Toxicity and Lethal Effect of Greenhouse Insecticides on Coccinella septempunctata (Coleoptera: Coccinellidae) as Biological Control Agent of Myzus persicae (Hemiptera: Aphididae)

Deltamethrin and imidacloprid are commonly used insecticides for controlling sub-sucking insects in greenhouses. However, their application may cause sublethal effects on the aphid coccinellid predator Coccinella septempunctata (Coleoptera: Coccinellidae). Here, we study (i) the toxicity and the effect of two sublethal doses (LD10 and LD30) of deltamethrin and imidacloprid on C. septempunctata in a laboratory microcosm and (ii) the residual toxicity of the two insecticides in a greenhouse. The results showed that both insecticides reduced fecundity, longevity, the intrinsic rate of increase, the finite rate of increase and the net reproductive rate. However, the developmental time of the fourth instar larvae was prolonged by both insecticides at LD10 and LD30. Deltamethrin residues were toxic 21 DAT (days after treatment) to C. septempunctata fourth instar larvae. In contrast, imidacloprid began in the slightly harmful category (75%) 1 DAT and declined to the harmless category (18.33%) 21 DAT. These results indicate that deltamethrin and imidacloprid have potential risks to C. septempunctata. This study provides information to guide the development of integrated pest management (IPM) strategies in greenhouses.


Introduction
Aphids (Hemiptera: Aphididae) comprise an insect group that feeds on a wide range of plant species. In greenhouse crops, aphids are amongst the most dominant and destructive pests, causing significant losses in quality and/or yield [1]. The green peach aphid, Myzus persicae (Sulzer) (Hemiptera: Aphididae), is extremely polyphagous, with great efficiency as a virus vector [2]. Myzus persicae can seriously damage crops by feeding on the vascular bundles of plants and/or by transmitting more than 100 plant viruses [3]. Chemical insecticides are widely used to control aphids and especially M. persicae [4]. However, long-term exposure of agricultural systems to insecticides has led to the development of resistance to many classes of insecticides, including pyrethroids and neonicotinoids [5,6].
Concerns about insecticide resistance development and the rapid emergence of insecticide resistance of M. persicae populations to new active ingredients [7] or insecticides with different MoA have increased the interest in integrated pest management (IPM) adoption for aphid control [4,6,8].
Biological control of aphids with parasitoids and/or predators is a critical component of IPM programs. Coccinellid predators are often utilized in greenhouses for aphid management [9], since many species can reduce aphid populations in greenhouses [10,11]. Coccinellid predator release can reduce the aphid population and is considered an alternative to insecticide applications [9,12].
The seven-spotted lady beetle, Coccinella septempunctata L. (Coleoptera: Coccinellidae), is an excellent biological control agent because it preys on more than 20 aphid species of Coccoidea, as well as species of Psylloidea and Tetranychidae, that infest crops both in the fields and in greenhouses [13] and can be found in a wide range of agricultural and/or natural habitats or crops all over the world [14]. Larvae and adults of C. septempunctata preying on aphids pose a major advantage since they could prevent aphid density increases from reaching the economic thresholds (ETs). However, when aphid population density exceeds the ET, the use of selective pesticides is critical for preserving C. septempunctata populations and achieving successful aphid predation rates [15].
Two of the most commonly used classes of insecticides applied to control sap-sucking pests in greenhouses and fields are neonicotinoids and pyrethroid [4,16]. However, in the European Union (EU), the neonicotinoids, imidacloprid, clothianidin and thiamethoxam in 2017 and thiacloprid in 2020 were banned for all outdoor uses [17,18] due to chronic adverse effects on honeybees [19]. In the present study, we aimed to evaluate the toxicity, developmental time, survival and adults' longevity by examining the pre-oviposition period (APOP), total pre-oviposition period (TPOP), fecundity and population growth parameters of C. septempunctata exposed to lethal and sublethal doses of imidacloprid and deltamethrin. Additionally, in this study, we tried to examine this assumption of differential mortality by directly assessing the residual toxicity of imidacloprid and deltamethrin of C. septempunctata in the greenhouse. The present findings could assist the conservation of C. septempunctata and the regulation of deltamethrin and imidacloprid in IPM programs in greenhouses.

