Short communication. Mutual pheromone antagonism in two sympatric corn borers, Sesamia nonagrioides and Ostrinia nubilalis, under field conditions

In previous studies, we demonstrated cross-antagonism in pheromone perception between pheromone components of the two corn (Zea mays L.) borers Sesamia nonagrioides Lefèbvre (Lepidoptera: Noctuidae) and Ostrinia nubilalis (Hübner) (Lepidoptera: Crambidae) in the laboratory and in the field. The two pheromone components identified as responsible for this cross-antagonism were Z 11-16:Ald, a minor component of S. nonagrioides pheromone, and Z 1114:Ac, the main component of the pheromone of the Z-strain of O. nubilalis, which inhibited the response of O. nubilalis and S. nonagrioides, respectively. Here, we study this antagonism phenomenon in the field by air permeation of maize plots with each of the two components separately and measurement of mating in caged couples of the two corn borers on treated and untreated plots during three years. A significant reduction in mating rates was observed on the permeated plots: 7% for S. nonagrioides and 12% for O. nubilalis. When dispenser charges (200 ng) were increased by 50% and 75% in the third year, no decrease in mating rates was recorded at either of the increased concentrations. On the other hand, the use of large cages resulted in an increase of 8% to 12% in the percentage of unmated females in each of the two corn borers suggesting that at more realistic field corn borer densities, 0,1 couples/plant instead of the 2 couples/plant used in this experiment, cross-antagonism in the two corn borers could be higher than that recorded in small cages. Additional key words: inhibition; maize; mating disruption.

In a previous paper (Gemeno et al., 2006) we recorded cross-antagonism in pheromone perception between pheromone components of the two corn borers. In wind tunnel and electroantennogram studies, we observed behavioural inhibition of O. nubilalis male response to its own pheromone when the latter was mixed with a certain amount of a S. nonagrioides pheromone component, Z 11-16:Ald. Conversely, response of S. nonagrioides males was inhibited when a component of the pheromone of the other corn borer, Z 11-14:Ac was added to the pheromone blend (Eizaguirre et al., 2007). This mutual behavioural antagonism was also confirmed in the field since male catches of each corn borer were practically zero when the above-mentioned components were cross-added to pheromone traps.
In order to confirm cross-behavioural antagonism between S. nonagrioides and O. nubilalis pheromone components in the field and its role in mating prevention, the effect of the two most active components as pheromone antagonists recorded in the laboratory was measured in the field. For this, the antagonistic effect of the S. nonagrioides pheromone Z 11-16:Ald on O. nubilalis mating and of the O. nubilalis pheromone Z 11-14:Ac on S. nonagriodes mating, male response was tested in the field.
Insects of S. nonagrioides and O. nubilalis used in the experiments came from the continuous rearing maintained in the laboratory and periodically renewed with wild individuals collected in the area study following the methods detailed in Eizaguirre and Albajes (1992) and Riba et al. (2005) respectively.
Trials were carried out during three consecutive years (2006 to 2008) in Lleida area (Catalonia, NE Spain). Treatments were replicated three times in a randomized complete block design (Gómez and Gómez, 1984). Plot size was 1500 m 2 ; the treatment consisted of air permeation with Z 11-16:Ald or Z 11-14:Ac, placing 75 solid dispensers (one per 5 × 5 m 2 ) and it was compared with an untreated plot. In the case of Z 11-16:Ald, the dispenser was a cylinder of PETPAI (PET051 ® , of 45 mm length and Ø13 mm with a cylindrical handle of 15 mm length and Ø8 mm for dispenser manipulation) and for Z-11-14:Ac, the dispenser was a bag of KVT (KVT ON ® ; size 5 × 5 cm 2 ). Both dispensers were provided by Sociedad Española de Desarrollos Químicos, SEDQ ® , Barcelona, Spain. Dispensers were charged with 200 ng of each pheromone component. They were tied to a maize leaf at 1.2 m height. The evaluation period covered 6-7 weeks from approximately twelfth-fourteenth leaf fully emerged to silking corn (V12-14 to R1, in Ritchie's nomenclature) (Ritchie et al., 1992).
Three 1-to 2-day-old virgin couples of S. nonagrioides or O. nubilalis were introduced into the cages [made with a wooden frame (0.5 × 0.4 m 2 and 0.5 m high) and covered with a plastic screen (mesh 2 × 2 mm 2 )] and these were placed on plots permeated with Z 11-14:Ac or Z 11-16:Ald, respectively and on the untreated plots. Couples were kept in cages for 2 to 3 days and then removed. The females were brought to the laboratory to assess mating status by observing spermatophore presence.
In order to improve the evaluation technique, influence of cage size on mating rates was studied by comparing the cages described above with larger ones (2.35 × 0.6 m 2 and 0.7 m high). The total number of S.nonagrioides and O. nubilalis females recovered from cages and then dissected was in 2006 287/344, in 2007 208/242 and in 2008 454/383, respectively. Additionally, the influence of the pheromone component amount on mating rates was evaluated by increasing the concentration in dispensers by 50 and 75%. These two treatments were added to each block in 2008.
