Abstract

Laboratory experiments were conducted to examine selection by Cotesia plutellae between host larvae feeding on two plant species: Chinese cabbage, Brassica campestris ssp. pekinensis and common cabbage, Brassica oleracea var. capitata. When C. plutellae wasps were provided with equal numbers of Plutella xylostella larvae on both species of plants in one arena, the parasitoid parasitised 4-15 fold more host larvae on the Chinese cabbage than on the common cabbage and this preference did not change with host density. However, an experience of oviposition or searching on a leaf of the less-preferred plant, the common cabbage, significantly increased the preference for parasitising host larvae on this plant. Plant volatiles from Chinese cabbage were more attractive to C. plutellae adult females than those emanating from common cabbage. Feeding by P. xylostella larvae increased the attraction of both plant species to C. plutellae, but the infestation and the presence of P. xylostella larvae on the plants did not affect their relative levels of attraction to the parasitoid. In parallel to the increased parasitism on common cabbage following experience, an oviposition in host larvae on this less-preferred plant significantly increased the response to volatiles emanating from the plant. These results indicate that host plants may strongly influence the foraging behaviour of C. plutellae, but their differential attractions to the parasitoid may be significantly offset by the learning behaviour of the insect.

Keywords

Plutella xylostella, host plants, host foraging, parasitoid learning

Introduction

Plants may mediate many of the interactions between herbivores and their insect parasitoids and thereby increase or decrease the effectiveness of natural enemies (Cortesero et al. 2000). Understanding these multitrophic effects may help explore the potential for manipulating crop-pest-parasitoid interactions for improved pest management (Bottrell et al. 1998, Verkerk et al. 1998).

Insect parasitoids are important biological control agents of the diamondback moth (DBM), Plutella xylostella (L.) (Lepidoptera: Plutellidae), yet their performance has not been examined in a multitrophic context until very recently (Talekar & Shelton 1993, Verkerk & Wright 1996). Field observations suggest that parasitoids may show different rates of parasitism of DBM on different crops (Verkerk & Wright 1997, Liu et al. 2000). Bogahawatte and van Emden (1996) showed that the endo-larval parasitoid, Cotesia plutellae Kurdjumov (Hymenoptera: Braconidae) preferred the odour of the host plant on which it had developed. Potting et al. (1999) showed that C. plutellae used plant volatiles in its in-flight searching behaviour, which was modified by its experience with plant material. Shiojiri et al. (2000) demonstrated that C. plutellae, being a host-specific parasitoid, showed very specific responses to volatile stimuli and exhibited differential responses between host- and nonhost-plant complexes.

Here, we investigate the relative foraging success for DBM larvae by C. plutellae on two species of host plants. We report the response of the parasitoid to volatiles emitted by the two host plants, and the role of experience in determining parasitism success on the two plants and the response to volatiles. Our objectives were to determine the effects of host plants and in particular their volatiles on the foraging behaviour and success of this parasitoid, and to determine whether these effects are modified by the parasitoid’s experience.

Materials and methods

Insects and plants

Three species of plants were used in this study: (1) Chinese cabbage, Brassica campestris L. ssp. pekinensis, cv. Zaoshou No. 5; (2) common cabbage, Brassica oleracea L. var. capitata, cv. Jingfeng No. 1; and (3) radish, Raphanus sativus L. cv. Yidianhong. They were grown from seed in potting mix in small pots (11 cm diameter) in screen houses to the 6-7 leaf growth stage when used.

Three cultures of DBM were started from a sample collected from a cabbage field in a suburb of Hangzhou, China and maintained on the three plant species using the procedures as described in Wang et al. (1999).

A culture of C. plutellae was started from a sample collected from a radish field in a suburb of Hangzhou. The culture was maintained by exposing approximately 500 II and III instar DBM larvae on radish plants to 10 mated female parasitoids (3-4 days after emergence) in a ventilated cage (55 × 55 × 55 cm, with a glass top, screen-covered opening on three sides and a clear plastic front door) for two days. The exposed larvae were reared until parasitoid cocoons developed. The cocoons were transferred to ventilated plastic containers for adult emergence and mating. The adults were fed with 10% honey solution upon emergence. To obtain parasitoids for the experiments, II or III instar DBM larvae were exposed individually to single mated female parasitoids and each larva was observed to receive oviposition only once to avoid superparasitism. The exposed DBM larvae were then reared until parasitoid cocoons developed. The cocoons were collected and placed in clean containers (i.e., without any plant or host material) for adult emergence and mating (provided with honey solution). Mated female wasps were collected 3-4 days after emergence for use in the observations. All insect cultures were maintained in constant temperature rooms at 25 ± 1°C, 14L:10D and 60-80% RH.

