization was achieved by electron impact (70 eV, source temperature 200ºC), and the data were collected with ChemStation software.Identifications were made by comparison of spectra with library databases (National Institute of Standards and Technology, 2008), or with published spectra, using retention indices (published at Pherobase and NIST Chemistry Web Book web sites), and confirmed by GC co-injection using authentic standards.
Gas chromatography -electroantennographic detection (GC-EAD) was used to determine compounds within mixtures that were detected by the male antennae.For this purpose, a Perkin Elmer Autosystem XL GC (NY, USA) was coupled to an EAD detector (Syntech, Inc., Hilversum, The Netherlands).The GC was equipped with a nonpolar DB-5 column (30 m x 0.25 mm ID, 0.25 µm film, J&W Scientific, Folsom, CA, USA), and a splitless injector with helium as the carrier gas.The oven temperature was programmed to start at 50 o C (2 min), then rise to 250 o C at 15 o C min -1 and hold at this temperature for 10 min.The effluent temperature to the GC-EAD system was kept at 195 o C. The antennae of one male were removed by using a small spring scissors and were immediately placed in stainless steel electrodes.The electric connection was achieved using conductive gel.The electrodes were connected to an autospike interface box and an AC/DC amplifier IDAC-2 (Syntech, Inc., The Netherlands).Preparations were done in a continuous humidified air flow (1 L min -1 ) with a Stimulus Controller CS-55 (Syntech, Inc., Hilversum, The Netherlands).Antennae of S. cosmioides males were tested using conspecific females gland extracts (N=5).Only peaks that showed depolarization and the repolarization of the antennae were considered as GC-EAD responses, and only those compounds that elicited responses in all tested antennae (N=5) were considered electrophysiologically active.A single antennal preparation was used for only one chromatography analysis.
Behavioural bioassays were conducted in a 1.5x0.5 x0.5 m (LxWxH) wind tunnel.Bioassays with S. cosmioides and S. frugiperda were conducted using 0.5 m s -1 airflow.Treatments were spotted on filter paper strips (1.5 cm long and 0.5 cm wide) (Whatman no.1), which were placed inside a metal mesh cage.The cage was placed on a support 15 cm above the wind tunnel floor and 30 cm from the upwind end of the tunnel.Males were released individually and, before testing, were allowed to acclimate for 5 min inside the wind tunnel while assembling the treatment cages.Male behavioural monitored steps were: taking flight, moving antennae, exposing the genitalia, and landing on the odour source.The first set of bioassays evaluated S. cosmioides and S. frugiperda male responses towards conspecific calling females, using the procedure described above, but with five calling females as the odour source.
The fourth set of bioassays verified the possibility of cross-attraction between Spodoptera spp.Therefore, cross-attraction was evaluated for S. frugiperda males to Mix-1, which attracted S. cosmioides males, and contained one compound that is not present in the sex pheromone blend of Brazilian populations of S. frugiperda and Mix-6 containing Z9-14:OAc + Z9,E12-14:OAc + Z7-12:OAc (5:0.5:0.02µg) was evaluated for both species.The ratio between the components was determined from previous experiments conducted in our laboratory.
All bioassays were conducted in a dark room, using a red light (14 Watts, Twister, Taschibra Indaial-SC, Brazil), at 27 o C and 65% relative humidity.For each treatment, 30 replicates were carried out and males were used only once.All bioassays were recorded from 5 to 9 hours within scotophase (Lalanne-Cassou et al., 1999).For the statistical analysis, only males that were capable of flying were ranked for: males that flew up wind but did not land on the odour source; and males that flew up wind and landed on the odour source.Each observation lasted 5 min, after which males were removed.
Field trials were conducted in Santa Helena de Goiás, GO, Brazil (17°48'S, 50°35'W) during April 2012.Four different treatments (N=5 for each treatment) were settled: A, grey rubber septa (11 mm) impregnated with 1 mg pentane, which was called Blank; B, two-day-old S. cosmioides females; C, Biofrugiperda, a commercial pheromone of S. frugiperda (Biocontrole, São Paulo, SP, Brazil), which contained three compounds Z9-12:OAc, Z11-16:OAc and Z7-12:OAc; and D, grey rubber septa impregnated with 1 mg of Z9-14:OAc and Z9,E12-14:OAc in the ratio (10:1).The experiment was performed in a cotton field at vegetative stage.Each treatment was hung up inside Delta traps (AR905-Plastic Delta traps -Isca Technologia, Ijuí, RS, Brazil), which were distributed in a completely randomized design and spaced 50 m from each other in a 5x4 grid.The traps were examined every three days when insects were identified and quantified, and virgin females of treatment B were replaced.After each monitoring period, the sticky floor of the traps was replaced and traps were reallocated to avoid positional bias.The total duration of the experiments was three weeks.
Statistical analysis was performed with R 3.0.1 (R Development Core Team, 2007).Data from wind tunnel bioassays were analysed using a GLM with a binomial distribution.The proportion and confidence interval (95%) of responding insects to each treatment were also calculated.In the field experiments, the total number of males of each species (S. frugiperda and S. cosmioides) captured in the traps, during all the experimental period, were used to compare the effect of treatment.The analyses were performed using a GLM with Poisson distribution of errors and the deviance analysis with 95% confidence level, using the number of insect per trap as a dependent variable, and treatment as the fixed effect.The mean number of insects captured by treatment was compared by contrast analysis.
In wind tunnel bioassays, conspecific S. cosmioides calling females used as the odour source stimulated 100% S. cosmioides males to display reproductive behaviour, which consisted of antennation, genital exposure and zigzag flights.Furthermore, 56% of these males landed on the odour source (Figure 4 A).Similarly, S. frugiperda males responded to calling conspecific females; 100% of these males showed   The RI was calculated using a DB-5-MS and DB-Wax column.
reproductive behaviour, and 66% of them landed on the odour source.When S. cosmioides males were stimulated with five calling S. frugiperda females, only 25% of males landed on the odour source, whereas 65% S. frugiperda males landed on the odour platform when five S. cosmioides females were used as odour source (Figure 4 B).The results suggest that the sex pheromone blend produced by S. frugiperda is more species-specific compared to that of S. cosmioides, considering the cross-attraction between these two species.Spodoptera frugiperda males response to S. cosmioides females might be related to the production of specific and shared components.
are precursors of the main components of the sex pheromone of Spodoptera sp. and they neither elicited any response from males, nor showed an antagonist effect.By contrast, Z7-12:OAc was added to the binary blend (Mix-6), only 10% of S. cosmioides males landed on the odour source, showing that this component is important for the reproductive isolation of both species.When Mix-6 was tested with S. frugiperda males only 47% of individuals landed on the odour platform.The presence of Z9,E12-14:OAc decreased the attraction of S. frugiperda in comparison to the percentage of individuals that landed when conspecific females were used as odour source.The same percentage of S. frugiperda males landed on the odour source, when the binary blend containing Z9-14:OAc and Z9,E12-14:OAc was evaluated.This shows that the presence of Z7-12:OAc did not increase the attraction; however, the presence of Z9,E12-14:OAc inhibited the attraction of some individuals.The results obtained in the bioassays indicate that both specific compounds Z7-12:OAc (S. frugiperda) and Z9E12:14OAc (S. cosmioides) confer species-specificity and reproductive isolation of the sex pheromone blend.
Field experiment results corroborate those obtained by the wind tunnel bioassays.Spodoptera cosmioides was attracted to the commercial pheromone blend, that contains Z7-12:OAc, but it was more attracted to the binary solution of Z9-14:OAc and Z9,E12-14OAc, showing that this binary blend might be used for monitoring this species in the field.Laboratory and field results, obtained in the present work, suggest that moths can distinguish the specific blends, but there is some degree of cross-attraction between S. frugiperda and S. cosmioides pheromone blends.The synthetic mixture of Z9-14:OAc and Z9,E12-14OAc was able to attract both S. cosmioides and S. frugiperda in field conditions, suggesting the possibility of using a single synthetic mixture to attract more than one species of Spodoptera in field conditions.Cross-attraction between Spodoptera species was also previously demonstrated for S. triturata (Walker), S. exempta (Walker) and S. littoralis (Boisduval) (Khasimuddin & Lubega, 1984), S. eridania and S. exigua (Hübner) (Mitchell & Tumlinson, 1994), and between native and exotic species in Florida, USA (Meagher et al., 2008).In general, cross-attraction in Spodoptera is related to the major compounds that are shared by species attracted to the same pheromone blend as observed for S. cosmioides and S. frugiperda in the present work.Total number (mean±SD) of Spodoptera frugiperda and S. cosmioides males captured in cotton field experiments using four different treatments: blank, grey rubber septum impregnated with pentane; two twoday-old S. cosmioides females; commercial pheromone of S. frugiperda (Biofrugiperda), which contained the compounds Z9-12:OAc, Z11-16:OAc, and Z7-12:OAc; and grey rubber septum impregnated with Z9-14:OAc and Z9,E12-14:OAc, at 10:1.N=5 traps for each treatment.Bars followed by different letters for each species indicate statistical differences in the proportion of responses between treatments (general linear models, deviance analyses and contrast analyses, p<0.05).
An interesting aspect of this study was that the field experiments showed that it is possible to use just the two major compounds found in the sex pheromone glands of S. cosmioides to monitor this species in field crops.Different species coexist at the same time in the field, and to use cheap and efficient pheromone traps that capture more than one species might be very useful for farmers.However, more intensive field work is required, in order to assess the effectiveness of these pheromone traps as monitoring tools, as well as to test the relationships between trap catch and population densities, to define more accurately the precise volatile pheromone blend for these two components, and to evaluate the possibility of capturing more Spodoptera species.
3. The wind tunnel bioassays and field experiment showed that there is pheromonal cross-attraction of S. frugiperda and S. cosmioides.
4. The response of S. frugiperda males is more species-specific than that of S. cosmioides.

Table 1 .
Compounds identified in female gland extracts of Spodoptera cosmioides, the retention index (RI), and the produced amount of each compound±standard error (SE) per gland, n=6.