Janet A. Morrison research interests: plant-pathogen interactions
I am interested in the relative influence of environmental variation and host genetic variation in the infection of host individuals in natural plant populations. While pathogen response to host genetic variation has received substantial research attention, its interaction with environmental variation has been studied less. Until this question is addressed fully, we can not adequately understand the evolutionary role of disease in nature. I also am interested in the role that interactions with pathogens may have during invasions of non-native plant species. To date, I have done research on three different study systems:
1. The rush Juncus dichotomusand the smut fungus Cintractia junci.
This rush is a major component of open meadow communities in the Long Island pine barrens. Stands can be heavily infected with the smut fungus, which invades the reproductive tissue, rendering the plant sterile.
Ph.D. thesis: Ecological and genetic factors influencing a natural plant-pathogen interaction. 1994. State University of New York at Stony Brook. PDF
Morrison, J.A. 1996. Infection of Juncus dichotomus by the smut fungus Cintractia junci : an experimental field test of the effects of neighboring plants, environment, and host plant genotype. Journal of Ecology 84: 691-702.
Morrison, J.A. 1995. Multiple ecological and genetic factors in an experimental plant-pathogen system: soil water, plant clone, and environmental heterogeneity. Abstract. Supplement to the Bulletin of the Ecological Society of America 76(2):190.
Morrison, J.A. 1994. An indirect effect of interspecific plant competition on a plant-pathogen interaction. Abstract. Supplement to the Bulletin of the Ecological Society of America 75(2):160.
Morrison, J.A. 1993. Infection of Juncus dichotomus by the smut fungus Cintractia junci: a field test of the effects of pathogen density, host density, and host genotype. Abstract. Supplement to the Bulletin of the Ecological Society of America 74(2):367.
Morrison, J.A. 1992. Microsite dependence of environmental and genetic effects in a natural plant-pathogen interaction. Abstract. Supplement to the Bulletin of the Ecological Society of America 73(2):278.
Morrison, J.A. 1991. Causal factors of spatial pattern in a plant-pathogen relationship. Abstract. Supplement to the Bulletin of the Ecological Society of America 72(2):200.
2. Flowering dogwood, Cornus florida, and dogwood anthracnose, Discula destructiva.
Dogwood anthracnose is a fungal disease that appeared about 30 years ago near New York City. It initially caused high mortality rates, but not all trees succumbed, so both trees and the fungus are still present in the woodlands near New York City where I have studied it. I used this system to look at the effect of the urban environment on decline and disease in forest trees, while I was a researcher at The New York Botanical Garden in Bronx, NY. I used GPS and GIS to map and analyze the spatial pattern of disease in the NYBG forest and I examined the disease along an urban-rural gradient.
Morrison J.A. 2000. Dogwood decline in the urban New York Botanical Garden Forest. Abstract, Botanical Society of America Annual Meeting 2000.
Much of the forest in the eastern United States exists as fragments surrounded by urbanizing development, yet little explicit attention has been paid to plant ecology in forests embedded in an urban matrix. In the 16 ha old-growth New York Botanical Garden Forest, in Bronx, NY, many native tree species exhibit decline and/or lack of recruitment. Cornus florida shows distinct symptoms of decline that may be due to dogwood anthracnose disease, age-related senescence, and/or environmental stress. In order to investigate spatial factors often associated with disease, such as host density and distance from the forest edge, I characterized spatial patterns of symptoms and mortality from 1995 to 1998, using GIS with positional data from a sub-meter accurate GPS. Most dogwoods in the forest were mature; only 2% had DBH < 3 cm in 1998 and no seedlings were found in either year. In both years the fungal pathogen Discula destructiva was present. Nearly all trees showed some symptoms (lower branch and twig die-back, conidiomata on leaves, leaf blotch). Of 219 trees found alive in 1995, 12% were dead by 1998, 80% were alive, but another 8% were not relocated. Dead trees had a smaller DBH on average, suggesting that older trees were not more likely to die. No clear spatial pattern was evident for symptoms, but there were patterns for mortality. Only one of the 41 trees at or close to the forest edge (within 15 m) died. The 81 trees within 2 m of another dogwood were also less likely on average to die (only 8.6%). So, the interior of this urban forest appears to present a challenge to isolated young dogwoods in particular. At the observed mortality rates, and without recruitment, the interior of the NYBG Forest may lose half of its dogwood trees by 2013.
Morrison, J.A. 1997. Dogwood decline and disease along an urban-rural gradient. Abstract. Supplement to the Bulletin of the Ecological Society of America 78.
