Genetic variability in tomato germplasm (solanum lycopersicum l ) using morphological characteristics and simple sequence repeat (SSR) markers

It was therefore, necessary to collate information and collect tomato genotypes from the farmers for morphological and molecular evaluation to assess the variations among and within the

KWAME NKRUMAH UNIVERSITY OF SCIENCE AND TECHNOLOGY

KUMASI, GHANA SCHOOL OF GRADUATE STUDIES DEPARTMENT OF CROPS AND SOIL SCIENCES

GENETIC VARIABILITY IN TOMATO GERMPLASM (Solanum

GENETIC VARIABILITY IN TOMATO GERMPLASM (Solanum lycopersicum

L.) USING MORPHOLOGICAL CHARACTERISTICS AND SIMPLE

SEQUENCE REPEAT (SSR) MARKERS

A Thesis submitted to the Department of Crop and Soil Sciences, Faculty of Agriculture, Kwame Nkrumah University of Science and Technology in Partial

Fulfilment of the Requirements for the Degree of

MASTER OF PHILOSOPHY

IN AGRONOMY (PLANT BREEDING)

JACINTA ADOMA OPOKU

(BSc Hons Agriculture)

AUGUST, 2015

DECLARATION

I hereby declare that this work is a direct result of my field and laboratory research undertaken and are supported by cited references in relation to other previous and similar work performed and therefore, this thesis has not been presented anywhere for a degree All references to other works have been duly acknowledged

Certified by:

ABSTRACT

Tomato is an essential ingredient in the daily diets of majority of Ghanaians and is cultivated in almost all the agro-ecological zones in Ghana, however, not much information is available on the genetic and morphological variability of tomatoes cultivated by farmers The need for varietal identification of tomato cannot be over-emphasised It was therefore, necessary to collate information and collect tomato genotypes from the farmers for morphological and molecular evaluation to assess the variations among and within the genotypes A survey was conducted and semi structured questionnaires administered to obtain information about farmers’ knowledge of varietal differences, preferences and production practices and constraints Preferred tomatoes were documented and fruits were obtained from the farmers’ field to extract seeds for the study A pot experiment was conducted to study the genetic variability among and within 12 tomato genotypes obtained from the farmers in three-selected tomato growing areas, namely Afari, Akomadan and Kumawu and their environs in Ashanti Region and planted in a randomised complete block design DNA was extracted from eight (8) plants per genotype without bulking

to check the purity of the genotypes using the CTAB method with modification by Takrama, (2000) CRIG adopted by Kirkhouse Trust Mobile Laboratory Data taken included 16 qualitative and 16 quantitative traits and scored using AVRDC descriptor list with additions from the IPRGI descriptor list It was realised that most (65 %) of the farmers saved seeds from previous harvest for the subsequent planting Analysis of variance revealed highly significant difference (P≤ 0.01) among the genotypes for all quantitative traits studied except plant height and number of diseased and pest damaged fruits Genotypic (GCV) and phenotypic coefficient of variation (PCV) were high but narrow for most of the traits studied except for days to

first flowering, days to 50 percent flowering, 100 percent flowering and plant height which recorded lower phenotypic and genotypic coefficients of variation (PCV and GCV), indicating the little environmental influence and hence, highly heritable Genotypes with higher fruit yield per plant (PF-AF and PM-AF) were from Afari and its environs The correlation analysis indicated that number of fruits per plant was highly significant and positive with fruit yield per plant Similar observation was made with plant height and days to 50 % flowering The SSR markers were highly informative as generated by the PowerMaker V3.25 software NTSyS PC software 2.2 grouped the 96 DNA samples of the 12 genotypes into two major groups The samples were reduced to 48 for clear dendrogram Samples from DW-KU were all put into one group while the rest were in the other group The outcome of this study should be useful to manage the germplasm conservation and future tomato genetic improvement However, all the genotypes may be cultivated over time at different locations on field to ascertain their stability and purity

DEDICATION

To the glory of God, I dedicate this thesis to my uncle, Mr Akwasi Adjei Adjekum

ACKNOWLEDGEMENT

I am forever grateful to the Almighty God for His protection throughout my studies I wish to express my sincere gratitude to my supervisors: Prof Richard Akromah (Provost, College of Agriculture and Natural Resources, KNUST) and Dr Charles Kwoseh (Head of Crop and Soil Science Department, KNUST) for their adept supervision and support that enabled me to complete this research successfully I thank Dr Daniel Nyadanu,(Crop and Soil Sciences Department, KNUST), Dr Amoah (CRI, Fumesua), Dr Francis Appiah (Head of Horticulture Department, KNUST), Dr B K Banful (Horticulture Department, KNUST), Mr Prince Pobi Owusu and all the lecturers and Staff of Department of Crop and Soil Sciences for their incessant support, affection and encouragement I acknowledge the following individuals for their immense technical support; Mr Ben Armooh (CRIG, Tafo), Mr Fuseini, Jamal, Ruchia (Biotechnology Lab), Mr Koranteng, Mr Osei, Mr Malik, Bismark, Pieterson, Eunice and Jemima (Pathology Lab), I also thank David Gaikpa, Kirpal, Ambrose, Ulzen, Gifty Marian, Naomi and Nana Yaw for their selfless support in data taken and analysis My utmost gratitude goes to my sponsors, AGRA (Alliance for Green Revolution in Africa) for the financial support throughout the Masters programme May God replenish everything you all lost while supporting me

To all my family and loved ones especially my mum, Georgina Cylacia Adjei, my late father, Mr Collins Opoku, Mad Faustina Akoto, my sibblings, all my aunties, uncle and cousins, I say God bless you profusely

TABLE OF CONTENT

DECLARATION i

ABSTRACT ii

DEDICATION iv

ACKNOWLEDGEMENT v

TABLE OF CONTENT vi

LIST OF FIGURES ix

LIST OF TABLES x

LIST OF PLATES xi

LIST OF APPENDICES xii

CHAPTER ONE 1

1.0 INTRODUCTION 1

CHAPTER TWO 4

2.0 LITERATURE REVIEW 4

2.1 Introduction 4

2.2 Tomato production 4

Tomato production in the world 4

Tomato Production in Ghana 5

2.3 Types or groups of tomato 6

2.4 Genotypes of tomatoes grown in Ghana 7

2.5 Genetic divergence 8

2.6 The tomato quality 9

2.7 Characterization 9

Morphological characteristics of Tomato 10

Molecular characterization 11

Marker Assisted Selection (MAS) 13

Molecular marker techniques 14

Non - PCR based technique 14

2.8 PCR-based techniques 16

Limitation of PCR 17

Optimal PCR-primers 17

Amplified Fragment Length Polymorphism (AFLP) 18

2.9 Simple Sequence Repeats (SSR) markers 19

2.10 Evaluation of molecular markers 19

CHAPTER THREE 21

3.0 MATERIALS AND METHODS 21

3.1 Introduction 21

Survey of tomato fields in selected major growing area and collection of tomato genotypes 21

Tomato genotypes obtained from farmers’ fields in the selected major growing areas 22

Extraction of tomato seeds from fruits collected from farmers during the survey 22

Field experiment: Evaluation of morphological characteristics of the tomato genotype 23

Soil used for the pot experiment 23

Nursery and transplanting of the tomato seedlings 24

Agronomic practices 24

3.2 Data collected 24

Field layout and experimental design 25

3.3 Morphological data analysis 25

Quantitative characters measured 27

Qualitative character studies 29

3.4 Laboratory experiment: Evaluation of 12 tomato genotypes from major tomato growing areas in the Ashanti Region of Ghana using SSR markers 29

Genomic DNA extraction and purification 31

DNA quality testing and quantification 32

Test run for SSR primers 33

Molecular markers and polymerase chain reactions 34

Agarose gel electrophoresis (AGE) 34

Gel scoring of DNA fragment 35

Statistical Data analysis 35

CHAPTER FOUR 37

4.0 RESULTS 37

4.1 Information on Farmers and the Study Areas 37

General socio-economic characteristics of the areas of study 37

Socio-demographic information on farmers and farming activities 37

4.2 Agro-morphological traits of the selected tomato genotypes 41

Variability in qualitative traits 41

Variability in quantitative traits 43

Cluster analysis of the morphological data 51

4.3 Correlation among yield components 53

4.4 Summary Statistics about the SSR markers used 55

Analysis of Molecular Diversity of the tomato genotypes 56

CHAPTER FIVE 61

5.0 DISCUSSIONS 61

5.1 Introduction 61

5.2 Survey of tomato fields in selected major growing area and collection of tomato genotypes 61

5.3 Agro-morphological traits of the tomato genotypes 62

5.4 Phenotypic and genotypic coefficient of variation and heritability estimates 63

5.5 Morphological relationship among genotypes 64

5.6 Correlation of yield and yield component 65

5.7 Evaluation of SSR markers in tomato genotypes using dendrogram 66

5.8 Genetic diversity within and among the genotypes 66

CHAPTER SIX 68

6.0 CONCLUSIONS AND RECOMMENDATION 68

6.1 Conclusion 68

6.2 Recommendation 70

REFERENCES 71

LIST OF FIGURES

Figure 4.1: Percentage of Gender of Respondents 38 Figure 4.2: Dendrogram generated using based on 16 traits of the twelve (12) tomato genotypes using UPGMA 51 Figure 4.3: Dendrogram based on 12 SSR markers 48 tomato accessions using SM similarity coefficient and UPGMA clustering 60

LIST OF TABLES

Table 3.1: List of tomato microsatellite markers used in the DNA fingerprinting 30 Table 4.1: Age distribution of farmers 38 Table 4.2: Educational background 39 Table 4.3: Sources of seeds 41 Table 4.4: Mean performance of twelve tomato genotypes for vegetative and

reproductive traits 45 Table 4.5: Means of morphological traits of the tomato genotypes 47 Table 4.6: Means of morphological traits of the tomato genotypes 49 Table 4.7: Range, broad sense heritability, phenotypic and genotypic coefficients of variation of the quantitative traits measured 50 Table 4.8: Correlation among yield components 54 Table 4.9: Summary Statistics of twelve SSR Markers 55 Table 4.10: Genotypes, codes and DNA sample numbers for generating dendrogram 57 Table 4.11: Clusters of DNA samples from dendrogram 58

LIST OF PLATES

Plate 4.1: (A) Skin colour of ripened fruit, (B) external colour of immatured fruits, external colour of matured fruits (C) and (D) various shapes of the transverse

section of the tomato 42

Plate 4.2: Glimpse of some phenotypic variations in the tomato genotypes 52

Plate 4.3: Primer TGS0010 showing genetic diversity in 96 DNA samples 59

Plate 4.4: Primer TGS0020 showing genetic diversity in 11 DNA samples 59

LIST OF APPENDICES

APPENDIX I: QUESTIONNAIRES FOR TOMATO FARMERS AT AGOGO,

AKOMADAN AND KUMAWU 87

APPENDIX II: Analysis of Physical and chemical properties of soil 89

APPENDIX III: Morphological Characteristics observed in the experiment based on IPGRI tomato descriptor (Darwin et al., 2003) 90

APPENDIX IV: Survey Results 92

APPENDIX V: Qualitative Traits Data in percentage 94

APPENDIX VI: ANOVA for 16 quantitative traits 96

APPENDIX VII: Dendrogram based on 12 SSR markers 96 tomato accessions using SM similarity coefficient and UPGMA 99

CHAPTER ONE

1.0 INTRODUCTION

Tomato (Solanum lycopersicum L.) belongs to the Solanaceae or nightshade family (Naz

et al., 2013) In the tropics, it is the most important vegetable crop (FAO, 2003) and the

second most important vegetable crop next to potato globally (FAOSTAT, 2014;

Fuseini, 2010; Al-Aysh et al., 2012)

In Ghana, tomato is an essential ingredient in the daily diets of majority of the

population It can be consumed as fresh vegetable or in the processed form (Al-Aysh et

al., 2012) It can also be used in the preparation of various cuisines including soups,

sauces, stew, salad and other meals Compared to other vegetables used in the country,

tomato is normally used in large quantities (Ellis et al., 1998) In the developed

countries, tomato is produced in larger quantities and processed However, in most West Africa, it is produced mainly for domestic consumption (Norman, 1992) The fruit is a rich source of fibre, potassium, vitamins C and A (AVRDC, 2004; Murphy and Trowbridge, 2011) and a good source of calcium, iron, phosphorus and sulphur It is also

a major source of lycopene found in high concentrations in processed tomato products (Di Mascio, 1998) Lycopene is an antioxidant known to combat premature aging (Wener, 2000) and quenches free radicals (Simon, 1992) Other health benefits include efficiency in neutralizing Reactive Oxygen Species (ROS), prevention of prostate cancer, heart disease and high cholesterol (Murphy and Trowbridge, 2011; AVRDC, 2004)

Genetic variability is the first step of plant breeding for crop improvement which should

be available from germplasm, the reservoir of variability for different characters (Dar and Sharma, 2011) Although tomato is cultivated in almost all the agro-ecological zones in Ghana, not much information is available on their genetic and morphological

variability (Osei et al., 2009; Robinson and Kolavalli, 2010) The problem of varietal identification in tomato by farmers cannot be over-emphasised (Blay, 1999) The

evaluation of genetic variability among and within populations of tomato genotypes can

be assessed by using morphological, biochemical and molecular characterization (Garcia

et al., 2004) Morphological characterization has been the popular tool used for the

improvement of new genotypes over the years where improved plants are developed by solely selecting plants with desirable phenotypes However, developing a new improved plant genotype by ways of phenotypic selection can easily exceed 10 years and is highly dependent on the environment for expression; hence, their ability to estimate genetic diversity in plants is reduced (Brunlop and Finckh, 2010)

The development of DNA (molecular) markers has enhanced plant genetics and plant breeding These molecular markers are effective tools for efficient selection of desired agronomic traits because they are based on the plant genotypes and are independent of

environmental variation (Franco et al., 2001) While there are several applications of

DNA markers in breeding, the most promising for cultivar development is the

marker-assisted selection (MAS) (Pervaiz et al., 2009)

Researchers have been estimating genetic variation in tomato landrace and cultivar collections using several molecular techniques including Amplified Fragment Length Polymorphism (AFLP), Restriction Fragment Length Polymorphism (RFLP), Random

Amplified Polymorphic DNA (RAPD) and Simple Sequence Repeats (SSR)

(Bredemeijer et al.,1998; Villand et al., 1998; Park et al., 2004 and Garcia-Martinez et

al., 2006) Comparatively, SSR provides co-dominant expression, highly reproducible,

locus-specific, high proportion of single-fragment amplification, high sensitivity that

allows distinctions even between closely related individuals (Bindler, 2007; Pervaiz, et

al., 2009) and cost-effectiveness (Loridon et al., 2005; Dongre and Parkhi, 2005;

Sharma and Sharma, 1999)

The main objective of the study was to determine the morphological and genetic variability of tomato genotypes grown by farmers in selected areas in the Ashanti Region of Ghana

The specific objectives were to:

i document tomatoes genotypes grown by farmers in selected areas

ii determine the variations among the tomato genotypes using morphological descriptors,

iii determine the genetic variability among the tomato genotypes using SSR markers

2.2 Tomato production

Tomato production in the world

In 2013, tomato was estimated to be about 161.8 million tonnes in the world (FAOSTAT, 2014) It is recorded that, China produces about one quarter of the world’s tomatoes, which makes them the most prominent producer followed by India, United States and Turkey (FAOSTAT, 2014) In Africa, Egypt is the leading tomato producer and the fifth producers in the world Other major producers in Africa include Nigeria, Tunisia and Morocco (Yeboah, 2011)

According to Nicole et al (2009), there is an increased importance of tomato production

in the world which has led to expansion of acreages and exportation share of many countries particularly to those located near the major importing countries China Business Intelligence Portal (2010) for instance reported that there was an increase in the world’s area of tomato with processing tomatoes output by 115.6% year-on-year to 42.317 million tonnes because of the hiked price of tomato sauce in 2008

Tomato adapts to a wide range of climatic conditions from temperate to hot humid tropics even though relatively cool, dry climate are required for high yield and better quality (Srinivasan, 2010) There is limitation in the production of tomatoes during the hot-wet season in tropical and subtropical climates which increases disease incidence

(Nicole et al., 2009)

Tomato Production in Ghana

Ghana’s commitment to the tomato sector began in the 1960s However, tomato production has failed to reach its potential in terms of the ability to sustain the few processing plants, improving the livelihoods of the households involved in tomato production and the tomato commodity chain (Robinson and Kolavalli, 2010)

Despite government interventions, farmers do not grow the right quality (prefer planting local genotypes) and could not produce the right quantity for commercial agro processing Robinson and Kolavalli (2010) observed decline in the tomato production

even though Ellis et al (1998) reported average increase in tomato production within a

year (three times a year) The average yields of tomato in Ghana is low, typically, less than 10 t/ha due to production seasonality, high perishability, poor market access and competition among the farmers and also from imports, some farmers are unable to sell

their tomatoes, which are left to rot in their fields Ellis et al (1998) also reported that,

the yield on a farm is low between 7.5 – 10 t/ha However, in this entire situation, they maintain that many farmers in Ghana prefer to cultivate tomatoes to other crops since the production is profitable (Robinson and Kolavalli, 2010)

2.3 Types or groups of tomato

Tomato has been grouped into two main types, hybrid and open-pollinated which often confuse new tomato growers (Sacco, 2008) The desirable traits of either parent of a hybrid include size of fruit; resistance to some diseases and taste whilst the negative aspect is where, their seeds could regress to either of the parent plant or would become sterile and will not produce seeds at all According to Gould (1992), even though not clearly documented, tomatoes were perhaps small-fruited initially Later stress on breeding for smooth-skinned cultivars pre-supposed that early cultivars had rough skin Open-pollination enhances gradual changes in the plant's production and immunities (Edlin, 2009) This, therefore, means that new genotypes are obtained over a period through open pollination, which averts further out-crossing Heirloom tomatoes are described as open-pollinated plants, developed directly from the seed of a previous fruit,

from a genotype that has been around for 50 years or more (Vavrina et al., 2003)

Despite the description, one chooses for heirloom because they all exhibit one peculiar

quality, open pollination (Edlin, 2009) Domesticated tomatoes (Solanum lycopersicum)

are naturally self-pollinating and usually do not outcross Hence, they quickly become homogenous and produce ‘true-to-type’ seed (Edlin, 2009) Colours of heirlooms vary from yellow, red, orange, purple, white, green, and bicolour combinations of them all There is also wide genotype in their shapes and sizes, which ranges from tiny cherries to huge two-pounders in the same garden, along with egg, flattened, globe, oblong,

pumpkin, pear, and pepper shaped fruits (Edlin, 2009)

2.4 Genotypes of tomatoes grown in Ghana

There are many tomato genotypes grown in Ghana These include Bolga and Ashanti

(Adubofour et al., 2010) and Powerano (Robinson and Kolavalli, 2010), as genotypes

mostly grown in Brong Ahafo Region under rain-fed conditions Other genotypes are

‘Rasta’, ‘Fadebegye’, ‘Mmoboboye’ and ‘Amo’ as described by Osei et al (2014),

Nimagent F1 supplied by Trusty Foods, grown under both irrigated and rain- fed conditions,’ In Greater Accra Region Nimagent and ‘Ada lorry tyre’ are grown under rain-fed conditions However, ‘Wosowoso’, Roma VF, Cac J, Pectomech, Laurano, Pectomech VF, Tropimech, Rio Grande, were the genotypes recommended by Ministry

of Food and Agriculture of Ghana (2008) which were obtained by farmers from reputable seed dealers and other farmers Robinson and Kolavalli (2010) found Pectomech genotype suitable for processing and preferred by consumers hence achieved

premium price over the local genotypes According to Clottey et al (2009), genotypes

Pectomech, Tropimech, and Roma grown mostly under rain-fed conditions in the Greater Accra Region were also found to be the major tomato genotypes grown in Vea

in the Upper East Region, a notable tomato growing area Some genotypes were recognised as major local open pollinated that farmers can wash and recycle These included genotypes such as ‘Power’ Powerano, Rasta’and Wosowoso, with Powerano

often being preferred due to its high tolerance and/or resistance to diseases Ellis et al

(1998) reported that the `Power' genotype was the genotype mostly cultivated in Ghana

2.5 Genetic divergence

The evaluation of genetic variability using quantitative traits has been of major significance in many contexts particularly, in distinguishing well defined populations Heterogeneous sets of groups exist in a self-pollinated crop in a germplasm since each

group is homozygous within itself (Yu et al., 2000) It is critical to select parents in such

crops for breeding programme because, the success of such programme depends upon the segregants of hybrid derivatives between the parents, mostly when the aim to

improve the quantitative characters such as yield (Yu et al., 2000) To help the breeder

in the process of identifying the parents, several methods of divergence analysis based

on quantitative traits been proposed (Walker and White, 2001) Among them, Unweighted Pair Group Method with Arithmetic averages (UPGMA) is one of the most prevalent procedures and an effective method to measure the degree of variability among genotypes, based on their pairwise similarities in relevant traits (Legendre and Legendre, 1998)

Currently, UPGMA has been mostly used to produce lead trees for more advanced phylogenetic reconstruction algorithms even though it was designed initially for use in protein electrophoresis studies (Walker and White, 2001) The UPGMA algorithm constructs a rooted tree (dendrogram) that reflects the structure present in a pairwise similarity or a dissimilarity matrix) The closest two clusters are joined into a higher- level cluster at each step The average of all distances between any pairs of objects “x”

in A and “y” in B is taken to be the distance between the two clusters A and B The mean distance between elements of each cluster is shown below:

