ABSTRACT
EFFECT OF ALLIARIA PETIOLATA INVASION ON ECTOMYCORRHIZAL COLONIZATION OF QUERCUS RUBRA
By Steven Michael Castellano
Exotic plant invasion may result in the disruption of symbioses between plants and soil biota, potentially affecting plant fitness and community composition. I tested whether the invasive biennial Alliaria petiolata (garlic mustard) affects abundance and community composition of ectomycorrhizal fungi (EMF) associated with experimentally planted Quercus rubra (northern red oak) seedlings. Proportional EMF colonization, EMF species richness and diversity, and Q. rubra survival were lower for seedlings in a stand with high density A. petiolata than in a stand without A. petiolata. Community composition of EMF also differed between stands. A comparison of naturally occurring Q. rubra seedlings in a moderate vs. a low density site showed a trend toward less EMF with more A. petiolata, while richness and diversity did not differ. Though confounded by uncontrolled environmental variables, these results are consistent with A. petiolata negatively impacting ectomycorrhizal fungi, warranting more research on this indirect effect on native plants.
EFFECT OF ALLIARIA PETIOLATA INVASION ON ECTOMYCORRHIZAL COLONIZATION OF QUERCUS RUBRA
A Thesis
Submitted to the
Faculty of Miami University
in partial fulfillment
of the requirements for the degree of
Master of Science
Department of Botany
by
Steven Michael Castellano
Miami University
Oxford, Ohio
2008
Advisor ______David L. Gorchov
Reader ______Nicholas P. Money
Reader ______Charles Kwit
Table of Contents
Introduction: 1 Methods: 5 Results: 13 Discussion: 15 Conclusions: 20 Literature Cited: 22 Appendices: Appendix A. 1938 aerial photograph of Tanglewood Preserve 46 Appendix B. Tree composition for all study sites 47 Appendix C. Aerial photograph of Reinhart Preserve and soil series description 49 Appendix D. Aerial photograph of Kramer Woods and soil series description 51 Appendix E. Aerial photograph of Tanglewood Preserve and soil series description 53 Appendiz F. Estimation of Alliaria petiolata cover at Tanglewood Preserve 55 Appendix G. Soil nutrient analysis of Tanglewood and Reinhart Preserves 56 Appendix H. Fungal morphotype description and total proportional abundance for 57 three sites Appendix I. Color plate of selected ECM morphotypes 58 Appendix J. Identities of successfully sequenced ITS samples of ECM morphotypes 59
ii List of Tables
Table 1. Contingency table of germination success in high and no Alliaria sites 30 Table 2. Contingency table of seedling survival in high and no Alliaria sites- 30 Harvest 1 Table 3. Contingency table of over-winter seedling survival in high and no Alliaria 31 sites- Harvest 2 Table 4. Three-way ANOVA of mycorrhizal colonization of planted seedlings 31 Table 5. Weighted one-way ANOVA of mycorrhizal colonization, comparative study 31 Table 6. Ectomycorrhizal morphotypes found on Quercus rubra seedlings from 32 temperate forest sites Table 7. Comparison of fungal richness and Shannon-Weiner index between high 33 and no Alliaria sites Table 8. Comparison of fungal richness, Shannon-Weiner index between moderate 33 and no Alliaria sites Table 9. ECM community comparison (MRPP) between high and no Alliaria sites 34 Table 10. ECM community comparison (MRPP) between moderate and no 34 Alliaria sites
List of Figures
Figure 1. Survival of seedlings from planting date to harvest 1 35 Figure 2. Proportional mycorrhizal colonization of planted seedlings: Harvest 1 36 Figure 3. Proportional mycorrhizal colonization of planted seedlings: Harvest 2 37 Figure 4. Proportional ECM colonization of field germinated and bare root seedlings 38 Figure 5. Proportional mycorrhizal colonization of naturally growing seedlings 39 Figure 6. Rank abundance curves of ECM morphotypes in high and no Alliaria 40 density sites- Harvest 1 Figure 7. Rank abundance curves of ECM morphotypes in high and no Alliaria 41 density sites- Harvest 2 Figure 8. Rank abundance curves of ECM morphotypes in moderate and no Alliaria 42 density sites Figure 9. NMDS ordination of ECM community in high and no Alliaria density 43 sites-Harvest 1 Figure 10. NMDS ordination of ECM community in high and no Alliaria density 44 sites-Harvest 2 Figure 11. NMDS ordination of ECM community in moderate and no Alliaria density 45 sites
iii
In Memory
Of
John W. Thieret (1926-2006)
Botanist, Teacher, and Friend
iv Acknowledgments
To the following people I am greatly indebted for their assistance on the project I present here. My advisor, Dr. David Gorchov, Professor of Botany, Miami University, who provided much needed guidance at all stages of this project, especially during the writing process as his editing skills greatly improved the quality of this thesis. I thank my other committee members: Nicholas P. Money, Professor of Botany, Miami University, whose delicate red pen danced the tango over every page of earlier drafts of the thesis, and to Charles Kwit, Visiting Professor, Miami University, for his ecological insights and his friendship, all of which improved the quality of my project. I am grateful for all the people who helped me with my field work: to Hayley Kilroy who braved the risk of frostbite with me to measure trees, to Jim Godby and Joe Fisher of the Cincinnati Park Board who helped plant acorns on a cold December morning, to my wife Meaghan and my daughters Savannah and Isabella who not only helped in the field but also in the home. Without my family’s support, love, and tolerance of me hiding in my office for days on end while I write, this project would have been doomed from the start. I thank everyone in the “Watson” lab, especially Aaron Kennedy, Pieter Pelser, Jenise Bauman, Susannah Fulton, and Melanie Link-Perez, for their help in learning molecular techniques and for providing reassurance when my reactions proved fruitless. I thank all my fellow graduate students for providing friendship and support during some of the more difficult stages of graduate education. I thank Hank Stevens, my “R Guru” for his help with some of the fancy code needed to produce the fancy figures in this thesis. I am grateful for the two pilots that drive our department, Barb Wilson and Vickie Sandlin, who do an amazing job at the helm and without whom, our department would not function. And finally, for my strength, my courage, and my perseverance even when my energy was low, and my burden great, I thank the wise and compassionate words of Jesus Christ. Thank you. Funding for this project was provided by MU Department of Botany Academic Challenge Grant, and the Summer Field Workshop in Botany. I thank Miami University Natural Areas and the Cincinnati Park Board for allowing me to conduct my field research on their property.
v Introduction: Some species that are introduced into areas outside their native range become invasive, and spread very rapidly. Invasive species are reported to be a major cause of species decline and loss of biodiversity (Wilcove et al. 1998) and result in an estimated $120 billion in environmental damages and loses in the United States each year, of which non-native plants are a major contributor (Pimental et al. 2005). Plant invasions have the potential to impact native plant communities through niche displacement, resource competition, allelopathy, alteration of nutrient cycling, hydrology, and fire regimes, and disruption of mutualistic relationships (Vitousek et al. 1997, Gordon 1998, Mack et al. 2000, Mooney and Cleland 2001, Ehrenfeld 2001, Ehrenfeld 2003, Orr et al. 2005, Stinson et al 2006) A relationship understudied in the past is that between invasive plants and their effects on below-ground mutualisms between native plants and soil microbes. Soil microbes, especially fungi, are important for decomposition and nutrient cycling ultimately controlling the availability of some nutrients to plants and thus, play an integral role in ecosystem functioning (Chapin et al.1997). Recently this important area of research has been gaining more attention, especially concerning relationships between invasive plants and beneficial soil mycorrhizal fungi (Mooney and Cleland 2001, Roberts and Anderson 2001, Reinhart and Callaway 2004, Stinson et al. 2006). The association of plants and mycorrhizal fungi is thought to be one of the most ubiquitous mutualisms on earth, occurring widely (Lilleskov et al. 2004, Peterson et al. 2004, Reinhart and Callaway 2006), involving an estimated 95 % of all plants (Smith and Read 1997) including most woodland herbs (Whigham 2004). These fungi form mutualistic relationships with plants, in which they contribute to plant acquisition of water and nutrients while utilizing the plant as a carbon source. While most land plants benefit from endomycorrhizae, colonizing interior tissues of the plant root, about 8000 (3%) seed plant species form associations with ectomycorrhizal (ECM) fungi (Taylor and Alexander 2005). ECM host plants, mostly woody perennials, including trees such as Betula (birch), Fagus (beech), Pinus (pine), and Quercus (oak), are typically dominant components of woodlands (Smith and Read 1997, Taylor and Alexander 2005). The diversity of ECM fungi is quite high and while exact species numbers are unknown, the global diversity has been estimated to be 7000-10,000 species (Taylor and Alexander 2005). Plants forming ectomycorrhizal associations benefit by having greater access
1 to mineral nutrients, increased nutrient absorption, protection from pathogens, and increased tolerance to environmental stresses such as water, salinity, pH, temperature, and heavy metal stress (Gupta et al. 2000). These benefits likely enhance host fitness, so long as costs due to carbon (photosynthate) losses are minimal. Some plants are recognized as having non-mutualistic interactions with fungi capable of mycorrhizal associations. It has been noted that non-mycorrhizal plants, such as members of the Brassicaceae (mustard family), may negatively affect important relationships between soil fungi and native plants (Mooney and Cleland 2001, Stinson et al. 2006, Callaway et al. 2008), potentially resulting in the decline of native plant species. Members of the Brassicaceae contain a diversity of secondary compounds, many of which deter herbivory (Freeland and Janzen 1974, Fahey et al. 2001), including cyanide containing compounds, flavonoids, and glucosinolates (Vaughn and Berhow 1999, Fahey et al. 2001, Cipollini and Gruner 2007, Callaway et al. 2008). Glucosinolates are organic compounds that are broken down by the enzyme myrosinase into several hydrolysis products including isothiocyanates, nitriles, and thiocyanates (Fahey et al. 2001), which are toxic to a variety of organisms including fungi (Mayton et al 1996). While Mayton’s study was concerned with pathogenic fungi, it is likely the results would be similar for beneficial mycorrhizal fungi associated with plant roots. Recent work by Callaway and colleagues (2008) found that flavonoids are also important antifungal compounds. A species capable of disrupting mutualistic associations may affect composition of plant communities with highly mycorrhizal-dependent systems, such as those consisting largely of hardwood tree seedlings (Stinson et al. 2006), while potentially enhancing its own spread through alteration of soil microbial communities and positive feedback (Bever 2003). This may result in lasting changes to natural plant communities. Here I investigate if Alliaria petiolata (M. Bieb) Cavara and Grande (Brassicaceae, garlic mustard), an introduced invasive forb, shows evidence of suppressing ECM interactions. Specifically, I am looking at the native hardwood tree Quercus rubra L. (northern red oak) and the associations it forms with ECM fungi. This species was chosen because it is an obligate ECM host plant (Smith and Read 1997) that generally benefits from ECM (Daughtridge et al. 1986, Reiske 2001, Courty et al 2007) and associates with a wide diversity of ECM fungi making it useful as “bait” for ECM (Walker et al 2005).
2 Originally from Eurasia, Alliaria petiolata is now widely recognized as an important invader (Rodgers et al. 2008). It was first documented in native communities in 1869 on Long Island, New York, and has since spread to much of the Midwest and Northeastern United States as well as some western states (Nuzzo 1991). As of 2000 A. petiolata occupied 34 US states and four Canadian provinces with populations growing exponentially (Nuzzo 1991, Nuzzo 1993a, Welk 2002). Alliaria petiolata is an obligate biennial producing overwintering rosettes in the first year, blooming from early spring through July of the second year, and producing fruit from June through September, after which the plant dies (Cavers et al. 1979, Baskin and Baskin 1992, Anderson et al. 1996, Byers and Quinn 1998). Alliaria petiolata is adapted for generalist pollinators, is capable of self-pollination (Anderson et al. 1996, Cruden et al. 1996), can produce as many as 9,500 to 107,000 seeds per m2 (Cavers et al. 1979, Nuzzo 1993b, Anderson et al.1996), and forms a seed bank viable for up to 5 years (Baskin and Baskin 1992, Nuzzo 2000). These characteristics allow a single A. petiolata plant to successfully found a new population (Cruden et al. 1996) that can rapidly grow into dense stands (Meekins and McCarthy 2002). While disturbance is associated with new A. petiolata establishment (Bartuszevige et al. 2007), it is not a prerequisite for successful establishment and survival (Meekins and McCarthy 2001, Bartuszevige et al. 2007). Once established, A. petiolata is difficult to eradicate (Nuzzo 2000, Slaughter et al. 2007) and becomes a permanent part of the forest community (Nuzzo 1999). Alliaria petiolata, while being outcompeted by Acer negundo, has been shown to outcompete seedlings of another native tree, Quercus prinus, compete equally with some native annuals such as Impatiens capensis (Meekins and McCarthy 1999), and reduce seed germination of the native perennial Geum laciniatum (Pratti and Bossdorf 2004). McCarthy (1997) also found that removal of A. petiolata resulted in an increase in cover of annuals, tree seedlings, and vines over the course of three years. A five year study evaluating the response of native plant communities following herbicide treatment of A. petiolata found that this treatment did not affect species richness or diversity, but it did increase cover of spring ephemerals and graminoids (Carlson and Gorchov 2004, Hochstedler et al. 2007). Stinson and colleagues (2007) found that while the response of native species to increasing A. petiolata varied among species, in general, species diversity was reduced with increasing A. petiolata and increased upon A. petiolata removal. More importantly they found, of all functional groups investigated, tree seedlings showed the
3 most notable decline with increasing levels of invasion and responded positively to partial removal of A. petiolata (Stinson et al. 2007). The impacts of A. petiolata on native plant communities may be linked to the disruption of mutualistic relationships between soil fungi and plants. Alliaria petiolata has shown to negatively affect spore germination and root colonization of endomycorrhizal fungi and interfere with the formation of mutualisms between fungi and plants, including native hardwood canopy trees (Roberts and Anderson 2001, Stinson et al. 2006), while being correlated with a reduction in host plant vigor (Callaway et al. 2008). These are traits not observed in its native region (Callaway et al. 2008). Cipollini and others (2008) provide further support for this as they found that A. petiolata extracts did not affect growth or reproduction of another non-mycorrhizal plant. The release of antifungal secondary compounds is believed to be a “novel weapon” (Callaway and Ridenour 2004) that kills non-resistant fungi ultimately adversely affecting the success of the mycorrhizal host plant. While A. petiolata’s effects on endomycorrhiza have been investigated (Roberts and Anderson 2001, Stinson et al. 2006, Callaway et al. 2008) similar investigations concerning affects on ECM are needed. Such a mechanism could simultaneously allow A. petiolata to successfully invade a forest ecosystem and lead to the decline of native host plants that benefit from ECM. This study compares the extent of ECM colonization of planted seedlings of a native, obligate ECM host, Quercus rubra L. (northern red oak), in sites of contrasting A. petiolata density, as well as colonization on naturally growing seedlings. My hypothesis was that A. petiolata has negative effects on the mutualism between Q. rubra and fungal symbionts. I predicted that Q. rubra grown in high density A. petiolata would have reduced ECM colonization. Because it is likely that some fungi are more tolerant of A. petiolata, I predicted lower diversity of fungi forming ectomycorrhizae, as less tolerant species are lost, and a shift in the fungal community composition to fewer, more tolerant species. I also predicted that seedling germination and survival would be lower at the high A. petiolata site. A negative effect on Quercus mycorrhizal colonization by A. petiolata would have implications for Quercus regeneration, possibly contributing to the decline of these trees in Eastern deciduous forests (Abrams 1992, Lorimer et al. 1994, Gribko et al. 2002) and a change in the composition of temperate forest communities.
4 Methods:
STUDY SITES. This study was conducted in three forested sites in southwestern Ohio, two in Butler County and one in Hamilton County. The study consisted of an experimental portion and a comparative portion. The experimental portion of this study was conducted at Reinhart Preserve (Butler CO, 39°31′22″N, 84°42′28″W) a Miami University Natural Area, and Bradford- Felter Tanglewood Preserve (hereafter referred to as Tanglewood Preserve) (Hamilton CO, 39°11′18″N, 84°33′24″W), a property of Cincinnati Park Board located in the Northside neighborhood of Cincinnati. Reinhart Preserve is contiguous with a larger natural area, Bachelor Reserve, and both are managed as a natural preserve. The land use history of this area was row crop farming and grazing, but after 1938 these activities declined dramatically and now successional areas and forests cover 80% of the immediate land area (Medley and Krisco 2007). The specific area where this study was conducted is closed canopy forest and aerial photographs reveal that this area has been tree covered since at least 1938. Tanglewood Preserve comprises a series of separate land acquisitions through donations and Nature Conservancy transfers beginning in 1938, with a large tract added in 1978, and smaller portions added in the early 1980’s for a total of 71.2 hectares (Cincinnati Parks 2008, TNC 2008). Tanglewood Preserve is kept in its natural state and aerial photographs reveal this site has been dominated by hardwood forests since at least 1938 (Appendix A). Sites used for the comparative study of this project were Reinhart Preserve, described above, and Kramer Woods (39°31′46″N, 84°42′59″W), a Miami University Natural Area which is a 4.9 hectare old regrowth stand of at least 100 years that was donated to the university in 1989 (Medley 1996). The tree canopy of all these sites is dominated by Acer saccharum (sugar maple) followed by Fraxinus spp. (ash) and Prunus serotina (wild black cherry) at Reinhart Preserve, Quercus rubra (northern red oak) and Fraxinus americana (white ash) at Tanglewood Preserve and Liriodendron tulipifera (tulip tree), Fraxinus spp. and, Fagus grandifolia (American beech) at Kramer Woods (Medley 1996, A. Maye and D. Gorchov, unpublished) (Appendix B). All three sites have important ectomycorrhizal host tree species abundant in the canopy (Appendix B) and are presumed to have ectomycorrhizal fungi present in the soil (Dickie et al. 2002). The soils of Reinhart Preserve and Kramer Woods consist of mostly moderately eroded Hennepin-
5 Miamian (HeE2) series silt loams with 18-25% slopes and at Tanglewood Preserve, mostly Eden (EcE) silty clay loam with 25- 40% slopes (NRCS 2007) (Appendix C-E). Alliaria petiolata density in these three sites vary from nearly absent at Reinhart Preserve, low at Kramer Woods, to extremely high at Tanglewood Preserve, contributing to 100% herb cover in some areas of this preserve (Castellano pers. observation). An estimate of A. petiolata density at Tanglewood Preserve, the high A. petiolata site, determined that the average density of second year plants across the research site in the summer of 2007 was approximately 6.7 / m2 and the density of first year plants was 33.4 / m2 (Appendix F). Although coverage was not quantified for other understory plants in this study, it was observed that both Reinhart Preserve and Kramer Woods had greater abundance of native understory herbs and tree saplings than Tanglewood, which was only sparsely populated with native herbs and was nearly devoid of tree seedlings. This was especially true of Quercus seedlings of which none were found anywhere near the Tanglewood study plot (Castellano pers. observation).
EXPERIMENTAL PLANTING DESIGN. The experimental portion of this project was conducted to investigate if planted seedlings in forests with very low and very high A. petiolata density (Reinhart Preserve and Tanglewood Preserve respectively) differ in their proportion of ectomycorrhizal (ECM) root tips, root fungal community composition, and overall survival. Quercus rubra acorns, collected in Richland County, Ohio (pers. comm. Lynn Brinley), were purchased from NN Seed Co. in Mansfield, Ohio. Prior to planting I discarded any acorns with weevil holes, and those that floated (indicating non-viability, Godman and Mattson 1980). Usable acorns were surface sterilized in a 10% bleach solution for approximately 10 minutes. I planted a total of 49 acorns at each site in the late fall and early winter of 2006 (26-27 Dec. 2006 at Tanglewood Preserve & 4 Jan. 2007 at Reinhart Preserve) to allow for required cold stratification in situ (Godman and Mattson 1980, Young and Young 1992). Single acorns were planted in mineral soil so that each was covered by approximately 0.5 cm of firmed soil, and covered with a thin layer of leaf litter (Sander 1990, Young and Young 1992). Each acorn was enclosed by a wire, vinyl-coated, mesh cage (mesh size 1.27 x 1.27 cm) to protect against seed predation and herbivory, from both above and below-ground. The top of the cages was anchored independently of the cage bottom to prevent acorn and or seedling disruption in the event the above ground portion of the cage was disturbed. Experimental plots were 15 x 15 m squares and planting design followed a systematic approach using a grid of 7 rows and 7 columns, spaced 2
6 m apart, with planting locations at every intersection and a buffer zone 1.5 m wide around the perimeter. Acorns that failed to germinate by May 2007 were replaced with greenhouse-germinated seedlings. These were derived from excess acorns that were cold stratified and germinated in moistened vermiculite at approximately 4° C during the winter of 2006 (Young and Young 1992), planted in coarse perlite in “cone tube” type pots, and grown under greenhouse conditions. Seedlings were regularly watered using greenhouse water supply which contained a very dilute (approximately 1 part fertilizer per 100 parts water) 21-7-7 fertilizer mix (pers. com. Harry Friedman). Prior to seedling planting, 10 seedlings were randomly selected for destructive sampling to ensure that ECM colonization did not occur in the greenhouse as some ECM species, such as members of the Thelephoraceae, are common greenhouse contaminates (Walker et al. 2005). Fine roots were inspected under a dissecting microscope for evidence of fungal hyphal or mantle development; none showed any indication of ECM colonization. A total of 31 seedlings were planted at Tanglewood Preserve (24 May 2007) and 25 seedlings were planted at Reinhart Preserve (25 May 2007) so that each site had a total of 49 living seedlings. My intention was to allow all the seedlings to grow throughout one growing season, and be harvested in the late fall of 2007. However, due to drought conditions during the summer and fall of 2007, seedling mortality and water stress occurred on both sites. In an effort to “rescue” samples from these dry conditions, a group of seedlings was harvested early in the fall (11 Sept. 2007 & 25 Sept. 2007). On each site the early harvest consisted of approximately half of the living seedlings on the site, specifically those appearing to be most affected by the drought. To ensure that sufficient time had elapsed for ECM colonization between the planting date and the date of the early harvest, three seedlings were randomly selected from each site, harvested, and evaluated for ECM activity prior to the harvest of the remaining seedlings. Colonization of ECM on these seedlings was apparent, and thus the time elapsed was sufficient for at least some degree of ECM colonization; the data collected from these six seedlings is included in the Harvest 1 data set. The remainder of the seedlings were left and harvested in the spring of 2008 (13-14 May 2008). Seedlings were stored and cleaned as described below.
SOIL ANALYSIS. In the spring of 2008 I collected ten soil samples, using systematic sampling, from the A horizon up to a maximum depth of 20 cm from each experimental site. Composite samples were made for each site, air dried at ambient temperature (Jones 2001) and
7 shipped to Spectrum Analytic Inc. (Washington Courthouse, Ohio) for general nutrient testing including available phosphorus and nitrogen. Lab methods include the Ion Selective Electrode method for N analysis and the Mehlich 3 method for other nutrients (pers. comm. Vernon Pabst, Spectrum Analytic).
COMPARATIVE STUDY- SEEDLING COLLECTION CRITERIA. This portion of the study was an observational approach to investigate if there are patterns in the abundance or composition of ectomycorrhizal fungi colonizing naturally occurring Q. rubra seedlings growing in forests with different densities of A. petiolata. Here I collected Q. rubra seedlings from Reinhart Preserve, a site with very low A. petiolata, and Kramer Woods, a site with a patchy distribution of A. petiolata. A total of 19 “A. petiolata present” seedlings were collected from Kramer Woods and 20 “A. petiolata absent” seedlings from Reinhart Preserve, in November of 2006. Although I did not determine the age of the seedlings, all collected seedlings were approximately 30 cm or less in height. A seedling considered to be “A. petiolata absent” had no trace of A. petiolata foliage, dead or alive, within a 2 m radius. This condition was easily met as the density of A. petiolata at the Reinhart plot was extremely low and I noticed only a few individual A. petiolata plants some distance from the study area. A seedling considered for the “A. petiolata present” group had at least one flowering or fruiting Alliaria stem within 20 cm and at least 4 flowering/fruiting stalks within 50 cm. Second year plants were used in deciding selection criteria because of the biennial nature of this plant; reproductive A. petiolata has been present on site for at least 2 years, at least one complete life cycle, allowing more time to effect a change in mycorrhizal activity. On both sites, Q. rubra seedlings collected were not closer than 1.5 m to another collected seedling to ensure independent sampling.
SEEDLING CLEANING AND STORAGE. Following collection, the seedlings and their intact root balls were stored in open plastic bags at approximately 4 ºC until they were analyzed. Seedling root balls were moistened as needed to prevent root and fungus desiccation (O’Dell et al. 1998). Prior to analysis, seedlings were soaked in distilled water to loosen adhering soil and carefully washed over a wire mesh screen to remove coarse and loosely adhered soil and organic material. Lateral roots were cut from the tap root and finer cleaning was done in distilled water under a dissecting microscope using forceps, fine metal probes, and squirt bottle (O’Dell et al.
8 1998). Once clean, roots were stored in distilled water at 4 ºC, until analysis (Visser 1995, Dickie and Reich 2005).
ROOT TIP SAMPLING. Fine root tips <1mm were evaluated for the presence or absence of ECM colonization. In the experimental study, fine roots were cut into 3-4 cm pieces and unbroken, living root tips were randomly selected for scoring until a total of 250 root tips per seedling were analyzed. The formation of a fungal mantle determined ECM colonization and colonization rate was determined as the proportion of ECM colonized tips to all root tips sampled; all ECM root tips were counted as individual tips regardless of whether they were individual monopoid mycorrhizae or part of larger clusters (Dickie and Reich 2005, Dickie et al. 2005). In the event that an individual did not have 250 root tips, counting included all live root tips present on the seedling. This sampling procedure was modified from an earlier sampling technique used in the comparative study in which roots were cut in 3-4 cm lengths and 10 lengths were randomly selected for scoring and all living, unbroken tips were considered with no maximum number, an average of 233 tips/plant. This procedure was modified to lessen the difficulty of root tip quantification and to provide a more standardized sample size. Ectomycorrhizae found on the fine roots were sorted into different morphotypes based on the color, growth pattern, and texture of the fungal mantle (Agerer 1993, Goodman et al. 1998). Abundance of each morphotype was quantified and an ECM sample of each was placed in micro centrifuge tubes, immediately frozen, and stored at -80° C for subsequent DNA extraction and molecular identification (see below).
COMPARISON OF ECTOMYCORRHIZAL COLONIZATION. The proportion of fine root ECM colonization to all root tips sampled was compared between experimentally planted seedlings from Reinhart Preserve and Tanglewood Preserve with fixed effects three-way Analysis of Variance (ANOVA) (α = 0.05) using the GLM procedure in SAS 9.1 (SAS Institute Inc. Cary, SC) with A. petiolata presence (site) as a factor, harvest date and seedling type (field germinated or bareroot) as fixed effects, and proportional colonization as the response variable. The UNIVARIATE procedure (SAS Institute Inc. Cary, SC) was used to ensure the assumption of normally distributed sample variance was met and proportional data was arcsine square root transformed to meet this condition.
9 The comparative study had no standard number of quantified root tips, so proportional colonization was compared between sites with a weighted one-way ANOVA (α = 0.05) with weighting applied to the number of root tips examined, as larger sample sizes more accurately represent the sample.
GERMINATION AND SEEDLING SURVIVAL. During the spring following planting each site was monitored regularly for seed germination until 24-25 May 2007, when all ungerminated acorns were deemed failures. Germination events were compared using a 2 x 2 contingency table analyzed with Fisher’s Exact Test in the statistical program R (Zar 1999, Dalgard 2002, R Development Core Team 2004). The survival of experimentally planted seedlings was compared between sites with high and very low density of A. petiolata. A seedling was considered to be alive if on 11 Sept. 2007 at Tanglewood and 25 Sept. 2007 at Reinhart it: 1) had a least some green, and presumably photosynthetically active, portions of the leaves, or 2) had living buds and a pliable stem for seedlings experiencing potentially drought-related early dormancy. Survival of all seedlings on each site, from the planting date to the date of Harvest one, were compared using 2 x 2 contingency tables and analyzed with Fisher’s exact test in the statistical program R (Zar 1999, Dalgard 2002, R Development Core Team 2004). Over-winter survival of the remaining seedlings (harvested 13-14 May 2008) was also compared in the same manner.
FUNGAL IDENTIFICATION. Mycorrhizal fungi were identified using direct genetic sequencing of the internal transcribed spacer (ITS) region of nuclear rDNA. This is a highly conserved repeat between the small and large subunit rDNA genes, with two variable non-coding regions, ITS-1 and the ITS-2, within the repeat (White et al. 1990, Gardes and Bruns 1993). The ITS region has been described as an effective target for the identification of fungal species and has been used often in mycorrhizal taxonomic studies (White et al. 1990, Gardes and Bruns 1993, Erland et al. 1997, Pritsch et al. 1997, Taylor and Bruns 1997, Mello et al. 1999, Gupta and Satyanarayana 2000, Walker 2005, Walker 2008). Multiple samples of ECM root tips of each morphotype were collected from seedlings on each site and washed with distilled water to remove soil debris and organic matter. Each sample was placed in a microcentrifuge tube, immediately fresh frozen and stored at -80 C° until DNA extraction. DNA was extracted from root tip samples using the DNeasy Plant Mini Kit protocol (Qiagen Inc.) and working aliquots were prepared and stored at 4° C. The ITS-1 and ITS-2
10 regions of extracted DNA were amplified with polymerase chain reaction (PCR) using the ITS1- F forward primer (5′ CTTGGTCATTTAGAGGAAGTAA 3′) and the ITS4 reverse primer (5′ TCCTCCGCTTATTGATATGC 3′). Used together these primers are specific to fungal sequences of both ascomycetes and basidiomycetes and effective in gene amplification for a variety of species of these fungal taxa (White et al. 1990, Gardes and Bruns 1993). Thermocycle parameters followed Gardes and Bruns (1993). Negative controls (no DNA) were run with every PCR to test for DNA contamination of reagent mixtures or buffers. PCR product was visualized using gel electrophoresis in 1% agarose gel, in 1X sodium borate buffer, for approximately 15-20 minutes at 130 mV, and stained with either ethidium bromide or Safeview™ Nucleic Acid stain (Applied Biological Materials Inc., Richmond, British Columbia) to verify that amplification of the ITS sequence was successful. Successful PCR product was cleaned using Promega (Madison, Wisconsin) PCR cleanup protocol and stored in nuclease-free water at 4° C until sequencing. In the event that PCR product from one root tip sample resulted in multiple fragment bands on the gel, the remaining PCR product from that sample was re-electrophoresed on a 1% agarose gel at a lower voltage of approximately 100 mV until there was enough distance between the bands to manually excise each from the gel; the time for adequate separation was typically 1- 1.5 hours. Excised bands were purified using the cleanup protocol above, re-amplified to increase concentration, and cleaned a second time. Clean PCR product was prepared for sequencing using BigDye ® Terminator v3.1Cycle Sequencing Kit (Applied Biosystems, Foster City, California) and protocol with the same primers as used in the initial PCR amplification; sequencing was carried out on an ABI Prism sequencer (Applied Biosystems, Foster City, California). Both the forward (ITS1-F) and reverse primer (ITS4) were used to sequence both strands of the DNA, aiding the editing process. These strands are complementary and missing bases were determined by comparison of the sequences of the two strands. Sequences were hand edited using the software application Squencher 4.8 (Gene Codes Corporation, Ann Arbor, Michigan) and compared to known ITS sequences using the UNITE database (Koljalg et al. 2005), a sequence database limited only to ECM fungal sequences obtained from voucher specimens. Sequence searches were performed using the galaxieBLAST search option which finds the best BLAST matches and performs web-based multiple phylogenetic alignments using a maximum parsimony model. This method is recommended to use for the identification of unknown ITS sequences as it can differentiate
11 similar, but evolutionarily unrelated, sequences (Nilsson et al.2004). Sequences containing highly conserved areas, such as the 5.8S in the ribosomal ITS region, may result in high BLAST scores using a similarity search even though the remaining sequences of the query are less well matched. The use of phylogenetic analysis allows the resolution of this situation, thus outperforming similarity searches (Nilsson et al. 2004). Species identity was determined by the best match, i.e. the species with lowest E-value, resulting from phylogenetic alignments. The E- value, or expected value, represents the number of sequence matches expected by random chance. When equally close matches to different species of the same genus were obtained for either one individual morphotype or two different morphotypes, taxon identification was described only by generic name. Any morphotype that was not successfully sequenced or that lacked BLAST matches was left as unidentified.
COMPARISON OF FUNGAL COMMUNITIES. To compare if A. petiolata is associated with different fungal community composition, I estimated diversity of ECM morphotypes for each seedling using richness (S, the total number of taxa), and the Shannon-Weiner index (H′ = -
∑pi[ln pi] where pi = proportion of total seedling ECM belonging to morphotype i) (Brower et al. 1998, Moser et al. 2005). Differences in values of each parameter between sites were tested by treating each seedling as a replicate and using a two sample T-test, or, when the parameters did not conform to assumptions of normality, a nonparametric two-sample Wilcoxon test. Separate tests were carried out for the two harvests in the experimental study. Rank abundance curves were drawn to visually compare the distribution of morphotype abundance among sites. Ordination using non-metric multidimensional scaling (NMDS), using the metaMDS function in the R package vegan (Oksanen 2008), was performed to compare differences in morphotype community composition among sites. Abundance data was square-root transformed and standardized using a double Wisconsin standardization. This method of standardization preserves the relevance of morphotype abundance, as opposed to presence/absence data, while diminishing the effect of highly dominant species and increasing the importance of more rare morphotypes (Oksanen 2008). The Bray-Curtis index was used as the dissimilarity measure to calculate the distance matrix of the standardized data. To test the null hypothesis that there was no difference in fungal morphotype composition between sites, I performed multiple response permutation procedure (MRPP) on Wisconsin double standardized abundance of morphotypes for each seedling with non-zero richness. This non-parametric, multivariate test calculates the
12 fraction of permuted pair-wise dissimilarities that are less than observed dissimilarities between sites (Oksanen 2008, Walker 2008). MRPP was performed using the mrpp function in the package vegan in R (Oksanen 2008); group size (n) was used as a weighting factor and a total of 10,000 permutations were run.
Results
SOIL ANALYSIS. Soil analysis revealed that Tanglewood Preserve had a higher pH and higher organic fraction than Reinhart Preserve and was much higher in all nutrients measured except magnesium (Mg) (Appendix G). Most striking were the much higher levels of nitrate
(NO3) and phosphorus (P) at Tanglewood than at Reinhart (17 vs. 2 ppm and 141 vs. 5 ppm, for + NO3 and P respectively. Ammonium (NH4 ) levels for these sites were not different (Appendix G).
GERMINATION OF FIELD PLANTED QUERCUS ACORNS. Germination of Quercus rubra was successful with 24 of 49 acorns germinating at Reinhart and 18 of 49 at Tanglewood (Table 1); these proportions were not significantly different using Fisher’s exact test (P = 0.307). It is possible that some of the ungerminated acorns may have germinated had they not been removed when replaced by bare-root seedlings.
SEEDLING SURVIVAL. Prior to the first harvest, 5 field-germinated seedlings died, 4 at Tanglewood and 1 at Reinhart, and 30 bareroot seedlings died, 17 at Tanglewood and 13 at Reinhart (Figure 1). No significant differences in survival were found for either seedling type across sites (P = 0.15 and P = 1.0) (Table 2). At both of these sites, many of the seedlings suffering mortality showed signs of water stress from the drought during the growing season of 2007 and browsing by white tailed deer. A significantly greater proportion of seedlings remaining on site after the first harvest died through the winter at Tanglewood, 6 of 15 seedlings, than at Reinhart, 1 of 18(P = 0.03) (Table 3).
ECTOMYCORRHIZAL COLONIZATION OF PLANTED SEEDLINGS. A combined total of 56 living seedlings were harvested from Reinhart Preserve and Tanglewood Preserve at two different times during this study. The first harvest (fall 2007) included 30 living seedlings. A total of 4,066 living root tips were observed from 17 seedlings harvested from Reinhart, 1,050 of which showed evidence of ECM colonization, an average proportion of 0.26. In sharp contrast only 248 of 3,250 tips analyzed from 13 harvested seedlings from Tanglewood had ECM, an
13 average proportion of 0.076 (Figure 2). A total of 26 living seedlings were harvested in the spring 2008, 17 from Reinhart Preserve and 9 from Tanglewood Preserve. From Reinhart, 1,085 of 4,144 root tips scored showed evidence of ECM colonization, an average proportion of 0.26. Only 133 of 2,046 root tips from Tanglewood, an average proportion of 0.06, showed signs of ECM colonization (Figure 3). Results from the three-way ANOVA show ECM colonization was significantly lower at Tanglewood Preserve, the high A. petiolata site, than at Reinhart Preserve, the no A. petiolata site (ANOVA: F = 45.72, P <0.0001), regardless of when the seedlings were harvested or whether they were planted as seeds or bareroot seedlings (all interactions not significant, Table 4). Seedlings grown from field germinated seed also had a higher proportional ECM colonization than the bare root planted seedlings (ANOVA: F = 4.48, P = 0.04) (Table 4) Figure 4).
COMPARATIVE STUDY- ECTOMYCORRHIZAL COLONIZATION. A total of 5466 living root tips were observed from 19 seedlings collected from ‘A. petiolata present’ sites at Kramer Woods, of which 1344 showed signs of ECM colonization, an average proportional colonization of 0.25 (Figure 5). A total of 3626 root tips were examined from 20 ‘A petiolata absent’ seedlings from Reinhart Preserve; of these 1184 were colonized, for a proportional colonization rate of 0.33 (Figure 5). There was a trend toward more colonization on the no A. petiolata samples but this difference was not significant (ANOVA: F = 3.19, P = 0.08) (Table 5).
FUNGAL MORPHOTYPE COMMUNITY. In all, 17 morphotypes were collected from the experimental and comparative study combined (Appendices H, I). ITS sequences of 13 of these types were successfully amplified and sequenced (Table 6). DNA from one type, C, was not successfully sequenced but the morphotype was identified as Cenococcum geophilum based on its distinct morphology (LoBugglio 1999); the remaining three types remain unidentified. Multiple root tips of the same morphotype often produced significant alignments with several different species (Appendix J). In each case the taxon is represented by genus only. It is likely that several closely related species would have very similar morphology, and multiple species were potentially catalogued as one morphotype. It should be noted that although several samples of each morphotype were collected, many samples were not successfully extracted and amplified. Thus, some species identities are based solely on one ITS sequence (Appendix J). Most of the genera identified have been recorded in the literature as growing on different species of Quercus (Table 6).
14 For both harvest dates, Reinhart (no A. petiolata) had significantly greater ECM morphotype richness and Shannon-Weiner diversity than Tanglewood (high A. petiolata) (Table 7). Richness and diversity tended to be slightly higher at Kramer (moderate A. petiolata) than at Reinhart, but significant differences were not found (Table 8). Rank abundance curves show that for both of the harvests, Reinhart had a large abundance of the most dominant types while other types were rarer (Figures 6,7); morphotype distribution at Tanglewood was more evenly distributed with dominant species being of less importance. Kramer and Reinhart were similar in the abundance of dominant and rare morphotypes (Figure 8). NMDS ordinations reveal differences in community composition for both harvests of planted seedlings (Figures 9, 10). MRPP confirmed that the ECM morphotype community compositions of Reinhart and Tanglewood were marginally different at the first harvest (P = 0.076), and differed significantly at the second harvest (P = 0.002) (Table 9). A combined total (both harvests) of 10 out 14 morphotypes were shared between Reinhart and Tanglewood. Four morphotypes were unique to Reinhart while no morphotypes were unique to Tanglewood. The only morphotype dominant at both sites was type Tomentella ellisii (Appendix H). NMDS ordination shows considerable overlap of community composition at Reinhart and Kramer (Figure 11); MRPP confirmed no differences in the composition of ECM between these sites (P = 0.105) (Table 10). Of 11 morphotypes, 8 were shared between sites, 1 was unique to Kramer, and 2 unique to Reinhart. Both sites had Russula sp. and Cenoccocum geophilum as dominant taxa (Appendix H).
Discussion
ECTOMYCORRHIZAL RESPONSE TO ALLIARIA PETIOLATA. My results are consistent with the hypothesis that A. petiolata has a negative impact on the mutualistic relationship shared by Q. rubra and ECM fungi. Results from the experimental portion of this project reveal significantly lower ECM root tip colonization at Tanglewood Preserve, a site with very high A. petiolata density, than at Reinhart Preserve, a site with extremely low, virtually non-existent, A. petiolata density. These results were similar for two separate harvests of Q. rubra seedlings, one harvested in the early fall and a second in the early spring, at each date only about 7% of root tips at Tanglewood were infected with ECM compared to about 26 % at Reinhart.
15 These results parallel earlier reports that the presence of A. petiolata potentially interferes with endomycorrhizal association (Rogers and Anderson 2001, Stinson et al. 2006, Callaway et al. 2008). Rodgers and others (2008) suggest this negative relationship may extend to the ectomycorrhizal fungi as well, based on unpublished data. To the best of our knowledge, no studies have been published with evidence that ECM is reduced in the presence of A. petiolata. ECM infection also tended to be lower on seedlings growing naturally near A. petiolata in an area of moderate A. petiolata density (Kramer) than in seedlings growing at the A. petiolata-absent site (Reinhart), but this trend was only marginally significant. This weaker effect at Kramer than at Tanglewood (experimental site with high A. petiolata densities) could be attributed to the lower densities of A. petiolata at Kramer. The lower diversity and richness of fungal morphotypes forming ECM associations at the high A. petiolata density site (Table 7) are also consistent with my hypothesis that A. petiolata negatively impacts ECM fungi. The community composition between sites was also different, albeit only marginally for the fall harvest (Table 8). I found that while all the morphotypes documented at Tanglewood, the high density site, were also documented at Reinhart, the latter had unique types. This raises the possibility that fungal species may differ in their resistance to A. petiolata effects and those less tolerant may be restricted from growing at the high density sites. Species more tolerant of A. petiolata would be expected to be more dominant in the community. While my results do not explicitly attribute a shift in dominant ECM species to A. petiolata, I did find that dominant species differed among sites. At the high A. petiolata density site dominant species included Tomentella ellisii and type L, whereas the dominant species at the low and moderate density sites were the unresolved Russula sp. and Cenoccocum geophilum (Appendix H). There were some minor shifts in dominant types on each site (Appendix H) from one harvest to the next, but these may potentially be seasonal trends. Though strong evidence of seasonal patterns is lacking, some studies show relative abundance and frequencies of some species varying with season (Koide et al. 2007, Walker et al. 2008). Walker and colleagues (2008) noted that Cenoccocum geophilum, a very common ECM type, was more common in samples collected in the fall than in summer; I report also that this species was more abundant in my fall samples than in my spring samples. However, in another study C. geophilum showed little seasonal variation (Koide et al. 2007). I also found a reduction in abundance of Russula
16 spp. and an increase in Lactarius sp., and Sebacina sp. from Harvest 1 to Harvest 2 for both experimental sites. Tomentella ellisii increased in Reinhart Preserve and decreased at Tanglewood Preserve between harvests (Appendix H). This further indicates that community dynamics may be site specific. While it wasn’t the focus of this study to investigate seasonal patterns of ECM, my data does support earlier hypotheses of temporal and seasonal dynamics (Koide et al. 2007, Walker et al. 2008) of ECM communities, and further illustrates the importance of multi-seasonal ECM collection when describing the ECM community of a site. In general, at least for the experimental part of this study, the predictions of reduced ECM colonization, reduced morphotype diversity and richness, and differences in community composition were mostly supported. Care, however, must be used in the interpretation of these results as there are potentially confounding environmental effects that were not controlled in this study. Lower ECM infection at Tanglewood might have been due to higher soil nutrients (Appendix G), at least during the spring when sampling took place. This is especially true of
nitrate nitrogen (N03), which was 8X greater than at Reinhart, and phosphorus (P), which was 28X greater. As nutrients become more available to plants, they are expected to form fewer associations with ECM, as carbon (C) allocation is adjusted, and fungi become C limited. In contrast as nutrients become more limiting, a greater C investment to mycorrhizal fungi is expected as associated fungi are beneficial in nutrient acquisition (Treseder 2004). In a meta- analysis comparing mycorrhizal response to these nutrients across field studies done in a variety of biomes, Treseder (2004) reports that in general, nitrogen fertilization tends to reduce mycorrhizae by about 14% and phosphorus reduces mycorrhizae by 32%, with no difference in response between ECM and arbuscular mycorrhizae (AM). However, this percent reduction in N represents data collected from studies using different quantification methods including hyphal length and percent colonization. Among studies only considering percent colonization, the method used in the current study, the reduction in colonization was somewhat lower (5.8%) (Treseder 2004). A more recent study found nitrogen fertilization caused no ECM reduction and actually increased arbuscular mycorrhizae (Garcia et al. 2008). In addition to reduced colonization of ECM, high N may also have affected community composition. Edgerton-Warburton and Allen (2000) observed that increasing nitrogen resulted in lower species richness and a shift in AM community composition. Because the high A. petiolata
17 stand was also higher in N and P, it is possible the lower ECM species richness and different community composition at this site are due to nutrients rather than A. petiolata. The high nutrient levels at Tanglewood might have been due to an outbreak of forest tent caterpillar (Malacosoma disstria) that occurred in spring 2007 and, to lesser extent, in the spring of 2008. Frost and Hunter (2004) found that eastern tent caterpillar (Malacosoma americanum) + frass deposition increased total soil N, the NH4 soil pool, and soluble NO3 in Q. rubra + mesocosms. My results from Tanglewood show a much higher soil NO3 while the NH4 level was comparable to Reinhart. Another hypothesis for the high N and P at Tanglewood considers A. petiolata and its relationship with the nutrient cycles in forests it invades. Ehrenfeld (2003) reviewed this relationship of a variety of introduced, invasive plants with soil nutrients and found in general, most were associated with higher levels of inorganic soil N, increased N mineralization, and increased nitrification. Rodgers et al. (2008) describe this trend of increased N for A. petiolata invaded soils as well; they also present preliminary data that suggest the same trend of increased P on invaded soils. Our data provide support for these conclusions, but can not distinguish if A. petiolata causes high nutrients or if A. petiolata is more likely to invade a nutrient rich site. Rodgers et al. (2008) also note that despite the increased nutrient levels found on sites with A. petiolata invasion, native plants do not show increased growth and survival, but rather negative responses are typically observed. This means that while potentially positively affecting soil quality, A. petiolata has negative effects as well. In addition to the direct effects summarized in the Introduction, A. petiolata may indirectly reduce ECM by creating eutrophic soil conditions unfavorable to infection or which cause the host tree to suppress infection. Potential interactions among N, P, ECM, and A. petiolata should be explored in greater depth using long-term manipulative experiments to distinguish between environmental variables and A. petiolata as potential stressors to successful ECM colonization, and how cumulative effects impact this important association and the diversity of the fungal community in general. Differences in herbivory between sites may have influenced the ECM-Quercus interaction. Herbivory from white-tailed deer and insects was evident at both sites. The seedlings at the Tanglewood Preserve suffered greater deer-related damage than those at Reinhart Preserve; in addition to this, Tanglewood Preserve also experienced the outbreak of forest tent caterpillars, which defoliated some seedlings and portions of the canopy. The effects that
18 herbivory has on ECM varies among reports. Rossow et al. (1997) found browsing mammals reduced the ECM abundance on willow and Mueller et al. (2005) report that Quercus turbinella had reduced ECM with increased insect herbivory. Other studies, however, report that herbivory did not affect ECM abundance and species richness, but did affect community composition (Saikkonen et al. 1999, Cullings et al. 2001). The greater herbivory at Tanglewood may have contributed to the low abundance and distinct community composition of ECM at this site.
QUERCUS GERMINATION AND SURVIVAL. Although Reinhart had a greater number of seeds germinate than Tanglewood, this difference was not significant. Thus, our prediction of reduced Q. rubra germination in high density A. petiolata is not supported by these data. It is important to consider here, that monitoring for germination events was halted in the spring of 2007 and all non-germinated acorns were considered failures though some potentially, were viable. Had the length of the study been longer, differences in germination may have been observed. However, if germination differences were observed, these would likely not be related to ECM activity but likely the result of direct effects of A. petiolata. Acorns were germinated in the absence of soil and soil fungi in our lab and resulting seedlings, after a season in the greenhouse, had no evident colonization by ECM prior to field planting. Contrary to my prediction that seedling survival at low A. petiolata density would be greater, survival to the first harvest did not differ between sites regardless of whether they were field-germinated or bare root. However the survival of bare root vs. field germinated seedlings was different within the sites, with the former having rapid mortality (Figure 1). Survival of seedlings grown from field-germinated seed was higher at each site when all seedlings are considered; interestingly these seedlings also showed a proportionally higher ECM colonization than the bare root seedlings. Field germinated seedlings were in the soil for a longer period of time than the bare root seedlings, perhaps this extra time allowed for greater ECM colonization enhancing survival. Also, transplanted seedlings possibly experienced some stress during transplant. The survival of seedlings left on site to be collected at Harvest 2 did, in fact, differ significantly between the sites when only deaths after Harvest 1 were considered; overwinter survival was greater at the no A. petiolata site, which is consistent with my prediction of increased survival with reduced A. petiolata (Table 3). ECM have been described as buffers against stressful environments (Gupta et al. 2000), ultimately contributing to host plant survival.
19 The lower overwinter mortality at the site without A. petiolata, and with greater ECM, is consistent with this view. Some of this mortality is likely explained by herbivore damage as this was evident on both sites, though deer browsing and forest tent caterpillar damage was more pronounced at Tanglewood Preserve than at Reinhart Preserve. Also, many of the seedlings at Reinhart Preserve that had been browsed showed regrowth from undamaged nodes and the root, and a flush of new leaves, showing resilience to some extent. This however was not the case at Tanglewood where regrowth was rare (Castellano pers. obs.). Most of the dead seedlings at each site showed evidence of water stress. During the summer and fall of 2008 both sites experienced drought conditions, which would be expected to cause mortality as soil moisture has been described as critical for survival through the first year (Gribko et al. 2002). The slower mortality at the no A. petiolata site may have been due to greater buffering of dry conditions by greater ECM colonization or richness. Swaty et al. (2004) found in their investigation of drought effects of ECM on a pine species, that sites with lower ECM richness had higher rates of mortality than sites with greater ECM richness. In general, while my data do not support my prediction that germination will be reduced at high A. petiolata densities, they do show support for the prediction of reduced Q. rubra seedling survival in heavily invaded areas during the first year. This trend was observed even though the high A. petiolata site had much greater nutrients, implying that mortality was not a result of nutrient limitation but was influenced by other factors, such as susceptibility to drought because of lack of ECM. My results here provide evidence that any negative effect A. petiolata has on the ECM community, either directly or indirectly, is likely to impact the successful establishment of Q. rubra, especially in exceptionally dry conditions.
Conclusions. I found that at the high A. petiolata site, ECM colonization was significantly reduced in both the fall and spring harvests of Q. rubra seedlings. In addition to this I report a significantly lower ECM species richness and diversity and different communities at sites with high and no A. petiolata. While germination was not affected, seedling survival through the first year was reduced at the high A. petiolata site. This negative relationship between A. petiolata and ECM abundance and diversity, coupled with a higher mortality at the high A. petiolata site suggests that the regeneration of Q. rubra is reduced at heavily invaded sites. This relationship
20 likely extends to other ECM trees and has the potential to cause lasting changes in plant communities. My data, while consistent with the hypothesis that A. petiolata disrupts belowground mutualisms, must be interpreted with caution. Interpretation is complicated by high levels on N and P on one site, drought, and herbivory, each of which could potentially influence the ECM activity as well as the survival of seedlings. While my study was not designed to distinguish between these environmental factors it does highlight the need for long term, manipulative experiments to tease apart the effects of secondary compounds from nutrient levels and environmental variables while controlling for herbivory. More complete knowledge of A. petiolata influence on beneficial soil fungi associating with canopy trees, and other native plants, is important in understanding the impact of this species on native communities. The results I report here may have implications for management of forests invaded with A. petiolata. Long term control of A. petiolata has proved to be a difficult task and the ability of a site to recover following control, is not fully known (Nuzzo 2000). Nuzzo (2000) suggests that high quality communities, containing greater diversity of native species and community structure, should recover on their own, provided that removal occurs before high A. petiolata densities are reached. Such communities likely have abundant and diverse assemblages of mycorrhizal fungi, much like Reinhart Preserve, with which naturally occurring as well as replanted species can associate. At high density A. petiolata sites, however, natural regeneration is less likely as these sites tend to be devoid of abundant native species, and replanting vegetation following A. petiolata removal may be required (Nuzzo 2000). The success of the replanting would likely be reduced if the soil lacks mycorrhizae. In that case, inoculation of fungi common to the particular area, or to the species planted, would greatly enhance recovery following A. petiolata removal. While my study shows an effect of A. petiolata on ECM, it does not include a “removal” treatment to evaluate ECM activity following removal. Of most concern would be the persistence of A. petiolata effect. In Stinson’s study (2006), tree seedlings planted in A. petiolata invaded soils had reduced fungal colonization and seedling growth than in non- invaded soils, even in the absence of A. petiolata individuals. This means residual effects may hinder replanting efforts even if they include inoculation, and the persistence of the effect is unknown. Long term studies involving removal treatments would provide key information that would greatly enhance restoration efforts following the removal of this invasive plant.
21 Literature Cited
ABRAMS MD. 1992. Fire and the development of oak forests. BioScience 42:346-353.
AGERER R. 1993. Some checklists for anatomical studies on ectomycorrhizae: Comments on check-list “A”, Morphological characteristics. In: Agerer editor. Colour atlas of ectomycorrhizae with glossary. Munich: Einhorn-Verlag. p 7i-16i.
ANDERSON RC, DHILLION SS, KELLEY TM. 1996. Aspects of the ecology of an invasive plant, garlic mustard (Alliaria petiolata), in central Illinois. Restoration Ecology 4:181-191.
BARTUSZEVIGE AM, HRENKO RL, GORCHOV DL. Effects of leaf litter on establishment, growth, and survival of invasive plant seedlings in a deciduous forest. American Midland Naturalist 158:472-477.
BASKIN JM, BASKIN CC. 1992, Seed-germination biology of the weedy biennial Alliaria petiolata. Natural Areas Journal 12:191-197.
BEVER JD. 2003. Soil community feedback and the coexistence of competitors: conceptual framework and empirical tests. New Phytologist 157:465-473.
BROWER JE, ZAR JH, VON ENDE CN. 1998. Field and laboratory methods for general ecology. McGraw Hill, Boston.
BYERS DL, QUINN JA. 1998. Demographic variation in Alliaria petiolata (Brassicaceae) in four contrasting habitats. Journal of the Torrey Botanical Society.
CALLAWAY RM, CIPOLLINI D, BARTO K, THELEN GC, HALLET SG, PRATI D, STINSON K, KILRONOMOS J. 2008. Novel weapons: invasive plant suppresses fungal mutualists in America but not in its native Europe. Ecology 89(4):1043-1055.
CALLAWAY RM, RIDENOUR WM. 2004. Novel weapons: invasive success and the evolution of increased competitive ability. Frontiers in Ecology and the Environment 2(8):436-443.
CARLSON AM, GORCHOV, DL. 2004. Effects of herbicide on the invasive biennial Alliaria petiolata (garlic mustard) and initial responses of native plants in a southwestern Ohio forest. Restoration Ecology 12(4):559-567.
CAVERS PB, HEAGY MI, KOKRON RF. 1979. The biology of Canadian weeds,35: Alliaria petiolata (M.Bieb.) Cavara and Grande. Canadian Journal of Plant Sciences 59:217-229
CHAPIN FS, WALKER BH, HOBBS RJ, HOOPER DU, LAWTON JH, SALA OE, TILMAN D. 1997. Biotic control over the functioning of ecosystems. Science 277:500-504.
CINCINNATI PARK BOARD. 2008. Park Board Web Site. http://www.cincinnati- oh.gov/cityparks/pages/-3036-/. Accessed 2008 July 3.
22 CIPOLLINI D, GRUNER B. 2007. Cyanide in the chemical arsenal of garlic mustard Alliaria petiolata. Journal of Chemical Ecology 33:85-94.
CIPOLLINI D, STEVENSON R, CIPOLLNI K. 2008. Contrasting effects of allelochemicals from two invasive plants on the performance of a non-mycorhizal plant. International Journal Plant Science 169(3):371-375.
COURTY PE, BREDA N, GARBAYE J. 2007. Relation between oak tree phenology and the secretion of organic matter degrading enzymes by Lactarius quietus ectomycorrhizas before and during bud break. Soil Biology and Biochemistry 39:1655-1663.
CULLINGS KW, VOGLER DR, PARKER VT, MAKHIJA S. 2001. Defoliation effects on the ectomycorrhizal community of a mixed Pinus contorta/Picea engelmannii stand in Yellowstone Park. Oecologia 127:433-539
CRUDEN RW, MCCLAIN AM, SHRIVASTAVA G. 1996. Pollination biology and breeding system of Alliaria petiolata (Brassicaceae). Bulletin of the Torrey Botanical Club 123:273-280.
DAUGHTRIDGE AT, PALLARDY SG, GARRETT HG, SANDER IL. 1986. Growth analysis of mycorrhizal and non-mycorrhizal black oak (Quercus velutina Lam.) seedlings. New Phytologist 103:473-480.
DALGAARD P. 2002. Introductory statistics with R. New York: Springer. 267 pgs.
DICKIE IA, KOIDE RT, STEINER KC. 2002. Influences of established trees on mycorrhizas, nutrition, and growth of Quercus rubra seedlings. Ecological Monographs 72:505-521.
DICKIE IA, REICH PB. 2005. Ectomycorrhizal fungal communities at forest edges. Journal of Ecology 93:244-255.
DICKIE IA, SCHNIZTER SA, REICH PB, HOBBIE SE. 2005. Spatially disjunct effects of co- occurring competition and facilitation. Ecology Letters 8:1191-1200.
EGERTON-WARBURTON LM, ALLEN EB. 2000. Shifts in arbuscular mycorrhizal communities along an anthropogenic nitrogen deposition gradient. Ecological Applications 10(2):484- 496.
EHRENFELD JG, KOURTEV P, HUANG W. 2001. Changes in soils fluctuations following invasions of exotic understory plants in deciduous forests. Ecological Applications 11(5):1287- 1300.
EHRENFELD JG. 2003. Effects of exotic plant invasions on soil nutrient cycling processes. Ecosystems 6:503-523.
ERLAND S, HENRION B, MARTIN F, GLOVER LA, ALEXANDER IJ. 1994. Identification of the ectomycorrhizal basidiomycete Tylosporsa fibrillose Donk by RFLP analysus of the
23 PCR-amplified ITS and IGS regions of ribosomal DNA. New Phytologist 126(3):525- 532.
FAHEY JW, ZALCMANN AT, TALALAY P. 2001. The chemical diversity and distribution of glucosinolates and isothiocyanates among plants. Phytochemistry 56:5-51.
FREELAND WJ, JANZEN DH. 1974. Strategies in herbivory by mammals: The role of plant secondary compounds. American Naturalist 108:269-289.
FROST CJ, HUNTER MD. 2004. Insect canopy herbivory and frass deposition affect soil nutrient dynamics and export in oak mescocosms. Ecology 85(12):3335-3347.
GARCIA MO, OVASAPYAN T, GREAS M, TRESEDER KK. 2008. Mycorrhizal dynamics under elevated CO2 and nitrogen fertilization in a warm temperate forest. Plant and Soil 303:301-310.
GARDES M, BRUNS TD. 1993. ITS primers with enhanced specificity for basidiomycetes- application the identification of mycorrhizae and rusts. Molecular Ecology 2:113-118.
GEBHARDT S, NEUBERT K, WOLLECKE J, MUNZENBERGER B, HUTTL RF. Ectomycorrhiza communities of red oak (Quercus rubra L.) of different age in the Lusatian lignite mining district, East Germany. Mycorrhiza 17:279-290.
GOODMAN RM, MATTSON GA. 1980. Low field temperatures optimum for field germination of northern red oak. Tree Planters’ Notes 32-34.
GOODMAN DM, DURALL DM, TROFYMOW JA. 1998. Describing ectomycorrhizae. In: Goodman DM, Durall DM, Trofymow JA, Berch SM, Eds. A manual of concise descriptions of North American ectomycorrhizae. Sidney BC: Mycologue Publications. p 3A1-3A5.
GORDON DR. 1998. Effects of invasive, non-indigenous plant species on ecosystem processes: lessons from Florida. Ecological Applications 8(4):975-989.
GRIBKO LS, SCHULER TM, FORD WM. 2002. Biotic and abiotic mechanisms in the establishment of northern red oak seedlings: a review. Gen. Tech. Rep. NE-295. Newton Square, PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 18 pgs.
GUPTA V, SATYANARAYANA T. 2000. Molecular genetics of ectomycorrhizal fungi. In: Mukerji KG, Chamola BP, Singh J. eds. Mycorrhizal Biology. New York: Kluwer Academic/ Plenum Publishers p.119-134.
GUPTA V, SATYANARAYANA T, GARG S. 2000. General aspects of mycorrhiza. In: Mukerji KG, Chamola BP, Singh J. eds. Mycorrhizal Biology. New York: Kluwer Academic/ Plenum Publishers p.29-44.
24 HOCHSTEDLER WW, SLAUGHTER BS, GORCHOV DL, SAUNDERS LP, STEVENS MHH. 2007. Forest floor plant community response to experimental control of the invasive biennial, Alliaria petiolata (garlic mustard). Journal of the Torrey Botanical Society. 134(2)155:165.
ISHIDA TA, NARA K, HOGETSU T. 2007. Host effects on ectomycorrhizal communities: insight from eight host species in mixed conifer-broadleaf forests. New Phytologist 174:430-440.
JONES JB. 2001. Laboratory Guide for Conducting Soil Tests and Plant Analysis. Boca Raton: CRC Press LLC 363 pp.
KOIDE RT, SHUMWAY DL, XU B, SHARDA JN. 2007. On temporal partitioning of a community of ectomycorrhizal fungi. New Pytologist 174:420-429.
KÕLJALG U, LARSSON K-H, ABARENKOV K, NILSSON RH, ALEXANDER IJ, EBERHARDT U, ERLAND S, HOILAND K, KJOLLER R, LARSSON E, PENNANEN T, SEN R, TAYLOR AFS, TEDERSOO L, VRALSTAD T, URSING BM. 2005. UNITE: a database providing web-based methods for the molecular identification of ectomycorrhizal fungi. New Phytologist 166:1063-1068
LILLESKOV EA, BRUNS TD, HORTON TR, TAYLOR DL, GROGAN P. 2004. Detection of forest satnd-level spatial structure in ectomycorrhizal fungal communities. FEMS Microbiology Ecology 49:319-322.
LOBUGLIO KF. 1999. Cenoccum. In: Cairney JWG, Chambers SM, Eds. Ectomycorrhizal fungi: key genera in profile. Berlin: Springer. p.287-305.
LORIMER CG, CHAPMAN JW, LAMBERT WD. 1994. Tall understory vegetation as a factor in the poor development of oak seedlings beneath mature stands. Journal of Ecology. 82:227- 237.
MACK RN, SIMBERLOFF D, LONSDALE WM, EVANS H, CLOUT M, AND BAZZAZ FA. 2000. Biotic invasions: causes, epidemiology, global consequences, and control. Ecological Applications 10(3):689-710.
MALLOCH DW, PIROZYNSKI KA, RAVEN PH. 1980. Ecological and evolutionary significance of mycorrhizal symbioses in vascular plants (a review). Ecology 77(4):2113-2118.
MAYTON HS, OLIVIER C, VAUGHN SF, LORIA R. 1996. Correlation of fungicidal activity of Brassica species with allyl isothiocyanate production in macerated leaf tissue. Phytopathology 86(3):267-271.
MCCARTHY BC. 1997. Response of a forest understory community to experimental removal of an invasive nonindigenous plant (Alliaria petiolata, Brassicaceae). Pages 117-130 in J. O. Luken and J. W. Thieret, editors. Assessment and Management of Plant Invasions. Springer, New York.
25 MEDLEY KE. 1996. Dieback in the native shrub, Lindera benzoin: A subtle effect of forest fragmentation. Ohio Journal of Science 96(4/5):76-80.
MEDLEY KE, KRISKO B. 2007. Physical site conditions and land use history as factors influencing the conservation of regrowth in a southwestern Ohio Nature Reserve. Natural Areas Journal 27:31-40.
MEEKINS JF, MCCARTHY BC. 1999. Competitive ability of Alliaria petiolata (garlic mustard, Brassicaceae), an invasive, nonindigenous forest herb. International Journal of Plant Science 160(4):743-752.
------. 2001. Effect of environmental variation on the invasive success of a non-indigenous forest herb. Ecological Application 11:1336-1348.
------. 2002. Effect of population density on the demography of an invasive plant (Alliaria petiolata, Brassicaceae) population in a southeastern Ohio forest. American Midland Naturalist 147:256-278.
MELLO A, GARNERO L, BONFANTE P. 1999. Specific PCR-primers as a reliable tool for the detection of white truffles in mycorrhizal roots. New Phytologist 141:511-516.
MOONEY HA, CLELAND EE. 2001. The evolutionary impact of invasive species. Proceedings of the National Academy of Science 98(10):5446-5451.
MOSCA E, MONTECCHIO L, SELLA L, GARBAYE J. 2007. Short-term effect of removing tree competition on the ectomycorrhizal status of a declining pedunculate oak forest (Quercus robur L.). Forest Ecology and Management 244:129-140.
MOSER MA, PETERSON CA, D’ALLURA JA, SOUTHWORTH D. 2005. Comparison of ectomycorrhizas of Quercus garryana (Fagaceae) on serpentine and non-serpentine soils in southwestern Oregon. American Journal of Botany 92(2):224-230.
MUELLER RC, STHULTZ CM, MARTINEZ T, GEHRING CA, WHITHAM TG. The relationship between stem-galling wasps and mycorrhizal colonization of Quercus turbinella. Canadian Journal of Botany 83(10):1349-1353.
NILSSON RH, LARSSON K-H, URSING BM. 2004. galaxie - CGI scripts for sequence identification through automated phylogenetic analysis. Bioinformatics 20:1447-1452
[NRCS] NATURAL RESOURCES CONSERVATION SERVICE. 2007. NCSS Web Soil Survey. Retrieved May 25, 2008 from Natural Resources Conservation Service, US Department of Agriculture.
NUZZO VA. 1991. Experimental control of garlic mustard [Alliaria petiolata (Bieb.) Cavara & Grande] in northern Illinois using fire, herbicide, and cutting. Natural Areas Journal 11:158-167.
26
------. 1993a. Distribution and spread of the invasive biennial Alliaria petiolata (garlic mustard) in North America. Pages 137-145 in B. N. McKnight, editor. Biological Pollution: the control and impact of invasive exotic species.Indianapolis: Indiana Academy of Science.
------. 1993b.Current and historic distribution of garlic mustard (Alliaria petiolata) in Illinois. Michigan Botanist 32:21-33.
------. 1999. Invasion pattern of the herb garlic mustard (Alliaria petiolata) in high quality forests. Biological Invasions 1:169-179.
------. 2000. Element stewardship abstract for Alliaria petiolata (Alliaria officinalis), garlic mustard. Arlington VA: The Nature Conservancy.
O’DELL TE, MASSICOTTE HB, KERNAGHAN G. 1998. How to store, clean, and photograph ectomycorrhizae and prepare voucher material. In: Goodman DM, Durall DM, Trofymow JA, Berch SM, Eds. A manual of concise descriptions of North American ectomycorrhizae. Sidney BC: Mycologue Publications. p.2.1-2.3.
OKSANEN J. 2008. Multivariate analysis on ecological comminutes in R: Vegan tutorial. Vegan: R functions for vegetation ecologists. R Documentation Page
ORR SP, RUDGERS JA, CLAY K. 2005. Invasive plants can inhibit native tree seedlings: testing potential allelopathic mechanisms. Plant Ecology DOI 10.1007/s11258-005-5698-6.
PETERSON RL, MASSICOTTE HB, MELVILLE LH. 2004. Mycorrhizas: Anatomy and Cell Biology. Ottawa: CABI Publishing, 173pp.
PIMENTEL D, ZUNIGA R, MORRISON D. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecological Economics 52(3):273-288.
PRATI D, BOSSDORF O. 2004. Allelopathic inhibition of germination by Alliaria petiolata (Brassicaceae). American Journal of Botany 91(2):285-288.
PRITSCH K, BOYLE H, MUNCH JC, BUSCOT F. 1997. Characterization and identification of black alder ectomycorrhizas by PCR/RFLP analysis of the rDNA internal transcribed spacer (ITS). New Phytologist 137(2):357-369.
R DEVELOPMENT CORE TEAM. 2004. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-00-3, URL http://www.R-project.org.
27 REINHART KO, CALLAWAY RM. 2006. Tansley review: Soil biota and invasive plants. New Phytologist 170:445-457.
RIESKE LK. 2001. Influence of symbiotic fungal colonization on oak seedling growth and suitability for insect herbivory. Environmental Entomology 30:849-854
ROBERTS KJ, ANDERSON RC. 2001. Effect of garlic mustard [Alliaria petiolata (Beib. Cavara & Grande)] extracts on plants and arbuscular mycorrhizal (AM) fungi. American Midland Naturalist 146:146-152.
RODGERS VL, STINSON KA, FINZI AC. 2008. Ready or not, garlic mustard is moving in: Alliaria petiolata as a member of eastern North American Forests. Bioscience 58(5):426-436.
ROSSOW LJ, BRYANT JP, KIELLAND K. 1997. Effects of above-ground browsing by mammals on mycorrhizal infection in an early successional taiga ecosystem. Oecologia 110:94-98.
SANDER IL. 1990. Quercus rubra. Pp 727-733 in Burns, R.M. and B.H. Honkala, Silvics of North America. v. 2. Hardwoods. Washington, D.C. : U.S. Dept. of Agriculture, Forest Service : Agriculture handbook no. 654.
SAIKKONEN K, AHONEN-JONNARTH U, MARKKOLA AM, HELANDER M, TUOMI J, ROITTO M, RANTA H. 1999. Defoliation and mycorrhizal symbiosis: a functional balance between carbon sources and below-ground sinks. Ecology Letters 2:19-26.
SLAUGHTER BS, HOCHSTEDLER WW, GORCHOV DL, CARLSON AM. 2007. Response of Alliaria petiolata (garlic mustard) to five years of fall herbicide application in a southern Ohio deciduous forest. Journal of the Torrey Botanical Society 134(1):18-26.
nd SMITH SE, READ DJ. 1997. Mycorrhizal Symbiosis. 2 ed. New York: Academic Press.
STINSON KA, CAMPBELL SA, POWELL JR, WOLFE BE, CALLAWAY RM, ET AL. 2006. Invasive plant suppresses the growth of native tree seedlings by disrupting belowground mutualisms. PLoS Biology 4(5): e140. DOI: 10.1371/journal.pbio.0040140.
STINSON K, KAUFMAN S, DURBIN L, LOWENSTEIN F. 2007. Impacts of garlic mustard invasion on a forest understory community. Northeastern Naturalist 14(1):73-88.
SWATY RL, DECKERT RJ, WHITAM TG, GEHRING CA. 2004. Ectomycorrhizal abundance and community composition shifts with drought: predications from tree rings. Ecology 85(4):1072-1084.
TAYLOR ASF, ALEXANDER I. 2005. The ectomycorrhizal symbiosis: life in the real world. Mycologist 19(3):102-112.
28 TAYLOR LD, BRUNS TD. 1997. Independent, specialized invasions of ectomycorrhizal mutualism by two nonphotosynthetic orchids. Proceedings of the National Academy of Science 94:4510-4515.
[TNC] THE NATURE CONSERVANCY. 2008. Land Protection Statistics. http://www.nature.org/wherewework/northamerica/states/ohio/preserves/art11493.html
TRESEDER KK. 2004. A meta-analysis of mycorrhizal responses to nitrogen, phosphorus, and atmospheric CO2 in field studies. New Phytologist 164: 347-355.
VAUGHN SF, BERHOW MA. 1999. Allelochemicals isolated from tissues of the invasive weed garlic mustard (Alliaria petiolata). Journal of Chemical Ecology 25(11):2495-2504.
VISSER S. 1995. Ectomycorrhizal fungal succession in jack pine stands following wildfire. New Phytol. 129:389-401.
VITOUSEK PM, MOONEY HA, LUBCHENCO J, MELILO JM. 1997. Human domination of earth’s ecosystems. Science 277:494-499.
WALKER JF, MILLER OK, HORTON JL. 2005. Hyperdiversity of ectomycorrhizal fungus assemblages on oak seedlings in mixed forests in the southern Appalachian Mountains. Molecular Ecology 14:829-828.
WALKER JF, MILLER OK, HORTON JL. 2008. Seasonal dynamics of ectomycorrhizal fungus assemblages on oak seedlings in the southeastern Appalachian Mountains. Mycorrhiza 18:123-132.
WELK E, SCHUBERT K, HOFFMAN MH. 2002. Present and potential distribution of invasive garlic mustard (Alliaria petiolata) in North America. Diversity and Distributions 8:219-233.
WHIGHAM DF. 2004. Ecology of woodland herbs in temperate deciduous forests. Annual Review of Ecology, Evolution, and Systematics 35:583-621.
WHITE TJ, BRUNS T, LEE S, TAYLOR J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR protocols. A guide to methods and applications. San Diego: Academic press, 315-322.
WILCOVE DS., ROTHSTEIN D, DUBOW J, PHILLIPS A, AND LOSOS E. 1998. Quantifying threats to imperiled species in the United States. BioScience 48(8):607-615.
YOUNG JA, YOUNG CG. 1992. Seeds of woody plants in North America, rev. and enl. ed. Portland, Or., Dioscorides Press, vii, 407 pp.
ZAR, J.H. 1999. Biostatistical analysis. 4th edition. Prentice Hall, Upper Saddle River, NJ.
29 Table 1. Contingency table reporting the number of planted Q. rubra acorns that germinated at Tanglewood (high Alliaria petiolata density) and Reinhart (no A. petiolata). Fisher’s exact test found no differences in germination success between sites (P = 0.31).
Germinated Ungerminated Total
Tanglewood 18 31 49
Reinhart 24 25 49
Total 42 56 98
Table 2. Contingency table reporting the number of seedlings from a) seed and b). bare root planting living at the fall 2007 harvest at Tanglewood (high Alliaria petiolata density) and Reinhart (no A. petiolata). Fisher’s exact tests found no difference of seedling survival between sites (P = 0.15 and P = 1.0 respectively). a. Alive Dead Total Seed Derived Tanglewood 14 4 18 Reinhart 23 1 24 Total 37 5 42 b. Alive Dead Total Bare Root Tanglewood 14 17 31
Reinhart 12 13 25 Total 26 30 56
30 Table 3. Contingency table reporting the number of seedlings (both bare root and seed derived) living at the spring 2008 harvest at Tanglewood (high Alliaria petiolata density) and Reinhart (no A. petiolata). Fisher’s exact test found over-winter survival was lower at Tanglewood (P = 0.03).
Alive Dead Total
Tanglewood 9 6 15
Reinhart 17 1 18
Total 26 7 33
Table 4. Three-Way ANOVA of the proportion of colonized root tips for Quercus rubra seedlings planted in Tanglewood (high Alliaria petiolata density) and Reinhart (no A. petiolata). Seedlings were harvested in Aug 2007 and May 2008. Seedling type refers to field-germinated and bareroot-derived seedlings. Data were arcsine square root transformed to meet assumptions of variance homogeneity. Non-significant three-way and two-way interactions were removed from model. Bold indicates significance at α = 0.05
Source Term df MS FP
Site 1 1.3505 45.72 <0.0001 Harvest Date 1 0.0009 0.03 0.86 Seedling Type 1 0.1324 4.48 0.04 Error 52 0.0295 Corrected Total 55
Table 5. Weighted one-way ANOVA comparison of the proportion of colonized root tips for Quercus rubra seedlings naturally growing in Kramer Woods (near Alliaria petiolata) and Reinhart (no-A. petiolata)
Source Term df MS FP
Site 1 14.18 3.19 0.082 Error 37 4.45 Corrected Total 38
31 Table 6. Ectomycorrhizal morphotypes found on Quercus rubra seedlings from the three forest stands studied. Names represent the closest match to ITS sequences catalogued in the UNITE database as determined by phylogenetic alignments using maximum parsimony method. I also show which genera have been recorded on various Quercus species and corresponding references. It should be noted that, in the case of types A and H, that these species are often decomposer fungi and may be contaminates amplified in place of the intended ECM. Because, however, the sequences were deposited in a database specific to ectomycorrhizal fungi, they possibly form these mutualistic relationships on some species. Literature on their relationship to Quercus, however, could not be found. In addition, type A also showed significant alignments, with only slightly lower E values, to a variety of ECM species (Appendix I). Cenoccocum geophilum was not successfully sequenced but distinctive morphology allowed identification of this morphotype.
Morph Documented on/near ID Closest galaxieBLAST match UNITE Accession# Score E Quercus A Coprinus comatus UDB001537 337 2.00E-93 Walker et al. 2005, B Russula sp. UDB001631 1019 0 Gebhart et al.2007 Walker et al. 2005, C Cenoccocum geophilum Gebhart et al.2007 Walker et al. 2005, D Lactarius sp. UDB000861 831 0 Gebhart et al.2007 Walker et al. 2005, E Russula odorata UDB001631 1035 0 Ishida et al.2007 F Inocybe sp.1 UDB000617 335 8.00E-93 Mosca et al. 2007 G Byssocorticium atrovirens UDB000075 1166 0 Ishida et al.2007 H Mycena purpureofusca UDB001610 938 0 I Unidentified 1 J Sebacina sp.1 UDB000773 819 0 Ishida et al.2007 K Unidentified 2 L Unidentified 3 M Tomentella ellisii UDB000219 936 0 Mosca et al. 2007 Walker et al. 2005, N Russula subrubens UDB002453 1015 0 Gebhart et al.2007 O Sebacina sp. 2 UDB000773 819 0 Ishida et al.2007 P Inocybe sp. 2 UDB000617 335 8.00E-93 Mosca et al. 2007 Q Tomentella stuposa UDB002429 1213 0 Ishida et al.2007
Table 7. Morphotype richness and Shannon-Weiner diversity index of ECM fungi found colonizing Quercus rubra planted Tanglewood (high Alliaria petiolata) and Reinhart (no A. petiolata). Test statistics are two-sample T-tests (T) or Wilcoxon (W) tests. Data are from two harvests (fall of 2007 and spring of 2008). Bold indicates significance at α = 0.05.
Harvest 1 Harvest 2 Harvest 1 Harvest 2 Test Test Parameter Tanglewood Reinhart Tanglewood Reinhart statistic P statistic P
Morphotype Richness (S) 1.46 4.24 1.44 4.82 T = -4.60 <0.0001 T = -5.87 <0.0001
Shannon's Diversity Index (H′) 0.31 1.00 0.30 1.11 W = 35 0.0010 W = 15 0.0009
Table 8. Morphotype richness and Shannon-Weiner diversity index of ECM fungi found colonizing Quercus rubra planted at Kramer (moderate Alliaria petiolata) and Reinhart (no A. petiolata). Test statistics are two-sample T-tests (T) or Wilcoxon (W) tests.
Moderate No Parameter A. petiolata A. petiolata T ,W P
Morphotype Richness (S) 3.68 3.40 W = 169.5 0.562
Shannon's Diversity Index (H′) 0.865 0.798 T =-0.567 0.574
Table 9. Effect of Alliaria petiolata density (high vs. no) on the relative abundance of ECM fungal morphotypes found growing on root tips of Quercus rubra in Reinhart Preserve ( no A. petiolata) and Tanglewood Preserve (high A. petiolata density) determined by multiple response permutation procedure (MRPP). The test statistic A is the chance corrected within-group agreement describing the amount of similarity within a group. P values represent the chance that expected delta (δ) values are less than observed delta values; expected delta values were obtained from 10,000 random permutations. Bold indicates significance at α = 0.05 and * indicates marginal significance 0.10 > P < 0.05.
Harvest Date δ Observed δ Expected AP
Fall 2007 0.7954 0.8126 0.0212 0.076*
Spring 2008 0.7642 0.8070 0.0530 0.002
Table 10. Effect of Alliaria petiolata density (moderate vs. no) on the relative abundance of ECM fungal morphotypes found growing on root tips of Quercus rubra in temperate deciduous forests as determined by multiple response permutation procedure (MRPP). The test statistic A is the chance corrected within-group agreement describing the amount of similarity within a group. P values represent the chance that expected delta (δ) values are less than observed delta values; expected delta values were obtained from 10,000 random permutations. No significant differences were found in composition between these two sites.
δ Observed δ Expected AP
0.7364 0.7450 0.0116 0.105
34 Figure 1. Seedling survival at Tanglewood (high Alliaria petiolata density) (red) and Reinhart (no-A. petiolata) (cyan) from time of planting (5-24-07 & 5-25-07) to harvest 1 (9-11-07 & 9-25- 07). Solid lines represent number of live seedlings derived from field-germinated seed and dashed lines represented bareroot planted seedlings.
35 Figure 2: Proportional colonization by mycorrhizal fungi on root tips of Quercus rubra, (harvested fall 2007), planted in forest sites with high (Tanglewood Preserve) and no (Reinhart Preserve) Alliaria petiolata. Crosses represent the mean of the distribution and the error bars are extended to 2x SEM
36 Figure 3: Proportional colonization by mycorrhizal fungi on root tips of Quercus rubra, (harvested spring 2008), planted in forest sites with high (Tanglewood Preserve) and no (Reinhart Preserve) Alliaria petiolata. Crosses represent the mean of the distribution and the error bars are extended to 2x SEM.
37 Figure 4. Proportional colonization by mycorrhizal fungi on root tips of Quercus rubra derived from field germinated seed and bare root planted seedlings. Crosses represent the mean of the distribution and the error bars are extended to 2x SEM
38 Figure 5: Proportional colonization by mycorrhizal fungi on root tips of Quercus rubra naturally growing in forest sites with moderate (Kramer Woods) and with no Alliaria petiolata (Reinhart Preserve). Crosses represent the mean of the distribution and the error bars are extended to 2x SEM
39 Figure 6. Rank abundance curve of ECM morphotypes growing on root tips of Quercus rubra in high (H) density and no (N) Alliaria petiolata sites (Harvest1). Open triangles (cyan) represent abundances from the no A. petiolata site and open circles (red) are those from high A. petiolata sites. A wider curve indicates higher species richness and evenness increases as curve becomes more horizontal.
40
Figure 7. Rank abundance curve of ECM morphotypes growing on root tips of Quercus rubra in high (H) density and no (N) Alliaria petiolata sites (Harvest 2). Open triangles (cyan) represent abundances from the no A. petiolata site and open circles (red) are those from high A. petiolata sites. A wider curve indicates higher species richness and evenness increases as curve becomes more horizontal.
41
Figure 8. Rank abundance curve of ECM morphotypes growing on root tips of Quercus rubra in moderate (M) density and no (N) Alliaria petiolata sites. Open triangles (cyan) represent abundances from the moderate A. petiolata site and open circles (red) are those from the no A. petiolata site. A wider curve indicates higher species richness and evenness increases as curve becomes more horizontal.
N
42 Figure 9. Non-metric multidimensional scaling (NMDS) ordination of ECM abundance at high density A. petiolata (H, dashed-red lines) and no-A. petiolata (N, solid-blue lines) sites for the fall 2007 harvest. Polygons represent range of data and ellipses represent 95% confidence intervals drawn around the data centroids. For this harvest the centroids overlapped only slightly showing these sites do differ slightly in ECM composition; the range of data show considerable overlap but this is influenced by a single point.
43 Figure 10. Non-metric multidimensional scaling (NMDS) ordination of ECM abundance at high density A. petiolata (H, dashed-red lines) and no-A. petiolata (N, solid-blue lines) sites for the spring 2008 harvest. Polygons represent range of data and ellipses represent 95% confidence intervals drawn around the data centroids. For this harvest the data range had some similarities but lack of centroid overlap indicates differences in ECM composition.
44 Figure 11. Non-metric multidimensional scaling (NMDS) ordination of ECM abundance at moderate density A. petiolata (M, dashed-red lines) and no-A. petiolata (N, solid-blue lines) sites. Polygons represent range of data and ellipses represent 95% confidence intervals drawn around the data centroids. These communities have considerable overlap and thus, are not different.
45 Appendix A. 1938 aerial photograph of Tanglewood Preserve showing that the site was allowed to return to forest from this time on. An earlier photograph from 1932 shows this site was likely in an “old field” successional stage in at that, earlier photos or an exact time of field abandonment could not be obtained. This site has remained in a forested state and has been managed as a preserve until present day (Photograph courtesy of Cincinnati Park Board, Land Management Section). The green border represents present day parcel boundaries and the blue box (not drawn to scale) shows approximate location of the 15 m x 15 m research plot.
