Seedling and adult plant resistance to leaf rust in 46 Chinese bread wheat landraces and 39 wheat lines with known Lr genes

ZHANG Pei-pei, Takele Weldu Gebrewahid , , ZHOU Yue, LI Qing-luo, LI Zai-feng , LIU Da-qun

1 Department of Plant Pathology, College of Plant Protection, Hebei Agricultural University, Baoding 071001, P.R.China

2 College of Agriculture, Aksum University-Shire Campus, Shire-Indaslassie 314, Ethiopia

3 College of Biochemistry and Environmental Engineering, Baoding University, Baoding 071001, P.R.China

Abstract Wheat leaf rust, caused by Puccinia triticina (Pt), is an important foliar disease that has an important inf luence on wheat yield. The most economic, safe and effective way to control the disease is growing resistant cultivars. In the present study, a total of 46 wheat landraces and 34 wheat lines with known Lr (leaf rust resistance) genes were inoculated with 16 Pt pathotypes for postulating seedling resistance gene(s) in the greenhouse. These cultivars and f ive wheat differential lines with adult plant resistance (APR) genes (Lr12, Lr22b, Lr34, Lr35 and Lr37) were also evaluated for identif ication of slow rusting resistance in the f ield trials in Baoding, Hebei Province of China in the 2014-2015 and 2015-2016 cropping seasons. Furthermore, 10 functional molecular markers closely linked to 10 known Lr genes were used to detect all the wheat genotypes. Results showed that most of the landraces were susceptible to most of the Pt pathotypes at seedling stage. Nonetheless, Lr1 was detected only in Hongtangliangmai. The f ield experimental test of the two environments showed that 38 landraces showed slow rusting resistance. Seven cultivars possessed Lr34 but none of the landraces contained Lr37 and Lr46. Lr genes namely, Lr9, Lr19, Lr24, Lr28, Lr29, Lr47, Lr51 and Lr53 were effective at the whole plant stage. Lr18, Lr36 and Lr45 had lost resistance to part of pathotypes at the seedling stage but showed high resistance at the adult plant stage. Lr34 as a slowing rusting gene showed good resistance in the f ield. Four race-specif ic APR genes Lr12, Lr13, Lr35 and Lr37 conferred good resistance in the f ield experiments. Seven race-specif ic genes, Lr2b, Lr2c, Lr11, Lr16, Lr26, Lr33 and LrB had lost resistance. The 38 landraces showed slow rusting resistance to wheat leaf rust can be used as resistance resources for wheat resistance breeding in China.

Keywords: Triticum aestivum L., wheat leaf rust, gene postulation, adult plant resistance

1. Introduction

Wheat leaf rust, caused by the fungus Puccinia triticina (Pt), is one of the most important foliar diseases in wheat (Triticum aestivum L.) worldwide. It occurs in a wide range of climates wherever wheat is grown and can cause yield losses of up to 40% under favorable conditions (Knott 1989). In China, wheat leaf rust affects about 15 million hectares of wheat annually. Frequent wheat leaf rust epidemics were raised in the southwestern and northwestern China, the middle and lower Yangtze River Valley, and the southern parts of Huang-Huai f lood plain (Huerta-Espino et al. 2011). In the last two years, wheat leaf rust had occurred with higher prevalence and severity in most of wheat growing areas which suggested that there was a potential for a new leaf rust disease epidemic. Planting resistant cultivars is the most eff icient and environmentally friendly way to minimize losses by the disease.

Genetic resistance to rust pathogens is generally divided into three general categories (Lan et al. 2017): race-specif ic seedling resistance, also known as all-stage resistance; race-specif ic APR; and race non-specif ic APR, also known as slow-rusting or partial resistance.

