Directory UMM :Data Elmu:jurnal:A:Applied Soil Ecology:Vol15.Issue1.Agust2000:

(1)

In search of biological indicators for soil health and

disease suppression

A.H.C. van Bruggen

a,∗

, A.M. Semenov

a_{Department of Plant Pathology, University of California, 1 Shields Ave., Davis, CA 95616, USA}

b_{Department of Microbiology, Biological Faculty, Moscow State University, Vorob’evy Gory, Moscow 119899, Russia}

Abstract

While soil quality encompasses physical and chemical besides biological characteristics, soil health is primarily an ecologi-cal characteristic. Ecosystem health has been defined in terms of ecosystem stability and resilience in response to a disturbance or stress. We therefore, suggest that indicators for soil health could be found by monitoring responses of the soil microbial community to the application of different stress factors at various intensities. The amplitude of a response and time to return to the current state before application of stress could serve as measures of soil health. Root pathogens are an integral part of soil microbial communities, and the occurrence of epiphytotics forms an indication of an ecosystem in distress. Disease suppression can be viewed as a manifestation of ecosystem stability and health. Thus, indicators for soil health could possibly also function as indicators for disease suppressiveness. Previously suggested indicators for soil health and disease suppression have mainly been lists of variables that were correlated to more or less disturbed soils (ranging from conventional to organic agricultural soils, grassland and forest soils) or to conduciveness to disease. We suggest a systematic ecological approach to the search for indicators for soil health and disease suppression, namely, measuring biological responses to various stress factors and the time needed to return to the current state. © 2000 Published by Elsevier Science B.V.

Keywords: Microbial diversity; Microbial succession; Resilience; Stability; Stress

1. Introduction

The concept of soil health dates back to ancient civ-ilizations (Doran et al., 1996). It has been considered more or less synonymous to soil quality, defined as “The capacity of a soil to function within ecosystem boundaries to sustain biological productivity, maintain environmental quality, and promote plant and animal health” (Doran et al., 1996). However, the Ad Hoc

∗_{Corresponding author. Present address: Organic Farming}

Systems Group, Dept. of Plant Sciences, Wageningen University, Marijkeweg 22, 6709 PG Wageningen, the Netherlands. Tel.:+31-317-478201; fax:+31-317-478213.

E-mail address: [email protected]

(A.H.C. van Bruggen)

Committee on Soil Quality of the Soil Science Soci-ety of America reserved this definition for soil quality (Karlen et al., 1997). Soil functions include life port processes, i.e. plant anchorage and nutrient sup-ply, water retention and conductivity, support of soil food webs, and environmental regulatory functions, such as nutrient cycling, source of microbial diver-sity, remediation of pollutants, and sequestration of heavy metals (Bezdicek, 1996). Until recently, many researchers defined soil quality primarily in chemical and physical terms but soil quality encompasses three basic components: biological, chemical and physical properties while soil health is determined primarily by ecological characteristics (Karlen et al., 1997). We concur with Karlen et al. (1997) and consider soil quality a broader concept than soil health.

(2)

Soil health can be considered a subset of ecosys-tem health. A healthy ecosysecosys-tem is characterized by integrity of nutrient cycles and energy flows, stability, and resilience to disturbance or stress (O’Neill et al., 1986). Thus, soil health may be associated with bio-logical diversity and stability. Plant and animal disease outbreaks can be considered as indicators of instability and poor ecosystem health. Therefore, there is likely also a link between soil health, the ability of the bi-ological community to suppress plant pathogens, the population density of plant pathogens in soil, and ul-timately disease incidence and severity (van Bruggen and Grunwald, 1996). For this reason, disease sup-pression could function as an indicator for a stable and healthy soil ecosystem.

In this paper, we will first explore the concept of soil health in more detail and discuss the relationship between soil health and soil ecosystem stability. Sec-ond, we will explore various stress factors, the micro-bial responses and resilience to a disturbance or stress. In the ecological literature, a distinction is made be-tween disturbance (short-term) and stress (longer-term or chronic), but we will refer to both of them as ‘stress’. Third, we will list some characteristics that have traditionally been suggested as indicators for soil health, and we will propose a different approach to measuring indicators for soil health. Fourth, we will relate soil health to root disease suppression. Fifth, we will mention traditional approaches to the search for indicators for disease suppressive soils, and finally we will discuss some alternative approaches to searching for such indicators.

2. Soil health

According to the definition of soil health given above, a healthy soil is a stable soil, with resilience to stress, high biological diversity, and high levels of internal cycling of nutrients (Elliott and Lynch, 1994). As pointed out above, ecosystem stability has been related to biodiversity and resilience in response to stress. Soil resilience was defined in terms of tolerance against stress, buffering capacity, and the ability to regenerate (Szabolcs, 1994), but practical methods to measure soil resilience have not been sug-gested so far. Similarly, a relationship between soil resilience and biodiversity has been suggested (Elliott

and Lynch, 1994), but methods to prove or disprove this relationship have not been proposed so far.

Biodiversity in soil refers to a variety of taxonomic groups including bacteria, fungi, protozoa, nematodes, earthworms and arthropods, but in this review we will focus on the first two groups. Microbial diversity in soil is normally assessed as species or genetic diversity rather than structural and functional diversity. How-ever, these last two measures of diversity may be more relevant to soil health (Visser and Parkinson, 1992). This statement is based on the assumption that there is functional redundancy in a healthy soil (Beare et al., 1995), so that the soil ecosystem will recover from a stress factor that eliminates part of the microbial com-munity. Besides the active microbial pool there is a reserve pool of quiescent micro-organisms (more di-verse than the active pool) which can respond to a dis-turbance such as addition of foreign substances to soil (Zvyagintsev et al., 1984). Soil homeostasis is main-tained by this diverse microbial pool. The larger the functional redundancy and diversity, the quicker the ecosystem can return to stable initial conditions after exposure to a stress or disturbance.

De Ruiter et al. (1995) calculated food web stability for various natural and agricultural soil ecosystems. The calculated stability was slightly higher in a na-tive shortgrass prairie soil than in some agricultural soils (at similar or higher latitudes but not at lower latitudes), and in agricultural field plots subjected to integrated farming methods than in companion plots subjected to conventional farming techniques. Soils of natural ecosystems and integrated farming systems are generally considered healthier than those of con-ventional farming systems, although this has not been proven conclusively. Despite the notion that a stable soil ecosystem would imply a healthy soil, microbial populations and species composition are seldom sta-ble but fluctuate with changes in environmental con-ditions.

