Aphid fecundity and grassland invasion: Invader life ... - ESA Journals

Ecological Applications, 19(5), 2009, pp. 1187–1196 Ó 2009 by the Ecological Society of America

Aphid fecundity and grassland invasion: Invader life history is the key ELIZABETH T. BORER,1 VINCENT T. ADAMS, GARETH A. ENGLER, AUTUMN L. ADAMS, CANAN B. SCHUMANN, AND ERIC W. SEABLOOM Department of Zoology, Oregon State University, Corvallis, Oregon 97331 USA

Abstract. Loss or gain of pathogens can determine the trajectory of biological invasions, and invasion by novel hosts also can alter pathogen dynamics to facilitate invasion. Recent empirical and theoretical work has implicated infection by barley and cereal yellow dwarf viruses (B/CYDV), a group of generalist pathogens of the Poaceae family (grasses), as a necessary precursor to the invasion of over 9 million hectares of California’s perennial grasslands by exotic annual grasses. The mechanism underlying this pathogen-mediated invasion hypothesis is elevated vector fecundity on exotic annual grasses. While empirical evidence supports this hypothesis, the links between aphid fecundity, host identity, and host resource supply have not been thoroughly assessed. We performed field and laboratory experiments to examine the fecundity and preference responses of three of the most common aphid vectors of B/CYDV, Rhopalosiphum padi (L.), R. maidis (Fitch), and Sitobion avenae (Fab.), to a combination of host life history (annual and perennial), host provenance (native and exotic), and nutrient supply (mineral N and P fertilization), controlling for host phylogenetic lineage. Aphids consistently had higher fecundity on annual grasses than perennials, regardless of host provenance, age, or nutrient fertilization. In addition, aphids preferentially colonized annual hosts when offered a choice among host species. Multigeneration studies have found that nutrient addition affects both host quality and composition in natural communities; our experimental results indicate that the indirect effects of nutrient fertilization in determining host community composition are of more importance than are the direct effects on host quality to aphid population dynamics. To summarize the applications of our results, we demonstrate that, in contrast to the current focus on the qualitative differences between invaders and natives, the impact of invasive exotic grasses is not due to host provenance, per se, but arises because the annual invaders differ qualitatively from the native species in interactions with shared pathogen vectors. More generally, our work demonstrates the importance of isolating whether the fate and impacts of an invader are, at their root, due to the provenance of the invader, or due to other characteristics that determine its functional uniqueness in the context of the native community. Key words: apparent competition; barley and cereal yellow dwarf viruses (BYDV); California grassland; disease ecology; exotic and native grasses; Poaceae; vector.

INTRODUCTION Biological invasions are one of the leading threats to imperiled species; invasion of exotic weeds and plant pests is estimated to cost the United States $80 billion every year (Wilcove et al. 1998, Pimentel et al. 2000). While gains or losses of pathogens have long been recognized as potential mediators of biological invasions (Park 1948, Dobson and Hudson 1986, Hudson and Greenman 1998, Mitchell and Power 2003, Torchin et al. 2003), it has only recently been demonstrated that the invasion process itself can alter pathogen dynamics (Tompkins et al. 2003, Borer et al. 2007). In one of the most extreme cases, recent work shows that the presence of a suite of aphid-vectored, RNA viral grass pathogens, Manuscript received 25 June 2008; revised 24 September 2008; accepted 6 October 2008; final version received 10 November 2008. Corresponding Editor: S. K. Collinge. 1 E-mail: [email protected]

collectively referred to as barley and cereal yellow dwarf viruses (B/CYDV), was a necessary precondition for one of the most spatially extensive and persistent biological invasions worldwide (Borer et al. 2007), in which 25% of California [USA] became dominated by exotic annual grasses from the European Mediterranean region (Heady 1977, Mooney et al. 1986). Theory suggests that the mechanism underlying the pathogen-mediated invasion of California’s grasslands is apparent competition, in which the aphid vectors for this viral group have higher fecundity on exotic annual grasses compared to the natives, thereby leading to increased pathogen transmission rates throughout the grass community in the presence of the invaders (Borer et al. 2007). Empirical work in this system has demonstrated that with increasing relative abundance of exotic annual grasses, B/CYDV infection rates increase in hosts (Malmstrom et al. 2005, Borer et al. 2009), and observed patterns of B/CYDV infection at broad spatial and temporal scales clearly point to the critical importance of

1187

1188

Ecological Applications Vol. 19, No. 5

ELIZABETH T. BORER ET AL.

vectors and host community in controlling infection prevalence in natural communities (Seabloom et al. 2009). Finally, the link between vector fecundity and disease transmission is quite strong: higher aphid densities consistently lead to higher likelihood of B/CYDV transmission (Jensen 1969, Jensen and D’Arcy 1995), elevated viral spread rates (Kendall et al. 1992, Power and Gray 1995), and increased numbers of infected plants (Burnett and Gill 1976). While this body of evidence supports apparent competition via this shared pathogen group as a critical precursor to the vast invasion of California’s grasslands, the existing empirical links between aphid fecundity, host identity, and host resource supply have not yet been thoroughly assessed. Previous work suggests that both host life history and the abiotic environment affect aphid fecundity: foraging alate (winged) aphids tend to choose annual grasses over perennials (Rautapaa 1970, Malmstrom et al. 2005), and nitrogen fertilizer is known to enhance aphid fecundity in crops (Vickerman and Wratten 1979). However, interactions in this system have only been assessed on exotic-annual and nativeperennial hosts (Malmstrom et al. 2005), because exoticperennial and native-annual grasses are uncommon in the interior grasslands where this earlier work was conducted. In contrast, exotic perennial grasses dominate grasslands in coastal and northern areas of the Pacific States (e.g., Oregon, Washington, and coastal California). In addition, while this earlier study provides some support of the apparent competition hypothesis, the fecundity of only one of several abundant West Coast vectors was assayed on a small subset of hosts at ambient soil fertility. Thus, the relative importance for aphid fecundity of host life history (annual vs. perennial), host provenance (native vs. exotic), host nutrition (N or P fertilization), and host phylogenetic lineage (genera or tribe) remain unclear. Although each of these factors could be influential for vector fecundity, they have differing implications for conservation. For example, if vector preference is primarily driven by life history (e.g., preference for perennials), we would expect exotic and native perennials to experience similar infection levels; aphid fecundity and B/CYDV infection would not be the primary driver of invasion by exotic perennial grasses into an existing community of native perennials. In contrast, if vector preference is driven by host provenance (e.g., preference for exotics), then we would expect greater vector fecundity, infection, and invasion in the presence of either annual or perennial invaders. Resolving the interaction between vector performance, host community and composition, and host resources has relevance far beyond B/CYDV and the California grassland invasion case study discussed here. Vectored and generalist pathogens cause some of the most common and widespread infections of humans, and many important emerging diseases of both animals and plants are vector-borne, generalist pathogens (Taylor et al. 2001, Power and Flecker 2003, Anderson

et al. 2004). The spread rate of vectored pathogens can be quite sensitive to changes in vector abundance, birth rates, and behavior (Ross 1911, MacDonald 1957, Porter et al. 1988, Van Buskirk and Ostfeld 1995, Holt et al. 1997, Ostfeld et al. 2006, Bigoga et al. 2007), and changes in the abiotic environment, such as temperature, precipitation, or nutrient availability, can cause changes in vector fecundity that alter host infection risk (Pope et al. 2005, Minakawa et al. 2006). In addition, we have a relatively underdeveloped understanding of disease in multihost communities compared to the vast literature on single host–pathogen systems (Anderson and May 1986, Begon et al. 1992, Norman et al. 1994, Woolhouse et al. 2001, Holt et al. 2003, Dobson 2004). Thus, in spite of the recognized global importance of this class of pathogens, the links between vector fecundity, the abiotic environment, and host preference are only beginning to be explored in natural systems (Collinge and Ray 2006). Here we systematically investigated the interactions between host provenance, phylogeny, and resource supply on the preference and performance of a suite of aphid vectors of barley and cereal yellow dwarf viruses. To do this, we performed field and greenhouse experiments to quantify the relative influence on aphid fecundity and host preference of a globally common and locally important aphid species, Rhopalosiphum padi (L.), to a factorial combination of life history, provenance, and nutrient status of host species. Because invasion by apparent competition requires concordant fecundity responses among B/CYDV vectors in these grasslands, we then examined the generality of this vector species’ preference and fecundity responses by repeating this experiment using three of the most common B/CYDV vectors on the U.S. West Coast, R. padi, R. maidis (Fitch), and Sitobion avenae (Fab.). METHODS Study system The barley and cereal yellow dwarf viruses (B/CYDV) are a group of globally distributed viral pathogens that infect the phloem-carrying structures of cereals and other grasses, leading to chlorosis and stunting (D’Arcy 1995). Although host species have variable susceptibility and viremia, B/CYDV generally acts to reduce fecundity and increase mortality of infected hosts (D’Arcy 1995, Malmstrom et al. 2005). This pathogen group can cause devastating effects on cereal crop production (D’Arcy 1995), and has been implicated as a causal factor in the invasion of nonagricultural grasslands (Malmstrom et al. 2005, Borer et al. 2007). Thus, this pathogen group and its vectors have major global importance for their effects on crops as well as in grassland conservation. The B/CYDV group is obligately transmitted by a suite of aphid vectors; these viruses cannot be transmitted directly among hosts or vertically to seeds. Twentyfive aphid species are known vectors for this viral group (Halbert and Voegtlin 1995), including our focal species,

July 2009

APHID FECUNDITY AND GRASSLAND INVASION

1189

TABLE 1. Experimental host species were chosen to create replicated host groups by life history (annual [A] or perennial [P]), phylogenetic lineage (group), and provenance (native [N] or exotic [E] to our study region). Species and authority

Common name

Life history

Provenance

Group

Bromus hordeaceus L. Bromus carinatus Hook & Arn. Avena fatua L. Koeleria cristata (L.) Pers. Arrhenatherum elatius (L.) J. Presl & C. Presl Taeniatherum caput-medusae (L.) Nevski Elymus glaucus Buckley Cynosurus echinatus L. Festuca roemeri (Pavlick) E. B. Alexeev Festuca arundinacea Schreb.

soft brome, soft chess California brome wild oat prairie junegrass tall oat medusa-head rye blue wild-rye bristly dogs-tail grass Romer’s fescue tall fescue

A P A P P A P A P P

E N E N E E N E N E

brome brome oat oat oat rye rye fescue fescue fescue

Note: Subsets of these species were used in experiments I–IV. Species used only in greenhouse experiments.

R. padi, R. maidis, and S. avenae, which occur commonly throughout West Coast grasslands. All three of these aphid species feed on a variety of crop and noncrop grasses, and are common, globally distributed, and competent vectors of B/CYDV (Halbert and Voegtlin 1995). Host phylogeny, life history, and provenance Experimental host species were chosen to create replicated host groups by two major host species traits: life history (annual vs. perennial) and provenance (native and exotic to our study region). Because aphid responses to hosts may be confounded by evolutionary constraints of different host groups (Agrawal et al. 2005), we accounted for phylogenetic constraints by employing species groups that nested our focal host traits (annual/perennial, native/exotic) within four groups with shared phylogenetic lineage (Strauss et al. 2006). By employing this design, we can confidently attribute observed differences to host traits as opposed to idiosyncratic differences among species, and we can quantitatively compare the relative importance of each factor to aphid fecundity. Using grass species that commonly occur in the U.S. West Coast flora, we identified four phylogenetic groups, bromes, oats, ryes, and fescues, with representative exotic annual, exotic perennial, and native perennial members (Table 1). We were unable to include native annual grasses because they are generally rare in the Pacific states flora (Seabloom et al. 2006). Short-term fecundity experiments All fecundity experiments adhered to the same protocol, following Malmstrom et al (2005). Sleeve cages (8 3 2 cm) of 118-lm polyester mesh (Sefar America Incorporated, Kansas City, Missouri, USA) were affixed to individual grass blades and received a single mature apterous (wingless) aphid, which was sealed into the bag. After four days, leaves with bags attached were clipped from the plants and returned to the laboratory, where survival of the original aphid was assessed, adult and young aphids were counted, and the area of the leaf enclosed in the bag was recorded. All

aphids used in experiments were raised on Avena sativa in Percival Intellus environmental chambers at 228C, 75% humidity, and 24 h of light. R. padi used in the field and laboratory fecundity and preference experiments (I– III) were from a mixed culture of R. padi individuals collected locally from corn plants found near Corvallis, Oregon; the three-species experiment (IV) used aphids from single clone cultures maintained at Cornell University (A.G. Power). I. Field fecundity: R. padi.—First, we sought to examine the importance of life history and provenance on aphid fecundity in mixed-age stands of naturally occurring grasses. We performed the field fecundity experiment in June 2006 at Baskett Slough National Wildlife Reserve in Oregon, USA (44858 0 N, 123815 0 W), in open oak savannah. The reserve is surrounded by grass seed agriculture (e.g., Dactylis glomerata, Festuca arundinacea, and Lolium perenne). The region receives 1000–1500 mm of precipitation annually, and the temperature ranges from a summer mean of ;278C to a winter mean of ;78C. To examine the relative effects of provenance and life history under field conditions, we found a location with naturally occurring populations of 8 of our 10 focal host species in close proximity on a single hillside at the reserve (Table 1). We deployed 80 sleeve cages (10 cages/species; for methods, refer to Short-term fecundity experiments) between 11 and 14 June 2006, and repeated this for a second temporal block from 20 to 23 June 2006. Individuals receiving sleeve cages were selected haphazardly for both temporal blocks. A total of 150 sleeve cages were deployed: 20 cages per host species, except for Taenatherium caput-medusae, which had senesced by the time we performed the second temporal block. II. Greenhouse fertility–fecundity: R. padi.—To examine the relative effects of nitrogen and phosphorus fertilization, provenance, and life history on aphid fecundity, we performed a controlled greenhouse fecundity experiment. In this experiment, we employed a total of 10 host species (Table 1). In contrast to the field fecundity experiment (I), we started all grass hosts from seed, so aphids were exposed to hosts of identical age. Seed for the original eight species was collected from

1190



naturally occurring populations at Baskett Slough Reserve in June 2006; seed for the two additional native species was purchased from local commercial seed growers (Koeleria cristata, Landmark Seed Company, Albany, Oregon, USA; Festuca romeri, Pacific Northwest Natives, Albany, Oregon, USA). We applied a factorial combination of nitrogen (0.16 g of N as Ca(NO3)2) and phosphorus (0.16 g of P as Ca(H2PO4 )2) to the 10 host species (5 replicates 3 4 NP treatments 3 10 species) for a total of 200 experimental units. These fertilization rates translate to ;10 g/m2 of mineral fertilizer, rates that have been shown to cause significant changes in plant productivity and diversity in West Coast grasslands (Huenneke et al. 1990, Seabloom et al. 2005, Harpole and Tilman 2007). All 200 black polyethylene planter cones (656 cm3) were filled with planting medium (Sunshine Grower’s Mix A, Sungro Incorporated, Vancouver, Canada). A single pregerminated start was planted in each cone, and placed at randomized locations in racks of 20 cones each. Sleeve cages and aphids were applied to all experimental units when plants were six weeks old, on 31 July 2006, removed after four days, and counted. After the aphids were counted, the aboveground growth was removed from each pot at the level of the growing medium, dried for 24 hours, and weighed. Dry tissue was ground and samples with .0.5 g dry tissue (n ¼ 119) were analyzed for %C and %N (Central Analytical Laboratory, Oregon State University, Corvallis, Oregon, USA). III. Preferential herbivory: R. padi.—To determine whether aphids had a consistent preference for hosts by life history or provenance, germinants of the 10 focal grass species (Table 1) were planted in a randomized radial pattern in 17 3.78-L black polyethylene pots and raised for five weeks in an environmental chamber at 21.18C. Just prior to application of aphids, the grasses were thinned to a single individual of each species per pot. At the start of each trial, 30 aphids were placed in the center of each pot, equidistant from all grass hosts. The pots and plants were immediately enclosed in hoods constructed of 118-lm polyester mesh and returned to the indoor environmental chamber. After 24 hours, we removed the hoods and counted the number of adult aphids on each individual plant. We then removed the aboveground growth of each plant at the level of the growing medium, recorded its identity, and weighed its dry mass. IV. Greenhouse multivector fecundity: R. maidis, R. padi, S. avenae.—To confirm that the fecundity responses by R. padi were general to a variety of vector species, we performed an experiment similar to the greenhouse fertility–fecundity experiment (II). Using our standard protocol, we exposed three different species of aphids to six phylogenetically matched annual and perennial host species (a subset of Table 1: A. fatua, K. cristata, B. hordeaceus, B. carinatus, T. caputmedusae, E. glaucus) and examined the effects of nitrogen fertilization (none and 1 g of Ca(NO3)2). We

replicated this across five blocks for a total of 180 experimental units. All 180 656-cm3 black polyethylene planter cones were filled with planting medium (Sunshine Grower’s Mix A) and nitrogen was added to the assigned pots. Pregerminated starts were planted in the cones, and placed at randomized locations in racks of 18 cones each. Sleeve cages and aphids were applied to all experimental units when plants were six weeks old, on 9 August 2007, removed after seven days, and juvenile aphids were counted. Statistical analyses All data were analyzed using R (v. 2.0.0, R Foundation for Statistical Computing, Vienna, Austria). We developed Poisson regression models using orthogonal contrasts to quantify the relative influences on aphid fecundity of host life history, provenance, and fertilization, after accounting for host phylogenic lineage (see Appendix). Only bags containing either nymphs or a live adult aphid at the end of the experiment were included in analyses of aphid fecundity. Tissue chemistry was available for only a subset of the greenhouse fertility–fecundity experiment (II), but results of models including (n ¼ 119) or excluding (n ¼ 189) the tissue chemistry data were qualitatively similar (in terms of both parameter estimates and significant terms). Here, we present results from the model including host tissue chemistry. We exclude from the tissue chemistry analysis one anomalously high %C outlier value (B. carinatus: 61.21%). Plant mass and tissue chemistry were analyzed using linear regression. We assessed the preference of aphids for hosts of differing life history, provenance, and phylogeny (estimated using aphid density on each plant after 24 hours, or number of aphids per aboveground host mass) using linear regression. The block term was never significant, so it was dropped from all final models. All final regression models are presented in the online Appendix. RESULTS I. Field fecundity: R. padi.—In field assays, R. padi aphids produced nearly 2.5 times more progeny on annual grasses than perennials (P , 0.001; Fig. 1a, b; Appendix: Table A1). Provenance also affected vector fecundity: aphids had twice the fecundity on native perennials compared to exotic perennials (P ¼ 0.003; Fig. 1b). These patterns in fecundity of R. padi were consistent among phylogenetic groups (P . 0.05 for all group comparisons). II. Greenhouse fertility–fecundity: R. padi.—In greenhouse assays, with even-aged hosts, aphid fecundity mirrored the field results: fecundity was lower on both native and exotic perennials than on phylogenetically matched annual grass hosts (P ¼ 0.001; Fig. 1c, d; Appendix: Table A2). Phylogeny influenced aphid fecundity: aphids had lower fecundity on host species in the rye group compared to the other three (P ¼ 0.01). A comparison between exotic and native perennial hosts

July 2009


1191

FIG. 1. Aphid short-term fecundity on annuals and perennials as a function of host age, phylogeny, and provenance. Consistent patterns of Rhopalosiphum padi fecundity are seen in both (a, b) naturally occurring host populations in the field, and (c, d) greenhouse trials using identical-aged host plants. Parts (a) and (c) show means by each species within phylogenetic groups; parts (b) and (d) summarize means by host life history and provenance. In all plots, error bars are 2 SE.

showed that aphid fecundity was independent of host provenance (P ¼ 0.16; Fig. 1d). Fertilization by N and P affected plant mass, tissue chemistry, and aphid fecundity. Overall, annual grass aboveground biomass did not differ from perennials (P ¼ 0.32; Appendix: Table A3), but the brome group had greater aboveground mass than any of the other three phylogenetic groups (P , 0.001). Phosphorus fertilization did not alter aboveground plant biomass (P ¼ 0.40). In contrast, nitrogen fertilization increased the mass of all plant types (native perennials, exotic perennials, and annuals; P , 0.001). Tissue chemistry of plants in this experiment (%C and %N in aboveground biomass) was primarily affected by nitrogen fertilization (Fig. 2; Appendix: Table A4). Perennial grasses had a greater %C in their tissues than annuals, even after just six weeks of growth (P , 0.001); tissue %C was not affected by fertilization with either N (P ¼ 0.98) or P (P ¼ 0.54). In contrast, adding N fertilizer tripled tissue %N in all 10 species, regardless of life history or phylogeny (means ¼ 1.13% with no N, 3.39% with N, P , 0.001). Although %N in the tissues

of unfertilized annuals and perennials did not differ (unfertilized perennials ¼ 1.1%, annuals ¼ 1.2%), in contrast to the patterns seen in %C, life history influenced the response to nitrogen fertilization; adding N to annual grasses caused disproportionately large increases in tissue %N compared to perennials (P , 0.001). Fertilization with P did not affect tissue %N (P ¼ 0.11). Phylogenetic group had a minor influence on tissue chemistry; tissue carbon was similar among groups, but species in the oat group had somewhat more nitrogen (P ¼ 0.005) in their tissues than bromes. This trend toward higher nitrogen in oat tissue was similar, though not significant, across the other groups. Finally, fertilization with nitrogen did not affect overall aphid fecundity (P ¼ 0.12); however, aphid fecundity increased substantially with tissue %N (slope in log space ¼ 0.277, P ¼ 0.002). After accounting for host life history, aphid fecundity slightly increased with %C in the perennials (slope in log space ¼ 0.094, P ¼ 0.003); however %C and life history are strongly collinear (slope ¼ 0.62, P , 0.001), so across all species, aphid fecundity declined with increasing %C.

1192



(estimated from control plants in experiment II) was a strong predictor of R. padi preference among hosts (correlation ¼ 0.81, P ¼ 0.004; Fig. 4b); however, vector preference among hosts was not associated with %N in unfertilized hosts (P ¼ 0.41; Fig. 4a). IV. Greenhouse multivector fecundity: R. maidis, R. padi, S. avenae.—Major aphid vector species, R. padi, R. maidis, and S. avenae, responded similarly to host life history: fecundity of all three vectors was consistently and substantially higher on annuals than perennials (P , 0.001, Fig. 5). Within host phylogenetic groups, vectors had broadly similar patterns of fecundity, although fecundity at the host group scale tended to be 50% lower on oats than on brome or rye species (P , 0.05) because of extremely low vector fecundity on Koeleria for all aphid species. Nitrogen fertilization did not affect vector fecundity (P ¼ 0.72; Appendix: Table A6). Because of substantial overdispersion in the data, the model results from this experiment are based on a quasi-Poisson, rather than Poisson, model (McCullagh and Nelder 1989). DISCUSSION

FIG. 2. Total tissue (a) %C and (b) %N of 10 grass hosts as a function of host life history and provenance. Plants are the same as those shown in Fig. 1c, d. Fertilization with nitrogen did not affect %C (not shown) but significantly increased %N in all species (part b). Error bars are 2 SE.

Adding phosphorus caused a moderate increase in aphid fecundity (11% increase, P , 0.001), but this was driven primarily by a threefold increase in aphid fecundity on Arrhenatherum elatius in response to addition of P. Significant higher order effects of nutrients (e.g., N 3 P 3 provenance) were largely driven by the response of this same species (A. elatius) to P fertilization (Appendix: Table A2). III. Preferential herbivory: R. padi.—In 24-hour preference assays, aphids preferred annuals over perennials (P ¼ 0.01; Fig. 3; Appendix: Table A5); in each phylogenetically matched group, exotic annuals had the same or higher density of colonists than did either of the perennial species. Thus, individuals of R. padi were willing to tolerate a higher density of conspecifics on annual grasses than on perennials. In contrast, provenance did not affect preference: aphids tolerated similar densities on native and exotic perennial grasses (P ¼ 0.91). Finally, the mean tissue %C of each species

B/CYD viral spread is sensitive to changes in aphid vector abundance and birth rates. Aphid density is positively associated with aphid emigration (Parry 1977, Donaldson et al. 2007). In addition, severity of infection in individual hosts (Jensen and D’Arcy 1995), viral spread rates (Kendall et al. 1992, Power and Gray 1995), and number of infected plants (Burnett and Gill 1976) all increase with aphid density. Similar patterns are observed in other host–pathogen systems (MacDonald 1957, Porter et al. 1988, Van Buskirk and Ostfeld 1995, Holt et al. 1997, Ostfeld et al. 2006). Thus, factors that significantly increase vector abundance are likely to have substantial impacts on viral prevalence in host commu-

FIG. 3. Aphid preference among grasses as a function of life history and provenance. Preference is measured as adult aphid density after 24 hours on each of 10 grass hosts; data shown summarize means by host life history and provenance. Error bars are 2 SE; n ¼ 17.

July 2009


FIG. 4. Correlation between mean density of Rhopalosiphum padi aphids after 24 hours (experiment III) and (a) mean %N and (b) mean %C in tissues (nutrients in tissues of unfertilized control plants from experiment II). Letters within the plot represent the genus–species epithet for each of the 10 plant species examined (see Table 1). The line in part (b) shows the significant correlation between %N and colonist density.

nities. Our results demonstrate that three of the most abundant West Coast cereal aphid species experience strong, concordant fecundity responses: all had higher growth rates on annual hosts compared to perennials. This result was consistent regardless of host age, aphid clone, and laboratory or field conditions, and host provenance had minimal effects on fecundity. In addition, aphids tended to preferentially colonize hosts that maximized their population growth rates (annuals). Fertilization of hosts, particularly by nitrogen, directly altered host tissue chemistry, but also indirectly altered aphid fecundity. Thus, the invasion of California by a suite of grasses that enhance the reproductive rates of important aphid vector species may have qualitatively changed the dynamics of the B/CYD virus group. Through a positive feedback, B/CYDV infection may facilitate viral transmission by increasing vector fecun-

1193

dity on, and subsequent movement from, infected hosts, thereby increasing levels of infection throughout this grassland community (Borer et al. 2007). Overall, aphids prefer and perform better on annual plants compared to perennials, a result that is concordant with previous experimental assays in which aphid populations were greatest after two weeks on fastgrowing annual grasses, followed by fast-growing perennials, and the fewest aphids were found on slowgrowing perennial grasses (Fraser and Grime 1999). Our results are also concordant with a previous study that demonstrated higher aphid fecundity on a suite of California annual grasses compared with two perennials (Malmstrom et al. 2005). However this prior study could not uncouple the effects of host life history and provenance (species were either exotic annual or native perennial) and did not correct for phylogeny; the current results clarify the importance of host life history, per se. In fact, grass life history and growth rate are strongly associated with susceptibility to attack by herbivores (Rosenthal and Dirzo 1997). This phenomenon, in the current context, suggests that invasive annual grasses in West Coast grasslands play a far more important role in increasing B/CYDV transmission than do the invasive perennial grasses. Grass life history covaries strongly with tissue %C, and both factors provide substantial predictive power for aphid preference and fecundity. Leaf toughness, positively associated with %C in grasses (Perez-Harguindeguy et al. 2000, Barbehenn et al. 2004), can reduce aphid feeding (Dixon 1985). Alternately, the ratio of C:N may constrain aphid nutrition and growth rates (Elser et al. 2000). Whether aphid fecundity is controlled by the C/N ratio, tissue %C, or by a factor for which %C is a surrogate (Powell et al. 2006), the widespread invasion of the West Coast’s formerly perennial

FIG. 5. Short-term fecundity of three species of aphid (Rhopalosiphum padi, R. maidis, and Sitobion avenae) commonly found on the U.S. West Coast on annual and perennial grass hosts. Error bars are 2 SE; n ¼ 15.

1194


grasslands by invasive annual grasses, particularly in the southern half of this vast grassland ecosystem, appears to have provided a context for substantially increased aphid reproduction. Although aphids had much greater preference for and fecundity on annuals compared to perennial grasses, it is interesting to note that tissue %N did not differ substantially among any of the unfertilized grasses, regardless of life history. Whole-leaf chemistry (i.e., %C or %N) can be a reasonable surrogate for phloem chemistry (Awmack and Leather 2002); however, this relationship does not always hold (Karley et al. 2003, 2008), and many plants can partition nitrogen into both nutritional and toxic compounds (Mattson 1980). The basis for aphid preference, acceptance, and performance on plants is an area of active research (Powell et al. 2006), and these results suggest that further examination of the relationship between phloem chemistry and the chemistry of whole leaves in grasses may clarify whether whole-tissue chemistry, particularly %N, provides adequate information about aphid nutrition in grasslands. Fertilization by N and P had variable effects on plant growth, but generally led to increased aphid fecundity. The primary nutritional challenge faced by aphids is the acquisition of nitrogen from phloem sap in the form of amino acids (Terra 1988). Nitrogen fertilization of a variety of grasses tends to increase the amount and alter the composition of amino acids in phloem, which can, in turn, increase aphid population growth rates (Weibull 1988, Fraser and Grime 1999, Isopp et al. 2000). Thus, our result that N fertilization increased plant biomass, tissue %N, and aphid fecundity is concordant with many previous studies (see review in Awmack and Leather 2002). Nutrient fertilization acts at both the individual and community levels, and can alter both tissue chemistry and community composition in a field setting. In a multigeneration field setting, fertilization can alter the biomass, relative abundance, and identities of grassland species, favoring invasive annual grasses in this system (Seabloom et al. 2003). In our short-term experiments, fertilization primarily affected tissue quality, but this effect of fertilization on aphid fecundity was not nearly as strong as the effect of grass life history. Thus, we expect that in natural grasslands, particularly those of the southern and central West Coast, chronic and increasing nitrogen deposition (Fenn et al. 2003a, b) will tend to shift the community toward annual grasses and may also increase tissue nitrogen in hosts (Seabloom et al. 2003, Henry et al. 2005), both of which should increase grain aphid populations. In choice tests, aphids were found at significantly higher densities on annual grasses compared to perennials; however, at least a few aphids were found on each of the grass species after 24 hours, indicating that preference for annuals was not complete. This result mirrors field choice tests in which R. padi aphids preferentially colonized the plant species on which they


had higher fecundity (Weibull 1988). This incomplete preference is critically important for disease transmission in West Coast grasslands: although annual grasses substantially increase aphid populations compared to perennials, perennial grasses serve as the among-season pathogen reservoir. For the B/CYDV group to be propagated, aphids must first visit an infected plant, which is most likely to be a perennial grass, early in the season. If aphids had a complete aversion to perennial grasses, the B/CYDV pathogen group could not persist in this system (Borer et al. 2007). Our results emphasize the importance of host context in controlling vector fecundity and host infection risk in complex natural communities. Previous studies have shown that aphid density on a plant, resulting from reproduction, is positively associated with production of alate (winged morph) aphids, as well as emigration of apterous (unwinged) aphids (Parry 1977, Donaldson et al. 2007), and elevated vector movement rates can substantially increase infection in a community (Burnett and Gill 1976, Kendall et al. 1992, Bailey et al. 1995, Power and Gray 1995). Our results, taken in the context of this previous work, suggest that aphid density is elevated above levels sustained prior to the extensive invasion of the U.S. West Coast by invasive annual grasses; invasive perennial grasses do not appear to play the same role. The impacts of invaders can be mediated through shared natural enemies (e.g., consumers, pathogens, vectors) via apparent competition (Tompkins et al. 2003, Borer et al. 2007, Orrock et al. 2008); however, this will occur only for exotic species that differ from natives in their effects on consumers. Our results demonstrate that the exotic pool is not uniform in its effects on vectors. Thus, in this case, the impact of novel species is not the result of host provenance, per se, but rather arises when invaders cause a change in the dynamics of a vector shared with natives. More generally, these results demonstrate that the impacts of invading species, and their responses to management, may not be determined by provenance (i.e., ‘‘eradication of exotics’’), but require a careful assessment of the most important characteristics determining their role in a community. In the case of West Coast grasslands, invasive annual grasses appear to cause a substantially greater impact on the dynamics of a shared pathogen (Malmstrom et al. 2005, Borer et al. 2009) and its vectors (our current results), compared to invasive perennial grasses. Unfortunately, the broad variety of management strategies employed to enhance native grasses in West Coast grasslands has generally been unsuccessful in eradicating annual grasses. However, these results do highlight the importance of recognizing that provenance alone may not be the most important characteristic determining the role and most effective management of novel invaders. Provenance, phylogeny, and life history can independently determine the role and best management of novel invaders.

July 2009

APHID FECUNDITY AND GRASSLAND INVASION ACKNOWLEDGMENTS

We thank S. Moore, B. Martin, and R. Allen for assisting with aphid wrangling, and M. Antolin for comments on an earlier version. We also thank two anonymous reviewers for insightful comments on this manuscript. Support for this project was provided in part by NSF/NIH EID 05-25666 (to E. Borer, A. Dobson, C. Mitchell, A. Power, and E. Seabloom), including REU funding for A. Adams associated with this grant, an OSU Research Office Undergraduate Research, Innovation, Scholarship, and Creativity grant (to V. Adams), and a grant from the Murdock Foundation (to E. Borer and G. Engler). LITERATURE CITED Agrawal, A. A., P. M. Kotanen, C. E. Mitchell, A. G. Power, W. Godsoe, and J. Klironomos. 2005. Enemy release? An experiment with congeneric plant pairs and diverse aboveand belowground enemies. Ecology 86:2979–2989. Anderson, P. K., A. A. Cunningham, N. G. Patel, F. J. Morales, P. R. Epstein, and P. Daszak. 2004. Emerging infectious diseases of plants: pathogen pollution, climate change and agrotechnology drivers. Trends in Ecology and Evolution 19:535–544. Anderson, R. M., and R. M. May. 1986. The invasion, persistence and spread of infectious diseases within animal and plant communities. Philosophical Transactions of the Royal Society B 314:533–570. Awmack, C. S., and S. R. Leather. 2002. Host plant quality and fecundity in herbivorous insects. Annual Review of Entomology 47:817–844. Bailey, S. M., M. E. Irwin, G. E. Kampmeier, C. E. Eastman, and A. D. Hewings. 1995. Physical and biological perturbations—their effect on the movement of apterous Rhopalosiphum padi (Homoptera, Aphididae) and localized spread of barley yellow dwarf virus. Environmental Entomology 24: 24–33. Barbehenn, R. V., D. N. Karowe, and Z. Chen. 2004. Performance of a generalist grasshopper on a C3 and a C4 grass: compensation for the effects of elevated CO2 on plant nutritional quality. Oecologia 140:96–103. Begon, M., R. G. Bowers, N. Kadianakis, and D. E. Hodgkinson. 1992. Disease and community structure: the importance of host self-regulation in a host–host-pathogen model. American Naturalist 139:1131–1150. Bigoga, J., L. Manga, V. Titanji, M. Coetzee, and R. Leke. 2007. Malaria vectors and transmission dynamics in coastal south-western Cameroon. Malaria Journal 6:5. Borer, E. T., P. R. Hosseini, E. W. Seabloom, and A. P. Dobson. 2007. Pathogen-induced reversal of native dominance in a grassland community. Proceedings of the National Academy of Sciences (USA) 104:5473–5478. Borer, E. T., C. E. Mitchell, A. G. Power, and E. W. Seabloom. 2009. Consumers indirectly increase infection risk in grassland food webs. Proceedings of the National Academy of Sciences (USA) 106(2)503–506. Burnett, P. A., and C. C. Gill. 1976. The response of cereals to increased dosage with barley yellow dwarf virus. Phytopathology 66:646–651. Collinge, S., and C. Ray. 2006. Community epidemiology. Pages 1–5 in S. Collinge and C. Ray, editors. Disease ecology. Oxford University Press, Oxford, UK. D’Arcy, C. 1995. Symptomology and host range of Barley Yellow Dwarf. Pages 9–28 in C. J. D’Arcy and P. A. Burnett, editors. Barley Yellow Dwarf: 40 years of progress. American Phytopathological Society, St. Paul, Minnesota, USA. Dixon, A. F. G. 1985. Aphid ecology. Blackie, Glasgow, UK. Dobson, A. 2004. Population dynamics of pathogens with multiple host species. American Naturalist 164:S64–S78. Dobson, A. P., and P. J. Hudson. 1986. Parasites, diseases and the structure of ecological communities. Trends in Ecology and Evolution 1:11–15.

1195

Donaldson, J. R., S. W. Myers, and C. Gratton. 2007. Densitydependent responses of soybean aphid (Aphis glycines Matsumura) populations to generalist predators in mid to late season soybean fields. Biological Control 43:111–118. Elser, J. J., W. F. Fagan, R. F. Denno, D. R. Dobberfuhl, A. Folarin, A. Huberty, S. Interlandi, S. S. Kilham, E. McCauley, K. L. Schulz, E. H. Siemann, and R. W. Sterner. 2000. Nutritional constraints in terrestrial and freshwater food webs. Nature 408:578–580. Fenn, M. E., J. S. Baron, E. B. Allen, H. M. Rueth, K. R. Nydick, L. Geiser, W. D. Bowman, J. O. Sickman, T. Meixner, D. W. Johnson, and P. Neitlich. 2003a. Ecological effects of nitrogen deposition in the western United States. BioScience 53:404–420. Fenn, M. E., R. Haeuber, G. S. Tonnesen, J. S. Baron, S. Grossman-Clarke, D. Hope, D. A. Jaffe, S. Copeland, L. Geiser, H. M. Rueth, and J. O. Sickman. 2003b. Nitrogen emissions, deposition, and monitoring in the western United States. BioScience 53:391–403. Fraser, L. H., and J. P. Grime. 1999. Aphid fitness on 13 grass species: a test of plant defence theory. Canadian Journal of Botany–Revue Canadienne de Botanique 77:1783–1789. Halbert, S., and D. Voegtlin. 1995. Biology and taxonomy of vectors of barley yellow dwarf viruses. Pages 217–258 in C. J. D’Arcy and P. A. Burnett, editors. Barley Yellow Dwarf: 40 years of progress. American Phytopathological Society, St. Paul, Minnesota, USA. Harpole, W. S., and D. Tilman. 2007. Grassland species loss resulting from reduced niche dimension. Nature 446:791–793. Heady, H. F. 1977. Valley grassland. Pages 491–514 in M. G. Barbour and J. Major, editors. Terrestrial vegetation of California. John Wiley and Sons, New York, New York, USA. Henry, H. A. L., E. E. Cleland, C. B. Field, and P. M. Vitousek. 2005. Interactive effects of elevated CO2, N deposition and climate change on plant litter quality in a California annual grassland. Oecologia 142:465–473. Holt, J., M. J. Jeger, J. M. Thresh, and G. W. Otim-Nape. 1997. An epidemiological model incorporating vector population dynamics applied to African cassava mosaic virus disease. Journal of Applied Ecology 34:793–806. Holt, R. D., A. P. Dobson, M. Begon, R. G. Bowers, and E. M. Schauber. 2003. Parasite establishment in host communities. Ecology Letters 6:837–842. Hudson, P., and J. Greenman. 1998. Competition mediated by parasites: biological and theoretical progress. Trends in Ecology and Evolution 13:387–390. Huenneke, L. F., S. P. Hamburg, R. Koide, H. A. Mooney, and P. M. Vitousek. 1990. Effects of soil resources on plant invasion and community structure in Californian serpentine grassland. Ecology 71:478–491. Isopp, H., M. Frehner, J. P. F. Almeida, H. Blum, M. Daepp, U. A. Hartwig, A. Luscher, S. Suter, and J. Nosberger. 2000. Nitrogen plays a major role in leaves when source-sink relations change: C and N metabolism in Lolium perenne growing under free air CO2 enrichment. Australian Journal of Plant Physiology 27:851–858. Jensen, S. G. 1969. Composition and metabolism of barley leaves infected with barley yellow dwarf virus. Phytopathology 59:1694–1698. Jensen, S. G., and C. D’Arcy. 1995. Effects of barley yellow dwarf on host plants. Pages 55–74 in C. J. D’Arcy and P. A. Burnett, editors. Barley Yellow Dwarf: 40 years of progress. American Phytopathological Society, St. Paul, Minnesota, USA. Karley, A. J., C. Hawes, P. P. M. Iannetta, and G. R. Squire. 2008. Intraspecific variation in Capsella bursa-pastoris in plant quality traits for insect herbivores. Weed Research 48: 147–156. Karley, A. J., J. W. Pitchford, A. E. Douglas, W. E. Parker, and J. J. Howard. 2003. The causes and processes of the midsummer population crash of the potato aphids Macrosiphum

1196


euphorbiae and Myzus persicae (Hemiptera: Aphididae). Bulletin of Entomological Research 93:425–438. Kendall, D. A., P. Brain, and N. E. Chinn. 1992. A simulation model of the epidemiology of barley yellow dwarf virus in winter sown cereals and its application to forecasting. Journal of Applied Ecology 29:414–426. MacDonald, G. 1957. The epidemiology and control of malaria. Oxford University Press, London, UK. Malmstrom, C. M., A. J. McCullogh, H. A. Johnson, and E. T. Borer. 2005. Invasive annual grasses indirectly increase virus incidence in California native perennial bunchgrasses. Oecologia 145:153–164. Mattson, W. J., Jr. 1980. Herbivory in relation to plant nitrogen content. Annual Review of Ecology and Systematics 11:119– 161. McCullagh, P., and J. A. Nelder. 1989. Generalized linear models. Second edition. Chapman and Hall, New York, New York, USA. Minakawa, N., E. Omukunda, G. F. Zhou, A. Githeko, and G. Y. Yan. 2006. Malaria vector productivity in relation to the highland environment in Kenya. American Journal of Tropical Medicine and Hygiene 75:448–453. Mitchell, C. E., and A. G. Power. 2003. Release of invasive plants from fungal and viral pathogens. Nature 421:625–627. Mooney, H. A., S. P. Hamburg, and J. A. Drake. 1986. The invasions of plants and animals into California. Pages 250– 272 in H. A. Mooney and J. A. Drake, editors. Ecology of biological invasions of North America and Hawaii. Springer, New York, New York, USA. Norman, R. M., M. Begon, and R. G. Bowers. 1994. The population dynamics of microparasites and vertebrate hosts: the importance of immunity and recovery. Theoretical Population Biology 46:96–119. Orrock, J. L., M. W. Witter, and O. J. Reichman. 2008. Apparent competition with an exotic plant reduces native plant establishment. Ecology 89:1168–1174. Ostfeld, R. S., C. D. Canham, K. Oggenfuss, R. J. Winchcombe, and F. Keesing. 2006. Climate, deer, rodents, and acorns as determinants of variation in Lyme-disease risk. Plos Biology 4:1058–1068. Park, T. 1948. Experimental studies of interspecific competition. I. Competition between populations of the flour beetles, Tribolium confusum and Tribolium castaneum. Ecological Monographs 18:267–307. Parry, W. H. 1977. Effects of nutrition and density on production of alate Elatobium abietinum on Sitka spruce. Oecologia 30:367–375. Perez-Harguindeguy, N., S. Diaz, J. H. C. Cornelissen, F. Vendramini, M. Cabido, and A. Castellanos. 2000. Chemistry and toughness predict leaf litter decomposition rates over a wide spectrum of functional types and taxa in central Argentina. Plant and Soil 218:21–30. Pimentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. Environmental and economic costs of nonindigenous species in the United States. BioScience 50:53–65. Pope, K., P. Masuoka, E. Rejmankova, J. Grieco, S. Johnson, and D. Roberts. 2005. Mosquito habitats, land use, and malaria risk in Belize from satellite imagery. Ecological Applications 15:1223–1232. Porter, C. H., R. C. Collins, and A. D. Brandlingbennett. 1988. Vector density, parasite prevalence, and transmission of Onchocerca volvulus in Guatemala. American Journal of Tropical Medicine and Hygiene 39:567–574. Powell, G., C. R. Tosh, and J. Hardie. 2006. Host plant selection by aphids: behavioral, evolutionary, and applied perspectives. Annual Review of Entomology 51:309–330. Power, A. G., and A. S. Flecker. 2003. Virus specificity in disease systems: are species redundant? Pages 330–346 in P.


Kareiva and S. A. Levin, editors. The importance of species: perspectives on expendability and triage. Princeton University Press, Princeton, New Jersey, USA. Power, A. G., and S. M. Gray. 1995. Aphid transmission of barley yellow dwarf viruses: interactions between viruses, vectors, and host plants. Pages 259–289 in C. D’Arcy and P. A. Burnett, editors. Barley Yellow Dwarf: 40 years of progress. American Phytopathological Society, St. Paul, Minnesota, USA. Rautapaa, J. 1970. Preference of cereal aphids for various cereal varieties and species of Gramineae, Juncaceae, and Cyperaceae. Annales Agriculturae Fenniae 9:267–277. Rosenthal, J. P., and R. Dirzo. 1997. Effects of life history, domestication and agronomic selection on plant defence against insects: Evidence from maizes and wild relatives. Evolutionary Ecology 11:337–355. Ross, R. 1911. The prevention of malaria. John Murray, London, UK. Seabloom, E. W., O. N. Bjornstad, B. M. Bolker, and O. J. Reichman. 2005. The spatial signature of environmental heterogeneity, dispersal, and competition in successional grasslands. Ecological Monographs199–214. Seabloom, E. W., W. S. Harpole, O. J. Reichman, and D. Tilman. 2003. Invasion, competitive dominance, and resource use by exotic and native California grassland species. Proceedings of the National Academy of Sciences (USA) 100:13384–13389. Seabloom, E. W., P. R. Hosseini, A. G. Power, and E. T. Borer. 2009. Diversity and composition of viral communities: coinfection of barley and cereal yellow dwarf viruses in California grasslands. American Naturalist, in press. Seabloom, E. W., J. W. Williams, D. Slayback, D. M. Stoms, J. H. Viers, and A. P. Dobson. 2006. Human impacts, plant invasion, and imperiled plant species in California. Ecological Applications 16:1338–1350. Strauss, S. Y., C. O. Webb, and N. Salamin. 2006. Exotic taxa less related to native species are more invasive. Proceedings of the National Academy of Sciences (USA) 103:5841–5845. Taylor, L. H., S. M. Latham, and M. E. J. Woolhouse. 2001. Risk factors for human disease emergence. Philosophical Transactions of the Royal Society B 356:983–989. Terra, W. R. 1988. Physiology and biochemistry of insect digestion—an evolutionary perspective. Brazilian Journal of Medical and Biological Research 21:675–734. Tompkins, D. M., A. R. White, and M. Boots. 2003. Ecological replacement of native red squirrels by invasive greys driven by disease. Ecology Letters 6:189–196. Torchin, M. E., K. D. Lafferty, A. P. Dobson, V. J. McKenzie, and A. M. Kuris. 2003. Introduced species and their missing parasites. Nature 421:628–630. Van Buskirk, J., and R. S. Ostfeld. 1995. Controlling Lyme disease by modifying the density and species composition of tick hosts. Ecological Applications 5:1133–1140. Vickerman, G. P., and S. D. Wratten. 1979. The biology and pest status of cereal aphids (Hemiptera: Aphididae) in Europe: a review. Bulletin of Entomological Research 69:1– 32. Weibull, J. 1988. Resistance in the wild crop relatives Avena macrostachya and Hordeum bogdani to the aphid Rhopalosiphum padi. Entomologia Experimentalis et Applicata 48:225– 232. Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying threats to imperiled species in the United States. BioScience 48:607–615. Woolhouse, M. E. J., L. H. Taylor, and D. T. Haydon. 2001. Population biology of multihost pathogens. Science 292: 1109–1112.

APPENDIX Full statistical model results for aphid field and laboratory experiments (I–IV) (Ecological Archives A019-045-A1).