. . arr. . ~t ' 3 .73.”? t l I $.75“; H. x. I\ 2 T .. .. . a a ‘4 2‘. w. - ya 1. . v .5 x. I ”it Ar, u. I . . a ‘5»: a . 3 3 .a\...J. .L 1.1. ‘ .3... THESIS :13 N MICHIG GAN STATE ) 3IIU'I3IIH l! UlflllllHill”!HHIIHIIHUINIHI 01410 6698 This is to certify that the dissertation entitled Genetic Diversity among Wheat Germplasm Pools with Diverse Geographical Origins presented by Hong-Sik Kim has been accepted towards fulfillment of the requirements for Doctoral degreein Plant Breeding & Genetics/ Crop & Soil Sciences / Major pr essor Date MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY l Michigan State! University ' PLACE N RETURN BOX to romanthb checkout from your noord. TO AVOID FINES Mum on or More data duo. DATE DUE DATE DUE DATE DUE MSU loAn Affirmative Action/Equal Opportunity Intuition Was-m GENETIC DIVERSITY AMONG WHEAT GERMPLASM POOLS WITH DIVERSE GEOGRAPHICAL ORIGINS By HONG—SIK KIM A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Plant Breeding and Genetics Program Department of Crop and Soil Sciences 1995 GE Since recognized in genetic bases requested to o. iml‘mi'ement ; chmcteristics ditcrsity in Ed Philogenetic i- ”5le RFLPS. L collections “3;; RFLP ABSTRACT GENETIC DIVERSITY AMONG WHEAT GERMPLASM POOLS WITH DIVERSE GEOGRAPHICAL ORIGINS By Hong-Sik Kim Since a significant level of genetic erosion and latent vulnerability has been recognized in some major crop species, more efforts have been made to expand their genetic bases. Systematic quantification and management of genetic diversity are requested to optimize exotic and elite germplasm resources and to further facilitate wheat improvement programs. The objectives of this research were to: (I) analyze RF LP characteristics and their association with coefficient of parentage for measures of genetic diversity in Eastern US. soft winter wheat lines, (2) investigate the genetic nature and phylogenetic relationships of T. tauschii and landraces of T. aestivum found in China using RFLPs, and (3) determine patterns of genetic diversity in world germplasm collections with different origins using RFLP and morphological characters. RFLP variation was very limited for the 22 Eastern US. sofi wheat lines. The highly pedigree-related soft white winter wheat lines (mean coefficient of parentage (COP) = 0- mm COP of DNA PC similarit.‘ K As c tam/111' p00 base. the Chi RFLP-based Tibetm \l'ect landtaees \xet tau‘st‘ht'i on th from Southttt Eacht institutional u RFLP variant “inter wheat : RFLP-based 3; determined b3. spite of sign}; mo’PhOIOgicaQ Ml." Correla (COP) = 0.51) was much less genetically diverse than the soft red winter wheat lines with mean COP of 0.15. Insertions and/or deletions might play an important role in the origin of DNA polymorphisms. As measures of genetic relationships, RFLP-based genetic similarity (GS) significantly correlated with COPS for all genotype pairs (r = 0.73"). As compared to the Southwest Asian gene pool of T. tauschii, the Chinese T. tauschii pool showed a low frequency of polymorphism. In spite of a narrow genetic base, the Chinese landraces of common wheat were divided by the cluster analysis of RFLP-based GS matrix into two subgroups: Xinjiang Rice wheat for one group and Tibetan Weedrace, Sichuan White and Yunnan Hulled wheats for the other. The Chinese landraces were more related to the Southwest Asian T. tauschii than to the Chinese T. tauschii on the basis of RFLP-based mean GS. An eastward decline of RF LP diversity from Southwest Asia was apparent for both T. tauschii and landrace pools of T. aestivum. Each of the 21 germplasm pools of common wheat grouped by geographical or institutional origins of 338 accessions showed significant level of genetic uniformity. RFLP variation was highest in the Turkish landrace pool. The Eastern US. sofi red winter wheat pool revealed the highest RFLP diversity in the advanced germplasm pools. RFLP-based genetic relationships among accessions or germplasm pools were largely determined by their common geographical origins, breeding history and/or coancestry. In spite of significant variation (p < 0.01) among germplasm pools for all examined morphological characters, morphology-based standard taxonomic distance estimates were poorly correlated with the RFLP-based GS estimates (r = -O.15 NS). To my parents, grandmother, sister and brother. iv guidanc. with the as the gre Al Thomas-ho members. gratefull} a The l'SDAl. Dr. lelllC'Cae Re Valuable \t hm APPTL‘ 38.0111 (K, ACKNOWLEDGMENTS The author wishes to express his sincere gratitude to Dr. R. W. Ward for his guidance and support during the entire course of doctoral program. He has always been with the author as the major professor who provides invaluable advice and assistance, and as the greatest friend who shares consistent encouragement and cooperation. Also, the author greatly thanks Drs. D. Douches, J. F. Hancock and M. F. Thomashow for their helpful suggestions and consultation as guidance committee members. Especially, the critical reading of the manuscript by Dr. J. F. Hancock is gratefully acknowledged. The author is grateful to Dr. H. E. Bockelman (National Small Grains Collection, USDA), Dr. C. J. Peterson (USDA-ARS, University of Nebraska) and Prof. C. Yen (Triticeae Research Institute, Sichuan Agricultural University, China) for providing valuable wheat germplasm with diverse geographical origins. Appreciation is also extended to Dr. M. E. Sorrells (Cornell University) and Dr. B. S. Gill (Kansas State University) for providing DNA clones. A special thanks is offered to Ms. J. L. Yang for conducting laboratory work for the Chinese wheat germplasm studies and to Dr. Y. Yen (University of Nebraska) for helpfiil discussion and information on some Chinese wheat materials. The author wishes to thank Drs. K. Hokanson, S. Gilmore and Mr. P. Callow for their technical assistance in molecular biological experiments, and Ms. H. Awale and Ms. S, Carlson Th: and Ms. E. The sister and b author \t'oul S. Carlson for their kind assistance in both field and laboratory work. The assistance and friendship of the wheat group people including Mr. D. Glenn and Ms. E. Jenkins is also sincerely appreciated. The author would like to express the greatest thanks to his parents, grandmother, sister and brother for their love and encouragement. Without their hearty support, the author would not attain his desired end. vi MO: USltDF llST OF LlSl‘()F ABBRE\l GENERAL 1. ESINA' WINTIZ R PARENT Abstr lntrod Mater Result DMD TABLE OF CONTENTS SECTION Page LIST OF TABLES ix LIST OF FIGURES xi LIST OF APPENDICES xiii ABBREVIATIONS xiv GENERAL INTRODUCTION 1 I. ESTIMATION OF GENETIC DIVERSITY IN EASTERN U.S. SOFT WINTER WHEAT BASED ON RFLPS AND COEFFICIENTS OF PARENT AGE 12 Abstract 12 Introduction 14 Materials and Methods 17 Results 25 Analysis of RFLP diversity 25 Analysis of coefficients of parentage (COP) 34 Relationship between RFLPS and COPs 35 Discussion 40 DNA polymorphism in Eastern US. soft winter wheats 40 Comparison of the germplasm bases of the Eastern US. SRW and SWW wheat gene pools 41 Relationship between marker-based GS and COP measures of diversity 42 vii II. EYC IN 7 T. Ut'. In M. Re Ill. PA TH GER‘ilPlA MORPHOL Abs: IIIIIU; Mater RESuI DlScu~ LITERATURE! II. EVOLUTIONARY ASPECTS OF RFLP-BASED GENETIC DIVERSITY IN Triticum tauschii Cosson AND FOUR LANDRACE POOLS OF T. aestivum L. FROM CHINA Abstract Introduction Materials and Methods Results and Discussion RFLP analysis of T. tauschii Cosson RFLP analysis of T. aestivum L. III. PATTERNS OF GENETIC DIVERSITY IN COMMON WHEAT GERMPLASM (T. aestivum L.) DETERMINED BY RFLPS AND MORPHOLOGICAL TRAITS Abstract Introduction Materials and Methods Results Genetic diversity and relationships determined by RFLPS Genetic diversity and relationships determined by qualitative and quantitative traits Discussion Polymorphisms in germplasm pools of common wheat Cluster analysis and genetic relationships of germplasm pools Efficacy of morphological trait similarity and its application Genetic diversity and breeding strategy APPENDICES LITERATURE CITED viii 45 45 47 50 56 56 65 77 77 79 82 91 91 112 125 125 128 130 131 133 136 [Lites in». l. I“ L" orig 3- Pro ;‘ me. I and. 4- Sun: band indl‘x Pain1 for th 6' thr Hind] (helm 30ft u NOrrn 48 PM tsntt Relaij, Comm WM 1. n1elis COmm3 LIST OF TABLES CHARIERJ Twenty-two Eastern US. soft wheat lines and their pedigree information, origin and year of release The list of 48 probes used in the Southern blot analysis Proportion of monomorphism and uniqueness among total bands revealed by 48 probes with different restriction enzymes in individual and combined gene pools of Eastern US. soft wheat Stunmary of average values of PIC, no. of banding patterns and no. of bands revealed by each probe with different restriction enzymes in individual and combined gene pools of Eastern US. soft wheat Pairwise comparisons of RF LPs and mean genetic similarity estimates for the combined and separate gene pools RFLP-based genetic similarity (above diagonal) generated by 48 probe- HindIII restriction enzyme combinations and coefficients of parentage (below diagonal) based on pedigree information for the 22 Eastern US. soft winter wheats Normalized Mantel statistic (r) values between similarity matrices developed from RFLPS with three different restriction enzymes and 48 probes for the Eastern US. soft winter wheat germplasms (SRW and SWW) Relationships between genetic similarity matrices based on RFLPS and coefficient of parentages CHABIEILII The list of germplasms of T. tauschii and four Chinese landraces of common wheat ix 18 21 26 26 29 31 34 38 52 b) h.) Lo) RI {CS List USCL The Statu. Paint germ} PlCi Mean obse Anal} 00mm 2. List of a set of 30 KSU clones for RFLP analysis 55 3. RFLP banding characteristics revealed by 75 probes and the HindIII restriction enzyme for T. tauschii gene pools 58 4. Estimated mean, minimum and maximum values of genetic similarity coefficients within and between gene pools 63 5. The magnitude of polymorphism within each landrace pool of Chinese hexaploid wheat 65 6. RFLP banding characteristics revealed by 63 probes and the HindIII restriction enzyme for four landrace gene pools of Chinese wheat 66 CHAEIERJII 1. List of 300 accessions of 17 germplasm pools of T. aestivum selectively used in the RFLP and morphological trait analysis 87 2. The frequencies of RFLPS in the 21 germplasm pools of common wheat 93 3. The frequencies of RF LPs in germplasm pools grouped by developmental status and geography 94 4. Pairwise comparisons of RFLPS over all possible genotype pairs in each germplasm pool 97 5. PIC index values of 30 probes for 21 germplasm pools of common wheat 99 6. Means and standard deviations (Means i SD) of morphological traits observed in 17 germplasm pools of common wheat 113 7. Analyses of variance for six quantitative traits in 17 germplasm pools of common wheat 1 14 E i gures, 50ft u "J 1.4 J LIST OF FIGURES Eigms CHAPIERJ Average relative frequency of polymorphic probes for individual enzymes for genotype pair-probe combinations which showed polymorphism with (1) no other enzyme (poly w/O), (2) one other enzyme (poly w/ 1), and (3) two enzymes (poly w/2) Dendrogram resulting from the cluster analysis of the RFLP-based Nei and Li’s genetic similarity matrix among 22 Eastern US. soft wheats Dendrogram resulting from the cluster analysis of coefficients of parentage among 22 Eastern US. soft winter wheats The plot of RF LP-based genetic similarity (63035) and expected GS (GSEXP) vs. coefficients of parentage for 231 pairs of the Eastern US. soft winter wheat genotypes W Geographical regions where accessions of T. tauschii and four Chinese landraces of T. aestivum were collected Frequency of probes revealing polymorphism in separate and combined gene pools of T. tauschii and T. aestivum Dendrogram resulted from cluster analysis of the RFLP-based genetic similarity matrix among 20 accessions of T. tauschii Dendrogram resulting from cluster analysis of the RF LP-based genetic similarity matrix among 39 accessions of Chinese hexaploid wheat Associations of 39 accessions of Chinese hexaploid wheat revealed by principal coordinate analysis of RFLP-based genetic similarity xi 30 33 36 39 53 57 64 7O C 0‘ Sp: CHAPIII 14) Der gen “he coefficients 7 1 6. The mean RF LP-based genetic similarity coeflicient values between species of the T. tauschii and T. aestivum gene pools 74 CHARTERJII 1. Dendrogram resulting from the cluster analysis of RFLP-based genetic similarity estimates among 292 accessions of common wheat germplasms 102 2. Dendrogram resulting from the cluster analysis of RFLP-based mean genetic similarity estimates within and between 21 germplasm pools of common wheat 108 3. Association of 21 germplasm pools of common wheat resulting from the cluster analysis of Nei’s genetic distance estimates based on (a) restriction band frequencies within a pool and (b) RFLP banding pattern frequencies within a pool 110 4. Dendrogram resulting from the cluster analysis of morphological trait-based taxonomic distance estimates among 265 accessions of common wheat germplasms 118 5. Dendrogram resulting from the cluster analysis of morphological trait-based mean taxonomic distance estimates within and between 17 germplasm pools of common wheat 124 xii Appendix A. Me. Sim; COD] ‘1‘: QCDC b .\'ei C Fig: of 11 com LIST OF APPENDICES Appendix Page A. Mean, minimum and maximum values of RFLP-based genetic similarity estimates and genotype pairs within gene pools of common wheat 133 B. RFLP-based mean genetic similarity estimates within and between 21 germplasm pools of common wheat. All genotype-pairwise genetic similarities within and between pools were computed by Nei and Li’ measures 134 C. Eigenvectors, eigen values, cumulative variation and its proportion of the phenotypic trait correlation matrix for the first six principal components (PC) 135 xiii AIG mu ARG BCD crass CH.\'_T\\' CH.\'_XR CIi\'_YH COO COP PM GER GD GS HRW Hm llll'gx PCR PIC AFG ANOVA BCD CHN_SW CHN_TW CHN_XR CHN_YH CD0 COP FRA GER GD GS HRW IRAN IWWSN PCR PIC ODESSA RAPD RFLP ROM ABBREVIATIONS Afghanistan common wheat (landrace) germplasm Analysis of Variance Argentine common wheat cultivars Barley cDNA clone Sichuan White wheat complex (landrace) from China Tibetan Weedrace wheat (landrace) from China Xinjiang Rice wheat (landrace) from China Yunnan Hulled wheat (landrace) from China Oat cDNA clone Coefficient of Parentage French common wheat cultivars German common wheat cultivars Genetic Distance Genetic Similarity Hard Red Winter (Wheat) Iranian common wheat (landrace) germplasm International Winter Wheat Screening Nursery lines Polymerase Chain Reaction Polymorphic Information Content Wheat breeding lines from the Institute of Plant Breeding and Genetics in Odessa, Ukraine Correlation coefficient Random Amplified Polymorphic DNA Restriction Fragment Length Polymorphism Romanian common wheat cultivars xiv Rt's SAHN SRW 5w II'R L'KR L'PGMA L'S_ER t's_E\t' I'S_GP I'S_.\tSL' L'S_\t‘ \l’G tl’G RUS SAHN SRW SWW UPGMA US_ER US_EW US_GP US_MSU us_w WG Russian common wheat cultivars Sequential, Agglomerative, Hierarchical, and Nest (clustering) Soft Red Winter (Wheat) Soft White Winter (Wheat) Turkish common wheat (landrace) germplasm pool Ukraine common wheat cultivars Unweighted Pair Group Method, Arithmetic average Eastern US. sofi red winter wheat cultivars and breeding lines Eastern US. sofi white winter wheat cultivars Hard red winter wheat cultivars in the US. Great Plains Eastern US. sofi white winter wheat breeding lines developed from Michigan State University Western US. soft white winter wheat cultivars and breeding lines Wheat genomic DNA clone Yugoslavian common wheat cultivars XV C or important 1 ago tRenfn S}stematic ' 1986; Brig-g introductior nitrogenous Nob, PTOductix it}- COUnlries_ 13 farmer‘s prir GENERAL INTRODUCTION Common hexaploid wheat (T. aestivum L. 2n=6x=42) has been one of the most important food sources for human beings since it was first cultivated at least 9,000 years ago (Renfrew, 1973). A steady expansion of wheat productivity has occurred after systematic breeding programs began to be established early in this century (Johnson, 1986; Briggle and Curtis, 1987). Particularly, increase in grain yields resulted from introduction of semi-dwarf and high-yielding varieties together with increasing use of nitrogenous fertilizers, herbicides and pesticides (Frankel, 1970). Nobody doubts the remarkable contribution of the “Green Revolution” to wheat productivity in specific regions including many under-developed and over-populated countries. However, significant genetic erosion occurred through the loss of local farmer’s primitive cultivars (Frankel, 1970; Fowler and Mooney, 1990). Agronomically poor performance and lack of knowledge of unadapted exotic germplasm resources are barriers to its immediate incorporation into crop breeding programs (Cox, 1991). Moreover, continuous demand of farmers and cereal processors for uniform and high-quality varieties tends to lead plant breeders to restrict their breeding materials to only a few elite breeding stocks (Committee on Genetic Vulnerability of Major Crops, 1972; Kronstad, 1986). After the corn blight in 1970, the significant level of latent vulnerability due to genetic rmiformity throughout the major crop species was recognized by the reports from the ‘Committee on Genetic Vulnerability of Major more than maize and P13 derelopme programs. sources of l 1976; Kron Sew germplasm 1 has been uti Characters. t. {3011115 for b based on pri. collections U Trad; qufintitatit'ei- compared SP? Spares of B 3 Similar dipim 3111973363 CfIlIeI‘ of dIVQ‘ 2 of Major Crops’ (1972). Surprisingly, only a small number (< 10) of varieties dominated more than 50% of acreage in the US. in each crop species such as wheat, soybean, rice, maize and potato. Plant breeders of today, therefore, are more challenged toward progress in both development of purely superior inbreds and expansion of genetic base in their breeding programs. Obviously, world germplasm collections have a great potential as existing sources of genetic diversity and favorable genes to facilitate crop improvement (Harlan, 1976; Kronstad, 1986). Several independent measures have been used to resolve the genetic nature of germplasm resources including elite materials. Basic information on genetic diversity has been utilized to estimate genetic variation, to identify agronomically important characters, to clarify taxonomic or phylogenetic relationships, and to clarify heterotic groups for breeding parental selection in crop improvement. Grouping of germplasms based on primary information may also provide guidance for the choice of lines for core collections of germplasm (Frankel and Brown, 1984). Traditionally, morphological (or agronomic) characters inherited qualitatively or quantitatively have been used for the above purposes. Sarkar and Stebbins (1956) compared spikelet morphologies of einkom and emmer wheat to deduce the progenitor species of B genome of hexaploid wheat. They concluded that Aegilops speltoides or a similar diploid species might be a possible B genome donor to tetraploid wheat. Jain et a1. (1975) reported that Ethiopia, the Mediterranean region, and India seemed to be the center of diversity for T. turgidum var. durum afier surveying phenotypic variation of more than Charaders phenol)?“ factOI’S- TI grmflh. Elm Sorrells. 1t; Reli relationship genetic mea crop specie; With regard suggested 21> that groupin, morphologie “her (COP) has ht probability ti allele of the ,\ Parentage rei Period for a 1*; Soybean (Del, 3 more than 3,000 USDA world collection lines. As for different contribution of six characters to the total variation, Jain et a1. (1975) postulated that a different pattern of phenotypic frequencies of a character might be adaptively related to ecogeographical factors. These adaptive characters were largely associated with plant developmental growth, and reported as a main source of variation among germplasm pools (Souza and Sorrells, 1991a; Kato and Yokoyama, 1992). Reliability of morphological characters to resolve genetic diversity and relationships among genotype pairs is generally low. Only partial agreement with other genetic measures based on coancestry or molecular marker traits has been reported in crop species (Smith and Smith, 1989b; Souza and Sorrells, 1991 a, b; Beer et al. 1993). With regard to this, combination of morphological measures with other type of traits was suggested as an alternative genetic measure. For example, Wrigley et al. (1982) found that grouping of 60 Australian wheat cultivars based on both gliadin composition and morphological characters was close to grouping based on pedigrees. When accurate pedigree records of lines are available, coefficient of parentage (COP) has been offered as an indirect measure of genetic diversity. COP is defined as the probability that a random allele from one locus of a genotype is identical to a random allele of the same locus of another genotype by descent (Kempthome, 1969). Many studies using COPs have routinely focused on analyses of amount of genetic diversity, parentage relationship and change of relative importance of ancestor lines over a certain period for a few crop species (barley (Martin et al. 1991), oat (Souza and Sorrells, 1989), soybean (Delannay et al. 1983; Gizlice et al. 1994), and wheat (Cox et al. 1986; Murphy er al. 1986 levels of p relationshi; significant traits of th compared t that genetic yield in a St (.199513150 Peil‘Omtanc As c SCemed to P the genetic 3' Moreover. 5 Similarity “a 1985 Grant n) r: - 4 et al. 1986; Beuningen, 1993; Souza et al. 1994)). Recently, the possibility to predict levels of phenotypic performance of progeny lines was examined by accessing pedigree relationships with important ancestral lines. For example, Beer et al. (1995) found significant genetic contribution of some particular ancestral genotypes to yield and other traits of the group of oat cultivars that was closely related with these ancestors when compared to those of other distantly related groups. Cox and Murphy (1990) reported that genetic divergence based on COP was poorly associated with F 2 heterosis of grain yield in a series of crosses of winter wheats. Cowen and Frey (1987), and Martin et al. (1995) also found COP-based genetic diversity as poor predictor of F1 hybrid performance in oat and wheat, respectively. As compared to molecular marker (MM)-based genetic similarity (GS), COP has seemed to be somewhat inadequate as a measure of genetic diversity because it ignores the genetic proportion identical not by descent but in state (Kempthome, 1969). Moreover, studies reporting poor association between COPs and MM-based genetic similarity (GS) pointed out the assumptions required for calculation of COP (Cox et al. 1985; Graner et al. 1994). These assumptions for inbred lines include (Cox et al. 1986): (i) All genotypes are homozygous and homogeneous, (ii) Male and female parents contribute alleles equally to their progeny, (iii) COP of unrelated ancestors or genotypes with unknown parentages equals 0, (iv) COP between a genotype and itself is 1, (v) COP between a genotype and a line selection from it is 0.75, (vi) COP between two selected lines from a genotype is (0.75)2 = 0.56. Despite this limitation, Messmer et al. (1993) reported highly significant rank correlations between COP and MM-based GS for pairs of flint (F0..- reported si, distance ar: were less r; (1:03? I. and found . (RAPD t-ba only when i In \\ glutenin an “eight (H.\ of the grou; arms Of‘the (Jackson 61 alleles per 13 populaIIOng distribmiOn adaptation genetic rel a: mid “hear I generic (133.3,I I 5 flint (r=0.71) and of dent (r—-0.86) European maize inbreds. Gerdes and Tracy (1994) reported significant relationships (r=0.54**) between DNA marker (RFLP)-based genetic distance and COP-based distance for sweet corn inbreds. However, isozyme markers were less reliable than DNA markers to resolve pedigree-relatedness in their studies (r=0.32*). Tinker et al. (1993) examined all pairs of 27 North American barley cultivars and found a significant, but moderate rank correlation (1:0.51’”) between DNA marker (RAPD)-based GS and COPS. This relationship between two measures was improved only when highly pedigree-related genotype pairs were considered (Tinker et al. 1993). In wheat, two classes of seed storage proteins related with bread-making quality, glutenin and gliadin, have been used as protein-based MM traits. The high molecular weight (HMW) glutenin subunits are encoded by alleles at the Glu-I loci in the long arms of the group 1 chromosomes (Payne et al. 1982), and the gliadin subunits in the short arms of the group 1 (Gli-I loci) and group 6 (Gli-2 loci) chromosomes, respectively (Jackson et al. 1983; Payne, 1987). Levy and Feldman (1988) reported that up to four alleles per locus of HMW glutenins were identified in the tetraploid wild emmer wheat populations (T. turgidum var. dicoccoides) in Israel. They also found that their allelic distribution was significantly associated with selectively regional or microenvironmental adaptation. Cox et al. (1985) analyzed allelic variation of gliadin loci to determine genetic relationships among the US. hard red winter wheat cultivars. As compared to wild wheat population, differentiation of allelic frequencies of gliadin in this wheat cultivar germplasm might be significantly affected by artificially intensive selection and genetic drift during inbred development (Cox et al. 1985). Se by more tl based .\I_\l polymorpi Natural sel different ra Mo made possi protein-hast throughout : al. 1991), ( successful}; quantitatiVe Iher Length POI} are bELSed On remlcllt‘m Cr a513011.35 (B remlctlon e3. Penman, V based Onami r | the de‘EIOpn 6 Several forms of an enzyme encoded by different alleles at a locus (allozyme) or by more than one locus (isozyme) have been extensively used as another class of protein- based MMs. Nevo et al. (1988) and Nevo and Beiles (1989) reported allozyme polymorphisms in wild emmer wheat populations distributed in Israel and Turkey. Natural selection by ecogeographical factors was postulated as the prime cause for different ranges of allelic frequencies at more than 40 putative loci in these populations. More comprehensive genetic studies of particular genes or germplasm pools were made possible by the introduction of a new class of DNA-based MMs. Compared to protein-based MMs, DNA marker loci are highly abundant and uniformly distributed throughout the whole genome and their polymorphism rates are usually high (Paterson et al. 1991). Consequently, MMs assigned in the highly saturated linkage maps have been successfully used to tag particular genes or to localize major and minor loci controlling quantitative traits (QTL) (Melchinger, 1990; Paterson et al. 1991; Stuber, 1992). There are two primary types of DNA-based MMs: RFLPS (Restriction Fragment Length Polymorphisms) and RAPDs (Random Amplified Polymorphic DNAs). RFLPS are based on variation in the length of DNA fragments produced by digestion with restriction enzymes and hybridization with unique or low c0py number DNA sequences as probes (Beckmann and Soller, 1983). This variation results from point mutation of restriction enzyme sites and/or length mutation arising from insertion/deletion events (Beckmann and Soller, 1983; McCouch et al. 1988). Recently developed RAPDs are based on amplification of DNA segments primed by a single short oligonucleotide with the development of polymerase chain reaction (PCR) (Williams et al. 1990). With some practical a artificiall} and Hoisir Pre genetic hit 1990). Ch combinatic pair of six Variation \\ including 3. 311101112 eigh e“lime con On I in a number LUbbCps CI E Iauschii aCC PTObabihn. 3 not identica; | Minorth 1' . | magenta. u .. , “Rage Emu emmer “hel 7 practical advantages over RFLPS in detection of polymorphisms with numerous artificially-made primers, RAPDs have also extensively used (Williams et al. 1990; Ragot and Hoisington, 1993). Preliminary studies surveying polymorphism at the RFLP loci revealed a narrow genetic base in common wheat (Chao et al. 1989; Kam-Morgan et al. 1989; Liu et al. 1990). Chao et al. (1989) found on average 8.7% of cDNA probe-restriction enzyme combinations being polymorphic at the RFLP loci of chromosome group 7 for random pair of six common wheat cultivars. Kam-Morgan et al. (1989) reported that no allelic variation was detected at five out of eight RFLP loci in analysis of 12 hexaploid wheats including 3 synthetic varieties. Liu et al. (1990) also found low frequency of RFLPS among eight common wheats where only 13% of wheat genomic DNA probe-restriction enzyme combinations were polymorphic in a genotype-pairwise analysis. On the other hand, extensive allelic variation at all surveyed RFLP loci was found in a number of accessions of T. tauschii (2n=14, DD) (Kam-Morgan et al. 1989). Lubbers et al. (1991) reported that 80% of RF LP loci were polymorphic among T. tauschii accessions collected from Southwest Asian regions. They computed the probability as an adjusted polymorphic index that two random alleles (or genotypes) are not identical at a given RFLP locus within a pool, which ranged from 0.06 to 0.74 for the polymorphic loci. These greater genetic polymorphisms allowed development of a linkage map of T. tauschii, in which 127 RFLP loci were mapped to seven D genome linkage groups (Gill et al. 1991). As for tetraploid wheat, Mori et al. (1995) reported that emmer wheat (T. dicoccoides, 2n=28, AABB) was more diverse than the Timopheevi wheat (I. L» within a pt‘ also contir: In . maps fore. and other I; Xie et al. 1 using 66 F; loci were rrl developmc. from barlc; anus by us: 'Chinese 8; A10: agmmmlca breeding pr mime“ (Hz; nematode (‘3 wheat. And Significanll} In or 8 wheat (T. araraticum, 2n=28, AAGG) on the basis of RFLP-based genetic distance within a pool. The phylogenetic relationship of emmer wheat with common wheat was also confirmed in their RF LP analysis (Mori et a1. 1995). In spite of significant barrier of low polymorphism in common wheat, linkage maps for each homoeologous chromosome group were developed using RFLP markers and other types of marker traits (Chao et al. 1989; Devos et al. 1992; Devos et al. 1993; Xie et a1. 1993). Liu and Tsunewaki (1991) constructed a linkage map of common wheat using 66 F2 plants from the cross of ‘Chinese Spring’ and T. spelta, in which 197 RFLP loci were mapped with total map size of 1800 cM. Anderson et a1. (1992) reported development of the wheat chromosome arm map on which 210 low-copy DNA clones from barley cDNA, oat cDNA, and wheat genomic DNA were assigned to chromosome arms by using well-established nullisomic-tetrasomic lines and ditelosomic lines of ‘Chinese Spring’. Along with development of linkage maps, some RFLP markers linked to agronomically important genes have enabled marker-assisted selection schemes in wheat breeding programs. Resistance genes for leaf rust (Schachermayr et al., 1994), powdery mildew (Hartl et al. 1993; Ma et al. 1994), Hessian fly (Ma et al. 1993) and cereal cyst nematode (Williams et al. 1994) were tagged with respective RFLP markers in common wheat. Anderson et al. (1993b) also identified eight regions of the wheat genome significantly associated with resistance to preharvest sprouting in the RFLP analysis. In order to identify more polymorphisms and to improve gene mapping studies in common wheat, the introduction of RAPD markers has been strongly suggested (D'Ot'idio of RAPD r (1993) rept electropho: winter whe primers. I] with DGGI it al. (.1994 mapped lo SIS~PCR ? hexaploid .\ ('Ialben et AS I SUCh as bar 19941., mai. 1994) and s I the pIOPOni then offer (r 9 (D’Ovidio et al. 1990; Devos and Gale, 1992). However, some difficulty in application of RAPD markers such as limited repeatability in PCR products in the agarose gel with standard protocol has been reported (He et al.,1992). Devos and Gale (1992) reported that RAPD patterns might be significantly influenced by the PCR process itself. Some modified methodologies have been proposed to improve RAPDs. He et al. (1992) reported that 38% of bands obtained from PCR plus denaturing gradient gel electrophoresis (DGGE, Fisher and Lerman, 1983) were polymorphic among 13 spring or winter wheats and that 90% of genotype pairs revealed polymorphism for a given pair of primers. Dweikat et al. (1993) also reported that RAPD makers from PCR combined with DGGE were reliable to resolve pedigree relatedness among wheat cultivars. Talbert et al. (1994) used specific primers constructed from DNA sequences of previously mapped low-copy RF LP clones to develop ‘sequence-tagged-site’ PCR products. The STS-PCR products digested with a restriction enzyme revealed polymorphism among 20 hexaploid wheat lines, which verified its potential as measures of genetic diversity (Talbert et al. 1994). As measures of genetic diversity and phylogenetic relationship within or between gene pools, the reliability of DNA-based MMs has been reported in a few crop species such as barley (Melchinger et al. 1994), common wheat (Chen et al. 1994; Siedler et al. 1994), maize (Melchinger et al. 1991; Messmer et al. 1993), oat (O’Donoughue et al. 1994) and soybean (Keim et al. 1992). Theoretically, DNA marker-based GS indicates the proportion of alleles at the genomic loci identical by descent and/or in state, which then offer good estimation of diversity of the genetic background (Bemado, 1993; Messme. ordinatic breeding (1991) re groups of (8885) ’ clearly id: in the ana. within a St (O‘Donou sported b; Rel andiof hete related gm] R0eer‘s dis alnong mai low “”613 POStuIated 1 DNA to 333133 10 Messmer et al. 1993). Cultivars or lines grouped by GS-based cluster analysis or ordination analysis were closely related with their common pedigree background, breeding history and/or adaptive ecogeographical origin. For example, Melchinger et al. (1991) reported that principal component analysis of RFLP data distinguished two major groups of 32 maize inbreds from the US. Corn belt, i.e., Iowa Stiff Stalk Synthetic (BSSS) / Reid Yellow Dent and Lancaster Sure Cr0p. O’Donoughue et al. (1994) also clearly identified spring and winter types among North American oat cultivar germplasm in the analysis of RF LP-based genetic distance (GD). The relationships between cultivars within a sub-cluster in dendrogram were further consistent with their common pedigrees (O’Donoughue et al., 1994). Similar patterns of clustering in barley germplasm were reported by Melchinger et al. (1994). Reliability of DNA-based genetic distance to predict F, hybrid performance and/or heterosis of single crosses was not as high as that of clustering genotypes into related groups. Lee et al. (1989) reported that RFLP diversity measured by modified Roger’s distance was moderately associated (r=0.462*) with grain yield of single crosses among maize inbreds. Melchinger et al. (1990) also reported positively significant, but low correlation of RFLP-based Roger’s distance with F 1 performance (r=0.32**) and specific combining ability (r=0.39**) for grain yield of maize. Melchinger et al. (1992) postulated that the range of genetic distance coefficients of parents in crosses might affect the correlation between estimated genetic distance and hybrid performance. In common wheat, Martin et al. (1995) reported that there was no significant rank correlation between DNA marker-based genetic distance and hybrid performance of diallel crosses of hard red spring “It might be t In relationsh. landraces. RFLPS. mt Cosson) at RFLP anal consen'atit future \k hci I. tuust‘bii Unique lam 11 spring wheats obtained by STS-PCR. They also suggested that this low relationship might be closely related with the narrow range of heterosis and estimated GS values. In the present study, we attempted to characterize the genetic diversity and relationships among germplasm pools of common wheat (T. aestivum L.) including landraces, old and current cultivars and breeding lines using different parameters such as RF LPs, morphological traits and COPS. A small set of D genome wheat (T. tauschii Cosson) accessions originated from Southwest Asia and China was also included in the RFLP analysis. The information from our analysis would be valuable to optimize the conservation and utilization of world germplasm collections and finally to facilitate future wheat improvement programs. Moreover, phylogenetic relationships with T. tauschii from Southwest Asia would help to trace the origin of T. tauschii and some unique landraces of hexaploid wheat found in China. Ge till red an. POIE‘morpli Variation g enzymes \\ 55% of pr I (PIC) inde' ”1310f role Wheat gen.- gellOlype P | 1. Estimation of genetic diversity in Eastern US. soft winter wheat based on RFLPS and coefficients of parentage ABSTRACT Genetic diversity of a set of the Eastern US. soft winter wheat germplasm pool (11 red and 11 white wheat genotypes) was assayed using restriction fragment length polymorphism (RFLP) genetic markers and coefficients of parentage (COP). RFLP variation generated by 48 cDNA and gDNA probes combined with three restriction enzymes was limited. On average, 78% of all bands were monomorphic. Fewer than 55% of probes revealed any polymorphism and probe polymorphic information content (PIC) indexes ranged from 0 to 0.73 with a mean of 0.2. Insertions/deletions might play a major role in these DNA polymorphisms. As compared to the soft red winter (SRW) wheat gene pool, the frequency of polymorphisms in the soft white winter (SWW) wheat gene pool was much lower and was dependent upon genotype pairs of specific cultivars or lines; most polymorphism in the SWW wheat gene pool was biased toward the genotype pairs of ‘C4828’, a breeding line developed from Michigan State University. UPGMA cluster analysis using RF LP-based genetic similarity estimates revealed that the SW wheat genotypes were more closely related to each other than the SRW wheat genotypes. 12 CC 0.02 to 0.9 SWW w he variation \i (mean C O with RFLP usociatior SWW whe respective.” l3 COP values for all genotype pairs of the Eastern US. soft wheat ranged from 0.02 to 0.9 with a mean of 0.21. The relatively close pedigree relationships among the SW wheat genotypes (mean COP = 0.51) were consistent with the low genetic variation within this gene pool. The SRW wheat gene pool had more complex parentages (mean COP = 0.15). As measures of genetic relationships, COPS correlated significantly with RFLP-based GS for all genotype pairs (r = 0.73, p < 0.01). However, the associations between the two measures were substantially weak when the SRW and SW wheat gene pools were considered individually (r values of 0.23 and 0.28, respectively). Till and manag det'eloped to practica L'sed as m sampling c dCVelOprng INTRODUCTION The efficiency of crop improvement is influenced by systematic quantification and management of genetic diversity within and among breeding populations. Recently developed DNA-based molecular markers (MM) provide new opportunities with respect to practical plant breeding and genetic studies (Tanksley et al. 1989; Paterson et al. 1991). Used as measures of genetic diversity, DNA markers can lead to more appropriate sampling of germplasm pools and to selection of more genetically distant parents for the development of breeding populations. The utility of MMs is significantly dependent on the frequency and distribution of polymorphisms (Helentjaris et al. 1985; Landry et al. 1987; Miller and Tanksley, 1990). Marked differences have been observed in the frequency of polymorphism in different species and at the different loci within species. For example, relatively high levels of polymorphism have been reported in maize, rice and Brassica species as compared to soybean and common wheat (Helentjaris et al. 1985; McCouch et al. 1988; Song et al. 1988; Liu et al. 1990; Keim et al. 1992). In common wheat (Triticum aestivum L.), lack of adequate genetic diversity (Chao et al. 1989; Kam-Morgan et al. 1989; Liu et a1. 1990) along with complexity derived from its large genome size (i.e., 1.45 x 1010 base pairs per haploid nucleus, Bennett et al. 1982) and polyploidy have made the employment of MMs in wheat breeding program more problematic compared to many other crop species. 