B. Kerry
Introduction
Exploitation of natural enemies
Environmental concerns
Development of a biological control agent for nematodes - An ecological approach
The next 20 years
In the past 20 years three developments have occurred which have had significant effects on the prospects and opportunities for the biological control of plant-parasitic nematodes. First, several nematicides have been withdrawn from the market because of health and environmental problems associated with their production and use (Thomason, 1987). As a result of this, and increasing public concern over the use of pesticides in food production, there has been increased interest in the development of alternative methods of control, including the use of biological agents. Second, it has been demonstrated in several soils that nematophagous fungi and bacteria increase under some perennial crops, and under those grown in monocultures, and so may control some nematode pests, including cyst and root-knot nematodes (Stirling, 1991). Such nematode-suppressive soils have been reported from around the world and include some of the best documented cases of effective biological control of nematode pests. Finally, a number of commercial products based on nematophagous fungi and bacteria have been developed, but all so far have had only limited success. Their use has been based on empirical research, and it is instructive to consider what might be the key factors for a successful biological control agent for nematodes in order to identify the reasons for the general failure of the products that have been developed.
Biological control is more inconsistent, less effective and slower acting than control normally achieved with chemicals. Although improvements in performance might be expected from more research on individual agents, it seems likely that these limitations are inherent in most biological control agents and that their successful application will depend on integration with other control measures.
The most studied example of natural control of a plant-parasitic nematode concerns the decline of populations of the cereal-cyst nematode, Heterodera avenae, under monocultures of susceptible cereals in many soils throughout northern Europe (Kerry, 1982). This is an example of an induced suppression in which the nematode increases to damaging population densities in the second and third cereal crops, but usually declines thereafter to infestations of <5 eggs/g soil, which cause little loss of yield in northern European conditions; it is essential that the nematode is abundant in the early years of the monoculture to support the build-up of the microbial parasites (Kerry, 1988). The decline in nematode populations is mainly caused by two parasitic fungi, Nematophthora gynophila and Verticillium chlamydosporium, which attack the developing female on the root surface; in suppressive soils 95 to 97 percent of the females and eggs are destroyed (Kerry, Crump and Mullen, 1982). Thus the natural control of cereal-cyst nematode in a range of soils is predictable and effective, but slow acting. Research on the manipulation of natural control has been limited and attention has concentrated on the introduction of specific agents to provide more rapid control that might be commercially exploitable.
It has proved difficult to develop a biological control agent that is effective worldwide for any soil-borne disease. Despite much research effort only two agents have had widespread success: Phlebia gigantea for the control of Heterobasidion annosum, which spreads from tree stumps to the roots of adjacent trees, and Agrobacterium radiobacter that is applied as a root dip to transplants for control of crown gall caused by A. tumefasciens. For both, the agents are applied in high concentrations as inundative treatments to a readily accessible site of action (the cut surface of a tree stump or the bare roots of a transplant) and protection from the diseases for a relatively short period provides long-term control (Deacon, 1991). In most situations where nematode control is required, there is a need to provide long-term protection to a relatively inaccessi- ble and growing root system without the use of inundative treatments that are likely to be impractical and uneconomic. Hence the development of biological control agents for plant-parasitic nematodes is likely to be difficult and to require a detailed understanding of the biology and ecology of the agent and the nematode target.
The farming system in which biological control is applied has a marked effect on the way the agent is used (Davies, de Leij and Kerry, 1991). In general, growers in developed agriculture require an agent that can be applied to crops grown in monocultures over large areas using standard application machinery, so a formulated product with a good shelf-life that can be applied at low dosages is required; seed treatments are preferred for most arable crops. Little research has been done on the mass production and formulation of biological control agents for nematodes. Some organisms, such as rhizosphere bacteria, can be applied as seed treatments (Oostendorp and Sikora, 1989), but such applications tend to provide short-term control and are only useful in reducing the invasion of roots by nematodes that have a single generation in the growing season. In subsistence farming systems, crops tend to be grown in mixed stands in relatively small areas (often less than one hectare) and labour inputs are often large. As a consequence, relatively large application rates (up to one tonne per hectare) of an unformulated agent could be mixed into the soil by hand, as long as the organism could be produced cheaply and locally. Thus, the initial exploitation of biological control agents for nematodes may be in developing countries (Hussey, 1990). However, if agents are only effective against specific nematode pests, and their efficacy is dependent on pest densities, then their effective use will require expert advice that may not be available in many developing countries.
