1. Introduction
Tomato (Solanum lycopersicum L.) is the second most important vegetable crop next to potato. In 2013, world production of tomatoes was reduced to 163.4 million tonnes, with Iran accounting for 3.8% of the total [1] . This crop is attacked by a wide range of pests. Invasive species represent a major threat to crops [2] [3] [4] and agricultural pests can reduce yield and increase production costs related to their management [5] . The tomato leaf miner, T. absoluta (Lepidoptera: Gelechiidae), has been a long-time pest of open field and greenhouse tomato in South America [6] [7] [8] .
This key pest started moving and invading towards the eastern regions, within being first reported from Spain in 2006. Today, T. absoluta represents a major threat to both natural and agronomic ecosystems in most countries of South America, Europe, the Middle East and some part of the pest status of T. absoluta has been argued to arise mainly from its high fecundity, high mobility, migratory potential as well as its high propensity to rapidly develop resistance against different classes of pesticides [1] . The larvae of T. absoluta mine the leaves producing large galleries and burrow into the fruit, causing a substantial loss of tomato production in protected and open field cultivations. Damage may reach to 100% of the yield if no control strategy is taken. T. absoluta can complete between 10 - 12 generations per year in appropriate climatic conditions and each female can lay between 250 - 300 eggs during her life time. A variety of methods, including monitoring, mass trapping as well as the use of chemical, cultural, and biological factors have been developed for control of T. absoluta, each of which may contribute partially to reduce damage on cultivations.
For example, the use of synthetic insecticides for control of T. absoluta is questionable with respect to the mine-feeding nature of the larvae [9] . On the other hand, as tomato is consumed as raw product in many countries, the use of synthetic pesticides has provoked concern about the harmful effects of these compounds on human health. Additionally, wide application of chemical pesticides is expected to threat the survival and normal activity of non-target organisms, particularly natural enemies and pollinators more than the target pest itself, thus negatively affecting on production by reducing the efficiency of pollinators and biological control agents.
Biological control, as an alternative to chemical control, has been attempted in several South American and European countries [10] [11] . Currently, two important groups of biological control agents, i.e. entomopathogenic bacteria and oophagous parasitic wasps have been considered as efficient agents for control of T. absoluta. The soil dwelling bacterium, Bacillus thuringiensis var. kurstaki, for example, have exhibited satisfactory efficacy against T. absoluta larval infestations in Spanish outbreaks. Delayed application of this bacterium in combination with the use of resistant tomato cultivars has been argued to cause higher larval mortality, because the larvae seem to become more susceptible to the pathogen after a longer period of feeding on the resistant crop. The egg parasitic wasps of the genus Trichogramma, are another important group of biological control agents that have provided promising results for control of T. absoluta under greenhouse conditions. T. achaeae, for example, has been identified as a suitable species for biological control of T. absoluta in South American. Under greenhouse conditions, the release of 30 adult female wasps per plant (i.e. 75 adults per m2) every 3 - 4 days during August and September of 2008 was reported to reduce the damage by 91.74% in the southeast of Spain [12] . Several other native species of Trichogramma have been also evaluated for controlling T. absoluta in many crops worldwide [10] - [17] . Given the host specificity of B. thuringiensis var. Krstaki against lepidopteran larvae, this bacterium is an ideal agent to be used in combination with other biological agents, such as Trichogramma wasps, in integrated management of T. absoluta. It has been reported that a combine application of T. pretiosum and B. thuringiensis resulted in the reduction of fruit damage by 2% in South America.
T. brassicae and T. embryophagum are among the most widespread species of Trichogramma in Iran [18] [19] . Before decision making about the combined use of biological agents, it is necessary to evaluate their unwanted side effects on each other’s in order to select for the best combination from available agents and so, prevent increased production costs.
In the current study, the efficiency of T. brassicae and T. embryophagum alone and in combination with B. thuringiensis was investigated against the tomato leafminer T. absoluta under greenhouse conditions. Additionally, the parasitism rate and some other biological characteristics of T. brassicae and T. embryophagum were investigated in presence of B. thuringiensis, in order to explore any detrimental effects of the entomopathogenic bacterium on these egg parasitoids.
2. Materials and Methods
2.1. Plants and Insects
Tomato plants (Solanum lycopersicum L.) were grown in a greenhouse (25˚C ± 5˚C, 70% ± 5% RH, 16L:8D), seeds were originated from the University of Tehran, Department of Plant Protection, Ecology and Behavior Lab (Karaj, Iran). Five week old plants were used in the experiments.
All used insects were collected from the University of Tehran. Mated females of T. absoluta were obtained by pairing a virgin male and a virgin female moth. The eggs of T. absoluta were obtained by placing adult males and females approximately 30 adult inside oviposition cages (50 cm × 50 cm) containing tomato plants offered as a host plant for egg laying and removed 24 h later, tomato plants with eggs were kept at the same conditions in the growth chambers until used for the experiments. Colonies of T. brassicae and T. embryophagum (Hymenoptera: Trichogrammatidae), were reared on eggs of the Mediterranean flour moth, Ephestia kuehniella (Lepidoptera: Pyralidae) in the University of Tehran, in a climate chamber (25˚C ± 5˚C, 70% ± 5% RH, L16:D8). Only mated 24 hour old wasps were used in the experiments.
2.2. Formulation of Bt
The Bt formulation used in the experiments was B. thuringiensis sbsp. kurstaki (109 spore/g) WP, Biolep, Biorun Company, Iran.
2.3. Test Facilities
The experiment was conducted in a 6 × 10 m2 research greenhouse located at the University of Tehran, Department of Plant Protection (Karaj, Iran). Experimental cages (95 × 70 × 120 cm) were screened with anti-thrips polyethylene mesh, and accessed by a separate door secured with a zipper.
2.4. Experimental Design
This experiment was conducted with four treatments (including control), each containing. Five tomato plants (30 cm in height) were placed inside each of the screened cages and one pair of adult T. absoluta per plant was released in each cage immediately after planting. In treatment 1 (T1), namely Tuta + Trichogramma, 20 adult Trichogramma females per plant was released in the cages one day following T. absoluta mating. In treatment 2 (T2), namely Tuta + Bt, 25 ml per plant of the Bt suspension (with a concentration of 106 spore/ml) was spra- yed on plant surfaces two times, once at the beginning of the experiment before T. absoluta release and the other one week later. In treatment 3 (T3), namely Tuta + Trichogramma + Bt, the release of Trichogramma wasps and the spraying of Bt suspension was simultaneously conducted exactly like what described for treatment 1 and 2, respectively. Finally, in treatment 4 (T4, i.e. the control), the plants were sprayed two times with water, with no release of Trichogramma wasps and no Bt treatment.
2.5. Sampling
Sampling was conducted for two week, 2 times. The presence of pest eggs was sampled separately. In the first treatment, 15 leaves for each replicate was selected at random from the upper, middle, and bottom third of tomato plants experimented in each treatment after 5 days from Trichogramma sp. release. Data were recorded two times once; at the fifth day after the beginning of the experiment and the other after two weeks. The parasitism rate, adult emergence rate percentage, number of females and adult longevity were assessed. Second treatment sampling was conducted for two weeks, 15 leaves selected at random from the upper, middle, and bottom third of the plant tomato plants experimented in each treatment. In the third treatment, 15 leaves for each replicate selected at random from the upper, middle, and bottom third of tomato plants experimented in each treatment after 5 days from Trichogramma sp. release. Data were recorded twice; first record after 5 days from 1st week and the second record after 2nd week. The rate of egg parasitism, adult emergence percentage, number of females and adult longevity were assessed. In the fourth treatment, 15 leaves (total 150 leaves) recently expanded were selected at random from the upper, middle, and bottom third of the tomato plants experimented in each treatment, in order to determine the preferred plant section for oviposition of T. absoluta. The leaves were collected, labeled, and transported to the laboratory and then examined under a stereoscopic microscope to determine whether eggs had hatched or had been killed. In addition, three leaves, selected along each of the five selected plants were examined for the presence of T. absoluta larvale mines from a randomly selected upper, middle and low in all treatments.
3. Statistical Analysis
For statistical analysis, each mean value is given with its standard error. The differences in parasitism rate, adult emergence, number of females and adult longevity between two Trichogramma species were compared using one way analysis of variance (ANOVA). When significant differences were detected, further comparisons were run using the Duncan’ test (p ≤ 0.05). Univariate general linear model (GLM) analysis was performed for comparison of the number of mines per plant. Statistical analysis was performed using the software SPSS Statistics for Windows Version 21 (SPSS 2012).
