Fingerprinting and Genetic Stability of Rubus Using Molecular Markers

AN ABSTRACT OF THE THESIS OF

Nina Rosa F. Castillo for the degree of Master of Science in Plant Physiology presented on September 13, 2006. Title: Fingerprinting and Genetic Stability of Rubus Using Molecular Markers

Abstract approved: ______

Barbara M. Reed

Two studies used DNA markers to assess their usefulness in germplasm identification and evaluation of genetic stability in four cryopreserved Rubus accessions that were stored for over 10 years in liquid nitrogen. In the first study, 12 robust Simple Sequence Repeat (SSR) markers were developed from two microsatellite-enriched libraries of ‘Meeker’ raspberry and ‘Marion’ blackberry. Of the 70 and 78 SSR-containing sequences identified, four SSR markers were obtained from the ‘Meeker’ library and eight from that of ‘Marion’. These twelve genomic

SSRs and one Expressed Sequence Tag- (EST)-SSR designed from an (AT)6- containing R. idaeus sequence (AF292369) from Genbank were used for fingerprinting 48 raspberry and 48 blackberry cultivars stored at the Repository. The SSR markers developed in Rubus were highly polymorphic. Twelve SSRs generated 96 alleles in raspberries and 177 in blackberries. These markers distinguished between the raspberries and blackberries except in ‘Logan’ and ‘Boysen’ clones. Neighbor Joining cluster analysis based on the proportion of shared allele distance using 13 SSRs separated the cultivars into two main groups: the raspberries and the blackberries. Hybrid berries and cultivars with uncommon ancestry grouped separately from the two major groups. The raspberry and blackberry groups were further divided according to their pedigrees. In the second study, two types of markers, SSRs and Amplified Fragment Length Polymorphisms (AFLP) were used to evaluate genetic fidelity of regrown cryopreserved Rubus shoot tips stored for 12 years in liquid nitrogen. Analyses were done on two groups of plants separated based on the length of time they were subcultured after storage. Group one plants were analyzed

after subculturing for seven months using Platinum Taq polymerase in the PCR reactions while Group two plants were analyzed immediately after recovery from cryopreservation using AmpliTaq Gold polymerase in the PCR reactions. No polymorphism was detected in either group of plants based on SSR analysis using 10 loci. Ten AFLP primer pairs amplified 547 fragments in R. grabowskii, 400 in ‘Mandarin’, 530 in ‘Silvan’, and 521 in ‘Hillemeyer’ Group one plants. An appreciably lower number of PCR products were amplified in Group two plants: 331 fragments in ‘Hillemeyer’, and 379 in ‘Silvan’. Differences in number of AFLP markers between Groups one and two were caused by use of different polymerases during the analysis. AFLP markers, with a high marker index, revealed polymorphism in three of four Rubus genotypes in Group one. However, no polymorphism was detected in Group two plants based on AFLP analysis. Recovery of plants from cryopreservation was low in the three accessions that exhibited AFLP polymorphisms (R. grabowskii, ‘Silvan’ and ‘Mandarin’). ‘Hillemeyer’ regrowth was 80% while R. grabowskii was 40%, ‘Silvan’, 20% and ‘Mandarin’, 10%, indicating less than ideal regrowth for the three genotypes. Such polymorphism might have been generated through somaclonal variants regenerated from callus tissue. Genotypic influence on stability may explain why those three genotypes were prone to variation while ‘Hillemeyer’ remained genetically stable despite long culture periods. High recovery rates and careful treatment and monitoring of regrown plants should therefore be employed to ensure maintenance of genetic fidelity of cryopreserved plants. The variation detected may also be transient and requires further morphological and molecular analysis of adult regrown cryopreserved plants that were transplanted and are growing in the greenhouse.

Fingerprinting and Genetic Stability of Rubus Using Molecular Markers

by Nina Rosa F. Castillo

A THESIS

submitted to

Oregon State University

in partial fulfillment of the requirements for the degree of

Master of Science

Presented September 13, 2006 Commencement June 2007

Master of Science thesis of Nina Rosa F. Castillo presented on September 13, 2006.

APPROVED:

Major Professor, representing Plant Physiology

Head of the Department of Horticulture

Dean of the Graduate School

I understand that my thesis will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my thesis to any reader upon request.

Nina Rosa F. Castillo, Author

ACKNOWLEDGEMENTS

My sincere appreciation goes to my major professor, Dr. Barbara Reed for accepting me as her advisee and giving me the opportunity to pursue the study. The support and encouragement had been valuable as I finish my degree. My sincere appreciation also goes to Dr. Nahla Bassil for accommodating me in her laboratory and for funding the research. The technical guidance, knowledge and skills imparted and unending patience are greatly appreciated. Thank you also to Dr. Chad Finn for sharing the resources that I need especially during the preparation of the manuscript; Dr. Kim Hummer for sharing valuable information especially during the final stages of the research; Dr. Valerian Dolja for reviewing the manuscript; members of the molecular genetics lab of the repository: especially to Barb Gilmore for the assistance throughout the experiment, Wambui Njuguna and Adrienne Oda for always lending a hand; members of the tissue culture lab, Jeanine DeNoma and Janine de Paz for helping me in media preparation and micropropagation; Brian Knaus and Peter Boches for helping me during the data analysis; my friends Vidyasagar, Sathuvalli and Jack Peters for all the support and help extended. Special thanks are due to my family, my dad, mom, sisters, Vicky and Joy, brothers-in-law, Greg and Rigor, my nephew, Jethro for all the emotional support and encouragement and most especially to my husband, Neil, and son, Nikki, for the patience and love always.

CONTRIBUTION OF AUTHORS

Dr. Nahla Bassil and Dr. Barbara Reed assisted with the experimental design, analysis, and writing of all chapters. Dr. Chad Finn provided all the information for the raspberries and blackberries used in fingerprinting in Chapter 2.

TABLE OF CONTENTS

Page

CHAPTER 1. INTRODUCTION ...... 1

The Genus Rubus ...... 2

Raspberries...... 4 Blackberries ...... 5

Plant Genetic Resources of Rubus ...... 6

Rubus Conservation ...... 8

Seed Storage...... 9 Field Genebanks...... 10 In Vitro Methods ...... 11

Genetic Variation ...... 12

Cryopreservation...... 14

Markers and Their Uses in Rubus...... 17

Molecular Marker Use in Assessing Genetic Fidelity ...... 27

Research Objectives...... 29

Literature Cited ...... 30

CHAPTER 2. MICROSATELLITES FOR CULTIVAR IDENTIFICATION IN RASPBERRIES AND BLACKBERRIES...... 44

Abstract ...... 45

Introduction...... 46

Materials and Methods...... 49

Results...... 60

Discussion ...... 63

TABLE OF CONTENTS (Continued)

Page

Literature Cited ...... 71

CHAPTER 3. EVALUATION OF GENETIC STABILITY OF CRYOPRESERVED Rubus MERISTEMS USING SSR AND AFLP MARKERS...... 77

Abstract ...... 78

Introduction...... 79

Materials and Methods...... 83

Results...... 88

Discussion ...... 94

CHAPTER 4. CONCLUDING REMARKS...... 104

Conclusion ...... 105

BIBLIOGRAPHY...... 107

APPENDICES ...... 126

LIST OF FIGURES

Figure Page

2.1. Neighbor Joining (NJ) tree of 96 raspberry and blackberry cultivars based on the proportion of shared alleles distance for 13 SSR loci...... 66

3.1. Neighbor Joining (NJ) dendrogram of cryopreserved and screenhouse grown Rubus plants based on Euclidean genetic distances derived from allele frequencies at 10 SSR loci...... 93

3.2. Neighbor Joining (NJ) dendrogram of cryopreserved and screenhouse grown Rubus plants based on AFLP data...... 93

LIST OF TABLES

Table Page

2.1. The 48 raspberry accessions analyzed with 12 SSR loci, USDA-ARS Plant Introduction numbers (P. I. no.), cultivar name or taxon for species, pedigree or origin and taxon are included...... 53

2.2. The 48 blackberry accessions analyzed with 12 SSR loci, USDA-ARS Plant Introduction numbers (P. I. no.), cultivar name or taxon for species, pedigree or origin and taxon are included...... 55

2.3. Thirteen SSR loci of the expected size...... 58

2.4. Raspberries and blackberries used for initial primer screening...... 59

3.1. List of Rubus accessions used for genetic fidelity assessment...... 84

3.2. SSR primers used in SSR analysis...... 85

3.3. Sequences of primers and adaptors used in AFLP analysis...... 88

3.4. AFLP polymorphisms of the three Rubus accessions (Group one) when compared to their screenhouse-grown counterparts using 10 AFLP primer pairs...... 91

LIST OF APPENDICES

Appendix Page

A. High Throughput DNA Extraction Protocol ……………………………………127

B. Microsatellite Enrichment Protocol ……………………………………………..131

C. List of Primer Pairs that Amplified a Product in Raspberries and/or Blackbrries 140

D. Characterization of the 13 Polymorphic SSR Loci in Raspberries and Blackberries ……………………………………………………………………………….143

E. DNA Fingerprints of 96 Raspberry and Blackberry Cultivars Using 13 SSR Loci ……………………………………………………………………………….146

F. AFLP Primer Pairs Tested for Polymorphism …………………………………...157

G. Electropherograms of the Three Rubus Accessions Showing AFLP Polymorphisms ………………………………………………………………………………160

H. DNA Fingerprints of 4 Screenhouse and Cryopresrved Rubus Accessions Using 10 AFLP Primer Pairs ………………………………………………………….174

LIST OF APPENDIX TABLES

Table Page

C.1. List of primer pairs that amplified a product in ‘Meeker’ and ‘Marion’...... 141

D.1. Characterization of the 13 SSR polymorphic loci in raspberries………………144

D.2.Characterization of the 13 SSR polymorphic loci in blackberries...... 145

E.1. DNA fingerprints of 48 blackberry cultivars using 12 SSR loci...... 147

E.2. DNA fingerprints of 48 raspberry cultivars using 12 SSR loci...... 153

F.1. AFLP primer pairs tested for polymorphism on four Rubus accessions...... 158

H.1. Fragment sizes of the 4 cryopreserved and screenhouse-grown Rubus accessions determined by 10 AFLP primer pairs...... 175

Fingerprinting and Genetic Stability of Rubus Using Molecular Markers

CHAPTER 1

INTRODUCTION

N. R. F. Castillo, N. V. Bassil and B. M. Reed

The Genus Rubus Raspberries and blackberries belong to the Rubus L. genus in the Rosaceae family. Rubus is a diverse genus and contains over 600 species. The United States Department of Agriculture (USDA) Germplasm Resources Information Network (GRIN) database (USDA, 2006) describes 13 subgenera within the genus. Most of the cultivated fruit species belong to two subgenera: the Idaeobatus (raspberries) and the Rubus formerly Eubatus (blackberries). Commercial Rubus fruit crops include the red (R. idaeus L.), black (R. occidentalis L.) and purple (red and black raspberry hybrid) raspberries, blackberries (Rubus species and hybrids), Andean blackberries and cloudberries (R. chamaemorus L.) (Thompson, 1997). The basic chromosome number of Rubus is seven. However, multiple ploidy levels exist in this genus and range from diploid (2n = 2x = 14) to (2n = 14x = 98) (Moore, 1984). Most raspberries are diploid except for a few natural triploids and tetraploids while blackberries range from 2n = 2x = 14 to 2n = 12x = 84 (Daubeny, 1996b). The hybrid berries (‘Loganberry’, ‘Tayberry’, ‘Sunberry’, ‘Tummelberry’, ‘Fertodi Botermo’ and ‘Phenomenal’) are produced from octaploid species (2n = 6x = 56) and tetraploid (2n = 4x = 28) raspberries (Jennings, 1988). Other hybrid cultivars produced using pre-existing hybrids resulted in septaploid cultivars (2n = 7x = 49) like ‘Boysenberry’ and ‘Youngberry’. Most Rubus species are perennial shrubs that vary in habit from erect to trailing and are deciduous with a few evergreen exceptions (Daubeny, 1996b). Economically important species have perennial roots and crowns that produce biennial flowering structures (canes) (Moore and Skirvin, 1990). New canes, referred to as primocanes, are vegetative growth produced from crowns or roots (root suckers). Vernalization shifts the primocanes to reproductive structures (floricanes) that die after producing fruits. Some species are able to flower without vernalization and produce a desirable trait, the primocane flowering type (Lopez-Medina and Moore, 1999).

The fruit is an aggregate containing a number of fleshy drupelets attached to a single receptacle. In the center of each drupelet is a pyrene that usually contains a single seed. The drupelets are attached to a central receptacle or torus, which may or may not remain when the berry is picked. Raspberries are differentiated from blackberries by the way the mature fruit separates from the torus. Raspberry fruits separate from the torus due to a weak abscission zone at the base of each drupelet. Blackberry drupelets adhere to the receptacle and both are part of the edible fruit (Moore and Skirvin, 1990). Native species of Rubus have widespread distribution from the lowland tropics to the subarctic regions (Thompson, 1995). Cultivated raspberries and blackberries however grow best and provide greater economic returns when cultivated in areas with mild winters and long and dry summers. The primary regions of production in North America include the Pacific Northwest (Oregon, Washington and British Columbia) and California although there are some large plantings in parts of Texas and Arkansas, as well as in temperate climates surrounding lakes like New York, Michigan, Pennsylvania and Ohio (Skirvin et al., 2005). Raspberry and blackberry fruits are produced for the fresh fruit market and for use in a number of processed food items (Donnelly and Daubeny, 1986). The demand is high for fresh fruits that are shipped over long distances or are sold locally and for processed fruit products like juice, yogurt, and dessert pies. Given the economic importance of the fruit, many breeding programs are dedicated to produce cultivars of excellent quality, higher yields, greater adaptation to adverse environment and increased pest and disease resistance. The wide diversity of Rubus species provides a potential source of novel traits. The USDA Agricultural Research Service (ARS), National Clonal Germplasm Repository (NCGR) in Corvallis, Oregon is responsible for collecting, maintaining, characterizing and distributing Rubus accessions. Our large and diverse Rubus collection is a great asset to plant breeders and scientists (Thompson, 1995).

Raspberries Idaeobatus has a northerly distribution principally in Asia, eastern and South Africa, Europe and North America. This subgenus is comprised of approximately 200 species. Current raspberry cultivars are derived from the European red raspberry (R. idaeus L.), North American red raspberry (R. strigosus Michx.), the black raspberry (R. occidentalis L.) and the purple raspberries (R. neglectus Peck) that are hybrids between red and black raspberries. Domestication of the red raspberry occurred the past 400 to 500 years. Medicinal uses were described before food uses. Raspberry blossoms were used as an eye ointment or for treating stomach problems. Ancient Greeks and Romans who lived in the city of Troy in the foothills of Mount Ida in Asia Minor were reported to appreciate the fruits 2000 years ago. The Greeks called the fruits “Ida” after Mount Ida and the people who cultivated them (Jennings, 1988). Linnaeus later used the specific epithet “idaeus” for the red raspberry as derived from the word “Ida” and Rubus as the genus name from the Latin word Ruber meaning red. Most red raspberry cultivars originated from two subspecies: European raspberry, R. idaeus, a diploid plant characterized by glandless inflorescence and thimble-shaped fruits; and the North American raspberry, R. strigosus, another diploid species characterized by glandular inflorescence and round fruits. Hybrids between these two subspecies have been successful. Five parent cultivars dominate the red raspberry ancestry (Dale et al., 1993). These include ‘Lloyd George’ and ‘Pyne’s Royal’ derived from R. idaeus, ‘Newburgh’ from R. strigosus and ‘Preussen’ and ‘Cuthbert’ from crosses between R. idaeus and R. strigosus. Breeders have recently utilized other Idaeobatus species (i.e. R. occidentalis, R. cockburnianus Hemsl., R. biflorus Buch., R. kuntzeanus Hemsl., R. parvifolius Hemsl., and R. pungens oldhamii [Mig]. Maxim), two Cyclatis species (R. arcticus L. and R. stellatus Sm.) and one Anoplobatus species (R. odoratus L.) to develop new cultivars (Daubeny, 1996b). Rubus occidentalis is used as a source of resistance to postharvest root rot mainly caused by the two pathogens Botrytis cinerea Pers. Ex. Fr. and Amphorophora idaei

Borh, and to the large raspberry aphid, vector of red raspberry mosaic virus in Europe (Ellis et al., 1991). This species has also been used to enhance fruit firmness and late ripening in British and Scottish cultivars, ‘Malling Leo’, ‘Malling Joy’, ‘Glen Prosen’ and ‘Glen Moy’ (Keep, 1984). In North America, the primocane fruiting cultivars ‘Amity’ and ‘Summit’ released from the OSU-USDA breeding program also have the genes from R. occidentalis. Extreme winter hardiness and early primocane fruiting were derived from the arctic raspberry, R. arcticus and resulted in winter-hardy cultivars ‘Merva’ and ‘Heija’ (Hiirsalmi and Sako, 1976) and primocane fruiting types ‘Malling’ and ‘Autumn Bliss’ (Keep, 1984). Two Asiatic species, R. kuntzeanus Hemsl. and R. parvifolius Nutto were sources of low chilling requirement and were found in the ancestry of ‘Van Fleet’, ‘Southland’ and ‘Dormanred’. Modern breeding program objectives in raspberries include high yield of large, easily harvested fruits of good quality from spine-free or nearly spine-free plants. Other desirable traits include upright growth habits, primocane fruiting, strongly attached fruiting laterals and disease and pest resistance (Daubeny, 1996b).

Blackberries The first mention of blackberries was in 370 B.C. by the Greek writer Theophrastus who reported that the plant was used in hedges to keep out invading forces 2000 years ago (Jennings, 1988). Blackberries were domesticated in Europe by the seventeenth century and in North America during the nineteenth century (Jennings, 1988). As settlers moved west, forests were cut and cleared promoting various native species of blackberries to cross in a natural breeding program (Darrow, 1937). These wild selections which include ‘Aughinbaugh’, ‘Dorchester’, ‘Lawton’, ‘Eldorado’, and ‘Lucretia’ were released as superior cultivars in the 1850s (Jennings, 1988; Moore, 1984; Ourecky, 1975). The superiority of these cultivars presumably contributed greatly to increased blackberry cultivation. Present day species have arisen from the intercrossing of primary diploid species and occasional polyploidization events followed by intercrossing between new

6 polyploid species (Jennings, 1988). Ancestral diploids that could have given rise to this range of species may have been numerous but only six diploid species have survived in Europe, the Mediterranean regions and on the Canary Islands (Jennings, 1988). These do not carry the range of genetic variability found today not even when raspberry species are included. It is likely that a number of ancestral diploids did not survive. The first improved cultivar came from Judge J. H. Logan’s home garden experiment to combine characters of a pistillate form of a Western American blackberry R. ursinus ‘Aughinbaugh’ and a domesticated blackberry ‘Texas Early’ (Hall, 1990). The European red raspberry ‘Antwerp’ was planted nearby and an unplanned crossing event took place between this red raspberry cultivar and ‘Aughinbaugh’ that resulted in the cultivar ‘Loganberry’. Another hybrid from this accidental cross was ‘Mammoth’ or ‘Black Logan’ (Jennings, 1981). Other blackberry red raspberry hybrids followed like ‘Phenomenal’ from ‘Aughinbaugh’ and ‘Cuthbert’ red raspberry and ‘Youngberry’ from ‘Phenomenal’ and ‘Austin Mayes’. Modern breeding objectives are similar to those desirable in raspberries and include cultivars with higher yields, better fruits and pest and disease resistance.

Plant Genetic Resources of Rubus Most of the past fruit germplasm selections can be considered a biased sampling of the population (Zagaja, 1983). Collectors searched only for advantageous characteristics based on phenotypic expression, discriminating against undesirable traits in the process. The preference for specific characters led to the discarding of potentially important germplasm. In red raspberry, domestication in Europe and North America was based upon a few chance selections that formed the basis for the subsequent controlled crosses. Most red raspberry cultivars originated from two subspecies of diploid R. idaeus L.: R. idaeus and R. strigosus (Darrow, 1937) and also from their crosses. The subspecies intercross with little or no sterility in the resulting seedlings (Ourecky, 1975). Four cultivars (‘Cuthbert,’ ‘Latham,’ ‘Herbert’ and

‘Ranere’) provided the source of most of the germplasm of R. strigosus in present-day cultivars and no more than six individual wild genotypes are represented in these four cultivars (Daubeny and Anderson, 1989). Even Europe used only slightly more cultivars, and crosses between European and North American cultivars involved only a few parents with ‘Lloyd George’ predominating. Black raspberries are derived from R. occidentalis and also lack genetic diversity (Weber, 2003). The species is self- compatible and most cultivars lack heterozygosity and show little segregation and much inbreeding (Jennings, 1988). Blackberries suffer a similar genetic setback as only a few species were used for breeding and improvement of this crop. Most cultivated blackberries of European origin are related to or entirely derived from R. ulmifolius (=R. rusticanus), R. nitidioides and R. thrysiger (Hall, 1990). However, in the eastern USA, the species R. allegheniensis was involved in almost all released erect cultivars (Ourecky, 1975) while trailing cultivars involving R. baileyanus and its hybrids R. trivialis (=R. rubrisetus) and R. ursinus were important in the south and west USA. However, in spite of the extensive breeding program and wide use of locally collected accessions, the germplasm base represented in cultivars released is very small, with very few clones from the species used for breeding improvement that gave rise to most of the improved cultivars (Hall, 1990). Blackberry breeders still have not tapped a vast reservoir of genetically variable species and hybrids occurring in most parts of the world (Hall, 1990). The remarkably narrow genetic base in Rubus may have negative consequences in the future as plants are continuously threatened by environmental stresses. Germplasm with improved horticultural characteristics may have the most immediate value for commercial fruit production but much lesser value for the preservation of a wider genetic base (Zagaja, 1983). Wild relatives of cultivated fruits may be an extremely rich source of genetic variability that can contribute greatly to the improvement of important traits particularly regarding pest and disease resistance,

8 winter hardiness, drought, vigor of growth and productivity. Wild populations of the domesticated species and relatives are now being screened for useful genes. In Rubus there is increased preference for developing cultivars adapted to less than optimum environments including particularly low or high temperature, short ripening season, low winter chilling, drought, excessive wind and adverse soil conditions (Daubeny, 1996b). Breeders discovered that features required for adaptation to the environment are often associated with the undesirable “wild” traits. This emphasizes the need for preservation of botanical diversity through germplasm conservation and maintenance since botanical diversity is a finite resource. In addition there is the rapid disappearance of germplasm worldwide brought by urbanization, industrialization or slash and burn agriculture. The search for greater diversity and minimization of genetic uniformity are greatly advantageous for the continuous survival of very important crops. The initiation of the Rubus collection by the USDA-ARS in Corvallis, Oregon (Jahn, 1982) enabled the conservation and maintenance of virus- free clones of species worldwide. A collection of Rubus’ seeds of species or cultivars was assembled and clonal material is maintained in greenhouses, screen houses, field, and tissue culture and some is cryopreserved. Plants grown in the greenhouses, screen houses or in the field are utilized for data collection and are also available to researchers as a resource (Hummer, 1987; Hummer, 1988a; Hummer, 1988b). The plants in tissue culture and in cryogenic storage are maintained for back-up to the pot or field-grown plants (Gupta and Reed, 2006).

Rubus Conservation Approaches for conservation of plant genetic resources may be classified either as in situ or ex situ. The former involves the maintenance of genetic resources in their natural habitat of occurrence. The latter, however, involves conservation outside the native habitat protecting plants against dangers of destruction, replacement or deterioration. In situ rather ex situ conservation is preferable. However, due to the threat of genetic erosion in the original location and the need for easy access for

9 exploitation, ex situ conservation techniques are necessary for crops and other species (Hawkes et al., 2000). Ex situ techniques provide a safety back-up for in situ conservation methods wherein in situ conservation in a genetic reserve or farm cannot guarantee long-term security for the plants. Ex situ conservation methods include seed storage, field genebanks and botanical gardens. Collection involves gathering of samples of a species from populations in the field or natural habitats for conservation and subsequent use (Rao, 2004). Depending on the species’ breeding system, seeds or vegetative propagules may be collected.

