UO.V.s. BIB1]OT~
IIUInive~rsiIty ~Fre~e S~tate~III~~~~~
34300000346928
Universiteit Vrystaat
........
o
IDENTIFICATION OF GENETIC
DISTANCES AND HETEROTIC GROUPS
OF INBRED MAIZE (Zea mays) LINES
USING DNA FINGERPRINTING.
EZANNE SWANEPOEL
Submitted in fulfilment of the degree
Magister Scientiae Agriculturae
Faculty of Agriculture
Department of Plant Breeding
University of the Orange Free State
BLOEMFONTEIN
December 1999
Supervisor: Prof. M.T. Labuschagne
Co-Supervisor: Dr. C.D. Viljoen
Dedicated to
Jenny
PREFACE
The results presented in this thesis follow from the study which was carried out at
the department of Plant Breeding, at University of the Orange Free State,
Bloemfontein, under the supervision of Prof M.T. Labuschagne and eo-
supervision ofDr. c.n Viljoen.
The results presented here are original and have not been submitted at any other
University.
-~- -pJ-----
Ezanne Swanepoel
ACKNOWLEDGEMENTS
I would like to thank the following people, organisations and institutions for their
contribution towards the success of this thesis:
G Prof. Maryke Tine Labuschagne for the excellent supervision and continued
support, motivation and enthusiasm.
Cj) Dr. Chris Viljoen for excellent (eo) supervision and valuable theoretical and
practical input.
e Personnel of Department of Plant Breeding at UOFS for providing excellent
support, encouragement and most of all their friendship.
oDepartment of Botany and Genetics for providing excellent research facilities.
i) All personnel and students of Molecular and Genetics Laboratory at
Department of Botany and Genetics for their outstanding technicaJ support and
assistance, research input, advice and their friendship.
o Molecular Biology Laboratory in the Department of Plant Breeding for
providing research facilities for data interpretation.
Q My best friend Jenny de Bruyn and her two daughters, Loma Lee and Christal,
for their encouragement, love, patience and support.
9 My parents and two sisters, Celecia and Zanelle, for continued support and
understanding.
o Advanta Africa Seeds for providing research funds.
o All the colleagues and friends that directly or indirectly have influenced my
studies.
o God, who made all this possible.
OUTLINE
Page
CHAPTER I 1
INTRODUCTION 1
CHAPTER2 4
LITERATURE REVIEW 4
CHAPTER3 31
Morphological characterisation of maize inbred lines. 31
CHAPTER4 60
Identification and genetic distance analysis of maize inbred lines using 60
AFLP fingerprinting.
CHAPTERS 84
Comparison of different methods of grouping maize inbred lines in 84
Clusters.
CHAPTER6 92
CONCLUSION 92
CHAPTER 7 94
SUMMARY 94
OPSOMMING 96
CHAPTER8 97
LIST OF REFERENCES 97
APPENDICES 113
CONTENTS
Page
OUTLINE
CONTENTS 11
LIST OF ABBREVIATIONS VlI
LIST OF FIGURES IX
LIST OF TABLES Xl
CHAPTER 1
lNTRODUCTION 1
CHAPTER2
LITERA TURE REVIEW 4
Introduction 4
Morphological characteristics as markers 6
DNA-based molecular marker systems (DNA fingerprinting) 9
Restriction Fragment Length Polymorphisms (RFLP's) technique 10
PCR-based technique 12
Random Amplified Polymorphic DNA markers (RAPD's) technique 13
Microsatellite markers (SSR's) 15
The Amplified Fragment Length Polymorphism (AFLP) technique 16
Principle of the AFLP technique 16
The powerful nature of AFLP markers 17
Optimisation of the AFLP technique 18
ii
DNA isolation procedures 18
Optimisation of the AFLP reaction 21
AFLP primers and adaptors 21
DNA template preparation 21
Identification of AFLP markers linked to specific genes of interest 22
The use of molecular markers for genetic analysis in crop plants 23
Cultivar identification 23
The use ofnear-isogenic lines to identify markers 23
The use of bulked segregant analysis (BSA) to identify markers 24
Parental selection and germplasm surveys 26
Genetic distance analysis of molecular markers 26
Development of genetic maps based on AFLP markers 27
Advantages of molecular markers over morphology 28
Comparison of major marker systems 28
CHAPTER3
Morphological characterisation of maize inbred lines 31
Abstract 31
Introduction 31
Materials and Methods 33
Cultivars 33
Geographic location and field design 33
Morphological parameters 35
Computer analysis 37
lil
Results 37
Morphological data 37
Phenotypic analysis 42
Discussion and conclusion 57
CHAPT.ER4
Identification and genetic distance analysis of maize inbred lines using
AFLP fingerprinting 60
Abstract 60
Introduction 60
Material and Methods 62
Plant material 62
DNA extraction 64
Primers 64
Pre-amplification and selective amplification 65
Gel analysis 66
Data collection and analysis ·66
Results 67
Quality of the genomic DNA 67
Identification of maize inbred lines 68
Genetic distance analysis of maize inbred lines 68
Grouping of inbred maize lines into clusters of genetically related lines 71
Discussion and conclusion 80
iv
CHAPTERS
Comparison of different methods of grouping maize inbred lines in clusters 84
Abstract 84
Introduction 84
Material and Methods 86
Plant material 86
Agronomic traits 86
Molecular marker assays 86
AFLP assays 86
RFLP assays 86
Data analysis 86
Results 87
Phenotypic analysis 87
Genetic distance analysis 87
AFLP cluster analysis 87
RFLP cluster analysis 87
Discussion and conclusion 88
Comparison of different methods of grouping 88
Application of these results in maize breeding programmes 90
CHAPTE.R6
CONCLUSION 92
CHAPTER 7
SUMMAR.Y 94
OPSOMMING 96
v
CHAPTER8
LIST OF REFERENCES 97
APPENDICES 113
VI
LIST OF ARBJREVJJ:ATiON§
AFLP Amplified Fragment Length Polymorphism
bp Base-pairs
lBSA Bulked Segregant Analysis
degrees Celsius
COAT Centro Internacional de Agricultura Tropical
D pairwise genetic distance
DNA Deoxyribonucleic Acid
dNTPs Deoxyribonucleotides
EDTA Ethylenediaminetetra-acetic Acid
If index of genetic similarity
kb kilobases (lkb = 103 base-pairs)
L litre
MAS Marker-assisted Selection
MgCh magnesium chloride
miv Mass per volume
mm millimetre
mM millimolar
Nael potassium chloride
NILs Near-isogenie Lines
PAUP Phylogenetic Analysis Using Parsimony
PCR Polymerase Chain Reaction
vii
PHYLI1P Phylogeny Inference Package
RAPD Random Amplified Polymorphic DNA
R.FLP Restriction Fragment Length Polymorphism
S similarity coefficient
seA Specific Combining Ability
SSR. Short Sequence Repeats
STS Sequence-tagged Site
TAE Tris-acetate-EDTA buffer
Tris Tris (hydroxymethyl) - aminomethane
UUFS University of the Orange Free State
U1PGMA Unweighted Pair-group Mean Arithmetic
UV Ultraviolet
v/v Volume per volume
w Watt
viii
LIST OF FIGURES
Page
Fig. 2.1 The process of the Amplified Fragment Length Polymorphism (AFLP) 19
technique (Perkin-Elmer, 1996).
Fig. 2.2 Illustration of the principle of amplification ofEcoRI-MseI fragments 20
(GIBCO BRL, 1996).
Fig. 3.1 Dendogram generated by Neighbour joining analysis from morphological 45
data for the 68 different maize lines.
Fig. 3.2 Dendogram generated by UPGMA analysis from morphological data 46
for the 68 different maize lines.
Fig. 3.3 Consensus tree generated by Neighbour joining analysis from morphological 47
data for the 68 different maize inbred lines.
Fig. 3.41 Consensus tree generated by UPGMA from morphological data 48
for the 68 different maize inbred lines.
Fig 4.1 Gel illustrating the presence or absence of DNA after DNA extraction. 67
Fig. 4.2 Dendogram generated by Neighbour joining analysis from genetic
distance matrix determined by Jukes and Cantor. The values on the
dendogram represent the genetic distances among different inbred lines. 73
Fig. 4.3 Dendogram generated by Neighbour joining analysis from genetic
distance matrix determined by Kimura. 74
Fig. 4.4 Consensus tree generated by Neighbour joining analysis from genetic
distance analysis determined by Jukes Cantor. 75
Fig. 4.5 Consensus tree generated by Neighbour joining analysis from genetic
distance analysis determined by Kimura. 76
IX
Fig. 4.6. Consensus tree generated by Neighbour joining analysis from the 77
combination of genetic distance matrixes determined by Jukes and Cantor,
and Kimura.
Fig. 4.7 Consensus tree generated by 1000 bootstrap replicates using DNA 78
parsimony.
Fig. 5.1 Dendogram generated by Neighbour joining cluster analysis of pairwise 88
distance data for the different maize inbred lines.
Fig. 5.2 Dendogram generated by UPGMA cluster analysis ofpairwise distance 88
data for the different maize inbred lines.
x
LIST OF TABLES
Page
Table 2.1 Comparison of major marker systems (Langridge and Chalmers, 1998). 29
Table 3.1 Maize lines used for morphological characterisation. 34
Table 3.2 Morphological traits scored three to four months after planting, for white 38
inbred lines
Table 3.3 Morphological traits scored three to four months after planting, for yellow 39
inbred lines
Table 3.4 Morphological traits scored at harvest, for white inbred lines 40
Table 3.5 Morphological traits scored at harvest, for yellow inbred lines 41
Table 3.6 Input data matrix used for UPGMA and Neighbour Joining analysis of 42
25 morphological markers
Table 3.7 Pairwise distance matrix based on the F statistic ofNei and Li (1979) 43
Table 4.1 Maize lines used for AFLP analysis.
Table 4.2 Oligonucleotide sequences used in AFLP analysis.
Table 4.3 Pairwise distance matrix using Kimura based on the F statistic ofNei 69
and Li (1979).
Table 4.4 Pairwise distance matrix using Jukes and Cantor based on the F statistic 70
ofNei and Li (1979).
xi
CHAPTER 1
INlfR. 0 IDlIJCT! ON
Southern Africa is a developing region with a relatively low agronomic potential.
Maize is currently the dominant crop and accounts for 79% of all grain production.
The dominance of maize is highlighted by the fact that maize production for the world
is 25% of the total cereal production compared to 79% for Southern Africa. There are
a number of factors in Southern Africa, together with cereal science and technology
that affect the maize industry. These factors include strong population growth, low
living standards, rapid urbanisation, democratisation, free markets and improved
scientific communication. Research needs for the maize market cover aspects such as
the maize food market, the maize feed market and the industrial market. Cownie
(1993) suggested that research in the following marketing areas should enjoy priority:
small-scale maize producers, commercial maize producers and food requirements of
the penurious sector.
Although never totally closed, the political developments in South Africa have opened
the door for better scientific communication with neighbouring countries. The new
dispensation in South Africa presents exciting possibilities. These developments can
all contribute to cereal science in the quest to invigorate the maize giant of Southern
Africa (Cownie, 1993).
Maize breeding has been effective in developing improved varieties and hybrids to
meet the rapidly changing cultural and environmental conditions of the past century.
The development of the commercial seed industry is testimony of successful breeding
methods that have evolved for the economical production of high-quality hybrid seed
that is accepted and sought after by the modem farmer (Hallauer and Miranda, 1988).
White maize is the staple food of the majority of South Africans, with an estimated
annual consumption of3.0 metric tons (mt). A further 3.5 mt of yellow maize is used
annually by various industries. High quality grain is in demand by both the consumers
and the industry and this compels farmers to produce grain of a given quality standard
(Du Plessis, 1993).
The availability of elite, adapted varieties of different genetic backgrounds allows the
farmer to hedge the risk of reduced performance by anyone variety under
unpredictable environmental stresses. Independent selection made in different
locations, according to different breeding methodologies and criteria, utilising
different germplasm, are all facets that have given rise to new genetic recombinants
(RusselI, 1974; Duvick, 1977; Castleberry et al., 1984). Every year breeders of
autogamous crops produce a multitude of potentially useful crosses and evaluate their
varietal ability (Gallais, 1979) by testing either selfed progenies or doubled haploid
lines in field experiments. If breeders could predict the prospect of crosses for line
development before producing and testing lines derived from them in field trials, this
would increase the efficiency of breeding programmes by concentrating the efforts on
the most promising crosses (Bohn et al., 1999).
One strategy used in breeding programmes to predict the prospects of crosses for line
development, is based on the assumption that the specific combining ability (SCA)
expressed by a hybrid is related to the genetic distance between its parental lines (Lee
et al., 1989). The ability to genetically define crop lines or breeding lines is useful for
maintaining genetic purity, identifying proprietary germplasm and estimating genetic
relationships. In breeding programmes, information on genetic relationships within
species is used for organising germplasm collections, identifying heterotic groups and
selecting breeding material (Prabhu et al., 1997). DNA marker systems are useful
tools for assessing genetic diversity between germplasm (Lee, 1995; Karp et al.,
1996). In comparison with morphological markers, DNA markers offer significant
advantages with respect to increased numbers of loci detectable, overall phenotypic
neutrality, and the ability to score the plant at any developmental stage (Prabhu et al.,
1997).
The aim of this study was to use AFLP fingerprinting for the identification and
genetic distance analysis of a collection of maize inbred lines in the Advanta Africa
genebanks. The knowledge of relative genetic variation and the genetic distance
between maize inbred lines will be used to facilitate identification of diverse parents
2
in order to maximise the expression of heterosis in their potential progeny. To this
end the inbred lines that have been used were assessed for genetic variation and then
compared with morphological data.
3
CHAP1rER2
JLli1'EIRA 1'URE JRIEVlIJEW
Introduction
Maize is well suited for genetic research as it is easy to grow, adapted to a wide range
of environmental conditions and has a large number of distinct hereditary variations.
Inbreeding or crossing is a simple and rapid procedure. Hundreds of kernels may be
obtained on an ear from a single pollination and each kernel represents a distinct
individual (Jugenheimer, 1976). The effective management of germplasm in terms of
resources and biodiversity in breeding programmes, requires characterisation of
material at both the agronomic (phenotypic) and genetic levels. Plant breeders focus
on morphological traits to generate new lines and to meet the legal requirements for
line registration, i.e. distinctness, uniformity and stability (Ellis et aI., 1997).
Determining the relationship of the different lines at the DNA level, can however hold
many advantages for the plant breeder, since it may increase the efficiency of
breeding efforts to improve crop species (Barrett et al., 1998). This may explain the
reason for the development of all the different marker systems.
The first marker-assisted selection in breeding programmes was carried out using
mutations at loci controlling plant morphology. Morphological markers have not,
however, been extensively used by breeders, both because the mutations available are
limited in number and most are not neutral in their phenotypic expression (Koebner
et al., 1994). Subjectivity in evaluating morphological characters and confinement of
expression of some characters to a particular stage of development, such as flowering,
presents further complications in line identification (MorelI et al., 1995).
The complications were overcome to some extent in the 1960's with the development
of gel-based assays for proteins and other gene products. The electrophoretic or
chromatographic differences between allelic gene products such as isozymes are
4
usually neutral in their effects on phenotype, i.e. they are not generally themselves the
basis of recognisable variation at the whole plant level. Furthermore they are usually
expressed eo-dominantly, allowing the breeder to unequivocally distinguish between
homozygous and heterozygous individuals for the expressed gene. Although isozyme
analysis provided more markers than were available through morphology, they were
still too few in number to use for line identification. In addition, because each
isozyme system requires its own particular staining, and in some cases electrophoretic
technique, their widespread use in breeding programmes has proven limited to a few,
very specific applications (Koebner et al., 1994).
An important breakthrough in the 1980's, was the development of RFLP's
(Restriction Fragment Length Polymorphisms), in human genetics and later for plants.
RFLP's are eo-dominant, and are effectively unlimited in number since they rely on
the ability to clone small (500-2000 nucleotide base pairs), random fragments from
complex genomes (more than a billion nucleotide base pairs). The RFLP's are
assayed by a common analytical technique, but require a relatively sophisticated assay
environment and are time-consuming and expensive. In many applications of the
technique there is a requirement for the use of radioactive assays, which present
further problems for plant breeders, particularly those from developing countries. The
late 1980' s saw the emergence of the polymerase chain reaction (PCR) and with it a
new generation of DNA marker techniques that are relatively simple and fast to use,
lend themselves to automation and are often eo-dominant (Koebner et al., 1994).
DNA molecular markers have found wide scale application in agricultural breeding
programmes through PCR-based technology (Koebner et al., 1994). Of these
techniques RAPD's (Random Amplified Polymorphic DNA markers) and AFLP's
(Amplified Fragment Length Polymorphisms), have proved to be the most suitable for
the identification of lines and varieties (Mo reIl et al., 1995). DNA fingerprinting has
the potential to distinguish between distantly and closely related individuals (Nybom,
1994).
5
Morphological characteristics as markers
Morphological characters, for which the variant allelic phenotypes are sufficiently
discrete to allow their segregation to be followed, are the easiest and generally most
economical of all markers to assay. The use of morphological characters dates back
to breeding and selection itself (Koebner et al., 1994). Phenotypic identification of
plants is based on morphological traits recorded in the field. The application of
morphological characterisation has been used for different purposes. It has been used
as a powerful tool in the classification of lines, to study taxonomic status,
identification, determining of genetic variation and correlation of characteristics with
agronomic potential (Centro Intemacional de Agricultura Tropical (CIAT), 1993;
Millan and Cubero, 1995).
There is a great range of phenotype variability among races, varieties, hybrids, and
inbred lines of maize in terms of plant type, ear type, tassel type, and maturity. These
phenotypic differences are relative to the germplasm used in breeding nurseries
(Hallauer and Miranda, 1988). For example, Kiesselbach (1980) realized that the total
number of leaves formed varies both between and within varieties. Hallauer and
Miranda (1988) suggested the opportunities for intererossing in maize are ample,
because it is essentially 100% cross-pollinated through wind movement. Jugenheimer
(1976) suggested that in maize the number and size of kernels contribute to grain
yield and that the number of kernels are determined by the length of the ear, number
of rows per ear, number of ears per plant, and number of plants per unit area.
Although breeders and geneticists use morphological characteristics such as leaf and
flower attributes to follow the segregation of genes in hybrids, most agronomic traits
are not easily observed (Kochert, 1994).
Bonnett (1954, 1960) divided the development of the maize plant into the vegetative,
transitional, reproductive, and seed stages. For these studies he used the methods of
collection and classification of lines described by Wellhausen et al. (1952). These
methods of classifying maize collections comprise four principal categories:
ó
(1) vegetative characters of the plant which include response to altitude, height, total
leaf number, number of leaves above the ear, width of ear-bearing leaf, venation-
index and internode patterns;
(2) tassel characters which include tassel length, peduncle length, length of branching
space, percentage of branching space, percentage of secondary and tertiary
branches and the total number of branches;
(3) external and internal ear characters which include ear diameter, length, and row
number, shank diameter and length, and denting, cob diameter, rachis diameter,
cob/rachis index, rachilla length, rachilla/kernel index, glume/kernel index, cupuie
hairs, rachis flag, lower and upper glume traits, rachis induration and teosinte
introgression;
(4) physiological, genetical, and cytological characters that include number of days
from planting to anthesis, pubescence of leaf sheath, plant colour, midcob colour,
chromosome knobs and B-chromosomes.
Morphological markers have been used successfully using the four principal
categories of maize classification, to estimate the diversity of historically important
sweetcorn inbreds (Gerdes and Tracy, 1994). For example, four independent
populations comprising testerosses of flint lines of maize were used by Lubberstedt et
al. (1998) to select for forage traits based on morphological and agronomic traits
(estimates of genotypic correlations among traits agreed well across the three
validation populations). Zanoni and Dudly (1989) identified inbred lines useful for
improving elite maize hybrids by evaluating them for grain yield, grain dry matter
percentage, earliness of flowering and plant and ear height. Zehr et al. (1992) used a
maize population as a donor of alleles for improvement of an elite hybrid (trait
measurements like grain yield, grain moisture at harvest, final stand, number of stalk
lodged plants, number of root lodged plants, plant and ear height and maturity were
recorded). Mock and Frey (1970) measured association of meiotic stage with several
morphological measurements including expanded leaf number, days after emergence,
plant height, and stem diameter. Expanded leaf numbers proved to be a better index of
meiosis than days after emergence, plant height, or stem diameter. However, use of
the index is limited to genetically homogeneous populations.
7
Kuleshov (1933) classified maize by endosperm types in the following groups:
1. Zea mays indurata: flour
2. Zea mays indentata: dent
3. Zea mays everta: popcorn
4. Zea mays amylacea: flint
5. Zea mays saccharata: sweet
6. Zea mays amylea saccharata: starchy-sugary
7. Zea mays ceratina: waxy
8. Zea mays tunicata: pod
Classification of maize by endosperm types was a breakthrough for genetic research
of maize in America (Anderson and Brown, 1952). These examples indicate that
morphological markers have contributed to the improvement of maize in the past and
that the technique has a role to play in the future, even if it is only a starting point in
determining distinctiveness (Gerdes and Tracy, 1994).
The study of taxonomy and genetic relationships for identification of genotypes in
many crops has become complicated due to a large number of spontaneous, as well as
artificial, crosses. Certification of new lines is usually based on the genetic purity of a
particular crop determined by botanical traits. However, morphology is often
ambiguous and has limited use for line identification (Stegemann, 1984; Koebner et
al., 1994). Plant breeders face the difficult task of having to select for traits that are
often under complex genetic control and subjected to environmental changes.
Hybridisation and introgression make this problem even more difficult. Furthermore,
many of these traits do not have the simple genetic control assumed in genetic models
thus making analysis difficult (Liu and Furnier, 1993). Morphological
characterisation requires mature plants, it usually displays dominant phenotype and
there are too few available in any single species (Koebner et al., 1994). The method
involves a lengthy survey of plant growth that is labour intensive and timely (CIAT,
1993).
Advances in molecular biology have provided new methodologies, which extend the
list of useful genetic markers (paterson et al., 1991). The first methods developed
included protein techniques such as isozyme analysis (Bassiri, 1976; Arus et al., 1982;
Stegemann, 1984; Nienhuis et al., 1995). Given a molecular map of sufficient
density, it is possible to identify and locate short regions of the chromosome that
account for much of the variation in yield or quality components of a cross. The
identification and location of quantitative trait loci (QTL) in F2 populations of maize
are also considered together with their manipulation using marker-facilitated
procedures (Stuber et al., 1990). However, these techniques were limited in use due
to laboratory constraints (Hu and Quiros, 1991). Furthermore, isozymes often show
low levels of polymorphism and problems with reproducibility arise due to tissue type
and environmental conditions.
With the advance 10 molecular techniques, DNA based procedures have been
proposed for line identification. DNA techniques have the advantage over
conventional methods in that the composition of DNA is consistent in similar tissue
types and is not affected by environmental changes (Beeching et al., 1993). The
development of DNA markers provides an opportunity to detect, monitor and
manipulate genetic variation (Yamamoto et al., 1994) more precisely than in the case
of morphological and expressed phenotypic markers. These techniques include a
variety of different methodologies commonly referred to as DNA fingerprinting
(Nybom et al., 1990). DNA molecular markers are potentially unlimited in number,
are not affected by the environment and can be mapped on linkage maps (Soller and
Beckmann, 1983; Winter and Kahl, 1995).
DNA-based! molecular marker systems (DNA fingerprinting)
A number of diverse problems in plant breeding practice can be addressed via a
genetic marker approach. The decision to exploit the possibilities opened up by the
technology is more often influenced by economical or practical, rather than by
technical considerations. Cost and labour efficiency has been steadily improving, and
this trend can confidently be expected to continue (Koebner et al., 1994).
9
Molecular markers provide a remarkable improvement in the efficiency and
sophistication of plant breeding. It is now generally accepted that molecular markers
represent the most significant advance in breeding technology that has occurred in the
last few decades and is currently the most important application of molecular biology
to plant breeding. The most common widespread application is indirect selection for
linked traits, but other applications include accelerated backcrossing, pyramiding
genes, identification of varieties and hybrids and detection of resistance genes
(Langridge and Chalmers, 1998).
The application of DNA fingerprinting holds great potential in the identification of
lines and species, and could help make breeding programmes more efficient through
the detection of genetic linkages between DNA and important quantitative traits. The
high variability of DNA fingerprinting described in humans, animals and plants
allows the identification of different individual genotypes and species (Lin et al.,
1993). Thus molecular biological techniques can be employed in a variety of ways to
facilitate crop improvement in the long term (Jones and Lindsey, 1990). DNA
fingerprinting includes various techniques, used to differentiate between individuals at
the species and subspecies levels (McClean et aI., 1994). There are two classes of
DNA markers: those based on DNA-DNA hybridization as used in RFLP's and those
based on the PCR amplification of genomic DNA sequences (Koebner et al., 1994).
With the exception of RFLP's, DNA markers lend themselves to automated analysis,
which can provide the fast, high throughput desired for examining large numbers of
samples that must typically be analysed in crop breeding (perkin-Elmer, 1996).
Detection frequency is always dependent on the level of genetic variance between the
DNA templates or genomes compared (Sobral and Honeycutt, 1994b).
Restriction Fragment Length Polymorphisms (RJFL.lP)technique
RFLP (Restriction Fragment Length Polymorphism) analysis is based on the detection
of differences in the length of restriction fragments generated by the complete
digestion of genomic DNA with restriction endonucleases (Botstein et al. 1980;
Tanksley et al., 1989; Hille et al., 1991). Fragment length polymorphism is generated
when a particular recognition site of a restriction enzyme is absent in one individual
10
and present in another, resulting in differently sized restriction fragments at that locus.
The polymorphic fragments are detected by resolving the DNA fragments using
electrophoresis and detection with probes (Southern, 1975).
DNA probes are used to detect single or multiple RFLP loci. Single copy RFLP's
have the advantage that their scoring is unambiguous and that allelism (eo-
dominance) can be determined. In contrast, RFLP probes, which give a multi-banded
profile have the advantage that more than one locus can be detected simultaneously.
In general, cDNA probes as well as many genomic DNA probes, produce single or
low copy hybridisation profiles. Some genomic DNA probes can detect repetitive
sequences, and therefore provide multi-banded patterns, which can find uses III
fingerprinting or varietal identification application (Koebner et al., 1994).
RFLP mapping is time consuming, costly and labour intensive. The technique is
difficult in some species with large and complex genomes (Marsan et al., 1993). The
complexity of the technique for performing RFLP analysis, coupled with the
widespread use of short-lived radio-isotopes has led to some limitation for routine
application in large scale crop improvement programs (Yamamoto et aI., 1994). In
addition, the RFLP technique requires a substantial amount of DNA (5-10 ug per
lane) (Beeching et al., 1995).
An advantage of the RFLP technique is that it generates a lot of detectable loci, which
are insensitive to environmental factors and can be used at any developmental stage of
the organism (Kelly, 1995). This has allowed the extensive use of RFLP analysis in
genetic studies (O'Toole, 1989), exploring evolutionary relationships among different
species and populations (Springer et al., 1994; Eubanks, 1997), for line and genotypic
identification (Ajmone-Marsan et al., 1992), gene mapping (Louie et al., 1991;
Burstin et al., 1994) and determining heterotic groups (Dudley et al., 1991). The
RFLP technique has been particularly useful in mapping species that display a high
level of intraspecific variation. Several maps have been compiled for maize (Zea
mays)(Burr et al., 1983; Helentjaris et al., 1986; Melchinger et al., 1990; Smith and
Smith, 1991; Boppenmaier et al., 1992; Prabhu et al., 1997; Gardiner et al., 1993).
11
The future role of genetic engineenng (gene identification, location, cloning and
introduction into a target genome) will require the generation of an extensive
information base for each trait. RFLP's have been suggested as molecular markers to
facilitate improvement of agronomic traits in maize and the RFLP technique can be
applied immediately (Lee et al., 1989; O'Toole, 1989). Accordingly, RFLP's have
been suggested as superior markers to protein based techniques for assessing diversity
and relationships in maize breeding germplasm (Murray et al., 1988; Walton and
Helentjaris, 1987) .
.!PeR-based! technique
The polymerase chain reaction (peR) based technology has produced a second
generation of molecular markers (Mullis and Faloona, 1987; Saiki et al., 1988;
Ehrlich et al.,1991; Koebner et al., 1994). Most peR programs include an initial
denaturation step of three to five min at 94-95°e that is intended to completely
denature complex genomic DNA so that primers can anneal after cooling (Rolfs et
al., 1992). During the first few temperature cycles with genomic DNA as template,
primers must perform a 'genomic screening' process. The most critical component
for optimising the specificity of any PeR-based assay is the choice of the annealing
temperature (Ruano et al., 1991) until they find complementary annealing sites. In
later cycles, denaturation temperatures may even be shortened in order to preserve
enzymatic activity (Gelfand, 1989). Yu and Pauls (1992) concluded that the best
results should be obtained by optimising for the shortest possible denaturing time.
Too many cycles may result in primer depletion and subsequent priming by
amplification products, which often leads to longer products and smears in the gel
(Rolfs et al., 1992). The last step of the peR program should not be a denaturation
step, because rapid cooling of single structures of even heteroduplexes, instead of
reannealing to complimentary fragments. This may disturb electrophoretic product
analysis (Rolfs et al., 1992). The main advantage of this technique over RFLP
analysis is its inherent simplistic analysis and the ability to conduct peR tests with
extremely small quantities of tissue for DNA extraction (Edwards et al., 1991;
Welsch et al., 1991). On the other hand, peR is limited in its usefulness because of
12
the time and expenses required to obtain the DNA sequence information required for
primer design (Samec and Nasinec, 1995).
Randlom Amplified Polymorphic DNA markers (RAPD's) technique
The discovery that PCR with arbitrarily selected primers will amplify a specific set of
arbitrarily distributed loci in any genome, laid the foundation for the high output of
genetic markers that can be used for a variety of purposes (Sobral and Honeycutt,
1994a). Williams ef al. (1990) was the first to use random primers in PCR on plant
samples and suggested that the technique be named RAPD's. RAPD's are obtained
by PCR amplification of DNA fragments from randomly distributed loci using single
arbitrary primers. The arbitrary primers used in the technique are usually 10bp in
length. Ideally they have a GC content of 50% to 80% and contain no palindromic
sequences. The number of DNA fragments amplified is dependent on the sequence of
the primer and the size of the genome being used as template. A single nucleotide
change in a primer usually results in a completely different amplification profile
although mismatch pairing can take place, at the first two nucleotides at the 5' end of
the primer (Yu et al., 1993; Hosaka and Hanneman, 1994).
Reaction conditions used in the RAPD procedure limit the size of the amplification to
a range of between 100 and 3000bp. Only DNA sequences that are within subsequent
primer recognition sequences in inverted orientation are amplified. Polymorphisms
occur as a result of either single base pair changes, deletions of primer sites or
insertions that increase the fragment length between primer sites. Size variants are
only rarely detected in RAPD analysis and as a rule, individual amplification products
represent one allele per locus (Devos and Gale, 1992; Waugh and Powell, 1992).
The frequency of detecting RAPD polymorphisms is generally expressed as the
number of polymorphisms per RAPD primer used. The detection frequency is always
dependent on the level of genetic variance between the DNA templates compared.
Consequently, the detection frequencies of RAPD's for lower taxa such as lines and
varieties tend to be lower than that reported between different species and genera
(Sobral and Honeycutt, 1994a).
13
The PCR technique usmg random pnmers (RAPD's) has identified DNA
polymorphisms useful for genetic mapping in a large variety of organisms (Nicholson
et al., 1993; Rowland and Levi, 1994; Wang et al., 1996). Although technically very
powerful, the use of arbitrary primers for genome mapping has the disadvantage of
characterising DNA sequences of unknown function. Thus, there is no reason to
anticipate that DNA fragments amplified by use of arbitrary primers will be enriched
for either transcribed or promoter sequences that may be conserved in evolution. For
these reasons, the arbitrarily primed PCR method was modified by using
oligonucleotide primers derived from conserved promoter elements and protein
motifs. Twenty-nine of these primers were tested individually and in pairwise
combinations for their ability to amplify genomic DNA from fowls, maize,
Drosophila, man and various inbred strains of laboratory mice and Mus spretus.
Using recombinant inbred strains of mice, the chromosomal location of 27
polymorphic fragments in the mouse genome was determined. The results
demonstrated that motif sequence-tagged PCR products are reliable markers for
mapping the mouse genome, and that motif primers can also be used for genomic
fingerprinting of divergent species (Birkenmeier et aI., 1992; Vaillancourt and Hanau,
1992).
Although the dominant nature ofRAPD's can be seen as a limitation in their use for
genetic mapping, there are specific cases where only dominant markers can be used.
For example, genetic mapping of polyploid species and tree species that contain high
amounts of DNA and long generation times have been brought out of a near standstill
by the application ofRAPD technology (Sobral and Honeycutt, 1994a). Using RAPD
profiles it is also possible to reliably and unambiguously cluster lines into groupings
(Vaillancourt and Hanau, 1992; Ting, 1994; Cao and Oard, 1997).
The success of the RAPD technique lies in the ability to generate a large number of
molecular markers using low level technology, which is accessible to a variety of
research environments. The advantage of using RAPb's include: (1) a universal set
of primers can be used for In s~ecies; (2) no probe library, radioactivity, Southern
transt~ts, ot pridt pHmer sëqU~fi.ce data are required; (3) only the oligoprimer
14
sequence is needed to use the technique; (4) the process lends itself to automation, and
(5) onl):' small quantities of DNA are needed (Williams et aI., 1990).
The automation of the RAPD technique has realised the potential for routine use of
molecular markers in breeding programmes, something that could not be possible
with RFLP technology. Many crop species that were previously excluded from
detailed mapping due to insufficient research funding, are now ideal candidates for
genetic studies, because the RAPD technique does not require any previous molecular
characterisation (Sobral and Honeycutt, 1994b). Unfortunately, the major problems
experienced in the application of the technique, have been the lack of both
reproducibility and polymorphism, which appear to be less severe in species with
small genome sizes (Koebner et al., 1994).
Microsatellite markers (SSR's)
Microsatellites or single sequence repeats (SSR's) are highly mutable loci, which are
present at many sites throughout a genome. The flanking sequences at each of these
sites are often unique. Specific primers can be designed according to the flanking
sequences, which then result in single locus identification. Alleles which differ in
length can be resolved using agarose gels or sequencing gels where single repeat
differences can be resolved and all possible alleles detected. SSR's are highly
informative because they are eo-dominant (unlike RAPD's and AFLP's) and
generally highly polymorphic (Jones et al., 1997).
Unfortunately, SSR markers are time consuming and costly to develop in that the
genomic regions carrying them must be identified and sequenced. Once the primers
are developed, the technique is, however, one of the most informative marker
systems available. Even between closely related individuals, the number of repeat
units at a locus is highly variable (Mazur and Tingey, 1995). SSR's are used to
cluster lines into groupings (Liu and Wu, 1998; Senior et al., 1998). SSR's occur in
many plant genomes including those of maize (Echeverrigaray et al., 1996; Phelps et
al., 1996; Bird et al., 1997), soybean (Akkaya et al., 1992), rice (WU and Tanksley,
1993), and barley (Saghai-Maroof et al., 1994).
15
The Amplified Fragment Length Polymorphism (AFLP)
technique
Amplified fragment length polymorphisms (AFLP's), developed by Zabeau and Vos
(1993), is a reproducible, multiplex assay with the ability to generate large numbers of
polymorphic genetic loci. AFLP's is a robust and rapid technique for determining
large numbers of DNA polymorphisms and is being used extensively for genetic
mapping and fingerprinting in plants. Use of this technique avoids problems, which
may be encountered with reproducibility, and optimisation of reaction conditions
when using arbitrarily primed peR (Money et al., 1996). This technology has been
used to analyse diversity in many species (Mackill et al., 1996; Maughan et al., 1996;
Paul et al., 1996; Travis et al., 1996). The generation of molecular marker profiles
based on AFLP's allows a retrospective analysis of the consequences of breeding and
selection on the production of new lines. In addition, it can facilitate the strategic
planning of new breeding approaches based on combining and selecting new
combinations of genotypes to maximize the rate of line improvement (Ellis et al.,
1997).
Principles of the AFLP technique
The AFLP technique is based on the selective peR amplification of restriction
fragments from an endonuclease digest of genomic DNA. The technique involves
three steps: (1) restriction digestion of DNA and ligation of oligonucleotide adaptors,
(2) pre-selective and selective amplification of restriction fragments, and (3) gel
analysis of the amplified fragments (Vos et al., 1995). peR amplification of
restriction fragments is achieved by using the ligated adaptors on to restriction ends as
target sites for primer annealing. The pre-selective and selective amplification is
achieved by using primers that extend into the restriction fragments, amplifying only
those fragments in which the primer extensions match the nucleotides flanking the
restriction sites (Fig. 2.1 )(Vos et al., 1995).
ló
Two restriction enzymes, a rare cutter and a frequent cutter (Fig. 2.2) generate the
restriction fragments for amplification. The rationale for using two restriction
enzymes is the following: (1) The frequent cutter will generate small DNA fragments
which will amplify well and are in the optimal size range for separation on denaturing
gels (sequence gels). (2) The number of fragments to be amplified is reduced by
using the rare cutter, since only the rare cutter/frequent cutter fragments are
selectively amplified. (3) Using two different restriction endonuclease gives the
greatest flexibility in selecting the number of fragments to be amplified. (4) Large
numbers of different fingerprints can be generated by using selective primers (Vos et
al., 1995).
Resolving the amplified DNA fragments on standard sequencing gels allows for the
detection of amplified fragment length polymorphisms (AFLP's). Similar to RAPD's,
AFLP's require no prior sequence knowledge, and is able to detect a ID-fold number
more of loci (20-100) than by RAPD analysis. Thus, using AFLP analysis it is
possible to rapidly screen thousands of independent genetic loci. The ability to screen
large numbers of loci has specific applications in linkage mapping studies when
combined with near-isogenie lines (Muehlbauer et al., 1988; Young et al., 1988) or
bulk segregant analysis (Michelmore et al., 1991b).
The powerful nature of AlFLP marken
The AFLP technique combines the reliability of the RFLP technique with the high
throughput ofPCR. In principle, AFLP's offer all the advantages expected ofRFLP's
and RAPD's, but in a robust PCR regime. However, the technology is largely untried
and is as yet, technically demanding (Koebner et al., 1994).
The AFLP approach is powerful because it requires no prior sequence characterisation
of the target genome and is readily applicable to a wide variety of crops.
Additionally, it is easily standardised and readily automated for high-throughput
application. Researchers can analyse thousands of polymorphic data points each day
(Perkin-Elmer, 1996). AFLP technology currently offers the fastest, most
reproducible, and most cost-effective way to generate high-density genetic maps for
17
marker-assisted selection of desirable traits. It is also the ideal tool for determining
varietal identity and assessing trueness to type (Perkin-Elmer, 1996).
Optimisation of the AFLP technique
DNA isolation procedures
Plants can be grown in a variety of environments and in different locations and still
provide starting material for DNA isolation (Young, 1994). Any part of a plant can be
used to extract DNA. The most common starting material is young leaves. They can
be either fresh, lyophilised, dried in an oven or in some cases dried at room
temperature (Kochert, 1994). Several methods for DNA extraction have been
developed (John, 1992; Tai and Tanksley, 1990; Dellaporta et al., 1983; Murrayand
Thompson, 1980). These methods all have the same goals of simplicity, speed, and
utilisation of a small amount of starting material (Lamalay et al., 1990). Simplicity
and speed are absolutely essential for processing large numbers of individuals. It is an
advantage if large quantities of starting material are hard to obtain (Young, 1994).
Yu and Pauls (1994) proposed a modification of the method described by Edwards et
al. (1991). The procedure does not require the use of expensive or hazardous
chemicals and uses only milligram quantities of leaf tissue. Approximately four to
eight hours are needed to process 50 to 100 samples, and each sample contains
enough DNA for at least] 00 Pf'R's. An important modification on the method of
Edwards et al. (1991) involves the addition of the extraction buffer. In addition, Yu
and Pauls (1994) used a pestle that fits tightly into an eppendorf tube and that was
attached to a motor revolving at approximately 500rpm. The purity of the DNA
preparation was also increased by a centrifugation step after dissolving the DNA.
IR
Fig. 21 The process of the Amplified Fragment Length Polymorphism (AFLP)
technique (Perkin-Elmer, 1996).
19
5------GAATIC ------- TIM --3"
3 - CTTAAG AATI ,5'
+Er:oR I
Msel
MIT-e------1
G AAT
+fcoR I aó~ter
TTAAs' Mse' adapter
EcoR Iadapter
5'T"~
Mse Iadapter
---A
.AATTCN ---- NTTA r-1 .
TIAAGN NAAT I --=s:JC
pr~~ectiVe
am pll1laatiOR wilt!
ECDR I primer +A
MSB Iprimer +C
primer +3 5' AAC GiTTA r--ï
_ .AATTCA ------- L--J
_TTAAGT CAATc::::J
MC 5'
selective
amplification
with pl1mers +3
_AAlTCAAC----......-lTGTTA c:J
_nAAGTTG MCAATCJ
1
denaturing polyacrylamide gel ulectro~horesis
c:::J Mse I adapter sequences
_ EeoR I adapter sequences
Fig. 2.2 Illustration of the principle of amplification of EcoRI-MseÏ fragments
(GIBeO BRL, 1996).
20
Optimisation of the AFLlP reaction
Several reaction components need to be optimised in order to give reproducible
results. Reaction components that should be optimised include template, primer,
MgCh, enzyme and dNTP concentration (Caetano-Anollés et al., 1991). Optimisation
of the PCR reaction usually relies on the sequential investigation of each reaction
variable. This approach requires a large number of PCR reactions in order to include
all the possible combinations of reaction conditions. However, in practice, the
optimum reaction conditions are rarely identified. Cobb and Clarkson (1994)
proposed a method in which the PCR reaction can be optimised using a modification
of Taguchi optimisation methods.
AFLP primers and adaptors
AFLP primers consist of three parts, a core sequence, an enzyme specific sequence
(EZ) and a selective extension (EXT) (Zabeau and Vos, 1993). AFLP-primers have
enzyme-specific parts corresponding to the respective enzymes. The AFLP
fingerprints show that large numbers of restriction fragments are amplified
simultaneously. In general, the simultaneous amplification of DNA fragments using
specific primer sets for each PCR fragments (multiple PCR) appears to be rather
troublesome. Results suggest that multi-fragment amplification is efficient provided
that all fragments use the same primer set for their amplification. This implies that the
differences in amplification efficiency of DNA fragments in PCR, are mainly primer-
associated and not fragment specific (Vos et al., 1995).
DNA template preparation
In general, template DNA quantities can vary from picogram to nanogram amounts,
depending on the genome size. However, if the template concentration is too high
then the PCR amplification often results in a smear without distinct bands (Yu et aI.,
1993). Hosaka and Hanneman (1994) demonstrated that certain amplified fragments
show continuous increase or decrease in band intensity depending on template
concentrations. This indicates that adjusting genomic DNA concentrations to the
21
same level in all samples may be an initial step m obtaining reproducible and
comparable banding patterns.
Identification of AFlLP markers linked to specific genes of
Hllllterest
It is a common practice to determine mode of inheritance, of plant traits using
classical genetics. In some instances the chromosomal location of the gene(s) coding
for the traits is deduced by monosomic and telosomic analysis. However, for most
traits, the location and characteristics of genes are unknown. Therefore, the
identification of tightly linked molecular markers represents an important first step
towards the molecular characterisation of plant genes (Yu and Pauls, 1994).
The identification of molecular markers for specific genes is usually a three step
process that involves (1) assessing the mode of inheritance of the plant trait and the
molecular marker, (2) verification of linkage between the marker and the trait through
segregation analysis and (3) calculation of the recombination fraction and linkage
distance (Yu and Pauls, 1994). The principle of Marker Assisted Selection (MAS) is,
that indirect selection for a trait can be performed by selecting for a marker closely
linked to a trait, rather than directly for the trait itself. Single traits can be targeted by
the 'bulk segregant' method, where a segregating population is scored for the trait and
DNA from individuals with two alternative bulked genotypes (Michelmore et al.,
1991a). The profiles of the two bulked genotypes are then compared with as many
markers as possible. For traits with more complex genetic control an adequate
complete genetic map is the primary prerequisite for successful MAS (Koebner et al.,
1994).
Although this study is concerned only with line identification, the applications of
other molecular technologies will also be discussed, to give the broad picture of the
uses of this technology.
22
'fine use of molecular markers for genetic analysis in. crop plants.
Line identification
DNA fingerprints of closely related organisms are generally very similar. Differences
in DNA sequence are observed as the presence or absence of bands. These
differences are characteristic and heritable. The fingerprints can be used to establish
the identities of individuals, or to assess the relatedness between individuals. The
fingerprints can also be used as a source of DNA markers for genetic linkage maps or
to identify markers linked to a specific locus (perkin-Elmer, 1996). Examples cited
include DNA fingerprinting of maize, developing herbicide and disease resistant
crops, and the production of a biodegradable natural polymer by bacteria (Doyle,
1993). By using DNA markers, rather than morphological markers, breeders can
select for desirable traits at the seedling stage, rather than at the adult plant stage.
This can save the time of an entire growing season (perkin-Elmer, 1996). The
suitability of AFLP analysis for line identification, is demonstrated by the large
number of reports recently published on the use of the technique for line identification
in a variety of plant species, such as tomato, soyabean, brassicas, sunflower, pepper,
sugar beet, lettuce (perkin-Elmer, 1996) wheat (Donini et al., 1997) and barley
(Pakniyat et al., 1997).
The use of near-isogennic lines to identify markers
Near isogenie lines (NIL's) are generated through repeated backerossing to a target
genotype and selection of the desirable trait in each backcross. The resuIt of this
process is the production of two genotypes that are in theory identical at all genetic
loci except for the chromosomal region containing the gene of interest. The high
probability that any polymorphism between NIL's will originate from the introgressed
chromosomal segment provides a powerful approach towards identifying closely
linked molecular markers (Kelly, 1995).
The size of the retained donor chromosome segment influences the efficiency of NIL
analysis. The size of the donor segment specifies the maximum distance expected
23
between a marker identified by NIL analysis and the target gene. The shorter the
donor segment, the tighter the required linkage of the marker. The size of the
introgressed chromosome segment is influenced by the number of times the isogenie
line was backcrossed to the recurrent parent, the position of the target gene on the
chromosome and the size of the genome and target chromosome (Stam and Zeven,
1981; Zeven et al., 1983). The AFLP technique has been used to generate molecular
markers for specific genes of interest in a number of crop plants. These include wheat
(Bohn et al., 1999), barley (Ellis et al., 1997), and soyabean (Maughan et al., 1996).
'rille use of bulked segregant analysis (RSA) to identify markers
AFLP fingerprinting of bulked segregants/isogenic lines is an efficient means to
identify introgression-specific DNA sequences, which can serve as a source of
template for the production of STS markers. Particularly when numerous AFLP
fragments are targeted, the bottleneck in the procedure is the identification of the
'correct' sequence among the heterogeneous population of clones retrieved from the
amplification product of DNA excised from the gel (Koebner et al., 1998).
The process of constructing NIL's is somewhat time-consuming and almost
impossible in a number of vegetatively reproducing species. Michelmore et al.
(1991b) developed an elegant method of creating 'in vitro near-isogenie lines'. The
process is called bulked segregant analysis (BSA) and involves the bulking of F2
individuals (segregants) from a cross between parents that are polarised for a specific
trait. DNA from F2 individuals, which represent the phenotypic extremes for a
segregating trait, is bulked in two separate pools. The DNA pools are then screened
for polymorphisms. It is assumed that the phenotypic distribution represent
individuals that have opposing homozygous alleles for the trait in question. All other
alleles are assumed to be in a heterozygous state because of the random distribution of
these unselected alleles (Michelmore et al., 1991b).
BSA can also be used to target multiple loci of highly heritable quantitative traits that
are controlled by a few loci with a large phenotypic effect (Michelmore et al., 1992).
Yield and quality factors, as well as resistance to some biotic and abiotic stresses, are
24
usually controlled by polygenic systems. In addition to their biological and
economical importance, these traits are very difficult to manipulate in breeding
programmes due to a poor understanding of the genetic basis of their inheritance.
Quantitatively determined traits are characterised by a continuous range of phenotypic
variation in individuals at the extremes of the phenotypic distribution (Waugh and
Powell, 1992).
BSA not only allows for rapid screening and enrichment for markers linked to specific
traits, but allows the generation of genetic maps for the traits under investigation
(Michelmore et al., 1991b). This map can be refined by selection of appropriate
progeny for further bulking and analysis, as well as individual progeny testing. These
maps can then be incorporated into existing genetic maps by cross-mapping those
regions in the standard mapping population (Michelmore et al., 1992).
Crop plants that are of utmost importance include all the major cereals (rice, maize,
wheat, barley, sorghum, millet, rye) and among arable crops, an ever-growing number
of leguminous (soybean, pea, bean), solanaceous (potato, tomato, pepper),
brassicaceous (oil seed rape) and other (sugar beet, lettuce) species (Mackill et al.,
1996; Paul et al., 1996; Powell et al., 1996; Tohme et al., 1996). The incorporation of
genes of agronomic importance into these plants depends critically on both the ability
to identify precisely alternative alleles in a segregating population at the target locus,
i.e. that the trait is qualitative, and that the population is adequately polymorphic with
respect to the markers being used (Koebner et aI., 1994). The potential ofMAS and
BSA in agriculture are considerable. Selection of genotypes can be made at very
early stages, and often at a juvenile age. This has implications for breeding programs,
as only relatively small populations of segregating material need to be grown to
maturity. Furthermore, the interaction between genotype and environment is
irrelevant for those traits selected by MAS. This means that, in theory, generations
can be advanced in off-season conditions or in locations where the environment
would otherwise be a limitation. A further benefit is that individuals carrying a
favorable recessive allele can be selected at each backcross generation, without the
need for a progeny test, as is required in conventional backcross programs (Koebner
et al., 1994).
25
Parental selection and germ plasm surveys
Parental selection is one of the most critical aspects of breeding. In general, the
probability of achieving transgressive segregation from a cross (or maximum
heterosis in a F1 hybrid) increases the more diverse the parents become. Genetic
markers provide an objective method for assessing this divergence, since they can be
used to sample polymorphisms across the entire genome, rather than just a few loci, as
in the case of phenotypic comparisons. Similar considerations relate to germplasm
surveys and the management of gene banks. Genetic markers may be useful in these
contexts to minimise duplication, thereby allowing a more rational use of scarce
germplasm storage capacity (Koebner ef aI., 1994).
The coefficient of parentage provides a statistical expectation, based on pedigrees, of
the parental contributions to a line or population. The coefficient of parentage is
calculated from pedigree data but is subject to errors stemming from deviations
between expected and actual parental contribution and incorrect pedigree information.
This coefficient cannot be determined when pedigree records are unavailable.
Molecular marker loci that are neutral and easily identified offer the opportunity to
track parental genetic contributions directly. Furthermore, the potential abundance of
molecular markers makes them ideal for use in determining genotype pedigree (Van
Toai ef al., 1997).
Genetic distance analysis of molecular markers
Two approaches are generally used to deduce phylogenetic relationships from
fingerprinting data. In the first approach, parsimony analysis, phenograms
representing phylogenetic relationships are constructed on the basis of the lowest
number of character state transformations that yields a particular phenogram. The
software program PAUP (phylogenetic Analysis Using Parsimony)(Swofford, 1991)
and PHYLIP (phylogeny Inference Package)(Felsenstein, 1993) is the most widely
used for this purpose and calculates the most parsimonious tree directly from an input
data matrix.
26
The second widely used approach involves the cluster analysis of pairwise genetic
I distances for the construction of dendograms. Pairwise genetic distances are
calculated directly from an input data matrix containing presence (1) or absence (0)
values for all markers. Each marker is considered a unique allele. Generally, the
fraction of alleles shared between any two individuals is used to calculate a similarity
coefficient (S) that is converted to a genetic distance value using either:
D = 1-S
or
D = -In(S)
This process is repeated for all the possible pairwise groupings of individuals and the
pairwise distance values tabled in a pairwise distance matrix. A number of formulas
have been used to calculate pairwise distance matrixes from discrete data. The index
of genetic similarity (F) of Nei and Li (1979), although initially developed for the
analysis ofRFLP data, is commonly used to analyse molecular data. The formula:
where F is the ratio of shared bandsIalleIes between individuals x and y, 2nxyis the
number of shared alleles, and nx and ny are the total number of alleles observed in
each of the individuals. The F value is used to calculate the distance value (D = I-F)
for every pairwise combination.
Development of genetic maps based on AFLP markers
Since the inception of the AFLP technique, the development of genetic maps for
agriculturally important crops has become priority. Plant species that have already
been mapped by this approach include potato (Rouppe van der Voort et al., 1998),
soybean (Keim et al., 1997), Astragalus cremnophylax var. cremnophylax (a critically
endangered plant)(Travis et al., 1996.), barley (Simons et al., 1997) and tomato
(Thomas et al., 1995). Many other crop plants are in the process of being mapped in
27
this way. The research that is being done underlines the importance of this
technology in plant biotechnology.
Advantages of molecular markers over morphology
Molecular DNA markers such as AFLP's offer numerous advantages over markers
traditionally used in plant mapping and selective breeding. In the past, breeders have
selected among closely related strains on the basis of morphological markers that are
readily observable and that are eo-inherited with the desired trait. Because of the
scarcity of such markers, this approach often fails to discriminate between closely
related lines that differ for the trait of interest (Perkin-Elmer, 1996).
Even when a useful morphological marker is identified, its application in breeding
programmes is time consuming and costly. Typically, an entire field of a particular
genotype must be grown and analysed. The task of examining the myriad of
individual plants IS arduous and time-consuming (Perkin-Elmer, 1996).
Consequently, breeders are abandoning this traditional approach in favor of faster,
highly discriminating and less costly, approach of using molecular markers such as
RFLP's, RAPD's, SSR's and AFLP's (Perkin-Elmer, 1996).
Comparison of major molecular systems
The marker systems that are currently available, and there relative advantages and
disadvantages, are summarised in Table 2.1.
During the past few years, the use of polymerase chain reaction-based techniques has
significantly increased the application of DNA markers to genotyping, genome
mapping and phylogenetics. The advantage of these techniques over other DNA
marker techniques includes the detection of a large number of polymorphisms from a
single experiment, thus reducing expense (Van Toai et al., 1997). AFLP and RAPD
technology is PCR-based and has the advantage over RFLP's, as it is readily
automatable. Unlike RAPD's, AFLP's have proven to be more reproducible.
2R
Furthermore, AFLP analysis requires no pnor sequence knowledge of the target
genome and, therefore, has no prior characterisation costs (perkin-Elmer, 1996).
Table 2.1 Comparison of major marker systems (Langridge and Chalmers, 1998)
Marker Loci! DNA Time/ % bands Informative Comments
system assay amount assay poly value
RFLP 3 5.01lg 5 days 10 0.3 Dominant, reliable
SSR 1 0.21lg 5 hrs 10 0.1 Dominant, reliable
AFLP 50 0.21lg 1 day 10 5.0 Mostly recessive, reliable
RAPD 10 O.21lg 5 hrs 20 2.0 Recessive, less reliable
The AFLP approach provides lO-fold more markers on average than do the other
molecular approaches. This translates into 10-100 times denser linkage maps.
AFLP's also tend to be more informative than RFLP's, RAPD's, or SSR's (Table
2.2). AFLP markers offer considerable advantages in terms of both the number of
informative data points generated per sample reaction and the number of markers
placed in a genome (Perkin-Elmer, 1996).
Using the AFLP approach allows the specific amplification of numbers of restriction
fragments. However, the number of fragments that can be analysed simultaneously, is
dependent on the resolving power of the detection system. The AFLP technique
provides a novel and very powerful DNA fingerprinting technique for DNA's of any
origin or complexity (Vos et al., 1995).
AFLP's were initially difficult to perform by most groups, whose first attempts
produced profiles in which over 50% of bands were not reproducible. However, with
greater familiarity, these problems were resolved and the AFLP profiles subsequently
obtained showed extremely high reproducibility. This confirms previous reports on
the reproducibility of AFLP profiles (Jones et al., 1997).
The data in Table 2.1 represents averages for these marker systems screened against
commercial varieties. Although Table 2.1 suggests that RAPD markers are the most
29
efficient, the poor reliability of these markers and the inability to transfer them
between crosses has greatly limited their use (Langridge and Chalmers, 1998). The
crucial question in all aspects of marker implementation is cost. The ramifications of
this question are the value of the trait or traits being monitored, the availability and
speed of alternative screening techniques and the likely segregation pattern of the
trait. It is not feasible to use markers to select complex traits that are determined by
five or more genes unless special strategies are introduced to keep the effective
population size small. The efficient implementation of markers will depend upon
devising new breeding strategies. The implementation of markers in breeding
programmes does not simply lie in replacing existing procedures but devising new
breeding strategies ( Langridge and Chalmers, 1998).
10
CHAPTER3
MORPHOJLOG KCAL CRARA.CTERlISA TlION OF MAiZE
JrNBRJEJI)LJfNE§.
Abstract
Sixty eight maize (Zea mays L.) inbred lines (both flint and dent) from the Advanta
Africa genebank:, representing five countries, were evaluated for genetic diversity to
select germplasm that can be used in breeding programmes to improve hybrid
development. The objective of this study was to evaluate the inbred lines, using 25
morphological characters. The morphological characters were coded into a discrete
binary data matrix based on the presence and absence of character states. The inbred
lines were grouped, using UPGMA and Neighbour joining, into clusters. The results
indicated that the grouping of inbred lines are correlated with origin. Morphology
was found to be a costly method to determine genetic relatedness.
bn trod anein0Il1l
Germplasm curators as well as plant breeders have an interest in quantification and
classification of genetic diversity. In germplasm collections, such information is used
to designate core collections and enhance efficiency of collection management and
utilisation (Brown et al., 1987). Van Beuningen and Busch (1997), two U.S. maize
(Zea mays L.) breeders, developed a new classification system using heterotic
grouping. This method of classifying maize collections has made a significant
contribution to maize breeding.