4. Results
The average number of T. absoluta larval mines on tomato leaves treated with Trichogramma species as well as Bt suspension have been summarized in Figure 1. Significant difference was observed in the number of larval mines of plants treated with T. brassicae and T. embryophagum alone or in combination with B. thuringiensis (F = 78.385; df = 5; p ≤ 0.001), Additionally there was significant difference in the number of larval mines between two weeks in nearly all experimental treatments (F = 4.524; df = 1; p ≤ 0.050). In first treatment (T1), an average of 18.2 and 12.10 mines/plant of T. absoluta were recorded during the first week when T. brassicae and T. embryophagum were alonely released respectively. A significant increase in the number of mines per plant was recorded in the scond week after treatment. When B. thuringiensis was used in combination with each of T. brassicae and T. embryophagum the number of larval mines was recorded as 10.50 and 2.80 mines/plant during the first week, respectively, However, in contrast to the application of Trichogramma alone, the integration of Bt and Trichogramma resulted in a significant decrease in the average number of larval mines in the second week, such that an average number of 6.10 and 0.50 mines/plant was recorded for T. brassicae and T. embryophagum respectively. Indeed, the mean number of T. absoluta mines in the treatment with parasitoid releases and B. thuringiensis was significantly lower than in the control (T4) (Figure 1). In particular, the best results in relation to T. absoluta control were obtained in the treatment where T. embryophagum was released in combination with B. thuringiensis application. The biological characteristics of T. brassicae and T. embryophagum have been presented in Figure 2-6. As the Figure 2, the results shows significant differences were observed among parasitism percentages of T. brassicae and T. embryophagum in presence of B. thuringiensis (F = 4.617; df = 3; p ≤ 0.010), with the highest parasitism rate was obtained for T. embryophagum with Bt was 31.18%. Otherwise, there was no significant difference
![]()
Figure 1. Mean number of Tuta absoluta mines per plant according to release T. brassicae and T. embryophagum in comparison to the control, in the first and second week (Treatments: F = 78.385, df = 5, p = 0.000) (Weeks: F = 4.524, df = 1, p = 0.036).
![]()
Figure 2. The parasitism of T. brassicae and T. embryophagum when treated with/ without Bt reared in the T. absolutaeggs incage inside greenhouse (F = 4.617, df = 3, p = 0.008).
in parasitism rate of other treatments. Additionally, we found no significant differences among mortality of two parasitoid species in presence or absence of Bt treatment (F = 1.365; df = 3; p ≤ 0.269). In this experiment, an average mortality of 21.92 and 37.25% was recorded for T. brassicae in presence and absence of B. thuringiensis, respectively, while the mortality of T. embryophagum was recorded as 24.80, 26.84% in presence and absence of B. thuringiensis respectively (Figure 3). Similarly, there was no significant difference in the emergence of adult rate of both was pspecies in presence or absence of B. thuringiensis (78.07 and 62.74% for T. brassicae and 75.18 and 73.85% for T. embryophagum in presence and absence of B. thuringiensis, respectively) (F = 1.408; df = 3; p ≤ 0.256) p = 0.256) (Figure 4). The ratio of females in untreated plants was significantly higher than that of Bt treated ones in both Trichogramma species (68.54% and 48.83% for T. brassicae and 67.54% and 43.63% for T. embryophagum in absence and presence of Bt respectively (F = 7.782; df = 3; p ≤ 0.001) (Figure 5). The number of females of two T. richogramma species is shown in Figure 5. Finally, we found significant difference in the longevity of adult parasitic wasps in presence or absence of Bt (F = 12.504; df = 3; p ≤ 0.000). The longevity of T. embryophagum adults reared on Bt treated eggs of T. absoluta (6.4 d) was significantly higher than that of untreated plants (5.6 d) (Figure 6). Although, the longevity of T. brassicae was not affected by Bt treatment, it was significantly lower than T. embryophagum (Figure 6).
![]()
Figure 3. The Mortality of T. brassicae and T. embryophagum when treated with/ without Bt reared in the T. absolutaeggs incage inside greenhouse (F = 1.365, df = 3, p = 0.269).
![]()
Figure 4. The adult emergence of T. brassicae and T. embryophagum when treated with/without Bt reared in the T. absoluta eggs in cage inside green- house (F = 1.408, df = 3, p = 0.256).
![]()
Figure 5. Females of T. brassicae and T. embryophagum when treated with/ without Bt reared in the T. absoluta eggs in cage inside greenhouse (F = 7.782, df = 3, p = 0.000).
![]()
Figure 6. Longevity of T. brassicae and T. embryophagum when treated with/without Bt reared in the T. absoluta eggs in cage inside greenhouse (F = 12.504, df = 3, p = 0.000).
5. Discussion
The present study showed an efficiency of T. brassicae and T. embryophagum released in combination with B. thuringiensis in controlling the damage of T. absoluta, when compared to the untreated control plants, Trichogramma sp. release and spray with B. thuringiensis only. The decrease of eggs was higher than the larvae in the first treatment when T. brassicae and T. embryophagum were released in the first week, which was consistent with. Results reported by [20] . However, significant differences between the two species were consistently observed in biocontrol activity against T. absoluta, where T. embryophagum performed slightly well than T. brassicae in the first week when released without B. thuringiensis, and better in the second week when released with B. thuringiensis. These results are in agreement with earlier investigations that demonstrated that the integration of these two organisms (Trichogramma sp. +B. thuringiensis) may produce desirable results in T. absoluta control [9] [21] . Also, compared with the control treatment, the decrease of mines numbers of T. absoluta was higher in Trichogramma sp. + B. thuringiensis treatments. The average number of larval mines per plant of mines per plant in Trichogramma sp. + B. thuringiensis treatments was significantly lower than that of Trichogramma release without Bt treatment.
Consistent with the previous findings, the data collected on 1st week after the treatment indicated that the lowest number of H. armigera larvae per plant was recorded in treatment (Trichocard having 300 parasitized eggs in combination with Neem extract) and treatment (Trichocard having 300 parasitized eggs in combination with 45 Chrysoperla 2nd instar larvae and neem extract) recorded 0.50 and 0.60 respectively [22] .
In this study, the parasitism rate of two Trichogramma species were tested in cage experiments inside greenhouse. Results showed that both T. richogramma species accepted T. absoluta eggs as host and the species of T. embryophagum exhibited promising (31.18%) parasitism when treated in combination with Bt vs. 23.05% of parasitism without Bt. Our result were in line with results obtained by Chailleux et al. [23] that investigated twenty-nine Trichogramma species- strains on T. absoluta eggs under laboratory conditions and recorded the highest rate of egg parasitism (35.4%) for T. achaeae. Furthermore, we found a significant increase in the number of parasitized eggs when the tomato plants were simultaneously treated with Bt. These results were significantly different by Vaez et al. [21] suggested that the functional response parameters of Trichogramma wasps were affected by Bt treatment such that those parasitic wasps which developed inside Bt-treated hosts were weaker than normal ones. They attributed these observations to the sub-lethal effects of Bt on T. brassicae females searching host eggs laid by H. armigera treated by LC20 of the bacterium. Moreover, studies have suggested that larger host eggs can positively influence parameters such as percentage of parasitism, number of parasitoids emerged per egg, longevity, and sex ratio [24] [25] [26] . The parasitism rate of Trichogramma wasps have been argued to vary with physical and chemical barriers as well as by the type and characteristics of host eggs such as their size, hardness, scales and kairomones [17] [19] [27] [28] . Similarly, the emergence rate of Trichogramma adults can also vary with the size and quality of the host egg, number of parasitoids that develop per egg, development period in host eggs and temperature [29] [30] . Adult longevity of T. embryophagum reared on eggs of Ephestia kuehniella Zeller has been reported as 12.37 days by [31] . These differences show that the host species may also influence adult longevity of parasitoid wasps. Also, reduction in host fecundity may show inverse effects of bioinsecticides on yolk deposition process and this may reduce both the quality and quantity of the eggs deposited by host as well as their size [21] , Therefore amongst the sub-lethal effects of microbial insecticides on hosts themselves are prolonged larval development, decreased adult body size, reduced pupal weight and reduced fecundity. Moreover, these events may lead to smaller parasitoids with lower fecundity, survival and searching ability [32] [33] [34] .
6. Conclusions
The results will undoubtedly be supported by the current scientific and applied research conducted with parasitoids and entomopathogens, which will help in managing T. absolutea in those areas or crop cycles. In this scenario, augmentative and conservation strategies intending to increase the role of these natural enemies in T. absoluta control could become the cornerstone in the reconfiguration of integrated pest management in the tomato crop.
This study showed that different species of Trichogramma wasps have parasitized eggs of T. absoluta as one of their herbivores. Additionally, we found no direct effect of Bt treatment on parasitism rate and some other biological characteristics of Trichogramma species. These findings may justify the combination of these two biocontrol organisms, especially T. embryophagum as a promising strategy for integrated management of T. absoluta under greenhouse conditions.
Acknowledgements
The authors wish to thank the Department of Plant Protection, University of Tehran, Karaj, Iran and the University of Basrah, Basrah, Iraq for providing the fund for this research work.