African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 6, Num. 2, 1998, pp. 179-188

African Crop Science Journal, Vol. 6. No. 2, pp. 179-188, 1998

Fungal pathogens for biological control of Striga hermonthica on sorghum and pearl millet in West Africa

A. A. ABBASHER, D.E. HESS¹ and J. SAUERBORN²

ICRISAT-Niger, B.P. 12404, Niamey, Niger, and University of Giessen, Tropical Crop Science, Schottstr. 2, 35390 Giessen, Germany
¹ICRISAT-Mali, B. P. 320, Bamako, Mali
²University of Gie1en, Tropical Crop Science, Schottstr. 2, 35390 Giessen, Germany

(Received 3 December, 1997; accepted 10 March, 1998)

Code Number:CS98020
Sizes of Files:
      Text: 47K
      Graphics: Line drawings and tables (gif) - 49K

ABSTRACT

The potential to employ natural enemies of Striga hermonthica as biocontrol agents in West Africa was evaluated in a collaborative project between the University of Giessen and International Crops Research Institute for the Semi-Arid Tropics (ICRISAT). Indigenous fungal plant pathogens in West Africa were surveyed and pathogenicity and host specificity of the isolates to S. hermonthica were evaluated. Numerous fungal genera were found to attack Striga including Alternaria spp. (A. alternata, A. chlamydospora), Aspergillus spp., Bipolaris spp., Curvularia spp. (C. eragrostidis and C. lunata), Drechslera spp., Fusarium spp. (F. equiseti, F. oxysporum, and F. verticillioides), Macrophomina phaseolina, Paecilomyces lilacinus, Phoma sorghina, Rhizopus spp., and Verticillium lecanii. Fusarium was the prevalent genus, with species isolated from more than 90% of diseased Striga samples collected in Burkina Faso, Mali, and Niger. Fusarium oxysporum was the predominant species. Pathogenicity tests conducted in pots in the glasshouse and out doors confirmed the pathogenicity of Fusarium species to S. hermonthica. Inoculum of these isolates applied pre-sowing at a rate of 5 g kg^-1 of soil resulted in reduced Striga emergence and increased shoot and grain yield of millet and sorghum. In specificity tests, no isolates attacked hosts other than Striga.

Key Words: Biocontrol, fungi, pathogenicity, Striga

La possibilité d'utiliser des ennemis naturels du Striga dans la lutte biologique a été évaluée dans le cadre d'un projet de collaboration entre l'Université de Giessen et l'ICRISAT. Des souches endémiques de champignons ont été collectées et leur capacité d'attaquer le Striga ainsi que leur spécificité pour cet h(tm)te ont été évaluées. Une large gamme de champignons s'attaquent au Striga, parmi eux: Alternaria spp. (A. alternata, A. chlamydospora), Aspergillus spp., Bipolaris spp., Curvularia spp. (C. eragrostidis et C. lunata), Drechslera spp., Fusarium spp. (F. equiseti, F. oxysporum, et F. verticillioides), Macrophomina phaseolina, Paecilomyces lilacinus, Phoma sorghina, Rhizopus spp., et Verticillium lecanii. Le genre le plus commun était le Fusarium, trouvé dans plus de 90% des échantillons prélevés au Burkina Faso, au Mali et au Niger. L'espèce la plus fréquente fut le F. oxysporum. Des tests de pathogénicité conduits dans des pots en serre et à l'air libre ont confirmé le pouvoir pathogène de plusieurs de ces isolats vis-à-vis de S. hermonthica. Des inoculums préparés à partir de ces isolats et appliqués avant le semis à un taux de 5 g kg-1 de sol ont été réduit à l'émergence du Striga et conduit à une amélioration du rendement en paille et en grains du petit mil et du sorgho. Aucun isolat qui attaque une plante autre que le Striga n'a été observé.

Mots clés: Lutte biologique, champignons, pathogénécité, Striga

The parasitic weed Striga hermonthica (Del.) Benth. is a serious constraint to cereal production in Sub-Saharan Africa. Sorghum (Sorghum bicolor [L.] Moench), maize (Zea mays L.), pearl millet (Pennisetum glaucum [L.] R. Br.) and rice (Oryza sativa L.) as well as sugar cane (Saccharum officinarum L.) are attacked by the parasite. Twenty-one million hectares of cereals in Africa are estimated to be infested by Striga, leading to an estimated annual grain loss of 4.1 million t (Sauerborn, 1991). In some locations and years, Striga infestation may lead to total crop failure. Severe infestation may make continued cultivation of cereals difficult, forcing farmers to change cropping strategy or abandon their fields (Debrah, 1994).