The pheromone released throughout the experimental period has been measured each year by analysis of the dispensers at the beginning and at the end of the experimental period. A total of 16 dispensers (8 loaded with each component) were placed to monitor release rate throughout the experimental period. The amount of active ingredient in the dispensers at the moment of installation in the field and the content of each pheromone component remaining in the 16 dispensers at the end of the experimental period (45 days after deployment in the field) were recorded. Extraction of each group of dispensers was performed in a Soxhlet apparatus using 200 mL of hexane at reflux for 10 h. Quantif ication was carried out using a HP-1 12 m × 0.2 mm × 0.33 mm capillary column with nonyl acetate as internal standard under the following chromatographic conditions: injection at 50°C and program of 10°C min -1 up to 270°C, which was maintained for 10 min. The total amount released during the experimental period was calculated by subtracting the mean amount remaining in the dispensers at day 45 from the amount found at day 0.
To analyse the efficacy of the treatment for each corn borer, we calculated the number of mated and unmated females during the year and the mean percentage of unmated females [100*(unmated)/(unmated + mated)] transformed by ASIN[SQRT(x100)] to normalise it. A split-plot-like model (Gómez and Gómez, 1984) was used to analyse this percentage.
Year was considered the main plot and treatment the subplot. All factors, except blocks, were considered fixed. Influence of cage size and pheromone component concentration on mating rate was analysed with a three-way (cage size, component concentration and disruption treatment) ANOVA and LSD test to compare means. Mean corrected efficacy during the experimental period was calculated with Schneider-Orelli's formula (Püntener, 1981): [corrected efficacy = 100* (% response in treatment -% response in control)/ (100 -% response in control)].
The percentage of unmated females was significantly higher for both corn borers in the treated plots, as shown in Table 1 regardless of the year. Corrected mean efficacy of antagonism at the end of the experiment was 10.8% for S. nonagrioides and 20.0% for O. nubilalis in base to the percentages of unmated females shown in Table 1.
Larger cages showed significantly higher percentages of unmated females in both corn borer species (Table 2), with an increase of about 12% for S. nonagrioides and about 8% for O. nubilalis. On the contrary, increases of 50% and 75% in the pheromone component concentration did not significantly decrease ma-ting rates (F = 0.00; df = 2,21; p = 0.99 for S. nonagrioides and F = 1.64; df = 2,21; p = 0.22 for O. nubilalis).
The total amount of pheromone components released during the experimental period is shown in Figure 1. In the case of Z 11-14:Ac to prevent S. nonagrioides mating, the amount actually released was 80 g ha -1 . In the case of the Z 11-16:Ald component to prevent O. nubilalis mating, the values were close to 80 g/ha in 2006 and 2008 but clearly lower in 2007.
Over three years of study in the field, the recorded mutual antagonism of pheromone components to male response in the two corn borers has confirmed the results previously obtained in the laboratory. The earlier studies had shown that when Z 11-16:Ald, a pheromone component of S. nonagrioides pheromone, was added to the O. nubilalis pheromone blend, it significantly reduced oriented flight and pheromone source contact in the wind tunnel of O. nubilalis males and nearly eliminated trap catch in the f ield (Gemeno et al., 2006). A similar antagonistic response was recorded in the male of S. nonagrioides when Z 11-14:Ac, the main component of local populations of O. nubilalis (Sans, 1996), was added to a pheromone blend of the former (Eizaguirre et al., 2007).
The results of the present work show that the release of the O. nubilalis pheromone component reduced    Albajes et al., 2002;Sole et al., 2008). However, as the main objective of this work was not to put a mating disruption system into practice but to confirm the cross-antagonistic action on pheromone perception by males of the two corn borers, the worst-case situation was chosen. To evaluate the potential of crossantagonism to disrupt mating in practice, experiments should be carried out in more realistic conditions. At least three kinds of experimental factors should be taken into account when a more realistic evaluation of cross-antagonism potential for mating disruption purposes is performed. First, common adult densities in the f ield are much lower than those used in our experiments within the cages. In the field, 0.1 couples per plant is a common adult density in the second generation adults whereas in this work, densities of 2 and 0.25 couples per plant, small and large cages respectively, have been tested. The lower mating rates found in females released into larger cages in comparison with smaller ones would conf irm that visual recognition might play a role in high densities as it had been observed in the field (López et al., 2002).
An insufficient amount of pheromone components released on the treated plots is a second experimental factor that could influence the recorded efficacy of mating disruption but it was shown in that study that an increased concentration in the dispenser charge did not lead to an increase in the antagonistic effect. Therefore, it does not seem that the low efficacy rates of mating disruption found were due to an insuff icient antagonist concentration on the treated plots. In fact, Fadamiro et al. (1999) found similar levels of disruption when they used 240 g ha -1 of the complete pheromone of O. nubilalis to disrupt its mating.
A third experimental factor to be considered in the evaluation of cross antagonism for mating disruption in the field concerns with dispensers and release rate. Improving release rates and uniformity of dispensers should be one of the objectives of the system.
In summary, the capacity of some of the pheromone components of each of the two corn borers as behavioural antagonists of the other has been confirmed in the field, as it has been observed in the laboratory. Such antagonistic behaviour may be used simultaneously for mating disruption of the two corn borers, although it may be expected that mating in O. nubilalis was more efficiently disrupted than in S. nonagrioides. However, for practical applications of cross-antagonism in the field for mating disruption, release rates of dispensers, concentration of antagonist in the dispenser and techniques to be used to assess efficacy should be further investigated.