Virulence of Puccinia coronata var avenae f. sp. avenae (oat crown rust) in Canada during 2010 to 2015

Abstract

Crown rust, caused by Puccinia coronata var. avenae f. sp. avenae (Urban & Marková) (Pca), is an economically significant problem in oat in Canada. The growing of crown rust resistant oat cultivars has been an important management practice. The development of oat cultivars with effective resistance requires knowledge of the virulence genes present in the pathogen population. This study’s objective was to determine the presence and frequency of virulence to 24 oat lines with single Pc genes in isolates of Pca collected in Canada during 2010 to 2015. A total of 692 distinct races were identified, of which 603 were from the eastern Prairie region (EPR) and 89 from eastern Canada. None of the 24 Pc resistance genes studied was effective against all of the Pca isolates from the EPR. The frequency of virulence increased on 15 of the Pc differential lines in the EPR between 2010 and 2015, with the most dramatic increase from 0% to 67% being to Pc91. The most effective resistance genes were Pc94, Pc50, Pc96, Pc97, Pc98 and Pc101. Virulence to Pc58, Pc94, Pc98 and Pc101 was not observed in Pca isolates from eastern Canada. The race structure of the Pca population in Canada was highly variable, with more than 80% of the races identified each year represented by one Pca isolate. The race structure of the Pca populations in eastern Canada and the EPR were distinct, as 80% of the races identified from Pca isolates collected in eastern Canada were not detected in the EPR. The results of this study indicate that the Pca populations in the EPR and eastern Canada continue to evolve in their virulence dynamics, creating a challenge to breed oat lines with effective and stable resistance.

Résumé

La rouille couronnée, causée par Puccinia coronata var. avenae f. sp. avenae (Urban & Marková) (Pca), constitue un problème économique important en ce qui a trait à la production d’avoine au Canada. L’utilisation de cultivars d’avoine résistants à la rouille couronnée est le mode de gestion de la maladie privilégié. Le développement de cultivars d’avoine offrant une résistance efficace requiert de connaître les gènes de virulence présents dans la population d’agents pathogènes. Le but de cette étude était de déterminer l’occurrence et la fréquence de la virulence à l’égard de 24 lignées d’avoine possédant des gènes individuels de Pc dans des isolats de Pca collectés au Canada de 2010 à 2015. En tout, 692 races distinctes ont été identifiées, parmi lesquelles 603 provenaient de la portion est des Prairies (PEP) et 89 de l’est du Canada. Aucun des 24 gènes de résistance de Pc étudiés n’a été efficace contre tous les isolats de Pca issus de la PEP. La fréquence de virulence s’est accrue chez 15 des lignées différentielles de Pc dans la PEP de 2010 à 2015, Pc91 affichant l’accroissement le plus spectaculaire, soit de 0% à 67%. Les gènes de résistance les plus efficaces étaient Pc94, Pc50, Pc96, Pc97, Pc98 et Pc101. La virulence à l’égard de Pc58, Pc94, Pc98 et Pc101 n’a pas été observée chez les isolats de Pca de l’est du Canada. La structure de race de la population de Pca au Canada variait énormément avec plus de 80% des races identifiées chaque année représentées par un isolat de Pca. La structure de race des populations de Pca de l’est du Canada et de celle de la PEP étaient distinctes, puisque 80% des races identifiées à partir d’isolats de Pca collectés dans l’est du Canada n’ont pas été détectées dans la PEP. Les résultats de cette étude indiquent que les populations de Pca de la PEP et de l’est du Canada continuent d’évoluer selon leur propre dynamique de virulence, ce qui pose un défi quant à la sélection de lignées d’avoine à la résistance stable et efficace.

Keywords:

• crown rust
• eastern Canada
• oat
• races
• virulence
• western Canada

Mots clés:

• avoine
• est du Canada
• ouest du Canada
• races
• rouille couronnée
• virulence

Introduction

Oat (Avena sativa L.) is an important cereal crop in Canada, with 1.2–1.4 million ha seeded and a harvest of 2.5–3.9 million metric tonnes annually during 2010 to 2015 (Statistics Canada Table 001–0010, http://www5.statcan.gc.ca/cansim/a26). Approximately 85–90% of the seeded acreage is located in Manitoba, Saskatchewan and Alberta, and another 10% is located in Quebec and Ontario.

Crown rust, caused by Puccinia coronata var. avenae f. sp. avenae (Urban & Marková) (Pca), is considered to be the most serious disease of oat (Harder & Haber, 1992). The areas where crown rust can be an economically significant problem in oat production in Canada include Quebec, Ontario, Manitoba and eastern Saskatchewan (Bailey et al., 2003). The pathogen can cause yield losses on susceptible oat cultivars, as well as a reduction in grain quality. Annual yield losses resulting from crown rust in the eastern Prairie region (EPR) of Canada (Manitoba and eastern Saskatchewan) were reported to average 5% during 2001 to 2005 (Chong et al., 2008). The growing of oat cultivars with genetic resistance has been one of the principal techniques used to control crown rust in Canada.

Breeding for crown rust resistance in new oat cultivars requires the identification of effective sources of resistance. Effective resistance can be defined as resistance to most if not all of the virulence phenotypes or races in the pathogen population. Breeding for resistance to oat crown rust in North America has been occurring since the 1920s. Early efforts utilized resistance genes from cultivated hexaploid oat until the 1950s, when all deployable sources of resistance from cultivated oat had become ineffective against the predominant pathogen races (Martens & Dyck, 1989; McCallum et al., 2007). New sources of genetic resistance from Avena sterilis L., a relative of oat native to the Mediterranean regions, were then utilized in oat breeding programmes (Simons et al., 1978), followed by resistance genes from other oat relatives such as A. magna Murphy and Terrell (Rooney et al., 1994), A. longiglumus Durr. and A. strigosa Shreb. (Aung et al., 1996). A number of effective genes from oat relatives were deployed in Canada, including resistance genes Pc38, Pc39, Pc48 and Pc68, but changes in the virulence patterns of Pca soon rendered these genes ineffective (Chong & Seaman, 1991; Chong & Zegeye, 2004; Chong et al., 2008, 2011).

Puccinia coronata var avenae f. sp. avenae is a highly variable pathogen, with two distinct populations in Canada; the eastern Canadian population (Ontario and Quebec), and the EPR (Chong & Kolmer, 1993). Both populations have characteristics typical of sexually reproducing cereal rust populations (Groth & Roelfs, 1982; Chong & Kolmer, 1993; Kolmer & Chong, 1993), with many infrequent races and few dominant races. Virulence assessments conducted from 2002 to 2009 found that 60–85% of races were represented by one isolate each year (Chong et al., 2008, 2011). Chong et al. (2008) found that only 6 (2.2%) of 272 races identified were observed in all five years from 2002 to 2006, with 16 (5.8%) races identified in four of the years, and 20 races (7.4%) identified in three of the years. Virulence assessments from isolates collected from 2007 to 2009 found that ~8% of 341 races occurred in all three years (Chong et al., 2011).

Virulence frequencies on specific Pc genes can also change rapidly. Resistance genes Pc38 and Pc39 were first deployed in combination in the oat cultivar ‘Dumont’ released in 1982, which was followed by the cultivars ‘Riel’, ‘Robert’ and ‘AC Marie’ (McCallum et al., 2007). There was no virulence to this gene combination found until 1987, when two isolates virulent to this gene combination were detected (Chong, 1988). By 1990, 92% of the isolates collected from commercial oat and 43% of isolates from wild oat possessed virulence to this gene combination (Chong & Seaman, 1991). A series of oat cultivars possessing the gene combination Pc38 and Pc39, augmented with the resistance gene Pc68 were then released between 1992 and 2003 (McCallum et al., 2007). Virulence to Pc68 was detected for the first time in one isolate of Pca from the EPR in 1991 (Chong & Seaman, 1993). By 2009, virulence to Pc68 was found in ~82% of isolates from commercial oat and 45% of isolates collected from wild oat (Chong et al., 2011).