Seed Storage Seed storage involves dehydration to low moisture contents and subsequent storage at low temperature for extended periods. Most agricultural crops produce orthodox seeds that can be dehydrated without the loss of their viability. For example, strawberry germplasm can be retained for extended periods of time by storage of their dry seeds in a cool temperature (40 °F) (Scott and Draper, 1970). Seeds were removed from the strawberries in a food blender and dried for about two weeks at room temperature on a filter paper before storage. The germination test performed revealed that germination is still possible after 23 years at 40 °F. In Rubus, seeds are also dried and stored at 4 to 5 °C in sealed containers (Clark, 1993). Long-term storage is a viable method of maintaining blackberry germplasm as observed in the high germination rates especially of the thorny populations but Rubus is also associated with dormancy problems. It has a complex dormancy mechanism frequently resulting in low seed germination causing problems in breeding programs (Ke et al., 1985). The hard endocarp of Rubus seeds is impermeable to water and air requiring scarification to facilitate germination. Traditionally, seed scarification is done using H2SO4 for 3 h in blackberry and 30 min to one h in raspberry followed by stratification for three to four months at 2 to 5 °C (Ourecky, 1975). Other scarification treatments employed are laborious like endocarp removal (Kerr, 1954), prolonged acid treatment (Moore et al., 1974), in vitro germination (Ke et al., 1985) and scarification

10 of freshly-harvested air-dried seeds (Dale and Jarvis, 1983) which all have unsatisfactory results. Seed storage is not applicable for crops in three categories: 1) vegetatively propagated plants that do not produce seeds like bananas and plantain; 2) sterile genotypes and genotypes that produce highly heterozygous orthodox seeds unfit for conservation like potato and sugarcane This is also true for cultivated species of plants such as Rubus that are not true to type from seeds; and 3) recalcitrant species of tropical origin with seeds that cannot be dried to low moisture levels to permit storage at low temperature (Engelmann, 2004). Therefore, genetic resources of all these species are conserved in field genebanks.

Field Genebanks A field genebank is the collection of seeds or living materials from one location and its transfer and planting at a second site (Hawkes et al., 2000). It is suitable for storage of recalcitrant species and provides easy access for characterization, evaluation and utilization of plants. However, field genebanks are highly at risk for destruction by natural calamities, pests and diseases (Engelmann, 2004). In Rubus, fungi and viruses cause many diseases making them prone to infection if they are conserved in field genebanks. The fungus Phytophthora fragariae var. rubi causes root-rot disease in raspberries. Cane botrytis, spur blight and cane spot also cause major losses in raspberries. In the Pacific Northwest, the Raspberry bushy dwarf virus (RBDV) is of much greater concern and reached epidemic proportions. RBDV is found in ‘Marion’ blackberry causing a substantial drop in the yield of infected plants (Finn and Knight, 2002). Viruses cause poor growth and low yields in raspberries and can even cause death after a few seasons. Field genebanks also involve large areas of land and there is a high requirement for labor costs and technical personnel. Rubus germplasm is commonly preserved in field genebanks as growing plants or as potted plants under insect-proof screens to reduce the risk of virus or virus-like disease.

In Vitro Methods Safety duplicates of living collections are established using alternate strategies for preservation like in vitro conservation options. The extreme diversity and large numbers of species associated with Rubus germplasm makes it difficult to maintain collection of these lines in either greenhouses or the field. In vitro methods may be more appropriate for a very diverse genus (Skirvin et al., 2005). In vitro techniques may be defined as those that utilize tissue culture methods in the maintenance, production or modification of plant material. It involves growth of tissue on sterile medium in sterile culture vessels (Ashmore, 1997). In vitro techniques for germplasm conservation include the micropropagation of shoots which can be modified to allow slow growth for medium-term storage. Long-term storage may involve cryopreservation that utilizes tissue-culture protocols in the storage of the frozen shoot tips. These techniques are highly space-efficient, minimize diseases and pest problems and allow manipulation and control of external variables that may threaten the plants (Turner et al., 2001). In addition, in vitro methods also offer a better alternative to be used for plants that are difficult to propagate using conventional techniques. In apples the most effective method for its mass clonal multiplication is through in vitro culture (Modgil et al., 2004). Almonds are also woody plants that are difficult to propagate using conventional techniques. In vitro culture offers the potential to increase the multiplication rates of elite genotypes as well as products of new and improved cultivars (Martins et al., 2004). Micropropagation of Rubus is valuable for propagation and producing virus- and disease-free plants. Most of the media used are a modification of the high salt formulation of Murashige and Skoog (McPheeters et al., 1990). Medium additives include various carbohydrate sources, vitamins, growth regulators, agar and other supplements. Culturing shoot tips in a modified MS medium with BA, GA3 and IBA followed by explanting into a liquid culture and transfer to an agar medium allowed for rapid shoot proliferation (Broome and Zimmerman, 1978). Tissue culture definitely offers a greater multiplication advantage compared to conventional

12 methods. Tissue culture is suitable for a wide variety of Rubus germplasm (Reed, 1990). The term “medium-term storage” encompasses slow-growth strategies ranging from temperature reduction or environmental manipulation to chemical additions in the growth medium (Reed and Chang, 1997). Reed (1993) upon evaluation of 250 accessions Rubus in the germplasm collection at NCGR, Corvallis showed that storage for 1-5 years is successful at 4 °C with 12 hours of light using plastic tissue-culture bags and growth medium without hormones. On the other hand, cold-sensitive genotypes of tropical species can be stored for 9 months in the growth room at 25 °C and 16-hour photoperiod.

Genetic Variation Like any other techniques, tissue culture techniques as well as cold storage have associated risks including loss of accession due to contamination, human error and somaclonal variation (mutations that occur spontaneously in tissue culture with increasing frequency with increasing subculturing) (Larkin and Scowcroft, 1981). The major problem of somaclonal variation among subclones of parental lines cause a direct consequence on in vitro culture of plants, cells, tissues and organs (Larkin and Scowcroft, 1981). These variations are most common in cell and callus culture but may also occur in micropropagated plants at a low level. The presence of variation is disadvantageous, especially when propagation of an elite tree is desired, where clonal fidelity is required to maintain the advantages of the desired elite genotypes (Modgil et al., 2004). Causes of variation include changes in the structure and/or chromosome number, noticeable point mutations, changes in the expression of a gene as a result of structural changes in the chromosome (heterochromatin and effects of position) or activation of transposable elements, chromatin loss, DNA amplification, somatic crossing over, somatic reduction and structural changes in the cytoplasmic organelle

DNA (Kaeppler et al., 2000). A number of factors can be identified that influence whether or not variation is produced and how much variation is generated (Karp, 1995). These factors are the degree of departure from organized meristematic growth, genetic constitution of the starting material, growth regulators in the medium and cultivation period. Growth in culture may occur from already established meristems or it may take a disorganized form as callus from which organized structures arise by somatic embryogenesis or organogenesis. Departure from organized growth is a key element in somaclonal variation. The greater the departure from organized growth and the longer time spent in this state, the greater is the chance of generating variation (Karp, 1995). This was proven in a study in barley wherein little somaclonal variation was observed when plants were derived from highly differentiated meristematic tissues (Bregitzer et al., 2002). Shoot meristematic cultures and modified embryogenic callus showed higher regeneration rates compared to standard embryogenic callus reflective of greater genomic stability. Plant genotype also has an important effect on regeneration and frequency of somaclonal variation. A genotypic effect on somaclonal variation was reported in embryogenic suspensions of Coffea arabica F1 hybrids (Etienne and Bertrand, 2001). In another study of Etienne and Bertrand (2003), also in embryogenic suspensions of Coffea arabica, somaclonal variation was observed in all the genotypes studied but differences existed in the frequency of variation. Hybrids belonging to the same family performed similarly with regards to variant frequency, confirming genotypic influence on somaclonal variation. A number of cases also attributed somaclonal variation to the hormones present in the medium. Addition of growth regulators to culture medium is known to have influence on the frequency of the karyotype alterations in the cell cultures (Bordallo et al., 2004). It is possible that growth regulators act as mutagens (Karp, 1995) with the auxin 2,4-D (2,4-Dichloro-phenoxyacetic acid) as the one responsible for much variation. In carrots, methylation levels increased with increasing 2,4-D (Kaeppler et

14 al., 2000). No variation was found in strawberries grown with increased concentrations of N6 benzyladenine (Kumar et al., 1999). Correlation between the culture duration and accumulation of variation was reported. The age of C. arabica embryogenic cell suspensions affected variant frequency. For all the genotypes studied, the frequency of variants increased exponentially with suspension age, indicating that, for true-to-type multiplication, it is essential to restrict embryogenic material multiplication times to less than six months (Etienne and Bertrand, 2003). Longer in vitro cultures also caused somaclonal variation in micropropagated tulips (Podwyszynska, 2005). The frequency of variation was less than 3.3% in progeny lines derived from cultures maintained in vitro for less than three years. However, maintenance from four years or longer resulted in changes that were also observed phenotypically. The fact that number of factors may cause somaclonal variation makes employment of conservation methods that preserve genetic fidelity a challenge. Any system that significantly reduces or eliminates tissue culture generated variability can be of much practical utility.

Cryopreservation Cryopreservation was developed to overcome the disadvantages as well as maximize the advantages offered by tissue culture and cold-storage techniques. This involves the storage of biological materials at ultra-low temperatures, usually in liquid nitrogen (-196 °C) (Engelmann, 2004). At this temperature cellular division as well as metabolic activities is stopped, hence plants can be stored without alteration or modification for a theoretically unlimited period of time. This method provides the potential for safe and cost-efficient, long-term germplasm conservation since materials are stored in small volumes protected from contamination and with little requirement for maintenance. It is especially useful for long-term conservation with the assumption that the genetic make-up of the plant is preserved. This method can be applied to a wide range of organisms and biological tissues (Benson et al., 1998). The choice of material for storage will depend on the plant species and include shoot

15 apices, pollen, somatic embryos, and seeds or excised zygotic embryos (Ashmore, 1997). The techniques currently used in cryopreservation aim to reduce the possibility of intracellular ice that causes irreversible cell damage. It includes the slow cooling technique based on freeze-induced dehydration (Kartha, 1985) as well as techniques based on vitrification (Sakai et al., 1990). Vitrification is the transition of an aqueous solution directly from liquid phase to an amorphous phase or glass, avoiding crystalline ice formation (Fahy et al., 1984). The two techniques differ in the way the plant material is prepared prior to immersion in liquid nitrogen and in the dependence upon a precisely controlled rate of cooling (Grout, 1995). The goal for successful cryopreservation is to maintain a level of integrated structure and function compatible with high viability and normal activity upon restoration to physiological temperature (Grout, 1995). Post-storage genetic stability is also an important consideration with the desire of cryoprotection and recovery methods not altering the genetic make-up of the plant. Cryoprotective techniques were improved to prevent two main causes involved in cryoinjury: ice crystal damage and dehydration injury. Traditional cryopreservation methods are usually dependent upon the short-term exposure of plant tissues to chemical cryoprotectants just prior to freezing that is achieved by applying controlled rate cooling regime in which temperature of cells is gradually reduced to an intermediate transfer temperature (usually within range of -30 to -80 °C) before final transfer to liquid nitrogen (Benson, 1998). At an optimum controlled rate of freezing, the cells lose just enough intracellular water before plunging into liquid nitrogen to vitrify on contact, so ice damage is eliminated. The use of cryoprotectants during the freezing process is one key to the viable preservation of plant materials. These cryoprotectants may be classified as non- penetrating or penetrating. Non-penetrating compounds include mannitol and sucrose that have a dehydrative effect on cells. Osmotic differentials cause water to withdraw from the cells resulting in less free water for intracellular ice crystallization or causing

16 delay or avoidance of crystallization (Finkle et al., 1984). On the other hand, penetrating compounds like dimethyl sulphoxide (DMSO) and glycerol exert a colligative action and may be considered as stabilizing solvents for the solute component of frozen cells as they prevent the damaging effects of cell dehydration and volume changes (Benson, 1998). These compounds are equally distributed in the extra- and intracellular compartments so when extracellular ice forms, water is equally lost in both compartments preventing the toxic concentration of the cell’s solute through dehydration. Damaging cell volume changes will be prevented and the high cryoprotectant content of the cell will depress the freezing point to a very low sub-zero temperature at which damage, if it does occur, can be tolerated (Meryman and Williams, 1984). Vitrification occurs when solute concentration of a biological system becomes so high that ice nucleation is prevented, then ice crystal formation and growth is inhibited by the highly viscous cytoplasm and water molecules form a glass (Benson, 1998). Several techniques like encapsulation-dehydration, vitrification, encapsulation vitrification, desiccation, pre-growth, pregrowth desiccation and droplet freezing involve direct vitrification. All the methods utilize a very rapid freezing process with samples plunged into liquid nitrogen once pretreatment is complete. The techniques differ in the pretreatment stages that include treatment with cryoprotectants, dehydration steps and encapsulation. Cryopreservation is in use as a backup to present methods of storage in Rubus at the USDA-ARS NCGR. In Rubus, shoot tip explants from tissue cultured plants were cryopreserved by slow-freezing (0.8 °C/min) to -40 °C and then rapidly to -196 °C in the presence of cryoprotectants (Reed and Lagerstedt, 1987). Initial screening of five accessions at the NCGR, Corvallis, showed that combination of polyethylene glycol, glucose and DMSO were the most successful cryoprotectant (Reed and Lagerstedt, 1987). The addition of one week of cold acclimation to the slow-freezing technique was more effective for in vitro grown Rubus meristems (Reed, 1988). Extending the cold acclimation period was more effective in increasing the regrowth

17 after storage (Chang and Reed, 1999). However, the combination of cold acclimation with ABA was found to improve survival of only one of five Rubus genotypes, R. cissoides (Reed, 1993a). Cryopreservation of Rubus shoot tips by vitrification in plant vitrification solution (PVS) 2 or encapsulation dehydration is also effective (Gupta and Reed, 2006).

Markers and Their Uses in Rubus Several markers exist and these include morphological and molecular (biochemical and DNA-based) markers. 1. Morphological Markers Traditionally, characterization of germplasm collections relied solely on the use of morphological descriptors (Fajardo et al., 2002) which consist of phenotypic traits like flower color and growth habit. Morphological analysis is the easiest and least complex technique for plant identification and characterization. It involves the description and visual monitoring of easily detectable plant characteristics like form and structure. However errors in scoring are common and may be attributed to environmental effects on morphological expression (Marinoni et al., 2003). Additional shortcomings of morphological traits as markers include changes in characteristics depending on the growth stage of the plant, insufficient variation and the length of time required for appearance of informative traits particularly in tree crops. Accurate identification becomes difficult in the process, lowering the reliability of morphological markers for germplasm characterization. 2. Molecular Markers Molecular markers provide additional tools for germplasm characterization and assessment of genetic relatedness and diversity in collections. Molecular methods include biochemical and DNA markers. Biochemical markers were introduced in the 1960s and involve protein and enzyme electrophoresis. These markers reveal differences between storage proteins or enzymes encoded by different alleles at one (allozymes) or more gene loci (isozymes) (Rao, 2004). The enzymes are of different

18 molecular forms but catalyze the same biochemical reactions. They are differently charged and can be separated by electrophoresis. Visualization is achieved by supplying the bands with the substrates and cofactors and observing the formation of colored products. The products represent protein products encoded by different alleles/genes and provide codominant markers. Isozyme analysis has been used for identification of cultivars (Bringhurst et al., 1981; Veasey et al., 2002; Weeden and Lamboy, 1985) and species (Buck and Bidlack, 1998). In Rubus, two-way paper chromatography was utilized to distinguish seven wild and cultivated raspberries according to their flavonoid spots and spotting patterns (Haskell and Garrie, 1966). Starch gel electrophoresis and isozyme staining were also effective in identification of raspberry cultivars (Cousineau and Donnelly, 1989a) and tissue-cultured raspberry shoots (Cousineau and Donnelly, 1989b). A total of 75 raspberry genotypes out of 104 (72%) were uniquely characterized using 6 isozymes (Cousineau and Donnelly, 1993). Unfortunately, despite the analysis of several enzymes, some Rubus cultivars remain undistinguishable. Like in black raspberries, the level of isozyme variation is too low for cultivar identification (Cousineau and Donnelly, 1992). This limits the value of isoenzyme analysis for fingerprinting mostly because of lack of or low level of variation in many cultivars and species (Fang et al., 1997). The use of additional powerful markers like DNA markers can circumvent the problem. DNA markers are unaffected by environmental conditions, organ specificity or growth stage and can detect single nucleotide changes. This access to fine scale genetic variation makes DNA-based methods the marker of choice for cultivar identification and characterization and for complementing traditional approaches. DNA markers are generally based either on the use of restriction enzymes that recognize and cut specific short sequences of DNA (Restriction Fragment Length Polymorphism, RFLP) or on the polymerase chain reaction (PCR) that involves DNA amplification of target sequences using short oligonucleotide primers. PCR-based techniques include Random Amplified Polymorphic DNA (RAPD), Amplified Fragment Length Polymorphism (AFLP), Internal Transcribed Spacer (ITS) region

19 and Simple Sequence Repeats (SSR) or microsatellites. The choice of markers depends on the study’s objectives, technical expertise and operational funds. DNA markers are used in many studies for cultivar identification (Becher et al., 2000; Guilford et al., 1997), species characterization (Ahmad and Southwick, 2003; Graham and McNicol, 1995) assessment of genetic variability (Jakse et al., 2001; Zhebentyayeva et al., 2003), evaluation of population structure (Aranzana et al., 2003) and breeding materials (Hernandez et al., 2003), detection of monogenic (Araujo et al., 2002) and quantitative trait loci (QTL) (Funatsuki et al., 2006) marker assisted selection (Yi et al., 2004) and sequence identification of useful candidate genes (Li and Garvin, 2003). Molecular methods are important tools for genebank management and have been used to develop future collecting strategies (van Treuren et al., 2001), to identify gaps in the collections (Carvalho and Schaal, 2001), to eliminate redundancies (Dean et al., 1999) and misidentified accessions (Dangl et al., 2001; Fossati et al., 2001; Khadari et al., 2003) and to validate core collections (Grenier et al., 2000). Restriction Fragment Length Polymorphism (RFLP) was the first DNA-based marker developed. RFLPs are detected as DNA fragments of different sizes generated after restriction endonuclease digestion and hybridization to a known DNA probe like cDNA clones or microsatellites (Staub and Serquen, 1996). DNA fragments are transferred by Southern blotting to a nitrocellulose or nylon membranes that are generally hybridized to a radioactively-labeled DNA probe. RFLPs are co-dominant markers and analysis of band profiles is straightforward. The use of multilocus probes was found to be effective in Rubus identification (Waugh et al., 1990). The variation in 20 diverse Rubus genotypes including red raspberries, blackberries, hybrid berries and wild relatives was investigated by hybridization of the chloroplast DNA sequence probes derived from Hordeum and Pisum. A phylogenetic tree was constructed to examine Rubus’ species relationship and genotypic differences were revealed that were not observed when morphological markers were used. Hybridization of a M13 bacteriophage probe to the DNA of 14 different Rubus plants revealed variable length

20 of repetitive sequences in the genome and demonstrated high genetic variability among the selected taxa (Nybom et al., 1990). This shows the ability of RFLP to reveal genetic differences among closely related Rubus’ species or taxa. RFLP was also utilized to resolve cases of cultivar misidentification like in the case of mixing ‘Lowden Sweet’ and ‘Royal’ (Nybom et al., 1989). The disadvantages of RFLPs include the requirement for large amounts of DNA and for radioactive labeling of the probes. PCR-based techniques in contrast require a small amount of DNA and do not require radioactive labeling. Random Amplified Polymorphic DNA (RAPD) markers and their variants are the most commonly used PCR-based marker type (Williams et al., 1990). The technique is based on the use of a single arbitrary primer, commonly a 10-mer or 20-mer, in a PCR reaction to amplify multiple copies of random genomic DNA sequences. RAPDs are relatively easy, require a modest laboratory set-up and are amenable for automation. Several studies in Rubus using RAPD markers had been reported. The first identification of raspberry cultivars utilized RAPDS (Parent et al., 1993). A total of 15 cultivars using 19 different RAPD primer combinations were assessed. The combination of three primer pairs was able to differentiate all the cultivars. Reproducible fingerprints were obtained from samples of one cultivar from different propagation methods, seasons and sources. Successful identification of another 10 red raspberry cultivars using 10 RAPD primers was also reported by Graham et al. (1994). RAPDs were also able to establish the genetic relationships among Rubus cultivars and differentiate closely related vegetatively propagated cultures (Graham and McNicol, 1995). RAPD markers were used on 13 different species (24 accessions) for the assessment of the degree of similarity between species from the important subgenera. The results were found to support the other findings concerning the origin of Rubus germplasm. Trople and Moore (1999) calculated the genetic similarities of 42 Rubus genotypes representing 37 species and found that ‘Tulameen’ and ‘Meeker’ were highly similar. Genetic relationship among and between black raspberry, red raspberry and a blackberry hybrid genotypes was also analyzed using RAPD markers

(Weber, 2003). Combined marker profiles from six RAPD primers were able to distinguish between the 16 black raspberry genotypes studied. Red raspberry and blackberry were distinguished from black raspberry by 27 and 29 of the 30 RAPD primers tested, respectively. Among the 16 black raspberry cultivars, an average of 81% similarity was calculated with a maximum similarity of 98% and a minimum of 70%. Average similarity between black raspberry and red raspberry was 41% and was 26% between black raspberry and blackberry. Differentiation of blackberry and raspberry cultivars for genetic identification by pedigree and RAPD analysis was also done by Stafne et al. (2003). Comparison of the clustering based in pedigree and RAPD data showed that the approximation of the relatedness in pedigree analysis can overestimate or underestimate percentage relationships. This may question the degree of certainty of the relationship showed by pedigree analysis. On the other hand, RAPDs proved to be a good method of assessing genetic relationships. As a result, pedigree and RAPD data do not correlate for the genotypes sharing many of the same parents because of the overestimation during pedigree analysis. Hence, extreme caution should be observed when using different analyses in genetic relationship determination as seen in the inconsistency of the results of the study. RAPD markers also have limitations like the irreproducibility of banding patterns preventing comparisons to be made between studies. In addition they are dominant markers and genetic homology of identical bands is not certain. Amplified Fragment Length Polymorphism (AFLP) is a DNA fingerprinting technique based on the amplification of subsets of genomic restriction fragments using PCR (Vos et al., 1995). The first step of the AFLP protocol involves digestion of the DNA with two restriction enzymes, a rare cutter like EcoRI and a frequent cutter like MseI. Double stranded (ds) adapters are ligated to ends of DNA fragments to generate template DNA for a two-step amplification process. Primers are designed to contain sequences that are complementary to those of the adapters and the restriction sites along with one (in step one) to three selective bases (in step two) added at their 3’ end. The use of selective bases allows amplification of only a subset of restriction

22 fragments. Polymorphisms are revealed after separating the amplified DNA fragments by electrophoresis on a sequencing gel, and visualized by silver staining, radioactive or fluorescent detection. A large number of bands is generated that facilitates the detection of polymorphisms (Gupta et al., 1999). AFLP reveals a high level of polymorphism and has a high marker index or diversity index (Russell et al., 1997). The high marker index or diversity index is a reflection of the efficiency of AFLPs to simultaneously analyze a large number of bands rather than the levels of polymorphism detected. The key feature of AFLP is its capacity for the simultaneous screening of different DNA regions that distributed randomly throughout the genome (Mueller and Wolfenbarger, 1999). AFLP fragments can be exploited as landmarks in genetic and physical maps with each fragment characterized by its size and the primers required for amplification bridging the gap between genetic and physical maps. AFLP was used in Rubus to demonstrate sexual recombination in R. armeniacus and R. bifrons (Kollmann et al., 2000) and to evaluate genetic diversity in R. alceifolius collected in different geographical locations (Amsellem et al., 2001a). Deviations from the common AFLP banding pattern observed in the pooled seed families of R. armeniacus and R. bifrons indicated sexual reproduction. The sexual offspring was observed to have new or missing bands for several AFLP primers (Kollmann et al., 2000). The use of AFLP also provided insights on the probable origin of R. alceifolius. Changes in genetic patterns were observed in the species from the native range of Southeast Asia and its areas of introduction like Indian Ocean and Australia. Changes in biology probably occurred as a result of the movement from one region to another. A Madagascar species is the probable ancestor of this Rubus and was confirmed by comparisons with the AFLP profiles of Indian Ocean and Australian species. Results showed that one genotype from each of the Indian Ocean and Australian populations tested were identical to the Madagascar species. This suggests that R. alceifolius may have been introduced in Madagascar initially and had moved to other regions afterwards.

