Genetic and Environmental Factors Affecting Improvement of Rootstocks for Tomato THESIS Presented in Partial Fulfillment Of

Genetic and Environmental Factors Affecting Improvement of Rootstocks for Tomato

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

Nancy Huarachi Morejon

Graduate Program in Horticulture and Crop Science

The Ohio State University

2013

Master's Examination Committee:

Dr. David M. Francis, “Advisor”

Dr. Pablo Jourdan

Dr. Matthew Kleinhenz

Copyrighted by

Nancy Huarachi Morejon

2013

Abstract

Grafting is a technique that has been used for fruit trees and vine crops for thousand years. Grafting to rootstocks is becoming popular in annual vegetable production to control soil-borne diseases, replace fumigation, increase yield, increase tolerance to abiotic stress, and impart vigor. Previous research indicates that inconsistent seed quality and lack of information about rootstock-scion compatibility affect the efficiency of grafting, raises cost, and inhibits adoption of the technology. The goals of this research were to address limitations in seed quality and graft efficiency. The specific objectives were: a) evaluate genetic and environmental factors affecting quality of seed in hybrids derived from interspecific crosses, b) improve grafting success through use of adhesives, and c) determine the genetic basis of graft failure between rootstock and scion.

Tomato is a model for grafting annual vegetables due to the importance of the crop and the extensive genetic resources available. To assess the potential to select for improved seed quality, experimental rootstocks were developed through pollination of cultivated

(Solanum lycopersicum L.) parental lines as female parents and 11 accessions of wild species as male parents. Seed quality was evaluated based on seed size (weight) and total germinability for each hybrid produced. Maternal effects and environment determined fruit set. Specific genotype combinations and environment determined seed yield. Seed size was mainly affected by genetic components, while seed germination was affected by

ii both genetics and environmental factors. Seed size can be used as selection criterion in breeding program for early selection of rootstock seed quality. To improve graft success, nine different tomato rootstocks were grafted using the traditional tube method of grafting and using adhesives. Despite wide variation across rootstock genotypes and grafting environment, grafting using adhesives resulted in higher grafting success. To determine the genetic basis of graft failure, molecular markers were exploited for quantitative trait locus (QTL) analysis. An advanced backcross population (BC3S4), derived from two S. lycopersicum parents (H7998 × OH881199), was evaluated for graft failure, using 10 plants from each family as rootstocks with the entire experiment repeated twice. The genotyping of the BC3S4 population was conducted with 78 polymorphic markers for the marker-trait analysis. A second population was developed as an F2 (H7998 × Ohio MR13) and evaluated for survival and scion height as a validation of the marker-trait associations. We identified potential regions of the genome affecting grafting failure or success on chromosome 9, and two QTL for scion height on chromosome 2 and 4. We were not able to validate these putative QTL due to the lack of overlapping markers across the two populations. The findings of this thesis work indicated that wide crosses could produce rootstocks with high seed quality. I identified key selection points in the breeding of new tomato rootstocks. At the same time, I demonstrated that control of the seed production environment is essential to obtain high quality seed from wide-cross hybrids. I also demonstrated that grafting success can be increased using adhesives in the process, and grafting failure appears to have a genetic basis, with failure associated with alleles from H7998.

iii

Dedication

Este trabajo va dedicado:

A mis abuelitas: Q.E.P.D. Rosa Mamani de Huarachi y benedicta Martínez de Morejon

A mis padres y familia que desde la distancia siempre me apoyaron

A Matthew Kost y la familia en Kost por su apoyo incondicional

Acknowledgments

I want to thank God for giving the opportunity to study at The Ohio State University.

Thanks to my parents, my siblings for all their support during some difficult times. Thank to my best friend and partner Matthew Kost for his support and patient during this two years of study. I would like to say thank to the Kost family for the amazing times in

Kansas.

I would also like to express my gratitude to my adviser, Dr. David Francis for his guidance and support. To my committee members, Dr. Pablo Jourdan and Dr. Matthew

Kleinhenz for their suggestions and guidance through this research. To Troy Aldrich for his valuable help in the establishment of the greenhouse studies and helping to take care my tomato plants and the process of pollination and emasculation.

I thank the support from many undergraduate students and Cassi for helping in the process of seed production and evaluation.

Finally, many thanks to our all amazing people that support me direct or indirectly in this process, you all have made these couple years an incredible experience.

Vita

July 28, 1986 ...... Puqui – Oruro, Bolivia

2008...... B.S. Agronomy

2011 to present ...... Graduate Research Associate, Department

of Horticulture and Crop Science, The Ohio

State University

Publications

Huarachi Morejón, N. (2008). Efecto del EM sobre el control de la mancha grasienta

(Mycosphaerella citri) en cítricos.Thesis.

Rodríguez GR, Moyseenko JB, Robbins MD, Huarachi Morejón N, Francis DM, van der

Knaap E (2010). Tomato Analyzer: A Useful Software Application to Collect Accurate and Detailed Morphological and Colorimetric Data from Two-dimensional Objects. JoVE

37.

Fields of Study

Major Field: Horticulture and Crop Science

Table of Contents

Abstract ...... ii

Acknowledgments...... v

Vita ...... vi

Publications ...... vi

Fields of Study ...... vi

List of Tables ...... viii

List of Figures ...... x

Chapter 1 : Introduction ...... 1

Chapter 2 : Genetic and environmental factors affecting seed quality of tomato rootstocks using wild species ...... 28

Chapter 3 : Cyanoacrylate adhesive improve grafting success between tomato rootstock and scion ...... 61

Chapter 4 : Analysis of Quantitative Trait Loci (QTL) associated with the failure of grafting tomatoes ...... 81

Chapter 5 : Conclusions ...... 104

References ...... 109

vii

List of Tables

Table 2.1: Female and male parents used in hybrid seed production...... 53

Table 2.2: Estimated variance components for female parents, male parents, experiment, and interactions on fruit set, seed yield, seed weight and seed germination...... 54

Table 2.3: Mean comparisons using Least Squares Means with adjustment for multiple comparisons using Dunnett’s test and control 578 (control 2) for evaluation of hybrid seed quality...... 55

Table 3.1: Analysis of variance for success rate of grafting Cherokee Purple scion onto nine different rootstocks using two different grafting methods...... 76

Table 3.2: Effect of rootstock genotype on success rates across all treatments at weekly intervals following the grafting process...... 77

Table 4.1: Polymorphic molecular markers used in the population derived from the cross

OH88119 × H7998 ...... 94

Table 4.2: Single Nucleotide Polymorphisms (SNP) molecular markers used in the population derived from the cross Ohio MR13 × H7998...... 98

Table 4.3: Single factor analysis for association in BC3S4 (OH88119 × H7998) population for grafting success at week 3...... 99

Table 4.4: Single factor analysis of association with F2 population for grafting success at week 1, week 2 and week 3 post-grafting...... 100

viii

Table 4.5: Single factor analysis of association with F2 population height of the scion at week 3 post-grafting...... 101

Table 4.6: Analysis of variance for survival mean along three weeks-post-grafting in three different grafting combinations...... 102

List of Figures

Figure 2.1: Effect of female parents on (A) Fruit set (%), (B) Number of seeds per fruit,

Figure 2.2: Effect of male parents on (A) Number of seeds per fruit, (B) Seed weight

(mg), and (C) Seed Germination (%)...... 57

Figure 2.3: Effect of Environment (Experiments) on (A) fruit set (%), (B) Number of seeds per fruit and (C) Seed germination (%)...... 58

Figure 2.4: Relationship between seed size (weight) and seed germination...... 59

Figure 2.5: Germination curves for seeds from eight tomato hybrids harvested at different stages of fruit maturation...... 60

Figure 3.1: Effects of rootstock genotype on grafting success across all five experiments and grafting methods (Adhesives and tube grafting methods)...... 78

Figure 3.2: Main effects of grafting method on grafting survival success across all five experiments and rootstock genotypes...... 79

Figure 3.3: Effects of two main effects (Rootstock genotype and grafting method) across experiments three, ...... 80

Figure 4.1: Phenotypic data distribution for grafting success at week-3 post grafting for

BC3S4 and F2 Population...... 103

Chapter 1 : Introduction

Tomato (Solanum lycopersicum L.) is an important vegetable consumed worldwide (Brown et al., 2005). World production of fresh tomato for 2009 was about

141 million tons planted on 4.5 million hectares in 144 countries (FAOSTAT, 2013).

From 1990 and 2004 global consumption of tomatoes increased ~ 4.5% each year

(Aherne et al., 2009). Fresh-tomato consumption represented about 76% of the total consumption during 1996 - 2009 (Branthome, 2010).

Soil-borne diseases and abiotic stress cause significant losses in yield every year

(Rivard and Louws, 2008). Many countries have reported substantial losses of yield due to soil-borne diseases, which could be controlled by grafting to disease resistant rootstocks (Alam et al., 1994; Hasna et al., 2009; Oda, 1999). In Bangladesh, root knot nematodes caused a 53% - 62% loss of yield (Ali et al., 1994). European greenhouses have reported yield losses of up to 75% due to corky root rot (Hasna et al., 2009). As a specific example, In Sweden corky root rot causes a 30% - 40% reduction of yield in greenhouse production (Hasna et al. 2009). In Japan, soil-borne diseases can cause loss of as much as 6% of the vegetables production (Oda, 1999).

Abiotic stresses limit production of many crops. These stresses include temperature (high or low), drought, light, atmosphere, nutrients, and salinity (Haferkamp,

1987; Criddle et al., 1997; Cramer et al., 2011). Estimates of the effect of abiotic stress on global agriculture suggest that up to 70% of crop production is affected by environmental constraints (Boyer, 1982; Cramer et al., 2011). According to FAOSTAT

(2006), only 3.5% of the worldwide land area is not negatively affected by any environmental factors during a cropping cycle.

The use of rootstocks offers many potential benefits, such as resistance to a wide range of pathogens in the soil, resistance to abiotic stress, and increased yield and fruit quality (Oda, 2002; Lee, 2003; Kubota et al., 2008). In the Netherlands, grafting was adopted for extending the harvest period (Lee, 2003). In the early 1990s, tomato rootstocks were not used commonly because grafting methods were inefficient and seed germination was low for interspecific hybrid rootstock (King et al., 2010). Adoption of rootstocks accelerated when interspecific crosses exhibited a high level of resistance to

Pyrenochaeta lycopersici (corky root rot) were introduced (Lee, 1994; Oda, 1995). Later it was observed that these hybrids also had increased cold tolerance and vigor (Lee, 1994;

Oda, 1995). In 1996 resistance to Fusarium oxysporum f. sp. Radices-lycopersici (crown root rot) was added to the rootstock resistance profile. Germination was improved due to commercial seed treatments, and the rootstock market grew to its current size, most of the hydroponic vegetables are grafted (King et al., 2010). Introduction of the ‘Japanese”

(tube) grafting method to Europe led to large scale grafting at significantly lower cost

(Oda, 1994; Lee, 1995). The use of grafting has the potential to replace the use of methyl

2 bromide to control disease (Miguel et al., 2004; King et al., 2010). Currently, interest in breeding for improved rootstock is increasing rapidly in the private sector due to cost effectiveness and expanded diseases resistance (King et al., 2010).

Objectives

The overall objective of this research was to address genetic and environmental factors limiting grafting technology using tomato as a model species. I measured the relative effect of gentic and environmental factors affecting seed production and quality when using wild relatives of the cultivated tomato for rootstock breeding. In addition, I evaluated the efficacy of adhesive in increasing grafting success between rootstock and scion. The central hypothesis of this research was that variance partitioning could be used to identify limitations in grafting technology and identify key control points for the improvement of rootstock seed, grafting technique, or genetics. The specific objectives of the research were to:

1. Evaluate genetic and environmental factors affecting quality of seed in

hybrids derived from interspecific crosses.

2. Improve grafting success using adhesives.

3. Assess the genetic basis of graft failure between rootstock and scion.

Use of grafting for tomato production

The practice of grafting in annual fruits and vegetables started in the late 1920s and early 1930s as a means to control soil-borne diseases of watermelon (Oda, 2002).

The gourd (Lagenaria siceraria) provided a compatible rootstock to control fusarium wilt (Fusarium spp.) in watermelon (Tateishi, 1927). In the United States, small farmers grafted vegetables onto weeds 60 years ago or more (Kubota et al., 2008). In the early

1960s, grafting was introduced into commercial production of tomato in many parts of the world (Harrison and Burgess, 1962; Bravenboer and Pet, 1962; Smith, 1966; Upstone and Finney, 1966; Smith, 1968 and Lee and Oda, 2003). In the early 1990s, grafting was re-introduced to European countries as a tool to control root diseases of soil and in hydroponic systems (Oda, 2002). More recently, use of the technique has expanded to the

Middle East, North America, Central America, and other parts of the world. Currently, grafting of annual vegetables has been applied to different degrees by different agricultural sectors. For example, nearly all-hydroponic tomato production is currently from grafted plants (Lee, 1994; Kubota, 2008; King et al., 2010). In contrast, few soil- based production systems outside of Asia have adopted the technique (Kubota et al.,

2008). However, interest is high and the market for grafted plants appears to be expanding in all sectors (Lee et al., 2010). Grafting is widely used in cucurbitaceous crops (cucumber, melon, and watermelon) and Solanaceous (tomato, eggplant, and peppers) (Oda, 2002).

Control of soil-borne diseases

Chemical control

Methyl bromide (CH3Br) is one of the most widely used pesticides in the world

(Shorter et al., 1995), and tomato production has the highest use, with 23% of the total used in agriculture (Braun and Supkoff, 1994; Schafer, 1999). In 1992, the parties of the

Montreal Protocol listed methyl bromide as an ozone-depleting compound (Braun and

Supkoff, 1994; Schafer, 1999 and USDA, 2000). In addition, this product is thought to cause health problems in exposed humans and animals (Calvert et al., 1998; Schafer,

1999; Barry et al., 2012). Recognition of negative environmental and health effects drove legislation to phase out use of methyl bromide completely by 2005 in developed countries and by 2015 in developing countries (Schafer, 1999 and USDA, 2000). Despite this legislation, methyl bromide is still widely used in many countries due its effectiveness and convenience for controlling soil-borne diseases in many crops (Akca et al., 2009; Barry et al., 2012; Walse et al., 2012).

Non-chemical control

Soil solarization is one alternative to methyl bromide. Solarization controls efficiently many pathogenic fungi, bacteria, weeds, and nematodes in vegetables and ornamental plants (Katan et al., 1976; Katan, 1981; Braun and Supkoff, 1994; Stapleton

2000, Tjamos et al., 2000; McSorley et al., 2009). Soil solarization involves heating the soil with sunlight or ultraviolet radiation under a transparent polyethylene cover, leading to lethally high temperatures for the pathogens (Katan et al., 1976; Katan, 1981;

Stapleton and DeVay, 1986; Stapleton, 2000; Tjamos et al., 2000; Wang and McSorley,

2008). This technique gained popularity in hot regions of Israel to control Verticillium dahliae and Fusarium oxysporum f. sp. lycopersici (Katan et al., 1976; Katan, 1981). It was later expanded to other countries, including cooler areas. For further information, see reviews (Stapleton and DeVay, 1986; Braun and Supkoff, 1994; Tjamos et al., 2000;

Stapleton, 2000; Ioannou, 2001; Wang and McSorley, 2008; McSorley et al., 2009). In developing countries, solarization has become common for disinfestation of seedbeds, containerized planting media, cold frames, and tomato supports (Stapleton, 1998;

Stapleton et al., 1999; Stapleton, 2000). Soil solarization is more effective in controlling soil-borne diseases and pests when combined with other methods including chemicals, organic amendments, biological control agents, and grafting (Braun and Supkoff, 1994;

Gamliel and Stapleton, 1997; Stapleton, 2000; Tjamos et al., 2000; Ioannou, 2001; Bello et al., 2002). The highest use of solarization appears to be in protected crops (Cartia,

1998; Stapleton, 2000). Despite the efficiency, ease, and safety, farmers have not broadly adopted soil solarization as an alternative to methyl bromide due to its labor-intensive nature, dependency on a hot climate for effectiveness, potential interference or delay of planting, and use of plastic film must be discarded (Braun and Supkoff, 1994; Stapleton,

2000; Tjamos, 2000). This last obstacle remains an issue despite successful recycling of the plastic film (Katan and DeVay, 1991). In addition, there are some heat-tolerant pests

(e.g. Marophomina phaseolina, Synchitrium lagenariae and S. trichosanthidis), which cannot be controlled by solarization (Katan, 1981; Katan, 1987; Stapleton and DeVay,

1995; Stapleton, 1998).

Genetic control

Another main alternative to methyl bromide is using resistant varieties, which may be the solution for many soil-borne diseases and pests (Braun and Supkoff, 1994;

Kaskavalci et al., 2009). Most contemporary hybrid tomato varies have multiple resistance to soil-borne pathogens, such as Verticillium wilt (race 1), fusarium wilt (F. oxysporum f. sp. lycopersici, races 1 and 2), and root-knot nematodes (Smith, 1944;

Ioannou, 2001; Kaskavalci et al., 2009; Labrada, 2008). One of the main disadvantages of resistance breeding is that most genes are only effective against a single pathogen or one race of a pathogen (Braun and Supkoff, 1994; Besri, 2003). In addition, resistant varieties may not always be available with the complete complement of quality attributes needed for a specific market. Many landrace or indigenous cultivars lack the breeding history of contemporary varieties and therefore lack resistance. Additional research is required to create resistant and marketable varieties through conventional breeding or genetic engineering techniques. Grafting onto resistant rootstock has the potential to address concerns about chemical use while decreasing production limitations. Furthermore, this technology is considered as a component of an Integrated Pest Management (IPM), which includes many other methods of control e.g. solarization (Ioannou, 2001), biofumigation, intercropping (Kaskavalci et al., 2009). Grafting also avoids loss of time and loss of important traits by introducing resistance genes into modern cultivars.

Disease resistance has been a driving influence for the adoption of rootstocks for annual fruit and vegetable production. The major soil-borne diseases affecting field tomatoes around the world are bacterial wilt of tomato (Ralstonia solanacearum),

7 fusarium wilt (Fusarium oxysporum Schlecht. f.sp. lycopersici (Sacc.) Snyder & Hansen, race 1, race 2, and race 3), fusarium crown and root rot (Fusarium oxysporum f. sp.radicis-lycopersici (FORL)), verticillium wilt (Verticillium dahliae, races 1 and 2),

Clavibacter michiganense (bacterial canker), Orobanche (Orobanche ramosa), Corky root rot (Pyrenochaeta lycopersici), and root knot nematodes (Meloidogyne spp.) (Rivero et al., 2003; Besri, 2008; King et al., 2008; Rivard and Louws, 2008; Rivard et al., 2010).

Over time, grafting has been expanded to control numerous soil-borne diseases. In Japan and Korea, the objective of grafting is now to control bacterial wilt, fusarium wilt, corky root, verticillium wilt and nematode in tomato (Lee, 1994; Oda, 1999; Lee, 2003; Oda,

2007).

Various rootstocks have been created for cucurbitaceous plants. Cucurbits are commonly grafted onto an interspecific hybrid between Cucurbita maxima and Cucurbita moschata (Shintozwa) (Lee, 1994; Lee et al., 2003; Oda, 2007). Other cucurbitaceous rootstocks (Watermelon, Cucumber and Melon) have been reviewed (Oda, 1999; Lee,

2003; Oda, 2007; Lee et al., 2010). Grafting in annual crops was initiated with cucurbit annual vegetables and by now, these crops are the most grafted vegetables, especially in countries where the land use is intensive.

In Solanaceous vegetables, S. Lycopersicum or interspecific hybrids (S. lycopersicum × S. habrochaites) are widely used as rootstock for tomato. Early reports from India demonstrated that ‘CRA 66’, a wild eggplant rootstock, could reduce bacterial wilt incidence in tomatoes, increasing survival rates from 0% (control) to 100% of the grafted plants at last harvest time (Tikoo, 1979). For more details about other solanaceous

8 vegetables (Eggplant and Sweet pepper) see previous reviews (Yamakawa, 1982; Lee,

1994; Oda, 1999; Lee et al., 2003; Oda, 2007 and Lee et al., 2010). Corky root was controlled using ‘Beaufort’ tomato rootstock, where the grafted plants were healthier and more uniform compared to non-grafted plants (Hasna et al., 2009). Southern blight

(Sclerotium rolfsii) and root-knot nematode (Meloidogyne spp.) can also be minimized by grafting (Rivard et al., 2010). The interspecific hybrid rootstock ‘Multifort’ (De Ruiters

Seeds, Bergschenhoek, The Netherlands; Now Monsanto, St. Luis, Mo.) shows tolerance under severe infestation of root-not nematode and the hybrid rootstock ‘Survivor’ (Takii

Seeds, Salinas, CA) significantly reduced root galling compared to ‘Multifort’ (Barrett et al., 2012). Given the effectiveness of rootstocks to control disease of the root system, it is not surprising that grafting is used commercially to control soil-borne diseases under different environments (Di Gioia et al., 2010; Rivard et al., 2010 and Louws et al., 2010).

Rootstock tolerance to abiotic stress

Cultivated tomato lacks variation and specific adaptation to low temperatures

(Paul et al., 1984; Nieuwhof et al., 1999; Schwarz et al., 2010). Temperature is an important environmental factor that affects tomato production (Hansen et al., 1994;

Venema et al., 2008; Schwarz et al., 2010; Venema et al., 2010). For soil temperature, the critical root temperature for tomato range from 15 °C to 34 °C (Martin and Wilcox, 1963;

Zijlstra and DenNijs, 1987). Low root temperature affects the productivity of the plant by

(1) delaying initiation of growth, (2) restricting water movement to roots, (3) decreasing permeability of the membrane on the root surface, and (4) delaying opening of the

9 stomata with a resulting reduction of daily photosynthesis (Haferkamp, 1987; Bloom et al., 2004). Wild species offer potential sources to increase the genetic variation and phenotypic variation for low temperature tolerance in cultivated tomato (Venema et al.,

2008; Bloom et al., 2004). For example, Solanum habrochaites grows in the Andes (from

500 up to 3300 meters above sea level) and grows at chilling temperatures that are detrimental for cultivated tomatoes (Vallejos and Tanksley, 1983; Bloom et al., 2004;

Venema et al., 2008). The cold tolerant accession LA 1777 (S. habrochaites) was used as a rootstock under two temperatures 25 °C and 15 °C (Venema et al., 2008). This study demonstrated that it is possible to graft a cultivated tomato onto a wild species without creating hybrids. However, the study did not evaluate the yield or other characteristics of the grafted plants. The only parameter evaluated was root development under low temperature, and grafted plants onto S. habrochaites LA 1777 had better developed roots than the self-grafted plants (Venema et al., 2008). Thus, grafting could combine a marketable tomato with rootstocks tolerant to low or high temperatures, thereby stimulating better growth and development under either thermal or chilling conditions.

High temperatures (greater than 35 °C) can reduce fruit set in tomato; physiologically it inactivates the metabolism of the plant and causes a reduction of foliar- biomass (Haferkamp, 1987; Hansen et al., 1994; Rivero et al., 2003; Schwarz et al.,

2010). For grafting, the use of more distantly related species as a source of new variation becomes possible. For example, wild eggplant rootstock for tomatoes has allowed plants to cope with hot-wet seasons, including flooding, waterlogging and high temperature

(Black et al., 2003; King et al., 2010). Rivero et al., (2003) grafted tomato onto a heat

10 tolerant rootstock (L. esculentum cv. ‘RX-335’), and demonstrated g resistance to high temperatures which resulted in superior plant biomass in grafted plants compared to non- grafted plants.

The soils of many production regions around the world are affected by high concentrations of salt (Yamaguchi and Blumwald, 2005; Munns, 2002; Martinez-

Rodriguez et al., 2008; FAOSTAT, 2009; He et al., 2009; Estañ et al. 2005; Colla et al.,

2010; Venema et al., 2010). Salinization is often caused by irrigation with poor-quality of water (Estañ et al., 2005; Santa-Cruz et al., 2002; Fernandez-Garcia et al., 2004; Venema et al., 2008). Because salinity affects many physiological and biochemical aspect of plants including ion balance, ion toxicity, and hyperosmotic stress (Santa Cruz et al.,

2002; Cuartero et al., 2006; Colla et al., 2010; Fernandez-Garcia et al., 2004), lands with high salt concentrations are no longer useful for agriculture. As with heat and cold resistance, wild species offer a potential source of tolerance to salt stress (e.g. L. pennellii accession LA0716). A genetic study based on cultivated tomato and a wild species

(LA0716) populations estimated the proportion of genetic to environmental variance

(heritability) for salt tolerance to be in the range of 0.3 - 0.45, suggesting that salt tolerance can be improved through genetic selection (Saranga et al., 1992). Wild tomatoes have salt tolerance genes, but it may be difficult to move this trait due to the complexity of the genetic architecture, which appears to be a quantitative multigenic- controlled trait (Jones, 1985; Saranga et al., 1992; Santa-Cruz et al., 2002; Estañ et al.,

2005). Rootstocks offer a potential solution to reduce the negative effects of high salt concentration and avoid damage by salinity (Cuartero et al., 2006; Martinez-Rodrigues et

11 al., 2008; Venema et al., 2008). Santa-Cruz et al. (2002) suggested that grafting might be a valid technique for tomato under saline conditions. Estañ et al. (2005) and Martinez-

Rodriguez et al. (2008) demonstrated that grafting onto appropriate rootstock could reduce ionic stress. Grafting provides an alternative way to confer salt tolerance on the scion through the rootstock.

The high chemical inputs used in contemporary agriculture (such as fertilizers, herbicides, and other chemicals), has resulted in high concentrations of heavy metals, pesticides, and nitrates in the soil (Savvas et al., 2010). Depending on the rootstock genotype, grafted plants may grow under conditions of heavy metal contamination by minimizing transport of heavy metals (Cu, B, Cr, Cd) into the fruit or leaves, using a mechanism of restricted uptake (Estañ et al., 2005; Arao et al., 2008; Savvas et al., 2009;

Savvas et al., 2010).

This review of the existing literature suggests that grafting has many benefits, from increase yield and fruit quality, disease resistance, abiotic stress tolerance, and ion and nutrient uptake efficiency. Rootstocks can improve resistance to diseases of the root system and they can impart tolerance to abiotic stress. Rootstocks can also restrict or enhance the uptake of specific nutrients compared to non-grafted plants, with some combinations having higher ability to take up nutrients and others restricting specific transport (Leonardi and Giuffrida, 2006). Given the diversity of potential uses, there is a need to design knowledge-based strategies of breeding rootstocks with high seed quality, and increase grafting success using new methods of grafting. In addition, it is important to understand the genetic basis of rootstock-scion grafting failure.

Goals of breeding tomato

Over time, the goals of tomato breeding have changed (Grandillo et al., 1999).

In the 1970s, the most important trait to breed for was increased yield. In the 1980s, extending tomato fruit shelf life was a priority. At present, a goal for many tomato breeding programs is sensorial and nutritional quality (Grandillo et al., 1999; Bai and

Lindhout, 2007). Quality is a general term must be applied to specific markets, and includes physical properties that contribute to flavor and compounds that provide nutrition and health benefits (Rouphael et al., 2010). With shifting breeding goals, it is sometimes difficult to combine traits. Thus, grafting offers an opportunity to combine resistance breeding with multiple quality traits.

Studies in many locations and countries have reported that yield was increased for a variety of rootstock-scion combinations. These increases are reported as an increased number of fruits per plant and fruit weight (Turhan et al., 2011). For example, the variety ‘Lemance’ grafted onto ‘Beaufort’ rootstock showed higher yield compared to non-grafted plants (Pogonyi et al., 2005). Plants grafted onto two different rootstocks

(‘He-man’ and ‘Primavera’) produced more fruit; increasing yield up to 32.5%

(greenhouse) and 11.1% (field) (Khah et al., 2006). When ‘Cecilia F1’ was grafted onto three rootstocks (‘Beaufort’, ‘He-man’ and local Syria tomato); productivity increased up to 21% (Mohammed et al., 2009). The combination ‘Beaufort’ used as the rootstock and

‘Monroe’ as the scion showed an increase in yield from 3.37 kg.plant-1 non-grafted to

4.46 kg.plant-1 grafted plants (Kacjan-Marsic and Osvald, 2004). Yield increases close to

20% compared to non-grafted plants have been reported (Gebologlu et al., 2011). In

Sweden, plants grafted onto ‘Beaufort’ rootstock showed yield increases and more uniform fruit size (Hasna et al., 2009). Gebologlu et al., (2011) showed data that yield increased from 13.85% to 37.72%, compared to non-grafted plants. However, reported yield increases are not universal. For example, ‘Belle’ as a scion grafted onto ‘PG 3’ rootstock had lower yield (Kacjan-Marsic and Osvald, 2004). In our work with 36 rootstocks grafted to Celebrity and grown in three, soil-based, open field production trial, grafted plants had an increase in yield relative to checks. However, only ‘Beaufort’,

‘Maxifort’, and ‘VFNT cherry’ of these were statistically significant, and several had significantly lower yields relative to non-grafted or self-grafted controls. Delayed maturity has also been observed on grafted plants.

Multiple reports have shown effects of grafting on fruit quality; some researchers have found positive effect, other negative effects of grafting on fruit quality.

For example, when “Beaufort” was used as rootstock and ‘Lemance’ as a scion, soluble solids °Brix, carbohydrates, and organic acids were lower in grafted plants than those from non-grafted ones (Pogonyi et al., 2005). On the other hand, the rootstock cv. ‘Radja’ increased fruit quality (soluble solid content and titratable acid) of fruit from scions

(Flores et al., 2010). A hybrid rootstock based on Solanum cheesmaniae × cultivated tomato improved the soluble solids content and titratable acid (Flores et al., 2010). The increase in total dissolved solids and decrease in lycopene content was reported

(Mohammed et al., 2009). Turhan et al. (2011) reported no effects of grafting on lycopene and pH content. In a study by Gebologlu et al. (2011), vitamin C, water-soluble dry matter, and titratable acidity were not significantly affected by different rootstock

14 used. Moreover, other studies have failed to observe a change in lycopene content on grafted plants under field or greenhouse conditions (Rouphael et al., 2010). Some studies reported negative effects of grafting. A reduction in vitamin C and phenolic content was reported in grafted plants (Vrcek et al., 2011). Similar to yield, studies indicate a range of responses from improved quality to negative effects. Thus, establishing which combinations of rootstock, scion, and environment improve desired traits should be a key focus of investigation.

