Postulation of leaf rust resistance genes in Iranian wheat cultivars and breeding lines

Abstract

Knowledge of the number and identity of the leaf rust resistance genes in wheat breeding material is essential for maximizing resistance in future-bred cultivars. The objective of this study was to test for seedling resistance genes to Puccinia triticina potentially present in 36 Iranian wheat cultivars and breeding lines, using 13 prevalent isolates of P. triticina. Eight known genes and some unidentified genes were postulated for resistance to leaf rust in 21 Iranian wheat genotypes. The most frequently occurring genes in Iranian wheat genotypes were Lr1 (71%), followed by Lr13 (62%), Lr10 (43%), Lr26 (38%), Lr23 (19%) and Lr17 (10%). Seven genotypes lacked any detectable seedling resistance gene and two cultivars along with three breeding lines were resistant to all isolates used in the study. It is concluded that there was little variation in the Lr genes carried by wheat cultivars commercially grown in Iran. Therefore, strategies for deploying resistance genes to prolong effective disease resistance are suggested.

Résumé

La connaissance du nombre et de l'identité des gènes de résistance à la rouille des feuilles, contenus dans le matériel utilisé pour l'amélioration du blé, est essentielle au développement maximum de la résistance chez les futurs cultivars. L'objectif de cette étude était d'analyser, au stade de semis, les gènes de résistance à Puccinia triticina qui pourraient être présents dans 36 cultivars de blé iranien et de souches généalogiques, et ce, à l'aide de 13 isolats dominants de P. triticina. Huit gènes connus et certains non identifiés ont été considérés en fonction d'une présumée résistance à la rouille des feuilles chez 21 génotypes de blé iranien. Les gènes apparaissant le plus fréquemment étaient Lr1 (71 %), Lr13 (62 %), Lr10 (43 %), Lr26 (38 %), Lr23 (19 %) et Lr17 (10 %). Sept génotypes étaient dépourvus de tout gène de résistance détectable au stade de semis, et deux cultivars ainsi que trois souches généalogiques se sont avérés résistants à tous les isolats au cours de l'étude. Nous en avons conclu qu'il y avait peu de variation chez les gènes Lr portés par les cultivars de blé utilisés commercialement en Iran. Par conséquent, on suggère des stratégies visant à déployer ces gènes pour prolonger la résistance effective à la maladie.

Keywords:

• disease resistance
• infection type
• Lr gene
• Puccinia triticina
• Triticum aestivum L

Mots clés:

• gène Lr
• Puccinia triticina
• Triticum aestivum L
• résistance à la maladie
• type d'infection

Introduction

Wheat leaf rust, caused by Puccinia triticina Eriks, is one of the most common diseases of wheat worldwide. Although leaf rust tends to be less damaging than stem rust or stripe (yellow) rust, it probably results in higher total annual losses worldwide because of its more frequent and widespread occurrence (Huerta-Espino et al., 2011). After stripe rust (P. striiformis Westend f. sp. tritici Eriks.), leaf rust is the most common and important wheat disease in Iran. It is estimated that 1.1 million hectares (20%) of Iran's wheat-growing lands are prone to leaf rust infection. Wheat is the most important crop in Iran, grown on 6.7 million hectares (more than 50% of total agricultural lands) (FAOSTAT, 2011). The use of resistant cultivars is the most efficient, economical and environmentally safe method to control leaf rust disease.

Although there are about 60 known genes involved in the resistance of wheat to P. triticina, a majority of these are not effective against current races of P. triticina (McIntosh et al., 1995). P. triticina has diverse virulence and is able to overcome resistance genes. The emergence of virulent pathotypes can restrict the durability and use of rust resistance genes. Accordingly, there is an ongoing need to identify, characterize and deploy new sources of resistance (Kolmer, 1996).

For assessing the vulnerability of the crop to leaf rust, knowledge of the major resistance genes present in the predominant wheat cultivars is a prerequisite. In turn, when this information is combined with data on virulence features of the P. triticina population in Iran, it is possible to make informed decisions for improving the leaf rust resistance in Iranian wheat cultivars (McIntosh et al., 1995). The presence of race-specific resistance gene(s) in a cultivar is postulated based on the gene-for-gene relationship (Flor, 1956), provided that an array of pathogen cultures with diverse combinations of avirulence and virulence genes is used. In wheat, the presence of a specific gene for Puccinia spp. resistance can be ascertained by an interaction with the Puccinia spp. culture that lacks the corresponding gene for virulence. Leaf rust isolates that produce distinct low infection type (IT) on specific Lr genes, will also produce low ITs on those cultivars that have the same resistance genes (Kolmer, 2003). A great deal of information on postulated leaf rust resistance genes has been collected from countries (including Australia, US, Canada, China, India, Pakistan and South Africa) where wheat is a major crop (Singh et al., 2001; Kolmer, 2003; Wamishe & Milus, 2003; Oelke & Kolmer, 2004; Pathan & Park, 2006).

