ΤΜΗΜΑ ΦΥΤΙΚΗΣ ΠΑΡΑΓΩΓΗΣ Crop Science Department ΕΡΓΑΣΤΗΡΙΟ ΒΕΛΤΙΩΣΗΣ ΦΥΤΩΝ ΚΑΙ ΓΕΩΡΓΙΚΟΥ ΠΕΙΡΑΜΑΤΙΣΜΟΥ Plant Breeding and Biometry Laboratory

Determination of the Genetic Relationships among Wheat Varieties Using Morphological, Molecular and Cytogenetical Markers

∆Ι∆ΑΚΤΟΡΙΚΗ ∆ΙΑΤΡΙΒΗ Philosophy of Doctorate (PhD) Thesis

Kamal Fouad Abdellatif Αthens, July/2007

1

ΓΕΩΠΟΝΙΚΟ ΠΑΝΕΠΙΣΤΗΜΙΟ ΑΘΗΝΩΝ ΤΜΗΜΑ ΦΥΤΙΚΗΣ ΠΑΡΑΓΩΓΗΣ ΕΡΓΑΣΤΗΡΙΟ ΒΕΛΤΙΩΣΗΣ ΦΥΤΩΝ ΚΑΙ ΓΕΩΡΓΙΚΟΥ ΠΕΙΡΑΜΑΤΙΣΜΟΥ

Ο προσδιορισµός των γενετικών σχέσεων µεταξύ ποικιλιών σιταριού µε τη χρήση µορφολογικών, µοριακών και κυτταρογενετικών δεικτών

Determination of the Genetic Relationships among Wheat Varieties Using Morphological, Molecular and Cytogenetical Markers

∆Ι∆ΑΚΤΟΡΙΚΗ ∆ΙΑΤΡΙΒΗ Kamal Fouad Abdellatif Αθήνα, 2007 2

TABLE OF CONTENTS Page

DEDICATION ACKNOWLEDGMENTS TABLE OF CONTENTS I. SUMMARY

1

1. English Summary

2

2. Greek Summary

4

II. REVIEW OF LITERATURE

7

1. Wheat production and improvement in Greece, Egypt and Cyprus

8

2. Genetic markers and diversity

12

3. Genetic diversity based on morphological characteristics

14

4. Genetic diversity based on SSR Markers

16

5. Genetic diversity based on ISSR Markers

23

6. Correlation between morphological and molecular markers

25

7. Cytogenetic studies

29

8. Study Objectives

32

III. MATERIALS AND METHODS

33

A. Plant material

34

B. Morphological Traits

36

B.1. Experimental design and data collection

36

B.2. Statistical analysis

37

C. Molecular Markers

43

C.1. DNA Isolation

43

C.2. Determination of DNA concentration

44

C.3. Simple Sequence Repeat (SSR) Reactions

45

C.4. Inter-Simple Sequence Repeat (ISSR) Reactions

45

C.5. Statistical analysis

47

D. In Situ Hybridization D.1. Slide treatment

48 48

3

D.2. Root tips preparation

49

D.3. Preparation of plant chromosomes:

49

D.4. In Situ Hybridization

50

IV. RESULTS A. Morphological Traits

54 54

A.1. Phenotypic plasticity

54

A.2. Two-way hierarchical Cluster Analysis

57

A.3. Cluster Analysis

60

A.4. Principal Coordinate Analysis (PCOORDA)

62

A.5. Discriminant Analysis

64

B. Simple Sequence Repeats (SSRs) Analysis

65

B. 1. Polymorphism and polymorphic content

77

B. 2. Cluster Analysis

79

B. 3. Principal Coordinate Analysis (PCOORDA)

81

C. Inter Simple Sequence Repeats (ISSRs) Analysis

83

C. 1. Cluster Analysis

83

C. 2. Principal Coordinate Analysis (PCOORDA)

88

D. Mantel’s Test

90

E. In situ hybridization

91

V. DISCUSSION

103

1. Phenotypic plasticity:

105

2. Cluster analysis:

107

3. Two-way hierarchical morphological Cluster Analysis

109

4. Principal Coordinate Analysis (PCOORDA)

110

5. Discriminant analysis

112

6. SSR and ISSR patterns

114

7. Correlation among similarity matrices

116

8. In situ hybridization

117

VII. REFERENCES

121

VI. APPENDICES

137

Appendix 1

138

4

Appendix 2

144

Appendix 3

146

Appendix 4

148

Appendix 5

152

Appendix 6

154

Appendix 7

156

Appendix 8

158

5

I. SUMMARY

6

1. English Summary Wheat is the most important cereal crop in the world. Because of the complexity of its genome different marker forms are needed to study its genetic diversity. Thirty nine morphological characteristics, 11 SSR primer pairs, nine anchored ISSR primers and in situ hybridization technique were used in order to study the genetic diversity and relationships among 45 wheat entries having different geographical origin (Greece, Egypt, Cyprus and Italy). Correlation among morphological and molecular markers was also computed in order to reveal the relationship among the two different types of markers. Morphological data were used for cluster and principal coordinate analyses (PCOORDA), and their phenotypic plasticity was calculated. Two-way hierarchical cluster using the morphological traits was compared to cluster analysis using the UPGMA (Unweighted Pair-Group Method with arithmetical algorithms Averages) method. Discriminant analysis was also carried out using JMP IN 5 software. For both types of molecular markers (SSR and ISSR), cluster analysis and PCOORDA were performed. The numbers of 5S and 18S-5.8S-26S rDNA sites were counted in 12 entries, representing different ploidy levels having different origins, using the in situ hybridization technique. A high percentage of phenotypic plasticity was obtained from the quantitative morphological traits compared to the qualitative traits. Wheat entries having different ploidy levels were grouped separately when using the cluster analysis, regardless of the type of marker (i.e. morphological traits, SSR and ISSR markers). On the other hand, wheat entries with diverse geographical background didn’t clearly separate from one other, and all entries were scattered inside the cluster regardless of their origin for all three analyses, except for the SSR analysis of the tetraploid wheat entries originating from Cyprus. The results of the two-way hierarchical cluster analysis (using Ward’s method) for the morphological characteristics were comparable to those obtained from cluster analysis using UPGMA method. Very high correlation values were found among some traits, ranging from 0.73 (between blade of flag leaf glaucosity and ear glaucosity traits) to 0.95 (between angle of flag leaf to culmn and angle of leaves to culmn traits). Based on the PCOORDA, hexaploid wheat entries were distinctly separated from the tetraploid wheats on the first principal coordinate (PC1) using the three different markers.

7

The results of the PCOORDA showed better resolution of genetic relationships compared to cluster analysis. According to discriminant analysis, the first canonical separated the wheat entries into two distinct groups, one containing the tetraploids and the other the hexaploids. Nine qualitative morphological traits and one quantitative characterized the tetraploid hard wheat entries, while the hexaploid wheat entries were characterized by four qualitative and two quantitative traits. Intermediate significant correlation values were obtained among morphological and molecular similarity matrices (r=0.54 between morphological traits and SSR, while r=0.57 between morphological and ISSR markers), while between the two types of molecular markers the correlation was very high (r=0.85). All tetraploid wheat entries used for in situ hybridization have four 5S rDNA sites on chromosomes 1A, 1B, 5A and 5B. Hexaploid wheat entries, on the other hand, have six 5S rDNA sites, except of ‘Yecora-E’ which has eight sites and ‘Nestos’ which had only four sites, similar to the tetraploid entries. A total of eight sites of 18S-5.8S-26D rDNA were revealed in tetraploid wheat entries, while 10 sites were visualized on the chromosomes for the hexaploid entries. Ten sites were counted for the Cypriot tetraploid entry ‘Mesaoria’ and 12 for the Greek hexaploid ‘Yecora-E’ and ‘Nestos’. One pair of homologous chromosomes having both rDNA types was observed in all wheat entries, except ‘Acheron’ which had two pairs and ‘Yecora-E’ which had three pairs. Further cytogenetical studies on ‘Yecora-E’ entry are needed to understand the nature and structure of ribosomal RNA genes in the specific variety which could promote our understanding of the evolution of the repetitive tandem arrays on DNA. All marker systems used, morphological, molecular and cytological proved to be useful in identifying genetic relationships among the entries and revealing their polymorphism. High correlation was obtained between molecular markers and higher than expected, according to the literature, between morphological and molecular markers. This may be due to the large number of morphological traits used.

8

2. Περίληψη Το σιτάρι είναι το σηµαντικότερο δηµητριακό στον κόσµο. Λόγω του περίπλοκου γονιδιώµατος, διαφορετικές µορφές δεικτών απαιτούνται για να µελετηθεί η γενετική παραλλακτικότητα στο σιτάρι . Τριάντα εννέα µορφολογικά χαρακτηριστικά, 11 ζευγάρια

SSR

εκκινητών,

εννέα

ISSR

εκκινητών

και

in

situ

υβριδισµοί

χρησιµοποιήθηκαν προκειµένου να µελετηθεί η γενετική παραλλακτικότητα και οι σχέσεις µεταξύ 45 καταχωρήσεων σιταριού µε διαφορετική γεωγραφική προέλευση (Ελλάδα, Αίγυπτος, Κύπρος και Ιταλία). Ο συσχετισµός µεταξύ των µορφολογικών και των µοριακών δεικτών πραγµατοποιήθηκε προκειµένου να αποκαλύψει τη σχέση µεταξύ των δύο διαφορετικών τύπων δεικτών. Τα µορφολογικά στοιχεία χρησιµοποιήθηκαν για τη ανάλυση κυρίων συνιστωσών (PCOORDA) καθώς και για τον υπολογισµό της φαινοτυπικής πλαστικότητας. Η διπλής κατεύθυνσης ανάλυση ιεραρχικής συστάδας (two-way hierarchical cluster analysis) πραγµατοποιήθηκε και συγκρίθηκε µε τα αποτελέσµατα της ανάλυσης συστάδας χρησιµοποιώντας τη µέθοδο UPGMA (Unweighted Pair-Group Method with arithmetical algorithms Averages). Οι δύο τύποι µοριακών δεικτών αναλύθηκαν µε τη µέθοδο PCOORDA. Ο αριθµός των 5S και 18S5.8S-26S rDNA ερευνήθηκε σε 12 καταχωρήσεις σιταριού που αντιπροσωπεύουν διαφορετικά

επίπεδα

πλοειδίας

και

διαφορετικές

γεωγραφικές

προελεύσεις,

χρησιµοποιώντας τη τεχνική του in situ υβριδισµού. Ένα υψηλό ποσοστό της φαινοτυπικής πλαστικότητας οφείλεται σε ποσοτικά µορφολογικά χαρακτηριστικά. Οι καταχωρήσεις σιταριού που έχουν διαφορετικό επίπεδο πλοειδίας οµαδοποιήθηκαν ξεχωριστά µε όλους τους τύπους δεικτών (µορφολογικά χαρακτηριστικά και µοριακοί δείκτες SSR και ISSR). Αφ' ετέρου, οι καταχωρήσεις από τις διαφορετικές γεωγραφικές περιοχές σαφώς δεν οµαδοποιήθηκαν, αλλά όλες διασκορπίστηκαν µέσα στο δενδρόγραµµα και στις τρεις αναλύσεις, εκτός από τις καταχωρήσεις σιταριού τετραπλοειδών από την Κύπρο µε την ανάλυση SSR. Τα αποτελέσµατα της ιεραρχικής ανάλυσης συστάδων (που χρησιµοποιεί τη µέθοδο του Ward) για τα µορφολογικά χαρακτηριστικά ήταν συγκρίσιµα µε εκείνα που υπολογίστηκαν από την ανάλυση συστάδων που χρησιµοποιεί τη µέθοδο UPGMA. Πολύ υψηλές τιµές συσχετισµού βρέθηκαν µεταξύ µερικών µορφολογικών χαρακτηριστικών που κυµάνθηκαν από 0,73 (µεταξύ του κηρώδους της λεπίδας του φύλλου και του

9

κηρώδους του σταχυού) ως 0,95 (µεταξύ της γωνίας του φύλλου σηµαία προς το στέλεχος και της γωνίας των φύλλων προς το στέλεχος). Με βάση την ανάλυση κύριων συνιστωσών, οι εξαπλοειδείς καταχωρήσεις σιταριού ξεχωρίσαν ευδιάκριτα από τους τετραπλοειδείς χρησιµοποιώντας την πρώτη κύρια συντεταγµένη (PC1) και µε τους τρεις δείκτες. Εννέα ποιοτικά µορφολογικά χαρακτηριστικά και ένα ποσοτικό ξεχώρισαν τις καταχωρήσεις σκληρού σιταριού, από τις καταχωρήσεις µαλακού σιταριού

που

χαρακτηρίστηκαν από τέσσερα ποιοτικά και δύο ποσοτικά χαρακτηριστικά. Οι τιµές συσχέτισης µεταξύ των µορφολογικών και µοριακών µητρών οµοιότητας κυµάνθηκαν σε ικανοποιητικά επίπεδα (r=0,54 µεταξύ των µορφολογικών χαρακτηριστικών και SSR, ενώ r=0,57 µεταξύ µορφολογικών και ISSR), ενώ ήταν πολύ υψηλές µεταξύ των δύο τύπων µοριακών δεικτών (r=0,85). Όλες οι τετραπλοειδείς καταχωρήσεις σιταριού που χρησιµοποιήθηκαν για τον in situ υβριδισµό έχουν τέσσερις περιοχές µε το 5S rDNA στα χρωµοσώµατα 1A, 1B, 5Α και 5B. Οι εξαπλοειδείς καταχωρήσεις έχουν έξι 5S rDNA περιοχές εκτός από την ‘Yecora-E’ που έχει οκτώ περιοχές και τη ‘Νέστος’ που είχε µόνο τέσσερις περιοχές, όπως οι τετραπλοειδείς καταχωρήσεις. Συνολικά οκτώ περιοχές 18S-5.8S-26S rDNA σηµάνθηκαν στις τετραπλοειδείς καταχωρήσεις σιταριού, ενώ 10 περιοχές βρέθηκαν στα χρωµοσώµατα των εξαπλοειδών καταχωρήσεων. Εξαίρεση ήταν η Κυπριακή τετραπλοειδή ποικιλία ‘Mesaoria’ µε 10 περιοχές 18S-5.8S-26S rDNA και οι εξαπλοειδές ‘Yecora-E’ και ‘Νέστος’ µε 12 περιοχές. Σε όλες τις καταχωρήσεις σιταριού βρέθηκε ένα ζευγάρι οµόλογων χρωµοσωµάτων που έχει και τους δύο τύπους rDNA εκτός από την ‘Αχερών’, στην οποία βρέθηκαν δύο ζευγάρια, και την ‘Yecora-E’ στην οποία

βρέθηκαν

τρία

ζευγάρια

των

οµόλογων

χρωµοσωµάτων.

Περαιτέρω

κυτταρογενετικές µελέτες για τη ποικιλία ‘Yecora-E’ απαιτούνται για να κατανοήσουµε τη φύση και τη δοµή των ριβοσωµατικών γονιδίων RNA στη συγκεκριµένη ποικιλία που θα µπορούσε να βοηθήσει την κατανόησή µας για την εξέλιξη των επαναλαµβανόµενων διαδοχικών αλληλουχιών του DNA. Όλοι

οι

δείκτες

που

χρησιµοποιήθηκαν,

µορφολογικοί,

µοριακοί

και

κυτταρογενετικοί βρέθηκαν ότι µπορούν να διαφοροποιήσουν τις καταχωρήσεις σιταριού που χρησιµοποιήθηκαν και να αναδείξουν τον πολυµορφισµό τους. Υψηλή συσχέτιση βρέθηκε µεταξύ των µοριακών δεικτών και υψηλότερη της αναµενόµενης, σύµφωνα µε

10

την βιβλιογραφία, µεταξύ µορφολογικών και µοριακών δεικτών, ίσως λόγω του µεγάλου αριθµού µορφολογικών χαρακτηριστικών που χρησιµοποιήθηκαν.

11

II. REVIEW OF LITERATURE

12

Determination of the Genetic Relationships among Wheat Varieties Using Morphological, Molecular and Cytogenetical Markers II. REVIEW OF LITERATURE 1. Wheat production and improvement in Greece, Egypt and Cyprus Two types of wheat are cultivated, the tetraploid Triticum durum Desf. and the hexaploid Triticum aestivum L. The tetraploid wheat has two genomes, A and B, while the hexaploid has the A, B and D genomes. The basic chromosome number is 7 and thus tetraploid wheat has 28 chromosomes (14 homologous chromosome pairs), while the hexaploid has 42 chromosomes (21 homologous chromosome pairs). In bread wheat (hexaploid), the 21 haploid chromosomes are arranged in seven homoeologous groups, each group having three chromosomes, one from each of the three genomes. It also has a genome size of 16.7 × 109 bp (17.33 pg/1C) (Hayden et al., 2001; Gupta, 1991; Bennett and Leitch, 1995; Mukai 2006), of which more than 80% is repetitive DNA. In durum wheat (tetraploid) each group has two homoeologous chromosomes (one from each of the two genomes A and B), and a genome size of 11.8 × 109 bp (12.28 pg/1C). Wheat, either bread or pasta, is the most important cereal crop in the world, with a total grain production of 580Mt/year (FAO, 2000). It has been the staple food of the major civilizations in Europe and North Africa for 8,000 years. The high protein content of wheat grain makes it the most important source of human nutrition in the world and consequently of social economy. Wheat yields in Southern European countries are much lower than the yields achieved in Middle Europe (FAO, 1995). In general, yields depend upon climatic and soil conditions, as well as other factors, such as choice of cultivar, size of kernel, etc. The major Mediterranean durum-producing countries, such as Italy, Spain and France, are large producers cultivating 1,069,339 hectares, 388,031 hectares and 420,000 hectares, respectively. Of the non-EU Mediterranean countries, Turkey is the

13

largest producer in terms of area with 44,552 hectares. In the Mediterranean countries, wheat is a major crop for Egypt, France, Greece, Israel, Italy, Lebanon, Portugal and Spain (De Castro et al., 2002). Durum wheat is a major crop in Greece (Bebeli and Kaltsikes, 1985) producing about 10 percent of total EU durum output (FAO, 2003). Comparing total wheat production of Greece, Egypt and Cyprus for the last 45 years (1960 till 2005), clear differences can be noted on policies in all wheat production components (area harvested, yield per hectare and production quantity) (Figure 1). Wheat area harvested in Egypt steadily increased from 581,409 hectares to 1,262,000 hectares. At the same time wheat harvested area in Greece and Cyprus decreased from 1,172,900 to 843,900 hectares and from 78,930 to 6,239 hectares, respectively (Figure 1A). Wheat yields (kg/ha) during the same period tripled in Egypt (from 2,470 in 1961 to 6,486 in 2005) and in Cyprus (from 579 in 1961 to 1,486 in 2005) while a slight increase was observed in Greece (from 1,303 in 1961 to 2,422 in 2005) (Figure 1B). Total wheat production, which is a combination of area harvested and yield per hectare, showed different trends for the three countries. In Egypt total wheat production increased by a factor of eight from 1961 to 2005 (from 1,435,926 to 8,185,000 tons), while in Greece a slight increase was recorded (from 1,527,870 to 2,044,149 tons) and in Cyprus a significant decrease was observed (from 45,720 to 9,274 tons) during the same period (Figure 1C). In 2004 total production of wheat in Greece reached 1.8 million tons compared to 2.1 million tons in 1964. Although yields increased from 165.3 tons/km2 in 1964 to 211.3 tons/km² in 2004, the total area harvested decreased from 12,631 km² in 1964 to 8,519 km² in 2004. However, wheat in Greece (cultivated in 21.5% of the arable land) has lower grain yields compared to those achieved in Middle European countries (IENICA, 1999). According to the Common Agricultural Policy (CAP, 2003), a new premium will be introduced to improve the quality and quantity of durum wheat used for semolina and

14

A

Cyprus Egypt Greece

Area Harvested (Hectar) 1400000 1300000 1200000 y = -6686.9x + 1E+06

1100000

Area Harvested

1000000 900000 800000 700000 600000 500000 y = 14467x + 382795 400000 300000 200000 100000 0 -100000

y = -1919.6x + 72284

1960 1

3

1965

5

7

1970

1975

1980

1985

1990

1995

2000

2005

9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45

Years

B

Cyprus

Yield (Kg/Ha)

Egypt Greece

7000 6000 5000

Kg/Ha

y = 102.34x + 1826 4000 y = 22.738x + 1741.7

3000 2000 1000 0

y = 34.413x + 728.19 1960 1 3

197013 151975 1985 1995 2005 5 1965 7 9 11 17 19 1980 21 23 25 27 29 1990 31 33 35 37 39 2000 41 43 45

Years

C

Production Quantity

Total Production

9000000

Cyprus Egypt Greece

8000000 y = 144920x - 63981

7000000 6000000

Tons

5000000 4000000 3000000

y = 9086x + 2E+06

2000000 1000000 y = -1635x + 67258 0 119603

5 1965 7 9 11 13 15 1975 17 19 1980 21 23 251985 27 29 1990 31 33 35 37 392000 41 43 45 1970 1995 2005

-1000000

Years

Figure 1: Wheat productivity of Greece, Egypt and Cyprus during the period 1960 till 2005 (A) area harvested (hectares); (B) yield (Kg/Ha); and (C) total production (tons) (FAOSTAT, 2006). 15

pasta production. The premium will be paid in traditional production zones to farmers who are using a certain quantity of certified seeds of selected varieties that will be selected to meet quality requirements. Wheat production in Egypt increased from 2 million tons in 1982 to 8.3 million tons in 2006, ranking fourth in the world productivity (FAOSTAT, 2006). This increase is due to the cultivation of high yielding and long spike varieties, in the context of the National Campaign for Wheat Improvement, and the price incentives offered by the State to wheat growers. The aim of the present development strategy is to optimize the cropping pattern and the use of agricultural and water resources, taking into consideration that wheat production in Egypt is assessed only under irrigated conditions. By 2017 it is planned that the cropping pattern should involve a gradual increase of the area planted with wheat from about 1 million ha in 1997 to about 1.4 million ha and raising wheat production to about nine million tons annually, in order to meet the increasing national demand resulting from the growing population. Wheat production in Cyprus covers two-thirds of domestic demand. Although mechanized farming had significantly improved cereal production, wheat production in 1961 was 45,720 tons, while in 2003 decreased by 68.9%, to 14,280 tons. The yield production decreased from 3,514 kg/ha in 1996 to 1,486 kg/ha in 2005. This was due to steady decline of wheat importance relative to barley during the 1980s till present, and to reduced governmental interest in the wheat breeding program, since greater subsidies were paid for barley. In 2004 imports of wheat and barley reached 364,200 tons, most of which was wheat, and thus there is a need for new improved wheat varieties to increase yields and to meet the domestic demand. Thus, both Greece and Cyprus need to develop and cultivate high producing wheat cultivars in order to cover expenses and need to maintain genetic diversity, while Egypt needs to develop cultivars that meet reform strategy goals (i.e. optimize cropping pattern and use of agricultural and water resources).

16

2. Genetic markers and diversity Analysis of genetic relationships in crop species is an important component of crop improvement programs, as it provides information about genetic diversity (Mohammadi and Prasanna, 2003) and sources of genetic variation for plant breeding programs. In the past, the complexity of the wheat genome led to a delay in developing and applying of molecular markers in plant breeding programs, so that it lagged behind rice, barley and maize in the availability of markers for different traits, and for understanding its genomic structure and organization. Molecular markers are now available for wheat, which may be specific either for individual genomes or chromosomes. Thus the availability and variability of a large number of molecular markers in wheat can be used in interspecific analyses, comparative analysis and gene introgression studies as well as in practical wheat breeding (Gupta et al., 1999). Kumar (1999) divided the genetic markers into two types (1) morphological markers that their expression is altered by epistatic and pleitropic effects and affected by genotype x environment interaction, and (2) molecular markers that reveal polymorphisms at the protein level (known as biochemical markers) or at the DNA level (known as DNA markers). He also classified the DNA markers into two categories depending upon the way this polymorphism is revealed: hybridization-based polymorphisms and PCR-based polymorphisms. He reported that morphological, isozyme and nuclear DNA markers (co-dominant markers) are inherited in a Mendelian manner. For any genome, the number of morphological and isozyme markers is limited compare to DNA markers which are ubiquitous and numerous. Molecular markers and marker mapping are part of the intrusive 'new genetic' that is thrusting its way into all areas of modern biology, from genomics to breeding, from transgenics to developmental biology, from systematics to ecology, and even into plant and crop physiology (Jones et al., 1997). It is mentioned that molecular markers (DNA markers) reveal neutral sites of variation at the DNA sequence level, having the

17

advantage of being more numerous than morphological markers. The assessment and use of the molecular markers can be achieved using restriction enzymes, electrophoretic separation of DNA fragments, southern hybridization, the polymerase chain reaction (PCR) and labeled probes (Jones et al., 1997). Molecular markers can be broadly classified in the following three groups according to Gupta et al. (1999): 1.

Hybridization-based DNA markers, such as restriction fragment length polymorphisms (RFLPs) and oligonucleotide fingerprinting.

2.

PCR-based DNA markers such as random amplified polymorphic DNA (RAPD) or arbitrarily primed PCR (AP-PCR), which can also be converted into sequence characterized amplified regions (SCARs); simple sequence repeats (SSRs) or microsatellites also known as allele-specific PCR (AS-PCR); intersimple sequences repeat amplification (ISSR or ISA) or microsatellites-primed polymerase chain reaction (MP-PCR); sequence-tagged sites (STS); amplified fragment length polymorphisms (AFLPs); cleaved amplified polymorphic sequence (CAPS); and amplicon length polymorphisms (ALPs). In addition to the above markers, DNA amplification fingerprinting (DAF) have also proved useful in the detection of polymorphism (Mohan et al., 1997).

3.

DNA chip and sequencing-based DNA markers such as single nucleotide polymorphisms (SNPs) and microarrays. The ideal DNA marker must have the following properties according to Joshi et

al. (1999): (1) to be highly polymorphic in nature, (2) to have co-dominant inheritance (determination of homozygous and heterozygous states in diploid organisms), (3) to have frequent occurrence in genomes, (4) to have selective nature behavior (the DNA sequences of any organism are neutral to changes of environmental conditions or management practices), (5) easy to assay, (6) to be high reproducible within and between laboratories. Of course, it is very difficult to find a molecular marker system meeting all of the above criteria but, depending upon the aim of the study, the molecular marker system selected must have at least a few of these characteristics.

