Temperature adaptation in Australasian populations of Puccinia striiformis f. sp. tritici

Authors


Abstract

Stripe rust, caused by Puccinia striiformis f. sp. tritici, is one of the major fungal pathogens of wheat. A new pathotype was introduced to Australia in 2002 and several derivative pathotypes were detected in subsequent seasons. It has been suggested that the severity of stripe rust outbreaks in Australia since 2002 could be as a result of traits other than virulence in the pathogen population. This study was conducted to investigate the hypothesis that the stripe rust pathogen population dominant in Australia since 2002 was better adapted to warm temperature conditions compared to previous pathogen populations. Sixteen pathotypes were selected to examine the influence of two contrasting temperature regimes during the 24 h incubation (10°C and 15°C) and the subsequent post-inoculation (17°C and 23°C) periods on latent period and infection efficiency on four susceptible wheat cultivars. In addition, the effect of two contrasting incubation temperatures on urediniospore germination was examined. The results indicated that pathotypes of P. striiformis f. sp. tritici detected after 2002 did not show evidence of adaptation to high temperatures, which suggests that other factors contributed to the observed increased aggressiveness.

Introduction

Stripe rust of wheat, caused by Puccinia striiformis f. sp. tritici (Pst), is one of the major diseases of wheat (Triticum aestivum) in a majority of wheat growing regions of the world (Wellings, 2011). The disease affects wheat grain quality and can significantly reduce grain yield, with reports of 100% yield loss in susceptible cultivars if infection occurred early in the growing season (Chen, 2005). In Australia, it has been estimated that if the disease was not controlled, it could cause grain yield losses up to A$83·43 per hectare and A$994 million at the national level (Murray & Brennan, 2009). In addition, the estimated cost of chemical control of the disease during 2003–2005 growing seasons was between A$40 and A$90 million, excluding application expenses (Wellings, 2007).

Stripe rust was first reported in Australia in 1979 (O'Brien et al., 1980) and was most probably introduced by means of spore-contaminated apparel or goods (Wellings et al., 1987). As the pathogen is usually considered a temperate climate organism, it was anticipated not to survive the harsh Australian summer conditions (Brown, 1984). However, since 1979 there have been regular reports of the disease in each season including during droughts, with significant losses experienced under epidemic conditions. In the period since arrival, more than 20 new pathotypes have been detected (Wellings & McIntosh, 1990; Wellings, 2007), with new pathotypes presumed to emerge by single step mutation from pre-existing pathotypes. There is also evidence for several incursions of new pathogenic variants. A new pathotype, designated 134 E16 A+, was detected for the first time in Western Australia in 2002 (Wellings et al., 2003). This region was among the last significant wheat producing regions of the world to succumb to stripe rust. The pathotype had a virulence/avirulence pattern contrasting by at least 11 loci with contemporary eastern Australian pathotypes suggesting a new pathotype incursion. This pathotype moved eastwards in 2003 and rapidly dominated wheat-growing regions in Australia in the following seasons. The adaptive capacity of pathotype 134 E16 A+ can be estimated from the emergence, within 8 years, of five new variant pathotypes with specific virulence associations to resistance genes deployed in commercial wheats and triticales (C. R. Wellings, unpublished data).

According to Wellings et al. (2003), pathotype 134 E16 A+ appeared to be of similar virulence/avirulence to pathotype PST-78, which was reported as a foreign incursion in the USA for the first time in 2000 (Chen et al., 2002). Prior to 2000, stripe rust predominated in the west coast regions of the USA (Washington, Idaho, Oregon and California). However, after 2000 the disease spread to the regions east of the Rocky Mountains into south central states and the central Great Plains, where the climate was considered to be comparatively warm (Chen, 2005). The prevalent pathotypes of stripe rust in these regions since 2000 were observed to be closely related variants of PST-78 (Chen, 2005; Wan & Chen, 2009).

The occurrence of stripe rust epidemics in warmer regions of the USA led Milus et al. (2006) to propose that Pst pathotypes originating in the USA post-2000 were aggressive because of adaptation to higher temperatures. Milus et al. (2006) presented evidence that post-2000 USA pathotypes had a shorter latent period at higher temperature profiles compared to those detected prior to 2000, and concluded that this formed a basis for an aggressive trait in the pathogen population that led to severe stripe rust epidemics in the seasons from 2000 to 2005.

The Australian Pst pathotype 134 E16 A+ and its single-step mutant derivatives are evidently similar in virulence/avirulence traits with the post-2000 USA Pst population and therefore an investigation of aggressiveness in comparable Australian pathotypes was undertaken. The objective of this study was to determine if pathotypes present in Australia since 2002 were adapted to higher temperatures and therefore might possess an advantage in environmental fitness compared to pathotypes existing in Australia prior to 2002. The influence of temperature on pre- and post-2002 Pst pathotypes was investigated by examining their response to contrasting incubation and post-inoculation temperatures in vivo, and urediniospore germination in vitro.