3. Soil stress factors and microbial succession

Three kinds of stress factors can be distinguished: physical, chemical, and biological. The most im-portant physical stress factors are extreme tempera-tures, extreme matric potentials (drying and rewetting cycles), osmotic potentials, and high pressure (for

(3)

example, by agricultural equipment). Chemical stress factors include pH, excess or shortage of inorganic and organic nutrients, anoxia, salinity, and biocides, such as heavy metals, radioactive pollutants, pesticides, and hydrocarbons. Biological stress factors include again nutrient deficiency or excess (oligotrophica-tion and eutrophica(oligotrophica-tion, respectively), introduc(oligotrophica-tion of exogenous organisms with a high competitive abil-ity, and uncontrolled growth of particular organisms, such as pathogens or predators. Individual stress fac-tors seldom operate in isolation: physico-chemical factors may enhance or weaken biological stress fac-tors, or multifunctional stresses may result from one particular interference, for example soil tillage or incorporation of organic amendments.

Any disturbance of soil will lead to a succession in bacteria and fungi and the associated food web. It will also lead to an initial decrease and then an increase in biodiversity. The extent and duration of the successional changes will depend on the inten-sity and duration of the disturbance. In this respect, we have to distinguish between short-term disturbance and long-term or chronic stresses. In response to a short-term perturbation biological communities in a healthy soil will return relatively quickly to the initial conditions. Long-term or chronic stress will result in long-term succession leading to a new dynamic equi-librium among ecosystem components.

Many researchers have studied effects of all kinds of disturbances like tillage, crop rotation, irrigation, organic amendments, or application of fertilizers or pesticides on soil processes or major groups of organ-isms in soil. For example, Bongers and co-workers studied the effects of both short-term disturbances and long-term stresses on nematode communities (Ettema and Bongers, 1993; Korthals et al., 1996; Bongers et al., 1997). The nematode maturity index decreased and then increased after a disturbance (Ettema and Bongers, 1993), while the community structure was permanently damaged, as evidenced by extinction of predacious nematodes, as a result of long-term stress (Korthals et al., 1996). Under enriched nutrient conditions, the maturity index decreased while the plant-parasite index increased (Bongers et al., 1997), characteristic for poor soil health.

Visser and Parkinson (1992) pointed to the impor-tance of changes in microbial community structure and microbial and functional diversity to assess the

Fig. 1. Ratio of CFUs of copiotrophic bacteria to total micro-scopic counts of bacteria 1 day before, 1 day after and 1, 2, 3, 5, and 7 weeks after incorporation of a vetch/oats cover crop (‘Cover crop’) or the same amount of vetch/oats cover crop foliage (‘Fallow+debris’) into soil, or after leaving the soil unamended (‘Unamended’).

extent of degradation by surface mining and the progress in reclamation of degraded soil. However, relatively few researchers have actually investigated changes in fungal and bacterial community structure over time in response to disturbance or stress. Dom-sch et al. (1983) reviewed the response of soil fungi to fungicide applications. They concluded that in soils with high fungal diversity and functional redundancy the effects on soil respiration and decomposition of particular substrates is generally short-lived because species insensitive to the fungicide will take over the functions of affected species. Another example is that eutrophication by incorporation of fresh organic matter or drying and rewetting leads to a temporary increase in microbial activity, CFU, and the ratio of CFU to total number of cells (Fig. 1) (Zvyagintsev et al., 1984; Staben et al., 1997; van Bruggen and Semenov, 1999). The amplitude of the temporary in-crease and subsequent dein-crease in CFU or microbial activity is larger in fallow soil than in cover-cropped soil (Fig. 1) (van Bruggen and Semenov, 1999) or conservation soil (Staben et al., 1997). Microbial di-versity (in terms of species evenness) is expected to decline immediately after nutrient addition since a limited number of species (fast-growing, copiotrophic species) respond quickly to excess nutrients (Fig. 2).

Maximal biodiversity is expected in climax ecosys-tems. For soil, this means under oligotrophic con-ditions with respect to available carbon sources and essential mineral nutrients, but mesotrophic or even eutrophic conditions in terms of total organic carbon.

(4)

Fig. 2. Ratio of CFUs of copiotrophic to oligotrophic bacteria 1 day before, 1 day after and 1, 2, 3, 5, and 7 weeks after incorporation of a vetch/oats cover crop (‘Cover crop’) or the same amount of vetch/oats cover crop foliage (‘Fallow+debris’) into soil, or after leaving the soil unamended (‘Unamended’).

This general concept was verified in a series of exper-iments in which the effects of various stress factors (temperature, mineral salts, drying-rewetting, and combinations of these factors) on cellulase activity and fungal communities were investigated for peat bogs (Nizovtseva and Semenov, 1995; Semenov and Nizovtseva, 1995). Maximal fungal diversity occurred when cellulase activity was minimal, namely under oligotrophic conditions without addition of mineral elements or drying-rewetting stress. Any of the ap-plied stress factors resulted in increased cellulase activity and a succession in the micromycete commu-nity leading to an initial decrease in fungal diversity followed by an increase.

In other studies, metabolic diversity (in terms of evenness as well as richness) decreased after tillage which results in an increase in available carbon compounds from necromass (Lupwayi et al., 1998). Similarly, scarification of forest soil resulted in a temporary lower metabolic diversity than in control plots (Staddon et al., 1997). Such a decline in di-versity after temporary eutrophication will likely be followed by an increase in diversity as the soil returns to an oligotrophic state. The same may hold for other disturbances, for example introduction of a denitrify-ing Pseudomonas fluorescens into a small-pore soil fraction resulted in a temporary decline in microbial diversity (White et al., 1994). From all these examples it appears that consideration of microbial succession resulting from various stress factors will likely pro-vide the means to determine the health status of a soil.

4. Indicators for soil health

Many authors have attempted to develop soil health indicators by measuring various soil characteristics and relating these to different management practices, productivity, environmental quality, or plant disease severity (Workneh et al., 1993; Pankhurst et al., 1995; Grunwald, 1997; Staben et al., 1997). Doran et al. (1996) presented a list of properties affecting soil ecological functions and quality, for example soil bulk density, water infiltration and holding capacity, total organic C and N, electrical conductivity, pH, plant-available nutrients, and measures of microbial biomass and activity. Although these properties may be useful as indicators for soil quality, they are not necessarily associated with soil health and the main-tenance of essential soil ecological functions. The general approach to measure as many variables as pos-sible and relate them to different uses (natural versus agricultural soil) or soil management practices (such as conventional versus alternative practices in terms of tillage, plant nutrition, or pest control) has not resulted in indicators that are consistently correlated with soil health (Table 1) (Pankhurst et al., 1995; Staben et al., 1997). One of the reasons for the inconsistencies may be the sensitivity of many of these measurements to the time of sampling in relation to significant man-agement or environmental events (tillage, irrigation, residue incorporation, fertilization, rainfall, etc.).

A different approach has been to search for in-dicator organisms associated with healthy or dete-riorated soil. Keystone species or higher taxonomic groups may function as indicators for soil health since

Table 1

Variables associated with severity of corky root (Pyrenochaeta

lycopersici) on tomatoa

Corky root field studies in 1989 and 1990 Variables selected

in both years

Variables selected in one of the years N mineralization potential Organic C

Tissue N Soil N

Soil NO3 Clay%

Soil NH4 Microbial activity

pH Aggregate stability

a_{Determined by stepwise discriminant analysis on physical,} chemical and microbial variables measured on soil and plant tissues from organic and conventional farming systems from a field study by Workneh et al. (1993).

(5)

these control the interactions between other species within an ecosystem. For example, certain springtail species were identified as keystone species in oak for-est soils due to their ability to promote detoxification of phenolic compounds and enhance microbial activity and nutrient cycling (Lee, 1994). Rhizobium species have also been considered as keystone species, and Rhizobium populations have been proposed as indi-cators of soil quality (Visser and Parkinson, 1992). Organisms that are sensitive to anthropogenic influ-ences and thrive only within a narrow range of environ-mental conditions could also act as indicator species. For example, Rhizobium and nitrifying bacteria could be used as indicators in this respect because of their high sensitivity to agrochemical (Domsch et al., 1983). Members of the soil fauna such as earthworms, nema-todes, collembola or predatory mites have also been suggested as potential indicators for soil health (Lee, 1994; Pankhurst et al., 1995; Lau et al., 1997), while soil algae, bacteria-feeding or predacious nematodes, or basidiomycete fungi could function as indicators of industrial pollution of soil (Shtina, 1981; Won-dratschek and Roeder, 1993; Korthals et al., 1996; Lau et al., 1997). However, searching for indicator species does not constitute a systematic approach either.

Instead of a random search for indicators of soil health, we need a unified concept of soil health, and base our search for indicators on this concept. First of all, we have to realize that biological indicators for soil health can be divided into general and specific indicators. General indicators are universal, while the usefulness of specific indicators depends on the geo-graphic zone, climate, soil type, and soil history. The soil characteristics mentioned in the previous para-graph mostly fall under the category of specific indi-cators, so that the interpretation of the values for these characteristics depends on the location and climatolog-ical conditions. Universal indicators may include bio-diversity (Parr et al., 1992), stability and self-recovery from stress. However, microbial diversity determined by phospholipid analysis and diversity of bacteria and fungi cultured on rich media did not differ among soils in conventional or low-input agroecosystems (Buyer and Kaufman, 1996). On the other hand, Moore et al. (1993) showed that the time needed for a soil to re-turn to steady state after perturbation (indicative of resilience) declined exponentially with an increase in primary productivity. Primary productivity in soil is

determined by the return of plant-derived organic mat-ter to the soil which may be higher in natural than in agricultural ecosystems, and higher in biologically than chemically based farming systems. Considering a healthy soil as a stable ecosystem with tolerance or resilience to stress, we propose that the responses of a soil sample to a variety of stresses may be better indi-cators of soil health than soil characteristics measured without imposing stress. Similar ideas were proposed by Kozhevin (1989) and Crowley (1997). However, these have not generally been adopted to measure soil health.

We suggest to apply stress factors (such as dry-ing and rewettdry-ing, mechanical disturbance, ex-cess nutrients or pollutants, or inundation with a micro-organism) to soil samples differing in man-agement history, and then monitor the amplitude of changes in microbial populations and the time needed to return to dynamic steady-state conditions as ob-served before application of stress. An important con-sideration with this approach is the frequency of the measurements, since trends in microbial populations could be missed or misinterpreted if data are collected too infrequently, especially if wave-like fluctuations result from a perturbation (Semenov et al., 1999). This was also shown for nematode successions (Ettema and Bongers, 1993).

Alternatively, thresholds of increasing levels of stress could be determined until the soil ecosystem would not rebound within a reasonable time (in terms of several weeks). Effects of increasing concentra-tions of heavy metals or hydrocarbons on soil fun-gus communities were studied by Guzev and Levin (1995), but only for a few days after the treatments. They distinguished different response zones, namely homeostasis, stress, resistance and repression zones, corresponding to particular concentration ranges of the pollutants. The concentration ranges for these zones could also be used to compare soil health of different soils.

The response of the microbial community to stresses could be measured with a large variety of microbio-logical techniques. The most common traditional mi-crobial measurements such as mimi-crobial activity and biomass by respiration or fumigation–extraction or metabolic profiles do not include important micro-bial groups such as oligotrophs (Hu and van Bruggen, 1998). Thus, these methods have severe limitations as

(6)

indicators for the response to stress, since the transi-tion from eutrophic to oligotrophic conditransi-tions likely is an important characteristic for soil health.

We suggest to test the following methods to track the bacterial succession after application of a stress fac-tor: (1) the ratio of CFUs to total microscopic counts (Fig. 1), coined the index of microbial succession stage (Zvyagintsev et al., 1984; Kozhevin, 1989), (2) the ra-tio of copiotrophic to oligotrophic CFU (Fig. 2) (van Bruggen and Semenov, 1999), or (3) the ratio of res-piration to microbial biomass (Visser and Parkinson, 1992) which all increase in ecosystems after pertur-bation and decrease with soil maturity. These indica-tors, and in particular, the second indicator reflect the nutrient stress tolerance of the species present in soil. Interestingly, measures of nutrient stress tolerance of the plant species in various communities were posi-tively correlated with resistance to frost and drought stress and negatively correlated with resilience after fire (MacGillivray et al., 1995).

