Effects of elevated temperature on chemistry of an invasive plant, its native congener and their herbivores

Abstract

Climatic warming affects plant growth and physiology, yet how warming alters chemistry in invasive plants and indirectly affects herbivorous insects remains largely unknown. Here, we explored warming-induced changes in leaf chemistry of the invasive plant Alternanthera philoxeroides and its native congener Alternanthera sessilis, and further examined how these changes affected the performance of the herbivores, Cassida piperata and Spodoptera litura. We conducted a simulated warming experiment to address its effects on 13 leaf chemical traits of A. philoxeroides and A. sessilis. We measured growth and development time of two herbivores reared on plants from warming or ambient controls. Warming significantly affected leaf chemistry composition for both the invasive and native Alternanthera. Warming decreased nitrogen concentration in A. philoxeroides and increased total flavonoid and total phenol concentration in A. sessilis. The effects of warming on nutrients (i.e. fructose, sucrose, total soluble sugar and starch) varied with individual chemicals and plant species. Weight of C. piperata pupal and S. litura larval reared on warming-treated A. sessilis significantly decreased compared with non-warmed control, and a similar pattern was observed for weight of S. litura larval feeding on warming-treated A. philoxeroides. In addition, warming-treated A. sessilis significantly prolonged larval development time of S. litura. These results indicate that warming can directly affect the leaf chemistry in both invasive plant and its native congener, but these effects vary by species. Such differences in warming-induced changes in plant chemistry could indirectly affect herbivorous insects associated with the invasive and native plants.

摘要

增温对莲子草属入侵植物与本地同属植物化学物质组成和天敌昆虫的影响

气候变暖影响植物生长和生理活动，然而气候变暖如何改变入侵植物化学物质组成并间接影响其与植食性昆虫互作还少有报道。本研究以入侵植物空心莲子草(Alternanthera philoxeroides)及其本地同属 植物莲子草(A. sessilis)为对象，探究增温对其叶片化学物质组成的影响并进一步检验这些变化如何影响两 种植食性昆虫虾钳菜披龟甲(Cassida piperata)和斜纹夜蛾(Spodoptera litura)的生长发育。通过模拟增温实验，探究增温对空心莲子草和莲子草13个叶片化学物质的影响，并用其饲养两种植食性昆虫，测量它们的生长和发育时间。研究结果显示，增温显著改变了空心莲子草和莲子草叶化学物质组成；增温降低了空心莲子草叶片氮浓度，增加了莲子草叶片总黄酮和总酚浓度；增温对其它营养物质(果糖、蔗糖、总可溶性糖和淀粉)随物种和具体物质发生改变；采用增温处理的莲子草饲养的虾钳菜披龟甲蛹重和斜纹夜蛾幼虫重量，以及增温处理的空心莲子草饲养的斜纹夜蛾幼虫重量，显著低于对照不增温处理；此外，采用增温处理的莲子草饲养的斜纹夜蛾幼虫发育时间显著延长。这些结果表明，增温对植物化学物质组成的影响随物种发生变化，增温对入侵植物和本地植物化学物质组成的影响间接改变了其与植食性昆虫的互作关系。

INTRODUCTION

Growth and development rates of invasive plants may be affected by both climate and feeding of herbivorous insects (Dukes et al. 2009; Lu et al. 2013, 2015, 2016). The effects of climate warming on invasive plants include changes in their physiological processes, growth, development, reproduction and phenology, as well as their distribution (Fletcher et al. 2020; Lu et al. 2019; Luo et al. 2020; Wu et al. 2016). Climate warming may also affect an insect’s biology and ecology, both directly and indirectly by changing host plant growth and resource allocation (i.e. food quality, plant biomass and photosynthesis) (Bauerfeind and Fischer 2013; Jamieson et al. 2015; Liang et al. 2013; Lin et al. 2010). Therefore, predicting the response of herbivorous insects to invasive plants under climate warming is complex. Such studies which could advance our understanding of the consequence of climate warming on invasive plants, insects and their interactions, however, are still rare.

Plant nutrients are sensitive to temperature and may be altered by climate warming. For instance, Hobbie et al. (2002) found that sugar concentration in Douglas fir needles (Pseudotsuga menziesii (Mirb.) Franco) decreased with elevated temperature. A meta-analysis showed that in general, plants species exposed to elevated temperature may have reduced total carbohydrates in leaves, causing reductions in leaf nutritional quality, while leaf nitrogen did not show any consistent pattern (Zvereva and Kozlov 2006). Many studies have found that both plant nitrogen and carbohydrates (i.e. sugar) are important components of herbivore nutrition and determinants of plant palatability (An et al. 2005; Lemoine et al. 2013; Zvereva and Kozlov 2006), affecting insect herbivores development. For example, Jamieson et al. (2015) found that greater nitrogen concentrations of leaves were associated with reduced development time and increased food conversion efficiency of insect herbivores. Similarly, changes in invasive plant nutrients induced by elevated temperature may also affect the development of herbivorous insects feeding on the affected invasive plants, but this prediction has not yet been tested.

Elevated temperature may also indirectly affect herbivorous insects by changing plant secondary metabolites. For instance, Sallas et al. (2003) found that warming increased concentrations of leaf terpenoids in both Norway spruce (Picea abies L.) and Scots pine (Pinus sylvestris L.), though it did not affect phenols. Dury et al. (1998) found that the higher condensed tannin levels in Pendunculate oak (Quercus robur L.) exposed to elevated temperature might negatively influence the development and fecundity of herbivorous insects feeding on the tree. Together, climate-induced changes in nutrition, palatability, digestibility or toxicity of plant leaves may affect insect performance. However, how climate warming affects such secondary chemicals and then indirectly changes the development of the plant’s herbivores in the context of plant invasions has received little attention. Addressing these questions could provide new insights into understanding of mechanisms governing interactions between invasive plants and insects under climate change.

Here, we report the impact of elevated temperature on the nutrients and secondary chemicals in the invasive plant alligator weed (Alternanthera philoxeroides (Mart.) Griseb (Amaranthaceae)), and the subsequent effects of these changes in plant chemistry on associated herbivorous insects. Alternanthera philoxeroides, originating from South America, has been widely invasive in parts of Asia, Australia and North America, where it causes serious threats to ecosystems and the economy (Julien et al. 1995). Across most of its distribution in China, several native insect herbivores, including the oligophagous beetle Cassida piperata Hope (Coleoptera: Cassididae), the generalist defoliator Spodoptera litura Fabricius (Lepidoptera: Noctuidae), Hymenia recurvalis Fabricius (Lepidoptera: Pyralidae), Plutella xylostella Linnaeus (Lepidoptera: Plutellidae), Herpetogramma basalis Walker (Lepidoptera: Crambidae), Oxya chinensis Thunberg (Orthoptera: Acrididae), the sucking insect various aphids (Hemiptera: Aphididae) and the introduced biocontrol beetle Agasicles hygrophila (Coleoptera: Chrysomelidae)—commonly feed on and damage this plant (Dai et al. 2014; Dan et al. 2015; Liu et al. 2018, 2022; Lu et al. 2019). These insects are also associated with Alternanthera sessilis (Linn.) R. Br. ex DC, a native congener of A. philoxeroides, with a similar distribution to the invader in China. Previous studies have reported that A. philoxeroides could expand its distribution further north in China due to climate warming, and this expansion may affect the distribution of its associated herbivorous insects (Lu et al. 2013, 2015, 2016). However, how warming will affect native herbivorous insects due to changes in A. philoxeroides’ nutrients and secondary chemicals is unknown. These knowledge gaps are critical for predicting the effects of A. philoxeroides invasion on native insects and for determining its management under climate change.