14 anal} sis al. 19831 Sorrells. trends in and relati. wheat ger (HRW) gc al. (1986) COP and 1 genetic rel. measured other hand declined fr import/meg 1h. varied dep; IOW C'OII'ELI fOUnd an“ conelatmn EmpIaSm 15 As an alternative measure of genetic similarity, coefficient of parentage (COP) analysis has been commonly applied in a few crop species such as soybean (Delannay et al. 1983), common wheat (Cox et al. 1986; Murphy et al. 1985) and oat (Souza and Sorrells, 1989). This approach has been used to identify important parental lines and trends in genetic diversity over time and space, as well as to quantify genetic diversity and relationships within and between gene pools. In common wheat, the US. red winter wheat germplasm was decomposed into soft red winter (SRW) and hard red winter (HRW) gene pools on the basis of cluster analysis of COPS (Murphy et al. 1985). Cox et al. (1986) monitored the change of genetic diversity of these two gene pools by using COP and its weighted values based on the acreage data of cultivars in a given year. The genetic relatedness of the US. SRW wheat gene pool did not change remarkably when measured by acreage-weighted COP (i.e., from 0.30 in 1919 to 0.22 in 1984). On the other hand, mean acreage-weighted COP within the HRW wheat gene pool significantly declined from 1.0 to 0.4 in the same period, which was primarily due to a reduction in the importance of the plant introduction, ‘Turkey’ (Cox et al. 1986). The relationship between MM and COP based measures of genetic diversity has varied depending on the crop species studied and on the germplasm materials sampled. A low correlation (r = 0.27) between seed storage protein (gliadin)-based GS and COP was found among pairs of US. HRW wheat (Cox et al. 1985). Similarly, poor rank correlation between RFLP-based GS and COP was reported in related winter type cultivars (r = 0.21) and in related spring type cultivars (r = 0.42) of European barley germplasms (Graner et al. 1994). On the other hand, Messmer et al. (1993) found good aSSOCIZII 0.71). at postulate influence sampled 851100119 In soft white COPsth relationsh ““0 di\'ers manipulat' YBgion. l6 association between RFLP-based GS and COP among related pairs of flint corn lines (r = 0.71), and among related pairs of dent corn lines (r= 0.86). These previous reports postulated that the association of two different estimates of genetic diversity might be influenced by the level of selection and genetic drifi, the type of molecular marker loci sampled in the genome, and the level of genetic identity among unrelated pairs of genotypes (i.e., the proportion of alleles alike in state). In the present study, we surveyed 22 wheat lines representing the Eastern US. soft white winter (SWW) wheat and SRW wheat germplasm pools using RFLPS and COPS; (i) to quantify and characterize genetic diversity, (ii) to analyze the genetic relationships within and between gene pools, (iii) to evaluate the correlation between the two diversity measures, and (iv) to find the appropriate strategies for the future manipulation of genetic diversity of soft winter wheat germplasm in the Eastern US. region. .. at q Iw'entj depending on ‘ the breeding p Which are aISo are generally I; released by Mi Pedigree. The and Wisconsin earlier in matur High m Seedling bullt's . micrograms of differem Esme Diem DNA , buffer at 133 3,01 Wiened by Ca MATERIALS AND METHODS Germplasm Twenty-two genotypes of Eastern US. wheats were grouped into two subsets depending on their kernel colors (Table 1). Most SWW genotypes were developed from the breeding programs of New York and Michigan in the US. and Ontario in Canada, which are also the main growing regions of this gene pool. Genotypes of this gene pool are generally late maturity and large grained. ‘Hillsdale’, a soft red winter wheat cultivar released by Michigan State University, was classified as a SWW wheat based on its pedigree. The other cultivars classified as the SRW wheat originated from Indiana, Ohio and Wisconsin. Compared to SWW gene pool, genotypes in this gene pool are generally earlier in maturity and have smaller kernels. RELRanalxais High molecular weight (HMW) DNA was isolated from leaves of 2-3 week-old seedling bulks by a modified CTAB procedure (Saghai-Maroof et al. 1984). Ten micrograms of DNA of each genotype was digested for 16 hours at 37°C with three different restriction enzymes (BamHI, EcoRV and HindIII, Boehringer Mannheim). Digested DNA was electrophoresed on a 0.8% agarose gel in Tris-Borate-EDTA (TBE) buffer at 1.3 volts/cm for 16-18 hrs. The gel-fractionated DNA fragments were then transferred by capillary action to Hybond N+ membranes (Amersham) using 0.4N NaOH 17 Tasle 1 . of reeas 11 lines 1 I Name Augusta 04828 Chesea C5688 C5107 Frauen: Geneseg Germ/a Harris Hilisgaie lonta Aflder Beater Calif-well Char"tar Clark Dynasw \ 18 Table 1. Twenty-two Eastern US. soft wheat lines and their pedigree information, origin and year of release. The first 11 lines from the top belong to soft white winter wheat gene pool and the next 11 lines to soft red winter wheat gene pool. Name Origin Parentage Year Augusta Michigan (Genesee/Redcoat. 82747) // Yorkstar 1979 C4828 Michigan (Genesee/\Mnoka, X0467) /5/ (82141. Suwpn 92/8revor/l N/A 5 Genesee /4/ Norin 10/8revor/lYorkyvin/3/3 Genesee) /6/ (85250, Talbot/Cl 8487 /3/ Genesee 4// Norin 10/ Brevor) Chelsea Michigan (Ledal3/Siete Cerros 66/Cianol/Calidad, SWD71242-16H-01H- 1993 OP)3/4/ 82141 l6/ (85219. Nadadores 63/Yorkstar/5/Cornell 595 2/Redcoat/4/ Norin 10/8revor/IYorkwin/3/ 3 Genesee) 05088 Michigan (Genesee'5/Redcoatl3/Cornell 595‘2/Redcoat/I2' R, 84062) /4/ N/A (84128, Genesee 5/ Purdue 42137/lYorkstar) /5/ (85312, Norin 10/8revor/l Yorkwin/Bl Genesee 4/ Purdue 4217) C5107 Michigan (Norin10/Brevor/lYorkwin/3/2'Genesee/4{Genesee'3/Redcoat. N/A 82218) /5/ (82142, Suwpn 92/8revor/3/3 Genesee/4/Norin 10/8revor/lYorkwin/3/ 3 Genesee) /6/ 85250 Frankenmuth Michigan (Norin10/8revor/Norkwin/3/2°Genesee, A3141) l4/ (A51 15, 1979 Genesee 3/ Redcoat) Genesee New York Yorkwin // (NY530c25-181-4—2, Honor *2/Fowvard) 1951 Geneva New York Ross Selection /3/ (NY5207a8-28-34, BurU/Genesee/CH 2658) 1983 I4/ Genesee Harus Ontario Frederick / Yorkstar 1985 Hillsdale Michigan Asosan/Genesee'4 /6/ (VA 66-54-10, Purdue F4126A9-n32-2 1983 l5/ VahartlFrondoso/Nahart/Cl 1 2658/3/Asosan/4INorin 10/ Brevor) Ionia Michigan Redcoat/3'Genesee 1969 Adder Indiana Abe/3lRedcoat/lKnox 62 sibllDular/4/Knox/l Centenano/Rio 1985 Negro l3/Riley sib /5/ Abe/Caldwell Becker Ohio Hart / Va.66-54-10 1985 Caldwell Indiana Purdue 572483-5P-8-2'2lSiete Cerros 1981 Cardinal Ohio Logan'2/3NA 63-52-12/Logan/I8lue boy 1986 Charrnany \Msconsin Lancota /5/ 2'(Wts. 265. Racine/4IKnox/3/ Brevor/ Norin 10—1// 1984 H483a-3-1-5) Clark Indiana Beau/[Purdue 65256A1-8-1/Purdue 6713785-16/4/ Sullivan /3/ 1987 Beau/l Purdue 551788-5-3-3/Logan Dynasty Ohio BE. 1 -5ILoganl2/Arthurl3/NY5726a8-38-23/ TN 1 403 1 987 19 Table 1 (cont’d). Name Origin Parentage Year P2548 Pioneera , Hadden‘2l3/GA1 123/lNorin 10-8revor/Tenmarq/4/MO W6582/ 1988 Indiana Redcoat/S/Coker 68-15/4/Etoile de Choisy/fThome/Clarkan/B/ Pawnee/Purdue 3848A5 P2550 Pioneer, Coker 68-15/Mo W7510 1982 Indiana P2555 Pioneer, IN4946A4/Hadden/3/KawvaleNigo/lDirectour Journee/4NV521 1986 Indiana Twain Agri Pro Knox 62/SW 14-74 1987 Seed, lnc., Indiana a Pioneer Hi-Bred lntemational, Inc. as the transfer buffer. Membranes were prehybridized with salmon testicle (ST) DNA (5 mg/ml, Sigma) and then hybridized overnight at 65 °C with 32F random labeled probes in a hybridization buffer that was composed of 0.6% SDS, 5% Dextran sulfate, 2.5mM EDTA and 5x Denhardt’s solution. Specific radioactivity of labeled probe was 1.0 x 108- 109 cpm/ug, and its concentration in the hybridization solution was 2.5 x 106 cpm/ml. Membranes were washed four times at 20-minute intervals at 65 °C except the first washing which was done at room temperature. The first and second washings were carried out in 2x SSC and 0.1% SDS, and the third in 1x SSC and 0.05% SDS. After a final washing with 0.5x SSC and 0.025% SDS, membranes were exposed to X-ray film (Kodak) in a cassette with intensifier screens at -80 °C for 1-6 days before development. Membranes were stripped with boiling 0.5% SDS solution following Amersham’s protocol and re-hybridized with additional probes (8-10 total probings per membrane). 20 For RFLP analysis, 48 low-copy DNA clones provided by Dr. M. Sorrells at Cornell University were used. These probes included 34 clones from genomic DNA (WG) of ‘Chinese Spring’ (T. aestivum L.), 12 from cDNA of barley (BCD), and 2 from oat cDNA (CDO) (Table 2). Chromosome arm locations for most probes were reported by Anderson et al. (1992). Probes were selected from each chromosome arm to gain uniform coverage of the wheat genome. Probes were amplified by using PCR prior to labeling and hybridization. E l' l . Pedigrees of each genotype were obtained from several data sources: release notices, documents of wheat genealogy (Zeven, 1976), the Gopher Graingenes internet data base file, and personal communication with wheat breeders. A data file of total 650 parental lines and F1 ’5 was constructed for calculation of COP among the 22 genotypes. COPS were computed for all pairwise genotypes on the basis of following formula (Kempthome, 1969): [l] COPA,B=COPA,CxD=1/2(COPA.C+COPA,D) where COP between the A and B genotypes is equal to the average of COPS between A and the parental lines of B (C and D). The COPS of ancestral lines that are of unknown parentage are normally assigned a value of zero. Other assumptions required for calculation of COP also followed Cox et al. (1986). 0‘. E0: 00.30500 Sm who: 00:353.: 9056) 0.0 .Umxm: 90 .000 mcmm 0.025500 Dem. Emkwamm and xwtt.‘ «0‘1» tattoo COCQZtQS 0.59.05300 Uta $502000. Eta. 023505.020 .ON‘m twat. .mEbECE .03 CgOEZOW mi: E UOmS MOQOCQ Q? \0 gm: OCR N Wit: 21 006 end 96 FF NF NF F N N N.F me 0>> cod and mvd v v F. N N N .50 .43 :26 F.F SN 02, 8.0 8.0 00.0 m m m F F F md .FVN 0>> cod cod cod v v v F F F 40v .mv .5. F.F NFN 0>> vnd omd mNd 9 NF NF m m w . N.o omF 0>> cod cod 8.0 N N N F F F N.F v3 95 8.0 cod 8.0 m m n F F F 9% .wov .49. .49. .m> cod cod 8.0 m m o F F F .EN .mmN rEv va .93. .49 J> 8.0 9.0 96 n v v F N N 4> oFd mmd and m m m N m m F.F 30F 0>> oo.o 0N.o and m N N F v v md NvoF 02F med ovd Nvd NF mF NF m m m 0m .3 .(m N.F oNoF O>> cod cod 8.0 N N N F F F ._25w+ $25? 32% 3mm >>mw >>>>w >>mw 3mm >>>>w 3mm ksmm 3% 25:33 35 39a OK 305 \ mucmn Co .02 39a \ «.5023 .0 .oz .5 5:33 05 as»: 68 mean comm Co Emu aim 829633.65: E9“. 859:8 2m mam; c.2952: mm:_m> OE 622. 2m .08 mama 8:588 new 9888 .8 x85 Gav Emucoo cozmctes oEBoEboa can mcozmog .Em 6880895 .37. toms .m_m>_mcm SB camcsow 2: 5 com: monoa 910 E. of. .N 638. >>>>w 31w 2(1)”... 9 >>Im >253... >>>>w >>~tw >>zm 0253; 25.6 aim ismm III l.l Iflilil l i 0.0. 0205 \ 50:15 NC 02 0205 \ nEb:mQ F0 02 l. 2.. ii a ...| .. |. I . .....:l . i. I... Mala. l .l1. l 68,. 7.1%., » utocmooq 3.: 05m cecal . u 8.4». ...I.r.l 05:00» N Souk 22 68 mean 82 .252, 22? tom .m.: 5993 05 .0 22388 FF ”25>w u .08 9.3 69¢ .953 no. tom .0: Eoummm as .6 206808 FF H>>mw a 2552:: __mEoo 65 E m__m:om ._2 .5 E 8265 295. mm; 986 some F0 5382 EH mEomoEoEo Fo coszaoFE 9; a .265. 9053 9: mm .5 mm mm. w.__mtow .5 E9: 20> 82.233 Soc mam 50m ucm O>>v mmcoB =mEoO 9:8 Co c0380. E5 .6 cozmctoFE 9F . cod 00.0 8.0 N N N F F F .50 Jmm .._> cod and mFd v v v F N N mmN .53 .8me .> end mod 98 m N N N m m 0N .mmN .m> cod mFd 96 v m m F N N woF~ .mmv J> Fwd mod 36 w o m m n v o.F NNN 02, 8.0 cod cod m m n F F F o.F omN O>> oo.o cod 86 m m n F F F m.F NNN 95 cod ood cod N N N F F F 0F oFN 03 and Fed 36 FF FF FF N N N ._n_N JmN .._> 9.0 9.0 mFd m m m N N N E .mmN .w> cod Fmd NF.o N m n F m m mm .mm .> cod ood cod o m o F F F o.F own 02, cod ood cod m m m F F F 0N .me .w> and $6 3.0 m N m N n v .EN JmN .._> cod 86 8.0 N N N F F F .EN .._mN .._> cod ood cod v v v F F F mm .mm .> cod cod cod m m m F F F .5m .3 PF mow O>> 8.0 8.0 cod FF FF FF F F F mF Fov 03 86 96 96 N m n F N N mom .wmm .w>>>w >>mw 3mm >>>>w 31m >>mw 33w 3mm n_>>~.._.w 35 .5333 mum 32a OE 303 F 3:3 .0 .oz 303 F mEozma Fe .02 .8258 N «53 '_i DNA 1 frequency of j formula: [31 where NP = thl gene pool. ant patterns per pr Information C 13‘] where P11 is 111 Genetic Simil; estimated USir [41 Where N\ an d bands in COmr were de‘efmip pair group me COPS N0nna correlation Ir ) 23 DataCQIIectionandAnaleiS DNA bands on autoradiograms were visually scored as present or absent. The frequency of polymorphic probes in each gene pool was determined by the following formula: [2] WM. where N. = the number of probes exhibiting any polymorphism among accessions of a gene pool, and NT = the total number of probes. The number and frequency of band patterns per probe were determined and used for calculation of the Polymorphic Information Content (PIC) Index (Anderson et al. 1993) for each gene pool: [3] PIC, = 1 - 2P7},- where Pij is the frequency among the assayed genotypes of the jth pattern of clone i. Genetic Similarity based on RFLP markers between two genotypes, x and y, was estimated using Nei and Li’s computation (1979): [4] Genetic Similarity (GS),y = 2 ny / (Nx + Ny) where Nx and Ny are the number of bands for each genotype, and ny is the number of bands in common between the two genotypes. Genetic relationships among genotypes were determined by the SAHN cluster routine (Rohlf, 1992) using UPGMA (unweighted pair group method, arithmetic average) based on the RFLP-based GS estimates or on COPS. Normalized Mantel statistic Z (1967) that is equivalent to product-moment correlation (r) was used to access the relationships among the various Similarity matrices. followir [51 where n RFLP-t Those 1; 24 Expected GS estimates were computed from COPS of all genotype pairs by the following formula (Messmer et al. 1993): [5] GS EXP for X-Y = GSo + [(1 ' G50) X COPX-Y] where mean GS (68,) of unrelated lines (COP=0) was estimated by computation of RFLP-based mean GS of 55 accessions of Southwest Asian landraces of T. aestivum. Those landraces were presumed to be unrelated to each other. M Genetic surveying more enzymes. As e three or more i restriction enZ) EcoRV and Ba 78% of all bar LIS. wheats ar revealed 3 hi gl 0f either the 11 [JD 5% of probe There i VS. 280 bands unique to the 5 the 101a] 307 b gene13001. Ge RESULTS ,1. [8131131 . Genetic variation of 22 Eastern US. soft wheat lines was determined by surveying more than 150 RFLP loci with 48 probes and three different restriction enzymes. AS expected for an allohexaploid, most probe DNA sequences hybridized to three or more bands of different sizes when restricted with a given enzyme. The restriction enzyme HindIII generated a total of 307 bands across the 22 genotypes, while EcoRV and BamHI produced 293 and 250 bands, respectively (Table 3). On average, 78% of all bands were monomorphic, which indicated that genetic base of the Eastern US. wheats analyzed in the present study was narrow. The HindIII restriction enzyme revealed a higher frequency of RFLPS than the other two enzymes from the perspective of either the number of bands or the banding patterns per probe (Table 4). A fewer than 55% of probes detected polymorphism for the entire germplasm. There were more bands scored in the SRW gene pool than in the SWW (e.g., 305 vs. 280 bands for HindIII), irrespective of restriction enzyme (Table 3). More bands were unique to the SRW gene pool as compared to the SWW gene pool (Table 3). Only two of the total 307 bands (in the case of HindIII-based RFLPS) were not detected in the SRW gene pool. Generally, the frequency of polymorphic probes in the SRW gene pool was higher than that in the SW gene pool (Table 4). 25 Table 3. Propon 48 probes with d: Eastern US. sofi T R+V= Hmdlll 30? EcoRV 29; SamHl 25.: \ :R+w: SRW + 5 ' Band monomor H\ inum ECORV 0“ .4 83min R+w SRW * 26 Table 3. Proportion of monomorphism and uniqueness among total bands revealed by 48 probes with different restriction enzymes in individual and combined gene pools of Eastern US. soft wheat. Total no. of bands Band monomorphism” No. of unique bands R+WI SRW SWW R+W SRW SWW SRW SWW Hindlll 307 305 280 0.75 0.75 0.89 27 2 EcoRV . 293 292 265 0.77 0.78 0.93 28 1 BamHI 250 249 237 0.82 0.83 0.91 13 1 " R+W: SRW + SWW " Band monomorphism = No. of monomorphic bands / Total no. of bands Table 4. Summary of average values of PIC, no. of banding patterns and no. of bands revealed by each probe with different restriction enzymes in individual and combined gene pools of Eastern US. soft wheat. Frequency of Mean PIC Mean no. of Mean no. of polymorphic probes patterns / probe bands / probe R+W“ SRW SWW R+W SRW SWW R+W SRW SWW R+W SRW SWW Hindlll 0.54 0.54 0.31 0.20 0.24 0.09 2.23 2.10 1.42 6.4 6.4 5.8 EcoRV 0.40 0.40 0.23 0.15 0.19 0.07 2.08 1.96 1.25 6.1 6.1 5.5 BamHl 0.35 0.35 0.25 0.15 0.16 0.08 1.75 1.67 1.31 5.2 5.2 4.9 ‘ R+W: SRW + SWW The lever was estimated b from Hindlll-R1 inIable 2. The I'mean=0.30). a1 “'0 190. BC D 3). The range I mean=0.2~l) \ This indicatec [fifths of the r Table Within the S e“ZYme Cor demoed 0 were m 051 comb-mar“. bob-“mm P001, hm: in a Pair CQnsi d E] 0‘.er a\‘ 27 The level of heterogeneity of RFLP marker states within and between gene pools was estimated by computation of PIC indexes. PIC values for each probe computed from Hindlll-RF LP data are summarized for the two gene pools separately and combined in Table 2. The range of probe PIC indexes for the entire germplasm was 0 to 0.73 (mean=0.20), and eight probes were recognized as being highly polymorphic (PIC 2 0.5): WG 190, BCD 348, BCD 21, WG 645, WG 1042, WG 514, WG 822, BCD 1086 (Table 2). The range of probe PIC index values for the SRW gene pool was 0 to 0.87 (mean=0.24) which was greater than that (0 to 0.50) of the SWW gene pool (Table 2). This indicated that the SRW cultivars were more genetically diverse than SWW lines in terms of the number and frequency of banding patterns per probe. Table 5 contains the summary of a pairwise analysis of RFLPS for all genotypes. Within the SW gene pool, the average relative frequency of polymorphic probe- enzyme combinations for a pair of genotypes was 8.5%. The frequency of polymorphism depended on which genotypes were considered. The SWW genotype pairs with ‘C4828’ were most polymorphic with an average relative frequency of 14% of probe-enzyme combinations for all possible pairs. The other SWW genotype pairs revealed polymorphism with fewer than 12% of probe-enzyme combinations. Within SRW gene pool, however, on average 22.1% of probe-enzyme combinations detected polymorphism in a pair of genotypes. This frequency was relatively independent of genotypes considered. As shown in Table 5, the mean frequencies of probes detecting polymorphism over all possible pairs of the Eastern US. wheat genotypes were 0.21 (HindIII), 0.15 (EcoRV) an by only poll probabilitie: by this mute Accordingl} with each re revealing po In co be detectable afiect the I‘m mutations an PTObes Ru 3 1 01.1101 Eng-mt. when [he 01hr p01imOTPhisr increasfil to 3 When bOLh Of observed fOr I 28 (EcoRV) and 0.16 (BamHI). If DNA polymorphism between two genotypes is generated by only point mutation in the recognition sequence of a restriction enzyme, the probabilities of other restriction enzymes detecting polymorphism will not be influenced by this mutation when they have different recognition sequences (McCouch et al. 1988). Accordingly, the mean frequency of polymorphic probes for a pair of genotypes (Table 5) with each restriction enzyme Should be unrelated to the number of restriction enzyme revealing polymorphism for the corresponding genotype pair. In contrast, polymorphism originating from insertions/deletions is more likely to be detectable with more than one restriction enzyme. This is because length mutations affect the fragments generated by any enzyme whose restriction sites flank both the mutations and the probe locus. AS shown in Fig.1, the relative frequency of polymorphic probes for a given enzyme was not independent of the occurrence of polymorphism with other enzymes. For example, only 12% of probes Showed polymorphism with Hindlll when the other two restriction enzymes (EcoRV and BamHI) did not detect polymorphism for a given pair of genotypes. However, this relative frequency was increased to 37% when either of two other enzymes revealed polymorphism and to 79% when both of the other enzymes revealed polymorphism (Fig. 1). A Similar pattern was observed for EcoRV and BamHI, respectively (Fig. 1). This indicates that insertions and/or deletions play an important role in polymorphism in the Eastern US. soft winter wheat germplasm. RFLP-based GS estimates between pairs of genotypes in the SRW gene pool were greater than the average GS estimate (= 0.984) between genotype pairs of the SWW gene Ucm .mmEo comm F0 mtma mazocmm mEEmOQ =m c2 bquEou memes mecca: m6. tcu aortic... rial}- .. . >n noEcocom cocoa 2:99:38 50 Nocmncmt new mncmn UEQEOENBQ ‘0 59:3: .mbcme Egg \0 $5532 .EOQQ 0:00 2.3QO.6 Ucm UQCEEOU 05 c9 mngawm btszEm ocmcwm cmmF: Ucm mlqlk ‘0 mCOmthEOO mmxéxml .m QQWN 29 .83: E20568 m... ucm 62 .5 88533 was b...m__E_m ozocom cmms. . .IEmm 5.3 $905 mv Fo cozchEoo u m >moom 5.3 monoa NV .6 coschEoo u m. :65: 5E 885 3.6 2.2.2558 n I .0 mmmd Food vmmd mead ommd vmmd mde 03.0 mood . bcm=§m 0.555 :85. moqbocom Fo :2. m .8 Em.F.EoE>.oa mod No.0 9.0 mFd FN.o NN.o oF.o mFd FN.o 9:265 mecca .0 5:262. :85. moabocom F0 :8 m .0. No.0 No.0 mod mod mod mod mod mod No.0 mncmo 0290822. F0 cofoaoa 53.2 39:33 Fe m m m 3 mm v.0. FF 3 2 :8 m .2 2:99:28 8:8 8 .9. :82 moaaocmm oNN mmN NON NmN NmN FNN omN va moN Fo 2.3 a .8 moans .0 .o: :35. m m. I m w I am am. «I .526 >>mw >>>>w + >>mw .moabocmm .o 9.8 .6. 85.8.8 203 mm:_m> camcozw :9... new .390 some .0 2.8 338% 03.32. __m .2 8.3.50 mam; mecca mv ucm moENNcm 5.853. .5556 we... .5 .8996ch mecca 2:89:28 ,6 .6533: new means 2:89:28 .0 PBS? .mucmn .99 Fe .3252 8.8.. mag 9883 new 69.588 9.. co. mmfiéfio NEQEB ozocom cmoE 6cm mafia Fo mcowtmano mm..s._mn. .m mam... 08 8 or .o 2 n. 2 0.61 5 E o 0.5 E 2' 8 04 ‘6 3‘ 03 C O 3 g 02 u. 0.1 0 Figureq' AVE Dali-probe CO r one 0mm 902 30 08 - 0.79 a . 3 0.7 1 .ECIORVl 063 2 IBamHI ‘ 0-52 ' 3' 0.6 « UHindILi .C S 0.5 - E ; 2' 8. 0.4 J. 0.37 “a > 0.3 l 3 I 0.23 0.23 3 . g 0.2 4 l 0.12 0.1 . 0.07 0.07 0 Fin F. Poly w/O Poly w/1 Poly w/2 No. of restriction enzymes detecting polymorphism Figure. 1. Average relative frequency of polymorphic probes for individual enzymes for genotype pair-probe combinations which showed polymorphism with (1) no other enzyme (poly w/O), (2) one other enzyme (poly WM), and (3) two enzymes (poly w/2). pool (Table 6). revealed that th ll. As expecte but was distine panicular subg Varieties were Subgrc depending on matrices (one were consider SRW gene pt (Table 7)_ Bi P001 and Han 3] pool (Table 6). The dendrogram derived from UPGMA clustering of RF LP-based GS revealed that the SWW genotypes grouped together apart from the SRW genotypes (Fig. 2). As expected from genotype-pairwise analysis, ‘C4828’ was close to ‘Hillsdale’, but was distinct from the other SWW genotypes. The SRW genotypes did not form any particular subgroup. ‘P2548’, ‘P2555’, ‘Dynasty’ and ‘Cardinal’ among the SRW varieties were genetically distant from others based on RF LP variation. Subgrouping patterns of genotypes based on estimated GS were slightly different depending on the restriction enzyme. Correlations among the three RF LP-based GS matrices (one for each enzyme) were relatively strong (r .>_ 0.72) when both gene pools were considered together (Table 7). However, r values between GS matrices for the SRW gene pool were relatively low compared to those for the SWW gene pool only (Table 7). Biased distribution of RFLP diversity between genotypes within a SWW gene pool and narrow range of GS estimates might affect different level of associations between GS matrices. $9.53 32:... at- t - wco:mc.oEoo oE>~cm eccoEmm. :EEIécEQ mv >2 15.: 32 owod wood wood NNFd otd mNFd K F .o wwmd o8d oFmd F F F.o oood wood wood woFd ooF .o oood oood woFd oood oF Fd £95k oood omod oood wwod oood Fmod owod oood omod wood Fwod oood owod owod omdd omod owod wmod omod omod mwod oooud oood onod oood mood Food oood wood wood on F .o mood Cod omod owod oood mnod m8d Food mood mnod mood oood ooomd owod wood oood ooFd oF Fd oood oF Fd no F .o ooFd wF Fd NoFd wood wood oood oood mood wood onod nood «nod NF Fd owomm owod oood Nwod owod owmd Food oowd ~F~d oo F .o oowd oowd oF F.o om F .o dwF .o o Fmd med om F .o NoFd Bod oo F .o owod 33:5 owod mood owod wood mwod omFd Food Food oo F .o oowd ooFd oood 09d 5 F .o moF .o on F .d mF Fd wad F.2d umFd wo~d x50 oood wood wood osod mwod wood 9 Fd «o F .o wood Fw F .o Food owod Food oood Rod mood oood Food oood onod oood €2,630 mood Food wood owod oood oood wood wood Fde -~d omwd -Fd mde Nde hood omwd woFd Food Food otd oowd .2..an mood wood oood owod omod oood oood wood o~ F d Rwd o F .o o8d Food wood 53 .o No F .o 3 Fd mm Fd soFd NF Fd soFd =03an mood oood onod Food owod oood oood oood oood Fo F .o Fde oF Fd mnod oood NF F d Fm F .o oood ooFd ode wood otd .oxoom owod owod oood mood oood oood mood oood owod oood on F .o oood oood oood oo F .o mo F .o 9 F .o om F .o ooFd ow F d ooFd 83¢ wood wood Food oood mood owod oood oood oood Kod mood mmwd omod F Fod oood ooud NFwd wood wood onod wFsd 2.5. wood wood Koo owod owod owod oood wood oood Food wood oood oowd Ewd :wd and oowd owmd owmd wood Fsmo 0.25:: wood Food onod oood mood Food oood Fnod oood onod hood wood hood ommd mood oood Cod Nuwd oowd oomd wowd 25: oood mood tod mood owod mood oood mnod oood Rod oood mood oood wood hood wowd oowd med oowd Fomd mmwd 26:00 oood oood Rod mood nood oood osod Koo wood oood mood oood Food oood oood hood oFod Fahd oond omod oo~d 089.60 oood wood oood Food mood owod oood mood owod mnod oood oood mood wood oood sood wnwd oood oood oood mood saEcoxcEm hood wood Food oood Food mood oood hood oood onod oood wood mood oood oood oood oood Nomd nomd mwmd twd cogozo mood oood «Nod Food mood owod mood oood wood wood owod Food mnod Food Rod mood o~o.o oood «mod owod oood BFoo oood wood osod owod Food owod oood mood owod onod oood oood oood oood hood hood oood hood oood oowd hood ooooo oood Fnod Rod owod owod wwod wood wood Food osod mood onod Food Rod ouod «Nod mood Food oood onod oowd omowo wood wood Food oood Food oood oood oood owod oood wood hood oood oood mood osod wood oood oFod wood oood 8263‘ >3 >52: .2 =2; 23 com 536:0 «on 2» £93. ooomd ooowa owowd .35 {so .550 4an .23 5x03 .32 2:2 -mEI mam: «>050 .050 $.an +26 330 ooooo owowo -ao:< .mummEs .253 :8 .w.: £93m. mm 9: .8 cOomEeooE mouofioa co comma cocoomfi 2668 oomdcoema “—0 35.0580 new 20:25:60 9:35 5:053. __ 651-305 ow E ooomamcoo cocoomfi 90an bro-Em ooocmo oommnddm .o mink FiSUte 2 CluSter at 33 0.9?50 0.9?00 0.9050 1.0000 F C4828 Hillsdale Chelsea Ionia Harus Geneva Genesee P—-——-Huqusta F“ C5107 P2550 Becker Calduell [ Tuatn Charmanu Clark Hdder P2548 -————— P2555 1 DunaSIg Lo* Cardinal Figure 2. Dendrogram resulting from the cluster analysis of the RFLP-based Nei and Li’s genetic similarity matrix among 22 Eastern US. soft wheats. GS matrix used for cluster analysis was constructed from RFLPs which were generated from 48 probes and three restriction enzymes. Table 7. Norm developed from the Eastern US Eco-RV BamHI ' p < 0.05; “ 34 Table 7. Normalized Mantel statistic (r) values between similarity matrices developed from RFLPs with three different restriction enzymes and 48 probes for the Eastern US. soft winter wheat germplasms (SRW and SWW). SRW°+SWW" SRW sww Hindlll Econ Hindlll EcoRV Hindlll EcoRV EcoRV 0.723" 0.154NS 0.764“ BamHl 0.715“ 0.762“ 0.158NS 0322* 0573* 0.677“ * p < 0.05; ** p< 0.01; ”3 non-significant l . E 55 . E 3:13; COPS were calculated for 23l genotype pairs of Eastern US. soft wheat lines (Table 6). There were 81 ancestral lines for the 22 Eastern US. soft wheats. COP values ranged from 0.02 (‘Hillsdale’-‘P2555’) to 0.90 (‘Genesee’-‘Ionia’) with a mean of 0.21. Mean COP between SW and SRW gene pools was 0.11. Mean COP within the SWW gene pool was 0.51 with a range from 0.25 for ‘Chelsea’-‘Hillsdale’ to 0.90 for ‘Genesee’-‘Ionia’. ‘Mediterranean’ was the most important (mean COP with the SWW gene pool = 0.23) among the 53 ancestral parent lines of the SWW gene pool. Eleven SWW lines formed a subgroup in the COP-based dendrogram due to their high mutual level of coancestry (Fig. 3). The prominence of the New York-developed cultivar, ‘Genesee’, in the pedigrees of SWW gene pool was remarkable (Tables 1 and 6). Mean COP of ‘Genesee’ with the ten SWW lines was 0.67 which W85 mu» released SW“ and 'Augusta' ‘ developed or re “16 Elk gene pool. Me 0.03 (‘Twain' J lines were ider were the most to the S\\'\\‘ g COP-based d cultivars Wit} example. 13' (AE3), ‘Cla fTOm the pi( 35 which was much higher than that with eleven SRW type cultivars (=0.13). The earlier released SWW type cultivars (before 1980’s) such as ‘Genesee’, ‘Ionia’, ‘Frankenmuth’, and ‘Augusta’ were more genetically related to each other than the other recently developed or released (Fig. 3). The Eastern US. SRW gene pool had more complex parentages than the SWW gene pool. Mean COP within this pool was 0.15 and the range of COP values was from 0.03 (‘Twain’-‘P2555’) to 0.49 (‘Adder’-‘Clark’) (Table 6). Sixty-five ancestral parent lines were identified in this gene pool; ‘Mediterranean’ and ‘Turkey Red (syn. ‘Turkey’)’ were the most predominant, with mean COPS of 0.17 and 0.08, respectively. In contrast to the SW gene pool, the SRW type cultivars did not form a simple subgroup in the COP-based dendrogram (Fig. 3). Nevertheless, grouping patterns of some related cultivars within this gene pool likely corresponded to their developmental origins. For example, ‘Dynasty’ and ‘Cardinal’ were from the Ohio Agricultural Experiment Station (AES), ‘Clark’, ‘Adder’ and ‘Caldwell’ from the Indiana ABS, and ‘P2548’ and ‘P2550’ from the Pioneer Hi-Bred International, Inc. (Fig. 3). B I. l' l EEIE 100E The correlation between RFLP-based GS (from all three restriction enzymes) and COP was moderate (r=0.73, p < 0.01) when all 231 genotype pairs were considered. (Table 8). However, the associations between the two different parameters within the individual SRW and SWW gene pools were substantially lower (r value of 0.23 and 0.28, respectively). RFLP-based GS estimates are plotted against COP values in Figure 4. F‘QUl’e 3. | am00S! 22 36 0.00 0.25 0.50 0.25 1.00 C4828 C5088 C5107 _ r—r::'0“*a Genesee F -——-[———‘——_— FY 8111: anUIh Augusta Harus Geneva Chelsea Hillsdale Dunasru f——l Cardinal Clark __£—1 Fidder CaldueH Tuarn D2518 {—_L 92550 Becker Charmanu P2555 Figure 3. Dendrogram resulting from the cluster analysis of coefficients of parentage among 22 Eastern US. soft winter wheats. Most genotype the plot along : between the S“ ranges of C OI correlation of GS was dilfere based on EcoR other “NO festr Expfietr Mean RFLRb 0.905, Which ‘ descem (.080) the COHFSpon GSExr: “as b “’0 Parameb Eastern L‘ .S The Somhm “one the . developed. 37 Most genotype pairs within the SRW gene pool were distributed only on the left side of the plot along the COP axis. In contrast, genotype pairs within the SWW gene pool or between the SWW and SRW gene pools were distributed throughout the plot. A wide ranges of COPs compared with the narrow ranges of RFLP-based GS caused a low correlation of the GS and COP matrices within either gene pool (Table 8). RFLP-based GS was differently correlated with COP depending on restriction enzyme; GS values based on EcoRV-RFLP data showed higher correlation with COP than those based on other two restriction enzymes (Table 8). Expected GS estimates were calculated by using the equation [5] and COP values. Mean RFLP-based GS for 55 unrelated landrace lines from Southwest Asia (COP=0) was 0.905, which was used as an estimate of the proportion of alleles alike in state not by descent (GSO). Expected GS values (GSEXP) plotted against COPs were compared with the corresponding RFLP-based GS ((35033) for a pair of genotypes (Fig. 4). Generally, GSEXP was below the level of 65033 along the COPs and significant differences between two parameters were observed as COP declined (Fig. 4). This suggests that GSo for the Eastern US. soft wheat gene pool may be higher than the estimate of GS0 derived from the Southwest Asian landraces. That may result from a higher degree of relatedness among the dominant ancestors from which the Eastern US. soft wheat gene pool was developed. o_n_mmoo __m do. 3 ococmfi BEE). noN=m .momchoemQ oo FCQUEUOU ncm maqlm co booms omoEmE btmtEN Eco: \3 nmEFEmomn mama 35¢.qu ecqtchi 3:: m ococmo twosome qumcoomEm .o mBmF 38 9:38 8858.. :08 .6 8 £8 :8: 829:8 mafia Eo: 859:8 5:9: 38:88 oomcmo 9:38 88:89 iEmm 5:5 888 ow 3 8:88: mafia Eco 859:8 xEmE 38.86 08:8 9:38 8858: >m8m 5:3 888 ow E 8.88: mafia Eco 8:588 58.: 38:86 08:8 mE>~8 88:89 :65: 53> 888 ow E 8.88: mafia Eo: 859:8 xEmE 38:88 ooocmo Emoo_:o_w-:o: mz ”Fod vo : Hood v a . H.001! mzomd mzomd .Nwd mzood 53 5.23650 some some : :3 some- as 8&8 2>¢m :28 :08 :03 2.8.0 $5.88 __< .158 >88 :65: __< .158 >m8m =65: U__< 0.18m n>88 .___e:_x 33m 5% 2>mw .mooEoE >::m__E_m ozocmo 883-8288“: *0 E20580 5:5 889:8 9oz, moE>~co m __o oo “8 Son 8:568 __m Eco .28 2153 98 .>m8m .__=u:=.c oE>Nco 885mm: :08 oo .8 Ban Fcoaotfi E0: 8059: 38:87... ozocmo oommnddm .208 ocoo down; :25; :8 .w.: Eoummm Fo $8 2988 __m :2 S 0888 .385. 8~__oFEoc >9 8:_E8Fmo m8; 8:88.: E2320 025 F0 mcooofitoo .mmooocoama oo E30580 ocm mnjmm :o 8me 888.: 58.86 oomcoo c828: 82808.0”. .o 28... Genetnc SH 10c .. 0990 090» 0 097+. use»! 095 > 094.» 093» 092 0. 0914» 090‘ FjQure 4. vs. coeffic genOtYDes i RiFLiP-gasegési R? = 0.53 Genetic Similarity 1.00 T 0.99 1' 0.98 <- o.97 «- 0.96 + 0.95 -- of 8 0.94 w- ’0‘" 0.93 ‘- o.92 ‘- o.91 r 0.90 i t 4. + 9 . . . . . 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Coefficient of parentage Figure 4. The plot of RFLP-based genetic similarity (GSOBS) and expected GS (GSEXP) vs. coefficients of parentage for 231 pairs of the Eastern US. soft winter wheat genotypes. DNA whmtv Very i the Eastern L' in common 0 extensive Rt Wh‘morphit Over all acce The 1031 to \x'hic sequences 1 revealed f0 the A- B. 21 Th: r EVEaled. COmDOSiti. 6~ba3e rec in 50me pj (Helentj at DISCUSSION Very low polymorphism was found with Southern blot analysis among 22 lines of the Eastern US. soft winter wheat germplasm. This is commensurate with other results in common wheat (Chao et al.1989; Kam-Morgan et al. 1989; Liu et al. 1990). In an extensive RFLP analysis of European wheat and spelt lines, only 11% of loci were polymorphic in pairwise comparisons, even though 44% of RFLP loci were polymorphic over all accessions (Siedler et al. 1994). The number of bands revealed by RFLP analysis is a function of the number of loci to which a given probe DNA sequence hybridizes and the position of restriction sequences flanking such loci. In our results and others, three or more bands were revealed for most probes, which indicates differentiation of homoeologous probe loci in the A, B, and D genomes of common wheat (Anderson et al. 1992). There were differences among restriction enzymes in how many RFLPS they revealed. The level of polymorphism is likely to be related with the number and composition of nucleotides in the recognition sequence of restriction enzymes; A-T rich 6-base recognizing restriction enzymes have been reported to detect more polymorphism in some plant species as compared to G-C rich 4-base recognizing restriction enzymes (Helentjaris et al. 1985; McCouch et al. 1988). Chao et al. (1989) also found that the ones with A-T rich recognition sequence among thirteen 6-base restriction enzymes 40 pro lt‘C pit! l.\l pol p05 the CXC POl dix 41 produced more RFLPS. In our study, BamHI restriction enzyme with less A-T in its recognition sequence produced lower polymorphism than Hindlll and EcoRV. If point mutation is the major source of RF LPs, detection of polymorphism by restriction enzymes will be independent with one another. As revealed in the present study (Fig. 1), however, the frequency of polymorphism of one restriction enzyme is a function of the number of restriction enzymes that detect polymorphisms with a given probe. This is evidence of insertion-deletion events as a major cause of polymorphism (McCouch et al. 1988). The discovery of transposable element-like structures in a highly polymorphic DNA clone constructed from T. aestivum (Liu et al. 1992) provides a possible mechanism for generation of insertions/deletions in common wheat. However, the relatively low correlation coefficients observed among some GS matrices can not exclude the possibility of point mutation-derived polymorphism in this study (Table 7). 2.1!. A l'e “1' f. P O 0.. Q . . '"0-0 .0. I "'0' N '0 . 0 . N , -. .. - _ . .- pools As revealed in the coefficient of parentage analysis, the Eastern US. SWW gene pool has been developed from a narrow genetic background, which results in low genetic diversity. A relatively small number of varieties which dominated cultivated hectares in the late 1940’s to the middle 1970’s have been recurrently used as breeding parents for development of the SWW gene pool (Patterson and Allan, 1981). Mean COPS of the SW gene pool analyzed here with these predominant varieties are 0.49 (‘Yorkwin’), 0.24 (‘Cornell 595’), 0.67 (‘Genesee’) and 0.58 (‘Yorkstar’). 42 The predominance of a few elite cultivars as breeding parents is also common in the SRW gene pool development. However, the genetic base of the SRW gene pool is wider than that of the SWW gene pool, primarily because of efforts to enhance disease and insect resistances. Major contributors classified on the basis of growth habits are: (1) hard red winter type (‘Malakot’ , ‘Hussar’, and ‘Hungarian’), (2) semihard red winter type (‘Kawvale’), (3) hard red spring type (‘Hope’, ‘Frondoso’, ‘Fronteira’, ‘Chinese’, ‘Transfer’, ‘Wisconsin 245’, ‘Kenya Farmer’, and ‘Supresa’), and (4) soft red winter type (‘Mediterranean’, ‘Fultz selection’, ‘Wabash’ and ‘Trumbull’) (for review, see Patterson and Allan. 1981). These source germplasms and the resulting recombined elite varieties such as ‘Redcoat’, ‘Knox’, ‘Arthur’, ‘Benhur’, ‘Lucas’, and ‘Coker 68-15’ have been commonly used as breeding parents for the development of the SRW gene pool (Murphy et al. 1986). As compared to the SWW gene pool, the broader genetic background of the SRW gene pool led to greater RFLP-based genetic diversity in the present study. t- ., u .. rm“. u... u -o-. 'a. -uc. 0' u‘-._ .. ‘ o q. ‘ The different range of RFLP-based GS estimates and COPS within and between the SRW and SWW gene pools seemed to influence the relationship between the two measures of genetic relationship. There was a significant correlation of two different measures of genetic relationships for all genotype pairs of the Eastern US. soft winter wheat lines. However, genetic relationships revealed by the two diversity measures were not always consistent, as revealed from their poor correlation when an individual gene 43 pool was considered (Table 8). Consequently, the relationship between RF LP-based GS and COPS might be systematically biased due to sampling effect. AS reviewed in other reports (Cox et al. 1985; Graner et al. 1994; Messmer et al. 1993), impractical assumptions required for computation of COPS may also be as source of the poor relationship of the two diversity measures. For example, alleles are not always transmitted equally from female and male parents to progeny. Also, released varieties or lines as breeding parents are not always completely homozygous and homogeneous. Moreover, unrelated ancestral lines may have unknown amount of alleles alike in state, so that COPS underestimate the true GS. As Cox et al. (1985) and Graner et al. (1994) postulated, poor relationships between the two measures may result partially from high level of GS between unrelated genotypes. In fact, the calculated GSo from the Southwest Asian landrace population is 0.905, while the estimated GSo from regression of RFLP-based (35033 to COPS for 231 genotype pairs of the Eastern US. soft winter wheat is approximately 0.96 (Fig. 4). This suggests that the actual GSo value is probably high for ancestral population of Eastern US. soft winter wheat germplasm pool. Our results also revealed that the range of RFLP-based GS was confined irrespective of the range of COP. The unknown effects of selection and genetic drift on the allelic frequency changes during inbred development may reduce the reliability of COPS as measures of genetic relationship. For example, transmission of alleles, especially those controlling qualitative traits with high heritability, is evidently influenced by intensive selection 44 pressure in the breeding program, which results in biased contribution from one parent with favorable alleles to the progeny generation (Cox et al. 1985; Souza and Sorrells, 1989). GS estimates based on RFLP data reflect existent genomic differentiation among lines for the assayed loci. Identical banding states may indicate that alleles at the assayed loci are identical by descent or alike in state, therefore, RFLP-based GS is an upwardly biased estimate of COP (Bemado, 1993; Messmer et al. 1993). Meanwhile, some co- migrating RFLP fragments are genuinely not the same, i.e., they arise from distinct loci but have the same size by chance. In allopolyploids like common wheat, the probability of this event is increased because of the chance that two or more homoeologous loci possess identical alleles. RFLP-based GS estimates can also be biased by errors from inadequate genomic sampling and misclassification of banding patterns (Lynch, 1988). Use of an appropriate number of probes and the ability of the probes to resolve polymorphism are critical determinants to estimate true GS. The Eastern US. SWW wheat gene pool exhibits very little RFLP diversity and high levels of pedigree relatedness. RFLP diversity within the Eastern US. SRW wheat gene pool is also low, but more putative loci are polymorphic than in the SWW gene pool. The near equivalence of SRW gene pool in performance (R. Ward, personal communication) along with its relatively great genetic diversity, points to its utility as a tool in the expansion of the germplasm base of SWW gene pool. II. Evolutionary aspects of RFLP-based genetic diversity in T riticum tauschii Cosson and four landrace pools of T. aestivum L. from China ABSTRACT The genetic nature and phylogenetic relationships of Triticum tauschii and hexaploid landrace wheats found in China were characterized with RFLP analysis. The Chinese gene pool of T. tauschii Cosson (2n=2x=l4, DD) and four landraces of T. aestivum L. (2n=6x=42, AABBDD) exhibited narrow genetic variation. An eastward decline of diversity and uniqueness in RF LP characteristics from Southwest Asia was apparent for these Chinese gene pools. As compared to the Southwest Asian gene pool, the Chinese T. tauschii accessions were highly homogeneous with a low frequency of polymorphic bands (16%) and banding patterns with fewer than 30% of the total 75 RF LP probe-Hindlll combinations. T. tauschii accessions from Afghanistan and Pakistan were genetically more similar to the Chinese T. tauschii than those from Iran. The widest range of genetic diversity was found in the Iran gene pool. Of 368 bands found for 39 accessions of the Chinese hexaploid wheats with 63 RF LP probe-Hindlll combinations, 28.3% were polymorphic with on average 2.6 banding patterns per probe and 5.0 bands per genotype. Tibetan Weedrace was genetically more diverse than Sichuan White wheat and Yunnan Hulled wheat with which it was clustered in UPGMA analysis. On the basis 45 46 of RF LP-based genetic Similarity, ‘Chinese Spring’ is a close relative of ‘Chengdu- guang-tou’, a famous landrace cultivar of Sichuan White wheat. Xinjiang Rice Wheat was genetically distinct from other Chinese wheat landraces. Chinese wheat landraces were more related to Southwest Asian T. tauschii than Chinese T. tauschii. V) Cy, INTRODUCTION The emphasis on core collection of germplasms and their potential for crop improvement has led to increased interest in the genetic diversity of wheat (Triticum aestivum L.) gene pools from China. Of the several landraces of wheat recognized in China, four have unique morphological traits: (1) Yunnan Hulled wheat (King, 1959) which is locally referred to as ‘Tiekemai’ or iron glume wheat, (2) Tibetan weedrace (Shao, 1983) known as ‘Duansuimai’ or fragile spike wheat, (3) Xinjiang Rice wheat (Udatsin and Miguschova, 1970) known as ‘Daosuimai’ or rice-head wheat, and (4) the Sichuan White wheat complex, which is the likely origin of the cultivar ‘Chinese Spring’ (Yen et al 1988). Several morphological and cytogenetic studies have described the origin and evolution of these four landrace pools (Shao et al. 1983; Chen et al. 1985; Chen et al. 1988; Yen et al. 1988; Yang et al. 1992). On the basis of chromosome pairing of F 1 hybrids, Riley et al. (1967) reported that the chromosomal structure of ‘Chinese Spring’ is not different from that of putative ancestral wheat species, T. turgidum L. ssp. dicoccoides (2n=28, AABB) and T. tauschii (2n=14, DD). The Yunnan, Tibet, and Xinjiang races also appeared to have a primitive chromosomal constitution for the D genome (Yen et al. 1988; Yang et al. 1992). One or two chromosomes (especially in the B genome) distinguish ‘Chinese Spring’ from members of the Yunnan, Tibet and Xinjiang races (Chen et al. 1988; Yang et al. 1992). Chen et al. (1 988) reported that chromosomes of the Yunnan race were less differentiated 47 48 from ‘Chinese Spring’ than those of the Tibet and Xinjiang races. Xinjiang Rice wheat’s polish wheat-like spike morphology and its relatively greater chromosomal differentiation led Yang et al. (1992) to suggest that they may have originated through an independent hexaploidization event from the group of the Tibetan Weedrace, Yunnan Hulled, and Sichuan White wheats. Chen et al. (1988) and Chen et al. (1985) also suggested introgression from T. polonicum or some other wild emmer wheat as the mechanism that led to the differentiation of these two groups. The discovery of Chinese T. tauschii has also led to the hypothesis that some Chinese hexaploid landraces might have originated from natural hybridization between cultivated emmer wheats and Chinese T. tauschii (Yen et al. 1983; Yang et al. 1992). Together with their Spatial separation from the Middle East center of origin of wheat, the morphological and cytological evidence suggest that the four Chinese landraces under discussion represent a unique set of wheat germplasm. In the present study, we surveyed a sample of Chinese T. tauschii and four landrace pools of hexaploid wheat using RFLP markers (i) to estimate genetic diversity and genetic relationships within and among Chinese gene pools, (ii) to examine the genetic relationships of Chinese gene pools with other gene pools from the center of genetic diversity of Triticeae, (iii) to clarify further evidence regarding the origin of four landraces of Chinese wheat in view of phylogenetic relationship with T. tauschii. The bottleneck enforced by its recent origin by hexaploidization renders wheat a relatively narrow species and heightens the importance of the resolution of any new genetic heterogeneity. Identification of groups of wheat that trace to distinct hexaploidization 49 events would be particularly valuable since such groups are likely to be highly polymorphic in comparison to each other. MATERIALS AND METHODS Siennnlasm Triticum tauschii (DD) Ten accessions of Chinese T. tauschii were used in the RF LP analysis. Nine accessions from the Shaanxi and Henan provinces of China were collected by J.L. Yang and B.R. Lu of the Triticeae Research Institute, Sichuan Agricultural University. One accession from natural vegetation of Xinjiang was collected by Profs. C. Yen and. J. L. Yang (Table 1 and Fig. 1). Ten accessions of T. tauschii collected from Southwest Asia were provided by Dr. B.S. Gill at the Kansas State University. These accessions represent five regions: Caspian Iran (4 accessions), northwestern Iran (1 accession), eastern Afghanistan (3 accessions), western Afghanistan (1 accession) and Pakistan (1 accession) (Table 1 and Fig. l). Triticum aestivum (AABBDD) Thirty-seven accessions of the Chinese landraces of T. aestivum were provided by Prof. C. Yen (Table 1). These accessions including ‘Chinese Spring’ are members of the four “special landraces” from China (Fig. l): (1) Sichuan White Wheat Complex, (2) Yunnan Hulled Wheat, (3) Tibetan Weedrace, and (4) Xinjiang Rice wheat. None of 50 51 these races has been found outside China. The Sichuan White Wheat Complex includes ‘Chinese Spring’ and is comprised of cultivated common wheats characterized by multifloret Spikelets and rounded glumes (Yen et al. 1988). The Yunnan Wheat accessions have hard glumes and are therefore difficult to thresh. Although they have brittle rachis like T. spelta, the Yunnan wheat accessions have the wedge type of disarticulation, whereas T. spelta has the barrel type (Shao et al., 1983; Tsunewaki et al., 1990). These Yunnan wheats are found in the valleys of the Nujiang and the Lanchangjiang Rivers in Yunnan province of China. The Tibetan Weedrace accessions are unique among wheats in that they have a naturally broken rachis with wedge-type disarticulation and consequently exhibit Spontaneous disarticulation of the rachis in the field (Shao et al., 1983; Tsunewaki et al., 1990). This race is found as a weed in fields of wheat and barley mixtures in Tibet. Rice Wheat accessions are found only in southwestern Xinjiang and their spikes are very similar to those of tetraploid “Polish Wheat” (T. turgidum L. var. polonicum). This unusual landrace was classified as a distinct subspecies (T. aestivum ssp. petropavlovsky) by Russian germplasm collectors (Udatsin and Miguschova, 1970). In order to examine the genetic relationships with other gene pools from different origins, RFLP-based genetic distances with 55 accessions of hexaploid wheat landraces from Turkey, Afghanistan and Iran were also analyzed. These landrace germplasm pools were obtained from the National Small Grain Collections, USDA, Idaho. T. aestivum ssp. typicum cv. Hongmangmai provided by Prof. C. Yen that is representative of the common wheat cultivars currently grown in China was also included in this study. 52 Table 1. The list of germplasms of T. tauschii and four Chinese landraces of common wheat. I. Triticum tauschii (DD) China 1? .. SI . ll 'AS 71 AS 74 AS 77 AS 81 AS 75 AS 79 AS 82 AS 76 AS 80 AS 83 Southwest Asia EAIQHamSILn Qasnianltan "TA 2371 TA 2378 TA 2413 TA 2454 TA 2533 TA 2470 TA 2529 mm Milan TA 2436 TA 2492 II. Triticum aestivum (AABBDD) W Chinese Spring Chengdu-Guang-Tou (AS 489) Chengdu-Guang-Tou, S-1 (AS 2228) Peng'an White wheat (AS 742) Wanxian White wheat (AS 743) Yongchuan White wheat (AS 745) Changning White wheat (AS 746) Guanghan White Wheat (AS 749) Rongjing White wheat (AS 750) Leshan Qianqian White wheat (AS 751) Tongjiang White wheat (AS 754) Yilong white wheat (AS 755) Langzhong White wheat (AS 764) Iugnan Hulled wheat (ssp. zunnanensis) AS 331 AS 332 AS 333 AS 334 AS 335 AS 336 AS 337 AS 338 AS 339 AS 340 Balustan TA 2379 Xinjiang Rig Wheat (ssp. petropav/ovskl) Akesu Rice wheat (AS 356) Akesu Rice wheat (AS 357) Tulufan Spring wheat (AS 358) Yutian Rice wheat (AS 360) Yutian Rice wheat (AS 361) Luopu Rice wheat (AS 362) Wushen Rice wheat (AS 363) Moyu Rice wheat (AS 364) W (ssp. tibetanum) AS 329 AS 330 AS 907 AS 908 AS 1025 AS 1026 AS 1027 hin h ul iv cv. Hongmangmai (AS 1727) ' AS # = ID. of accessions in the Triticeae Research Institute, Sichuan Agricultural University, China b TA # = ID. of accessions in the Wheat Genetics Resource Center, Kansas State University, USA 53 Caspian Sea Figure 1. Geographical regions where accessions of T. tauschii and four Chinese landraces of T. aestivum were collected. 1, Xinjiang; 2, Shaanxi; 3, Henan; 4, Sichuan; 5, Tibet (syn. Xizang); 6, Yunnan; 7, Pakistan; 8, east Afghanistan; 9, west Afghanistan; 10, Caspian Iran; 11, northwest Iran. 54 RELEanalxsis Southern blot analysis for RFLPS was performed as described in the previous chapter. For DNA digestion, only one restriction enzyme, Hindlll was used in the present study. Prolxs A set of thirty clones from genomic DNA of T. tauschii was provided by Dr. B.S. Gill at the Kansas State University (KSU) (Gill et al. 1991). Genomic DNA inserts of KSU clones were amplified using bacterial culture and plasmid miniprep routine (Maniatis et al. 1989). The list of KSU clones including their insert Size, genomic location, and probe usage information is presented in Table 2. All 30 KSU clones were used to probe T tauschii gene pool, whereas 18 of them were selectively used as probes for the Chinese wheat landrace pools. F orty-five cDNA and genomic DNA clones (Cornell) used in the previous study (see Table 2 in Chapter I) were employed again except WG 296, WG 401 and WG 645 as probes for both T. tauschii gene pool and landrace pools of Chinese wheat. i ll . l l . The methodologies for RFLP data collection and evaluation of polymorphism were performed as described in the previous chapter. Genetic relationships within and between gene pools were estimated by two methods: (1) SAHN clustering of the matrix of RFLP-based genetic Similarity estimates using UPGMA, and (2) principal coordinate 55 Table 2. List of a set of 30 KSU clones for RFLP analysis. Genomic DNA inserts were derived from T. tauschii. Individual clones were selectively used as probes for T. tauschii gene pool and/or four landraces of Chinese hexaploid wheat. Probe Gene pool Source of Insert Genome Genome applied " clone Size location" of locationb of (Kb) T. tauschii T. aestivum A1 T, CW KSU 1.2 7D 7DS A3 T, CW KSU 1.4 50 A5 T, CW KSU 1.5 7D 7DL D16 T, CW KSU 0.9 SD 10 D17 T. CW KSU 1.2 6D 6DL D19 T KSU 1.6 3D 30S DZ T, CW KSU 1 70 022 T. CW KSU 1.5 20 20L 027 T, CW KSU 1.5 6D D30 T KSU 1 SD D7 T KSU 1.4 7D 3D D8 T KSU 1.5 20 ZD E16 T, CW KSU 1.2 ZD E18 T KSU 1.5 10 E19 T. CW KSU 1.3 10 E6 T. CW KSU 1.9 40 EB T KSU 1.8 10 E9 T. CW KSU 1.7 4D F15 T KSU 0.8 ZD F41 T KSU 0.9 20 20 F48 T, CW KSU 0.7 70 F8 T KSU 1.2 4D G12 T. CW KSU 1.35 5D,7D G14 T. CW KSU 1.55 SD G34 T, CW KSU 0.9 1D G48 T. CW KSU 1.6 60 GS T KSU 1.5 ZD 20 G8 T, CW KSU 1.3 6D H16 T KSU 0.4 ZD H7 T KSU 1.5 30 * Gene pool applied: Gene pool which was probed, T: T. tauschii gene pool CW: T. aestivum gene pool from China " b The information of chromosome arm location of each clone was kindly provided by Dr. Gill’s lab at the Kansas State University. analysis based on Similarity matrix of RF LP markers in order to develop a scatterplot with accessions of each gene pool. RESULTS AND DISCUSSION EEIE 1' £1: E" Analysis of the RFLP patterns generated with seventy-five probes revealed a total of 309 bands, 46.9% of which were monomorphic across the entire set of T. tauschii. There were on average 2.7 RFLP banding patterns per probe and 2.9 bands per probe/accession combination. The genetic base of the Chinese T. tauschii assessed by RFLPS was substantially narrow (Fig. 2 and Table 3). Fewer than 30% of probes detected polymorphism (Fig. 2) and the number of RFLP banding patterns per probe (mean=1.33) was low. The range of PIC (Polymorphic Information Content, see equation [3] in the previous chapter I for computation) index values was 0 to 0.59 with a mean of 0.11 (Table 3). About 16% of the 233 bands scored were polymorphic, and only 9 bands were unique to the Chinese T. tauschii. On the other hand, more RF LP variation was observed in the Southwest Asian T. tauschii. Among 300 distinct bands scored, 50% of them were polymorphic and 70 bands were unique to the Southwest Asian gene pool. F ifty-four probes (72%) detected polymorphism within a pool (Fig. 2), giving on average 2.5 RFLP banding patterns per 56 57 probe with a mean PIC index of 0.4 (Table 3). As Lubbers et al. (1991) reported, the accessions from Iran had more genetic diversity than the others in this study. There were Frequency of polymorphic probes (%) 90.0 .KSU .CORNELL DALLl Alizx CHN sw All sw YH rw XR Asia 6X T. tauschii gene pools T. aestivum gene pools Figure 2. Frequency of probes revealing polymorphism in separate and combined gene pools of T. tauschii and T. aestivum. (Note: CHN = T. tauschii from China, SW Asia = T. tauschii from Southwest Asia, All2X = CHN + SW Asia, SW = Sichuan White wheat complex, YH = Yunnan Hulled wheat, TW = Tibetan weedrace, XR = Xinjiang Rice wheat, and All6X = SW + YH + TW + XR). 58 Table 3. RF LP banding characteristics revealed by 75 probes and the Hindlll restriction enzyme for T. tauschii gene pools. Number of distinct RF LP banding patterns and number of distinct bands produced by each probe were scored, and Polymorphic Information Content (PIC) index were calculated for separate and combined gene pools. PIC No. of patterns No. of bands Probe Alla CHN” SWAc All CHN SWA All CHN SWA A1 0.375 0.000 0.500 2 1 2 9 8 9 A3 0.481 0.000 0.640 3 1 3 5 4 5 A5 0.480 0.000 0.320 2 1 2 1 0 1 D16 0.000 0.000 0.000 1 1 1 5 5 5 D17 0.000 0.000 0.000 1 1 1 1 1 1 019 0.600 0.343 0.660 4 2 3 7 6 7 DZ 0.445 0.000 0.580 3 1 3 3 1 3 D22 0.000 0.000 0.000 1 1 1 1 1 1 D27 0.533 0.000 0.649 3 1 3 4 2 4 D30 0.000 0.000 0.000 1 1 1 5 5 5 D7 0.660 0.000 0.673 4 1 4 7 6 7 D8 0.428 0.000 0.493 2 1 2 3 3 3 E16 0.555 0.420 0.500 3 2 2 5 3 4 E18 0.696 0.589 0.620 5 3 3 4 3 3 E19 0.481 0.000 0.640 4 1 4 5 4 5 E6 0.255 0.420 0.000 2 2 1 2 2 1 E8 0.565 0.420 0.640 3 2 3 8 7 8 E9 0.348 0.000 0.565 3 1 3 5 4 5 F 15 0.797 0.500 0.660 6 2 4 7 6 6 F41 0.625 0.000 0.700 4 1 4 6 2 6 F48 0.180 0.000 0.320 2 1 2 2 1 2 F8 0.340 0.000 0.560 3 1 3 3 1 3 G12 0.720 0.540 0.700 6 3 4 12 10 10 G14 0.180 0.000 0.320 2 1 2 5 4 5 G34 0.375 0.000 0.500 2 1 2 6 5 6 G48 0.000 0.000 0.000 1 1 1 1 1 1 GS 0.000 0.000 0.000 1 1 1 5 5 5 GB 0.710 0.340 0.680 6 3 5 5 3 5 H16 0.625 0.320 0.620 3 2 3 3 2 3 H7 0.395 0.000 0.580 3 1 3 3 1 3 BCD 1066 0.000 0.000 0.000 1 1 1 1 1 1 BCD 1069 0.000 0.000 0.000 1 1 1 2 2 2 BCD 1086 0.455 0.320 0.500 2 2 2 2 2 2 BCD 120 0.180 0.000 0.320 2 1 2 3 2 3 BCD 1230 0.340 0.000 0.560 3 1 3 4 2 4 BCD 1278 0.640 0.420 0.700 5 2 4 9 5 9 BCD 21 0.320 0.000 0.480 2 1 2 3 2 3 BCD 327 0.000 0.000 0.000 1 1 1 2 2 2 BCD 348 0.420 0.000 0.680 5 1 5 6 2 6 BCD 386 0.505 0.000 0.620 3 1 3 3 2 3 r}. We“ 59 Table 3 (Cont’d). PIC values No. of patterns No. of bands Probe All“ CHN" SWA“ All CHN SWA All CHN SWA BCD 442 0.420 0.320 0.460 2 2 2 2 2 2 BCD 606 0.575 0.343 0.640 3 2 3 4 4 4 one 1396 0.160 0.000 0.320 2 1 2 3 2 3 000 920 0.000 0.000 0.000 1 1 1 4 4 4 we 1026 0.000 0.000 0.000 1 1 1 4 4 4 we 1042 0.180 0.320 0.000 2 2 1 2 2 2 we 1044 0.203 0.000 0.367 3 1 3 3 2 3 we 114 0.000 0.000 0.000 1 1 1 1 1 1 we 160 0.525 0.000 0.700 4 1 4 3 1 3 we 181 0.625 0.320 0.620 3 2 3 4 3 4 we 164 0.755 0.420 0.740 6 2 5 9 9 6 we 190 0.675 0.000 0.700 5 1 4 11 7 10 we 212 0.000 0.000 0.000 1 1 1 2 2 2 we 241 0.555 0.000 0.620 3 1 3 3 3 3 we 266 0.000 0.000 0.000 1 1 1 1 1 1 we 341 0.375 0.000 0.500 2 1 2 2 1 2 we 363 0.000 0.000 0.000 1 1 1 1 1 1 we 405 0.605 0.180 0.800 6 2 6 11 8 11 we 419 0.405 0.320 0.460 3 2 3 2 2 2 we 466 0.375 0.000 0.500 2 1 2 2 1 2 we 514 0.375 0.500 0.000 2 2 1 2 2 1 we 522 0.395 0.000 0.560 3 1 3 5 3 5 we 530 0.645 0.320 0.700 4 2 4 10 7 10 we 583 0.160 0.000 0.320 2 1 2 2 1 2 we 605 0.000 0.000 0.000 1 1 1 2 2 2 we 669 0.415 0.000 0.660 4 1 4 7 4 7 we 666 0.375 0.000 0.500 2 1 2 5 4 5 we 710 0.095 0.000 0.160 2 1 2 6 5 6 we 727 0.000 0.000 0.000 1 1 1 5 5 5 we 750 0.000 0.000 0.000 1 1 1 1 1 1 we 622 0.375 0.000 0.500 2 1 2 4 3 4 we 676 0.550 0.340 0.660 6 3 5 6 5 6 we 9 0.180 0.000 0.320 2 1 2 2 1 2 we 909 0.405 0.000 0.620 3 1 3 3 1 3 we 933 0.255 0.000 0.420 2 1 2 2 1 2 " All = all 20 accessions (CHN and SWA) of T. tauschii " CHN = 10 accessions of T. tauschii from China ° SWA = 10 accessions of T. tauschii from Southwest Asia 60 1.68 RFLP banding patterns per probe in the Iran gene pool, which was greater than those of accessions from Afghanistan (1.39) and China. In spite of the small number of accessions, about 60% of banding patterns in the Iran gene pool did not appear in other gene pools including the Chinese T. tauschii. Only 16.3% and 16% of the RFLP patterns were unique to the Afghanistan and Chinese gene pools, respectively. The different range of RFLP variation between the Chinese and Southwest Asian T. tauschii pools may be associated with heterogeneity based on the frequencies of taxonomic subspecies within a pool as well as based on ecogeographical habitats from which each gene pool was collected. T. tauschii can be divided into two sub-species, ssp. eusquarrosa and ssp. strangulata in terms of spike morphology; the latter has moniliform shape in which its rachis segment is curved, and is also longer and narrower than the adjacent spikelet (Kimber and F eldman, l987). Two accessions from Iran (TA 2454 and TA 2470) analyzed in this study belong to ssp. strangulata. In fact, the distribution of ssp. strangulata seems to be restricted to near the Caspian Sea region (Lubbers et a1. 1991). Lubbers et al. (1991) attributed the high level of genetic diversity in this region to the distribution of various taxonomic subgroups of T. tauschii. As for the Chinese T. tauschii, however, no accession belonging to ssp. strangulata has been recognized in China (Y. Yen, personal communication, 1994). UPGMA cluster analysis based on the GS coefficient matrix generated two isolated subclusters: one for the accessions from Iran and the other for those from China, Afghanistan and Pakistan (Fig. 3). Differentiation of these two subgroups was explained with 53.1% of the total variation that was quantified by the principal coordinate analysis [helm iRo‘m‘i apps lies: pron M35 in fro 61 based on the GS matrix. In this analysis, only the first principal coordinate that separated the Iran accessions from others was statistically significant under the broken-stick model (Rohlf, 1992). Close genetic relationships among the accessions of Chinese T. tauschii were apparent in the dendrogram derived from UPGMA cluster analysis of GS matrix (Fig. 3). Most accessions collected from the Yellow River valley region (Shaanxi and Henan provinces) were grouped as a subcluster (Fig. 3). The high similarities among Chinese accessions at RFLP loci are consistent with the monomorphism previously reported at the esterase isozyme loci (Yen et al. 1983) or at the spacer region of rDNA at the Nor locus (Lagudah et al. 1991b). As expected, the accessions of Chinese T. tauschii were genetically close to those from the adjacent countries such as Afghanistan and Pakistan. Five accessions from these regions were subclustered to the Chinese group in the dendrogram (Fig. 3). As for genetic relationships among gene pools, Lagudah et al. (1991b) previously reported that there was a single rDNA banding pattern among accessions of T. tauschii from Afghanistan, Pakistan and China after surveying length variation of spacer region of rDNA at the Nor locus. This pattern (the spacer length type 2) was predominant (60%) in T. tauschii ssp. eusqarrosa var. lypica and indicated complete homogeneity of these three gene pools at the Nor-D3 locus (Lagudah et al. 1991b). RFLP-based genetic similarity (GS) between accessions within a pool was inverser correlated with the geographical distance from the Middle East center of 62 d1versrty (1.6., GS Iran/Iran < GS Afghanistan/Afghanistan < GS Chinese T. tauschii/Chinese T. tauschii) (Table 4). The trends of eastward decline of genetic diversity from the geographic center of T. tauschii and close genetic relatedness with adjacent gene pools in the present study indicate that the Chinese T. tauschii may originate from Caspian Sea region following the “founder effect” model (Mayr, 1942). According to this model, development of populations at new habitats may be initiated by a small number of individuals (i.e., the founders) from the source population. The genetic nature of newly colonized populations must have been characterized by only a portion of the total genetic variation of their origin. Consequently, limited genetic variation within new populations is expected when compared to that of the origin. As for the origin of the Chinese T. tauschii, it is likely that a small subset of individuals from the adjacent gene pools in the West Asia was introduced and colonized as founders. Additionally, Yen et a1. (1983) presumed from isozyme gel electrophoresis that T. tauschii growing as a weedrace in the habitat of the Yellow River region originated from the natural vegetation of T. tauschii in Xinjiang. 63 Table 4. Estimated mean, minimum and maximum values of genetic similarity coefficients within and between gene pools. Genetic Similarity‘ Gene Pool Mean Minimum Maximum Within T. tauschii from China (CHN) 0.9693 0.926 1.000 7'. tauschii from South West Asia (SWA) 0.8482 0.759 0.963 T. taus. (AFG)‘ 0.9470 0.940 0.957 T. taus. (IRN)b 0.9120 0.858 0.963 T. taus. (PAK)c --- --- -- T. aestivum from China 0.9560 0.907 0.998 Sichuan White W. (SW) 0.9721 0.948 0.998 Yunnan Hulled W. (Y H) 0.9784 0.971 0.997 Tibetan Weedrace (TW) 0.9603 0.943 0.997 Xinjiang Rice W. (XR) 0.9691 0.947 0.987 BeMeen Chinese T. taus. and SW Asia T. taus. 0.8602 0.751 0.970 Chinese T. taus. and T. taus. (AFG) 0.9366 0.899 0.970 Chinese T. taus. and T. taus. (IRN) 0.7866 0.751 0.821 Chinese T. taus. and T. taus. (PAK) 0.9445 0.920 0.958 Chinese T. aestivum SW and YH 0.9640 0.946 0.980 SW and TW 0.9617 0.936 0.982 SW and XR 0.9379 0.907 0.964 YH and TW 0.9621 0.939 0.990 YH and XR 0.9366 0.917 0.967 TW and XR 0.9360 0.920 0.967 * Genetic similarity was determined by the measure of Nei and Li (1979) on the basis of RFLP data. a 4 accessions of T. tauschii collected from Afghanistan " 5 accessions of T. tauschii collected from Iran ° 1 accession of T. tauschii collected from Pakistan 64 0.7500 0.8125 0.6250 0.9375 1.0000 as71 fiS338(YH) 68907(Tu) fiS333(YH) '__T fiS337(YH) RS335(YH) fiS334(YH) l [ fiS336 0.26): WG 1042, BCD 348, BCD 21, WG 190, WG 822, WG 1026, CD0 920, WG 341, WG 686, BCD 1086, and WG 514. All these probes except BCD 1086 revealed more bands per genotype than the others. The relationship between the mean PIC values and the number of bands generated by each of 30 probes was highly significant (r = 0.68") over all 21 germplasm pools. For BCD 442, WG 181, and WG 583, more multilocus variations were observed in the germplasm pools from Southwest Asia as compared to the advanced germplasm pools and the Chinese landraces. RFLP variation for WG 363 was not detected except in the Eastern US. wheat cultivars (U S_ER and US_EW) (Table 5). Genetic similarity (GS) coefficients in all 42,486 pairs of the 292 accessions calculated on the basis of restriction bands ranged from 0.844 [PI 243650 (IRAN) - ‘Freedom’ (U S_ER)] to 1.0 for several pairs of accessions [PI 243767 (IRAN) - PI 243781 (IRAN), PI 222682 (IRAN) - PI 243637 (IRAN), AS 360 (CHN_XR) - AS 361 (CHN_XR), ‘Tam 105’ (U S_GP and IWWSN) - ‘Tam 107’ (U S_GP), and 093-44/ 99 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000 0>> 50.0 0.10 F F.0 F00 00.0 00.0 0V0 0F.0 0 F .0 FN.0 0F.0 0N0 00.0 0 F .0 0F.0 0V0 00.0 V5.0 00.0 0N0 0 F .0 000 0>> 00.0 00.0 00.0 00.0 0F.0 00.0 Fv.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0N0 00.0 00.0 00.0 00.0 00.0 00.0 0 0>> 00.0 0.40 0v.0 00.0 00.0 0Y0 00.0 00.0 N00 0V0 v0.0 05.0 05.0 00.0 00.0 Nvd 50.0 0V0 05.0 #00 50.0 NNO 0>> 0N0 00.0 000 5.4.0 0F.0 0Y0 0v.0 00.0 0F.0 00.0 Nvd 0Y0 0V0 Nvd 00.0 00.0 Fv.0 00.0 0Y0 0N0 0F.0 000 O>> FF.0 00.0 00.0 00.0 00.0 00.0 0N0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 F~.0 Fv.0 0F.0 00.0 000 0>> F F.0 00.0 F F.0 F F.0 00.0 00.0 00.0 00.0 0 F .0 00.0 0 F .0 00.0 00.0 00.0 5V0 0V0 0V0 0V0 FN.0 00.0 00.0 mg .95 0N0 00.0 34.0 00.0 00.0 FN.0 V0.0 0V0 0v.0 0F.0 0V0 0N0 0F.0 v0.0 00.0 00.0 00.0 00.0 Fv.0 0F.0 0F.0 va 0>> 00.0 00.0 00.0 00.0 00.0 FF.0 :0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000 0>> FF.0 00.0 00.0 0Y0 0F.0 FN.0 0Y0 00.0 0F.0 00.0 Nvd 0F.0 No.0 0v.0 00.0 00.0 0V0 00.0 0v.0 0N.0 00.0 Fvn 0>> 00.0 0N0 00.0 F F.0 0N0 F00 0V0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0 F .0 00.0 Fv.0 00.0 00.0 5N.0 00.0 00w 0>> 00.0 00.0 N50 N00 00.0 00.0 05.0 05.0 00.0 FN.0 00.0 F5.0 N00 00.0 00.0 00.0 50.0 00.0 00.0 05.0 05.0 00F 0>> 0N0 0N0 NN.0 N00 0F.0 FN.0 00.0 00.0 0F.0 00.0 0F.0 00.0 0F.0 00.0 0F.0 «40.0 v~.0 F~.0 FF.0 0F.0 00.0 #3 O>> 00.0 00.0 00.0 FF.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 FOF 02F 0N0 0N0 50.0 FN.0 0N0 00.0 0F.0 00.0 500 N00 0F.0 00.0 00.0 0F.0 0F.0 00.0 00.0 00.0 00.0 00.0 00.0 00F 0>> 00.0 00.0 00.0 00.0 00.0 00.0 0F.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 :40 FN.0 FF.0 00.0 0F.0 VFF 0>> 00.0 05.0 N00 00.0 F5.0 00.0 00.0 F00 50.0 55.0 00.0 00.0 F400 00.0 00.0 F400 N00 00.0 50.0 00.0 N00 NvoF 0>> F50 F00 05.0 N00 50 V00 0.0 05.0 .00 FN.0 0V0 00.0 00.0 00.0 00.0 0.40 50.0 0V0 00.0 0v.0 N00 0NOF 0>> 00.0 3.0 00.0 0v.0 3.0 N00 00.0 .00 F00 00.0 V0.0 000 0F.0 NF..0 0N0 00.0 34.0 FN.0 05.0 05.0 F5.0 0N0 000 00.0 00.0 FF.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0F.0 00.0 00.0 00.0 00.0 FF.0 00.0 0F.0 80F 000 F F.0 00.0 F F.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 3.0 00.0 Fv.0 00.0 0Y0 0v.0 0V0 N3 00m F F.0 3.0 F F.0 50.0 0N0 NN.0 Fv.0 00.0 0 F .0 00.0 00.0 00.0 0 F .0 00.0 540 00.0 0V0 00.0 0Y0 N10 5V0 000 000 N00 00.0 0.0 00.0 05.0 NN.0 00.0 00.0 00.0 00.0 05.0 00.0 00.0 05.0 00.0 00.0 05.0 00.0 05.0 00 N00 0F.0 now 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0F.0 0F.0 00.0 00.0 00.0 00.0 0F.0 00.0 00.0 00.0 00.0 F F.0 00.0 00.0 5N0 000 0.0 N50 V5.0 00.0 0Y0 9.0 05.0 v0.0 F00 00.0 00.0 00.0 0V0 00.0 0V0 V00 N00 00.0 F00 05.0 00.0 FN 000 00.0 00.0 0V0 00.0 00.0 Fm.0 VNO 00.0 00.0 00.0 00.0 0 F .0 00.0 00.0 0N0 00.0 00.0 3.0 F F.0 00.0 00.0 05NF Dom FF.0 00.0 00.0 00.0 00.0 00.0 00.0 0F.0 00.0 FN.0 00.0 00.0 0F.0 00.0 0N0 00.0 00.0 00.0 00.0 0F.0 00.0 00NF 000 00.0 00.0 00.0 FF.0 00.0 00.0 F00 00.0 00.0 00.0 00.0 00.0 00.0 0F.0 0F.0 00.0 00.0 00.0 00.0 00.0 00.0 0NF 00m FF.0 3.0 00.0 F00 vwd FN.0 00.0 0F.0 00.0 0V0 00.0 00.0 0F.0 00.0 00.0 0F.0 vwd 00.0 0V0 N10 0N0 002 com 0N0 3.0 F F.0 NN.0 00.0 00.0 vmd 00.0 0 F .0 00.0 Nvd 00.0 3.0 0 F .0 00.0 00.0 00.0 00.0 F F.0 0F.0 00.0 003 000 >> Q0 :02 >>w mm (mm >>w I> 2F mx Zm>>>>_ Om< m: w: w: w: 03 «GO (an. 03> xx: 03m 20m woo ZIO ZIO 2:0 2:0 m3... z xonE OE .0 can... 100 (IWWSN?) - ES84/(IWWSN10)]. The mean GS within the TUR gene pool was lowest (0.925) among the 21 germplasm pools (Appendix A). Mean GS coefficients of all possible genotype pairs within and between germplasm pools are presented in Appendix B. Generally, larger mean GS values were observed within a pool than between gene pools for Afghanistan, Iran, China and Eastern European regions as well as in the Eastern US. SWW wheat germplasms (U S_EW and US_MSU) (see Appendix B). This reflects higher levels of homogeneity existing in these germplasms as compared to others. The detailed genetic relationships among accessions and their germplasm pools are revealed by dendrograrns resulting from the cluster analysis of RFLP-based GS among all 292 accessions (Fig. 1) or of mean GS values (see Appendix B) within and between germplasm pools (Fig. 2). The goodness of fit of the cluster analysis was tested with the correlation between the RFLP-based GS matrix and the clustered data-derived cophenetic value matrix. As for the clustering of 292 accessions based on RFLP-GS, the degree of fit was very poor but statistically significant (r = 0.68**). This indicates that classifications of some accessions resulting from cluster analysis do not reflect true genetic relationships. Since the possibility of biased clustering of accessions existed in the dendrogram, the analyzed data should be interpreted with caution. Generally, accessions closely related to each other due to their common geographical origins and/or high level of coancestry had a tendency to be in the same cluster. There were also several subgroups within germplasm pools (Fig. 1). This trend of clustering was most apparent in germplasm pools with high mean GS within a pool, such as the US_EW (0.966), US_MSU (0.963), CHN_XR (0.959), CHN_SW (0.959), 101 and CHN_YH (0.964) germplasm pools (Fig. 1). Most accessions from these germplasm pools formed distinct clusters depending on their origins. About 80% of accessions from both Afghanistan and Iran were close together forming a large subcluster. In spite of its relatively high heterogeneity within a pool, six out of 16 accessions of the TUR germplasm pool fell into a single subgroup. There was no particularly distinct subgroup for the advanced lines originating from Eastern European regions (Russia, Ukraine, Romania, Yugoslavia and Odessa). This was due to the fact that mean GS values within a pool and between pools in the Eastern European wheat germplasm were almost equivalent. Therefore, a large subcluster of the Eastern European wheat lines was formed. This subcluster included about 65% (1 1/ 17) of HRW wheat cultivars in the US_GP. A small number of accessions from other germplasm pools such as the IWWSN, the US_ER, the FRA, the GER and the US_W pools were also partially associated with the Eastern European wheat cultivars. Several small subclusters of accessions with similar coancestries were apparent for the FRA, the GER and the US_W pools, respectively. However, the SRW cultivars and breeding lines in the Eastern US. did not form any distinct subgroup in the dendrogram. This pattern of clustering reflects a substantially diverse genetic background of the US_ER pool. The genetic relationships among germplasm pools were summarized with UPGMA clustering based on the mean GS within and between pools (Fig. 2). The goodness of fit for this clustering was significant (r = 0.89"). Several subgroupings of germplasm pools were notable because of their genetic relatedness depending on geographical proximity. These were (i) Afghanistan and Iran landraces, (ii) Romania, 102 Figure 1. Dendrogram resulting from the cluster analysis of RFLP-based genetic similarity estimates among 292 accessions of common wheat germplasms. Some accessions are labeled with their own cultivar names together with the source of germplasm pools, but others are identified with Pl/Cl or breeding line numbers (Table 1). (Note: AFG - Afghanistan; IRAN - Iran; TUR - Turkey; CHN_XR - Xinjiang Rice wheat (China); CHN_TW - Tibetan Weedrace (China); CHN_YH - Yunnan Hulled wheat (China); CHN_SW - Sichuan White wheat (China); ROM - Romania; RUS - Russia; UKR - Ukraine; YUG - Yugoslavia; FRA - France; GER - Germany; US_ER - Eastern US. SRW wheat; US_EW - Eastern US. SWW wheat; US_MSU - SWW wheat breeding lines from Michigan State University; US_GP - U.S. Great Plains; US_W - Western US. SWW wheat; ARG - Argentina; ODESSA - wheat breeding lines from Odessa research station; IVWVSN - wheat breeding lines from International Vlfinter Wheat Screening Nursery) 103 0.6675 0.9250 0-9925 1.0000 ,——F 91470434<100> 1 PI311629 ,gngFF DI341740<1UP> L PIl67772 ____{_-{ 91268311(IQN) Pll?3501(TUD) P1172563(TUQ) PI382002 l l Ukrarnka Barbarna)a(DUS) DI372345 Moldova D2510 Km0r(US_U) DI350678 ______1_'i Karl(US_GD) Tualn(US_[P) i1 CarstenUHI f—TAC: LOU29/[IIJLISN111 C126-[IHNSN21] [ DI372126(PUS) [ODESSH_7I Odesskaj652(UKD __l [ [ODESSH_11 . Tomahauk(US_GD) L0vr1n_2l(QON) DI345691(QUS) P‘— BLLlIUHSN3l l [ODESSH_18] Hrbor _____(————______l 01166866<109> OH498 Elq1n(US_U) Qelg Cordral 441-— Ullmor1n27(FDR) PI171682(FQH) r-Stephens(US_U) ————_{ l——~ru-301 Solo,50(HRG) PI315990 He1ne,UlI(GER) Palmaress(FQfi) 91180584(GER) Nakeiield(USED) Pl36718l(9F6) PI317011 l__ PI367IOO PIZ15554<9FG) Pl220125(9FG) P1367062(6FG) 91367116<9F01 “—l l: 91367075<0r01 91367047rerc> ,.91213761 cs ”—{:iEE 63-469 NE87615m_ 60:02 05 mm 0 new _m>o_ 6026. 65 mm 0 cmoéon 0%: 2o; $.QO 05am N .68 0:00 0 £53, 3008 205650 800 3230 So; 2800300 050530 F 8 5.0 H F.~ 8% 0.0 H F0 :9 0.0 H 0.v0F 6.86 5.0 H 0.00 .8 «.0 H vF 66 NN H 0.5F 0F zw>>>>_ a 0F H 5.0 6.8 5.0 H 0.0 ; 0F H 0.NOF 88 0.0 H 0.00 :96 0 H 0F .66 0F H 0.5F 0 9?. u 0.0 H N0 00.5. H 0.5 6 0F H 0.00F 96 5.0 H 5.05 86 F0 H 0F on FF H 5.5F 0F 2,100 8 F.F H 0.0 .6 0.0 H 0.5 :0 vF H 0.~0F 9% 0.5 H 0.05 3 F0 H 0F F0.F H N.0F 5F 0010: 8 0.0 H 0.0 m 0.0 H 0.9 an 0.0 H 0.00F 6 NF H 0.3 .8 F0 H VF 68 0.0 H 0.0F 0 00.210: 6 «.0 H 0.0 an 0.0 H 0.0 96 0F H F.00F an 0.0 H 0.00 :98 F.0 H vF co 0F H 0.0F 0F >510: u 0.0 H N0 .68 5.0 H 0.0 3 F0 H 0.09 .8 0.0 H 0.00 9% 0 H 4F .8 N0 H 0.5F 5F «.010: u F.0H F.0 60F H0.0F «0.0H005F 8 0.5H0.00 a ~.0H5.F 6v.~H5.0N 0F «00 u v.0 H F0 6 0.0 H 0.9 a 0.4 H N.N5F 96 0.5 H 0.05 6 F0 H 0F 8 F0 H 0.F~ vm <0“. 8 0.0 H v.0 .68 0.0 H 4.0 :9 0.0 H ~.v0F 9 0.0F H 5.05 68 0.0 H 4F 68 5F H N.0F 0 00> 86 0.0 H 0F m 0.0 H F.0F o 0.0 H 0.00F on 0.0 H 5.00 :96 0 H FF 8 0F H Now NF 55 Ba NF H 0F 88 NF H 0.0 68 0.0 H 0.00F a NFF H 0.40 an 0.0 H vF 8 0.0 H 0.0F 0F 00m 66 0.0 H «N 8 5.0 H 5.0 :9 0.0 H N.v0F 68 0.0F H 0.00 :96 F0 H 0F 68 5F H F.0F 0F .29... 8 0.0 H 4.0 0% 0.0 H 0.0 c 5.0 H 0.~0F m 0.0 H F05 on 0 H 0F 68 F.F H 0.9 0F «00000 a v.0 H 0.0 an 0.0 H F0 6 0.0 H 0.00F 86 0.0 H F00 5 0 H 0F 68 0.0 H NOF mm «5... a 0.0 H 0.5 86 0F H 0.0 8 0.F H 5.50F one 0.0 H 0.00 :96 0.0 H F.F 68 0F H 0.2 F0 23: a 0.0 H 0.0 66 0F H «.0 68 5.N H F.50F 86 F0 H v.00 :9 0 H 0F 68 0.0 H F.0F F0 0”? A63 03 8.. €60 €60 is 505. E0: £855 £06; £23 £05. mco_mmooo< .000 ch_0uo._ exam 9 960 205 Re. 00:. 0mm. 00E .0 .oz Emma—E00 AMOS; EOE—too Fo £000 Emma—Emu 5F E 002300 mgmz .mo_0o_oca._oE Fo row H memos: 300950 20053 0cm 2me .0 0.00... 114 Table 7. Analyses of variance for six quantitative traits in 17 germplasm pools of common wheat. Germplasm No. of Flag Flag Plant Days Spike Lodging Pool1 Accessions leaf leaf height to length length width (cm) anthesis (cm) (cm) (cm) AFG 31 it t it *t *4! *4! IRAN 31 NS “I it it it *4 TUR 22 it 41 NS 4* it 18* ODESSA 10 NS NS NS "‘ * NS ROM 13 NS NS " ** * NS RUS 13 I” “I IIHII *4! #4! NS UKR 12 NS NS " " " * YUG 9 NS * ** ** ** NS FRA 24 IIHII NS I” till it NS GER 15 ** ** NS ** ** NS US_ER 17 n: 4: an 4:4- 4:4: 414: US_EW 10 NS NS " * ** NS US_MSU 3 NS NS NS NS NS NS US_GP 17 * NS “‘ " NS " US_W 16 NS NS ** * ** ** ARG 6 NS NS * ** " NS IWWSN 16 * "' NS ** ** ** * P< 0.05; ** P< 0.01; NS Non-Significant 1 AFG - Afghanistan; IRAN - Iran; TUR - Turkey; ROM - Romania; RUS - Russia; UKR - Ukraine; YUG - Yugoslavia; FRA - France; GER - Germany; US_ER - Eastern US. SRW wheat; US_EW - Eastern US. SVWV wheat; US_MSU - SVWV wheat breeding lines from Michigan State University; US_GP - U.S. Great Plains; US_W - Western US. SWW wheat; ARG - Argentina; ODESSA - wheat breeding lines from Odessa research station; IVWVSN - wheat breeding lines from International Winter Wheat Screening Nursery 115 days to anthesis (Appendix C). Clustering of 265 accessions based on the morphological trait-based standard taxonomic distance matrix is shown in Fig. 4. The goodness of fit for this cluster analysis was poor (r = 0.70") indicating that some subclusters of accessions were not informative. Only a fraction of the accessions from certain germplasm pools with similar geographical origins or highly related genetic background tended to group together, leading to several small subgroups in the dendrogram. This pattern was especially apparent for (i) the Eastern U.S. SWW wheat cultivars and breeding lines (U S_EW and US_MSU), (ii) the Western U.S. soft white wheat cultivars and breeding lines (U S_W), (iii) the French and German wheat cultivars and (iv) the Russian and Ukraine wheat cultivars. Most of the landrace accessions from Southwest Asia commonly revealed very poor performance for the agronomic traits in the field. The phenotypic similarities within/between pools of Southwest Asia led to formation of a relatively large subcluster. Seventeen accessions (77.3%) from Turkey, 28 accessions (90.3%) from Iran and 23 accessions (74.2%) from Afghanistan were included in this subcluster. On the other hand, some Southwest Asian landrace accessions, i.e., l accession (‘Yumuzak’) from Turkey, 6 accessions (PI 367146, Pl 367100, PI 367075, PI 366984, PI 268465 and PI 221482) from Afghanistan and 2 accessions (PI 382070 and PI 243793) from Iran revealed close relationships with ‘Elgin’ and its derivative cultivars/lines of the Western U.S. soft white wheat pool (U S_W) (Fig. 4). They were taxonomically ssp. compactum containing the same elliptical spike types. 116 Most of the French and German wheat cultivars were distantly related with other germplasm pools. Only a few number of cultivars from Russia, Yugoslavia and Romania were closely associated with them within the subcluster (Fig. 4). The French and German cultivars contained in this subcluster showed relatively late maturing and high seed sterility in our experimental field in Michigan (East Lansing). The genetic relationship among germplasm pools based on morphological/ agronomic traits might be summarized with the cluster analysis based on the mean standard taxonomic distance within and between germplasm pools (Fig. 5). The goodness of fit for the classification from cluster analysis was highly significant (r = 0.89"). The associations of particular germplasm pools are shown in Fig. 5: (i) the Southwest Asian landrace pool (Afghanistan, Iran and Turkey), (ii) the Eastern U.S. SWW wheat pool (US_EW and US_MSU), (iii) the Western European wheat pool (France and Germany) and (iv) Russia and Ukraine wheat pool. Based on the clustering patterns of morphological traits, it appears unlikely that phenotypic diversity in the field can fully represent genotypic diversity. For all possible pairs of the 219 accessions used in both RFLP and morphological trait analyses, the r value between the GS and GD matrices was -0.15 NS. The correlation between the mean taxonomic distance estimates and the RFLP-based mean GS estimates as measures of the genetic relationships among 17 germplasm pools was significant, but its value was not high (r = -0.46**). In addition, the mean genetic distance estimates based on band frequencies or banding pattern frequencies were poorly correlated with the mean taxonomic distance estimates, respectively (r = 0.40” and 0.36"). This indicates that 117 the standard taxonomic distance based on morphological traits is not commensurate with RF LP-based genetic (dis)similarity. It can not therefore be expected that sampling of germplasm based on phenotypic performance per se lead to adequate strategies for the conservation and utilization of genetic diversity. 118 Figure 4. Dendrogram resulting from the cluster analysis of morphological trait-based taxonomic distance estimates among 265 accessions of common wheat germplasms. Some accessions are labeled with their own cultivar names together with the source of germplasm pools, but the others are referred to by their Pl/Cl or breeding line numbers (Table 1). (Note: AFG - Afghanistan; IRAN - Iran; TUR - Turkey; ROM - Romania; RUS - Russia; UKR - Ukraine; YUG - Yugoslavia; FRA - France; GER - Germany; US_ER - Eastern U.S. SRW wheat; US_EW - Eastern U.S. SWW wheat; US_MSU - SWW wheat breeding lines from Michigan State University; US_GP - U.S. Great Plains; US_W - Western U.S. SWW wheat; ARG - Argentina; ODESSA - wheat breeding lines from Odessa research station; IWWSN - wheat breeding lines from International Winter Wheat Screening Nursery) 119 1.8750 1.2500 0.6250 0.0000 DI470434lTUD] PI367047l6F61 PI243760lIRN] Pl347153lflF6] PI347099£HFGJ PI347041[6FG] Pl268311lIDN] PI367192[6FG] DI243786[IPN1 Pl367181l6FG] Pl243711llDN] Pl367087l6FG] D1367027[6FGJ Pl245621l6F0] PI245598£6FGJ 98-5/[IUHSN9] DI282917£6901 GDKlluHSN4] Dll?8077[TUR] 91243748llQN] P1341629[TUQ] Pli66653lTUD] PllB2446lTUQ] Pl341740[TUQ] Pl367109l6FG] PIl7IOI4[TUP] PI341487lTUQ] D1167772LTUD] Pl167746lTUQ] DI347023[6FG] DII67727[TUQ] 91167650[TUQ] P1166866[TUP] 01341673lTUQ] DI213658£IQN1 DI347031£6FGJ Cloe/[US_U_15] Pl172563lTUQ] DI347048l6FG] 91137741[IDN] Pl367062l6FG] PI367051[AFGJ Dl245388l6FG] Pl366586l6F6] Pl222671llQN1 PI]37760[IPNJ Cl6501lIQN] Pl24377lllQN] Pl243683llDN1 DI243733ilPN] 01167753llUQ] P1245554[6FG] PI220125[6FG] PI250949llPN] Pl243739llPN] PI243694llQN] PI222653llPN] Pl243667lIPN] PI222664llPNJ 00ffifi fiflfiii iii ii 1.8159 Figure 4. (Cont’d). 120 1.2500 0.6250 0.0000 L— Ll F—L __{—l ——l E q—d —l—C: 01243667[IDN1 PI222661£IQNJ PI245583l6FG] 9124543SLRFG] PI382002ilQN] AU//[IUUSN18J r—-——————-—-PlZ43650[IPN] TL--—-——-'PI202827[IQN] 01173501lTUQ] PI243767[IQN] DI243781£IQN1 Pl243677llDNJ Pl243764lIPN] DI347081[6FG] DI243637lIQN] 01222682[IQN1 Pl341471lTUQJ LarnedlUS_GD] 0dvos_241[DOMJ Favorltlpoml Doqu88[IUUSN8] BIHIHNSN3] Klein_6tlas[HQG Solo_50[HQG] SCOut66lUS_GP] DI]67667[TUQ] Pl243756lIDNJ KrgmkalUKP] CluylllQOH] PI262602lUKQ] Cenad,ll7[DOM1 Cenad_512[POMJ DI267135£QUSJ CharmanglUS_EQl SiouxlandlUSGD] firqeelUS_EQJ C114047[HDGJ Kharkol[US-u1 DI284668lYUO] DI245493l6FG] Baragan_l2[QOHJ KrusevaclYUG] Dl346431iYUG] 093-44/[IUUSN7] h8uCur95txllPOh] hfiaqnil_li4[6961 CRQ7l—[quSN15] PI326306[DUSJ Odesskaja52lUKP LamarlUS_Gp] SkorospelkalDON Abilene[US_GD] Tam-202[US_GP] UonalUS_GP] TomahauklUS_GPJ anastulUS_EQ] 1.8720 1.2500 0.6250 0.0000 7‘ TomahauklUS_GD] Comm. [ODESSH-91 SaugeriUS_EP] HakelieldiUS_ER Hdder[US _EP] r___: RedlandlUS_ 6p] firapahoelUS_ GP] __{: CTK_ 78[US _Gp] NE87615lUS_ GD] Klein_puma[HPG] CQP[US_GPJ D2555iUS_EQ] E: Tam—200[US GP] P2518lUS_EPl GuzankalUKQJ [, HoldovalPON] WC ExcelsmrlPOl‘l] [ODESSH_11 DacralPOMJ — [00Essa_31 l_[ [00mm] [::::::::::::[ODESSR-71 [ODESSH_12] [ [00ESSH_191 FreedomlUS_ER] T‘ Brkul)a_4[YUG] Tam-105[IUNSN61 ___E Tam-105tus,0m Tam-107lUS_GPJ HUHl IUUSNI 7] PI245480l6FG] PrrboinKQ] '— StepnaJ310[PUS] DI383362[RUS] BezllUUSNl] CardinallUS_ER] [ODESSR_2] [ODESSH_26] Tua1n[US_EPJ D2550[US-EP] YNH/[INHSNI3] CalduelllUS_EP] Hrbor[US,EQJ D2510lUS_EPJ DunaleUG] '_—______L___{ [ODESSH_181 “‘ Becker[US_ERJ —L ClarklUS_EPJ DI]67031[TUP] HugustaiUS_ENl C5107[US_HSU] Hillsdale[USEU] Frankenmuth[US_EUJ ‘ ' Karena[US_EU] Genesee[US_ENJ IonralUS_EU] Figure 4. (Cont’d). 1.8750 Figure 4. (Cont’d). 1.2500 122 0.6250 0.0000 L LIL-i —-——{: P—“'_l fli: ______r“‘———{ —rL L15 Genesee[US_E01 IonialUS_EU1 C5088[US_HSU1 HaruslUS_EHJ C4828[US_HSU] Geneva[US,EU] D2737NEUS_EH1 LoueHlUS,EU1 55-[INHSN16] FN-301lUS_U] SPG/SDNLUS_U_14 SON/[US,U_35] StephenslUS_U] HadseniUS_H] YHH/[US_N_10] KmorlUS_UJ C126-[IUUSN21] PIl67601lTUQ] PI345683lUKQJ DI372345£UKQJ Ukrainka[UKQl Htag-SSllUUSNS] ZenitkalUKQ] PI372120£QUSJ PI280455lUKQJ P1267148lUKQ] PI345694lUKR] Pl372326[PUSJ 91345691[RUS] BarhatnajalpUS] PI372126[PUSJ Lovr1n2llPOH] Lovrxn23llPOMJ BrennuleQH] UukalYUG] SumadijalYUG] PI323641£YU01 NS-32[YUG] NontJOIelFQAJ GaillardiFQHJ palmaresleRHJ Bezostaua4[PUS] BezostaJal[PUSJ PIIS3108iFPRJ LOU29/[luuSN11] BezlluuSNZOJ KavkaleUS] 91350678[QUS] SperberlGEQ] Heine_UIl[GED] HahndorlerlGER] DoroslGEQ] HochlanleEP] C111771[GEQJ DernburqerlGER] CI7076[FRHJ 1.8750 123 {SE EE—l: : ._____{__{ Figure 4. (Cont’d). fl lie £50000 DernburqerlGEQ] CI7076[FQHJ HannolGEQ] firromancheleQA RimpauslGEQ] DilOthED] Cordial[FQHJ Cappelle_[FQfl] Pll?4682[FQHJ LuroanPfi] Bizelifpfil SommelFRfi] HabichthERI KranichlGEQ] HarnelFPH] TrlolFRHJ CII2664lFDH] DI315980£FDHJ Uilmorin27lFDHJ CarStenUflllGEQ BeaufortlFQfll Pl239088lGEQ] ChanteclairlFPR NoraliterlFQH] PI315990iFPH] CoteD'oriFDH] 91262227lF96] Pl]80584[GEQ] 91166594[TUQ] DI367IOOl6FG] DI268465LRF6] Pl243793llDN] Dl366984l6FG] pl367075l6FG] DI367146l6FG] ElginlUS_U] Pl221482l6FG] pHHfl//[US_H_13] Pl382070llPN] horinS-U] TresiUS_U] HuaklUS_U] PeluiUS_U] UDH/[US_U_16] 124 2.2g0 _ 1.500 0.750 0.000 Afghanistan &‘ Iran Turkeu Odessa US_ER US_GP ~ Argentina Romania [UHSN Yugoslav1a DUSSLB Ukraine US_EH US_MSU US,“ r France Germanu ”‘fi Figure 5. Dendrogram resulting from the cluster analysis of morphological trait-based mean taxonomic distance estimates within and between 17 germplasm pools of common wheat. (Note: US_ER - Eastern U.S. SRW wheat; US_EW - Eastern U.S. SVWV wheat; US_MSU - SWW wheat breeding lines from Michigan State University; US_GP - U.S. Great Plains; US_W - Western U.S. SWW wheat; IWWSN - wheat breeding lines from lntemational Vlfinter Wheat Screening Nursery) DISCUSSION E l 1 . . l l E l The apparently narrow genetic base of common wheat has been postulated to be related to hexapolyploidization of only one or a few individuals of T. tauschii (2n=14, DD) and T. turgidum (2n=28, AABB) (Gill et a1, 1991; Lagudah et al. 1991a; Kimber and Sears, 1987). There has also been a relatively short time (i.e., about 10,000 years since polyploidization) for evolutionary differentiation (Gill et al, 1991; Lagudah et al. 1991a; Kimber and Sears, 1987). Since wheat is a highly self-pollinated crop species, the effects of linkage disequilibrium and intolerance to mutation of undesirable recessive alleles may be also involved in the limitation of genetic diversity as compared to outbreeding crop species (Jain, 1975). In spite of the limited genetic diversity of wheat, different levels of RF LP variation were detected among the 21 germplasm pools. The accessions from Southwest Asia were not only highly heterogeneous but genetically distant from the other germplasm pools. The genetic nature of landrace wheat populations is basically dependent upon the interaction between selective environments and pre-existing genetic variation within a population (Antonovics et a1. 1988). Significant variations of spatial and seasonal environments which are typical of Southwest Asia, have been reported as main causes of highly heterogeneous populations (Tahir and Valkoun, 1994). Kato and Yokoyama (1992) and Jaradat (1991) reported that landrace wheat populations have 125 126 evolved through developmental characters as the primary ‘adaptation strategy’ to fit to heterogeneous environments. Our multivariate analysis using morphological traits also showed that flowering time was one of the main sources of variation among 17 germplasm pools. This indicates that spatial distribution of genetic variation may result from optimization of fitness of each gene pool to its ecological environment. The genetic variation of Southwest Asian landrace pools was unexpectedly low in comparison to some advanced germplasm pools when it is taken into account that they are one of Vavilov’s centers of genetic diversity. In fact, all the accessions that were classified as landraces from Southwest Asia may not be correct (Dr. H. Bockelman, personal communication). Some accessions with identified genotype names such as ‘Saficha’, ‘Tirmai’, ‘Gandum’, ‘Karakilcik’ and ‘Bugday’ (Table 1) indicate that they may have originally come from another area (Dr. H. Bockelman, personal communication). Additionally, estimation of genetic diversity is ofien biased by sampling error through evaluation of regional collections or broader geographic collections within a country (Goffreda et al. 1992; Beer et al. 1993). The advanced germplasm pools developed by local breeding programs also showed relatively low levels of RF LP variation. This may result from (i) the influence of selective environments, particularly, intensive plant breeding in order to improve fitness to abiotically and biotically stressful environments, and to suit to farmer’s and consumer’s end use/needs, (ii) the limitation of available genetic variation, i.e., narrow genetic base of current cultivar germplasm pools derived from a limited number of plant 127 introductions (Cox, 1991), and (iii) the recurrent use of a few elite genotypes with good performance as breeding parents. For example, the progress of wheat breeding in European countries has been achieved on the basis of selection and hybridization with a few dominant cultivars (Lupton, 1987; Siedler et al. 1994). ‘Vilmorin 23’, ‘Vilmorin 27’, ‘Vilmorin 29’ and ‘Cappelle Desprez’ for the French wheat breeding, ‘Derenberger Silber’ and ‘Heine VII’ for the German wheat breeding, and ‘Bezostaja’ and its derivatives (e. g., ‘Aurora’ and ‘Kavkaz’) for the Russian wheat breeding are recognized as the historically important cultivars (for review, see Lupton, 1987). Siedler et al. (1994) reported that 60% of the European winter wheat lines surveyed for RFLPS were pedigree-related to eight cultivars that produced 93% of the RF LP bands. This suggests that the genetic base of each European winter wheat breeding program is very narrow, resulting in low frequency of RFLPS. Certain U.S. wheat germplasm pools also have narrow breeding histories, for instance, the hard red spring and winter wheat cultivars grown in the northern Great Plains largely descended from the single genotypes, ‘Marquis’ and ‘Turkey’, respectively (Cox, 1991). On the other hand, the germplasm base of the SRW wheat in the Eastern U.S. is much broader. This is due to the contribution of heterogeneous introduction lines from the European continent in the early days and the continuously additional introgression from other sources (Patterson and Allan, 1981;’Cox et al., 1986; Murphy et al., 1986; Cox, 1991). Consequently, a higher frequency of polymorphism exists within the US_ER pool than in any other U.S. or European germplasm pools. 128 :l l'l'l'l'El 1 Both pedigree-related and geographical proximity-related genetic relationships were revealed among accessions by the cluster analysis of RFLP-based genetic similarity matrices. A few cultivars from Russia and Ukraine such as ‘Kharkof (C1 1442)’, ‘Krymka (PI 355731)’, ‘Ukrainka (PI 326299)’, ‘Bezostaja’, ‘Bezostaja 1 (PI 345685)’, ‘Bezostaja 4 (PI 262645)’, and ‘Kavkaz (PI 367698)’ were important as the core of accessions in subgroups of the Eastern European wheat germplasms (Fig. 1). Ninety-one h~ percent of all restriction bands scored in the Eastern European wheat germplasms were from these cultivars. In particular, ‘Bezostaja 1’ is the pureline selection fiom ‘Bezostaja J - 4’ and is one of the parents of ‘Kavkaz’. ‘Ukrainka’, a landrace cultivar in Ukraine, is involved in the parentage of ‘Bezostaja 4’ (Martynov et al. 1992). This high pedigree- relatedness of pivotal accessions led their subgroups to cluster together. The Crimean group of HRW wheat introduced from Ukraine into the U.S. Great Plains by Russian Mennonite immigrants in about 1874 was grown under the name of ‘Turkey’ in the U.S. (Clark and Bayles, 1935). This introduction line is locally referred to as ‘Kharkof’ or ‘Krimka (Crimean)’ at its geographical origin and is presumed to be highly heterogeneous (Clark and Bayles, 1935; Cox, 1991). In spite of no geographical proximity, the US_GP pool is therefore genetically related with Russian and Ukraine gene pools, and this relationship is revealed by cluster analysis in the present study (Fig. 1 and 2). The effect of particular cultivars on subgroups of pedigree-related accessions is also apparent in the French and German wheat germplasms: (i) ‘Alliés’, ‘Hybride de 129 Joncquois (‘Vilmorin 23’x ‘Institut Agronomique’)’, ‘Vilmorin 27 (PI 125093)’, ‘Cappelle Desprez (PI 262223; ‘Hybride de Joncquois’x ‘Vilmorin 27’)’, and ‘Etiole de Choisy (PI 193108)’ for subgroups of French wheat germplasm pool, and (ii) ‘Kronen’, ‘Heine VII (PI 325877)’ and ‘Derenberger Silber’ for subgroups of German wheat germplasm pool (Fig. 1). In the present RFLP analysis, 92% of total restriction bands scored in the French wheat pool were common to three cultivars (‘Vilmorin 27’, ‘Cappelle Desprez’ and ‘Etiole de Choisy’). As for the German wheat pool, ‘Heine VII’ represented 88% of total restriction bands scored within this pool. Common pedigree- caused interrelatedness between some accessions from France and Germany were found from cluster analysis, too. For example, two French wheat cultivars, ‘Inversable Bordeaux (PI 315990)’ and ‘Palmaress (PI 316001)’, were closely related to ‘Heine VII’, which was probably because their pedigrees included this German cultivar and ‘Hatif Inversable’, one of the parents of ‘Heine VII’. The Western U.S. soft white wheat (US_W) accessions are of relatively diverse genetic background. Introduction lines such as ‘Turkey’, ‘Federation (Australian white spring wheat)’, and ‘Fortyfold’ contribute significantly to ‘Nugaines’, ‘Stephens’ and ‘Kmor’. ‘VPMl (a synthetic line derived from interspecific crosses, Ae. ventricosum / T. persicum /2/ 3* ‘Marne’)’ is involved in the pedigrees of ‘Madsen’ and ‘Hyak’. ‘Elgin’, club wheat cultivar released in 1943, also significantly contributes to ‘Moro’, ‘Tres’, and ‘Rely’. Accordingly, primary subgrouping within the US_W pool seems to follow this pedigree relationships (Fig. l). A particular subgroup of the US_W pool (i.e., ‘Madsen’, 130 ‘Hyak’, and ‘Stephens’ and its related lines) is also close to French wheat subgroup due to its pedigree-relatedness (Fig. 1). EEE [1].]...]. 1' 1.. Phenotypic traits have not only been used as cultivar descriptors and genetic markers but as a means to assess genetic distance on the assumption that phenotypic resemblance reflects genetic similarity (Beuningen, 1993; Souza and Sorrells, 1991a, b). Although significant variations for morphological traits exist among germplasm pools, and even within individual pools, these differences were not reliable as an estimator of genetic distance as compared to DNA sequence polymorphism in the present study. As for relationships of morphological traits with other parameters, Beer et al. (1993) found higher correlations of morphology-based GD with RFLP-based GS (I rl = 0.17 - 0.35) than with allozyme-based GS (l r1 = 0.11 - 0.16) in Avena sterilis. Beuningen (1993) reported a correlation value of 0.39 between GD matrices based on pedigrees and quantitative traits for spring wheat cultivars (T. aestivum). Poor association between measures revealed in our study suggested inadequate sampling of loci of morphological traits and possible effects of interaction between genotype and environment on phenotype. Even though morphological trait data are less informative with regard to accurate assessment of genetic relationships, the knowledge of clustering structure based on them may be still valuable. For example, as Souza and Sorrells (1991a) suggested, when adaptive characters largely distinguish genotypes and/or gene pools between clusters, it 131 may be advantageous for plant breeders to sample distantly related genotypes within a cluster that is adapted to a particular environment. 3 . 1° . l l 1. Theoretically, genetically distant genotypes as breeding parents should provide a wide range of recombinants in segregating generations. When information on genetic relationships of germplasms is available, it can therefore benefit plant breeders by facilitating selection of ideal breeding parents. In this study, genetically diverse groups of cultivars or landraces were identified by multivariate analysis based on DNA marker polymorphisms. Relatively high levels of homogeneity within a pool due to narrow genetic bases probably resulted in development of heterogeneity among gene pools. Since several distinct subgroups are clarified by the geographical origins and/or the coancestry, sampling of different germplasm among heterogeneous groups by these categories may ensure maintenance or expansion of the current genetic diversity of wheat cultivars and breeding lines. In order to optimize their potential as use in breeding programs, however, more details on the adaptability and other agronomic characters of each material to a particular agricultural environment should be identified. The genetic diversity of some advanced germplasm pools was reduced by the predominance of small number of elite cultivars or breeding lines in their pedigrees. However, as shown in this study, part of the advanced germplasm pools such as the Eastern U.S. SRW wheat gene pool and the HRW wheat gene pool of the U.S. Great Plains still have significant levels of genetic diversity. This range of diversity 132 undoubtedly results from systematic management of previous and/or current genetic variation suitable for plant performance such as yield, time of harvest and pest resistance (Patterson and Allan, 1981; Cox et al. 1986). APPENDICES 133 AS 2925: - c. 292$: 08.? c. 29);: 5.0 - a 2925; 3 5m mood mead zw2<<= 3223 E 3; E85. - .833 E 25E 53. mmmd 33$ .9 on cam - 53.3 E mean. so: momd ommd 0m< 32E - 2285 nmmd 228a - 222 wmwd mmmd 31w: m2 ES. 2: ES coo... 2282 - 2.3.280 End mvmd dOImD 880 - 880 2a 880 - 2&5 mmmd 33.0 - 820 wwmd mmmd 32219. 2393 - cSEcexch 500.0 232:: - 053200 mmmd mmmd >>wlw3 mom? - opmwd Nhad Emacs - Siam wad :56 ”film: .853” E SE - A853” EC 8.8 03.0 683” EC 32%: - 58% E __> 86: vomd ovmd mwO 883? E R 585.5 - :68. .8 3.38 298E Sad $082 E1396 8 22m - 80%; E 855522 mmwd mmmd (an. 8&8” E v «:35 - $898 Ev £638 oood 88va E 22 368962 - 848% EC wwwz Nvmd mowd 03> 3083 E: «£29803. - 63.0.5 EV mExEoEoEoO ommd 3&0me E: mom £39293 . 32.an Ev .ontd mde mmmd 1!: @988 EC 8 «598°ch - A825 E a £3928 Sad 885m EC 528. - GNPNB EC 3 «533833 mwmd mmmd max 588 E 8:38.95 - 83on Eto_m_8xm ommd 8825 E a :858m - 8835 EC 3 3.0 Rad mmmd .201 :1_ E:E_c__2 cams. Boa 0:00 Smog; coEEoo Co m_oon_ ocmm £53, mafia cabocom new woumEzmo b_..m__E_m ozocom nommndfim ho mo:_m> E:E_me ncm E:E_c_E .522 .< x_ucoaa< 134 000 000.0 000.0 0.00 500.0 500.0 N000 0000 #000 Nv00 500 0.0.0 500 0.00 00.0 500.0 N_.0.0 0N0.0 5N00 0N0.0 0N0.0 000.0 0N0.0 000.0 000.0 500.0 5N00 F000 0000 500.0 000.0 20.0 000.0 500.0 5N0.0 00.0 00.0 NN0.0 VN00 00.0 0N0.0 000.0 000.0 500.0 000.0 0N0.0 0N0.0 000.0 000.0 000.0 500.0 000.0 5000 _.N00 000.0 050.0 NN0.0 0N0.0 0.0.0 NN0.0 0v00 000.0 0V00 0N0.0 VN00 5N00 000.0 30.0 0V00 53.0 #00 N00 000.0 00.0 0N0.0 000.0 NN0.0 vN00 000.0 N000 000.0 000.0 0.00 0v00 .000 000.0 500.0 000.0 000.0 5N00 0N0.0 0N0.0 0N0.0 5N00 000.0 000.0 000.0 000.0 0000 0000 000.0 000.0 3.00 000.0 000.0 0N0.0 0N0.0 000.0 000.0 0N0.0 500.0 «00.0 0N0.0 5N00 N000 N000 #000 0000 00.0 00.0 000.0 00.0 _.N0.0 N00 300 0N0.0 00.0 000.0 000.0 000.0 000.0 000.0 000.0 00.0 «00.0 500.0 050.0 590.0 050.0 00.0 000.0 N000 000.0 000.0 000.0 000 NN0.0 :00 00.0 090.0 0N0.0 N00 N00 000.0 0.00 00.0 0.0.0 000 N00 0500 000 N00 0000 0N0.0 000.0 000.0 V000 N000 000.0 00.0 #000 N00 0N0.0 00.0 —N0.0 .N00 0000 000.0 000.0 00.0 0000 30.0 N000 000.0 VN00 0N0.0 000.0 000 _.N00 000.0 0900 F000 N000 0N0.0 0N0.0 00.0 050.0 000.0 N500 NN0.0 0N0.0 000 «N00 0000 300 0.0.0 0N0.0 ~00 50.0 0N0.0 V000 0.0.0 0.00 500.0 000.0 30.0 0v00 0N0.0 000.0 30.0 NN0.0 000.0 NN0.0 00.0 050.0 0N0.0 0.0.0 0N0.0 30.0 000.0 .300 202,3. 92 2,10: E010: 00.210: 3010: mmiwn «mo 5.... 00> 55 03. .29. 5008 301210 z>uzzo 2:125 mxuzzo «E 25.. 0"? 20 >> 00 :05. >>m mm <00 >>0 I> >>5 mx >25. Om< 03 w: 03 03 03 «.00 (m... 03> xx: 03m 20m 000 2:0 2:0 210 210 m3... Z> 0000205 xEmE be .0800? 05 E E0nEzc 0E. .005000E .3 0:0 _02 E u0§ano 0.02, 0.80 E00500 0cm 55:5 m0:_cm__E_m 000:00 00_E_0a-0aboc00 =< 000:2, coEEoo 0o 0.80 EmmEE00 rN c0053 0:0 55:5 0000E=m0 b_.0__E_m 000:00 c00E 00090.00”. .0 50:09? Appendix C. Eigenvectors, eigen values, cumulative variation and its 135 proportion of the phenotypic trait correlation matrix for the first six principal components (PC). PC axis PC1 P02 PC3 PC4 PC5 Traits Eigenvectors Flag leaf length 0.839 0.172 0.156 0.155 -0.158 Flag leaf width 0.738 -0.236 -0.169 -0.240 0.022 Plant height -0.119 0.419 0.653 -0.043 -0.513 Days to anthesis 0.755 0.368 0.173 0.114 0.022 Lodging -0.357 0.792 0.156 0.031 -0.064 Awnedness -0.641 0.402 0.161 0.321 0.113 Spike type 0.191 0.359 -O.729 -0.001 -0.448 Spike length 0.405 -0.001 0.774 -0.107 0.243 Glume color -0.060 0.705 -0.335 -0.062 0.402 Seed color 0.131 -0.377 -0.026 0.873 0.011 Seed shrinkage 0.562 0.649 -0.163 0.173 0.181 Eigenvalue 2.908 2.375 1.830 1.007 0.760 Proportion of 0.264 0.216 0.166 0.092 0.069 vanafion Cumulative 0.264 0.480 0.646 0.738 0.807 vaflafion LITERATURE CITED LITERATURE CITED Anderson, J .A., G.A. Churchill, J .E. Autrique, S.D. Tanksley, and M.E. Sorrells. 1993a. Optimizing parental selection for genetic linkage maps. Genome 36: 181-186. Anderson, J .A., M.E. Sorrells, and SD. Tanksley. 1993b. RFLP analysis of genomic regions associated with resistance to preharvest sprouting in wheat. Crop Sci. 33:453-459. Anderson, J .A., Y. Ogihara, M.E. Sorrells, and SD. Tanksley. 1992. Development of a chromosomal arm map for wheat based on RF LP markers. Theor. Appl. Genet. 83:1035-1043. Antonovics, J ., N.C. Ellstrancl, and RN. Brandon. 1988. Genetic variation and environmental variation: expectations and experiments. pp. 275-303. In. L.D. Gottlieb and SK. Jain. (ed.) Plant evolutionary biology. Chapman and Hall. London, New York. Beckmann, J .S. and M. Soller. 1983. Restriction fragment length polymorphisms in genetic improvement: methodologies, mapping, and costs. Theor. Appl. Genet. 67: 35-43. Beer. S.C., E. Souza and M.E. Sorrells. 1995. Prediction of genotype performance from ancestral relationship in oat. Crop Sci. 35: 69-73. Beer, S.C., .ll. Goffreda, TD. Phillips, .U.P. Murphy, and M.E. Sorrells. 1993. Assessment of genetic variation in Avena sterilis using morphological traits, isozymes, and RFLPS. Crop. Sci. 33: 1386-1393. Bennett, M.D., J .B. Smith, and J .S. Heslop-Harrison. 1982. Nuclear DNA amounts in angiosperrns. Proc. R. Soc. London, B. 216:179-199. 136 137 Bemado, R. 1993. Estimation of coefficient of coancestry using molecular markers in maize. Theor. Appl. Genet. 85: 1055-1062. Beuningen, L.T.V. 1993. Genetic diversity among North American spring wheat cultivars as determined from genealogy and morphology. Ph.D. Diss. The Univ. of Minnesota. Botstein, D., R.L. White, M. Skolnick, and R.W. Davis. 1980. Construction of a linkage map in man using restriction fragment length polymorphisms. Am. J. Hum. Genet. 32: 314-331. Briggle, L.W. and BC. Curtis. 1987. Wheat worldwide. pp. 1-32. In. E.G. Heyne (ed.). Wheat and wheat improvement. 2nd ed. ASA, CSSA, and SSSA. Madison, WI. Briggle, L.W. and LP. Reitz. 1963. Classification of T riticum species and of wheat varieties grown in the United States. USDA Tech. Bull. No. 1278. Washington, DC. Chao, S., P.J. Sharp, A.J. Worland, EJ. Warham, R.M.D. Koebner and MD. Gale. 1989. RFLP-based genetic maps of wheat homologous group 7 chromosomes. Theor. Appl. Genet. 78:495-504. Chen, H.B., J .M. Martin, M. Lavin, and LE. Talbert. 1994. Genetic diversity in hard red spring wheat based on sequence-tagged-site PCR markers. Crop Sci. 34: 1628- 1632. Chen, P.D., DJ. Liu, G.Z. Pei, L.L. Qi and L. Huang. 1988. The chromosome constitution of three endemic hexaploid wheats in Western China. Proc. 7th Intern. Wheat Genet. Symp. Cambridge, England: 75-80. Chen, Q., Y. Sun, and Y. Dong. 1985. Cytogenetic studies on interspecific hybrids of Xinjiang wheat. Acta Agronomica Sinica 11: 23-28. Clark, J .A. and BB. Bayles. 1935. Classification of wheat varieties grown in the United States. USDA Tech. Bull. No. 459. Washington, DC. 138 Clements, R.L. 1987. A study of gliadins of soft wheats from the Eastern United States using a modified polyacrylamide gel electrophoresis procedure. Cereal Chem. 64(6): 442-448. Committee on Genetic Vulnerability of Major Crops. 1972. Genetic vulnerability of major crops. Natl. Acad. Sci., Washington, DC. Cowen, NM. and KJ. Frey. 1987. Relationship between genealogical distance and breeding behavior in cats (A vena sativa L.). Euphytica 36: 413-424. Cox, TS. 1991. The contribution of introduced germplasm to the development of U.S. wheat cultivars. pp. 25-47. In H.L. Shands and LE. Wiesner (ed.) Use of plant introductions in cultivar development, Part I. CSSA Spec. Pub]. No. 17. CSSA, Madison, WI. Cox, TS. and J .P. Murphy. 1990. The effect of parental divergence on F2 heterosis in winter wheat crosses. Theor. Appl. Genet. 79:241-250. Cox, T.S., J .P. Murphy, and D.M. Rodgers. 1986. Changes in genetic diversity in the red winter wheat regions of the United States. Proc. Natl. Acad. Sci. USA. 83:5583- 5586. Cox, T.S., G.L. Lookhart, D.E. Warker, L.G. Harrell, L.D. Albers, and D.M. Rodgers. 1985. Genetic relationships among hard red winter wheat cultivars as evaluated by pedigree analysis and gliadin polyacrylamide gel electrophoretic patterns. Crop Sci. 25:1058-1063. Delannay, X., D.M. Rodgers, and R.G. Palmer. 1983. Relative genetic contributions among ancestral lines to North American soybean cultivars. Crop Sci. 23:944- 949. Devos, K.M., T. Millan, and MD. Gale. 1993. Comparative RF LP maps of the homoeologous group 2 chromosomes of wheat, rye and barley. Theor. Appl. Genet. 85:784-792. 139 Devos, K.M., M.D. Atkinson, C.N. Chinoy, C. Liu, and MD. Gale. 1992. RFLP-based genetic map of the homoeologous group 3 chromosomes of wheat and rye. Theor. Appl. Genet. 83: 931-939. Devos, KM. and MD. Gale. 1992. The use of random amplified polymorphic DNA markers in wheat. Theor. Appl. Genet. 84: 567-572. D’Ovidio, R., O.A. Tanzarella, and E. Porceddu. 1990. Rapid and efficient detection of genetic polymorphisms in wheat through amplification by the polymerase chain reaction. Plant Mol. Biol. 15:169-171. Dweikat, I. S. Mackenzie, M. Levy, and H. Ohm. 1993. Pedigree assessment using RAPD-DGGE in cereal crop species. Theor. Appl. Genet. 85: 497-505. Falconer, D.S. 1981. Introduction to quantitative genetics. 2nd ed. Longman. London and New York. Fisher, S.G. and LS. Lerman. 1983. DNA fragments differing by single base-pair substitutions are separated in denaturing-gradient gels: correspondence with melting theory. Proc. Natl Acad. Sci. USA. 80:1579-1583. Fowler, C. and P. Mooney. 1990. Shattering: food, politics, and the loss of genetic diversity. The Univ. of Arizona. Tucson. Frankel, OH. and A.H.D. Brown. 1984. Plant genetic resources today: a critical appraisal. pp. 249-257. In. J .H.W. Holden and J .T. Williams (ed.) Crop genetic resources: conservation & evaluation. George Allen & Unwin Ltd. London. Frankel, OH. 1970. Genetic dangers in the green revolution. World Agric. 19: 9-13. F ribe, B, Y. Mukai, and BS. Gill. 1992. C-banding polymorphisms in several accessions of Triticum tauschii. Genome 35: 192-199. Gerdes, J .T. and W.F. Tracy. 1994. Diversity of historically important sweet corn inbreds as estimated by RFLPS, morphology, isozymes and pedigree. Crop Sci. 34: 26-33. 140 Gill, K.S., E.L. Lubbers, B.S. Gill, J. Raupp, and TS. Cox. 1991. A genetic linkage map of Triticum tauschii (DD) and its relationship to the D genome of bread wheat (AABBDD). Genome 34: 362-374. Gizlice, Z., T.E. Carter, Jr., and J .W. Burton. 1994. Genetic base for North American public soybean cultivars released between 1947 and 1988. Crop Sci. 34: 1143- 1 15 1 . Goffreda, J.C., W.B. Bumquist, S.C. Beer, S.D. Tanksley, and M.E. Sorrells. 1992. Application of molecular markers to assess genetic relationships among accessions of wild oat, Avena sterilis. Theor. Appl. Genet. 85: 146-151. Goodman, M.M. 1968. The race of maize: 11. Use of multivariate analysis of variance to measure morphological similarity. Crop Sci. 8: 693-698. Graner, A., W.F. Ludwig, and A.E. Melchinger. 1994. Relationships among European Barley Germplasm: 11. Comparison of RFLP and pedigree data. Crop Sci. 34:1199-1205. Harlan, J .R. 1976. Genetic resources in wild relatives of crops. Crop Sci. 16: 329-333. Hartl, L., F .J . Weiss, and A. Jahoor. 1993. Use of RFLP markers for the identification of alleles of the Pm3 locus conferring powdery mildew resistance in wheat (Triticum aestivum L.). Theor. Appl. Genet. 86: 959-963. He, 8., H. Ohm, and S. Mackenzie. 1992. Detection of DNA sequence polymorphisms among wheat varieties. Theor. Appl. Genet. 84: 573-578. Helentjaris, T., G. King, M. Slocum, C. Siedenstrang, and S. Wegman. 1985. Restriction fragment polymorphisms as probes for plant diversity and their development as tools for applied plant breeding. Plant Mol. Biol. 5:109-118. Heun, M. A.E. Kennedy, J .A. Anderson, N.L.V. Lapitan, M.E. Sorrells, S.D. Tanksley. 1991. Construction of a restriction fragment length polymorphism map for barley (Hordeum vulgare). Genome 34: 437-447. 141 Jackson, E.A., L.M. Holt, and PI. Payne. 1983. Characterization of high-molecular- weight gliadin and low-molecular-weight glutenin subunits of wheat endosperm by two-dimensional electrophoresis and the chromosomal localization of their controlling genes. Theor. Appl. Genet. 66:29-37. Jain, SK. 1975. Population structure and the effects of breeding system. pp. 15-36. In. OH. Frankel and J .G. Hawkes (ed.) Crop genetic resources for today and tomorrow. Cambridge university press. Jain, S.K., C.O. Qualset, G.M. Bhatt, and K.K. Wu. 1975. Geographical patterns of phenotypic diversity in a world collection of durum wheats. Crop Sci. 15:700- 704. J aradat, A.A. 1991. Levels of phenotypic variation for developmental traits in landraces genotypes of durum wheat from Jordan. Euphytica 51: 265-271. Johnson, VA. 1986. World wheat production. pp. l-5. In. E.L. Smith (ed.) Genetic improvement in yield of wheat. CSSA, Madison, WI. Kam-Morgan, L.N.W., B.S. Gill, and S. Muthukn'shnan. 1989. DNA restriction fragment polymorphism: a strategy for genetic mapping of D genome of wheat. Genome 32: 724-732. Kato, K. and H. Yokoyama. 1992. Geographical variation in heading characters among wheat landraces, Triticum aestivum L., and its implication for their adaptability. Theor. Appl. Genet. 84: 259-265. Keim, P., W. Beavis, J. Schupp, and R. Freestone. 1992. Evaluation of soybean RFLP marker diversity in adapted germplasm. Theor. Appl. Genet. 85: 205-212. Kempthome, O. 1969. An introduction to genetic statistics. Iowa State Univ. Press, Ames, IA. Kimber, G. and M. Feldman. 1987. Wild wheat. An introduction. Special report 353, College of Agriculture, University of Missouri, Columbia. 142 King, SB. 1959. The kinds and distributions of wheats in China. 21-22. Nanjing Agricultural College. Kronstad, W.E. 1986. Germplasm: The key to past and future wheat improvement. pp. 41-53. In. E.L. Smith (ed.) Genetic improvement in yield of wheat. CSSA, Madison, WI. Lagudah, E.S., R. Appels, H.D. Brown, and D. McNeil. 1991a. The molecular-genetic analysis of Triticum tauschii, the D-genome donor to hexaploid wheat. Genome 34: 375-386. Lagudah, E.S., R. Appels, D. McNeil. 1991b. The Nor-D3 locus of T riticum tauschii: natural variation and genetic linkage to markers in chromosome 5. Genome 34: 387-395. Landry, B.S., R. Kesseli, H. Leung and R.W. Michelmore. 1987. Comparison of restriction endonucleases and sources of probes for their efficiency in detecting restriction fragment length polymorphisms in lettuce (Lactuca sativa L.). Theor. Appl. Genet. 74:646-653. Lee, M., E.B. Godshalk, K.R. Lamkey, and W.W. Wooden. 1989. Association of restriction fragment length polymorphisms among maize inbreds with agronomic performance of their crosses. Crop Sci. 29: 1067-1071. Levy, A.A. and M. Feldman. 1988. Ecogeographical distribution of HMW glutenin alleles in populations of the wild tetraploid wheat Triticum turgidum var. dicoccoides. Theor. Appl. Genet. 75: 651-658. Liu, Y.G., T.M. Ikeda and K. Tsunewaki. 1992. Moderately repeated, dispersed, and highly variable (MRDHV) genomic sequences of common wheat usable for cultivar identification. Theor. Appl. Genet. 84:535-543. Liu, Y.G. and K. Tsunewaki: 1991. Restriction fragment length polymorphism (RFLP) analysis in wheat. II. Linkage maps of the RF LP sites in common wheat. Jpn. J. Genet. 66:617-633. 143 Liu, Y.G., N. Mori, and K. Tsunewaki. 1990. Restriction fragment length polymorphism (RFLP) analysis in wheat. I. Genomic DNA library construction and RFLP analysis in common wheat. Jpn. J. Genet. 65:367-380. Lubbers, E.L., K.S. Gill, T.S. Cox, and BS. Gill. 1991. Variation of molecular markers among geographically diverse accessions of Triticum tauschii. Genome 34: 354- 361. Lupton, F .G.H. 1987. History of wheat breeding. pp. 51-70. In. F.G.H. Lupton (ed.) Wheat breeding. Chapman and Hall. London and New York. Lynch, M. 1988. Estimation of relatedness by DNA fingerprinting. Mol. Biol. Evol. 5: 584-599. Ma, Z.Q., M.E. Sorrells, and SD. Tanksley. 1994. RFLP markers linked to powdery mildew resistance genes Pm], Pm2, Pm3, and Pm4 in wheat. Genome 37:871- 875. Ma, Z.Q., B.S. Gill, M.E. Sorrells, and SD. Tanksley. 1993. RFLP markers linked to two Hessian fly-resistance genes in wheat (Triticum aestivum L.) from Triticum tauschii (coss.) Schmal. Theor. Appl. Genet. 85:750-754. Mackay, MC. 1990. Strategic planning for effective evaluation of plant germplasm. pp. 21-25. In. J .P. Srivastava and AB. Damania (ed.) Wheat genetic resources: Meeting diverse needs. John Wiley & Sons. Chichester, UK. Maniatis, T., E.F. F ritsch, and I. Sambrook. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbour Laboratory, Cold Spring Harbour, New York. Mantel, NA. 1967. The detection of disease clustering and a generalized regression approach. Cancer Res. 27: 209-220. Martin, J .M., L.E. Talbert, S.P. Lanning, and N.K. Blake. 1995. Hybrid performance in wheat as related to parental diversity. Crop Sci. 35: 104-108. 144 Martin, J .M., T.K. Blake, and EA. Hockett. 1991. Diversity among North American spring barley cultivars based on coefficients of parentage. Crop Sci. 31: 1131- 1137. Mayr, E. 1942. Systematics and the origin of species. Columbia University Press, New York. McCouch, S.R., G. Kochert, Z.H. Yu, Z.Y. Wang, G.S. Khush, W.R. Coffrnan, and SD. Tanksley. 1988. Molecular mapping of rice chromosomes. Theor. Appl. Genet. 76:815-829. Melchinger, A.E., A. Graner, M. Singh, and M.M. Messmer. 1994. Relationships among European barley germplasm: 1. Genetic diversity among winter and spring cultivars revealed by RFLPS. Crop Sci. 34: 1191-1199. Melchinger, A.E., J. Boppenmaier, B.S. Dhillon, W.G. Pollmer, and R.G. Herrmann. 1992. Genetic diversity for RFLPS in European maize inbreds: II. Relation to performance of hybrids within versus between heterotic groups for forage traits. Theor. Appl. Genet. 84: 672-681. Melchinger, A.E., M.M. Messmer, M. Lee, W.L. Woodman, and KR. Lamkey. 1991. Diversity and relationships among U.S. maize inbreds revealed by restriction fragment length polymorphisms. Crop Sci. 31: 669-678. Melchinger, A.E., M. Lee, K.R. Lamkey, and W.L. Woodman. 1990. Genetic diversity for restriction fragment length polymorphism: Relation to estimated genetic effects in maize inbreds. Crop Sci. 30: 1033-1040. Melchinger, A.E. 1990. Use of molecular markers in breeding for oligogenic disease resistance. Plant Breeding 104: 1-19. Messmer, M.M., A.E. Melchinger, R.G. Herrman, and J. Boppenmaier. 1993. Relationships among early European Maize Inbreds: 11 Comparison of pedigree and RFLP data. Crop Sci. 33:944-950. 145 Miller, J .C. and SD. Tanksley. 1990. Effect of different restriction enzymes, probe source, and probe length on detecting restriction fragment length polymorphism in tomato. Theor. Appl. Genet. 80:385-389. Monte, J .V., C.L. McIntyre, J .P. Gustafson. 1993. Analysis of phylogenetic relationships in the Triticeae tribe using RFLPS. Theor. Appl. Genet. 86:649-655. Mori, N., Y.G. Liu, and K. Tsunewaki. 1995. Wheat phylogeny determined by RFLP analysis of nuclear DNA. 2. Wild tetraploid wheats. Theor. Appl. Genet. 90:129- 134. Murphy, J .P., T.S. Cox, and D.M. Rodgers. 1986. Cluster analysis of red winter wheat cultivars based upon coefficients of parentage. Crop Sci. 26:672-676. Nei, M. and W. Li. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. USA. 76: 5269-5273. Nei, M. 1972. Genetic distance between populations. Amer. Naturalist. 106: 283-292. Nevo, E. and A. Beiles. 1989. Genetic diversity of wild emmer wheat in Israel and Turkey. Theor. Appl. Genet. 77: 421-455. Nevo, E., A. Beiles, and T. Krugman. 1988. Natural selection of allozyme polymorphisms: a microgeographical differentiation by edaphic, topographical, and temporal factors in wild emmer wheat (Triticum dicoccoides). Theor. Appl. Genet. 76: 737-752. O’Donoughue, L.S., E. Souza, S.D. Tanksley, and M.E. Sorrells. 1994. Relationships among North American oat cultivars based on restriction fragment length polymorphisms. Crop Sci. 34: 1251-1258. Paterson, A.H., S.D. Tanksley, and M.E. Sorrells. 1991. DNA markers in plant improvement. In: Advances in Agronomy, Sparks, D.L.(ed.) 46:39-90. Academic Press, New York. 146 Patterson, FL. and RE. Allan. 1981. Soft wheat breeding in the United States. pp. 33-98. In: Yamazaki, WT. and GT. Greenwood (ed.) Sofi Wheat: production, breeding, milling, and uses American association of cereal chemicals. Inc. St. Paul. Minnesota. Payne, RI. 1987. Genetics of wheat storage proteins and the effect of allelic variation on bread-making quality. Ann. Rev. Plant. Physiol. 38:141-153. Payne, P.I., L.M. Holt, A.J. Worland, and C.N. Law. 1982. Structural and genetical F studies on the high molecular-weight subunits of wheat glutenin. 3. Telocentric - mapping of the subunit genes on the long arms of the homoeologous group 1 chromosomes. Theor. Appl. Genet. 63: 129-138. Ragot, M. and DA. Hoisington. 1993. Molecular markers for plant breeding: E comparisons of RF LP and RAPD genotype costs. Theor. Appl. Genet. 86: 975- 984. Renfrew, J .A. 1973. Palaeoethnobotany. The prehistoric food plants of the Near East and Europe. Methuen, London. Riley, R., H. Coucoli and V. Chapman. 1967. Chromosomal interchanges and the phylogeny of wheat. Heredity 22: 233-248. Rohlf, F .J . 1992. NTSYS-pc, numerical taxonomy and multivariate analysis system. ver. 1.7. Exter Publishing Ltd., Setauket, New York. Saghai-Maroof, M.A., K.M. Soliman, R.A. Jorgensen, R.W. Allard. 1984. Ribosomal DNA spacer length polymorphism in barley: Mendelian inheritance, chromosomal location, and population dynamics. Proc. Natl. Acad. Sci. USA. 81. 8014-8018. Sarkar, P. and G.L. Stebbins. 1956. Morphological evidence concerning the origin of the B genome in wheat. Amer. J. Bot. 43: 297-304. Schachermayr, G., H. Siedler, M.D. Gale, H. Winzeler, M. Winzeler, and B. Keller. 1994. Identification and localization of molecular markers linked to the Lr9 leaf rust resistance gene of wheat. Theor. Appl. Genet. 88:110-115. 147 Shao, Q, C. Li and C. Basang. 1983. Semi-wild wheat from Xizang (Tibet). Proc. 6th Intern. Wheat Genet. Symp. Kyoto, Japan: 111-114. Siedler, H., M.M. Messmer, G.M. Schachermayr, H. Winzeler, M. Winzeler, and B. Keller. 1994. Genetic diversity in European wheat and spelt breeding material based on RFLP data. Theor. Appl. Genet. 88:994-1003. Smith, J SC. and 08. Smith. 1989a. The description and assessment of distance between inbred lines of maize: I. The use of morphological traits and descriptors. Maydica 34:141-150. Smith, J .S.C. and 0.8. Smith. 1989b. The description and assessment of distance between inbred lines of maize: II. The utility of morphological, biochemical, and genetic descriptors and a scheme for the testing of distinctiveness between inbred lines. Maydica 34: 151-161. Sneath, P.H.A. and RR. Sokal. 1973. Numerical taxonomy. Freeman. San Francisco. Song, K.M., T.C. Osborn, and RH. Williams. 1988. Brassica taxonomy based on nuclear restriction fragment length polymorphisms (RFLPS). l. Genome evolution of diploid and amphidiploid species. Theor. Appl. Genet. 75: 784-794. Souza, E., P.N. Fox, D. Byerlee, and B. Skovmand. 1994. Spring wheat diversity in irrigated areas of two developing countries. Crop Sci. 34: 774-783. Souza, E. and M.E. Sorrells. 1991a. Relationships among 70 North American oat germplasms: I. Cluster analysis using quantitative characters. Crop Sci. 31: 599- 605. Souza, E. and M.E. Sorrells. 1991b. Relationships among 70 North American oat germplasms: II. Cluster analysis using qualitative characters. Crop Sci. 31: 605- 612. Souza, E. and M.E. Sorrells. 1989. Pedigree analysis of North American oat cultivars released from 1951 to 1985. Crop Sci. 29:595-601. 148 Stuber, CW. 1992. Biochemical and molecular markers in plant breeding. pp. 37-61. In: Plant Breeding Reviews, Janick, J. (ed.). Wiley, New York. Tahir, M. and J. Valkoun. 1994. Genetic diversity in wheat - An international approach in its evaluation and utilization. Wheat Information Service. 78: 1-12. Talbert, L.E., N.K. Blake, P.W. Chee, T.K. Blake, G.M. Magyar. 1994. Evaluation of “sequence-tagged-site” PCR products as molecular markers in wheat. Theor. Appl. Genet. 87: 789-794. Tanksley, S.D., N.D. Young, A.H. Paterson, and M.W. Bonierbale. 1989. RFLP mapping in plant breeding: new tools for an old science. Bio/Technology 7:257- 264. Tanksley, SD. 1983. Molecular markers in plant breeding. Plant molecular biology reporter 1: 3-8. Tinker, N.A., M.G. F ortin, and DE. Mather. 1993. Random amplified polymorphic DNA and pedigree relationships in spring barley. Theor. Appl. Genet. 85: 976-984. Tsunewaki, K., S. Yamada and N. Mori. 1990. Genetical studies on a Tibetan semi-wild wheat, Triticum aestivum ssp. tibetanum. Jpn. J. Genet. 65:353-365. Udatsin, RA. and BF. Miguschova. 1970. Novoe v poznanni roda Triticum L. Vestnik s/ch nauki. 9: 20-23. Williams, J .G.K., A.R. Kubelik, K.J. Livak, J .A. Rafalski, and S.V. Tingey. 1990. DNA polymorphism amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res. 18: 7213-7218. Williams, K.J., J .M. Fisher and P. Langridge. 1994. Identification of RFLP markers linked to the cereal cyst nematode resistance gene (Cre) in wheat. Theor. Appl. Genet. 89:927-930. 149 Wrigley, C.W., P.J. Robinson, and WT. Williams. 1982. Relationships between Australian wheats on the basis of pedigree, grain protein composition, grain quality and morphology. Aust. J. Agric. Res. 33:419-427. Xie, D.X., K.M. Devos, G. Moore, and MD. Gale. 1993. RFLP—based genetic maps of the homoeologous group 5 chromosomes of bread wheat (Triticum aestivum L.) Theor. Appl. Genet. 87: 70-74. Yang, W.Y., C. Yen, and J .L. Yang. 1992. Cytogenetic study on the origin of some special Chinese landraces of common wheat. Wheat Information Service 75: 14- 20. Yen, C., M.C. Luo, and J .L. Yang. 1988. The origin of the Tibetan weedrace of hexaploid wheat, Chinese Spring, Chengdu-guang-tou and other landraces of the White Wheat complex from China. Proc. 7th Intern. Wheat Genet. Symp. Cambridge, England: 175-179. Yen, C, J .L. Yang, X.D. Liu and LR. Li. 1983. The distribution of Aegilops tauschii cosson in China and with reference to the origin of the Chinese common wheat. Proc. 6th lntem. Wheat Genet. Symp. Kyoto, Japan: 55-58. Zeven, A.C., N.C. Zeven-Hissink. 1976. Genealogies of 14,000 wheat varieties. Issued by: The Netherlands Cereals Centre-NGC, Wageningen and The International Maize and Wheat Improvement Center-CIMMYT, Mexico. "illllillllllillliilli