Front Cover
 Title Page
 Table of Contents
 Color plates
 List of Tables
 Executive summary
 Introductions : 15 years of progress...
 Perennial and annual wheat relatives...
 Interspecific crosses : hybrid...
 Intergeneric crosses : hybrid production...
 Production of polyhaploid wheat...
 Applications of tissue culture...
 Applications of biochemical markers...
 Applications of molecular markers...
 Conclusions and a look to...
 Appendix 1
 Appendix 2
 Appendix 3
 Appendix 4
 Appendix 5
 Appendix 56

Group Title: Utilizing wild grass biodiversity in wheat improvement : 15 years of wide cross research at CIMMYT
Title: Utilizing wild grass biodiversity in wheat improvement
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00077532/00001
 Material Information
Title: Utilizing wild grass biodiversity in wheat improvement 15 years of wide cross research at CIMMYT
Series Title: CIMMYT research report
Physical Description: xxiv, 140 p. : ill. (some col.) ; 26 cm.
Language: English
Creator: Mujeeb-Kazi, A
Hettel, Gene P
Publisher: CIMMYT
Place of Publication: Mexico D.F
Publication Date: 1995
Subject: Wheatgrasses   ( lcsh )
Biodiversity   ( lcsh )
Wheat -- Breeding   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references (p. 107-118).
Statement of Responsibility: edited by A. Mujeeb-Kazi and G.P. Hettel.
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00077532
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 50420051
isbn - 968692308X
issn - 0188-2465 ;


This item has the following downloads:

ResRep02MKazi ( PDF )

Table of Contents
    Front Cover
        Front cover
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
        Page v
        Page vi
    Color plates
        Page vii
    List of Tables
        Page viii
        Page ix
        Page x
        Page xi
        Page xii
        Page xiii
        Page xiv
        Page xiv
        Page xiv
    Executive summary
        Page xv
        Page xvi
        Page xvii
        Page xviii
        Page xix
        Page xx
        Page xxi
        Page xxii
        Page xxiii
        Page xxiv
    Introductions : 15 years of progress in wheat wide crosses at CIMMYT
        Page 1
        Page 2
        Page 3
        Page 4
    Perennial and annual wheat relatives in the Triticeaea
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
    Interspecific crosses : hybrid production and utilization
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
    Intergeneric crosses : hybrid production and utilization
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Production of polyhaploid wheat plants using maize and Tripsacum
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
    Applications of tissue culture in wheat wide crosses
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
    Applications of biochemical markers in wheat wide crosses
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
    Applications of molecular markers in wheat wide crosses
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
    Conclusions and a look to the future
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
    Appendix 1
        Page 119
    Appendix 2
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
    Appendix 3
        Page 126
        Page 127
    Appendix 4
        Page 128
    Appendix 5
        Page 129
    Appendix 56
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
Full Text


Utlzn Wil Grs Bioia a i

Edte b A. Mj -az an

le, 41`1



Utilizing Wild Grass Biodiversity
in Wheat Improvement:
15 Years of
Wide Cross Research

Edited by A. Mujeeb-Kazi and
G.P. Hettel

CIMMYT is an internationally funded, nonprofit scientific research and training organization.
Headquartered in Mexico, the Center is engaged in a worldwide research program for maize, wheat,
and triticale, with emphasis on improving the productivity of agricultural resources in developing
countries. It is one of 16 nonprofit international agricultural research and training centers supported by
the Consultative Group on International Agricultural Research (CGIAR), which is sponsored by the
Food and Agriculture Organization (FAO) of the United Nations, the International Bank for
Reconstruction and Development (World Bank), and the United Nations Development Programme
(UNDP). The CGIAR consists of a combination of 40 donor countries, international and regional
organizations, and private foundations.

CIMMYT receives core support through the CGIAR from a number of sources, including the
international aid agencies of Australia, Austria, Belgium, Brazil, Canada, China, Denmark, Finland,
France, Germany, India, Italy, Japan, Mexico, the Netherlands, Norway the Philippines, Spain,
Switzerland, the United Kingdom, and the USA, and from the European Economic Commission, Ford
Foundation, Inter-American Development Bank, OPEC Fund for International Development, UNDP,
and World Bank. CIMMYT also receives non-CGIAR extra-core support from the International
Development Research Centre (IDRC) of Canada, the Rockefeller Foundation, and many of the cole
donors listed above.

Responsibility for this publication rests solely with CIMMYT.

Correct Citation: Mujeeb-Kazi, A., and G.P Hettel, eds. 1995. Utilizing Wild Grass Biodiversity in
Wheat Improvement: 15 Years of Wide Cross Research at CIMMYT. CIMMYT Research Report No. 2.
Mexico, D.F.: CIMMYT

ISBN: 968-6923-08-X
ISSN: 0188-2465
AGROVOC descriptors: Wheats, biodiversity, germplasm, interspecific hybridization,
intergeneric hybridization, hybrids, triticales, plant 1beding.
AGRIS category codes: F30; A50.
Dewey decimal classification: 631.523.

Designers: Miguel Mellado E. and Eliot Sanchez P
Printed in Thailand.


v Illustrations/Photographs

vii Color Plates

viii Tables

x Acknowledgments

xi Abbreviations/ Acronyms

xii Preface

xiv Contributors, Reviewers, and Editors

xv Executive Summary

xx Resumen

1 1 Introduction: 15 Years of Progress in Wheat Wide Crosses at CIMMYT
Abdul Mujeeb-Kazi

5 2 Perennial and Annual Wheat Relatives in the Triticeae
Abdul Mujeeb-Kazi and Richard R.-C. Wang

14 3 Interspecific Crosses: Hybrid Production and Utilization
Abdul Mujeeb-Kazi

22 4 Intergeneric Crosses: Hybrid Production and Utilization
Abdul Mujeeb-Kazi

47 5 Production of Polyhaploid Wheat Plants Using Maize and Tripsacum
Abdul Mujeeb-Kazi, Oscar Riera-Lizarazu, and Manilal D.H.M. William

66 6 Applications of Tissue Culture in Wheat Wide Crosses
Abdul Mujeeb-Kazi, Nitschka ter Kuile, Reagan Waskom, and Murray W. Nabors

76 7 Applications of Biochemical Markers in Wheat Wide Crosses
Manilal D.H.M. William and Abdul Mujeeb-Kazi

93 8 Applications of Molecular Markers in Wheat Wide Crosses
Manilal D.H.M. William and Abdul Mujeeb-Kazi

102 9 Conclusions and a Look to the Future
Abdul Mujeeb-Kazi

107 References

119 Appendix 1- Durum wheat cultivars or lines used by CIMMYT in crosses with Triticum
tauschii ( 1. 1,3,-/ squarrosa), T. monococcum, T. boeoticum, and T. uratu to develop
D genome synthetics and A genome hexaploids.

120 Appendix 2- D genome synthetic hexaploids (2n=6x=42; AABBDD) produced at CIMMYT by
crossing durum x Triticum tauschii.

126 Appendix 3- A genome hexaploids (2n=6x=42; AAAABB) produced at CIMMYT by crossing
durum x various accessions of A genome diploid species (Triticum monococcum,
T. boeoticum, T. urartu).

128 Appendix 4- Partial or complete cross combinations of Triticum aestivum and T. ',. ;,1,,. with
some perennial Triticeae.

129 Appendix 5- Amphiploids resulting from combinations of Triticum aestivum and T. r. ;.1,...
with some Triticeae species and their expected chromosome status.

130 Appendix 6- Some intergeneric and trigeneric hybrids produced from crosses between
Triticeae species and maintained at CIMMYT, El Batan, Mexico, under
greenhouse conditions.

133 Glossary


16 Figure 3.1. Schematic showing the production of synthetic hexaploids derived from
crossing Triticum I ., i.;,1. .i x T. tauschii and their utilization.
18 Figure 3.2. Somatic chromosome numbers of synthetic hexaploids from Triticum r. ;1,,l i x
T. tauschii: 2n=6x=42, hyperploid with 43 chromosomes, 42 with a telocentric,
and hypoploid with 41 chromosomes.
19 Figure 3.3. Schematic demonstrating alien transfers from Triticum tauschii (2n=2x=14, DD) to
elite T. aestivum cultivars via direct crossing and backcrossing.
21 Figure 3.4. Schematic showing extraction of the AABB component from an elite hexaploid
wheat cultivar, derivation of a synthetic hexaploid by crossing it to Triticum
tauschii (2n=2x=14, DD), and utilization of the doubled derivative (2n=6x=42,
32 Figure 4.1. Somatic details at metaphase for Thinopyrum dijliigati /Secale cereale F1
(2n=3x=21), and Th. Ollwilga i1/2* S. cereale backcross I with 49 chromosomes
34 Figure 4.2. Mitotic and meiotic chromosomal details of some backcross I derivatives of
Triticum aestivum/Thinopyrum ... ,7.' .... f. aestivum shown in somatic cells and
35 Figure 4.3. A hypoploid backcross II derivative from Triticum aestivum/Thinopyrum scirpeum
//2* T aestivum shown in 48 somatic chromosomes, 7 univalents + 4 rod bivalents
+ 15 ring bivalents + 1 trivalent at metaphase I, and 8 univalents + 3 rod
bivalents + 17 ring bivalents at metaphase I.
36 Figure 4.4. Spike showing a C-banded chromosome M of Leymus racemosus that
spontaneously substitutes for chromosome 6A6A or 6D6D of T. aestivum and
produces well developed grains.
38 Figure 4.5. Aneuploidy in amphiploid advance of Triticum ',.1 .1,.... L. cv. Laru/A,.;:1,.!.
variabilis: 55 chromosomes + 1 telocentric, 56 chromosomes + 1 telocentric, 57
chromosomes, and 56 chromosomes.
42 Figure 4.6. Schematic showing genetic manipulation option using the phlb locus of Triticum
aestivum cv. Chinese Spring by direct crossing or by the maize-mediated
polyhaploid system.
54 Figure 5.1. Spikes of wheat polyhaploids (n=3x=21, ABD).
55 Figure 5.2. Mitotic metaphase spread of Triticum aestivum L. (2n=6x=42, AABBDD) and
mitotic spread of a T aestivum L. polyhaploid (n=3x=21, ABD).
55 Figure 5.3. Mitotic spread of Triticum ,:. .'1, l. cv."Altar 84" (2n=4x=28, AABB); mitotic
spread of a T. r.,. .; 1.... cv. "Altar 84" polyhaploid (n=2x=14, AB); a polyhaploid
of Triticum aestivum with 20 chromosomes; and a Triticum aestivum polyhaploid
with 21 chromosomes including a telocentric.
57 Figure 5.4. Polyhaploid chromosome configurations of Triticum aestivum L. at Metaphase I
of meiosis showing variable univalents and bivalents.

58 Figure 5.5. SDS-PAGE separation of seed proteins from durum and bread wheat cultivars
and their extracted doubled haploids.
58 Figure 5.6. Grain esterase profiles of durum and bread wheat cultivars and their extracted
doubled haploids.
59 Figure 5.7. f-Amylase profiles of durum and bread wheat cultivars and their extracted
doubled haploids.
62 Figure 5.8. Somatic and meiotic cytology of a n=3x=21, ABD polyhaploid derived from a
hexaploid wheat x Tripsacum dactyloides cross.
68 Figure 6.1. Schematic showing callus induction, transfer, and regeneration protocol from
crosses of Triticum spp. x 1.. ;,..1 variabilis.
69 Figure 6.2. N-banded chromosome 5B from Triticum aestivum L. cv. Chinese Spring showing
consistency of banding sites on both arms despite stage of contraction of the
chromosomes and their extraction for the figure from different cells and root
72 Figure 6.3. Embryogenic callus formation in Triticum aestivum cultures; Early regeneration in
T. aestivum cultures; Advanced regeneration in T aestivum; A spontaneously
doubled (2n=8x=56) T. h, ..,.... tic cell.
73 Figure 6.4. Spike morphology variations in R-0 plants (callus regenerated) of Triticum
74 Figure 6.5. Number of Triticum aestivum regenerated plants and their somatic chromosome
numbers; Number of T. ',. ;1:....r regenerated plants and their somatic
chromosome number.
82 Figure 7.1. Separation of Beta-Amylase (f-AMY) isozymes on native-PAGE (8.,, gels) for
Thinopyrum bessarabicum, Chinese Spring, the amphiploid, and addition lines.
86 Figure 7.2. SDS-PAGE of high molecular weight glutenin subunits of Triticum tauschii
87 Figure 7.3. SDS-PAGE of high molecular weight glutenin subunits of synthetics with the
durum and Triticum tauschii parent combinations.
90 Figure 7.4. Grain glucose phosphate isomerase banding profiles of two different
combinations of parents and their F1 progeny.
91 Figure 7.5. Gliadin banding profiles of two parents and their F1 progeny.
91 Figure 7.6. Grain glucose phosphate isomerase (GPI) banding patterns of 1B,1B durums,
1BL/1RS, 1BL/1RS durums and the heterozygote 1B, 1BL/1RS durums on IEF
(pH 5-8.5).
100 Figure 8.1. Amplification of genomic DNA (20 ng) of Triticum aestivum cv. Chinese Spring,
Thinopyrum bessarabicum, and the amphiploid of CS/Th. bessarabicum.
101 Figure 8.2. Amplification of genomic DNA of Triticum aestivum cv. Chinese Spring (20 ng)
using the primers N 1 and N 3.


137 Plate 1.

138 Plate 2.

139 Plate 3.

140 Plate 4.

Detection of Thinopyrum bessarabicum chromosomes in the amphiploid of Triticum
aestivum cv. Chinese Spring/Th. bessarabicum (2n=8x=56, AABBDDJJ) using genomic
in situ hybridization.
a) A fluorescence micrograph of X Triticosecale (2n=6x=42, AABBRR) with filter 9.
b) le chromosome 1R disomic addition line to T. aestivum cv. Chinese Spring
(2n=6x=42+lR1R). c) A translocated 1BL/1RS wheat cultivar with 42 chromosomes in
which the two rye segments (1RS) are visible. d) The same cell as in c) viewed under
filter 23 showing the diagnostic for 1RS segments.
a) A 1BL/1RS homozygous substitution in a durum wheat (2n=4x=28) showing the
bright yellow rye arm (1RS) when rye DNA was used as a probe and unlabeled
wheat DNA was used for blocking with filter 9. b) An interphase mitotic cell of a
homozygous 1AL/1RS segment using filter 23. c) A homozygous 1AL/1RS
T. aestivuncultivar with the 1RS rye segment detected with filter 2. d) 1AL/1RS
metaphase cell with 42 chromosomes.
a) A homozygous double translocation in wheat involving rye segments 5RL and 1RS
onto wheat chromosomes 5AS and 1BL, respectively, detected with filter 9 and rye
DNA as a probe. b) A homozygous 6BS/6RL translocation wheat with 42
chromosomes showing the two rye (6RL) segments as detected under filter 9 with rye
DNA as a probe. c) Double labeling on an aneuploid trigeneric hybrid of T. aestivunr
Th. I. '.. .iii... '. cereale with 40 chromosomes. Wheat DNA was unlabeled;
Th. bessarabicunand S. cereale were labeled with biotin and digoxigenin, respectively.
d) A homozygous rye DNA insert associated with the leaf rust Lr25 gene.


7 Table 2.1. The genomes and ploidy levels of the perennial genera of the tribe Triticeae.
9 Table 2.2. A synonym list of annual ,, .1 ;,. iTriticum species.
10 Table 2.3. Genome designations of the diploid Triticum/IA 1,. .. I -p~ cki- according to
various workers.
11 Table 2.4. Genome designations of polyploid Triticum i, 1../. species.
13 Table 2.5. The genomes of Triticum ,l. ; 1.37, and their variability for alien gene transfer.
20 Table 3.1. Five synthetic hexaploids selected as resistant to Helminthosporium sativum
compared with their durum wheat parents.
20 Table 3.2. Five synthetic hexaploids selected as resistant to Fusarium graminearum
compared with their durum wheat parents.
20 Table 3.3. Five synthetic hexaploids that have tested positive for the Na:K discrimination
trait associated with salinity tolerance in hydroculture testing compared with
their durum wheat parents.
21 Table 3.4. Five synthetic hexaploids selected as resistant to Karnal bunt under the
greenhouse screening test.
31 Table 4.1. Some backcross derivatives (BCI) from Triticum aestivum /Thinopyrum ,louigatl:
(2n=10x=70)/ /T. aestivum with aneuploid chromosome numbers and some
N-banded chromosome compositions.
32 Table 4.2. Apomictic progeny in BCI derivatives of Hordeum. .,,... Triticum 'i.,. ;1,...
and H. ,,.. 1 T. aestivum with cytological details.
39 Table 4.3. Triticeae relatives with a promise for salinity tolerance based upon literature
reports and collaborative research findings.
40 Table 4.4. Dry weight (g) and Na and K cell sap values (mol/m3) in some wheat
cultivars and their alien derivatives under hyhdculture at 50 mol/m3, then at
200 mol/nA NaC1.
41 Table 4.5. Thinopyrum bessarabicum (2n=2x=14, JJ) disomic addition lines identified on the
basis of cytological, morphological, and biochemical markers and tentatively
assigned to the seven homoeologous groups of Triticum aestivum L.
41 Table 4.6. Salinity hydroculture screening of some promising wheat /Thinopyrum
bessarabicum addition lines with 44 chromosomes and multiple disomics at
150 mol/nNaCl measured after 50 days of full stress.
44 Table 4.7. Triticeae germplasm screened for aluminum tolerance under laboratory
conditions in hydroculture with tolerance response )1.
50 Table 5.1. Number of florets pollinated, embryos rescued, and wheat polyhaploid plants
regenerated from wheat x maize crosses using three 2,4-D treatment techniques.
51 Table 5.2. Wheat polyhaploid embryo recovery frequencies i from Triticum aestivum L.
x Zea mays L. crosses under three 2,4-D treatment techniques.
53 Table 5.3. Embryos produced, recovery percentage, plant regeneration, and colchicine-
induced doubling frequencies of Triticum aestivum L., T. '., ;',.... L., and
T. ',.T;,l.1, x T tauschii lines following crosses with Zea mays.

56 Table 5.4. Mean chromosome pairing with ranges in parentheses at metaphase I in some
polyhaploids of Triticum aestivum L. and T. ',,. .:1, ... x T. tauschii synthetic
60 Table 5.5. Embryo recovery and plant regeneration from hybridization of some synthetic
hexaploids (Triticum '1,. ;,1r, .. x T. tauschii) and T. aestivum and T. 1.1 ;,...
cultivars with Tripsacum dactyloides.
61 Table 5.6. Embryo recovery frequencies in crosses between Triticum aestivum cv. Ciano T 79
and T. r;'1,1i.... cv Altar 84 with Tripsacum dactyloides after various treatments.
63 Table 5.7. Mean chromosome pairing with ranges in parentheses at metaphase I in some
polyhaploids of Triticum aestivum L. and T. ',. .;1,.... x T. tauschii synthetic
63 Table 5.8. Spontaneous doubling in polyhaploids of T. 'r,. :; 1,,i. cv. Ruff'S' x T tauschii;
spikes, seed number, and somatic chromosome counts of three seeds per
doubled plant.
64 Table 5.9. Polyhaploid embryo production of three Fl DNA polymorphic crosses
between Triticum aestivum (cvs. Buckbuck, Opata M 85, Ciano T 79) and a
synthetic hexaploid (T. ', i i1,,. / T. tauschii) using the maize polyhaploid
induction system.
65 Table 5.10. Disomic Thinopyrum liigatliim additions to wheat variety Goshawk'S'
developed by doubled haploidy and identified by isozyme analysis for
67 Table 6.1. Data from regenerated Triticum aestivum and T. 'r,. ;,1, plants from callus
maintained up to seven months.
69 Table 6.2. Partial data for plants regenerated from Triticum aestivum x 1, A ;37 variabilis
callus during different passages (T).
70 Table 6.3. Meiotic associations in hybrids of Chinese Spring (CS) Ph x .;1, I. variabilis
(13E) with low pairing; of CS phlb x Ae. variabilis (13E) with high pairing; in a CS
Ph x Ae. variabilis (13E) callus-regenerated F with modified increased pairing.
71 Table 6.4. Regenerated plants of Fl hybrid with Ae. variabilis (13E) of Triticum aestivum and
T. *'1. ;1:.... ,r -l. i n, cytologically doubled progeny.
74 Table 6.5. Some Triticum aestivum x alien species combinations where embryos were
excised but could not be differentiated into hybrid plants.
78 Table 7.1. Some isozyme markers identified with isoelectric focusing for alien species with
potential of being applied to wide hybridization in the Triticeae.
85 Table 7.2. Biochemical markers that have been identified at CIMMYT using Chinese
Spring/ Imperial rye addition lines.
97 Table 8.1. Germplasm used for FISH analyses with cytological and origin details.
97 Table 8.2. Specific details of the FISH mixture used for different germplasm comprising of
Triticum aestivum, T. '1, .;',1t ... Secale cereale, Thinopyrum bessarabicum,
TI: ..; .i. ... and T. tauschii.


This research report emanates from the
efforts of our CIMMYT-based support
staff that has expanded from only
Silverio RoldAn in 1979 to Victor Rosas, Silverio
Cano, Roman Delgado, Angel Ramirez, and
Alejandro Cortes. The project has also advanced
through inputs of post-doctoral fellows,
associate scientists, and trainees, who brought in
different expertise. These included Harnek
Sandhu, Lesley Sitch, Robert Asiedu, Manilal
William, Masanori Inagaki, Nitschka ter Kuile,
Oscar Riera-Lizarazu, Ali Moustafa Moussa,
Michel Bernard, Shafqat Farooq, Qamar Jahan,
and Bai Dapeng.

The program has been aided by budgetary
support from the Rockefeller Foundation, the
Australian Centre for International Agricultural
Research, the Overseas Development Agency,
the Government of Japan, and the United States
Agency for International Development. This
became the basis of several international
collaborative projects. To further our cause, we
recognize the inputs of several national
programs, professional visitors to the program,
external review teams, and the assistance of
several CIMMYT colleagues that facilitated our
forging ahead from basic/strategic research
aspects to applied goals.

The nine chapters of this report were sent to
external reviewers noted for their respective
expertise: Roland von Bothmer, John Carman,
Andre Comeau, Ardeshir Damania, Bob
Forsberg, Pat Heslop-Harrison, Howard Rines,
and J. Valkoun. For their comments and
suggestions, which enabled us to better present
complicated topics more clearly, we are highly
appreciative. The internal review by George
Varughese provided added insight and is also

Throughout this 15-year research effort, the
Wheat Program directing staff teams of Glenn
Anderson/Art Klatt, Byrd Curtis/Art Klatt, and
Tony Fischer/George Varughese are gratefully
recognized for their support and guidance.

However, the individual who truly made this
program possible is N.E. Borlaug, whose vision
jump-started the wide crosses research effort at
CIMMYT during the early 1970s, which led to
my support at Kansas State (1974-1979) and later
employment at CIMMYT. Norm has been an
ardent supporter of our wide cross efforts and
his continuing research discussions up to the
present are fondly acknowledged.

Abdul Mujeeb-Kazi
El Batan, Mexico
May 30,1994


ADH-Alcohol dehydrogenase.
BC-Backcr oss.
CS-Chinese Spring.
DNA-Deoxyribonucleic acid.
EST-Seed esterase.
FISH-Fluorescent In situ hybridization.
GA3-Gibberellic acid.
GOT-Glutamate oxaloacetate transaminase.
GPI-Glucose phosphate isomerase.
HMW-High molecular weight.
IAA-Indole-3-acetic acid.
IEF-Isoelectric focusing.
ITMI-International Triticeae Mapping
KB-Karnal bunt.
LS-Linsmaeier and Skoog (1965) medium.
MDH-Malate dehydrogenase.

MS-Murashige and Skoog (1962) basal
MTT-3-(4,5 dimethyl-thiazol-2-yl)-2,5-
diphenyl-tetrazolium bromide.
NAD-B-nicotinamide adenine dinucleotide.
NADP-B-nicotinamide adenine dinucleotide
PAGE--Polyacrylamide gel electrophoresis.
PCR-Polymerase chain reaction.
pI-Isoelectric point.
RAPD-Randomly amplified polymorphic DNA
RFLP-Restriction fragment length
SDS-Sodium dodecyl sulfate.
SKDH-Shikimate dehydrogenase.
SOD-Superoxide dismutase.
TCCP-Tissue Culture for Crops Project.
2,4-D-2,4-dichlorophenoxyacetic acid.


The CIMMYT Research Report Series
documents specific CIMMYT research
efforts and is directed toward technical
audiences. Each publication synthesizes the
results of research that has usually occurred over
an extended period. This report, the second
installment in the series, relates the
accomplishments of 15 years of wheat wide cross
research at CIMMYT. This body of work has
been conducted and accumulated by Dr. Abdul
Mujeeb-Kazi, head of the Wheat Wide Crosses
Section, in conjunction with specialists, who
have spent time at CIMMYT or who have
collaborated in various ways through their
respective centers of excellence in developed and
developing countries.

The CIMMYT Wheat Wide Crosses Section
assists our cereal breeding programs in adding
new variability to the wheat gene pool by
making crosses between: 1) wheat and the
annual grasses within the genus Triticum
interspecificc) and 2) wheat and more distant
relatives in other genera of the Triticeae tribe
(intergeneric). CIMMYT has achieved an
international reputation for the production and
characterization of these two types of wide
hybrids through basic and applied research and
by significant scientific reporting in peer-
reviewed publications and at international
symposia. Although relying on established plant
breeding techniques, the Section makes use of
emerging technologies to better understand
genetic relationships and to introgress genetic

The research has shown that an alien grass
species can be screened for a specific resistance
or tolerance to a disease or abiotic stress and

then be hybridized with wheat to introgress the
desired trait. Alternatively an alien species can
be first hybridized to wheat and then the
derivative plants can be screened for the trait.
The key to either strategy is wide hybrid
production, which normally involves embryo
rescue and chemical treatment to double the
chromosome number of the resulting hybrid
plants. These plants are the critical base from
which alien genetic material can be utilized by
CIMMYT breeders and others in cereal
improvement programs.

This report details the outstanding
contributions-to-date of new knowledge and
techniques emanating from our wide crossing
efforts. These include unique approaches to
production of wheat polyhaploids involving
crosses with maize and Tripsacum, application of
tissue culture to demonstrate the potential of
inducing genetic variability or alien
introgressions, and the use of biochemical and
molecular markers to confirm the introgression
of alien chromosomes. As explained in the
report, we are using isozyme analyses, randomly
amplified polymorphic DNA sequences
(RAPDs), and in situ hybridization techniques to
detect the presence in wheat of inserted
chromosome segments and entire alien

At the onset of our wide crosses effort in 1979,
we did not anticipate producing on-the-farm
products as rapidly as we have. Recent releases
in Pakistan of salt- and drought-tolerant wheat
varieties, derived from CIMMYT's alien genetic
material, are products already in the hands of
farmers. Wheat lines-arising from our wide
crosses work to obtain resistance to spot blotch,

a devastating disease of wheat in warmer
areas-are being field-tested in a number of
countries and will soon be available to farmers.
These lines, together with accessions of goat
grass (Triticum tauschii) and advanced
derivatives of wheat with immune responses to
Karnal bunt disease, have been registered as
genetic stocks with the Crop Science Society of
America. Over the last four years, we have also
produced an impressive number (nearly 525) of

synthetic bread wheats (durum wheat x Triticum
tauschii), which represent another major
contribution toward increasing variability in the
wheat gene pool.

I believe that this Research Report is a major
document that provides relevant information in
a very challenging area of research and trust that
it will stimulate discussion among researchers in
the related disciplines.

Donald Winkelmann
Director General


Abdul Mujeeb-Kazi, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program,
CIMMYT, Mexico
Nitschka ter Kuile, CIMMYT Wide Crosses Section and Dept. of Biology, Colorado State University
Fort Collins, Colorado, USA
Murray W. Nabors, Dept. of Botany, Colorado State University, Fort Collins, Colorado, USA
Oscar Riera-Lizarazu, CIMMYT Wide Crosses Section and Dept. of Agronomy and Plant Genetics,
University of Minnesota, St. Paul, Minnesota, USA
Richard R.-C. Wang, USDA-ARS, Forage and Range Research, Utah State University, Logan, Utah, USA
Reagan Waskom, Dept. of Agronomy Colorado State University, Fort Collins, Colorado, USA
Manilal D.H.M. William, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program, and
Applied Molecular Genetics Laboratory, CIMMYT, Mexico


Roland von Bothmer, Department of Plant Breeding Research, The Swedish University of Agricultural
Sciences, Svalov, Sweden (Chapters 2 and 9)
John G. Carman, Department of Plants, Soils, and Biometeorology, Utah State University, Logan, Utah,
USA (Chapter 6)
Andre Comeau, Agricultural Canada, Ste. Foy, Quebec, Canada (Chapter 4)
Ardeshir B. Damania, Genetic Resources Unit, International Center for Agricultural Research in the
Dry Areas (ICARDA), Aleppo, Syria (Chapter 3)
Robert A. Forsberg, College of Agricultural and Life Sciences, Department of Agronomy, University of
Wisconsin-Madison, Madison, Wisconsin, USA (Chapter 7)
J.S. (Pat) Heslop-Harrison, Department of Cell Biology, John Innes Centre, Norwich, England
(Chapters 1, 8, and 9)
Howard Rines, Midwest Area Plant Science Research Unit of the United States Department of
Agriculture, Agricultural Research Service, St. Paul, Minnesota, USA (Chapter 5)
Jan Valkoun, Genetic Resources Unit, International Center for Agricultural Research in the Dry Areas
(ICARDA), Aleppo, Syria (Chapter 3)


Abdul Mujeeb-Kazi, Head, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program,
CIMMYT, Mexico
Gene P. Hettel, Science Writer/Editor, Information Services, CIMMYT, Mexico


Abdul Mujeeb-Kazi, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program,
CIMMYT, Mexico
Nitschka ter Kuile, CIMMYT Wide Crosses Section and Dept. of Biology, Colorado State University
Fort Collins, Colorado, USA
Murray W. Nabors, Dept. of Botany, Colorado State University, Fort Collins, Colorado, USA
Oscar Riera-Lizarazu, CIMMYT Wide Crosses Section and Dept. of Agronomy and Plant Genetics,
University of Minnesota, St. Paul, Minnesota, USA
Richard R.-C. Wang, USDA-ARS, Forage and Range Research, Utah State University, Logan, Utah, USA
Reagan Waskom, Dept. of Agronomy Colorado State University, Fort Collins, Colorado, USA
Manilal D.H.M. William, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program, and
Applied Molecular Genetics Laboratory, CIMMYT, Mexico


Roland von Bothmer, Department of Plant Breeding Research, The Swedish University of Agricultural
Sciences, Svalov, Sweden (Chapters 2 and 9)
John G. Carman, Department of Plants, Soils, and Biometeorology, Utah State University, Logan, Utah,
USA (Chapter 6)
Andre Comeau, Agricultural Canada, Ste. Foy, Quebec, Canada (Chapter 4)
Ardeshir B. Damania, Genetic Resources Unit, International Center for Agricultural Research in the
Dry Areas (ICARDA), Aleppo, Syria (Chapter 3)
Robert A. Forsberg, College of Agricultural and Life Sciences, Department of Agronomy, University of
Wisconsin-Madison, Madison, Wisconsin, USA (Chapter 7)
J.S. (Pat) Heslop-Harrison, Department of Cell Biology, John Innes Centre, Norwich, England
(Chapters 1, 8, and 9)
Howard Rines, Midwest Area Plant Science Research Unit of the United States Department of
Agriculture, Agricultural Research Service, St. Paul, Minnesota, USA (Chapter 5)
Jan Valkoun, Genetic Resources Unit, International Center for Agricultural Research in the Dry Areas
(ICARDA), Aleppo, Syria (Chapter 3)


Abdul Mujeeb-Kazi, Head, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program,
CIMMYT, Mexico
Gene P. Hettel, Science Writer/Editor, Information Services, CIMMYT, Mexico


Abdul Mujeeb-Kazi, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program,
CIMMYT, Mexico
Nitschka ter Kuile, CIMMYT Wide Crosses Section and Dept. of Biology, Colorado State University
Fort Collins, Colorado, USA
Murray W. Nabors, Dept. of Botany, Colorado State University, Fort Collins, Colorado, USA
Oscar Riera-Lizarazu, CIMMYT Wide Crosses Section and Dept. of Agronomy and Plant Genetics,
University of Minnesota, St. Paul, Minnesota, USA
Richard R.-C. Wang, USDA-ARS, Forage and Range Research, Utah State University, Logan, Utah, USA
Reagan Waskom, Dept. of Agronomy Colorado State University, Fort Collins, Colorado, USA
Manilal D.H.M. William, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program, and
Applied Molecular Genetics Laboratory, CIMMYT, Mexico


Roland von Bothmer, Department of Plant Breeding Research, The Swedish University of Agricultural
Sciences, Svalov, Sweden (Chapters 2 and 9)
John G. Carman, Department of Plants, Soils, and Biometeorology, Utah State University, Logan, Utah,
USA (Chapter 6)
Andre Comeau, Agricultural Canada, Ste. Foy, Quebec, Canada (Chapter 4)
Ardeshir B. Damania, Genetic Resources Unit, International Center for Agricultural Research in the
Dry Areas (ICARDA), Aleppo, Syria (Chapter 3)
Robert A. Forsberg, College of Agricultural and Life Sciences, Department of Agronomy, University of
Wisconsin-Madison, Madison, Wisconsin, USA (Chapter 7)
J.S. (Pat) Heslop-Harrison, Department of Cell Biology, John Innes Centre, Norwich, England
(Chapters 1, 8, and 9)
Howard Rines, Midwest Area Plant Science Research Unit of the United States Department of
Agriculture, Agricultural Research Service, St. Paul, Minnesota, USA (Chapter 5)
Jan Valkoun, Genetic Resources Unit, International Center for Agricultural Research in the Dry Areas
(ICARDA), Aleppo, Syria (Chapter 3)


Abdul Mujeeb-Kazi, Head, Wide Crosses Section, Genetic Resources Subprogram, Wheat Program,
CIMMYT, Mexico
Gene P. Hettel, Science Writer/Editor, Information Services, CIMMYT, Mexico


their effort to meet the increasing
worldwide demand for food, plant breeders
are finding less and less appropriate
germplasm with desired traits among cultivated
crops themselves with which to make needed
improvements. Fortunately, useful genetic
resources (i.e., important traits for use in crop
improvement) are being found among
uncultivated plants in the wild. As stated in
Chapter 1, the challenge is to be able to
incorporate this "new" germplasm routinely into
existing food crops through a technique called
wide crossing.

