HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Knott, D.R. Search for Related Content PubMed Articles by Knott, D.R. Agricola Articles by Knott, D.R.

CROP BREEDING, GENETICS & CYTOLOGY

Inheritance of Resistance to Stem Rust in Medea Durum Wheat and the Role of Suppressors

^a Dep. of Plant Sciences, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK, Canada, S7N 5A8

knott{at}sask.usask.ca

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
The durum wheat (Triticum turgidum L.) cultivar Medea Ap9d has been used as a supplementary differential in the identification of isolates of stem rust (Puccinia graminis f. sp. tritici Eriks. & Henn.). Genes for stem rust resistance from Medea were transferred to a susceptible hexaploid wheat (T. aestivum L.), LMPG, so that they could be used more efficiently in race identification and wheat breeding. Medea was backcrossed five times to LMPG with selection for resistance to stem rust races LBB and MCC. Ten near-isogenic lines (NILS) were produced and tested with isolates of 10 stem rust races. The NILs fell into four types. Three of the four types had resistance to races to which Medea was susceptible. Apparently, Medea possesses suppressors for some of its genes for resistance and the suppressors were lost during the backcrossing to LMPG. This was confirmed when the cross, Medea/Glossy Hugenot, and the backcross, Medea/2 x Glossy Hugenot, segregated for resistance to races RCH and TMH(15B-4), to which both parents are susceptible. The segregations could not be explained on the basis of complementary genes. Medea was also crossed and backcrossed to a susceptible durum, Glossy Hugenot, for a genetic study using races MCC and TMB. The genetic analysis showed that Medea has four genes that condition resistance to races MCC, two of which also condition resistance to race TMB. None of the four types of NILs are resistant to both. Apparently, neither of the latter genes was transferred to LMPG. The presence of suppressors in Medea suggests that they may confer a selective advantage on plants that carry them.

Abbreviations: IT, infection type • NILS, near-isogenic lines • R, resistant • S, susceptible • seg, segregating

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
THE TUNISIAN DURUM WHEAT cultivar Medea Ap9d has been used at the USDA-ARS Cereal Disease Laboratory, University of Minnesota, as a supplementary differential in the identification of isolates of stem rust (Roelfs and McVey, 1979). Medea is known to carry the gene Srdp-2 for stem rust resistance (Roelfs and McVey, 1979), but probably carries other genes as well.

In identifying isolates of stem rust, differential lines carrying single Sr genes have considerable advantage. Perhaps most important, by using single-gene differentials one can determine exactly which corresponding genes for virulence the isolate carries and to which isolates the resistance gene conditions resistance. With a multigene differential, it is impossible to tell which gene for resistance or combination of genes results in a particular low-infection type.

A program was started to transfer stem rust resistance from Medea to hexaploid wheat with the objective of producing one or more single-gene lines that could be useful as stem rust differentials. It was anticipated that the genes might also be useful in breeding stem rust–resistant bread wheats. When it became evident that Medea carried a number of genes for resistance, a genetic study was undertaken to determine the inheritance of its resistance to two races of stem rust, MCC and TMB.

	Materials and methods

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
Seed of the tetraploids Medea Ap9d and Glossy Hugenot was obtained from Dr. A.P. Roelfs, Cereal Disease Laboratory, USDA, St. Paul, MN. LMPG, a daylength-insensitive hexaploid line susceptible to many races of stem rust, was produced by the author (Knott, 1990) for use as a recurrent parent in producing a set of near-isogenic lines (NILs) for the genes for stem resistance in wheat.

Medea was crossed with LMPG and backcrossed five times. During the backcrossing, seedlings were selected for resistance to races LBB and MCC of stem rust. Race LBB was used in the studies because it is relatively avirulent and, therefore, detects many resistance genes. Race MCC is an old race that has been used in many genetic studies at the University of Saskatchewan. LMPG gives an infection type (IT) 4 to both races, whereas Medea gives a 1⁼ X⁼ IT with race LBB and a 11⁺ with race MCC (using the 0 to 4 scale of Stakman et al., 1962). Infection types 0, 1, 2, and X were considered resistant, and 3 and 4 susceptible. Particularly in the early backcrosses, the seedlings were tested with both races. As it became apparent that a particular family had resistance to only one race, only that race was used in further testing. Plants with each combination of different ITs to the two races were backcrossed further to LMPG. In some cases, resistance was recessive and it was necessary to grow F₂ families to recover the resistance and continue the backcrossing. As the backcrossing progressed, families were obtained that were segregating for only one IT. After five backcrosses, the material was selfed until homozygous lines were obtained in the BC₅F₃ or BC₅F₄. Usually for each combination of ITs with races MCC and LBB, at least two homozygous lines with uniform ITs were produced. Sixteen to 20 plants per line were tested. After five backcrosses with selection for a specific IT, it was assumed that the lines carried single genes. However, it is possible that two linked genes could have been carried along together undetected. Ten lines and the parents were tested with 10 races of stem rust: LBB(111), MCC(56), TMH(15B-1), TMB(15B-1L), TMH(15B-4), RCH(11), HFC(29-1), HFC(48A), HDG(C15), and QCL(C65) (the old race number is given in brackets). Races 29-1 and 48A are very similar and key out to the same code under the new system (Roelfs and Martens, 1988), as do 15B-1 and 15B-4.