Materials and methods

Pst isolates

Isolates in the Cereal Rust Working Collection of the Plant Breeding Institute, The University of Sydney (Cobbitty, NSW, Australia) were selected to represent the major pathotypes of Pst collected and characterized over the past 30 years and stored in liquid nitrogen. Sixteen representative pathotypes of Pst were selected on the basis of year and location of collection (Table 1). The isolates collected before and after 2002 were referred to as ‘old’ and ‘new’ pathotypes, respectively.

Table 1. Pathotype designation, year of origin and location of Australasian Puccinia striiformis f. sp. tritici isolates selected for this study
PathotypeYear of originCulture numberLocation
Old pathotype group
104 E137 A+1983414Forbes, NSW, Australia
360 E137 A+1985431Hillston, NSW, Australia
104 E153 A–1985434Narrabri, NSW, Australia
108 E205 A+1987446Devenish, VIC, Australia
104 E153 A+1987447Trangie, NSW, Australia
109 E141 A–1987445Christchurch, New Zealand
110 E143 A+1987444Richmond, TAS, Australia
108 E13 A+1988449Yanco, NSW, Australia
111 E143 A+1989457Southland, New Zealand
106 E139 A–27+1994505Lincoln, New Zealand
104 E137 A–17+1999544Cobbitty, NSW, Australia
111 E143 A–27+ 1999541Dunedin, New Zealand
New pathotype group
134 E16 A+2002572Newdegate, WA, Australia
150 E16 A+2005598Wolseley, SA, Australia
134 E16 A+17+2006599Coleambally, NSW, Australia
134 E16 A+J+2007602Bingara, NSW, Australia

Urediniospores of individual pathotypes were increased on seedlings of the susceptible cultivar Morocco for all experiments. For this purpose about 30 seeds of cv. Morocco were grown in 9-cm diameter pots with a soil mix consisting of four parts of composted pine bark and one part of sand and grown at 17–20°C in the greenhouse. The pots were fertilized with a complete fertilizer (Aquasol, 2·5 g L−1 water) before planting. At 5–6 days, seedlings were treated with 50 mL per pot water–maleic hydrazide solution (Slow Grow 270; Kendon Chemical & Minfg. Co. Pty Ltd; at a concentration of 2 mL L−1 water) in order to increase urediniospore yield by restricting growth to the primary seedling leaf. Seedlings were inoculated with urediniospores suspended in light mineral oil (Shellsol, Mobil Oil; 5–10 mg urediniospores per 5 mL oil per 100 pots) using a hydrocarbon propellant pressure pack. The oil was allowed to evaporate from the seedlings for 15 min. The pots with inoculated seedlings were placed in water trays (5 cm deep) in an incubation room and sealed under polyethylene hoods. The incubation temperature was set at 10 ± 2°C for 24 h in the dark. Incubated seedlings were then placed in greenhouse microclimate chambers at 17 ± 2°C under natural light conditions. The pots were again fertilized with nitrogen fertilizer (Nitram, 25 g per 10 L water) 5–7 days after inoculation. To minimize the possibility of pathotype contamination, recently inoculated seedlings were inserted into clear plastic bottles (1 L) with a side opening and covered and secured with cellophane film. At 15–17 days post-inoculation the plastic bottles were gently shaken to dislodge urediniospores into 50 mL glass tubes. Harvested urediniospores were kept in glass tubes in a cabinet (15°C, 55% relative humidity, RH) for 1 day before experimental use. In this manner, fresh urediniospores were produced in order to ensure uniform Pst inoculum across all experiments.

Plant material

Four wheat cultivars known to be highly susceptible in seedling tests to Pst were selected: Morocco (Australian Winter Cereals Collection accession number AUS5105), Avocet S (AUS90661), Jupateco 73 S (AUS18651) and Nyabing (AUS99243). Two or three seeds of each cultivar were planted in 5-cm diameter pots in the soil mix described above. After emergence, plants were thinned to one per pot and grown at 17–20°C until the seedling leaf was fully expanded and the second leaf was partially exposed. The plants were not treated with maleic hydrazide in the temperature treatment experiments.

Experiment 1: Response of Australasian Pst pathotypes to contrasting post-inoculation temperatures

The aim of this experiment was to compare the latent periods and infection efficiency of the old and new pathotypes when exposed to two contrasting post-inoculation temperatures. The experiment consisted of two repetitions with two post-inoculation temperatures (17 ± 2°C and 23 ± 2°C) considered as the main experimental blocks. The sub-blocks consisted of 16 Pst pathotypes and four susceptible wheat cultivars with 20 plants evaluated within each cultivar. The temperature in the greenhouse rooms was set using the central greenhouse climate operation system and calibrated using data loggers (Fourier Systems).

Four susceptible wheat cultivars were grown as described above. Space limitations in the climate-controlled rooms allowed only eight pathotypes to be examined in a single experiment. Plants were inoculated with the stripe rust pathotypes as described with quantities of urediniospores increased to 30–40 mg per 5 mL oil per 100 pots to minimize the chance of plants escaping infection. After incubation at 10 ± 2°C for 24 h in the dark, plants were placed in two different microclimate chambers set at either 17 ± 2°C or 23 ± 2°C.

Two parameters were measured in this experiment: latent period and infection efficiency. The latter was measured as the percentage of plants with sporulating lesions on the last day of observation. Latent period was measured as the number of days from inoculation to the appearance of Pst pustules on the leaf surface of the last seedling to develop pustules (Milus et al., 2006). Observations were made from day 8 post-inoculation until day 15, after which no additional symptom development was observed on individual plants. The experiment was repeated twice.

Experiment 2: Response of Australasian Pst pathotypes to contrasting incubation temperatures

The aim of this experiment was to investigate the effect of two contrasting incubation temperatures on selected Pst pathotypes in relation to latent period and infection efficiency. The design used in this experiment was the same as in experiment 1, with two different incubation temperatures (10 ± 2°C and 15 ± 2°C) considered as the main block. Plants were inoculated as described above and the material was split into two incubation chambers with the different temperature regimes. After 24 h incubation, plants were removed to a single microclimate room at 17 ± 2°C. Latent period and infection efficiency were measured as described in experiment 1. The experiment was repeated twice.

Experiment 3: The effect of temperature on in vitro urediniospore germination among Australasian Pst pathotypes

For assessment of urediniospore germination, 1·5% Nobel agar media was prepared in Petri dishes as described by Milus et al. (2006). A uniform distribution of urediniospores on agar surfaces was achieved by applying an atomized mineral oil suspension of urediniospores as described above for plant inoculation. Petri dishes were covered and wrapped in aluminium foil to allow high humidity for germination under dark conditions; Petri dishes were then placed into two incubation rooms with contrasting temperatures. The temperature regimes during incubation were identical to those of experiment 2 (10 ± 2°C and 15 ± 2°C). The rates of urediniospore germination were assessed after 3, 6, 9, 12 and 24 h by removing and assessing individual Petri dishes. Assessments were conducted by light microscopy at ×500 magnification, using digital images of three random fields of view. Petri dishes were discarded after each sampling time. Digital images of germinated and non-germinated urediniospores were digitally enhanced using Paint.NET software (v. 3.31) in order to provide contrasting colours (red and yellow, respectively) for germination assessment. The spore images were analysed using Assess v. 2.2 (The American Phytopathological Society). The experiment was repeated twice.

Statistical analysis

Data from all three experiments were subjected to analysis of variance (anova) using the mixed procedure of sas v. 8 (SAS Institute). For experiment 1 and 2 the effects of the linear model were built from the fixed effects of temperature, pathotype and cultivar, and the random effects of replication (individual plants) and repetition (over time). For experiment 3 the fixed effects were temperature and pathotype and the random effect was replication. The percentage data for infection efficiency, latent period and urediniospore germination were transformed using arcsine transformation prior to anova. In all three experiments, the models were constructed separately for individual pathotypes and pathotype groups (new versus old).

Pairwise comparisons were used for comparing pathotype groups, individual pathotypes and their interactions with temperature and cultivars using lsmeans and estimate statements (Milus et al., 2006). The P values were adjusted for multiple testing using the Tukey–Kramer test (Dunnett, 1980).

Results

Experiment 1: Response of Australasian Pst pathotypes to contrasting post-inoculation temperature

Infection efficiency

The data in Table 2 indicates that the higher post-inoculation temperature generally resulted in lower infection efficiency for all pathotypes used in this study. Comparisons between the two pathotype groups showed evidence of significant (< 0·0001) effects of temperature, pathotype group and cultivar, with no significant interactions. Examination of infection efficiency at different post-inoculation temperatures (Table 2) revealed that one new pathotype (134 E16 A+ 17+) and three old pathotypes (111 E143 A+, 111 E143 A–27+  and 104 E153 A–) produced no significant differences across all cultivars and post-inoculation temperature regimes. In these instances, the percentages of plants successfully infected with these pathotypes at both post-inoculation temperatures were not significantly different (= 0·05).

