Close this search box.

Meta-QTL analysis and identification of candidate genes for multiple-traits associated with spot blotch resistance in bread wheat – Scientific Reports

In recent years, molecular breeding involving the use of DNA markers for MAS has become routine in conventional plant breeding. This is particularly true, when improvement of quantitative traits is involved. However, for MAS, robust markers that are suitable to be used for MAS are not always available. In such studies, MQTLs derived from QTLs identified in earlier studies may sometimes prove useful. For spot blotch and related traits, the present meta-QTL study is the first of its kind. The study utilized all known QTLs not only for spot blotch resistance but also those associated with five other related traits, which often accompany spot blotch resistance (for identity of five traits, see above).

As reported in results, in the present study, using 228 QTLs that were available from published literature, only 84.0 (36.8%) could be projected onto the consensus map, because only these had the complete information needed for MQTL analysis. This gave 31 MQTLs, of which 15 MQTLs were based each on a single QTL and were therefore rejected, thus leaving only 16 MQTLs derived from 69 QTLs. The remaining 144 QTLs could not be projected due to one of the following reasons: (i) there were no flanking markers that the original and consensus maps shared, and (ii) the CIs were relatively large19. Notably, SG trait was the only trait besides spot blotch resistance that was directly targeted in QTL mapping within the context of spot blotch23. In previous studies24,25,26,27,28, the remaining traits (except BPR), namely GLAD, GLNS, FLS, were generally also subjected to mapping of tolerance against abiotic stress factors in wheat. A 14.2 -fold reduction in the number of genomic areas or QTLs linked to spot blotch resistance in wheat was achieved through the identification of 16 MQTLs from 228 QTLs. This projection success percentage is less than that of previous studies on MQTL analysis conducted for different wheat disease resistances, where the following projection rates were available: 44.0, 60.62, 66.6, and 75.2%15,16,19. The present study also adds to the examples of MT-MQTLs involving disease resistance, because 12 of the 16 MQTLs were MT-MQTLs. Earlier MQTL studies in wheat have been carried out for a variety of traits, including tolerance against abiotic and biotic stresses. For instance, Liu et al.14 identified 20 MQTLs for tan spot using 106 QTLs, Amo et al.17 reported 35 MQTLs for leaf rust using 128 QTLs, and Jan et al.15 discovered 61 MQTLs for stripe rust using 184 QTLs. MQTLs for multiple disease resistance (MDR) were also identified in two of our own studies (see later for some details).

Although our focus in the present study was primarily on SBR, a five other traits known to be associated with SBR were also included. Among these associated traits,, SG trait was of special interest, since it has the unique characteristic of delaying senescence in leaves and other plant parts, enabling them to maintain high chlorophyll levels and sustain photosynthetic activity for an extended period. The association of SG with spot blotch has also been reported in wheat, where a significant negative correlation (0.73) between SG and AUDPC (a measure of spot blotch severity) has been reported23,37. In a number of studies, SG has often been used as a selection criterion for tolerance against abiotic stresses also, including heat and drought16,24,25,26,28,38. Thus, this adaptive trait (SG) is also particularly useful in adverse environments, since it facilitates an improved grain-filling process27. The genotypes that are characteristics of SG and SBR also exhibit tolerance/resistance against heat/drought, suggesting a shared physiological mechanism for response to abiotic and biotic stresses including SBR39.

Since six different traits were used in the present study, it was also possible to identify MT-MQTL, which can be used for improvement of more than one traits using the same MQTL (see later in Discussion). In the published literature, only two other studies from our own work were available, which involved identification of MT-MQTLs, both involving resistance against a number of diseases. In one study, Saini et al.19 reported MQTLs for MDR involving five unrelated diseases (STB, SNB, FHB, KB, and LS), while Pal et al.20 reported MDR MQTLs for only the three rusts, including leaf rust, stem rust, and stripe rust. In the present study, out of 16 MQTLs, 12 involved more than one trait and were therefore described as MT-MQTL (involving 2–4 traits); no MT-MQTL was available for 5 or 6 traits (Table 1). Among these 12 MT-MQTLs, on the basis of priority, three MT-MQTL are being recommended for MAS involving SBR with other associated traits (see later).

The merit of MQTLs relative to the corresponding QTLs used for MQTL analysis, depends on four key attributes, including PVE, LOD score. CI and the number of QTLs involved in each individual MQTL. Following are some details of comparison of these four attributes in MQTL and QTLs used for MQTL analysis, suggesting that MQTL showed an improvement over QTLs: (i) The PVE of the MQTLs ranged from 1.6 to 55.2% (with a mean of 13.6%), while that of the QTLs ranged from 5 to 10% in 72 QTLs, followed by 10–15% in 60 QTLs, and 15 to 20% in 39 QTLs. This suggested that PVE% improved in MQTLs. (ii) The LOD scores of MQTLs ranged from 3.0 to 11.0; in contrast, in QTLs, the LOD score ranged from 2 to 4 in 83 QTLs, from 4 to 6 in 75 QTLs, and 6–8 in 48 QTLs, suggesting that the LOD score also improved in MQTL. (iii) The CI in MQTLs largely ranged from 1.5 to 20.7 cM (except five MQTLs, where it ranged from 36.1 to 57.8 cM) as against 0.62–272 cM for the QTL used for MQTL analysis, suggesting a major reduction in the length of the CI, thus making these MQTL relatively more robust.(iv) Similarly, 8 MQTLs were each based on two QTLs, and MQTL3 was based on 17 QTLs, suggesting that the number of QTLs associated with individual MQTLs is large (at least in some MQTLs), thus improving the credibility of the utility of MQTLs (Table 1).

The results of the present MQTL study on SB in wheat can also be compared with those available for several other wheat diseases, including three rusts9,15,16,18, fusarium head blight (FHB)10 and tan spot14. Following are the some details about the MQTLs for disease resistance identified in other studies: (i) The PVE ranged from 4.7 to 51.0% for three rusts, 6.2–22.3% for FHB, and 6.3–27.0% for tan spot14. (ii) The LOD score for MQTL ranged from 3.0 to 28.0 for three rusts20, from 1.0 to 62.0 for FHB12, and from 3.9 to 17.0 for tan spot14. (iii) The CI in these earlier studies ranged from 0.04 to 83.5 cM for three rusts20, < 10.0–95.0 cM for FHB10 and 0.9–8.1 cM for tan spot14. For MT-MQTLs for MDR in earlier studies, the PVE ranged from 2.2% to 51.2%, the LOD score ranged from 2.9 to 48.4 and the CI ranged from 0.04 to 15.2 cM19.

For validation, a study of colocalization of MQTLs identified in the present study with MTAs identified in published literature was also undertaken. For this purpose, MTAs were available for only SBR, SG, and BPR; for other traits, no GWAS were available. For these three traits, 13 (> 81.0) of the 16 MQTs were colocalized with GWAS-MTAs (Supplementary Table S4. These results can be compared with similar validation studies undertaken earlier. In these earlier studies, co-localization ranged from 38.7 to 90.5%17,20,40,41,42,43,44. The high rate of colocalization suggests the presence of causal polymorphisms in MQTLs identified in the present study.

Another component of the present study was the identification of CGs associated with MQTLs. The distribution of CGs ranged from none in MQTL7 to a maximum of 157 in MQTL2, with an average of ~ 33 CGs per MQTL (Supplementary Table S5). This data suggests a potential role for these CGs in spot blotch resistance, warranting further investigation. The R domains in proteins encoded by CGs are known to be involved in plant defense mechanisms, strengthening the hypothesis of their association with disease resistance. In wheat, similar studies on CGs associated with MQTLs were earlier conducted for traits like fusarium head blight12, stripe rust tolerance15,16, leaf rust17, tan spot resistance14, MDR19,20.

Among a large number of CGs identified to be associated with MQTLs in the present study, 71 DECGs were available, which encode proteins carrying domains with known roles in plant defense. Some of these genes were downregulated (FC < − 2.0), while others were upregulated, when a resistant genotype was compared with a susceptible genotype at two different durations after inoculation (R24 vs. S24 and R48 vs. S48). The protein domains involved in downregulation included the following: nucleic acid-binding domain, RING-type zinc-finger domain, and palmitoyltransferase domain. Similarly, the upregulated genes encoded proteins carrying the following domains: AOX protein family, AOX protein family, NBS-LRR and VQ protein. These results are also in agreement with the results of several earlier studies conducted either in Arabidopsis or in wheat45,46,47,48,49,50,51,52, suggesting that these DECGs deserve further detailed study.

In the present study, we also validated 13 (> 81.0%) of the 16 MQTLs using the previously conducted GWAS. This also allowed selection of the following for MQTLs based on high PVE, LOD value, and lower CIs: MQTL3, MQTL4, MQTL5, and MQTL8. Genomic selection models can also utilize these markers to enhance the accuracy of resistance prediction. On the basis of present study, we also recommend further research to clone and functionally characterize the identified candidate genes (CGs). These genes have the potential to be used to improve wheat resistance but require validation through techniques like gene cloning, reverse genetics, or omics approaches.

In the present study, among the four R genes (Sb1 to Sb4) known for spot blotch, Sb1 is the only Sb gene, which colocalize with one of the 16 MQTLs, namely MQTL16, suggesting that the list of MQTLs identified is certainly not exhaustive and exclusive and that there must be more SB MQTLs (associated with Sb2, Sb3 and Sb4) to be identified in future.

Among the 12 MT-MQTLs also, three MT-MQTLs (MQTL3, MQTL5 and MQTL8) are recommended for MAS on the basis of criteria outlined above for selection of MQTLs. MQTL16 carrying the R gene Sb1 is also recommended for MAS. Only future QTL and MQTL studies will allow identification of MQTLs associated with the remaining R genes, namely Sb1, Sb3 and Sb4, Some of the DECGs identified during the present study may also be utilized for further studies.