Races of Puccinia graminis on wheat, oat, and barley in Canada in 2009 and 2010

Abstract

Stem rust, caused by Puccinia graminis, is an important disease on wheat, oat, and barley crops worldwide. Because the disease is primarily controlled in Canada using resistant cultivars, determining the virulence structure in pathogen populations is necessary for the early detection of novel virulent races. Surveys of wheat (Triticum aestivum), oat (Avena sativa), and barley (Hordeum vulgare) fields and trap nurseries were conducted in 2009 and 2010 across Canada to provide information on incidence and severity of stem rust infection, and to identify the virulence structure in the pathogen populations. Stem rust was not found on cultivated wheat, and was at trace levels in oat and barley fields in both years. Race QFCSC of P. graminis f. sp. tritici was dominant in 2009 (88.0% of all samples), and was the only race detected in Canada in 2010. Races MCCFC, RFCSC, and TPMKC were found at low levels (<5%) in 2009. Thirteen races of P. graminis f. sp. avenae were found in 2009, with TGD (30.4%), TJN (16.2%), TGN (14.9%), TJJ (11.2%), TJS (9.9%), and TGB (8.6%) the most frequent. Twelve races of P. graminis f. sp. avenae were found in 2010, and TGN (27.3%), TGB (19.8%), TGD and TJN (14.0% each), and TJS (9.9%) were the most frequent. Race TJJ, which is virulent on all Canadian oat cultivars and was first detected in 1998, fell to only 5.2% frequency (total samples) in 2010. One new race (TJL) was detected in three samples from Manitoba in 2009, but is avirulent to gene Pg13 and does not pose a threat to oat production in Canada.

Résumé

La rouille de la tige, causée par Puccinia graminis, est une grave maladie du blé, de l’avoine et de l’orge, et ce, à l’échelle de la planète. Parce qu’au Canada la maladie est principalement gérée au moyen de cultivars résistants, il importe de déterminer la structure de virulence chez les populations d’agents pathogènes afin de détecter toute nouvelle race virulente. Des études ont été menées en 2009 et 2010 d’un bout à l’autre du Canada dans des champs de blé (Triticum aestivum), d’avoine (Avena sativa) et d’orge (Hordeum vulgare) ainsi que dans des pépinières-pièges afin de fournir de l’information sur l’incidence et la gravité de la rouille de la tige, ainsi que pour déterminer la structure de virulence chez les populations d’agents pathogènes. Durant les deux années, nous n’avons pas trouvé de rouille dans les champs de blé et que des traces dans les champs d’avoine et d’orge. En 2009, la race QFCSC de P. graminis f. sp. tritici prédominait (88.0% de tous les échantillons), et c’est la seule qui a été détectée au Canada en 2010. En 2009, les taux des races MCCFC, RFCSC et TPMKC étaient faibles (< 5%). En 2009, on a trouvé 13 races de P. graminis f. sp. avenae, les plus courantes étant TGD (30.4%), TJN (16.2%), TGN (14.9%), TJJ (11.2%), TJS (9.9%) et TGB (8.6%). En 2010, on en a trouvé 12 races: TGN (27.3%), TGB (19.8%), TGD et TJN (14.0% chacune) et TJS (9.9%) étaient les plus courantes. La race TJJ, qui est virulente à l’égard de tous les cultivars canadiens d’avoine et qui a été détectée pour la première fois en 1998, n’affichait plus, en 2010, qu’une fréquence de 5.2% (tous les échantillons). Une nouvelle race (TJL) a été détectée dans trois échantillons provenant du Manitoba en 2009, mais elle est avirulente à l’égard du gène Pg13 et ne pose aucune menace à la production d’avoine au Canada.

Keywords:

• barley
• oat
• race
• rust survey
• stem rust
• virulence
• virulence phenotype
• wheat

Mots clés:

• avoine
• blé
• études sur la rouille
• orge
• phénotype de virulence
• race
• rouille de la tige
• virulence

Introduction

Stem rust of wheat (Triticum aestivum L. and Triticum turgidum L.) and barley (Hordeum vulgare L.), caused by Puccinia graminis Pers.:Pers. f. sp. tritici Eriks. & E. Henn., is an important disease of these crops worldwide. Epidemics of stem rust on wheat occurred frequently in North America in the first half of the 20th century, with substantial losses in 1916, 1923, 1927, and 1935 (Kolmer 2001). In Canada, the last major epidemics on spring wheat occurred from 1953–1955 (Peturson 1958), but all spring and durum wheat cultivars currently recommended for production in the prairie region are resistant to prevalent races of stem rust in the North American population (Fetch et al. 2015). In contrast, the last stem rust epidemic on winter wheat occurred in 1986, causing severe losses in Saskatchewan and an estimated $3.7 million in losses in Manitoba (Martens & McFadden 1988). Most winter wheat cultivars currently grown in Canada are now resistant to stem rust.