The recognition and exploitation of heterotic patterns among genetically divergent
groups of germplasm are fundamental in hybrid breeding (Hallauer et al., 1988).
11
Classification of elite germplasm into heterotic groups and assignment of new inbred
lines to establish heterotic groups are major decisions 10 any maize breeding
programme. For these reasons maize breeders have a keen interest in the
characterisation of genetic diversity among and within existing heterotic groups as
well as in the relationships between current and historically important lines (Messmer
ef al., 1992).
More recently, an interest in the quantification of genetic diversity of pure line crops
has developed. Transgressive segregation may be more likely to occur when parents
in a cross are less similar, allowing different favourable alleles to be combined in the
offspring (Cowen and Frey, 1987). The perspective that hybrid lines and
consequently the detection of heterotic patterns may become important for grain crops
adds to the importance of quantification of genetic distances and relatedness patterns
in these crops (Cregan and Busch, 1978; Cox and Murphy, 1990).
Traditionally, distance estimation and classification were based entirely on
morphological markers and quantitative traits (Goodman, 1972). Distance estimations
and classification was greatly enhanced in the last decade with the introduction of
novel genetic marker systems (Keim ef al., 1997). Recent studies have focused on
genetic markers in the form of gene products such as isozymes (Messmer ef al., 1991)
and direct DNA markers such as Restriction Fragment Length Polymorphisms
(RFLP's) (Tanksley, 1983), Random Amplified Polymorphic DNA (RAPD's) (Wang
et al., 1996), microsatellites (SSR's) (phelps et al., 1996) and Amplified Fragment
Length Polymorphisms (AFLP' s) (Ellis et al., 1997).
An assumption underlying the use of phenotypic similarity estimates based on
quantitative and qualitative traits is that such estimates are an accurate reflection of
genotypic similarity (Van Beuningen and Busch, 1997). Morphological
characterisation is useful in breeding programmes because: (i) existing databases on
germplasm collection or breeders crossing block entries can often be used for genetic
analysis, (ii) statistical procedures for processing morphological characterisation are
readily available, (iii) morphological information adds to an understanding of
ideotype-performance relations and (iv) explanations of heterosis may be enhanced by
adding morphological measures of distance as another independent variable (Cox and
32
Murphy, 1990). Therefore, the grouping oflines based on morphological similarities
will enable breeders to make better decisions about which crosses will generate higher
combining ability. The reason is that the further the genetic distance between two
individuals, the more likely to expect a better combining ability.
The aim of this study was to determine clustering patterns of a large collection of
maize inbred lines on the basis of morphology and plant development. Furthermore,
to evaluate the relationship between distance based on quantitative and qualitative
morphological traits by dividing them in heterotic groups for later molecular genetic
analysis. The data would then be a starting point for breeders to do better planning for
field trials to get the maximum benefit from the growing season.
Materials and Methods
Lines
The plant material used in this study was obtained from the Advanta group (Table
3.1). Out ofa total of68 lines studied (both flint and dent), 36lines were from South
Africa (local), three from Zimbabwe (CIMMYT), three from Argentina, 12 from
Garst (USA) and six from Thailand. Each lines name and country of origin is
summarised in Table 3.1. (Arbitrary codes were assigned to lines to protect
intellectual property rights and confidentiality.) Seed was obtained from the Advanta
grain germplasm depository at Petit, South Africa and represents a broad range of
diversity in origin. All 68 inbred lines are currently used in the maize-breeding
programme at Bapsfontein, South Africa, with the aim of making crosses between
lines, which have the best combining ability. Morphological data from this material
was collected at Bapsfontein Research Station during the 1998/1999 seasons.
GeograpllIicaD location and! field design
The average minimum temperature of the area is 8.9°C and the average maximum
temperature is 28.7°C. The relative humidity is approximately 60%. The seasonal
rainfall is approximately 700mm, lasting from September to April and is inconsistent.
:n
It is classified as a summer rainfall- and semi-arid area (Weather Bureau, JHB
International).
Table 3.1 Maize lines used for morphological characterisation
NQ Lines Origin NQ Lines Origin
1 Y012 Argentina 35 YM2 Local
2 Y013 Argentina 36 YM3 Local
3 Y014 Argentina 37 YI1 Local
4 YS1 Brazil (tropical) 38 YI2 Local
5 W08 Brazil (tropical) 39 YI3 Local
6 W09 Brazil (tropical) 40 YR1 Local
7 W010 Brazil (tropical) 41 YS3 Local
8 W011 Brazil (tropical) 42 YF4 Local
9 W05 (CIMMYT) Zimbabwe 43 W01 Local
10 W06 (CIMMYT) Zimbabwe 44 WF1 Local
11 W07 (CIMMYT) Zimbabwe 45 WF2 Local
12 YM4 GARST (USA) 46 WI2 Local
13 YL2 GARST (USA) 47 WP3 Local
14 YL3 GARST (USA) 48 WK1 Local
15 YR2 GARST (USA) 49 WK2 Local
16 YR3 GARST (USA) 50 Y08 Local
17 YR5 GARST (USA) 51 Y09 Local
18 YL1 GARST (USA) 52 Y010 Local
19 YL4 GARST (USA) 53 Y016 Local
20 YL5 GARST (USA) 54 Y017 Local
21 Y011 GARST (USA) 55 WM1 Local
22 WR1 GARST (USA) 56 WM2 Local
23 WR2 GARST (USA) 57 W03 Local
24 Y01 Local 58 WI1 Local
25 Y02 Local 59 WP1 Local
26 Y03 Local 60 WP2 Local
27 YF1 Local 61 W02 Local
28 YF2 Local 62 W04 Local
29 YF3 Local 63 YR4 Thailand
30 Y04 Local 64 YR6 Thailand
31 Y05 Local 65 YR7 Thailand
32 Y06 Local 66 YS2 Thailand
33 Y07 Local 67 YS4 Thailand
34 YM1 Local 68 Y015 Thailand
Y = yellow, W = white, M = M heterotic group, F = F heterotic group, 1= Il37TN, R = Reid, L =
Lancaster, S = Suwan, P = Potch Pearl, K = K64R, 0 = Other heterotic groups
The collection was planted at Bapsfontein Research Station in the last week of
October and harvested in May of the following year. Each plot consisted of a single
14
S.7m long row, with 0.96m-row spacing. Plots were planted at 30 kernels per plot
row and thinned to 20 plants per plot. This produced an average plant density of
approximately 30 000 plants per ha. Two guard rows were planted around the trials,
but not between plots. Weeds, insects, and foliar diseases were controlled as follows:
Lamba-cyhalothrin (pyretroid)(6g aiha-l)(insecticide), Sulcotrione, Atrazine and other
Triazines (200g aiha", 46S,6g aiha", l4,4g aiha-l)(herbicide) and Acetochlor
(Chloroacetanilide)(840g aiha-l)(herbicide) were applied at planting, followed by 6g
aiha-I Lamba-cyhalothrin (Pyretroid) and 200g aiha-l Sulcotrione after planting.
Asetochlor (Chloroacetanilide)(840g aiha-l)(herbicide) was applied once after
planting. Supplementary irrigation was applied during the growing season.
Morphological parameters
A total of 25 morphological and developmental characters were scored for all entries
during the growing season. The plants were maintained and self-pollinated. The data
was collected from five individual plants from the middle of each row and the
averages were determined. Three to four months after planting the following traits
were measured:
1. Plant height (m): Height from the base to the lowest tassel branch determined
from five plants that were visually judged to represent an average of each plot.
2. Ear height (m): Height from the base to the ear node determined from five plants
that were visually judged to represent an average of each plot.
3. Ear prolificacy: The number of ears per plant, to represent an average of the five
middle plants of the plot. (With a plant density of 30000 and the narrow row
width, ear prolificacy may have been underestimated.)
4. Tillers: The presence or absence of tillers in the plot.
5. Tassel colour: The observation of colour variations ranging from cream to purple.
6. Silk colour: The observation of colour variations ranging from white to purple.
7. Flowering date: The number of days from planting to 50% pollen shed.
8. Leaftype: The direction (upwards or hanging downwards) of leaf growth.
9. Rust: Scoring on a scale ofO-9, where 0 is susceptible and 9 is resistant.
10. Stalkborer: Scoring on a scale ofO-9, where 0 is susceptible and 9 is resistant.
35
11. Germination (%): The number of plants that germinated, as a percentage of the
number of kernels planted.
At harvest time, plant and ear characteristics were recorded as follows:
12. Percentage root lodging: The number of plants leaning more than 30° from the
vertical at harvest divided by the total stand of each plot.
13. Mass per ear (g): The average mass per ear of the selected ears.
14. Ear length (cm): The length from the base to the tip of the ear, determined from
five plants to represent the average of each plot.
15. Ear width (cm): The width in the middle of the ear, determined from five plants to
represent the average of each plot.
16. Kernel colour: The intensity of white or yellow respectively.
17. Kernel rows: The number of kernel rows at the base of the ear, determined from
five plants to represent the average of each plot.
18. Diplodia ear rot: The presence or absence ofDiplodia ear rot.
19. Low ear: If the height from the base to the ear node is low enough to cause ear
damage by rodents, birds etc., or not.
20. Wide inter row spacing: The presence or absence of wide kernel rows.
21. Kernel abortion: The presence or absence of aborted kernels.
22. Tapering ears: The presence or absence of an ear that becomes narrower from the
base to the tip of the ear.
23. Grain type: Flint or dent type.
24. Cap: Presence or absence of a cap (starchy endosperm).
25. Cob colour: The observation of colour variations ranging from white to purple.
3ó
Computer analysis
For qualitative characters, presence (0) or absence (1) was scored and summarised in
an input matrix (Table 3.6). For quantitative characters, averages were calculated and
individual values scored for being above (1) or below (0) the average. All subsequent
calculations were performed using the computer programme PHYLIP (Phylogeny
Inference Package)(Felsenstein, 1993).
The index of genetic similarity (F) of Nei and Li (1979) was used to calculate
pairwise genetic distance (D) for all lines:
D = I-F
where Nxy= the number of shared traits between any two lines x and y and N,and Ny
are the number of traits for lines x and y, respectively (Wang and Tanksley, 1989).
Unweighted Pair Group Mean Arithmetic Analysis (upGMA)(Sneath and Sokal,
1973) and Neighbour joining methods were used for cluster analysis of the pairwise
distance matrix representing the distances among the maize lines, based on the
similarity matrix. To estimate the reliability of the analysis, a thousand bootstrap
replicates were performed and a consensus tree was drawn.
Results
Morphological data
Morphological data are provided in Tables:
37
'fable 3.2 Morphological traits scored three to four months after planting, for white
inbred lines
NQ Line Plant Ear Ear Tillers Tassel Silk color FD Leaf Rust Stalk Germination
height height prolificacy color (days) Type borer %
1 WM1 2-3 0 wIp P 96 upright 8 7 56.67
2 WM2 1.83 0.65 2 0 creamlp p 101 floppy 7 8 56.67
3 WF1 1.70 0.55 1-2 0 creamlp w 105 floppy 7 8 73.33
4 WF2 1.15 0.38 1-2 0 pur w 94 upright 7 8 56.67
5 WI1 1.67 0.50 0 oly w 102 upright 7 8 26.67
6 WI2 1.40 0.44 1 0 P p/w 97 upright 7 9 60.00
7 WP1 1.60 0.80 2 0 w w 109 upright 7 8 50.00
8 WP2 1.60 0.73 1-2 0 P w 111 upright 8 8 60.00
9 WP3 2.03 0.75 0 wIp w 103 upright 6 8 80.00
10 WR1 1.67 0.65 0 o/p/w w/p/pur 79 upright 6 9 63.33
11 WR2 1.64 0.60 0 w w 79 upright 5 8 80.00
12 WK1 1.41 0.50 1-2 0 w w 103 upright 8 8 63.33
13 WK2 1.68 0.60 2 0 wIp wIp 106 upright 8 8 46.67
14 W01 1.50 0.50 1-2 0 w wIp 97 floppy 7 7 83.33
15 W02 1.83 0.60 2 0 p/pur w 102 upright 6 8 40.00
16 W03 1.87 0.80 2-3 0 P P 113 floppy 6 8 66.67
17 W04 1.67 0.60 2 0 w/o wIp 100 floppy 8 7 60.00
18 W05 1.50 0.57 0-1 0 olp wIp 123 upright 8 8 80.00
19 W06 1.54 0.60 0 w pur 105 upright 8 9 76.67
20 W07 1.40 0.70 0 w w 102 upright 7 8 93.33
21 W08 1.50 0.65 1-2 0 p/w 115 floppy 6 8 36.67
22 W09 1.60 0.70 1-2 0 w w 94 floppy 5 9 80.00
23 W010 2.30 0.90 0 w P 109 floppy 7 8 86.67
24 W011 1.67 0.66 1-2 0 olp plpur 111 upright 8 8 46.67
FD = flowering date, wIp = white/pink, pur = purple, oly = orangel yellow, p = pink, w = white, plpur = pink/purple
Table 3.3 Morphological traits scored three to four months after planting, for yellow
inbred lines
No Line Plant Ear height Ear Tillers Tassel Silk FD Leaf Rust Stalk Germination
height (m) prolificacy color color (days) Type borer %
m
YMl 1.50 0.50 2 o w wIp 98 floppy 8 8 43.3
2 YM2 1.55 0.70 2-3 o w P 89 floppy 7 6 73.3
3 YM3 1.66 0.59 2 o wIp 90 floppy 8 5 53.3
4 YM4 1.56 0.50 o w wIp 89 upright 7 9 76.7
5 YFl 1.43 0.41 1-2 o w w 94 floppy 8 8 73.3
6 YF2 1.17 0.57 2 o w/y w 79 floppy 7 6 73.3
7 YF3 1.22 0.34 1-2 o w w 94 floppy 8 8 60.0
8 YF4 1.76 0.53 2 o o w 93 floppy 8 7 63.3
9 Yll 1.30 0.42 1-2 w w 80 upright 8 6 56.7
10 YI2 1.39 0.40 o w w 89 upright 8 9 73.3
11 YI3 0.80 0.20 o w w 97 upright 8 8 70.0
12 YRl 1.38 0.40 o y w 89 floppy 6 5 43.3
13 YR2 1.74 0.61 1-2 o w wIp 85 upright 8 6 73.3
14 YR3 1.25 0.40 1 o olp wIp 82 upright 5 7 70.0
15 YR4 1.30 0.50 o oly wIp 86 upright 7 8 76.7
16 YR5 1.37 0.48 o w w 79 upright 5 6 63.3
17 YR6 1.40 0.40 1 w wIp 89 floppy 6 7 70.0
18 YR7 1.55 0.50 1-2 o w w/p/pur 78 upright 4 7 33.3
19 YL1 1.44 0.60 o w w 78 upright 7 8 53.3
20 YL2 1.33 0.49 o w w 82 floppy 8 8 56.7
21 YL3 1.55 0.55 o wIp wIp 78 upright 7 8 46.7
22 YL4 1.38 0.50 2 o ylo wIp 79 upright 7 7 73.3
23 YL5 1.30 0.60 1-2 o wIp wIp 85 upright 7 7 46.7
24 YSl 1.50 0.50 o oly w 95 upright 8 8 63.3
25 YS2 1.28 0.50 o w wIp 84 upright 6 8 66.7
26 YS3 1.OD 0.38 o w w 95 upright 8 7 46.7
27 YS4 1.20 0.50 2 o w w 89 floppy 7 7 53.3
28 YOl 1.47 0.48 o w w 91 floppy 7 8 83.3
29 Y02 1.34 0.40 o ylo w 89 upright 6 8 40.0
30 Y03 1.53 0.64 o o w 79 upright 7 6 73.3
31 Y04 1.46 0.60 o w/y w/p/pur 80 upright 7 8 73.3
32 Y05 1.44 0.55 w w 74 upright 4 6 53.3
33 Y06 1.40 0.45 o w w 91 upright 8 8 76.7
34 Y07 1.62 0.64 o w w 94 floppy 8 7 63.3
35 Y08 1.21 0.40 o w/y wIp 94 upright 8 8 76.7
36 Y09 1.60 0.60 o o wIp 93 floppy 8 7 73.3
37 Y010 1.40 0.45 o w 85 floppy 7 8 53.3
38 YOll 1.35 0.38 o w w 77 upright 6 7 50.0
39 Y012 1.18 0.48 o w w 85 upright 6 8 70.0
40 Y013 1.44 0.60 o pur pur 84 upright 6 8 50.0
41 Y014 1.34 0.40 o w/o wIp 84 upright 8 8 60.0
42 Y015 1.04 0.30 1 o w wIp 93 upright 6 8 63.3
43 Y016 1.92 0.80 2 (big) o w wIp 99 floppy 7 7 20.0
44 Y017 1.53 0.50 o w w 93 floppy 7 7 66.7
FD = flowering date, wIp = white/pink, pur = purple, oly - orangel yellow, p = pink, w = white, plpur = pink/purple
39
'fable 3.41Morphological traits scored at harvest, for white inbred lines
No Line % Root Mass Ear Ear Kernel Kernel Diplodia Low Wide Kernel Tapering Grain type Cap Cob color
lodging per ear length width color rows ear rot ear inter abortion ears
(g) (cm) (cm) row
1 WM1 0 42 150 45 offwhite 10 o o o o o dent 0 white
2 WM2 0 87 160 50 lemon 14 1 flinUdent 0 white
3 WF1 0 62 120 40 offwhite 10 o o o o flint 0 white
4 WF2 0 32 140 45 cream 10 o 1 o dent 1 white
5 WI1 0 54 165 45 lemon 12 o o o dent 0 white
6 WI2 0 31 145 45 offwhite 10 o o o o dent 1 white
7 WP1 0 19 100 35 white 10 1 1 o flint 0 white
8 WP2 0 19 100 40 offwhite 10 o o o o flint 0 white
9 WP3 0 23 120 45 opaque 10 o o o dent 0 white
10 WR1 0 100 140 50 white 16 o o o dent white
11 WR2 0 42 150 45 white 10 o 1 o o o dent 1 white
12 WK1 5 60 145 50 offwhite 12 o o dent 0 white
13 WK2 0 86 180 45 white 12 o o o o dent 0 white
14 W01 0 64 145 50 white 12 o o o o 1 flint 0 white
15 W02 0 99 190 50 lemon 12 1 o o flinUdent 0 white
16 W03 0 29 130 45 offwhite 12 o o o 1 1 dent 0 white
17 W04 0 39 120 45 white 10 o o o o o flint 0 white
18 W05 0 35 160 50 white 10 1 o flint o white
19 W06 0 94 165 50 off white 12 o 1 o dent 1 white
20 W07 0 53 150 50 offwhite 14 o o flint o white
21 W08 0 61 140 40 white 14 1 o o flinUdent white
22 W09 0 32 85 40 white 12 o o 1 o o dent white
23 W010 0 97 180 50 white 14 o o o o flinUdent white
24 W011 0 50 155 45 white 14 o o o o dent white
40
Table 3.5 Morphological traits scored at harvest, for yellow inbred lines
No Line % Root Mass Ear Ear Kernel eolor Kernel Diplodia Low Wide Kernel Tapering Grain type Cap Cob
lodging per ear length width rows ear rot ear inter abortion ears eolor
(g) (cm) (cm) row
1 YM1 8 30 60 45 yellow 12 1 o o o o dent 0 white
2 YM2 0 48 135 50 pale yellow 14 o o o 1 flint 0 white
3 YM3 0 20 120 40 orange 12 1 o 1 dent 0 white
4 YM4 13 45 120 35 orange 10 o o 1 flint 0 white
5 YF1 9 36 125 40 yellow 12 o o o o dent 0 white
6 YF2 0 22 125 35 yellow 10 o 1 o o dent 0 white
7 YF3 28 24 135 30 yellow 10 1 o o o o flinUdent 0 white
8 YF4 0 56 140 40 yellow 12 o o o o o flint 0 white
9 YI1 6 38 110 45 yellow 14 o o o o flint 0 white
10 YI2 0 58 100 40 yellow 10 o 1 o o dent white
11 YI3 0 16 110 40 yellow 12 o o o o o flinUdent white
12 YR1 0 29 150 35 light yellow 12 o 1 dent 0 red
13 YR2 0 61 135 45 dark yellow 12 1 o dent red
14 YR3 0 42 110 40 yellow 14 o 1 1 o o dent red
15 YR4 0 42 130 35 yellow 14 1 o o o dent pink
16 YR5 0 50 105 45 yellow 14 o o o o o dent 1 white
17 YR6 0 36 100 40 orange 10 1 o o o o flint 0 white
18 YR7 0 47 120 45 yellow 12 o 1 1 dent 0 red
19 YL1 0 28 135 40 yellow 10 o o o o dent white
20 YL2 0 78 130 40 yellow 10 o o o dent red
21 YL3 0 49 100 50 yellow 14 1 o o o o dent red
22 YL4 0 26 160 40 yellow 12 o o o o dent 1 white
23 YL5 0 62 150 45 yellow 12 o o 1 dent 0 white
24 YS1 0 25 130 45 orange 16 o o flint 0 white
25 YS2 o. 33 110 35 orange 12 o o o flint 0 white
26 YS3 0 34 115 35 orange 10 o 1 o 1 1 flint 0 white
27 YS4 0 46 140 55 orange 16 o o o o o flint 0 white
28 Y01 12 16 95 40 yellow 12 o o o o flinUdent 0 white
29 Y02 17 45 140 40 orange 12 o o 1 flint 0 white
30 Y03 0 20 125 35 yellow 10 o o o dent red
31 Y04 5 44 150 40 orange 12 1 flint red
32 Y05 0 75 145 40 yellow 14 o o dent pink
33 Y06 0 71 170 50 yellow 12 o o dent white
34 Y07 0 45 130 45 orange 12 o o o o dent white
35 Y08 0 24 95 45 orange 12 1 o o 1 dent white
36 Y09 5 55 135 45 light yellow 10 o o o o dent white
37 Y010 0 48 125 40 yellow 12 o o o o dent white
38 Y011 0 42 125 45 yellow 10 o o o o denUflint red
39 Y012 0 50 125 40 orange 12 o 1 o o flint 0 red
40 Y013 0 57 130 40 orange 10 o o o o flint 0 white
41 Y014 0 32 115 35 orange 12 o o o o o flint 0 white
42 Y015 0 40 100 35 dark yellow 10 o o o o flinUdent 0 white
43 Y016 0 74 185 40 orange 12 o o 1 o flint 0 white
44 Y017 0 30 130 40 dark yellow 12 o o o o denUflint white
41
Phenotypic analysis
The data matrix of morphological markers scored for genetic distance analysis is
provided in Table 3.6 and 3.7.
Table 3.6 Input data matrix used for UPGMA and Neighbour Joining analysis of 25
morphological markers
YM1 1111121121222222221212222 Y08 2112222221122122211212211
YM2 2111121112221112121121121 YOg 1111122211121221121221221
YM3 2112122111212111222211222 Y010 2211212221111221222222122
YM4 1112121222221121212222111 YOll 2221211222121221221221112
YF1 1111211121222222222212221 Y012 2222211222221221212222111
YF2 2211212111222122212221121 Y013 2212222212221222222212112
YF3 1111211122222222122212222 Y014 1212222222222222222222212
YF4 2111112122221222122221221 Y015 2111221222222222222212111
YIl 1211211122212122221121212 Y016 2112121212221121122221122
YI2 2111211221121221222212211 Y017 2111111222122122222221121
YI3 2111111222122222222222211 WMl 2112222121221222121221222
YR1 2121212221222111112221122 WM2 2111122112221111111112122
YR2 2221121111121111121211211 WF1 2112112112221222222212121
YR3 2221222221121221212121111 WF2 2112212121122121111222112
YR4 2221222221121122222112111 WIl 2111112221221212111222112
YR5 2211211221121222221121111 WI2 2112222221122212121222112
YR6 2112221222212222222211121 WP1 2112222122222112212212112
YR7 2221121121221111212221112 WP2 2112112112222212222222212
YL1 2211211211122222122212112 WP3 2112112211222221221212121
YL2 2121211221121221222212222 WR1 2212122211121212221112111
YL3 2221122211121222221112112 WR2 2212111211121222111222111
YL4 2211222121122122122221111 WK1 1112211121221211121212211
YL5 2211222111221111121221112 WK2 2112122111221212121222212
YS1 2112112222222212211112211 WOl 2112121122221122121221121
YS2 2212221222222122212222111 W02 2111112112221211121222112
YS3 2112211222222112212221212 W03 2112122111222112221222121
YS4 2112211122221222121121122 W04 2112122112222222221221222
Y01 1111211222222222222212121 W05 2112122212222111121212211
Y02 1112212222221111122222112 W06 2112121211121211111222211
Y03 2221112211122i21222222121 W07 2112211212221112121112111
Y04 1222221212121111112211111 W08 2112122112121212212112122
Y05 2221211211111211112111112 WOg 2112111111122221222222121
Y06 2111211221121121121212211 W010 2112121212121212121122121
Y07 2112111211121222221211221 WOll 2112122111121212121122212
42
I !i-_
!. i
'I· I, j iI i· ..j 1 ~~~i
?; j i I·1 I_ I !~ F':'. I
I ! I 1
I
I <=.JIG ~ :1
I"F!:.':
_.,
,I 1- ~,l~
, I
, s :S~!S:, .], el:~ ~ I !:::':",
i
1, ] 1 ;,. '<:~s;e S.S.~
~.J ~ 1;:=!"!;:'. ~:~i~~(
· fI , I''j .' I I
" ! i i
'I I ·1
i I I
'Ii 1I,·1' i! i
I; 1 li !
I I
i i
'1 I I
'I jl
'I ! \ I
'i, I I i
! I 1
H ! i I
I I ! I
'I I I ,
l , JIIiIlli \ I
"I') I..i
ii I i i I
': I i I I
i I I
'1 i I
"i II
i
i i
'i Il ii
!j \ ! i
'I li'
I ,II
'j ~ I I I
'.I I: II I, i
'I i I I
1i iOl
'I I i
! I !
')il IiI i
I i I !
i I f r
+, j'l I ij i
I I I i
t ••
II !L
I, !
j.
i
The resulting dissimilarities were subjected to clustering by the genetic distances. The
values on the phenograms represent the genetic distances among different inbred
lines. To estimate the reliability or effectiveness of the analysis, a thousand bootstrap
replicates were performed. A thousand trees were produced and the computer
programme determined a consensus tree (Fig 3.3 and 3.4). The consensus tree gives a
good idea of how effective the total analysis is. The numbers at the forks indicate the
number of times the group consisting of the species which are to the right of that fork
occurred among the trees, out of 1000 trees.
Based on the Neighbour joining dendogram and consensus tree, the lines were
grouped into 21 major clusters (Fig. 3.1 and 3.3). Clusters from different trees were
consolidated and those occurring in both trees were used. When comparing the
dendogram (using Neighbour joining) and Consensus (using Neighbour joining)
generated trees it is evident that these trees are similar.
In cluster I, there were two subgroups. Subgroup a consisted of lines YO 13 and
Y014. Subgroup b contained lines YR4, YR3, YL3 and WRl. In the subgroup b,
lines YR4 and YR3 were closer related to each other, and YL3 and WR1 were closer
related to each other. In the dendogram the two subgroups were separated and in the
consensus tree the two subgroups were grouped together. In general the genotypes of
this cluster had pink or purple tassel and .silk colour, an upright leaf type and ears
longer than 130mm.
Cluster II had five lines: YF2, YL4, YL1, Y017 and YB. This cluster was divided
into two subgroups. Subgroup a consisted of lines YF2, YL4 and YL1, where YF2
and YL4 were closer related to each other than YLl. Subgroup b consisted of lines
YO 17 and YB. In the dendogram the two subgroups were separated and in the
consensus tree the two subgroups were grouped together. These lines had some
common characters, they had a white cob colour, no kernel abortion or wide inter row
spacing, and they had more than 12 kernel rows per ear.
44
+-----------&05
I XII +----11'011
I +----4
I I +Vlt2
I +-59
I I I XIII+-----802
I I +---43
I I +------BIl
-66-65
I I +-------lK2
I I XX+-12
I I +-29 +-----"8'010
I I I I
I +-51 +--------801
I I
I IXXI +--------WM2
I +-----1
I +------808
I
+-------YL5
I
+---14 +----------Ya1
I I I
I +-10 VII +-------------Y05
I I +---1
+-40 +--6 +---------Y04
I I I
I I +----------YR2
I I
I IlIa +--------Yr2
I +---20
+--49 +-------YL4
I I
I I +----------YL1
I I I
I I I +----Y1l4
I I I +-----2
I +-38 Ib I +-------Y1l3
I I +---13
I I I I +---01
+-55 " +---5
, +-31 +------Y13
I I
I I +---------Y03
, +--17
I +--------Y010
I
I +-------812
I I
+-51 +--------------YRl
, +-29
+-44 +--------Y02
I
+---------&?2
VI +-------'YM4
+----19
I I +--------Y012
I +-18
+-61 +-49 +----YS2
I I I
I I 'la +-------Y014
I I +--25
I , +----Y013
I I
I I +-------Y1l6
I I +---22
I +-52 , +--Y015
I , , IV I
I , I +-36 +----R3
I I I " +--3
I I I , , +---9 +---Yr1
I I , I I , I
I I I I +--24 +--Y01
I I I I I
I I +-47 +-------'YM1
I I I
I I I XVIII+-----------------Yl1
I +-54 I +- ---21
I , " I +-----YR5
+-62 I " +--11
I I I +-39 +-------YOll
I I I
I I 'IXb +---YI3
I , +---23
I , +-----Y017
I I
I I xvU+-------IIP1
I I +-26
I I I +--------YB3
I +-50
I I +----------YS1
I +-33
I +---------Y08
I
I +-----YL2
I XIV +--8
I +--15 +-n2
I I I
I +----30 +-----Y06
+-64 I I I
I , +----------1I'1U.