In this study, we conducted a simulated warming experiment, laboratory chemical analysis and insect performance bioassays to investigate the effects of elevated temperature on A. philoxeroides nutrients and secondary chemicals and their effects on the growth and development of its associated native herbivores, S. litura and C. piperata. These two insects are mostly found on the plant cross the distribution range, thus well representing native insects feeding on the invader in China (Liu et al. 2018; Wei et al. 2016). We also included the native congener A. sessilis to examine whether the patterns of growth and development of two native herbivores vary with host plants (invasive versus native) under simulating climate warming. Specifically, we addressed the following questions: (i) Whether the effects of elevated temperature on the leaf chemistry differed between the invasive plant A. philoxeroides and the native congener A. sessilis? (ii) How do these changes in nutrients and secondary metabolites affect the growth and development of two native herbivorous insects?

MATERIALS AND METHODS

Study species

Alternanthera philoxeroides is an amphibious perennial plant that reproduces asexually from stem or root buds in invaded areas, where it outcompetes native plants in both terrestrial and aquatic ecosystems. Introduced into China in the 1930s, it has become invasive in southern China (Julien et al. 1995). It rarely sets seed, and seeds produced are usually not viable in its invaded ranges. Its only native congener in China, A. sessilis, is an annual or perennial herb, co-occurring with A. philoxeroides in many habitats that can reproduce both asexually and through seeds (Lu et al. 2014; Wang et al. 2019).

Spodoptera litura is a polyphagous insect with a broad host range including plants in approximately 40 families (Rao et al. 1993). It is widely distributed across Asia, Australia, Europe, Africa, the south Pacific and Hawaii (Nagoshi et al. 2011). It has four to five generations per year and overwinters 3–5 cm below the soil surface as pupae. In China, an oligophagous leaf beetle, C. piperata, feeds on several plants in the Amaranthaceae and Chenopodiaceae (Dai et al. 2014). Naturally occurring in northeast Asia and most areas in southeast Asia, C. piperata has three to four generations per year, and both its larvae and adults feed on leaves of A. philoxeroides and A. sessilis.

For this study, C. piperata were collected from an experimental field of Henan University (34°49′22″ N, 114°18′25″ E) in Kaifeng, Henan Province, China, from Chenopodium album Linn. (Chenopodium: Chenopodiaceae) and then reared on A. philoxeroides and A. sessilis in an insectary. Field collected insects were reared for one generation before the experiment to reduce possible maternal effects. The offspring of the field collected insects were used as larvae (first instar) for the experiment. We obtained S. litura pupa from a commercial supplier (Ke Yun Biocontrol Company, Jiaozuo, Henan Province, China) in May 2018.

Field experiment

To assess the effects of A. philoxeroides grown at elevated temperature on the growth and development of native herbivores, we conducted a simulated warming experiment in a farmland with infrared heating on the open-air test field for 2 years (2017–2018) at Henan University in Kaifeng (34°49′22″ N, 114°18′25″ E, 73 m a.s.l.). This experimental field was located in a temperate monsoon region characterized by hot and rainy summers, and cold and dry winters. Mean precipitation is 650 mm and mean annual temperature is 14 °C (http://ha.cma.gov.cn/kaifeng/).

Before the experiment, the field was thoroughly ploughed and the dominant weed species were suppressed by mowing and hand weeding. We set up six test plots (3 m × 4 m), three of which had MSR-2420 infrared heaters (2000 W, Kalglo Electronics, Bethlehem, PA, USA), suspended 2.05 m above the soil surface, that were capable of increasing the air temperature by about 2 °C above the 10 cm above ground. In the remaining three control plots, we used a dummy heater of the same shape and size as the infrared one. The position of the heating and control plots were randomly assigned and the distance between plots was >1.5 m. In early August 2017, we dug six soil pits (10 cm in depth) on the six test plots and filled with thoroughly mixed growth media of 2:1 top soil and fertile soil (Peilei Organic Fertilizer, Zhenjiang, China) with the content of pH of 5.5–7.5, soil EC value of 1.0–3.5 and total nutrient content (N + P₂O₅ + K₂O) of 1.0%–5%. We watered each plot every day during the course of the experiment to minimize the potential effects of infrared heaters on soil water content (Lu et al. 2016; Wan et al. 2002).