In contrast to spring wheat, stem rust epidemics on barley in North America recently occurred in 1990 and 1991 (Steffenson 1992). These were caused by race QCCJ, which was first detected in 1988 (Martens et al. 1989). This race is virulent on gene Rpg1, which has been durable in North America since the introduction of the barley variety ‘Kindred’ (Steffenson 1992). Race QCCJ has been detected at a low frequency in Canada, and likely is explained by the limited production of winter wheat cultivars that are susceptible to this race in the southern United States (Dill-Macky & Roelfs 1998). Race QCCJ has not caused any significant losses in barley production in Canada to date.

Stem rust of oat (Avena sativa L.), caused by Puccinia graminis Pers.:Pers. f. sp. avenae Eriks. & E. Henn., causes sporadic epidemic losses. The most recent oat stem rust epidemic occurred in 2002, with an estimated $14 million yield loss that was caused by race TJJ (Fetch 2005). This race was first detected in 1998 and has been prevalent in the P. g. avenae population since then. In addition to this race, several new races have recently been detected, including TJG (1999), TJS (2005), and TGN and TJN (2006) (Fetch et al. 2015). While races TJJ, TJG, and TJS are virulent on all currently grown Canadian oat cultivars, races TGN and TJN are avirulent on gene Pg13, which likely is present in most oat cultivars classified as resistant (Mitchell Fetch & Fetch 2011).

Annual surveys have been conducted in Canada since 1919 to describe the incidence, severity, and virulence spectrum of P. graminis f. sp. tritici and P. graminis f. sp. avenae populations (Johnson & Green 1957). These surveys are critical to detect novel virulent rust races that could overcome the pyramid of stem rust resistance genes (Sr, Pg, and Rpg genes for wheat, oat, and barley, respectively) in currently grown cereal cultivars. They also provide a historical record of virulence frequency of races in P. graminis populations, which have been used for analyses of population genetics.

The objective of this paper is to document changes on the incidence, severity, and virulence spectrum of P. graminis f. sp. tritici and P. graminis f. sp. avenae populations in Canada during 2009 and 2010.

Materials and methods

Collection and processing of isolates

Methods for collection of infected stem samples, inoculation of differential lines, and race determination were as previously reported (Fetch et al. 2015). Commercial fields, patches of wild barley (Hordeum jubatum L.) and wild oat (Avena fatua L.), and susceptible lines of spring wheat (‘Klein Titan’, ‘Little Club’, ‘Morocco’), barley (‘Wolfe’), and oat (‘Triple Crown’, ‘Makuru’) from trap-plot nurseries were assessed for stem rust incidence and severity in Manitoba and Saskatchewan from July to September in 2009 and 2010. Additional barley and wheat samples were obtained from collaborators in Ontario and Quebec. Infected stems (about five per site or line) were collected to provide isolates of P. graminis f. sp. tritici and P. graminis f. sp. avenae to characterize the virulence spectrum of the 2009 and 2010 stem rust populations. Samples were air-dried at room temperature for at least 24 h and then stored in a refrigerator at 3–5°C. Urediniospores were scraped from stems using a sterile spatula sprayed with distilled water and transferred to seedling leaves of either ‘Little Club’ wheat (P. g. tritici) or ‘Pc94ʹ oat (P. g. avenae). Inoculated seedlings were incubated in the dark for 16 h at 20°C and near 100% relative humidity in a dew box misted with ultrasonic humidifiers. Once the dew period was finished, the door of the dew box was slightly opened and T5 fluorescent lights (200 μmol·m⁻²·s⁻¹) mounted over the clear Plexiglas top of the box were turned on for 2–3 h to provide a period of slow-drying and light to finish infection. Plants were then removed from the dew box and moved to a greenhouse at 20 ± 4°C with a 14 h light: 10 h dark photoperiod supplemented with high-pressure sodium lighting. Each pot containing inoculated ‘Little Club’ or ‘Pc94ʹ seedlings was covered with a plastic lamp cover, with a hole in the base connected to an air source of light positive pressure to prevent cross-contamination of isolates. At 14 days post-inoculation, a single isolate consisting of several random pustules was made from each sample using a cyclone spore collector and size 00 gelatin capsule. Additionally, bulk collections (about 20 random samples per bulk) were made from the initial inoculations using the cyclone spore collectors.