Considerable research has been done on the biology and control of Striga. However, conventional approaches have had limited impact, particularly when employed singly. It is now recognised that an integrated control approach is required (Berner et al., 1996; Hess et al., 1996). One of the components of integrated Striga management could be the use of fungal antagonists.

The virulence of parasitic weeds, their regional occurrence, and their parasitic life-style render them ideal targets for biocontrol using a mycoherbicidal approach. Two mycoherbicides are already in use to control parasitic weeds: Product F, employing Fusarium oxysporum f. sp. orthoceras, developed in the former Soviet Union, is used against broomrape (Orobanche aegyptiaca) in tomato, melon, and cabbage (Panchenko, 1991); and Luboa II, employing Colletotrichum gloeosporioids f. sp. cuscutae was developed and used against dodder (Cuscuta chinensis) in China (Zhang, 1985).

In West Africa, several plant pathogenic fungi have been isolated from diseased S. hermonthica and were evaluated as possible biocontrol agents against Striga (Zummo, 1977; Charudattan, 1985; Abbasher et al., 1995). However, to employ plant pathogens as biocontrol agents, they should have several features including easy production of inoculum, survival, sporulation and infectivity at a wide range of temperature and relative humidity, and non-pathogenicity to crops.

The present study was done to survey native Striga pathogens in Burkina Faso, Mali, and Niger and to explore the feasibility of using some of them as biocontrol agents to control Striga hermonthica in sorghum and pearl millet.

Surveys. In Burkina Faso, Mali, and Niger, field surveys for pathogens of S. hermonthica were carried out from August through October 1995. Samples were collected from 24 fields at 11 locations in Burkina Faso, 33 fields at 13 locations in Mali, and 69 fields at 23 locations in Niger. Striga plants with probable symptoms of infection including browning, wilting, lesions or death were collected. They were brought to the laboratory where they were dried between sheets of newspaper at room temperature for long-term storage.

Isolation, culture, and identification of fungi. Diseased Striga plant parts were cut into small pieces (ca. 5-10 mm), which were surface-sterilised with 70% ethanol for 3 min., rinsed 3 times in sterile distilled water, dried with sterilised filter paper and placed on water agar (15 g of agar in 100 ml of distilled water). Pure cultures were prepared by hyphal-tip isolation and maintained on potato dextrose agar (PDA: Sera, Heidelberg, Germany) or special nutrient poor agar (SNA) in tubes according to Nirenberg (Nirenberg, 1976). All fungal isolates were identified based on shape and colour of the sporophores (fruiting bodies) and the shape, size, colour, and placement of spores on the sporophores. Identifications of fungi were verified by the Federal Biological Research Centre for Agriculture and Forestry, Berlin, Germany.

Pathogenicity to Striga plants. Sorghum was grown in 13 cm plastic pots in the glasshouse at ICRISAT Sahelian Centre, Niamey, Niger from October through December 1995. Pots were filled with 5 kg of soil mix (sterilised manure, sand, and clay, 2:5:1, v/v/v). Five replicates were used. Prior to sowing, pots were infested by mixing 16,000 viable Striga seeds into the soil layer 5-10 cm below the surface. Eight weeks after planting, when Striga plants were 5-15 cm tall, they were inoculated with different fungal isolates at a rate of 20 g of inoculum per pot using 15-day-old inoculum produced on sterilised sorghum seeds. Inoculum was placed around the base of Striga plants at different stages of growth (5-25 cm). Following inoculation, Striga plants were assessed for disease symptoms at 3-day intervals for a month. Fungal isolates were re-isolated from diseased Striga onto PDA to test Koch's Postulates.

Pathogenicity to Striga (seeds and plants). Fusarium isolates collected from the three locations were screened for their pathogenicity to Striga and also for their effect, if any, on host crop performance. The experiments were carried out from October 1995 through July 1996. Temperature and relative humidity in the glasshouse ranged between 28 to 451/4C and 25 to 50%, respectively. In the first experiment, nine isolates were evaluated on a population of Striga adapted to pearl millet and 10 isolates on a population of Striga adapted to sorghum. In the second experiment, 79 isolates were evaluated for pathogenicity to sorghum Striga. Inoculum for all trials was prepared on sterilised chopped millet straw as described by Abbasher and Sauerborn (1992).

Plastic pots of 28-cm diameter were used in the first experiment and filled with 10 kg of soil mix containing sterilised manure, sand, and clay (2:5:1, v/v/v). The 13-cm diameter pots employed in the second experiment were filled with 5 kg of the same soil mix. Prior to sowing sorghum or pearl millet, pots were infested by mixing 16,000 viable Striga seeds into the soil layer 5-10 cm below the surface. Inoculum of the biocontrol fungus was subsequently applied to the infested soil at a rate of 5 g kg-1 of soil. Six sorghum (variety Mota Maradi) or pearl millet (landrace Bengou local) seeds were sown per pot. Controls consisted of sorghum and pearl millet grown in fungus uninoculated pots with and without Striga infestation. Irrigation was applied when necessary. Seven days after sowing, host plants were thinned to one per pot. To control insects and mites on the host, bifenthrin [2, methyl (1,1-biphenyl) ethyl 3-(2-chloro-3,3,3-trifluro-1-propenyl)-2-2-dimethylcyclopropan-ecarboxylate] was applied weekly. Experiments lasted for four months. Parameters assessed to study the effect of Fusarium isolates on crop performance and Striga were: Striga incidence (weekly counts of emerged Striga plants), host plants heights (measured every 2 weeks after the Striga emergence in the first pot), and dry weight of sorghum and millet shoots at harvest (determined after oven drying at 60íC for 7 days). Data were analysed using the ANOVA procedure.

Outdoor pot experiments. The effect of environmental factors on pathogenicity of 11 Fusarium isolates to S. hermonthica, using sorghum and millet as host crops was investigated in an outdoor pot trial (between June and September 1996) at ICRISAT Sahelian Centre. The trial differed from the previous one in that an unsterilised soil mix was used to fill 28-cm-diameter pots to which 10 g of inoculum was applied per kg of soil (total of 100 g inoculum per pot). The same parameters were measured and the analyses was conducted as in the preceding experiments.

Host range. The reaction of 10 crop species [sorghum, pearl millet, maize, fonio (Digitaria exilis Kippist), cowpea (Vigna unguiculata [L.] Walp.), cotton (Gossypium hirsutum L.), okra (Hibiscus esculentus L.), sesame (Sesamum indicum L.), groundnut (Arachis hypogaea L.), and tomato (Lycopersicon esculentum Mill.)] to 21 Fusarium isolates was evaluated in pots. Inoculum was prepared as previously described and applied pre-sowing at a rate of 5 g kg^-1 of soil.

Isolation and identification of fungi. Fusarium spp. were isolated from more than 90% of the diseased Striga samples collected in Burkina Faso, Mali and Niger. However, the dominant species (93% of the total Fusarium isolates) was F. oxysporum. Other fungi isolated were: Alternaria spp. (A. alternata, A. chlamydospora), Aspergillus spp., Bipolaris spp., Curvularia spp. (C. eragrostidis and C. lunata), Drechslera spp., Fusarium spp. (F. equiseti, F. verticillioides), Macrophomina phaseolina, Paecilomyces lilacinus, Phoma sorghina, Rhizopus spp., and Verticillium lecanii.

Pathogenicity of Fusarium spp. to Striga. Young Striga plants were generally more susceptible to infection and were killed more rapidly than older plants. However, isolates differed in pathogenicity to emerged Striga. Certain Fusaria were effective in killing Striga at the flowering stage. The highest infection rate (91%) was observed with Fusarium isolate 12-Sh-Mai collected in Mali, whereas isolate 9-1-BH from Burkina Faso produced the highest (71%) Striga mortality rate (Fig. 1).

Figure 1. Percentage of emerged Striga hermonthica plants infected and killed by ten Fusarium isolates applied as soil drench post emergence.

Figure 2. Effect of four Fusarium isolates on Striga emergence and yield of pearl millet in pots maintained in the glasshouse.

Pathogenicity to Striga (seeds and plants). Of the nine Fusarium isolates tested on millet Striga in the first experiment, four were found to reduce Striga infestation on millet by 72-99% (Fig. 2). Pearl millet shoot and panicle dry weights were similar to, or greater than those of the uninfested control (Fig. 2). Of the nine Fusarium isolates tested on sorghum Striga, six reduced S. hermonthica emergence by 75-95% (Fig. 3). Sorghum shoot weight increased by 86%, whereas panicle yield increased by 400-650% compared to the uninfested control (Fig. 3). In the uninfested control, shoot dry weight and panicle dry weight were higher (89% and 14%, respectively) than those in infested control.

In the second experiment, 29 isolates reduced the number of emerged Striga by 61% to 99% (Table 1), whereas the remaining 50 isolates were considerably less effective. On average, the use of fungal pathogens to control Striga increased sorghum shoot dry weight by 89% (58-135%). As expected, the two uninfested controls (sorghum and sorghum with sterilised, uninoculated straw) outperformed the infested control for both shoot and panicle yield. Yield was somewhat suppressed in the control with sterilised, uninoculated straw (Table 1).

Figure 3. Effect of six Fusarium isolates on Striga emergence and yield on Sorghum in pots maintained in the glasshouse.

TABLE 1. Striga hermonthica emergence and sorghum shoot and panicle dry weight as influenced by different Fusanum isolates collected in Burkina Faso (BF), Mali and Niger and applied pre-sowing

Out door pot experiments. Of the 11 isolates tested, four (11-2, 4-3-B, 6-8 and 5-S-Kol.) reduced the total number of emerged Striga by 92, 85, 62 and 48%, respectively, while the remainder had little or no effect on Striga emergence (Fig. 4). Sorghum shoot dry weight was generally enhanced in the presence of the fungal isolates. For isolates 4-3-B and 6-8, Striga-infested sorghum plus the fungal pathogens was as or more productive (panicle yield) than the uninfested control (Fig. 4).

Figure 4. Effect of four Fusarium isolates on Striga emergence and yield of Sorghum in outdoor pots.

Host range. No sorghum, pearl millet, maize, fonio, cowpea, cotton, okra, sesame, groundnut, or tomato plants were affected by pre-sowing application of the Fusarium isolates to the soil. Non-host plants growing in both infested and (uninfested) control pots were identical in appearance and productivity (biomass production). No disease symptoms whatsoever were in evidence. The Fusarium isolates investigated were pathogenic only to Striga.

In the course of our surveys, fungi associated with infected Striga plants were given more attention than other microorganisms (e.g., bacteria). Fungi were favoured since so many are available, they are usually aggressive parasites, and they are generally host-specific (Templeton, 1982; Charudattan, 1985). Most of the fungi encountered are facultative parasites or saprophytes. This character renders them easily cultured in vitro where abundant sporulation permits efficient preparation of inoculum. There is conjecture that soil-borne pathogens may play an important role in "Striga suppressive soils" and there has been interest in locating sites where Striga has declined for otherwise unexplained reasons. In Nigeria, in the course of a 1990 survey (Lagoke et al., 1990), up to 40% of emerged Striga plants were observed to die before reaching the reproductive stage. It was proposed that natural mortality was due to soil-borne pathogens. In recent surveys of fungal antagonists of Striga in West Africa, soil-borne pathogens predominated over foliar pathogens (Abbasher et al., 1995; Ciotola et al., 1995). Among Striga pathogens encountered in northern Ghana, Fusarium spp. were isolated from over 90% of the infected samples (Abbasher et al., 1995). Similarly, in the 1995 Burkina Faso-Mali-Niger survey, Fusarium spp. were widely prevalent with F. oxysporum the predominant species. Ciotola et al. (1995) reported Fusarium spp. to be among the most frequently isolated fungi in an earlier (1991) sampling conducted in the same three countries. Fusarium spp. were also found to attack Striga spp. other than S. hermonthica.

Pathogenicity tests clearly demonstrated the ability of Fusarium spp. to infect and kill seeds and unemerged seedlings of S. hermonthica. Under controlled conditions, Fusarium spp. reduced Striga incidence by up to 99%. Their use as bioherbicides could contribute to reduction in the soil seed bank of this noxious parasitic weed.

Two environmental factors which limit the utilisation and effectiveness of biological control agents are temperature and moisture (Greaves et al., 1989; Heale et al., 1989). Incorporation of inoculum into soil lends some protection from environmental extremes. Under natural conditions in the Sahel, we observed four Fusarium isolates to improve sorghum growth and yield through decreased incidence of S. hermonthica. Since these isolates are indigenous to the region, they also seem to be adapted to the prevalent high temperatures.

Host specificity tests carried out in the glasshouse clearly demonstrated the specificity of the experimental organisms to the Striga host. Further host range tests will be conducted under natural field conditions with isolates that appear most promising for Striga control.

In glasshouse and outdoor pot trials, the ability of certain Fusarium isolates to limit Striga infestation of pearl millet and sorghum hosts has been established. Work is continuing to evaluate the performance of indigenous isolates under natural field conditions in Burkina Faso, Mali and Niger.

The authors are grateful to the Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) and ICRISAT for financial support. We thank Harouna Dodo for the technical assistance.

Abbasher, A.A. and Sauerborn, J. 1992. Fusarium nygamai, a potential bioherbicide for Striga hermonthica control in sorghum. Biocontrol 2:291-296.

Abbasher, A.A., Kroschel, J. and Sauerborn, J. 1995. Microorganisms of Striga hermonthica in Northern Ghana with potential as biocontrol agents. Biocontrol Science and Technology 5:157-161.

Berner, D.K., Alabi, M.O., Di-Umba, U. and Ikie, F.O. 1996. Proposed integrated control program for Striga hermonthica in Africa. In: Advances in Parasitic Plants Research. Moreno, M.T., Cubero, J.I., Berner, D., Joel, D., Musselman, L.J. and Parker, C. (Eds.), pp. 817-825. Dirección General de Investagición Agraria, Sevilla, Spain.

Charudattan, R. 1985. The use of naturally and genetically altered strains of pathogens for weed control. In: Biological Control in Agriculture IPM Systems. pp. 347-372. Academic Press, New York.

Ciotola, M., Watson, A. K. and Hallett, S. G. 1995. Discovery of an isolate of Fusarium oxysporum with potential to control Striga hermonthica in Africa. Weed Research 35:303-309.

Debrah, S. K. 1994. Socio-economic constraints to the adoption of weed control techniques: the case of Striga control in the West African Semi-Arid tropics. International Journal of Pest Management 40:153-158.

Greaves, M.P., Bailey, J. A. and Hargreaves, J. A. 1989. Mycoherbicides: opportunities of genetic manipulation. Pesticide Science 26:93-101.

Heale, J.B., Isaac, J. E. and Chandler, D. 1989. Prospects for strain improvement in entomopathogenic fungi. Pesticide Science 26:79-82.

Hess, D.E., Obilana, A. B. and Grard, P. 1996. Striga research at ICRISAT. In: Advances in Parasitic Plants Research. Moreno, M.T., Cubero, J.I., Berner, D., Joel, D., Musselman, L.J. and Parker, C. (Eds.), pp. 828-834. Dirección General de Investagición Agraria, Sevilla, Spain.

Lagoke, S.T.O., Shebayan, J. Y., Weber, G., Olufajo, O., Elemo, K., Adu, J. K., Emechebe, A. M., Singh, B. B., Zaria, A., Awad, A., Ngawa, L., Olaniyan, G.O., Olafare, S. O., Adeoti, A. A. and Odion, C. 1990. Survey of Striga problem and evaluation of Striga control methods and packages in crops in the Nigerian savanna. In: Improving Striga Management in Africa. Lagoke, S.T.O., Hoevers, R., M'Boob, S.S. and Traboulsi, R. (Eds.), pp. 91-120. FAO, Accra, Ghana.

Nirenberg, H.I. 1976. Untersuchungen über die morphologische und biologische Diffe-renzierung in der Fusarien Sektion. Liseola. Mitt. Biol. Bundesanstalt Land Forstwirtsch. Berlin-Dahlem, Germany 169:1-117.

Panchenko, V. P. 1981. The biological protection of water melons and tomatoes from broomrape in Astrkhan Province. Doklady Vsesoyuznoi Akademii Sel' Skookhozyaistvennykh Nauk imeni VI. Lenina 8:25-27.

Sauerborn, J. 1991. The economic importance of the phytoparasites Orobanche and Striga. In: Proceedings of the 5th International Symposium on Parasitic Weeds. Ransom, J.K., Musselman, L.J., Worsham, D. and Parker, C. (Eds.), pp. 137-143. CIMMYT, Nairobi, Kenya.

Templeton, G. E. 1982. Status of weed control with plant pathogens. In: Biological Control of Weeds with Plant Pathogens. Charudattan, R. and Walker, H.L. (Eds.), pp. 29-44. Wiley, New York.

Zhang, T. Y. 1985. A forma specialis of Colletotrichum gloeosporioides on Cuscuta sp. (in Chinese). Acta Mycologica Sinica 4:223-239.

Zummo, N. 1977. Diseases of giant witchweed, Striga hermonthica in West Africa. Plant Disease Reporter 61:428-430.

Copyright 1998, African Crop Science Society

The following images related to this document are available:

Line drawing images