The quantity of generated information, replicability, resolution, ease of use and cost efficiency of AFLP markers are at least as good, if not superior, to those of other standard molecular markers (Mueller and Wolfenbarger, 1999). Other advantages offered by AFLP are the following. The technique permits the detection of restriction fragments in any background or complexity including pooled DNA samples and cloned DNA segments. Therefore, AFLP can generate fingerprints of any DNA regardless of origin and complexity. It has a broad taxonomic scope hence, AFLP markers can be developed in any organism with DNA with no prior knowledge of the organism’s genomic make-up needed. AFLP is utilized not only in plants but in bacteria (Lin et al., 1996), fungi (Forche et al., 2000) and animals (Liu et al., 1998). The high stringency requirement results in low error levels for the method. It is also highly reproducible and highly efficient reflected by the high ratio of polymorphisms generated per experiment. The unlimited number of markers that can be generated by different primer combinations also results in its high resolution. In addition, just like the other PCR-based technique, it requires only a minimum amount of DNA. Internal Transcribed Spacer Region The internal transcribed spacer (ITS) region has been used in numerous systematic studies at the genera and species level of a wide array of plant taxa (Baldwin et al., 1995). Two internal spacers (ITS-1 and ITS-2) are located between genes encoding the 5.8s, 18s and 26s nuclear ribosomal RNA (nRNA) subunits (Baldwin, 1992). The ITS-1, ITS-2 and 5.8s nRNA make up the ITS region (Baldwin, 1992). The short length and highly conserved nature of the flanking ribosomal subunit genes makes ITS region easily amplified from small amounts of genomic DNA using PCR. Previous studies on Rubus involving nuclear ribosomal DNA ITS region sequences have focused at the species level (Alice, 2002; Alice and Campbell, 1999; Alice et al., 2001). Rubus ITS sequences were found to be informative among subgenera with low variability between closely related species (Alice and Campbell, 1999). Since the lineages of most blackberry cultivars involve several distinct species, discrimination among the cultivars based on the ITS sequences may be possible. However, ITS region sequences did not appear to

24 differentiate among closely related blackberry genotypes (Stafne and Szalanski, 2003). ITS sequence may distinguish between red raspberry and blackberry genotypes but probably not within the cultivars of thos groups unless they contain highly divergent ancestral species. Microsatellite Markers or Simple Sequence Repeats (SSR) are tandemly repeated motifs of one to six bases that are found in most prokaryotic and eukaryotic genomes analyzed to date (Zane et al., 2002). They are found in coding and non-coding regions and are highly polymorphic. Advantages of SSRs include their multi allelic nature, codominant transmission, reproducibility, ease of detection by PCR, relative abundance and extensive genome coverage (Powell et al., 1996). These markers are amenable for automation and are easily shared between labs as primer sequences providing a common language for collaborative research and acting as universal genetic mapping anchors (Powell et al., 1996). Polymorphism results mostly from either the gain or loss of repeat units (Schlotterer and Tautz, 1992). Two mutational mechanisms were proposed to explain the high rates of mutation: DNA polymerase slippage or recombination (Ellegren, 2004). The slippage model appears as the most probable cause of variability. During this event, DNA polymerase pauses during replication and dissociates from the DNA (Levinson and Gutman, 1987; Schlotterer and Tautz, 1992). On dissociation, the terminal portion of the newly synthesized strand may separate from the template and anneal to another repeat unit. As replication continues after misalignment, repeat units may be inserted or deleted relative to the template strand. The mismatch repair system of the DNA polymerase may correct the primary mutation and those that are not repaired end up as microsatellite mutation events. Thus, SSR reliability can represent a balance between the generation of replication errors by slip strand mispairing and the correction of some of these errors by exonucleolytic proofreading and mismatch repair (Li et al., 2002). Microsatellite-mutation may also be caused by recombination-like processes like cross-over or gene conversion. Cross-over is the reciprocal transfer of genetic

25 information while gene conversion is the non-reciprocal transfer of information which has recently emerged as the major cause of tandem repeat instability (Richard and Paques, 2000). Environmental conditions affect the efficiency of the two mutational mechanisms. Factors like repeated motif, allele size, chromosome position, GC content in flanking DNA, cell division (meiotic vs. mitotic), sex and genotype affect the mutation rate at the SSR loci (Li et al., 2002). The high information content (which is a feature of the number and frequency of alleles) detected and ease of genotyping increased the utility of SSRs (Powell et al., 1996). SSRs can distinguish between closely related individuals. This discrimination power is valuable for identification of plant species that have a narrow genetic base like that found in Rubus. When compared to RFLPs, AFLPs, and RAPDs, thirteen SSR markers resulted in 100% polymorphism in barley accessions as opposed to the other methods (Russell et al., 1997). Up to 54% of the fragments generated by AFLP analysis were monomorphic. In fruit trees, SSRs are considered the marker of choice for genetic fingerprinting purposes (Ahmad et al., 2004). They have been used for cultivar identification in many fruit crops including apple (Hokanson et al., 2001), Asian and European pear (Yamamoto et al., 2001), blueberry (Boches, 2005), kiwi fruit (Zhen et al., 2004), peach (Testolin et al., 2000), sweet cherry (Struss et al., 2003), grapes (Dangl et al., 2001; Fossati et al., 2001), apricot (Zhebentyayeva et al., 2003) and pistachio (Ahmad and Southwick, 2003). The need for de novo isolation of microsatellite sequences in species examined for the first time (Zane et al., 2002) constitutes the major drawback of microsatellite marker use. High costs are incurred in the process of de novo microsatellite development. Cross-transportability between species however may reduce the development costs (Gupta and Varshney, 2000). Cross-species transportability was widely reported in stone fruits (Cipriani et al., 1999; Zhebentyayeva et al., 2003), in pome fruits (Yamamoto et al., 2001) and in hazelnuts (Bassil et al., 2005). Ten peach microsatellite loci amplified in other Prunus species like plums, apricots, almonds, nectarine, sweet and sour cherry as well as in apple (Cipriani et al., 1999) while 12

26 peach SSRs amplified 14 polymorphic loci in apricot and were able to highly discriminate between 63 of 74 cultivars tested (Zhebentyayeva et al., 2003). SSRs isolated from apple were highly conserved in pear (Yamamoto et al., 2001) and used to distinguish 36 pear accessions. In Rubus, Graham et al. (2002b) developed the first ten SSR markers from the red raspberry ‘Glen Moy’. These markers were highly polymorphic in a group of 50 Rubus genotypes consisting of raspberries, blackberries and their hybrids. Eight additional SSR markers were identified from the invasive weed R. alceifolius in the Malachobatus subgenus (Amsellem et al., 2001a) and were used to study the reproductive biology of Rubus species from Thailand and Vietnam. The first genetic linkage map in Rubus was also constructed based on genomic and EST-SSRs as well as on AFLP (Graham et al., 2004). The genetic linkage map (789 cM) of the red raspberry had been constructed from a cross between two phenotypically different cultivars; the recent European cultivar, ‘Glen Moy’ and the older North American cultivar, ‘Latham’. ‘Latham’ was used in controlled crosses for its resistance to raspberry root rot but not in commercial cultivation because of its small fruit. ‘Glen Moy’ is highly susceptible to root rot and is spine-free (Jennings, 1988). The cross between these cultivars exhibited wide range of root rot resistance and other plant characteristics like structure and fruit quality. Information from the cross was utilized in the linkage map. Two or more genes were proposed to control spineness in the raspberries. Stafne (2005) used the primers developed by Graham et al. (2002), Amsellem et al. (2001), Lewers et al. (2005) (derived from Fragaria x ananassa Duch) and Rosa (derived from Genbank sequence) to differentiate progeny and assess genetic similarity within four segregating raspberry and blackberry populations. Four genotypes were tested with 142 unique SSR primers. A total of 60 primers were found useable for mapping raspberries and 45 for blackberries. Up to 32 of the primers may be useful for genetic mapping in both the blackberry populations and at least one of the raspberry populations. The low numbers generated indicated the need for developing additional primer pairs. A specific blackberry primer may be needed

27 too. In the case of the strawberry-derived SSR markers, a low success rate was observed in blackberries and raspberries suggesting poor cross-generic use of these primers. SSRs were also used in wild raspberries to examine gene flow (Graham and Smith, 2002). Genetic diversity appeared to increase as the extent of the sample area increased implying that genetic variation was spatially dependent in wild raspberry (Graham et al., 1997), indicating that gene flow is infrequent. SSRs can also be used to develop markers linked to pests or diseases like root rot (Graham and Smith, 2002).

Molecular Marker Use in Assessing Genetic Fidelity The high discriminatory power of DNA markers is valuable for characterizing and evaluating genetic integrity of conserved plants. DNA markers can be utilized with morphological markers to evaluate the reliability of the different storage methods especially that of cryopreservation, which is the only technique for long-term maintenance. Several studies showed that cryopreservation indeed maintains the genetic fidelity of plants. Cryopreserved Prunus was assayed using RAPD and AFLP (Helliot et al., 2002). Comparison of the DNA pattern of frozen and non-frozen materials did not show any polymorphism. This was also true for cryopreserved grape and kiwi, (Zhai et al., 2003) and peanut (Gagliardi et al., 2003) shoot tips using RAPD markers. The same analysis suggested that the cryopreservation method used preserved the genetic stability of cryopreserved Scots pine embryogenic cultures (Haggman et al., 1998). Genetic stability was also observed in the cryopreserved shoot tips of Solanum tuberosum using microsatellites (Harding and Benson, 2001). AFLP assay of tissue cultured, cold stored and cryopreserved shoot apices Anigozanthos viridis generated a total of 95 fragments for 3 primer pairs with no differences detected across treatments (Turner et al., 2001). The genetic fidelity of A. viridis was apparently maintained after 12 months of storage. In a study by Dixit, et al., (2003) on Dioscorea bulbifera L., RAPD fragments showed identical profiles between cryopreserved embryogenic tissues and in vitro maintained control plantlets indicating stability in the genomic primer region. Amplified fragment patterns of

28 plants regenerated from cryopreserved embryogenic tissues were identical to those of in vitro grown control plants for 9 of the 10 primers tested. One band was found to be variable within one cryopreserved plant out of the 4960 bands produced by the 10 primers indicating somaclonal variation. However, the extremely low frequency of variation (0.02%) detected should be considered a marginal finding as this minor variation may have been induced during the induction and maintenance of embryogenic tissues through repeated subculturing before cryogenic treatments (Dixit et al., 2003). RAPD analysis of non-frozen embryogenic tissue can further confirm this assumption. Somaclonal variation was also detected in some cryopreserved embryogenic cultures of white spruce two and 12 months after they were reestablished following cryopreservation but not in the corresponding regenerated trees using RAPD markers (De Verno et al., 1999). These results suggest that cryopreservation may not affect callus or embryogenic tissues and even if variation due to the in vitro culture process appears, it may not affect stability of regenerated plants.

Research Objectives Rubus is one of the important small fruits of the Pacific Northwest of America because of its popularity as a fresh fruit and as processed food products. Given the economic importance of the fruit, breeding programs were established to produce cultivars with excellent quality. Since the wide diversity of Rubus is the potential source of novel traits, the USDA Agricultural Research Service (ARS), National Clonal Germplasm Repository (NCGR) in Corvallis, Oregon collects, maintains, characterizes and distributes Rubus accessions from all over the world. Genetic characterization of the stored Rubus accessions can guide future collection strategies, evaluate genetic relationships among accessions and ensure the diversity of the core collection. Molecular markers like DNA markers are important tools that can help manage the Rubus collection.

The major objectives of the research were: 1. Construction of microsatellite-enriched libraries of ‘Meeker’ red raspberry and ‘Marion’ blackberry 2. Development of SSR markers for red raspberry and blackberry 3. Fingerprinting 48 accessions each of important red raspberry and blackberry cultivars using SSR markers 4. Evaluation of genetic stability of regrown cryopreserved Rubus meristems using SSR and AFLP markers

Fingerprinting of the Rubus accessions using SSR markers can characterize the accessions to ease cultivar identification and identification of redundancies in the collection. The assessment of the genetic stability of cryopreserved Rubus meristems using SSR and AFLP markers can establish the reliability of cryopreservation as a long-term storage technique.

Literature Cited

Ahmad, F. and S. Southwick. 2003. Identification of pistachio (Pistachia vera L.) nuts with microsatellite markers. J Amer Soc Hort Sci. 128: 898-903.

Ahmad, R., D. Potter, and S.M. Southwick. 2004. Genotyping of peach and nectarine cultivars with SSR and SRAP molecular markers. J Amer Soc Hort Sci. 129: 204-210.

Amsellem, L., C. Dutech, and N. Billotte. 2001. Isolation and characterization of polymorphic microsatellite loci in Rubus alceifolius Poir. (Rosaceae), an invasive weed in La Réunion island. Mol Ecol Notes. 1: 33-35.

Aranzana, M.J., J. Carbo, and P. Arus. 2003. Microsatellite variability in peach [Prunus persica (L.) Batsch.]: cultivar identification, marker mutation, pedigree inferences and population structure. Theor Appl Genet. 106: 1341- 1352.

Araujo, L.G., A.S. Prabhu, and M.C. Filippi. 2002. Identification of RAPD marker linked to blast resistance gene in a somaclone rice cultivar Araguaia. Fitopatol. Bras. 27: 181-185.

Ashmore, S.E. 1997. Status report on the development and application of in vitro techniques for the conservation and use of plant genetic resources. International Plant Genetic Resources Institute, Rome, Italy.

Baldwin, B.G. 1992. Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: an example from the Compositae. Mol Phylogenetics Evol. 1: 3-16.

Baldwin, B.G., M.J. Sanderson, J.M. Porter, M.F. Wojciechowski, C.S. Campbell, and M.J. Donoghue. 1995. The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Ann Missouri Bot Garden. 82: 247-277.

Bassil, N.V., R. Botta, and S.A. Mehlenbacher. 2005. Microsatellite markers in hazelnut: isolation, characterization, and cross-species amplification. J Amer Soc Hort Sci. 130: 543-549.

Becher, S.A., K. Steinmetz, K. Weising, S. Boury, D. Peltier, J.P. Renou, G. Kahl, and K. Wolff. 2000. Microsatellites for cultivar identification in Pelargonium. Theor Appl Genet. 101: 643-651.

Benson, E., P. Lynch, and G. Stacey. 1998. Advances in plant cryopreservation technology: current applications in crop plant biotechnology. Agbiotech News and Info. 10: 133-140.

Benson, E.E. 1998. Development of plant cryopreservation technology applications in agroforestry and forestry. In: Recent advances in biotechnology for tree conservation and management. IFS Workshop, Stokholm, Sweden.

Boches, P. 2005. Microsatellite marker development and molecular characterization in highbush blueberry (Vaccinium corymbosum L.) and Vaccinium species, Oregon State University, Corvallis.

Bordallo, P., D. Silva, J. Maria, C.Cruz, and E. Fontes. 2004. Somaclonal variation on in vitro callus culture potato cultivars. Hort Brasileira. 22: 300-304.

Bregitzer, P., S. Zhang, M. Cho, and P. Lemaux. 2002. Reduced somaclonal variation in barley is associated with culturing highly differentiated, meristematic tissues. Crop Sci. 42: 1303-1308.

Bringhurst, R.S., S. Arulsekar, J.F. Hancock, and V. Voth. 1981. Electrophoretic characterization of strawberry cultivars. J Amer Soc Hort Sci. 106: 684-687.

Broome, O.C. and R.H. Zimmerman. 1978. In vitro propagation of blackberry. HortScience. 13: 151-153.

Buck, G.W. and J. Bidlack. 1998. Identification of Quercus and Celtis species using morphological and electrophoretic data. Proc Okla Acad Sci. 78: 23-33.

Carvalho, L. and B. Schaal. 2001. Assessing genetic diversity in the cassava (Manihot esculenta Crantz) germplasm collection in Brazil using PCR-based markers. Euphytica. 120: 133-142.

Chang, Y. and B.M. Reed. 1999. Extended cold acclimation and recovery medium alteration improve regrowth of Rubus shoot tips following cryopreservation. CryoLetters. 20: 371-376.

Cipriani, G., G. Lot, W. Huang, M. Marrazzo, and E. Peterlunger. 1999. AC/GT and AG/CT microsatellite repeats in peach [Prunus persica (l) Batsch]: isolation, characterization and cross-species amplification in Prunus. Theor Appl Genet. 99: 65-72.

Clark, J.R. 1993. Longevity of Rubus seeds after long-term cold storage. HortScience. 28: 929-930.

Cousineau, J.C. and D.J. Donnelly. 1989a. Identification of raspberry cultivars by starch gel electrophoresis and isoenzyme staining. Acta Hort. 262: 259-265.

Cousineau, J.C. and D.J. Donnelly. 1989b. Identification of raspberry cultivars in vivo and in vitro using isoenzyme analysis. HortScience. 24: 490-492.

Cousineau, J.C. and D.J. Donnelly. 1992. Use of isoenzyme analysis to characterize raspberry cultivars and detect cultivar mislabeling. HortScience. 27: 1023- 1025.

Cousineau, J.C. and D.J. Donnelly. 1993. Characterization of red raspberry cultivars and selections using isoenzyme analysis. HortScience. 28: 1185-1186.

Dale, A. and B.C. Jarvis. 1983. Studies on germination of raspberry (Rubus idaeus L.). Crop Res. 23: 73-81.

Dale, A., P.P. Moore, R.J. McNicol, T.M. Sjulin, and L.A. Burmistrov. 1993. Genetic diversity of red raspberry varieties throughout the world. J Amer Soc Hort Sci. 118: 119-129.

Dangl, G., M. Mendum, B. Prins, M. Walker, C. Meredith, and C. Simon. 2001. Simple sequence repeat analysis of a clonally propagated species: a tool for managing a grape germplasm collection. Genome. 44: 432-438.

Darrow, G.M. 1937. Blackberry and Raspberry Improvement, p. 496-533. Yearbook, 1937.

Daubeny, H.A. 1996. Brambles. In: Fruit breeding. John Wiley and Sons Inc., New York.

Daubeny, H.A. and A. Anderson. 1989. Germplasm enhancement in the British Columbia raspberry breeding program. Acta Hort. 262: 61-64.

De Verno, L.L., Y.S. Park, J.M. Bonga, and J.D. Barrett. 1999. Somaclonal variation in cryopreserved embryogenic clones of white spruce [Picea glauca (Moench) Voss.]. Plant Cell Rpt. 18: 948-953.

Dean, R.E., J.A. Dahlberg, M.S. Hopkins, S.E. Mitchell, and S. Kresovich. 1999. Genetic redundancy and diversity among 'Orange' accessions in the US National Sorghum Collection as assessed with Simple Sequence Repeat (SSR) markers. Crop Sci. 39: 1215-1221.

Dixit, S., B. Mandal, S. Ahuja, and P. Srivastava. 2003. Genetic stability assessment of plants regenerated from cryopreserved embryogenic tissues of Dioscorea bulbifera L. using RAPD, biochemical and morphological analysis. CryoLetters. 24: 77-84.

Donnelly, D.J. and H.A. Daubeny. 1986. Tissue culture of Rubus species. Acta Hort. 183: 305-314.

Ellegren, H. 2004. Simple sequences with complex evolution. Nature Rev. 5: 435-445.

Ellis, M.A., R.H. Converse, R.N. Williams, and B. Williamson. 1991. Compendium of raspberry and blackberry diseases and insects. APS Press, St. Paul, MN.

Engelmann, F. 2004. Plant cryopreservation: progress and prospects. In Vitro Cell Dev Biol- Plant. 40: 427-433.

Etienne, H. and B. Bertrand. 2001. Trueness-to-type and agronomic characteristics of Coffea arabica trees micropropagated by the embryogenic cell suspension technique. Tree Physiol. 21: 1031-1038.

Etienne, H. and B. Bertrand. 2003. Somaclonal variation in Coffea arabica: effects of genotype and embryogenic cell suspension age on frequency and phenotype of variants. Tree Physiol. 23: 419-426.

Fahy, G.M., D.R. MacFarlane, C.A. Angell, and H.T. Meryman. 1984. Vitrification as an approach to cryopreservation. Cryobiology. 21: 407-426.

Fajardo, D., D. Bonte, and L. Jarret. 2002. Identifying and selecting for genetic diversity in Papua New Guinea sweet potato Ipomea batatas (L.) Lam. Germplasm collected as botanical seed. Genet Res Crop Evol. 49: 463-470.

Fang, D., M. Roose, R. Krueger, and C. Federicic. 1997. Fingerprinting trifoliate orange germplasm accessions with isozymes, RFLPs and inter-simple sequence repeat markers. Theor Appl Genet. 95: 211-219.

Finkle, B.J., M.E. Zavala, and J.M. Ulrich. 1984. Cryoprotective compounds in the viable freezing of plant tissues. In: Cryopreservation of plant cells and organs. CRC Press Inc, Boca Raton, Florida.

Finn, C. and V.H. Knight. 2002. What’s going on in the world of Rubus breeding? Acta Hort. 585: 31-38.

Forche, A., J. Xu, R. Vilgalys, and T. Mitchell. 2000. Development and characterization of a genetic linkage map of Cryptococcus neoformans var. neoformans using amplified fragment length polymorphisms and other markers. Fungal Genet Biol. 31: 189-203.

Fossati, T., M. Labra, S. Castiglione, O. Failaa, A. Scienza, and F. Sala. 2001. The use of AFLP and SSR molecular markers to decipher homonyms and synonyms in grapevine cultivars: the case of the varietal group known as “Schiave.” Theor Appl Genet. 102: 200-205.

Funatsuki, H., M. Ishimoto, H. Tsuji, K. Kawaguchi, M. Hajika, and K. Fujino. 2006. Simple sequence repeat markers linked to a major QTL controlling pod shattering in soybean. Plant Breeding. 125: 195-197.

Gagliardi, R., G. Pacheco, L. Carneiro, J. Valls, M. Vieira, and E. Mansur. 2003. Cryopreservation of Arachis species by vitrification of in vitro-grown shoot apices and genetic stability of recovered plants. CryoLetters. 24: 103-110.

Graham, J. and R.J. McNicol. 1995. An examination of the ability of RAPD markers to determine the relationships within and between Rubus species. Theor Appl Genet. 90: 1128-1132.

Graham, J. and K. Smith. 2002. DNA markers for use in raspberry breeding. Acta Hort. 585: 51-55.

Graham, J., K. Smith, K. MacKenzie, L. Jorgenson, C. Hackett, and W. Powell. 2004. The construction of a genetic linkage map of red raspberry (Rubus idaeus subsp. idaeus) based on AFLPs, genomic-SSR and EST-SSR markers. Theor Appl Genet. 109: 740-749.

Graham, J., G.R. Squire, B. Marshall, and R.E. Harrison. 1997. Spatially dependent genetic diversity within and between colonies of wild raspberry Rubus idaeus detected using RAPD markers. Mol Ecol. 6: 1001-1008.

Grenier, C., M. Deu, S. Kresovich, P.J. Bramel-Cox, and P. Hamon. 2000. Assessment of genetic diversity in three subsets constituted from the ICRISAT sorghum collection using random vs non-random sampling procedures B. Using molecular markers. Theor Appl Genet. 101: 197-202.

Grout, B.W. 1995. Introduction to the in vitro preservation of plant cells, tissues and organs. In: Genetic preservation of plant cells in vitro. Springer-Verlag, Heidelberg, NY.

Guilford, P., S. Prakash, J.M. Zhu, E. Rikkerink, S. Gardiner, H. Bassett, and R. Forster. 1997. Microsatellites in Malus x domestica (apple): abundance, polymorphism and cultivar identification. Theor Appl Genet. 94: 249-254.

Gupta, P., R. Varshney, P. Sharma, and B. Ramesh. 1999. Molecular markers and their applications in wheat breeding. Plant Breeding. 118: 369-390.

Gupta, P.K. and R.K. Varshney. 2000. The development and use of microsatellite markers for genetic analysis and plant breeding with emphasis on bread wheat. Euphytica. 113: 163-185.

Gupta, S. and B.M. Reed. 2006. Cryopreservation of shoot tips of blackberry and raspberry by encapsulation-dehydration and vitrification. CryoLetters. 27: 29- 42.

Haggman, H., L. Ryynanen, T. Aronen, and J. Krajnakova. 1998. Cryopreservation of embryogenic cultures of Scots pine. Plant Cell Tissue Organ Cult. 54: 45-53.

Hall, H.K. 1990. Blackberry Breeding. Plant Breeding Review. 8: 249-312.

Harding, K. and E.E. Benson. 2001. The use of microsatellite analysis in Solanum tuberosum L. in vitro plantlets derived from cryopreserved germplasm. CryoLetters. 22: 199-208.

Haskell, G. and J.B. Garrie. 1966. Fingerprinting raspberry cultivars by empirical paper chromatography. J Sci Food Agri. 17: 189-192.

Hawkes, J.G., N. Maxted, and B. Ford-Lloyd. 2000. In: The ex situ conservation of plant genetic resources. Kluwer Academic Publishers, USA.

Helliot, B., D. Madur, E. Dirlewanger, and M.T.D. Boucaud. 2002. Evaluation of Genetic Stability in Cryopreserved Prunus. In Vitro Cell Dev Biol -Plant. 38: 493-500.

Hernandez, P., G. Dorado, R.C. Ramirez, D.A. Laurie, J.W. Snape, and A. Martin. 2003. Development of cost-effective Hordeum chilense DNA markers: molecular aids for marker-assisted cereal breeding. Hereditas. 138: 54-58.

Hiirsalmi, H. and J. Sako. 1976. The nectar raspberry, Rubus idaeus x Rubus arcticus- a new cultivated plant. Acta Hort. 60: 151-157.

Hokanson, S., W. Lamboy, A. McFadden, and J. McFerson. 2001. Microsatellite (SSR) variation in a collection of Malus (apple) species and hybrids. Euphytica. 118: 281-294.

Hummer, K. 1987. New accessions at the repository. Newsletter. October 1987. National Clonal Germplasm Repository, Corvallis.

Hummer, K. 1988a. New accessions at the repository. Newsletter. April, 1988. National Clonal Germplasm Repository, Corvallis.

Hummer, K. 1988b. New accessions at the repository. Newsletter. November, 1988. National Clonal Germplasm Repository, Corvallis.

Jahn, O.L. 1982. Maintenance of clonal plant germplasm. HortScience. 17: frontspiece and cover.

Jakse, J., K. Kindlhofer, and B. Javornik. 2001. Assessment of genetic variation and differentiation of hop genotypes by microsatellite and AFLP markers. Genome. 44: 773-782.

Jennings, D.L. 1981. A hundred years of loganberries. Fruit Var J. 35: 34-37.

Jennings, D.L. 1988. In: Raspberries and Blackberries: Their Breeding, Diseases and Growth. Academic Press Inc., USA.

Kaeppler, S., H. Kaeppler, and Y. Rhee. 2000. Epigenetic aspects of somaclonal variation in plants. Plant Mol Biol. 43: 179-188.

Karp, A. 1995. Somaclonal variation as a tool for crop improvement. Euphytica. 85: 295-302.

Kartha, K.K. 1985. Meristem culture and germplasm preservation. In: Cryopreservation of plant cells and organs. CRC Press Inc., Boca Raton, Florida.

Ke, S., R.M.Skirvin, K.D. McPheeters, A.G. Otterbacher, and G. Galleta. 1985. In vitro germination and growth of Rubus seeds and embryos. HortScience. 20: 1047-1049.

Keep, E. 1984. Breeding Rubus and Ribes crops at East Malling. Sci Hort. 35: 54-71.

Kerr, A. 1954. Seed development in blackberries. Can J Bot. 32: 654-672.

Khadari, B., C. Breton, N. Moutier, J. Roger, G. Besnard, A. Beville, and F. Dosba. 2003. The use of molecular markers for germplasm management in a French olive collection. Theorl Appl Genet. 106: 521-529.

Kollmann, J., T. Steinger, and B.A. Roy. 2000. Evidence of sexuality in European Rubus (Rosaceae) species based on AFLP and allozyme analysis. Amer J Bot. 87: 1592-1598.

Kumar, M.K., R.E. Barker, and B.M. Reed. 1999. Morphological and molecular analysis of genetic stability in micropropagated Fragaria x ananassa cv. Pocahontas. In Vitro Cell Dev Biol- Plant. 35: 254-258.

Larkin, P.J. and W.R. Scowcroft. 1981. Somaclonal variation: a novel source of variability from cell cultures for plant improvement. Theor Appl Genet. 60: 197-214.

Levinson, G. and G.A. Gutman. 1987. High frequencies of short frameshifts in poly- CA/TG tandem repeats borne by bacteriophage M13 in Escherichia coli K-12. Nucl Acid Res. 15: 5323-5338.

Lewers, K.S., S.M. Styan, S.C. Hokanson, and N.V. Bassil. 2005. Strawberry genbank-derived and genomic simple sequence repeat (SSR) markers and their utility with strawberry, blackberry, and red and black raspberry. J Amer Soc Hort Sci. 130: 102-115.

Li, L. and D.F. Garvin. 2003. Molecular mapping of Or, a gene inducing beta-carotene accumulation in cauliflower (Brassica oleracea L. var. botrytis). Genome. 46: 588-594.

Li, Y., A. Korol, T. Fahima, A. Beiles, and E. Nevo. 2002. Microsatellites: genomic distribution, putative functions and mutational mechanisms: a review. Mol Ecol. 11: 2453-2465.

Lin, J., J. Kuo, and J. Ma. 1996. A PCR-based DNA fingerprinting technique:AFLP for molecular typing of bacteria. Nucl Acid Res. 24: 3649-3650.

Liu, Z., A. Nichols, P. Li, and R. Dunham. 1998. Inheritance and usefulness of AFLP markers in channel catfish (Ictalurus punctatus), blue catfish (I. furcatus), and their F1, F2, and backcross hybrids. Mol Gen Genet. 258: 260-268.

Lopez-Medina, J. and J.N. Moore. 1999. Chilling enhances cane elongation and flowering in primocane-fruiting blackberries. HortScience: 638-640.

Marinoni, D., A. Akkak, G. Bounous, K. Edwards, and R. Botta. 2003. Development and characterization of microsatellite markers in Castanea sativa (Mill.). Mol Breeding. 11: 127-136.

Martins, M., D. Sarmento, and M. Oliviera. 2004. Genetic stability of micropropagated almond plantlets, as assessed by RAPD and ISSR markers. Plant Cell Rpt. 23: 492-496.

McPheeters, K.D., R.M. Skirvin, and H.K. Hall. 1990. Brambles (Rubus spp.). In: Biotechnology in agriculture and forestry. Volume II: Crops II. vol. 2. Springer-Verlag, Berlin, Germany.

Meryman, H.T. and R.J. Williams. 1984. Basic principles of freezing injury to plant cells; natural tolerance and approaches to cryopreservation. In: Cryopreservation of plant cells and organs. CRC Press Inc., Boca Raton, Florida.

Modgil, M., K. Mahajan, S.K. Chakrabarti, D.R. Sharma, and R.C. Sobti. 2004. Molecular analysis of genetic stability in micropropagated apple rootstock MM106. Scientia Hort. 104: 151-160.

Moore, J.N. 1984. Blackberry Breeding. HortScience. 19: 183-185.

Moore, J.N., G.R. Brown, and C. Lundergan. 1974. Effect of duration of scarification on endocarp thickness and seedling emergence of blackberries. HortScience. 9: 204-205.

Moore, J.N. and R.M. Skirvin. 1990. Blackberry management. In: Small fruit crop management. Prentice Hall, Englewood Cliffs, NJ.

Mueller, U. and L. Wolfenbarger. 1999. AFLP genotyping and fingerprinting. Tree. 14: 389-394.

Nybom, H., S.H. Rogstad, and B.A. Schaal. 1990. Genetic variation detected by use of the M13 "DNA fingerprint" probe in Malus, Prunus, and Rubus (Rosaceae). Theor Appl Genet. 79: 153-156.

Nybom, H., B.A. Schaal, and S.H. Rogstad. 1989. DNA "fingerprints" can distinguish cultivars of blackberries and raspberries. Acta Hort. 262: 305-310.

Ourecky, D.K. 1975. Brambles. In: Advances in fruit breeding. Purdue University Press, West Lafayette, Indiana.

Parent, J., M.G. Fortin, and D. Page. 1993. Identification of raspberry cultivars by random amplified polymorphic DNA (RAPD) analysis. Can J Plant Sci. 73: 1115-1122.

Podwyszynska, M. 2005. Somaclonal variation in micropropagated tulips based on phenotype observation. J Fruit Ornamental Plant Res. 13: 109-122.

Powell, W., G. Machray, and J. Provan. 1996. Polymorphism revealed by simple sequence repeats. Trends Plant Sci. 1: 215-222.

Rao, N.K. 2004. Plant Genetic Resources: Advancing conservation and use through biotechnology. African J Biotechnol. 3: 136-145.

Reed, B.M. 1988. Cold acclimation as a method to improve survival of cryopreserved Rubus meristems. CryoLetters. 9: 166-171.

Reed, B.M. 1990. Multiplication of Rubus germplasm in vitro: A screen of 256 accessions. Fruit Var J. 44: 141-148.

Reed, B.M. 1993. Improved survival of in vitro-stored Rubus germplasm. J Amer Soc Hort Sci. 11: 890-895.

Reed, B.M. and Y. Chang. 1997. Medium- and long-term storage of in vitro cultures of temperate fruit and nut crop. In: Conservation of plant genetic resources in vitro. Volume 1: General Aspect. Science Publishers, USA.

Reed, B.M. and H.B. Lagerstedt. 1987. Freeze preservation of apical meristems of Rubus in liquid nitrogen. HortScience. 22: 302-303.

Richard, G. and F. Paques. 2000. Mini- and microsatellite expansions: the recombination connection. EMBO Reports. 11: 122-126.

Russell, J., J. Fuller, M. Macaulay, B. Hatz, A. Jahoor, W. Powell, and R. Waugh. 1997. Direct comparison of levels of genetic variation among barley accessions detected by RFLPs, AFLPs, SSRs and RAPDs. Theor Appl Genet. 95: 714- 722.

Sakai, A., S. Kobayashi, and I. Oiyama. 1990. Cryopreservation of nucellar cells of navel orange (Citrus sinensis Osb var. brasiliensis Tanaka) by vitrification. Plant Cell Rpt. 9: 30-33.

Schlotterer, C. and D. Tautz. 1992. Slippage synthesis of simple sequence DNA. Nucl Acid Res. 20: 211-215.

Scott, D.H. and A.D. Draper. 1970. A further notice on longevity of strawberry seed in cold storage. HortScience. 5: 439.

Skirvin, R.M., S. Motoike, M. Coyner, and M. Norton. 2005. Rubus spp. cane fruit. In: Biotechnology of fruit and nut crops. Cabi Publishing, Cambridge, MA.

Stafne, E.T. and A.L. Szalanski. 2003. Nuclear ribosomal ITS region sequences for differentiation of Rubus genotypes. J Arkansas Acad Sci. 57: 176-180.

Staub, J. and F. Serquen. 1996. Genetic Markers, Map construction, and their application in plant breeding. HortScience. 31: 729-741.

Struss, D., R. Ahmad, and S. Southwick. 2003. Analysis of sweet cherry (Prunus avium L.) cultivars using SSR and AFLP markers. J Amer Soc Hort Sci. 128: 904-909.

Testolin, R., M. Marrazzo, G. Cipriani, R. Quarta, I. Verde, M. Dettori, M. Pancaldi, and S. Ansavini. 2000. Microsatellite DNA in peach (Prunus persica L. Batsch) and its use in fingerprinting and testing the genetic origin of cultivars. Genome. 43: 512-520.

Thompson, M.M. 1995. Chromosome numbers of Rubus species at the National Clonal Germplasm Repository. HortScience. 301: 1447-1452.

Thompson, M.M. 1997. Survey of chromosome number in Rubus (Rosaceae: Rosoideae). Ann Missouri Bot Garden. 84: 128-164.

Turner, S., S.L. Krauss, E. Bunn, T. Senaratna, K. Dixon, B. Tan, and D. Touchdell. 2001. Genetic fidelity and viability of Anigozanthos viridis following tissue culture, cold storage and cryopreservation. Plant Sci. 161: 1099-1106.

USDA, ARS, National Genetic Resources Program. 2006. Germplasm Resources Information Network- (GRIN)[Online Database]. National Germplasm Resources Laboratory, Beltsville, Md, URL: . van Treuren, R., L.J.M.v. Soest, and T.H.J.L.v. Hintum. 2001. Marker-assisted rationalization of genetis resources collections: a case study in flax using AFLPs. Theor Appl Genet. 104: 144-152.

Veasey, E., R. Vencovsky, P. Martins, and G. Bandel. 2002. Germplasm characterization of Sesbania accessions based on isozyme analyses. Genet Resource Crop Evol. 49: 449-462.

Vos, P., R. Hogers, M. Bleeker, M. Reijans, T.V.d. Lee, M. Hornes, A. Frijters, J. Pot, J. Peleman, M. Kuiper, and M. Zabeau. 1995. AFLP: a new technique for DNA fingerprinting. Nucl Acid Res. 23: 4407-4414.

Waugh, R., M.v.d. Ven, S. Millam, R. Brennan, and W. Powell. 1990. The potential use of restriction fragment length polymorphism in Rubus breeding. Acta Hort. 280: 541-545.

Weber, C.A. 2003. Genetic diversity in black raspberry detected by RAPD markers. HortScience. 38: 269-272.

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

Williams, J., A. Kubelik, K. Livak, J. Rafalski, and S. Tingey. 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucl Acid Res. 18: 6531-6535.

Yamamoto, T., T. Kimura, Y. Sawamura, K. Kotobuki, Y. Ban, T. Hayashi, and N. Matsuta. 2001. SSRs isolated from apple can identify polymorphism and genetic diversity in pear. Theor Appl Genet. 102: 865-870.

Yi, G., S.K. Lee, Y.K. Hong, T.Y.C. Cho, M.H. Nam, S.C. Kim, S.S. Han, G.L. Wang, T.R. Hahn, P.C. Ronald, and J.S. Jeon. 2004. Use of Pi5 (t) markers in marker-assisted selection to screen for cultivars with resistance to Magnaporthe grisea. Theor Appl Genet. 109: 978-985.

Zagaja, S.W. 1983. In: Methods in fruit breeding. Purdue University Press, West Lafayette, Indiana.

Zane, L., L. Bargelloni, and T. Patarnello. 2002. Strategies for microsatellite isolation: a review. Mol Ecol. 11: 1-16.

Zhai, Z., Y. Wu, F. Engelmann, R. Chen, and Y. Zhao. 2003. Genetic stability assessments of plantlets regenerated from cryopreserved in vitro cultured grape and kiwi shoot tips using RAPD. CryoLetters. 24: 315-322.

Zhebentyayeva, T., G. Reighard, V. Borina, and A. Abbott. 2003. Simple sequence repeat analysis for assessment of genetic variability in apricot germplasm. Theor Appl Genet. 106: 435-444.

Zhen, Y., Z. Li, and H. Huang. 2004. Molecular characterization of kiwifruit (Actinidia) cultivars and selections using SSR markers. J Amer Soc Hort Sci. 129: 374-382.

CHAPTER 2

MICROSATELLITES FOR CULTIVAR IDENTIFICATION IN RASPBERRIES AND BLACKBERRIES

N. R. F. Castillo, B. M. Reed, N. V. Bassil

Abstract Twelve microsatellites from Rubus were isolated and characterized for genetic fingerprinting and differentiation of 48 cultivars each of raspberries and blackberries from the collections of the National Clonal Germplasm Repository, Corvallis, Oregon. Small-insert libraries from the raspberry, ‘Meeker’ (R. idaeus) and blackberry ‘Marion’ (R. hybrid) were enriched for microsatellites. Of the 74 fragments sequenced from the ‘Meeker’ library, 70 contained microsatellite motifs and 78 from the 90 fragments sequenced from ‘Marion’. Four primer pairs from the ‘Meeker’ library and 8 primer pairs from the ‘Marion’ library amplified clearly and produced interpretable PCR products during the initial test on 12 each of the raspberry and blackberry cultivars. The twelve primer pairs developed from Rubus and one primer pair from R. idaeus Genbank sequence were used in the fingerprinting of all the raspberry and blackberry genotypes. Primer pair RhM031 failed in raspberries while primer pair RiG001 failed in blackberries. In raspberries, the 12 SSRs generated 3 to 16 alleles per locus, with an average of 8 and a total of 96 alleles. In blackberries, the 12 primer pairs amplified 6 to 31 alleles per locus, with an average of 15 and a total of 177 alleles. NJ cluster analysis was used to describe genetic variation and relatedness among the 96 accessions. The NJ dendrogram revealed three major clusters corresponding to the raspberries, the blackberries and an intermediate heterogeneous group that consisted of cultivars that contained unusual Rubus species in their ancestry while two hybrid berries failed to join any group. The raspberry and blackberry clades were further divided on the basis of their founding clones or pedigree. The SSR markers developed in Rubus were highly polymorphic and were able to distinguish most of the raspberries and blackberries from the unique DNA fingerprints generated.

Key words: Rubus, genetic fingerprinting, alleles, NJ cluster, pedigree

Introduction The genus Rubus L. belongs to the Rosaceae family (Jennings, 1988) and contains 13 subgenera (USDA, 2006). Most of the cultivated fruit species in Rubus belong to two subgenera: the Idaeobatus (raspberries) and the Rubus formerly Eubatus (blackberries). Idaeobatus has a northerly distribution principally in Asia and in east and South Africa, Europe and North America (Graham and McNicol, 1995). This subgenus contains approximately 200 species which include the European red raspberry (R. idaeus L.), the North American red raspberry (R. strigosus Michx.), the black raspberry (R. occidentalis L.) and the purple raspberries (R. neglectus Peck). The subgenus Rubus is found worldwide except in desert regions but it is most prevalent in temperate regions of the northern hemisphere (Moore and Skirvin, 1990). Raspberries and blackberries grow best and provide greater economic returns when cultivated in areas with mild winters and long and dry summers. The primary regions of production in North America include the Pacific Northwest (Oregon, Washington and British Columbia) and California although there are some large plantings in parts of Texas and Arkansas, as well as in temperate climates surrounding lakes like New York, Michigan, Pennsylvania and Ohio (Skirvin et al., 2005). In the Pacific Northwest, raspberries and blackberries are important temperate small fruit crops where they are sold as fresh fruit and in a number of processed items like juice, yogurt and desserts (Donnelly and Daubeny, 1986). In raspberries, early domestication was based upon a few chance selections that formed the basis of the current controlled crosses (Daubeny and Anderson, 1989). Self-compatibility and lack of genetic diversity in some cultivated species also contributed to the narrowness of the genetic base. Important present-day red raspberry cultivars are primarily derived from R. idaeus, R. strigosus and also from their crosses. Several cultivars include the black raspberry (R. occidentalis L.), ‘Cumberland’ in their ancestry. Additionally, R. arcticus L., R. chamaeomorus L., R. cockburnianus

Hemsl., R. kunzeanus Hemsl., R. odoratus L. and R. parvifolius are in the pedigree of several red raspberry cultivars released since 1960 (Dale et al., 1993). However, only five parent cultivars dominated the ancestry of red raspberries: ‘Lloyd George’ and ‘Pyne’s Royal’ of R. idaeus, ‘Newburgh’ of R. strigosus, ‘Preussen’ and ‘Cuthbert’ of R. idaeus and R. strigosus crosses (Daubeny, 1996a). In blackberries, breeders utilized only few species in breeding improvement and still have not tapped a vast reservoir of genetically variable species and hybrids (Hall, 1990). For example, the Arkansas breeding program relied heavily on genotypes derived from R. allegheniensis Porter., R. argutus Link. and R. frondosus Bigel. ‘Darrow’, ‘Brazos’ and ‘Thornfree’ were the cultivars widely used as parents at the beginning of the blackberry breeding program (Stafne and Clark, 2003). The wide diversity of the genus is a potential source of novel traits. Unfortunately, cultivated Rubus has a narrow genetic base since its early domestication resulted in the reduction of its morphological and genetic diversity (Jennings, 1988) with the modern cultivars being genetically similar (Dale et al., 1989; Graham and McNicol, 1995). Up to 90% of red raspberry cultivars have ‘Lloyd George’ in their pedigree (Dale et al., 1989, 1993). Limited genetic variability in cultivated Rubus encourages inbreeding with losses of vigor, yield, sources of disease resistance and fruit size. Phenotypic plasticity can result from narrow genetic diversity in Rubus and is reflected in erroneous cultivar identification based solely on morphological traits. Identification also depends on fruit and flower characteristics that are not always present. Improvement in cultivar identification methods is therefore a high priority in Rubus. The National Clonal Germplasm Repository in Corvallis maintains 2094 Rubus accessions from 193 species and 64 countries. This collection is a great asset to breeders and researchers interested in conserving and increasing genetic diversity in cultivated raspberries and blackberries (Jahn, 1982). Efficient germplasm management requires species characterization and genetic diversity assessment using morphological traits and molecular marker analyses. Since the accurate

48 characterization of Rubus germplasm using morphological markers is difficult, molecular markers provide more reliable means for cultivar identification and to assess genetic relatedness and diversity in these collections. Simple Sequence Repeats (SSRs) are tandemly repeated motifs of one to six bases found in most prokaryotic and eukaryotic genomes analyzed to date (Zane et al., 2002). SSRs are found in coding and non-coding regions but are more abundant in the latter, and are usually characterized by high levels of polymorphisms. Advantages of SSRs as molecular markers include multiallelism, codominant mode of transmission, ease of detection by PCR, relative abundance, extensive genome coverage and the requirement for a small amount of starting DNA (Powell et al., 1996). SSR markers are easy to share between labs as primer sequences thus providing a common language for collaborative research and are more polymorphic than other PCR-based markers (Russell et al., 1997). SSRs are powerful genetic markers that can distinguish between closely related individuals, a particularly useful property in crops that have a narrow genetic base like Rubus. Microsatellite markers are not available in blackberry but a limited number of SSRs were isolated from the invasive weed R. alceifolius Poir. (Amsellem et al., 2001a), an Azorean island endemic species R. hochstetterorum Seub., (Lopes et al., 2006) and in red raspberry (Graham et al., 2004; Graham et al., 2002). Eight microsatellite markers were first isolated from R. alceifolius (Amsellem et al., 2001a). These eight SSRs were found to detect sexuality in a half-sib population of this simple-leafed bramble in its native range in Vietnam (Amsellem et al., 2001a) and to indicate apomixes in its area of introduction in the Indian Ocean islands of Madagascar and La Reunion (Amsellem et al., 2001b). Cross-species amplification of 4 of these 8 SSRs on 14 other tropical Rubus species was also demonstrated (Amsellem et al., 2001a). Fifteen microsatellite markers were recently identified from a genomic library of R. hochstetterorum (Lopes et al., 2006). Up to 100 genomic and 15 EST-SSR primer pairs were also isolated from the red raspberry cultivar ‘Glen Moy’ (Graham et al., 2002). Ten genomic SSRs were used to examine 50 Rubus

49 genotypes including European and North American red raspberry cultivars (R. idaeus), black raspberry (R. occidentalis), purple raspberry (R. idaeus x R. occidentalis), blackberries (R. fructicosus L.) and hybrids (Graham et al., 2002). These 10 SSRs were polymorphic and cross-amplified in the species evaluated. Thirty genomic SSRs and four EST-SSRs were placed on a genetic linkage map (789 cM) of the red raspberry constructed from a cross between ‘Glen Moy’ and ‘Latham’ (Graham et al., 2004). Stafne and Clark (2005) evaluated available Rubus SSR markers (8 from R. alceifolius and 84 from R. idaeus) for amplification and polymorphism in the parental genotypes of two existing mapping populations of raspberry (NY 322 x ‘Jewel’) and blackberry (‘Prime-Jim’ x ‘Arapaho’). Only two out of eight R. alceifolius markers amplified a product in the four Rubus genotypes examined. However, 51 ‘Glen Moy’- derived SSRs were useful for mapping in red raspberry while a smaller number, 24, was useful for mapping in blackberries. These results illustrate the need to develop additional SSR markers for mapping in raspberries and blackberries. It also suggests the need for blackberry-specific microsatellite primer pairs. In this study, we developed four microsatellite markers from a genomic library of ‘Meeker’ raspberry and eight SSRs from a library of ‘Marion’ blackberry. These 12 SSRs and one marker identified from a R. idaeus GenBank sequence were used to identify 48 cultivars each of raspberry and blackberry.

Materials and Methods Plant Materials DNA was extracted from young expanding leaves of 48 raspberry (Table 2.1) and 48 blackberry cultivars (Table 2.2) using a modified protocol based on the PUREGENE® kit (Gentra Systems Inc., MN). Modifications included an RNAse A digestion, a Proteinase K digestion and two protein precipitation steps. The detailed protocol is described in Appendix A. Construction of Microsatellite-enriched Libraries Two DNA libraries were constructed. In raspberry, DNA was extracted from ‘Meeker’ while ‘Marion’ was the

50 source of DNA in blackberry. A phenol: chloroform step was added to the modified PUREGENE (Gentra Systems Inc., MN) DNA extraction protocol described above. Microsatellite enrichment followed the protocol of Glenn and Schable (2005) (Appendix B). Total genomic DNA (100-200 ng) was digested with RSAI (New England Biolabs, MA) at 37 °C for 1 h and blunt-end ligated to phosphorylated double stranded superSNX linkers (Qiagen Inc., CA) with T4 DNA ligase (New England Biolabs, MA). Linker-ligated DNA was amplified by PCR with the single stranded SNX forward primer. Ten μL of the PCR product was added to a cocktail of 6X SSC [0.9 M sodium chloride, 0.09 M sodium citrate], 0.1 μM biotinylated mixture of microsatellite oligonucleotides [(AG)12, (AAC)6, (AAG)8, (AAT)12, (ACT)12, (ATC)8, and (TG)12] and 50 µL washed streptavidin-coated magnetic beads (Dynal, Oslo, Norway). The mixture was hybridized in an MJ thermocycler (MJ Research, Inc., MA) programmed for 5 min at 95 °C, 5 min at 70 °C, 50 touchdown cycles decreasing by 0.2 °C every 10 seconds, 50 °C for 10 min, 50 touchdown cycles decreasing by 0.5 °C every 5 s, and 1 h at 50 °C. The beads were then captured with a magnet and subjected to the following washes: twice in 400 μL 2X SSC, twice in 400 μL 1X SSC and twice in 400 μL 1X SSC. For the last 1X SSC wash, the beads were heated at 50 °C for 10 min in a water bath. The beads were then rinsed once in 200 μL TLE [10 mM Tris, 0.1 mM EDTA] and resuspended in 25 μL TLE. A second round of hybridization-enrichment for the microsatellites was performed to increase the amount of DNA using the following protocol. Ten μL of the PCR product recovered from the dynabeads on the first enrichment was added to a cocktail of 6X SSC [0.9 M sodium chloride, 0.09 M sodium citrate], 0.1 μM biotinylated mixture of microsatellite oligonucleotides [(AG)12, (AAC)6, (AAG)8, (AAT)12, (ACT)12, (ATC)8, and (TG)12] and 25 µL aliquot of washed streptavidin-coated magnetic beads (Dynal, Oslo, Norway) set aside during the first enrichment. The mixture was hybridized in an MJ thermocycler (MJ Research, Inc., MA) programmed for 5 min at 95 °C, 5 min at 70 °C, 50 touchdown cycles decreasing by 0.2 °C every 10 s, 50 °C for 10 min, 50

51 touchdown cycles decreasing by 0.5 °C every 5 s, and 1 h at 50 °C. The beads were then captured with a magnet and subjected to the following washes: twice in 400 μL 2X SSC, twice in 400 μL 1X SSC, and twice in 400 μL 1X SSC. For the last 1X SSC wash, the beads were heated at 50 °C for 10 min in a water bath. The beads were then rinsed once in 200 μL TLE [10 mM Tris, 0.1 mM EDTA] and resuspended in 25 μL TLE. The resuspended beads (2 μL) were used as PCR template with the SNX forward primer. PCR products were ligated into the pCR®4-TOPO® vector and inserted into TOP10 E. coli cells using the TOPO TA Cloning® kit version K (Invitrogen, CA). Colony PCR was used to determine insert size. DNA was purified from colonies containing 500-1000 bp inserts with the Perfectprep Plasmid 96 Spin DB kit (Eppendorf, Hamburg, Germany). Inserts were sequenced on a CEQ 8000 genetic analyzer (Beckman Coulter, CA) using primers designed for use with the CEQ 8000 (TopoF: 5’ -CGCCAAGCTCAGAATTAACCCTCAC-3’ and TopoR: 5’- CGACGGCCAGTGAATTGTAATACG-3’). The Genome Lab Methods Development Kit (MDK) (Beckman Coulter, Inc., CA) was used for DNA sequencing. This kit is recommended for difficult template DNA that cannot be sequenced using dITP-based sequencing chemistries like we encountered in Rubus. SSR Detection and Primer Design Sequences from the genomic library were screened for the presence of microsatellites with the SSRIT software (Temnykh et al., 2001) and aligned using CAP 3 software (Huang and Madan, 1999). Primers were designed for sequences that contained a minimum of seven dinucleotide repeats and five trinucleotide repeats. Primer 3 software (Rozen and Skaletsky, 2000) and PBC Public (http://hornbill.cspp.latrobe.edu.au/cgi-binpub/ssrprimer/indexssr.pl) were used for primer design. The parameters consisted of an optimum primer length of 22, optimum temperature of 60 ˚C and optimum GC content of 50%. Primers were obtained from Operon technologies (Qiagen Inc., CA). Primer Testing and PCR Reactions Primers were screened for amplification and polymorphism in 12 raspberry and 12 blackberry cultivars (Table 2.3). PCR reactions

were performed in 10 µL volume and contained 1X reaction buffer, 2 mM MgCl2, 0.2 mM dNTPs, 0.3 µM of each primer, 0.025 U of Biolase Taq DNA polymerase (Bioline USA Inc., MA) and 2.5 ng genomic DNA. DNA was amplified for 35 cycles in an Eppendorf gradient thermocycler (Brinkman Instruments Inc., NY) or an MJ Research Tetrad thermocycler (MJ Research Inc., MA) programmed for a 40 s denaturation step at 94 ˚C, 40 s annealing step at the optimum annealing temperature of the primer pair and 40 s extension step at 72 ˚C. The optimum annealing temperature for each primer pair was determined by gradient PCR from 55 ˚C to 65 ˚C using an equimolar mix of the DNA templates used for screening. PCR products were separated on a 3% agarose gel, stained with ethidium bromide and visualized on a UV transilluminator using a Bio-Rad GelDoc 2000 digital imaging system (Bio-Rad laboratories, CA). Primer pairs were scored for amplification of a product within 100 base pairs of the expected size in raspberries and blackberries and for polymorphism within each crop type.

Table 2.1. The 48 raspberry accessions analyzed with 12 SSR loci, USDA-ARS Plant Introduction numbers (P. I. no.), cultivar name or taxon for species, pedigree or origin and taxon are included.

P. I. no. Name Pedigree or origin Taxon 553495 Amber Taylor x Cuthbert R. idaeus 553487 Amity OSC 1839 (Fallred x OSC 1347) x OSC 1835 (Malling 791/45 x Heritage) R. idaeus 2323 Anne Amity x Glenn Garry R. idaeus 553325 Autumn Bliss Complex of R. strigosus, R. arcticus and R. occidentalis R. idaeus 2308 Caroline GEO-1 (Autumn Bliss x Glen Moy) x Heritage R. idaeus 618456 Centennial Meeker x Skeena R. idaeus 553508 Chief Latham x Newburgh R. idaeus 553537 Chilcotin Sumner x Newburgh R. idaeus 553503 Chilliwack BC 64-10-198 (Sumner x Carnival) x Skeena R. idaeus 618392 Coho Lewis x ORUS 520-48 (OSC 1586 x OSC 1655) R. idaeus 553504 Comox BC 64-9-81(Creston x Willamette) x Skeena R. idaeus 553363 Cuthbert Chance seedling of R. idaeus var. strigosus R. idaeus 553425 Dormanred R. parvifolius x Dorsett R. hybrid 618582 Glen Moy Complex of several cultivars R. idaeus 553511 Glen Prosen Complex of several cultivars R. idaeus 553457 Goldenwest Cuthbert x Lloyd George R. idaeus 553382 Heritage (Milton x Cuthbert) x Durham R. idaeus 2330 Josephine Amity x Glenn Garry R. idaeus 553479 Killarney Chief x Indian Summer R. idaeus 618311 Kitsilano Comox x EM 3909/4 R. idaeus 553564 Latham King x Loudon R. idaeus 2004 Lloyd George Selection from wild red raspberry in Dorsetshire, England R. idaeus 638206 Malahat Meeker x BC/SCRI7853/116 (complex parentage, including R. idaeus 553515 Malling Enterprise Preussen x (Pyne’s Royal selfed x Lloyd George selfed) R. idaeus

Table 2.1. (Continued)

P. I. no. Name Pedigree or origin Taxon 553493 Mandarin (R. parvifolius x Taylor) x Newburgh R. idaeus 553384 Meeker Willamette x Cuthbert R. idaeus 553369 Newburgh Newman x Herbert R. idaeus 553448 Newman Chance seedling from a mix of several cultivars R. idaeus 553480 Nova Southland x Boyne R. idaeus 553525 Preussen Thought derived from Superlative x Marlboro cross R. idaeus 553408 R. strigosus R. idaeus var. strigosus from Anaconda, Idaho R. strigosus 553385 Reveille [VPI 1 (Indian Summer x Sunrise) x September] R. idaeus 553468 Rubin Bulgarski Lloyd George x Preussen R. idaeus 553377 September Marcy x Ranere R. idaeus 553366 St. Regis Selection of red raspberry R. idaeus 553502 Summit OSC 1838 (Fallred x OSC 1347) x OSC 1842 (NY 600 x OSC 1347) R. idaeus 553451 Sumner Washington x Tahoma R. idaeus 553470 Tahoma Latham x Lloyd George R. idaeus 554015 Trailblazer Augustred x Cheyenne 63-16 R. strigosus 618441 Tulameen Nootka x Glen Prosen R. idaeus 553471 Veten Asker x Lloyd George R. idaeus 553368 Viking Cuthbert x Marlboro R. idaeus 553380 Washington Cuthbert x Lloyd George R. idaeus 553640 Wawi Washington x Willamette R. idaeus 553362 Willamette Newburgh x Lloyd George R. idaeus 553520 WYO US 68-21 (2) Selection of red raspberry R. strigosus 553428 Zeva Herbsternte (Indian Summer x Romy) x Romy R. idaeus 553473 Zzopska Alena Selection of red raspberry from Kostinbrod, Bulgaria R. idaeus

Table 2.2. The 48 blackberry accessions analyzed with 12 SSR loci, USDA-ARS Plant Introduction numbers (P. I. no.), cultivar name or taxon for species, pedigree or origin and taxon are included.

P.I. no Name Pedigree or origin Taxon 553300 Anderson Unknown eastern U.S. erect type R. hybrid 553278 Ashton Cross selection of wild European blackberry 4x thorny R. hybrid 553260 Aurora OSC 616 (Zielinski x Logan) x OSC 73 (Logan x Austin Thornless) R. hybrid 553292 Austin T. Mayes Open-pollinated seedling selection R. hybrid 553243 Bailey R. allegheniensis hybrid R. hybrid 553277 Bedford Giant Self of Vietchberry, hexaploid R. hybrid 553318 Benenden R. trilobus x R. deliciosus R. hybrid 638257 Black Diamond Kotata x NZ 8610L-163 (E90 x N-71) R. hybrid 638260 Black Pearl ORUS 1117-11 (OSC 1122 x OSC 2009) x ORUS 1122-1 (Olallie x ORUS 728-3) R. hybrid 553272 Black Satin SIUS 47(US 1482 X Darrow) X Thornfree R. hybrid 553336 Boysen presumed R. ursinus x R. idaeus R. hybrid 553341 Boysen 43 Clonal selection of Boysen R. hybrid 553244 Brazos F2 of (Lawton x Nessberry) R. hybrid 553245 Brison F2 of (Brainerd x Brazos) x Brazos R. hybrid 553261 Carolina Austin Thornless x Lucretia R. hybrid 553246 Cascade Zielinski x Logan R. hybrid 553273 Chehalem Santiam x Himalaya R. hybrid 553247 Cherokee Darrow x Brazos R. hybrid 553322 Chester T. SIUS 47(US 1482 X Darrow) x Thornfree R. hybrid 553250 Darrow NY 15826 (Eldorado x Brewer) x Hedrick R. hybrid 553251 Dirksen T. SIUS 47(US 1482 X Darrow) x Thornfree R. hybrid 553319 Ebano F2 of Comanche x (Thornfree x Brazos) R. hybrid 618384 Eldorado Unknown (R. allegheniensis x R. argutus) R. hybrid

Table 2.2. (Continued)

P.I. no. Name Pedigree or origin Taxon 553312 Flordagrand F2 from Regal Ness x R. trivialis R. hybrid 553348 Illini Hardy NY 95 x Chester Thornless R. hybrid 638183 Kiowa Ark. 791 x Ark. 1058 R. hybrid 553293 Kotata OSC 743 (Pacific x Boysen) x OSC 877(Jenner 1 x Eldorado) R. hybrid 553314 Lincoln Logan Histogenic manipulation of Thornless Loganberry R. hybrid 638182 Loch Ness complex hybrid derived from Comanche, Chehalem, R. hybrid 553258 Logan T. Thornless mutation of Logan R. hybrid 553253 Lucretia unknown R. hybrid 553254 Marion Chehalem x Olallie R. hybrid Merton 553276 Thornless John Innes selfed R. hybrid 638262 Metolius Douglass x Kotata R. hybrid 553343 Navaho Ark. 583 (Thornfreee x Brazos) x Ark. 631 (Ark. 550 x Cherokee) R. hybrid 638263 Nightfall Marion x Waldo R. hybrid 638259 Obsidian ORUS 828-43 (OSC 1122 x OSC 1683) x ORUS 1122-1 (Olallie x ORUS 728-3) R. hybrid 553255 Ollalie Black Logan x Young fro Mt. Hebo, Oregon R. hybrid 638265 ORUS1843-3 GP 9-24 (selection of R. ursinus) x Waldo R. hybrid 553256 Raven Dewblack x Eldorado 4x R. hybrid 618393 Rosborough F2 of (Brainerd x Brazos) x Brazos R. hybrid 553321 Shawnee Cherokee x Ark. 583 (Thornfree x Brazos) R. hybrid 553324 Sunberry R. ursinus x self tetraploid mutant of Malling Jewel R. hybrid 553351 Tayberry Aurora x SHRI 626/67 (4x raspberry) R. hybrid 553354 Tillamook Selected from the wild in Oregon R. hybrid 553334 Waldo OSC 1122 (Marion x OSC 878) x OSC 1367 (OSC 1083 x NC 3735M-2) R. hybrid R. 271519 Watlab Seedling collection from seed collected wild in India armeniacus

Simple Sequence Repeat Genotyping Thirteen primer pairs were used for fingerprinting 48 raspberries and 48 blackberries. Fluorescent forward primers were ordered from Sigma-Proligo (Sigma-Aldrich, MO). PCR reactions were performed separately for each primer pair using a fluorescently labeled forward primer and an unlabeled reverse primer. Reactions were performed in 15 µL volumes containing 1X reaction buffer, 2 mM MgCl2, 0.2 mM dNTPs, 0.15 µM of each primer, 0.025 U of Biolase Taq DNA polymerase (Bioline USA Inc., MA) and 3 ng genomic DNA. A 5 µL aliquot of each PCR product was reserved for fragment analysis after verification of amplification success by separating the remaining PCR product by 1.5% agarose gel electrophoresis. Failure to amplify was scored as a null genotype after three failed attempts on a DNA template that amplified using other primers. Fragment analysis was determined after separation on a Beckman CEQ 8000 genetic analyzer (Beckman Coulter Inc., Fullerton, CA). PCR products were loaded in 26 µL of Sample Loading Solution that contained 0.6 µL of CEQ-600 size standard and 1 µL of diluted and undiluted PCR products. The optimal amount of PCR product was determined experimentally. Up to two primer pairs were multiplexed after PCR. Allele sizing and visualization were performed using the fragment analysis module of the CEQ 8000 software. Alleles were scored by fitting the peaks into bins less than 1 nucleotide. Data Analysis Statistical Analysis of SSR Data Due to the multiple ploidy levels of the raspberry and blackberry cultivars, individuals were scored for the presence or absence of each allele and each allele was treated as a separate locus. A Perl script was used to convert the codominant data into a binary data format. Genetic distance matrices were computed using PowerMarker (Version 3.25) (Liu and Muse, 2004) with the proportion of shared alleles distance, Dsa (Chakraborty and Jin, 1993): 1 m aj Dsa = ∑∑min( pij, qij) m j==11i

Table 2.3. Thirteen SSR loci of the expected size. The optimum annealing temperature (Ta) as determined by the gradient PCR, the expected size of the PCR amplicon and the forward (Fwd) and reverse (Rev) primers used to amplify them are included.

Locus Ta (°C) Size Sequence Fwd: CGACACCGATCAGAGCTAATTC RiM015 62 350 Rev: ATAGTTGCATTGGCAGGCTTAT Fwd: GAAACAGGTGGAAAGAAACCTG RiM017 59 194 Rev: CATTGTGCTTATGATGGTTTCG Fwd: ATTCAAGAGCTTAACTGTGGGC RiM019 59 176 Rev: CAATATGCCATCCACAGAGAAA Fwd: AGCAACCACCACCTCAACTAAT RiM036 51 315 Rev: CTAGCAGAATCACCTGAGGCTT Fwd: GGTTCGGATAGTTAATCCTCCC RhM001 51 232 Rev: CCAACTGTTGTAAATGCAGGAA Fwd: CCATCTCCAATTCAGTTCTTCC RhM003 50 200 Rev: AGCAGAATCGGTTCTTACAAGC Fwd: AAAGACAAGGCGTCCACAAC RhM011 56 280 Rev: GGTTATGCTTTGATTAGGCTGG Fwd: CACCAATTGTACACCCAACAAC RhM018 54 379 Rev: GATTGTGAGCTGGTGTTACCAA Fwd: CAGTCCCTTATAGGATCCAACG RhM021 50 282 Rev: GAACTCCACCATCTCCTCGTAG Fwd: CGACAACGACAATTCTCACATT RhM023 53 196 Rev: GTTATCAAGCGATCCTGCAGTT Fwd: CAACCTAATGACCAATGCAAGA RhM031 50 396 Rev: GCAGAATCCATTCTCTTGTTGA Fwd: GGACACGGTTCTAACTATGGCT RhM043 56 373 Rev: ATTGTCGCTCCAACGAAGATT

where pij and qij are the frequencies of the ith allele at the jth locus, m is the number of loci examined, and aj is the number of alleles at the jth locus. Cluster analysis based on the Neighbor Joining (NJ) and Unweighted Pair Group Method with Arithmetic Mean (UPGMA) algorithms used MEGA version 3.1 (Kumar et al., 2004) to create the dendrograms.

Table 2.4. Raspberries and blackberries used for initial primer screening. Their cultivar name, USDA Plant Introduction numbers (P.I. no.), taxon and ploidy level are included when available.

P. I. no. Name Taxon Ploidy Raspberries 553382 Heritage R. idaeus 14 553392 Malling Joy R. idaeus 14 553513 Krupna Dvorodna R. idaeus 14 2004 Lloyd George R. idaeus 14 553384 Meeker R. idaeus 14 553438 Pathfinder R. idaeus 14 553385 Reveille R. idaeus 14 553430 Rose de Cote d'Or R. idaeus 14 553366 St. Regis R. idaeus 14 553470 Tahoma R. idaeus 14 553362 Willamette R. idaeus 14 553428 Zeva Herbsternte R. idaeus 14 Blackberries 553292 Austin Thornless R. hybrid 56 553243 Bailey R. hybrid 28 553247 Cherokee R. hybrid 28 553322 Chester Thornless R. hybrid 28 638183 Kiowa R. hybrid 28 553293 Kotata R. hybrid 49 553254 Marion R. hybrid 42 553343 Navaho R. hybrid 28 553255 Olallie R. hybrid 42 553324 Sunberry R. hybrid 42 553351 Tayberry R. hybrid 42 271519 Watlab R. armeniacus 14

Results Isolation and characterization of microsatellite loci A total of 96 clones each of ‘Meeker’ and ‘Marion’ was screened by colony PCR for the presence of inserts. Out of 74 fragments sequenced from the ‘Meeker’ library, 70 contained microsatellite motifs and of 90 fragments sequenced from ‘Marion’, 78 motifs were found. Clustering of the fragments using CAP 3 software identified 45 unique microsatellite- containing sequences in ‘Meeker’ and in ‘Marion’. Twenty four primer pairs were designed in microsatellite-flanking sequences of ‘Meeker’ DNA and 30 pairs were designed to amplify SSR-containing sequences in ‘Marion’. Short flanking sequences prevented primer design in the remaining SSR-containing sequences. Out of seven motifs used in the enrichment the TG motif was the most abundant repeat isolated (25%) followed by AG (8.33%) and ATC (8.33%) repeats. The AAC, AAG, AAT, ACT were not enriched in the sequences. Fifteen out 24 primer pairs designed in red raspberry failed to amplify in ‘Meeker’ and 20 of 30 primer pairs isolated from blackberry sequences did not produce a PCR product in ‘Marion’. The remaining 9 SSRs from ‘Meeker’ and 10 SSRs from ‘Marion’ were tested for amplification in a set of 12 accessions each of raspberry and blackberry (Table 2.3). Monomorphic PCR fragments were generated from two primer pairs designed from ‘Meeker’ sequences and one pair isolated from ‘Marion’ sequences. Four primer pairs (3 from ‘Meeker’ and 1 from ‘Marion’) failed to amplify and produced unclear bands that were difficult to interpret in some of the 12 raspberry and blackberry accessions tested. These seven primer pairs that generated monomorphic bands, produced unclear bands or failed to amplify in some accessions were discarded. Level of Polymorphism A total of 4 primer pairs from the ‘Meeker’ library and 8 primer pairs from the ‘Marion’ library generated polymorphic DNA fragments of the expected size in the 12 raspberry and blackberry test set. These 12 primer pairs and 1 primer pair designed from an (AT)6-containing R. idaeus sequence (AF292369) from

Genbank were used to fingerprint 48 raspberries and 48 blackberries (Table 2.1 and Table 2.2). RhM031 failed to amplify in raspberries and RiG001 failed to amplify in blackberries. In raspberries, the 12 SSRs generated 3 to 16 alleles per locus, with an average of 8 and a total of 96 alleles. In blackberries, the 12 primer pairs amplified 6 to 31 alleles per locus, with an average of 15 and a total of 177 alleles. Shannon’s index ranged from 0.34 to 7.15 in raspberries and 0.44 to 15.79 in blackberries. The average Shannon’s index among the blackberries (7.22) was significantly higher than among the raspberries (3.63) (two sided p-value from paired t-test <0.003). Relatedness of Cultivars and Variety Identification NJ cluster analysis was used to describe genetic variation and relatedness among the 96 accessions (Fig. 2.1). The NJ dendrogram separated the 96 cultivars into two major clusters corresponding to the raspberries and the blackberries. Cultivars that contained unusual Rubus species in their ancestry formed a heterogeneous group separate from the two major groups and contained: ‘Trailblazer’, ‘August Red’ x ‘Cheyenne’ progeny; ‘Dormanred’, from ‘Van Fleet’ x R. parvifolius; ‘Zeva Herbstente’, from ‘Indian Summer’ x ‘Romy’; ‘Flordagrand’, ‘RegalNess’ x R. trivialis progeny; and ‘Benenden’, a R. trilobus x R. deliciosus progeny. Hybrid berries like ‘Sunberry’ and a descendant of R. rusticanus (spine-free variant of R. ulmifolius), ‘Bedford Giant’ failed to group with the remaining accessions. The first major cluster contained the raspberries that grouped according to their pedigree. The groupings based on the species of their founding clones were evident. The raspberries were divided based on the presence of R. strigosus R. idaeus or the R. idaeus x R. strigosus hybrids in their ancestry. The R. strigosus group was divided into three subclusters based on the parents ‘Newman’ and ‘Ranere’ (‘St. Regis’) plus cultivars based on ‘Newman’ but with the intermediate parent ‘Durham’. The first subcluster contained ‘Newman’ and its derivatives via ‘Newburgh’. It included the cultivars, ‘Nova’, ‘Chief’ and ‘Latham’. Cultivars that are based on ‘Newman’ but have ‘Durham’ as its intermediate parent formed a separate group. The third group from R. strigosus included ‘Ranere’ and its derivatives. ‘Reveille’ and ‘Autumn Bliss’

62 formed the clade at the R. strigosus and R. idaeus boundaries since both species are found in their ancestry especially of ‘Reveille’. Cultivars based on R. idaeus formed the next group of raspberries separated based on the founding clones ‘Lloyd George’ and ‘Hudson River Antwerp’ through the parents ‘Cuthbert’, ‘Marlboro’ and ‘Creston’. The first subgroup includes ‘Lloyd George’ and its derivatives. The next subgroup includes the cultivars based on the ‘Hudson River Antwerp’ through the intermediate parents ‘Cuthbert’ and ‘Marlboro’. ‘Creston’ derivatives formed another subgroup with ‘Skeena’ as its intermediate parent. ‘Creston’ is of unknown origin but is believed to be 50% ‘Lloyd George’ (Dale et al., 1993). Included in the subgroup are the British selections ‘Glen Moy’, ‘Glen Prosen’ and its derivative ‘Tulameen’ that are known to have high ‘Lloyd George’ content. A mixed subgroup of R. idaeus and R. strigosus descendants included ‘Newburgh’ and the rest of the cultivars. Blackberries were subdivided into two clades: the first contained Eastern tetraploid cultivars and the other consisted of Western cultivars of higher ploidy levels. In the Eastern clade, the erect thorny cultivars that are mostly derived from the Arkansas breeding program formed the first subcluster based on the parents ‘Brazos’ and ‘Darrow’. The second subcluster included the thornless populations based on ‘Thornfree’ or ‘John Innes’ parents that are either semi-erect or trailing. ‘Hull Thornless’, ‘Black Satin’, ‘Chester’ and ‘Dirksen Thornless’ that resulted from the ‘SIUS 47’ and ‘Thornfree’ cross grouped together. ‘Illini Hardy’ was also included in this group even though it is thorny due to the presence of ‘Chester’ in its pedigree (NY 95 x ‘Chester’). The Western clade contained trailing thornless cultivars that were derived from ‘Austin Thornless’ and thorny cultivars released by the USDA-ARS breeding program in Corvallis, Oregon. SSR analysis did not distinguish between two pairs of clones: ‘Logan Thornless’ and ‘Lincoln Logan’; ‘Boysen’ and ‘Boysen 43’.

