Abstract

Isolates of Puccinia coronata were collected from nurseries of Avena sativa on the Menashe Plateau and A. sterilis lines near Tel Aviv in 1996 and from a nursery of A. sativa lines at Menashe in 1998. Virulence to differential oat lines with resistance genes from A. sativa occurred at high frequencies, especially in 1998. Frequencies of virulence to 34 differential lines with resistance genes from A. strerilis were more similar between the collections from A. sativa at Menashe in 1996 and from A. sterilis at Tel Aviv in 1996 than between the collections from A. sativa at Menashe in 1996 and 1998. Associations of virulence to 10 pairs of single-gene differentials were statistically significant in collections from both 1996 and 1998.

Introduction

Israel is a center of diversity for Avena sterilis, the progenitor of cultivated oat, A. sativa (Leggett & Thomas, 1995). Crown rust caused by Puccinia coronata occurs abundantly on wild populations of A. sterilis in Israel (Wahl et al., 1984). There is little or no production of cultivated oat, A. sativa, in Israel so the pathogen population is largely free from agricultural influences. Populations of P. coronata persist throughout the year in Israel through the production of teliospores that survive the hot dry summers and germinate with winter rains to produce basidiospores that infect the alternate host, Rhamnus palaestina (Wahl et al., 1984).

The natural populations of A. sterilis have been shown to be a rich source of diversity for genes for race-specific resistance to crown rust (Wahl, 1970). Previous virulence surveys showed that the populations of P. coronata on A. sterilis in Israel are highly polymorphic for virulence corresponding to the race-specific resistances found in A. sterilis accessions from Israel and other countries of the Mediterranean region (Leonard et al., 2004). Earlier, those P. coronata populations were also shown to be polymorphic for virulence to the standard set of oat crown rust differentials of Simons & Murphy (1955), which contain genes for resistance identified in A. sativa, although virulences to the A. sativa genes tended to occur at greater frequencies than virulences to A. sterilis genes for crown rust resistance (Brodny, 1988; Leonard et al., 2004; Wahl, 1959). The objectives of this study were: 1) to extend the information on fluctuations in frequencies of virulence to resistances derived from both A. sativa and A. sterilis in the Israeli populations of P. coronata, and 2) to compare patterns of virulence found in collections of P. coronata from nurseries of A. sativa and A. sterilis at two locations in Israel.

Materials and Methods

Isolates of P. coronata were collected from rusted plants in two nurseries in Israel in 1996 and from one in 1998. In 1996, 25 samples were isolated from A. sativa plants in a nursery on the Menashe Plateau, an upland area southeast of Haifa where natural infection of P. coronata is abundant and where resistance in native A. sterilis plants has been shown to be prevalent (Wahl, 1970). In 1996 18 samples were also isolated from plantings of A. sterilis accessions in a nursery near Tel Aviv. In 1998 42 samples of P. coronata were collected again from the A. sativa nursery at Menashe.

The crown rust infected leaves of A. sativa and A. sterilis were air dried and sent to the United States Department of Agriculture, Agricultural Research Service, Cereal Disease Laboratory in St. Paul, Minnesota, USA, for testing. At the Cereal Disease Laboratory urediniospores were collected from the leaves, suspended in a light mineral oil, and sprayed onto seedlings of oat cv. Marvellous, which has no known genes for resistance to crown rust. After the mineral oil carrier had evaporated, the inoculated seedlings were incubated overnight in a dew chamber at 18°C and then returned to the greenhouse where the plants inoculated with each leaf sample were separated in polyethylene isolation cells to minimize cross contamination between isolates. When uredinia appeared, the seedlings inoculated with each sample were trimmed to leave a single uredinium from which urediniospores were collected to establish a single-uredinial isolate. The isolates were stored as dry urediniospores at -65°C until testing during winter when no susceptible hosts existed in the fields in Minnesota.

