Selenium hyperaccumulation offers protection from cell disruptor herbivores

Abstract

Background: Hyperaccumulation, the rare capacity of certain plant species to accumulate toxic trace elements to levels several orders of magnitude higher than other species growing on the same site, is thought to be an elemental defense mechanism against herbivores and pathogens. Previous research has shown that selenium (Se) hyperaccumulation protects plants from a variety of herbivores and pathogens. Selenium hyperaccumulating plants sequester Se in discrete locations in the leaf periphery, making them potentially more susceptible to some herbivore feeding modes than others. In this study we investigate the protective function of Se in the Se hyperaccumulators Stanleya pinnata and Astragalus bisulcatus against two cell disrupting herbivores, the western flower thrips (Frankliniella occidentalis) and the two-spotted spider mite (Tetranychus urticae).

Results: Astragalus bisulcatus and S. pinnata with high Se concentrations (greater than 650 mg Se kg(-1)) were less subject to thrips herbivory than plants with low Se levels (less than 150 mg Se kg(-1)). Furthermore, in plants containing elevated Se levels, leaves with higher concentrations of Se suffered less herbivory than leaves with less Se. Spider mites also preferred to feed on low-Se A. bisulcatus and S. pinnata plants rather than high-Se plants. Spider mite populations on A. bisulcatus decreased after plants were given a higher concentration of Se. Interestingly, spider mites could colonize A. bisulcatus plants containing up to 200 mg Se kg(-1) dry weight, concentrations which are toxic to many other herbivores. Selenium distribution and speciation studies using micro-focused X-ray fluorescence (μXRF) mapping and Se K-edge X-ray absorption spectroscopy revealed that the spider mites accumulated primarily methylselenocysteine, the relatively non-toxic form of Se that is also the predominant form of Se in hyperaccumulators.

Conclusions: This is the first reported study investigating the protective effect of hyperaccumulated Se against cell-disrupting herbivores. The finding that Se protected the two hyperaccumulator species from both cell disruptors lends further support to the elemental defense hypothesis and increases the number of herbivores and feeding modes against which Se has shown a protective effect. Because western flower thrips and two-spotted spider mites are widespread and economically important herbivores, the results from this study also have potential applications in agriculture or horticulture, and implications for the management of Se-rich crops.

Figure 1

Western flower thrips feeding on A. bisulcatus. A, B: Thrips piercing A. bisulcatus leaflets, as viewed from the top (A) and side (B). C, D: Representative leaflets from high-Se (C) and low-Se (D) A. bisulcatus plants after exposure to thrips herbivory. Herbivory damage is apparent as white patches with black spots.

Figure 2

Selenium reduces thrips herbivory to A. bisulcatus in both choice and non-choice experiments. A-C: Thrips non-choice feeding experiment where thrips were offered only high-Se or low-Se plants. Herbivory was quantified as the percentage of entire A. bisulcatus young, medium and old leaves that showed herbivory (A) and as the percentage of leaflets per leaf that showed herbivory (B). The leaf Se concentration of the high-Se and low-Se plants used in the non-choice studies is shown in panel C. D-F: Thrips choice feeding experiments where thrips were provided with a choice between high-Se and low-Se plants. Herbivory was quantified as the percentage of A. bisulcatus young, medium and old leaves that showed herbivory (D) and as the percentage of leaflets per leaf (E) that suffered herbivory. The leaf Se concentration of the plants used in the choice study is shown in panel F. Values are means +/- SE. An asterisk above a pair of bars represents a significant difference between the high-Se and low-Se treatments (t-tests, α = 0.05, n = 6 for both high-Se and low-Se non-choice experiments, n = 4 for choice experiments).

Figure 3

Growth of spider mite populations feeding on high-Se or low-Se A. bisulcatus over the course of a non-choice feeding study (A) and a choice feeding study (B). Values are means +/- SE. An asterisk between data points in the non-choice or choice feeding experiments represents a significant difference between high-Se and low-Se plants (t-tests, α = 0.05).

Figure 4

Selenium added to spider mite-populated A. bisulcatus plants reduced spider mite population growth. A: Percent population change of established spider mite populations on A. bisulcatus over the course of a 3-week high-Se or low-Se treatment. B: Selenium concentration of plants at the beginning and end of the experiment. Values show means +/- SE. An asterisk between data points (A) or bars (B) represents a significant difference between the high- and low-Se treatments (t-test, α = 0.05, n = 10 for non-choice experiments, n = 7 for choice experiments).

Figure 5

Selenium speciation in spider mites collected from Se-rich A. bisulcatus plants. X-ray analysis of near-edge spectra (XANES) revealed that two-spotted spider mites collected from Se-rich A. bisulcatus plants (shown in panel A) contained primarily methylselenocysteine (B, C). The XANES Se spectra of the spider mites and of the methylselenocysteine standard compound are shown in panel C.

Figure 6

Selenium prevents thrips herbivory to S. pinnata in both choice and non-choice studies. A: Choice feeding experiment quantifying thrips herbivory of S. pinnata plants treated with or without Se, quantified as percentage of leaves per plant showing herbivory. B: Selenium concentration in plants treated with Se, comparing leaves that experienced thrips herbivory with leaves showing no herbivory. C: Elemental concentration of Fe, Mg, Mn, Mo, S and Zn in plants treated with Se, comparing leaves that experienced thrips herbivory with leaves showing no herbivory. Values are means +/- SE. An asterisk between a pair of bars represents a significant difference between the two treatments (t-tests, α = 0.05, n = 18).

Figure 7

S. pinnata plants show an age-related difference in leaf Se concentration. Shown is Se concentration in three young leaves from consecutive nodes. Values are means +/- SE. (Tukey-Kramer test, α = 0.05, n = 6).