Abstract

In the United States, barley stripe rust, caused by Puccinia striiformis f. sp. hordei, has spread in the south central and western United States and has caused localized damages in California and the Pacific Northwest since it was first reported in southern Texas in 1991. Barley stripe rust has been monitored using trap nurseries and field survey. Sixty-nine races of P. striiformis f. sp. hordei have been identified using a set of 12 barley genotypes. Races that were detected in California and/or Texas were also detected in the Pacific Northwest, indicating the spread of the races among the different regions and lack of selection from cultivars grown in various regions. Races with a relatively narrow virulence spectrum tended to become predominant. Major barley cultivars with non-race specific high-temperature, adult-plant resistance and the barley cropping system have played significant roles in reduction of damage by barley stripe rust.

Introduction

In the US, stripe rust (also called yellow rust) of barley, caused by Puccinia striiformis f. sp. hordei, was first observed in a barley nursery near Uvalde, Texas in April 1991 (Marshal & Sutton, 1995; Roelfs et al., 1992). The pathogen was postulated to be introduced from Europe to Colombia, where the rust was first reported in 1975 (Dubin & Stubbs, 1986). The pathogen spread southward and northward and reached Mexico in 1987 (Chen et al., 1995; Brown et al., 2001; Line, 2002). Since 1991, stripe rust of barley has quickly spread and established in the south-central and western US.

Through co-operation with pathologists, breeders and growers, stripe rust of barley has been monitored throughout the barley growing areas in the United States, especially in the south central states, California and the Pacific Northwest. Identification of races of P. striiformis f. sp. hordei races has been conducted in the US Department of Agriculture, Agricultural Research Service, Wheat Genetics, Quality, Physiology and Disease Research Unit at Washington State University, Pullman, Washington. Barley germplasms and breeding lines were annually screened in greenhouse and field plots. Yield losses were estimated based on field survey and nursery trials of commonly grown cultivars. Foliar fungicides were tested for their efficacy in stripe rust control. The management of the disease has been primarily through use of resistant cultivars and application of foliar fungicides.

This paper is to report epidemics of barley stripe rust, document races of P. striiformis f. sp. hordei and discuss environmental and crop managerial factors that affect stripe rust epidemics.

Materials and Methods

Monitoring barley stripe rust
Stripe rust was monitored by surveying commercial fields during the growing season through collaborators throughout the barley growing season. Trap plots consisting of the set of differential genotypes (Table 1) and other cultivars either susceptible or resistant to P. striiformis f. sp. hordei were planted at various locations. For commercial field survey, infection type, severity (percentage of leaf areas infected), prevalence (percentage of plants infected), and information on cultivar, location, and growth stage were recorded. Rusted leaf samples were collected for race identification. For the trap plots, infection type and severity were recorded one to three times during the growing season. The rust data were used for forecasting stripe rust epidemic and for determining virulences of the stripe rust population.

Yield loss estimation
Field experiments for determining yield losses caused by stripe rust and gains obtained by fungicide application were conducted in Washington almost every year since 1995. The 2002 experiment is used as an example. Eight barley cultivars 'Topper', 'Steptoe', 'Harrington', 'Morex', 'Baronesse', 'Farmington', 'WA8681', and 'Bancroft' were tested using randomized split-block design with four replicates. Field blots were planted on 30 April. Field cultivation, planting, and managements were the same as the commercial fields in the area. Fields were under natural infection of the pathogen. Before fungicide application, each plot was cut from the middle to separate it into two sub-plots, one was then sprayed with a foliar fungicide and the other was not sprayed. Each sub-plot was measured as 5 X 1.34 m. On 28 June, Quadris was sprayed at the rate of 6.2 fl. oz/A in water mixed with Agridex 10 ml/l when stripe rust developed to 5%-10% on susceptible cultivars like 'Topper' and 'Steptoe' in the heading stage. Stripe rust severity data were recorded three weeks after fungicide application. Grain yield was measured for each sub-plot after harvested in late August. Data of stripe rust severity and grain yield were compared between fungicide-sprayed and non-sprayed plots with the t test using the MS Excel package. The experimental data were used to estimate yield losses for the State of Washington. Cultivar acreage and rust severity were considered in yield loss estimation. The data were also used to make recommendations for growers to choose cultivars and use fungicide to control stripe rust.

Race identification
Races of stripe rust samples were determined using a set of differential genotypes following the methods described by Chen et al. (1995). Currently, 12 barley genotypes (Table 1) were used to differentiate races of P. striiformis f. sp. hordei. Samples of infected leaves that were usually shipped in glassine envelopes were used to inoculate seedlings of 'Steptoe' or 'Topper', which are susceptible to all races of P. striiformis f. sp. hordei, for increasing urediniospores. For samples of poor quality (received more than 7 days after collection or with limited uredia), leaf pieces were placed on moist filter paper in a petri dish that was kept at temperatures of 4-12°C overnight. Fresh spores produced on the leaves were used in inoculation for spore increase. Inoculated plants were kept in a dew chamber at 10°C for 18 to 24 hours for infection and then grown in a greenhouse growth chamber at a diurnal temperature cycle gradually changing from 4°C at 2:00 am to 20°C at 2:00 pm. The light period consisted of day light supplemented with metal halide lights to extend the photoperiod to 16 hours. Urediniospores that were collected 16 to 30 days after inoculation were used to inoculate seedlings of the set of differential genotypes. Inoculated plants were kept in the dew chamber for infection and then grown in a growth chamber for symptom development under the same conditions as for spore increase described above. Infection type (IT) data were recorded 18-22 days after inoculation according to the 0-9 scale described by Line and Qayoum (1991) as the following: 0 = no visible signs or symptom, 1 = necrotic and/or chlorotic flecks; no sporulation, 2 = necrotic and/or chlorotic blotches or stripes; no sporulation, 3 = necrotic and/or chlorotic blotches or stripes; trace sporulation, 4 = necrotic and/or chlorotic blotches or stripes; light sporulation, 5 = necrotic and/or chlorotic blotches or stripes; intermediate sporulation, 6 = necrotic and/or chlorotic blotches or stripes; moderate sporulation, 7 = necrotic and/or chlorotic blotches or stripes; abundant sporulation, 8 = chlorosis behind sporulating area; abundant sporulation, and 9 = no necrosis or chlorosis; abundant sporulation.

Results and Discussion

Spread and distribution of the pathogen
Figure 1 shows occurrence and spread of barley stripe rust in the US since it was first observed in southern Texas in 1991. In 1992, barley stripe rust was reported in Oklahoma, New Mexico, and Colorado. In 1993, the disease was found in Arizona, southern Idaho, and Montana. In 1994, the disease was found in California and Utah. Barley stripe rust was not found in Oregon, Washington, and northern Idaho until the 1995 growing season (Chen et al., 1995). The disease has not been found in North Dakota, one of the major barley growing states, which may be due to the big geographic gaps between Texas and North Dakota, where limited barley has been grown in that region.

Stripe rust damage
Table 2 shows the results of the yield loss experiment in 2002. Under natural infection at Pullman, Washington, stripe rust caused more than 40% of yield loss on highly susceptible cultivars like 'Topper' that was not commercially grown. The commercially grown susceptible cultivars like 'Harrington' and 'Steptoe' had more than 20% of yield losses. The fact that these commercial cultivars are maturing faster than 'Topper' might reduce the yield losses. 'Baronesse', the most grown cultivar and having moderate level of high-temperature, adult-plant (HTAP) resistance (XM Chen, unpublished data), had 12% yield loss in the experimental field and had much lower rust severity in commercial fields.

Table 3 shows estimated yield losses in states that had stripe rust epidemics from 1992 to 2003. California suffered the biggest yield losses. The losses in 1996 reached 20%, more than 2.5 million bushels. From 1992 to 2003, California had 7.6 million bushels of yield loss, 6.5% the total potential production. In addition, the barley acreage decreased about 73.6% from 220,000 acres in 1994 to 58,000 in 2003. The Pacific Northwest (Idaho, Oregon and Washington) did not have as severe epidemics as did California. The worst yield losses in the Pacific Northwest occurred in 1998 and 2000; ranging from 2 to 5%. This region had about 4.9 million bushels of yield losses in the last decade. The weather conditions in this region are favorable to stripe rust (Line & Qayoum, 1992). Stripe rust of wheat (P. striiformis f. sp. tritici) is most destructive in this region. However, barley in this region has not suffered as much damage as wheat by stripe rust. In 2002 and 2003, severe stripe rust epidemics occurred on wheat while stripe rust on barley was very low in commercial fields. This difference can be attributed to the following three factors: 1) the barley acreage was much smaller than the wheat acreage; 2) almost all barley acreage was grown with spring barley while both winter and spring wheat were grown in the same region, which provide hosts all year along; and 3) the most predominant barley cultivar in this region has been 'Baronesse', which has moderate level of HTAP resistance. The region east of the Rocky Mountains (Montana, Colorado, and Texas) had localized stripe rust damage in some of these years. In general, yield losses were less than 0.1% in these states. In the US, stripe rust caused more than 12.6 million bushels of yield losses during the last 12 years.

Races of Puccinia striiformis f. sp. hordei
Before barley stripe rust was first observed in the United States in 1991, stripe rust had been sometimes found on barley fields and the collections had been always identified as races of P. striiformis f. sp. tritici that was occurring in adjacent wheat fields (Chen & Line, 1995). The rust had never caused significant damage to barley crops. When P. striiformis f. sp. hordei reached Mexico and later in the United States, the barley stripe rust pathogen was practically referred as a single "race 24" following Stubbs' race designations (Stubbs, 1985). The rust population in the US was soon found of a mixture of different virulence groups (Chen et al., 1995; Marshall & Sutton, 1995). Chen et al. (1995) selected 11 barley genotypes (Table 1) to differentiate races of P. striiformis f. sp. hordei. Most of the differential genotypes were of European origin and used by Stubbs (1985). In 2001, 'Bancroft', one of the first barley cultivars developed in the United States for resistance to stripe rust (Wesenberg et al., 2001), was added to the differential set. Table 4 shows 69 races identified so far. From 1993 to 1995, a limited number of isolates were obtained for race identification. However, these isolates were identified as diverse races. Nine races were identified from 14 isolates in 1993 and seven races were identified from 8 isolates in 1994 (Chen et al., 1995). In 1995, 17 new races were identified and none of the new and previously identified races had five or more isolates. Similarly, 32 races were identified from 52 isolates in 1996, 32 races from 136 isolates in 1997, and 38 races from 166 isolates in 1998. Relatively small numbers of races were identified from 1999 to 2003; nine, 21, 14 and nine races were identified from 56, 39, 67, 35 and 36 isolates, respectively. Since 1996, certain races have become predominant. Race PSH-33 was the most predominant in 1996 (13.5%), 1998 (14.5%), 1999 (30.5%) and 2000 (25.6%). Races PSH-46, 56 and 54 were the most predominant in 1997 (17.6), 2001 (19.7%) and 2002 (31.4%), and 2003 (27.8%), respectively. These races belong to a group of races sharing virulences on 'Topper' and 'Abed Binder 12', which is susceptible to 77% of the races. They differ in virulence on 'Hiproly', 'Trumpf', and/or 'Bancroft'. The predominant races have relatively narrow spectra of virulences, which may provide them with an advantage in fitness or adaptation in the environment and cropping system. Because major cultivars have either non race-specific HTAP resistance (e.g. 'Baronesse') or susceptible to all races (e.g. 'Morex' and 'Harrington') (XM Chen, unpublished data), the selection pressure of the host has been low, and as a result, the rust population still consists of numerous races. The HTAP resistance in the major barley cultivars has kept stripe rust under control in the Pacific Northwest. Barley breeders and pathologists are taking efforts to maintain and improve the level of HTAP resistance in new cultivars with better yield potential and quality.

Acknowledgements

I would like to thank Dr. Roland F. Line and Mrs. Mary Moore for collecting stripe rust samples and identifying races, Drs. Lee Jackson, Patrick Hayes, Steve Ullrich, David Marshall, and others for sending samples of barley stripe rust and sharing the disease information in various states. I also like to thank David Long and Mark Hughes for compiling yield loss data from various states. The financial support from the US Department of Agriculture, Agricultural Research Service and Washington Barley Commission is highly appreciated.

References

Brown WM Jr, Hill JP, Velasco VR, 2001. Barley yellow rust in North America. Annual Review of Phytopathology 39, 367-384.

Chen XM, Line RF, 2003. Identification of genes for resistance to Puccinia striiformis f. sp. hordei in 18 barley genotypes. Euphytica 129, 127-145.

Chen XM, Line RF, Leung H, 1995. Virulence and polymorphic DNA relationships of Puccinia striiformis f. sp. hordei to other rusts. Phytopathology 85, 1335-1342.

Dubin HJ, Stubbs RW, 1986. Epidemic spread of barley stripe rust in South America. Plant Disease 70, 141-144.

Line RF, 2002. Stripe rust of wheat and barley in North America: a retrospective historical review. Annual Review of Phytopathology 40, 75-118.

Line RF, Qayoum A, 1991. Virulence, aggressiveness, evolution, and distribution of races of Puccinia striiformis (the cause of stripe rust of wheat) in North America, 1968-87. USDA-ARS, Technical Bulletin 1788, 44 pp.

Marshall D, Sutton RL, 1995. Epidemiology of stripe rust, virulence of Puccinia striiformis f. sp. hordei, and yield loss in barley. Plant Disease 79, 732-737.

Roelfs AP, Huerta-Espino J, Marshall D, 1992 Barley stripe rust in Texas. Plant Disease 76, 538.

Stubbs RW, 1985. Stripe rust. Pages 61-101 in AP Roelfs and WR Bushnell (ed.) The Cereal Rusts Volume II. Academic Press, INC. Orlando.

Wesenberg DM, Burrup DE, Brown Jr WM, Velasco VR, Hill JP, Whitmore JC, Karow RS, Hayes PM, Ullrich SE, Liu CT, 2001. Registration of 'Bancroft' barley. Crop Science 41, 265-266.