| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Mycology and Plant Pathology |
Department of Biological Sciences, Wright State University, 3640 Colonel Glenn Highway, Dayton, Ohio 45435 USA
Received for publication December 22, 2006. Accepted for publication September 6, 2007.
ABSTRACT
Garlic mustard (Alliaria petiolata) is an invasive biennial that negatively impacts plant and animal communities throughout North America and lacks significant herbivory in its invasive range. Throughout Ohio, many garlic mustard populations support the powdery mildew fungus Erysiphe cruciferarum, although disease incidence varies among populations and environments. Effects of infection on plant growth, as well as both plant and fungal responses to drought and light conditions, were examined on greenhouse-grown, first-year garlic mustard plants. Also, the effects of the fungus on plant growth and fitness were studied in a naturally growing population of second-year plants in the field. Powdery mildew significantly reduced growth of first-year plants in the greenhouse, eventually causing complete mortality. Simulated drought slowed both plant growth and disease development, independent of light conditions. In the field, plants with little incidence of disease after their first year grew taller during their second year, producing significantly more siliques and twice as many seeds as heavily diseased plants did. Seed germination rates did not differ between plants with different levels of disease severity. Consistent reductions in survival, growth, and fitness caused by fungal infection may reduce populations of garlic mustard. These effects may be more evident in moist sites that favor fungal development.
Key Words: Alliaria petiolata • biocontrol • Erysiphe cruciferarum • infection • invasive species • plant fitness • plant growth
Because of their importance in determining agricultural productivity, the effects of plant pathogens on survival, growth, and reproduction of crop plants have been widely reported (e.g., Williamson and McFarlane Smith, 1986). Less studied are the effects of plant pathogens on wild plant populations (e.g., Jarosz and Davelos, 1995
), which can range from a mild depression in individual plant growth rate to widespread mortality and population declines. The literature also reveals that infection rates and impacts of plant pathogens on agricultural or wild plant populations can vary with environmental conditions, often being more severe in moist environments or during wet years (e.g., Jarosz and Davelos, 1995
). Because herbivores and pathogens may exert top-down control on plant populations, escape from population regulation by herbivores and pathogens can contribute to the dramatic success of invasive plants in novel habitats (Carpenter and Cappuccino, 2005
). The introduction of biocontrol agents to control invasive plants that have otherwise escaped attack is an approach that has succeeded in some circumstances (Culliney, 2005
).
Garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande (Brassicaceae), is a Eurasian biennial herb that is invasive in North America. It has the potential to negatively impact native plant, insect, and microbial communities (Renwick et al., 2001
; Prati and Bossdorf, 2004
; Stinson et al., 2006
). Factors contributing to its success include a high reproductive rate, self-compatibility, early season germination and growth, survival in poor soil conditions, allelopathy, and a lack of herbivores (Meekins and McCarthy, 2000
; Susko and Lovett-Doust, 2000a
; Blossey et al., 2001
; Prati and Bossdorf, 2004
). Widespread but apparently insignificant attacks have been reported by some insects and several fungal pathogens, but none of these have prevented range expansion of garlic mustard (Blossey et al., 2001
). Several insects from the native range of garlic mustard are currently being evaluated for their potential to control garlic mustard populations (Blossey et al., 2001
; Davis et al., 2006
), with some monophagous weevils in the genus Scrobicolus slated for test releases. Impacts of pathogens on garlic mustard individuals and populations have never been evaluated.
The incidence of a powdery mildew fungus on garlic mustard populations has been increasing throughout southwestern Ohio (D. Cipollini and S. Enright, personal observation). This fungus was accidentally introduced to our greenhouse on some infected garlic mustard plants from a local population. We identified it as Erysiphe cruciferarum Opiz ex. L. Junell (Koch and Slusarenko, 1990
; Plotnikova et al., 1998
), which has also been noted on garlic mustard populations in Indiana (Blossey et al., 2001
). Erysiphe cruciferarum is distributed worldwide and causes the main powdery mildew of crucifers (Koch and Slusarenko, 1990
), including Brassica crops (Williamson and MacFarlane Smith, 1986
; Karakaya et al., 1993
; Koike, 1997
; Kumar and Saharan, 2002
) and Arabidopsis thaliana (Koch and Slusarenko, 1990
; Kunkel, 1996
). There are reports of E. cruciferarum infecting garlic mustard in the British Isles (Ellis and Ellis, 1997
; Newsletter of the Warwickshire Fungus Survey, 2004
), Germany (Erysiphales Collection at the Botanische Staatssammlung München, Global Diversity Information Facility, http://data.gbif.org/datasets/resource/1440 [accessed 9 November 2006), and Armenia (Erysiphales Collection at the University Halle-Wittenberg, http://www.gbif-mycology.de/DatabaseClients/HALcoll/index.html [accessed 9 November 2006]), indicating that the fungus infects garlic mustard in both the native and introduced ranges of the plant.