18

The term DNA-fingerprinting was introduced for the first time by Jeffery et al. (1985) to describe bar-code-like DNA fragment patterns generated by multilocus probes after electrophoretic separation of genomic DNA fragments. The emerging patterns made up a unique feature of the analyzed individual and are currently considered to be the ultimate tool for the biological individualization. Recently, the term DNA fingerprinting/ profiling is used to describe the combined use of several single locus detection systems that are being used as versatile tools for investigating various aspects of plant genomes. These include characterization, genetic variability, genome fingerprinting and mapping, gene localization, analysis of genome evolution, population genetics taxonomy, plant breeding and diagnosis (Joshi et al., 1999). It is vital for plant breeding programs to have sufficient genetic diversity available in order to develop new varieties that are aimed towards the increase of crop productivity and to withstand damage from biotic and abiotic factors. In this respect, efforts have also been made to predict the prospects of developing superior genotypes from a cross by calculating the genetic similarity (GS) or genetic distance (GD) between the parents, since the later can be used as an estimation of expected genetic variances in different sets of segregating progenies derived from different crosses (Korzun, 2003).

3. Genetic diversity based on morphological characteristics It is useful for the plant breeder to determine the genetic relationships among the genotypes of the available breeding material. The relationship between genotypes, according to Schut et al. (1997), is usually based on three sources of information: (1) geographic information about the origin of the genotypes, (2) pedigree information, and (3) information about plant characteristics. Geographic information can be helpful especially when other information for the genotypes is lacking. This is often the case for gene-bank material. Pedigrees of varieties and breeding lines are often well documented. Most of the times they trace back to landraces and wild accessions. However, pedigrees sometimes contain erroneous or incomplete information. Plant characteristics are the only

19

source of relationship information that is available for any set of genotypes. Such characteristics can be divided into three groups: (a) agronomic/morphological characters (used to distinguish between the varieties), (b) biochemical characters (e.g. storage proteins and isozymes), and (c) molecular (DNA) markers. Differences between genotypes with regard to any of these characteristics are either indirect or direct representations of differences at the DNA level and are therefore expected to provide information about genetic relationships (Schut et al., 1997). Genealogical analysis of diversity was studied in Russian winter hexaploid wheat cultivars from 1929 to 2002 (Martynov and Dobrotvorskaya, 2006). Their results showed an increase in diversity of Russian cultivars due to the use of foreign material in their breeding programs. Since 1970s the number of landrace ancestors was steadily increased. On the other hand, genetic erosion of the released diversity occurred, whereas, about 50% of the Russian landrace ancestors were lost (thrown out) in the 1950s and 1960s. For wheat, UPOV (Union pour la Protection Obtentions Vegetales, 2002) provides a set of morphological descriptors that can be used to characterize the varieties. They contain morphological characters of wheat seed and plants at different growth stages. The characteristics may be either qualitative [nominal (ex. growth habit and color of awns at maturity), ordinal (ex. auricle color and glaucosity), or binary (ex. apical rachis hairiness of the convex surface)] or quantitative [continuous (ex. seeds weight and plant height) or discrete (ex. heading date and number of seeds per head)]. Genetic diversity for tetraploid Ethiopian wheat populations based on morphological characteristics was determined (Bechere et al. 1996; Pecetti and Damania, 1996; Eticha et al., 2006). Bechere et al. (1996) measured eleven qualitative and quantitative traits to differentiate 27 tetraploid wheat landraces from north and northcentral regions of Ethiopia, while Pecetti and Damania (1996) evaluated tetraploid wheat landraces from two Ethiopian provinces (Shewa and Tigray) using 25 morphological traits (13 quantitative and 12 qualitative). They found that these landraces were distinctly different, attributing their differences to the environmental conditions between the two provinces. They also observed differences among the populations within each province.

20

Eticha et al. (2006) compared 32 tetraploid wheat landrace populations from two regions in Ethiopia (Wollo and Bale) using seven morphological traits. A total of 2559 individual plants (45-110 plants representing each population) were classified into their species components except for five tetraploid (T. durum, T. turgidum L., T. aethiopicum Jakubz., T. polonicum, T. dicoccon Schrank, 2n=28) and one hexaploid (T. aestivum L., 2n=42) wheat species which were found to be mixtures in varying proportions. They supported their results using discriminant analysis, revealing 91.5% correct classification of the wheat species. Similar studies were carried out for Austrian wheat landraces (Zeven and Schachl, 1989) using 49 morphological characters in order to differentiate 66 accessions originating from 15 bread wheat populations, as well as for a sample of Argentinean wheat landrace (Tranquilli et al., 2000), for Italian durum wheats (Motzo et al., 2004), and more recently for twelve durum wheat landraces originating from two areas in Jordan (Ajloun and Karak) using 14 morphological and agronomic traits (Rawashdeh et al., 2007). All of them revealed a wide range of phenotypic diversity among the landraces tested. Genetic diversity in hexaploid bread wheat was also studied using morphological characteristics in combination with AFLPs (Lage et al., 2003; Moghaddam et al., 2005), RAPDs (Cao et al., 2000; Maric et al., 2004), and SSRs (Hamza et al., 2004). Cao et al. (2000) studied the phylogenetic relationships of five morphological groups of hexaploid wheat (T. aestivum) based on RAPD analysis. They found that macha, spelta, and common wheat accessions formed separate group-specific clusters, whereas semi-wild wheat (SWW) and vavilovii did not fall within the clusters of the other wheat groups, nor were they grouped together. They reported that their results were in agreement with those based on morphological classification, suggesting that common wheat is most closely related to SWW, followed by spelta, vavilovii, and macha.

4. Genetic diversity based on SSR markers Simple Sequence Repeats (SSR) or microsatellites are small DNA segments,

21

usually 2 to 6 bp in length, that repeat tandemly a number of times. They are abundant, dispersed throughout the genome and show higher levels of polymorphisms compared to other genetic markers. Microsatellites appear to be ubiquitous in higher organisms, although their frequencies vary between species. In addition to their ease detection they are inherited in a co-dominant manner. Due to their advantages compared with other types of molecular markers, microsatellites have recently become important genetic markers in cereals, including wheat (Holton 2001). The SSRs are easier to use than RFLPs due to the smaller amount of DNA required, the higher existing polymorphism and the ability to automate the assay. The SSR assays are more robust than RAPDs and more transferable than AFLPs (dominant marker) in which the polymorphisms are often difficult to transfer to more sequencespecific PCR applications. This disadvantage of AFLPs may hinder the assignment of linkage groups to chromosomes unless they also contain other types of DNA markers of known chromosomal location, or whose chromosomal origin can be easily determined. The SSRs are now widely used in combination with AFLPs to produce detailed genetic maps of crop plants and are replacing RFLPs. The co-dominant nature of SSRs is also an advantage for genetic mapping (Holton 2001). Röder et al. (1995) compared 15 microsatellite markers to 14 RFLP markers tested on 12 wheat lines and six synthetic wheat lines. The synthetic wheat lines were developed by crossing a tetraploid wheat (T. durum, genome AABB) with a diploid (Triticum tauschii, genome DD) followed by chromosome doubling. The microsatellite markers were found to be significantly more variable than the RFLP markers. The average variability detected by microsatellite markers as well as RFLP markers was increased when synthetic lines were included in the survey. However, the Polymorphism Information Content (PIC) obtained was higher when only the accessions were screened with microsatellites than when the accessions and the synthetic lines were screened with RFLP markers. They concluded that microsatellites are a useful class of markers and complementary to other types of genetic markers in wheat, although their development is costly and time consuming.

22

In another comparison between RFLP and SSR markers in mapping of genes for dwarfing (Rht12) and vernalization response (Vrn1) on chromosome 5A of wheat, Korzun et al. (1997a), found that five of 17 RFLP probes previously located on Triticeae homoeologous group 5 (one RFLP probe) or 4L (four RFLP probes) gave polymorphism with at least one restriction enzyme and could therefore be mapped to chromosome 5AL. Another five RFLP probes were polymorphic for chromosomes 4B or 5B or 4D only, whereas seven probes were not polymorphic. On the other hand, four of the five microsatellites were polymorphic and could be incorporated and mapped on chromosome 5AL. Adhikari et al. (2004) studied the link between microsatellite markers and the Stb2, and Stb3 genes for resistance to Septoria tritici blotch of wheat. They found SSR markers linked to Stb2 and Stb3 genes at distance of 0.9 and 3.7 cM for Stb2 gene and at distance of 3.0 cM for Stb3 gene. In contrast to the above studies, Bohn et al. (1999) carried out a study to assess the level of genetic diversity among German and Austrian winter wheats on the basis of RFLPs, AFLPs and SSRs, and their use for predicting progeny variance. The average PIC was not found significantly different between the marker systems. Fifty-nine out of 191 RFLP probes yielded polymorphic banding patterns, while 117 out of 559 AFLP fragments were polymorphic across all 11 winter wheat cultivars, and finally 13 out of 22 SSR primer pairs produced 33 polymorphic bands from a total of 47 fragments. In comparison, the AFLPs were more reliable than either SSRs or RFLPs in fingerprinting each wheat cultivar. They recommended AFLPs for fingerprinting wheat cultivars in plant variety protection, quality control as well as for identification of the wheat varieties. Microsatellite markers have been used to detect the genetic diversity of wheat (Plaschke et al., 1995; Dograr et al., 2000; Huang et al., 2002; Lima et al., 2003; RibeiroCarvalho et al., 2004; Khleshtkina et al., 2004; Dreisigacker et al., 2004; Yifru et al., 2006; Zhang et al., 2006; Landjeva et al., 2006). Plaschke et al. (1995) used 23 SSR primer pairs to estimate the genetic diversity among 40 wheat cultivars and lines. The microsatellite markers were able to detect genetic variation in closely related elite material of hexaploid wheat and to identify groups with different levels of genetic

23

variation. The correlation between the similarity values, based on the microsatellite primer pairs and pedigrees was only moderate but significant (r=0.55). So, they concluded that it is possible to use a small number of microsatellite primers to select lines or cultivars with high genetic diversity and to carry out phylogenetic studies. However, Varshney et al. (1998) mentioned that microsatellite probes showed low levels of polymorphism in bread wheat. Twenty-three microsatellite probes were used in combination with 14 different restriction enzymes to detect the polymorphism among nine diverse bread wheat genotypes. A very low level of polymorphism was detected when using as many as 142 probe-enzyme combinations. They suggested that the genome of bread wheat differs from that of other plants in the organization and distribution of SSR and microsatellites are not useful for fingerprinting purposes in wheat. Determination of genetic diversity, however, based on SSRs (microsatellites) and pedigree data were accomplished. Parker et al. (2002) screened 124 Australian wheat cultivars and breeding lines using 19 microsatellite (SSR) loci generating 160 scorable bands that were used to calculate a genetic distance (GD) matrix. A distance matrix based on coefficient of parentage (COP) scores was also generated for the cultivars for which good pedigree records were available. The SSR and COP data for 101 of the wheat cultivars were compared with genetic distance scores obtained using 1898 scorable restriction fragment length polymorphism (RFLP) bands. Dendrograms were generated based on SSR, RFLP and combined (SSR, RFLP and COP) data. The matrices obtained from the SSR and RFLP data were significantly correlated (r=0.51), although a large number of SSR loci were required for determining robust genetic relationships when large numbers of cultivars were used. In addition, accurate pedigree records were needed to determine genetic relatedness using COP. Korzun et al. (1997b) used the microsatellite markers to distinguish inter-varietal chromosome substitution lines of wheat (T. aestivum L.). Forty-five primer pairs (one to three for each chromosome) were used to analyze the complete set of ‘Cappelle-Desprez’ and ‘Bezostaya 1’ substitution wheat lines. The microsatellites were able to provide polymorphisms between chromosomes of ‘Bezostaya 1’ and ‘Cappelle-Desprez’. They

24

reported that microsatellites provide an abundance of polymorphic markers evenly distributed over the genome, and are highly suitable as nuclear genetic markers in wheat for testing the correctness of genetic stocks. Microsatellite markers can also exploite the utilization of the stock to detect genes of agronomic importance, as well as to provide readily detectable markers for these genes allowing them to be handled efficiently in segregating breeder’s populations. In another study Stachel et al. (2000) proved that the resolving power of microsatellites is in varietal differentiation, breed under specific environmental conditions and for different end-use. A small number of microsatellite primer pairs were able to clearly differentiate between the cultivars originating from three agroecological areas (AEAs), Germany, Austria and Hungary, and even between quality categories within AEAs. Their results have proven the excellent resolving power of microsatellites in varietal differentiation and for different end-use. Moreover, they concluded that microsatellites should be a valuable tool for establishing a complete phylogenetic history of the entire present-day European wheat gene pool for the benefit of plant breeding. Dreisigacker et al. (2004) studied the determination of the genetic diversity among CIMMYT wheat lines from different megaenvironments (ME) based on SSRs and pedigree analyses. Eight to 15 CIMMYT advanced spring wheat lines were selected, out of 68 advanced lines from crosses made during 1989 to 1996, representing five spring bread wheat MEs. Ninty-seven SSR markers were used to assess the genetic diversity within and among the sets of the lines. The correlation between GS and COP estimates was low (r= 0.43) but significant, whereas in the study of Plaschke et al. (1995) it was estimated as r= 0.55. Dreisigacker et al. (2004) speculated that by assembling more diversified germplasm pools for different MEs and use specific ME key sites for yield evaluation higher genetic differentiation could achieved. However, increased genetic diversity does not necessarily lead to higher productivity or adaptation. While SSRs are powerful tool for genotype identification, their usefulness for revealing genotype differentiation regarding specific traits such as adaptation to certain MEs could not be proven in their study and warrants further research.

25

Langridge et al. (2001) reviewed the trends in genetic and genome analyses in wheat. The wheat genome is one of the most complexes for genetic analysis due to its size and structure. They reviewed the development of molecular techniques for genetic analysis and in particular the molecular markers to monitor DNA sequence variation between genotypes. A microsatellite map of wheat was constructed by Röder et al. (1998) using the variety ‘Chinese Spring’ as a source for genomic DNA, which was used for microsatellite isolation from various λ phage libraries. The highest number of microsatellites was found in the B genome (115) and the least in D genome (71), while the A genome contained 93 microsatellites. They found that the best way to increase the microsatellite isolation efficiency of functional primer pairs is to use the undermethylated fraction of the wheat genome. Although this map provides a general picture of the wheat genome based on microsatellites, complimentary work is needed in order to construct a saturated map of the wheat genome, based on microsatellite, especially for the agronomically important genes and quantitatively inherited traits. Korzun et al. (1999), however, integrated the dinucleotide microsatellites from hexaploid bread wheat into a genetic linkage map of durum wheat. They used 98 primer pairs of wheat microsatellites located on A and B genomes of hexaploid wheat. They reported that microsatellite developed from hexaploid bread wheat could also be utilized for tetraploid durum wheat. The integration of microsatellites from hexaploid bread wheat into a genetic linkage map of durum wheat will accelerate the transfer of knowledge from bread wheat to durum wheat and facilitate the development of new durum wheat varieties. A genetic linkage map of durum wheat was constructed by Blanco et al. (1998) based on RFLP markers. Other molecular genetic maps have been constructed for group 1 chromosomes of Triticeae species (Van Deynz et al., 1994) and for group 7 chromosomes of bread wheat (Hohmann et al., 1994). The map construction was carried out using RFLP and RAPD markers for the group 1 chromosomes of Triticeae and using only RFLP markers for the group 7 chromosomes of bread wheat. The RFLP maps were

26

not able to show high levels of polymorphism in wheat because hexaploid bread wheat has low level of RFLPs. Stephanson et al. (1998) mapped fifty new microsatellite loci for the wheat genome, most of them located on the A genome (21 loci) and the fewest on the D genome (13 loci). They reported that microsatellite markers are likely to be of limited use for comparative analyses or introgression studies involving wild species related to wheat. On the other hand, they suggested that microsatellite markers are the markers of choice for practical wheat breeding today because of their locus specificity and high level of polymorphism associated with them. Recently, Gupta et al. (2002) mapped 66 new microsatellite loci on bread wheat that were distributed on 20 out of the 21 chromosomes (no marker found for chromosome 6D). They generally classified all molecular markers that have been used for mapping of bread wheat into three categories: (1) those having homoeology (one locus on each of the three chromosomes of a homoeologous group), (2) those having multiple loci, but not on homoeologous chromosomes, and (3) those that are chromosome specific, each within a single locus. Microsatellite primers usually amplify a specific locus each, and therefore they belong to the third category according to Gupta et al. (2002). Isolation and mapping of microsatellite markers specific for the D genome of bread wheat had been done by Pestsova et al. (2000). The diploid progenitor of the D genome of wheat, Aegilops taushii, was used as a source of microsatellite markers for the hexaploid bread wheat. Sixty-five functional microsatellite markers were isolated from Ae. taushii and amplified well in hexaploid wheat. They concluded that microsatellite markers developed from Ae. taushii sequences are as useful for wheat genome analysis as markers from hexaploid wheat. Selectively amplified microsatellite (SAM) analysis was used to assist in construction of a genetic linkage map from an interspesific cross between the bread wheat cultivars ‘Sunco’ and ‘Tasman’ (Hayden et al., 2001). Twenty-two (38%) of the SAMs were located on the short arms of the chromosomes and 36 (62%) were located on the long arm. They reported that SAM analysis may be useful for developing markers for

27

an abundant class of short SSRs that is generally inaccessible to traditional hybridizationbased methods.

5. Genetic diversity based on ISSR Markers Inter simple sequence repeat (ISSR) is a technique that utilizes a single primer which anneals to microsatellite sequences and during the PCR DNA fragments flanked by the primer are amplified. The primer for this technique is constructed to anneal to homologous microsatellite sequences with an additional nucleotide at the 3'-end. The extra nucleotide will allow amplification only if the whole primer sequence binds to the 5'-end of a microsatellite with a homologous nucleotide in the flanking sequence. Such additional nucleotides are termed anchors and they assure that amplification will always start from the 5'-end of the microsatellite. Fragment amplification will only take place if microsatellite arrays, homologous to the primer including the additional nucleotide, are in a distance suitable for PCR amplification. The ISSR markers act either as dominant markers, if changes occur at the anchoring nucleotides, or as co-dominant markers, if the length of the intervening sequence between the microsatellites has changed. The ISSR markers are scored by the presence/absence of bands. The absence of a band can attributed to the divergence occurred at either one of the two primer binding sites, or it could also be that a SSR site (ISSR primer binding site) might have been lost, or a chromosomal rearrangement may have taken place (Wolfe and Liston, 1998). The ISSR technique was first reported by Zietkiewicz et al. (1994). They obtained complex, species-specific patterns and intraspecies polymorphisms from a variety of eukaryotic taxa. They concluded that the ISSR approach is applicable for taxonomic and phylogenetic comparisons as well as a mapping tool in a wide range of organisms.

28

Genetic diversity of different plant species has been studied using ISSRs. Hou et al. (2005) determined the genetic diversity of barley accessions and landraces originating from west China using RAPD and ISSR. According to their results ISSR was found to be better than RAPD in revealing genetic diversity among the barley accessions. Because of a poor correlation (r=0.11) between the two sets of genetic similarity estimates, it was concluded that these markers multiply unrelated parts of the genome. In sorghum and banana, the ISSR technique has also been used to detect genetic diversity (Godwin et al., 1997). Analysis with ISSRs detected a higher level of polymorphism compared to RFLP or RAPD analyses. According to the results this was not due to greater genetic polymorphism, but rather due to technical reasons related to the detection methodology used for ISSR analysis. ISSR markers were used for cultivar identification in durum wheat (Pasqualone et al., 2000). High level of ISSR marker distinguishing efficiently a set of 30 Italian durum wheat cultivars and 22 breeding lines was reported. Two primers were found to be sufficient for identifying all durum wheat entries. Ben El Maati et al. (2004) studied the polymorphism of 28 Moroccan and Mexican wheat varieties using ISSR markers. The results obtained from the ISSR markers were in agreement with the geographical origin of wheat varieties. They reported that ISSR markers can be of help in improvement strategies during selection of resistant varieties to Moroccan wheat diseases. The genetic diversity among selected bread wheat lines differing in heat tolerance was studied by Motawei et al. (2007) using both ISSR and RAPD markers. The dendrogram based on RAPD markers was not in accordance with the dendrogram based on ISSR markers in all cases. The results of the combined dendrogram, generated by both ISSR and RAPD markers, better agrees with the classification depending on pedigree and heat tolerance background than the dendrogram generated by RAPD or ISSR data alone. For Indian tetraploid wheat, Pujar et al., (2002) reported that a small number of ISSR primers (15 out of 100 primers) could detect genetic diversity and the results were consistent across using different distance coefficient in the analysis. A good

29

correspondence between pedigrees of genotypes and ISSR marker results was also obtained. The ISSR markers were found superior compared to RAPD markers. On the other hand, in another study comparing the ISSR, RFLP and RAPD markers in wheat, Nagaoka and Ogihara (1997) reported that the genetic relationships of wheat accessions estimated by the polymorphism of ISSR markers were identical with those inferred by RFLP and RAPD. They reported that polymorphism, generation of information and ease of handling of ISSR markers support their applicability to the analysis of genotypes as well as to the construction of PCR-based genome maps in wheat. Genetic linkage map using ISSR and RAPD markers of Einkorn wheat had been constructed by Kojima et al. (1998) in relation to RFLP markers. Although 44 ISSR fragments and 29 RAPD fragments statistically showed a 3:1 segregation ratio in the F2 population, only 9 markers from each method were able to be mapped on the RFLP linkage map. The comparison of the genetic linkage map of Einkorn with linkage and cytological maps of common wheat revealed that the order of the markers between the two maps coincided, except for chromosome 4A, which harbors rearrangements specific for polyploid wheats, indicating conservation between the two genomes.

6. Correlation between morphological and molecular markers Existence and use of genetic diversity is a prerequisite for any plant breeding program. Determination of this genetic diversity can be achieved through the use of morphological characteristics, biochemical, molecular, or even cytogenetical markers. However, for assessment of genetic diversity, more than one of the above mentioned markers should be used at the same time. Genetic marker used for determination of genetic diversity is important to be reliable. Morphological characteristics are influenced by the environment but molecular markers overcome this defect because they detect diversity at the DNA level. In general, assessment of genetic diversity depends on the type of polymorphism detected and the proportion of the genome covered by the marker system used (Roy et al., 2004).

30

Association between morphological and molecular marker systems had been determined in a number of plant taxa including pepper (Lefebvre et al., 2001; Geleta et al., 2004), olive (Belaj et al., 2003; Hagidimitriou et al., 2005), maize (Dillmann et al., 1997), and barley (Schut et al., 1997; Hamza et al., 2004; Hou et al., 2005). The correlation between genetic similarity estimated by morphological traits and molecular markers for all of the above mentioned studies was low, except for Lefebvre et al. (2001), where molecular genetic distances were significantly correlated with distances based on a set of discriminating agronomic traits. The association between AFLP markers, pedigree data and morphological traits in barley, studied by Schut et al. (1997), revealed a poor to moderate correlation between genetic similarities (gs) based on AFLP markers and pedigree-based coefficients of co-ancestry (f) within a group of European two-row spring barleys. They suggested that this poor relationship might be caused by inappropriate assumptions in the calculation of f as well as marker-sampling error and biased representation of genomic differences revealed by AFLPs. Morphological distances (md) showed no significant relationship with gs or f. This may be caused by biased and insufficient representation of the whole genome using morphological traits. In maize, a comparison of RFLP and morphological distances had been made in a set of Zea mays L. inbred lines (Dillmann et al., 1997). The relationship between molecular and morphological distances was not linear and appeared to be expressed in triangular shape (e.g. low marker distances were systematically associated with low morphological distances and, on the other hand, high marker distances were associated either to high or low morphological distances). It was also observed that molecular divergence behaved as a limiting factor for morphological divergence. Burstin and Charcosset (1997) reported similar results, since the relationship between marker and phenotypic distances in maize, computed from quantitative traits, displayed a triangular shape. The relationship between quantitative and marker distances was affected by the linkage disequilibrium between marker loci and the QTLs considered for distance estimation. Lefebvre et al. (2001) evaluated the genetic distances between pepper-inbred lines using AFLP, RAPD and phenotypic data. The relationships between genetic distance indexes (GDI) estimated

31

with molecular markers and a genetic distance based on morphological traits was studied. Forty-one agronomic traits were recorded recommended by UPOV for assessing distinctiveness, uniformity and stability criteria. The results showed a general agreement between AFLP and RAPD markers (r=0.95). The molecular genetic distances were correlated with distances based on the discriminating agronomic traits. The relationship between molecular and morphological distances appeared to be triangular. In wheat, relationship among morphological traits and molecular markers give diverse results. Assessment of genetic diversity in bread wheat using AFLP markers and agronomic traits revealed low correlation between the genetic distance estimates (Lage et al., 2003). In another study (Moghaddam et al., 2005), when the agronomic performance of the Iranian bread wheat was compared with the AFLP diversity analysis, five of the six drought susceptible genotypes clustered together in the agronomic dendrogram and were located in the same cluster in the AFLP dendrogram. Genetic diversity based on morphological traits and RAPD markers was also studied in hexaploid wheat by Cao et al. (2000) and Maric et al. (2004). Cao et al. (2000) found agreement between the RAPDs and the morphological classification (cluster analysis), while Maric et al. (2004) reported that there was no significant correlation (r=0.12) between RAPD markers and morphological traits of Croatian bread wheat cultivars. In this respect, the efficiency of RAPD, ISSR and RFLP markers in revealing the polymorphism in wheat were compared (Nagaoka and Ogihara, 1997; Kojima et al., 1998). Nagaoka and Ogihara (1997) obtained comparable clustering results using the three marker systems, while Kojima et al. (1998) obtained similar results from RFLP and ISSR markers, but different results from RAPD markers. Recently Khan et al. (2005) used RAPDs, ISSRs and SSRs to identify the markers linked to rust resistance genes Lr3a and Sr22 in Australian wheat. The ISSR marker UBC 840540 was found to be linked with Lr3a gene in repulsion at a distance of 6.0 cM. They also reported that the use of these markers in combination could identify the presence or absence of Sr22 gene in breeding populations. The SSRs had been used for genetic diversity determination studies in wheat in combination with AFLP markers and morphological traits (Bohn et al., 1999; Barrett et

32

al., 2002; Roy et al., 2004; Stodart et al., 2005; Roy et al., 2006). In some cases; molecular markers were associated with important agronomic traits. Molecular markers were found to be linked with important traits, such as protein content, and mapped on chromosome arm 2DL of bread wheat (Prasada et al., 1999 and 2003). Other markers were found to be associated with grain protein content in T. turgidum L. var. dicoccoides using RFLP markers (Mesfin et al., 1999). Linkage of RFLP markers to powdery mildew resistance genes Pm1, Pm2, Pm3 and Pm4 in wheat were studied by Ma et al. (1994). According to their results all of the above genes are located on chromosome 7A except of Pm2 gene, which may be close to the centromere of chromosome 5D. In another association study a microsatellite marker was found close to a QTL responsible for the 1000 grain weight trait in bread wheat (Varshney et al., 2000). They found that one primer pair (wmc333) showed an association of the marker locus xwmc333 with grain weight, confirmed by using a single marker linear regression approach. The marker has been located on chromosome arm 1AS, and the QTL was designated as QQW1.ccsu-1A. Shariflou and Sharp (1999) examined the potential polymorphism of an (AT)n microsatellite at the 3' end of waxy genes in Australian bread wheat. A distinguishable fragment was amplified from chromosome 7D that was absent whenever a plant was null for the waxy gene. Eight alleles on chromosome 7A were also revealed from 135 Australian mainly wheat cultivars. The association of pre-harvest sprouting (PHS) resistance with RFLP and microsatellite molecular markers and phenotypic data [falling number (FA) and α-amylase activity (AA)] was studied by Zanetti et al. (2000) using 226 F2 recombinant inbred lines originating from a Swiss wheat X spelt cross. The two phenotypic traits were high negatively correlated (r= - 0.91). They detected 12 and 13 QTLs for the two traits, respectively. They proposed that these QTLs could be important for marker-assisted selection for PHS resistance in both wheat and spelt germplasm.

