Black stem rust biology and threat to wheat growers

(from a presentation to the Central Plant Board Meeting
February 5-8, 2001, Lexington, KY)

Kurt J. Leonard
U.S. Department of Agriculture, Agricultural Research Service,
Cereal Disease Laboratory, University of Minnesota, St. Paul 55108,
now Adjunct Professor Emeritus, Plant Pathology Department,
University of Minnesota, St. Paul 55108

Production and sale of various species and cultivars of Berberis (barberry), Mahonia, and Mahoberberis in the nursery trade in the United States are affected by the Black Stem Rust Quarantine, which was imposed to protect wheat crops in regions vulnerable to stem rust epidemics. For producers and distributors of barberry bushes, the quarantine is a burden that naturally raises the questions - Is wheat stem rust still a concern in the quarantined regions? - Is barberry still considered a threat?

To answers these questions requires a brief review of the biology of the wheat stem rust fungus (Puccinia graminis f.sp. tritici), the role of barberry in the stem rust epidemics, and the current status of stem rust in North America. Wheat stem rust is caused by a parasitic fungus that reproduces only in living plants. Symptoms of the disease in wheat consist of erumpent pustules primarily on the stems and leaf sheaths (Fig. 1). Each pustule is the result of an infection by a single rust spore. The initial infections produce no obvious symptoms until about 7-10 days after infection. Then the fungal mycelium, which has been growing within the plant tissue, masses directly under the plant epidermis and begins producing thousands of spores that rupture the epidermis and emerge as powdery, rust colored piles of urediniospores. Each urediniospore has the potential to produce a new infection that will cause similar damage on the same plant or another wheat plant. Multiple cycles of infection, sporulation, and re-infection can produce very destructive epidemics in wheat fields within just a few weeks.

The stem rust life cycle

The complete life cycle of P. graminis involves barberry as well as wheat. Late in their development, the stem rust infections on wheat plants convert from producing urediniospores to production of a new type of spore known as teliospores. Teliospores remain attached to the infected plants and are commonly left in the field on the straw residue from the crop. The teliospores are specialized survival structures for the fungus. They remain dormant in the field until the following spring when they germinate and immediately produce another type of spore known as basidiospores. Basidiospores cannot infect wheat plants. Instead, they infect young leaves of common barberry (Berberis vulgaris) or other susceptible Berberis, Mahonia, or Mahoberberis species or cultivars. On barberry, the resulting infections produce specialized infection structures called pycnia, which play an essential role in the sexual stage of the fungus.

The sexual stage of P. graminis begins with events within the teliospores during their dormant period. Like urediniospores, the teliospores of P. graminis have two nuclei per cell. Each nucleus contains a single set of chromosomes. In these spores the nuclei are paired so that one nucleus is of a + mating type and the other is of a - mating type. During the apparently inactive dormant period of teliospores, the + and - mating type nuclei fuse to produce a single diploid nucleus (two sets of chromosomes). The chromosomes pair and the diploid nucleus quickly begins to divide in meiosis, which eventually results in four haploid nuclei with recombination of genes from the paired chromosomes. During the dormant period, this process is suspended at an intermediate stage of meiosis in which the teliospores pass the winter. In the spring with the onset of warm wet weather, meiosis resumes and the teliospores germinate. Each of the four haploid nuclei per teliospore cell migrate to one of the four developing basidiospores (Fig. 2), where they quickly divide to produce two haploid nuclei per mature basidiospore. The mature basidiospores are forcibly ejected and carried away be air currents. If they reach a susceptible barberry leaf, basidiospores germinate and penetrate the leaf.

The sexual stage of P. graminis, in which nuclei of + and - mating types are reunited, is completed on barberry. Pycnia, which result from infection on young barberry leaves by basidiospores, contain two key elements for the sexual process. Pycniospores are produced in a sugary nectar within the pycnia and function as male gametes. They consist of little more than a nucleus to fertilize the receptive hyphae of another pycnium of compatible mating type. The receptive hyphae function as the female gametes. The sugary nectar released by pycnia helps spread the pycniospores. Insects are attracted to the nectar and often visit several pycnia in succession, fertilizing them much as bees pollinate flowers. Self fertilization is prevented in P. graminis, because only + mating type pycniospores can fuse with - mating type receptive hyphae, and vice versa.

When a pycnium has been fertilized by pycniospores from a mating type compatible pycnium, its cells quickly change from their original haploid uninucleate state to a new dikaryotic condition in which they have paired + and - nuclei in each cell. The fertilized structure becomes an aecium (Fig. 3), which produces chains of aeciospores surrounded by a cup-like enclosure of fungal cells. Like the urediniospores and like the cells of the aecium, each aeciospore contains two nuclei. And like the urediniospores, aeciospores infect wheat, not barberry. Pycnia typically form on the upper side of barberry leaves, and aecia form within 5-7 days after fertilization on the lower side of the leaf directly below each fertilized pycnium.

The life cycle of P. graminis illustrates two reasons for wanting to exclude susceptible barberry plants from major wheat producing regions. First, barberry serves as the bridge to carry the fungus from one wheat crop to the next. During the interval between the harvest of one wheat crop and the emergence of the next wheat crop, P. graminis can survive as dormant teliospores, which through production of basidiospores, may give rise to new infections only on barberry. Without barberry, the teliospores represent a dead end in the fungus life cycle. When barberry bushes grow adjacent to wheat fields, they can serve as foci for the spread of stem rust into the wheat (Fig. 4) where the epidemic is sustained and spread by repeated cycles of infection and production of urediniospores. The second reason for excluding barberry is that, as the site of sexual reproduction of P. graminis, it serves as the breeding ground for new pathogenic races of the fungus. More will be said on this subject later.

Barberry eradication

In 1918 in response to two devastating wheat stem rust epidemics in the early 1900s the U.S. Department of Agriculture coordinated a barberry eradication program for important wheat producing states in regions where barberry presented a serious threat. North Dakota and South Dakota had already begun barberry eradication programs in 1917. Preventing wheat stem rust epidemics was deemed a national priority not only to protect the U.S. food supply, but also to maintain food exports to U.S. allies during the First World War. The barberry eradication program focused on the northern U.S., because P. graminis teliospores either do not germinate or barberry bushes fail to become infected in regions where winters are mild. In the first few years of the barberry eradication program, the efforts were concentrated on removal of barberry hedges and individual bushes planted at farm sites and villages. By the end of 1918, 1,690,000 bushes had been destroyed.

In 1920, the barberry eradication program began a farm to farm survey in 13 states involving 1,900,000 farms in an area of 750,000 square miles. It became apparent from that survey that the common barberry, Berberis vulgaris, had spread from the original planting sites and was rapidly becoming naturalized after its introduction by settlers from Europe. By 1933 over 18 million bushes had been destroyed.in the eradication area. Of these, about 75% were in sites of natural spread away from the original plantings. Most of the bushes destroyed were of B. vulgaris in the north central U.S. from Ohio to Nebraska and the Dakotas (Fig. 5). In addition, the North American species Allegheny barberry (B. canadensis) in Virginia and Colorado barberry (B. fendleri) in Colorado were targeted in their native habitats, because they were found also to be susceptible to the stem rust fungus. As the barberry eradication program progressed other states joined the original 13. Washington joined in 1923; Missouri, Pennsylvania, Virginia, and West Virginia joined in 1935; and Kansas joined in 1955.

By 1930, barberry had been largely eliminated from the vicinity of wheat fields, where it posed an immediate threat, and the frequency of stem rust epidemics in wheat had begun to decline. However, the role of barberry as a breeding ground for new stem rust races had not yet been broken. A new race of P. graminis f.sp. tritici, termed race 15B, was able to overcome all known sources of stem rust resistance and caused devastating epidemics in 1953 and 1954. Race 15B was first found on a barberry bush near Fort Dodge, Iowa in 1939. Although Iowa was not a major wheat producing state in the 1930s, race 15B was able to move from its site of origin and become established in the Great Plains wheat production areas.

By 1972, 98% of the barberry eradication area indicated in Fig. 5 had been cleared of all susceptible barberry plants. To be declared barberry-free, a site on which a fruiting barberry bush had been removed had to be inspected at intervals over a 15 year period to be sure that no new seedlings from dormant seed in the soil survived. With most of the bushes gone, the cost of finding and removing each remaining bush greatly increased. The U.S. Department of Agriculture opted to withdraw funding for the eradication program in 1981 and turned full responsibility for the program to individual states within the barberry eradication area. Those states gradually reduced funding for the program, so that by 1990 the eradication effort had essentially ceased. Most of the states in the eradication program also were part of the barberry quarantine area (Fig. 6). Wyoming joined the quarantine area in 1990. The quarantine is still in effect, and transportation of stem rust susceptible barberry plants into the quarantine area or from state to state within the area is prohibited.

Wheat stem rust epidemics

The impact of barberry eradication was seen first as a reduction in the frequency of significant stem rust epidemics in states of the eradication area. Within the decade from 1918 through 1927, there was an average of 5.1 states per year in which losses to stem rust in wheat exceeded 1%. (Fig. 7). By the following decade, the average frequency of epidemics in the original 13-state eradication area had dropped to 3.0 states per year. Thereafter, the frequency of stem rust epidemics remain fairly constant until after the mid-1950s. The drop in epidemics in the 1920s can be attributed to the effect of removing barberry bushes as primary sources of disease inoculum (as in Fig. 4). However, it was clear that barberry was not the only possible source of stem rust inoculum for wheat in the northern states. The stem rust fungus also survives during winters on fall-planted wheat in the southern U.S. near the coast of the Gulf of Mexico where the winters are very mild. In years with spring weather conducive to rust infection and spread of spores, stem rust spreads north as the weather warms and winter wheat resumes growth until it eventually reaches the fields of spring-planted wheat in the northern U.S. Spread of rust is fastest in the Great Plains states in which there is a nearly continuous belt of wheat production from Texas to the Prairie Provinces of Canada (Fig. 8). The average of 2-3 states per year with significant wheat stem rust epidemics in the barberry eradication area from 1928-1967 represents approximately the frequency of years in each decade in which stem rust is able to spread in significant amounts from the Gulf Coast to the north central U.S. in the absence of barberry.

Of course, during the years from 1928-1967, wheat breeders were busy finding new sources of genes for stem rust resistance and breeding new rust resistant wheat varieties. Unfortunately, the genes for resistance that were used in the new wheat varieties were race-specific. That is, they were effective against common races of the stem rust fungus, but not against some rare races of the fungus. These rare races, which were often unknown at the time, were selected on the new resistant wheat varieties and within a few years were present in great enough amounts to cause major epidemics. Thus, the effect of barberry eradication initially was to reduce the frequency, but not the intensity, of stem rust epidemics, especially in the important spring wheat area of North Dakota, South Dakota, and Minnesota (Fig. 9). In those states, the wheat stem rust epidemics of 1935, 1953, and 1954 were as severe as any that occurred before barberry eradication. In those years, stem rust destroyed up to 50% of the wheat crop in North Dakota and Minnesota and 20% or more of the crop in South Dakota (Figs. 10, 11). Nevertheless, barberry eradication played an essential role in the eventual suppression of wheat stem rust epidemics following 1954 (Fig. 7).

Current status of wheat stem rust

Once the barberry bushes were essentially eliminated from major wheat growing areas, the production of new virulent stem rust races essentially stopped. Without its alternate host for the sexual stage of its life cycle, the stem rust fungus no longer had a source of genetic recombination to reshuffle its genes for virulence. From then on, new races had to arise as mutations, which are extremely rare and which change only one virulence at a time. Gradually the number of races in the stem rust population dwindled as selection weeded out the less fit races until only a few were left in the rust population surviving by spread out of the Gulf Coast region each spring. This made it much easier for wheat breeders to put together combinations of race specific resistance genes in new wheat varieties that would remain effective for decades. In turn, the stem rust population declined until in recent decades it has become difficult to find and collect rusted wheat samples in stem rust surveys carried out by the Cereal Disease Laboratory (Fig. 12). In some years, fewer than five wheat fields were found with stem rust among hundreds of fields surveyed throughout the Great Plains each year. Surveys now rely on wheat breeders nurseries and special trap nurseries of stem rust susceptible wheat varieties to monitor the rust races present in the Great Plains each year.

Hard red spring wheat cultivars currently grown in the northern Great Plains and hard red winter wheat cultivars grown in the central and southern Great Plains commonly have combinations of two or more genes for race-specific resistance that are effective against known stem rust races that still exist in the Great Plains (Table 1). Races TPMK and QFCS, which have been among the most common races in the Great Plains, have been controlled effectively by several different resistance genes that are generally found in current wheat varieties. In addition, other genes for resistance are known and are available for use against other races should they increase in frequency. Therefore, the future looks bright for controlling wheat stem rust. But this bright future depends on the continued absence of barberry from the wheat producing areas.

Barberry and the future of wheat stem rust

As an example of what could happen if barberry returned and the wheat stem rust fungus were to change from the current asexual population back to a sexual population, we can consider the current situation with a related rust, oat crown rust. Oat crown rust produces its sexual stage on the alternate host common buckthorn (Rhamnus cathartica), a species introduced from Europe that has become naturalized and extremely common in the north central U.S. Hundreds of races of oat crown rust occur in the oat crown rust population, and oat varieties with race specific resistance to crown rust rarely remain resistant for more than 5 years. It is virtually impossible for oat breeders to produce varieties with resistance that will remain highly effective against crown rust for more than a few years. See Table 2 for a comparison of diversity found in three recent years of surveys of wheat stem rust and oat crown rust. In the years 1997-1999 the Cereal Disease Laboratory made 42 to 73 collections of wheat stem rust from which they identified only 6 to 8 races each year (Table 2). In the same years, the Cereal Disease Laboratory made 102-217 collections of oat crown rust from which they identified 80 to 134 races each year. Almost every new collection of crown rust yielded a new race. The great diversity of races in oat crown rust is directly due to the abundance of common buckthorn, its alternate host for the sexual stage of crown rust. If common barberry ever becomes as abundant as common buckthorn in the U.S., it will become virtually impossible to control wheat stem rust epidemics with race-specific resistance. Epidemics causing yield losses of 20-100 million bushels could once again ravage the spring wheat region at the cost of hundreds of millions dollars per epidemic.

It is important to recognize that the massive barberry eradication campaign in the United States did not completely eliminate all stem rust susceptible barberry bushes. When funding for barberry eradication ran out in the late 1980s, there were still many sites remote from wheat production in rough, wooded areas of Minnesota and Wisconsin where barberry seedlings escaped eradication. Although those barberry plants do not pose a direct threat to wheat production, we must realize that barberry does not necessarily remain in its current locations. The seedlings that survived eradication have grown to large mature bushes (Fig. 13). Each mature bush can produce up to 20,000 berries per year (Fig. 14). The berries are food to a variety of bird species. Barberry seeds can survive passage through the bird's digestive tract, so they are easily spread by movement of birds that feed on barberries.

Conclusion

There are three main points to consider regarding the relationship of barberry to wheat stem rust. First, the stem rust fungus, formerly the most destructive pathogen of wheat in North America, has not been eradicated from the United States. Stem rust still exists, although at greatly reduced amounts, in the main wheat producing regions of the United States. Second, destroying susceptible barberry bushes reduced stem rust by getting rid of local sources of epidemics and, more importantly, by removing barberry as the breeding ground for new stem rust races. Limiting the development of new stem rust races allowed wheat breeders to develop new varieties with multiple genes for resistance to the remaining races of the pathogen. Third, it is important to realize that barberry will not stay out of wheat growing regions without active efforts to keep it out. The quarantine and selective eradication are still needed to keep stem rust susceptible bushes from being distributed and planted in the major wheat producing states. Even if susceptible barberry bushes are not planted next to wheat fields, they will produce berries with seeds that may be deposited by birds in fence rows or hedge rows near the fields. In a matter of years, this could undo the success that wheat breeders and pathologists have had in suppressing stem rust epidemics in wheat crops for more than 40 years. The result would be a return to cycles of major rust epidemics several times per decade such as currently occur with oat crown rust, except that the economic consequences would be far greater with losses from $50 million to $100 million or more in any given epidemic.

References

Leonard, K.J. 2001. Stem rust - future enemy? Pages 119-146 in P.D. Peterson, ed., Stem Rust of Wheat: from Ancient Enemy to Modern Foe. APS Press, St. Paul, MN.

Long, D.L. and M.E. Hughes. 2001. Small grain losses due to rust. Publication No. CDL-EP#007 at http://www.cdl.umn.edu (accessed March, 2001).

Peterson, P.D. 2001. The campaign to eradicate the common barberry in the United States. Pages 16-50 in P.D. Peterson, ed., Stem Rust of Wheat: from Ancient Enemy to Modern Foe. APS Press, St. Paul, MN.

Roelfs, A.P. 1978. Estimated losses caused by rust in small grain cereals in the United States - 1918-76. U.S. Dep. Agric., Agricultural Research Service, Miscellaneous Publication 1363. 85 pp.

Roelfs, A.P. 1982. Effects of barberry eradication on stem rust in the United States. Plant Disease 66:171-181.

USDA,ARS Cereal Disease Laboratory. 2001. Barberry testing at the Cereal Disease Laboratory. at http://www.cdl.umn.edu (accessed March, 2001).