1 ∑∑ d(x,y)

The

This method is generally attributed to Sokal and Michener (1958) Fionn Murtagh

(1983) found a time optimal (n 2) time algorithm to construct the UPGMA tree

2.6 The tomato quality

The improvement of crop species has been a principal human goal since cultivation began Tomato is a major economically important crop with several characteristics used

in establishing a model system for segmentation of genetic factors of quantitative trait loci (QTL) Several wild-related species in tomato have been confirmed to be unexploited sources of appreciated genetic variability These include pathogen-resistance genes, and industrial and nutritional quality traits Recent medical research on fruit compositional quality for social health has also motivated crop improvement

strategies (Pestana et al., 2003) This is seen in the interest shown to enhance antioxidant

compounds present in fruit and vegetables, which may help, prevent chronic diseases

such as cancer, arthritis and heart disease in the past century (Harrigan et al., 2007) The

tomato quality traits of interest to both fresh market and processing tomato industries include fruit size, shape, total solids, colour, firmness, nutritional quality and flavour The pH, titratable acidity, vitamin contents and low seed load are the other important fruit quality characteristics of tomato (Foolad, 2007)

2.7 Characterization

The expression of plants of highly heritable characters is determined by characterization These characters range from phenotypical or agronomical triats to seed proteins or

molecular markers (Valpuesta and Botella, 2004) Characterization based on differences between morphological and phenological of high heritability is approximated The variations in the germplasm of the plants are estimated, therefore the diversity in a collection of germplasm can only be said to have been studied when characterization is done This kind of variations can only be expressed by molecular markers Such variation may also include characteristics whose expression is little influenced by the

environment (Bhatt et al., 2004) Suitable characterization for morphologic and

agronomic traits is important to facilitate the use of germplasm by breeders This could

be achieved by characterizing germplasm accessions of all crops for morphological and agronomic traits in batches over the years (Tanksley, 2004)

Morphological characteristics of Tomato

Traditionally, cultivar identification and morphological characterization in tomato have

been done by evaluating stem, flower and fruit characteristics (Susic et al., 2002) Susic

et al (2002) acknowledged the significance of morphological characterization in

identifying duplicate accessions, discovery of exceptional traits and in conservation of the population structure, thus saving on storage space and streamlining selection by plant breeders It has been used for studies of genetic diversity patterns and correlation with characteristics of agronomic importance (CIAT, 2007) Tomato cultivars are normally renowned based on morphological traits that have a wide variability of botanic characteristics

2.7.1.1 Descriptor list of tomato

Descriptor lists are those characteristics by which germplasm can be identified and its prospective usefulness determined (IPGRI, 2003) Descriptors used for characterization allow comparatively easy discrimination among phenotypes Those related to phenotypic traits mostly relate to the morphological description of the plant and its architecture (CIAT, 2007) Standard descriptor lists provide an international format thereby producing a universally understood language for plant genetic resource data (IPGRI, 2003)

2.7.1.2 Limitations in morphological studies

There are limitations in assessing tomato diversity morphologically due to the tendency

of a species changing in appearance substantially in response to environment

(Weerasingh et al., 2008) They identified disease as one of the key environmental

factors for the conservation of genetic variations in plant populations Pathogens such as viruses can tremendously affect the structure, diversity and functioning of the plant populations (Brummell and Harpster, 2001) Variations in environmental conditions could permit variations in results obtained if morphological characterization is repeated spatially and periodically (Miller and Tanksley, 1990) Therefore, observation of genetic diversity based on only high morphological diversity among the tomato accessions may not be conclusive indication of genetic diversity (Zamir, 2001)

Molecular characterization

It is very imperative to complement morphological characterization with molecular characterization due to changes in environment and sometimes mutation in tomatoes

The biochemical and molecular markers can be used to study genome itself directly These methodologies give greater accuracy in locating genes of interest but do not

consider the environmental effect on the expression of those genes (Tomar et al., 2004)

Molecular biology has advanced especially in the development of the polymerase chain reaction (PCR) for amplifying DNA, DNA sequencing and data analysis, hence powerful techniques can be used for screening, characterization and evaluation of genetic diversity DNA fingerprinting has become an important tool for cultivar identification in plant breeding and for germplasm management A number of 14 different molecular assays have been applied in tomato These include DNA Amplification Fingerprinting, Random Amplified Polymorphic DNA (RAPD), Selective Amplification of Microsatellite Polymorphic Loci, Inter Simple Sequence Repeat, Simple Sequence Repeat (SSR), Amplified Fragment Length Polymorphism (Zhang and Stommel, 2000) One significant objective of germplasm characterization is to identify the accessions of a germplasm collection so that, they can be distinguished or individualized clearly (CIAT, 2007) Many molecular markers, especially those associated with PCR based methods, including Restriction Fragment Length Polymorphism (RFLP), Random Amplified Polymorphic DNA (RAPD) and microsatellites or Simple Sequence Repeats (SSR), have been widely used to estimate

genetic variability and phylogenetic studies in banana (Creste et al., 2004; Amorim et

al., 2011) Of these, microsatellite technique has higher potential use because it permits

the detection of greater polymorphism and of co-dominant inheritance, and offers high reproducibility and easy interpretation The introduction of molecular biology in the 80s elevated great hopes in terms of characterization of the genetic diversity present in both wild and cultivated sections There has been a great expectation since the development

of molecular techniques to ‘pinpoint’ genomic regions involved in targeted traits Ad hoc techniques from quantitative genetics have made it possible for dissection of the genetic control of complex traits leading to the identification of key alleles involved in diverse agronomic traits, which originated from several wild relatives Now, the tomato genome is fully sequenced and new phase in the knowledge on tomato diversity with the

so called “-omics” and next generation sequencing techniques is coming (Goncalves et

al., 2009) These technologies and related data analysis permits a complete and

combined reading of genomes and related levels of expression (transcriptome, proteome, metabolome) in a high throughput way Among the new approaches, QTL mapping techniques in natural populations or genome wide association studies facilitate the genetic characterization of complex traits and germplasm management of both wild and cultivated tomatoes

Marker Assisted Selection (MAS)

DNA-based markers for genetic study and management of significant agronomic traits have become a progressively valuable tool in plant breeding MAS is a technique of

choosing desirable organisms and their traits values in a breeding scheme based on DNA

molecular marker patterns The introduction of DNA-based genetic markers has now facilitated the identification of large numbers of markers dispersed throughout the genome of any species of interest MAS reduces cost of field evaluation, increases

breeding efficiency, and allows simultaneous selection, for example, disease resistance

and other agronomic traits DNA-based markers have been useful in backcrossing of

resistance loci into elite cultivars (Babu et al., 2004) and in selection of alleles with

major effect across multiple populations (Ejeta et al., 2000) However, their greatest

prospective seems to be in increasing the rate of gain from selection for desirable

genotypes and in the manipulation of quantitative trait loci (QTL) that form complex economic traits DNA markers also allow the location of chromosome of various

interacting genes that condition complex agronomic traits to be identified (Pervaiz et al.,

2009) Genetic mapping is important for effective manipulation of important genes

Molecular marker techniques

Major mutants such as morphological, anatomical, or behavioural differences have increased the ability to understand genetic relationship among organisms at the molecular level

Molecular markers have several merits over morphological markers because relatively large number of alleles can be found, codominant mode of inheritance exhibited and virtually having no effect on the phenotype (Nadarajan and Gunasekaran, 2005) Also, it has the ability to determine genotypes at a very early developmental stage, which allows early screening methods to be applied Many molecular marker techniques could be employed for genetic linkage mapping They can easily be divided as based whether they are amplified on a PCR or not

Non - PCR based technique

2.7.5.1 Restriction Fragments Length Polymorphism (RFLP)

RFLPs have yielded very fascinating results in its application in crop genetics and

breeding as genetic marker (Bhatt et al., 2004) This technique alters restriction

fragments length by digesting with restriction-enzymes type, therefore, reflecting the differences in homologous DNA sequences, which results from translocations or

inversions of base pairs at the recognition site of restriction enzyme or from internal deletion/insertion events Restriction fragments are separated according to their size by

agarose gel electrophoresis RFLP markers have codominant expression, which do not

have pleiotropic effects on agronomic traits, which provides virtually infinite number of

possible markers (Bernardo, 2002) In addition, RFLP can be used to detect the patterns

in hybridization of all the available probes with the same laboratory method, unlike biochemical markers such as isozymes Isozymes require different staining and electrophoretic techniques for each isozyme and detect much more polymorphism than biochemical markers because many of the probes are non-coding or less conserved

sequences (Bakhsh et al., 2006)

RFLPs have made possible the dissection of quantitative traits into Mendelian factors This method has been most used for genetic mapping to obtain detailed maps of genetic linkage in many crops such as rice lettuce, potato, tomato, maize and barleys which can

be used to examine QTLs and also enhance the selection efficiency for traits It has also been used in identification of genotypes in barleys, beets, maize and potato

(Hariprasanna et al., 2006)

2.7.5.2 Drawbacks of RFLPs assays

RFLPs assays are time consuming and require expensive laboratory supplies and

manipulation of radioactive isotopes however, the development of non-radioactive

detection methods is reducing the cost and simplifying it (Arun et al., 2011) It remains

laborious, and limits the level of polymorphism particularly, for crops with a narrow

genetic base such as cotton, soybean, tomato and wheat (Liu et al., 2003a) For instance,

in cultivated wheat, the polymorphism level of RFLP is low ranging from 20 to 38%,

because of this limited polymorphism, gene and genome mapping has required the use

of populations derived from wide crosses However, mapping many agronomically

important genes or QTL, which is a major goal in plant breeding, needs informative

markers in an intraspecific context (Nachit et al., 2001) This is mostly true for

marker-assisted selection RFLPs identified with single-copy genomic and cDNA clones are

exceptionally powerful for comparative mapping approaches (Yu et al., 2000) They are

only limited for intraspecific molecular analysis of agronomic traits

2.8 PCR-based techniques

One of the most useful techniques among numerous methods for the detection of DNA polymorphism appears to be PCR, which allows effective amplification of target known sequences (Benito, 1993) This makes the technique specific, reliable, and repeatable

(Lashermers et al., 1994) PCR has become the standard technique in plant molecular

biology because of its efficiency, ease, and versatility PCR offers a less expensive,

technically demanding and more rapid methodology than RFLP The major advantage of PCR is that, it permits synthesis and sharing of primer sequences, which prevent their exchange between biological laboratories, which is vital to RFLP clones Principally, this method can be developed for any part of the genome targeted provided the nucleotide-sequence information is available or can be obtained from RFPL probes readily

In addition, this method permits the efficient screening of large populations PCR

primers also known as Simple Tagged Sequences (STS) are short, unique sequence that can be amplified by PCR and that identifies a known location on a chromosome This

makes them superior especially in map construction to multilocus DNA marker types DNA polymorphism obtained with the PCR has been used as genetic markers to tag

genes, to fingerprinting plants, viruses, fungi, bacteria, and humans as well as to

determine genetic relationships (Premalatha et al., 2006) It has been used in genomics

to gene identification

Limitation of PCR

There has been a recurrent observation that results from a particular primer may differ

between laboratories (He et al., 2003) hence, limit sharing of primer sequences among

laboratories PCR can only effectively amplify within a certain size range of DNA, and

Taq DNA polymerase can introduce errors He et al (2003) reported that the

accumulated mutation rate after 20-30 cycles was reported to be 0.3-0.8 % In addition, the primer must be highly specific to the target sequence to increase priming and resultant amplification that is reproducibility

Optimal PCR-primers

There have been studies conducted on the definition of ideal Sequence-Tagged Sites (STS) The following are the criteria: the primer must be specific to be maintain specific and should be short preferably18 and 22 bp to give sufficient base pairing for stable

duplex formation (Peirson et al., 2003)

Stable duplex formation means that both primers in a PCR reaction should have similar melting temperature to obtain the same hybridization kinetics during the template-annealing phase The most preferred primers are the one with overall G+C of 45-55 %, a high GC content resulting in lower repetition (Borneman and Jack Hartin, 2000) The stability of the 5’ and 3’ end must be considered High 3’stability inhibits repetition

hence further information which are not always unavailable on the DNA sequence would be required

2.8.2.1 Avoidance of hairpin

In addition, the primers should be free of dimers and hairpins (palindromes (complementary) within themselves This is to obviate the folding back of the primers resulting in stable intrastand structure which could inhibit primer annealing to the template DNA leading to unproductive priming event The primers should be free of dimers means, they should not include nucleotide sequences particularly those that would complement at the 3’ end, which would permit one primer molecule to anneal to itself, or to other primer used in a PCR reaction (primer dimer formation) This would avoid competition for amplification, which could decrease the overall signal that would

be obtained

Amplified Fragment Length Polymorphism (AFLP)

AFLP technique is highly reproducible and can be applied to all species (Yousef and Juvik, 2001) This technique is based on the selective amplification of a limited number

of DNA restriction fragments cut out of complex plant genomic DNA by restriction enzymes This technique could bridge the gap between genetic and physical maps since most AFLP fragments correspond to unique positions on the genomes, which can be exploited as landmarks in genetic and physical maps (Yousef and Juvik, 2001) However, it is less efficient than RFLP in terms of synteny studies (Tanksley, 2004) AFLPs markers are locus specific but only at species level The technique provides concurrent coverage of many loci in a single assay This can be used to generate DNA

fingerprints of the complexity required by altering the number of selective bases employed Hence, proving to be an invaluable tool for studies of diversity, especially in species where other generation’s markers, such as microsatellites are unavailable

(Bindler et al., 2007)

2.9 Simple Sequence Repeats (SSR) markers

Simple sequence repeats (SSR) are regions of DNA that consist of short, tandem repeated units (2-6 bp in length) found within the coding or non-coding Regions of all

eukaryotic organisms (Sulodhani et al., 2005) Particular primers (generally 20–25 bp)

can be designed to amplify the microsatellite by PCR if nucleotide sequences in the flanking regions of the microsatellite are recognised Conversed DNA sequences flanking the SSR as primers can be used to detect different alleles detected at a locus by

PCR (Levin, 2004) Levin (2004) has reported the success in using SSR markers to

detect polymorphism between parent cultivars SSR markers are both easy and inexpensive for analysis even though they are costly to develop relatively They are highly polymorphic in nature (high information content) and excellent genetic markers for many types of investigations, including marker assisted selection and fingerprinting

of germplasm collections (Anne and Donald, 2010) They have the ability to detect different alleles at a locus by PCR using conserved DNA sequences flanking the SSR as primers (Singh and Westermann, 2002)

2.10 Evaluation of molecular markers

Hildebrand and Subrahmanyam (1994) reported that, polymorphic information content (PIC) value is an important primary data used to determine the informativeness and

usefulness of a molecular or DNA marker Markers with high PIC values are said to be highly polymorphic and thus detect higher level of genetic variation in an organism That is, PIC value of 0.70 and above is described as highly informative while a value of 0.44 is moderately informative

Survey of tomato fields in selected major growing area and collection of tomato genotypes

There were plethora of commercially available tomato genotypes that had been cultivated by the respondent farmers before but were not being cultivated again since the market women no longer demand them due to their unavailability and consumer preference, among other reasons Twenty farmers from each of the three selected tomato-growing areas in Ashanti Region were interviewed using semi-structured questionnaires (Appendix I) to find out the type of tomatoes that they cultivated, among others The areas selected were Akomadan, Afari, Kumawu, and their environs,

respectively A total of 60 tomato farmers were randomly selected from these areas and interviewed after which two farms were visited at each area Matured tomato fruits were harvested directly from the plants for all the genotypes on the farms in order to get true-to-type fruits

Tomato genotypes obtained from farmers’ fields in the selected major growing areas

Thirteen tomato genotypes were obtained from the farmers’ fields From Afari and its environs, six tomato genotypes, namely ‘Atoa’, ‘Daagyine’, ‘Local 1’, ‘Power’, Pectofake 1 and Pectomech were obtained At Akomadan, three genotypes, ‘Akoma’, Pectofake 2 and Powerano while the four genotypes were obtained from Kumawu and its environs They were ‘Bolga’, ‘Dwidwi’ (cherry), ‘Local 2’ and Rano Unfortunately, tomato genotype Rano did not germinate hence, the genotypes used for this research was limited to twelve (12) genotypes

Extraction of tomato seeds from fruits collected from farmers during the survey

Tomato seeds were extracted from fruits of each of the genotypes collected using the

fermentation method (Vanangamudi et al., 2006) The seeds of each genotype were

squeezed out into a containers filled with water enough to cover the seed and left overnight The seeds were then extracted using a sieve and left to dry under shade in an open area for the days

Field experiment: Evaluation of morphological characteristics of the tomato genotype

The field experiment was conducted in 6 liter pots at the Department of the Crop and Soil Sciences experimental field at the Faculty of Agriculture, KNUST, Kumasi It lies between latitude 6.68o N, and longitude 1.57o W

Soil used for the pot experiment

The physical and chemical components of the soil were analysed and results shown in Appendix II Soil samples were thoroughly mixed, bulked, air-dried and composite samples taken for physico-chemical analyses and biological assays using standard protocols Soil pH was determined according to the electrometric method described by

Page et al (1982) in a suspension 1:2.5 soil to distilled water (soil:water) ratio The

modified Walkley and Black procedure as described by Nelson and Sommers (1982) was used to determine organic carbon content in soil sample The Kjeldahl method involving digestion and distillation as described by the Soil Laboratory Staff (1984) was used in the determination of total nitrogen The readily acid-soluble forms of phosphorus were extracted with Bray No 1 solution (HCl:NH4F mixture) (Bray and Kurtz, 1945) Particle size distribution was determined by the hydrometer method (Day, 1953)

Potassium was determined in 1 M ammonium acetate (NH4OAc) extract (Black, 1986) Sterilisation was done by heating and allowed to cool after three hours This was used to fill up to the first line mark in the pots which weighed approximately 19.2 kg

Nursery and transplanting of the tomato seedlings

Seeds from the various tomato genotypes were planted separated in wooden boxes with the sterilised sandy loan soil In the nursery, the germination percentages of the seeds of the tomato genotypes were determined After three weeks, the tomato seedlings were transplanted into pre-weighed pots containing sterilized well-drained sandy-loam soil

Agronomic practices

At transplanting, YaraMila Winner (150 kg /ha) and YaraLiva Nitrabor (AI: Nitrate-N, calcium and boron) (50 kg/ha) were used as basal fertilizer respectively Watering was done immediately after transplanting The plants were watered at least once a day manually to prevent waterlog NPK 30:10:10 fertilizer was applied at 10 days after transplanting and urea at two weeks afterwards Weeds were controlled as and when they emerged in the pot by hand Staking of plants was done after a month to permit easier observation Power 76WPTM 40-60 g - fungicide (AI: cymoxanil (ethyl urea) 60 g/kg + propineb (dithiocarbamate) 700 g/kg) dissolved in 15 liters of water was sprayed every two weeks The insecticides were used as follow: 20 ml Golan SL (acetamiprid

200 g/l) in 15 liters of water at vegetative stage, 35 ml Rimon Star 65EC(AI: Novaluron

35 g/l + Bifenthrin 30 g/l) in 15 liters of water after one week at flowering stage and at fruiting stage 50 ml of Karat in 15 liters of water

3.2 Data collected

All the plants in one trial were used for the recording for the morphological characters and were based on modified IPGRI (International Plant Genetic Resources Institute)

together with modified AVRDC tomato descriptors (Darwin et al., 2003) shown in

Appendix III

Field layout and experimental design

The size of the field for the trial was 26.27 m2 (3.55 m x 7.40 m) The experiment was laid out in a randomized complete block design with three replications Each replication contained 12 tomato genotypes as treatments Each replication measured 6.39 m2 (3.55

m x 1.80 m) with 1 m alley between blocks Each plant was planted per pot There were five plants per tomato genotypes within a block The size of each pot was 6 L The average weight per pot filled with steam-sterilized sandy-loam soil was 19.20 kg The surface area of each of the pots was 471.44 cm2 The height of each of the pots measured

20 cm From the ground layers three black polythene sheets were spread on the field before the pots were placed on it The total plant population was 180

3.3 Morphological data analysis

SPSS statistics V16 was used to analyse the frequencies and computed the percentages

of the qualitative data The quantitative data were subjected to Analysis of Variance (ANOVA) using the GenStat Statistical package version 11.1 (GenStat, 2010) to calculate the relationship between the traits of the genotypes Least significant difference (lsd) at 5% was used to separate the treatments means

Components of variance, σ2g = genotypic variance, σ2p= phenotypic variance, σ2e = error variance, PCV= phenotypic coefficients of variation and GCV= genotypic coefficients of variation were estimated using the following formula (Wricke and Weber, 1986);

PCV and GCV were grouped as:

0-10 % = low, 10-20 % = moderate and 20 % and above = high (Shivasubramanian and Menon, 1973)

Heritability in broad sense was computed using the following formula (Allard, 1960; Weber and Moorthy, 1952)

Heritability (H2bs) = σ2g x 100 and classified as;

0-30 % = low,

30-60 % = moderate and

60 % and above = high (Robinson et al., 1949)

The genetic advance (GA) expressed under selection in broad sense, assuming selection

intensity of 5 % was estimated in accordance with the formula illustrated by Johnson et