Erysiphe spp. can infect any aboveground plant part and can cause heavy yield losses in Brassica crops by reducing plant growth and consequently, the quantity and quality of seeds (Williamson and MacFarlane Smith, 1986
; Adam and Somerville, 1996
; Penaud, 1999
; Dange et al., 2002
; Kumar and Saharan, 2002
). Climatic factors such as temperature and precipitation can influence the development of powdery mildews on plant surfaces (Williamson and MacFarlane Smith, 1986
; Asher and Williams, 1991
; Penaud, 1999
), which indicates that effects of this fungus will vary across habitats and seasons. Fungal infections can also generally reduce the cold hardiness of plants, increasing the amount of frost damage (Paul and Ayres, 1986
). As the infection progresses, spreading chlorosis, dehydration, and necrosis can reduce plant fitness (Adam and Somerville, 1996
). In the field, any surviving garlic mustard plants that were infected in their first year are often smaller and have more morphological abnormalities at the start of their second year than those not infected (S. Enright and D. Cipollini, personal observation).
The quantitative effects of powdery mildews in the genus Erysiphe have been widely studied on agricultural crops and on some wild plant populations (Jarosz and Davelos, 1995
) but never on garlic mustard. Because garlic mustard largely lacks herbivores in North America, E. cruciferarum has the potential to serve as an important natural control for this species and may be manipulated for biocontrol. While infection by biotrophic fungal pathogens does not always increase mortality (Hatcher, 1995
), the indirect or sublethal long-term effects may limit the success of future garlic mustard populations. The overall goal of this study was to determine whether powdery mildew has the potential to regulate populations of garlic mustard and to examine the influence of some environmental factors on disease development and plant growth. First, we examined effects of the infection on first-year plant growth in the greenhouse. Second, we examined the effects of drought and shading on plant growth and disease development in the greenhouse because drought and shading appear to contribute to variation in disease severity on garlic mustard in the field (S. Enright and D. Cipollini, personal observation). Third, we examined the effects of disease severity during its first year on the growth and fitness of garlic mustard during its second year in the field.
MATERIALS AND METHODS
Experiments were conducted on garlic mustard plants from two local populations near Dayton, Ohio, USA (Wright State University Forest Preserve and Taylorsville MetroPark) that have been susceptible to powdery mildew in the field. For experiments on first-year plants in the greenhouse, garlic mustard seeds were collected from the Wright State University Forest Preserve and moist stratified at 4°C until germination (approximately 3 mo). Germinated seedlings were planted in moistened ProMix BX potting medium (Grace-Sierra; Premier Horticulture, Red Hill, Pennsylvania, USA) in 18-cell flats (280 mL per cell) that were randomly placed in a greenhouse with supplemental fluorescent light on a 14 h light/10 h dark cycle. Experiments on second-year plants were conducted on a garlic mustard population growing naturally in the field at Taylorsville MetroPark.
The program SAS (version 9.1, SAS Institute, Cary, North Carolina, USA) was used for all statistical analyses. Means were compared using Tukey's test. Any transformations suggested by the program were used and are indicated in parentheses.
Impact of infection on first-year plant growth
For this greenhouse experiment, 28 garlic mustard seedlings were planted, and half of the plants were protected from powdery mildew with a weekly spray of Daconil fungicide (SOLARIS Group of Monsanto, San Ramon, California, USA) at rates appropriate for Brassica crops (active ingredient: 29.6% chlorothalonil; (0.65 mL/L distilled water). In a pilot study with bolting, second-year plants with no incidence of disease, fungicide-treated plants were significantly shorter than untreated plants (F1,10 = 14.25, P = 0.0036) after 5 wk of once-weekly fungicide treatment; the mean (SE) heights were 36.6 (2.71) cm for the control and 20.2 (3.47) cm for the fungicide-treated plants. The area of the largest leaf, however, was not significantly affected by fungicide treatment (F1,10 = 0.138, P = 0.7176); the area was 44.9 (3.93) cm2 for the control and 42.6 (4.61) cm2 for the fungicide-treated plants. In the current study, untreated plants naturally acquired infection and were visibly diseased at 2 wk of age. While we are able to manually infect garlic mustard with spore suspensions of powdery mildew and intended to do so here, we simply took advantage of the natural infection that occurred on our experimental plants. Spores from this fungus germinate on plant surfaces and form white, star-shaped colonies of mycelia that produce conidiophores with asexual spores (Xiao et al., 1997
). In this experiment, external symptom development was confirmed by the appearance of these colonies, which also appear after we manually inoculate plants with powdery mildew spores.