7.

33

Cytogenetic studies Cytogenetics, according to Singh (2003), is a hybrid science that combines cytology (the study of chromosomes and other cell components) and genetics (the study of inheritance). The science includes chromosome handling (chromosome staining techniques), function and movement of chromosomes (cell division, mitosis, meiosis), numbers and structure of chromosomes (karyotype analysis), and numerous modifications of structure and behavior as they relate to recombination, transmission, and expression of genes. Most of the DNA in eukaryotic genomes is repetitive. Repetitive DNA is found in two forms, tandem arrays or dispersed repeats. Dispersed repeats are either transposon (move and amplify in DNA form) or retrotransposon (move by means of intermediate RNA). Tandem DNA repeats can vary in length from two base pairs to many thousands base pairs (Flavell et al., 1986) and include satellite DNA (often localized to C-banded regions) and rRNA genes or ribosomal DNA (rDNA) (Katsiotis et al., 1996). Two major types of rRNA genes (rDNA) are the 18S-5.8S-26S rDNA and 5S rDNA arrays. These ribosomal RNA genes are arranged in tandem arrays clustered in specific sites. The rDNA sites can be determined by means of in situ hybridization, a powerful technique that is used to characterize and locate nucleic acid sequences within the nucleus (Schwarzacher and Heslop-Harison, 2000). This method in molecular cytogenetics became increasingly important over the last decade, and has had an important impact on many different areas of life sciences. Use of biotin labeled probes was found to be rapid, consistent, and reliable technique to detect repeated DNA sequences by in situ hybridization (Rayburn and Gill, 1985). In situ hybridization technique was used to detect the sites of rRNA genes in wheat (Rayburn and Gill, 1985; Flavell et al., 1986; Mukai et al., 1990 and 1991; Lou and Devorak., 1998; Lima-Brito et al., 2006). Nucleolus organizing regions (NORs) are the sites of active 18S-5.8S-26S rRNA genes (Luo et al., 1998). In hexaploid wheat, six loci of 5S rRNA genes were identified on the short arm of the chromosomes of homoeologous group 1 and 5 (1A, 1B, 1D, 5A, 5B, and 5D) (Mukai

34

et al. 1990), whereas18S-5.8S-26S rDNA loci were mapped on the short arm of 1A, 1B, 6B, and 5D chromosomes, and the long arm of 7D chromosome (Mukai et al. 1991). Genomic in situ hybridization (GISH) technique was used to determine and identify the Thinopyrum chromosomes in wheat (Schwarzacher et al., 1992; Miller et al., 1995; Chen et al., 1999). The method was able to identify and detect the alien chromatin incorporated into wheat. They reported that GISH is a fast, sensitive, accurate and informative technique. The method was also used in wheat-rye hybrids to identify A, B, D and R genomes (Sanchez-Moran et al., 1999) and fluorescent GISH was used to identify the inverted chromosome and characterize the mildew resistant in wheat-rye substitution lines (Forsstrom et al., 2002). GISH distinguished simultaneously the wheat genomes (A, B and D) as well as the R genome of rye. Their results also showed the presence of rye germplasm in seven different mildew resistant wheat lines derived from crosses between triticale and bread wheat. Identification of the different genomes was also carried out in wheat-barley hybrids (Malysheva et al., 2003) and in wheat T. aestivum L x Jointed goatgrass Ae. Cylindrical host BC1 plants (Wang et al., 2002). Schneider et al. (2003) determined the polymorphism among different hexaploid wheat cultivars with fluorescence in situ hybridization (FISH) using two repetitive DNA clones, pSc119.2 and pAs1. Even though low levels of polymorphism were obtained, it was possible to identify 17 pairs of chromosomes according to their hybridization pattern. Differences were observed for chromosomes 4A, 5A, 1B, 2B, 5B, 6B, 7B, 1D, 2D, 3D and 4D. They assumed that variation in hybridization patterns are caused by chromosome structural rearrangements and by differences in the amount and location of repetitive sequences in the cultivars analyzed. A cytogenetically based physical map of chromosome 1B in common wheat was constructed utilizing a total of 18 homozygous deletion stocks by Kota et al. (1993), dividing chromosome 1B into sub-regions. Nineteen genetic markers are physically mapped to nine sub-regions of chromosome 1B. Comparison of the cytological map of chromosome 1B with an AFLP-based genetic linkage map of T. tauschii revealed that the linear order of the markers was maintained between chromosome 1B in hexaploid wheat

35

and 1D of T. tauschii. They observed striking differences between physical and genetic maps in relation to the relative distances between the genetic markers. The genetic markers that were clustered in the middle of the genetic map were physically located in the distal regions of both arms of chromosome 1B. They reported that it was not clear whether the increased recombination in the distal regions of chromosome 1B was due to specific regions of increased recombination or a more broadly distributed increase of recombination in the distal regions of Triticeae chromosomes. Another cytologically based physical map (CBPMs) of the group-2 chromosomes of common wheat T. aestivum L. was constructed by Delaney et al. (1995). They used 21 homozygous deletion lines for 2A, 2B and 2D to allocate RFLP loci in 19 deletioninterval regions. A consensus CBPM was collinearly arranged with a consensus genetic map of group-2 chromosomes, having a higher frequency of recombination in the distal regions. They concluded that targeted-mapping of specific chromosomal regions containing a gene of interest should be feasible. Their results suggested that the pattern of marker distribution and recombination is similar to those described for the group 7B, 6B and 1B chromosomes of wheat. Recently, genomic in situ hybridization (GISH) and microsatellite markers were used to describe the genomic constitution of wheat-barley hybrids from two backcross generations (Malysheva et al., 2003). Molecular markers detected introgression of small segments of barley genome with only one marker, which probably resulted from recombination between wheat and barley chromosomes. The screening of the backcrossed populations from intergeneric hybrids could be effectively conducted using both, GISH and microsatellite markers. They reported that GISH images presented a general overview of the genome constitution of the hybrid plants, while microsatellite analysis revealed the genetic identity of the alien chromosome segments introgressed. Lima-Brito et al. (2006) analyzed the morphological, yield, cytological and molecular characteristics of bread wheat x tritordeum F1 hybrids (2n = 6x = 42; AABBDHch) and their parents. Morphological differences were found among the hybrids

36

and their parents. The hybrids were slightly bigger than either parent, had more spiklets per spike, and tillered more profusely. Complete homologous pairing (14 bivalents) plus (14 univalents) was obtained in only 9.59% of the pollen mother cells (PMCs) analyzed. The average number of wheat genome univalents (6.8) was higher than of the Hch genome (5.4). Fifty RAPD and 31 ISSR polymorphic bands were amplified between the F1 hybrids and both parents. They reported that the complementary use of morphological, molecular cytogenetic techniques and molecular markers allowed a more accurate evaluation and characterization of their hybrids.

8. Study Objectives: The present study aimed to determine the genetic relationships among 45 wheat entries having different ploidy levels originating from Greece, Egypt, Cyprus and Italy, using morphological traits, molecular markers (SSR and ISSR) and in situ hybridization. Comparisons among morphological, molecular and cytological studies was performed in order to reveal the genetic relationships among the wheat entries used. The aims are summarized as following: 1.

determination of the genetic diversity among 45 bread and durum wheat entries originating from Greece, Cyprus, Egypt and Italy using three types of data: morphological, molecular (SSR and ISSR markers) and cytogenetical markers,

2.

analyze the differences among the wheat entries having different origin, and

3.

compare the results obtained from morphological, molecular and cytological analyses.

37

III. MATERIALS AND METHODS

38

III. MATERIALS AND METHODS The present work was carried out at the Plant Breeding and Biometry Laboratory, Crop Science Department of the Agricultural University of Athens, Greece during the period 2004-2007. The main objective of the present study was to determine the genetic diversity among forty five bread and durum wheat entries originating from Greece, Cyprus, Egypt and Italy using three types of data: morphological, molecular (SSR and ISSR markers) and cytogenetical markers. The morphological data were collected from all entries during two growing seasons (2004/2005 and 2005/2006) in the field and the laboratory, in order to determine the genetic diversity based on the phenotype. The genetic diversity at the molecular level was determined by SSR and ISSR markers and cytogenetical comparisons among entries having different origin and ploidy level was done by the in situ hybridization technique.

A. Plant material: A total of 45 wheat varieties or landraces were used, originating from Greece (20 durum wheat entries and eight bread wheat entries), Egypt (seven bread wheat entries), Cyprus (seven durum wheat entries) and Italy (two durum wheat entries and one bread wheat variety). Some of the entries have known pedigrees while others have unknown background (Table 1). All 45 entries were used for the molecular markers analyses while morphological data were collected from 42 of them during two growing seasons (2004/2005 and 2005/2006) (Table 1). Twelve entries were used for the in situ hybridization representing the four different regions and the two different ploidy levels (Table 2).

39

Table 1: Wheat entries used for morphological traits and molecular marker analyses No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Variety Durum Wheat (4X) Athos (C) Agias (C) Anna (C) Electra* (OC) Electra -A Electra -M Kallithea (C) Limnos (OC) Mexikali- 81 Sifnos (C) Kornos Mavraghani Heraklio* Mavraghani Heraklio -A Mavraghani Heraklio -M Myrina Papadakis (C) Pondos (C) Sanda (OC) Simeto (C) Kyperounda* Kyperounda -A Kyperounda -M Aronas Mesaoria Makedonia Karpasia Selas (C) Skiti (C) Skyros Pietrafitta Bread Wheat (6X) Sakha 61 (C) Sakha 69 (C) Sakha 94 (C) Gemmiza 7 (C) Gemmiza 9 (C) Giza 168 (C) Sids 1 (C) Yecora-E Acheron (C) Verghina (C) Genorozo-E (C) Strymonas (C) Nestos (C) Elisavet (C) Acheloos (C)

Origin

Pedigree

Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Greece Italy Cyprus Cyprus Cyprus Cyprus Cyprus Cyprus Cyprus Greece Greece Greece Italy

HOHENHEIMER/PETIT-QUINQUIN//COTE-D-OR[39] [1790] UNKNOWN MEXIKALI- 81/ SANDA S. Capelli/4/LIMNOS//Florence/Arditto/3/Sinai S. Capelli/4/LIMNOS//Florence/Arditto/3/Sinai S. Capelli/4/LIMNOS//Florence/Arditto/3/Sinai UNKNOWN Local durum wheat selection for landrace Asprostachi Selection from CIMMYT`s variety Mexicali 75 UNKNOWN UNKNOWN Landrace Landrace Landrace UNKNOWN ATHOS/MEXIKALI- 81//MEXIKALI- 81 UNKNOWN Selection from irradiated variety METHONI

Egypt Egypt Egypt Egypt Egypt Egypt Egypt Greece Greece Greece Italy Greece Greece Greece Greece

CAPEITI-8/VALNOVA [1620] [1622] [1623] [1625] [1666]

AMBER AMBER AMBER (RAE (RIALE-E) /4 X TC60 // STEWART 63 /3/ AA “S”) =CISNE or COCORIT 71 (AA “S”/VOLUNTEER) = MIA UNKNOWN PLC “S”/RUFF “S”//GTA “S”/RTTE Selection from the variety Storck "S" Selection of CR "S"/ T.DIC S VERNUM-GLL"S" UNKNOWN Grazia/Isa 61 INIA/RL4220//7C-3/YR"s"CM15430-2S-5S-0S INIA/RL4220//7C/YR"s"CM15430-2S-1S-0S OPATA/RAYON//KAUZ 7CMH74A-630/SX//SERI82/AGENT ALD"s"/HUAC//CMH74A-630/SX MRL/BUC//SERI HD2172/PAVON"s"//1158-57/MAGA74"s" Individual selection from CIMMYT`s variety Yecora 70. LOCAL CULTIVAR

G-38290/YG-3297 (see AIGES) AUTONOMIA//AQUILA/AUTONOMIA [39] [221] [2289]; AUTONOMIA/MARA, ITA[2494]

YAQUI 50/ENANO//KALYANSONA from CIMMYT Selection from segregated material of INIA 66R//HBGN/DRC. UNKNOWN Siete Cerros T-66//Weibulls-Karin/Yt 54B II–7518-3c-2h-1h.

* Varieties not used in the morphological traits experiment; (C) Cultivar, (OC) Old cultivar

40

Table 2: Wheat varieties used for in situ hybridization Variety

Origin Pedigree

Tetraploid varieties

Agias Pondos Kallithea Aronas Mesaoria Pietrafitta Hexaploid varieties

Sakha 94 Giza 168 Acheron Nestos Yecora-E Genorozo-E

Greece Greece Greece Cyprus Cyprus Italy

UNKNOWN UNKNOWN UNKNOWN RAE (RIALE-E) /4 X TC60 // STEWART 63 /3/ AA “S” = CISNE = COCORIT 71 AA “S”/VOLUNTEER = MIA GRAZIA/ISA

Egypt Egypt Greece Greece Greece Italy

OPATA/RAYON//KAUZ MRL/BUC//SERI LOCAL CULTIVAR Selection from segregated material of INIA 66R//HBGN/DRC. Individual selection from CIMMYT`s variety Yecora 70. AUTONOMIA//AQUILA/AUTONOMIA [39] [221] [2289]; AUTONOMIA/MARA, ITA[2494]

B. Morphological Traits: B.1. Experimental design and data collection:

The morphological traits were collected during two growing seasons (2004/2005 and 2005/2006) at the Agricultural University of Athens farm (Appendix 4). One hundred seeds from each variety were spaced planted in three one-meter rows (30 cm between rows) and replicated twice. The entries were randomized in each replication. All morphological characteristics were collected only from the middle row for both replications during the two growing seasons. Thirty entries were planted during the 2004/2005 growing season. It was observed that entries ‘Electra’, ‘Mavraghani Heraklio’ and ‘Kyperounda’ had heads with either black or white awns during the harvest stage. During the growing season 2005/2006, these two types were planted as separate entries [black (M) and white (A) awns] in addition to the 36 entries. Thirty-nine morphological characteristics were measured according to UPOV 2002 and literature (Bohn et al., 1999; Lage et al., 2003) in the field and the laboratory (Table 3). The traits were recorded as

41

qualitative including binary and categorical (ordinal or nominal) traits or quantitative (including continuous and discrete) traits (Table 3 and Figure 2). Most of the data were collected once per growing season, except for five morphological traits (total ear length, head length, number of seeds per head, head weight and seeds weight per head traits) for which data were recorded as an average of ten readings. Another two traits have been scored as a mean of five readings in the field (length and width of the second leaf from top). Fifty seeds from each variety were tested for coloration with phenol for both growing seasons. The seeds were soaked in water for 18 hours in 10-cm Petri dishes and then drained without leaving any water on the seed’s surface. The seeds afterwards were placed with crease downwards. One per cent freshly made phenol solution was applied to the seeds until ¾ of the seeds were covered. The Petri dishes were left in day light (out of direct sunshine) for four hours and after that the color of the seeds was recorded as no coloration (1), light (3), medium (5), dark (7) or very dark (9) (Figure 2B).

B.2. Statistical analysis:

The morphological traits were scored for their phenotypic plasticity. Three categories were formed for each morphological trait using the observations of the entries from each replication and year. To calculate the categories for each morphological trait, scoring data from 33 entries that were planted during both growing seasons were used. The original data used for calculating these percentages are presented in Appendix 4. The first category was ‘stability’ and was calculated as the percentage of the entries having the same score for both replications and during both growing seasons. Traits in this category remain the same and different growing seasons (different environmental conditions) did not affect them. The second category was ‘phenotypic plasticity’ and was calculated as the percentage of the entries having the same score for both replications but different scores during the two growing seasons, indicating that the environment can change their phenotype. Finally, ‘instability’ was obtained when different scores were recorded between replications and growing seasons.

42

The averages of the morphological traits were calculated for each variety (the averages of the two seasons and replications). In order to minimize the environmental effect of the two growing seasons, data for the morphological quantitative traits were normalized using the Z-value and then all morphological data were standardized (Appendix 3). The similarity coefficient matrix was calculated using the simple matching algorithm, which was used to construct a dendrogram based on the Unweighted PairGroup Method with arithmetical algorithms Averages (UPGMA) method (Sneath and Sokal, 1973) and the correlation cophenetic coefficients were calculated. The principal coordinate analysis (PCOORDA) was performed using the NTSYS PC 2.0 (Rohlf 1998) software based on the similarity matrix using the standardized centered data. Morphological data were also used for discriminant analysis using JMP IN 5 software (Lehman et al., SAS institute Inc, 2005). Discriminant analysis was carried out in order to study the effect of wheat type (hard or soft) on morphological traits and also to determine the morphological traits characterizing the entries for each wheat ploidy level. Two-way hierarchical cluster analysis was also done using JMP IN 5 software to insure the results obtained from NTSYS PC software and to determine the relationships among the morphological traits used.

43

Table 3: Wheat morphological traits recorded according to UPOV, 2002 and literature. CODE

TRAIT

SCORING SYSTEM

GROWING STAGE+

1 erect – 3 semi erect – 5 intermediate – 7 semi prostrate – 9 prostrate 1 absent or weak – 2 medium – 3 strong 1 absent or weak – 2 medium – 3 strong 1 absent or weak – 2 medium – 3 strong 1 absent or weak – 2 medium – 3 strong 3 narrow – 5 medium – 7 broad 1 thin – 2 medium - 3 thick 1 slopping – 2 straight – 3 elevated with 2nd beak 1 straight – 2 moderately curved – 3 strongly curved 1 absent – 3 weak – 5 medium – 7 strong 1 very loose – 3 loose – 5 medium – 7 stiff – 9 very stiff 1 whitish – 2 light purple – 3 purple – 4 dark purple 3 acute 90 3 bent – 5 slightly bent – 7 stiff 3 acute 90 3 bent – 5 slightly bent – 7 stiff 3 lax – 5 medium - 7 compact 1 very short or absent – 3 short – 5 medium – 7 long – 9 very long 1 straight – 2 moderate – 3 curved 1 white – 2 colored – 3 brown 1 whitish – 2 light brown or brown –3 black 1 upright – 5 intermediate – 7 lodged 1 straight – 2 moderately curved – 3 curved 3 avoid – 5 semi elongated – 7 elongated 1 none or very light – 3 light – 5 medium – 7 dark – 9 very dark 3 short – 5 medium – 7 long

25 – 29 60 – 69 60 – 69 60 – 69 60 – 69 80 – 92 80 – 92 80 – 92 80 – 92 80 – 92 80 – 92 47 – 51 50 – 53 50 – 53 50 – 53 50 – 53 90 – 92 90 – 92 90 – 92 92 92 92 90 - 92 92 92 92

A:Categorical (Ordinal or Nominal) Traits T1 T3 T4 T5 T6 T7 T8 T10 T11 T12 T13 T14 T15 T16 T17 T18 T22 T23 T24 T26 T27 T28 T29 T30 T31 T32

Growth habit Flag leaf: blade glaucosity Ear glaucosity Culmn glaucosity Peduncle shape Lower glume shoulder width Straw pith in cross section Lower glume shoulder shape Lower glume beak shape Lower glume: External surface hairiness Snap back Auricle color Angle of flag leaf to culmn Rigidity of flag leaf Angle of leaves to culmn Rigidity of leaves Ear density Lower glume beak length Lowest lemma beak shape Ear color at Maturity Awn color at maturity Lodging Ear shape at maturity Grain shape Grain coloration with phenol Grain brush length

B: Quantitative (Continuous or Discrete) traits T2 T19 T20 T25 T33 T34 T35 T36 T37 T38 T39

Heading date (days from planting date until emergence of 50% of the ears) Length of the second leaf from top in cm. Width of the second leaf from top in cm. Plant height in cm. 1000 grains weight (grams) Test weight (g/l) Total ear length (including awns in cm). Head length (cm ). Number of seeds per head Head weight (grams) Seeds weight per head (grams).

50 – 52 50 – 69 50 – 69 75 – 92 92 92 92 92 92 92 92

C: Qualitative Binary Traits T9 T21

Apical Rachis Segment: Hairiness of Convex Surface 1 absent – 2 present Homogeneity 1 absent – 2 present

80 – 92 80 – 92

+ days from planting

44

A

C

D

B

E

Light

Non

Medium

Very dark Dark

45

C Short or absent

Medium

Long

D

Absent

Present

E Thick

Medium

Thin

46

F

White

Yellow

Brown

Black

G

White

Brown

Black

Medium

Compact

H

Lax

Figure 2: Some morphological characteristics measured for the wheat entries, (A) plants in the field; (B) grain coloration with phenol; (C) lower glume peak length; (D) peduncle shape; (E) stem thickness in cross section; (F) Ear color at maturity; (G) awns color at maturity; and (H) ear density at maturity. 47

C. Molecular Markers: C.1. DNA Isolation: Thirty seeds from each entry were planted in plastic containers in a growing chamber at 20oC with 16/8 hours light/dark photoperiod. One-week old healthy leaves were harvested and bulked, immersed in liquid nitrogen and stored at -70oC until DNA extraction. Total genomic DNA was isolated from the stored leaves using the standard CTAB method and then preserved at -20oC until used. DNA was isolated according to the following protocol: 1.

One-week old seedling pathogen-free leaves were collected directly in aluminum foil and immersed immediately in liquid nitrogen. The leaves were preserved at 70oC until used.

2.

Sampled leaves were ground in liquid nitrogen using a pestle and mortar. The tissue powder was transferred in 50-ml falcon tubes; about 12 ml of 2% CTAB solution were added and gently mixed.

3.

The tubes were incubated in a water bath at 65oC for 30 minutes, gently shaken every 5 minutes.

4.

An equal volume (12 ml) of chloroform: isoamyl alcohol (IAA) (24:1) solution was added to the tubes and were centrifuged at 3200 rpm for 12 minutes at RT.

5.

The supernatant was transferred in another clean falcon tube. 5% CTAB solution was added at one fifth of the supernatant’s volume followed by an equal volume of chloroform: IAA (24:1) volume.

6.

The tubes were centrifuged at 3200 rpm for 12 minutes at RT.

7.

The supernatant was taken in another clean falcon tube and then three volumes of cooled ethanol were added in order to precipitate the DNA. The DNA was collected using a glass hook and transferred to a 1.5 ml eppendorf tube.

8.

The DNA was left 30 minutes to dry and 500 µl of TE buffer (1X) was added to dissolve it.

9.

7 µl of RNase A (10mg/ml) were added to the dissolved DNA, gently mixed and

48

incubated at 37oC for one and half hours in order to digest the RNA. 10. 500 µl of phenol were added to the eppendorf, mixed and centrifuged at 14000 rpm for 4 minutes at RT. 11. The supernatant was transferred in a clean 1.5-ml eppendorf tube. 250 µl of chloroform:IAA (24:1) and 250 µl of phenol were added to the DNA solution, mixed and centrifuged at 14000 rpm for 4 minutes at RT. 12. 500 µl of chloroform:IAA (24:1) were added to the supernatant in another clean eppendorf tube, mixed and centrifuged at 14000 rpm for 4 minutes at RT. 13. One tenth volume of sodium acetate (CH3COONa) and about 1 ml (three volumes) of cooled ethanol were added to the supernatant in another clean eppendorf tube and mixed gently to precipitate the DNA. 14. The eppendorfs were centrifuged at 14000 rpm for 4 minutes. The supernatant was discarded and the pelt was left to dry for about 30 minutes at RT. 15. 50 µl of TE buffer were added to the precipitated DNA, dissolved and preserved in -20o C until used.

16. Isolated total genomic DNA was run on a 1% agarose gel electrophoresis. C.2. Determination of DNA concentration: The optical densities (OD) of the DNA samples were determined at 260 nm and 280 nm wave lengths using a spectrophotometer device (HITACHI, U-2001, 121-0032). The OD at both wave lengths was recorded for 5 µl of isolated DNA added in 495 µl of de-ionized water. A spectrophotometer reading of 1.000 indicates concentration of 50 µg/ml of double-stranded DNA. The OD ratio A260:A280 represent the purity of DNA; the desired ratio for molecular experiment should be close to 1.8. DNA concentration was determined using the following equation: DNA concentration (µg/µl) = [(A260/OD) X 50 X dilution factor (500/5)]/1000 The final concentration of DNA was 25 ng/µl.