The present study is a continuation of annual crown rust surveys that have been conducted by Agriculture and Agri-Food Canada personnel at the Cereal Research Centre, Winnipeg, MB, and now, the Morden Research and Development Centre, Morden, MB, since 1929 (Peturson, 1935). The main objectives of these surveys have been: (1) monitor the incidence and severity of the crown rust pathogen, (2) identify pathogenic races to determine the effectiveness of particular crown rust resistance genes for incorporation into new varieties, and (3) determine if current oat varieties have effective resistance against current Pca populations in Canada. This manuscript reports incidence and severity of Pca, and the presence and frequency of virulence in Pca populations in the various oat producing regions of Canada during 2010 to 2015.

Materials and methods

Surveys for crown rust in the EPR were conducted from 29 July to 19 August, and in Ontario from 5 July to 19 August in 2010 (Menzies et al., 2013a). In 2011, the surveys were conducted from 21 July to 21 September in the EPR and between 18 July and 18 August in Ontario and Quebec (Menzies et al., 2013b). In 2012, the surveys were conducted between 23 July and 12 September in the EPR and between 6 July and 31 July in Ontario, Quebec and Prince Edward Island (Menzies et al., 2014). In 2013, the surveys occurred between 7 August and 28 August in the EPR and between 16 July and 7 August in Ontario, Quebec and Prince Edward Island (Menzies et al., 2015). In 2014, the surveys were conducted between 12 August and 22 August in the EPR, and 16 July and 7 August in Ontario, Quebec and Prince Edward Island (Menzies et al., 2016). In 2015, the surveys occurred between 2 August and 10 August in the EPR and during July in Ontario (Menzies et al., 2017). Samples collected in eastern Canada (Ontario, Quebec and Prince Edward Island) and the EPR were used for virulence assessment. The surveys in the EPR and eastern Ontario were conducted to determine the incidence and severity of crown rust in fields of cultivated and wild oats. The modified Cobb scale (Peterson et al., 1948) was used in the EPR, and the per cent leaf area covered by pustules was used in eastern Ontario to estimate crown rust severity.

Collections of Pca uredinia were obtained from cultivated oat fields and nurseries, and from fields with infestations of wild oats (Avena fatua L.). Wild oats were included as a non-selective host of Pca, because there have been no reports of resistance to crown rust in wild oats in Canada. Uredinia collections were air-dried at room temperature overnight and stored at 5°C until processed as described by Chong et al. (2008). Urediniospores from each collection were transferred using a sterile spatula and a small amount of water to seedlings of ‘Makuru’ which had been treated with maleic hydrazide (0.28 g L⁻¹ water) to suppress emergence of secondary leaves. Inoculated seedlings were incubated overnight at 18°C in a dew chamber and subsequently incubated in a greenhouse at 18–25°C with a 16 h photoperiod supplemented with high-pressure sodium lights. The seedlings in each pot were grown inside a transparent plastic cone to minimize cross contamination. Seedlings were thinned to one plant per pot with the primary leaf trimmed to contain a single uredinium at ~7 days after inoculation, when pustules were beginning to form. Urediniospores from a single uredinium (a single pustule isolate or spi) were collected with a micro-cyclone collector into size #00 gelatin capsules at two weeks post-inoculation. The urediniospores from each spi were mixed with 400 μL of light industrial oil (Bayol, Esso Canada) and sprayed individually onto susceptible ‘Makuru’ seedlings to increase the urediniospores for inoculation onto a set of differential oat lines.

Each spi was subsequently inoculated onto a set of differential oat lines to determine the race as described by Chong et al. (2000). The oat differential set consisted of 16 standard and eight supplemental differential lines (Table 1). The cultivar ‘Makuru’ was included as a susceptible check. Inoculated differential sets were incubated overnight in a dew chamber and subsequently incubated in a greenhouse as described above. Seedlings were assessed for infection type (IT) at 12–14 d after inoculation using the 0–4 scale of Murphy (1935). Infection types of 0, 1 and 2 were considered to be an avirulent reaction, and ITs of 3 and 4 were considered to be a virulent reaction.

Table 1. The differences in frequency of virulence to crown rust differential oat lines of isolates of Puccinia coronata var. avenae f. sp. avenae collected from wild oat and commercial oat plants in the eastern Prairies (eastern Saskatchewan and Manitoba), and from oat plants in eastern Canada (Ontario, Quebec and Prince Edward Island) in 2010 to 2015.

	Per cent of isolates virulent to the oat line
	Eastern Prairies
Oat lines and Pc gene present^a	Isolates from commercial oat	Isolates from wild oat	Eastern Canada
Pc38	98.0a^b	96.5a	86.5b
Pc39	92.5a	95.7a	62.2b
Pc40	29.2ab	42.7a	13.9b
Pc45	30.7a	29.9a	2.6b
Pc46	43.8a	47.2a	27.1b
Pc48	23.7b	19.0b	77.5a
Pc50	8.9	7.1	5.2
Pc51	63.8a	59.2a	6.2b
Pc52	24.6ab	17.9b	44.6a
Pc54	2.5	6.2	6.2
Pc56	81.6a	81.0a	62.9b
Pc58	4.6	4.8	0.0
Pc59	9.4a	10.0a	0.8b
Pc62	14.7	9.9	8.8
Pc64	12.5	15.1	11.4
Pc68	64.5ab	54.6b	76.5a
Pc91	30.9a	19.1a	1.4b
Pc94	2.0	0.6	0
Pc96	3.4	5.2	4.8
Pc97	3.1	5.0	4.8
Pc98	5.3a	4.2a	0.0b
Pc101	3.2	2.9	0.0
Pc103–1	11.7a	8.4ab	2.5b
Pc104	15.2	17.3	12.8
Total number of isolates	201	919	115

^a The oat differential set consisted of a standard set of 16 primary differentials (Pc38 to Pc68; Chong et al., 2000) and eight supplemental differential lines. All are single gene lines, each with a different Pc gene for resistance to P. coronata var. avenae f. sp. avenae, except for Pc58 and Pc59. Hoffman et al. (2006) reported that Pc58 resistance is conditioned by three loosely linked genes, and Brouwer (1983) reported Pc59 resistance to be conditioned by three unlinked genes.

^b Values in the same row followed by different letters are significantly different at P <0.05.

The data of percentage of virulent isolates on the 24 oat Pc genes were subjected to analysis of variance without transformation. The two sources of collection, cultivated oat and wild oats, were treated as replicates in the analysis to examine the possible significant shifting in virulence spectrum in the Prairies over the years 2010 to 2015. When the different effectiveness of the 24 resistance genes was compared, the years of collection were treated as replicates. The analysis of variance was conducted using the general linear model in SAS version 9.3 for personal computer (SAS Institute Inc., Cary, NC, USA). The means of each year for a specific Pc gene or the means of each Pc gene in all six years were separated by Fisher’s Least Significant Difference (LSD_0.05) test, if the F-tests were significant (P < 0.05). Correlations for virulence among the 24 Pc genes during 2010 to 2015 were determined using Fisher’s Exact Test and Pearson’s Correlation Coefficient Test using the SAS software.

Results

Crown rust was observed on cultivated and wild oat plants each year in the EPR and on cultivated oat in eastern Ontario (Table 2). In the EPR, the percentage of cultivated oat fields with Pca infected plants ranged from 17% in 2013 to 69% in 2015. The mean incidence of infected plants in infested fields ranged from 6% in 2012 to 55% in 2011, while the mean severity of infection ranged from 1 MS in 2012 to 9 MS in 2010. Puccinia coronata var. avenae f. sp. avenae was found in 60% (2010) to 92% (2014) of fields infested with wild oats in the EPR (Table 2). The per cent of plants infected in wild oat fields was highest in 2010 at 84%, and then generally decreasing each year until 2015, when it was 26%. The mean severity on wild oat plants ranged from 15 S to 26 S in 2010 to 2014, and was only 2 MS in 2015.

Table 2. Incidence and severity of Puccinia coronata var. avenae f. sp. avenae in the Eastern Prairie Region (eastern Saskatchewan and Manitoba), and Eastern Ontario during 2010 to 2015.^a.

	Eastern Prairie Region
Year	Host	Number of fields assessed	Number (percentage) of fields with crown rust	Mean incidence^b (range among fields)	Mean severity^c (range among fields)
2010	Wild Oat	203	122 (60%)	84% (trace to 100%)	26 S (trace to 80 S)
2011	Wild Oat	218	140 (64%)	62% (trace to 100%)	19 S (trace to 100 S)
2012	Wild Oat	191	138 (72%)	60% (trace to 100%)	24 S (Trace to 90 S)
2013	Wild Oat	100	89 (89%)	60% (trace to 100%)	15S (trace to 80 S)
2014	Wild Oat	71	65 (92%)	43% (trace to 100%)	16 MS (trace to 60 S)
2015	Wild Oat	100	82 (82%)	26% (trace to 100%)	2 MS (trace to 20%)
2010	Common Oat	53	29 (55%)	51% (trace to 100%)	9 MS (trace to 60 S)
2011	Common Oat	39	24 (62%)	55% (trace to 100%)	7 MS (trace to 60 S)
2012	Common Oat	30	12 (40%)	6% (trace to 100%)	1 MS (trace to 90 S)
2013	Common Oat	18	3 (17%)	17% (trace to 100%)	3 MS (trace to 60S)
2014	Common Oat	20	12 (60%)	25% (trace to 100%)	6 I (trace to 30 S)
2015	Common Oat	66	45 (69%)	23% (trace to 100%)	3 MS (trace to 10 MS)
	Eastern Ontario
2010	Common Oat	15	15 (100%)		1% (1%)
2011	Common Oat	12	12 (100%)		18% (10 to 25%)
2012	Common Oat	15	9 (60%)		4% (1 to 10%)
2013	Common Oat	19	18 (95%)		22% (10 to 30%)
2014	Common Oat	22	21 (95%)		17% (5 to 30%)
2015	Common Oat	21	19 (90%)		9% (1 to 30%)

^a The table is a summary of information previously published in the Canadian Plant Disease Survey. For EPR, see Menzies et al. (2013a, 2013b, 2014, 2015, 2016, 2017). For eastern Ontario, see Xue & Chen (2011, 2012, 2013, 2014, 2015, 2016)).

^b Incidence refers to the percentage of plants in the field infected with Puccinia coronata var. avenae f. sp. avenae.

^c The modified Cobb scale (Peterson et al., 1948) was used in the EPR, and the per cent leaf area covered by pustules was used in eastern Ontario to estimate crown rust severity.

The percentage of cultivated oat fields in eastern Ontario found to have Pca infected plants was between 90% and 100% in all years, except for 2012, where 60% of the fields had infected plants (Table 2). The lowest severities on infected oat plants were observed in 2010 (1%) and 2012 (4%), and the highest severity was in 2013 (22%).

Virulence of Puccinia coronata var. avenae f. sp. avenae

A total of 1,120 spi were established from samples collected from cultivated oat and wild oat in the EPR, and 115 spi from collections from oat disease nurseries and cultivated oat from eastern Canada, during 2010 to 2015. The mean number of virulence genes per spi collected in Canada were similar from 2010 to 2012 (5.4 in 2010, 4.8 in 2011 and 5.2 in 2012), but increased to 6.1 in 2013, 7.4 in 2014 and 8.3 in 2015. The mean number of virulence genes possessed by spi from wild oat (6.6) or commercial oat lines (6.8) in the EPR over the six years were not different, but were higher than the number of virulence genes possessed by spi from eastern Canada (5.2).

There were no differences in frequency of virulence for each Pc gene in the spi from the EPR between spi collected from wild oat or from commercial oat for the combined six years of virulence frequency data (Table 1). The frequencies of virulence to Pc38, Pc39, Pc45, Pc46, Pc51, Pc56, Pc59, Pv91 and Pc98 were significantly less in eastern Canada as compared with wild oat or commercial oat in the EPR, while the frequency of virulence to Pc48 was greater in eastern Canada. Frequency of virulence to Pc40 in spi from eastern Canada was less than the virulence frequency observed from spi from wild oat, but not commercial oat, while the frequencies of virulence to Pc52 and Pc68 were higher in eastern Canada as compared with wild oat. The frequency of virulence found in eastern Canada for Pc103-1 were less than the frequency found from spi on commercial oat, but not wild oat in the EPR. There were no differences in the frequency of virulence of the spi from eastern Canada, wild oat and commercial oat from the EPR for Pc50, Pc54, Pc58, Pc62, Pc64, Pc94, Pc96, Pc97, Pc101 and Pc104. Frequencies of virulence of Pca isolates collected in the EPR on the 24 differential lines is presented in Table 3. Virulence to Pc38 and Pc39 were greater than 94% each year in the EPR, except for Pc39 in 2014 which was 80.4%. The next most common virulence was to Pc56 which was between 66.1 and 97.8%. Virulence frequencies of 25–83% each year were observed with Pc40 (except 2014), Pc46, Pc51 and Pc68. Frequencies of virulence to genes Pc50, Pc54, Pc94, Pc96, Pc97, Pc98 and Pc101 were under 13% each year. Pc94 was the most effective resistance gene during 2010 to 2015 in the EPR, with 4.8% of spi in 2010, 1.3% in 2011, 1.7% in 2015 and none of the spi in the other years having virulence to this gene. There were significant increases in the frequencies of virulence to Pc45, Pc46, Pc48, Pc51, Pc52, Pc54, Pc56, Pc58, Pc59, Pc62, Pc91, Pc96, Pc101, Pc103-1 and Pc104 from 2010 to 2015. The most dramatic increase in virulence was to Pc91, in which the frequency of virulence increased from 0% in 2010 and 2011 to 66.6% in 2015 in the EPR. The only significant decrease in frequency of virulence was to Pc40.

Table 3. Percentage of virulent isolates of Puccinia coronata var. avenae f. sp. avenae collected from the eastern Prairie Region (eastern Saskatchewan and Manitoba) of Canada on 16 standard and eight supplemental crown rust differential oat lines in 2010 to 2015.

Oat lines and Pc genes present^a	Per cent of isolates virulent to the oat line
	2010	2011	2012	2013	2014	2015
Pc38	97.5	96.4	97.4	96.1	96.5	99.6
Pc39	95.0	99.5	97.4	94.1	80.4	98.3
Pc40	44.8a^b	47.2a	45.2a	37.3ab	14.8b	26.2ab
Pc45	12.3c	17.4bc	24.4b	18.1bc	50.5a	58.9a
Pc46	36.5c	29.6c	25.9c	50.8b	69.5a	60.9ab
Pc48	5.7b	8.9b	9.9b	8.6b	45.6a	49.2a
Pc50	11.7	7.1	12.9	6.7	4.2	5.4
Pc51	47.4c	46.0c	59.0bc	56.3c	82.5a	77.9ab
Pc52	5.7b	8.4b	12.0b	7.1b	50.1a	44.2a
Pc54	4.1b	0.5b	2.1b	1.6b	4.9ab	12.9a
Pc56	72.6b	66.1b	83.4ab	81.5ab	86.7ab	97.8a
Pc58	2.9b	0.7b	1.6b	2.8b	2.1b	18.1a
Pc59	8.9b	6.1b	6.7b	7.5b	7.4b	21.6a
Pc62	10.5bc	2.1c	10.2bc	15.0b	11.9b	24.0a
Pc64	21.7	10.6	14.2	20.8	1.4	14.2
Pc68	68.5	63.4	47.6	67.8	50.9	59.1
Pc91	0.0b	0.0b	5.7b	19.0b	58.9a	66.6a
Pc94	4.8	1.3	0.0	0.0	0.0	1.7
Pc96	2.2b	0.5b	4.5ab	8.2a	1.4b	9.0a
Pc97	0.8	2.8	0.5	7.1	7.4	5.7
Pc98	7.3	4.1	0.5	2.7	6.7	7.1
Pc101	0.3b	1.0ab	2.7ab	0.8ab	6.0ab	7.6a
Pc103–1	3.5b	2.5b	0.5b	5.5b	26.6a	21.7a
Pc104	10.4bc	6.8c	15.0b	13.7bc	36.5a	15.2b
# of isolates	254	246	218	149	85	168

^aThe oat differential set consisted of a standard set of 16 primary differentials (Pc38 to Pc68; Chong et al., 2000) and eight supplemental differential lines. All are single gene lines, each with a different Pc gene for resistance to P. coronata var. avenae f. sp. avenae, except for Pc58 and Pc59. Hoffman et al. (2006) reported that Pc58 resistance is conditioned by three loosely linked genes, and Brouwer (1983) reported Pc59 resistance to be conditioned by three unlinked genes.

^bValues in the same row followed by different letters are significantly different at P <0.05.

Virulence frequencies of Pca isolates collected in eastern Canada are presented in Table 4. Virulence frequencies of 50% or greater were found for Pc38, Pc48 and Pc68 in all six years in eastern Canada. Virulence to Pc39, Pc52 and Pc56 was also common at frequencies of 22–100%. Isolates with virulence to Pc58, Pc94, Pc98 and Pc101 were not found, and isolates with virulence to Pc59 and Pc91 were found at frequencies of 5% or less.

Table 4. Percentage of virulent isolates of Puccinia coronata var. avenae f. sp. avenae collected from Eastern Canada (Ontario, Quebec and Prince Edward Island) on 16 standard and eight supplemental crown rust differential oat lines in 2010 to 2015.

Oat lines and Pc genes present^a	Per cent of isolates virulent to the oat line
	2010	2011	2012	2013	2014	2015
Pc38	85.0	70.8	80.6	87.5	95.0	100.0
Pc39	55.0	33.3	38.7	81.3	65.0	100.0
Pc40	10.0	12.5	3.2	12.5	20.0	25.0
Pc45	0.0	0.0	3.2	12.5	0.0	0.0
Pc46	30.0	12.5	22.6	37.5	35.0	25.0
Pc48	80.0	58.3	67.7	68.8	90.0	100.0
Pc50	5.0	4.2	3.2	18.8	0.0	0.0
Pc51	5.0	0.0	9.7	12.5	10.0	0.0
Pc52	40.0	25.0	22.6	50.0	30.0	100.0
Pc54	10.0	4.2	3.2	0.0	20.0	0.0
Pc56	25.0	45.8	58.1	68.8	80.0	100.0
Pc58	0.0	0.0	0.0	0.0	0.0	0.0
Pc59	0.0	0.0	0.0	0.0	5.0	0.0
Pc62	15.0	8.3	3.2	6.3	20.0	25.0
Pc64	15.0	29.2	3.2	6.3	15.0	0.0
Pc68	75.0	50.0	51.6	87.5	95.0	100.0
Pc91	0.0	0.0	3.2	0.0	5.0	0.0
Pc94	0.0	0.0	0.0	0.0	0.0	0.0
Pc96	10.0	12.5	6.5	0.0	0.0	0.0
Pc97	0.0	12.5	6.5	0.0	10.0	0.0
Pc98	0.0	0.0	0.0	0.0	0.0	0.0
Pc101	0.0	0.0	0.0	0.0	0.0	0.0
Pc103–1	5.0	0.0	0.0	0.0	10.0	0.0
Pc104	5.0	8.3	3.2	25.0	10.0	25.0
# of isolates	20	24	31	16	20	4

^aThe oat differential set consisted of a standard set of 16 primary differentials (Pc38 to Pc68; Chong et al., 2000) and eight supplemental differential lines. All are single gene lines, each with a different Pc gene for resistance to P. coronata var. avenae f. sp. avenae, except for Pc58 and Pc59. Hoffman et al. (2006) reported that Pc58 resistance is conditioned by three loosely linked genes, and Brouwer (1983) reported Pc59 resistance to be conditioned by three unlinked genes.

Associations for virulence among the 24 crown rust resistance genes were common (Table 5), with the virulence to each of the Pc genes being correlated to the virulence on a minimum of one other Pc gene. There were 276 possible combinations of genes, and 109 of these combinations were positively or negatively associated, 10 being negative associations and 99 being positive associations. Virulence to Pc45 and Pc56 was associated to virulence to 16 other Pc resistance genes, with all associations being positive except for the association between virulence genes Pc45 and Pc68. Virulence to Pc91 was associated to virulence to 15 other Pc genes, with only one association being negative; Pc91 and Pc40. Virulence to Pc50 was associated only to virulence to Pc94, which was a positive association. Virulence to Pc94 and Pc97 was positively associated to virulence to only two other Pc genes; Pc50 and Pc98, and Pc54 and Pc103-1, respectively. The greatest number of negative associations were observed with virulence to Pc68, which was negatively associated with virulence to five other Pc lines, followed by the negative associations of Pc40 to the virulence of three other Pc genes.

Table 5. The association for virulence to 24 crown rust resistance genes in the Canadian population of Puccinia coronata var. avenae f. sp. avenae collected from wild oat and commercial oat plants in the eastern Prairies (eastern Saskatchewan and Manitoba), and from oat plants in Eastern Canada (Ontario, Quebec and Prince Edward Island) in 2010 to 2015.

Pc gene^a	Pc40	Pc45	Pc46	Pc50	Pc38	Pc39	Pc48	Pc68	Pc51	Pc52	Pc58	Pc59	Pc54	Pc56	Pc62	Pc64	Pc94	Pc96	Pc91	Pc97	Pc98	Pc101	Pc103–1
Pc45	++
Pc46		+++
Pc50
Pc38		+	+
Pc39	+++	+++			+++
Pc48	–	+++	++			–
Pc68		–			++	+	++
Pc51		+++	+++		+	+++		–
Pc52	–	+++	+++				+++		+++
Pc58		+++	++			+	+++		+++	+++
Pc59	+++	+++	++						++	+	+++
Pc54		+++					++
Pc56		+++	+++		+++	+++	+		+++	+++	++	+++
Pc62							++	–	++				+++	++
Pc64	++											-		+	+++
Pc94				+
Pc96	+++							–							+++	+++
Pc91	–	+++	+++			+++	+++		+++	+++	+++	+++	+++	+++	+++
Pc97													+++
Pc98						+								+	+		+++
Pc101		+++	+					-	++	++	+++	+++		+					+++
Pc103–1		+++	+++				++		+++	++			+++	+++	+++	+			+++	+++	+++
Pc104		+++	++			++	++		++	+++				++					+++				+++

^a +, ++, and +++ indicate significant positive associations and -, –, and – indicate significant negative associations among the Pc genes at the 0.05, 0.01 and 0.001 levels of probability, respectively as determined using Fisher’s Exact Test and Pearson’s correlation coefficient test.

Races of Puccinia coronata var. avenae f. sp. avenae

A total of 603 races were identified from collections in the EPR, and 89 races from those from eastern Canada during 2010 to 2015, of which 463 (77%) of the races from the EPR, and 80 (91%) of the races from eastern Canada were detected in only one year (Table 6).There were 450 unique races from wild oat spi, 94 unique races from commercial oat spi, and 59 races common to both wild oat and commercial oat spi in the EPR. The most common race was BRBG, which was represented by 35 spi (3%) and was found every year except 2014. The next most common race was LRLG, which was represented by 24 (2%) spi, and was the only race detected every year. Fourteen other races were detected as 10 to 22 spi, with races BQBB, BRLG, DRLG, GQLG, LRBG and NRLG detected in five of the six years. None of these latter isolates were observed in 2014 except BRLG, which was observed in 2014 but not 2015. In eastern Canada, there were 71 of 89 races identified that were not detected in the EPR. Of these, 74 (83%) races were detected as only one spi over the six years (Table 6). Twelve of the other 15 races occurred as 3 or fewer spi. Race BBBB (avirulent to all Pc genes) was represented by 4 spi (3.5%) from 2011. Two races were represented by 5 spi (4.3%); race BRBG, which was found in 2011, 2012 and 2013, and race BTGG, which was found in 2010, 2011, 2014 and 2015.

Table 6. Races of Puccinia coronata var. avenae f. sp. avenae identified from collections made in the eastern Prairies (eastern Saskatchewan and Manitoba) and eastern Canada (Ontario, Quebec and Prince Edward Island) during 2010 to 2015.