From wk 5 through 10 after planting, the areas of the third, fourth, and fifth true leaves of plants in each treatment were calculated by first measuring the diameter at the widest part of each leaf. Because garlic mustard leaves are approximately circular, the area of each leaf was calculated using the formula 
r2. A repeated measures ANOVA was used to examine the effect of the infection and the interaction of infection and time on area of each leaf.
Drought and shading effects on first-year plant growth and mildew symptom development
For this greenhouse experiment, 40 garlic mustard plants were divided into four groups of 10 plants each, with each group given one of the following treatments: regular watering and ambient light, drought conditions and ambient light, regular watering and shading, or drought conditions and shading.
Moisture and light availability in the field will vary between years and locations, and therefore the levels selected for this experiment were certainly within a range that could be encountered in the field in this geographical area. Drought and shading treatments began when plants were 4 wk of age. At this time, all plants had naturally acquired infection from spores present in the greenhouse and had external powdery mildew symptoms on 5–10% of the total leaf area. During the experiment, the 20 plants that were regularly watered were given approximately 75 mL water when the soil surface started to dry. The other 20 plants that were exposed to drought conditions were given approximately 25 mL water when more than half of those plants started to wilt (every 2–3 d). Ten plants from each water treatment were exposed to ambient light conditions in the greenhouse, while the other 10 were shaded under a canopy of two layers of thin polyester mesh (Kleen Test Products, Milwaukee, Wisconsin, USA). Ambient light levels ranged from 120–740 µmol·m2·s PAR, with shaded plants receiving approximately 50% ambient light.
Areas of the fourth and fifth true leaves were measured (as described earlier) when they were fully expanded, i.e., 4 wk after initiation of the drought and shading. The percentage of the fourth true leaf area occupied by disease symptoms was also followed weekly for 5 wk from the beginning of the drought and shading treatments.
A two-way ANOVA was used to analyze the areas of the fourth and fifth true leaf, with water treatment, light treatment, and their interaction as fixed effects and leaf area (square-root transformed) as the response. A repeated measures ANOVA was used to analyze powdery mildew disease development over time on the fourth true leaf, with the same between-subject factors, and time as the within-subject factor.
Impact of first-year disease severity on second-year growth and fitness
The field site at Taylorsville MetroPark contained first-year garlic mustard plants with visible disease symptoms in 2004. This site was selected because the percentage of total leaf area occupied by powdery mildew symptoms varied greatly from plant to plant. We discovered this site in late November and intended to categorize each plant based on the percentage of total leaf area with visible powdery mildew, but freezing temperatures induced abscission of noticeably diseased leaves before the plants were categorized. Therefore, approximately 120 plants known to have been diseased were tagged, and early in their second year (April 2005), plants were assessed based on the degree of purpling or malformation on their leaves, which are stress responses likely resulting from the pathogen infection (among other possible stressors) (Gandikota et al., 2001
; Gould, 2004
). Plants known to have been infected with powdery mildew have these symptoms in both the greenhouse and field, even in the absence of leaves with external fungal symptoms (D. Cipollini and S. Enright, personal observation). Each plant was assigned a number from 0–3 for the degree of purpling and for the degree of malformation, based on the following classification: 0 = no signs of purpling or malformation, 1 = <30% of whole plant affected, 2 = 30–60% of whole plant affected, and 3 = >60% of whole plant affected. Scores for purpling and malformation were added together, with a possible maximum score of 6 (>60% of whole plant with both purpling and malformation). Plants were then categorized based on their final scores: 0 = mildly diseased group, 1–3 = intermediately diseased group, 4–6 = severely diseased group. Plants were not visibly affected by any other pathogens but were not tested for the presence of other pathogens.
Once bolting began in mid-April of the second year (2005), mildly diseased plants were sprayed biweekly with Daconil (0.65 mL/L distilled water) in an attempt to minimize disease severity through their second year. As our pilot study indicated, Daconil can decrease the height of bolting second-year plants. Because we expected powdery mildew infection to also decrease height in these plants, differences in height between mildly diseased fungicide-treated plants and more heavily diseased, untreated plants should be conservative. Fungicide was carefully sprayed only on aerial plant parts of the target plant with minimal overspraying onto the soil to minimize fungicide effects on soilborne fungi or neighboring plants. Heights of all plants were measured biweekly until maximum height was reached. Herbivore damage was examined but was too minimal and infrequent to analyze. By late summer, visible powdery mildew symptoms were very light in this population, so leaf samples were collected from 50 plants and cleared and stained with trypan blue following a modified protocol from Vogel and Somerville (2000)
, then microscopically observed to determine presence of powdery mildew infection during this second year. Siliques were collected as they matured, then counted and air-dried in the lab. The maximum number of seeds per silique was determined by counting seeds from the three largest siliques per plant. The average mass per seed was calculated by weighing 20 randomly chosen seeds from each plant. The total seed mass for each plant was measured, and the total seed number was calculated by dividing total seed mass by average mass per seed. For measuring germination rates of seeds from plants in each disease group, 20 randomly chosen seeds from each plant were moist stratified at 4°C, as described earlier. From the onset of germination to approximately 3 mo later, the number of germinated seeds was recorded every 4–5 d. A repeated measures ANOVA was used to analyze differences in heights and seed germination rates over time among the three levels of severity. A one-way ANOVA was used to examine reproductive output with disease severity as the factor and each of the following as responses: total silique number (log10 transformed), number of seeds per silique (squared transformed), mass per seed, and total seed number (log10 transformed).