Hybrid seeds

The use of hybrid seeds has increased from the late 1960’s to the current situation where hybrids dominate markets due to advantages compared to open pollinated varieties. Hybrid cultivars have vigor, uniformity, diseases resistance, stress tolerance, and desirable horticultural traits (Shull, 1911; Maxon-Smith, 1966; Rick, 1980; Opeña et al., 2001). Charles Darwin (1876) first observed hybrid vigor or heterosis in many plant species including tomato. Later, Shull defined the concept of heterosis as superior performance of a hybrid genotype in maize resulting from the cross of two inbred lines

(Shull, 1908; Shull, 1911).

In tomato, studies on heterosis were initiated almost at the same time as in maize (Hedrick and Booth, 1907; East and Hayes, 1912). Heterosis has not been a driving force in tomato breeding. Rather, hybrids were used by the seed industry to protect germplasm from misappropriation. Hedrick and Booth (1907) conducted a study of hybrid vigor on tomato by performing crosses of different varieties and growing the

15 progeny next to both parents. They observed that the F1 progeny had higher performance in terms of growth, leaf color, and fruit production. Although heterosis in tomato is widely documented in the academic literature, with many studies having demonstrated high levels of heterosis for yield and yield component (Kravchenko, 1990; Suresh et al.,

1995; Atanassova and Georgiew, 2007), there are few documented examples of heterosis for economic traits using commercially relevant germplasm. The disconnect between observed heterosis in growth and marketable yield is not an issue in rootstock breeding as heterosis in growth of the root system can be exploited, despite the fact that most rootstocks do not produce marketable fruit. Studies also demonstrated that heterotic performance might enhance the ability of plants to cope with an unfavorable environment

(Khanna-Chopra et al., 1993), and increases fruit quality (Patil and Patil, 1988) and uniformity (Yordanov, 1983; Atanassova and Georgiew, 2007). Thus, the use of hybrid seeds in grafting technology has been expanted due to its high performance in different environments.

Seed quality

The definition of seed quality consists of several components, including, analytical, genetic or species purity, cultivar purity, germination capacity, vigor, size, uniformity,moisture content, and freedom from weeds an diseases (Desai, 2004). High- seed quality represents high profits for farmers and growers. Therefore, seed producers have responsibility to produce high seed quality.

As commonly described in tomato, the use of seed directly from wild species or from hybrid seed derived from crosses to wild species may negatively affect seed quality.

In tomato, the standards of high quality seed are genetic purity of 99%, germination rate above 85%, and moisture content less than 8% (International Seed Federation, 2013).

Wild species and hybrids (wild × cultivated tomato) were evaluated for uniform germination, and the variation was large (Ibrahim et al., 2001). The germination percentage of wild Solanum species and hybrids ranged from low rates up to 86%. Large variation in germination exists between wild species and hybrids (wild × cultivated tomato), ranging from 8% to 86% (Ibrahim et al., 2001). These observations are not surprising as part of the domestication process involves selection for uniform germination.

The use of F1 hybrids as tomato rootstock has become common in tomato grafting. Early rootstock hybrids of tomato were produced using L. esculentum as a female and L. hirsutum var ‘Glabratun’ as a male (currently S. lycopersicum and S. habrochaites, respectively) (Bravenboer and Pet, 1962). In 1966, ‘GCR 66’ was issued as a new parent for rootstocks (Maxon-Smith, 1966). This new parent came from the combination of two lines with specific traits (Green stem and potato leaf positional sterile). Rootstocks derived from ‘GCR 66’ contained resistance to verticillium wilt (Ve) and fusarium wilt, race 1, which comes from the donor parent S. habrochaites (Maxon-

Smith, 1968).

Reproductive isolation barriers

Hybrids from wild species and cultivated species increase the ability to combine disease resistance, but may reduce seed quality. Wild species have poor seed germination rate, poor uniformity in seed germination and strong seed dormancy (Ibrahim et al. 2001).

Gibberellic acid (GA3) can be used to break the dormancy and encourage uniform seed germination; however, this technique is difficult to apply at the grower level.

Wide crosses offers the possibility of combining specific traits such as resistance to biotic and tolerance to abiotic stress. At the same time, seed quality may suffer decay due to reproductive barriers. Both genetic and ecological barriers have a role. Ecological isolation, temporal and seasonal isolation, mechanical isolation, may prevent fertilization in nature (Stebbins, 1958; Sano and Kita, 1978; Hadley and

Openshaw, 1980). Genetic barriers may result in hybrid sterility, when two species may succeed in producing a viable even vigorous F1, but the structural differences between parental chromosomes causes hybrid sterilization (Hadley and Openshaw, 1980;

Rieseberg, 1997). Although wide crosses offer potential advantages of hybrid vigor, genetic barriers may also lead to poor performance of hybrids resulting from such crosses. Thus, selection based on seed quality must be investigated to create new hybrids using wild relatives as a germplasm variation.

Grafting methods

Many grafting methods have been developed over time and specialized for each crop (Oda, 1999). In tomato, “cleft grafting” has high success, as defined by the number

18 of plants that survive the grafting process (Oda, 1995), but this method is labor intensive.

In the late 1990s, a new grafting method was introduced in Japan, called “tube grafting” or the “Japanese method” (Oda, 1995). Using this new method it is possible to graft younger plants at a rate of 300-500 grafts per hour, depending on the worker’s skill

(Kubota et al., 2008). The advantage of tube grafting is that it is faster than cleft grafting because it requires only one cut, instead of three cuts (Oda, 1994; Oda, 1995; Oda, 1999).

The tube grafting method also frees space in the healing chamber because smaller plants may be grafted. The use of tube grafting has allowed for the creation of robotic systems, which accelerate the grafting processes and have success rates of over 93%, (Oda, 1995;

Chen et al., 2010). Despite of its advantages of speed and space, this method is restricted to similar stem size of rootstock and scion. Sometimes it is difficult to obtain uniform seedlings without using an extra seeds.

In addition, adhesives such cyanoacrylate have been used in grafting. Oda and

Nakajima (1992) reported the used of adhesives to graft Chinese cabbage onto turnip.

Adhesives have been used to graft cucumber, eggplant, and grape (Oda, 1995). The use of cyanoacrylates in agriculture parallel those in medical applications to close wounds

(Quintino and Pires 2004; Bozkurt and Saydam, 2008), in veterinary medicine (Andrade et al., 2001; Albuquerque et al., 2006; Endo et al., 2007), in dentistry (Grisdale, 1998;

Leggat et al., 2004), and in aesthetic facial plastic surgery (Kamer et al., 1989). Using cyanoacrylates to join biological tissues is both inexpensive and easy. Using adhesives could help to improve efficiency and may be utilized in the automation of the grafting process using robotic systems.

Grafting failure

Many woody ornamentals, most fruit trees, and recently annual vegetables are propagated by grafting. Sometimes the graft union fails, resulting in the main stem breaking off, dieback, poor growth or death of the grafted plant (Hartmann et al., 1990;

Andrews and Marquez, 1993).

The success or failure of grafting depends on various factors including taxonomy, environment, availability of oxygen and water, physiological stage of rootstock/scion, herbicide toxicity, the skill of the grafter, mechanical damage of the graft union, and graft incompatibility (Andrews and Marquez, 1993). Graft failure occurs when the rootstock and scion are partially or completely separate from each other. However, it is often difficult to establish the exact cause of the graft failure. Grafting failure often occurs at early stages (Martinez-Ballesta, 2010), which can be classified into three categories: (1) failed bud growth, (2) mechanical obstruction at the union, and (3) abnormal union structure (Herrero, 1951; Andrews and Marques, 1993). Grafting failure may also occur in later stages such as the fruiting stage.

In plants, the more closely related rootstock-scion are (taxonomy), the better the chances for the graft to be successful (Copes and Oliver, 1970). A successful graft begins a sequence of events during the healing process including: (1) callus proliferation from rootstock and scion, (2) callus bridge formation (3) vascular differentiation, and (4) production of secondary xylem and phloem (McCully, 1983; Moore, 1984; Andrew and

Marquez, 1993; Kawaguchi et al., 2008). However, grafting can fail at healing process or when plants are fully-establish with limited warning. When tomato cultivars and wild (S.

20 habrochaites) species were used as a rootstock in eggplant grafting, the grafting success was low (Gisbert et al., 2011), this might be attributed to taxonomic incompatibility between graft partners or other reasons.

Grafting incompatibility is often misused as a synonym of “grafting failure” or vice versa. Many definitions of graft incompatibility are based on physiological and biochemical processes (Andrew and Marquez, 1993) or physiological intolerance at the cellular level (Moore and Walker, 1981). Often grafting incompatibility is used to mean failure or interruption of the development of the new grafted seedling. This interruption can occur for number or reasons, including taxonomic differences, cellular recognition and diseases present in the grafter partners (Garner, 1979; Andrew and Marquez, 1993).

However, in the literature both terms are used as a synonym, thus, define that grafting failure can be caused by grafting incompatibility, and this former has specific reasons.

The normal growth of a grafted plant may be interrupted at any stage of development due to incompatibility between scion and rootstock. Incompatibility could be directly related to undergrowth or overgrowth of the scion relative to the rootstock

(Lee, 2007). The rootstock is the portion of the plant that controls the uptake, synthesis, and translocation of water and minerals from the soil and the scion must be able to transport and use what the rootstock delivers (Lee and Oda, 2003). The perfect combination of these parts results in a successful plant that can respond to both abiotic and biotic stress in a given environment without decreasing yield or fruit quality.

In grafting tomatoes, one of the main limitations is thought to be Tobacco

Mosaic Virus (TMV) infection. There are three different resistant alleles to TMV: Tm-1,

Tm-2 and Tm-2a. Tm-1 is symptomless or tolerant, Tm-2a confers a hypersensitive response, and Tm-2 is intermediate (Hall, 1980; Yamakawa, 1982; Oda, 1995; King et al., 2010). In 1981, Yamakawa suggested different combinations of these alleles for tomato grafting to avoid problematic combinations. When the rootstock and the scion have the Tm-1 allele, the probability of graft to success is higher. The same is true with

Tm-2 and Tm-2a; both the scion and rootstock must have the same allele. Problems could occur when the rootstock has Tm-1 and the scion Tm-2 or vice versa. The most problematic combination could be when the rootstock has Tm-2a and scion has Tm-1. In this case, the virus may be transmitted from the rootstock to the scion inducing a systemic necrosis. In addition, the graft may be lost if the rootstock has Tm-1 and scion has Tm-2a

(Yamakawa, 1982; Oda, 1995; Morra, 2004; King et al., 2010). Therefore, it is important to understand the different mechanisms that may affect graft failure.

Alternatively, grafting compatibility is the ability of two plants (scion and rootstock) to grow successfully and reproduce as a single plant after they are joined

(Hartmann et al., 1990; Santamour and Frank, 1988). Marketable rootstocks must be compatible with scions without causing negative effects on yield or fruit quality. Many of the tomato rootstocks available in the market originated from pre-existing germplasm; few were developed by making crosses to generate variation and selecting superior genotypes, especially using multiple S. habrochaites wild lines (King et al., 2010). The factors that determine how a rootstock and scion interact to produce a strong graft are still not well understood. Most of the studies focused mainly on the physiological and biochemical factors of the graft formation, especially in tree species (Moore, 1983;

Andrew and Marquez, 1993; Kawaguchi et al., 2008). For that reason, it is important to investigate the genetic basis that causes graft failure, using tomato as a model plant for annual grafted crops.

Grafting quality and efficiency depend upon many factors, but the rootstock uniformity is a major contributor. The uniformity of the rootstock depends upon seed germination, which is affected by seed quality. In addition, grafting failure decreases the efficiency of the technology, and it depends on many factors, from the environment to the genetic level of failure. Therefore, it is important to answer the following questions (1)

Can genetic selection improve rootstock seed quality? (2) Can the environment of seed maturation improve rootstock seed quality? (3) Can the use of adhesives increase grafting success? finally (4) Do rootstock and scion failure have a genetic basis? The following research will aim to answer these questions.

QTL-analysis

In the last decades, genomic technology advanced in gigantic steps on creating techniques and strategies, to understand the molecular level of the plant function

(Somerville and Somerville, 1999; Hamilton and Robin Buell, 2012). A Quantitative

Trait Locus (QTL) is a region of the genome containing an allelic difference that causes a change in phenotype. Many agriculturally important traits exhibit complex genetic architecture, including phenotypes including seed yield, plant architecture and many abiotic tolerances (Burr et al., 1983; Lamkey and Lee, 1993).

Molecular markers are signs along the organism DNA that indicates genetic traits or genetic differences i.e. polymorphisms, which distinguish alleles in a population.

Multiple types of markers have been created from hybridization based, polymerase chain reaction, sequencing and DNA array (Agarwal et al., 2008; McCouch et al, 2002;

Rafalski, 2002; Staub et al., 1996; Phillips and Vasil, 2001). Microsatellite or simple sequence repeat (SSR), insertion/deletions (INDELS) and single nucleotide polymorphism (SNP) are sequence-based DNA markers. These makers have been widely used in crop breeding, in creation of molecular maps (genetic and physical maps) and other genetic studies and applications (Litt and Luty, 1989; Weber and May, 1989;

Varshney et al., 2007; Phillips and Vasil, 2001, Robbins et al., 2009). SNPs are the most common type of sequence variation in plant species (Lipshutz et al., 1999; Rafalski,

2002; Baird et al., 2008), in tomato this type of markers has been extensively applied in many studies and SNP arrays have created as well (Sim et al., 2009; Robbins et al., 2011)

QTLs with large effects on phenotypic variation are the easiest to identify and analyze. However, not all QTL have large effects, furthermore, most QTL have small average effects on complex traits (Mackay, 2001). Thus, while theory and experiment suggest that a large fraction of the variation of many phenotypes will be explained by

QTL with smaller effect sizes (Fisher, 1930; Paterson, 2010; Shao et al., 2008; Yang et al. 2011), our current understanding of complex traits is based primarily on analyses of

QTL with the largest effect sizes.

There are many factors affecting the efficiency of QTL mapping including (1) number of genes involved in the trait under control, (2) heritability of the trait (higher

24 better) (3) type and size of mapping population (4) density and coverage of markers, and

(4) statistical methodology applied (Lippman et al., 2007). The structure of population determine their effectiveness for decipher the genetic component of a trait and their application in plant breeding. Traditional population structures for genetic studies require parents that differ widely in the trait with the effectiveness of a population determined by the separation of parents (segregation) (Fisher, 1919). There are multiple methods for

QTL mapping studies to detect QTL. Single Marker Approach, is often referred as the single factor analysis of variance, and has been used widely, mainly with isozymes

(Tanksley et al., 1982). The methodology applied in a QTL study to measure a trait should be objective and reproducible.

Rationale and significance

The production cost of grafted tomato seedlings is high. Currently, the price of a grafted seedling ranges from $0.90 to $1.12 per plant (Kubota et al., 2008; Rivard et al.,

2010). The cost of grafted plants is due to the high cost of rootstock seed, the intensive work needed to graft, efficiency of the procedure relative to success, and post-graft care

(Lee, 2003; Kubota et al., 2008; Lee et al., 2010; Rivard et al., 2010). My research addressed seed quality and graft success in an attempt to identify key steps where efficiency could be increased. Previous studies have indicated that rootstock germination is low (Smith, 1966) and physiological incompatibility between scion and rootstock add cost to the system (Lee and Oda, 2003). However, there are many potential benefits of grafting including increased yield, extended growth, disease resistance, and low and high temperature tolerance (Lee, 2003; Kacjan-Marsic and Osvald, 2004; Mohammed et al.,

2009; Savvas et al., 2010; Rivard et al., 2010). Thus, addressing limitations to the production of grafted plants is desirable.

It is difficult to address some limitations due to industry trade secrets, inexperience with wild species used to produce rootstock, and the biology of grafting.

Information about flowering conditions, day-length sensitivity, mating system, and the native habitat is available for many wild accessions (TGRC, 2012), but translating this information into a production setting will require effort. Remaining limitations include issues of seed quality and germination, and graft quality and efficiency (Ibrahim et al.,

2001; Gisbert et al., 2011).

This research identified key restrictions in the process of developing new rootstocks and grafting rootstock to scion. Rootstock seed quality can be improved through a combination of genetic selection and manipulation of the seed production environment. Similarly, graft success can be improved through a combination of genetic and environmental manipulation. As a result of this research, I identified specific crosses with high seed quality. Similarly, I identified specific rootstock-scion combinations with high rates of success. I also demonstrate the improved success of grafting through use of cyanoacrylate adhesives. Finally, studies initiated to dissect the genetic basis of graft compatibility suggest a genetic component, but have been inconclusive at revealing the underlying genetic architecture.

Chapter 2 : Genetic and environmental factors affecting seed

quality of tomato rootstocks using wild species

Abstract

Rootstocks are currently used in annual vegetable production to control soil-borne diseases, replace fumigation, and to impart vigor. The goals of this study were to evaluate the genetic and environmental factors affecting quality of hybrid tomato seeds. We developed a wide variety of hybrid tomato rootstocks based on cultivated parental lines as female parents (Solanum lycopersicum L.) and a range of wild species as male parents.

Environmental effects were assessed through replicated experiments in two different seed production seasons (winter-spring) and (summer-autumn). Fruits were harvested at three maturation stages (breaker, full ripe and overripe) to evaluate the effect of fruit maturation on seed quality. Seed quality was assessed based on seed size and germination rates. During the experiments, fruit set and yield of seed were assessed. Our results show that seed production is affected by genetic components and environment, while seed quality showed higher genetic effect than environment. We found that fruit set rate was affected by maternal parent and environment (P < 0.05). Seed yield was determined by both parents and environment (P < 0.05). The paternal and maternal parents were a significant factor for determining seed size (P < 0.05) and seed 28 germination was determined by genetic components and environment factors. The effect of fruit maturation was significant on seed germination (P < 0.05). Larger seeds resulted in a higher germination rate. This study illustrates that both parents (female and male) contributed to seed quality for tomato rootstocks. In addition, our results suggest the possibility of selecting combinations of cultivated parental lines and wild species that produce high seed quality. The knowledge gained from this experiment will enable the selection and breeding of new tomato rootstock cultivars with multiple traits, which will improve the sustainability and profitability of both conventional and organic tomato production.

Introduction

Tomato (Solanum lycopersicum L.) is one of the most important vegetable crops in the world (Lucier et al., 2000). In the United States, between 2007 and 2012, tomato was planted on over 82,000 ha per year, with an annual crop value of approximately 1.8 billion dollars for the combination of processed and fresh-market crops (USDA-NASS,

2013). Based on economic value, fresh-market tomato is the most important fresh vegetable (USDA-NASS, 2013). In addition to its economic value, tomato provides vitamin C and pro-vitamin A (β-carotene) to the human diet (Rouphael et al., 2010).

Therefore, per capita consumption of tomato has increased from 14.9 lb. to 17.8 lb. in the interval between 1985 and 2000 (Jones, 2012). Given projected population growth, it is necessary to continue to improve production efficiency for all crops while maintaining nutritional and market quality standards.

Tomato producers often face production problems related to soil-borne diseases and abiotic stresses that can reduce the yield and quality of fruit (Alam et al., 1994; King et al., 2008; Hasna et al., 2009; Cramer et al., 2011 and McAvoy et al., 2012). A number of methods are available to control soil-borne diseases, including: (1) host resistance

(e.g., use of resistant varieties); (2) cultural control (e.g., crop rotation), (3) organic amendments, (4) physical control methods (e.g., solarization), (5) chemical fumigants

(e.g., methyl bromide), and (6) biological control, which is under development (Schafer,

1999; Koike et al., 2003; McSorley et al., 2009).

Grafting is now a well-established technique in annual crops that has already shown promising results in diverse production environments. Grafting onto resistant rootstocks has the potential to address concerns about chemical use while increasing production efficiency (McAvoy et al., 2012). Furthermore, this technology is considered a component of Integrated Pest Management (IPM), which includes many other methods to control soil-borne diseases such as solarization (Ioannou, 2001), bio-fumigation, and intercropping (Kaskavalci et al., 2009). Grafting onto resistant rootstocks can be used to control soil-borne diseases (Harrison and Burgess, 1962; Bradley, 1968; Grimault and

Prior, 1994; Paplomatas et al., 2000; Rivard et al., 2010; Barret et al., 2012 and McAvoy et al., 2012), increase yield (King et al., 2010), and impart tolerance to abiotic stresses

(Black et al., 2003; Rivero et al., 2003 and Venema et al., 2008). Grafting can also avoid the loss of desirable traits and the time consuming processes associated with introducing resistance into contemporary cultivars. Many tomato rootstocks have been released to the market in recent years. However, only a few are widely used in practice because of the

30 superior performance or the availability of seeds in the market (Kubota et al., 2008; King et al., 2010). There are multiple reasons to increase the diversity of rootstock used by the industry. The overuse of the same rootstocks by growers may permit pathogen populations to evolve, resulting in the breakdown of resistance, as occurred in Japan with watermelon rootstock (Acquaah, 2007 and King et al., 2008). Currently, interest in breeding for improved rootstock is increasing rapidly in the private sector due to cost effectiveness and expanded disease resistance (King et al., 2010).

Rootstock seed quality is important because it affects seedling emergence, uniformity of growth, and diameter of the stem (Pet and Garretsen, 1983; Nieuwhof et al.,

1989). Lack of uniformity creates inefficiencies and difficulties in the grafting process

(Lee et al., 2010). Seed production and quality are often affected by genetic components, the maternal environment, and harvest time of the fruit (Baskin and Baskin, 1998;

Contreras et al., 2008). Seed quality is measured by seed size, germination rate, genetic purity, freedom of contaminates, and moisture content. In tomato, the standards of high seed quality are genetic purity of 99%, germination rate above 75% (AASCO, 2013) or

85% (ISF, 2013), and moisture content less than 8%.

The use of F1 hybrids as tomato rootstocks has become common in tomato grafting. Early rootstock hybrids of tomato were produced using S. lycopersicum

(formerly Lycopersicon. Esculentum Mill) as a female parent and S. habrochaites

(formerly Lycopersicon hirsutum var. ‘glabratun’) as a male parent (Bravenboer and Pet,

1962). Hybrid seed production from wide hybridization involves the fusion of the female and male gametes, where the aim of crossing is to combine important traits from the wild

31 species and the cultivated species. However, in some wide crosses reproductive barriers between parents affect the production of hybrid seeds. Crosses between cultivated tomato and its wild relatives, lead to vines displaying increased growth which is interpreted as increased vigor, and combine multiple traits (e.g. disease or insect resistance, salt tolerance, cold tolerance) (Rick, 1982; Rick et al., 1978; Thomas and Pratt, 1982; Rick et al., 1986 and Rick and Chetelat, 1995). At the same time, there are prezygotic and postzygotic interspecific crossing barriers between cultivated tomatoes and its wild relatives which may result in cross failure, seeds with low quality, or seed sterility in hybrids (Quiros, 1991; Rick et al., 1978 and Bedinger et al., 2011). The use of seed directly from wild species or from hybrid seed derived from wild species may, therefore, have a negative effect on seed quality.

To obtain high seed quality, environmental components of production and harvesting should also be identified and maximized. Ideally, seeds should be harvested when they reach physiological maturity (PM), which is when no additional assimilates are deposited in the developing embryo. However, there are discrepancies in the literature between and within species when defining PM. Some seeds achieve PM when the seed reaches the maximum dry mass (Harrington, 1977; Deluche, 1980) and others before they attain the maximum dry mass (Tekrony and Egli, 1997). Species where the seeds develop in fleshy fruits (tomato, peppers) may reach maximum quality after maximum dry mass is achieved (Valdes and Gray, 1998; Demir and Samit, 2001).

In recent years tomato breeders have been developing plant material with high lycopene content, which have health benefits because its antioxidant functions (Croteau

32 et al., 2000; Takeoka et al., 2001). Plant material with high lycopene levels may present negatives effects on plant development including delayed germination or reduced plant growth, which was reported previously (Jarret et al., 1984; Berry and Bewly, 1992).

Furthermore, high levels of carotenoids might inhibit the production of essential germination promoters including gibberellins (Fray et al., 1995), which will reduce the germinability of the seeds or delayed germination. Lycopene production is associated with fruit maturation (Arias et al., 2000). Therefore, it is essential to identify the ideal harvest time of tomato fruit for seed production.

To address the importance of genetic and environmental effects on rootstock seed quality, we evaluated a large number of wide crosses, with parents selected based on their potential contribution to resistance against biotic and abiotic stress. The objectives of this study were: (1) to quantify the influence of genetic and environmental factors on seed production and quality in hybrid crosses, (2) to assess quality relationships in order to define early selection criteria for rootstock breeding, and (3) to identify seed harvest time that maximize seed quality of hybrid tomato rootstocks.

Materials and Methods

Plant material and growing conditions

We obtained seeds from seven accessions of tomato and wild relatives from the

Tomato Genetics Resource Center (TGRC) and another 13 accessions were used from our breeding program. Specific crosses were conducted using 10 inbred parental lines as female parents (S. lycopersicum), and 11 accessions of wild species as male parents (S.

33 arcanum, S. chilense, S. corneliomulleri, S. habrochaites, S. pennellii, S. peruvianum, and

S. pimpinellifolium). The selection of these parents was based on traits such as resistance to soil-borne diseases and tolerance to abiotic stresses. Details are presented in (Table

2.1). All seeds were sown in 288-cell (Hummert, EARTH City, MO-USA) trays with a cell volume of 13 mL, using soilless mix PRO-MIX (Premier Hotticultural, Quakertown,

PA – USA) and placed in a greenhouse. After four to five weeks, plants were transplanted into 3.78 L plastic pots (Hummert, EARTH City, MO-USA) using PRO-

MIX.

In the greenhouse at the Ohio Agricultural Research and Development Center

(OARDC) in Wooster, OH, day time temperatures were set between 27 ºC – 30 °C and nighttime temperatures between 20 ºC – 25 °C, with 12 h light and 12 h dark (long-day conditions). Plants were irrigated daily and uniformly fertilized with 250 mg l-1 dose of

20N-20P-20K (Peters Professional All-Purpose Fertilizer, Scotts-Sierra Horticultural

Products Co., Marysville, OH). Daily temperature data were obtained from the Ohio

Agricultural Research and Development Center (OARDC) weather stations system

(http://www.oardc.ohio-state.edu/newweather/dailyinfo.asp?id=1).

Pollination

All crosses were performed by hand pollination. Flowers from female plants were emasculated before flowers opened, typically a day before the anthesis. Pollen was collected from each male parent and placed on the stigma of the emasculated female flower. Each pollinated flower was labeled with a white tag containing the pollination

34 date and the parental information. The number of pollinated flowers per plant was 30 –

50 based on fruit set and flower production of the female plants. Non-pollinated flowers setting fruit were removed from female plants to avoid introducing selfed seed into evaluation of quality. Pollination took place over 1.5 months (experiment 1) and 3 months (experiment 2), generating 9 and 26 pollination dates for experiments 1 and 2, respectively.

Pollen collection was carried out for all wild male parents in bulk, and stored to maintain availability. Pollen was collected in a 0.2-mL eppendorf tubes by vibrating an open flower from the male plants. This collection was conducted during 2011 and 2012.

Tubes containing pollen were placed into glass vials 50 mL (USA Scientific, Inc.

Orlando FL) with full filled of drierite desiccant (W.A. Hammond Drierite Co., Xenia,

Ohio) and stored at -80 °C until used (Sacks and Clair, 1996). On pollination day, the vial was removed from the -80 °C freezer 20-30 min before pollinating, warmed to room temperature and wiped dry to remove moisture before opening the glass vial and removing the eppendorf tube containing pollen.

The experiment was divided in two phases. In Phase 1, ten female and four male parents were used and the entire experiment was repeated twice: Jan. – Jun. 2012 and

Jun. – Dec. 2012. Ten female plants were arranged in a randomized complete block design (RCBD). In the second phase, an experiment was conducted to compare 70 new crosses with two controls crosses used to normalize data using an augmented design. This second experiment involving these 70 new crosses was used to expand the genetic base of potential rootstock crosses using criteria identified from the first phase. The experiment

35 was conducted using an augmented randomized complete block design with four blocks consisting of 4 controls and 18 new crosses. Blocking was used to account for environmental gradients along rows and quadrants in the greenhouse.

Separate experiments were performed using a reduced number of genotypes in order to evaluate the effect of fruit harvest stage on seed production and quality. Four S. lycopersicum accessions were used as female parents (Hawaii 7998, Ohio MR13, FG02-

188 and, Fla.7775). A partially inbred accession developed from LA2204 (S. habrochaites), LA2204B, and (S. pimpinellifolium) LA1589 were used as the male parents. These eight crosses were selected based on experience suggesting a range of seed quality (from poor to high) and on positive performance data of the resulting hybrids. A randomized complete block design with harvest time as a factorial with three replications was used. The experiment was repeated twice Jun. 2012 – Dec. 2012 and Dec. 2012 –

May. 2013. Normally, the tomato seed producers harvest the fruit at breaker stage to full ripe, due to the high cost of maintaining the field for an extra time (Demir and Samit,

2001). In addition, over-ripe of fruit in mother plant increases lycope content, which may inhibit important germination promoters causing reduction in seed quality (Berry and

Bewley, 1992). In this study our objective was to identify the best harvest time for wide- hybrid crosses. Thus, our main treatment was fruit maturity at harvest time: early

(breaker stage), medium (full ripe), and late (over-ripe).

Harvesting of tomato fruit was carried out daily until the end of the season. A fruit was considered set if enlarged to > 0.5 cm diameter. Fruit set rate was determined as the total number of fruit divided by the total number of pollinated flowers on each plant.

Seed yield was determined as the total seed obtained divided by the total fruit harvested from each plant.

Seed extraction

Seed extraction was conducted by hand following the standard protocol used by the Tomato Genetics and Breeding Program, OARDC, Wooster, Ohio. All harvested fruit were cut in a half, and seed removed into a plastic beaker. 10% Hydrochloric acid (3 N

HCl) was added to the seed and left standing for 30 minutes at room temperature (~ 25

°C). The seeds were rinsed under running tap water and placed back in the beaker.

Trisodium phosphate (TSP) (10%) was added to the beaker containing seeds and soaked for 15 minutes. Then, seeds were rinsed again and dispersed on a screen. The screens with seeds were immediately placed in a dryer with at ~ 25 °C with air movement, and left overnight. The next day, the seeds were packed in coin envelopes and stored until the evaluation of seed quality.

For experiments aimed at measuring genetic effects, seeds were extracted three times a week during the peak of season and once a week toward the end of the seed production season. For the harvest timing experiment, seed extraction was done immediately or the day after harvesting the fruit.

Seed quality evaluation

We randomly pooled 200 seeds per hybrid and divided them into four subsamples

(50 seeds). Seed size was determined by weighing replicates of these 50 seeds. The

37 average weights of the subsamples were used as an estimate of seed size. Germination tests were carried out according to International Seed Testing Association (ISTA) rules

(ISTA, 2013). We evenly distributed 50 seeds in each Petri dish containing 9.0 cm × 2nm blue blotting paper (Anchor Paper, Co., St. Paul, MN) previously saturated with double distilled water (ddH2O). The Petri dishes were then placed in a controlled environment chamber (Hoffman Manufacturing Inc. Albany, OR) at 25 ± 1 °C in the dark.

For the first replication of experiments designed to explore the relationship between fruit maturity and seed quality, germination test was extended for 7 d longer than recommended (ISTA, 2013) to enable more-dormant seeds to germinate.

Germinated seeds were counted 7, 14 and 21 d after seeding. For the second experimental replication, the germination period was 14 d and germinated seeds were counted every day. For all experiments, newly emerged seedlings were removed from the Petri dishes, and seeds were regularly watered with double distilled water. Seeds were considered germinated when the radicle protruded from the seed coat. Seedlings were also evaluated as having normal or abnormal visual morphology.

Statistical analysis

Statistical analysis was performed using RStudio computer software version

0.97.320 (RStudio, 2012) in concert with the R core package version 2.15.3 (R Core team, 2011). In order to assess the relative contribution of genetic and environmental effects on fruit set, seed yield and seed quality data were treated as using a Random

Effect statistical model in order to extract variance components using the Lme4 package

(Bates et al., 2011). The statistical model contained all main effects and all possible two- way interactions between the factors.

The Agricolae package in R (De Mendiburu, 2012) was used to perform mean separation. Mean separations were based on Tukey’s test (P < 0.05). The Sciplot package was used to place error bars on the graphics (Morales et al., 2012). Seed size (weight) was related to germination with scatter plot, superimposed Locally Weighted Polynomial regression (LOESS) (Cleveland and Devlin, 1988) with second degree polynomial and the parameter alpha, controlling the size of the local neighborhood, set at 1 using the package ggplot2, with commands geom_smooth and stat_smooth, and overlapping two levels of confidential interval 95% C.I and 99.5% C.I, respectively (Wickham, 2009).

Data for the augmented design were analyzed based on heterogeneity within

Blocks (Federer, 1961). The analysis of the data was performed using PROC GLM of

SAS (SAS Institute Inc., Cary, NC, USA, 2010). This analysis takes into account variability amongst blocks, measured by the control hybrids. The values for new hybrid entries are adjusted based on the experimental model, allowing for comparisons of a larger number of hybrids in order to assess those that are worth selecting for further evaluation for potential as rootstock.

Results Since the study was conducted in two season productions, air temperature data were collected from the OARDC weather system. In the winter-spring season

(experiment 1), the daily maxima and minima air temperature ranged from -5.9 ºC – 33.7

ºC and -18 ºC – 20 ºC, respectively. For the summer-autumn season (experiment 2) the 39 daily maxima and minima air temperature ranged from -2.8 ºC – 36 ºC and -8.7 ºC – 22

ºC, respectively.

Over 5,900 flowers were pollinated using 10 female cultivated genotypes and 4 male wild species to create 40 hybrids. These hybrids were evaluated for four seed production and quality traits including fruit set, seed yield (seeds/fruit), seed weight, and germination. Data were approximately normally distributed and analysis was done on untransformed data. Total variance for each trait was partitioned into the components associated with genetics (female and male parent), crossing season environment

(experiment), the interaction between parents, and the interaction between parental genotypes by environment and block within environment (Table 2.2).

Variance in seed production

Seed production was highly influenced by both genetic components (female and male parent) and environment (seed production season).The maternal effect explained

16% of the total variation associated with fruit set. No significant effect was detected for male genotype on fruit set. For seed yield, an equal effect was identified for both parents, each contributing 14% to the total variation. The environmental effect was significant for both traits. Experiment to experiment differences explained 22% of the total phenotypic variation for fruit set and 27% for seed yield. The portion of variance explained by block within experiment was negligible for fruit set and 2% for seed yield. The variance explained by the interaction between parents was also low, 0% and 3% for fruit set and seed yield, respectively. No significant difference was detected for block within

40 experiment, or the interaction between parental genotypes, for either trait. The variance explained by the interaction between female genotype and the environment was 11% for fruit set and 8% for seed yield. The interaction between the male genotype and the environment explained 3% of the total variation for fruit set and 1% for seed yield.

Uncontrolled error accounted for 47% and 30% of the total variation for fruit set and seed yield, respectively (Table 2.2).

Crosses in which Fla.7547, Fla.7775, VFNT Cherry, NC-HS-1 and Mogeor were used as a female parent the mean percentage of fruit set was above 50%. While, for

H7998, UC-T338, FG02-188, H7997 and Ohio MR13 the rate of fruit set was below 50%

(Figure 2.1-A). The lack of paternal (main) effects on fruit set suggests that this trait was mainly determined by maternal parent in wide crosses. The percentage of fruit set was

20% higher during the experiment 1 (winter-spring) than experiment 2 (summer-autumn) across all the female and male parents (Figure 2.3-A), suggesting an important role for environmental control during hybrid production.

Significant interaction was observed between female genotype × environment and male genotype × environment for fruit set (Table 2.2). The significant female genotype by environment interaction for fruit set was due to a change in rank and change in magnitude of the differences among the genotypes. Likewise, the significant male genotype by environment interaction for fruit set was due to change in rank among the genotypes. In experiment 1, hybrids formed using LA2204A were the highest seed producers. This rank changed for experiment 2, where hybrids produced with LA1589 were the higher seed producers. This result could be influenced by a temperature effect

41 on the pollen viability of the S. habrochaites parent. In previous studies, heat tolerance was reported for S. pimpinellifolium pollen (Villareal et al., 1978; Abdul-Baki, 1991;

Abdul-Baki and Stommel, 1995).

Seed yield ranged from 11 seeds/fruit (Ohio MR13) to 54 seeds/fruit (NC-SH-1) for female parents across male parents and environment (Figure 2.1-B). In contrast to fruit set, there was significant variation among paternal genotypes for seed yield

(seeds/fruit). LA1589 (S. pimpinellifolium) resulted 31%, 43%, and 46% higher seed yield compared to LA2204B, LA2204A (S. habrochaites) and LA0716 (S. pennellii), respectively (Figure 2.2-A). The average yield of seed (seeds/fruit) was 43% higher during experiment 1 (winter-spring) compared to experiment 2 (summer-autumn) (Figure

2.3-B).This may be atribuited to the extreme high temperatures occurred during the experiment 2. The female genotype by environment interaction was significant for seed yield. This significance was due to a change in the magnitude of the differences among genotypes between two experiments with minor changes in rank.

Variance in seed quality

The variances associated with genetic components were larger than the variance associated with environmental effects for seed size (seed weight) and seed germination

(Table 2.2). The role of the female parent was highly significant for seed weight and germination, explaining 27% and 22% of the variance, respectively. The male parent was also highly significant for seed size and seed germination, contributing 57% and 17% to the total variance, respectively. The crossing season accounted for 13% of the total

42 variation for seed germination and none for seed weight. The female by male parent interaction was significant only for seed germination, contributing 26% to the total variation. Paternal genotype by environment interaction was significant, but explained only 2% and 3% of the variation for seed size and seed germination, respectively. We did not detect significant effects for female genotype by crossing season interaction and block within experiment interactions for either trait. Error accounted for 12% and 19% of the total variation for seed weight and germination, respectively (Table 2.2).

The seed weight for female parents ranged from 1.034 mg (VFNT cherry) to 3.14 mg (Fla.7775). This range indicates differences in the amount of resources allocated to the seeds among maternal parents. Such maternal effects on seed size are normal in plants

(Roach and Wulff, 1987; Schwaegerle and Levin, 1990). Differences among paternal parents also contributed greatly to variation for seed size. For seeds resulting from pollinations with LA1589, the seed weight across female plants was 3.68 mg.

Pollinations by LA2204B resulted in an average seed size of 1.97 mg, followed by

LA2204A with 1.94 mg, and LA0716 with the smallest seeds at 1.42 mg (Figure 2.2-B).

The ranks of seed weight by paternal parents were consistent across all the maternal plants. These ranks correspond to the genetic distance between the wild species, with S. pimpinellifolium most closely related to the cultivated types.

There was a significant male genotype by environment interaction effect for seed weight, although the magnitude of the interaction was relatively small in comparison to the main effect due to male genotype. This observation suggests that seed size is influenced more by male genotype than environment.

The germination for female parents across male parents and environment varied significantly, ranging from 34% (VFNT Cherry) to 83% (H7997) averaged across all crosses (Figure 2.1-D). Across female parents and environment for male parents the average germination varied from 49% (LA2204A) to 85% (LA1589) (Figure 2.2-C). In experiment 1, conducted in the (winter-spring) season the percentage of seed germination was higher 74%, compared to experiment 2, conducted in the (summer-autumn) season, which averaged 59% of final germination across parental genotypes used in the study

(Figure 2.3-C).

The seed germination was also significantly affected by female and by male parent interactions, revealing genetic variation among hybrids for germination response

(data not shown). Of ten female parents used in this study, eight had high germination when hybrids were formed with LA1589. Seven had high germination when crossed to

LA0716. Four females had high germination rates when crossed to LA2204B, and only one had high germination when crossed to LA2204A. Variation was also detected for germination based on female parents. The multi-resistant female parent Mogeor had relatively high seed germination across all male parents used in pollination. The germination percentage was low for Ohio MR13 female parent independent of male parent used as a pollen donor.

The effect of paternal parent by crossing season interaction was significant, which was due to change in the magnitude of differences among paternal genotypes between experiments. Seeds from experiment 2 pollinated with LA2204A had lower germination rate compared to experiment 1 and other male parents within experiment 2.

Seed weight and germination

We wanted to know if selection for rootstock seed could be performed before germination tests, without negatively influencing which hybrids might be carried forward in a breeding program. The mean of the final germination percentage was plotted over the mean of seed weight, a curve fit to the data, and confidence intervals of 95% and 99.5% were superimposed (Figure 2.4). Seed weight ranged from 0.04 mg to 4.56 mg with the large differences in seed weight among the hybrids also reflected in wide differences in germination which ranged from 15% to 97%. Germination from large seeds was better than from intermediate and small seeds (Figure 2.4). These results indicate that selection is possible based on seed weight, thus avoiding to the need to perform germination tests for clearly inferior material.

In the second phase, an experiment was conducted to compare the genetic potential of 70 new crosses. Control crosses were replicated in each block, in order to correct for spatial variation. The controls were chosen based on evaluation during phase

1. Using this approach, we were able quickly to evaluate a larger number of crosses, using available space, time, and human resources. The results for seed production and quality of the 70 new crosses and the 2 controls are shown in Table 2-3. Least squares means for fruit set across hybrid ranged from adjusted values of -11% to 98%. Seed yield ranged from adjusted values of 1 seed/fruit to 117 seeds/fruit. Twenty-four hybrids were dropped from seed weight and seed germination tests due to low seed number.

Based on seed weight and germination, of the 70 new hybrids, only 11 would be worth pursuing as potential rootstocks. FG12-501 to FG12-510 were crossed with

LA2533 (S. pimpinellifolium), FG12-511 to FG12-520 with LA2157 (S. arcanum),

FG12-521 to FG12-530 with LA 1929 (S. peruvianum), FG12-531 to FG12-540 with

LA1937 (S. corneliomulleri), FG12-541 to FG12-550 with LA1959 (S. chilense), FG12-

551 to FG12-560 with LA2931 (S. chilense), and FG12-561 to FG12-570 were crossed with LA0407 (S. habrochaites). The seed weight among the hybrids ranged from 0.001 mg (FG12-520) to 4.56 mg (FG12-507), and seed germination varied from -7% (FG12-

516) to 93% (FG12-504) (Table 2-3). Based on these results, only LA2533 (S. pimpinellifolium) and LA0407 (S. habrochaites) produced any hybrids with seed quality that would warrant further evaluation and optimization (Table 2.3).

In order to investigate the potential for improved seed quality, we investigated the effect of fruit maturation on seed germination. Selected parents were used for this study H7998, OhioMR13, FG02-188, and Fla.7547 as female parents and LA2204B and

LA1589 as male parents. Harvest of fruit was timed to correspond to breaker stage, fully ripe and over ripe. Time of harvest was significantly associated with germination rates (P

= 0.003). Seed germination for seeds harvested at full ripe stage was higher compared to breaker stage for most of the hybrids. However, we detected differences among hybrids for germination and germination speed. Seed of the hybrids FG02-188 × LA1589 and

FG02-188 × LA2204B were less affected by fruit harvested at the breaker stage than the other six hybrids (Figure 2.5-A). From the results of our study, FG02-188 × LA1589and

FG02-188 × LA2204B hybrids can be harvested early and high seed quality can be obtained. For the crosses, OhioMR13 × LA2204B, Fla. 7547 × LA2204B, OhioMR13 ×

LA1589 and Fla.7547 × LA1589 final germination was significantly increased when

46 fruits were harvested at the overripe stage. The wide hybrid FG12-604 was limited by genetic compatibility between the female (H7998) and male (LA2204B) parents (Figure

2.5-C).

Discussion

High seed production and quality are important in hybrid seed production to justify the extra cost added by emasculation and pollination, which is labor intensive, and therefore, expensive (Georgiew, 1991; Opeña et al., 2001).

In the present study, seed quality was the product of interactions between environmental and genetic components (female and male parents). While seed size was maintained between environments (experiment), seed germination was higher during winter-spring season (experiment 1) compared to summer – autumn season (experiment

2). This may due to the extreme high temperatures occurred during experiment 2, which may cause dormancy on some hybrids or abnormal germination. In addition, we identified that seed size is an important predictor for seed germination at early selection point.

There was a wide variation in fruit set and seed yield, which can be attributed to a maternal parent by environment interaction. These results are not surprising given that maternal parent determines the number of ovules, provides resources to the new embryo and produces the seed coat (Roach and Wulff, 1987).

Tomato fruit set is optimal between 17 - 18 °C night and 20 – 25.6 °C d (Went,

1944; Leopold and Scott, 1952; Charles and Harris, 1972). According to the NOAA’s

47 report, the summer of 2012 was the third hottest summer on record for the United States.

Extreme climate conditions could lead to overheat the greenhouse during summer time and to expose the plants to heat stress. Cooling a greenhouse in a hot summer is difficult and complicated task compared to heating a greenhouse during wintertime (Nilsen et al.,

1984; Baille et al., 1994; Baille and Leonardi, 2000; Kittas et al., 2005).

The low fruit set and seed yield obtained in experiment 2 (summer-autumn season) can be attributed to high temperature in the greenhouse, which is well documented that extreme temperatures has negative effects on yield (Alam et al., 2010), and reproductive organs. High temperature can affect fruit set and seed yield by affecting meiosis in the pollen and ovule mother cells, development of the endothecium in the anther, then reducing the dehiscence and pollen shed, inhibiting pollen germination, slowing pollen tube growth, drying the stigma surface and causing low pollen retention, reduce growth of the endosperm pre-embryo and early stage zygote. Reducing the growth on fertilized embryo could result in small undeveloped seeds, which could be lost during seed extraction process (reduce seed yield) (Iwahori, 1965, 1966; Picken, 1984; Dinar and Rudich, 1985; Dane et al., 1991; Pet et al., 1997; Hanson et al., 2002; Sato et al.,

2000; Max et al., 2009). Thus, it is important to maintain the ideal temperature in order to obtain high seed yield when producing hybrid seed tomatoes.

These results suggest that the ideal season for seed production under controlled environment is winter-spring season. In addition, selecting both female and male parent based on their response to environment seed production can be improved under harsh environment conditions.

Based in the results we conclude that seed size is determined mainly by paternal genotype, contradicting previous studies (Pet and Garretsen, 1983), where they found that seed size is mainly determined by maternal genotype. These differences might be due that they used all inbred lines (female and male parents), while in this study we produced wide hybrid seeds using wild relatives, where paternal parent had stronger influence.

Therefore, when using wild relatives it must be keep in mind that seed size can either increase or decrease.

Marked differences due to genetic components were obtained for seed germination. In general, the genetic distance between female and male parent can predict seed germination, but some wide crosses that can perform the same or better than crosses with short genetic distance parents. In this study we identified seven wide hybrids crossed with LA0716 (S. pennellii), three with LA2204B (S. habrochaites) and only one with

LA2204A (S. habrochaites) that reached the minimum national seed germination standard 75% (AASCO, 2012).

Data presented here (Figure 2. 4), out of 40 plotted germination points, the six smallest hybrids showed the lowest germination rate. Thus, selection could be performed before the germination test by cutting off at 1.4 mg. of seed weight. At this selection point, we could be omitting 33% seeds with low quality and retaining a 100% of high quality seeds for germination test. This selection point can be increased to 1.8 mg. where

43% of seeds with low quality can be omitted and retained 74% of high seed quality for seed germination test. In the extreme scenarios, we can draw a line at 2.2 mg. for seed

49 germination where 62% of low seed quality can be omitted and retained only 53% of high seed quality.

The effect of seed weight on final germination may be due to variation in food reserves in the seeds and influenced by the interaction of the parents. H7998 and VFNT

Cherry resulted in the lowest seed weight and final germination when they were crossed with LA2204A, LA2204B and LA0716. In contrast when crossed with LA1589, the seed sized increased up to 90% for H7998 and 88% for VFNT Cherry. The final germination also increased up to 80% and 83% for H7998 and VFNT Cherry, respectively.

Part of the seed weight effect on seed germination seemed to be parentally determined, since the hybrids crossed with S. pimpinellifolium showed higher seed weight and high seed germination. Hybrids crossed with S. pennellii, even the seed weight was lowest the germination was the second highest compared to other male parents. The hybrids crossed with S. habrochaites had the lowest seed germination but seed weight was the second largest compared to other male parents.

Hybrids crossed with the two partially inbred selection S. habrochaites showed a correlation in seed size where LA2204B had 22% higher seed germination and 2% higher seed weight compared to LA2204A. This result indicates that these two male parents are different between each other, due to selection over the time in the greenhouse.

Previous studies looked at the correlation between seed size and time to germinate

(Whittington et al., 1965). In this study, we evaluated the association of seed weight and final germination percentage. The effect of seed weight on germination could be due to

50 food reserve, which can be influenced by the maternal environment during seed production.

The maximum dry weight of tomato seeds is attained before physiological maturity, which is at breaker stage (Valdes and Gray, 1998; Demir and Samit, 2001). In this study, over the range of fruit ripeness from breaker stage to overripe no changes were detected in seed weight. This result indicates that wide hybrid crosses in tomato reaches the maximum mass at the same time as self-seeds breaker stage (Valdes and Gray, 1998;

Demir and Samit, 2001). However, maximum seed quality in terms of seed germination is not obtained at breaker stage for the majority of hybrids.

Fruit maturation can be one of the factors to manipulate in order to increase seed quality, but not all hybrids can be improved. Seeds from fruits harvested at breaker stage had lower final germination; it could be due to high percentage of immature seeds. The majority of the hybrids produced in this study showed higher seed germination when the fruits were left on the plant mother until overripe stage. While, only two out eight hybrids resulted to have high germination when harvested at breaker stage, this seems to be maternal determined (FG02-188).

The female parent FG02-188 crossed with both male parents showed a higher final germination regardless of the harvesting time. However, there was higher speed germination when fruits were harvested at overripe stage. In general, the germination speed was higher in seeds from fruits harvested at overripe stage (Figure 2.5-A-B-C) and has a genetic component, which was observed before (Demir and Ellis, 1992; Berry and

Bewley, 1992).

Fruit maturation can affect the final seed germination and speed germination in wide hybrid seeds, both traits are important for seed industry. These findings indicated that wide crosses could produce rootstocks with high seed quality. In addition, fruit maturation a key point to manipulate in order to increase seed quality in wide hybrids. At the same time, control of the seed production environment is essential for better quality seed production in wide-cross hybrid tomato seeds. Finally, we identified key selection points in the breeding of new tomato rootstock cultivars.

Tables Accession Variety Name Sourcez Parent Species Resistance/tolerance

LA3471 Mogeor TGRC Female S. lycopersicum Frl, I, I^2, Mi, pyI, Tm2^2, Ve LA2939 UC-T338 OHIO Female S. lycopersicum I, Mi, Tm2^2, Ve LA1221 VFNTCherry OHIO Female S. lycopersicum I, Mi, Tm2^2, Ve LA4025 Fla.7547 OHIO Female S. lycopersicum I, I^2, I^3 LA3847 NC HS-1 OHIO Female S. lycopersicum I, I^2, Ve, heat OhioMR13 Ohio MR13 OHIO Female S. lycopersicum Frl, Tm2^2 Fla.7775 Fla.7775 OHIO Female S. lycopersicum Frl,, Tm2^2, I, I^2, FG02-188 FG02-188 OHIO Female S. lycopersicum Clavibacter michiganensis

LA3856 Hawaii7998 OHIO Female S. lycopersicum Xanthomonas campestris pv. vesicatoria; Ralstonia solanacearum

LA3202 Hawaii7997 OHIO Female S. lycopersicum Ralstonia solanacearum LA1589 LA1589 OHIO Male S. pimpinellifolium 53 LA2533 LA2533 OHIO Male S. pimpinellifolium Phytophthora infestans Xanthomonas gardneri

LA2157 LA2157 TGRC Male S. arcanum Mi^9, Clavibacter michiganensis, alternaría solani, heat LA1929 LA1929 TGRC Male S. peruvianum Drought LA1937 LA1937 TGRC Male S. corneliomulleri Drought LA1959 LA1959 TGRC Male S. chilense Peanut Bud Necrosis Virus, drought LA2931 LA2931 TGRC Male S. chilense Stress, salt

LA2204A LA2204 OHIO Male S. habrochaites Clavibacter michiganensis

LA2204B LA2204 OHIO Male S. habrochaites Clavibacter michiganensis LA0407 LA0407 TGRC Male S. habrochaites Clavibacter michiganensis LA0716 LA0716 OHIO Male S. penellii I^3, Xanthomonas campestris, Alternaria alternata, salt, drought Verticillium dahliae (Ve), Fusarium oxysporum f.sp. Lycopersici race 1(I), Fusarium oxysporum f.sp. Lycopersici race 2 (I^2), Fusarium oxysporum f.sp. Lycopersici race 3 (I^3), Fusarium oxysporum f. sp. radicis- lycopersici (Frl), Tomato mosaic virus (TM2^2), Meloidogyne incognita (Mi) and Pyrenochaeta lycopersici (pyl). zTGRC: Tomato Genetics Resource Center, Department of Vegetable Crops, University of California, Davis, CA 95616, USA; OHIO: Tomato Genetics and Breeding Program, Department of Horticulture and Crop Science, Ohio Agricultural Research and Development Center - The Ohio State University, Wooster, Ohio 44691, USA.

Table 2.1: Female and male parents used in hybrid seed production.

Fruit set Seed yield Seed weight Germination Source of (%) (Seeds/fruit) (mg) (%) variance Estimate % of Estimate % of Estimate % of Estimate % of variance variance variance variance Female parentz 0.004 16*** 96.21 14*** 0.45 27*** 0.020 22*** Male parenty 0.000 0NS 95.60 14*** 0.94 57*** 0.016 17*** Female:Male 0.000 0NS 21.97 3NS 0.02 1NS 0.024 26*** Environemntx 0.006 22*** 177.87 27*** 0.00 0NS 0.012 13*** Block(Env) 0.000 0NS 10.75 2NS 0.00 0NS 0.000 0NS ** * NS NS 54 Female:Env 0.003 11 55.57 8 0.02 1 0.000 0

Male:Env 0.001 3* 5.17 1NS 0.03 2* 0.002 3* Error 0.013 47 202.43 30 0.20 12 0.018 19 NS, *, **, *** Nonsignificant or significant at P ≤ 0.05, ≤ 0.01 and ≤ 0.001, respectively.

z Ten inbred parental lines (S. lycopersicum) were used as female parents to evaluate fruit set (%), seed yield (number of seeds per fruit), seed weight (mg) and seed germination (%). Females were Mogeor, UC-T338, VFNT Cherry, Fla. 7547, NC HS-1, Ohio MR13, Fla. 7775, FG02-188, Hawaii 7998, and Hawaii 7997. yFour wild accessions were used as male parents LA1589 (S. pimpinellifolium), LA0716 (S. pennellii) and two partially inbred accessions developed from LA2204 (S. habrochaites), LA2204A and LA2204B. xExperiments were repeated two times, January – June 2012, and June – December 2012 to evaluate the effects of environment.

Table 2.2: Estimated variance components for female parents, male parents, experiment, and interactions on fruit set, seed yield, seed weight and seed germination.

Hybrid FS SY SW SG Hybrid FS SY SW SG (%) (seeds/fruit) (mg) (%) (%) (seeds/fruit) (mg) (%) FG12-504 64 69* 3.77 93 FG12-567 36 8 0.59 1 FG12-501 15 33 4.26 93 FG12-535 33 32 0.14 1 FG12-509 9 26 2.64 92 FG12-536 9 33 0.10 1 FG12-507 94 22 4.54 90 FG12-520 1 40 0.00 1 FG12-502 28 5 2.77 90 FG12-531 -1 19 0.08 1 FG12-503 19 26 2.16 90 FG12-533 -5 12 0.02 1 C-2 67 16 4.83 87 FG12-517 66 12 0.29 -7 FG12-506 31 22 3.78 86 FG12-514 55 24 0.27 -7 FG12-505 66 33 3.49 84 FG12-515 32 35 0.94 -7 C-1 9 9 2.59 80 FG12-511 22 9 0.44 -7 FG12-510 4 28 2.76 65 FG12-512 19 1 0.18 -7 FG12-565 -5 117 0.54 20 FG12-516 10 4 0.37 -7 FG12-561 -4 16 0.67 15 FG12-539 44 4 x x FG12-562 31 9 0.45 9 FG12-545 34 18 x x FG12-544 98 15 0.13 3 FG12-522 31 5 x x FG12-547 75 14 0.39 3 FG12-542 31 8 x x FG12-554 54 18 0.40 3 FG12-521 27 7 x x FG12-557 46 25 0.35 3 FG12-549 26 13 x x FG12-555 19 15 0.48 3 FG12-529 25 6 x x FG12-546 18 24 0.45 3 FG12-528 24 18 x x FG12-541 15 53 0.30 3 FG12-552 19 6 x x FG12-550 14 27 0.06 3 FG12-553 15 12 x x FG12-543 7 18 0.08 3 FG12-560 10 9 x x FG12-527 67 17 1.33 3 FG12-558 9 16 x x FG12-537 62 56 0.18 3 FG12-556 8 6 x x FG12-524 55 27 0.20 3 FG12-508 7 3 x x FG12-525 37 25 0.20 3 FG12-548 6 19 x x FG12-538 35 20 0.02 3 FG12-559 3 5 x x FG12-526 28 26 1.22 3 FG12-519 2 16 x x FG12-540 26 19 0.06 3 FG12-513 2 1 x x FG12-530 23 19 0.07 3 FG12-518 1 29 x x FG12-523 19 18 0.06 3 FG12-570 -3 13 x x FG12-551 16 26 0.35 3 FG12-566 -4 2 x x FG12-534 92 18 0.16 1 FG12-569 -8 12 x x FG12-532 48 11 0.09 1 FG12-563 -11 2 x x FG12-564 48 10 0.58 1 FG12-568 -11 2 x x C-1: control-1; C-2: Control- 2; X: Absence of test due to low seed or zero; FS: Fruit set; SY; Seed yield; SW: Seed weight; SG: Final germination.

Table 2.3: Mean comparisons using Least Squares Means with adjustment for multiple comparisons using Dunnett’s test and control C-2 for evaluation of hybrid seed quality.

A B

C D

Figure 2.1: Effect of female parents on (A) Fruit set (%), (B) Number of seeds per fruit, (C) Seed weight (mg), and (D) Seed Germination (%). Bars represent SE (n = 16). Letters above the SE bar are based on Tukey’s mean separation, when the same letter is present there is no significant difference at P ≤ 0.05.

A B C

Figure 2.2: Effect of male parents on (A) Number of seeds per fruit, (B) Seed weight (mg), and (C) Seed Germination (%). Bars represent SE (n = 40). Letters above the SE bar are based on Tukey’s mean separation, when the same letter is present there is no significant difference at P ≤ 0.05.

A B C

Figure 2.3: Effect of Environment (Experiments) on (A) fruit set (%), (B) Number of seeds per fruit and (C) Seed germination (%). Experiment 1 was conducted January-June 2012 and Experiment 2 was conducted June – December 2012. Data represent the means and error bars indicate ± SE of two independent experiment, ten female and four male parents (n = 80). Letters above the SE bar are based on Tuke’s mean separation, when the same letter is present there is no significant difference at P ≤ 0.05.

Figure 2.4: Relationship between seed size (weight) and seed germination. Data for 40 hybrids were used to fit a curve line and superimpose confidence interval (shaded). Germination of 75% (horizontal line ------) and 85% (horizontal line ) are considered the legal minimum standard (AASCO, 2012; ISF, 2013) respectively.

Figure 2.5: Germination curves for seeds from eight tomato hybrids harvested at different stages of fruit maturation. The small vertical bars represent SE for each point. In A: Fruits were harvested at the breaker stage. In B: seeds were extracted from fruits at the full ripe. In C: seeds were extracted from overripe stage fruits. Hybrids FG12-601 to FG12- 604 the pollen donor is LA2204B and FG12-605 to FG12-608 the pollen donor is LA1589. Germination of 75% (horizontal line ) and 85% (horizontal line ) are considered the legal minimum standard (AASCO, 2012; ISF, 2013) respectively.

Chapter 3 : Cyanoacrylate adhesive improve grafting success

between tomato rootstock and scion

Abstract

Grafting annual fruits and vegetables can improve crop yield and increase disease resistance; however, the process is labor intensive and economically unviable for most crops. We studied an alternative method to increase grafting success of annual vegetables using tomato as a model. Nine rootstocks were used, including accessions of the wild species Solanum pimpinellifolium (LA1589) and S. habrochaites (LA2204). We also included a processing tomato S. lycopersicum (FG02-188), an unimproved variety Hawaii

7998 (H7998), hybrids derived from crossing these parents (SGH07-315, SGH07-316,

SGH07-320, SGH07-326), and a commercial rootstock ‘Maxifort’ The experiment was repeated five times. Tube grafting, a widely used method, was compared to grafting using ethyl-2-cyanoacrylate adhesive. The scion used was Cherokee purple, an heirloom variety. We also evaluated survival rate in open field studies and yield data was collected from two rootstocks. Grafting success was significantly affected by the methods and rootstock genotypes (P < 0.05) in the greenhouse study. Using the adhesives method resulted in a better survival rate. Rootstocks Maxifort and SGH07-315 showed higher grafting success over all the experiments and grafting methods. Success for H7998 was low, suggesting that this parent contributed to poor scion-rootstock compatibility.

In the open field study, grafting success was affected by rootstocks (P < 0.05) and no differences were observed between grafting methods. Yield data revealed that using adhesives in the grafting process did not have negative effects. Grafting success was dependent on the choice of rootstock and could be improved by using adhesives.

Introduction

Grafting of annual crops has been used in Japan and Korea since the late 1920s.

Grafting of annual fruits and vegetables was first documented for watermelon (Citrullus lanatus Matsum. et Nakai) using pumpkin (Cucurbita spp.) rootstock to provide resistance to Fusarium wilt (Fusarium oxysporum) (Tateishi, 1927). In the United States, small farmers grafted vegetables onto weeds in small scale in the middle of 1940’s

(Kubota et al., 2008). Later the technique was applied to the production of cucumber and tomato (Lee, 1994; Oda, 1995, 1999 and 2002). In the early 1960s, grafting was introduced into commercial production of tomato in many parts of the world (Harrison and Burgess, 1962; Bravenboer and Pet, 1962; Upstone and Finney, 1966 and Lee and

Oda, 2003). This technology was primarily introduced into annual crops to control soil- borne diseases caused by bacteria, fungi, and nematodes. Production of annual horticultural crops using grafted seedlings has become a common practice in many parts of the world, especially in Asia and parts of Europe where land use is intensive. Grafting also dominates hydroponic production environments in North America (Lee, 1994; Oda,

2002; Lee and Oda, 2003; Kubota et al., 2008). In the United States and other western countries where land use is not intensive, grafting is less common for soil-based

production systems. Nevertheless, grafting is becoming more widely used because of the benefits in resistance, vigor, and increase production (Lee, 1994; Fernandez-Garcia et al.,

2003; King et al., 2010; Rivard and Louws, 2008; Savvas et al., 2009).

Interest in grafting Solanaceous plants has increased in recent years due to promising results such as resistance to a wide range of pathogens in the soil, tolerance to abiotic stress, and increased yield and fruit quality (Oda, 2002; Lee, 2003; Kubota et al.,

2008). Grafting may provide a chemical free solution to tomato production (Miguel et al.,

2004; Pogonyi et al., 2005; Rivard and Louws, 2008; Kubota et al., 2008; King et al.,

2010). The harvesting season could also be extended using grafted plants in vegetables

(Lee, 2003). Adoption of rootstocks accelerated when interspecific crosses exhibited a high level of resistance to Pyrenochaeta lycopersici (corky root rot) were introduced

(Lee, 1994; Oda, 1995). It was later observed that these hybrids also had increased cold tolerance and high vigor (Lee, 1994; Oda, 1995). In 1996, resistance to Fusarium oxysporum f. sp radicis-lycopersici (FORL) (crown root rot) was added to the rootstock resistance profile. Later, seed germination was improved by commercial seed treatments, and the rootstock market grew to its current size (King et al., 2010). Introduction of the

‘Japanese” (tube) grafting method to Europe led to significantly lower the cost of grafting in large scale (Kurata, 1994; Lee, 1994 and Oda et al., 1994)

For tomato, there are two common grafting methods; cleft grafting and tube grafting (Lee, 1994; Oda, et al., 1994; Oda, 1995 and 1999). For the cleft grafting method, the stem of the scion is cut in a wedge, and this tapered scion is fit into a cleft cut in the end of the rootstock. The graft is then held firm with a plastic tube or clip (Oda,

1999). Using cleft grafting method the success of grafting is high and plants with different stem diameters can be matched excellently. The disadvantage of the cleft graft method is that the grafting process requires multiple cuts and is therefore labor intensive.

Another method widely used in Korea and Japan is tube grafting. Tube grafting makes it possible to use younger plants; therefore, more plants can be grown per plug tray, and the grafting process is two to three times faster than with cleft grafting (Oda, 1999). Smaller plants also require less space in healing chambers or acclimation rooms. For this method, the rootstock and scion are cut once at the same angle. The cut ends are then joined by a tube or silicon clip, splicing the cut surface together (Oda, 1999). This method is potentially faster, but it requires the diameter of the scion and rootstock to be well matched. In our experience, the failure rate is higher with tube-grafted plants.

Problems associated with grafting remain to be addressed before the technique will be widespread for soil-based production systems. Grafting is labor intensive, which adds additional cost to the production system (Lee, 1994; Leonardi and Romano, 2004).

This cost can be magnified when incompatibility between scion and rootstock limits the rate of survival (Davis et al., 2008). Despite of favorable reports of grafting, reductions in yield and fruit quality have also been documented due to incompatibility between the scion and rootstock (Edelstein, 2004; Kacjan-Marsic and Osvald, 2004). An increased survival rate is important to compensate for the additional costs of grafting. Therefore, new methods should be explored in order to take full advantage of this technique.

The use of cyanoacrylates as a surgical closure for wounds (Andrade et al., 2001;

Albuquerque et al., 2006 and Endo et al., 2007), and the related adhesive, histocryl, for

aesthetic facial plastic surgery (Kamer and Joseph, 1989) suggest their potential use in other biological systems to join living tissue. The wide use of cyanoacrylate adhesives is facilitated by the low price and ease of use. Grafting eggplant, cucumber, Chinese cabbage and grapevine is facilitated using adhesives (Oda, 1995; Lee and Oda, 2003).

The objective of this study was to increase the success rate of grafting using ethyl-2- cyanoacrylate adhesives (Krazy Glue; Super glue) for tomato grafting. The aim of using adhesives such as ethyl-2-cyanoacrylate is to simplify the procedure and improve the efficiency in grafting tomato.

Materials and methods

Plant material and growing conditions

We used nine tomato genotype rootstocks including two wild species LA1589 (S. pimpinellifolium), an inbred selection from LA2204 (S. habrochaites), one processing material FG02-188 (S. lycopersicum), H7998 (Hawaii 7998 an unimproved); the hybrids

SGH07-316, SGH07-315, SGH07-320, SGH07-326; and ‘Maxifort’ as a control for all experiments. The scion was ‘Cherokee Purple’, an heirloom variety.

Experiments were conducted in the greenhouse facilities at The Ohio State

University, Ohio Agricultural Research and Development Center (OARDC), Wooster

OH. Sowing times of the scion were staggered over three weeks to provide a variety of stem diameters to match with the rootstock at the time of grafting. One week before the rootstocks, at the same time as the rootstocks, and one week after the rootstocks were seeded. All seeds were planted in plastic plug 288-cell (Hummert, EARTH City, MO)

trays with a cell volume of 13 mL (Hummert, EARTH City, MO) using PRO-MIX

(Premier horticulture, Quakertown, PA - USA). Twenty-five days after seeding, the rootstocks were transplanted into 40-cell trays with cell volume of 68 mL (Hummert,

EARTH City, MO) using PRO-MIX. Temperatures were set at 24/18 °C day/night and a

12-h photoperiod. Fertilization was applied using 20N-20P-20K (Peters Professional All-

Purpose Fertilizer, Scotts-Sierra Horticultural Products Co., Marysville, OH) with every irrigation time; plants were irrigated one or twice a day.

Grafting process

Seven days following transplanting, rootstocks were grafted using tube grafting

(Japanese top-grafting; Oda, 1999, 2002) or a modification using ethyl-2- cyanoacrylate adhesive (Krazy Glue, Columbus, OH). Both rootstock and scion were cut at 35° angles.

The scion was aligned with the rootstock to ensure contact vascular tissue between the two plants. Tube grafting followed procedures described previously (Oda, 1995; Oda,

2002 and Rivard and Louws, 2008), with clips replaced by Tygon tubing (1.5 mm and 2 mm) cut to 1 cm length and slit down one side. The ethyl-2-cyanoacrylate grafting followed the same steps, but replaced the Tygon-tubing clip with a brushing adhesive to the outside of the graft. Care was taken to avoid introduction of adhesive between the cut surfaces of both plants.

Healing conditions

The new grafted plants were placed in a 5.5 m × 1.2 m × 0.7 m dimension healing chamber constructed from 0.75-inch PVC pipe as a frame and 0.08-mm plastic covering to maintain high humidity (95% - 100%). The greenhouse bench was covered with a capillary mat to keep constant humidity (Agriculture solutions, Columbus, OH), which was continually saturated with water. Shade cloth (60% shade) was used to cover on the plastic for the first week to minimize water loss of the new seedlings through transpiration. Temperatures in the greenhouse were set at 30 °C day/25 °C night and no artificial light was provided during the healing period. Seven days after grafting, the shade cloth was gradually removed by rolling up the sides, and the healing chamber was opened to increase light and lower humidity. Fourteen days post-grafting, the shade cloth, and plastic were completely removed.

Experimental Design

The experiment was conducted in a randomized complete block design with block replication occurring over time. These experiments were conducted at different seasons of the year including late autumn (Experiments 1 and 2), spring (Experiment 3) and late spring (Experiments 4 and 5). Rootstock genotypes were randomized in each of five experiments (blocks). Genotypes H7998, LA2204 and SGH07-316 were omitted from the second experiment because poor germination of seed prevented sufficient replication.

The treatments included tube grafting and adhesive grafting. Twenty-five plants per rootstock were used for each treatment in each of the five experiments. Data collection

began one week post-grafting and continued for three weeks. Grafting success was evaluated by determining the number of grafted plants that survived (Alive) and showed a growth of the leaves at week one, two and three.

Field experiment

An open field experiment was carried out in Wayne County, OH (long.

40°46′46″N, lat. 81°55′30″W) to determine survival rate (Alive plant at harvesting time) and productivity of grafted plants using adhesives and tube grafting. From the last greenhouse experiment, late spring (Experiment 5), six rootstocks were selected for the field study based on the plant availability and the results of greenhouse experiment. The rootstocks chosen were ‘Maxifort’ and SGH07-315 (with high grafting success), LA1589 and FG02-188 (with medium grafting success), and LA2204 and H7998 (with low grafting success) at week 3 in greenhouse conditions. In the field trials, the two grafting methods for each rootstock genotype were compared. Each rootstock genotype-grafting method consisted of four plants per plot in a randomized complete block design with three replications. Transplants were set into a raised-bed plasticulture system. Plants were irrigated as necessary. Plant spacing was set at 61 cm within the row and 3.4 m between rows. Harvest was carried out on 21, 28 Sept. and 11 Oct. 2012 for ‘Maxifort’ and

SGH07-315 rootstocks to determine the effect of the grafting method on total productivity. Total fruit per plant was weighted for each combination (‘Maxifort’ rootstock grafted using adhesives and tube grafting method; SGH07-315 rootstock grafted using adhesives and tube grafting methods).

Statistical analysis

Statistical analysis was performed using the statistical package in “Rstudio” computer software version 0.97.320 (RStudio, 2012) in concert with the R core package version 2.15.3 (R Core team, 2012). The Agricolae package (De Mendiburu, 2012) was used to perform mean separation. The Sciplot package (Morales et al., 2012) was used to place error bars on the graphics. Data were analyzed using analysis of variance

(ANOVA) with main effects and all possible interactions between the factors.

Differences between grafting methods and genotype were assessed by least significant differences (LSD) test at P = 0.05 after a significant F-test in the ANOVA. A fixed effects model was used to estimate LSMEANS based on a linear model. Mean values were separated by paired t-test for survival rate of grafting method for each rootstock genotype at P < 0.05. The following model was used to test the effect of grafting using adhesives on grafting success at week 1, week 2 and week 3 post-grafting:

Where Yijk was the grafting success data, µ was the overall mean, Ti was the effect of grafting methods, Gj was the effect due to rootstock genotype, Ek was the experiment effect. The interactions terms were TGij was the effect due to grafting method by rootstock genotype interaction, GEik was the effect of interaction between rootstock genotype x experiment, TEik was the effect due to grafting method x experiment interaction, and ɛijk was the experimental error.

Normalization of the data was accomplished within our fixed effects model by estimating LSMEANS. In replicate 2, data were missing for genotype H7998, LA2204

and SGH07-316. Due to missing data, LSMEANS for each genotype, experiment, and treatment were extracted using the fixed effects model:

Where Yijk was the grafting success value (%), was the overall mean, TGEijk was the interaction between grafting method, rootstock genotype and experiment, and was the experimental error. Analysis of variance was repeated using LSMEANS values to verify that our results were not influenced by missing data.

Results

We grafted over 2245 plants; representing nine tomato rootstocks with one common scion, using two grafting methods and repeated the experiment five times. High variation was observed among the experiments. Late autumn grafting success across rootstock ranged from 4% to a maximum of 88% (Exp. 1 and 2), spring grafting ranged from 19% to 100% (Exp. 3) and late spring ranged from 25% to 100% (Exp. 4 and 5).

Success for grafting the control ‘Maxifort’ ranged from 52% to 100% (using adhesive) and 16% to 80% (using tube grafting). This high variation between experiments was attributed to the seasonal weather variation. The air temperature, light, and humidity might have an influence on the greenhouse environment and subsequently on the healing chamber. The air temperatures ranged from 18 °C to 25 °C (Exp. 1 and 2), 25 °C to 27 °C

(Exp. 3) and 24 °C to 28 °C (Exp. 4 and 5). The total natural light ranged from 37 µmol m-2 s-1 to 397 µmol m-2 s-1 (Exp. 1 and 2), 31 µmol m-2 s-1 to 823 µmol m-2 s-1 (Exp. 3) and 42 µmol m-2 s-1 to 187 µmol m-2 s-1 ( Exp. 4 and 5). The Relative humidity of the air

ranged from 17% to 61% RH (Exp. 1 and 2), 27% to 59% RH (Exp. 3) and 37% to 70%

RH (Exp. 4 and 5). Despite the observed variation, there were significant differences observed for the main effects of rootstock genotype and grafting method (Table 3.1).

Rootstock genotypes

There were significant differences (P = 0.0126) among the rootstock genotypes for grafting success on the first and final week (week 1 and week 3) (Table 3.1). The rootstocks Maxifort (commercial variety) and SGH07-315 (hybrid) exhibited higher rates of success across all experiments at three weeks. Hawaii 7998 (H7998), LA2204 and

SGH07-326 had the lowest rate of grafting success, with < 50% success observed at the final week (Table 3.2; Figure 3.1). No differences existed between rootstock genotype for grafting success on the second week (week 2) (Table 3.2).

Grafting methods

There were also significant differences (P < 0.0001) in grafting success between grafting methods (Table 3.1). Across the five experiments and nine rootstocks, the survival success was higher using adhesives, compared to tube grafting over three weeks

(Figure 3.2). There were no significant interactions between grafting methods and rootstock genotypes (Table 3.1). In addition, the grafting success was higher than 50% for seven out of nine rootstocks using adhesive. On the other hand, only ‘Maxifort’ and

SGH07-315 survived above 50% using the tube-grafting method across all the experiments (Figure 3.3).

Field experiment

Survival in the field was unaffected by grafting method (P = 0.131), but was affected by the rootstock genotype (P = 0.012). ,Maxifort, and SGH07-315 rootstocks had the highest survival rates of 100% and 96%, respectively. The H7998 genotype rootstock had the lowest survival rate of 49% in the field study. Even though, no statistic differences were identified in yield between grafting methods and rootstock genotypes (P

= 0.131; P = 0.134, respectively), a numerical differences were detected between grafting method for both rootstock evaluated. The commercial rootstock ‘Maxifort’ had the highest yield per plant (14 kg•plant-1) when grafted using adhesive compared to only (11 kg•plant-1) using tube grafting method. The hybrid rootstock SG07-315 showed a similar trend of (12 kg•plant-1) and (9 kg•plant-1) for adhesive and tube grafting methods respectively.

Discussion

In addition to genetic backgrounds and grafting methods, many other factors influence grafting success, including post-grafting environmental conditions, plant vigor, carbohydrate content, and the proper match of vascular bundles (Bisognin et al., 2005).

Despite using the same rootstocks, scion and protocol, differences among experiments were noticed in this study (Table 3.1). The explanation for this variation may be due to fluctuation in temperature, light and humidity in greenhouse during the healing process, which may affect the environment on the healing chamber. Fluctuation in the temperature was observed between experiments, where the higher variation was in late autumn and

late spring (7 °C and 4 °C, respectively) , while during in spring temperature fluctuation was of (2 °C in average) during the three weeks of the study. Light intensity also fluctuated among experiments where in spring light intensity went from lower to upper gradually as the plants were healing. On the other hand in late autumn was inverted and the fluctuation was highly between days (data not shown). The late spring (2012) the light intensity registered in the greenhouse was lower compared to previous experiment.

We attribute this environment variation in the greenhouse to natural fluctuation because for all the experiments, temperature was set at 25/30 °C nigh/day and no additional light was provided.

Since the experiments were carried out during different seasons of the year including late autumn (Exp. 1 and 2), spring (Exp. 3), and late spring (Exp. 4 and 5), the day length could affect the grafting success due to limited or excess light available. In addition to day length, light intensity varies among the seasons. Direct sunlight may cause the healing chamber to heat up excessively or cause the grafted plants to lean towards the light, pulling the graft apart. This natural environmental variation still affects the plant in the healing chamber, specifically the temperature, relative humidity and amount of light absorbed by the shade cloth and plastic of the healing chamber (Johnson and Miles, 2011). Regardless of variance among experiments, survival rate were consistently higher using adhesives and for future work this factors should be, took into account to maximize the grafting success.

Graft healing and survival depends on the compatibility of scion and rootstock combinations, which can be affected by anatomical, physiological, and genetic variables

(Backhouse et al., 2008; Edelstein, 2004). Differences in graft success among tomato rootstocks are attributed to incompatibility between rootstocks and the scion used in this study. When tomato cultivars and wild plants (S. habrochaites) were used as a rootstock in grafting eggplant, the grafting success was low (Gisbert et al., 2011). The factors that determine how a rootstock and scion interact to produce a strong graft are still not very well understood. However, grafting failure often occurs at early stages of tissue joining

(Martinez-Ballesta, 2010), and could be classified into four categories: (1) failed bud growth, (2) virus-caused graft failures, (3) mechanical obstruction at the union and (4) abnormal union structure (Herrero, 1958; Andrews and Marques, 1993). In this study, failure of the scion meristem to grow (category 1) and abnormal union (category 4) were observed during the three weeks of data collection. We have previously observed poor success using H7998 as a rootstock and hybrids derived from this bacterial wilt tolerant parent (unpublished results).

The benefits of adhesive cyanoacrylates or related products have been shown extensively in other applications for joining tissues (Kinloch, 1997; Andrade et al., 2001;

Quintino and Pires, 2004; Albert and Job, 2004; Albuquerque et al., 2006; Pawar et al.

2008; Saska et al., 2009 and Singer et al., 2008). Adhesives were used in grafting plants in combination with grafting robots (Oda and Nakajima, 1992; Lee and Oda, 2003).

However, performance data has not been published (Kurata, 1994). In this study, we evaluated the benefits of ethyl-2-cyanoacrylate (Krazy Glue) to increase grafting success in tomato seedlings (Figure 3.3). Results revealed that survival was raised for Cherokee

Purple grafted onto nine different rootstocks using adhesive. A higher survival rate is

necessary to justify the added cost and labor of grafting and increase the returns for the farm or grafted plant grower.

Cyanoacrylate adhesives increased the rate of success in all rootstocks. Grafting success depends on the rootstock-scion interactions. The perfect combination of the rootstocks and scion parts will result in a successful plant with the potential to impart both abiotic and biotic stress in a given environment without decreasing yield or fruit quality. The improvement of grafting success across all genotypes suggests that adhesives may increase tissue union between rootstock and scion. This improvement could be due to the fast dry of the adhesives, and then the new seedling grafted is firmly held by this production. Thus, by drying fast it leaves minimum or no space between rootstock and scion. The same time alignment can be simplified and more tissues can be matched between the scion and rootstock, which will result in increased grafting success. Yield in open field study was not negatively affected by the use of adhesives in the grafting process. Furthermore, it was detected numerically higher yield compared to tube grafting for both rootstocks evaluated.

TABLES

Week 1w Week 2 Week 3 Source of variation df P-value df P-value df P-value z 8 0.0426 8 8 Genotype (Gen) 0.0727 0.0126 y 1 <0.0001 1 1 Grafting method (Trt) 0.0002 <0.0001 x 4 <0.0001 4 4 Experiment (Exp) <0.0001 <0.0001 32 0.0516 32 32 Gen:Exp 0.8293 0.8501 4 <0.0000 4 4 Treat:Exp 0.1278 0.0685 Gen:Treat 8 0.1452 8 0.7154 8 0.3965 Nine rootstock genotypes were compared for graft success LA1589 (S. pimpinellifolium, an inbred selection from LA2204 (S. habrochaites), FG02-188 (S. lycopersicum), Hawaii 7998 (H7998); the hybrids SGH07-316, SGH07-315, SGH07-320, SGH07-326; and Maxifort). yStandard tube graft using a Tygon tubing clip; grafting using adhesive in place of the clip. xExperiments were repeated five times, the first and second experiment in late autumn (17 and 24 November 2009 respectively), third experiment in spring (24 March 2010), fourth and fifth late spring (17 and 24 May 2012 respectively). wData were collected at week 1, week 2 and week 3 post-grafting.

Table 3.1: Analysis of variance for success rate of grafting ‘Cherokee Purple’ scion onto nine different rootstocks using two different grafting methods.

Week 1 Week 2 Week3 Genotype Grafting success (%)z Grafting success (%) Grafting success (%) SGH07-315 93 abcy 80 a 76 a Maxifort 94 ab 71 a 68 ab LA1589 93 abc 62 a 54 bc FG02-188 95 ab 63 a 53 bc SGH07-320 90 bc 62 a 53 bc SGH07-316 92 bc 64 a 50 bc H7998 94 ab 66 a 49 c LA2204A 89 c 48 a 44 c SGH07-326 98 a 57 a 42 c zTwenty-five plants were grafted per genotype /grafting method and 5 experiments were conducted over time. ySimilar letters in each column indicate nonsignificant differences among rootstocks at P = 0.05 based on least significant difference (LSD).

Table 3.2: Effect of rootstock genotype on success rates across all treatments at weekly intervals following the grafting process.

FIGURES

Figure 3.1: Effects of rootstock genotype on grafting success across all five experiments and grafting methods (Adhesives and tube grafting methods). Y-axis is the percentage of plants that were alive and growing after three weeks post-grafting. In the X-axis nine rootstock genotypes were compared for graft success LA1589 (S. pimpinellifolium, an inbred selection from LA2204 (S. habrochaites), FG02-188 (S. lycopersicum), Hawaii 7998 (H7998); the hybrids SGH07-316, SGH07-315, SGH07-320, SGH07-326; and ‘Maxifor’t) using a common scion (Cherokee Purple). Data represent the means and error bars indicate ± SE of five independent experiment (n=125).

Figure 3.2: Main effects of grafting method on grafting survival success across all five experiments and rootstock genotypes. Y-axis is the percentage of plants that were alive and growing after one, two and three weeks post-grafting. In the X-axis tow grafting methods (A = Grafting using adhesives; T = tube grafting) were compared. Data represent the means and error bars indicate ± SE of five independent experiment and nine rootstocks (n=1125).

Figure 3.3: Effects of rootstocks genotype and grafting method on grafting success at week 3 post-grafting. Data represent the means and error bars indicate ± SE of three independent experiment (n=75). (A = Using adhesives; T = Tube grafting) three experiments (During Spring to late spring).

Chapter 4 : Analysis of Quantitative Trait Loci (QTL)

associated with graft failure in tomato

Abstract

Grafting failure can be caused by multiple factors and can occur at early or late life cycle of the grafted plant. This failure often is attributed to grafting compatibility between scion and rootstock. The ability to predicted failure or success of grafting would help to avoid use of non-compatible rootstock and scion partners. We examined grafting failure and conducted a quantitative trait loci (QTL) analysis to elucidate the genetic basis of grafting failure. We evaluated an inbred backcross population (BC3S4) and an F2 population developed from crosses elite lines of Solanum lycopersicum L. (H7998 ×

OH88119) and (H7998 × Ohio MR13), respectively. H7998 had been identified as a parent with poor grafting success. We collected survival data for 3 weeks post-grafting.

Scion height was also recorded for the F2 population at week 3 post-grafting. We used single factor analysis of variance to detect marker-trait association. Our results suggest that a putative QTL is underlying grafting failure on chromosomes nine, and two QTLs for scion growth (height) on chromosome 2 and 4. If confirmed, these results can be exploited for rootstock improvement and to gain further understanding of QTL underlying graft failure during the healing process. 81

Introduction

Grafting is the ability to unify two plants, the scion (upper plant) and the rootstock (bottom plant) in order to produce a single plant (Kramer, 1979; Andrews and

Marquez, 1993). This technique has historically been used in fruit trees due to their extended reproductive lifecycle (Kramer, 1979). During the early 1920’s this technique was introduced into annual crops to reduce diseases caused by soil pathogens (Lee,

1993). Later, it was used to impart abiotic tolerance and to enhance nutrient uptake in vegetables (Lee et al., 1998; Kubota et al., 2008; Rivard et al., 2010).

The success of grafting depends on various factors including taxonomic differences, environment, availability of oxygen and water, physiological stage of rootstock/scion, diseases, pests, herbicide toxicity and the ability of the grafter to perform successful grafts (Andrews and Marquez, 1993). A success in grafting encompasses a complex biochemical and structural sequence of events: (1) wound response, from rootstock and scion; (2) callus bridge formation; (3) vascular differentiation; and (4) production of secondary xylem and phloem. Extensive reviews detailing these steps have been published elsewhere (McCully, 1983; Moore, 1984; Andrew and Marquez, 1993;

Kawaguchi et al., 2008; Martinez-Ballesta et al., 2010). The success of grafting in plants depends on the successful development of the events noted above between the scion and rootstock.

Grafting is not always successful; multiple factors contributed to grafting failure

(Masa, 1989; Leoni et al., 1990; Edelstein, 1999, 2004). Poor development of the graft union due to anatomical mismatching, poor craftsmanship, adverse environmental

conditions, disease, and graft incompatibility (Pease, 1933; Herrero, 1951; Andrews and

Marques, 1993; Hartmann et al., 2002).Graft incompatibility is defined as the interruption or delay in the development of the new grafted seedling or growing plant (Santamour and

Frank, 1988; Kollman and Glockman, 1985). This interruption can have an anatomical basis (Mosse, 1962), be affected by physiological intolerance at the cellular level (Feucht,

1987; Moore and Walker, 1981), and be caused by virus or phytoplasma transmission

(Mircetich et al., 1980; Uyemoto, 1989; Hartman et al., 2002). The examples of virus- induced incompatibility include black line in English walnut caused by cherry leaf roll virus (Mircetich et al., 1980), prune brown line caused by a strain of tomato ring spot virus (99; Mircetich and Hoy, 1981) and apple union necrosis and decline caused by virus

(Cummins and Gonsalves, 1982). Incompatibility could also be directly related to undergrowth and overgrowth of the scion relative to that of the rootstock, due to unknown reasons (Salesses and Al Kai, 1985; Lee, 2006).

In tomatoes, a limitation to grafting success is proposed to be Tobacco Mosaic

Virus (TMV) and resistance genes. Three different alleles (Tm-1, Tm-2 and Tm-2a) against TMV have been identified. Plants possessing the Tm-1 allele are symptomless or tolerant to TMV , those possessing the Tm-2a allele confer a hypersensitive response, and those with the Tm-2 allele are intermediate (Hall, 1980; Yamakawa, 1982; Oda, 1995;

King et al., 2010). In 1981, Yamakawa suggested that different combinations of these alleles in tomato rootstocks and scions could influence whether the grafting process is successful or not. To our knowledge, this is the closest genetic study that has been conducted for grafting compatibility due allelic interaction between scion and rootstock.

Most studies of grafting failure have focused on the anatomical, physiological and biochemical factors (Moore, 1983; Kawaguchi et al. 2008 and Andrew and Marquez,

1993). Association between genetic relationships and graft failure was suggested for grafts peach-plum (Grasselly, 1968), almond-plum (Kester et al., 1965) and Copes

(1973), studied the inheritance of graft compatibility in Douglas-fir, suggesting multiple genes are involved in this process. On the other hand, studies in human, frogs and mice demonstrated the relationship between genetics and graft incompatibility (Ceppellini et al., 1966; Bistoni et al., 2012). Despite of numerous studies conducted on grafting success in plants, the information of genetic basis of grafting failure was not revealed.

The objective of this study was to identify the genetic basis of grafting failure between tomato rootstocks and scions during the early portion of the lifecycle.

Materials and Methods

Plant materials and experimental design

Two segregating populations were used to determine the genetic basis of graft failure using marker-trait analysis. Our experience with H7998 suggested that this parent had a low rate of success during grafting with multiple scions (e.g. Celebrity, NC84173,

Cherokee Purple and Ohio FG99-218). The initial population was created by crossing the

H7998 (low grafting success) × OH88119 (high grafting success). An inbred backcross

(BC3S4) population was developed using OH88119 as a recurrent parent details of this population are found somewhere else (Yang et al., 2005). The population size for this study consisted of 172 BC3S4 families. The second population consist of 288 F2 plants

from a cross between H7998 (low grafting success) and Ohio MR13 (high grafting success). A common scion was used in both experiments a fresh marker material

“NC84173”.

All tomato seeds were sown in 288-cell flats with a volume of 13 mL (Hummert,

EARTH City, MO-USA) using PRO-MIX (Premier horticulture, Quakertown, PA-USA).

A month after seeding, the rootstock seedlings were transplanted into 40-cell trays with a cell volume of 68 mL (Hummert, EARTH City, MO-USA) using PRO-MIX. The seedlings were grown for one additional week prior the grafting process in the greenhouse. Temperatures were set at 24/18 °C day/night and a 12 h of photoperiod.

Two independent experiments were conducted in the greenhouse facilities at the

Ohio Agricultural Research and Development Center (OARDC) in Wooster, Ohio. The initial experiment was a randomized complete block design (RCBD) with two blocks.

The BC3S4 families (OH88119 ×H7998) were used as rootstocks and NC84173 was used as the scion. Ten plants were grafted per each family. In the second experiment, F2 plants derived from OhioMR13 × H7998 were used as rootstocks and NC84173 was used as the scion. A completely randomized design (CRD) was used distributing 288 F2 plants in the healing chamber after the grafting process.

DNA extraction

Leaf tissue of approximately 1-cm2 was collected from each rootstock used in the grafting process for the F2 population. The leaf tissue for BC3S4 was bulked from the 10 plants used as a rootstock for each family. Total genomic DNA was extracted using the

modified CTAB protocol (Kabelka et al., 2002) scaled for 96-well format. Polymerase chain reaction (PCR) was conducted as described previously (Yang et al., 2005; Robbins et al., 2010; and Wang et al., 2011).

Molecular markers and genotyping

The initial BC3S4 population was genotyped using 78 polymorphic markers including, simple repeat (SSR), Insertion/deletion (INDELS), and Single Nucleotide

Polymorphism (SNP) were selected based on their ability to distinguish H7998 and

OH88119 (Yang et al., 2005). These markers were distributed across the genome with 2-

17 markers per chromosome. The F2 population was genotyped using 40 Single

Nucleotide Polymorphism (SNP) markers as implemented in the Infinium Golden-gate assay (Illumina Inc., San Diego, CA, USA) to detect polymorphisms between H7998 and

OhioMR13. The forty SNP markers were scattered throughout the genome.

Genotyping was conducted on three platforms, one for size polymorphism (SSR and INDELS). Size polymorphisms were detected using polyacrylamide gels on the Li-

Cor-IR2 4200 system (LI-COR bioscciences, Lincoln, NE) or agarose gels. SNPs were detected by an Allele Specific Primer Extension (ASPE) assay (Lee et al., 2004). Samples were loaded at 50 ng/ul on an Ilumina BeadXpress analyzer system (Luminex

Corporation, Austin, TX). For the second experiment, Golden-Gate assay was performed following manufacture’s protocol of Ilumina (Ilumina, San Diego, CA).

Grafting and healing process

Seven days following transplanting, rootstocks were grafted using cleft grafting method (Oda, 1999). The stem of the scion were cut in a wedge, and the tapered scion was fit into a cleft cut in the end of the rootstocks. The graft union was held firm with a plastic tube to ensure contact vascular tissue between partners (scion-rootstock) (Oda,

1999).

The grafted plants were placed in a 5.5 m × 1.2 m × 0.7 m dimension healing chamber constructed from ¾ - inch PVC pipe as a frame and 0.08-mm plastic covering to maintain high humidity. The greenhouse bench was covered with capillary mat

(Agriculture solutions, Columbus, IN) which was maintained saturated with water. Shade cloth (60 shade) was used for the first week to minimi e water loss through transpiration. Temperatures were maintained at 28-30 C. After 7 days, the shade cloth was gradually removed and the healing chamber was opened to increase light and lower humidity. By 14 days post-grafting, the shade cloth and plastic were completely removed.

Phenotype for grafting success

Grafting failure was measured as percent survival of the 10 rootstocks grafted per each BC3S4 family at week-3 post-grafting. For the F2 population, survival was determined on each individual rootstock; data was collected at week 1, week 2 and week

3. We also collected the height of the scion grafted during the final week (week 3). An independent experiment was carried out during springtime of 2013. We grafted 120 plants in combinations: H7998 self-grafted (autograft), H7998 with NC84173

(homograft) and NC84173 self-grafted (autograft). The experimental design was a randomized complete block design with four blocks. Survival data was collected along the three weeks post grafting.

Statistical analysis

Best Linear Unbiased Predictors (BLUPs) were estimated in order to account for the variance associated with different grafters and experimental replications for BC3S4 population (H7998 × OH88119). The model used estimation of BLUPs was:

Yijk = μ + Ri + Gj + Ek + REik) + RGEijk + εijklm where Yijk was the percentage of grafting success μ was the overall mean, Ri was the effect due to ith rootstock genotype, Gj was the effect due to grafter, Ek was the effect due to kth experiment. Two and three way interactions were added to the model including, REik was the effect due to the interaction between rootstock genotype and experiment, RGEijk was the effect due to the three-way interaction between the tree main effects, and εijklm was the experimental error. In this design, all effects were random.

For the estimation of BLUPs Lme4 package (Bates et al., 2011) was used in “Rstudio” computer software version 0.97.320 (RStudio, 2012) in concert with the R core package version 2.15.3 (R Core team, 2012).

To determine the association between grafting success and DNA-based markers, data were analyzed using analysis of variance (ANOVA) based on general linear model.

For statistical analysis, we used the statistical package in “Rstudio” computer software

version 0.97.320 (RStudio, 2012) in concert with the R core package version 2.15.3 (R

Core team, 2012). The model used was:

Yij = μ + Mi + εij where Yij was the BLUPs value for grafting success (phenotype data), μ was the overall mean, Mi was the effect due to marker class, and εij was the experimental error. The approximate F test for marker-trait associations (Mi) was Mi/εij.

Marker-trait associations were tested in the F2 population using analysis of variance (ANOVA) through the general linear model. Phenotypic data was represented by the mean values of survival (dead = 0, stunted = 1, alive = 2) of the grafted seedling during three weeks post-grafting. For the last week (week 3), the phenotypic value was represented also by the mean values height scion. The model used was:

Yij = μ + Mi + εij where Yij was the quality measured (survival values and high of scion), μ was the overall mean, Mi was the effect due to marker class, and εij was the experimental error. The statistical package in RStudio computer software version 0.97.320 (RStudio, 2012) was used in concert with the R core package version 2.15.3 (R Core team, 2012).

For the independent experiment, ANOVA was performed to determine grafting capacity of H7998. The Agricolae package in R (De Mendiburu, 2012) was used to perform mean separation. Mean separations were based on Tukey’s test (P < 0.05)

Results

The marker data was obtained from genotyping the populations BC3S4 with a total of 78 markers. Markers were distributed from 2 to 19 markers per chromosome. In F2, population markers were mapped from 0 to 7 markers per chromosome, with a total of 40 markers. Our aim was to identify at least one marker per chromosome; however the marker coverage was uneven with only two markers on chromosomes 8 and 12 in BC3S4 population and no marker was present on chromosome 5 in F2 population.

The frequency distribution of the grafting success/failure effect in the BC3S4 families were continuous (Figure 4.1-A), indicating that grafting success/failure is controlled by multiple genes. The BLUPs-value ranged from -100 to 130 (Figure 4.1-A).

The survival data for F2 population show binary distribution (Figure 4.1-B); it shows a clear segregating population. The scion height phenotype for F2 population show skewed distribution (Figure 4.1-C). The skewed distribution appears to be due to the high number of plants that died before week 3, which is, continuous and not binary.

Analysis of marker-trait associations demonstrated potential linkage between markers and grafting failure on chromosome 9 and chromosome 5. On chromosome 9, a marker SSR70 is highly associated with percentage of graft failure (P = 0.002196).

Alleles from the H7998 parent are associated with grafting failure (BLUPs value = -

28.69). There is some evidence for a negative association with OH88119 alleles on chromosome 5 (Table 4.3).

F2 population (H7998 × Ohio MR13) was developed to confirm the genetic basis of grafting failure in allele coming from H7998. To facilitate the analysis of marker-trait

association, we decided to collect one objective phenotype data (height of scion) and survival data, at final week (week 3). Using these data, we detected two markers on chromosome 2 and 4 that were significantly associated with low scion height (Table 4.4;

Table 4.5).

The survival data for F2 population did not follow 1:3 segregation ratios, which indicates that multiple genes determine grafting failure. This result agrees with the BC3S4 phenotype data, since it revealed continuous distribution of the BLUPs-value. In F2 population, the markers sol_cap_snp_sl_20325 and sol_cap_snp_sl_8464 on chromosome 2 were highly associated with grafting failure. The rank of means indicates alleles associated with Ohio MR13 have higher grafting survival compared to H7998, and the heterozygous resulted in the survival mean (Table 4.3). This result is inconsistent with the allelic effect observed in the BC3S4 population, where H7998 genotypes had the lowest graft compatibility with the scion NC84173.

The growth of scion varied markedly among the F2 rootstock population.

Association was detected on chromosome 2, 4 for growth of scion. On chromosome 2, a marker Sol_cap_snp_ls_20052 was highly associated with scion growth (P = 0.04600).

Alleles from H7998 parent is associated with reduction of scion growth, the mean height

3.59 cm compared to 7.65 cm for plants associated with alleles from Ohio MR 13. On chromosome 4 we observed the same trend with marginal significance on association between low scion height and alleles from H7998 parent (Table 4.5).

After a period of 3 weeks, the grafting failure expressed in success value for

H7998-self-grafted decreased from 1.1 to 0.3, H7998 × NC84173 1.4 to 0.8, and

NC84173 self-grafted 1.7 to 1.5 (P < 0.001). This result strongly suggests that H7998 presents grafting rejection not only foreign tissue also to its own (i.e. intolerance to grafting) (Table 4.6). This phenomenon was before reported in Douglas-fir (Copes,

1974).

Discussion

The goal of this research was to assess the genetic basis of grafting failure in populations derived from H7998 × OH88119 and H7998 ×Ohio MR13 via single factor

QTL analysis.

We used grafting success and scion height for evaluation of failure at an early stage in tomato. Two populations were used as rootstock and one common scion in two independent experiments. Grafting success was recorded along the 3 weeks-post grafting.

In the BC3S4 population (H7998 × OH88119), survival of 10 plants was used as percentage of survival, while in F2 population (H7998 × Ohio MR13), the data collected data was categorical including; dead = 0, stunted = 1 and alive =2. We identified a potential marker from H7998 that was associated with the grafting failure. Single-marker analysis in BC3S4 population implicates one genomic location chromosome 9. The phenotype data suggest that this trait is a quantitative trait. This result supports the early suggestion that grafting compatibility is a quantitative trait that is controlled polygenically (Copes, 1973).

Subsequent testing in F2 population, we fail to validate the marker on chromosome 9 associated with grafting failure with allele from H7998 parent.

Nevertheless, the phenotype data (scion height) show a continuous distribution, confirming that grafting failure is a trait controlled by multiple genes. Even we fail to validate the marker on chromosome 9 we found association of grafting failure on chromosome 2 associated to allele of H7998 parent. These results prove the importance of confirming the findings when associating traits of interest.

Our results from two independent populations have not validated the markers associated with grafting failure unanimously. QTL effect size, high environment influence, lack of overlap marker set between populations limited our ability to identify the QTL associated with grafting failure. Quantitative trait loci (QTL) with small effects on phenotypic variation can be difficult to detect and analyze.

A high level of failure was observed when H7998 was used as a rootstock even when it was auto-grafted. This observation may explain why this specific accession performs poorly when used as a rootstock. Grafting rejection was before reported in

Douglas-Fir (Copes, 1974) who suggested it was caused by genetic additive, thus, grafting for compatible rootstock would be possible by genetic selection.

TABLES

Locus Type Chr Pos (bp) ForwardPrimer ReversePrimer Temz REy CosOH47 SNP 1 1,096,992 TTGCTGATTTTCTTCCCATTTT GCAGCTGGAGTGAGAGGAAC 54 BstU I LEOH36 SNP 1 2,648,884 TCACAAAAATGGCGATGAGA CCACCTGTGGATCCTTGACT 56 Bcl I SSR134 SSR 1 -- CCCTCTTGCCTAAACATCCA CGTTGCGAATTCAGATTAGTTG 50 .. SL10939 SNP 2 1,514,029 GGGAGCTTCTTGTTTGTGATG AAACGGGCAACTGGACAA 55 .. SL10351 SNP 2 1,578,934 CTGTGCTGCTAGTGCTACTGCT AACTAGGGCAACACTTCCCA 55 .. TOM11 SSR 2 24,729,349 ATTGTAATGGTGATGCTCTTCC CAGTTACTACCAAAAATAGTCAAACAC 45 .. SSR71 SSR 2 26,961,020 AAATGGCATGGAGAATGGAA CATCCACTGAGAGCCCAAAG 40 .. LEOH23.3 SNP 2 33,674,756 CTATGCGTTTGTCGGTCGT CAAGGTAGTTGAAGGTATGACCA 54 Tsp509 I SSR104 SSR 2 34,116,722 TTCCATTTGAATTCCAACCC CCCACTGCACATCAACTGAC 50 .. SSR96 SSR 2 34,290,813 GGGTTATCAATGATGCAATGG CCTTTATGTCAGCCGGTGTT 50 .. SL10050 SNP 2 34,879,926 AAGGTGGTGATGTAATGGTGC TTCTATGGCTGCATGTACGG 55 .. SL10649 SNP 2 34,922,754 ATGGGTTGGCTCCTCACTC CAAGAGGTCCAAGGTTAATGC 55 .. SL10790 SNP 3 2,053,125 CCCTCCGATAGCAACATCAC TGCACCACTTTCTGGGTCTA 55 ..

94 SSR111 SSR 3 52,956,347 TTCTTCCCTTCCATCAGTTCT TTTGCTGCTATACTGCTGACA 50 ..

TOM59 SSR 3 60,133,677 TAACACATGAACATTAGTTTGA CACGTAAAATAAAGAAGGAAT 45 .. SL10042i INDEL 3 60,801,881 ATTTGCCGACACGTCATTTC CACTCATCATGACAAAGACAACC 45 .. CosOH51 SNP 3 60,856,997 CTCATTTGATACCTCTATTTGTGGTG TGAGATCTTAAAAGAAACACATGAGG 56 Rsa I SL10952 SNP 4 222,322 GGTTAAAGCCAGCCGTGATA AGCAAGGTCGCATTCATCTC 55 .. SSR310 SSR 4 41,915,678 GCGATGAGGATGACATTGAG TTTACAGGCTGTCGCTTCCT 50 .. SL10485 SNP 4 46,569,638 AGCACTCGGTGAGCTTGTTAG ATCCTCATCCTCCAGTGCAT 55 .. SSR450 SSR 4 47,205,488 AATGAAGAACCATTCCGCAC ACATGAGCCCAATGAACCTC 50 .. SSR306 SSR 4 47,205,495 ACATGAGCCCAATGAACCTC AACCATTCCGCACGTACATA 50 .. SL10809 SNP 4 54,865,292 ACCAATGTGTCAAAGGGTGT CATCAGACGTTCAATCATGC 55 .. yRE, restriction enzyme; z tem. Annealing temperature of the marker.

Continued

Table 4.1: Polymorphic molecular markers used in the population derived from the cross OH88119 × H7998

Table 4.1 continued

Locus Type Chr Pos (bp) ForwardPrimer ReversePrimer Temz REy SL10215 SNP 4 54,981,611 TCCCGTAATCTGGCGTTACT GAAACTGGCATCCACATGAA 55 .. LEOH26 SNP 4 61,335,167 GAAGATTCGGAGGTCAAACG AAAGGCTTCACATCTTCACCA 52 Fok I SL10136 SNP 4 62,155,819 GTCTCGGCCTATCTGCTTTG GCTTCTCAAACTCCTTCCCTT 55 .. SL10556 SNP 4 62,476,187 CACAAGTGGTGCTGAGGAAAT TGAGGCGGAATGAAGAAGAT 55 .. SL10184 SNP 4 63,392,677 GATCGGAGATGAAACGGAGA CACAATGCTTCAATCCAACG 55 .. SL10184i INDEL 4 63,393,232 TCTCAGAAAATTCTCCATGTAATTC CGATCCACATCTGGTTATCAAAG 45 .. LEOH37 SNP 4 -- TTGATATATTCCATGTGTGTCTC AACTACAAATTAACAAACTTAAATGG 51 NmuC I SSR146 SSR 4 61,314,392 TATGGCCATGGCTGAACC CGAACGCCACCACTATACCT 50 .. SSR43 SSR 4 2,127,725 CTCCAAATTGGGCAATAACA TTAGGAAGTTGCATTAGGCCA 50 .. SSR603 SSR 4 5,290,324 GAAGGGACAATTCACAGAGTTTG CCTTCAACTTCACCACCACC 50 .. SSR627 SSR 4 -- TACAGAATAGGGTTTGCCATA GTTTTAGTGGGTTGTGTTGAA 48 .. SSR638 SSR 4 -- TGTTGGTTGGAGAAACTCCC AGGCATTTAAACCAATAGGTAGC 50 .. TOM210 SSR 4 -- CGTTGGATTACTGAGAGGTTTA ACAAAAATTCACCCACATCG 45 ..

95 SL10639 SNP 5 8,214,475 GGTCTCACAATGTTGCTTACTG TTCGCAATCAGTTTAGGGTC 55 ..

SL10469 SNP 5 13,141,819 TGTCGATTTCTGGTTCACCT CTAACTACCACCTCTCTTAACCACC 55 .. SL10526 SNP 5 18,738,337 GAACTGTCTACCATACTTGAAGGG AGGTCGGAGTTTCCAGAAGA 55 .. LEOH63 SNP 5 31,653,002 CCCCTCGTGTAGGTGTCACT ATCCTCCGATCATCAGCAGT 52 Taq I LEOH9.1 SNP 5 34,412,895 GGCAATGCCACTGACTTACA CTCTCTGCTGCTTCGGCTAC 55 Aci I LEOH20.1 SNP 5 34,412,915 CAGACCTAACAAGACAGGCAAA ATCAGGCATGACCATGGAAG 55 Hae III SL10373i INDEL 5 50,850,031 CCGGAAAGCATTGTTGTAGC ACAGTTGAAATTGCCTGAGC 45 .. TOM49 SSR 5 59,033,822 AAGAAACTTTTTGAATGTTGC ATTACAATTTAGAGAGTCAAGG 45 .. SL20210i INDEL 5 59,256,534 GAGGTATCAAAGTTATGCTTTCAC TCGATTAGTTGAGCTAGTTATTCC 45 .. SL20097 SNP 5 60,290,332 TCTGCAATAGCCATACTGGAAG GGATGGATCTTTGGAACCAG 55 .. yRE, restriction enzyme; z tem. Annealing temperature of the marker.

Continued

Table 4.1 continued

Locus Type Chr Pos (bp) ForwardPrimer ReversePrimer Temz REy SL10591 SNP 5 61,000,438 CTGATCTCATTGCCTTTAACCC ACTCCAGTATGACCAGCTCCA 55 .. Rx3-L1 SNP 5 62,493,894 CTCCGAGCGAAGAGTCTAGAGTC GAAGGCAAAAGGAAAAGGAGAAGGATGG 60 BsrB I CosOH73 SNP 5 63,554,544 CTTCCCGACAAGCACAAAAA CGAATGCTCTGTACCATTTCC 56 Alu I SL10963 SNP 5 -- TTGCCAAGGAATAGGACAGA GAAGATACGCGAAACAATGG 55 .. TOM152 SSR 5 -- ATTCAAGGAACTTTTAGCTCC TGCATTAAGGTTCATAAATGA 45 .. SL10401 SNP 6 3,290,013 GGGAGACTCATCCTGATTGC ATTTGTTCCTCCCAATGCTG 55 .. SL10187 SNP 6 12,751,418 GTGAAAGAAATCCGGGTGAA AGATTATGCCCTCCTCGATGT 55 .. SL20059 SNP 6 80,405,248 CCAATTCCTCTTCGTGATGC CTTCCATTACCACCTTGTCCA 55 .. LEOH343 SNP 8 2,583,051 CAAATGGGTTTGGCTGAAAA CGCAAACTGATTTGAACAGC 52 Mnl I SL10044 SNP 8 61,310,899 AATGGTATTGCTGAGCGGTT ACATCATATCCAGCCCAAGTG 55 .. LEOH8.4 SNP 9 1,296,246 CCACTGATCAATGTGGTGGA CAACCACAAATGGCTCCTAAA 55 Hae III Cosi52 INDEL 9 3,490,540 GCCTTTCTTCCAGGATGCTA CCCATTTTCCTTCTTCCTAGA 52 30 bp indel SSR69 SSR 9 3,545,307 TTGGCTGGATTATTCCTGTTG GCATTTGATAGAAGGCCAGC 50 ..

96 SSR70 SSR 9 3,628,939 TTTAGGGTGTCTGTGGGTCC GGAGTGCGCAGAGGATAGAG 50 ..

SL10004 SNP 9 4,062,850 CACGACATTAGCTGTTGGGTT TGGAGCTGGAATCAGTGTTG 55 .. TOM236 SSR 9 65,218,304 GTTTTTTCAACATCAAAGAGCT GGATAGGTTTCGTTAGTGAACT 45 .. SL10807 SNP 10 4,062,850 CACGACATTAGCTGTTGGGTT TGGAGCTGGAATCAGTGTTG 55 .. LEVCOH15 INDEL 10 58,124,437 GCAACCACCAATGTTCATTACA AAGCTAAATCTGGCTTGTGGAG 52 .. SSR248 SSR 10 52,294,227 GCATTCGCTGTAGCTCGTTT GGGAGCTTCATCATAGTAACG 50 .. SSR318 SSR 10 -- GCAGAGGATATTGCATTCGC CAAACCGAACTCATCAAGGG 50 .. TOM196 SSR 11 5,983,764 CCTCCAAATCCCAAAACTCT TGTTTCATCCACTATCACGA 45 .. TOM144 SSR 11 14,565,287 CTGTTTACTTCAAGAAGGCTG ACTTTAACTTTATTATTGCGACG 45 .. SL10915 SNP 11 34,441,452 GGCAAAGGAAGAAATCCCAT GAGTTCCACCATGAGACAAGC 55 .. yRE, restriction enzyme; z tem. Annealing temperature of the marker.

Continued

Table 4.1 continued

Locus Type Chr Pos (bp) ForwardPrimer ReversePrimer Temz REy SL10781 SNP 11 36,270,048 CTGCAACTGGGAATGAGGTC ACACCAGACTAGACGCCTTGA 55 .. SL10737i INDEL 11 47,457,610 CCACTCCTGGGACTCAAATC TGGACCCACAGGTAATGAGG 45 .. CosOH57 SNP 11 49,863,314 TGCCCAAAAGCACAGTACAA CGCCTCCTATCTTCCAAACTT 56 Tth111 I SL20181 SNP 11 50,391,207 TCGACTATGCCATTTGCTTG TTCAGGTGCAGTAGAAAGCTCA 55 .. SL10027 SNP 11 53,186,408 GATCCAGTTTCCCTCTTACCG CCATTAGAGCCAAGACGCTC 55 .. SL10615i INDEL 11 -- TTTCCACAGGTCATTCTTCC TGGGGGATGAGAGTTGTAATG 45 .. SL10715 SNP 11 -- CATTGGGACTGCTTGCAC GAGACAATTCCCAGAATCCA 55 .. SL10831 SNP 11 -- AATTCGGCAGGCTTTCTGT TTTGTGGACCTGTTGCTCC 55 .. SL10953i INDEL 12 3,785,881 CTGTCTCTCGCTTTTCTCCTG ACGGAACACACCCTAAGTGC 45 .. SSR20 SSR 12 6,759,379 GAGGACGACAACAACAACGA GACATGCCACTTAGATCCACAA 50 .. yRE, restriction enzyme; z tem. Annealing temperature of the marker.

Locus Chr Pos_(bp) SNP solcap_snp_sl_8655 1 2,446,679 [A/G] solcap_snp_sl_8656 1 2,446,475 [T/C] solcap_snp_sl_9758 1 76,901,815 [A/G] 241_2F_264_241_2b_60_b 2 37,376,666 [T/C] solcap_snp_sl_12647 2 21,285,017 [T/C] solcap_snp_sl_20052 2 48,684,930 [T/G] solcap_snp_sl_20325 2 31,821,452 [T/C] solcap_snp_sl_36443 2 30,745,666 [A/G] solcap_snp_sl_8464 2 33,122,529 [A/C] SGN-U565563_snp1042_solcap_snp_sl_62467 3 57,379,505 [T/C] solcap_snp_sl_19508 3 7,988,023 [T/C] solcap_snp_sl_20723 3 61,134,917 [C/G] CL017321-0298_solcap_snp_sl_47277 4 61,090,851 [A/G] solcap_snp_sl_14104 4 53,220,225 [A/G] solcap_snp_sl_1698 4 7,217,339 [C/G] solcap_snp_sl_21376 4 2,927,112 [T/C] solcap_snp_sl_25082 4 3,504,152 [C/G] solcap_snp_sl_29326 4 63,478,227 [A/T] solcap_snp_sl_47540 4 62,129,421 [T/C] solcap_snp_sl_15515 6 1,605,118 [A/G] solcap_snp_sl_24361 6 34,952,365 [A/C] solcap_snp_sl_34975 6 3,502,335 [A/G] solcap_snp_sl_11171 7 1,890,767 [T/C] solcap_snp_sl_22770 7 53,094,298 [A/T] solcap_snp_sl_51838 7 55,250,609 [T/C] solcap_snp_sl_5840 7 57,417,265 [A/T] solcap_snp_sl_7028 7 60,956,980 [A/G] solcap_snp_sl_34763 8 58,879,586 [T/C] solcap_snp_sl_31883 9 17,881,415 [T/C] solcap_snp_sl_36856 9 65,167,417 [T/C] CL017176-0241 10 63,002,557 [A/G] solcap_snp_sl_13202 10 1,163,845 [T/C] solcap_snp_sl_61192 10 62,487,601 [T/C] solcap_snp_sl_8000 10 46,998,997 [A/G] solcap_snp_sl_8825 10 63,847,842 [T/C] solcap_snp_sl_12406 11 11,934,846 [T/C] solcap_snp_sl_14367 11 7,715,510 [A/C] solcap_snp_sl_17571 11 52,801,129 [A/C] solcap_snp_sl_29505 11 8,626,274 [T/C] solcap_snp_sl_1572 12 4,038,515 [T/C]

Table 4.2: Single Nucleotide Polymorphisms (SNP) molecular markers used in the population derived from the cross Ohio MR13 × H7998.

Marker Chromosome Allele BLUPs-value P-value Genotypez Mean y SSR70 9 H7998 -28.69 bx 0.002196 SSR70 9 OH88119 34.68 a SSR70 9 H - Cosi52 9 H7998 -32.12 a 0.0714 Cosi52 9 OH88119 33.49 a Cosi52 9 H - SL10591 5 H7998 52.77 a 0.01814 SL10591 5 OH88119 -56.08 b SL10591 5 H -56.21 b x BLUPs values were estimated in order to account for the variance associated with different grafters and experimental replications. y Entries with the same grouping letter are not significantly different based on the LSD (0.05). z H7998 = Homozygous for the Hawaii 7998 allele, OH88119 = Homozygous for the OH88119 allele; H= heterozygote.

Table 4.3: Single factor analysis for association in BC3S4 (OH88119 × H7998) population for grafting success at week 3.

Survival at week 1 Marker Chromosome Allele Genotypez Survival Meany P-value sol_cap_snp_sl_36443 2 H7998 1.71 abx sol_cap_snp_sl_36443 2 OhioMR13 1.89 a 0.04866 sol_cap_snp_sl_36443 2 H 1.59 b sol_cap_snp_sl_20052 2 H7998 1.71 ab sol_cap_snp_sl_20052 2 OhioMR13 1.94 a 0.07576 sol_cap_snp_sl_20052 2 H 1.63 b sol_cap_snp_sl_8000 10 H7998 1.60 a sol_cap_snp_sl_8000 10 OhioMR13 1.60 a 0.08621 sol_cap_snp_sl_8000 10 H 1.83 a Survival at week 2 Marker Chromosome Allele Genotypez Survival Meany P-value sol_cap_snp_sl_8464 2 H7998 1.47 ax sol_cap_snp_sl_8464 2 OhioMR13 1.39 ab 0.02691 sol_cap_snp_sl_8464 2 H 0.91 b sol_cap_snp_sl_20325 2 H7998 1.43 a sol_cap_snp_sl_20325 2 OhioMR13 1.35 ab 0.0587 sol_cap_snp_sl_20325 2 H 0.93 b sol_cap_snp_sl_20052 2 H7998 1.07 b sol_cap_snp_sl_20052 2 OhioMR13 1.61 a 0.08454 sol_cap_snp_sl_20052 2 H 1.06 b Survival at week 3 Marker Chromosome Allele Genotypez Survival Meany P-value sol_cap_snp_sl_20325 2 H7998 1.30 abx sol_cap_snp_sl_20325 2 OhioMR13 1.61 a 0.04697 sol_cap_snp_sl_20325 2 H 0.79 b sol_cap_snp_sl_8464 2 H7998 1.33 ab sol_cap_snp_sl_8464 2 OhioMR13 1.39 a 0.01329 sol_cap_snp_sl_8464 2 H 0.76 b xData were recorded as (Alive = 2; Stunted =1; Dead = 0) y Entries with the same grouping letter are not significantly different based on the LSD (0.05). z H7998 = Homozygous for the Hawaii 7998 allele, OhioMR13 = Homozygous for the Ohio MR13 allele; H= heterozygote.

Table 4.4: Single factor analysis of association with F2 population for grafting success at week 1, week 2 and week 3 post-grafting.

100

Marker Chromosome Allele Height Mean P-value Genotypez (cm)y sol_cap_snp_sl_20052 2 H7998 3.59 bx 0.04600 sol_cap_snp_sl_20052 2 OhioMR13 7.65 a sol_cap_snp_sl_20052 2 H 5.09 ab sol_cap_snp_sl_1698 4 H7998 3.25 b 0.0512 sol_cap_snp_sl_1698 4 OhioMR13 6.06 ab sol_cap_snp_sl_1698 4 H 6.18 a sol_cap_snp_sl_31883 9 H7998 3.01 a 0.08987 sol_cap_snp_sl_31883 9 OhioMR13 4. 83 a sol_cap_snp_sl_31883 9 H 6.06 a xScion height was measure from the graft union to the meristem y Entries with the same grouping letter are not significantly different based on the Tukey’s test (0.05). z H7998 = Homozygous for the Hawaii 7998 allele, OhioMR13 = Homozygous for the Ohio MR13 allele; H= heterozygote.

Table 4.5: Single factor analysis of association with F2 population height of the scion at week 3 post-grafting.

101

Treatmentsz Week Survival Meany P-value H7998 (autograft) 1 1.1 bx H7998xNC84173 1 1.4 ab *** NC84173 (autograft) 1 1.7 a H7998 (autograft) 2 0.3 c H7998xNC84173 2 1.1 b *** NC84173(autograft) 2 1.6 a H7998(autograft) 3 0.3 c H7998xNC84173 3 0.8 b *** NC84173(autograft) 3 1.5 a ***, significant at < 0.001 level xData were recorded as (Alive = 2; Stunted =1; Dead = 0) yValues with different letters were significantly different ( P < 0.05) zThree grafting combinations were conducted H7998 self-grafted, H7998 grafted with NC84173, and NC74173 self- grafted.

Table 4.6: Analysis of variance for survival mean along three weeks-post-grafting in three different grafting combinations.

102

FIGURES

Figure 4.1: Phenotypic data distribution for grafting success at week-3 post grafting for BC3S4 and F2 Population.

A) BC3S4 population, 172 families Hawaii7998 × OH89119 (BLUPs-value). Population mean = -0.9111, B) F2 population 288 individual plants Hawaii7998 × Ohio MR13 (0 = dead, 1 =stunted, 2 = alive) Population mean =1.16, C) F2 population 288 individual plants Hawaii7998 × Ohio MR13 scion height (cm). Population mean =6.117 cm. 103

Chapter 5 : Conclusions

These studies were conducted to expand our understating of grafting technology and to identify key selection points in a breeding program. Improvements on grafting include increased seed quality when using wild species for tomato rootstocks and increased grafting success using a variety of rootstocks. Grafting offers multiple benefits including disease resistance and tolerance to abiotic stress by using genetic diversity.

Rootstock with multiple resistance and tolerance to biotic and abiotic stresses are required in order justify the extra cost added in the production. At the same time it is important to obtain high rootstock seed quality. Screening multiple inbred lines crossed with multiple wild relatives can help to achieve these goals. The production of seeds is a complex interaction of genetics and environmental factors.

The grafting process itself also has genetic and environmental components affecting its efficiency, which are measured as grafting success/failure. Increasing its efficiency by implementing new methods could make grafting available to more farmers.

Another approach is to create a rootstock that is highly compatible with multiple varieties

(scion). Finding the genetic basis of grafting failure may allow for breeding rootstock with multiple resistance genes and high success rates.

104

The first objective of this study was to optimize screening and early selection of new potential crosses that produce high seed yield and quality. Each cross was evaluated in terms of seed production (fruit set and seed yield) and seed quality (seed size and germination). The greatest source of variation for fruit set was found among environments (experiments) which accounted for 22% of the total phenotypic variation.

In contrast, for seed yield the genetic variation and environment was similar 28% and

27%, respectively. The source of variation for seed weight was mainly genetic components, which accounted for 84% of the phenotypic variation. Finally, seed germination was influenced by genetic components 65% and environment 13% of the phenotypic variation. We thus suggest to increase seed production environment must be optimize, while increase seed quality genetic selection must be performed before the crosses are conducted.

In addition, the effect of fruit maturation on seed quality was determined from eight different crosses harvested at three different times (Breaker, full ripe and overripe).

The seeds harvested from the overripe fruits resulted to have high final seed germination and the speed was higher. However, we identified a female parent FG02-188 that produces high seed quality regardless of the male parent used and harvesting time. Fruit maturation can be used as a tool to increase seed quality in hybrid seed production, also can be used to select some hybrids that have high seed germination regardless of fruit harvesting.

Early selection is important in order to save resource and reduce cost in production system. Seed size was used as a predictor for seed germination with the

105

objective of to identify a selection point based on this trait. The selection point was based on the minimum national and international standard seed germination 75% (AASCO,

2013) and 85% (ISF, 2013). We identified three selection points with different percentage of discharging bad seed quality and retaining high seed quality for seed germination test.

Selection at 1.4 mg seed weight, 100% of hybrids with high seed quality can be saved the same time 33% of the hybrids with low seed quality can be omitted. The selection point can be increase to 1.8 mg where 74% of the hybrid with high seed quality can be retained and omitted 43% of hybrids with low seeds quality. Finally, at 2.2 mg it can be retained only 53% of the hybrids high seed quality and omitted 62% of the hybrids with low seed quality. These results can be used to draw thresholds depending on resources available for further seed testing (germination).

The second objective was to increase efficiency of grafting success using adhesives in the grafting process. Nine different rootstocks were used grafted with two different methods tube grafting and using adhesives. The experiments were repeated in different seasons under greenhouse environments. Grafting success was measure a survival plant at week 3 post-grafting. Additionally, one season of field experiment was conducted to evaluate the effect of grafting methods on open field conditions. The analysis of variance (ANOVA) showed that grafting method had a significant effect of grafting success and we identified high environmental effect and rootstock genetics.

Thus, adhesives can be a tool to improve grafting success in tomato or other vegetables.

The third objective was to identify the genetic basis of grafting failure between scion-rootstocks via quantitative trait loci (QTL) analysis. The population consisted of

106

advanced backcross BC3S4 population derived from OH88119 and H7998, and F2 population derived from Ohio MR 13 and H7998. The initial experiment phenotype data was collected as a percentage of survival for 10 plants from each BC3S4 family. In the second F2 population data was collected as dead = 0, stunted = 1, alive=2, and height of scion was recorded at week-3 post grafting. QTL analysis was based on single factor analysis of variance. The initial findings suggest on BC3S4 population one QTL was identified for grafting failure. In the second F2 population, a new QTL was identified on chromosome 2 that is associated with low height scion. Additionally, we determined that

H7998 is intolerant to the grafting process, which would explain the poor grafting success in previous studies. However, it is unknown the cause of poor performing of H7998 to the grafting process, it may be genetically limited to tolerate high humidity, unable to overcome the stress caused by grafting (poor wound response) or other causes.

Combining fruit setting ability, high seed yield, seed size and seed germination, will be possible to increase seed yield and quality for new hybrid tomato seeds for rootstocks. We were able to identify potential rootstocks with high seed quality and high seed production. In general, grafting efficiency can be improved using adhesives as tool to bind the grafted seedling in the process. The results of this study show that grafting failure in tomato is associated with alleles from the H7998 parent. We also identified that grafting rejection happens in H7998 even when is self-grafted. More research is needed to identify a marker for the alleles that causes the grafting failure in H7998 or other causes. Finding markers or genes that causes the grafting rejection would help to

107

accelerate the breeding of highly compatible rootstocks with many scions, which is the most desirable trait for a farmer or grafter producer.

108

References Abdelhafeez, A.T., H. Harssema, and K. Verkerk. 1975. Effect of Air Temperature Soil Temperature and Soil Moisture on Growth and Development of tomato itself and Grafted on its own and Eggplant Rootstock. Abdelmageed, A.H.A. and N. Gruda. 2009. Influence of Grafting on Growth, Development and some Physiological Parameters of Tomatoes under Controlled Heat Stress Conditions. European Journal of Horticultural Science 74(1):16-20. Acquaah, G. 2007. Principles of Plant Genetics and Breeding. Blackwell. Agarwal, M., N. Shrivastava, and H. Padh. 2008. Advances in molecular marker techniques and their applications in plant sciences. Plant Cell Rep. 27(4):617-631. Agele, S. and S. Cohen. 2009. Effect of genotype and graft type on the hydraulic characteristics and water relations of grafted melon. Journal of Plant Interactions 4(1):59-66. Aherne, S.A., M.A. Jiwan, T. Daly, and N.M. O’Brien. 2009. Geographical location has greater impact on carotenoid content and bioaccessibility from tomatoes than variety. Plant foods for human nutrition 64(4):250-256. Akca, E.T., S. Serpil, U. Sezer, E. Ozlem, G. Ayse, C. Canan, B. Hakan, K. Ozgur, O. Banu, and G. Hulya. 2009. Health profiles of methyl bromide applicators in greenhouses in Turkey. Ann. Acad. Med. Singap. 38:707-713. Alam, M.Z., M. Ali, A.M. Akanda, D.A.M. Choudhury, N.M.M. Haque, M.M. Hossain, and K. Ogata. 1994. Grafting technology: An integrated pest management component for eggplant and tomato. Bulletin of the Institute of Tropical Agriculture Kyushu University 17(0):85-91. Albuquerque, D.S., L.F. Gominho, and R.A. Santos. 2006. Histologic evaluation of pulpotomy performed with ethyl-cyanoacrylate and calcium hydroxide. Brazilian Oral Research 20(3):226-230. Al-Chaabi, S., O. Koutifani, M.H. Safeih, A. Sedawi, and J. Asmar. 2009. Management of Root-Knot Nematodes and Corky Root Disease of Pepper Plants by Grafting Technique onto Resistant Rootstocks under Plastic House. Arab Gulf J. Sci. Res. 27(3):178-186. Andrade, J., S.E. Cuevas, C.L. Maniscalco, S.A. Stefanes, J. Junqueira, and P.P. Santos. 2001. Uso do etil-cianoacrilato na síntese da parede vascular em cães. Ars veterinaria (Brasil).V 17 (3):172-176. Andrews, P.K. and C.S. Marquez. 1993. Graft incompatibility. Horticultural reviews 15:183-232. Anwar, S.A. and M.V. McKenry. 2008. Induction of Resistance and Susceptibility in Tomato by Two Meloidogyne incognita Populations. J. Nematol. 40(4):276-280.

109

Arao, T., H. Takeda, and E. Nishihara. 2008. Reduction of cadmium translocation from roots to shoots in eggplant (Solanum melongena) by grafting onto Solanum torvum rootstock. Soil Sci. Plant Nutr. 54(54):555-559. Arroyo, A. and I.W. Selman. 1977. Effects of Rootstock and Scion on Tobacco Mosaic - Virus Infection in Susceptible, Tolerant and Immune Cultivars of Tomato. Ann. Appl. Biol. 85(2):249-255. Asins, M.J., M.C. Bolarin, F. Perez-Alfocea, M.T. Estan, C. Martinez-Andujar, A. Albacete, I. Villalta, G.P. Bernet, I.C. Dodd, and E.A. Carbonell. 2010. Genetic analysis of physiological components of salt tolerance conferred by Solanum rootstocks. What is the rootstock doing for the scion? Theor. Appl. Genet. 121(1):105-115. Atanassova, B. and H. Georgiev. 2007. Expression of Heterosis by Hybridization. Genetic improvement of Solanaceous 2:113-151. Backhouse, D., M. Kawaguchi, M. Oda, and A. Taji. 2008. Anatomy and physiology of graft incompatibility in solanaceous plants. Journal of Horticultural Science and Biotechnology 83(5):581-588. Bai, Y. and P. Lindhout. 2007. Domestication and Breeding of Tomatoes: What have we gained and what can we gain in the future? Annals of Botany 5:1085-1094. Baird, N.A., P.D. Etter, T.S. Atwood, M.C. Currey, A.L. Shiver, Z.A. Lewis, E.U. Selker, W.A. Cresko, and E.A. Johnson. 2008. Rapid SNP discovery and genetic mapping using sequenced RAD markers. PloS one 3(10):e3376. Balliu, A., G. Vuksani, S. Kaciu, T. Nasto, and L. Haxhinasto. 2008. Grafting Effects on Tomato Growth Rate, Yield and Fruit Quality under Saline Irrigation Water. p. 1161-1166. In: S.M.G.S.M.A.S.E. DePascale (ed). Proceedings of the International Symposium on High Technology for Greenhouse System Management, Vols 1 and 2. Bang, H., S. Hwang, H. Ham, and J. Lee. 2000. Seed germination and seedling characteristics of cucurbits as affected by dry heat treatment and subsequent handling. HortScience 35(3):432. Barrett, C.E., X. Zhao, and A.W. Hodges. 2012. Cost Benefit Analysis of Using Grafted Transplants for Root-knot Nematode Management in Organic Heirloom Tomato Production. Horttechnology 22(2):252-257. Barry, K.H., S. Koutros, J.H. Lubin, J.B. Coble, F. Barone-Adesi, L.E. Beane Freeman, D.P. Sandler, J.A. Hoppin, X. Ma, and T. Zheng. 2012. Methyl bromide exposure and cancer risk in the Agricultural Health Study. Cancer Causes and Control :1- 12. Batchelor, T. 2004. The impact of the Montreal Protocol and European Union controls on methyl bromide. Bateman, A.J. 1955. Grafting experiments between the tomato varieties, Golden Apple and Oxheart. Nature 175(4469):1118-1120. Bausher, M. 2008. Study of quality and growth characteristics of tomato variety camelia grafted onto commercial rootstocks. HortScience 43(4):1257-1257.

110

Bausher, M.G. 2009. Commercial Tomato Rootstock Performance When Exposed to Natural Populations of Root-Knot Nematodes in Florida. HortScience 44(4):1021- 1021. Bausher, M.G. and D.O. Chellemi. 2010. Effect of Rootknot Nematode on Tomato Rootstocks in Non-fumigated Open Field Production. HortScience 45(8):S112- S112. Bausher, M.G., E. Rosskopf, and N. Burelle. 2011. Grafting Technique to Eliminate Rootstock Suckering of Grafted Tomatoes (vol 46, pg 596, 2011). HortScience 46(11):1568-1568. Beckman, T.G. and G.A. Lang. 2003. Rootstock breeding for stone fruits. p. 531-551. In: J. Janick (ed). Genetics and Breeding of Tree Fruits and Nuts. Bello, A., J. Lipez-Perez, and A. Garcea-Elvares. 2002. Biofumigation as an alternative to methyl bromide. Bernstein, L. 1976. Physiological basis of salt tolerance in plants. Basic Life Sci. 8:283- 290. Besri, M. 2008. Cucurbits grafting as alternative to Methyl Bromide for cucurbits production in Morocco. Black, L.L., D.L. Wu, J.F. Wang, T. Kalb, D. Abbass, and J.H. Chen. 2003. Grafting tomatoes for production in the hot-wet season. Asian Vegetable Research & Development Center.AVRDC Publication (03-551):6. Bletsos, F.A. 2005. Use of grafting and calcium cyanamide as alternatives to methyl bromide soil fumigation and their effects on growth, yield, quality and Fusarium wilt control in melon. J. Phytopathol. 153(3):155-161. Bloom, A.J., M.A. Zwieniecki, J.B. Passioura, L.B. Randall, N.M. Holbrook, and D.A. St. Clair. 2004. Water relations under root chilling in a sensitive and tolerant tomato species. Plant, Cell and Environment 27:971-979. Bomblies, K. and D. Weigel. 2007. Hybrid necrosis: autoimmunity as a potential gene- flow barrier in plant species. Nature Reviews Genetics 8(5):382-393. Boyer, J.S. 1982. Plant Productivity and Environment. Science 218(4571):443-448. Bozkurt, M.K. and L. Saydam. 2008. The use of cyanoacrylates for wound closure in head and neck surgery. European Archives of Oto-Rhino-Laryngology 265(3):331-335. Bradley, J. 1968. Tomato Grafting to Control Root Diseases. New Zealand Journal of Agriculture 16:26-27. Branthôme, F. 2010. Trends in tomato products consumption compared to total tomato consumption. The world Processing Tomato Council (WPTC). Braun, A.L. and D.M. Supkoff. 1994. Options to methyl bromide for the control of soil- borne diseases and pests in California with references to the Netherlands. Environmental Protection Agency, . Bravenboer L. Pet G. 1962. Control of soil-borne diseases in tomatoes by grafting on resistant rootstock, Brussels, Belgium, August 31 - September 8. Brown, P., T. Lumpkin, S. Barber, E. Hardie, K. Kraft, K. Luedeling, T. Rosenstock, K. Tabaj, D. Clay, G. Luther, P. Marcotte, R. Paul, S. Weller, F. Youssefi, and M. Demment. 2005. Global Horticulture Assessment. Gent-Oostakker, Belgium.

111

Burr, B., S.V. Evola, F.A. Burr, and J.S. Beckmann. 1983. The application of restriction fragment length polymorphism to plant breeding. p. 45-59. In: Anonymous Genetic engineering. Springer. Buttaro, D., F. Serio, and P. Santamaria. 2012. Soilless greenhouse production of table grape under Mediterranean conditions. Journal of Food Agriculture & Environment 10(2):641-645. Calvert, G.M., C.A. Mueller, J.M. Fajen, D.W. Chrislip, J. Russo, T. Briggle, L.E. Fleming, A.J. Suruda, and K. Steenland. 1998. Health effects associated with sulfuryl fluoride and methyl bromide exposure among structural fumigation workers. Am. J. Public Health 88(12):1774-1780. Canene-Adams, K., J.K. Campbell, S. Zaripheh, E.H. Jeffery, and J.W. Erdman. 2005. The tomato as a functional food. J. Nutr. 135(5):1226-1230. Cansev, A. and M. Ozgur. 2010. Grafting cucumber seedlings on Cucurbita spp.: Comparison of different grafting methods, scions and their performance. Journal of Food Agriculture & Environment 8(3-4):804-809. Cantliffe, D.J., N. Quang Vinh, X. Zhao, N.L. Shaw, and S.A. Sargent. 2009. Evaluating Rootstocks for Greenhouse Tomato Production. HortScience 44(4):1057-1057. Cao, Z., D. Dong, X. Wang, and Y. Chen. A method of cultivating tomato resisting root- knot nematodes. Univ China Agric, . Cardoso, S.C., A.C. Fermino Soares, A.D. Santos Brito, L.A. De Carvalho, and C.A. Da Silva Ledo. 2006. Potential of Hawaii 7996 hybrid as rootstock for tomato cultivars. Bragantia 65(1):89-96. Carmen Martinez-Ballesta, M., C. Alcaraz-Lopez, B. Muries, C. Mota-Cadenas, and M. Carvajal. 2010. Physiological aspects of rootstock-scion interactions. Scientia Horticulturae 127(2):112-118. Cartia, G. 1998. Solarization in integrated management systems forgreenhouses. FAO, Rome. Chatzissavvidis, C., I. Papadakis, and I. Therios. 2008. Effect of calcium on the ion status and growth performance of a citrus rootstock grown under NaCl stress. Soil Sci. Plant Nutr. 54(6):910-915. Chen, G.X., X.P. Fu, S.H. Lips, and M. Sagi. 2003. Control of plant growth resides in the shoot, and not in the root, in reciprocal grafts of flacca and wild-type tomato (Lysopersicon esculentum), in the presence and absence of salinity stress. Plant Soil 256(1):205-215. Chen, G.X., S.H. Lips, and M. Sagi. 2002. Biomass production, transpiration rate and endogenous abscisic acid levels in grafts of flacca and wild-type tomato (Lycopersicon esculentum). Functional Plant Biology 29(11):1329-1335. Chen, S., Y.C. Chiu, and Y.C. Chang. 2010. Development of a tubing-grafting robotic system for fruit-bearing vegetable seedlings. Appl. Eng. Agric. 26(4):707-714. Cheng, Y.W. and S.E. Chua. 1976. The possible use of grafted tomato brinjal plants for tomato production in lowland tropics. Singapore Journal of Primary Industries 4(2):94-102. Chia, P. and C. Kubota. 2009. Efficacy of End-of-day Far-red Light in Controlling Tomato Rootstock Height. HortScience 44(4):1051-1052.

112

Chiu, Y.C., M.Y. Chang, G.J. Wu, and C.C. Chen. 2008. Development of an automatic outward-feature properties measurement system for grafted tomato seedlings. Appl. Eng. Agric. 24(1):101-113. Choi, J.M., C.S. Kang, and J.W. Ahn. 2008. Effect of root substrate formulations on the growth of grafted tomato plug seedlings. p. 351-357. In: D.I. Leskovar (ed). Proceedings of the Ivth International Symposium on Seed, Transplant and Stand Establishment of Horticultural Crops: Translating Seed and Seedling Physiology into Technology. Choi, Y. and H. Chung. 2003. Interspecific differences in H2O2 production and antioxidant activity in chilling treated tomato (Lycopersicon spp.) plants. Journal of the Korean Society for Horticultural Science 44(5):644-648. Chung, H. and Y. Choi. 2002. Growth responses on varying soil EC and selection of salt- tolerant rootstock of tomato (Lycopersicon spp.). Journal of the Korean Society for Horticultural Science 43(5):536-544. Chung, H., S. Youn, and Y. Choi. 1997. Effects of rootstocks on yield, quality and components of tomato fruits. Journal of the Korean Society for Horticultural Science 38(6):603-607. Colla, G., M. Cardarelli, A. Fiorillo, Y. Rouphael, A. Salerno, and E. Rea. 2012. Can Grafting in Watermelon Plants Enhance Tolerance to Bicarbonate in Nutrient Solution?. p. 323-329. In: Castilla,N.VanKooten O.Sase S.Meneses J.F.Schnitzler W.H.VanOs (ed). Xxviii International Horticultural Congress on Science and Horticulture for People. Colla, G., Y. Roupahel, M. Cardarelli, and E. Rea. 2006. Effect of salinity on yield, fruit quality, leaf gas exchange, and mineral composition of grafted watermelon plants. HortScience 41(3):622-627. Colla, G., Y. Rouphael, M. Cardarelli, D. Massa, A. Salerno, and E. Rea. 2006. Yield, fruit quality and mineral composition of grafted melon plants grown under saline conditions. Journal of Horticultural Science & Biotechnology 81(1):146-152. Colla, G., Y. Rouphael, C. Leonardi, and Z. Bie. 2010. Role of grafting in vegetable crops grown under saline conditions. Scientia Horticulturae 127(2):147-155. Copes, D.L. 1971. Seed Source and Graft Compatibility in Douglas - Fir. For. Sci. 17(4):499-&. Copes, D.L. 1973. Inheritance of Graft Compatibility in Douglas - Fir. Botanical Gazette 134(1):49-52. Copes, D.L. and M.D. Oliver. 1970. Relationship Between Douglas - Fir Graft compatibility and Wood Specific Gravity. J. For. 68(11):726-&. Cortada, L., F.J. Sorribas, C. Ornat, I. Kaloshian, and S. Verdejo-Lucas. 2008. Variability in infection and reproduction of Meloidogyne javanica on tomato rootstocks with the Mi resistance gene. Plant Pathol. 57(6):1125-1135. Cortada, L., F.J. Sorribas, C. Ornat, I. Kaloshian, and S. Verdejo-Lucas. 2009. Response of Tomato Rootstock Carrying the Mi-resistance gene to populations of Meloidogyne arenaria, M. incognita and M. javanica. Eur.J.Plant Pathology 124:337-343.

113

Cramer, G.R., K. Urano, S. Delrot, M. Pezzotti, and K. Shinozaki. 2011. Effect of abiotic stress on plants: a systems biology perspective. BMC Plant Biology 11(163):14. Criddle, R.S., B.N. Smith, and L.D. Hanse. 1997. A respiration based description of plant growth rate response to temperature. Planta 201:441-445. Cuartero, J., M.C. Bolarin, M.J. Asins, and V. Moreno. 2006. Increasing salt tolerance in the tomato. J. Exp. Bot. 57(5):1045-1058. Darwin, C. 1876. The effects of cross and self-fertilization in the vegetable kingdom. J. Murray. Dasgan, H.Y., V. Romheld, I. Cakmak, and K. Abak. 2002. Physiological root responses of iron deficiency susceptible and tolerant tomato genotypes and their reciprocal F-1 hybrids. Plant Soil 241(1):97-104. Date, H., K. Ozaki, T. Shirakawa, H. Nasu, M. Hatamoto, and Y. Okamoto. 1992. Bacterial wilt of cucumber plant grafted on pumpkin rootstock caused by Pseudomonas solanacearum. Ann. Phytopathol. Soc. Japan 58(2):220-227. Davis, A.R., P. Perkins-Veazie, R. Hassell, A. Levi, S.R. King, and X. Zhang. 2008. Grafting effects on vegetable quality. HortScience 43(6):1670-1672. Davis, A.R., P. Perkins-Veazie, Y. Sakata, S. Lopez-Galarza, J.V. Maroto, S. Lee, Y. Huh, Z. Sun, A. Miguel, S.R. King, R. Cohen, and J. Lee. 2008. Cucurbit grafting. Crit. Rev. Plant Sci. 27(1):50-74. DeVay, J.E., J.J. Stapleton, and C.L. Elmore. 1991. Soil solarization : proceedings of the First International Conference on Soil Solarization, Amman, Jordan, 19-25 February 1990. Food and Agriculture Organization of the United Nations, Rome. Di Gioia, F., F. Serio, D. Buttaro, O. Ayala, and P. Santamaria. 2010. Influence of rootstock on vegetative growth, fruit yield and quality in 'Cuore di Bue', an heirloom tomato. Journal of Horticultural Science & Biotechnology 85(6):477- 482. Djian-Caporalino, C., S. Molinari, A. Palloix, A. Ciancio, A. Fazari, N. Marteu, N. Ris, and P. Castagnone-Sereno. 2011. The reproductive potential of the root-knot nematode Meloidogyne incognita is affected by selection for virulence against major resistance genes from tomato and pepper. Eur. J. Plant Pathol. 131(3):431- 440. Djidonou, D. and X. Zhao. 2010. Rootstock Effects on Tomato Yield and Nutrient Uptake under Greenhouse Conditions. HortScience 45(8):S130-S130. East, E.M. 1936. Heterosis. Genetics 21(4):375. East, E.M. and H.K. Hayes. 1912. Heterozygosis in evolution and in plant breeding. Govt. Printing Off. Edelstein, M., M. Ben-Hur, and Z. Plaut. 2008. Response of grafted and non-grafted melon plants to effluent and boron under greenhouse conditions. p. 325-328. In: D.I. Leskovar (ed). Proceedings of the Ivth International Symposium on Seed, Transplant and Stand Establishment of Horticultural Crops: Translating Seed and Seedling Physiology into Technology. Edmond Ghanem, M., I. Hichri, A.C. Smigocki, A. Albacete, M. Fauconnier, E. Diatloff, C. Martinez-Andujar, S. Lutts, I.C. Dodd, and F. Perez-Alfocea. 2011. Root-

114

targeted biotechnology to mediate hormonal signalling and improve crop stress tolerance. Plant Cell Rep. 30(5):807-823. Endo, M.S., J.V. Costa, M.R.M. Natali, and A.F. Queiroz. 2007. Efeito in vivo do etil- cianoacrilato como isolamento absoluto em gengiva inserida. Revista de Odontologia da UNESP 36(3):287-292. Estañ, M.T., M.M. Martinez-Rodriguez, F. Perez-Alfocea, T.J. Flowers, and M.C. Bolarin. 2005. Grafting raises the salt tolerance of tomato through limiting the transport of sodium and chloride to the shoot. J. Exp. Bot. 56(412):703-712. Estan, M.T., I. Villalta, M.C. Bolarin, E.A. Carbonell, and M.J. Asins. 2009. Identification of fruit yield loci controlling the salt tolerance conferred by solanum rootstocks. Theor. Appl. Genet. 118(2):305-312. Etehadnia, M., D. Waterer, H. De Jong, and K.K. Tanino. 2008. Scion and rootstock effects on ABA-mediated plant growth regulation and salt tolerance of acclimated and unacclimated potato genotypes. J. Plant Growth Regul. 27(2):125-140. Fan, M., Z. Bie, A. Krumbein, and D. Schwarz. 2011. Salinity stress in tomatoes can be alleviated by grafting and potassium depending on the rootstock and K- concentration employed. Scientia Horticulturae 130(3):615-623. FAOSTAT, F.A.d. 2009. Food and Agriculture Organization of the United Nations. Rome, Italy. FAOSTAT, F.A.d. 2013. Food and Agriculture Organization of the United Nations. Rome, Italy. FAOSTAT, F.A.d. 2006. Food and Agriculture Organization of the United Nations. Rome, Italy. Felix, S. 1973. A method for combating bacterial wilt and tomato nematodes. Revue Agricole et Sucriere de l'Ile Maurice 52(1):12-14. Fernández-García, N., A. Cerdá, and M. Carvajal. 2003. Grafting, a useful technique for improving salinity tolerance of tomato?. Fernandez-Garcia, N., M. Carvajal, and E. Olmos. 2004. Graft union formation in tomato plants: Peroxidase and catalase involvement. Annals of Botany 93(1):53-60. Fernandez-Garcia, N., V. Martinez, A. Cerda, and M. Carvajal. 2002. Water and nutrient uptake of grafted tomato plants grown under saline conditions. J. Plant Physiol. 159(8):899-905. Fernandez-Garcia, N., V. Martinez, A. Cerda, and M. Carvajal. 2004. Fruit quality of grafted tomato plants grown under saline conditions. Journal of Horticultural Science & Biotechnology 79(6):995-1001. Fisher, R.A. 1919. The Correlation between Relatives on the Supposition of Mendelian Inheritance. Trans. R. Soc. Edinb. 52(02):399-433. Fisher, R.A. 1930. The evolution of dominance in certain polymorphic species. Am. Nat. :385-406. Flaishman, M.A., K. Loginovsky, S. Golobowich, and S. Lev-Yadun. 2008. Arabidopsis thaliana as a model system for graft union development in homografts and heterografts. J. Plant Growth Regul. 27(3):231-239. Flores, F.B., P. Sanchez-Bel, M.T. Estan, M.M. Martinez-Rodriguez, E. Moyano, B. Morales, J.F. Campos, J.O. Garcia-Abellan, M.I. Egea, N. Fernandez-Garcia, F.

115

Romojaro, and M.C. Bolarin. 2010. The effectiveness of grafting to improve tomato fruit quality. Scientia Horticulturae 125(3):211-217. Francis, D., A. Clivati-McIntyre, and M. Kleinhenz. 2007. Grafting tomato improves yield potential but presents challenges in rootstock-scion compatibility and retention of desirable scion traits. HortScience 42(4):851-851. Freeman, J.H., T. McAvoy, S. Rideout, and A. Wimer. 2010. Utilization of Grafted Tomato Seedlings for Open Field Production. HortScience 45(4):499-499. Fujiwara, K., S. Isobe, and M. Iimoto. 2001. Optimum conditions of low light irradiation- CA storage for quality preservation of grafted tomato plug seedlings. Environment Control in Biology 39(2):111-120. Fujiwara, K., Y. Kimura, and K. Kurata. 2003. Effect on the quality of grafted tomato plug seedlings of blue-light PPFD percentage during red and blue LEDs low light irradiation storage. Environment Control in Biology 41(4):361-368. Gajc-Wolska, J., J. Radzanowska, and M. Lyszkowska. 2009. The influence of grafting and biostimulators on physical and sensorial traits of greenhouse tomato fruit (Lycopersicon esculentum Mill.) in field production. Acta Scientiarum Polonorum-Hortorum Cultus 8(3):37-43. Gamliel, A. and J.J. Stapleton. 1997. Improvement of soil solarization with volatile compounds generated from organic amendments. Phytoparasitica 25(1):S31-S38. Garibaldi, A., M. Baudino, A. Minuto, and M.L. Gullino. 2008. Effectiveness of Fumigants and Grafting against Tomato Brown Root Rot Caused by Colletotrichum coccodes. Phytoparasitica 36(5):483-488. Garner, R.J. 1947. The grafter's handbook. The grafter's handbook. Gaytan-Mascorro, A., J. Zaragoza Castellanos-Ramos, S. Villalobos-Reyes, J.C. Diaz- Perez, and F. Camacho-Ferre. 2008. Response of grafted tomato plants (Lycopersicon esculentum Mill.) to leaf pruning and nutrient solution concentration. Journal of Food Agriculture & Environment 6(3-4):269-277. Geboloğlu, N., E. Yılma , P. Çakmak, M. Aydın, and Y. Kasap. 2011. Determining of the yield, quality and nutrient content of tomatoes grafted on different rootstocks in soilless culture. Sci.Res.Essays 6(10):2147-2153. Giannakou, I.O. and D.G. Karpouzas. 2003. Evaluation of chemical and integrated strategies as alternatives to methyl bromide for the control of root-knot nematodes in Greece. Pest Manag. Sci. 59(8):883-892. Gilreath, J.P., J.P. Jones, B.M. Santos, and A.J. Overman. 2004. Soil fumigant evaluations for soil-borne pest and Cyperus rotundus control in fresh market tomato. Crop Protection 23:889-893. Gisbert, C., J. Prohens, and F. Nuez. 2011. Performance of eggplant grafted onto cultivated, wild, and hybrid materials of eggplant and tomato. International Journal of Plant Production 5(4):367-380. Goto, R., L.F. Sirtori, J.D. Rodrigues, and M.C. Lopes. 2010. Production of tomato, momotaro hibrid according to seedling stage and grafting. Ciencia E Agrotecnologia 34(4):961-966. Graham, C.W. 1966. Potato root eelworm and tomato rootstocks. Plant Pathol. 15(2):76- 82.

116

Grandillo, S., D. Bernacchi, T.M. Fulton, D. Zamir, and S.D. Tanksley. 1999. Advanced backcross QTL analysis: a method for the systematic use of exotic germplasm in the improvement of crop quality. p. 283-290. In: Anonymous Genetics and Breeding for Crop Quality and Resistance. Springer. Grandillo, S., H.M. Ku, and S.D. Tanksley. 1999. Identifying the loci responsible for natural variation in fruit size and shape in tomato. Theor. Appl. Genet. 99(6):978- 987. Grigoriadis, I., I. Nianiou-Obeidat, and A.S. Tsaftaris. 2005. Shoot regeneration and micrografting of micropropagated hybrid tomatoes. Journal of Horticultural Science & Biotechnology 80(2):183-186. Grimault, V. and P. Prior. 1994. Grafting Tomato Cultivars Resistant or Susceptible to Bacterial Wilt - Analysis of Resistance Mechanisms. Journal of Phytopathology- Phytopathologische Zeitschrift 141(3):330-334. Grisdale, J. 1998. The use of cyanoacrylates in periodontal therapy. Journal-Canadian Dental Association 64:632-635. Grun, P. and M. Aubertin. 1966. The inheritance and expression of unilateral incompatibility in Solanum. Heredity 21:131-138. Guan, W., X. Zhao, R. Hassell, and J. Thies. 2012. Defense Mechanisms Involved in Disease Resistance of Grafted Vegetables. HortScience 47(2):164-170. Hadley, H.H. and S.J. Openshaw. 1980. Interspecific and intergeneric hybridization. Hybridization of crop plants (hybridization):133-159. Hafeez, A.T.A. and P. Cornillon. 1976. Studies of the Effect of Grafting on Tomato Quality. Sudan Journal of Food Science and Technology 8:80-84. Hall, T.J. 1980. Resistance at the Tm-2 locus in the tomato to tomato mosaic virus. Euphytica 29(1):189-197. Hamilton, J.P. and C. Robin Buell. 2012. Advances in plant genome sequencing. The Plant Journal 70(1):177-190. Hanna, H.Y. 2012. Producing a Grafted and a Non-Grafted Tomato Plant from the Same Seedling. Horttechnology 22(1):72-76. Hansen, L.D., M. Afzal, R.W. Breidenbach, and R.S. Criddle. 1994. High and low temperature limits to growth of tomato cells. Planta 195:1-9. Harrison, A.L., P.A. Young, G.E. Alstatt, and T.A.E. Station. 1939. Control of tomato diseases in the seed bed and cold frame. Texas Agricultural Experiment Station, College Station, Tex. Harrison, D.J. and P.G. Burges. 1962. Use of rootstock resistance for controlling Fusarium wilt of tomatoes. Plant Pathol. 11:23-25. Hartmann, H.T., D.E. Kester, and F.T. Davies. 1990. Theoretical aspects of grafting and budding. Plant propagation 5:305-348. Hartmann, H.T., D.E. Kester, F.T. Davies Jr, and R.L. Geneve. Plant Propagation— Principles and Practices. 1997. Prentice Hall). Englewood Cliffs, New Jersey, . Hasna, M.K., E. Ogren, P. Persson, A. Martensson, and B. Ramert. 2009. Management of corky root disease of tomato in participation with organic tomato growers. Crop Protection 28(2):155-161.

117

He, Y., Z.J. Zhu, J. Yang, X.L. Ni, and B. Zhu. 2009. Grafting increases the salt tolerance of tomato by improvement of photosynthesis and enhancement of antioxidant enzymes activity. Environ. Exp. Bot. 66(2):270-278. Herrero, J. 1951. Studies of compatible and incompatible graft combinations with special reference to hardy fruit trees. J.hort.Sci 26:186-237. Hikichi, Y., Y. Nakazawa-Nasu, S. Kitanosono, K. Suzuki, and T. Okuno. 1999. The behavior of lux-marked Ralstonia solanacearum in grafted tomato cultivars resistant or susceptible to bacterial wilt. Ann. Phytopathol. Soc. Japan 65(6):597- 603. Hoyos Echevarria, P., G. Rollon Martinez, and B. Galvez Rodriguez. 2012. Influence of Grafting on the Yield and Quality of Tomato Cultivars Grown in Greenhouse in Central Spain. p. 449-454. In: Castilla,N.VanKooten O.Sase S.Meneses J.F.Schnitzler W.H.VanOs (ed). Xxviii International Horticultural Congress on Science and Horticulture for People. Huang, Y., Z.L. Bie, Z.X. Liu, A. Zhen, and X.R. Jiao. 2011. Improving cucumber photosynthetic capacity under NaCl stress by grafting onto two salt-tolerant pumpkin rootstocks. Biol. Plant. 55(2):285-290. Huang, Y., Z. Bie, S. He, B. Hua, A. Zhen, and Z. Liu. 2010. Improving cucumber tolerance to major nutrients induced salinity by grafting onto Cucurbita ficifolia. Environ. Exp. Bot. 69(1):32-38. Huang, Y., R. Tang, Q. Cao, and Z. Bie. 2009. Improving the fruit yield and quality of cucumber by grafting onto the salt tolerant rootstock under NaCl stress. Scientia Horticulturae 122(1):26-31. Huh, Y., D. Kim, S. Lee, K. Park, D. Park, Y. Woo, and J. Lee. 2005. Growth and development of watermelon plants grafted onto Citrullus rootstocks selected for disease resistance at two temperature regimes. HortScience 40(4):1053-1054. Huh, Y., Y. Woo, D. Kim, J. Lee, and Y. Om. 2003. Salt tolerance of watermelon grafted onto Citrullus rootstocks selected for disease resistance. Journal of the Korean Society for Horticultural Science 44(5):655-660. Hwang, S., H. Bang, J. Hong, and J. Lee. 2001. Seed germination and seedling vigor of gourd (Lagenaria siceraria) as affected by dry-heat treatments. HortScience 36(3):587. Ibrahim, M., M.K. Munira, M.S. Kabir, A.K.M.S. Islam, and M.M.U. Miah. 2001. Seed Germination and Graft Compatibility of Wild Solanum as Rootstock of Tomato. Online Journal of Biological Sciences 1(8):701-703. Ioannou, N. 2001. Integrating soil solarization with grafting on resistant rootstocks for management of soil-borne pathogens of eggplant. Journal of Horticultural Science & Biotechnology 76(4):396-401. Iott, M., R.M. Welker, C.L. Rivard, and F.J. Louws. 2011. Potential role of grafting as a method to manage Verticillium dahliae race 2 in tomato production systems. Phytopathology 101(6):S79-S79. Ito, H. and S. Kawai. 1991. Cultivation of cherry tomato using rooted cuttings and grafts from lateral shoots. Research Bulletin of the Aichi-ken Agricultural Research Center 0(23):177-184.

118

Johnson, S.J. and C.A. Miles. 2011. Effect of Healing Chamber Design on the Survival of Grafted Eggplant, Tomato, and Watermelon. Horttechnology 21(6):752-758. Jones, O.P. 1985. Endogenous growth regulators and rootstock/scion interactions in apple and cherry trees. Jones, J.P., J.P. Gilreath, and A.J. Overman. 1995. Control of soil-borne diseases of mulched tomato by fumigation. Proc.Fla.State.Hort.Soc. 108:201-203. Jung, K., B. Kang, Y. Park, and J. Lee. 2004. Producing Double-stemmed Tomato Seedlings by Cutting Propagation, Pinching, and Plant Bioregulators Application. HortScience 39(4):758-+. Jung, K., Y. Park, and J. Lee. 2005. Production of double-stemmed grafted tomato plug seedlings. HortScience 40(4):1056. Kacjan Marsic, N. and J. Osvald. 2004. The influence of grafting on yield of two tomato cultivars (Lycopersicon esculentum Mill) grown in plastic house. Acta Agriculture Slovenica 83:243-249. Kakinuma, M., K. Masuda, and H. Kinoshita. Method to graft fruit vegetables to stock plants. Takii Shubyo Kk; Daiichi Ceramo Ltd, . Kamer, F.M. and J.H. Joseph. 1989. Histoacryl: its use in aesthetic facial plastic surgery. Archives of Otolaryngology—Head & Neck Surgery 115(2):193. Kang, X. and I. Miyajima. 1997. Fruit quality of grafted tomato plants under sand culture. Science Bulletin of the Faculty of Agriculture Kyushu University 52(1-2):1-4. Kang, X. and I. Miyajima. 1997. Growth and yield of grafted tomato plants under sand culture. Science Bulletin of the Faculty of Agriculture Kyushu University 52(1- 2):5-9. Kaskavalci, G., Y. Tuzel, O. Dura, and G.B. Oztekin. 2009. Effects of Alternative Control Methods Against Meloidogyne incognita in Organic Tomato Production. Ekoloji 18(72):23-31. Katan, J. 1981. Solar heating (Solarization) control of soil-borne pests. Ann.Rev.Phytopathology 19:211-236. Katan, J. and J.E. DeVay. 1991. Soil solarization. CRC, Boca Raton. Katan, J., A. Greenberger, H. Alon, and A. Grinstein. 1976. Solar heating by polyethylene mulching for the control of diseases caused by soil-borne pathogens. Phytopathology 66:683-688. Katoh, N., M. Yui, S. Sato, T. Shirai, H. Yuasa, and M. Hagimori. 2004. Production of virus-free plants from virus-infected sweet pepper by in vitro grafting. Scientia Horticulturae 100(1-4):1-6. Kaya, C., A.L. Tuna, M. Guneri, and M. Ashraf. 2011. Mitigation Effects of Silicon on Tomato Plants Bearing Fruit Grown at High Boron Levels. J. Plant Nutr. 34(13):1985-1994. Keyworth, W.G. 1964. Hypersensitivity of Monogenic Resistant Tomato Scions to Toxins Produced in Bonny Best Rootstock Invaded by Fusarium Oxysporum F. Lycopersici. Ann. Appl. Biol. 54(1):99-&. Khah, E.M. 2005. Effect of grafting on growth, performance and yield of aubergine (Solanum melongena L) in the fieldand greenhouse. Journal of Food Agriculture & Environment 3(3-4):92-94.

119

Khah, E.M. 2011. Effect of grafting on growth, performance and yield of aubergine (Solanum melongena L.) in greenhouse and open-field. International Journal of Plant Production 5(4):359-366. Khanna-Chopra, R., M. Maheswari, D.G. Rao, and S.K. Sinha. 1993. Expression of heterosis-a physiological analysis, Ludhiana, 23-24 February. Kim, D. and J. Lee. 2000. Storability of dry-heat-treated gourd seeds. HortScience 35(3):427. Kim, D. and J. Lee. 2001. Effects of seed sterilization treatment on germination and seedling growth of bottle gourd (Lagenaria siceraria). Journal of the Korean Society for Horticultural Science 42(2):131-136. King, S.R., A.R. Davis, W. Liu, and A. Levi. 2008. Grafting for disease resistance. HortScience 43(6):1673-1676. King, S.R., A.R. Davis, X. Zhang, and K. Crosby. 2010. Genetics, breeding and selection of rootstocks for Solanaceae and Cucurbitaceae. Scientia horticulturae 127(2):106-111. Koren, A. and M. Edelstein. 2004. Advantages and Limitations of Grafted Vegetable Transplants in Israel. HortScience 39(4):873-873. Kowalczyk, K. and J. Gajc-Wolska. 2011. Effect of the Kind of Growing Medium and Transplant Grafting on the Cherry Tomato Yielding. Acta Scientiarum Polonorum-Hortorum Cultus 10(1):61-70. Kraevoi, I.M. 1978. Development of New Tomato Forms and Their Use in Practical Breeding. Sel'skokhozyaistvennaya Biologiya 13(3):375-380. Kravchenko, V.A. 1990. Complex hybridization in breeding tomato for earliness. Selektsiya i Semenovodstvo (Moskva) (2):14-16. Kroggel, M. and C. Kubota. 2003. A preliminary study on the effects of EC and grafting on growth, gas exchange and fruit quality of hydroponic tomato (Lycopersicon esculentum). HortScience 38(5):789. Kubota, C. 2008. Use of grafted seedlings for vegetable production in North America. p. 21-28. In: Oh,D.Kubota C.Mitra S.K.Park H.Shigyo M.Shahak Y. (ed). Proceedings of the International Symposium on Cultivation and Utilization of Asian, Sub-Tropical, and Underutilized Horticultural Crops. Kubota, C., M.A. McClure, N. Kokalis-Burelle, M.G. Bausher, and E.N. Rosskopf. 2008. Vegetable grafting: History, use, and current technology status in North America. HortScience 43(6):1664-1669. Kuniyasu, K. and K. Yamakawa. 1983. Control of Fusarium wilt of tomato caused by Fusarium-oxysporum f. sp. lycopersici race J-3 by grafting to KNVF and KNF rootstock of the interspecific hybrids between Lycopersicon esculentum and Lycopersicon hirsutum. Ann. Phytopathol. Soc. Japan 49(5):581-586. Kwon, W. and J. Lee. 1997. Changes of cytokinins and gibberellin contents during low temperature storage of dehisced ginseng seeds. Journal of the Korean Society for Horticultural Science 38(2):111-115. Kwon, W., B. Min, and J. Lee. 1998. Separation of cytokinins in ginseng seeds during after-ripening by HPLC. Journal of the Korean Society for Horticultural Science 39(3):371-375.

120

Lapushner, D. and R. Frankel. 1967. Practical aspects, and the use of male sterility in the production of hybrid tomato seed. Euphytica 16(2):300-310. Laterrot, H. 1989. The tomato. Advantages and use of wild varieties for varietal creation. Rev.Hort 295:13-17. Lee, J., H. Bang, and H. Ham. 1998. Grafting of vegetables. Journal-Japanese Society for Horticultural Science 67:1098-1104. Lee, J., C. Kubota, S.J. Tsao, Z. Bie, P.H. Echevarria, L. Morra, and M. Oda. 2010. Current status of vegetable grafting: diffusion, grafting techniques, automation. Scientia Horticulturae 127(2):93-105. Lee, J. and M. Oda. 2003. Grafting of herbaceous vegetable and ornamental crops. Horticultural Reviews - Westport then New York 28:61-124. Lee, H., S. Lee, C. Chun, and J. Lee. 2006. Growth and salt tolerance of grafted hot pepper seedlings. HortScience 41(4):1072-1073. Lee, H., S. Lee, and J. Lee. 2005. Rooting response of grafted seedlings of hot peppers (Capsicum annuum). HortScience 40(4):1056. Lee, J. and J. Lee. 2006. Promotion of fruit enlargement of 'Whangkeumbae' Asian pear fruits by gibberellins. Horticulture Environment and Biotechnology 47(4):183- 187. Lee, J. 2003. Advances in Vegetable Grafting. Chronica Horticulturae 43:13-19. Lee, J., J. Jung, K. Park, and I. Yu. 1994. Outline of vegetable growing and research in the Republic of Korea. Journal of the Korean Society for Horticultural Science 35(SUPPL.):68-86. Lee, J. and D. Kim. 2000. Moisture balance in dry-heat-treated vegetable seeds. HortScience 35(3):498. Lee, S., J. Choi, K. Kim, J. Chung, and Y. Lee. 1997. Effect of rootstocks and grafting methods on the growth and fruit quality of tomato (Lycoperiscum esculentum MILL). RDA Journal of Horticulture Science 39(2):15-20. Leggat, P.A., U. Kedjarune, and D.R. Smith. 2004. Toxicity of cyanoacrylate adhesives and their occupational impacts for dental staff. Ind. Health 42(2):207-211. Leonardi, C. and F. Giuffrida. 2006. Variation of plant growth and macronutrient uptake in grafted tomatoes and eggplants on three different rootstocks. European Journal of Horticultural Science 71(3):97-101. Li, W., S. Royer, and R.T. Chetelat. 2010. Fine mapping of ui6. 1, a gametophytic factor controlling pollen-side unilateral incompatibility in interspecific solanum hybrids. Genetics 185(3):1069-1080. Lin, C., S. Hsu, K. Tzeng, and J. Wang. 2008. Application of a preliminary screen to select locally adapted resistant rootstock and soil amendment for integrated management of tomato bacterial wilt in Taiwan. Plant Dis. 92(6):909-916. Lippman, Z.B., Y. Semel, and D. Zamir. 2007. An integrated view of quantitative trait variation using tomato interspecific introgression lines. Curr. Opin. Genet. Dev. 17(6):545-552. Lipshutz, R.J., S.P.A. Fodor, T.R. Gingeras, and D.J. Lockhart. 1999. High density synthetic oligonucleotide arrays. Nat. Genet. 21:20-24.

121

Litt, M., J. Luty, M. Kwak, L. Allen, R.E. Magenis, and G. Mandel. 1989. Localization of a human brain sodium channel gene (< i> SCN2A) to chromosome 2. Genomics 5(2):204-208. Liu, N., B. Zhou, X. Zhao, B. Lu, Y. Li, and J. Hao. 2009. Grafting Eggplant onto Tomato Rootstock to Suppress Verticillium dahliae Infection: The Effect of Root Exudates. HortScience 44(7):2058-2062. Liu, S.C., B.E. Liedl, and M.A. Mutschler. 1995. Alterations of the manifestations of hybrid breakdown in Lycopersicon esculentum L. pennellii F 2 populations containing L. esculentum versus L. pennellii cytoplasm. Sexual plant reproduction 8(6):361-368.

Lockwood, J.L., D.L. Yoder, and M.K. Beute. 1970. Grafting Eggplants-D on Resistant Rootstock as a Possible Approach for Control of Verticillium wilt. Plant Disease Reporter 54(10):846-848. Loos, R.A., F.R. Brunele Caliman, and D.J. Henriques da Silva. 2009. Grafting, production and quality of tomato growth in protected environment. Ciencia Rural 39(1):232-235. Lopez-Perez, J.A., M. Escuer, M.A. Diez-Rojo, L. Robertson, A. Piedra Buena, J. Lopez- Cepero, and A. Bello. 2011. Host Range of Meiloidogyne arenaria (Neal, 1889) Chitwood, 1949 (Nematoda:Meloidogynidae) in Spain. Nematropica 41(1):130- 140. Louws, F.J., C.L. Rivard, and C. Kubota. 2010. Grafting fruiting vegetables to manage soil-borne pathogens, foliar pathogens, arthropods and weeds. Scientia Horticulturae 127(2):127-146. Lowman, M.S. and J.W. Kelly. 1946. The presence of mydriatic alkaloids in tomato fruit from sicon grown on Datura stramonium rootstock. Proceedings of the American Society for Horticultural Science 48(DEC):249-259. Luckwill, L.C. 1937. Studies in the Inheritance of Physiological Characters: IV. Hybrid Vigour in the Tomato: Part 2. Manifestations of Hybrid Vigour during the Flowering Period. Annals of Botany 1(3):379-408. Lykas, C., C. Kittas, and A. Zambeka. 2008. Water and Fertilizers Use Efficiency in Grafted and Non Grafted Tomato Plants on Soilless Culture. p. 1551-1555. In: S.M.G.S.M.A.S.E. DePascale (ed). Proceedings of the International Symposium on High Technology for Greenhouse System Management, Vols 1 and 2. Mackay, T.F.C. 2001. The genetic architecture of quantitative traits. Annu. Rev. Genet. 35(1):303-339. Malcolm, P., P. Holford, B. McGlasson, J. Conroy, and I. Barchia. 2007. Growth and its partitioning in Prunus rootstocks in response to root zone temperature. Scientia Horticulturae 112(1):58-65. Manuel Estevez-Caparros, J., M. Diaz-Perez, and F. Camacho-Ferre. 2011. Influence of several rootstocks on yield of cultivars of pear cherry tomato cultivated under mesh greenhouse. Journal of Food Agriculture & Environment 9(1):364-368.

122

Marsic, N.K. and J. Osvald. 2004. The influence of grafting on yield of two tomato cultivars (Lycopersicon esculentum Mill.) grown in a plastic house. Acta agriculturae slovenica 83:243-249. Martin, G.C. and G.E. Wilcox. 1963. Critical Soil temperature for tomato plant growth. Proc.Am.Soil.Sci.Soc. 27:565-567. Martinez-Rodriguez, M.M., M.T. Estan, E. Moyano, J.O. Garcia-Abellan, F.B. Flores, J.F. Campos, M.J. Al-Azzawi, T.J. Flowers, and M.C. Bolarin. 2008. The effectiveness of grafting to improve salt tolerance in tomato when an 'excluder' genotype is used as scion. Environ. Exp. Bot. 63(1-3):392-401. Martinez-Rodriguez, M.M., A. Santa-Cruz, M.T. Estan, M. Caro, and M.C. Bolarin. 2002. Influence of rootstock in the tomato response to salinity. p. 455-460. In: Aksoy,U.Anac D.Anac S.Beltrao J.BenAsher J.Cuartero J. and T.J.H.S. Flowers (eds.). Proceedings of the International Symposium on Techniques to Control Salination for Horticultural Productivity. Martorana, M., F. Giuffrida, C. Leonardi, and S. Kaya. 2007. Influence of rootstock on tomato response to salinity. p. 555-561. In: A.S.W.H. Hanafi (ed). Proceedings of the VIIIth International Symposium on Protected Cultivation in Mild Winter Climates: Advances in Soil and Soilless Cultivation Under Protected Environment. Masayuki, O. New Grafting Methods for Fruit-Bearing Vegetables in Japan. a): A 92(43,972):58.5. Masuda, M., A. Todaka, and T. Tanaka. 1985. Effects of Rootstocks on the Fruit Quality and Chemical Composition of Grafted Tomato Plants. Bulletin of the Faculty of Agriculture Miyazaki University 32(1):187-196. Matsubara, Y. and N. Tanaka. 2005. Cultivar difference in growth of tomato rootstocks under low temperature stress. Environment Control in Biology 43(2):139-142. Matsuzoe, N., H. Aida, K. Hanada, M. Ali, H. Okubo, and K. Fujieda. 1996. Fruit quality of tomato plants grafted on Solanum rootstocks. Journal of the Japanese Society for Horticultural Science 65(1):73-80. Matsuzoe, N., H. Nakamura, H. Okubo, and K. Fujieda. 1993. Growth and Yield of Tomato Plants Grafted on Solanum Rootstocks. Journal of the Japanese Society for Horticultural Science 61(4):847-855. Matsuzoe, N., H. Okubo, and K. Fujieda. 1993. Resistance of tomato plants grafted on Solanum Rootstock to Bacterial Wilt and Root-Knot Nematode. Journal of the Japanese Society for Horticultural Science 61(4):865-872. Matsuzoe, N., H. Aida, K. Hanada, M. Ali, H. Okubo, and K. Fujieda. 1995. Mineral content in leaves of tomato plants grafted on Solanum rootstocks. Science Bulletin of the Faculty of Agriculture Kyushu University 49(3-4):143-148. Mattos, L.M., J.B. Pinheiro, J.L. de Mendonca, and J.P. de Santana. 2011. Wild Solanaceae: Potential for the Use as Rootstocks Resistant to Root-Knot Nematode (Meloidogyne spp.). p. 243-247. In: C. Hale (ed). Xxviii International Horticultural Congress on Science and Horticulture for People. Mavi, K., S. Ermis, and I. Demir. 2006. The effect of priming on tomato rootstock seeds in relation to seedling growth. Asian Journal of Plant Sciences 5(6):940-947.

123

Maxon-Smith, J.W. 1966. A new Tomato Rootstock Seed Parent. Euphytica 15(3):395- &. Maxon-Smith, J.W. 1968. A New Wilt-Resistant Parent for Producing Tomato Rootstock Seed. Euphytica 17(3):414-&. McAvoy, T., J.H. Freeman, S.L. Rideout, S.M. Olson, and M.L. Paret. 2012. Evaluation of Grafting Using Hybrid Rootstocks for Management of Bacterial Wilt in Field Tomato Production. HortScience 47(5):621-625. McCouch, S.R., L. Teytelman, Y. Xu, K.B. Lobos, K. Clare, M. Walton, B. Fu, R. Maghirang, Z. Li, and Y. Xing. 2002. Development and mapping of 2240 new SSR markers for rice (Oryza sativa L.). DNA research 9(6):199-207. McCully, M.E. 1983. Structural aspects of graft development. Vegetative Compatibility Responses in Plants.Baylor Univ.Press, Waco, TX :71-88. McSorley, R., K.H. Wang, E.N. Rosstopt, N. Kokalis-Burelle, H.N. Hans Petersen, H.K. Gill, and R. Krueger. 2009. Nonfumigant alternatives to methyl bromide for management of nematodes, soil-borne disease, and weeds in production of snapdragon (Antirrhinum majus). Int. J. Pest Manage. 55(4):265-273. Mian, I.H., M. Ali, and R. Akhter. 1995. Grafting on Solanum rootstocks to control root- knot to tomato and bacterial wilt of eggplant. Bulletin of the Institute of Tropical Agriculture Kyushu University 18(0):41-47. Miguel, A., J.V. Maroto, A. San Butista, C. Baixauli, V. Cebolla, B. Pascual, S. Lopez, and J.L. Guardiola. 2004. The grafting of triploid watermelon is an advantageous alternative to soil fumigation by methyl bromide for control of Fusarium wilt. Scientia Horticulturae 103:9-17. Minuto, A., G. Gilardi, M.L. Gullino, and A. Garibaldi. 2008. Increasing severity of attacks of Colletotrichum coccodes on grafted tomatoes. p. 101-106. In: M.S.J.W.B.G. Cirulli (ed). Proceedings of the Fifteenth Meeting of the Eucarpia Tomato Working Group. Mohammed, S.M.T., M. Humidan, M. Boras, and O.A. Abdalla. 2009. Effect of Grafting Tomato on Different Rootstocks on Growth and Productivity under Glasshouse Conditions. Asian Journal of Agricultural Research 3(2):47-54. Moore, R. and D.B. Walker. 1981. Studies of vegetative compatibility-incompatibility in higher plants. I. A structural study of a compatible autograft in Sedum telephoides (Crassulaceae). Am. J. Bot. :820-830. Moore, R. 1981. Graft Compatibility and Incompatibility in Higher - Plants. Dev. Comp. Immunol. 5(3):377-389. Moore, R. 1982. Studies of Vegetative Compatibility - Incompatibility in Higher Plants. 5. A Morphometric Analysis of the Development of a Compatible and an Incompatible Graft. Canadian Journal of Botany-Revue Canadienne De Botanique 60(12):2780-2787. Moore, R. 1983. Studies of Vegetative Compatibility - Incompatibility in Higher - Plants. 4. The development of Tensile - Strength in a Compatible and an Incompatible Graft. Am. J. Bot. 70(2):226-231. Morra, L. 2004. Grafting in vegetable crops. The Production in the Greenhouse After the Era of the Methyl Bromide, Comiso, Italy :147-154.

124

Morra, L. 2004. Grafting in vegetable crops. The Production in the Greenhouse After the Era of the Methyl Bromide, Comiso, Italy :147-154. Mosse, B. 1962. Graft-incompatibility in fruit trees with particular reference to its underlyiung causes. Tech.Commun.Bur.Hort.Plant.Crops, E.Malling (28):36. Munns, R. 2002. Comparative physiology of salt and water stress. Plant, Cell and Environment 25:239-250. Nakaho, K. and S. Takaya. 1993. Resistance of Tomato Rootstock Cultivars to Pseudomonas solanacearum Evaluated by Infection Rate Under Different Testing Conditions. . Nakaho, K. 1997. Distribution and multiplication of Ralstonia solanacearum (Synonym Pseudomonas solanacearum) in tomato plants of resistant rootstock cultivar LS-89 and susceptible Ponderosa. Ann. Phytopathol. Soc. Japan 63(2):83-88. Neau, L.J.M. Grafting method for vegetable plants of same family|involves using tongue and groove graft bound with raffia and covered with plastics bag. Neau L J M, . Nieuwhof, M., L.C.P. Keizer, S. Zijlstra, and P. Lindhout. 1999. Genotypic variation for root activity in tomato (Lycopersicon esculentum Mill.) at different root temperatures. J. Genet. Breed. 53(4):271-278. Nobuoka, T., M. Oda, and H. Sasaki. 1996. Effects of relative humidity, light intensity and leaf temperature on transpiration of tomato scions. Journal of the Japanese Society for Horticultural Science 64(4):859-865. Obrero, F.P., M. Aragaki, and E.E. Trujillo. 1971. Tomato Bacterial Wilt Inoculation of Susceptible Scions Grafted to Resistant Rootstock. Plant Disease Reporter 55(6):521-&. O'Connell, S., M. Peet, C.L. Rivard, F.J. Louws, and C.D. Harlow. 2009. The Grafted Heirloom Tomato System for Organic Production in High Tunnels: Are There Advantages in the Absence of Diseases? HortScience 44(4):1056-1056. Oda, M. 2002. Grafting of vegetable crops. Sci. Rep. Agr. & Biol. Sci. Osaka Pref. Univ. 54:49-72. Oda, M. 2008. Use of grafted seedlings for vegetable production in Japan. p. 15-20. In: Oh,D.Kubota C.Mitra S.K.Park H.Shigyo M.Shahak Y. (ed). Proceedings of the International Symposium on Cultivation and Utilization of Asian, Sub-Tropical, and Underutilized Horticultural Crops. Oda, M., M. Maruyama, and G. Mori. 2005. Water transfer at graft union of tomato plants grafted onto Solanum rootstocks. Journal of the Japanese Society for Horticultural Science 74(6):458-463. Oda, M., M. Nagaoka, T. Mori, and M. Sei. 1994. Simultaneous Grafting of Young Tomato Plants Using Grafting Plates. Scientia Horticulturae 58(3):259-264. Oda, M. and T. Nakajima. 1992. Adhesive grafting of Chinese cabbage on turnip. HortScience 27(10):1136-1136. Oda, M., K. Tsuji, and H. Sasaki. 1993. Effect of Hypocotyl Morphology on Survival Rate and Growth of Cucumber Seedlings Grafted on Cucurbita spp. Japan Agric. Res. Q. 26:259-259.

125

Oda, M. 2007. Vegetable seedling grafting in Japan. Ihc 2006: Proceedings of the Xxvii International Horticultural Congress on Global Horticulture: Diversity and Harmony, Introduction (759):175-180. Oda, M., K. Okada, and H. Sasaki. 2000. Effects of transplant container and Solanum rootstocks on the incidences of overgrowth and unmarketable fruits in tomato plants planted with plug seedlings. Environment Control in Biology 38(4):273- 280. Oka, Y., R. Offenbach, and S. Pivonia. 2004. Pepper rootstock graft compatibility and response to Meloidogyne javanica and M-incognita. J. Nematol. 36(2):137-141. Opeña, R.T., J.T. Chen, T. Kalb, and P. Hanson. 2001. Seed production of open- pollinated tomato lines. Asian Vegetable Research and Development Center publication (01-528). Oztekin, G., Y. Tuzel, A. Gul, and I.H. Tuzel. 2007. Effects of grafting in saline conditions. p. 349-355. In: D.J. Cantliffe (ed). Proceedings of the International Symposium on Advances in Environmental Control, Automation and Cultivation Systems for Sustainable, High-Quality Crop Production under Protected Cultivation. Oztekin, G.B., F. Giuffrida, Y. Tuzel, and C. Leonardi. 2009. Is the vigour of grafted tomato plants related to root characteristics? Journal of Food Agriculture & Environment 7(3-4):364-368. Oztekin, G.B. and Y. Tuzel. 2011. Comparative Salinity Response Among Tomato Genotypes and Rootstocks. Pak. J. Bot. 43(6):2665-2672. Oztekin, G.B. and Y. Tuzel. 2011. Salinity response of some tomato rootstocks at seedling stage. African Journal of Agricultural Research 6(20):4726-4735. Pae, D., J. Yoon, and J. Lee. 1993. Non-destructive screening for resistance to multiple diseases using excised plant parts in pepper (Capsicum species). HortScience 28(5):207. Pae, D., J. Yoon, and J. Lee. 1997. Non-destructive screening for resistance to Phytophthora capsici using excised shoots in pepper. Journal of the Korean Society for Horticultural Science 38(5):474-477. Palada, M.C. and D.L. Wu. 2007. Increasing off-season tomato production using grafting technology for peri-urban agriculture in Southeast Asia. p. 125-131. In: K.K. Chow (ed). Proceedings of the International Conference & Exhibition on Soilless Culture. Palada, M.C. and D.L. Wu. 2008. Evaluation of chili rootstocks for grafted sweet pepper production during the hot-wet and hot-dry seasons in Taiwan. p. 151-157. In: R.K.B.S.D. Prange (ed). Proceedings of the International Symposium on Sustainability through Integrated and Organic Horticulture. Palada, M.C. and M. Ali. 2007. Evaluation of technologies for improving year-round production of safe vegetables in peri-urban agriculture of Southeast Asia. p. 271- 281. In: T.A.W.I.J. Lumpkin (ed). Proceedings of the International Symposium on Horticultural Plants in Urban and Peri-Urban Life.

126

Paplomatas, E.J., K. Elena, A. Tsagkarakou, and A. Perdikaris. 2000. Control of Verticillium wilt of tomato and cucurbits through grafting of commercial varieties on resistant rootstocks. Paplomatas, E.J., K. Elena, and A. Tsagkarakou. 2000. Screening tomato and cucurbit rootstocks for resistance to Verticillium dahliae. Bulletin OEPP 30(2):239-242. Paret, M.L., J.H. Freeman, T. McAvoy, S.L. Rideout, and S.M. Olson. 2011. Grafting of a commercially important but bacterial wilt susceptible tomato variety with disease resistant rootstocks for open field production. Phytopathology 101(6):S136-S137. Park, Y., D. Cha, K. Jung, B. Kang, and J. Lee. 2005. Quality changes in oriental melon as affected by washing after fruit harvest and 1-MCP treatment. HortScience 40(4):1004. Passam, H.C., M. Stylianou, and A. Kotsiras. 2005. Performance of eggplant grafted on tomato and eggplant rootstocks. European Journal of Horticultural Science 70(3):130-134. Paterson, A.H. 2010. Molecular dissection of complex traits. CRC Press. Patil, A.A. and S.S. Patil. 1988. Preliminary studies on combiners and combinations for quality traits in tomato (Lycopersicon esculentum Mill.). Plant Foods for Human Nutrition 38(1):43-49. Patrick, B., D. Montague, L. Thomas, B. Sarah, H. Erin, K. Kraig, L. Eike, R. Todd, T. Kristi, Y. Farbod, and D. M. 2005. Global Horticulture Assessment. Gent- Oostakker, Belgium. Paul, E.M.M., R.C. Hardwick, and P.F. Parker. 1984. Genotypic variation in the response to sub-optimal temperatures of growth in tomato (Lycopersicon esculentum Mill). New Phytology 98:221-230. Peet, M., S. O'Connell, C. Rivard, C. Harlow, and F. Louws. 2009. Physiological Disorders in Grafted Heirloom Tomatoes Grown in High Tunnels Using Organic Production Practices. HortScience 44(4):979-979. Pek, Z., A. Pogonyi, and L. Helyes. 2007. Effects of rootstock on yield and fruit quality of indeterminate tomato (Lycopersicon lycopersicum L.) karsten). Cereal Research Communications 35(2):909-912. Peralta, I.E. and D.M. Spooner. 2007. History, origin and early cultivation of tomato (Solanaceae). Genetic improvement of solanaceous crops 2:1-27. Peres, L.E.P., R.F. Carvalho, A. Zsogon, O.D. Bermudez-Zambramo, W.G.R. Robles, and S. Tavares. 2005. Grafting of tomato mutants onto potato rootstocks: An approach to study leaf-derived signaling on tuberization. Plant Science 169(4):680-688. Philis, J. 1970. Resistance of F1 Tomato Hybrid Rootstocks to Potato Cyst Nematode, Heterodera rostochiensis woll. Horticultural Research 10(1):78-&. Phillips, R.L. and I.K. Vasil. 2001. DNA-based markers in plants. Springer, . Pina, A. and P. Errea. 2005. A review of new advances in mechanism of graft compatibility-incompatibility. Scientia Horticulturae 106(1):1-11. Poessel, J., M. Faurobert, M. Loonis, G. Olivier, M.N. Corre, V. Restier, J.M. Audergon, and M. Masse. 2006. Physiological and genetic studies on apricotl/Prunus

127

rootstocks graft compatibility. Proceedings of the XIIth ISHS Symposium on Apricot Culture and Decline, Vols 1 and 2 (701):87-93. Pogonyi, A., Z. Pek, L. Helyes, and A. Lugasi. 2005. Effect of grafting on the tomato's yield, quality and main fruit components in spring forcing. Acta Aliment. 34(4):453-462. Puskas, J.D., W.H. Williams, P.G. Duke, J.R. Staples, K.E. Glas, J.J. Marshall, M. Leimbach, P. Huber, S. Garas, and B.H. Sammons. 2003. Off-pump coronary artery bypass grafting provides complete revascularization with reduced myocardial injury, transfusion requirements, and length of stay: a prospective randomized comparison of two hundred unselected patients undergoing off-pump versus conventional coronary artery bypass grafting. J. Thorac. Cardiovasc. Surg. 125(4):797. Qaryouti, M.M., W. Qawasmi, H. Hamdan, and M. Edwan. 2006. Tomato fruit yield and quality as affected by grafting and growing system. p. 199-206. In: A.N.S.W.Q.M. Fardous (ed). Proceedings of the 1st International Symposium on Fresh Food Quality Standards: Better Food by Quality and Assurance. Quintino, L. and I. Pires. 2004. Overview of the technology of adhesive bonding in medical applications. Businessbriefing: Medicaldevicemanufacturing&Technology . Rafalski, A. 2002. Applications of single nucleotide polymorphisms in crop genetics. Curr. Opin. Plant Biol. 5(2):94-100. Ricardez, M., N. Rodriguez, M. Diaz, and F. Camacho. 2008. Influence of Rootstock, Cultivar and Environment on Tomato Yield under Greenhouse. p. 443-448. In: C.K.M. Kubota (ed). Proceedings of the International Workshop on Greenhouse Environmental Control and Crop Production in Semi-Arid Regions. Rick, C.M. 1980. Tomato. Hybridization of crop plants (hybridization):669-680. Rick, C.M. 1986. Reproductive isolation in the Lycopersicon peruvianum complex. Solanaceae: biology and systematics :477-495. Rick, C.M., J.W. DeVerna, R.T. Chetelat, and M.A. Stevens. 1986. Potential contributions of wide crosses to improvement of processing tomatoes. Rieseberg, L.H. 1997. Hybrid origins of plant species. Annu. Rev. Ecol. Syst. :359-389. Rivard, C.L., S. O'Connell, M.M. Peet, and F.J. Louws. 2010. Grafting tomato with interspecific rootstock to manage diseases caused by Sclerotium rolfsii and southern root-knot nematode. Plant Dis. 94(8):1015-1021. Rivard, C.L. and F.J. Louws. 2007. Grafting for soil-borne disease management in organic heirloom tomato production. Phytopathology 97(7):S99-S99. Rivard, C.L., F.J. Louws, S. O'Connell, and M.M. Peet. 2009. Grafting tomato with interspecific rootstock provides effective management for southern blight and root- knot nematodes. Phytopathology 99(6):S109-S109. Rivard, C.L., F.J. Louws, M.M. Peet, and S. O'Connell. 2008. High tunnels and grafting for disease management in organic tomato production. Phytopathology 98(6):S133-S133. Rivard, C.L. and F.J. Louws. 2008. Grafting to Manage Soil-borne Diseases in Heirloom Tomato Production. HortScience 43(7):2104-2111.

128

Rivard, C.L., O. Sydorovych, S. O'Connell, M.M. Peet, and F.J. Louws. 2010. An Economic Analysis of Two Grafted Tomato Transplant Production Systems in the United States. Horttechnology 20(4):794-803. Rivero, R.M., J.M. Ruiz, and L. Romero. 2003. Role of grafting in horticultural plants under stress conditions. Journal of food agriculture and environment 1:70-74. Rivero, R.M., J.M. Ruiz, and L. Romero. 2004. Iron metabolism in tomato and watermelon plants: Influence of grafting. J. Plant Nutr. 27(12):2221-2234. Robbins, M.D., A. Darrigues, S. Sim, M.A.T. Masud, and D.M. Francis. 2009. Characterization of hypersensitive resistance to bacterial spot race T3 (Xanthomonas perforans) from tomato accession PI 128216. Phytopathology 99(9):1037-1044. Robbins, M.D., S. Sim, W. Yang, A. Van Deynze, E. van der Knaap, T. Joobeur, and D.M. Francis. 2011. Mapping and linkage disequilibrium analysis with a genome- wide collection of SNPs that detect polymorphism in cultivated tomato. J. Exp. Bot. 62(6):1831-1845. Romano, D. and A. Paratore. 2001. Effects of grafting on tomato and eggplant. p. 149- 153. In: J.A.M.P.F.C.N. Fernandez (ed). Proceedings of the Fifth International Symposium on Protected Cultivation in Mild Winter Climates: Current Trends for Sustainable Technologies, Vols I and Ii. Rouphael, Y., M. Cardarelli, G. Colla, and E. Rea. 2008. Yield, mineral composition, water relations, and water use efficiency of grafted mini-watermelon plants under deficit irrigation. HortScience 43(3):730-736. Rouphael, Y., M. Cardarelli, E. Rea, and G. Colla. 2008. Grafting of cucumber as a means to minimize copper toxicity. Environ. Exp. Bot. 63(1-3):49-58. Rouphael, Y., D. Schwarz, A. Krumbein, and G. Colla. 2010. Impact of grafting on product quality of fruit vegetables. Scientia Horticulturae 127(2):172-179. Ruiz, J.M., B. Blasco, R.M. Rivero, and L. Romero. 2005. Nicotine-free and salt-tolerant tobacco plants obtained by grafting to salinity-resistant rootstocks of tomato. Physiol. Plantarum 124(4):465-475. Sacks, E.J. and D.A.S. Clair. 1996. Cryogenic storage of tomato pollen: effect on fecundity. HortScience 31(3):447-448. Salehi, R., A. Kashi, J. Lee, M. Babalar, M. Delshad, S. Lee, and Y. Huh. 2010. Leaf Gas Exchanges and Mineral Ion Composition in Xylem Sap of Iranian Melon Affected by Rootstocks and Training Methods. HortScience 45(5):766-770. Salehi-Mohammadi, R., A. Kashi, S.G. Lee, Y.C. Huh, J. Lee, M. Babalar, and M. Delshad. 2009. Assessing the Survival and Growth Performance of Iranian Melon to Grafting onto Cucurbita Rootstocks. Korean Journal of Horticultural Science & Technology 27(1):1-6. San Bautista, A., A. Calatayud, S.G. Nebauer, B. Pascual, J. Vicente Maroto, and S. Lopez-Galarza. 2011. Effects of simple and double grafting melon plants on mineral absorption, photosynthesis, biomass and yield. Scientia Horticulturae 130(3):575-580. Sanchez-Rodriguez, E., M. del Mar Rubio-Wilhelmi, B. Blasco, R. Leyva, L. Romero, and J. Manuel Ruiz. 2012. Antioxidant response resides in the shoot in reciprocal

129

grafts of drought-tolerant and drought-sensitive cultivars in tomato under water stress. Plant Science 188:89-96. Sanchez-Rodriguez, E., J. Manuel Ruiz, F. Ferreres, and D.A. Moreno. 2011. Phenolic Metabolism in Grafted versus Nongrafted Cherry Tomatoes under the Influence of Water Stress. J. Agric. Food Chem. 59(16):8839-8846. Sanchez-Rodriguez, E., J.M. Ruiz, F. Ferreres, and D.A. Moreno. 2012. Phenolic profiles of cherry tomatoes as influenced by hydric stress and rootstock technique. Food Chem. 134(2):775-782. Sano, Y. and F. Kita. 1978. Reproductive barriers distributed in Melilotus species and their genetic bases. Canadian Journal of Genetics and Cytology 20(2):275-289. Santa-Cruz, A., M.M. Martinez-Rodriguez, F. Perez-Alfocea, R. Romero-Aranda, and M.C. Bolarin. 2002. The rootstock effect on the tomato salinity response depends on the shoot genotype. Plant Science 162(5):825-831. Santamour Jr, F.S. 1988. Graft incompatibility related to cambial peroxidase isozymes in Chinese chestnut. J.Environ.Hort 6(2):33-39. Saranga, Y., A. Cahaner, D. Zamir, A. Marani, and J. Rudich. 1992. Breeding tomatoes for salt tolerance: inheritance of salt tolerance and related traits in interspecific populations. TAG Theoretical and Applied Genetics 84(3):390-396. Sasaki, T. Grafting tool of seedling e.g. aubergine, has vertical crack which is formed in inner periphery of housing and extended from upper end to lower end of housing. Chubu Nozai Kk, . Savvas, D., G. Colla, Y. Rouphael, and D. Schwarz. 2010. Amelioration of heavy metal and nutrient stress in fruit vegetables by grafting. Scientia Horticulturae 127(2):156-161. Savvas, D., D. Papastavrou, G. Ntatsi, A. Ropokis, C. Olympios, H. Hartmann, and D. Schwarz. 2009. Interactive Effects of Grafting and Manganese Supply on Growth, Yield, and Nutrient Uptake by Tomato. HortScience 44(7):1978-1982. Schafer, K.S. 1999. Methyl Bromide Phase-out Strategies: A Global Compilation of Laws and Regulations. United Nations Environment Programe, . Schwarz, D., Y. Rouphael, G. Colla, and J.H. Venema. 2010. Grafting as a tool to improve tolerance of vegetables to abiotic stresses: Thermal stress, water stress and organic pollutants. Scientia Horticulturae 127(2):162-171. Scott, J.W. 1978. Breeding and pollination studies of heterostylous genotypes for hybrid seed production in Lycopersicon esculentum Mill. (tomato). Shakya, S.M. 1982. Effects of environment and flower maturity at pollination on hybrid seed production and selfing contamination of several tomato genotypes. Ohio State University. Shao, H., L.C. Burrage, D.S. Sinasac, A.E. Hill, S.R. Ernest, W. O'Brien, H. Courtland, K.J. Jepsen, A. Kirby, and E.J. Kulbokas. 2008. Genetic architecture of complex traits: large phenotypic effects and pervasive epistasis. Proceedings of the National Academy of Sciences 105(50):19910-19914. Sheela, K.B., P.K. Gopalakrishnan, and K.V. Peter. 1984. Resistance to Bacteria Wilt in a Set of Eggplant Breeding Lines. Indian J. Agric. Sci. 54(6):457-460.

130

Shibata, H. Graft support for junction part of e.g. tomato, has cylindrical elastic body, and slit formed at one end of sapling insertion hole, where inner side edge of slit and sapling insertion hole is formed as curved surface. Nippon Piaresu Kogyo Kk, . Shibuya, T., H. Nakashima, K. Shimizu-Maruo, and T. Kawara. 2007. Improvement of graft development in tomato and eggplant grafted cuttings by supplying warmed water to graft union during low-air-temperature storage. Journal of the Japanese Society for Horticultural Science 76(3):217-223. Shimizu, K., T. Shibuya, T. Kawara, and K. Tsuchiya. Seedling raising method for graft seedling of vegetables e.g. aubergine, involves maintaining temperature of scion lower than warm water temperature and rooting of rootstock and scion to form graft seedling. Bergearth Co Ltd; Taiyo Kogyo Kk; Univ Osaka Prefecture; Mkv Platech Kk, . Shorter, J.H., C.E. Kolb, P.M. Crill, R.A. Kerwin, R.W. Talbot, M.E. Hines, and R.C. Harriss. 1995. Rapid degradation of atmospheric methyl bromide in soils. Nature 377(6551):717-719. Shull, G.H. 1908. The composition of a field of maize. J. Hered. (1):296-301. Shull, G.H. 1911. The genotypes of maize. Am. Nat. 45(532):234-252. Sim, S., M. Robbins, C. Chilcott, T. Zhu, and D. Francis. 2009. Oligonucleotide array discovery of polymorphisms in cultivated tomato (Solanum lycopersicum L.) reveals patterns of SNP variation associated with breeding. BMC Genomics 10(1):466. Smith, L.M., K. Bomblies, and D. Weigel. 2011. Complex evolutionary events at a tandem cluster of Arabidopsis thaliana genes resulting in a single-locus genetic incompatibility. PLoS genetics 7(7):e1002164. Smith, P.G. 1944. Embryo Culture of a Tomato Species Hybrid. Proceeding of the American Society for Horticultural Science 44:413-416. Solt, M.L. 1957. Nicotine Production and Growth of Tobacco Scions on Tomato Rootstocks. Plant Physiol. 32(5):484-490. Solt, M.L. and R.F. Dawson. 1958. Production, Translocation and Accumulation of Alkaloids in Tobacco Scions Grafted to Tomato Rootstocks. Plant Physiol. 33(5):375-381. Somerville, C. and S. Somerville. 1999. Plant functional genomics. Science 285(5426):380-383. Stapleton, J.J. and J.E. DeVay. 1986. Soil solarization: A non-chemical approach for management of plant pathogens and pests. Crop Protection 5:190-198. Stapleton, J.J. 2000. Soil Solarization in various agricultural production systems. Crop Protection 19:837-841. Stapleton, J.J., J.E. DeVay, C.L. Elmore, and Nations.,Food and Agriculture Organization of the United. 1998. Soil solarization and integrated management of soil-borne pests : proceedings of the Second International Conference on Soil Solarization and Integrated Management of Soil-borne Pests, Aleppo, Syrian Arab Republic, 16-21 March 1997. Food and Agriculture Organization of the United Nations, Rome.

131

Staub, J.E., F.C. Serquen, and M. Gupta. 1996. Genetic markers, map construction, and their application in plant breeding. HortScience 31(5):729-741. Stebbins, G.L. 1958. The inviability, weakness, and sterility of interspecific hybrids. Adv. Genet. 9:147-215. Stephens, S.G. 1950. The internal mechanism of speciation in Gossypium. The Botanical Review 16(3):115-149. Stevens, M.A. and C.M. Rick. 1986. Genetics and breeding. The tomato crop: a scientific basis for improvement :35-109. Sugahara, S., Y. Sakurai, T. Suzuki, K. Ito, F. Yamashita, and Y. Takase. 1989. Breeding of Tsueeze a New Tomato Rootstock Cultivar With a Wide Spectrum of Disease Resistance. Research Bulletin of the Aichi-ken Agricultural Research Center (21):158-169. Suresh, K., M.K. Banerjee, P.S. Partap, and S. Kumar. 1995. Heterosis study for fruit yield and its components in tomato. Ann.Agricult.Res 16(2):212-217. Tateishi, K. 1927. Grafting Watermelon onto Pumpkin. Journal Of Japanese Horticulture 39:5-8. Thorpe, H.J. and W.R. Jarvis. 1981. Grafted Tomatoes Escape Fusarium Foot and Root- rot. Canadian Journal of Plant Science 61(4):1027-1028. Tikoo, S. 1979. Successful Graft Culture of Tomato in Bacterial Wilt Sick Soil. Curr. Sci. 48:259-260. Tjamos, E.C., P.P. Antoniou, and S.E. Tjamos. 2000. Implementation of soil solarization in Greece: conclusions and suggestions. Crop Protection 19:843-846. Turhan, A., N. Ozmen, M.S. Serbeci, and V. Seniz. 2011. Effects of grafting on different rootstocks on tomato fruit yield and quality. Horticultural Science 38(4):142-149. Tzortzakakis, E.A., F.A. Bletsos, and A.D. Avgelis. 2006. Evaluation of Solanum rootstock accessions for control of root-knot nematodes and tobamoviruses. Journal of Plant Diseases and Protection 113(4):188-189. Upstone, M.E. 1968. The effects of methyl bromide fumigation and grafting on yield and root diseases of tomatoes. Plant Pathol. 17(3):103-107. Upstone, M.E. and J. Finney. 1966. Tomato Rootstock and Control of Phytophthora Root Rot. Plant Pathol. 15(1):15-&. USDA. 2000. Economic Implications of the Methyl Bromide Phaseout. Vallejos, C.E. and S.D. Tanksley. 1983. Segregation of isozyme markers and cold tolerance in an interspecific backcross of tomato. TAG Theoretical and Applied Genetics 66(3):241-247. Van der Burg, W.J., J.W. Aartse, R.A. van Zwol, H. Jalink, and R.J. Bino. 1994. Predicting tomato seedling Morphology by X-ray Analysis of Seeds. J.Amer.Soc.Hort.Sci. 119(2):258-263. Van der Burg, W.J., H. Jalink, R.A. van Zwol, J.W. Aartse, and R.J. Bino. 1994. Non- destructive seed evaluation with impact measurements and x-ray analysis. Acta Horticulturae 362:149-157. Varshney, R.K., K. Chabane, P.S. Hendre, R.K. Aggarwal, and A. Graner. 2007. Comparative assessment of EST-SSR, EST-SNP and AFLP markers for

132

evaluation of genetic diversity and conservation of genetic resources using wild, cultivated and elite barleys. Plant Science 173(6):638-649. Venema, J.H., B.E. Dijk, J.M. Bax, P.R. van Hasselt, and J.T.M. Elzenga. 2008. Grafting tomato (Solanum lycopersicum L.) onto the rootstock of a high-altitude accession of Solanum habrochaites improves suboptimal-temperature tolerance. Environ. Exp. Bot. 63(1-3):359-367. Verdejo-Lucas, S., L. Cortada, F.J. Sorribas, and C. Ornat. 2009. Selection of virulent populations of Meloidogyne javanica by repeated cultivation of Mi resistance gene tomato rootstocks under field conditions. Plant Pathol. 58(5):990-998. Verdejo-Lucas, S. and L. Cortada. 2010. Root- Knot Nematodes in Tomato. Verdejo-Lucas, S. and F.J. Sorribas. 2008. Resistance response of the tomato rootstock SC 6301 to Meloidogyne javanica in a plastic house. Eur. J. Plant Pathol. 121(1):103-107. Villareal, R.L. 1970. Control of Bacterial Wilt in Solanaceous Vegetables Through Grafting and Cross Protection. Proceedings of the International Horticultural Congress 1:94. Villareal, R.L., E.S. Sarreal, N.G. David, and B.T. Pamplona. 1970. Grafting for the Control of Bacterial Wilt in Solanaceous Vegetables. Philippine Agriculturist 53(10):527-538. Vrcek, I.V., V. Samobor, M. Bojic, M. Medic-Saric, M. Vukobratovic, R. Erhatic, D. Horvat, and Z. Matotan. 2011. The effect of grafting on the antioxidant properties of tomato (Solanum lycopersicum L.). Spanish Journal of Agricultural Research 9(3):844-851. Walker, D.E. 1952. Intergeneric grafts and their influence on fruit size in the tomato. Univ. of California, June 1952., [Berkeley,. Walse, S.S., R. Krugner, and J.S. Tebbets. 2012. Postharvest treatment of strawberries with methyl bromide to control spotted wing drosophila, Drosophila suzukii. Journal of Asia-Pacific Entomology . Wang, K.H. and R. McSorley. 2008. Exposure Time to lethal temperatures for Meloidogyne incognita Suppression and its implication for Soil solarization. J. Nematol. 40(1):7-12. Weber, J.L. and P.E. May. 1989. Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Am. J. Hum. Genet. 44(3):388. Westerich, J.N., J.M. Osorio Rosa, and S.R. Siciliano Wilcken. 2011. Comparative study of biology of Meloidogyne enterolobii (= M. mayaguensis) and Meloidogyne javanica in tomatoes with Mi gene. Summa Phytopathologica 37(1):35-41. Xia, L. Grafting capsicum planting method, involves taking hybrid tomato with material as rootstock and capsicum seedling as cion, grafting rootstock and cion by splicing, and forming hole for grafting on cross section. Xia L, . Yamaguchi, T. and E. Blumwald. 2005. Developing salt-tolerant crop plants: Challenges and opportunities. Thends in Plant Science 10(12):615-620. Yamakawa, K. 1982. Use of rootstocks in Solanaceous fruit-vegetable production in Japan. Japan Agricultural Research Quarterly, Ibaraki 15(3):175-180.

133

Yamazaki, H., S. Kikuchi, T. Hoshina, and T. Kimura. 2000. Calcium uptake and resistance to bacterial wilt of mutually grafted tomato seedlings. Soil Sci. Plant Nutr. 46(2):529-534. Yamazaki, H., S. Kikuchi, T. Hoshina, and T. Kimura. 2000. Effect of calcium concentration in nutrient solution on development of bacterial wilt and population of its pathogen Ralstonia solanacearum in grafted tomato seedlings. Soil Sci. Plant Nutr. 46(2):535-539. Yang, M., G. Ding, L. Shi, F. Xu, and J. Meng. 2011. Detection of QTL for phosphorus efficiency at vegetative stage in Brassica napus. Plant Soil 339(1-2):97-111. Yang, L.F., Y.L. Zhu, C.M. Hu, Z.L. Liu, and T. Maruo. 2008. Separation and Identification of Specific Proteins in Leaves of Grafted Cucumber under NaCl Stress. p. 183-190. In: C.L.J.M. Chun (ed). Proceedings of the International Symposium on Seed Enhancement and Seedling Production Technology. Yetisir, H. and V. Uygur. 2009. Plant Growth and Mineral Element Content of Different Gourd Species and Watermelon under Salinity Stress. Turkish Journal of Agriculture and Forestry 33(1):65-77. Yordanov, M. 1983. Heterosis in the tomato. Monogr.Theor.Appl.Genet 6:189-219. Zarrouk, O., J. Aparicio, Y. Gogorcena, M. Moreno, J. Pinochet, and R. Infante. 2006. Graft compatibility for new peach rootstocks in nursery. Proceedings of the VIth International Peach Symposium (713):327-329. Zhang, H., S. Yang, W. Zhu, L. Zhu, Y. Wan, L. Yu, and X. Tai. Cultivation method for improving salt tolerance of tomato plant, involves grafting weak salt tolerance variety of tomato on stem of rootstock by performing substrate culture type or field culture type cultivation. Shanghai Agric Sci Acad, . Zhang, Z.J., Y.M. Wang, L.K. Long, Y. Lin, J.S. Pang, and B. Liu. 2008. Tomato rootstock effects on gene expression patterns in eggplant scions. Russian Journal of Plant Physiology 55(1):93-100. Zhang, Z., H. Li, Y. Zhag, Z. Huang, K. Chen, and S. Liu. 2010. Grafting Enhances Copper Tolerance of Cucumber Through Regulating Nutrient Uptake and Antioxidative System. Agricultural Sciences in China 9(12):1758-1770. Zhao, X., E.H. Simonne, and R.C. Hochmuth. 2009. Grafting as an Alternative to Methyl Bromide in Field Tomato Production. HortScience 44(4):1058-1058. Zhen, A., Z. Bie, Y. Huang, Z. Liu, and Q. Li. 2010. Effects of scion and rootstock genotypes on the anti-oxidant defense systems of grafted cucumber seedlings under NaCl stress. Soil Sci. Plant Nutr. 56(2):263-271. Zhu, J., Z. Bie, Y. Huang, and X. Han. 2008. Effect of grafting on the growth and ion concentrations of cucumber seedlings under NaCl stress. Soil Sci. Plant Nutr. 54(6):895-902. Zijlstra, S. and A.P.M. Dennijs. 1987. Effects of Root Systems of Tomato Genotypes on Growth and Earliness, Studied in Grafting Experiment at Low- Temperature. Euphytica 36(2):693-700.

134