It is somewhat more difficult to track fungal suc-cession in soil. CFUs obtained from dilution plating are more indicative of fungal sporulation than of hy-phal growth. It is therefore preferable to use a direct soil plating technique. Since many fungi have both oligotrophic and copiotrophic capacities, it is not very useful to distinguish trophic groups on media differ-ing in nutrient contents. However, it is easier to iden-tify fungi than bacteria using traditional classification techniques. Characteristic ‘sugar fungi’ such as the Mucorales and Moniliales occur early in succession, while cellulolytic and ligninolytic fungi occur typi-cally later in succession.

The above-mentioned techniques are quite insen-sitive to subtle changes in microbial communities. Moreover, only a small proportion of micro-organisms can be isolated from soil. Techniques assessing mi-crobial composition by measuring phospholipid fatty acid (PLFA) profiles and various DNA hybridization or DNA/RNA fingerprinting techniques are more sen-sitive and culture-independent (Torsvik et al., 1996; Bossio et al., 1998). For example, microbial composi-tion as assessed by PLFA profiles was more affected by specific farming operations (disturbances) than by farming system (organic, integrated, or conventional), but the time to return to the initial composition (re-silience) was not compared for the different farming systems (Bossio et al., 1998). Besides monitoring

changes in microbial succession after a perturba-tion, PLFA and DNA- or RNA-based techniques also give the opportunity to estimate microbial diversity (Torsvik et al., 1996). These techniques can then be used to compare recovery of the microbial community from stress for soils varying in microbial diversity (Crowley, 1997).

The problem with these techniques is that the extraction efficiencies of phospholipids and DNA de-pend on soil type and microbial community (Zhou et al., 1996), and that they give little if any information about changes in functional capabilities. Moreover, fatty acids or DNA sequences characteristic for olig-otrophs have not been identified yet. For soil health assessment, the number and identity of phylogenetic, physiological and trophic groups is probably more important than biodiversity as such. Hence, charac-teristic PLFA peaks and function-specific DNA or mRNA probes or primers are necessary to track spe-cific microbial populations after a disturbance (Liu et al., 1997).

Alternatively, substrate utilization tests such as pro-vided by ‘Biolog’ could be used to trace shifts in metabolic activities by microbial communities (Bossio and Scow, 1995; Grunwald, 1997). Differences in sub-strate utilization patterns were detected after carbon enrichment (Bossio and Scow, 1995; Grunwald, 1997) and soil moisture treatment (Bossio and Scow, 1995). However, segments of the microbial community are excluded on these substrate utilization plates, in partic-ular, slower growing organisms such as gram-positive bacteria and oligotrophs (Wuensche and Babel, 1996; Verschuere et al., 1997). Thus, all biological methods listed above have some deficiencies. It is therefore im-portant to use more than one method to monitor mi-crobial succession in soil.

In addition to these potential indicators, inoculum levels of root pathogens and subsequent disease were also considered as potential bioindicators of soil health (Visser and Parkinson, 1992; Pankhurst et al., 1995), but this idea was rejected by Hornby and Bateman (1997) for a variety of reasons, in particular, the de-pendence of root pathogens on cropping history.

5. Soil health and disease suppression

Despite the fact that plant disease suppression can-not be equated with soil health according to Hornby

(7)

and Bateman (1997), disease suppression can be an important function of a healthy soil. Disease suppres-sion is the phenomenon that less disease is incited than would be expected in the presence of a suscep-tible host and a virulent plant pathogen, in a physi-cal environment conducive for infection. Various soil factors, including physical, chemical and biological factors, can contribute to disease suppression (or en-hancement) (Hoeper and Alabouvette, 1996). In this review, we focus on biological factors contributing to disease suppression, although we realize that the extent of biological suppression is affected by envi-ronmental conditions, for example mineral nutrients (Hoeper and Alabouvette, 1996).

Analogous to the distinction between general and specific indicators of soil health, there are two kinds of disease suppression in soil: general and specific (Cook and Baker, 1983). General suppression is a function of antagonism and the nutrient and energy supply avail-able for growth of the pathogen through soil and on the root surface. This kind of root disease suppression has often been observed in natural ecosystems or organi-cally compared to conventionally farmed soil (Fig. 3) (Workneh et al., 1993; van Bruggen, 1995). The mech-anisms and orgmech-anisms responsible for this form of sup-pression are mostly unknown. Specific supsup-pression, on the other hand, is due to a specific interaction be-tween a plant pathogen and one or more antagonists, for example an antibiotic producer or parasite. Spe-cific disease suppression can occur after monocrop-ping, this kind of suppression is called disease decline. A well-known example is decline in take-all of cereal crops caused by Gaeumannomyces graminis (Cook and Baker, 1983; Hornby and Bateman, 1997). This decline phenomenon has been attributed to increases in populations of specific antagonists like phlorogluci-nol producing fluorescent pseudomonads (Raaijmak-ers and Weller, 1998).

Although researchers often focus on one group of organisms at a time, general disease suppression can be dependent on communities of micro-organisms. These communities may be associated with a sub-strate at a particular stage of decomposition under certain environmental and management conditions (Boehm et al., 1993, 1997). The composition of func-tional groups (rather than individual species) may determine the character of the community while indi-vidual species within a functional group may be

terchangeable. This functional composition is tightly linked to microbial succession during decomposition of organic matter. Thus, similar to soil health, disease suppression is likely associated with particular stages in microbial succession depending on the pathogen in question. It has long been known that the effects of organic amendments on disease depend on the specific material used, its chemical composition and C:N ratio, and the time elapsed since incorporation (Papavizas et al., 1968). Many plant pathogens are facultative saprophytes and can compete quite well with the soil microflora for colonization of fresh or-ganic matter. If a cash crop is planted too soon after incorporation of a cover crop, the cash crop may succumb to damping-off caused by Pythium spp. or Rhizoctonia solani (Cook and Baker, 1983). We now look at this established concept in a new light. Dur-ing the decomposition process of organic matter in