46 Appendix B- Tree composition for (a) Reinhart Preserve, (b) Kramer Woods (A. Maye and D. Gorchov, unpublished), and (c) Bradford-Felter Tanglewood Preserve (Castellano, unpublished) measured using the point-quarter method. Relative basal area (RBA) was calculated as the total basal area of each species / total basal area of all species, relative density (RD) as the number of each species / total number of trees, relative frequency (RF) as the number of quadrants a species was found in / total number of quadrats sampled, and importance value (IV) as [(RBA*100) + (RD*100) + (RF*100)]/3. Names marked with (*) are species forming ECM associations. Data were collected near study areas and does not represent composition of entire stands. a. Tree Composition: Reinhart Preserve
Tree Species RBA RD RF IV Acer saccharum 0.18 0.206 0.157 18.055 Fraxinus sp. 0.11 0.111 0.098 10.510 Prunus serotina 0.11 0.095 0.098 10.172 Quercus rubra* 0.11 0.095 0.078 9.460 Liriodendron tulipifera 0.07 0.079 0.098 8.169 Quercus muhlenbergii* 0.04 0.063 0.078 6.158 Platanus occidentalis 0.09 0.032 0.020 4.681 Carya tomentosa* 0.03 0.048 0.059 4.551 Juglans nigra 0.05 0.032 0.039 4.062 Celtis occidentalis 0.03 0.032 0.039 3.508 Aesculus glabra 0.02 0.032 0.039 3.174 Quercus coccinea* 0.05 0.016 0.020 2.776 Carya cordiformis* 0.03 0.032 0.020 2.661 Cornus florida 0.01 0.032 0.039 2.607 Fagus grandifolia* 0.03 0.016 0.020 2.144 Maclura 0.03 0.016 0.020 2.096 Ulmus rubra 0.00 0.016 0.020 1.336 Tilia americana 0.00 0.016 0.020 1.318 Cercis canadensis 0.00 0.016 0.020 1.284 Carpinus caroliniana* 0.00 0.016 0.020 1.277 ECM Species Total: 0.29 b. Tree Composition: Kramer Woods
Tree Species RBA RD RF IV Acer saccharum 0.30 0.516 0.341 38.705 Liriodendron tulipifera 0.22 0.109 0.114 14.891 Fraxinus sp. 0.16 0.109 0.159 14.347 Fagus grandifolia* 0.12 0.125 0.182 14.220 Quercus rubra* 0.13 0.031 0.045 6.976 Juglans nigra 0.04 0.047 0.068 5.306 Prunus serotina 0.01 0.016 0.023 1.496
47 Celtis occidentalis 0.00 0.016 0.023 1.364 Ostrya virginica* 0.00 0.016 0.023 1.355 Carya cordiformis* 0.00 0.016 0.023 1.339 ECM Species Total: 0.25 c. Tree Composition: Tanglewood Preserve
Tree Species RBA RD RF IV Acer saccharum 0.40 0.725 0.500 54.124 Quercus rubra* 0.33 0.075 0.125 17.519 Fraxinus americana 0.13 0.075 0.125 10.850 Acer nigrum 0.06 0.038 0.075 5.828 Fraxinus sp. 0.01 0.025 0.050 2.830 Ostrya virginiana* 0.01 0.025 0.050 2.828 Quercus alba* 0.04 0.013 0.025 2.427 Quercus muhlenbergii* 0.02 0.013 0.025 1.827 Ulmus rubra 0.02 0.013 0.025 1.767 ECM Species Total: 0.40
48 Appendix C: Aerial photograph of Reinhart Preserve. Square shows description of soil series (see key) in the general vicinity of research site. The star represents the approximate location of 15 m x 15 m research plot. Blue line represents Harkers Run creek. Source: NRCS 2007.
50 Appendix D: Aerial photograph of Kramer Woods and square shows description of soil series (see key) in the general vicinity of research area where seedlings were collected. Source: NRCS 2007.
51
52 Appendix E: Aerial photograph of Bradford-Felter Tanglewood Preserve. Square shows description of soil series (see key) in the general vicinity of research site. The star represents the approximate location of 15 m x 15 m research plot. Source: NRCS 2007.
53
54 Appendix F: Estimation of Alliaria petiolata density at Tanglewood Preserve
Cover class % Cover Cover class % Cover
1 0-20 4 61-80
2 21-40 5 81-100 3 41-60
A 50 cm radius plastic ring with was centered on each planting location, and the percentage of ground covered by Alliaria petiolata within that radius was visually estimated and assigned to a cover class (above). In addition, the distance to the nearest first year (DGm1) and nearest second year (DGm2) A. petiolata plants was measured and the total number of flowering stalks was counted within the 50 cm radius. Density for second year plants (6.6 / m2) was 2 determined as the average number of Gm2 in the 50cm radius; density of first year plants (33.4 / m ) was calculated by closest individual method (D=1/(2r)2 where D is estimated density and r is average point to plant distance) using
DGm1. Data collection occurred 6-5-07.
ID DGm2 (cm) DGm1 (cm) Cover Class #Gm2 50 cm(radius) A1 42 6 3 1 A2 33 7 2 7 A3 40 10 2 1 A4 18 7 3 23 A5 37 7 2 5 A6 10 9 1 12 A7 24 8 2 3 B1 12 10 3 10 B2 26 1 5 2 B3 8 6 3 9 B4 17 10 3 8 B5 40 7 3 4 B6 2 5 5 9 B7 30 8 5 2 C1 20 10 2 2 C2 9 20 3 9 C3 27 7 2 7 C4 58 15 2 0 C5 11 3 4 20 C6 9 2 4 33 C7 30 12 1 3 D1 73 5 2 0 D2 47 7 3 2 D3 64 12 1 0 D4 18 6 1 7 D5 22 9 3 7 D6 21 14 2 2 D7 31 9 4 1 E1 56 11 1 0 E2 104 10 1 0 E3 116 4 1 0 E4 70 6 4 0 E5 35 8 3 5 E6 48 13 3 1 E7 41 11 5 1 F1 23 7 2 6 F2 42 9 1 1 F3 50 9 2 1 F4 81 6 5 0 F5 20 7 3 9 F6 49 9 4 1 F7 20 2 2 5 G1 20 25 0 2 G2 32 12 2 4 G3 15 10 2 7 G4 33 15 2 1 G5 94 9 3 0 G6 36 4 3 7 G7 13 5 1 16 Mean 36.27 8.65 2.57 5.22
55 Appendix G- Nutrient levels, nitrogen, phosphorus, and other soil properties at Reinhart Preserve and Bradford-Felter Tanglewood Preserve in the spring of 2008.
56 Appendix H. Fungal morphotypes growing on Quercus rubra root tips from three forested sites with low (Reinhart), moderate (Kramer), and high (Tanglewood) Alliaria petiolata density. Proportional data is provided and represents the proportion of each morphotype out of all ECM root tips for each site. For Reinhart (exp.) and Tanglewood, proportion is the sum of both harvests; data in parentheses () are from Harvest 1and data in brackets [] Harvest 2 of the experiment only. Kramer and Reinhart (c) are from comparative study only.
ID Species ID Reinhart (exp) Reinhart (c) Kramer Tanglewood Morphotype A Coprinus comatus 0.061 (0.026) [0.078] 0.08 0.077 0.042 (0.000) [0.120] Gold/Brown B Russula sp. 0.210 (0.247) [0.195] 0.19 0.376 0.026 (0.040) [0.000] White/Cream smooth C Cenoccocum geophilum 0.331 (0.356) [0.160] 0.47 0.245 0.092 (0.105) [0.068] Black w/ black emanating hyphae D Lactarius sp. 0.084 (0.036) [0.131] 0.08 0.034 0.092 (0.040) [0.188] Gold, smooth E Russula odorata 0.077 (0.114) [0.042] 0.07 0.142 0.105 (0.113) [0.090] Grey smooth F Inocybe sp.1 0.016 (0.030) [0.018] 0.00 Gray/Black G Byssocorticium atrovirens 0.017 (0.008) [0.001] 0.04 0.008 0.003 (0.004) [0.000] Silver/blue metallic shiny w/some hyphae H Mycena purpureofusca 0.017 (0.000) [0.000] 0.05 0.047 Black/Brown I Unidentified 1 0.001 (0.000) [0.000] 0.00 Glassy J Sebacina sp.1 0.00 0.001 Yellow/Gray K Unidentified 2 0.028 (0.000) [0.068] 0.02 0.068 White/Gray Fluff L Unidentified 3 0.011 (0.029) [0.006] 0.239 (0.239) [0.241] Black/gray-gold hyphae some branching M Tomentella ellisii 0.104 (0.047) [0.273] 0.284 (0.381) [0.105] White-Cream w/ Fluffy Hyphae/some branching N Russula subrubens 0.022 (0.056) [0.014] 0.026 (0.040) [0.000] Gray Hairy O Sebacina sp. 2 0.005 (0.000) [0.014] 0.092 (0.040) [0.188] Black- No hyphae P Inocybe sp. 2 0.015 (0.047) [0.000] Cream w/ hyphae-not fluffy, branching Q Tomentella stuposa 0.002 (0.006) [0.000] Brown wooly Total ECM count: 2135 (1050) [1085] 1184 1344 381 (248) [133]
Appendix I. Select color images of morphotypes described in Appendix H.
A B C
D E G
K L M
N O P
58 Appendix J. Identities obtained from multiple maximum parsimony phylogenetic alignments using the galaxieBLAST application of the UNITE database (Nilsson et al.2004, Koljalg et al. 2005) of multiple samples of each morphotype. In some cases multiple species are assigned to one morphotype indicting the likelihood of multiple species with similar morphology. Note some identities are based on one ITS sample only as many samples were not successfully amplified. Species identities are based on the closest match, E- value closest to 0, to database sequences.
Extraction # morph Sample Name UNITE Accession# and Identities Score E-value 4 a QR 14KW-2 UDB001537 Coprinus comatus 337 2.00E-93 6 a QR 44KW- 3 UDB000991 Geopyxis alpina 272 1.00E-73
1 b QR 12RP UDB000617 Inocybe lacera 335 8.00E-93 2 b QR 14KW-5 UDB002453 Russula subrubens 908 0 3 b QR 44KW-2 UDB001631 Russula odorata 1019 0 9 b QR 14KW- 6 UDB000617 Inocybe lacera 335 9.00E-93 15 b QR 3RP- 3 UDB000893 Russula mustelina 815 0 5a b QR 3RP UDB000893 Russula mustelina 783 0 5b b QR 3RP UDB000093 Elaphomyces muricatus 178 9.00E-46 5c b QR 3RP UDB002254 Sistotrema muscicola 178 9.00E-46
8 d QR 14RP- 1 UDB002413 Lactarius quietus 634 0 19 d QR 27RP- 1 UDB000861 Lactarius hepaticus 831 0 20 d QR 27RP- 2 UDB000379 Lactarius fulvissimus 823 0
18 e QR 16RP- 2 UDB001387 Xerocomus communis 817 0 21 e QR 5RP- 1 UDB001610 Mycena purpureofusca 242 6.00E-65 23 e QR 4KW- 2 UDB001782 Tricholoma virgatum 115 1.00E-26 25 e TW F7-1 UDB001631 Russula odorata 1035 0 40 e RP D6 - 1 UDB001631 Russula odorata 797 0
39 f RP B5 - 1 UDB000617 Inocybe lacera 335 8.00E-93
10 g QR 18RP- 1 UDB000075 Byssocorticium atrovirens 1156 0 12 g QR 18RP- 4 UDB000075 Byssocorticium atrovirens 1146 0 14 g QR 18RP- 3 UDB000075 Byssocorticium atrovirens 1162 0 13a g QR 18RP- 2 UDB000075 Byssocorticium atrovirens 1166 0
16 h QR 16RP- 3 UDB000991 Geopyxis alpina 248 1.00E-66 16 h QR 16RP- 3 UDB001610 Mycena purpureofusca 297 2.00E-81 22 h QR 9RP- 3 UDB001610 Mycena purpureofusca 938 0
37 j TW G3 - 2 UDB000773 Sebacina sp. 819 0 38 j TW G3 - 1 UDB000773 Sebacina sp. 811 0
30 m TW E5-1 UDB000219 Tomentella ellisii 936 0
32 n RP C1-2 UDB002453 Russula subrubens 1015 0 35 n RP A6-2 UDB000937 Entoloma sp. 387 e-108
26 o TW A7-1 UDB000773 Sebacina sp. 819 0
41 p RP F2 - 1 UDB000617 Inocybe lacera 335 8.00E-93
31 q RP G2-2 UDB002429 Tomentella stuposa 1213 0 34 q RP G2-1 UDB002429 Tomentella stuposa 1027 0
59