To date, more than 100 Lr genes have been reported in wheat and its relative genomes (Mc Intosh et al. 2017). Seventy-nine of them are permanently catalogued in wheat (Qureshi et al. 2018). Most of them are race-specif ic Lr genes that confer hypersensitive reactions and interact with the pathogen in a gene-for-gene manner. Ten genes, among the race-specif ic genes, Lr12, Lr13, Lr22, Lr35, Lr37, Lr48, Lr49, Lr74, Lr75 and Lr77 are race-specif ic APR genes. This kind of resistance often loses effectiveness after deployment in agriculture for several years. Race non-specif ic resistance is usually effective at the adult plant stage. It is associated with longer latent period, lower infection frequency, smaller uredinial size, reduced duration of sporulation, and less spore production per infection site (Caldwell 1968). This kind of resistance is more durable. At present, only four Lr genes, Lr34, Lr46, Lr67 and Lr68 conferred slow rusting resistance. Therefore, it is important to identify new resistance genes to cope with dynamic and rapidly evolving virulent pathogen populations.

Gene postulation based on multi-pathotype response has been widely used by researchers to identify leaf rust resistance genes. In this procedure, the responses of test lines are compared with the responses of an array of lines with known resistance genes when inoculated with sets of Pt pathotypes whose avirulence/virulence phenotypes are known (Kolmer 2003). Although the Lr genes of the seedling can be postulated on the basis of gene-for-gene strategy, there were obvious limitations due to the inability to identify all the Lr genes. It is inevitable that gene postulation will be affected by an uncontrolled environment and the discriminability of a couple of pathotypes. It is diff icult to identify all leaf rust resistance genes especially when several genes were present in a cultivar.

In recent years, molecular markers including simple sequence repeats (SSR), amplif ied fragment length p olymorphisms (AFLP), resistanc e gene analog polymorphisms (RGAP), expressed sequence tags (EST), single nucleotide polymorphisms (SNP) and genotypingby-sequencing (GBS), are all useful tools for markerassisted selection (MAS). Molecular markers based on cloned gene sequences are accurate but the accuracy of linked markers in predicting the presence of a target gene depends on the linkage distance. Linked markers can solve problems associated with gene postulation, including prediction of individual genes in gene combinations and APR genes.

In China, lack of information about resistance genes in wheat cultivars made it diff icult to utilize resistant cultivars to control leaf rust. The Chinese wheat landraces might be good resistant resources to leaf rust. Therefore, in the present study, 46 bread wheat landraces from China and 39 wheat differentials with known Lr genes were tested to identify seedling Lr and APR genes.

2. Materials and methods

2.1. Plant materials and Pt pathotypes

A set of 46 Chinese wheat landraces and 34 differential lines, mostly near-isogenic lines in the background of Thatcher with known Lr genes were used to test the seedling responses to 16 Pt pathotypes collected from China in the greenhouse (Tables 1 and 2). The differential lines were kindly provided by CIMMYT. Pt pathotypes were designated following the coding system of Long and Kolmer (1989), with the addition of a four-letter code for reactions to a fourth set of differentials (http://www.ars.usda.gov/SP2UserFiles/ad_hoc/36400500Cerealrusts/pt_nomen.pdf). Infection types (ITs) displayed by the genotypes at the seedling tests were used as the basis for postulating resistance genes (Tables 1 and 2). The 46 Chinese landraces and 39 differential lines (including f ive lines with APR genes, Lr12, Lr22b, Lr34, Lr35 and Lr37) were also used in the f ield tests. CIMMYT line Saar with typical slow leaf rusting resistance and susceptible line Zhengzhou 5389 were used as slow rusting and susceptible controls, respectively.

2.2. Evaluation of seedling responses in the greenhouse

All the wheat landraces and differential lines with known Lr genes were tested with 16 Pt pathotypes in the greenhouse (Tables 1 and 2). Seedlings were grown in a growth chamber (30 cm×50 cm). When the f irst leaf was fully expanded, inoculation was performed by brushing urediniospores from fully infected susceptible genotype Zhengzhou 5389 onto the seedlings to be tested. Inoculated seedlings were placed in plastic-covered cages and incubated at 15°C and 100% relative humidity (RH) for 24 h in darkness. They were then transferred to a growth chamber programmed with 12 h light/12 h darkness at 18 to 22°C and 70% RH. ITs were scored 10 to 14 days after inoculation according to the Stakman scale as modif ied by Roelfs et al. (1992). 0=no uredinia or other macroscopic sign of infection, ;=no uredinia but small hypersensitive necrotic or chlorotic f lecks present, 1=small uredinia surrounded by necrosis, 2=small to medium uredinia surrounded by necrosis or chlorosis (green islands may be surrounded by necrotic or chlorotic border), 3=medium uredinia with or without chlorosis, 4=large uredinia without chlorosis, X=heterogeneous, similarly distributed over the leaves, C=more chlorosis than normal for the IT, +=uredinia somewhat larger than normal for the IT. Plants with ITs 0 to 2 were considered to be resistant and those with ITs of 3 to 4 were susceptible. The gene postulation was performed following Dubin et al. (1989).

Table 1 Seedling infection types on 34 wheat lines with known leaf rust resistance genes when tested with 16 pathotypes of Puccinia triticina collected from China

Table 2 Seedling infection types on 46 wheat landraces and Zhengzhou 5389 tested with 16 Puccinia triticina pathotypes collected from China1)

Table 2 (Continued from preceding page)

2.3. Field testing

All the 46 wheat landraces, slow rusting check Saar, susceptible check Zhengzhou 5389 and a set of 39 differential lines were planted in a randomized complete block design with two replicates in Baoding, Hebei Province in the 2014-2015 and 2015-2016 cropping seasons. About 30 seeds of each line were sown in a single-row plot with 1.5-m length and 30 cm between rows. Spreader rows of Zhengzhou 5389 were planted perpendicular and adjacent to the test rows. Leaf rust epidemics was initiated by spraying aqueous suspensions of urediniospores of mixed Pt pathotype (THTT, THJS and PHTP) to which a few drops of Tween 20 (0.03%) was added onto the spreader rows at tillering stage. The three Pt pathotypes were selected as dominant pathotypes due to wide virulence to the testing genotypes at the seedling stage in the greenhouse (Tables 1 and 2). Disease severities were assessed thrice at weekly intervals with the f irst scoring four weeks after inoculation using the modif ied Cobb scale (Peterson et al. 1948). The host response to infection in adult plant was determined according to Roelfs et al. (1992), where, R=resistant, visible chlorosis or necrosis, no uredia are present; MR=moderately resistant, small uredia are present and surrounded by either chlorotic or necrotic areas; M=intermediate, variable sized uredia are present; some with chlorosis, necrosis, or both; MS=moderately susceptible, medium sized uredia are present and possible surrounded by chlorotic areas; and S=susceptible, large uredia are present, generally with little or no chlorosis and no necrosis. Cultivars and lines that were seedling-susceptible to the mixed Pt pathotypes and with lower maximum disease severity (MDS, less than 15%) in f ield trials were considered to be slow rusting genotypes (Li et al. 2010).

2.4. DNA extraction and molecular marker detection

Genomic DNA of all tested genotypes was extracted and purif ied from seven-day old wheat seedlings based on the modif ied CTAB method (Sharp et al. 1988). The STS, SCAR and CAPS molecular markers of ten known Lr genes (Lr1, Lr9, Lr10, Lr19, Lr20, Lr24, Lr26, Lr34, Lr37 and Lr46) were used for gene detection. The primer sequences, references and PCR amplif ication conditions for the target Lr genes are presented in Table 3. PCR reactions were performed in a volume of 20 μL containing 10 μL 2×Taq MasterMix (Beijing ComWin Biotech Co., Ltd., China, http://www.cwbiotech.com/), 6 pmol of each primer and 100 ng of template DNA. CAPS marker cs LV46G22 specif ic to Lr46 was kindly provided by Dr. Lagudah E (Commonwealth Scientif ic and Industrial Research Organization, Plant Industry, Canberra, Australia). Primers were synthesized by Sangon Biotech (Shanghai) Co., Ltd., in China. The PCR experiments were repeated twice independently. The PCR product was separated on a 1.5% agarose gel in 1× TAE buffer and stained with ethidium bromide, then photographed under UV light.

