Molecular investigation of genetic variation to improve uniformity of cauliflower … · Cauliflower production in Australia is export oriented. The industry aims for uniform, high

Molecular investigation of genetic variation to

improve uniformity of cauliflower production

Ida Ayu Astarini

B.Sc. (Hort), Bogor Agricultural University, West Java, Indonesia

M.Sc. (Hort), The University of Western Australia, Perth, WA

This thesis is presented for the degree of

Doctor of Philosophy

of The University of Western Australia

School of Plant Biology

Faculty of Natural and Agricultural Sciences

2006

Abstract

Cauliflower production in Australia is export oriented. The industry aims for

uniform, high quality product and is based on Fi hybrid cauliflower cultivars. However

non-uniform products still occur. The lack of uniformity may be due to genetic

variation evident in the seed, seedling or production stage.

There are many cauliflower cultivars released each year and there is a need to

correctly identify cultivars for seed companies and growers. W e developed

fingerprinting keys using molecular markers to correctly identify cultivars at any stage,

and therefore without the need for field trials. Australian Fi hybrid cultivars and open

pollinated cultivars of Indonesia were assessed genetically using R A P D markers.

Genetic variation within and between cultivars of both countries were investigated.

Dendograms were constructed using Neighbor-Joining analysis based on Phylogenetic

Analysis Using Parsimony (PAUP). D N A fingerprinting keys were developed and

genetic relationships among cultivars were determined. Comparison between

Indonesian and Australian based cultivars indicated that Indonesian cultivars have

unique genotypes and would be good sources of genes for future crop improvement.

Results proved that R A P D markers can be used for the routine identification of

cauliflower cultivars.

Hybrid cauliflowers have been developed to exploit heterosis and to improve

uniformity of production. However, 'morphological sib' plants, assumed to be self

inbred, often contaminate hybrid seed lots in the SI system and contribute up to 2 0 % of

the total harvest. Sibs produce very small and non-marketable curds. Whilst self

inbreeding is not possible in the C M S system, plants that look like sibs often occur up to

4 0 % of the crop. In this study, microsatellite markers for male and female cauliflower

parent lines of both SI and C M S systems were developed. T w o pairs of markers were

chosen for purity testing of F] hybrid seeds. Microsatellite analysis, glasshouse and

field trials confirmed that morphological sib plants from the SI system were not always

self-inbred. In contrast, most self inbred plants showed normal growth. All

morphological sibs from the C M S system showed hybrid bands. This suggests that

morphological sibs were not always due to selfing but possibly to an interaction

between genetic and environmental factors and this requires further investigation.

Variation in curd maturity which results in spread of harvesting time is another

problem in the cauliflower industry and contributes to up to a 2 0 % loss. This

phenomenon prevents the use of harvest machinery and increases cost of harvesting as

i

several manual harvests are required. Morphological variation from seed to harvest is

due to genetic variation interacting with environmental conditions and here the genetic

factors were investigated using R A P D markers. Multivariate analysis based on

principle coordinates analysis was employed to correlate morphological traits with

molecular markers across cultivars. Markers associated with seed weight, germination

rate, shoot length, root length, fresh weight and dry weight were identified.

In summary, successful application of molecular markers to screen cauliflower

plants in every stage of production, from choosing the right cultivar, screening for

particular traits to reduce seedling variability, and screening for abnormality will

significantly improve uniformity in cauliflower production.

ii

Declaration

I declare that this thesis contains no experimental materials that have been previously

presented for any degree at any other university or institution.

I and my supervisors A/Prof Julie Plummer, Dr Guijun Yan and Ms Rachel Lancaster

discussed and decided the research topics and techniques. The preparation of this

thesis, including published papers and manuscript was done by myself with feedback

from m y supervisors.

March, 2006

Ida Ayu Astarini

iii

Dedication

I dedicated this thesis to the memory of my father, the late Dr Ida Bagus

Astawa, M.P.H. Although you never lived long enough to see me through

this study, your encouragement has always been a guiding light in my life.

IV

Acknowledgments

First and foremost, I express my sincere thanks to my principal supervisor

Associate Professor Julie Plummer for her advice, patience, time, effort and supervision

during m y study. There is so much I learnt from her, not only supervision for m y

projects, but also beyond m y PhD study. Sincere thanks to m y supervisors Dr Guijun

Yan for his encouragement and guidance throughout m y study and M s Rachel Lancaster

for her suggestions and valuable feedback on m y projects, comprehensive information

on cauliflower production and access to Manjimup Horticulture Research Institute.

Thorough and detailed supervision from all of them has made m y study a success.

Financial support was generous and I would like to thank AusAID and

Department of Agriculture Western Australia for supporting this study. M y special

thanks to M s Rhonda Haskell, the AusAID liaison officer for being very supportive and

helpful during m y study.

I thank M s Anouska Cousins for technical guidance, Dr Matthew Nelson for

advice on microsatellite technique and help to make m y visit to Dutch seed companies

possible. Thank you to Associate Professor Wallace Cowling for useful advice and

comments on m y projects. Also, the Plant Breeding and Molecular Genetics Discussion

Group has broadened up m y knowledge about plant breeding. A Field trial in Manjimup

would not have been possible without the kind assistance from M r John Doust, Dr

Kristen Stirling, M r David Tooke and Grazi.

Thank you to Henderson Seeds, Enza Zaden Australia, South Pacific Seeds,

Syngenta Seeds, Lefroy Valley Seeds and Bejo Seeds for supplying seeds. Kind help

from Ahmad Rivani, Sitawati, Agus Suryanto, Dewa Okayadnya and Professor IGP

Wirawan during sample collection in Indonesia was greatly appreciated.

M y colleagues in room 2.105, Leida, Nic, Cam, Chris, Claire, Nader, Leila,

Bambang and Sharmin, it has been a fun experience sharing a study room with all of

you. I improved m y English and learned different cultures. For m y fellow Indonesian

students, Ila, Anne, Suzie, Ndari, Iin, Ita, Pharma and Titik, your companionship has

made this study an enjoyable experience.

To m y m u m , thank you for love and support. Special thanks to m y parents in

law, Peranda Gede and Peranda Istri, for always praying for m y success. M y sons,

Guntur and Ari, you are the best kids in the world!

Finally, to m y husband, Ida Bagus Gunawan, your endless support,

understanding and sacrifice has nothing to compare. Thank you!!

v

Thesis Outline

This thesis consists of 8 chapters. The first chapter contains background

information about problems in cauliflower production around the world, with particular

reference to the Western Australian cauliflower industry. This chapter also introduces

the usefulness of molecular techniques in crop improvement programs. The second

chapter, Literature Review, contains details about cauliflower with emphasis on

breeding systems and justification of current molecular techniques used in this study

and successfully applied in Brassica and other vegetable crops.

The research program consists of five projects completed during m y P h D

candidature. These are presented in Chapters 3-7. Chapter 3 is on the development of

a fingerprinting technique using R A P D marker systems on hybrid cultivars commonly

grown in Australia. This paper has been published in the Australian Journal of

Agricultural Research. In Chapter 4, an extension of the R A P D fingerprinting

technique was applied in open pollinated cultivars from Indonesia. The paper was

presented at an International/Australian Society of Horticultural Science conference

entitled "Harnessing the Potential of Horticulture in the Asian-Pacific region in Coolum,

Queensland, 1 - 3 September 2004, where it was awarded 'Young Scientist Award' for

the best student presentation. The paper has been published in Acta Horticulturae

(2005) 694, 149-152. In Chapter 5, genetic distance between Indonesian and Australian

cultivars were revealed. The paper from this chapter has been accepted in Scientia

Horticulturae.

In Chapter 6, the use of R A P D and microsatellites to identify a marker to

distinguish between male and female parent lines, hybrid and non-hybrid (commonly

known as 'sibs') plants was investigated. The manuscript of this chapter is currently

under review in Theoretical and Applied Genetics. In Chapter 7, the association

between molecular markers and morphological traits of cauliflower seedlings was

investigated. A number of associations were found and these may be useful in

molecular-assisted selection in breeding programs. A manuscript of this chapter is

currently under review with the Australian Journal of Agricultural Research.

Chapter 8 is the General Discussion of the thesis. All chapters/papers are

brought together and justify the molecular techniques employed to assist in improving

uniformity of cauliflower production.

vi

List of Publications, Conferences Attended and Awards

Publications

Astarini IA, Plummer JA, Yan G, Lancaster R A (2006) 'Sib' plants in hybrid

cauliflowers may be hybrid or self-inbred progeny. Proceeding 13 Australasian

Plant Breeding Conference. Christchurch, N e w Zealand (accepted).

Astarini IA, Plummer JA, Yan G, Lancaster R A (2006) Molecular markers correlated

with morphological traits in cauliflower seedlings. Australian Journal of

Agricultural Research (under review).

Astarini IA, Plummer JA, Yan G, Lancaster R A (2006) Identification of 'sib' plants in

hybrid cauliflowers using microsatellite markers. Theoretical and Applied

Genetics (under review).

Astarini IA, Plummer JA, Yan G, Lancaster R A (2006) Genetic diversity of Indonesia

cauliflowers and their relationship with Australian grown hybrid cultivars.

Scientia Horticulturae 108, 143-150.

Astarini IA, Plummer JA, Yan G, Lancaster R A (2005) Genetic diversity of open

pollinated cauliflower cultivars in Indonesia. Acta Horticulturae 694, 149-152.

Astarini IA, Plummer JA, Yan G, Lancaster R A (2004) Fingerprinting of cauliflower

cultivars using R A P D markers. Australian Journal of Agricultural Research 55,

117-124.

Conferences attended and visits

1. The 12th Australasian Plant Breeding Conference, 15-20 September 2002, Perth,

Western Australia

2. International/Australian Society of Horticultural Science (ISHS/AUSHS)

Conference: Harnessing the Potential of Horticulture in the Asian-Pacific

Region, 1-3 September 2004, Coolum, Queensland.

3. ComBio 2004,26-30 September 2004, Perth, Western Australia.

4. International Society of Horticultural Sciences (ISHS) Symposium on Brassicas,

24-28 October 2004, Daejeon, South Korea.

5. Visits to Seminis Vegetable Seeds and Enza Zaden, Enkhuizen, The

Netherlands, 1-3 November 2004.

vii

Awards

1. Young Scientist Award (2004). The best student presentation in ISHS/AUSHS

conference: Harnessing the Potential of Horticulture in the Asian-Pacific

Region, Coolum, Queensland.

2. Mary Janet Lindsay of Yanchep Travel Award (2004). Faculty of Natural and

Agricultural Science, The University of Western Australia.

viii

Table of Contents

Abstract i

Declaration iii

Dedication iv

Acknowledgments v

Thesis Outline vi

List of Publications, Conferences Attended and Awards vii Publications vii Conferences attended and visits vii Awards viii

Table of Contents ix

List of Tables xi

List of Figures xiii

Chapter 1 1

General Introduction 1 Background 2 Hypotheses 5 Objectives 6

Chapter 2 7

Literature Review 7 Cauliflower 8

Origin, distribution and taxonomy 8 Cultivars 9 Seedling production 12

Brassica breeding 14 Breeding objectives 14 Floral biology, pollination and seed production 14 Breeding systems 15 Genetic purity in Fi hybrid cultivars 20

Molecular Markers technology and its applications 22 Types ofDNA markers 23 Other markers, marker combinations 25 Application of molecular markers 25

Chapter 3 Error! Bookmark not defined.

Fingerprinting of Cauliflower Cultivars using R A P D Markers Error! Bookmark not defined. Abstract Error! Bookmark not defined. Introduction Error! Bookmark not defined. Materials and Methods Error! Bookmark not defined.

Results Error! Bookmark not defined. Discussion Error! Bookmark not defined. Acknowledgements Error! Bookmark not defined. References Error! Bookmark not defined.

ix


Genetic Diversity of Open Pollinated Cauliflower Cultivars in Indonesia Error! Bookmark not defined.

Abstract Error! Bookmark not defined. Introduction Error! Bookmark not defined. Material And Methods Error! Bookmark not defined. Results And Discussion Error! Bookmark not defined. Conclusions Error! Bookmark not defined. Acknowledgements Error! Bookmark not defined. Literature cited Error! Bookmark not defined.


Genetic Diversity of Indonesian Cauliflower Cultivars and Their Relationships with Hybrid Cultivars Grown in Australia Error! Bookmark not defined.

Abstract Error! Bookmark not defined. Introduction Error! Bookmark not defined. Materials and methods Error! Bookmark not defined. Results Error! Bookmark not defined. Discussion Error! Bookmark not defined. Acknowledgements Error! Bookmark not defined. References Error! Bookmark not defined.

Chapter 6 64

Identification of 'Sib' Plants in Hybrid Cauliflowers using Microsatellite Markers 64 Abstract 65 Introduction 66 Materials and methods 67 Results 70 Discussions 72 Acknowledgments 75 References 75

Chapter 7 80

Molecular Markers Correlated with Seedling Traits in Cauliflower Varieties 80

Abstract 81 Introduction 81 Materials and Methods 82

Results 84 Discussion 86 Acknowledgments 89 References 89

Chapter 8 98

General Discussion 98

Chapter 9 103

References 103

List of Tables

Chapter 2

Table 1. Types of genetic markers commonly generated using P C R method (Rafalski

and Tingey, 1993) 26

Chapter 3

Table 1. Decamer primers used in this study. Primers were synthesised by Life Technologies. Primers marked with an asterisk (*) generated polymorphisms

33

Table 2. Useful R A P D markers for the identification of 25 cauliflower cultivars. Polymorphic bands are described as Primer code followed by molecular weight indicated by number of base pairs. Bands were P=Present or A = absent 34

Table 3. Pairwise differences between 25 cultivars of cauliflowers. The data were calculated using Phylogenetic Analysis Using Parsimony based on R A P D bands from all the random primers tested 36

Chapter 5

Table 1. R A P D markers for the identification of 12 cauliflower cultivars. Bands are described as primer code followed by molecular weight. Bands were present (P) or absent (A) 59-61

Table 2. Pairwise distances between cauliflower cultivars. The data were calculated using Phylogenetic Analysis Using Parsimony based on R A P D bands from all primers tested 62-63

Chapter 6

Table 1. Number of plants grown (n), number of morphological sibs (nm) and the proportion, number of genetic sibs (ng) and the proportion, plant height, leaf number and curd weight of each cauliflower lines in Field Trial 2. Genetic sib was confirmed using microsatellites analysis, where only one female band was present. * Proportion of genetic sibs on normal plants was based on 20 plants tested, except for selfed SI, 10 plants tested. **Proportion of genetic sibs of sib plants = ng/nm, also expressed as a percentage. N = narrow, W = w a v y (Fig 2) 78

Chapter 7

Table 1. Cauliflower cultivars and their characteristics. SI=Self incompatibility breeding system, CMS=cytoplasmic male sterility, W = week. Plant characteristics were provided by Seed Companies 92

Table 2. Comparison of morphological characteristics observed at 7 days and 6 weeks. Values are mean ± s.e. It = length, Germ = germination, # = number, wt= weight, F W = fresh weight, D W = dry weight, d= day, w = week 93

Table 3. Correlation coefficient (r) of seedlings traits across cultivars. wt = weight, It = length, Germ = germination, # = number, F W = fresh weight, D W = dry weight, d=day, w = week 94

XI

Table 4. Correlation between seed weight with other seedling traits within cultivars. F W = fresh weight, Germ = germination, d=day, w = week 95

Table 5. Significance of markers associated with higher or lower values for seed weight, germination rate (Germ), shoot length (It), root length, total length and fresh weight (FW) at 7 days (d) for 21 cultivars. Values given are/?-values..96

Table 6. Significance of markers associated with higher or lower values for fresh weight (FW), dry weight ( D W ) , leaf number (#), shoot length (It) and root length at harvest after 6 weeks (w) for 21 cultivars. Values given are upvalues 96

xii

List of Figures

Chapter 2

Figure 1. Genomes of Brassica (U, 1935) 9

Figure 2. Comparison of breeding programs to produce hybrid vegetable and seed commodities. M F = male fertile, M S = male sterile, C M S = Cytoplasmic male sterility, Rf = nuclear restoration gene for C M S trait, - = absence of trait, + = presence of trait (Makaroff, 1995) 19

Chapter 3

Figure 1. R A P D amplification profiles obtained with primer UBC106. Standard bands are indicated by arrows. Molecular weight of standard are indicated in base pairs. M , 100 bp D N A ladder; C, Control lane; 1, Monarch; 2, Donner; 3, M3444; 4, Cauldron; 5, Gibralter; 6, Chaser; 7, CF535; 8, SPS716; 9, Liberty; 10, Omeo; 11, G389; 12, Advantage; 13, CLF33902; 14, CF536; 15, G376; 16, Virgin; 17, J3195; 18, Sirente; 19, Morpheus; 20, Fremont; 21, Plana; 22, Alabama; 23, SPS3074; 24, Celeste; 25, Discovery 33

Figure 2. Fingerprinting key for cauliflower cultivars generated from R A P D markers. Bold indicates that the cultivars has been differentiated 35

Figure 3. Dendogram of 25 cauliflower cultivars, constructed using P A U P based on Neighbor-Joining (NJ) analysis. The numbers at each node represent NJ coefficient of differences 37

Chapter 4

Figure 1. Dendogram of Indonesian cauliflower cultivars, constructed by unweighted pair-group method with arithmetic averages ( U P G M A ) based on total character differences. Numbers above branches represent branch length and numbers below branches indicate bootstrap values 43

Chapter 5

Figure 1. R A P D amplification profiles of 12 cultivars obtained with primer SL-01 and SL-08. Standard bands are indicated by arrows. M , marker ladder, 1, Harli; 2, Blaster; 3, Broad; 4, Manalagi, 5, Gembel; 6, Bandung; 7, Malang; 8, Bedugul; 9, Atlantis; 10, Omeo; 11, Monarch; 12, Plana; M , Marker ladder..55

Figure 2. Fingerprinting key for cauliflower cultivars. Bold indicates that the cultivar has been differentiated. Markers=primer-number of base pairs, P= Present, A=Absent 56

Figure 3. Dendogram of Indonesian cultivars ('Harli', 'Broad', 'Bandung', 'Gembel', 'Malang', 'Blaster', 'Manalagi', 'Bedugul'), Australian-bred cultivars ('Atlantis', 'Omeo') and European-bred cultivars ('Monarch', 'Plana'), constructed by unweighted pair-group method with arithmetic averages ( U P G M A ) based on total character differences. Numbers adjacent to cultivars indicate collection number. Numbers above branches represent branch length and numbers below branches indicated bootstrap values 57-58

xiii

Chapter 6

Figure 1. Reproductive organs of two types of CMS flowers compared to normal flowers (petals removed), a = shrunken anthers, b = petaloid anthers, c = normal anthers 79

Figure 2. Abnormal plant types (8-weeks old) observed in the field, a = wavy leaf (W), b = narrow leaf (N), c = blind apex, d = normal 79

Figure 3. Banding patterns of male parent ($), female parent (?), Fi hybrids (H), manual crosses (C) and self pollinated plants (S) using primer Nal2-E06b on SI system plants. M L = Marker Ladder. *= abnormal plant such that H*= abnormal hybrid, C*= abnormal crosses, S*= abnormal selfed 79

Chapter 7

Figure 1. a. Seedling size variation at day 7, from lightest to heaviest (left to right), b. Largest and smallest seedlings at 6 weeks, c. Biggest and smallest root mass at 6 weeks 97

Figure 2. R A P D profiles of 21 cauliflower cultivars using primer U B C 106. A, Markers for root length, total length at day 7, shoot length, fresh weight and dry weight at week 6. M , Marker Ladder; 1, Plana; 2, Donner; 3, Discovery; 4, Fremont; 5, Monarch; 6, CF0284; 7, Virgin; 8, Morpheus; 9, Cauldron; 10, Arctic; 11, Atlantis; 12, L3368; 13, Belot; 14, Lateman; 15, Phanter; 16, Jerez; 17, Fandango; 18, Megan; 19, Delfur; 20, SPS 716; 21, Omeo 97

xiv

Chapter 1

General Introduction

Background

Cauliflower (Brassica oleraceae var. botrytis, Family Brassicaceae) is a major

vegetable crop in Australia and around the world. World production of cauliflower was

16.4 million tonnes in 2004. The major cauliflower-producing countries are China, India,

Spain, Italy and U S A , and Australia produces about 128 thousand tonnes (FAOSTAT,

2005).

In Western Australia, cauliflower is the fourth most valuable vegetable crop grown

with the majority being exported to Asian countries. More than 8 0 % of Australia's export

cauliflower is grown and packed in the Manjimup district of Western Australia. The

industry is valued at about A U D $ 25 million per year (Lancaster and Burt, 2001; A B S ,

2005; A U S V E G , 2005). Western Australia has the advantages of a suitable climate and

soils, available water, advanced agronomic practices and excellent postharvest handling

facilities which make it an ideal place to supply a high quality produce to domestic and

international markets.

Western Australia has a strong cauliflower industry. Industry members aim for a

high level of customer satisfaction by offering desired varieties, quality and the ability to

consistently supply markets (Lancaster and Pasqual, 1999). However, high labour and

shipping costs have lead to increased economic pressure and competition from other

suppliers such as China and U S A (Mattingley, 2002). To compete in international markets,

all costs must be kept to a minimum and waste product must be minimized, yet around 1 0 %

of hybrid cauliflower crops grown are lost due to non-uniformity. Variation makes the

industry very labour intensive and reduces export income.

Improved uniformity is the first priority of breeding programs in Brassica around

the world (Monteiro and Lunn, 1999). Improving uniformity is a priority research area for

the cauliflower industry (Warren Cauliflower Group, 2003). A high degree of uniformity

in cauliflower quality is required for the domestic and export markets. Greater uniformity

would reduce cultivation and harvesting costs and improve crop quality. Distinctness,

uniformity and stability are criteria that new Fi hybrid varieties have to accomplish before

patent registration (Ruffio-Chable et al, 2000; Raparelli and Menesatti, 2000).

Variation between cauliflowers occurs at all stages in production and reduces

profitability at harvest. Cauliflower shows considerable variation in germination and

harvest date (Hadley and Pearson, 1998). Variation may be caused by genetic or

2

environmental differences or the interaction of these factors. Moreover the curd, which is

an immature inflorescence, is highly perishable and harvest must occur at a specific stage

(Wurr, 1990). Several harvests are commonly required due to plant to plant variation in

maturity and non-uniform curd development (Rubatzky and Yamaguchi, 1996).

The uneven and unpredictable curd development periods of cauliflower are a

problem all over the world (Fujime and Okuda, 1996; Raparelli and Menesatti, 2000;

Ruffio-Chable et al., 2000). In Western Australia, many selective harvests are required at

two to three day intervals to obtain each curd at the optimum quality. Consequently all

picking is done by hand and may extend for up to 30 days. This is very labour intensive as

leaves are folded over cauliflower heads to prevent sun damage and these must be removed

and replaced at each inspection to determine head maturity. M a n y curds are left in the field

as they mature outside the profitable picking time range (Shellabear, 1994). This variation

also prevents the use of mechanical harvesting systems. A grower's profit is reduced first

by the unpicked curds and secondly by the high labour costs required for several sweeps of

checking, covering and harvesting the curds. Less variable plants would improve harvest

uniformity and reduce losses from early and late maturity dates where harvestable numbers

are small and uneconomic. A uniform cauliflower field would also make grading much

easier and cheaper.

Several approaches have been taken to address the existing variation, such as

agronomic management (improved approaches to nutrition, irrigation) and physiological

approaches (vernalisation, temperature and hormonal control) with limited success (Stirling

and Lancaster, 2005; Charsley, 1998). The alternative approach, examining genetic aspects

requires further attention. This approach will benefit and complement existing

management options.

Variation begins at the seed and seedling stage and these differences in growth are

exacerbated during field production. Elimination of variation at the seed or seedling stage

would greatly enhance uniformity of the crop. Identification of specific molecular markers

associated with seedling traits would assist in the selection of superior, uniform seedlings.

Associations between seedling characters and genetic data exist in radish (Raphanus

sativus) (Pradhan et al., 2004b). Molecular markers linked with particular traits are

available for Arabidopsis and the related Brassica oilseeds (Allonso-Blanco et ah, 1999;

Barret et al., 1998; Somers et al., 2001). Associations and molecular markers linked with

3

cauliflower seedling traits would be of considerable value in cauliflower and will be

investigated.

Cauliflower identification is important for accurate classification of cultivars.

Intensive breeding in cauliflower has resulted in difficulties and irregularities in their

classification and cultivar identification. Problems with plant classification based on

morphological characters occur in cauliflower and broccoli (Malatesta and Davey, 1996).

Identification based on morphological characters is also time consuming and requires

expensive field trials and evaluation. Molecular ( D N A ) markers have an advantage over

other tests for cultivar identification, in that D N A is not affected by environmental factors

or the developmental stage of the plant. Molecular markers can be used to study

cauliflower diversity and to develop fingerprinting keys for selected cultivars.

The use of Fi hybrids, which can produce a genetically uniform population is still

developing. T w o breeding systems for hybrid production are currently employed, self

incompatibility and cytoplasmic male sterility. However, hybrid varieties usually contain

up to 2 0 % 'sib' plants, thought to be from self-inbred seed (Holland and McNeilly, 1985;

Crockett et al, 2000; Crockett et al., 2002). Sibs can be smaller and have darker green leaf

colour than the hybrid, be taller and have paler green leaf colour, or have much weaker

growth habit than the hybrid. Sib plants usually flower very early and produce a very small

curd which is not suitable for export (Lancaster and Burt, 2001). Registration of new

varieties allows a m a x i m u m of 5 % sibs, and seed company breeders need to meet this

criteria (Ruffio-Chable et al, 2000; Harvey and Smith, 1987).

There are several possible causes of sibs which are assumed to be non-hybrid plants

(Holland and M c Neilly, 1985). This includes pollinating vector behaviour, the presence of

S-allele modifying genes, inappropriately matched parents and the effect of temperature

and humidity within the crop. Brassica oleraceae possesses a single locus, multi-allelic,

sporophytic incompatibility system. Breeder's lines are usually maintained by bud

pollination. All incompatibility alleles are not equally effective and varying amounts of

self-fertilization m a y occur in inbred lines homozygous for S-alleles. In addition, the

incompatibility reaction m a y be weakened by environmental factors such as high

temperature. The ratio of selfing to crossing m a y be affected by the behavior of pollinating

insects and the availability of foreign pollen (Wills et al, 1980). These factors mean that

methods to screen for sibs or potential sibs would be desirable.

4

Aberrant plants, which are also unsuitable for harvest, are often observed in

cauliflower production utilizing the Cytoplasmic Male Sterility ( C M S ) breeding system.

Phenotypes of aberrant plants predominantly involve the modification of curd size, leaf

shape, size and thickness (Ruffio-Chable et al, 2000). Occurrence is more frequent at the

end of the vegetative cycle. Aberrations occur in up to 4 0 % of plants in cauliflower

varieties. Again a method to screen for potential aberrant plants would be desirable.

To address the existing problems, knowledge of molecular genetics in cauliflower is

required. M a n y molecular techniques can be used to define identity, purity and stability of

a given plant. The range of techniques available for immediate application varies, with

major species offering a wide range of established molecular markers and minor or less

well studied species requiring evaluation of potential marker systems. The choice of

method for any particular application will depend upon the difficulty of the distinction to be

made and factors such as time, facilities and funds available. The nature of the sample may

determine the techniques to be used (Henry, 1997). The techniques currently available

include R A P D (Random Amplified Polymorphic D N A ) , R F L P (Restriction Fragment

Length Polymorphism), A F L P (Amplified Fragment Length Polymorphism), S S R (Simple

Sequence Repeats), R N A (Ribonucleic Acid) and protein analysis. Scientists have used

R A P D markers to investigate seed purity of hybrid cabbage and broccoli (Crockett et al,

2000; 2002), which are the same species as cauliflower. It would be very useful to apply

molecular techniques to seed purity studies in cauliflower.

Hypotheses

The overall hypothesis of this thesis is that variation in cauliflower production is

due to genetic and environmental factors. Here the genetic factors were investigated. This

can be divided into four specific hypotheses:

1. Hybrid cultivars have a narrow or limited genetic base resulting in a very similar

morphological appearance among cultivars. Cultivar identification should be based

on genetic make up instead of morphology. Variation can be detected on a genetic

level, which is independent of the environment.

2. Different breeding systems result in different degrees of variation in their progeny.

Variation within open pollinated cultivars is greater than control pollinated

cultivars. Greater variation occurs between distantly related cultivars.

5

3. The occurrence of sibs in hybrid seed is mainly due to genetic factors, i.e. self

pollination of the female parent. Development of molecular markers for

identification of sib plants will allow selection for more uniform crops.

4. Morphological variation at the seedling stage is in part caused by genetic variation

and genes controlling morphological traits can be identified using molecular

markers.

Objectives

These hypotheses reflect four specific aims of this study:

1. T o fingerprint the c o m m o n cauliflower cultivars using molecular markers (Chapter

3,4).

2. To study diversity/variation in open pollinated cultivars and genetic distance

between Australian grown and Indonesian grown cultivars (Chapter 4, 5).

3. To develop markers for the identification of sib and normal plants (Chapter 6).

4. To search for molecular markers associated with genes controlling morphological

traits related to plant growth and development (Chapter 7).

6

Chapter 2

Literature Review

Cauliflower

Origin, distribution and taxonomy

Brassica contains around 40 species. This genus has great commercial value and

plays a major role in feeding the world population. Brassicas include nutritious vegetables,

mustards and oil seeds, animal feed, cover crops and weeds (Rubatzky and Yamaguchi,

1996).

Six of the most important Brassica species are closely interrelated. B. rapalB.

campestris (turnip, Chinese cabbage), Brassica nigra (black mustard) and B. oleracea

(cabbage, cauliflower, broccoli, kale, kohlrabi, Brussel sprouts) are monogenomic. Their

genomic composition has been labelled as A, B and C. The other three are amphidiploid

species, B. juncea (leaf mustard), B. napus (rutabaga, oil rape) and B. carinata (Abyssinian

mustard) identified as A B , A C and B C respectively (Rubatzky and Yamaguchi, 1996).

According to the widely accepted scheme known as the U-triangle (Figure 1), the three

Brassica amphidiploids have been derived from interspecific hybridisation between the

three diploid species (Song et al, 1996).

Large genetic diversity exists among and within the three cultivated amphidiploid

species. Based on genetic diversity, B. napus seems to be the most ancient amphidiploid,

followed by B. juncea and B. carinata. T w o major factors are responsible for the genetic

diversity within amphidiploids, one is multiple hybridizations between different diploid

parents and the other is genome modification after polyploidization.

Cauliflower appears to have been domesticated in the Mediterranean region. The

first written description of cauliflower appeared in 1544 and Italy is widely regarded as the

centre of diversity of cultivated B. oleracea (Rubatzky and Yamaguchi, 1996; Massie et al,

1996; Wien and Wurr, 1997). The simple cultivated variety of the 16th Century possessed a

small, quick-bolting inflorescence. Today, specially adapted varieties and spatial and

temporal distribution of production allow year-round supply (Sauer, 1993).

8

B. oleracea, n = 9

Cabbage, Cauliflower, Broccoli

CC

n = 9 n = 9

BBCC

B. carinata, 2n = 34

Abyssinian

mustard

AACC

B. napus, 2n = 38

.Canola

BB

B. nigra,

2n=16

Black mustard

AABB

B. juncea,

2n = 36

Leaf mustard

Figure 1. Genomes of Brassica (U, 1935).

AA

B. campestris

2n = 20

Chinese cabbage

Cauliflower {Brassica oleracea var botrytis) is one of the most popular Brassica

vegetables. It is cultivated worldwide in different climatic conditions, ranging from

temperate to tropical regions and is available year round in the market. Major producers of

cauliflower are India, China, France, Italy, United Kingdom, USA, Spain, Poland,

Germany, Pakistan (Sharma et al, 2004) and Australia (Lancaster and Burt, 2001). They

are also cultivated in tropical zones of Africa, Central and South America, and Oceania

(Sauer, 1993).

Cultivars

A range of cauliflower cultivars has been developed through breeding and selection,

to produce satisfactory yields in specific environments ranging from tropics to temperate

areas (Wien and Wurr, 1997). Proper variety selection is crucial for cauliflower

production. Cultivars have biological clocks which trigger the curd to develop at a specific

9

time based on plant age and past and present ambient temperature. Depending on the

cultivar, the period of vegetative growth may be only a couple of weeks or more than a

month. Cultivars grown out of their appropriate season will not develop satisfactorily.

C o m m o n defects include bolting (early flowering), riceyness, yellowing, light weight curds

and breaking-apart of the florets. These occur due to inappropriate planting period for a

given variety, curds exposed to sunlight or when the crop grows during adverse weather

(Mayberry, 2000).

Important cultivar variables are curd weight, size, shape, compactness, surface

texture and colour. Traditionally pure white curds are preferred, although cultivars

producing cream, purple, green and orange curds are also grown (Rubatzky and

Yamaguchi, 1996). In Australia, curd size for export must be heavier than 0.5 kg, with the

best size being approximately 1 kg - 1.2 kg. In addition, curds with the following defects

are discarded: leaves in the curds, surface dirt, bruising, severe shape distortion, excessive

furriness or 'riceyness', unevenness, yellow, brown or pink discolouration and insect and

disease damage (Lancaster and Burt, 2001).

Cauliflower cultivars can be grouped into three major maturity types: early (for

summer and early autumn harvest), intermediate (late autumn and early winter) and late

(winter and spring). Late-maturing types require vernalization for curd initiation. Some of

the major cultivar include: 1. Italian cultivars of varied curd form and colour, grown as

annuals and biennials, for example Jezi, Romanesco, Flora and Blanca, 2. Northern

European cultivars, grown as annuals during summer and autumn, for example Alpha and

Snowball, 3. Northwestern European cultivars grown as biennials for late winter and spring

harvesting, for example Roscoff and St Malo, 4. Australian cultivars grown as annuals

using cultivars mostly developed from European sources, 5. Asian cultivars grown as

annuals and adapted for high temperature regions, often lacking in uniformity, curd

compactness and colour, for example Patna (Rubatzky and Yamaguchi, 1996).

Most cauliflowers available in the Australian market are Fi hybrid cultivars. These

varieties result from intensive breeding programs conducted by seed companies throughout

the world. Seed companies in developed countries such as those in Europe and Northern

America often exchange their breeding materials. Consequently, most hybrid varieties are

very closely related and the same variety may be released under different names.

10

Since hybrid varieties are very closely related and difficult to distinguish

morphologically, there is a need for more accurate identification of the varieties released to

the market. Identification based on morphological characters is time consuming and

requires expensive field trials and evaluation. Methods using molecular markers may

provide an accurate fingerprinting method as they are based on genetic information and are

not affected by environmental factors (Henry, 1997).

Most cauliflower cultivars available in Indonesia are open pollinated. There is little

information about cultivars or landraces in Indonesia. Cultivars may have been introduced

from India during the Dutch period in Indonesia more than a hundred years ago. These

cultivars became locally adapted. They are grown in highlands throughout the country,

mainly for domestic consumption.

Before introducing any vegetable species to the tropics, it is necessary to identify

the optimal growth temperatures and photoperiodic requirements to ensure they are

compatible with local climatic conditions. It is also important to utilise existing local

varieties of the species to be introduced, even though the quality of the produce m a y not be

high. This is because such traditional, local varieties m a y have an important role to play in

future plant improvement schemes since they represent an irreplaceable source of genetic

variability (Messiaen, 1992). More importantly, local varieties often have a good level of

resistance to pests and diseases (William et al. 1991).

The lack of adequate taxonomy has seriously affected the systematic collection and

assessment of cauliflower genetic resources (cultivars). This has had two damaging effects.

Firstly, many genotypes may have become extinct because the range of variation in

cauliflower was unknown to genetic conservationists. Secondly, the genetic variation of

cauliflower and its closest relatives has not been exploited by breeders, instead a

disproportionate effort m a y have been spent making wide crosses within B. oleracea or

even with other species, in order to introduce desirable traits into cauliflower. For example,

attempts to backcross resistance to the clubroot disease into cauliflower from cabbage were

unsuccessful because of the difficulty in regaining an acceptable cauliflower phenotype

(Sharma et al, 2004). There is a need for the adoption of molecular techniques for cultivar

identification and their application in selection of superior characters and individuals for

future breeding programs.

11

Seedling production

Cauliflower is commercially propagated from seeds usually by a specialist nursery.

Seeds are commonly sown in individual cell trays. Growing conditions are closely

controlled to ensure morphologically uniform seedlings are produced. Seedlings are ready

for transplanting into the field when they have three to four true leaves and can be

transplanted until they are 7 or 8 weeks old. Planting older seedlings increases the

likelihood of premature curd production (Madhavi and Gosh, 1998). Weekly to fortnightly

plantings are usually made to ensure continuity of supply throughout the season. Specialist

nurseries provide good quality plants, reduce risk of soil diseases, reduce transplanting

shock and variation in seedling size leading to greater uniformity at maturity (Lancaster and

Burt, 2001).

Seed and seedling variation

It is important to begin with high quality seed that will produce a uniform stand of

vigorous seedlings. Seedling uniformity is critical since variation in vigour results in

shading of small plants and slower plant growth (Webster, 1964). Variation begins at the

seed and seedling stage and differences in growth are exagerated during field production.

Minimising or elimination of variation at the seed or seedling stage would greatly enhance

uniformity of the crop.

Seed size variation influences early seedling performance and subsequent adult

growth (Bretagnolle et al, 1995). Seed size m a y vary within species, among populations,

within populations, individuals and within fruits in an individual plant. In Alliariapetiolata

(Brassicaceae), individual seed weight varies 2.5-fold to 7.5-fold within populations and

nearly eightfold among populations (Susko and Lovett-Doust, 2000).

Variation in seed weight is caused by environmental, maternal or genetic factors.

Environmental effects include differences in temperature, light, water and nutrient levels

(Gutterman, 2000). Position of seeds in the mother plants affects seed weight. Within an

infructescence of fruit, individual seed mass decreases from basal fruits to distal fruits.

Furthermore, seed mass decreases within fruits from basal to distal seed positions (Susko

and Lovett-Doust, 2000). Basal fruits within an infructescence and basal seeds within fruits

may behave as strong sinks for limited parental resources, such as nutrients and

photosynthate. Thus, competition for limited resources m a y influence the maturity of a

seed, as well as the mass of the seed. Early-initiated, basal fruits produce larger seeds than

12

fruits in the middle or the tip of an infructescence. Cauliflower plants produce many

inflorescences in a plant with many siliques (pods) within an inflorescence. Approximately

16 seeds are produced in a silique. Variation in seed weight may occur within a silique and

within an inflorescence. The detected differences may be due to genetic and/or

environmental factors (Susko and Lovett-Doust, 2000).

Genetic factors are also involved in seed size though to date these are poorly

understood. At least 11 seed size and seed weight quantitative trait loci contribute to seed

size variation by affecting both cell number and cell size in Arabidopsis (Alonso-Blanco et

al, 1999). The gene AP2 (APETALA2) plays an important role in the control of seed mass

and seed yield, by affecting seed size, embryo size, seed weight and the accumulation of

seed reserves. AP2 acts in the maternal sporophyte and endosperm perhaps by influencing

source-sink relations, and it is required for normal seed coat development (Jofuku et al,

2005; Ohto et al, 2005). In tomato, Sw4.1 was described as a major Q T L for seed weight

variation (Orsi and Tanksley, 2005). Genes or QTLs for large seed size are unknown in B.

oleracea. The advantages associated with larger seeds and the potential for increasing yield

through seed size indicate the importance of identifying the genes involved in the

determination of seed size and seed mass. Three ISSR markers are found linked with low

seed weight in wheat (Ammiraju et al, 2001), 13 SSR markers are associated with seed

size in soybean (Hoeck et al, 2003), 11 R L F P loci for mungbean (Humpry et al, 2005) and

4 Q T L for seed weight have been identified based on R A P D , ISSR and phenotypic markers

in chickpea, (Cho et al, 2002). Association of seedling traits and molecular markers in

cauliflower were investigated.

Effect of seed weight variation on seedling traits

Seed weight can greatly affect seedling traits including germination (Van Molken et

al, 2005) and seedling size (Schaal, 1980). Germination responses depend on the species

(Baskin and Baskin, 1998) and rate and percentage germination can increase, decrease or

remain unaffected by differences in seed size. In Alliaria petiolata, smaller seeds

germinate earlier but larger seedlings produce higher total plant biomass (Susko and

Lovett-Doust, 2000). In Cakile edentula (Zhang, 1993) and Erodium brachycarpum

(Stamp, 1990), small seeds germinate earlier than large seeds. This may be because small

seeds have greater access to water as a result of their higher surface to volume ratios.

Hence, small seeds imbibe water faster and germinate sooner. In Arabidopsis, the RGL2

13

gene is responsible for seed germination and it probably functions as an integrator of

environment and endogenous cues to control seed germination (Lee et al, 2002).

Seed size and seed weight are important determinants in seedling dry matter and

seedling leaf area in Cercis canadensis (Couvillon, 2002). Larger seeds produce larger

seedlings, with greater fresh and dry weight and leaf area than small seeds. Small seed size

may influence other aspects early seedling growth and establishment. In Pastinaca sative

seedlings, the maximum ratio of root length to total leaf area is negatively related to seed

weight at 10 and 20 days after emergence (Hendrix et al, 1991).

In a number of Brassica oleracea species, differences in seed vigour contribute to

differences in seed germination and seedling variability. Seeds with high vigour usually

germinate fast and produce more uniform seedlings. Seeds with low vigour, as a result of

ageing, germinate more slowly and produce smaller and more variable seedlings (Powell et

al, 1991). Genes or Q T L s for seed and seedlings traits are unknown in Brassica oleracea.

However, in other Brassica, R A P D markers have been linked to seed coat colour in

Brassica napus and B. rapa-alboglabra (Somers et al, 2001; Heneen and Jorgensen, 2001).

In conclusion, genetic and environmental factors contribute to seed and seedling

variation in plants. Identification of specific molecular markers associated with seedling

traits such as seed weight, germination rate, fresh weight, dry weight, shoot length and root

length, would greatly assist in the selection of superior, uniform seedlings.

Brassica breeding

Breeding objectives

Breeding objectives can be addressed to satisfy both growers and consumers. These

need be considered in terms of crop production and product improvement. The main

criteria for crop production are yield, resistance to disease or environmental stress,

uniformity and continuity of cropping. Breeding for appearance, commercial quality, shelf

life, taste and nutritional value is part of product improvement. The most important

objective is crop uniformity which makes grading much easier and reduces harvest time

(Monteiro and Lunn, 1999).

Floral biology, pollination and seed production

Cauliflower flowers are typical of the Brassicaceae family, with four sepals, four

symmetrical yellow petals and six stamens (2 short and 4 long). Anthers rarely dehisce

14

before flower opening, even though the stigma may be level with the anthers within the

flower bud (Crisp and Tapsell, 1993; Sharma et al, 2004).

Cauliflower is a cross-pollinated crop, mainly pollinated by insects. Honeybees are

the usual pollinating agents, although bumble bees and flies may also be responsible for

pollination. Wind can also be the pollinating agent. The stigma of Brassica is receptive 5

days before and 4 days after flower opening. The period from pollination to fertilization

generally takes 24 - 48 hours, depending on temperature, with the ideal temperature being

12-18°C. Higher day temperatures cause pollen sterility, resulting in poor seed

development. Pod maturity for harvest of pods may require 50-90 days from the date of

flowering, depending on climatic conditions. The fruit is a siliqua but often called a pod.

The seeds are small, globular, smooth and dark brown. There are normally 1 2 - 2 0 seeds

per pod and nearly 350 seeds weigh one gram (Sharma et al, 2004).

Breeding systems

Controlled pollination is essential for production of hybrid seed. Plants grown from

hybrid seeds benefit from the heterotic effect of crossing two genetically distinct breeding

lines. Heterosis or hybrid vigour is the increased vigour of plants when compared with

parents. The agronomic performance of the hybrid progeny is superior to both parents in

terms of yield, vigour, adaptability and uniformity. In order to produce hybrid seed

uncontaminated with self-pollinated seed, control methods need to be developed to stop self

pollination (Bhalla and Singh, 1999). The two major methods applied for the production of

Fi hybrids in Brassica are self-incompatibility (SI) and cytoplasmic male sterility (CMS).

Open pollination is still employed, particularly in developing countries (Williams et al,

1991).

Self-incompatibility system in Brassica

Self-incompatibility (SI) is a natural mechanism that prevents self-fertilization and

promotes outcrossing and m a x i m u m recombination in Angiosperms (Watanabe and Hinata,

1999). Brassica oleracea posseses a single locus, multi-allelic, sporophytic incompatibility

system. Phenotype of the pollen is determined by the diploid genotype of the pollen

producer, the sporophyte. Pollen rejection occurs when both the pollen and the pistil

exhibit the same S-phenotype, although they m a y have different genotypes (Bateman,

1955). All incompatibility alleles are not equally effective and varying amounts of self-

15

fertilization may occur in inbred lines homozygous for S-alleles. Not all species naturally

possess SI (McCubbin and Dickinson, 1997). In addition, the incompatibility reaction may

be weakened by environmental factors such as high temperature and high humidity (Zur et

al, 2003). A weakening in self-incompatibility is likely to increase selfing and selfed seed,

especially during the non-coincident phases of flowering (Gowers, 2000).

Developing self-incompatible breeding lines for hybrid seed production is costly

since the stabilisation of inbred parental lines requires several generations of selfing, and

the maintenance of breeding lines is labour intensive. In order to self SI plants, the

mechanism needs to be overcome or avoided. The incompatibility system becomes

operative two or three days before anthesis, so self-incompatible plants can be self-

pollinated by opening immature buds and placing pollen on the exposed stigma (Crisp and

Tapsell, 1993). Alternatively the stigma can be removed and pollen placed on the cut

surface of the style in bud stage. This avoids the SI mechanism which is located in the

stigma. Bud pollination is commonly used to overcome self-incompatibility in cauliflower

(Hallidri and Pertena, 2002). Sodium chloride solution and carbon dioxide can break down

the incompatibility mechanism in Brassica rapa (Mohring et al, 1999).

SI is widely used in the production of Fi hybrids in vegetable Brassica oleracea

(Sharma et al, 2004). All of the new hybrid Brussels sprouts, cabbages and kales owe their

origins to breeding programs employing SI. Unfortunately, there is potential for

breakdown of self-incompatibility due to adverse environmental factors in the hybrid seed

production field resulting in contamination of hybrid seed with selfed seed, commonly

known as sibs (Bhalla and Singh, 1999).

Cytoplasmic Male Sterility (CMS)

Cytoplasmic male sterility ( C M S ) is a convenient method for the production of

hybrid seeds. It is a more advanced and reliable system than SI. C M S occurs in a wide

variety of higher plants and is characterized by a very low level or complete absence of

pollen production. C M S is caused by different factors linked to mitochondria and it is

assumed to be a consequence of mitochondrial dysfunction (Araya et al, 1998). In C M S ,

sterility is carried out by the cytoplasm and therefore through the maternal line.

C M S systems have been characterized by the restorer genes required to overcome

them and to provide male-fertile progeny in the male-sterile cytoplasm. Female fertility is

generally not affected by C M S , so that male-sterile plants can set seed if viable pollen is

16

provided (McVetty, 1997). The affected organs and tissues in C M S plants are the stamens

(anther and filament) and pollen grain (microspores). Abnormal behaviour of the tapetum

in the anthers is frequently identified with C M S (McVetty, 1997).

There are several ways to create C M S . C M S can arise spontaneously in breeding

lines, following mutagenesis, as a result of wide crosses or through interspecific exchange

of nuclear and cytoplasmic genomes (Schnable and Wise, 1998).

Types of CMS

At present, there are several C M S systems (Schnable and Wise, 1998; Makaroff,

1995). These include:

1. "Pol" (Polima). This system was identified in China, in Polish Brassica napus cv Polima.

This C M S system is relatively but not completely temperature stable. The availability of

maintainer and restorer lines of the pol cytoplasm, in addition to the relative temperature

stability of the male-sterile phenotype, has made the pol cytoplasm one of the most

advantageous C M S systems for the production of hybrid rapeseed.

2. Ctr (Bronowsky). This is the most recent system found in B. napus. It arose in the F2

generation of triazine resistant lines Tower x Bronowski. This system is temperature

unstable.

3. Nap (SHIGA). First discovered in B. napus cytoplasm following crossing of two Japanese

varieties, Hokuriku x Isuzu. This system is unstable at high temperatures (26 - 30°C),

when it reverts to fertility and therefore it is unsuitable for most field locations.

4. Nigra. This system has been transferred to broccoli and cauliflower and other B. oleracea

vegetables and rapeseed. It has very stable sterility in B. oleracea but not in B. napus. It

also has problems with seed set. The anthers develop as petals (petaloid sterility).

5. Tour (tournefotii). This C M S system resulted from spontaneous interspecific hybridisation

between B. tournefortii and B. juncea.

6. Mur. This system was found after transferring B. napus nuclei into Diplotaxis muralis

cytoplasm. It has been transferred to turnip (B. rapa) and rapeseed. The sterility is

complete and stable.

7. Ogu. This C M S was found by Ogura in a Japanese radish variety. Fi hybrid seed

production in B. oleracea has been achieved using this system.

17

C M S has not been found in cauliflower but it has been introduced from several

sources. Commercial Fi hybrid production has been achieved in B. oleracea using the

improved "Ogura" cytoplasm obtained by Pelletier et al. (1989). F, hybrids of various B.

oleracea types (cauliflower, sauerkraut cabbage, garden cabbage, savoy cabbage) have

been registered (Leviel, 1998).

Morphological changes associated with male sterility

Achievement of an effective C M S system has often been impeded by difficulties

such as the instability of the male sterility, the absence of maintainer or restorer lines,

chlorophyll deficiencies and deformed flowers which lack nectarines (Delourme and Budar,

1999). In some forms of C M S , the absence of functional nectarines prevents commercial

production of Fi hybrid seed using normal insect pollinators.

In the Ogura system, low temperature induces chlorosis in the seedling stage, which

is expressed in the field grown Fi commercial crop as a loss of vigour. The Ogura source

has also been associated with low seed set and poor curd quality in Fi cauliflower. In B.

napus and B. oleracea, which use the Ogura system, chlorophyll deficiency is corrected by

obtaining hybrids via protoplast fusion (Delourme and Budar, 1999). In the nap, pol and

tour C M S , flowers are characterized by narrow petals but this has not been found or

reported in Brassica oleracea, C M S system.

The stability of male sterility is largely dependent on the environment or maintainer

lines. Fertile pollen grains may be produced at temperatures higher than 25 - 30°C in nap

and polima systems or at low temperatures depending on the maintainer lines in the polima

system (Delourme and Budar, 1999). In the field this could lead to selfing and production

of self-inbred plants, which will contaminate hybrid seed production. Temperature effect

on particular C M S system requires further investigation.

CMS for production of commercial hybrids seed in Brassica

In developing a C M S system for breeding purposes, it is important to consider the

crop. Breeding programs for the generation of Fi hybrids of a seed commodity, such as

canola, are different from those that are used to develop vegetable hybrids like broccoli,

cabbage or cauliflower. Production of Fi hybrid seed for the seed commodity requires the

development of both maintainer lines, which maintain the C M S trait and male fertility

restorer lines, which contain nuclear genes that suppress the C M S trait and result in male-

18

fertile plants. Production of Fi hybrid vegetable seed only requires the development of

maintainer lines because vegetable crops are grown for the plant and not the seed

(Makaroff, 1995). A simplified breeding program for the two different crop types is

outlined in Figure 2.

Vegetable commodity Seed commodity

1. C M S line x maintainer line

(MS.CMS+, Rf-) (MF:CMS-, Rf-)

1 hybrid seed sold to farmers

(MS:CMS+, Rf-)

1. C M S line x maintainer line

(MS:CMS+, Rf-) (MF:CMS-, Rf-)

1 2. C M S line x restorer line

(MS:CMS+, Rf-) (MF:CMS+/-, Rf+)

i hybrid seed sold to farmers

(MF:CMS+, Rf+)

Figure 2. Comparison of breeding programs to produce hybrid vegetable and seed commodities. M F = male fertile, M S = male sterile, C M S = Cytoplasmic male sterility, Rf = nuclear restoration gene for C M S trait, - = absence of trait, + = presence of trait

(Makaroff, 1995).

Use of CMS lines is gaining more commercial status, since there should be no risk

of selfed seed. The male sterile lines in commercial Fi production are planted with the

pollen parent in the ratio of 2:1 or 3:1 or 4:1 depending upon varietal characters. In

selecting the pollen parent, factors that need to be considered besides good combining

ability are similarity in morphological characters including plant height and synchrony of

flowering with the male sterile plant. Immediately after pod setting, the pollen parent is

removed to avoid mixture and provide sufficient space for the female plant to produce

hybrid seed (Sharma et al, 2004; Frankel and Galun, 1977).

Seed production in open-pollinated cultivars

Several techniques are used to produce basic seeds from selected plants of open-

pollinated cultivars. In all cases the mother plants are grown in their normal season and

19

final selections are made after confirmation of plant characters and curd quality. W h e n

environmental conditions in the field are expected to remain favourable for further flower

development, anthesis and seed maturity, the selected mother plants can be left in situ. This

is the normal practice in northern Europe and North America for early summer

cauliflowers. It is also a c o m m o n method in Asian countries (Sharma et al, 2004).

Genetic purity in Fj hybrid cultivars

High uniformity has been almost impossible to achieve with open-pollinated

varieties owing to the cross-pollinating nature of brassicas. The introduction of Fi hybrids

is a major advance. Fi hybrid cultivars are the result of crossing two inbred lines which

have been maintained under the control of plant breeders and are known to produce a

desirable hybrid. The advantages of Fi hybrid cultivars include uniformity, increased

vigour, earliness, high yield and resistance to specific pests and pathogens (Crisp and

Tapsell, 1993).

In theory, all plants in an Fi hybrid cultivar resemble each other exactly. Fi hybrid

seed production necessitates the use of a hybridization control system. In Brassica

oleracea such systems exploit either self-incompatibility (SI) or cytoplasmic male sterility

( C M S ) (Ruffio-Chable et al, 1993). With the SI system, due to some self-pollination of

the female parent used in the cross, some plants which are not Fi hybrids may occur and

they are usually morphologically different. These off-types in an Fi hybrid are assumed to

be the result of accidental self-pollination of the female parent and are generally known as

'sibs' (Hodgkin, 1981).

Sibs are a worldwide problem in vegetable Brassica. The key characteristics

associated with sibs are not identifiable in the seedlings stage and therefore they are planted

in the field. Sibs have weak plant habit and produce small, unmarketable curds (Holland

and McNeilly 1985; Lancaster and Burt, 2001). The limit for off type plants in registration

of new varieties of cauliflower is 5 % (Ruffio-Chable et al, 2000). This has been difficult

to meet with some desirable selections. Seed companies cannot afford to market hybrid

seed with appreciable amounts of sib seeds (Hodgkin, 1981).

In addition to the problems associated with sibs in an Fi hybrid seed lot, there are

increased production costs compared to open-pollinated cultivars. The development of the

initial breeding program, subsequent maintenance of the inbred parents, extra land required

for male parents, care with sowing, isolation and harvesting, high labour input for manual

20

emasculation of female flowers, lower seed yield per unit of land and high cost of seeds for

farmers all add to the cost of Fi hybrid seed production (George, 1999).

It is assumed that it is possible to identify sibs by their distinctive plant phenotype

and these will be here referred to as 'morphological sibs'. Also it is assumed that sibs are

self inbred, therefore are genetically determined and could be identified by genetic markers.

Here plants from a self inbred parent will be referred to as 'genetic sibs'.

CMS-based seed parent lines do not produce pollen and thus do not risk self-

pollination (Bhalla and Singh, 1999). Fi cauliflower hybrid derived from C M S system

often produce developmentally aberrant plants, which are unsuitable for harvest.

Phenotypes of aberrant plants mainly involve the modification of three characters: leaf

shape, size and thickness (Ruffio-Chable et al, 2000; Fujime and Okuda, 1996). The

proportion of aberrant plants ranges from 5 % to 4 0 % in cauliflower production.

The high cost associated with hybrid seed production and losses during harvest has

lead many studies to determine sib or other types of aberrant in the Fi harvest. M u c h

research has gone into methods to identify sibs but a definitive method is still required for

cauliflower and other brassica vegetables. For example, image analysis (Fitzgerald et al,

1997), high-pressured liquid chromatography (HPLC) (Mennella et al, 1996), isozyme

analysis (Harvey and Smith, 1987; Zheng and Liu, 1994) and analysis of ploidy levels by

flow cytometry (Ruffio-Chable et al, 2000). The basic features of the image analyses

system include image capture and digitalization, image processing into pixel intensity

numbers and image analyses using mathematical tools to sort and compare images.

Accuracy of this technique is limited by genetic relatedness of cultivars and heterogeneity

of characters within cultivars (Cooke, 1999).

Seed or cotyledon extracts from a seed lot are analysed for isozymes of acid

phosphatase by P A G E of Fi hybrid Brussels sprout varieties (Harvey and Smith, 1987).

Seeds that show homozygous alleles are regarded as sibs. The level of sib content of bulk

seed lots can be above 10%. There are some limitations to isozyme analysis, such as

insufficient polymorphisms among closely-related genotypes and variations affected by

environmental factors, seed vigour and growing stage (Meng et al, 1998).

Molecular marker techniques such as R A P D and R F L P have also been employed in

other brassica vegetables. There are up to 1 4 % of morphological sibs in cabbage (Crockett

et al. 2000) and up to 4 5 % morphological sibs in broccoli observed in the field (Crockett et

al. 2002). A similar proportion of genetic sibs were identified using R A P D technique on

21

seed lots suggested the genetic sibs would have been morphological sibs and therefore

causal, but this has yet to be proved. Thus, there is a need to develop a purity test for

cauliflower and other brassicas to prevent unacceptable levels of sib and aberrant seed

being released in the market.

Improving crop purity standards (environmental factors)

Members of the Brassica oleracea group, i.e. cabbage, cauliflower, Brussels

sprouts, broccoli, kale and other wild allies freely cross with each other, generating

potential causes of variation via genetic pollution/impurity. To maintain seed purity, seed

producers need to consider a range of factors (Stewart, 2002).

The first and key factor is isolation distance. Proper isolation among crops and

fields of different cultivars within each crop are required. A n isolation border of 3000 m

for breeder seed and 1500 m for certified seed is recommended. This is to minimize pollen

flow caused by honeybees, bumble bees and wind (Sharma et al, 2004). Field history of

Brassica seed production and Brassica weeds need to be recorded and taken into account.

Brassica seeds m a y survive 10-15 years in the soil so a cropping history free of Brassica

will be required for a clean crop. The overlap of flowering time with nearby crops, off-

types and weeds also needs to be considered. A s flowering time of a Brassica crops is

quite extended, there is a large window of opportunity for an overlap of flowering. The

propensity of the crop to accept outside pollen is another factor. This increases from self-

compatible to self-incompatible to hybrid plants. Furthermore, C M S hybrid types would

have an even greater propensity to accept outside pollen. Seed need to be screened for size

to remove any potential hybrids. Dressing to remove small seed can significantly reduce

the contamination from other species but is unlikely to affect the contamination caused by

pollen of the same species. Harvest machinery cleanliness also needs to be maintained. All

machinery and seed boxes must be completely free of seed.

To summarise, impurity in Fi hybrid cauliflower may be caused by genetic and

environmental factors and agronomic and postharvest management. Genetic factors were

investigated in this thesis.

Molecular Markers technology and its applications

Theoretically, all molecular techniques can be used to define identity, purity and

stability of a given plant but some techniques are better suited for particular species or

22

cultivars. The range of techniques available for immediate application varies, with major

species offering a wide range of established molecular markers and minor or less well-

studied species requiring evaluation of potential marker systems (Table 1). The choice of a

method for any particular application will depend upon the difficulty of the distinction to be

made and factors such as time, facilities and funds available. The nature of the sample will

vary and the appropriate techniques to be used will require investigation (Henry, 1997).

Types of DNA markers

D N A markers can be classified into hybridization and P C R based techniques.

RFLP is the major hybridization-based marker system, whereas PCR-based markers

include R A P D , AFLP, CAPS, EST, S C A R and SSR. Besides these, high-throughput

techniques of micro-arrays have been developed, which employ both P C R and

hybridization and these can be used to simultaneously analyse a large number of loci

(Lakshmikumaran et al, 2003).

RFLP (Restriction fragment length polymorphism)

RFLP markers remain extremely useful in research applications. These markers

provide co-dominant bands where heterozygote can be distinguished from homozygote

bands that are easily interpreted and amenable to population genetic analysis (Nybom,

2001; Rafalski and Tingey, 1993).

The utility of RLFP is hampered by the large amount of D N A required (5-10 ug) for

restriction digestion and southern blotting. Further, the requirement of a radioactive isotope

makes the analysis relatively expensive and hazardous. The assay is time-consuming and

labour-intensive, often producing low numbers of polymorphic bands (Nybom, 2001;

Rafalski and Tingey, 1993).

RAPD (Random amplified polymorphic DNA)

R A P D technique has been adopted most widely. R A P D markers are preferred for

their ease of use. The key innovation of this method is the use of D N A amplification to

generate genetic markers that require no prior knowledge of the target D N A sequence. The

technique uses P C R with short oligonucleotide primers of arbitrary (random) sequence to

generate genetic markers (William et al, 1990). R A P D is attractive for breeding

applications because it requires only a small amount of D N A (15-25 ng), a non radioactive

23

assay that can be performed in several hours and a simple experimental set up (Rafalski et

al, 1994).

The main issue associated with the use of this technique is to ensure reproducibility

of amplification profiles. Both the quantity and quality of the template D N A preparation

substantially influences results. Besides template D N A , magnesium concentration, cycling

temperatures and times, and base composition of the primer affect the success of R A P D

analyses (Henry, 1997). These factors need to be carefully controlled to ensure reproducible

results. R A P D is also dominant markers, meaning that in a segregating population the

homozygote of the parental type cannot be distinguished from the heterozygote (Nybom,

2001).

Simple Sequence Repeats (SSRs or Microsatellites)

Tandem arrays of short nucleotide repeats from 1 to 5 bases per unit are usually

called microsatellites or SSRs (simple sequence repeats). In particular, the dinucleotide

repeats (AC)n, (AG)n and (AT)n are abundant and highly polymorphic in eukaryotic

genomes (Rafalsky and Tingey, 1993).

In microsatellite analysis, a P C R is conducted using primers that are complementary

to the D N A sequences of regions flanking a highly variable microsatellite. The resulting

polymorphic bands are caused by differences in the number of D N A repeat units and

provide locus-specific, co-dominantly inherited bands with a high level of polymorphism.

SSRs have gained increasing importance in plant genetics and breeding. High abundance

and extensive polymorphism make them an ideal marker system for genetic mapping and

the characterization of germplasm, particularly in very closely related and inbreeding

species (Plieske and Struss, 2001; Saal et al, 2001). A major drawback is the time-

consuming and expensive development of suitable primers which are usually transferable

only among related species (Nybom, 2001).

Microsatellites of Brassica species are well documented. A large number of

microsatellites from rape seed (B. napus) have been identified and characterized. Many

rape seed microsatellite flanking primer pairs are functional in the A and C genome species

within the genus Brassica, but are not useful as markers for a wide range of species in the

family Brassicacea (Saal et al, 2001). Cauliflowers are among diploid C genome species

and therefore may utilize microsatellite information established in B. napus.

24

AFLP (Amplifiedfragment length polymorphism)

A F L P method combines both RFLP and R A P D . There are two rounds of

amplification, the first with primers that will amplify numerous fragments and then a

second with more specific primers. Numerous bands are produced for each primer pair.

These bands are dominantly inherited just like R A P D . The main difference between A F L P

and R A P D is that A F L P yields more bands but is technically more demanding. Thus A F L P

is cost-effective in situations where many analyses need to be performed (Henry, 1997;

Nybom, 2001). The main disadvantage of A F L P is the difficulty in identifying

homologous markers (alleles), rendering this method less useful for studies that require

precise assignment of allelic states, such as heterozygosity analyses (Mueller and

Wolfenbarger, 1999).

ISSR (Inter simple sequence repeat)

ISSR involves the use of microsatellite sequences as primers in a polymerase chain

reaction to generate multilocus markers. It is a simple and quick method that combines

most of the advantages of SSRs and A F L P to the universality of random amplified

polymorphic D N A (RAPD). ISSR markers are highly polymorphic and are useful in

studies on genetic diversity, phylogeny, gene tagging, genome mapping and evolutionary

biology (Reddy et al, 2002).

ISSR have high reproducibility possibly due to the use of longer primers (16-25

mers) as compared to R A P D primers (10-mers), which permit the use of high annealing

temperature (45-60°C) leading to higher stringency (Reddy et al, 2002). ISSR segregate

mostly as dominant markers following simple Mendelian inheritance (Gupta et al, 1994).

Other markers, marker combinations

Other markers such as C A P S and various modifications and combinations of the

above described methods (Table 1) have been developed and proven useful (Gupta et al,

1999; Wolfe and Liston, 1998).

Application of molecular markers

Cultivar identification

The analysis of genetic variation or diversity in plants has been traditionally

assessed by analysis of morphological or biochemical traits. The assessment of phenotype

25

may not be a reliable measure of genetic difference because of the influence of environment

on gene expression. The analysis of plant D N A allows the direct assessment of variation in

genotype (Henry, 1997).

The type of molecular method used to measure genetic distances in plants will vary

depending upon the magnitude of the genetic differences being assessed. Techniques such

as R A P D analysis may be useful for distinguishing different genotypes within a plant

cultivar while sequence analysis of the ribosomal genes may allow species or higher level

analysis (Henry, 1997).

Table 1. Types of genetic markers commonly generated using P C R method (Rafalski and

Tingey, 1993).

Charac

teristics

Principle

Type of

polymorphis

m

Inheritance

Need prior

sequence

knowledge?

Need radio-

labelling?

R A P D

Arbitrary

primers used

for D N A

amplification

Nucleotide

substitutions

indels

Dominant

No

No

SSR

Site-specific

amplification

of SSRs

Changes in

number of units

in repeated

motif

Codominant

Yes

Sometimes

Genetic Markers

ISSR

Non-specific

amplification

of SSRs

Nucleotide

substitutions

indels

Dominant

No

No

AFLP

Amplification

of fragment

length

polymorphisms

Nucleotide

substitutions

indels

Dominant

No

Sometimes

CAPS, MRSP

Restriction

digest of PCR

amplification

Nucleotide

substitutions

indels

Codominant

Yes

No

R A P D has been widely used to study genetic diversity in Brassica. Geraci et al.

(2001) utilized R A P D markers to assess genetic complexity of Sicilian populations of

Brassica using bulked seeds. They reveal that R A P D technique is useful to confirm

morphological differences among populations and to differentiate extremely similar

populations. Similarly, Cansian and Echeverrigaray (2000) successfully discriminated 16

26

commercial cabbage cultivars using R A P D analysis. Yuan et al. (2004) employed R A P D

markers to study genetic diversity among populations and breeding lines in Brassica napus.

R A P D analyses of individuals of open pollinated cultivars and landraces of collard in the

United States indicated that intra-population genetic variance accounts for as much

variation as that observed between populations (Farnham, 1996). The R A P D technique has

been utilized to discriminate between 14 broccoli and 12 cauliflower cultivars from the U S

and European seed companies. Amplification products from only four random primers

were sufficient to differentiate between cultivars (Hu and Quiros, 1991).

In other vegetable crops, R A P D technology is a rapid, precise and sensitive

technique for identification of pea genotypes (Pisum sativum) (Samec and Nasinec, 1996),

melon (Cucumis melo) (Staub et al, 2004), chicory (Chicorium intybum) (Bellamy et al,

1996) and radish (Raphanus sativus) (Pradhan et al, 2004a).

Besides R A P D , other markers such as AFLP, ISSR and SSR have been employed

for cultivar fingerprinting. A F L P was successfully used to estimate levels of genetic

diversity among Brassica crop species, particularly B. carinata, B. juncea, B. nigra

(Warwick and Soleimani, 2001) and B. rapa (Zhao et al, 2005). ISSR is used for genetic

diversity studies in several Brassica species and Arabidopsis (Bornet and Branchard, 2004).

SSR markers are also used to study the genetic relationships of Brassica vegetables

(Tongue and Griffiths, 2004) and B. napus (Lowe et al, 2004).

D N A fingerprinting techniques exhibit a great potential as a tool for a wide range of

areas in plants, including genotype identification, population genetics, taxonomy and plant

breeding. It appears that R A P D is a powerful technique for cultivar distinction and

therefore was used in this study to distinguish Fi hybrid cultivars commonly grown in

Australia and open pollinated cultivars (landraces) from Indonesia.

Genetic purity analyses

Determination of genetic purity of Fi hybrid seeds is a quality control requirement

in the production of hybrid Brassica vegetable seeds. R A P D technique is employed to

determine levels of self-inbred plants (genetic 'sibs') in a number of hybrid Brassica crops

such as cabbage (Crockett et al, 2000), broccoli (Crockett et al, 2002), Chinese cabbage

(Meng et al, 1998) and cauliflower (Boury et al, 1992) and other vegetable crop such as

chicory (Bellamy et al, 1996).

27

To test for the level of genetic sibs, D N A was isolated from seed or germinated

seeds and then examined using R A P D analyses. Sib contamination percentage obtained by

R A P D analyses was similar to that from field trials. This shows R A P D analyses can be

used for seed purity testing of commercial hybrid seeds (Crockett et al, 2000; 2002, Meng

etal, 1998).

Application of other molecular techniques to detect self-inbred seed in Fi hybrids

has not been reported. In this study, R A P D and SSR techniques were employed to

distinguish parent lines and use the markers generated to differentiate sib and hybrid plants

at the genetic level and compare these with morphological sibs (phenotypically identified

sib plants).

Markers association with morphological traits

Molecular marker associations with agronomically important traits have been

reported in a number of plants such as wheat, soybean, mungbean and Brassica. R A P D

technique is mainly used to find associations, with some reports using other techniques

such as RFLP, ISSR and SSR.

Molecular markers associated with seed and seedling traits using R A P D markers

were reported in radish (Pradhan et al, 2004b). A number of studies have identified R A P D

markers linked with seed coat colour in Brassica napus (Somers et al, 2001), linoleic acid

in Brassica napus (Jourdren et al, 1996), rust resistance in B. juncea (Prabhu et al, 1998),

leaf shape, period to bolting and self incompatibility in B. campestris (Nozaki et al, 1997)

and siliqua shatter resistance in B. rapa (Mongkolporn et al, 2003).

In crop plants, 3 ISSR markers are associated with low seed size in wheat

(Ammiraju et al, 2001), and 13 SSR markers were identified as being associated with seed

size in soybean. R L F P analyses on mungbean (Vigna radiata) revealed 11 loci for seed

weight. T w o of them co-localised with hard seededness (Humpry et al, 2005). RLFP

markers associated with seed weight also exist for soybean (Glycine max). U p to nine

independent loci were associated with seed weight in two different populations (Mian et al,

1996).

Bulk segregant analysis is commonly used to link molecular markers with

phenotypic traits to generate linkage maps. A well-defined linkage map can describe the

linkage relationships of genetically characterized markers with traits of interest (Bert and

Lydiate, 2003; Quijada et al, 2004). Recently a different approach based on statistical

28

analysis was introduced to find associations between molecular markers and morphological

traits (Pradhan et al, 2004b). The new approach employing multivariate analysis and

generating correlations based on principal coordinate analyis is expected to be simpler and

quicker as there is no need for segregating population and a huge number of primers to be

screened. The use of multivariate analysis was explored in this thesis to find links between

seedling traits and molecular markers.

In summary, there are a range of molecular marker techniques currently available

for research in plant genetics and breeding. The marker techniques will be applied in this

study for cultivar identification, purity testing and finding associations between molecular

markers and morphological traits.

29

Chapter 3

Fingerprinting of Cauliflower Cultivars using RAPD Markers

This chapter has been published in Australian Journal of Agricultural Research and is

presented in its P D F format. Citation: Astarini IA, Plummer JA, Lancaster R A , Yan G

(2004) Fingerprinting of cauliflower cultivars using R A P D markers. Australian Journal of

Agricultural Research 55:117-124.

CS1R0 PUBLISHING

www.publish.csiro.au/joumals/ajar Australian Journal of Agricultural Research, 2004, 55, 117-124

Fingerprinting of cauliflower cultivars using R A P D markers

Ida A. AstariniA'C, Julie A. Plummet, Rachel A. Lancaster^, and Guijun YanA

APlant Biology, Faculty of Natural and Agricultural Sciences, The University of Western Australia, 35 Stirling Highway, Crawley, W A 6009, Australia.

BDepartment of Agriculture Western Australia, Bunbury District Office, P O Box 1231, Bunbury, W A 6231, Australia.

Corresponding author; email: [email protected]

Abstract. Randomly amplified polymorphic D N A (RAPD) was used to investigate genetic relationships among 25 cultivars of cauliflower (Brassica oleracea var. botrytis L.). Forty decamer primers were examined, among which 15 primers produced polymorphism. Twenty-five polymorphic bands were observed, ranging in size from 428 to 1646 bp. A fingerprinting key was generated using these polymorphic bands. A dendogram was constructed using neighbour-joining analysis based on phylogenetic analysis using parsimony (PAUP). Results indicate that R A P D markers can be used for the routine identification of cauliflower cultivars within B. oleracea var. botrytis L.

Additional keywords: Brassica oleracea var. botrytis L., DNA markers, DNA polymorphism, genetic relationships.

Introduction

Cauliflower is one of the most important commercial crops of Brassica oleracea. Many open-pollinated cultivars and F, hybrids are commercialised around the world and new cultivars are continuously being released. Traditional cultivar identification in cauliflower, as in other crops, is based on a laborious evaluation of phenological and morphological characteristics (Cansian and Echeverrigaray 2000). Other techniques such as ion-exchange high performance liquid chromatography (IE-HPLC) (Mennella et al. 1996) and isozyme analysis (Zheng and Liu 1994) have also been attempted for varietal identification in Brassica.

Cauliflower identification is important for accurate classification of the cultivar, for example, when there is controversy over ownership with regard to Plant Breeder's Rights. Problems with plant classification based on morphological characters have been reported in cauliflower and broccoli (Malatesta and Davey 1996). Intensive breeding within these crops has resulted in irregularities in their classification and cultivar identification. Identification based on morphological characters is also time consuming and requires expensive field trials and evaluation. Morphological differences may be due to environmental influences on plants. Environmental factors such as space, amount of irradiance, temperature, water, and mineral nutrition affect physiological processes and morphology of plants (Kumar et al. 1998).

DNA-based markers have an advantage over other tests for cultivar identification, in that D N A is not affected by environmental factors or the developmental stage of the

© CSIRO 2004

plant. In recent years, identification of Brassica cultivars has been attempted using D N A markers such as restriction fragment length polymorphisms (RFLPs) (Dos Santos et al. 1994), randomly amplified polymorphic D N A sequences (RAPDs) (Williams et al. 1990), microsatellites (Charters etal. 1996), and amplified fragment length polymorphisms (AFLP) (Das et al. 1999).

Despite the development of the newer techniques, R A P D methodologies have retained their advantage in that they are fast, require no radioactive handling facilities, and the costs are relatively low. In Brassica, R A P D markers are considered to be as efficient as R F L P markers for estimating intraspecific genetic relationships among genotypes (Dos Santos et al. 1994). R A P D has been used successfully for genetic fingerprinting in B. oleracea and B. rapa (Kresovich et al. 1992), B. oleracea var. capitata (Phippen et al. 1994), B. rapa ssp. pekinensis (Lamboy et al. 1994), and B. napus (Dulson et al. 1998). R A P D markers have been used to analyse genetic variability among cultivar collections of cauliflower, cabbage, and kale populations in France (Margale et al. 1995). R A P D also appears to be a useful tool to confirm the U triangle relationship between diploid and amphidiploid Brassica taxa (Demeke et al. 1992).

Limited work has been reported on the identification and evaluation of genetic relationships between cultivars or germplasm entries of cauliflower. R A P D markers have been used to quickly estimate genetic distances between cauliflower cultivars in France (Boury et al. 1992) and between broccoli and cauliflower cultivars in the U S A (Hu and Quiros 1991). A pairwise distance matrix was developed

10.1071/AR03012 0004-9409/04/020117

31

http://www.publish.csiro.au/joumals/ajar

mailto:[email protected]

18 Australian Journal of Agricultural Research I. A. Astarini et al.

sing the computer program phylogenetic analysis using

arsimony ( P A U P ) to determine the relationships a m o n g

ultivars.

A fingerprinting key using R A P D markers has not been

sported for cauliflower cultivars. A key could be based on

ie same principals as conventional classification keys.

however, conventional keys use morphological

laracteristics, whereas D N A markers could be used in a

ngerprinting key. Markers could be developed from

mding patterns of polymerase chain reaction ( P C R )

roducts from each cultivar.

The aim of this study was to provide a protocol for routine

lentification of cauliflower cultivars within B. oleracea var.

otrytis L. using R A P D markers based on a simple

ngerprinting key and to determine the genetic relationships

nong these cultivars.

Iaterials and methods

'ant material

ie plant material used in this study included 18 cauliflower cultivars ithin B. oleracea var. botrytis L., grown in a variety trial in the field at e Department of Agriculture Western Australia Horticultural ;search Institute at Manjimup, and seedlings of 7 other cultivars pplied by G & S Seedling Nursery, The Seedling Factory, and South icific Seeds. The cultivars were Advantage, Alabama, Cauldron,

:leste, CF535, CF536, Chaser, CLF33902, Discovery, Donner, emont, Gibralter, G376, G389, J3195, Liberty, Monarch, Morpheus, 3444, Omeo, Plana, Sirente, SPS716, SPS3074, and Virgin. These

ltivars were supplied by Henderson Seeds, Lefroy Valley Seeds,

'ngenta Seeds, South Pacific Seeds, and Yates Vegetable Seeds. iltivars were selected by seed companies as the most likely to be

ccessfiil in the Manjimup district, which is the major region in

estern Australia for the production of export cauliflowers. Individual

if samples were collected from the field, transported to the laboratory i ice, and then stored at -80°C.

VA extraction

Â was extracted from leaf samples following the C T A B method

scribed by (Yan et al. 2002). Leaf tissue (1 g) from 4 young leaves

is ground in liquid nitrogen using a mortar and pestle. The extract was

msferred to 50-mL centrifuge tubes containing 10 m L C T A B

traction buffer [ 2 % CTAB, 100 m M TRIS (pH 8), 20 m M EDTA,

1M NaCl], and |J-mercaptoethanoI (20 uL) was then added. The

xture was swirled gently and incubated in a waterbath at 60°C for 15

n. Chloroformrisoamyl alcohol (24:1, 10 m L ) was added, mixed

:11, and centrifuged at 2236G for 20 min at 20°C. The supernatant was

llected in a fresh centrifuge tube and an equal amount of isopropanol

is added, mixed well and then refrigerated at -20°C for 2 h. The lution was then centrifuged at 2236G for 10 min at 20°C to separate

: D N A pellet. Isopropanol was poured off and the pellet was washed th 7 0 % and then 100% ethanol. The pellet was then dryed in a

siccator at 37°C for 30 min. The pellet was resuspended in 500 uL '• buffer [10 m M TRIS-HC1 (pH 8) and 1 m M E D T A (pH 8)] and then > uL RNAse (10 ng/mL) was added to remove R N A contamination,

i the tube was tapped gently to mix it thoroughly. The mixture was

itedat37°Cforatleast3h.

IA quantification

e D N A extract was diluted with sterile deionised water to 1/250 and

'A concentration was measured using a U V absorbance D U R 640

spectrophotometer (Beckman, U S A ) at 260 ran. Readings were taken 3 times for each sample and the average calculated. Absorbance at

260 n m was used to calculate the D N A concentration in the sample

(Eqn 1) and the ratio between absorbance at 260 nm and 280 nm was used to estimate D N A purity. The required D N A concentration

(60 ng/uL) was prepared for each genotype using injection water.

Absorbance at 260 nm x 250 * 50 = fig/mL DNA cone. (1)

RAPD analysis

Forty arbitrary decamer primers were examined for P C R amplification. All primers were synthesised by Life Technologies customer primer

program and published sequences are indicated in Table 1. The P C R reaction was performed in a final volume of 25 uL containing injection water, 1 x Taq polymerase buffer (Promega), 1.5 units of Taq polymerase (Promega), 0.05 m M of each dNTP (dATP, dCTP, dGTP,

dTTP; Promega), 1 p.M of primer, 1.5 m M MgCl2, and 120 ng template

D N A . A negative P C R tube containing all components except genomic D N A was used with each primer to check for contamination. P C R was performed in an iCycler (Bio-Rad, U S A ) using the following cycling

program: 10 times 5-s cycles of denaturation at 94°C, annealing at 35°C for 30 s, elongation at 72°C for 1 min, 25 times 5-s cycles of denaturation at 94°C, annealing at 45°C for 30 s, elongation at 72°C for

1 min, and finally 1 cycle including an elongation step at 72°C for 2 min. The iCycler was programmed to retain the samples at 4°C until they were collected and stored at-20°C.

Gel electrophoresis

Each sample of R A P D products (10 uL) was mixed with 6 x gel loading

buffer (2 uL) a nd loaded onto an agarose (1.5% w/v) gel for electrophoresis (Bio-Rad, N S W ) in 1 x T A E buffer (50 x T A E buffer

contains 242 g TRIS base, 57.1 g glacial acetic acid, 100 m L 0 . 5 M EDTA,

and distilled water to 1 L) at 60 V for 2 h. A 100-bp D N A ladder (5 pX of D N A ladder and 1 uL of gel loading buffer, Promega) was included

on both sides as a molecular standard. Amplification products separated

by gels were stained in ethidium bromide solution (2 uL Etbr/100 m L 1 x T A E buffer) for 30 min and then photographs were taken under U V

light using a digital camera (Kodak D C 120) and the images were recorded with a Macintosh Kodak ID 2.0 computer program.

Analysis of data

A data matrix was created from photographs of gels by scoring 1 for present bands or 0 for absent bands. The size of amplification products

in base pairs was estimated using the D N A marker with bands of known

molecular weight. The regression of distance run against the molecular

weight of each band of the 100-bp D N A ladder was used to calculate the equivalent molecular weight in base pairs for each band. Only

clearly scorable bands with the size between 300 and 1700 bp were

included in the analysis. A pairwise distance matrix was generated based on total and mean

R A P D band differences in PAUP, using a Power Macintosh 7600/120

(Swofford 1993). The data were subsequently used to construct a

dendogram using neighbour-joining analysis. To test the validity of phylogenetic relationships revealed by neighbour-joining, the data were

also used to generate fingerprinting keys.

Results

All 40 primers produced multiple PCR fragments in each

cultivar. Ninety bands were scored, of which 25 were

polymorphic. Fifteen out of the 4 0 primers tested had

polymorphic bands (Table 1). N o bands were observed in the

control lane (Fig. 1).

32

Fingerprinting cauliflower using RAPD markers Australian Journal of Agricultural Research 119

Table 1. Decamer primers used in this study Primers were synthesised by Life Technologies. Primers marked with

an asterisk (*) generated polymorphisms

Primer name

AOI A

A02 A

A03 A

*A04A

D12B

D20B

Hong-H

LISA-1

LISA-2

•OPA-07C

•OPB-04D

OPB-08C

*OPB-12D

•OPH-01E

OPH-03E

OPH-06E

OPH-09E

*UBC-106F

UBC-I27F

UBC-147F

Nucleotide sequence (5' -> y)

AAGACGACGG

AATCCGCTGG

AGTCGGCCCA

AACAGGGCAG

CACCGTATCC

ACCCGGTCAC

GTCACTGCTC

GGCCTTGAGT

GGTCCTCAGG

GAAACGGGTG

GGACTGGAGT

GTCCACACGG

CCTTGACGCA

GGTCGGAGAA

AGACGTCCAC

ACGCATCGTG

TGTAGCTGGG

CGTCTGCCCG

ATCTGGCAGC

GTGCGTCCTC

Primer name

*UBC-250F

SF-04

•SF-06

SF-08

SF-09

SF-13

SF-17

•SL-01

SL-03

SL-07

•SL-08

SL-12

•SK-01

SK-02

•SK.-03

•SK-09

*SK-14

SK-17

SK-18

•SK-19

Nucleotide sequence

(S'-+3')

CGACAGTCCC

GGTGATCAGG

GGGAATTCGG

GGGATATCGG

CCAAGCTTCC

GGCTGCAGAA

AACCCGGGAA

GGCATGACCT

CCAGCAGCTT

AGGCGGGACC

AGCAGGTGGA

GGGCGGTACT

CATTCGAGCC

GTCTCCGCAA

CCAGCTTAGG

CCCTACCGAC

CCCGCTACAC

CCCAGCTGTG

CCTAGTCGAG

CACAGGCGGA

Obtained from: A Hu and Quiros (1991); BBoury el al. (1992); cCrockett et al. (2002); "Crockett et al. (2002); EMeng el al. (1998); FMailer and May (1999).

The molecular weight of bands amplified from 15 primers (Table 1) ranged from 428 to 1646 bp (Table 2). Two to 6 bands were scored per primer. A minimum of 12 markers (coded as primer-number of base pairs: A04-1427, OPA07-498, OPB04-797, OPB12-1159, OPH01-1646, OPH01-1460, SL01-1062, SL08-698, SK14-1294, SK14-575, UBC106-525, and UBC250-839) obtained from 10 primers was required to distinguish between cultivars. A fingerprinting key was developed for 18 cultivars (Fig. 2), but 7 cultivars could not be separated. Cultivar CLF33902 showed identical R A P D profiles to G376, as did cultivar Plana with Alabama, cultivar Donner with SPS3074 and with Sirente. A dendogram showing the relationship among cultivars was generated (Fig. 3). Three major clusters were obtained. Monarch had a very distant relationship to other cultivars. Cultivars CLF33902 and G376 were very closely related, as revealed by their neighbour-joining (NJ) coefficient being zero. Zero NJ coefficient was also obtained between cultivars Plana and Alabama, and between cultivars Donner, SPS3074, and Sirente.

The pairwise distance matrix generated by the P A U P program was used to quantify the differences among all

M l 2 3 4 5 1 1 1 1

6 7 8 9 10 1 2 3 4 1 1 1 1 1 2 2 2 2 2 5 6 7 8 9 20 1 2 3 4 5 C M

Fig. 1. R A P D amplification profiles obtained with primer UBC106. Standard bands are indicated by arrows. Molecular weights of the standard are indicated in base pairs. M, 100 bp D N A ladder; C, control lane; 1, Monarch; 2, Donner; 3, M3444; 4, Cauldron; 5, Gibralter; 6, Chaser; 7, CF535; 8, SPS716; 9, Liberty; 10, Omeo; 11, G389; 12, Advantage; 13, CLF33902; 14.CF536; 15.G376; 16, Virgin; 17.J3195; 18, Sirente; 19, Morpheus; 20, Fremont; 21, Plana; 22, Alabama; 23, SPS3074; 24, Celeste; 25, Discovery.

33

Australian Journal of Agricultural Research

•a

5 CQ

3 --u ex u 11 u a- w

£ « O J3

5 *•" 3 i_ « 0) U JO

*> E r« 3 i- c

« >> .2 -a

2 «

E •= a) — T3 -C — 60

II u !=. .2 — o ©

Hi vA M

» o n

bu vA o> cfl o "">

C/3 o

J OO \o

•J OO "1 C/3 O ©

*u v ~i ©

W 4 05 IZI Z <N

n_ • OO

OSS * J. 2 O o ^

o o 2

2: ~ "* 0 © S

CN °°

U

g _, t—

Q • CN < • * • • *

d, <s ^2

o - -PQ 1 r-

O © r-

& • 00 PL, t_~ ON o © •*

2S S

o-a.<n-cua.cu<o-D-&.o-D-a.p_ti.D-a.a.a.a-a,D-ci-D-

a - < < e u < e u a . c u f t . < c u < o . < o - C L . p - a . c u a , < < < & , <

< < 0 , < < < < < < < < < < < < ; < < < < < < < < a . <

< < < < < < < < » . < : < < < < < < < < : < < < < : < < : <

o - < < a - < c i . < < e L , < e L , C L , p L , < c u p u , < < : a . o - < o , < e L , ( i ,

C U C U C W C u A ^ C U C U C U O - i P w C U t C u ^ i P - i P k O - t O - t C u O ^ C u C U O u C W ^ C U

c u & - a - a - c u a . c u c u D - c u C u c u o - c u c i ^ c i - & - c i - o - c i - c u c ^ c ^ c ^ <

a . G . Q . D - & . f t . f l . a , & . a , f l , Q - P ^ G . f l . & . a . B - f i . c i . o . c u C L a . <

a > < < < < < < o . O H e u , P L , < < < < C L , < < < < < < < f t - <

< < < < < < < » - < < < < < < < < < < ; < < < < < < <

C u C u < C d - d - Q - C u < C C u £ X C l - C l , D - C U O - C l - C u C u C u ( X C u C l - P - C u C u

cua.<D-cuA-cu<a-<a-cuCi,cup-CL.cua-ci-o-ci-a.p-cucu

< < e u < < < < a . < < < < < < < < < < < < < < < < <

£ 1 , < C U C 1 - < C U C 1 - ( X < < C 1 - P - P U < & - C L 1 < P - & - ( X < 0 , < C 1 . 0 -

< e u , < - < c u < < : < ; < ( i . < < < e L , < C L , c u < < < 0 H O , a , c u <

< < < < < < < < < < < < ( I , < C L , < < ; < ; < < < < < < <

< < < < < < < < < < < < < < < < < < < < < < < < p -

<<<<a*<e*<<<<<<<<<*.<<<<<<<<

< < < < Q * P « < < < < < P « < < < < P * < < < < < < < 0 *

P - C u < C U ( X D - C u < < ; c i H j X D « C X , C X 1 C l - C U C U P H P - f X . C u C u C L - a - C l .

<ja,<;<<a,<<<ai<'<eu-<a.aceLiPuewaiai<eL,eL,-<

< < p u < < < < a , - < < < p - . < < ; < ; < e L < < ; < < < < < < <

< < a , - < < - < < a , e L i a < c u i P L , a , - 0- S 8 °> §

a> CN bo o 3 ON - <n vo

O - S O *rl « c a - o s S n P t gov § « m vo .S S g « S n » 0 Sro a£jCU<a.-S BnTJJlLn.S-' H ^ — cL, u

3 e?

«, l- u

M 2 P

Fingerprinting cauliflower using R A P D markers A ustralian Journal of Agricultural Research 121

1. U B C 106-525

1.1. O P B 04-797 P 1.1.1. A04-1427 P 1.1.2. A04-1427 A 1.1.2.1. OPH01-1646 1.1.2.2. OPH01-1646 1.1.2.2.1. UBC250-839 1.1.2.2.2. UBC250-839

1.2. O P B 04-797 A

1.2.1. OPB12-1159 1.2.2. OPB 12-1159

1.2.2.1. A04-1427 1.2.2.1.1. OPH01-1460 P 1.2.2.1.2. OPH01-1460 A

1.2.2.1.2.1. SK14-1294 1.2.2.1.2.2. SK14-1294

1.2.2.2. A04-1427 A 1.2.2.2.1. OPH01-1646 P 1.2.2.2.2. OPH01-1646 A

1.2.2.2.2.1. SK 14-1294 1.2.2.2.2.2. S K 14-1294 1.2.2.2.2.2.1. SL01-1062 1.2.2.2.2.2.2. SLO1-1062

A

Cauldron, Chaser, CF535, Liberty, G389, Advantage, CLF33902, G376, J3195, Plana. Alabama, Celeste, Morpheus, Discovery Liberty, G389, Advantage, CLF33902, G376 CLF33902, G376 Liberty, G389, Advantage P Advantage A Liberty, G389 P Liberty A G389 Cauldron, Chaser, CF535, J3195, Plana Alabama, Celeste. Morpheus, Discovery J3195 Cauldron, Chaser, CF535, Plana, Alabama, Celeste, Morpheus, Discovery

2. U B C 106-525

2.1. O P A 07^198 P 2.1.1. OPB 04-797 P 2.1.1.1. SL08-698 P 2.1.1.2. SL08-698 A

2.1.2. OPB 04-797 A 2.1.2.1. A04-1427 2.1.2.2. A04-1427

2.2. O P A 07-498 A

2.2.1. OPB 04-797 2.2.2. OPB 04-797

P A

2.2.2.1. A04-1427 2.2.2.1.1. SK. 14-575 2.2.2.1.2. SK 14-575

2.2.2.2. A04-1427 2.2.2.2.1. OPH01-1460 2.2.2.2.2. OPH01-1460

Chaser, Plana, Alabama, Morpheus Chaser Plana, Alabama, Morpheus P Morpheus A Plana, Alabama Cauldron, CF535, Celeste, Discovery Discovery Cauldron, CF535, Celeste P SPS3075 A Cauldron, CF535 P Cauldron A CFS3S

Monarch, Donner, M3444, Gibralter, SPS716. Omeo, CF536, Virgin, Fremont, SPS3074, Sirente Monarch, M3444, SPS716, Fremont M3444.SPS716 SPS716 M3444 Monarch, Fremont P Fremont A Monarch Donner, Gibralter, Omeo, CF536, Virgin, SPS3074, Sirente Omeo Donner, Gibralter, CF536, Virgin, SPS3074, Sirente P Donner, Virgin, SPS3074, Sirente P Virgin A Donner, SPS3074, Sirente A Gibralter, CF536 P Gibralter A CF536

Fig. 2. Fingerprinting key for cauliflower cultivars generated from R A P D markers.

Bold indicates that the cultivar has been differentiated.

cultivars (Table 3). The pairwise difference between cultivars

ranged from 0 to 16 markers.

Discussion

Eighteen cultivars were differentiated in the fingerprinting key, showing that each of these cultivars had a different genetic background. Seven other cultivars were clustered together into 3 different groups, indicating that they are closely related. Cultivars from the same company were often clustered together, possibly indicating similar parentage and a high level of genetic similarity. Cultivar Donner was the same as SPS3074, which was the breeding line number

(South Pacific Seeds, pers. comm.). Most crops are given a commercial name upon release, eliminating the requirement for identification numbers (Noli et al. 1999).

Cultivars CLF33902 and G376 were obtained from different seed companies; however, they may have similar parental lines. The same germplasm may be used in different breeding programs, resulting in similar progeny and released varieties (Cansian and Echeverrigaray 2000). It should be noted that a small number of genes could make a substantial difference to the performance of a variety. This is highlighted, for example, in herbicide, pest, or disease resistance. R A P D , by definition, uses

35

Australian Journal of Agricultural Research I. A. Astarini et al.

II -S 8. 2 E C a> .—

is -ON

8 g 00

I- O

3

u

I oo o

I »o in *n

I V"l O oo ©

I \o m \o Tt vo

I o vo nô-t NO

I es <N Tf »o Tf V£) oo

I ôô\ovor-,oooo

I OOTl-Tj-Tj-Tl-lOTt-^-O

I VOOO^(SNOO<nOO,000

I ôôoTrNMooinooôoo

I v->oo*oor-.r^-*nvia\Tj-cj\c^v>

I ^ôooc«^ôo\»o^«^^no^','^1•o^«/^^••-

I oooo\rfou-icj\c^cômoorno\m

I o\r^r-~voo\voooooovoisDor-oooo

I 0-0\«tNnNNTfO(NM'<Ntrn'tTj->0

I '*v1"C0C--O'^Tfc000C«"l»O»OC*"»'—'—• t*- •* Is- <o <o

I oo M cs -•w-io(N'^-mvorr»u->r^fn'—• ~* r- <N r-- v> <n

I iNhnh^Tromo\ox^MTt^TtvoôioOMO

I r^^^côoTtTfc>^c«r^vow^vou^rna\rocncot^Ti-t---»or--

a> cN 00 o 8 os

u a>

§ §J5N 3 J3 « J " 0 0

2Q2OOUUKJOO<:UOO>R£E<SUM2O

^ R O N « n S o -S « 2 a S m « E p. 5

36

Fingerprinting cauliflower using R A P D markers Australian Journal of Agricultural Research 123

0.5

0.5

0.3

0.5

1.2

0.5

1

0.4

0.4

0.2

0.4

2

.0

0,1

1.9

0.1

0.9

0.7

0.9

1.1

0.3

0.6

0.7

1.3

0

0

0.4

2.6

1.3

S

6.4

0.6

0

0

0

0

1.6

3.4

1.1

3.9

0

0

1.7

0.3

1.8

2.2

2.8

0.2

_ i

— Gibralter

- C F 5 3 6

— Monarch

Fig. 3. Dendogram of 25 cauliflower cultivars, constructed using

PAUP based on neighbour-joining (NJ) analysis. The numbers at each node represent NJ coefficient of differences.

randomly marked sections of the genome and any associated genes are not necessarily critical in distinguishing cultivar performance (Heneen and Jorgensen 2001). Also note that the R A P D technique tends to provide only dominant markers. Individuals containing 2 copies of an allele are not distinguished quantitatively from those containing only 1 copy (Williams et al. 1990).

The dendogram also indicated possible genetic relationships among cultivars (Fig. 3) and similarities of these relationships with economic traits and source of seeds were also noticed. For example, in trials in Manjimup, W A , in 2002, Monarch was clearly distinguished from other cultivars. High similarity between Donner and Sirente corresponded to the morphology of these cultivars. Both Donner and Sirente are recommended for spring and autumn harvest and both have pure white curds (McArthur 2001). In contrast, Plana, Discovery, and Fremont, which were in different groups of the dendogram, are all summer cultivars (McArthur 1999).

The neighbour-joining coefficient varied between 0.1 and 6.4, which means that the genetic diversity ranged from little

to reasonably wide. A few cultivars appeared to have the same or similar parent lines. Cauliflower has been bred for a long time, and the present day cultivars have a very narrow genetic base. This is typical of many present-day crops. Following R A P D analysis on cabbage and pea, Jaccard's coefficient ranged from 0.72 to 0.87 and from 0.49 to 0.98, respectively, indicating little genetic diversity. Many cabbage or pea cultivars have the same ancestors (Samec and Nasinec 1996; Cansian and Echeverrigaray 2000).

Identification of cauliflower cultivars with R A P D markers has been reported by H u and Quiros (1991). They used 12 cultivars from 4 American seed companies. Successful cultivar identification by diagnostic markers was developed using 4 primers (AOl, A 0 2 , A 0 3 , and A04). With our cultivars, which came from European, Australian, and N e w Zealand based seed companies, only A 0 4 produced polymorphic bands. This may indicate that different seed companies have used parent lines with different genetic background. However, these studies indicated that a RAPD-based key encompassing all worldwide cultivars is a feasible goal for future research.

Bulked D N A samples were used in this experiment in order to detect the c o m m o n genetic base of each cultivar. Bulked D N A is commonly used for R A P D analysis, e.g. in cauliflower, cabbage, and kale local cultivars from France (Margale et al. 1995), B. napus cultivars (Dulson et al. 1998; Mailer and May 1999), and Sicilian wild populations of Brassica (Geraci et al. 2001). Use of bulked samples may allow identification of a distinctive profile of R A P D markers. Moreover, the large number of populations in Brassica collections makes genetic diversity studies based on plant-to-plant analysis impracticable.

Hybrid cultivars used in this experiment were produced using C M S (cytoplasmic male sterility) and SI (self-incompatibility) systems. In the C M S systems, male sterile plants are used as the female parent and a specific maintainer line is used as the male parent. C M S allows the production of 1 0 0 % hybrid seed (Ruffio-Chable et al. 2000). Self-incompatibility avoids the production of self-inbred plants and is commonly used for hybrid seed production of cauliflower, cabbage, and broccoli (Crockett et al. 2002).

In Brassica, R A P D markers are considered to be as efficient as R F L P markers for estimating genetic relation-ships among genotypes. A study on 45 B. oleracea genotypes indicated that R A P D provides a level of resolution equivalent to RFLPs for determination of the genetic relationships among genotypes (Dos Santos et al. 1994).

The present study demonstrated that R A P D analysis provides a simple and reliable method for cultivar identification. Using R A P D markers to identify genetic diversity within B. oleracea var. botrytis L. is important, providing breeders with genetic information for the improvement of crops. Identification of genetic diversity/similarity may help in selection of appropriate

37

;4 Australian Journal of Agricultural Research I. A. Astarini et al.

eeding lines. Future work to systematically identify R A P D

arkers associated with economic traits, origin, and general

;netic diversity would be beneficial.

cknovvledgments

re thank AusAID for providing a scholarship to Ida Ayu

starini. Thanks also to Henderson Seeds, Lefroy Valley

;eds, Syngenta Seeds, South Pacific Seeds, and Yates

jgetable Seeds for providing seeds, and to G & S Seedlings

id The Seedlings Factory for providing s o m e seedlings for

is project. This project was supported by grants from the

estern Australia Department of Agriculture and Plant

lology, The University of Western Australia.

eferences

iury S, Lutz I, Gavalda M-C, Guidet F, Schlesser A (1992)

Empreintes genetiques du chou-fleur par R A P D et verification de la

purete hybride FI d'un lot de semences. Agronomie 12, 669-681.

nsian RL, Echeverrigaray S (2000) Discrimination among cultivars

of cabbage using randomly amplified polymorphic D N A markers.

HortScience 35, 1155-1158. larters Y M , Robertson A, Wilkinson MJ, Ramsay G (1996) P C R

analysis of oilseed rape cultivars (Brassica napus L. ssp. oleifera) using 5'-anchored simple sequence repeat (SSR) primers. Theoretical and Applied Genetics 92, 442—447. doi: 10.1007/S001220050147 ockett PA, Singh M B , Lee C K , Bhalla PL (2002) Genetic purity analysis of hybrid broccoli (Brassica oleracea var. italica) seeds using R A P D PCR. Australian Journal of Agricultural Research 53,

51-54. doi:10.1071/AR01022 s S, Rajagopal J, Bhatia S, Srivastava PS, Lakshmikumaran M

(1999) Assessment of genetic variation within Brassica campestris

cultivars using amplified fragment length polymorphism and

random amplification of polymorphic D N A markers. Journal of

Biosciences 24,433-440. meke T, Adams RP, Chibbar R (1992) Potential taxonomic use of

random amplified polymorphic D N A (RAPD): a case study in Brassica. Theoretical and Applied Genetics 84, 990-994.

s Santos JBD, Nienhuis J, Skroch P, Tivang J, Slocum M K (1994) Comparison of R A P D and RFLP genetic markers in determining

genetic similarity among Brassica oleracea L. genotypes.

Theoretical and Applied Genetics 87, 909-915. lson J, Kott LS, Ripley V L (1998) Efficacy of bulked D N A samples

for R A P D D N A fingerprinting of genetically complex Brassica

napus cultivars. Euphytica 102, 65-70.

doi:10.1023/A:1018378304701 raci A, Divaret I, Raimondo FM, Chevre A M (2001) Genetic

relationships between Sicilian wild populations of Brassica

analysed with R A P D markers. Plant Breeding 120, 193-196.

doi: 10.1046/J. 1439-0523.2001.00589.X

leen W K , Jorgensen R B (2001) Cytology, R A P D , and seed colour of progeny plants from Brassica rapa-alboglabra aneuploids and

development of monosomic addition lines. Genome 44,1007-1021. J, Quires CF (1991) Identification of broccoli and cauliflower

cultivars with R A P D markers. Plant Cell Reports 10, 505-511. sovich S, Williams JGK, McFerson JR, Routman EJ, Schaal B A

(1992) Characterization of genetic identities and relationships of

Brassica oleracea L. via a random amplified polymorphic D N A

assay. Theoretical and Applied Genetics 85, 190-196.

http://www.publish.

Kumar PP, Yau JCK, Goh CJ (1998) Genetic analyses of Heliconia

species and cultivars with randomly amplified polymorhic D N A (RAPD) markers. Journal of the American Society for Horticultural

Science 123, 91-97. Lamboy WF, McFerson JR, Li R, Kresovich S (1994) Relationships

among Chinese vegetable brassicas using R A P D markers.

Cruciferae Newsletter 16,44—44. Mailer RJ, May C E (1999) Heterogeneity of random amplified

polymorphic D N A sequences in individual seedlings and bulked

samples of four cultivars of Brassica napus. Plant Breeding 118,

465^170. doi: 10.1046/J. 1439-0523.1999.00428.X Malatesta M , Davey JC (1996) Cultivar identification within broccoli,

Brassica oleracea L. var. italica Plenk and cauliflower, Brassica

oleacea var. botrytis L. Acta Horticulturae 407, 109-113. Margate E, Herve Y, H u J, Quiros C F (1995) Determination of genetic

variability by R A P D markers in cauliflower, cabbage and kale local cultivars from France. Genetic Resources and Crop Evolution 42,

281-289. McArthur S (1999) Winter newsletter. South Pacific Seeds,

Christchurch, N Z . McArthur S (2001) Winter newsletter. South Pacific Seeds,

Christchurch, N Z . Meng X, Hong M , Zhang W, Wang D (1998) A fast procedure for

genetic purity determination of head Chinese cabbage hybrid seed based on R A P D markers. Seed Science and Technology 26,

829-833. Mennella G.Iori A, Sanaja VO, Magnifico V (1996) Broccoli and

cauliflower cultivars identification through IE-HPLC seed protein

analysis. Acta Horticulturae 407, 115-121, Noli E, Conti S, Maccaferri M , Sanguineti M C (1999) Molecular

characterization of tomato cultivars. Seed Science and Technology 27, 1-10.

Phippen W B , Kresovich S, McFerson JR (1994) Assessing genetic

identity and relatedness in cabbage with RAPDs. Cruciferae Newsletter 16, 46-46.

Ruffio-Chable V, Chatelet P, Thomas G (2000) Developmentally

'aberrant' plants in F, hybrids of Brassica oleracea. Acta Horticulturae 539, 89-94.

Samec P, Nasinec V (1996) The use of R A P D technique for the

identification and classification of Pisum sativum L. genotypes.

Euphytica 89, 229-234. Swofford D L (1993) 'PAUP: Phylogenetic analysis using parsimony,

version 3.1.' (Illinois Natural History Survey: Champaign, IL) Williams JGK, Kubelik AR, Livak KJ, Rafalski JA, Tingey S V (1990)

D N A polymorphisms amplified by arbitrary primers are useful as

genetic markers. Nucleic Acids Research 18, 6531-6535. Yan G, Shan F, Plummer JA (2002) Genetic relationship within Boron ia

(Rutaceae) as revealed by karyotype analysis and R A P D molecular

markers. Plant Systematics and Evolution 233, 147-161. Zheng XY, Liu Y (1994) Inbred testing of Chinese cabbage F( varieties

by peroxidase and esterase isozyme analysis. Acta Horticulture

Sinica 21, 65-70.

Manuscript received 9 January 2003, accepted 1 September 2003

csiro.au/journals/aj ar

38

http://www.publish

http://csiro.au/journals/aj

Chapter 4

Genetic Diversity of Open Pollinated Cauliflower

Cultivars in Indonesia

This chapter has been presented in an International/Australian Society of Horticultural

Sciences (ISHS/AUSHS) conference entitled: Harnessing the potential of horticulture in the

Asian-Pacific region. The paper has been published in Acta Horticulturae and is presented

here in the published format. Citation: Astarini IA, Plummer JA, Yan G, Lancaster R A

(2005) Genetic diversity of open pollinated cauliflower cultivars in Indonesia. Acta

Horticulturae 694, 149-152.

Genetic diversity of open pollinated cauliflower cultivars in Indonesia

LA. Astarini, J.A. Plummer and G.Yan R.A. Lancaster Plant Biology, F N A S Department of Agriculture Western Australia The University of Western Australia Bunbury District Office 35 Stirling Highway, Crawley W A 6009 P O Box 1231 Bunbury, W A 6231, Australia Australia

Keywords: Brassica oleracea, DNA fingerprinting, DNA markers, genetic relationships.

Abstract Eight cauliflower cultivars collected from three production regions in

Indonesia were evaluated using R A P D markers. The objectives of this study were to investigate genetic variation and relationships between cultivars and to evaluate variation within cultivars as all of them are open-pollinated. D N A was extracted using Nucleon Phytopure Plant D N A extraction kit, followed by treatment with RNAse. D N A polymorphism generated from 10 polymorphic primers was used to construct a dendogram using the unweighted pair-group method with arithmetic averages ( U P G M A ) . The R A P D technique indicated that variation occurred both within and between cultivars.

INTRODUCTION Cauliflower is gaining popularity as a vegetable crop in Indonesia. Cauliflowers

attract middle and upper income Indonesian, because the price is higher than other vegetables available in the market. Cauliflower is produced in cooler highland regions across The Indonesian archipelago. Lembang, Malang and Bedugul are central production areas for vegetables in West Java, East Java and Bali respectively. In Bali, cauliflowers are grown to supply the tourist industry.

Most cauliflower cultivars available in Indonesia are open-pollinated (OP) lines. Therefore, variation in product frequently occurs. Variation is found in curd size, maturity time and resistance to diseases such as club root. Little is known about the genetic make up of these cultivars and there has never been a systematic evaluation of cauliflower cultivars in Indonesia. The first step in crop improvement in developing countries should be a full assessment of the local materials (Williams et al, 1991).

The objectives of this study were to determine the diversity present among Indonesian cultivars and their relationships, to evaluate within cultivar variation and to examine the potential of Indonesian cauliflowers as new sources in Brassica oleracea gene

pools.

MATERIAL AND METHODS

Plant materials and DNA extraction Eight cauliflower cultivars were collected, they are 'Harli', 'Broad' and 'Blaster'

(from West Java), 'Manalagi', 'Bandung' and 'Gembel' (from East Java), 'Malang' and 'Bedugul' (from Bali). Since liquid nitrogen was not available at Biotechnology Laboratory, Udayana University, Bali, leaf samples, together with mortars and pestles were frozen in the -80°C freezer for 30 minute before grinding. Once the leaves were frozen, they were ground immediately into powder. D N A was extracted using Nucleon Phytopure

Proc. IS on Hort. in Asian-Pacific Region 40 (149)

Ed. R. Drew Acta Hort. 694, ISHS 2005

Plant D N A extraction kit (Amersham Biosciences, UK). Purified D N A was kept in 1.5 ml tubes in a cool container and brought to Perth, Western Australia for analysis.

R A P D analysis

Three cultivars ('Blaster', 'Manalagi' and 'Bedugul) were chosen for inter-variety differences with eight individuals for each cultivar. To test between cultivars variation, D N A from 4 individuals of were bulked randomly. Ten arbitrary decamer primers were screened for PCR amplification. Primer name and nucleotide sequence (5' —• 3') were: SK-01 (CATTCGAGCC), SK-14 (CCCGCTACAC), SK-19 ( C A C A G G C G G A ) , SL-01 (GGCATGACCT), SL-08 ( A G C A G G T G G A ) , SF-06 (GGGAATTCGG), UBC-106 (CGTCTGCCCG), UBC-250 (CGACAGTCCC), OPA-07 ( G A A A C G G G T G ) , OPB-04 (GGACTGGAGT). All primers were synthesized by Life Technologies customer primer

program. The P C R reaction was performed in a final volume of 15 uL containing injection water, lx Taq polymerase buffer (Promega), 1 unit of Taq polymerase (Promega), 0.05 m M

of each dNTP (dATP, dCTP, dGTP, dTTP; Promega), 1 u M of primer, 1.5 m M MgCl2 and 40 ng template D N A . P C R was performed in an iCycler ™ (Bio-Rad, USA) using the following cycling program: 10 times 5 s cycles of denaturation at 94°C, annealing at 35°C for 30 s, elongation at 72°C for 1 min, 25 times 5 s cycles of denaturation at 94°C, annealing at 45°C for 30 s, elongation at 72°C for 1 min and finally 1 cycle included an elongation step at 72°C for 2 min. The iCycler was programmed to retain the samples at 4°C until they were collected and stored at -20°C. The P C R products were examined using 1.8 % agarose gel electrophoresis in T A E buffer and stained with ethidium bromide. A 100 bp ladder (Promega) was used as a size marker.

Data analysis A data matrix was created based on photographs of gels by scoring 1 for present

bands or 0 for absent bands. The molecular weight in base pairs for each band was estimated using regression of distance run against the molecular weight of the 100 bp D N A ladder. A pairwise distance matrix was generated based on total R A P D band differences in PAUP (Phylogenetic Analysis Using Parsimony), using a Power Macintosh 7600/120 (Swofford, 1993). The data was subsequently used to construct a dendogram using U P G M A analysis.

RESULTS AND DISCUSSION All ten R A P D primers produced polymorphic bands. These primers have proved

useful in distinguishing Australian cultivars (Astarini et al, 2004). This experiment confirmed that R A P D technique is reproducible and is a reliable, rapid method for D N A

fingerprinting. A total of 65 bands were scored and 34 of these were polymorphic. The molecular

weight of amplified bands ranged from 380 to 1800 base pairs. Four to eight bands were scored per primer. A dendogram showing the relationship between and within cultivars was generated (Figure 1). Two major clusters were obtained. 'Bedugul' has a distant relationship to other cultivars. 'Bedugul' was cultivated and bred locally in Bali. This indicating 'Bedugul' may have different origins from the rest of the cultivars.

'Harli' and 'Broad' were cultivated in Lembang, West Java. These cultivars may have been introduced from India in the 19th century and have been reproduced locally since then. 'Blaster' and 'Manalagi' are intermixed in the dendogram, suggesting that these cultivars are similar, although they are cultivated in different regions, West Java and East

Proc. IS on Hort. in Asian-Pacific Region 41 (150)

Ed. R. Drew Acta Hort. 694, ISHS 2005

Java respectively. Gene flow is more likely to occur within the island than between islands. 'Malang', another Bali line, had a close relationship with 'Gembel' from East Java. A number of cauliflower growers in Bali have bought cauliflower seeds from East Java, and this explains the similarity between these two cultivars.

There was substantial within variety variation and almost all individuals were separated on the dendogram. Seed production of each variety is done by local farmers and variation within variety maybe due to cross-pollination, as isolation of plants during seed production is poor. Populations have been selected and traditionally multiplied by growers and they will therefore possess genetic adaptation to local conditions. This variability may contribute to diversity for breeding purposes.

The sale of seeds among local farmers, including lots of commercial seed from seed companies, m ay also contribute to genetic variability. In this region, cauliflowers are planted around vegetable gardens and seeds are often collected from the best individuals without specific varietal isolation. Frequent intercrossing among different local varieties in the same area may increase the genetic variability of populations from the same region. In West Java, cauliflowers are planted as intercropping plants, usually with chilli or spring onion.

CONCLUSIONS In conclusion, the variability among cauliflower cultivars could be related primarily

to their geographical origin. Identification of genetic diversity with molecular markers may help in selection of appropriate breeding lines and the time for new variety development can be reduced.

ACKNOWLEDGEMENTS W e would like to thank AusAID for providing a scholarship to Ida Astarini. Thanks

also to Pak Ah m a d Rivani, Dr. Ir. Agus Suryanto M S , Ir. Sitawati, M S and Pak D e w a Okayadnya for providing information about cauliflower production in Indonesia and provision of field grown cauliflowers. Sincere thanks to Prof. I G.P. Wirawan for permitting Ida Astarini to extract D N A in his Biotechnology Lab in Bali. This project was supported by grants from The Department of Agriculture Western Australia and Plant Biology, The University of Western Australia.

Literature cited Astarini, I.A., Plummer, J.A., Lancaster, R.A. and Yan, G. 2004. Fingerprinting of

cauliflower cultivars using R A P D markers. Australian Journal of Agricultural Research

55:117-124. Swofford, D. L. 1993. 'PAUP:Phylogenetic analysis using Parsimony, version 3.1' (Illinois

Natural History Survey: Champaign, Illinois) Williams, C.N., Uzo, J.O. and Peregrine, W.T.H. 1991. Vegetable Production in the

Tropics. Intermediate Tropical Agriculture Series. Longman Scientific and

Technical, U K .

Proc. IS on Hort. in Asian-Pacific Region

Ed. R Drew Acta Hort. 694, ISHS 2005

42(151)

3.0

0

0 5

0.5

2.1

2.3 51

2.6 77

1.9

0 3.0

3.0

1.6 53

0.2

0.3 55

0.5 72

2.5

2.5

0

0

0.5

1.7

1.0 59

1.5 60

0.2

1.0

1.0

U.5

0.5

0.2

1.3

1.2

u

0

0.5

0.5

1.5

1.5

0.8

6.2

0.8

0.6

0.5

i..\j

2.0

3.0

1.0 ZR

3.5

<£.U

2.0

1.1

3.4 9 3

3.0

30

1.5

1.5

Harli

Broad

Bandung

Gembel

Malang

Blaster-1

Blaster-2

Manalagi-2

Blaster-3

Manalagi-4

Manalagi-7

Manalagi-1

Manalagi-3

Manalagi-8

Manalagi-5

Manalagi-6

Blaster-4

Blaster-6

Blaster-7

Blaster-8

Blaster-5

Bedugul-1

Bedugul-2

Bedugul-7

Bedugul-8

Bedugul-3

Bedugul-4

Bedugul-5

Bedugul-6

Fig. 1. Dendogram of Indonesian cauliflower cultivars, constructed by unweighted pair-

group method with arithmetic averages ( U P G M A ) based on total character differences.

Numbers above branches represent branch length and numbers below branches indicate

bootstrap values.

Proc. IS on Hort. in Asian-Pacific Region Ed. R. Drew Acta Hort. 694, ISHS 2005

43 (152)

Chapter 5

Genetic Diversity of Indonesian Cauliflower

Cultivars and Their Relationships with Hybrid

Cultivars Grown in Australia

This chapter has been accepted and currently in press in Scientia Horticulturae and is

presented here in its submitted format. Citation: Astarini IA, Plummer JA, Yan G,

Lancaster R A (2006) Genetic Diversity of Indonesian Cauliflowers and Their

Relationships with Hybrid Cultivars Grown in Australia. Scientia Horticulturae (in

press)

Genetic diversity of Indonesian cauliflower cultivars and their

relationships with hybrid cultivars grown in Australia

Ida A. AstariniAB, Julie A. PlummerA, Rachel A. Lancaster0, and Guijun Y a n A

School of Plant Biology, Faculty of Natural and Agricultural Sciences, The University of

Western Australia, 35 Stirling Highway, Crawley W A 6009, Australia.

Author for correspondence: Ida Ayu Astarini, email: arin(a).student.uwa. edu.au

Fax: +61 8 6488 1108, Phone: +61 8 6488 1992

department of Agriculture, Western Australia, Bunbury District Office, P O Box 1231

Bunbury, W A 6231, Australia.

Abstract

The objectives of this study were to investigate genetic variation and relationships

between Indonesia, Australian and European based cultivars and to evaluate variation

within Indonesia cultivars as all cultivars are open-pollinated. Eight cauliflower cultivars

collected from three production regions in Indonesia and four Fi hybrids cultivars grown in

Australia were evaluated using R A P D and ISSR markers. D N A polymorphisms generated

from 10 polymorphic R A P D primers were used to construct a dendogram using the

unweighted pair-group method with arithmetic averages ( U P G M A ) and to generate a

fingerprinting key. ISSR marker analysis using 14 primers were attempted but D N A

polymorphisms could not be clearly identified. The R A P D technique indicated that

variation occurred both within and between Indonesian cultivars. Comparison between

Indonesian, Australian and European based cultivars showed that Indonesian cultivars have

unique genotypes and would be good sources of genes for future crop improvement.

Key words: Brassica oleracea; D N A fingerprinting; D N A markers; ISSR, R A P D .

1. Introduction

Cauliflower is gaining popularity as a vegetable crop in Indonesia. Cauliflowers are

mainly consumed by middle and upper income Indonesians, because they are still less

common and only seasonally available, and therefore attract a higher price compared to

traditional vegetables available in the market. Cauliflower is produced in cool highland

regions across the Indonesian archipelago. Lembang, Malang and Bedugul are central

45

http://edu.au

production areas for vegetables in West Java, East Java and Bali respectively. In Bali,

cauliflowers are grown to supply the tourist industry.

Most cauliflower cultivars available in Indonesia are open-pollinated lines.

Therefore variation in product frequently occurs. Variation is found in curd size, maturity

time and resistance to diseases such as club root. Little is known about the genetic make up

of these cultivars and there has never been a systematic evaluation of cauliflower cultivars

in Indonesia.

Major constraints to increasing cauliflower production in Indonesia include lack of

varieties adapted to tropical conditions, inadequate supplies of seeds of improved varieties

and a high incidence of pests and diseases (Asandhi and Sastrosiswojo, 1988; Darmawan

and Pasandaran, 2000). Indonesian research stations in collaboration with overseas seed

companies and growers from major cauliflowers growing areas have tested introduced

cultivars in Indonesia. To date these cultivars have not been successful and breeding

cauliflower for tropical conditions is required.

The first step in crop improvement in developing countries should be a full

assessment of the local materials, including collection, evaluation and molecular

characterization of germplasm lines. Often, local varieties are of excellent quality and

flavour, have a good level of resistance to pests and diseases and may be superior to exotic

materials (Williams et al, 1991)

Cauliflower cultivars grown in Australia are mainly Fi hybrid cultivars, which have

a limited genetic source. In the long term, there is a need for new or novel genes for

cultivar improvement. It is essential that w e avoid loosing genetic variation and many

Western cultivars were highly bred (Sharma et al., 2004). Investigating genetic distance

between Indonesian and Australian grown cultivars will provide useful information for

future breeding programs in both countries.

Inter Simple Sequence Repeat (ISSR) and Random Amplified Polymorphic D N A

( R A P D ) were employed to investigate the genetic diversity. ISSR markers are highly

polymorphic and are useful for genetic diversity, phylogeny, genome mapping and

evolutionary studies (Reddy et al., 2002). It is a simple and quick method, although

optimisation of the P C R reaction of each primer needs to be tested before applying the

technique (Pharmawati et al., 2004). R A P D has been widely used to study genetic

relationships between cultivars in Brassica (Astarini et al, 2004, H u and Quiros, 1991;

Divaret et al., 1999; Mailer and May, 1999), radish (Pradhan et al., 2004), Cucumis spp.

46

(Zhuang et al., 2004), and banana (Onguso et al., 2004). This is because the technique is

simple, requires small amounts of D N A , does not require information on D N A sequence

and it is economical (William et al., 1990). The numbers of R A P D primers utilized to

determine relationships between cultivars varies. H u and Quiros (1991) could distinguish

14 broccoli and 12 cauliflower cultivars using 4 primers that produce 40 bands. Mailer and

May (1999) also employed 4 primers producing 61 bands to differentiate four Brassica

napus cultivars. Genetic diversity of some core collections of Brassica oleracea were

determined using 8 primers that produced R A P D bands ranging from 39 to 57 depended on

sources of D N A (Divaret et al., 1999). In radish, 5 primers producing 52 bands were

sufficient to distinguish cultivars (Pradhan et al., 2004) and 19 R A P D primers were

employed to characterize banana cultivars in Kenya (Onguso et al., 2004).

The objectives of this study were to investigate genetic variation and relationships

between Indonesian and Australian and European-based cultivars and to fingerprint

Indonesia cultivars based on D N A markers.

2. Materials and methods

2.1. Plant materials and DNA extraction

Eight cauliflower cultivars were collected from Indonesia, they are 'Harli', 'Broad'

and 'Blaster' from West Java, 'Manalagi', 'Bandung' and 'Gembel' from East Java, and

'Malang' and 'Bedugul' from Bali. Since liquid nitrogen was not available, leaf samples,

together with mortars and pestles were frozen in -80°C freezer for 30 min to maintain low

temperature during grinding of leaf samples into fine powder. It is important to keep the

materials, mortars and pestles cold to prevent D N A degradation. D N A was extracted using

a Nucleon Phytopure Plant D N A extraction kit (Amersham Biosciences,

Buckinghamshire). Purified D N A was kept in 1.5 ml tubes in a cool container and brought

to Perth, Western Australia for analysis. D N A of 4 hybrid cultivars commonly grown in

Australia, 'Atlantis' and 'Omeo' (from an Australian-based breeding program), 'Monarch'

and 'Plana' (from a European-based breeding program) were extracted at The University of

Western Australia.

2.2. ISSR analysis

Fourteen ISSR primers (UBC, Vancouver) were screened for polymerase chain

reaction (PCR) amplification. The optimum annealing temperature for P C R was

47

determined for each primer. The PCR reaction was performed in a final volume of 15 uL

containing injection water, lx Taq polymerase buffer (Promega, California), 1 unit of Taq

polymerase (Promega, California), 0.05 m M of each dNTP (dATP, dCTP, dGTP, dTTP;

Promega, California), 0.3 u M of primer, 1.5 m M MgCl2 and 40 ng template D N A . D N A

amplifications were performed in an iCycler™ (Bio-Rad, California) using the following

cycling program: 15 min at 95°C for initial activation step, followed by 45 cycles of 30 s at

94°C, 45 s at annealing temperature (50°C, 54°C depending on the primers used) and a 2

min extension at 72°C. The iCycler was programmed to retain the samples at 4°C until they

were collected and stored at -20°C. P C R amplification was repeated 2 times using separate

D N A samples.

2.3. RAPD analysis

Three cultivars ('Blaster', 'Manalagi' and 'Bedugul') were chosen to study inter-

variety variations with 8 individuals from each cultivar. To test between cultivar variation,

D N A was bulked from 4 individuals for the 5 other Indonesia cultivars and 4 Australian

grown hybrid cultivars. Ten R A P D primers suitable for use in cauliflowers (Astarini et al,

2004) were employed for PCR amplification. Primer names and nucleotide sequences

were: SK-01 (CATTCGAGCC), SK-14 (CCCGCTACAC), SK-19 ( C A C A G G C G G A ) , SL-

01 (GGCATGACCT), SL-08 ( A G C A G G T G G A ) , SF-06 ( G G G A A T T C G G ) , UBC-106

(CGTCTGCCCG), UBC-250 (CGACAGTCCC), OPA-07 ( G A A A C G G G T G ) , OPB-04

(GGACTGGAGT). All primers were synthesized by Life Technologies Inc. (Madison)

customer primer program. The P C R reaction was performed in a final volume of 15 uL

containing injection water, lx Taq polymerase buffer (Promega, California), 1 unit of Taq

polymerase (Promega, California), 0.05 m M of each dNTP (dATP, dCTP, dGTP, dTTP;

Promega, California), 1 u M of primer, 1.5 m M MgCl2 and 40 ng template D N A . P C R was

performed in an iCycler ™ (Bio-Rad, California) using the following cycling program: 10

times 5 s cycles of denaturation at 94°C, annealing at 35°C for 30 s, elongation at 72°C for

1 min, 25 times 5 s cycles of denaturation at 94°C, annealing at 45°C for 30 s, elongation at

72°C for 1 min and finally 1 cycle including an elongation step at 72°C for 2 min. The

iCycler was programmed to retain the samples at 4°C until they were collected and stored at

-20°C. R A P D analysis on each primer was repeated 2 times using separate D N A samples

to confirm results.

48

2.4. Gel Electrophoresis

Each sample of R A P D or ISSR products (10 uL) was mixed with 6x gel loading

buffer (2 uL) and loaded onto an agarose gel (1.8 % w/v) for electrophoresis (Bio-Rad,

California) in lx T A E buffer (50x T A E buffer contains 242 g Tris base, 57.1 g glacial

acetic acid, 100 m L 0.5 M E D T A and distilled water to 1 L) at 80 volt for 1.5 h. Ethidium

bromide solution (2 uL Etbr/100 m L lx T A E buffer) was added to the gel. A 100 bp D N A

ladder (Promega, California), 5 uL of D N A ladder and 1 uL of gel loading buffer was

included on one sides of the gel as a molecular standard. Amplification products separated

by gels were photographed under U V light using a Kodak D C 120 digital camera and the

images were recorded with a Macintosh Kodak ID 2.0 computer program.

2.5. Data analysis

A data matrix was created based on photographs of gels by scoring 1 for present

bands or 0 for absent bands. The molecular weight in base pairs for each band was

estimated using regression of distance run against the molecular weight of the 100 bp D N A

ladder. Only clearly scorable bands were included in the analysis. A pairwise distance

matrix was generated based on total R A P D band differences in P A U P (Phylogenetic

Analysis Using Parsimony), using a Power Macintosh 7600/120 (Swofford, 1993). The

data were subsequently used to construct a dendogram using unweighted pair-group

method with arithmetic averages ( U P G M A ) . Bootstrap analysis was done based on 1000

reiterates to show the degree of confidence of each branch/node on the dendogram. The

higher bootstrap value (presented in percent), the higher degree of confidence (Swofford,

1993). To help the molecular identification of the cultivar tested, the data were also used to

generate fingerprinting keys.

3. Results

ISSR analysis was difficult to conduct. Only one primer ( G A C A ) 4 produced clear

bands but no polymorphisms were observed, four primers produced faint bands and nine

others did not produce bands.

All ten R A P D primers produced polymorphic bands. R A P D technique was

reproducible and was a reliable, rapid method for D N A fingerprinting. A total of 65 bands

were scored and 46 of these were polymorphic (Table 1). Four to eight bands were scored

per primer and molecular weight ranged from 250 to 1915 base pairs (Figure 1).

49

A fingerprinting key was developed for 10 cultivars (Figure 2). A minimum of 8

markers (coded as primer-number of base pairs): OPB04-1172, OPB04-760, OPB04-330,

OPA07-733, OPA07-509, UBC250-935.4, SL01-1915 and SL01-1076 obtained from 4

primers were required to distinguish between cultivars (Table 1). T w o cultivars, 'Blaster'

and 'Manalagi' could not be separated, as they had identical R A P D profiles.

A dendogram showing the relationship between and within cultivars was generated

(Figure 3). T w o major clusters were obtained, Australian grown cultivars were clustered

and separated from Indonesian cultivars and this was supported by high Bootstrap values of

88%. Australian-bred and European-bred cultivars were much more closely related and

were clustered together. 'Bedugul' had a distant relationship with other Indonesian

cultivars, while 'Manalagi' and 'Blaster' were intermixed. 'Harli', 'Broad', 'Bandung',

'Gembel' and 'Malang' were closely related.

Pairwise distance matrix produced from the P A U P program was used to quantify

differences among cultivars (Table 2). The pairwise difference between cultivars ranged

from 0 to 34. 'Atlantis' (Australian grown hybrid cultivar) tended to have the greatest

difference from Indonesian cultivars.

4. Discussion

Australian grown varieties are all hybrid cultivars from breeding programs based in

either European or/and Australian seed companies. These hybrids have been highly bred,

resulting in narrow genetic diversity. Replacement of open-pollinated cultivars with Fi

hybrids of a narrow genetic base has resulted in the genetic erosion of cauliflower cultivars.

In the long term, genetic variability in the form of landraces and primitive types will

disappear unless efforts are made to define and preserve them (Sharma et al., 2004).

Three ISSR primers which proved useful to generate polymorphisms in Brassica

oleracea (Leroy et al., 2000) and 11 primers useful for Leucadendron (Pharmawati et al.,

2004) were employed in this study. However, only one primer from B. oleracea produced

clear ISSR patterns. Primers from Leucadendron resulted in smear bands or did not

produce bands at all. The smeared ISSR bands obtained indicated that P C R conditions

need to be further refined. Smeared ISSR patterns are c o m m o n but not useful (Gupta et al.,

2000, Pharmawati et al., 2004). Annealing temperatures are usually the main factors

affecting pattern quality and reproducibility of ISSR fingerprints and are primer-specific

50

(Bomet and Branchard, 2001). Different annealing temperatures have been tested for each

primer in this experiment but no clear bands were obtained.

A fingerprinting key is useful for correct identification of cultivars. Correct

identification of a cultivar is an important step in a breeding program, to ensure the right

line for breeding purposes is chosen and not the same line under different names.

Fingerprinting keys can also be used for protection of new cultivars. Similarly, a

dendogram is a practical way to show relationships among cultivars tested. W h e n using

new or distinct materials for a breeding program, the dendogram will show the distance

between new or distinct materials and existing cultivars. This will assist plant breeders in

choosing which cultivars will be used in their breeding program.

In this study, R A P D showed clear bands and a high level of polymorphism. In

R A P D technique, it is important to maintain consistent reaction conditions that have been

optimised for reproducible D N A amplification. Several factors including template D N A

concentration, magnesium concentration, primer annealing temperature, primer length and

primer base composition all affect the reaction (William et al., 1990) and were carefully

controlled.

The distant relationships between Indonesian and Australian grown cultivars

suggest that the two gene pools have been separated and there has been no or limited gene

flow between the pools. For future breeding programs, the available cultivars could be

used as genetic sources for broadening the development of new cultivars, including hybrid

cultivars suitable for warmer climates.

'Harli' and 'Broad' were cultivated in Lembang, West Java. These cultivars may

have been introduced from India in the 19th century and have been reproduced locally since

then (Rukmana, 1994). 'Blaster' and 'Manalagi' were intermixed in the dendogram and

could not be separated by the fingerprinting key, suggesting that these cultivars are similar.

Although they are cultivated in different regions, West Java and East Java respectively,

gene flow is more likely to occur within an island than between islands. 'Malang', another

Balinese line, had a close relationship with 'Gembel' from East Java. A number of

cauliflower growers in Bali have purchased cauliflower seeds from East Java (Okayadnya,

pers. c o m m ) and this would explain the close similarity and perhaps past cross pollination

between these two cultivars.

There was substantial within-variety variation in the Indonesian cultivars and almost

all individuals were separated on the dendogram. Seed production of each variety is carried

51

out by local farmers and variation within variety maybe due to cross-pollination, as there is

little effort to isolate plants during seed production. Populations have been selected and

traditionally multiplied by growers and they will therefore possess genetic adaptation to

local conditions. This variation m a y contribute to diversity for breeding purposes.

The sale of seeds among local farmers, including lots of commercial seed from seed

companies, m a y also contribute to genetic variability. In West Java, cauliflowers are

intercropped, usually with chilli or spring onion. Cauliflower seeds are usually collected

from the best individuals without specific varietal isolation. Frequent intercrossing among

different local varieties in the same area may increase the genetic variability of populations

from the same region.

In Indonesia, most cauliflower seeds available for growers are open pollinated. The

main hindrance to the adoption of Fj hybrids by growers is their lack of availability and

cost. Attempts to introduce hybrid cultivars have not been successful. The current hybrid

cultivars have poor performance in the field, produce very small non-harvestable curds and

sometimes pink coloured curds due to high temperature (Grubben, 1977; Okayadnya,

pers.comm). This suggests that the introduced cultivars are not adapted to warm tropical

climates, even though they were cultivated in the cooler mountainous regions. The

temperature in Indonesia varies depending on altitude and distance from the sea. Average

temperatures are 28°C near the coasts and around 22°C in the mountains. Indonesia has an

average relative humidity between 7 0 % and 9 0 % . Therefore, there is tremendous potential

for the development of hybrid cauliflower cultivars for tropical countries such as Indonesia.

Identification of genetic diversity with molecular markers m a y help in selection of

appropriate breeding lines and the time for new variety development can be reduced. In the

future, research in the tropics should be aimed primarily to develop hybrid cultivars and

seed production technologies that provide better adapted varieties that will increase the

availability and quality of vegetables (Darmawan and Pasandaran, 2000).

In conclusion, the variability among cauliflower cultivars could be related primarily

to their geographical origin. This study demonstrated that Indonesian cauliflower cultivars

have unique genotypes that could be used for future breeding program.

52

Acknowledgements

W e thank AusAID for providing a scholarship to Ida Ayu Astarini. Thanks also to

M r Ahmad Rivani, Dr. Ir. Agus Suryanto M S , Ir. Sitawati, M S and M r Dewa Okayadnya

for provision of field grown cauliflowers in Indonesia. Sincere thanks to Prof. I. G. P.

Wirawan for permitting Ida Astarini to extract D N A in his Biotechnology Laboratory in

Bali. This project was supported by grants from The Department of Agriculture Western

Australia and Plant Biology, The University of Western Australia.

References

Asandhi, A. A., Sastrosiswojo, S., 1988. Research on vegetable in Indonesia. In: McLean,

B. T. (Ed.). Vegetable Research in Southeast Asia. Asian Vegetable Research and

Development Center, Taipei.

Astarini, I. A., Plummer, J. A., Lancaster, R. A., Yan, G., 2004. Fingerprinting of

cauliflower cultivars using R A P D markers. Aust. J. Agric. Res. 55, 117-124.

Bornet, B., Branchard, M., 2001. Nonanchored inter simple sequence repeat (ISSR)

markers: reproducible and specific tools for genome fingerprinting. Plant Moi. Biol.

Reporter 19,209-215.

Darmawan, D. A., Pasandaran, E., 2000. Indonesia, dynamics of vegetable production,

distribution and consumption in Asia. In: AH, M . (Ed.), Asian Vegetable Research and

Development Center, Taiwan.

Divaret, I., Margale, E., Thomas, G., 1999. R A P D markers on seed bulks efficiently assess

the genetic diversity of a Brassica oleracea L. collection. Theor. Appl. Genet. 98,

1029-1035.

Grubben, G. J. H., 1977. Tropical vegetable and their genetic resources. International

Board for Plant Genetic Resources, Rome.

Gupta, P. K., Varshney, R. K., 2000. The development and use of microsattelite markers

for genetic analysis and plant breeding with emphasis on bread wheat. Euphytica 113,

163-185.

Hu, J., Quiros, C. F. 1991. Identification of broccoli and cauliflower cultivars with R A P D

markers. Plant Cell Reports 10, 505-511.

Leroy, X. J., Leon, K., Branchard, M., 2000. Characterisation oi Brassica oleracea L. by

microsatellite primers. Plant Syst. Evol. 225, 235-240.

53

Mailer, R. J., May, C. E., 1999. Heterogeneity of random amplified polymorphic D N A

sequences in individual seedlings and bulked samples of four cultivars of Brassica

napus. Plant Breeding 118, 465-470.

Onguso, J. M., Kahangi, E. M., Ndiritu, D. W., Mizutami, F., 2004. Genetic

characterization of cultivated bananas and plantains in Kenya by R A P D markers. Sci.

Hort. 99, 9-20.

Pharmawati, M., Yan, G., McFarlane, I. J., 2004. Application of R A P D and ISSR markers

to analyse molecular relationship in Grevillea (Proteaceae). Aust. Syst. Bot. 17,49-61.

Pradhan, A., Yan, G., Plummer, J. A., 2004. Development of D N A fingerprinting keys for

the identification of radish cultivars. Aust. J. Exp. Agric. 44, 95-102.

Reddy, M. P., Sarla, N., Siddiq, E. A., 2002. Inter simple sequence repeat (ISSR)

polymorphism and its application in plant breeding. Euphytica 128, 9-17.

Rukmana, R., 1994. Budidaya Kubis Bunga dan Brokoli. Penerbit Kanisius, Yogyakarta.

Sharma, S. R., Singh, P.K., Chable, V., Tripathi, S. K., 2004. A review of hybrid

cauliflower development. Journal of N e w Seeds 6, 151-193.

Swofford, D. L., 1993. PAUP: Phylogenetic analysis using Parsimony, version 3.1 Illinois

Natural History Survey, Champaign, Illinois.

William, J. G. K., Kubelik, A. R., Livak, K. J., Rafalski, J. A., Tingey, S. V., 1990. D N A

polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic

Acids Res. 18,6531-6535.

Williams, C. N., Uzo, J. O., Peregrine, W . T. H., 1991. Vegetable production in the tropics.

Intermediate Tropical Agriculture Series. Longman Scientific and Technical, Essex.

Zhuang, F. Y., Chen, J. F., Staub, J. E., Qian, C. T., 2004. Assessment of genetic

relationships among Cucumis spp. by SSR and R A P D marker analysis. Plant Breeding

123, 167-172.

54

Figure 1

RAPD amplification profiles of 12 cultivars obtained with primer SL-01 and SL-08.

Standard bands are indicated by arrows. M, marker ladder, 1, Harli; 2, Blaster; 3, Broad; 4,

Manalagi, 5, Gembel; 6, Bandung; 7, Malang; 8, Bedugul; 9, Atlantis; 10, Omeo; 11,

Monarch; 12, Plana; M, Marker ladder.

1 23 4 56 7 8910 1112 123 4567 8 9 10 11 12 M

^^m9- -tUUft

Hi r f

A 4lilf4|W«i

1500 bp

1000 bp

500 bp

55

Figure 3

Dendogram of Indonesian cultivars ('Harli', 'Broad', 'Bandung', 'Gembel', 'Malang',

'Blaster', 'Manalagi', 'Bedugul'), Australian-bred cultivars ('Atlantis', 'Omeo') and

European-bred cultivars ('Monarch', 'Plana'), constructed by unweighted pair-group

method with arithmetic averages ( U P G M A ) based on total character differences. Numbers

adjacent to cultivars indicate collection number. Numbers above branches represent branch

length and numbers below branches indicated bootstrap values.

57

1.1

3.0

3.0

2.6 78

3.0

1.6 53

2.5

2.5

0

2.1

1.9

\J.C

0.3

1.5 63

1.3

6.2

1.2 57

0.5 73 0

0.5

1.7

1.0 61

1.0

1.0 n c

0.2 n

0.5 53

0.2

1.2

0.8

0

0.5

0.5

1.5

1.5

2.0

2.0

2.3

08

0.6

0.5

1.1 54

3.4 9 1

3.0 o n

1.0 50

3.5

2.0

3.0

3.0

1.5

1.5

2.5

3 9 88

5.5 99

3.5 89

2.5

4.5

4.5

Harli

Broad

Bandung

Gembel

Malang

Blaster-1

Blaster-2

Manalagi-2

Blaster-3

Manalagi-4

Manalagi-7

Manaiagi-1

Manalagi-3

Manalagi-8

Manalagi-5

Manalagi-6

Blaster-4

Blaster-6

Blaster-7

Blaster-8

Blaster-5

Bedugul-1

Bedugul-2

Bedugul-7

Bedugul-8

Bedugul-3

Bedugul-4

Bedugul-5

Bedugul-6

Atlantis

O m e o

Monarch

Plana

ON

CO

fi

.fi DO

• »—t

1)

o

~o

6

T3 1)

O

T3 O o ;-<

s S-i

CI, GO

CO T3 <D Xi . *~t

o 00



fi

o t-l

O

s a o CN

o fi

o o fi

X! <

.2

<2 CO t-l <L> M »-c

6 Q PH

2

CO

cd t-c

o ^—s '& -i->

fi 0) CO 0)

OH

o

ON

t

00

O i

VI

in . — i

in

ro in

m m NO

"t

in so

NO *o so

o o o

ro ro

so

m

60 00 NO

ro m

in

so CN ON

in ir

-

o ON CN

SO

so

ON

o

00 NO NO

in

o o ON

O 00 o

hi

at

|

o

<

0.

<

<

OH

PH

<

PH

<

<

PH

PH

PH

PH

<

<

<

P.

PH

<

<

<

TE sS

<

p.

<

<

PH

P.

<

P.

<

<

P.

PH

PH

PH

<

<

<

PH

O.

<

<

<

u U H-»

J3

m

<

PH

<

<

OH

OH

<

0-

<

<

0-

' O H

O.

OH

<

OH

<

o.

0-

<

<

<

"S 2 m

<

OH

<

<

0-

0-

<

o.

PH

<

<

<

OH

<

<

<

<

P.

PH

<

<

<

'5b as

<

p.

<

<

p.

PH

<

PH

<

<

PH

P-

O.

OH

<

PH

<

O H

O H

<

<

0-

13

a O

<

OH

<

<

OH

OH

<

OH

<

<

OH

<

o.

OH

<

<

<

a

OH

<

<

<

60

c 3 T3

§ P3

<

OH

<

<

OH

O.

<

0-

<

<

OH

OH

OH

0-

<

OH

<

O,

P,

OH

OH

0.

60

"3

<

OH

<

<

o.

o.

OH

0,

<

<

o.

O.

0-

OH

<

OH

OH

O.

OH

OH

OH

OH

"3 60 3 T3 <D

OQ

<

OH

OH

<

OH

<

<

OH

OH

OH

<

<

<

<

O H

<

<

O.

OH

o.

<

o.

C _eS

<

<

0.

o.

<

OH

OH

<

o.

OH

OH

<

<

<

<

OH

<

<

OH

O H

OH

<

OH

o £ O

<

OH

<

CU

OH

<

<

O H

<

o.

o.

<

OH

0.

OH

<

<

OH

O.

OH

OH

o,

J3

1 c o

<

OH

<

OH

OH

OH

<

OH

<

OH

<

<

OH

<

OH

<

<

OH

OH

OH

OH

OH

I E-

-a c c3

03

NO

o

6 5

NO O

i

HH 00

oo o

00

_ o J 00

0>

OO

in ro oo

ro

ro oo ro

O in

oo

m ro <n

o CN CN

O

o in

ro ro 00

oo ON

ro

O •n in

o

o ON

r--

o in Ov

in in •3-

o in in

ON

ON

NO

o

in

ON

O •n CN

in

ro -3-

u 08

S

u

o.

<

<

<

O H

<

OH

<

<

OH

<

a

<

<

0 H

OH

O-

<

<

<

<

X

O H

O H

<

<

O-

<

OH

<

<c

OH

<

0H

<

<

OH

OH

OH

<

<

<

<

PQ

OH

OH

<

OH

OH

<

0.

PH

<

O H

<

OH

<

<

OH

OH

OH

<

<

<

<

T3

2

OH

<

<

<

OH

<

OH

<

<

O H

<

O,

<

<

OH

OH

<

<

<

<

OH

'5b nS d C

o.

o.

<

<

OH

<

OH

<

OH

OH

<

OH

<

<

O.

OH

PH

<

0.

<

<

13

E

a

OH

O.

<

<

o>

<

o.

OH

<

o.

<

o.

<

<

OH

O H

O H

<

<

<

<

60

c 3

pa

O.

o>

<

<

o.

<

OH

O.

o.

OH

<

OH

<

<

O H

OH

OH

<

OH

<

<

60

"3

OH

o.

<

<

OH

<

O.

OH

<

OH

<

OH

<

<

OH

OH

OH

<

OH

<

<

"3 60 3 T3 U 03

O H

<

<

<

OH

OH

0.

<

<

<

<

<

OH

OH

OH

O.

OH

O,

<

<

<

c

<

OH

<

<

<

OH

OH

OH

<

<

<

OH

<

OH

OH

OH

O.

OH

O H

<

<

<

O O

E O

OH

<

<

<

O H

O.

0-

<

<

<

OH

O H

<

<

O H

0-

<

<

0.

<

<

0

c 0

OH

<

<

<

OH

OH

OH

<

OH

<

OH

<

OH

<

O.

OH

<

<

OH

<

<

US

s OH

o ro ro

m so ro

ir m in

so ON NO

u-o oa NO PH t-

o

f-CN

NO NO NO

ON O •n

CN CN NO

ON 00

T3

* t~~ ro P3 O ro

PH

° s ON SO CN

in 00 CN

< < < < < < < < < < < , < ( .

P H O H O H O H O H O H O H O H O H O H O H O H

P H O H O H O H O H O H O H O H < O . - < < !

P H O H O H < O H < < , O H O H O H < O H

P H < < < < < < < ; < < < <

< < < < < < < < ; < < p H f i H

< < < < < < < < ; q . p H C L 1 f N H

< < < < < < < < < < < <

o

< < O H < < < < < ; O , < < < ;

O H O H O H O H O H O H O H O H O H O H O H O H

O H O H P H P H O H O H O H O H O H O H O H O H

< < < < < < a « < <

O H O H O H O H O H O H O H O H ^ C

< < <

< < <

in

ON in t-H

o in N ON

in ro ON

O O ro

2 i

u f? r-D ON

NO 83

I u


P H P H O H O H O H O H O H O H O H O H O H O H




< < < < < < < < | 1 , < < P H



< < < O H < O H < H < O H O H O H <

'5b ô — cT T5 S M

EN -2 o

S J3 -5 PQ

•g 1 -g -s s 3 g -; - t •-: ^ •o jj ? - a § ffl S O « 2 ffl <

O e« -S

22 C C E o 3

o S S

CN NHO

T3 H-» CO

CO l-l

4>

PH

C N

»—<

OS

a o 00

*0

(N ro

O ro

ON

<N

00 CN

CN

NO CN

CN

CO CN

CN CN

O CN

i OO —

If i ro ON —

O — — — ON 00

NO OO ON r- —

ON r- ON OO -3- t~-

— O NO O N r- — — NO f-

1

-

in

o

.

«n

^t

CN

in

•

in

CN

in

o

CN

—

CN

-

—

CN

NO

in

NO

in

m

NO

—

CN

ro

CN

CN

—

ro

CN

ro

CN

CN

CO

m

NO

r-

NO

NO

m

NO

r~

OO

r-

r-

NO

H3-

ro

H *

ro

ro

ir

•n

it

<n

u-

1t

in

i CN ro ro

i ro —• NO CN

O -*f —• ro U- CN ro CN ro CN — — O N — . — — — —

r o r ^ c N ^ - N O t - ~ - i n N O

ro O CN —< — ' C N — • C N O C N m r o O N N O C N

o O O — i

o — ON

- ^ - H s f O s N O O o r - t ^ O O t -

Nt xr NO - CN o m — CN m

O O O m r o i n c N O

o CN —• r~ NO CN

OO — — — ON — — — ON

ro

O O O C N t ^ O N t — O O O r O — — C N — ' — • —• — C N C N

— r o t — r - o o r ~ t ^ o o N o r ~ o o t ~ -— m CN CN oo r-~ r o c N

i in — T H / - ^ - N O t - - T H - N o i n i n N O ' n N O O o CN — — ro •* o —

1 NO NO

1

1 ON

oo in

NO ON

NO —

1

ON CN

it CN

NO CN

NO CN

it CN

1

in

NO CN

CN

ro CN

in CN

CN

1

r-^ oo

--

ON

OO

O CN

CN

O tN

ON

it — • '

in

*—' CN CN

>n CN

»—H

CN

r-CN

CO CN

1

o CN

ON

It CN

ON

CN

ro

i

o

o CN

ON

1t CN

ON CN

CO

ro

ro

CN

O CN

in CN

o ro

^t

CN

NO

NO

NO CN

CO CN

NO CN

ON CN

O

—

—

—

in CN

o CN

ON CN

oo CN

ON

o CN

OO

in

in

ro CN

it CN

m CN

o ro

it

CN

CN

r-

r-

ro CN

CN CN

CN

o CO

NO

CN

It

ON

ON

in CN

NO CN

ON CN

Tt ro

ro

OO

It

NO

•t

1t

CN

CN

NO CN

ON CN

• *

ro

in

in

-

-

CN

OO

CO CN

oo CN

ro

CN

• *

• *

ro

CO

ro CN

O CN

in CN

oo CN

ro

CN .—H

It

It

It

• ^ -

CN CN

ON

CN CN

CN

CN

i—«

CO

ro

it

it

it CN

CN

it CN

CN

CN

—

ro

ro

in

ON

CN

O CN

in CN

oo CN

ro

CN

• *

it

Htt

• *

CN CN

ON

CN CN

CN

CN

~

CO

ro

ro

ro

ro CN

O CN

in CN

oo CN

ro

CN

it

•<3-

ON

ON

CN

O CN

CO CN

00 CN

ro

NO

it

00

ro

ro

ro CN

CN CN

in CN

00 CN

in

o CN

OO

CN

OO

oo

CN CN

CN

NO CN

ON CN

OO

r-

ON

r-

o

o

oo

ON

o CN

ro CN

1t

m

in

r-

CN

CN

O CN

l>

NO CN

r-CN

oo

CN

r-

ro CN

ro

ro

CN

oo

r-CN

OO CN

r-

o CN

NO

CN CN

~

CO CN

O CN

in CN

oo CN

r»

oo

00

o CN

CN

CN

CN CN

ro CN

it CN

ON

CN

it

in

in

t-

ro co CN — N O r t N O o o i n H - t T t r o c o - * r o H H > - N O O t - ~ i n O N O o o r -— — — — — — — — CN — — — CN — —

3

o es

a:

XI cd

o u PQ

JD

E u O

60 S 3 -a

PQ

60

£ cd

"3 3 o u

E O

J3 O cd C

o

2

cd

J OH

— CN co it <n NO r- oo "Sb'Sb "S> "So '5b "5b '5b '5p -g -~ ~~ — -~ -~ -~ ~

i i i i i A A A.2.2.--..2.2.2.2.2 0 O M M & 0 6 O M 6 0 & 0 t\icacdcdcdc3cdcqcdcdcdcdCTcq.o3tafl>fl>flaa>Hja>Hjqj

Chapter 6

Identification of 'Sib' Plants in Hybrid

Cauliflowers using Microsatellite Markers

This chapter was presented in the 4th International Society of Horticultural Sciences

(ISHS) Symposium on Brassica in Dejeon, South Korea, October 2004. Manuscript of

this chapter was submitted to Theoretical and Applied Genetics and is currently under

review. Paper is presented here in its submitted format.

IDENTIFICATION OF 'SIB' PLANTS IN HYBRID CAULDTLOWERS USING

MICROSATELLITE AND RAPD MARKERS

Ida A. Astarini^, Julie A. PlummerA, Rachel A. Lancaster0 and Guijun Y a n A

APlant Biology, Faculty of Natural and Agricultural Sciences, The University of


BAuthor for correspondence: Ida A y u Astarini

email: [email protected] Fax: +61 8 6488 1108, Phone: +61 8 6488 1992

department of Agriculture, Western Australia, Bunbury District Office, P O Box 1231


Abstract Hybrid cauliflowers have been developed to obtain heterosis and to improve

uniformity of production. T w o breeding systems are commonly employed, self-

incompatibility (SI) and cytoplasmic male sterility (CMS). Sibs, assumed to be self

inbred, often contaminate hybrid seed lots in the SI system and whilst self inbreeding is

not possible in the C M S system, plants that look like sibs occur. The objective of this

study was to develop microsatellite markers for male and female cauliflower parent

lines of both SI and C M S systems and to use them to identify sibs and aberrant plants in

Fi hybrids. Fifty six microsatellite primers were screened and 8 primers produced co-

dominant markers in parent plants and two markers were chosen for purity testing of Fj

hybrid seeds. Controlled pollinations were conducted in the glasshouse to produce

hybrid and selfed-seeds. These seeds were grown in a field trial to identify

morphologically normal and sib plants and to assess the reliability of microsatellite

markers in detecting sibs and aberrant plants. Microsatellite analysis of morphological

sib plants from the SI system revealed that sibs were not always self-inbred, in contrast,

most self inbred plants showed normal growth. Similarly, all morphological sibs from

the C M S system had hybrid bands. This suggests that morphological sibs were not due

to selfing but possibly to an interaction between genetic and environmental factors and

this requires further investigation.

Keywords Brassica oleracea var. botrytis, self-inbred, self-incompatibility,

cytoplasmic male sterility, S S R (simple sequence repeats).

65


Introduction

Identifying breeding lines and determining Fi hybrid purity are important quality

controls in vegetable breeding and seed production. Fi hybrids have been developed to

improve the uniformity of Brassica crops. In cauliflower, Fi hybrid selection aims for

earliness, high yield, better curd quality with regard to compactness and colour, uniform

maturity, and resistance to insects, diseases and unfavourable weather conditions (Crisp

andTapsell, 1993).

T w o systems have been employed for hybrid seed production, cytoplasmic male

sterility ( C M S ) and self-incompatibility (SI). Both systems were developed to prevent

self-pollination. M a n y commercial hybrids of cauliflower utilising SI system have a

substantial proportion of abnormal plants usually known as 'sibs', non-hybrid plants, or

selfing of parent plants. It is assumed that it is possible to identify sibs by their

distinctive plant phenotype and these will be here referred to as morphological sibs.

Compared to the hybrid, morphological sibs can be smaller and have darker green

leaves, have much weaker growth habit or be taller, with paler green leaves. Leaves are

unusually narrow or wavy. Curd size is usually small and not marketable (Holland and

McNeilly 1985). It is therefore necessary to carry out quality control on all hybrid seed

harvests to prevent unacceptable levels of sib seed being released in the market

(Crockett, 2002).

Fi cauliflower hybrid derived from C M S system often produce aberrant plants,

which are unsuitable for harvest. Phenotypes of aberrant plants mainly involve the

modification of three characters: leaf shape, size and thickness (Ruffio-Chable et al.

2000; Fujime and Okuda 1996). In cauliflower varieties, the proportion of aberration

ranges from 5 % to 4 0 % . The main concern is the economic consequence for growers.

M a n y attempts have been made to identify and screen out morphological sib or

aberrant plants in hybrid populations. Automated phenotypic examination such as

image analysis on Brussels sprout and cabbage seedlings, have been developed

(Fitzgerald 1997). This technique is not a universal approach as morphological

differences m a y be difficult to detect and are affected by environmental factors.

It is assumed that sibs are self inbred and therefore are genetically determined

and could be identified by genetic markers. Here plants from a self inbred parent will

be referred to as genetic sibs. Isoenzymes of acid phosphatase using P A G E can detect

genetic sib content in Fi hybrid Brussels sprout varieties (Harvey and Smith 1987).

Isozymes are also used in seed purity testing of Chinese cabbage (Brassica campestris)

66

(Zheng and Liu 1994). There are some limitations to isozyme analysis, such as

insufficient polymorphisms among closely related genotypes, and variation affected by

environmental factors, seed vigour and growing stage (Meng et al. 1998). Ploidy levels

have been examined by flow cytometry, but they do not reveal any correlation between

ploidy level (haploid, diploid or tetraploid) and abnormalities observed in plants

(Ruffio-Chable et al. 2000).

Molecular techniques such as R A P D have been employed to determine genetic

purity in Brassica oleracea crops such as cauliflower (Boury et al. 1992), broccoli

(Crockett et al. 2002) and cabbage (Crockett et al. 2000) and Brassica campestris such

as Chinese cabbage (Meng et al. 1998). R A P D P C R was successfully employed to

identify female and male parent lines and detect genetic sibs in cabbage and broccoli

(Crockett et al. 2000; 2002). The proportion of genetic sib contamination was similar to

that of morphological sib observed in the field trial.

In recent years microsatellites, also referred to as simple sequence repeats

(SSR), have gained increasing importance in plant genetics and breeding. High

abundance and extensive polymorphism make them an ideal marker system for genetic

mapping and characterization of germplasm, in particular in very closely related and

inbreeding species (Saal et al. 2001). Microsatellites are highly variable with regard to

repeat number and show co-dominant inheritance. Therefore they are considered

suitable for use in genetic purity testing in hybrid populations of crops such as

cauliflower.

Microsatellites of Brassica species are well documented. A large number of

microsatellites from rapeseed canola (Brassica napus) have been identified and

characterized. M a n y B. napus microsatellite flanking primer pairs are functional in the

A and C genome species within the genus Brassica, but are not useful as markers for a

wide range of species in the family Brassicacea (Saal et al. 2001). However, some

primers m a y be useful as markers for cauliflower, which has C genomes.

The hypothesis in this study was that morphological sibs are equal to genetic

sibs, a view held by both the farming and scientific community. The aim of this study

was to generate R A P D and microsatellite markers for discrimination of cauliflower

parent lines and the subsequent identification of genetic sibs in Fi hybrid cauliflower.

Materials and methods

Plant materials

67

A commercial cultivar 'Discovery' was provided by South Pacific Seeds Pty Ltd. This

hybrid cultivar is produced using the SI system. T w o pairs of parent lines and their

hybrid progenies were provided by Henderson Seeds Pty Ltd. One pair utilised the SI

system (5038 SI) and another pair utilised the C M S system (5038 C M S ) . Seeds were

stored at 4°C until required.

Field Trial 1

Plants of 'Discovery' were grown under standard commercial field conditions at the

Horticultural Research Institute, Western Australian Department of Agriculture at

Manjimup, from September to December 2001. Leaf morphology and curd weight were

observed. Leaf samples were collected from 10 normal plants and all sib plants at

harvest and were stored at -20°C until required for R A P D analysis.

Controlled pollination in glasshouse grown plants

Cauliflower seeds supplied by Henderson Seed were germinated using the top of paper

method (ISTA 2003). Twenty seeds of each parent line were germinated on top of filter

paper (2 x Whatman N o 1) in 15 cm petri dishes filled with 15 ml distilled water. Seed

were germinated in a 20°C growth chamber with a 12h/12h light/dark regime for 7 days.

Germinated seeds were transferred to a glasshouse at day 7. Ten vigorous,

uniform seedlings were chosen. The seedlings were maintained in 100 m m diameter

pots for 6 weeks and then transferred to 255 m m diameter pots until mature. Plants

were regularly watered and fertilized. Glasshouse temperature ranged from 20°C to

27°C. N o mature plants were morphological sibs.

Crossing was done on both SI and C M S pairs. One hundred flowers from 4

plants were cross-pollinated for each pair. Flowers were bagged after pollination and

seed set was recorded. Selfing is difficult in SI female parent lines since plants are self

incompatible. T w o methods for overcoming self incompatibility in SI systems were

investigated. In the first method, selfing was performed on the 5038 SI line, 10 min

after spraying both stigma and stamens of open flowers with 5 % salt (NaCl) solution

(Fu et al. 1992). The second method used bud pollination, where immature, non-

receptive stigmas (1-2 days prior to anthesis) were pollinated with mature pollen of the

same plant (Hallidri and Pertena 2002).

To induce pollen fertility in C M S female parent lines, four plants at flower bud

stage were kept in a 30°C growth chamber for two weeks. T w o plants were maintained

in the glasshouse as controls. Plants were self pollinated under both conditions. Flower

68

morphology and pollen production was observed. Pollen viability was examined using

the fluorescein diacetate staining method (Heslop-Harrison and Heslop-Harrison 1970).

Counts of pollen were made under the microscope. All seed produced from crossing

and selfing were collected and stored at room temperature until required for Field Trial

2. Leaf samples of female and male parents of SI and C M S lines were collected for

molecular analysis.

Field Trial 2

Parent lines, hybrid seeds and plants from controlled pollination (crossed and selfed

seeds obtained from the glasshouse) were grown at a seedling nursery in Manjimup.

Seed was sown manually, one seed per cell and each line was allocated to a separate

tray. After 6 weeks, seedlings were transplanted to the field at the Manjimup

Horticultural Research Institute, Western Australian Department of Agriculture, at a

spacing of 40 c m between plants within a row and 80 c m between rows. They were

grown under standard commercial field conditions.

Plants were observed at 2, 9 and at harvest at 13 weeks after transplanting.

Observations included normal and unusual characteristics of leaves, plant height, leaf

number and curd weight of normal and morphological sib plants. The difference

between normal and sibs were analysed using paired two sample t-XesX on the mean

values of plants from different lines using Genstat 7.0. Leaf samples of normal and

morphological sib plants were collected and stored at -20°C for molecular analysis.

Molecular analysis

T w o types of molecular markers, R A P D and microsatellite were examined for their

suitability in testing for parent lines, self and hybrid plants.

RAPD analysis

Twelve normal hybrids and 12 sibs of 'Discovery' were chosen for D N A extraction.

D N A was extracted following the C T A B method (Astarini et al. 2004). R A P D P C R

was performed following the method described in Astarini et al. (2004).

Microsatellite analysis

First step in microsatellite analysis was to find markers for parent lines. Eight

individual plants of each parent line were chosen for this purpose. Fifty six

microsatellite primers suitable for Brassica C genome were screened for P C R

69

amplification. Canola Breeders Western Australia (Perth) provided 50 primers and 6

other primers were synthesized by Life Technologies Inc. (Madison) customer primer

program. The next step, specific markers that able to distinguish parent lines was then

tested on hybrids, self inbred and morphological sib plants to differentiate true hybrids

and genetic sibs. Twenty normal plants of each line, self-inbred and all morphological

sibs were examined. The P C R reaction was performed in a final volume of 20 uL

containing injection water, 10 uL AmpliTaq Gold P C R Master Mix (Applied

Biosystems, Foster City, C A ) , 0.5 u M of each forward and reverse primer and 40 ng

template D N A .

P C R was performed in an iCycler ™ (Bio-Rad, Hercules, C A ) using the

following cycling program : 1 cycle of 5 min at 94°C, 40 cycles of denaturation at 94°C

for 1 min, annealing at 53°C for 2 min, elongation at 72°C for 2 min, and finally 1 cycle

of elongation step at 72°C for 10 min. The iCycler was programmed to retain the

samples at 4°C until collected and stored at -20°C.

Each sample of P C R product (10 uL) was mixed with 6x gel loading buffer (1

uL) and loaded onto an agarose ( 4 % w/v) gel for electrophoresis (Bio-Rad, Hercules,

C A ) in lx T A E buffer at 100 volts for 2 hours. Fifty times T A E buffer contained 242 g

Tris base, 57.1 g glacial acetic acid, 100 m L 0.5 M E D T A and distilled water to 1 L. A

100 bp D N A ladder (5 uL) and 1 uL of gel loading buffer (Promega, Madison), was

included on both sides of the gel as a molecular standard. Ethidium bromide solution (2

uL/100 m L ) was incorporated into the gel. Amplification products separated by gels

were then photographed under U V light using a digital camera (Kodak D C 120) and the

images were recorded with a Macintosh Kodak ID 2.0 computer program.

Results

Controlled pollination

Crosses were 1 0 0 % successful in both pairs of the C M S and SI parent lines. Some

fertilisation (16%) occurred following selfing of the SI female line using bud

pollination, while salt spray treatment only produced empty pods with no seed.

T w o types of female C M S parent line plants were observed. The first type, had

flowers with reduced and shrunken stamens (Fig. la). A n average of 5.3 ± 0.5

pollen/microscope field was counted, and 1 2 % pollen was viable. The second type, had

flowers with petaloid stamens (Fig. lb). N o pollen was observed in anthers of this

flower type and so self pollination was not possible. Non-CMS plants had flowers with

70

abundant fertile pollen, 46 ± 2.8 pollen/microscope field, and 8 8 % of pollen was viable

(Fig. lc).

Maintaining the female C M S line in the 30°C growth chamber did not induce

pollen fertility. Pollen viability decreased from 1 2 % to 5 % with high temperature

treatment. N o seed formation was observed on any C M S plants.

Field Trial 1

O n 'Discovery', 1.2% sibs were observed. Average curd weight of normal plants was

1148 ± 13 g, compared to 101 ± 22 g on sib plants.

Field Trial 2

Two weeks after transplanting (8 week-old seedlings), abnormal seedlings were

distinguishable from normal seedlings by their unusual morphology. Abnormal

seedlings observed in the field had wavy leaves (Fig. 2a), narrow leaves (Fig. 2b), or

blind apexes (Fig. 2c). Abnormal plants with wavy leaves or narrow leaves were noted

as morphological sibs.

At 9 weeks after transplanting, average plant height and leaf number were vary

between lines (Table 1). Significance differences were found between normal plants

and morphological sibs on plant height (p value < 0.001) and leaf number (p value =

0.014).

At harvest, all morphological sibs observed at 8-week old seedlings (2 weeks

after transplanting) continued to have unusual vegetative growth and produced small

curds. Sib plants produced very small curds, below 200 g, compared to more than 700 g

of normal hybrid plants (Table 1). Most self-inbred plants (92%) exhibited normal

growth in the field and produced curd around 417 g. Morphological sibs found in all

lines tested except male SI parent, with the proportion of 1.3% to 8%.

RAPD analysis

N o polymorphic bands were obtained from 36 primers tested during R A P D analysis of

'Discovery'. The same banding patterns were observed on normal and sib plants grown

in Field Trial 1.

Microsatellite analysis

Out of 56 microsatellite primers screened, 8 primers had polymorphic bands.

Furthermore, only two primers were suitable as specific markers for both 5038 C M S

71

and 5038 SI. These primers, Nal2-E06b (Forward: C A T A T A G G G G A A T C A T C A T

C G G C , Reverse: A G A C C A A T T A G C A T C T C G C C ) and O110-G08 (Forward: T G C T

T A A T T G A T T A G G G C A G , Reverse: T T A C C T C A T C A G G T G G A G G C ) showed

distinct co-dominant bands between female and male parents.

Microsatellite analysis of normal Fi hybrids and manual crosses from the SI and

C M S systems had both female and male bands present, confirming they were all true

hybrids. All self-inbred plants only had the female band indicating they were genuinely

self-inbred (Fig. 3). Microsatellite analysis on morphological sibs on parent plants

revealed no genetic sibs on the parent plants.

Microsatellite analysis of morphological sib plants from the Fi hybrids and

manual crosses of the SI system indicated that 3 3 % and 3 8 % plants respectively (Table

1), only had the female band, and were therefore genetically self inbred, whilst the

remainder had both female and male bands. All morphological sibs in Fi hybrids and

crosses from the C M S system had male and female bands (Fig. 3).

Discussions

It has been assumed that abnormal plants in Fi hybrid utilising SI systems were due by

self-inbreeding and were known as sibs. There are up to 1 4 % sibs in cabbage (Crocket

et al 2000) and 45%> sibs in broccoli (Crocket et al 2002). Our studies show that plants

with sib characteristics occurred not only in Fi hybrid lines, but also in parents and

selfed lines. Morphological sib plants varied from 1.2% in 'Discovery' to 8 % in

manually crossed 5038 SI and selfed 5038 SI. This proved that sibs are not always self

inbred.

Barriers to selfing were not overcome in the C M S line and so no selfed seeds

were produced. Exposure to 30°C did not induce fertility of pollen on C M S cauliflower

lines, which means the C M S line was stable at high temperatures. C M S lines in

Brassica are derived from a number of systems such as 'Pol', 'Ctr', 'Nap', 'Ogura',

Anand and 'Nigra' (Makaroff, 1995). Stability of pollen sterility in C M S lines depends

on their systems. 'Ctr' and 'Nap' are temperature unstable, 'Pol' is relatively stable but

can be broken down at 24°-30°C while 'Ogura' and 'Nigra' have very stable sterility.

The cauliflowers C M S lines used here may be derived from a very stable system or may

need a lower temperature range to break pollen sterility (Ruffio-Chable et al. 1993).

Unusual flower morphology is often observed in C M S plants (Cardi and Earle

1997). In the cauliflower C M S plants grown here, two distinct types were observed

within one line, flowers with petaloid stamens and flowers with shrunken stamens (Fig

72

1). Petaloid stamens are also found in B. oleracea (Cardi and Earle 1997) and B. juncea

(Malik et al. 1999). The petaloid stamens did not produce any pollen, so could not

produce any self-inbred plants. The shrunken stamens had dehiscent anthers and

produced pollen with 1 2 % viability. Only a small amount of pollen was produced, an

average of 5 pollen/microscope field, compared to 45 pollen/microscopic field in non-

C M S plants. This indicated that although the chances were small, it may be possible for

C M S parent plants to produce self-inbred plants as not all flowers were sterile.

The C M S breeding system was developed to eliminate the possibility of sibs or

inbred seed being produced, as found in less efficient systems, such as SI. However

plants with unusual growth were observed in Fi hybrid cauliflowers derived from the

C M S system. Our glasshouse, field trial and microsatellite examination confirmed that

these abnormalities were not due to self-inbreeding. Abnormal growth may require an

environmental trigger. Ruffio-Chable et al. (2000) suggest the abnormality in C M S

hybrids is due to the occurrence of epigenetic phenomena during plant development

causing modifications in gene expression. The causes of these abnormalities should be

further investigated.

Self-incompatibility mechanisms were overcome. Self incompatibility

mechanisms are often undeveloped at the bud stage in B. oleracea (Crisp and Tapsell

1993). Here placing mature pollen on immature stigmas overcame the incompatibility

mechanisms in cauliflower and allowed successful pollination, fertilization and

production of viable seeds. This enabled analysis of self inbred plants from female SI

lines of cauliflowers.

Overcoming self incompatibility with 5 % salt spray 10 minutes before selfing

was not successful in cauliflower (B. oleracea). The same or a similar technique was

suitable for overcoming incompatibility in B. napus (Fu et al. 1992) and B. campestris

(Monteiro et al. 1988). Salt spray inactivates the incompatibility substance on the

stigmatic surface and allows pollen to germinate and penetrate the stigma and style

resulting in fertilization and seed formation in other Brassica species. Further

investigation on the appropriate stage at which flowers should be sprayed and

optimising salt concentration may improve pollination success in cauliflower.

Unusual growth habits in cauliflower were easily distinguished from normal

growth in 8-weeks old plants in the field (2 weeks after transplanting). Particular

abnormal growth types, such as plants with a blind apex or wavy leaves, were obvious

during transplanting at 6 weeks. Early detection and removal of abnormal plants would

be useful to avoid further profit loss due to production costs and labour costs at harvest.

73

Growers m a y be able to detect and discard abnormal seedlings during transplanting at 6

weeks.

Self-pollinated SI plants produced the same height and leaf shape as Fi hybrid

plants and produced marketable size curds. Morphologically, they could not be

distinguished from normal hybrid plants. Thus abnormalities in cauliflower hybrid are

not caused by self-inbreeding, which has been assumed to be the cause of

morphological sibs.

All R A P D primers trialed produced clear and scorable bands, indicating the

technique was easy to conduct. R A P D primers, including several of those examined

produce polymorphism in other Brassica crops including Chinese cabbage {Brassica

rapa) (Meng et al. 1998). This technique could also identify and distinguish sibs in

cabbage {B. oleracea var. capitata) (Crockett et al. 2000) and broccoli (B. oleracea var.

italica) (Crockett et al. 2002). This technique however was not able to generate markers

for genetic sibs in the Fi hybrid cauliflower 'Discovery'.

Parent lines were not available for R A P D analysis of 'Discovery', therefore it

was difficult to distinguish true hybrids from self inbred plants as there was no genetic

information on the parents. It is useful to develop genetic markers for the parent lines

prior to identification of self-inbred plants from their progeny.

R A P D is a dominant marker and employs random primers (Henry, 1997). This

marker m a y not be suitable for distinguishing specific male and female bands, which are

usually co-dominant (Saal et al. 2001), especially when parent lines are closely related.

It was therefore necessary to investigate co-dominant markers, such as microsatellites.

Microsatellite analysis was a powerful technique which could be used to

distinguish specific male and female bands and to detect true hybrids and self-inbred

plants. Difficulties m a y be encountered in screening for appropriate primers but the

selection process was reduced by choosing primers from a closely related species with

similar genome. There are hundreds of Brassica microsatellites publicly available

(Bond et al. 2004) which can be used in Brassica vegetables.

In cabbage and broccoli, Crockett et al. (2000; 2002) found a similar proportion

of morphological sibs (abnormal plants observed in the field) and genetic sibs (detected

using R A P D markers). They therefore assumed that self inbreeding was the genetic

cause of morphologically identifiable sibs. In this experiment, morphological sibs

identified in the field were not the same plants as genetic sibs as revealed by

microsatellite analysis (Fig. 3). Thus self-inbreeding was not the only cause of

morphological sibs in cauliflower production.

74

In conclusion, microsatellite analysis was a reliable technique for identification

of parent line, hybrid and self-inbred plants. The lack of uniformity in hybrid

cauliflowers observed in the field was not solely due to contamination by self inbred

plants (genetic sibs). Self inbred parents did produce seed which grew into normal

plants. Morphological sibs were not all self-inbred. Thus the small size plants usually

referred to as sibs were not only from self-inbred parent lines. Environmental triggers

or other genes may be involved in this abnormal plant formation. The molecular

evidence that sib is not the only cause of contamination in hybrid cauliflower is new and

is further supported by the evidence that controlled self-pollinated plants exhibited

normal growth in the field.

Acknowledgments W e thank AusAID for providing a scholarship to Ida Astarini.

Thanks to South Pacific Seeds and Henderson Seeds for providing seeds. Sincere

thanks to M s Anouska Cousins, Dr Matthew Nelson and Associate Professor Wallace

Cowling for technical advice and provision of primers, M r John Doust, Dr Kristen

Stirling, M r David Tooke and M r Grazi Giadresco for invaluable assistance during the

field trial at Manjimup. Financial assistance from Horticulture Australia Limtd and

Department of Agriculture Western Australia are gratefully acknowledged.

References

Astarini IA, Plummer JA, Lancaster R A , Yan G (2004) Fingerprinting of cauliflower

cultivars using R A P D markers. Aust J Agric Res 55:117-124.

Bond JM, M o g g RJ, Squire G R, Johnstone C (2004) Microsatellite amplification in

Brassica napus cultivars: Cultivar variability and relationship to a long-term feral

population. Euphytica 139:173-178.

Boury S, Lutz I, Gavalda M-C, Guidet F, Schlesser A (1992) Genetic fingerprinting in

cauliflower by the R A P D method and determination of the level of inbreeding in a

set of Fi hybrid seeds. Agronomie 12:669-681.

Cardi T and Earle E D (1997) Production of new C M S Brassica oleracea by transfer of

'Anand' cytoplasm from B. rapa through protoplast fusion. Theor Appl Genet

94:204-212

Crisp P and Tapsell C R (1993) Cauliflower, Brassica oleracea L. In: Kalloo G, Bergh

B O (Eds). Genetic Improvement of Vegetables Crops. Pergamon Press, Oxford.

Crockett PA, Bhalla PL, Lee CK, Singh M B (2000) R A P D analysis of seed purity in a

commercial hybrid cabbage {Brassica oleraceae var. capitata) cultivar. Genome

43:317-321.

75

Crockett PA, Singh M B , Lee CK, Bhalla PL (2002) Genetic purity analysis of hybrid

broccoli {Brassica oleracea var. italica) seeds using R A P D PCR. Aust J Agric Res

53:51-54.

Fitzgerald D M , Barry D, Dawson PR, Cassells A C (1997) The application of image

analysis in determining sib proportion and aberrant characterization in Fj hybrid

Brassica population. Seed Sci Technol 25:503-509.

Fu T, Ping S, Xiaoniu Y, Guangsheng Y (1992) Overcoming self-incompatibility of

Brassica napus by salt (NaCl) spray. Plant Breeding 109:255-258.

Fujime Y, Okuda N (1996) The physiology of flowering in Brassicas, especially about

cauliflower and broccoli. Acta Hort 407:247-254.

Hallidri M , Pertena D (2002) Self-incompatibility test in cabbage {B. oleracea var

capitata). Acta Hort 579:117-122.

Harvey E, Smith B M (1987) A recent survey of sib content in Fi hybrid Brussel sprout

varieties. Cruciferae Newsl 12:122-123.

Henry RJ (1997) Practical Applications of Plant Molecular Biology. Chapman and

Hall, London.

Heslop-Harrison J, Heslop-Harrison Y (1970) Evaluation of pollen viability by

enzymatically-induced fluorescence; intracellular hydrolysis of fluorescein

diacetate. Stain Tech 45:115-120.

Holland RL, McNeilly T (1985) Genotype environment interaction and sib content in

Fi hybrid Brussels sprouts. Euphytica 34:371-376.

International Seed Testing Association (ISTA) (2003) International Rules for Seed

Testing. Bassersdorf, CH-Switzerland.

Makaroff C A (1995) Cytoplasmic male sterility in Brassica species. In Levings III CS,

Vasil IK (Eds). The Molecular Biology of Plant Mitochondria. Kluwer Academic

Publ. London.

Malik M , Vyas P, Rangaswamy N S , Shivanna K R (1999) Development of two new

cytoplasmic male-sterile lines in B. juncea through wide hybridization. Plant

Breeding 118:75-78.

Meng X, M a H, Zhang W , Wang D (1998) A fast procedure for genetic purity

determination of head Chinese cabbage purity seed based on R A P D markers. Seed

Sci Technol 26:829-833.

Monteiro A A , Gabelman W H , William P H (1988) Use of sodium chloride solution to

overcome self-incompatibility in Brassica campestris. HortSci 23:876-877.

76

Ruffio-Chable V, Chatelet P, Thomas G (2000) Developmental^ "Aberrant" plants in

Fi hybrids of Brassica oleracea. Acta Hort 539:89-94.

Ruffio-Chable V, Bellis H and Herve Y (1993) A dominant gene for male sterility in

cauliflower {Brassica oleracea var botrytis): phenotype expression, inheritance, and

use in FI hybrid production. Euphytica 67:9-17.

Saal B, Plieske J, Quiros C, Struss D (2001) Microsatellite markers for genome

analysis in Brassica. II. Assignment of rapeseed microsatellites to the A and C

genomes and genetic mapping in Brassica oleracea L. Theor Appl Genet 102:695-

699.

Zheng X Y , Liu Y (1994) Inbred testing of Chinese cabbage Fi varieties by peroxidase

and esterase isozyme analysis. Acta Hort Sin 21:65-70.

77

cu

*3

OS

co X

O

*3 CD

o o o co cd

+3 " cu

c CU OC

C)-H

O <D

3 3

c o o a o u

a cu

x *3

co X)

cd o '5b o "o X

e-o

<4-l

o u X

E 3 3

15

Hi

c

a CN

"cd E -g

u

CO



c _cd

a, o S-CD X E 3

2 vH

4>

3 et H

x .5? '53 X H->

s cd

5, c o '•S o a o s-O H

CO

cd

CO H — >

3 cd

cd

O 3

c o co X 'co o '-M

CU 3 CU bo O s

PU -2 .FJ o

o. o fcH

PU * c CD co cu s-

o. CO

cd *3

X

cd

CM



CH-H

o 3 O

"•6 o a o s-OH

* *

*3 cu +-1 CO

cu

I O. O

C/.

-3 CM



3 cu o (-<u OH

cd CO

cd

-a cu 1/5 CO

cu s-O H

X CU

33 *

M C/5

.£f

s s a

X3

C/3

s

s « •NM

Sf '3 a a

g, a

•-C cu 0

o

* * JO

55

et

S u

o

1-5 £

CU

«s i U

<u 3 O

CU 3 O

2 in CN

-H

cu 3 O

Z

m o -H

cu 3 O

2

00

o -H m

o

3 cu Cd OH

2

-H

CN

00

-H ô in m

in

-H CN

OO

o -H

CN

-H

in CN -H oo

o in

3 cu t-

cd OH M

S cn

oo

-H

00 m -H oo 00

m o

-H >n

©

-H oo

-H cn m

CN -H cn oo

cn cn

0 s

^*

CN

O

o cn

c/. -3

'E X >. X

2 CN

NHO

, - H < ^

+ in cn

-H l>

o

m

o -H

>n

o -H N H O

M3 ,-H

— CN

-H

m

oo

-H c^

OO cn

oo

00

o o

CN •N©

o

CN

-H cn CN

in cn

in

-H

O

o cn .-H

oN O

o

00

Ô Of

T3 NV*) w T3

o 2=1 1- CD

cn

-H

ON

-H -H !>• O •—I in '*• cn

CN

-H

o -H

in

o -H m

cn CN

-H

!=H <=>

""-1-

>n CN

-H o C?\ 00

o

in

o -H

oo CN

CN

-H

in

CN

-1 CN —

CN

-H in cn

-H o '-t oo

-H

oo o -H oo

o -H

in

cn CN

-H Os

o -H

m 'St

-H '-t oo

00

r-H t>

-H '=t

m o -H oo

00 '-t

-H '-t m

in

o -H Os

NT in

in

o m

H-> 3 CU cd O, 00

s O Ô

o in

H-l

3 CU Cd OH

OO

2 u Of

O O cn

t/.

2 U -3 'E X

X 1

& H

O

o

CZJ

u co cu co C/5 O s-

u

Fig 1 Reproductive organs of two types of C M S flowers compared to normal flowers (petals removed), a = shrunken anthers, b = petaloid anthers, c = normal anthers.

Fig 2 Abnormal plant types (8-weeks old) observed in the field, a = wavy leaf (W), b = narrow leaf (N), c = blind apex, d = normal

* * • • -

ML S S 9 ? H H H* H* C C C* C* S S S S* ML

300 bp

200 bp

Fig. 3 Banding patterns of male parent (6"), female parent (°), Fi hybrids (H), manual crosses (C) and self pollinated plants (S) using primer Nal2-E06b on SI system plants. M L = Marker Ladder. *= abnormal plant such that H*= abnormal hybrid, C*= abnormal crosses, S*= abnormal selfed.

79

Chapter 7

Molecular Markers Correlated with Seedling

Traits in Cauliflower Varieties

The manuscript of this chapter was submitted to the Australian Journal of Agricultural

Research on 14 November 2005 and is presented here as the submitted format.

Molecular markers correlated with seedling traits in cauliflower

varieties

Ida A. AstariniAB, Julie A. P l u m m e t , Rachel A. Lancaster* and Guijun Y a n A

APlant Biology, Faculty of Natural and Agricultural Sciences, The University of


Correspondence: Ida Astarini [email protected] Fax: +61 8 6488 1108

cDepartment of Agriculture, Western Australia, Bunbury District Office, P O Box 1231


Abstract. Cauliflower production is hindered by variation in curd quality and maturity.

Morphological variation from seed to harvest is due to genetic variation interacting with

environmental conditions and here the genetic factors were investigated. The aim of

this study was to search for D N A markers linked to seedling traits, facilitating earlier

selection for cauliflower production. Cauliflower seed lines were germinated in Petri

dishes (20°C, 7 days) and seedlings were transferred to pots and grown under

glasshouse conditions. Seed weight and various seedling characters were measured

until harvest at 6 weeks. D N A was extracted using C T A B method and R A P D markers

were identified using 17 primers. Multivariate analysis based on principle coordinates

analysis was employed to correlate morphological traits with molecular markers across

cultivars. Markers associated with seed weight, germination rate, shoot length, root

length, fresh weight and dry weight were identified.

Additional key words: Brassica oleracea var. botrytis, R A P D markers.

Introduction

Western Australia produces 8 5 % of Australia's export cauliflowers and to remain

internationally competitive, growers need to improve the uniformity of curd maturity

and size (Stirling and Lancaster 2005). Curds must be picked within one day of

maturity and variation in harvest time increases the number of picks, which is labour

intensive and hinders the use of harvest machinery. These variations are determined by

genetic and environmental factors. Several methods can be employed to reduce

variation including screening for uniform seed weight and size and management

practices. Agronomic studies, such as increasing plant density and irrigation application

(Stirling and Lancaster 2005), have partially reduced variation but genetic approaches

are also required.

81


Variation begins at the seed and seedling stage and these differences in growth

are exacerbated during field production. Elimination of variation at the seed or seedling

stage would greatly enhance uniformity of the crop. Identification of specific molecular

markers associated with morphological traits would greatly assist in the selection of

superior, uniform seedlings.

Nowadays, molecular marker techniques have become practical for cultivar

identification and selection for breeding purposes (Hu and Quiros 1991). The use of

molecular markers for morphological traits will assist in removing undesirable traits

such as small or malformed seedlings and assist in the selection of more vigorous and

uniform seedlings. A correlation between morphological traits and molecular markers

has been achieved for a few species. Leaf shape, period to bolting and leaf hairiness in

Brassica campestris have been linked with R A P D markers by Q T L and linkage

analyses (Nozaki et al. 1997). Somers et al. (2001) identified a major gene for yellow

seed coat colour in canola linked to R A P D markers. Hirai et al. (2004) found two

R A P D markers linked to clubroot resistance in Brassica rapa. In radish, morphological

traits correlate with R A P D markers both within and across cultivars (Pradhan et al.

2004).

Molecular markers associated with seedlings traits are potentially useful in

seedling selection for interesting traits. Selection of seedlings very early in their

development for the desired traits m a y provide early selection of desirable seedlings

traits for nursery production and field establishment.

The aim of this study was to search for D N A markers correlated to seedling

traits, facilitating early selection in cauliflower production. Molecular markers

associated with morphological traits will potentially contribute to a wide range of

research. This can provide complementary information in the development of improved

hybrids.

Materials and Methods

Plant materials

Twenty one cauliflower {Brassica oleracea) cultivars were utilized (Table 1). Cultivars

varied in their characteristics and origin; some were produced from C M S (cytoplasmic

male sterility), SI (self incompatibility) or O P (open pollinated) breeding systems, they

included summer, autumn, winter and spring cultivars, they had mini, green or white

curds, some were well-covered curds, they had a short or long period to maturity and

they were sourced from different seed companies. Seeds were provided by Bejo,

82

Syngenta, Enza Zaden (formerly Yates Seeds), South Pacific Seeds and Henderson

Seeds.

Seed germination

Individual seed weight was recorded before sowing. Seeds (25) were sown in four

replicates of 9 cm Petri dishes, on 2 layers of Whatman N o 1 filter paper soaked with 10

ml distilled water (ISTA 2003). Petri dishes were sealed with Parafilm to minimise

evaporation. Seeds were kept at 20°C in a growth chamber under a 12h light/12h dark

regime for 7 days. Seed germination was monitored on a daily basis. Radicle tip

emergence indicated the time of germination for each individual seed. Germination rate

was measured as 1/t, where t was the germination time. Individual seedling vigour was

measured by root length, shoot length, total length and fresh weight on 7 day-old

seedlings.

Seedling growth

Seven-day old seedlings, 5 of each cultivar were transferred from petri dishes to 100

m m pots containing a mix of composted bark, coco peat and river sand (2:1:1).

Seedlings were maintained in the glasshouse for 6 weeks (20-27°C, 7 5 % RH).

Seedlings were watered daily and fertilized with a complete fertilizer (Thrive®) once a

week. Length of the longest leaf (measure from where the leaf joints the stem to leaf

tip) and number of leaves were measured every 2 weeks. Shoot and root length, fresh

weight and dry weight were measured when plants were harvested at 6 weeks.

RAPD analysis

For each cultivar, leaves from 4 plants (approximately 1 mg, F W ) were pooled for D N A

extraction. Genomic D N A was isolated using C T A B method as described by Astarini et

al. (2004).

Twenty arbitrary decamer primers were examined for P C R amplification. All

primers were synthesized by Life Technologies Inc. (Madison) customer primer

program. The P C R reaction was performed in a final volume of 25 uL containing

injection water, lx Taq polymerase buffer, 1.5 units of Taq polymerase, 0.05 m M of

each dNTP (dATP, dCTP, dGTP, dTTP) (Promega, Madison), 1 u M of primer, 1.5 m M

MgCl 2 and 40 ng template D N A . A negative P C R tube containing all components

except genomic D N A was used with each primer to check for contamination. P C R was

performed in an iCycler ™ (Bio-Rad, Hercules, C A ) using the following cycling

program: 10 x 5 s cycles of denaturation at 94°C, annealing at 35°C for 30 s, elongation

at 72°C for 1 min, 25 x 5 s cycles of denaturation at 94°C, annealing at 45°C for 30 s,

83

elongation at 72°C for 1 min and finally 1 cycle included an elongation step at 72°C for

2 min. The iCycler was programmed to retain the samples at 4°C until they were

collected and stored at -20°C.

Each sample of R A P D products (10 uL) was mixed with 6x gel loading buffer

(2 uL) and loaded onto an agarose (1.5 % w/v) gel for electrophoresis (Bio-Rad,

Hercules, C A ) in lx T A E buffer (50x T A E buffer contained 242 g Tris base, 57.1 g

glacial acetic acid, 100 m L 0.5 M E D T A and distilled water to 1 L) at 80 volt for 1 hour

30 min. A 100 bp D N A ladder (Promega, Madison, 5 uL of D N A ladder and 1 uL of

gel loading buffer) was included on both sides as a molecular standard. Ethidium

bromide solution (2 uL Etbr/100 m L lx T A E buffer) was incorporated in the agarose

solution. Gel photographs were taken under U V light using a digital camera (Kodak

D C 120) and the images were recorded with a Macintosh Kodak ID 2.0 computer

program.

Statistical analysis

Bands produced on gels were scored as 0 (absent) and 1 (present). Data set were

analysed using Genstat 7.0 software (Digby et al. 1989). Multivariate analysis based on

principal coordinate analysis was employed to generate -values. The program was run

with both phenotypic and genotypic data sets. The program produced associations

between morphological traits and molecular markers. Strong associations of markers

with traits were identified with high t-values. P-values were calculated based on t-

values and degree of freedom (d.f.) to indicate strength of association. Morphological

traits included in the analysis were seed weight, germination rate, shoot length (1, 2, 4, 6

weeks), root length (1, 6 weeks), total length (1 week), fresh weight (1, 6 weeks),

number of leaves (2, 4, 6 weeks) and dry weight (6 weeks). Correlations between

morphological traits were determined, presented as correlation coefficient (r).

Results

Morphological variation across cultivars

Morphological traits varied across cultivars. Average seed weight ranged from 2.5 m g

in 'Atlantis' to 5.8 m g in 'Delfur' (Table 2). Average germination rate ranged from

0.33 d"1 in 'Arctic' to 0.88 d"1 in 'Belot' and 'Panther'. Overall average of germination

rate was 0.55 d"1 with 1 0 0 % germination of all cultivars within 4 days.

Seedling vigour at 7 days varied across cultivars (Fig. la). Average shoot length

ranged from 1.8 c m to 3.7 cm, 'Morpheus' had the shortest shoots and 'Belot' the

longest. Average root length ranged from 2.6 c m in 'Arctic' and 6.4 cm in 'Megan'.

84

'L3389' had the highest fresh weight of 74 mg, while 'Atlantis' was less than a third of

this (23 m g ) (Table 2).

Seedling vigour measured at 6 weeks indicated the tallest plants were 'Omeo', at

21 cm, while 'Megan' plants only half the length (14.5 cm, Fig lb). 'Plana' had the

highest fresh weight (31.2 g) and dry weight (3.7 g), while 'Belot' had the lowest fresh

weight (11.2 g) and 'Discovery' and 'Panther' had the lowest dry weight (0.8 g).

Average leaf number ranged from 5.8 in 'Donner' to 9 in 'Plana' (Table 2).

Correlations between seed weight and other traits across cultivars

Seed weight was positively correlated with most seedling vigour parameters at day 7 but

not by week 6. The strongest correlation was with fresh weight on day 1 {r = 0.6)

followed by number of leaves at week 4 {r = 0.4). Other traits had weak correlations

with seed weight. These ranged from r = 0.06 for shoot length at week 2 to r = 0.36 for

total length at day 7. There was no correlations found between seed weight and

germination rate, shoot length, leaf number, fresh weight and dry weight at 6 weeks,

where r close to 0 (Table 3).

Correlations among seedling traits across cultivars

Germination rate was positively correlated with shoot length, root length, total length

and fresh weight at day 7 and root length at week 6 (Table 3). Shoot length at day 7 had

a moderate correlation with total length (r = 0.57) and fresh weight (r = 0.44) at day 7.

Overall shoottotal length ratio at day 7 was 1:3, while shootroot ratio was 1:2. Root

length at day 7 had a very strong correlation with total length at day 1 {r = 0.93) (Fig

lc). Overall roottotal length ratio at day 7 was 2:3.

Very strong correlation was also observed between fresh weight and dry weight

at week 6 (r = 0.98), and number of leaves at week 2 with number of leaves at week 4 {r

= 0.77). Shoot length at week 4 strongly correlated with shoot length (r = 0.55), fresh

weight {r = 0.74) and dry weight at week 6 (r = 0.72).

Correlations between individual seed weight and other traits within cultivars

Eight cultivars tested had a positive correlation between seed weight and germination

rate within cultivars {r = 0.13 - 0.99) (Table 4). Thirteen other cultivars had no

correlation between seed weight and germination rate, where r ranging from -0.01 in

'Donner' to -0.99 in 'Plana'.

Correlation between individual seed weight and fresh weight at day 7 was

generally positive and strong, ranging from r = 0.4 in 'Lateman' to r = 0.99 in

'Atlantis'. Only 'Discovery' and 'Fremont' had a negative correlation between seed

weight and fresh weight at day l{r = -0.1 and -0.9) (Table 4).

85

Correlation between individual seed weight and fresh weight and dry weight at

week 6 varied among cultivars. Very strong correlations were found in 'Plana' and

'Megan' (r = 0.8), 'L3668', 'Panther', 'Jerez', 'Cauldron', 'Arctic' {r = 0.9). Some

cultivars had a negative correlation such as 'Belot', 'Discovery' and 'Fremont' (r = -0.5

to -0.9) (Table 4).

Molecular markers associated with morphological traits across cultivars

Out of 20 random primers tested, 17 showed clear polymorphisms, and a total of 88

R A P D bands were used as genetic markers. Only clear and reproducible bands were

used as genetic markers (Fig 2). A number of associations were found between

molecular and morphological traits among 21 cultivars tested. Based on the t-values, a

total of 24 molecular markers were identified as having a strong association with high or

low values for seedling traits.

Seed weight was associated with only one D N A marker SL08-626, which was

linked to lighter seed weight (Table 5). Germination rate associated with 7 markers, 4

of them linked with faster germination (A04-833, A04-185, D20-277, SK01-606).

Markers AO2-1038 and SL03-761 had a moderate association with heavier fresh weight

at day 7. Five markers had a strong association with heavier fresh weight at week 6 and

3 markers associated strongly with lighter fresh weight at week 6. Similarly, 4 markers

had a strong association with heavier dry weight at week 6 and 3 markers strongly

associated with lighter dry weight at 6 weeks (Table 6).

A number of molecular markers were associated across traits. SK01-606

associated with faster germination, longer shoot and total length at day 7. Conversely,

SK01-824 associated with slower germination and lower values of shoot length and

total length at day 7. SKI4-723 and U B C 106-404 associated with shorter root and total

length at day 7. OPH15-1794, SL03-562, UBC106-404 and UBC106-200 were strongly

associated with high values of fresh weight and dry weight at 6 weeks, while SL03-

1047, SL03-791 and SL03-595 were linked with fewer leaves, lower fresh weight, and

lower dry weight.

Discussion

Within and between cultivar variations were observed on all traits measured, even

though almost all cultivars tested were Fi hybrids and they were expected to be very

uniform. Variation occurred in morphological analysis and in molecular genetic

analysis and these was correlated.

86

Seed weight varied within and across cultivars. Seed weight is determined by

several factors such as environment, position of seeds in the mother plants (Gutterman

2000) and genetic factors (Alonso-Blanco et al. 1999). Environmental effects mediated

via the maternal plants contribute to seed weight variation. Nutrients, light and water

regime which the mother plants were subjected to during the growing season affects

seed weight (Fenner 1993).

Cauliflower seed are borne in a silique within a complex inflorescence.

Variation in seed weight both within the fruit and across the inflorescence occurs in

Brasicaceae (Susko and Lovett-Doust 2000). Early-initiated, basal fruits produce larger

seeds than fruits in the middle or the tip of an infructescence and these factors will

interact with genetic effects.

Genetic factors also impact on seed weight variation. In Arabidopsis at least 11

seed size quantitative trait loci contribute to seed size variation. Five loci control seed

size via maternal components affecting ovule number or reproductive resource

allocation in the mother plant. Six other loci are involved specifically in seed

development process (Alonso-Blanco et al. 1999). The APETALA2 gene contributes to

the determination of seed weight and size in Arabidopsis. This gene is required for

normal seed coat development (Jofuku et al. 2005) and also controls embryo cell

number and size (Ohto et al. 2005). In crop plants, 3 ISSR markers are associated with

low seed weight in wheat (Ammiraju et al. 2001) and 13 SSR markers are associated

with seed size in soybean (Hoeck et al. 2003). However, in general w e still know little

about the genetic regulation of seed size.

Seed weight influenced seedling vigour. Across cultivars, seed weight

correlated positively with seedling vigour up to 4 weeks and correlated strongest with

fresh weight at day 7. Positive correlations between seed size and fresh weight in 1-

week-old seedlings are c o m m o n in vegetables (Pradhan 2004; Soffer and Smith 1974).

Within cultivars, strong correlation was found between seed weight and seedling traits,

indicating that variations in seed weight were carried through seedling stage and this

would greatly benefit the successful establishment of individual plants in the nursery

and in the field (Kidson and Westoby 2000).

Smaller seeds germinated earlier in many of the cauliflower cultivars tested,

similar to Alliaria petiolata {Brassicaceae) (Susko and Lovett-Doust 2000), Cakile

edentula (Zhang 1993) and Erodium brachycarpum (Stamp 1990). Small seeds have

greater access to water as a result of their higher surface to volume ratios and so small

seeds m a y imbibe water faster and germinate sooner (Stamp 1990).

87

Whilst maternal and environmental factors affect germinability (Gutterman,

2000), certain genes are also involved in the control of seed germination. In

Arabidopsis, RGL2 regulates seed germination probably by functioning as an integrator

of environment and endogenous cues to control seed germination (Lee et al. 2002).

Germination rate was positively correlated with almost all seedling parameters,

therefore seeds that germinated faster tended to have more vigorous seedlings. In

general, cauliflower seeds examined here germinated within 5 days, much faster than

the ISTA 10 day standard for germination of most Brassica oleracea species (1STA,

2003).

Seedling vigour measurements of 7-day-old seedlings indicated root length was

the major component contributing two thirds to the total length. So within the first

week from germination, growth was biased in favour of root establishment rather than

shoot growth. Root establishment is critical for field establishment.

Seedling vigour at very early age (seedlings before transplanted) carried through

to later growth as shown by strong correlations between number of leaves in week 2 and

week 4, shoot length in week 2, with shoot length in weeks 4 and 6 and with fresh and

dry weight at week 6. There is usually a strong relationship between seed vigour,

seedling vigour and field performance (Kidson and Westoby 2000). Seed vigour

influences crop development and yield in cauliflower (Finch-Savage and McKee, 1990),

beetroot and carrot (Karuna and Aswathaiah, 1989). So, further study of seed and

seedling traits may eventually improve field performance.

A lot of marker associations were identified at 6 weeks, suggesting that seedling

screening/analysis should be performed at 6 weeks. This is a critical time in seedling

development as it marks the developmental stage when seedlings are transferred to the

field.

R A P D markers were employed to investigate the correlations between

morphological traits and genetic markers and consistent bands were found on all

primers tested. Out of 88 markers produced, 24 markers showed association with a

number of seedlings traits across cultivars. The variation observed in certain

morphological traits may be under strong genetic control, as indicated by a strong

association between traits and molecular markers. These potential markers could be

linked to genes controlling these traits and this requires further investigation for their

use in marker-assisted breeding for particular traits.

Several studies have associated R A P D markers with particular traits in Brassica.

Bulk segregrant analysis links R A P D markers with particular traits such as silique

88

shatter resistance in B. rapa (Mongkolporn et al. 2003) and club root resistance in B.

rapa (Hirai et al. 2004). With bulk segregrant analysis, a large number of primers and

populations need to be examined to find associations and validate results. Our method

employed a relatively small number of primers and plants and association was

determined using multivariate analyses based on genetic information generated from

R A P D analyses. This is quite a new approach and has been applied recently on radish

(Pradhan et al. 2004). Although a preliminary phase, this method indicates some

tightness of the association as shown by the range of /^-values. Here w e developed our

range as follow: 0.00 - 0.01 tightly linked, > 0.01 - 0.05 moderate linked, > 0.05 not

linked.

This investigation was a preliminary study in the process of identification of

particular traits using molecular markers. Association of molecular markers with

seedlings traits established here may contribute to marker-assisted selection. Bulk

segregant analysis will be required to examine populations and confirm the reliability of

this simpler alternative technique. Other markers such as SSR and S C A R may also be

employed and compared for their reliability. In the future, marker-assisted selection of

seedling traits could be conducted more efficiently, and this would assist in the

screening of plants at an early stage. This technique can be applied to any traits of

interest.

Acknowledgments W e thank AusAID for providing a scholarship to Ida Astarini.

Sincere thanks to Bejo, Syngenta, Enza Zaden (formerly Yates Seeds), South Pacific

Seeds and Henderson Seeds for providing seeds for this project. Financial assistance

from Department of Agriculture Western Australia is gratefully acknowledged.

References

Alonso-Blanco C, Blankestijn-De Vries H, Hanhart CJ, Koorneef, M (1999) Natural

allelic variation at seed size loci in relation to other life history traits of

Arabidopsis thaliana. Proceedings of the National Academic of Sciences USA

96,4710-4717.

Ammiraju JSS, Dholakia B B , Santra D K , Singh H, Lagu M D , Tamhankar SA, Dhaliwal

H S , Rao V S , Gupta V S , Ranjekar P K (2001) Identification of inter simple

sequence repeat (ISSR) markers associated with seed size in wheat. Theoretical

and Applied Genetics 102, 726-732.

89

Astarini IA, Plummer JA, Lancaster RA, Yan G (2004) Fingerprinting of cauliflower

cultivars using R A P D markers. Australian Journal of Agricultural Research

55,117-124.

Digby P, Galwey N, Lane P (1989) Genstat 7.0, Clarendon Press, Oxford.

Fenner M (1993) Environmental influences on seed size and composition.

Horticultural Reviews 13, 183-213.

Finch-Savage W E , McKee J M T (1990) The influence of seed quality and

pregermination treatment on cauliflower and cabbage transplant production and

field growth. Annals of Applied Biology 116, 365-369.

Gutterman Y (2000) Maternal effects on seed during development. The Ecology of

Regeneration in Plant Communities. Fenner M (ed). C A B International,

Wallingford, UK.

Hirai M , Harada T, Kubo N, Tsukada M , Suwabe K, Matsumoto S (2004) A novel

locus for clubroot resistance in Brassica rapa and its linkage markers.

Theoretical and Applied Genetics 108, 639-643.

Hoeck JA, Fehr W R , Shoemaker RC, Welke GA, Johnson SL, Cianzio SR (2003)

Molecular markers analysis of seed size in soybean. Crop Science 43, 68-74.

Hu J, Quiros CF (1991) Identification of broccoli and cauliflower cultivars with R A P D



Testing. Bassersdorf, Switzerland.

Jofuku K D , Omidyar PK, Gee Z, Okamura JK (2005) Control of seed mass and seed

yield by the floral homeotic gene APETALA2. Proceedings of the National

Academic of Sciences USA 102, 3117-3122.

Karuna M N , Aswathaiah B (1989) Effect of seed vigour on field performance in

beetroot and carrot. Seeds and Farms Sept-Oct, 40-46.

Kidson R, Westoby M (2000) Seed mass and seedling dimensions in relation to seedling

establishment. Oecologia 125, 11-17.

Mongkolporn O, Kadkol GP, Pang ECK, Taylor PWJ (2003) Identification of R A P D

markers linked to recessive genes conferring siliqua shatter resistance in Brassica

rapa. Plant Breeding 122, 479-484.

Nozaki T, Kumazaki A, Koba T, Ishikawa K, Ikehashi H (1997) Linkage analysis

among loci for RAPDs, isozymes and some agronomic traits in Brassica

campestris L. Euphytica 95, 115-123.

90

Ohto M , Fischer RL, Goldberg R B , Nakamura K, Harada JJ (2005) Control of seed

mass by APETALA2. Proceedings of the National Academic of Sciences USA

102,3123-3128.

Powell A A , Thornton JM, Mitchell JA (1991) Vigour differences in Brassica seed and

their significance to emergence and seedling variability. Journal of Agricultural

Science 116, 369-373.

Pradhan A, Yan G, Plummer JA (2004) Correlation of morphological traits with

molecular markers in radish {Raphanus sativus). Australian Journal of

Experimental Agriculture 44, 813-819.

Soffer H, Smith O E (1974) Studies on lettuce seed quality: IV. Individually measured

embryo and seed charactereistics in relation to continuous plant growth (vigor)

under controlled conditions. Journal of the American Society for Horticultural

Science 99, 270-275.

Somers DJ, Rakow G, Prabhu V K , Friesen K R D (2001) Identification of a major gene

and R A P D markers for yellow seed coat colour in Brassica napus. Genome 44,

1077-1082.

Stamp N E (1990) Production and effect of seed size in a grassland annual {Erodium

brachycarpum, Geraniaceae). American Journal of Botany 11, 874-882.

Stirling K, Lancaster R (2005) Effect of alternative planting configurations on

cauliflower development. Acta Horticulturae (in press).

Susko DJ, Lovett-Doust L (2000) Patterns of seed mass variation and their effects on

seedling traits in Alliaria petiolata (Brassicaceae). American Journal of Botany

87, 56-66.

Zhang J (1993) Seed dimorphism in relation to germination and growth of Cakile

edentula. Canadian Journal of Botany 71, 1231 -123 5.

91

£ •* *

se CU

a w

a CO

"S e

U

a 0) ••* CA

5? WD

.a -o CU 4» U

a

a o u

.a cu

CN o

09 CU

co g

* a 09

•c 0)

-a cu

s *

-a g -a Q,

5 « OJ CU

et > •B 3 u L. CU

0

i a A u

•

fH CU

2 « h

<*•

09

09

'C cu

U et 4a u •** a et PH

* cu *

09

•-a 09

c cu

13 « k. et -S u

u cu a

'S * A et

09

a o 09 « CU 09

M i k.

&

a o w T3 CU

cu

k.

i si

U

(U GO

O S-i

O 00

O CD VH

CD

cd

O

CD

.fi

o T3 CD tH CD

> o o

>

3 o

CD

?* -t-"

-fi DO

CNJ

00 «~I CD - >" R i—

M f^H -4-> 1—1 1—I I—I I—I I—I NN^H I—I

*

£ GO

• i-H S-I

OH

D i—r 0 0

0/)

i •4—1 cd fi CD

a u

CD

3 It) <D i -

is <D

• H

cd

a o OH

fi

oo O

a -a

o

CD CD

O 00 00

fi o

• »—I

H->

o CD +H

o 1-1

PH

13 CD

cn =3

O >-O GO

t/3 W W W

CN vo in ,-., r—I i—I t—I *-_*

cn o

m

I I

o •—I CN cn o

m CN --i- CN '-' C N N-H CN

I I

O - t c N - i C N N - i r t t ^ C ^ t ^ ( S I N - H N — l , - H ^ H ^ H , - H , - M , - H

I I I I os cn CN N—i

I I I I I o o o >n m

^H

CD

3 CO

oo ~&o

.s .s PH OH

&0 CO 00

CO CO CO CO

*o *o "d *o CD (D <D <D CD 1) <D <D

W W W W it-<

PH

« .i—i

o cd OH

5

OH

1 cd cd cd cd C2 CD CD Dfi GO

CD

GO

c CD

00 fi

c CD

-3 N cd N fi

c <1> Ti cd N cd

S

fi aj TTJ cd N cd

a

« CD T3 cd N cd N

c CD CD <D CD CD CD

M W M ^ o Q c w w w S w w W O H P Q p q p Q o q p Q

fi

cd

N cd

a w

CO

CD CD 00 o « % &H

-s fi o 00

c o co S-i

CD

-a fi CD

09

9 cu 43 & o

a P 2

09

•-a a et

3

a C4

O

PH

o

3 £ % £ s «

Vfi

PH

o cu

a o

I

i s

•

-H a

a B et 09

u

>

i CU

u

•a

a « 03

-a

•8

cu

u

II

fr 73

II 73

•ft

fr A 73

73 II

^ I e ja 09 M>

V 09 •J* tU

2 ^ « II

2 *r

? -ft — V

.s * § a "2 fl

3, "

a * o

U •

cu

2 5

a o •-C

.a cu

«4

I a?

I Sa § ^ ja

hH

N O N O N n M V ) m H ; ( N f N l f n r N | - n r H M " s t M o o o o o o o o o o o o o o o -H-H-H-H-H-H-H-H-H-H-H-H-H-H-H

OS O 00 © ~ CN

o o o o o o -H -H -H -H -H -H C ^ C N O O ^ ^ C N I ^ I - ^ O S C D C N ^ H C S O S O O O S C T N O

c D C N d ^ ~ c N N - ^ t N d ^ . - J — I d d © © — ^ ^

^ ^ ^ r ^ ^ < R 0 ° < 3 ^ < - ? ° ® 0 > > c N ' V 0 ^ > N ? N t - ~ i ^ ' ^ t o c N r - -cn '"fr CN CN © "fr ~ CN CN CN —" CNi I-H" CN © CN CN —' >—'- cn CN

- H - H - W - H - H - H - H - H - H - H - H - H - H - H H H - H - H - H - H - H - H C N O O r - - - , ' - f r S O i D O \ o o r - - - O s ^ H i D C N © C D a s c D i D ' — ' © s o -i C N - H t-~ r--

C D C N -^ —• N—< O SO © CN CN N— CN »—

"fr >n "fr ^ - . ^ H O — ' O CN cn ID

~ — CN

t ^ C D " f r © © s q o o i D O > C s t ^ O S N O c N C N — < C ^ C ^ C - - - - ' ' 4 - c D ^ C N O C D ^ d c N ^ - d d ^ ' o ' d c N C N C N C D ^ d ^ ^ - H - H - H - H - H - H 4 H - H - H - H - H - H - H - H - H - H - H - H - H - H - H

m f t ' H o o q ' t ^ C N i ^ i n N O N O N o a N ' N t N c i H H M O ' H r n

^ s 6 " f r d d d d c N © o s o s t ^ c D ' ^ " f r ^ t - ^ o s o d s d a s C N ' - N — I C N C N C N C N C N C N — H ^ H — H C N C N C N C N C N - ^ — ' - ^ - ^

2 a cu

«^ 133

£ a

= a * a r-i 73

a a

73 cu cu 03

O ~ ,-H' © -H' —! CN —^ © © -H -H -H -H -H ~ •« rH rt; NO « N£) -H

O — O O O O ^ — ' O — ' —

- H - f l - H - H - H - H - W - H - H ^ - H - ^ - H - H r - N o i o r o i n O o o H ^ s o r ^ i n N O H ^ - o

OS SO 00 Os © CN

Os Os Os Os Os o CN

c-~"frm'^-oo"-rv©r~-«-H

* N O ^ N ( n n c N j N c n ^ n * ( * ) N O N N e ( N j h ; c n ^ l ; i c i O N - ^ O O O O O O O O O O O O O O O O O O O

- H - H - H - H - H - H - H - W - H - H - H - H - H - H - H - H - H - H - H - H - H © o © c D c D i n i D O O c D i n o o i n © i D i n o o i n c D O O i n o o c D ( - y \ i D t ^ r ^ t ^ i ^ t ^ o 6 c ^ r ^ r ^ t ^ s d t ^ t ^ t ^ o 6 o d

"fr

o o -H SO

"fr

CD

o -H CD SO

so

o o -H "fr

"fr

o

o o -H o in

CD

o -H CD SO

"fr

o o -H SO

"fr

"fr

o o -H so

"fr

"fr

o -H ID

o

o o -H o in

o o -H CD CD

CD

o -H CD SO

"fr

o -H oo ID

CD

o +1 oo 00

r-

o -H ID

CD

O

-H oo

oo

SO

o -H so ID

ID

O o -H CN

"fr

ID

o o -H CN

"fr

"fr O o -H 00 CD

o o -H 00 CD

o o o -H o ID o o o o o o o o o o o o o o o o o o o o o

—i o o —i —i CN

CD CD 00

-H -H -W -H -H -H -H

CN

oo ^ — CN

+1 « r t o o i - ^ > - H i D i D r - - ^ ^ ^ o s o ^ o s o i n

- H - H - H - H - H - H - H - H - H - H - H - H - H CD OS ^ -*f ,-H CD 1" CD CN SO

CD CD

O s o o r - ^ H C D " H r c D C - - O S ' s ( - O s c D ' - H ' ^ - C D C N c D C N C - - ' * f r " f r C D i D " l - i D " r

ID CD CD "fr

o m t-- so —<" o c> o -H -H -H -H r^ >-; -n- N-H so Os so C-

H ^ N O O O N O O O O C D O O O S ^ - ; ^

O O O O O - — I C 5 - - ^ O O ^ H

- H - H - H - H - H - H - H - H - H - H - H ^ - t ~ - " ^ - - - r - ~ o s " f r o o r - f - o

ID o Os o ID t---H' -^ O -H N-H O

"H ~H ~H ~n ~H ~H r- CN OS ID SO "fr

00 CD

d d -H -H CN t-; OS "d- so "t

so

d -H -H

r - i n s o s o N O H - j - v o s o o o o o o s ^ o o o o o o o f - i n

^NON^(or^t^CDCDt^CNCNCNO\00C3sCDNO o ' d d d d d d ^ n d d ^ ^ ^ o o o ^ H O - H - H - f l - H - H - H - H - H - H - H - H - H - H - H - H - H - H l O h ^ ^ N O N O r H r H r H O O N t N N O ' t r H i n O ) ^ ' ( ^ H ^ ' T ^ « ^ c N H - f r ^ i o < / ^ s d " f r i D s d s o ' i n c N

C D ^ C N O ^ O C D ^ C N i n ^ i D C D ^ C D ^ C N C N C N C D C D

iDN*ô.-HNOOOoo^cDCDC^côocînsq«n(D--;iD

CN CN CN CN CN CN CN —" CN CN CN CN CD CN CN CN CN CN CN CN CN

01

a

SO

o -H o

ID

o -H i—i

CN O

-H CD

ID

O

-H o

CD ID CD CD

C-~ "fr CD CD CN

d c> d d d 41 +1 -H -H -H

Os in cN c- so CD "fr "fr CN

C N C N ' ^ i D C N C D C D ' ^ -

c i c i c i d d d c i c i -H -H -H -H o in oo "fr "fr" CN "fr "fr

t? ~ k, cu

- » > 2 et a o a

1 s J> I PH Q Q fe

et d

i

00 fS * s 1

> S u

09

J ro .

3 ^ m •< iJ P5

-H

CD

-H -H -H CN t~- CN

ID

d

-H

o

d

-H oo

^1- -5-j- in in m

m so

d d -H -H —i CM "fr" "f

NH

k

cu

O

en

S a i* ^

•3 et 3 r-

a ^ I ^

so

o

u

P_. a

§5 o

CD OS

hi

%

a a

I •SH

A

r *

•9 vi

cu c

i .1 to "ft .2 T II 5 a +? (n ft ©

H <T

* 2 cu

E S et 73

a w 09 09

0 k, u Sj HS

?•

b 09 W)

•5 73

09

e 'C w • * *

a '5

e cu

0 u a e v a ! o

•

CU

2 a H

a a U s ft 09

|| *

J CU u

II * •s

fr 73 II 73 rf A WD CU

fr 73 II

Q

"ft •M CU *

-a 09 9

VI

a a u

s ft

I 1

PH S © it o so •™ HW

GO S

© it

GO —

o o -3 GO

% CM

S3 cu

& <s

,1 so

cu "*r

"a *

*u a

HSS

£ O 73 O i>

A *-&0 —

73 CU B

GO

— o

00

<fr — ©

CN cn

— d

* r-; — ©

CN ON

CD © © ©

i •

m CN

—< ©

73

* * * CD ON

—, ©

r-~ t-CN m © ©

CN SO CD CO © ©

SO

d

oo © CN "fr

© ©

• - ID — CN

CN ro

d

00

©

"fr CN ©

"fr

"f

"fr SO ID

00 00 © — © ©

—. cn

1 ? 2-N "*

SO ~?

73 73

t * x ss f N- " W 73 cu cu GO

©

o A GO

et O <vw

q o PH H

73

cu HM

2

I 8

00 CN

% tM OS

cu NH

"fr "fr

ro ro

SO CN

o ©

00 CN

CN

CN ro

00

o

CN ID

* * © 00

© ro

so o ©

ro ©

ON

o

NO ©

c

o o

ro O

ro ro

"fr "fr

ro

©

© ©

* D r-so ©

CN © ©

ro ©

SO

so ©

"fr CN

© CN

SO

ro

© so —i io © CN

o d d

ro CD ~ CN

9 d

o

00 CN

* * 00

ON — ©

©

q —i ©

2 ©

9 5

"fr

ro

00 ©

1 f £ es cu J

1 % «4-93 CU HH

•M

o o A GO

1 H— O o -a GO

1 H-

O o A GO

1 +j • * *

o &

CN CD

"fr d

00

"fr ©

o CO

CO

o d

* "fr SO

* CN

ID ID

CO

— r-< — >

* 00 NO

* © r-

© "fr

CN

"fr ©

Os

"fr

SO

"fr ro

>D

ro

NO © ©

"t Os

Table 4. Correlation between seed weight with other seedling traits within

cultivars. F W = fresh weight, G e r m = germination, d=day, w = week, *=

significantly different (p< 0.05), **=higbly significant (p< 0.01), ***= very highly

significant (p< 0.001).

Correlation coefficient (r) between seed weight with traits

Cultivars

Plana

Donner

Discovery

Fremont

Monarch

CF284

Virgin

Morpheus

Cauldron

Arctic

Atlantis

L3368

Belot

Lateman

Panther

Jerez

Fandango

Megan

Delfur

SPS 716

Omeo

FW7d

0.78**

0 87***

-0.60*

-0 98***

0.53

0.73*

0.62*

0.79*

0.69*

0.98***

0 99***

0.59

0.65*

0.40

0.11

0.72*

0.54

0.53

0.62*

0.62*

0.53

Germ rate

-0.01

-0.98***

-0.09

-0 98***

-0.17

-0.51

0.14

0.33

n 90***

0.86***

-0.13

0.84**

-0.58

-0.11

-0.78**

0.83**

-0.22

-0.54

0.34

0.34

-0.54

FW6w

0.78**

-0.02

-0.80**

-0.96***

-0.15

0.58

0.65*

0.06

n 99***

0 99***

0.13

0 99***

-0.56

0.62*

0 97***

0 99***

0.73**

0.81**

0.38

0.38

0.86***

DW6w

0.83**

0.18

-0.87***

-0.99***

-0.88**

0.70*

0.10

0.20

n 99***

0 99***

-0.00

0 99***

-0.49

0.62*

0 97***

0 9c***

0 91***

0.80**

0.37

0.37

0.96***

Table 5. Significance of markers associated with higher or lower values for seed

weight, germination rate (Germ), shoot length (It), root length, total length and

fresh weight (FW) at 7 days (d) for 21 cultivars. Values given are /7-values.

Markers

A02-1038 A04-833 A04-395 A04-185 D20-277 SKO1-606 SL03-761

A04-1059 A04-833 OPA07-707 SKO 1-824 SK14-723 SK19-1443 SL08-626 UBC106-404

Seed wt

n.s. n.s n.s. n.s n.s n.s. n.s

n.s. n.s. n.s. n.s. n.s. n.s. 0.043 n.s.

G e r m rate Markers for

n.s. 0.029 n.s. 0.02 0.02 0.00 n.s

Shoot It Root It • higher values n.s. n.s n.s. n.s n.s 0.001 n.s

n.s. n.s 0.043 n.s n.s n.s. n.s

Markers for lower values 0.02 0.022 0.02 0.001 n.s. n.s. n.s. n.s.

n.s. n.s. n.s. 0.002 n.s. 0.003 n.s. n.s.

n.s. n.s. n.s. n.s. 0.019 n.s. n.s. 0.03

Total It

n.s. n.s n.s. n.s n.s 0.032 n.s

n.s. n.s. n.s. 0.038 0.038 n.s. n.s. 0.023

FW7d

0.016 n.s n.s. n.s n.s n.s. 0.04

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

Table 6. Significance of markers associated with higher or lower values for fresh

weight (FW), dry weight (DW), leaf number (#), shoot length (It) and root length at

harvest after 6 weeks (w) for 21 cultivars. Values given are/>-values.

Markers

D12-1864 D12-1194 D20-1074 OPH15-1794 SL03-562 SK14-723 UBC 106-404 UBC 106-200

A04-833 D12-199 SKO 1-824 SKI 9-953 SL03-1047

SL03-791 SL03-761 SL03-595

FW6w

0.042 n.s. n.s. 0.021 0.000 n.s. 0.000 0.022

n.s. n.s. n.s. n.s. 0.001 0.001 n.s. 0.001

DW6w Markers for

n.s. n.s. n.s. 0.032 0.000 n.s. 0.002 0.026 Markers for

n.s. n.s. n.s. n.s. 0.000 0.000 n.s. 0.000

Leaf# higher values

n.s. 0.03 n.s. n.s. n.s. n.s. n.s. n.s.

• lower values 0.006 0.04 n.s.

n.s. 0.029 0.029 n.s. 0.029

Shoot It

n.s. n.s. 0.025 n.s. n.s. 0.017 0.000 n.s.

n.s.

n.s. n.s.

n.s. n.s. n.s. n.s.

n.s.

Root It

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

n.s.

n.s. 0.009

0.003 n.s. n.s. 0.048

n.s.

96

Fig 1. a. Seedling size variation at day 7, from lightest to heaviest (left to right),

b. Largest and smallest seedlings at 6 weeks, c. Biggest and smallest root mass at 6

weeks.

M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15161718 19 20 21 M t

ML 1000bp

i."-=.il"lallils'r" ^•|3_5oobP

Fig 2. RAPD profiles of 21 cauliflower cultivars using primer UBC 106. A,

Markers for root length, total length at day 7, shoot length, fresh weight and dry

weight at week 6. M, Marker Ladder; 1, Plana; 2, Donner; 3, Discovery; 4,

Fremont; 5, Monarch; 6, CF0284; 7, Virgin; 8, Morpheus; 9, Cauldron; 10, Arctic;

11, Atlantis; 12, L3368; 13, Belot; 14, Lateman; 15, Phanter; 16, Jerez; 17,

Fandango; 18, Megan; 19, Delfur; 20, SPS 716; 21, Omeo.

97

Chapter 8

General Discussion

Lack of uniformity is a critical issue limiting profitability in the cauliflower

industry. Improving uniformity, machine harvesting and developing new export

markets are regarded as high priority research issues for the cauliflower industry

(Warren Cauliflower Group, 2003). Non-uniform production is caused by variation,

occurring within cultivar seed and in the development of seedlings. This variation is

exacerbated in the field environment to produce plants with uneven curd maturity and

the production of sibs. Procedures were developed to identify sources of genetic

variation with the aim of improving uniformity using molecular genetic techniques.

This thesis demonstrated successful application of molecular marker techniques

to identify cultivars, to develop fingerprinting keys, to reveal genetic relatedness among

cultivars (Chapter 3, 4, 5), to differentiate between male and female parent lines, and

between hybrid and non hybrid plants (Chapter 6), and to link molecular markers with

seedling traits (Chapter 7). Research and findings of this thesis will substantially

increase our ability to visualise and detect sources of genetic variation.

R A P D technique was successful in differentiating 25 Fi hybrid cauliflower

cultivars commonly grown in Australia (Chapter 3). The 25 polymorphic markers

clearly distinguished 18 cultivars and identified genetically identical cultivars. A

minimum of 12 markers obtained from 10 primers was required to separate cultivars

and generate a fingerprinting key.

The ability of R A P D technique to differentiate closely related varieties will be

very useful for the cauliflower industry in Western Australia and around the world. As

the number of cultivars available in the market increases, the ability to distinguish them

on the basis of morphological traits becomes more difficult (Lombard et al, 2000) and

molecular markers are an accurate option, as no environmental factors such as weather,

seasonal and agronomy issues are involved.

The R A P D markers also successfully identified within and between cultivar

variability in open pollinated cultivars from Indonesia (Chapter 4). This technique was

able to differentiate geographical/growing areas of the cultivars. This is the first report

of Indonesian cauliflower cultivar identification using molecular markers and will

certainly encourage further research into cauliflowers, other Brassica vegetables and

other vegetable crops particularly in developing countries.

The study confirmed the ability of R A P D technique to work on genetically

distinct groups of cauliflowers, i.e. control pollinated, Fi hybrids from temperate

regions and open pollinated lines from tropical regions. This indicates the technique

m a y be able to discriminate a wide range of cauliflower cultivars and could be used as a

99

fast method to verify cultivars worldwide. R A P D fingerprinting will be extremely

useful for future cauliflower identification, avoiding possible mislabeling and fraud.

At present, distinctness, uniformity and stability are the criteria of Plant

Breeders's Rights for the purpose of registration of new plant varieties ( U P O V - B M T ,

2002). A s per the guidelines of the International Union for the Protection of N e w

Varieties of Plants, distinctness, uniformity and stability testing is currently primarily

based on essential morphological characters. The distinctness, uniformity and stability

testing based on phenotype and isozyme expression suffers from the limited number of

target traits and genotype by environment interactions when the candidate variety is

evaluated across environments. Cultivar registration based on D N A markers will

provide accuracy as they are not affected by environmental factors. Consequently, it

would significantly reduce costs as no field trials are required to confirm identity. D N A

markers are also independent from the developmental stage of the plant, while many

protein-based markers are, therefore providing more accuracy compared to for example,

isozyme markers that are being used in some crops such as maize, wheat and barley.

The use of D N A markers will be implemented in the future to establish the

distinctness, uniformity and stability of plant variety trials (Nandakumar et al, 2004;

U P O V - B M T , 2002). A D N A marker-based registration test will substantially enhance

the process of discrimination of candidate varieties and hybrids. Information that is

likely to be broadly applicable for cauliflower identification is n o w available for both

narrow genetic based Fi hybrid cultivars and high variability open pollinated cultivars.

Chapters 3 and 4 of this thesis provide valuable information to be considered in the

process of establishing DNA-marker based cultivar registration.

Genetic distance between Indonesian and Australian cultivars was confirmed in

this thesis using R A P D markers (Chapter 5). Comparison between open pollinated and

Fi hybrid cultivars shows that more variation occurs within cultivars in open pollinated

plants. All Indonesian cultivars tested were separated from Australian grown cultivars

confirming they have a long independent breeding history. Distant relationships

between Europe and East Asian Brassica accessions have also been reported in Brassica

rapa (Zhao et al, 2005).

The study opens the opportunity for breeding better cauliflowers in both

countries. Indonesian cultivars will be good resources to broaden genetic diversity and

bring the potential characteristics such as resistance to abiotic and biotic stresses for

cultivar improvement programs. Hybridization of cultivated species with distantly

related species, with unimproved 'wild' relatives or aiming for better productivity and

100

disease resistance has been attempted for a number of crops such as canola (Voss et al,

2000; Brown et al, 1996) and chickpea (Singh et al, 2005; Crosser et al, 2003). For

future work, collection and characterisation of vegetable Brassica and their wild

relatives using R A P D markers would provide important germplasm information for

potential breeding programs both within Australia and world wide. Comprehensive

research should be focused on finding markers for particular traits such as disease

resistance and high yield.

Fi hybrid cultivars were originally established with the aim to improve yield and

uniformity. However, a considerable proportion of non-uniform curds, assumed to be

caused by self-inbred plants are found in all production areas. These are commonly

known as 'sibs'. A n improvement from the 2-20% of losses currently considered to be

'sibs' could benefit growers by at least $330 per hectare, with a potential benefit of

about $3300 per hectare if sib losses were high. In Chapter 6, microsatellite markers

were proved powerful enough to distinguish female and male parent lines, hybrid and

non-hybrid plants. Controlled pollination experiments in the glasshouse, a field

production trial and microsatellite marker examination proved that self-inbreeding was

not the only cause of sibs. It has been observed in the field that poor management

increases the percentage of 'sib-like' aberration (Lancaster, Pers. C o m m ) . It has also

been suggested that aneuploidy, a missing or extra chromosome (Gerard Korevaar;

Duane Falk, Pers. C o m m ) may be responsible. A recent investigation on aberrant

cauliflowers using A F L P and M S A P (Methylation Sensitive Amplification

Polymorphism) indicated a low level of polymorphism between 'normal' and 'aberrant'

plants at the end of the vegetative development, suggesting 'sib-like' aberrations in

cauliflower m a y be under epigenetic control (Salmon et al, 2004).

Future research using cytology examination of chromosome karyotyping would

be useful to explore chromosome abnormalities in sib or aberrant plants. Further

investigation into the interaction between genetic factors and the environment should

also be pursued to find the true cause of the abnormality.

Variation in production begins at the seed and seedling stage. High quality

seeds should have high purity and germination, with good vigour. In Chapter 7,

variation within and between cultivar seedlings traits was revealed. R A P D markers

associated with seedlings characteristics were identified, indicating there are genetic

differences. The interaction of genetic differences and environmental conditions is

expressed in morphological variation. This chapter shows that variations at the early

growth stage are identifiable, screenable and removable. Screening at the seedling stage

101

will assist in breeding and selection of superior seedlings. It will also substantially

reduce seedling variability which will be an important factor in reducing production and

labour costs during growing, reducing losses from early and late harvest dates where

harvestable numbers are small and are considered uneconomic to recover. It is

projected that 1 % increase in exportable yield would represent at least $170 per hectare

return to growers.

Finding associations between molecular markers and seedling morphological

traits is still in its preliminary stage. The idea is to offer a simpler method to identify

genes controlling morphological traits. To confirm the benefit and accuracy of identified

markers, this technique needs to be further tested using currently available methods

such as bulk segregant analyses on recombinant inbred line populations. In the future,

seedling nurseries and breeders would be able to use this information to screen out

unfavorable plants.

In conclusion, procedures for cultivar fingerprinting using R A P D markers,

distinguishing male and female parent lines, hybrids and non-hybrids using

microsatellite markers and finding associations between molecular markers with

morphological traits have been developed in this thesis. Being able to screen cauliflower

plants in every stage of production, from choosing the right cultivar, screening for

particular traits to reduce seedling variability, and screening for abnormality will

significantly improve uniformity in cauliflower production. Molecular techniques

established in this thesis provide a promising and reliable approach to cultivar

identification, purity testing and screening for particular traits.

102

Chapter 9

References

A B S (2005) Australian Bureau of Statistics. Australia.

Alonso-Blanco C, Blankestijn-De Vries H, Hanhart CJ, Koorneef M (1999) Natural allelic variation at seed size loci in relation to other life history traits of Arabidopsis thaliana. Proceedings of the National Academic of Sciences USA 96, 4710-4717.

Ammiraju JSS, Dholakia BB, Santra D K , Singh H, Lagu M D , Tamhankar SA, Dhaliwal HS, Rao VS, Gupta VS, Ranjekar P K (2001) Identification of inter simple sequence repeat (ISSR) markers associated with seed size in wheat. Theoretical and Applied Genetics 102, 726-732.

Araya A, Zabaleta E, Blanc V, Begu D, Hernould M , Mouras A, Litvak S (1998) R N A editing in plant mitochondria, cytoplasmic male sterility and plant breeding. Electronic Journal of Biotechnology 1, 31-39.

Asandhi A A , Sastrosiswojo S (1988) Research on vegetable in Indonesia. In McLean B T (Ed.). Vegetable Research in Southeast Asia. Asian Vegetable Research and

Development Center, Taipei.

Astarini IA, Plummer JA, Lancaster RA, Yan G (2004) Fingerprinting of cauliflower cultivars using R A P D markers. Australian Journal of Agricultural Research 55,

117-124.

A U S V E G (2005) Commodity spotlight, cauliflower. A U S V E G , Victoria, Australia.

Barret P, Delourme R, Foisset N, Renard M (1998) Development of a S C A R (sequence characterized amplified region) marker for molecular tagging of the dwarf BREIZH (Bzh) gene in Brassica napus L. Theoretical and Applied Genetics 97,

828-833.

Bateman AJ (1955) Self-incompatibility systems in Angiosperms. III-Cruciferae.

Heredity 9, 53-68.

Baskin CC, Baskin J M (1998) Seeds: Ecology, Biogeography, and Evolution of Dormancy and Germination. San Diego, Academic Press.

Bellamy A, Vedel F, Bannerot H (1996) Varietal identification in Cichorium intybus L. and determination of genetic purity of Fi hybrid seed samples, based on R A P D

markers. Plant Breeding 115,125-132.

Bert K E , Lydiate DJ (2003) Genetic analysis and genome mapping in Raphanus.

Genome 46, 423-430.

Bhalla PL, Singh M B (1999) Molecular control of male fertility in Brassica. Proceedings of the 10th International Rapeseed Congress, Canberra, Australia.

Bond JM, Mogg RJ, Squire GR, Johnstone C (2004) Microsatellite amplification in Brassica napus cultivars: Cultivar variability and relationship to a long-term feral

population. Euphytica 139, 173-178.

Bornet B, Branchard M (2001) Nonanchored inter simple sequence repeat (ISSR) markers: reproducible and specific tools for genome fingerprinting. Plant

Molecular Biology Reporter 19, 209-215.

Boury S Lutz I, Gavalda M-C, Guidet F, Schlesser A (1992) Empreintes genetiques du chou-fleur par R A P D et verification de la purete hybride Fi d'un lot de semences.

Agronomie 12, 669-681.

104

Bretagnolle F, Thompson JD, Lumaret R (1995) The influence of seed size variation on seed germination and seedling vigour in diploid and tetraploid Dactylis glomerata L. Annals of Botany 76, 607-615.

Brown J, Thill D C , Brown AP, Brammer TA, Nair H (1996) Gene transfer between canola {Brassica napus) and related weed species. In Proceedings of the 8th

Symposium on Environmental Releases of Biotechnology Products: Risk Assessment Methods and Research Progress, Ottawa, Canada.

Cansian RL, Echeverrigaray S (2000) Discrimination among cultivars of cabbage using randomly amplified polymorphic D N A markers. HortScience 35, 1155-1158.

Cardi T and Earle E D (1997) Production of new C M S Brassica oleracea by transfer of 'Anand' cytoplasm from B. rapa through protoplast fusion. Theoretical and Applied Genetics 94, 204-212

Charsley T N (1998) Reducing the Harvest Period of Cauliflower {Brassica oleracea var. botrytis L.) with Pre-treatments of Cold and Gibberellic Acid. B.Sc. thesis. Plant Sciences, Faculty of Agriculture, The University of Western Australia.

Charters Y M , Robertson A, Wilkinson MJ, Ramsay G (1996) P C R analysis of oilseed rape cultivars {Brassica napus L. ssp. oleifera) using 5'-anchored simple sequence repeat (SSR) primers. Theoretical and Applied Genetics 92, 442-447.

Cho S, Kumar J, Shultz JL, Anupama K, Tefera F, Muehlbauer FJ (2002) Mapping genes for double podding and other morphological traits in chickpea. Euphytica 128, 285-292.

Cooke RJ (1999) Modern methods for cultivar verification and the transgenic plant challenge. Seed Science and Technology 27, 669-680.

Couvillon G A (2002) Cercis canadensis L. seed size influences germination rate, seedling dry matter, and seedling leaf area. HortScience 37, 206-207.

Crisp P and Tapsell C R (1993) Cauliflower, Brassica oleracea L. In Kalloo G, Bergh B O (Eds). Genetic Improvement of Vegetables Crops. Pergamon Press, Oxford.

Crockett PA, Bhalla PL, Lee CK, Singh M B (2000) R A P D analysis of seed purity in a commercial hybrid cabbage {Brassica oleraceae var. capitata) cultivar. Genome

43,317-321. "

Crockett PA, Singh M B , Lee CK, Bhalla PL (2002) Genetic purity analysis of hybrid broccoli {Brassica oleracea var italica) seed using R A P D PCR. Australian

Journal of Agricultural Research 53, 51-54.

Croser JS, Ahmad F, Clarke HJ, Sidiqque K H M (2003) Utilisation of wild Cicer in chickpea improvement-progress, constraints, and prospects. Australian Journal of

Agricultural Research 54, 429-444.

Darmawan D A , Pasandaran E (2000) Indonesia, dynamics of vegetable production, distribution and consumption in Asia. In Ali M . (Ed.), Asian Vegetable Research

and Development Center, Taiwan.

Das S, Rajagopal J, Bhatia S, Srivastava PS, Lakshmikumaran M (1999) Assessment of genetic variation within Brassica campestris cultivars using amplified fragment length polymorphism and random amplification of polymorphic D N A markers.

Journal ofBioscience 24, 433-440.

Delourme R, Budar F (1999) Male Sterility. In Gomez-Campo C (Ed.). Biology of

Brassica Coenospecies. Elsevier Science B.V. Amsterdam.

105

Demeke T, Adams RP, Chibbar R (1992) Potential taxonomic use of random amplified polymorphic D N A (RAPD): a case study in Brassica. Theoretical and Applied Genetics 84, 990-994.

Digby P, Galwey N, Lane P (1989) Genstat 7.0, Clarendon Press, Oxford.

Divaret I, Margale E, Thomas G (1999) R A P D markers on seed bulks efficiently assess the genetic diversity of a Brassica oleracea L. collection. Theoretical and Applied Genetics 98,1029-1035.

Dos Santos JBD, Nienhuis J, Skroch P, Tivang J, Slocum M K (1994) Comparison of R A P D and RFLP genetic markers in determining genetic similarity among Brassica oleracea L. genotypes. Theoretical and Applied Genetics 87, 909-915.

Dulson J, Kott LS, Ripley V L (1998) Efficacy of bulked D N A samples for R A P D D N A fingerprinting of genetically complex Brassica napus cultivars. Euphytica 102, 65-

70.

F A O S T A T Database. FAO. http://apps.fao.org/page/collection?subset=agricuture. Last

update 14 July 2005.

Farnham M W (1996) Genetic variation among and within United States collard cultivars and landraces as determined by randomly amplified polymorphic D N A markers. Journal of American Society for Horticultural Sciences 121, 374-379.

Fenner M (1993) Environmental influences on seed size and composition. Horticultural

Reviews 13, 183-213.

Finch-Savage W E , McKee J M T (1990) The influence of seed quality and pregermination treatment on cauliflower and cabbage transplant production and

field growth. Annals of Applied Biology 116, 365-369.

Fitzgerald D M , Barry D, Dawson PR, Cassells A C (1997) The application of image analysis in determining sib proportion and aberrant characterization in Fi hybrid

Brassica populations. Seed Science and Technology 25, 503-509.

Frankel R, Galun E (1977) Pollination Mechanism, Reproduction and Plant Breeding.

Springer-Verlag, Berlin, Germany.

Fu T, Ping S, Xiaoniu Y, Guangsheng Y (1992) Overcoming self-incompatibility of

Brassica napus by salt (NaCl) spray. Plant Breeding 109, 255-258.

Fujime Y, Okuda N (1996) The physiology of flowering in Brassicas, especially about

cauliflower and broccoli. Acta Horticulturae 407, 247-254.

George R A T (1999) Vegetable Seed Production. 2nd Ed. CABI Publishing Wallingford,

UK. Geraci A, Divaret I, Raimondo F M , Chevre A M (2001) Genetic relationships between

Sicilian wild populations of Brassica analysed with R A P D markers. Plant

Breeding 120,193-196.

Gowers S (2000) A comparison of methods for hybrid seed production using self-incompatibility in Swedes {Brassica napus ssp. napobrassica). Euphytica 113,

207-210.

Grubben G J H (1977) Tropical Vegetable and Their Genetic Resources. International

Board for Plant Genetic Resources, Rome.

Gupta PK, Varshney R K (2000) The development and use of microsatellite markers for genetic analysis and plant breeding with emphasis on bread wheat. Euphytica

113, 163-185.

106

http://apps.fao.org/page/collection?subset=agricuture

Gupta PK, Varshney R K , Sharma PC, Ramesh B (1999) Molecular markers and their applications in wheat breeding. Plant Breeding 118, 369-390.

Gupta M , Chyi Y-S, Romero-Severson J, Owen JL (1994) Amplification of D N A markers from evolutionarily diverse genomes using single primers of simple-sequence repeats. Theoretical and Applied Genetics 89, 998-1006.

Gutterman Y (2000) Maternal effects on seed during development. The Ecology of Regeneration in Plant Communities. Fenner M (Ed). CABI Publishing, Wallingford, U K .

Hadley P, Pearson S (1998) Effects of environmental factors on progress to crop maturity in selected Brassica crops. Acta Horticulturae 459, 61-70.

Hallidri M , Pertena D (2002) Self-incompatibility test in cabbage {B. oleracea var capitata). Acta Horticulturae 519, 117-122.

Harvey E, Smith B M (1987) A recent survey of sib content in Fi hybrid Brussels sprout varieties. Cruciferae Newsletter 12, 122-123.

Hendrix SD, Nielsen E, Nielsen T, Schutt M (1991) Are seedlings from small seeds always inferior to seedling from large seeds? Effects of seed biomass on seedling growth in Pastinaca sativa L. New Phytologist 119, 299-305.

Heneen W K , Jorgensen R B (2001) Cytology, R A P D , and seed colour of progeny plants from Brassica rapa-alboglabra aneuploids and development of monosomic

addition lines. Genome 44, 1007-1021.

Henry RJ (1997) Practical Applications of Plant Molecular Biology. Chapman and

Hall, London.

Heslop-Harrison J, Heslop-Harrison Y (1970) Evaluation of pollen viability by enzymatically-induced fluorescence; intracellular hydrolysis of fluorescein

diacetate. Stain Technology 45, 115-120.

Hirai M , Harada T, Kubo N, Tsukada M , Suwabe K, Matsumoto S (2004) A novel locus for clubroot resistance in Brassica rapa and its linkage markers. Theoretical

and Applied Genetics 108, 639-643.

Hodgkin T (1981) Some aspects of sib production in Fi cultivars of Brassica oleracea.

Acta Horticulturae 111, 17-24.

Hoeck JA, Fehr W R , Shoemaker RC, Welke G A , Johnson SL, Cianzio SR (2003) Molecular markers analysis of seed size in soybean. Crop Science 43, 68-74.

Holland RL, McNeilly T (1985) Genotype environment interactions and sib content in

Fi hybrid-brussels sprouts. Euphytica 34, 371-376.

Hu J, Quiros CF (1991) Identification of broccoli and cauliflower cultivars with R A P D


Humpry M E , Lambrides CJ, Chapman SC, Aitken EAB, Imrie BC, Lawn RJ, Mclntyre CL, Liu CJ (2005) Relationships between hard-seedness and seed weight in mungbean {Vigna radiate) assessed by Q T L analysis. Plant Breeding 124, 292-

298.


Testing. Bassersdorf, CH-Switzerland.

Jofuku K D , Omidyar PK, Gee Z, Okamura JK (2005) Control of seed mass and seed yield by the floral homeotic gene APETALA2. Proceedings of the National

Academic of Sciences USA 102, 3117-3122.

107

Jourdren C, Barret P, Horvais R, Delourme R, Renard M (1996) Identification of R A P D markers linked to linolenic acid genes in rapeseed. Euphytica 90, 351-357.

Karuna M N , Aswathaiah B (1989) Effect of seed vigour on field performance in beetroot and carrot. Seeds and Farms Sept-Oct, 40-46.

Kidson R, Westoby M (2000) Seed mass and seedling dimensions in relation to seedling establishment. Oecologia 125, 11-17.

Kresovich S, Williams JGK, McFerson JR, Routman EJ, Schaal B A (1992) Characterization of genetic identities and relationships oi Brassica oleracea L. via a random amplified polymorphic D N A assay. Theoretical and Applied Genetics 85,190-196.

Kumar PP, Yau JCK, Goh CJ (1998) Genetic analyses of Heliconia species and cultivars with randomly amplified polymorhic D N A (RAPD) markers. Journal of the American Society for Horticultural Science 123, 91-97.

Lamboy W F , McFerson JR, Li R, Kresovich S (1994) Relationships among Chinese vegetable Brassicas using R A P D markers. Cruciferae Newsletter 16, 44.

Lancaster R, Pasqual G (1999) Cauliflowers from Western Australia at a glance. Bulletin 4398. Department of Agriculture, Western Australia.

Lancaster R, Burt J (2001) Cauliflower Production in Western Australia. Bulletin 4521. Department of Agriculture, Western Australia.

Lakshmikumaran M , Mohapatra T, Gupta VS, Ranjekar P K (2003) Molecular markers in improvement of wheat and Brassica. In: Plant Breeding-Mendelian to Molecular Approaches. Jain H K , Kharkwal M C (Eds.). Narosa Publishing House,

N e w Delhi, India.

Lee S, Cheng H, King KE, Wang W , He Y, Hussain A, Lo J, Harberd NP, Peng J (2002) Gibberellin regulates Arabidopsis seed germination via RGL2, a GAI/RGA-like gene whose expression is up regulated following imbibition. Gene

and Development 16, 646-658.

Leroy XJ, Leon K, Branchard M (2000) Characterisation of Brassica oleracea L. by microsatellite primers. Plant Systematics and Evolution 225, 235-240.

Leviel R (1998) La sterilite male chez le chou-fleur. PHM, Revue Horticole 388, 31-33.

Lombard V, Baril CP, Dubreuil P, Blouet F, Zhang D (2000) Genetic relationships and fingerprinting of rapeseed cultivars by AFLP: consequences for varietal

registration. Crop Science 40, 1417-1425.

Lowe AJ, Moule C, Trick M , Edwards KJ (2004) Efficient large-scale development of microsatellites for marker and mapping applications in Brassica crop species. Theoretical and Applied Genetics 108, 1103-1112.

Madhavi DL, Ghosh SP (1998) Cauliflower. In Handbook of Vegetable Science and Technology. Production, Composition, Storage and Processing. Salunkhe DK,

Kadam SS (Eds). Marcel Dekker, Inc. N e w York.

Mailer RJ, M a y C E (1999) Heterogeneity of random amplified polymorphic D N A sequences in individual seedlings and bulked samples of four cultivars of Brassica

napus. Plant Breeding 118, 465-470.

Makaroff C A (1995) Cytoplasmic male sterility in Brassica species. In The Molecular Biology of Plant Mitochondria. Levings III CS, Vasil IK (Eds). Kluwer Academic

Publishers. London.

108

Malatesta M , Davey JC (1996) Cultivar identification within broccoli, Brassica oleracea L. var. italica Plenk and cauliflower, Brassica oleacea var. botrytis L. Acta Horticulturae 407, 109-113.

Malik M, Vyas P, Rangaswamy NS, Shivanna KR (1999) Development of two new cytoplasmic male-sterile lines in B. juncea through wide hybridization. Plant Breeding 118, 75-78.

Margale E, Herve Y, Hu J, Quiros CF (1995) Determination of genetic variability by R A P D markers in cauliflower, cabbage and kale local cultivars from France. Genetic Resources and Crop Evolution 42, 281-289.

Massie IH, Astley D, King GJ (1996) Patterns of genetic diversity and relationships between regional groups and populations of Italian landrace cauliflower and broccoli {Brassica oleracea L. var. botrytis L. and var italica Plenck). Acta Horticulturae 407, 45-53.

Mattingley P (2002) Cauliflowers in Western Australia, an Industry Plan. Department of Agriculture, Western Australia.

Mayberry K S (2000) Sample cost to establish and produce cauliflower. U.C. Cooperative extension. Imperial County, California.

McArthur S (1999) Winter Newsletter. South Pacific Seeds, Christchurch, N e w

Zealand.

McArthur S (2001) Winter Newsletter. South Pacific Seeds, Christchurch, N e w

Zealand.

McCubbin A, Dickinson H (1997) Self-incompatibility. In Pollen Biotechnology for Crop Production and Improvement. Shivanna KR, Sawhney V K (Eds).

Cambridge University Press, N e w York.

McVetty P B E (1997) Cytoplasmic male sterility. In Pollen Biotechnology for Crop Production and Improvement. Shivanna KR, Sawhney V K (Eds). Cambridge

University Press, N e w York.

Meng X, M a H, Zhang W , Wang D (1998) A fast procedure for genetic purity determination of head Chinese cabbage hybrid seed based on R A P D markers.

Research Note. Seed Science and Technology 26, 828-833.

Mennella G, Iori A, Sanaja V O , Magnifico V (1996) Broccoli and cauliflower cultivars identification through IE-HPLC seed protein analysis. Acta Horticulturae 407,

115-121.

Messiaen C M (1992) The Tropical Vegetable Garden. Principles for improvement and increased production with application to the main vegetable types. The Macmillan

Press, London.

Mian M A R , Bailey M A , Tamulonis JP, Shipe ER, Carter Jr TE, Parrott W A , Ashley D A , Hussey RS, Boerma H R (1996) Molecular markers associated with seed weight in two soybean populations. Theoretical and Applied Genetics 93, 1011-

1016

Mohring S Esch E, Wricke G (1999) Breeding hybrid varieties in winter rapeseed using recessive self-incompatibility. Proceedings of the 10th International Rapeseed

Congress, Canberra.

Mongkolporn O, Kadkol GP, Pang ECK, Taylor P W J (2003) Identification of R A P D markers linked to recessive genes conferring siliqua shatter resistance in Brassica

rapa. Plant Breeding 122, 479-484.

109

Monteiro A A , Lunn T (1999) Trends and Perspectives of vegetable Brassica breeding

world-wide, http://www.tropical-seeds.com/tech_forum/pubs_res/brassica.html

Monteiro A A , Gabelman W H , William P H (1988) Use of sodium chloride solution to overcome self-incompatibility in Brassica campestris. HortScience 23, 876-877.

Mueller U G , Wolfenbarger L L (1999) AFLP genotyping and fingerprinting. Tree 14, 389-394

Nandakumar N, Singh A K , Sharma RK, Mohapatra T, Prabhu K V , Zaman F U (2004). Molecular fingerprinting of hybrids and assessment of genetic purity of hybrid seeds in rice using microsatellite markers. Euphytica 136, 257-264.

Noli E, Conti S, Maccaferri M, Sanguineti MC (1999) Molecular characterization of tomato cultivars. Seed Science and Technology 27, 1-10.

Nozaki T, Kumazaki A, Koba T, Ishikawa K, Ikehashi H (1997) Linkage analysis among loci for RAPDs, isozymes and some agronomic traits in Brassica campestris L. Euphytica 95, 115-123.

Nybom H (2001) D N A markers for different aspects of plant breeding research and its applications. Acta Horticulturae 560, 63-66.

Ohto M , Fischer RL, Goldberg RB, Nakamura K, Harada JJ (2005) Control of seed mass by APETALA2. Proceedings of the National Academic of Sciences USA

102,3123-3128.

Onguso JM, Kahangi E M , Ndiritu D W , Mizutami F (2004) Genetic characterization of cultivated bananas and plantains in Kenya by R A P D markers. Scientia


Orsi CH, Tanksley S D (2005) Sw4.1, The major Q T L for seed weight variation in tomato: mapping and characterization during seed development. Plant and Animal

Genomes XIII conference, San Diego, California.

Pelletier G, Ferrault M , Lancelin D, Boulidard L (1989) C M S Brassica oleracea cybrids and their potential for hybrid seed production. 12th Eucarpia Congress, Gottingen.

Pharmawati M , Yan G, McFarlane IJ (2004) Application of R A P D and ISSR markers to analyse molecular relationship in Grevillea (Proteaceae). Australian Systematic

and Botany 11, 49-61.

Phippen W B , Kresovich S, McFerson JR (1994) Assessing genetic identity and relatedness in cabbage with RAPDs. Cruciferae Newsletter 16, 46.

Plieske J, Struss D (2001) Microsatellite markers for genome analysis in Brassica. I. Development in Brassica napus and abundance in Brassicaceae species.

Theoretical and Applied Genetics 102, 689-694.

Powell A A , Thornton JM, Mitchell JA (1991) Vigour differences in Brassica seed and their significance to emergence and seedling variability. Journal of Agricultural

Science 116, 369-373.

Prabhu K V , Somers DJ, Rakow G, Gugel R K (1998) Molecular markers linked to white rust resistance in mustard Brassica juncea. Theoretical and Applied Genetics 97,

865-870.

PradhanA Yan G, Plummer JA (2004a) Development of D N A fingerprinting keys for

the identification of radish cultivars. Australian Journal of Experimental

Agriculture 44, 95-102.

110

http://www.tropical-seeds.com/tech_forum/pubs_res/brassica.html

Pradhan A, Yan G, Plummer JA (2004b) Correlation of morphological traits with molecular markers in radish {Raphanus sativus). Australian Journal of Experimental Agriculture 44, 813-819.

Quijada PA, Udall JA, Polewicz H, Vogelzang R D , Osborn T C (2004) Phenotypic effects of introgressing French winter germplasm into hybrid spring canola. Crop Science 44, 1982-1989.

Raparelli E, Menesatti P (2000) Quality and Technological Characterization of two Cauliflower hybrids {Brassica oleraceae L. convar. botrytis L.). Acta Horticulturae 539, 109-113.

Rafalski A, Tingey S, Williams J G K (1994) Random amplified D N A (RAPD) markers. Gelvin, S. B. and R. A. Schilperoort (Eds). Plant Molecular Biology Manual. 2nd

Ed. Kluwer Academic Publ. Dordrecht. Section H/4.

Rafalski JA, Tingey S V (1993) Genetic diagnostics in plant breeding: RAPDs, microsatellites and machines. Trends in Genetics 9, 275-280.

Reddy M P , Sarla N, Siddiq E A (2002) Inter simple sequence repeat (ISSR) polymorphism and its application in plant breeding. Euphytica 128, 9-17.

Rubatzky V E , Yamaguchi M (1996) World Vegetables, Principles, production and nutritive values. 2nd Ed. International Thomson Publ. Singapore.

Ruffio-Chable V, Chatelet P, Thomas G (2000) Developmentally "Aberrant" Plants in Fi hybrids oi Brassica oleracea. Acta Horticulturae 539, 89-94.

Ruffio-Chable V, Bellis H and Herve Y (1993) A dominant gene for male sterility in cauliflower {Brassica oleracea var botrytis): phenotype expression, inheritance,

and use in Fi hybrid production. Euphytica 67,9-17.

Rukmana R (1994) Budidaya Kubis Bunga dan Brokoli. Penerbit Kanisius,

Yogyakarta.

Saal B, Plieske J, Quiros C, Struss D (2001) Microsatellite markers for genome analysis in Brassica. II. Assignment of rapeseed microsatellites to the A and C genomes and genetic mapping in Brassica oleracea L. Theoretical and Applied

Genetics 102, 695-699.

Salmon A, Manzanares-Dauleux M , Renard M , Chable V (2004) Epigenetic control of a phenotypic aberration in Brassica oleracea. Joint meeting of the 14 Crucifer Genetics Workshop and the 4th ISHS Symposium on Brassica. South Korea.

Samec P, Nasinec V (1996) The use of R A P D technique for the identification and classification of Pisum sativum L. genotypes. Euphytica 89, 229-234.

Sauer JD (1993) Historical geography of crop plants - a select roster. C R C Press, Boca

Raton, Florida.

Schaal B A (1980) Reproductive capacity and seed size in Lupinus texensis. American

Journal of Botany 67, 703-709.

Schnable PS, Wise R P (1998) The molecular basis of cytoplasmic male sterility and

fertility restoration. Trends in Plant Science 3,175-180.

Sharma SR, Singh PK, Chable V, Tripathi S K (2004) A review of hybrid cauliflower

development. Journal of New Seeds 6, 151-193.

Shellabear M (1994) Export cauliflower improvement project 1993 and 1994. Western Australian Department of Agriculture and Horticultural Research and

Development Corporation, Agriculture Western Australia, Perth.

Ill

Singh S, Gumber R K , Joshi N, Singh K (2005). Introgression from wild Cicer reticulatum to cultivated chickpea for productivity and disease resistance. Plant Breeding 124, 477-480.

Soffer H, Smith O E (1974) Studies on lettuce seed quality: IV. Individually measured embryo and seed charactereistics in relation to continuous plant growth (vigor) under controlled conditions. Journal of the American Society for Horticultural Science 99, 270-275.

Somers DJ, Rakow G, Prabhu V K , Friesen K R D (2001) Identification of a major gene and R A P D markers for yellow seed coat colour in Brassica napus. Genome 44, 1077-1082.

Song K, Tang K, Osborn TC, Lu P (1996) Genome variation and evolution of Brassica

amphidiploids. Acta Horticulturae 407, 35-44.

Stamp N E (1990) Production and effect of seed size in a grassland annual {Erodium brachycarpum, Geraniaceae). American Journal of Botany 11, 874-882.

Staub JE, Lopez-Sese Al, Fanourakis N (2004) Diversity among melon landraces {Cucumis melo L.) from Greece and their genetic relationships with other melon

germplasm of diverse origins. Euphytica 136, 151-166.

Stewart A V (2002) A review of Brassica species, cross-pollination and implications for pure seed production in N e w Zealand. Agronomy New Zealand 32, 63-82.

Stirling K, Lancaster R (2005) Alternative planting configurations influence cauliflower

development. Acta Horticulturae 694, 301-305

Susko DJ, Lovett-Doust L (2000) Patterns of seed mass variation and their effects on seedling traits in Alliaria petiolata {Brassicaceae). American Journal of Botany

87,56-66.

Swofford D L (1993) PAUP: Phylogenetic analysis using Parsimony, version 3.1 Illinois

Natural History Survey, Champaign, Illinois.

Tongue M , Griffiths P D (2004) Genetic relationships of Brassica vegetables determined using database derived simple sequence repeats. Euphytica 137, 193-201.

U N (1935) Genomic analysis of Brassica with special reference to the experimental formation of Brassica napus and its peculiar mode of fertilization. Japanese

Journal of Botany 1, 389-452.

U P O V - B M T (2002) BMT/36/10 Progress Report of the 36th session of the technical committee, the technical working parties and working group on biochemical and molecular techniques and DNA-profiling in particular, Geneva.

Van Molken T, Jorritsma-Wienk LD, Van Hoek P H W , de Kroon H (2005) Only seed size matters for germination in different populations of the dimorphic Tragopogon pratensis subsp. pratensis (Asteraceae). American Journal of Botany 92, 432-

437 Voss A, Snowdon RJ, Liihs W (2000) Intergeneric transfer of nematode resistance from

Raphanus sativus into the Brassica napus genome. Acta Horticulturae 539, 129-

134

Warren Cauliflower Group (2003) Priorities for research and development for the export cauliflower industry. Warren Cauliflower Group Incorporated, Manjimup.

Warwick SI, Soleimani V (2001) Genetic diversity in Brassical carinata, K juncea and B. nigra based on molecular A F L P markers. Cruciferae Newsletter 23, 15-16.

112

Watanabe M , Hinata K (1999) Self-Incompatibility. In Biology of Brassica Coenospecies. Gomez-Campo C (Ed.). Elsevier Science B.V. Amsterdam.

Webster R E (1964) Effect of size of seed on plant growth and yield of Fordhook 242 bush lima bean. Proceeding of the American Society of Horticultural Sciences 84, 327-331.

Wien HC, Wurr DCE (1997) Cauliflower, Broccoli, Cabbage and Brussel Sprouts. In The Physiology of Vegetable Crops. Wien H C (Ed.). CABI Publishing, Wallingford, U K .

William JGK, Kubelik AR, Livak KJ, Rafalski JA, Tingey SV (1990) DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18, 6531-6535.

Williams CN, Uzo JO, Peregrine W T H (1991) Vegetable Production in the Tropics. Intermediate Tropical Agriculture Series. Longman Scientific and Technical, Essex.

Wills A B , Fyfes K, Wiseman E M (1980) Testing Fi hybrids of Brassica oleracea for sibs by seed isoenzyme analysis. Annals of Applied Biology 91, 263-270.

Wolfe A D , Liston A (1998) Contributions of PCR-based methods to plant systematics and evolutionary biology. In Molecular Systematics of Plants II, D N A Sequencing. Soltis DE, Soltis PS, Doyle JJ (Eds). Kluwer Academic Publishers, Dordrecht, The Netherlands.

Wurr D C E (1990) Prediction of the time of maturity in cauliflowers. Acta


Yan G, Shan F, Plummer JA (2002) Genetic relationship within Boronia {Rutaceae) as revealed by karyotype analysis and R A P D molecular markers. Plant Systematics

and Evolution 233, 147-161

Yuan M , Zhou Y, Liu D (2004) Genetic diversity among populations and breeding lines from recurrent selection in Brassica napus as revealed by R A P D markers. Plant

Breeding 123, 9-12.

Zhang J (1993) Seed dimorphism in relation to germination and growth of Cakile edentula. Canadian Journal of Botany 71,1231-1235.

Zhao J, Wang X, Deng B, Lou P, W u J, Sun R, X u Z, Vromans J, Koornneef M, Bonnema G (2005). Genetic relationships within Brassica rapa as inferred from AFLP fingerprints. Theoretical and Applied Genetics 110, 1301-1314.

Zheng X Y , Liu Y (1994) Inbred testing of Chinese cabbage F! varieties by peroxidase and esterase isozyme analysis. Acta Horticulture Sinica 21, 65-70.

Zhuang FY, Chen JF, Staub JE, Qian C T (2004) Assessment of genetic relationships among Cucumis spp. by SSR and R A P D marker analysis. Plant Breeding 123,

167-172.

Zur I, Klein M , Dubert F, Samek L, Waligorska H, Zuradzka I, Zawislak E (2003) Environmental factors and genotypic variation of self-incompatibility in Brassica

oleracea L. var. capitata. Acta Biologica Cracoviensia 45,49-52.

113

Molecular investigation of genetic variation to improve uniformity of cauliflower … · Cauliflower production in Australia is export oriented. The industry aims for uniform, high

Documents