Performance of Trichogramma pretiosum Riley (Hymenoptera: Trichogrammatidae) on eggs of Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae)

Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) is a polyphagous pest with a wide geographic distribution. This pest first arrived in Brazil in 2013, and since then studies on possible control methods for it have been necessary. A possible method for the control of H. armigera is using the egg parasitoid Trichogramma pretiosum Riley (Hymenoptera: Trichogrammatidae). Therefore, the objective of this study was to evaluate the performance of T. pretiosum on H. armigera eggs, which are known to represent suitable hosts for the development of this parasitoid species in the laboratory. Parasitism and emergence rates and the duration of the egg-to-adult period of T. pretiosum were investigated following 24- and 48-h exposures of this parasitoid to H. armigera and Corcyra cephalonica (Stainton) (Lepidoptera: Pyralidae) eggs. The longevity of offspring after the 24-h exposure was studied, as well as the frequency of parasitism and emergence, host preference, and the emergence of offspring from eggs of different ages or oviposited by lepidopterans on different days. Parasitism was 14.4 and 34.9% more frequent on C. cephalonica than on H. armigera after 24 and 48 h of exposure, respectively. In C. cephalonica, parasitism was 27.2% higher after 48 h. Parasitism was more frequent on C. cephalonica eggs collected on the second day of oviposition (76.2%), and on H. armigera on the third day of oviposition (71.1%). Parasitism frequency was lower on 2-day-old C. cephalonica eggs (63.3%) and on 3-day-old H. armigera eggs (41.3%). When tested with a chance of choice between hosts, T. pretiosum preferred H. armigera, while in the test with no chance of choice there was no difference in preference. Thus, T. pretiosum may be considered a tool for the Integrated Pest Management (IPM) of H. armigera.

parasitism by T. pretiosum on H. armigera eggs. Eggs were used that were up to 24 h of age (i.e. up to 24 h after female oviposition), from the first day of female oviposition on C. cephalonica (alternative host) and H. armigera (natural host). The egg volume of H. armigera is 0.08 mm 3 21 , and that of C. cephalonica is 0.036 mm 3 22 . The eggs were glued to light blue paper (3.5 × 1.5 cm) with arabic gum (50%) diluted in deionized water. C. cephalonica eggs were handled with a soft-bristled brush, while those of H. armigera were cut out individually from the paper that was used as an oviposition substrate and glued to the light blue paper. Both types of eggs were exposed to germicidal light for 45 min to arrest their development. Each experimental replicate corresponded to a flat-bottomed glass tube (8.0 cm high × 2.0 cm in diameter), which contained a piece of light blue paper with eggs and a female parasitoid, and was covered with polyvinyl chloride plastic film. A droplet of honey was placed on the inner surfaces of the tubes to feed the females. Twenty replicates were used, with 30 eggs per host. Two sets of females were used in different treatments, one in which the female was allowed to parasitize the eggs for 24 h, while in the other treatment it was allowed to parasitize the eggs for 48 h, after which they were withdrawn and discarded. Female wasps used in the experiments were up to 24-h-old. The tubes were subsequently checked for adult emergence once per day. Parasitism frequency was evaluated by counting the number of dark eggs (signaling the occurrence of parasitism), and the emergence rate was determined by counting the number of dark eggs with holes. The duration of the period elapsed from the egg to adult parasitoid was also determined, and was measured from the day of parasitism to the day of offspring emergence. Laboratory conditions were controlled at 25 ± 2 °C with a 70 ± 10% RH and a 12-h photoperiod during these experiments.

Longevity of T. pretiosum.
