Wheat (Triticum spp.) with over 760 million tons of annual global production from a total area of more than 219 million hectares is one of the major food crops that provide the majority of calories worldwide (FAO, 2021). Several biotic and abiotic stresses, reducing cultivable fertile land and rapid climate change conditions, are the major limiting factors for global food production. Therefore, ensuring global food and nutrition security is frightening with a soaring world population, which is expected to reach 9 billion by 2050 (Rosegrant and Agcaoili, 2010). Among several constraints to wheat production, biotic stresses including fungal and insect damage could cause up to 21.5% yield reduction (Savary et al., 2019). Furthermore, climate change effects on temperature, relative humidity, rainfall, photoperiod, wind speed, and direction could potentially upset global wheat production directly or indirectly by maneuvering the growth, multiplication, pathogenesis, dissemination, and survival of plant pathogens (Prasad et al., 2021b).

Wheat rusts caused by Puccinia spp. are among the most threatening and economically important crop diseases. Stem or black rust caused by Puccinia graminis (Pers.) f. sp. tritici Eriks. & E. Henn. (Pgt) is one of the most detrimental diseases of wheat and has continued to threaten global wheat production since time immemorial. It can destroy all the crops if the infection happens in the early crop season and could cause 100% yield loss (Singh et al., 2011). Several stem rust epidemics were documented in the major wheat-growing regions of the world until the late 1950s (Saari and Prescott, 1985). After that era, this disease was efficiently managed through the extensive use of host resistance and eradication of alternate host, barberry (Berberis vulgaris L.), especially in Europe and the USA. One of the most important historical events during this period was the introduction of rye (Secale cereale L.) gene (1B/1R translocation), carrying Lr26/Sr31/Yr9/Pm8, into bread wheat (Mettin et al., 1973; Zeller, 1973). The 1B/1R translocation contributed 12%–20% yield jump and imparted resistance to major biotic and abiotic stresses (Cox et al., 1995). This introduction, popularly known as Sr31 alone or with other Sr genes, kept the stem rust fungus under control for several decades. However, such a contentment did not last for too long and an Sr31 virulent isolate, named TTKSK based on the North American nomenclature system and widely known as Ug99, was detected from Uganda during 1998 (Pretorius et al., 2000). The emergence of Ug99 race, virulent on at least 80% of world’s wheat varieties grown during that period, has augmented the risk of stem rust epidemics worldwide (Singh et al., 2011). The Ug99 and 15 known variants in its lineage have been reported from 14 countries in East Africa and the Middle East (https://rusttracker.cimmyt.org/?page_id=22; Singh et al., 2011; Patpour et al., 2016; Prasad et al., 2016; Nazari et al., 2021). The occurrence and movement of Ug99, its variants, and other new races caused severe localized stem rust outbreaks and epidemics in East Africa (Olivera et al., 2015), Europe (remained wheat stem rust-free since the mid- 20th century; Lewis et al., 2018), and Central Asia (Shamanin et al., 2020), indicating that stem rust is re-emerging as a challenging menace to wheat production worldwide.

Wheat stem rust is largely prevalent in nearly 7 million hectares of Central and Southern Indian states (Bhardwaj et al., 2019). India has witnessed several stem rust outbreaks in the past. Some notable epidemics were reported from Central India during 1786, 1805, 1827, 1828–1829, 1831–1832, 1879, 1894–1895, and 1907. Subsequently, a devastating stem rust outbreak was reported from the southern district of Bombay in 1946–1947 and again from Central India during 1948–1949 (Nagarajan and Joshi, 1975). Since then, stem rust was kept under control through the effective monitoring of pathogen variability and deployment of stem rust-resistant wheat cultivars and no epidemic was witnessed during last seven decades (Bhardwaj et al., 2019). Nevertheless, efforts must continue to combat possible future stem rust outbreaks due to continuous evolution in local Pgt races and/or exotic incursions.