Year	Collection location and host	Races identified^a
2010	Eastern Prairies, wild oats	BGBB, BJBB, BLBH, BLCG, BLLG-103–1, BMBG, BQBB, BQBD-97, BQBG (5), BQBK, BQLB, BQLG (2), BQLG-104, BQLH, BRBB (6), BRBG (7), BRBG-98, BRBG-104, BRBK, BRFG, BRLB, BRLB-98, BRLG (4), BRLG-103–1, BRLG-104, BRLH, BTGQ, CQBG-104, CQLB, CRBB, CRBG, CSQH, DGLC, DLBG, DMLB, DMLG, DQBB, DQBG (3), DQBJ-104, DQCH, DQLB (3), DQLB-98, DQLG (3), DQLG-104, DQLK, DRBG (4), DRBH, DRCG, DRLB (3), DRLG (4), DRMG, DTQG, DTQK, FLLG, FRBG (2), FRLG-97, GHBH, GQLG (2), GQLG-103–1, GQLG-104, GQMG-96 98, GRLB, GRLG-104, GSQG (2), GSQK, GTQQ-104, JQBG, JQLB, JQLG, JRLG, JSGG, JSGG-103–1, LHBH, LHMQ, LLBB, LLLG, LMBB, LQBB (2), LQBB-104, LQBG (2), LQBG-104, LQBK, LQBK-96, LQLB, LQLG, LQLG-98, LQLG-104, LQLH-103–1, LQLK, LQLK-96, LRBB (4), LRBG (2), LRBG-98 104, LRBG-104, LRBH, LRBJ, LRCB, LRCB-97, LRCG (4), LRLB (3), LRLB-103–1, LRLG (7), LSGG, LTQG, MRBB, MRBG (2), MRLG, NBMG, NGLH, NHLB, NPBG-104, NQBB-104,NQBG-103–1 104, NQBK, NQLG, NQLJ (2), NQMG, NRBB, NRBB-104, NRBG (2), NRCG, NRLB, NRLC, NRLG (5), NRLH, NRLL, NRPB, NTGG-103–1 104, NTQG, QQBB, QQCG-96 101, QQPG-103–1, QSQG, SQMG, SRLG, SRLG-97, TQLB, TTQG-96 104
2010	Eastern Prairies, cultivated oats	BMBG, BQPG, BRBC, BRBG-98 (2), BRBH, BRLD, BRLG, BRLG-98, BRPK, CQLB, CQLG, CRLG, DHBB-94, DQBB-103–1, DRBG (4), DRBH-94, DRBH-104, DRBR, DRLG, DRLK, FRLK-98, GLBH, GQLG-104, JQLG, JRLG-98, LQBB, LQCB, LQLG, LRBB-104, LRBF, LRBG, LRBH-98, LRBH-104, LRBQ, LRLB, LRLG (2), LRLH, LSGG, LTHG, MRLB-94, MRLD-94, NRBN, NRLG, NRLG-104, PRBG-96, QRBK, RRBB-94
2010	Eastern Canada, cultivated oats	BBBC, BLBB-96, BPBB, BPBG, BSBG, BSGD, BTBB (2), BTBC, BTGB-96, BTGF-103–1 104, BTGG (2), CPQB, DCBB, DPGB, DTBB, DTBS, LPBL, NBBG
2011	Eastern Prairies, wild oats	BGLB, BGLH, BLBB, BMLG, BQBB (7), BQBB-98, BQBG (2), BQBK, BQCB-98, BQLB (3), BQLG (2), BRBB (5), BRBC-97, BRBG (11), BRBG-103–1 104, BRBH, BRBQ, BRCB, BRLB (2), BRLG (5), BRLG-98, BRLH (3), BRLM-104, BSGG-97, BSQG, BTCH, BTGC, BTQG-104, CQLG, CQLH, CRLB (2), DQBB, BQBG, DQLG, DQLJ, DRBB (2), DRBB-103–1, DRBC, DRBG (4), DRBG-96, DRLB, DRLC, DRLG, DRLG-104, DSQB, DSQB-98 104, DSQG-104, GQBG, GQBH, GQLG, GQLG-97, GQLG-97 104, GQMG-101, GRLG, JQBG-96, JQBH, JQLG, JSGG, JSQG, JSQJ, KQBB, LGLB, LLBG, LQBB (4), LQBC, LQBG (3), LQBG-103–1, LQBG-104, LQBH, LQLB (2), LQLG (3), LQLH, LQMG, LRBB (3), LRBG (6), LRBG-97, LRBG-104, LRBH, LRBH-98, LRBH-104, LRCB, LRCG, LRDG-104, LRLB, LRLG (4), LRLH, LRLH-103–1, LRLJ-104, LRMG, LSGG, LTQC, MGRG, MQBB, MQLG, MRBB, MRBH, MRLG, NGBH, NGLG, NMBB, NQBG, NQCG-104, NQLB, NQLG (2), NQLG-103–1, NQLH, NRBB (3), NRBB-104, NRBG (7), NRBG-98, NRLG (4), NSGG-104, NSTH, NTQB, PQBG, PRBG, PTGG (2), QQBG, QQLB, QQLG, QRBG, QRBG-97, QRLG, RQBB-101 104, RQMG-101, SQBB-104, SQLG (2), SRBG, SSMG, STQB, TRBG-101
2011	Eastern Prairies, cultivated oats	BQLB, BQLG, BRBB, BRBB-94, BRBG, BRBH-98 103–1, BRCG, BRLG (3), BTGB, BTBG-104, CQLC, DQLB, DQLG, DRBB, DRBG (2), DTQG, GGLG-104, GQLG-97, GRLG, LHBB, LQLG, LRBB (2), LRBF, LRLB, LRLG (2), LRMG, NRBG, NRLG, QQBG, QQLG, QRBG, RRBB, SSLG
2011	Eastern Canada, cultivated oats	BBBB (4), BBBB-96, BBBC-96, BJBB, BPBC-96, BPBG, BPBH, BPBH-97, BPGG, BQBH, BRBG, BSBB-97, BSBJ, BTGG, BTGH-104, DLBB, DLBG-97, DTBB, LPGB, LPGF-104, MPGQ,
2012	Eastern Prairies, wild oats	BGBB, BLBB, BLBG, BLLG, BMBB, BMBL, BQBB (3), BQBG (8), BQBG-104, BQBH, BQLB (2), BQLF, BQLG (2), BQLL, BRBB (5), BRBB-104, BRBG (9), BRBG-104 (3), BRGG, BRLB (4), BRLB-104, BRLG (3), BSGB, BTGG (2), BTSG (2), CHLQ-103–1, CQBB, CRBG (2), CRBH, DHMG-96, DNGB-97, DQBB, DQBG (2), DQLG (4), DQLK-91, DRBB-104, DRBG (2), DRBL, DRLB, DRLG, FQBG, FRLB, GLLB, GQBG, GQLB, GQLG (4), GQMG-104, GRLG, GSQG-104, GTGG-96 104, HLLG, JLBH, JQBB, JQBG, JQBK, JQBM, JQLB, JQLG (2), JQLG-104, JQLJ, JQLQ-91, JRLG (2), JRLG-98, JSGG-103–1, JSSG-104, LGCG, LGLB, LGLG, LHLG-104, LJBB, LQBG (3), LQBG-97, LQBK, LQBK-96 (3), LQCB, LQLB, LQLG (3), LQLG-104, LQLK, LQLK-96 (3), LRBB (3), LRBB-104, LRBG (3), LRBG-104, LRBL-104, LTQG, MQLN, MRBB, BRBG, MRLG (2), NHCB, NQBB, NQBG-104, NQCB, NQLG (3), NRBG (3), NRBH, NRLB-104, NRLG, NRLG-103–1 104, NRLG-104, NSGG-91, NTQG, PQBG, PTJB, QQBG, QQBG-98 104, QQLB, QQLG, QQLH, QRBB, QRLG, QSQH, QSTG, QSTG-91 101, SMBL, SRLF-104, SRLG, STQG, TQLG
2012	Eastern Prairies, cultivated oats	BQLB, BRLG (2), BSQG, CQBH, CRLG, FRBG, GQBG-91, GQBK, GQLG, GQLG-91 104, GQQG, GSQG-104, GTJK-91, LQBG, LQLK-96, LRBG (3), LRCG, NRBG, NRMG, PRLK-104, SQMG-101
2012	Eastern Canada, cultivated oats	BBBB-96, BBBG (2), BFBB, BFBG, BKBG, BMBG, BNBB, BNBG, BNGB, BNQG, BQBB, BRBB, BRBF, BRBG (3), BRBG-96 97, BRGB, BSGG (2), BTGB, BTGG-104, DQBB, DQLB, DRBB, DRBG (2), DSGB-97, JQLQ-91, MNBB
2013	Eastern Prairies, wild oats	BLBG, BQBB, BQBL, BQLG, BQLG-96, BQMG, BRBB (2), BRBG (4), BRBG-98, BRBG-104, BRBH, BRLH, BRLH-104, BRLQ, BRLS-97 103–1 104, BSGB-91, DLLG-97 104, DMBH, DMLG (2), DQBG (4), DQLG, DRBB, DRBG, DRBK, DRLG, DRLG-98 (2), DRLG-104, DRLH, DRLH-97, DRLH-97 104, DRLK-91 104, DSQH-91, DTBG, DTQL-103–1, FTQH-104, GHLH-91 104, GLLH, GQBG, GQLG, GRLC, GSQG, HTBF-97, JNQD, JQLG, JQPG-98 103–1, JRBG (2), JRCG. JRLH-91 97 104, JSLH, JTQG-91, LGLK, LMBM-96 97, LQBG, LQBG-96, LQBK-96 (4), LQCG, LQLB, LQLF-96, LQLK, LQLK-96 (2), LRBG (2), LRBG-91 104, LRBG-103–1, LRCG-103–1, LRLG (4), LRLG-104, LRLH, LTGG, LTLH-104, MGLG, MRBG-96, NHBB-97, NQBH, NQBK-98, NQLF-97 98, NQLG-104 (2), NQLH, NQLH-104, NQLJ-98, NQLK, NQMG-96 97 104, NQMG-104 (2), NRBG (2), NRBG-97, NRBH-97 103–1, NRCG, NRLG (4), NRLG-91 96, NRLG-96 104, NRMG, NSGG-91 104, NSGG-96, NTQG, QQBB, QQLH, QQMG, SLBG, SQBC-103–1, SQBG-101, SQLG-91, SQLK, SRCG-101 104, SRLG, SRLH-97 103–1, TTGG-91
2013	Eastern Prairies, cultivated oats	BQBB, BRBG (2), BRLB, BRLB-97, BRLG (2), CRLG, DQLG-91 96, DRBG, DRBJ-91, DRLG (2), DRMJ-91, JQLG-91 103–1, LLLC-104, MHBG, NRBG, NRBG-98, SRLB-91
2013	Eastern Canada, cultivated oats	BDGB, BPGB, BRBB-104, BRBG, BRBG-104, BTBK, BTGB (2), DPGG, DQBG-104, FTGG, HTBG, KKGG, NTGG, NTLG
2014	Eastern Prairies, wild oats	BHLB-91, BRBG-104, BSLG-104, BTGG-103–1, BTQG, CQBG, DQLG-91 (2), DQLG-96, DQLG-104, DRLG-91, DRLG-91 103–1 104, DRLG-104 (2), DRLK, DRLS-103–1, DSLG-91, DTQG-91 104, DTQG-91 103–1 104, DTQJ, GJGG-104, GLRG-104, GQLG-91 (2), HRLG-104, JQLB, JGLG-91 (2), JQLG-91 104, JRLB, JRLG-91, JRLG-91 103–1, JSQG, JSQG-91, JSQG-91 104, JSQG-104, JTGG-103–1, JTQG-91, JTQG-91 104 (2), JTRG-91 98 101 103–1, KQLQ-91 98 103–1 104, KQQC, KSQG-91 104, LBGG, LHBG, LLBB, LQNG, LRLG, LRLG-103–1 104, LSQG-96 97 98, NQLJ-104, NRLG-91, NRLH-97 103–1, NRLJ-97, NRLQ-91 97, 103–1 104, NTQG-91 104, PRNG, QTTG-91 101, SQBB-91 104, SQLQ-104, SRBG-104, SRLG-97, SSQQ-91, SSRG-91 104, STQG-101 103–1
2014	Eastern Prairies, cultivated oats	BLBG, BNRJ-91 104, DLLG-91 97, DNBB, DRQG-91 103–1 104, GTQG-91 98 103–1, JQLG-91 103–1, JRLG-91, JSQL-91 101, JTQG-91 104, JTQG = 103–1 104
2014	Eastern Canada, cultivated oats	BQBG-91, BPBG, BPBK-103–1, BRGG, BTBB, BTBK, BTBN, BTBT, BTGG, BTLQ, DPBB, DPBG-97, DPGF, DTBG, DTGG, DTQG-104 (2), LPHG, LTBG, LTBG-97
2015	Eastern Prairies, wild oats	BQBB, BQBG, BQLB-91, BQLJ-96 104, BRBG, BRBG-98, BRBQ, BRMG, BTGG, BTQG, BTQG-98, CRLG, DQLF, DRBJ-94 98, DRLG, DRLQ-91 97, DRNK-103–1, DTGG-91, DTGG-103–1, DTQK-91 97 103–1, DTRG-91, FGLG-103–1, GLBG, GQBG, GQLG, GQLG-91 (2), GQLG-103–1 104, GQLS-91 97, GQPG-91, GRLG-91, GRLG-91 104, GRLG-96 103–1, GSQG-91 104, GTLQ-91 103–1, GTQG-91, GTQJ-91 96, GTQK-91, GTQQ-91, GTQQ-91 97 104, GTQT-101 103–1, JQLG-91, JQLG-103–1, JQLQ-91 97 103–1, JQLT-91, JQPG-91, JQPG-91 101, JQPJ-91, JRLG, JRLG-91, JRLG-96 103–1 104, JRMT-91, JTNG-91, JRPG-91 104, JRPS-91 97 101 104, JSGG-91, JSQG-91, JSQG-91 103–1, JSQG-91 104, JSQK-91, JSQK-91 98 103–1 104, JSQS-97 103–1 104, JSRG-91 101, JTQB-91, JTQG-91 97 98, JTQG-97 104, JTQT-91 103–1, JTPG-91, JTRG-91, JTRG-91 101, JTRQ-91, JTTG-91 97 101, KQDG-91 98 103–1 104, KTQQ-91, LLCG, LMBG, LQLQ-91 97, LQLK-96 (2), LRLG, LRLG-98, LTQG, NQBG, NQBS, NQCT-103–1, NQFB-104, NQLG, NQLG-91 103–1, NRBG-91 96, NRLG, NRLG-91 98 103–1, NRLG-96, PRPG-91, PRSG-91 104, QQBG, QQLG-91 101, QQLK-91, QRLK-98, QTQG-91, QTQK-91 104, QTRG-91 104, SLBH, SQBK-98, SQLQ-91 103–1 104, SQPG-91 101, SRLG-91, SRLG-91 103–1, SRLG-91 104, SRMH-91, SRPG-91 101, SRPQ-91 101, SSQG-91 101 104, STLG-103–1
2015	Eastern Prairies, cultivated oats	BRBG, BRBJ, BRLG-91 98 103–1, BTFG-91, BTGG-97 103–1, BYQG-91, BTTH, DQLG-91, DQLG-91 104, DRLQ-96, DSQG-91, DSQJ-91 (2), DTQG-91, DTQJ-91, DTQQ-91, GQBJ-91 103–1 104, GQLJ-91 98, GQPG-91 101, GSNJ-91 103–1, GSQG-91 96, GSRG-91 96, GTLQ-91, GTQG-91 09, GTQK-91 103–1, HRLG, HTQG-91 103–1 104, JQLG-91, JQLG-91 103–1, JQLJ-91, JQPG-91 101 103–1, JRBG-91, JRLG-91, JRLG-91 103-1, JRLH-91 103-1 104, JSBK-91 103-1, JSGJ-91 103-1, JSSG-91, JSRQ-91, JSTG-91 101 104, JTQG-91, JTQJ-91, JTQQ-91 103-1 104, JTTG-91, LRBG, LTGT, NRLG, NTQG, NTQJ-104, NTRG-96, NTTG-96, SSGG-91 96, TTKG-91
2015	Eastern Canada, cultivated oats	BTGG, BTGG-104, BTGJ, NTGG

^a The oat differential set consisted of a standard set of 16 primary differentials (Pc38 to Pc68; Chong et al., 2000) and eight supplemental differential lines. All are single gene lines, each with a different Pc gene for resistance to P. coronata var. avenae f. sp. avenae, except for Pc58 and Pc59. Hoffman et al. (2006) reported that Pc58 resistance is conditioned by three loosely linked genes, and Brouwer (1983) reported Pc59 resistance to be conditioned by three unlinked genes. The four letter code designations for the virulence phenotypes are determined using the 16 primary differentials (Chong et al., 2000), followed by a listing of supplemental differentials to which the pathogen isolate is virulent. Numbers in parentheses indicate the number of single pustule isolates of that race which were identified in the year.

Discussion

Crown rust developed well on cultivated oat crops on the EPR in 2010 and 2011, in contrast to 2008 and 2009, when crown rust development was light (Chong et al., 2011). This was followed in 2012 by the lowest crown rust incidence and severity in cultivated oats over the six years of this study. The incidence and severity of crown rust on cultivated oats was higher in 2013 to 2015 than in 2012, but did not reach the levels observed in 2010 and 2011. Incidence and severity of crown rust on wild oats did not follow this trend over the six years of the study. Incidence and severity of crown rust on wild oats was the highest in 2010, and the lowest in 2015. The differences among years in the incidence and severity of crown rust is a reflection of the environmental conditions on the EPR and in the USA for crown rust infection and development. Primary inoculum of urediniospores for infection of oat plants in the spring is generally considered to be introduced into the EPR by winds from the south (Kolmer & Chong, 1993). These urediniospores develop on infected oat crops in the upper Midwest of the USA, and environmental conditions in this region that affect crown rust development will influence the level of primary inoculum that migrates to the EPR. The percentage of fields with crown rust infected plants and severity of crown rust in eastern Canada was monitored in eastern Ontario only, where the disease was generally found at low to moderate levels of infection in most surveyed fields. The severity of crown rust was low in eastern Ontario in 2010 and 2012 as a result of prevailing hot and dry weather conditions during the growing season (Xue & Chen, 2011, 2013).

The frequencies of Pca virulence to several Pc genes changed significantly in the EPR over the six years. Chong et al. (2008, 2011) observed that there can be significant differences between spi from wild oat and those from cultivated oat in some years, but an analysis of these differences over the combined six years of this study did not find differences. Therefore the data for virulence frequencies of spi from the EPR are reported as combined data from wild oats and cultivated oats in Table 1. There were significant increases in frequency of virulence to 15 of the 24 Pc genes in the differential set, and a decrease in the frequency of virulence to Pc40 on the EPR. This increase in frequency of virulence to this number of Pc genes may be a reflection in the increase in mean number of virulence genes per spi from 2010 to 2015.

The most dramatic increase in virulence frequency was to Pc91, which was not detected in 2010 and 2011 in Canada and increased each subsequent year in the EPR until 2015, when virulence was possessed by 66.6% of all spi. Virulence to Pc91 was detected in one spi in 2009 (Chong et al., 2011), which was attributed to the cultivar ‘HiFi’ being grown on about 3% of the acreage in Manitoba. ‘HiFi’ was reported to contain Pc91 (McMullen et al., 2005). The acreage of ‘HiFi’ decreased to less than 1% of the acreage in Manitoba and Saskatchewan by 2011 (Manitoba Agricultural Services Corporation (https://www.masc.mb.ca) and Saskatchewan Crop Insurance Corporation (https://www.saskcropinsurance.com) variety market share reports). The oat cultivar ‘Souris’ was grown on ~11% of the acreage in Manitoba and Saskatchewan in 2010, and increased in popularity to ~20% of the acreage in 2015. ‘Souris’ has a similar pedigree to ‘HiFi’ (https://dl.sciencesocieties.org/publications/cm/articles/5/1/2006-71), and likely also contains Pc91. The increase in acreage of ‘Souris’ likely resulted in strong selection pressure for virulence to Pc91. Virulence to Pc48 increased in 2014 and 2015, which may be attributed to the increase in acreage of the oat cultivar ‘Summit’. ‘Summit’ is reported to possess resistance genes Pc38, Pc39, Pc48 and Pc68 (Mitchell Fetch et al., 2011), and its acreage in Manitoba and Saskatchewan rose from 0 acres in 2010 to 13% of the acreage in 2015. Virulence to Pc48 rose from frequencies of 6–14% from 2007 to 2013 (Chong et al., 2011, Table 1) to just under 50% in 2014 and 2015.Virulence to Pc38, Pc39 and Pc68 did not increase over 2010 to 2015, but the frequency of virulence to these genes was already common, with frequencies of virulence to Pc38 being >96%, Pc39 being > 80%, and Pc68 being between 47 and 69%. This is not surprising since 82% of the oat acreage in Manitoba and Saskatchewan in 2007 was planted to cultivars with these three genes in combination (Chong et al., 2011). In 2010, cultivars that possessed these three resistance genes (‘Furlong’ (Mitchell Fetch et al., 2006), ‘Ronald’ (Mitchell Fetch et al., 2003) and ‘AC Assiniboia’ (Brown et al., 2001)) accounted for 25% of the oat acreage. The acreage of cultivars with the Pc38, Pc39 and Pc68 combination decreased to 8% in 2015, but the frequency of virulence to these genes remained high. It should be noted that during 2010 to 2015, ‘AC Morgan’ increased from 12% of the acreage in 2010 to 26% of the acreage in 2015 in Manitoba and Saskatchewan (Manitoba Agricultural Services Corporation (https://www.masc.mb.ca) and Saskatchewan Crop Insurance Corporation (https://www.saskcropinsurance.com) variety market share reports). ‘AC Morgan’ was the most widely grown oat cultivar in 2011 to 2015, and is susceptible to Pca (Kibite & Menzies, 2001).

The increase in frequency of virulence to Pc45, Pc51, Pc52, Pc58 and Pc59 cannot be attributed to an increased acreage of an oat cultivar possessing these resistance genes in the EPR. These genes have been used in commercial oat cultivars in the USA (Simmons et al., 1978; Chong et al., 2011), and virulence to these genes was commonly identified in pathogen isolates collected in the USA from wild oat and domesticated oat from 2001 to 2005 (Carson, 2008). It is likely that these increases in virulence are the result of Pca isolates migrating to Canada from the oat growing regions of the USA. The increase in frequency of virulence to Pc46, Pc54, Pc56, Pc62, Pc96, Pc101, Pc103-1 and Pc104 occurred mostly in 2014 and 2015. The frequency of virulence to Pc45, Pc46, Pc51, Pc52, Pc54, Pc56, Pc58, Pc59, Pc62, Pc96, Pc101, Pc103-1 and Pc104 will need to be monitored in the future to determine if this is a normal fluctuation or a shift in the pathogen population for increased virulence on these Pc genes. For example, the frequency of virulence to Pc62 in 2015 was similar to what was reported in 2007 (Chong et al., 2011), and the frequency of virulence to Pc96 was higher in 2007 than what is reported in Table 3, so the differences over time may just reflect fluctuations in virulence frequency.

The decrease in frequency of virulence to Pc40 on the EPR was the only significant decrease observed in this study. Chong & Kolmer (1993) observed virulence to Pc40 to fluctuate between 20 and 60% between 1974 and 1990. The frequency of virulence to this gene from 2010 to 2013 was similar to that reported by Chong et al. (2011) for 2007 to 2009, but in 2014, it was lower than the observed frequencies in 2010 to 2012. This may represent a fluctuation in virulence frequency that occurs occasionally.

The frequencies of virulence for the 24 Pc genes in spi from eastern Canada for 2010 to 2015 are presented in Table 4. Statistical analysis was not conducted on these data because of the low number of spi that were assessed each year, especially in 2015. The most common virulences were associated with Pc38, Pc39, Pc 48 and Pc68, and their frequencies of virulence appeared to increase from the previous reports from eastern Canada by Chong et al. (2008). Resistance genes Pc38 and Pc39 were present in many commercial oat cultivars released in Ontario in the 1980s and 1990s, and Pc48 and Pc68 were used in cultivars released in the mid-1990s (Chong & Zegeye, 2004). No virulence was observed on Pc58, Pc94, Pc98 or Pc101 from 2010 to 2015 from eastern Canada. Chong & Zegeye (2004) did not observe virulence to Pc58 and Pc94 from 1999 to 2001, and the differentials for Pc98 and Pc101 were not used in their study.

One hundred and nine (39%) of 276 possible virulence combinations were found to be associated in this study. Ten (4%) of these combinations were negatively associated and 99 (36%) were positively associated. Leonard et al. (2005) observed that 49%, 32% and 36% of 276 possible virulence combinations were significantly associated in isolates obtained from Israel, the Northern Plains of the USA, and Texas, USA, respectively, while Carson (2008) observed 41% and 26% of virulence combinations were associated in isolates from the winter and spring oat regions of the USA, respectively. The oat differential sets used by Leonard et al. (2005) and Carson (2008) differed somewhat from the differential set used in this study, but the percentage of gene combinations observed to be associated were similar to our studies. Kolmer & Chong (1993) studied virulence associations in Canadian populations of Pca on 10 single Pc gene lines from data collected during 1974 to 1990 and observed only 9% of the virulence gene combinations were associated on the Prairies, and only 1% of combinations were associated in eastern Canada. They observed that no association between pairs of virulence genes lasted more than three years. Chong & Zegeye (2004) suggested that virulence to Pc48 and Pc52 is positively associated. Leonard et al. (2005) found these two genes to be strongly associated, and our results agree. Kolmer & Chong (1993) and Leonard et al. (2005) identified five pairs of virulence genes combinations that were positively associated in both studies, namely Pc38 and Pc39, Pc45 and Pc46, Pc45 and Pc54, Pc46 and Pc54, and Pc35 and Pc40. We observed strong associations between Pc38 and Pc39, Pc45 and Pc46, and Pc45 and Pc54, but no association between Pc46 and Pc54. Virulence to Pc35 was not assessed in our study. Our results and those of Leonard et al. (2005) found a strong association between virulence to Pc58 and Pc59. Leonard et al. noted that virulence to Pc62 and to Pc68 were not associated with any other virulence, while Carson (2008) observed virulence to Pc62 to be positively associated to virulence to Pc64 and the virulence to Pc68 to be positively associated with the virulences to Pc38, Pc39, Pc55 and Pc63.We found that virulence to Pc62 was positively associated with virulence to nine other Pc genes (including Pc64) and negatively associated with virulence to Pc68, and virulence to Pc68 was negatively associated with virulence to five other Pc genes and positively associated with three other Pc genes (including Pc38 and Pc39). Virulence to Pc40 and Pc48 was negatively associated in our study, which corresponds to Leonard et al. (2005), who separated these two virulences into different virulence association groups. Carson (2008) did not observe virulence to Pc40 and Pc48 to be associated in either the spring or winter oat production areas of the USA.

The association of virulences to different resistance genes is of interest in practical terms for breeding for resistant host lines. Positive associations between different virulences would suggest that the corresponding Pc resistance genes would not make an effective gene combination in a variety, while negative associations would suggest that the resistance genes could be effective in controlling the pathogen. These associations, or linkage disequilibria, can develop in several ways (Leonard et al., 2005). Founder effects resulting from the establishment of pathogen populations in new areas, or genetic drift as pathogen populations migrate to new areas which may or may not have endemic pathogen populations, may result in linkage disequilibria. Linkage disequilibria can also develop in response to the deployment of resistant host genotypes. Resistant host genotypes can exert a very strong selection pressure for associations between virulence genes, depending on the diversity of the host genotypes and the extent of the area of cultivation of these genotypes. In general, from the studies mentioned above, the associations between the virulences to the different Pc genes are mostly positive in nature, and as Carson (2008) concluded, it would be difficult to identify strong consistent negative associations between virulence genes to exploit.

The race structure of the Pca populations from the EPR and eastern Canada indicated highly genetically variable populations in both regions. The differences in the races between the EPR and eastern Canada support the conclusion of Chong & Kolmer (1993) that the eastern Canada population and the EPR population of Pca are distinct populations. Eighty per cent of the races identified in eastern Canada were not detected in the EPR. The initiation of infection on oat crops in Canada can arise each year from two sources. Uredia can migrate into Canada from infected oat crops in the USA, or infection of oat crops can occur from sexually produced aecia off of the pathogen’s alternate host, Rhamnus cathartica L. (buckthorn) (Harder & Haber, 1992). The eastern Canadian populations of Pca are likely more strongly influenced by the presence of R. cathartica, which has its most common distribution in Canada in the southern Ontario and south-western Quebec regions (Qaderi et al., 2009). The aecia formed on R. cathartica can have greater variability in virulence than the uredia that arise from oats (Simons et al., 1979). The oat crops grown in Manitoba and Saskatchewan are generally considered to be infected by uredia being blown north from oat crops in the USA along the ‘Puccinia Pathway’ (Harder & Haber, 1992). Rhamnus cathartica can be found in Manitoba and Saskatchewan (Qaderi et al., 2009), but is not as common or widespread as in southern Ontario or south-western Quebec.

Previous studies have shown that the majority of races of Pca are represented by one spi (Chong & Zegeye, 2004; Chong et al., 2008, 2011). Chong et al. (2011) found that 73–80% of the races were represented by a single spi during 2007 to 2009. This was also seen in our study, with ~80–86% of the races in 2010 to 2013, 93% of the races in 2014, and 98% of the races in 2015 identified from the EPR being represented by one spi. In eastern Canada, the percentage of races that were represented by one spi ranged from 85 to 95% in 2010 to 2014, and was 100% in 2015 when only 4 spi were assessed. The fewer races represented by more than one spi in the latter years may be a reflection of the increase in the mean number of virulence genes per spi in 2013, 2014 and 2015. Carson (2008) reported that 71–73% of the races detected in the winter oat and spring oat areas of the USA during 2001 to 2005 were represented by a single isolate.

The lack of a dominant race in Pca populations from the EPR and eastern Canada is a phenomenon that has occurred over the last 15 to 20 years. Chong & Zegeye (2004) found that race BQBB was the dominant phenotype across Canada in 1999 to 2001. BQBB was represented by 14%, 8% and 7% of the spi assessed in 1999, 2000 and 2001 respectively on the EPR, and was 36%, 20% and 31% of the spi during these three years in Eastern Canada. Races LQMB and LBMB represented 27% and 25% of assessed spi from the EPR in 2002 (Chong et al., 2008). In this study, dominant races were represented by 5% or less of the spi collected from the EPR, and dominant races from eastern Canada were represented by 2 spi, except for 2011 where race BBBB was represented by 4 spi (17%). Race LRLG was the only race found in all six years of this study. It was detected each year from wild oat and every year except 2010 and 2011 on cultivated oat in the EPR, but not detected in eastern Canada. Seven races were detected in five of the six years of the study from the EPR (BQBB, BRBG, BRLB, DRLG, GQLG, LRBG and NRLG), and 16 races were detected in four of the years. In Eastern Canada, despite fewer spi being assessed, race BTGG was identified in four of the six years and races BPBG and BRBG were identified in three of the years.

The populations of Pca from the EPR and eastern Canada possess a wide diversity in virulence. This study indicates that these populations continue to evolve in their virulence dynamics, creating a challenge to breed oat lines with effective and stable resistance. Resistance gene Pc94 was the most effective resistance gene across Canada and in eastern Canada, virulence to Pc58, Pc94, Pc98 and Pc101 was not detected in 2010 to 2015. The use of these genes, or other effective resistance genes, should be considered for incorporation into new oat lines for commercial cultivation. However, these genes should be pyramided with other effective resistance genes to try to prevent the pathogen from developing races that could overcome these genes.

Acknowledgements

We would like to thank all the cooperators who submitted crown rust infected leaves for analysis.