Influence of polyploidy on insect herbivores of native and invasive genotypes of Solidago gigantea (Asteraceae)

Abstract

Herbivores are sensitive to the genetic structure of plant populations, as genetics underlies plant phenotype and host quality. Polyploidy is a widespread feature of angiosperm genomes, yet few studies have examined how polyploidy influences herbivores. Introduction to new ranges, with consequent changes in selective regimes, can lead to evolution of changes in plant defensive characteristics and also affect herbivores. Here, we examine how insect herbivores respond to polyploidy in Solidago gigantea, using plants derived from both the native range (USA) and introduced range (Europe). S. gigantea has three cytotypes in the US, with two of these present in Europe. We performed bioassays with generalist (Spodoptera exigua) and specialist (Trirhabda virgata) leaf-feeding insects. Insects were reared on detached leaves (Spodoptera) or potted host plants (Trirhabda) and mortality and mass were measured. Trirhabda larvae showed little variation in survival or pupal mass attributable to either cytotype or plant origin. Spodoptera larvae were more sensitive to both cytotype and plant origin: they grew best on European tetraploids and poorly on US diploids (high mortality) and US tetraploids (low larval mass). These results show that both cytotype and plant origin influence insect herbivores, but that generalist and specialist insects may respond differently.

Polyploidy, or the possession of more than two sets of homologous chromosomes, is a fundamental force in angiosperm evolution.^1,2 Many plant species or species complexes consist of multiple cytotypes that may occur sympatrically;³ this is an important source of genetic structure in plant populations that is often overlooked.⁴ Possession of multiple genomes may confer advantages to polyploid plants such as increased heterozygosity, a decreased probability of inbreeding depression, or a greater gene pool available for selection; these traits contribute to the widespread success of polyploids and may make them prone to invasiveness.^5,6 In a recent article,⁷ we examined the functional consequences of polyploidy for different cytotypes of Solidago gigantea Ait. (Asteraceae), collected from both its native range (North America) and its introduced range (Europe). In this addendum, we show how cytotype and continent of origin influence interactions of S. gigantea with insect herbivores. Interactions with herbivores are expected to vary with cytotype because of phenotypic changes associated with polyploidy, but this area has received little study (reviewed in refs. ^8–11). Plant origin, from either the native range or an introduced range, should also influence herbivores. Plants may escape from their specialist natural enemies in the introduced range, thereby experiencing reduced herbivore pressure from an insect community dominated by generalists.^12,13 Given sufficient time, plants from the introduced range may evolve to decrease investment in anti-herbivore defenses, particularly those effective against specialists.¹⁴ While a growing body of research has addressed whether plant defenses against herbivory are lower in the introduced range,^12,15,16 few of these studies have also examined the influence of cytotype.¹⁷

Three cytotypes of S. gigantea can be found in its native range in North America (diploid, tetraploid and hexaploid, 2n = 18, 36 and 54 respectively). These are morphologically indistinguishable and not generally treated as separate species.¹⁸ In Europe, where S. gigantea was introduced in the mid 18^th century,¹⁹ tetraploids are the dominant cytotype but diploids also occur. S. gigantea supports a diverse array of insect herbivores in its native range, but has few natural enemies in its introduced range.²⁰ We report here on experiments using both a generalist and a specialist leaf-chewing insect. The generalist, Spodoptera exigua (Lepidoptera: Noctuidae) is widely distributed and highly polyphagous, while the specialist Trirhabda virgata (Coleoptera: Chrysomelidae) feeds only on closely-related species within the genus Solidago. T. virgata is an outbreak insect that can be a major defoliator of S. gigantea and related species in North America.²¹ We grew plants originating from 10 populations in the US and 20 populations in Europe in common gardens at the University of Wisconsin-Milwaukee Field Station in Saukville, Wisconsin. There were five plant origin-cytotype combinations: three cytotypes from the US and two from Europe. Insects were reared on detached leaves from a single plant (Spodoptera) or on potted host plants (Trirhabda), for a set period of 21 d (Spodoptera) or until pupation (Trirhabda). We recorded insect survival and mass at the end of 21 d (Spodoptera) or at pupation (Trirhabda) (reviewed in ref. 22).