Developing and deploying stem rust-resistant wheat cultivars complemented by understanding the pathogen variability and distribution is an effective, sustainable, and environmentally safe approach to manage this disease (Prasad et al., 2020a). Monitoring the pathogen variability and distribution through continuous surveillance is critical for breeding rust-resistant wheat. The Puccinia spp. causing wheat rusts are highly dynamic and continue to evolve, through mutation, somatic recombination, introduction, or selection, to produce new isolates periodically (Patpour et al., 2016). Such evolving nature of these pathogens renders resistant varieties susceptible in a short span. Therefore, continuous breeding for rust resistance including anticipatory breeding, search for new resistance sources, and monitoring of pathogen variability, structure, and distribution are prerequisite and continuous approaches for effective stem rust management (Abebe et al., 2012). The present study reports the following: (i) the distribution pattern of Pgt isolates in the Indian sub-continent from 2016 to 2022, (ii) virulence-based phenotypic and microsatellite marker-based genotypic diversity among these isolates, (iii) stem rust response of recently developed wheat varieties at seedling and adult plant stages at five different geographical locations, and (iv) the presence of Sr genes in these varieties postulated through gene matching technique and genetic analysis using closely linked molecular markers to these genes.

Wheat stem rust has re-emerged as a serious menace to wheat production globally since the evolution of Ug99 in Uganda. To date, about 15 variants in the Ug99 lineage have been identified within its lineage. Some of these variants have spread to several wheat-growing areas in different countries including Zimbabwe, South Africa, Sudan, Yemen, Egypt, and Iran. The Ug99 variants and other new races reported from East Africa (Digalu race; Olivera et al., 2015), Europe (Olivera Firpo et al., 2017; Lewis et al., 2018), and Central Asia (Shamanin et al., 2020) have caused severe localized stem rust outbreaks and endangered wheat productivity and food security (Lewis et al., 2018). The further spread of the races in the Ug99 lineage to major wheat-growing countries in Asia, including India where almost 7 million hectares area in Central and Southern India is highly prone to stem rust outbreaks, and beyond could devastate wheat production in the affected areas (Olivera et al., 2018). Therefore, survey for monitoring the evolution or introduction of Pgt virulences supported by virulence phenotypes and molecular genotypes is crucial and must be undertaken regularly. Moreover, replacing stem rust-susceptible wheat varieties with resistant ones is a systematic approach to minimize yield losses and thwart pathogen survival (Randhawa et al., 2018; Prasad et al., 2020b). The present study was designed to monitor the variability and distribution pattern of Pgt isolates in India and neighboring countries from 2016 to 2022 through exhaustive surveillance, virulence phenotyping, and SSR genotyping, evaluate recently released wheat varieties for stem rust response at seedling and adult plant stages during two crop seasons at five stem rust-prone geographically different areas, and to undertake genetic analysis of wheat varieties to identify Sr genes in them.

The occurrence of wheat stem rust was sporadic during the crop years 2016 to 2022. Stem rust incidence was extremely low during 2015–2016 and 2021–2022 as we could detect only a few isolates of pathotypes 11, 21-1, 21A-2, 40A, 40-2, 40-3, and 122 from 54 samples during 2015–2016, and 11, 21, 34-1, 40A, and 40-2 from 62 samples during 2021–2022 ( Table 3 ). Most of the advanced wheat material was stem rust-free while some old genotypes and admixtures had up to 80S severity under natural conditions. The availability of Sr31 and other known and unknown Sr genes in recently developed wheat varieties, effective against all known pathotypes of stem rust in India and neighboring countries, accompanied by poor rainfall and a significant reduction in relative humidity over the years could have hampered the survival and infection rate of the natural population of Pgt and, therefore, poor stem rust incidence (Ayliffe et al., 2008; Prank et al., 2019). As revealed previously (Prasad et al., 2018; Prasad et al., 2022), the Pgt population reported from India and Nepal remained avirulent to Sr7a, Sr26, Sr27, Sr31, Sr32, Sr33, Sr39, Sr40, Sr43, SrTmp, and SrTt3 to date. Some of these genes, for instance, Sr26, have been found to be effective against the Pgt population including Ug99 globally. Hence, these genes can be targeted for prudent wheat breeding for stem rust resistance through their pyramiding with other effective major and minor genes (Singh et al., 2015). The Sr26 has been extensively used in the Australian wheat breeding program. Still, its use in India is seldom and no Indian wheat variety, including the tested varieties in this study, is reported to contain it.