Little information is available on Lr genes present in Iranian wheat cultivars. The objective of this study was to postulate genes for seedling resistance to P. triticina in Iranian wheat cultivars and breeding lines, mostly derived from CIMMYT germplasm. Knowledge of the number and identity of the leaf rust resistance genes in these cultivars and lines will be useful in understanding their field reaction to changing P. triticina populations and in using these breeding materials as parents for improving future wheat cultivars.

Materials and methods

Plant materials

Thirty-six Iranian wheat cultivars and breeding lines (Table 1) were assessed for their response to P. triticina isolates. The land race cultivar, ‘Bolani’, was used as a susceptible control. Sets of differential lines containing known resistance genes used in Australia to characterize pathotypes of P. triticina (NILs; developed by the late Dr P.L. Dyck) were also included. The set consisted of the 30 ‘Thatcher’ near isogenic lines (NILs ) carrying the resistance genes: Lr1, Lr2a, Lr2b, Lr2c, Lr3ka, Lr3, Lr3bg, Lr9, Lr10, Lr11, Lr13, Lr14a, Lr14b, Lr15, Lr16, Lr17, Lr18, Lr19, Lr20, Lr21, Lr23, Lr24, Lr25, Lr26, Lr28, Lr29, Lr30, Lr32, Lr33 and Lr36. The NILs were kindly provided by Dr Ravi Singh, CIMMYT, Mexico.

Table 1.  Iranian wheat cultivars/lines tested for resistance to wheat leaf rust

Wheat cultivar/ lines	Pedigree
Alemoot	CM85836-4Y-OM-OY-8M-OY-OPZ,Attila
Alvand	1-27-6275/CF1770
Atrak	Kauz”s”
Chamran	Attila,CM85836-50Y-OM-OY-3M-OY
Darab2	Maya“s”/Nac
Dez	Kauz*2/Opata//Kauz
Fong	Vee 9/3 / Coox / Doves
Ghods	Rsh/5/Wt/4/Nor10/K54*2//Fn/3/Ptr/6/Omid//Kal/Bb
Hyrmand	Byt/4/Jar//Cfn/Sr70/3/Jup”s”
Kavir	Stm/3/Kal//V534/Jit716
Mahdavi	Ti/Peh5/Mt48/3/Wt*// Nar59/Tota63/4/Mus
Mv17	Slavia/Mv Tf/ZG443(Hungre Culrivar)
Nicknejad	F13471/Crow”s”
Sablan	(908*FnA12)*1-32-4382
SD-2	Not available
Shahriar	Kvz/Ti71/3/Maya s”//Bb/Inia/4/Karaj2/5/Anza/3/Pi/Nar//Hys
Shiraz	GV/D630//ALD “S”/3/AZD
Shirodi	Attila, CM85836-4Y-OM-OY-8M-OY-OPZ
Tajan	Bow”s”/Nkt”s”(CM67428-GM-LR-5M-3R-LB-Y
Tos	Spn/Mcd//Cama/3/Nzr
Zarin	PK12841
Vee/Nac	Vee/Nac
C-79-16	KinacI 97 951327 Sum 12289-78-OM
C-78-7	BOF13367//P102/3/Strampelli
C-78-18	VATA
C-80-4	Sh#4494/Crow“s”//Kvz/6/1-68-20/5/Gds/4/…
C-80-20	Vee“s”//Nac//1-66-22
C-80-11	Gds/4/Anza/3/Pi/Nar//Hys/5/1-66-75
C-80-10	Omid//H7/4p839/Omid/Tdo/5/Icw HA 81-1473
C-80-16	Jup/4/CIIF/3/II 14.53/Odin//CI 13431/…
M-79-6	Bow“s”/Vee“s”//1-60-3
M-79-7	Bloyka ICW84-0008-013AP-300L-3AP-3000L-0AP
N-75-16	Shanghi7//Hahn“s”/2Prl“s”/CM95119-3y-OM-oy
S-75-11	Twvaco/Chill
S-78-11	Bow“s”/CM 34798/3/Snb/Pewee“s”//Snb/Mus
S-79-10	Pbw343
Bolani	Land race

P. triticina isolates

In spring of 2005–2006, isolates of P. triticina were collected from trap nurseries located at Ahvaz, Boroojerd, Doruod, Gharakhil, Ghazvin, Gonbad, Iranshahr, Karaj, Kelardasht, Maneh, Mashhad, Shaavar and Saffiabad where leaf rust occurs annually. To obtain sufficient spores for inoculation from each sample, urediospores from a single pustule on an infected leaf were propagated on the susceptible cultivar ‘Bolani’. When sufficient inoculum was obtained it was applied to the differential set.

Inoculation and disease assessment

Each isolate was tested in two replicates; first, the seedlings were inoculated by suspension of urediospores in 1000 mL water containing two to three droplets of Tween 80; and second, the mixture of talcum powder and P. triticina urediospores (3:1) was used for inoculation. Ten seeds were sown per pot and 12-day-old seedlings (two-leaf stage) were inoculated by prevalent Iranian leaf rust isolates collected from different regions in 2005–2006. After inoculation, seedlings were moved to a dark, cool room (16 ± 2 °C and 100% RH for 20–24 h). They were then transported to a greenhouse maintained at 22 ± 2 °C. The leaf rust reactions on primary leaves were recorded 12–14 days after inoculation using the IT scale outlined by McIntosh et al. (1995), where 0 = no visible uredia, ‘;’ = hypersensitive flecks, 1 = small uredinia with necrosis, 2 = small to medium sized uredinia with green islands surrounded by necrosis or chlorosis, 3 = medium sized uredinia with or without chlorosis, 4 = large uredia without chlorosis, and X = heterogenous. Designations of ‘+’ and ‘−’ were used with the 0 to 4 scale to indicate larger and smaller uredinia than normal, respectively. ITs 3–4 were considered susceptible (high ITs), and ITs of 0 to 3⁻ and X were considered as resistant responses (low ITs).

Gene postulation

The presence of resistance genes in each cultivar/line was postulated based on the differential response to different isolates. Isolates which produced distinct low ITs on specific Lr genes will also produce low ITs on those cultivars that have the same resistance genes. In order to postulate Lr genes in wheat cultivars used in the present study, the ITs on genotypes at the seedling stage were compared to those produced on specific Lr genes in Thatcher NILs.

To characterize the leaf rust resistance genes in the tested cultivars, a SAS Macro program developed by Wamishe et al. (2004) was used. The program's outputs are in two steps. To exclude Lr genes from a line and obtain a relatively short list of Lr genes that could be present in a line, step 1 of the program utilized data from races virulent on a line as determined from IT data in the line data set. Lines susceptible to a given race cannot have any of the Lr genes for which the race is avirulent as determined from the isoline data set. Step 2 of the program considered only Lr genes not excluded in step 1 and utilized data from races that were avirulent on the line. To facilitate postulating the presence of Lr genes in each line by matching infection types, output from step 2 listed the low ITs produced on the line and the NIL of Lr genes being considered. To postulate Lr genes by matching ITs, the low ITs listed in the output of step 2 were compared visually. Of the Lr genes listed in step 2, a gene was considered present if the low IT produced on the line by one or more races matched the low IT on the corresponding NIL (Wamishe et al., 2004).

Results

In 90% of the cases, NILs had more or less similar ITs using both inoculation methods. However, the final decision for the reaction type was based on the data resulting from inoculation with the suspension of spores in water, due to a cleaner background of leaves and consequently easier scoring for resistance to leaf rust in this method. The isolates were distinguished as virulent and avirulent on the differential set of ‘Thatcher’ NILs (Table 2). Only NILs that showed differential response between isolates could be used to distinguish isolates. The Lr genes present in these NILs would be characterized in resistant cultivars. Therefore, in this study, the NILs with resistance genes Lr3a, Lr11, Lr14b and Lr30 that showed susceptibility to all tested isolates were ignored because they were not informative. The values showed that some NILs (e.g. Lr2b, Lr2c, Lr3ka, Lr3bg, Lr15, Lr16 and Lr20) were susceptible to a high number of isolates whereas others (e.g. Lr9, Lr19 and Lr25) were susceptible to none of the tested isolates (Table 2). Therefore, this experiment was not able to assess whether Lr2b, Lr2c, Lr3ka, Lr3bg, Lr9, Lr15, Lr16, Lr19 and Lr25 are present in Iranian wheat lines.

Table 2.  Seedling infection types displayed by near-isogenic lines with known Lr genes with 13 different isolates of P. triticina.

Isolate	84-4	84-15	84-31	84-33	85-1	85-3	85-4	85-5	85-6	85-22	85-27
Lr1	3+	3+	3	3+	3+	3+	3+	3+	3+	;1	0
Lr2a	3	1	1	;1	0;	;1	;1	;1	0	0;	0;
Lr2b	3+	3	1+	3+	;	3+	2	1	0	;	;1
Lr2c	3+	3+	3+	3+	3+	3+	3+	2+	2+	2+	3+
Lr3	3+	3+	3+	3+	3+	3+	4	3+	3+	3+	3+
Lr3ka	3	3	3+	2	3+	3+	3+	2C	3+	3+	3+
Lr3bg	3	3+	3+	3+	3+	3+	2	3+	3+	3+	3+
Lr9	0	0;	0	0;	0	0	0	0	0;	0	0
Lr10	3	3	3	2	;1	3+	3+	;1	3+	1	3+
Lr11	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+
Lr13	4	3	X	3	3+	3+	3+	3+	3+	3+	3+
Lr14a	X	X	2	X	2	2+	3+	2	3+	2	2
Lr14b	4	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+
Lr15	3	3	2	3	3+	3+	2C	3+	3+	3+	3+
Lr16	4	3	3+	3	3+	3+	3+	3+	2+C	3+	2+C
Lr17	1	3	3	3-	;1	X	1	X	3+	1	2
Lr18	3	2	3	2+	3+	3+	3+	3+	2+	2+	2
Lr19	0	0	0;	0	0	0	0;	0;	0	0	0
Lr20	3+	3	3+	3+	3+	3+	3+	X	3+	2+N	–
Lr21	1+	2	2+	2	2−	3+	2+	2+	3+	2−	1
Lr23	3	2+	2	2	2	2	1	3+	3+	;1	;1
Lr24	;1	0;	;1	1	2C	3+	2+	2C	2C	;1	;1
Lr25	0	0;	0	0;	0;	0;	0	0	0;	0;	0;
Lr26	3+	1+	1	3+	;1	3+	2	2	3	0;	0
Lr28	0;	0;	0	0	0;	0;	0;	0;	0;	0;	0;
Lr29	;1	;1	;1	;1	;1	;1N	;1	3+	;1	2N	;1
Lr30	3	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+
Lr32	2	2	2	2	;1	2	1	3+	3+	2	2
Lr33	3+	1	1	1	3+	3+	3+	2+	2	2	2
Lr36	0;	0;	1	0;	1N	1N	;1N	1N	;	3+	;1N
The ITs 0 to 3− and X considered as resistance responses and 3+ to 4 considered as susceptible responses; (–) indicates missed data.

The cultivars and breeding lines displayed a range of ITs and were classified into three main groups based on their seedling responses (ITs) to 13 isolates of P. triticina, which is summarized in Table 3, and these groups are described below. Table 4 shows wheat cultivars and breeding lines with their non-omitted and postulated Lr genes.

Table 3.  Seedling infection types displayed by Iranian wheat genotypes with known Lr genes with 13 different isolates of P.triticina.

Isolate	84-4	84-15	84-31	84-33	85-1	85-3	85-4	85-5	85-6	85-14	85-19	85-22	85-27
Alemoot	X	X	2	2+	;1	;1CN	;1C	2	2	;1N	0;CN	;1	;1CN
Alvand	4	3+	3+	3+	3+	4	3+	3+	3+	3+	4	3+	3+
Atrak	2	2	2	23	0	3+	0;	2	;1	0;	0;	0	0
Chamran	2	2	3	3+	3	4	3+	3+	3+	;1	3+	;1	3+
Darab2	0;	3+	3	2	;	3+	0;	;1	1	1C	3+	1+	;1N
Dez	2+	2+	3−	3−	2+	3+	1+	X	3	;1CN	3+	1	2+
Fong	3+	2	2	2	0;	3+	;1	2	;1	2−	;1	0;	1+
Ghods	X	X	3−	X	2	3+	2	–	3+	X	–	2	3+
Hyrmand	2	X	3−	3	0;	;	;1	3+	1+	;1N	0;CN	;	;1
Kavir	2	2	3−	23^−	;1	3+	1	3+	2	;1N	1CN	;1	1+N
Mahdavi	3−	X	3−	3	;1	–	2N	3−	1+	0;	;1	3+	2
MV17	2		2	23	0	3	0	;1	;1	0;	0;	0	0
Nicknejad	3+	3+	3+	2	1+C	–	1+	1	1+	3+	3+	1+	3+
Sablan	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+
SD-2	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+
Shahriar	2	X	2	2	;1	;1CN	;1C	2	2	;1N	';1CN	;1	1+
Shiraz	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+	3+
Shirodi	2	2	2	3+	0	2	0	;1	;1	0;	0;	0	0
Tajan	X	X	3−	3	;1	;1	;1	3+	3+	1C	2CN	;1	;1
Tos	X	3+	3−	3	1+	3+	X	3−	3+	;1C	;1C	1	2CN
Vee/Nac	2	2	0;	X	0	3+	2+	;1	1+	0;	0;	0	0;
Zarin	X	2	3+	3	X	3+	1C	3+	3+	;1N	;1CN	;1	;1
C-78-18	2	2	3−	3	;	0;	;1	1	;1	;1CN	0;	;1	0
C-78-7	3+	3+	3+	3+	X	3+	3+	3+	3	3+	3+	3+	2
C-79-16	2	3+	3+	2	;1	3+	3+	3+	;1	;1CN	3+	2+	2
C-80-10	X	X+	3+	3+	1+	3+	3+	3+	3+	2+	X	2−	2
C-80-11	2	2	;1	3−	;1	;1CN	;	2	2	;1N	0	;1	;1
C-80-16	;1	2	;1	0;	0;	;1	0;	1+	3+	0;	;1	;1	1
C-80-20	2	2+	2+	3	;1	3+	2	2	2	;1N	0;	;1	1
C-80-4	X	X	X	X	;1	3+	3+	2	1+	X	3+	;1	3+
M-79-6	X	2	3−	;1	;1	;1CN	2	2	3+	1N	;1CN	;1	1
M-79-7	2	;1	2	2	;1	;1CN	0;	1+	3+	0CN	0;CN	;1	1+CN
N-75-16	X	X	2	2	;	3+	1+	2	2+	;1CN	;1CN	;1	;1
S-75-11	2	2	;1	3−	0	0;	0	;1	;1	0;	0;	0;	0
S-78-11	2	X	2+	2	1+C	3+	3+	2	;1	;1CN	;1C	1N	;1
S-79-10	2−	2	3−	2+	0;	–	1+	1N	1+	0;	;1	;1−	–
Bolani	4	4	3+	3+	3+	3+	3+	4	3+	4	3+	4	4
The ITs 0 to 3− and X considered as resistance responses and 3+ to 4 considered as susceptible responses; (–) indicates missed data.

Table 4.  Wheat cultivars and breeding lines with their non-omitted Lr genes and postulated Lr genes

Cultivar/ line	Non-omitted Lr gene(s)	Postulated genes	Cultivar/line	Non-omitted Lr gene(s)	Postulated genes
Alemoot	All Lr genes	U, Lr26	Tos	Lr13	Lr13,U
Alvand	None	No seedling Lr gene	Zarin	None	U
Atrak	Lr1,2b,2c,3ka,10,13,14a,15, 16,18,20,21,24,26,33	Lr1,Lr26,U	Vee/Nac	Lr1,10,20,26	Lr1,Lr10,Lr26
Chamran	None	U	C-79-16	None	U
Darab2	Lr10,20	Lr10,U	C-78-7	None	No seedling Lr gene
Dez	Lr3ka,10,15,20,26	Lr3ka,Lr26	C-78-18	Lr1,3ka, 10, 13, 14a, 15, 17,20	Lr1,Lr10,Lr13,Lr17
Fong	Lr1,13,20,23,26,33	Lr1,Lr13,Lr23,Lr26	C-80-4	Lr13	Lr13,U
Ghods	Lr13	Lr13,U	C-80-20	Lr1,10,20,26	Lr1,Lr10,Lr26
Hyrmand	Lr1,13,16	Lr1,Lr13,U	C-80-11	All Lr genes	U
Kavir	Lr1, Lr13, Lr16	Lr1,Lr13,U	C-80-10	None	U
Mahdavi	Lr13	Lr13,U	C-80-16	Lr1,3ka,10,13,15,17,20,21,23	Lr1,Lr10,Lr17,Lr23
Mv17	Lr1,2b,2c,3ka,10,13,14a,15, 16,18,20,21,23, 24,26,33	Lr1,Lr26,U	M-79-6	Lr1,3ka,10,13,15,17,20,21,23	Lr1,Lr10,Lr13,Lr23,U
Nicknejad	None	U	M-79-7	Lr1,10,13,15,16,17,20,21,23	Lr1,Lr10,Lr13,Lr23,U
Sablan	None	No seedling Lr gene	N-75-16	Lr1,2c,10,13,15,20,24,26	Lr1,Lr10,Lr13,Lr26
SD-2	None	No seedling Lr gene	S-75-11	All Lr genes	U
Shahriar	All Lr genes	U, Lr26	S-78-11	Lr1,10,13,16,18	Lr1,Lr10,Lr13,U
Shiraz	None	No seedling Lr gene	S-79-10	All Lr genes	U
Shirodi	Lr1,20,26	Lr1,Lr26	Bolani	None	No Lr gene
Tajan	Lr1,13,15	Lr1,Lr13,U

Group 1: cultivars without Lr genes

This group comprised ‘Alvand’, ‘Sablan’, ‘Shiraz’ and SD-2 which displayed high ITs (3+ to 4) with all 13 tested isolates. These genotypes do not possess any seedling resistance genes to prevalent leaf rust isolates in Iran.

Group 2: cultivars with unknown Lr genes

This group included ‘Chamran’, ‘Zarrin’ and ‘Nicknejad’ cultivars along with breeding lines C-78-7, C-79-16 and C-80-10. These lines had high ITs against different isolates which were avirulent for NILs having leaf rust resistance genes: Lr1, 2a, 2b, 2c, 3ka, 3bg, 9, 10, 13, 14a, 15, 16, 17, 18, 19, 20, 21, 23, 24, 25, 26, 28, 29, 32, 33 and 36. However, they still showed resistance to some of the tested isolates (Table 3). Therefore, it was not feasible to identify the gene(s) which confers resistance to leaf rust isolates in these genotypes. Also, three additional breeding lines (C-80-11, S-75-11 and S-79-10) along with ‘Alemmot’ and ‘Shahriar’ cultivars had low IT to all tested isolates, and were classified in this group. It was impossible to exclude any leaf rust resistance genes in these breeding lines.

Group 3: cultivars with both known and unknown Lr genes

This group consisted of 13 cultivars and breeding lines which possessed different combinations of Lr genes. Of the Lr genes identified as being present among the 13 wheat cultivars and lines, Lr1 was most frequent (71%), followed by Lr13 (62%), Lr10 (31%), Lr26 (31%) and Lr23 (15%). Our results revealed the presence of Lr1 and Lr26 genes in ‘Atrak’, and ‘MV17’ cultivars. However, the IT patterns of these cultivars to isolates 84-4, 84-33 and 85-3 (virulent to Lr1 and Lr26) showed variation. Based on these findings, there should also be unknown resistance gene(s) in these cultivars to account for the different responses observed.

Lr1, Lr13 and Lr23 were the most likely contributing genes for resistance against leaf rust that was present in ‘Hyrmand’, ‘Kavir’ and ‘Tajan’ cultivars. Moreover, the presence of Lr16 in ‘Hyrmand’ and ‘Kavir’, and Lr15 in ‘Tajan’ was postulated in this study. Lr15 originated from common wheat but it was not known to be deployed in agriculture (McIntosh et al., 1995). Therefore, the combination of Lr1 and Lr13 is the most likely gene combination present in ‘Tajan’ cultivar for resistance to leaf rust. The Lr16 gene carried by ‘Hyrmand’ originated from common wheat. This gene is not highly effective when present alone but its interaction with some other Lr genes can give enhanced levels of resistance. Therefore, higher resistance of ‘Hyrmand’ compared with ‘Tajan’ could be attributed to Lr16 or some other unidentified Lr gene(s) in the former cultivar. According to McIntosh et al. (1995), Lr16 in combination with Lr13 in ‘Columbus’ and ‘Kenyon’ cultivars is considered to provide more durable resistance.

Breeding lines M-79-6 and M-79-7 had similar IT patterns to almost all tested isolates. The combination of Lr1, Lr10, Lr13 and Lr23 is postulated as the most likely for resistance to P. triticina in M-79-6 and M-79-7.

The responses of ‘Ghods’, ‘Mahdavi’, ‘Tos’ and breeding line C-80-4 against the evaluated isolates were similar. These genotypes produced low ITs and high ITs to isolates which were respectively avirulent and virulent to Lr13. Resistance was recorded for these genotypes against isolates virulent for Lr13, implying the presence of some unknown Lr gene (s).

Group 4: cultivars with known Lr genes

Eight genotypes, including four cultivars (‘Dez’, ‘Fong’, ‘Shirodi’ and ‘Vee/Nac’) and four lines (C-78-18, C-80-20, C-80-16 and N-75-16) fell within this group. Lr1 was the most common resistance gene among the wheat genotypes of this group (seven out of eight cultivars and breeding lines), followed by Lr26 (six out of eight cultivars and breeding lines), Lr10 (five out of eight cultivars and breeding lines), and Lr13 (three out of eight cultivars and breeding lines). Lr17 and Lr23 were postulated to be present in two wheat genotypes (Table 4).

The responses of ‘Vee/Nac’ and breeding line C-80-20 against the evaluated isolates were similar. These genotypes produced low ITs and high ITs to isolates which were respectively avirulent and virulent to Lr1, Lr10 and Lr26. Therefore, a combination of Lr1, Lr10 and Lr26 is postulated as the most likely for resistance to P. triticina in ‘Vee/Nac’ and C-80-20.

Discussion

To the best of our knowledge, this is the first report of leaf-rust seedling resistance genes present in Iranian bread wheat cultivars and breeding lines. We identified eight known genes and several unidentified genes for seedling leaf rust resistance in a range of advanced breeding lines and wheat cultivars grown in Iran. Genes masked by suppression in the seedling stage or under the given environment conditions could remain undetected, although they may still have an effect on resistance in the field (Kolmer, 2003). The frequency of each gene was highest for Lr1 (71%), followed by Lr13 (62%), Lr10 (43%), Lr26 (38%), Lr23 (19%) and Lr17 (10%).

The seedling resistance genes were postulated on the basis of gene-for-gene specificity. However, there are obvious limitations to this approach for the analysis of leaf rust resistance genes in wheat. The P. triticina isolates used in this study were inadequate to identify all the seedling Lr genes that were present in the cultivars and breeding lines. A more diverse collection of P. triticina isolates may have allowed the postulation of leaf rust resistance genes in ‘Chamran’, ‘Nicknejad’ and ‘Zarrin’ cultivars and C-78-7, C-79-16 and C-80-10 breeding lines. Breeding lines C-80-11, S-75-11 and S-79-10 along with ‘Alemoot’ and ‘Shahriar’ cultivars showed low ITs with all isolates used in the study. These lines would be appropriate candidates for further genetic analysis to conclusively determine the number and identity of leaf rust resistance genes they carry. According to pedigree analysis, genes Lr1, Lr13 and Lr34 could possibly be present in C-80-11, since this line includes cultivar ‘Ghods’ postulated to carry Lr1 and Lr13 (this study), and ‘Anza’, postulated to carry Lr34, within its pedigree. Also, in this study, similar IT patterns were observed for ‘Alemoot’ and ‘Shahriar’ cultivars. This could be explained by their similar pedigrees. Due to presence of ‘Kavkaz’ and ‘Inia’ (Lr26) in their pedigree, Lr26 is presumably one of the Lr genes present in ‘Alemoot’ and ‘Shahriar’.

The gene Lr1 was the most commonly postulated gene in the Iranian wheat cultivars and breeding lines tested. There is a report demonstrating low level of virulence on Lr1 (Kolmer, 1998). However, virulence to Lr1 is common in Iran (present study), Australia, South Africa and Canada (Singh & Rajaram, 1991; Singh et al., 2001). This gene is, therefore, regarded as being of little use on its own, but it does continue to play a role in gene combinations with genes such as Lr13, Lr17, Lr26 and Lr26 in Iran (Tables 2 and 4).

The second most commonly postulated gene in the tested cultivars and breeding lines was Lr13. On a global scale, Lr13 is probably the most widely distributed gene for resistance to P. triticina (McIntosh et al., 1995; Winzeler et al., 2000). Although Lr13 was not effective against more than 50% of the P. triticina isolates in Iran, it is still considered important because it is present in several cultivars bred in CIMMYT in combination with other adult plant resistance genes which continue to give excellent leaf rust protection (Huerta-Espino et al., 2011). According to Singh et al. (2001), there are different reports showing high levels of susceptibility of some wheats carrying Lr13 in South Africa, Mexico and South America, but this gene is one of the most important genes in other parts of the world. Lr13 was the most commonly detected gene present in the UK (Singh et al., 2001), Sweden (Hysing et al., 2006) and European wheat cultivars (Winzeler et al., 2000; Pathan & Park, 2006).

The Lr10 gene originated from the T. aestivum L. genome and it exists in many old Australian as well as North American cultivars, and in lines created by the CIMMYT breeding programs (Singh & Rajaram, 1991; McCallum & Seto-Goh, 2010). Lr10 occurred in combination with other identified Lr genes in ‘Kavir’ and ‘Vee/Nac’ cultivars and breeding lines including C-78-18, C-80-16, C-80-20, M-79-6, M-79-7, N-75-16 and S-78-11.

The Lr23 gene, located on the short arm of chromosome 2B, was postulated in 19% of genotypes tested in the present study. Due to the low levels of virulence on Lr23 in Iran, it seems that the gene could be a good candidate to incorporate in Iranian wheat cultivars. Although there are different reports suggesting high virulence to Lr23 (McIntosh et al., 1995; Wamishe & Milus, 2003), this gene had been considered useful by Oelke & Kolmer (2004). They showed the ‘Thatcher’ line with Lr23 had effective resistance in wheat plots and likely contributes some effective resistance in the hard red spring wheat cultivars.

The results of this study showed that Lr26 was one of the postulated genes in 38% of cultivars and breeding lines. Lr26 was derived from rye chromosome arm 1RS which also carries Sr31, Yr9 and Pm8 genes (McIntosh et al., 1995). Resistance genes Yr9 and Lr26 became ineffective because of the shift of P. triticina and P. striiformis f. sp. tritici phenotypes to virulence to Yr9 and Lr26, respectively (Torabi et al., 1995; Mesterhazy et al., 2000).

Pyramiding genes has been suggested as a method to achieve more durable resistance against pathogens with low genetic diversity, high gene flow and asexual mating systems (McDonald & Linde, 2002; Hysing et al., 2006). The combination of several effective resistance genes into a single cultivar should extend the period of resistance since mutation at several avirulence loci would be required to produce new virulent isolates. Slow rusting or partial resistance has been reported to be a more durable resistance than single seedling resistance genes (Singh et al., 2001). The pyramiding of such differently functioning genes is simplified by the use of molecular markers that have been developed for most genes for leaf rust resistance. These markers already have been developed for slow rusting resistance genes Lr34 (Lind & Gultyaeva, 2007) and Lr46 (Lagudah et al., unpublished data). The presence of some genes conferring slow rusting phenotypes can be predicted by pedigree analysis of each cultivar. Considering this method, it would be suggested that ‘Chamran’ and ‘Shirodi’ cultivars carry Lr46 due to the presence of ‘Attila’ in their pedigree. The genotypes ‘Shahriar’ and ‘Alemoot’, ‘Vee/Nac’ and C-80-20 possibly carry Lr34 from having ‘Anza’ and ‘Vee/Nac’ as parents in their pedigree, respectively.

The results of the pathogen survey in this research revealed that almost 80% of NILs showed moderate to high levels of susceptibility (30 to 100%) to tested isolates. The frequencies of virulence to genes Lr9, Lr19, Lr25, Lr29, Lr32, Lr36, Lr21 and Lr14a were low in Iran, so these genes have the potential for introduction into Iranian wheat cultivars. Regardless of several reports indicating lack of virulence to Lr21 in North America and Australia, it seems there are virulent leaf rust isolates to this gene in Europe (Hanzalova & Bartos, 2006; Hanzalova et al., 2008; Huerta-Espino et al., 2011) and Iran (Afshari et al., 2006; Huerta-Espino et al., 2011). Furthermore, virulence to Lr32 is low worldwide, although we observed 23% incidence of virulence to this gene. This can be explained by two reasons. Firstly, taking into account that leaf rust is an endemic disease in Iran, there might be some virulent isolates to this gene. Secondly, the seed source for Thatcher-Lr32 may be contaminated with other unknown seeds. In order to address these hypotheses, another seed source of Lr32 is required.

The knowledge of which leaf rust seedling resistance genes are present facilitates future studies and the use of adult plant resistance genes in the wheat cultivars. The genes Lr1, Lr10, Lr13, Lr17, Lr23 and Lr26 were the most frequent seedling leaf rust resistance genes postulated to be present in Iranian wheat cultivars and breeding lines. Therefore, there is relatively inadequate variation in Lr genes carried by wheat cultivars commercially grown in Iran. Future host selection pressure on the pathogen could be further decreased by rotating genes through time and space by mixtures or regional deployment of cultivars with different effective resistance genes. Nevertheless, classical genetic and molecular marker analyses will be needed to further validate and expand the findings of the present study regarding the Lr genes responsible for both seedling and adult plant resistance to leaf rust in Iranian wheat genotypes.

Acknowledgements

We would like to thank Dr Peter Chandler from CSIRO, Canberra, Australia for editing the manuscript. We also thank Dr Eugene Milus from the University of Arkansas for his helpful comments and editing the first draft of the manuscript. Special thanks go to Dr Kevin Thompson from the University of Arkansas for his help in the statistical analysis of the data.