Genomic compatibility and inheritance of hexaploid‐derived Fusarium head blight resistance genes in durum wheat

Hexaploid‐derived resistance genes exhibit complex inheritance and expression patterns in tetraploid backgrounds. This study aimed to characterize the inheritance patterns and genomic compatibilities of hexaploid‐derived Fusarium head blight (FHB) resistance genes in tetraploid durum wheat (Triticum durum Desf.). Evaluation of FHB resistance for F1 hybrids of hexaploid ‘Sumai 3’ crossed with tetraploid and hexaploid wheats indicated that Sumai 3‐derived FHB resistance genes exhibit a dominant phenotypic effect seen only in hexaploid hybrids. Alternately, the hexaploid‐derived FHB resistance genes from PI 277012 exhibited complete dominance in the crosses with both tetraploid and hexaploid wheat. FHB evaluation of the F1 hybrids of Sumai 3 and PI 277012 crossed with ‘Langdon’ (LDN)–‘Chinese Spring’ D‐genome substitution lines suggested that chromosomes 2B, 3B, 4B, 5B, 6B, 3A, 4A, 6A, and 7A contain genes that suppress expression of the Sumai 3‐derived FHB resistance, whereas chromosomes 4A, 6A, and 6B contain genes required for expression of PI 277012‐derived FHB resistance. A wide range of segregation for FHB severity (10–90%) was observed in the F2 generation of Sumai 3 crossed with durum cultivars LDN and ‘Divide’, but the distribution of F3 families derived from the most resistant F2 segregants was skewed towards susceptibility. Similar segregation trends were observed in the crosses of PI 277012 with other durum wheats, whereby FHB resistance became slightly diluted over successive generations. These results suggest tetraploid durum wheat contains the unique alleles at multiple gene loci on different chromosomes that positively and/or negatively regulate the expression of hexaploid‐derived FHB resistance genes, which complicate efforts to deploy these genes in durum breeding programs.


INTRODUCTION
Fusarium head blight (FHB), also called scab, is one of the most devastating diseases in wheat (Triticum aestivum L.) worldwide. It is mainly caused by the fungus Fusarium graminearum Schwabe in North America. The major epidemic regions of FHB in the United States are in the midwestern states comprising the Great Plains and in some eastern states (Stack, 1999). In the Great Plains, economic losses due to FHB from 1993 to 2001 were estimated at $2.5 billion dollars in wheat and barley (Hordeum L.) (Buerstmayr et al., 2020;Nganje et al., 2004). Globally, wheat yield losses due to FHB have been estimated as high as 21.5%, which was second only to leaf rust (Savary et al., 2019). Host plant resistance to FHB is considered the most efficient approach to reduce yield and quality losses compared to other management tactics, such as rotation and chemical control (McMullen et al., 1997;Rudd et al., 2001).
To date, most of the commonly used sources of FHB resistance are derived from hexaploid wheat. 'Sumai 3′, a Chinese hexaploid wheat cultivar, has been a widely used source of resistance to FHB in wheat breeding programs worldwide. Previous studies have implicated chromosomes 2B, 3B, 6B, and 7A harbored quantitative trait loci (QTL) and FHB resistance genes in Sumai 3 (Yao et al., 1997;Waldron et al., 1999;Zhou et al., 2002;Liu & Anderson, 2003). Among these genes, the major gene on 3B, designated Qfhs.ndsu-3BS (Fhb1), has explained as much as 60.0% of the phenotypic variation for FHB resistance (Buerstmayr et al., 2002(Buerstmayr et al., , 2009Jayatilake et al., 2011). In addition, many other genes and QTLs have been associated in contributing both major and minor resistance to FHB; for example, between 2009-2019 alone, 64 different QTL mapping studies were conducted in wheat using linkage and association mapping procedures. QTL for FHB resistance have been identified on all wheat chromosomes, using differing inoculation and phenotyping procedures to assess Type I and Type II FHB resistance, FHB severity, FHB incidence, FHB index, deoxynivalenol accumulation and Fusarium-damaged kernels (Bai et al., 1989;Van Ginkel et al., 1996;Buerstmayr et al., 2020;Ma et al., 2020;

Core Ideas
• Durum contains genes that enhance or suppress hexaploid-derived FHB resistance. • The hexaploid-derived FHB resistance genes were inherited differently in durum. • Genomic compatibility for FHB resistance was assessed by D genome substitutions. • Sumai 3-derived FHB resistance was suppressed by loci on nine durum chromosomes. • Three durum chromosomes contain loci needed for PI 277012-derived FHB resistance. Zhu et al., 2020). Both Type II (resistance to the spread of the pathogen) and Type V (low accumulation of mycotoxins) resistance have been identified in Sumai 3 (Jayatilake et al., 2011). Preliminary studies seeking to clone the gene underlying resistance mechanisms of Fhb1 were carried out, and a diagnostic molecular marker UMN10 was developed by Liu et al. (2008). More recently, Su et al. (2019) and Li et al. (2019) reported that a mutation of a histidine-rich calciumbinding protein gene (named as TaHRC) at the Fhb1 locus confers resistance against FHB. The introgression of FHB resistance from Sumai 3 has been successfully used to develop many hexaploid wheat germplasm and cultivars worldwide (He et al., 2001;Mergoum et al., 2006Mergoum et al., , 2008Teresa et al., 2013;Waldron et al., 1999). In the 1980s, two hexaploid cultivars 'Een 1′ and 'Yangmai 4′ with moderate resistance to FHB were first developed from Sumai 3 and released in China (He et al., 2001). In the United States, the first FHB resistant cultivar, 'Alsen', derived from Sumai 3, was released in the late 1990s . After that, many more FHB-resistant cultivars, including 'Faller', 'Glenn', and 'Freyr' were developed from Sumai 3 (Mergoum et al., , 2008. Recently, another hexaploid spring wheat accession with high level of FHB resistance 'PI 277012′ was identified with a level of FHB resistance deemed equivalent to Sumai 3 (Chu et al., 2011). Molecular mapping research using doubled haploids derived from PI 277012 identified two FHB resistance QTL positioned on the short arm and long arm of chromosome 5A, respectively (Chu et al., 2011). Moreover, it appeared that the PI 277012-derived FHB resistance genes could normally express when placed in a tetraploid wheat genetic background and hence could be very useful in tetraploid wheat breeding programs (Chu et al., 2011).
To identify novel FHB resistance sources in tetraploid wheat, extensive screening of wild relatives of tetraploid wheat has been carried out (Buerstmayr et al., 2012;Cai et al., 2005;Oliver et al., 2007Oliver et al., , 2008Ruan et al., 2012). For example, several FHB resistance QTL have been identified in wild emmer (Triticum dicoccoides) wheat accessions (Otto et al., 2002;Stack & Faris, 2006;Kumar et al., 2007). However, most of those identified QTL confer only moderate to low levels of resistance to FHB (Fakhfakh et al., 2011). Of these QTL, Qfhs-ndsu-3AS located on the short arm of chromosome 3A has been more extensively characterized than others (Chen et al., 2007;Otto et al., 2002;Zhu et al., 2015). The poor understanding of the chromosomal region where this QTL is positioned and the associated linkage drag of undesirable genes present in tetraploid wild relative introgressions has made it difficult to utilize these FHB resistance QTL directly in tetraploid wheat breeding programs.
One of the strategies to improve FHB resistance in tetraploid wheat is to transfer hexaploid-derived FHB resistance genes to tetraploid wheat. However, progress using FHB resistance derived from Sumai 3 in tetraploid wheat breeding programs has been very limited. The possible mechanisms hindering the progress have been ascribed to (a) the difficulty in recovering the desired configuration of gene combinations for the successful introgression of Sumai 3 resistance (Liu & Anderson, 2003;Basnet et al., 2012); (b) complications arising from the absence of wheat D genome in tetraploid wheat, which possibly affects expression of FHB resistance genes upon introgression in a tetraploid genetic background (Rudd et al., 2001;Fakhfakh et al., 2011;Zhu et al., 2016); and, conversely, (c) suppressor genes found in tetraploid wheat's A or B genomes that may inhibit expression of Sumai 3-derived FHB resistance (Rudd et al., 2001;Gilbert et al., 2002). For example, researchers were unable to utilize an introgression of tetraploid-derived leaf and stem rust resistance genes when placed in a hexaploid genetic background due to the discovery of suppressor genes found on chromosomes 1D, 2D, 3D, and 4D that compromised the expression of the resistance (R) genes due to complementary gene action (Kerber & Green, 1980;Bai & Kbitt, 1992). Genomic incompatibilities provide yet another obstacle that breeders must overcome to successfully deploy newly discovered R genes via recombination, chromosomal translocations, or alien introgressions between tetraploid and hexaploid wheats. Hence, dissection of the genetic factors influencing expression of hexaploid-derived FHB resistance genes in tetraploid wheat may facilitate improved utilization of hexaploid R genes in tetraploid breeding programs.
The objective of the present study is to investigate the effects and compatibility of durum's tetraploid genomic background on the expression and inheritance of hexaploid-derived FHB resistance genes from Sumai 3 and PI 277012.

Plant genetic materials and inheritance analysis
Four FHB-susceptible tetraploid wheat cultivars-'Langdon' (LDN), 'Grenora', 'Alkabo', and 'Divide'-were used as female parents in crosses made with Sumai 3 and PI 277012, two hexaploid wheat accessions highly resistant to FHB. The F 1 hybrids along with their parents were evaluated for FHB severity in four greenhouse seasons (Fall 2011, Fall 2012, Spring 2013, and Spring 2014. Four hexaploid wheat accessions highly susceptible to FHB ('2398′, 'Choteau', 'AC Vista', and 'AC Lillian') were also used as female parents to develop F 1 hybrids with Sumai 3, which were evaluated for FHB resistance in two greenhouse seasons (Spring 2013 andFall 2013).
In addition, a complete set of LDN-'Chinese Spring' (CS) D-genome substitution lines, developed by the USDA-ARS Cereal Crops Research Unit, Fargo, ND, were crossed with Sumai 3 and PI 277012. In each substitution line, a pair of homologous A-or B-genome chromosomes in LDN was substituted by a pair of its corresponding homoeologous Dgenome chromosomes from CS. The F 1 hybrids of these crosses were evaluated for FHB resistance in either three (Spring 2013, Fall 2013, and Spring 2014 or two greenhouse seasons (Spring 2013 andFall 2013).
F 1 hybrids in all crosses were confirmed by visual inspection of seed morphology and molecular marker analysis. Some of the F 1 hybrids were further verified by spike morphological comparisons. True F 1 hybrids were employed for the evaluation of FHB resistance and advanced to the subsequent generations for inheritance studies.
To study inheritance patterns, evaluation of FHB resistance in segregating progeny was performed from the F 2 to F 4 generations for the crosses of LDN × Sumai 3, Divide × Sumai 3, and LDN × PI 277012. One spike of each plant was selfpollinated to derive the next generation. The advanced F 5 families were formed by the bulked seeds of resistant segregants identified in the previous generations.

Molecular marker analysis
Polymerase chain reaction (PCR) amplification was carried out in a 20-μl mixture containing 40 ng genomic DNA, 0.5 μM each of forward and reverse primers, 1× PCR buffer, 1.5 mM MgCl 2 , 0.25 mM dNTP, and 0.25U of Taq DNA polymerase. Polymerase chain reaction was performed as the following protocol: 94˚C for 3 min; 45 cycles of 94˚C for 1 min, 55˚C for 1 min, and 72˚C for 1.5 min; then with a final 72˚C for 7 min. PCR products were separated on 8% polyacrylamide gel and visualized by ethidium bromide staining (Chen et al., 2007).

Experimental design and FHB evaluation
All plants were grown in 6-in plastic pots with one plant per pot for each F 1 hybrid, two plants per pot for parental controls, and each LDN-CS D-genome substitution line. Five to ten spikes in each pot were inoculated using the single floret inoculation method, described in detail below. The pots were randomly arranged on the greenhouse benches. Each pot was regarded as one replicate. The number of replicates for each entry in each experiment ranged from two to five depending on the successful germination of the F 1 seeds. A completely randomized design (CRD) with unbalanced data was used for statistical analysis using SAS version 9.3 (SAS Institute, 2011). The data obtained in different greenhouse environments were pooled for combined analysis if the Bartlett's homogeneity test of error variance was not indicative of significant difference (P = .05).
The greenhouse temperature was maintained at approximately 16 and 18˚C at night and daytime, respectively, with a 16-h photoperiod in the greenhouse before anthesis. During the inoculation period, the temperature was increased to approximately 25˚C. The single-floret inoculation method was used to infect plants with the inoculum as described by Stack et al. (2002). For inoculation, a spore suspension at a concentration of 1 × 10 5 conidiospores per ml was prepared as inoculum with a mixture of equal number of spores produced by four F. graminearum isolates (two 3ADON type isolates and two 15ADON type isolates), collected from North Dakota (Puri & Zhong, 2010), and ∼10 μl of the inoculum was injected into a central floret of each spike during anthesis. The inoculated spikes were covered with plastic bags, which were water-misted on the inside to maintain a relatively high humidity to facilitate disease development for 72 h. At 21 d postinoculation, the percentage of diseased spikelets was recorded for each spike to evaluate Type II resistance and the mean percentage of FHB severity over all spikes in each pot was calculated and recorded as the value of one replicate.

Production of F 1 hybrids
Initially, the morphology of F 1 hybrid seeds from all crosses was visually studied to validate that crossing was successful. True hybrid seeds derived from the crosses of hexaploid wheat lines 2398, Choteau, AC Vista, and AC Lillian by each of Sumai 3 and PI 277012 were significantly smaller than the normal self-pollinated seeds of their female parents (Table 1; Figure 1). The true F 1 seeds from the crosses of Sumai 3 and PI 277012 with tetraploid wheat lines LDN, Grenora, Alkabo, Divide, and the LDN-CS D-genome substitution lines were shriveled in comparison to the plump seeds of their female tetraploid parents (Figure 2). Ballester and de Vicente (1998) reported that molecular marker analysis is a more reliable method to verify presence of true F 1 hybrids compared with other methods. Herein, molecular marker UMN10, which is commonly deployed by breeders for selection of Fhb1 in Sumai 3, was used to validate presence of Fhb1 at the seedling stage in each F 1 hybrid derived from the crosses of Sumai 3 by tetraploid parents LDN, Grenora, Alkabo, Divide, and each LDN-CS D-genome substitution line. Each tetraploid wheat parent and each LDN-CS D-genome substitution line did not yield amplicons at the UMN10 locus, whereas F I G U R E 3 Polymerase chain reaction (PCR) amplification products from tetraploids and hexaploids at the molecular marker loci UMN10 and Xwgc1079 (internal control for PCR), respectively, for (from left to right): hexaploid cultivar 'Sumai 3′, tetraploid cultivars 'Grenora', 'Alkabo', 'Langdon' (LDN) and 'Divide', the LDN-Chinese Spring D-genome substitution lines (n = 14), and tetraploid × hexaploid F 1 hybrids from the crosses of LDN × Sumai 3 ( PCR amplification of products from each true F 1 hybrid produced the same amplicon as Sumai 3 at the UMN10 locus ( Figure 3). The STS marker Xwgc1079 (GATGGCCTAA-CAAATGATGT; TCCATCAAGCATACAGATGA) that constantly amplified a product (337 bp) in all genotypes involved in this experiment was used as an internal control for PCR (Figure 3). Lastly, spike morphology was used to confirm that crossing was successful for each F 1 hybrid derived from the crosses of PI 277012 by tetraploid parents LDN, Grenora, Alkabo, Divide, and each LDN-CS D-genome substitution line. Visual observations confirmed presence of a speltoid (spear-shaped) spike morphological phenotype for each true F 1 hybrid derived from PI 277012, thereby confirming presence of the q gene on chromosome 5A from PI 277012 (Chu et al., 2011) (Figure 4).

FHB Resistance in F 1 hybrids derived from PI 277012
The F 1 hybrids derived from the tetraploid (susceptible) × hexaploid (resistant) crosses of 'LDN × PI 277012′, 'Divide × PI 277012′, 'Grenora × PI 277012′, and 'Alkabo × PI 277012′ exhibited a resistance level not significantly different (P > .05) than paternal parent PI 277012 (Table 5). The mean FHB severity of all the F 1 hybrids across populations was 14.3%, indicating complete dominance of the resistance genes in PI 277012 over the susceptible alleles in each maternal tetraploid parent. Comparison of the FHB mean severity ratings for the F 1 hybrids derived from PI 277012 that were crossed with the complete set of LDN-CS D-genome substitution lines revealed that LDN4D(4A), LDN6D(6A), and LDN6D(6B) F 1 hybrids had significantly higher (P < .05) mean FHB severity ratings than F 1 hybrids of LDN × PI 277012, whereas the F 1 hybrids derived from the remaining LDN-CS D-genome substitution lines exhibited similar levels of resistance as F 1 hybrids derived from the cross of LDN × PI 277012 (Tables 5, 6).

Inheritance analysis
A total of 57 F 2 plants from the cross of LDN × Sumai 3 were evaluated for FHB resistance by calculating the mean percentage of FHB severity ratings following the single-floret spike inoculation method in the greenhouse . Of the 57 plants, two F 2 plants with FHB severity ratings of 14.8 and 15.8%, respectively, were bulk harvested to form an F 3 family consisting of 36 plants (Table 7). Of those 36 plants, two F 3 plants with FHB severity ratings of 13.8 and 17.9%, respectively, were bulked to form an F 4 family consisting of 38 plants. The same selection criteria were applied to the F 2 population consisting of 37 plants from the cross of Divide × Sumai 3, which were evaluated for FHB resistance. In this case, four F 2 plants were identified with FHB severity ratings from 7.4 to 16.3%, which were bulk harvested to form an F 3 family consisting of 52 plants (Table 7). In the F 3 generation, a single F 3 plant with FHB severity rating of 12.9% yielded 22 seeds, which were used to establish an F 4 family from the Divide × Sumai 3 cross. For the cross of LDN × PI 277012, 41 F 2 plants were evaluated for FHB severity ratings. Of those 41 single plants, two F 2 plants with FHB severity ratings of 6.8 and 8.4%, respectively, were selected and subsequently bulked to form an F 3 family comprised of 59 plants (Table 7). In the F 3 generation, two single plants were selected with FHB severity ratings of 12.9 and 15.0%, respectively, which were again bulk harvested to form an F 4 family from the cross of LDN × PI 277012. A wide variation in FHB severity was observed in the F 2 population of the LDN × Sumai 3 cross. Approximately 53% of the plants exhibited FHB severity ratings of less than 30% and only 3.5% of the plants were identified with FHB severity ratings lower than 10% ( Figure 5). All other segregants had FHB severity ratings that ranged from 30 to 90%. However, only 22.2% of the evaluated plants in the F 3 generation and just 2.6% of the plants evaluated in the F 4 generation were identified with an FHB severity rating of less than 30% ( Figure 5). Individuals with an FHB severity rating of less than 20% were not observed in the F 4 generation. In addition, over 50% and 80% individuals in the F 3 and F 4 families, respectively, were identified with FHB severity ratings higher than 50% ( Figure 5). The F 3 and F 4 families in the cross of LDN × Sumai 3 had higher susceptibility compared with the F 2 population. Similar segregation patterns for FHB severity ratings were observed for families in the F 2 , F 3 , and F 4 generations derived from the cross of Divide × Sumai 3 in comparison to that of LDN × Sumai 3. In the F 2 population, approximately 25% of the individuals had FHB severity ratings of less than 20%. However, over 80% of the individuals exhibited an FHB severity rating greater than 50% in each of the F 3 and F 4 families ( Figure 5). For families derived from the cross of LDN × PI 277012, segregation patterns of FHB severity ratings were initially similar in the F 2 generation to those patterns observed in the LDN × Sumai 3 and Divide × Sumai 3 crosses, respectively, as approximately 40% plants displayed an FHB severity rating of less than 20%. However, FHB severity ratings observed in the F 3 and F 4 generations clearly segregated differently than that of the LDN × Sumai 3 and Divide × Sumai 3 populations; for example, approximately 55% and 85% of plants in the F 3 and F 4 generations, respectively, recorded FHB severity ratings of less than 30%, which was a much higher frequency of resistant plants than that observed for the other two tetraploid by hexaploid populations ( Figure 5).

DISCUSSION
Utilization of hexaploid-derived FHB resistance in tetraploid wheat breeding programs has achieved limited success. It has been proposed that expression of hexaploid-derived FHB resistance genes may be influenced by genetic factors present in tetraploid (Rudd et al., 2001;Fakhfakh et al., 2011). In this study, all F 1 hybrids produced from the crosses of Sumai Note. The populations were advanced using bulked seeds of resistant segregants identified in the previous generation. FHB severity ratings for resistant plants were calculated in a greenhouse experiment following the single-floret spike inoculation method .
3 with FHB susceptible hexaploid wheats, 2398, Choteau, AC Vista and AC Lillian, exhibited a level of FHB resistance intermediate to their respective parents. However, the F 1 hybrids of Sumai 3 with tetraploid wheats LDN, Grenora, Alkabo, and Divide, exhibited levels of FHB resistance similar to or lower than that of their maternal tetraploid parent. Similar results were also reported in the F 1 hybrids of other tetraploid wheat crossed with FHB-resistant hexaploids including Sumai 3 (Gilbert et al., 2002). Conclusively, Sumai 3-derived FHB resistance genes were normally expressed in the F 1 hybrids produced from hexaploids, but not expressed in the F 1 hybrids produced with the tetraploid cultivars as much as what were expressed in the hexaploid background. The mid-parental FHB resistance level observed in the F 1 hybrids of Sumai 3 with hexaploid wheat is consistent with polygenic inheritance in the hexaploid genetic background. It was proposed that suppression of Sumai 3-derived resistance genes in tetraploid wheat was caused by suppressor genes in the tetraploid wheat genetic background (Gilbert et al., 2002). However, PI 277012-derived FHB resistance genes are normally expressed in the F 1 hybrids of each tetraploid wheat crossed with PI 277012 (Table 5). Apparently, FHB resistance genes in PI 277012 are completely dominant over the susceptible alleles in tetraploid wheat, which showed introgression efficiency of FHB resistance from various sources. The F 1 hybrids of Sumai 3 by the LDN-CS D-genome chromosomal substitution lines of LDN2D(2B), LDN3D(3A), LDN4D(4A), LDN4D(4B), LDN5D(5B), LDN6D(6A), LDN6D(6B), and LDN7D(7A) had significantly higher FHB resistance than their substitution line parents, which suggests LDN chromosomes 2B, 3B, 4B, 5B, 6B and 3A, 4A, 6A, and 7A, respectively, may contain genes that suppress expression of the Sumai 3-derived FHB resistance genes. No significant increase of FHB resistance was observed in the F 1 hybrids derived from LDN1D(1A), LDN2D(2A), LDN5D(5A), LDN1D(1B), and LDN7D(7B) by Sumai 3, indicating that LDN chromosomes 1A, 2A, 5A, 1B, and 7B may not influence expression of Sumai 3-derived FHB resistance genes. The F 1 hybrids derived from LDN4D(4A), LDN6D(6A), and LDN6D(6B) by PI 277012 exhibited an FHB resistance level lower than PI 277012, whereas the F 1 hybrids of PI 277012 crossed with other substitution lines had a similar resistance level as PI 277012. These results suggest that LDN chromosomes 4A, 6A, and 6B likely contain genes required for expression of PI 277012-derived FHB resistance. Moreover, these expression results clearly demonstrate that there are likely different mechanisms and/or different FHB resistance genes underlying the Sumai 3-and PI 277012-derived FHB resistance. Zhuang et al. (2012) suggested that FHB resistance in Sumai 3 could be conferred by reducing the susceptibility rather than producing an active resistance reaction.
Wide variation for FHB severity ratings were observed in the F 2 populations, F 3 and F 4 bulked families of the crosses of Sumai 3 by LDN and Divide. Furthermore, it was observed that the frequencies of plants with high levels of FHB resistance decreased from the F 2 to the F 4 generations. Alternately, a high frequency of plants with high levels of FHB resistance was retained over the F 2 to F 4 generations in the cross of PI 277012 by LDN. The difference in inheritance patterns between Sumai 3-and PI 277012-derived FHB resistance could be caused by several factors. Firstly, FHB resistance QTL have been identified on several chromosomes including 7A, 2B, 3B and 6B in Sumai 3 (Yao et al., 1997;Waldron et al., 1999;Zhou et al., 2002;Liu & Anderson, 2003). However, two FHB resistance QTL were mapped on the same chromosome 5A in PI 277012 (Chu et al., 2011). Thus, Sumai 3-derived FHB resistance QTL recombine more F I G U R E 5 Distribution of Fusarium head blight (FHB) severity ratings observed in the F 2 , F 3 , and F 4 generations in segregating susceptible tetraploid by resistant hexaploid populations derived from the crosses of (a) LDN × Sumai 3; (b) Divide × Sumai 3; (c) LDN × PI 277012. The populations were advanced using bulked seeds of resistant segregants identified in the previous generation frequently than those previously identified in PI 277012. Secondly, it was reported that introgression of a single FHB resistance gene from Sumai 3 could not provide an FHB resistance level comparable to Sumai 3 (Mergoum et al., , 2008. However, one of the two QTL in PI 277012 mapped to chromosome 5A could provide a level of FHB resistance comparable with PI 277012 in other FHB susceptible wheat genetic backgrounds (http://www.uky.edu/Ag/Wheat/wheat_ breeding/New%20Folder/Steve%20Xu.pdf). Lastly, herein we report that expression of Sumai 3-derived hexaploid resistance genes was possibly suppressed by multiple genes on different tetraploid chromosomes. However, PI 277012-derived FHB hexaploid resistance was normally expressed and the expression is possibly influenced by fewer genes in tetraploid.

CONCLUSION
The expression of hexaploid-derived FHB resistance genes in Sumai 3 was suppressed by multiple genes on different chromosomes in a tetraploid genetic background, based on FHB severity ratings of segregating populations derived from tetraploid and hexaploid wheats and the complete set of LDN-CS D-genome chromosome substitution lines. Because of these genomic incompatibilities, it may be difficult to utilize Sumai 3-derived FHB resistance genes for improvement of tetraploid wheat. However, complete dominance was observed for PI 277012-derived resistance genes over the susceptible tetraploid alleles, indicating that introgression of these resistance genes would be less complicated. The differing inheritance patterns and expression of FHB resistance genes between Sumai 3 and PI 277012, observed herein, illustrate that proper selection of hexaploid FHB resistance donor sources could provide an opportunity to improve FHB resistance in tetraploid breeding programs.

A C K N O W L E D G M E N T S
The authors thank the members in Xiwen Cai's and Shaobin Zhong's labs for their assistance to this research. This material is based upon work supported by the U.S. Department of Agriculture, under Agreement No. 58-3042-1-017. This is a cooperative project with the U.S. Wheat & Barley Scab Initiative. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture.

C O N F L I C T O F I N T E R E S T
All authors declare that they have no conflict of interest.