Ne1 induced the necrotic phenotype and yield decrease
In the RIL population of ZN17/YBM, 24 out of 188 RILs showed different degrees of necrosis (Supplementary Fig. S1). Most of the necrotic lines exhibited necrosis at the leaf tips during the seeding stage, preceded gradually to the middle and bottom of the leaves at jointing and flowering stage, and dried at the late grain filling stage (Supplementary Fig. S1a-c), while the two parents ZN17 and YBM grew normally. At the grain filling stage, six lines showed severe necrosis that resulted in premature death of the leaves (Supplementary Fig. S1d) and shriveled grains; three lines exhibited weak necrosis at only flag leaf tips (Supplementary Fig. S1e); two lines had yellow leaves (Supplementary Fig. S1f).
The necrotic phenotype in the NIL-Ne1YBM firstly appeared on the lower leaves at the tillering stage (Fig. 1a) and became more evident during the heading period in the field. A more detailed observation indicated that the leaf necrosis initiated from the bottom leaves and progressed gradually to the upper leaves, and advanced from the tip to base of the leaf (Fig. 1b). Necrosis formation in NIL-Ne1YBM leaves appeared to follow a developmental pattern, in which the severity of necrosis is correlated positively with the age of leaves. At the grain filling stages, only the flag leaf or upper leaves remained partially green, and the rest of leaves were necrotic or dried in the NIL-Ne1YBM, while at the same stage the leaves of NIL-ne1ZN17 remained green and healthy (Fig. 1c). In addition, the NIL-Ne1YBM displayed smaller leaves and fewer tillers, with a shorter plant height (Fig. 1a, c) compared to those of the NIL-ne1ZN17.
Yield-related agronomic traits of the NIL-Ne1YBM and NIL-neZN17 lines were evaluated after harvest. Compared to the normal NIL-ne1ZN17, the NIL-Ne1YBM showed fewer tillers (Fig. 2a), shorter plant height (Fig. 2b), smaller (Fig. 2c) and thinner grain size (Fig. 2d). The thousand-kernel weight of the NIL-Ne1YBM (34.62 g) was 33.13% smaller than that of the NIL-ne1ZN17 (51.77 g; Fig. 2e).
Genetic mapping of Ne1
The two parents and 188 RILs of ZN17/YBM population were genotyped with the wheat660K SNP array. After removing the SNPs with > 5 percentage data missing values and multiple mapping sites on RefSeq v1.0 (Keeble-Gagnere et al. 2018), 146,351 high-quality polymorphic SNPs were obtained. The associations of SNPs and the necrotic phenotype were conducted with the function of GLM implemented using software TASSEL Version 5.0. There were 1,844 SNPs significantly associated with the phenotype at the threshold of P-value (3.42 × 10− 7). Among these SNPs, 916 (49.67%) and 870 (47.18%) SNPs were mapped on chromosome 2BS and 5BL, respectively, and the rest 58 SNPs (3.15%) were distributed on 2A, 3A, 5A, 2D, 4D and 5D chromosome (Fig. 3a). These consisted with previous reports that hybrid necrosis was controlled by Ne1 on 5B and Ne2 on 2B in common wheat (Chu et al. 2006; Nishikawa et al. 1974; Zeven 1972). The significant SNPs on 5BL were in ~ 297 Mb to 399 Mb and those on 2BS located in ~ 110 Mb to 180 Mb (Fig. 3b). These results suggested that Ne1 and Ne2 were located to the regions of ~ 297 Mb − 399 Mb on 5BL and ~ 110 Mb − 180 Mb on 2BS, respectively.
Genetic analysis and fine mapping of Ne1 with HIFs
To test the genetics of the Ne1 locus on 5B, 252 individuals were randomly selected from HIFs-5B population (HIF2), which segregated at the Ne1 locus. Among them, 53 individuals showed severe necrosis, 134 plants were characterized as moderate necrosis, and 65 plants displayed a normal phenotype at the grain filling stage (Supplementary Fig. S2). The segregation ratio of severe necrosis: moderate necrosis: normal fitted the expected ratio of 1:2:1 (χ2 = 2.16, P < 0.05). These results suggested that the necrotic phenotype of HIFs-5B population was controlled by a semi-dominant nucleus gene.
Based on the IWGSC RefSeq v1.0 and the flanking sequence of the associated SNPs, 50 molecular markers were developed in the Ne1 region (~ 297 Mb − 399 Mb) on chromosome 5BL. Among them, four markers were found to be polymorphic between ZN17 and YBM. The four polymorphic markers were applied to assess the genotype of 682 randomly selected individuals from HIFs-5B population. With these four markers, Ne1 was delimited into the region flanked by markers 5B-378 and 5B-388. Then, the rest 3402 individuals of HIFs-5B population were analyzed with these two markers for screening additional recombinants. A total of 23 recombinants were identified between the markers 5B-378 and 5B-388. Subsequently, we developed new polymorphic molecular markers (InDel and SNP markers) and genotyped the recombinant plants to further narrow the candidate interval of Ne1. Eventually, Ne1 was mapped to a 4.06-Mb physical interval delimited by markers 5B-383 and SN-2142 (Fig. 3c, d).
The ne1 allele is associated with a 2.89-Mb fragment deletion on chromosome 5BL in ZN17
In order to clone Ne1 gene, we designed 57 SSR markers in the range of ~ 383 Mb to 387 Mb between the markers 5B-383 and SN-2142 on 5BL based on IWGSC RefSeq v1.0, and found most SSR markers were of presence/absence between ZN17 and YBM, but not a product size differences. We speculate that this phenomenon should be caused by missing a large segment in this interval in ZN17. In order to confirm this speculation, we re-sequenced ZN17 and YBM for 10X sequencing depths. Sequence reads were aligned to the reference genome of Chinese Spring (RefSeq v1.0), and SNPs and InDels located in the region between the markers 5B-383 and SN-2142 were called by the HaplotypeCaller module (Chai et al. 2018). We found continuously missing sequences from ~ 383,441,497 to 386,325,646 (RefSeq v1.0) in ZN17, but not in YBM (Supplementary Table S3). This suggested that there is an approximate 2.89-Mb segment deletion between the markers 5B-383 and SN-2142 in ZN17. To verify the missing segment in ZN17, we designed primers to amplify six high confident genes located on the missing segment of chromosome 5BL in the NILs and the two parents ZN17 and YBM (Supplementary Table S2). All six genes were detected in YBM and NIL-Ne1YBM, but not in ZN17 and NIL-ne1ZN17. These results further confirmed the deletion of the Ne1 gene candidate region in ZN17. Therefore, it is not possible to clone Ne1 with the population developed from the cross between ZN17 and YBM.
Development, validation and application of the diagnostic marker for Ne1
To confirm that the missing of the 2.89-Mb fragment on 5BL is associated with Ne1 gene, we screened a new HIF (HIF3) from a RIL (F8 generation) population from the cross between ‘Zhengzhou 6903’ and ‘Yumai 14’. The HIF3, as HIF1 and HIF2, segregated in necrotic phenotype in the field, and showed the same PCR product pattern as in HIF2 with the molecular marker 5B-378. We also found that all six genes located in the 2.89-Mb fragment could be identified in Yumai 14 (carrying Ne1), but not in Zhengzhou 6903 (carrying ne1). This indicated that the large fragment deletion in ne1 varieties might not be rare. Therefore, we picked 5B-InDel385 as the diagnosis marker for Ne1 gene among the six markers. 5B-InDel385 divided wheat into two major haplotypes (H1 and H2) based on the presence/absence of the 2.89-Mb fragment on 5BL. The H1 haplotype was the necrotic-Ne1 allele carrying the 2.89-Mb fragment; oppositely, H2 haplotype possesses the normal-ne1 allele with the 2.89-Mb fragment deletion.
To test the efficiency of the diagnostic marker 5B-InDel385, 1034 individuals of HIF3-5B population (derived from HIF3) were genotyped and analyzed. As expected, all normal plants were H2 haplotype and all necrotic individuals were H1 haplotype. Collectively, this data revealed that Ne1 was co-segregated with the marker 5B-InDel385 on chromosome 5BL in wheat.
The 5B-InDel385 was also validated in some cultivars - carriers of known alleles of the gene Ne1 or Ne2/Lr13: including Kubanka (T. durum, Ne1s), Chinese Spring (Ne1w, ne2), Vakka (Ne2w), Blackhull (Ne2s), Sonalika (Ne2m), Manitou (Lr13), and Frontana (Lr13). Notably, all the above-mentioned varieties grow normally. The marker 5B-InDel385 assay showed that all of the cultivars carrying Ne2/Lr13 belonged to H2 type containing ne1 allele. The other two cultivars, Kubanka and Chinese Spring known to possess Ne1 allele were of the H1 type. Based on these results, we concluded that the marker 5B-InDel385 can accurately distinguish Ne1 from ne1.
Using the diagnostic marker, we firstly characterized 259 wheat accessions (29 landraces and 230 cultivars) from China. As shown in Fig. 4a, the Ne1 allele was frequent in landraces (62%). Conversely, the frequency of Ne1 decreased sharply in the modern cultivated wheat varieties (32%) in China (Fig. 4b). We studied the geographical distribution of Ne1 in China and found the frequency of Ne1 was highest in Henan (41.51%), followed by Shaanxi (41.38%), Shanxi (33.33%), Beijing (31.25%), Hebei (28.57%), Jiangsu (29.41%), and Shandong (5.13%; Table 1). Furthermore, 242 common wheat landraces/cultivars from diverse origin were detected with 5B-InDel385. The results showed that Ne1 and ne1 were widely distributed throughout the world. The Ne1 allele was present in 104 of 171 landraces (61%) from USDA collections (from South, West and Central Asia; Fig. 4c). On the contrary, in a subset of 71 common wheat cultivars from USA, the Ne1 allele was just observed in 7% of the accessions (Fig. 4d). In total, 122 landraces (61%) showed the presence of Ne1 (Fig. 4e), whereas only 79 modern cultivars (26%) carried Ne1 (Fig. 4f). This suggested that Ne1 allele was subjected to a high selected pressure in wheat breeding.
Table 1
Distribution of Ne1 in wheat cultivars from China
Agroecological region
|
NO. of accessions
|
5B-InDel385
|
H1 (Ne1)
|
H2 (ne1)
|
NO. of accessions
|
Percentage of the allele
|
NO. of accessions
|
Percentage of the allele
|
Henan
|
53
|
22
|
41.51%
|
31
|
58.49%
|
Shaanxi
|
29
|
12
|
41.38%
|
17
|
58.62%
|
Shanxi
|
15
|
5
|
33.33%
|
10
|
66.67%
|
Beijing
|
16
|
5
|
31.25%
|
11
|
68.75%
|
Hebei
|
42
|
12
|
28.57%
|
30
|
71.43%
|
Jiangsu
|
18
|
5
|
27.78%
|
13
|
72.22%
|
Shandong
|
39
|
2
|
5.13%
|
37
|
94.87%
|
Sichuana
|
11
|
8
|
-
|
3
|
-
|
Anhuia
|
3
|
2
|
-
|
1
|
-
|
Zhejianga
|
2
|
1
|
-
|
1
|
-
|
Yunnana
|
2
|
0
|
-
|
2
|
-
|
Total
|
230
|
74
|
32.17%
|
156
|
67.83%
|
a Wheat accessions from these regions were not included for the Ne1 frequency analsysis because of limited entries in this study. |
Putative Ne1 candidate genes
To identify the candidate genes for Ne1, we analyzed the predicted genes on the 5BL from 383.03 Mb to 387.10 Mb of the Chinese Spring RefSeq v.1.0 sequence. Based on the RefSeq v1.1 annotation, 54 genes were identified in this region. There are 28 genes in the 2.89-Mb missing fragment, 6 genes are in the region between the marker 5B-383 and the deletion region, and 20 genes between the marker SN-2142 and the deletion region (Supplementary Table S4). The re-sequencing of ZN17 and YBM showed that the major difference in the Ne1 candidate region between the two parents was the 2.89-Mb deletion and there was no amino acid difference in the rest 26 genes outside the deletion bin though there were 385 SNPs/InDels in the candidate interval (Supplementary Table S5).
To predict the Ne1 gene, we analyzed the expression profiles of the genes in the candidate region of Chinese Spring using the wheat expVIP expression platform (http://www.wheat-expression.com/). It is known that the necrotic phenotype is caused by the interaction of Ne1 and Ne2, and starts from seedlings, and ‘Chinese Spring’ carries the Ne1w allele (Hermsen 1963a; Zhang et al. 2016). Therefore, we hypothesize that Ne1 should express in leaves and shoots at its whole growth period. We found that 18 of 54 candidate genes expressed (above two transcripts per million) in at least three RNA-seq samples of Chinese Spring (leaves/shoots and roots, n = 40) at different developmental stages (Supplementary Table S6). Among them, 9 of 18 genes are high-confidence genes, including two auxin-responsive SAUR protein, two serine protease HtrA-like, two alpha/beta-Hydrolases superfamily protein, one serine protease HTRA3, one initiation factor 4F subunit (DUF1350), one RING/U-box superfamily protein, one trypsin-like serine protease (Supplementary Table S7). We speculated that one of the 9 genes might be the Ne1 gene.