RESULTS
Impact of infection on first-year plant growth
Plants had visible powdery mildew symptoms at 2 wk of age, and leaf areas were measured on control and infected plants from wk 5 through 10. The third true leaf was already significantly larger on control plants than on infected plants at 5 wk of age (Fig. 1A). Leaves in each treatment group continued to grow until fully expanded at 7 wk of age, when the third true leaf was approximately 1.8 times larger on control plants than on infected plants (time: F2,25 = 81.93, P < 0.0001; treatment: F1,25 = 57.37, P < 0.0001; time x treatment: F2,25 = 21.53, P < 0.0001). As the fourth and fifth true leaves expanded, their response to the treatments mirrored those of the third true leaf (Fig. 1B, C) (fourth leaf; time: F4,23 = 84.38, P < 0.0001; treatment: F1,23 = 113.4, P < 0.0001; time x treatment: F4,23 = 64.36, P < 0.0001) (fifth leaf; time: F2,25 = 117.76, P < 0.0001; treatment: F1,25 = 61.56, P < 0.0001; time x treatment: F2,25 = 68.9, P < 0.0001). Once fully expanded, the fourth true leaf was four times larger and the fifth true leaf was 15 times larger on control plants than on infected plants, indicating that the effects of infection increased with each new leaf. Developmentally, the fifth true leaf emerged 1–2 wk later on infected plants than on controls, and also expanded more slowly. In turn, the number and size of visible colonies increased through time on infected plants until they covered all aboveground plant parts. By 14 wk of age, all severely diseased plants had died, while all control plants remained alive.
|
|
|
Throughout the growing season, mildly diseased plants were significantly taller than the plants in the intermediate or severe groups (Fig. 4; time x trt: F8,178 = 6.6, P < 0.0001). The mildly diseased group also produced significantly more seeds per silique (Fig. 5A; F2,89 = 11.03, P < 0.0001), twice as many siliques per plant (Fig. 5B; F2,88 = 11.35, P < 0.0001), and twice as many seeds per plant (Fig. 5C; F2,89 = 12.74, P < 0.0001) than the other two severity groups, and had a significantly heavier mean seed mass than the severely diseased group (Fig. 5D; F2,89 = 4.708, P = 0.0114). The largest seed-producing plant produced almost 1900 seeds. Germination rates of seeds did not significantly differ among severity groups (Fig. 6; time x trt: F14,16 = 0.71, P = 0.7402), although seeds from the severely diseased plants tended to germinate more slowly than those from the mildly diseased group.
|
|
|
DISCUSSION
Garlic mustard is an important invasive plant in North America for which significant herbivory is largely lacking. The objective of our study was to determine the extent to which powdery mildew caused by E. cruciferarum is capable of reducing the growth and fitness of garlic mustard and to explore some environmental influences on its impacts. To our knowledge, this is the first report of the impacts of powdery mildew on growth and fitness of garlic mustard, and of the influence of the environment on disease development on this plant.
Powdery mildew infection substantially reduced the growth of young plants in the greenhouse, eventually leading to mortality. Plants not protected with fungicide had visible powdery mildew symptoms by 2 wk of age. At this time, leaf areas were already smaller on infected plants than on control plants, so the disease can quickly reduce plant growth and eventually kill the plant. This effect may have been facilitated by environmental conditions favorable to mildew development in the greenhouse, that is, consistently warm temperatures and elevated humidity. In the field, external powdery mildew symptoms typically start appearing in late June or early July in southwestern Ohio (D. Cipollini and S. Enright, personal observation), when climatic conditions (including temperature and humidity) permit spore germination and disease spread. By this time, first-year plants can be 3 mo old and may be large enough to survive powdery mildew, but the fungus may still reduce plant growth. However, if garlic mustard was infected early in its first year in the field, it would likely not survive to its second year. The disease could also impact other plant processes such as allocation to storage resources and induction of costly defenses, which could severely stress the plants over the winter and reduce their size entering the second year. Because plant size is a well-known predictor of fitness in garlic mustard (Susko and Lovett-Doust, 2000b
), reductions in size resulting from the disease that carry over into the next growing season should greatly reduce fitness, whether or not plants are infected in their second year.
Drought stress strongly reduced plant growth and delayed disease development. Plants in the field typically encounter abiotic stresses that influence plant or pathogen responses (Paul and Ayres, 1986
). Among abiotic stresses, moisture availability greatly affects plant growth and reproduction (Reddy et al., 2004
) and greatly affects pathogen development (Williamson and MacFarlane Smith, 1986
; Asher and Williams, 1991
; Penaud, 1999
). In water-limited environments, osmotic stress or stomatal closure in plants may reduce a pathogen's ability to enter through the leaf (Fredeen et al., 1991
; Creelman and Mullet, 1997
; Reddy et al., 2004
; Thaler and Bostock, 2004
). In our study, although drought apparently slowed disease development, plants were much smaller under drought conditions, with powdery mildew occupying the total leaf area by the end of the experiment. This indicates that powdery mildew can likely affect plant fitness under a wide range of moisture conditions in the field but may be particularly detrimental in moist, productive habitats.
Shading did not affect plant growth and pathogen development as much as drought stress. A similar pattern was seen in the resistance of barley to the powdery mildew Blumeria graminis (Wiese et al., 2004
), suggesting that leaf water status may be generally more important than light exposure to the severity of mildew symptoms. These findings indicate that the higher incidence of powdery mildew in the shade in the field (our anecdotal observations) is due to higher moisture rather than reduced light in the shade.
Powdery mildew substantially reduced growth and fitness of garlic mustard in the field; with mildly diseased plants growing taller and producing significantly heavier seeds, more siliques, and more seeds per silique than the intermediately or severely diseased plants. The positive correlation between height and reproductive output in our study agrees with other studies of fitness correlates in garlic mustard (Susko and Lovett-Doust, 2000b
; Smith et al., 2003
). The pathogen likely indirectly decreased fitness by reducing shoot height and overall plant biomass, thus reducing the amount of internal resources available for reproduction. Despite reduced growth, plants in the intermediate and severe disease groups still produced a relatively large number of seeds, thus the pathogen did not entirely prevent host reproduction in the field. Results of our pilot experiment suggested that Daconil treatment directly reduced the height of the mildly diseased plants, which makes the difference in fitness between them and the other disease groups conservative. Protection from other foliar or root fungal pathogens conferred by Daconil treatment, while possible, was unlikely given that overspray was minimal and that very few, if any, other fungal pathogens are known to associate with garlic mustard in North America (Blossey et al., 2001
).
Our results suggest that the fitness effects of powdery mildew on second-year plants extend largely from its effects on growth during the first year. In our field study, external mildew symptoms on second-year plants were minimal until very late in the season when climatic conditions permitted spore germination and when plants were already maturing seeds. At this time, most plants sampled had evidence of internal fungal hyphae, regardless of the disease severity determined at the start of the season. Nonetheless, new infections this late in the season on second-year plants are unlikely to have been responsible for the observed fitness effects of mildew. In a study of the fitness impacts of defoliation of second-year plants, removal of 50% of the leaf area of healthy plants at the bolting stage had no impact on seed production in garlic mustard (D. Cipollini, unpublished data). Factors that influence the size of garlic mustard entering the reproductive stage (like previous mildew infection) appear to be particularly important in determining fitness in this plant. We assume that the variation in disease severity within this garlic mustard population is due to variation in exposure to inoculum among plants. Although the resulting differences in plant growth and fitness could also be caused by within-population variation in susceptibility to the powdery mildew, garlic mustard often has little within-population variation in growth and production of chemical defenses (e.g., Cipollini, 2002
).
We observed no effect of previous disease incidence on seed germination rates, suggesting that powdery mildew is not transmitted through the seed. Similarly, infection with turnip mosaic virus had no effect on germination of garlic mustard seeds, even though the infected maternal plants produced smaller siliques and smaller seeds (Stobbs and van Schagen, 1987
). Susko and Lovett-Doust (2000a)
found that garlic mustard seed germination varied with seed mass in plants not infected. Seed mass differed by up to 45% between groups in their study, while it differed by only up to 13% in our study. While seeds from severely diseased plants tended to germinate at a lower percentage, variation among groups in seed size may have been too small to generate significant differences in germination rates. Furthermore, growth and development of these seedlings may have differed but were not investigated.
While not capable of immediately eliminating garlic mustard populations, mildew-induced reductions in survival, growth, and reproduction that extend across several years should slow population expansion. According to a modeling study of the impacts of four specialist weevils being considered for introduction in North America, biocontrol agents capable of causing both rosette mortality and reductions in seed output will most strongly affect the demography of garlic mustard populations (Davis et al., 2006
). Based on our limited laboratory and field data, powdery mildew may be as capable as, or even better than, any of the weevils examined by Davis et al. (2006)
at reducing garlic mustard survival and population spread. This modeling study also revealed that single biocontrol agents may be incapable of controlling garlic mustard in highly productive habitats (Davis et al., 2006
). Plants from the population that we studied produced a higher mean number of seeds per plant than have been reported for uninfected garlic mustard in other field studies (Meekins and McCarthy, 2000
; Susko and Lovett-Doust, 2000a
; Smith et al., 2003
). This may relate to the relatively low density of garlic mustard in our field site that enabled plants to grow large (Meekins and McCarthy, 2000
; Bossdorf et al., 2004
). Nonetheless, powdery mildew was capable of reducing seed output by 50% in severely diseased plants and should reduce fitness even more in higher density, or otherwise lower yielding, populations.
On a large scale, biocontrol may be the only effective method for reducing garlic mustard populations (Anderson et al., 1996
), especially because other means of control are difficult to implement in natural ecosystems (McFadyen, 1998
). With its dramatic effects on growth and fitness, powdery mildew has the potential to be an important natural control of this plant and may contribute to the benign nature of garlic mustard in its native range. There are programs that control invasive plant species through biological means, especially using insect herbivores (Blossey et al., 2001
; Davis et al., 2006
), but pathogens, primarily fungi, have also been used to control weedy plant species worldwide with great success (McFadyen, 1998
; Ellison and Barreto, 2004
; Trujillo, 2005
). For example, E. cynoglossi significantly reduced plant growth and fitness in houndstongue, an invasive biennial plant in forested rangelands of Canada and the northwestern USA (DeClerck-Floate, 1999
).
Conidia of powdery mildew are windblown, but the rate of their spread among garlic mustard populations is currently unknown, as are the long-term impacts on garlic mustard populations. In terms of biocontrol, garlic mustard populations can be manually inoculated with the conidia. For the greatest impact on fitness, the first-year rosettes should be inoculated. However, the distance that the conidia are naturally dispersed from a host needs to be tested to determine an adequate distance from other wild and cultivated Brassicaceous plants that may be susceptible. A growing concern about biocontrol is the potential damage to nontarget organisms (McFadyen, 1998
). As a crop pathogen, powdery mildew is sure to receive scrutiny as a biocontrol for garlic mustard, but the fact remains that it is already present in the field and may be appropriate to use in some circumstances. Our powdery mildew isolate was capable of infecting Brassica kaber and B. rapa in the greenhouse (both European weeds in North America) but infected B. napus (cv. Westar) only minimally and did not infect Arabidopsis thaliana (Ecotype Columbia) (D. Cipollini and S. Enright, unpublished data). We have not yet tested the susceptibility of any native crucifers to this fungus, but the phenology of many of the spring ephemeral crucifers native to forest understories where garlic mustard occurs may allow them to temporally escape infection. In turn, different strains of E. cruciferarum likely exist that vary in their virulence on garlic mustard and other crucifers, and garlic mustard strains may similarly vary in susceptibility to the fungus. If substantial variation in susceptibility does exist, then this powdery mildew could serve as an important selective agent on resistance of garlic mustard. However, the same would be true for weevils used for biocontrol if susceptibility to weevils varies substantially. Thus, more strains of both the garlic mustard and powdery mildew need to be examined to fully understand this host–pathogen interaction. Preferably, all possible consequences should be thoroughly studied beforehand because once the fungus is released, it cannot be eradicated (McFadyen, 1998
). For powdery mildew, that opportunity has passed in locations where it is already present. Although infected plants may still survive to produce seed, garlic mustard may not need to be completely eliminated to control its invasion and spread. Successful biocontrol does not eliminate the target plant but rather gradually reduces its spread so native vegetation can gradually compete and replace the plant (McFadyen, 1998
). With reduced vigor of individual plants and smaller population sizes of garlic mustard, native plant species may be able to better compete for resources.
FOOTNOTES
1 The authors thank K. Barto, K. Pinson, C. Rainey, J. Mbagwu, and M. Enright for technical support and Five Rivers MetroParks for land use. Comments by two anonymous reviewers substantially improved this manuscript. This research was supported by Wright State University and a grant from the Ohio Board of Regents. 
2 Author for correspondence (e-mail: stephanie.enright{at}wright.edu
)
LITERATURE CITED
Adam L. Somerville S.. 1996. Genetic characterization of five powdery mildew disease resistance loci in Arabidopsis thaliana. Plant Journal 9: 341-356..[CrossRef][ISI][Medline]
Anderson R. C. Dhillion S. S. Kelley T. M.. 1996. Aspects of the ecology of an invasive plant, garlic mustard (Alliaria petiolata), in central Illinois. Restoration Ecology 4: 181-191..[CrossRef][ISI]
Asher M. J. C. Williams G. E.. 1991. Forecasting the national incidence of sugar-beet powdery mildew from weather data in Britain. Plant Pathology 40: 100-107..[CrossRef][ISI]
Blossey B. Nuzzo V. Hinz H. Gerber E.. 2001. Developing biological control of Alliaria petiolata (M.Bieb.) Cavara and Grande (garlic mustard). Natural Areas Journal 21: 357-367..[ISI]
Bossdorf O. Prati D. Auge H. Schmid B.. 2004. Reduced competitive ability in an invasive plant. Ecology Letters 7: 346-353..[CrossRef][ISI]
Carpenter D. Cappuccino N.. 2005. Herbivory, time since introduction and the invasiveness of exotic plants. Journal of Ecology 93: 315-321..[CrossRef][ISI]
Cipollini D. F.. 2002. Variation in the expression of chemical defenses in Alliaria petiolata (Brassicaceae) in the field and common garden. American Journal of Botany 89: 1422-1430..
Creelman R. A. Mullet J. E.. 1997. Biosynthesis and action of jasmonates in plants. Annual Review of Plant Physiology & Plant Molecular Biology 48: 355-381..[CrossRef][ISI][Medline]
Culliney T. W.. 2005. Benefits of classical biological control for managing invasive plants. Critical Reviews in Plant Sciences 24: 131-150..[CrossRef][ISI]
Dange S. R. S. Patel R. L. Patel S. I. Patel K. K.. 2002. Assessment of losses in yield due to powdery mildew disease in mustard under North Gujarat conditions. Journal of Mycology and Plant Pathology 32: 249-250..
Davis A. S. Landis D. A. Nuzzo V. Blossey B. Gerber E. Hinz H. L.. 2006. Demographic models inform selection of biocontrol agents for garlic mustard (Alliaria petiolata). Ecological Applications 16: 2399-2410..[CrossRef][ISI][Medline]
De Clerck-Floate R.. 1999. Impact of Erysiphe cynoglossi on the growth and reproduction of the rangeland weed Cynoglossum officinale. Biological Control 15: 107-112..[CrossRef][ISI]
Ellis M. B. Ellis J. P.. 1997. Microfungi on land plants. Richmond Publishing, Slough, UK..
Ellison C. A. Barreto R. W.. 2004. Prospects for the management of invasive alien weeds using co-evolved fungal pathogens: a Latin American perspective. Biological Invasions 6: 23-45..[CrossRef][ISI]
Fredeen A. L. Gamon J. A. Field C. B.. 1991. Responses of photosynthesis and carbohydrate-partitioning to limitations in nitrogen and water availability in field-grown sunflower. Plant, Cell and Environment 14: 963-970..[CrossRef]
Gandikota M. De Kochko A. Chen L. Ithal N. Fauquet C. Reddy A. R.. 2001. Development of transgenic rice plants expressing maize anthocyanin genes and increased blast resistance. Molecular Breeding 7: 73-83..[CrossRef][ISI]
Gould K. S.. 2004. Nature's Swiss army knife: the diverse protective roles of anthocyanins in leaves. Journal of Biomedicine and Biotechnology 5: 314-320..
Hatcher P. E.. 1995. Three-way interactions between plant pathogenic fungi, herbivorous insects and their host plants. Biological Reviews of the Cambridge Philosophical Society 70: 639-694..
Jarosz A. M. Davelos A. L.. 1995. Effects of disease in wild plant populations and the evolution of pathogen aggressiveness. New Phytologist 129: 371-387..[CrossRef][ISI]
Karakaya A. Gray F. A. Koch D. W.. 1993. Powdery mildew of Brassica spp. in Wyoming. Plant Disease 77: 1063..
Koch E. Slusarenko A. J.. 1990. Fungal pathogens of Arabidopsis thaliana (L.) Heyhn. Botanica Helvetica 100: 257-268..[ISI]
Koike S. T.. 1997. First report of powdery mildew, caused by Erysiphe cruciferarum, on broccoli raab in California. Plant Disease 81: 1093..
Kumar S. Saharan G. S.. 2002. Sources of multiple disease resistance in Brassica spp. Journal of Mycology and Plant Pathology 32: 184-188..
Kunkel B. N.. 1996. A useful weed put to work: genetic analysis of disease resistance in Arabidopsis thaliana. Trends in Genetics 12: 63-69..[CrossRef][ISI][Medline]
McFadyen R. E.. 1998. Biological control of weeds. Annual Review of Entomology 43: 369-393..[CrossRef][ISI][Medline]
Meekins J. F. McCarthy B. C.. 2000. Responses of the biennial forest herb Alliaria petiolata to variation in population density, nutrient addition and light availability. Journal of Ecology 88: 447-463..[CrossRef][ISI]
Paul N. D. Ayres P. G.. 1986. The impact of a pathogen (Puccinia lagenophorae) on populations of groundsel (Senecio vulgaris) overwintering in the field. Journal of Ecology 74: 1069-1084..[CrossRef][ISI]
Penaud A.. 1999. Chemical control and yield losses caused by Erysiphe cruciferarum on oilseed rape in France. In Proceedings of the 10th International Rapeseed Congress, N. Wratten, P. A. Salisbury [eds.], The Regional Institute Ltd., Canberra, Australia. Website http://www.regional.org.au/au/gcirc/3/327.htm [accessed 3 August 2004]..
Plotnikova J. M. Reuber T. L. Ausubel F. M.. 1998. Powdery mildew pathogenesis of Arabidopsis thaliana. Mycologia 90: 1009-1016..[CrossRef][ISI]
Prati D. Bossdorf O.. 2004. Allelopathic inhibition of germination by Alliaria petiolata (Brassicaceae). American Journal of Botany 91: 285-288..
Reddy A. R. Chaitanya K. V. Vivekanandan M.. 2004. Drought-induced responses of photosynthesis and antioxidant metabolism in higher plants. Journal of Plant Physiology 161: 1189-1202..[CrossRef][ISI][Medline]
Renwick J. A. A. Zhang W. Haribal M. Attygalle A. B. Lopez K. D.. 2001. Dual chemical barriers protect a plant against different larval stages of an insect. Journal of Chemical Ecology 27: 1575-1583..[CrossRef][ISI][Medline]
Smith G. R. Dingfender H. A. Vaala D. A.. 2003. Effects of plant size and density on garlic mustard reproduction. Northeastern Naturalist 10: 269-276..[ISI]
Stinson K. A. Campbell S. A. Powell J. R. Wolfe B. E. Callaway R. M. Thelen G. C. Hallett S. G. Prati D. Klironomos J. N.. 2006. Invasive plant suppresses the growth of native tree seedlings by disrupting belowground mutualisms. Public Library of Science (PLoS), Biology 4: 727-731..
Stobbs L. W. Van Schagen J. G.. 1987. Occurrence and characterization of a turnip mosaic virus isolate infecting Alliaria petiolata in Ontario, Canada. Plant Disease 71: 965-968..[CrossRef][ISI]
Susko D. J. Lovett-Doust L.. 2000a. Patterns of seed mass variation and their effects on seedling traits in Alliaria petiolata (Brassicaceae). American Journal of Botany 87: 56-66..
Susko D. J. Lovett-Doust L.. 2000b. Plant-size and fruit-position effects on reproductive allocation in Alliaria petiolata (Brassicaceae). Canadian Journal of Botany 78: 1398-1407..
Thaler J. S. Bostock R. M.. 2004. Interactions between abscisic-acid-mediated responses and plant resistance to pathogens and insects. Ecology 85: 48-58..[CrossRef][ISI]
Trujillo E. E.. 2005. History and success of plant pathogens for biological control of introduced weeds in Hawaii. Biological Control 33: 113-122..[CrossRef][ISI]
Vogel J. Somerville S.. 2000. Isolation and characterization of powdery mildew-resistant Arabidopsis mutants. Proceedings of the National Academy of Sciences, USA 97: 1897-1902..
Warwickshire Fungus Survey.. 2004. Newsletter of the Warwickshire Fungus Survey for 2004. Affiliated with the British Mycological Society. Website http://freespace.virgin.net/william.moodie/wfsXNews04.rtf [accessed 9 November 2006]..
Wiese J. Kranz T. Schubert S.. 2004. Induction of pathogen resistance in barley by abiotic stress. Plant Biology 6: 529-536..[CrossRef][Medline]
Williamson C. J. MacFarlane Smith W. H.. 1986. Fungicidal control of powdery mildew and its effect on yield, digestibility and chemical composition of eight forage rape cultivars. Journal of Agricultural Science, Cambridge 107: 385-391..
Xiao S. Ellwood S. Findlay K. Oliver R. P. Turner J. G.. 1997. Characterization of three loci controlling resistance of Arabidopsis thaliana accession Ms-0 to two powdery mildew diseases. Plant Journal 12: 757-768..[CrossRef][ISI][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.05.