Wide Crosses

Most efforts to transfer alien germplasm from
wild plants into cultivated crops have involved
the Triticum grass species-with the greatest
emphasis being placed on improving bread
wheat (T. aestivum L.). Over the past 15 years,
CIMMYT has been a part of this endeavor as it
has vigorously pursued bread wheat
improvement not only with interspecific
hybridization (crosses made among annual
grasses within the Triticum IA ;7.'/. group), but
also with intergeneric hybridization (crosses
with wheat using some of the 250 perennial
grasses in the Triticeae tribe). The perennials are
important because their natural habitats provide
the possibility that they could be potent sources
of resistance for several biotic and abiotic
stresses. Chapter 2 sets the stage for our wide
cross efforts by describing the complex genome
make-up of the perennials and annuals in the
Triticeae. As described in Chapter 2, some
perennial species have recently undergone
taxonomic readjustments based on new
knowledge about their genomes.

Interspecific hybridization (Chapter 3)
When utilizing the varied gene pools within
other Triticum species for improving bread wheat,
a high priority for breeders is to utilize the many
alien accessions among these species that have
genomes similar to the A, B, or D genomes of
bread wheat. This not only makes accomplishing
alien gene introgression easier because of
genomic similarities, but it is also compatible
with field research and sets the stage for a high
success rate in accomplishing multiple-gene
polygenicc) transfers. Thus, our interspecific
work for bread wheat improvement focuses on
using these closely related genomes.

CIMMYT has been concentrating on exploiting
accessions of a wild relative of wheat called goat
grass (Triticum tauschii syn. A .. 7, squarrosa).
We believe this wild relative's diversity and
distribution across Eurasia provide a unique
opportunity for exploiting novel genetic
variability for bread wheat improvement. In
addition, T. tauschii, unequivocally accepted as
being the grass that contributed the D genome to
bread wheat, is attributed to have a wide range of
resistances and tolerances to diseases and abiotic

As described in Chapter 3, the best way to exploit
T. tauschii variability is to first reliably screen the
accessions for desired resistances or tolerances.
The selected accessions can then be crossed
directly with bread wheat (T. aestivum)-if a
program has embryo rescue and chromosome
doubling capabilities-or crossed with durum
wheat (T. I,.1 ;',,....--t,. produce what are called
synthetic hexaploids, which can in turn be easily
crossed with bread wheat by any conventional
breeding program. We have made successful

crosses between Karnal bunt (Tilletia indica)-
susceptible bread wheats and several KB-
resistant T. tauschii accessions. We have
identified synthetics that are resistant to spot
blotch (Helminthosporium sativum), Septoria tritici,
and scab (Fusarium graminearum), which can then
be crossed to bread wheat. Several synthetics
have shown tolerance to salt stress in initial field
screening at La Paz, Baja California Su; Mexico.
To date, we have produced nearly 525 synthetic
hexaploids-most involving a unique T. tauschii
accession (see Appendix 2)-for use in crosses
with bread wheat.

Intergeneric hybridization (Chapter 4)
The different gene pools within the annual and
perennial species of the Triticeae can provide
tremendous genetic variability for wheat
improvement. However, in contrast to the
Triticum /A,. ;. ,.. group, the species we deal
with in our intergeneric crosses are quite diverse
genomically and rather difficult to cross with
wheat. Even when successfully combined, the
resulting hybrids exhibit little or no
intergenomic chromosome association. Despite
these limitations, significant successes and
advancements have been made by centers of
excellence over the past 20 years. CIMMYT's
principal objectives in intergeneric crosses have
been to obtain tolerances to toxic levels of
aluminum and salt; copper uptake efficiency;
and resistances to H. sativum, F. graminearum,
and T. indica. We also anticipate the eventual
transfer of other traits such as resistance to
barley yellow dwarf virus (BYDV), S. tritici, and
Russian wheat aphid (Diuraphis noxia).

Hybrids and, in most cases, amphiploids have
been produced in crosses between species of the
genera Hordeum, AJ. ,i ...i Elymus, Secale,
Taeniantherum, Eremopyrum, and Haynaldia. The
range of new hybrids with more distantly related

species is constantly increasing and it is expected
that a greater range of genotypes will become
available for introgressing novel genetic
variability into wheat.

Search for New Applications
and Knowledge

Although the CIMMYT Wheat Wide Crosses
Section's primary thrust has been to assist
CIMMYT breeders in adding new variability to
the wheat gene pool through interspecific and
intergeneric crosses, it has also contributed a
significant body of knowledge on new
techniques and applications in plant

Polyhaploid production (Chapter 5)
Over the last four years, we have been able to
produce high frequencies of polyhaploid wheat
plants using either maize or Tripsacum pollen.
Polyhaploid plants are important in our efforts
to reduce the number of generations it takes to
fix the homozygosity of wheat and other cereal
plants. Homozygosity is required in basic
research projects, such as our collaborative work
with Cornell University to produce RFLP maps
of the wheat and barley genomes.

CIMMYT has obtained a high recovery of
polyhaploid wheat plants from crosses between
the wheat cultivar 'Morocco' and CIMMYT
maize population 'Pool 9A'. The taxonomic
proximity of Tripsacum dactyloides (L.) to maize
has encouraged us to evaluate cross
combinations involving Tripsacum and wheat (T.
aestivumnd T. ;nd T. I,r and T. ',, n rn .... x T.
tauschii amphiploids. We felt that Tripsacum
could serve as a novel and alternate sexual route
for the production of cereal polyhaploids and,
indeed, wheat x Tripsacum crosses have resulted
in the production of polyhaploid wheat plants of
various genotypes.

Unlike wheat anther culture or sexual
hybridization of wheat with H. bulbosum,
genotypic specificity and aneuploidy are absent
in maize- and Tripsacum-mediated polyhaploid
production, which makes them both superior
systems. The potential of stored maize and
Tripsacum pollen needs to be explored because it
could be a significant factor in extending the use
of the methodology to countries where cropping
cycles are separated or where adequate facilities
are lacking for growing plants under controlled

Tissue culture (Chapter 6)
Tissue culture applications have been essential to
the production of complex hybrids within the
Triticeae. These techniques will presumably even
widen the existing range of hybridization
possibilities. CIMMYT has exploited long-term
callus culture and regeneration to demonstrate
the potential of inducing variability within
various groups of the Triticeae for morphological,
biochemical, and cytological characteristics. Two
operational constraints in intergeneric
hybridization are associated with alien gene
introgression and amphiploid induction. Callus
culture methodology has significantly helped us
to overcome these constraints by enhancing
chromosome pairing analogous to that
characteristic of the Ph locus on chromosome 5B
and by inducing amphiploidy in two intergeneric
hybrid combinations.

Callus culture also: 1) provides advantages in
inducing variability in euploid wheat cultivars, 2)
facilitates in vitro screening for stress- or toxin-
producing pathogens, and 3) furnishes the
capacity to alter chromosomes structurally. In
addition, callus culture might be used to modify
recombination frequencies in otherwise low-

pairing complex hybrids as well as facilitate
recovery of hybrid derivatives with double the
number of chromosomes.

Confirming alien introgressions
(Chapters 7 and 8)
Although in some cases we have been using a
radical methodology that purposely de-
emphasizes the confirmation of alien gene
introgressions and chromosomal interchanges,
the Wide Crosses Section has been somewhat
involved in looking at biochemical and
molecular markers as means to accomplish such
confirmation when it is feasible and/or desirable
to do so. Initial identification and
characterization of alien introgressions can be
done by using relatively inexpensive, less
complicated cytological techniques and
biochemical markers. Once the material is
characterized, molecular markers could be
established and subsequently used to detect the
presence of minute and harder to detect
chromosomal interchanges.

Biochemical markers-These marker
techniques, which utilize isozymes and seed
storage proteins, are applicable in distinguishing
alien chromosomes in wheat for both
intergeneric and interspecific hybridizations.
More than 100 structural genes for isozyme
markers have been identified and located on
different chromosomal segments in wheat. The
major advantage to using isozymes is the speed
with which material can be screened because
there is adequate polymorphism. Information
regarding the homoeology of the alien
chromosomes in the addition lines can be
ascertained by identifying the genes they possess
that are orthologous to sets of T. aestivum genes
of which the chromosomal locations are known.

This can also be done by studying the ability of
the alien chromosomes to substitute for and pair
with specific wheat chromosomes.

When biochemical markers are first identified, it
is necessary to study the banding profiles for a
particular enzyme system in the two parental
species. When using these markers, analysis of
the two parental species and the amphiploid is
important in characterizing the alien genetic
material. Sometimes the alien species may show
a certain degree of polymorphism for a particular
enzyme system that results from allelic
differences and/or the mutually incompatible
nature of the alien species. We can identify
markers when the two parental species show
remarkably different banding profiles.

Molecular markers-Molecular markers are
becoming increasingly important in detecting
alien introgressions and chromosomal
interchanges-especially those involving small
segments of alien chromatin in wheat
backgrounds. The molecular techniques that
CIMMYT uses in the wheat wide crosses
laboratory currently include in situ hybridization
and Randomly Amplified Polymorphic DNA
sequences (RAPDs), which are based on
polymerase chain reactions (PCRs). Wide
adaptability of in situ hybridization procedures
was made possible by the development of
nonradioactive labeling techniques.

Since it is an important source of salinity
tolerance, CIMMYT has used Thinopyrum
bessarabicum to produce disomic addition lines.
Its combination with bread wheat has allowed us
to characterize quite a few biochemical markers.
With the objective to develop molecular markers
for tracking Th. bessarabicum chromatin in wheat

backgrounds and subsequently detect subtle
introgressions, we have conducted genomic in
situ hybridizations using the amphiploids of T.
aestivum cv. Chinese Spring (CS) x Th.
bessarabicum. More recently, we have begun
successfully exploiting the fluorescent in situ
hybridization technique (FISH) to detect alien


We believe that our radical approach of
advancing wide hybrids is serving as an
important rapid mechanism to get needed
germplasm to CIMMYT breeders. With this
strategy of leaving scientific questions
unanswered-at least for the time being-we
have been able to distribute germplasm with
needed attributes that had not been obtained by
breeders in their conventional programs. To a
considerable extent, slower basic research will
address questions that we leave unanswered.
However, over the last four or five years, we
have embarked on a more meticulous
methodology with our intergeneric
hybridizations that may allow for more basic
research later.

The structure of the CIMMYT Wide Crosses
Section is designed to link plant-level
manipulation with cellular and molecular
approaches-two aspects that are essential to the
program's function and effectiveness. We
anticipate that a number of very desirable
approaches will subsequently emerge to aid
cereal crop improvement. When these research
breakthroughs are refined and made applicable,
they will find complementary use in wheat
improvement and may even have the potential
to replace several conventional stages of genetic

We anticipate that the successful use of wheat
polyhaploids will receive greater application in
our program as well as in the breeding/
molecular areas. Viable stored pollen may
provide an additional boost to the application of
the wheat x maize or wheat x Tripsacum
techniques for producing polyhaploids. We also
expect to obtain wheat polyhaploids-as well as
desirable diversification-from sexual crosses
with sorghum. And we may be able to extend
the procedure to range grasses where analysis of

the resulting polyhaploids could help clarify
some complicated genomic relationships in the

In our 15 years of investigations, we have
progressed to a stage that allows us to project a
prosperous future. Historically, wide crosses at
CIMMYT were not anticipated to yield on-the-
farm products in a short time frame nor to
provide answers for each and every aspect of
development. However, we have had notable
achievements in these areas. So, we are
optimistic about additional achievements over
the next five years.


En el esfuerzo por satisfacer la creciente
demand de alimentos, los
fitomejoradores encuentran que entire las
species cultivadas hay cada vez menos
germoplasma adecuado con las caracteristicas
que requieren para mejorar los cultivos.
Afortunadamente, en la actualidad se estan
encontrando recursos geneticos (es decir,
caracteristicas tiles en el fitomejoramiento)
entire las plants no cultivadas. Como se
menciona en el Capitulo 1, el reto es elaborar,
mediante una tecnica denominada cruzas
amplias, un procedimiento sistematico que
permit incorporar este germoplasma "nuevo"
en los cultivos alimentarios existentes.

Cruzas amplias
Las species del genero Triticum han formado
parte de la mayoria de los trabajos orientados a
transferir genes de otras species,
particularmente, al trigo harinero (T. aestivum L.).
En los 61timos 15 afos, el CIMMYT ha
participado en ese esfuerzo, tratando
vigorosamente de mejorar el trigo harinero no
s6lo mediante la hibridizaci6n interespecifica
(cruzas entire gramineas anuales del grupo
Triticumln/A, ..7. ), sino tambien la intergenerica
(cruzas con trigo en las que han participado
algunas de las 250 gramineas perennes de la
tribu Triticeae). Las gramineas perennes son de
critica importancia debido a que, gracias a los
habitats donde se originaron, podrian ser fuentes
de fuerte resistencia a various factors bi6ticos y
abi6ticos adversos. El Capitulo 2 establece el
context de nuestro trabajo con cruzas amplias al
describir la compleja configuraci6n gen6mica de
las species perennes y anuales de las Triticeae.
Como se indica en ese capitulo, la clasificaci6n
taxon6mica de algunas de las species perennes

se ha reajustado con base en nuevos
conocimientos de su configuraci6n gen6mica.

Hibridizaci6n interespecifica
(Capitulo 3)
Cuando los mejoradores utilizan los variados
complejos geneticos de otras species Triticum
para mejorar el trigo harinero, una de sus
prioridades es utilizar en cruzas interespecificas
las innumerables accesiones extrafas cuyos
genomios son muy semejantes a los genomios A,
B o D del trigo harinero. Esto no s61o facility la
transferencia de genes extrafos debido a las
similitudes gen6micas, sino que tambien result
compatible con la investigaci6n en campo y,
ademas, permit alcanzar un alto grado de exito
en las transferencias de genes ml6tiples
(poligenicas). Asi pues, en los trabajos
interespecificos orientados al mejoramiento de
trigo harinero nos centramos en utilizar
genomios muy afines.

En el CIMMYT nos hemos concentrado en la
explotaci6n de las accesiones de Triticum tauschii,
sin. ;,./ -. squarrosa, pariente silvestle del trigo,
porque creemos que su gran diversidad y
distribuci6n en Europa y Asia. nos brindan una
oportunidad unica de utilizar variabilidad
genetica nueva en el mejoramiento del trigo. For
otra parte, se le atribuye a T. tauschii (que se
acepta sin lugar a dudas como la especie que
don6 el genomio D al trigo harinero) una gran
diversidad de resistencias y tolerancias a factors
adversos, tanto bi6ticos como abi6ticos, que
podrian contribuir a mejorar el trigo harinero.

Como se describe en el Capitulo 3, la tecnica
ideal para explotar la variabilidad de T. tauschii
en el mejoramiento del trigo harinero require
que se haga una selecci6n eficaz para encontrar

accesiones con distintas resistencias o tolerancias
deseables. Las accesiones seleccionadas entonces
se pueden cruzar directamente con trigo
harinero (T. aestivum) -siempre que el program
tenga la capacidad de efectuar rescate de
embriones y duplicar cromosomas-o con trigo
durum (T. i. .;.1,,i a fin de producer lo que
denominamos haploides sinteticos, que a su vez
pueden ser fAcilmente cruzados con trigo
harinero utilizando un program fitotecnico
conventional. Hemos efectuado cruzas entire
trigos harineros susceptibles al carb6n parcial
(Tilletia indica) y varias accesiones de T. tauschii
resistentes a ese hongo. Hemos identificado
trigos sinteticos que son resistentes al tiz6n foliar
(Helminthosporium sativum), Septoria tritici y rofia
de la espiga (Fusarium graminearum) y que
posteriormente se podran cruzar con trigo
harinero. Varios de ellos mostraron tolerancia a
la sal en la selecci6n inicial en campo realizada
en La Paz, Baja California Sur, M4xico. Hasta la
fecha, hemos producido casi 525 hexaploides
sinteticos, la mayoria de ellos a partir de una
accesi6n inica de T. tauschii (vease el Apendice
2), para usarlos en cruzas con trigo harinero.

Hibridizaci6n intergendrica
(Capitulo 4)
Los distintos complejos geneticos de las
gramineas anuales o perennes de la tribu
Triticeae pueden aportar una enorme
variabilidad genetica al mejoramiento de trigo.
Sin embargo, en contrast con lo que ocurre en el
grupo Triticum/Aegilops, las species que
participan en las cruzas intergenericas son
gen6micamente muy diversas y bastante dificiles
de cruzar con el trigo. Incluso cuando se logra
combinarlas, los hibridos que produce
muestran poca o ninguna asociaci6n
cromos6mica intergen6mica. A pesar de estas
limitaciones, algunos centros de excelencia han
logrado avances significativos en los l6timos 20
afos. Los principles objetivos del CIMMYT en

los trabajos intergenericos han sido obtener
tolerancia a niveles t6xicos de aluminio y sal;
captaci6n eficiente del cobre, y resistencia a H.
sativum, E graminearum y Tilletia indica.
Asimismo, a la larga esperamos poder transferir
otras caracteristicas como la resistencia al virus
del enanismo amarillo de la cebada (BYDV), a
Septoria tritici y al pulg6n ruso del trigo
(Diuraphis noxia).

Se han producido hibridos y, en la mayoria de
los casos, anfiploides entire species de los
generos Hordeum, A. i- ...i Elymus, Secale,
Taeniatherum, Eremopyrum y Haynaldia. Va en
aumento la cantidad de hibridos nuevos que son
product de cruzas con parientes mas lejanos y
se espera que pronto habra un ndmero mayor de
genotipos que podran utilizarse para introducir
variabilidad genetica nueva en el trigo.

La bisqueda de nuevos
conocimientos y aplicaciones

Aunque el trabajo primordial de la Secci6n de
Cruzas Amplias de Trigo del CIMMYT ha sido
ayudar a los mejoradores del mismo Centro
agregando nueva variabilidad al complejo
genetico del trigo mediante cruzas
interespecificas e intergenericas, tambien ha
contribuido un acervo considerable de nuevas
tecnicas y aplicaciones en el Area de la
biotecnologia vegetal.

Producci6n de polihaploides
(Capitulo 5)
En los l6timos cuatro afos, hemos podido lograr
altas frecuencias de plants polihaploides
utilizando polen de maiz o de Tripsacum. Las
plants polihaploides son tiles en nuestros
esfuerzos por reducir el ndmero de generaciones
necesarias para fijar la homocigocidad del trigo y
otros cereales. Se require homocigocidad en
proyectos de investigaci6n bAsica, como nuestro

trabajo conjunto con la Universidad de Cornell,
cuyo objeto es elaborar mapas de los RFLPs de
los genomios del trigo y de la cebada.

El CIMMYT ha logrado un alto grado de
recuperaci6n de trigos polihaploides en las
cruzas entire la variedad de trigo Morocco y el
Pool 9A, una poblaci6n de maiz del CIMMYT. La
proximidad taxon6mica de Tripsacum dactyloides
(L.) al maiz nos ha llevado a evaluar distintas
combinaciones de cruzas entire Tripsacum y el
trigo (T. aestivumy T. '1r, .; ;,.. .) y anfiploides de
T. '., ;', l.,. x T tauschii. Pensabamos que
Tripsacum podria proporcionar una forma sexual
novedosa de producer polihaploides de cereales
y asi ha sido, pues las cruzas entire el trigo y
Tripsacum han conducido a la producci6n de
trigos polihaploides de distintos genotipos.

A diferencia de lo que sucede en el cultivo de
anteras de trigo o la hibridizaci6n sexual del
trigo con H. bulbosum, no hay especificidad
genotipica ni aneuploidia en la producci6n de
polihaploides mediada por el maiz y el
Tripsacum, lo cual los hace sistemas superiores.
Es necesario explorer el potential del polen de
maiz y de Tripsacum que ha estado almacenado
porque podria ser un factor significativo para
extender el uso de la metodologia a paises donde
los ciclos de cultivo no son sucesivos o donde no
existen instalaciones adecuadas para el cultivo
de plants bajo condiciones controladas.

Cultivo de tejidos (Capitulo 6)
Las aplicaciones del cultivo de tejido han sido de
vital importancia para la producci6n de hibridos
complejos de las species Triticeae. Su
manipulaci6n presumiblemente ampliara las
hibridizaciones posibles actualmente. El
CIMMYT ha explotado el cultivo y regeneraci6n
de calls a largo plazo a fin de demostrar las
posibilidades de inducir variabilidad para
caracteristicas morfol6gicas, bioquimicas y

citol6gicas en various grupos de las species
Triticeae. Existen dos limitaciones operacionales
en la hibridizaci6n intergenerica que estan
relacionadas con la introgresi6n de genes ajenos
y la inducci6n de anfiploides. La metodologia
del cultivo de calls nos ha ayudado
grandemente a superar ambas limitaciones
gracias a que ha aumentado el apareamiento
cromos6mico analogo a aquel que es
caracteristico del locus Ph en el cromosoma 5B e
inducido anfiploidia en dos combinaciones
hibridas intergenericas.

Asimismo, el cultivo de calls: 1) nos
proporciona ventajas en la inducci6n de
variabilidad en variedades de trigo euploides, 2)
facility la selecci6n in vitro para factors de estres
o pat6genos que produce toxinas, y 3) nos da la
capacidad de alterar la estructura de los
cromosomas. Por otra parte, es possible que el
procedimiento modifique las frecuencias de
recombinaci6n en hibridos complejos que de otra
forma tienen bajas frecuencias de apareamiento
y facility tambien la recuperaci6n de derivados
hibridos con el double de cromosomas.

Confirmaci6n de la introgresi6n de
genes (Capitulos 7 y 8)
Si bien en algunos casos hemos utilizado
metodologias radicales que deliberadamente no
dan importancia a la confirmaci6n de la
introgresi6n de genes de otras species y los
intercambios cromos6micos, la Secci6n de
Cruzas Amplias hasta cierto punto ha estado
explorando los marcadores bioquimicos y
moleculares como medios de efectuar tal
confirmaci6n cuando result factible y/o
convenient hacerlo. La identificaci6n y
caracterizaci6n iniciales de introgresiones
extrafas pueden realizarse con tecnicas
citol6gicas y marcadores bioquimicos que son
menos complicados y relativamente baratos. Una
vez que se ha caracterizado el material, los

marcadores moleculares pueden establecerse y
despu6s usarse para detectar la presencia de
pequefifsimos intercambios de cromosomas con
otras species que son mas dificiles se detectar.

Los marcadores bioquimicos. Estos marcadores,
que utilizan isoenzimas y proteinas almacenadas
en la semilla que le permiten a 6sta germinar,
pueden usarse para distinguir cromosomas
extrafos en el trigo en hibridizaciones tanto
intergen6ricas como interespecificas. MAs de 100
genes estructurales para los marcadores
isoenzimaticos han sido identificados y
localizados en distintos segments
cromos6micos del trigo. La mayor ventaja de
utilizar los isoenzimas es la rapidez con la que el
material puede ser seleccionado debido a que
hay polimorfismo adecuado. La informaci6n
sobre la homoeologia de los cromosomas
extrafos en las lines de adici6n puede
confirmarse identificando los genes que poseen y
que son ort6logos a conjuntos de genes de T.
aestivum cuyas ubicaciones en los cromosomas se
conocen. Esto tambi6n puede hacerse estudiando
la capacidad de los cromosomas extranos de
remplazar cromosomas especificos del trigo.

Cuando los marcadores bioquimicos son
identificados por primera vez, es necesario
estudiar los perfiles de las bandas
correspondientes a un sistema enzimatico
determinado utilizando las dos species
progenitoras. El analisis de las dos species
progenitoras y del anfiploide es important
cuando se utilizan estos marcadores para
caracterizar materials gen6ticos extrafos. A
veces las species extrafias muestran cierto grado
de polimorfismo para un determinado sistema
enzimatico como resultado de diferencias
al6licas y/o porque no son compatibles entire si.
Podemos identificar marcadores cuando las dos
species progenitoras muestran perfiles de
bandas notablemente distintos.

Marcadores moleculares. Los marcadores
moleculares se estan volviendo cada vez mas
importantes en detectar introgresiones de genes
extraios e intercambios cromos6micos,
especialmente cuando se trata de pequefios
segments de cromatina extrafia en una
configuraci6n de trigo. Las t6cnicas moleculares
que actualmente se utilizan en el laboratorio de
cruzas amplias de trigo del CIMMYT incluyen la
hibridizaci6n in situ y las secuencias de ADN
polim6rfico amplificado al azar (RAPD) basadas
en las reacciones en cadena de polimerasa (PCR).
La amplia adaptabilidad de los procedimientos
de hibridizaci6n in situ es possible gracias a la
generaci6n de t6cnicas no radiactivas de

Debido a que constitute una fuente important
de tolerancia a la salinidad, Thinopyrum
bessarabicum se ha utilizado en el CIMMYT para
producer lines de adici6n dis6micas. Su
combinaci6n con trigo harinero nos ha permitido
caracterizar un gran nfmero de marcadores
bioquimicos. Con el objeto de desarrollar
marcadores moleculares que rastreen cromatina
de Th. bessarabicum en configuraciones de trigo y,
posteriormente, detectar introgresiones sutiles,
hemos realizado hibridizaciones gen6micas in
situ utilizando anfiploides de T. aestivum cv.
Chinese Spring x Th. bessarabicum. MAs
recientemente, hemos comenzado a emplear la
tecnica fluorescent de hibridizaci6n in situ
(FISH) para detectar ADN extrafo, con
excelentes resultados.


Estamos convencidos de que nuestro enfoque
radical para avanzar en la producci6n de
hibridos mediante cruzas amplias, es un
mecanismo rapido important que permit
entregar a los mejoradores del CIMMYT el
germoplasma que necesitan. Siguiendo nuestra

estrategia de dejar interrogantes cientificos sin
responder (al menos por el momento, hemos
podido distribuir germoplasma con
determinados atributos que los mejoradores no
habian podido obtener con los programs
fitot6cnicos tradicionales. Las investigaciones
bAsicas realizadas a un ritmo menos acelerado
responderan en gran media a los interrogantes
que hemos dejado pendientes. No obstante, cabe
mencionar que en los l6timos cuatro o cinco afos
hemos adoptado una metodologia mas
meticulosa para realizar nuestras hibridizaciones
intergen6ricas que quizA permit que
posteriormente efectuemos investigaciones mas

La secci6n de cruzas amplias de trigo del
CIMMYT esta estructurada de tal manera que
vincula la experimentaci6n con plants con
metodos moleculares y celulares -dos aspects
que son esenciales para el buen funcionamiento
y efectividad del program. Confiamos en que
mas adelante tambi6n surgiran m6todos muy
eficaces que ayudaran a mejorar los cereales.
Cuando esos avances cientificos se perfeccionen
y sean aplicables, tendran usos complementarios
en el mejoramiento de trigo y quizA hasta
lleguen a remplazar por complete algunas etapas
convencionales de la manipulaci6n gen6tica.

Creemos que se encontrarAn otras formas de
aplicar los trigos polihaploides tanto en nuestro
program como en las Areas fitot6cnica y
molecular. Es possible que el uso de polen que ha
estado almacenado y que sigue siendo viable d6
un impulse adicional a la aplicaci6n de las
tecnicas trigo x maiz o trigo x Tripsacum que
inducen polihaploidia. Esperamos producer
polihaploides de trigo-asi como lograr
diversificaci6n- a partir de cruzas sexuales con
sorgo. Tambi6n es possible que podamos utilizar
el procedimiento de los polihaploides con pastos
naturales y esclarecer mediante el analisis de los
polihaploides resultantes, las complejas
relaciones gen6micas entire las species Triticeae.

En los 15 afos que llevamos en la investigaci6n,
hemos progresado hasta un punto en que
podemos predecir un future pr6spero para
nuestra secci6n. Cabe mencionar que no se
esperaba que la secci6n de cruzas amplias del
CIMMYT fuera a general products para el
campo en un corto plazo ni proporcionar las
soluciones requeridas en cada aspect de su
generaci6n. No obstante, hemos logrado avances
notables en esas Areas y, por tanto, somos
optimistas respect a lo que lograremos en los
pr6ximos cinco afos.


Introduction: 15 Years of Progress in

Wheat Wide Crosses at CIMMYT
Abdul Mujeeb-Kazi

Experts predict today's worldwide
population of 5.5 billion people will
grow by 1 billion over the next decade,
and double to 11 billion in 40 years (Beamish
1994). By 2050, 12 billion people will crowd the
planet, with more than -,i of the growth
occurring in developing nations. These ominous
circumstances are placing a formidable task
before agricultural scientists and the food
management sector. On one front where plant
breeders are involved in crop improvement
efforts to meet the ever-increasing demand for
food, they are finding less and less appropriate
germplasm with desired traits among cultivated
crops themselves with which to make the
needed improvements (Harlan and deWet 1971).
Fortunately, new and useful genetic resources
are being found in wild, uncultivated plants. The
challenge is to incorporate this germplasm into
existing food crops.

Wide Crosses in the Triticeae

Because cereal crops provide the structural base
for world food production, it is fortuitous that
most alien genetic transfers, to date, have
involved the Triticum grass species within the
tribe Triticeae-where the greatest emphasis has
been placed on using these introgressions to
improve bread wheat (T. aestivum L.). Wheat has
received the most attention because of its global
importance and because genetic manipulation
techniques have become well established for
wheat and its relatives. Using wide crosses to
improve bread wheat is an area that CIMMYT

has pursued vigorously over the past 15 years.
Our research efforts and active collaborations
involving wide crosses are detailed in this
research report.

Of the approximately 325 perennial and annual
grasses within the Triticeae tribe, relatively few
have been hybridized with wheat. Perennials
used have been predominantly among
Thinopyrum spp. Among the 75 or so annuals,
particular successes have been achieved with
A. .;,:. Hordeum, and Secale spp. Over the last
decade and a half, noteworthy successes at
CIMMYT and various other laboratories have
been achieved in the production of complex
hybrids among species in the Triticeae, which
now provide a potential stock of invaluable alien
germplasm. CIMMYT is continually adding to
this stock with its growing number of A genome
hexaploids and D genome synthetic hexaploids,
which are being developed through crosses
between durum wheat and a number of diploid
grass species (see Chapter 3 and Appendices 2
and 3), and other hybrid combinations produced
by crossing wheat with various perennial species
in the Triticeae (See Chapter 4 and
Appendices 4-Ji

Utility of Wide Hybrids

There are different methodologies for
transferring desired resistances or tolerances
from the alien species to wheat (Chapters 3 and
4). Irrespective of the procedure adopted,
production of hybrids (intergeneric or

interspecific) is the key to accomplishing useful
genetic transfers. There are two ways to
accomplish this: 1) alien species can be screened
for specific resistances or tolerances and then be
hybridized with wheat or 2) the alien species are
first hybridized with wheat and then the
advanced derivatives can be screened.

Wide hybrids provide cytological data and
evolutionary information about the parental
species, as well as the practical motivation to
improve wheat by transferring significant
characteristics from alien species. Alien transfers
can diversify variability for both dynamic biotic
situations and static abiotic circumstances. Alien
sources of variability are normally inaccessible to
breeders who work in conventional crop
improvement programs and hence wide
hybridization is considered additive to
traditional plant breeding efforts.

Intergeneric vs interspecific
Actual successes of incorporating usable alien
genetic variation have been relatively few
(Sharma and Gill 1983a, Mujeeb-Kazi and
Kimber 1985) and for the most part have
involved simply inherited genetic traits. So, in
order to ensure faster practical returns to
agriculture, we believe that simply inherited
traits should be the major emphasis when
making intergeneric crosses (see Chapter 4),
while traits with complex heritability (involving
the introgression of several genes
simultaneously) and ill-defined genetic
information should be limited to less complex
interspecific crosses (see Chapter 3). Some of the
complex traits receiving our attention are
associated with resistances or tolerances to
Helminthosporium sativum, Fusarium graminearum,
Tilletia indica (syn. Neovossia indica), and salinity.

As major problems that limit the use of wide
hybrids, i.e., crossability and embryo
development, continue to be solved, it has
encouraged researchers to examine the
possibilities of yet even wider hybrids, e.g.,
crossing wheat with maize, millet, sorghum,
teosinte, and Tripsacum (see Chapter 5).

Evolution of Wide Cross
Research at CIMMYT

In the late 1970s, the research mandate for wheat
wide crossing at CIMMYT was highly specific in
its emphasis to refrain from activities that
emulated those of other established centers of
excellence. We were not to set our investigations
towards highly specific cytogenetic research.
Our efforts were to focus upon exploiting the
applied advantages for CIMMYT's mandate of
wheat crop improvement. Presumably, we
would lose some precise scientific explanations,
but, through opening up collaborative linkages
with basic research centers of excellence, it
would be their job to unravel the howss" and
"whys" of the new germplasm we were
developing. This was essentially "CIMMYT's
look toward year 2000"-a wheat wide crosses
investigative framework to span two decades

It was unlikely that there would be speedy pay-
offs from such a risky endeavor. Further, instead
of handling a restrictively narrow objective
through one hybrid combination (wheat x one
alien source), the two-decade span would allow
us the time to exploit a wide array of alien
germplasm. It would also permit us to adapt
readily to new situations that would most likely
emerge with significant scientific discoveries, a
changing research environment, budgetary
decreases or increases, and client needs.

2 Chapter 1

At the onset of the current wide crosses program
in 1979, the goals were to:

* Produce classical intergeneric hybrids with
cytological validation over a three-year
* Design a breeding methodology that
permitted advance of the hybrids for field
evaluations with elite plant types that
dispensed with or minimized chromosome
analysis to confirm actual introgression.
* Emphasize complex genes and polygenically
controlled traits instead of simply inherited
* Select and stabilize the advanced hybrid
derivatives by transferring improved
germplasm to CIMMYT base and national
breeding programs.
* Proceed with advances of F1 hybrids and
basic research activities along classical
cytogenetical lines, preferably through
collaborative research, which would require
scientific communication through various

Subsequently a changing research environment
after 1987 placed considerable emphasis upon
basic and strategic research-a new stage that
the wide crosses program needed to assess
accordingly. Hence, there emerged a slight
reduction in the original research structure and
some additional modifications that involved:

* Initiating an interspecific hybridization
program based upon genomic proximity of
the closely related Triticeae species with the
three genomes (A, B, and D) of wheat. This
offered an avenue for relatively rapid, short-
term returns.
* Transferring targeted traits from distant
wheat relatives via intergeneric crosses.

* Producing polyhaploids in wheat using a
sexual cross system (wheat x maize) in efforts
to generate doubled haploids for the wheat
RFLP (restriction fragment length
polymorphism) mapping project (see
Chapter 5). These techniques would have
subsequent ramifications for future
cytogenetics and breeding activities.
* Using tissue culture to facilitate alien
introductions (see Chapter 6).
* Exploring the application of biochemical and
molecular markers to enhance the detection
of alien introgressions (see Chapters 7 and 8).
* Publishing results in refereed scientific

Prerequisites for Success

Two major prerequisites for success in wide
cross research are long-term commitment and
collaboration among specialists and institutions.

Long-term commitment
Over the years, certain advances in wide cross
research have set the stage for current support.
Some of these well-recognized global endeavors

* Nearly a century of accomplishments with
wheat x rye (triticale) and wheat x barley
* Nearly 50 years of astounding successes (by
the late E.R. Sears) with wheat cytogenetic
* Some 20 years of successes (by the late
E. Sebesta) in the production of wheat/alien
translocations such as 1AL/1RS. This
translocation and the spontaneous 1BL/1RS
translocation are undoubtedly the most
significant examples of successful alien
introgressions in wheat breeding; the
1BL/1RS contribution (more than 5 million

Introduction: 15Years of Progess 3

hectares planted to such wheat varieties) is
believed-by some-to be the most important
natural genetic manipulation made to date.
*Some 15 years of work involving alien genetic
stock developments of Dasypyrum villosum,
Thinopyrum ,ljog,at'n. Th. bessarabicum,
Hordeum I,..,. and Triticum tauschii-in
each case by independent researchers
working in major programs.

In addition to the above examples, the known
complexities that are firmly established in the
scientific literature show that long-term
commitments must prevail for such involved
research to be successful. Regardless of the
financial investment, both researchers and
administrators must recognize this inescapable

Collaboration among
specialists and institutions
A successful wide crosses program like the one at
CIMMYT requires the commitment and
cooperation of literally hundreds of specialists
located at CIMMYT and research centers
worldwide. Interconnected disciplines include
genetics, cytogenetics, pathology entomology,
physiology biotechnology, breeding, nutrition,

and agronomy. For a who's who listing of
specialists involved, just scan the references cited
in this research report.

Collaboration with other research centers has
been particularly important to recent successes
at CIMMYT. Just as one example, scientists at the
USDA-ARS Forage and Range Research
Laboratory, Logan, Utah, have provided
germplasm of many species, facilitated crosses
with Triticeae species in the grass nursery
maintained at Logan, and shared knowledge on
genome compositions of various Triticeae
species. Other institutions have provided
valuable information on sources of various
disease resistances and environmental stress

Strengths of other Laboratories

We recognize that non-CIMMYT workers with
basic research strengths are located in other
laboratories and we do not pretend to compete.
With diversified mandates, some are most likely
more ideally suited than CIMMYT's applied
program to conduct specific sophisticated
aspects of wide cross work aimed at global crop
improvement. We acknowledge-but do not
identify-these researchers in this particular

4 Chapter 1


Perennial and Annual Wheat

Relatives in the Eiticeae
Abdul Mujeeb-Kazi and Richard R.-C. Wang

Of the approximately 325 species in the
tribe Triticeae, about 250 are
perennials and 75 are annuals (Dewey
1984). Relatively few perennials have been
intergenerically crossed with wheat because of
the complexity of doing so and embryo
rescue/ regeneration constraints. The perennials,
which include many important forage grasses,
have the potential to serve as a vital genetic
reservoir for the improvement of the annual
grasses, which include the major cereals (bread
wheat, durum wheat, triticale, barley, and rye).
Perennials that have been successfully utilized
for improving wheat are predominantly in the
Thinopyrum group.

Wide hybrids in the Triticeae have been
attempted and studied for more than 100 years.
The first such hybrid was between wheat and
rye (Wilson 1876). Rimpau (1891) described 12
plants recovered from seed of a wheat-rye
hybrid that represented the first triticale. Farrer
(1904) made similar early studies of wheat-
barley hybridization; however, Shepherd and
Islam (1981) concluded that it is improbable that
these were true hybrids. Several perennial
grasses were hybridized with wheat as early as
the 1930s with the objectives of transferring
disease resistance and perenniality into annual
crops (Tsitsin 1960, 1975). Many hybrids
involving Triticum and severalA,. 1;, ./. species
were made during the 1920s and 1930s (Kihara
1937) from which the genomic relationships of
the two genera were derived (Lilienfeld 1951).
The large-scale practical use of the hybrids,

however, was delayed until the advent of
colchicine treatment (Eigsti and Dustin 1955) in
the late 1930s. The ability to double the
chromosome number of hybrids using colchicine
had both practical and theoretical consequences.
The production of fertile amphiploids provided
the way to develop triticale as a new cereal crop
(Gupta and Priyadarshan 1982) and also
advanced evolutionary studies when McFadden
and Sears (1946) resynthesized T. aestivum and
thus discovered T. tauschii to be the D genome
donor to bread wheat.

With the advancement of hybridization
techniques (Kruse 1973) and embryo culture
(Murashige 1974), wide hybridization became a
more common practice and involved more
perennial species. In reviews of the progress of
wide hybridization, Dewey (1984) and Wang
(1989) clearly showed intense interest among
breeding programs in utilizing the genetic
resources available in the perennial Triticeae for
cereal improvement.

Since 1980, CIMMYT has been conducting a
vigorous gene transfer program in which species
of the perennial Triticeae are utilized for wheat
improvement. We have emphasized the
perennials over the annuals because of greater
biotic/abiotic resistances that the perennials
most likely possess because of their habitats. In
the quest to acquire genetic diversity, we have
found that obtaining usable alien characters
requires precise transfer of the controlling genes
of a desirable character that comes from a donor

species with different genomes. An ideal gene
transfer involves normal introgression of the
alien material without negative background
effects on grain yield and quality. To utilize
effectively the Triticeae gene reservoir, we need
to know:

* The genome constitutions of the donor
* The genomic relationships between the donor
and recipient species;
* The chromosomal locations) of the desirable
* The number of the gene(s) conferring the
desirable trait and the mode of inheritance;
* Whether the donor's gene(s) can be expressed
in the recipient species; and
* Whether any negative effects ensue from the

CIMMYT is meeting the above prerequisites
with regard to various desirable characteristics,
including disease resistances and abiotic stress

Genome Make-Up in the Triticeae

Perennial species
In the genomic system of classification of the
perennial Triticeae (Dewey 1984), 13 genera with
defined genomes or genome combinations are
recognized. Due to a lack of sufficient
information, we do not consider two of these:
Hordelymus andFestucopsis. However, the
remaining 11 genera with their type species and
genome compositions are: Agropyron
(A. cristatuipP) It. .ili, -, .... (A. pectinatum;
W ), ?'.,. ,-,,... ....1;, .i (P. strigosa; S), Psathyrostachys
(Ps. 'I.ir, ;..... N), Critesion (C. jubatum; H; also
Hordeum), Thinopyrum (Th. junceum; J-E), Eli .
(E. repens; SX), Elymus (E. sibiricus; SHY) Leymus
(L. arenarius; XN), Pascopyrum (Pa. smithii;
SHXN), and Secale (S. montanum; R).

As new genomic compositions were found in the
Triticeae species (Wang et al. 1986; Liu and Wang
1989; Jensen 1990a,b; Torabinajad and Mueller
1993; Assadi and Runemark 1994), the genomic
classifications were modified. The closeness
between the J and E genomes also necessitated a
genome symbol change from E to je (Wang and
Hsiao 1989). With these changes, a more detailed
system became warranted. Table 2.1 lists the
perennial genera under the new genomic system
where specific genome combinations are
assigned to known species.

,1. ,', ..i remains a small genus consisting of
P-genome species at three ploidy levels.
Psathyrostachys is comprised of N-genome
diploids, although autotetraploid cytotypes have
been both discovered and synthesized. Pending
new information, the genome make-ups of the
genera Pascopyrum, Australopyrum, and Secale
presently remain unchanged. See the Chapter 7
discussion of the genomic status of tetraploid
Leymus racemosus.

With the discovery of natural species having the
SSPP genomes, i.e., .,. 1 .,....., ... I;.i tauri (Wang
et al. 1986) and P. deweyi (Jensen et al. 1992),
.,.,1, ... 1 .i ': ,, ..-% has a section nam ed
Pseudopyron to accommodate these SSPP species.
These species exemplify the inadequacy of
morphology alone and the added importance of
genome analysis when studying species
relationships. Similarly, only genome analysis
can separate species within the genus Elymus
into the SH, SHY, SY, and SYP groups with any
certainty. Chinese taxonomists (Keng 1965, Yen
and Yang 1990) have given genus names to the
SSYY and SSYYPP species (i.e., ',. ...., ;.i and
T-,. .j ;l;.i respectively). However, we have kept
them at the sectional level to avoid massive
name changes since Elymus is the largest genus

6 Chapter 2

Table 2.1.The genomes and ploidy levels of the perennial genera of the tribeTriticeae.a,b

Genus Section Ploidy levels) Species


Pseudopyron SSPP





From HH to
HHHHHH (i.e., 2x to 6x)



Elongatum EE

Intermedium EESS





Th. elongatum
Th. scirpeum
Th. curvifolium
Th. turcicum
Th. ponticum

Th. caespitosum, Th. nodosum, Th. scythicum

Th. intermedium

E repens

L. arenarius, L. racemosus,
L. mollis, L. triticoides, L. salinus, L. cinereus,
L. innovatus, L. chinensis
L. angustus

P smithii

A. pectinatum

S. montanum

Perennial and Annual Wheat Relatives 7

A. cristatum, A. mongolicum, A. fragile
A. desertorum, A. michnoi
A. cristatum

P strigosa, P libanotica, P stipifolia, P spicata
P tauri, P deweyii

E sibiricum, E canadensis, E arizonicus, E caninus,
E vaillantianus
E dahuricus, E drobovii, E tsukushiensis, E kamoji
E scaburus, E rectiselus

E ciliaris, E parviglume, E longearistatus, E strictus,
E gmelinii, E pendulinus, E abolinii, E panormitanus,
E shandongensis, E ugamicus

E alatavicus, E batalinii, E grandiglumis, E kengii

P lanuginosa, P juncea, P fragilis, P huashanica,
P kronenburgii.

H. bogdanii, H. violaceum,
H. jubatum, H. brevisubulatum, H. iranicum, etc.

Th. junceum,
Th. bessarabicum
Th. junceiforme, Th. sartorii, Th. distichum















a This table does not provide details of genera hybridized in the CIMMYT program that are elaborated in Chapter 4. For a
more detailed listing of species, readers should refer to Dewey (1984).
b For simplicity, the jb and Je genomes (discussed in the text) are replaced by J and E, respectively, in this table.

in the Triticeae consisting of approximately 150
species. Only a small number of these species
has been genomically analyzed (Lu and von
Bothmer 1993).

Although the diploid species in Australopyrum
contain the W genome (Hsiao et al. 1986), the
polyploid species in some Australian Elymus
species appear to be allopolyploids that contain
other genomes in combination with W
Therefore, Elymus may include some other
genomic combinations presently unknown to us.
For example, E. rectisetus and E. scabrus both
appear to have the genome composition
SSYYWW (Torabinajad and Mueller 1993).

Critesion is now recognized as a section of
Hordeum (Bothmer et al. 1986). Genetic
regulators (promoters and suppressors) of
chromosome pairing might have confused the
make-up and earlier classification of the
Hordeum genus, which contains both perennial
and annual species (Dewey 1984). For example,
a gene has been shown to be responsible for the
low pairing between H. violaceum and
H. bgi/niil/\ang et al. 1991). It may also have
led to misinterpretation of some cytogenetic data
since the number of hybrid plants obtained from
certain combinations has been very low. The
bivalentization mechanism may also cause an
autopolyploid to behave like an allopolyploid
(Wang and Hsiao 1989). Further research being
pursued by Bothmer and his colleagues should
lead to a better understanding of the genomic
relationships of the perennials in Hordeum.

The genus Thinopyrum is embroiled in the most
recent controversy. The two diploid species,
Th. bessarabicunand Th. il gtiit, were
originally given the genome symbols J and E,
respectively (L6ve 1984). Lately, cytogenetic data
have demonstrated the closeness between the

two genomes. Thus, they have been merged into
a single, basic genome symbol as proposed by
Dvorak (1981), Dewey (1984), McGuire (1984),
Wang (1985), Pienaar et al. (1988), and Wang and
Hsiao (1989). See the discussion on
Th. bessarabicunin Chapter 7. Different opinions
do exist (Jauhar 1988, 1990). Liu and Wang
1993a) suggest that Th. junceum, the type species
(JbJbJe), possesses two Jb genomes and a modified
Je genome, which is also present in Th. sartorii.

On the other hand, Th. junceiforme (JbJe2) and
Th. I.' '.." -"l' I' share another modified
genome (J~2), differing from the respective
second genomes, which are Jb and Je The two
modified je genomes probably arose from
recombination between jb and Je through the
pivotal-differential evolution of the polyploid
species. As a result, J'e is closer to Je while Je2 is
closer to jb. Therefore, all the triploid and
tetraploid hybrids involving Thinopyrum species
have meiotic pairing patterns closer to those for
autoploid plants than those for strict alloploids.
The fertility in the hybrids of Th. scirpeum /
2*Th. bessarabicutfTh. ,~l,,ig. iiti, which have the
genome constitution jeje2jbj (Wang 1992), further
supports the pivotal-differential hypothesis.

Liu and Wang (1989) have shown the presence of
an S genome in Th. caespitosum and the JeeSS
genome composition has since been found in
Th. nodosunand Th. scythicum (Liu and Wang
1993b). Depending on the accessions,
Th. intermediuncan be represented by the
genome formula JeeJXJXSS, where Jx can be any
version of the J genome.

Annual species
The annual plants of the Triticeae are confined
largely to the Triticum and A,: ;',... species; some
notable exceptions include species from
Hordeum, Secale, Haynaldia, Eremopyrum,

8 Chapter 2

Heteraunthelium, Taeniantherum, and Henrardia.
The large number of generic and specific names
of the interrelated Triticum and A,..1..' 'p
has led to considerable confusion over the years.
The multitude of names not only expresses the
whims of various taxonomists, but also

represents the diversity of the species themselves.
To reduce some of the confusion, Kimber and
Feldman (1987) have compiled a synonym list of
the most commonly used names among the
Triticum, /A..., ', groups (Table 2.2).

Table 2.2.A synonym list of annual AegilopslTriticum species (Kimber and Feldman 1987).

Ae. aucheri
Ae. bicornis
Ae. biuncialis
Ae. caudata
Ae. columnaris
Ae. comosum
Ae. crassa
Ae. cylindrica
Ae. geniculata
Ae. heldreichii
Ae. juvenalis
Ae. kotschyi
Ae. ligustica
Ae. longissima
Ae. longissima
var. sharonensis
Ae. lorentii
Ae. markgrafii
Ae. mutica
Ae. neglect
Ae. ovata
Ae. peregrina
Ae. persica
Ae. recta
Ae. searsii
Ae. sharonensis
Ae. speltoides
Ae. squarrosa
Ae. tauschii
Ae. triaristata (4x)
Ae. triatistata (6x)
Ae. tripsacoides
Ae. triuncialis
Ae. turcomanica
Ae. umbellulata
Ae. uniaristata
Ae. variabilis
Ae. vavilovi
Ae. ventricosa

T aegilops
T aegilopoides
T aaraticum
T bicorne
T boeoticum

T speltoides (aucheri)
T bicorne
T machrochaetum
T dichasians
T columnare
T comosum
T crassum
T cylindricum
T ovatum
T comosum
T juvenile
T kotschyi
T speltoides (ligustica)
T longissimum

T sharonense
T macrochaetum
T dichasians
T tipsacoides
T neglect, T triaristatum (4x)
T ovatum
T peregrinum
T triunciale
T recta, T tiaristatum (6x)
T searsii
T sharonense
T speltoides (ligustica)
T tauschii
T tauschii
T neglect
T recta
T tipsacoides
T triunciale
T juvenile
T umbellulatum
T uniaistatum
T peregrinum
T syiacum
T ventficosum

T tauschii
T monococcum
T timopheevii
Ae. bicornis
T monococcum

Perennial and Annual Wheat Relatives 9

T carthlicum = T turgidum
T columnare = Ae. columnaris
T comosum = Ae. comosa, Ae heldreichii
T crassum = Ae. crassa, Ae vavilovi
T cylindricum = Ae. cylindrica
T dichasians = Ae. caudata, Ae markgrafii
T dicoccoides = T turgidum
T dicoccum = T turgidum
T durum = T turgidum
T juvenile = Ae. juvenalis, Ae turcomanica
T kotschyi = Ae. kotschyi
T longissimum = Ae. longissima
T macrochaetum = Ae. biuncialis, Ae. lorentii
T monococcum = T aegilopoides, T boeoticum, T urartu
T neglect = Ae. neglect, Ae. triaristata (4x)
T ovatum = Ae. ovata, Ae. geniculata
T peregrinum = Ae. peregrina, Ae variabilis
T persicum = T turgidum
T polonicum = T turgidum
T recta = Ae. triaristata (6x)
T searsii = Ae. searsii
T sharonense = Ae. sharonensis,
= Ae. longissima var sharonensis
T speltoides
(aucheri) = Ae. aucheri
T speltoides
(lingustica) = Ae. speltoides Ae. ligustica
T syriacum = Ae. vavilovi, Ae crassa var
vavilovi or var palaestina
T tauschii = Ae. squarrosa
T timopheevii = T timopheevii, T araraticum
= T timopheevii var zhukovskyi
T timopheevii
var. zhukovskyi = T timopheevii
T triaristatum = Ae. triaristata, Ae recta, Ae. neglect,
T rectum
T tipsacoides = Ae. mutica
T triunciale = Ae. triuncialis
T turgidum = T carthlicum, T dicoccoides,
T dicoccum, T durum, T persicum,
T polonicum
T umbellulatum = Ae. umbellulata
T uniaristatum = Ae. uniaristata
T urartu = T monococcum
T venticosum = T ventficosa

Despite its limitations, analysis of chromosome
pairing, which has evolved from the "analyzer
method" of H. Kihara (Lilienfeld 1951), can
provide an insight into genomic relationships
among the annuals. Although universal
acceptance on usage may not exist, those
interested in in-depth treatment of this topic are
referred to Morris and Sears (1967), Bowden
(1959, 1966), and proceedings of recent
International Wheat Genetics Symposia
(Sakamoto 1983, Miller and Koebner 1988,
IWGS 1993).

In general, hybrids, in which the alien parent is
the diploid genome donor, exhibit
approximately seven bivalents because the
Triticeae tribe has a basic chromosome number
of seven. All hybrids involving nondonor
diploid "analyzers" should have 21 or more

univalents at meiotic metaphase I. This approach
has worked quite well in enabling cytogenetists
to assign genome designations to the annuals
(Tables 2.3 and 2.4). Only a few cytogenetic
studies have resulted in any significant changes
to the genome designations made using the
analyzer method.

Genome Relationships among
Perennial and Annual Species

Increased understanding of genome
relationships among the perennial and annual
species of the Triticeae would contribute much to
accomplishing successful genetic transfers.
Unfortunately, researchers have done very little
work on these relationships. However,
extrapolated interpretations of meiotic data from
polyploid hybrids suggest very low homology

Table 2.3. Genome designations of the diploid TriticumlAegilops species, according to various workers.

Diploid Kihara, in Kimber & Kimber &
TriticumlAegilops species Lilienfeld (1951) Sears (1983) Tsunewaki (1988)

T monococcum L. A
T urartuTum. A

Ae speltoides Tausch S
Ae. bicorne (Forsk.) Jaub. & Sp. Sb
Ae. longissimum Schweinf. & Muschli in Muschli SI
Ae. sharonensis Eig SI
Ae. searsii Feldman & Kislev Ss

Ae. mutica Boiss. Mt* T

Ae. tauschiiCoss. D

Ae. comosa Sibth. & Sm. M

Ae uniaristata Vis. Mt Un N

Ae. caudata L. C

Ae. umbellulata Zhuk. Cu U



Ohta (1990) proposed that the genome Mt be changed to Sm

10 Chapter 2

between any of the genomes (P, S, N, H, J, W, X,
and Y) in the perennials and some of the
genomes (A, B, D, G, M, T, and U) in the annuals.
Most early alien gene transfers to wheat
involved species of Thinopyrum. This suggests a
closer homoeologous relationship between the
Thinopyrum genomes and the wheat genomes
than that present in the other perennial genomes.
However, the chromosome pairing between
these two genomic groups in the absence of the
Phl gene is still much lower than the pairing
among the A, B, and D genomes of wheat alone.
Therefore, gene transfer attempts require
complex, long-term genetic manipulations to
facilitate recombination and alien introgression.

Table 2.4. Genome designations of polyploid TriticumlAegilopsspecies.

Polyploid Kihara, in Kimber & Latest
TriticumlAegilops species Lilienfeld (1951) Sears (1983) changes

T dicoccoides KUrn
T dicoccum Schrank
T durum Desf.
T turgidum L.
T persicum (Percival)
Vavilov ex Zhukovsky

T aestivum L. em. Thell.
T spelta L.
T compactum Host
T sphaerococcum Perc.
T macha Dek. et Men.

T timopheevii Zhuk.
T zhulovskyi Men. et Er

Ae. ovata L.
Ae biuncialis Vis.
Ae. columnaris Zhuk.
Ae. triaristata Willd.
Ae. recta (Zhuk.) Chen.
Ae. variabilis Eig
Ae. triuncialis L.

Ae. cylindrica Host
Ae. crassa (4x) Boiss.
Ae. crassa (6x) Boiss.
Ae. vavilovi(Zhuk.) Chen.
Ae ventricosa Tausch
Ae. juvenile (Thell.) Eig









Hybridization between Wheat and
Perennial Triticeae

Since Wang (1989) reviewed intergeneric crosses
involving perennial Triticeae, a few new hybrids
have been obtained. All genomes of the
perennial Triticeae have been combined, either
singly or in combination, with the A, B, and D
genomes of bread wheat (Wang 1989, Mujeeb-
Kazi et al. 1994b). To date, there have been 89
different combinations made involving the three
genomes of wheat and the eight genomes of the
perennial species. Of these, 52 involve hexaploid
wheats, 30 tetraploid wheats, and seven diploid
primitive wheats.

Additional Triticum, I ,, i .
hybrids include those from
crosses of T. aestivum and
A. u.........h.,,,2,,= 2x=14, PP),
A. desertorutor A. michnoi
(2n=4x=28, PPPP) (Chen et al.
1990); and crosses of T. aestivum
and A. cristatum (2n=4x=28,
PPPP) (Jauhar 1992). New
Triticum /Thinopyrum hybrids are
crosses of T. durum and
Th. '. .i '.il.. i(2,=2x=14, JJ)
(Jauhar 1991), Th. scirpeum
(4x=28, JeJejee), and
Th. junceiform44x=28, JbJbJeJe)
(Liu and Wang 1993c).


UM Newly obtained Triticum lElymus
combinations include
E. 1:..in',,..... : ,. T. aestivum
UMX (SYABD; Lu and Bothmer 1989),
USI E. kamoji x T. aestivum (SHYABD
and SHYAABBDD; Weng and
Liu 1989), T.' i ..: -l,. x Ps. juncea
DcX (Mujeeb-Kazi et al. 1994e) and
CcXSs T. aestivunx E. rectisetus

Perennial and Annual Wheat Relatives 11

(ABDSSYYWW; Wang et al. 1993). Meiotic
pairing data were published for previously
obtained hybrids T aestivum x A. cristatum
(ABDP and ABDPP; Limin and Fowler 1990)
T. aestivunx A. desertorum (ABDPP; Li and Dong
1991a), T. aestivum x A. michnoi (ABDPP; Li and
Dong 1991b), and T aestivum x Ps. juncea (ABDN
and ABDNN; Plourde et al. 1990).

The wide array of genetic variability residing in
the above Triticeae relatives supplies a superb
arsenal of new defenses against biotic and
abiotic stresses in cereals. Inevitably, the use of
this variation has its constraints since genomic
homoeology does not offer a satisfactory level of
chromosomal association in the Fl hybrids to
promote alien gene transfers. Use of the phl
locus may provide a way to overcome the
recombination constraint, but the ensuing
complexities to obtain backcross derivatives and
stable advanced progenies are going to be fairly
long-term. Other genetic manipulations exist
that progressively revolve around production of
alien disomic chromosome additions or
substitutions, which could lead to translocations
or subtle exchanges through cytogenetic and

novel manipulative procedures. Although the
process will be slow, the benefits of
incorporating these diverse genetic resources
into wheat will be extremely high.

Hybridization between Wheat and
the Annual Triticeae

Production of interspecific hybrids involves the
diploid Triticum /,:,. ;. 1.. spp. In Chapter 3, we
elaborate on the utilization of these diploids that
have high genetic proximity to the A, B, and D
genomes of bread wheat-particularly Triticum
tauschii ( 1,. 1./. -, squarrosa). Although variation
in the annual species resides predominantly in
the diploids, it does extend somewhat to the
tetraploids and hexaploids. Triticum/,. 1, 7/.
spp. with different genomic make-ups (Tables 2.3
and 2.4) offer an additional gene pool also worth
considering in our intergeneric crossing efforts.

Kimber (1993) compiled a list (Table 2.5) of all
the Triticum/A. .;1,. -, spp. available for alien
transfer based on genome relationships
determined by mathematical measurements
made to date.

12 Chapter 2

Table 2.5. The genomes of TriticumlAegilops and their availability for alien gene transfer (Kimber 1993).

Genome Species Optimum technique, difficulty, and availability

T syriacum

G T timopheevi

M T comosum
T columnare
T crassum
T juvenile
T macrochaetum
T ovatum
T syriacum
T triaristatum

N T uniaristatum
T venticosum

Sd T speltoides
T bicorne
T kotschyi
T longissimum
T searsii
T sharonensis

T T tipsadoides

U T umbellatum
T kotschyi
T columnare
T juvenile
T macrochaetum
T ovatum
T triaristatum
T triunciale

Recombination, meiotic difficulty, genome repatterning, poor availability

Pairing modification or ionizing radiation, meiotic difficulty,
considerable genome repatterning, poor availability

Pairing modification or ionizing radiation, meiotic difficulty,
considerable genome repatterning, poor availability

Recombination and/or pairing modification (some S genome accessions
contain suppressors of Ph), meiotic difficulty considerable genome
repatterning, available

Recombination and/or pairing modification (some T genome accessions contain
suppressors of Ph), meiotic difficulty considerable genome repatterning, available

Ionizing radiation, meiotic difficulty,
considerable genomic repatterning, poor availability

a See Table 2.2 for Triticum/Aegilops synonyms.
b Genome repatterning indicates modifications of the genome in the donor.
c Meiotic difficulty indicates possible complications in the introduction of alien variation due to the presence of
nonhomologous genomes or translocations.
d S genome also similar to the B genome designation of T. speltoides.

Perennial and Annual Wheat Relatives 13

A T monococcum Recombination, possible partially unreduced gametes, some genome repatterning, good

T turgidum Recombination, some genome repatterning, good availability

T timopheevi Recombination, genome repatterning, meiotic difficultyc, available

T aestivum Recombination from landraces, some genome repatterning, very good availability

C T dichasians Pairing modification, meiotic difficulty considerable genome repatterning, poor availability

T cylindricum Pairing modification, meiotic difficulty considerable genome repatterning, poor availability

D T tauschii Recombination, possible partially unreduced gametes, good availability

T cylindricum Recombination, meiotic difficulty, genome
T venticosum repatterning, available

T crassum Recombination, meiotic difficulty, considerable
T juvenile genomic repatterning, poor availability


Interspecific Crosses:

Hybrid Production and Utilization
Abdul Mujeeb-Kazi

Conventional improvement of bread wheat
(Triticum aestivum L.; 2n=6x=42; AABBDD),
breeders have normally made crosses
between varieties. Such crosses have few
constraints and invariably all associations of
parental traits and segregation are based on
genetic recombination. The next step in bread
wheat improvement is to tap into the varied
gene pools of other Triticum species. In such
interspecific crosses, breeders utilize the
numerous alien accessions among these species
that have genomes similar to the A, B, or D
genomes of bread wheat. These crosses allow for
relatively easy alien gene transfers, are
compatible with normal field research, and set
the stage for the successful introgression of
several genes simultaneously (i.e., en-bloc

Germplasm Used

A number of accessions of diploid wild relatives,
which have either the A, B, or D genomes, are
potential candidates for use in interspecific
crosses. We maintain working collections of wild
grass accessions with these genomes. For the
D genome, thea is T. tauschii (Appendix 2); for
the A genome, there is T. monococcum (= T. urartu
and T. boeoticum; Appendix 3); and for the B (S)
genome, there are the T. speltoides and related
accessions. Kimber and Feldman (1987) discuss
other potential sources. The procedures used to
incorporate such alien variability and the choice
of genome to work with differ among

Use of T. tauschii at CIMMYT

Why T. tauschii?
CIMMYT has been concentrating on exploiting
accessions of the annual wild relative T. tauschii
(goat grass) because we believe the wide
diversity and distribution of this species across
Eurasia (see Kimber and Feldman 1987) provide
a unique opportunity for exploiting new genetic
variability. T. tauschii has a wide range of
resistances or tolerances to biotic or abiotic
stresses (Valkoun et al. 1990), such as Karnal
bunt (Tilletia indica), scab (Fi -..i ;'I .i. i lu. .., ,i.
spot blotch (Helminthosporium sativum syn.
Bipolaris sorokiniana), leaf rust (Puccinia recondita),
stripe rust (P. striiformis), salinity, and drought.
The wild grass also appears to be a potent source
of new variability for important yield
components such as 1000-grain weight and
increased photosynthetic rate not to mention
improved bread making quality.

In addition, T. tauschii (also called A,. :;,.7.
squarrosa) is unequivocally accepted as being the
donor of the D genome to bread wheat (Kimber
and Feldman 1987). We consider crosses with
T. tauschifo be interspecific instead of
intergeneric because of T. tauschii's diploid
nature and its D genome status, i.e., 2n=2x=14,
DD. Note that we consider most crosses with
diploid Triticum /. .1, ,..g spp. (see Table 2.3) to
be interspecific because of their genomic
similarity to bread wheat. As explained in
Chapter 2, we place all other crosses with

different TriticumI:. .;, 1. spp. having
dissimilar genomes into the intergeneric
category (see Table 2.4).

Sources of the accessions
For our work with the D genome, we obtained
490 T. tauschii accessions-mostly of winter
habit-from CIMMYT's wheat germplasm bank
and from researchers in Pakistan (N.I. Hashmi of
the National Agricultural Research Council,
Islamabad), the UK (C. Law, Institute of Plant
Science Research, Norwich), and the USA (B.S.
Gill, Kansas State University; G. Kimber,
University of Missouri; R. Metzger, Oregon State
University; and G. Waines, University of
California, Riverside).

Plant generation
Seeds of selected T. tauschii accessions are
planted in jiffy-7 peat pots and vernalized at 80C
with eight hours light for eight weeks. The
seedlings are transplanted in pots containing a
steam-sterilized mixture of soil, sand, and peat
moss. Controlled greenhouse conditions involve
22/140C day/night temperatures, approximately
,-.II relative humidity, and 14 hours of natural
light. We vernalize the winter habit T. tauschii
seedlings and transplant them during
CIMMYT's normal wheat crop cycles at four
Mexican locations:

* Ciudad Obregon (November to May) for
crossing with T. ,' .. i
* El Batan (May to October) for crossing with
T. I,. ;,, ... .
* Toluca (May to October) for F. graminearum
screening; and
* Poza Rica (November to April) for H. sativum

At Ciudad Obregon, vernalization of the
T. tauschiiccessions results in very vigorous
growth; flowering takes place from 90 to 135

days of age. Growth is less vigorous at El Batan,
Poza Rica, and Toluca; flowering is extremely
delayed. We recognize that photoperiod
response may have alleviated the late flowering,
but logistically, this is possible only in Cd.

We are using the T tauschii accessions in the
following ways:

* Producing synthetic hexaploids by crossing
T. ',.;1:r.. cultivars with T tauschii

* Crossing elite, but susceptible T. aestivum
cultivars with resistant T tauschii accessions
and backcrossing the ABDD F1 hybrids with
the elite T. aestivum cultivar used in the
initial oass.

* Extracting the AABB genomes from
commercial T. aestivum cultivars and then
developing hexaploids by crossing with
desired T. tauschii accessions.

For any of the above techniques, it is important
that the desired traits in the D genome from
T. tauschibe identified since genetic factors in
the A and B genomes may mask or modify its
expression. However, this may not be a general
rule. For example, Multani et al. (1988) observed
that synthetic hexaploids, which were produced
with a KB-susceptible durum parent, expressed
the KB resistance of the T. tauschii parent.

We can screen the T. tauschii accessions for their
many desired attributes and then cross selected
ones to T. aestivum. However, when screening the
T. tauschii accessions is sometimes a major
constraint, we can hybridize them with
T. 'i,,~;,1. and then screen the resulting
synthetic hexaploids.

Interspecific Crosses 15

Resistance screening of the T. tauschii accessions
for H. sativum at Poza Rica and F. graminearum at
Toluca have been inconclusive. Growing
conditions at these two Mexican locations can
adversely affect the alien species, but logistically
the sites are ideal for disease resistance
screening. Because of the screening constraints,
winter habit, and the tendency for shattering
(which could cause a weed problem at the
stations) of the T tauschii accessions themselves,
we decided to indiscriminately cross T. ,. 1.;,. i
with T tauschii accessions. This has allowed us to
screen the resulting synthetic hexaploids more
adequately for our objectives without having to
deal with vernalization. In addition, when we
find a positive attribute (the durum parent being
susceptible), the breeding program can
immediately use the synthetic hexaploid. Even if
the synthetic expresses diluted resistance, the
end-product resistance is far superior to that
encountered in the best wheat germplasm

Variability of T tauschii accessions
See Chapter 7 for discussions on: 1) evaluation of
the variability for seed storage proteins and
isozymes associated with some T. tauschii
accessions and 2) comparison of the wide
variability of T tauschii accessions with the
variability of the synthetic hexaploids.

Synthetic Hexaploid Production

The hybridization process is quite simple when
using any of several manipulative crossing
procedures (durums as the female parent;
Appendix 1) described by Kruse (1973), Sharma
and Gill (1983a), Mujeeb-Kazi and Asiedu (1990),
or Riera-Lizarazu and Mujeeb-Kazi (1990). For
the crossing cycles, we always plant the durum
cultivars obtained from CIMMYT's durum

section over at least three planting intervals so
that T. ',. ;'6.lr,. flowering coincides with
flowering of the T tauschii accessions.
Procedures for embryo rescue, embryo culture,
and plantlet management are similar to those
described by Mujeeb-Kazi et al. (1987). We
transplant plantlets to a potted soil mix and
maintain them in the greenhouse at El Batan.

The 21 chromosomes of the F1 hybrids are
doubled (induced with colchicine or
spontaneous) to produce 42-chromosome
synthetic hexaploids (2n=6x=42, AABBDD;
Figure 3.1; Appendix 2). If found to be resistant
or tolerant after appropriate disease or stress

Triticum turgi
(2n=4x=28, A,


F1 hybrid
(21 chromosomes, ABD)

Doubling of the F1 hybrid
(Induced or spontaneous)

Synthetic hexaploid
(2n=6x=42; AABBDD)

Screening and detecting
(2n=6x=42; AABBDD)


I. aestivum cultivars
(2n=6x=42, AABBDD)


riticum tauschii
2n=2x=14, DD)

elected synthetics
ving desired trait
:6x=42; AABBDD)

(2n=6x=42, AABBD)

Selection of resistant/tolerant derivatives

Figure 3.1. Schematic showing the production of
synthetic hexaploids derived from crossing
Triticum turgidum x T. tauschii and their utilization.

16 Chapter 3

screening, the synthetic hexaploids can be
crossed conventionally with other bread wheat

Chromosome analysis (cytology)
To analyze the chromosomes in resulting
hybrids, root tips are collected from young
growing plants and processed according to the
schedule of Mujeeb-Kazi and Miranda (1985). Fl
hybrids with 2n=3x=21 chromosomes are treated
with 0.1% colchicine + 2. I dimethyl-sulfoxide
for six hours via aerated root-treatment. We
grow the colchicine-treated seedlings in the
greenhouse and place a glassine bag over each
spike after emergence from the boot. The seeds
that set on such plants after this treatment are
planted; after germination, they are analyzed for
chromosome number. Resulting plants are
maintained in the greenhouse and a glassine bag
is placed over each spike. For each doubled
fertile plant, we increase seed to a reserve of
50 g. W use surplus seed supplies for testing
resistance and tolerance to biotic and abiotic

We have produced, to date, nearly 525 synthetic
hexaploids-most involving a unique T tauschii
accession-over several crossing cycles
(Appendix 2). Whenever possible, we have
screened them for selected resistances to diseases
and abiotic stresses. We have identified
synthetics resistant to H. sativum and
F. graminearumSeveral have shown tolerance to
salt stress in initial field screening at La Paz, Baja
California Sur, Mexico. When screening shows
the synthetics to have these positive attributes,
we must be cognizant of the interaction of the A
and B genomes with the D genome as well as the
dilution effect of the resistance or tolerance of

the D genome in the hexaploid plant. Villareal et
al. (1990) have studied agronomic and taxonomic
traits of a few synthetics, such as days to
anthesis, plant growth, maturity, biomass,
harvest index, yield, pigmentation, and

With the development of synthetic hexaploids in
mind, all genuine Fl hybrids are stable for
2n=3x=21 chromosomes. After colchicine
doubling, the seeds generally possess 42
chromosomes. Although there are some
hypoploids and hyperploids among the resulting
synthetics (Figure 3.2), they can be subsequently
eliminated through additional chromosome
analysis (cytology) and seed increase. We have
identified synthetics that are resistant to
H. sativunand T indica. Synthetics resistant to
F. graminearunand tolerant to salt await
additional testing before we turn them over to
the breeders.

Molecular applications of the
Scientists involved in the collaboration of
CIMMYT, Cornell University, and the
International Triticeae Mapping Initiative (ITMI)
have found that some synthetic hexaploids
resulting from the T. I1,. ;ll.... x T tauschii crosses
are highly polymorphic (M. Sorrells, pers.
comm.). This is aiding molecular laboratories in
the development of the RFLP linkage map for
the D genome. The CIMMYT/Cornell
collaboration has produced synthetics that have
ultimately led to the development of doubled
polyhaploid plants (see Chapter 5 for the
significance of this). Other synthetics with
desirable agricultural attributes can be subjected
to molecular analysis.

Interspecific Crosses 17

New synthetics and
AA genome hexaploids
We are producing new synthetics using
additional T. tauschii accessions maintained in
the CIMMYT wide crosses working collection.
The CIMMYT Wide Crosses Laboratory has also
produced about 155 A genome hexaploids by
crossing durum wheat with A genome diploid
species (Appendix 3). It is anticipated that these
approaches will contribute to the availability of
additional genetic variability for wheat breeding
efforts, germplasm conservation, and global
germplasm distribution. There is merit for
international distribution of the synthetic and
A genome hexaploids, which would enable
national agricultural research programs with
specific objectives to do their own screening of
these accessions.

I /

Vt 's-00

I -

Direct Crossing of
T. tauschii with Bread Wheat

The most ideal, efficient technique for exploiting
T. tauschii variability for bread wheat
improvement is to achieve direct transfers from
resistant/ tolerant T. tauschii accessions to bread
wheat. This methodology rapidly produces
improved BCI derivatives with the six genomes
(AABBDD), five of which (AABBD) resemble the
elite wheat cultivar used in the cross (Figure 3.3).
With this methodology, we may have to contend
with the ensuing aneuploidy in the BC
derivatives and hence it may have lesser value in
mapping programs for recognized quantitative
traits. Cox et al. (1990) report on the numerous


I / s 1
5 If4111 |

c d

Figure 3.2. Somatic chromosome numbers of synthetic hexaploids from Triticum turgidumx T. tauschii:
a) 2n=6x=42, b) Hyperploid with 43 chromosomes, c) 42 with a telocentric arrowed, and d) Hypoploid with
41 chromosomes.

18 Chapter 3

Before crossing with bread wheat, reliable
screening of the T tauschiaccessions for
resistances and tolerances to diseases and abiotic
stresses is critical. Alonso and Kimber (1984),
Cox et al. (1990, 1991), and Gill and Raupp (1987)
unequivocally placed priority on direct
T. tauschitrossing with bread wheat cultivars.
Based on the transfer of stem rust resistance
from T. tauschii to the bread wheat cultivar
Chinese Spring, Alonso and Kimber (1984)
determined that one backcross onto the F1
hybrids re-instated -'2 of the genotype of the
recurrent parent.

When there are constraints to direct screening of
T. tauschii accessions, such as less than reliable
identification of resistance or tolerance to
H. sativutF. graminearum, T. indica, and salinity
we believe that screening of the synthetic
hexaploids, resulting from T. '.,. r.... x

Triticum aestivum
(2n=6x=42; AABBDD)
e.g., Bacanora

Triticum tauschii
(2n=2x=14; DD)

F1 hybrid
(28 chromosomes; ABD))

F1 as female Tfiticum aestivum
(ABD) (2n=6x=42; AABBDD)
Se.g., Bacanora

Backcross I

BC (and

Figure 3.3. Schematic demonstrating alien transfers
from Triticum tauschii (2n=2x=14, DD) to elite
TI aestivum cultivars via direct crossing and

T. tauschicrosses, is an alternative (Tables
3.1-3.4) especially where the durums are
susceptible. The information obtained from
screening the synthetics allows us to target
specific T. tauschii accessions for direct crossing
with susceptible elite bread wheat cultivars, e.g.,
'Ciano T 79' and 'Bacanora T 88' for H. sativum
resistance, 'Seri M 82' and 'Opata M 85' for
E graminearuresistance, and 'Oasis F 86' for
salt tolerance. Using these cultivars, we have
duplicated the crossing successes that Alonso
and Kimber (1984) had with the cultivar Chinese
Spring. Several other options are also available
for achieving additional crossing successes
(Gill and Raupp 1987, Riera-Lizarazu and
Mujeeb-Kazi 1990).

We have satisfactorily screened T. tauschii
accessions for KB resistance (Warham et al. 1986,
and unpublished data) and identified several i
infection types. Successful crosses have been
made between KB-susceptible bread wheat
cultivars Seri M 82 and Bacanora T 88 and
several of these T. tauschii accessions. The above
procedure is highly efficient and, from about 270
F1 hybrids, -''*-, are normal with 2n=4x=28

Extracting the AABB genomes
Extracting the AABB genomes from commercial
T. aestivum cultivars and then developing
hexaploids by crossing with desired T. tauschii
accessions allow for very clear analysis of the
genetic contribution of the alien D genome.
There is negligible interference from
recombinant segregation of the A and B genomes
that is rampant in the first hybridization
procedure (Figure 3.4). However transmission
of paternal clhmosomes and aneuploidy in the
backcross generations can complicate the

Interspecific Crosses 19

Table 3.1. Five synthetic hexaploids selected as resistant to Helminthosporium sativum compared with
their durum wheat parents.

Seed source Disease score
Durum wheat parent and identifier
Synthetic hexaploid (Synthetic ID no.a) Leaf damageb Seed damagec

CPIIGediz'S'I3/Goo'S'IIJo'S'ICR'S' 99 4
CPIIGediz'S'I3/Goo'S'IIJo'S'ICR'S'/41T tauschii 41 93 2
TK SN1081 98 4
TKSN1081/T tauschii 54 93 2
Gan'S' 96 3
Gan'S'/T tauschii 69 93 2
Decoyl 97 3
Decoyl/T tauschii 114 93 2
Decoyl/T tauschii 128 93 2
CianoT 79 (susceptible bread wheat) 99 5
BH1146 (resistant bread wheat) 97 3

a See Appendix 2.
b Two-digit scoring system:first digit = height of infection; i.e., five = up to center of plant, 9 = upto the flag leaf; second digit
= disease severity on infected leaves, 1 = low and 9 = total leaf destroyed.
c Grain infection scored as: 1 = low and 5 = high seed blemish at embryo points.

Table 3.2. Five synthetic hexaploids selected as
resistant to Fusarium graminearum compared with
their durum wheat parents.

Seed source
identifier Disease
Durum wheat parent and (Synthetic score
synthetic hexaploid I.D. no.a) on spikesb

Altar 84 5
Altar 841T tauschii 64 0

Laru'S' 5
Laru'S'/T tauschii 82 0
Gan'S' 5
Gan'S'/T tauschii 134 0
Crocethia 1'S' 5
Crocethia 1'S'IT tauschii 145 0
Crocethia_1'S'IT tauschii 161 0
Seri M 82
(Susceptible bread wheat) 5

a See Appendix 2.
b Fungal presence on spikes recorded as: 0 = no infection;
and 5 = severe infection detected through pink colored
fungal exudate on nodes and spikelets.

Table 3.3. Five synthetic hexaploids that have tested
positive for the Na:K discrimination trait associated
with salinity tolerance in hydroculture testing
compared with their durum wheat parents.a

Seed source
Durum wheat parent and (Synthetic K:Na
synthetic hexaploid I.D no.b) ratios

Rokel'S'/Kamilario'S' 1.2
Rokel'S'/Kamilario'S'IIT tauschii 39 7.7

PBW 34 1.2
PBW 114/T tauschi 13.3
CPI/Gediz/3/Gool/Jo'S'ICR'S' 1.1
/Jo'S'/CR'S'14/T tauschii 24 16.4

Scoop_1 1.5
Scoop_l/T tauschii 111 17.7
Decoyl 0.7
Decoyl/T tauschii 128 3.5

a Levels recorded 50 days after a 50 mol/m3 NaCI
concentration was reached.
b See Appendix 2.
c K:Na discrimination ratios; higher values are positive for
salinity tolerance.
d Synthetic obtained from H. Dhaliwal. Instead of the durum
PBW114, we have used PBW34 in the evaluation, since
both are susceptible.

20 Chapter 3

Table 3.4. Five synthetic hexaploids selected as
resistant to Karnal bunt under the greenhouse
screening test.

Seed source Percent
identifier Karnal
Durum wheat parent and (Synthetic bunt
synthetic hexaploid I.D. noa) infection

Sora 19.4
Sora/T tauschii 14 0
Altar 84 7.4
Altar 84/T tauschii 23 0
D67.2/P66.270 25.5
D67.2/P66.270// Ttauschii 59 0
Yar'S' 64.2
Yar'S'/T tauschii 125 0
68112/Ward 21.8
68112/Wardl/Ttauschii 101 0

a See Appendix 2.

Triticum aestivum
cv. Pavon
(2n=6x=42, AABBDD)

Triticum aestivum
cv Pak-81

(2n=4x=28; AABB)

Pentaploid F1
(2n=5x=35, AABBD)

Cross I derivatives
(2n=5x=35; AABBD)

Continue up to backcross 5 or 6
as for cross 1 to obtain

Backcross 5 or 6 derivatives
(2n=5x=35; AABBD)

Self and recover progeny of
28 chromosomes with phenotypic
similarity to Pavon

Pavon type but

T tauschii accessions
(2n=2x=14; DD)


F1 hybrids
(2n=3x=21; ABE)

Colchicine double derivatives
(2n=6x=42; AABBO)

1. Screen for tolerance/resistance
2. Use in transfers to other
wheat cultivars

Figure 3.4. Schematic showing extraction of the
AABB component from an elite hexaploid wheat
cultivar, derivation of a synthetic hexaploid by
crossing it to Triticum tauschii (2n=2x=14; DD), and
utilization of the doubled derivative (2n=6x=42;

Interspecific Crosses 21


Intergeneric Crosses:

Hybrid Production and Utilization
Abdul Mujeeb-Kazi

he different gene pools within the annual
and perennial grasses of the Triticeae
Tribe can provide tremendous genetic
variability (Dewey 1984) for improving wheat.
However, in contrast to the Triticum /A. .3,
spp. discussed in the previous chapter, the
species we deal with in our intergeneric crosses
are quite diverse genomically and rather difficult
to cross with wheat. Even when successfully
combined, the resulting hybrids exhibit little or
no intergenomic chromosome association. Hence,
accomplishing beneficial alien transfers through
intergeneric hybridization is quite time-
consuming. Despite these limitations, significant
successes and advancements have been made
over the past 20 years (Kruse 1973; Islam et al.
1981; Sharma and Gill 1983a,b,c; Mujeeb-Kazi
and Kimber 1985; Mujeeb-Kazi et al. 1987, 1989;
Mujeeb-Kazi and Asiedu 1989, 1990; Gill 1989).
The CIMMYT Wide Crosses Section now has a
significant number of hybrids derived from
intergeneric crosses among its genetic stocks
(Appendices 4-6).

This chapter describes the methodologies
currently available for accomplishing successful
intergeneric crosses within the Triticeae and
using the resulting hybrids for wheat
improvement-particularly in the areas of abiotic
stress tolerance (e.g., aluminum, salt, and copper
uptake efficiency). We also anticipate that greater
emphasis will soon be attached to biotic traits
such as resistances to spot blotch
(Helminthosporium sativum syn. Bipolaris

sorokiniana), scab (Fusarium graminearum), Karnal
bunt (Tilletia indica), barley yellow dwarf virus
(BYDV), Septoria spp., and Russian wheat aphid
(Diuraphis noxia).

Hybrid Production

Germplasm used
The germplasm we utilize in our intergeneric
hybridization program includes accessions of
Triticum I, .;,:l .... T. aestivum, Hordeum l.., ,
Secale cereale cv. Prolific, and numerous
perennials mentioned in the sections below.

Mujeeb-Kazi and Rodriguez (1984) described the
procedures involving growth, hybridization,
embryo culturing, somatic analyses, meiosis,
backcross seed production, and amphiploid
induction. The cytological techniques are similar
to those employed by Mujeeb-Kazi and Miranda
(1985). The chromosomal variations reported
extend over investigations that have been in
progress for more than a decade. Mujeeb-Kazi et
al. (1984, 1987) describe the methodology for
situations where the hybrids are produced under
field conditions.

Application of the simplest techniques of
emasculation and pollination used in
conventional wheat breeding accomplished
production of the earliest intergeneric hybrids.
Many important hybrids are still produced with
these techniques. The extensive series of crosses
made by Kihara (1937), his colleagues, and

others can be cited as examples. Kimber and
Abu Bakar (1979) pmvided a tabulation of
hybrids involving wheat and its relatives. These
contributions coupled with the pioneering work
of Kruse (1967, 1969, 1973) led to an increased
momentum in the area of intergeneric
hybridization that extended the range of
combinations among the Triticeae (Sharma and
Gill 1983a,b,c; Mujeeb-Kazi and Bernard 1985a,b;
Mujeeb-Kazi et al. 1987, 1989).

Some intergeneric hybrids, e.g., T. aestivum x
A,,. 1. '/.1 cylindrica, are quite easy to produce;
while others, e.g., Hordeum -l,....' x T. aestivum,
are much more difficult. Some other difficult
intergeneric combinations involve wheat crosses
with Psathymstachys juncea, all 1. I' 'i species,
and the small-anthered Elymus species. Regular
production of inter generic hybrids between
Hordeum and Triticum, which may have been
made as early as 1904 by Farrer, had to await the
discovery by Kruse (1973) that embryos could be
rescued by applying gibberellic acid to the
developing ovule with subsequent culturing of
the embryo on an artificial medium.

Choice of parents
With the intergeneric combinations that have
already been made, choice of parents and cross
direction become the paramount considerations
in producing newer complex hybrids. The low
frequency of the production of viable embryos
in some hybrid combinations indicates the
significance of placing together pollen and
ovules that are genetically compatible. Presently
the only way to determine compatibility of the
parents is actually to make a cross. Therefore, if
difficulty is experienced with a particular hybrid
combination, the only practical solution is to
increase the range of parental genotypes
involved and to attempt making reciprocal
crosses (i.e., the alien species as the female

Brink and Cooper (1940) described and
discussed the effects of particular genotypes,
ploidy level, and the species choice for the
female parent in hybridization. In both
interspecific and intergeneric crosses, it has been
the general practice to use the higher ploidy
species as the female parent because there
appears to be less imbalance between the
chromosome numbers of the embryo and
endosperm and it is generally easier to
emasculate the hexaploid T. aestivum than the
tightly invested florets of most alien species.
However, the early production of hybrids from
wheat x barley crosses was accomplished with
barley (lower chromosome number) as the
female parent. Similarly, Sharma and Gill
(1983a,b,c) Lu-1J 1.. i'' ..' ciliare and A. yezoense
as the female parents in crosses with T aestivum;
and in the following crosses, the alien species
was used as the female parent: A. trachycaulum x
T. aestivum (Mujeeb-Kazi 1980), A. fibrosum x
T. aestivun(Mujeeb-Kazi and Bernard 1982), and
Elymus canadensis x T aestivum (Mujeeb-Kazi and
Bernard 1985a,b). So, it would appear that other
hybrids, which have not been possible to date,
may be recovered if we make the reciprocal
combination. Fertility restoration in such
combinations will vary.

Emasculation procedures can also affect hybrid
production. At CIMMYT, we clip the tops of the
glumes of the female parent and extract the
anthers with forceps (Mujeeb-Kazi and Bernard
1985a,b). Other workers prefer not to clip the
glume tops. At very high temperatures, there
may be some advantage in not clipping, as it
reduces the chance of drying the stigma, which
can be a major problem in some of the wild
species. It is almost always advantageous to
remove the awns.

Intergeneric Crosses 23

Male sterility
The development of cytoplasmic male sterility in
T. aestivum eliminates the need for mechanical
emasculation of the female parent. However, it is
possible that the system producing the male
sterility may also cause sterility in the hybrid or
its derivatives. Male sterility has been induced in
many species by cold treatment of the
developing spike prior to microsporogenesis.
The availability of chemicals that can induce
male sterility without affecting ovule or embryo
development may greatly enhance hybrid

Researchers have used pre-pollination chemical
treatments with varying success to overcome
fertilization barriers. These treatments likely
induce pollen-tube growth, gynoecium
longevity, micropylar barriers, and the delivery
of the male gametes through the pollen tube.
Treatments with immuno-suppressants (Bates et
al. 1976); 2,4-dichlorophenoxyacetic acid (2,4-D)
(Kruse 1974a,b); and gibberellic acid (GA3)
(Larter and Chaubey 1965) have been reported.
GA3 has emerged as the most popular treatment.

Characteristically, a rapid elongation of cells that
produces a fluffy appearance in the stigmata of
Triticeae species indicates receptiveness to
pollination. In addition, the glumes of some
species tend to gape at this time or slightly later.
Pollination is usually done at this time; however,
Mujeeb-Kazi et al. (1984, 1987, 1989) made
previously unattainable hybrids by pollinating
before the stigmata showed the visible
receptivity indicator. It would seem that early
pollination circumvents fertilization barriers that
develop as the stigma matures (Sitch 1984). In
addition, Kruse (pers. comm.) found that the
first pollination with inactivated pollen followed
by normal pollen enhanced the recovery
frequency of hybrids of T. aestivum x H. ,1.. ,.

Kruse (1973) demonstrated the significance of a
post-pollination treatment, which consisted of
one application of a 75-ppm GA3 solution to the
stigma and ovule walls to assist the
developing embryoAlthough up to 10 daily
post-pollination applications have been tried, the
single GA3 application has been equally effective
and, at the same time, it decreases labor and
reduces the risk of the invasion of accidental
out-pollination. In the divergent crosses of wheat
with maize or Tripsacum for polyhaploid
production (see Chapter 5), a 2,4-D post-
pollination treatment is considered to be
essential. We suggest a 2,4-D post-pollination
treatment be tried in other wide hybrids. It is,
however, unlikely that 2,4-D would modify the
effect of crossability genes in such crosses.

Embryo rescue and culture
Embryo excision and culture on artificial media
are the developments that have advanced the
production and utilization of wide hybrids more
than any other techniques. Although various
media are employed in different laboratories, we
use those developed by Murashige and Skoog
(1962) and Taira and Larter (1978).

Embryo rescue and culture are aimed at
removing the embryo aseptically as late as
possible in its development, yet still early
enough to allow its continued development on
artificial media. Endosperm degeneration may
start earlier and seems to be closely related to the
cessation of embryo growth. Characteristically,
embryo development in wide hybrids tends to
slow down about eight days after pollination
and, in 10 to 14 days, the embryo often ceases
development. Embryos of the hybrid derived
from Hordeum ,., ..,. -.. Secale cereale (Fedak
1977a,b) were rescued and successfully cultured
only 12 to 14 days after pollination; however,

24 Chapter 4

most embryo rescues are accomplished 16 to 18
days after pollination under field conditions and
around 20 days under growth chamber or
greenhouse conditions.

Recently developed media (A. Comeau, per.
comm.) will undoubtedly allow the recovery of
even younger, less developed embryos. This area
of study currently seems to attract little
attention, yet the potential of increasing the
range of producible wide hybrids is substantial.
The difficulty of producing mature plants from
hybrids of only a few cells is probably
considerably less than that from artificially
fused somatic cells because some level of
compatibility has already been demonstrated by
the fact that sexual fusion has taken place.
Therefore, it would seem that the improvement
of embryo-rescue techniques would present a
greater potential for utilizing the alien genetic
material found in distant, but related species.

Bridge crosses and polyploidy
We have also obtained wide hybrids through the
use of bridge crosses and the contribution of
polyploidy (Mujeeb-Kazi and Asiedu 1990).
Examples of the former are possible
hybridization of T. aestivum with Heteranthelium
by using the Fl hybrid of the Heteranthelium x
Thinopyrum ,~lgiti, i cross and combining
T. aestivunmx A..i 'i desertorum by crossing
the amphiploid of the A. desertorum x Th. repens
with T aestivum. We have demonstrated the
utility of polyploidy by obtaining the wheat
combination with the Psathyrostachys juncea
combination. With this, the alien-induced
tetraploid source not only facilitates hybrid
production, but also simplifies obtaining the
amphiploid when the self-sterile Fl hybrid
(2n=5x=35, ABDNN or 2n=4x=28,ABNN) is
backcrossed to T aestivum or T. '~. .; i

Cytological verification
Production of an Fl plant does not necessarily
mean a successful intergeneric cross has been
made. Morphological recognition that a wide
hybrid has actually been produced is usually
very unreliable because important characters
may be completely suppressed (Kimber 1983).
Cytological verification is more convincing, but
this too can result in erroneous conclusions
due to:

* Technical problems in the collection and
preparation of material;
* Unreliability of the chosen cytological
* Somatic chromosome elimination or
chimeras; and
* Misinterpretation of meiotic data.

Root tip preparations from plants in plant pots
can provide the first indication of hybridity. We
usually delay root tip collection from a suspected
hybrid for cytological analysis until it is growing
in a pot because of the small number (often only
one) of roots produced in culture. After taking
roots from the pots, the number of dividing cells
is usually less than that found in normal
seedlings that have germinated in a petri dish.
Further, silica particles adhering to the roots can
spoil the preparation of good squashes. Feulgen
or carmine/orcein staining of prefixed cells
validates that a genuine hybrid plant has been
produced. A technique we developed (Mujeeb-
Kazi and Miranda 1985) can result in very clear
preparations from which we can determine
chromosome number, arm ratios, secondary
constrictions, and intergeneric differences in
chromatid thickness (e.g., Thinopyrum/ Secale). If
both parents of a hybrid have the same
chromosome number, somatic chromosome
counts, at best, can only give an indication of

Intergeneric Crosses 25

hybridity if there are large and characteristic
karyotypic differences between chromosomes of
the two parents.

The reliability of a particular cytological
technique must also be considered. If the parents
differ in chromosome number, simple counting
of somatic chromosome number can be reliable.
If the somatic chromosome numbers are the
same, the parents might still differ in their ability
to show C or N bands. The reliability of hybrid
recognition on the basis of chromosome banding
is directly proportional to the number, intensity,
and chromosomal distribution of the bands. If
the arm ratio of the chromosomes of the parents
is sufficiently different, this too may be used as
an indication of hybridity. Again, the reliability
of the arm ratio as an indicator of hybridity
depends on the number of easily recognized
differences between the parents. The presence or
absence of secondary constrictions is not a good
method of recognizing hybrids since
amphiplasty does contribute to suppression.
Resolution of secondary constrictions requires
that superior somatic preparations be made and
that no nucleolar organizer competition prevails.
The length of somatic chromosomes generally
provides a very poor method for the
identification of hybridity. Low reliability can be
attributed to: 1) the general absence of large
differences in relative chromosome length in the
Triticeae and 2) the inaccuracies involved in
measuring chromosome length (Kimber 1970).

Even if the zygote of an intergeneric cross is
recovered, it is still possible that the seedling
may not be a hybrid due to chromosome
elimination in the early zygotic divisions.
Haploid barleys and polyhaploid wheats,
produced by pollination of wheat by Hordeum
bulbosum and Zea mays, are examples of this type

(Kasha and Kao 1970, Barclay 1975; see
upcoming section on genome elimination and
Chapter 5). The spontaneous production of
chimeras (cells with different chromosome
numbers) may also hinder the recognition of
hybrids or make their utilization more difficult.
Several authors have recognized chimeras. For
example, Kasha and Sadasivaiah (1971) recorded
the expected chromosome number in only 41I,
of the somatic cells of a diploid hybrid of
Hordeum I..i,. -.. H. bulbosum.

Clearly, when such difficulties abound, any
claim of hybridity must be accompanied by rigid
chromosome analysis. The interpretation of such
analysis is quite important because both species
relationships and the choice of the most suitable
method for the introduction of alien variation
depend on the ability of the chromosomes to
pair. In the earliest work, although judgments
were largely subjective, correct conclusions were
still often reached (Lilienfeld 1951). More
recently, numerical methods for the analysis of
meiosis in hybrids have provided some
objectivity in determining genomic relationships
(Kimber et al. 1981, Alonso and Kimber 1981,
Kimber and Alonso 1981, Espinasse and Kimber
1981, Kimber and Pignone 1982).

Ideally genomic analysis is conducted using a
triploid hybrid that results from a cross between
a tetraploid plant and a diploid analyzer;
however, analyses are often made at other
polyploid levels. In general, diploid hybrids can
provide little if any genomic information, for
there must be competition for chromosome
pairing partners in order to recognize differences
in the genomes present. The pairing patterns at
higher levels of polyploidy can be very
confusing because of the large number of pairing
possibilities between both the homologous (if
present) and homoeologous chromosomes

26 Chapter 4

within any homoeologous group. These practical
limitations result in a useful range of triploid to
pentaploid hybrids from which we can reliably
obtain information. If telocentric chromosomes
are available, they can provide unequivocal
information about the frequency with which
particular chromosomes are pairing, but their
usefulness is usually limited to measurements of
relationships with the A, B, and D genomes of
T. aestivum

Clear proof of hybridity can only come from
chromosome analysis of the supposed hybrids.
The investigation of species relationships from
backcrossed hybrids can, in some cases, be
accomplished; however, complications
introduced by the production of unreduced
gametes or the random elimination of
chromosomes can give rise to incorrect

Chromosomal variations
Chromosomal variations in intergeneric hybrids
among the Triticeae occur fairly consistently and
at various phases in the formation and the
development of the hybrid. To a certain extent,
the events are fortuitous, but in other situations
the variation effects may pose a serious
constraint in systematic wide crossing programs.
Stable amphiploids are a prerequisite for the
development of alien chromosome addition
lines. The production of these lines could be
hindered greatly if amphiploids cannot be
produced. This forces researchers to adopt the F1
self-sterile based route to generate F1 backcross I
(BCI) progeny that has the potential to induce
alien structural chromosomal modifications. In
the "shot-gun" approach, chromosomal
variations have a decided breeding advantage,
more so for complex polygenically controlled
characters. The occurrence of these variations
augments those associated with callus culture

and mutagenesis. During the cytological
analyses of some intergeneric hybrids or of their
advanced derivatives, we have observed unique
chromosomal variations.

Genome elimination
A normal F1 hybrid resulting from an
intergeneric cross possesses half the
chromosome number of each parent involved in
the combination. In some situations, however,
the alien genome may be totally or partially
eliminated, which results in the production of
polyhaploid or aneuploid F1 hybrids. This
phenomenon has been observed in a number of
hybrids resulting from intergeneric crosses
using species within the Titiceae as both parents
and in crosses where one parent is outside of
the iliticeae.

Crosses within the Triticeae include:
Hordeum x Titicum (Kruse 1974a,b; Mujeeb-Kazi
et al. 1978), Triticum x Hordeum (Fedak 1980,
Finch and Bennett 1980, Islam et al. 1981),
Hordeum x Secale (Kruse 1967, Fedak 1977a),
Triticum x Elymus (Mujeeb-Kazi and Bernard
1985a), Agropyron x Triticum (Mujeeb-Kazi and
Bernard 1985a,b), and Triticum x 1. "I. i.n
(Mujeeb-Kazi, unpublished). Crosses involving
a paant outside the Triticeae include:
T. aestivunx Zea miip4Laurie and Bennett 1986),
T. aestivum x Sorghum bicolor (Laurie and Bennett
1988b, Ohkawa et al. 1992), T aestivum x Z. mays
ssp. mexicana (Ushiyama et al. 1991), and
T. aestivumn Tripsacum dactyloides (Riera-
Lizarazu and Mujeeb-Kazi 1993).

The most widely documented and high
frequency method of polyhaploid production
involves the hybridization of T. aestivum with
Hordeum bulbosum (Barclay 1975). Wheat
polyhaploids were obtained from pollination of

Intergeneric Crosses 27

T. aestivum cv. Chinese Spring (CS) by both
diploid and tetraploid forms of H. bulbosum.
Studies on the early embryo and endosperm
development of these hybrids demonstrated that
fertilization was followed by the elimination of
the H. bulbosum chromosomes during
subsequent divisions of the embryo and
endosperm. The same phenomenon has been
observed in both diploid and tetraploid
reciprocal H. l..i.,,. H. bulbosum
hybridizations, resulting in a high frequency of
H.. i.., .. haploids (Davis 1960, Kao and Kasha
1969, Bennett et al. 1976).

We do not know what controls genome
elimination in T. aestivum x H. bulbosum
hybridizations. However, studies of chromosome
elimination in H. ,l.., ,. H. bulbosum
hybridizations have revealed that many factors
are involved. Three genes have been identified
and located in chromosome arms 2S, 2L, and 3S
of H.. ,. ..,.. which control H. bulbosum genome
elimination (Barclay et al. 1972, Ho 1972, Ho and
Kasha 1974). In addition, the H. bulbosum
genotype (Pickering 1979, Simpson et al. 1980,
Fukuyama and Hosoya 1983), the H.. l,..i,,.'
genotype (Simpson et al. 1980), and the balance
between the parental genomes appear to be

Pollination of T. ventricosum by tetraploid
H. bulbosunhas also resulted in polyhaploids of
T. ventricosunbecause the H. bulbosum genomes
were eliminated (Fedak 1983).

Low frequencies of genome elimination have
been observed in other Triticum x Hordeum
hybridizations, which involve species other than
H. bulbosum. Within the 20 Fl progeny of
hybridization between CS and H. i.,. ,.- .cv.
Betzes, Islam et al. (1981) identified three
21-chromosome plants that phenotypically

resembled wheat polyhaploids, implicating the
elimination of the H.. I. l.. ,.'genome. Fedak
(1980) obtained one polyhaploid from the same
hybrid combination. Finch and Bennett (1982)
obtained a particularly high frequency I'1, of
wheat polyhaploids among the progeny of a
cross between T. aestivum cv. TH3929 and
H. r,..,r i.v = 2x=14) Paavo Line P-4. The
elimination of the H. rl.., .. genome was
suspected due to variations in the somatic
chromosome number between and within a
number of nonviable embryo-derived plantlets;
these variations ranged from the polyhaploid
(n=3x=21) to the hybrid chromosome number

Genome elimination and the consequent
production of wheat polyhaploids have also
been observed in the reciprocal hybridization,
H. I...,r -.. T. aestivum (Kruse 1974a,b), in our
own studies, and in the cross between H.. ,1....
x T. 'i.,: ;.1, .. (Mujeeb-Kazi et al. 1978).
Hybridizations between H. -i, ..,.' and S. cereale
have also produced a low frequency of
polyhaploid H. i,'.. ,. pi, ..-i following the
elimination of the rye genome. Kruse (1967)
reported the production of one polyhaploid and
two hybrid progenies from this intergeneric
hybridization. Later, Fedak (1977a) noted the
production of barley haploids at a frequency of
less than I in barley x rye hybridizations.

We obtained a 21-chromosome Fl progeny with
a wheat polyhaploid phenotype from
intergeneric hybridizations of T. aestivum x
Elymus !".i i. '' .. i. (Mujeeb-Kazi and Bernard
1985a,b). From the reciprocal cross of E.fibrosus x
T. aestivum, we obtained a polyhaploid of
E. fibasus (Mujeeb-Kazi 1994a). Laurie and
Bennett (1986) observed elimination of the maize
genome six days after a T aestivum x Zea mays
hybridization, which has provided an alternative

28 Chapter 4

for producing T. aestivum polyhaploids; this
technique, which has had up to a 4.4% recovery
of wheat polyhaploid plants (Laurie and Bennett
1988c), functions satisfactorily and is
independent of the krl kr2 loci (Laurie and
Bennett 1987). A modification of this technique,
which uses 2,4-D and an altered mode of
application, has significantly improved the
recovery of wheat polyhaploids (Inagaki and
Tahir 1990, Riera-Lizarazu and Mujeeb-Kazi
1990, Suenaga and Nakajima 1989). Fertilization
of T aestivum by S. bicolor has had a,; -* success
rate following which elimination of the S. bicolor
chromosomes occurs. Polyhaploid wheat plants
have since been recovered (Ohkawa et al. 1992).

Aneuploid Fl progeny
Reports of the occurrence of aneuploid Fl
progeny from intergeneric hybridizations
involving the Triticeae are restricted to crosses
between T. aestivum and H.. ,1. .i,.' (Islam et al.
1981). Meiotic instability within complete Fl
hybrids has, however, been more widely
reported, namely within Triticum x Hordeum
reciprocal hybrids (Fedak 1977b, 1980; Islam et
al. 1981; Mujeeb-Kazi and Rodriguez 1983a,b)
and within Hordeum x Secale hybrids (Finch and
Bennett 1980).

Islam et al. (1981) reported a high frequency of
aneuploid Fl progeny from a cross between CS
and H.. ,1..i,,.' cv. Betzes. Of 20 Fl progeny, 19
showed nonhybrid chromosome numbers
ranging from 21 to 36. Meiotic analyses of these
plants indicated that aneuploidy in five plants
was associated with the addition of 1, 2, 3, 4, or 6
H.. -l..i, C ii', .. 'ii-. to the polyhaploid
wheat genome. Islam et al. (1981) postulated that
this was due to spindle abnormalities at the early
zygotic divisions of the hybrid embryo, causing
the elimination of one or more H.. i,...-..'
chromosomes. In 11 plants, the duplication and/

or deficiency of wheat and barley chromosomes
were apparent. For example, one 35-
chromosome plant showed a meiotic pairing
pattern of 1511 + 51. Although the 28-
chromosome Fl hybrid resulting from the wheat
x barley cross exhibited stability in the somatic
cells and in the majority of the pollen mother
cells, some mosaic cells with aneuploid and
polyploid chromosome numbers were observed,
which indicated meiotic instability in this
complete hybrid (Islam et al. 1981). Similarly,
from an examination of the meiosis of complete
H. ,.1... .. T. aestivum Fl hybrids, we found
meiotic instability and the precocious separation
of five or six chromosomes at meiosis
(Mujeeb-Kazi and Rodriguez 1983a).

We also observed meiotic instability in H.. ,...,..
x T. ', ;1, .... Fl hybrids (Mujeeb-Kazi and
Rodriguez 1983b) in the form of hyperploid and
hypoploid pollen mother cells, aneuploid cells
showing trivalent and quadrivalent associations,
and premature chromosome separation. We
speculate that the chromosomes showing
premature disjunction may undergo synapsis
without the formation of chiasmata, resulting in
more rapid chromosome separation. Fedak
(1980) observed a differential degree of meiotic
instability in reciprocal H.. rl,..-, ,. T. aestivum
hybrids, the frequency of euploid pollen mother
cells being lower in Fl hybrids that possess
H.. l..,r c, ,,.p1.-m Fedak (1979) also observed
a number of meiotic abnormalities, namely
unequal disjunction, multipolar mitoses, and the
presence of isochromosomes, which implies
misdivision at anaphase I.

Finch and Bennett (1980) observed similar
meiotic instability in hybrids derived from
H. jubatunssp. breviaristatum x S. africanum,
which had hyperploid pollen mother cells
containing 22 to 32 chromosomes and

Intergeneric Crosses 29

occasionally 47 chromosomes. No aneuploid F1
hybrids have been reported in this or other
Hordeum x Secale hybridizations. In a triploid
hybrid (2n=3x=21) obtained from the cross of
H. l .. ,r Betzes x S. vavilovii, the mean
somatic chromosome number was 19.7 with a
range of 7 to 24 between individual cells. In
meiocytes, the mean chromosome number was
18.3 with a range of 14 to 26 (Fedak and
Nakamura 1982).

F1 hybrids produced from a cross between CS
and the hybrid resulting from the intergeneric
cross of Thinopyrum repens/A. desertorum also
showed a high degree of aneuploidy. We found
that chromosome numbers of the F1 progeny
ranged from 2n=35 to 2n=57 where the expected
complete chromosome complement was
2n=8x=56. In the 35-chromosome hybrids, the
elimination of three genomes had presumably
occurred (Mujeeb-Kazi et al. 1989).

Although rare, we found chromosome
irregularities in F1 hybrids of H.. I. ..',.. x
E. canadensisvhere meiocytes at metaphase I
possessed chromosomal compositions exceeding
the normal 2n=3x=21 complement (Mujeeb-Kazi
and Rodriguez 1982).

Variations in backcross I (BCI)
Aneuploid BCI progeny-In wide crosses, the
self-sterile F1 hybrids, upon colchicine
treatment, classically result in fertile
amphiploids that may then have practical utility.
Triticale (X Triticosecale Wittmack) is a
noteworthy result of such a process-both at the
hexaploid and octoploid polyploidy levels. In
other cases, the fertile amphiploids are sources of
BCI derivatives (amphiploid/Triticum source),
from which the eventual production of alien
disomic addition lines-after cytogenetic

manipulations-can lead to subtle alien genetic
transfers. The BCI derivatives are generally
expected to have normal wheat and normal alien
chromosome complements-the former in a
double dosage and the latter in a single dosage.

Alternatively, the F1 hybrid can be pollinated by
wheat to produce the BCI derivatives in a low
frequency. This occurs via fusion of the wheat
pollen with an unreduced egg cell of the F1
hybrid. This common procedure is a rapid way
of meeting applied research goals, although it is
beset with considerable aneuploidy that is
maternally contributed. The unreduced egg cell
could be an assemblage of wheat/wheat, wheat/
alien, or alien/alien translocations and may have
drastic aneuploid changes to be expressed as
hyperploid or hypoploid progeny in the BCI

Hybridization of T. aestivum with decaploid
(2n=10x=70) Th. ,higatitl, (E_ ij, .:.:. i pontica)
produced F1 hybrids with 56 chromosomes.
Subsequent pollination of these Fls with
T. aestivunwas expected to produce BCI
progeny with 77 chromosomes. The chromosome
numbers in 143 BCI plants ranged from 42 to 62
chromosomes and N-band analyses indicated a
random variation for the presence of specific
maternal chromosomes. As expected, we
obtained plants that were disomic for
chromosome 5B, but several plants were mono-
5B which, upon selfing or further backcrossing,
may lead to selfed nulli-5B derivatives
(Table 4.). These derivatives may enhance
recombination events between wheat and alien
species. We found a similar trend in BCI
derivatives of the hybrid of T. aestivum/ Aegilops
variabilis (Jewell and Mujeeb-Kazi 1982).
Variation in chromosome number of the BCI
appears to be common in other intergeneric
hybrids (Mujeeb-Kazi and Bernard 1982,

30 Chapter 4

1985a,b), but such events seem independent of
alien genomic composition. We have found a
high frequency of 56-chromosome BCI
derivatives for the T. aestivum /Th. scirpeum/ /
T. aestivuncross (Mujeeb-Kazi and Miranda
1984) where Th. scirpeum contributes the E1E2
genomes to the above BCI derivatives with the
AABBDD wheat genomic combination; Sharma
et al. (1987) obtained similar results. Replacing
Th. scirpeum with the tetraploid Th. curvifolium,
which also has the E1E2 genomes, in the above
cross results in a far lower frequency of 56-
chromosome BCI progeny. This not only creates
doubt about the E1E2 genomic designation in
Th. scirpeunand Th. curvifolium, but also
presumably reflects the varied effects of wheat's
regulator genes for meiotic pairing. The expected
chromosome count of a BCI derivative should
not be considered normal because there is ample
opportunity for structural changes to arise in
both the wheat and alien genome(s) contributed
to the F1 hybrid. The applied aspects of a wide
cross program are not hampered by this BCI
variability because the chromosomal
reorganizations may actually be advantageous
for breeding. Most F1 hybrids derived from
intergeneric crosses do not exhibit enough
chromosome association to indicate a direct alien
transfer. Aspects of induced pairing
manipulation do promote such F1 T aestivum/

alien chromosomal associations, but the BCI
derivatives (F1 hybrid x T aestivum) are highly
aneuploid (Asiedu et al. 1989) and extremely
valuable-an example of exploitation of

The BCI variations are often a constraint to
producing alien addition lines. We believe that
producing amphiploids prior to BCI formation
can alleviate this problem (ter Kuile et al. 1988).
However, disomic addition lines of H.. il..,..' to
wheat have been produced (Islam et al. 1975,
1978, 1981) through the BCI approach when
amphiploids of T aestivum/H. r 1.., .- could not
be produced. Thus, each hybrid should be
regarded as a unique entity but we feel that so-
called normal BCI derivatives will be obtained if
a large BCI population is produced.

Apomixis-In traditional intergeneric crosses
(Mujeeb-Kazi 1994b), the F1 hybrid is always
self-sterile. Upon placing wheat pollen on the
self-sterile F1 spikes, BCI seed set does occur at
reasonable frequencies, which leads to normal or
aneuploid progeny. In rare situations, the BCI
progenies have a chromosomal complement
identical to the F1 hybrid and are classified as
apomictic (Mujeeb-Kazi 1981a,b). Such a
phenomenon has occurred more often with the
BCIs of H. I.., iT,../ 'I, .; 6... than with those of

Table 4.1. Some backcross derivatives (BCI) from Triticum aestivum/Thinopyrum elongatum(2n=10x=70)//
T. aestivum with aneuploid chromosome numbers and some N-banded chromosome compositions.

BCI derivatives Some N-banded chromosomes
with chromosomal
variations 1B 2B 3B 4B 5B 6B 7B 4A 7A

50 chromosomes + + + + + + + + +
+ + + +

40 chromosomes + + + + + + + + +
+ + + +

49 chromosomes + + + + + + + + +
+ + + + +

Intergeneric Crosses 31

H. i. l.. ,. 'T. aestivum (Table 4.2). In our initial
BCI crosses, we used the same wheat variety as
was used in the production of the F1 hybrid, so
we were not able to categorically state that the
progenies were apomictic. There was a
possibility that all the maternal wheat
chromosomes were eliminated, then the so-
called 'apomictic progeny' could have resulted
from the fertilization of the egg-cell (with seven
barley chromosomes) by wheat pollen. In
subsequent BCI crosses, we used different wheat
varieties and utilized awn markers to
unequivocally demonstrate that apomixis was
indeed occurring (Mujeeb-Kazi 1994b). We also
used a trigeneric system to further document the
findings in the T. aestivum/Leymus racemosus / /
Th. lon,,ga'tio cross where, in addition to the
expected 70-chromosome progeny (21+14+35),
apomictic 35-chromosome plants resulted with

Table 4.2. Apomictic progeny in backcross I derivatives
H. vulgare/2* T aestivum with cytological details.

21+14 chromosomes whose banding and meiotic
pairing were similar to those we found
independently in the F1 hybrid of T aestivum
L. i, iitolAl.tNiMujeeb-Kazi et al. 1983).

Doubled egg cell-A doubled chromosome
number in the egg cell of the hybrid is rare-
considering the wide range of hybrids we
produce (Mujeeb-Kazi 1984, Mujeeb-Kazi and
Bernard 1982). However, we did observe it in a
BCI of the Th. .. ..i.. i. 1 '. cereale combination.
The F1 possessed 21 chromosomes where the
seven rye chromosomes could be identified on
the basis of their larger size, although C-banding
also aided identification. We pollinated the F1 of
Th. i ..ir... '- cereale cross with diploid S.
cereale; the resulting BCI derivative possessed 49
chromosomes (Figures 4.1a,b). We inferred this
to be the product of the fertilization of a

of Hordeum vulgare/2* Triticum turgidum and

Meiotic association at metaphase I
chromosome I II II III IV VI
Cross Combination number Rods Rings
H. vulgare/T turgidum 21 19.3 0.4 0.5 0.02
H. vulgare/2* T turgidum 21 19.3 0.4 0.5 0.03
H. vulgare/T aestivum 28 23.9 1.4 0.5 0.1 0.03 0.01
H. vulgare/2* T aestivum 28 23.9 1.4 0.4 0.1 0.04

a b

S- rr'OP
,I "

Figure 4.1. Somatic details at metaphase for: a) Thinopyrum elongatuniSecale cerealeF1 (2n=3x=21), and b)
Th. elongatuml2* S. cereale backcross I with 49 chromosomes (2n=7x=49).

32 Chapter 4

spontaneously doubled egg cell (28+14) by
seven-chromosome S. cereale pollen.
Chromosome size and C-banding indicated this
to be the case. Initially, determining chromosome
size differences often led to erroneous
conclusions (Mujeeb-Kazi and Kimber 1985), but
a procedure we developed in 1985 eliminates
this problem (Mujeeb-Kazi and Miranda 1985).

Selfed progenies of backcross I
derivatives; BCIF1 and production of
modified genomes
Complete synthetic genomes-The F1 hybrids
of crosses between T. aestivum and Th. junceiforme
(2n=4x=28), Th. curvifolium (2n=4x=28), and
Th. scirpeun(2n=4x=28) possess 35 chromosomes
and are self-sterile. Upon pollination of these Fls
with wheat, seed set occurred; somatic analyses
indicated a predominance of 56-chromosome
progenies for Th. scirpeum; for Th. junceiforme and
Th. curvifolium, there were some variations
around the expected number of 56. Some spikes
on each BCI derivative were covered with
glassine bags; seed set took place on these
spikes, proving them to be self-fertile (BCIF1).
The seed set frequency was Th. scirpeum >
Th. curvifoliunm Th. junceiforme. Somatic analysis
showed these seeds to possess 56 chromosomes
(Figure 4.2d) or near this number (Figures 4.2a,
b, c). Chromosomal associations ranged from
high bivalency (Figures 4.2e, f) to 21 bivalents
and 14 univalents. The meiosis in the 56-
chromosome derivatives of T aestivuni
Th. scirpeur/ IT. aestivunwas the most
organized due to the high pairing. Sharma et al.
(1987) further substantiated this where the
multivalency they observed was attributed to
genomic structural variations. Where 14
univalents were observed, extremely low seed
set led to the perpetuation of this BCIF1

This self-fertility phenomenon in the three
partial autopolyploids is indicative of a genomic
composition that may provide novel options for
plant-level genetic manipulation. Repeated
selfings of the partial autopolyploid-based BCIF1
derivatives could lead to synthetic genome
formation, i.e., a means of aggregating complex
polygenic recessive traits in a modified package
(Mujeeb-Kazi and Miranda 1984).

After repeated selfings of the BCI, additional
backcrosses to wheat resulted in derivatives that
predominantly had 49 chromosomes but some
variation as well in chromosome number, i.e., 48
(Figure 4.3a), associated as 19 bivalents + 7
univalents + 1 trivalent (Figure 4.3b), or as 20
bivalents + 8 univalents (Figure 4.3c) for
Th. scirpeumbased progeny. In the near 49-
chromosome Th. curvifoliumor Th. junceiforme-
based progeny, aneuploidy was more
pronounced. It is our contention that addition
lines from selfed Th. scirpeunBCIF D will be
different than those from where the BCI is
directly advanced to BCII or BCIII, etc., i.e.,
devoid of several BCI selfing steps. We are
currently testing this theory and we should be
able to validate it in all tetraploids where
aneuploidy is not noticeable. Other tetraploids
that respond to the synthetic genome concept are
hybrids of crosses between T. aestivurand
Th. distichumiTh. n...,... ;, Th. scythicupand
Th. repens/A. desertorum (a 35-chromosome Fl)
(Mujeeb-Kazi et al. 1987, 1989).

Asymmetric synthetic genomes-In segmental
allohexaploids (Th. junceum; 2n=6x=42) or
segmental autoallohexaploids (Th. intermedium
and synonymous species), there are two related
genomes (E E2) and a distinctly different third
genome (X or Z). Taking Th. intermedium and its
relatives as an example, the genomic constitution
of a hybrid resulting from a cross with wheat
would be ABDEE,2X or ABDE E2Z. When such

Intergeneric Crosses 33

self-sterile F1 hybrids are pollinated with wheat,
the BCI derivatives usually possess 63
chromosomes and have a genomic make-up of
AABBDDE1E2X. These are partially self-fertile
and the BCIF1 progenies as well as progenies of
additional selfings usually possess derivatives
with 56 chromosomes. We believe that an
asymmetric genome loss occurs in the BCI
octoploid (2n=8x=56) yielding BCIF1 progeny.

This prevalent phenomenon has advantages and
disadvantages. It has been described as a
mechanism known as "genome splitting" or
"asymmetric genome reduction" by Gottschalk
(1971), Cauderon (1977), Ladizinsky and
Fainstein (1978), and Dewey (1980). These
workers provided information concerning the
extraction of different genomes of the complex
polypl .iJ I. I .. species, which facilitated

Figure 4.2. Mitotic and meiotic chromosomal details of some backcross I derivatives of Triticum aestivum/
Thinopyrum scirpeum/IT aestivum shown in a somatic cell with a) 51 chromsomes, b) 53 chromosomes, c) 56
chromosomes including a telocentric, and d) 56 chromosomes and in a meiocyte with e) 6 univalents + 12 rod
bivalents + 13 bivalents at metaphase I and f) 6 univalents + 11 rod bivalents + 14 ring bivalents at metaphase I.

34 Chapter 4

cytogenetic analysis as conventionally done for a
diploid alien species. The data presented by the
above workers show a range of variation that is
explained by the genome splitting phenomenon,
which we do not deal with here.

Variations in advanced backcross
In analyses of BC progeny of T aestivum /
L. racemosud /n*T. aestivum, unusual
chromosome number variants were observed. In
one instance, wheat polyhaploids originated as
twin seedlings in the selfed progeny of a
44-chromosome double monosomic addition
line. N-banding revealed 16 typical wheat
chromosomes in the polyhaploids. Other
variants in the backcross progeny included
mixoploids with 43 and 86 chromosomes.

After selling of the BC derivatives (with 49
chromosomes) of T. aestivum/Th. junceiforme / /
2*T. aestivum, 56-chromosome derivatives were
recovered. One explanation implies fertilization
of a 28-chromosome egg cell with pollen that
maintained a high number of Th. junceiforme
chromosomes in the transmission. Such events
have also been observed in BCIIF1 derivatives of

wheat crossed with Th. I ... ... ;, Th. curvifolium,
and Th. scirpeum. In a similar way, Cauderon
(1963, 1977) obtained a 56-chromosome
amphiploid from a T. aestivum /Th. intermedium/
/T. aestivum selfed derivative, although the
transmitted genome was unidentified. The
occurrence is common in the 49-chromosome
BCI derivatives of T. aestivum/ Th. bessarabicum/ /
T. aestivum. These BCI plants are fairly self-fertile
and, upon analysis, the BCIF1 progeny shows a
range in chromosome number from 23 to 51.
This provides an opportunity to obtain
homologous alien chromosomes as multiple
disomic additions (Mujeeb-Kazi et al. 1988);
thereby complex polygenes can be accumulated;
an advantage commonly associated with
induced amphiploids or spontaneously
doubled pcgeny.

Under certain conditions, individual alien
chromosomes may be preferentially substituted
for wheat chromosomes. Such chromosomes
may compensate adequately for the wheat
chromosome that is being replaced.
Chromosome F of L. racemosus replaced 6B6B of
wheat (Mujeeb-Kazi et al. 1983) and we later
observed that chromosome M of L. racemosus




I 'k


Figure 4.3. A hypoploid backcross II derivative from Triticum aestivum/Thinopyrum scirpeum/R* T aestivum
showing in: a) 48 somatic chromosomes, b) 7 univalents + 4 rod bivalents + 15 ring bivalents + 1 trivalent at
metaphase I, and c) 8 univalents + 3 rod bivalents + 17 ring bivalents at metaphase I.

Intergeneric Crosses 35

I;a re

`Y f
Ile I
~it r


replaced chromosome 6D6D of wheat (Figure
4.4). We are critically analyzing these
replacements using ditelocentric wheat stocks of
chromosomes 6DS or 6DL. In another derivative,
chromosome M disomically replaced 6A6A-this
was verified by electrophoretic analysis
(unpublished data).

Upon further backcrossing, the preferentially
substituted chromosomes enable the
development of two univalents in the derivative
progeny (M+6A or M+6D) that could, via centric
break and fusion, form new derivatives through
translocation of the alien chromosomes to wheat.
It was presumably such a univalent misdivision
and its selling that formed the natural

homozygous 1BL/ 1RS translocation in
T. aestivumnarieties that has proven to be so
valuable (Merker 1982, Mujeeb-Kazi 1982,
Rajaram et al. 1983, Jahan et al. 1990).

Variations in amphiploid maintenance
There are not too many examples of fertile
amphiploids in diverse wide crosses presumably
because few researchers have an interest in
pursuing their production. Also, there are some
genetic constraints despite the availability of an
efficient technique. Fertile amphiploids have
been obtained in the following diverse crosses:
T. aestivuriTh. bessarabicum (2x) (Forster and
Miller 1985, Kimber and Sallee 1980),
T. aestivunrHordeum chilense (Chapman and

Figure 4.4. a) Spike showing a C-banded chromosome M of Leymus racemosus (inset) that spontaneously
substitutes for chromosome 6A6A or 6D6D of T. aestivum and b) well developed grains produced as a result.

36 Chapter 4

Miller 1978), H. '., ... /.i/ T. timopheevii (Kimber
and Sallee 1976), T. aestivum/Thinopyrum
distichum (Pienaar 1980), H. pubiflorum /Secale
africanum (Fedak 1985), and H. californicum /
T. aestivun(Fedak, unpublished). Although
numerous efforts failed to produce fertile
amphiploids from the T aestivum /H. i ..-,..
cross (Fedak 1977b; Islam et al. 1975, 1978, 1981;
Mujeeb-Kazi and Rodriguez 1983b), success was
finally achieved (Molnar-Lang and Sutka 1993).
Mujeeb-Kazi et al. (1987) successfully produced
fertile amphiploids from the T. aestivum/
Th. i.. ::; ... cross and earlier from T. aestivum/
Th. junceum cross (Mujeeb-Kazi and Bernard
1985a). These amphiploids usually possessed 70
and 84 chromosomes, respectively, but in both
amphiploids, aneuploidy was observed. Five
amphiploid plants derived from the T. aestivum/
Th. junceum cross had from 72 to 84
chromosomes. Despite a high degree of
asynapsis (up to six univalents) in the
H. californicunT T aestivum-derived amphiploids,
progenies studied so far have had the expected
chromosome number of 2n=8x=56 (Fedak,
unpublished). Recently, several primary triticale
hexaploids have been produced that express a
high degree of normal chromosome counts over
three generations of selling (unpublished data)
and, contrastingly, we have reported several
amphiploids derived from intergeneric crosses
that show variable extents of aneuploidy
(Mujeeb-Kazi et al. 1994c).

We found a rare example of an amphiploid
derived from Elymus fibrosus/T. 'I, ;1,;, ,. where
the alien species was the maternal parent
(Mujeeb-Kazi 1994a). E. trachycaulum, E. ciliare,
and E. caninus have also been used as the female
parents in intergeneric crosses with wheat, but
no amphiploids have resulted after colchicine
treatment (Sharma and Gill 1981, 1983a,b,c;
Mujeeb-Kazi and Bernard 1985a; Sharma and

Baenzinger 1986). The same is true for hybrids of
Hordeum I...,. 'T. .,, 1 ,.. and H.. i. .'/
T. aestivum

In essence, each hybrid combination must be
treated individually and generalities should be
avoided as illustrated above. The aneuploidy
referred to above is for limited individual
chromosomes (Figures 4.5a-d), but complete
genome(s) may be lost in the advance to
amphiploids. There are several examples in the
range grasses (Dewey 1980) where 10- to 12-
ploid amphiploids have spontaneously
stabilized at the octoploid level of 56
chromosomes. A T. 'n, .;, 61r /Leymus mollis-
derived amphiploid was stabilized at 42
chromosomes (N.V. Tsitsin; information to A.
Mujeeb-Kazi via D. Dewey, Logan, Utah, USA,
and A. Merker, Svalov, Sweden).


Although, as mentioned earlier, there are
numerous traits to be utilized from intergeneric
crosses in the areas of biotic and abiotic stresses,
we are particularly enthusiastic about
introgressing tolerance to various abiotic stresses
and emphasize this aspect in our discussion of
the utilization of intergeneric hybrids. We briefy
discuss some of our successes involving biotic
(disease) stress resistances in the conclusion of
this chapter and in the section on the radical
approach in Chapter 9.

Substantial information is available on the
contribution of alien species to stress tolerance in
wheat (Mujeeb-Kazi et al. 1991a, Manyowa and
Miller 1991, Paull et al. 1991). Many of these
attributes require situation-specific improvement
strategies. We have been specifically working
with the transfer of traits related to salt
tolerance, copper efficiency, and aluminum

Intergeneric Crosses 37

Salt tolerance
Through a collaborative research program
involving CIMMYT; the Institute of Plant Science
Research, Norwich, UK; and the laboratory at
Bangor, Wales, we have been able to identify
several salt-tolerant alien genera and
conventional sources, further substantiated by
other literature reports (Table 4.3). This
cumulative information has led to the
formulation of a comprehensive list of alien
variability. Targeted alien species with salt
tolerance include Thinopyrum bessarabicum and
Th. .loi,aiimni; both are diploids (2n=2x=14). An
additional source of salt tolerance includes
selected accessions of Triticum tauschii. Mujeeb-
Kazi et al. (1987, 1989) and Mujeeb-Kazi and
Asiedu (1990) document the hybrid production

and the advance procedures involved. We
closely follow the hydroculture salinity
screening methodology for advanced derivatives
adopted in the Bangor Laboratory (Gorham et al.
1987, Gorham 1990).

Tolerance in conventional germplasm-
CIMMYT has established a tester set that
contains wheat cultivars that are either salt-
tolerant (CS, Kharchia 65, Shorawaki, and
Lu26S) or salt-susceptible (Yecora 70, Oasis F86,
and PBW 34). Some of these entries like CS have
a long history of being classified as salt-tolerant
and have been extensively studied in several
laboratories. Tolerant Shorawaki has had little
publicity; some cultivars (Sakha 8, SNH-9, WH-
157, and Candeal) still require rigid evaluations.



4 j ,



% .

s 7/V
I m q

Nc ,

,, ,,

*1 j O


" Ift j1' I

n~ hII


Figure 4.5. Aneuploidy in amphiploid advance of Triticum turgidum L. cv. Laru/Aegilops variabilis: a) 55
chromosomes + 1 telocentric, b) 56 chromosomes + 1 telocentric, c) 57 chromosomes, and d) 56
chromosomes. The normal chromosome number expectation of this amphiploid is 2n=8x=56.

38 Chapter 4



We plan to study a new Indian release
(KRL 1-4)-a derivative fwm a cross involving
Kharchia 65, considered an elite cultivar for
saline-sodic soils-reported to have superb
growth and high yield even at pH 9.6
(K.N. Singh, pers. comm.).

Tolerance in alien germplasm-With such a
limited source of salt tolerance in conventional
germplasm, it is prudent to search for additional
genetic diversity in the alien species. There has
been a flurry of reports (especially over the past
five years) about alien germplasm possessing
salt tolerance. How close we might be in

receiving benefits from use of such alien species
is still an open question. However, we believe
the prognosis is quite encouraging. As
mentioned above, Th. clon ,igal-i (2n=2x=14) and
Th. bessarabicum (2n=2x=14) are particular
standouts (Gorham et al. 1985, Dvorak et al.

In hydroculture tests, we re-evaluated some salt-
tolerant characteristics of these species and our
observations support their potential as assessed
through their amphiploids derived from crosses
with T aestivum cultivars (Table 4.4). Even
though diploid alien sources are a priority we

Table 4.3. Triticeae relatives with a promise for salinity tolerance based upon literature reports and
collaborative research findings.

Germplasm designation Origin or polyploidy Reference

Conventional sources

T aestivum cultivars
Kharchia 65
KRL 1-4
Yecora 70, Oasis
(susceptible checks)

Alien sources

Thinopyrum elongatum*
(Elytrigia elongata)
Th. scirpea (E scirpea)
Th. elongatum(E pontica)
Th. junceiforme (E junceiformis)
Th. distichum(E disticha)
Th. bessarabicum*
Leymus racemosus (E giganteus)
Aegilops squarrosa
(Triticum tauschii)
Ae. umbellulata
Ae. comosa
Ae. mutica
Psathyrostachys juncea*



McGuire & Dvorak (1981)
McGuire & Dvorak (1981)
McGuire & Dvorak (1981)
McGuire & Dvorak (1981)
McGuire & Dvorak (1981)
Gorham et al. (1985)
McGuire & Dvorak (1981)

Gorham (1990)
Gorham (1990)
Gorham (1990)
Gorham (1990)
Dewey (1984)

Priority source in CIMMYT's Wide Crossing Program.

Intergeneric Crosses 39



recognize that, where diverse genomic distance
prevails, the salt-tolerant trait may not be simply
inherited. This is clear from the observations of
Dvorak et al. (1988), who found three (3E, 4E, 7E)
of seven disomic addition lines derived from
T. aestivun(CS)/ Th. dlo.,gih tio to give positive
salt tolerance responses. This poses several
constraints for introgressing genes from these
three addition lines into wheat and a further
constraint in transferring from CS the tolerance
into a commercial cultivar that will contribute to
agricultural productivity in salt-prone areas.
This will take time to accomplish. W anticipate
more success in exploiting the alien cytoplasm
by working with the reciprocal cross
(Th. ..i .... ir !. aestivum). This may also
facilitate obtaining derivatives in agronomically
superior plant types. Because multiple alien
chromosomes are involved, we will also produce
multiple disomics, incorporate the maize-
mediated polyhaploid system (see Chapter 5),
and then attempt the alien introgression by Ph
locus manipulation.

The disomic 5J addition of Th. bessarabicum
(2n=2x=14) to CS imparts salt tolerance; its 2J
addition is salt-susceptible while the amphiploid
(2n=8x=56) is tolerant (Forster et al. 1987, 1988).
Further verification of 5J's reported positive
effect is warranted since our results have been
varied (Mujeeb-Kazi et al. 1991a). We are
skeptical as to whether just one alien
chromosome, such as 5J, can contribute to
acquiring a sufficient level of salt tolerance. So,
we have proceeded to produce the complete
addition set (seven chromosomes in total) in a
background that is superior to CS.
Th. bessarabicunis also believed to be a source of
tolerance to copper, aluminum, and manganese
toxicities (Mujeeb-Kazi et al. 1992) and the
complete additive line set may contribute
significantly to gaining more abiotic stress
tolerance in wheat.

By backcrossing a commercial bread wheat
cultivar (Genaro T 81) onto the Fl hybrid of
CS/ Th. bessarabicum and then selling, we have
selected several 44-chromosome derivatives.

Table 4.4. Dry weight (g) and Na and K cell sap values (mol/m3) in some wheat cultivars and their alien
derivatives under hydroculture at 50 mol/m3, then at 200 mol/m3 NaCI. Growth data measured 50 days after
reaching 50 mol/m3, Na + K measured in plants grown at 50 mol/m3 NaCI.

Cultivar or line Dry weight (g) Na (mol/m3 plant sap) K (mol/m3 plant sap)

Chinese Spring (CS) 4.42+1.14 31 + 3 225 + 5
(268 + 21) (207 + 4)*

CSITh. bessarabicum 2.30+0.30 17 + 2 196 + 7
(amphiploid) (169 + 28) (285 + 8)*

Awnless + solid stem 3.42 + 0.40 22 + 2 240 + 9
CSITh. bessarabicum (123 + 15) (321 + 9)*

Goshawk (GH'S') 1.85+0.26 20 + 3 243 + 1
(220 + 27) (270 + 9)*

Th. elongatuniGH'S' 5.33+0.67 17 + 2 233 + 7
(105 + 20) (285 + 17)*

CSITh. elongatum 6.23+0.86 18 + 2 263 + 7
(99 + 12) (290 + 11)*

* Na and K measured in plants grown at 200 mol/m3 NaCI.

40 Chapter 4

These meiotically stable (22 bivalents)
derivatives possess superior agronomic
characters, are highly fertile, and-based upon
morphological, cytological, and biochemical
diagnostics-appear to have characteristics that
are associated with each of the seven wheat
homoeologous groups (Table 4.5). Through

limited screening of some of these 44-
chromosome derivatives (Table 4.6), we have
identified a positive salt tolerance response for
two lines and two multiple disomic derivatives.
The latter two are a consequence of the selling of
BCI derivatives that possess 49 chromosomes
(2n=7x=49; AABBDDJ). In the process of selling,

Table 4.5. Thinopyrum bessarabicum (2n=2x=14, JJ) disomic addition lines identified on the basis of
cytological, morphological, and biochemical markers and tentatively assigned to the seven homoeologous
groups of Triticum aestivum L.

Line characteristic
Tentative disomic line
designation Cytological Morphological Biochemical

1J 44 (22 II)* MDH and Glu

2J 44 (22 11)* Tapering spike SOD

3J 44 (22 11)* Solid stem EST

4J 44 (22 II)* Blue aleurone PGM (?)

5J 44 (22 II)* Club-shaped spike f-AMY

6J 44 (22 11)* GOT

7J 44 (22 1)* ca-AMY

* Mean relationships over 40 meiocytes estimated. Predominantly 22 bivalents observed.

Table 4.6. Salinity hydroculture screening of some promising wheatlThinopyrum bessarabicumaddition lines
with 44 chromosomes and multiple disomics at 150 mol/m3 NaCI measured after 50 days of full stress. Data
tabulated for dry weight (g) and Na plus K from plant sap.

Amphiploid and
addition lines Dry weight (g) Na (mol/m3 plant sap) K (mol/m3 plant sap)

CSITh. bessarabicum 3.65 + 0.65 26 + 9.0 240 + 32
Addition lines

91-103 1.41 + 0.23 19 + 6 303 + 17
91-156 2.74 + 0.64 49 + 22 212 + 17
91-164 2.07 + 0.46 36 + 8 264 + 16

91-195 1.39 + 0.27 8 + 1 290 + 19

Multiple disomics
89-3171 1.77 + 0.43 40 + 11 231 + 8
89-3179 2.39 + 0.45 21 + 6 264 + 7
89-3186 3.16 + 0.88 31 + 5 218 + 23

Intergeneric Crosses 41

the alien chromosomes are paternally
transmitted in a high enough frequency to obtain
BCIF1 derivatives with chromosomal
complements of more than one disomic alien
addition (e.g., 23 bivalents). Subsequent genetic
manipulation procedures using desirable single
chromosome disomic additions or multiple
disomic additions are underway.

The role of the CS phlb genetic stock seems very
crucial to acquiring complex characters like salt
tolerance. Even though difficulties have been
encountered in exploiting its full potential
(Sharma and Gill 1986), the merits warrant
additional effort. Since the report of Sharma and
Gill (1986), we have advanced phlb x alien
species-although it has been difficult (Rosas
and Mujeeb-Kazi 1990). We are looking at an
alternate route involving the CS (Ph) x
Th. bessarabicunhybrid (Figure 4.6), which we
believe offers more promise. Results from the
backcross to the phlb source and subsequent
manipulation promise to prove quite effective in
handling alien transfers for salt tolerance. After
establishing a methodology for genomic in situ
hybridization with Th. bessarabicum (Rayburn et
al. 1993), the exploitation of the phlb locus
should be a logical extension for its application
aimed at detecting subtle wheat/alien
homoeologous exchanges.

Copper uptake efficiency
It appears that copper (Cu) deficiency affects the
wheat plant's reproductive stage more than its
vegetative (Graham 1975). Graham (1978)
studied the symptomology of copper (Cu)
deficiency on triticale, wheat, and rye by
evaluating the effects on grain yield of various
Cu levels ranging from 0 to 4.0 mg/pot. This
screening led him to identify Secale cereale cv.
Imperial as a rye variety that can efficiently use

available Cu. Chromosome 5R from S. cereale,
disomically added to CS, was further identified
as being responsive and subsequently the 5RL
rye arm of this chromosome addition was
positively associated with Cu use efficiency.

We obtained the 5AS/5RL translocation line
from the Plant Breeding Institute in Cambridge
and used it as the male parent in crosses onto
some CIMMYT spring wheat cultivars for
generating the 5A,5AS/5RLheterozygote F1
combinations. Subsequent backcrosses of these
Fls with their respective recurrent parents, up to
BCVIII followed by an eventual selling of the
best heterozygote BC derivative, should result in
elite T. aestivum materials that are near-isogenic
substitution lines with the 5AS/5RL
chromosome disomic. The 5AS/5RL

Chinese Spring x
(Ph Ph)

F1 hybrid
(2n=4x=28, ABDJ)

Backcross I derivatives x
(2n=7x=49, AABBDDJ)
Ph ph I

Th. bessarabicum

Chinese Spring
(phlb phlb)

(a) Chinese Spring
(phlb phlb)

(b) Zea mays

a) Identify phlb phlb derivatives
with J chromosomes

b) Identify phlb polyhaploid
with J chromosomes

Doubled phlb phlb

Figure 4.6. Schematic showing genetic manipulation
option using the phlb locus of Triticum aestivum cv.
Chinese Spring by direct crossing or by the maize-
mediated polyhaploid system.

42 Chapter 4

heterozygote is identified at each BC through the
morphological presence of the hairy peduncle
(hp) marker mapped on 5RL, controlled by a
dominant gene and also effective in the
hemizygous stage. Differential C-banding checks
on the heterozygote BC derivatives were
integrated to ensure adequate accuracy in the
generation advance procedures. We have
delayed screening for Cu efficiency until the BC
program, currently at BCIV, is completed.

Tolerance to copper toxicity
Since Cu is bound strongly to soil particles, Cu
toxicity is rare, but over-fertilization with Cu in
acid soils and the element's use in fungicide
applications can sometimes create a toxicity
problem. Although alternative remedial
solutions exist, their integration with genetic
tolerance does provide an advantage. Manyowa
and Miller (1991) have identified Th. bessarabicum
as a potent alien source to contribute tolerance
genes. S. cereale is another identified source-
specifically chromosome 2R. The partial
availability of disomic addition lines allowed
Manyowa and Miller (1991) to associate
chromosomes 2J and 5J (also 2Eb and 5Eb) as
disomics that contribute to tolerance. The potent
5J/6J translocation (5EbL/6EbL) may be a
positive contributor because of 5EbL. Th. repens
has also been identified as a potential source,
however, its hexaploid status may complicate its
use. We are now focusing on Th. bessarabicum,
then we will move on to S. cereale as a secondary
source of tolerance.

Aluminum tolerance
Screening-Polle et al. (1978) and Takagi (1983)
developed selection systems involving direct
observation of wheat seedling roots under
aluminum (Al) stress. L6pez-Cesati et al. (1986)
described a modified screening methodology

employed at CIMMYT. The process, based on the
fact that Al tolerance in wheat is largely a
function of Al exclusion from the roots, involves:

* Immersing the roots in a nutrient solution
containing 46 ppm Al;
Staining the roots with a 0.2% aqueous
hematoxylin solution;
Observing any continued root growth; and
Scoring on a 1-to-3 scale the corresponding Al
tolerance based continued root growth.

Based upon the above screening schedule, we
conducted experiments using: 1) conventional
germplasm, 2) alien species with their wheat
amphiploids, and 3) some S. cereale cultivars
(Table 4.7). The Al test levels were 0 and 46 ppm
for the conventional germplasm and alien
species and 0, 46, 70, and 95 ppm for the S. cereale
cultivars. Gustafson and Ross (1990) reported on
the suppression of rye gene expression by wheat,
which suggests that full expression of the rye
genes may not occur. Continued root growth
scores after hematoxylin staining allowed us to
estimate Al tolerance at various concentration
levels (Table 4.7).

Genetics and wide hybridization studies-
Genetic studies. Earlier unpublished observations
in our laboratory have provided information that
Glennson M 81 is a highly susceptible cultivar at
46 ppm; CS is medium-tolerant; and Maringa
and CNT-1 are highly tolerant. Al tolerance in
wheat varies across cultivars. The genes
controlling tolerance have been reported to:
1) range fom one to the additive effect of two or
more, 2) be dominant in action, and 3) be located
genomically and chromosomally (see review and
update by Manyowa and Miller 1991). Since a
monosomic series is now available in Glennson
M 81-highly susceptible to Al-we designed a
monosomic study using CNT-1 as the tolerant

Intergeneric Crosses 43

parent. Glennson 81 monosomic plants with 41
chromosomes were cytologically identified and
crossed with CNT-1. The Fl seed obtained forms
the basis of the monosomic analytical study that
is currently underway.

Wide hybridization studies. With the effectiveness
of the phlb locus demonstrated in CS (Sears
1977), we have incorporated CS as the female
parent in crosses with S. cereale cultivars and
A .: ;1,.. variabilis, which has, in addition to
superb resistance to Karnal bunt (see Chapter 6),
excellent Al tolerance. First, we repeatedly
screened S. cereale cultivars, and then grew
tolerant seedlings for controlled seed increase.
After at least near homozygosity was achieved,
hybridizations were made with CS (phlb phlb).
Mujeeb-Kazi and Miranda (1985) and Mujeeb-
Kazi et al. (1987, 1989) describe T. aestivum/
Ae. variabilihybrid production, embryo excision,
regeneration, transplanting, and cytological
procedures for validation. The self-sterile Fl

hybrids were advanced by crossing with wheat
and obtaining BCI seed from which embryos had
to be excised, presumably a consequence of the
presence of the ph locus. Our subsequent
backcrosses produced reasonably well filled seed
and embryo culture was not necessary.
Cytological analysis of the advanced BC
derivatives is essential to facilitate selection of
plants possessing 42 chromosomes and stable
meiosis. However, we have not yet achieved
such normal BC-selfed derivatives. Even though
somatic counts are 42 or close to it, the meiotic
relationships have been highly abnormal. We
have also crossed Ae. variabilis with Pavon F 76,
which has a Ph-dominant locus. Backcrosses to
Pavon F 76 and eventual selfings have resulted
in numerous plants with 44 chromosomes with
bivalent meiosis. Once we complete seed
increase, Al screening will hopefully help
identify alien chromosome(s) associated with Al
tolerance. Further incorporation, although time-
consuming, will follow routine genetic
manipulation procedures.

Table 4.7. Triticeae germplasm screened for aluminum tolerance under laboratory conditions in
hydroculture with tolerance response (%).

Cultivars Al'* concentration (ppm)
Germplasm Accesions 0 46 70 95

Triticum aestivum Chinese Spring 100 24
CNT-1 100 100
Glennson M 81 100 0
Maringa 100 100
Pavon F 76 100 56
Parents and Chinese Spring 100 20
amphiploids Aegilops variabilis 100 90
T turgidum cv. Laru 100 0
CSIAe. variabilis 100 82
LarulAe. variabilis 100 0

Secale cereale Short T-4776 100 100 95 65
Sardev T-4777 100 100 100 90
Doukala T-4778 100 100 100 95
Turkey T-4779 100 100 100 90
Prolific T-4781 100 100 100 90
Blanco T-4783 100 100 95 95

44 Chapter 4

Berzonsky and Kimber (1989) reported that alien
species with the N genome possess Al tolerance.
Although we could utilize,,. :.1, ..~ (Triticum)
ventricosum (DDNN) or T. rectum (UUMMNN),
we prefer to exploit Ae. (Triticum) uniaristata
(NN) because of its diploid status. To the best of
our knowledge, no direct hybrids exist between
Ae. uniaristata and T. aestivum, but with the
current successes of numerous divergent crosses
like wheat x maize (Riera-Lizarazu et al. 1992),
achieving this hybrid combination should be


It is now over 38 years since Sears (1956)
introgressed the Lr9 gene for leaf rust resistance
from Ae. umbellulata into bread wheat by
irradiation. Since then, the range and potential of
the techniques available for this type of
manipulation have increased dramatically. This
improvement has been both in the production of
wide hybrids and in the cytogenetic
manipulations possible on the derivatives of the

Most of the species of the Triticum /Aegilops
group can now be easily hybridized with the
cultivated wheats. Their genomic relationships
are well understood (Kimber 1984b). Once
desirable variation has been recognized in the
wild species and its expression in a hybrid is
established, then the choice of methodology for
the introduction of the alien variation follows
(Kimber 1984a), logically from measurements of
the relative affinity of the chromosomes
involved. This ability can only greatly increase
the range of variation upon which plant breeders
can exercise selection.

Hybrids and, in most cases, amphiploids have
been produced in crosses between species of the
genera Hordeum, Agropyron, Elymus, Secale,

Taeniatherum, Eremopyrum, and Haynaldia and
various species of the Triticum /,A. .;' g/. group as
defined by Morris and Sears (1967), Sakamoto
(1973), Mujeeb-Kazi (1982), Sharma and Gill
(1983a,b,c). The range of new hybrids with more
distantly related species is constantly increasing
and it is to be expected that a greater range of
genotypes will become available for
introgressing novel genetic variability into

It is not possible to predict the future genetic
demands that may be placed on wheat as new
races of pathogens appear or as cultivation is
extended into new areas. Consequently, a stock
of alien genetic material introgressed from wide
hybrids may prove to be of great value.

The CIMMYT program's structure and its
linkage with other disciplines demonstrate what
may be called a form of pre-breeding. After the
complex production of hybrids, development of
genetic stocks under controlled environments,
and identification of derivatives with resistance
and tolerance to biotic and abiotic stresses, the
germplasm is transferred to CIMMYT's breeders
and national program collaborators. Through
this pre-breeding structure, germplasm derived
from CIMMYT wide crosses has already resulted
in varietal releases in Pakistan. Lines resistant to
Karnal bunt and H. sativum have been registered
as genetic stocks. During the 1993-94 cycle in
Poza Rica, Mexico, -'i1 of the material selected
by the wheat breeding program for its H. sativum
resistance was derived from the wide crosses
germplasm. This developmental scheme
demonstrates how genetic resources used in a
pre-breeding effort can provide novel variability
for use in breeding programs. It also illustrates
how our support infrastructure assists in the
utilization of genetic stocks.

Intergeneric Crosses 45

The practical potential of wide hybridization in
the Triticeae is probably greater than in most
other tribes of the Grass Family, partly because
of the ease of hybridization, partly because of the
clear understanding of the cytogenetical
relationships, and partly because of the immense
importance of wheat. Consequently, there may
not be as much need to utilize techniques such as
gene splicing or somatic cell fusion within the
Triticeae as in other crops. Further such
techniques may have limitations in that the

introgressed material may not integrate well
with the wheat genotype. The introgression of
genetic material from species with relatively
close evolutionary ties to wheat would be
expected to have the most potential. In addition,
the ability to induce recombination between
homoeologous chromosomes in the Triticeae
would tend to place introgressed segments of the
grass species in the best location within the
recipient wheat chromosomes.

46 Chapter 4


Production of Polyhaploid Wheat Plants

Using Maize and Tripsacum
Abdul Mujeeb-Kazi, Oscar Riera-Lizarazu,
and Manilal D.H.M. William

several workers have successfully crossed
Triticum aestivum L. with Zea mays L.
(Zenkteler and Nitzsche 1984, Laurie and
Bennett 1986) and Tripsacum dactyloides (Riera-
Lizarazu and Mujeeb-Kazi 1993), which has led
to documented production of polyhaploid
plants. There have also been successful crosses
between Z. mays and T. 'I,. .;. i.... L. as well as
other Triticum and ,. 1..7/ spp. (O'Donoughue
and Bennett 1988).

Successful fertilizations have also been
accomplished in crosses between wheat and
Sorghum bicolor L. Moench, sorghum (Laurie and
Bennett 1988a,b); Pennisetum glaucum R. Br., pearl
millet (Laurie 1989); Z. mays ssp. mexicana,
teosinte (Ushiyama et al. 1991); Hordeum ol,.. ,..
L., barley (Laurie and Bennett 1988c); and Secale
cereale L., rye (Laurie et al. 1990).

Crosses between wheat and the above species (as
the pollen parent) provide an alternative means
of producing polyhaploid haploidd if the species
is a diploid) wheat plants through the natural
elimination of the pollen parent's chromosomes
in the early stages of embryo development (see
Chapter 4). Also, there is the possibility of
exploiting the genetic variability of the diverse
gene pools within these alien species for wheat
improvement if, for instance, maize or Tripsacum
chromosomes could actually be retained in a
wheat background.

Over the last four years, we have been
producing high frequencies of polyhaploid
wheat plants in crosses using either maize or
Tripsacum pollen. We believe that both of these
polyhaploid production procedures for wheat
are better than anther culture or wheat x
Hordeum bulbosum crosses.

Polyhaploid plants are important in our efforts
to reduce the number of generations it takes to
fix the homozygosity of wheat and other cereal
plants. A homozygous plant is obtained when a
polyhaploid's chromosomes are doubled. This
homozygosity is required in basic research
projects such as our collaborative work with
Cornell University and the International
Triticeae Mapping Initiative (ITMI) to produce
RFLP maps of the wheat and barley genomes.

After successful fertilization occurs in any of the
above crosses, chromosomes of the male parent
are eliminated very early, thus producing a
polyhaploid embryo with the chromosomes of
the female parent. Normally, the embryo soon
aborts; however, exogenous treatment with the
synthetic auxin 2,4-dichlorophenoxyacetic acid
(2,4-D) promotes seed and embryo development
until the embryo can be excised and plated onto
a synthetic medium for continued growth and
plantlet regeneration (Laurie et al. 1990).

Using this methodology, polyhaploid cereal
plants have been recovered from crosses of bread
wheat (T. aestivum) x maize (Comeau et al. 1988,

Laurie and Bennett 1988c, Suenaga and
Nakajima 1989, Inagaki and Tahir 1990, Rines et
al. 1990, Riera-Lizarazu and Mujeeb-Kazi 1990,
Laurie and Reymondie 1991); durum wheat x
maize (Riera-Lizarazu and Mujeeb-Kazi 1993);
wheat x pearl millet (Ahmad and Comeau 1990);
bread wheat x sorghum (Ohkawa et al. 1992);
bread wheat x teosinte (Ushiyama et al. 1991);
barley x maize (Furusho et al. 1991); and wheat x
Tripsacum (Riera-Lizarazu and Mujeeb-Kazi

Until recently, polyhaploid production in the
Triticeae had relied mostly on anther culture and
sexual crossings with the perennial barley
relative Hordeum bulbosum L. The occurrence of
somaclonal variation, aneuploidy, and genotypic
specificity (Picard 1989) are major limitations of
anther culture in polyhaploid production. The
homoeologous group 5 crossability loci (Krl,
Kr2, Kr3) are the major limiting factors of the H.
bulbosum sexual crossings (Snape et al. 1979; Falk
and Kasha 1981, 1983; Sitch and Snape 1986,
1987; Mujeeb-Kazi and Asiedu 1990). In order to
avoid tissue culture-associated somaclonal
variation, the sexual route to polyhaploid
production seemed to be more desirable;
however, we needed a substitute for the
troublesome H. bulbosum technique. So, we have
been exploring Zea mays L. (Laurie and Bennett
1986, 1988a,c; O'Donoughue and Bennett 1988;
Laurie et al. 1990) and Tripsacum dactyloides
(Riera-Lizarazu and Mujeeb-Kazi 1993) as
alternative sexual routes for polyhaploid
production in the Triticeae.

Wheat x Zea mays Hybridization

Since maize pollen growth and fertilization
activity appear to be insensitive to the Kr
crossability alleles of wheat (Laurie and Bennett
1987), polyhaploids can be recovered across
different genotypes (Suenaga and Nakajima

1989, Inagaki and Tahir 1990). This makes it
superior to the H. bulbosum system. In addition,
gametoclonal variation induced in doubled
polyhaploid lines using the maize system was
similar to that found in doubled polyhaploids
obtained from wheat x H. bulbosum crosses
(Laurie and Snape 1990).

The use of 2,4-D appears to be critical in
promoting seed set and embryo formation in
wheat x maize crosses (Laurie and Bennett 1988c,
Inagaki and Tahir 1990). Techniques using 2,4-D
treatment include: floret culture (Laurie and
Bennett 1988c), tiller injection (Suenaga and
Nakajima 1989, Inagaki and Tahir 1990), spike
spraying (Rines et al. 1990), and floret treatment
(Riera-Lizarazu and Mujeeb-Kazi 1990).
Detached tillers (Riera-Lizarazu and Mujeeb-
Kazi 1990) and detached spikelets (Laurie and
Bennett 1988c) offer more flexibility because
experimental material can be transferred to
locations where conditions can be more easily
controlled and monitored.

We first obtained a high recovery of wheat
polyhaploids from crosses between the wheat
cultivar 'Morocco' and CIMMYT maize
population'Pool 9A'. Subsequently, we achieved
successful polyhaploid embryo production for
additional T. aestivum and T. ',. ;.,r... cultivars
and for the T. l .. x T. tauschii synthetic
hexaploids, using a detached tiller culture

Plant material
We used two sets of plants that were field-grown
at El Batan, CIMMYT, Mexico:
* T. aestivum cv. "Morocco" andZ. mays
population "Pool 9A".
* T. aestivum, T. i. 1I .i Secale cereale,
T. ',.;1,,.?, /T. tauschii-derived amphiploids,
and Z. mays (bulk pollen sample from several
cross-pollinating maize populations).

48 Chapter 5

Crossing procedures
and detached tiller culture
We hand-emasculated spikes before anthesis and
covered them with glassine bags. When the
stigmatic surface was receptive (three to four
days after emasculation), the spikes were
pollinated with fresh maize pollen. The tillers of
pollinated spikes were detached 5 cm below the
peduncular node and placed in a beaker with an
aqueous solution of 100 mg 2,4-D/L. The basal
halves of detached tillers were then surface-
sterilized in a 211, (v/v) chlorine bleach (5.25%
sodium hypochlorite) solution for 5 min., rinsed
six times in sterile deionized water, and
transferred to test tubes (45 ml) containing liquid
MS (Murashige and Skoog 1962) basal medium
components with 100 mg 2,4-D/L without agar
(Riera-Lizarazu and Mujeeb-Kazi 1990). We
placed the test tubes with detached tillers in a
Styrofoam box containing ice-water in the
greenhouse under regimes of 25/120C (day/
night), 16-hour photoperiod, and 45 to a-1, I
relative humidity. Detached tillers were kept in
the 2,4-D medium for 48 hours and then
transferred to a growth regulator-free medium
for 12 days.

Embryo rescue, plant regeneration,
and transplanting
For each of the three crossing techniques, we
collected seeds 14 days after pollination and
sterilized them in a chlorine bleach solution ,1i I
v/v) for 15 min. Embryos were excised under a
stereomicroscope (2x) in a laminar flow hood
decontaminated with 75% ethanol. Excised
embryos were transferred to vials containing half
strength MS basal medium supplemented with
20 g sucrose/L, 0.4 mg indole-3-acetic acid
(IAA)/L, 0.1 mg 6-benzylaminopurine (BAP)/L,
and 2 g Gelrite (Scott Laboratories, Inc., West
Warwick, RI, USA)/L. Vials with embryos were
kept in the dark at room temperature for 1 to 2

weeks. After germination, we transferred the
regenerated plantlets to peat pots and eventually
to soil in pots kept in the greenhouse.

Somatic chromosome analysis of all regenerated
plants was conducted according to the method
of Mujeeb-Kazi and Miranda (1985). For meiotic
analysis, the young spikes were fixed in 6:3:1
[ethanol ;-'*- ): chloroform: glacial acetic acid]
for 48 hours and stored in 7i, ethanol solution
in the freezer (-10C) until needed. Anthers at
metaphase I were stained in alcoholic carmine
(Snow 1963), then processed according to the
modified procedure of Mujeeb-Kazi et al. (1994a)
for high contrast, intense staining, and reduced
stickiness. Mean metaphase I pairing
associations were calculated from 25 meiocytes
for some bread wheat and synthetic hexaploid

Colchicine treatment
We treated cytologically identified polyhaploid
plants with colchicine (Mujeeb-Kazi et al. 1987)
in order to induce chromosome doubling. We
presumed successful doubling had occurred if
we observed seed set.

Protein separation
Some female wheat parents and their doubled
polyhaploid progenies were analyzed by
studying the banding profiles of their seed
storage proteins (glutenin) and isozymes
(Esterase, E.C. 3.1; and f-Amylase, a-1,4-glucan
maltohydrolase E.C. The endosperm
halves of mature kernels were used to analyze
protein separation and the isozymes.

The high molecular weight glutenin subunits
were separated by using a slight modification of
the SDS-Polyacrylamide gel electrophoresis
procedure of Ng et al. (1988). Stacking gels of 2
cm and 1ii, separation gels of 15.5 cm were

Production of Polyhaploid Wheat Plants 49

used. Thickness and width of the gels were 0.15
and 16 cm, respectively. Each gel was run at 20
mA constant current for 1 hour followed by 30
mA constant current for 4 hours on a Bio-rad
protean II electrophoresis unit. The temperature
was maintained at 150C during electrophoresis.
Esterase and f-Amylase isozymes were
separated by isoelectric focusing using precast
Pharmacia PAG plates with pH gradients of 3.5-
9.5 for Esterase and 4-6.5 for f-Amylase. The
running conditions and the staining protocols
were similar to those of William and Mujeeb-
Kazi (1992).

Morocco x Z. mays pool 9A
Table 5.1 summarizes the data for the number of
florets pollinated, embryos rescued, and plants
regenerated for three 2,4-D treatment techniques:
detached tiller culture, tiller injection, and floret
spray (Riera-Lizarazu and Mujeeb-Kazi 1990).
Plant regeneration frequencies, as a percentage
of embryos excised, did not significantly differ
from embryos originating from spikes receiving
different treatment procedures. Plant

regeneration frequencies averaged 75% across
the three techniques. Table 5.2 summarizes the
percent embryo recovery for the different
treatments and individual spikes per treatment.
Embryo recovery of wheat polyhaploids ranged
from 0 to 10. I 0 to :Ii 11, and 13.63 to 41.67%
for the floret spray, tiller injection, and detached
tiller culture techniques, respectively. Embryo
recovery using the detached tiller method was
significantly higher (x=28.71 p <0.05) than
from crosses that received 2,4-D tiller injections
(x=12 ). In turn, tillers treated with an
injection of 2,4-D in the uppermost internode had
significantly higher (p<0.05) embryo recovery
frequencies than plants that received 2,4-D
sprays made 24 hours prior to maize pollination
(x=2. .

Cytological analysis confirmed that the wheat
plants possessed the expected polyhaploid
complement of n=3x=21 chromosomes.
Chromosomes 1B and 6B were consistently
identified because of their secondary
constriction; occasionally, a 5D chromosome with
its secondary constriction was also identified.

Table 5.1. Number of florets pollinated, embryos rescued, and wheat polyhaploid plants regenerated from
wheat x maize crosses using three 2,4-D treatment techniques.

2,4-D Treatments
Floret spray Tiller injection Detached tillers
Florets Embryos Plants Florets Embryos Plants Florets Embryos Plants
pollinated rescued regenerated pollinated rescued regenerated pollinated rescued regenerated
(%) (%) (%)

24 1 0 24 3 67 24 8 100
24 0 24 0 24 6 100
20 0 18 0 22 7 100
20 1 100 22 5 23 21 5 80
20 2 100 20 4 75 20 6 50
20 0 20 6 83 22 3 100
20 1 0 22 4 50 20 3 100
20 0 22 5 80 22 7 29
24 0 22 0 22 9 78
24 1 100 20 2 100 24 10 90
20 1 100 -
216 6 67 234 30 77 221 64 81

50 Chapter 5

Laurie and Bennett (1988c) reported that embryos
in caryopses, allowed to develop on the plants
without growth regulator treatments, had poor
viability, whereas spikelets cultured in solid MS
medium with 2,4-D two days after pollination
resulted in increased embryo recovery from 0.17
to 2,. Suenaga and Nakajima (1989) reported
equal to better embryo recovery frequencies (18.0
to 31.- ) by injecting the uppermost stem
internode with 100 mg 2,4-D/L. Exogenous
treatments with 2,4-D appear to enhance embryo
viability, although the mechanisms are not clear.

In our study, embryo recovery was unexpectedly
low when we applied 2,4-D in the field with the
tiller injection and floret spraying methods.
Recovery was consistently high when we applied
2,4-D in the greenhouse with the detached tiller
method (Table 5.1). The field environment was
exceedingly wet and cold during the experiment,
which might have negatively affected cross
fertilization and seed development in the injected
or sprayed spikes. Spraying of the florets was

Table 5.2. Wheat polyhaploid embryo recovery
frequencies (%) from Triticum aestivum L. x Zea
mays L. crosses under three 2,4-D treatment

2,4-D Treatments


Floret spray


Tiller injection

Detached tillers


particularly ineffective, probably due to the
exposure of unfertilized ovaries to 2,4-D 24
hours prior to pollination. Effective embryo
recovery has been reported when 2,4-D spray
applications were made 24 hours post
pollination (Rines et al. 1990). Also, reasonably
good frequencies of embryos were recovered (0.6
to 2r. when the spray procedure was used 24
hours post-pollination in wheat x Tripsacum
crosses described later.

Suenaga and Nakajima (1989) also observed a
reduction in embryo recovery when tillers were
injected one to two days before pollination. They
speculated that 2,4-D treatments prior to
pollination induced morphological and
physiological changes in unfertilized florets that
were detrimental to cross fertilization (see
Marshall et al. 1983). On the other hand, 2,4-D
treatments prior to pollination in wide crosses
have been shown to improve embryo recovery
frequencies (Kruse 1974a, Riera-Lizarazu and
Dewey 1988). Thus, other factors besides 2,4-D
applications prior to pollination may have
affected the differences in embryo recovery.

In the greenhouse, detached tillers were drier
and the caryopses larger than those obtained
from the field material. Translocation and seed
development were probably better under the
greenhouse conditions as well. In another
controlled experiment, we found the detached
tiller technique to be significantly better than
tiller injection across several wheat genotypes
(unpublished data). So, we conclude that the use
of detached tillers offers the most practical and
versatile alternative for wheat polyhaploid
production when crossing wheat x maize.
However, as we point out later in this chapter,
some modifications may be in order to improve
the detached tiller system, especially when
crossing other Triticum and Triticeae species with

Production of Polyhaploid Wheat Plants 51

a Column means followed by the same letter are not
significantly different (p<0.05).
b Standard error of the mean for treatment averages.

Although plant production frequencies from
recovered embryos did not dramatically vary
among the different treatments (frequencies
ranged from 67 to I we found embryo
germination could be increased with improved
embryo culture procedures or by enhancing
embryo development on the crossed spikes. We
rescued embryos 14 days after pollination.
Allowing embryos to remain on the spikes
longer might be appropriate if differentiated
embryos are desired. Although polyhaploid
frequency using detached tillers averaged about
2 : (average embryo recovery frequency of
28.711, x average plant regeneration frequency
of 1 I it could potentially be as high as 42% if
we consistently obtained 1II differentiation
and high embryo recovery.

In summary, the embryo excision/plantlet
regeneration/polyhaploid production
frequencies (all percentages) obtained so far are:

* 28.7/81.3/23.3 with detached tillers;
* 12.8/76.6/9.8 with tiller injection.

The recent success rates of two other laboratories
not using the detached tiller procedure have

* 25.1/83.6/20.9 (Suenaga and Nakajima 1989);
* 21.7/43.7/9.5 (Inagaki and Tahir 1990).

Triticeae species x diverse pollen
mixtures of Z. mays
As reported by Inagaki and Tahir (1990) and
Laurie and Reymondie (1991), we have also
recovered polyhaploid embryos using an
assortment of wheat genotypes (Table 5.3). In
addition, our results suggest that using detached
tillers in the maize system (as described above)
can be extended to recover polyhaploids in
durum wheats and T. ,,. I.1,. x T. tauschii
derived amphiploids (Table 5.3).

In this study, we obtained a wide range of
embryo recovery frequencies among hexaploid
wheats, tetraploid wheats, and the synthetic
hexaploids, averaging 15.6, 16.9, and 19. ,
respectively (Table 5.3). We recovered no
embryos from S. cereale x maize crosses although
Laurie et al. (1990) have reported embryo
initiation in such crosses. Mean plant
regeneration frequencies for bread wheats,
durum wheats, and the synthetic hexaploids
were 68.5, 73.9, and 74 respectively.
Successful chromosome doubling (Figure 5.1)
with colchicine averaged 60.7, for T. aestivum
cultivars, 6'-' ;, for T. ', .1.,,. cultivars, and
,.:. -i. for the synthetic hexaploids (Table 5.3).

Production frequencies of 1 t, 4 have been
considered to be acceptable for the economic
production of polyhaploids (Comeau et al. 1988).
In our study, the average doubled polyhaploid
recovery for T. aestivum, T. '~, 1 ...' and the
synthetic hexaploids (based on florets pollinated)
ranged from 6.5 to 9.4%, with average embryo
recovery frequencies of 14.7 to 19.4%, mean plant
regeneration frequencies of 68.5 to 74 and
successful doubling frequencies of 60.7 to 6'-'
(Table 5.3). Although the polyhaploid plant
frequencies we obtained for wheat in this study
more than adequately meet economic threshold
levels, Suenaga and Nakajima (1989), Inagaki
and Tahir (1990), and Riera-Lizarazu and
Mujeeb-Kazi (1990) have reported higher
frequencies across genotypes.

We attribute our lower recovery-compared to
earlier results of Riera-Lizarazu and Mujeeb-
Kazi (1990)-to continuous rainfall during tiller
collection in the field. This led to a lack of
complete tiller sterilization, which resulted in
progressive decay of the spike culm base in the
culture medium, in turn affecting normal
nutrient translocation and seed development.

52 Chapter 5

For such situations in the future, it may be best to 1983), hence modifications may be needed before
use intact spikes (Suenaga and Nakajima 1989) or we can apply the procedure to other Triticum and
to modify the detached tiller process. Triticeae species. One modification involves the
use of sulfurous acid to suppress contamination
The detached tiller system was specifically in the culture solution and culm decay as
designed to study nutrient translocation and seed reported by Kato et al. (1990). Also, we can avoid
development physiology in wheat spikes (Jenner humid/wet environments by making crosses in
1970, Donovan and Lee 1977, Singh and Jenner environmentally controlled greenhouses.

Table 5.3. Embryos produced, recovery percentage, plant regeneration, and colchicine-induced doubling
frequencies of Triticum aestivum L., T turgidum L., and T. turgidum x T. tauschii lines following crosses
with Zea mays.

Cultivars Embryos Percent embryo Plants Plants
and lines produced recovery regenerated doubled

Triticum aestivunm
AgaA/6*Yecora 70 32 16.4 28 19
Alondra'S'/Pavon F 76'S' 44 19.6 34 21
Bagula'S' 55 21.2 38 25
Bobwhite'S'/Pavon F 76'S' 48 25.3 30 20
F12.71/Coc//GenN 17 11.7 9 4
Fukuhokomugi 35 16.7 29 17
GenaroT 81 26 14.1 22 14
Glennson M 81 16 9.6 10 6
Gov/Az//Mus'S' 16 11.2 10 7
Kauz'S' 40 18.6 27 15
Mirlo'S'/Buckbuck'S' 31 22.3 19 7
Opata M 85 16 8.8 9 4
Papago'S' 11 6.5 8 4
Pavon F 76'S'/Buckbuck'S' 4 6.7 1 1
Seri M 82 22 15.8 13 9
Tesia 79 7 7.4 6 4
Thornbird'S'/Kea'S' 22 18.3 10 7
Total/Average 442 14.7 303 184
Percentage 60.7
Triticum turgiduni
CandolEnte//Arlequin 27 18.0 22 17
Altar 84 44 15.4 37 27
Laru'S' 18 11.8 13 6
Crocethia 1'S' 21 19.4 12 9
Arlequin 32 22.2 21 14
Total/Average 142 17.4 105 73
Percentage 69.5
T. turgidum x T. tausdcii lines
Duergand_2/T tauschii 23 16.4 19 11
Ruff/T tauschii 43 24.2 35 27
Yarmouk/T tauschii 20 17.4 13 6
Gan'S'/T tauschii 20 22.5 12 7
Decoyl/T tauschii 12 16.4 9 5
Total/Average 118 19.4 88 56
Percentage 63.6
a Triticum aestivum cv. "Fukuhokomugi" was obtained from G. Fedak, Plant Research Centre, Agriculture Canada, Ottawa,
Ontario, Canada. All other cultivars are part of CIMMYT's breeding germplasm.
b Obtained from CIMMYT's Wheat Wide Crosses Section.

Production of Polyhaploid Wheat Plants 53

In another example, we observed severely
reduced tiller viability when detached tillers of
Secale cereale cv. "Prolific" and "Sardev" were
used. No embryos were recovered in this rye x
maize combination. It appears that, in this
particular case, an in vivo approach may hold
more promise. If the detached tiller method
prevails, then the constitution of the nutrient
solution and the place of tiller detachment plus
the constraints expressed earlier will have to be

In this study, seeds produced from crosses
between the Triticeae species and maize lacked
normal endosperm. In addition, the embryos

were found floating in a watery solution inside
the seeds. Generally, any embryo recovered from
seed lacking normal solid endosperm is a
polyhaploid. This could serve as a
morphological diagnostic tool for screening
selfed versus cross-pollinated products.

Cytological analysis of plants recovered from
wheat x maize crosses showed them to possess
the expected polyhaploid complement of
n=3x=21 chromosomes for T. aestivum (Figure
5.2b) and n=2x=14 chromosomes for T. 'r., .;',i...
(Figure 5.3b), where each wheat parent had the
euploid number of 2n=6x=42 (Figure 5.2a) or
2n=4x=28 (Figure 5.3a), respectively. Two

Ia I b I I
Figure 5.1. Spikes of wheat polyhaploids (n=3x=21, ABD): a) side and frontal views of a sterile polyhaploid
(n=3x=21) spike without seeds; b) side and frontal views of a fertile spike as a consequence of colchicine
treatment of a n=3x=21 sterile polyhaploid derived from wheat x maize hybridization.

54 Chapter 5

T. aestivunpolyhaploids were aneuploids with 20
chromosomes (Figure 5.3c) of which one died at
the seedling stage. Another anomaly was a

T. aestivunpolyhaploid that possessed 21
chromosomes (Figure 5.3d) including a

Figure 5.2. a) mitotic metaphase spread of Triticum aestivum L. (2n=6x=42, AABBDD); b) mitotic spread of a
T. aestivumL. polyhaploid (n=3x=21, ABD).




% 1 0 #

- I


0 ,1


'I i



Figure 5.3. a) mitotic spread of Triticum turgidum cv."Altar 84" (2n=4x=28, AABB); b) mitotic spread of a
T. turgidumcv. "Altar 84" polyhaploid (n=2x=14, AB); c) a polyhaploid of Triticum aestivumwith 20
chromosomes; d) a Triticum aestivumpolyhaploid with 21 chromosomes including a telocentric (t).

Production of Polyhaploid Wheat Plants 55

p' N





Polyhaploids of T. aestivum cultivars and the
synthetic hexaploids showed very low A, B, and
D genome association, i.e., allosyndetic pairing
(Table 5.4 and Figure 5.4). Ring bivalents were
rare; the chiasmata ranged from 0.44 to 1.72/
meiocyte (Table 5.4). Riley and Chapman (1958)
reported chromosome associations of wheat
polyhaploids (n=3x=21) to be 18.05 univalents +
1.38 bivalents + 0.07 trivalents. Subsequently,
Kimber and Riley (1963) reported a mean
frequency for bread wheat of 19.18 univalents +
0.90 bivalents + 0.008 trivalents from analyses of
eight euhaploids-mean chromosome pairing
values indicating very low allosyndetic pairing.
These chromosome pairing relationships are
consistent with our data where the T. aestivum
polyhaploids of several cultivars gave a mean
metaphase I chromosome association frequency
of 18.6 univalents + 0.01 ring bivalents + 1.24 rod
bivalents + 0.06 trivalents (extracted from Table
5.4). Values for the synthetic (T. 1 ;'l, i. x
T. tauschjipolyhaploids were 20.1 univalents +
0.44 bivalents. This low pairing occurred because
the wheat cultivars and the synthetic hexaploids

used had the dominant Ph locus (one that
remains intact over the polyhaploid induction
process), which restricts homoeologous pairing.

Genes for high molecular weight glutenins have
been located on the long arms of homoeologous
group 1 chromosomes (Payne and Lawrence
1983); grain Esterase genes are on the long arms
of homoeologous group 3 chromosomes
(Ainsworth et al. 1984); and those for f-Amylase
are on group 4 and 5 chromosomes (Ainsworth
et al. 1983). Extensive allelic variations have also
been reported for all three systems. Figures 5.5-
5.7 show banding profiles of high molecular
weight glutenins, isozymes of seed esterase, and
isozymes of f-Amylase. We observed extensive
variations in the banding profiles for all the
above three systems among different cultivar
families-probably as a consequence of allelic
variation. Parental banding profiles of HMW
glutenin and esterase were identical to those
present in the doubled polyhaploid progenies.
For f-Amylase, there were some minor
differences in the banding profiles within some
families (Figure 5.7). This isozyme variation may

Table 5.4. Mean chromosome pairing with ranges in parentheses at metaphase I in some polyhaploids of
Triticum aestivum L. and T. turgidum x T. tauschii synthetic hexaploids.

Metaphase I chromosomal associations (25 meiocytes)
Bivalents (II) Trivalents (III) Chiasmata
Polyhaploid per
cultivars I Rings Rods Total II Chain Pan Total III meiocyte

Fukuhokomugi 18.8 0 1.12 1.12 0 0 0 1.12
(15-21) (0-3)
Bagula'S' 18.4 0.08 0.96 1.04 0.08 0.08 0.16 1.52
(16-21) (0-1) (0-2) (0-2) (0-1) (0-1) (0-1)
Bobwhite'S'/Pavon F 76'S' 17.7 0.04 1.48 1.52 0.08 0 0.08 1.72
(15-21) (0-1) (0-3) (0-3) (0-1) (0-1)
GenaroT 81 19.0 0 1 1 0 0 0 1
(17-21) (0-2) (0-2)
Kauz'S' 18.7 0 0.96 0.96 0.12 0 0.12 1.2
(16-21) (0-2) (0-2) (0-1) (0-1)
Mirlo'S'/Buckbuck'S' 19.0 0.04 0.96 1 0 0 0 1.04
(17-21) (0-1) (0-2) (0-2)
Duergand_2/T tauschii 20.1 0 0.44 0.44 0 0 0 0.44
(17-21) (0-2) (0-2)

56 Chapter 5

be partially attributed to post translational
modifications (Ainsworth et al. 1983), whereas
some of the band intensity differences may also
be accounted for by variation in endosperm
protein concentration. The close similarity in the
banding profiles of the doubled polyhaploid

progenies and their parents suggests stable
transmission of genetic information by this
procedure. It also indicates that the parental
genetic information for the evaluated enzyme
systems is fixed in the doubled polyhaploid
progeny without alteration.

_ S E


I '








Figure 5.4. Polyhaploid chromosome configurations of Triticum aestivum L. at Metaphase I of meiosis
showing variable univalents and bivalents as in: a) 13 univalents + 3 rod bivalents + 1 ring bivalent; b) 15
univalents + 3 rod bivalents; c) 17 univalents + 2 rod bivalents; d) 15 univalents + 3 rod bivalents (1
separated); e) 17 univalents + 1 rod bivalent + 1 ring bivalent; and f) 19 univalents + 1 rod bivalent.

Production of Polyhaploid Wheat Plants 57

r ~G

j t4


I a

go UV


- -
,@ 0 --p. .0e ON-E

wpm so-4

gsa~~ ~


A 1 S 1 2 3 K 1 2 3 4 CS
Figure 5.5. SDS-PAGE separation of seed proteins from durum and bread wheat cultivars and their extracted
doubled haploids. From left to right: Altar 84 (A) and a doubled haploid (1); Seri M 82 (S) and three doubled
haploids (1, 2, 3); Kauz (K) and four doubled haploids (1, 2, 3, and 4); and Chinese Spring (CS).


-a p 4 1 1bNE mp 4F q-

AIL A 40 Z A S 4d P^
^ SXfe $ee&*^

K 1 2 3 S 1 2 3 CS A 1 M 1 P 1 2
Figure 5.6. Grain esterase profiles of durum and bread wheat cultivars and their extracted doubled haploids.
From left to right: Kauz'S' (K) and three doubled haploids (1, 2, and 3); Seri M 82 (S) and three doubled
haploids (1,2, and 3); Chinese Spring (CS); Altar 84 (A) and its doubled haploid (1); Mirlo'S'/Buckbuck'S' (M)
and its doubled haploid (1); and Papago'S' (P) with two doubled haploids (1, 2).

58 Chapter 5

The use of the maize system for polyhaploid
production in the Triticeae is very encouraging
since genotype specificity does not exist.
Reaching homozygosity in earlier generations
will certainly accelerate work in cereal breeding
programs. Despite the current, presumably site-
specific, contamination problem we encountered
with the detached tiller method, the potential for
its application in polyhaploid production
research in cereals looks promising. Laurie and
Reymondie (1991) corroborate this contention
where high frequency polyhaploid production
has been reported in spring and winter wheat x
maize crosses. More durum wheat and rye
genotypes need to be tested to further evaluate
the detached tiller method.

Wheat x Tripsacum dactyloides

The taxonomic proximity of eastern gamagrass
(Tripsacum dactyloides L.) to maize (Doebley 1983)
has encouraged us to evaluate cross
combinations involving wheat (T. aestivum and

T. I,,~;1,,,r and T. ', .;,l.... xT. tauschii
amphiploids with Tripsacum as a novel and
alternate sexual route for the production of
cereal polyhaploids. It may also facilitate
extending the crossing cycle in Mexico by at least
eight weeks.

Plant materials
Cultivars of T. aestivum, T. '1. r .l.... and
amphiploids derived from T. 'r,. ;.1,, r. /T. tauschii
were grown in outdoor pots at El Batan,
CIMMYT, Mexico, and used as female parents in
crosses with Tripsacum dactyloides also grown
outdoors (Table 5.5).

Crossing, embryo rescue,
plant regeneration, and transplanting
Spikes were hand-emasculated before anthesis
and covered with glassine bags. When the
stigmatic surface was receptive (three to four
days after emasculation), the spikes were
pollinated with fresh Tripsacum pollen. One day
after pollination, the emasculated floral cups
were flooded with an aqueous solution of 50 mg

S 1 2 3 CS A 1 M 1 P 1 2 3 SY 1
Figure 5.7. R-Amylase profiles of durum and bread wheat cultivars and their extracted doubled haploids.
From left to right: Seri M 82 (S) and three doubled haploids (1,2, and 3); Chinese Spring (CS); Altar 84 (A) and
its doubled haploid (1); Mirlo'S'/Buckbuck'S' (M) and its doubled haploid (1); Papago'S' (P) and three doubled
haploids (1,2, and 3); and the synthetic hexaploid Ruff/T tauschii (SY) and its doubled haploid (1).

Production of Polyhaploid Wheat Plants 59

2,4-D/L and 150 mg gibberellic acid (GA3)/L. To
evaluate the effect of 2,4-D on embryo recovery
crosses involving the hexaploid wheat cultivar
Ciano T 79 and the tetraploid wheat cultivar
Altar 84 were given three treatments:

* Some spikes did not receive 2,4-D;
* Some spikes received 2,4-D, but were not
* Other spikes were pollinated and treated
with 2,4-D (Table 5.6).

Embryo rescue, plantlet regeneration, and
transplanting procedures were similar to those
reported in the section on wheat x maize
hybrids. The cytological processes for mitosis

and meiosis were also identical to those earlier
reported with the exception that we integrated a
modified procedure (Mujeeb-Kazi et al. 1994a).

Importance of 2,4-D treatment
As mentioned earlier, Suenaga and Nakajima
(1989) and Inagaki and Tahir (1990) found that
2,4-D treatment of the spikes is critical to
recovering seeds and embryos from wheat x
maize crosses. Our preliminary trials show that
2,4-D is also important for embryo recovery in
wheat x Tripsacum crosses (Table 5.6). In crosses
involving T aestivum cv. Ciano T 79 and
T. 'r~i.r... cv. Altar 84, we recovered embryos
only from pollinated florets treated with 2,4-D.
We did not recover embryos from unpollinated

Table 5.5. Embryo recovery and plant regeneration from hybridization of some synthetic hexaploids (Triticum
turgidum x T tauschii) and T. aestivum and T turgidum cultivars with Tripsacum dactyloides.

Cultivar or line Florets pollinated Embryos recovered Plants regenerated

Synthetic hexaploids
T turgidum x T tauschii 1 68 10 8
T turgidum x T tauschii2 74 7 5
T turgidum x T tauschii3 62 11 8
T turgidum x T tauschii4 86 18 13
T turgidum x T tauschii5 40 16 12
T turgidum x T tauschii6 40 11 8
T turgidum x T tauschii7 40 18 14
T turgidum x T tauschii8 40 15 12
Total 450 106 80
Percentage 23.5 75.5

T.aestivum cultivars
Glennson M 81 40 12 10
Seri M 82 40 11 9
Opata M 85 40 9 7
Bacanora T 88 40 13 9
Alondra/Pavon F 76 40 12 10
Spinebill 156 20 16
Bagula 150 27 21
Bobwhite/Pavon F 76 148 31 24
Total 654 135 106
Percentage 20.6 78.5

T. turgidum cultivars
Altar 84 88 19 13
Arlequin 40 12 9
Crocethia 1'S' 40 14 8
Total 168 45 30
Percentage 26.8 66.7

60 Chapter 5

pistils after 2,4-D treatment or from pollinated
florets without a 2,4-D treatment (Table 5.6).
Exogenous 2,4-D treatments may be important in
early stages of embryo development in wheat x
Tripsacum crosses.

In all crosses receiving 2,4-D and GA3 treatments
24 hours after pollination, we obtained a wide
range of embryo recovery frequencies. The mean
frequencies were 211 I for T. aestivum, 2r. for
T. ,.1.1~ .... and 2 for the synthetic
hexaploids (Table 5.5). There was no apparent
genotype specificity, implying that Tripsacum,
like maize and other species of the Panicoideae,
is also insensitive to the Kr crossability alleles of
wheat. A more detailed study is needed to reveal
the extent of this insensitivity in different
Tripsacum accessions because Suenaga and
Nakajima (1989) observed variation among
maize cultivars.

Embryo recovery frequencies were slightly low
in this experiment-perhaps due to variations in
technique (Comeau et al. 1992). Embryos were
smaller (averaging 0.5 mm long) than those
resulting from wheat x maize crosses (averaging
1 mm). In order to reduce the number of daily
applications, we doubled the GA3

Table 5.6. Embryo recovery frequencies in crosses
between Triticum aestivum cv. CianoT 79 and
T. turgidum cv. Altar 84 with Tripsacum dactyloides
after various treatments.

Cultivars Florets Pollination 2,4-D* Embryos
pollinated status applied recovered

CianoT 79 144 Yes No No
148 No Yes No
126 Yes Yes Yes
Altar 84 102 Yes No No
156 Yes Yes Yes

* 2,4-D = 2,4-dichlorophenoxyacetic acid.

concentration-to 150 mg/L (Suenaga and
Nakajima 1989, Furusho et al. 1991). This
doubling might have been detrimental to normal
embryo development. The GA3 variable needs
further evaluation to determine whether embryo
size could be improved by using a lower GA3
concentration or by omitting it altogether. We
anticipate that with normal embryo
development better germination frequencies will

As with the wheat x maize crosses, seeds
produced from wheat x Tripsacum lacked a
normal endosperm. Embryos were lodged at the
micropylar end of shriveled seeds or were
floating in a watery solution (probably
translocated solutes) in well-developed seeds. In
spikes treated with 2,4-D after pollination, the
ovary tissues were enlarged as happens in
normal seed development, turgid but filled with
liquid (Suenaga and Nakajima 1989, Inagaki and
Tahir 1990, Riera-Lizarazu and Mujeeb-Kazi
1990). Sometimes embryos were found, other
times not.

Cytological analyses showed the T. aestivum
polyhaploids to possess 21 chromosomes (Figure
5.8a), the T. 1~,.. ;'.,1, polyhaploids to possess 14
chromosomes, and polyhaploids from the
synthetic hexaploids to possess 21 chromosomes.
The secondary constriction site resolution readily
identified the 1B and 6B chromosomes (Figure
5.8a) in all samples.

Meiotic analyses of some ABD polyhaploids
(n=3x=21) demonstrated negligible allosyndetic
chromosome pairing at metaphase I (Table 5.7,
Figures 5.8b-d). Riley and Chapman (1958) and
Kimber and Riley (1963) reported similar low
chromosome pairing relationships-data fairly
consistent with our observations (Table 5.7). We
detected no chromosome abnormalities.

Production of Polyhaploid Wheat Plants 61

Plant regeneration frequencies from recovered
embryos were 66.7, for durum wheats, 7 -
for bread wheats, and 7; ; for the synthetic
hexaploids-similar to the earlier regeneration
frequencies of 73.9, 68.5, and 74 ; respectively,
of polyhaploids from maize crosses (Riera-
Lizarazu et al. 1992). In the maize studies, we
found colchicine doubling ranged between 63.6
and 69.,, --.,1 aspect we did not incorporate
into the Tripsacum investigation.

Because of our diversified research interests in
the synthetic hexaploids, we placed their
polyhaploids in a glasshouse where we bagged
each spike in an operational maintenance

procedure. We obtained spontaneous seed set on
seven T. 'r,. ;., r.. cv. Ruff'S' x T tauschii
polyhaploids (Table 5.8) and somatic analyses
supported the anticipated chromosome count of
2n=6x=42, AABBDD. As mentioned earlier, each
polyhaploid possessed n=3x=21 chromosomes,
hence a meiotic restitution-related process seems
to have produced the doubled seed progeny-an
event of frequent occurrence in intergeneric and
interspecific hybrids.

Crosses between wheat and Tripsacum resulted in
the production of wheat polyhaploids of various
genotypes. Unlike wheat anther culture or sexual





U 1



Figure 5.8. Somatic and meiotic cytology of a n=3x=21, ABD polyhaploid derived from a hexaploid wheat x
Tripsacum dactyloides cross: a) 21 somatic chromosomes with 1B and 6B satellite chromosomes (arrows);
b) a meiocyte at metaphase I with 21 univalents; c) metaphase I cell with 21 univalents; d) a metaphase I cell
with 21 chromosomes associated as 19 univalents + 1 stretched rod bivalent (arrows).

62 Chapter 5







r, II 10


hybridization of wheat with H. bulbosum,
troublesome genotypic specificity and
aneuploidy were absent. As with maize, this
makes Tripsacum-mediated polyhaploid
production a superior system for producing

The merits of using Tripsacum instead of maize or
a combination of both are worthy of
consideration and further evaluation. In the field
at El Batan, Mexico, Tripsacum dactyloides flowers
six to eight weeks earlier than maize, which
would allow a prolonged crossing cycle if both
maize and Tripsacum are used as pollen donors.
Regardless of which of these are used as male
parents, polyhaploid production through such
hybridizations will aid in accelerating progress
in cereal breeding programs; other cytogenetic
applications will be enhanced as well (Mujeeb-
Kazi et al. 1991b). Easier production of doubled
polyhaploid populations of different genotypes
will facilitate genetic and genome mapping
studies in cereals.

Finally, a long-term utility of Triticum x Tripsacum
hybridizations is the possibility of transferring to
wheat some of Tripsacum's desirable traits, such
as drought tolerance and insect resistance.
Earlier, Laurie and Bennett (1986) theorized a
similar concept for transferring the more efficient
C-4 photosynthetic pathway from maize to
wheat. Retention of the alien chromosomes in
wheat will be a crucial step if such introgressions
are ever to materialize.

Table 5.8. Spontaneous doubling in polyhaploids of
T. turgidum cv. Ruff'S' x T tauschii; spikes, seed
number, and somatic chromosome counts of three
seeds per doubled plant.

Somatic root tip counts
Polyhaploid Spike seeds Doubled
identification number per plant Polyhaploid seed

B91-7086 6 29 n=3x=21 42
B91-7087 9 5 n=3x=21 42
B91-7088 7 7 n=3x=21 42
B91-7089 5 20 n=3x=21 42
B91-10327 8 40 n=3x=21 42
B91-10328 8 12 n=3x=21 42
B91-10329 6 20 n=3x=21 42

Table 5.7. Mean chromosome pairing with ranges in parentheses at metaphase I in some polyhaploids
of Triticum aestivum L. and T turgidum x T tauschii synthetic hexaploids.

Metaphase I Chromosomal Associations
Polyhaploid Bivalents (II) Chiasmata
n=3x=21, entries I Rings Rods Total II Trivalents per meiocyte

Bobwhite/Pavon F 76 17.4 0 1.8 1.8 0 1.8
(15-19) (1-3)
Opata M 85 17.6 0 1.7 1.7 0 1.8
(15-19) (1-3)
Bacanora T 88 18.2 0 1.4 1.4 0 1.4
(17-19 (1-2)
Ruff/T tauschii 1 20.8 0 0.1 0.1 0 0.1
(19-21) (0-1)
Ruff/T tauschii2 21.0 0 0 0 0 0
Ruff/T tauschii3 21.0 0 0 0 0 0
Ruff/T tauschii4 21.0 0 0 0 0 0
Ruff/T tauschii5 21.0 0 0 0 0 0
T aestivuma 18.05 1.38 0.07
T aestivumb 19.18 0.90 0.008
T aestivum 8.6 0.01 1.24 1.25 0.06

a Riley and Chapman (1958). b Kimber and Riley (1963).c Riera-Lizarazu et al. (1992).

Production of Polyhaploid Wheat Plants 63

Applications of Polyhaploidy

RFLP genome mapping in wheat
Fl recombinants of inbred doubled polyhaploids
can shorten the time it takes to obtain valuable
homozygous lines. In the process commonly
known as haplo-diploidization, a homozygous
line is instantly obtained when the chromosomes
of a polyhaploid plant are doubled. In wheat,
where polymorphisms at the DNA level are
relatively low, this system can be used to obtain
polyhaploid plants from a cross that shows
polymorphisms. Upon doubling the
chromosomes of these polyhaploids using
colchicine treatment, we can produce a
population of homozygous plants that
represents the variation in the initial cross. These
progeny can then be used for RFLP mapping of
the cereal genomes.

Since polymorphic loci in hexaploid wheat
appear to be rare, RFLP linkage mapping can be
achieved by using populations of wild
progenitors where polymorphisms are more
prevalent. Of these wheat relatives, Triticum
tauschii accessions, which share complete
homology with the D genome of hexaploid
bread wheat, have been found to be highly
polymorphic at the DNA level. RFLP mapping
of hexaploid wheat is now feasible with the use
of these synthetic hexaploids-the result of
crossing T. ', .;', ,. (AABB) with T. tauschii
(DD)-see Chapter 3. When the chromosomes
are doubled, a reconstituted hexaploid wheat is
produced (AABBDD). Our procedure is the
following. First, we cross the durum cultivar
Ruff with T. tauschii to produce a highly
DNA-polymorphic synthetic hexaploid. We then
cross this synthetic with hexaploid bread wheat
cultivars such as Buckbuck, Opata M 85, and
Ciano T 79. We cross the resulting Fl derivatives
with maize to produce the polyhaploids. We

then double the chromosomes of these
polyhaploid plants to produce homozygous

Our polyhaploid production procedure has been
routinely effective, so we have not emphasized
recording the number of embryos excised from
pollinated florets. Typically, enough embryos can
be excised to allow a regeneration frequency of
between 70 and -1 and a doubling frequency
of between 60 and 7II, (Table 5.9). To date,
we have plduced at least 300 doubled
polyhaploid plants for our collaborators at
Cornell University, who are involved in the
genome mapping project.

Production of alien chromosome
addition lines
In wheat wide crosses, polyhaploidy can be
further exploited for the production of alien
chromosome addition lines from populations
that have varying chromosome numbers.
Preferably, plants with 22 chromosomes (21
chromosomes of wheat plus 1 alien
chromosome) are recovered. The final product
after colchicine treatment is a plant with 44
chromosomes (42 wheat plus an alien pair).
This process not only simplifies our production
of disomic addition lines, but also resolves the
constraints of paternal transmission of alien

Table 5.9. Polyhaploid embryo production of three
F1 DNA polymorphic crosses between Triticum
aestivum (cvs. Buckbuck, Opata M 85, Ciano T 79)
and a synthetic hexaploid (T. turgidum/T tauschii)
using the maize polyhaploid induction system. Also
included are values for plants regenerated and

Characteristic Buckbuck Opata 85 Ciano 79
observed synthetic synthetic synthetic

Number of embryos 245 260 207
Plants regenerated 172 180 154
Plants doubled 107 136 115

64 Chapter 5

chromosomes. In addition, it reduces the
analyses necessary for recovering 44-
chromosome disomic derivatives following the
selfing of a 43-chromosome plant containing
21 bivalents plus 1 univalent.

If a wide cross program were built exclusively
around the wheat cultivar "Chinese Spring", the
Hordeum bulbosum procedure (40 to 45%
polyhaploid recovery) would be satisfactory
However, in our program where commercial
wheat cultivars are used, the H. bulbosum
technique is ineffective and we logically favor
the wheat x maize methodology We have
initially applied the procedure to derivatives of
Thinopyrum dlo-,iai, x T. aestivum. From 180
backcross derivatives, with somatic chromosome
numbers of 43, 44, and 45, we have obtained
seed set after colchicine treatment on 62 plants.
Doubled haploids with 44 chromosomes have so
far allowed diagnostics of four homoeologous
groups through isozyme applications (Table
5.10). More 44-chromosome plants will be
produced for completing the addition set. The
wheat cultivar Goshawk'S'-involved in the
hybrid, its amphiploid, and in its backcrosses-
has poor crossability with S. cereale-indicative
of a dominant crossability Kr locus.

Table 5.10. Disomic Thinopyrum elongatum
additions to wheat variety Goshawk'S' developed
by doubled haploidy and identified by isozyme
analysis for homoeology.

Homoeologous Isozyme
group line marker Identification numbers

1 HMW-Glu INVO 92-6704
2 EST INVO 92-6870
5 P-Amylase INVO 92-6899
7 c-Amylase INVO 92-6840

BCI-selfed derivatives
We are also applying the maize procedure to
BCI-selfed derivatives of T. aestivum x
Th. bessarabicunwhere more than one alien
chromosome is present in a derivative. We
anticipate this will allow us to fix multiple
disomic additions. The procedure will simplify
the introgression of complex genes (for traits
such as salt tolerance) located on different
Th. bessarabicumihromosomes.

Some Closing Impressions

* The potential of stored maize and Tripsacum
pollen is being explored because it could be a
significant factor in extending the use of the
methodologies discussed in this chapter to
countries where cropping cycles are
separated or where adequate facilities are
lacking for growing plants under controlled
* Simplification of genetic studies, pyramiding
of simple genes (e.g., for leaf rust resistance),
and applications in wide crosses to
homozygosity and molecular mapping
populations are just a few avenues that could
be further exploited and diversified.
* Equally promising is the development of
doubled polyhaploids from Fl combinations
for traits like salt tolerance where the soil
heterogeneity makes genetic studies almost
* Just as 2,4-D is unequivocally recognized as
being an essential exogenous regulator in this
methodology, we argue that the quality and
quantity of maize or Tripsacum pollen are
equally critical.
* Can maize and Tripsacum chromosomes be
retained in a wheat background? If so, will
any characters be expressed? Only future
research will provide the answer.

Production of Polyhaploid Wheat Plants 65


Applications of Tissue Culture in

Wheat Wide Crosses
Abdul Mujeeb-Kazi, Nitschka ter Kuile,
Reagan Waskom, and Murray W. Nabors

The use of tissue culture technologies to
grow wheat plants from somatic or
polyhaploid cells has significant
potential to aid plant breeders in their efforts to
develop improved wheat cultivars. Callus
induction from immature embryos-including
its maintenance and eventual plantlet
regeneration-is a way to tap into the heritable
variability of Triticum aestivum for desired traits
that are either simply inherited or under more
complex polygenic control. Tissue culture
applications in the Triticeae associated with
embryo culture and hybrid plantlet
differentiation have become routine in
producing hybrids from intergeneric and
interspecific crosses. X Triticosecale Wittmack
(triticale) is a notable example.

In the 1980s, the United States Agency for
International Development (USAID) provided
funding that supported a tissue culture program
at Colorado State University. This evolved into a
larger internationally recognized effort known as
the Tissue Culture for Crops Project (TCCP). In a
five-year collaboration between CIMMYT and
the Colorado effort, a TCCP researcher was
stationed in our wheat wide crosses laboratory.
The TCCP researcher in our laboratory focused
on long-term callusing and regeneration of
certain wheat cultivars and their utility in
facilitating alien genetic transfers from wild
grasses to wheat. The collaboration terminated at

the conclusion of USAID's long-term grant to the
TCCP, but the impacts of the association are still
being realized through the development of
cytogenetic stocks and germplasm that is
resistant or tolerant to Tilletia indica (Karnal bunt,
KB), salt, and other stresses.

CIMMYT-Based Tissue
Culture Activities

We have used long-term tissue (callus and
embryo) culture and regeneration to
demonstrate the potentials of inducing
variability within various groups of the Triticeae
for morphological, biochemical, and cytological

Callus culture
Operational constraints in intergeneric
hybridization are associated with alien gene
introgression and obtaining hybrid plants with a
doubled chromosome number (amphiploid).
Callus culture has significantly helped us
overcome these constraints by:

* Promoting chromosome pairing in wheat/
alien species hybrids similar to the
chromosome associations in wheats
possessing the recessive ph locus on
chromosome 5B.
* Inducing amphiploidy in two intergeneric
hybrid combinations mediated by altered
chromosome division in the regenerated

The test systems we used for callus culture were
T. aestivumor T. ., .;r... x I .,:I ;,7, variabilis. In
the following, we discuss the cytogenetical and
practical implications of our observations.

Why use Ae. variabilis?-An Ae. variabilis
accession (no. 13E in the CIMMYT Wheat Wide
Crosses Working Collection) was reported to
possess a remarkable level of resistance to KB-a
quarantinable seedborne disease that can
seriously restrict international movement of
wheat seed. The accession showed I infection
under the boot inoculation procedure (Warham
et al. 1986). Several cytogeneticists (Sears 1977,
Jewell 1983, Jewell and Mujeeb-Kazi 1982)
concluded that the F1 hybrid-resulting from the
cross T. aestivum cultivar Chinese Spring (CS) x
Ae. variabilis-exhibits a low wheat/ alien
chromosome association frequency. This shows
that there is a limitation in the genetic exchange
between wheat and Ae. variabilis through normal
chromosomal recombination. Conventional
cytogenetic procedures provide some

Table 6.1. Data from regenerated
seven months.

opportunity for homoeologous chromosome
exchange by enhancing chromosomal pairing
through chromosome 5B manipulation; like
using the ph locus. As an alternative, we decided
to apply the callus culture procedure in an
attempt to induce random chromosomal
exchanges in the crosses of wheat/Ae. variabilis
(Vahidy et al. 1989).

Long-term callusing and regeneration-
Cultivars of T. aestivum and T. 'I. .;'. ...
(Table 6.Jwere grown in pots in a 2:1:1
sterilized mixture of soil:sand:peat moss. We
maintained the plants under greenhouse
conditions of 240C day/140C night, 15 hours
natural light, and approximately 65% relative
humidity. Immature embryos excised at 15 days
post-anthesis were cultured on LS medium
(Linsmaeier and Skoog 1965) with 2,4-D for both
callus induction and maintenance. The calli from
both species were maintained until the seventh
one-month passage. At each monthly passage,
we regenerated some embryogenic (E) callus into

Triticum aestivumand T. turgidum plants from callus maintained up to

Months in callus
3 4 5 6 7
Wheat cultivars Number of plants regenerated Total

Triticum aestivum
Mirlo'S'/Buckbuck'S' 21 33 26 6 0 86
Bobwhite'S' 1 17 12 0 0 30
Alondra'S'/Pavon F 76'S' 24 36 21 15 11 107
Goshawk'S' 30 68 33 6 7 144
Pavon F 76 0 0 3 0 0 3
Total 370

T. turgidum
Crocethia 1'S' 0 8 0 0 0 8
Laru'S' 12 12 4 10 0 38
Duergand_2 9 0 0 0 0 9
Arlequin 22 3 0 0 0 25
Rokel'S'/Kamilario'S' 2 0 0 0 0 2
Total 82

Grand Total 452

Applications of Tissue Culture 67

plants. We placed the plants in pots and
maintained them under the above greenhouse
conditions. These regenerated plants were
phenotypically observed, cytologically analyzed,
and individually harvested to obtain R1 seed.

Fl hybrids of Triticum spp. xAe. variabilis-
Additional plantings of T. aestivum (including
CS), T. ', l, .... and Ae. variabilis(CIMMYT
accession 13E) were maintained in the
greenhouse. The Triticum spp. were crossed with
Ae. variabilis (as the pollen parent); immature
embryos were excised 15 days after pollination.
We plated the embryos on MS medium
(Murashige and Skoog 1962) for plantlet
differentiation-these plantlets served as the
cytogenetic control. We plated the remaining
immature embryos on LS medium for up to 22
months with monthly transfers (or passages) of E
callus/calli. At each monthly passage, a portion
of the callus was regenerated into plants. The
plants were transferred to greenhouse growing
conditions, cytologically analyzed, and
advanced by backcrossing with appropriate
wheat cultivars. The procedures for
hybridization, mitotic cytology, and meiotic
analysis were similar to those described by
Mujeeb-Kazi and Miranda (1985) and Mujeeb-
Kazi et al. (1987, 1989).

Results with Ae. variabilis-The callus-
mediated approach to introgress alien genes into
Triticum spp. seems workable for crosses with
Ae. variabilis. This alien accession crosses easily
with T. aestivum and T. 'r., .lr .. is positive for
N-banding, and has several characteristic
biochemical markers. It is now unequivocally
accepted as a cytogenetic standard where
resulting hybrids derived from the Ae. variabilis
cross with CS have 35 chromosomes and express
a mean chromosomal association frequency of
less than one open bivalent per meiocyte.

Additionally, no fertile amphiploid derivatives
have been obtained, whereas BCI derivatives
obtained by pollinating the F1 hybrids with
T. aestivunwere highly aneuploid (Jewell 1980,
1983; Jewell and Mujeeb-Kazi 1982; Mujeeb-Kazi
and Asiedu 1989) and a negligible number of
normal 56-chromosome BCI derivatives were

Intergeneric hybrids derived from CS and
Ae. vari.,'14ill 'L, were produced at a high
frequency. The homozygous krl, kr2, and kr3
genes on homoeologous group 5 promoted
hybrid formation. The endosperm was well
formed, but despite this all embryos were
excised for direct regeneration (control) or plated
in LS media for callus induction and
maintenance for up to 22 months. Callus
portions were being regenerated at each monthly
passage (T2 to Tn; Figure 6.1 and Table 6.2). All
control F1 hybrids possessed the normal
35-chromosome characteristic of the hybrid
combination and had low pairing. The

Embryos from
Triticum spp. x Ae. vaiiabilis

Passage *

Portion regenerated 4

Regenerated 4

Regenerated 4

Callus induction





22 months

Figure 6.1. Schematic showing callus induction,
transfer, and regeneration protocol from crosses of
Triticum spp. x Aegilops variabilis.

68 Chapter 6

regenerated Fl hybrids expressed a certain
degree of aneuploidy (hyperploid or hypoploid
coupled with inclusion of telocentric
chromosomes), which could not be correlated
with the length of time in callus. In the 35-
chromosome regenerated plants, aneuploidy
involving structural chromosomal changes was
prevalent-and in one plant it influenced the 5B
chromosome. When N-banded, both short and
long arms of chromosome 5B had characteristic
banding sites that were stable across different
cells and had varied chromosome contraction
stages (Figure 6.2). The regenerated plant with
the 5B structural change indicates total absence
of bands on the 5B short arm and serves as an
ideal cytological marker for this critical
chromosome where, by backcrossing and
selection, it should be possible to obtain
derivatives homozygous for the 5B chromosome
marker. This then could be easily exploited in
transfers of the CS phlb mutant stock. So far, we
have not obtained backcross derivatives from
this 5B modified F1 hybrid.

Chromosome analysis of the regenerated plants
revealed interesting variation from the standard
low pairing characteristic (Table 6.3). Several
plants had highly paired multivalent

Table 6.2. Partial data for plants regenerated from
Triticum aestivum x Aegilops variabilis callus
during different passages (T).

Passages (T) Plants regenerated

T2 54
T3 5
T4 15
T5 12
T6 64
T7 32
T8 9
T9 10
T11 12
T12 11
T13 5
T15 20
Total 249

associations, which suggests that control
mechanisms of chromosomal pairing have been
influenced (Table 6.3)-some even up to the level
of multivalency that prevails in CS phlb hybrids
with Ae. variabilis (Asiedu et al. 1989). Although
the dominant Ph gene in chromosome 5B
suppresses homoeologous chromosome pairing,
there are other suppressors like those on



r= r


Figure 6.2. N-banded chromosome 5B from
Triticum aestivum L. cv. Chinese Spring showing
consistency of banding sites on both arms despite
stage of contraction of the chromosomes and their
extraction for the figure from different cells and
root tips.

Applications of Tissue Culture 69

chromosomes 3AS, 3DS, and 4D (minor
contribution) (Mello-Sampayo and Canas 1973,
Driscoll 1973). Since the callus-induced influence
is random, genetic changes in the suppressors
could be similarly influenced-not discounting
the fact that pairing promoter genes may also
contribute positively to the observed meiotic
associations of some regenerated plants (Table
6.3). These modifier gene(s) may also be present
in the 13E accession-an aspect that we would
have to critically analyze under optimum
environmental conditions.

Amphiploids have significant advantages in
germplasm distribution, maintenance, and
cytogenetic manipulation. However, efforts over
the last 30 years to produce an amphiploid from
the CS x Ae. variabilis combination have not been
successful. In our studies, this seems important
to achieve because of the practical significance of
Ae. variabilis' important traits for KB resistance
and aluminum tolerance and also because the
BCI progeny-produced by pollinating with
wheat the F1 derived from CS x Ae. variabilis-
was highly aneuploid (Jewell and Mujeeb-Kazi
1982). As a consequence, it was probable that all
14 alien disomic additions would not be
obtained, thus decreasing our chances to transfer
the alien chromosome(s), which confer KB
resistance or aluminum tolerance. Jewell (1980,

1983), who used Fl-based backcrossing, did
manage to create an incomplete set of alien
disomic chromosome addition lines. This
accomplishment encouraged us to attempt:

* Obtaining an amphiploid from CS x
Ae. variabiliwith 70 chromosomes; and then

* Deriving the alien additions by making
backcrosses onto the amphiploid, which may
complete the set of alien disomic addition

We were unsuccessful at inducing chromosome
doubling via direct colchicine treatment of the F1
hybrids. However, we observed occasional seed
set on a few plants in advanced growth stages of
the callus-regenerated F1 hybrids of CS x
Ae. variabilismd T. r, .;, 1... x Ae. variabilis
(Table 6.4). Apparently, seed set is a random
event. After germination, cytological analysis
revealed these seeds to possess the anticipated
70 or 56 chromosomes with meiotic regularity;
they were also self-fertile as evident from the
C-1 plus C-2 derivatives pmduced (Table 6.4).
We attribute the initial seed set on F1 regenerants
to meiotic restitution since the plants that set
seeds possessed either 35 chromosomes
(T aestivunr Ae. variabil~or 28 chromosomes
(TI ;r, i.... xAe. variabilis). One meiotic division

Table 6.3. Meiotic associations in hybrids of Chinese Spring (CS) Ph x Aegilops variabilis(13E) with low
pairing; of CS phlb x Ae. variabilis (13E) with high pairing; in a CS Phx Ae. variabilis (13E) callus-
regenerated F1 with modified increased pairing.

Mean Meiotic
Metaphase I Chromosomal Association
Rings Rods

CS (Normal) 34.68 0.16
CS (phlb) 9.50 1.6 7.2 2.1 0.25 0.08 0.04
CS (Regenerated) 13.20 0.70 6.60 1.60 0.40 0.20

Source: Asiedu et al. (1989).

70 Chapter 6

would lead to chromatid separation in the 35- or
28-chromosome hybrids. This, if coupled with
meiotic restitution, would subsequently produce
male and female gametes of 35 or 28
chromosomes that, upon fusion, are capable of
forming progeny with 70 or 56 chromosomes. A
chimerically doubled sector (shoot tip) may be
another explanation. The 70-chromosome
progenies of the T aestivum x Ae. variabilis
combination should serve as the base for
developing normal BCI derivatives, which
should lead to a more complete set of alien
disomic addition lines.

Applications of callus culture-In callus
culture, the variation that emanates from long-
term callusing and regeneration can have
significance for wheat breeding programs.
However, we are not exactly sure how callus-
induced variability (Scowcroft 1989) differs in
quality and quantity from that obtained through
applications of ionizing, non-ionizing, and
chemical mutagenic sources. Nevertheless, the
findings of Larkin and Scowcroft (1981), Lorz et
al. (1988), and Scowcroft (1989) demonstrated
callus culture-induced variation, which has been
called "somaclonal variation". This aspect was
exploited in the first phase of our study with
several tetraploid and hexaploid wheat cultivars
where we measured callusing and regeneration
responses (Table 6.1). We observed the resulting

progenies from each cultivar for morphological,
cytological, and biochemical responses.

As shown in Table 6.1, bread wheats were more
prolific in producing E callus (Figure 6.3a) and
possessed a higher number of regenerated plants
(370) than durum wheats (82). Figures 6.3b and
6.3c show the early and advanced regeneration
stages for T. aestivum. There were several
abnormalities in spike development (Figures
6.4a-d) that at the R-0 stage may be transient
changes. Any heritable changes remain to be
determined from study of the advanced

Somatic root tip cytology was done on 255
bread wheats and 46 dunm wheats. We also
performed meiotic analyses on a random sample
of spikes. Somatic root-tip count data
predominantly expressed counts of 42 for
T. aestivunand 28 for T. ~,1 .1l'... (Figures
6.5a-b). The chromosome numbers for
T. aestivunranged from 39 to 45 and for
T. ',~;'.i,. from 28 to 56; there were one or two
telocentric chromosomes in the T. 'n ;.1, I..11 b.-J
germplasm. There was a single T. ',n .;.,lr plant
with 56 chromosomes (Figure 6.3d), attributed to
spontaneous doubling, which was male-sterile
but female-fertile; it set seed after backcrossing
to T. ', .l..b The meiotic analyses of a few 42-
and 28-chromosome plants provided evidence of

Table 6.4. Regenerated plants of F1 hybrid with Ae. variabilis (13E ) of Triticum aestivumand T turgidum
showing cytologically doubled progeny.

Somatic Somatic
chromosome C-O seed chromosome C-1 seed C-2 seed
Female parent number Passages number number number number

T aestivum 35 12 1 70 14 Not advanced
T turgidum 28 7 23 56a 57 Not advanced
T turgidum 28 8 4 56 51 69
T turgidum 28 10 9 56 56 633
a Some aneuploidy (mixoploidy).

Applications of Tissue Culture 71

cytological variations where-apart from the
normal bivalent formation-meiocytes possessed
several univalents, trivalents, and quadrivalents.
We advanced all R-O derivatives since the
intrinsic R-O meiotic changes implied a great
potential for selecting stable variants in
advanced generations. So far, we have partially
selected in advanced generations for plant
height, days to anthesis, maturity, solid stem,

and isozymic electrophoretic banding
differences. These materials are being evaluated
further in yield trials by the respective CIMMYT
wheat breeding sections.

In addition to inducing variability, we are
exploiting callus induction procedures to achieve
alien gene transfers in intergeneric
hybridization. Generally, these divergent hybrids


tlK sIE 8'
lid P~

Figure 6.3. a) Embryogenic callus formation in Triticum aestivum cultures; b) Early regeneration in
T aestivum cultures; c) Advanced regeneration in T aestivum, d) A spontaneously doubled (2n=8x=56)
T. turgidum somatic cell.

72 Chapter 6

11" 11

^ two

-igure b.4. spiKe morphology variations (a-d) In K-u plants (callus regenerated) of iriticum aestivum

Applications of Tissue Culture 73

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs