Structure and Function in Agroecosystem Design and Management - Chapter 9 pot

184 Effect of Diseases on the Structure of Plant Communities and Plant Population.. However, over the greater part of the history of plant ecology it has been convenient to assume that t

CHAPTER 9 Plant Diseases and Plant Ecology Nikolaos E Malathrakis and Dimitrios G Georgakopoulos

CONTENTS

Introduction 184

Effect of Diseases on the Structure of Plant Communities and Plant Population 185

Age Structure 185

Spatial Structure 185

Plant Density 186

Temporal Structure-Succession 187

Competition 188

Diversity 189

Diversity within Plant Communities 189

Diversity within Plant Populations 189

The Effect of Pathogen Attributes 191

Type of Dispersal 191

Wind Dispersal 191

Rain Dispersal 192

Insect-Transmitted Inoculum 192

Virulence 193

Pathogen Survival 194

Effect of the Type of Epidemic 194

Some Major Plant Epidemics: Ecological Aspects 195

Dutch Elm Disease [Ophiostoma (Ceratocystis) ulmi] 195

Chestnut Blight [Cryphonectria (Endothia) parasitica] 195

Dieback Caused by Phytophthora cinnamomi 196

Potato Late Blight (Phytophthora infestans) 196

Tristeza 196

Other Pandemics 197

183 0-8493-0904-2/01/$0.00+$.50

Weed Control with Fungal Pathogens 198

Epilogue 199

Acknowledgments 199

References 200

INTRODUCTION

Plant pathogens are among the main biotic factors of any ecosystem and may play a major role in its dynamics However, over the greater part of the history of plant ecology it has been convenient to assume that the structure and composition of plant communities is mainly determined by macro- and microclimate, soil conditions, and interactions among the plants themselves (Harper, 1990)

In natural ecosystems the role of plant pathogens has tended to be neg-lected Recently, however, attention has been paid to the importance of plant pathogens and the relevant diseases on the pattern of plant com-munities (Dobson and Crawley, 1994) Dinoor and Eshed (1984) number sev-eral reasons for the growing interest in diseases in the wild The dramatic eco-logical impact of several plant pandemics, such as Dutch elm disease, was probably the most important However, there are other diseases with less evi-dent, but no less important, effects on plant communities that merit great attention

In agroecosystems, on the other hand, there has always been a great deal

of concern about plant diseases, but they were mostly considered from a directly economical viewpoint Well-known examples of destructive diseases

in agricultural systems include the great potato famine, which devastated the population of Ireland from 1846–1851, and the 1943 great famine in Bengal due to rice blast (Strange, 1993), but we know much less about the impact of these, and several other epidemics, on the ecology of their hosts This is prob-ably due to the difficulty in studying this aspect of the consequences of dis-ease and man’s interference, which jeopardizes the potential interactions of plants and plant diseases

Harper (1990) questioned the existence of convincing evidence with respect to the role of pests and pathogens on plant communities and posed fundamental questions which should be answered Although those questions are far from being answered, several publications, which have appeared since then, provide increasing evidence that plant diseases may affect plant ecology through the innumerable interactions taking place in any plant com-munity and plant population

The present chapter approaches the following aspects of the subject: (1) the effect of diseases on the structure of plant communities, (2) the contribu-tion of some major pathogen attributes and the type of epidemics, (3) the eco-logical impact of selected plant pandemics, and (4) the effect of weed control

by pathogenic fungi on regulation of weed populations

EFFECT OF DISEASES ON THE STRUCTURE OF PLANT COMMUNITIES AND PLANT POPULATION

Studies to elucidate plant-disease interactions and their effect on plantecology are few Data from such studies supporting the potential effect ofplant pathogens on several aspects of the structure of plant communities andplant populations are presented below

Age Structure

In the wild, it is assumed that newly established populations are moresusceptible to diseases than are older ones (Harper, 1970) Carlsson et al.(1990) carried out comparative studies of many populations in areas wherepopulation age can be estimated to test this assumption They compared dis-ease incidence caused by three host-specific systemic fungal pathogens on

host plant populations of Valeriana sambucifolia, Trientalis europea, and Silena dioica They found that in all three pathosystems, disease incidence was higher

during an early-intermediate phase of population development Populations

of individual species with an estimated age of over 50, 400, and 300 yearsrespectively showed low disease incidence (10%) In other pathosystems,

natural infections depended upon environmental conditions Armillaria spp.,

for instance, is a well-known group of root rot-inducing pathogens wide They may cause both primary infections of healthy trees as well as sec-ondary infections of stressed trees Primary infections tend to diminish withstand age of over 20–30 years Since only a small proportion of the total pop-ulation is usually infected, aged trees may prevail in infected areas However,

world-in drier, world-inferior forests, contworld-inuworld-ing mortality world-in all age classes is common world-inmany stands (Kile et al., 1991) Trees infected by Dutch elm disease may sur-vive for some years after infection Elms planted in the areas where the dis-ease is prevailing usually survive for less than 20 years Given the destructiveeffect of the disease, old trees in such stands should be rare

In agroecosystems, plant diseases affect the aged structure of standingcrops in two ways First, an established plantation is maintained as long as it

is healthy enough to produce a good yield Second, in several cases early tivars are grown in order to avoid plant diseases The theory behind this is toreduce the time of the epidemic’s progress, as shown in Vanderplank’s equa-

cul-tion describing disease progress, X  X0ert (where X denotes the amount of disease at time t, X 0 the amount of initial inoculum, and r the rate of disease

progress), and keep infection at a low level

Spatial Structure

The effect of abiotic factors on the spatial structure of plants within plantcommunities is much more evident than is the effect of plant pathogens It is

well known, for instance, that soil acidity and soil salinity, among others, aremajor factors that determine the spatial structure of plant communities Therole of plant pathogens is sometimes hidden under the effect of the diseasethat may be exacerbated by abiotic factors For instance, in soils with high

pH, potatoes often fail to thrive not because of the soil pH per se but because

potato scab, which is favored by such soil conditions, becomes a production

constraint Mal secco disease of citrus, caused by Phoma tracheiphila, is

deter-mining which citrus species are grown in several areas in the Mediterranean.Lemon, the most susceptible species, is grown only in the least windy areaswhere infections of wind-damaged shoots are fewer

In the wild, one of the most extensively documented cases is the disaster

caused in Australian forests by the fungus Phytophthora cinnamomi It

pro-duces a typical epidemic which may worsen over approximately five years,but, about three years after infection, field-resistant species colonize the floor

of the diseased forest, thus completely changing the spatial pattern of plants

in the community (Weste and Mark, 1987) Due to its wide host range, theinvasion of this pathogen in an area exerts a definite regulatory effect on anentire set of plants which may be the main component of the local flora Other

plant diseases with a large range of hosts may play a similar role Xylella tidiosa, a xylem-limited bacterium, is the principal factor preventing the development of high quality Vitis vinifera and V labrusca grapes in the south-

fas-eastern U.S where it is endemic (Hopkins, 1989) It is assumed that because

of its very large host range of cropped and wild plants favored by the sameenvironmental conditions, the structure of entire plant communities in thesame area may be affected (Newhook and Podger, 1972)

Plant Density

In natural systems, plant density is the result of the established tions of all the biotic and abiotic factors Generalizations about patterns ofdensity-dependent mortality and reproduction are a subject of plant popula-tion ecology (Harper, 1977) However, there is limited information on theeffect of plant pathogens on the relevance of these generalizations for plantpopulations growing in the presence as opposed to absence of different plantpathogens (Mihail et al., 1998) Dense stands contribute to increased diseaseinfection because of the establishment of microclimatic conditions such ashigh relative humidity, which favor pathogen infections, increase root con-tacts that enhance transmission of root diseases, etc (Burdon and Chilvers,1982), indicating the regulatory effect of diseases on host populations Severalfungal pathogens and all bacteria require free moisture to produce disease,while infections by most other fungal pathogens are favored by high relativehumidity (Harrison et al., 1994) The spread of root diseases, such as white rot

interac-of onions caused by Sclerotium cepivorum, is positively correlated with plant

density and can be controlled by spacing the host to eliminate root contactbetween adjacent plants (Scott, 1956) Mihail et al (1998) reported that

Rhizoctonia solani and Pythium irregulare reduced the number of plants and the total biomass of the annual legume Kummerowia stipulacea Reduction was

higher in plots with higher sowing densities Burdon (1978) claims that “theinteraction between plant density and disease has certain features of a self-regulatory feedback system and as such has special interest in the considera-tion of all plant communities Thus, because of its faster rate of dissemination,

a pathogen is likely to kill more plants at high than at low plant densities.This death of plants reduces plant density and this in turn tends to curb thepathogen through its effect on transmission from plant to plant.” In somepathosystems, thinning has been adapted as a standard practice for diseasemanagement, indicating the regulatory effect of pathogens on plant density.For example, thinning of high risk trees (over 50% girdle) is recommended formanagement of infected mature plants of southern pines infected by fusiform

rust caused by Cronartium quercuum f.sp fusiforme in the U.S (Powers et al.,

1981) Evidence of the regulatory effect of diseases on host population can befound in several other studies (Augspurger, 1988) Ingvarsson and Lundberg

(1993), using a mathematical model to study the effect of Ustilago violacea on the population density of Lychnis viscaria, found three different outcomes of

this interaction: (1) extinction of the fungus, (2) a stable coexistence betweenplant and fungus, and (3) extinction of both plant and fungus Virus particlenumbers may decrease with increasing host density due to the difficulties ofinsect vectors in spreading disease in dense stands (Boudreau & Mundt,1997) Hence, it appears that diseases are an important regulatory factor forplant densities, but their effect seems to be disease specific

Temporal Structure-Succession

Succession is the process whereby one plant community changes intoanother Although the deterministic concept with respect to succession inplant communities was initially accepted, the role of randomness in succes-sion is rather universally adopted now (Crawley, 1994) Stemming from thisnew concept, the role of plant epidemics, which appears as an accident ratherthan as a sequence of events, could also be considered It is reported that dur-ing primary succession in areas where no life pre-existed, the first colonistsare cryptogams (Crawley, 1994) However, we are not aware of any reportthat plant pathogens may interfere in primary succession There are severalmodels on pathways of secondary succession, but all have an intrinsic deter-ministic concept Models based on the facilitation of succession of one organ-ism by another, such as the replacement of fast growing species by slowergrowing ones, etc., nearly predetermine plant succession in the community

on the basis of plant characteristics and available resources None of these

models consider the possible effect of plant pathogens Nevertheless, severalrecent publications regard disturbances mediated by host-specific pathogens

as underlining factors that determine successional relationships in a

commu-nity Holah et al (1997) studied the effect of Phellinus weirii, a native root rot pathogen of Pseudotsuga menziesii (Douglas fir), an early species during suc-

cessional development of infected forests in the lower Cascade and Coast

ranges of western Oregon They found that the presence of P weirii in these sites appears to push changes towards the late successional species, Tsuga het- erophylla, Thuja plicata, and Taxus brevifolia At least in the Cascade mountain

sites, not only was there an increase of the late successional species withininfection centers, but the trajectory along which disease had “pushed” withinthese sites was common to all three areas studied

Competition

Nutritional resources are the most studied factors affecting competition

in plant communities (Tilman, 1994) However, there is increasing evidence

of plant pathogen interference on interspecific competition among plants inthe wild The main evidence is the flourishing of species introduced into

areas in the absence of their pathogens Chondrilla juncea, a common but not

dangerous weed in Mediterranean countries, became a noxious weed

throughout Australia As soon as the fungus Puccinia chondrillina, a pathogen

of this plant in its origin, was introduced, C juncea populations declined

(Hassan, 1988) Several other reports indicate that rust fungi and otherbiotrophic pathogens reduce the ability of infected plants to compete withhealthy ones Burdon and Chilvers (1977) found that mildew reduced thecompetitive ability of barley when grown in mixtures with wheat Paul and

Ayres (1987) noticed reduced competition of Senecio vulgaris infected with the species-specific rust fungus Puccinia lagenophorae over lettuce (Lactuca sativa)

when grown in mixtures Finally, Paul (1989) studied the effect of the same

fungus on the competitive behavior of S vulgaris versus the weed Euphorbia peplus and found that infected S vulgaris was less competitive than the

healthy There are fewer, but not less important, examples from soil-bornediseases Van der Putten and Prters (1997) found strong evidence that when

Ammophila arenaria was exposed to its soil-borne pathogens, it was peted by Festuca rubra spp arenaria, especially under nutrient limitation.

out-com-The main issue is how pathogens affect the competitive ability of infectedplants Many factors are involved, such as the number of competing geno-types, pathogen type, infection time, and environmental factors, making itdifficult to draw an overall conclusion Reduction of seed production due topathogen infection might reduce the competitive ability of the infected plant

In S vulgaris infected by P lagenophorae, seed production decreased by 60%

over that of the healthy plants (Paul and Ayres, 1986) It seems that in eachpathosystem the reaction is different and not always easy to identify

Diversity within Plant Communities

The role of plant pathogens in plant community diversity, neglected for

a long time, has recently been recognised both for aerial (Alexander et al.,1996; Burdon, 1987) and root-infecting pathogens (Bever et al., 1997; Burdon,1987) Peters and Shaw (1996) executed an experiment on plots of rough

grassland dominated by Holcus lanatus Plots were cleared of vegetation in

three successive years and allowed to regenerate One third of plots was leftuntreated, one third of plots was regularly sprayed with propiconazole toreduce fungal diseases, and the last third was inoculated with urediospores

of Puccinia coronata f.sp holci on the second and third years of the study and with conidia of the leaf-spotting fungus Ascochyta leptospora in the third year.

Vegetation cover and disease severity were regularly monitored The authorsconcluded that, in communities dominated by grasses, foliar pathogenstended to decrease the abundance of perennial herbs and, therefore,decreased the diversity in regenerating plots by favoring grasses

Mills and Bever (1998) assume that soil community as a whole can tribute to the maintenance of diversity within plant communities They claimthat negative feedback occurs when the presence of a plant alters the soilmicrobial community in a manner resulting in growth reduction of that par-ticular plant species relative to other species, with the potential interference

con-of soil-borne pathogens Assuming that the negative feedback was related to

the species-specific soil pathogens, they tested the effect of Pythium spp on

the growth of plant species in which negative feedback through soil nity had previously been observed Their results suggest that accumulation

commu-of species-specific soil-borne pathogens could account for this negative back and conclude that soil pathogens may themselves contribute to themaintenance of plant species diversity

feed-Diversity within Plant Populations

The effect of plant-pathogen interactions on pathogen populations hasbeen well studied in great detail in a number of agricultural and naturalpathosystems Little work has been done, however, on the long-term effect ofdisease on plant populations, although this situation has started to changewith the use of modern molecular genetic techniques, such as the variouselectrophoretic methods for detection of DNA polymorphism or allozymeanalysis

Plant resistance to pathogens has long been explained by the for-gene theory (Flor, 1971), where a single plant resistance gene interactswith a matching pathogen avirulence gene to produce a resistance reaction.This type of resistance was based on specific interactions between certainplant cultivars and pathogen races Several plant resistance and pathogen

gene-avirulence genes have now been cloned and sequenced, although their mode

of action still remains to be explained Although the gene-for-gene theorywas initially based on an agricultural pathosystem, it has been well docu-mented in wild plant pathosystems as well (Thompson and Burdon, 1992).Single gene plant resistance, however, is not the only type of resistance innature A broad and quantitative type of resistance to pathogens is also verycommon, but it has been less studied, perhaps due to its inherent complexity.This type of resistance also exists in natural plant populations, but its long-term effect has not been elucidated

In natural plant populations, plant resistance genotypes have co-evolvedwith pathogen virulence genotypes, interacting in a perpetual “arms race”where selection of resistance plant genotypes is followed by the reciprocalselection of pathogen virulent genotypes Although this procedure is greatlyinfluenced by environment (Paul, 1990) and spatial features of the surroundingvegetation (Morrison, 1996), a few cases have been documented in which dis-ease altered in time the composition of host plant genotypes in a population

Murphy et al (1982) examined the competitive ability of five oats (Avena sativa) multilines in a mixture over four consecutive years in the presence and absence of infections by the crown rust pathogen Puccinia coronata Each year, plants were inoculated with a mixture of five P coronata races and were either

treated with fungicide during the growing season to prevent infection or leftuntreated During the course of the experiment, the frequency of certain mul-tilines in the population started to rise while others were reduced, but no sta-tistically significant difference was observed in treated and untreated plants

It would be interesting to see whether this trend would be maintained if theexperiment was continued for a number of years This study generates thehypothesis that disease has the potential of reducing genotypic variability in

a population of plants over time

A recent study on the effect of oak wilt epidemic caused by Ceratocystis fagacearum is in accord with the former assumption The genetic structure of

oak trees before and after an epidemic wave was determined with allozymeanalysis of wood samples (McDonald et al., 1998) Post-epidemic trees weresurvivors of a 20-year epidemic Allozyme analysis indicated that geneticdiversity of post-epidemic oak trees was lower than pre-epidemic diversityfor two out of the four allozyme loci tested Data analysis considered theeffect of spatial distribution of trees and suggested that disease was the majorfactor driving this shift in oak forest genetic structure

A hypothesis proposed by Clay and Kover (1996) similarly suggests thatsystemic plant pathogens may sometimes promote host plant genetic unifor-mity Several systemic plant pathogens are known to induce asexual repro-duction of their host or enforce self-fertilization, thus reducing geneticrecombination This provides the pathogen a selective advantage, because asusceptible genotype is perpetuated in a plant population and the pathogencan be vertically transmitted with seed Direct experimental data are needed

to support this hypothesis

Shifts in host plant genotypes effected by disease have been observed in

a number of cases In Australia, an attempt to stop the spread of the

compos-ite weed Chondrilla juncea was undertaken by using the rust pathogen Puccinia chondrillina as a biocontrol agent Plants belonged to three pheno-

typically different genotypes, one of them being the most abundant Afternine years of biocontrol a complete shift in genotype composition wasrecorded, with the formerly most important genotype reduced to extinction

in most areas and the two other genotypes prevailing and becoming the newtarget weeds for control (Burdon et al., 1981)

THE EFFECT OF PATHOGEN ATTRIBUTES

Plant pathogens share some attributes, such as type of dispersal and ulence Each of them, alone or in combination, clearly affects the interaction

vir-of plants and diseases and finally their effect on plant ecology The effects vir-ofsome of these attributes are briefly discussed below

Type of Dispersal

Dispersal of pathogens or their carriers is closely related to the spread ofany epidemic and plays a major role on disease appearance in new areas (forreviews see Fitt et al., 1989; McCartney, 1989) Pathogens are spread in sev-eral ways but for simplicity we mention only wind dispersal, rain dispersal,and insect transmission of inoculum

Wind Dispersal

Airborne spores may travel intercontinentally and cause disease sands of miles away from the original infection For instance, spores of wheatstem rust are transferred each year from Mexico to the U.S and Canada as

thou-well as from India to Scandinavia Coffee rust, caused by the fungus Hemileia vastatrix, is also transferred via airborne spores It was discovered early in

1970 in Bahia, Brazil, and four years later it had spread in South America over

an area equivalent to the size of Central America Coffee rust possibly came

to Brazil from Angola with trade winds across the Atlantic in 5 to 7 days(Schieber, 1975)

For long distance pathogen migration by air currents, propagules shouldreach high altitudes in the atmosphere by eddy diffusion Otherwise theyremain in the lower atmospheric layers and disperse over rather short dis-

tances Studies for dispersal of Cronartium ribicola, the causal agent of

white-pine blister rust indicated that it is spread about 0.4 km away from infected

Ribes Other wind-borne pathogens, such as Venturia inaequalis (apple scab),

follow the same pattern (Meredith, 1973)

Rain Dispersal

Bacterial plant pathogens, as well as fungi producing mucilaginousspores, are dispersed by rain splash since mucilage prevents dispersal bywind alone Distance of dispersal depends on the size of rain drops andrarely exceeds 1 m

Insect-Transmitted Inoculum

The majority of viral diseases and many bacterial and fungal diseases aretransmitted by insects or other animals, such as nematodes Many of theknown catastrophic pandemics are animal-dispersed Chestnut blight is

spread by birds and insects, Dutch elm disease by the beetle Scolytes spp and tristeza by several aphids such as Toxoptera citricida (Agrios, 1997) The dis-

tance of animal-dispersed diseases depends on many factors, including

ani-mal activity, type of crop, plant species and pathogen strain T citricida, for

instance, is 25 times more efficient in transmitting the tristeza virus than

Aphis gossypii Although some strains of the virus are more easily transmitted

by A gossypii than others, their transmissibility, by either species, is markedly

affected by the source plant used for acquisition feeding (Raccah et al., 1978;Bar-Joseph, 1989) Most viruses spread within crops and cause diseases of the

“compound interest” type However, the ultimate proportion of infectedplants and the rate at which new infections appear vary widely among dif-ferent viruses and for different crops Viruses that infect annual crops spreadmore rapidly than those of trees and shrubs In a typical orchard in California,the citrus tristeza virus spreads to an average of two citrus trees a year for eachinfected one already present By contrast, cauliflower mosaic virus spreadsfrom a single infected plant to as many as 131 in one season Invariably,viruses such as citrus tristeza, cacao swollen shoot, and plum pox take severalyears to spread throughout plantations Nevertheless, their ecological impact

is important since trees are far larger and take longer to grow (Thresh, 1974).Long distance transport of several wind-borne diseases is one of theirmain characteristics with respect to their epidemiology and their effect onplant ecology Coffee rust, a wind-borne disease, spread to South Americawithin four years, 1971–1974 (Schieber, 1975), but it took approximately twodecades for Dutch elm disease, another fast-spreading insect-borne disease,

to spread across Europe (Gibbs, 1978; Ingold, 1978), while chestnut blightspread in the U.S at a rate of about 37 km/year (Anagnostakis, 1987).Man himself also acts as the main long distance transporting agent ofmany diseases Several pathogens have been transferred to Europe from theNew World during the last century and changed the structure of severalcrops as well as of natural plant communities For example, potato late blightand downy mildew of grapes were introduced in Europe around 1845 and

1875, respectively, from America (Strange, 1993; Agrios, 1997), and chestnutblight was introduced in the U.S probably from Japan or China Citrus