Plant sulphur nutrition influencing host-plant selection and performance of insect herbivores

In the past several years, sulphur (S) deficiencies in agricultural crops have been reported with increasing frequency and therefore more interest has been directed into plant nutrition and fertilisation with particular respect to this element. In this context, special attention has been given to the economically important oilseed rape crop (Brassica napus), because of its high S need. Oilseed rape removes between 20 and 30 kg S/ha from the soil, corresponding approximately to double the demand of cereals. Until now, the effect of different S-supplies to plants in relationship to their pests has not attracted much attention. The aim of this study was therefore to investigate the influence of S-nutrition of rape on the preferences and performance of two main crucifer pests, the diamondback moth, Plutella xylostella (L.) (Lepidoptera; Plutellidae) and the cabbage root fly, Delia radicum (L.) (Diptera: Anthomyiidae). The larvae of P. xylostella attack the aerial part while D. radicum infest the roots of numerous wild and cultivated plants belonging to the Cruciferae family. Since the larvae of both insects have only a limited capacity to select alternative plants or roots in the soil autonomously, their survival depends largely on the host choice made by the ovipositing female. To assess the effect of S plant nutrition on the oviposition behaviour, females of both insects were exposed to plants grown under three different S-regimes: S-free (S0), normalS (Sn) and S-rich (S+) plants. In both insects the lack of S in the plant nutrient solution resulted in reduced oviposition, while differences between the two S fertilisation levels (normal and double field concentration) were smaller and not significant. In order to identify the plant characteristics influenced by S and perceived by the female insects, the ovipositional responses of the two crucifer pests were further tested by means of methanolic leaf surface extracts of the three types of B. napus. Using surrogate leaves treated with extracts of plants with an equivalent fresh weight, the same preferences were observed as in corresponding experiments with intact leaves, namely decreased oviposition in the absence of S. Also in both insects, the duration of larval development, from hatching to emergence, was significantly shorter and adults were heavier on Sn than on S0 plants. Comparing these same two parameters in Sn and S+ plants, a somewhat shorter development time on plants rich in S was recorded, but again this trend was not statistically significant. Larval feeding preferences of the diamondback moth were tested in a dual choice assay, using leaf discs. A significantly higher number of larvae preferred leaf discs of Sn plants rather than those of S0 plants. Further, the larvae preferred S+ to Sn discs. In the case of the cabbage root fly, the larval performance was evaluated using three additional intermediate sulphur levels between S0 and Sn. The percentage pupation at the end of larval feeding ranged from 6% (S0) to 32% (Sn), and the mean adult fresh weight of the emerging flies varied between 3.2 mg (S0) and 8.17 mg (Sn). Both the percentage of pupation and the adult fresh weight were positively correlated with the S content of the plant nutrient solution. Since GSLs and their volatile metabolites, the isothiocyanates, are known to stimulate oviposition in different insects, an analysis was made of the GSLs composition at the leaf surface of the three S variants. The analytical (GSL) data correlated with the host preferences of both the diamondback moth and the cabbage root fly. Furthermore, in the case of the cabbage root fly, a non-GSL compound, called CIF ("cabbage identification factor", thiatriaza-fluorenes) and known to be a powerful oviposition stimulant, was also more abundant in S fertilised plants. The electroanntenogram recordings (EAGs) obtained
from P. xylostella antennae confirmed that olfaction is an
important modality for the stimulation of oviposition in this
insect, as extracts from S fertilised plants caused more
stimulation than those from S0 plants. This might be due to
the higher content of isothiocyanates. In the case of D.
radicum, we recorded the response of the tarsal contact
chemoreceptor neurons to methanolic extracts. The specific
receptor neurones for CIF and GSL reacted more strongly
to the extracts from the two preferred plants (Sn and S+).
This shows that the fluorene compound CIF, together with
indolyl and benzyl GSLs, is involved in host acceptance
and confirms the analytical results. Our findings are in
agreement with earlier publications reporting that contact
chemoreception is the most important modality for the
stimulation of oviposition in the cabbage root fly.
In the final part of this work it was found that in dual
oviposition choice bioassays leaf extracts of Arabidopsis
thaliana plants stimulated oviposition in D. radicum. Both
the CIF and the GSL fractions of the plant extract stimulated
receptor neurons in the tarsal sensilla. This is an indication
that in addition to the known GSLs, A. thaliana also contains
CIF.
In conclusion, it appears that even a higher than optimal S
fertilisation will not lead to a significantly increased
accumulation or population outbreaks of the two harmful
pests investigated. Further, these findings confirm that
secondary plant metabolites play a crucial role in host
recognition by the insect. Plant nutrient solutions not only
affect directly the nature and concentration of secondary
metabolites but also indirectly the host-plant relationship,
modulating the host-choice and the plant suitability for bi-
(herbivores) and possibly tritrophic-(parasitoids) interactions.