The biology, ecology and potential of biological control agents for nematodes have been extensively reviewed in recent years (Kerry, 1987; Stirling, 1991; Sayre and Walter, 1991; Sikora, 1992). Nematologists have identified natural enemies with a range of modes of action similar to those currently studied by plant pathologists for the control of soil-borne diseases. Comments on the advantages and limitations of the major groups of organisms with potential as biological control agents are summarized (Table 3). It must be stressed that several organisms that are effective natural enemies of nematodes in the field, may have limited potential as biological control agents for application by growers. For example Nematophthora gynophila is a causal agent of cereal-cyst nematode decline in many soils, but its limited host range, complex requirements for in vitro culture, and need for soil moisture levels to be at field capacity to ensure infection, mean that its potential use is too limited to warrant further development as a biological control agent.
The stage in the nematode life cycle attacked by the control agent has a profound effect on the damage to the crop and the level of population control. Therefore, trapping fungi and rhizosphere bacteria that attack the second-stage juveniles of cyst and root-knot nematodes may significantly improve crop growth but is unlikely to prevent nematode populations increasing, especially those species that have more than one generation in a growing season. In contrast, those parasites that attack developing females and eggs act like a partially resistant cultivar in that initial nematode invasion and plant damage is not prevented but multiplication of the nematode is significantly reduced. Root colonizing fungi such as the mycorrhizae and endophytic species such as Fusarium spp. may reduce both nematode invasion and development.
Sikora (1992) reviewed a range of control measures, including crop rotation, partial soil sterilization, soil amendments and nematicides, that could be combined to increase the activity of naturally occurring biological control agents. Such measures could also be used to improve the performance of agents added to soil. Partial soil sterilization by methods such as solarization reduces nematode infestations and also reduces the competition from the residual soil microflora, enabling the biological control agent to establish more readily. Soil amendments may also reduce nematode infestations and, by providing an energy source, help to increase numbers of facultative parasites; pre-colonized substrates are most effective in establishing nematophagous fungi in soil. Soils that are naturally suppressive to some nematodes may be used to shorten rotations of susceptible crops and improve the performance of nematicides and resistant cultivars (Kerry, 1990). The use of crop cultivars which are tolerant of nematode attack is likely to be very important for the successful deployment of biological control agents, that provide less effective nematode control than most nematicides. Agents such as rhizosphere bacteria and rhizosphere competent fungi depend on root exudates for their proliferation on roots; exudation from roots differs markedly between plant species and cultivar and affects the efficacy of these agents (Kerry and de Leij, 1992; Sikora, 1992).
TABLE 3
The advantages and limitations of potential biological control agents with different modes of action against plant-parasitic nematodes
|
Type of agent |
Mode of action |
Comments |
|
Facultative parasites Trapping fungi |
Traps produced on modified mycelium give rise to infective trophic hyphae |
Advantages: easily produced in vitro; some species rhizosphere competent; wide host range. |
|
Paecilomyces lilacinus |
Hyphal penetration |
Advantages:easily produced in vitro; rhizosphere competent; attacks the eggs of several nematode species; treatment of planting material (e.g. seed tubers) can be effective. Limitations: requires high soil temperatures; has given variable control in range of conditions; large numbers of propagules (106/g soil) required for nematode control; some isolates are pathogenic to humans. |
|
Verticitlium chlamydosporium |
Hyphal penetration |
Advantages: easily produced in vitro; some isolates rhizosphere competent, and virulent (103 propagules/g soil required for nematode control); resistant resting spores produced; survives throughout growing season in soil. Limitations: seed treatments ineffective; efficacy dependent on nematode species, density and plant host. |
|
Obligate parasitesPasteuria spp. |
Adhesive spores |
Advantages: most isolates highly virulent; infective spores resistant to drying; good shelf-life; reduce infectivity of nematodes as well as fecundity. |
|
Hirsutella spp. |
Adhesive spores |
Advantages: relatively easy to culture in vitro; attack infective nematodes in soil. Limitations: poor saprophytic competitor; limited spread in soil. |
|
Rhizosphere bacteria |
Toxins or modification of root exudates |
Advantages: easy to culture in vitro; can be applied as seed treatments; reduce plant damage. |
|
Endophytic fungi(non-pathogenic root-infecting fungi and mycorrhizae) |
Competition in roots and modification of root exudates |
Advantages:include agents with potential to control migratory endoparasitic nematodes in roots; may improve plant growth even in absence of nematodes; reduce damage caused by wide range of nematodes and limit their multiplication; can be mass produced and formulated; could be applied to seeds or transplant material; may reduce fungal root rots. |
Sources: Kerry, 1987; Stirling, 1991; Sikora, 1992.
Soil conditions, nematode species, rate of development and density and host plant all have a considerable effect on the establishment and activity of biological control agents for nematodes. It is essential to understand these interactions for effective control.
In temperate zones soil temperatures range between 10° and 15°C for much of the growing season whereas in warmer climates temperatures of 20° to 25 °C are normal at 10 cm depth. Too often, research workers in temperate climates have tested the effects of biological control agents in pot experiments in greenhouses at temperatures considerably above those which occur in the field; data from such tests are of limited value. Temperature has a direct effect on the growth and sporulation of the agent and on the rate of development of the nematode target. Moisture in soil is rarely limiting for the growth of most fungi but affects spore dispersal, especially of zoospores. Bacteria are more sensitive to soil moisture levels than fungi, but are unlikely to be affected by levels that enable nematodes to remain active. Soil texture and structure influence nematode activity and the growth and spread of micro-organisms. The great bulk of soil (2 500 tonnes per hectare) to plough depth (20 cm) makes the thorough incorporation of biological control agents difficult and broadcast treatments are unlikely to be economic. The use of agents in arable crops is likely to depend on those organisms that proliferate in the rhizosphere from in-row applications or from seed treatments. The residual soil microflora competes with the introduced agent for scarce energy sources and can significantly affect the performance of the agent even when it is added to soil in a pre-colonized substrate.
Most plant-parasitic nematodes remain mobile throughout their life cycle and microbial parasites must produce traps or adhesive spores to infect them. Nematodes with sedentary stages such as cyst and root-knot nematodes may be parasitized from vegetative hyphae in the rhizosphere without the production of specialized infective structures apart from an appressorium. Fungi such as Verticillium chlamydosporium infect female cyst nematodes that are exposed on the root surface for several weeks during their maturation; depending on the time of infection, the fungus has a range of effects including a reduction in the number of mature females and their fecundity, and parasitism of the eggs. Usually only the eggs and not the females of root-knot nematodes are exposed in the rhizosphere, and these complete their embryonic development and hatch within about ten days at 25 °C. Hence, egg-parasitic fungi have less time to kill root-knot nematode eggs compared with those of cyst nematodes. Non-parasitic micro-organisms that degrade soil amendments and release nematicidal compounds, such as the bacterium which degrades chitin to produce ammonia (Spiegel et al., 1991), are likely to kill most nematodes in soil. Similarly, many endophytic fungi may compete with a wide range of nematodes in roots although those which are sedentary are likely to be more susceptible.
The host plant influences the rate of nematode development and fecundity and the establishment of some natural enemies in the rhizosphere. Despite these effects the role of the host plant has rarely been considered in research on the biological control of nematodes.
Too often, biological control agents have failed because they have been used before a basic knowledge of their ecology and biology has been established. The importance of such knowledge can be seen from work at Rothamsted aimed at the development of V. chlamydosporium as a biological control agent for root-knot nematodes. Stirling (1991) describes the design and methods used in the conduct of biological control experiments, which are more complex than those required to assess the efficacy of a nematicide. He identified five key aspects in setting up an experiment to evaluate a biological control agent and these are presented below in a slightly modified form.
· The test organism and any organic amendment should be applied at practical application rates; 0.1 percent w/w soil is equivalent to 2.5 tonnes/ha and should represent a maximum dose. Tests should always be performed in a non-sterilized soil with a natural residual soil microflora.· Appropriate treatments as well as an untreated control should be included if the organism is added with a substrate. These treatments should include the substrate alone, the organism alone, and the autoclaved colonized substrate. Too often, untreated controls are compared only with large applications of the organism and substrate and this does not allow separation of the effects of the agent from the effects of the substrate. In several tests reported in the literature, application of the substrate alone has decreased nematode populations to the same extent as the substrate colonized by the agent, and there is no clear evidence of biological control.
· Population densities of the agent under test should be monitored to ensure that it has survived in soil throughout the period that activity against the nematode target is required. Such monitoring may require the development of selective media, which can be a difficult and time-consuming task.
· Nematode mortality caused by the organism under test should be measured to assess whether differences between nematode population densities in treated and untreated soil relate to the levels of kill caused by the agent. Infection levels are relatively straightforward to estimate for most parasites, but repeated sampling is required to determine total kills. The effects of agents which produce toxins or have indirect effects on nematodes through competition, the modification of root exudates or the colonization of feeding cells, can only be measured by assessing their impact on nematode development.
· The impact of the soil environment, host plant and nematode should be tested as these are likely to affect the efficacy of the biological control agent, and could account for the lack of activity of potential agents in specific test conditions.
In a review of the literature less than 15 percent of experiments purporting to demonstrate biological control caused by Paecilomyces lilacinus satisfied the above criteria (Kerry, 1990). Although this is an unsatisfactory situation that must be remedied, the difficulties in conducting carefully controlled and monitored experiments should not be underestimated.
Isolates of V. chlamydosporium differ markedly in their growth and sporulation in vitro (Irving and Kerry, 1986), and in their virulence, saprophytic competitiveness and rhizosphere competence (Kerry and de Leij, 1992). Such differences between isolates of the same species of micro-organism are common and there is a need for simple laboratory-based screening methods to select the most promising isolates for further testing. Verticillium chlamydosporium isolates were collected from infested nematode females and eggs in suppressive soils around the world. Hence, they were isolated from the niche in which they would be required to be active as a biological control agent and from soils most likely to yield active isolates. Once isolated in pure culture using standard techniques, the in vitro growth requirements were determined.
Virulent isolates were selected by counting the number of nematode eggs parasitized after exposure to the fungus on agar in a standard test (Irving and Kerry, 1986). Tests for proliferation in the rhizosphere (de Leij and Kerry, 1991) and growth in soil (Kerry, 1991) were also used to select promising isolates. Although simple laboratory-based screens help eliminate many isolates that show insufficient activity to justify further testing, selected isolates will not necessarily be active in the field. As it may take 10 to 16 weeks to investigate adequately the performance of different isolates against cyst and root-knot nematodes in pot tests, relatively few can be screened.
The method of mass culturing of V. chlamydosporium for experiments can have a marked effect on the subsequent survival and proliferation of the fungus in soil. Inoculum produced in shaken liquid cultures consists mostly of hyphae and conidia, which require an energy source to ensure proliferation in soil (Kerry, 1987), whereas on solid media large numbers of chlamydospores are produced and these can be added to soil in aqueous suspension and rapidly establish the fungus (de Leij and Kerry, 1991).
The development of a semi-selective medium (de Leij and Kerry, 1991) has enabled detailed studies to be made on changes in relative abundance of the fungus in soil and on roots. Some isolates of V. chlamydosporium may be extremely abundant in soil but unless they are capable of colonizing the rhizosphere they do not parasitize the eggs of root-knot nematodes. Growth in the rhizosphere differs markedly between plant species, e.g. tomato, cabbage and maize roots support much growth, whereas sorghum, pepper and cotton are poor hosts (Table 4). Colonization by the fungus is confined to the rhizosphere and rhizoplane and there is no spread into root tissue; no lesions have been observed on roots grown in soil treated with V. chlamydosporium and there have been no detrimental effects on the growth of a range of crop species. A tenfold reduction (from 104 to 103 chlamydospores/g soil) in the amount of fungus applied to soil had no effect on the extent of colonization in the rhizosphere (de Leij, Davies and Kerry, 1992); the ability to proliferate on the root surface where the fungus is required to control nematodes is an important characteristic which may allow significant reductions in the amount of inoculum applied to soil.
The efficacy of V. chlamydosporium as a biological control agent for root-knot nematodes is affected by three key factors: the amount of fungus in the rhizosphere (Table 4); the rate of development of eggs in the egg masses; and the size of the galls in which the female nematodes develop. In large galls female root-knot nematodes may produce egg masses which remain within the gall and are not exposed to parasitism by V. chlamydosporium, which is confined to the rhizosphere. Hence, V. chlamydosporium is less effective in controlling root-knot nematodes in heavily infested soils and on highly susceptible crops because large galls are formed on the roots and many eggs escape parasitism. Verticillium chlamydosporium is unlikely to be useful in these situations where a grower would normally apply a nematicide. Also, at temperatures above 25°C eggs may complete their embryonic development and hatch before the fungus has completely colonized the egg mass; at 30°C about 30 percent of eggs of three root-knot species hatched and the second-stage juveniles escaped from the egg mass before the eggs were killed (de Leij, Dennehy and Kerry, 1992). These studies in pot tests, if supported by field experiments, help to define the conditions in which V. chlamydosporium might be used successfully for control of root-knot nematodes. It is only from such detailed studies that the limitations and requirements of the fungus can be assessed.
TABLE 4
The extent of colonization of the rhizosphere by Verticillium chlamydosporium on several plants grown on soil treated with 5 000 chlamydospores/g soil and the control of Meloidogyne incognita
|
Host plant |
Rhizosphere colonization |
Control (%) |
|
Tomato |
229 |
58 |
|
Maize |
216 |
71 |
|
Cabbage |
199 |
68 |
|
Potato |
88 |
79 |
|
Tobacco |
43 |
40 |
|
Pepper |
29 |
36 |
|
Sorghum |
25 |
0 |
|
Wheat |
15 |
5 |
1 cfu are the colony-forming units which develop on the semi-selective medium (methods described in de Leij and Kerry, 1991).Source: de Leij, 1992.
Verticillium chlamydosporium is not a replacement for nematicides but, despite its limitations, it may be a useful management tool. In most tests biological control agents have been applied to protect susceptible crops. However, application of the fungus to a relatively poor host for the nematode on which small galls are produced so that most egg masses are exposed in the rhizosphere might provide more effective control; soil population densities of the nematode would be reduced to non-damaging levels before a susceptible crop was planted. Bridge (1987) recommends the following rotation for the management of root-knot nematodes: susceptible host - poor host - poor host - resistant or non-host -susceptible crop. Application of the fungus before the first or second poor host may permit the shortening of rotations, or the replacement of the resistant or non-host crop, without increasing nematode damage to the susceptible crops. Currently, at Rothamsted Experimental Station, United Kingdom, poor hosts for root-knot nematodes are being screened for their ability to support V. chlamydosporium in their rhizospheres. It is hoped that by combining the host plant and the biological control agent, more effective and consistent control can be achieved.
The past 20 years have seen a significant increase in the number of scientists involved in research on the biological control of nematodes. Surveys and empirical tests are being replaced by quantitative experimentation and basic research on the modes of action, host specificity and epidemiology of selected organisms. Such basic information is essential for a realistic appraisal of the impact of molecular biology on the improvement of microbial agents and monitoring the spread and survival of released organisms, and for the development of rational strategies for control.
Current experience suggests that biological control agents will not replace the use of nematicides but, integrated with other control measures including chemicals, they could play an important role in the development of integrated control strategies in both developed and developing agriculture. The urgent need to reduce the dependence on nematicides should provide the necessary impetus for the considerable amount of research and development still required to ensure the successful use of such agents.
REFERENCES
Bridge, J. 1987. Control strategies in subsistence agriculture. In R.H. Brown & B.R. Kerry, eds. Principles and practice of nematode control in crops, p. 389-420. Sydney, Australia, Academic Press.
Davies, K.G., de Leij, F.A.A.M. & Kerry, B.R. 1991. Microbial agents for the biological control of plant-parasitic nematodes in tropical agriculture. Tropical Pest Management, 37: 303-320.
Deacon, J.W. 1991. Significance of ecology in the development of biocontrol agents against soil-borne plant pathogens. Biocon. Sci. Technol., 1:5-20.
de Leij, F.A.A.M. 1992. Significance of ecology in the development of Verticillium chlamydosporium as a biological control agent against root-knot nematodes (Meloidogyne spp.). University of Wageningen, the Netherlands. (Ph.D. thesis)
de Leij, F.A.A.M. & Kerry, B.R. 1991. The nematophagous fungus, Verticillium chlamydosporium, as a potential biological control agent for Meloidogyne arenaria. Revue Nématol., 14: 157-164.
de Leij, F.A.A.M., Davies, K.G. & Kerry, B.R. 1992. The use of Verticillium chlamydosporium and Pasteuria penetrans alone and in combination to control Meloidogyne incognita on tomato plants. Fund. Applied Nematol., 15: 235-242.
de Leij, F.A.A.M., Dennehy, J.A. & Kerry, B.R. 1992. The effect of temperature and nematode species on interactions between the nematophagous fungus Verticillium chlamydosporium and root-knot nematodes (Meloidogyne spp.). Nematologica, 38: 65-79.
Hussey, N.W. 1990. Agricultural production in the third world - a challenge for natural pest control. Exp. Agric., 26: 171-183.
Irving, F. & Kerry, B.R. 1986. Variation between strains of the nematophagous fungus, Verticillium chlamydosporium Goddard. II. Factors affecting parasitism of cyst nematode eggs. Nematologica, 32: 474-485.
Kerry, B.R. 1982. The decline of Heterodera avenae populations. EPPO Bulletin, 12: 491-496.
Kerry, B.R. 1987. Biological control In R.H. Brown & B.R. Kerry, eds. 92 Biological control of nematodes: prospects and opportunities Principles and practice of nematode control in crops, p. 233-263. Sydney, Australia, Academic Press.
Kerry, B.R. 1988. Fungal parasites of cyst nematodes. Agric. Ecosys. Environ., 24: 293-305.
Kerry, B.R. 1990. An assessment of progress towards microbial control of plant-parasitic nematodes. J. Nematol., 22: 621-631.
Kerry, B.R. 1991. Methods for studying the growth and survival of the nematophagous fungus, Verticillium chlamydosporium Goddard, in soil. In B.R. Kerry & D.H. Crump, eds. Methods for studying nematophagous fungi. IOBC/WPPRS Bulletin XIV/2: 34-38.
Kerry, B.R., Crump, D.H. & Mullen, L.A. 1982. Studies of the cereal-cyst nematode, Heterodera avenae under continuous cereals, 1975-1978. II. Fungal parasitism of nematode eggs and females. Ann. Appl. Biol., 100:489-499.
Kerry, B.R. & de Leij, F.A.A.M. 1992. Key factors in the development of fungal agents for the control of cyst and root-knot nematodes. In Biological control of plant diseases, p. 139-144. London, Plenum.
Oostendorp, M. & Sikora, R.A. 1989. Seed treatment with antagonistic rhizobacteria for the suppression of Heterodera schachtii early root infection of sugar beet. Revue Nematol., 12: 77-83.
Sayre, R.M. & Walter, D.E. 1991. Factors affecting the efficacy of natural enemies of nematodes. Ann. Rev. Phytopathol., 29: 149-166.
Sikora, R.A. 1992. Management of the antagonistic potential in agricultural ecosystems for the biological control of plant-parasitic nematodes. Ann. Rev. Phytopathol., 30: 245-270.
Spiegel, Y., Cohn, E., Galper, S., Sharon, E. & Chet, I. 1991. Graduation of a newly isolated bacterium, Pseudomonas chitinolytica sp.nov., for controlling the root-knot nematode Meloidogyne javanica. Biocon. Sci. Technol., 1: 115-125.
Stirling, G.R. 1991. Biological control of plant-parasitic nematodes. Wallingford, UK, CAB International. 282 pp.
Thomason, I.J. 1987. Challenges facing nematology: environmental risks with nematicides and the need for new approaches. In JA. Veech & D.W. Dickson, eds. Vistas on nematology, p. 469-476. Hyattsville, USA, Society of Nematologists.