After the emergence of the parasitoids in response to the 24-h treatment described in the previous section ("Parasitism by Trichogramma on H. armigera eggs"), 40 females from both hosts were randomly selected and used in further experiments to determine the longevity of the adults. The females were kept individually in flat-bottomed glass tubes (8.0 cm high × 2.0 cm in diameter), which contained a honey droplet on their inner surface for food and were covered with polyvinyl chloride plastic film. Each wasp was provided honey ad libitum, and thus food level did not affect how long each wasp lived. Laboratory conditions were controlled at 25 ± 2 °C with a 70 ± 10% RH and a 12-h photoperiod.
parasitism of T. pretiosum on eggs oviposited by H. armigera on different days. Eggs from the first, second, third, fourth, fifth, and sixth day of C. cephalonica and H. armigera oviposition were used in this test. Thirty eggs from each host collected on different days of oviposition were exposed to a T. pretiosum female for 24 h in a flat-bottomed glass tube (8.0 cm height × 2.0 cm in diameter) containing a droplet of honey on the inner-side surface and covered with polyvinyl chloride plastic film. The eggs used in the different treatments were at the same developmental stage (<24 h post-oviposition). The eggs were attached to light blue paper and placed under germicidal light for 45 min, as described in a previous section ("Parasitism by Trichogramma on H. armigera eggs"). Fifteen replicate tests were performed, with 30 eggs used for each host, in which parasitism and parasitoid emergence frequencies were evaluated. Female wasps used in the experiment were up to 24-h-old. Laboratory conditions were controlled at 25 ± 2 °C with a 70 ± 10% RH and a 12-h photoperiod. parasitism by T. pretiosum on eggs of H. armigera of different ages. Each female parasitoid was placed in a glass tube (8.0 cm height × 2.0 cm in diameter) covered with polyvinyl chloride plastic film and with a honey droplet provided on the inner wall for food. Thirty C. cephalonica eggs that were 1-, 2-, 3-, or 4-days-old were obtained from laboratory rearing, glued onto light blue paper (3.5 × 1.5 cm), and then placed into the glass tubes. This procedure was repeated with H. armigera eggs. The eggs used in this experiment were not placed under germicidal light because in this experiment we wanted to study the development of the embryo after parasitism. Female wasps used in the experiment were up to 24-h-old. After 24 h of parasitism, T. pretiosum females were removed and discarded. Fifteen replicates were observed, with 30 eggs for each host. The tubes containing light blue paper were kept in a heated room, under the conditions described previously, until the offspring emerged. The percentages of eggs that were parasitized and underwent parasitoid emergence were also evaluated. Laboratory conditions were controlled at 25 ± 2 °C with a 70 ± 10% RH and a 12-h photoperiod. Host preference of T. pretiosum. Arenas were set up that were 4-cm high, made of transparent polyethylene acrylic, and contained four Duran tubes arranged equidistantly from the covering holes ( Fig. 1) 23,24 . The insects could move between the different eggs in each tube to choose between them based on touch (egg-size measuring) and odor. In the test with a double chance of two choices, light blue pieces of paper (0.4 × 2.0 cm) containing 15 C. cephalonica eggs each were placed in two opposing tubes, and 15 H. armigera eggs were placed in the other two tubes. In the no-chance tests, only two tubes containing 15 eggs of either C. cephalonica or H. armigera were placed in the arena. The eggs were placed under germicidal light for 45 min before being used in the test. A female was released into each arena, through a hole located in the top part of the cap. After 24 h, the Duran tubes with the eggs were removed from the arena, covered with polyvinyl chloride plastic film, and kept in a heated room until the adults emerged. Fifteen replicates were observed for each treatment, and the frequencies of parasitism and emergence of offspring were evaluated. Laboratory conditions were controlled at 25 ± 2 °C with a 70 ± 10% RH and a 12-h photoperiod.
Data analyses. Data on the parasitism, emergence, and egg-to-adult period of T. pretiosum were submitted to the Kolmogorov and Bartlett tests to verify the normality of their residuals and the homogeneity of their variances, respectively. The data that met these assumptions were then submitted to analysis of variance (ANOVA). When there were two treatments, the Student's t-test was used to compare different treatment conditions to each other, and when there were more than two treatments the Student-Newman-Keuls test was used (P < 0.05). When the data did not meet the requirements for ANOVA, the most adequate transformation was used, and if they still did not present normal residuals and homogeneous variances the data were then submitted to non-parametric tests. For the non-parametric tests, the Wilcoxon test was used to compare two treatments and the Kruskal-Wallis test was used to compare three or more treatments (p < 0.05). All analyses were conducted using the SAS software 25 . Survival curves were also constructed using survival data at specific ages, and were compared according to the Kaplan-Meyer methodology 26 and analyzed using SAS software 25 .
The frequency data from the choice tests were analyzed using Proc FREQ 25 and interpreted by the chi-square (χ 2 ) test, in which 1:1 was the null hypothesis assumed if the parasitoid had no preference for one host over the other.  The duration of the egg-to-adult period of T. pretiosum was 10 days on both hosts after 24 h of egg exposure. At 48 h, the duration of the developmental period was longer on C. cephalonica eggs (11.3 days) than on H. armigera eggs (10.9 days) (z = 2.36, df = 1, p = 0.0182) ( Table 2).
Longevity of T. pretiosum. Following emergence in host eggs parasitized for 24 h, T. pretiosum females from C. cephalonica had a longevity of up to 12 days, which differed in relation to those from H. armigera, who had a longevity of up to 13 days. However, 3 days after emergence the percentage of surviving adults that had emerged from H. armigera eggs was reduced to 50%. The equivalent reduction was observed only on day 8 for insects that had emerged from C. cephalonica eggs. The survival of insects from C. cephalonica eggs was lower than that of those from H. armigera eggs on day 11 only (Fig. 2).

Host preference of T. pretiosum.
In the test where the parasitoid was given a chance of choice between hosts (χ 2 = 20.12, df = 1, p < 0.0001), T. pretiosum preferred to parasitize H. armigera eggs (80%) over those of C. cephalonica (20%). In the tests with no chance of choice (χ 2 = 1.1843, df = 1, p = 0.2765), there was no preference of the parasitoid for one host over the other (Fig. 3).
There was no difference between the hosts in the emergence of adults observed; in the tests with a chance of choice (χ 2 = 0.0179, df = 1, p = 0.8937) and no chance of choice (χ 2 = 1.2678, df = 1, p = 0.2602), T. pretiosum grew equally well on both C. cephalonica and H. armigera eggs.

Discussion
The present study contributes basic information useful for studies aiming to achieve the management of an introduced pest in Brazil, H. armigera, using the egg parasitoid T. pretiosum, which is efficient at controlling some lepidopteran pests in several crops of economic importance 16,27 .
The parasitism frequency of T. pretiosum over 24 and 48 h of exposure to C. cephalonica and H. armigera eggs was higher in the alternative host, C. cephalonica, which is used to maintain laboratory rearing cultures of the parasitoid. The percent parasitism at 24 and 48 h on H. armigera eggs was lower than that reported by Ballal and Singh 28 , which was 76.7% under laboratory conditions when using an Indian population of the same pest. However, before tests were done the authors of the present study reared the parasitoid for two generations on H. armigera eggs. Therefore, it is possible that the better performance of the parasitoid on C. cephalonica eggs was related to behavioral adaptation (i.e. to insect conditioning), as it was maintained for successive generations in  the laboratory on the eggs of this host [29][30][31] . In the laboratory, insects can adapt their behavior to the conditions in which they are maintained, both in adults and in the young 29 . However, previous results showed that the measures of laboratory performance used (fecundity and offspring sex ratio) were good predictors of field success in T. pretiosum 32 . However, the parasitism frequency on H. armigera eggs observed herein was approximately 50% (49.7%), which suggests that the release of T. pretiosum in the field would need to be combined with other tools to control the pest effectively. T. pretiosum can be released in conjunction with the application of chemical and biological products, which must be efficient for use in the control of the pest but also selectively harmless to the parasitoid 33,34 . In sweetcorn crops in eastern Australia, Trichogramma release was conducted along with the use of bacteria-and virus-based biological products in IPM for the control of H. armigera 35 . In China, Trichogramma species are also used for pest management in maize, including the control of the eggs of H. armigera 36 .
In Turkey, the release of 120,000 Trichogramma parasitoids per hectare to control H. armigera on cotton crops resulted in parasitism frequency being reduced to 52.5% of its initial value 37 , highlighting the potential of Trichogramma for use in the biological control of H. armigera.
The percent emergence was similar between the hosts tested following 24 h of exposure, which suggests that the eggs of both species permit the complete development of the parasitoid 38 . However, with 48 h of exposure the emergence rate was reduced in both species to values below the ideal value of 85% 10 . It is possible that by remaining in contact with the eggs for a longer period, some females may deposit their eggs onto a host that is already parasitized, either by itself or by another female, which results in superparasitism. This behavior may impair the development of the parasitoid within the egg, or even kill both the developing parasitoid and the host. When a host is superparasitized, some or all of the immature parasitoids are insufficiently nourished, and thus fail to fully develop and die. In some cases, insects are born with a smaller size or with deformations [39][40][41] . Superparasitism may therefore reduce the parasitoid's reproductive success 42 . Furthermore, in larger hosts gregarious development may occur, as is the case for H. armigera eggs and could explain the apparently reduced levels of parasitism on this species 43 .
In practice, our results regarding exposure time differences suggest that there is a need for periodic releases of the parasitoid, since the parasitoid will not maintain its population at the necessary level required in the field to control the pest once emergence has declined. EMBRAPA (the Brazilian Agricultural Research Corporation) recommends that the parasitoid is released when, during sampling with adhesive pheromone traps, the first adults of H. armigera are collected 44 . However, in this study the highest frequencies of parasitism occurred on eggs collected on the third day of pest oviposition. These results are of great importance to the release of T. pretiosum as they will help the parasitoid to be released at the most appropriate time to best contribute to the efficient control of H. armigera in the field.
Regarding the ages of the parasitized eggs, there was reduced parasitism on 2-day-old C. cephalonica eggs and 3-day-old H. armigera eggs compared to that at other ages. The age of host eggs may influence the performance of egg parasitoids used in biological control 45 , since eggs may undergo morphological and physiological changes that may interfere with their acceptance by the female 46,47 . As the host egg is in a transitional stage of development, the parasitoid must kill the embryo and prevent its development, and then subsequently oviposit its eggs 48 .
In 3-day-old H. armigera eggs, it is possible that the embryo is already developing and occupying most of the total volume of the egg, with a high level of sclerotization, which increases the protection of the host and decreases the amount of food available for the development of a Trichogramma larva inside the egg 48 .
In general, it can be assumed that there was parasitism on H. armigera eggs of all ages, indicating the acceptance of the eggs of this species by the parasitoid, even at older stages. If the environmental conditions are favorable to the parasitoid insect for a longer period in the field, it may continue to parasitize pest eggs of all ages present on the crop, preventing the larvae from hatching.
The age of the egg following parasitism did not affect the emergence of the parasitoid offspring in this study, as was also previously observed with T. pretiosum on Mocis latipes eggs (Guenée) (Lepidoptera: Noctuidae) 49   (Linnaeus) (Lepidoptera: Plutellidae) 51 . Therefore, the offspring and their descendants should be able to establish persistent populations in the areas in which they are released.
In the preference tests conducted, when a chance of choice between host species was provided to the parasitoid, it preferred the eggs of the natural host, H. armigera. Several factors influence host preference, such as egg size, which is a critical factor in host selection. In general, most Trichogramma species tend to prefer to oviposit on medium-to large-sized eggs 52 . In the present study, the eggs of H. armigera and C. cephalonica had an approximate volume of 0.08 mm 3 and 0.036 mm 3 , respectively 21,22 , which may be one of the factors explaining this preference. Further studies on the mechanisms associated with host selection should be performed with T. pretiosum and H. armigera to guarantee the success of the application of this parasitoid in the biological control of the pest.
The results obtained in this study demonstrated the possibility of using T. pretiosum in the control of H. armigera. Semi-field and field tests should be performed to adjust and optimize the release methodology, as well as studies of the association of this parasitoid with other pest control techniques.