Parasitism of host larvae on two plant species

The day before the parasitism was measured, II and early III instar DBM larvae from the culture maintained on Chinese cabbage were collected and placed on the test Chinese cabbage plants, and DBM larvae from the culture maintained on common cabbage were collected and placed on the test common cabbage plants. Two Chinese cabbage and two common cabbage plants each bearing the same number of DBM larvae were placed in alternate position at the four corners of a cage (size and structure as described above), the top of which was covered with tracing paper to defuse the light from above. Wooden boards were placed near the four lateral sides of each cage to avoid interference from lateral light. The cages were placed in a constant temperature room at 25 ±1°C with minimum or no airflow. Two naïve parasitoid females (with no experience of oviposition or searching on a plant) were introduced into each cage for oviposition for 6 h and then discarded. The exposed DBM larvae on each of the test plants were collected and dissected to determine parasitism. The trial was conducted with four levels of host density (5, 10, 20 or 30 larvae per plant) with 10 replicates for each level of density.

In a separate trial, one Chinese cabbage and one common cabbage plant each bearing the same number of DBM larvae were placed 25 cm apart near the back of a cage. One naïve female parasitoid was released from a point near the front door of the cage (with equal distance to the two plants) and its behaviour was observed continuously for 30 minutes to record time spent on each of the two plants (with a stop watch) and the number of ovipositions in larvae on each of the two plants. The trial was conducted with four levels of host density (5, 10, 20 or 30 larvae per plant) with 30 females observed for each host density. The positions of the two plants in a cage were alternated between observations.

Effect of adult experience on host foraging

The experimental set-up was similar to that used for parasitism of host larvae in a cage, except that the female parasitoids were given various experiences immediately prior to their introduction into the cage. Six pre-treatments of parasitoid females were conducted (see Table 3): the females were provided with a DBM-damaged leaf of either the Chinese cabbage (1) or common cabbage (2) to search for 10 minutes; or each female was allowed to oviposit once in a DBM larva, which had been reared on either the Chinese cabbage (3) or common cabbage (4), in a test tube without plant material; or each female was allowed to search for 10 minutes and oviposit once in a DBM larva on a host-damaged leaf of either the Chinese cabbage (5) or common cabbage (6). Two Chinese cabbage and two common cabbage plants each bearing 20 DBM larvae were placed in alternate position at the four corners of a cage. Two pre-treated females were introduced into each cage for 6 h and 10 replicates were carried out for each pre-treatment.

Response to volatiles of different plants

The response of parasitoid females to volatile chemicals emitted by different odour sources was investigated in a Y-tube olfactometer. The olfactometer consisted of a Y-shaped glass tube 2.8 cm in internal diameter. The stem and the two arms (at a 75° angle) of the Y-tube were 16 and 28 cm in length, respectively. Each arm was connected via a Teflon hose to an odour source chamber consisting of a glass box measuring 25 × 25 × 35 cm, large enough to hold a whole test plant plunged in a small water bottle. Air was drawn by the negative pressure of an electric pump, filtered through an activated-charcoal filter and humidified by bubbling through distilled water before being pulled into the odour source chamber at approximately 200 ml/min. The observations were made at 25±1°C in a box with defused light above the Y-tube.

Female parasitoids were released individually into the base of the stem of the Y-tube, and each of them was given 5 minutes to move upwind towards the ends of the arms of the tube. When a female penetrated more than 10 cm into one of the two arms and remained there for more than 30 seconds, it was recorded as a choice for that arm. The connections of the odour sources to the olfactometer arms were exchanged after testing five parasitoids to remove any asymmetrical bias in the set-up. The olfactometer tube was washed with alcohol after testing 10 females.

For each of the two species of plants, four types of odour sources were tested: (1) intact plants; (2) mechanically damaged plants: the plants were each punched with 20 5-mm holes in the leaves 24 h before the test and were punched again in the wounds 10 minute prior to the test; (3) infested plants: each plant was infested with 20 II instar DBM larvae 24 h prior to the test and the larvae were left on the plant during the test and (4) previously infested plants: the plants were infested as in (3) but the larvae were removed 10-20 minutes prior to the test.

Four pairs of odour sources were compared between the two plant species and one pair was compared within each of the two plant species (Figure 1, Results section). For each pair of odour sources, about 40 females were individually tested. A goodness-of-fit G-test, with the application of Williams’s correction, was applied to analyse the numbers of females that made a choice in each pair of odour sources with the null hypothesis of no preference (Sokal & Rohlf 1995).

Effect of adult experience on response to plant volatiles

The test materials and methods were the same as above, except that, 5-10 minutes prior to introduction into the Y-tube, each test female was allowed to search and oviposit once in a III instar DBM larva feeding on a host-damaged leaf of either Chinese cabbage or common cabbage.

Results

Parasitism of host larvae on two plants species

When provided with equal numbers of DBM larvae feeding on the two plant species, the female parasitoids parasitised 5-16 times more larvae on the Chinese cabbage than on the common cabbage (Table 1). The relative proportion of DBM larvae parasitised on Chinese cabbage increased with host density in the range of 5-20 larvae per plant and then declined as host density further increased to 30 per plant (Table 1). The results of direct observations showed that female parasitoids spent a much longer time and had many more ovipositions in DBM larvae on Chinese cabbage than on common cabbage (Table 2). Note that the number of ovipositions does not equal the number of DBM larvae parasitised, because superparasitism was common under these conditions especially at higher host density.

Table 1. Number of Plutella xylostella larvae parasitised by Cotesia plutellae on plants of two species with the same host density in choice tests for 6 h at 25°C

Host density	No. of	Mean ± SE number of larvae parasitised per plant
(Larvae/plant)	replicates	B. campestris (a)		B. oleracea (b)		a/b
5	10	2.0 ± 0.2		0.3 ± 0.2		6.7
10	10	6.8 ± 0.4		0.7 ± 0.3		9.7
20	10	10.9 ± 0.9		0.7 ± 0.3		15.6
30	10	11.9 ± 1.2		2.5 ± 0.5		4.8

Note: At each of the four levels of host density, the two mean numbers of DBM larvae parasitised on the two species of plants differ significantly (P<0.01 in all cases, Student-t test).

Table 2. Time spent and number of ovipositions in host larvae by Cotesia plutellae on two plant species with the same host density in choice tests for 30 minutes at 25°C

Host density	n	Mean ± SD duration (s) spent on			Mean ± SE number of ovipositions on
(larvae/plant)		B. campestris	B. oleracea		B. campestris	B. oleracea
5	30	216 ± 27	9 ± 3		0.97 ±0.14	0.23 ± 0.08
10	30	564 ± 44	24 ± 4		4.10 ± 0.35	0.57 ± 0.12
20	30	842 ± 16	29 ± 4		8.90 ± 0.13	0.77 ± 0.16
30	30	1073 ± 18	41 ± 5		12.20 ± 0.39	1.40 ± 0.20

Note: At each level of host density, both the mean durations between the two species of plants and the mean numbers of ovipositions between the two plants differ significantly (p<0.01 in all cases, Student-t test).

Effect of adult experience on host foraging

Compared with the much higher number of DBM larvae parasitised on Chinese cabbage than on common cabbage by naïve females (Table 1), females that had searched on a leaf of common cabbage parasitised similar numbers of larvae on the two plants. Females that had searched and oviposited in a larva feeding on a leaf of common cabbage, parasitised twice as many larvae on common cabbage than that on Chinese cabbage (Table 3). Similarly, females that had experience of searching on a leaf of Chinese cabbage or search coupled with an oviposition in a larva feeding on the leaf increased their preference for hosts on this plant further, resulting in parasitism on Chinese cabbage only and no parasitism on common cabbage. In contrast, females that had oviposited in a DBM larva without access to the plant, regardless of the plant species from which the larva had been reared, did not change their preference for host larvae between the two plants (Table 3).

Table 3. Number of Plutella xylostella larvae parasitised by Cotesia plutellae on plants of two species in choice tests for 6 h at 25°C when the parasitoid female adults had different prior search experiences

Prior search experience	Mean ± SE number of larvae parasitised/planta
	B. campestris (a)	B. oleracea (b)	a/b
(1) Search on a B. campestris leaf for 10 min	15.2 ± 0.3	0	-
(2) Search on a B. oleracea leaf for 10 min	11.5 ± 0.5a	10.2 ± 0.8a	1.1
(3) Oviposition in a larva previously reared from B. campestris in a tube without plant material	11.9 ± 0.7a	1.1 ± 0.3b	10.8
(4) Oviposition in a larva previously reared from B. oleracea in a tube without plant material	11.7 ± 0.7a	0.9 ± 0.3b	13.0
(5) Oviposition in a larva feeding on a B. campestris leaf	15.9 ± 0.5	0	-
(6) Oviposition in a larva feeding on a B. oleracea leaf	4.6 ± 0.8b	11.5 ± 0.7a	0.4

^a Ten replicates in each treatment, means in the same row followed by the same letters do not differ significantly (P>0.05, Student-t test).

Response to volatiles of different plants

When female parasitoids were offered a choice between airflows carrying volatiles from the two plant species, about twice as many of them moved towards the Chinese cabbage as towards the common cabbage, although the preference in the test with intact plants did not reach statistical significance due to the low number of females that responded (Figure 1). Within each of the two plant species, the females showed high preference for infested plants compared with intact ones (Figure 1).

Figure 1. Numbers of Cotesia plutellae female adults showing a response to plant volatiles of Brassica campestris (Chi-cabbage) and B. oleracea (cabbage). Asterisks indicate statistically significant preferences in a choice test (*: P<0.05, **: P<0.01, ***: P<0.001). NS: not significant. n: number of parasitoid wasps tested. The number of wasps that did not choose either of the odour sources is listed under “No choice”. M-damaged: mechanically damaged. P-infested: previously infested.

Effect of adult experience on response to plant volatiles

Compared with naïve female parasitoids, those with a prior experience of ovipositing in a DBM larva feeding on a leaf of the common cabbage increased their preference for volatiles emitted from this plant to those emitted from Chinese cabbage, when the plants were infested, previously infested or mechanically damaged (Figure 2B). Interestingly, the response of the parasitoid to intact plants of the two species was not affected by prior experience (Figure 2B). When female parasitoids had an experience of ovipositing in a DBM larva feeding on a leaf of the Chinese cabbage, their preference for Chinese cabbage was further increased when the plants were infested with DBM larvae, but was unaffected when the plants were intact, mechanically damaged or previously infested (Figure 2A).

Figure 2. Numbers of Cotesia plutellae female adults showing a response to plant volatiles of Brassica campestris (Chinese cabbage) and B. oleracea (cabbage) after the adults had oviposited either in a host larva feeding on a Chinese cabbage leaf (A) or in a host larva feeding on a cabbage leaf (B). Asterisks indicate statistically significant preferences in a choice test (*: P<0.05, **: P<0.01, ***: P<0.001). NS: not significant. n: number of parasitoid wasps tested. The number of wasps that did not choose either of the odour sources is listed under “No choice”.

Discussion

When offered a choice between Chinese cabbage and common cabbage, C. plutellae showed a strong preference for searching and parasitising DBM larvae on the former (Tables 1 and 2). This innate preference, exhibited by naïve female parasitoids, is apparently mediated, at least in part, by the different levels of attraction from volatiles emitted by the two plants (Figure 1). However, this innate preference could be immediately and significantly modified by experience of searching and oviposition by the female parasitoids, to the extent that a ten-minute search coupled with a single oviposition experience on a host-damaged leaf of the common cabbage resulted in higher parasitism on this innately less-preferred plant than that on the innately-preferred plant (Table 3). This modification of preference by experience was evidently associated with learning of plant volatiles and to a much less extent with learning of host-derived stimuli (Figure 2).

Cotesia plutellae is a specific larval parasitoid of DBM, the latter can feed and reproduce on plants of at least 28 genera of the family Brassicaceae (Talekar & Shelton 1983). In fact, DBM has been observed to occur in high numbers on crops of many Brassica species in many parts of the world in the last 30 years (Talekar & Shelton 1993, other reports in this volume). Such an association between the three trophic levels, i.e., specific connection between the parasitoid and herbivore, but diverse connections between the herbivore and its host plants, has been speculated to favour the evolution of learning by the parasitoid to deal with the variability of plant cues (Vet et al. 1995). Our results offer support for this theory. In several parasitoids the learned responses were shown to wane and disappear in a short time relative to the adult’s longevity as a consequence of another experience (Vet et al. 1995, Fukushima et al. 2001). If this happens in C. plutellae, one may expect that the parasitoid could be more efficient in parasitising DBM on Chinese cabbage than on common cabbage when small plots of the two plants are grown in close proximity. Direct evidence for such differential levels of parasitism between these two plant species in the field is lacking. Liu et al. (2000) showed that rates of parasitism of DBM larvae by C. plutellae on mustard, B. juncea, a species more closely related to Chinese cabbage than to common cabbage, were usually higher than on common cabbage, when crop growth periods and levels of DBM density were similar between fields of the two plant species.

Our observation of the effects of plant volatiles and experience on the foraging behaviour of C. plutellae agrees with the earlier reports by Bogahawatte and van Emden (1996) and Potting et al. (1999), in that this parasitoid uses mainly plant volatiles in its orientation towards infested plants and experience of plant volatiles increases the preference to these cues. While the study by Bogahawatte and van Emden (1996) showed the influence of the host plant on which the parasitoid had developed and searched after emergence, our study and the study by Potting et al. (1999) demonstrate the dramatic effects of a brief experience by the adults on their foraging behaviour. Jiang (2001) further showed that experience acquired during the development of immature stages had limited influence on the preference for plants by the adults. This rapid learning ability by the adult females of C. plutellae may be utilised to increase its effectiveness for biological control. For example, mass-reared parasitoids pre-treated with odour of a target crop may search more efficiently and remain longer in the field of release.

Acknowledgements

This work was funded jointly by the China National Natural Science Foundation (Project No. 39930120) and the Australian Centre for International Agricultural Research (Project No. ACIAR CS2/1998/089). We thank Myron Zalucki, Department of Zoology and Entomology, the University of Queensland, Australia for helpful comments on the manuscript.

References