Determination of physiologic races

Race typing of isolates was done using differential lines (5 seeds/line) planted in RL98 conetainers (Steuwe & Sons, Inc., Tangent, OR) filled with moist #4 Sunshine Mix (Sun Gro Horticulture, Agawam, MA). For samples from wheat or barley, five sets of single-gene differential lines (Sr5, Sr21, Sr9e, Sr7b; Sr11, Sr6, Sr8a, Sr9g; Sr36, Sr9b, Sr30, Sr17; Sr9a, Sr9d, Sr10, SrTmp; Sr24, Sr31, Sr38, SrMcN) were used to determine the physiological race using the letter-code nomenclature for P. graminis f. sp. tritici (Jin et al. 2008). For samples from oat, three sets of single-gene differential lines (Pg1, Pg2, Pg3, Pg4; Pg6, Pg8, Pg9, Pg10; Pg12, Pg13, Pg15, Pg16) were used to determine the physiological race of each isolate using the letter-code nomenclature for P. graminis f. sp. avenae (Fetch & Jin 2007). Oat isolates also were inoculated onto the lines ‘Pg-a’, ‘Alpha’, and ‘Omega’ (Pg-a complex; Martens et al. 1981), and ‘Wisconsin X-1588ʹ (Pg10; Harder 1999). Bayol oil (Esso Canada, Toronto) was added to each gelatin capsule containing an isolate, and spores were inoculated onto 8-day-old seedlings of differential lines using a rust inoculator (G-R Manufacturing, Manhattan, KS) pressurized to 20 kPa by an air pump. Inoculated plants were incubated as described previously, then moved to a greenhouse at 20 ± 4°C with a 14 h light: 10 h dark photoperiod supplemented by artificial lighting. Infected plants were assessed for infection type (IT; Stakman et al. 1962) at 14 days post-inoculation. Isolates were identified to race based on their reaction on the differential lines (Fetch & Jin 2007; Jin et al. 2008). In the event where both resistant and susceptible ITs were found on a single differential line, single-pustule isolates were generated to determine the identity of each race.

In order to quickly detect novel wheat stem rust virulence, bulked collections were inoculated on 22 wheat lines containing a single Sr gene, and also on Canadian cultivars with resistance to historic North American races of P. graminis f. sp. tritici. Bulked collections were inoculated, incubated, and assessed as previously described on: (i) lines with resistance genes Sr7a, Sr8b, Sr12, Sr13, Sr14, Sr15, Sr16, Sr18, Sr20, Sr22, Sr25, Sr26, Sr27, Sr28, Sr29, Sr32, Sr33, Sr34, Sr35, Sr37, Sr39 and Sr40; (ii) lines RL6076 and Tr129; (iii) the T. aestivum cultivars ‘AC Barrie’, ‘AC Domain’, AC ‘Karma’, ‘AC Michael’, ‘AC Splendor’, ‘AC Vista’, ‘Burnside’, ‘Canthatch’, ‘Columbus’, ‘Harvest’, ‘Kanata’, ‘Lillian’, ‘Little Club’, ‘McKenzie’, ‘Peace’, ‘Selkirk’, ‘Snowbird’, ‘Somerset’, ‘Superb’, and ‘5701PR’; and (iv) the T. turgidum cultivars ‘AC Navigator’, ‘Sceptre’, and ‘Strongfield’.

Results and discussion

Environmental conditions

Environmental conditions were generally favourable for stem rust infection in Canada in 2009 and 2010 across the Prairie region (Agriculture & Agri-Food Canada 2015). From early July to early September 2009, it was dry (40–85% of normal rainfall) in southern Manitoba and very dry (<40% to 85% of normal rainfall) in northern Alberta, but normal (85–115% of normal rainfall) to wet (115–150% of normal rainfall) in northern Manitoba, Saskatchewan, and southern Alberta. Temperatures were cool (−1 to −2°C below average) in June, colder (−1 to −3°C lower than average) in July, and normal (−1 to +1°C from average) in August, but severely hot (>4°C above average) in September 2009 across the ‘rust area’ of Manitoba and eastern Saskatchewan. In 2010, rainfall was high (115–150% of normal) in Manitoba and Alberta, and very high (150–200% of normal) in Saskatchewan from early July to early September. Temperatures were normal (−1 to +1°C from average) across the Prairies from June to August, but a bit cool (−1 to −2°C from average) in Alberta in September. Temperatures were cold (−2 to −4°C from average) in Alberta and Saskatchewan in May that led to delayed planting in 2010.

Incidence and severity

Despite favourable environmental conditions that were conducive for stem rust infection in 2009 and 2010, only trace infection was found in Canada on susceptible wheat, oat, and barley lines in trap-plot nurseries, on cultivated oat and barley, and on wild oat (A. fatua) and wild barley (H. jubatum). Stem rust infection was not detected in commercial spring wheat fields in Canada in 2009 or 2010. No significant losses in yield occurred in wheat, oat or barley in 2009 or 2010 in Canada that could be attributed to stem rust. In the United States, stem rust severity on wheat was mostly at low levels in 2009, but up to 40% severity was found in several susceptible winter wheat fields, causing trace losses (Long 2009). Stem rust severity on oat was generally at low levels in the United States Midwest region, but was severe (20–40%) in Texas and Louisiana in 2009 (Long 2009). Stem rust severity was mostly at low levels on wheat and barley in the United States in 2010, but high levels of stem rust was observed in some plots in Texas and Mississippi (Long 2010). Stem rust on oat in the United States was at low levels in 2010, but some plots in College Station in central Texas had high severity (Long 2010). Low levels of stem rust inoculum from the United States likely explain the low stem rust severity found in Canada in 2009 and 2010, since weather conditions were favourable for infection. Additionally, most fields of wheat, oat, and barley in Canada were sprayed with foliar fungicides, which may have contributed to the low levels of stem rust infection.