49

C.3. Simple Sequence Repeat (SSR) Reactions: Fourteen wheat specific primer pairs were tested for SSR analysis, most of them located on different chromosomes according to literature (Korzun et al., 1997b; Gupta et al., 2002). Eleven primer pairs were selected as the most informative ones to carry out the analysis (Table 4). The primer sequences were constructed by Invitrogen (Germany). The lyophilized primers were dissolved in de-ionized sterile water at a 10-µM concentration. The PCR amplification reactions were performed in a 25 µl volume using 50 ng DNA containing 0.5 µmoles of each primer pair, 100 µM of dNTPs, 5 µl (1X) of Taq polymerase buffer, 1 mM MgCl2 and 0.5 U Taq DNA polymerase (Promega). The SSR reactions were carried out using Touchdown PCR program. The main program was: 7 cycles at 94oC for 1 min, 59oC for 1 min, decreasing 1oC in every cycle, and 72oC for 1 min, followed by 28 cycles at 94oC for 1 min, 52oC for 1 min and 72oC for 1 min. The previous cycles were preceded by a denaturation step at 94oC for 3 minutes. An extension step at 72oC for 5 minutes was added at the end of the program. PCR products were run on 2% agarose gel electrophoresis to ensure that there was a product present before loading onto the polyacrylamide gel. The PCR products were separated on a 6% denaturing polyacrylamide gel electrophoresis (Appendix 1). Denaturing polyacrylamide gel was stained with silver staining kit (Promega Silver Sequence DNA Staining Reagents, No. Q4134) according to the manufacturer’s protocol, dried and scanned as a permanent image. C.4. Inter-Simple Sequence Repeat (ISSR) Reactions: Nine SSR-anchored primers were selected in order to carry out the ISSR analysis (Table 4). The 25 µl reaction volume contained 5 µl of reaction buffer (1X), 200 µM dNTPs, 1.5 mM MgCl2, 0.75 µM of primer, 0.75 U Taq DNA polymerase (Promega) and 25 ng of genomic DNA. The PCR program included a denaturation step at 94 oC for 3

50

51

minutes, followed by 28 cycles at 94oC for 40 sec, 47oC for 45 sec and 72oC for 2 min, and a final extension step at 72 oC for 7 minutes. The PCR products were run on 2% agarose gel electrophoresis to ensure that there was a product present before loading onto the polyacrylamide gel. The PCR products were separated on a 6% denaturing polyacrylamide gel electrophoresis and silver stained, using similar conditions described for SSR analysis. C.5. Statistical analysis: Both SSR and ISSR gels were scored as 0/1 for absence/presence of the DNA bands, respectively. Similarity coefficient matrices were calculated using the Jaccard similarity algorithm (Jaccard 1908) for ISSR markers, which is a dominant marker, and the simple matching algorithm for the SSR codominant marker. Polymorphism information contents (PIC) were calculated according to Anderson et al. (1993) using the following simplified formula, assuming that the inbred wheat lines are homozygous. PICi = 1 ‫ ـــ‬Σp2ij Where pij is the frequency of the jth allele for marker ith summed across all alleles for the locus. This value provides an estimate of discriminatory power of a microsatellite locus by taking into account not only the number of alleles per locus, but also their relative frequencies in the population studied (Ribeiro-Carvalho et al., 2004). Dendrograms were constructed using the UPGMA method and the correlation cophenetic coefficients were calculated. Principal coordinate analysis (PCOORDA) was also performed on the basis of the distance matrices, using the standardized centered data. For the above mentioned analyses, the NTSYS PC2.0 software was used (Rohlf 1998). Correlations

among

all

obtained

similarity

matrices

(molecular

and

morphological) were performed using the Mantel's test (Mantel, 1967) of the NTSYS PC2.0 software.

52

D. In Situ Hybridization: Twelve wheat entries were used for in situ hybridization (Table 2). The selected entries represented the four different origins (Greece, Egypt, Cyprus and Italy) and the two different ploidy levels (tetraploid and hexaploid). The cytogenetical experiment (in situ hybridization) was carried out in order to identify and determine the differences among the wheat entries used based on hybridization sites of the 5S and the 18S-5.8S26S ribosomal DNA (rDNA). The in situ hybridization technique is described by Schwarzacher and Heslop-Harrison (2000). Two probes were used to detect rDNA sites (5S and 18S-5.8S-26S). The probes were labeled using two types of labels (biotin and digoxigenine). For labeling, biotin-16dUTP and digoxigenin-16-dUTP (Roche Diagnostics) were incorporated in separate reactions. PCR was used for labeling the probes. Chromosomes were counterstained with DAPI (4’, 6-diamidino-2-phenylindole) using 100 µl for each slide. The metaphase preparations were photographed using an epifluorescence microscope and then the photos were optimized through Adobe Photoshop CS2 software. The number of chromosomes, 5S rDNA and 18S-5.8S-26S rDNA were counted for each selected metaphase preparation. The method was carried out as follows: D.1. Slide treatment: a. Place slides in chromium trioxide solution in 80% (w/v) sulpheric acid (referred to as chromic acid) or 11N HCl for at least 3 hours at room temperature. b. Wash the slides with running tap water for 5 minutes. c. Rinse the slides with distilled water and then dry them at 37oC overnight.

53

d. Immerse the slides in 100% ethanol alcohol until used. D.2. Root tips preparation: a. In 10-cm diameter Petri dishes, put about 20 seeds of each variety onto filter paper, saturated with distilled water, at 25oC in the dark for 24 hours. b. Put the Petri dishes at 4oC for 24 hours in order to increase the metaphase index and synchronize the dividing cell population. c. Return the Petri dishes at 25oC for 24 hours to recover approximately one cell cycle period. d. Cut 10-mm long root tips from the seedlings and immerse them in ice cold water (in 1.5 ml eppendorff tubes) for 24 hours to accumulate metaphases. e. Transfer the root tips in a 3:1 fixative solution (100% ethanol: glacial acetic acid) and store at -20oC until used. D.3. Preparation of plant chromosomes: a. Remove the fixative solution from the root tips by washing for 10 minutes in 5 ml of 1X washing solution (prepared from 40 ml of 100 mM citric acid + 60 ml of 100 mM tri-sodium-citrate, pH 4.8). b. Transfer the root tips in 1.5 ml enzyme solution [2% (w/v) cellulose from Aspergillus niger (Calbiochem, 21947, 4000 units g-1, final concentration: 80 units ml-1) and 3% (v/v) pectinase (Sigma, P-4716) from A. niger] and digest at 37 oC for about 50 minutes. c. Transfer root tips into 5 ml of 1X washing solution in Petri dishes to wash. Prepare 50% acetic acid solution. d. Dissect one root tip on an acid-cleaned slide to make chromosome preparations, drop about 30 µl (one drop) of 50% acetic acid freshely made to increase dispersion of cytoplasm, try to remove the root cap and tease out the cells from

54

the remaining tissue. e.

Apply a glass coverslip to the material without trapping air bubbles, tap gently the coverslip with a needle or with a pencil eraser to disperse the cells and then squash the cells using the thumb with light pressure that just turns the nail white.

f. Immerse the slide in liquid nitrogen for 30 seconds or place the slide in -70oC for about one hour immediately after preparation and then immediately remove the coverslip with help of a razor blade. Allow the slide to air-dry. g. Screen the slides under light microscope to find the coordinates with suitable cells for in situ hybridization. The slides can be stored at -20oC for up to three months. D.4. In Situ Hybridization: a. Pepsin treatment: 1. Mark the slide using a diamond pencil. 2. Add 100 µl of pepsin solution [1 pepsin (500 µg/ml) : 75 of 0.01 M HCl] on the slide, cover with a plastic coverslip and incubate at 37 oC for one hour. 3. Wash the slides for 5 minutes in 2X SSC. b. RNase treatment: 1. Dilute RNase stock solution (10 mg/ml) in 2X SSC (1 RNase : 100 2X SSC). 2. Add 100 µl from diluted RNase solution to the slide and cover with plastic coverslip. Incubate at 37 oC for one hour. 3. Wash the slides twice for 5 minutes in 2X SSC solution while shaking at room temperature. c. Fixation: 1. Add 4 g of paraformaldhyde in 60 ml of water. Dissolve at 60oC until the solution becomes milky in color and then add 20 ml of 0.1 M NaOH.

55

Bring the final volume to 100 ml by adding ultrapure water. 2. Incubate the slides in the fixative solution for 10 minutes at room temperature with shaking. 3. Wash the slides twice for 5 minutes in 2X SSC solution while shaking at room temperature. d. Drying: 1. Incubate the slides into 70% ethanol for 3 minutes at room temperature. 2. Incubate the slides into 100% ethanol for 3 minutes at room temperature. 3. Let the slides to air-dry. e. Hybridization: 1. Prepare the hybridization solution (30 µl) for each slide as follows: Formamide

15 µl

20X SSC

3 µl

50% Dextran Sulphate (w/v)

6 µl

Blocking DNA

1 µl

5S probe

2 µl

18S-5.8S-26S probe

2 µl

2. Prepare the hybridization solution for all the slides in one vial and heat for 10 minutes at 70oC in order to denaturate the probes. 3. Place the eppendorf in ice for 5 minutes. 4. Drop 30 µl of the hybridization solution on the slide and cover immediately with the plastic coverslip. 5. Put the slides in the in situ hybridization apparatus. Use the following program: 70oC for 5 minutes, 55oC for 2 minutes, 50oC for 1 minute, 45oC for 2 minutes and 40oC for 2 minutes. 6. Leave the slides at 37oC overnight (12-16 hours) for hybridization in the hybridization oven. f. Discard non-homologous hybridized conjugates (stringent wash): 1. Take out the slides from the hybridization oven and wash them for 5

56

minutes in 2X SSC at 42oC with shaking. 2. Wash the slides in 50% formamide solution diluted in 2X SSC (stringent wash solution) for 10 minutes at 42oC. 3. Wash the slides twice for 5 minutes each in 2X SSC solution with shaking at 42oC. 4. Wash the slides twice for 5 minutes each in 2X SSC solution with shaking at room temperature. g. Detection: 1. Wash the slides once for 5 minutes in 4X SSC/Tween (0.2%) solution with shaking at room temperature. 2. Add 100 µl of detection solution on each slide. 3. Incubate the slides at 37oC for one hour. 4. Wash the slides twice for 5 minutes in 4X SSC solution at room temperature while shaking. 5. Add 100 µl DAPI on each slide for 5 minutes. 6. Wash the slides in 4X SSC at room temperature for 5 minutes.

7. Add a drop (about 30 µl) of anti-fade (Chem. Lab. Canterbury, CT2 7NH) solution and apply a glass coverslip.

57

IV. RESULTS

58

IV. RESULTS A. Morphological Traits A.1. Phenotypic plasticity: The ability of an organism with a given genotype to change its phenotype in response to changes of the environment is called phenotypic plasticity. Such plasticity in some cases is expressed as several highly morphologically distinct results; in other cases, continuous norm of reactions describe the functional interrelationship of a range of environments to a range of phenotypes (Bradshaw, 1965). Organisms of fixed genotype may differ in the amount of phenotypic plasticity they display when exposed to the same environmental change. Hence phenotypic plasticity can evolve and be adaptive, if fitness is increased by changing phenotype. Immobile organisms such as plants have well developed phenotypic plasticity giving a clue to the adaptive significance of plasticity (Wu et al., 2004). According to the above mentioned definition, some qualitative morphological traits were found to be stable across the different growing seasons (different environmental conditions) and some other expressed phenotypic plasticity (different scores between the two growing seasons). The first group included 14 qualitative morphological traits and showed scoring stability between the replications for both seasons (in more than 50% of the wheat entries) (Table 5). The stability of these traits ranged from 97% (for the angle of flag leaf to culmn) to 51.5% (for culmn glaucosity). These traits showed low levels of phenotypic plasticity ranging from 0% (for angle of flag leaf to culmn, awn color at maturity, lodging and ear color at maturity) to 18.2% (for culmn glaucosity) (Table 5). The second group included 14 qualitative morphological traits, showing higher levels of plasticity compared to the previous group, while at the same time their stability through the two growing seasons was less than 50%. For these

59

traits, phenotypic plasticity ranged from 36.4% (for peduncle shape trait) to 9.1% (for straw pith in cross section, lower glume beak shape, lower glume external surface hairiness and ear shape at maturity) (Table 5). These traits seem to be affected by the environment more than the previous traits (first group). For all the qualitative traits (both groups), instability scores (between the replications and the growing seasons) ranged from 3% (angle of flag leaf to culmn) to 63.6% (rigidity of flag leaf). The quantitative traits showed different percentage of phenotypic plasticity. This means, as expected, that quantitative traits are affected by the environmental changes more than the qualitative traits and that a number of genes control each of these traits (Abercrombie et al., 1990; Robinson, 2003). Heading date, plant height, 1000 grains weight, head weight, test weight (g/l) and seeds weight per head showed the highest percentage of phenotypic plasticity (Table 5). On the other hand, width and length of second leaf from top showed the lowest levels of phenotypic plasticity from quantitative traits accounting for 12.1% and 15.1%, respectively (Table 5). The percentage stability for heading date, plant height and test weight traits was zero, while the instability for all quantitative traits ranged from 0% to 42.4% (Table 5).

60

Table 5: Stability, instability and plasticity for 39 morphological traits of two replications during two growing seasons. Trait % Stability % Plasticity % Instability Qualitative Traits High stability & Low plasticity Angle of flag leaf to culmn Angle of leaves to culmn Awn color at maturity Grain coloration with phenol Lodging Ear density Ear color at Maturity Apical Rachis Hairiness of Convex Surface Grain brush length Homogeneity Snap back Grain shape Ear glaucosity Culmn glaucosity

97 87.9 87.9 87.9 81.8 78.8 75.8 66.7 66.7 60.7 54.5 54.5 51.5 51.5

0 3 0 6 0 9.1 0 9.1 15.1 6 6 3 12.1 18.2

3 9.1 12.1 6 18.2 12.1 24.2 24.2 18.2 33.3 39.4 42.4 36.4 30.3

36.4 21.2 36.4 36.4 30.3 45.5 18.2 36.4 39.4 24.2 45.5 33.3 33.3 39.4

36.4 30.3 27.2 24.2 21.2 21.2 21.2 15.2 12.2 12.1 9.1 9.1 9.1 9.1

27.3 48.5 36.4 39.4 48.5 33.3 60.6 48.5 48.5 63.6 45.5 57.6 57.6 51.5

81.8 57.6 39.3 39.3 24.2 12 6.1 3 0 0 0

12.1 15.1 36.4 30.3 33.3 88 93.9 97 100 97 91

6.1 27.3 24.2 30.3 42.4 0 0 0 0 3 9

Low stability & High plasticity Peduncle shape Lowest lemma beak shape Growth habit Flag leaf: blade glaucosity Lower glume shoulder width Auricle color Rigidity of leaves Lower glume beak length Lower glume shoulder shape Rigidity of flag leaf Straw pith in cross section Lower glume beak shape Lower glume: External surface hairiness Ear shape at maturity Quantitative Traits Width of the second leaf from top in cm Length of the second leaf from top in cm Total ear length (including awns in cm). Head length (cm). Number of seeds per head Seeds weight per head (grams). Head weight (grams) 1000 grains weight (grams) Heading date (days) Plant height in cm. Test weight (g/l)

61

A.2. Two-way hierarchical Cluster Analysis: A two-way hierarchical cluster analysis was carried out using JMP IN 5 software for the 45 wheat entries and the 39 morphological characters. According to the above mentioned analysis, the wheat entries were separated into two main groups (A for the tetraploid and B for the hexaploid wheat entries, Figure 3). The tetraploid wheat group (A group) separated into two sub-groups, A1 and A2. Sub-group A2 included the three mixed color awn entries (i.e. ‘Elektra’, ‘Mavraghani Hiraklio’ and ‘Kyperounda’). Although they are grouped in the same cluster, the different awn-color types (white or black) were not found to be closely related except for ‘Kyperounda’. This sub-group also contained ‘Myrina’, ‘Limnos’ and ‘Kornos’ which are all Greek landraces. Sub-group A1 contained the other tetraploid wheat entries including Greek, Cypriot and Italian originating entries (Figure 3). The closest two entries in this subgroup were ‘Mexikali-81’ (a selection from ‘Mexikali’) and ‘Selas’ (a Greek variety) which were grouped along with ‘Pondos’, ‘Sifnos’ (Greek landraces) and ‘Anna’ (‘Mexikali-81’ is a parent for ‘Anna’). The hexaploid wheat group (group B) formed three clusters B1, B2 and B3. The first cluster (B1) contained the three varieties of ‘Sakha’ (61, 69 and 94) along with ‘Yecora-E’ (selection from ‘Yecora’) (Figure 3). ‘Yecora’ is a common ancestor in the above mentioned varieties. The second cluster (sub-group B2) included all the other Egyptian wheat varieties along with ‘Acheron’, ‘Elisavet’, ‘Strymonas’ and ‘Acheloos’ (Greek varieties). The third cluster (B3) contained two Greek varieties (‘Verghina’ and ‘Nestos’) besides ‘Genorozo-E’ (a Greek selection from the Italian variety ‘Genorozo’), (Figure 3). Based on the results, no clustering according to the geographic origin of the entries was observed for both ploidy levels (Figure 3). The correlation among all morphological traits was calculated using the multivariate analysis in JMP IN 5 software, and are presented in Appendix 5. Traits having high correlation values (≥0.73) are included in Table 6, revealing the relationships among these traits.

62

Growth habit Heading date Width of the second leaf from top in cm Lower glume beak shape Lowest lemma beak shape Lower glume: External surface hairiness Ear density Awns color at maturity Lodging Plant height in cm Homogeneity Grain shape Total ear length including ear and awns in cm Ear Color at Maturity Length of the second leaf from top in cm Test weight (g/hl) Auricle color Straw pith in cross section Rigidity of flag leaf Rigidity of leaves Peduncle shape 1000 grains weight in grams Head weight in gram Seeds weight per head in gram Angle of flag leaf to culmn Angle of leaves to culmn Ear shape at maturity Head length in cm Grain coloration with phenol Grain brush length Apical Rachis Segment: Hairiness of Convex Surface Lower glume shoulder width Snub back Number of seeds per head Lower glume shoulder shape Lower glume beak length Flag leaf: blade glaucosity Culm glaucosity Ear glaucosity

ATHOS PAPADAKIS KALLITHEA AGIAS SIMETO ARONAS MAKEDONIA SKITI SKYROS SANDA MESAORIA KARPASIA PIETRAFITTA ANNA MEXIKALI-81 SELAS PONDOS SIFNOS ILEKTRA-A MAVRA_HERAK-A MYRINA ILEKTRA-M MAVRA_HERAK-M LIMNOS KORNOS KYPEROUNDA-A KYPEROUNDA-M YECORA-E SAKHA-69 SAKHA-61 SAKHA-94 GEMMIZA-7 SIDS-1 ACHERON GEMMIZA-9 GIZA-168 ELISAVET STRYMONAS ACHELOOS VERGHINA GENEROZO-E NESTOS

A1

A

A2

B1

B2

B

B3

a b1

b

b2

Figure 3: Two-way hierarchical clustering of wheat entries using Ward’s method in JMP IN 5 software.

63

In the second way of the hierarchical clustering (traits clustering), the morphological traits were separated into two groups, a and b (Figure 3). Group a included nine qualitative morphological traits along with six quantitative morphological traits and one binary qualitative. The two morphological traits showing high correlation in this group were lodging (qualitative) and plant height (quantitative) (r=0.84, Table 6). Group b was divided into two sub-groups, b1 and b2. Sub-group b1 included three quantitative morphological traits and five qualitative. A high correlation value (r=0.93) was obtained between two qualitative traits (rigidity of flag leaf and rigidity of leaves) and between two quantitative traits (head weight and seeds weight per head) (Table 6). Sub-group b2 contained twelve qualitative morphological traits in addition to two quantitative traits and one binary (hairiness of convex surface of apical rachis segment). Many pairs of traits were found to be highly correlated in this sub-group; one pair was a qualitative trait and a quantitative one (ear shape at maturity and head length, r=0.76), (Figure 3 and Table 6). The same qualitative trait was negatively correlated with ear density (r= -0.90) (Table 6). A highly related pair of qualitative traits was also found in this group (rigidity of flag leaf to culmn and rigidity of leaves to culmn, r=0.95). Another pair of qualitative traits that were also highly correlated was grain coloration with phenol and grain brush length (r=0.79). The results also showed that three qualitative traits concerning the glaucosity were highly correlated (glaucosity of flag leaf blade, culmn glaucosity and ear glaucosity). The correlation between the first two traits was found to be 0.88 and between culmn and ear glaucosity was 0.84, while between blade of flag leaf and ear glaucosity was 0.73 (Figure 3 and Table 6).

64

Table 6: Highly correlated morphological characters.

First trait (type) Lodging (ordinal) Rigidity of flag leaf (nominal) Angle of flag leaf to culmn (nominal) Ear shape at maturity (nominal) Ear density (nominal) Grain coloration with phenol (ordinal) Head weight in grams (continuous) Blade of flag leaf glaucosity (ordinal) Blade of flag leaf glaucosity (ordinal) Culm glaucosity (ordinal)

Second trait (type)

Correlation value 0.84 Plant height in cm (continuous) 0.93 Rigidity of leaves (nominal) 0.95 Angle of leaves to culmn (nominal) 0.76 Head length in cm (continuous) -0.90 Head length in cm (continuous) 0.79 Grain brush length (ordinal) 0.93 Seeds weight per head in grams (continuous) 0.88 Culm glaucosity (ordinal) 0.73 Ear glaucosity (ordinal) 0.84 Ear glaucosity (ordinal)

A.3. Cluster Analysis: According to cluster analysis, wheat entries were clustered into two major groups (i.e. A, tetraploid durum wheat entries and B, hexaploid bread wheat entries; Figure 4). Hexaploid wheat entries were found to be well separated from tetraploids. Similarity coefficient for tetraploid entries ranged from 0.13 (between ‘Athos’ and ‘Sifnos’; ‘Mexikali-81’ and ‘Kyperounda-A’; ‘Pondos’ and ‘Kyperounda-A’; ‘Mavraghani Hiraklio-M’ and ‘Pondos’; and ‘Mavraghani Hiraklio-A’ and ‘Papadakis’) to 0.56 (between ‘Kyperounda-A’ and ‘Kyperounda-M’; and ‘Skiti’ and ‘Selas’). The similarity coefficient for hexaploid entries ranged from 0.13 between ‘Verghina’ and ‘Yecora-E’ to 0.54 between ‘Genorozo-E’ and ‘Verghina’ (Appendix 6). The tetraploid wheat entries formed two sub-groups (A1 and A2, Figure 4). Sub-group A1 included the three mixed color awn entries (i.e. ‘Elektra’, ‘Mavraghani Hiraklio’ and ‘Kyperounda’) along with ‘Myrina’, ‘Limnos’ and ‘Kornos’, which they are all Greek landraces. High similarity coefficient between ‘Kyperounda-A’ and ‘Kyperounda-M’ (0.564) was recorded. The other two mixed color awn entries were not found to cluster like the ‘Kyperounda’ types, although they are located in the same cluster (sub-group A1). The rest of the tetraploid wheat entries formed another sub-group including Greek, Cypriot and Italian originating varieties (sub-group A2). Entries ‘Selas’ and ‘Skiti’ showed the highest morphological homogeneity having a 0.56 similarity coefficient.

65

ATHOS KALLITHEA PAPADAKIS SANDA SIMETO MESAORIA SKYROS AGIAS

A2

ANNA PONDOS MEXIKALI-81 SELAS SKITI

4X

SIFNOS MAKEDONI A ARONAS

A

KARPASIA PIETRAFITTA ILEKTRA-A MYRINA LIMNOS MAVRA_HERAK-M KYPEROUNDA-A KYPEROUNDA-M

A1

ILEKTRA-M MAVRA_HERAK-A KORNOS YECORA-E SAKHA-61

B1

SAKHA-94 SAKHA-69 GIZA-168

B2

ELISAVET SIDS-1

6X

NESTOS

B

ACHERON GEMMIZA-7 VERGHINA GENEROZO-E

B3

GEMMIZA-9 STRYMONAS ACHELOOS

0.18

0.28

0.38 Coefficient

0.48

0.58

Figure 4: Cluster analysis of wheat entries based on morphological traits using the simple matching coefficient and the UPGMA clustering method (cophenetic correlation coefficient r=0.73).

66

In group B, three sub-groups were formed namely B1, B2 and B3. Three entries were included in the first sub-group (B1), ‘Yecora-E’, ‘Sakha-61’ and ‘Sakha-94’. The second sub-group (B2) contained three Egyptian entries ‘Sakha-69’, ‘Giza-168’ and ‘Sids-1’, and three Greek entries, ‘Elisavet’, 'Nestos’ and ‘Acheron’ (Figure 4). The third sub-group (B3) included entries ‘Acheloos’, ‘Strymonas’, ‘Verghina’, ‘Genorozo-E’ and the Egyptian entries ‘Gemmiza-7’ and ‘Gemmiza-9’. The highest morphological similarity in group B was obtained between ‘Verghina’ and ‘Genorozo-E’ (sub-group B3) with a 0.54 similarity coefficient (Figure 4). The general trend for both hexaploid and tetraploid wheat groups is that the entries were dispersed throughout the cluster regardless of their country of origin. A good goodness of fit (r = 0.73) between the clustering method (UPGMA) and the simple matching similarity coefficient was revealed by the cophenetic correlation coefficient. A.4. Principal Coordinate Analysis (PCOORDA): Principal coordinate analysis separated all wheat entries by using the first three principal coordinates PC1, PC2 and PC3, accounting for 22%, 14% and 10.6% of the total genetic variance, respectively (totaling 46.6%). Wheat entries were separated according to their ploidy level by the first principal coordinate (PC1) (Figure 5). The tetraploid wheat entries formed four sub-groups. Most of the Greek entries were included the first (sub-group 1) (loaded very high on the second principal coordinate PC2) and the fourth (sub-group 4) (loaded low on PC2) sub-groups (Figure 5). The variables in these two sub-groups showed intermediate variation on the third principal coordinate (PC3) compared to the other sub-groups. Sub-group 2 contained the Cypriot entries, except the ‘Kyperounda’ entries (which were included in sub-group 4) in addition to the Italian originating entries ‘Simeto’ and ‘Pietrafitta’ and the two Greek tetraploid wheat entries ‘Skyros’ and ‘Papadakis’. ‘Simeto’ loaded low on PC3, while ‘Papadakis’ loaded maximum variation on PC3. Sub-group 3 contained the three Greek entries ‘Kallithea’, ‘Athos’ and ‘Sanda’. The entries in this sub-group showed different

67

PAPADAKIS

KALLITHEA PONDOS SIFNOS ANNA

PC1= 22% PC2= 14% PC3= 10.6%

MEXIKALI-81 SAKHA-69 YECORA-E

AGIAS

ELISAVET

GIZA-168 SAKHA-94

ATHOS

SELAS SKITI

ILEKTRA-A MAVRA_HERAK-A MYRINA

MAKEDONIAARONAS

ILEKTRA-M MAVRA_HERAK-M

SKYROS KARPASIA

ACHERON

ACHELOOS GEMMIZA-9 SIDS-1

MESAORIA

KYPEROUNDA-M

PIETRAFITTA

KYPEROUNDA-A LIMNOS

STRYMONAS

PC3

SIMETO

SAKHA-61

KORNOS

1 2 VERGHINA NESTOS SANDA

GEMMIZA-7

3

GENEROZO-E

4

PC2 B A

Bread entries

Durum entries

PC1

Figure 5: 3-Dimensional PCOORDA analysis for wheat varieties obtained by the morphological traits.

