Introduction

The zoophytophagous predator Nabis stenoferus Hsiao (Hemiptera: Nabidae) is a natural enemy of prey such as aphids, moths, and spider mites in Korea, Japan, and China1,2,3. Nabis stenoferus can also be a pest in some crop plants, such as chrysanthemum [Chrysanthemum spp. (Asterales: Asteraceae)] due to its zoophytophagous habit1. Generally, this predator lives in grasslands around agricultural fields, and it has been regarded principally as a biodiversity component of agroecosystems rather than a pest or biocontrol agent2,3,4. However, N. stenoferus has the potential to be a new biological control agent for use in augmentation or conservation programs because of its generalist feeding habit and adaptation to temperate regions3,4. Generalist predators can be especially useful as biological control agents in crops where several pest species occur at the same time5. Also, a generalist may be better able to survive under low pest densities than a specialist as alternative food sources are available to them5,6,7.

Efficient mass-rearing of natural enemies is essential if agents are to be used in augmentative biocontrol8. For efficient mass-rearing, it is crucial to find suitable environmental conditions such as temperature and relative humidity9,10,11, however, adequate diet is also a critical factor12,13. The food item used for insect mass-rearing should be inexpensive, easily obtained, and suitable for both the species’ development and oviposition14.

Also, even though a particular food may not be part of the insect’s natural diet, such foods may be used for rearing insects if they provide sufficient nutrients14. Eggs of stored product moths such as Ephestia kuehniella Zeller (Lepidoptera: Pyralidae) are frequently used for mass-rearing generalist predators15,16. This moth is a pest of stored grain and is not found in Korea, but its frozen eggs have been imported into Korea for the mass-rearing of insect predators17,18. However, it can be argued that foods that are part of a species’ diet under natural conditions might be more effective than foods never yet encountered by a generalist predator19. Myzus persicae (Sulzer) (Hemiptera: Aphididae) is a common agricultural pest in crops in the families Asteraceae, Brassicaceae, Cucurbitaceae, and Solanaceae20,21,22,23. This aphid is found in open fields in agricultural areas in Korea, Japan, and China, and its host plants and habitats overlap with those of N. stenoferus1,3,20. Also, this aphid has sometimes been used as a supplementary food for the rearing of generalist predators24.

Life table analysis allows the population parameters of insects under specific conditions to be calculated. Among life table parameters, the intrinsic rate of increase is a comprehensive and intuitive parameter describing the population potential of an insect reared under specific conditions that can be calculated from data on developmental time, fecundity, longevity, sex ratio, and survivorship25,26. Population parameters derived from life table analysis can be statistically compared using bootstrap or jackknife methods25,27. Thus, life table analysis is a very effective tool for comparing the fitness of insects under different conditions. However, life table analysis of N. stenoferus has rarely been done.

Our goal in this study was to find a suitable diet for mass-rearing N. stenoferus and to better understand its biology. We compared its life history characteristics using life table analysis for groups reared using three diets: (1) aphids only (M. persicae), (2) moth eggs only (E. kuehniella), and (3) a mixed diet of aphids and moth eggs.

Results

Immature stages

There were no significant differences among the three treatments (aphids only, eggs only, or both) in the development periods of the egg to 3rd nymph stages (Table 1) (egg, χ2 = 0.13, df = 2, P = 0.935; 1st nymph, χ2 = 3.70, df = 2, P = 0.157; 2nd nymph, χ2 = 5.51, df = 2, P = 0.064; 3rd nymph, χ2 = 4.53, df = 2, P = 0.104). However, the developmental periods of the 4th and 5th nymphal instars, as well as that of the total immature period for groups fed a mixed diet were significantly shorter than in either the aphid-only or moth egg-only diets (Table 1) (4th nymph, χ2 = 17.78, df = 2, P < 0.001; 5th nymph, χ2 = 27.45, df = 2, P < 0.001; total immature, χ2 = 31.50, df = 2, P < 0.001).

Table 1 Developmental periods (days, mean ± SE) of immature stages of Nabis stenoferus fed three different diets. *Means followed by the same letter within a column are not significantly different at α = 0.05. DSCF multiple comparisons as a post hoc test for the Kruskal–Wallis test.
Full size table

Adult females

There were significant differences among the diet treatments in the life history characteristics of adult females of N. stenoferus (Table 2). The preoviposition period was significantly shorter in the group fed a mixed diet than in aphid-only or moth egg-only group (F = 16.11, df = 2, 17; P < 0.001). The oviposition periods varied from 5.7 to 41.6 days among the treatments, but without statistical significance (χ2 = 5.85, df = 2, P = 0.054). The postoviposition period in the group fed the aphid-only diet was significantly longer than in moth egg-only group (1.4 days) (χ2 = 7.50, df = 2, P = 0.024). There was no significant difference in female adult longevity among the diets tested (F = 1.76, df = 2, 25; P = 0.193). The total fecundity per female of bugs fed the mixed diet (aphids + moth eggs) was significantly higher than in groups fed the other two diets (F = 14.75, df = 2, 25; P < 0.001).

Table 2 Oviposition period, longevity, and fecundity (mean ± SE) of Nabis stenoferus reared from eggs on three different diets. *Means followed by the same letter within a column are not significantly different at α = 0.05. Tukey’s studentized range test as a post hoc test for analysis of variance or DSCF multiple comparisons as a post hoc test for the Kruskal–Wallis test.
Full size table

Life table parameters by treatment

The age-specific survival rate of N. stenoferus for nymphs fed the aphid-only diet decreased more rapidly than did rates of groups fed the other diet (Fig. 1). Fecundity was also much lower in the group fed the aphid-only diet (Fig. 1). Different diet treatments significantly affected the population parameters of N. stenoferus (Table 3). The intrinsic rate of increase and finite rate of increase were highest in insects fed the mixed diet, followed by the moth egg-only group and the aphid-only group. The net reproductive rate also was highest in insects fed the mixed diet. Furthermore, the net reproductive rate in the group fed the aphid-only diet was relatively lower compared to insects on the mixed diet, but was not significantly different compared to moth eggs only diet. There was no significance among treatments in the mean generation time.

Figure 1
figure 1

Age-specific survival rate (lx) and fecundity (mx) curves of Nabis stenoferus fed three different diets.

Full size image
Table 3 Population parameters (estimate ± SE) of Nabis stenoferus fed three different diets. *Means followed by the same letter within a column are not significantly different at α = 0.05, Tukey’s studentized range test after jackknife estimates.
Full size table

Discussion

In this study, we found significant effects of different food sources on the life history characteristics of N. stenoferus. Moth eggs alone did provide sufficient nutrients for both insect development and adult oviposition by N. stenoferus, but use of the aphid-only diet yielded sterile, or nearly sterile, adults. The mean fecundity in insects reared on the mixed diet was 87 times higher than bugs reared on the aphid-only diet. However, the fecundity of bugs reared on the moth egg diet was not significantly different from the group reared on aphid-only diet probably due to the very low number of ovipositing females obtained. Furthermore, the aphid-only diet lowered adult fitness of N. stenoferus by prolonging both the pre- and post-oviposition periods. Aphids are known to be low-quality foods for generalist predators, likely due to the presence of toxins or feeding deterrents24. However, in our study, there were no negative effects of an aphid-only diet on the development of the immature stage of N. stenoferus, and, consequently, we surmise that low adult fertility might be due to a nutritional deficiency in the aphid-only diet. Both protein and lipid diet components are known to significantly affect insect fecundity28,29,30. Proteins and lipids comprise 51.0 and 33.6% of the dry mass of Ephestia eggs31. In contrast, aphid-only diets have a lower lipid content than Ephestia eggs28,29,30. Different dietary needs for development in the immature stage versus reproduction in the adult stage are well known32,33. Different food exploitation patterns between immature and adult stages have frequently been reported in predatory mites. For example, immature stages of Phytoseiulus persimilis Athias‐Henriot (Acari: Phytoseiidae) can consume thrips and complete their stage’s development despite a lower survival rate of this food34. However, the adult P. persimilis was known to rarely consume thrips as food34. Thus, a M. persicae-only diet might be suitable for the development of N. stenoferus nymphs but not suitable for adult maturation is consistent with other systems.

However, there was a synergistic effect of aphid and moth egg diets for the life history characteristics of N. stenoferus in both the immature and adult stages. Numerous studies have reported the advantages of a mixed diet on insect fitness35,36,37. However, the noteworthy finding in our study is that the diet on which N. stenoferus completed its nymphal development (but yielded sterile adults) had a synergistic effect when mixed with other diet. Toft et al.30, in which Ephestia eggs and/or Rhopalosiphum padi (L.) aphids were used as food for Orius majusculus Reuter (Hemiptera: Anthocoridae), found lower fecundity in the aphid diet than in Ephestia egg diet. However, unlike our study, there was no synergistic effect on O. majusculus’ fecundity from a mixed diet. In contrast, we found that the fecundity of N. stenoferus fed a mixed diet of aphids and moth eggs more than doubled compared to the moth eggs-only diet. The developmental period of the immature stage and the preoviposition period of N. stenoferus were also significantly shorter for bugs reared on the mixed diet. Furthermore, the mixed diet resulted in insects with a higher intrinsic rate of increase than did the other two treatments. This is the first finding of such contrasting contributions of diets on the development and reproduction of Hemiptera as far as we know.

Zoophytophagous hemipteran predators facultatively consume plant sap to obtain nutrients and water, but they can complete their life cycles only by feeding on animal prey such as insects or mites, as shown for N. stenoferus in this study38,39,40,41,42. However, some zoophytophagous predators can complete their life cycles by feeding only on plants without prey. For example, Nesidiocoris tenuis (Reuter) (Hemiptera: Miridae) can successfully increase its population by feeding only on sesame plants [Sesamum indicum L. (Lamiales: Pedaliaceae)]43, although it was not able to do so when feeding only on tomato plants [Solanum lycopersicum L. (Solanales: Solanaceae)]44. The Ephestia egg diet did not show synergistic effects on the fitness of N. tenuis when paired with a tomato plant39. However, Ephestia egg diet did have synergistic effects on development periods and fecundity of N. tenuis when paired with a sesame plant43, showing that plant-derived components from a suitable host plant could positively affect the fitness of zoophytophagous predators45. In our study, we supplied Chinese cabbage [Brassica rapa L. subsp. pekinensis (Brassicales: Brassicaceae)] leaves to N. stenoferus as its water source and oviposition substrate. Chinese cabbage might not be able to increase the population size of N. stenoferus without prey because the aphid-only diet in our study did not show good adult fitness. However, the nutrients derived from Chinese cabbage might be converted and enriched in the aphids, and these nutrients in a mixed diet might enhance the fitness of N. stenoferus. Thus, in our study, M. persicae in a mixed diet with eggs of E. kuehniella might have played a similar role for N. stenoferus as a beneficial plant did for N. tenuis in the previous example.

Even though M. persicae alone is not a complete food source for N. stenoferus, it might be an error to regard this predator as an ineffective biological control agent against M. persicae or other aphid species. A diet that improves a predator's fitness does not necessarily engender a higher preference for that diet compared to others, even ones with intrinsically poorer nutrient profiles30. In Toft et al.30, O. majusculus, when reared on Ephestia eggs only or on a mixed diet of eggs and aphids, still preferred the poorer quality aphid-only diet. However, this predator, when reared on an aphid-only diet showed no preference between Ephestia eggs and aphids. Even though a diet might not be able to provide all the nutrients a predator might need, if the diet has the essential nutrients such as vitamins and amino acids, the predator may still consume the diet to prevent nutrient deficiency46. Unlike laboratory conditions, where the type of diet is artificially restricted, under natural conditions predators have opportunities to exploit mixed host resources efficiently for the best fitness gain46. The natural prey of N. stenoferus are known to be moth eggs or larvae, spider mites, and aphids3; and N. stenoferus should maximize its fitness by exploiting mixed diets in nature. However, in agricultural areas, especially in greenhouses, a predator’s diet choices may be limited by the simplification of the agroecosystem47,48. In predator-based augmentative biological control programs against aphid species in greenhouses, supplemental provision of Ephestia eggs might improve control49. Also, it may be possible to plant chrysanthemums in greenhouses as banker plants for N. stenoferus because this predator can complete its life cycle on chrysanthemums alone1. However, in open grassy fields, where N. stenoferus is an indigenous predator2,3, the species would be appropriate for conservation biological control50,51. Building up refuges with chrysanthemum plants can be suitable to enhance this predator than probably food spraying of Ephestia eggs, which can be used by other antagonists such as ants protecting aphids52,53.

In conclusion, the overall fitness of N. stenoferus was higher on a diet of E. kuehniella eggs than a diet of M. persicae, and a mixed diet of M. persicae and E. kuehniella eggs had a synergistic effect on the fitness of N. stenoferus in both the immature and adult stages. Therefore, we propose that M. persicae can be a supplementary food source for the mass-rearing of N. stenoferus, with the eggs of E. kuehniella the primary food source. Further studies might be needed on the nutritional composition of both E. kuehniella eggs and M. persicae to identify essential nutrients that might be responsible for the better fitness of N. stenoferus. To establish an efficient mass-rearing system for N. stenoferus, studies on the effects of environmental conditions such as temperature and humidity on N. stenoferus would also be needed. Moreover, as the use of Ephestia eggs for mass breeding of N. stenoferus may become expensive in certain regions or under specific conditions, additional research might be necessary to investigate the feasibility of utilizing relatively inexpensive diets such as brine shrimp eggs54.

Methods

Food sources

The eggs of E. kuehniella were used to feed our laboratory colony of N. stenoferus. Combinations of both E. kuehniella eggs and M. persicae (nymphs and adults mixed) were tested for the diet suitability. Both eggs and aphids were purchased from the Osang Kinsect, Namyangju, Korea. The eggs of E. kuehniella were frozen in the bottles received and taken out and used whenever necessary. Seedlings (about 10 cm height) of Chinese cabbage (B. rapa subsp. pekinensis) were purchased from the market in Andong, Korea, and planted in each pot (10 cm × 9.7 cm; diameter × height) and kept at 27 ℃, 60–80% RH, and a 16:8 (L:D) h photoperiod in an incubator (DS-50CPL, Dasol Scientific Co., ltd., Suwon, Korea) until plants grow to 20 cm. Myzus persicae stages were reared in an acrylic cage (30 × 30 × 30 cm) on three pots of Chinese cabbage at 27 ℃, 60–80% RH, and a 16:8 (L:D) h photoperiod in an incubator. Chinese cabbage pots were replaced at two to three week intervals. All methods were carried out in accordance with the relevant guidelines and regulations of Republic of Korea.

Laboratory rearing of Nabis stenoferus

Nabis stenoferus was obtained from the Gyeonggi-do Agricultural Research and Extension Service in Hwaseong, Korea, and bugs were reared individually from egg to adult in Petri dishes (35 mm × 10 mm; diameter × height; SPL Life Science, Pocheon, Korea) at 27 ℃, 60–80% RH, and a 16:8 (L:D) h photoperiod in an incubator. Eggs of E. kuehniella, attached on parchment paper (1 × 1 cm), were provided as food for N. stenoferus rearing, and the egg papers were replaced daily. A piece of water-saturated cotton (0.8 × 0.8 × 0.8 cm) was placed in the rearing Petri dishes to supply water and as an oviposition substrate. When N. stenoferus individuals became adults, a pair of predators were kept in the Petri dish for mating.

Life table experiments

The experiment on rearing diets was conducted at 27.9 ± 0.76 ℃, 50.1 ± 6.4% RH, and a 16:8 (L:D) h photoperiod in an incubator. Twenty individuals (as newly laid eggs) of N. stenoferus were used for each food treatment. To obtain newly laid N. stenoferus eggs, ten adult females were randomly collected from the rearing colony for each treatment. Each female was allowed to lay eggs on water-saturated cotton in a Petri dish (35 mm × 10 mm; diameter × height) for 24 h. After 24 h, each female’s cotton pieces were collected into a larger Petri dish (90 mm × 15 mm; diameter × height; SPL Life Science) and held for egg hatch. The eggs developmental period was recorded.

The newly emerged nymphs were randomly selected and placed individually into experimental Petri dishes (50 mm × 15 mm; diameter × height; SPL Life Science) containing water-saturated cotton and a Chinese cabbage leaf disc (50 mm diameter). Diets were provided in Petri dishes as three treatments: (1) aphids only, (2) moth eggs only, and (3) both aphids and moth eggs. Aphid numbers in each Petri dish with aphids were maintained at least 40 individuals daily. The daily supply of aphids might be sufficient for Nabis sp. to meet its feeding needs (about 14.4 aphids were consumed per day in a previous study55) with no observable food shortage. For treatments with moth eggs, one parchment paper (1 × 1 cm2) with several hundred eggs was present in each Petri dish and was replaced daily.

When nymphs molted to adults, individual females and males were paired and held in new experimental Petri dishes. When female could not be immediately paired due to discrepancy in the number of molted males, we used male adults collected from the rearing colony. In the adult stage, only the life history characteristics of females were assessed, according to Maia et al.25,26. Developmental times of each life stage, female adult longevity, and daily fecundity were observed daily until all females had died.

Data analysis

The data from any individuals lost during the experiment were discarded before analysis. Females that did not lay eggs were excluded from the calculation of mean oviposition periods. The developmental periods for nymphs, as well as the oviposition and postoviposition periods were compared among treatments using the Kruskal–Wallis test in PROC NPAR1WAY in SAS56 because of the non-normal distribution of the data. The preoviposition periods, as well as female adult longevity and fecundity were compared among treatments by analysis of variance using PROC GLM in SAS56.

Life table analysis and jackknife estimation of population parameters were carried out using the R program57 by referring to Maia et al.25,26. Age-specific survival rates (\({l}_{x}\)) and fecundity (\({m}_{x})\) for each treatment were calculated using the following equations:

$${l}_{x}=SURV\times \frac{{NSF}_{x}}{NF}$$
$${m}_{x}={NEGG}_{x}\times SR$$

where \(SURV\) is the survival rate from egg to adult, \({NSF}_{x}\) is the number of surviving females at age \(x\), and \(NF\) is the initial number of females. \({NEGG}_{x}\) is the mean number of eggs laid at age \(x\), and \(SR\) is the sex ratio of each treatment group. The population parameters were calculated by the following equations25.

The net reproductive rate (\({R}_{0}\))

$${R}_{0}=\sum_{x=0}^{\infty }{l}_{x}{m}_{x}$$

The mean generation time (\(T\))

$$T=\frac{\sum_{x=0}^{\infty }x{l}_{x}{m}_{x}}{\sum_{x=0}^{\infty }{l}_{x}{m}_{x}}$$

The intrinsic rate of increase (\({r}_{m}\))

$$\sum_{x=0}^{\infty }{e}^{-{r}_{m}x}{l}_{x}{m}_{x}=1$$

The finite rate of increase (\(\lambda\))

$$\lambda ={e}^{{r}_{m}}$$

The population parameters such as net reproductive rate (\({R}_{0}\)), mean generation time (\(T\)), intrinsic rate of increase (\({r}_{m}\)), and finite rate of increase (\(\lambda\)) were compared by Tukey's studentized range test after jackknife estimation25.