Physiological specialization

Puccinia graminis f. sp. tritici

There were 42 viable isolates from 175 collections in 2009, and 80 viable isolates from 147 collections in 2010, for evaluation of P. graminis f. sp. tritici virulence on wheat and barley. Race QFCSC was dominant (88% of total isolates) in 2009 (Table 1) and was the only race found in 2010 (data not presented). QFCSC was also predominant in the United States in 2009 (Long 2009) and 2010 (Long 2010). Race RFCSC was detected at two locations in Ontario in 2009, and also was reported in several Midwestern states of the United States in 2009 (Long 2009). Race MCCFC was found at one location on wild barley in Manitoba in 2009. Race TPMKC was found at Carmen, MB in 2009 in two plots of wheat.

Table 1. Frequencies of races of Puccinia graminis f. sp. tritici obtained from wheat trap plots, cultivated barley, and wild barley in Manitoba, Ontario, Quebec, and Saskatchewan, Canada in 2009.

		Wheat		Cultivated barley		Wild barley		Total
Race	Virulence spectrum (ineffective Sr genes)	No.	%	No.	%	No.	%	No.	%
Manitoba
MCCFC	5, 7b, 9g, 17, 10, Tmp, McN	0	0	0	0	1	14.3	1	6.3
QFCSC	5, 21, 8a, 9g, 17, 9a, 9d, 10, McN	6	75.0	1	100	6	85.7	13	81.2
TPMKC	5, 21, 9e,7b, 11, 8a, 9g, 36, 17, 9d, 10, Tmp, McN	2	25.0	0	0	0	0	2	12.5
Ontario
QFCSC	5, 21, 8a, 9g, 17, 9a, 9d, 10, McN	0	0	5	83.3	0	0	5	71.4
RFCSC	5, 21, 7b, 8a, 9g, 17, 9a, 9d, 10, McN	1	100	1	16.7	0	0	2	28.6
Quebec
QFCSC	5, 21, 8a, 9g, 17, 9a, 9d, 10, McN	1	100	0	0	0	0	1	100
Saskatchewan
QFCSC	5, 21, 8a, 9g, 17, 9a, 9d, 10, McN	3	100	0	0	15	100	18	100
Total
MCCFC	5, 7b, 9g, 17, 10, Tmp, McN	0	0	0	0	1	4.5	1	2.4
QFCSC	5, 21, 8a, 9g, 17, 9a, 9d, 10, McN	10	76.9	6	85.7	21	95.5	37	88.0
RFCSC	5, 21, 7b, 8a, 9g, 17, 9a, 9d, 10, McN	1	7.7	1	14.3	0	0	2	4.8
TPMKC	5, 21, 9e,7b, 11, 8a, 9g, 36, 17, 9d, 10, Tmp, McN	2	15.4	0	0	0	0	2	4.8

Single-gene differential lines used for race pathotyping are Sr5, 21, 9e, 7b, 11, 6, 8a, 9g, 36, 9b, 30, 17, 9a, 9d, 10, Tmp, 24, 31, 38, and McN.

Three races of P. graminis f. sp. tritici were found in Canada in 2009 and only one in 2010. Race QFCSC has been dominant in the P. graminis f. sp. tritici population since 2004, although ‘old’ races such as TPMKC and MCCFC were found in minor frequency and are probably surviving on wild grass hosts. Since stem rust rarely overwinters in Canada, the frequency of races in the Midwestern United States population of P. graminis f. sp. tritici is a major determinant of races that will be found in Canada, blowing north on the ‘Puccinia pathway’ during the growing season. Race QFCSC has been dominant in both the United States, and Canadian populations of P. graminis f. sp. tritici since 2004. Since this is a relatively avirulent race, the dominance of QFCSC is best explained by a lack of stem rust resistance in winter wheat cultivars grown in the southern United States. Otherwise, if United States winter wheat cultivars had moderate levels of stem rust resistance, highly virulent strains such as TPMKC would be expected to be dominant. Race QFCSC also could possibly be a more aggressive competitor compared with other races and proliferate better than other races on wild grass species such as Hordeum jubatum. The latter would require further experiments to confirm this hypothesis.

Virulence frequencies of P. graminis f. sp. tritici isolates from 2009 on the single-gene differential lines are presented in Table 2. All isolates were virulent to Sr5, Sr9g, Sr10, Sr17, and SrMcN, and most were virulent to Sr8a, Sr9a, Sr9d, and Sr21, while all were avirulent on Sr6, Sr9b, Sr24, Sr30, Sr31, and Sr38. Since only one race (QFCSC) was detected in 2010, genes Sr5, Sr8a, Sr9a, Sr9d, Sr9g, Sr10, Sr17, Sr21, and SrMcN were ineffective and all other genes were effective (data not presented). Stem rust resistance in Canadian wheat is mainly derived from ‘Neepawa’ (Sr5, Sr7a, Sr9b, Sr12, Sr16; Kolmer et al. 1991), which is in the pedigree of most current Canadian western red spring wheat cultivars (McCallum & DePauw 2008). Genes Sr6, Sr7a, and Sr9b are effective against most races (including QFCSC) in the North American population of P. graminis f. sp. tritici, and along with Sr57 (Lr34) are probably present in many Canadian wheat cultivars. Additionally, the gene SrCad is in the cultivars ‘AC Cadillac’ and ‘Peace’ (Hiebert et al. 2011), and also in the newly released variety ‘AAC Tenacious’ (Brown et al. 2015; Fetch unpublished data). SrCad is on chromosome 6DS and is effective against the highly virulent African race TTKSK (Ug99) (Hiebert et al. 2011). Gene Sr42 also is on chromosome 6DS (Ghazvini et al. 2012) at the same locus, thus SrCad appears to be Sr42, an allele, or closely linked.

