Herbivory and Attenuated UV Radiation Affect Volatile Emissions of the Invasive Weed Calluna vulgaris

Abstract

Calluna vulgaris (heather) is an aggressive invasive weed on the Central Plateau, North Is., New Zealand (NZ), where it encounters different environmental factors compared to its native range in Europe, such as high ultraviolet radiation (UV) and a lack of specialist herbivores. The specialist herbivore Lochmaea suturalis (heather beetle) was introduced from the United Kingdom (UK) in 1996 as a biocontrol agent to manage this invasive weed. Like other plant invaders, a novel environment may be challenging for heather as it adjusts to its new conditions. This process of "adjustment" involves morphological and physiological changes often linked to phenotypic plasticity. The biochemical responses of exotic plants to environmental variables in their invaded range is poorly understood. The production and release of volatile organic compounds (VOCs) is essential to plant communication and highly susceptible to environmental change. This study therefore aimed to explore the VOC emissions of heather in response to different levels of UV exposure, and to feeding damage by L. suturalis. Using tunnel houses clad with UV-selective filters, we measured VOCs produced by heather under NZ ambient, 20% attenuated, and 95% attenuated solar UV treatments. We also compared VOC emissions in the field at adjacent sites where L. suturalis was present or absent. Volatiles produced by the same target heather plants were measured at four different times in the spring and summer of 2018-2019, reflecting variations in beetle's abundance, feeding stage and plant phenology. Heather plants under 95% attenuated UV produced significantly higher amounts of (E)-β-farnesene, decanal, benzaldehyde, and benzeneacetaldehyde compared to 25% attenuated and ambient UV radiation. We also found significant differences in volatiles produced by heather plants in beetle-present versus beetle-absent sites on most sampling occasions. We also recorded a lower number of generalist herbivores on heather at sites where L. suturalis was present. Interactions between invasive plants, a novel environment, and the native communities they invade, are discussed.

Keywords: biocontrol agents; heather beetle; plant ecophysiology; plant secondary metabolites; plant volatiles; ultraviolet radiation; volatile organic compounds.

Figure 1

Proportions of major chemical classes in heather plants exposed to ambient and attenuated UV (n = 10 for each treatment). Y-axis shows log10x + 1 transformed mean ± SE proportion of compound classes. Different letters indicate significant differences.

Figure 2

(a) Principal component analysis (PCA) biplot based on the volatile compounds produced by heather under ambient, 20% attenuated and 95% attenuated UV. (b) The relative proportion of compounds with higher contributions to PC1 and PC2 between the three UV treatments (n = 10). Y-axis of barplot (b) shows log10x + 1 transformed mean ± SE emission rate compounds and different letters indicate significant differences. Abbreviations in PCA: (E)-β-farnesene (EBF), (Z)-3-hexenal (Z3Hnal), (Z)-3-hexenol (Z3Hnol), (Z)-3-hexenyl acetate (Z3Ha), (Z)-3-Hexenyl butyrate (Z3Hbut), (Z)-3-hexenyl 2-methylbutyrate (Z3H2M).

Figure 3

Developmental stage and abundance of Lochmaea suturalis during the four volatile organic compound (VOC) measurements periods. Bars indicate the total number of L. suturalis present and pie charts show the proportion of each developmental stage at a given sampling time.

Figure 4

Comparison of major chemical classes between sites at (a) 14 November 2018, (b) 11 December 2018, (c) 31 January 2019 and (d) 25 March 2019. Y-axis shows log10x + 1 transformed mean ± SE proportion of chemical classes and asterisk indicate a significant difference. n = 7 for the beetle-present site in (c), otherwise 8 replicates per treatment at each sampling time. Asterisk indicates significant differences (* p < 0.05).

Figure 5

Non-metric multidimensional scaling (NMDS) plots showing the first (NMDS1) and second (NMDS2) axes of VOCs identified from heather at beetle-present and beetle-absent sites in (a) November 2018, (b) December 2018, (c) January 2019 and (d) March 2019. n = 7 for the beetle-present site in (c), otherwise 8 replicates per treatment were done in all sampling times. Abbreviations: (Z)-3-hexenyl benzoate (Z3Hben), δ-cadinene (dCad), (E)-2-hexanal (E2Hnal), (Z)-3-hexenyl 2-methylbutyrate (Z3H2M), (Z)-3-hexenyl valerate (Z3Hval), (E)-4,8-dimethyl-1,3,7-nonatriene (E-DMNT), (Z)-2-hexenol (Z2Hnol), germacrene D (GerD), (E,E)-α-farnesene (aFar), (Z)-3-hexenyl acetate (Z3Ha), (Z)-3-hexenol (Z3Hnol), (Z)-3-hexenal (Z3Hnal), (Z)-3-hexenyl butyrate (Z3Hbut), (Z)-3-hexenyl isobutyrate (Z3Hiso), (Z)-β-ocimene (Zbo), α-pinene (aPin), (E)-β-caryophyllene (EbC), o-cymene (oCym), hexyl acetate (HexAce), β-myrcene (bMyr), (E)-β-farnesene (EBF), (Z)-3-hexenyl hexanoate (Z3Hhex), (Z)-3-hexenyl isovalerate (Z3Hisoval), geranyl nitrile (GN), phenylethyl alcohol (PA), α-bourbonene (aBour), α-cubebene (aCub), benzyl alcohol (BA), α-gurjunene (aGur), α-selinene (aSel), α-terpineol (aTerp), β-pinene (bPin), γ-elemene (gEle), δ-guaiene (dGua), (E)-β-ocimene (EbO), (E)-2-hexenyl acetate (E2HAce), (E)-2-nonenal (E2Non).

Figure 6

Comparing the abundance of arthropods other than L. suturalis on heather between sites. Samples collected on (a) November 2018, (b) December 2018, (c) January 2019 and (d) March 2019 using the beating tray technique (n = 3). Bars show mean ± SE and asterisk indicates a significant difference between sites. Asterisk indicates significant differences (* p < 0.05).