Augmenting Nesidiocoris tenuis (Nesidiocoris) with a Factitious Diet of Artemia Cysts to Control Bemisia tabaci (Gennadius) on Tomato Plants under Greenhouse Conditions

Simple Summary Nesidiocoris tenuis (Hemiptera: Miridae) is an important biocontrol agent of several key arthropod pests, such as greenhouse and tobacco whiteflies, the South American tomato pinworm, thrips, plant mites, and other pests in the greenhouse. However, optimizing N. tenuis utilization in the greenhouse to control greenhouse pests such as whiteflies still needs further studies, especially in Japan. Here, we showed that factitious supplementary dietary in the form of Artemia cysts enhanced with high fructose corn syrup and honey, and delivered using a hemp rope could promote N. tenuis proliferation and spread among tomato plants. Nesidiocoris tenuis spread among tomato plants therefore, can maintain whitefly (Bemisia tabaci) eggs and nymph numbers at minimum in the greenhouse conditions. Abstract Natural predators such as Nesidiocoris tenuis are known for their role in managing greenhouse pests. However, techniques in maximizing the biological control potential of N. tenuis under field conditions are still lacking. We evaluated under greenhouse conditions the prospects of Artemia cysts enhanced with high fructose corn syrup and honey, and delivered using hemp strings (hemp rope) as supplementary factitious dietary in augmenting the proliferation and spread of N. tenuis on tomato plants. Results showed that N. tenuis supplemented with hemp rope could establish, proliferate and disperse among tomato plants compared to the N. tenuis supplemented with banker plants. Even though N. tenuis proliferated exponentially on banker plants, their movement and relocation to tomato plants, as expected, were only congested on tomato plants near the banker plants. However, as the survey continued, they relocated to the rest of the tomato plants. Furthermore, the number of Bemisia tabaci eggs and nymphs, a serious greenhouse pest of tomato, was observed to be significantly reduced in hemp rope greenhouse compared to banker plants and the negative control (no pest control system) greenhouses. This study, therefore, establishes foundational data on the usage of Artemia cysts enhanced with isomerized sugar (high fructose corn syrup) and honey under greenhouse conditions as factitious supplementary dietary in supporting N. tenuis establishment and spread, traits that are essential towards development of whitefly Integrated Pest Management (IPM) system. enhanced with isomerized sugar (high fructose corn syrup) and honey.


Introduction
Tomato (Lycopersicon esculentum Mill.) production, almost always under the greenhouse conditions across the whole of Japan was estimated at around 724,200 tons in 2018 with 11,800 hectares of total land use, yielding productivity of around 6140 kg per 0.1 ha [1]. is known to prefer these plants and can proliferate on these plants in high numbers in a short period [9].
Nesidiocoris tenuis populations were reared on verbena, sesame, and cleome at Ibaraki Horticultural Research Institute insectary rooms under optimized conditions (25 ± 1 • C, 65 ± 10% RH, and 16:8 L:D photoperiod). Bemisia tabaci populations were originally collected from capsicum fields in Kamisu City (Ibaraki Prefecture, Japan) in 2011 and were sparingly maintained on eggplants (Solanum melongena), kidney bean (Phaseolus vulgaris L.), and green bell pepper (Capsicum annuum) as a food source at Ibaraki Horticultural Research Institute insectary following the methods described by Wari et al. [18].

Greenhouse Preparations and Experiment Settings
Field studies were performed in two consecutive years, 2019 and 2020. The experiments were conducted in the Horticultural Research Institute inside a 10 m by 5 m vinyl pipe greenhouse (See Appendix A Figure A1). Tomato Momotaro Peace cultivar, a TYLCV tolerant cultivar (Takii Seed Co., Ltd., Kyoto, Japan) was used in this study. Basic greenhouse preparations such as soil fumigation, plowing, rotary, fertilizer applications, bedding, and mulching were performed according to the tomato farming schedules practiced by Ibaraki Prefecture tomato farmers. In the case of this study, these basic greenhouse preparations were performed between April and May. Tomato seeds were sowed in early May, then two weeks later, transplanted onto vinyl pots (105 mm). After a month in vinyl pots, the 1.5 months old tomato seedlings were transplanted into the greenhouses. A total of 38 to 40 tomato plants per greenhouse were planted in two rows of beds (See Appendix A Figure A1). In the same week, banker plants (verbena, sesame, and cleome) used in propagating N. tenuis, were also planted in the greenhouse (negative control treatment, see below). Bemisia tabaci adults propagated in the Horticultural Research Institute insectary were released on tomato seedlings (two months after transplanting) at a rate of one B. tabaci adult per tomato plant. Two weeks after release of B. tabaci, N. tenuis was released onto tomato seedlings containing the hemp rope (for hemp rope treatment) and banker plants (for banker plants treatment) at the same rate, i.e., one N. tenuis (irrespective of the growth stage) per two tomato seedlings as recommended in the manual of NARO (National Agriculture Research Organization) manual [9]. Hemp rope was set on the same day N. tenuis was released. The assessment was started one month after the release of N. tenuis.

Experimental Treatment Settings
To examine if Artemia cysts enhanced with high fructose corn syrup and honey and delivered using a hemp rope could promote N. tenuis propagation, proliferation, and spread among the tomato plants in the greenhouse, thereby maximizing its predation potential, four test treatments were considered. These include (i) Hemp rope, (ii) Banker plants, (iii) Positive control, and (iv) Negative control treatment. In the Hemp rope treatment greenhouse, N. tenuis nymphs and adults were randomly released on tomato plants containing the hemp rope with rates as mentioned earlier. In a 10 m by 5 m vinyl pipe greenhouses, two rows of 8 m beds were raised as discussed above and hemp rope (2.5 m each) were set on each bed adjacent to each other to cover every tomato seedling ( Figure A1A,D). Banker plants are widely reported to provide shelter, food sources, and medium for breeding and, therefore, are advantageous for N. tenuis propagation and proliferation. Thus, a banker plants treatment greenhouse was included to compare the movement and relocation of N. tenuis. In the banker plant treatment greenhouse, N. tenuis was released together with the banker plants (verbena sesame and cleome). Banker plants were planted at one end of the vinyl pipe greenhouse as per the recommendation by NARO [9] in integrating N. tenuis and banker plants to control B. tabaci under greenhouse conditions.
To evaluate the role of N. tenuis in managing B. tabaci in the greenhouse, we included both a positive and a negative control treatment. Positive control treatment greenhouse involved the use of synthetic insecticides to control B. tabaci while negative control had no means of any pest control measures. Nesidiocoris tenuis was not released in positive and Insects 2021, 12, 265 4 of 13 negative control treatments. Bemisia tabaci was added in all treatments (i.e., hemp rope, banker plants, positive control, and negative control) to substantiate the role of N. tenuis when augmented with a factitious diet. As mentioned previously, this study was performed annually in two different years (i.e., 2019 and 2020) from April to October. Each annual year study was considered as replicates.

Sampling and Quantification of Test Subjects
In order to manage greenhouse pests that are almost always distributed among all the plants in the greenhouse, natural enemies must be scattered among all crops to maximize control impact. To determine if N. tenuis (both nymphs and adults) are scattered among the tomato plants in the greenhouse, we quantified the distribution of the N. tenuis individuals on all the tomato plants planted in the greenhouse. Tomato plants were grouped into clusters making four zones along the greenhouse. Nesidiocoris tenuis individuals on every tomato plants in each zone were quantified. Nesidiocoris tenuis individuals on tomato plants were quantified by visually inspecting each tomato plant in the greenhouse. Nesidiocoris tenuis nymphs and adults on the underside and topside of tomato leaves, as well as on the flower stalks and shoots, were also counted and recorded. To minimize errors in quantifying N. tenuis on tomato plants, as adults are prone to disperse when disturbed, disturbance of tomato plants were minimized. Furthermore, the entire counting was performed by one person for consistency and to keep track of the N. tenuis movement. Nesidiocoris tenuis nymph and adult counts for each zones (10~12 tomato plants per zone) were pooled together and the average determined to represent each zone. To determine the population dynamics of N. tenuis on tomato plants in the greenhouse throughout the study period, N. tenuis (both nymph and adult) counts from all tomato plants (40~48 tomato plants per greenhouse) were pooled together and the average calculated to represent the number of N. tenuis per tomato plant. The averages for each sampled week (among zones and all plants in the greenhouse) were treated as replicate for each treatment per annual study. Nesidiocoris tenuis on the banker plants was quantified by placing a 40 cm by 20 cm white bath container 10-15 cm beneath the banker plants and tapping ten times on the banker plants. Nesidiocoris. tenuis nymphs and adults that fell onto the white bath container were quantified and recorded. Sampling of N. tenuis on banker plants were performed separately for each banker plants, i.e., verbena, cleome and sesame.
Bemisia tabaci eggs, nymphs, and adults were enumerated following the methods described by Wari et al. [18]. In brief, twenty leaves per treatment were sampled every week. Eggs and nymphs (including 1st, 2nd, 3rd, and 4th instar nymphs) underneath the tomato leaves were counted using a Leica S6E stereo microscope mounted with a Leica KL300 LED illumination. Bemisia tabaci egg and nymph counts for the twenty leaves were pooled together and the average calculated to designate the B. tabaci egg and nymph counts per tomato leaf in a given treatment. The averages for each sampled week are treated as replicate for each treatment per annual study. One 200 cm 2 yellow sticky trap (Arysta Lifesciences, Tokyo, Japan) was set in the middle of each treatment and collected after three days to determine B. tabaci adult population. Sticky trap setting and collection (together with leaves) were performed on weekly basis for 13 weeks. After leaf and sticky trap sampling, synthetic insecticides were applied in the positive control greenhouse treatment as per the application schedule in Appendix A (Table A1).

Statistical Analyses
Nesidiocoris tenuis and B. tabaci (eggs, nymphs and adults separately) count were tested for normality (Shapiro-Wilk test and Lilliefors test) and homogeneity (Bartlett's Test) using an open-source software OpenStat (http://openstat.info/OpenStatMain.htm, accessed on 10 February 2021). The total number of N. tenuis (nymphs and adults) and the eggs, nymphs and adults of B. tabaci were transformed using 'x + 1' to meet the assumptions of analysis of variance (ANOVA). The effects of "treatment (hemp rope and banker plants)" and "sampling interval (weeks)" factors and their interaction on the population trends of N. tenuis on tomato were analyzed using univariate repeated-measures ANOVA. While the effect of N. tenuis on B. tabaci eggs, nymphs and adults was analyzed using a repeatedmeasures ANOVA model with treatment and sampling interval considered as fixed factors. Nesidiocoris tenuis distribution among the zones in the greenhouses between the hemp rope and banker plants treatments were analyzed using contingency chi-square test for two-way tables to test the relationship between treatments and zones. In addition, threeway contingency table for treatment, zones and weeks (sampling interval) was analyzed using log-linear regression model. Furthermore, standard deviations (SD) were calculated from averages of pooled data (40~48 tomato plants per greenhouse for N. tenuis counts per treatment, 10~12 tomato plants per zone for N. tenuis counts per zone, 20 tomato plant leaves per treatment for B. tabaci eggs and nymphs, and one 200 cm 2 yellow sticky trap per treatment for B. tabaci adults) per sampled week from two annual studies to represent the difference between the years.

Nesidiocoris tenuis Population Dynamics on Tomato Plants
A total of six weeks after N. tenuis was released onto the tomato plants with hemp rope, N. tenuis (inclusive of nymphs and adults) were observed to be gradually increasing in numbers per tomato plant. Significant increase could be seen on the fifth week from the start of the survey and subsequently peaking at week eight at 8.80 ± 0.76 individuals per tomato plant ( Figure 1). Comparatively, N. tenuis trends on tomato plants in the banker plant treatment greenhouse was stationary until week 6 of the survey, then peaking gradually until reaching its highest peak at the ninth week with 5.03 ± 0.47 individuals (nymphs and adults inclusive) per tomato plants ( Figure 1). Both hemp rope and banker plants promoted N. tenuis development on tomato plants. However, proliferation on tomato plants was delayed when N. tenuis was augmented with banker plants (Figure 1). A significant increase in N. tenuis during time post-release was observed on tomato plants in both treatments. However, the onset of the increase was significantly earlier in hemp rope treatment than in banker plants treatment, leading to the significant interaction in the sampling interval (p = 0.0001) ( Table 1).

Impact of N. tenuis on B. tabaci Eggs, Nymphs, and Adults on Tomato Plants
To examine if N. tenuis augmented with hemp rope could impact the population of B. tabaci, population densities of B. tabaci eggs, nymphs, and adults were determined. Results showed that the interaction between treatments, sampling intervals (weeks) or between treatments and sampling intervals in the hemp rope, banker plants, positive and negative control treatments did not significantly impact the number of B. tabaci eggs on tomato plants under greenhouse conditions (Table 2). Similarly, the interaction between treatments, sampling intervals (weeks) or between treatments and sampling intervals in the hemp rope vs. banker and hemp rope vs. positive control did not significantly affect B. tabaci nymphs and adults. However, interaction between treatments (hemp rope vs. negative control) yielded significant differences in the population densities of B. tabaci nymphs and adult in hemp rope vs. negative control (p ≤ 0.001 and p = 0.003, respectively) ( Table 2). Interaction between sampling intervals or between treatments and sampling intervals did not significantly affect B. tabaci nymphs and adults. These observations are reflective of the population densities of B. tabaci nymphs and adults between treatments shown in Figure 2 where B. tabaci nymphs and adults are shown to be considerably lower in the hemp rope, as well as positive control treatments compared to banker plants and negative control treatments. It can be deduced that the reduced number of B. tabaci nymphs in hemp rope treatment is the direct impact of N. tenuis while the reduced B. tabaci densities in positive control treatment is due to the aggressive insecticide application approach (Table A1).

Nesidiocoris tenuis Population Trends on Banker Plants
Banker plants are known for their role in augmenting the growth and development of natural enemies by providing them with shelter, food, and a habitat to lay eggs and propagate. To avoid early damage by N. tenuis on tomato seedling, banker plants (verbena, sesame, and cleome) were used. To ensure that N. tenuis is present and is moving onto tomato plants after propagating on banker plants, population trends of N. tenuis on banker plants were determined. The results showed that the number of N. tenuis in the banker plants increased as the survey progressed. While cleome and sesame contributed larger quantities of N. tenuis, the same was not observed for the verbena (Figure 3).  (Table 2). Interaction between sampling intervals or between treatments and sampling intervals did not significantly affect B. tabaci nymphs and adults. These observations are reflective of the population densities of B. tabaci nymphs and adults between treatments shown in Figure 2 where B. tabaci nymphs and adults are shown to be considerably lower in the hemp rope, as well as positive control treatments compared to banker plants and negative control treatments. It can be deduced that the reduced number of B. tabaci nymphs in hemp rope treatment is the direct impact of N. tenuis while the reduced B. tabaci densities in positive control treatment is due to the aggressive insecticide application approach (Table A1).  propagate. To avoid early damage by N. tenuis on tomato seedling, banker plants (verbena, sesame, and cleome) were used. To ensure that N. tenuis is present and is moving onto tomato plants after propagating on banker plants, population trends of N. tenuis on banker plants were determined. The results showed that the number of N. tenuis in the banker plants increased as the survey progressed. While cleome and sesame contributed larger quantities of N. tenuis, the same was not observed for the verbena (Figure 3).

Nesidiocoris tenuis Distribution among Tomato Plants
To determine if N. tenuis (both nymphs and adults) are scattered among the tomato plants in the greenhouse, we observed the distribution of the N. tenuis individuals on tomato plants. Nesidiocoris tenuis were observed to be equally distributed among the tomato

Nesidiocoris tenuis Distribution among Tomato Plants
To determine if N. tenuis (both nymphs and adults) are scattered among the tomato plants in the greenhouse, we observed the distribution of the N. tenuis individuals on tomato plants. Nesidiocoris tenuis were observed to be equally distributed among the tomato plants two weeks from the start of the survey when augmented with hemp rope. However, N. tenuis in the banker plant treatment greenhouse were unevenly distributed among the zones (Figure 4). This is evident as N. tenuis counts were highly significantly different (p ≤ 0.001) between the hemp rope and banker plants treatment as well as between the zones from week one of the survey until week seven (Table 3). On the other hand, there was only a moderate correlation (r = 0.576, r 2 = 0.3295) between the treatments (hemp rope and banker plants), zones (zone 1, 2, 3 and 4) and weeks (sampling intervals from week 1 to 13) implying that while there were significant differences between the treatments and the zones in the start of the survey, the significant differences diminished as the survey continued. To determine if different growth stages of N. tenuis counts (adults and nymphs) would provide insights into such trends, N. tenuis adults and nymphs were analyzed separately. While significant differences (p ≤ 0.001) between the hemp rope and banker plants treatment as well as between the zones could be observed, significant difference in N. tenuis adults were observed until week 7 of the survey while nymphs a week later ( Figure A1 and Table A1). This observation is of no surprise as the nymph emergence and or occurrences are expected to follow that of the adults. As expected, N. tenuis individuals clustered mainly in zone 1 in the first half of the study in the banker plant treatment as zone 1 was adjacent to the banker plants and progressively moved onto other zones as the study continued. On the other hand, N. tenuis were expected to scatter among the tomato plants in all zones in the hemp rope treatment. Presumably, the presence of the factitious diet during the release of N. tenuis assisted with their spread.  different (p = <0.001) between the hemp rope and banker plants treatment as well as between the zones from week one of the survey until week seven (Table 3). On the other hand, there was only a moderate correlation (r = 0.576, r 2 = 0.3295) between the treatments (hemp rope and banker plants), zones (zone 1, 2, 3 and 4) and weeks (sampling intervals from week 1 to 13) implying that while there were significant differences between the treatments and the zones in the start of the survey, the significant differences diminished as the survey continued. To determine if different growth stages of N. tenuis counts (adults and nymphs) would provide insights into such trends, N. tenuis adults and nymphs were analyzed separately. While significant differences (p = <0.001) between the hemp rope and banker plants treatment as well as between the zones could be observed, significant difference in N. tenuis adults were observed until week 7 of the survey while nymphs a week later ( Figure A1 and Table A1). This observation is of no surprise as the nymph emergence and or occurrences are expected to follow that of the adults. As expected, N. tenuis individuals clustered mainly in zone 1 in the first half of the study in the banker plant treatment as zone 1 was adjacent to the banker plants and progressively moved onto other zones as the study continued. On the other hand, N. tenuis were expected to scatter among the tomato plants in all zones in the hemp rope treatment. Presumably, the presence of the factitious diet during the release of N. tenuis assisted with their spread.

Discussion
Nesidocoris tenuis has been a success story in managing key pests of tomato crops in the Mediterranean basin as well as in Japan [6,19]. Releasing and maintaining the N. tenuis on tomato plants at a rate of 0.5 to 1 insect per plant over the whole cultivation period is usually recommended to minimize N. tenuis-mediated damages on tomato [8,9]. To minimize proliferation beyond the recommended rate, supplementing N. tenuis with banker plants such as sesame or verbena has been recommended and been practiced in the western part of Japan (9). The whole idea behind supplementing N. tenuis with banker plants is so that N. tenuis can proliferate on banker plants to the maximum number of individuals in a given time and can move to the tomato plants to prey on B. tabaci, T. absoluta and T. palmi when these prey are present. While relocation of N. tenuis from the banker plants to tomato plants is always guaranteed, distribution among all the tomato plants under the field conditions may be delayed and, thus, delay control of a given pest, as shown in our data (Figures 1 and 4). A system that promotes natural enemy proliferation is therefore needed to enhance propagation, relocate, and spread among the cultivated crops.
Factitious prey is preferentially used in supplementing and augmenting biological control agents. Factitious prey can be used as a live specimen, frozen, irradiated, or lyophilized insects, mites, and crustaceans. Artemia cysts are live aquatic crustaceans having the potential to promote natural enemy growth and development [7,[10][11][12][13][14][15][16][17]. Sugar has also been documented to nutritionally supplement N. tenuis [20]. Combining Artemia cysts with sugar, Owasi et al. [7] showed in vitro that life-history traits of N. tenuis could be improved and recommended for future studies. Building on the research by Owasi et al. [7], we showed under greenhouse conditions that N. tenuis proliferation was accelerated when Artemia cysts were supplemented with high fructose corn syrup and honey as sugar source. The increased proliferation thus encouraged the spread of N. tenuis among the tomato plants. As a result of the increase in proliferation and spread among tomato plants, the number of B. tabaci eggs and nymphs was suppressed as compared to the control treatment. With all variables constant, the use of Artemia cysts enhanced with high fructose corn syrup and honey as factitious dietary to augment N. tenuis seem promising for B. tabaci control in a protected system, as such in tomato production.
While successful N. tenuis proliferation, spread, and control of B. tabaci eggs and nymphs could be observed in the hemp rope supplemented greenhouse, an accelerated increase in N. tenuis individuals was also observed. According to Calvo et al. [8] and the N. tenuis "how to use" manual by NARO [9], 0.5 to 1 individual per two tomato plants is recommended. However, more than 0.5 to 1 N. tenuis insect per tomato plant (maximum of nine individuals per plant at the highest peak) was observed in our study. While it can be argued that the increased number of N. tenuis can be advantageous in managing pests, scarcity in prey numbers as a result of over-predation can result in plant-feeding. There are three models proposed by Gillespie and McGregor [21] to determine the feeding behavior of omnivorous Heteropterans; (i) the amount of plant-feeding decreases with increased prey feeding, (ii) the amount of plant-feeding increases with increased prey feeding, and (iii) the amount of plant-feeding is independent of the amount of prey feeding. Calvo et al. [8] observed that N. tenuis, a classic example of zoophytophagous insect that plant-feeds when animal feeds are scarce seemed to follow the first model. They further concluded that damage to tomato plants decreased in response to the greater availability of B. tabaci. Arno et al. [22] also made similar observations under laboratory conditions. However, in our study, necrotic rings were observed on the younger shoots of tomato plants (data not shown) even prey was still present. Artemia cysts are promising, in that they accelerate N. tenuis proliferation; however, a higher number of N. tenuis individuals on tomato plants may not only prey on the pests such as whiteflies but also supplementing themselves with nutrients through feeding on tomato plants, yielding plant damage.
Artemia cysts enhanced with high fructose corn syrup and honey are a cheap and cost-effective source of factitious supplementary dietary. Our study, under greenhouse conditions, further supports the in vitro studies by Owasi et al. [7] establishing Artemia cysts mixed with high fructose corn syrup and honey are equally efficient as banker plants in augmenting N. tenuis growth and development while improving their spread among the tomato plants. While an increase in N. tenuis numbers on tomato plants can improve pest control, an increase beyond recommended rates (0.5-1 individual per plant) can negatively impact the crop (tomato) growth and productivity. Further field studies are needed to optimize the use of Artemia cysts enhanced with high fructose corn syrup and honey in augmenting N. tenuis under both open and protected field settings, at the same time, management of N. tenuis using selective pesticides to minimize plant-feeding. These parameters, therefore, are required in the optimization and development of the Integrated Pest Management (IPM) system in tomato against whiteflies. Furthermore, this study was performed in two consecutive years in a 10 m by 5 m greenhouses. Multiple consecutive annual studies and in large scale systems is needed to categorically substantiate these results.

Conclusions
In the past decade, studies have been performed on the suitability of Artemia cysts as factitious supplementary dietary for natural enemies and/or predators under in vitro or semi-field conditions [7,10,11,[13][14][15][16][17][23][24][25][26]. We showed under greenhouse conditions the Artemia cysts enhanced with high fructose corn syrup and honey could augment and improve the growth and development of N. tenuis. The improved growth and development can hypothetically enhance quicker settlement of the natural enemies on the crops thus, unsettling and decelerating the establishment of the intended pest before it can cause any further economic damages. The results in this study establish foundational data in building towards an IPM system integrating N. tenuis, Artemia cysts enhanced with isomerized sugar (high fructose corn syrup) and honey, and selective pesticides to control greenhouse pests.     *: Insecticides were mixed in a 15 L sprayer at the rates specified above.