(8)

Fig. 4. Damping-off of tomato seedlings caused by Pythium

ultimum and Pythium aphanidermatum naturally occurring in soil

collected 1 day before, 1 day after and 1, 2, 3, and 5 weeks af-ter incorporation of a vetch/oats cover crop (‘Cover crop’) or the same amount of vetch/oats cover crop foliage (‘Fallow+debris’) into soil, or after leaving the soil unamended (‘Unamended’) (no soil collected 7 weeks after incorporation, compare with Figs. 1 and 2).

soil, the soil ecosystem is subjected to oligotrophi-cation, and the ratio of oligotrophic to copiotrophic micro-organisms changes during microbial succession (Grunwald, 1997; van Bruggen and Semenov, 1999). It is therefore likely that a particular range of this ratio is associated with general disease suppression. The actual range may depend on the pathogen and its position on the scale from R- to K-strategists. For example, Pythium ultimum, a typical R-strategist, was not suppressed immediately after cover crop incor-poration (Fig. 4), but also not in highly decomposed peat, when the proportion of ‘putative’ oligotrophs was high (Boehm et al., 1997). Soil with organic mat-ter at an inmat-termediate level of decomposition may be most suppressive in this case. On the other hand, R. solani, a typical K-strategist, was suppressed at later stages of decomposition of organic debris in soil and at higher ratios of oligo- to copiotrophic bacteria than Pythium aphanidermatum (Grunwald, 1997).

6. Indicators for disease suppression

Similar to the search for indicators of soil qual-ity or soil health, the search for indicators of dis-ease suppression has not always been systematic. A variety of physical, chemical, and microbial charac-teristics of soil has been tested for their relationship to root disease suppression (Hoeper and Alabouvette,

1996; van Bruggen and Grunwald, 1996; Grunwald, 1997; Oyarzun et al., 1998), unfortunately with mixed results.

Numerous investigations have been conducted to find individual microbial species that may be respon-sible for the suppressiveness of soils to a variety of root diseases, with the ultimate aim to find suitable biocontrol agents. If there were individual organisms responsible for disease suppression, these could then also function as indicators for soils suppressive to that particular disease. However, this strategy has been successful only for a limited number of cases of specific disease suppression, in particular, for take-all decline which is primarily caused by phloroglucinol producing P. fluorescens strains (Raaijmakers and Weller, 1998). Since the gene coding for the antibi-otic has been identified, the presence of quantities of this gene in soil above a threshold level can be used as indicator for suppressiveness of take-all (Raaij-makers, personal communication). Another success story is the somewhat less specific suppression of wilt-inducing formae speciales of Fusarium oxyspo-rum by non-pathogenic strains of the same species (Hoeper and Alabouvette, 1996; Larkin and Fravel, 1998). However, the genetic characteristics of the sup-pressive strains have not been identified yet. Several well-known fungal antagonists can often be found in soils with general disease suppressiveness, for exam-ple Trichoderma, Fusarium, Gliocladium, Penicillium and Acremonium spp. (Castejon-Munoz and Oyarzun, 1995; Kurakov and Kostina, 1998). Similarly, cer-tain bacterial genera, such as Pseudomonas, Bacillus, Burkholderia, and actinomycetes are often found in high populations in soils with general disease suppres-siveness (Workneh and van Bruggen, 1994; Larkin and Fravel, 1998). However, there is not always a clear relationship between the effectiveness of strains as biocontrol agents and the suppressiveness of the soil from which they are isolated (Castejon-Munoz and Oyarzun, 1995). In fact, most investigations aimed at finding individual species responsible for disease suppression have been unsuccessful.

Therefore, several investigators have tried to find other microbiological variables that could function as indicators for general disease suppression. Flu-orescein diacetate (FDA) hydrolysis proved to be a good indicator of suppressiveness to several root pathogens, including P. ultimum (Boehm et al., 1997),

(9)

Pyrenochaeta lycopersici (Workneh et al., 1993; Workneh and van Bruggen, 1994), and Phytophthora parasitica (Workneh et al., 1993). However, FDA hy-drolysis was not a reliable predictor for damping-off caused by R. solani and sometimes not even for Pythium damping-off (Grunwald, 1997). In a study on cover crop decomposition, high FDA hydrolysis activity coincided with high activity of P. aphanider-matum (Grunwald, 1997).

Analogous to the search for soil health indicators, the usefulness of general variables like microbial biomass and activity as indicators for disease sup-pression is limited as they are strongly dependent on the time of sampling in relation to significant events and the composition of the organic food base. There-fore, various characteristics of the organic food base itself have been investigated as potential indicators for root disease suppression. The carbon to nitrogen ratio is one of the most frequently tested characteris-tics; indeed a moderately high C:N ratio was usually least conducive for root disease expression (Papavizas et al., 1968). Grunwald (1997) found that the total C and N content and C:N ratio of coarse organic debris extracted from soil was most consistently associated with growth of P. aphanidermatum and Pythium root rot in discriminant analyses with 20 variables. Several techniques for determining the biochemical compo-sition of the organic food base, such as cross polar-ization magic angle spinning 13C nuclear magnetic resonance (CPMAS 13C-NMR) (Boehm et al., 1997), have been used to search for indicators associated with certain stages of organic matter decomposition when disease suppression was observed. This research established a direct relationship between the concen-tration of pre-colonized cellulose in the substrate and disease suppression.

As mentioned above, the composition and diversity of microbial communities changes during decompo-sition of active organic matter in soil. Indeed, a rela-tionship has sometimes been found between microbial diversity and root disease suppression (Nitta, 1991; Workneh and van Bruggen, 1994). Diversity of fun-gal genera isolated from roots was higher in rotated than in monocropped fields, and in fields amended with manure or crop residues than in unamended fields. Fungal diversity was negatively correlated with incidence of brown stem rot of adzuki bean caused by Acremonium gregatum (Nitta, 1991). In another

study, the diversity index of functional groups of actinomycetes from tomato rhizospheres was higher in organically than in conventionally managed soil, and negatively correlated with corky root severity caused by P. lycopersici (Workneh and van Bruggen, 1994). However, in a study with peat mixes at differ-ent stages of decomposition, the diversity of bacterial species isolated from cucumber root tips was not re-lated to suppression of Pythium damping-off although a species composition effect was identified (Boehm et al., 1993). Copiotrophic taxa, including fluorescent Pseudomonas spp., predominated in peat mix sup-pressive to P. ultimum (a typical copiotrophic organ-ism) whereas oligotrophs predominated in conducive substrate.

The difficulties encountered in the search for re-liable indicators for disease suppression point at the need for a more systematic approach. If we assume that disease suppressiveness is a measure of ecosys-tem stability and health, it is logical to investigate the relationships between diversity, resilience to dis-turbance or stress, and disease suppression. Similar to our approach to searching for indicators of soil health, we propose to view disease suppression in relation to resistance and resilience in the face of a disturbance or stress. Thus, we propose that changes in microbial community structure and the time required to return to the initial state after application of various distur-bances or stresses could be characteristic for disease suppressive soils (compare Figs. 1 and 2 with Fig. 4). Alternatively, the amount of force needed to perma-nently damage the microbial community could be an indicator of disease suppressiveness and soil health. Several characteristics of the microbial community could be measured over time as pointed out in the sec-tion on indicators for soil health: the copiotrophic to oligotrophic ratio (van Bruggen and Semenov, 1999), the index of microbial succession stage (Zvyagintsev et al., 1984), metabolic profiles (Bossio and Scow, 1995), PLFA analyses (Bossio et al., 1998), or various DNA finger printing techniques after increase of soil microbial DNA using PCR (Liu et al., 1997). This ap-proach seems to be promising based on the results of Boehm et al. (1997) who found a consistent associa-tion between bacterial composiassocia-tion (as determined by fatty acid analyses and growth in culture) at different stages of decomposition of peat with suppression of P. ultimum.

(10)

7. Conclusion

In this review, we showed that there are very clear parallels between the search for indicators of soil health and disease suppression. In both cases, the search has been rather random (with some excep-tions), and the results are very difficult to interpret. Therefore, we suggest a more systematic approach to the search for indicators for soil health and dis-ease suppression. In particular, we suggest to subject soil samples to various stress factors and monitor the microbial (and microfaunal) response at regular in-tervals after application of stress. We suggest to use microbial measurements that will allow observation of succession from copiotrophic to oligotrophic or-ganisms or from R- to K-strategists. This approach has been followed successfully by nematologists (Ettema and Bongers, 1993) but not explicitly by soil microbial ecologists or plant pathologists. By using the same approach in our search for both soil health and disease suppression, we may find that similar or even the same indicators may function for both of these important soil characteristics.

Acknowledgements

We thank Cecilia Jones and Joe Wakeman for shar-ing some of their data.

References

Beare, M.H., Coleman, D.C., Crossley Jr., D.A., Hendrix, P.F., Odum, E.P., 1995. A hierarchical approach to evaluating the significance of soil biodiversity to biogeochemical cycling. Plant Soil 170, 5–22.

Bezdicek, D.F., 1996. Development and evaluation of indicators for agroecosystem health. Agriculture in Concert with the Environment ACE Research Projects Western Region, 1991–1995, 6 pp.

Boehm, M.J., Madden, L.V., Hoitink, H.A.J., 1993. Effect of organic matter decomposition level on bacteria species diversity and composition in relationship to Pythium damping-off severity. Appl. Environ. Microbiol. 59, 4171–4179.

Boehm, M.J., Wu, T., Stone, A.G., Kraakman, B., Iannotti, D.A., Wilson, G.E., Madden, L.V., Hoitink, H.A.J., 1997. Cross-polarized magic-angle spinning 13C nuclear magnetic resonance spectroscopic characterization of soil organic matter relative to culturable bacterial species composition and

sustained biological control of Pythium root rot. Appl. Environ. Microbiol. 63, 162–168.

Bongers, T., vanderMeulen, H., Korthals, G., 1997. Inverse relationship between the nematode maturity index and plant parasite index under enriched nutrient conditions. Appl. Soil Ecol. 6, 195–199.

Bossio, D.A., Scow, K.M., 1995. Impact of carbon and flooding on the metabolic diversity of microbial communities in soils. Appl. Environ. Microbiol. 61, 4043–4050.

Bossio, D.A., Scow, K.M., Gunapala, N., Graham, K.J., 1998. Determinants of soil microbial communities: effects of agricultural management, season, and soil type on phospholipid fatty acid profiles. Microbiol. Ecol. 36, 1–12.

Buyer, J.S., Kaufman, D.D., 1996. Microbial diversity in the rhizosphere of corn grown under conventional and low-input systems. Appl. Soil Ecol. 5, 21–27.

Castejon-Munoz, M., Oyarzun, P.J., 1995. Soil receptivity to

Fusarium solani f. sp. pisi and biological control of root rot of

pea. Eur. J. Plant Pathol. 101, 35–49.

Cook, R.J., Baker, K.F., 1983. The Nature and Practice of Biological Control of Plant Pathogens. The American Phytopathological Society, St. Paul, MN, 539 pp.

Crowley, D.E., 1997. Microbial diversity and resilience of agricultural ecosystems: an examination of the diversity-stability hypothesis. http://www.cnr.berkeley.edu/∼gsposito/ Kearney/Proposals/crowley.

De Ruiter, P., Neutel, A.-M., Moore, J.C., 1995. Energetics, patterns of interaction strengths, and stability in real ecosystems. Science 269, 1257–1260.

Domsch, K.H., Jagnow, G., Anderson, T.H., 1983. An ecological concept for the assessment of side-effects of agrochemical on soil micro-organisms. Residue Rev. 86, 65–105.

Doran, J.W., Sarrantonio, M., Liebig, M.A., 1996. Soil health and sustainability. Adv. Agron. 56, 2–54.

Elliott, L.F., Lynch, J.M., 1994. Biodiversity and soil resilience. In: Greenland, D.J., Szabolcs, I. (Eds.), Soil Resilience and Sustainable Land Use. CAB International, Wallingford, UK, pp. 353–364.

Ettema, C.H., Bongers, T., 1993. Characterization of nematode colonization and succession in disturbed soil using the maturity index. Biol. Fertil. Soils 16, 79–85.

Grunwald, N.J., 1997. Characterization of soil nutrient and microbial variables associated with P. aphanidermatum and R.

solani growth and tomato damping-off during short-term cover

crop decomposition. PhD Thesis, University of California at Davis, 190 pp.

Guzev, V.S., Levin, S.V., 1995. Technogenic changes in soil microbiota. Physiol. Gen. Biol. Rev. 9, 1–56.

Hoeper, H., Alabouvette, C., 1996. Importance of physical and chemical soil properties in the suppressiveness of soils to plant diseases. Eur. J. Soil Biol. 32, 41–58.

Hornby, D., Bateman, G.L., 1997. Potential use of plant root pathogens as bioindicators of soil health. In: Pankhurst, C., Doube, B.M., Gupta, V.V.S.R. (Eds.), Biological Indicators of Soil Health. CAB International, Wallingford, UK, pp. 179–200. Hu, S., van Bruggen, A.H.C., 1998. Efficiencies of chloroform fumigation in soil: effects of physiological states of bacteria. Soil Biol. Biochem. 30, 1841–1844.

(11)

Karlen, D.L., Mausbach, M.J., Doran, J.W., Cline, R.G., Harris, R.F., Schuman, G.E., 1997. Soil quality: a concept, definition, and framework for evaluation (a guest editorial). Soil Sci. Soc. Am. J. 61, 4–10.

Korthals, G.W., Alexiev, A.D., Lexmond, T.M., Kammenga, J.E., Bongers, T., 1996. Long-term effects of copper and pH on the nematode community in an agroecosystem. Environ. Toxicol. Chem. 15, 979–985.

Kozhevin, P.A., 1989. Microbial Populations in Nature. Moscow State University Press, Moscow, Russia, 174 pp. (in Russian). Kurakov, A.V., Kostina, N.V., 1998. Saprotrophic micromycetes in the rhizoplane of tomatoes, cucumbers. and soddy-podzolic soil and their ability to suppress Fusarium root rot. Pochvovedenie 2, 193–199 (in Russian).

Larkin, R.P., Fravel, D.R., 1998. Efficacy of various fungal and bacterial biocontrol organisms for control of Fusarium wilt of tomato. Plant Dis. 82, 1022–1028.

Lau, S.S., Fuller, M.E., Ferris, H., Venette, R.C., Scow, K.M., 1997. Development and testing of an assay for soil ecosystem health using the bacterial feeding nematode Cruznema tripartitum. Ecotoxicol. Environ. Safety 36, 133–139.

Lee, K.E., 1994. The biodiversity of soil organisms. Appl. Soil Ecol. 1, 251–254.

Liu, W.T., Marsh, T.L., Cheng, H., Forney, L.J., 1997. Characterization of microbial diversity by determining terminal restriction fragment length polymorphism’s of genes encoding 16S rRNA. Appl. Environ. Microbiol. 63, 4516–4522. Lupwayi, N.Z., Rice, W.A., Clayton, G.W., 1998. Soil microbial

diversity and community structure under wheat as influenced by tillage and crop rotation. Soil Biol. Biochem. 30, 1733–1741. MacGillivray, C.W., Grime, J.P., Integrated Screening Programme

(ISP) Team, 1995. Testing predictions of the resistance and resilience of vegetation subjected to extreme events. Func. Ecol. 9, 640-649.

Moore, J.C., De Ruiter, P.C., Hunt, H.W., 1993. Influence of productivity on the stability of real and model ecosystems. Science 261, 906–908.

Nitta, T., 1991. Diversity of root fungal floras: its implications for soil-borne diseases and crop growth. Jpn. Agric. Res. Q. 25, 6–11.

Nizovtseva, D.V., Semenov, A.M., 1995. Influence of moisture on cellulase activity of micro-organisms in upper peat. Microbiology 64, 841–846.

O’Neill, R.V., DeAngeles, D.L., Waide, J.B., Allen, T.F.H., 1986. A hierarchical concept of ecosystems. Princeton University Press, Princeton, NJ, 263 pp.

Oyarzun, P.J., Gerlagh, M., Zadoks, J.C., 1998. Factors associated with soil receptivity to some fungal root rot pathogens of peas. Appl. Soil Ecol. 10, 151–169.

Pankhurst, C.E., Hawke, B.G., MacDonald, H.J., Kirkby, C.A., Buckerfield, J.C., Michelsen, P., O’Brien, K.A., Gupta, V.V.S.R., Doube, B.M., 1995. Evaluation of soil biological properties as potential bioindicators of soil health. Aust. J. Exp. Agric. 35, 1015–1028.

Papavizas, G.C., Lewis, J.A., Adams, P.B., 1968. Survival of root infecting fungi in soil. II. Influence of amendment and soil carbon-to-nitrogen balance on Fusarium root rot of beans. Phytpathology 58, 365–372.

Parr, J.F., Papendick, R.I., Hornick, S.B., Meyer, R.E., 1992. Soil quality: attributes and relationship to alternative and sustainable agriculture. Am. J. Alternative Agric. 7, 5–11.

Raaijmakers, J.M., Weller, D.M., 1998. Natural plant protection by 2,4 diacetylphloroglucinol-producing Pseudomonas spp. in take-all decline soils. Mol. Plant-Microbe Interactions 11, 144– 152.

Semenov, A.M., Nizovtseva, D.V., 1995. Influence of temperature and mineral supply on cellulase activity and micromycete development in samples of peat from the upper bog. Microbiology 64, 97–103.

Semenov, A.M., van Bruggen, A.H.C., Zelenev, V.V., 1999. Moving waves of bacterial populations and total soluble organic carbon along roots of wheat. Microbiol. Ecol. 37, 116–128. Shtina, E.A., 1981. Soil algae as indicators of environmental state.

In: Kovda, V.A., Glazovskaya, M.A. (Eds.), Problems of Soil Science. Nauka, Moscow, pp. 74–77.

Staben, M.L., Bezdicek, D.F., Smith, J.L., Fauci, M.F., 1997. Assessment of soil quality in conservation reserve program and wheat-fallow soils. Soil Sci. Soc. Am. J. 61, 124–130. Staddon, W.J., Duchesne, L.C., Trevors, J.T., 1997. Microbial

diversity and community structure of postdisturbance forest soils as determined by sole-carbon-source utilization patterns. Microbiol. Ecol. 34, 125–130.

Szabolcs, I., 1994. The concept of soil resilience. In: Greenland, D.J., Szabolcs, I. (Ed.), Soil Resilience and Sustainable Land Use. CAB International, Wallingford, UK, pp. 33–39. Torsvik, V., Sorheim, R., Goksoyr, J., 1996. Total bacterial diversity

in soil and sediment communities: a review. J. Ind. Microbiol. Biotechnol. 17, 170–178.

van Bruggen, A.H.C., 1995. Plant disease severity in high-input compared to reduced-input and organic farming systems. Plant Dis. 79, 976–984.

van Bruggen, A.H.C., Grunwald, N.J., 1996. Tests for risk assessment of root infection by plant pathogens. In: Doran, J.W., Jones, A.J. (Eds.), Methods for Assessing Soil Quality. Soil Science Society of America, Madison, WI, pp. 293–310. van Bruggen, A.H.C., Semenov, A.M., 1999. A new approach to

the search for indicators of root disease suppression. Aust. J. Plant Pathol. 28, 4–10.

Verschuere, L., Fievez, V., Van Vooren, L., Verstraete, W., 1997. The contribution of individual populations to the biolog pattern of model microbial communities. FEMS Microbiol. Ecol. 24, 353–362.

Visser, S., Parkinson, D., 1992. Soil biological criteria as indicators of soil quality: soil micro-organisms. Am. J. Alternative Agric. 7, 33–37.

White, D., Crosbie, J.D., Atkinson, D., Killham, K., 1994. Effect of an introduced inoculum on soil microbial diversity. FEMS Microbiol. Ecol. 14, 169–178.

Wondratschek, I., Roeder, U., 1993. Monitoring of heavy metals in soils by higher fungi. In: Markert, B. (Ed.), Plants as Biomonitors. VCH, Weinheim, pp. 345–363.

Workneh, F., van Bruggen, A.H.C., 1994. Microbial density, composition, and diversity in organically and conventionally managed rhizosphere soil in relation to suppression of corky root of tomatoes. Appl. Soil Ecol. 1, 219–230.

(12)

Workneh, F., van Bruggen, A.H.C., Drinkwater, L.E., Shennan, C., 1993. Variables associated with corky root and Phytophthora root rot of tomatoes in organic and conventional farms. Phytopathology 83, 581–589.

Wuensche, L., Babel, W., 1996. The suitability of the Biolog Automated Microbial Identification System for assessing the taxonomical composition of terrestrial bacterial communities. Microbiol. Res. 151, 133–134.

Zhou, J., Bruns, M.A., Tiedje, J.M., 1996. DNA recovery from soils of diverse composition. Appl. Environ. Microbiol. 62, 316–322.

Zvyagintsev, D.G., Kochkina, G.A., Kojhevin, P.A., 1984. New approaches to the investigation of microbial succession in soil. In: Mishustin, E.N. (Ed.), Soil Organisms as Components of Biogeocenosis. Nauka, Moscow. pp. 81–103.

(1)

in-Fig. 3. Number of corky root (Pyrenochaeta lycopersici) lesions per tomato root (A), and percentage of root tips with softrot (Phytophthora or Pythium sp.) (B) in organic 4-year rotation, low-input 4-year rotation, conventional 4-year rotation, and con-ventional 2-year rotation plots at the sustainable agriculture farm-ing systems (SAFS) field site at UC Davis in 1997 and 1998. terchangeable. This functional composition is tightly linked to microbial succession during decomposition of organic matter. Thus, similar to soil health, disease suppression is likely associated with particular stages in microbial succession depending on the pathogen in question. It has long been known that the effects of organic amendments on disease depend on the specific material used, its chemical composition and C:N ratio, and the time elapsed since incorporation (Papavizas et al., 1968). Many plant pathogens are facultative saprophytes and can compete quite well with the soil microflora for colonization of fresh or-ganic matter. If a cash crop is planted too soon after incorporation of a cover crop, the cash crop may succumb to damping-off caused by Pythium spp. or

Rhizoctonia solani (Cook and Baker, 1983). We now

look at this established concept in a new light. Dur-ing the decomposition process of organic matter in

(2)

Fig. 4. Damping-off of tomato seedlings caused by Pythium

ultimum and Pythium aphanidermatum naturally occurring in soil

solani, a typical K-strategist, was suppressed at later

stages of decomposition of organic debris in soil and at higher ratios of oligo- to copiotrophic bacteria than

Pythium aphanidermatum (Grunwald, 1997).

6. Indicators for disease suppression

1996; van Bruggen and Grunwald, 1996; Grunwald, 1997; Oyarzun et al., 1998), unfortunately with mixed results.

oxyspo-rum by non-pathogenic strains of the same species

(Hoeper and Alabouvette, 1996; Larkin and Fravel, 1998). However, the genetic characteristics of the sup-pressive strains have not been identified yet. Several well-known fungal antagonists can often be found in soils with general disease suppressiveness, for exam-ple Trichoderma, Fusarium, Gliocladium, Penicillium and Acremonium spp. (Castejon-Munoz and Oyarzun, 1995; Kurakov and Kostina, 1998). Similarly, cer-tain bacterial genera, such as Pseudomonas, Bacillus,

Burkholderia, and actinomycetes are often found in

high populations in soils with general disease suppres-siveness (Workneh and van Bruggen, 1994; Larkin and Fravel, 1998). However, there is not always a clear relationship between the effectiveness of strains as biocontrol agents and the suppressiveness of the soil from which they are isolated (Castejon-Munoz and Oyarzun, 1995). In fact, most investigations aimed at finding individual species responsible for disease suppression have been unsuccessful.

(3)

Pyrenochaeta lycopersici (Workneh et al., 1993;

Workneh and van Bruggen, 1994), and Phytophthora

parasitica (Workneh et al., 1993). However, FDA

hy-drolysis was not a reliable predictor for damping-off caused by R. solani and sometimes not even for

Pythium damping-off (Grunwald, 1997). In a study

on cover crop decomposition, high FDA hydrolysis activity coincided with high activity of P.

aphanider-matum (Grunwald, 1997).

Pseudomonas spp., predominated in peat mix

sup-pressive to P. ultimum (a typical copiotrophic organ-ism) whereas oligotrophs predominated in conducive substrate.