The single-uredinial isolates were tested for virulence on 7- to 10- day-old seedlings of two sets of oat crown rust differentials. The first set of differentials consisted of the 10 standard differentials (Simons & Murphy, 1955) which have resistance genes from A. sativa and which were used in the crown rust race surveys of Wahl (1959) and Brodny (1988) (Table 1). The second differential set consisted of 34 oat lines with crown rust resistance genes from A. sterilis plus one line with Pc-14, a resistance gene derived from the Brazilian oat cv. Ascencao (Table 2). Twenty four of the differentials in the second set have known Pc genes for crown rust resistance, many of which came from accessions of A. sterilis collected in Israel (Leonard et al., 2004). The other 11 lines were from the Iowa multiline breeding programme with single unidentified resistance genes from A. sterilis introduced into a cultivated oat background through three or more backcross generations.

Seedlings of both differential sets were inoculated by spraying them with uredinospores suspended in light mineral oil and incubating them in a dew chamber as described above. At 10 to 14 days after inoculation, the plants were evaluated for infection type. Differential lines with an abundance of heavily sporulating uredinia were scored as susceptible, and lines with no visible reaction, flecks, or small uredinia surrounded by chlorosis and producing few spores were scored as resistant.

The three collections were compared for similarity of frequencies of virulence to each differential and for mean virulence complexity based on the average number of virulences detected per isolate in each collection. In addition, virulence associations were determined for all possible pairs of resistances in the set of single-gene differentials. Significance was determined by Fisher's exact test, and statistically significant associations were compared between collections.

Results and Discussion

Nearly all of the isolates of P. coronata from Menashe in 1998 were virulent on nine of the 10 Murphy & Simons (1955) crown rust differentials (Table 1). Only the diploid cultivar Saia (A. strigosa) was resistant to most isolates. Frequencies of virulence to differentials 'Bond' and 'Landhafer' increased markedly from 1996 to 1998. The frequencies of virulence on these differentials were high in Israel also in 1991-1994 with 74-100% of isolates virulent to 'Bond' and 65-100% virulent to 'Landhafer' (Leonard et al., 2004). In 1956-1959, Wahl (1959) reported that 70% of the isolates of P. coronata from Israel were virulent on 'Bond' and 'Landhafer', but in 1976-1980 Brodny (1988) found only 24-36% virulence to 'Bond' and 16-22% virulence to 'Landhafer' in isolates of P. coronata from Israel. Also, Brodny found only 6-10% virulence to 'Victoria' in his collections, whereas our collections showed 100% virulence in 1996 and 1998. Thus, it appears that frequencies of virulence in P. coronata can be moderately variable in Israel even though the pathogen population is not influenced by commercial oat production or breeding.

The collections of P. coronata from Menashe and Tel Aviv in 1996 were polymorphic for virulence to 32 and 31, respectively, of the 35 single-resistance-gene differentials. (Table 2). In combination, the 1996 collections from the two locations were polymorphic for virulence 33 of the 35 differentials. The 1998 collection from Menashe was polymorphic for virulence to 31 of the 35 differentials. Mean virulence complexity on this differential set ranged from 15.7 for Menashe, 1998 to 17.6 for Tel Aviv, 1996. The Tel Aviv, 1996 collections were made from A. sterilis, which may account for the slightly greater virulence complexity, because all but one of the 35 differentials have resistance genes derived from A. sterilis. As in earlier studies, (Leonard et al., 2004; Manisterski & Wahl, 1995), no virulence to Pc-68 was found in Israel.

The similarity of virulence in the three collections of P. coronata was compared by determining the absolute difference in virulence frequency between paired collections on each differential and then averaging the values over all 35 differentials. By this measure the virulence patterns of the two 1996 collections were more similar to each other than either was to the 1998 collection (Table 3). This is in spite of the fact that the Menashe, 1996 isolates came from A. sativa while the Tel Aviv, 1996 isolates came from A. sterilis. Thus, there were greater differences in virulence between collections from the same host in the same location in Israel over 2 years than between collections from different hosts and different locations in the same year.