Discussion SSR markers were successfully isolated from ‘Meeker’ and ‘Marion’ and were used as genetic markers in 48 genotypes each of raspberries and blackberries. The hybridization capture method (Kandpal et al., 1994) was used in the enrichment for microsatellite DNA. Hybridization capture is the predominant strategy in use because it allows selection before cloning and thus, is faster and easier to do with multiple samples (Glenn and Schable, 2005). It is effective for sequences that have less abundant microsatellites. In this study, high enrichment efficiency was achieved by using a double round of enrichment. 94.5% of the fragments in ‘Meeker’ and 87% of the fragments in ‘Marion’ were found to contain SSRs. This may suggest that microsatellites are abundant in the Rubus genome. Compared to other studies, enrichment efficiency is in excess of 20% as in Pelargonium following two rounds of enrichment (Becher et al., 2000). Low enrichment efficiency suggests that microsatellites may be relatively rare in the genome (Becher et al., 2000). The TG repeat was the most frequent motif in the enriched genomic library. This is similar to studies in mammals where AC/TG repeat was the most common (Hamada and Kakunaga, 1982; Stallings et al., 1991). The second common motif was the AG repeat which also occurs frequently in both EST libraries and in the enriched genomic library in most plant species (Cardle et al., 2000; Toth et al., 2000). Isolation of microsatellite loci involves a series of steps that can lead to a small number of robust microsatellite primer pairs. The low number of SSR markers developed out of the 24 primer pairs designed for ‘Meeker’ (4.2%) and 30 primer pairs designed for ‘Marion’ (8.3%) is not unusual. Starting with the successfully sequenced clones, attrition could occur due to the lack of microsatellites and the presence of duplicate or chimeric clones, failure of primer design and non-Mendelian or non- polymorphic products (Squirrel et al., 2003). Twenty-five microsatellite sequences occurred more than once in ‘Meeker’ and 33 microsatellite seqeunces in ‘Marion’ libraries. Duplicate clones were observed in other microsatellite-enriched libraries and are most likely a consequence of the PCR step that follows the affinity capture

(Becher et al., 2000). Chimeric sequences having one flanking region similar to another clone (Squirrel et al., 2003) may be generated during the initial restriction/ligation step or during the PCR step of the enrichment cloning procedure (Koblizkova et al., 1998). Only 24 sequences in ‘Meeker’ and 30 sequences in ‘Marion’ were suitable for primer design mostly due to a short flanking region or unsuitable base composition of one or both flanking regions. In the case of pistachio, from a total of 151 clones only 67 (44%) had sufficient flanking sequence for primer design (Ahmad and Southwick, 2003). In sunflower 54% of the sequences with microsatellites were suitable for primer design (Paniego et al., 2002). In this study, the standard Dye-labeled dideoxy Terminator Cycle Sequencing (DTCS) Quick Start kit used for routine sequencing failed to generate clean sequences and the Methods Development (MD) kit, intended for sequencing difficult templates was used. This MD kit dNTP(G) mix solution chemistry is recommended for difficult G-C rich and polymerase hard stops regions and secondary structures. However, band compression is a common occurrence and may have contributed to low yield in the number of successful microsatellite loci obtained as well as the failure of primers pairs developed. Initial test on the 12 raspberry and blackberry cultivars resulted in 62.5% of the primer pairs failing in ‘Meeker’ while 66.7% failed in ‘Marion’. For Begonia, of the 24 primer pairs designed and tested, 6 (25%) failed to amplify a product (Hughes et al., 2002). Further attrition of loci happened during the initial evaluation of amplification in a set of 12 raspberry and 12 blackberry cultivars (Table 2.3). Most of the motifs of the monomorphic primer pairs were found to be relatively short like in Actinidia (Weising et al., 1996). For Actinidia 4 out of the 16 microsatellites tested produced a single band for all the 8 plants tested. In ‘Meeker’ two primer pairs were monomorphic while in ‘Marion,’ one primer pair was monomorphic. Four primer pairs (three in ‘Meeker’ and one in ‘Marion’) were further discarded because of their failure to amplify in some cultivars and the production of uninterpretable banding patterns.

The final number of primer pairs developed was four for ‘Meeker’ and eight for ‘Marion’. In the study the percentage of attrition was quite high with 95% for ‘Meeker’ and 91% for ‘Marion’. The average attrition rate from 71 plant studies is 83% from obtaining sequences with microsatellites to the designing of the primers. Producing 10 microsatellite loci will require an average of 60 clones for sequencing, identifying 38 microsatelllites and designing 20 primer pairs (Squirrel et al., 2003). Despite the low frequency of SSR markers developed from ‘Meeker’ and ‘Marion’ the resulting 12 primer pairs and the EST-SSR primer pair identified from GenBank were polymorphic and efficiently identified the raspberry and blackberry cultivars tested. The observed allele range of 3-16 alleles per locus in raspberries is lower than what was reported previously in Rubus: 15-30 in the native range of R. alceifolius (Amsellem et al., 2001a) and is within the range of 7-16 alleles per locus observed in R. idaeus (Graham et al., 2002). Compared to R. alceifolius, the lower number of alleles per locus obtained in raspberries may be due to their diploid nature as opposed to the suspected tetraploid nature of R. alceifolius. The range of 6-31 alleles per locus observed in blackberries is within the range of what was reported previously. In raspberries, the limited genepool and their diploid nature makes them less polymorphic compared to blackberries. Raspberry cultivars released since the 1960s were derived from a few founding clones and have a narrow genetic base (Dale et al., 1989). Blackberry is a highly heterozygous plant and many current cultivars are either tetraploid, hexaploid or of higher ploidy levels (Stafne and Clark, 2003). Polyploid blackberries have varied reproductive stages (sexual, facultatively apomictic and obligately apomictic (Hall, 1990), cytological conditions (auto- and allopolyploidy) and inheritance strategies (disomic and tetrasomic). Wide crossing is a feature of the blackberry breeding programs (Hall, 1990). The high ploidy levels present have provided potential for wide interspecific crossing within Rubus subgenus and between the Rubus and Idaeobatus subgenera (Jennings, 1981). The two subgenera are interfertile with the genomes sufficiently differentiated to enforce preferential pairing of like genomes in allopolyploids and to ensure regular mitosis

GlenProsen GlenMoy Tulameen Comox Raspberries Chilliwack Kitsilano Trailblazer Dormanred ZevaHerbsternte Flordagrand Benenden Rosborough Brison Kiowa Raven Anderson Shawnee Darrow Bailey Eldorado Cherokee Brazos HullThornless BlackSatin Chester DirksenThornless Navaho LochNess MertonThornless IlliniHardy Ebano AshtonCross Watlab Waldo Blackberries ORUS18433 Nightfall Marion BlackPearl Obsidian Chehalem Carolina AustinThornless Lucretia LoganThornless LincolnLogan Olallie Boysen43 Boysen Tayberry Aurora Metolius Kotata BlackDiamond Tillamook Cascade BedfordGiant Sunberry Nova Newman Latham Chief WYOUS68212 StRegis September Rstrigosus Heritage Caroline Amber Summit Amity Josephine Anne Reveille AutumnBliss Raspberries LloydGeorge Killarney Malahat Centennial Veten RubinBulgarski ZzopskaAlena Newburgh Tahoma Sumner Mandarin Chilcotin Preussen MallingEnterprise Viking Goldenwest Cuthbert Washington Coho Wawi Meeker Willamette

0.02

Fig. 2.1. Neighbor Joining (NJ) tree of 96 raspberry and blackberry cultivars based on the proportion of shared alleles distance for 13 SSR loci.

67 and good fertility (Thomas, 1940). But despite the variation produced from the wide crosses, the genetic base of cultivated blackberries is still narrow (Jennings, 1981).The lower total number of alleles and Shannon’s index obtained from raspberries compared to blackberries also show the less polymorphic nature of raspberries. Raspberries had a total of 96 alleles and Shannon’s index of 3.63 while blackberries had a total of 177 alleles and a Shannon’s index of 7.22. A large amount of allelic diversity was observed in blackberries as opposed to raspberries. Shannon’s index is a measure of diversity that takes into account the number of classes (allelic richness) and the distribution of individuals among classes (allelic evenness) (Shannon and Weaver, 1949). Therefore Shannon’s index may accurately measure diversity in polyploids where determination of copy number of alleles is uncertain and allele frequencies cannot be determined. High allele number may reflect the ability of SSR markers to provide unique genetic profiles for individual genotypes. Most of the primer pairs designed from ‘Meeker’ was functional in the blackberry accessions and vice versa. This shows the conserved nature of microsatellite loci within a genus as observed in other plant genomes as well (Becher et al., 2000; Weising et al., 1996). These results did not agree with the study of Stafne and Clark (2005) wherein only 28.5% of raspberry-derived SSR markers amplified a product in blackberries. The high failure of amplification was attributed to the few shared species within the respective backgrounds of the SSR marker derived from ‘Glen Moy’ and the blackberry cultivars tested (Stafne and Clark, 2005). Of the eight primer pairs designed from ‘Marion’, one primer pair, RhM031, did not amplify a product in raspberries. EST-SSR primer pair RiG001 that was obtained from an R. idaeus sequence (AF292369) did not cross-amplify in blackberries. Two primer pairs, RhM003 and RhM031, failed to amplify in five blackberry cultivars. RhM003 failed in ‘Brazos’ while RhM0031 failed in ‘Benenden’, ‘Ebano’, ‘Tayberry’ and ‘Watlab’. The presence of null alleles may be due to a mutation in the primer binding sites. Mismatches in the annealing region or to artifact inserts originated during the enrichment period can cause failure of amplification (Marinoni et al., 2003).

Scoring of the allele quality also revealed the presence of two PCR artifacts: stutter, the production of additional peaks different from the true peak by a multiple of the repeat unit and split peaks resulting from the unequal non-template addition of adenosine by Taq polymerase (Appendix D). Stutter bands are generated by the slippage of Taq polymerase during PCR (Ginot et al., 1996; Litt et al., 1993). Stutter peaks were more common in raspberries (31%) and blackberries (38%) than split peaks wherein both raspberries and blackberries had 15% of the split peaks. The presence of null alleles and the PCR artifacts are of particular concern since they can create consistent allelic and genotypic scoring bias that may, in turn, bias data interpretation (DeWoody et al., 2006). Detection of errors and mitigation of scoring errors is advisable for accurate SSR analysis. In the NJ dendrogram, Rubus cultivars clustered according to their ancestry into two large groups and a heterogeneous group consisting of cultivars containing uncommon species in their background while two hybrid berries failed to join any group (Fig. 2.1). Raspberries from the Idaeobatus subgenus were grouped in the first cluster while blackberries from the subgenus Rubus grouped in the second cluster. The raspberry and blackberry groups were separated based on their founding clones as well as on the breeding program that released them. In raspberries, by tracing the species derivation of the founding clones, most are found to be derived from R. idaeus and a few from R. strigosus. This is due to the three cycles of breeding in raspberries that were employed: 1) many cultivars were selected within the R. idaeus subsp. idaeus and subsp. strigosus either from chance discoveries or from families of open-pollinated seedlings; 2) some of these were used to produce R. idaeus x R. strigosus hybrid varieties; and 3) some of these hybrid varieties and some of the earlier chance discoveries were used to produce modern varieties (Jennings, 1988). The raspberry breeding strategy employed resulted in the development of cultivars based on few parents. Parents like ‘Lloyd George’, ‘Preussen’ and ‘Newburgh’ were the popular parents for a number of cultivars. In a study by Dale et al., (1993), 4 founding clones occurred in more than 90 raspberry

69 pedigrees tested. These are ‘Lloyd George’, ‘Hudson River Antwerp’, ‘English Globe’ and ‘Highland Hardy’. The latter three are found in the intermediate parents, ‘Cuthbert’ and ‘Marlboro’. Both are derived from ‘Hudson River Antwerp’ with ‘Marlboro also derived from ‘English Globe’ and ‘Highland Hardy’. The results of the NJ analysis revealed the groupings of the raspberries similar to the results obtained by Dale et al. (1993). The grouping of raspberries based on the contribution of R. idaeus, R. strigosus and crosses of R. idaeus and R. strigosus on their genetic make-up was evident in the dendrogram. R. strigosus contributed to the ancestry of raspberries through ‘Newman’ or ‘Ranere’. Cultivars based on these intermediate parents grouped together and formed subclusters. The next group in the raspberries is based from the R. idaeus background separated into subgroups based on the founding clones, ‘Lloyd George’, Hudson River Antwerp’ and ‘Creston’. Dale et al., (1993) determined the pedigrees of 137 raspberries released throughout the world since 1960. All 137 varieties of known parentage were found to be derived from only 50 founding clones. The mean genetic contribution of the founding clones ranged from <0.1% to 21%. This range indicates that raspberries were developed from a limited germplasm base and explains the closeness of the raspberries as shown in our dendrogram and the low allelic diversity observed in the study. Since the narrow genetic base observed in raspberries is alarming to breeders and growers, several strategies are employed to increase genetic diversity. These include use of more parents of different ancestry per generation, introduction of unrelated germplasm, introduction of germplasm from wild R. idaeus and R. strigosus and introduction of unimproved germplasm from other species of Idaeobatus (Dale et al., 1993). Blackberries were subdivided into two primary clades: the eastern tetraploid cultivars and the western cultivars of higher ploidy levels. The eastern North American blackberries can further be classified by cane architecture that is either semi-erect or erect-caned (Clark, 2005) and by their thornlessness. In the first clade,

70 blackberries from the eastern region formed closely related groups within which the thorny and thornless populations formed subgroups. The erect thorny cultivars that are mostly derived from the Arkansas breeding program formed one cluster. Cultivars in the subgroup have mostly R. allegheniensis Porter as their founding clone consistent with the study of Stafne et al. (2003). According to Stafne and Clark (2003), R. allegheniensis had the highest mean genetic contribution of 18.93% indicating that it comprised nearly 20% of the total genetic make-up in all the of 13 blackberry cultivars they tested. R. frondosus and R. argutus had the next greatest mean genetic contributions with 17.07% and 12.02% respectively. The genetic contribution of these founding clones to the cultivars may be observed from the intermediate parents ‘Brazos’ and ‘Darrow’, (’Eldorado’ x ‘Brewer’) x ‘Hedrick’. ‘Brazos’ and ‘Darrow’ were used extensively in breeding programs because of the proclivity of producing exceptional offsprings (Stafne et al., 2004). ‘Anderson’ has an unknown pedigree but grouped with the eastern blackberries possibly because of the origin of the erect architecture of the plant. The second eastern North American blackberry cluster contained thornless cultivars that are either semi-erect or trailing. Thornlessness is a desirable trait in blackberries. R. ulmifolius var. inermis has been used to transmit the important recessive thornless gene to modern blackberry cultivars through the intermediate parents ‘John Innes’ and ‘Thornfree’ (Stafne et al., 2003). A thorny, wild selection from Europe had surprisingly joined the group; ‘Ashton Cross’ a successful cultivar in northern Britain is known for its ability to flower late. The Scottish breeders combined the rapid fruit maturation of ‘Ashton Cross’, spinelessness of ‘Thornfree’ and erectness and early flowering of ‘Darrow’, resulting in ‘LochNess’ after several generations of breeding (Jennings, 1988). The second clade of blackberries is comprised of the trailing blackberries or dewberries of the Pacific Coast of North America. Rubus ursinus was widely used in breeding Western blackberries while R. laciniatus (‘Evergreen’ and ‘Oregon Evergeeen’) and R. armeniacus (‘Himalaya’, ‘Himalaya Giant’) were also imported from Europe as potential crops for the region. Rubus species or R. argutus L. were

71 also used to breed for thornlessness, cold-hardiness or primocane-fruiting characteristic in western blackberries. Rubus ursinus made a genetic contribution to the intermediate parents ‘Aughinbaugh’, ‘Santiam’ and ‘Zielinski’. ‘Austin Mayes’ served as the intermediate parent for Rubus species and are present in the cultivars released from Oregon (Finn et al., 2005). Clonal variations were not distinguished in the dendrogram. ‘Boysen 43’ is a variation of ‘Boysen’ and ‘Lincoln Logan’ and ‘Logan Thornless’ is another pair of clones with similar SSR profiles. Like in the study of Yamamoto et al., (2001) and Hokanson et al., (1998), mutant pear cultivars and apple sport mutants were found to be indistinguishable by SSR analysis. The successful development of SSR markers from both raspberry and blackberry enabled the efficient fingerprinting of raspberries and blackberries. These SSR markers can complement traditional morphological markers used for the identification of Rubus cultivars. It enables an easy and accurate genetic diversity assessment helpful in the identification of gaps in the collections, identification of redundancies and development of core collections. DNA markers like SSRS can guide future collection strategies as well as the management of genetic resources stored at the Repository.

Literature Cited

Ahmad, F. and S. Southwick. 2003. Identification of pistachio (Pistachia vera L.) nuts with microsatellite markers. J Amer Soc Hort Sci. 128: 898-903.

Amsellem, L., C. Dutech, and N. Billotte. 2001a. Isolation and characterization of polymorphic microsatellite loci in Rubus alceifolius Poir. (Rosaceae), an invasive weed in La Réunion Island. Mol Ecol Notes. 1: 33-35.

Amsellem, L., J.-L. Noyer, and M. Hossaert-McKey. 2001b. Evidence for a switch in the reproductive biology of Rubus alceifolius (Rosaceae) towards apomixis, between its native range and its area of introduction. Am J Bot. 88: 2243-2251.

Becher, S.A., K. Steinmetz, K. Weising, S. Boury, D. Peltier, J.P. Renou, G. Kahl, and K. Wolff. 2000. Microsatellites for cultivar identification in Pelargonium. Theor Appl Genet. 101: 643-651.

Cardle, L., L. Ramsay, D. Milbourne, M. Macaulay, D. Marshall, and R. Waugh. 2000. Computational and experimental characterization of physically clustered simple sequence repeats in plants. Genetics. 156: 847-854.

Clark, J.R. 2005. Intractable traits in eastern U.S. blackberries. HortScience. 40: 1954- 1955.

Dale, A., R.J. McNicol, P.P. Moore, and T.M. Sjulin. 1989. Pedigree analysis of red raspberry. Acta Hort. 262: 35-39.

Dale, A., P.P. Moore, R.J. McNicol, T.M. Sjulin, and L.A. Burmistrov. 1993. Genetic diversity of red raspberry varieties throughout the world. J Amer Soc Hort Sci. 118: 119-129.

Daubeny, H.A. 1996. Brambles. In: J. Janick and J.N. Moore (eds.). Fruit Breeding, Volume II: Vine and Small Fruits Crops.

Daubeny, H.A. and A. Anderson. 1989. Germplasm enhancement in the British Columbia raspberry breeding program. Acta Hort. 262: 61-64.

DeWoody, J., J.D. Nason, and V.D. Hipkins. 2006. Mitigating scoring errors in microsatellite data from wild populations. Mol Ecol.

Donnelly, D.J. and H.A. Daubeny. 1986. Tissue culture of Rubus species. Acta Hort. 183: 305-314.

Finn, C.E., B. Yorgey, B.C. Strik, B. Yorgey, and J. DeFrancesco. 2005. 'Obsidian' trailing blackberry. HortScience. 40: In press.

Ginot, F., I. Bordelais, S. Nguyen, and G. Gyapay. 1996. Correction of some genotyping errors in automated fluorescent microsatellite analysis by enzymatic removal of one base overhangs. Nucl Acid Res. 24: 540-541.

Glenn, T.C. and N.A. Schable. 2005. Isolating microsatellite DNA loci. Methods Enzymol. 395: 202-222.

Graham, J. and R.J. McNicol. 1995. An examination of the ability of RAPD markers to determine the relationships within and between Rubus species. Theor Appl Genet. 90: 1128-1132.

Graham, J., K. Smith, M. Woodhead, and J. Russell. 2002b. Development and use of simple sequence repeat SSR markers in Rubus species. Mol Ecol Notes. 2: 250-252.

Hall, H.K. 1990. Blackberry Breeding. Plant Breeding Review. 8: 249-312.

Hamada, H. and T. Kakunaga. 1982. Potential Z-DNA forming sequences are highly dispersed in the human genome. Nature. 298: 396-398.

Hokanson, S., W. Lamboy, A. McFadden, and J. McFerson. 2001. Microsatellite (SSR) variation in a collection of Malus (apple) species and hybrids. Euphytica. 118: 281-294.

Huang, X. and A. Madan. 1999. Cap3: A DNA sequence assembly program. Genome Res. 9: 868-877.

Hughes, M., J. Russell, and P.M. Hollingsworth. 2002. Polymorphic microsatellite markers for the Socotran endemic herb Begonia socotrana. Mol Ecol. 1: 22-24.

Jahn, O.L. 1982. Maintenance of clonal plant germplasm. HortScience. 17: frontspiece and cover.

Jennings, D.L. 1981. A hundred years of loganberries. Fruit Var J. 35: 34-37.

Jennings, D.L. 1988. In: Raspberries and Blackberries: Their Breeding, Diseases and Growth. Academic Press Inc., USA.

Kandpal, R.P., G. Kandpal, and S.M. Weissman. 1994. Construction of libraries enriched for sequence repeats and jumping clones and hybridization selection for region-specific markers. Proc. Natl. Acad. Sci. 91: 88-92.

Koblizkova, A., J. Dolezel, and J. Macas. 1998. Subtraction with 3' modified oligonucleotides eliminates amplification artifacts in DNA libraries enriched for microsatellites. Biotechniques. 25: 32-38.

Kumar, S., K. Tamura, and M. Nei. 2004. MEGA 3: Integrated software for molecular evolutionary genetics analysis and sequence alignment. Briefings in Bioinformatics. 5: 150-163.

Litt, M., X. Hauge, and V. Sharma. 1993. Shadow bands seen when typing polymorphic dinucleotide repeats-some causes and cure. Biotechniques. 15: 280.

Liu, K. and S. Muse. 2004. Powermarker: new genetic data analysis software. Version 3.0. 10 October 2005. http://www.powermarker.net.

Lopes, M.S., B. Maciel, D. Mendonca, F.S. Gil, and A.D. Machado. 2006. Isolation and characterization of simple sequence repeat loci in Rubus hochstetterorum and their use in other species from the Rosaceae family. Mol Ecol Notes. 6: 750-752.

Marinoni, D., A. Akkak, G. Bounous, K. Edwards, and R. Botta. 2003. Development and characterization of microsatellite markers in Castanea sativa (Mill.). Mol Breeding. 11: 127-136.

Moore, J.N. and R.M. Skirvin. 1990. Blackberry management. In: Small fruit crop management. Prentice Hall, Englewood Cliffs, NJ.

Paniego, N., M. Echaide, M. Munoz, L. Fernandez, S. Torales, P. Faccio, I. Fuxan, M. Carrera, R. Zandomeni, E. Suarez, and H. Hopp. 2002. Microsatellite isolation and characterization in sunflower (Helianthus annuus L.). Genome. 45: 34-43.

Powell, W., G. Machray, and J. Provan. 1996. Polymorphism revealed by simple sequence repeats. Trends Plant Sci. 1: 215-222.

Rozen, S. and H.J. Skaletsky. 2000. Primer 3 on the WWW for general users and for biologist programmers. In: Bioinformatics Methods and Protocols: Methods in Molecular Biology. Humana Press, Totowa, NJ.

Shannon, C.E. and W. Weaver. 1949. The mathematical theory of communication. University of Illinois Press, Urbana, IL.

Skirvin, R.M., S. Motoike, M. Coyner, and M. Norton. 2005. Rubus spp. cane fruit. In: Biotechnology of fruit and nut crops. Cabi Publishing, Cambridge, MA.

Squirrel, J., P.M. Hollingsworth, M. Woodhead, J. Russell, A.J. Lowe, and M. Gibby. 2003. How much effort is required to isolate nuclear microsatellites from plants? Mol Ecol. 12: 1139-1348.

Stafne, E.T. and J.R. Clark. 2003a. Genetic Similarity Among Arkansas Blackberry Cultivars Based on Pedigree Analysis. Horticulture Studies: 29-31.

Stafne, E.T. and J.R. Clark. 2005. Simple Sequence Repeat (SSR) markers for genetic mapping of raspberry and blackberry. J Amer Soc Hort Sci. 130: 722-728.

Stafne, E.T., J.R. Clark, M.C. Pelto, and J.T. Lindstrom. 2003. Discrimination of Rubus cultivars using RAPD markers and pedigree Analysis. Acta Hort. 626: 119-124.

Stafne, E.T., A.L. Szalanski, and J.R. Clark. 2004. Nuclear ribosomal ITS region sequences for differentiation of Rubus genotypes. J. Ark. Acad. Sci. 57: 176- 180.

Stallings, R.L., A.F. Ford, D. Nelson, D.C. Torney, and C.E. Hildebrand. 1991. Evolution and distribution of (GT)n repetitive sequences in mammalian genomes. Genomics. 10: 807-815.

Temnykh, S., G.D. Clerck, A. Lukashova, L. Lipovich, S. Cartinhour, and S. McCouch. 2001. Computational and experimental analysis of mirosatellites in rice (Oryza sativa L.): genetic marker potential. Genome Res. 11: 1441-1452.

Thomas, P.T. 1940. The origin of new forms in Rubus III. The chromosomal constitution of R. laganobaccus Bailey, its parents and derivatives. J Genet. 40: 141-156.

Toth, G., Z. Gaspari, and J. Jurka. 2000. Microsatellites in different eukaryotic genomes: survey and analysis. Genome Res. 10: 967-981.

USDA, ARS, National Genetic Resources Program. 2006. Germplasm Resources Information Network- (GRIN)[Online Database]. National Germplasm Resources Laboratory, Beltsville, Md, URL: .

Zane, L., L. Bargelloni, and T. Patarnello. 2002. Strategies for microsatellite isolation: a review. Mol Ecol. 11: 1-16.

CHAPTER 3

EVALUATION OF GENETIC STABILITY OF CRYOPRESERVED Rubus MERISTEMS USING SSR AND AFLP MARKERS

N. R. F. Castillo, N. V. Bassil, B. M. Reed

Abstract The genetic stability of cryopreserved Rubus shoot tips from four accessions regrown after 12 years of storage in liquid nitrogen was examined with 10 SSR markers and 10 AFLP primer pairs. The four accessions consisted of one R. grabowskii accession (CRUB 48.001), two blackberries (‘Hillemeyer’ and ‘Silvan’) and one raspberry (‘Mandarin’). No phenotypic differences were observed in 10- month old regrown cryopreserved plants when compared to the control screenhouse- grown plants. SSR and AFLP analyses were performed on two groups of plants: Group one plants were subcultured for seven months and Group two plants were analyzed immediately after recovery from cryopreservation. Ten SSR primers developed from ‘Marion’ and ‘Meeker’ microsatellite-enriched libraries were used. The 10 primer pairs amplified 1-15 alleles per locus, with an average of 7 alleles and a total of 70 alleles in the four genotypes tested. No SSR polymorphisms were observed between cryopreserved plants and the corresponding screenhouse grown plants irrespective of subculture. In Group one plants, the 10 AFLPs primer pairs produced 547 amplified fragments in R. grabowskii CRUB 48.001, 400 fragments in ‘Mandarin’, 530 fragments in ‘Silvan’ and 521 fragments in ‘Hillemeyer’. In Group two plants, the lower number of amplified fragments (331 in ‘Hillemeyer’ and 379 in ‘Silvan’) generated was mostly due to the use of AmpliTaq Gold polymerase as opposed to Platinum Taq polymerase used in Group one plants. AFLP revealed polymorphism in three out of the four Rubus genotypes subcultured for seven months while no polymorphism was detected in either genotypes analyzed immediately after recovery from cryopreservation. Recovery of plants from cryopreservation was low in the three accessions that exhibited AFLP polymorphisms (R. grabowskii, ‘Mandarin’ and ‘Silvan’). Such polymorphism might have been generated through somaclonal variantion of these plantlets if regenerated from callus tissue. Genotypic influence on stability may explain why those three genotypes were prone to variation while ‘Hillemeyer’ remained genetically stable despite long culture periods. Careful treatment and monitoring of regrown plants should therefore be employed to ensure

79 maintenance of genetic fidelity of cryopreserved plants. The variation detected may also be transient and requires further morphological and molecular analysis of adult regrown cryopreserved plants that were transplanted and are growing in the greenhouse. Key Words: genetic fidelity, in vitro culture, somaclonal variation, Taq DNA polymerase

Introduction The genus Rubus belongs to the Rosaceae family and contains cultivated raspberries, blackberries and hybrid berries and a large number of species (Jennings, 1988). Raspberry and blackberry fruits are produced for the fresh fruit market and for use in a number of processed food items (Donnelly and Daubeny, 1986) and are important worldwide. The Pacific Northwest (Oregon, Washington and British Columbia) is a major production area in the world. Many breeding programs are actively working on releasing cultivars with excellent quality, high yields, greater adaptation to adverse environmental conditions and increased pest and disease resistance. The wide diversity of Rubus species provides a potential source of novel traits. The USDA Agricultural Research Service (ARS), National Clonal Germplasm Repository (NCGR) in Corvallis, Oregon is responsible for collecting, maintaining, characterizing and distributing Rubus accessions. Rubus species and cultivars are clonally propagated and maintained in greenhouses, screenhouses, field collections (Hummer, 1987; Hummer, 1988a; Hummer, 1988b) and as tissue cultured plantlets and cryopreserved shoot tips (Gupta and Reed, 2006). The NCGR’s large and diverse Rubus collection consists of 2094 accessions representing 193 species from 64 countries and is a great source of genetically diverse genotypes for use by plant breeders and scientists across the world (Thompson, 1995). Cryopreservation is used for preservation of Rubus and involves the storage of biological materials at ultra-low temperatures usually in liquid nitrogen (-196 ˚C)

80 allowing for long-term and contamination-free storage of plant genetic resources (Engelmann, 2004). Advantages of cryopreservation include low maintenance and small storage space (Helliot et al., 2002). Cryopreservation utilizes tissue culture methods in the production and regeneration of the frozen material. As plants are cryopreserved, cell division and metabolic activities are stopped by exposing plants to ultra-low temperatures allowing plant storage without alteration and modification for an unlimited period of time. Maintenance of genetic fidelity of meristem derived plants following cryopreservation was supported by RAPD studies in Vitis and Actinidia, (Zhai et al., 2003), Arachis (Gagliardi et al., 2003), Pinus sylvestris L embryogenic cultures (Haggman et al., 1998) and Prunus (Helliot et al., 2002); AFLP studies in Prunus (Helliot et al., 2002), Diospyros virginiana L. (Ai and Zhengrong, 2005) and Anigozanthos viridis Endl. (Turner et al., 2001) and SSR study in Solanum tuberosum L. (Harding and Benson, 2001). However, since cryopreservation utilizes in vitro methods, associated risks to the maintenance of the genetic fidelity in stored plants may be present. Changes occurring during propagation in tissue culture are either epigenetic or genetic (Larkin and Scowcroft, 1981). Genetic changes are heritable and are known as somaclonal variation which may result in modification of chromosome number or methylation pattern, chromosome breakage, transposon activation, deletion, genome rearrangement, polyploidy or nucleotide substitution (Bhatia et al., 2005). Variations may pre-exist in the natural population of plants from field collection or genebank or it may be developed de novo as a result of the tissue culture technique. Somaclonal variation is described as mutations that occur spontaneously in tissue culture with increasing frequency with increasing subculturing and is well documented (Larkin and Scowcroft, 1981). In a study by Dixit et al., (2003) on Dioscorea bulbifera L., amplified fragment patterns of plants regenerated from cryopreserved embryogenic tissues were identical to those of in vitro grown control plants for 9 of 10 primer pairs tested. One band was variable in one cryopreserved plant out of the 4960 bands

81 produced by the 10 primers. However, the extremely low frequency of variation (0.02%) detected was reported to arise during the induction and maintenance of embryogenic tissues through repeated subculturing before cryogenic treatments (Dixit et al., 2003). In Saccharum officinarum L. meristem culture, tissue culture was found to be responsible for the generation of variation with a 7-fold increase in the rate of polymorphism observed with the variation also showing up as phenotypic differences (Zucchi et al., 2002). It is clear in studies that cryopreservation did not cause stress nor produce genetic variation but pre- or post-cryopreservation procedures may do. Maintenance of genetic fidelity is essential for a successful cryopreservation strategy emphasizing the need for genetic stability evaluation of cryopreserved plants. The development of molecular techniques in recent years provided additional means for assessing genetic fidelity in plants. The objective of this study was to use Simple Sequence Repeat (SSR) and Amplified Fragment Length Polymorphism (AFLP) markers to evaluate genetic stability of regrown cryopreserved Rubus stored in liquid nitrogen for more than ten years. SSRs are tandemly repeated motifs of one to six bases in most prokaryotic and eukaryotic genomes analyzed to date (Zane et al., 2002). They are found in coding and non-coding regions and are highly polymorphic. SSRs were utilized to assess genetic fidelity of post-cryopreserved Solanum plants (Harding and Benson, 2001) and identify somaclonal variation in tissue cultured Actinidia (Palombi and Damiano, 2002), Theobroma (Rodriguez et al., 2004), and Populus (Rahman and Rajora, 2001). SSR analysis was utilized in Solanum to detect DNA sequence length variation in encapsulation-dehydration cryopreserved cultivars, ‘Brodick’ and ‘Golden Wonder’. Identical SSR profiles were observed in plants regrown from cryopreserved apices, parental plants and their progeny (Harding and Benson, 2001). This indicates the stable somatic inheritance of the genomic sequences with the SSRs. SSRs are also capable of detecting somaclonal variation as seen in Actinidia (Palombi and Damiano, 2002). One SSR marker showed genetic variation between in vivo and in vitro plants

82 of the cultivar ‘Tomuri’. Somaclonal variation was also detected in Populus at two SSR loci (Rahman and Rajora, 2001). Variation at the PTR2 locus resulted in the appearance of a new allele of increased size while variation at the PTR5 locus resulted in the appearance of a third allele. The most important advantage of the use of AFLP is its capability to inspect an entire genome for polymorphism and its reproducibility (Blears et al., 1998). The ability to reveal many polymorphic bands makes AFLP a very efficient technique. AFLP was utilized to assess genetic fidelity of several species like Fragaria and Anigozanthos after cryopreservation (Hao et al., 2002; Turner et al., 2001). AFLP analysis of cryopreserved Prunus plants and the non-frozen control revealed two polymorphic fragments of a 135 bp fragment and a missing 90 bp fragment (Helliot et al., 2002). The frequency of the polymorphic pattern increased from 18% for the non- frozen plants to 37% from the in vitro plants regrown from cryopreserved apices. In citrus callus cultures, AFLP was performed on cryopreserved and non-cryopreserved single-cell sibling lines (Hao et al., 2002). Analysis showed that the amplified fragments were identical in number and size except for the addition of one additional fragment in the cryopreserved samples. This additional fragment can be attributed to the change in DNA methylation status. The change in methylation status can be confirmed by performing the Methylation Sensitive Amplified Polymorphism (MSAP) assay that relies on the differential sensitivity of restriction enzymes to methylated DNA sequences (Harding, 2004). Since DNA markers have different efficiencies in detecting polymorphism and different levels of DNA changes, the use of more than one marker should be more reliable during genetic stability studies. In poplar, somaclonal variation was detected using RFLP, RAPDs and SSRs (Wang et al., 1996) while Goto (1998) did not find any differences using only RAPD markers in the same species. Therefore, the use of more than one molecular marker increases the reliability of genetic stability assessments.

Materials and Methods Plant Materials In vitro grown plantlets of one wild European blackberry accession of R. grabowskii Weihe ex Gunther et al. (PI 379534), two blackberry cultivars, ‘Hillemeyer’ (PI 553275), and ‘Silvan’ (PI 553308) and one R. idaeus red raspberry cultivar ‘Mandarin’ (PI 553493) were cryopreserved between 1990 and 1993 by slow cooling (Reed, 1993). The plants were regrown in 2005 and analyzed for genetic stability using morphological and DNA markers. A total of 11 regrown plants that were subcultured for 7 months (Group 1 plants) were used for molecular analysis: 4 in R. grabowskii, 5 in ‘Hillemeyer’, and 1 each in ‘Silvan’ and ‘Mandarin’ (Table 3.1). Ten plants were immediately analyzed after recovery from cryopreservation (Group two plants) and consisted of eight plants of ‘Hillemeyer’ and two plants of ‘Silvan’ (Table 3.1). Cryopreserved plants subcultured for 7 months were grown in the greenhouse for 10 months and were checked for phenotypic differences in leaf shape and spinelessness. Cryopreservation and Regrowth of Rubus Plants The slow-cooling cryopreservation protocol used between 1990 and 1993 consisted of cold-acclimating the in vitro plantlets for one week, excising 0.8 mm shoot tips and placing them on MS medium (Murashige and Skoog, 1962) with 5% dimethylsulfoxide (DMSO) for 2 days. Shoot tips in 1.2 ml cryotubes were cryoprotected with PGD (10% each polyethylene glycol, glucose and DMSO in MS medium) and cooled at 0.5 °C per min to -40 °C then plunged into liquid nitrogen. Storage was under liquid nitrogen from 1990 to 2005. To regrow the Rubus plantlets, cryovials were warmed for 1 min in 45 °C water and 1 min in 25 °C water, rinsed in MS medium and plated on recovery medium (Chang and Reed, 1999). Plantlets were micropropagated on NCGR Rubus medium to produce enough tissue for analysis (Reed, 1990).

Table 3.1. List of Rubus accessions used for genetic fidelity assessment.

Genotypes Name Crop Type Propagation Origin* 48 P R. grabowskii Wild blackberry Screenhouse 48 A R. grabowskii Wild blackberry Cryopreserved, TC 48 B R. grabowskii Wild blackberry Cryopreserved, TC 48 C R. grabowskii Wild blackberry Cryopreserved, TC 48 D R. grabowskii Wild blackberry Cryopreserved, TC 252 P Hillemeyer Blackberry cultivar Screenhouse 252 E Hillemeyer Blackberry cultivar Cryopreserved, TC 252 F Hillemeyer Blackberry cultivar Cryopreserved, TC 252 H Hillemeyer Blackberry cultivar Cryopreserved, TC 252 I Hillemeyer Blackberry cultivar Cryopreserved, TC 252 L Hillemeyer Blackberry cultivar Cryopreserved, TC 252-1 Hillemeyer Blackberry cultivar Cryopreserved, NTC 252-2 Hillemeyer Blackberry cultivar Cryopreserved, NTC 252-3 Hillemeyer Blackberry cultivar Cryopreserved, NTC 252-4 Hillemeyer Blackberry cultivar Cryopreserved, NTC 252-5 Hillemeyer Blackberry cultivar Cryopreserved, NTC 252-6 Hillemeyer Blackberry cultivar Cryopreserved, NTC 252-7 Hillemeyer Blackberry cultivar Cryopreserved, NTC 252-8 Hillemeyer Blackberry cultivar Cryopreserved, NTC 633 P Silvan Blackberry cultivar Screenhouse 633 A Silvan Blackberry cultivar Cryopreserved, TC 633-1 Silvan Blackberry cultivar Cryopreservation, NTC 633-2 Silvan Blackberry cultivar Cryopreservation, NTC 743 P Mandarin Raspberry cultivar Screenhouse 743 B Mandarin Raspberry cultivar Cryopreserved, TC *TC= analyzed after 7 month of subculture NTC= analyzed immediately after recovery

DNA Extraction Genomic DNA was extracted from regrown cryopreserved Rubus plants and four screenhouse-grown plants with the PUREGENE kit (Gentra Systems Inc., MN) using the optional RNAse A treatment followed by phenol:chloroform extraction (Sambrook et al., 1989) (Appendix A). SSR Analysis Ten primer pairs isolated from ‘Meeker’ (R. idaeus) and ‘Marion’ (R. hybrid) (Table 3.2) were used for assessment of genetic stability of regrown cryopreserved Rubus. Fluorescent forward primers were ordered from Sigma-Proligo

Table 3.2. SSR primers used in SSR analysis.

Primer Sequence Fwd: GGTTCGGATAGTTAATCCTCCC RhM001 Rev: CCAACTGTTGTAAATGCAGGAA Fwd: CCATCTCCAATTCAGTTCTTCC RhM003 Rev: AGCAGAATCGGTTCTTACAAGC Fwd: AAAGACAAGGCGTCCACAAC RhM011 Rev: GGTTATGCTTTGATTAGGCTGG Fwd: CACCAATTGTACACCCAACAAC RhM018 Rev: GATTGTGAGCTGGTGTTACCAA Fwd: CAGTCCCTTATAGGATCCAACG RhM021 Rev: GAACTCCACCATCTCCTCGTAG Fwd: CGACAACGACAATTCTCACATT RhM023 Rev: GTTATCAAGCGATCCTGCAGTT Fwd: GGACACGGTTCTAACTATGGCT RhM043 Rev: ATTGTCGCTCCAACGAAGATT Fwd: CGACACCGATCAGAGCTAATTC RiM015 Rev: ATAGTTGCATTGGCAGGCTTAT Fwd: GAAACAGGTGGAAAGAAACCTG RiM017 Rev: CATTGTGCTTATGATGGTTTCG Fwd: AGCAACCACCACCTCAACTAAT RiM036 Rev: CTAGCAGAATCACCTGAGGCTT

(Sigma-Aldrich Co., Mo.). PCR reactions were performed separately for each primer pair using a fluorescently labeled forward primer and an unlabeled reverse primer. Reactions were performed in 15 µL volumes containing 1X reaction buffer, 2 mM

MgCl2, 0.2 mM dNTPs, 0.15 µM of each primer, 0.025 U of Biolase Taq DNA polymerase (Bioline USA Inc., MA) and 3 ng genomic DNA. A 5 µL aliquot of each PCR product was reserved for fragment analysis after verification of amplification success by separating the remaining PCR product by 1.5% agarose gel electrophoresis. Fragment analysis was determined after separation on a Beckman CEQ 8000 genetic analyzer (Beckman Coulter Inc., Fullerton, CA). PCR products were loaded in 26 µL of Sample Loading Solution that contained 0.6 µL of CEQ-600 size standard and 1 µL of diluted and undiluted PCR products. The optimal amount of PCR product

86 was determined experimentally. Up to two primer pairs were multiplexed after PCR. Allele sizing and visualization were performed using the fragment analysis module of the CEQ 8000 software. Alleles were scored by fitting the peaks into bins less than one nucleotide. AFLP Analysis Genomic DNA (200 to 500 ng) was restricted with 2 µL of EcoRI and MseI at 37 °C for three hours using 1.25 U of the enzymes and 5 µL of 5X restriction buffer [50 mM Tris-HCl (pH 7.5), 50 mM Mg-acetate, 250 mM K-acetate] in a final volume of 25 µL. DNA was run on thin 1.5% agarose gel to verify complete digestion. Adaptor ligation followed by adding 24 µL of adapter/ligation solution [EcoRI/MseI adapter, 0.4 mM ATP, 10 mM Tris-HCl (pH 7.5), 10 mM Mg-acetate, 50 mM K-acetate], 1 µL of 400-1000 U T4 DNA Ligase to 24 µL of each double- digested DNA samples (25 µL final volume) and incubating at 16 °C overnight. A pre-amplification step was performed with primers complementary to the adaptor sequences and carrying an additional selective nucleotide. Pre-amplification of the adapter-ligated DNA was performed using A and C as selective nucleotides (EcoRI+ A and MseI + C respectively). A 1:10 dilution of the digested and adaptor-ligated DNA was used as a template for preselective amplification. PCR was carried out in a total volume of 25 µL containing 2.5 µL of 10X biolase buffer, 0.75 µL of 50 mM

MgCl2, 2.5 mM of each dNTP, 0.75 µL of 10 µM each of primer EcoRI+ A and MseI+ C, 0.125 µL of 5 U Biolase (Bioline USA Inc., MA) and 5 µL of DNA. The MJ thermocycler (MJ Research Inc., Reno, Nev.) was used with the following cycling parameters: 20 cycles of 30 s at 94 ˚C, 60 s at 56 °C and 120 s at 72 °C. For selective amplification, the pre-amplified DNA was diluted 50-fold with TE buffer and used as template DNA. Ten µL of each sample were checked on a thin 1.5% agarose gel where a smear is visible. EcoRI and MseI primers with three selective bases at the 3’ end were used for selective amplification. For detection, the EcoRI-based primers were fluorescently labeled with Well-Red (Sigma Aldrich Co., Mo.) fluorescent dye. Sixteen primer pair combinations or 64 primer pairs were initially tested to select for the most polymorphic set of primer pairs. The PCR amplification mixture (15 µL

final volume) consisted of 1.5 µL of 10X PCR buffer, 1.2 µL of 25 mM MgCl2, 2.5 mM of each dNTP, 0.938 µL of labeled 10 µM EcoRI+ 3 primer, 0.938 µL of unlabeled 10 µM MseI+ 3 primer, 0.075 µL of either 0.75 U Platinum Taq DNA Polymerase (Invitrogen, Co., CA) in Group 1 plants or 0.75U AmpliTaq Gold DNA polymerase (Applied Biosystems, CA.) in Group 2 plants and 3 µL of diluted pre- amplification product. Selective amplification was carried out in an MJ thermocycler using the following temperature profile: an initial denaturation step of 94 °C for 2 min; 9 cycles of 94 °C for 30 s, 65 °C for 30 s which decreases by 1°C /cycle for those 9 cycles; 72 °C for 2 min followed by 24 cycles of 94 °C for 30 s, 56 °C for 30 s and 72 °C for 2 min, with one final cycle of 72 °C for 3 min. Three µL of the selective amplification product was run on a thin 1.25% gel to confirm amplification success. Ten primer pairs that amplified 331 to 547 bands were chosen for AFLP analysis (Table 3.3). The fluorescently labeled amplified fragments were analyzed by capillary gel electrophoresis using the CEQ 8000 Genetic Analyzer (Beckman Coulter, CA). The inclusion of internal size CEQ-600 size standard in each lane enabled accurate sizing and scoring (presence/absence) of DNA fragments between 85-500 base pairs. Statistical Analysis of SSR and AFLP Data SSR and AFLP products were scored as present (1) or absent (0) to create a binary matrix. A Perl script converted the dominant data into a binary data format. Genetic distance matrices (Euclidean distance) were computed using NTSYS-PC (Numerical Taxonomic System, Exeter Software), version 2.1 (Rohlf, 2000). The Euclidean distances were calculated as follows: 2 1/2 Eij= [Σk (xki – xkj) ] th where Eij is the genetic distance between individual i and j; xki and xkj are the i band scores (1 or 0) for individuals i and j. Cluster analysis was performed on standardized data based on the Euclidean distance coefficient and unweighted pair-group method. The dendrogram was generated using the TREE sub-program of the software package NTSYS-pc.

Table 3.3. Sequences of primers and adaptors used in AFLP analysis. The three selective nucleotides in the selective primers are in bold.

EcoRI preselective primer 5’-GACTGCGTACCAATTCA-3’ MseI preselective primer 5’-GATGAGTCCTGAGTAAC-3’ EcoRI adaptor 5’-GACTGCGTACCAATTC-3’ EcoRI selective primers (+3) 5’-GACTGCGTACCAATTCAAC-3’ EcoRI selective primers (+3) 5’-GACTGCGTACCAATTCAAG-3’ EcoRI selective primers (+3) 5’-GACTGCGTACCAATTCACT-3’ EcoRI selective primers (+3) 5’-GACTGCGTACCAATTCACG-3’ EcoRI selective primers (+3) 5’-GACTGCGTACCAATTCAGC-3’ EcoRI selective primers (+3) 5’-GACTGCGTACCAATTCAGG-3’ MseI adaptor 5’-GATGAGTCCTGAGTAA-3’ MseI selective primers (+3) 5’-GATGAGTCCTGAGTAACAA-3’ MseI selective primers (+3) 5’-GATGAGTCCTGAGTAACAC-3’ MseI selective primers (+3) 5’-GATGAGTCCTGAGTAACAT-3’ MseI selective primers (+3) 5’-GATGAGTCCTGAGTAACTA-3’ MseI selective primers (+3) 5’-GATGAGTCCTGAGTAACTC-3’ MseI selective primers (+3) 5’-GATGAGTCCTGAGTAACTT-3’

Results Morphological and DNA variation The regrown cryopreserved plants were categorized into two groups based on whether they were subcultured. Group one accessions were R. grabowskii, ‘Hillemeyer’, ‘Silvan’ and ‘Mandarin’ and were subcultured for seven months before DNA extraction. Group two accessions were ‘Hillemeyer’ and ‘Silvan’ and DNA was immediately extracted after recovery from cryopreservation. Visual assessments of approximately 10-month old regrown cryopreserved plants and greenhouse-grown plants showed no gross morphological differences. Each of the Group one cryopreserved plants was the same as its corresponding greenhouse-grown genotype. All plants had the same leaf shape and presence or absence of spines. Ten SSR primers developed from ‘Meeker’ and ‘Marion’ was used in the SSR analysis. The 10 primer pairs amplified 1-15 alleles per locus, with an average of 7 alleles and a total of 70 alleles in the 4 genotypes evaluated. Primer RhM018 was

89 monomorphic and primer RhM043 was the most polymorphic of the 10 primers showing 15 alleles. SSR analysis of Group 1 and Group 2 cryopreserved plants did not show any variation from the screenhouse grown plants at the 10 SSR loci examined. Ten AFLP primer pairs were used to evaluate genetic stability in 11 Group 1 plantlets and 10 Group 2 plants. These AFLP primer pairs produced 547 amplified fragments in R. grabowskii, 400 fragments in ‘Mandarin’, 530 fragments in ‘Silvan’, and 521 fragments in ‘Hillemeyer’ Group 1 plants. An appreciably lower number of amplified fragments were generated in Group 2 plants: 331 in ‘Hillemeyer’ and 379 in ‘Silvan’. Average number of fragments per primer pair was 55 in R. grabowskii, 40 in ‘Mandarin’, 53 in ‘Silvan’, and 52 fragments in ‘Hillemeyer’ Group 1 plants. The average number of fragments per primer pair was lower in Group 2 plants: 33 in ‘Hillemeyer’, 38 in ‘Silvan’. All Group one cryopreserved plants, except ‘Hillemeyer’, showed polymorphism when compared to the screenhouse grown plants. In R. grabowskii, the primer pairs AAC-CTA, AAG-CTC and ACG-CTT revealed 12 polymorphisms when compared with the corresponding electropherogram of the screenhouse grown plant. AAC-CTA showed differences at fragments 126, 132, 166, 181, 193 and 365. Primer pair AAG-CTC was different at fragments 240, 246, 247 and 353 and ACG-CTT was different at fragments 107 and 202. There were differences between ‘Silvan’, Group one and ‘Silvan,’ screen house grown results using two AFLP primer pairs, ACG- CAC and AAG-CTC. ACG-CAC was different at fragments 88, 89, 111 and 178 and AAG-CTC was different at fragment 336. ‘Mandarin’, exhibited greatest number of polymorphisms of the four Rubus genotypes used. Primer pair ACT-CTT showed differences at fragments 114, 116, 117, 135, 146, 147, 195, 219, 248 and 270. AGG- CTT was different at fragments 90, 128, and 163 and 307. Primer pair AGC-CTA was different at fragments 108, 119, 124, 165, 167, 188, 199, 320 and 438 while ACG- CAC was different at fragments 88, 118, 125, 130, 163, 193, 198, 216, 322 and 441.

Group 2 plants did not show any variation compared to the screenhouse-grown plants with the selected 10 different primer pair combinations. Cluster Analysis The NJ cluster analysis was used to describe genetic variation between the screenhouse-grown plants and each of the cryopreserved plants. The SSR NJ dendrogram separated the four genotypes into four clades (Fig 3.1). The R. hybrid cultivars, R. grabowskii, ‘Silvan’ and ‘Hillemeyer’ were close together while the R. idaeus cultivar, ‘Mandarin’ was separated. Each clade corresponding to one genotype showed no differences within the cryopreserved plants and the screenhouse-grown plant. In the AFLP dendrogram, six clades can be observed (Fig. 3.2). “Hillemeyer’ Group one and Group two plants were separated into two groups since different PCR conditions were used. But analysis of the two clades revealed no differences within the group. Cryopreserved ‘Hillemeyer’ were the same to the screenhouse plant in both Groups one and two as indicated in the dendrogram. Different PCR conditions were also employed for analyzing Group one and two of ‘Silvan’ hence, Group one and two plants also occupied different clades. Analysis of each clade revealed that a difference existed in Group one of ‘Silvan’ but not in Group two. Within R. grabowskii, three cryopreserved plants (48A, B and 48C) were observed to be different from the screenhouse plant. In ‘Mandarin’ the cryopreserved plant was also different from the screenhouse plant as seen in the dendrogram.

Table 3.4. AFLP polymorphisms of the three Rubus accessions (Group one) when compared to their screenhouse-grown counterparts using 10 AFLP primer pairs.

R. grabowskii A B C AAC-CTA_126 -126 -126 -126 AAC-CTA_132 -132 AAC-CTA_166 -166 -166 -166 AAC-CTA_181 -181 AAC-CTA_193 -193 -193 -193 AAC-CTA_365 -365 -365 -365 AAG-CTC_240 -240 AAG-CTC_246 -246 -246 AAG-CTC_247 +247 +247 AAG-CTC_353 -353 -353 ACG-CTT_107 -107 -107 -107 ACG-CTT_202 -202 -202 -202 Silvan A ACG-CAC_88 +88 ACG-CAC_89 -89 ACG-CAC_111 +111 ACG-CAC_178 +178 AAG-CTC_336 -336 Mandarin B ACT-CAT_114 +114 ACT-CAT_116 -116 ACT-CAT_117 +117 ACT-CAT_135 +135 ACT-CAT_146 +147 ACT-CAT_147 -147 ACT-CAT_195 +195 ACT-CAT_219 +219 ACT-CAT_248 +248 ACT-CAT_270 +270 AGG-CTT_90 -90 AGG-CTT_128 -128 AGG-CTT_163 +163 AGG-CTT_307 +307 AGC-CTA_108 +108 AGC-CTA_119 +119 AGC-CTA_124 -124 AGC-CTA_165 +165 AGC-CTA_167 +167 AGC-CTA_188 +188

Table 3.4. (Continued) Mandarin B AGC-CTA_194 -194 AGC-CTA_320 +320 AGC-CTA_438 -438 ACG-CAC_88 +88 ACG-CAC_118 +118 ACG-CAC_125 +125 ACG-CAC_130 +130 ACG-CAC_163 -163 ACG-CAC_193 +193 ACG-CAC_198 +198 ACG-CAC_256 -256 ACG-CAC_322 +322 ACG-CAC_441 -441

Fig. 3.1. Neighbor Joining (NJ) dendrogram of cryopreserved and screenhouse grown Rubus plants based on Euclidean genetic distances derived from allele frequencies at 10 SSR loci.

Fig. 3.2. Neighbor Joining (NJ) dendrogram of cryopreserved and screenhouse grown Rubus plants based on AFLP data.

Discussion Visual assessment of approximately 10-month old regrown cryopreserved plants and plants growing in the screenhouse showed no phenotypic differences. This result was not unexpected since organized structures such as apical and axillary meristems produce plantlets with little or no variation among them due to their origin from pre-existing meristems (Vasil and Vasil, 1980; Wang and Charles, 1991). However since morphological markers depend on the growth stage and are easily affected by the environment, visual evaluation may not accurately reflect variations that may occur within the plants. Morphological markers require extensive observation of the plants until maturity. Furthermore, some changes induced by in vitro culture cannot be observed because the structural difference in the gene product does not always alter its biological activity to such an extent that it modifies a phenotype (Palombi and Damiano, 2002). The use of molecular makers provides more sensitive tools for the detection of variation. The use of two DNA markers in the study established a more reliable means in the assessment of variation. DNA markers detect different levels of polymorphism and different levels of DNA changes. For this reason, the use of more than one marker can increase the probability of variation detection. The use of SSR and AFLP provided certain advantages in terms of the high level of polymorphism and the capacity to evaluate a larger part of the genome. Scoring of bands for both SSR and AFLP was conservative, excluding the weak bands in the process. It is the aim of the study to obtain reproducible and clear data so that only robust and reliable bands are evaluated. In our study, no differences between screenhouse-grown and cryopreserved plants were observed in either Group 1 or Group 2 plants using 10 SSR loci. This suggests that there were no changes in the genetic fidelity of the cryopreserved plants due to cryopreservation. This was also the case in Solanum wherein microsatellite sequence of plants regrown from cryopreserved apices were identical to the profiles of the parent plants and their progeny (Harding and Benson, 2001). No

95 structural changes were observed in the in vitro control or in the Solanum plants grown from the cryopreserved germplasm indicating stable inheritance of SSR sequences in the somatic progeny (Harding and Benson, 2001). Despite being highly polymorphic and co-dominant, SSRs may be clustered and distributed unevenly in certain chromosome locations. Fisher et al. (1996) observed the presence of additional internal microsatellites for some of the ISSR markers in Pinus radiata. This observation implies that there may be clustering of microsatellites in some genomic regions. Lower coverage of the genome or a low number of primers may also fail to detect variability as in the study of Wolff et al., (1995) wherein no RAPD or SSR- PCR differences were detected between phenotypically different members of a Chrysanthemum family that were either tissue-cultured or vegetatively propagated. The results led them to suppose that they had failed to show polymorphism because of the part of the gneome studied was too reduced or the number of primers used was too low. In a study of the genetic stability of Prunus, 565 RAPD and 85 AFLP markers used were considered to be too low to cover the full plant genome and to detect any variation present in the genome (Helliot et al., 2002). Deverno (1999) also came to the same conclusion wherein identical RAPD fragments were observed in freshly thawed embryogenic cultures of Picea glauca using 10 different primers. It is also probable that the detection of somaclonal variation would require an extremely high frequency of mutation for detection. The use of a multilocus marker can increase the chances of detecting variation. AFLP is a multilocus marker and was able to detect differences in three out of the four accessions studied. AFLP polymorphism was detected in R. grabowskii, ‘Silvan’ and ‘Mandarin but not in ‘Hillemeyer’ of Group one plants. However, Group two plants ‘Hillemeyer’ and ‘Silvan’, showed no differences based on AFLP analysis. In Group one R. grabowskii only one plant remained identical to the parent while three other plants were different. In Group one, the one ‘Silvan’ and one ‘Mandarin’ cryopreserved plants were different from the screenhouse plant. Somaclonal variation is known to occur spontaneously in tissue culture with increasing frequency

96 with increasing subcultures especially in callus cultures (Larkin and Scowcroft, 1981). It is likely that the variation in Group one plants was due to the regeneration of the plants via a callus phase. This hypothesis is based on the few plants that survived and regrew following thawing. One problem encountered in cryopreservation of shoot tips is the maintenance of meristem organization and its subsequent regrowth without callus formation. Previous studies showed that blackberry and raspberry genotypes can be cryopreserved with high survival following storage but with lower percentage of shoot formation and with many genotypes producing callus (Reed, 1993b; Reed and Chang, 1997; Reed and Lagerstedt, 1987). The departure from an organized meristematic growth made the plantlets more prone to somaclonal variation. Like in cauliflower (Brassica oleracea var. botrytis L.), genetic variations were observed to take place during the callus step (Leroy et al., 2001). Genomic instability occurred during an early stage of calli formation as revealed by ISSR markers. In barley, reduced somaclonal variation was observed in plants derived from highly differentiated meristematic tissues (Bregitzer et al., 2002). Extended cold acclimation and optimization of the composition of the recovery medium was found to improve the regrowth of cryopreserved Rubus shoot tips (Chang and Reed, 1999) and reduce callus formation. The study showed that increased cold acclimation duration and reduction of IBA in the recovery medium greatly improved the regrowth of cryopreserved R. parvifolius and R. caesius meristems with less callus formation. A number of factors can be identified that influence whether or not variation is produced and how much variation is generated (Karp, 1995). These factors are the degree of departure from organized meristematic growth, genetic constitution of the starting material, growth regulators in the medium and cultivation period. The cryopreservation procedure may not cause the change, however regrowth from callus could. The variation may exist prior to cryopreservation or may be generated via tissue culture techniques (Harding, 2004). ‘Hillemeyer’ had a high regrowth rate following thawing and had no variation. The genotypic influence on somaclonal variation also explains the different susceptibility of genotypes to change. In Coffea

97 somaclonal variation frequency was observed to vary in different families (Etienne and Bertrand, 2003). The variation observed was illustrated in the dendrograms constructed for both markers. SSR dendrogram revealed the genetic stability of the cryopreserved plants for all the genotypes studied while differences existed in R. grabowskii, ‘Silvan’ and ‘Mandarin’ of Group one as shown by the AFLP dendrogram. The ability of the AFLP marker to detect variability may be attributed to its high marker or diversity index (Russell et al., 1997; Vendrame et al., 1999). It reflects the efficiency of AFLP to simultaneously analyze a larger number of bands rather than the polymorphism detected and is considered more powerful compared to RFLP, RAPDs and SSRs (Russell et al., 1997). AFLP proved to be an efficient method for the identification of Arabidopsis thaliana (Polanco and Ruiz, 2002) and pecan somatic embryos (Vendrame et al., 1999). The study in A. thaliana showed that 34 of the 51 regenerated plants showed at least 1 polymorphic AFLP marker when compared to other plants obtained from the same callus. The changes obtained may not be artifacts since the experimental procedure was repeated to confirm the changes and eliminate the possibility of DNA contamination (Polanco and Ruiz, 2002). Repetition was also done in the study to verify the robustness and the reliability of the fragments obtained. In pecan differences among tissue culture lines were detected using a single primer combination only. This shows the efficiency of AFLP to detect polymorphism not possible in other markers. In the AFLP analysis two PCR conditions were employed in the first and second Group of plants. Changes in PCR conditions also affect genetic profiles. Amplification efficiencies changed with different polymerases. Group one utilized the Platinum Taq polymerase while Group two utilized AmpliTaq Gold polymerase while the rest of the PCR parameters remained the same. Comparison of the effect of the enzyme revealed that Platinum Taq polymerase is able to amplify more bands compared to AmpliTaq Gold polymerase. In the first Group ‘Hillemeyer’ produced up

98 to 521 bands while ‘Silvan’ produced 530 bands. In the second Group, only 331 and 379 bands were produced by ‘Hillemeyer’ and ‘Silvan’ respectively. The difference was also observed in the dendrogram wherein Group one and two plants of ‘Hillemeyer’ and ‘Silvan’ formed different clades. It is not uncommon to find DNA fragment profile changes caused by different PCR reactions: DNA preparation of the starting sample, thermo-cycling machines, operators and laboratories and preparation of Taq DNA polymerase may all contribute to changes in genetic profiles (Harding, 2004). The use of Platinum Taq polymerase is therefore recommended because of the higher information content that can be obtained from the profiles that it produces. The results of the study showed the need for genetic analysis of cryopreserved plants especially after regrowth from cryopreservation. The detection of variation caused by possible regeneration from callus emphasized the need for careful treatment of the plants after regrowth from cryopreservation to ensure genetic stability. Optimization of the recovery medium must be employed to avoid callus formation and to ensure the maintenance of genetic stability of the stored plants. Analysis of variation also highlights the need for careful assessments like using more than one marker to verify differences since genome coverage is uncertain and variation may or may not be detected. Since no differences were detected morphologically between the cryopreserved and greenhouse grown plants, the possibility of transient change may had occurred during subculturing that diminished when the plants were grown in the screenhouse. To test this probability, it is also suggested that the cryopreserved plants grown in the greenhouse and field be further tested using the DNA markers. Plants should also be field grown and morphological markers be evaluated.

99 Literature Cited

Ai, P. and L. Zhengrong. 2005. Cryopreservation of dormant vegetative buds and genetic stability analysis of regenerated plantlets in persimmon. Acta Hort. 685: 85-92.

Bhatia, P., N. Ashwath, T. Senaratna, and S. Krauss. 2005. Genetic analysis of cotyledon-derived regenerants of tomato using AFLP markers. Current Sci. 88: 280-284.

Blears, M.J., S.A. deGrandis, H. Lee, and J.T. Trevors. 1998. Amplified fragment length polymorphism (AFLP): a review of the procedure and its applications. J Industrial Microbiology Biotechnol. 21: 99-114.

Bregitzer, P., S. Zhang, M. Cho, and P. Lemaux. 2002. Reduced somaclonal variation in barley is associated with culturing highly differentiated, meristematic tissues. Crop Sci. 42: 1303-1308.

Chang, Y. and B.M. Reed. 1999. Extended cold acclimation and recovery medium alteration improve regrowth of Rubus shoot tips following cryopreservation. CryoLetters. 20: 371-376.

Donnelly, D.J. and H.A. Daubeny. 1986. Tissue culture of Rubus species. Acta Hort. 183: 305-314.

Engelmann, F. 2004. Plant cryopreservation: progress and prospects. In Vitro Cell Dev Biol- Plant. 40: 427-433.

Etienne, H. and B. Bertrand. 2003. Somaclonal variation in Coffea arabica: effects of genotype and embryogenic cell suspension age on frequency and phenotype of variants. Tree Physiol. 23: 419-426.

100

Fisher, P.J., R.C. Gardner, and T.E. Richardson. 1996. Single locus microsatellites isolated using 5' anchored PCR. Nucl Acid Res. 24: 4369-4371.

Haggman, H., L. Ryynanen, T. Aronen, and J. Krajnakova. 1998. Cryopreservation of embryogenic cultures of Scots pine. Plant Cell Tissue Organ Cult. 54: 45-53.

Hao, Y., C. You, and X. Deng. 2002. Analysis of ploidy and the patterns of amplified fragment length polymorphism and methylation sensitive amplified polymorphism in strawberry plants recovered from cryopreservation. CryoLetters. 23: 37-46.

Harding, K. 2004. Genetic integrity of cryopreserved plant cells: a review. CryoLetters. 25: 3-22.

Harding, K. and E.E. Benson. 2001. The use of microsatellite analysis in Solanum tuberosum L. in vitro plantlets derived from cryopreserved germplasm. CryoLetters. 22: 199-208.

Helliot, B., D. Madur, E. Dirlewanger, and M.T.D. Boucaud. 2002. Evaluation of genetic stability in cryopreserved Prunus. In Vitro Cell Dev Biol -Plant. 38: 493-500.

Hummer, K. 1987. New accessions at the repository. Newsletter. October 1987. National Clonal Germplasm Repository, Corvallis.

Hummer, K. 1988a. New accessions at the repository. Newsletter. April, 1988. National Clonal Germplasm Repository, Corvallis.

Hummer, K. 1988b. New accessions at the repository. Newsletter. November, 1988. National Clonal Germplasm Repository, Corvallis.

Jennings, D.L. 1988. In: Raspberries and Blackberries: Their Breeding, Diseases and Growth. Academic Press Inc., USA.

101

Karp, A. 1995. Somaclonal variation as a tool for crop improvement. Euphytica. 85: 295-302.

Larkin, P.J. and W.R. Scowcroft. 1981. Somaclonal variation: a novel source of variability from cell cultures for plant improvement. Theor Appl Genet. 60: 197-214.

Leroy, X.J., K. Leon, J.M. Hily, P. Chaumeil, and M. Branchard. 2001. Detection of in vitro culture-induced instability through inter-simple sequence repeat analysis. Theor Appl Genet. 102: 885-891.

Murashige, T. and F. Skoog. 1962. A revised medium for rapid growth and bio-assays with tobacco tissue cultures. Physiol Plant. 15: 473-478.

Palombi, M.A. and C. Damiano. 2002. Comparison between RAPD and SSR molecular markers in detecting genetic variation in kiwifruit (Actinidia deliciosa A. Chev). Plant Cell Rpt. 20: 1061-1066.

Polanco, C. and M.L. Ruiz. 2002. AFLP analysis of somaclonal variation in Arabidopsis thaliana regenerated plants. Plant Sci. 162: 817-824.

Rahman, M. and O. Rajora. 2001. Microsatellite DNA somaclonal variation in micropropagated trembling aspen (Populus tremuloides). Plant Cell Rpt. 20: 531-536.

Reed, B.M. 1993. Responses to ABA and cold acclimation are genotype dependent for cryopreserved blackberry and raspberry meristems. Cryobiology. 30: 179-184.

Reed, B.M. and H.B. Lagerstedt. 1987. Freeze preservation of apical meristems of Rubus in liquid nitrogen. HortScience. 22: 302-303.

Rodriguez, C.M., A.C. Wetten, and M.J. Wilkinson. 2004. Detection and quantification of in vitro-culture induced chimersim using simple sequence repeat (SSR) analysis in Theobroma cacao (L.). 110: 157-166.

102

Rohlf, F.J. 2000. NTSYS-PC Numerical Taxonomy and Multivariate Analysis System. Version 2.1, Exeter Publ., NY.

Rout, G.R., P. Das, S. Goel, and S.N. Raina. 1998. Determination of genetic stability of micropropagated plants of ginger using Random Amplified Polymorphic DNA (RAPD) markers. Bot Bul Acad Sinica. 39: 23-27.

Thompson, M.M. 1995. Chromosome numbers of Rubus species at the National Clonal Germplasm Repository. HortScience. 301: 1447-1452.

Vasil, I.K. and V. Vasil. 1980. Clonal Propagation Int. Rev. Cytol. (Suppl 11A). 145- 173.

Vendrame, W.A., G. Kochert, and H.Y. Weitzstein. 1999. AFLP analysis of variation in pecan somatic embryos. Plant Cell Rpt. 18: 853-857.

Wang, P.J. and A. Charles. 1991. Micropropagation through meristem culture. In: High technology and micropropagation. Biotechnology in Agriculture and Forestry. Vol. 17. Springer Berlin, Heidelberg, NY.

Wolff, K., E. Zietkiewicz, and H. Hofstra. 1995. Identification of chrysanthemum cultivars and stability of DNA fingerprint patterns. Theor Appl Genet. 91: 439- 447.

Zane, L., L. Bargelloni, and T. Patarnello. 2002. Strategies for microsatellite isolation: a review. Mol Ecol. 11: 1-16.

103

Zucchi, M.I., H. Arizono, V.A. Morais, and M.H.P. Fungaro. 2002. Genetic instability of sugarcane plants derived from meristem cultures. Genet Mol Biol. 25: 91- 96.

104

CHAPTER 4

CONCLUDING REMARKS

N. R. F. Castillo

105

Conclusion The importance of the utilization of DNA markers in the management of the Rubus collection at the National Clonal Germplasm Repository (NCGR) was established in the study. SSR markers are useful in various aspects of molecular genetic studies and are superior to other techniques for fingerprinting or varietal identification because of their high polymorphism and somatic stability (Varshney et al., 2005). This superiority was demonstrated in the study wherein the SSRs developed accurately identified various raspberry and blackberry cultivars stored at the NCGR. It proved useful in a number of ways to improve the conservation and management of Rubus genetic resources. In particular, the genetic diversity data obtained is helpful in the identification of duplicate accessions, verification of synonyms and homonyms and determination of misidentifed cultivars. The SSRs can also be utilized in the determination of gaps in terms of coverage in gene pools to ensure the maximum capture of genetic diversity. The SSRs developed in the study also answered the need for additional SSR markers needed for the genetic mapping of Rubus. The developed SSR markers add to the limited number of SSRs isolated from red raspberry (Graham et al., 2004; Graham et al., 2002), from an Azorean island endemic species R. hochstetterorum Seub., (Lopes et al., 2006) and from the invasive weed R. alceifolius Poir. (Amsellem et al., 2001a). This is also the first report on isolation of SSR markers from blackberry. The information obtained in the study can help in the development of molecular breeding in Rubus which is at its infancy. The evaluation of the storage methods in the Repository by the use of DNA markers also shed light on the reliability of the various storage methods currently employed. DNA markers can complement morphological assessment traditionally used for evaluation. The use of DNA markers enabled a fast and easy detection of variation that may occur during or after storage. The use of more than one DNA marker enabled wider genome coverage hence, enabling a more dependable genetic assessment of variability. Since the maintenance of genetic fidelity is critical, careful treatment of plants after recovery from cryopreservation is crucial for avoiding

106 somaclonal variation. Improvement of in vitro procedures to avoid somaclonal variation may increase the reliability of the storage methods employed.

107

BIBLIOGRAPHY

Ahmad, F. and S. Southwick. 2003. Identification of pistachio (Pistachia vera L.) nuts with microsatellite markers. J Amer Soc Hort Sci. 128: 898-903.

Ahmad, R., D. Potter, and S.M. Southwick. 2004. Genotyping of peach and nectarine cultivars with SSR and SRAP molecular markers. J Amer Soc Hort Sci. 129: 204-210.

Ai, P. and L. Zhengrong. 2005. Cryopreservation of dormant vegetative buds and genetic stability analysis of regenerated plantlets in persimmon. Acta Hort. 685: 85-92.

Araujo, L.G., A.S. Prabhu, and M.C. Filippi. 2002. Identification of RAPD marker linked to blast resistance gene in a somaclone rice cultivar Araguaia. Fitopatol. Bras. 27: 181-185.

Baldwin, B.G. 1992. Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: an example from the Compositae. Mol Phylogenetics Evol. 1: 3-16.

108

Bassil, N.V., R. Botta, and S.A. Mehlenbacher. 2005. Microsatellite markers in hazelnut: isolation, characterization, and cross-species amplification. J Amer Soc Hort Sci. 130: 543-549.

Becher, S.A., K. Steinmetz, K. Weising, S. Boury, D. Peltier, J.P. Renou, G. Kahl, and K. Wolff. 2000. Microsatellites for cultivar identification in Pelargonium. Theor Appl Genet. 101: 643-651.

Benson, E., P. Lynch, and G. Stacey. 1998. Advances in plant cryopreservation technology: current applications in crop plant biotechnology. Agbiotech News and Info. 10: 133-140.

Bhatia, P., N. Ashwath, T. Senaratna, and S. Krauss. 2005. Genetic analysis of cotyledon-derived regenerants of tomato using AFLP markers. Current Sci. 88: 280-284.

Boches, P. 2005. Microsatellite marker development and molecular characterization in highbush blueberry (Vaccinium corymbosum L.) and Vaccinium species, Oregon State University, Corvallis.

Bordallo, P., D. Silva, J. Maria, C.Cruz, and E. Fontes. 2004. Somaclonal variation on in vitro callus culture potato cultivars. Hort Brasileira. 22: 300-304.

Bregitzer, P., S. Zhang, M. Cho, and P. Lemaux. 2002. Reduced somaclonal variation in barley is associated with culturing highly differentiated, meristematic tissues. Crop Sci. 42: 1303-1308.

109