Insect Rearing
In total, more than 200 adults of C. septempunctata were collected from tobacco and peach fields infested by M. persicae in Katerini, Greece, and transferred to the Laboratory of Agricultural Entomology & Zoology of the University of the Peloponnese (Kalamata City, Greece). The green peach aphid, M. persicae, was used to rear C. septempunctata. Laboratory cultures of C. septempunctata were held on Chinese cabbage Brassica rapa pekinensis Hanelt (Brassicaceae) plants in cylindrical acrylic glass cages (DL: 30 × 50 cm) at 25 ± 1 • C and a relative humidity (RH) of 60-70% with a photoperiod of 16:8 h light: darkness (L:D).

Dose Response to Topical Application Bioassays
Independent assays were performed to find the acute toxicity of deltamethrin and imidacloprid against 4th instar larvae of C. septempunctata. The assays were carried out by topical application following a modified protocol of the method of [20,21]. Imidacloprid was dissolved in acetone to prepare a concentration gradient of 100, 200, 400, 600, 800, 1000, 1200 and 1600 ng of active ingredient per insect. The corresponding values for deltamethrin were 0.20, 0.40, 0.80, 1.60, 3.20, 4.80 and 6.40 ng of active ingredient per insect. Fourth instar larvae of C. septempunctata (<24 h old) were transferred to a Blackman box [22], and 1 µL of insecticide at each concentration was applied to the mesonotum of C. septempunctata using a 10 µL Hamilton microsyringe. Larvae treated with 1 µL acetone alone were set as control. Each larva was placed separately in a Blackman box with more than five hundred live M. persicae aphids. There were seventeen treatments including control, and every treatment was repeated three times using twenty larvae per treatment. Treated larvae were maintained in a controlled environmental chamber at 23 ± 1 • C, L16:D8 and 50 ± 5% RH. Mortality data for C. septempunctata were scored after 72 h. Larvae were considered dead if they did not move when gently pushed by a fine brush.

Evaluation of Low and Sublethal Effects on Fourth Instar Larvae
To calculate the life table parameters for C. septempunctata, a total of 361 eggs (12-24 h old) were randomly collected and maintained in Petri dishes (9 cm diameter). We used 65 eggs for the control group, 70 and 75 eggs in the deltamethrin LD 10 and LD 30 groups, respectively, and 71 and 80 eggs in the imidacloprid LD 10 and LD 30 groups, respectively. The hatch rate and incubation period of C. septempunctata eggs were recorded daily. After egg hatching, each first instar larva was transferred individually into a Blackman box, at the base of which there was a piece of water-saturated moss. Mortality and development time were recorded daily until the fourth instar larvae. To assess the low and sublethal effects on C. septempunctata population parameters, fourth instar larvae (<24 h old) topically treated (as described in the dose response to topical application bioassays section) to LD 10 and LD 30 doses of deltamethrin (0.34 ng a.i. and 0.63 ng a.i. per insect, respectively) and to LD 10 and LD 30 doses of imidacloprid (357.96 ng a.i. and 519.13 ng a.i. per insect, respectively), which were calculated from the toxicity regression equation, and as a control, we used only acetone (65 eggs). One fourth instar larva was released in each Blackman box. Larvae were reared and treated as noted in the topical bioassays section. After exposure to the insecticide, development time and mortality were recorded daily until the emergence of adults. After eclosion, adult females and males of each dose were randomly paired and transferred to a new Blackman box. Each pair was fed ad libitum with M. persicae every 24 h. The survival rate and fecundity were scored daily until all adults were dead. All insects were maintained in a controlled environmental chamber at 23 ± 1 • C, L16:D8 and 50 ± 5% RH.

Greenhouse Residual Toxicity Test
Seven hundred and twenty plants of Capsicum annuum L. were maintained in a greenhouse at 20 ± 3 • C with additional light (16 h light: 8 h dark) and were grown individually (Supplementary File). Plants at the five-leaf stage were used for the experiment. Pepper plants were sprayed (at the highest dose recommended on the label) until run off with deltamethrin (17.5 mg a.i L −1 ) or imidacloprid (60 mg a.i L −1 ) or tap water for control plants using a manual backpack sprayer. Bioassays on pepper plants 1, 3, 10 and 21 days after insecticide application (DAT) were conducted by placing fourth instar larvae (<24 h old) of C. septempunctata. In addition, about more than 300 frozen M. persicae aphids were placed on each plant to assure that all larvae had access to food during the experiment, and each plant-larvae system was covered by muslin to avoid insects' escape. Each plant counted as a replicate. A total of sixty larvae per treatment were used. Larvae mortality was recorded 3 and 7 days after exposure.

Statistical Analysis
The predator dose-mortality relationship of LD 50 and sublethal (LD 10 and LD 30 ), doses of the insecticides, 95% confidence intervals (CI), slopes and chi-square were calculated by probit analysis using SPSS version 25.0 (SPSS Inc., Chicago, IL, USA). The software two-sex MSChart [23] was used to analyze the life history data of C. septempunctata after exposure to sublethal doses of the insecticides imidacloprid and deltamethrin, based on the age-stage two-sex life table theory [24,25]. The life table parameters (l x ), (m x ), (e xj ), (v xj ) and (s xj ) (age-specific survival rate, age-specific fecundity, age-stage life expectancy, age-stage reproductive value and age-stage survival rate, respectively) were calculated. The pre-adult developmental duration time, pre-oviposition and total pre-oviposition period (APOP and TPOP, respectively) fecundity, male and female duration time, intrinsic rate of increase (r), finite rate of increase (λ), net reproductive rate (R 0 ) and mean generation time (T) were also calculated. The paired bootstrap test was used to analyze the difference among each treatment group for all population parameters [26].
The residual effect of each insecticide on C. septempunctata mortality was compared using the χ 2 test. When χ 2 was significant, pairwise comparisons were performed using the Bonferroni correction.

Toxicity of Deltamethrin and Imidacloprid to C. septempunctata
The toxicity of deltamethrin and imidacloprid against fourth instar larvae of C. septempunctata was determined after 3 days ( Table 1). The LD 50 , LD 10

Sublethal Effects of Deltamethrin and Imidacloprid on C. septempunctata
The developmental time of fourth instar larvae of C. septempunctata at all doses was significantly longer than that of the control group (Table 2). There were no significant differences between the deltamethrin and imidacloprid LD 10 and LD 30 groups. Furthermore, the developmental time of pupa did not differ in deltamethrin LD 30 , two imidacloprid and the control groups, while deltamethrin LD 10 resulted in a significantly lower developmental time (4.64 days) than the control group. Table 2. Sublethal effects of imidacloprid and deltamethrin on the developmental time (mean ± SE) of C. septempunctata adults exposed to insecticide from the fourth instar larval stage. Adult longevity was significantly shorter in imidacloprid and deltamethrin treatments than in the control treatment (Table 3). There were no significant differences between imidacloprid and deltamethrin treatments in terms of female adult longevity. In addition, female adult longevity was significantly shorter in the imidacloprid treatments compared with the control treatment. Male adult longevity did not differ significantly between imidacloprid LD 10 (61.23 days) and the control treatment (70.11 days). Male adult longevity was significantly lower in the treatments of deltamethrin LD 30 and LD 10 (52.67 and 50.79 days, respectively) and imidacloprid LD 30 (50.33 days) compared to the control treatment (70.11 days). No significant differences were recorded between the imidacloprid, deltamethrin and control group in terms of TPOP and APOP ( Table 3). The number of eggs laid by females of C. septempunctata decreased by 52.07% and 69.89% in imidacloprid LD 10 and LD 30 doses, respectively, and by 57.53% and 31.72% in deltamethrin LD 10 and LD 30 doses, respectively, when compared with the control treatment. Furthermore, female fecundity did not differ significantly between the deltamethrin LD 30 treatment and the control treatment, while imidacloprid treatments (LD 10 and LD 30 doses) and deltamethrin (LD 10 dose) resulted in significantly reduced fecundity compared to the control treatment. Table 3. Sublethal effects of imidacloprid and deltamethrin on the life parameters (mean ± SE) of C. septempunctata adults exposed to insecticide from the fourth instar larval stage. The population growth parameters are shown in Table 4. Treatment with imidacloprid had a significant effect on the finite rate of increase (λ) and intrinsic rate of increase (r) compared to those in the control group. Treatment with deltamethrin and imidacloprid resulted in a significantly smaller net reproductive rate (R 0 ) in comparison to the control group. However, the mean generation time (T) had no significant effect between treatments. Table 4. Sublethal effects of imidacloprid and deltamethrin on the population growth parameters (mean ± SE) of C. septempunctata adults exposed to insecticide from the fourth instar larval stage.  Figure 1 presents the fecundity of the total population (m x ), age-specific survival rate (l x ) and the net maternity (l x m x ) of C. septempunctata. The analysis of the age-specific survival rate, l x , demonstrates a more rapid decrease in the deltamethrin and imidacloprid treatment groups than in the control group. The highest m x (16.8) and l x m x (9.1) in the control group occurred on day 41. In comparison, the lowest m x (6.6) peak occurred at age 42 days in the deltamethrin LD 10 group, while the l x m x (1.7) peak was recorded at 43 days in the imidacloprid LD 30 treatment group. Age-specific survival rate (l x ), age-specific fecundity (m x ), and age-specific maternity (l x m x ) after fourth instar C. septempunctata larvae exposed to sublethal deltamethrin and imidacloprid doses.
The age-stage life expectancy e xj of C. septempunctata is shown in Figure 2. All individuals treated by imidacloprid or deltamethrin have a lower e xj than the control group. For example, a newly hatched C. septempunctata egg was supposed to survive 51.67 days in the control group compared to the life expectancies in deltamethrin LD 10 and LD 30 groups, which were 34.69 and 33.28 days, respectively. The corresponding values for imidacloprid were 34.55 and 27.86 days in the LD 10 and LD 30 treatments, respectively.
The age-stage reproductive value (Vxj) of newly hatched C. septempunctata eggs was significantly lower in the imidacloprid LD10 and LD30 treatments compared to the control and deltamethrin groups (Figure 3). The Vxj began to increase when females started to produce offspring. The peak Vxj value of the untreated control C. septempunctata was 192.58 days −1 at 40 days. In the deltamethrin LD10 and LD30 groups, the peak Vxj values were 83.08 Life expectancy (e xj ) values after fourth instar C. septempunctata larvae exposed to sublethal deltamethrin and imidacloprid doses.
The age-stage reproductive value (V xj ) of newly hatched C. septempunctata eggs was significantly lower in the imidacloprid LD 10 and LD 30 treatments compared to the control and deltamethrin groups (Figure 3). The V xj began to increase when females started to produce offspring. The peak V xj value of the untreated control C. septempunctata was 192.58 days −1 at 40 days. In the deltamethrin LD 10 and LD 30 groups, the peak V xj values were 83.08 and 105.95 days −1 found at 42 and 85 days, respectively. In the imidacloprid LD 10 and LD 30 groups, the corresponding values were 118.80 and 106.01 days −1 found at 54 and 94 days, respectively.  The age-stage survival rates (Sxj) of C. septempunctata were negatively affected in the deltamethrin and imidacloprid treatment groups compared to the control group ( Figure  4). The peak Sxj values for male and female adults in the control group were 0.28 and 0.26, Figure 3. Age-stage specific reproductive values (V xj ) values after fourth instar C. septempunctata larvae exposed to sublethal deltamethrin and imidacloprid doses.
The age-stage survival rates (S xj ) of C. septempunctata were negatively affected in the deltamethrin and imidacloprid treatment groups compared to the control group (Figure 4). The peak S xj values for male and female adults in the control group were 0.28 and 0.26, respectively. The peak S xj values for male and female adults treated with deltamethrin (i.e., LD 10 : 0.20 for males and females; LD 30 : 0.20 for males and 0.17 for females) and imidacloprid (i.e., LD 10 : 0.18 for males and females; LD 30 : 0.15 for males and females) decreased by increasing the insecticide doses.  . Age-stage-specific survival rate (Sxj) after fourth instar C. septempunctata larvae exposed to sublethal deltamethrin and imidacloprid doses.
The projection population size of C. septempunctata at 140 days following different insecticide treatments is shown in Figure 5. The population size of C. septempunctata after 140 days in the control group was projected to be 7.6-fold greater than the initial Figure 4. Age-stage-specific survival rate (S xj ) after fourth instar C. septempunctata larvae exposed to sublethal deltamethrin and imidacloprid doses.
The projection population size of C. septempunctata at 140 days following different insecticide treatments is shown in Figure 5. The population size of C. septempunctata after 140 days in the control group was projected to be 7.6-fold greater than the initial population. The corresponding values for deltamethrin were 6.8 and 6.7 in the LD 10 and LD 30 groups, respectively. Population size was also affected by imidacloprid at 140 days in the LD 10 (5.6-fold) and LD 30 (5.0-fold) groups.
Toxics 2023, 11, x FOR PEER REVIEW 11 population. The corresponding values for deltamethrin were 6.8 and 6.7 in the LD10 LD30 groups, respectively. Population size was also affected by imidacloprid at 140 d in the LD10 (5.6-fold) and LD30 (5.0-fold) groups.

Figure 5.
Population projection for C. septempunctata larvae exposed to LD10 and LD30 doses of tamethrin and imidacloprid.

Discussion
Combining biological control and insecticide use in IPM programs requires information on how insecticides affect not only the target pest but also their natural enemies [27,28]. In the present study, the toxicity and the sublethal and residual effect of deltamethrin and imidacloprid on the seven-spot ladybeetle, C. septempunctata, were examined. Pyrethroid and neonicotinoid insecticides can have lethal and sublethal effects on coccinellid predators [28][29][30][31][32][33][34][35][36][37][38][39]. Deltamethrin and imidacloprid, which are frequently used insecticides in greenhouses to control aphids, have direct toxic effects on the fourth instar larvae of C. septempunctata, but among the two insecticides tested, deltamethrin was much more toxic than imidacloprid. Deltamethrin, a widely used pyrethroid, was found to be 685 times more toxic for the fourth instar larvae of C. septempunctata than imidacloprid. This difference in toxicity between the two insecticides was likely due to the resistance development in C. septempunctata to imidacloprid due to its frequent use to control M. persicae in tobacco and peach orchards for more than three decades in northern Greece [5,40], in addition to detoxification enzymes or activity in target site sensitivity by each insecticide [38]. The same toxic results to coccinellid predators by pyrethroids have been reported by many researchers [29,33,41] in Adalia bipunctata (L.) and Ceratomegilla undecimnotata (Schneider, 1792) [37]. Furthermore, residues of the insecticide deltamethrin were more toxic for the fourth instar larvae of C. septempunctata than those of imidacloprid. The mortality of the fourth instar larvae of C. septempunctata for imidacloprid began in the slightly harmful category (75%) 1 DAT and declined to the harmless category (18.33%) 21 DAT. In contrast, deltamethrin was placed in the harmful category (100%) during the entire examined period. Briefly, deltamethrin was toxic to biological control agents for aphid control in greenhouses, and the number of days for release after treatment should be carefully considered when using C. septempunctata in IPM programs.
A sublethal dose of deltamethrin and imidacloprid increased the developmental time of fourth instar larvae of C. septempunctata. Prolonged developmental time has been reported by several authors in C. septempunctata after fourth instar larvae are treated with sublethal doses of imidacloprid [42,43]. Moreover, increased developmental time was caused by imidacloprid in fourth instar larvae of C. undecimnotata and Hippodamia variegata (Goeze) (Coleoptera: Coccinellidae) [36,37]. Increased developmental time was caused by bifenthrin to the second instar larvae of C. septempunctata [35]. The prolonged developmental time compared to the control group may be due to the fact that deltamethrinor imidacloprid-treated larvae groups used their energy to detoxify the insecticides rather than for their development [44] or/and decreased pest consumption and, as a result, reduced their energy supply [30,31,36].
Although the duration of fourth instar larvae was impressively increased by the sublethal doses (LD 10 and LD 30 ) of deltamethrin and imidacloprid, the adult total longevity decreased compared to the control group. Furthermore, fecundity was significantly reduced after treatment with the LD 10 of imidacloprid and deltamethrin and the LD 30 of imidacloprid. These results are similar to those of other studies, where imidacloprid and deltamethrin decreased the adult longevity and fecundity of coccinellid predators [36,37,39,45]. The decreased fecundity could be attributed to the reduced food intake at the fourth instar larvae of C. septempunctata due to the insecticide treatment, which affected adult fitness [38,46]. Decreased fecundity may be based on the direct toxic effect of the insecticide and/or malformations of organs [47]. In addition to fecundity, the developmental time and longevity, value of the intrinsic rate of increase (r), net reproductive rate (R 0 ) and finite rate of increase (λ) can be useful for understanding the predator coccinellid population dynamics. Our results showed that the r, λ and R 0 of C. septempunctata decreased under the sublethal doses of imidacloprid compared to the control group. Sublethal doses of deltamethrin decreased the R 0 of C. septempunctata compared to the control group. The results indicate that both sublethal doses of imidacloprid and deltamethrin can produce detrimental effects on the physiology of C. septempunctata [34]. Sublethal doses of neonicotinoids have been reported to lower population growth parameters for many coccinellid predators and/or other insects [30,32,36,37,39,47].
Furthermore, in our study, we found that imidacloprid and deltamethrin sublethal doses affected the two-sex life table parameters of C. septempunctata. Values such as m x , l x , l x m x , e xj , V xj and S xj show a decreasing pattern, indicating that the life table parameters were affected by deltamethrin and imidacloprid at sublethal doses of LD 10 and LD 30 . The reduction in the life table parameters such as m x and l x might be due to the fact that insecticides kill the more sensitive individuals, while the more resistant reduce prey consumption [32,38,46,48]. This result supports the idea that the reduced population parameters of the coccinellid predator verify that sublethal doses of deltamethrin and imidacloprid can reduce the survival and reproduction, thereby minimizing its efficacy as an aphid predator in greenhouse IPM strategies.

Conclusions
In conclusion, our study showed that the residual toxicity of C. septempunctata varied between imidacloprid and deltamethrin. Imidacloprid had lower residual toxicity than deltamethrin to fourth instar larvae of C. septempunctata. In the present laboratory experiments, deltamethrin and imidacloprid had negative effects on the population parameters and survival of C. septempunctata. These findings indicate that both insecticides should not be preferred for IPM programs of M. persicae and other aphids. In particular, deltamethrin should be avoided due to its extreme toxicity to fourth instar larvae of C. septempunctata, which lasts for up to three weeks after application. These data provide the basis for new studies on the residual toxicity of the insecticides tested on coccinellid predators beyond 21 DAT performed under greenhouse conditions.