Table 2. Frequency of virulence of Puccinia graminis f. sp. tritici isolates collected from wheat trap plots, cultivated barley, and wild barley in Manitoba, Ontario, Quebec, and Saskatchewan, Canada in 2009 to single-gene stem rust differential lines.

	Wheat						Cultivated barley						Wild barley
Resistance	Manitoba		Ontario		Quebec		Saskatchewan		Manitoba		Ontario		Manitoba		Saskatchewan
Gene	No.	%	No.	%	No.	%	No.	%	No.	%	No.	%	No.	%	No.	%
Sr5	8	100	1	100	1	100	3	100	1	100	6	100	7	100	15	100
Sr21	8	100	1	100	1	100	3	100	1	100	6	100	6	86	15	100
Sr9e	2	25	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr7b	2	25	1	100	0	0	0	0	0	0	1	17	1	14	0	0
Sr11	2	25	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr6	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr8a	8	100	1	100	1	100	3	100	1	100	6	100	6	86	15	100
Sr9g	8	100	1	100	1	100	3	100	1	100	6	100	7	100	15	100
Sr36	2	25	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr9b	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr30	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr17	8	100	1	100	1	100	3	100	1	100	6	100	7	100	15	100
Sr9a	6	75	1	100	1	100	3	100	1	100	6	100	6	86	15	100
Sr9d	8	100	1	100	1	100	3	100	1	100	6	100	6	86	15	100
Sr10	8	100	1	100	1	100	3	100	1	100	6	100	7	100	15	100
SrTmp	2	25	0	0	0	0	0	0	0	0	0	0	1	14	0	0
Sr24	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr31	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Sr38	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
SrMcN	8	100	1	100	1	100	3	100	1	100	6	100	7	100	15	100

Bulked samples of collections from Manitoba and Saskatchewan also were inoculated on Canadian wheat cultivars considered broadly resistant and on wheat lines with single Sr genes in order to detect new virulent races. Virulence to genes Sr8b, Sr12, Sr14, Sr15, Sr16, Sr18, Sr20, Sr28, Sr34, Sr35, and Sr37 was detected (data not shown). Race QFCSC is virulent to most of these genes including Sr35, which is effective against Ug99 and most North American races of P. graminis f. sp. tritici. With the exception of cultivar ‘Little Club’, no susceptible pustules were detected on wheat lines or Canadian varieties. No new virulence was found in the 2009 or 2010 population of P. graminis f. sp. tritici in Canada that would threaten wheat or barley production.

Puccinia graminis f. sp. avenae

There were 303 isolates derived from 242 collections in 2009, and 196 isolates derived from 145 collections in 2010, for evaluation of P. graminis f. sp. avenae virulence on oat. Many collections had mixed infections, resulting in the determination of two races from a single collection in many instances. In 2009, there were nine races identified in Saskatchewan and 12 in Manitoba (Table 3), and in 2010, the same races were found in Saskatchewan but only 11 were found in Manitoba (Table 4). The most prevalent races from cultivated oat samples in 2009 were TGD (27%), TJJ (22%), TJN (20%), and TGN (12%), and in 2010 were TGN (20%), TJN and TJS (18% each), TGB (13%), and TGD (11%). The most prevalent races from wild oat samples were TGD (31%), TJN (16%), and TGN (15%) in 2009, while races TGN (30%), TGB (22%), TGD (15%), and TJN (13%) were dominant in 2010. One new race (TJL) was detected in three collections from Manitoba in 2009, but would not threaten oat production in Canada as this is avirulent to gene Pg13, which would be present in stem rust resistant oat cultivars (Mitchell Fetch & Fetch 2011). In the United States, the races identified in 2009 were TGN, TJN, TGD, TGL, and TGB, while in 2010 the races found were TJN, TJS, TGS, TGN, SGD, and TGD (Y. Jin, http://www.ars.usda.gov/Main/docs.htm?docid=9757).

Table 3. Frequencies of races of Puccinia graminis f. sp. avenae obtained from cultivated and wild oat in Saskatchewan and Manitoba, Canada in 2009.

	Virulence spectrum	Cultivated oat		Wild oat		Total
Race	(ineffective Pg genes)	No.	%	No.	%	No.	%
Saskatchewan
SGD (NA63)	1, 2, 3, 8, 15	0	0	2	2.4	2	2.3
TGB (NA27)	1, 2, 3, 4, 8	0	0	3	3.7	3	3.4
TGD (NA29)	1, 2, 3, 4, 8, 15	1	20.0	26	31.7	27	31.0
TGN (2006)	1, 2, 3, 4, 8, 12, 15, Pg-a	1	20.0	12	14.7	13	14.9
TJB (NA68)	1, 2, 3, 4, 8, 9	0	0	1	1.2	1	1.1
TJD (NA30)	1, 2, 3, 4, 8, 9, 15	0	0	9	11.0	9	10.3
TJJ (NA67)	1, 2, 3, 4, 8, 9, 13, 15	2	40.0	3	3.7	5	5.7
TJN (2006)	1, 2, 3, 4, 8, 9, 12, 15, Pg-a	1	20.0	20	24.4	21	24.1
TJS (2005)	1, 2, 3, 4, 8, 9, 12, 13, 15, Pg-a	0	0	6	7.3	6	6.9
Manitoba
SGB (NA23)	1, 2, 3, 8	0	0	1	0.6	1	0.5
TGB (NA27)	1, 2, 3, 4, 8	2	5.6	21	11.7	23	10.6
TGD (NA29)	1, 2, 3, 4, 8, 15	10	27.8	55	30.6	65	30.1
TGL (NA28)	1, 2, 3, 4, 8, 12, Pg-a	1	2.8	1	0.6	2	0.9
TGN (2006)	1, 2, 3, 4, 8, 12, 15, Pg-a	4	11.1	28	15.6	32	14.8
TJB (NA68)	1, 2, 3, 4, 8, 9	0	0	4	2.2	4	1.9
TJD (NA30)	1, 2, 3, 4, 8, 9, 15	1	2.8	3	1.7	4	1.9
TJG (NA76)	1, 2, 3, 4, 8, 9, 13	0	0	1	0.6	1	0.5
TJJ (NA67)	1, 2, 3, 4, 8, 9, 13, 15	7	19.4	22	12.2	29	13.4
TJL (new*)	1, 2, 3, 4, 8, 9, 12, Pg-a	1	2.8	2	1.1	3	1.4
TJN (2006)	1, 2, 3, 4, 8, 9, 12, 15, Pg-a	7	19.4	21	11.7	28	13.0
TJS (2005)	1, 2, 3, 4, 8, 9, 12, 13, 15, Pg-a	3	8.3	21	11.7	24	11.1
Total
SGB (NA23)	1, 2, 3, 8	0	0	1	0.4	1	0.3
SGD (NA63)	1, 2, 3, 8, 15	0	0	2	0.8	2	0.7
TGB (NA27)	1, 2, 3, 4, 8	2	4.9	24	9.2	26	8.6
TGD (NA29)	1, 2, 3, 4, 8, 15	11	26.8	81	30.9	92	30.4
TGL (NA28)	1, 2, 3, 4, 8, 12, Pg-a	1	2.4	1	0.4	2	0.7
TGN (2006)	1, 2, 3, 4, 8, 12, 15, Pg-a	5	12.2	40	15.3	45	14.9
TJB (NA68)	1, 2, 3, 4, 8, 9	0	0	5	1.9	5	1.7
TJD (NA30)	1, 2, 3, 4, 8, 9, 15	1	2.4	12	4.6	13	4.3
TJG (NA76)	1, 2, 3, 4, 8, 9, 13	0	0	1	0.4	1	0.3
TJJ (NA67)	1, 2, 3, 4, 8, 9, 13, 15	9	22.0	25	9.5	34	11.2
TJL (new*)	1, 2, 3, 4, 8, 9, 12, Pg-a	1	2.4	2	0.8	3	1.0
TJN (2006)	1, 2, 3, 4, 8, 9, 12, 15, Pg-a	8	19.5	41	15.6	49	16.2
TJS (2005)	1, 2, 3, 4, 8, 9, 12, 13, 15, Pg-a	3	7.3	27	10.3	30	9.9

The letter code race designation described by Fetch and Jin (2007) uses 12 single-gene differential lines (Pg1, 2, 3, 4, 6, 8, 9, 10, 12, 13, 15, and 16) grouped into three subsets of four lines each. The race is determined by the seedling reaction (low or high) on each line. The equivalent North American (NA) race code or year of discovery is provided in parentheses.

*Denotes that TJL is a novel race found in 2009 that was not previously described.

Table 4. Frequencies of races of Puccinia graminis f. sp. avenae obtained from cultivated and wild oat in Saskatchewan and Manitoba, Canada in 2010.

	Virulence spectrum	Cultivated oat		Wild oat		Total
Race	(ineffective Pg genes)	No.	%	No.	%	No.	%
Saskatchewan
SGB (NA23)	1, 2, 3, 8,	0	0	3	7.3	3	5.7
TGB (NA27)	1, 2, 3, 4, 8	2	16.7	9	22.0	11	20.8
TGD (NA29)	1, 2, 3, 4, 8, 15	3	25.0	8	19.5	11	20.8
TGL (NA28)	1, 2, 3, 4, 8, 12, Pg-a	1	8.3	0	0	1	1.9
TGN (2006)	1, 2, 3, 4, 8, 12, 15, Pg-a	3	25.0	10	24.4	13	24.5
TJD (NA30)	1, 2, 3, 4, 8, 9, 15	0	0	1	2.4	1	1.9
TJJ (NA67)	1, 2, 3, 4, 8, 9, 13, 15	0	0	2	4.9	2	3.8
TJN (2006)	1, 2, 3, 4, 8, 9, 12, 15, Pg-a	3	25.0	5	12.2	8	15.1
TJS (2005)	1, 2, 3, 4, 8, 9, 12, 13, 15, Pg-a	0	0	3	7.3	3	5.7
Manitoba
SGD (NA63)	1, 2, 3, 8, 15	1	3.0	1	1.2	2	1.7
TGB (NA27)	1, 2, 3, 4, 8	4	12.1	19	22.1	23	19.3
TGD (NA29)	1, 2, 3, 4, 8, 15	2	6.1	11	12.8	13	10.9
TGL (NA28)	1, 2, 3, 4, 8, 12, Pg-a	0	0	3	3.5	3	2.5
TGN (2006)	1, 2, 3, 4, 8, 12, 15, Pg-a	6	18.2	28	32.6	34	28.6
TJB (NA68)	1, 2, 3, 4, 8, 9	1	3.0	2	2.3	3	2.5
TJD (NA30)	1, 2, 3, 4, 8, 9, 15	1	3.0	0	0	1	0.8
TJG (NA76)	1, 2, 3, 4, 8, 9, 13	1	3.0	2	2.3	3	2.5
TJJ (NA67)	1, 2, 3, 4, 8, 9, 13, 15	4	12.1	3	3.5	7	5.9
TJN (2006)	1, 2, 3, 4, 8, 9, 12, 15, Pg-a	5	15.2	11	12.8	16	13.4
TJS (2005)	1, 2, 3, 4, 8, 9, 12, 13, 15, Pg-a	8	24.2	6	7.0	14	11.8
Total
SGB (NA23)	1, 2, 3, 8	0	0	3	2.4	3	1.7
SGD (NA63)	1, 2, 3, 8, 15	1	2.2	1	0.8	2	1.2
TGB (NA27)	1, 2, 3, 4, 8	6	13.3	28	22.0	34	19.8
TGD (NA29)	1, 2, 3, 4, 8, 15	5	11.1	19	15.0	24	14.0
TGL (NA28)	1, 2, 3, 4, 8, 12, Pg-a	1	2.2	3	2.4	4	2.3
TGN (2006)	1, 2, 3, 4, 8, 12, 15, Pg-a	9	20.0	38	29.9	47	27.3
TJB (NA68)	1, 2, 3, 4, 8, 9	1	2.2	2	1.6	3	1.7
TJD (NA30)	1, 2, 3, 4, 8, 9, 15	1	2.2	1	0.8	2	1.2
TJG (NA76)	1, 2, 3, 4, 8, 9, 13	1	2.2	2	1.6	3	1.7
TJJ (NA67)	1, 2, 3, 4, 8, 9, 13, 15	4	8.9	5	3.9	9	5.2
TJN (2006)	1, 2, 3, 4, 8, 9, 12, 15, Pg-a	8	17.8	16	12.6	24	14.0
TJS (2005)	1, 2, 3, 4, 8, 9, 12, 13, 15, Pg-a	8	17.8	9	7.1	17	9.9

The letter code race designation described by Fetch and Jin (2007) uses 12 single-gene differential lines (Pg1, 2, 3, 4, 6, 8, 9, 10, 12, 13, 15, and 16) grouped into three subsets of four lines each. The race is determined by the seedling reaction (low or high) on each line. The equivalent North American (NA) race code or year of discovery is provided in parentheses.

The virulence frequencies of races of P. graminis f. sp. avenae from wild oat collections continued to shift in 2009 and 2010. Race TJJ, which was first detected in 1998 and is virulent on all Canadian oat cultivars, fell to only 4% frequency in 2010 compared with the high of 54% in 2003 (Fetch et al. 2015). Equally surprising was the decline of race TJS to only 7% in 2010 compared with its peak of 15% in 2007, since it is virulent on all Canadian oat cultivars. The frequencies of races TGB and TGD, which were dominant in the 1990s prior to the arrival of race TJJ, changed substantially in 2009 and 2010. Race TGB increased from only 4% in 2008 to 22% in 2010, while race TGD decreased from 53% in 2008 to 15% in 2010. Race TJN rose slightly from 7% in 2008 to 16% in 2009 and declined slightly to 13% in 2010. The race that is steadily increasing is TGN, which rose from 2% when it was first detected in 2007 to 15% in 2009 and became dominant (30%) in 2010. The dominance of races TGB, TGD, TJN, and TGN in Canada is somewhat surprising since cultivars with resistance to these races are available. The prevalence of races TJN and TGN may be explained by their virulence on gene Pg12 and the Pg-a complex, which is deployed in oat cultivars grown in the southern United States. Races TJN and TGN were commonly detected in the United States in 2009 and 2010, thus it is likely that these races increase there during the winter and spring, then move north into Canada on prevailing winds. The resurgence of races TGB and TGD is more difficult to explain, since they were not detected in the United States in either 2009 or 2010. Resistance to these races is present in most oat cultivars grown in both the United States and Canada. Possibly, TGB and TGD are more fit than others in the population of P. graminis f. sp. avenae, but there is no empirical evidence to support this at present.

Virulence frequencies of P. graminis f. sp. avenae isolates on the 12 single-gene differential oat lines and the Pg-a complex are presented in Tables 5 and 6. All isolates were virulent to Pg1, Pg2, Pg3, and Pg8, and none were virulent on Pg6, Pg10 or Pg16. Virulence was very high to Pg4 (98–100% and 97–99%) and Pg15 (80–90% and 70–87%) in cultivated and wild oat, respectively. Virulence to Pg9 (51–54% and 28–44%) and Pg13 (29% and 13–20%) was common in collections from cultivated oat but less in collections from wild oat, respectively. Compared with 2008, virulence to Pg12 (44–58% and 42–52%) and the Pg-a complex (32–36% and 26–39%) rose dramatically in 2009 and 2010, respectively. The increase in frequency of races TJN, TJS, and particularly TGN during 2009 and 2010 can explain the rapid rise in virulence to Pg12 and the Pg-a complex. While race TJS is virulent on all Canadian oat cultivars, it is fortunate that races TJN and TGN are avirulent to gene Pg13, which is present in Canadian oat cultivars with an intermediate level of resistance. Reliance on only a single gene to protect the Canadian oat crop from a potentially devastating stem rust epidemic is a recipe for disaster; thus the importance of the search for new oat stem rust resistance genes cannot be understated.

Table 5. Frequencies of virulence of Puccinia graminis f. sp. avenae isolates collected from cultivated and wild oat in Saskatchewan and Manitoba, Canada in 2009 to single-gene stem rust differential lines and the Pg-a gene complex.

	Cultivated oat						Wild oat
Resistance	Saskatchewan		Manitoba		Total		Saskatchewan		Manitoba		Total
Gene	No.	%	No.	%	No.	%	No.	%	No.	%	No.	%
Pg1	5	100	36	100	41	100	82	100	180	100	262	100
Pg2	5	100	36	100	41	100	82	100	180	100	262	100
Pg3	5	100	36	100	41	100	82	100	180	100	262	100
Pg4	5	100	36	100	41	100	80	97.6	179	99.4	259	98.9
Pg6	0	0	0	0	0	0	0	0	0	0	0	0
Pg8	5	100	36	100	41	100	82	100	180	100	262	100
Pg9	3	60.0	19	52.8	22	53.7	39	47.6	74	41.1	147	43.9
Pg10	0	0	0	0	0	0	0	0	0	0	0	0
Pg12	2	40.0	16	44.4	18	43.9	38	46.3	73	40.6	111	42.4
Pg13	2	40.0	10	27.8	12	29.3	9	11.0	44	24.4	53	20.2
Pg15	5	100	32	88.9	37	90.2	78	95.1	150	83.3	228	87.0
Pg16	0	0	0	0	0	0	0	0	0	0	0	0
Pg-a	1	20.0	12	33.3	13	31.7	25	30.5	44	24.4	69	26.3

Table 6. Frequencies of virulence of Puccinia graminis f. sp. avenae isolates collected from cultivated and wild oat in Saskatchewan and Manitoba, Canada in 2010 to single-gene stem rust differential lines and the Pg-a gene complex.

	Cultivated oat						Wild oat
Resistance	Saskatchewan		Manitoba		Total		Saskatchewan		Manitoba		Total
Gene	No.	%	No.	%	No.	%	No.	%	No.	%	No.	%
Pg1	12	100	33	100	45	100	41	100	86	100	127	100
Pg2	12	100	33	100	45	100	41	100	86	100	127	100
Pg3	12	100	33	100	45	100	41	100	86	100	127	100
Pg4	12	100	32	97.0	44	97.8	38	92.7	85	98.8	123	96.9
Pg6	0	0	0	0	0	0	0	0	0	0	0	0
Pg8	12	100	33	100	45	100	41	100	86	100	127	100
Pg9	3	25.0	20	60.6	23	51.1	11	26.8	24	27.9	35	27.6
Pg10	0	0	0	0	0	0	0	0	0	0	0	0
Pg12	7	58.3	19	57.6	26	57.8	18	43.9	48	55.8	66	52.0
Pg13	0	0	13	39.4	13	28.9	5	12.2	11	12.8	16	12.6
Pg15	9	75.0	27	81.8	36	80.0	29	70.7	60	69.8	89	70.1
Pg16	0	0	0	0	0	0	0	0	0	0	0	0
Pg-a	4	33.3	12	36.4	16	35.6	17	41.5	33	38.4	50	39.4

Acknowledgements

Ken Dunsmore and Courtney Wolfe are thanked for their expert technical work on the collection and virulence phenotyping of survey samples. Dr Andre Comeau from St. Foy, Quebec is thanked for providing samples from infected wheat.