Metropolitan areas present problematic environmental conditions for natural plant populations remaining in parks and woodlots, but which evolved in more pristine conditions. Plant resistance to disease is one important facet of plant ecology that may be affected. Flowering dogwood is subject to dogwood anthracnose, caused by the fungus Discula destructiva. As an initial test of whether trees in forested areas within urban centers exhibit greater decline and more disease, in the New York City region I sampled 18 trees each from three urban, three suburban, and three rural forests with similar dogwood densities. Urban forest trees had significantly more epicormic sprouting and somewhat more lower branch die-back ; both are anthracnose symptoms and/or response to other stressors. Urban trees also had larger DBH, indicting possibly greater age, which could influence disease susceptibility and decline. However, there was no relationship between DBH and either symptom. Compared to the urban and suburban forests, rural forest trees were significantly less likely to have Discula present and sporulating on leaves, and there was no relationship between presence of fungus and DBH. These results present some evidence that tree decline is more pronounced at the urban end or urban-rural gradients, and that it might be due to increased effects of pant pathogens in metropolitan environments.
3. Andropogon virginicus(broomsedge) and the smut fungus Sporisorium ellisii
This grass is widespread in the eastern United States, where it is a dominant perennial in abandoned agricultural fields. It also is found in California and in the Pacific basin, including in Hawaii, where it is considered invasive. As are many grass species, A. virginicus is subject to infection with smut fungi that invade its reproductive structures, rendering it sterile. I use the A. virginicus – Sporisorium ellisii study system to address three major research questions (partially funded by a grant to J.A. Morrison from the USDA #NJR – 2004 -01992).
- How does disease affect the population biology of plants? I tackled this question by studying two A. virginicus populations in central New Jersey over four years, tracking changes in infection rates and the fates of individual infected and uninfected plants.
- How variable is disease incidence among populations, and what factors are associated with its variation? I surveyed many populations from Pennsylvania to South Carolina and assessed them for the presence and severity of the smut fungus disease, plant density, and plant size. We also are using ISSR markers to investigate the population genetic structure of the pathogen within host plants, among hosts within populations, and among host populations, and to compare the population genetic diversity of infected and uninfected A. virginicus populations.
- Does escape from disease promote the weediness of plants? I also surveyed many A. virginicus populations in the introduced ranges of California and Hawaii and assessed them for the presence of the smut fungus disease (it was absent), plant density, and plant size. We are comparing these populations with those from the native range, using field data, greenhouse experiments, and ISSR markers.
Collaboration with Dr. Dennis Shevlin’s lab at The College of New Jersey
Presentations and Posters
(All co-authors are TCNJ undergradute students)
August 2013. Ecological Society of America Annual Meeting, Minneapolis, MN. and April 2013. Mid-Atlantic Chapter of the Ecological Society of America Annual Conference, Delaware State University.
Giacopelli, B., M. Readinger, M. Wyles, A. Lugo, and J.A. Morrison. 2013. Comparison of genetic variation in healthy and diseased populations of the old-field grass Andropogon virginicus. (http://esa.org/meetings_archive/2013/webprogram/Paper44314.html). PDF
Wild plant populations are commonly afflicted by disease, but the prevalence can vary among populations. Some populations may exhibit epidemic levels of disease while other populations of the same host plant species appear disease-free. One possible contribution to variation in prevalence is the genetic structure of the host plant populations. Populations that are more genetically homogeneous should be more likely to become diseased, for two reasons. First, a pathogen that can successfully infect a very common plant genotype will spread easily among those abundant, common genotypes. Second, a pathogen should be able to more easily evolve virulence in a population with fewer different genotypes. We tested this idea for healthy and infected populations of Andropogon virginicus (broomsedge), a common old-field grass in the eastern U.S, where only about 50% of populations are infected with the smut fungus Sporisorium ellisii. We did PCR for ISSR markers, on 20 individuals each from three healthy and three infected populations. We extracted DNA with DNeasy spin columns by Qiagen, from greenhouse plants grown from field-collected seed. Every test plant had a different seed mother. We calculated percent polymorphic loci and Nei’s gene diversity for each population, using methods for dominant markers. Sporisorium ellisii has strong negative consequences for A. virginicus; it eliminates reproduction and increases mortality. In addition to our expectation of disease developing in host populations with lower genetic variation, these negative effects could also result in decreased variation. However, our results did not show this pattern. Percent polymorphism ranged from 33-67% in the healthy populations and 42-50% in the infected populations, with no significant difference between the means for healthy (50%) and infected (47%) populations. Similarly, gene diversity ranged from 0.10-0.15 in the healthy populations and 0.08-0.16 in the diseased populations, again with no significant difference between the means (healthy = 0.12, diseased = 0.11). This suggest that infection of A. virginicus populations by the smut fungus S. ellisii is not related to host genetic variation. There are a number of possible explanations. First, these results are based on 11 polymorphic markers; addition of more populations and markers will provide more information. Second, ISSRs provide a genetic fingerprint across the entire genome and are not focused on specific disease resistance genes, which could exhibit different patterns of genetic structure. Third, environmental variables may be the stronger drivers of disease development in this system.
Readinger, M., B. Giacopelli, M. Wyles, A. Lugo, and J.A. Morrison. 2013. Comparison of genetic variation in native and non-native populations of the grass Andropogon virginicus. (http://esa.org/meetings_archive/2013/webprogram/Paper44400.html). PDF
Are introduced, non-native populations more or less genetically variable than native populations? This is an important question for understanding the ecological and evolutionary potential of non-native populations and also for planning biocontrol. Introduced populations may be expected to be less variable because of the founder effect. Alternatively, if they are very successful and are spreading widely, they may be expected to be more genetically variable, allowing them to exploit a wider range of environments. We examined genetic variation in native and non-native populations of Andropogon virginicus (broomsedge), a grass that is common and native in old-fields of the eastern U.S., introduced and naturalizing in mesic habitats of the Central Valley of California, and introduced and widely invasive in the Hawaiian Islands, notably in Hawaii Volcanoes National Park. We conducted PCR using ISSR primers on 20 individuals each in 3-6 populations from each of the regions. We extracted DNA with DNeasy spin columns by Qiagen, from greenhouse plants grown from field-collected seed. Every test plant had a different seed mother. We calculated percent polymorphism and Nei’s gene diversity for each population, using methods for dominant markers. Native A. virginicus populations exhibited, on average, somewhat higher percent polymorphism than the non-native populations (means: Native, 48.6%; California, 35.4%, Hawaii 38.9%), and included the population with the highest value (66.7%). The lowest value was in California (25%). However, ANOVA did not detect significant variation among the regions (P=0.11). The gene diversity results showed that populations from California were, on average, somewhat less genetically variable than the others (means: California, 0.09; Hawaii, 0.13; native, 0.12), but the regions were not significantly different (P=0.18). These results suggest the possibility that Californian populations of A. virginicus are generally less diverse than other populations (addition of more ISSR markers and populations will provide more information). This may be due to a founder effect but additionally, in the seasonally dry Central Valley, the perennial A. virginicus is restricted to lakeshores, riparian corridors, seeps, etc. This restriction may have acted to select for a narrow range of genotypes. The plant is found in a much wider range of habitats where native. In the remote Pacific islands of Hawaii, where a strong founder effect might be expected, it grows in variable niches that range from very xeric lava flows resembling asphalt to pastures and submontane forests.
August 2010. Plant Biology 2010: Joint Annual Meeting of the American Society of Plant Biologists and the Canadian Society of Plant Physiologists, Montreal, QU.
Harth. J. and J.A. Morrison. 2010. Smut fungus detection and infection in the perennial grass Andropogon virginicus. PDF
Andropogon virginicus, a dominant grass in mid-Atlantic old fields, can be attacked by the smut fungus Sporisorium ellisii. This pathogen causes systemic or partial infections, reduces/eliminates seed production, and increases mortality. Our ability to establish infections experimentally is crucial for further research on host resistance and the modes of transmission and infection. We collected 25 partially infected plants from a New Jersey population. These had some tillers infected, indicating susceptibility and providing a source of fungal inoculum (teliospores), and some uninfected tillers, providing a source of seeds matched to inoculum. We prepared four inoculum types: teliospores matched from the same plant as the target seed, mixed teliospores, mycelial cultures grown from single, matched teliospores, and mixed mycelial cultures. We applied each inoculum to four seeds, four seedlings, and four seed-grown adults per maternal plant, with controls. The inoculated adult plants flowered, at which point infection can be detected (sporulation in the grass spikelets), but we saw no infections for any inoculation type. We are monitoring plants inoculated at the seed and seedling stage until flowering. Since visual detection of infection must wait until flowering, we also are developing a molecular method of detection, which tests for the presence of S. ellisii DNA within the plant. We have developed primers that amplify S. ellisii DNA but not A. virginicus DNA, and have tested different plant tissues from field-collected infected plants. So far we have detected smut DNA only in root crown tissue.
April 2010. Mid-Atlantic Chapter of the Ecological Society of America Annual Conference, University of Mary Washington, Fredericksburg, VA.
Nowicki, E. and J.A. Morrison. 2010. Genetic analysis of the smut fungus Sporisorium ellisii using direct polymerase chain reaction (PCR).
Abstract: To understand the interaction between a fungal plant pathogen and its host plant population, it is necessary to have genetic information for both populations, but little work has been done to characterize the population genetics of wild plant pathogens. We seek to develop and use an efficient method for acquiring molecular population genetic data from fungal pathogens. Specifically, we are interested in fungal genetic diversity within hosts, among hosts, and among populations of Sporisorium ellisii, a smut fungus in the Basidiomycete family. This fungus causes systemic infection, decreased fecundity, and increased mortality in its host plant, the perennial bunchgrass Andropogon virginicus. Our goal was to design a more efficient method of obtaining genetic data from fungal spores. Individual fungal spores within a plant may have different genotypes, thus necessitating the culturing of single spore races for genetic work, but these cultures take weeks to grow and often have high contamination rates. We have developed a technique of performing polymerase chain reaction (PCR) directly on germinated teliospores of S. ellisii, without first extracting DNA from cultures. Using this technique with inter-transcribed sequence (ITS) markers from ribosomal genes, we have shown that DNA bands can be produced on a gel, extracted and sequenced for genotype. While three spores per PCR reaction showed a higher success rate (80%), sequenceable DNA was also produced from single spores (50% success rate). We are currently optimizing our method for use with highly variable inter-simple-sequence-repeat (ISSR) markers.
Wong, J., L. Frazee, and J.A. Morrison. 2010. Optimizing the use of inter-simple-sequence repeat (ISSR) markers to detect diversity in Andropogon virginicus. PDF
Andropogon virginicus is an indigenous grass of eastern North America that plays a significant role during succession in old fields throughout the mid-Atlantic Abstracts 32 region. It is introduced in California, where it is naturalizing in hydric soils of the Central Valley, and in Hawaii where it ranks as a serious invasive species. About 50% of native populations are infected with a smut fungus pathogen, but this disease is absent in the introduced ranges. We seek to characterize the population genetic structure of this grass in the context of its invasion and disease ecology, so we are optimizing the use of highly variable ISSR (inter simple-sequence-repeat) molecular markers for use with A. virginicus. To date, no molecular markers have been developed for use in this important species. We selected a group of 18 ISSRs to screen, based on their published use in related grasses. We experimented with different PCR (polymerase chain reaction) conditions to try and produce clear, repeatable, DNA bands that exhibit polymorphism among plants that we collected from a population in Tyler State Park, PA. We experiment with different combinations of PCR annealing temperatures (45 – 50C) and particular ISSRs, with PCR products run through 1.5% agarose gels. Of the initial 12 ISSRs tested thus far, four show repeatable, polymorphic bands. We aim eventually to use an expanded set of ISSR markers to compare the level of genetic variation among infected and uninfected native populations, and between native, naturalizing, and invasive populations.
August 2009. Ecological Society of America Annual Meeting, Albuquerque, NM.
Morrison, J.A., J. Harth, L. Frazee, R. Doherty, and A. Romanchuk. 2009. The effect of mowing disturbance on a wild plant-pathogen interaction in a successional, post-agricultural community. (http://esa.org/meetings_archive/2009/Paper18472.html) PDF
Plant-pathogen research in non-agricultural species focuses primarily on intact natural communities, yet many host plants live in anthropogenic landscapes. The perennial grass Andropogon virginicus (broomsedge) persists in post-agricultural old-field communities with regular disturbance (e.g. mowing); otherwise it decreases during succession. Of populations from Pennsylvania to South Carolina, 50% were infected by the smut fungus Sporisorium ellisii, which decreases reproduction and increases mortality. In a New Jersey population, disease rates declined from 2003 to 2006 as the community underwent succession without major disturbance. We hypothesize that regularly mown plants are more likely to be diseased, due to spreading of fungal teliospores, reduction of woody competitors, increased host density, and production of wound sites for pathogen entry. We tested this idea by mowing three of five 20×100 m strips in our study site in 2007, and then comparing changes in density, disease, and host size from 2006 to 2008 in the mown and unmown strips. Also, aboveground disturbance may select for a mechanism of pathogen persistence in basal host tissues. We developed DNA primers and a PCR technique to detect S. ellisii DNA within host plants, and tested for its presence in various broomsedge tissues. Prior to mowing, in 2006, host density in 480 1m2 plots was not significantly different in the strips assigned to mowing vs. not mowing (3.51 vs. 3.11 plants/m2). After the 2007 mowing, 2008 average density decreased to 1.45 in unmown and 1.41 in mown plots. Average disease prevalence per plot in 2006 was 0.07 in unmown plots and 0.05 in plots slated for mowing; in 2008 it was 0.10 and 0.07, respectively. The decrease in density was not significantly different among treatments, but was significant among strips within treatments (P<0.0001). The increase in prevalence was not significant among or within treatments. Mown plants were 76% as tall as unmown plants (P<0.0001), but had similar numbers of tillers and severity of infection. PCR detection of S. ellisii within plants from the field identified smut DNA only in the root crown of overwintering broomsedge. Broomsedge commonly inhabits frequently mown landscapes, so a pathogen strategy of localizing to the root crown fosters pathogen fitness. However, a single mowing disturbance was not important for plant or pathogen dynamics, even though it negatively affected plant height. Other factors acting at the scale of approximately 20 m may be more influential.
March 2009. Mid-Atlantic Chapter, Ecological Society of America Annual Conference, held jointly with the Mid-Atlantic Chapter of the Society for Ecological Restoration, Ewing, NJ.
Doherty, R., L. Frazee, J. Harth, and J.A. Morrison. 2009. Plant-pathogen interactions in an anthropogenic landscape: contrasting effects of undisturbed succession and mowing. PDF
The warm season, perennial grass Andropogon virginicus (broomsedge) can dominate successional fields in eastern North America, particularly on post-agricultural poor soils. It persists in communities with regular disturbance (e.g. mowing); otherwise it decreases as woody vegetation increases. A survey of populations from Pennsylvania to South Carolina showed that 50% of populations were infected by the smut fungus Sporisorium ellisii (Basidiomycota, Ustilaginales), which decreases photosynthesis and host size, reduces/eliminates reproduction, and increases mortality. We have studied a 100×100 m section of an infected population in New Jersey since 2003, in order to understand host and pathogen population dynamics and their reciprocal influences. Disease rates declined from 2003 to 2006, while the community underwent succession without major disturbance. In the human-modified landscapes common to broomsedge, mowing is a frequent disturbance that may act to increase disease rates by spreading fungal teliospores, reducing woody competitors and increasing host density, and producing multiple wound sites for pathogen entry. We tested this idea by mowing three of five 20×100 strips once in the study site in 2007, and then comparing changes in host size, density, and disease from 2006 to 2008, in the mowed and unmowed strips. Mowed plants were only 76% as tall as unmowed plants on average. Broomsedge density decreased and disease frequency increased across the site, with significant variation among strips, but independent of the mowing treatment. This suggests that other environmental factors acting at the scale of 20 m may influence both host and pathogen dynamics more than does physical disturbance. (Poster 15, Plant Ecology)
Nowicki, E. and J.A. Morrison. 2009. Optimizing direct polymerase chain reaction (PCR) with germinated smut fungus (Sporisorium ellisii) teliospores, for use in ecological studies. PDF
Plant pathogens can strongly affect plant populations by influencing the structure and dynamics of natural populations and by decreasing crop yield and quality, yet they have received far less attention from ecologists than have herbivores. In many cases, little is known about even the most fundamental aspects of fungal pathogen ecology, such as the in-situ life cycle and basic population genetic features. Sporisorium ellisii is a smut fungus (Phylum Basidiomycota, Order Ustilaginales) that infects populations of the common, old-field perennial grass Andropogon virginicus in its native range within the eastern United States. Infection can cause host sterility and mortality. A goal of our laboratory group is to document the population genetic structure of S. ellisii within and among individual host plants, populations, and regions. Currently, nothing is known about genetic variation in this ecologically important fungal pathogen. Identification of DNA markers in fungi traditionally requires costly and time consuming DNA extraction from slow-growing fungal cultures or scarce teliospores, prior to PCR. We are optimizing a new technique to perform PCR directly on newly germinated teliospores, without first culturing the fungus and extracting DNA. The direct PCR method is most successful when three germinated, one-day-old teliospores are used per reaction. Single germlings produce unacceptably inconsistent results. If all teliospores within an individual host are genetically identical, the three-spore direct PCR method will enable very efficient and inexpensive genetic data collection at the multiple-population scale, making this technique very useful for fungal pathogen population genetics. (Poster 16, Plant Ecology)
August 2008. Ecological Society of America Annual Meeting, Milwaukee, WI.
Morrison, J.A., E. Bojdani, B. Graf, and A. Romanchuk. 2008. Native grass-pathogen dynamics and non-native escape from disease: Andropogon virginicus and smut fungus. (http://esa.org/meetings_archive/2008/P14047.HTM) PDF.
Andropogon virginicus is a C4 perennial grass dominating successional old fields in its native range in eastern North America. It is naturalizing on hydric soils in California, and is invasive in the Hawai’ian islands. In many eastern populations this grass is attacked by the smut fungus Sporisorium ellisii, which replaces the plant’s reproductive structures with its own teliospores. If this pathogen influences its host’s individual and population growth, then escape from the disease may promote A. virginicus invasion in the introduced range, as suggested by the enemy release hypothesis and the evolution of increased competitive ability hypothesis. Infected native populations were censused annually from 2003-2006, with measures of host and disease density in 1m2 plots, and mortality, size, infection status, and photosynthesis rates in focal plants. Multiple populations in the eastern U.S., California, and Hawai’i were sampled in 1 m belt transects for plant density and disease presence, and up to 80 collected plants per population were measured for height and mass, and examined for smut fungus. The populations were compared in two common-greenhouse experiments. One measured seed germination percentages and time to germination, and the other measured photosynthesis rates and competitive ability against a phytometer. In the main native population, mean plant density fluctuated across years, from 3.26-5.47 plants/m2, while mean disease frequency declined, from 29.8% to 5.9% infected plants/m2. In plots with disease present, disease was negatively correlated with plant density (2004: r = -0.50, 2005: -0.45, 2006: -0.63; all P < 0.001, with correction for spatial autocorrelation). Fully infected plants were smaller than healthy plants and produced no seeds. Although 21.2% of infected plants recovered, 70% died after two years, compared to 43% of healthy plants. Eastern populations averaged 2.96 plants/m2. Out of 41 populations, 22 were infected.California populations are much less common, restricted to wet soils, have 2.66 plants/m2, and smut fungus was absent. In Hawai’i, A. virginicus has colonized large expanses of nearly barren lava, pasture lands, and early successional forest, especially in Hawai’i Volcanoes National Park. Plant density averaged 2.97 plants/m2 and smut was absent. Hawaiian populations had lowest mean germination speed and percentage, but no differences in competition or photosynthesis compared to natives. Even though smut infection has adverse effects on individual hosts and plant density, escape from this disease in the invaded range does not appear to have resulted in denser populations or more vigorous plants.
April 2008. Mid-Atlantic Chapter, Ecological Society of America Annual Meeting, Wilkes-Barre, PA.
Bojdani, E., B. Graf. E. Nowicki, and J.A. Morrison. 2008. Have Andropogon virginicus plants from Hawaii evolved to be more competitive that their native counterparts from eastern North America?
Plant invasion may occur due to evolved increased competitive ability (EICA) because of a reduced need for defense against enemies who were left behind in the native range. Andropogon virginicus (broomsedge) is a C4 perennial grass native to the eastern US, and has been introduced into various places including Hawaii, where it is identified as a widespread invasive species. The EICA hypothesis was tested by analyzing the effect of broomsedge from different regions (East, Hawaii, California, Australia) on the growth characteristics of a phytometer, Bermuda grass, in a common-greenhouse competition experiment. Dry mass of shoots and roots of Bermuda grass were greatly reduced by competition from A. virginicus, while the root:shoot ratio increased, but this effect did not differ between A. virginicus from the East (native) or Hawaii (invasive). Hawaiian A. virginicus grew larger than eastern plants, and Bermuda grass reduced the size of A. virginicus plants in competition. This effect was different among A. virginicus from the different ranges, with the Hawaiian plants experiencing a greater negative effect. Even though plants from the invaded range were more vigorous than natives, this did not translate into stronger competitive ability. This suggests that A. virginicus plants in Hawaii have not evolved increased competitive ability per se, but nevertheless have higher inherent growth rates that promote their ability to thrive.
Ligocki, A., E. Nowicki, B. Corbett, A. Romanchuk, and J.A. Morrison. 2008. Three years of change in a natural plant-pathogen interaction.
Plant diseases can be important in ecological and evolutionary processes of their host plants. We have studied the interaction between the grass Andropogon virginicus (broomsedge) and its specialist pathogen, the smut fungus Sporisorium ellisii. Infected plants experience reduced or complete loss of reproduction because the fungus uses the inflorescence for its sporulation. We are interested in the changing pattern of disease and how it may influence the population dynamics of A. virginicus. We established 480 1 m 2 plots arrayed in a clustered, regular pattern across a 50 m x 150 m section of old-field. Annually in each plot we counted the number of infected and uninfected plants. Infection across the site increased from 2005 (9% plants/plot on average) to 2006 (11%), due to an increase of plots with infected plants (80 to 150). A much greater disease increase occurred in 2007 (19%) because more plots showed disease (194) and because infection became more severe in many plots. In plots with disease present, higher infection rate was correlated with lower plant density within years (e.g. in 2006, correlation with CRH procedure to account for spatial autocorrelation: -0.45, P<0.01). However, infection rate in the previous year either had no correlation with the change in plant density the next year (2005-2006) or was positively correlated with density change (2006-2007, CRH correlation: 0.23, P=0.05). Rather, decreased plant density was strongly correlated with higher plant density the previous year. These results illustrate a complex interplay between the two populations.
July 2007. Botany 2007: Botanical Society of America and American Society of Plant Biology Joint Annual Meeting, Chicago, IL.
Romanchuk, A. and J.A. Morrison. 2007. Effects of smut fungus infection on the early successional grass Andropogon virginicus. (ASPB SURF recipient). PDF
Andropogon virginicus is an important early successional species. This perennial C4 bunchgrass often is found dominating old fields along the East coast of United States, and is introduced in California and Hawaii. In the east, populations of A. virginicus often harbor a pathogenic smut fungus, Sporisorium ellisii. Infected plants are smaller and exhibit lower photosynthesis rates than healthy plants, and mortality is greater in infected plants (70% compared to 43% in healthy plants in one population from 2005 through 2006. This fungus remains largely asymptomatic until the grass flowers, when infected individuals display fungal sori instead of normal flowers and fruits. As a result, we do not yet know the pattern of infection throughout the host tissue; while most infected plants exhibit systemic infection (all flowers on all shoots are replaced by sori), we can not tell if fungal hyphae are growing throughout the plant. Additionally, infections appear to be perennial; most plants infected one year remain infected the next year, but it is possible that this may actually be re-infection. Therefore, to facilitate our further study of this plant-pathogen interaction we are developing a molecular marker method to diagnose S. ellisii infection within host plant tissue, by utilizing S. ellisii specific primers developed from within the ribosomal ITS sequence.
Morrison J.A. 2007. Spatial and temporal population dynamics of a smut fungus – grass interaction. PDF
Disease dynamics and host population dynamics in natural plant pathosystems remain poorly understood, with only a handful of well-studied systems. We investigated an interaction in which the common perennial old-field grass Andropogon virginicus (broomsedge) is attacked by the parasitic smut fungus Sporisorium ellisii, which replaces the plant’s reproductive structures with its own teliospores. From 2004-2006, in a 100 x 100 m population, we investigated the relationship between host population density and disease frequency by counting infected and healthy plants within 480 permanent 1 m2 plots each year. The effect of disease on individuals was measured on focal plants. Mean plants/m2 ranged from 5.47 in a 2002 pilot study to 3.59 in 2004, 5.52 in 2005, and 3.26 in 2006. Plant density in plots was positively correlated across all pairs of years (all r > 0.64, all P< 0.0001, correlation with CRH correction for spatial autocorrelation). Mean disease frequency in plots ranged from 29.8% in the 2002 pilot study to 10.6% in 2004, 8.9% in 2005, and 5.9% in 2006, and was positively correlated across adjacent years only (2004-2005, P<0.0001; 2005-2006, P = 0.0005). In plots with disease present, disease frequency was negatively correlated with plant density in each year (r = -0.50 in 2004, -0.45 in 2005, -0.63 in 2006; all P < 0.001, with CRH correction). Fully infected plants were smaller than healthy plants, were at significantly higher risk for mortality, and produced no seeds. Although 21.2% of infected individuals recovered from disease, 70% of infected plants died after two years compared to 43% mortality in healthy plants. Sporisorium ellisii has a negative effect on A. virginicus and may act both in population regulation and, potentially, as an agent of natural selection that drives the evolution of A. virginicus populations toward increased resistance.
March 2007. Mid-Atlantic Chapter, Ecological Society of America Annual Meeting, York, PA.
Morrison, J.A. 2007. Smut fungus disease and host plant density in native, naturalized, and invasive populations of the grass Andropogon virginicus (broomsedge). Morrison J.A. Presentation 2007
Andropogon virginicus is a very common early successional old field grass and pasture weed in eastern North America. Its range also includes California, where it is introduced and naturalizing, and Hawaii, where it is identified as an invasive nonnative species. I visited populations throughout each range and measured plant density in 1 m wide belt transects, examined plants for the presence or absence of a smut fungus disease caused by Sporisorium ellisii, and collected aboveground plant material and seeds for future comparative research on weediness in this species. Eastern native populations are widespread and occurred primarily in open fields and roadsides. They averaged 2.96 plants / m . Out of 41 sampled populations, smut 2 fungus disease was present in 22. California populations occur in the Central Valley region. They are much less common than in the east, and are restricted to continuously moist soils such as riverbanks, pond margins and seeps, presumably because A. virginicus can not tolerate the prolonged summer drought in the Central Valley. Plant density in the California populations averaged 2.66 plants / m and smut 2 fungus was not present in any of the 8 sampled populations. In Hawaii A. virginicus has colonized large expanses of old lava flows of various ages, especially in Volcano National Park; its habitat ranges from nearly barren lava to pasture lands to early successional forest. Plant density averaged 2.97 plants / m and no smut was present 2 in any of the 18 populations I sampled on Hawaii and Maui.
Romanchuk, A.M., J.J. Decker, and J.A. Morrison. 2007. Seed germination in native, naturalized, and invasive populations of the grass Andropogon virginicus (broomsedge) grown in a common environment. PDF
Andropogon virginicus is a C4 perennial grass that colonizes old fields in its native range in eastern North America. It is naturalized on moist soils in California and has become invasive in the Hawaiian islands. We collected A. virginicus from each of these regions and compared seed germination in a common greenhouse environment. For weedy and invasive plants both germination speed and quantity may be important components of their success during colonization in competition with other vegetation. Also, our study system is characterized by the presence of Sporisorium ellisii, a smut fungus, which appears to be restricted to certain native broomsedge populations. Early germination and growth may be advantageous when smut fungus is present, since the youngest germlings are often the most vulnerable to infection. We planted 50 seeds (a seed family) from each of 20 seed mothers from each sampled population, and measured two-week germination rates and maximum germination rates for each family. There was highly significant variation among populations within regions for both variables, but not among regions. Mean maximum germination ranged from 14.4% to 49.3% in eastern, 12.3% to 63.0% in Californian, and 0.0% to 52.9% in Hawai’ian seed families. Mean two-week germination ranged from, 2.0% to 46.9% in eastern, 8.1% to 60.8% in Californian, and 0.0% to 47.0% in Hawaiian seed families. Germination did not differ among families from smut-infected versus healthy populations. Local rather than regional conditions appear to be driving genetic divergence among populations for the key life-history component of seed germination.
Bojdani, E., B. Graf, R. Shupak, A. Romanchuk, J. Decker, and J.A. Morrison. 2007. Smut fungus infection in the perennial grass Andropogon virginicus (broomsedge): individual and population level consequences. PDF
Pathogens have important effects on individual plants and their populations, but have not been well studied. W e investigated a plant-pathogen system in which the perennial old-field grass Andropogon virginicus (broomsedge) is attacked by the parasitic smut fungus Sporisorium ellisii. This smut replaces its host plant’s reproductive structures with its own teliospores. In a three-year field study conducted in New Jersey from 2004-2006, we investigated the relationship between host population density and disease frequency by counting infected and healthy plants within 480 permanent 1 m2 plots each year. The effect of disease upon individual plants was measured on focal plants. Data from a 2002 pilot study of 192 plots provided a population baseline, with mean plant density 5.47 / m and, on average, 29.8% plants infected per plot. By 2006 2 mean density had decreased to 3.26 and mean disease to 5.9%. Higher disease rates were correlated with lower plant density in plots where disease was present, suggesting that smut fungus infection has a negative effect on the broomsedge population. Fully infected plants were smaller than healthy plants, were at significantly higher risk for mortality, and produced no seeds. Although 21.2% of infected individuals recovered from disease, 70% of infected plants died after two years compared to 43% mortality in healthy plants. Therefore S. ellisii has a negative effect upon A. virginicus and may act both in population regulation and, potentially, as an agent of natural selection that drives the evolution of A. virginicus populations towards increased resistance.
October 2005. Principal Investigators Meeting on Invasive Plants: held jointly by US Department of Agriculture and National Science Foundation, Washington, D.C.
Morrison, J.A. 2005. Escape from disease: its role in broomsedge invasiveness and weediness. PDF
March 2005. Northeast Ecology and Evolution Conference, Pennsylvania State University.
Halloran, S. and J.A. Morrison. 2005. A study of the factors influencing infection by Sporisorium ellisii in a population of Andropogon virginicus. PDF
August 2004. Ecological Society of America Annual Meeting.
Morrison, J.A. 2004. Smut fungus in broomsedge populations: infection frequency, photosynthesis, host morphology, and density. PDF
March 2004. Northeast Ecology and Evolution Conference, University of Connecticut.
Smith, S. and J. A. Morrison. 2004. Physiological changes of Andropogon virginicus due to smut fungus infection.
April 2003. Northeast Ecology and Evolution Conference, Rutgers University.
Sabelnik, T and J.A. Morrison. 2003. Environmental and genetic effects in a broomsedge – smut fungus plant- pathogen system.