Table 3 Primer sequences and PCR amplification programs for different primer combinations

3. Results

3.1. Gene postulation from seedling reactions

Variations of ITs on the 34 differential lines carrying known Lr genes inoculated with 16 Pt pathotypes (Table 1) indicated the possibility of identifying 15 Lr genes (Lr1, Lr2a, Lr2b, Lr3ka, Lr10, Lr14a, Lr15, Lr18, Lr20, Lr21, Lr26, Lr30, Lr36, Lr44 and LrB). Lr9, Lr19, Lr24, Lr28, Lr29, Lr47, Lr51 and Lr53 were effective against all pathotypes. Postulation of Lr2c, Lr3, Lr3bg, Lr11, Lr13, Lr14b, Lr16, Lr17, Lr23, Lr33 and Lr45 were not possible because high ITs were recorded with most of Pt pathotypes.

Lr2b, Lr10, Lr14a, Lr26 and LrB showed susceptible to most of the pathotypes. Lr2b, Lr26 and LrB each showed resistance to only one pathotype, PHTS, PGTS and THSP, respectively. Lr10 was resistant to two Pt pathotypes, THSP and PHTP. Lr14a showed intermediate reaction to THTK and THSP but it was susceptible to the other Pt pathotypes. Lr21 was resistant to four Pt pathotypes. Each of Lr1 and Lr30 displayed low reactions to nine Pt pathotypes. Three genes, Lr2a, Lr3ka and Lr15 were resistant to ten Pt pathotypes. Lr18 conferred resistance to 11 Pt pathotypes. Both Lr36 and Lr44 gave low ITs to 12 Pt pathotypes. Lr20 produced low reactions to 13 Pt pathotypes.

ITs of the 46 tested landraces are listed in Table 2. Forty-three of the tested wheat landraces showed high ITs to most Pt pathotypes, indicating that it was impossible to postulate the Lr genes in these lines. Xiaohongmang showed medium resistance (2+) to only one Pt pathotype (PGTS). Lr1 combined with Lr20 and Lr36 might be present in Hongtangliangmai since it displayed low ITs against all avirulent pathotypes to Lr1, Lr20 and Lr36. Baiheshang showed low ITs with all avirulent pathotypes to Lr14a, Lr2b and Lr21, indicating it may contain Lr14a, Lr2b and Lr21.

3.2. Slow rusting resistance in f ield tests

The MDS value of the susceptible check, Zhengzhou 5389, was 100% in both the 2014-2015 and 2015-2016 cropping seasons in the f ield experimental site, indicating the disease developed well in both seasons.

Eleven Lr genes, viz., Lr9, Lr18, Lr19, Lr24, Lr28, Lr29, Lr36, Lr45, Lr47, Lr51 and Lr53 showed resistance with IT resistance (R) and MDS less than 5% at the adult plant stage (Appendix A), indicating these genes were effective. Lr34 and Lr14b showed high ITs (MS or S) in the f ield but the MDS of the two genes were less than 15%, indicating that the two differential lines showed slow rusting resistance. The MDS of Lr17, Lr21, Lr23 and Lr44 was less than 30% with ITs of M. Therefore, these lines also may have good resistance at the adult plant stage. Four race-specif ic APR genes, Lr12, Lr13, Lr35 and Lr37 showed good resistance to the mixture Pt pathotypes in the f ield conditions with ITs R (Lr35), MR (Lr12) or M (Lr13 and Lr37) and disease severity less than 10%. Another race-specif ic gene Lr22b was susceptible with IT of S and disease severity 80%. A total of eight Lr genes, Lr2a, Lr3, Lr3ka, Lr3bg, Lr10, Lr14a, Lr20 and Lr30 with the MDS more than 30% but less than 60%, indicated minor resistance may be present in these lines. The MDS of Lr1, Lr2b, Lr2c, Lr11, Lr15, Lr16, Lr26, Lr33 and LrB was more than 60%, indicating that these differential lines had lost resistance in the f ield.

All the 46 tested wheat landraces showed high ITs to the Pt pathotypes in the f ield. The mean MDS of these landraces in the 2014-2015 and 2015-2016 cropping seasons are listed in Appendix A. A total of 38 wheat landraces of the tested genotypes had mean MDS less than 15%, indicating these lines showed slow rusting resistance in the f ield. Seven wheat landraces (Baitutou, Hongpimai, Daqingsui, Wutongmangmai, Hongbanmang, Dalatou and Kulouding) had the mean MDS ranged from 15 to 40%. Only one wheat landrace (Baimangbiansui) showed the MDS 80%. The result showed that most of the landraces might carry unknown minor or slow-rusting resistance genes.

3.3. Molecular detection

All the 39 differential lines with known Lr genes were used to verify the specif icity of the molecular markers linked to Lr1, Lr9, Lr10, Lr19, Lr20, Lr24, Lr26, Lr34, Lr37 and Lr46. The results showed that they were specif ic and robust for gene detection. These markers were then used to test the 46 Chinese landraces to conf irm the postulated Lr genes. Lr9, Lr10, Lr19, Lr20 and Lr24 were not present in all the 46 tested landraces. Lr1 was detected in Hongtangliangmai. None of the landraces contained Lr26. It was not possible to postulate Lr34, Lr37 and Lr46 in the seedling tests because all of these are APR genes. Therefore, the molecular markers for these three APR genes were used to test all the landraces. Seven lines, Zijielumai, Xinlijun, Hongtutou, Hongguangtou, Baitutou, Baimazhatou and Baiguozitou had the 150 bp specif ic band for STS marker csLV34, indicating that these landraces contained Lr34. Lr46 and Lr37 were not detected in all the 46 wheat landraces.

4. Discussion

4.1. Known Lr genes from seedling and adult plant reactions

Eight Lr genes, viz., Lr9, Lr19, Lr24, Lr28, Lr29, Lr47, Lr51 and Lr53 showed high resistance both at the seedling and adult plant stages. However, these genes are rarely used in Chinese released wheat cultivars (Li et al. 2010). Hence, these genes can be used as effective genes in wheat breeding program. Lr18, Lr36 and Lr44 showed resistance to most of the Pt pathotypes at the seedling stage and displayed high APR in the f ield. This showed that these genes can also be used as effective genes for wheat breeding.

Lr45 derived from rye was susceptible to all the Pt pathotypes at the seedling stage but showed high resistance at the adult plant stage (McIntosh et al. 1995). This indicated that it might be an APR gene. Lr45 is probably associated with agronomic def iciencies that will prevent its exploitation in wheat cultivars (Mc Intosh et al. 1995).

Lr2b, Lr2c, Lr11, Lr16, Lr26, Lr33 and LrB (derived from common wheat) showed susceptible ITs to most of the Pt pathotypes at the seedling stage and their MDS in the f ield were more than 60%. Therefore, these genes had lost resistance in China at present.

Lr1 and Lr15 were effective to some of the Pt pathotypes at the seedling stage but showed high MDS at the adult plant stage. These genes can be used in the seedling gene deployment. Lr17, Lr21 and Lr23 gave high ITs to most of the pathotypes at the seedling stage but the MDS was less than 30% with IT M in the f ield. These genes can also be used in Chinese wheat breeding program. Lr14b and Lr34 gave susceptible ITs but had low MDS in the f ield. Lr14b was originally transferred to Thatcher from the South American cultivar Maria Escobar (Herrera-Foessel et al. 2012). Maria Escobar carried the closely linked gene Lr68 (a slow rusting gene), this would explain why the Thatcher near-isoline for Lr14b displayed APR in the f ield trials (Herrera-Foessel et al. 2012). Four race-specif ic APR genes, Lr12, Lr13, Lr35 and Lr37 showed good resistance in the f ield experiments with ITs R, MR or M and disease severity less than 10%, indicating these genes can be used in APR breeding. Another APR gene, Lr22b in Thatcher showed IT S and disease severity 80%, showed that it had lost resistance in the f ield. However, Zhang et al. (2017) reported that the residue resistance of Lr22b also plays a minor effect on wheat rust.

On the other hand, Lr2a, Lr3, Lr3ka, Lr3bg, Lr10, Lr14a, Lr20 and Lr30 were highly susceptible with MDS ranged from 30 to 60% in the f ield indicating that these genes had lost resistance in China.

4.2. Application of landraces for breeding

Among the 46 wheat landraces, a total of 38 wheat landraces showed slow-rusting resistance. The landraces were bred by farmers according to the performance in the f ield under the disease conditions. At present, the landraces are often ignored by breeders as its poor agronomic performance. However, the landraces had rich diversities and most of them showed high resistance to leaf rust in the f ield. Therefore, they may contain new effective Lr genes. Many researches on gene mapping for wheat disease on Chinese landraces had been reported. The known powdery mildew resistance genes, viz., Pm24 (Huang et al. 1997), Pm24b (Xue et al. 2012), and Pm47 (Xiao M G et al. 2013) were identif ied in Chinese landraces Chiyacao, Baihulu and Hongyanglazi, respectively. Five QTLs for Fusarium head blight resistance were found in the Chinese wheat landraces Huangfangzhu and Haiyanzhong, respectively (Li et al. 2011; Tao et al. 2012). Stripe rust resistance gene YrHong244 (Ning et al. 2014) and YrChk (Liu et al. 2007) were mapped in Hongcaomai and Chike, respectively. Wangshuibai carried a major Fusarium resistance QTL Fhb1 (Xiao J et al. 2013). Lr34 was found in 359 Chinese landraces among 422 tested wheat landraces (Yang et al. 2008). The Chinese wheat landrace Pingyuan 50 was an important landrace in the Yellow and Huai f lood plain of China in the 1950s and showed good APR to powdery mildew, stripe rust and leaf rust (Lan et al. 2010; Asad et al. 2014). Therefore, the Chinese landraces contained abundant effective resistance genes to wheat leaf rust and had been proven to be valuable resistance sources for wheat resistant breeding.

Thus, it is necessary to identify effective resistance genes in Chinese wheat landraces. Genetic populations (RIL, DH and F2lines) can be constructed by crossing these resistant lines with other susceptible cultivars. Phenotypic data, bulk segregation analysis (BSA) method combined with molecular markers (SSR, SNP, GBS and others) can be used to identify effective Lr genes within the created populations. Genome-wide association study (GWAS) as a standard method can also be used for gene discovery in the landraces. This research f inding could be benef icial to Chinese bread wheat resistant breeding.

5. Conclusion

For the 46 landraces, Lr1 was present only in one landrace. A total of 38 landraces showed slow rusting resistance in the f ield. Seven landraces possessed Lr34. The known Lr genes, Lr9, Lr19, Lr24, Lr28, Lr29, Lr47, Lr51 and Lr53 were effective at all plant growing stages. Lr18, Lr36 and Lr45 showed high resistance at the adult plant stage. Lr34 as a slowing rusting gene showed good resistance in the f ield. Race-specif ic APR genes, Lr12, Lr13, Lr35 and Lr37 conferred good resistance in the f ield experiments. Seven genes, viz., Lr2b, Lr2c, Lr11, Lr16, Lr26, Lr33 and LrB had lost resistance.

Acknowledgements

This study was supported by the National Key Research and Development Program of China (2017YFD0300906-07).

Appendixassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm