Citation
Genetics of resistance to fusarium wilt in tomato

Material Information

Title:
Genetics of resistance to fusarium wilt in tomato
Creator:
Bournival, Brian, 1961-
Publication Date:
Language:
English
Physical Description:
vii, 81 leaves : ill., photos ; 29 cm.

Subjects

Subjects / Keywords:
Chromosomes ( jstor )
Diseases ( jstor )
Enzymes ( jstor )
Fusarium ( jstor )
Gels ( jstor )
Genes ( jstor )
Genetic loci ( jstor )
Genetics ( jstor )
Pathogens ( jstor )
Tomatoes ( jstor )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1989.
Bibliography:
Includes bibliographical references (leaves 71-80).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Brian Bournival.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Brian Bournival. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
022623916 ( ALEPH )
22712602 ( OCLC )

Downloads

This item has the following downloads:


Full Text












GENETICS OF RESISTANCE TO FUSARIUM WILT IN TOMATO

















By

BRIAN BOURNIVAL


















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1989














This dissertation is dedicated in loving memory to my aunt, Arline Bournival, who died during my tenure at the University of Florida.














ACKNOWLEDGEMENTS


I would like to extend my gratitude to Dr. Jay Scott and Dr. Eduardo Vallejos, my coadvisors in this project. Without their assistance and guidance, this study would not have been possible. I would also like to thank my committee members, Dr. Corby Kistler, Dr. Curt Hannah, and Dr. J.P. Jones, for their invaluable contributions to the project. Sincere gratitude is also extended to Dr. Dan Cantliffe for making his growth chambers available for the inoculations. I also thank Bruce Ritchings for watering my plants every other weekend and allowing me to leave Gainesville occasionally. Thanks is also extended to Patty Hart for being a special friend and getting me out of the lab and into other activities. Lastly, I would like to thank my parents, Herb and Neva Bournival, and my aunt and cousins in St. Petersburg for their continued support throughout my work here.












iii















TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS . .................. iii

KEY TO ABBREVIATIONS . ............... . v

ABSTRACT . . . . . . . . . . . . . . . . . . . . ... vi

CHAPTERS

1 INTRODUCTION . ................. 1

2 LITERATURE REVIEW . ............. . 4

ISOZYMES in BREEDING and GENETICS . ... . 4 FUSARIUM WILT of TOMATO . ....... . . 10

3 MAPPING OF NEW ENZYME LOCI . .......... 20

INTRODUCTION . .. ........... . 20
MATERIALS and METHODS . . . . . . . . . . . 21
RESULTS and DISCUSSION . ...... .. . 23

4 ANALYSIS AND MAPPING OF GENES CONTROLLING
RESISTANCE TO RACE 3 OF FUSARIUM WILT . . . . 36

INTRODUCTION . ............. . 36
MATERIALS and METHODS . . . . . . . . . . . 37
RESULTS and DISCUSSION . ....... . . 41

5 GENETIC ANALYSIS OF RESISTANCES TO RACES 1
AND 2 OF FUSARIUM WILT . ........... 54

INTRODUCTION . ..... ........ . . 54
MATERIALS and METHODS . . . . . . . . . . . 55
RESULTS . ................. . 57
DISCUSSION . ........ ....... 60

6 SUMMARY AND CONCLUSIONS . . .. ... . . . . 67

APPENDIX . . . . . . . . . . . . . . . . . . . . ... 69

REFERENCES .............. ... .... 71

BIOGRAPHICAL SKETCH . ................. 81

iv















KEY TO ABBREVIATIONS


Name Abbreviation

Acid phosphatase Aps Aconitase Aco Alcohol dehydrogenase Adh a-Mannosidase Aman P-N-Acetyl glucosaminidase Bnag Diaphorase Dia Esterase Est Glutamate oxaloacetate transaminase Got Hexokinase Hk Mannose-6-phosphate dehydrogenase Mpi Peroxidase Prx Phosphoglucose isomerase Pgi Phosphoglucomutase Pgm 6-Phosphogluconate dehydrogenase 6Pgdh Punctate Pn Shikimate dehydrogenase Skdh Sucrose synthase Ss Superoxide dismutase Sod







V














Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

GENETICS OF RESISTANCE TO FUSARIUM WILT IN TOMATO By

Brian Bournival

December 1989

Chairman: Dr. John W. Scott
Major Department: Vegetable Crops

The inheritance and linkage relationships of

resistances to races 1, 2, and 3 of Fusarium oxysporum f. sp. lycopersici derived from Lycopersicon pennellii (LA 716) were analyzed in an interspecific backcross to L. esculentum. Progenies from each backcross (BC1) individual were inoculated with each of the races, and their responses measured according to a visual rating system; progeny responses were used to calculate a mean disease rating for each BC1 individual for each race. For all three races, the frequency distribution of the disease ratings was bimodal, indicating that resistance was controlled by one major locus in each case. To map resistance gene(s), disease ratings were compared between homozygotes and heterozygotes at each of 23 segregating marker enzyme loci. For all three races, highly significant differences were detected for the chromosome segment marked by the Got-2 locus on chromosome vi








7. This indicated that Got-2 was linked to a major gene or genes controlling resistance to the three races. It was not possible with available data to determine the number of genes on chromosome 7 conferring resistance to the three races.

To determine whether Got-2 could serve as a selectable marker for the race 3 resistance locus, designated 1-3, the genotype of each BC1 individual at 1-3 was determined using cluster analysis of the disease ratings; a test for independent assortment between I-3 and Got-2 revealed strong linkage, with an estimated map distance of 2.5 cM. The Got2 locus is proposed as a selectable marker to expedite the transfer of resistances to races 1, 2, and 3 into commercial tomato cutivars.

In addition, a minor factor controlling resistances to all three races was detected on chromosome 8 linked to Aps2; this locus has been designated Tfw for "tolerance to Fusarium wilt." Tfw might also serve as a source of genetic resistance to Fusarium wilt by increasing stability of resistance and reducing the probability of new race formation by the pathogen.












vii














CHAPTER 1

INTRODUCTION


Fusarium wilt of tomato, caused by races 1 and 2 of Fusarium oxysporum (Schlecht.) f. sp. lycopersici (Sacc.) Snyder and Hansen, is a potentially devastating disease found in many tomato growing regions. Control is mainly through the use of genetically resistant cultivars; chemically, only fumigation before planting has proven effective. Monogenic resistances to races 1 and 2 (I and I2, respectively), derived from accessions of Lycopersicon pimpinellifolium (Bohn and Tucker, 1939, 1940; Alexander and Hoover, 1955; Stall and Walter, 1965), have been transferred to many cultivars. Race 3, a new race with the ability to infect cultivars resistant to races 1 and 2, has been observed in Florida (Volin and Jones, 1982), Australia (Grattidge and O'Brien, 1982), and recently in California (Davis et al., 1988). Currently, no cultivars carry resistance to race 3. Dominant monogenic resistance to race 3, designated 1-3, has been observed in two accessions of the wild tomato species L. pennellii, PI414773 (McGrath et al., 1987) and LA 716 (Scott and Jones, 1989).

In a standard breeding program, transfer of a disease resistance gene from a wild species to a domesticated

1








2

cultivar is a time-consuming process requiring inoculations each generation. Tanksley and Rick (1980b) proposed the use of isozymes as selectable markers to expedite the transfer of important genes from one genetic background to another. Previous investigators analyzing crosses between L. esculentum and L. pennellii have identified several polymorphic enzyme loci dispersed throughout the tomato genome that could serve as potential selectable markers for important traits (Tanksley and Rick, 1980b; Chetelat, 1989). Use of an isozyme locus as a marker for a disease resistance gene would have many advantages including the following: 1. it is nondestructive to the plant; 2. screening can be done in a matter of hours on very young plants; 3. the enzyme genotype does not affect the overall plant phenotype; 4. migration rate of bands is not affected by the environment; 5. isozymes are codominantly expressed allowing for the differentiation of homozygotes and heterozygotes; 6. additional traits can be screened; and

7. the breeder in uninfected regions would not have to work with the pathogen and risk introduction.

A major problem associated with the use of genetic

resistance as a control for a disease is the overcoming of resistance by the pathogen through mutation. In addition to race 3 resistance, L. pennellii (LA 716) represents a potential new monogenic source of resistance to races 1 and

2 (Scott and Jones, 1989). Incorporation of additional









3

genetic resistance to these two races may reduce the likelihood of races 1 and 2 of the pathogen developing into a new race. The objectives of this work were to identify and map new polymorphic enzyme loci, analyze race 3 resistance derived from L. pennellii (LA 716) and tag I-3, and determine the genetic relationship between LA 716 resistances to races 1, 2, and 3.















CHAPTER 2

LITERATURE REVIEW


ISOZYMES in BREEDING and GENETICS Uses of Isozymes

Isozymes can be described as multiple molecular forms of the same enzyme. To analyze the isozymes of a given enzyme system, they must first be extracted and separated. This is usually done through gel electrophoresis. In this technique, plant tissue samples are ground in an extraction buffer and loaded into a gel. The gel usually consists of starch or polyacrylamide and any of a wide variety of buffers ranging in pH from 4.6 to 9.0 (Shaw and Prasad, 1970). For starch gel electrophoresis, loading is done by absorbing the extract onto filter paper wicks and placing the wicks in a cut made approximately 1/3 of the way up from the bottom of the gel, whereas for polyacrylamide gel electrophoresis, the liquid extract is placed directly into preformed wells at the top of the gel. An electrical potential is applied across the gel, and the enzymes migrate in the gel according to their molecular weight and charge. After separation, zones of enzyme activity are resolved by incubating the gel in a specific activity stain.


4









5

The banding pattern of an enzyme system is referred to as a zymogram. Bands produced by the same locus are referred to as allozymes, whereas bands produced by different loci are termed isozymes. Herein, to avoid confusion, all variant bands of the same enzyme system will be referred to as isozymes.

There are a variety of reasons why plants have more

than one locus producing the same enzyme. Enzyme loci can be tissue and/or organelle specific. For example, there are two phosphoglucomutase (PGM) loci in most plant species. One locus is active in the chloroplasts and is specific to the green tissues of the plant, whereas the other locus is expressed in the cytoplasm and is not tissue specific (Gottlieb, 1981). Both loci are required for normal plant growth. Chloroplast PGM is involved in starch metabolism, and cytoplasmic PGM is necessary for sucrose synthesis.

Having more that one locus for a critical enzyme affords the plant a backup mechanism in case one locus fails. In maize there are two alcohol dehydrogenase (ADH) loci (Schwartz, 1969; Freeling and Birchler, 1980). ADH catalyzes the reduction of acetaldehyde to ethanol, generating NAD+ oxidizing power which is important for plant survival under anaerobic conditions. Schwartz (1969) compared seeds lacking Adh-1 activity to wild type seeds, and determined that Adh-l- seeds could not survive the same flooding conditions as Adh-l+ seeds. Normal aerobic growth








6

was not influenced by the Adh-1 genotype. On the other hand, no effect on flooding tolerance was observed by substituting Adh-2- for Adh-2+. Seedlings lacking activity at both ADH loci were extremely flood sensitive and did not grow well even under aerobic conditions. These results lead to the conclusion that Adh-1 is the main locus involved in flood tolerance; Adh-2 also contributes to flood tolerance, but is only beneficial when Adh-l was inactive (Schwartz, 1969; Freeling and Birchler, 1980).

Another reason for the existence of isozymes is

maintenance of optimal enzyme activity over a range of environments. For example, there are two alcohol dehydrogenase genes in tomato (Tanksley, 1979; Tanksley and Jones, 1981). Adh-l is constituitively expressed in pollen and seed tissue (Tanksley, 1979). However, Adh-2 is only expressed in response to anaerobic induction with no tissue specificity (Tanksley and Jones, 1981). Apparently Adh-l is required in the early developmental stages of the plant, whereas Adh-2 is only active when the plant is subjected to an anaerobic environment.

Isozymes have several useful properties in plant

breeding and genetics (Tanksley and Rick, 1980b). 1. They are codominantly expressed, meaning both alleles in a diploid can be observed. 2. Electrophoretic mobility is not affected by environmental factors. 3. Sampling is nondestructive. 4. Isozyme genotypes can be determined on








7

young seedlings in a matter of hours. 5. Also, isozymes have a neutral effect on the overall phenotype of the plant.

For an isozyme locus to be useful to geneticists,

electrophoretic polymorphism must exist among genotypes. Polymorphic isozyme loci can be used to differentiate between plant genotypes (Bassiri and Rouhani, 1977; Santamour and Demuth, 1980; Weeden, 1984; Weeden and Lamb, 1985), to separate accidental self-pollinations from hybrids in a breeding program (Soost et al., 1980), to determine phylogenetic relationships between species (Crawford, 1983), to analyze the genetics of a population (Weber and Stettler, 1981; Brown and Weir, 1983), to identify loci controlling quantitative traits (Tanksley et al., 1982; Weller et al., 1988), or as selectable genetic markers for important horticultural traits (Rick and Fobes, 1974; Tanksley, 1983; Tanksley et al., 1984; Weeden et al., 1984).


Tomato Isozymes

Isozymes of tomato have been studied extensively.

Currently, 39 isozyme loci have been mapped, with at least one locus on each of the 12 tomato chromosomes (Chetelat, 1989). For a list of enzyme abbreviations, see page v. Mapped loci include: Idh-l, Prx-l, Skdh-l, Est-3, and Nir-1 on chromosome 1; Est-7, Prx-2, and Fdh-1 on chromosome 2; Prx-7, Pgm-l, Mdh-l, and Prx-6 on chromosome 3; 6Pqdh-1, Tai-2, Got-l, Pm-2, and Adh-1 on chromosome 4; Dia-l,









8

6Pqdh-3, and Prx-5 on chromosome 5; Aps-1, Sod-3, and Adh-2 on chromosome 6; Got-3, Got-2, Aco-2, and Mdh-3 on chromosome 7; Aps-2 and Got-4 on chromosome 8; Dia-3 and Est-2 on chromosome 9; Est-8 and Prx-4 on chromosome 10; Sod-1 on chromosome 11; and Est-4, 6Pqdh-2, Pgi-l, Mdh-4, and Aco-1 on chromosome 12. Of these loci, Prx-1, Skdh-1, Nir-1, Est-7, Prx-2, Fdh-1, Prx-7, Tpi-2, pgm-2, Adh-1, Dia1, ADs-l, Got-2, Aps-2, Est-2, Est-8, Prx-4, Sod-1, Est-4, 6Pqdh-2, Pqi-1, and Aco-l are polymorphic between L. esculentum and L. pennellii (Chetelat, 1989).

Tomato isozyme loci have served a number of purposes to previous investigators. Tanksley et al. (1981) analyzed isozyme loci in gametophytic and sporophytic tissues, and found that most loci (18 of 30) were expressed in both tissues. They proposed the use of gametophytic selection for some traits to obtain desirable sporophytic phenotypes. Tanksley et al. (1981) also determined that gametophytic enzyme loci were expressed postmeiotically because they did not observe intragenic heterodimeric bands at heterozygous dimeric enzyme loci. For intragenic heterodimers to be produced, transcription and translation must take part in diploid cells of premeiotic stages.

Isozymes have also been used to map several genes of economic importance. Tanksley et al. (1984) found tight linkage between a nuclear male-sterile locus, ms-10, and Prx-2 on chromosome 2. Rick and Fobes (1974) reported tight








9

linkage between Mi, a nematode resistance gene, and Aps-l on chromosome 6. Prx-2 and Aps-l have been used extensively in breeding programs as selectable markers for ms-10 and Mi, respectively (Rick and Fobes, 1974; Medina-Filho and Stevens, 1980; Tanksley et al., 1984). Tanksley and LoaizaFigueroa (1985) mapped the Lycopersicon gametophytic selfincompatibility locus, S, to chromosome 1 linked to Prx-1. This was the first report proving monogenic control of gametophytic self-incompatibility in plants.

Quantitative traits have also been examined using

isozymes. Tanksley et al. (1981) analyzed four quantitative traits (leaf shape, stigma exsertion, fruit weight, and seed weight) and 12 polymorphic isozyme loci in a backcross population. They detected a significant correlation between mean heterozygosity (i.e., the portion of heterozygous isozyme loci in each backcross individual) and measurements for each of the quantitative traits. They concluded that the isozyme genotypes could be used to predict the quantitative trait phenotypes. Tanksley et al. (1982) were able to map 21 loci controlling these traits using the 12 isozyme loci as markers. Vallejos and Tanksley (1983) examined the effect of the chromosome segments marked by 17 isozyme loci on cold tolerance derived from a high altitude L. hirsutum. They found that one marker, Pqi-l on chromosome 12, was linked to a locus that significantly improved growth at low temperatures.








10


FUSARIUM WILT of TOMATO


Taxonomy

Fungi are taxonomically classified according to their sexual cycle, or perfect stage. Since no sexual cycle has been observed for Fusarium oxysporum, it is classified as a Deuteromycete, or Fungi Imperfecti, a group of about 15,000 species with no apparent perfect stage (Moore-Landecker, 1972). E. oxysporum was initially classified by Wollenweber and Reinking (1935) based on the following four factors:

1. whether or not macroconidia were produced on sporodochia,

2. the color of stroma, 3. the presence or absence of sclerotia, and 4. the length, width, and number of septae in macroconidia. Initially, F. oxysporum was included in the Section Elegans which had three subsections, Orthocera, Constrictum, and Oxysporum (Wollenweber and Reinking, 1935). In 1940, Snyder and Hansen reclassified members of the Section Elegans so that all belonged to one species, E. oxysporum. Individuals within the F. oxysporum species were subdivided into formae speciales based on pathogenicity to different hosts; Booth (1971) has classified 76 different formae speciales of F. oxysporum. Formae speciales were further subdivided into races based on varying pathogenicity to genotypes within a host species; race classifications were first proposed by the International Botanical Congress in 1935 (Ainsworth et al., 1943). Three races of F.









11

oxysporum f. sp. lycopersici, the Fusarium wilt pathogen of tomato, have been observed (Massee, 1895; Alexander and Tucker, 1945; Grattidge and O'Brien, 1982; Volin and Jones, 1982).


Anatomy, Growth, and Reproduction of the Fungus

F. oxysporum reproduces asexually by means of spores. Three types of asexual spores, or conidia, are produced by the fungus including macroconidia, microconidia, and chlamydospores. Macroconidia are multiseptated, thin-walled spores with a pointed apical cell and a distinct foot cell; they are usually produced on branched conidiophores (specialized hyphae that give rise to conidia) in sporodochia on the surface of infected plant parts (Nelson, 1981). Microconidia are uni- or di-septated spores found on short microconidiophores in the aerial mycelium; they are the predominant form involved in spread of the fungus through the plant. Chlamydospores are uni- or di-septated thick-walled conidia formed from macroconidia or in hyphae (Toussoun and Nelson, 1976). They are produced in the final stages of wilt disease development, and represent a dormant stage of the fungus; chlamydospores have the ability to survive for years in soil or plant debris in the absence of a host (Nelson, 1981).

During asexual reproduction, macroconidia and

microconidia are produced on conidiophores. Conidiophores are produced singly, either in sporodochia or in flat








12

clusters called pionnotes (Puhalla and Bell, 1981). The conidia are generated on flask-shaped terminal cells of conidiophores called phialides and remain dormant on mycelia until dispersed (Puhalla and Bell, 1981). Germination is inhibited by the action of volatile compounds produced by the fungus (Robinson and Garrett, 1969).

After dispersal, conidia can germinate under a wide

variety of conditions. Water potential as low as -90 bars, a temperature range of 240 to 32 oC, a pH range of 4.0 to 9.0, or an atmosphere devoid of oxygen permit germination (Stotzky and Goos, 1965; Mozumder et al., 1970; Ioannou et al., 1977; Puhalla and Bell, 1981). The presence of essential nutrients increases germination rate and percentage (Puhalla and Bell, 1981). Macroconidia and chlamydospores have been shown to contain large quantities of lipids which appear to be utilized as a carbon source during germination (Marchant, 1966; Van Eck and Schippers, 1976). Macroconidia germinate by forming one or two germ tubes, whereas microconidia and chlamydospores normally give rise to only one germ tube (Griffin, 1970).

Once germinated, fungal growth can either be vegetative as a mycelium or yeast-like; this trait is called dimorphism (Beckman et al., 1953; Malca et al., 1966). In the yeast form, conidia produce one or two phialides that directly give rise to more conidia with no germ tube being produced (Buckley et al., 1969). Yeast-like growth appears to be








13

important in pathogenicity; variants of Ceratocystis ulmi and Verticillium EM. that have lost their ability to grow as a yeast are avirulent (Tolmsoff, 1973). Also, yeast-like growth requires de novo protein synthesis; addition of pfluorophenylalanine, a protein synthesis inhibitor, repressed yeast-like growth of C. ulmi and F. oxvsporum (Biehn, 1973).

F. oxysporum mycelia are made up of septate hyphae that are usually uninucleate (Howard and Maxwell, 1975); these septa have central pores, but nuclei do not migrate through them (Typas and Heale, 1976). Fungal metabolites called diffusible morphogenic factors (DMF) can inhibit hyphal elongation and induce lateral branching; succinic acid and bikaverin, the red pigment in hyphae and chlamydospores, have been shown to have DMF activity (Robinson et al., 1969). The lateral branches can become a part of the growing front, give rise to conidiophores, or anastomose with other hyphae (Hoffmann, 1966; Puhalla and Mayfield, 1974).


Optimal Fungal Growth Conditions

F. oxysporum is a soil borne fungus with greatest potential for infection at soil and air temperatures of approximately 28 OC (Clayton, 1923a). Soil moisture levels that are optimal to the plant promote disease development; moisture levels that are too high or low inhibit the pathogen (Clayton, 1923b). Short day length and low light








14

intensity were also found to enhance disease development (Foster and Walker, 1947; Walker, 1971). The pathogen also prefers acid soil with an optimum pH range of 4.4 to 5.2 (Sherwood, 1923). Jones and Woltz (1969, 1970) found that liming soil to raise pH was effective in retarding disease development; the increase in pH limited the availability of micronutrients such as iron, manganese, and zinc to the pathogen. Bacteria and actinomycetes that compete for nutrients and produce antibiotics that can retard fungal development are also more prevalent at a higher pH (Marshall and Alexander, 1960). Other nutrient studies have indicated that deficiencies in potassium and calcium, and increased levels of nitrogen and sulfur promote infection (Fisher, 1935; Walker and Foster, 1946; Edgington and Walker, 1958). Life Cycle

Dormant F. oxysporum chlamydospores germinate when

brought into contact with host roots or fresh plant debris (Stover, 1962; 1970). After host penetration, the fungus migrates to the vascular tissue, specifically the xylem (Nelson, 1981). Once in the xylem, the pathogen spreads by means of mycelia or microconidia (Nelson, 1981). Spread of hyphae laterally from primary to secondary xylem is through pit pairs, whereas upward movement is through the simple perforation plates of xylem vessel elements (Nelson, 1981). During latter stages of disease infection, the fungus can move to phloem, cambium, pith, and cortex tissue (Nelson,








15

1981). After death of the host, sporodochia form on the surface of the plant and produce large numbers of macroconidia, as well as some microconidia (Baker, 1953; Phipps and Stipes, 1976). Macroconidia frequently convert to chlamydospores and return to the soil on the host debris, completing the life cycle. F. oxvsporum f. sp. lyvcopersici can also colonize several weed genera including Oryzopsis, Digitaria, Amaranthus, and Malva with no disease symptoms being evident; eggplants (Solanum melongena) can also be colonized with some stunting (Katan, 1971). This type of nonhost colonization allows the fungus to survive longer periods of time in the absence of the host. In addition, Armstrong and Armstrong (1948) showed that F. oxysporum f. sp. batatas, the wilt pathogen of sweet potato (Ipomea batatas), can colonize on tomato without expression of symptoms.


Host Defense Mechanism

Successful colonization of Fusaria within the host is dependent on the ability of the pathogen to circumvent the defense mechanisms of the plant. Initially the soil-borne pathogen encounters a number of physical barriers associated with the plant root. Young roots are protected by a continuous epidermal layer. Within the epidermis are cortex, endodermis, and pericycle layers surrounding the vascular tissue. Older roots are surrounded by a heavily suberized corky tissue which arises from the pericycle









16

(Talboys, 1958a). When hyphae penetrate epidermis or cortex tissue, host cell walls are degraded. However, surrounding living cells lay down callose deposits called papillae that block further hyphal penetration (Bishop and Cooper, 1983). Papillae eventually become lignified through infusion of secondary metabolites making them resistant to enzymatic and chemical degradation by the fungus. This response is similar in both resistant and susceptible genotypes (Bishop and Cooper, 1983). Hyphae not stopped by the papillae are physically inhibited by the endodermis with suberized Casparian strips (Beckman, 1987).

Once hyphae reach the vascular tissue, the pathogen is localized in individual xylem vessels (Beckman, 1987). Within the vessel, hyphae move laterally through pits into adjacent contact cells where they derive their nutrients. Pits are plugged with callose deposits to inhibit this movement; these deposits lignify and become papillae (Beckman et al., 1982). Any invading organism, as well as wounding, can initiate this response in both resistant and susceptible genotypes (Beckman, 1987).

Longitudinal movement of hyphae is inhibited by vascular gels and tyloses (Wardlaw, 1930; Beckman and Halmos, 1962). Gels are composed of pectins, calcium pectates, hemicellulose, and traces of protein (Beckman and Zaroogian, 1967). In a resistant response, gels persist several days and are resistant to both physical and chemical









17

degradation (Beckman et al., 1974), whereas gels of susceptible genotypes become weakened and break under transpirational tension (Beckman et al., 1962). Tyloses are outgrowths of vascular parenchyma contact cells through vessel pits (Wardlaw, 1930). The cells grow into the xylem vessel, effectively barricading the fungus (Talboys, 1958b). If pits are too small to allow passage of enlarging contact cells, vascular elements are crushed; this also walls off the invading pathogen (Chattway, 1949). Since plants have the ability to transport more water than necessary, loss of some xylem vessels is of little consequence (Beckman, 1987).

Another aspect of the defense mechanism is the release of secondary metabolites including phenolics and phytoalexins. Phenolics are stored in a reduced state in specialized cells and are released in response to wounding or infection (Mace, 1963). Once released, they are enzymatically oxidized and polymerize to form lignin which is incorporated into cell walls and tissues (Beckman, 1987); these tissues are impermeable to most invading organisms (Friend, 1976).

After localization of the pathogen, phytoalexins are mobilized to the site of infection; also, tyloses produce and release phytoalexins directly into the lumina of infected vessels (Hutson and Smith, 1980). These phytoalexins inhibit growth of the pathogen. In tomato, chlorogenic acid, scopolin, tomatine, rishitin, and a number








18

of flavonoids have been shown to have phytoalexin activity (Stoessl et al., 1976). Additional host defense responses include production of a protein that inhibits fungal polygalacturonase (Albersheim and Anderson, 1971) and production of P-1,3-glucanase and chitinase fungal cell wall degrading enzymes (Pegg, 1976; Pegg and Young, 1981). Both responses are common in resistant and susceptible host reactions.


Genetics of Fusarium Wilt Resistance

Fusarium wilt of tomato was first reported in England by Massee in 1895. Some early cultivars such as 'Marglobe' and 'Norton' carried moderate levels of genetic tolerance; however, frequently these cultivars would succumb to the disease under optimal environmental conditions (Richards, 1925; Harrison, 1941). Bohn and Tucker (1939, 1940) were the first to introduce monogenic resistance to Fusarium wilt derived from a L. pimpinellifolium accession called 'Red Currant' (U.S.D.A. Accession 160). This gene conferred complete immunity to the disease, and was designated I for "immunity". However, it has been observed that race 1 can occasionally infect plants carrying the I gene; this incomplete penetrance is influenced by genetic background, environmental conditions, and whether the plant is heterozygous or homozygous at the I locus (Stall, 1961; Alon et al., 1974). In 1945, Alexander and Tucker observed a new race in Ohio, designated race 2, that consistently









19

infected lines with 'Red Currant' resistance. Race 2 did not spread as rapidly as race 1. Monogenic resistance to this race, 1-2, was first observed in 1955 (Alexander and Hoover, 1955) in another accession of L. pimpinellifolium (P.I. 126915). 'Walter' was the first cultivar introduced that incorporated resistances to both races 1 and 2 (Strobel et al., 1969). I and I-2 have been mapped to chromosome 11, where I-2 is tightly linked to the mps (miniature phosphorus syndrome) gene (Paddock, 1950; Latterot, 1976).

Race 3, a new race with the ability to infect cultivars resistant to races 1 and 2, has been observed in Australia (Grattidge and O'Brien, 1982), Florida (Volin and Jones, 1982), and more recently in California (Davis et al., 1988). Accessions of the wild species L. pennellii, P.I. 414773 (McGrath et al., 1987) and LA 716 (Scott and Jones, 1989), have been shown to carry dominant monogenic resistance to race 3. Both reports have designated the resistance gene as I-3 (McGrath et al., 1987; Scott and Jones, 1989). Preliminary data indicate these two resistances are allelic; further work is needed to confirm this (JW Scott, personal communication, 1989). L. pennellii accession LA 716 also carries resistance to races 1 and 2 (Scott and Jones, 1989).















CHAPTER 3

MAPPING OF NEW ENZYME LOCI


INTRODUCTION

Genetic markers have been used by plant breeders to expedite transfer of important genes from one genetic background to another (Rick and Fobes, 1974; Weeden et al., 1984). There are three types of markers commonly used: morphological, restriction fragment length polymorphisms (RFLP's), and isozymes. Morphological markers have a number of potential disadvantages: 1. they can be deleterious to the overall plant phenotype; 2. environmental conditions can influence marker phenotypes; 3. frequently they are not identifiable until late in the plant life cycle; and 4. they are usually only expressed in the homozygous recessive condition. RFLP use is limited due to time requirements, expense, and the necessity of probes. Isozymes have the following advantages as markers: 1. they have a neutral effect on the overall plant phenotype; 2. isozyme genotypes are not influenced by the environment; 3. sampling for the isozyme analysis is nondestructive; 4. isozymes can be identified very early in the plant life cycle; 5. they are codominantly expressed; and 6. they are relatively inexpensive and safe to analyze (Tanksley and Rick, 1980b).

20








21

However, use of isozyme markers is limited to the availability of polymorphism and the relatively small number of loci that have been mapped. For a recent report on the map locations of enzyme loci in the genus Lycopersicon, see Chetelat (1989).

In this report, the map locations and tissue

specificities of eight additional enzyme loci dimorphic betweeen L. esculentum and L. pennellii (LA 716) were identified by conducting linkage analyses to 15 previously mapped isozyme loci. The new loci included Diaphorase-2 (Dia-2), B-N-acetyl-Qlucosaminidase-1 (Bnaq-l), Mannose-6phosphate isomerase-1 (Mpi-l), Sucrose synthase-1 (Ss-l), Esterase-9 (Est-9), Esterase-10 (Est-10), Hexokinase-l (Hk1), and a-Mannosidase-1 (Aman-l). These loci were used in chapters 4 and 5 as potential markers of gene(s) for resistance to races 1, 2, and 3 of Fusarium oxysporum f. sp. lycopersici derived from L. pennellii (LA 716).


MATERIALS and METHODS

To examine segregation and linkage analyses of the new enzyme loci, the following backcross was generated: L. esculentum cv. Bonny Best x [Bonny Best x L. pennellii (LA 716)]. Isozyme procedures have been described previously (Rick et al., 1977; Tanksley and Rick, 1980b). The mannose6-phosphate isomerase stain included the following components: tris*HCl (Sigma Trizma base) 0.1M, pH 7.5; MgC12*6H20 20mM; mannose-6-phosphate 0.2 mg/ml; NADP+ 0.15








22

mg/ml; MTT 0.2 mg/ml; phenazine methosulphate (PMS) 0.04 mg/ml; phosphoglucose isomerase 0.5 units/ml; and glucose-6phosphate dehydrogenase 0.3 units/ml. The sucrose synthase stain involved a triple enzyme coupling mechanism with the following reagents: trisHC1 0.1M, pH 7.5; MgC12*6H20 20mM; sucrose 100mM; ADP 0.2 mg/ml; ATP 1.3 mg/ml; hexokinase 1 units/ml; phosphoglucose isomerase 0.5 units/ml; glucose-6phosphate dehydrogenase 0.3 units/ml, NADP+ 0.15 mg/ml; MTT

0.2 mg/ml; and PMS 0.04 mg/ml. Staining for both systems was done at room temperature in the dark. a-Mannosidase was stained using a filter paper overlay with the following protocol: 2mg of 4-methylumbelliferyl-a-mannopyrannoside was dissolved in iml of dimethyl sulfoxide; this was added to 5ml of 0.1M acetic acid pH 4.5 (adjust pH with NaOH). Bands were observed using an ultraviolet light box. Other staining systems have been described previously (Vallejos, 1983; Weeden, 1984, 1986). The following gel systems were used: histidine (gel: histidineHC1, 5mM, pH 7.0, adjusted with NaOH; electrode: tris, 135mM/citric acid 43mM, pH 7.0, adjusted with tris or citrate) for Ss-1, BnaQ-l, and Aman-l; tris-EDTA-borate (Heath-Pagliuso et al., 1984)) for Dia-2 and Bnaq-l; and tris citrate (Tanksley and Rick, 1980b) for MDi1l, Hk-1, Est-9, and Est-10.

All samples were collected simultaneously from the same 'Bonny Best' and LA 716 plants for the tissue specificity analysis; tissues included the fol lowing: leaf, petiole,









23

stem, pedicel, peduncle, flower, immature fruit, and root. The segregation for the new and 15 previously mapped enzyme loci was recorded for all backcross individuals; the latter 15 loci included: Prx-1 and Skdh-l on chromosome 1; Est-7 and Prx-2 on chromosome 2; Prx-7 on chromosome 3; Pgm-2 on chromosome 4; Aps-l on chromosome 6; Got-2 on chromosome 7; Aps-2 on chromosome 8; Prx-4 on chromosome 10; Sod-1 on chromosome 11; and Est-4, 6Pqdh-2, Pai-l, and Aco-1 on chromosome 12. The morphological marker, Pn on chromosome 8, was also scored. Tests for independent assortment were conducted using two-way contingency tables on a main frame computer using the Proc Freq program of the Statistical Analysis System (SAS Institute Inc., SAS Circle, Box 8000, Cary, NC 27511.


RESULTS and DISCUSSION

Electrophoretic migration distances of the eight new enzyme loci are shown in Table 3-1. One zone of activity was detected for AMAN, BNAG, and MPI indicating the presence of a single locus for these enzymes. There were at least two zones of activity associated with HK, the most anodal of which was assigned as the Hk-l locus. Four zones of DIA activity appeared in the tris-EDTA-borate gel system. The large zone of activity that migrated between Dia-1 and Dia3, two loci previously mapped to chromosomes 5 and 9 respectively (Chetelat, 1989), has been designated Dia-2 (Table 3-1); the slowest migrating zone of DIA activity was








24

not segregating in this cross and has not been assigned a chromosomal location.

At least five zones of activity were observed in the

EST leaf zymogram. The most anodal two loci, Est-2 and Est3 have been previously mapped to chromosomes 9 and 1, respectively (Tanksley and Rick, 1980a). The slowest migrating zone of activity, Est-8, was mapped to chromosome 10 (Tanksley et al., 1988). Two additional zones of activity appeared between Est-3 and Est-8 (Table 3-1). These two zones assorted independently and have been assigned as Est-9 and Est-10 with Est-9 being anodal to Est10.

Six zones of activity appear after staining leaf tissue for sucrose synthase. However, not all of these bands represent sucrose synthase activity. Sequential removal of coupling enzymes allowed for the determination of the origin of most bands. The slowest migrating zone of activity was the only zone that did not appear when coupling enzymes were removed. This zone of activity has been designated Ss-1. Even Ss-1 is only tentatively assigned as sucrose synthase. This locus was specific to green tissues (Table 3-1); however, a western blot analysis of tomato root and leaf extracts using maize sucrose synthase antibody only detected two root specific loci. No protein associated with sucrose synthase was detected in the leaf sample. Since the sucrose synthase stain involves a three enzyme coupling








25

reaction, it is possible that an intermediate produced in one of the reactions may have provided the necessary substrate for a different enzyme to produce the bands observed for Ss-l.

Table 3-1 also shows the tissue specificities of each of the new enzyme loci. For all loci, some level of activity was present in all green tissues analyzed; in addition, activity in the roots was detected for all loci except Est-9, Est-10, and Ss-l. Relatively equal levels of activity were observed for Aman-l, Bnag-l, Dia-2, and Est-10 in all tissues where these loci were expressed (Table 3-1). Loci which showed an increased level of activity in a specific tissue included: Est-9 and Mpi-l in pedicel and flower tissues, Hk-1 in stem and peduncle tissues, and Ss-l in flower tissue (Table 3-1).

Tests of independent assortment between each new locus and all segregating loci (the previously mapped loci as well as the other new loci) were conducted to determine the chromosome location of the eight enzyme loci (Table 3-2). Tables 3-3, 3-4, and 3-5 show contingency tables of loci pair combinations where significant deviations from expected ratios were observed. Linkage was detected between Bnaq-l, Dia-2, and Est-9 to the Skdh-1 locus known to be on chromosome 1 (Tanksley and Rick, 1980b, Table 3-3). Thus, these loci were assigned to chromosome 1, and the following








26

gene order was deduced by three point mapping using Prx-1 and Skdh-1 as reference points (Tables 3-6 and 3-7): Prx-l--20cM--Skdh-l,Bnaq-l--29cM--Dia-2--6cM--Est-9.


Due to tight linkage, further testing is necessary to determine the side of Skdh-l on which Bnaq-1 lies.

Mi-l was assigned to chromosome 2 due to its linkage to Est-7 and Prx-2, two loci previously mapped to this chromosome (Rick and Fobes, 1977; Tanksley and Rick, 1980b, Table 3-4). Three point mapping between these three loci could not definitely discern a gene order with the available data (data not shown). Linkage was also detected between Ss-l and Prx-7 (27cM) known to be on chromosome 3 (Tanksley and Rick, 1980b), Aman-1 and Aps- (31cM) previously assigned to chromosome 6 (Rick and Fobes, 1974), and Est-10 and Sod-1 (3cM) previously reported on chromosome 11 (Tanksley et al., 1988, Table 3-5). No linkage was detected between Hk-1 and any of the segregating marker loci.

Analysis of monogenic segregation ratios of the marker loci in the BC1 population showed that most conformed to the expected 1:1 ratio (Table 3-8). However, a significant excess of heterozygotes was detected on chromosome 1 at Prx1, Dia-2, and Est-9, on chromosome 8 at Pn, on chromosome 11 at Est-10, and on chromosome 12 at 6Pqdh-2 and Pqi-1; Got-2 on chromosome 7 had significantly more homozygotes than expected (Table 3-8). It is likely that loci linked to








27
these markers contributed to the ability of the pollen grain to fertilize the female gametophyte. Skewed ratios at loci known to be monogenic have been reported previously in interspecific Lycopersicon backcrosses (Rick, 1969; Tanksley et al., 1982).

Most of the new enzyme loci were not tightly linked to a previously mapped marker locus. In future gene mapping studies, including the tagging of genes for resistance to Fusarium oxysporum f. sp. lycopersici, use of these new loci will effectively increase the portion of genome covered by enzyme markers.









28

Table 3-1. Migration distances and tissue specificities of eight new enzyme loci. Migration distances are a measure of the distance that the zone of activity associated with the locus migrated from the origin toward the anode (+) end of the gel. Tissue sample analysis shows a comparison of the relative levels of activity of the tissue samples for each locus. Samples were taken from the same plants at the same time; the fruit sample was at the immature green stage. ee L. esculentum homozygotes; ep L. pennellii heterozygotes.a


Migration Tissue samples
distance (cm) Petiole Peduncle Flower Root Locus ee ep Leaf Stem Pedicel Fruit


Aman-1 1.0 1.6 + ++ ++ ++ ++ ++ ++ ++ Bnaq-1 1.8 2.4 ++ ++ ++ ++ ++ ++ ++ ++ Dia-2 5.5c 6.3 + ++ ++ ++ ++ ++ ++ + Est-9 7.5d 7.8 + ++ ++ ++ +++ +++ ++ Est-10 6.5d 6.9 ++ ++ ++ ++ ++ + + Hk-1 9.5 9.3 + ++ +++ +++ ++ + ++ + Mpi-l 9.0 8.7 ++ ++ + ++ +++ +++ ++ + Ss-1 1.7 1.2 + + + + + +++ ++


aTable only represents comparisons of activity within each locus. Comparisons between loci can not be made (e.g. Aman1 and Dia-2 leaf activities are not necessarily equal). 5(-) no activity, (+) low activity, (++) intermediate activity, (+++) greater activity. cDia-1l, Dia-3, and Dia-4 had migration distances of 7.0, 4.0, and 2.0, respectively for both L. esculentum and L. pennellii.
aLeaf esterase loci, Est-2, Est-3, and Est-8, had migration distances of 9.1, 8.3, and 5.0, respectively for L. esculentum.









29

Table 3-2. Chromosomal location of new enzyme loci. Independent assortment analyses comparing each new locus with previously mapped loci as well as the other new loci were conducted using two-way contingency tables. Numbers represent X2 values in each analysis.


Chromo- New loci
Locus some Aman-1 Bnaq-1 Dia-2 Est-9 Est-10 Hk-1 Mpi-1 Ss-1


Prx-1 1 NS 25.3** 6.7* NS NS NS NS NS Skdh-1 1 NS 73.1** 24.7 19.4 NS NS NS NS Est-7 2 NS NS NS NS NS NS 8.79* NS Prx-2 2 NS NS NS NS NS NS 7.31* NS Prx-7 3 NS NS NS NS NS NS NS 24.2 Pgm-2 4 NS NS NS NS NS NS NS NS Aps-i 6 8.4* NS NS NS NS NS NS NS Got-2 7 NS NS NS NS NS NS NS NS Aps-2 8 NS NS NS NS NS NS NS NS Pn 8 NS NS NS NS NS NS NS NS Prx-4 10 NS NS NS NS NS NS NS NS Sod-1 11 NS NS NS NS 86.1 NS NS NS Est-4 12 NS NS NS NS NS NS NS NS 6Pqdh-2 12 NS NS NS NS NS NS NS NS Pgi-l 12 NS NS NS NS NS NS NS NS Aco-1 12 NS NS NS NS NS NS NS NS Aman-1 - -- NS NS NS NS NS NS NS Bnaq-1 - NS -- 14.1 NS NS NS NS NS Dia-2 - NS 14.1 -- 52.9 NS NS NS NS Est-9 - NS NS 52.9 -- NS NS NS NS Est-10 - NS NS NS NS -- NS NS NS Hk-1 - NS NS NS NS NS -- NS NS Mpi-l - NS NS NS NS NS NS -- NS Ss-1 - NS NS NS NS NS NS NS -*, **Significant at the 0.01 and 0.001 levels, respectively.









30

Table 3-3. Independent assortment analysis of chromosome 1 isozyme loci using two-way contingency tables. ee L. esculentum homozygotes; ep L. pennellii heterozygotes; p percent recombination.a


Locus
Locus Prx-1 Skdh-1 Bnaq-1 Dia-2 Est-9
ee ep ee ep ee ep ee ep ee ep

ee 46 6 25 4 25 24 12 20 Prx-l
ep 19 57 13 35 20 52 11 30

X2=49.8** X2=25.3** X2=6.7* X2=NS p=19.5% p=9.1% p=36.4%

ee 38 0 36 26 21 18 Skdh-1
ep 1 38 8 49 2 32

X2=73.1** X2=24.7** X2=19.4** p=1.3% p=28.6% p=27.4%

ee 18 18 9 10 Bna-1 --ep 3 31 2 16

X2=14.1** X2=NS
p=30.0%

ee 20 3 Dia-2
ep 1 46

X2=52.9
p=5.7%


, **Significant at the 0.01 and 0.001 levels, respectively.

aDeduced gene order:

Prx-l--20cM--Skdh-l,Bnag-l--29cM--Dia-2--6cM--Est-9









31

Table 3-4. Independent assortment analysis of chromosome 2 isozyme loci. ee L. esculentum homozygotes; ep L. pennellii heterozygotes; p percent recombination.


Est-7 Prx-2 Mpi-i
ee ep ee ep ee ep

ee 50 5 21 11 Est-7
ep 4 61 13 29

X2=86.5** X2=8.8* p=7.5% p=32.4%

ee 19 9 Prx-2
ep 13 25

X2=7.3*
p=33.3%


, **Significant at the 0.01 and 0.001 levels, respectively.









32

Table 3-5. Independent assortment analyses using two-way contingency tables for the following linked loci: Ss-1 and Prx-7 on chromosome 3, Aman-1 and Aps- on chromosome 6, and Est-10 and Sod-i on chromsome 11. ee L. esculentum homozygotes; ep L. pennellii heterozygotes; p percent recombination.


Ss-1 Aman-1
ee ep ee ep

ee 40 17 ee 17 11 Prx-7 Apasep 12 40 ep 5 19

X2=24.2** X2=8.42* p=26.6% p=30.8%


Est-10
ee ep

ee 39 2
Sod-i
ep 1 56

X2=86.1**
p=3.1%


, **Significant at the 0.01 and 0.001 levels, respectively.









33

Table 3-6. Deduced chromosomal location of Dia-2 with respect to Prx-1 and Skdh-1 on chromosome 1 using threepoint mapping techniques. Prx-1 and Skdh-1 were mapped previously (Tanksley and Rick, 1980). Table shows segregation ratios and the number of double recombinants for each putative gene order. ee L. esculentum homozygotes; ep L. pennellii heterozygotes.


Prx-1/Skdh-1 Putative Double
ee/ee ee/ep ep/ee ep/ep gene order Recombs


ee 24 1 12 7 Prx-1/Skdh-i/Dia-2a 6 Dia-2 Prx-l/Dia-2/Skdh-1 27
ep 20 3 5 46 Dia-2/Prx-/Skdh-1 15

aCorrect gene order because it showed the least number of double recombinants.









34

Table 3-7. Deduced chromosomal location of Est-9 with respect to Skdh-1 and Dia-2 on chromosome 1 using threepoint mapping techniques. Est-9 and Prx-1 were not linked (Table 3-2). Skdh-1 was mapped previously, and the location of Dia-2 was determined in Table 3-5. This table shows segregation ratios and the number of double recombinants for each putative gene order. ee L. esculentum homozygotes; ep L. pennellii heterozygotes.


Skdh-1/Dia-2 Putative Double
ee/ee ee/ep ep/ee ep/ep gene order Recombs


ee 19 0 1 1 Skdh-1/Dia-2/Est-9a 0 Est-9 Skdh-i/Est-9/Dia-2 4
ep 3 15 0 30 Est-9/Skdh-/Dia-2 16


aCorrect gene order as judged by the lack of double recombinants.









35

Table 3-8. Analysis of monogenic segregations of marker loci. ee L. esculentum homozygotes; ep L. pennellii heterozygotes.


Locus Chromosome ee ep Monogenic X2


Prx-l 1 53 80 5.48* Skdh-1 1 66 63 NS Bnaq-1 1 39 39 NS Dia-2 1 45 77 8.39 Est-9 1 23 51 10.59 Est-7 2 62 72 NS Prx-2 2 54 66 NS Mpi-l 2 34 40 NS Prx-7 3 64 57 NS Ss-1 3 61 59 NS Pm-2 4 62 70 NS Aps-1 6 69 55 NS Aman-1 6 23 32 NS Got-2 7 81 50 7.34 Avs-2 8 74 55 NS Pn 8 26 64 16.04 Prx-4 10 66 63 NS Sod-1 11 43 57 NS Est-10 11 49 79 7.03 Est-4 12 43 59 NS 6Pqdh-2 12 42 88 16.28 Pi-1 12 44 89 15.22 Aco-1 12 57 73 NS Hk-1 ? 44 53 NS


*, **Significant at the 0.05 and 0.01 levels, respectively.
















CHAPTER 4

ANALYSIS AND MAPPING OF GENES CONTROLLING RESISTANCE
TO RACE 3 OF FUSARIUM WILT


INTRODUCTION

Races 1 and 2 of Fusarium oxvsporum f. sp. lycopersici (Sacc.) Snyder and Hansen which causes Fusarium wilt of tomato are widespread in most production regions. Single dominant genes I and 1-2, which control resistances to races 1 and 2, respectively, have been derived from accessions of Lycopersicon pimpinellifolium (Bohn and Tucker, 1939; 1940; Alexander and Hoover, 1955; Stall and Walter, 1965). These genes have been mapped to chromosome 11 (Paddock, 1950; Latterot, 1976), and have been incorporated into most cultivars.

Race 3, a new race with the ability to infect cultivars resistant to races 1 and 2, has been reported in Australia (Grattidge and O'Brien, 1982), Florida (Volin and Jones, 1982), and more recently in California (Davis et al., 1988). The wild species L. pennellii represents a potential source of resistance for new cultivars. This species has been reported to carry monogenic dominant resistance to race 3 in accessions PI414773 (McGrath et al., 1987) and LA 716 (Scott and Jones, 1989).

36









37

Tanksley and Rick (1980b) proposed the use of isozymes as scorable markers to facilitate the efficient transfer of genes from a wild species into domesticated cultivars. Rick and Fobes (1974) were the first to report tight linkage between the enzyme locus Aps-l, and Mi, a gene responsible for nematode resistance in tomato. This linkage has had a major impact on tomato breeding programs; the Aps-i locus is now widely used as a scorable marker for nematode resistance (Medina-Filho and Stevens, 1980). In pea, the enzyme locus Pqm-D can be used as a marker for Mo, a gene which controls for resistance to bean yellow mosaic virus (Weeden et al., 1984).

The objective of this portion of the project was to find a molecular marker for resistance to race 3 of F. oxysporum f. sp. lycopersici detected in L. pennellii (LA 716). Two independent approaches were taken. First, tests for independent assortment were conducted between race 3 resistance and each of 25 previously mapped enzyme loci in an interspecific backcross. Second, isozyme analysis was conducted on five breeding lines selected for race 3 resistance derived from LA 716 to test whether any LA 716 alleles had been co-selected.


MATERIALS and METHODS

Plant Materials

The following interspecific backcross was generated for segregation and linkage analyses: [L. esculentum x (L.








38

esculentum x L. pennellii, LA 716)]. 'Bonny Best' was used as the L. esculentum parent due to its extreme susceptibility to race 3. In addition, five breeding lines that had been independently produced using conventional breeding techniques were analyzed [E427 (BC1S3), I3R-1 (BC2S2), 520 (BC2S2), 572 (BC3S1), and 604 (BC1S2)]. All lines included the initial cross 'Hayslip' x LA 716. With the exception of LA 716, all lines and cultivars included in the pedigrees of the breeding lines were race 3 susceptible. The cultivars 'Manapal' (resistant to race 1) and 'Hayslip' (resistant to races 1 and 2) were also used as controls in race 3 inoculations. In addition, five L. esculentum lines carrying L. pennellii (LA 716) chromosome substitutions for chromosome 2, 4, 6, 8, and 11 (Rick, 1969) were inoculated. Isozyme Analysis

The starch gel electrophoresis and enzyme staining

techniques employed have been described elsewhere (Rick et al., 1977; Tanksley and Rick, 1980b; Vallejos, 1983). The genotypes of 79 BC1 individuals and the race 3 resistant breeding lines were determined at each of the following segregating enzyme loci: Prx-l, Skdh-l, Bnag-l, Dia-2, and Est-9 on chromosome 1; Mpi-l, Est-7, Prx-2 on chromosome 2; Prx-7 and Ss-1 on chromosome 3; PQm-2 on chromosome 4; Aps-1 and Aman-1 on chromosome 6; Got-2 on chromosome 7; Aps-2 on chromosome 8; Prx-4 on chromosome 10; Sod-1 and Est-10 on chromosome 11; Est-4, 6Pqdh-2, PQi-1, and Aco-1 on








39

chromosome 12; and Hk-l which has not been assigned a

chromosomal location. The morphological marker, Pn, on

chromosome 8 was also scored. Following isozyme analysis,

the BC1 individuals were raised to maturity and selfpollinated. The BC1S1 progenies, along with the breeding

lines, were then inoculated with race 3.


Inoculations

Inoculum was prepared by incubating spores of race 3 of

F. oxysporum f. sp. lycopersici (SC761 isolated) in

sterilized 100% potato dextrose broth (Difco) at 280 C with

continuous shaking at 125 rpm for 3-4 days. After

incubation, the culture was filtered through four layers of

cheesecloth to remove mycelia and diluted to a concentration

of 1.10 - 1.25 X 108 spores/ml; a hemacytometer was used to

determine spore concentration. The phenotype of each

backcross individual with respect to resistance to race 3

was determined through progeny tests; at least 20 BC1S1

progeny were tested per BC1 analysis. Progeny tests were

used to increase reliability of BC1 scores and to allow for

the eventual determination of BC1 responses to races 1 and 2. 'Bonny Best', LA 716, and F1 seedlings were inoculated

as controls with every set of BC1S1 families. Also, another

set of controls was incubated in distilled water only.



1SC761 is a race 3 strain isolated in 1982 by Dr. J.P. Jones at the University of Florida, Gulf Coast Research and Education Center, at Bradenton, FL.








40

Seedlings of 'Manapal' and 'Hayslip' were periodically inoculated to monitor the specificity of race 3 inoculum.

Seedlings at the 2-4 true leaf stage were inoculated by pruning roots and incubating in inoculum for 3-5 hours at 280 C. They were then transplanted into a sterilized peat:sand:soil (1:1:1) medium (pH 6.0) four cm apart and transfered to a growth chamber. Growth conditions were as follows: 280 C constant temperature, 80% �20% humidity, and a 10h photoperiod. The photon flux density (umol quanta-m2.sec-1) in the chamber changed according to the following schedule: 90 for 2h, 130 for 2h, 200 for 2h, 130 for 2h, 90 for 2h, and darkness for 14h. Fertilizer (Peters 20-20-20) was applied approximately one week after inoculation. Disease severity was estimated at least twice. The first estimation was carried out shortly after susceptible 'Bonny Best' plants began showing wilt symptoms, usually 10-14 days after inoculation. The purpose of this evaluation was to determine the cause of death of dying plants. Stems of dying plants were cut and examined for vascular browning; living plants were judged as wilting, stunted, or healthy. A second estimation of disease severity was carried out approximately one month after inoculation; for this evaluation, the degree of vascular browning was determined by making a series of thin longitudinal cuts from the first true leaf down to the root tip with a razor blade. Scores were assigned according to the following scale: 1 = no sign








41

of browning, or browning only at root tip, 2 = at least some browning found above the root tip but not continuous from root to cotyledon, 3 = continuous browning from root to cotyledon in only one vascular tract, and 4 = continuous browning from root to cotyledon in more than one vascular tract. Dying plants showing external wilt symptoms were given a rating of 5. Plants rated from 1 - 4 had no external wilt symptoms. A weighted mean disease rating was calculated for each backcross individual using the progeny scores. All statistical analyses were performed on a main frame computer using the Statistical Analysis System (SAS Institute Inc., SAS Circle, Box 8000, Cary, NC 27511). The following programs were used: Proc Univariate, Proc Freq, Proc Cluster, Proc Nparlway, Proc Glm, and Proc Sort.


RESULTS and DISCUSSION

In all inoculations, seedlings of L. pennellii (LA 716) and the interspecific F1 consistently showed resistance to race 3; whereas, seedlings of the L. esculentum cultivars, 'Bonny Best', 'Manapal', and 'Hayslip', showed susceptibility. Responses to race 3 for BC1 individuals were deduced from mean disease ratings of their progenies. Analysis of the frequency distribution of the mean disease ratings (Fig. 4-1) using the Kolomogorov D statistic (Kolomogorov, 1933) indicated a significant deviation from normality (D=0.128, a<0.01). In this procedure, the frequency distribution is compared to a normal distribution








42

generated using the mean and variance of the BC1 disease ratings; the test statistic, D, represents the total deviation between the two distributions (Kolomogorov, 1933). Furthermore, the distribution proved to be bimodal according to the coefficient of bimodality (b = (m32+1)/(m4+3); m3 = skewness, m4 = kurtosis; b = 0.586, a < 0.05). Skewness and kurtosis describe types of nonnormal distributions. A distribution shows skewness if a significant excess of samples appear on one side of the mean (Snedecor and Cochran, 1980). Distributions with long tails have a positive kurtosis, whereas a flat-topped distribution has a negative kurtosis (Snedecor and Cochran, 1980). This bimodal distribution indicated that race 3 resistance was a trait controlled by one major locus which was designated I-3 by Scott and Jones (1989).

The possible association between race 3 resistance and a segregating marker locus was analyzed using a one-tailed nonparametric analysis of variance -- the Wilcoxon rank sum test (Wilcoxon, 1945). This test is similar to the parametric two-sample t-test. The nonparametric analysis was implemented because the disease ratings had a limited range (1 to 5) and were based on a qualitative evaluation of disease symptoms. In this procedure, BC1 individuals were first ranked according to their mean disease rating; then, the mean of the rankings of all homozygous plants was compared to that of the heterozygotes at each marker locus.









43

Highly significant differences were detected for the chromosome segment marked by Got-2. BC1 individuals carrying L. pennellii alleles for Got-2 were significantly more resistant than those homozygous for L. esculentum (Table 4-1). These results clearly demonstrated linkage between Got-2 and 1-3. An additional significant effect was also detected for the chromosome segment marked by Aps-2, but the effect of this locus was not as great as the one observed for I-3 (Table 4-1). The presence of this minor factor was supported by the results obtained from inoculations of five L. pennellii (LA 716) chromosome substitution lines developed by Rick (1969). The chromosome 8 line (carrying ADs-2) had a mean disease rating of only 3.00 compared to 4.62, 4.38, 4.85, and 4.84 for lines of chromosome 2, 4, 6, and 11, respectively. Other substitution lines were unavailable. Since the chromosome 8 substitution line showed tolerance to race 3, the locus linked to Aps-2 is being designated Tfw for "tolerance to Fusarium wilt." It was also noted that heterozygotes at the Pqm-2 locus were significantly more susceptible than homozygotes (Table 4-1). Such transgressive behavior has previously been reported in tomato (Tanksley et al., 1982; Vallejos and Tanksley, 1983). Nevertheless, upon further investigation, this observation proved to be an artifact (see below).








44

The percent recombination between Got-2 and I-3 was

estimated to evaluate the usefulness of Got-2 as a marker for resistance. Although the bimodal distribution of mean disease ratings and the nonparametric analyses of variance

indicated that race 3 resistance was controlled by one major locus, it was difficult to classify some BC1 individuals for

their I-3 genotype. To overcome this difficulty, the

average linkage clustering method (Sokal and Michener, 1958) was used to classify all BC1 individuals according to their

mean disease ratings into two clusters: resistant and

susceptible. This method sequentially clusters individuals, or groups of individuals, with the least distance2 until the

desired number of clusters is attained -- two in this case.

The genotype of each BC1 individual at the I-3 locus was

inferred according to its cluster classification:

susceptible (i-3/i-3; R=4.01 �0.63) or resistant (I-3/i-3; x=2.11 �0.32). A test for independent assortment between

Got-2 and I-3 using a two-way contingency table showed

significant deviations from expected ratios; the

recombination between the two loci was estimated at 2.5%

(Table 4-2). Fig. 4-2 shows a GOT starch gel including the

Got-2 phenotypes of 'Bonny Best', LA 716, and their F1.

The evidence presented above indicated that race 3

resistance was controlled by one major gene, 1-3. However,

2DKL = (RK - XL)2 + SK2 + s 2 DKL = distance between
clusters K and L. XK, XL and sK , sL2 = means and variances of clusters K and L, respectively.








45

the variability observed within clusters (Fig. 4-1) suggested the presence of minor factors such as Tfw. To detect additional minor factors that may have been overshadowed by I-3, two-way factorial (2x2) analyses of variance were conducted for all I-3/marker locus pair combinations. The results of these analyses (Table 4-3) showed again that Tfw, linked to Aps-2, had a significant effect on the response to race 3. Furthermore, Tfw appeared to act independently from 1-3, as suggested by the lack of a significant interaction (Table 4-3). Also, a significant interaction was detected between I-3 and the chromosome segment marked by Dia-2 (Table 4-3). However, one-way analyses of variance showed no significant differences between mean disease rankings of Dia-2 homozygotes and heterozygotes within each I-3 genotype (data not shown); thus, the I-3/Dia-2 interaction appeared to be an artifact. Additional factors affecting resistance may have existed in parts of the genome not covered by the marker loci. The factorial analysis also indicated that the chromosome segment marked by Pqm-2 had no significant effect on resistance (Table 4-3). A two-way contingency table between Pqm-2 and I-3 (Table 4-4) showed a slight trend to skew in favor of recombinants (64%). This resulted in an excess of Pqm-2 homozygotes in the I-3/i-3 class, and an excess of heterozygotes in the i-3/i-3 class. Thus, the cause of the significant difference in mean disease rankings








46

observed at the Pgm-2 locus (Table 4-1) could be attributed to the slight skewing. Finally, a two-way analysis of variance of mean disease rankings for all possible pair combinations between marker loci did not show any significant interactions.

Additional evidence for the presence of linkage between Got-2 and I-3 was obtained through an analysis of breeding lines previously selected for resistance to race 3. The genotype at all segregating marker loci, and the response to race 3 was determined for each line. Three lines, I3R-1, 520, and 604 were uniformly resistant to race 3 and homozygous for the L. pennellii allele at Got-2; whereas, E427 and 572 were segregating for both resistance and Got-2. Thus, there had been no recombination between Got-2 and I-3 in any of the five lines. The lines were homozygous for L. esculentum alleles at the other marker loci with the following exceptions: E427 was segregating at Prx-l, ADs-l, Prx-4, and Aco-1, and 520 was homozygous for the L. pennellii allele at Aco-1. E427 is a breeding line known to be fixed for resistance to race 3. However, the E427 seed used in this study originated from a seed lot suspected of contamination by cross pollination which occurred in the field during seed increase.

Currently, race 3 of the Fusarium wilt pathogen is primarily confined to Florida, Australia, and parts of California; however, since races 1 and 2 have a worldwide








47

distribution, it is probable that race 3 will eventually

become a problem in other tomato growing regions. The

results of this work demonstrate that Got-2 can be used as a

scorable marker to facilitate the transfer of I-3 into

commercial cultivars. Using the 2.5% recombination

estimate, only two plants (N=2) would need to be selected using Got-2 as a marker to maintain a 99.9% chance3 of coselecting 1-3.

It should be noted that only one isolate of race 3 was

used as a source of inoculum in this study. However,

breeding lines carrying I-3 have also maintained resistance

against the race 3 isolates from Australia (D.J. McGrath,

personal communication, 1988) and California (M. Kuehn,

personal communication, 1988).






















3p = 1-rN; N = In(l-P)/ln(r). P = probability, r = percent recombination, N = no. of plants.









48

Table 4-1. Analyses of monogenic segregations of marker loci and comparisons of BC1 mean disease ratings using a one-way nonparametric analysis of variance. Mean disease ratings were calculated for BC1 individuals based on response to race 3 of their progeny. To perform the one-way nonparametric analysis of variance, the Wilcoxon rank sum test was used. In this procedure, BC1 individuals were ranked according to their mean disease rating, and homozygote and heterozygote mean disease rankings were compared at each marker locus. ee L. esculentum homozygotes; ep L. pennellii heterozygotes.


Mean Mean Wilcoxon Chromo- No. Disease Disease rank sum
some Plants Ratingm Ranking test Locus No. ee ep ee ep ee ep Z stat


Prx-1 1 27 51 3.23 3.24 38.8 39.9 NS Skdh-1 1 31 44 3.34 3.06 41.3 35.6 NS Bnaq-1 1 22 30 3.35 3.08 28.5 25.0 NS Dia-2 1 23 48 3.33 3.14 34.7 38.6 NS Est-9 1 11 29 3.38 3.16 21.6 20.1 NS Mpi-l 2 24 21 2.96 3.27 21.4 24.8 NS Est-7 2 41 38 3.37 3.10 42.9 36.9 NS Prx-2 2 37 34 3.29 3.13 34.5 37.4 NS Prx-7 3 34 39 3.15 3.28 35.3 38.5 NS Ss-1 3 31 39 3.32 3.27 37.0 34.3 NS Pgm-2 4 36 42 2.92 3.52 32.1 45.8 2.66b Aps-l 6 35 38 3.09 3.29 38.8 35.1 NS Aman-1 6 15 22 3.00 3.42 16.8 20.5 NS Got-2 7 47 32 4.00 2.13 55.8 16.8 -7.40** Aps-2 8 43 34 3.40 3.00 43.1 33.8 -1.82* Pn 8 14 53 3.16 3.22 34.2 33.2 NS Prx-4 10 42 36 3.42 3.02 43.2 35.1 NS Sod-1 11 23 33 3.48 3.10 32.4 25.8 NS Est-10 11 29 50 3.48 3.10 45.0 37.1 NS Est-4 12 21 37 3.50 3.15 32.7 27.7 NS 6Pqdh-2 12 20 55 3.29 3.29 37.7 38.1 NS Pgi-i 12 21 58 3.16 3.27 38.3 40.6 NS Aco-1 12 29 50 3.08 3.34 36.4 42.1 NS Hk-1 ? 25 27 3.45 3.12 29.3 23.9 NS

* **Significant at 0.05 and 0.0001 levels, respectively. aBC1 progeny were scored according to a visual rating system; 1 = resistant to 5 = susceptible. significant at 0.01 level using a two-tailed nonparametric analysis of variance with heterozygotes being more susceptible than homozygotes.









49

Table 4-2. Two-way contingency table between Got-2a and 1-3. ee L. esculentum homozygotes; ep L. pennellii heterozygotes.


I-3
ee ep Totals

ee 46 1 47 Got-2
ep 1 31 32

Totals 47 32 79


X2 = 70.92.
Prob < 0.001.
p (% recombination) = 2.5%.








50

Table 4-3. Detection of minor factors taking into account the effect of 1-3. Two-way (I-3 by each marker locus) factorial analyses of variance (2x2) comparing BC1 mean disease rankings were utilized. Main effects were calculated according to Steel and Torrie (1960). ee L. esculentum homozygotes; ep L. pennellii heterozygotes.


Mean Disease Ranking Main Effects
InterLoci ee/ee ee/ep ep/ee ep/ep I-3 Marker action


I-3 Aps-2 58.15 53.00 19.72 13.28 -39.1** 5.80* -0.64
I-3 Dia-2 50.78 58.52 21.57 15.39 -36.2** 0.78 -6.96*
1-3 Pqm-2a 56.09 56.42 13.58 21.38 -38.8** 4.06 3.74


*, **Significant at the 0.05 and 0.0001 levels, respectively.
apgm-2 analysis included to illustrate that significance observed in the nonparametric analysis of variance (Table 41) was an artifact; no significance existed after accounting for 1-3.









51

Table 4-4. Two-way contingency table comparing I-3 and Pgm2. ee L. esculentum homozygotes; ep L. pennellii heterozygotes.


I-3
ee ep Totals

ee 16 20 36 P m-2
ep 30 12 42

Totals 46 32 78


X2 = 5.834.
Prob < 0.05.
p (% recombination) = 64.1%.








52









16 14
>12
O
C 10 3 8 Q)6
LL 4
2 0
1.0 1.4 1.8 2.2 2.6 3.0 3.4 3.8 4.2 4.6 5.0
Mean Disease Rating







Fig. 4-1. Frequency distribution of backcross (BC1) mean disease ratings. For each BC1, at least twenty progeny (BC1S1) were inoculated with race 3 and scored for their level of vascular browning. Higher scores represented greater levels of vascular browning, and thus greater susceptibility. A mean disease rating for each BC1 individual was calculated by taking a weighted mean of its progeny scores.









53

































Fig. 4-2. GOT starch gel. The origin is at the bottom of the gel; the anode electrode is at the top of the gel. Got2 is the second zone of activity from the top. BB is L. esculentum cv. Bonny Best. 'Bonny Best', LA 716, and F1 controls are in the center of the gel. The other wells are F2 progeny from the cross 'Bonny Best' x LA 716. The LA 716 allele migrates approximately 9 mm further than the L. esculentum allele. The central band in the F1 is an intragenic heterodimer indicating that Got-2 is a dimeric enzyme. Plants selected using Got-2 are 97.5% likely to have the same genotype at the I-3 locus.















CHAPTER 5

GENETIC ANALYSIS OF RESISTANCES TO RACES 1 AND 2 OF FUSARIUM WILT


INTRODUCTION

Race 3 of the tomato Fusarium wilt pathogen Fusarium oxysporum f. sp. lycopersici (Sacc.) Snyder and Hansen is spreading throughout important growing regions worldwide. It has been reported in Australia (Grattidge and O'Brien, 1982), Florida (Volin and Jones, 1982), and California (Davis et al., 1988). A dominant gene, 1-3, that controls resistance to race 3 has been detected in the wild tomato species Lycopersicon pennellii (Corr.) D'Arcy, accession LA 716 (Scott and Jones, 1989), and found to be tightly linked (2.5 cM) to the Got-2 locus on chromosome 7 (see Chapter 4). Since LA 716 was also found to be resistant to races 1 and 2 (Scott and Jones, 1989), a genetic analysis was conducted to elucidate the relationships among the resistances to the three races.

Two parallel approaches were taken for this analysis. First, although most commercial cultivars already carry resistances to races 1 and 2 (I and I-2 on chromosome 11) derived from L. pimpinellifolium accessions (Bohn and Tucker, 1939; 1940; Paddock, 1950; Alexander and Hoover,


54








55

1955; Stall and Walter, 1965; Latterot, 1976), it was of interest to determine whether LA 716 carried additional genes for resistance to these two races. For this reason, allelism of I and I-2 with those resistances present in LA 716 was tested. In the second approach, the chromosomal location of genes controlling resistances to races 1 and 2 found in LA 716 were determined using linkage analyses to 17 previously mapped enzyme loci.


MATERIALS and METHODS

Plant Material

Two progenies were generated for the allelism test using the L. esculentum cultivar Hayslip; this cultivar carries I and 1-2. The first progeny was an F2 of the cross 'Hayslip' x LA 716. Detection of susceptible individuals in this progeny would indicate that the genes were nonallelic. A second progeny [Bonny Best x (Hayslip x LA 716)] was also analyzed because of the increased chances of observing susceptible plants in a nonallelic interaction.

The interspecific backcross, L. esculentum x (L.

esculentum x L. pennellii, LA 716) analyzed previously for the genetics of race 3 resistance (see Chapter 4) was used in this study to analyze the responses to races 1 and 2 through progeny (BC1S1) tests. 'Bonny Best' was used as the L. esculentum parent because of its extreme susceptibility to all three races. The 54 BC1 individuals used in this analysis (those for which BC1S1 seed was still available)









56

were a subset of the 79 individuals previously used to locate I-3 (see Chapter 4). Chi-square analyses indicated that, for all marker loci, ratios of homozygotes to heterozygotes in the subset did not differ significantly from those observed in the overall population. In addition, a Z-test indicated that the race 3 mean disease rating of the subset was not different from that of the overall population.


Isozyme Analysis and Inoculations

Isozyme analysis was conducted on the BC1 individuals as described previously (see Chapter 4) to locate genes responsible for resistances to races 1 and 2. Backcross individuals were scored for each of 23 enzyme loci including Prx-l, Skdh-l, Bnaq-l, Dia-2, and Est-9 on chromosome 1; Est-7, Prx-2, and Mpi-l on chromosome 2; Prx-7 and Ss-l on chromosome 3; Pgm-2 on chromosome 4; Aps- and Aman-1 on chromosome 6; Got-2 on chromosome 7; Aps-2 on chromosome 8; Prx-4 on chromosome 10; Sod-1 and Est-10 on chromosome 11; Est-4, 6Pqdh-2, Pgi-l, and Aco-l on chromosome 12 and Hk-1 whose chromosome designation has not been determined. The morphological marker, Pn on chromosome 8, was also scored. BC1 individuals were self-pollinated to produce BC1S1 seed for progeny tests.

BC1S1 individuals and the progenies used in the allelism test were inoculated according to previously described procedures (see Chapter 4) with races 1 and 2 of








57

F. oxysporum f. sp. lycopersici (race 1: strain SC626, isolated by Dr. M. Cirulli in Italy, also called the "Oristano" isolate; race 2: strain SC548, isolated by Dr. J.M. Walter at the University of Florida at Bradenton). Included in each set of inoculations were 'Bonny Best', LA 716, 'Manapal' (resistant to race 1 only), 'Hayslip', and the F1 (Bonny Best x LA 716) to monitor the activity and specificity of inoculum. Also, another set of controls was treated with distilled water only. Plants were evaluated 10-14 days after inoculation and again approximately one month after inoculation using the visual rating system of Chapter 4. A mean disease rating was calculated for each backcross individual for each race using the progeny scores; at least 20 progeny were used for each analysis. All statistical analyses were performed on a main frame computer using the Statistical Analysis System (SAS Institute Inc., SAS Circle, Box 8000, Cary, NC 27511). Procedures used included the following: Proc Univariate, Proc Cluster, Proc Freq, and Proc Nonparlway.


RESULTS

The detection of susceptible plants in F2 progenies

inoculated with race 1 or 2 indicated that neither I nor I-2 was allelic to the resistances carried by LA 716 (Table 51). The same conclusion was more decisively reached with the results of the L. esculentum backcross progeny where









58

several susceptible plants were observed for each race: 39.6% for race 1 and 39.4% for race 2 (Table 5-1).

Frequency distributions of BC1 mean disease ratings for races 1 and 2 were found to be bimodal according to the coefficient of bimodality (b = (m32+l)/(m4+3), m3 = skewness, m4 = kurtosis; race 1: b = 0.563; race 2: b =

0.584; a < 0.05). A bimodal distribution was also observed for the race 3 mean disease ratings (see Chapter 4). These results indicated that a single major locus was responsible for resistances to races 1 and 2, as was the case for race

3.

The possible association of resistance genes with any

of the 24 segregating enzyme markers was tested by comparing the mean disease ratings of homozygotes and heterozygotes at each marker locus; a one-tailed nonparametric analysis of variance was used for these tests as previously described (see Chapter 4). Heterozygotes for the chromosome segment marked by Got-2 were markedly more resistant than homozygotes for both races 1 and 2 (Table 5-2). Since I-3, a gene conferring race 3 resistance, was also tightly linked to Got-2 (see Chapter 4), these results indicated that resistances to the three races were either controlled by I-3 alone or gene(s) tightly linked to 1-3. Moreover, as previously reported for race 3 (see Chapter 4), significant effects were also observed for the chromosome segment marked by Aps-2 for both races (Table 5-2) suggesting that








59

Tfw, a factor linked to Aps-2, was active against all three races. The chromosome segment marked by Prx-4 also appeared to have an effect for both races; however, this was determined to be an artifact. A two-way contingency table comparing Got-2 and Prx-4 (Table 5-3) showed slight skewing (a < 0.05) in favor of parental types (66%). This resulted in more BC1 individuals that were homozygous or heterozygous for both loci than expected. Since the chromosome segment marked by Got-2 was significant for both races (Table 5-2), the significance observed for Prx-4 can be attributed to skewing with the resistance gene(s) linked to Got-2. In addition, significance for the Prx-4 marker was not observed in the original population (N = 79) analyzed for race 3 resistance (see Chapter 4). Other marker loci assorted as previously reported (Tanksley and Rick, 1980b; Chetelat, 1989); no significant deviations in expected ratios were detected between unlinked loci.

In order to discern whether I-3 alone or other tightly linked loci were responsible for resistances to races 1 and 2, the genotype (resistant or susceptible) at each of the putative race specific loci was deduced for all BC1 individuals based on mean disease ratings of their progeny. For this purpose, the average linkage clustering analysis was used as described previously (see Chapter 4). Two-way contingency tables were used to test for independent assortment between the putative loci that control








60

resistances to the three races. No recombinants were detected between resistances to races 2 and 3; however, seven BC1 individuals appeared to be recombinants between race 1 resistance and the other two resistances (Table 5-4).


DISCUSSION

The data presented here seem to indicate that I-3

confers resistance to races 2 and 3, and that another linked gene on chromosome 7 controls for resistance to race 1. However, a closer examination of the seven putative recombinants between race 1 resistance and resistances to races 2 and 3 suggests that they may have been misclassified by the cluster analysis. First, mean disease ratings of the putative recombinants were very close to the cuttoff point for the resistant class for race 1 (Table 5-4). In addition, all seven recombinants appeared in the same class

-- resistant for race 1 and susceptible for races 2 and 3 (Table 5-4). This was due to the fact that the upper limit of the race 1 resistant cluster was much greater than the limits for races 2 and 3 (Table 5-4). The presence of minor factors, such as Tfw, increases the frequency of individuals with intermediate disease ratings; these individuals have a greater probability of being misclassified. An additional reason for the discrepancy in clustering could have been due to the slightly greater aggressiveness of race 1 compared to the other two races. For instance, all 'Bonny Best' x LA 716 -- F1 plants evaluated were given a rating of 1 or 2 for








61

all three races (Table 5-1). However, a 2 rating was given to 16.7% of the Fl's analyzed for race 1 compared to only 7.1% and 4.1% for races 2 and 3, respectively. Though no clear recombinants were observed between race resistances, the possibility of more than one resistance gene on chromosome 7 could not be ruled out with the available data.

In the past, the Fusarium wilt pathogen has been able

to overcome host genes for resistance (I and I-2) and infect commercial tomato cultivars. Races 1 and 2 have been spread to most tomato growing regions, and still pose the threat of forming a new race. Since Got-2 is tightly linked to monogenic resistances to all three races, it can be used as a scorable marker to incorporate genetic resistances to each of the races into new cultivars. This might reduce the probability of a new race forming in the near future. In addition, if Tfw were tagged with a chromosome 8 marker, its incorporation into new cultivars might offer even more stable resistance to each of the races. It should be noted that only one isolate of each of the races was used as a source of inoculum in this study, and therefore, these results should be verified with other isolates.

Previous reports have concluded that genes for

resistance to obligate parasites such as Puccinia, Uromyces, and Erysiphe are the product of coevolution between these pathogens and their hosts (Anikster and Wahl, 1979; Eshed








62

and Dinoor, 1981; Moseman et al., 1983; 1984). E. oxvsporum is a facultative parasite with the ability to survive away from its host. An interesting feature of the L. pennellii resistance to Fusarium wilt is that it may not be the result of selection pressure on the host. The natural habitat of L. pennellii (LA 716) is in the deserts of southern Peru where it grows on the slopes of rocky soils with nearly no precipitation (Rick and Tanksley, 1981). Either the tomato Fusarium wilt pathogen can adapt to the extreme environment of L. pennellii, or the resistance to the fungus is not the result of coevolution between pathogen and host. A survey of Fusarium populations and an assessment of their pathogenic activity in the area of distribution of L. pennellii could elucidate whether LA 716 may have had an evolutionary interaction with F. oxysporum.









63

Table 5-1. Results of the inoculations for allelism tests for I and I-2 resistances with LA 716 resistances to races 1 and 2. Visual ratings were given one month after inoculations.


Visual rating Total % (No. of plants) No. SuscepCrossb Race 1 2 3 4 5 Plants tiblec


Controls
Bonny Best 1 0 0 0 0 29 29 100.0 2 0 0 0 0 34 34 100.0

LA 716 1 21 0 0 0 0 21 0.0 2 14 0 0 0 0 14 0.0

Bonny Best x LA 716 1 25 5 0 0 0 30 0.0 2 39 3 0 0 0 42 0.0

Manapal 1 17 8 1 1 0 27 7.4 2 1 1 7 14 28 51 96.1

Hayslip 1 15 5 1 0 0 21 4.8 2 13 7 2 1 0 23 13.0


Allelism proQenies
F2 of Hayslip x LA 716 1 32 1 0 0 1 34 2.9 2 32 2 0 0 1 35 2.8

BB x (Hayslip x LA 716) 1 31 4 9 8 6 58 39.6 2 48 9 7 6 24 94 39.4


**Significant at the 0.05 and 0.01 levels, respectively. aDenotes the number of plants falling into each visual rating category; 1, resistant to 5, susceptible. bBB Bonny Best: i/i, i-2/i-2; Manapal I/I, i2/i-2; Hayslip:
1/I, 1-2/I-2.
cPlants with a visual rating of 3 - 5 were classified as susceptible; whereas, ratings of 1 and 2 were considered resistant.









64




Table 5-2. Comparisons of BC1 mean disease ratings using a one-way nonparametric analysis of variance -- the Wilcoxon rank sum test. In this procedure, BC1 individuals were ranked according to their mean disease rating, and homozygote and heterozygote mean disease rankings were compared at each marker locus for each race. ee L. esculentum homozygotes; ep L. Dennellii heterozygotes; R1, R2, and R3 races 1, 2, and 3, respectively.a


R1 mean R2 mean R3 mean Wilcoxon Chromo- No. disease disease disease rank sum test some plants ranking rankina ranking (Z statistic)
Locus No. ee ep ee ep ee ep ee ep R1 R2 R3


Prx-1 1 16 38 26.5 27.9 28.3 27.7 (26.4 28.0) NS NS (NS) Skdh-1 1 16 35 27.5 25.3 28.7 24.8 (28.2 25.0) NS NS (NS) Bna-l1 1 14 29 22.0 22.0 23.2 21.4 (22.6 21.7) NS NS (NS) Dia-2 1 12 35 24.7 23.8 28.2 22.5 (28.1 22.6) NS NS (NS) Est-9 1 7 20 15.3 13.6 13.4 15.6 (15.4 13.5) NS NS (NS) Est-7 2 30 24 27.2 27.8 28.2 26.7 (27.8 27.1) NS NS (NS) Prx-2 2 25 23 23.3 25.8 23.5 25.6 (22.6 26.5) NS NS (NS) MDi-l 2 18 15 15.2 19.2 13.9 20.7 (14.9 19.5) NS NS (NS) Prx-7 3 24 25 25.4 24.6 23.0 26.9 (23.3 26.6) NS NS (NS) Ss-1 3 21 30 26.7 25.5 26.3 25.8 (27.0 25.3) NS NS (NS) Pgi-2 4 26 28 24.0 30.7 25.5 29.4 (24.2 30.5) NS NS (NS) ADs-1 6 28 23 25.9 26.2 26.6 25.2 (25.3 26.8) NS NS (NS) Aman-1 6 13 17 14.9 15.9 14.8 16.0 (13.6 16.9) NS NS (NS,, Got-2 7 29 25 39.6 13.5 40.0 13.0 (40.0 13.0) -6.1 -6.3 (-6.3
ADs-2 8 31 23 31.4 22.2 31.6 22.0 (31.6 22.0) -2.1* -2.2* (-2.2*)
Pn 8 8 39 23.9 24.0 24.2 24.0 (25.9 23.6) NS NS (NSb Prx-4 10 30 23 32.1 20.4 30.2 22.8 (31.2 21.6) -2.7* -1.7 (-2.2)b
Sod-1 11 18 19 20.6 17.5 20.0 18.1 (20.7 17.4) NS NS (NS) Est-10 11 21 33 31.6 24.9 31.3 25.1 (30.2 25.8) NS NS (NS) Est-4 12 15 25 23.9 18.5 24.6 18.1 (22.9 19.1) NS NS (NS) 6Padh-2 12 15 37 26.5 26.5 26.7 26.4 (25.8 26.8) NS NS (NS) Pqi-1 12 14 40 26.8 27.8 26.8 27.8 (25.5 28.2) NS NS (NS) Aco-1 12 20 34 25.2 28.8 23.5 29.8 (24.5 29.3) NS NS (NS) Hk-1 ? 17 19 18.4 18.6 18.5 18.5 (19.1 18.0) NS NS (NS)


*, ** Significant at 0.05 and 0.0001 levels, respectively. aRace 3 mean disease rankings and Z-statistics calculated using only the 54 individuals analyzed for races 1 and 2. bin the original population (N = 79) analyzed for race 3, the chromosome segment marked by Prx-4 was not significant (see Chapter 4).









65

Table 5-3. Two-way contingency table comparing Got-2 and Prx-4. ee L. esculentum homozygotes; ep _L. pennellii heterozygotes.


Got-2
ee ep Totals

ee 20 10 30 Prx-4
ep 8 15 23

Totals 28 25 53


X2 = 5.31.
Prob < 0.05.
% parental types = 66.0%.









66

Table 5-4. Mean disease ratings and cluster classifications of putative recombinants between resistances to races 1 and 2/3. BC1 individuals were classified as resistant or susceptible for each race using the average linkage clustering method based on their mean disease ratings. R1, R2, and R3 for races 1, 2, and 3, respectively; R and S resistant and susceptible.a


Cluster
classification Mean disease rating
BC# R1 R2 R3 R1 R2 R3


12 R S S 3.15 3.78 3.49 22 R S S 3.29 3.46 3.64 41 R S S 3.64 4.08 3.92 65 R S S 2.90 3.30 3.05 71 R S S 3.59 4.00 3.96 92 R S S 3.36 3.77 3.06 130 R S S 3.24 3.63 3.70

Ranges
Resistant cluster 1.55 - 3.64 1.39 - 2.86 1.50 - 2.70 Susceptible cluster 3.78 - 4.88 3.30 - 4.93 3.05 - 4.97

aData for race 3 clustering analysis was obtained from Chapter 4.














CHAPTER 6

SUMMARY AND CONCLUSIONS


Fusarium wilt of tomato, caused by races 1, 2, and 3 of the pathogen Fusarium oxysporum f. sp. lycopersici, is a potentially devastating disease found in most growing regions worldwide. Currently, use of race 3 for screening purposes by breeders is restricted due to fear of introduction of this race into new regions. Use of Got-2 as a scorable marker to incorporate monogenic resistances to all three races into commercial cultivars will have two benefits. One, it will allow for production of race 3 resistant breeding lines and cultivars without introducing the pathogen into new growing regions; and second, it will result in the incorporation of additional monogenic resistances to races 1 and 2. This may reduce the likelihood of future new race formation by these two races. Also, more durable resistance may be acquired by incorporating Tfw into new cultivars.

Several future studies should be conducted. Using near isogenic lines segregating only for Got-2 and resistances to races 1, 2, and 3 in a L. esculentum background, it will be possible using progeny tests to determine the number of genes controlling these resistances and a gene order if more 67








68

than one gene exists. The Tfw resistance should be analyzed to determine whether it confers field tolerance against the pathogen, and whether it acts as a dominant or recessive allele. Also, a mapping project must be conducted to identify a scorable marker (Aps-2 or an RFLP on chromosome 8) for this gene so that it can be incorporated along with the chromosome 7 resistances into new cultivars. Modes of action of 1, I-2, and the L. pennellii resistances could be analyzed and compared by observing plant responses to each of the races in near isogenic lines carrying one or more of the resistance genes.














APPENDIX

VEGETATIVE COMPATIBILITY AMONG RACES 1, 2, AND 3 OF
FUSARIUM OXYSPORUM F. SP. LYCOPERSICI


Genetic relationships among races 1, 2, and 3 of Fusarium oxysporum f. sp. lycopersici were explored by analyzing vegetative compatibility through heterokaryosis. Heterokaryosis is the process of anastomosis by fungal hyphae, forming a cell with two different nuclei in the same cytoplasm. Genotypes are only compatible if all vic (vegetative incompatibility) loci are identical. Previously, Sidhu and Webster (1979) were able to generate heterokaryons between races 1 and 2 using amino acid auxotrophs.

Nitrate nonutilizing (nit) auxotrophic mutants were

developed for each race using the techniques of Cove (1976). Mutants were paired on Czapek-Dox minimal medium and allowed to grow at 28 oC until they came into contact. The appearance of thick wild type growth at the intersection indicated complementation and a potential heterokaryon. Prototrophic colonies were tested for the possibility that wild type growth could be due to crossfeeding. Dialysis tubing (permeable to MW 12,000 - 14,000) was placed between complementary mutants; growth along the tubing would have 69








70

indicated crossfeeding. Also, in another crossfeeding test, one nit mutant was grown in liquid culture at room temperature for 1 - 2 weeks at 100 rpm and filtered through a 0.45 um micropore to collect diffusable metabolites. The filtrate was then incorporated into cooled (40 OC) CzapekDox agar and a complementary mutant plated. The appearance of wild type growth would have indicated crossfeeding.

Complementation was observed between nit mutants of

races 2 and 3; both crossfeeding tests indicated that this was due to heterokaryosis and not crossfeeding. None of twenty race 1 nit mutants showed complementation to nit mutants of races 2 and 3. However, race 1 mutants did not show complementation among themselves either. These data indicate that races 2 and 3 are vegetatively compatible, and likely share a similar phylogenetic background. However, no conclusions about vegetative compatibility or phylogeny between race 1 and races 2 and 3 can be made from these results.















REFERENCES


Ainsworth, G.C. and G.R. Bisby. 1943. A dictionary of the
fungi. The Imperial Mycological Institute. Kew and Surrey.
p 228.

Albersheim, P. and A.J. Anderson. 1971. Proteins from plant
cell walls inhibit polygalacturonases secreted by plant
pathogens. Proc. Natl. Acad. Sci. U.S.A. 68:1815-1819.

Alexander, L.J. and M.M. Hoover. 1955. Disease resistance in
wild species of tomato. Ohio Agr. Exp. Sta. Res. Bull.
752.

Alexander, L.J. and C.M. Tucker. 1945. Physiological
specialization in the tomato wilt fungus Fusarium
oxysporum f. sp. lycopersici. J. Agr. Res. 70:303-313.

Alon, H., J. Katan, and N. Kedar. 1974. Factors affecting
penetrance of resistance to Fusarium oxysporum f. sp.
lycopersici in tomatoes. Phytopathology 64:455-461.

Armstrong, G.M. and J.K. Armstrong. 1948. Nonsusceptible
hosts as carriers of wilt Fusaria. Phytopathology
38:808-826.

Baker, K.F. 1953. Fusarium wilt of China aster. In:
Stefferud, A. (ed.). Plant diseases. U.S. Dept. Agric.
Yearbook of Agriculture. Washington, D.C., pp 572-577.

Bassiri, A. and I. Rouhani. 1977. Identification of broad
bean cultivars based on isozyme patterns. Euphytica
26:279-286.

Beckman, C.H. 1987. The nature of wilt diseases of plants.
APS Press. The American Phytopathological Society. St.
Paul, MN. 175 p.

Beckman, C.H. and S. Halmos. 1962. Relation of vascular
occluding reactions in banana roots to pathogenicity of
root-invading fungi. Phytopathology 52:893-897.





71









72

Beckman, C.H., S. Halmos, and M.E. Mace. 1962. The
interaction of host, pathogen, and soil temperature in relation to susceptibility to Fusarium wilt of bananas.
Phytopathology 52:134-140.

Beckman, C.H., J.E. Kuntz, A.J. Riker, and J.G. Berbee.
1953. Host responses associated with the development of
oak wilt. Phytopathology 43:448-454.

Beckman, C.H., W.C. Mueller, and M.E. Mace. 1974. The
stabilization of artificial and natural cell wall
membranes by phenolic infusion and its relation to wilt
disease resistance. Phytopathology 64:1214-1220.

Beckman, C.H., W.C. Mueller, B.J. Tessier, and N.A.
Harrison. 1982. Recognition and callose deposition in
response to vascular infection in Fusarium wiltresistant or susceptible tomato plants. Physiol. Plant
Pathol. 20:1-10.

Beckman, C.H. and G.E. Zaroogian. 1967. Origin and
composition of vascular gel in infected banana roots.
Phytopathology 57:11-13.

Biehn, W.L. 1973. p-Fluorophenylalanine selectively inhibits
sporulation of two wilt fungi. Phytopathology 63:421423.

Bishop, C.D. and R.M. Cooper. 1983. An ultrastructural study
of root invasion in three vascular wilt diseases.
Physiol. Plant Pathol. 24:277-289.

Bohn, G.W. and C.M. Tucker. 1939. Immunity to Fusarium wilt
in the tomato. Science 89:603-604.

Bohn, G.W. and C.M. Tucker. 1940. Studies on Fusarium wilt
of the tomato. I. Immunity in Lycopersicon
pimpinellifolium and its inheritance in hybrids. Mo. Agr. Exp. Sta. Res. Bull. p 311.

Booth, C. 1971. In: The Genus Fusarium. Commonwealth Mycol.
Inst. Kew, Surrey, and England, p 233.

Brown, A.H.D. and B.S. Weir. 1983. Measuring genetic variability in plant populations. In: Tanksley, S.D. and T.J. Orton (eds.) Isozymes in genetics and breeding. Elsevier. Amsterdam, Oxford, and New York, pp 219-240.

Buckley, P.M., T.D. Wyllie, J.E. DeVay. 1969. Fine structure of conidia and conidium formation in Verticillium alboatrum and V. niQrescens. Mycologia 61:240-250.








73

Chattway, M.M. 1949. The development of tyloses and
secretion of gum in heartwood formation. Aust. J. Sci.
Res., Ser. B 2:227-240.

Chetelat, R.T. 1989. Isozyme gene linkage map of tomato
(Lycopersicon esculentum). Isozyme Bulletin 22:19.

Clayton, E.E. 1923a. The relation of temperature to the
Fusarium wilt of the tomato. Amer. J. Bot. 10:71-88.

Clayton, E.E. 1923b. The relation of soil moisture to the
Fusarium wilt of the tomato. Amer. J. Bot. 10:133-147.

Cove, D.J. 1976. Chlorate toxicity in Aspergillus nidulans:
the selection and characterisation of chlorate resistant
mutants. Heredity 36:191-203.

Crawford, D.J. 1983. Phylogenetic and systematic inferences
from electrophoretic studies. In: Tanksley, S.D. and T.J. Orton (eds.) Isozymes in breeding and genetics.
Elsevier. Amsterdam, Oxford, and New York, pp 257-288.

Davis, R.M., K.A. Kimble, and J.J. Farrar. 1988. A third
race of Fusarium oxysporum f. sp. lycopersici
identified in California. Plant Disease 72:453.

Edgington, L.V. and J.C. Walker. 1958. Influence of calcium
and boron nutrition on development of Fusarium wilt of
tomato. Phytopathology 48:324-326.

Fisher, P.L. 1935. Responses of tomato in solution cultures
with deficiencies and excesses of certain essential
elements. Md. Agr. Exp. Sta. Bull. 375.

Foster, R.E. and J.C. Walker. 1947. Predisposition of tomato
to Fusarium wilt. J. Agr. Res. 74:165-185.

Freeling, M. and J.A. Birchler. 1980. Mutants and variants
of the alcohol dehydrogenase-1 gene in maize. In:
Genetic engineering. Setlow, J.K. and A. Hollander (eds.).
Plenum Press. New York, vol 3, pp 223-264.

Friend, J. 1976. Lignification in infected tissue. In:
Friend, J. and D.R. Threlfall (eds.). Biochemical aspects of plant-parasite relationships. Academic Press. New York, pp 291-303.

Gottlieb, L.D. 1981. Electrophoretic evidence and plant populations. Progress in Phytochemistry 7:1-46.









74

Grattidge, R. and R.G. O'Brien. 1982. Occurrence of a third
race of Fusarium wilt of tomatoes in Queensland. Plant
Disease 66:165-166.

Griffin, G.J. 1970. Carbon and nitrogen requirements for
macroconidial germination of Fusarium solani: dependence
on conidial density. Can. J. Microbiol. 16:733-740.

Harrison, A.L. 1941. Studies on tomato wilt. Tex. Agr. Exp.
Sta. Rept. 1940:88.

Heath-Pagliuso, S., R.C. Huffaker, and R.W. Allard. 1984.
Inheritance of nitrite reductase and regulation of
nitrate reductase, nitrite reductase, and glutamine
synthetase isozymes. Plant Physiol. 76:353-358.

Hoffmann, G.M. 1966. Untersuchungen uber die
heterokaryosebildung und den parasexualcyclus bei
Fusarium oxysporum. I. Anastomosenbildung im mycel und
kernverhaltwisse bei der conidienentwicklung. Arch.
Mikrobiol. 53:336-347.

Howard, R.J. and D.P. Maxwell. 1975. Cytology of
Ceratocystis ulmi. Phytopathology 65:816-819.

Hutson, R.A. and I.M. Smith. 1980. Phytoalexins and tyloses
in tomato cultivars infected with Fusarium oxysporum f.
sp. lycopersici or Verticillium albo-atrum. Physiol.
Plant Pathol. 17:245-257.

Ioannou, N. R.W. Schneider, and R.G. Grogan. 1977. Effect of
oxygen, carbon dioxide, and ethylene on growth,
sporulation, and production of microsclerotia by Verticillium dahliae. Phytopathology 67:645-650.

Jones, J.P. and S.S. Woltz. 1969. Fusarium wilt (race 2) of
tomato: Calcium, pH, and micronutrient effects on
disease development. Plant Dis. Rep. 53:276-279.

Jones, J.P. and S.S. Woltz. 1970. Fusarium wilt of tomato:
Interaction of soil liming and micronutrient amendments
on disease development. Phytopathology 60:812-813.

Katan, J. 1971. Symptomless carriers of the tomato Fusarium
wilt pathogen. Phytopathology 61:1213-1217.

Kolomogorov, A.N. 1933. Sulla determinazione empirica di una
legge di distribuzione. G. Inst. Ital. Attuari. 4:8391.









75

Latterot, H. 1976. Mapping of I-2 allele in tomato,
controlling the genetic resistance to pathotype 2 of
Fusarium oxysporum f. sp. lycopersici wilt. Ann. Amelior.
Plant. 26(3):485-491.

Mace, M.E. 1963. Histochemical localization of phenols in
healthy and diseased banana roots. Physiol. Plant
16:915-925.

Malca, I., D.C. Erwin, W. Moje, and B. Jones. 1966. Effect
of pH and carbon and nitrogen sources on the growth of
Verticillium albo-atrum. Phytopathology 56:401-406.

Marshall, K.C. and M. Alexander. 1960. Competition between
soil bacteria and Fusarium. Plant Soil 12:143-148.

Massee, G. 1895. The "sleepy disease" of tomatoes. Gard.
Chron. Ser. 3, 17:707-708.

Marchant, R. 1966. Fine structure and spore germination in
Fusarium culmorum. Ann. Bot. 30:441-445.

McGrath, D.J., D. Gillespie, and L. Vawdrey. 1987.
Inheritance of resistance to Fusarium oxysporum f. sp.
lycopersici races 2 and 3 in Lycopersicon pennellii.
Aust. J. Agric. Res. 38:729-733.

Medina-Filho, H.P. and M.A. Stevens. 1980. Tomato breeding
for nematode resistance: survey of resistant varieties for horticultural characteristics and genotype of acid
phosphatase. Acta. Hortic. 100:383-393.

Moore-Landecker, E. 1972. In: Fundamentals of the fungi.
Prentice Hall, Inc. Englewood Cliffs, N.J., pp 132-143.

Mozumder, B.K.G., N.E. Caroselli, and L.S. Albert. 1970.
Influence of water activity, temperature, and their
interaction on germination of Verticillium albo-atrum
conidia. Plant Physiol. 46:347-349.

Nelson, P.E. 1981. Life cycle and epidemiology of Fusarium
oxysporum. In: Fungal wilt diseases of plants. Mace,
M.E., A.A. Bell, and C.H. Beckman (eds.). Academic
Press. New York, London, Toronto, Sydney, and San
Francisco, pp 51-80.

Paddock, E.F. 1950. A tentative assignment of the Fusariumimmunity locus to linkage group 5 in tomato. Genetics
35:683-684.









76

Pegg, G.F. 1976. The occurrence of P-1,3-glucanase in
healthy and Verticillium albo-atrum-infected
susceptible and resistant tomato plants. J. Exp. Bot.
27:1093-1101.

Pegg, G.F. and D.H. Young. 1981. Changes in glycosidase
activity and their relationship to fungal colonization
during infection of tomato Lycopersicon esculentum
cultivar Craigella by Verticillium albo-atrum. Physiol.
Plant Pathol. 19:371-382.

Phipps, P.M. and R.J. Stipes. 1976. Histopathology of Mimosa
infected with Fusarium oxysporum f. sp. perniciosum.
Phytopathology 66:839-843.

Puhalla, J.E. and A.A. Bell. 1981. Genetics and biochemistry
of wilt pathogens. In: Mace, M.E., A.A. Bell, and C.H.
Beckman (eds.). Fungal wilt diseases of plants. Academic
Press. New York, London, Toronto, Sydney, and San
Francisco. pp 145-192.

Puhalla, J.E. and J.E. Mayfield. 1974. The mechanism of
heterokaryotic growth in Verticillium dahliae. Genetics
76:411-422.

Richards, B.L. 1925. Canning crop diseases. Utah Agr. Exp.
Sta. Bull. 192:59-60.

Rick, C.M. 1969. Controlled introgression of chromosomes of
Solanum pennellii into Lycopersicon esculentum:
segregation and recombination. Genetics 62:753-768.

Rick, C.M. and J.F. Fobes. 1974. Association of an allozyme
with nematode resistance. Rep. Tomato Genet. Coop.
24:25.

Rick, C.M. and J.F. Fobes. 1977. Linkage relations of some
isozymic loci. Rep. Tomato Genet. Coop. 27:22-24.

Rick, C.M., J.F. Fobes, and M. Holle. 1977. Genetic variation in Lycopersicon pimpinellifolium: evidence of evolutionary change in mating systems. Plant Syst.
Evol. 127:139-170.

Rick C.M. and S.D. Tanksley. 1981. Genetic variation in
Solanum pennellii: comparisons with two other sympatric
tomato species. Pl. Syst. Evol. 139:11-45.

Robinson, P.M. and M.K. Garrett. 1969. Identification of volatile sporostatic factors from cultures of Fusarium oxvsporum. Trans. Br. Mycol. Soc. 52:293-299,









77

Robinson, P.M., D. Park, and W.K. McClure. 1969.
Observations on induced vacuoles in fungi. Trans. Br.
Mycol. Soc. 52:447- 450.

Santamour, F.S. and P. Demuth. 1980. Identification of
Callery pear cultivars by peroxidase isozyme patterns.
J. Heredity 71:447-449.

Schwartz, D. 1969. Alcohol dehydrogenase in maize: genetic
basis for multiple isozymes. Science 164:585-586.

Scott, J.W. and J.P. Jones. 1989. Monogenic resistance in
tomato to Fusarium oxysporum f. sp. lycopersici race 3.
Euphytica 40:49-53.

Shaw, C.R. and R. Prasad. 1970. Starch gel electrophoresis
of enzymes -- a compilation of recipes. Biochem. Genet.
4:297-320.

Sherwood, E.C. 1923. Hydrogen-ion concentration as related
to the Fusarium wilt of tomato seedlings. Amer. J. Bot.
10:537-553.

Sidhu, G.S. and J.M. Webster. 1979. A study of heterokaryosis and its influence on virulence in Fusarium oxysporum lycopersici. Can. J. Bot. 57:548-555.

Snedecor, G.W. and W.G. Cochran. 1980. Statistical methods
(7th ed.). The Iowa State University Press. Ames, Ia. p
78-81.

Snyder, W.C. and H.N. Hansen. 1940. The species concept in
Fusarium. Amer. J. Bot. 27:64-67.

Sokal, R.R. and C.D. Michener. 1958. A statistical method
for evaluating systematic relationships. Univ. of Ks.
Science Bulletin 38:1409-1438.

Soost, R.K., T.E. Williams, and A.M. Torres. 1980.
Identification of nucellar and zygotic seedlings of
Citrus with leaf isozymes. HortScience 15:728-729.

Stall, R.E. 1961. Development of Fusarium wilt on resistant varieties of tomato caused by a strain different from race 1 isolates of Fusarium oxysporum f. lycopersici.
Plant Dis. Rep. 45:12-15.

Stall, R.E. and J.M. Walter. 1965. Selection and inheritance
of resistance in tomato to isolates of races 1 and 2 of the Fusarium wilt organism. Phytopathology 55:12131215.









78

Steel, R.G.D. and J.H. Torrie. 1960. Principles and
procedures of statistics. McGraw-Hill Book Company,
Inc. New York, Toronto, and London, pp 194-231.

Stoessl, A., J.B. Stothers, and E.W.B. Ward. 1976.
Sesquiterpenoid stress compounds of the Solanaceae.
Phytochemistry 15:855-872.

Stotzky, G. and R.D. Goos. 1965. Effect of high CO2 and low
02 tensions on the soil microbiota. Can. J. Microbiol.
11:853-868.

Stover, R.H. 1962. Studies on Fusarium wilt of bananas. IX.
Competitive saprophytic ability of F. oxysporum f.
cubense. Can. J. Bot. 40:1473-1481.

Stover, R.H. 1970. In: Toussoun, T.A., R.V. Bega, and P.E.
Nelson (eds.). Root diseases and soil-borne plant
pathogens. Univ. California Press. Berkeley, Ca., pp
197-200.

Strobel, J.W., N.C. Hayslip, D.S. Burgis, and P.H. Everett.
1969. Walter, a determinate variety resistant to races 1
and 2 of the Fusarium wilt pathogen. Fl. Agr. Exp. Sta.
Cir. S-202.

Talboys, P.W. 1958a. Some mechanisms contributing to
Verticillium-resistance in the hop root. Trans. Br.
Mycol. Soc. 41:227-241.

Talboys, P.W. 1958b. Association of tylosis and hyperplasia
of the xylem with vascular invasion of the hop by
Verticillium albo-atrum. Trans. Br. Mycol. Soc. 41:249260.

Tanksley, S.D. 1979. Linkage, chromosomal association, and
expression of Adh-1 and Pam-2 in tomato. Biochem.
Genet. 17:1159-1167.

Tanksley, S.D. 1983. Introgression of genes from wild
species. In: Tanksley, S.D. and T.J. Orton (eds.)
Isozymes in genetics and breeding. Elsevier. Amsterdam,
Oxford, and New York, pp 331-338.

Tanksley, S.D. and R.A. Jones. 1981. Effects of 02 stress on
tomato alcohol dehydrogenase activity: description of a
second ADH coding gene. Biochem. Genet. 19:397-409.

Tanksley, S.D. and F. Loaiza-Figueroa. 1985. Gametophytic
self-incompatibility is controlled by a single major
locus on chromosome 1 in Lycopersicon peruvianum. Proc. Natl. Acad. Sci. 82:5093-5096.









79


Tanksley, S.D., H. Medina-Filho, and C.M. Rick. 1981. The
effect of isozyme selection on metric characters in an interspecific backcross of tomato -- basis of an early
screening procedure. Theor. Appl. Genet. 60:291-296.

Tanksley, S.D., H. Medina-Filho, and C.M. Rick. 1982. Use of
naturally occurring enzyme variation to detect and map
genes controlling quantitative traits in an
interspecific backcross of tomato. Heredity 49:11-25.

Tanksley, S.D., J.C. Miller, A.H. Paterson, and R.
Bernatzky. 1988. Molecular mapping of plant chromosomes.
In: Gustafson, J.P. and R. Appels (eds.). Chromosome
structure and function. Plenum Publishing Co. pp 157-173.

Tanksley, S.D. and C.M. Rick. 1980a. Genetics of esterases
in species of Lycopersicon. Theor. Appl. Genet. 56:209219.

Tanksley, S.D. and C.M. Rick. 1980b. Isozymic gene linkage
map of the tomato: applications in genetics and
breeding. Theor. Appl. Genet. 57:161-170.

Tanksley, S.D., C.M. Rick, and C.E. Vallejos. 1984. Tight
linkage between a nuclear male-sterile locus and an enzyme marker in tomato. Theor. Appl. Genet. 68:109113.

Tolmsoff, W.J. 1973. Life cycles of Verticillium species.
In: Verticillium wilt of cotton. U.S. Dept. Agric.
Publ. ARS-S-19, pp 20-38.

Toussoun, T.A. and P.E. Nelson. 1976. A pictorial guide to
the identification of Fusarium species (2nd ed.). Pa.
State Univ. Press. University Park, Pa.

Typas, M.A. and J.B. Heale. 1976. Heterokaryosis and the
role of cytoplasmic inheritance in dark resting
structure formation in Verticillium SDR. Mol. Gen.
Genet. 146:17-26.

Vallejos, C.E. 1983. Enzyme activity staining. In: Tanksley
S.D. and T.J. Orton (eds.). Isozymes in plant genetics
and breeding. Elsevier. Amsterdam, Oxford, and New
York, pp 469-516.

Vallejos, C.E. and S.D. Tanksley. 1983. Segregation of
isozyme markers and cold tolerance in an interspecific
backcross of tomato. Theor. Appl. Genet. 66:241-247.









80

Van Eck, W.H. and B. Schippers. 1976. Ultrastructure of
developing chlamydospores of Fusarium solani f.
cucurbitae in vitro. Soil Biol. Biochem. 8:1-6.

Volin, R.B. and J.P. Jones. 1982. A new race of Fusarium
wilt of tomato in Florida and sources of resistance.
Proc. Fl. State Hort. Soc. 95:268-270.

Walker, J.C. 1971. Fusarium wilt of tomato. Monograph 6. The
American Phytopathological Society. St. Paul, MN.

Walker, J.C. and R.E. Foster. 1946. Plant nutrition in
relation to disease development. III. Fusarium wilt of
tomato. Amer. J. Bot. 33:259-264.

Wardlaw, C.W. 1931. The biology of banana wilt (Panama
disease). III. An examination of sucker infection
through root bases. Ann. Bot. 45:381-399.

Weber, J.C. and R.F. Stettler. 1981. Isoenzyme variation
among ten populations of Populus trichocarpa Torr. et Gray in the Pacific Northwest. Silvae Genetica 30:2-3.

Weeden, N.F. 1984. Distinguishing among white seeded bean
cultivars by means of allozyme genotypes. Euphytica
33:199-208.

Weeden, N.F. 1986. Genetic confirmation that the variation
in the zymograms of 3 enzyme systems is produced by
allelic polymorphism. Ann. Rep. Bean Improv. Coop.
29:117-118.

Weeden, N.F. and R.C. Lamb. 1985. Identification of apple
cultivars by isozyme phenotypes. J. Amer. Soc. Hort.
Sci. 110:509-515.

Weeden, N.F., R. Provvidenti, and G.A. Marx. 1984. An
isozyme marker for resistance to bean yellow mosaic
virus in Pisum sativum. J. Heredity 75:411-412.

Weller, J.I., M. Soller, and T. Brody. 1988. Linkage
analysis of quantitative traits in an interspecific
cross of tomato (Lycopersicon esculentum x Lycopersicon pimpinellifolium) by means of genetic markers. Genetics
118:329-339.

Wilcoxon, F. 1945. Individual comparisons by ranking methods. Biometrics 1:80.

Wollenweber, H.W. and O.A. Reinking. 1935. Die Fusarien,
ihre Beschreibung, Schadwirkung aund Bekampfung. Paul
Pareyk, Berlin. p 335.














BIOGRAPHICAL SKETCH


Brian Bournival was born June 12, 1961 in Cleveland,

Ohio, the son of Herbert and Neva Bournival. He earned his Bachelor of Science degree from Ohio State University in March of 1983. In January of 1986, he was granted a Master of Science degree from the University of Illinois, where he studied isozymes and their inheritance in the apple.
































81








I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Jfn W. Scott, Chair
Associate Professor of
Horticultural Science

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




C. Eduardo Valljos, Cochair Assistant Professor of
Horticultural Science

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




H. Corby Kistler
Assistant Professor of Plant Pathology

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




L. Curtis Hannah
Professor of Horticultural
Science









I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.




Jo P. Jone (
Pro essor of Plant Rathology

This dissertation was submitted to the Graduate
Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy.



December 1989
Dean, Cogege of Agrilture



Dean, Graduate School
























































UNIVERSITY OF FLORIDA

I 111111 II Il Illl III9lII i i III 111II 11Il
3 1262 08556 7930




Full Text

PAGE 1

GENETICS OF RESISTANCE TO FUSARIUM WILT IN TOMATO By BRIAN BOURNIVAL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1989

PAGE 2

This dissertation is dedicated in loving memory to my aunt, Arline Bournival, who died during my tenure at the University of Florida.

PAGE 3

ACKNOWLEDGEMENTS I would like to extend my gratitude to Dr. Jay Scott and Dr. Eduardo Vallejos, my coadvisors in this project. Without their assistance and guidance, this study would not have been possible. I would also like to thank my committee members, Dr. Corby Kistler, Dr. Curt Hannah, and Dr. J. P. Jones, for their invaluable contributions to the project. Sincere gratitude is also extended to Dr. Dan Cantliffe for making his growth chambers available for the inoculations. I also thank Bruce Ritchings for watering my plants every other weekend and allowing me to leave Gainesville occasionally. Thanks is also extended to Patty Hart for being a special friend and getting me out of the lab and into other activities. Lastly, I would like to thank my parents, Herb and Neva Bournival, and my aunt and cousins in St. Petersburg for their continued support throughout my work here. iii

PAGE 4

TABLE OF CONTENTS Page ACKNOWLEDGEMENTS iii KEY TO ABBREVIATIONS V ABSTRACT vi CHAPTERS 1 INTRODUCTION 1 2 LITERATURE REVIEW 4 ISOZYMES in BREEDING and GENETICS 4 FUSARIUM WILT of TOMATO 10 3 MAPPING OF NEW ENZYME LOCI 20 INTRODUCTION 20 MATERIALS and METHODS 21 RESULTS and DISCUSSION 23 4 ANALYSIS AND MAPPING OF GENES CONTROLLING RESISTANCE TO RACE 3 OF FUSARIUM WILT .... 36 INTRODUCTION 3 6 MATERIALS and METHODS 37 RESULTS and DISCUSSION 41 5 GENETIC ANALYSIS OF RESISTANCES TO RACES 1 AND 2 OF FUSARIUM WILT 54 INTRODUCTION 54 MATERIALS and METHODS 55 RESULTS 57 DISCUSSION 60 6 SUMMARY AND CONCLUSIONS 67 APPENDIX 69 REFERENCES 71 BIOGRAPHICAL SKETCH 81 iv

PAGE 5

KEY TO ABBREVIATIONS Name Acid phosphatase Aconitase Alcohol dehydrogenase a-Mannosidase /3-N-Acetyl glucosaminidase Diaphorase Esterase Glutamate oxaloacetate transaminase Hexokinase Mannose-6-phosphate dehydrogenase Peroxidase Phosphoglucose isomerase Phosphoglucomutase 6-Phosphogluconate dehydrogenase Punctate Shikimate dehydrogenase Sucrose synthase Superoxide dismutase Abbreviation Aps Aco Adh Aman Bnag Dia Est Got Hk Mpi Prx Pgm 6Pgdh Pn Skdh Ss Sod v

PAGE 6

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy GENETICS OF RESISTANCE TO FUSARIUM WILT IN TOMATO By Brian Bournival December 1989 Chairman: Dr. John W. Scott Major Department: Vegetable Crops The inheritance and linkage relationships of resistances to races 1, 2, and 3 of Fusarium oxysporum f. sp. lycopersici derived from Lycopersicon pennellii (LA 716) were analyzed in an interspecific backcross to L. esculentum . Progenies from each backcross (BC]^) individual were inoculated with each of the races, and their responses measured according to a visual rating system; progeny responses were used to calculate a mean disease rating for each BC;l individual for each race. For all three races, the frequency distribution of the disease ratings was bimodal, indicating that resistance was controlled by one major locus in each case. To map resistance gene(s) , disease ratings were compared between homozygotes and heterozygotes at each of 23 segregating marker enzyme loci. For all three races, highly significant differences were detected for the chromosome segment marked by the Got-2 locus on chromosome vi

PAGE 7

7. This indicated that Got-2 was linked to a major gene or genes controlling resistance to the three races. It was not possible with available data to determine the number of genes on chromosome 7 conferring resistance to the three races. To determine whether Got-2 could serve as a selectable marker for the race 3 resistance locus, designated 1-3 . the genotype of each BC 1 individual at 1-3 was determined using cluster analysis of the disease ratings; a test for independent assortment between 1-3 and Got-2 revealed strong linkage, with an estimated map distance of 2.5 cM. The Got2 locus is proposed as a selectable marker to expedite the transfer of resistances to races 1, 2, and 3 into commercial tomato cutivars. In addition, a minor factor controlling resistances to all three races was detected on chromosome 8 linked to Aps— ^ — 2; this locus has been designated Tfw for "tolerance to Fusarium wilt." Tfw might also serve as a source of genetic resistance to Fusarium wilt by increasing stability of resistance and reducing the probability of new race formation by the pathogen. vii

PAGE 8

CHAPTER 1 INTRODUCTION Fusarium wilt of tomato, caused by races 1 and 2 of Fusarium oxysporum (Schlecht.) f. sp. lycopersici (Sacc.) Snyder and Hansen, is a potentially devastating disease found in many tomato growing regions. Control is mainly through the use of genetically resistant cultivars; chemically, only fumigation before planting has proven effective. Monogenic resistances to races 1 and 2 (I and J± 2, respectively) , derived from accessions of Lvcopersicon pimpinellifolium (Bohn and Tucker, 1939, 1940; Alexander and Hoover, 1955; Stall and Walter, 1965), have been transferred to many cultivars. Race 3, a new race with the ability to infect cultivars resistant to races 1 and 2, has been observed in Florida (Volin and Jones, 1982) , Australia (Grattidge and O'Brien, 1982), and recently in California (Davis et al. , 1988) . Currently, no cultivars carry resistance to race 3 . Dominant monogenic resistance to race 3, designated 1-3 . has been observed in two accessions of the wild tomato species L. pennellii . PI414773 (McGrath et al. , 1987) and LA 716 (Scott and Jones, 1989). In a standard breeding program, transfer of a disease resistance gene from a wild species to a domesticated 1

PAGE 9

2 cultivar is a time-consuming process requiring inoculations each generation. Tanksley and Rick (1980b) proposed the use of isozymes as selectable markers to expedite the transfer of important genes from one genetic background to another. Previous investigators analyzing crosses between L. esculentum and L. pennellii have identified several polymorphic enzyme loci dispersed throughout the tomato genome that could serve as potential selectable markers for important traits (Tanksley and Rick, 1980b; Chetelat, 1989) . Use of an isozyme locus as a marker for a disease resistance gene would have many advantages including the following: 1. it is nondestructive to the plant; 2. screening can be done in a matter of hours on very young plants; 3. the enzyme genotype does not affect the overall plant phenotype; 4. migration rate of bands is not affected by the environment; 5. isozymes are codominantly expressed allowing for the differentiation of homozygotes and heterozygotes; 6. additional traits can be screened; and 7. the breeder in uninfected regions would not have to work with the pathogen and risk introduction. A major problem associated with the use of genetic resistance as a control for a disease is the overcoming of resistance by the pathogen through mutation. In addition to race 3 resistance, L. pennellii (LA 716) represents a potential new monogenic source of resistance to races 1 and 2 (Scott and Jones, 1989) . Incorporation of additional

PAGE 10

genetic resistance to these two races may reduce the likelihood of races 1 and 2 of the pathogen developing into a new race. The objectives of this work were to identify and map new polymorphic enzyme loci, analyze race 3 resistance derived from L. pennellii (LA 716) and tag 1-3 , and determine the genetic relationship between LA 716 resistances to races 1, 2, and 3.

PAGE 11

CHAPTER 2 LITERATURE REVIEW ISOZYMES in BREEDING and GENETICS Uses of Isozymes Isozymes can be described as multiple molecular forms of the same enzyme. To analyze the isozymes of a given enzyme system, they must first be extracted and separated. This is usually done through gel electrophoresis. In this technigue, plant tissue samples are ground in an extraction buffer and loaded into a gel. The gel usually consists of starch or polyacrylamide and any of a wide variety of buffers ranging in pH from 4.6 to 9.0 (Shaw and Prasad, 1970) . For starch gel electrophoresis, loading is done by absorbing the extract onto filter paper wicks and placing the wicks in a cut made approximately 1/3 of the way up from the bottom of the gel, whereas for polyacrylamide gel electrophoresis, the liguid extract is placed directly into preformed wells at the top of the gel. An electrical potential is applied across the gel, and the enzymes migrate in the gel according to their molecular weight and charge. After separation, zones of enzyme activity are resolved by incubating the gel in a specific activity stain. 4

PAGE 12

cO The banding pattern of an enzyme system is referred to as a zymogram. Bands produced by the same locus are referred to as allozymes, whereas bands produced by different loci are termed isozymes. Herein, to avoid confusion, all variant bands of the same enzyme system will be referred to as isozymes. There are a variety of reasons why plants have more than one locus producing the same enzyme. Enzyme loci can be tissue and/or organelle specific. For example, there are two phosphoglucomutase (PGM) loci in most plant species. One locus is active in the chloroplasts and is specific to the green tissues of the plant, whereas the other locus is expressed in the cytoplasm and is not tissue specific (Gottlieb, 1981) . Both loci are required for normal plant growth. Chloroplast PGM is involved in starch metabolism, and cytoplasmic PGM is necessary for sucrose synthesis. Having more that one locus for a critical enzyme affords the plant a backup mechanism in case one locus fails. In maize there are two alcohol dehydrogenase (ADH) loci (Schwartz, 1969; Freeling and Birchler, 1980). ADH catalyzes the reduction of acetaldehyde to ethanol, generating NAD + oxidizing power which is important for plant survival under anaerobic conditions. Schwartz (1969) compared seeds lacking Adh-1 activity to wild type seeds, and determined that Adh-1 ~ seeds could not survive the same flooding conditions as Adh-1 + seeds. Normal aerobic growth

PAGE 13

6 was not influenced by the Adh-1 genotype. On the other hand, no effect on flooding tolerance was observed by substituting Adh-2 ~ for Adh-2 + . Seedlings lacking activity at both ADH loci were extremely flood sensitive and did not grow well even under aerobic conditions. These results lead to the conclusion that Adh-1 is the main locus involved in flood tolerance; Adh-2 also contributes to flood tolerance, but is only beneficial when Adh-1 was inactive (Schwartz, 1969; Freeling and Birchler, 1980). Another reason for the existence of isozymes is maintenance of optimal enzyme activity over a range of environments. For example, there are two alcohol dehydrogenase genes in tomato (Tanksley, 1979; Tanksley and Jones, 1981) . Adh-1 is constituitively expressed in pollen and seed tissue (Tanksley, 1979) . However, Adh-2 is only expressed in response to anaerobic induction with no tissue specificity (Tanksley and Jones, 1981) . Apparently Adh-1 is required in the early developmental stages of the plant, whereas Adh-2 is only active when the plant is subjected to an anaerobic environment. Isozymes have several useful properties in plant breeding and genetics (Tanksley and Rick, 1980b). 1. They are codominantly expressed, meaning both alleles in a diploid can be observed. 2. Electrophoretic mobility is not affected by environmental factors. 3. Sampling is nondestructive. 4. Isozyme genotypes can be determined on

PAGE 14

young seedlings in a matter of hours. 5. Also, isozymes have a neutral effect on the overall phenotype of the plant. For an isozyme locus to be useful to geneticists, electrophoretic polymorphism must exist among genotypes. Polymorphic isozyme loci can be used to differentiate between plant genotypes (Bassiri and Rouhani, 1977; Santamour and Demuth, 1980; Weeden, 1984; Weeden and Lamb, 1985) , to separate accidental self-pollinations from hybrids in a breeding program (Soost et al. , 1980) , to determine phylogenetic relationships between species (Crawford, 1983) , to analyze the genetics of a population (Weber and Stettler, 1981; Brown and Weir, 1983), to identify loci controlling quantitative traits (Tanksley et al. , 1982; Weller et al. , 1988), or as selectable genetic markers for important horticultural traits (Rick and Fobes, 1974; Tanksley, 1983; Tanksley et al. , 1984; Weeden et al. , 1984) . Tomato Isozymes Isozymes of tomato have been studied extensively. Currently, 39 isozyme loci have been mapped, with at least one locus on each of the 12 tomato chromosomes (Chetelat, 1989) . For a list of enzyme abbreviations, see page v. Mapped loci include: Idh-1 . Prx-1 , Skdh-1 , Est-3 , and Nir-1 on chromosome 1; Est-7 . Prx-2 , and Fdh-1 on chromosome 2; Prx-7 , Pgm1 , Mdh-1 . and Prx-6 on chromosome 3; 6Pgdh-l . Tpi-2 , Got-1 , Pgm2 , and Adh-1 on chromosome 4; Dia-1 .

PAGE 15

6Pqdh-3 , and Prx-5 on chromosome 5; Aps-1 . Sod-3 . and Adh-2 on chromosome 6; Got-3 , Got-2 . Aco-2 , and Mdh-3 on chromosome 7 ; Aps-2 and Got-4 on chromosome 8 ; Dia-3 and Est-2 on chromosome 9; Est-8 and Prx-4 on chromosome 10; Sod-1 on chromosome 11; and Est-4 , 6Pqdh-2 , Pqi-1 . Mdh-4 , and Aco-1 on chromosome 12. Of these loci, Prx-1 , Skdh-1 , Nir-1 , Est-7 , Prx-2 , Fdh-1 . Prx-7 . Tpi-2 . Pqm-2 , Adh-1 . Dia1, Aps-1 . Got-2 , Aps-2 , Est-2 , Est-8 . Prx-4 . Sod-1 . Est-4 , 6Pqdh-2 , Pqi-1 . and Aco-1 are polymorphic between L. esculentum and L. pennellii (Chetelat, 1989) . Tomato isozyme loci have served a number of purposes to previous investigators. Tanksley et al. (1981) analyzed isozyme loci in gametophytic and sporophytic tissues, and found that most loci (18 of 30) were expressed in both tissues. They proposed the use of gametophytic selection for some traits to obtain desirable sporophytic phenotypes. Tanksley et al. (1981) also determined that gametophytic enzyme loci were expressed postmeiotically because they did not observe intragenic heterodimeric bands at heterozygous dimeric enzyme loci. For intragenic heterodimers to be produced, transcription and translation must take part in diploid cells of premeiotic stages. Isozymes have also been used to map several genes of economic importance. Tanksley et al. (1984) found tight linkage between a nuclear male-sterile locus, ms-10 . and Prx-2 on chromosome 2. Rick and Fobes (1974) reported tight

PAGE 16

linkage between Mi, a nematode resistance gene, and Aps-1 on chromosome 6. Prx-2 and Aps-1 have been used extensively in breeding programs as selectable markers for ms-10 and Mi , respectively (Rick and Fobes, 1974; Medina-Filho and Stevens, 1980; Tanksley et al . , 1984). Tanksley and LoaizaFigueroa (1985) mapped the Lycopersicon gametophytic selfincompatibility locus, S, to chromosome 1 linked to Prx-1 . This was the first report proving monogenic control of gametophytic self-incompatibility in plants. Quantitative traits have also been examined using isozymes. Tanksley et al. (1981) analyzed four guantitative traits (leaf shape, stigma exsertion, fruit weight, and seed weight) and 12 polymorphic isozyme loci in a backcross population. They detected a significant correlation between mean heterozygosity (i.e., the portion of heterozygous isozyme loci in each backcross individual) and measurements for each of the guantitative traits. They concluded that the isozyme genotypes could be used to predict the guantitative trait phenotypes. Tanksley et al. (1982) were able to map 21 loci controlling these traits using the 12 isozyme loci as markers. Vallejos and Tanksley (1983) examined the effect of the chromosome segments marked by 17 isozyme loci on cold tolerance derived from a high altitude L. hirsutum . They found that one marker, Pqi-l on chromosome 12, was linked to a locus that significantly improved growth at low temperatures.

PAGE 17

10 FUSARIUM WILT of TOMATO Taxonomy Fungi are taxonomically classified according to their sexual cycle, or perfect stage. Since no sexual cycle has been observed for Fusarium oxysporum . it is classified as a Deuteromycete , or Fungi Imperfecti, a group of about 15,000 species with no apparent perfect stage (Moore-Landecker, 1972) . F. oxysporum was initially classified by Wollenweber and Reinking (1935) based on the following four factors: 1. whether or not macroconidia were produced on sporodochia, 2. the color of stroma, 3. the presence or absence of sclerotia, and 4. the length, width, and number of septae in macroconidia. Initially, F. oxysporum was included in the Section Elegans which had three subsections, Orthocera, Constrictum, and Oxysporum (Wollenweber and Reinking, 1935) . In 1940, Snyder and Hansen reclassified members of the Section Elegans so that all belonged to one species, F. oxysporum . Individuals within the F. oxysporum species were subdivided into formae speciales based on pathogenicity to different hosts; Booth (1971) has classified 76 different formae speciales of F. oxysporum . Formae speciales were further subdivided into races based on varying pathogenicity to genotypes within a host species; race classifications were first proposed by the International Botanical Congress in 1935 (Ainsworth et al . , 1943). Three races of F.

PAGE 18

11 oxysporum f. sp. lycopersici , the Fusarium wilt pathogen of tomato, have been observed (Massee, 1895; Alexander and Tucker, 1945; Grattidge and O'Brien, 1982; Volin and Jones, 1982) . Anatomy. Growth, and Reproduction of the Fungus F. oxysporum reproduces asexually by means of spores. Three types of asexual spores, or conidia, are produced by the fungus including macroconidia, microconidia, and chlamydospores. Macroconidia are multiseptated, thin-walled spores with a pointed apical cell and a distinct foot cell; they are usually produced on branched conidiophores (specialized hyphae that give rise to conidia) in sporodochia on the surface of infected plant parts (Nelson, 1981) . Microconidia are unior di-septated spores found on short microconidiophores in the aerial mycelium; they are the predominant form involved in spread of the fungus through the plant. Chlamydospores are unior di-septated thick-walled conidia formed from macroconidia or in hyphae (Toussoun and Nelson, 1976) . They are produced in the final stages of wilt disease development, and represent a dormant stage of the fungus; chlamydospores have the ability to survive for years in soil or plant debris in the absence of a host (Nelson, 1981) . During asexual reproduction, macroconidia and microconidia are produced on conidiophores. Conidiophores are produced singly, either in sporodochia or in flat

PAGE 19

12 clusters called pionnotes (Puhalla and Bell, 1981) . The conidia are generated on flask-shaped terminal cells of conidiophores called phialides and remain dormant on mycelia until dispersed (Puhalla and Bell, 1981). Germination is inhibited by the action of volatile compounds produced by the fungus (Robinson and Garrett, 1969) . After dispersal, conidia can germinate under a wide variety of conditions. Water potential as low as -90 bars, a temperature range of 24° to 32 °C, a pH range of 4.0 to 9.0, or an atmosphere devoid of oxygen permit germination (Stotzky and Goos, 1965; Mozumder et al. , 1970; Ioannou et al. , 1977; Puhalla and Bell, 1981). The presence of essential nutrients increases germination rate and percentage (Puhalla and Bell, 1981). Macroconidia and chlamydospores have been shown to contain large quantities of lipids which appear to be utilized as a carbon source during germination (Marchant, 1966; Van Eck and Schippers, 1976) . Macroconidia germinate by forming one or two germ tubes, whereas microconidia and chlamydospores normally give rise to only one germ tube (Griffin, 1970) . Once germinated, fungal growth can either be vegetative as a mycelium or yeast-like; this trait is called dimorphism (Beckman et al. , 1953; Malca et al. , 1966). In the yeast form, conidia produce one or two phialides that directly give rise to more conidia with no germ tube being produced (Buckley et al. , 1969). Yeast-like growth appears to be

PAGE 20

13 important in pathogenicity; variants of Ceratocystis ulmi and Verticillium spp . that have lost their ability to grow as a yeast are avirulent (Tolmsoff , 1973) . Also, yeast-like growth reguires de novo protein synthesis; addition of pfluorophenylalanine, a protein synthesis inhibitor, repressed yeast-like growth of C. ulmi and F. oxysporum (Biehn, 1973). F. oxysporum mycelia are made up of septate hyphae that are usually uninucleate (Howard and Maxwell, 1975) ; these septa have central pores, but nuclei do not migrate through them (Typas and Heale, 1976) . Fungal metabolites called diffusible morphogenic factors (DMF) can inhibit hyphal elongation and induce lateral branching; succinic acid and bikaverin, the red pigment in hyphae and chlamydospores, have been shown to have DMF activity (Robinson et al. , 1969) . The lateral branches can become a part of the growing front, give rise to conidiophores, or anastomose with other hyphae (Hoffmann, 1966; Puhalla and Mayfield, 1974). Optimal Fungal Growth Conditions F. oxysporum is a soil borne fungus with greatest potential for infection at soil and air temperatures of approximately 28 °C (Clayton, 1923a) . Soil moisture levels that are optimal to the plant promote disease development; moisture levels that are too high or low inhibit the pathogen (Clayton, 1923b) . Short day length and low light

PAGE 21

14 intensity were also found to enhance disease development (Foster and Walker, 1947; Walker, 1971). The pathogen also prefers acid soil with an optimum pH range of 4.4 to 5.2 (Sherwood, 1923). Jones and Woltz (1969, 1970) found that liming soil to raise pH was effective in retarding disease development; the increase in pH limited the availability of micronutrients such as iron, manganese, and zinc to the pathogen. Bacteria and actinomycetes that compete for nutrients and produce antibiotics that can retard fungal development are also more prevalent at a higher pH (Marshall and Alexander, 1960) . Other nutrient studies have indicated that deficiencies in potassium and calcium, and increased levels of nitrogen and sulfur promote infection (Fisher, 1935; Walker and Foster, 1946; Edgington and Walker, 1958). Life Cycle Dormant F. oxysporum chlamydospores germinate when brought into contact with host roots or fresh plant debris (Stover, 1962; 1970). After host penetration, the fungus migrates to the vascular tissue, specifically the xylem (Nelson, 1981) . Once in the xylem, the pathogen spreads by means of mycelia or microconidia (Nelson, 1981) . Spread of hyphae laterally from primary to secondary xylem is through pit pairs, whereas upward movement is through the simple perforation plates of xylem vessel elements (Nelson, 1981) . During latter stages of disease infection, the fungus can move to phloem, cambium, pith, and cortex tissue (Nelson,

PAGE 22

15 1981) . After death of the host, sporodochia form on the surface of the plant and produce large numbers of macroconidia, as well as some microconidia (Baker, 1953; Phipps and Stipes, 1976) . Macroconidia frequently convert to chlamydospores and return to the soil on the host debris, completing the life cycle. F. oxysporum f. sp. lycopersici can also colonize several weed genera including Oryzopsis . Digitaria . Amaranthus , and Malva with no disease symptoms being evident; eggplants (Solanum melongena) can also be colonized with some stunting (Katan, 1971) . This type of nonhost colonization allows the fungus to survive longer periods of time in the absence of the host. In addition, Armstrong and Armstrong (1948) showed that F. oxysporum f. sp. batatas , the wilt pathogen of sweet potato (Ipomea batatas ) , can colonize on tomato without expression of symptoms . Host Defense Mechanism Successful colonization of Fusaria within the host is dependent on the ability of the pathogen to circumvent the defense mechanisms of the plant. Initially the soil-borne pathogen encounters a number of physical barriers associated with the plant root. Young roots are protected by a continuous epidermal layer. Within the epidermis are cortex, endodermis, and pericycle layers surrounding the vascular tissue. Older roots are surrounded by a heavily suberized corky tissue which arises from the pericycle

PAGE 23

16 (Talboys, 1958a) . When hyphae penetrate epidermis or cortex tissue, host cell walls are degraded. However, surrounding living cells lay down callose deposits called papillae that block further hyphal penetration (Bishop and Cooper, 1983) . Papillae eventually become lignified through infusion of secondary metabolites making them resistant to enzymatic and chemical degradation by the fungus. This response is similar in both resistant and susceptible genotypes (Bishop and Cooper, 1983) . Hyphae not stopped by the papillae are physically inhibited by the endodermis with suberized Casparian strips (Beckman, 1987) . Once hyphae reach the vascular tissue, the pathogen is localized in individual xylem vessels (Beckman, 1987) . Within the vessel, hyphae move laterally through pits into adjacent contact cells where they derive their nutrients. Pits are plugged with callose deposits to inhibit this movement; these deposits lignify and become papillae (Beckman et al . , 1982). Any invading organism, as well as wounding, can initiate this response in both resistant and susceptible genotypes (Beckman, 1987) . Longitudinal movement of hyphae is inhibited by vascular gels and tyloses (Wardlaw, 1930; Beckman and Halmos, 1962) . Gels are composed of pectins, calcium pectates, hemicellulose, and traces of protein (Beckman and Zaroogian, 1967) . In a resistant response, gels persist several days and are resistant to both physical and chemical

PAGE 24

17 degradation (Beckman et al. , 1974), whereas gels of susceptible genotypes become weakened and break under transpirational tension (Beckman et al. , 1962). Tyloses are outgrowths of vascular parenchyma contact cells through vessel pits (Wardlaw, 1930) . The cells grow into the xylem vessel, effectively barricading the fungus (Talboys, 1958b). If pits are too small to allow passage of enlarging contact cells, vascular elements are crushed; this also walls off the invading pathogen (Chattway, 1949) . Since plants have the ability to transport more water than necessary, loss of some xylem vessels is of little conseguence (Beckman, 1987) . Another aspect of the defense mechanism is the release of secondary metabolites including phenol ics and phytoalexins. Phenol ics are stored in a reduced state in specialized cells and are released in response to wounding or infection (Mace, 1963) . Once released, they are enzymatically oxidized and polymerize to form lignin which is incorporated into cell walls and tissues (Beckman, 1987) ; these tissues are impermeable to most invading organisms (Friend, 1976) . After localization of the pathogen, phytoalexins are mobilized to the site of infection; also, tyloses produce and release phytoalexins directly into the lumina of infected vessels (Hutson and Smith, 1980) . These phytoalexins inhibit growth of the pathogen. In tomato, chlorogenic acid, scopolin, tomatine, rishitin, and a number

PAGE 25

18 of flavonoids have been shown to have phytoalexin activity (Stoessl et al. , 1976) . Additional host defense responses include production of a protein that inhibits fungal polygalacturonase (Albersheim and Anderson, 1971) and production of /9-1,3-glucanase and chitinase fungal cell wall degrading enzymes (Pegg, 1976; Pegg and Young, 1981) . Both responses are common in resistant and susceptible host reactions. Genetics of Fusarium Wilt Resistance Fusarium wilt of tomato was first reported in England by Massee in 1895. Some early cultivars such as 'Marglobe* and 'Norton' carried moderate levels of genetic tolerance; however, freguently these cultivars would succumb to the disease under optimal environmental conditions (Richards, 1925; Harrison, 1941). Bohn and Tucker (1939, 1940) were the first to introduce monogenic resistance to Fusarium wilt derived from a L. pimpinellifolium accession called 'Red Currant 1 (U.S.D.A. Accession 160). This gene conferred complete immunity to the disease, and was designated I for "immunity". However, it has been observed that race 1 can occasionally infect plants carrying the I gene; this incomplete penetrance is influenced by genetic background, environmental conditions, and whether the plant is heterozygous or homozygous at the I locus (Stall, 1961; Alon et al. , 1974). In 1945, Alexander and Tucker observed a new race in Ohio, designated race 2, that consistently

PAGE 26

19 infected lines with 'Red Currant 1 resistance. Race 2 did not spread as rapidly as race 1. Monogenic resistance to this race, 1-2 . was first observed in 1955 (Alexander and Hoover, 1955) in another accession of L. pimpinellifolium (P.I. 126915). 'Walter' was the first cultivar introduced that incorporated resistances to both races 1 and 2 (Strobel et al. , 1969) . I and 1-2 have been mapped to chromosome 11, where 1-2 is tightly linked to the mps (miniature phosphorus syndrome) gene (Paddock, 1950; Latterot, 1976). Race 3, a new race with the ability to infect cultivars resistant to races 1 and 2, has been observed in Australia (Grattidge and O'Brien, 1982), Florida (Volin and Jones, 1982), and more recently in California (Davis et a_l. , 1988). Accessions of the wild species L. pennellii , P.I. 414773 (McGrath et al. , 1987) and LA 716 (Scott and Jones, 1989), have been shown to carry dominant monogenic resistance to race 3. Both reports have designated the resistance gene as 1-3 (McGrath et al. , 1987; Scott and Jones, 1989). Preliminary data indicate these two resistances are allelic; further work is needed to confirm this (JW Scott, personal communication, 1989) . L. pennellii accession LA 716 also carries resistance to races 1 and 2 (Scott and Jones, 1989) .

PAGE 27

CHAPTER 3 MAPPING OF NEW ENZYME LOCI INTRODUCTION Genetic markers have been used by plant breeders to expedite transfer of important genes from one genetic background to another (Rick and Fobes, 1974; Weeden et al. , 1984) . There are three types of markers commonly used: morphological, restriction fragment length polymorphisms (RFLP's), and isozymes. Morphological markers have a number of potential disadvantages: 1. they can be deleterious to the overall plant phenotype; 2. environmental conditions can influence marker phenotypes ; 3 . frequently they are not identifiable until late in the plant life cycle; and 4. they are usually only expressed in the homozygous recessive condition. RFLP use is limited due to time requirements, expense, and the necessity of probes. Isozymes have the following advantages as markers: 1. they have a neutral effect on the overall plant phenotype; 2. isozyme genotypes are not influenced by the environment; 3. sampling for the isozyme analysis is nondestructive; 4. isozymes can be identified very early in the plant life cycle; 5. they are codominantly expressed; and 6. they are relatively inexpensive and safe to analyze (Tanksley and Rick, 1980b) . 20

PAGE 28

21 However, use of isozyme markers is limited to the availability of polymorphism and the relatively small number of loci that have been mapped. For a recent report on the map locations of enzyme loci in the genus Lvcopersicon . see Chetelat (1989). In this report, the map locations and tissue specificities of eight additional enzyme loci dimorphic betweeen L. esculentum and L. pennellii (LA 716) were identified by conducting linkage analyses to 15 previously mapped isozyme loci. The new loci included Diaphorase-2 ( Dia-2 ) , ff-N-acetyl-qlucosaminidase-l ( Bnaq-l ) , Mannose-6phosphate isomerase-1 ( Mpi-1 ) , Sucrose svnthase-1 ( Ss-1 ) , Esterase-9 ( Est-9 ) , Esterase-10 ( Est-10 ) , Hexokinase-1 ( Hk1) , and a-Mannosidase-1 ( Aman-1 ) . These loci were used in chapters 4 and 5 as potential markers of gene(s) for resistance to races 1, 2, and 3 of Fusarium oxysporum f. sp. lycopersici derived from L. pennellii (LA 716) . MATERIALS and METHODS To examine segregation and linkage analyses of the new enzyme loci, the following backcross was generated: L. esculentum cv. Bonny Best x [Bonny Best x L. pennellii (LA 716) ] . Isozyme procedures have been described previously (Rick et al . , 1977; Tanksley and Rick, 1980b). The mannose6-phosphate isomerase stain included the following components: tris-HCl (Sigma Trizma base) 0.1M, pH 7.5; MgCl2*6H 2 20mM; mannose-6-phosphate 0.2 mg/ml ; NADP + 0.15

PAGE 29

22 mg/ml; MTT 0.2 mg/ml; phenazine methosulphate (PMS) 0.04 mg/ml; phosphoglucose isomerase 0.5 units/ml; and glucose-6phosphate dehydrogenase 0.3 units/ml. The sucrose synthase stain involved a triple enzyme coupling mechanism with the following reagents: tris-HCl 0.1M, pH 7.5; MgCl 2 "6H 2 20mM; sucrose lOOmM; ADP 0.2 mg/ml; ATP 1.3 mg/ml; hexokinase 1 units/ml; phosphoglucose isomerase 0.5 units/ml; glucose-6phosphate dehydrogenase 0.3 units/ml, NADP+ 0.15 mg/ml; MTT 0.2 mg/ml; and PMS 0.04 mg/ml. Staining for both systems was done at room temperature in the dark. a-Mannosidase was stained using a filter paper overlay with the following protocol: 2mg of 4-methylumbelliferyl-a-mannopyrannoside was dissolved in 1ml of dimethyl sulfoxide; this was added to 5ml of 0.1M acetic acid pH 4 . 5 (adjust pH with NaOH) . Bands were observed using an ultraviolet light box. Other staining systems have been described previously (Vallejos, 1983; Weeden, 1984, 1986). The following gel systems were used: histidine (gel: histidine'HCl, 5mM, pH 7.0, adjusted with NaOH; electrode: tris, 135mM/citric acid 43mM, pH 7.0, adjusted with tris or citrate) for Ss-1 , Bnaq-1 . and Aman-1 ; tris-EDTA-borate (Heath-Pagliuso et al. , 1984)) for Dia-2 and Bnaq-1 ; and tris citrate (Tanksley and Rick, 1980b) for Mpi-1 , Hk-1 . Est-9 . and Est-10 . All samples were collected simultaneously from the same •Bonny Best 1 and LA 716 plants for the tissue specificity analysis; tissues included the fol lowing: leaf, petiole,

PAGE 30

23 stem, pedicel, peduncle, flower, immature fruit, and root. The segregation for the new and 15 previously mapped enzyme loci was recorded for all backcross individuals; the latter 15 loci included: Prx-1 and Skdh-1 on chromosome 1; Est-7 and Prx-2 on chromosome 2 ; Prx-7 on chromosome 3 ; Pqm-2 on chromosome 4 ; Aps-1 on chromosome 6 ; Got-2 on chromosome 7 ; Aps-2 on chromosome 8; Prx-4 on chromosome 10; Sod-1 on chromosome 11; and Est-4 . 6Pqdh-2 , Pgi-1 , and Aco-1 on chromosome 12. The morphological marker, Pn on chromosome 8, was also scored. Tests for independent assortment were conducted using two-way contingency tables on a main frame computer using the Proc Freg program of the Statistical Analysis System (SAS Institute Inc., SAS Circle, Box 8000, Cary, NC 27511. RESULTS and DISCUSSION Electrophoretic migration distances of the eight new enzyme loci are shown in Table 3-1. One zone of activity was detected for AMAN, BNAG, and MPI indicating the presence of a single locus for these enzymes. There were at least two zones of activity associated with HK, the most anodal of which was assigned as the Hk-1 locus. Four zones of DIA activity appeared in the tris-EDTA-borate gel system. The large zone of activity that migrated between Dia-1 and Dia3, two loci previously mapped to chromosomes 5 and 9 respectively (Chetelat, 1989) , has been designated Dia-2 (Table 3-1) ; the slowest migrating zone of DIA activity was

PAGE 31

24 not segregating in this cross and has not been assigned a chromosomal location. At least five zones of activity were observed in the EST leaf zymogram. The most anodal two loci, Est-2 and Est3 have been previously mapped to chromosomes 9 and 1, respectively (Tanksley and Rick, 1980a) . The slowest migrating zone of activity, Est-8 . was mapped to chromosome 10 (Tanksley et al. , 1988). Two additional zones of activity appeared between Est-3 and Est-8 (Table 3-1) . These two zones assorted independently and have been assigned as Est-9 and Est-10 with Est-9 being anodal to Est10. Six zones of activity appear after staining leaf tissue for sucrose synthase. However, not all of these bands represent sucrose synthase activity. Seguential removal of coupling enzymes allowed for the determination of the origin of most bands. The slowest migrating zone of activity was the only zone that did not appear when coupling enzymes were removed. This zone of activity has been designated Ss-1 . Even Ss-1 is only tentatively assigned as sucrose synthase. This locus was specific to green tissues (Table 3-1) ; however, a western blot analysis of tomato root and leaf extracts using maize sucrose synthase antibody only detected two root specific loci. No protein associated with sucrose synthase was detected in the leaf sample. Since the sucrose synthase stain involves a three enzyme coupling

PAGE 32

25 reaction, it is possible that an intermediate produced in one of the reactions may have provided the necessary substrate for a different enzyme to produce the bands observed for Ss-1 . Table 3-1 also shows the tissue specificities of each of the new enzyme loci. For all loci, some level of activity was present in all green tissues analyzed; in addition, activity in the roots was detected for all loci except Est-9 , Est-10 , and Ss-1 . Relatively equal levels of activity were observed for Aman-1 , Bnag-1 , Dia-2 , and Est-10 in all tissues where these loci were expressed (Table 3-1) . Loci which showed an increased level of activity in a specific tissue included: Est-9 and Mpi-1 in pedicel and flower tissues, Hk-1 in stem and peduncle tissues, and Ss-1 in flower tissue (Table 3-1) . Tests of independent assortment between each new locus and all segregating loci (the previously mapped loci as well as the other new loci) were conducted to determine the chromosome location of the eight enzyme loci (Table 3-2) . Tables 3-3, 3-4, and 3-5 show contingency tables of loci pair combinations where significant deviations from expected ratios were observed. Linkage was detected between Bnag-1 . Dia-2 , and Est-9 to the Skdh-1 locus known to be on chromosome 1 (Tanksley and Rick, 1980b, Table 3-3) . Thus, these loci were assigned to chromosome 1, and the following

PAGE 33

26 gene order was deduced by three point mapping using Prx-1 and Skdh-1 as reference points (Tables 3-6 and 3-7) : Prx-1 — 20cM — Skdh-1 , Bnaq-1 — 29cM — Dia-2 — 6cM — Est-9 . Due to tight linkage, further testing is necessary to determine the side of Skdh-1 on which Bnag-l lies. Mpi-1 was assigned to chromosome 2 due to its linkage to Est-7 and Prx-2 , two loci previously mapped to this chromosome (Rick and Fobes, 1977; Tanksley and Rick, 1980b, Table 3-4) . Three point mapping between these three loci could not definitely discern a gene order with the available data (data not shown) . Linkage was also detected between Ss-1 and Prx-7 (27cM) known to be on chromosome 3 (Tanksley and Rick, 1980b) , Aman-1 and Aps-1 (31cM) previously assigned to chromosome 6 (Rick and Fobes, 1974) , and Est-10 and Sod-1 (3cM) previously reported on chromosome 11 (Tanksley et al . , 1988, Table 3-5). No linkage was detected between Hk-1 and any of the segregating marker loci. Analysis of monogenic segregation ratios of the marker loci in the BC^ population showed that most conformed to the expected 1:1 ratio (Table 3-8). However, a significant excess of heterozygotes was detected on chromosome 1 at Prx1, Dia-2 , and Est-9 . on chromosome 8 at Pn, on chromosome 11 at Est-10 . and on chromosome 12 at 6Pgdh-2 and Pqi-1 ; Got-2 on chromosome 7 had significantly more homozygotes than expected (Table 3-8) . It is likely that loci linked to

PAGE 34

27 these markers contributed to the ability of the pollen grain to fertilize the female gametophyte. Skewed ratios at loci known to be monogenic have been reported previously in interspecific Lycopersicon backcrosses (Rick, 1969; Tanksley et al. , 1982) . Most of the new enzyme loci were not tightly linked to a previously mapped marker locus. In future gene mapping studies, including the tagging of genes for resistance to Fusarium oxysporum f. sp. lvcopersici , use of these new loci will effectively increase the portion of genome covered by enzyme markers.

PAGE 35

28 Table 3-1. Migration distances and tissue specificities of eight new enzyme loci. Migration distances are a measure of the distance that the zone of activity associated with the locus migrated from the origin toward the anode (+) end of the gel. Tissue sample analysis shows a comparison of the relative levels of activity of the tissue samples for each locus. Samples were taken from the same plants at the same time; the fruit sample was at the immature green stage. ee L. esculentum homozygotes ; ep L. pennellii heterozygotes. 3 Migration Tissue sample ^ 1.8 2.4 ++ ++ ++ ++ ++ ++ ++ ++ 5.5 C 6.3 + ++ ++ ++ ++ ++ ++ + 7.5 d 7.8 + ++ ++ ++ +++ +++ ++ 6.5 d 6.9 ++ ++ ++ ++ ++ + + 9.5 9.3 + ++ +++ +++ ++ + ++ + 9.0 8.7 ++ ++ + ++ +++ +++ ++ + 1.7 1.2 + + + + + +++ ++ distance (cm) Petiole Peduncle Flower Root Locus ee ep Leaf Stem Pedicel Fruit Aman-1 1.0 1.6 + ++ ++ ++ ++ ++++++ Bnag-1 Dia-2 Est-9 Est-10 Hk-1 Mpi-1 Ss-1 a Table only represents comparisons of activity within each locus. Comparisons between loci can not be made (e.g. Aman1 and Dia-2 leaf activities are not necessarily equal) . "(-) no activity, (+) low activity, (++) intermediate activity, (+++) greater activity. c Dia-l . Dia-3 , and Dia-4 had migration distances of 7.0, 4.0, and 2.0, respectively for both L. esculentum and L. pennellii . "Leaf esterase loci, Est-2 , Est-3 , and Est-8 , had migration distances of 9.1, 8.3, and 5.0, respectively for L. esculentum.

PAGE 36

29 Table 3-2. Chromosomal location of new enzyme loci. Independent assortment analyses comparing each new locus with previously mapped loci as well as the other new loci were conducted using two-way contingency tables. Numbers represent X 2 values in each analysis. Ch romosome New loci Locus Aman-1 BnacrL Dia-2 Est-9 Est-10 Hk-1 Mpi-1 Ss-1 Prx-1 1 NS 25.3** 6.7* NS NS NS NS NS Skdh-1 1 NS 73.1** 24.7** 19.4*" NS NS NS NS Est-7 2 NS NS NS NS NS NS 8.79* NS Prx-2 2 NS NS NS NS NS NS 7.31* NS Prx-7 3 NS NS NS NS NS NS NS 24.2 Pqm-2 4 NS NS NS NS NS NS NS NS Aps-1 6 8.4* NS NS NS NS NS NS NS Got-2 7 NS NS NS NS NS NS NS NS Aps-2 8 NS NS NS NS NS NS NS NS Pn 8 NS NS NS NS NS NS NS NS Prx-4 10 NS NS NS NS NS NS NS NS Sod-1 11 NS NS NS NS 86.1** NS NS NS Est-4 12 NS NS NS NS NS NS NS NS 6Padh-2 12 NS NS NS NS NS NS NS NS Pqi-1 12 NS NS NS NS NS NS NS NS Aco-1 12 NS NS NS NS NS NS NS NS Aman-1 — NS NS NS NS NS NS NS Bnaq-1 — NS — 14.1** NS NS NS NS NS Dia-2 NS 14.1** -52.9** NS NS NS NS Est-9 NS NS 52.9** — NS NS NS NS Est-10 NS NS NS NS — NS NS NS Hk-1 NS NS NS NS NS — NS NS Mci-1 NS NS NS NS NS NS — NS Ss-1 ~ NS NS NS NS NS NS NS ~~ ** *, **Significant at the 0.01 and 0.001 levels, respectively.

PAGE 37

Table 3 -3. Independent assortment analys is of chromosome 1 isozyme loci using two-way contingency tables. ee L. esculentum homozygotes; ep L. oennellii heterozygotes; p percent recombination . a Locus Locus Prx-1 Skdh-1 ee ep ee ep Bnaq-1 Dia-2 Est-9 ee ep ee ep ee ep ee 46 6 25 4 25 24 12 20 Prx-1 ep 19 57 13 35 20 52 11 30 X 2 =49.8** X 2 =25.3** X 2 =6.7* p=19.5% p=9.1% p=36.4% X 2 =NS ee 38 36 26 21 18 Skdh-1 ep 1 38 8 49 2 32 X 2 =73.1** X 2 =24.7** X 2 =19.4** p=1.3% p=28.6% p=27.4% ee 18 18 9 10 Bnaq-1 ep 3 31 2 16 X 2 =14.1** X 2 =NS p=30.0% 30 ee 20 3 Dia-2 ep 1 46 X 2 =52.9** p=5.7% *, **Signif icant at the 0.01 and 0.001 levels, respectively. a Deduced gene order: Prx-1 — 20cM— Skdh-1 , Bnaq-1 — 29cM — Dia-2 — 6cM— Est-9

PAGE 38

31 Table 3-4. Independent assortment analysis of chromosome 2 isozyme loci, ee L. esculentum homozygotes ; ep L. pennellii heterozygotes; p percent recombination. Est-7 Prx-2 Est-7 ee ep ee ep ep Prx-2 ee ep 50 5 4 61 X 2= =86.5 P= =7 . 5% ** MPJ-1 ee ep 21 11 13 29 X 2= P= =8.8* =32.4% 19 9 13 25 X 2= =7 .3* p=33.3% *, **Significant at the 0.01 and 0.001 levels, respectively.

PAGE 39

32 Table 3-5. Independent assortment analyses using two-way contingency tables for the following linked loci: Ss-1 and Prx-7 on chromosome 3 , Aman-l and Aps-1 on chromosome 6 , and Est-10 and Sod-1 on chromsome 11. ee L. esculentum homozygotes; ep L. pennellii heterozygotes ; p percent recombination . Ssee -1 ep Aman-l ee ep ee 40 17 ee 17 11 Prx-7 Aps•1 ep 12 40 ep 5 19 X 2 =24.2*" p=26.6% r X 2 =8.42* p=30.8% Estee -10 ep ee 39 2 Sod-1 ep 1 X 2 =86. 56 1** p=3.1* *, **Signif icant at the 0.01 and 0.001 levels, respectively.

PAGE 40

33 Table 3-6. Deduced chromosomal location of Dia-2 with respect to Prx-1 and Skdh-1 on chromosome 1 using threepoint mapping techniques. Prx-1 and Skdh-1 were mapped previously (Tanksley and Rick, 1980) . Table shows segregation ratios and the number of double recombinants for each putative gene order. ee L. esculentum homozygotes ; ep L. pennellii heterozygotes. Prx-l/Skdh-1 Putative Double ee/ee ee/ep ep/ee ep/ep gene order Recombs ee 24 1 12 7 Prx-l/Skdh-l/Dia-2 a 6 Dia-2 Prx-l/Dia-2 /Skdh-1 27 ep 20 3 5 46 Dia-2 /Prx-l/Skdh-1 15 a Correct gene order because it showed the least number of double recombinants.

PAGE 41

34 Table 3-7. Deduced chromosomal location of Est-9 with respect to Skdh-1 and Dia-2 on chromosome 1 using threepoint mapping techniques. Est-9 and Prx-1 were not linked (Table 3-2) . Skdh-1 was mapped previously, and the location of Dia-2 was determined in Table 3-5. This table shows segregation ratios and the number of double recombinants for each putative gene order, ee L. esculentum homozygotes ; ep L. pennellii heterozygotes. Skdh-l/Dia-2 ee/ee ee/ep ep/ee ep/ep Putative Double gene order Recombs ee 19 Est-9 ep 15 1 Skdh-l/Dia-2/Est-9 a Skdh-l/Est-9/Dia-2 4 30 Est-9/Skdh-l/Dia-2 16 a Correct gene order as judged by the lack of double recombinants .

PAGE 42

35 Table 3-8. Analysis of monogenic segregations of marker loci. ee L. esculentum homozygotes; ep L. pennellii heterozygotes . Locus Chromosome ee ep Monogenic X 2 Prx-1 Skdh-1 Bnag-1 Dia-2 Est-9 Est-7 Prx-2 Mpi-1 Prx-7 Ss-1 Pqm-2 Aps-1 Aman-1 Got-2 Aps-2 Pn Prx-4 Sod-1 Est-10 Est-4 6Pqdh-2 Pqi-1 Aco-1 Hk-1 1 53 80 5.48 1 66 63 NS 1 39 39 NS 1 45 77 8.39 1 23 51 10.59 2 62 72 NS 2 54 66 NS 2 3 34 64 40 57 NS NS 3 4 6 61 62 69 59 70 55 NS NS NS 6 23 32 NS 7 81 50 7.34 8 74 55 NS 8 26 64 16.04 10 66 63 NS 11 43 57 NS 11 49 79 7.03 12 12 12 43 HO 42 44 59 88 89 NS 16.28 15.22 12 57 73 NS -> 44 53 NS ** ** ** ** ** ** ** *, **Significant at the 0.05 and 0.01 levels, respectively.

PAGE 43

CHAPTER 4 ANALYSIS AND MAPPING OF GENES CONTROLLING RESISTANCE TO RACE 3 OF FUSARIUM WILT INTRODUCTION Races 1 and 2 of Fusarium oxysporum f. sp. lycopersici (Sacc.) Snyder and Hansen which causes Fusarium wilt of tomato are widespread in most production regions. Single dominant genes I and 1-2 , which control resistances to races 1 and 2, respectively, have been derived from accessions of Lycopersicon pimpinellifolium (Bohn and Tucker, 1939; 1940; Alexander and Hoover, 1955; Stall and Walter, 1965) . These genes have been mapped to chromosome 11 (Paddock, 1950; Latterot, 1976), and have been incorporated into most cultivars. Race 3, a new race with the ability to infect cultivars resistant to races 1 and 2, has been reported in Australia (Grattidge and O'Brien, 1982), Florida (Volin and Jones, 1982) , and more recently in California (Davis et al. , 1988) . The wild species L. pennellii represents a potential source of resistance for new cultivars. This species has been reported to carry monogenic dominant resistance to race 3 in accessions PI414773 (McGrath et al . , 1987) and LA 716 (Scott and Jones, 1989) . 36

PAGE 44

37 Tanksley and Rick (1980b) proposed the use of isozymes as scorable markers to facilitate the efficient transfer of genes from a wild species into domesticated cultivars. Rick and Fobes (1974) were the first to report tight linkage between the enzyme locus Aps-1 , and Mi, a gene responsible for nematode resistance in tomato. This linkage has had a major impact on tomato breeding programs; the Aps-1 locus is now widely used as a scorable marker for nematode resistance (Medina-Filho and Stevens, 1980) . In pea, the enzyme locus Pgm-p can be used as a marker for Mo, a gene which controls for resistance to bean yellow mosaic virus (Weeden et al . , 1984) . The objective of this portion of the project was to find a molecular marker for resistance to race 3 of F. oxysporum f. sp. lycopersici detected in L. pennellii (LA 716) . Two independent approaches were taken. First, tests for independent assortment were conducted between race 3 resistance and each of 2 5 previously mapped enzyme loci in an interspecific backcross. Second, isozyme analysis was conducted on five breeding lines selected for race 3 resistance derived from LA 716 to test whether any LA 716 alleles had been co-selected. MATERIALS and METHODS Plant Materials The following interspecific backcross was generated for segregation and linkage analyses: [L. esculentum x (L.

PAGE 45

38 esculentum x L. pennellii . LA 716)]. 'Bonny Best 1 was used as the L. esculentum parent due to its extreme susceptibility to race 3. In addition, five breeding lines that had been independently produced using conventional breeding technigues were analyzed [E427 (BC 1 S 3 ) , I3R-1 (BC 2 S 2 ), 520 (BC 2 S 2 ), 572 (BC3S!) , and 604 (BC 1 S 2 ) ] . All lines included the initial cross 'Hayslip' x LA 716. With the exception of LA 716, all lines and cultivars included in the pedigrees of the breeding lines were race 3 susceptible. The cultivars 'Manapal' (resistant to race 1) and 'Hayslip' (resistant to races 1 and 2) were also used as controls in race 3 inoculations. In addition, five L. esculentum lines carrying L. pennellii (LA 716) chromosome substitutions for chromosome 2, 4, 6, 8, and 11 (Rick, 1969) were inoculated. Isozyme Analysis The starch gel electrophoresis and enzyme staining techniques employed have been described elsewhere (Rick et al. . 1977; Tanksley and Rick, 1980b; Vallejos, 1983). The genotypes of 79 BC^ individuals and the race 3 resistant breeding lines were determined at each of the following segregating enzyme loci: Prx-1 , Skdh-1 . Bnaq-l , Dia-2 , and Est-9 on chromosome 1; Mpi-1 , Est-7 , Prx-2 on chromosome 2; Prx-7 and Ss-1 on chromosome 3 ; Pgm-2 on chromosome 4 ; Aps-1 and Aman-1 on chromosome 6 ; Got-2 on chromosome 7 ; Aps-2 on chromosome 8; Prx-4 on chromosome 10; Sod-1 and Est-10 on chromosome 11; Est-4 . 6Pqdh-2 , Pqi-1 . and Aco-1 on

PAGE 46

39 chromosome 12 ; and Hk-1 which has not been assigned a chromosomal location. The morphological marker, Pn, on chromosome 8 was also scored. Following isozyme analysis, the BC^ individuals were raised to maturity and selfpollinated. The BC^Si progenies, along with the breeding lines, were then inoculated with race 3. Inoculations Inoculum was prepared by incubating spores of race 3 of F. oxysporum f. sp. lycopersici (SC761 isolate 1 ) in sterilized 100% potato dextrose broth (Difco) at 28° C with continuous shaking at 125 rpm for 3-4 days. After incubation, the culture was filtered through four layers of cheesecloth to remove mycelia and diluted to a concentration of 1.10 1.25 X 10 8 spores/ml; a hemacytometer was used to determine spore concentration. The phenotype of each backcross individual with respect to resistance to race 3 was determined through progeny tests; at least 20 BC^^Sj^ progeny were tested per BC^ analysis. Progeny tests were used to increase reliability of BC^ scores and to allow for the eventual determination of BCq^ responses to races 1 and 2. 'Bonny Best', LA 716, and Fj seedlings were inoculated as controls with every set of BC^Sq^ families. Also, another set of controls was incubated in distilled water only. 1 SC761 is a race 3 strain isolated in 1982 by Dr. J. P. Jones at the University of Florida, Gulf Coast Research and Education Center, at Bradenton, FL.

PAGE 47

40 Seedlings of 'Manapal' and 'Hayslip 1 were periodically inoculated to monitor the specificity of race 3 inoculum. Seedlings at the 2-4 true leaf stage were inoculated by pruning roots and incubating in inoculum for 3-5 hours at 28° C. They were then transplanted into a sterilized peat: sand: soil (1:1:1) medium (pH 6.0) four cm apart and trans fered to a growth chamber. Growth conditions were as follows: 28° C constant temperature, 80% ±20% humidity, and a lOh photoperiod. The photon flux density (umol quanta -m" 2 -sec -1 ) in the chamber changed according to the following schedule: 90 for 2h, 130 for 2h, 200 for 2h, 130 for 2h, 90 for 2h, and darkness for 14h. Fertilizer (Peters 20-20-20) was applied approximately one week after inoculation. Disease severity was estimated at least twice. The first estimation was carried out shortly after susceptible 'Bonny Best' plants began showing wilt symptoms, usually 10-14 days after inoculation. The purpose of this evaluation was to determine the cause of death of dying plants. Stems of dying plants were cut and examined for vascular browning; living plants were judged as wilting, stunted, or healthy. A second estimation of disease severity was carried out approximately one month after inoculation; for this evaluation, the degree of vascular browning was determined by making a series of thin longitudinal cuts from the first true leaf down to the root tip with a razor blade. Scores were assigned according to the following scale: 1 = no sign

PAGE 48

41 of browning, or browning only at root tip, 2 = at least some browning found above the root tip but not continuous from root to cotyledon, 3 = continuous browning from root to cotyledon in only one vascular tract, and 4 = continuous browning from root to cotyledon in more than one vascular tract. Dying plants showing external wilt symptoms were given a rating of 5. Plants rated from 1-4 had no external wilt symptoms. A weighted mean disease rating was calculated for each backcross individual using the progeny scores. All statistical analyses were performed on a main frame computer using the Statistical Analysis System (SAS Institute Inc., SAS Circle, Box 8000, Cary, NC 27511). The following programs were used: Proc Univariate, Proc Freq, Proc Cluster, Proc Nparlway, Proc Glm, and Proc Sort. RESULTS and DISCUSSION In all inoculations, seedlings of L. pennellii (LA 716) and the interspecific F-± consistently showed resistance to race 3; whereas, seedlings of the L. esculentum cultivars, 'Bonny Best*, 'Manapal', and 'Hayslip', showed susceptibility. Responses to race 3 for BC;l individuals were deduced from mean disease ratings of their progenies. Analysis of the frequency distribution of the mean disease ratings (Fig. 4-1) using the Kolomogorov D statistic (Kolomogorov, 1933) indicated a significant deviation from normality (D=0.128, a<0.01). In this procedure, the frequency distribution is compared to a normal distribution

PAGE 49

42 generated using the mean and variance of the BC^ disease ratings; the test statistic, D, represents the total deviation between the two distributions (Kolomogorov, 1933) . Furthermore, the distribution proved to be bimodal according to the coefficient of bimodality (b = (m 3 2 +l)/ (m 4 +3) ; m 3 = skewness, m 4 = kurtosis; b = 0.586, a < 0.05). Skewness and kurtosis describe types of nonnormal distributions. A distribution shows skewness if a significant excess of samples appear on one side of the mean (Snedecor and Cochran, 1980) . Distributions with long tails have a positive kurtosis, whereas a flat-topped distribution has a negative kurtosis (Snedecor and Cochran, 1980) . This bimodal distribution indicated that race 3 resistance was a trait controlled by one major locus which was designated 1-3 by Scott and Jones (1989) . The possible association between race 3 resistance and a segregating marker locus was analyzed using a one-tailed nonparametric analysis of variance — the Wilcoxon rank sum test (Wilcoxon, 1945) . This test is similar to the parametric two-sample t-test. The nonparametric analysis was implemented because the disease ratings had a limited range (1 to 5) and were based on a qualitative evaluation of disease symptoms. In this procedure, BC;l individuals were first ranked according to their mean disease rating; then, the mean of the rankings of all homozygous plants was compared to that of the heterozygotes at each marker locus.

PAGE 50

43 Highly significant differences were detected for the chromosome segment marked by Got-2 . BC^ individuals carrying L. pennellii alleles for Got-2 were significantly more resistant than those homozygous for L. esculentum (Table 4-1) . These results clearly demonstrated linkage between Got-2 and 1-3 . An additional significant effect was also detected for the chromosome segment marked by Aps-2 , but the effect of this locus was not as great as the one observed for 1-3 (Table 4-1) . The presence of this minor factor was supported by the results obtained from inoculations of five L. pennellii (LA 716) chromosome substitution lines developed by Rick (1969) . The chromosome 8 line (carrying Aps-2 ) had a mean disease rating of only 3.00 compared to 4.62, 4.38, 4.85, and 4.84 for lines of chromosome 2, 4, 6, and 11, respectively. Other substitution lines were unavailable. Since the chromosome 8 substitution line showed tolerance to race 3, the locus linked to Aps-2 is being designated Tfw for "tolerance to Fusarium wilt." It was also noted that heterozygotes at the Pgm-2 locus were significantly more susceptible than homozygotes (Table 4-1) . Such transgressive behavior has previously been reported in tomato (Tanksley et al . , 1982; Vallejos and Tanksley, 1983) . Nevertheless, upon further investigation, this observation proved to be an artifact (see below) .

PAGE 51

44 The percent recombination between Got-2 and 1-3 was estimated to evaluate the usefulness of Got-2 as a marker for resistance. Although the bimodal distribution of mean disease ratings and the nonparametric analyses of variance indicated that race 3 resistance was controlled by one major locus, it was difficult to classify some BC]^ individuals for their 1-3 genotype. To overcome this difficulty, the average linkage clustering method (Sokal and Michener, 1958) was used to classify all BC]^ individuals according to their mean disease ratings into two clusters: resistant and susceptible. This method sequentially clusters individuals, or groups of individuals, with the least distance 2 until the desired number of clusters is attained — two in this case. The genotype of each BC^ individual at the 1-3 locus was inferred according to its cluster classification: susceptible ( i-3/i-3 ; x=4.01 ±0.63) or resistant ( I-3/J-3 ; x=2.11 ±0.32). A test for independent assortment between Got-2 and 1-3 using a two-way contingency table showed significant deviations from expected ratios; the recombination between the two loci was estimated at 2.5% (Table 4-2) . Fig. 4-2 shows a GOT starch gel including the Got-2 phenotypes of 'Bonny Best', LA 716, and their F^. The evidence presented above indicated that race 3 resistance was controlled by one major gene, 1-3 . However, 2 °KL = ( X K " *l) 2 + S K 2 + S L 2 • D KL = distance between clusters K and L. x K , x L and s K ^, s L 2 means and variances of clusters K and L, respectively,

PAGE 52

45 the variability observed within clusters (Fig. 4-1) suggested the presence of minor factors such as Tfw. To detect additional minor factors that may have been overshadowed by 1-3 . two-way factorial (2x2) analyses of variance were conducted for all I-3/ marker locus pair combinations. The results of these analyses (Table 4-3) showed again that Tfw, linked to Aps-2 , had a significant effect on the response to race 3. Furthermore, Tfw appeared to act independently from 1-3 . as suggested by the lack of a significant interaction (Table 4-3) . Also, a significant interaction was detected between 1-3 and the chromosome segment marked by Dia-2 (Table 4-3) . However, one-way analyses of variance showed no significant differences between mean disease rankings of Dia-2 homozygotes and heterozygotes within each 1-3 genotype (data not shown) ; thus, the I-3/Dia-2 interaction appeared to be an artifact. Additional factors affecting resistance may have existed in parts of the genome not covered by the marker loci. The factorial analysis also indicated that the chromosome segment marked by Pgm-2 had no significant effect on resistance (Table 4-3) . A two-way contingency table between Perm2 and 1-3 (Table 4-4) showed a slight trend to skew in favor of recombinants (64%) . This resulted in an excess of Perm2 homozygotes in the I-3/J-3 class, and an excess of heterozygotes in the i-3/i-3 class. Thus, the cause of the significant difference in mean disease rankings

PAGE 53

46 observed at the Perm2 locus (Table 4-1) could be attributed to the slight skewing. Finally, a two-way analysis of variance of mean disease rankings for all possible pair combinations between marker loci did not show any significant interactions. Additional evidence for the presence of linkage between Got-2 and 1-3 was obtained through an analysis of breeding lines previously selected for resistance to race 3. The genotype at all segregating marker loci, and the response to race 3 was determined for each line. Three lines, I3R-1, 52 0, and 604 were uniformly resistant to race 3 and homozygous for the L. pennellii allele at Got-2 ; whereas, E427 and 572 were segregating for both resistance and Got-2 . Thus, there had been no recombination between Got-2 and 1-3 in any of the five lines. The lines were homozygous for L. esculentum alleles at the other marker loci with the following exceptions: E427 was segregating at Prx-1 , Aps-1 , Prx-4 . and Aco-1 , and 520 was homozygous for the L. pennellii allele at Aco-1 . E427 is a breeding line known to be fixed for resistance to race 3. However, the E427 seed used in this study originated from a seed lot suspected of contamination by cross pollination which occurred in the field during seed increase. Currently, race 3 of the Fusarium wilt pathogen is primarily confined to Florida, Australia, and parts of California; however, since races 1 and 2 have a worldwide

PAGE 54

47 distribution, it is probable that race 3 will eventually become a problem in other tomato growing regions. The results of this work demonstrate that Got-2 can be used as a scorable marker to facilitate the transfer of 1-3 into commercial cultivars. Using the 2.5% recombination estimate, only two plants (N=2) would need to be selected using Got-2 as a marker to maintain a 99.9% chance 3 of coselecting 1-3 . It should be noted that only one isolate of race 3 was used as a source of inoculum in this study. However, breeding lines carrying 1-3 have also maintained resistance against the race 3 isolates from Australia (D.J. McGrath, personal communication, 1988) and California (M. Kuehn, personal communication, 1988) . 3 P = l-r N ; N = ln(l-P)/ln(r) . P = probability, r = percent recombination, N = no. of plants.

PAGE 55

48 Table 4-1. Analyses of monogenic segregations of marker loci and comparisons of BC-^ mean disease ratings using a one-way nonparametric analysis of variance. Mean disease ratings were calculated for BCj individuals based on response to race 3 of their progeny. To perform the one-way nonparametric analysis of variance, the Wilcoxon rank sum test was used. In this procedure, BC^ individuals were ranked according to their mean disease rating, and homozygote and heterozygote mean disease rankings were compared at each marker locus. ee L. esculentum homozygotes; ep L. pennellii heterozygotes . Mean Mean Wilcoxon Chromo No. Disease Disease rank sum some Plants Ratina 3 Rank incr test Locus No. ee ep ee ep ee ep Z stat Prx-1 1 27 51 3.23 3.24 38.8 39.9 NS Skdh-1 1 31 44 3.34 3.06 41.3 35.6 NS Bnacr-1 1 22 30 3.35 3.08 28.5 25.0 NS Dia-2 1 23 48 3.33 3.14 34.7 38.6 NS Est-9 1 11 29 3.38 3.16 21.6 20.1 NS Mpi-1 2 24 21 2.96 3.27 21.4 24.8 NS Est-7 2 41 38 3.37 3.10 42.9 36.9 NS Prx-2 2 37 34 3.29 3.13 34.5 37.4 NS Prx-7 3 34 39 3.15 3.28 35.3 38.5 NS Ss-1 3 31 39 3.32 3.27 37.0 34.3 NS Pqm-2 4 36 42 2.92 3.52 32.1 45.8 2.66 b Aps-1 6 35 38 3.09 3.29 38.8 35.1 NS Aman-1 6 15 22 3.00 3.42 16.8 20.5 NS Got-2 7 47 32 4.00 2.13 55.8 16.8 -7.40** Aps-2 8 43 34 3.40 3.00 43.1 33.8 -1.82* Pn 8 14 53 3.16 3.22 34.2 33.2 NS Prx-4 10 42 36 3.42 3.02 43.2 35.1 NS Sod-1 11 23 33 3.48 3.10 32.4 25.8 NS Est-10 11 29 50 3.48 3.10 45.0 37.1 NS Est-4 12 21 37 3.50 3.15 32.7 27.7 NS 6Padh-2 12 20 55 3.29 3.29 37.7 38.1 NS Pai-1 12 21 58 3.16 3.27 38.3 40.6 NS Aco-1 12 29 50 3.08 3.34 36.4 42.1 NS Hk-l •p 25 27 3.45 3.12 29.3 23.9 NS *, **Significant at 0.05 and 0.0001 levels, respectively. a BC2 progeny were scored according to a visual rating system; 1 = resistant to 5 = susceptible. Significant at 0.01 level using a two-tailed nonparametric analysis of variance with heterozygotes being more susceptible than homozygotes.

PAGE 56

49 Table 4-2. Two-way contingency table between Got-2 a and 1-3 . ee L. esculentum homozygotes; ep L. pennellii heterozygotes, ee 46 Got-2 ep 1 Totals 47 1-3 ep Totals 1 47 31 32 32 79 X z 70.92. Prob < 0.001. p (% recombination) = 2.5%.

PAGE 57

50 Table 4-3. Detection of minor factors taking into account the effect of 1-3 . Two-way ( 1-3 by each marker locus) factorial analyses of variance (2x2) comparing BC;l mean disease rankings were utilized. Main effects were calculated according to Steel and Torrie (1960) . ee L. esculentum homozygotes ; ep L. pennellii heterozygotes. Mean Disease Ranking Main Effects InterLoci ee/ee ee/ep ep/ee ep/ep 1-3 Marker action 1-3 Aps-2 58.15 53.00 19.72 13.28 -39.1** 5.80* -0.64 1-3 Dia-2 50.78 58.52 21.57 15.39 -36.2** 0.78 -6.96* 1-3 Pqm-2 a 56.09 56.42 13.58 21.38 -38.8** 4.06 3.74 *, **Significant at the 0.05 and 0.0001 levels, respectively . a Pqm-2 analysis included to illustrate that significance observed in the nonparametric analysis of variance (Table 41) was an artifact; no significance existed after accounting for 1-3.

PAGE 58

51 Table 4-4. Two-way contingency table comparing 1-3 and Pqm2. ee L. esculentum homozygotes; ep L. pennellii heterozygotes . 1-3 ee ep Totals ee 16 20 36 Perm2 ep 30 12 42 Totals 46 32 78 X 2 = 5.834. Prob < 0.05. p (% recombination) = 64.1%.

PAGE 59

52 1.0 1.4 1.8 2.2 2.6 3.0 3.4 3.8 4.2 4.6 5.0 Mean Disease Rating Fig. 4-1. Frequency distribution of backcross (BC3J mean disease ratings. For each BC^, at least twenty progeny (BC-^S-l) were inoculated with race 3 and scored for their level of vascular browning. Higher scores represented greater levels of vascular browning, and thus greater susceptibility. A mean disease rating for each BCj individual was calculated by taking a weighted mean of its progeny scores.

PAGE 60

53 Fig. 4-2. GOT starch gel. The origin is at the bottom of the gel; the anode electrode is at the top of the gel. Got2 is the second zone of activity from the top. BB is L. esculentum cv. Bonny Best. 'Bonny Best 1 , LA 716, and F^^ controls are in the center of the gel. The other wells are F 2 progeny from the cross 'Bonny Best' x LA 716. The LA 716 allele migrates approximately 9 mm further than the L. esculentum allele. The central band in the F;l is an intragenic heterodimer indicating that Got-2 is a dimeric enzyme. Plants selected using Got-2 are 97.5% likely to have the same genotype at the 1-3 locus.

PAGE 61

CHAPTER 5 GENETIC ANALYSIS OF RESISTANCES TO RACES 1 AND 2 OF FUSARIUM WILT INTRODUCTION Race 3 of the tomato Fusarium wilt pathogen Fusarium oxysporum f. sp. lycopersici (Sacc.) Snyder and Hansen is spreading throughout important growing regions worldwide. It has been reported in Australia (Grattidge and O'Brien, 1982) , Florida (Volin and Jones, 1982) , and California (Davis et al. . 1988). A dominant gene, 1-3 , that controls resistance to race 3 has been detected in the wild tomato species Lycopersicon pennellii (Corr.) D'Arcy, accession LA 716 (Scott and Jones, 1989) , and found to be tightly linked (2.5 cM) to the Got-2 locus on chromosome 7 (see Chapter 4). Since LA 716 was also found to be resistant to races 1 and 2 (Scott and Jones, 1989) , a genetic analysis was conducted to elucidate the relationships among the resistances to the three races. Two parallel approaches were taken for this analysis. First, although most commercial cultivars already carry resistances to races 1 and 2 (X and 1-2 on chromosome 11) derived from L. pimpinellifolium accessions (Bohn and Tucker, 1939; 1940; Paddock, 1950; Alexander and Hoover, 54

PAGE 62

55 1955; Stall and Walter, 1965; Latterot, 1976), it was of interest to determine whether LA 716 carried additional genes for resistance to these two races. For this reason, allelism of I and 1-2 with those resistances present in LA 716 was tested. In the second approach, the chromosomal location of genes controlling resistances to races 1 and 2 found in LA 716 were determined using linkage analyses to 17 previously mapped enzyme loci. MATERIALS and METHODS Plant Material Two progenies were generated for the allelism test using the L. esculentum cultivar Hayslip; this cultivar carries I and 1-2 . The first progeny was an F 2 of the cross 'Hayslip' x LA 716. Detection of susceptible individuals in this progeny would indicate that the genes were nonallelic. A second progeny [Bonny Best x (Hayslip x LA 716)] was also analyzed because of the increased chances of observing susceptible plants in a nonallelic interaction. The interspecific backcross, L. esculentum x (L. esculentum x L. pennellii . LA 716) analyzed previously for the genetics of race 3 resistance (see Chapter 4) was used in this study to analyze the responses to races 1 and 2 through progeny (BC^Sj) tests. 'Bonny Best' was used as the L. esculentum parent because of its extreme susceptibility to all three races. The 54 BCj^ individuals used in this analysis (those for which BC]^ seed was still available)

PAGE 63

56 were a subset of the 79 individuals previously used to locate 1-3 (see Chapter 4) . Chi-square analyses indicated that, for all marker loci, ratios of homozygotes to heterozygotes in the subset did not differ significantly from those observed in the overall population. In addition, a Z-test indicated that the race 3 mean disease rating of the subset was not different from that of the overall population. Isozyme Analysis and Inoculations Isozyme analysis was conducted on the BC-^ individuals as described previously (see Chapter 4) to locate genes responsible for resistances to races 1 and 2. Backcross individuals were scored for each of 23 enzyme loci including Prx-1 , Skdh-1 , Bnaq-1 . Dia-2 , and Est-9 on chromosome 1; Est-7 , Prx-2 , and Mpi-1 on chromosome 2; Prx-7 and Ss-1 on chromosome 3 ; Pqm-2 on chromosome 4 ; Aps-1 and Aman-1 on chromosome 6 ; Got-2 on chromosome 7 ; Aps-2 on chromosome 8 ; Prx-4 on chromosome 10; Sod-1 and Est-10 on chromosome 11; Est-4 . 6Pqdh-2 , Pqi-1 , and Aco-1 on chromosome 12 and Hk-1 whose chromosome designation has not been determined. The morphological marker, Pn on chromosome 8, was also scored. BC-l individuals were self-pollinated to produce BC-]^ seed for progeny tests. BC]^ individuals and the progenies used in the allelism test were inoculated according to previously described procedures (see Chapter 4) with races 1 and 2 of

PAGE 64

57 F. oxysporum f. sp. lycopersici (race 1: strain SC626, isolated by Dr. M. Cirulli in Italy, also called the "Oristano" isolate; race 2: strain SC548, isolated by Dr. J.M. Walter at the University of Florida at Bradenton) . Included in each set of inoculations were 'Bonny Best', LA 716, 'Manapal' (resistant to race 1 only), 'Hayslip', and the F 1 (Bonny Best x LA 716) to monitor the activity and specificity of inoculum. Also, another set of controls was treated with distilled water only. Plants were evaluated 10-14 days after inoculation and again approximately one month after inoculation using the visual rating system of Chapter 4 . A mean disease rating was calculated for each backcross individual for each race using the progeny scores; at least 20 progeny were used for each analysis. All statistical analyses were performed on a main frame computer using the Statistical Analysis System (SAS Institute Inc., SAS Circle, Box 8000, Cary, NC 27511). Procedures used included the following: Proc Univariate, Proc Cluster, Proc Freg, and Proc Nonpar lway. RESULTS The detection of susceptible plants in F 2 progenies inoculated with race 1 or 2 indicated that neither I nor 1-2 was allelic to the resistances carried by LA 716 (Table 51) . The same conclusion was more decisively reached with the results of the L. esculentum backcross progeny where

PAGE 65

58 several susceptible plants were observed for each race: 39.6% for race 1 and 39.4% for race 2 (Table 5-1). Frequency distributions of BC^ mean disease ratings for races 1 and 2 were found to be bimodal according to the coefficient of bimodality (b = (m 3 2 +l)/ (m 4 +3) , m 3 = skewness, m 4 = kurtosis; race 1: b = 0.563; race 2: b = 0.584; a < 0.05). A bimodal distribution was also observed for the race 3 mean disease ratings (see Chapter 4) . These results indicated that a single major locus was responsible for resistances to races 1 and 2, as was the case for race 3. The possible association of resistance genes with any of the 24 segregating enzyme markers was tested by comparing the mean disease ratings of homozygotes and heterozygotes at each marker locus; a one-tailed nonparametric analysis of variance was used for these tests as previously described (see Chapter 4) . Heterozygotes for the chromosome segment marked by Got-2 were markedly more resistant than homozygotes for both races 1 and 2 (Table 5-2) . Since 1-3 . a gene conferring race 3 resistance, was also tightly linked to Got-2 (see Chapter 4) , these results indicated that resistances to the three races were either controlled by 1-3 alone or gene(s) tightly linked to 1-3 . Moreover, as previously reported for race 3 (see Chapter 4) , significant effects were also observed for the chromosome segment marked by Aps-2 for both races (Table 5-2) suggesting that

PAGE 66

59 Tfw . a factor linked to Aps-2 , was active against all three races. The chromosome segment marked by Prx-4 also appeared to have an effect for both races; however, this was determined to be an artifact. A two-way contingency table comparing Got-2 and Prx-4 (Table 5-3) showed slight skewing (a < 0.05) in favor of parental types (66%). This resulted in more BC]^ individuals that were homozygous or heterozygous for both loci than expected. Since the chromosome segment marked by Got-2 was significant for both races (Table 5-2), the significance observed for Prx-4 can be attributed to skewing with the resistance gene(s) linked to Got-2 . In addition, significance for the Prx-4 marker was not observed in the original population (N = 79) analyzed for race 3 resistance (see Chapter 4) . Other marker loci assorted as previously reported (Tanksley and Rick, 1980b; Chetelat, 1989) ; no significant deviations in expected ratios were detected between unlinked loci. In order to discern whether 1-3 alone or other tightly linked loci were responsible for resistances to races 1 and 2, the genotype (resistant or susceptible) at each of the putative race specific loci was deduced for all BC]^ individuals based on mean disease ratings of their progeny. For this purpose, the average linkage clustering analysis was used as described previously (see Chapter 4) . Two-way contingency tables were used to test for independent assortment between the putative loci that control

PAGE 67

60 resistances to the three races. No recombinants were detected between resistances to races 2 and 3; however, seven BC^ individuals appeared to be recombinants between race 1 resistance and the other two resistances (Table 5-4) . DISCUSSION The data presented here seem to indicate that 1-3 confers resistance to races 2 and 3, and that another linked gene on chromosome 7 controls for resistance to race 1. However, a closer examination of the seven putative recombinants between race 1 resistance and resistances to races 2 and 3 suggests that they may have been misclassif ied by the cluster analysis. First, mean disease ratings of the putative recombinants were very close to the cuttoff point for the resistant class for race 1 (Table 5-4) . In addition, all seven recombinants appeared in the same class — resistant for race 1 and susceptible for races 2 and 3 (Table 5-4) . This was due to the fact that the upper limit of the race 1 resistant cluster was much greater than the limits for races 2 and 3 (Table 5-4) . The presence of minor factors, such as Tfw, increases the frequency of individuals with intermediate disease ratings; these individuals have a greater probability of being misclassif ied. An additional reason for the discrepancy in clustering could have been due to the slightly greater aggressiveness of race 1 compared to the other two races. For instance, all 'Bonny Best' x LA 716 — F;l plants evaluated were given a rating of 1 or 2 for

PAGE 68

61 all three races (Table 5-1) . However, a 2 rating was given to 16.7% of the F-l's analyzed for race 1 compared to only 7.1% and 4.1% for races 2 and 3, respectively. Though no clear recombinants were observed between race resistances, the possibility of more than one resistance gene on chromosome 7 could not be ruled out with the available data. In the past, the Fusarium wilt pathogen has been able to overcome host genes for resistance (I and 1-2 ) and infect commercial tomato cultivars. Races 1 and 2 have been spread to most tomato growing regions, and still pose the threat of forming a new race. Since Got-2 is tightly linked to monogenic resistances to all three races, it can be used as a scorable marker to incorporate genetic resistances to each of the races into new cultivars. This might reduce the probability of a new race forming in the near future. In addition, if Tfw were tagged with a chromosome 8 marker, its incorporation into new cultivars might offer even more stable resistance to each of the races. It should be noted that only one isolate of each of the races was used as a source of inoculum in this study, and therefore, these results should be verified with other isolates. Previous reports have concluded that genes for resistance to obligate parasites such as Puccinia . Uromyces . and Ervsiphe are the product of coevolution between these pathogens and their hosts (Anikster and Wahl , 1979; Eshed

PAGE 69

62 and Dinoor, 1981; Moseman et al. , 1983; 1984). F. oxysporum is a facultative parasite with the ability to survive away from its host. An interesting feature of the L. pennellii resistance to Fusarium wilt is that it may not be the result of selection pressure on the host. The natural habitat of L. pennellii (LA 716) is in the deserts of southern Peru where it grows on the slopes of rocky soils with nearly no precipitation (Rick and Tanksley, 1981) . Either the tomato Fusarium wilt pathogen can adapt to the extreme environment of L. pennellii , or the resistance to the fungus is not the result of coevolution between pathogen and host. A survey of Fusarium populations and an assessment of their pathogenic activity in the area of distribution of L. pennellii could elucidate whether LA 716 may have had an evolutionary interaction with F. oxysporum .

PAGE 70

63 Table 5-1. Results of the inoculations for allelism tests for I and 1-2 resistances with LA 716 resistances to races 1 and 2. Visual ratings were given one month after inoculations. Cross* Race Visual rating Total % (No. of plants) 5 No . Suscep1 2 3 4 5 Plants tible c Controls Bonny Best LA 716 Bonny Best x LA 716 Manapal Hayslip 1 29 29 100.0 2 34 34 100.0 1 21 21 0.0 2 14 14 0.0 1 25 5 30 0.0 2 39 3 42 0.0 1 17 8 1 1 27 7.4 2 1 1 7 14 28 51 96.1 1 15 5 1 21 4.8 2 13 7 2 1 23 13.0 Allelism progenies F 2 of Hayslip x LA 716 1 32 2 32 BB x (Hayslip x LA 716) 1 31 2 48 1 1 34 2.9 2 1 35 2.8 4 9 8 6 58 39.6 9 7 6 24 94 39.4 *, **Signif icant at the 0.05 and 0.01 levels, respectively. a Denotes the number of plants falling into each visual rating category; 1, resistant to 5, susceptible. b BB Bonny Best: i/i, i-2/i-2 ; Manapal I/I, i-2/i-2 ; Hayslip: I/I, 1=2/1=2. c Plants with a visual rating of 3 5 were classified as susceptible; whereas, ratings of 1 and 2 were considered resistant.

PAGE 71

64 Table 5-2. Comparisons of BC]_ mean disease ratings using a one-way nonparametric analysis of variance — the Wilcoxon rank sum test. In this procedure, BC]_ individuals were ranked according to their mean disease rating, and homozygote and heterozygote mean disease rankings were compared at each marker locus for each race, ee L. esculentum homozygotes ; ep L. Dennell ii hetero zygotes ; Rl, R2, and R3 r< aces 1, 2, and 3, respect ively. a Rl mean R2 ] nean R3 mean Wilcoxon Chromo No. disease disease disease rank sum t 2SZ some No. olants ee ep ranking ee ep rankina ee ep rankina ee ep (1 statis' tici Locus Rl R2 R3 Prx-1 1 16 38 26.5 27.9 28.3 27.7 [26.4 28.0) NS NS (NS) Skdh-1 1 16 35 27.5 25.3 28.7 24.8 r 28.2 25.0) NS NS (NS) 3naa-l 1 14 29 22.0 22.0 23.2 21.4 '22.6 21.7) NS NS (NS) Dia-2 1 12 35 24.7 23.8 28.2 22.5 '28.1 22.6) NS NS (NS) Est-9 1 7 20 15.3 13.6 13.4 15.6 '15.4 13.5) NS NS (NS) Est-7 2 30 24 27.2 27.8 28.2 26.7 '27.8 27.1) NS NS (NS) Prx-2 2 25 23 23.3 25.8 23.5 25.6 '22.6 26.5) NS NS (NS) Hni-1 2 18 15 15.2 19.2 13.9 20.7 '14.9 19.5) NS NS (NS) Prx-7 3 24 25 25.4 24.6 23.0 26.9 23.3 26.6) NS NS (NS) Ss-1 3 21 30 26.7 25.5 26.3 25.8 27.0 25.3) NS NS (NS) Pam-2 4 26 28 24.0 30.7 25.5 29.4 24.2 30.5) NS NS (NS) Aos-1 6 28 23 25.9 26.2 26.6 25.2 25.3 26.8) NS NS (NS) Aman-l Got-2 6 7 13 29 17 25 14.9 39.6 15.9 13.5 14.8 40.0 16.0 13.0 13.6 40.0 16.9) 13.0) NS -6.1** NS -6.3** (-6.3**) (-2.2*) Aps-2 8 31 23 31.4 22.2 31.6 22.0 | 31.6 22.0) -2.1* -2.2* Pn 8 8 39 23.9 24.0 24.2 24.0 ( 25.9 23.6) NS NS (NS i h (-2.2*) b Prx-4 10 30 23 32.1 20.4 30.2 22.8 ( 31.2 21.6) -2.7* -1.7* Sod-1 11 13 19 20.6 17.5 20.0 18.1 ( 20.7 17.4) NS NS (NS) Est-10 11 21 33 31.6 24.9 31.3 25.1 ( 30.2 25.8) NS NS (NS) Est-4 12 15 25 23.9 18.5 24.6 18.1 ( 22.9 19.1) NS NS (NS) 6Padh-2 12 15 37 26.5 26.5 26.7 26.4 ( 25.8 26.8) NS NS (NS) Pqi-1 12 14 40 26.8 27.8 26.8 27.8 ( 25.5 28.2) NS NS (NS) Aco-1 12 20 34 25.2 28.8 23.5 29.8 ( 24.5 29.3) NS NS (NS) Hk-1 p 17 19 18.4 18.6 18.5 18.5 ( 19.1 18.0) NS NS (NS) *, ** Significant at 0.05 and 0.0001 levels, respectively. a Race 3 mean disease rankings and Z-statistics calculated using only the 54 individuals analyzed for races 1 and 2. b In the original population (N = 79) analyzed for race 3, the chromosome segment marked by Prx-4 was not significant (see Chapter 4) .

PAGE 72

65 Table 5-3. Two-way contingency table comparing Got-2 and Prx-4 . ee L. esculentum homo zygotes; ep L. pennellii heterozygotes . Totals 30 23 53 Got-2 fcr fcr ep ee 20 10 Prx-4 ep 8 15 Totals 28 25 X 2 = 5.31. Prob < 0.05. % parental types = 66.0%.

PAGE 73

66 Table 5-4. Mean disease ratings and cluster classifications of putative recombinants between resistances to races 1 and 2/3. BC;l individuals were classified as resistant or susceptible for each race using the average linkage clustering method based on their mean disease ratings. Rl, R2, and R3 for races 1, 2, and 3, respectively; R and S resistant and susceptible. 9 Cluster class if icatior Rl R2 R3 l Mean disease ratincr BC# Rl R2 R3 12 R S S 3.15 3.78 3.49 22 R S S 3.29 3.46 3.64 41 R S S 3.64 4.08 3.92 65 R S S 2.90 3.30 3.05 71 R S S 3.59 4.00 3.96 92 R S S 3.36 3.77 3.06 130 R S S 3.24 3.63 3.70 Ranqes Resistant cluster 1. 55-3. 64 1 39 2. 86 1.50 2.70 Susceptibl e cluster 3. 78 4. 88 3. 30 4. 93 3.05 4.97 a Data for race 3 clustering analysis was obtained from Chapter 4 .

PAGE 74

CHAPTER 6 SUMMARY AND CONCLUSIONS Fusarium wilt of tomato, caused by races 1, 2, and 3 of the pathogen Fusarium oxysporum f . sp. lycopersici , is a potentially devastating disease found in most growing regions worldwide. Currently, use of race 3 for screening purposes by breeders is restricted due to fear of introduction of this race into new regions. Use of Got-2 as a scorable marker to incorporate monogenic resistances to all three races into commercial cultivars will have two benefits. One, it will allow for production of race 3 resistant breeding lines and cultivars without introducing the pathogen into new growing regions; and second, it will result in the incorporation of additional monogenic resistances to races 1 and 2. This may reduce the likelihood of future new race formation by these two races. Also, more durable resistance may be acguired by incorporating Tfw into new cultivars. Several future studies should be conducted. Using near isogenic lines segregating only for Got-2 and resistances to races 1, 2, and 3 in a L. esculentum background, it will be possible using progeny tests to determine the number of genes controlling these resistances and a gene order if more 67

PAGE 75

68 than one gene exists. The Tfw resistance should be analyzed to determine whether it confers field tolerance against the pathogen, and whether it acts as a dominant or recessive allele. Also, a mapping project must be conducted to identify a scorable marker ( Aps-2 or an RFLP on chromosome 8) for this gene so that it can be incorporated along with the chromosome 7 resistances into new cultivars. Modes of action of I, 1-2 . and the L. pennellii resistances could be analyzed and compared by observing plant responses to each of the races in near isogenic lines carrying one or more of the resistance genes.

PAGE 76

APPENDIX VEGETATIVE COMPATIBILITY AMONG RACES 1, 2, AND 3 OF FUSARIUM OXYSPORUM F. SP. LYCOPERSICI Genetic relationships among races 1, 2, and 3 of Fusarium oxysporum f. sp. lycopersici were explored by analyzing vegetative compatibility through heterokaryosis. Heterokaryosis is the process of anastomosis by fungal hyphae, forming a cell with two different nuclei in the same cytoplasm. Genotypes are only compatible if all vie (vegetative incompatibility) loci are identical. Previously, Sidhu and Webster (1979) were able to generate heterokaryons between races 1 and 2 using amino acid auxotrophs . Nitrate nonutilizing ( nit ) auxotrophic mutants were developed for each race using the techniques of Cove (1976) . Mutants were paired on Czapek-Dox minimal medium and allowed to grow at 28 °C until they came into contact. The appearance of thick wild type growth at the intersection indicated complementation and a potential heterokaryon. Prototrophic colonies were tested for the possibility that wild type growth could be due to crossfeeding. Dialysis tubing (permeable to MW 12,000 14,000) was placed between complementary mutants; growth along the tubing would have 69

PAGE 77

70 indicated crossfeeding. Also, in another crossfeeding test, one nit mutant was grown in liquid culture at room temperature for 1-2 weeks at 100 rpm and filtered through a 0.45 urn micropore to collect diff usable metabolites. The filtrate was then incorporated into cooled (40 °C) CzapekDox agar and a complementary mutant plated. The appearance of wild type growth would have indicated crossfeeding. Complementation was observed between nit mutants of races 2 and 3 ; both crossfeeding tests indicated that this was due to heterokaryosis and not crossfeeding. None of twenty race 1 nit mutants showed complementation to nit mutants of races 2 and 3. However, race 1 mutants did not show complementation among themselves either. These data indicate that races 2 and 3 are vegetatively compatible, and likely share a similar phylogenetic background. However, no conclusions about vegetative compatibility or phylogeny between race 1 and races 2 and 3 can be made from these results.

PAGE 78

REFERENCES Ainsworth, G.C. and G.R. Bisby. 1943. A dictionary of the fungi. The Imperial Mycological Institute. Kew and Surrey, p 228. Albersheim, P. and A.J. Anderson. 1971. Proteins from plant cell walls inhibit polygalacturonases secreted by plant pathogens. Proc. Natl. Acad. Sci. U.S.A. 68:1815-1819. Alexander, L.J. and M.M. Hoover. 1955. Disease resistance in wild species of tomato. Ohio Agr. Exp. Sta. Res. Bull. 752. Alexander, L.J. and CM. Tucker. 1945. Physiological specialization in the tomato wilt fungus Fusarium oxysporum f. sp. lycopersici . J. Agr. Res. 70:303-313 Alon, H. , J. Katan, and N. Kedar. 1974. Factors affecting penetrance of resistance to Fusarium oxysporum f . sp. lycopersici in tomatoes. Phytopathology 64:455-4 61. Armstrong, G.M. and J.K. Armstrong. 1948. Nonsusceptible hosts as carriers of wilt Fusaria. Phytopathology 38:808-826. Baker, K.F. 1953. Fusarium wilt of China aster. In: Stefferud, A. (ed.). Plant diseases. U.S. Dept. Agric. Yearbook of Agriculture. Washington, D.C., pp 572-577. Bassiri, A. and I. Rouhani. 1977. Identification of broad bean cultivars based on isozyme patterns. Euphytica 26:279-286. Beckman, C.H. 1987. The nature of wilt diseases of plants. APS Press. The American Phytopathological Society. St. Paul, MN. 175 p. Beckman, C.H. and S. Halmos. 1962. Relation of vascular occluding reactions in banana roots to pathogenicity of root-invading fungi. Phytopathology 52:893-897. 71

PAGE 79

72 Beckman, C.H., S. Halmos, and M.E. Mace. 1962. The interaction of host, pathogen, and soil temperature in relation to susceptibility to Fusarium wilt of bananas. Phytopathology 52:134-140. Beckman, C.H., J.E. Kuntz, A.J. Riker, and J.G. Berbee. 1953. Host responses associated with the development of oak wilt. Phytopathology 43:448-454. Beckman, C.H., W.C. Mueller, and M.E. Mace. 1974. The stabilization of artificial and natural cell wall membranes by phenolic infusion and its relation to wilt disease resistance. Phytopathology 64:1214-1220. Beckman, C.H., W.C. Mueller, B.J. Tessier, and N.A. Harrison. 1982. Recognition and callose deposition in response to vascular infection in Fusarium wiltresistant or susceptible tomato plants. Physiol. Plant Pathol. 20:1-10. Beckman, C.H. and G.E. Zaroogian. 1967. Origin and composition of vascular gel in infected banana roots. Phytopathology 57:11-13. Biehn, W.L. 1973. p-Fluorophenylalanine selectively inhibits sporulation of two wilt fungi. Phytopathology 63:421423. Bishop, CD. and R.M. Cooper. 1983. An ultrastructural study of root invasion in three vascular wilt diseases. Physiol. Plant Pathol. 24:277-289. Bohn, G.W. and CM. Tucker. 1939. Immunity to Fusarium wilt in the tomato. Science 89:603-604. Bohn, G.W. and CM. Tucker. 1940. Studies on Fusarium wilt of the tomato. I. Immunity in Lycopersicon pimpinellifolium and its inheritance in hybrids. Mo. Agr. Exp. Sta. Res. Bull, p 311. Booth, C 1971. In: The Genus Fusarium. Commonwealth Mycol . Inst. Kew, Surrey, and England, p 23 3. Brown, A.H.D. and B.S. Weir. 1983. Measuring genetic variability in plant populations. In: Tanksley, S.D. and T.J. Orton (eds.) Isozymes in genetics and breeding. Elsevier. Amsterdam, Oxford, and New York, pp 219-240. Buckley, P.M., T.D. Wyllie, J.E. DeVay. 1969. Fine structure of conidia and conidium formation in Verticillium alboatrum and V. nigrescens . Mycologia 61:240-250.

PAGE 80

73 Chattway, M.M. 1949. The development of tyloses and secretion of gum in heartwood formation. Aust. J. Sci. Res., Ser. B 2:227-240. Chetelat, R.T. 1989. Isozyme gene linkage map of tomato ( Lycopersicon esculentum ) . Isozyme Bulletin 22:19. Clayton, E.E. 1923a. The relation of temperature to the Fusarium wilt of the tomato. Amer. J. Bot. 10:71-88. Clayton, E.E. 1923b. The relation of soil moisture to the Fusarium wilt of the tomato. Amer. J. Bot. 10:133-147. Cove, D.J. 1976. Chlorate toxicity in Aspergillus nidulans : the selection and characterisation of chlorate resistant mutants. Heredity 36:191-203. Crawford, D.J. 1983. Phylogenetic and systematic inferences from electrophoretic studies. In: Tanksley, S.D. and T.J. Orton (eds.) Isozymes in breeding and genetics. Elsevier. Amsterdam, Oxford, and New York, pp 257-288. Davis, R.M. , K.A. Kimble, and J.J. Farrar. 1988. A third race of Fusarium oxysporum f. sp. lycopersici identified in California. Plant Disease 72:453. Edgington, L.V. and J.C. Walker. 1958. Influence of calcium and boron nutrition on development of Fusarium wilt of tomato. Phytopathology 48:324-326. Fisher, P.L. 1935. Responses of tomato in solution cultures with deficiencies and excesses of certain essential elements. Md. Agr. Exp. Sta. Bull. 375. Foster, R.E. and J.C. Walker. 1947. Predisposition of tomato to Fusarium wilt. J. Agr. Res. 74:165-185. Freeling, M. and J. A. Birchler. 1980. Mutants and variants of the alcohol dehydrogenase-1 gene in maize. In: Genetic engineering. Setlow, J.K. and A. Hollander (eds.). Plenum Press. New York, vol 3, pp 223-264. Friend, J. 1976. Lignif ication in infected tissue. In: Friend, J. and D.R. Threlfall (eds.). Biochemical aspects of plant-parasite relationships. Academic Press. New York, pp 291-303. Gottlieb, L.D. 1981. Electrophoretic evidence and plant populations. Progress in Phytochemistry 7:1-46.

PAGE 81

74 Grattidge, R. and R.G. O'Brien. 1982. Occurrence of a third race of Fusarium wilt of tomatoes in Queensland. Plant Disease 66:165-166. Griffin, G.J. 1970. Carbon and nitrogen requirements for macroconidial germination of Fusarium solani : dependence on conidial density. Can. J. Microbiol. 16:733-740. Harrison, A.L. 1941. Studies on tomato wilt. Tex. Agr. Exp. Sta. Rept. 1940:88. Heath-Pagliuso, S., R.C. Huffaker, and R.W. Allard. 1984. Inheritance of nitrite reductase and regulation of nitrate reductase, nitrite reductase, and glutamine synthetase isozymes. Plant Physiol. 76:353-358. Hoffmann, G.M. 1966. Untersuchungen uber die heterokaryosebildung und den parasexualcyclus bei Fusarium oxysporum . I. Anastomosenbildung im mycel und kernverhaltwisse bei der conidienentwicklung. Arch. Mikrobiol. 53:336-347. Howard, R.J. and D.P. Maxwell. 1975. Cytology of Ceratocystis ulmi. Phytopathology 65:816-819. Hutson, R.A. and I.M. Smith. 1980. Phytoalexins and tyloses in tomato cultivars infected with Fusarium oxysporum f . sp. lycopersici or Verticillium albo-atrum . Physiol. Plant Pathol. 17:245-257. Ioannou, N. R.W. Schneider, and R.G. Grogan. 1977. Effect of oxygen, carbon dioxide, and ethylene on growth, sporulation, and production of microsclerotia by Verticillium dahliae . Phytopathology 67:645-650. Jones, J. P. and S.S. Woltz. 1969. Fusarium wilt (race 2) of tomato: Calcium, pH, and micronutrient effects on disease development. Plant Dis. Rep. 53:276-279. Jones, J. P. and S.S. Woltz. 1970. Fusarium wilt of tomato: Interaction of soil liming and micronutrient amendments on disease development. Phytopathology 60:812-813. Katan, J. 1971. Symptomless carriers of the tomato Fusarium wilt pathogen. Phytopathology 61:1213-1217. Kolomogorov, A.N. 1933. Sulla determinazione empirica di una legge di distribuzione. G. Inst. Ital . Attuari . 4:8391.

PAGE 82

75 Latterot, H. 1976. Mapping of 1-2 allele in tomato, controlling the genetic resistance to pathotype 2 of Fusarium oxysporum f . sp. lycopersici wilt. Ann. Amelior. Plant. 26(3) :485-491. Mace, M.E. 1963. Histochemical localization of phenols in healthy and diseased banana roots. Physiol. Plant 16:915-925. Malca, I., D.C. Erwin, W. Moje, and B. Jones. 1966. Effect of pH and carbon and nitrogen sources on the growth of Verticillium albo-atrum . Phytopathology 56:401-406. Marshall, K.C. and M. Alexander. 1960. Competition between soil bacteria and Fusarium . Plant Soil 12:143-148. Massee, G. 1895. The "sleepy disease" of tomatoes. Gard. Chron. Ser. 3, 17:707-708. Marchant, R. 19 66. Fine structure and spore germination in Fusarium culmorum . Ann. Bot. 30:441-445. McGrath, D.J., D. Gillespie, and L. Vawdrey. 1987. Inheritance of resistance to Fusarium oxysporum f . sp. lycopersici races 2 and 3 in Lycopersicon pennellii . Aust. J. Agric. Res. 38:729-733. Medina-Filho, H.P. and M.A. Stevens. 1980. Tomato breeding for nematode resistance: survey of resistant varieties for horticultural characteristics and genotype of acid phosphatase. Acta. Hortic. 100:383-393. Moore-Landecker, E. 1972. In: Fundamentals of the fungi. Prentice Hall, Inc. Englewood Cliffs, N.J., pp 132-143. Mozumder, B.K.G., N.E. Caroselli, and L.S. Albert. 1970. Influence of water activity, temperature, and their interaction on germination of Verticillium albo-atrum conidia. Plant Physiol. 46:347-349. Nelson, P.E. 1981. Life cycle and epidemiology of Fusarium oxysporum . In: Fungal wilt diseases of plants. Mace, M.E., A. A. Bell, and C.H. Beckman (eds.). Academic Press. New York, London, Toronto, Sydney, and San Francisco, pp 51-80. Paddock, E.F. 1950. A tentative assignment of the Fusariumimmunity locus to linkage group 5 in tomato. Genetics 35:683-684.

PAGE 83

76 Pegg, G.F. 1976. The occurrence of /3-1, 3-glucanase in healthy and Verticillium albo-atrum -infected susceptible and resistant tomato plants. J. Exp. Bot. 27:1093-1101. Pegg, G.F. and D.H. Young. 1981. Changes in glycosidase activity and their relationship to fungal colonization during infection of tomato Ly copers icon esculentum cultivar Craigella by Verticillium albo-atrum . Physiol. Plant Pathol. 19:371-382. Phipps, P.M. and R.J. Stipes. 1976. Histopathology of Mimosa infected with Fusarium oxysporum f. sp. perniciosum . Phytopathology 66:839-843. Puhalla, J.E. and A. A. Bell. 1981. Genetics and biochemistry of wilt pathogens. In: Mace, M.E., A. A. Bell, and C.H. Beckman (eds.). Fungal wilt diseases of plants. Academic Press. New York, London, Toronto, Sydney, and San Francisco, pp 145-192. Puhalla, J.E. and J.E. Mayfield. 1974. The mechanism of heterokaryotic growth in Verticillium dahliae . Genetics 76:411-422. Richards, B.L. 1925. Canning crop diseases. Utah Agr. Exp. Sta. Bull. 192:59-60. Rick, CM. 1969. Controlled introgression of chromosomes of Solanum pennellii into Lycopersicon esculentum : segregation and recombination. Genetics 62:753-768. Rick, CM. and J.F. Fobes. 1974. Association of an allozyme with nematode resistance. Rep. Tomato Genet. Coop. 24:25. Rick, CM. and J.F. Fobes. 1977. Linkage relations of some isozymic loci. Rep. Tomato Genet. Coop. 27:22-24. Rick, CM., J.F. Fobes, and M. Holle. 1977. Genetic variation in Lycopersicon pimpinellifolium : evidence of evolutionary change in mating systems. Plant Syst. Evol. 127:139-170. Rick CM. and S.D. Tanksley. 1981. Genetic variation in Solanum pennellii : comparisons with two other sympatric tomato species. PI. Syst. Evol. 139:11-45. Robinson, P.M. and M.K. Garrett. 1969. Identification of volatile sporostatic factors from cultures of Fusarium oxysporum . Trans. Br. Mycol . Soc. 52:293-299,

PAGE 84

77 Robinson, P.M., D. Park, and W.K. McClure. 1969. Observations on induced vacuoles in fungi. Trans. Br. Mycol. Soc. 52:447450. Santamour, F.S. and P. Demuth. 1980. Identification of Callery pear cultivars by peroxidase isozyme patterns. J. Heredity 71:447-449. Schwartz, D. 1969. Alcohol dehydrogenase in maize: genetic basis for multiple isozymes. Science 164:585-586. Scott, J.W. and J. P. Jones. 1989. Monogenic resistance in tomato to Fusarium oxysporum f. sp. lycopersici race 3. Euphytica 40:49-53. Shaw, C.R. and R. Prasad. 1970. Starch gel electrophoresis of enzymes — a compilation of recipes. Biochem. Genet. 4:297-320. Sherwood, E.C. 192 3. Hydrogen-ion concentration as related to the Fusarium wilt of tomato seedlings. Amer. J. Bot. 10:537-553. Sidhu, G.S. and J.M. Webster. 1979. A study of heterokaryosis and its influence on virulence in Fusarium oxysporum lycopersici . Can. J. Bot. 57:548-555. Snedecor, G.W. and W.G. Cochran. 1980. Statistical methods (7th ed.). The Iowa State University Press. Ames, la. p 78-81. Snyder, W.C. and H.N. Hansen. 1940. The species concept in Fusarium. Amer. J. Bot. 27:64-67. Sokal, R.R. and CD. Michener. 1958. A statistical method for evaluating systematic relationships. Univ. of Ks. Science Bulletin 38:1409-1438. Soost, R.K., T.E. Williams, and A.M. Torres. 1980. Identification of nucellar and zygotic seedlings of Citrus with leaf isozymes. HortScience 15:728-729. Stall, R.E. 1961. Development of Fusarium wilt on resistant varieties of tomato caused by a strain different from race 1 isolates of Fusarium oxysporum f . lycopersici . Plant Dis. Rep. 45:12-15. Stall, R.E. and J.M. Walter. 1965. Selection and inheritance of resistance in tomato to isolates of races 1 and 2 of the Fusarium wilt organism. Phytopathology 55:12131215.

PAGE 85

78 Steel, R.G.D. and J.H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill Book Company, Inc. New York, Toronto, and London, pp 194-231. Stoessl, A., J.B. Stothers, and E.W.B. Ward. 1976. Sesquiterpenoid stress compounds of the Solanaceae. Phytochemistry 15:855-872. Stotzky, G. and R.D. Goos. 1965. Effect of high C0 2 and low 2 tensions on the soil microbiota. Can. J. Microbiol. 11:853-868. Stover, R.H. 1962. Studies on Fusarium wilt of bananas. IX. Competitive saprophytic ability of F. oxysporum f. cubense . Can. J. Bot. 40:1473-1481. Stover, R.H. 1970. In: Toussoun, T.A. , R.V. Bega, and P.E. Nelson (eds.). Root diseases and soil-borne plant pathogens. Univ. California Press. Berkeley, Ca., pp 197-200. Strobel, J.W., N.C. Hayslip, D.S. Burgis, and P.H. Everett. 1969. Walter, a determinate variety resistant to races 1 and 2 of the Fusarium wilt pathogen. Fl . Agr. Exp. Sta. Cir. S-202. Talboys, P.W. 1958a. Some mechanisms contributing to Verticillium -resistance in the hop root. Trans. Br. Mycol. Soc. 41:227-241. Talboys, P.W. 1958b. Association of tylosis and hyperplasia of the xylem with vascular invasion of the hop by Verticillium albo-atrum . Trans. Br. Mycol. Soc. 41:249260. Tanksley, S.D. 1979. Linkage, chromosomal association, and expression of Adh-1 and Pqm-2 in tomato. Biochem. Genet. 17:1159-1167. Tanksley, S.D. 1983. Introgression of genes from wild species. In: Tanksley, S.D. and T.J. Orton (eds.) Isozymes in genetics and breeding. Elsevier. Amsterdam, Oxford, and New York, pp 331-338. Tanksley, S.D. and R.A. Jones. 1981. Effects of 2 stress on tomato alcohol dehydrogenase activity: description of a second ADH coding gene. Biochem. Genet. 19:397-409. Tanksley, S.D. and F. Loaiza-Figueroa. 1985. Gametophytic self-incompatibility is controlled by a single major locus on chromosome 1 in Lycopersicon peruvianum . Proc. Natl. Acad. Sci. 82:5093-5096.

PAGE 86

79 Tanksley, S.D., H. Medina-Filho, and CM. Rick. 1981. The effect of isozyme selection on metric characters in an interspecific backcross of tomato — basis of an early screening procedure. Theor. Appl . Genet. 60:291-296. Tanksley, S.D., H. Medina-Filho, and CM. Rick. 1982. Use of naturally occurring enzyme variation to detect and map genes controlling guantitative traits in an interspecific backcross of tomato. Heredity 49:11-25. Tanksley, S.D., J.C Miller, A.H. Paterson, and R. Bernatzky. 1988. Molecular mapping of plant chromosomes. In: Gustafson, J. P. and R. Appels (eds.). Chromosome structure and function. Plenum Publishing Co. pp 157-173. Tanksley, S.D. and CM. Rick. 1980a. Genetics of esterases in species of Ly coper s icon . Theor. Appl. Genet. 56:209219. Tanksley, S.D. and CM. Rick. 1980b. Isozymic gene linkage map of the tomato: applications in genetics and breeding. Theor. Appl. Genet. 57:161-170. Tanksley, S.D., CM. Rick, and CE. Vallejos. 1984. Tight linkage between a nuclear male-sterile locus and an enzyme marker in tomato. Theor. Appl. Genet. 68:109113. Tolmsoff, W.J. 1973. Life cycles of Verticillium species. In: Verticillium wilt of cotton. U.S. Dept. Agric. Publ. ARS-S-19, pp 20-38. Toussoun, T.A. and P.E. Nelson. 1976. A pictorial guide to the identification of Fusarium species (2nd ed.). Pa. State Univ. Press. University Park, Pa. Typas, M.A. and J.B. Heale. 1976. Heterokaryosis and the role of cytoplasmic inheritance in dark resting structure formation in Verticillium spp . Mol. Gen. Genet. 146:17-26. Vallejos, CE. 1983. Enzyme activity staining. In: Tanksley S.D. and T.J. Orton (eds.). Isozymes in plant genetics and breeding. Elsevier. Amsterdam, Oxford, and New York, pp 469-516. Vallejos, CE. and S.D. Tanksley. 1983. Segregation of isozyme markers and cold tolerance in an interspecific backcross of tomato. Theor. Appl. Genet. 66:241-247.

PAGE 87

80 Van Eck, W.H. and B. Schippers. 1976. Ultrastructure of developing chlamydospores of Fusarium solani f . cucurbitae in vitro . Soil Biol. Biochem. 8:1-6. Volin, R.B. and J. P. Jones. 1982. A new race of Fusarium wilt of tomato in Florida and sources of resistance. Proc. Fl. State Hort. Soc. 95:268-270. Walker, J.C. 1971. Fusarium wilt of tomato. Monograph 6. The American Phytopathological Society. St. Paul, MN. Walker, J.C. and R.E. Foster. 194 6. Plant nutrition in relation to disease development. III. Fusarium wilt of tomato. Amer. J. Bot. 33:259-264. Wardlaw, C.W. 1931. The biology of banana wilt (Panama disease). III. An examination of sucker infection through root bases. Ann. Bot. 45:381-399. Weber, J.C. and R.F. Stettler. 1981. Isoenzyme variation among ten populations of Populus trichocarpa Torr. et Gray in the Pacific Northwest. Silvae Genetica 30:2-3. Weeden, N.F. 1984. Distinguishing among white seeded bean cultivars by means of allozyme genotypes. Euphytica 33:199-208. Weeden, N.F. 1986. Genetic confirmation that the variation in the zymograms of 3 enzyme systems is produced by allelic polymorphism. Ann. Rep. Bean Improv. Coop. 29:117-118. Weeden, N.F. and R.C. Lamb. 1985. Identification of apple cultivars by isozyme phenotypes. J. Amer. Soc. Hort. Sci. 110:509-515. Weeden, N.F., R. Provvidenti, and G.A. Marx. 1984. An isozyme marker for resistance to bean yellow mosaic virus in Pi sum sativum . J. Heredity 75:411-412. Weller, J.I.', M. Soller, and T. Brody. 1988. Linkage analysis of guantitative traits in an interspecific cross of tomato ( Lycopersicon esculentum x Lycopersicon pimpinellifolium ) by means of genetic markers. Genetics 118:329-339. Wilcoxon, F. 1945. Individual comparisons by ranking methods . Biometrics 1:80. Wollenweber, H.W. and O.A. Reinking. 1935. Die Fusarien, ihre Beschreibung, Schadwirkung aund Bekampfung. Paul Pareyk, Berlin, p 335.

PAGE 88

BIOGRAPHICAL SKETCH Brian Bournival was born June 12, 1961 in Cleveland, Ohio, the son of Herbert and Neva Bournival. He earned his Bachelor of Science degree from Ohio State University in March of 1983. In January of 1986, he was granted a Master of Science degree from the University of Illinois, where he studied isozymes and their inheritance in the apple. 81

PAGE 89

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Jdhn W. Scott, Chair Associate Professor of Horticultural Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. C. Eduardo Valle^jos, Cochair Assistant Professor of Horticultural Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of .Doctor of Philosophy. H. 'Corby Kistler Assistant Professor of Plant Pathology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. 2 0
PAGE 90

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. *vJl JoHrif P. Jones ( j Processor of Plant Pathology This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1989 'cuA
PAGE 91

UNIVERSITY OF FLORIDA 3 1262 08556 7930 '


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EDH7AM2IP_J03D0F INGEST_TIME 2015-03-25T20:37:32Z PACKAGE AA00029739_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES