Protecting Chestnut Trees from Blight

PP044 (11/97)

By Dr. Sandra L. Anagnostakis
Department of Plant Pathology and Ecology
The Connecticut Agricultural Experiment Station
123 Huntington Street
P. O. Box 1106
New Haven, CT 06504-1106

Telephone: (203) 974-8498 Fax: (203) 974-8502

American chestnut trees (Castanea dentata) were once an important part of the forests of the eastern United States (Anagnostakis, 1987). The chestnut blight fungus (Cryphonectria parasitica) entered the United States on Japanese chestnut trees (C. crenata) imported before the turn of the century, and is now throughout the native range of American chestnuts (Maine to Georgia, along the Appalachian Mountain range) (Anagnostakis, 1992). European chestnuts (C. sativa), once extensively planted here for their nuts (Anagnostakis, 1989), are also very susceptible to this fungus disease and large orchards were wiped out by the blight epidemic. Some of the cultivars of Chinese chestnut (C. mollissima) that were planted as orchard replacements, also suffer from blight, as do some hybrids selected for nut size and quality. One of the long-term solutions to this problem is breeding for blight resistance, while selecting for superior orchard trees for different climates. Such work is in progress, but even keeping trees alive for breeding can be difficult.

Chestnut blight cankers on American chestnut trees are usually easily recognized, because the thin bark of young trees or sprouts becomes orange where the fungus has grown. Thick bark may have dots of orange, fungal stromata in the fissures, but the most telling signs of cankers in such bark are the epicormic sprouts that form below the canker when the cambium is killed. Sprouts under circles of dead, cracking bark on branches of partially resistant Chinese hybrid chestnuts help locate cankers on such trees.


When homeowners call me for advice about controlling blight I always suggest that they try mudpacks. This is a method developed by W. H. Weidlich (1978) when he was at Michigan State University, and it works very well. All that is needed for this treatment is some soil that has not been treated with pesticides (such as forest soil, or compost), some plastic (a bag, cut open, or an old drop-cloth or shower curtain) and some string. Water is added to turn the soil into thick, gooey mud, and this is plastered all over the blight canker. Wrapping the plastic around the trunk or branch keeps the mud damp, and the plastic can be secured with string. After two months, the plastic can be removed (the tree bark should have started to swell with callus growth), and the blight fungus in the canker should be dead. Although this canker is "cured" the tree will continue to become infected in other places, so repeated mudpack treatments will be necessary. This is obviously not an ideal solution for someone who has many trees.


Hypovirulence is a disease of the blight fungus that is caused by a virus, and was first described by French scientist J. Grente in 1965. The virus keeps the fungus from killing trees, and can be passed through the fungal population, from one individual to another, when genetically similar strains fuse. As hypovirulent strains spread through the chestnut orchards of Italy and France, the trees began to survive longer, "healing" over the blight cankers with lumpy bark tissue (Grente and Berthelay-Sauret, 1978; Heiniger and Rigling, 1994). This spurred searches in this country for hypovirulent strains, and similar "healing" cankers were found in Michigan, Tennessee, Virginia, and West Virginia, which yielded strains of the blight fungus that were also less able to kill chestnut trees (Day et al., 1977). Virologists have now confirmed that these American strains, and the European hypovirulent strains of C. parasitica are infected with viruses of three very different types (three "families"), with wide variation in their effects on the fungus, and with varied effectiveness for biological control of chestnut blight (Choi and Nuss, 1992; B. I. Hillman, personal communication).

When scientists put bits of a hypovirulent blight fungus into holes in the bark around killing cankers, viruses can move into the virulent strains that caused the cankers. The cankers then stop expanding, and the tree's natural defenses of walling off invaders succeeds in protecting the tree's living cambium. Once hypovirulence has been established in a chestnut blight population, hypovirulent spores are moved around in orchards and in the forest by every creature that moves up and down the trees. It is likely that everything that walks across the cankers can pick up virus-infected spores on feet, fur, feathers, and beaks, and move them to new cankers (Sharf and DePalma, 1981). We know that carpenter ants will feed on the fungal stromata and carry live spores and fungal particles to feed comrades (Anagnostakis, unpublished). Tree-climbing slugs migrate up and down the trees, and many tiny insects are found in and around chestnut blight cankers (Turchetti and Chelazzi, 1984; Wendt et al., 1983).

In 1972, Grente sent us hypovirulent cultures, which we tested in the laboratory and the greenhouse. When we had the permission of the U.S. Plant Quarantine Division (in 1973), tests were begun on trees growing outside at The Experiment Station Farm in Hamden, Connecticut. Richard Jaynes planted an orchard of 3-year-old American chestnut trees in 1976, using seedlings raised by E. Thor of the University of Tennessee. The seed for these came from trees in Michigan and Wisconsin; outside the native range of American chestnut, where chestnut blight had not yet killed all of the mature trees. The trees became infected with the chestnut blight fungus, and were treated by Jaynes and J. E. Elliston for four years in a row (from 1978 through 1981) with a mixture of hypovirulent strains of the blight fungus. No cankers on these trees have been treated for the last 14 years. In 1981, 43 of the original 71 main stems were still alive. In 1995, 28 of the original stems survived, and there were 28 sprout-clumps with stems that reached the canopy. New cankers form each season, but usually become superficial (and non-lethal) before they kill any cambium. When cankers kill twigs or branches, they are pruned off in late July (the best time to prune chestnuts to avoid infections in the wounds).

In 1980, Elliston recorded 40 cankers on 59 trees; in 1995 they are uncountable. Many stems have cankered and calloused bark from their base up to 15 feet above the ground. Most of the cankers in this orchard have the appearance of hypovirulent cankers, and we assume that animals, birds, insects, and arthropods are moving hypovirulent strains of the fungus from cured cankers to newly-formed, killing cankers.

Some trees seem to be surviving better than others. Since we have no records of the exact source of each of these trees, it is not possible to draw conclusions about resistance. Perhaps some trees are simply better able to thrive even though they are covered with hypovirulent cankers. Nut production is heavy, and since 1989, these trees have been used in a back-cross breeding program.

It is clear that hypovirulence is not succeeding as well at controlling chestnut blight in West Virginia as it is in Connecticut (William MacDonald, unpublished). I have thought a lot about the differences between our results, and several possibilities occur to me. First, American chestnut trees may be different in each population (provenance) in the United States. Second, there are clearly many different kinds of viruses in the blight fungus population in West Virginia, but we have never found "native" viruses in our C. parasitica cultures from Connecticut. If these other viruses interfere with the infection of the fungus by hypovirulence viruses it could certainly reduce the possibility of a biological control. The third "southern-factor" is the presence, in states from Pennsylvania south, of two fungal species that are closely related to the blight fungus, and neither has ever been found in Connecticut. One is Endothia gyrosa, found on oaks and chestnuts, and the other is Cryphonectria radicalis. The latter has not been found in the U.S. since chestnut blight reached its native range. These fungi might be interacting (or might have interacted) in some way that we have not yet discovered.

Hypovirulent strains have been widely used in the U.S., since their use and transport became unrestricted, and scientists in at least 23 states have used them to treat chestnut trees with chestnut blight.

Genetically Engineered Hypovirulent Strains

About 10% of the asexual spores (conidia) produced by our "natural" hypovirulent strains do not contain the virus. This means that there is always virulent inoculum available to cause killing cankers. In addition, the large number of strains of the fungus present in the U.S. prevents easy transmission of viruses from one strain to another in our forests and orchards. This strain difference can be partially overcome by using mixtures of different strains, all containing the hypovirulence viruses, as Jaynes and Elliston did in the American chestnut orchard at The Experiment Station farm in Connecticut (Jaynes and Elliston, 1980; Turchetti and Maresi, 1988). An even better solution, would be to have hypovirulence viruses enter the spores produced after mating, when many new strain types may be produced by genetic recombination. Our newest tools in fighting the blight are new strains of the blight fungus with the genes of a hypovirulence virus inserted in among the genes of the fungus (Fig. 1). Donald Nuss, Gil Choy, and Bao Chen have produced these strains and studied them in the laboratory (Chen et al., 1993), and Nuss and I are now testing them in a forest in Connecticut for biological control of chestnut blight. In Nuss’s new strains, the virus genes are carried with the fungal genes, and half of the sexual spores produced by a mating of an engineered strain and a normal, killing strain will be hypovirulent. In addition, all of the conidia produced by these strains contain the virus. Viruses are produced in the cells, just as they are in other hypovirulent strains, and these can pass from one strain to another after contact, if the strains are similar.

Since these strains were produced by genetic engineering, their use is restricted by state and federal regulations. We first tested them in an insect-proof greenhouse on American chestnut trees and a variety of other woody plants (eleven species) growing in pots. The strains survived, but did not cause any disease. We considered the pitfalls of using such strains on chestnut trees in the forest, and decided that there should be no problems. If the virus genes were somehow rejected by the fungal strains, the resulting strains would be two killing strains that are already present in the forests of Connecticut. We applied for state and federal permission, and were allowed to start the work in June 1994. We used two engineered, hypovirulent strains to treat blight cankers on American chestnut trees in northwestern Connecticut. Some trees had spores of the strains painted onto cankers (Fig. 2), and other trees were inoculated with engineered strains above, and on the sides of cankers (Fig. 3). By November 1994, we had proof that cankers on five of the twelve treated trees were already producing sexual spores containing virus genes. Now we must regularly check these trees, and others in the plots, to see whether the engineered, hypovirulent strains are more efficient at controlling chestnut blight.

Our long-term plan for usable timber trees and orchard trees in the U.S. requires production of a variety of hybrids by careful breeding. These should be suited to different climates and soil types. We also want to reduce the overall virulence of the chestnut blight fungus, either by the presence of hypovirulence viruses in the fungal population, or by the eventual dispersal of strains with the viral genes in the fungal nuclei. All of these things working together should allow us to cope with the disaster begun when the blight fungus gained a foothold here in the last century.


1. Anagnostakis, S. L. 1987. Chestnut blight: The classical problem of an introduced pathogen. Mycologia 79:23-37.

2. Anagnostakis, S. L. 1989. An historical reference for chestnut introductions into North America. Annual Report of the Northern Nut Growers Assn. 80:132-143.

3. Anagnostakis, S. L. 1993. Chestnuts and the introduction of chestnut blight. Annual Report of the Northern Nut Growers Assn. 83:39-42.

4. Chen, B., G. H. Choi, and D. L. Nuss. 1993. Mitotic stability and nuclear inheritance of integrated viral cDNA in engineered hypovirulent strains of the chestnut blight fungus. EMBO J. 12:2991-2998.

5. Day, P. R., J. A. Dodds, J. E. Elliston, R. A. Jaynes, and S. L. Anagnostakis. 1977. Double-stranded RNA in Endothia parasitica. Phytopathology 67:1393-1396.

6. Grente, J. 1965. Les formes Hypovirulentes d'Endothia parasitica et les espoirs de lutte contre le chancre du châtaignier. C. R. Hebd. Séances Acad. Agr. France 51:1033-1037.

7. Grente, J. and S. Berthelay-Sauret. 1978. Biological control of chestnut blight in France. pp 30-34. In: Proceedings of the American Chestnut Symposium. Eds. W. L. MacDonald, F. C. Cech, J. Luchock, and C. Smith. West Virginia University, Morgantown. 122 pp.

8. Heiniger, U. and D. Rigling. 1994. Biological control of chestnut blight in Europe. Annu. Rev. Phytopathol. 32:581-599.

9. Jaynes, R. A. and J. E. Elliston. 1980. Pathogenicity and canker control by mixtures of hypovirulent strains of Endothia parasitica in American chestnut. Phytopathology 70:453-456.

10. Sharf, S. S. and N. K. DePalma. 1981. Birds and mammals as vectors of the chestnut blight fungus (Endothia parasitica). Can. J. Zool. 59:1647-1650.

11. Turchetti, T. and G. Chelazzi. 1984. Possible role of slugs as vectors of the chestnut blight fungus. Eur. J. For. Path. 14:125-127.

12. Turchetti, T. and G. Maresi. 1988. Mixed inoculum for the biological control of chestnut blight. Bulletin OEPP 18:67-72.

13. Weidlich, W. H. 1978. A preliminary report on a method of biological control of the chestnut blight not involving the use of a hypovirulent strain of Endothia parasitica. pp 79-83. In: Proceedings of the American Chestnut Symposium. Eds. W. L. MacDonald, F. C. Cech, J. Luchock, and C. Smith. West Virginia University, Morgantown. 122 pp.

14. Wendt, R. J. Weidhaas, G. J. Griffin, and J. R. Elkins. 1983. Association of Endothia parasitica with mites isolated from cankers on American chestnut trees. Plant Disease 67:757-758.


Chestnut trees with blight cankers can be cured with mud packs applied to each canker, or protected with a biological control based on a virus that keeps the blight fungus from killing trees.