Medea was also crossed to Glossy Hugenot, a relatively stem rust–susceptible durum, and backcrossed once for a genetic analysis of rust resistance. The BC₁F₁ plants were tall and weak and seed set was often poor, limiting population sizes. The BC₁F₁ families were tested at the seedling stage in the greenhouse using races MCC and TMB to which Medea is resistant. Race LBB could not be used since both Medea and Glossy Hugenot are resistant. The author is not aware of a durum line that is susceptible to race LBB. Except for families with very limited seed, the tests were repeated once and often twice.

The rust tests were carried out in a greenhouse except for the tests of the 10 lines with 10 rust races, which were done in a growth chamber. The greenhouse was maintained as near as possible to 20°C during the day and 15°C at night. Supplementary lighting was used to provide a 16-h day length. In most cases, eight seeds were planted in a 15-cm-diameter plastic pot. Seedlings at the two-leaf stage were inoculated with rust spores suspended in soltrol, a light mineral oil. The inoculated plants were kept moist for 18 h with a humidifier. After either 13 or 14 d, the ITs were rated on a 0 to 4 scale, according to the system of Stakman et al. (1962). In some tests, the same plants were rusted twice, first with one race and 7 or 8 d later with a second race. The infection with the first race was evaluated after 13 or 14 d and the infected leaves removed. After another 7 or 8 d, the infection with the second race was evaluated. Because the BC₁F₂ plants were often weak, lower leaves were sometimes shaded and rust developed somewhat poorly. A few backcross families segregated for a very heavy physiological flecking, which made rust evaluation difficult on some plants. Despite these problems, the distinction between segregating and susceptible families was usually clear.

For tests in the growth chamber, plants were grown under a 16-h light period at 20°C and an 8-h dark period at 15°C. Inoculations were done as in the greenhouse.

	Results

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
Genetic Analysis
In a backcross, if one gene for resistance is segregating, the expected ratio of BC₁F₂ families is 1 segregating (seg) for resistance (R): 1 susceptible (S). If two genes are segregating the expected ratio is 3 seg:1S. As shown later, Medea carries suppressors for at least some of its genes for resistance. The presence of a suppressor does not change the frequencies of segregating and susceptible families; that is, if there is one gene for resistance and one suppressor the expected ratio is still 1 seg:1S, and if there are two genes for resistance and two suppressors the expected ratio is still 3 seg:1S. However, the segregation within the segregating families is changed. With one dominant gene for resistance, all of the segregating families will segregate 3R:1S. However, if there is one dominant gene for resistance and one dominant suppressor, one-half of the segregating families will segregate 3R:1S, but the other half will segregate 3R:13S. If there are two genes for resistance and two suppressors, the segregating families will be of six types ranging from 15R:1S to 3R:13S. Large samples of each BC₁F₂ family would have to be grown to distinguish among the various ratios — much larger samples than were actually grown in this study. However, the major effect of suppressors is to reduce the overall frequency of resistant plants in segregating BC₁F₂ families. For example, with one dominant gene for resistance, the overall ratio changes from 3R:1S with no suppressor to 15R:17S with a dominant suppressor. With two dominant genes for resistance, the overall ratio changes from 15R:1S to 517R:507S (or 1.02:1).

In the backcross of Medea to Glossy Hugenot, 64 BC₁F₂ families segregated and 24 were susceptible to race TMB, a good fit to a 3:1 ratio (Table 1) . Thus, Medea carries two genes for resistance to race TMB. All 64 families that segregated for resistance to race TMB also segregated for resistance to race MCC, showing that the two genes condition resistance to both races. However, of the 24 families that were susceptible to race TMB, 14 segregated for resistance to MCC and 10 were susceptible, indicating that Medea carries at least one and possibly two additional genes conditioning resistance to race MCC. In total, 78 BC₁F₂ families segregated for resistance to race MCC and 10 were susceptible. The segregation fits either a 7:1 or a 15:1 ratio, although the fit is much better to a 7:1 ratio (Table 1). The results indicate that Medea has either three or four genes for resistance to race MCC, including the two that also give resistance to race TMB. If the results for the two races are combined and it is assumed that there are three genes, the expected ratio is six segregating for resistance to both races, zero segregating for resistance to TMB and susceptible to MCC, one segregating for resistance to MCC and susceptible to TMB, and one susceptible to both races. With four genes, the expected ratio is 12:0:3:1. The observed segregation (64:0:14:10) is a good fit to the former (P = 0.50–0.75) and a reasonable fit to the latter (P = 0.25) (Table 1).

In two of the three tests on the BC₁F₂ families, the seedlings were rusted first with MCC and then with TMB. In the third test, the order was reversed. In the latter test, in 14 of 65 families tested, some plants had a first leaf that was resistant to TMB (often a 1⁺ IT) and a second leaf that was susceptible (IT 3 or 4). Apparently, Medea carries a gene that conditions resistance to TMB in the early seedling stage, but the resistance is quickly lost.

Near-Isogenic Lines
The 10 NILS tested with 10 races of stem rust fell into four types, each having a different pattern of ITs (Table 2) . Since five backcrosses were made in producing the NILs and selection was for a uniform IT, I assumed that each carries a single gene for resistance. If a line, by chance, still carried two genes conditioning different ITs, this would have been detected during backcrossing.

The two lines of Type 1 were resistant only to races MCC and QCL, a pattern of resistance different than that for any of the other lines tested. The lines must carry either one of the more recently identified genes that I have not yet tested or a gene not previously identified.

The six lines of Type 2 and the single line of Type 3 were resistant to five races, but their reactions differed for races HFC(29-1) and HDG. The Type 2 lines gave the same pattern of resistance and susceptibility as a NIL carrying Sr34. However, they were distinctly more susceptible to two races than an Sr34 NIL (1^+c vs. 0;1^- to LBB, and X^- vs. 1 to QCL). In any case, the gene Sr34 was transferred into common wheat from T. camosum (Sibth. & Sm.) Richter (McIntosh et al., 1982) and is unlikely to be present in a durum wheat. The Type 3 line has a different response pattern than any other line I have tested.

The single line of Type 4 was resistant to seven races and susceptible to only three, TMH (15B-1), TMB, and TMH(15B-4), all members of the 15B complex. In previous tests, a NIL carrying Sr9e gave the same pattern of resistance and susceptibility, although 15B-4 was not tested (Knott, 1990). However, the Type 4 line derived from Medea gave a higher IT to each avirulent race than did the Sr9e line (e.g., 2^- vs. 1^-1 to QCL), suggesting that the two genes are different. A test cross between the Type 4 line and an Sr9e NIL segregated 35R:5S, confirming that the Type 4 line does not carry Sr9e. The segregation fits the 15R:1S ratio expected with two genes (P = 0.10–0.25).

Anomalous Results
Several completely unexpected results occurred in the data. Three of the four types of lines carrying genes from Medea had resistance to one or more races to which Medea was susceptible. Lines of Type 2 were resistant to HFC(29-1) and HFC(48A); Type 3 to HFC(48A); and Type 4 to RCH, HFC(29-1), and HFC(48A). The origin of the anomalous resistance in these lines cannot be identified in the rust pedigrees of the lines because races RCH, TMH(15B-4), HFC(29-1), and HFC(48A) were not used for selection during the backcrosses. However, selection for resistance to races MCC and LBB must have resulted in selection for genes that also gave resistance to the races to which Medea is susceptible.

One possible explanation for these results is that Medea carries suppressors of rust resistance that are not present in LMPG and were lost during backcrossing. To account for the development of three types of lines resistant to races to which Medea is susceptible would require the presence of three resistance genes and three suppressors, unless some of the latter suppress more than one gene.

Further evidence comes from comparing results from the genetic analysis with those from testing the near-isogenic lines. The analysis of the backcross of Medea to Glossy Hugenot showed that Medea carries four genes that condition resistance to race MCC, two of which also condition resistance to race TMB. All four types of NILs are resistant to MCC, but none are resistant to TMB. Thus, in total, six genes must control resistance to race MCC. Two of the genes giving resistance to MCC but not to TMB were not detected in the backcross to Glossy Hugenot. The simplest explanation is that both Medea and Glossy Hugenot carry the same suppressor or suppressors for the two genes, so they were not detected. Two explanations are possible for the failure to transfer to LMPG the two genes controlling resistance to both races MCC and TMB. LMPG could carry suppressors for both genes. Alternatively, with six genes segregating in the backcrosses to LMPG, if the two genes did not condition distinctive ITs, they could easily have been lost by chance during backcrossing.

Medea is resistant to races TMH(15B-1) and TMB, but none of the NILs are resistant to either race. Thus, Medea must have at least one more gene for resistance that was not transferred to LMPG.

Since NILs of Types 2, 3, and 4 were resistant to races to which Medea was susceptible, their genes were presumably suppressed in Medea. Line 1 does not condition resistance to races LBB and HDG, and if the resistance of NILs of Types 2, 3, and 4 is suppressed in Medea, then the resistance of Medea to the two races must be controlled by one or more additional genes.

During the backcrosses to LMPG, if selection had been practiced for resistance to additional races to which Medea is resistant, more genes for resistance would have been identified and transferred.

Proof of the Presence of Suppressors
To prove that suppressors are present in Medea, F₁ and F₂ plants from reserve seed of the cross, Medea/Glossy Hugenot, were tested with races RCH and TMH(15B-4), to which both parents are susceptible. The F₁ plants were susceptible to race RCH (IT 34) and moderately susceptible to TMH(15B-4) (IT 2⁺3). However, when tested with race RCH, F₂ plants gave a range of ITs from 1 to 4. It was not possible to separate the F₂ plants into completely distinct classes. However, in taking the rust readings, only those seedlings that appeared to be clearly susceptible were classified as IT 3 or higher. If a division was made between ITs 2⁺ and 3, the segregation was 36R:41S. The resistance cannot result from complementary, dominant genes since the F₁ plants were susceptible, nor can it result from complementary recessive genes since the F₂ segregation does not fit a 1R:15S ratio (P < 0.001). The alternative is that the resistance to race RCH had been suppressed, presumably in Medea, and also in the F₁ plants. Segregation for the suppressors occurred in the F₂ generation and resistance was expressed. The expected ratio for a dominant gene for resistance and a dominant suppressor, would be 3R:13S, which the data do not fit (P < 0.001). Evidently, the genetic explanation is more complicated. Possibly there are two dominant genes for resistance and two dominant suppressors.

The same F₂ plants were also tested with race TMH(15B-4), but the plants were rather vigorous and floppy, and the rust infection was erratic as a result of shading. Most plants gave similar ITs with both races, suggesting that at least one gene conditioned resistance to both. However, two plants were susceptible to race RCH and resistant to race TMH(15B-4), and another showed the reverse pattern. Thus, one gene may provide resistance only to race RCH (Type 4) and another may provide resistance only to race TMH(15B-4).

Seed was available from 16 BC₁F₁ plants from the backcross Medea/2 x Glossy Hugenot. Small BC₁F₂ families of 15 to 22 plants were tested with races RCH and TMH(15B-4). With race RCH, 14 families segregated and two were susceptible. The results fit the 3:1 ratio expected if two genes for resistance and two suppressors are segregating (P = 0.25–0.50), confirming the F₂ results. Because of the small number of families tested, the segregation will also fit the 7:1, 15:1, or 31:1 ratios expected with 3, 4, or 5 resistance genes, respectively.

The same plants were tested with race TMH(15B-4) and all 16 BC₁F₂ families segregated, including the two that were susceptible to race RCH. Thus, at least one gene gives resistance to TMH(15B-4) but not to race RCH. No NIL of this type was obtained. No BC₁F₂ families were susceptible to race TMH(15B-4) but segregating to race RCH. However, seven families segregated plants that were resistant to race RCH but susceptible to race TMH(15B-4). Thus, at least one gene conditions resistance to race RCH but not to TMH(15B-4). This is true for the line of Type 4.

	Discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
The first reaction to the results given in this paper may well be skepticism. When I first organized the data on the NILs and compared them with the genetic analysis, I was skeptical as well. However, as I reviewed the data, I concluded that they are valid. The fact that the four types of NILs did not appear to be similar to any of the other lines I have worked with reduces the likelihood that they resulted from outcrossing or seed contamination. The fact that the cross, Medea/Glossy Hugenot, and the backcross, Medea/2 x Glossy Hugenot, segregated for resistance to races RCH and TMH(15B-4), to which both parents are susceptible, proves that resistance to these races is suppressed in Medea.

The inheritance of stem rust resistance in Medea is extraordinarily complex. Four genes appear to have been isolated from it and, at least four more genes are present that were not transferred to LMPG. Furthermore, the genes carried in three of the four types of NILs appear to be suppressed in Medea. Unless one suppressor suppresses more than one resistance gene, Medea must carry three suppressors. Additional genes for resistance are likely to be identified if additional races of stem rust are used to transfer them to LMPG.

Suppressors of stem and leaf rust resistance in wheat have been identified previously (McIntosh and Dyck, 1975; Kerber and Green, 1980; Kerber, 1991; Bai and Knott, 1992; Williams et al., 1992; Nelson et al., 1997). Suppression can be genetically complex. Bai and Knott (1992) found that `Chinese Spring' carries three complementary genes on chromosomes 1D, 2D, and 4D which suppressed stem rust resistance in three accessions of T. turgidum var. dicoccoides. The suppressors of rust resistance were on the D-genome chromosomes of a hexaploid wheat cultivar and suppressed genes on the A- and B-genome chromosomes of tetraploid wheats. In the present case, the suppressors are on the A or B genome chromosomes of Medea and suppress resistance genes in Medea. As stated by Bai and Knott (1992), it is difficult to understand why suppressors of rust resistance should accumulate in a cultivar. Presumably, they must have some selective advantage. Possibly some genes for rust resistance are deleterious in the absence of rust and the suppressors suppress both the resistance and the deleterious effects. Alternatively, the suppressors may have some other favorable effects. Clearly, there is a great deal more work to be done to clarify the inheritance of stem rust resistance in Medea.

	ACKNOWLEDGMENTS

 
The author is pleased to acknowledge the financial support of the Natural Sciences and Engineering Council and the Saskatchewan Agricultural Development Fund for this research.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 
This research received financial support from the Natural Sciences and Engineering Research Council of Canada and the Saskatchewan Agriculture Development Fund.

Received for publication July 16, 1998.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and methods Results Discussion REFERENCES

 

• Bai D., Knott D.R. Suppression of rust resistance in bread wheat (Triticum aestivum L.) by D-genome chromosomes. Genome 1992;35:276-282.
• Kerber E.R. Stem-rust resistance in `Canthatch' hexaploid wheat induced by a nonsuppressor mutation on chromosome 7DL. Genome 1991;34:935-939.
• Kerber E.R., Green G.J. Suppression of stem rust resistance in hexaploid wheat cv. Canthatch by chromosome 7DL. Can. J. Bot. 1980;58:1347-1350.
• Knott D.R. Near isogenic lines of wheat carrying genes for stem rust resistance. Crop Sci. 1990;30:901-905.[Abstract/Free Full Text]
• McIntosh R.A., Dyck P.L. Cytogenetical studies in wheat. VII. Gene Lr23 for reaction to Puccinia recondita in Gabo and related cultivars. Aust. J. Biol. Sci. 1975;28:201-211.
• McIntosh R.A., Miller T.E., Chapman V. Cytogenetical studies in wheat XII. Lr28 for resistance to Puccinia recondita and Sr34 for resistance to P. graminis tritici. Z. Pflanzenzucht. 1982;89:295-306.
• Nelson J.C., Singh R.P., Autrique J.E., Sorrells M.E. Mapping genes conferring and suppressing leaf rust resistance in wheat. Crop Sci. 1997;37:1928-1935.[Abstract/Free Full Text]
• Roelfs A.P., Martens J.W. An international system of nomenclature for Puccinia graminis of sp. tritici. Phytopathology 1988;78:526-533.[ISI]
• Roelfs A.P., McVey D.V. Low infection types produced by Puccinia graminis f.sp. tritici and wheat lines with designated genes for resistance. Phytopathology 1979;69:722-730.
• Stakman, E.C., D.M. Stewart, and W.Q. Loegering. 1962. Identification of physiologic races of Puccinia graminis var. tritici. USDA Bull. E617 (revised).
• Williams N.D., Miller J.D., Klindworth D.L. Induced mutations of a genetic suppressor of resistance to wheat stem rust. Crop Sci. 1992;32:612-616.[Abstract/Free Full Text]

This article has been cited by other articles:

				J. M. Bonman, H. E. Bockelman, Y. Jin, R. J. Hijmans, and A. I. N. Gironella Geographic Distribution of Stem Rust Resistance in Wheat Landraces Crop Sci., September 1, 2007; 47(5): 1955 - 1963. [Abstract] [Full Text] [PDF]

				D.R. Knott Inheritance of Resistance to Stem Rust in `Triumph 64' Winter Wheat Crop Sci., September 1, 2000; 40(5): 1237 - 1241. [Abstract] [Full Text]

This Article Abstract Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Knott, D.R. Search for Related Content PubMed Articles by Knott, D.R. Agricola Articles by Knott, D.R.