A comparison of Pgt pathotypes identified in the present study revealed some dissimilarity with previous reports (Jain et al., 2013; Prasad et al., 2018). Pathotypes 42, 117, and 117-6, which were not detected during 2009–2015, reappeared during this period, whereas pathotype 40-1, virulent on Sr24, was present during 2009–2015 but did not appear afterward (Prasad et al., 2018). Pathotypes 42 (2019–2020, Karnataka), 117 (2017–2018, Gujarat), and 117-6 (2017–2018, Karnataka) made their reappearance after a long span of more than 20, 19, and 9 years, respectively (Bhardwaj et al., 2003; Bhardwaj et al., 2006; Jain et al., 2013). The Pgt pathotypes in virulence group 21, earlier used to dominate northern India and parts of Nepal, were also detected from Gujarat, Madhya Pradesh, Karnataka, and Tamil Nadu ( Table 3 ). Likewise, a primitive Sr24 virulent pathotype 34-1, first identified from Madhya Pradesh in 1991 and prevalent on some local wheat landraces only in far north Leh Ladakh region afterwards, was identified again from Madhya Pradesh during 2021–2022. Such a shift in virulence pattern over time is not surprising, as the prevalence of some pathotypes in a particular crop season and their establishment in a new region depend on the diversity and resistance level of wheat varieties cultivated during that season (Singh, 1991), and to some extent on the prevailing environmental conditions such as temperature, rainfall, and relative humidity in particular (Roelfs et al., 1992). A similar shift in virulence pattern for stem rust of wheat has also been reported in many other countries including Canada, Mexico, the USA, Australia, and South Africa (Le Roux and Rijkenberg, 1987; Singh, 1991; Park and Wellings, 1992; Fetch, 2005; Jin, 2005; Terefe et al., 2016); for leaf (brown) rust from USA, Pakistan, India, and Egypt (Kolmer et al., 2017; Khadegah et al., 2019; Kolmer, 2019; Bhardwaj et al., 2021); and for stripe (yellow) rust from Canada, India, and other Asian countries (Hovmoller et al., 2011; Gangwar et al., 2022). There was a major shift in the frequency of predominant pathotypes of Pgt in India from 2016 to 2022. For instance, the frequency of pathotype 40A, which was the most predominant pathotype for almost three decades (Bhardwaj, 2012), reduced significantly after 2013–2014. Pathotype 11 remained most predominant since 2013–2014 throughout the current study period except during 2016–2017, when 40A surpassed pathotype 11 again ( Figure 5 ). Although both pathotypes 40A and 11 are thought to have more virulence and better fitness potential over others, the modifications in the environmental conditions and altered varietal spectrum could have better-supported pathotype 11 over 40A during these years. Similar assumptions were made to justify the dominance of one pathotype over another in the case of sunflower rust caused by Puccinia helianthi Schw. (Prud'homme and Sackston, 1990), wheat leaf rust (Bhardwaj et al., 2014), and stripe rust (Hovmoller et al., 2011). The optimum temperature, sufficient moisture, and cultivation of diverse wheat genotypes supported maximum virulence diversity in Pgt populations during 2018–2019 and 2020–2021 (Prasad et al., 2019; Prasad et al., 2021c). Eight pathotypes with different virulence structures were identified during these two crop seasons in Gujarat, Himachal Pradesh, Karnataka, Madhya Pradesh, Maharashtra, Tamil Nadu, and Nepal.

The phylogenetic analysis of 14 Pgt pathotypes, using 37 Puccinia sp.-specific polymorphic SSRs, clustered them in three molecular groups ( Figure 4 ), indicating the existence of very few lineages in the whole Pgt population. The heterozygosity and PIC values of these SSRs suggested that these markers had sturdy differentiating power as suggested by Shete et al. (2000) and Botstein et al. (1980). The molecular grouping of Pgt pathotypes, generated through unweighted neighbor-joining (NJ) cluster analysis in DARwin 6.0.14, corroborated, up to some extent, virulence-based grouping of these pathotypes, which was in agreement with our previous findings (Prasad et al., 2022). Pathotypes 21, 21-1, and 21A-2; 11 and 11A; 117 and 117-6; and 40A, 40-2, and 40-3, sharing almost matching avirulence/virulence structure as reported previously (Prasad et al., 2018; Prasad et al., 2022), were clustered in one MG/subgroup ( Figure 4 ). Such an association of virulence phenotypes and SSR genotypes among P. triticina (Wang et al., 2010) and P. striiformis tritici (Hovmoller et al., 2002) isolates is also reported, which is not uncommon if the genes responsible for virulence/pathogenicity are extensively distributed in pathogen genome (Ordonez and Kolmer, 2009). The comprehensive evolutionary pattern of all these Pgt pathotypes and others reported from the Indian sub-continent is already discussed previously (Prasad et al., 2022). Lower genetic diversity among Pgt pathotypes may be attributed to the clonal multiplication and long-distance migratory nature of the pathogen in the Indian sub-continent (Wingen et al., 2013).

Stem rust response of 40 wheat varieties revealed substantial resistance in these varieties against 23 pathotypes of Pgt ( Supplementary Table 2 ) at seedling and against 11, 21A-2, 40A, 117-6, 122, and other naturally occurring isolates at adult plant stages from five geographically different locations ( Supplementary Table 3 ; Table 5 ). At the seedling stage, 14 varieties exhibited resistant response against all 23 pathotypes, which included resistance conferred by Sr31 in 10 varieties and by Sr24 in 2 varieties. Despite the fact that 2 (34-1 and 40-1) of the 23 pathotypes are particularly virulent on Sr24 (Jain et al., 2013), these pathotypes were unable to establish a compatible relationship with two Sr24-containing wheat varieties (CG1029 and MP3288). Such interactions could result from the presence of some additional unknown effective Sr gene other than Sr24, which might impede pathotypes 34-1 and 40-1 from establishing compatible interactions with these varieties (Dakouri et al., 2013). The Indian population of Pgt possesses very low virulence frequencies on Sr31, Sr24, and Sr30 (Prasad et al., 2022), which are among 12 Sr genes postulated in 39 varieties ( Table 4 ), and therefore, these varieties could be targeted for skillful gene deployment. Generally, Sr2, Sr5, Sr6, Sr7a, Sr7b, Sr8a, Sr8b, Sr9b, Sr9e, Sr11, Sr12, Sr13, Sr17, Sr21, Sr24, Sr25, Sr30, and Sr31 are frequently occurring Sr genes in Indian wheat varieties. Among these, Sr7b, Sr9e, Sr11, and Sr13 are more common in durum wheat, while Sr2, Sr11, and Sr31 are mostly postulated in bread wheat (Jain et al., 2013). The frequency of Sr2, postulated in 67.5% of varieties based on micro flecking: an Sr2-linked morphological marker (Khan et al., 2017), was the highest. The Sr2, along with other slow rusting Sr genes such as Sr55, Sr56, Sr57, and Sr58, could be a valuable resource to have and utilize in combination with other effective major genes, i.e., Sr26, for achieving durable stem rust resistance (Singh et al., 2015).

The slow rusting/APR/field response of 40 wheat varieties and a susceptible check (Agra Local) were assessed based on the valuation of the final disease score (FDS), ACI, AUDPC, and apparent infection rate. The findings showed that the tested wheat varieties had more available resistance than Agra Local, which was severely infected during both the crop years at all five locations. Generally, the slow rusting behavior in wheat has been estimated and reported through assessment of disease severity, AUDPC, CI values, and apparent infection rates (Broers et al., 1996; Pathan and Park, 2006). The FDS is also used as a parameter to assess the slow rusting character of wheat genotypes by many earlier researchers (Ali et al., 2009); however, ACI, AUDPC, and apparent infection rate are considered most appropriate and preferred. Among 40 test varieties, 30 had ACI values ≤ 17.24% and 23 had AUDPC values ≤ 8.94% of the check ( Table 5 ), which specify high to moderate levels of slow rusting resistance character of these varieties (Ali et al., 2009). Furthermore, most of the varieties had MR/MS-type infection responses, which are the characteristics of the slow rusting genes, conferring partial field resistance (Parlevliet, 1988). The Wilcoxon signed-rank test categorized HI1655, HI8826, HI1654, HI8830, CG1036, HI1633, HI1650, VL907, HD2733, and HS507, with ≤3.72 ACI and ≤103.40 AUDPC scores, as the 10 best-performing varieties for stem rust APR. Among these, the Sr31 was postulated in HI1633, HI1650, VL907, HD2733, and HS507, based on their response to 23 Pgt pathotypes at the seedling stage and its linkage to Lr26/Yr9. Lower ACI and AUDPC values in these five varieties are implicit since Sr31 virulences are not reported from South Asian countries (Bhardwaj et al., 2019). Although Sr31 lost its effectiveness in East Africa and the Middle East due to Ug99 and other variants in its lineage, yet it is valued plus owing to its contribution to yield advantage and effectiveness to stem rust and other diseases in many countries including those in South Asia (Cox et al., 1995; Yu et al., 2014). Some weak stem rust resistance genes, i.e., Sr13+, Sr7b+2+, Sr7b+, and Sr2+, were postulated in the remaining (5) top 10 varieties HI1654, HI8826 (d) and HI8830 (d), CG1036, and HI1655, respectively. The genetic analysis of wheat germplasm using rust resistance gene linked molecular markers is another substitute to traditional gene postulation and could help in the swift and precise identification of resistance genes (Goutam et al., 2013). The molecular markers closely linked to rust resistance genes have frequently been identified and utilized in marker-assisted breeding for rust resistance (Mago et al., 2013). Besides being quick and unambiguous, this method could also circumvent the problems associated with traditional gene postulation such as plant stage-dependent gene expression, temperature, and other abiotic factor-dependent gene interactions. In the present study, genetic analysis of wheat varieties using molecular markers closely linked to Sr31 and Sr24 confirmed the gene postulation results in the majority of the cases except for Sr24 in the wheat variety CG1036 and Sr31 in CG1029. The gene postulation using the gene matching technique listed the presence of Sr7b and Sr24 in CG1036 and CG1029, respectively ( Table 4 ). None of the six Sr genes could be identified in five varieties (HD2733, PBW771, NIDW1149, HI8498, and KRL19) through genetic analysis, indicating the presence of some other known or unknown ASR/APR genes. All these markers have been extensively used to identify Sr genes or Yr and Lr genes linked to Sr genes. For instance, marker iag95 was effectively used to identify Sr31 in some Chinese wheat cultivars (Xu et al., 2017; Xu et al., 2018). Similarly, Sr38 linked to Yr17 and Lr37 was reported from wheat cultivars in China and Nebraska using VENTRIUP-LN2, an Sr38/Yr17/Lr37 linked marker (Peng et al., 2013; Xue et al., 2014; Xu et al., 2017; Mourad et al., 2019). Yang et al. (2008) and Alma et al. (2012) identified the Sr57 gene in some Chinese and Central Asian wheat cultivars, respectively, using the csLV34 marker. The WMC44 marker was applied for the molecular analysis of slow rusting resistance gene Sr58/Lr46/Yr29 in triticale cultivars (Skowrońska et al., 2020). Herrera-Foessel et al. (2014) used CFD71 marker to detect Sr55 in wheat varieties that conferred APR to stem rust and powdery mildew in wheat. The marker Sr24#50 was used to evaluate and identify Sr24 in wheat lines from Gansu province in China (Xu et al., 2017). This marker indicated the presence of Sr24 in CG1036 contrary to gene postulation, which indicated the presence of a relatively weak Sr7b in the same variety ( Tables 4 , 5 ). The susceptibility of CG1036 to several Pgt pathotypes, other than Sr24 virulences (34-1 and 40-1, Supplementary Table 3 ), also ruled out the presence of Sr24 in it (Prasad et al., 2018). The SSR marker Xgwm533 has been extensively used for the identification of Sr2/Yr30 in wheat worldwide (Mago et al., 2011; Khan et al., 2017; Rahmatov et al., 2019). We could identify Sr2 in 24 wheat varieties using this marker. The Sr2 was confirmed in 13 varieties using both gene postulation and marker analysis approaches. However, when looking at specific methods individually, it was identified in 11 varieties through marker analysis and in 14 varieties through gene postulation ( Tables 4 , 5 and Supplementary Figure 1 ). Similar observations were reported previously (Khan et al., 2017). The failure of the Xgwm533 marker to validate the micro- flecking-based postulation of Sr2 in some varieties could be justified by the fact that many of these varieties might lack Sr2 and micro-flecking could have occurred due to external environmental factors influencing host–pathogen interactions (Mago et al., 2011). Another probability could be that the variety with micro-flecking might have separated from Sr2 or that the marker was not diagnostic (Rahmatov et al., 2019). Similar observations were reported for Lr34/Yr18/Pm38 linked to more accurate molecular markers, where markers reported the presence of these genes even in susceptible wheats (Lagudah et al., 2009). The collective gene postulation and marker analysis results indicated that wheat varieties HI1655 and HI1654 might harbor some effective known/unknown Sr genes other than Sr2 to confer adequate APR. Therefore, these varieties could be appraised for durable stem rust resistance breeding in wheat. More importantly, further genetic analysis of these varieties could reveal some interesting loci that confer resistance to all Indian Pgt pathotypes. The locus contributing to stem rust APR, which includes the Sr2 gene, has been found to be effective and recommended for use in breeding programs (Rutkoski et al., 2014; Singh et al., 2015). The deployment of Sr2 is largely advised in combination with other effective Sr genes as it (alone) could not offer adequate protection under epidemic situations (Haile and Roder, 2013).

In conclusion, 14 Pgt pathotypes were identified as a result of surveys on the prevalence of wheat stem rust from 2016 to 2022 and pathotype analysis of the samples gathered during the surveys. Pathotypes 11 and 40A remained the most predominant from the previous reports, while pathotypes 42, 117, and 117-6 made their reappearance after more than 9 years. Remarkable stem rust resistance, conferred by some effective known ASR (Sr24, Sr31 and Sr38) and APR (Sr2, Sr57, and Sr58) genes, as revealed by gene postulation and genetic analysis using closely linked markers to these genes, was observed among most wheat varieties. The values of adult plant slow rusting parameters in these varieties were highly encouraging and could prove useful for stem rust management through their deployment or as durable stem rust resistance sources for future breeding program. These findings would be useful in deploying appropriate resistant varieties and streamlining wheat breeding approaches to combat stem rust. These conclusions provide good opportunity to explore some novel resistances where already known slow resisting loci are not identified. In view of the dynamic nature, virulence diversity, and long-distance migration of Pgt isolates, their monitoring the shifting virulence pattern, searching for new sources of resistance, and their transfer into high-yielding cultivars must be sustained to mitigate the impact of stem rust menace on wheat.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

PP and SB conceived the experiment. PP, RT, SS, OG, CL, SA, SuK, SoK, and AM conducted the lab and greenhouse experiments. TP, SN, GH, BG, and KM conducted field experiments. PP, HK, VG, CM, and StK conducted statistical analysis. PP wrote the first draft of the article. SB, SudK, and GS edited and improved the article. All authors read the final version of the manuscript and approved it.