On 11 August 2017, we collected 60 stem cuttings each (approximately 15 cm in length, each with three nodes) of both A. philoxeroides and A. sessilis from local populations on the campus of Henan University. These stems cuttings were randomly assigned into each plot, with each receiving 10 stem cuttings of A. philoxeroides and 10 stem cuttings of A. sessilis. In each plot, stem cuttings of the two species were interplanted in two rows, and the distance between each stem cutting in a row was 40 cm. All the plots were immediately covered with 3 m × 4 m × 1.5 m nylon cages to exclude unwanted herbivores during the course of the experiment. We then started year-round warming. We monitored A. philoxeroides and A. sessilis individuals per plot in early spring 2018. The plants of A. philoxeroides, which asexually propagate from internodes or belowground tissues, emerged on 3 March, while the seedlings of A. sessilis, which propagate from seeds, emerged on 21 April. As A. philoxeroides grows quickly, dominating over and inhibiting the growth of A. sessilis, we cut and thinned A. philoxeroides seedlings to the same number and size of A. sessilis in each plot on 21 May 2018.

Laboratory bioassay of plant effects on target herbivores

To test the indirect effects of warming on the growth and development of two native insects—S. litura and C. piperata, we conducted a laboratory bioassay from July to August in 2018. We collected leaves of both plant species from both the warming and control plots and fed the leaves to the two native herbivores that were held in an insectary (27 ± 1 °C, 65%–80% RH and a 14 L:10 D h photoperiod). We had 20 biological replicates for each treatment. Newly hatched C. piperata larvae were placed in labeled Petri dishes (9 cm diameter) lined with a moistened wet filter paper, while newly hatched S. litura larvae were placed in plastic box (142 mm × 112 mm × 52 mm) and fed individually. Fresh leaves were provided daily for larvae throughout the experiment until pupation. We estimated development time through the daily monitoring and removal of molts for each life stage of C. piperata and S. litura from hatching to adult death. We measured S. litura larval weight (from hatching to 15 days) and pupal weight using an electronic balance (GL124-1SCN, SARTORIUS), recorded development time from hatching to pupation and calculated the survival rate from the first instar of larvae to the adult stage of development. We also used the above method to determine the pupal weight and the survival rate of C. piperata.

Chemical analyses of plant leaf tissue

Leaf samples for chemical analysis were collected from plants used to feed insects in the experiment described above. The leaves were haphazardly collected from individual plants (from the three to four top stem nodes), and the leaves were taken immediately to the laboratory and held at −80 °C before analysis. Thirteen leaf chemical traits were divided into two categories related to nutrients and secondary metabolites (Table 1). Normally, we had one or two replicates per plot for each species to assess leaf chemical traits. Leaves were dried for 48 h using a vacuum freeze drier (Freeze dry system, Labconco Corp, Kansas City, MO, USA), and then samples were ground for two minutes at a frequency of 30 times per second in grinding machine (Retsch MM400, Haan, Germany), and then kept sealed under dry conditions until further preparation for chemical analyses.

Plant total carbon and total nitrogen were analyzed using an elemental analyzer (Thermo Electron Corporation, Waltham, MA, USA), and then the ratio of C to N was calculated. Total soluble sugar and cellulose were determined using anthrone–sulfuric acid colorimetry (Dubois et al. 1956), with absorption at 620 nm recorded on a spectrophotometer (Thermo Spectronic Genesys 10 UV, Thermo Fischer Scientific Inc., Waltham, MA, USA). We determined fructose and sucrose following Albalasmeh et al. (2013), using a resorcinol method in which these sugars were measured spectrophotometrically at 480 nm. Starch was determined using a 3,5-dinitrosalicylic acid method, with a maximum absorption at 540 nm (Yu et al. 1997). Lignin was extracted in 1 mL 25% bromoacetyl-acetic acid and 0.1 mL perchlorate, heated in a water bath to 70 °C for 30 min, and then adding 2 mL 2 mol L⁻¹ sodium hydroxide and 5 mL acetic acid, and blending the mixture. Finally, the volume was adjusted to 10 mL with acetic acid, and the absorbance by UV spectrophotometry at 280 nm was measured (Fukushima and Hatfield 2001). Total phenol content was determined by the Folin–Ciocalteu assay, in which the sample was dissolved in 70% ethanol, at a 1:3 ratio of material/liquid, and the absorbance measured at 765 nm, using gallic acid as a standard (Kamtekar et al. 2014). Total flavonoid content was measured by the aluminum nitrate colorimetric method, in which the sample was dissolved in 70% ethanol, at a 1:3 ratio of material/liquid, subjected to ultrasound for 40 min for ultrasonic time, and the absorbance measured at 510 nm, using rutin as a standard (Kamtekar et al. 2014). Tannin content was measured using the H₂SO₄-Vanillin assay (Yang et al. 2010), in which the medium was comprised of 1 mL of sample, 2.5 mL of 1% vanillin solution in methanol and 2.5 mL of 5 mol L⁻¹ H₂SO₄ solution in methanol, and the reaction was carried out at 30 °C for 20 min. The absorption of the resultant solution was then measured at 500 nm, using catechin as a reference standard. The determination of total triterpenoid saponins by spectrophotometry was measured by absorbance at 215 nm, using oleanolic acid as standard (Wang 2011).

Statistical analyses

For plant data, non-metric multidimensional scaling (NMDS) was used on 13 chemical traits to depict the variation in leaf chemistry among the different species identity and warming treatments. A permutational multivariate analysis of variance (PERMANOVA) was also performed to detect the effects of species, warming treatment and their interaction on leaf chemistry. We further calculated the dissimilarity in leaf chemical traits using Euclidean distance, with a higher Euclidean distance indicating greater dissimilarity in leaf chemistry among different species and warming treatments. One-way analysis of variance (ANOVA) was used to test differences in leaf chemical traits among species and warming treatment for each individual trait. For insect data, two-way ANOVA was first used, in which plant species identity and warming were fixed factors, to test their effects on the performance of C. piperata (i.e. pupal weight and survival rate) and S. litura (i.e. larval and pupal weight, survival rate and larval development time). One-way ANOVA was performed when marginal or significant interaction between species identity and warming was detected. Tukey HSD post hoc comparisons with Benjamini–Hochberg correction for P values were used to test differences among treatment means. Prior to analyses, the normality of data and homogeneity of variances were tested using the Shapiro–Wilk’s test and Bartlett’s test, respectively. Ratio of C to N and lignin was log transformed while C. piperata pupal weight, S. litura larval weight and survival rate were square-root transformed to improve normality. All statistical analyses were performed using R v.3.5.1 (R Core Team 2018) with packages of vegan (Oksanen et al. 2019), MASS (Venables and Ripley 2002) and multcomp (Hothorn et al. 2008).

RESULTS

Effects of warming on leaf chemicals of Alternanthera philoxeroides and A. sessilis

The NMDS ordination revealed that leaf chemistry among different species and warming treatments separated distinctly from each other along the two first NMDS axes (Fig. 1), which were closely related to chemicals, such as starch, total sugar, sucrose, total flavonoid and total phenol. The PERMANOVA results further confirmed that leaf chemical traits composition differed significantly between different species, warming treatment and their interaction (Supplementary Table S1). Specifically, warming increased sucrose concentration and decreased starch concentration in leaf for both A. philoxeroides and A. sessilis (Table 1). For A. philoxeroides, warming decreased nitrogen and total sugar concentration, but increased fructose concentration in leaf (Table 1). For A. sessilis, warming decreased fructose concentration, but significantly increased total sugar, total flavonoid and total phenol concentration in leaf (Table 1). In addition, Euclidean distance induced by warming in A. sessilis was greater than in A. philoxeroides (Supplementary Table S2), indicating leaf chemistry in A. sessilis was more sensitive to warming than that in A. philoxeroides.

Effect of warming-treated plants on insect performance

The pupal weight of C. piperata reared on A. sessilis was higher than those reared on A. philoxeroides in ambient condition, but this difference disappeared under warming conditions (Table 2: marginal interaction; Fig. 2a). Warming significantly reduced S. litura larval weight, but this inhibitive effect was greater in A. sessilis than in A. philoxeroides (Table 2: significant interaction; Fig. 3a). The survival rate and pupal weight of S. litura were higher in A. sessilis than in A. philoxeroides (Table 2; Fig. 3b and c). Warming-treated plants marginally reduced pupal weight and significantly prolonged larval development time of S. litura, respectively (Table 2; Fig. 3c and d, but the increased larval development time was only observed in A. sessilis (Table 2: marginal interaction). There was no difference in other treatments (Table 2).

DISCUSSION

The responses of invasive plants to climate warming have received considerable attentions (Ren et al. 2021; Sorte et al. 2013; Walther et al. 2002; Wu and Ding 2019); however, little is known about how climate warming affects insect performance indirectly via changes in the plant chemistry in the context of plant invasion. In this study, we found that warming indirectly reduced weight and prolonged larval development time of native herbivores through altering plant nutrients and secondary chemicals concentrations in the native plant A. sessilis, while changes in nutrients concentrations of the invasive A. philoxeroides only impacted S. litura weight. These findings might help us understand how climate warming influences the interactions of invasive and native plants with insects, and suggest that elevated temperature may shift the interaction of native herbivores, native plants and invasive plants.

Variation in leaf chemistry induced by elevated temperature

Temperature plays a key role in regulating plant growth and physiology by influencing leaf chemical traits (Bidart-Bouzat and Imeh-Nathaniel 2008; Zvereva and Kozlov 2006). A previous study found that elevated temperature could reduce leaf carbohydrates, causing decreases in leaf nutritional quality (Zvereva and Kozlov 2006). In our study, elevated temperature affected leaf chemistry composition of both the invasive and congeneric native Alternanthera, but dissimilarity induced by warming in leaf chemistry of native A. sessilis was greater than in invasive A. philoxeroides. These findings support the prevailing views and our first hypothesis that warming influences plant metabolites in multiple ways, but the magnitude and direction of impacts are strongly species dependent.

Although the invasive A. phioxeroides and native A. sessilis belong to same genus of Alternanthera, sharing similar phylogeny and morphological traits, they exhibited significant differences in individual chemical response to elevated temperature. Plant nutrients and secondary chemicals are well known to influence plant fitness, varying with abiotic environments and natural enemies (War et al. 2012). Thus variation in these chemical traits under warming condition and among species may result in changes in plant growth and plant–herbivores interaction, further influencing plant survival, competitiveness and distribution (Jamieson et al. 2015; Lemoine et al. 2013; Lu et al. 2013, 2015). In this study, we found elevated temperature significantly reduced nitrogen concentration in invasive A. philoxeroides, while increased the contents of total phenol and total flavonoid in native A. sessilis, suggesting that the chemistry of invasive and native plants may reacted differently to elevated temperature (Bhattarai et al. 2017; Kuebbing and Nunez 2016). Relative to native plants, invasive plants normally face less stress from herbivores and allocate more resources to traits associated with growth, which may result in different chemical responses in native and invasive species under elevated temperature. Further work needs to identify factors triggering heterogenous variation in chemistry between native and invasive plants in response to warming.

Effects of warming-treated plants on native herbivores

The most novel finding of our study was that warming shifted the interactions of native herbivores with the native plant more strongly than invasive plant, as indicated by C. piperata pupal weight, S. litura larval weight and larval development time. This finding supports the prevailing view that changes in chemistry in host plants will further affect insect performance. Our results showed that the weight of C. piperata pupal reared on A. sessilis decreased under warming conditions, even though C. piperata showed higher larval survival rate on A. sessilis than on A. philoxeroides. Moreover, warming treatment significantly prolonged S. litura larval development time and such negative effects only occurred in A. sessilis. These results suggest that the indirect impact of warming on the insect is stronger via the native plant than the invader. These findings, combined with changes in plant nutrients and secondary chemicals, indicate that rapidly increased temperature could drive changes in plant metabolites, which may in turn modify the ways plants interact with other organisms, and may even alter the competitive outcome between invasive and native congeners in the future climate context.

Here, we propose several hypotheses to explain the differential responses of native and invasive plants interactions with native herbivores under warming conditions. First, trade-offs between nutrients and secondary chemicals in plants may affect insect performance. A meta-analysis demonstrated that insect herbivores will likely prefer nutritious individuals and avoid plant with high levels of non-nutritional plant chemicals, including secondary compounds such as phenolics and flavonoids (Zust and Agrawal 2017). Second, the ability of native herbivores to exploit invasive plants often differs between specialists/oligophagous and generalists. For example, specialist Gadirtha inexacta Walker (Lepidoptera: Noctuidae) grew larger on and consumed more plant biomass when fed on invasive Chinese tallow Triadica sebifera (Lour.) than on native tallow populations, while the generalist Cnidocampa flavescens Walker (Lepidoptera: Limacodidae) showed the same performance (i.e. larval biomass, larval development time) between them (Huang et al. 2010). In our study, we found that warming reduced fructose and starch, but increased total phenol and flavonoids in A. sessilis, which may have been the reason for the reduction in the larval weight of S. litura, the prolong of the larval development time, and the decrease of the pupal weight of C. piperata. Moreover, the influence of individual chemicals on insect performance depends on the relative proportions of nutrients and secondary metabolites, such as changes in proteins, carbohydrates, tannins and phenolics (Behmer 2009; Jamieson et al. 2015).

Climate warming may affect insect performance via a number of mechanisms, including direct effects on insect physiology as well as indirect effects on insects through plants. In this experiment, we attempted to address the phytochemical mechanisms behind the effects of climate warming on non-native plants and native herbivores insects. Jamieson et al. (2015) found that long-term study is needed to evaluate the underlying phytochemistry mechanisms affecting plant response to abiotic and biotic factors. However, warming disproportionately affected the content of plant secondary metabolites in this experiment, which may be partially because the short-term warming had little effect on chemical defense. Moreover, this study only included one pair of invasive and native host plants and two native herbivores. Long-term, large-scale experiments involving multiple species are needed to clarify the phenomenon and associated mechanisms. In addition, climate warming is often taking place in conjunction with other components of global change, such as nitrogen deposition and greenhouse gas emissions. Thus, studying the joint effects of climate warming with other factors on plant defense metabolomics, the relationship between invasive plants and native herbivores may be important in predicting alien plant invasions in the future.

CONCLUSIONS

In summary, our results suggest that warming could directly influence plant chemical traits and then indirectly affect native herbivores performance, but these effects vary with species identity. In addition, a shift in the interaction between native herbivores and native plants may result in increasing competitive ability of native species against the invader and a higher preference for insects to choose invasive species as their host plant, which in turn, may increase the impact of herbivores on invasive plants. Understanding these relationships is critical for predicting the effects of alien plant invasions on native insects and improvement of management for invasive plants under climate change.

Supplementary Material

Supplementary material is available at Journal of Plant Ecology online.

Table S1: Results of PERMANOVA statistics for the effects of species identity (Species), warming treatment (Warming) and their interaction on leaf chemical traits.

Table S2: Dissimilarity of leaf chemical traits among different species identity and warming treatment, as indicated by Euclidean distance.

Acknowledgements

We are grateful for the assistance of all staff members in the Biotic Interaction and Biosecurity Laboratory, Henan University at Kaifeng, Henan, China.

Funding

This work was supported by the National Key Research and Development Program (2017YFC1200104).

Conflict of interest statement. The authors declare that they have no conflict of interest.

REFERENCES

Author notes