Another measure of similarity is the commonality of virulence associations between two collections. Among all possible pair-wise combinations of polymorphic virulences, the proportions of paired virulences showing significant (P < 0.01) associations were 3.2% for Menashe, 1996, 6.2% for Tel Aviv, 1996, and 5.2% for Menashe, 1998. Data for collections from Menashe and Tel Aviv in 1996 were combined for comparisons of virulence associations between 1996 and 1998 in Israel (Table 4). Ten pairs of virulence were significantly (P < 0.05) associated in both years. For eight of the pairs the significance level was P < 0.01 in both years. The association of virulence to Pc-39 and Pc-55 was expected, because these two resistance genes are either identical or differ nnly in sensitivity to high temperatures at which the resistance breaks down (Bonnett, 1996; Chong & Kolmer, 1993). Virulences to Pc-36, Pc-56, X534, and X716 were associated as a group with the virulences occurring at 10% of less in 1996 and at 60% or more in 1998. The frequencies of the associated virulences to Pc-45 and Pc-54 dropped from 91% in 1996 to 5 and 17% in 1998. Leonard (2003) previously reported that virulences to Pc-45 and Pc-54 are associated in the United States.

Our results support previous evidence of a high level of polymorphism in populations of P. coronata in Israel for virulence corresponding to resistance genes in both A. sterilis and A. sativa (Brodny, 1988; Leonard et al., 2004; Wahl, 1959). The data are consistent with the previous observation (Leonard et al., 2004) that virulence to resistance genes from A. sativa generally occur at greater frequencies than virulence to resistance genes from A. sterilis. Whether the source of isolates of P. coronata was from plants of A. sterilis or A. sativa in Israel appeared to have little influence on the frequency of virulence in the collections that we tested. This suggests that, although genes for race-specific resistance are prevalent in at least some populations of A. sterilis in Israel, the individual Pc genes likely do not occur at high frequencies. Thus, the corresponding genes for virulence are not selected strongly enough to cause them to become fixed in the P. coronata population. Likewise, stabilizing selection does not appear to be strong enough to substantially reduce frequencies of virulence within a few generations on plants of A. sativa that lack most, if not all, of the resistance genes that occur in A. sterilis populations and influence the population structure of P. coronata in natural populations in Israel.

Acknowledgements

The technical assistance of Gerald Ochocki in maintaining cultures and setting up virulence tests is gratefully acknowledged.

References

Bonnett DG, 1996. Host:pathogen studies of oat leaf rust in Australia. Ph.D. Thesis, University of Sydney, Australia.

Brodny U, 1988. Physiologic races of oat crown rust identified in Israel during 1976-1988. Canadian Journal of Plant Pathology 10, 53-56.

Chong J, Kolmer JA, 1993. Virulence dynamics and phenotypic diversity of Puccinia coronata f. sp. avenae in Canada from 1974 to 1980. Canadian Journal of Botany 71, 248-255.

Leggett JM, Thomas H, 1995. Oat evolution and cytogenetics. In: RW Welch, ed. The Oat Crop: Production and Utilization. London, Chapman & Hall, 120-146.

Leonard KJ, 2003. Regional frequencies of virulence in oat crown rust in the United States from 1990 through 2000. Plant Disease 87, 1301-1310.

Leonard KJ, Anikster Y, Manisterski J, 2004. Patterns of virulence in natural populations of Puccinia coronata on wild oat in Israel and in agricultural populations on cultivated oat in the United States. Phytopathology 94, 505-514.

Manisterski J, Wahl I, 1995. Studies on oat crown rust in Israel. Petria 5, 50-54.

Simons MD, Murphy HC, 1955. A comparison of certain combinations of oat varieties as crown rust differentials. United States Department of Agriculture, Agricultural Technology Bulletin 1112.

Wahl I, 1959. Physiological races of oat crown rust identified in Israel in 1956-1959. Bulletin of the Research Council of Israel, 8D, 25-30.

Wahl I, 1970. Prevalence and geographic distribution of resistance to crown rust in Avena sterilis. Phytopathology 60, 746-749.

Wahl I, Anikster Y, Manisterski J, Segal A, 1984. Evolution at the center of origin. In: WR Bushnell and AP Roelfs, eds. The Cereal Rusts, vol. 1. New York, Academic Press, 165-190.