68

loads of variation (minimum loading for ‘Sanda’ to very high loading for ‘Kallithea’) on PC3 (Figure 5). ‘Papadakis’, ‘Pondos’, ‘Kallithea’, ‘Sifnos’, ‘Anna’, ‘Mexikali-81’, and ‘Athos’ loaded maximum on PC3, while ‘Simeto’ loaded low on the same principal coordinate and ‘Sanda’ had no load (Figure 5). Most of hexaploid bread wheat entries were aggregated together on the middle of PC2 except for ‘Nestos’ and ‘Sakha-61’, which were found separated near the edges. ‘Verghina’ and ‘Genorozo-E’ were sub-grouped together somewhere between ‘Nestos’, ‘Sakha-61’ and the main group of hexaploid entries (Figure 5). Based on the third principal coordinate (PC3), most of the hexaploid bread wheat entries of the main group showed similar variation proportion except for ‘Gemmiza-7’ which had low variation on PC3 (Figure 5). A.5. Discriminant Analysis: The discriminant analysis was carried out in order to study the effect of morphological traits on ploidy level (tetraploid and hexaploid) and to determine which morphological traits characterize ploidy level of wheat. Wheat entries were separated into two distinct groups (hard-tetraploid and soft-hexaploid) on the first discriminant canonical with 100% of the genetic variance (Figure 6). Tetraploid wheat entries (hard or H) were grouped at the left side of the first canonical while hexaploid wheat entries (soft or S) were grouped at the right side of the same canonical (Figure 6). Tetraploid (hard) wheat entries were characterized mainly by nine qualitative morphological traits (grain shape, ear density, angle of leaves to culmn, straw pith in cross section, rigidity of leaves, ear glaucosity, lower glume beak shape, growth habit and peduncle shape) and only one quantitative trait (length of second leaf from top). Hexaploid (soft) wheat entries, on the other hand, were characterized by four qualitative morphological traits [ear color at maturity (nominal), glaucosity of blade of flag leaf (ordinal), lower glume shoulder width (ordinal) and apical rachis hairiness (binary)] and two quantitative traits [plant height

69

(continuous) and number of seeds per head (discrete)]. The other morphological traits had a neutral effect in separating soft and hard wheat entries (Figure 6).

10

8

S

H

Canonical2

9

7

6

5 -7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

Canonical1 Figure 6: Discriminant analysis for the two wheat types (hard, H and soft, S).

B. Simple Sequence Repeats (SSRs) Analysis The PCR-products were first checked in a 3% agarose gel and they were then loaded in 6% polyacrylamide gel. Since polyacrylamide gel was more informative than agarose gel for all primer pairs, SSR data were collected only from the former. In general,

70

6

the number of alleles obtained in hexaploid was higher than in tetraploid wheat entries. Many stutter bands were obtained over and under the expected SSR molecular weights. These stutter bands were not taken into account during the analysis and in the calculation of the allele number. The stutter fainter bands are produced by the error of the polymerase during the PCR amplification as non-specific products (Manifesto et al., 2001). Two or three co-migrating bands were obtained in a number of SSR primer pairs. Thirty six bands were counted in the polyacrylamide gel for primer pair WMS 52, which amplified a SSR site found on chromosome 3DL. Only 12 of these bands (representing a locus) were considered in the analysis giving a total of 12 alleles (Figure 7.1). The sizes of these bands ranged from 145-185bp and the pair of co-migrating bands, having a constant molecular weight difference, was considered as a single allele. Primer pair Wmc 233 amplified a SSR site found on chromosome 5DS, producing a total of nine bands, three of which were polymorphic representing one locus (Figure 7.2). Three alleles were found ranging from 280-310bp in size. Both primer pairs, WMS 52 and Wmc 233, produced bands only in the hexaploid wheat entries as expected (D genome). Primer pair Wmc 256 amplified a SSR site found on chromosome 6D and another locus on chromosome 6A (Figure 7.3). The two loci ranged from 105 to 140bp in size and had 10 alleles for 6D and five alleles for 6A. Another SSR primer pair that amplified sequences localized on chromosomes 2DS, 2AS and 2BS was Wmc 25. The number of alleles was 15 for 2DS, seven for 2BS and three for 2AS. The band sizes ranged from 140bp to 290bp. The 2DS alleles had two co-migrating bands (Figure 7.4).

71

6X

4X

abc defghi j i ki i i l

Figure 7.1: PCR amplification of microsatellite WMS 52 on chromosome arm 3DL. The same letters represent one allele.

72

6X

4X

a ca bc aa bbbbb ba c

Figure 7.2: PCR amplification of microsatellite Wmc 233 on chromosome arm 5DS. The same letters represent one allele.

73

6X

4X

a b c d e e f c g h i d e e j e k l m k k l mmmm n k l mo l l l m l o mmmm om o m

Figure 7.3: PCR amplification of microsatellite Wmc 256 on chromosomes 6D and 6A. The same letters represent one allele

74

6X

4X

2BS

2DS

6X

4X

aba cc c c c c c c c c c c bb c a a a a aa a a c c a a a a cabbbbbaa b

2AS

Figure 7.4: PCR amplification of microsatellite Wmc 25 on chromosome arms 2AS, 2BS and 2DS. The same letters represent one allele.

75

Four SSR primer pairs amplified SSR fragments on B genome chromosomes, WMS 375 (4BL chromosome), WMS 234 (5BL chromosome), WMS 297 (7BS chromosome) and Xgwm 644 (7BL chromosome). Primer pair WMS 375 (4BL), amplified 25 polymorphic bands. Thirteen alleles in two co-migrating bands were obtained, nine in the hexaploid wheat entries and four in the tetraploid (Figure 7.5). The sizes of the alleles ranged between 150-190bp. Twenty nine bands in total were produced by primer pair WMS 234 (5BL). Fifteen alleles for this locus were obtained by three comigrating bands. Nine alleles were obtained in the hexaploid wheat entries and six in the tetraploid entries (Figure 7.6). The size of these alleles ranged between 130 and 210bp. Primer pair WMS 297 (7BS) produced 15 polymorphic bands, representing 13 alleles for this locus, that were repeated about three times but repetitions were stutter bands. Seven alleles were found in the hexaploid entries and six in the tetraploid. Their sizes ranged from 160-180bp (Figure 7.7). The fourth primer pair in this group, Xgwm 644 (7BL), produced 16 bands in total but only 12 of them were polymorphic. Six alleles were counted from these polymorphic bands, three in each ploidy level. The size of the alleles ranged between 160-210bp. Three A genome SSR primer pairs were Xgwm 136 (1AS), WMS 95 (2AS), and WMS 218 (3AS). Primer pair Xgwm 136 produced 17 bands, 14 of them being polymorphic. Seven alleles were counted in the hexaploid wheat entries and only two in the tetraploids. The allele sizes ranged between 180-310bp, containing three co-migrating bands. Twenty five bands were obtained from primer pair WMS 95. Two co-migrating main bands were amplified along with 14 stutter bands. The locus consisted of 10 alleles, six in tetraploid and four in hexaploid wheat entries (Figure 7.8). The allele sizes were between 110 and 140bp. For the last primer pair WMS 218 twenty seven polymorphic bands were obtained. Thirteen alleles were counted from these bands while the rest were stutter bands. Nine alleles were obtained in the hexaploid wheat entries, while only four obtained in the tetraploid ones. The size of the alleles ranged between 140 and 185bp (Figure 7.9).

76

6X

4X

a a b c d e e f d g d d d f g f hh h i i h h h h h i i h h i i i i i i h h i j j k k h i h

Figure 7.5: PCR amplification of microsatellite WMS375 on chromosome arm 4BL. The same letters represent one allele.

77

6X

4X

a b c d e f f d e g d h g d i d ___j_____ k l l l ______j______ m j j j n j j o j j j

Figure 7.6: PCR amplification of microsatellite WMS 234 on chromosome arm 5BL. The same letters represent one allele.

78

6X

4X

ab cd ef ebe g ff f f f f h i hhkhh h j l h j hj i mi i i k k i i i i i kki

Figure 7.7: PCR amplification of microsatellite WMS 297 on chromosome arm 7BS. The same letters represent one allele.

6X

4X

79

6X

4X

a ab a a c dae a eee afa g aah ga i i i gbaa ggggg aag cg jc baa

Figure 7.8: PCR amplification of microsatellite WMS 95 on chromosome arm 2AS. The same letters represent one allele.

80

4X

6X

a b a b c c c b c a a ---------c------- d d d b b d d d d c e f g h f i j i k k k l l l k m i l

Figure 7.9: PCR amplification of microsatellite WMS 218 on chromosome arm 3AS. The same letters represent one allele.

81

B. 1. Polymorphism and polymorphic content: The total number of bands and of polymorphic bands was counted for each SSR primer pair (Table 7). The eleven SSR primer pairs produced in total 238 bands (21.6 bands on average for each primer pair) from which 227 were polymorphic (95.4%). The total number of bands per primer pair ranged from six for Wmc 233 (5DS) to 35 for WMS 52 (3DL), while polymorphic bands ranged from eight to 33 bands for the same primer pairs, respectively. Generally, SSR band sizes ranged from 105 bp, for primer pair Wmc 256 (6A, 6D), to 310 bp for Wmc 233 (5DS) and Xgwm 136 (1AS). Most of the SSR bands were about 150 bp (Table 7). The number of SSR alleles ranged from three, for primer pair Wmc 233 (5DS), to 25 for primer pair Wmc 25 (2AS, 2BS, 2DS). The polymorphic information content (PIC) that was calculated for each SSR primer pair based on allele numbers according to Anderson et al. (1993) ranged from 0.67 for Wmc 233 (5DS) to 0.93 for Wmc 25 (2AS, 2BS, 2DS) (Table 5).

82

83

B. 2. Cluster Analysis: Using cluster analysis, SSRs were able to separate bread wheat from durum wheat entries (groups A and B, Figure 8). Similarity coefficient for tetraploid entries ranged from 0.63 between ‘Kyperounda-M’ and ‘Sifnos’, to 0.97 between ‘Kyperounda-A’ and ‘Kyperounda’, while for hexaploid entries ranged from 0.59 between ‘Sids-1’ and ‘Sakha-69’ to 0.91 between ‘Gemmiza-7’ and ‘Gemmiza-9’ (Appendix 7). Durum tetraploid wheat entries (group A) were divided in two sub-groups (A1 and A2). Cypriot origin entries were separated distinctly from the Greek ones (sub-group A1) along with ‘Athos’, ‘Sanda’ and ‘Skiti’ (Greek entries) (Figure 8). The highest similarity coefficient was obtained between Cypriot entries ‘Kyperounda’ and ‘Kyperounda-A’ with a 0.97 similarity coefficient, which clustered with ‘Kyperounda-M’ (0.92 similarity coefficient) (Appendix 7). The homogeneity within this sub-group was high compared to the other sub-groups in the dendrogram (Figure 8). The other Greek durum wheat entries were clustered in sub-group A2 along with the Italian-originating ones (‘Simeto’ and ‘Pietrafitta’). ‘Mavraghani Hiraklio’ and ‘Mavraghani Hiraklio-A’ along with ‘Pondos’ and ‘Selas’ showed the highest homogeneity in this sub-group (0.94 similarity coefficient). The former are grouped with

‘Mavraghani Hiraklio-M’ and

‘Myrina’ (Figure 8, І), while the latter are grouped with ‘Papadakis’, ‘Simeto’ and ‘Skyros’ (Figure 8, IІ). ‘Kallithea’ was found to differ from all entries of sub-group A2. The other members of this sub-group were dispersed along with ‘Pietrafitta’ (Italian origin) (Figure 8). Hexaploid bread wheat entries were separated in two sub-groups, namely B1 and B2. The first sub-group (B1) was further divided into two minor groups, B1.1 and B1.2. Minor group B1.1 contained all the Egyptian entries (except ‘Sakha-69’) along with ‘Strymonas’ and ‘Acheloos’ (Greek entries). Minor group B1.2 contained Greek bread wheat entries along with ‘Yecora-E’ and ‘Sakha-69’ (Figure 8). The closest two entries in sub-group B1 were ‘Gemmiza-7’ and ‘Gemmiza-9 having a 0.91 similarity coefficient (Appendix 7). The second sub-group (B2) contained both ‘Verghina’ and ‘Generozo-E’.

84

ATHOS SANDA MAKEDONIA KARPASIA ARONAS MESAORIA

A1

SKITI KYPEROUNDA KYPEROUNDA-A KYPEROUNDA-M AGIAS ANNA MEXIKALI-81 SIFNOS ILEKTRA-A

A

4X

MAVRA_HERAK MAVRA_HERAK-A

I

MAVRA_HERAK-M MYRINA PAPADAKIS PONDOS

II

SELAS SIMETO SKYROS PIETRAFITTA ILEKTRA

A2

ILEKTRA-M LIMNOS KORNOS KALLITHEA YECORA-E ACHERON NESTOS ELISAVET

B1.2

SAKHA-69 SAKHA-61 STRYMONAS

B1

SAKHA-94 GEMMIZA-7

B1.1

B

GEMMIZA-9 GIZA-168 SIDS-1 ACHELOOS

B2

VERGHINA GENEROZO-E

0.62

0.71

0.80 Coefficient

0.89

0.98

Figure 8: Cluster analysis of wheat entries based on SSR alleles using the simple matching algorithm and the UPGMA clustering method (Cophenetic correlation coefficient r=0.84).

85

6X

The cophenetic correlation coefficient between the simple matching algorithm and UPGMA clustering method was high (r = 0.84) indicating the reliability of clustering obtained. B. 3. Principal Coordinate Analysis (PCOORDA): Using the principal coordinate analysis (PCOORDA), wheat entries were separated by the first principal coordinate according to their ploidy level into two groups (A-tetraploid and B-hexaploid, Figure 9). About one third of the total genetic variance (32.7%) was represented by the first three principal coordinates PC1, PC2 and PC3, accounting for 18.4%, 8.1% and 6.2%, respectively. The tetraploid wheat entries were separated into four sub-groups (Figure 9). The first sub-group (1) contained the three ‘Kyperounda’ entries and loaded maximum on PC2 (Figure 9). The second sub-group (2) contained the other Cypriot entries (loaded high on PC2 and maximum on PC3) along with Greek entries ‘Skiti’ and ‘Sanda’. The third sub-group (3) included four Greek entries (‘Athos’, ‘Kallithea’, ‘Pondos’ and ‘Selas’) which loaded high variation on the PC3, and the Italian originating variety ‘Pietrafitta’. The fourth sub-group (4) contained the remaining Greek entries along with ‘Simeto’ (Italian origin). ‘Limnos’ loaded low on PC3, while ‘Kornos’ had no load for that principal coordinate (Figure 9). From hexaploid wheat entries (B, Figure 9), ‘Sids-1’, ‘Acheloos’, ‘Verghina’ and ‘Generozo-E’ entries loaded low on PC2, from which ‘Generozo-E’ loaded high on PC3 (Figure 9). Variety ‘Sakha-69’ loaded minimum on PC1, explained high variation on both PC2 and PC3 and separated from all other entries. The other hexaploid wheat entries explained intermediate variation on both PC2 and PC3. The only exception of these entries was ‘Giza-168’ which was loaded very low based on PC3 (Figure 9).

86

MAKEDONIA

GENEROZO-E

KARPASIA ARONASMESAORIA SANDA SKITI

PC1 = 18.4% PC2 = 8.1% PC2 = 6.2

KYPEROUNDA-M KYPEROUNDA-A KYPEROUNDA

ATHOS

KALLITHEA

ANNA

SIFNOS PONDOS SELAS SKYROS MEXIKALI-81 AGIAS

ACHERON

GEMMIZA-9

ELISAVET

SIMETO PAPADAKIS

SAKHA-69 NESTOS

GEMMIZA-7

MYRINA PIETRAFITTA

VERGHINA

SAKHA-94

ILEKTRA-A ILEKTRA-M MAVRA_HERAK-M MAVRA_HERAK-A STRYMONAS

YECORA-E

MAVRA_HERAK

SAKHA-61

ILEKTRA

PC3 ACHELOOS

SIDS-1

1

GIZA-168

LIMNOS

2

PC2 3

B. Bread entries

KORNOS

4

A. Durum entries PC1

Figure 9: Three-Dimensional PCOORDA analysis for wheat entries obtained using SSRs.

87

C. Inter Simple Sequence Repeats (ISSRs) Analysis Four hundred and eight bands were obtained from nine ISSR primers, from which 324 bands were polymorphic (79.4%) (Table 7). The total number of bands per ISSR primer ranged from 15 (for primer UBC 814, Figure 10.2) to 67 (for primer UBC 848). Number of polymorphic bands per ISSR primer ranged from 7 (for primer UBC 814) to 63 (for primer UBC 836). Polymorphic information content (PIC) was high, ranging from 0.92 (for primer UBC 814) to 0.98 (for primers UBC 825, UBC 835, UBC 836 and UBC 848) (Table 7). A sample of polyacrylamide gel image is presented in Figure 10.1 (anchored primer UBC 815). C. 1. Cluster Analysis: Wheat entries were clustered according to their ploidy level when using the ISSR generated data (Figure 11). Similarity coefficient for tetraploid entries ranged from 0.68 (between ‘Kyperounda’ and ‘Anna’; ‘Skyros’ and ‘Kallithea’; and ‘Skyros’ and ‘Mavraghani Hiraklio-M’) to 0.92 (between ‘Pondos’ and ‘Papadakis’), while for hexaploid entries it ranged from 0.71 (between ‘Elisavet’ and ‘Sakha-94’) to 0.88 (between ‘Elisavet’ and ‘Nestos’) (Appendix 8). In the cluster containing the tetraploid wheat entries ‘Pietrafitta’, ‘Kallithea’ and ‘Elektra-A’ were found as the outer most. The other tetraploid wheat entries were clustered into two sub-groups, A1 and A2 (Figure 11). Sub-group A1 was divided into two branches, I and II. Variety ‘Athos’ was clustered to the A1.I and A1.II branches. Branch I was divided into two minor branches, the first included ‘Limnos’, ‘Mavraghani Hiraklio’ and ‘Myrina’, while the second contained ‘Kornos’, ‘Mavraghani Hiraklio-A’ and ‘Mavraghani Hiraklio-M’. The two most related entries in branch I were ‘Mavraghani Hiraklio’ and ‘Myrina’ having a 0.85 similarity coefficient, and ‘Mavraghani Hiraklio-A’ and ‘Mavraghani Hiraklio-M’ having a 0.83 similarity coefficient (Appendix 8). Branch II contained ‘Sanda’, ‘Simeto’ and ‘Skiti’ in addition to the three types of ‘Kyperounda’.

88

Sub-group A2 was divided into two branches, III and IV. Branch III contained ‘Agias’, ‘Anna’, ‘Elektra’ and ‘Elektra-M’ (the last two having a 0.87 similarity coefficient). Branch IV was also divided into two minor branches; the first contained all Cypriot wheat entries (except of ‘Kyperounda’ and its derivatives) along with ‘Skyros’. The most related entries in this branch were ‘Makedonia’ and ‘Karpasia’ (0.84 similarity coefficient) (Appendix 8). The second minor branch included ‘Mexikali-81’, and ‘Sifnos’ (having 0.86 similarity coefficient) separated from ‘Selas’, ‘Papadakis’ and ‘Pondos’ (the last two having a 0.89 similarity coefficient) (Figure 11 and Appendix 8). Hexaploid wheat entries were clustered into two sub-groups, B1 and B2. Egyptian hexaploid bread wheat varieties were grouped in sub-group B1 along with ‘Yecora-E’ (Figure 11). The homogeneity within Egyptian hexaploid wheat entries was high. The most related varieties in this sub-group were ‘Sakha-69’ and ‘Sakha-94’ with a 0.85 similarity coefficient, followed by ‘Gemmiza-7’ and ‘Gemmiza-9’ with a 0.81 similarity coefficient (Appendix 8). Hexaploid bread wheat entries from Greece (sub-group B2) were separated from the Egyptian ones except for ‘Sids-1’ which was grouped with the Greek entries. The Greek varieties were more heterogeneous compared to the Egyptian varieties. The closest two entries in B2 were ‘Nestos’ and ‘Elisavet’ having a 0.80 similarity coefficient. The cophenetic correlation coefficient between the genetic similarity matrix generated by the Jaccard algorithm and the UPGMA clustering method was very high (r = 0.95), indicating the reliability of clustering.

89

6X

4X

Figure 10.1: ISSR pattern of anchored primer UBC 815 on polyacrylamide gel using 45 wheat entries. The arrows show band presence either in tetraploid or in hexaploid wheat entries.

90

91

A1

ATHOS LIMNOS MAVRA_HERAK MYRINA KORNOS MAVRA_HERAK-A MAVRA_HERAK-M SANDA SIMETO SKITI KYPEROUNDA KYPEROUNDA-A KYPEROUNDA-M AGIAS ANNA ILEKTRA ILEKTRA-M MEXIKALI-81 SIFNOS PAPADAKIS PONDOS SELAS ARONAS SKYROS MESAORIA MAKEDONIA KARPASIA ILEKTRA-A KALLITHEA PIETRAFITTA YECORA-E GIZA-168 SAKHA-61 SAKHA-69 SAKHA-94 GEMMIZA-7 GEMMIZA-9 SIDS-1 STRYMONAS ACHELOOS ACHERON VERGHINA GENEROZO-E NESTOS ELISAVET

I

II

III A2

IV

A

B1

B

B2

0.46

0.57

0.68

0.79

0.90

Coefficient Figure 11: Cluster analysis for wheat entries based on ISSR markers using the Jaccard algorithm and the UPGMA clustering method (Cophenetic correlation coefficient r=0.95).

92

4X

6X

C. 2. Principal Coordinate Analysis (PCOORDA): According to the principal coordinate analysis (PCOORDA), wheat entries were grouped into two distinct groups, A and B (Figure 12). The first three principal coordinates PC1, PC2 and PC3 accounted for 31%, 7.8% and 5 % of the total genetic variance, respectively. Wheat entries were separated by the first principal coordinate (PC1) according to their ploidy level (group A-durum wheat- and group B-bread wheat). All hexaploid bread wheat entries loaded very low on the PC1 (the left side in Figure 12), while all the tetraploid wheat entries loaded very high on PC1 (the right side). This coordinate was the most important since it accounted for 31% of the total genetic variance. All tetraploid wheat entries loaded very high on PC1 except for ‘Pietrafitta’ variety which loaded high on the same PC and explained maximum variation on PC3 (Figure 12). Tetraploid wheat entries grouped into four sub-groups depending on the second principal coordinate (PC2). The first sub-group explained minimum variation on PC2 and contained six Greek entries ‘Anna’, ‘Agias’, ‘Skyros’, ‘Selas’, ‘Elektra’ and ‘Elektra-M’ (Figure 12). These entries showed intermediate (‘Selas’ and ‘Skyros’) to high variation (the rest of the entries) on PC3. Sub-group 2 explained low variation on PC2 and contained two Cypriot entries (‘Aronas’ and ‘Makedonia’) that showed intermediate variation on PC3 along with ‘Papadakis’ and ‘Pondos’. Another two Greek tetraploid wheat entries were included in this sub-group (‘Mexikali-81’ and ‘Sifnos’) that loaded high on PC3. The third sub-group (sub-group 3) showed intermediate variation on both PC2 and PC3, except for ‘Elektra-A’, which showed high variation on PC3, and ‘Skiti’, which showed low variation on PC3. This sub-group contained four Cypriot entries (‘Kyperounda’, ‘Kyperounda-A’, ‘Mesaoria’ and ‘Karpasia’), seven Greek tetraploid wheat entries (‘Mavraghani Hiraklio’, ‘Myrina’, ‘Skiti’, ‘Elektra-A’, ‘Sanda’, ‘Athos’ and ‘Limnos’) in addition to ‘Simeto’ (Italian origin). The last sub-group (sub-group 4) explained high variation on PC2, containing three Greek entries ‘Kallithea’, ‘Mavraghani Hiraklio-A’ and ‘Kornos’, and one Cypriot, ‘Kyperounda’. These entries showed intermediate variation on PC3 except of ‘Kallithea’, which loaded high on PC3

93

PIETRAFITTA

SIFNOS ILEKTRA-A

ELISAVET

PC1 = 31% PC2 = 7.8% PC3 = 5%

NESTOS GENEROZO-E

MAVRA_HERAK-M

MEXIKALI-81

KALLITHEA

ANNA

ILEKTRA

MAVRA_HERAK-A KORNOS

ACHELOOS

ATHOS LIMNOS

KYPEROUNDA SIMETO SANDA

ACHERON

KARPASIA MESAORIA MYRINA

VERGHINA

STRYMONAS

ILEKTRA-M AGIAS

PONDOS

KYPEROUNDA-M

MAKEDONIA PAPADAKIS

KYPEROUNDA-A MAVRA_HERAK

SELAS ARONAS

SKYROS

PC3

YECORA-E

SIDS-1 GIZA-168 SKITI

GEMMIZA-9 GEMMIZA-7

4 SAKHA-61

3

2 A. Durum

2

PC2

SAKHA-69

1

SAKHA-94

B. Bread

PC1

1

Figure 12: Three-Dimensinal PCOORDA analysis for wheat entries obtained by ISSR markers.

94

(Figure 12). ‘Mavraghani Hiraklio-M’ was separated from all tetraploid wheat entries with maximum variation on PC2. Hexaploid bread wheat entries formed two sub-groups based on PC2. The first sub-group contained three Egyptian entries (‘Sakha-69’, ‘Sakha-94’ and ‘Giza-168’) in addition to ‘Yecora-E’ variety (Figure 12). This sub-group explained low variation on PC2 and various on PC3. ‘Sakha-69’ and ‘Sakha-94’ had minimum variation on PC3, while ‘Yecora-E’ and ‘Giza-168’ showed high variation on PC3. The second sub-group had all the other Greek hexaploid wheat entries in addition to ‘Sakha-61’, ‘Gemmiza-7’ and ‘Gemmiza-9’ Egyptian entries. This sub-group showed intermediate to high variation on PC2 and showed different variation based on PC3, ranging from low (Gemmiza-7’ and ‘Gemmiza-9’) to very high variation (‘Elisavet’), with the other entries falling in between (Figure 12).

D. Mantel’s Test: Mantel's test was used to determine the correlation among morphological and molecular similarity matrices. Mantel’s test was carried out using NTSYS PC 2.0 software. A moderate positive correlation was revealed between the morphological and the molecular similarity matrices, while the correlation between the two types of molecular markers was high (Table 8).

Table 8: Correlation (r) between similarity matrices based on morphological and molecular markers. Morphological SSRs ISSRs

Morphological

SSRs

0.54 0.57

0.85

ISSRs

95

E. In situ hybridization: In situ hybridization, using the 5S and 18S-5.8S-26S rDNA probes was performed on a selected, according to the origin and ploidy level, number of entries. Twenty eight chromosomes were counted for the metaphase preparation of ‘Agias’, confirming its tetraploid level. Four 5S rDNA sites were found on its chromosomes and eight 18S-5.8S-26S rDNA sites (Figure 13D). Four sites from the latter were major (based on the intense red color signal), two sites were minor (less intense color) and another two were minute sites (Table 9). Using the combined photo for 5S rDNA and 18S-5.8S-26S rDNA sites (Figure 13B), two chromosomes were found to have both sites and thus are considered homologous. Another two chromosomes had minute sites of 18S-5.8S-26S rDNA. Both rDNA types on the homologous chromosomes are contiguous and are found on the short arms (Figure 13B). ‘Pondos’ had also twenty eight chromosomes (DAPI, Figure 14A). Four 5S rDNA sites were visualized (Figure 14C). Using the 18S-5.8S-26S rDNA probe, two major and two minor sites were revealed (Figure 14D). Another four minute sites of 18S-5.8S-26S rDNA were also observed (Table 9). Comparing the hybridization signals resulting from the two probes on the chromosomes, two homologous chromosomes having both rDNA sites were found contiguous on the short arms (Figure 14B). Another two homologous chromosomes had only one site of 5S rDNA on the short arm of each (Figure 14B). Twenty eight chromosomes were counted from DAPI filter for ‘Kallithea’ (Table 9). Four sites of 5S rDNA were obtained and four major sites for 18S-5.8S-26S rDNA were observed. An additional four minor and two minute 18S-5.8S-26S rDNA sites were also revealed (Table 9). Two homologous chromosomes had one site from each rDNA type on the short arms. The Italian variety ‘Pietrafitta’ has twenty eight chromosomes in its genome (Figure 15A). Four sites of 5S rDNA were counted on the chromosomes (Figure 15C), while four major and two minor sites of 18S-5.8S-26S rDNA were obtained (Figure 15D and Table 9). Another two minute 18S-5.8S-26S rDNA were also observed. Two pairs of

96

homologous chromosomes could be identified, the first pair having one site of both rDNA types on the short arm and the second had only one site of 5S rDNA (Figure 15B). Table 9: 5S and 18S-5.8S-26S rDNA sites for twelve wheat entries having different origin and different ploidy level. Entry

Origin

Ploidy level

Agias Pondos Kallithea Aronas Mesaoria Pietrafitta Nestos Acheron Yecora-E Giza-168 Sakha-94 Genorozo-E

Greece Greece Greece Cyprus Cyprus Italy Greece Greece Greece Egypt Egypt Italy

4X (28) 4X (28) 4X (28) 4X (28) 4X (28) 4X (28) 6X (42) 6X (42) 6X (42) 6X (42) 6X (42) 6X (42)

5S rDNA 18S-5.8S-26S rDNA Major Minor minute 2 2 4 4 4 2 2 4 2 4 2 4 2 2 4 4 4 2 4 4 2 2 4 4 4 4 4 4 2 4 4 6 2 6 4 8 4 2 4 6 4 2 4 6 4 2 4 6

The tetraploid Cypriot variety ‘Aronas’ has four 5S rDNA sites. Four major sites and two minor were counted for 18S-5.8S-26S rDNA on its chromosomes (Table 9). Another two minute sites were also obtained revealing homology between these two chromosomes (Table 9). Two more homologous chromosomes having both rDNA types and two having only 5S rDNA site were identified (Table 9). The sites on the homologous chromosomes are found on the short arms.

97

A

B

C

D

Figure 13: In situ hybridization of 5S (green color) and 18S-5.8S-26S (red color) sites for wheat variety ‘Agias’ (4X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

98

A

B

C

D

Figure 14: In situ hybridization of 5S (green color) and 18S-5.8S-26S (red color) sites for wheat variety ‘Pondos’ (4X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

99

A

B

C

D

Figure 15: In situ hybridization of 5S (green color) and 18S-5.8S-26S (red color) sites for wheat variety ‘Pietrafitta’ (4X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

100

A

B

C

D

Figure 16: In situ hybridization of 5S (green color) and 18S-5.8S-26S (red color) sites for wheat variety ‘Nestos’ (6X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

101

The last tetraploid wheat entry used for in situ hybridization was Cypriot variety ‘Mesaoria’. Four sites were identified having the 5S rDNA sites and another ten for 18S5.8S-26S rDNA (Table 9). From these, four major, two minor and four minute sites were identified (Table 9). The two homologous chromosomes which carry the two rDNA types on the short arms were observed. Forty-two chromosomes were counted for the Greek variety ‘Nestos’ (Figure 16A). Four 5S rDNA sites were visualized for this variety (Figure 16C and Table 9). For the 18S-5.8S-26S rDNA, twelve sites were observed. Four major, four minor and four minute sites were determined depending on the intensity of the signal (Figure 16D). Two homologous chromosomes had one site from both rDNA types (Figure 16B). The rDNA sites in the homologous chromosomes were on the short arms of NORs region. Forty two chromosomes were counted from the metaphase preparation stained with DAPI for the entry ‘Acheron’, which is a hexaploid Greek wheat variety (Figure 17A). Six sites of 5S rDNA were counted for this entry (Figure 17D). Four major sites with strong signal were observed for 18S-5.8S-26S rDNA (Figure 17C). Another four minor and two minute sites were also visualized using the FITC filter (Table 9). Two pairs of homologous chromosomes, each having the two rDNA sites, were visualized from combining the three photos (DAPI, FITC and Rhode). The rDNA types on the homologous chromosomes are in close proximity and localized on the short arms (Figure 17B). ‘Yecora-E’ is a Greek selection from ‘Yecora’ variety and has 42 chromosomes as it appears from the DAPI filter (Figure 18A). The highest number of 5S rDNA sites was counted for this entry, since eight sites were observed (Figure 18D). Out of the twelve 18S-5.8S-26S rDNA sites (Table 9), four of them were major sites, six were minor, and two minute sites (Figure 18C). Three pairs of homologous chromosomes, each having the different two types of rDNA, were identified after combining the three pictures using the Adobe Photoshop CS2 software (Figure 18B). The 18S-5.8S-26S rDNA sites were found at the NORs region.

102

The two hexaploid Egyptian varieties which were used for in situ hybridization, ‘Giza-168’ and ‘Sakha-94’, had the same number of sites for 5S rDNA (six sites, Figure 19D and Table 9) and for 18S-5.8S-26S rDNA (four major, two minor and four minute sites, Figure 19C and Table 9). Only two homologous chromosomes having the two types of rDNA were observed (Figure 19B). The Italian originating entry ‘Genorozo-E’ with 42 chromosomes (Figure 20A), had six sites for the 5S rDNA (Figure 20D) and a total of 10 sites for the 18S-5.8S-26S rDNA (Figure 20C). Four from the latter were major sites, two were minor and four sites were minute, similarly to the Egyptian hexaploid wheat entries (Table 9). Two homologous chromosomes carrying sites for the two types of rDNA were also identified on the short arms (Figure 20B).

103

A

B

C

D

Figure 17: In situ hybridization of 5S (red color) and 18S-5.8S-26S (green color) sites for wheat variety ‘Acheron’ (6X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

104

A

B

C

D

Figure 18: In situ hybridization of 5S (red color) and 18S-5.8S-26S (green color) sites for wheat variety ‘Yecora-E’ (6X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

105

A

B

C

D

Figure 19: In situ hybridization of 5S (green color) and 18S-5.8S-26S (red color) sites for wheat variety ‘Giza-168’ (6X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

106

A

C

B

D

Figure 20: In situ hybridization of 5S (red color) and 18S-5.8S-26S (green color) sites for wheat variety ‘Genorozo-E’ (6X) (A) DAPI filter, (B) combined photo through Adobe Photoshop software, (C) FITC filter and (D) Rhodamine filter.

107

V. DISCUSSION

108

V. DISCUSSION Increase and maintenance of genetic diversity is vital for any plant breeding program. In general, there is wealth of genetic diversity, but human manipulation led to drifting and biasing of genetic diversity in elite genetic pools (Helgason et al., 2003). The ability to recognize and manipulate the genetic diversity, however, needs further improvement (Able et al., 2007). Increasing the plant genetic diversity can be accomplished through introgression of novel alleles, found in wild species, landraces etc in plant breeding programs (Reif et al., 2005). Because of its large size and complex structure, wheat genome is considered one of the most complex in crop species for genetic analysis (Langridge et al., 2001). Genetic diversity of wheat can be monitored by different forms of analyses using morphological, biochemical, molecular or even cytogenetical markers. Many studies had been carried out to detect the genetic diversity in wheat using the above mentioned markers, but each one of them has advantages and disadvantages. Each marker type scans different areas of the genome, thus using more than one marker should improve the estimation of genetic diversity. At the same time, correlation among the estimates of genetic similarities by using different types of markers should be accomplished in order to strengthen and support the results. In the present study, 39 morphological characteristics, 11 SSR primer pairs, nine anchored ISSR primers and in situ hybridization technique were used in order to study the genetic diversity and relationships among 45 wheat entries with different geographical origins (Greece, Egypt, Cyprus and Italy). Correlation among morphological and molecular markers was also accomplished in order to reveal the relationship among the two different types of markers. The morphological characteristics were recorded during two growing seasons (2004/2005 and 2005/2006) from two replications, with the entries randomly distributed. Morphological data were standardized in order to decrease the season effect to the minimum level and were then used for cluster and principal coordinate analyses. Two-way hierarchical cluster and discriminant analyses were also carried out in order to compare the obtained results with those obtained from cluster

109

analysis using UPGMA method and also to determine the diversity and the correlation among the morphological characteristics. Out of 16 SSR primers specific for wheat genome, 11 were selected (the informative primers) for the analysis and nine anchored primers were used for ISSR analysis. For both types of molecular markers, cluster analysis and PCOORDA were performed. ISSR markers (dominant markers) have high information content in the pair wise present/present (1/1) or present/absent (1/0) comparisons in relation to the pair wise absent/absent (0/0) comparisons and thus they have more chance of representing homologous states. For that reason, Jaccard’s algorithm (Jaccard 1908), a proximity algorithm that place no weight on 0/0 matches, was used to analyze them (Spooner et al. 2005). On the other hand, the simple matching coefficient was used to analyze SSRs data (co-dominant markers) because this algorithm provides equal weight to all pair-wise combinations including the absent/absent (0/0) matching. The cophenetic correlation coefficient between each similarity matrix and its corresponding cluster tree was calculated. The correlation among the different matrices was also calculated using Mantel’s test. The number of 5S and 18S-5.8S-26S rDNA sites was investigated in 12 entries representing different ploidy levels and different origins, using the in situ hybridization technique. The probes were biotin-16-dUTP or digoxigenin-11-dUTP labeled and were detected using streptavidin Cy3 conjugate or antidigoxigenin-fluorescein, respectively. The chromosome numbers were counted using DAPI.

1. Phenotypic plasticity: Phenotypic plasticity may play an important role in plant adaptation and evolution by combining a physiological buffering to poor environmental conditions with an improved response to favorable conditions (Jackson et al., 1990). The understanding of how phenotypic diversity is generated by the coherent change of other integrated traits is a key challenge in evolutionary biology. In this study phenotypic plasticity was studied using 39 morphological qualitative and quantitative traits measured for 33 wheat varieties

110

and landraces. In general, the varieties showed higher stability percentage for the traits compared to landraces. On the other hand, higher percentage of phenotypic plasticity was obtained from the landraces (Appendix 4). This indicates that wheat varieties are genetically more stable compared to landraces, which may contain alleles with diverse phenotypic expressions in different environments. High stability was calculated for the qualitative traits compared to the quantitative traits (Table 5). Fourteen qualitative morphological traits showed scoring stability between the replications for both seasons, while only two quantitative morphological traits showed stability in more than 50% of the wheat entries. This supports the fact that one or a few genes are controlling the heritability of qualitative traits, while a set of genes the quantitative traits, which are affected by the changes of the environmental conditions much more than qualitative traits. This effect can be seen from the percentage of the phenotypic plasticity obtained for both traits. The phenotypic plasticity reached maximum (100%) for the heading date and above 90% for 1000 grains weight, plant height, head weight and test weight. On the other hand, phenotypic plasticity did not exceed 40% for the qualitative traits and for some of these traits it was 0% (angle of flag leaf to culmn, awn color at maturity, lodging and ear color at maturity). The accurate scoring of the quantitative traits is better than the qualitative traits, since, low percentages of score instability (0%-42.4%) were calculated for the former comparing to the latter (3%-63.6%). This could be due to the optical scoring which can be affected by many factors including time of scoring, the scorer, the accurate determination of the order of the trait, etc. Quantitative traits seem to be affected by these factors less than the qualitative traits, because the actual score recorded in this case depending on stable metric measurements. Genetic variation and genotype X environment interactions (GEI) for four traits (age at maturity, fertility, egg size and growth rate) were studied in the nematode Caenorhabditis elegans and their phenotypic plasticity was measured at 12 and 24oC (Gutteling et al., 2007). The five QTL associated with age at maturity, fertility and growth rate showed QTL x environment interaction. These colocalized with plasticity QTL for the respective traits suggesting allelic sensitivity to temperature. Similar results

111

were reported for Arabidopsis by Ungerer et al. (2003), who reported the identification of QTL showing allelic sensitivity under different ecologically relevant photoperiod environments for two recombinant inbred mapping populations (Ler X Col and Cvi X Ler) using a combination of quantitative genetics and QTL mapping. Hausmann et al. (2005), also in Arabidopsis, detected genetic variation and significant genotype-byenvironment interactions for many traits related to water use, suggesting that the phenotypic plasticity is genetically based. Using a two-dimensional genome scan, they detected epistatic (QTL-QTL) interactions for the phenotypic plasticity regardless of additive genetic effects.

2. Cluster analysis: Wheat entries having different ploidy levels were grouped separately when using cluster analysis, regardless of the type of marker (i.e. morphological traits, SSR markers and ISSR markers). However, entries were dispersed regardless of their geographical origin in all three analyses. The only exception were the tetraploid wheat entries from Cyprus, that grouped together in SSR cluster analysis along with Greek origin entries ‘Athos’, ‘Sanda’ and ‘Skiti’ (sub-group A1 in Figure 8). Although Dreisigacker et al. (2004) reported that SSRs represent a powerful tool to quantify genetic diversity in wheat, SSRs could not differentiate between the CIMMYT wheat lines from different megaenvironments (MEs). The MEs represented different environmental conditions (i.e., full irrigation, reduced irrigation, drought and heat stress, etc.) and generally combining various production areas. On contrast, Lima et al. (2003) found that wheat entries clustered according to their geographical origin when using SSRs. Our SSR results are in between the previous results; while the Cypriot entries were found to be clustered together the rest were not. The Egyptian and the Greek hexaploid wheat entries were not separated from each other because some of them have common ancestors. ‘Yecora’ is a CIMMYT variety that was used as a parent for ‘Sakha-61’and ‘Sakha-69’, and at the same time ‘Yecora-E’ is a Greek selection from ‘Yecora’. These varieties did not group

112

close, although ‘Yecora’ is their common ancestor. These results are in agreement with those reported by Dreisigacker et al. (2004), since they reported low correlation (r=0.43) between pedigrees and SSR-based distance estimates among CIMMYT wheat lines targeted to different megaenvironments. They reported that selection and drift are presumably important factors in reducing the correlation between genetic similarity (GS) and coefficient of parentage (COP) due to a shift in genomic contribution of parental lines, particularly during the early selfing generations. On contrast, Almanza-Pinzon et al. (2003) and Parker et al. (2002) found general agreement of diversity estimates measured based on both molecular markers (including AFLP, RFLP and SSR) and COPs. They reported, however, that accurate pedigree records are needed to determine genetic relatedness using COP. The three black and white awned entries (i.e. ‘Elektra’, ‘Mavraghani Hiraklio’ and ‘Kyperounda’) grouped together in one cluster according to morphological traits analysis (Figure 4) while they didn’t group together according to both SSR and ISSR cluster analyses (Figures 8 and 11). The two different awn-colored types of ‘Kyperounda’ (i.e. ‘Kyperounda-A’ and ‘Kyperounda-M’) were closely associated (0.56, 0.92 and 0.90 similarity coefficients for morphological, SSR and ISSR analyses, respectively). According to SSR and ISSR analyses, ‘Kyperounda’ was clustered with ‘Kyperounda-A’, closer than ‘Kyperounda-M’, which is expected since ‘Kyperounda’ has white awns. The black awns for ‘Kyperounda-M’ could be explained by a mutation that could affect the phenotype. This mutation creates de novo genetic variation which might be shown by both classical and molecular genetic analyses and may be caused by gene amplification and transposable elements (Rasmusson and Phillips, 1997). Ford and Gottlieb (1992) reported a natural occurring homoeotic mutant in Clarkia concinna, in which petals are replaced with sepal-like organs, suggesting that mutations of large effect may occasionally arise and be maintained within generations and thus contribute in the evolution of the germplasm. The other two black and white awned entries (i.e. ‘Elektra’ and ‘Mavraghani Hiraklio’) didn’t group as closely as the ‘Kyperounda’ types. Although the two types of ‘Elektra’ and ‘Mavraghani Hiraklio’ (A and M) were grouped in the

113

same sub-group (A1 in Figure 4) based on the morphological results, they are still far away from each other. Depending on the molecular analysis results, ‘Elektra’ was grouped close to ‘Elektra-M’ suggesting that ‘Elektra’ may has black awns and its white awns are produced from seed contamination. Although ‘Mavraghani Hiraklio’ means ‘black-awned’, the results suggest that ‘Mavraghani Hiraklio’ landrace has white awns (SSR cluster, Figure 8). In addition to the previous explanation for the other type of awn color, epistasis and activation of transposable elements could be added as another source of variation that affects the genetic background. Liao et al. (2001), in their study of effects of genetic background and environment on QTLs and epistasis for rice (Oryza sativa L.) panicle number, reported that epistasis is more effective on genetic background and environment than main-effect quantitative trait loci (QTLs).

3. Two-way hierarchical morphological Cluster Analysis: According to Mohammadi and Prasanna (2003), distance-based clustering methods can be categorized into two groups: hierarchical and nonhierarchical. Hierarchical clustering methods (known as “agglomerative hierarchical” methods) are more commonly employed in analysis of genetic diversity for crop species. These methods proceed either by a series of successive mergers or by a series of successive divisions for a group of individuals, starting from a single entry. Thus, there are initially as many clusters as individuals. The most similar individuals are first grouped and these initial groups are merged according to their similarities. Among various agglomerative hierarchical methods, the UPGMA (Unweighted Paired Group Method using Arithmetic averages) (Sneath and Sokal, 1973; Panchen, 1992) is the most commonly adopted clustering algorithm, followed by the Ward’s minimum variance method (Ward, 1963). Two-way hierarchical cluster analysis for the morphological characteristics, using Ward’s method, was carried out with JMP IN 5 software. The results were comparable to those obtained from cluster analysis carried out with NTSYS PC using the UPGMA method (Figure 3). Ward’s method gave better separation for the wheat entries from

114

Cyprus and Egypt, than the UPGMA method. The Greek wheat entries were dispersed across all the dendrogram. It can be concluded that Ward’s method gave better resolution of the genetic diversity for the wheat entries used than the UPGMA method. Mohammadi and Prasanna (2003) obtained similar results in rapeseed (Brassica spp.). Although they found very high correlation between distance matrices obtained through use of different coefficients, they reported that Ward’s method was more suitable as it avoided the chaining effects that are often observed when using the UPGMA method. Dubreuil et al. (1996) reported also similar observations in analysis of genetic diversity among maize inbred lines based on RFLP data. The second way of the two-way hierarchical cluster analysis was carried out among the morphological characteristics (Figure 3). The morphological traits were grouped in three main groups containing both qualitative and quantitative traits. The morphological data were also used to determine the correlation values among the morphological traits through multivariate analysis in JMP program. Very high correlation values were found among some traits ranging from 0.73 (between blade of flag leaf glaucosity and ear glaucosity traits) to 0.95 (between angle of flag leaf to culmn and angle of leaves to culmn traits) (Table 6). Some morphological traits are highly correlated, which means that could be handled as a group. For example the angle of flag leaf to culmn can be grouped with the angle of leaves to culmn and used as a single trait for genetic diversity determination.

4. Principal Coordinate Analysis (PCOORDA): Both morphological and molecular data were standardized, similarities matrices were calculated using SimInt option, decentered, and then the eigenvectors and the eigenvalues were calculated using the ordination option in NTSYS PC software. The three-dimensional figures were obtained for the three different types of data. Hexaploid wheat entries were distinctly separated from the tetraploid wheats based on the first principal coordinate (PC1) using all three different markers (Figures 5, 9 and 12).

115

Comparing to the other markers, ISSRs gave the best separation of the hexaploid entries from the tetraploid; whereas, the hexaploid wheat entries were loaded very low on the PC1, the tetraploid entries were loaded very high on the same principal coordinate (Figure 12). These results could suggest that the ISSR markers are better than both morphological traits and SSR markers in differentiating the hexaploid wheat entries from the tetraploid ones. Depending on the second principal coordinate (PC2), there was no obvious trend in separating the wheat entries according to their geographic origin, except for the Cypriot tetraploid wheat entries that were loaded high for both PC1 and PC2 (Figure 5). The wheat genetic material adapted in the approximately similar environmental conditions in the Mediterranean countries might be the reason for this complexity in the separation of the wheat entries from these countries. Similar results were observed by Stodart et al. (2005). They used AFLP and SSR markers to discriminate among bread wheat accessions collected from North Africa, South Europe, the Middle East and southern and eastern Asia. Using PCOORDA, the accessions from North Africa and South Europe separated from those obtained from the other regions. For the present results, based on the third principal coordinate (PC3), higher level of diversity was obtained within the tetraploid durum wheat entries group than the hexaploid bread entries group for all types of analyses. The level of diversity for PC3 was higher in the morphological traits (eign. = 10.6%) than the molecular markers (SSR eign. = 6.2%, ISSR eign. = 5%). These results suggest that although the wheat entries from the Mediterranean area have broad genetic variation based on the morphological results (46.6% of the total genetic variation for the first three PCs.), they still have limited genetic base according to the molecular markers results (43.8% and 32.7% of total genetic variation for both ISSR and SSR markers, respectively). Cluster analysis refers to “a group of multivariate techniques whose primary purpose is to group individuals or objects based on the characteristics they possess, so that individuals with similar descriptions are mathematically gathered into the same cluster” (Hair et al., 1995). The resulting clusters of individuals should then exhibit internal homogeneity (within cluster) and external heterogeneity (between clusters). On

116

the other hand, PCOORDA is a scaling or ordination method that starts with a matrix of similarities or dissimilarities, between a set of individuals, and aims to produce a lowdimensional graphical plot of the data in such a way that distances between points in the plot are close to original dissimilarities (Mohammadi and Prasanna 2003). PCOORDA is usually used to reduce big amount of data in groups and to obtain a new set of uncorrelated variables which is known as principal coordinates (PCs). The first PC summarizes most of the variability present in the original data relative to all remaining PCs. The second PC explains most of the variability not explained by the first PC and uncorrelated to the first, and so on. Mohammadi and Prasanna (2003) reported that principal component analysis (PCA) and PCOORDA (the ordination methods) can be used in combination with cluster analysis for genetic diversity determination purposes, particularly when the first two or three PCs explain more than 25% of the genetic variation. In the present study, the first three PCs explained about a third (32.7%) to about a half (46.6%) of the total genetic variance for SSRs and morphological analyses, respectively, while the first three PCs for ISSR analysis explained 43.8% of the total genetic variance. That means that PCOORDA can be used either alone or in combination with cluster analysis to discuss the genetic diversity of the wheat entries. The results of the PCOORDA also showed better resolution for the genetic diversity than cluster analysis. Melchinger (1993) reported that the ordination methods provided faithful portrayal of the relationships between major groups of maize and barley lines comparing to the cluster analysis.

5. Discriminant analysis: Discriminant analysis was carried out for the morphological data in order to discriminate among the two types of wheat entries (hard or tetraploid-durum and soft or hexaploid-bread). Morphological traits that characterize each wheat type (tetraploid or hexaploid) were also observed. The first discriminant canonical accounted for all the

117

genetic variation present in all wheat entries. According to this canonical, wheat entries were separated into two distinct groups, hard and soft (Figure 6). Nine qualitative morphological traits were found to characterize the tetraploid hard wheat entries, while some of these traits clearly discriminate the tetraploids. However some other traits, such as angle of leaves to culmn, straw pith in cross section, lower glume beak shape and ear glaucosity traits did not clearly discriminate between the two types of wheats. The tetraploid wheat entries were characterized by the intermediate growth habit (5 in scoring system of growth habit trait in Table 3), while the hexaploid entries characterized by the erect growth habit. Rigidity of leaves was another trait discriminating the tetraploid (slightly bent) and the hexaploid (bent) entries. Furthermore, the tetraploid wheat entries had a sigma shape culmn (peduncle shape), while it was almost absent in the hexaploid entries. The tetraploid wheat entries were also characterized by the compact spikes where it was almost loose in bread wheat entries. The grain shape is another important morphological trait that characterizes the durum wheat. Hexaploid wheat entries, on the other hand, were characterized by four quantitative morphological traits. Two of them didn’t clearly discriminate the hexaploid entries (lower glume shoulder width and ear color at maturity), while the other two (apical rachis hairiness and glaucosity of blade of flag leaf) characterized the bread wheat by the presence of the hairiness on the uppermost (apical) rachis and the strong glaucosity of the flag leaf blade. Durum wheat entries didn’t have hairs on the apical rachis and the glaucosity of flag leaf (the blade) was absent or medium. For quantitative traits, the length of the second leaf from top in hexaploid bread wheat entries was on average 33 cm compared to the tetraploid durum wheat entries (30 cm). The hexaploid entries had also more seeds per head (61) compared to the tetraploid entries (55). The plant height result suggests that hexaploid wheat entries are relatively shorter than tetraploid plants; the average plant height for both years of bread wheat entries was about 115 cm where for durum wheat was 130 cm. These results are in disagreement with those obtained by Eticha et al. (2006), since a considerable proportion (49.3%) of the hexaploid wheat was classified into the tetraploid wheat. The failure to

118

distinguish the tetraploids from the hexaploids was attributed to the characters considered for the analysis. They used only seven morphological qualitative traits, and thus it was very difficult to distinguish the tetraploid from the hexaploid wheat entries. The morphological traits they used were glume hairiness, awn color, awn length, beak awn, glume color, seed color and spike density. The morphological traits they used to discriminate between the tetraploid and the hexaploid wheat entries, except of ear density, were not found, by the present study, to be discriminative. Some morphological traits that discriminate between the tetraploid and the hexaploid wheat entries in our study were stable during the two growing seasons, such as angle of leaves to culmn (87.9% stability) and ear density (78.8%). Some others were found less stable, such as plant height (0% stability and 100% plasticity) and rigidity of leaves (18.2% stability), but they can still be used based on annual data records. The present results suggested that some traits can be used to identify and classify the wheat varieties (bread or durum), but further studies are needed to ensure that these morphological traits are not influenced by the environment.

6. SSR and ISSR patterns: The number of SSR alleles obtained in hexaploid wheat entries was higher than in tetraploid wheat entries. On contrast, Zhang et al. (2006) reported that durum wheat landraces from Oman had a higher genetic diversity (depending on SSR markers) than did the bread wheat landraces. In this respect, other studies are still needed to compare the number of alleles among tetraploid and hexaploid wheat genotypes using SSR markers well dispersed over all the wheat genomes. Many stutter bands were obtained above and below the expected SSR allele bands. Production of such stutter bands in wheat SSR reactions is well documented (Manifesto et al., 2001; Lima et al., 2003; Naghavi et al., 2004). Manifesto et al. (2001) reported that these stutter fainter bands produced by error “slippage” of the polymerase during the PCR amplification as non-specific products, while Lima et al. (2003) reported that each of these bands was able to generate all other bands when it was used as a template for re-amplification under the same conditions.

119

Lima et al. (2003) mentioned that these bands did not prevent comparisons between separate cultivars. Although Lima et al. (2003) concluded that the reason for the presence of these bands remain unexplained, the results of the co-migrating bands suggest that they belong to the same chromosome because they were amplified with a stable space between them in all entries (Bryan et al., 1997). Even though the amplification of these comigrating bands from two or three SSR sites on the same chromosome could be suggested, the presence of these bands with the specific SSR bands is still accompanied with a question mark. The largest number of alleles per locus was obtained in B genome since a total of 53 alleles were derived from five SSR primer pairs specific for B genome, giving an average of 10.6 alleles per locus. A total of 43 alleles were obtained from five SSR primer pairs specific for A genome (with average of 8.6 alleles per locus), while 36 total alleles were obtained from four SSR primer pairs specific for D genome (with average of 9 alleles per locus). These results suggest that B genome is more polymorphic than either A or D genomes. The B genome chromosomes appear to be more polymorphic than the other genomes in various classes of repetitive DNA (Yifru et al., 2006). These results are in agreement with a number of studies (Huang et al., 2002; Zhang et al., 2002; Khlestkina et al., 2004; Medini et al., 2005; Yifru et al., 2006). On contrast, Roussel et al. (2004) obtained the highest number of alleles in the A genome (15.2 alleles per locus), followed by the D genome (14.4 alleles per locus) and the B genome (13.8 alleles). The allele size in the present results ranged from 105 bp, for primer pair Wmc 256 (6A, 6D), to 310 bp, for primer pairs Wmc 233 (5DS) and Xgwm 136 (1AS), indicating that both A and D genomes have the smallest and the biggest SSR repeats in size, while B genome has intermediate SSRs sizes for the primer pair set that was used. Microsatellite (SSR) marker is a co-dominant single-locus marker that is very informative and detects the variation among genotypes, depending on the number of the di- or tri-nucleotide repetition for a specific locus. ISSR, on the other hand, is a dominant marker detecting the variation among genotypes, depending on the presence or absence of a band. A total of 408 bands were obtained from the analysis of nine ISSR primers with

120

45.33 bands per primer, from which 324 (79.4%) bands were polymorphic (Table 7). For SSR analysis, a total of 238 bands were obtained with 21.6 bands per primer, from which 227 (95.4%) bands were polymorphic (Table 7). Although ISSR markers produced higher total number of bands than SSR markers, they produced less polymorphic bands (16%) than those produced by SSR markers (Table 7). The same results were obtained by Khan et al. (2005) in determining molecular markers linked to stem and leaf rust resistance in wheat. They found that SSR markers were more polymorphic and more informative than either RAPD or ISSR markers. Polymorphic information content (PIC) was high in general for both markers, ranging from 0.67 for SSR primer pair Wmc 233 to 0.98 for ISSR primers UBC 825, UBC 835, UBC 836 and UBC 848 (Table 7). PIC for ISSR markers was higher than for SSR markers. This is because PIC is calculated for ISSR markers using the total number of polymorphic bands, while for SSR markers depends on the number of alleles. Godwin et al. (1997) reported that ISSR analysis detected higher levels of polymorphism than RFLP or RAPD. They attributed this rather on technical reasons related to the detection methodology used for ISSR analysis. Nagaoka and Ogihara (1997), on contrast, concluded that ISSRs are highly polymorphic and especially informative for estimating genetic relationships in wheat compared to RAPD and RFLP markers.

7. Correlation among similarity matrices: Intermediate correlation values were obtained among morphological and molecular similarity matrices (r=0.54 between morphological traits and SSR, while r=0.57 between morphological and ISSR markers). A poor correlation values were obtained from a number of genetic diversity determination studies based on morphological characteristics in combination with SSR markers [(r = -0.09) Roy et al., 2004; (r=0.25) Hamza et al., 2004], RAPD [Maric et al., 2004], and AFLP [Lage et al., 2003]. It seems that in the present study a higher correlation was obtained between morphological characteristics and molecular markers. This can be related to the higher

121

number of morphological traits used in the determination of genetic diversity. Moderate correlation was found between pedigree-based genetic diversity estimates and SSR (r=0.43, Dreisigacker et al., 2004), or AFLP (r=0.42-0.46; Soleimani et al., 2002 and Barrett et al., 1998) in wheat. The correlation among the two types of molecular markers, SSR and ISSR, was very high according to Mantel’s test (0.85). These results are in agreement with those obtained by Nagaoka and Ogihara (1997), since they obtained identical dendrograms of wheat accessions created by ISSR, RFLP and RAPD marker data. Moderate correlation values ranging from 0.46 to 0.57 were also reported between SSR and AFLP markers in wheat genetic diversity determination studies (AlmanzaPinzon et al., 2003; Medini et al., 2005). On the other hand, a weak correlation was obtained among molecular markers such as RAPD and ISSR in barley (r=0.11), (Hou et al., 2005), SSR and RAPD in wheat (r=0.19), (Naghavi et al., 2004), SSR and AFLP in wheat (r=0.27), (Manifesto et al., 2001). The weak correlation between morphological traits and molecular markers (compared to the correlation between two types of molecular markers) could be attributed to different reasons according to Spooner et al. (2005). These could be considered as advantages for molecular markers and disadvantages for morphological at the same time and include: 1) the independence from environmental and pleiotropic effects for molecular markers; while morphological traits can vary under different environments (phenotypic plasticity) and subject to epistatic control; 2) the unlimited number of independent markers are available for molecular markers, unlike some morphological traits that are actually specific sites fixed through consequently selection during plant breeding program; 3) Molecular marker data can be more easily scored as discrete states of alleles or DNA base pairs, while some morphological traits and field evaluation data must be scored as continuously variable characters that are less amenable to robust analytical methods; and finally 4) Many molecular markers are selectively neutral. These advantages do not imply that other more traditional data used to characterize biodiversity are not valuable. On the contrary, morphological data will continue to provide practical and often critical information needed to characterize genetic resources. For the above

122

mentioned reasons, it can be concluded that molecular markers are more reliable in determining the genetic diversity in wheat than morphological traits. Furthermore, the strong correlation between SSR and ISSR markers in our results indicates their reliability in determination of genetic diversity in wheat.

8. In situ hybridization: The two types of ribosomal RNA genes (rDNA), 5S and 18S-5.8S-26S, are arranged in eukaryotes in tandem arrays on separated sites of the same chromosome or of different chromosomes, whereas, they are in juxtaposition in lower eukaryotes such as yeast and moss (Appels and Honeycutt, 1986). The 18S-5.8S-26S is associated with the nucleolus organizing origin (NOR) (Mukai et al., 1990; Flavell et al., 1986). The 5S rDNA sites are found on the short arms of the homoeologous wheat chromosomes 1 and 5 (Mukai et al., 1990), while the 18S-5.8S-26S rDNA sites are found on the short arms of chromosomes 1A, 1B, 6B, 5D and the long arm of chromosome 7D (Mukai et al., 1991). Different number of 5S rDNA sites was obtained for hexaploid wheat entries compared to tetraploids. All tetraploid wheat entries used for in situ hybridization have four sites of 5S rDNA. These sites according to Mukai et al. (1990) are found on chromosomes 1A, 1B, 5A and 5B. Hexaploid wheat entries, on the other hand, have six 5S rDNA sites, except of ‘Yecora-E’ which had eight sites and ‘Nestos’ which had only four sites, similar to the tetraploid entries. The expected number is six sites of 5S rDNA on the chromosomes of hexaploid wheat entries, while the natural amphiplasty may have occurred in ‘Nestos’, in which the rDNA sites were deleted during the polyploidy formation according to Yamamoto (1994). ‘Yecora-E’ which had eight sites of 5S rDNA is a Greek selection from ‘Yecora’ variety that may contain chromosomes that carry 5S rDNA sites from different ancestors, but the explanation of this increase number of 5S rDNA sites remains unknown. A total of eight sites of 18S-5.8S-26D rDNA were identified in tetraploid wheat entries, while 10 sites were visualized on the chromosomes of the hexaploid entries.

123

However, 10 sites were counted for the Cypriot tetraploid entry ‘Mesaoria’ and 12 for the Greek hexaploides ‘Yecora-E’ and ‘Nestos’. Four major sites were observed to all wheat entries used for in situ hybridization except for ‘Pondos’ and ‘Kallithea’ that had two major sites. The four major sites of 18S-5.8S-26D rDNA are assigned to the short arms of the chromosomes 1A, 1B, 6A and 5D according to Mukai et al. (1991). The number of the minor sites of 18S-5.8S-26D rDNA was two in tetraploid wheat entries (except in ‘Kallithea’ which had four). Two minor sites were also observed in the hexaploid Egyptian varieties along with ‘Genorozo-E’, while the Greek entries had four sites and ‘Yecora-E’ had six. The minute sites of 18S-5.8S-26D rDNA did not differentiate the wheat entries neither according to their ploidy level nor according to their origin. In general, the number of 18S-5.8S-26D rDNA minute sites ranged between two to four sites. It can be concluded that Egyptian hexaploid varieties ‘Giza-168’ and ‘Sakha-94’ were distinguished from the Greek hexaploid ones depending on the number of 18S-5.8S26S rDNA sites along with the Italian originating variety ‘Genorozo-E’. One pair of homologous chromosomes having both rDNA types was observed in all wheat entries except of ‘Acheron’, for which two homologous pairs were obtained, and ‘Yecora-E’, for which three pairs of homologous chromosomes were counted. Three pairs of homologous chromosomes carrying both rDNA sites (5S and 18S-5.8S-26S) were also obtained by Mukai et al. (1990) in hexaploid wheat and were assigned to the pairs of homologous chromosomes of 1A, 1B and 5D. Changes in copy number of the rDNA sites or complete loss during the polyploidy formation could be suggested in the other cases, in which one or two pairs of homologous chromosomes were obtained by Yamamoto (1994) and Mukai (2006). It can be concluded that in situ hybridization showed that hexaploid wheat entries, as expected, have different number of rDNA sites from tetraploid entries. Based on the number of 5S rDNA sites, hexaploid wheat entries could be characterized by six sites (with some exceptions) while tetraploid entries have only four sites. The cytogenetical results could differentiate among the wheat entries from different origins. Further cytogenetical studies on ‘Yecora-E’ entry are needed to understand the nature and

124

structure of ribosomal RNA genes in the specific variety which could promote our understanding of the evolution of the repetitive tandem arrays on DNA.

125

VI. REFERENCES

126

VII. REFERENCES Abercrombie M., Hickman M., Johnson M. L. and Thain M. (1990). The New Penguin Dictionary of Biology. Eighth edition pp: 581. Able J. A., Langridge P. and Milligan A. S. (2007). Capturing diversity in the cereals:many options but little promiscuity. Trends in Plant Science Vol.xxx No.x. Adhikari A. B., Wallwork H. and Goodwin S. B. (2004). Microsatellite markers linkedto the Stb2 and Stb3 genes for resistance to Septoria Tritici blotch in wheat. Crop Science 44:1403-1411. Almanza- Pinzon M.I., Khairallah M., Fox P.N. and Warburton M.L. (2003). Comparison of molecular markers and coefficients of parentage for the analysis of genetic diversity among spring bread wheat accessions. Euphytica 130: 77–86. Anderson J. A., Sorrels M. E. and Tanksley S. D. (1993). RFLP analysis of genomic regions associated to preharvest sprouting in wheat. Crop Sci 33:453–459. Appels R. and Honeycutt R. L. (1986). rDNA evolution over a billion years. In: Dutta SK (ed.) DNA Systematics. CRC Press, Boca Raton, USA, pp. 81-125. Barrett B., Bayram M. and Kidwell K. (2002). Identifying AFLP and microsatellite markers for vernalization response gene Vrn-B1 in hexaploid wheat using reciprocal mapping populations. Plant Breeding (121) 400-406. Barrett B. A., Kidwell K. K., and Fox P. N. (1998). Comparison of AFLP and pedigreebased genetic diversity assessment methods using wheat cultivars from the Pacific Northwest. Crop Sci. 38:1271-1278. Bebeli P. J. and Kaltsikes P. J. (1985). Karyotypic analysis of durum wheat varieties. CAN. J. GENET. CYTOL. Vol. 27:617-621. Bechere E., Belay G., Mitiku D. and Merker A. (1996). Phenotypic diversity of tetraploid wheat landraces from Northern and north-central regions of Ethiopia. Hereditas 124: 165-1 72. Belaj A., Satovic Z., Cipriani G., Baldoni L., Testolin R., Rallo L. and Trujillo I. (2003). Comparative study of the discriminating capacity of RAPD, AFLP and SSR

127

markers and of their effectiveness in establishing genetic relationships in olive. Theor. Appl. Genet. 107:736-744 DOI 10.1007/s00122-003-1301-5. Ben El Maati F., Jlibéne M. and Moumni M. (2004). Study of the polymorphism of common wheat using ISSR markers. J. Food, Agric. & Environ. Vol.2 (3&4): 121125. Bennett M. D. and Leitch I. J. (1995). Nuclear DNA amounts in angiosperms. Ann. Bot. 76: 113-176. Blanco A., Bellomo M. P., Cenei A., De Geovanni C., D’Ovidio R., Iacono E., Laddomada B., Simeone R., Tanzarella O. A. and Porceddu E. (1998). A genetic linkage map of durum wheat. Theor. Appl. Genet. 97:721-728. Bohn M., Friedrich H. U., Albrecht and Melchinger E. (1999). Genetic similarities among winter wheat cultivars determined on the basis of RFLPs, AFLPs, and SSRs and their use for predicting progeny variance. Crop Sci 39:228–237. Bradshaw AD (1965). Evolutionary significance of phenotypic plasticity in plants. Adv. Genet. 13: 115–155. Bryan G. J., Collins A. J., Stephenson P., Orry A., Smith J. B. and Gale M. D. (1997). Isolation and characterisation of microsatellites from hexaploid bread wheat. Theor. Appl. Genet. 94: 557-563. Burstin J. and Charcosset A. (1997). Relationship between phenotypic and marker distances: theoretical and experimental investigations. Heredity 79: 477-483. Cao W., Scoles G., Hucl P. and Chibber R. N. (2000). Phylogenetic relationships of five morphological groups of hexaploid wheat (Triticum aestivum L. em Thell.) based on RAPD analysis. Genome 43:724-727. CAP European Commission, Directorate-General for Agriculture (2003). Medium-term prospects and impact assessment of the CAP reform, EU-15 & EU-25, 2003 2010. Available at:http://europa.eu.int/comm/agriculture/publi/caprep/prospects2003b/pres.pdf. Chen Q., Conner R.L., Laroche A., Ji W.Q., Armstrong K.C., and Fedak G. (1999). Genomic in situ hybridization analysis of Thinopyrum chromatin in a wheat – Th.

128

intermedium partial amphiploid and six derived chromosome addition lines. Genome 42: 1217–1223. De Castro F., Fersino V. and Petruzzella M. (2002). Organic agriculture in Mediterranean area. CIHEAM. Instituto Agronomico Meditteraneo Bari. http://www.medobs.org/themes/agri_orga/organic_agriculture_complet.pdf (/pdf/decastro.pdf) Delaney D. E., Nasuda S., Endo T. R., Gill B. S. and Hulbert S. H. (1995). Cytologically based physical maps of the group-2 chromosomes of wheat. Theor. Appl. Genet. 91: 568–573. Dillmann C., Bar-Hen A., Guerin D., Charcosset A. and Murigneux A. (1997). Comparison of RFLP and morphological distances between maize Zea mays L. inbred lines. Consequences for germplasm protection purposes. Theor. Appl. Genet. 95:92-102. Dreisigacker S., Zhang P., Warburton M. L., Van Ginkel M., Hoisington D., Bohn M., and Melchinger A. E. (2004). SSR and Pedigree Analyses of Genetic Diversity among CIMMYT Wheat Lines Targeted to Different Megaenvironments. Crop Sci. 44:381–388. Dograr N., Akin-Yalin S., and Akkaya M. S. (2000). Discriminating durum wheat cultivars using highly polymorphic simple sequence repeat DNA markers. Plant Breeding 119:360-362. Dubreuil P. et al. (1996). Organization of RFLP diversity among inbred lines of maize representing the most significant heterotic groups. Crop Sci. 36, 790–799. Eticha F., Belay G. and Bekele E. (2006). Species diversity in wheat landrace populations from two regions of Ethiopia. Genet Res Crop Evol 53:387-393 DOI 10.1007/s10722-004-6095-z. FAO (Food and Agriculture Organization of the United Nations). (2006). FAOSTAT online statistical service. Rome: FAO. Available on-line at http://apps.fao.org. FAO (1995). FAO Statistical Databases. http://apps.fao.org/ FAOSTAT (2000). Food and Agriculture Organisation. Database collection at:

129

http://apps.fao.org/ FAO (2003). FAO Statistical Databases. http://apps.fao.org/ Flavell R. B., O’Dell M., Sharp P., Nevo E., and Beiles A. (1986). Variation in the Intergenic Spacer of Ribosomal DNA of Wild Wheat, Triticum dicoccoides, in Israel. Mol. Biol. Evol. 3(6):547-558. Ford V. S. and Gottlieb L. D. (1992). Bicalyx is a natural homeotic floral variant. Nature 358: 967-971. Forsstrom P-O., Mrker A. and Schwarzacher T. (2002). Characterization of mildew resistant lines and identification of an inverted chromosome by fluorescent in situ hybridization. Heredity 88: 349-355. Geleta L. F., Labuschagne M. T. and Viljoen C. D. (2004). Relationship between heterosis and genetic distance based on morphological traits and AFLP markers in pepper. Plant Breeding 123:467-473. ISSN 0179-9541. Godwin, I. D., Aitken E. A. B., and Smith L. W. (1997). Application of inter-simple sequence repeat (ISSR) markers to plant genetics. Electrophoresis 18:1524-1528. Gupta P. K. (1991). Cytogenetics of wheat and relative wild relatives – Triticum and Aegilos. In plants: genetics, breeding, evolution, part A. Elsevier Science Publisher, pp. 243-262. Gupta M., Chyi Y. S., Romero-Severson J. and Owen J. L. (1994). Amplification of DNA markers from evolutionary diverse genomes using single primers of simplesequence repeats. Theor. Appl. Genet. 89:998-1006. Gupta P. K., Balyan H.S., Edwards K. J., Isaac P., Korzun V., Röder M, et al. (2002). Genetic mapping of 66 new microsatellite (SSR) loci in bread wheat. Theor. Appl. Genet. 105:413-422 (DOI 10.1007/s00122-002-0865-9). Gupta P. K. and Varshney R. K. (2000). The development and use of microsatellite markers for genetic analysis and plant breeding with emphasis on bread wheat. Euphytica 113:163–185. Gupta P. K., Varshney R. K., Sharma P. C. and Ramesh B. (1999). Molecular markers and their applications in wheat breeding. Plant Breed 118:369-390.

130

Gutteling E. W., Riksen J. A. G., Bakker J. and Kammenga J. E. (2007). Mapping phenotypic

plasticity and genotype–environment interactions affecting life-history traits in Caenorhabditis elegans Heredity 98, 28–37. doi:10.1038/sj.hdy.6800894. Hagidimitriou M., Katsiotis A., Menexes G., Pontikis C. and Loukas M. (2005). Genetic diversity of major Greek olive cultivars using molecular (AFLPs and RAPDs) markers and morphological traits. J. Amer. Soc. Hort. Sci. 130 (2): 211-217. Hair J. R., Anderson R. E., Tatham R. L. and Black W. C. (1995). Multivariate data analysis with readings. 4th edition, Prentice-Hall, Englewood Cliffs, N. J. Hamza S., Ben Hamida W., Rebaï A. and Harrabi1 M. (2004). SSR-based genetic diversity assessment among Tunisian winter barley and relationship with morphological traits. Euphytica 135: 107–118. Hausmann N. J., Juenger T. E. , Sen S., Stowe K. A., Dawson T. E.and Simms E. L. (2005). Quantitative trait loci affecting d13C and response to differential water availability in Arabidopsis thaliana. Evolution 59: 81–96. Hayden M. J., Khatkar S. and Sharp P. J. (2001). Targetting microsatellites (SSRs) in genetic linkage maps of bread wheat. Aus. J. Agric. Res. 52:1143-1152. Helgason A., Nicholson G., Stefanson K. and Donnelly P (2003). A reassessment of genetic diversity in Icelanders: Strong evidence from multiple loci for relative homogeneity caused by genetic Drift. Annals of Human Genetics 67:281-297. Hohmann U., Endo T. R., Gill K. S. and Gill B. S. (1994). Comparison of genetic and physical maps of group 7 chromosomes from Triticum aestivum L. Mol. Gen. Genet. 245:644-653. Holton T. A. (2001). Plant Genotping by analysis of microsatellites. Chapter 2 of Plant Genotyping: the DNA Fingerprinting of plants book (ed. R. J. Henery) pp: 15-27. Hou Y., Yan Z., Wei Y. and Zheng Y. (2005) Genetic diversity in barley from west China based on RAPD and ISSR analysis. Barley Genetics Newsletter 35:9-22. Huang X. Q., Börner A., Röder M. S. and Ganal M. W. (2002). Assessing genetic diversity of wheat (Triticum aestivum L.) germplasm using microsatellite markers. Theor. Appl. Genet. 105:699–707 DOI 10.1007/s00122-002-0959-4.

131

IENICA Report from the state of Greece forming part of the IENICA project Athens, October 1999. Jaccard P. (1908). Nouvelles recherches sur la distribution florale. Bull. Soc. Vaudoise Sci.Nat. 44: 223-270. Jackson R. B., Manwaring J. H. and Caldwell M. M. (1990). Rapid physiological adjustment of roots to localized soil enrichment. Nature 344:58-60. Jeffreys A. J., Wilson V. and Thein S. L. (1985). Hypervariable “minisatellite” regions in human DNA. Nature Vol. 314:67-73. Jones N., Ougham H. and Thomas H. (1997) Markers and mapping: we are all geneticists now. New Phytol. 137:165-177. Joshi S. P., Prebhakar K., Ranjekar P. K. and Gupta V. S. (1999). Molecular markers in plant genome analysis. Curr. Sci. 77: 230-240. Katsiotis A., Schmidt T. and Heslop-Harrison J. S. (1996). Chromosomal and genomic organization of Ty1-copia-like retrotransposon sequences in the genus Avena. Genome 39:410–417. Khan R. R., Bariana H. S., Dholakia B. B., Naik S. V., Lagu M. D., Rathjen A. J., Bhavani S. and Gupta V. S. (2005). Molecular mapping of stem and leaf rust resistance in wheat. Theor. Appl. Genet. 111: 846–850 DOI 10.1007/s00122-0050005-4. Khlestkina E. K., Roder M. S., Efremova T. T., Borner A. and Shumny V. K. (2004). The genetic diversity of old and modern Siberian varieties of common spring wheat as determined by microsatellite markers. Plant Breeding 123: 122-127. Kojima, T., Nagaoka T., Noda K., and Ogihara Y. (1998). Genetic linkage map of ISSR and RAPD markers in Einkorn wheat in relation to that of RFLP markers. Theor. Appl. Genet. 96: 37-45. Korzun V. (2003). Molecular markers and their applications in cereals breeding. Cited in: Marker assisted selection: A fast track to increase genetic gain in plant and animal breeding, Session I: MAS in plants P. 18-22. Korzun V., Roder M., Worland A. J. and Borner A. (1997a). Interchromosomal mapping

132

of genes for dwarfing (Rht12) and vernalization response (vern1) in wheat by using RFLP and microsatellite markers. Plant Breeding 116:227-232. Korzun V., Borner A., Worland A. J., Law C. N. and Roder M. (1997b). Application of microsatellite markers to distinguish inter-varietal chromosome substitution lines of wheat (Triticum aestivum L.). Euphytica 95:149-155. Korzun V., Roder M. S., Wendehake K., Pasqualone A., Lotti C., Ganal M. W. and Blanco A. (1999). Integration of dinucleotide microsatellites from hexaploid bread wheat into a genetic linkage map of durum wheat. Theor. Appl. Genet. 98:12021207. Kota R. S., Gill K. S., Gill B. S. and Endo T. R. (1993). Acytogenetically based physical map of chromosome 1B in common wheat. Genome 36: 548–554. Kumar L. S. (1999) DNA markers in plant improvement: An overview. Biotechnology Advances 17:143-182. Lage J., Warburton M. L., Crossa J., Skovmand B. and Andersen S. B.

(2003).

Assessment of genetic diversity in synthetic hexaploid wheats and their Triticum dicoccum and Aegilops tauschii parents using AFLPs and agronomic traits. Euphytica 134:305-317. Landjeva S., Korzun V. and Ganeva G. (2006). Evaluation of genetic diversity among Bulgarian winter wheat (Triticum aestivum L.) varieties during the period 19252003 using microsatellites. Genet Res Crop Evol 53:1605-1614. Langridge P., Lagudah E. S., Holton T. A., Appels R., Sharp P. J. and Chalmers K. J. (2001). Trends in genetic and genome analyses in wheat: a review. Aust. J. Agric. Res. 52:1043-1077. Lefebvre V., Goffinet B., Chauvet J. C., Caromel B., Signoret P., Brand R. and Palloix A. (2001). Evaluation of genetic distances between pepper inbred lines for cultivar protection purposes: comparison of AFLP, RAPD and phenotypic data. Theor. Appl. Genet. 102: 741-750. Lehman A., Creighton L., Sall J., Jones B. Vang E., Blackwelder M., Zangi A. D. and Corcoran B. (2005). JMP IN 5 software, SAS Inc., Ltd. Nc. JMP.

133

Liao C.Y., Wu P., Hu B., Yi K.K. (2001). Effects of genetic background and environment on QTLs and epistasis for rice (Oryza sativa L.) panicle number. Theor. Appl. Genet. 103:104–111 Lima V. L. D., Seki H. A. and Rumjanek F. D. (2003). Microsatellite polymorphism in wheat from Brazilian cultivars; inter- and intra-varietal studies. Genet. Mol. Biol. Vol. 26 (3): 349-353. Lima-Brito J., Carvalho A., Martin A., Heslop-Harison J. S. and Guedes-Pinto H. (2006). Morphological, yield, cytological and molecular characterization of a bread wheat X tritordeum F1 hybrid. J. Genet. Vol. 85 (2): 123-131. Lou M, Yng Z., Devorak J. (1998). Position of ribosomal RNA multigene loci on meiotic recombination in wheat. Genetics 149: 1105-1113. Ma Z. Q., Sorrels M. E. and Tanksley S. D. (1994). RFLP markers linked to powdery mildew resistance genes pm1, pm2, pm3, and pm4 in wheat. Genome 37:871-875. Malysheva L., Sjaksta T., Matzk F., Roder M. and Ganal M. (2003). Molecular cytogenetic analysis of wheat-barley hybrids using genomic in situ hybridization and barley microsatellite markers. Genome 46: 314-322. Manifesto M. M., Schlatter A. R., Hopp H. E., Suarez, E. Y. and Dubcovsky J. (2001). Quantitative evaluation of genetic diversity in wheat germplasm using molecular markers. Crop Sci. 41:682–690. Mantel N. (1967). The detection of disease clustering and a generalized regression approach. Cancer Res 27:209-220. Maric S., Bolaric S., Martincic J., Pejic I. and Kozumplik V. (2004). Genetic diversity of hexaploid wheat cultivatrs estimated by RAPD markers, morphological traits and coefficients of parentage. Plant Breed 123:366-369. Martynov S.P. and Dobrotvorskaya T.V. (2006). Genealogical analysis of diversity of Russian winter wheat cultivars (Triticum aestivum L.). Genet. Res. Crop. Evol. 53: 379–386. DOI: 10.1007/s10722-004-0572-2. Medini M., Hamza S., Rebai A. and Baum M. (2005). Analysis of genetic diversity in Tunisian durum wheat cultivars and related wild species by SSR and AFLP

134

markers. Genet. Res. Crop. Evol. 52: 21-31. Mesfin A., Frohberg R. C. and Anderson J. A. (1999). RFLP markers associated with high grain protein from Triticum turgidum L. var. dicoccoides introgressed into hard red spring wheat. Crop Sci. 39:508-513. Melchinger A. E. (1993). Use of RFLP markers for analysis of genetic relationship among breeding materials and prediction of hybrid performance. p. 621.628. In D.R. Buxton et al. (ed.) International. Crop science I. CSSA, Madison, WI. Miller T. E., Reader S. M., Purdie K. A., Abbo S., Dunford R. .P. and King L. P. (1995). Fluorescent in situ hybridization as an aid to introducing alien genetic variation into wheat. Euphytica 85: 275-279. Moghaddam M. E., Trethowan R. M., William H. M., Rezai A., Arzani A. and Mirlohi A. F. (2005). Assessment of genetic diversity in bread wheat genotypes for tolerance to drought using AFLPs and agronomic traits. Euphytica 141: 147-156. Mohammadi S. A. and Prasanna B. M. (2003). Analysis of genetic diversity in crop plants—salient statistical tools and considerations. Crop Sci. 43:1235–1248. Mohan M., Nair S., Bhagwat A., Krishna T. G. and Yano M. (1997). Genome mapping, molecular markers and marker-assisted selection in crop plants. Molec Breed 3:87–103. Motawei M. I., Al-Doss A. A. and Moustafa K. A. (2007). Genetic diversity among selected wheat lines differing in heat tolerance using molecular markers. J. Food Agric. and Environ. Vol. 5 (1) 180-183. Motzo R., Fois S. and Giunta F. (2004). Relationship between gain yield and quality of durum wheats from different areas of breeding. Euphytica 140: 147-154. Mukai Y. (2006). Perspectives in molecular cytogenetics of wheat. Frontiers of Wheat Bioscience: 17-30 Memorial Issue, Wheat Information Service No.100. Available in http://www.shigen.nig.ac.jp/wheat/wis/No100/p17/p17.1.html. Mukai Y., Endo T. R., and Gill B. S. (1990). Physical Mapping of the 5S rRNA Multigene Family in Common Wheat. J. Heredity 81:290-295. Mukai Y, Endo TR and Gill BS (1991). Physical mapping of the 18S.26S rRNA

135

multigene family in common wheat: Identification of a new locus. Chromosoma 100: 71-78. Nagaoka T. and Ogihara Y. (1997). Applicability of inter-simple sequence repeat polymorphisms in wheat for use as DNA markers in comparison to RFLP and RAPD markers. Theor. Appl. Genet. 94:597-602. Naghavi M. R., Mardi M., Ramshini H. A. and Fazelinasab B. (2004). Comparative analyses of the genetic diversity among bread wheat genotypes based on RAPD and SSR markers. Iranian J. Biotech. Vol. 2 (3): 195-202. Panchen A. L. (1992). Classification, evolution and the nature of biology. Cambridge Univ. Press, Cambridge, England. Parker G. D., Fox P. N., Langridge P., Chalmers K., Whan B. and P. F. Ganter (2002). Genetic diversity within Australian wheat breeding programs based on molecular and pedigree data. Euphytica 124: 293–306. Pasqualone A., Lotti C., Bruno A., De Vita P., Di Fonzo N. and Blanco A. (2000). Use of ISSR markers for cultivar identification in durum wheat. In Royo C. (ed.), Nachit M. M. (ed.), De Fonzo H. (ed.), Araus J. (ed.). Durum wheat improvement in the Mediterranean region: New challenges. CIHEAM-IAMZ, 2000. P. 157-161. Pecetti L. and Damania A. B. (1996). Geographic variation in tetraploid wheat (Triticum turgidum ssp. Turgidum convar, durum) landraces from two provinces in Ethiopia. Genet. Res. Crop. Evol. 43: 395-407. Pestsova E., Ganal M. W. and Roder M. S. (2000). Isolation and mapping of microsatellite markers specific for the D genome of bread wheat. Genome 43:689697. Plaschke J. M., Ganal M. W. and Röder M. S. (1995). Detection of genetic diversity in closely related bread wheat using microsatellite markers. Theor. Appl. Genet. 91:1001-1007. Prasada M., Kumar N., Kulwal P. L., Roder M. S., Balyan H. S., Dhaliwal H. S. and Gupta P. K. (2003). QTL analysis for grain protein content using SSR markers and validation studies using NILs in bread wheat. Theor. Appl. Genet. 106:659-

136

667. Prasada M., Varshney R. K., Kumar A., Balyan H. S., Sharma P. C., Edwards K. J., Singh H., Dhaliwal H. S., Roy J. K. and Gupta P. K. (1999). A microsatellite marker associated with a QTL for protein content on chromosome arm 2DL of bread wheat. Theor. Appl. Genet. 99:341-345. Pujar S., Tamhankar S. A., Gupta V. S., Rao V. S. and Ranjekar P. K. (2002). Diversity analysis of Indian wheat using intersimple sequence repeat markers reveals their superiority over random amplified polymorphic DNA markers. Biochemical Genetics Vol. 40 (1/2) 63-69. Rasmusson D. C. and Phillips R. L. (1997). Plant breeding progress and genetic diversity from De Novo variation and elevated epistasis. Crop. Sci. 37: 303-310. Rawashdeh N. K., Haddad N. I., Al-Ajlouni M. M. and Turk M. A. (2007). Phenotypic diversity of durum wheat (Triticum durum Desf.) from Jordan. Genet Res Crop Evol (2007) 54:129–138 Springer 2006 DOI 10.1007/s10722-005-2636-3. Rayburn A. L. and Gill B. S. (1985). Use of biotin-labeled probes to map specific DNA sequences on wheat chromosomes. The Journal of Heredity 76:78-81. Reif J. C., Zhang P., Dreisigacker S., Warburton M. L., Van Ginkel M., Hoisington D., Bohn M. and Melchinger A. E. (2005). Wheat genetic diversity trends during domestication and breeding. Theor. Appl. Genet. 110: 859–864 DOI 10.1007/s00122-004-1881-8. Ribeiro-Carvalho C., Guedes-Pinto H., Igrejas G., Stephenson P., Schwarzacher T. and Heslop-Harrison J. S. (2004). High levels of genetic diversity throughout the range of the portuguese wheat landrace ‘Barbela’ . Annals of Botany 94: 699–705. Robinson R. (2003). Quantitative Traits. Genetics Volume 4Q–Z, Macmillan Reference, USA, Thomson Gale Inc. pp : 1-3. Röder M. S., Korzun V., Wendehake K., Plaschke J., Tixier M., Leroy P. and Ganal M. W. (1998). A microsatellite map of wheat. Genetics 149:2007-2023. Röder M. S., Plaschke J., Konig S. U., Borner A., Sorrells M. E., Tanksley S. D. and Ganal M. W. (1995). Abundance, variability and chromosomal location of

137

microsatellites

in

wheat.

Mol.

Gen.

Genet.

246:327-333

DOI

10.1007/BF00288605. Rohlf F.J. (1998). NTSYSpc. Numerical taxonomy and multivariate analysis system, version 2.02c. Exeter Software, New York. Roussel V., Koenig J., Beckert M. and Balfourier F. (2004). Molecular diversity in French bread wheat accessions related to temporal trends and breeding programmes. Theor. Appl. Genet. 108:920-930 DOI 10.1007/s00122-003-1502-y. Roy J. K., Bandopadhyay R., Rustgi S., Balyan H. S. and Gupta P. K. (2006). Association analysis of agronomically important traits using SSR, SAMPL and AFLP markers in bread wheat. CURR SCI 90 (5) 10 MARCH 683-689. Roy J. K., Lakshmikumaran M. S., Balyan H. S. and Gupta P. K. (2004). AFLP-based genetic diversity and its comparison with diversity based on SSR, SAMPL, and phenotypic traits in bread wheat. Biochem Genet 42(1/2) February 43-59. Sanchez-Moran E., Benavente E. and Orellana J. (1999). Simulatneous identification of A, B, D and R genomes by genomic in situ

hybridization in wheat-rye

derivatives. Heredity 83: 249-252. Schneider A., Linc G. and Molnar-Lang M. (2003). Fluorescence in situ hybridization polymorphism using two repetitive DNA clones in different cultivars of wheat. Plant Breeding 122: 396-400. ISSN 0179-9541. Schut, J.W., Qi X. and Stam P. (1997). Association between relationship measures based on AFLP markers, pedigree data and morphological traits in barley. Theor. Appl. Genet. 95: 1161–1168. Schwarzacher T., Anamthawat-Jonsson K., Harrison G. E., Islam A. K. M. R., Jia J. Z., King I. P., Leitch A. R., Miller T. E., Reader S. M., Rogers W. J., Shi M. and Heslop-Harison J. S. (1992). Genomic in situ hybridization to identify alien chromosomes and chromosome segments in wheat. Theor. Appl. Genet. 84: 778786. Schwarzacher T. and Heslop-Harison P. (2000). Practical in situ hybridization. First ed. BIOS Scientific Publishers Ltd.

138

Shariflou M. R. and Sharp P. J. (1999). A polymorphic microsatellite in the 3` end of ‘waxy’ genes of wheat, Triticum aestivum.

Short Communication in Plant

Breeding 118:275-277. Singh R. J. (2003). Plant Cytogenetics. 2nd edition CRC. Press LLC. Ram J. Singh. Sneath P. H. A. and Sokal R. R. (1973) Numerical taxonomy. Freeman, San Francisco. Soleimani V. D., Baum B. R., Johnson D. A. (2002). AFLP and pedigree-based genetic diversity estimates in modern cultivars of durum wheat [Triticum turgidum L. subsp. durum (Desf.) Husn.]. Theor. Appl. Genet. 104:350–357. Spooner D., van Treuren R. and de Vicente M. C. (2005). Molecular markers for genebank management. IPGRI Technical Bulletin No. 9 pp.11-17. Stachel M., Lelley T., Grausgruber H. and Vollmann J. (2000). Application of microsatellites in wheat (Triticum aestivum L.) for studing genetic differentiation caused by selection for adaptation and use. Theor. Appl. Genet. 100:242-248. Stephenson P., Bryan G., Kirby J., Collins A., Devos K., Busso C. and Gale M. (1998). Fifty new microsatellite loci for the wheat genetic map. Theor. Appl. Genet. 97:946-949. Stodart B. J., Mackay M. and Raman H. (2005). AFLP and SSR analysis of genetic diversity among landraces of bread wheat (Triticum aestivum L. em. Thell) from different geographical regions. Aust. J. Agr. Res. 56:691-697. Tranquilli1 G., Appendino M. L., Pflüger L. A., Bullrich L., Lewis S. and Suarez E. Y. (2000). Morphoagronomic and biochemical variation in an Argentinean landrace of wheat. Genet Res Crop Evol. 47: 281–284. Ungerer M. C., Halldorsdottir S. V., Purugganan M. D. and Mackay T. F. C. (2003). Genotype-by environment interactions at quantitative trait loci affecting inflorescence development in Arabidopsis thaliana. Genetics 165: 353–365. UPOV (Union pour la Protection Obtentions Vegetales, 2002. Van Deynze A. E., Dubcovsky J., Gill K. S., Nelson J. C., Sorrells M. E., Dvorak J., Gill B. S., Lagudah E. S., McCouch S. R. and Appels R. (1994). Molecular-genetic maps for group 1 chromosomes of Triticeae species and their relation to

139

chromosomes in rice and oat. Genome 38:45-59. Varshney R. K., Sharma P. C., Gupta P. K., Balyan H. S., Ramesh B., Roy J. K., Kumar A. and Sen A. (1998). Low level of polymorphism detected by SSR probes in bread wheat. Short Communication in Plant Breeding 117:182-184. Varshney R. K.,Prasada M., Roy J. K., Kumar N., Singh H., Dhaliwal H. S., Balyan H. S. and Gupta P. K. (2000). Identification of eight chromosomes and a microsatellite marker on 1AS associated with QTL for grain weight in bread wheat. Theor. Appl. Genet. 100:1290-1294. Wang Z., Zemetra R. S., Hansen J., Hang A., Mallory-Smith C. A., and Burton C. (2002). Determination of the Paternity of Wheat (Triticum aestivum L) _ Jointed Goatgrass (Aegilops cylindrica Host) BC1 Plants by Using Genomic In Situ Hybridization (GISH) Technique. Crop Sci. 42:939–943. Ward J. H. Jr. (1963). Hierarchical grouping to optimize an objective function. J. Am. Statist. Assoc. 58: 236-244. Wolfe A. D. and Liston A. (1998). Contributions of PCR based methods to plant systematic and evolutionary biology. In: Soltis DE, Soltis PS, Doyle JJ (eds) Molecular systematics of plants ІІ: DNA sequencing. Kluwer, Dordrecht, pp 4386. Wu R., Grissom J. E., McKeand S. E. and O'Malley D. M. (2004). Phenotypic plasticity of fine root growth increases plant productivity in pine seedlings. BMC Ecology 4:14 doi: 10.1186/1472-6785-4-14. Yamamoto M. (1994). Molecular-cytogenetic analysis of the rDNA region in Triticum and Aegilops. PhD thesis, Kyoto University, Japan, pp. 248. Yifru T., Hammer K., Huang X. Q. and Roder M. S. (2006). Regional patterns of microsatellite diversity in Ethiopian tetraploid wheat accessions. Plant Breeding 125: 125-130. Zanetti S., Winzeler M., Keller M., Keller B. and Messmer M. (2000). Genetic analysis of pre-harvest resistance in a wheat × spelt cross. Crop Sci. 40:1406-1417. Zeven A. C. and Schachl R. (1989). Groups of bread wheat landraces in Austrian alps.

140

Euphytica 41:235-246 Zhang X. Y., Li C. W., Wang L. F., Wang H. M., You G. X. and Dong Y. S. (2002). An estimation of the minimum number of SSR alleles needed to reveal genetic relationships in wheat varieties. I. Information from large-scale planted varieties and cornerstone breeding parents in Chinese wheat improvement and production. Theor. Appl. Genet. 106:112–117 DOI 10.1007/s00122-002-1016-z. Zhang P., Dreisigacker S., Buerkert A., Alkhanjari S. and Melchinger A. E. (2006). Genetic diversity and relationships of wheat landraces from Oman investigated with SSR markers. Genet. Res. Crop. Evol. 53: 1351-1360. Zietkiewicz E., Rafalski A. and Labuda D. (1994). Genomic fingerprinting be simple sequence repeat (SSR)-anchored polymerase chain reaction amplification. Genomics 20:176-183.

141

VII. APPENDICES

142

Appendix 1 Polyacrylamide gel electrophoresis 1. Prepare and run Polyacrylamide gel 6% (w/v): A. MATERIALS: Acrylamide

Bis-acrylamide

Ammonium persulfate

TEMED

Urea

Ethanol

Glacial Acetic Acid

Kimwipes

Binding Silane

Repel Silane ES (AppliChem)

Vaseline

B. EQUIPMENT: Vertical Electrophoresis unit

Power supply Glass plates

Gilson Pippets

Spacers

Tape& clips

C. THE METHOD: 1. Soak the glass plates first in 10% sodium hydroxide (200 g in 2 L distilled H2O) and then in 1N hydrochloric acid (170 ml of 11N HCl added to 1830 ml of distilled H2O) for two hours in each. 2. Clean each plate and the spacers well with soap and rinse with a lot of water to remove all soap, with a final rinse using de-ionized water. Hold the plates by the edges or wear gloves, in order not to deposit dirt on working surface of the plates. Mark the inner and outer surfaces of the plates. Lay the glass plates aside to dry. Wipe the inner surface of each plate with 95% ethanol (wear gloves and don’t touch anything except the glass plates). The glass plates must be free of grease spots to prevent air bubbles forming in the gel.

143

3. For the small glass plate: a. Prepare three 1.5 ml eppendorf tubes each one containing 6.5 µl glacial acetic acid, 6.5 µl binding silane (AppliChem, A3797, 0010) and 1287 µl ethanol. Pour the solution from one tube onto the middle of the inner surface of the small plate and spread it all over with a kimwipe. Let dry for 3 minutes. b. Repeat the previous step twice using the other two eppendorf tubes, but let the plate 5 minutes to dry. c. Remove the excess of the detergent used by wiping with a kimwipe wetted with 95% ethanol. Let the plate dry while treat the other plate. d. Change the gloves before the treating the large plate so you don’t transfer any binding silane on it. 4. For the large glass plate: a. Coat the inner surface of the large plate with 1.6 ml Repel Silane ES (Pharmacia Biotech., Code No. 17-1332-01) using a kimwipe. Let dry for 7 minutes. b. Repeat the previous step leaving the plate to dry for 10 minutes. c. Use a kimwipe to remove the excess materials from the surface of the plate. Preparation of both plates must be done under a hood to avoid respiration of materials that can be hazardous. 5. Assembling the plates:

Put

the spacers on the sides of the large plate and place the small plate on it. Don’t touch the inner surfaces of the plates. Line up the edges of plates and spacers and then clamp the edges. Place the tape on the bottom making a corner. Put tape on the sides with the same manner to prevent leak of the polyacrylamide before its polymerization. Tighten the two glass plates using clips.

144

6. Prepare 6% polyacrylamide solution as follows: a. Prepare 200 ml of 30% (19: 1) acrylamide: bisacrylamide solution, dissolve 58 g of acrylamide and 2 g of bisacrylamide in 120 ml of preheated ddH2O to 37oC. b. The pH of the solution should be less than 7 in order to allow the running buffer to flow in the gel during electrophoresis. Test the pH of the solution and then bring the volume to 200 ml using ddH2O. This solution can be stored up to one month. c. To prepare 80 ml of 6% polyacrylamide solution, mix 16 ml 30% polyacrylamide with 14.4 g of Urea, 8 ml of 10 X TBE buffer, 35 µl of TEMED and bring the volume 80 ml by adding ddH2O. Add 550 µl of 10% ammonium persulfate (APS) directly before pouring the gel and swirl to mix. 7. Tilt the glass plate sandwich so that the one bottom corner is lower than the other and the tip edge is elevated at about a 30 degree angle from horizontal. Let the polyacrylamide solution flow in so that it runs down one side filling the lowermost corner, then spreading across the bottom and finally up the other side. Lay the gel flat. If bubbles are present near the top they can be pulled out with the comb. Insert comb flat side to gel, making sure it is perpendicular to the edge of the gel. Pour a small amount of polyacrylamide over the top of the comb, and then put clips over the comb edge of the gel. Put water-wetted tissue paper over the plates to prevent sticking polymerized acrylamide bits on the place where the comb is. 8. Leave the gel flat two and half hours to polymerize. 9. After polymerization, remove the clips from the plates. Gently remove the comb and clean off any polyacrylamide from the outside surfaces of the plates using 95% ethanol. Clean the top edge where the comb was using TBE buffer (1X) and clean the comb. Remove the stick tape. 10. Pour about 400 ml of TBE 1X buffer into the bottom reservoir of vertical electrophoresis and then tilt the gel into place with the short plate (small glass

145

plate) against the apparatus and the taller plate (large glass plate) facing you. Tighten all four clamps. Don’t over tighten and check that the drain valve is closed, before adding TBE buffer in the top tank. Check for leaks, and use Vaseline to seal. If there are no leaks, fill the top tank past the top of the shorter plate (small plate). 11. Use a syringe to rinse the top part of the gel and to remove all polyacrylamide debris, dissolved chemicals and air bubbles. Insert the comb until the tips just press into the polyacrylamide. 12. Samples are usually loaded at 4 to 5 µl, but this can be reduced or increased depending on the samples. The mix contains loading buffer (Bromophenol Blue 0.25%, Xylene Cyanol 0.25% and Sucrose 40%) in addition to the PCR product. 13. After all samples and the size standards are loaded, close the reservoir covers, insert the wires into the gel electrophoresis unit and the power supply. Run the gel at 400 volts for about 10-12 hours. The gel is running usually overnight, so you have to prevent leaks of the buffer during the running period. 14. During the run there are two different blue lines, the first is light-colored (low molecular weight) and it is from bromophenol blue, while the other is more dark (high molecular weight) and it is from xylene cyanol. The electrophoresis is complete when the line of xylene cyanol (the high molecular weight) is 8-cm far from the end of the plate. 15. After the run is completed, turn off power, remove the wires, drain the top reservoir, remove and clean well the gel. 16. Put the plates in a tray containing distilled water and try gently to separate the glass plates without destroying the polymerized. The small plate has to be on top; use a plastic wedge to gently separate the plates apart.

146

2. Silver staining polyacrylamide gel: 1. It is important to use high purity chemicals and distilled water to avoid yellowing or browning of the gels. Use silver staining kit (Promega Silver Sequence DNA Staining Reagents, No. Q4134) to stain polyacrylamide gel. 2. Chill 2 liters of de-ionized water per gel overnight in the cold room one day before staining. 3. Freshly, prepare 2 liters of 10% acetic acid. Put it into one of the washing trays on big shaker. 4. Put the small glass plate (with the gel side up) in the tray which contains the 2 liters of acetic acid 10% and shake it for 20 minutes. 5. While shaking, prepare the silver nitrate solution in 2 liters of de-ionized water, 2 g of silver nitrate and 3 ml of 30% formaldehyde solution. 6. When the gel has soaked in the acetic acid for 20 minutes, soak the gel in 2 liters of distilled water for about 4 minutes until the acetic acid has leached out. Keep aside the acetic acid in its tray to use it again at the end of the process. 7. After 4 minutes in water, place the gel in the silver nitrate stain solution and shake for 30 minutes. Note that extending any step for too long may allow DNA to diffuse. 8. During this period, prepare the Sodium Carbonate solution. In 2 liters of chilled water, add 60 g of sodium carbonate, 3 ml of 30% formaldehyde solution and 4 mg of sodium thiosulfate; mix together as follows: in a large container with the water and a magnetic stirrer, add slowly the sodium carbonate. The sodium thiosulfate and the formaldehyde are added some minutes before using the solution. 9.

After the completion of step No. 8, remove the gel plate from the silver nitrate solution, rinse briefly in distilled water and then soak it directly in the sodium carbonate solution while shaking until the bands appear. Note that the bands start

147

to appear after about 5 minutes. Over-staining will produce too much background and poor contrast. 10.

Place the gel plate in 10% acetic acid to fix for 5 minutes.

11.

Rinse out the acetic acid with distilled water. Let the gel dry.

148

Appendix 2 Stock solutions 1. 20 X SSC solution NaCl (3M)

175 g

Na3Citrate. 2H2O (0.3M)

88.2 g

Adjust the volume with water to one liter and autoclave. 2. 4 X SSC/Tween solution Tween 20

1 ml

20 X SSC solution

100 ml

Adjust the volume with water to 500 ml. 3. 10 X TE buffer Tris base (0.1M)

6.06 g

EDTA (0.01M)

1.86 g

Adjust the volume with water to 500 ml and autoclave. 4. 50 X TAE buffer Tris base (2M)

121 g

EDTA (0.5M)

50 ml

Glacial acetic acid

28.55 ml

Adjust the volume with water to 500 ml and autoclave. 5. 0.5 M EDTA solution Dissolve 18.6 g in 100 ml water. Autoclave the solution. 6. 3 M NaAcetate solution Dissolve 24.61 g of sodium acetate (CH3COONa) in 100 ml water. 7. 1 M Tris-HCl solution, pH 6.8 Dissolve 12.1 g of Tris base in 80 ml water. Adjust the pH using HCl to 6.8. Adjust the volume with water to one liter and autoclave.

149

8. 10 X Enzyme buffer Citric acid monohydrate (0.1 M) (MW=210.1)

40 ml

Trisodium cytayle. 2 hydrate (0.1M) (MW=294.1)

60 ml

9. Root tip digestion enzyme solution a. 2% cellulose (CALBIOCHEM)

1.6 g

20% pectinase

16 ml

Water

64 ml

b. 2% cellulose (ONOZUKA)

0.4 g

20% pectinase

4 ml

Water

16 ml

Mix (a) and (b) solutions and store at -20oC till use. Use 1-2 ml of this mixture to digest the root tips. 10. Chloroform/iso-amyl alcohol (IAA) 24:1 (500 ml) Chloroform

480 ml

Iso amyl alcohol (IAA)

20 ml

11. 2% CTAB solution (total volume 250 ml) CTAB

5g

NaCl

20.45 g

PVP

2.5 g

1 M Tris-HCl 25 ml 0.5 M EDTA 10 ml 12. 5% CTAB (total volume 100 ml) CTAB

5g

NaCl

4.09 g

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164