Table 2. Infection efficiency of Australasian Puccinia striiformis f. sp. tritici pathotype groups assessed as mean percentage of infected plants in four susceptible host cultivars when exposed to post-inoculation temperatures of 17 and 23°C
 MoroccoAvocet SJupateco 73 SNyabing
17°C23°C17↔23°C17°C23°C17↔23°C17°C23°C17↔23°C17°C23°C17↔23°C
  1. Means within host cultivar and temperature columns followed by the same superscript letter were not significantly different (Tukey's multiple range test).

  2. Significant differences between temperature means for each pathotype and host combination (17↔23°C) are represented by asterisks: * 0·05, ** 0·005, ***< 0·0001.

Old pathotype group
104 E137 A+74·5a53·8abcd * 72·6ab39·2bcd *** 82·4a60·5ab ** 64·2a27bcd ***
360 E137 A+71·5a44·0abcdefg *** 69·3ab56·2abc 75·6a38·9bc *** 66·5a47·2ab 
104 E153 A–51·7b39·8bcdefg 71·6ab62·9a 77·6a65·2a 68·9a51·7ab 
108 E205 A+74·5a38·5bcdefg *** 72·2ab10·4e *** 78·6a15·6d *** 69·5a12·2de ***
104 E153 A+65·1ab34·9defg *** 63·1ab13·5e *** 68·7a16·5d *** 60·0a3·96de ***
109 E141 A-71·4a30·6defg *** 55·3b39·0bcd 74·8a41bc *** 66·2a7·3de ***
110 E143 A+67·3ab49·5abcd 70·5ab36·7bcd *** 76·5a49·7abc *** 62·0a41·4abc *
108 E13 A+78·4a47·5abcdef *** 76·0a52·6abcd *** 86·5a42·4bc *** 73·6a40bc ***
111 E143 A+64·9ab46·9abcdef 62·9ab50·6abcd 69·5a53·7abc 59·8a50·4ab 
106 E139 A–27+74·8a58·9a 72·4ab48·6bcd *** 78·4a65·4a 69·7a58·5ab 
104 E137 A–17+71·8a26·8defg *** 69·7ab16·1e *** 76·0a11·2d *** 66·7a5·88de ***
111 E143 A–27+ 72·6a56·5abc 70·7ab52·0abcd 76·5a63·1a 67·7a59·7a 
New pathotype group
134 E16 A+75·3a58·0ab 73·2ab62·2ab 79·3a64·8a 70·2a41·3abc ***
150 E16 A+76·5a47·3abcdef *** 68·9ab63·1a 85·7a65·3a * 71·5a30bcd ***
134 E16 A+17+71·1a54·7abc 68·3ab58·5ab 74·7a61·4ab 65·4a52·1ab 
134 E16 A+J+74·9a30·4defg *** 73·3ab15·3e *** 79·2a49·6abc *** 70·2a27·4bcd ***

One new pathotype (134 E16 A+J+) and six old pathotypes (108 E205 A+, 104 E153 A+, 104 E137 A–17+, 104 E137 A+, 106 E139 A–27+  and 108 E13 A+) consistently had lower percentages of infected plants at the high post-inoculation temperature of 23°C than at 17°C across all cultivars (Table 2). The response of two new pathotypes (150 E16 A+ and 134 E16 A+) and three old pathotypes (109 E141 A−, 360 E137 A+, and 110 E143 A+) to the post-inoculation temperature treatments showed evidence of variation in infection efficiency between susceptible wheat cultivars (Table 2).

Analysis of infection efficiency for individual pathotypes within each cultivar and post-inoculation temperature regime (Table 2) revealed that pathotypes were not significantly different at 17°C for cvs Jupateco 73 S and Nyabing and varied only slightly in cvs Morocco and Avocet S. However, greater variation between pathotypes in infection efficiency was observed at 23°C for all cultivars. The rankings of two new pathotypes (134 E16 A+17+  and 134 E16 A+) and two old pathotypes (111 E143 A+ and 111 E143 A−27+) did not change across all cultivars and temperature regimes.

Latent period

The higher temperature generally extended the latent period for all pathotypes and cultivars (Table 3). Analysis of variance for the effect of temperature on latent period duration of new and old pathotype groups indicated that all main effects were significant ( 0·05) although there were no significant (= 0·05) effect interactions (data not shown).

Table 3. Latent period of Australasian Puccinia striiformis f. sp. tritici pathotype groups assessed as mean number of days in four susceptible host cultivars when exposed to post-inoculation temperatures of 17 and 23°C
 MoroccoAvocet SJupateco 73 SNyabing
17°C23°C17↔23°C17°C23°C17↔23°C17°C23°C17↔23°C17°C23°C17↔23°C
  1. Means within host cultivar and temperature columns followed by the same superscript letter were not significantly different ( 0·05).

  2. Significant differences between temperature means for each pathotype and host combination (column 17↔23°C) are represented by asterisks: * 0·05, ** 0·005, ***< 0·0001.

Old pathotype group
104 E137 A+9·5a12·5ab ** 10ab14b *** 9a11·5ab 11ab14·5bc **
360 E137 A+10ab12·5ab 10·5ab12·5ab 10·5a11·5ab 11·5ab12·5abc 
104 E153 A−12·5b13·5ab 10ab11a 9a11a 10·5ab12ab 
108 E205 A+9·5a12·5ab ** 10·5ab13·5ab * 9·5a14b *** 11·5ab14·5bc *
104 E153 A+12b13ab 10·5ab14b ** 9a13·5ab *** 12ab13abc 
109 E141 A–11·5ab14b 10·5ab14b ** 9a12ab * 11ab15c ***
110 E143 A+11ab13·5ab 10ab12ab 9·5a11·5ab 11·5ab12·5abc 
108 E13 A+9a13·5ab *** 9a12ab * 9a13ab *** 10a13·5abc **
111 E143 A+10ab12ab 12b12·5ab 10·5a12ab 13b12ab 
106 E139 A–27+10ab11a 10·5ab12·5ab 9a11a 10·5ab11·5a 
104 E137 A–17+11ab12ab 10ab14b *** 10a14b *** 11ab13·5abc 
111 E143 A–27+ 10ab11·5ab 10·5ab12·5ab 9·5a11a 10·5ab11·5a 
New pathotype group
134 E16 A+10ab12·5ab 10·5ab12ab 9a12ab * 11ab12ab 
150 E16 A+9·5a13·5ab ** 11ab11a 9a11a 10a14·5bc ***
134 E16 A+17+11ab12ab 10·5ab12ab 10a11·5ab 11ab12·5abc 
134 E16 A+J+9a14ab *** 10ab14b *** 9·5a12·5ab * 10·5ab13abc 

Examination of the data (Table 3) revealed that the latent periods of pathotypes 134 E16 A+17+, 111 E143 A+, 360 E137 A+, 111 E143 A–27+, 106 E139 A–27+, 104 E153 A− and 110 E143 A+ were not significantly (= 0·05) influenced by temperature within a susceptible cultivar. Pathotypes 108 E205 A+ and 108 E13 A+ had significantly ( 0·05) longer latent periods at 23°C than at 17°C post-inoculation when tested on all wheat cultivars. However the response of new pathotypes 134 E16 A+, 134 E16 A+J+ and 150 E16 A+ and old pathotypes 104 E153 A+, 104 E137 A–17+, 109 E141 A– and 104 E137 A+ to the contrasting post-inoculation temperature treatments varied according to cultivar (Table 3).

Comparisons of latent period for each pathotype, within each cultivar and temperature regime, showed greater variation on Nyabing at the higher temperature.

Experiment 2: Response of Australasian Pst pathotypes to incubation temperatures

Infection efficiency

Analysis of variance for the final percentage of infected plants at incubation temperature regimes of 10°C and 15°C for individual pathotypes revealed that all factors (temperature, pathotype, cultivar) were significant (< 0·0001), although there were no significant (= 0·05) interaction effects (data not shown). The anova for infection efficiency of new and old pathotype groups treated with two contrasting incubation temperatures (10°C and 15°C) indicated that responses differed significantly between cultivars (= 0·005) while none of the interaction effects were significant (= 0·05; data not shown).

The infection efficiency for pathotypes across all cultivars and both incubation temperature regimes (Table 4) indicated that the percentage of infected plants was not significantly different between the majority of pathotypes. The new pathotype 150 E16 A+ had the highest percentage of infected plants (71·6%), but was not significantly different (< 0·05) from five of the old pathotypes (Table 4). One new and one old pathotype (134 E16 A+17+  and 111 E143 A+, respectively) had the lowest percentages of infected plants (43·1% and 40·3%, respectively) but were not significantly different from three other old and one new pathotypes (Table 4).

Table 4. Mean infection efficiency for Australasian Puccinia striiformis f. sp. tritici pathotypes across four wheat cultivars and two incubation temperature regimes (10 and 15°C)
PathotypeInfection efficiency (% infected plants)
  1. Means between pathotypes followed by the same letter were not significantly different ( 0·05).

Old pathotype group
104 E137 A+52·5bcd
360 E137 A+52·0bcd
104 E153 A–61·0abc
108 E205 A+61·5abc
104 E153 A+59·1bc
109 E141 A–66·3abc
110 E143 A+68·7ab
108 E13 A+67·3ab
111 E143 A+40·3de
106 E139 A–27+60·1bc
104 E137A–17+55·2bc
111 E143 A–27+ 48·1bcde
New pathotype group
134 E16 A+49·8bcde
150 E16 A+71·6a
134 E16 A+17+43·1de
134 E16 A+J+60·5bc

Latent period

Latent period varied from 10·3 to 12·5 days among pathotypes across all cultivars and both incubating temperatures (data not shown). anova for latent period duration among individual pathotypes (data not shown) treated at two incubation temperatures indicated that there was a significant effect of pathotype and cultivar ( 0·05) and pathotype × cultivar interaction (= 0·005). The significant effect of pathotype suggested that the latent periods of some pathotypes were different. However there was no evidence for the effect of temperature, temperature × pathotype or temperature × pathotype × cultivar interactions on latent period.

Experiment 3: The effect of temperature on in vitro urediniospore germination among Australasian Pst pathotypes

anova for the percentage of germinated urediniospores for individual pathotypes at two different incubation temperatures (data not shown) indicated that main effects (temperature and pathotype) and interactions were significant ( 0·05). Urediniospore germination of individual pathotypes within each temperature regime (Table 5) revealed that there was minor variation at 10°C. Although greater variation was observed at 15°C, the germination percentages were similar across all pathotypes. A comparison of percentage of germinated urediniospores at 24 h incubation among individual pathotypes at the two germination temperature regimes (Table 5) revealed that six old pathotypes (111 E143 A+, 360 E137 A+, 109 E141 A–, 104 E137 A+, 104 E153 A– and 110 E143 A+) had significantly ( 0·05) higher percentages of germinated urediniospores at 10°C compared to 15°C. None of the new pathotypes and the remaining six old pathotypes were significantly different ( 0·05) for germination between the two temperature regimes.

Table 5. Percentage of in vitro germinated urediniospores at 24 h incubation for individual Australasian Puccinia striiformis f. sp. tritici pathotypes at two temperatures (10 and 15°C)
PathotypeGermination (%)
10°C15°C10↔15°C
  1. Means for pathotypes within each temperature treatment column followed by the same letter were not significantly different ( 0·05).

  2. Significant differences between temperature means for each pathotype (column 10↔15°C) are represented by asterisks: * 0·05, **  0·005, *** < 0·0001.

Old pathotype group
104 E137 A+80·3ab67·3cdef ***
360 E137 A+81·7ab71·0bcde ***
104 E153 A–80·5ab73·0bcd **
108 E205 A+80·7ab77·4abcd 
104 E153 A+79·7ab81·3a 
109 E141 A–80·9ab64·2ef ***
110 E143 A+82·8a72·7bcde ***
108 E13 A+82·8a77·8ab 
111 E143 A+75·7b63ef ***
106 E139 A–27+80·8ab76·9abcd 
104 E137 A–17+83·8a81·9a 
111 E143 A–27+82·2ab79·3ab 
New pathotype group
134 E16 A+80·0ab74·2bcd 
150 E16 A+82·4a81·3a 
134 E16 A+17+81·4ab77·3abcd 
134 E16 A+J+83·6a80·3ab 

The mean percentage of germinated urediniospores of individual pathotypes over time of observation within each temperature regime is presented in Table 6. At 10°C, maximum urediniospore germination of three old pathotypes (111 E143 A+, 108 E205 A+, 108 E13 A+) was observed at 9 h, whereas the remaining pathotypes (both new and old) developed maximum urediniospore germination at 6 h.

Table 6. The influence of incubation time period and temperature on urediniospore germination in a collection of Australasian Puccinia striiformis f. sp. tritici pathotypes
PathotypeGermination (%) at 10°CGermination (%) at 15°C
3 h6 h9 h12 h24 h3 h6 h9 h12 h24 h
  1. Means for pathotypes followed by the same letter between time periods for each temperature treatment were not significantly different (≤ 0·05).

  2. Shaded cells indicate the earliest time period to reach maximum germination.

Old pathotype group
104 E137 A+49·4a73·3b77·4b78·4b80·3b43·7a58·0b63·2bc64·2bc67·3c
360 E137 A+51·4a77·6b81·7b82·2b81·7b48·5a61·4b68·2bc70·1c71·0c
104 E153 A–54·5a74·9b80·5b79·7b80·5b47·2a66·3b71·1b73·1b73·0b
108 E205 A+53·1a68·0b76·0c78·4c80·7c59·9a65·2a76·4b77·8b77·4b
104 E153 A+59·8a76·9b78·6b79·2b79·7b72·2a75·2ab80·1b81·5b81·3b
109 E141 A–36·9a76·0b76·1b80·9b80·9b25·7a44·4b51·9c55·3c64·2d
110 E143 A+69·2a80·4b79·0b82·1b82·8b60·0a71·6b71·7b72·8b72·7b
108 E13 A+56·5a75·2b77·4bc77·9bc82·8c71·3a73·5a74·0a77·7a77·8a
111 E143 A+34·4a64·5b72·6c74·0c75·7c29·3a35·7a57·5b58·2b63·0b
106 E139 A–27+58·7a77·1b79·2b79·2b80·8b55·6a70·5b71·5b75·1b76·9b
104 E137 A–17+76·1a78·9ab81·1ab83·5b83·8b78·0a78·1a78·6a81·9a81·9a
111 E143 A–27+56·1a76·3b75·1b79·8b82·2b45·6a67·1b76·2c78·3c79·3c
New pathotype group
134 E16 A+57·0a74·9b78·4b80·7b80·0b48·1a66·3b71·9b74·1b74·2b
150 E16 A+66·1a79·9b82·0b81·9b82·4b46·6a78·7b81·4b82·2b81·3b
134 E16 A+17+57·6a76·5b77·3b80·7b81·4b68·5a75·1ab76·2b76·9b77·3b
134 E16 A+J+54·8a77·7b79·6b81·8b83·6b69·0a70·7ab77·9bc79·0c80·3c

Maximum urediniospore germination at 15°C for one new (134 E16 A+J+) and five old pathotypes (111 E143 A+, 360 E137 A+, 108 E205 A+, 104 E137 A+ and 111 E143 A–27+) was observed at 9 h after incubation (Table 6). Maximum germination for one old pathotype (109 E141 A–) was observed at 24 h, in contrast with two old pathotypes (104 E137 A–17+ and 108 E13 A+) at just 3 h at 15°C. The remaining pathotypes achieved maximum germination after 6 h at 15°C.

The urediniospore germination response of three new pathotypes (134 E16 A+17+, 134 E16 A+ and 150 E16 A+) and four old pathotypes (104 E153 A+, 106 E139 A–27+, 104 E153 A– and 110 E143 A+) did not change between the two temperature regimes. Old pathotypes 108 E13 A+ and 104 E137 A–17+  developed maximum percentage urediniospore germination after 9 h at 10°C, whereas this was shortened to 3 h at 15°C (Table 6).

Discussion

Various issues associated with the concept of pathogen aggressiveness were reviewed in detail by Shaner et al. (1992). Definitions of aggressiveness in fungal pathogens vary depending on the pathogen and the preference of individual authors. For example, Green & Campbell (1979) referred to aggressiveness as the ability of the pathogen (P. graminis f. sp. tritici) to survive in nature, whereas Roelfs et al. (1989) referred to aggressiveness of the pathogen (P. graminis f. sp. avenae) as the ability to initiate epidemics regardless of virulence. Milus et al. (2006) concluded that a number of Pst pathotypes detected in the USA after 2000 were more aggressive because of better adaptation to high temperatures. The authors also concluded that the shorter latent period of new pathogen isolates at high temperatures was an important component of aggressiveness.

In contrast, various authors attributed the ability of a pathogen to sustain and persist in the environment as parasitic fitness (Fleming, 1981; Barrett, 1983; Tooley & Fry, 1985). Fitness is referred to as the ability of isolates of a pathogen to make a major contribution to the population structure and in doing so replace the previously existing pathogen population (Shaner et al., 1992). Similar to the situation in the USA, where pathotypes originating after 2000 replaced the previous Pst populations, pathotype 134 E16 A+ detected in Western Australia in 2002 replaced the Pst population that had developed in eastern Australia since 1979 (Wellings, 2007). Indeed molecular evidence based on AFLP analysis suggests a close relationship between the recent Pst populations in both continents (Hovmøller et al., 2008). On this basis it could be suggested that the recent Pst populations emerging in the USA and Australia carry increased parasitic fitness, which contributed to a major shift in the genetic structure of Pst in both countries. Other components of aggressiveness that have been investigated in Pst have included urediniospore production (Johnson & Taylor, 1972, 1976; Milus et al., 2009), lesion length (Priestley & Doling, 1974) and urediniospore germination (Volin & Sharp, 1973).

The absence of significant temperature interactions for new and old Pst pathotypes in Australian studies reported here indicated little evidence to support temperature adaptation as a basis for the current dominance of the new pathotype group. The effect of post-inoculation temperature on infection efficiency and latent period demonstrated that there was variation between individual pathotypes, but this was clearly independent of the categorization into new and old pathotype groups. One new and three old pathotypes demonstrated a level of adaptation or fitness elasticity because infection efficiency did not vary between the optimal (17°C) and the elevated (23°C) post-inoculation temperatures or between the susceptible wheat cultivars. Infection efficiency was low at the elevated post-inoculation temperature in one new and six old pathotypes across all cultivars, indicating little evidence of adaptation to elevated post-inoculation temperature in Pst pathotypes grouped as old or new. The remaining two new and three old pathotypes showed a lower degree of post-inoculation temperature adaptation as the effects of those temperatures varied depending on the host.

The latent period response of some pathotypes to post-inoculation temperature treatments varied between susceptible wheat cultivars. These results suggested that there were minor cultivar × temperature interactions that influenced the latent period duration. This would be a useful avenue for further investigation to determine if there is a heritable genetic component influencing latent period in susceptible genotypes.

Infection efficiency and latent period duration were generally unaffected in experiments comparing optimum (10°C) and elevated (15°C) incubation temperatures of Pst in Australia. de Vallavieille-Pope et al. (1995) reported 8–12°C as the optimum temperature range for spore germination of Pst and demonstrated that the highest germination temperature was 20°C. In the same study, these authors concluded that infection efficiency was optimal from 5–12°C and reached maximum at 15°C. However, infection efficiency calculated by de Vallavieille-Pope et al. (1995) was different from the current study, and hence results for this parameter are difficult to compare. McCracken & Burleigh (1962) studied Pst urediniospore germination in vitro and concluded that although urediniospores would germinate between 2 and 20°C, the optimum temperature for urediniospore germination was considered 2–5°C. Similarly, Sharp (1965) concluded that 7°C was the optimum incubation temperature for urediniospore germination in vivo. The present work suggests that temperatures from 10 to 15°C will continue to provide successful incubation conditions for Pst pathotypes in Australia. The data provided evidence for variation within and between new and old pathotype groups in response to contrasting germination temperatures (10 and 15°C). It was concluded that neither pathotype group could be differentiated based on adaptation to elevated incubation temperatures.

Milus et al. (2006) evaluated the adaptation of new and old isolates of Pst in USA to different temperatures. They compared old isolates, collected prior to 2000, with new post-2000 isolates for latent period and urediniospore germination at 12 and 18°C. The temperatures were selected to represent optimal and near maximum for development of stripe rust based on the work of de Vallavieille-Pope et al. (1995), although these temperature contrasts are not evident in the latter work. The suggestion that 18°C is an elevated temperature for assessing latent period (Milus et al., 2006) does not appear reasonable. The temperature contrasts selected in the current study were 10 and 15°C for incubation (i.e. germination and infection), and 17 and 23°C for the post-inoculation period from infection until urediniospore production. These temperature contrasts were considered to be more representative of optimal and elevated temperatures under epidemic conditions in Australia. Although the aims of the current study and those of Milus et al. (2006) were similar, the contrasting temperatures used in the current study were different, and therefore it is difficult to directly compare the outcomes of both studies.

The present work has provided no conclusive evidence for temperature adaptation to explain the dominance of a set of new pathotypes in Australian Pst populations since 2002. However, the overwhelming dominance of this new pathotype lineage clearly indicates an adaptive capacity that must find its basis in other features of fitness and aggressiveness. Nazari (2006) investigated adult plant resistance (APR) genes in field populations of BC1F2 families from crosses between several Australian wheat cultivars in 2002 using pathotype 110 E143 A+ and again in BC1F3 using pathotype 134 E16 A+ in the following year. No seedling resistance gene was detected in the segregating populations. Single gene lines segregating for APR in 2002 were completely susceptible in 2003, clearly implicating the role of virulence for APR in the new pathotype. This data strongly indicated a gene-for-gene interaction between the new Pst pathotypes and at least one APR gene, although this hypothesis needs to be tested further experimentally.

In a similar but more detailed study, Bariana et al. (2010) suggested the presence of at least one APR gene in each of the two Australian wheat cultivars Kukri and Janz. The APR gene present in cv. Janz was ineffective against the post-2002 pathotype 134 E16 A+ but effective against pre-2002 pathotype 110 E143 A+, whilst the APR gene in cv. Kukri was ineffective against 110 E143 A+ but effective against 134 E16 A+. Both cultivars were equally susceptible to both pathotypes in the seedling stage, indicating that seedling genes were not involved in the interactions.

There is also recent evidence in Australia that an APR gene in triticale cv. Tobruk was overcome by a pathotype carrying a corresponding virulence allele (Wellings, 2010). Preliminary results (C. R. Wellings, unpublished data) suggest that this pathotype was related to the post-2002 pathotype 134 E16 A+J+, but also carries additional virulence for the APR resistance in cv. Tobruk. These reports add weight to the suggestion that virulence for APR was an important basis for the evident aggressiveness of the new pathotype lineage of Pst in Australia. Given the evidence for a close relationship between Australian and USA Pst isolates and the unequivocally dominant nature of these pathotype groups in both regions, it will be of interest to establish if there are APR genes in common among commercial wheats in both countries.

Acknowledgements

The first author acknowledges the financial support provided through The University of Sydney, Endeavour International Postgraduate Research Scholarship. The continuing support of the Australian Grains Research and Development Corporation through the Australian Cereal Rust Control Program is gratefully acknowledged.

Ancillary