Selection for Chemical Trait Remixing in an Invasive Weed After Reassociation with a Coevolved Specialist

Home , Isopimpinellin

Selection for chemical trait remixing in an invasive SPECIAL FEATURE weed after reassociation with a coevolved specialist

A. R. Zangerl*, M. C. Stanley†, and M. R. Berenbaum*‡

*Department of Entomology, 320 Morrill Hall, University of Illinois at Urbana-Champaign, 505 South Goodwin, Urbana, IL 61801-3795; and †School of Biological Sciences, The University of Auckland, Private Bag 92019, Auckland Mail Centre, Auckland 1142, New Zealand

Edited by Jerrold Meinwald, Cornell University, Ithaca, NY, and approved December 19, 2007 (received for review November 1, 2007)

The interaction between Depressaria pastinacella (parsnip web- predictable because the decoupling of defense and herbivory worm) and wild parsnip (Pastinaca sativa), in its native Europe and that has been observed in interactions between invasive plants in its longstanding nonindigenous range in the midwestern United and indigenous herbivores (11) is less likely to occur upon States, is characterized by chemical phenotype matching, ostensi- reassociation with a coevolved specialist. Whereas compelling bly mediated by reciprocal selective responses. The first appear- evidence for rapid contemporary evolution of size, fecundity, ance of D. pastinacella on P. sativa in New Zealand in 2004 provided and leaf area of invasive plant species has been found (12), the an opportunity to quantify selective impacts of a coevolved her- potential for evolutionary change of chemical traits, upon which bivore and calibrate rates of phytochemical response in its host resistance may be based, upon reassociation with a coevolved plant. Webworms in 2006 reduced seed production up to 75% in enemy has yet to be explicitly evaluated. The geographic mosaic New Zealand populations, and in 2007 infestations increased in theory of coevolution (2) specifically predicts rapid reciprocal severity in all populations except one. Most New Zealand popu- responses, including trait remixing, to result from intense selec- lations fall into a furanocoumarin phenotype cluster distinct from tion exerted by a reassociated coevolved specialist, and historical European and U.S. phenotypes, although one heavily attacked evidence (13) is suggestive of such responses. population clusters with two U.S. populations and one European A system in which the chemical consequences of reassociation population long associated with webworms. Multivariate selection with a coevolved herbivore may be examined involves the analysis substituting realized fitness (with webworms present) for introduced Eurasian weed Pastinaca sativa, wild parsnip, and its potential fitness (absent webworms) as the dependent variable European insect associate Depressaria pastinacella, the parsnip revealed that reassociation with a coevolved specialist in a non- webworm. This interaction has been well characterized within its indigenous area profoundly altered the selection regime, favoring native range in Europe and in its introduced range in North trait remixing and rapid chemical changes in parsnip populations, America, and the ecological effects of a diversity of plant as predicted by the geographic mosaic theory. That uninfested chemicals on the principal specialized herbivore have been populations of New Zealand parsnips contain higher amounts of determined. Brought to America by the earliest colonists, the octyl acetate, a floral volatile used by webworms for orientation, parsnip was ‘‘common’’ by 1630 (14). Escaped from cultivation, suggests that plants that escape from specialized enemies may also it is regarded as noxious because all aerial parts produce EVOLUTION experience selection to increase kairomones, as well as to reduce furanocoumarins, phototoxic allomones that, on contact with allomones. human skin, cause blistering and hyperpigmentation (15). Very few native North American insects have colonized P. sativa (16); insect–plant interactions ͉ Lepidoptera ͉ Pastinaca sativa ͉ parsnip currently, the principal (and occasionally only) herbivore webworm ͉ herbivore throughout its range in North America is D. pastinacella,acci- dentally introduced from Europe and first reported in 1869 in Ontario, Canada (17). D. pastinacella webs together and feeds on lthough the ability of herbivorous insects to act as selective the reproductive structures of species in the closely allied genera agents on the chemistry of their host plants is an essential A Pastinaca and Heracleum; since their introduction, webworms component of coevolutionary theory (1, 2), experimental dem- have become established widely in North America (18). onstrations of such impacts are few (3–5). Accordingly, some Certain furanocoumarins in wild parsnip function as resis- skepticism remains as to the efficacy of herbivores as selective tance factors against webworms in North America (19, 20) and agents (6), and alternative explanations, including top-down Europe (21). Webworm damage in the Midwest United States selection from natural enemies of herbivores, have proliferated selects for increased concentrations of three furanocoumarins: (albeit also without an abundance of experimental demonstra- xanthotoxin, bergapten, and sphondin (3, 20). Genotypes with tions). In addition to its importance in a theoretical context, high levels of these furanocoumarins experience lower fitness in quantifying selection and rates of response is critical for evalu- the absence of herbivores than genotypes with lower furanocou- ating the stability of host-plant resistance traits in agricultural marin content (22), indicative of a cost of producing these crops as well as the sustainability of classical biocontrol of weeds. compounds. These three furanocoumarins act as resistance One way to examine the rate and magnitude of evolutionary factors in part because webworms metabolize them less effi- change in phytochemical profiles in response to insect herbivory ciently by cytochrome P450 monooxygenases (23). In both is to examine an interaction involving coevolved species that are Europe and North America, interactions between parsnip separated and then, by sequential range extensions, reassociated. Reassociation with a specialized coevolved enemy in an area of nonindigeneity is likely to have a profound and more predictable Author contributions: A.R.Z. and M.R.B. designed research; A.R.Z. performed research; effect on host-plant chemistry. Although little quantitative and M.C.S. contributed new reagents/analytic tools; A.R.Z. and M.R.B. analyzed data; and A.R.Z. ecologically relevant information is available on phytochemical and M.R.B. wrote the paper. changes in plants that occur after introduction into a nonindig- The authors declare no conflict of interest. enous area and release from interactions with longtime insect This article is a PNAS Direct Submission. associates (7–10), even less information is available on phyto- ‡To whom correspondence should be addressed. E-mail: [email protected]. chemical changes that ensue when coevolved enemies that are This article contains supporting information online at www.pnas.org/cgi/content/full/ demonstrated reciprocal selective agents resume interacting with 0710280105/DC1. a host plant in an area of invasion. Such changes should be more © 2008 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710280105 PNAS ͉ March 25, 2008 ͉ vol. 105 ͉ no. 12 ͉ 4547–4552 Downloaded by guest on September 27, 2021 Fig. 1. Realized fitness of wild parsnips in six N.Z. populations after consumption of reproductive parts by parsnip webworms. Average potential fitness (number of male flowers) of plants within a population is given below each pie chart. Estimates of potential fitness for many plants was not possible because webworms in addition to consuming all of the umbels had as well consumed the stalks bearing the umbels.

webworms and wild parsnips are characterized by chemical near the center of the outbreak in the city of Dunedin were phenotype matching: close correspondence between furanocou- heavily infested in 2006: Crimp, Lumber, and Townley. A fourth marin defense profile of the plants and detoxification capacity of population, Warrington, at the northern edge of the range and the webworms. Ϸ15 km from the presumed point of introduction [Port Chalmers Before the appearance of webworms in North America, (25)], was lightly infested. By 2007, the northernmost edge of the parsnips exhibited reduced furanocoumarin defenses in com- range extended 5 km to include two additional populations of parison with contemporaneous European parsnips, as deter- wild parsnip (Rock and Ocean). mined by examination of herbarium specimens collected in Webworms in 2006 removed half or more of the total fitness North America over 152 years spanning the period before of Ϸ50% of the plants in the Crimp and Townley populations and introduction of webworms to the present, and European speci- eliminated seed production by Ͼ75% of the plants in the Lumber mens from the 19th century to the present, for webworm damage population (Fig. 1). In 2007, infestation severity increased in all and furanocoumarin content (24). Concomitant with the rise in of these populations except Crimp; 75% of the plants in the webworm infestation between 1890 and 1909, levels of all five Lumber and Townley populations failed to produce seed, and furanocoumarins (those previously associated with resistance as nearly half of the Warrington plants, only lightly infested in 2006, well as imperatorin and isopimpinellin) increased significantly failed to produce seeds. and continued to increase thereafter. Although it is likely that Ripe seeds of N.Z. wild parsnips differ substantially in chem- webworm herbivory contributed to selection for the phytochemical composition from those of European and U.S. parsnips (Fig. ical shifts, it is impossible to assign responsibility conclusively a 2). Almost all of the N.Z. populations fall into a furanocoumarin century after the fact. phenotype cluster distinct from the phenotypes of Europe and The discovery of webworms for the first time attacking wild the United States, characterized by lower levels of imperatorin, parsnip populations in New Zealand (N.Z.) in 2004 (25, 26) bergapten, and isopimpinellin and slightly above-average levels provided an extraordinary opportunity to quantify selection by of xanthotoxin and bergapten. Occupying its own cluster is a coevolved herbivore on host-plant chemistry during the earliest population BB, low in all of the linear furanocoumarins and phase of reassociation. As in North America, parsnips were above-average in sphondin. One heavily attacked N.Z. popula- introduced to New Zealand as a food source by European tion, Lumber, clusters with two U.S. populations and one colonists (English and Scottish), escaped from cultivation, and European population long associated with webworms (21); this became naturalized, as noted by Hooker in 1867 (27). Conse- cluster is characterized by high levels of imperatorin, bergapten, quently, P. sativa has in New Zealand enjoyed at least 140 and xanthotoxin and low levels of isopimpinellin and sphondin. generations free of webworms. Based on characteristics of this Populations of N.Z. parsnips were no less phenotypically vari- interaction both in its native Europe and in its 150-year tenure able than their North American relatives; in fact, standard in North America, we predicted that N.Z. populations newly deviations of linear furanocoumarin and angular furanocouma- infested by webworms should show rapid changes in furanocou- rin concentrations ranged higher in New Zealand than in North marin chemistry after infestation, with their chemistry converg- America (8.13 vs. 5.11 ␮g/mg for linear furanocoumarins, 0.47 vs. ing on that of U.S. plants with long histories of webworm 0.36 ␮g/mg for angular furanocoumarins). association—i.e., plants in infested populations should be se- Because the number of traits used in the selection analyses was lected for increased amounts of furanocoumarins associated limited in certain populations by the number of plants available with resistance (xanthotoxin, bergapten, sphondin). In contrast, within a population (the number of independent variables in the N.Z. populations that have not been infested by webworms regressions must be less than the number of samples), compar- should be in equilibrium and should not exhibit differences in isons among populations and years are constrained. More in- chemical traits over the time period during which webworms formative are the within-population differences in selection have expanded their range elsewhere in New Zealand. coefficients that occur as result of substituting realized fitness (in the presence of webworms) for potential fitness (estimated to Results have occurred had webworms not been present) as the depen- Two webworm host plants were present within the survey area: dent variable. Changes in coefficients are attributable only to the P. sativa and hogweed, Heracleum sphondylium. In 2006, we effect of webworms. For all but the Warrington population in found 18 locations of hogweed populations ranging in size from 2006, which was minimally infested that year (average fitness was 1 to hundreds and 11 populations of wild parsnip ranging in size reduced by only 3.5%), webworms altered selection patterns from 17 to hundreds [supporting information (SI) Table 3]. (Table 1). Populations affording larger sample sizes permitted Three wild parsnip populations that were far from hogweeds and specification of greater numbers of traits and second-order

4548 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710280105 Zangerl et al. Downloaded by guest on September 27, 2021 acetate and bergapten. Patterns of nonlinear negative selection on isopimpinellin in the Ocean 2007 population and stabilizing SPECIAL FEATURE selection for isopimpinellin in the Townley 2006 population were diminished by webworms. In both of these populations, a variety of significant coefficients involving correlational selection and nonlinear selection were diminished by webworms (Table 1). In the case of the Townley 2006 population, negative nonlinear selection on octyl butyrate, known to act as a webworm feeding deterrent (28), was replaced by positive directional selection for the trait after fitness was adjusted by webworms. Finally, in the Townley 2007 population, again from a sample-size-limited model, selection by webworms against xanthotoxin was the only significant coefficient detected. Despite the limitations of regression, particularly when important variables are likely omit- ted, these results nevertheless suggest an extraordinary degree of selection for trait remixing. In some of these cases, webworms intensified (made significant) like-signed but nonsignificant coefficients detected by potential fitness. We suspect that, rather than amplifying the same selection pattern, our estimates of potential fitness may have been influenced by infestation; parsnips do attempt to compensate for losses to webworms by converting a larger percentage of flowers to seeds in higher- order umbels and also producing more higher-order umbels (29). To the extent that the response involves increased umbel stalk diameter and greater numbers of umbels, potential fitness, as measured, could have been influenced by infestation. Even so, the changes in significance and/or sign between fitness estimates for coefficients indicate that webworms are exerting significant selection on the chemistry of the populations. A high percentage (r2 ϭ 0.87) of the variation in pupal size of N.Z. webworms was accounted for by seven constituents of developing seeds (Table 2). Increases in the angular furanocoumarins sphondin and angelicin (the latter of which is the precursor of sphondin and typically not found in ripe seed)

Fig. 2. Clustering of 33 wild parsnip populations (involving 1,117 individuals) together with the linear furanocoumarin imperatorin and the EVOLUTION on the basis of average phenotypic furanocoumarin seed content; North octyl butyrate ester were associated with decreased pupal size, American and European data are from Berenbaum and Zangerl (21). Popula- whereas increases in the octyl acetate ester, isopimpinellin, and tions in red have had either long-term associations with webworms [in the xanthotoxin were associated with increased pupal size. Addition cases of North American, prefixed ‘‘na,’’ or European populations (green) of bergapten to the model (results not shown) diminished these from Austria, The Netherlands, or Germany, prefixed ‘‘aust,’’ ‘‘neth,’’ or relationships dramatically, likely due to high correlation with ‘‘germ’’] or only a recent association (between 2004 and 2006 in New Zealand, ‘‘nz’’). Populations in black have had no, or only rare, interaction with web- xanthotoxin in all of these populations, and substitution of worms. Clustering is based on squared Euclidean distances (SPSS 14, Chicago, bergapten for xanthotoxin failed to produce a significant model. IL). Bar graphs show percent deviations (ϩ or Ϫ) from mean furanocoumarin In addition to differences in furanocoumarin composition of content of each of six clusters for five furanocoumarins (left to right: impera- ripe seeds among N.Z. and U.S. populations (Fig. 2), all non- torin, bergapten, isopimpinellin, xanthotoxin, and sphondin). With one ex- furanocoumarin components of buds and male flowers, except ception (nzLUMB), N.Z. populations are clustered separately from all others. the ocimenes, differed substantially between countries (Fig. 3). Abbreviated N.Z. population names are LUMB, Lumber; TOWN, Townley; Midwest U.S. populations were Ͼ2-fold higher in octyl butyrate WAR, Warrington; COTT, Cottage; and OCEA, Ocean. and myristicin, higher in the sesquiterpenes caryophellene, ␣-trans-bergamotene, cis-␤-farnesene, and ␤-cubebene, and higher in 2-methylhexyl propanoate but lower in 1-octanol, octyl interactions. In these populations, complicated patterns of non- acetate, ␥-palmitolactone, and ␣-farnesene. linear or correlational selection were evident. In the Crimp 2006 population, selection for potential fitness favored a decline in Discussion isopimpinellin (directional selection), but selection incorporat- Wild parsnip populations in New Zealand newly infested by a ing the influence of webworms favored a reduction in sphondin coevolved specialist experienced major fitness reductions, pos- (directional selection) and positive or negative correlations sibly for the first time in the region, and experienced substantial among isopimpinellin, xanthotoxin, imperatorin, bergapten, and selection pressure for phytochemical trait remixing. It is too early octyl butyrate (Table 1). In 2007, in a sample-size-limited model to tell whether selection resulting from reassociation with the for the Crimp population, webworms intensified selection for an coevolved specialist D. pastinacella will favor convergence with uncoupling of xanthotoxin from sphondin. In the Lumber pop- North American and European parsnip populations long sub- ulation in both years, with sample-size-limited models, there jected to webworm herbivory. Some trends are consistent with were no significant selection patterns for potential fitness, but that direction, most clearly the selection for octyl butyrate, which with webworms there was selection against isopimpinellin (Table not only is low in N.Z. parsnip but also is associated with reduced 1). Webworms also caused intense nonlinear positive selection webworm pupal size. The trajectory of furanocoumarins is more for isopimpinellin in the Warrington 2007 population as well as difficult to assess, as there are three branches in the biosynthetic selection for positive coupling of octyl butyrate and sphondin. In pathway, all of which could be targeted by correlational selec- the Rock population, webworms exerted disruptive selection on tion. Potentially constraining genetic correlations among furano- sphondin as well as negative correlational selection for octyl coumarins and differences in the availability of additive genetic

Zangerl et al. PNAS ͉ March 25, 2008 ͉ vol. 105 ͉ no. 12 ͉ 4549 Downloaded by guest on September 27, 2021 Table 1. Selection in 2006 and 2007 on chemical traits of seeds of wild parsnip in N.Z. populations after reassociation with parsnip webworms Directional selection (␤) Nonlinear or correlational selection (␥) Population (Model) N % Potential ﬁtness/realized ﬁtness

Crimp 2006 (B, OB, IM, IS, S, X) 85 47.3 Sphondin ␤ ϭϩ0.02/؊0.25 Isopimpinellin ␤ ϭ ؊0.38/Ϫ0.19 Isopimpinellin/xanthotoxin ␥ ϭϪ2.20/ ؊4.46 Octyl butyrate/xanthotoxin ␥ ϭϩ2.60/؉5.28 Imperatorin/bergapten ␥ ϭϩ2.16/؉5.67 Imperatorin/xanthotoxin ␥ ϭϪ2.91/؊5.33 Octyl butyrate/bergapten ␥ ϭϪ2.42/؊4.98 Crimp 2007 (B, OB, S, X) 19 85.4 Xanthotoxin ␤ ϭ ؉0.80/ϩ0.82 Xanthotoxin/sphondin ␥ ϭϪ4.79/؊10.83 Lumber 2006 (OB, IM, IS) 21 34.9 Isopimpinellin ␤ ϭϪ0.14/ ؊0.58 Lumber 2007 (OA, IS, S) 12 18.4 Isopimpinellin ␤ ϭϩ0.18/ ؊0.94 (Ocean 2007 (IM, IS, B, S, X) 26 82.7 Isopimpinellin ␥ ϭ ؊6.18/Ϫ3.87 (negative Isopimpinellin/sphondin ␥ ϭ ؊8.0/Ϫ4.04 Isopimpinellin/bergapten ␥ ϭ ؉39.0/ϩ10.86 (Rock 2007 (B, IM, OA, OB, S) 74 85.2 Sphondin ␥ ϭϩ1.16/؉1.52 (disruptive Octyl butyrate/sphondin ␥ ϭ ؉1.81/؉2.057 Octyl acetate/bergapten ␥ ϭϪ0.11/؊2.01 Townley 2006 (B, IM, IS, OA, OB, S) 50 52.4 Octyl butyrate ␤ ϭϩ0.24/؉0.38 (Octyl butyrate ␥ ϭ ؊2.91/ϩ0.30 (negative (Isopimpinellin ␥ ϭ ؊3.152/Ϫ2.233 (stabilizing (Octyl acetate ␥ ϭ ؉2.272/ϩ1.461 (disruptive Isopimpinellin/sphondin ␥ ϭ ؉5.928/ϩ0.506 Octyl acetate/octyl butyrate ␥ ϭ ؊3.104/Ϫ1.514 Octyl acetate/sphondin ␥ ϭ ؊4.42/Ϫ2.998 Octyl acetate/isopimpinellin ␥ ϭ ؉3.4/ϩ2.21 Imperatorin/octyl butyrate ␥ ϭ ؉11.6/ϩ3.62 Imperatorin/isopimpinellin ␥ ϭ ؊11.2/Ϫ1.34 Bergapten/isopimpinellin ␥ ϭ ؉3.86/ϩ1.51 Townley 2007 (OB, IM, S, XB) 18 17.9 Xanthotoxin ␤ ϭϩ0.57/؊1.65 Warrington 2006 (B, IM, IS, OB, S, X) 29 96.5 Not signiﬁcant (Warrington 2007 (IS, OB, S, X) 19 38.4 Isopimpinellin ␥ ϭϩ5.88/؉26.25 (positive Octyl butyrate/sphondin ␥ ϭϩ9.55/؉20.369

Multivariate selection analysis according to Lande and Arnold (45) was performed separately for potential fitness (had webworms been absent) and for realized fitness (after consumption of reproductive parts by webworms). N gives the sample size for the multiple regression, and % refers to the average realized fitness as a percentage of potential fitness. Standardized coefficients in boldface were different from zero at P Ͻ 0.05. For nonlinear coefficients, the nature of the selection is shown in parentheses. In parentheses to the right of each population name are the traits included in the model: B, bergapten; I, isopimpinellin; S, sphondin; IS, isopimpinellin; IM, imperatorin; X, xanthotoxin; OA, octyl acetate; OB, octyl butyrate.

variance for selection to act on could limit the eventual out- populations will be necessary to determine the ultimate pheno- comes, and, indeed, different response trajectories are predicted type configuration resulting from webworm selection. by the geographic mosaic theory (2). Analysis comparisons of Rapid reacquisition of chemical defenses because of selection subsequent generations of plants in infested and noninfested pressure from a coevolved specialist may account for findings that appear inconsistent with ecological theories of invasion. The evolution of increased competitive ability (EICA) hypothesis, Table 2. Regression of mean pupal mass per stem for parsnip e.g., predicts that introduced plants, by leaving coevolved asso- webworms from 11 N.Z. plants on chemical composition of ciates behind, should have reduced investments in defense in the ؍ 2 wild parsnip (r 0.874) area of introduction (30); however, common garden experiments Model df Mean squares FPwith Senecio jacobaea, a weed native to Europe, failed to reveal a decline in pyrrolizidine alkaloid production in areas of invasion Regression 7 8.661 10.937 0.038 (31). Moreover, invasive populations possessed greater alkaloid Residual 3 0.792 content and greater resistance to Tyria jacobaeae, the cinnabar moth, a European specialist (31). Stastny and Schaffner (32) also Independents BtP found that plants in areas of invasion produced greater quantities Angelicin Ϫ4.723 4.19 0.025 of alkaloids than European populations and suggested that the Imperatorin Ϫ0.476 4.69 0.018 EICA hypothesis fails to account for their findings. The North Isopimpinellin 6.392 6.85 0.006 American populations in these studies, however, include local- Xanthotoxin 0.247 3.7 0.034 ities in Oregon and British Columbia, where T. jacobaeae has Sphondin Ϫ3.422 5.64 0.011 been established as a biocontrol agent for Ͼ30 years (33). Octyl acetate 4.333 4.31 0.023 Reassociation with a specialist may have selected for enhanced Octyl butyrate Ϫ2.05 4.27 0.024 chemical defense and thus account for the pattern found.

4550 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710280105 Zangerl et al. Downloaded by guest on September 27, 2021 ‘‘vigor’’ (38), may also result from increased production of volatile compounds mediating interactions with pollinating mu- SPECIAL FEATURE tualists. Such responses could account for seemingly paradoxical findings of enhanced production of certain ‘‘defense’’ chemicals in areas of invasion; these defense chemicals may serve other ecological roles. That herbivores and pollinators can both exert selection pressures on plant chemical traits is well established (39). Parsnip webworm larvae orient to octyl acetate to find suitable reproductive tissue to consume (28), but octyl acetate is among the volatile compounds that are attractive to the pro- miscuous pollinators of this plant and its relatives (40, 41). That N.Z. flowers produce higher amounts of octyl acetate than do U.S. plants is consistent with the suggestion that release from its specialist enemy may free the plant from constraints on production of volatile attractants and enhance its ability to attract pollinators (41). Reassociation with a major flower-feeding herbivore may alter pollination success of P. sativa; infestation by only a single webworm reduces pollinator visitation, likely because of changes in volatile cues used by pollinators (40). Whether, after reassociation with webworms, octyl acetate levels will continue to fall in N.Z. flowers and whether kairomones, like allomones, generally change in response to release from specialist herbivores remain to be seen. Methods Whereas P. sativa was first recorded in New Zealand in 1867, and the European import hogweed, H. sphondylium in 1939 (42), the webworm was reported for the first time in Port Chalmers on the South Island in 2004; an extensive survey at that time revealed that webworms were sufficiently numerous around Port Chalmers and nearby Dunedin as to be considered beyond eradication (26). In February 2006, populations of webworms and webworm host plants P. sativa and H. sphondylium were located in and around the presumed location of webworm introduction, Port Chalmers (25), and extending as far north as Christchurch, N.Z. All host populations were surveyed for the prevalence of webworms (SI Table 3). Although most webworms had pupated by the time

of the surveys, damage characteristic of webworm feeding, namely, presence EVOLUTION Fig. 3. Differences between N.Z. and Midwest U.S. parsnips in chemical of webbed umbellets together with frass and partially or wholly consumed Ϯ composition of buds and male flowers (means SE). All differences but those reproductive organs, was recorded for each plant surveyed. For large host- Ͻ for cis- and trans-ocimene between countries were significant (P 0.05) in a plant populations, a haphazardly selected subset of plants was surveyed. two-way ANOVA with country and plant part (bud or flower) as main effects. Relative fitness was quantified for plants within four infested wild parsnip There were no significant interactions and only three compounds differed populations in 2006 (Crimp, Lumber, Townley, and Warrington) and in the between buds and flowers: flowers were higher in octyl butyrate and lower in same four populations as well as two newly infested populations in 2007 (Rock myristicin and cis-ocimene (data not shown). and Ocean). Both realized fitness (after webworm activity) and potential fitness (had webworm not been present) were measured by estimating the proportion (in increments of 0.1) of umbellets in each umbel not destroyed by The potential for rapid phytochemical response of an invasive webworms. The diameter of each umbel stalk, as an approximate predictor of weed in an area of introduction to reassociation with a coevolved the number of reproductive units (units ϭ 91.6 ϫ umbel stalk diameter ϩ 98.3, specialist is instructive in two contexts. Not only does it provide n ϭ 10, r2 ϭ 0.80, P ϭ 0.0005) was also recorded to the nearest 0.1 mm with an example of the selective impact of an insect herbivore on plant Vernier calipers for each primary (stem apex), secondary (branch apex), and chemistry, a long-held assumption underlying theories of insect– tertiary (secondary branch apex) umbel on the plant. Total plant fitness after plant coevolution, it also demonstrates the potential for rapid webworm damage was then the sum of the products of the proportions of umbel undamaged times stalk diameter, or, for potential fitness had web- evolution of enhanced resistance in nonindigenous environ- worms not been present, simply the sum of stalk diameters. Because parsnips ments. If wild parsnip is typical of invasive weeds in possessing are temporally dioecious and not all flowers in secondary and higher-order chemical defenses that are genetically variable in areas of umbels transition to female flowers and fruits (each fruit potentially bearing introduction, then long-term control by introduced coevolved two seeds), application of the regression model to the sum of stalk diameters specialists may be an elusive goal. Moreover, theories relating to provides an estimate of the total number of male flowers. Each such flower in evolutionary changes in invasive plants are generally couched in the primary umbel typically turns into a female flower and eventually, if terms of reduced allocation to allomones—i.e., phytochemicals pollination is successful, a fruit (29). Transition of flowers on secondary and tertiary umbels to female flowers and fruits is conditional on plant resources with adverse effects on herbivores. Release from adapted spe- and webworm damage (29, 43). Relative fitness was calculated by dividing cialists, however, may have other impacts on plant chemistry. plant fitness by average fitness in the population. Specialist herbivores generally rely on kairomones, or host-plant Ripe seeds were collected from all wild parsnip populations in 2006 and recognition cues, to find feeding and oviposition sites. Many from most of them in 2007, including every plant for which fitness was phytochemicals that are defenses against generalist herbivores estimated. Four seeds from each plant were weighed to the nearest 0.1 mg, cut may serve as attractant kairomones for specialists (1, 34). In the in half with scissors, and placed in a 4-ml glass autosampler vial. The end of the absence of specialists, plants may experience selection to in- scissors was held inside the vial as 2 ml of ethyl acetate was added to ensure that residues on the scissors were also added to the vial. After2hatroom crease production of certain volatile compounds that attract not temperature, 200 ␮l of extract was placed in autosampler vials for analysis by only herbivores but also mutualists such as pollinators (35–37). gas chromatography, and the remainder was analyzed by high-pressure liquid Enhanced reproductive success of invasive species in areas of chromatography. Sample analysis by gas chromatography (GC) was per- invasion, although attributed to greater competitiveness or formed on a Hewlett–Packard 5890 instrument (Agilent) with a flame ioniza-

Zangerl et al. PNAS ͉ March 25, 2008 ͉ vol. 105 ͉ no. 12 ͉ 4551 Downloaded by guest on September 27, 2021 tion detector and an Alltech EC-1 capillary column (30 m, 0.32 mm inside sheet of paper, and photographed next to a Vernier caliper. Larvae almost diameter, 0.25-␮m coating). Two microliters of extract was injected into a always pupate within the stem of the plant on which they develop; however, 200°C inlet, and separation of compounds was accomplished with the follow- to reduce the likelihood of larval movement before pupation, pupae were ing program: 60°C for 0.5 min, 30°C/min, 170°C, 5 min. The predominant collected from plants at least 2 m from their closest neighbor. A sample of ripe non-furanocoumarin essential oil components in ripe seeds, octyl acetate and seeds was also collected from each plant. Cross-sectional area (mm2)ofthe octyl butyrate, were quantified. Furanocoumarins were autosampler-injected pupae was quantified by using Scion for Windows Imaging freeware. Mean (15 ␮l), separated, and quantified on a Waters HPLC. The separation was pupal areas were compared among plants by ANOVA and were regressed accomplished isocratically (1.5 ml/min flow, 55% cyclohexane, 42% diisopro- against furanocoumarin and octyl ester content of seeds to discern whether pyl ether, and 2% butanol) on a 250-mm long, 4.6mm inside diameter, and how pupal size is related to chemistry. Sperisorb 5-␮m silica column (Waters). The furanocoumarins imperatorin, Although the chemistry of ripe seeds of P. sativa can account for up to 70% bergapten, isopimpinellin, xanthotoxin, and sphondin were quantified. All of the variation in resistance to D. pastinacella (3), webworms consume buds, chemical concentrations were calculated in ␮g/mg of dry seed mass. Average flowers, and developing seeds over the course of their development. To furanocoumarin concentrations of N.Z. parsnip populations were compared investigate whether such reproductive structures of N.Z. wild parsnips differ with published mean values for U.S. and European populations (21) by em- from those of U.S. plants, we collected male flowers and buds from several N.Z. ploying hierarchical analysis (SPSS 14 2006) based on Euclidean distances and populations and compared them to flowers and buds from plants in U.S. average between-groups linkage (33 populations in all). populations. Eleven male flower and nine bud samples were collected from Estimates of directional, nonlinear, and correlational selection by web- four N.Z. populations (Ocean, Cottage, Townley, and Rock, one to six plants worms were obtained by multiple regression according to Lande and Arnold per population). Sixteen male flower and 12 bud samples were collected from five Midwest U.S. populations (two to five plants per population) in the vicinity (44). Limitations on sample size in some of the populations (all individuals, of Urbana, Illinois. Entire umbellets were collected, air-dried, weighed, and even those with as few as two or three ripe seeds, were analyzed in these placed in 2-ml centrifuge tubes. A glass bead was added to the tube and the populations) necessitated reducing the number of plant traits analyzed to sample was powdered in a shaker (modified Wig-L-Bug, Crescent Dental). accommodate analysis of nonlinear and correlational selection. Models were Ethyl acetate (0.5 ml) was added to each tube and after 1 h the particulates constructed that explained high amounts of the variation in relative realized were spun down in a centrifuge and the extract was analyzed by GC as fitness. A separate regression was performed for relative potential fitness described for octyl esters except for the oven program (50°C, 2 min, 5°C/min, employing the same traits that were specified in the model for realized fitness. 200°C, 3 min). Identity of peaks was confirmed by gas chromatography–mass The regressions were run again with the products of the dependent variables spectrometry. With the exceptions of 1-octanol and 2-methylhexyl propano- added to estimate nonlinear and correlated selection. The shape of the ate, all components quantified in this analysis were previously reported in P. trait–fitness relationship for traits under significant nonlinear selection (co- sativa (45). Where standards were available (octyl acetate and butyrate, efficients for squared terms in the regressions) were evaluated by plotting myristicin, ocimenes, caryophellene) component concentrations were ex- relative fitness values, predicted from the linear and quadratic terms by using pressed in ␮g/mg. For sesquiterpenes for which standards were not available, coefficients from the selection model, against the observed trait values from concentrations were calculated as caryophellene equivalents, and for all other the population. In this way, simple nonlinear selection for or against the trait, constituents the concentrations were calculated as hexadecane equivalents. disruptive selection, or stabilizing selection could be differentiated. To determine whether N.Z. P. sativa vary in suitability to D. pastinacella,we ACKNOWLEDGMENTS. We thank Andrew Suarez of the University of Illinois quantified the size of pupae inside stems of 11 substantially infested plants at Urbana–Champaign for comments on the manuscript. This work was sup- near the Warrington population. Each plant was cut at the base of its stem and ported by National Science Foundation Grant SGER 612376 (to M.R.B. and split lengthwise with a utility knife. Pupae were removed, placed on a white A.R.Z.).

1. Ehrlich P, Raven P (1964) Evolution (Lawrence, Kans) 18:586–608. 24. Zangerl AR, Berenbaum MR (2005) Proc Natl Acad Sci 102:15529–15532. 2. Thompson JN (2005) The Geographic Mosaic of Coevolution (Univ of Chicago Press, 25. Patrick B (2004) Weta 27:8–12. Chicago). 26. Wilson JA, Stephenson BP, Gill GSC, Randall JL, Vieglais CMC (2004) N Z Plant Protect 3. Berenbaum MR, Zangerl AR, Nitao JK (1986) Evolution (Lawrence, Kans) 40:1215–1228. 57:156–160. 4. Mauricio R, Rausher MD (1997) Evolution (Lawrence, Kans) 51:1435–1444. 27. Hooker JD (1867) Handbook of the New Zealand Flora (Reeve, London), p 759. 5. Shonle I, Bergelson J (2000) Evolution (Lawrence, Kans) 54:778–788. 28. Carroll MJ, Berenbaum MR (2001) J Chem Ecol 28:1365–1375. 6. Jermy T, Szentesi R (2003) Oikos 101:196–204. 29. Hendrix SD, Trapp EJ (1992) Am J Bot 79:365–375. 7. Daehler CC, Strong DR (1997) Oecologia 110:99–108. 30. Blossey B, No¨tzold R (1995) J Ecol 83:887–889. 8. Willis AJ, Thomas MB, Lawton JH (1999) Oecologia 120:632–640. 31. Joshi J, Vrieling K (2005) Ecol Lett 8:704–714. 9. Maron JL, Vila M, Arnason J (2004) Ecology 85:3243–3253. 32. Stastny M, Schaffner U, Elle E (2005) J Ecol 93:27–37. 10. Lewis KC, Bazzaz FA, Liao Q, Orians CM (2006) Oecologia 148:384–395. 33. Harris P, Wilkinson ATS, Neary ME, Thompson LS (1971) in Biological Control Pro- 11. Siemann E, Rogers WE (2003) Oecologia 135:451–457. grammes Against Insects and Weeds in Canada 1959–1968 (Commonwealth Institute 12. Maron JL, Vila M, Bommarco R, Elmendorf S, Beardsley P (2004) Ecol Monogr 74:261– of Biological Control, Commonwealth Agricultural Bureaux, Slough, UK), Technical 280. Communication 4, pp 97–104. 13. Zangerl AR, Berenbaum MR (2003) Evolution (Lawrence, Kans) 57:806–815. 34. Feeny P (1992) in Herbivores, Their Interactions with Secondary Plant Metabolites, eds 14. Sturtevant EL (1890) Am Nat 24:30–49. Rosenthal GA, Berenbaum MR (Academic, New York), Vol 2, pp 1–35. 15. Berenbaum MR, Zangerl AR (1996) in Phytochemical Diversity and Redundancy in 35. Dudareva N, Pichersky E (2000) Plant Physiol 122:627–633. Ecological Interactions, eds Romero JT, Saunders JA, Barbosa P (Elsevier, New York), pp 36. Knudsen JT, Tollsten L, Bergstrom LG (1993) Phytochemistry 33:253–280. 1–24. 37. Pichersky E, Gershenzon J (2002) Curr Opin Plant Biol 5:237–243. 16. Berenbaum M (1981) Ecology 62:1254–1266. 38. Hinz HL, Schwarzlaender M (2004) Weed Technol 18:1533–1541. 17. Bethune CJS (1869) Can Entomol 2:1–4. 39. Adler LS, Karban R, Strauss SY (2001) Ecology 82:2032–2044. 18. Hodges R (1974) Gelechioidea: Oecophoridae (in Part). Moths of America North of 40. Lohman DJ, Zangerl AR, Berenbaum MR (1996) Am Midl Nat 136:407–412. Mexico (Classey and RBD Publications, London), Fasc. 6.2. 41. Ju¨rgens A, Do¨tterl S, Meve U (2006) New Phytol 172:452–468. 19. Berenbaum MR (1990) Annu Rev Entomol 35:319–343. 42. Webb CJ (1978) NZJBot16:387–390. 20. Zangerl AR, Berenbaum MR (1993) Ecology 74:47–54. 43. Nitao JK, Zangerl AR (1987) Ecology 68:521–529. 21. Berenbaum MR, Zangerl AR (2006) Ecology 87:3070–3081. 44. Lande R, Arnold SJ (1983) Evolution (Lawrence, Kans) 37:1210–1226. 22. Zangerl AR, Berenbaum MR (1997) Am Nat 150:491–504. 45. Zangerl AR, Green ES, Lampman RL, Berenbaum MR (1997) Phytochemistry 44:825– 23. Berenbaum MR, Zangerl AR (1992) Evolution (Lawrence, Kans) 46:1373–1384. 831.

4552 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0710280105 Zangerl et al. Downloaded by guest on September 27, 2021