Overall survival was much better for the specialist Trirhabda than for the generalist Spodoptera (91% vs. 72%). Spodoptera larvae are not generally found on S. gigantea in the field, and while they are able to complete development, we found that this plant was not an ideal host. Spodoptera larvae were more sensitive to differences among cytotype and plant origin than were Trirhabda larvae. Percent survival was particularly poor for Spodoptera larvae reared on diploids from the US, where slightly more than half of the caterpillars survived for 21 days (Table 1). Trirhabda survival was consistently higher and did not show as great a range across the five ploidy-plant origin combinations. Mass of surviving larvae was also more variable for Spodoptera than Trirhabda (Fig. 1). Trirhabda pupal mass was remarkably consistent across the five ploidy-plant origin combinations. In contrast, Spodoptera larvae responded to both cytotype and continent of origin. Surviving Spodoptera larvae did particularly well on tetraploid plants from the introduced range (Europe), and particularly poorly on tetraploids from the US (Fig. 1). We have previously reported that Spodoptera grow better on plants from Europe;²² our current results reveal that this difference is due exclusively to better growth on tetraploid plants. However, our results also show that both diploids and tetraploids from the US were poor hosts for Spodoptera: diploids because they caused high mortality and tetraploids because they resulted in poor growth. These results indicate that plants from the introduced range have reduced defenses against herbivores, even when accounting for polyploidy.

Table 1.

Number and percent of insects surviving rearing experiments on host plants of different cytotypes of Solidago gigantea originating from the US (native range) or Europe (introduced range)

Spodoptera					Trirhabda
	No. Surviving	Initial No.	% Survival	No. Surviving	Initial No.	% Survival
US-Diploid	21	39	54	37	39	95
US-Tetraploid	70	93	75	82	92	89
US-Hexaploid	16	24	67	23	24	96
EU-Diploid	15	23	65	23	24	96
EU-Tetraploid	101	129	78	114	129	88

Insects were reared on a single genotype of each cytotype-origin combination for 21 days (Spodoptera) or until pupation (Trirhabda). Sample sizes for each cytotype-origin combination vary because cytotypes were not known at the time plants were collected; these distributions represent frequencies of cytotypes in our collections.

Figure 1.

Mass ± se of S. exigua (A) and T. virgata (B) larvae reared on host plants of different cytotypes of Solidago gigantea originating from the US (native range) or europe (introduced range). Means in A followed by different letters are significantly different at p < 0.05 (ANOVA followed by multiple Student's t-tests with Bonferroni correction). There were no significant differences in (B). Sample sizes for (A and B) shown in Table 1 (surviving larvae). Data for (A) In-transformed to improve normality for analysis.

Effects of the host plant on Spodoptera were probably driven, at least in part, by changes in secondary chemistry. We have previously shown that foliar terpenoids, chemicals known to influence insect herbivores,^23,24 are affected by both cytotype and continent of origin.⁷ It is surprising that Trirhabda larvae were not more sensitive to these differences in secondary chemistry among the five ploidy-origin combinations, given that Trirhabda is known to respond to host-plant chemistry.²³ We have previously reported that Trirhabda growth does not differ on European and US plants²² and show here that accounting for cytotype does not change this conclusion. In a recent study on the closely-related Solidago altissima, Halverson et al.¹¹ reported that the effects of plant cytotype on 5 gall-making herbivores were complex and not easily characterized. All five herbivores responded to plant cytotype, but for four of the five insects the most preferred cytotype was not consistent across sites. It is possible in our study that Trirhabda were responding to cytotype at a finer scale than that examined here. There may be differences due to cytotype that shift among the populations that we sampled, and that are averaged out when examined at the continental scale. We lack sufficient replication of cytotypes within populations to test this possibility. Even so, our results reported here reveal that plant cytotype can be an important source of variation affecting insect herbivores, but that generalist and specialist insects may respond differently.

Addendum to: Hull-Sanders HM, Johnson RH, Owen H, Meyer GA. Effects of polyploidy on secondary chemistry, physiology and performance of native and invasive genotypes of Solidago gigantea (Asteraceae) Am J Bot. 2009;96:762–770. doi: 10.3732/ajb.0800200.