+-53
I +-----------Y09
I +-32
I I +-------Y07
+-42
IVIIl+-----&R2
+---16
+-------806
+-------V!'4
+--41
I IXVI+------mll
I +--35
+--515 I +-------YS4
I I +-27
" +----801
I I
I +-----------Y016
I
+-153 +------809
, +-37
I +--45 +-----VP3
I I I
I I +------11'1'1
I I
+-60 +----&04
I +-34
I +-415 +-------------'lM3
lXIX' I
+--59 +-----1103
+--------vP2
Fig. 3.1 Dendogram generated by Neighbour joining analysis from morphological data for the 68 different maize lines.
45
+---------YL5
X +-32
+---39 +---------'19.1
I I
+-----58 +----------Ya.2
I I
I IVII+-----------Y05
I +--44
I +-----------Y04
I +----------YL1
I I
I +-40 +---YL2
I I I +--2
I xrvt +---10 +---YI2
I +-51 I
I I I +-----Y06
I I I
I I I +---------Y03
I I +-35
I I +---------YOI0
+--55
I XYIII+------YR5
I +--19
I I +------YOll
I +-39
I I lIb +------YR4
I I +--14
+-50 +------YR3
I
I Ib +-----WRl,
+------5
+-----YI3
+------YM4
VI +--5
+--20 +------YS2
I I
+-63 +-41 +-------Y012
I I I
I IlIa +------Y014
I I +---12
I I +------Y013
I +-53
I I I XVII+--------.P1
I I I +-21
+-6 I I I +-33 +--------YS3
I I I I I
I I +-48 +---------.P2
I I I
I I I +---------YSl
I I +-34
I I +---------Y08
I I
I I IV +-----YIl6
+-61 +---8
I I I +---YOI5
I I +--3
I +-30 +---YOl
I I I I
+-64 I I IIIb+-----YI3
I I I I +---7
I I I +-42 +-----YOI7
I I I I I
I I I I +---Yr3
I I I I IV +--4
I I I +-----9 +---YI'1
I I I I
+-52 +-----'4U
I
I +------&051
I +-13
I +-19 +------03
I I I
+--28 +-------VI'I
I
+---------Y07
I I
I I +-----------YRl
I I +-43
I I +---45 +-----------9J'2
I I I I
I +-60 +-----------Y02
I I
I IlIa +--------yp2
I +------23
I +--------YL4
I
+-65 +--------VI2
I +-22
-67 I I lXII +--fl011
I I I +-----1
I I +-31 +--BK2
I I I I
I I I IXIII+------&02
I I I +---16
I I +-49 +------vt1
I I I
I I I VIII+------n.2
I I I +---17
I I I I +------1i06
I +-54 +-46
I I I IXX+--------V010
I I I +-25
I I I +--------&01
I I I
I +-59 +--------------iJlC.
I I I
I I I +--------------Y09
I I I I
I I I I +--------n4
I +-56 +-21
I I I I +------9'04
I I I +-11
I I I I +------1iMl.
I I +--41
+-62 I +--------YS4
I I +-26
I I I I +-----YH2
I +-29 +--6
I I +-----1iOl
I I
I +---------YOI6
I
I XXI +--------&H2
I +----24
I I +--------"08
+-57
I +---------V05
I +-36
+--37 +---------1103
I
+----------~
+-----------------------YIl
Fig. 3.2 Dendogram generated by UPGMA analysis from morphological data for the 68 different maize lines.
46
re +----YOlJ
+------24.0
I +----Y014
I
+------------1.0 +----Y1l4
I I +-30.0
I I Ib I +----YD.3
I +--5.0
I I +----YL3
I +-33.0
I +----&lU
I +----YT2
I IIa +-15.0
I +--2.0 +----YL4
I I
+--1.0 +---------YLl
I I
I I IIb +----Y017
I +------30.0
I +----Y73
+--1.0
I I +----YFl
I I +-38.0
I I +-25.0 +----YF3
I Iuv I II I +---------n41
I +--4.0
+--1.0 I +----Y015
I I I +-30.0
I I +-22.0 +----YR6
I I I
I I +---------YOl
I I
I I v +----W!'1
I +----------------11.0
I +----YF4
I +----Y012
I VI +-18.0
I +-12.0 +----YS2
I I I
+------------1.0 +---------nt4
I I
I I VIII +----W06
I +------21.0
I +----02
I
I IX +----YD.l
I +-15.0
I I +----YR1
I +--2.0
I I lVII +----Y05
I , +-42.0
+------------4.0 +----Y04
I I
I I X +----nS
I +------15.0
I +----YR2
I XI +----V!'2
I +-----------19.0
I I +----YOe
I I
+--4.0-------1.0 +---------WI2
I I I +--1.0
I I I I lXII +----82
I I I I +-52.0
I I +--1.0 +----'11011
I I I
I I I XIII+----WIl
I I +------15.0
I I +----W02
I I
I I III +----W09
I I +-11.0
I +-----------------8.0 +----Y03
I I I
I I +---------WP3
I I
I +----YI2
I +-55.0
I +-25.0 +----YL2
I I I
+------------2.0 +---------Y06
I I
I +--------------Y010
I
I I xv +----0.1
I +---------------------10.0
I I +----Y02
I I
I I +----W01
I I +-14.0
I I I +----Y016
I +-----------------1.0
+-1$.0 I IXVI +----mtl
I I +-11.0
I I +----Y64
I I
I I XVII +----1rP1
I +---------------------15.0
I I +----YB3
I I
I I +----&010
I I +-12.0
I I XXI I +----&09
I +-----------------9.0
I I I +----1iIli2
I I +-32.0
I I +----'m2
I I
I I +----Y09
I +---------------------12.0
+100.0 I +----Y01
I I
I I XVIII +---------YI1
+-----------------5.0
I I +----YB5
I +-111.0
I +----Yoll
I
I xx +----&01
I +-10.0
+-----------------3.0 +----YS1
I
+---------w05
IX +----&04
+------------ ---- ---- ------15.0
+----1JP2
+ ------------ -------- ---- ------------ --&03
+--------------------------------------------'lM3
0.8 0.7 0.6 0..5 0.4 0.3 0.2 0.1 0.0
1-----1-----1-----1----1----+----[-----1-----1
Fig. 3.3 Consensus tree generated hy Neighbour joining analysis from morphological data for the 68 different maize inhred lines.
47
XI +---------&12
+--9.0
I I +----Y08
I +-20.0
+--1.0 +----R2
I I
IXVI +----YG4
+------23.0
+----iMl
+----»(3
+-----------10.0
I'XIV +----&05+--------------Y010
+--7.0
I , +---------Y06
I +-25.0
I ,+----Yl:2
, +-57.0
,'xv +----YL2+----BJQ
+------------7.0
, +----Y02
I
, III +----Y03
, +-18.0
, I +----W09
, +--7.0
I I IlIb +----Yl:3
, I +-44.0
+--1.0 +----Y017
I I
I IHa +---------YLl
I +--3.0
I I +----YIA
I +-18.0
, +----YI'2
lXIX +---------&03
+-------3.0
, +----VP2
+-13.0
+----'11'04
+----Y07
+------11.0
, +----9P3
+--1.0
I , VIII +----&015
I , +-23.0
, +--5.0 +----BR2
I I
I +---------Y09
I
I +----YOll
lXVIII +-30.0
+-------5.0 +----YllS
I I
I +---------Yl:1
I
+----&01
+-30.0
+------11.0 +----'m2
I I
I +---------Y0115
I
I ..-.---Ya3
I +-38.0
lIb I +----n.4
+-------4.0
I I +----YL3
I +-42.0
I +----1JRl
I
I V +----i1I'1
+-----------115 .0
I +----Y!'4
I
I +----YS2
I VI ..-.25.0
+------18.0 +----nt4
I I
I +---------Y012
I
I VI I +---- y05
+-----------20.0
I +----Y04
I
I XIII +----VI1
, +-19.0
I I +----1102
+-17.0 +-------1.0
I I lXII +----!iU
I I +-44.0
I 1 ... ----&011
I I
I I ..-.---W010
I I xx +--8.0
I +-------2.0 +----W07
I I I
I I +---------Y81
I I
I I la +----Y014
I +-----------26.0
I I +----Y013
I I
I I XVII +----YS3
I +-----------14.0
I I ..-.---BP1
+100.0 I
I I I X +----YL5
I +-26.0
I I +----Yil2
+-------1.0
I IIX +----YR7
I +-20.0
I +----YlU
I
I XXI +----VH1
+-----------22.0
+----W08
IV +----YR6
+----------------31.0
+----Y015
+----YI'1
N +-36.0
---------------28.0 +----Y!'3
I
N +---------Y01
+-- - --- ---- ---- - - - - ---- ---- ---- ---- nc.
u U U U ~ U U U M
[-----[-----[-----[----[-----[-----[-----[-----[
Fig. 3.4 Consensus tree generated by UPGMA from morphological data for the 68 different maize inbred lines.
4R
In the consensus tree cluster III consisted of lines W09 and Y03, but the lines were
not grouped together in the dendogram. Common features have been observed within
each cluster group. The genotypes of this cluster possessed wide inter row spacing,
pink silk colour, dent grain type, good germination, no kernel abortion and floppy leaf
type.
Cluster IV consisted of six lines: YF1, YF3, YM1, Y01S, YR6 and Y01. In both
trees these lines were grouped together, but the subgroups in the cluster were grouped
differently. This was the biggest cluster in both the trees. These lines had some
common characteristics, they had a white cob colour, no kernel abortion or wide inter
row spacing, and they had more than 12 kernel rows per ear.
Cluster V consisted of lines WF1 and YF4 in the consensus tree. In the dendogram
the lines were not grouped together. These lines had some common characters, they
had a white cob colour, no kernel abortion or wide inter row spacing, and they had
more than 12 kernel rows per ear.
Cluster VI comprised the lines YM4, Y012 and YS2. Line Y012 and YS2 were the
closest related within this cluster. The lines in this cluster both had upright leaf type,
there were more than 12 kernel rows per ear, orange grain colour and good
germination.
Cluster vn comprised lines Y04 and VOS. In the consensus tree this cluster was
closely related to cluster IX and in the dendogram this cluster was closely related to
cluster X. Common features that have been observed within the cluster are kernel
abortion, wide inter row spacing and susceptibility to stalk borer.
Cluster VITI consisted of two lines, W06 and WR2. In the consensus tree this cluster
was closely related to cluster VI and in the dendogram this cluster was closely related
to lines Y09 and Y07. The lines in this cluster both had upright leaf type, there
were more than 12 kernel rows per ear, orange grain colour and good germination.
49
In the consensus tree cluster IX consisted of two lines (YRI and YR7). In the
dendogram these lines were not grouped together. In general the genotypes of this
cluster had common traits such as kernel abortion, less kernel rows per ear and
susceptibility to stalk borer.
Cluster X consisted of only two lines, YL5 and YR2. In the dendogram these two
lines were also grouped together, but not alone like in the consensus tree. The
genotypes were characterised by kernel abortion, wide inter row spacing and
susceptibility to stalk borer.
Cluster XI consisted of three lines in the consensus tree: WF2, WI2 and Y08. In the
dendogram Y08 was not included and the other two lines were not grouped alone.
The major features of this cluster were white cob colour, no root lodging, no tillers,
upright leaf and purple tassel colour.
Cluster XII consisted of two lines: WOII and WK2. In the consensus tree this cluster
was closely related to cluster XI and XIII and in the dendogram this cluster was
closely related to cluster XIII, XX and XXI. The genotypes were characterised by
white cob colour, no root lodging, no tillers, upright leaf and purple tassel colour.
Cluster XIII comprised of the lines WIl and W02. Common features were observed
within this cluster group. The genotypes possessed white cob colour, no root lodging,
no tillers, upright leaf and purple tassel colour. Cluster XII and XIII were closely
related to each other.
In cluster XIV there were four lines in the consensus tree, YOIO, Y06, YL2 and YI2.
Line YL2 and YI2 were the closest related in this cluster. In the dendogram YO 1°
was excluded from the cluster. The genotypes of this cluster possessed pink silk
colour, presence of a cap, less than 12 kernel rows per ear, a dent grain type, less
kernel rows per ear and no kernel abortion.
50
In the consensus tree cluster XV comprised two lines, WK1 and Y02. The lines were
however not grouped together in the dendogram. The genotypes were characterised
by upright leaf type, root lodging, white cob colour, ear length less than 130 mm and
wide inter row spacing.
In cluster XVI there were only two lines, WMl and YS4, in the consensus tree. In the
dendogram a third line, WOl, was included in the cluster. The genotypes had ear
length longer than 130 mm, floppy leaf type, stalk borer susceptibility, Diplodia ear
rot tolerance and no cap.
Cluster XVII consisted of two lines, YS3 and WP 1. In the consensus tree this cluster
was grouped on its own and in the dendogram this cluster was closely related to lines
YS1 and Y08. The genotypes were characterised by flint grain type, upright leaf
type, less than 12 kernel rows per ear and tapering ears.
Cluster XVIII comprised of three lines, Yl1, YRS and Y011, in both trees. Line YRS
and YOll were closer related to each other than Yl1. The genotypes were
characterised by purple silk and tassel colour, upright leaf type, an ear width less than
40 mm, an ear length more than 130 mm, no kernel abortion and Diplodia ear rot
tolerance.
In cluster XIX there were four lines, W04, W03, WP2 and YM3. In the consensus
tree W03 and YM3 had less resolution and were not as closely related to WP2 and
W04 as in the dendogram. The genotypes were characterised by weight per ear of
more than 40 g, prolificacy, flint grain type, more than 12 kernel ears per ear and rust
tolerance.
Cluster XX had two lines, W07 and YS 1, in the consensus tree, but they were not
grouped together in the dendogram. The genotypes of this cluster possessed tapering
ears, good germination, upright leaf type, stalk borer tolerance, flint grain type and
susceptibility to Diplodia ear rot.
I 5 I
51
Cluster XXI comprised of two lines, W08 and WM2 in both trees. In the consensus
tree this cluster was closely related to lines YM2 and WO 10 and in the dendogram
this cluster was closely related to cluster XX and line YM2. The major features of
this cluster were flint grain type, high plant height, no tillers, kernel abortion and
floppy leaf type. Clusters la, llb, Ill, V, VII, vrn, IX, X, Xll, XIll, XV, XVII, XX
and XXI all only had two lines.
The clustering position of lines WOl, YOI6, WOlO, YM2, Y09, Y07, WOS and
YM3 were uncertain. The clusters' composition indicated absence of relation
between geographic and genetic diversity as estimated from morphological attributes.
Therefore the genetic relationship between the above mentioned lines are also
uncertain.
Based on the UPGMA dendogram and consensus tree, the lines were grouped into 21
major clusters (Fig. 3.2 and 3.4). Clusters from different trees were consolidated and
those occurring in both trees were used. When comparing the dendogram (using
UPGMA) and Consensus (using UPGMA) generated trees it is evident that these trees
are similar.
In cluster I, there were two subgroups. Subgroup a was composed of lines YO13 and
YOI4. The subgroup b consisted of lines YR3, YR4, YL3 and WRl. In both trees
the subgroups were not grouped together. In general the genotypes of this cluster had
pink or purple tassel and silk colour, an upright leaf type and ears longer than 130
mm.
Cluster II had five lines: YF2, YL4, YLI, YOl7 and YB. This cluster was divided
into two subgroups. Subgroup a consisted of lines YF2, YL4 and YLI, where YF2
and YL4 were closer related to each other than YLI. Subgroup b consisted of lines
YO17 and YB. In the dendogram subgroups a was grouped with lines Y02, WF2
and WRI and subgroup b was grouped with cluster lY. In the consensus tree the two
subgroups were grouped together with cluster ill. These lines had some common
characters, they had a white cob colour, no kernel abortion or wide inter row spacing,
and they had more than 12 kernel rows per ear.
52
Cluster III consisted of lines W09 and Y03 in the consensus tree, but the lines were
not grouped together in the dendogram. Common features have been observed within
each cluster group. The genotypes of this cluster possessed pink silk colour, dent
grain type, good germination, no kernel abortion and floppy leaf type.
Cluster IV consisted of six lines: YFl, YF3, YMl, YOI5, YR6 and YOl. In both
trees these lines were grouped together. However, in the dendogram the cluster
included subgroup lib in the cluster. This was the biggest cluster in both the trees.
These lines had some common characters, they had a white cob colour, no kernel
abortion or wide inter row spacing, and they had more than 12 kernel rows per ear.
Cluster V consisted of lines WFl and YF4 in the consensus tree. In the dendogram
the lines were not grouped together. These lines had some common characteristics,
they had a white cob colour, no kernel abortion or wide inter row spacing, and they
had more than 12 kernel rows per ear.
Cluster VI comprised of the lines YM4, Y012 and YS2. Line YM4 and YS2 were
the closer related to each other than Y012. The lines in this cluster had upright leaf
type, tolerance to stalk borer and Diplodia ear rot, there were more than 12 kernel
rows per ear, orange grain colour and good germination.
Cluster VII comprised of lines Y04 and Y05. In the dendogram this cluster was
closely related to cluster X. Common features that have been observed within the
cluster are kernel abortion, wide inter row spacing and susceptibility to stalk borer.
Cluster VIII consisted of two lines, W06 and WR2. In the consensus tree this cluster
was closely related to lines Y09, Y07 and WP3 and in the dendogram this cluster
was closely related to cluster xx. The lines in this cluster had upright leaf type,
tolerance to stalk borer and Diplodia ear rot, there were more than 12 kernel rows per
ear, orange grain colour and good germination.
53
Cluster IX consisted of two lines: YR I and YR7. In the consensus tree this cluster is
closely related to cluster X. In the dendogram these lines were not grouped together,
YR7 was grouped in cluster X and YRl was grouped with subgroup IIa. In general
the genotypes of this cluster had common traits such as kernel abortion, wide inter
row spacing and susceptibility to stalk borer.
Cluster X consisted of two lines, YLS and YR2. In the dendogram YR7 was also
included. The genotypes were characterised by kernel abortion, wide inter row
spacing and susceptibility to stalk borer.
Cluster XI consisted of three lines in the consensus tree: WF2, WI2 and Y08. Lines
Y08 and WF2 were closer related to each other than WI2. In the consensus tree this
cluster is closely related to cluster XVI. In the dendogram the lines were not grouped
together. The major features of this cluster were white cob colour, no root lodging, no
tillers, upright leaf and purple tassel colour.
Cluster XII consisted of two lines: Wall and WK2. In both trees this cluster was
closely related to cluster XIII. The genotypes were characterised by white cob
colour, no root lodging, no tillers, upright leaf and purple tassel colour.
Cluster XIII comprised of the lines WIl and W02. Common features were observed
within this cluster group. The genotypes possessed white cob colour, no root lodging,
no tillers, upright leaf and purple tassel colour. Cluster XII and XIII were closely
related to each other.
In cluster XIV there were four lines in the consensus tree, YOIO, Y06, YL2 and YI2.
Line YL2 and YI2 were the closest related in this cluster. In the dendogram Y03 was
included in the cluster and YLI was closely related to this cluster. The genotypes of
this cluster possessed pink silk colour, presence of a cap, less than 12 kernel rows per
ear, a dent grain type, no kernel abortion.
54
Cluster XV comprised of two lines, WKl and Y02, in the consensus tree. The lines
were however not grouped together in the dendogram. The genotypes were
characterised by upright leaf type, root lodging, white cob colour, ear length less than
130 mm and wide inter row spacing.
In cluster XVI there were two lines: WMl and YS4. In the consensus tree the cluster
is closely related to cluster XI. In the dendogram the cluster included YF4, W04,
YM2, WOl and YOI6. The genotypes had ear length longer than 130 mm, floppy
leaf type, stalk borer susceptibility, Diplodia ear rot tolerance and no cap.
Cluster xvrr consisted of two lines, YS3 and WPI. IIi the consensus tree this cluster
was grouped on its own and in the dendogram this cluster was closely related to line
WP2. The genotypes were characterised by flint grain type, upright leaf type, less
than 12 kernel rows per ear and tapering ears.
Cluster XVllI comprised of three lines, YlI, YR5 and YOII, in both trees. Line YR5
and YOII were closer related to each other than YlI. The genotypes were
characterised by purple silk and tassel colour, upright leaf type, an ear width less than
40 mm, an ear length more than 130 mm, no kernel abortion and Diplodia ear rot
tolerance.
In cluster XIX there were three lines, W04, W03 and WP2, in the consensus tree. In
the dendogram these lines were not grouped together. The genotypes were
characterised by mass per ear of more than 40 g, prolificacy, flint grain type, more
than 12 kernel ears per ear and rust tolerance.
Cluster XX had three lines, W07, WO 10 and YS 1, in the consensus tree. There were
only two lines, W07 and WOlO, grouped together in the dendogram. The genotypes
of this cluster possessed tapering ears, good germination, upright leaf type, stalk borer
tolerance, flint grain type and susceptibility to Diplodia ear rot.
55
Cluster XXI comprised of two lines, W08 and WM2 in both trees. In the consensus
tree this cluster was grouped on its own and in the dendogram this cluster was closely
related to lines WOS, W03 and YM3. The major features of this cluster were flint
grain type, high plant height, no tillers, kernel abortion and floppy leaf type. Clusters
la, llb, III, V, VII, VIII, IX, X, XII, XIII, XV, XVI, XVII and XXI all only had two
lines.
The position of lines YM3, WOS, Y07, WP3, Y09, YM2, WOl and Y016 was
placed out of any cluster. The clusters' composition was indicating absence of
relation between geographic and genetic diversity as estimated from morphological
attributes. Therefore the genetic relationship between the above mentioned lines are
also uncertain.
If the Neighbour joining and UPGMA consensus trees are compared with each other,
there were a few differences. In the Neighbour joining tree lines YFl and YF3
clustered close toYMl, but in the UPGMA tree they clustered close to YOl (in
neighbour joining tree lines YR6 and YO 15 clustered close to YO 1). If the grouping
oflines Y012, YS3 and YM4 were compared in the two trees, Y012 and YS2 were
closest related of the three lines in the Neighbour joining tree, and in the UPGMA tree
YS2 and YM4 were closest related of the three. Lines Y03 and W09 were closest
related to WP3 in the Neighbour joining tree, and closest related to YB and Y017 in
the UPGMA tree. In the Neighbour joining treeYM3 was the most distant line in the
tree, and in the UPGMA tree YMl was the most distant related to other lines. It is
clear that the differences between the trees are minor. Common features were
observed within each cluster group. If the grouping pattern in the two different
analysis methods were compared, there were similar patterns observed, indicating
increased confidence in the constructed trees.
5ó
Discussion and conclusion
Morphological characters have proven to be a useful tool in the Zea mays
classification (Whythe and Gevers, 1988; Gerdes and Tracy, 1994; Lubberstedt et al,
1998). Looking at the Neighbour joining-, as well as the UPGMA dendograms and
cluster analysis, the lines were grouped very similarly. Therefore the morphological
characterisation method has gained confidence and certain patterns of grouping were
identified.
There was a correlation of origin with groupmg of inbred lines throughout the
consensus trees. There was a tendency of American inbred lines grouping together
(Cluster I, X and XVIII), local inbred lines grouping together (Cluster II, IV, V, VII,
XI, XV, XVII AND XIX), Thailand inbred lines grouping together (Cluster IV) and
Argentinean inbred lines grouping together (Cluster I). Although the inbred lines
differ in origin, it was revealed in the cluster patterns that the material in the Advanta
group has been interbred to a large extent, therefore a lot of overlapping of groups
was found.
The maize lines included in this study displayed high levels of polymorphism.
Calculation of pairwise genetic distances (Table 3.7) revealed an average distance of
8.08% (l.28 to 14.88) between the maize lines. These results suggest that the genetic
distance data based on morphological characterisation of maize lines can be a useful
tool when selecting the most suitable parental maize lines used to construct
populations and hybrids. However, this is not the only criterion for the selection of
parental maize lines used to construct populations and hybrids. Plant breeders
generally prefer to use recurring parents that have the highest number of desirable
traits (Hallauer, 1987).
A problem that has been experienced is that in general when breeders make selections
in field breeding programmes, they are based on intimate knowledge of the line, as
well as selective characteristics. However, the history of the inbred lines used in this
study, was unknown and it was not a natural population where selection occurred
naturally, over a long period of time. This group was merely a selection of
commercial strains that were determined by their ability to cross and yield. These
57
lines were selected purely on the basis of their importance to the maize breeding
programme of Advanta Africa Seeds in South Africa. The aim of the genetic distance
analysis was to determine the levels of genetic variability that exists between a group
of potential recurrent or donor parents. There was a wide variability in the
morphological characteristics indicating a high degree of interspecific hybridisation,
assuming different characters could distinguish the genotypes. In this study all the
characters contributed equally, there were no specific major features used to group the
lines.
Salhuana et al. (1998) evaluated nearly 12 000 maize accessions in 12 countries to
identify the heterotic group in which the accessions belonged, using morphological
characterisation. Varying responses of the accessions to different environmental
conditions were recognised, and five homologous areas were defined according to
elevation and latitude. Therefore environment interaction also plays an important role
in determining the expression of the characteristics of the maize plant. Reports from
Goodman and Brown (1988), Smith and Smith (1989), Lee (1995) and Lubberstedt et
al. (1998) have shown that characterisation of maize germplasm accessions should be
based on descriptors that are not influenced by the environment. For this reason it is
also problematic to use morphological characterisation.
Reports on morphological characterisation of maize in Africa are fairly scarce,
because evaluation of maize lines in extensive field tests is the most costly and time
consuming part in modem hybrid breeding of grain and forage maize (Boppenmaier et
al., 1992). Therefore, maize breeders often need to make a "by-chance" selection in
breeding programmes, based solely on prior experience. However, the approach used
in this study makes it possible for the breeder to evaluate morphological characters in
a scientific manner to determine which inbred lines crossed will give good combining
ability. Using morphological characters in maize breeding will remain priority
especially when the characteristics are sought after in new lines.
In conclusion, the results of this study revealed that definite groups could be defined
based on morphological data. These groupings are based on similarity in
morphological characters. Therefore breeding lines that cluster together would not
result in good combining ability if crossed and should not be used to make crosses.
5R
Maize lines should have the largest possible genetic distance between them, even if
based on morphological characters, to yield the best hybrids. Measurements of
genetic distance facilitate identification of diverse parents in order to maximise the
expression of heterosis (Smith et al., 1990). The statistical analysis of morphological
characters will allow the breeder to make objected choices for the selection of
parental material to produce new hybrids. When looking at improving the variation
within populations, in other words the source of parental material, inbred lines from
the same groupings would be crossed with each other. This would enable the breeder
to increase variability within the parental material. The dendograms in this study
showed significant grouping, but more objective methods are needed to confirm the
results. These results will however contribute to the breeding programme, since it
will make the possibility of higher combining ability in crosses, as well as improved
populations, higher.
59
CHAPTER4
J[JIlIENTlIWliCA '[lION ANJI]) GJENlE'[J[C lIJ)ITS']Af'NClE
ANA1L )YSJ[S OW MA.JZIE llNlRRJElIJ) 1LllNES USING AWL]?
FIrN G JE1R1PJRllN'[ITNG
Abstract
In maize (Zea mays L.) breeding programmes it is important to determine the
genetic relatedness between inbreds, as the amount of relatedness will directly
affect the amount of heterosis expressed in the resulting hybrids. The objective of
this research was to utilise AFLP's (Amplified Fragment Length Polymorphic
markers) to characterise and determine groupings of 50 maize inbred lines (both
flint and dent) from the Advanta Africa genebank (representing five countries).
The AFLP data was coded using binary character states and analysed using
PHYLIP (Phylogeny Inference Package) Version 3.5p (Felsenstein, 1993). AFLP
fingerprinting of maize inbred lines allowed the efficient detection of large
numbers of polymorphic amplified fragments. The inbred lines were clustered
into two major clusters. Both clusters were further divided into two subgroups,
which were divided into a number of minor groups. Correlation was found using
different methods of analyses. These results have identified genetically distant
related lines, which ifused in breeding programmes will increase the expression of
heterosis in hybrids. The incorporation of molecular markers in a breeding
programme has the potential to reduce cost and time.
Introduction
The development of techniques for analysis and quantification of genetic
relatedness has generated a variety of new applications that can be used to
complement classical plant breeding techniques. A variety of DNA fingerprinting
techniques is presently available, most of which use PCR for detection of
fragments. The identification of lines, estimation of genetic distance, phylogenetic
analysis, genome mapping and gene tagging in particular, have benefited from the
60
application of DNA fingerprinting techniques (Yu et al., 1993; Nybom, 1994; Vos
et al., 1995).
Amplified fragment length polymorphism's (AFLP's) is a PeR-based method
which first involves restriction digestion of the genomic DNA. Adapters are
ligated to the ends of the restriction fragments and pre-selection is performed with
two AFLP primers having a single selective nucleotide, followed by a second
amplification using primers having longer selective extensions (Zabeau and Vos,
1993; Vos et al., 1995). The resulting amplification products are separated on a
sequencing gel and can be visualised using radioactive or fluorescent labelling.
Studies suggest that AFLP's are comparable to RFLP's (Restriction Fragment
Length Polymorphisms) in terms of reproducibility and should therefore be highly
suited to use in experiments (Jones et al., 1997).
AFLP analysis has found widespread application in agriculture (Donini et al.,
1997; Pakniyat et al., 1997). It is more cost-effective, reproducible and readily
automatable for high marker output than RFLP's and RAPD's (Perkin-Elmer,
1996; Van Toai et al., 1997; Barrett and Kidwell, 1998). AFLP analysis tends to
produce more informative data points and does not have some of the limitations
associated with RFLP's, RAPD's and SSR's (Keim et al., 1997). The AFLP
technique does not require sequence data or access to probe libraries and is,
therefore, accessible to research environments previously restricted by these
requirements (Mazur and Tingey, 1995).
The accurate identification of plant breeding material is extremely important, not
only for protection of breeder's rights, but also to drive plant-breeding
programmes. DNA-based identification methods have proved useful in cases
where line identification is based on mature plant characteristics such as flowering
or fruit ripening and where growth conditions may influence plant qualities
(MorelI et aI., 1995). AFLP analysis has been used to establish phylogenetic
relationships among species, subspecies and lines (Pakniyat et al., 1997) and to
measure genetic variation in populations (Barrett and Kidwell, 1998) and species
(Van Toai et al., 1997).
61
Since morphological variation among maize lines is not very distinct, phenotypic
identification based on morphological traits is difficult. AFLP's have been used as
an alternative for line identification in some crop species. The aim of this study
was to use AFLP fingerprinting to identify and determine the genetic distance
between inbred maize lines from the Advanta Africa gene bank. All 68 maize
inbred lines are currently in use in maize breeding programmes at Bapsfontein
Research Station, South Africa, with the aim to make crosses with the best
combining ability. Morphological data from this material was collected during the
1998/1999 season.
Material ami Methods
Plant maternal
A total of 50 maize inbred lines were used in this study. The material was
obtained from the Advanta Africa genebank (Building number 9, Old Trafford,
Corner of Leith and Trichardt Road, Bartlett, Boksburg). The maize inbred lines
used in this study are summarised in Table 4.1.
The germplasm used included lines from Africa (local), Argentina, Zimbabwe,
Garst (USA) and Thailand. Six individuals (three per pot) of each inbred line were
grown under glasshouse conditions at University of the Orange Free State
(UOFS), in Bloemfontein, South Africa. Growth conditions were set at 20°C night
temperature and 26°C day temperature.
62
Table 4.1 Maize lines used for AFLP analysis.
NQ Lines Origin NQ Lines Origin
1 Y012 Argentina 26 YM1 Local
2 Y013 Argentina 27 YM2 Local
3 W08 Brazil (tropical) 28 YM3 Local
4 W09 Brazil (tropical) 29 YI2 Local
5 W010 Brazil (tropical) 30 YS3 Local
6 W011 Brazil (tropical) 31 YF4 Local
7 W05 (CIMMYT) Zimbabwe 32 W01 Local
8 W06 (CIMMYT) Zimbabwe 33 WF1 Local
9 W07 (CIMMYT) Zimbabwe 34 WF2 Local
10 YM4 GARST (USA) 35 WI2 Local
11 YL2 GARST (USA) 36 WP3 Local
12 YR5 GARST (USA) 37 WK1 Local
13 YL1 GARST (USA) 38 WK2 Local
14 YL4 GARST (USA) 39 Y09 Local
15 YL5 GARST (USA) 40 Y010 Local
16 Y011 GARST (USA) 41 WM1 Local
17 WR1 GARST (USA) 42 WM2 Local
18 WR2 GARST (USA) 43 W03 Local
19 Y01 Local 44 WI1 Local
20 Y02 Local 45 WP1 Local
21 YF1 Local 46 WP2 Local
22 YF2 Local 47 W02 Local
23 YF3 Local 48 W04 Local
24 Y04 Local 49 YR6 Thailand
25 Y06 Local 50 YS2 Thailand
ó1
DNA extraction
DNA was extracted from young fresh leaves using the modified DNA isolation
method described by Dellaporta et al. (1983). Plant material was homogenised
using a mortar and pestle. The leaf material was frozen in liquid nitrogen. The
macerated tissue was transferred to a 50 ml centrifuge tube containing 5 ml
extraction buffer (100 mM Tris-HCI, 500 mM NaCl, 50 mM EDTA, 1.25% (m/v)
SDS, 20mM Na2S205, pH 8.0) and mixed vigorously. The homogenate was
incubated at 65°C for 20 minutes, with periodic shaking every ten minutes.
Thereafter chloroform-isoamylalcohol extraction was performed using 5 ml
chloroform-isoamylasetate (24:1). Centrifugation of the extraction was performed
at 10 000 rpm for ten minutes at 20°C. After three or four extractions the DNA
was precipitated using isopropanol (l:1 v/v). The DNA was spooled and washed
in 70% Ethanol. The DNA pellet was resuspended in one ml of sterile water
(Sabax, Non Pyrogenic). The DNA concentration was determined
spectrophotometrically. The DNA samples were diluted to a final concentration of
250 ng/ul and aliquoted for storage at -20°C.
Primers
AFLP reactions employed two oligonucleotide primers, one corresponding to the
PstI-ends and one corresponding to the Msel-ends. The PstI-primer was
fluorescentJy labelled. DNA was digested with MseI (frequent 4-base cutter) and
PstI (rare 6-base cutter) as described by Vos et al. (1995). Three primer pairs
(Life Technologies Ltd., Scotland, UK) (Table 4.2) were used for AFLP analysis
of the maize inbred lines (Vos et al., 1995; Keim et al., 1997). Oligonucleotide
sequences used for adapters and primers are listed in Table 4.2.
64
Table 4.2 Oligonucleotide sequences used in AFLP analysis.
Primers
Msel: 5'-GAC GAT GAG TCC TGA G-3'
Pstl: 5'- CTC GTA GAC TGC GTA CAT GCA-3'
Adapters
MseAdl: 5'-GAC GAT GAG TCC TGA-3'
MseAd2: 3'-TAC TCA GGA CTC AT-5'
PstAdl: 5'-CTC GTA GAC TGC GTA CAT GCA-3'
PstAd2: 3'-CAT CTG ACG CAT GT-5'
PstPI: 5'-GAC TGC GTA CAT GCA C-3'
PstP2: 5'-GAC TGC GTA CAT GCA A-3'
Primers labeled with 6-FAM labels:
PstAl: 5'-GAC TGC GTA CATGCA ATA-3'
PstA2: 5' -GAC TGC GTA CAT GCA CTC-3'
PstA3: 5'-GAC TGC GTA CAT GCA AGG-3'
Primers labeled with HEX labels:
PstA4: 5'-GAC TGC GTA CAT GCA ACC-3'
PstA5: 5'-GAC TGC GTA CAT GCA CAA-3'
PstA6: 5'-GAC TGC GTA CAT GCA CGA-3'
Pre-amplification and selective amplification
After adapter ligation, non-selective amplification (pre-amplification) of DNA
fragments was performed using non-selective (zero base extension) primers. The
peR products of the pre-amplification reaction were then diluted and used as
template for the second AFLP reaction using primers both having three selective
nuc1eotides. Selective amplification was carried out using various primer
combinations of primer MseI and primers Pst! (2-4 base pair extension) (Table
4.2) (Zabeau and Vos, 1993; Vos et al., 1995).
65
Pre-selective amplification reactions were as follows: 20 cycles of 30 sec at 94°C,
one minute at 56°C followed by one minute at noc. Selective amplification
reactions were as follows: 36 cycles of 30 sec at 94°C, 30 sec at 65°C reduced to
56°C using a ramp of -0.7°C per cycle, followed by one minute at noc. All
amplification reactions were performed in a Hybaid thermocycler (Vos et al.,
1995).
Gel analysis
Amplification products through the AFLP process were resolved by
electrophoresis on 2 % agarose gels (Seakem LE, FMC) in a TAB buffer (40 mM
Tris-acetate, one mM EDTA, pH 8.0). Electrophoresis was performed at constant
power, 110 W, for two hours. Amplification products were detected under UV
light after staining with ethidium bromide (0.5 ug/ml) (Vos et al., 1995). The
AFLP profiles were resolved using an ABI3 77 automatic sequencer.
Data colleetien and analysis
AFLP fragment data for three primer pairs was coded using a binary system and
summarised in a data matrix (Appendix, Table 8.1). The PHYLIP (Phylogeny
Inference Package) Version 3.5p (Felsenstein, 1993) was used to determine the
genetic distances and to group the inbred lines into clusters. Kimura (Table 4.3)
and Jukes and Cantor (Table 4.4) methods were used to calculate the genetic
distances between the inbred lines. The number of polymorphic bands was
determined (Barrett and KidweIl, 1998). The index of genetic similarities (F) of
Nei and Li (1979) was used to calculate pairwise genetic distances (D) for all the
lines:
F = 2NlI.)'/(NxNy)
D= I-F
Where Nxy= the number of shared bands between any two lines x and y and N, and
Nyare the number of bands for line x and y, respectively (Wang and Tanksley,
1989). Neighbour joining was used in cluster analysis of the pairwise distance
matrix. Support for clusters was evaluated by 1000 bootstrap replicates using
óó
Kimura and Jukes and Cantor algorithms with Neighbour joining analysis to
produce consensus trees (Barrett and Kidwell, 1998). The DNA parsimony
algorithm version 3.5p was also used to evaluate 1000 bootstrap replicates (Fig.
4.10).
Results
Quality of the genomic DNA
The DNA quality was determined spectrophotometrically, by taking readings at
260nm and 280nm. The 260: 280 ratio was l.8 or higher and the average DNA
yield was 1095.5 ug/ul. The quality of the DNA was good.
Fig 4.1 Gel illustrating the presence or absence of DNA after DNA extraction.
ó7
Identification of maize inbred lines
A line fingerprinting profile was compiled using unambiguously scorable AFLP
polymorphism's from the least number of primers allowing differentiation of all
the lines in the study. The fingerprinting profile was used to construct a molecular
identification key that displayed the sequence and combination of AFLP primers
needed to identify each of the lines in the study. Analysis of the 50 Zea mays
accessions with three AFLP primer pairs (Table 4.1) identified a total of 107
fragments. Within an individual plant the three primer pairs each detected
different AFLP profiles between 0.5 to 2.0 kilobases (kb). Of these, all 107 bands
were polymorphic (no monomorphic bands were determined). The average
number of bands detected per primer for all plants was 17.8. This also reflects the
high genetic similarity among the maize lines and necessitated the use of more
than one AFLP marker to distinguish any of the maize lines. Polymorphic loci
that distinguish all the lines used in the analysis allowed the lowest number of
primers for line identification.
Genetic distance analysis of maize inbred lines
Pairwise genetic distances (Table 4.3 and 4.4) based on the F statistic of Nei and
Li (1979), revealed an average distance of 4.38% (ranging from 0.09 to 8.67) for
all lines. The lines were originally from different sources and countries of origin
(Table 3.1). The biggest pairwise distance (8.67%) was observed between line
W02 and WP!. This is further illustrated by the fact that Neighbour joining
cluster analysis grouped them separately (Fig. 4.3). The smallest pairwise distance
was observed between WOl and W02, as well as YS2 and YOIO. These
calculations were based on the presence or absence of specific AFLP markers
among the lines (Appendix, Table 8.1).
The maize inbred lines were grouped into groups (major clusters), which in turn
were subdivided into subgroups and these subgroups were again subdivided into
minor groups. When looking at the clustering pattern, among the maize inbred
lines from group I, the average distance between them was 4.39% (ranging from
0.1 to 8.67). The lines from group II had an average distance between them of
fJll~~~li';~!-~~J~~~~=-=~'-~i=~::~~~~~~:~[T~~~~-L:=~~_-m~;__~~~~'~~~:-=J'
~t_::-~~---~f'I~-''---.--.= -', ._.-.. ----··=L._ --.=~--~r-_.--=~-. ::=--~-~-'-r--+--~--'-'~'==--:=====-===--=~==-=--=~~=~~~----~~=-'--=J
~;--~·t-!-i"f-'ïF~ t311 ;1&1' f0 -~-. ---.--. -~-----'.----',--' --.--. ---. -+-.. ..-~-----: --. -'. . " " ,.----,----- ..-----.--.------"--".-.i-- ,1c==f=E;~EE~~=~-=l-rt=:r:=--=~~=:_-~~=:=::~=~~~:~_~::~=~-J
: -·-·-:~1~·--{~--~,-·~;!~_-·::~-~-:--~--~tlf·---~=.;--· ..-------~--~-~-------.-._---'- .. ._----_ .._.- .._..--_-_._--_._j--~.--j_~. i.__:j__._.. ~-~._.'. .------ ...-.--.----....-.-..... - ......----..-.-..-.- ..- ..-...-.- .-.---
Wl-I--;E·--~I{-·I19~··I31-!]··4.-M-·i.II·-)!O'--llI--)}l-'-;T-i~C- G'--'~' ';---.-"-- _.-~:--T-'" - ...-,-~ ...-.- .-.-.-- ..-.~. -.-~--~-.:- .._, -- ','-' -- ~-.-,-.-----.--- ...--. . ..--
~~__:J:i Im_j["-4~ .l!, II1:J~ ~--4!_jll "ii 3IJ om 0--. '--.----,----.-. -.--.,- ~- ..__ • -. ~ •. -=--=::-- __ ===--~_====~~~-=:-=-.--c--.-----~
\\Ol 1313i31l3K13K13!1ll1US3m31l1!1II1UIUI 0'. .:,... ,
~·=_.:~=J:~-~~:3:=j~=jt=::~~i~=~1~;=-:j~:~~_=:_::i:t..~2f:.r-==!-f=~._-.~==~=:~·-=--.~==_:=_.=~_=~--.~:~~--._~_~~:~=~==-==--~~~.-.~.:__:_==~.:~=-.-"=~.~~~-~~~~ ...~~~.~~~..~.~, .===~-~_:~~~~
-m--. '-U,I--U-I -)(-Ij -.1-.1-1-,M.-I-J7.-I-II--.I-II--l.-IS-)-$-.Hl--lil-._li"' -l_55-.-_l.ro-'-..2.£-1-J13-- 0.-".-:.---, -..~-,t--• . . . ... .... -------.------------'---.-.---<
'!!!!_. _l'!__l;lI_J~__!_~_!!_~__J_97_J.!_3!__3_D_!:-:.~~-4~-.--J-~-JI£-!E~2113D! 2& 0 __ .. _
\\Ol 211 JM 1.111.11 151 3.1&115 3ll 3.15251' 2.m·3.e 3JJ JJJ 3JJ 2.193D! 251 2f17 0 :', . . . • ' . . . ,
~-_. _31..1JI_!,-j'~-.-!~--!~, UI. 3~._~i- 3ll~ 31!~.Jil:-.Ji,--311-=}12li Jwi._ _•2!Z2•9_1_.iU_,._2ffi 0_·· __C~.__·_ , .. _L_:_ ._.L_L_ ....;_~_ .._ '._-= ~=-=--. ~=__=-:~~~-~~~_==_=-==._._ ..j~ ---;Sm:00 !"--iUlIl-41•4--5•J-j-11~ 0;--~-;-$-sr~o·--'~11I-~11l-lml1:-IWS1'·g--431~-;B~j---ïD~i'"4i-ïrïm-ï£{,-ïll~rï.Is•--ï~uil'l- r ·o~-·''''--~_--_:._~_._+-._..c ._. .._._.,.- ..._· ._. --- .... -- ...--_. - -.- '- --. -- ----.-.- .. -- .. _- •
~ • 1~- 3:'ól~ 3J..J Ui ~S7 3~7,_~~l.iJ' ~. ï;-ï(_3U~X~ ___J_~,_j9__ ~~ JfJIjfli2_~~& __ 0__ ._~ __ . __ ' __ ._~_' __ __";___ ' _~-_._.----=~=:==--=-~---_.~
im---i:'il-i1il5s1i--III~I )i1{1-36~3-11' Jll 3i -3IIiiI4J~J-l-l-·t1li.-1i1-·l.~~·.ii1-ï.1ll1Ï4I]1ii:-I-IlIï~1ïï$'--42ï~mi'i'3-ïKir1-ï1~1-I1'i'3'l.$1C3-"3i1.i9ru'J;~:---03's-r-' UI II! 3.16 0 . ' --: ;-.~_, : ... ' . .,'-..---:-----;-----.---.--.---'..--.--•-.. .,...! --. ----~
~:··=~~:~~;:3~=·:;~=~;r-:l:=~~::~ --:~~-~::~~~~jr·1::=~·i-:-=:~f·;~.='-1-~fl~:I~:~-j,'~~~.t.-J-.-~=.~:~~~.:~==~-~_.--:==-~-~.-====..=.._=.~~~~~~=..:_=-::.~=-.~_.~ -.-..-,.-.-~
'ff! • liS liS 311 3m 210 3D! 139 l~ 3i l~ lM Im 311 35!/3.lJ Jll l.I3 3.153.IS2.19vs liS 3i ll6 3] 3ll 211 211 0 . • • • ' •. •.. '.
m • -4(_3i _!_!'_~Bi_ 3~. liT 5!5~'-~. iil 211.·~-UrUi:-3]'1ffi~ir!JrJi~·-;;sjr3i4~_ 3B "ii ;~ ii i'ïC3is-: -o·-·"--·--~-.-------~--. --: --·---~~==_=_=__j==_=_ l
~WlIIO.~:t~~.,:~~=3~.M~I]=1l.>:S~·Jl-Il-1:2f-113.:>'S-3·.9:7rJ~KI:J~lI-l={5~.4:6~:lK~lr~CJZ~l-l=4.1:1'~-l~4:1~i5;-I·]=~3.:5~:'n-l' ·~JDj-! -lI-!-l~4:J'l:=:J~!7~)M;r' :~-I~fII::;'i:4-~1~1::3·~fJ1:!·~-~l~IS,-,I-J-'~:0i~F_£~:·-- .=-=ó~=-c=~.•--. -.-~:-~-,-.~=.=.-~--~ '.:.~:=.=:-=~--~-~--====~
i·-. 1~'1~-·llï,-ï;rïr3:ïs-· -34J3M---ïïi'3T!i1-' -3ii'ii3ilili-' -1~:-ui'--iTIDI..,.--li-J~'ï~:wrll1-:-·1$-lïi'-_;,_i;r3$"-3Jiï.. Oill Oill 0 .--~ .. -.----------.----.- ... ---.-_.-_.-----, ---.
~-3iSJi-1i,i3J~i.i~~-·i]~~1~'4ilïia'ïÏ7 ::=~~:- ']i il! 3.1&31! lil 3B Il4 4i,-12i'sê{3i3ïi'-ïït_;~-I.sï-3Ï7~iir-l~ JIS 0 .--.---- ..-.- .-..------- ..,.._-.~~....:.:-:-:--i~"-~~~-=--~:!~. i~:f-~~.:-~{;:-·-~:,·1~i-i: i.:-· ;i; ~: :::-i:---~i:·~:r-·-:}-~:~·-~f::--~-:·:.~i.~- ~-f~r_;~,-.o:.~·-'---~---,---=-=~===·~=~~=~~=.=-..-·..-;.------'
~!-~-lr2r·-]1l ;--3i--·ï))'-·2~'-·-·3$~ï)),--ïffi··-ïll:l.lrï.~~-ïJ{--2Jr--217'·--;IS:-117~1D7;--iJ7,.-··I.;r·illC'-2)):-li-·i.\or---1.4.21~'"··-2i}.l)r;·-·-ïi[ll·--li--0~-oi-···o:r-2m--·O.~8--I$·~--·-.o,--" .---...-...- ...... ..,.- .. - ...-. ---- ..:--_ ..~_. '''-'-,
\ïs--' -ji-ïDi ..l$-ii)Bï~i.ii-i.iiIJr-i.ii-· -li-t~5~:-5i-5iJ3$ï~-ÏÏJ~--3rJID~~-lïï--m-ïi1:-JID--W:-3i2£14ffi~-J~--ï~-3~:--ïï11re--iii--~0-----· ..·-----·-·_-----,.--.---
:';--2~rn-' -3~~~1iï3;ïw-Uï1;-1;--m 251 2lI 2& lil) 217' Ist lil-' -I.iiïil 2lI 210 us ~iï~ig'-m1T'I]3 11i-;-~-OJgOiw____oiln-oi,--i51--0-----·--· ~
i~-3J6iir-~____;;_JM~-2Q~3$3Ji-m,--i$ï.riITi:~-219Jj61E1~S~l$-l~ii;-m--iMm3il1ï4~~--ïïl-lil 1$ 3M III m IJlw-ii--o----- ....-·---
M--11l-lfJI1JS-4.11 ~lJs·l.lr411 --lJS--ï~~-3JJ'lO'·_·I.~~lSl--ïZl:--1D!~-2li,--lJj-l]--2J9'-·2))·s.IS~-ii'----1IS:·--iT-2-1jr1l'J-sï."l-s"ïl--,JJ---l14 --'IM-'-lM-'ïr"31!'--'l})-'-2{'-lii--lll'-I£2 --'2~-" '0"'"
-ió2-'-----i_JS.c__'-'-w_-~-·-·1-1-I.'-3_!r_·ïJ2'--_-.--i:i-----4--..1_1._·-----·-_Ui~._S-·_·-1-11-'·--ï_{-.-i~"'--w'-l.ir-lffi'·--1l12:-.~·--'Hi--llS'-I~r2m"-lr-lll' --li--'31I1'---lr--I[l~ --114'--ï.IS·3.IS"·4))-I!{"-I$~-ill·--i.i(-·1))ï-)-)'--3.~·-·"0"12l{))-'2.19-0·..··..... -... -..-----.. --- ,---'.-. .._~..-_._•. ill --~--a--~--.211--•-_-ill---1-1-i-i-ll_.W_-_u_s -m--~m --m--m--_....._._-_._-m-_._~.._-~-'-. _._---_ ..-.-----m-~-,....--.~-_._----.._---'----_._---_._-_ ... ,,_- ..._._._-.,._-----_._----------m 31l m 2lI W m u-s-.m --m-'-_-'-i.ll -~--'---'~--.--i-ll_.-2lI ~v.s --~. -_2. lI 11i ill ~ ill ill2lI 11i ..•.-. -~-_g.._-~--•g ~ ~.~•---~-_g m ~ ~ ~ go. ..m m w..-. -_m._--,_.__~.~--~-..~-_._~-.--_g...-:----_._-.-----:---_.'"~ ill ill 3IJ m ~ ill lli ~ ~ ~ ~ ~ ill ill ill ~ ill ill ~ O, •
~•: ..._!~Ï~__!m;:~=!~ 3i=iïmi J~~._j~_}]=!~,-_2~ __ir~~.-mj~~-~J~~ ...-!mji~--~J19=_'~jjJiJ~,~.__•i~_J~:-_jï_j~~_..2mi;:=J!o_j2 2f.-=TsC:_~~jji._-2aï4..-.i-o~=-O.ImS...J~$_ ._i(_oÏ8 __jiï_j]':::=2~_j9==~ïl=-.Ji4=-.ï=ool(-~Oi=:_T .__~=~31l JM ill _ ill ~ ill 2lI ill ~ 2lI ill Hi ill ~ ~ ill 11i g ill g ill U 2lI ~ ~ ~ ~ • ~ m 0
MI- -·-1.?7~lbI--5(ó·-ill-lJ7---UI--4$'-·I.IS--·~-i~ll:--3.i-l3l- 5.15"----,41b51'--il"-' iS!--"i15~'bI:--ï!1'-"ï~$- &.13--;.11---ïm-sro---4.II-"):0'-11I-S.6-31-1'I-'1I'-'3JI'-JJo'---3-141i1: -- i~-lKl·'·-l.IS'-"4{-2E1-t-il' - ill" 11l 3fJ.1. lD!-'·l.IS---ii1'0--- ....--.-
ii-' -m2si--ïi3i"1J7~ffi-' -l~-2:r3i1-1JJ-'ijsl~-ïl1-' -2~___;s;..~2-~-m-2-lI-·ï]-ïil---I~-1]-2i1t:iO-2-Q-I$..--ïMl;\ïi~2lIQis_o.iOi,--lsi-~liO·lslioi--ïJl·i~-lffi··--~r--Q~-· O~ O~ 3.16 0
_:_r_o__;_i_-_-3iï-ï]--.3.__~-·J._]_2m.. -3.16JjJ-ï~~ii-2~~iT..ï_;.1~-3i1li. •__ ._._. -r-1:ï.9_#'""ïi.1m~.l,.JJ--UO·-ïli-.i]m~iis-.•i_..i.-2.t1iim....__~~ï_ï3ï.j_S_ 1.J..J--oï3-·0.. _]_-ïj9..·__-2~.-_-018.. "1$_·0..5_~J--;...-·o]-._-.i_i_j-2~_ï7·..--ll)r.-_._-.-i~·--Q~.._-._oi-.·.•o__.i-·... .._._ ... ,.. _~ ..__ .._.-31__6-6..$---_0
~ï~~=~~~_~:-~::~~~_~::_~:~:V::i~ï :;~-::.~~ -~:]~ï_~]i6_:~--:;~i::~: :~~~~j~=--.~i:~:~·: :·;l~ij ~:J~=~~_~§:_J0ii]~~C:~ .................... _.__ ..!·m~~\tI \-1-1_---_\t! \Il..-_ ...._ .._._._._._,Y._S.-l ._ \'SJ VOl V01I V011 VOlJ...
.'00 s$l oi 1
__;_-._..__l __ . ..._--+ __ -+__-+ : ;__. : ~ _ , :
._. ----'--------'--+--
: i i .
.__..-f-. r-'' .....:.......--!'--._. __ _._.- _ . .:-.._._~._._._.__ ._._._.- -.-_._._ _._ _ _ _ ..•._._ _._ __ ..•.. _ .._ _ .._ . I i .
ltl .---~.--__.__---i--;----___._---
m ~---_._-----j------+--+--!-------~
\'Ijl
vm S$I 6D1! 811 fil
WJ3 4JSI 4;si 1St Bl J11 4.19 JJ1 49l
I'lli S13' 4wl I_~. 601 lIJ 5.11 H9 l$
'Ir}, 4JS] 4JSI 1.$ 6$ ll1. 6El 6Jl J$ 513 4JS.
'Ir}, ~ ~ ~ ill ~ m ~ a m m
I,,!!_ : _~E! 'I!\ __ ~ __ ~~ __ ___4__l9; __ ~ __ ~~ _ _!!6 4!~ ~~_i_~~S.~__ 4~__ lii[
~m -•t~~•~~.L• :~:m ~ ~ : "ml ~~ •"l 'm. "l " " ,,! ~:_:___;_ ~ ----,----iill ~ ~ ~ m ill W ~ ~ 0
\1!1 4mi 4;si ll11 ll1 1l1: 4j) tID 21)1 4.10 lj, m t49 4Bl 4mi 1.9]i l16: ll1i 2j)' tIS 2j)' m l!1 m 0
00 ~ ~ ~ ~ ~ m ~ ~ ill lli ~ ~ m ~ ~ ~ ~ ~ ill ~ ~ m ~ m
~ :~I:~!: :: :;: ;~ :~ :~ :~ :: :~':: :: ::j :~I:::j ::1 ::' :~ :~: :: :~ :~ : ::: U:i 0 " L___ ___~=~:_;___ --+,-~----'
vn :~' :~:i:::: :~:;:::~:~:: :~i:~:: :~:~:~i::' ::: :~:: ::;:~:: :: :: :: ~. :!' :~, :~: 4& 0' ,_.__ ',__; '--+-__ ~---
2lJI 4Jl, 4;S, 25S: J16 t~ 259 lJl Jj)' 2]4' m lll: 2lli 2lI1 l16i 2i1i lj)' 2j) 21)1: lill 1.9] lB 115 Jffi !lS 2lI 312' I' , , .
~ ~ ~ ill m lli 2lI m Jj)' ill ill m ~ ~ ~ ~ ~ ~ ~ ~ 1.9] 2lI ~
~ ~ ~ 2lI ill lli • 2lI * ill ~ m ~ ~ ~ ~ ~ m ~ ill a 2lI 2lI
~ ~ ~ ill ~ ill m ~ ~ • • ~ ~ ~ ~ ~ m m m a
m! 4JO! 11l HS: ss lSl H5 1.9] l16.._El: l$ II l:E~! l:2.~~: :ll: 2,6i ~41: _J_R IIJI ._I!_ ._. _:: ~::: ::::~~_:~::~:~~!:: ::; lj! ll, 11J; _ •.
.YSl ; -~~r}![=~{~~~=~~-:ï~I~~~:~::~~~~:~~L~{~I4=SJ ~3$:' 4.10 1.11 4.10 3112l! 2m: Jl I~:~~1~1~32j)~~l161~I~ f-UJ4~~141:. -1115.15259!' l!!li 1$: 1$\ 059 0.19' HU' 114 OJl lill Ol Jj)
\'S3 ~ ~ m ill • • w mo. mOg ~ ~ ~ ~ m m ill ~ HS 0 ~ g • • g • ~ ~ ~ ~ ill • ill 111 ~ g ~ !
JOl
VOl l~1 3$) ll1i :~ :. :~---~~ :~ :~ :~ -~ ::--:~ :~' ::: ::i :~:~"---]~-_i+--1}-i:-:-~:~,:~ :~ ::i ~~:----}~~, . --~; ~~ ::! :~ ~ :j :~{.~_i~+--t--+
\'Ol 2lI ill 111 0 ill 25S ill 2lI 111 m • ~ • ~ ill ill ~ ill m 1.9] ~ 1.9] 25S ~ ill ~ ~ ~ D m li W D ill a m ~ ~ ~ ~
a ~ ~ g m ill ~ a ill 1.9] m 2lI 111 • ~ ~ ~ ~ ill ill ~ ill m 111 ~ 1.9] 25S ~ m ~ ~ ~ g m li wa. a m ~ ~ ~ ~
:0 :~l:l:~:.: :~::~:~:: :~:: ~~:~: ::! :: ~~:~l:'~:: :~:,~:: :: :~~:~:::~::~::1 :: ::' : :::: :: :~' :~ :: :: :: :~~+:~;-,;-;;;i ijj;: 0
VOlI 4j)i 6;si HJ lID 4jJ: 1.11 ~ 6IJ 1.11 5ID ll1 6$ SIJ 6~; 613: 5$' lID' 113 6Jl 3Jl: 413 811 1.11 HS ~ ill 4JO 513 ~i 513! 5ID! 4Jij III 4~ 511, lSI Hll 313 113 iil 5.11 513 l~; .
ï~2-'--ïïr--ïir- ï~'--l~'---i£-ïl6---ï.ïa----l~·---ï;--ll3-1iï )iJ- --ïïi---i1i;--iDii" ïïii---ii'-··-ïïi- 259 i1i~--lii--- ïi--ïiJ "-15ï---ïi4--ïis---w--ïi'---,~1---i$;---13Jl---~i9-oi--lm-- -ïi~-o~--lIS···- ·O.19···--.ïÏ01"]·- ·····iBiH-·ij···' ··1.15.
_-!-O-_1-~c_._-~-_ ]~.-L--+-- ...•mm_____2_lI_ _mIn --.---U--O- 2j) -- 316- .....2j).'-- 214 -3l - mm __m;_ -, m_ .-----~- - _j~~--I~[Ji- -!! ~_!.__~~_::~i::~~~_--J]Ijï---iBil_: _j_L_2~L~:o-j~ C}j-j~l:=_~!i.=~16-- O,~_:_Iï~_~_~.m:.:I~.__J;J) ._I~~ _
2.93% (ranging from 0.01 to 5.88). Within the subgroups, the maize inbred lines
from subgroup A had an average distance between them of 2.83% (ranging from
0.1 to 5.56), in subgroup B the average distance between the lines were 3.79%
(ranging from 2.02 to 5.56), in subgroup C the average distance between the lines
were 3.47% (ranging from 2.45 to 4.49) and in subgroup D the average distance
between the lines were l.62% (ranging from 0.01 to 3.22). Looking at the minor
groups the average distances were all lower that in their subgroup. Therefore, the
trend of the clustering pattern was more variation in the major clusters with less
variation between lines in subgroups and even less variation within the minor
groups. It was also further illustrated that the local and foreign inbred lines were
not grouped separately.
Grouping of inbred lines into dusters of genetically related lines
When the different dendograms and consensus trees were compared with each
other (Fig. 4.2 to 4.7), the lines were grouped into two major groups (I and II).
Group I consisted of two subgroups (A and B) and group II consisted of two
subgroups (C and D). Subgroup A was divided in three minor groups: 1 (W05
and W07), 2 (WOII, WOl, W02 and WK2) and 9 (W04, WIl, YMI and W08),
subgroup B was divided in five minor groups: 7 (YF3, YL2, W03 and WMl),
4(YOI, WFI and WOlO), 5 (YR6, YL5 and YM3), 6 (WP3, YM2 and YS3) and 3
(WRI, WF2, YM4, WP! and WM2), subgroup C was divided in two minor
groups: 8 (YOll, WKI and YF2) and la (WR2, WP2 and Y02), and subgroup D
was divided in four minor groups II (YF1, YOI2, YL4 and YR5), 12 (YOl3 and
W09), 13 (YLI, YI2 and Y09) and 14 (Y04, Y06, YOIO, YS2 and W06). Wl2
and YF4 were not contained in any specific cluster.
If the Jukes Cantor distance tree (Fig. 4.2) and - consensus tree are compared (Fig.
4.4) with each other, they are very similar. Group I in the dendogram is divided in
subgroup A, C and B, with group II only consisting of subgroup D. In the
consensus tree group I is divided in subgroup A and B, and group II is divided in
subgroup C and D. In subgroup A of the dendogram the minor groups are grouped
starting with minor group two, then one and then nine. In the consensus tree
subgroup A is divided in minor group one, followed by two, followed by nine. In
71
subgroup B the minor groups in the dendogram grouped as follows: minor group
three, followed by four and five, followed by seven and six. In the consensus tree,
subgroup B grouped as follows: minor groups four and seven, followed by three,
followed by five and six. Subgroup C and D had the same minor groups in both
trees. Looking at the minor groups, WO Il was not included in minor group two
of the dendogram, but it was included in the consensus tree. In minor group seven
of the distance tree, W03 clustered with WMl and W06, but last mentioned was
not included in minor group seven in the consensus tree. In the distance tree Y09
was included in minor group 12, but in the consensus tree it was included in minor
group 13. It was also clear that the dendogram had more resolution at the bottom
of the tree, than the consensus tree. In general, however, the two trees were
closely correlated.
If the Kimura distance tree (Fig. 4.3) and - consensus tree (Fig. 4.5) are compared
with each other, they are also very similar to each other. Group I and II consisted
of the same subgroups in both trees. In subgroup A of the dendogram the minor
groups are grouped starting with minor group two, then one and then nine. In the
consensus tree subgroup A is divided in minor group one, followed by two,
followed by nine. In subgroup B the minor groups in the dendogram grouped as
follows: minor group three, followed by four and five, followed by seven and six.
In the consensus tree, subgroup B grouped as follows: minor groups six and five,
followed by seven, followed by three and four. Subgroup C and D had the same
minor groups in both trees. In minor group two of the consensus tree the line
WO Il was included in the cluster, but not in the dendogram. In minor group
seven in the dendogram W03 clustered with WMl and W06, but they were not
included in minor group seven in the consensus tree. In the distance tree Y09 was
included in minor group 12, but in the consensus tree it was included in minor
group 13. In these trees, it was also clear that the dendogram had more resolution
at the bottom of the tree, than the consensus tree. In general however, the two
trees were closely correlated.
72
+YS2
+-47
+Y010
-
48-Y04 -
+W01
2 +---------1
+---2 +W02
+-27 +-----------WK2
].+-----W05 A
+--6
+------W07
+-37
+-----W11
+--8
+-14 +----------YM1
! 9
+-38 +-21 +-------W04
+---W08 -
+-------YF2
!10!
+-28 +--------WR2
+-12
+-22 +-------WP2
+--Y02
c
+------YF4
8
+-25 +------WK1
+--9
+-16 +------------Y011
+-31 +------------WI2
+-----------=m-z--
+--5
!3 +--------WRl
+-41 +-23
+-------YM4
+--3
+-32 +-19 +-------------WPl
+---------WM2
B
+----------WFl
4 +--4
+-17 +-------W010
+-----Y01
+-24
+-36 +--------YL5
l5 +--7
+-10 +--------YR6
+--------YM3
+--------yF3
+-11
+-18 +-----YL2
+-45 +-39 ! 7 +-----------WM1
+-26
+-------W03
+-15
+-20 +-------W05
+-----------W011
+-------WP3
!6 +-13
+-29 +---YM2
+-----YS3 -- -
11 +----YR5
!14! +-40
+-46 +YU
+-34
+Y012
+-43 +-30 D II
+YFl
+------W09
!12+-33
+-44 +-42 +Y013
+YOg
!13+YL1
+-35
+YI2 - -
+Y06
Fig. 4.2 Dendogram generated by Neighbour joining analysis from genetic distance matrix determined by Jukes and Cantor.
The values on the dendogram represent the genetic distances among different inbred lines.
73
+YS2 -
~8-Y010
+W01
+--------1
+--2 +W02
A
+-26 +----------WK2
!1 +-----WOS
+--5
+------W07
+-37
+-----WIl
+--8
+-14 +---------YM1
! 9
+-21 +------W04
+---W08
+-------YF4
+----------WF2
+--6
+-38 +-------WR1
+-35 +-23
+------YM4
+--4
+-19 +------------WP1
+---------WM2
+-30
+---------WFl
4 +--3
+-17 +------W010
+-----Y01
+-36 +-2~
+-------YL5
!5 +--7
+-12 +-------YR6 B
+-39
+--------YM3
+-------YF3
+-13
+-18 +----YL2
!7 t----------WMl
+-25
+------W03
+-15
+-20 +-------W05
+----------W011
+-41 +-------WP3
!6 +-10
+-28 +---YM2 -
+-----YS3 -
+------WK1
B +--9
+-16 +-----------Y011
+-45 +-22 +----------WI2
+---C---WR2 c
+-29 +-11
!10! +-------WP2
+-31
+-------YF2
+---Y02
II
+Y04 -
!14! 11+----YR5
+-47 +-40
+YL4
+-32
+Y012
+-43 +-27
+YFl
+------W09
!12+-33
+-44 +-42 +YOl3 D
+Y09
+-46 !13+YL1
+-34
+YI2
+Y06 - -
Fig. 4.3 Dendogram generated by Neighbour joining analysis from genetic distance matrix determined by Kimura.
74
+----W05
+----------535.0
+----W07
+-56.0 +----W02
+1000.0
+501.0 +----WOl
+136.0 +---------WK2 A
+--2.0 +--------------WOll
+----WIl
+313.0
+16!. 0 +----YMl
9
+------83.0 +---------W04
+--------------W08
4 +---------YOl
+----------------95.0
+----WFl
+314.0
+----W010
+----YL2
+184.0
+-89.0 +----YF3
7
+-----------11.0 +---------W03 B
+--------------WMl
+--1. 0
+----WPl
+289.0
+-57.0 +----YM4
3
+------------5.0 +---------WM2
+----WF2
+-----250.0
+----WRl
+----YL5
5 +237.0
+104.0 +----YR6
+--5.0 +---------YM3
+------------2.0
+----WP3
6 +343.0
+-50.0 +----YM2
+---------YS3
+----YF4
+---------------------62.0
+----WI2
8 +---------YF2
+--5.0 +-44.0
+----YOll
+258.0
+-----------------2.0 +----WKl c
l.0 +---------Y02
+-86.0
+----WP2
+270.0
+----WR2
+101.0
+---------YL4
l.l. +536.0
+----YOI2
+--------------------117.0 +571.0
+----YFl
l.4
+246.0 +--------------YR5
l.2 +----Y013 o II
+-----------------------------------147.0
+----W09
+265.0 l.3 +---------Y09
+-----------------------------------242.0
+----YI2
+582.0
+353.0 +----YLl
+ - ------- - -------- - - - - - -- ----- - - --------- - Y04
+682.0
+ - - - ---- - - -- --- - - - - ------- - - - - - - - - ----- --------- Y06
+999.0
+ - - - ------ - - - - -------- ---- - --- - - - - - - - -- YO10
+ ----------------YS2
+ ---------------------W06
Fig. 4.4Consensus tree generated by Neighbour joining analysis from genetic distance analysis determined by Jukes Cantor.
75
1 +----W05
+----------546.0
+----W07
+-62.0 +----W02
+1000.0
+509.0 +----WOl
2
+143.0 +---------WK2 A
+--2.0
+--------------WOll
+---------W04
+166.0
9 +----WIl
+------87.0 +315.0
+--_-YMl
+--------------W08
+----YM2
6 +358.0
+-58.0 +----WP3
+---------Y53
+------------3.0
+----YR6
5 +241. 0
+105.0 +----YL5 B
+---------YM3
+--------------WMl
+--1.0 7
+-----------12.0 +----YF3
+186.0
+-88.0 +----YL2
+---------W03
+----YM4
+287.0
+-58.0 +----WPl
+------------8.0 +---------WM2
+----WF2
+-----252.0
+-10.0 +----WRl
+----WOI0
+318.0
+----------------97.0 +----WFl
+---------YOl
+----W12
+---------------------67.0
+----YF4
8 +---------YF2
+-43.0
+-15.0 +----WKl
+259.0
+-----------------2.0 +----Y011 c
+----WR2
10+266.0
+-79.0 +----WP2
+---------Y02
+157.0
+----Y012
+594.0
+583.0 +----YFl
11
+--------------------147.0 +---------YL4
+293.0 +-------------- YR5
12 +----Y013
+ 157.0
+----W09 D II
14
+264.0 + 13
+---------Y09
251.0
+----YLl
+595.0
+392.0 +----Y12
+ - - - ---- - - - - - ----- - - - - Y04
+658.0 + --- - - ---- ---- - - ---- Y06
+999.0 + - ----- - - - - - ------- YO10
+ ----------------Y52
+ \'06
Fig. 4.5 Consensus tree generated by Neighbour joining analysis from genetic distance analysis determined by Kimura.
76
1 +----W05
+ 1.0
+----W07
+--1. 0 + W011
+--2.0 +----W01
+--2.0
+--2.0 +----W02 A
+--1. 0
+---------WK2
+---------W04
+--2.0
9 +----WIl
+-------2.0 +--2.0
+----YM1
+--------------W08
+----YF3
+--2.0
+--2.0 +----YL2
7
+------------2.0 +---------W03
+--------------WM1
+ +---------Y012.0
+----WFl
+--2.0
+----W010 B
+--1. 0
+----YR6
5 +--2.0
+--2.0 +----YL5
+---------YM3
+------------2.0
+----WP3
6 +--2.0
+--2.0 +----YM2
+---------YS3
+----vlR1
+-------2.0
3 +----WF2
+--1. 0 +------------2.0
+----YM4
+--2.0
+--2.0 +----WP1
+---------WM2
+ +----WI22.0
+----YF4
+----YOll
8 +--2.0
+--1. 0 +--2.0 +----WK1
+---------YF2
+ 2.0
+----WR2 C
10+--2.0
+--2.0 +----WP2
+---------Y02
+--1. 0
+----YFl
+--2.0
+--2.0 +----Y012
+ 112.0 +---------YL4
+ YR5
+--1. 0
+ 12 +----Y0132.0
+----W09 D II
+----YL1
+--1. 0
13 +--2.0
+ 2.0 +----YI2
14
+---------Y09
+--1. 0
+ - - -- - ---------- Y04
+--1. 0 + ------Y06
+--2.0 + YO10
+ YS2
+ W06
Fig.4.6 Consensus tree generated by Neighbour joining analysis from the combination of genetic distance matrixes
determined by Jukes and Cantor, and Kimura,
77
+----wp2
+--5.0
l.0 +----WR2
+--2.0
+----'{04
+--4.0
+------------3.0 +----Y02
+----Y010
l.4 +--5.0
+--4.0 +----Y06
+---------Y52
+----WI2
+--3.0
+----'{F4
+------------1.0
+----WM1
+--2.0
+----'{53
+----W01
+--5.0
+--5.0 +----W02
2
+--4.0 +---------WK2
+--------------W011
+--1. 0
+--3.0 +-------------- '{R5
11!
+--3.0 +----'{Fl
+--5.0
+--5.0 +----'{012
+---------YL4
+----WFl
4 +--5.0
+--5.0 +----W010
+-------2.0 +---------'{01
6 +----WP3
+-------5.0
+----YM2
+----YM1
+--5.0
+--5.0 +----WIl
9
+--4.0 +---------W04
+--1.0---2.0 +--------------W08
! 1 +----W07
+------------5.0
+----W05
l.3 +---------'{09
+--5.0
+--5.0 +----Y12
+--5.0
+-------1.0 +----YL1
l.2 +----W09
+-------5.0
+----'{013
+----'{F3
+--5.0
7 +----YL2
+------------1.0
+----W03
+--4. 0
+----W06
+----WR1
3 +--5.0
+------------2.0 +----WF2
+---------WM2
+----YL5
5 +--5.0
+------------5.0 +----YR6
+---------YM3
+----WP1
+-----------------4.0
+----YM4
8 +----WK1
+---------------------------5.0
+----YOl1
+...........•.•••••. ····················YF2
Fig. 4.7 Consensus tree generated by 1000 bootstrap replicates using DNA parsimony
7R
When comparing the Jukes Cantor and Kimura (Fig. 4.2 and 4.4) generated trees it
was evident that these trees were similar. Group I in the Jukes Cantor dendogram,
was divided in subgroup A, C and B, with group II only consisting of subgroup D.
In the Kimura dendogram, group I was divided in subgroup A and B, and group II
was divided in subgroup C and D. The subgroups for both trees were the same.
Line YF4 was grouped in minor group eight in the Jukes Cantor tree, but in the
Kimura tree it was grouped separately.
If the two consensus trees (Fig. 4.4 and 4.5) are compared, one using the Jukes
Cantor algorithm and the other one using the Kimura algorithm to determine the
genetic distances, the trees look very similar. The only difference between the two
trees is that in subgroup B, the order in which the minor groups clustered, differed.
In the Jukes Cantor tree the groups clustered in the following order: four, seven,
three, five and six; where as in the Kimura tree they clustered in the following
order: six, five, seven, three and four. Although the order of the groups were
different, all the groups had the same inbred lines clustered together and the rest of
the trees were the same.
If the two consensus trees (Fig. 4.3 and 4.5) were compared individuall y with the
combined consensus tree (Fig. 4.6), the only differences that occur once again, is
the order in which the minor groups cluster together in subgroup B. In the
combined consensus tree the minor groups clustered in the following order: seven,
four, five, six and three.
If the combined consensus tree and the DNA parsimony consensus tree were
compared, the groups and subgroups were clustered differently. Looking at the
minor groups of the two trees, there were only a few differences. The parsimony
tree had minor group three and five combined in one cluster, but in the combined
tree they were clustered separately. Minor group six didn't include YS3 in the
parsimony tree, but it was included in the combined tree. Minor group seven didn't
include WMl in the parsimony tree (like in the combined tree), but it included
W06 additionally. In the parsimony tree minor group ten had an additional line
Y04. It was also clear that the resolution at the end of the parsimony tree was
better than that of the combined consensus tree. The lines that formed part of
79
minor group 14 in the combined consensus tree (that had poor resolution), seemed
to be the lines that grouped differently in the parsimony tree.
Discussion and conclusion
The results of this study indicated that, the lines were distributed into two major
groups, four subgroups and 14 minor groups. From the distribution of lines in the
dendogram the coherence of the AFLP technique and the analytical methods can
clearly be seen, indicating its potential as a powerful tool for line identification.
Although there were different groupings, the genetic distances were not that high,
in other words the inbred lines are related to a large degree.
There were no definite groupings with hierarchical structure according to the
genetic distances, the reason was that there was no hierarchical grouping in the
group of maize inbred lines. The group of inbred lines were not developed in a
natural selection system, and the relationship between the inbred lines were close.
The clusters indicated the absence of any relation between geographic diversity
and genetic diversity. This therefore, revealed a relatively low degree of genetic
variability within the species. A possible reason can be that the inbred lines have
been used aU over the world to improve existing hybrids. This interbreeding has
caused the variation in the group of material, in terms of place of origin, to be
much less.
Initial experiments with AFLP fingerprinting of a number of plant DNA's
indicated that AFLP primers with at least three selective nucleotides at both the
Pst! and MseI primer were required to generate useful band patterns. Because
primers with three selective bases tolerate a low level of mismatch amplification a
two-step amplification strategy was developed for AFLP fingerprinting of
complex DNAs. With the pre-amplification strategy the background smears in the
fingerprint patterns were reduced, and fin&erprints with particular primer
combinations lacked one or more bands compared with fingerprints generated
without preamplification. This is best explainded assuming that the direct
preamplification with AFLP primers having three selective nucleotides resulted in
a low level of mismatch amplification products, which caused the background
80
smears and gave discrete amplified fragments corresponding to repeated restriction
fragments. An aditional advantage of the two-step amplification strategy is that it
provided a virtually unlimited amount of template DNA for AFLP reactions (Vos
etai., 1995).
Only the reproducible amplification products that could be unambiguously scored
in all the lines were included in the study. Bands that did not qualify were either
too faint to score accurately or not reproducible. Some of the minor (low
intensity) amplification products were not reproducible in all cases and were
consequently discarded. Major (high intensity) products were generally
reproducible across a range of near-optimum amplification conditions (Micheli et
al., 1994).
Lines WI2 and YF4 were very inconsistent, a possible reason for this could be
because the lines are very inbred, and they had no predominant characters that
could group them with a specific cluster, so they grouped together. This may also
be the reason why their position in the trees was not consistent. A more consistent
result may have been obtained if white and yellow types were treated separately.
The number of DNA markers required to distinguish any group of lines depends
on the genetic variability among the lines and the size of the group (Myburg et aI.,
1996). AFLP analysis was found to be highly effective in distinguishing
genotypes from geographically separate areas (Pakniyat et al., 1997). Good
coverage has been demonstrated for AFLP's from PstI-MseI DNA digests in
different lines (Powell et al., 1997). Barrett and Kidwell (1998) have used 16
AFLP primers to detect genetic relationships among 54 wheat lines. The AFLP
primer pairs detected 229 polymorphic bands. Bands excluded were those that in
the amplification profiles appeared or disappeared when repeating the reaction
across time and materials. A high level of polymorphism was detected among the
AFLP markers and none of them were unique to distinguish the lines. Clustering
of different lines have been done using AFLP's, for example wheat (Bohn et al.,
1999; Barrett et aI., 1998) and barley (Pakniyat et al., 1997).
81
AFLP analysis has accelerated the use of DNA-based fingerprinting data for the
identification and genetic analysis of lines, varieties and breeding lines (Vos et al.,
1995). This is certainly, in part, due to the fact that the technique does not require
any previously obtained sequence data. This requirement has thus far precluded
the use of most genome analysis techniques in all but the major crop species. We
found that optimised reaction conditions were an absolute requirement for the
generation of reproducible and informative AFLP profiles in maize. DNA samples
were found to be stable and reproducible. These results indicated that the AFLP
analysis could be used for the identification of closely related maize lines.
The results from this study will contribute to a large extent to the decisions that the
breeder will make in the breeding programme. Since the inbred lines are grouped
into clusters and the genetic distances between the lines are known, it is possible to
make crosses between lines that will yield higher combining ability. The reason is
that the shorter the genetic distance between two lines, the closer related the lines
are. The closer related two lines are, that are crossed with each other, the lower
the combining ability. The breeder will thus cross lines that have a larger genetic
distance between them.
The implication of these results to the breeder and the breeding programme are
significant. Knowing how closely related lines are to each other, is a good
indication of how good the combining ability would be if two lines were to be
crossed. This in return could make the whole breeding programme more labour
effective, time effective, and cost effective, since less unnecessary crosses will be
made, hybrids can possibly be developed quicker and because of the breeding
programme being more structured in terms of quicker returns and possibly less
input costs.
In conclusion, the results of this study indicate that AFLP analysis can be used for
the identification and genetic analysis of maize lines. This study generated a
combination of polymorphic AFLP loci that was used to construct a line
fingerprinting profile. Genetic distance analysis of the maize inbred lines based on
107 AFLP loci indicated high genetic similarity among the groups and high
genetic variation within the groups of lines included in the study. This data allows
82
the breeder, when making crosses between individuals that have been tested, to
make crosses with higher combining ability, because of their genetic relatedness
being known. That in return makes a breeding programme much more effective
and saves a lot of time. This documentation will therefore allow further scientific
studies on crosses that are made in the maize programme. It will also possibly
contribute to more effective hybrid breeding.
83
CHAPTER 5
COMIP A.JRTISON OF JD)TI1FFlEIRJENM1fETHODS OF
GIR01IJIPlING MA.TIZIE TINB1RlElIJJ)LTINIESTINCJLUSTIEIRS.
Abstract
The study of genetic variation is a primary concern in population biology and is
fundamental in the improvement of agricultural plants (Melchinger, 1993).
Comparing results of different genetic diversity estimation methods may be indicative
of their utility as a tool for plant breeders. Sixty-eight maize inbred lines from the
Advanta Africa genebank were used to compare morphological characterisation,
restriction fragment length polymorphism (RFLP) and amplified fragment length
polymorphism (AFLP)-based genetic diversity estimates. The AFLP and RFLP
analysis detected a similar clustering pattern of genetic diversity among the maize
inbred lines, but it was difficult to compare the two methods since the available RFLP
data were limited. The morphological clustering was different to those of the DNA
based techniques. This emphasised the problems related to morphological
characterisation, and indicated that DNA methods would be more reliable. The DNA
based methods succeeded to identify groups that were genetically further apart, which
will yield better hybrid progeny. Looking at the comparison of these methods to
cluster the different inbred lines into clusters, the DNA techniques are definitely more
reliable than the morphological characterisation.
Obtaining accurate estimates of genetic diversity levels among germplasm sources
may increase efficiency of breeding efforts to improve crop species (Barret et al.,
1998). Assignment of inbreds to heterotic groups and selection of donor lines for
improving parents of an elite single cross are major decisions facing any maize
breeder (Dudley et al., 1991). Various methods, including morphological marker
(Van Beuningen and Busch, 1997) and DNA marker (Karp et al., 1996) analysis, have
been used to quantify genetic diversity and relationships among genotypes.
84
Most, if not all, currently important heterotic groups in maize haye been established
empirically by relating the heterosis observed in crosses with the origin of their
parents. Further classification criteria were based on morphological traits (Messmer
et al., 1992). Although morphology has proved useful for classifying maize races and
populations (Goodman and Brown, 1988), it may not be appropriate for elite breeding
germplasm. Furthermore, morphological characteristics can be affected by
environmental conditions (Goodman and Patemiani, 1969). Smith and Smith (1989)
concluded that morphological data may be a starting point in determining
distinctiveness, but that RFLP (or AFLP) data offer better resolution of relatedness of
maize inbreds (Gerdes and Tracy, 1994).
Restriction fragment length polymorphisms (RFLP) detect DNA polymorphisms
through restriction endonuclease digestions followed by visualisation via DNA blot
hybridisation (Mazur and Tingey, 1995). RFLP's offer several advantages as
molecular markers. RFLP's are unaffected by environment, and show eo-dominant
inheritance. However, RFLP's are more polymorphic, more numerous and are better
distinguished throughout the genome. Accordingly, RFLP's have been suggested as
superior markers to assess diversity and relationships in maize breeding germplasm
(Murray et al., 1988; Walton and Helentjaris, 1987).
Amplified fragment length polymorphism (AFLP) is a recently developed DNA
marker analysis system based on a novel, powerful combination of polymerase chain
reaction (Pf'R) and restriction enzyme analysis (Vos et al., 1995). AFLP's assay the
presence/absence of restriction enzyme sites in combination with sequence
polymorphisms adjacent to these sites (Mazur and Tingey, 1995). AFLP's are highly
efficient compared with other DNA marker systems, they are reproducible, and
exhibit intraspecific homology (Rouppe van der Voort et al., 1997; Powell et al.,
1996; Tohmeetal., 1996).
Since this technology was recently developed, the relationship between AFLP-based,
RFLP-based analysis and morphological characterisation has not been determined for
maize. The objective of this study was to compare AFLP, RFLP and morphological
genetic diversity estimates generated for 68 maize inbred lines from Advanta Africa
Seeds.
85
Material and Methods
Plant Material
Sixty-eight maize inbred lines from the Advanta Africa genebank, originating from
five different countries, were used in this study. These lines were all analysed using
the morphological characterisation and AFLP analysis. However, only seven (YR5,
YOll, YLl, YL2, YL4, YM4 and YL5) of the 68 lines were used in the RFLP
analysis.
Agronomic traits
As described in chapter three.
Molecular marker assays
AFLP assays
As described in chapter four.
RFLP assays
GARST Laboratories (Advanta Africa's sister company) did the RFLP analysis on
these seven inbred lines prior to the AFLP analysis. They presented us with the
results of the research study. Advanta Africa Seeds in South Africa interpreted the
data.
Data analysis
Associations among the 50 lines were revealed by cluster analysis using Kimura and
Jukes and Cantor algorithms with Neighbour joining analysis. Neighbour joining
analysis was also applied to the RFLP and morphological data.
86
Results
Phenotypic analysis
The dendograms were given in chapter three.
Genetic distance analysis
AFL1P duster analysis
The consensus trees were given in chapter four.
If the AFLP dendograms and the morphological dendograms are compared with each
other, there were hardly any similarities between these two techniques. The only
similarities were as follows: W05 and W07 clustered together in both trees, but in
the morphological tree YSI was also included in the cluster. Cultivars WOll, W02
and WK2 also clustered together in both trees, but in the AFLP tree WO 1 was
included in the cluster. In the morphological tree WO I was replaced by WIl.
Cultivars W02 and WKI, and Y06 and YOIO also clustered together in both trees.
The rest of the groups clustered together very differently.
RFL1P duster analysis
If the RFLP dendograms (Fig. 5.1 and 5.2) and the AFLP dendograms are compared
with each other, the lines in the RFLP clusters were grouped similarly to that of the
AFLP dendograms. It was difficult to compare these two techniques with each other
since there were only seven lines included in the RFLP analysis in comparison with
the 50 lines included in the AFLP analysis. The clusters could not really be compared
with each other, since the lines included in the RFLP analysis were representatives
from different clusters. In the RFLP analysis both the UPGMA and Neighbour
joining analysis results were similar. It was significant that all the Lancaster lines
(YL5, YL4, YLI and YL2) grouped together. Only YL4 and YL2 exchanged places
in the two different trees. There were no correlations between the RFLP results and
the morphological characteristics.
87
+----YR5
+-99.0
+-89.0 +----Y011
+-51.0 +---------YM4
+-73.0 +--------------YL5
+----YL1
+-----------99.0
+----YL2
+------------------------YL4
Fig. 5.1 Dendogram generated by Neighbour joining cluster analysis ofpairwise
distance data for the different maize inbred lines.
+----YR5
+-94.0
+-82.0 +----Y011
+-76.0 +---------YM4
+100.0 +--------------YL5
+100.0 +-------------------YL4
+------------------------YL1
+-----------------------------YL2
Fig. 5.2 Dendogram generated by UPGMA cluster analysis of pairwise distance data
for the different maize inbred lines.
Discussion and conemsion
Comparison of different methods of grouping
Different cluster groups were formed using morphological, RFLP and AFLP analysis.
Clusters drawn using morphological characteristics were different to those drawn with
AFLP and RFLP data. Morphological traits had too much of an overlap to be able to
correlate morphology with the genotype.
88
Although genotype is reflected in the morphology it is unfortunately inconsistent,
because environment has an effect on how the morphology is expressed. From the
comparison of methods it was clear that morphological characterisation is subjective,
because the characters were measured visually. The variation within a certain
character was also fairly small, thus making it difficult to cluster the inbred lines
according to these parameters. The breeder assigned groupings based on non-
selective characteristics and then tried to scientifically correlate the groupings.
AFLP's and RFLP's are scientific methods and are not influenced by the
environment. They are reliable and difficult to manipulate.
The development of inbred lines by self-pollination and then evaluation for hybrid
performance is the basic procedure in most applied maize breeding programmes.
Most breeders develop inbred lines with a certain idiotype in mind, but many
correlation studies have shown that the values for agronomic traits of the parental
lines have minimal predictive value for hybrid performance, particularly grain yield
(RusselI, 1974).
Controversial results have been reported in many crops with the use of morphological
characterisation alone. Efisue (1993) has found that the combination of
morphological and isozyme techniques has offered a better approach than
morphological descriptors alone. Li.ibberstedt et al. (1998) concluded that RFLP
analysis may serve as a major source of information for separation of closely related
accessions, especially when integrated with phenotypic measures.
Some pairs of genotypes in the RFLP and AFLP consensus trees were observed
together in both analyses. This suggests that the results of both analyses do have
some correlation, but with the amount of RFLP data available it is difficult to really
draw any conclusion. The similarity obtained in both dendograms produced by
different analytical methods proves the clustering to be acceptable. This positive
relationship between these different methods shows that AFLP's and RFLP's are both
good techniques to cluster maize germplasm (Lubberstedt et al., 1998; Gerdes and
Tracy, 1994).
89
For the morphological analysis we used data collected in the 1998/1999 season on 25
characters to compile the dendograms. With the AFLP technique, data were collected
over a period of three months. One hundred and seven AFLP fragment loci were used
in our analysis to construct the dendogram. Using Neighbour joining analysis,
specific distances between lines were obtained. It was possible to prove that by using
the AFLP technique a large set of informative characters can be obtained in less time
than with a morphological analysis. By using the AFLP technique, measures of
genetic distance among the accessions were established, thus providing an identity for
each line.
A major obstacle to increased use of molecular techniques in systematics is cost.
Morphological data can be collected with minimal expenditures on supplies and
equipment. The greatest barrier for morphological systematises probably is the initial
set-up of the laboratory. Costs vary by specific discipline, but most molecular
laboratories require tens of hundreds of thousands of dollars to establish and maintain.
Because systematics is a relatively poorly funded subdiscipline of biology, these costs
can be prohibitive. However, the value and need for molecular data in systematics is
recognised, in spite of the expense (Hillis, 1987). Therefore, although morphological
characterisation is influenced by the environment and is time consuming in general,
among other disadvantages in relation to AFLP's and RFLP's, it can still be an
important and practical means of making progress in germplasm evaluation for
farmers and breeders (Liibberstedt ef al., 1998).
Application of these results in maize breeding programmes
In breeding programmes, parental lines should have a high degree of heterozygosity,
as well as optimum genetic distance (Varghese ef aI., 1997). Measurement of genetic
distance facilitates identification of diverse parents in order to maximise the
expression of heterosis (Smith ef aI., 1990). A desired and high magnitude of
heterosis is not always directly related to pronounced parental divergence
(Arunachalam ef aI., 1984) and intermediate divergent classes may also have a high
probability of producing heterotic hybrids (Thakur and Zarger, 1989). When this
procedure is applied for desired characters, it should result in better exploitation of
heterosis in the F 1 hybids. Further investigations between parents and hybrids at
90
molecular level in maize should reveal useful information on the degree of heterosis
and genetic distance between parents and its correlation with the performance of the
progeny.
The improvements in marker techniques are likely to accelerate the process of finding
markers associated with traits of interest. An important target of marker development
work has been to provide tools for more efficient and sophisticated breeding. The
close interaction between breeders and biotechnologists will be crucial to the effective
implementation of molecular markers. However, it is equally clear that molecular
markers offer a powerful set of tools for analysing the inheritance of a wide range of
traits (Langridge and Chalmers, 1998).
In conclusion, to show congruency between morphological and molecular data is very
difficult. The reason is that although molecular data are supposed to reflect the
overall genotype of an organism, and the morphological data are supposed to be an
expression of the genotype, the interaction of the environment on genotype is
subjective. Environment affects the expression of the genotype. Therefore,
morphology can be a valuable asset, but must not be seen as an independent entity to
reflect the relatedness between lines or cultivars.
91
CHAPTElR6
CONCLUSION
In conclusion, usmg AFLP fingerprinting for the identification and genetic
distance analysis of a collection of maize inbred lines, we found that there were no
significant differences between the groupings according to the genetic distances.
The reason was that there was no hierarchical grouping in the group of maize
inbred lines. The group of inbred lines was not developed in a natural selection
system, and the relationship between the inbred lines was close. The clusters
indicated the absence of any relation between geographic diversity and genetic
diversity. This therefore, revealed a relatively low degree of genetic variability
within the species. A possible reason can be that the inbred lines have been used
all over the world to improve existing hybrids. This interbreeding has caused the
variation in the group of material, in terms of place of origin, to be much less.
The results from this study will however, contribute to a large extent to the
decisions that the breeder will make in the breeding programme. Since the inbred
lines are grouped into clusters and the genetic distances between the lines are
known, it is possible to make crosses between lines that will yield higher
combining ability. The reason is that the shorter the genetic distance between two
lines, the closer related the lines are. The closer related two lines are, that are
crossed with each other, the lower the combining ability. The breeder will thus
cross lines that have a bigger genetic distance between them.
The implication of these results to the breeder and the breeding programme are
significant. Knowing how closely related cultivars are to each other, is a good
indication of how good the combining ability would be if two lines were to be
crossed. The closer related the lines are, the poorer the combining ability. The
results of this study will thus enable breeders to make better decisions about which
lines to cross with each other. This in return could make the whole breeding
92
programme more labour effective, time effective, and cost effective. The results
of this study thus indicate that AFLP analysis can be used for the identification
and genetic analysis of maize inbred lines.
When looking at the morphological characterisation, the statistical analysis will
allow the breeder to make objective choices for the selection of parental material
to produce new hybrids. The dendograms in this study showed significant
grouping, but more objective methods are needed to confirm the results since
environment does influence the expression of the phenotype. These results will
however, contribute to the breeding programme, since it will make the possibility
of higher combining ability in crosses, as well as improved populations, higher.
Therefore, to show congruency between morphological and molecular data is very
difficult. The reason being that although molecular data are supposed to reflect the
overall genotype of an organism, and the morphological data are supposed to be an
expression of the genotype, the interaction of the environment with genotype is
significant, which affects the expression of the genotype. Therefore, morphology
alone can not be used to reflect the relatedness between cultivars, scientifically
based molecular marker systems must follow.
Morphological and molecular systematic techniques each have distinct advantages
for phylogenetic reconstruction. Morphological techniques are applicable to an
enormous range of plant material, and a large portion of organisms will continue
to be studied primarily from morphological information. On the other hand, the
potential molecular data set is incredibly extensive and, when fully utilised, should
provide a detailed record of the history of life. Studies that combine the two
approaches can thereby maximise both information content and usefulness (Billis,
1987). In this study it was obvious that such combinations of molecular and
morphological studies provide a truly comprehensive view of the genetic variation
within a group of cultivars.
93
CHAPTER ï
SUMMARY
AFLP (Amplified Fragment Length Polymorphism) analysis has found widespread
use in DNA fingerprinting because of its relative simplicity, low cost and high
marker output. AFLP analysis also does not have the limitations associated with
morphological characterisation and the use of other molecular markers such as
RFLP's (Restriction Fragment Length Polymorphisms) and RAPD's (Random
Amplified Polymorphic DNA markers). Furthermore, the AFLP technique has
found widespread use in the identification of cultivars, varieties and breeding
lines. The accurate identification of breeding material is extremely important, not
only for the protection of breeder's rights, but also to accelerate plant-breeding
programmes. The aim of this study was to use AFLP fingerprinting for the
identification and genetic distance analysis of a collection of maize inbred lines
from the Advanta Africa genebanks.
Fifty maize inbred lines from Africa (local), Argentina, Zimbabwe (CIMMYT),
Garst (USA) and Thailand were fingerprinted using AFLP's. A total of 107
scored AFLP loci were used to calculate pairwise genetic distances. This revealed
an average genetic distance of 4.38% between all lines studied. The trend of the
clustering pattern was more variation in the major clusters with less variation
between lines in subgroups and even less variation within the minor groups. It
was also further illustrated that the local and foreign inbred lines were not grouped
separately. Neighbour joining cluster analysis of the genetic distance data yielded
a dendogram that indicated the absence of relation between geographic diversity
and genetic diversity.
The resulting knowledge of genetic distance and identification of maize inbred
lines in this study will contribute towards maize breeding programmes in Advanta
Africa Seeds. It permits an organisation of germplasm resources and identification
of parents for crossing blocks. This will enable the breeder to make more
94
scientific based choices, where both additive and non-additive sources of genetic
variation contribute to the gain. Our results have shown that AFLP technology is a
rapid, informative and precise technique for identification of maize inbred lines in
this study.
Key terms: Amplified Fragment Length polymorphism (AFLP), AFLP markers,
DNA fingerprinting, genetic distance, genetic variation, maize inbred
lines, morphological characterisation, genetic maps.
95
OJP>SOMMThTG
AFLP (Amplified Fragment Length Polymorphism) analise het 'n wye toepassing
gevind in DNA bandpatroon identifikasie a.g.v. die relatiewe eenvoudigheid
daarvan, lae kostes en 'n hoë merker uitset. AFLP analise het ook nie die
beperkings geassosieer met morfologiese karakterisering, en die gebruik van
sommige molekulêre merkers soos RFLP's (Restriction Fragment Length
Polymorphisms) en RAPD's (Random Amplified Polymorphic DNA markers) nie.
Verder het die AFLP tegniek 'n wye toepassing gevind in identifikasie van
kultivars, variëteite en teellyne. Die akkurate identifikasie van teelmateriaal is
geweldig belangrik, nie net vir die beskerming van plantetelersregte nie, maar ook
om teelprogramme te versnel. Die doel van hierdie studie was om die AFLP
tegniek te gebruik vir die identifikasie en genetiese afstandsbepaling van 'n
versameling van ingeteelde mielielyne wat teenwoordig is in Advanta Africa
Seeds geenbanke.
Vyftig ingeteelde lyne van Afrika (lokaal), Argentinië, Zimbabwe (CIMMYT),
Garst (Amerika) en Thailand se bandpatrone is bepaal met AFLP analise. 'n
Totaal van 107 vasgestelde AFLP loci is gebruik om paargewyse genetiese
afstande te bereken. Dit het 'n gemiddelde genetiese afstand van 4.38% aangetoon
tussen al die ingeteelde lyne wat bestudeer is. Die algemene neiging van die
groepering was meer variasie in die hoofgroepe, met minder variasie III die
subgroepe en selfs minder variasie in die kleiner groepe. Dit was verder
geïllustreer deurdat die lokale en oorsese lyne nie apart gegroepeer was nie.
''Neighbour joining" analise van die genetiese afstand data het 'n dendogram
gegee wat aangedui het dat daar geen verwantskap is tussen geografiese- en
genetiese diversiteit nie.
Die kennis wat III hierdie studie gegenereer is van genetiese afstande en
identifikasie van ingeteelde mielielyne sal bydra tot verbetering van mielie
teelprogramme in Advanta Africa Seeds. Dit verskaf die kennis om kiemplasma
bronne te organiseer, en om geskikte ouers te kies vir kruisingsblokke. Dit sal die
teler in staat stelom meer wetenskaplik gebaseerde besluite te neem, waar beide
additiewe en nie-additiewe bronne van genetiese variasie bydra tot gewas
verbetering. Hierdie studie het aangetoon dat AFLP tegnologie vinnig, informatief
en presies is vir die identifikasie van die ingeteelde mielielyne wat bestudeer is.
96
CHAPTER8
List of References
AJMONE-MARSAN, P., LIVINI, C., MESSMER, M.M., MELCHINGER A.E. and
MOTTO, M., 1992. Cluster analysis ofRFLP data from related maize inbred
lines of the BSSS and LSC heterotic groups and comparison with pedigree
data. Euphytica 60(2): 139-148.
AKKAYA. M.S., BHAGWAT, A.A. and CREGAN, P.B., 1992. Length
polymorphisms of simple sequence repeat DNA in soybean. Genetics 132:
1131-1139.
ANDERSON, E. and BROWN, W.L., 1952. Origin of Corn Belt maize and its
genetic significance. In: Heterosis. J.W. Gowen (ed.). pp 124-148. Iowa State
College Press, Ames, lA.
ARUNACHALAN, V., BANDOYPADHYAY, A., NIGAN, S.N., GIBBONS, R.W.,
1984. Heterosis in relation to genetic divergence and specific combining
ability in groundnut (Arachis hypogea L.). Euphytica 33: 33-35.
ARUS, P., TANKSLEY, S.D., ORTON, T.J. and JONES, R.A., 1982.
Electrophoretic variation as tool for determining seed purity and for breeding
hybrid varieties of Brassica oleracea. Euphytica 31: 417-428.
BARRETT, B.A. and KIDWELL, K.K., 1998. AFLP-Based genetic diversity
assessment among wheat cultivars from the Pacific Northwest. Crop Sci. 38:
1261- 1271.
BARRETT, B.A., KIDWELL, K.K. and FOX, P.N., 1998. Comparison of AFLP and
pedigree-based genetic diversity assessment methods using wheat cultivars
from the Pacific Northwest. Crop Sci. 38: 1271-1278.
BASSIRI, A., 1976. Barley cultivar identification by use of isozyme electrophoretic
patterns. Can. J Plant Sci. 56: 1-6.
97
BEECHING, J.R., MARMEY, P., HUGLES, M.A. and CHARRIER, A., 1995.
Evolution of molecular markers and approaches for determining genetic
diversity in cassava germplasm. In: Proceedings of the 2nd Scientific Meeting
of Cassava Biotechnology Network, Bogor, Indonesia. W. Roca & A.M. Thro
(eds.). pp. 80-89. Working Document 150. CIAT, Cali, Colombia.
BEECHING, J.R., MARMEY, P., GAVALDA, M-C., NOIROT, M., HAYSON,
H.R., HUGHES, M.A. and CHARRIER, A., 1993. An assessment of genetic
diversity within a collection of cassava (Manihot esculenta Crantz)
germplasm using molecular markers. Annuals of Botany 72(6): 515-520.
BIRD, R.M., GONZALEZ DE LEON, D. and HOISINGTON, D., 1997. Designation
ofSSR morphs. Maize Genetics Cooperation Newsletter, No. 71(63): 77-79.
BIRKENMEIER, E.H., SCHNEIDER, U. and THURSTON, S.J., 1992.
Fingerprinting genomes by use of PCR with primers that encode protein
motifs or contain sequences that regulate gene expression. Mammalian
Genome 3(10): 537-545.
BOHN, M., UTZ, H.F. and MELCHINGER, A.E., 1999. Genetic similarities among
winter wheat cultivars determined on the basis ofRFLPs, AFLPs, and SSRs
and their use for predicting progeny variance. Crop Sci. 39: 228-237.
BONNETT, O.T., 1960. The inflorescences of maize. Crop Sci. 120(3107): 77-87.
BONNETT, O.T., 1954. The developmental anatomy of the corn plant. Proc. is"
Corn Res. Conf., pp. 40-47. Amer. Seed Trade Assoc.
BOPPENMAIER, J., MELCHINGER, A.E., BRUNKLAUS-JUNG, E., GEIGER,
H.H. and HERRMANN, R.G., 1992. Genetic Diversity for RFLPs in
European Maize Inbreds: I. Relation to performance of Flint x Dent Crosses
for Forage Traits. Crop Sci. 32: 895-902.
BOTSTEIN, D., WHITE, R.L., SKOLNICK, M.H. and DAVIES, R.W., 1980.
Construction of a genetic linkage map in man using restriction fragment length
polymorphisms. Am. J Hum. Genet. 32: 314-331.
BROWN, A.H.D., GRACE, lP. and SPEER, S.S., 1987. Designation ofa "core"
collection of perennial Glycine. Soybean Genet. News!. 14: 59-70.
BURR, B., EVOLA, S.V., BURR, F.A. and BECKMANN, J.S., 1983. The
application of restriction fragment length polymorphism to plant breeding. In:
Genetic engineering principle and methods. lK. Setlow & A. Hollaender
(eds.). pp. 45-49. Plenum Press, London.
98
BURSTIN, J., VIENNE D., DE DUBREUIL, P., DAMERVAL, C. and DE VIENNE,
D., 1994. Molecular markers and protein quantities as genetic descriptors in
maize. I. Genetic diversity among 21 inbred lines. Theor. Appl. Genet. 89 (7-
8): 943-950.
CAETAO-ANOLLÉS, G.C., BASSAM, BJ. and GRESSHOFF, P.M., 1991. DNA
amplification fingerprinting: a strategy for genome analysis. Plant Mol. Bioi.
Rep. 9: 553-557.
CAO, D. AND OARD, lH., 1997. Pedigree and RAPD-based DNA analysis of
commercial U.S. rice cultivars. Crop Sci. 37: 1 630-1 635.
CASTLEBERRY, R.M., CRUM, C.W. and KRULL, C.F., 1984. Genetic yield
improvement ofU.S. maize cultivars under varying fertility and climatic
environments. Crop Sci. 24: 33-36.
CIAT, 1993. Biotechnology research Unit. Annual Report. Cali. Colombia.
COBB, D.B. and CLARKSON, J.M., 1994. A simple procedure for optimizing the
polymerase chain reaction (PCR) using modified Taguchi methods. Nuc/.
Acid. Res. 22(18): 3801-3805.
COWEN, N.M. and FREY, KJ., 1987. Relationship between genealogical distance
and breeding behavior in oats (Avena sativa L.). Euphytica 36: 413-424.
COWNIE, P.J., 1993. The maize giant in a changing Africa and its challenge for
cereal science and technology. In: Cereal science and technology impact on a
changing Africa. lR.N. Taylor, P.G. Randall and J.H. Viljoen (eds.). pp.31-
52. CSIR, Pretoria, South Africa.
COX, T.S. and MURPHY, J.P., 1990. The effect of parental divergence on F2
heterosis in winter wheat crosses. Theor. Appl. Genet. 79: 241-250.
CREGAN, P.B. and BUSCH, R.H., 1978. Heterosis, inbreeding, and line
performance in crosses of adapted spring wheat. Crop Sci. 18: 247-251.
DELLAPORTA, S.L., WOOD, r. and HICKS, lB., 1983. A plant molecular DNA
minipreparation version: II. Plant Mol. Bioi. Rep. 1: 19-21.
DEVOS, K.M. and GALE, M.D., 1992. The use of random amplified polymorphic
DNA markers in wheat. Theor. Appl. Genet. 84: 567-572.
DONINI, P., ELIAS, M.L., BOUGOURD, S.M. and KOEBNER, R.M.D., 1997.
AFLP fingerprinting reveals pattern differences between DNA extracted from
different plant organs. Genome 40: 521-526.
99
DOYLE, P., 1993. Biotechnology in action. Science and Public Affairs London,
Winter.
DUDLEY, J.W., SAGHAl MAROOF, M.A., RUFENER, G.K. and MAROOF,
M.A.S., 1991. Molecular markers and grouping of parents in maize breeding
programs. Crop Sci. 31(3): 718-723
DUVICK, D.N., 1977. Genetic rates of gain in hybrid maize yield during the past 40
years. Maydica 22: 187-196.
DU PLESSIS, 1.G., 1993. Maize yield and grain quality as affected by low input
production practices. In: Cereal science and technology impact on a changing
Africa. 1.R.N. Taylor, P.G. Randall and J.H. Viljoen (eds.). pp.163-172.
CSIR, Pretoria South Africa.
ECHEVERRIGARA Y, S., DELAMARE, A.P.L., TOSELLO, G., GAL, 0., HILLEL,
1. and LAVI, U., 1996. Evaluation of the genetic diversity among maize lines
(Zea mays L.) by DNA fingerprinting analysis. Ciencia Rural 26(3): 501-503.
EDWARDS, K., JOHNSTONE, C. and THOMPSON, C., 1991. A simple and rapid
method for the preparation of plant genomic DNA for PCR analysis. Nuc/.
Acid. Res. 19: 1349- 1458.
EFISUE, A.A., 1993. Characterisation of lITA elite clones of cassava, yam and sweet
potato by morphological and isozyme technique. B.Sc. (Hons) thesis.
University of Ibadan. Ibadan, Nigeria.
ELLIS, R.P., MCNICOL, J.W., BAIRD, E., BOOTH, A., LAWRENCE, P.,
THOMAS, B. and POWELL,W., 1997. The use of AFLPs to examine genetic
relatedness in barley. Molecular Breeding 3: 359-369.
EHRLICH, H.A., GELFLAND, D.H. and SNINSKY, 1.J., 1991. Crop Sci. 252:
1643-1650.
EUBANKS, M.W., 1997. Molecular analysis of crosses between Tripsacum
dactyloides and Zea diploperennis (Poaceae). Theor. Appl. Genet. 94(6-7):
707-712.
FELSENSTEIN, 1., 1993. PHYLIP: Phylogeny Inference Package, version 3.5p.
Computer program distributed by the University of Washington, Washington.
GALLAIS, A., 1979. The concept of varietal ability in plant breeding. Euphytica 28:
811-823.
100
GARDINER, J.M., COE, E.H., MEUA-HANCONCK, S., HOISINGTON, D.A. and
CHOA, S., 1993. Development of a core RFLP map in maize using an
immortalized F2 population. Genetics 134: 917-930.
GELFAND, D.H., 1989. Taq DNA polymerase. In: PCR Technology Principles and
Applications for DNA amplification. H.A. Erlich (ed). pp 17-22. Stockton
Press New York.
GERDES, J.T. and TRACY, W.F., 1994. Diversity of Historically Important Sweet
Corn Inbreds as Estimated by RFLPs, Morphology, Isozymes, and Pedigree.
Crop Sci. 34: 26-33.
GIBCO BRL, 1996. AFLP analysis system I. Technical Bulletin 34.
GOODMAN, M.M. and BROWN, W.L., 1988. Races of Maize. In: Com and com
improvement. 3rd ed. G.F. Sprague and J.W. Dudley (eds). pp 33-79. Agron.
Monogr. 18. ASA, Madison, WI.
GOODMAN, M.M., 1972. Distance analysis in biology. Syst. Zoo/. 21: 174-186.
GOODMAN, M.M. and PATERNIANI, E., 1969. The races of maize: III. Choices
of appropriate characters for racial classification. Econ. Bot. 23: 265-271.
HALLAUER, A.R. and MlRANDA, J.B., 1988. Quantitative genetics in maize
breeding (Second Edition). Iowo State University Press, Ames Iowa
HALLAUER, A.R., RUSSELL, W.A. and LAMKEY, K.R., 1988. Corn breeding. In:
Com and com improvement. 3rd ed. G.F. Sprague and J.W. Dudley (eds.).
pp.463-564. Agron. Monogr. 18. ASA, CSSA, and SSSA, Madison, WI.
HALLAUER, A.R., 1987. Maize. In: Principles of cultivar development. W.R.Ferh
(ed). Macmillan Publishing Co., New York.
HELLENTJARIS, T., SLOCUM, M., WRIGHT, S., SCHAEFER, A. and
NIENHUIS, 1., 1986. Construction of genetic linkage maps in maize and
tomato using restriction fragment length polymorphism. Theor. App/. Genet.
72: 761-769.
HILLE, J., ZABEL, P. and KOORNNEEF, M., 1991. Genetic transformation of
tomato and prospects for gene transfer. In: Genetic improvement of tomatoes.
G. Kalloo (ed.). Monographs on Theoretical and applied Genetics Vol. 14. pp.
283. Springer- Verlag, Berlin Heidelberg.
HILUS, D.M., 1987. Molecular versus morphological approaches to systematics.
Ann. Rev. Eco/. Syst. 18: 23-42.
101
HOSAKA, K. and HANNEMAN, E., 1994. Random amplified polymorphic DNA
markers detected in a segregating hybrid population of Solanum chacoense x
S. phureja. Jpn. J Genet. 69: 53-66.
HU, J. and QUIROS, C.F., 1991. Identification of broccoli and cauliflower cultivars
with RAPD markers. Plant Cel! Rep. 10: 505-511.
JOHN, M.E., 1992. An efficient method for isolation of RNA and DNA from plants
containing polyphenolics. Nuc/. Acid. Res. 20 (9): 55-59.
JONES, C.J., EDWARDS, KJ., CASTAGLIONE, S., WINFIELD, M.O., SALA, F.,
VAN DE WIEL, C., BREDEMEIJER, G., VOSMAN, B., MATTHES, M.,
DALY, A., BRETTSCHNEIDER, R., BETTINI, P., BUIATTI, M.,
MAESTRI, E., MALCEVSCHI, A., MARMIROLI, N., AERT, R.,
VOLCKAERT, G., RUEDA, J., LINACERO, R., VAZQUEZ, A. and KARP,
A., 1997. Reproducibility testing ofRAPD, AFLP, and SSR markers in plant
by a network of European laboratories. Molecular Breeding 3: 381-390.
JONES, M.G.K. and LINDSEY, K., 1990. Plant biotechnology. In: Molecular
biology & biotechnology (second edition). J.M. Walker and E.B. Gingold
(eds.). pp. 117- 158. Whitstable Litho Ltd, Whitstable, Kent.
JUGENHEIMER, R.W., 1976. Corn improvement, seed production, and uses. John
Wiley & Sons, Illinois, USA.
KARP, A., SEBERG, O. AND BUIATTI, M., 1996. Molecular techniques in the
assessment of botanical diversity. Ann. Bot. 78:143-149.
KEIM, P., SCHUPP, J.M., TRAVIS, S.E., CLAYTON, K., ZHU, T., SHI, L.,
FERREIRA, A. and WEBB, D., 1997. A High-Density Soybean Genetic
Map Based on AFLP Markers. Crop Sci. 37: 537-543.
KELLY, J.D., 1995. The use of random amplified polymorphic DNA markers in
breeding for major gene resistance to plant pathogens. Hart Science. 30(3):
461-465.
KIESSELBACH, T.A., 1980. The Structure and Reproduction of corn. University of
Nebraska Press, Nebraska USA.
KOCHERT, G., 1994. RFLP technology. In: DNA-based markers in plants. Ronaid
L. Phillips & Indra K. Vasil (eds.). pp. 9-35. Kluwer Academic Publishers,
Netherlands.
102
KOEBNER, R.M.D., KIRSCH, F., THORPE, C. and PRINS, R., 1998. AFLP as a
source of STS markers in alien introgression. In: Proceedings of the 9th
International wheat genetics symposium. A.E. Slinkard (ed), Saskatoon
Saskatchewan, Canada. pp.118-122.
KOEBNER, R.M.D., DEVOS, K.M. and GALE, M.D., 1994. Advances in the
application of genetic markers in plant breeding. Asian Seed 1994, Chiang
Mai, Thailand.
KULESHOV, N.N., 1933. World's diversity of phenotypes of maize. Journal of the
American Society of Agronomy 25: 688-700.
LAMALA Y, r.c, TEJWANI, R., RUFENER, G.K.l., 1990. Isolation and analysis of
genomic DNA from single seeds. Crop Sci. 30: 1079-1084.
LANGRIDGE, P. and CHALMERS, K., 1998. Techniques for marker development.
In: Proceedings of the 9th International wheat genetics symposium. A.E.
Slinkard (ed). pp. I07-117. Saskatoon Saskatchewan, Canada.
LEE, M., 1995. DNA markers and plant breeding programs. Adv. Agron. 55: 265-
344.
LEE, M., GODSHALK, E.B., LAMKEY, K.R. and WOODMAN, W.W., 1989.
Association of restriction fragment length polymorphisms among maize
inbreds with agronomic performance of their crosses. Crop Sci. 29(4):
1067-1071. Dep. Agron., Iowa State Univ., Ames, lA 50011, USA.
LIN, D., HUBBES, M. and ZSUFFA, L., 1993. Differentiation of poplar and willow
clones using RAPD fingerprints. Tree Physiology 14: 1097- 1105.
LIU, X.C. and WD, lL., 1998. SSR heterogenic patterns of parents for marking and
predicting heterosis in rice breeding. Molecular Breeding 4: 263-268.
LID, Z. and FDRNIER, G. R., 1993. Comparison ofallozyme, RFLP, and RAPD for
revealing genetic variation within and between trembling aspen and bigtooth
aspen. Theor. Appl. Genet. 87: 97- 105.
LODIE, R., FINDLEY, W.R., KNOKE, J.K. and MCMDLLEN, M.D., 1991. Genetic
basis of resistance in maize to five maize dwarf mosaic virus strains. Crop Sci.
31(1): 14-18.
LUBBERSTEDT, T., MELCHINGER, A.E., FÁHR, S., KLEIN, D., DALLY, A. and
WESTHOFF, P., 1998. QTL Mapping in Testerosses of Flint Lines of Maize:
Ill. Comparison across Populations for Forage Traits. Crop Sci. 38: 1278-
1289.
103
MACKILL, DJ., ZHANG, Z., REDONA, E.D. and COLOWIT, P.M., 1996. Level of
polymorphism and genetic mapping of AFLP markers in rice. Genome 39:
969-977.
MARSAN, P.A., EGIDY, G., MONFREDINI, G., DI SILVESTRO, S. and MOTTO,
M., 1993. RAPD markers in maize genetic analysis. Maydica 38: 259-264.
MAUGHAN, P.J., SAGHAI-MAROOF, M.A. and BUSS, G.R, 1996. Amplified
fragment length polymorphism (AFLP) in soybean species diversity,
inheritance, and near-isogenie line analysis. Theor. Appl. Genet. 93: 392-401.
MAUGHAN, P.J., SAGHAl MAROOF, M.A., BUSS, G.R and HUESTIS, G.M.,
1996. Amplified fragment length polymorphism (AFLP) in soybean: species
diversity, inheritance, and near-isogenie line analysis. Theor. Appl. Genet. 93:
392-401.
MAZUR, B.J. and TINGEY, S.V., 1995. Genetic mapping and introgression of genes
of agronomic importance. Current opinion in Biotechnology 6: 175-182.
McCLEAN, P.L., ERWING, 1., LINCE, M. and GRAFTON, K., 1994. Development
of a RAPD map of Phaseolus vulgaris L. Ann. Rpt. Bean. Coop. 37: 79-80.
MELCHINGER, A.E., 1993. Use of RFLP markers for analysis of genetic
relationships among breeding materials and prediction of hybrid performance.
Crop Sci. 28: 621-628.
MELCHINGER, A.E., LEE, M., LAMKEY, K.R and WOODMAN, W.L., 1990.
Genetic diversity for restriction fragment length polymorphisms: Relation to
estimated genetic effects in maize inbreds. Crop Sci. 30: 1033-1040.
MESSMER, M.M., MELCHINGER, A.E., BOPPENMAlER, 1., BRUNKLAUS-
JUNG, E. and HERRMANN, RG., 1992. Relationship among early European
maize inbreds: 1. genetic diversity among flint and dent lines revealed by
RFLPs. Crop Sci. 32: 1301-1309.
MESSMER, M.M., MELCHINGER, A.E., LEE, M., WOODMAN, W.L., LEE, E.A.
and LAMKEY, K.R., 1991. Genetic diversity among progenitors and elite
lines from the Iowa Stiff Stalk Synthetic (BSSS) maize population.
Comparison of allozyme and RFLP data. Theor. Appl. Genet. 83: 97-107.
MICHELI, M.R, BOVA, R, PASCALE, E., D' AMBROSIO, E., 1994. Reproducible
DNA fingerprinting with the random amplified polymorphic DNA (RAPD)
method. Nuc/. Acid. Res. 22: 1921-1922.
104
MICHELMORE, RW., KESSELI, R.V., FRANCIS, D.M., PARAN, I., FORTIN,
M.G. and YANG, C.H., 1992. Strategies for cloning plant disease resistance
genes. Molecular plant pathology 2. A practical approach. Oxford University
Press, Oxford.
MICHELMORE, RW., PARAN, I. and KESSELI, RV., 1991a. Proceedings of the
National Academy of Sciences of the USA. 88: 9828-9832.
MICHELMORE, RW., PARAN, I. and KESSELI, R.V., 1991b. Identification of
markers linked to disease-resistance genes by bulked segregant analysis: a
rapid method to detect markers in specific genomic regions by usmg
segregating populations. Proc. Nat!. Acad. Sci. USA 88: 9 828-9 832.
MILLAN, T.T. and CUBERO, A.M., 1995. Varietal identification in Rosa by
isozymes and RAPD markers. In: Anonymous. 2nd Int Rosa Symp.
ISHS/INRA Antibes. France (in press).
MOCK, JJ. AND FREY, K.J., 1970. Leaf number- An index of meioses in corn.
Crop Sci. 10(6): 659-661.
MONEY, T., READER, S., QU, L.J., DUNFORD, RP. and MOORE, G., 1996.
AFLP-based mRNA fingerprinting. Nucl. Acid. Res. 24(13): 2616-
2617.
MORELL, M.K., PEAKALL, R., APPELS, R, PRESTON, L.R and LLOYD, H.L.,
1995. DNA profiling techniques for plant variety identification. Aus. J Exp.
Agric. 35: 807-819.
MUEHLBAUER, GJ., SPECHT, J.E., THOMAS-COMPTON, M.A., STASWICK,
P.E. and BERNARD, RL., 1988. Near-isogenie lines- a potential resource in
the integration of conventional and molecular marker linkage maps. Crop
Sci. 28: 279-735.
MULLIS, K.B. and FALOONA, F.A., 1987. Methods Enzymol. 155: 335-350.
MURRAY, M.G., MA, Y., ROMERO-SEVERSON, J., WEST, D.P. and CRAMER,
J.H., 1988. Restriction fragment length polymorphisms: What are they and
how can breeders use them. In: Proceedings of the 43rd Annual Corn and
Sorghum Research Conference. D. Wilkinson (ed.). pp. 72-87. ASTA,
Washington, DC.
MURRA Y, H.G. and THOMPSON, W.F., 1980. Rapid isolation of higher molecular
weight DNA. Nucl. Acid. Res. 8: 4321.
105
MYBURG, A.A., BOTHA, A-M., WINGFIELD, B.D., 1996. Characterisation of
RAPD markers linked to Russian wheat aphid resistance. Plant Physiology
111: 89.
NEl, M. and LI, W.H., 1979. Mathematical model for studying genetic variation in
terms of restriction endonucleases. Proc. Nat!. Acad. Sci. USA 76: 5269-
5273.
NIENHUIS, J., TIVANG, 1. and SKOCK, P., 1995. Genetic relationships among
cultivars and landraces of lima bean (P. lunatus L.) as measured by RAPD
markers. JAmer. Soc. Hort. Sci. 120: 300- 306.
NICHOLSON, P., REZANOOR, H.N. and SU, H., 1993. Use of random amplified
polymorphic DNA (RAPD) analysis and genetic fingerprinting to differentiate
isolates of race 0, C and T of Bipolaris maydis. Journal of Phytopathology
139(3): 261-267.
NYBOM, H., 1994. DNA fingerprinting-A useful tool in fruit breeding. Euphytica
77: 59-64.
NYBOM, HY., ROGSTAD, S.H. and SCHAAL, B.A., 1990. Genetic variation
detected by use of the M13 'DNA fingerprint' probe in Malus, Prunus and
Rubus (Rosaceae). Theor. Appl. Genet. 79: 153-156.
0' TOOLE, J.C., 1989. Breeding for drought resistance in cereals: emerging new
technologies. In: Drought resistance in cereals. F.W.G. Baker (ed). pp.81-94.
CAB International; Wallingford; UK.
PAKNIYAT, H., POWELL, W., BAIRD, E., HANDLEY, L.L., ROBINSON, D.,
SCRIMGEOUR, e.M., NEVO, E., HACKETT, C.A., CALIGRA, P.D.S. and
FORSTER, B.P., 1997. AFLP variation in wild barley with reference to salt
tolerance and associated ecogeography. Genome 40: 332-341.
PATERSON, A.H., TANKSLEY, S.D. and SORRELLS, M.E., 1991. DNA markers
in plant improvement. Advances Agron. 46: 39-90.
PAUL, S., WACHlRA, F.N., POWELL, W. and WAUGH, R., 1996. Diversity and
genetic differentiation among populations ofIndian and Kenyan tea (Camellia
sinensis(L.) O. Kuntze) revealed by AFLP markers. Theor. App!. Genet. 94:
255-263.
PERKIN-ELMER, 1996. AFLP Plant Mapping Kit- the PCR marker of choice for
plant mapping.
106
PHELPS, T.L., HALL, A.E. and BUCKNER, B., 1996. Microsatellite repeat
variation within the yl gene of maize and teosinte. Journal of Heredity 87(5):
396-399.
POWELL, W., THOMAS, W.T.B., BAIRD, E., LAWRENCE, P., BOOTH, A,
HARROWER, B., McNICOL, J.W. and WAUGH, R., 1997. Analysis of
quantitative traits in barley by the use of amplified fragment length
polymorphism. Heredity. In press.
POWELL, W.M., MORGANTE, C.A, HANAFREY, M., VOGEL, J., TINGEY, S.
and RAFALSKI, A, 1996. Comparison ofRFLP, RAPD, AFLP, and SSR
(microsatelite) markers for germplasm analysis. Molecular Breeding 2: 225-
228.
PRABHU, R.R., WEBB, D., JESSEN, H., LUK, S., SMITH, S. AND GRESSHOFF,
P.M., 1997. Genetic relatedness among soybean genotypes using DNA
amplification fingerprinting (DAF), RFLP, and Pedigree. Crop Sci. 37: 1590-
1595.
ROLFS, A, SCHULLER, I., FINCKH, U. and WEBER-ROLFS, I., 1992. PCR:
Clinical Diagnostics and Research. Springer- Verlag, New York.
ROUPPE VAN DER VOORT, J.N.AM., VAN ECK, HJ., DRAAISTRA, J., VAN
ZANDVOORT, P.M., JACOBSON, E. and BAKKER, J., 1998. An online
catalogue of AFLP markers covering the potato genome. Molecular Breeding
4: 73-77.
ROUPPE VAN DER VOORT, J.N.AM., VAN ZANDVOORT, P., VAN ECK, H.J.,
FOLKERTSMA, R.T., HUTTEN, R.C.B., DRAAISTRA, J., GOMMERS,
FJ., JACOBSEN, E., HELDER, J. and BAKKER, J., 1997. Use of allele
specificity of comigrating AFLP markers to align genetic maps from different
potato genomes. Mol. Gen. Genet. 255: 438-447.
RUANO, G., BRASH, D.E. and KIDD, K.K., 1991. The first few cycles.
Amplifications: A Forum for PCR users 4: 5-7.
RUSSELL, W.A, 1974. Comparative performance for maize hybrids representing
different eras of maize breeding. In: Proc. Annu. Corn and Sorghum Ind. Res.
Conf. 29t\ Chicago, IL., 10-12 Dec. Am. Seed Trade Assoc. pp.81-101.
Washington, DC.
107
SAGAI-MAROOF, M.A, BIYASHEV, R.M., YANG, G.P., ZHANG, Q. and
ALLARD, R.W., 1994. Extraordinarily polymorphic micro satellite DNA in
barley: Species diversity, chromosomal locations, and population dynamics.
Proc. Nat!. Acad. Sci. 91: 5466-5470.
SAIKI, R.K., GELFLAND, D.H. STOFFEL, S., SCHART, S.J., HIGUCHI, R.,
HORN, G.T., MULLIS, K.B. and EHRLICH, H.A, 1988. Science 239: 487-
491.
SALHUANA, W., POLLAK, L.M., FERRER, M., PARATORI, O. AND VIVO, G.,
1998. Breeding potential of maize accessions fromn Argentina, Chile, USA
and Uruguay. Crop Sci. 38: 866-872.
SAMEC, P. and NASINEC, V., 1995. Detection of DNA polymorphism among pea
cultivars using RAPD technique. Biologia Plantarum 37(3): 321-327.
SENIOR, M.L., MURPHY, J.P., GOODMAN, M.M. and STUBER, C.W., 1998.
Utility of SSR's for determining genetic similarities and relationships in maize
using an agarose gel system. Crop Sci. 38: 1088-1098.
SIMONS, G., VAN DER LEE, T., DIERGAARDE, P., VAN DAELEN, R.,
GROENENDIJK, J., FRIJTERS, A, BUSCHGES, R., HOLLRICHER, K.,
TOPSCH, S., SCHULZE-LEFERT, P., SALAMINI, F., ZABEAU, M. and
VOS, P., 1997. AFLP-based fine mapping of the Mlo gene to a 30-kb DNA
segment of the barley genome. Genomies 44: 598-604.
SMITH, J.S.C. and SMITH, O.S., 1991. Restriction fragment length polymorphisms
can differentiate among U.S. maize hybrids. Crop Sci. 31: 893-899.
SMITH, O.S., SMITH, J.S.C., BOWE, S.L., TEMBORG, R.A, WALL, S.J., 1990.
Similarities among a group of elite maize inbreds as measured by pedigree Fl
grain yield, grain yield, heterosis and RFLP. Theor. App!. Genet. 80: 833-840.
SMITH, J.S.C. and SMITH, O.S., 1989. The description and assessment of distances
between inbred lines of maize: I. The use of morphological traits as
descriptors. Maydica 34: 141-150.
SNEATH, P.H.A and SOKAL, R.R., 1973. Numerical taxonomy .. The principles
and practice of numerical classification. W.H. Freeman and Co., San
Francisco.
SOBRAL, B.W.S. and HONEYCUTT, R.J., 1994a. The polymerase chain reaction.
In: Genetics, plants, and the polymerase chain reaction. K.B Mullis, F., Freré,
Gibbs, R.A (eds). Birkhauser, Boston, pp. 304-319.
108
SOBRAL, B.W.S. and HONEYCUTT, RJ., 1994b. High output genetic mapping in
polyploid using PCR-generated markers. Theor. Appl. Genet. 86: 105-112.
SOLLER, M. and BECKMANN, lS., 1983. Genetic polymorphism in varietal
identification and genetic improvement. Theor. Appl. Genet. 67: 25-33.
SOUTHERN, E.M., 1975. Detection of specific sequences among DNA fragments
separated by gel electrophoresis. J Mol. Bioi. 98: 503-517.
SPRINGER, P.S., EDWARDS, KJ. and BENNETZEN, J.L., 1994. DNA class
organization on maize Adhl yeast artificial chromosomes. Proceedings of the
National Academy of Sciences of the United States of America 91(3): 863-867.
STAM, P. and ZEVEN, A.C., 1981. The theoretical proportion of the donor genome
in near-isogenie lines of self-fertilizers bred by backcrossing. Euphytica 30:
227-238.
STEGEMANN , H., 1984. Retrospect on 25 years of cultivar identification by protein
patterns and prospects for the future. Seed Sci. & Technol. 15: 20-31.
STUBER, C.W., BROWN A.H.D., CLEGG M.T., KAHLER A.L. and WEIR B.S.,
1990. Molecular markers in the manipulation of quantitative characters.
Sinauer Associates Inc.; Sunderland, Massachusetts; USA.
SWOFFORD, D.L., 1991. PAUP: Phylogenetic analysis using parsimony, version
3. Os. Computer program distributed by the Illinois Natural History Survey,
Champaign, IL.
TAl, T.H. and TANKSLEY, S.D., 1990. A rapid and inexpensive method for
isolation oftotal DNA from dehydrated plant tissue. Plant. Mol. Biol. Rep. 9:
297-303.
TANKSLEY, S.D., YOUNG, N.D., PATERSON, A.H. and BONIERBALE, M.W.,
1989. Bio/Technology 7: 257-264.
TANKSLEY, S.D., 1983. Molecular markers in plant breeding. Plant Mol. Bioi.
Rep. 1: 3-8.
THAKUR, H.L. and ZARGER, M.A., 1989. Heterosis in relation to genetic
divergence and specific combining ability in indian mustard Brassica
juncea L. Czern. Indian J Genet. Plant Breed. 49:223-226.
THOMAS, e.M., VOS, P., ZABEAU, M., JONES, D.A., NORCOTT, K.A.,
CHADWICK, B.P. and JONES, lD.G., 1995. Identification of amplified
fragment length polymorphism (AFLP) markers tightly linked to the tomato Cl
9 gene for resistance to Cladosportum fulvum. Plant 8: 785-794.
109
TING, Y.C., 1994. Molecular markers of anther culture-derived plants. : Maize-
Genetics-Cooperation-Newsletter, No.68, 19-20.: Boston College, Chestnut
Hill, Boston, MA 02167, USA.
TOHME, J., GONZALEZ, D. 0., BEEBE, S. and DUQUE, M. C., 1996. AFLP
analysis of gene pools of a wild bean core collection. Crop Sci. 36: 1375-
1384.
TRAVIS, S.E., MASCHINSKI, 1. and KEIM, P., 1996. An analysis of genetic
variation in Astragalus cremnophylax var. cremnophylax, a critically
endangered plant, using AFLP markers. Mol. Ecol5: 735-745.
VAN BEUNINGEN, L.T. and BUSCH, R.H., 1997. Genetic diversity among North
American Spring Wheat Cultivars: Ill. Cluster analysis based on quantitative
morphological traits. Crop Sci. 37: 981-988.
VAN TOAI, T.T., PENG, 1. and ST. MARTIN, S.K., 1997. Using AFLP markers to
determine the genomic contribution of parents to populations. Crop Sci. 37:
1370-1373.
VAILLANCOURT, LJ. and HANAU, R.M., 1992. Genetic and morphological
comparisons of Glomerella (Colletotrichum) isolates from maize and from
sorghum. Experimental Mycology 16(3): 219-229.
VARGHESE, Y.A., KNAAK, K., SETHURAJ, M.R., ECKE, J., 1997. Evaluation of
random amplified polymorphic DNA (RAPD) markers in Hevea braziliensis.
Plant Breeding 116: 47-52.
VOS, P., HOGERS, R., BLEEKER, M., REIJANS, M., VAN DE LEE, T., HORNES,
M., FRIJTERS, A., POT, J., PELEMAN, 1., KUIPER, M. and ZABEAU, M.,
1995. AFLP: a new technique for DNA fingerprinting. Nuc/. Acid. Res.
23(21): 4407-4414.
WALTON, M., and HELENTJARIS, T., 1987. Application of restriction fragment
length polymorphism (RFLP) technology to maize breeding. In: Proc. 42nd
Ann. Com and Sorghum Res. Conf. American Seed Trade Association. ASTA.
D. Wilkinson (ed.). pp. 48-75. Washington, DC.
WANG, Z., WANG, B. and ZENG, M., 1996. RAPD analysis ofmtDNAs from
multiplasrnic CMS lines. Maize-Genetics-Cooperation-Newsletter, No. 70, 12.
Academia Sinica, Beijing, China.
110
WANG, Z.Y. and TANKSLEY, S.D., 1989. Restriction fragment length
polymorphism in Oryza sativa L. Genome 32: 1113-1118.
WAUGH, R., and POWELL, W., 1992. Using RAPD markers for crop improvement.
Tibtech 10: 186-192.
WELLHAUSEN, E.I., ROBERTS, L.M., HERNANDEZ, E. and
MANGELSDORF, P.e., 1952. Races of maize in Mexico. Bussey Inst.
Harvard Univ. Press, Cambridge.
WELSCH, r.,HONEYCUTT, R.I., MCCLELLAND, M. and SOBRAL, B.W.S.,
1991. Parentage determination in maize hybrids using the arbitrary primed
polymerase chain reaction (AP-PCR). Theor. Appl.Genet. 82: 473-476.
WILLIAMS, l.G., KUBELIK, A.R., LIVAK, K.l., RAFALSKI, l.A. and TINGEY,
S.V., 1990. DNA polymorphisms amplified by arbitrary primers are useful as
genetic markers. Nucl. Acid. Res. 18: 6 531- 6 535.
WINTER, P. and KAHL, G., 1995. Molecular marker technologies for plant
improvement. World J Mierob. & Biotech. 11: 438-448.
WHYTHE, I.V. and GEVERS, H.O., 1988. Diallel analysis of resistance of eight
maize inbred lines to Sphacelotheca reiliana. Phytopathology 78(1): 65-68.
WU, K.S. and TANKSLEY, S.D., 1993. Abundance, polymorphism and genetic
mapping of microsatelite in rice. Molec. Gen. Genet. 241: 225-235.
YAMAMOTO, T., NISHlKA WA, A. and OEDA, K., 1994. DNA polymorphisms in
Oryza sativa L. amplified by arbitrary primer PCR. Euphytica 78: 143-148.
YOUNG, N. D., 1994. Constructing a plant genetic linkage map with DNA markers.
Ronald L. Phillips & Indra K. Vasil (eds). pp. 39- 58. Kluwer Academic
Publishers. Netherlands.
YOUNG, N.D., ZAMIR, D., GANAL, M.W. and TANKSLEY, S.D., 1988. Use of
isogenie lines and simultaneous probing to identify DNA markers tightly
linked to the Tm-2a gene in tomato. Genetics 120: 579-585.
YU, K. and PAULS, K.P., 1994. Optimization of DNA-extraction and PCR
procedures for random amplified polymorphic DNA (RAPD) analysis in
plants. In: PCR Technology: Current innovations. Griffin, H.G., Griffen, A.M.
(eds). CRC Press, Inc.
YU, K. and PAULS, K.P., 1992. Optimization of the PCR program for RAPD
analysis. Nucl. Acids Res. 20: 2606.
III
YD, K., VAN DEYNZE, A. and PAULS, K.P., 1993. Random amplified
polymorphic DNA (RAPD) analysis. In: Methods in plant molecular biology
and biotechnology. pp. 287-301. CRC Press, Inc.
ZABEAU, M. and VOS, P., 1993. Selective restriction fragment amplification: a
general method for DNA fingerprinting. European Patent Application number
92402629.7. Publication number 0534858 AI.
ZANONI, U. and DUDLEY, J.W., 1989. Comparison of different methods of
identifying inbreds useful for improving elite maize hybrids. Crop Sci. 29:
577-582.
ZEHR, B.E., DUDLEY, J.W., CHOJECKI, J., SAGHAl MAROOF, M.A. and
MOWERS, R.P., 1992. Use ofRFLP markers to search for alleles in a maize
population for improvement of an elite hybrid. Theor. Appl. Genet. 83: 903-
911.
ZEVEN, A.C., KNOTT, D.R. and JOHNSTON, R., 1983. Phylogenetic distribution
and genetic mapping of a (CCC)n microsatellite from rice (Oryza sativa L.).
Plant. Mol. Bioi. 21: 607-614.
112
APPENDICES
Data matrix of AFLlP markers scored for genetic distance analysis
Table 9.1 Input data matrix used for Neighbour joining analysis of 107 AFLP markers.
WM1
111111111111111111111110000000000110000000000000000000000000000000
00000000000000000000000000000000000000000
WM2
000111000100000000000001111111111001111111000000000000000000000000
00000000000000000000000000000000000000000
WF1
101000101101010000000001000000000001000110111111111111111111000000
00000000000000000000000000000000000000000
WF2
101001100000001000000010100100100100101100100011111000011000111111
11000000000000000000000000000000000000000
WI1
000000000001001010000010000010000101110000001000000001001001000000
00111111000000000000000000000000000000000
WI2
101100010001000111000000001001000000110000100000011000001100100101
00010000111100000000000000010000000000000
WP1
010001000110110000100001100000000111000100010000011001001000001010
10010100010011111000000000000000000000010
WP3
111000100001000000000010110000000001010011000000000000000000000100
00010010011000000100001001011100000000000
WR1
101100001001000000000010010000010000000000001001001001100001010110
00000000001000100110000000000000000000000
WR2
000101001000000100001011000000000100101110100000001000001000011000
00000100010001000001111000000000000000000
113
WK1
001000010000000000000010010000000000000011110000101100001101000000
00010000010000000110001111100000000000000
WK2
011001000000000000000001000000000000110101100011000000000001000010
00000000111000010000000000011000010000000
WOl
000010010000101010001011000000000000000001010001000000000001000001
10000000000001011000000000000111101111100
W02
001110001001000000000010000000000000000001000010001000000101001000
01000000000010000000100011100100001111101
W03
001110001001000000000010000000000000000001000010001000000101001000
01000000000010000000100011100100001111101
W04
101000100000010100000000100000010000001100010100001000100000000000
00011000000000000110100000000000000100000
WOS
0100000000010100000000101000100000010100101001000000001000D1000010
00000001000000000101000000001000001010000
W06
111010000000010000000010000010010000000010010000000000100000001000
00001011000000100010010000001001100000000
W07
000000000001010010000010000000001000000010000000001001010001101001
00000000000001010001010100000000000000100
W08
001000010001010000100010010000001000000010000000011010000001001000
00000000000001100001000010001010000000010
WOg
000000010001011000100010001010000000000001100000000000000000000000
00000000010000000001000010000000000000000
WOII
000000000010100000000000000010000000000000010000010000000000100000
00000110000011000010110000000000100000000
YM1
000000001001010000000000011100000000100100000000100111001110001000
00000000001010000000000000010100000000010
114
YM2
111000100101000010000000000010011000000000100000000000000000001000
01000000001001000010001110000000100101100
YM3
000000000001011000001110001010000001010010100000100101000000000000
00101100000100100101000010100000100001000
YM4
100000000000000000000010000100000000000000001000101001100000000010
00000000000001000100000000000000110000000
YF2
101000000000010001000000000001000000001010010010001000111010000001
00100001010000000000000000100111000000000
YF3
001101010100000100100000100001001001000000010000010010000100000000
00010100100000001000000000000000010000000
YF4
001100000000000000000000000000000000000000000000000000000000000000
00010000000000000000000000000000100000001
WP2
000000000000010000001010000000000010100001000100000000000000001001
00010110011010010000000001000000000000000
W010
001010100001011011010010000000010001000000110000001000000000000000
00100001100000000010100100100001000000000
YF1
001001000001010010010000000010100000100000000000010100000000000100
00100000010000000000001000000000000000001
YI2
000101010000010000000000000001000010000000000000000000000000000000
00001100000000000000000000000000100000000
YR5
000000001011010000010010000000001100000000010011011001100000000011
00100101101000000000000001000001000000000
YR6
010000000000000000000000000000000000000000000000000000000000010000
00000000000000000000000000000000000000000
115
YL1
000000100001011100000000000000010100001010110000000000001010000000
00000010100000000000100000000000000000000
YL2
001000000000000000000000000000000000000000000000000000000000000000
00000000000000000001000000000000100000000
YL4
100001000110011001010000000010011001001010000010010101100000000001
00000000000100000001000000000001000000000
YL5
000000000000000000000010000000000000000000000000000000000000000000
00000000000000000000000000000000000000000
YS1
000100000001001000000110000001000000100000000001000000100000000101
01000000000000000000000000001000010000000
YS3
011100000000010010000010001000000001001010101000001011000010000000
00000000000000000000010000000000000000000
Y01
000000000000000000000010000000000000000000100000000000000000000000
00010010000000010100011000000101000000000
Y02
000000000000000000000010000000000000000000100000000000000000000000
00000000000000000000000000000000000000000
Y04
010000000000000000000010000000000000000000000000000000000000000000
00000000000000000000000000000000000000000
Y06
000000000000000000000000000000000000000000000000000000000000000000
00000000000010000000000000000000000000000
Y09
000000000000000000000010000000000000000000000000000000000000000000
00000000000000000000000000000000000000000
Y010
001000000010111101100010001000100000010001000010100100010000100010
00000111110100010000000000000100010000000
116
Y011
001000000000000000000000000000000000000000000000000000000000000000
00010000000000000000000000000000100000000
Y012
000000000000000000000000000010000000000000000000100000000000000000
00000000000001000000000000000000000000000
Y013
110000100010011100000000000000000010100000000000000000011000000001
00001101000000000000000000000000100000001
Each test string shows the name of the cultivar or line and the presence (1) and
absence (0) of AFLP markers.
117