‘2 ,- ,-I'I_’>~Aw-_ ’ .- _. 4va 2:11.: _ I'vvvzwn< .fll'r’l 'V I .I, vv-u 1.31.11.— ,1- £511" u-‘vv «ax-r1 var—1w. —, ~ "in... ,pi-n on- ur- ’5-.. -t 3‘ v _. §_1'121g11‘1'.31IE'-. 1‘1";'1"151~31'1“;‘1111'-‘ ”2'15“. 'I éII‘III' 1'1! -'1=,'11'-1:;.4'1I‘17I1:':",." "1". “11-4.. 1 1‘ {'11 '2‘ _-._.—.‘ . A__ ,..4. ”2,... .- w .12; 7 4-. <-_ n.- , 1.11 . 1 ""'-I ‘12}:11‘1 v . o. l . ._ 71,; , “3"”: ' ”flips? 2 ’ $1.”. 1‘ 1;. . tI’I 'III“ ,.1 1‘! 1'“ ' u r v' "I' '2 u. .11'3191‘1' ’ frfi" v.1 '1'“ :‘I I 3.! Iv “II 3 I 1~1~II1 212".II"".1-':- '13 '5‘". V“. I. ' l‘il111‘1311'1 13"“ '1' If! - 1 . .:I-21‘1.‘!.." .'11‘II"1 9" 1’91 '. 1 1.1'-':..'. . . 071-1.; ‘1 1:1'. " 1 ;' " '1‘“. . 1'12 '1“-‘11‘.:' 11511.2 1" 1» 1' = I- 113131111; ‘ ‘1‘11'1-11‘1112 1,111 ‘3“: 1‘l' I ' “ ‘11“ - {?1I1 2: 1.11.'1I~I1’}. 1 “(1121;13' - (-2 :E' ~-11'1°'I" '2“'1I'1 i: '11.:111‘” " .‘2'1 ‘1" I1 I1. ‘ 111.11" 2""r'.:‘31‘1-‘ v'-"'I"I'1"",IEII'min"???" . .11. '1'?“ I1; ,2. 1, 1.11:] ”11113121“ 11,, 1h” 4111111119 ‘1 '1'1’ ‘13' “11* 1'6"" 2‘ (I21. ' “'15:“; 11' 1 11M» :1 1'1 1‘ 1, '1¢‘1"""2i"1'1IIII"..I.11111;"I" 1""211‘11'. .1. . m211111'11111..2.-l111.111.-III '4? 1"" I '11 "‘11' I L' I I 'I'II'I"l ' 11I'W1'IyuIIIIII'IL'I'g 1 "1II':'I"1"":I'IIII1Z‘IIII'é'I'1"flap?"I":3“?th 1,1111” 131,111 - 1‘1".“ 11:1'111111'. I 1 ifthII'fl 11",!» 'I'11I12' It: 'I'r'IiI'; 11.11112, 111','.'I{"I11:_§i‘; ' "-' I11_,;1':II2[§,111I '1' ' ,I"- "“"1' 1‘19” 1' 'J " ' 'l 11'1‘1151.I'1I'1"1”1'I'IIi," 31:11:" I11" 311111" " 113,1I1113. 1 In '1""'I 11"‘"1”1i“1'1~I-' 1" . 1"1‘141. “1’211', ."1I'1':" II‘Ifl‘IA "'J "I'1I"'|""""1?IIIH'I"I'I'I""'E"‘"1"III'I',."""'"""'1""I"II'"""" 'I. '1") """-I'1' 1.911 ‘I' 'I'd'g‘r' I “I“ "J‘. “at.“ I": :1IJI'|V: "fill 613811 .1III',, III] '121'1 "'1', 11' I I'IIIILIW1’1‘1’ I [1 “ ' 1'11. {111% '1" 7:212 ‘ J"I1'1'1'111 ' 1 1 2'-.111.' I1I -|'1I1'I' 'II'.‘ " "'I'L'TI' I191 'I'Il'n' 2 " , ""' ' 2. "I- 111 ...'II"| ‘111112‘2‘1. ”'1'111‘111I‘1. 1"1'I'1'I|'1Im"1 1.1" II” '1111'2'" IqIII'IIhI'}! I”! 111m? {1291' III'I'I""1'1'1.'1III W2. 1.111% 1'11'11III'IIIQ 1 "1'13" 1'5 “1“I1‘1511 III '1'I‘1'1I'1I1L'1'11..II.."II2I .2II'11.. 2I.21."'1'.'."1'1II"I"‘ THESiS Z illlllilllilllllllllllHHllllllHHIIHIIIIIIllllllllilillllll 3 1293 01581 5693 LIBRARY Michigan State University This is to certify that the ‘ _ dissertation entitled S'eJ‘OOnIC D \‘dercnfht1-ion an EVO'HF:°’\ Examinailan '01" 5"“ fi'aiwcw" Hykrfid of: Dif’Ir/az/ i/aammm cur/“ow, anal Terraplo-‘(I V. (a17n7bt’5wn‘ in Btwebéfl’)’ 1 An presented by Lu? 3 n f) (2 a has been accepted towards fulfillment; . . . of the requirements for E“ 0 degree in E'OI‘i lli'ch’d/"j (ind figment 5‘ I Major professor Date JillLJL—_——\ MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 PLACE ll RETURN BOXto removethie checkout from your record. To AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE MSU ieAn Affinnetive Action/Bond Opportunity lnetituion mm: GENOMIC DIFFERENTIATION AND EVOLUTION IN BLUEBERRY: AN EXAMINATION OF AN INTERSPECIFIC HYBRID OF DIPLOID VACCINI UM DARROWI AND TETRAPLOID V. COR WBOSUM By Luping Qu A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Horticulture Plant Breeding and Genetics Program 1997 James F. Hancock ABSTRACT GENOMIC DIFFERENTIATION AND EVOLUTION IN BLUEBERRY: AN EXAMINATION OF AN INTERSPECIFIC HYBRID OF DIPLOID VACCHVI UM DARROWI AND TETRAPLOID V. COR YMBOS UM By Luping On A series of experiments were conducted to study the genomic difi'erentiation and evolution of blueberry using an interspecific hybrid, US 75, derived from Fla 4B [ a wild selection of Vaccinium darrowi (2x)] x ‘Bluecrop’ [(a cultivar of V. corymbosum (4x)]. First, the inheritance pattern of US 75 and the mode of Fla 4B’s 2n gamete ‘ production were determined using RAPD markers. US 75 was found to contain about 70% of Fla 4B’s heterozygosity, a value attributed to a first division restitution (FDR) mode of . 2n gamete production. Crossovers during 2n gamete formation were evidenced by the absence of 33 dominant alleles of Fla 4B in US 75, and direct tests of segregation in a diploid involving Fla 4B. RAPD markers that were present in both Fla 4B and US 75 were used to determine the mode of inheritance in a tetraploid segregating population of US 75 x V. corymbosum cv. ‘Bluetta’. Sixty-five duplex loci were identified which segregated in a 5:1 ratio, indicating US 75 undergoes tetrasomic inheritance. Second, a genetic linkage map of US 75 was generated. One hundred and forty markers tmique for Fla 48 that segregated 1:1 in the tetraploid population were mapped into 29 linkage groups that cover a total genetic distance of 1288.2 cM, with a range 1.6 - 33.9 cM between adjacent markers. The map is essentially of V. darrowi because US 75 was produced via a 2n gamete from Fla 4B and only unique markers for Fla 48 were utilized. Third, the meiotic pairing configurations and genomic similarity of US 75 and its parents were studied using cytogenetic observations and genomic in situ hybridization (GISH). US 75 was found to have a lower than expected number of multivalents for an autopolyploid, but it had a significantly higher number of quadrivalents than V. corymbosum, and one PMC (pollen mother cell) was observed with 11 of all 12 homologous groups being paired in quadrivalents. Normal distributions of chromosomes were observed at anaphase I and II by both cytological observation and pollen viability. 6181-! revealed that both parents labeled all the chromosomes in the hybrid. These findings suggest that little genomic divergence has deve10ped between the Vaccinium species, and that genes may be freely transferred from diploid to tetraploid species via unreduced gametes. Finally, four new mechanisms of Zn pollen formation were found in blueberry by examining the cytogenetics of PMCs and sporad configurations of selfed progenies of US 75. These were: 1) Premeiotic doubling (PD), where chromosome doubling in a PMC during mitosis resulted in the formation of four 2n gametes; 2) Tripolar spindle (TPS), where the spindles at anaphase II were fused at one pole resulted in one 2n and two n gametes; 3) Incomplete spindle at anaphase I (IS 1 ), where the spindle functioned briefly at anaphase I and then failed producing two 2n gametes; and 4) Incomplete spindle at anaphase II in one daughter nuclei (182-1), where the spindle at anaphase II in one daughter nuclei functioned briefly and then failed resulting in one 2n and two 11 gametes. The TPS and 181 produced FDR 2n pollen. The allelic constitution of gametes formed by PD is similar to FDR produced gametes, although the cytological process is different. The 182-1 mechanism resulted in SDR (second division restitution)2n pollen. Surveys of mature pollen from the selfed progeny, its parent and a tetraploid grandparent revealed sporad associations consistent with all these mechanisms. ACKNOWLEDGMENTS I thank my committee members: Drs. James F. Hancock, Amy F. Iezzoni, James D. Kelly and David Douches for the thoughtful reviews of my manuscripts and dissertation and helpful advice during the course of my study here at Michigan State University. I thank Dr. Joanne Whallon, Peter W. Callow, Dechun Wang and Roger May for their support and valuable discussion. I-must especially express my gratitude to my major professor Dr. Hancock for his great support in every possible way over the years. From helping improvement of my English to advising research his input has been in every step of my progress, and some great help I am unable to express in words. Thanks Jim! Last I thank my wife J inghua and daughter Jing for their love, understanding and support throughout the course of this study. iv Guidance committee: The thesis is divided into four chapters. Chapter 1 has published in Theoretical and Applied Genetics. Chapter 2 is in press in the Journal of American Society for Horticultural Science. Chapter 3 has been submitted to American Journal of Botany. Chapter 4 has been submitted to Theoretical and Applied Genetics. TABLE OF CONTENTS LIST OF TABLES .............................................................................................. LIST OF FIGURES ............................................................................................ GENERAL INTRODUCTION ........................................................................... AUTOPOLYPLOIDS ......................................................................................... POLYPLOIDY IN BLUEBERRY ...................................................................... LITERATURE CITED ....................................................................................... CHAPTER 1: NATURE OF 2N GAMETE FORMATION AND MODE OF INHERITANCE IN INTERSPECIFIC HYBRIDS OF DIPLOID VACCINIUM DARROWI AND TETRAPLOID V. CORYWOSUM ................ ABSTRACT ........................................................................................................ INTRODUCTION .............................................................................................. MATERIALS AND METHODS ........................................................................ RESULTS AND DISCUSSION ......................................................................... CONCLUSIONS ................................................................................................. LITERATURE CITED ....................................................................................... CHAPTER 2: RAPD-BASED GENETIC LINKAGE MAP OF BLUEBERRY DERIVED FROM AN INTERSPECIFIC CROSS BETWEEN DIPLOID VA CCINI UM DARROWI AND TETRAPLOID V. COR YMBOS UM ........................................................................................... ABSTRACT ........................................................................................................ INTRODUCTION .............................................................................................. MATERIALS AND METHODS ........................................................................ RESULTS AND DISCUSSION ......................................................................... LITERATURE CITED ....................................................................................... vi viii ix 1 2 7 14 20 21 22 25 28 55 56 59 60 61 63 65 70 TABLE OF CONTENTS (continued) CHAPTER 3: EVOLUTION IN AN AUTOPOLYPLOID GROUP DISPLAYING PREDOMINANTLY BIVALENT PAIRING AT MEIOSIS: GENOMIC SIMILARITY OF DIPLOID VACCINIUM DORROWI AND TETRAPLOID V. CORYMBOSUM ................................................................. 75 ABSTRACT ........................................................................................................ 76 INTRODUCTION .............................................................................................. 77 MATERIALS AND METHODS ........................................................................ 80 RESULTS ........................................................................................................... 84 DISCUSSION ..................................................................................................... 94 LITERATURE CITED ....................................................................................... 100 CHAPTER 4: FOUR MECHANISMS OF 2N POLLEN FORMATION IN INTERSPECIFIC BLUEBERRY (VACCINIUM) HYBRIDS .......................... 106 ABSTRACT ........................................................................................................ 107 INTRODUCTION .............................................................................................. 108 MATERIALS AND METHODS ........................................................................ 110 RESULTS ........................................................................................................... 1 1 1 DISCUSSION ..................................................................................................... l 14 LITERATURE CITED ....................................................................................... 119 vii LIST OF TABLES Chapter 1 RAPD markers that fit a 1:1 (present : absent) segregation ratio in a progeny population of US 75 x ‘Bluetta’ .................................................... RAPD markers that fit a 5:1 (present : absent) segregation ratio in a progeny population of US 75 x ‘Bluetta’ .................................................... RAPD markers which did not statistically fit a 5:1 or 1:1 segregation ratio .............................................................................................................. RAPD markers which were present in Fla 4B and US 75, but absent in the progeny of US 75 x‘Bluetta’ ................................................................ Primers which produced amplification products in Fla 4B that were absent in US 75 ............................................................................................ Segregation ratios in the diploid population Fla 48 x NC84 6-5 for the RAPD markers which segregated 5:1 or had skewed segregation ratios in thetetraploid population US 75 x ‘Bluetta’ ................................................. Chapter 3 Chromosome pairing configurations at diakinesis and metaphase I ........... viii 29 36 41 46 50 91 LIST OF FIGURES Chapter 1 DNA amplification profiles in the tetraploid progeny population US 75 x ‘Bluetta’ using primer OPSl3. The arrow shows a RAPD markers segregating in a 1:1 ratio (present : absent). From right to left, first lane is a 123 bp DNA ladder, lanes 2-4 are Fla 4B, ‘Bluecrop’, US 75 and ‘Bluetta’, respectively. The remaining lanes are from 23 progeny ....................................................................................................... 35 DNA amplification profiles in the tenaploid progeny population US 75 x ‘Bluetta’ using primer OPVO6. The arrow shows a RAPD marker segregating in a 5:1 ratio (present : absent). From right to left, first lane is a 123 bp DNA ladder, lanes 2-4 are Fla 4B, ‘Bluecrop’, US 75 and ‘Bluetta’, respectively. The remaining lanes are from 23 progeny ........................................................................................................ 40 DNA amplification profiles in the tetraploid progeny population US 75 x ‘Bluetta’ using primer OPIO9. The arrow shows a RAPD marker that was present in Fla 4B and US 75, but absent in the population . From right to left, first lane is a 123 bp DNA ladder, lanes 2-4 are Fla 4B, ‘Bluecrop’, US 75 and ‘Bluetta’, respectively. The remaining lanes are from 25 progeny .......................................................................................... 43 DNA amplification profiles in the tetraploid progeny population US 75 x ‘Bluetta’ using primer OPP17. The arrow shows a RAPD marker that was present in Fla 4B, but absent in US 75 . From right to lefi, first lane is a 123 bp DNA ladder, lanes 2-4 are Fla 4B, ‘Bluecrop’, US 75 and ‘Bluetta’, respectively. The remaining lanes are from 25 progeny ....................................................................................... - ................ 48 DNA amplification reactions in the diploid progeny population Fla 43 x NC84 6-5 using primer OPC06. The arrow shows a RAPD marker segregating in a 1:1 ratio (present : absent). This same marker segregated LIST OF FIGURES (continued) in a 5:1 ratio in the tetraploid population US 75 x ‘Bluetta’ (Figure 6). From right to left, first lane is a 123 bp DNA ladder, lanes 2 and 3 are Fla 4B and NC84 6-5, respectively. The remaining lanes are from 15 progeny ........................................................................................................ 52 DNA amplification reactions in the tetraploid progeny population US 75 x ‘Bluetta’ using primer OPC06. The arrow shows a RAPD marker segregating in a 5:1 ratio (present : absent). This same marker segregated an a 1:1 ratio in the diploid population Fla 48 x NC84 6-5 (Figure 5). From right to left, first lane is a 123 bp DNA ladder, lanes 2-4 are Fla 4B, ‘Bluecrop’, US 75 and ‘Bluetta’, respectively. The remaining lanes are from 39 progeny .................................................................................... 54 Chapter 2 RAPD-based genetic linkage map of blueberry derived from a cross of US 75 (V. darrowi x V. corymbosum) x ‘Bluetta’ (V. corymbosum x V. angustifolium). Linkage groups are numbered from longest to shortest. Marker names are shown to the left of each linkage group and the base number of the analyzed fragments is shown on the right. Distances between adjacent markers (in cM) are indicated between the brackets. The unlinked markers and sizes were: BC244-550, BC523-1720, OPC06—290, OPFO4-850, OPF12-1230, OPJ04-2700, OPJ 14-1020, OPK19-640, OPM17-1960, OPQ04-5000, OPSl3-860, OPX17-1830, OPZO4-610 and OPZO6- 610 ...................................................................... 66 Chapter 3 Somatic chromosome numbers of the tetraploids (2n = 4x = 48). A) US 75, B) ‘Bluecrop’, C) US 755, and D) CEL ................................................ 85 LIST OF FIGURES (continued) A) Somatic chromosomes of Fla 4B (2n=2x=24), and B) A metaphase I configuration with 12 ring I] ........................................................................ 86 Representatives of the meiotic chromosome configurations in the tetraploids. US 75 (A - F): A) Diplotene, exhibiting 11 quadrivalents (IV) and 2 bivalents (II, arrow), B) Late diplotene with 24 II, C) Diakinesis with6ring IV, 1 chainIV (large arrow), 1 III(small arrow), 7 II and 31 (arrow head), D) Diakinesis with 5 ring IV and 14 II, E) Late anaphase 1, showing lagging chromosomes (arrow), F) Early telophase I with 24 chromosomes/pole; ‘Bluecrop’ (G and H): G) Diakinesis with 2 IV [one possible broken ring IV (arrow)] and 20 11, H) Early anaphase I with 24 II; US 75s (I and J): I) Diakinesis with 3 ring IV and 18 II, J) Diakinesis with 24 11 with ring (large arrow), chain (small arrow), bar (large arrow head) and x (small arrow head) configurations; and CEL: K) Metaphase I with 1 IV and 22 H .................................................................................................... 88 Re-arrangement of Figure 2-B by chromosome size. Note high numbers of attachment sites in both large and small chromosomes ................................. 89 Pollen viability of US 75: A) Typical microscope field of pollen stained with acetocannine, B) A dayd pollen sporad (left) which may represent 2n gametes, and C) Typical microscope field of germinated pollen ............................................................................................................ 92 Genomic in situ hybridization. A) US 75 chromosomes hybridized by the ‘Bluecrop’ labeled genomic probe. Before it was transmitted, the color is green yellow; B) The ‘Bluecrop’ labeled genomic probe hybridized on its own chromosomes (some chromosomes were lost during the slide . treatment). Before it was transmitted, the color is yellow green; and C) US 75 chromosomes blocked with ‘Bluecrop’ genomic DNA and hybridized with labeled Fla 4B probe. Before it was transmitted, the color is brown red ................................................................................................................... 94 LIST OF FIGURES (continued) Chapter 4 Representatives of normal meiotic stages of US 755. A) Metaphase I, B) Telophase I, C) Telophase II, and D) Immature tetrad ............................ Representatives of irregular meiotic stages of US 755 which resulted in 2n pollen Formation. 1) Premeiotic doubling (PD): A) chromosome doubled before meiosis, showing 96 chromosomes, and B) A possible PD PMC at anaphase I ( arrow); 2) Tripolar spindle (TPS): C) Late anaphase II of a TPS PMC, D)Telophase II of a TPS PMC; 3) Incomplete spindle at anaphase I (181): E) chromosomes at anaphase I which were well separated by IS, but remain in one pole (two chromosomes were lost during slide preparation), F) A possible 181 PMC at telophase II; and 4) incomplete spindle in one daughter nuclei (182-1): G) Early anaphase II, chromosome at one pole undergoing normal disjunction, but the spindle at the other pole has failed, H) Another example of spindle failure in one pole during anaphase II ................................................................................. Representatives of unreduced sporads. A) Giant tetrad (left), which probably represented a PD product; B) triad with one large pollen grain (left), which Probably represented TPS product, C) Dyad, which probably represented 181 product, and D) A monad (left) ............................................ xii 112 .113 .115 GENERAL INTRODUCTION GENERAL INTRODUCTION Witt Polyploid classification, origin, genetic consequences and evolution have long been an important focus of studies in plant science. The fact that polyploidy has played an important role in both plant evolution and breeding has been recognized for at least 50 years (Stebbins, 1950). It is estimated that 30-50% of all angiosperms and more than 70% of the crop species are polyploid (Grant, 1971; Hancock, 1992), and new polyploids are being continuously created by nature and human beings. There are three major types of polyploids: 1) allopolyploids or amphipolyploids, 2) autopolyploids, and 3) segmental allopolyploids (Stebbins, 1947). It is generally considered that allopolyploids are of interspecific origin and can be recognized by their having a combination of characters of the parental species, meiotic pairing restricted to exactly two homologous chromosomes, and disomic inheritance. Autopolyploids are generated from an individual plant or different plants within the same species and are characterized by their having a phenotype similar to their parental species, random pairing among all the homologous chromosomes, and polysomic inheritance. Segmental allopolyploids are formed between partially divergent progenitors, display irregular pairing behavior between auto- and allopolyploids and mixed inheritance. Conclusions and inferences of early studies (before 1980) on polyploids were mainly based on morphological, cytological and ecological observations. Much important basic information was generated in these studies on origins (interspecific for allopolyploids, 3 intraspecific for autopolyploids), meiotic chromosome behaviors (regular bivalents for allopolyploids, variable multivalents for autopolyploids) and segregation ratios (disomic for allopolyploids, polysomic for autopolyploids); however, considerable debate was also stimulated, especially regarding the frequency and adaptation of autopolyploids ( Stebbins, 1971; Levin, 1983; Soltis and Rieseberg, 1986; Soltis and Soltis, 1993 and 1995). Autopolyploidy has been considered by a few to be a creative source of phenotypic variation in plant evolution (Levin, 1983), but most investigators have considered it to be a conservative force and a hindrance to long-term evolutionary success (Stebbins, 1950 and 1971; Briggs and Walters, 1984; Soltis and Riesberg, 1986). After reviewing the available research results from biochemical, physiological , developmental and genetical sources, Levin (1983) pointed out that autopolyploids could occupy habitats beyond the limits of its diploid progenitor. However, many investigators were convinced that autopolyploidy was maladaptive because of inherent low fertility, direct competition with progenitors, and the poor vigor of artificially produced polyploids by colchicine treatment (Sakai, 1956; Dewey, 1980; Hancock, 1996). Induced autopolyploids were usually characterized by slower development and reduced fertility (Stebbins, 1947), and autopolyploid breeding was a general failure (reviewed by Dewey, 1980; Hancock, 1996). In recent years, the application of molecular techniques has revealed a number of more positive aspects of autopolyploidy, including: 1) their having a higher level of allelic diversity than their diploid progenitors (Soltis and Rieseberg, 1986; Krebs and Hancock, 1989; Soltis and Soltis, 1993), 2) their being repeatedly formed (Soltis and Soltis, 1993), and 3) their having high fertility through the evolution of bivalent pairing at meiosis (Crawford 4 and Smith, 1984; Soltis and Rieseberg, 1986; Samuel et al., 1990; Cockerham and Galletta, 1976; Krebs and Hancock, 1989). Most researchers now believe that the origin of polyploids is generally via 2n gametes (Harlan and deWet, 1975; deWet, 1980; Bretagnolle and Thompson, 1995), and as a result, raw autopolyploids can have more heterozygosity than their diploid progenitors. In some instances, natural polyploids have even been shown to outcompete their progenitors under the same conditions (Maceira et al., 1993). It is clear that many more positive features of autopolyploidy await to be discovered. The use of molecular techniques in polyploid evolution studies began in the 1980's. Crawford and Smith (1984) surveyed 3 diploid and one hexaploid variety of Coreopsis grandiflora for 14 isozymes and found that all alleles detected in the hexaploid were also found in the diploids, and no fixed heterozygosity was found at any gene in any population of the hexaploid. They concluded that the hexaploid was an autoploid. Soltis and Rieseberg (1986) also used isozyme markers to document that T olmiea menziesii was an autopolyploid. They found there were an average of three or four alleles at each single locus and, as a result, heterozygosity was substantially higher in the tetraploid than the diploid cytotype. These genetic consequences had been predicted for autopolyploidy, but until this work, it had not been directly demonstrated (Soltis and Rieseberg, 1986). Since these studies, many more autopolyploids have been examined by molecular markers and in all cases, high levels of heterozygosity have been found (see review of Soltis and Sotis, 1993; Gutierrez et al., 1994). By determining if a polyploid has fixed heterozygosity or polysomic segregation patterns, molecular markers have revealed another important evolutionary aspect of 5 polyploidy - that predominant or even complete bivalent pairing at meiosis is not restricted to a110polyploids. Numerous studies have now shown polysomic inheritance in polyploid species that have predominant and even complete bivalent pairing (Soltis and Rieseberg, 1986; Crawford and Smith, 1984; Krebs and Hancock, 1989; Samuel et al., 1994). These findings not only challenge the traditional criteria that regular bivalent pairing signals only allopolyploidy, but also suggests that autopolyploids can achieve high levels of fertility through bivalent pairing. Indeed, autotetraploids of domesticated Vaccinium have higher pollen viability than related diploid species (Cockerham and Galletta, 1976; Krebs and Hancock, 1989), approximating 100% (Stushnoff and Hough,1968). Winge's hypothesis (Winge, 1917) that zygotic and somatic chromosome doubling were the main causes of polyploidy was prevalent until the 19703 (Bretagnolle and Thompson, 1995). However, numerous species have now been found to produce 2n gametes, and it is currently believed that the driving force behind the origin and evolution of polyploids is sexual polyploidization as a result of 2n gamete formation ( Harlan and De Wet, 1975). Autopolyploids produced via 2n gametes have two obvious advantages over those produced via somatic doubling. First, they have increased heterozygosity. No matter what mechanism generates the Zn gametes (see review of Bretagnolle and Thompson, 1995), the heterozygosity incorporated into the polyploids can not be lower through 2n gametes than somatic doubling and, in fact, FDR (first division restitution) and similar mechanisms can transfer as much as 80 to 100% of the parental heterozygosity to the autopolyploid (Peloquin, 1982; Hermsen, 1984; Vorsa and Ortiz, 1992; Bretagnolle and Thompson, 1995). Such high heterozygosity is often associated with higher plant vigor in both diploid and polyploid 6 species (T omekpe and Lumaret, 1991; Werner and Peloquin, 1991). Secondly, the unification of Zn gametes minimizes levels of inbreeding depression in the raw autopolyploid. Since somatic doubling is the most extreme case of inbreeding, it is not surprising that autopolyploids generated in this way have displayed markedly lower fertility and vigor (Hague and Jones, 1987; Dewey, 1980; Hancock, 1996). Sexual polyploidization via 2n gametes produces autopolyploids with more genetic diversity than their progenitors which can be of direct adaptive benefit, and with polysomic inheritance this higher allelic diversity is available for future differentiation. Probably the most important stumbling block in the establishment of a raw polyploid concerns competition with its already successful diploid progenitors. Several ways around this problem have been demonstrated including: 1) Partial separation (geographically or physiologically) between the new autopolyploids and their closely related diploid progenitors, minimizing the reductions in fertility caused by intercrossing between them (a reduction in the triploid block) (Borrill and Linder, 1971; Lumaret, 1985; Lumaret et al., 1987; Lumaret, 1988; Lumaret and Barrientos, 1990); 2) Chromosome doubling accompanied incidentally with biochemical, physiological, and developmental changes that immediately adapted the new polyploids to greater stress environments ( reviewed by Levin, 1983); and 3) Heterosis resulting in the autopolyploid being competitively superior (F elber, 1991). Maceira et a1. (1993) recently demonstrated using morphologically indistinguishable, sympatric diploid and tetraploid Dactylis glomerata that after two years of growth together, the tetraploid showed greater competitive ability than their diploid relatives. The autopolyploid substitution rate (4x : 2x) increased fi'om 1.8 in the first year to 3.9 in the 7 second year. The tetraploids also had heavier seeds and faster leaf production in early spring than the diploids, and they flowered earlier. This experiment was particularly important, as the plant materials were from natural populations, rather than between colchicine-induced autopolyploids and their diploid progenitors (Sakai and Suzuki, 1955; Sakai, 1956; Hagberg and Ellerstrom, 1959). As previously mentioned, such artificially formed autopolyploids have been shown repeatedly to have reduced vigor and fertility presumably due to inbreeding depression (McCollum, 1958; Borrill, 1978;). E I I . I . HI I Blueberry domestication began at the beginning of this century in the eastern US (Galletta, 1975; Galletta and Ballington, 1995). Through the efforts of breeders, dramatic improvements in terms of yield, berry quality, berry size, ripening season and geographical adaptation (Galletta, 1975; Galletta and Ballington, 1995) have been achieved. Interest in expanding blueberry production is now a worldwide goal ( Hanson and Hancock, 1990; Galletta and Ballington, 1995). Blueberries are in genus Vaccinium of the Ericaceae.They are all in the section Cyanococcus which comprises diploid, tetraploid and hexaploid species (Camp, 1945; Vander Kloet, 1983, and 1988). Three polyploids species: tetraploid lowbush, V. anguistlfolium Ait., highbush, V. corymbosum L., and hexaploid rabbiteye, V. ashei, are the foundations of the commercial blueberry industry. Early breeding efforts involved the selection of elite wild individuals and then intercrossing them for further selection. Conwntrating on only a few elite wild selections led 8 initially to the development of a very narrow germplasm base. The nuclear genome of the most widely cultivated highbush blueberries were derived primarily from only 3 native selections, 'Brooks', 'Sooy' and 'Rubel' (Hancock and Siefker, 1982), and their cytoplasms came from only 4 sources (Hancock and Krebs, 1986). The predominant rabbiteye cultivars are derived fi'om only 4 native selections, 'Myers', 'Black Giant', 'Ethel' and 'Clara' (Lyrene, 198 8). One of the negative consequences of such a narrow germplasm base is inbreeding depression (Hellman and Moore, 1983; Lyrene, 1983; Krebs and Hancock, 1988, Luby et al., 1991; Krebs and Hancock, 1990). Also, numerous traits have been noted in wild diploid species that have not been exploited in improving the current cultivars (Draper et al., 1982; Ballington, 1990; Hancock et al., 1995). The success of future breeding efl‘orts are considered to be dependent on increasing the germplasm base to broaden climatic and soil adaptation, increase yield and disease resistance, and improve fruit quality (Galletta and Ballington, 1995). There have been several recent examples where new blueberry cultivars were developed through interspecific and interploidy crosses (Lyrene, 1990; Ballington, 1990; Hancock et al., 1995). An important step in further promoting gene flow between the difl‘erent ploidies will be a better understanding of the polyploid nature of blueberry (auto vs. allo) including: 1) the degree of genomic diversity within and between species, ploidies and newly derived interspecific polyploid hybrids, 2) modes of gene exchange between the different ploidies, and 3) elucidating the role of all the aforementioned processes on the evolution of blueberry. Much progress has been made in studying the evolution of blueberry polyploids. 9 Originally, most of the tetraploid blueberries were considered to be allopolyploids (Camp, 1945; Eek, 1966; Vander Kloet, 1988) based primarily on the morphological traits. Cockerham and Galletta (1976) also suggested that V. corymbosum and four related tetraploid taxa were allopolyploids because the tetraploids as a group exhibited higher pollen stainability than a group of seven diploid species. Predominant bivalent associations at diakinesis and metaphase I were observed in all the tetraploids examined (J elenkovic and Hough, 1970; Jelenkovic and Harrington, 1971; Vorsa, 1987 and 1995) except an artificially doubled genotype of V. elloittii (Dweikat and Lyrene, 1991). The non-randomness of chromosome association of hi ghbush blueberry were interpreted as the result of obligatory pairing and localized distal chiasma (J elenkovic and Hough, 1970). However, tetrasomic inheritance ratios have been demonstrated in highbush V. corymbosum (Draper and Scott, 1971; Krebs and Hancock, 1989) and lowbush V. angustrfolium (Hokanson and Hancock, 1993), suggesting they are functionally autopolyploids. Tetraploid V. angustrfolium and V. corymbosum are also completely interfertile, indicating the group as a whole may be autopolyploid (Krebs and Hancock, 1989). Sexual hybridization via 2n gametes has been proposed as the origin of polyploid blueberries (Ortiz et al., 1992a) based on the findings that 2n gamete production is prevalent in blueberries. Unreduced pollen has been found in 2x (Ortiz et al., 1992b; Megalos and Ballington, 1988), 3x (Vorsa and Ballington, 1991), 4x (Ortiz et al., 1992a), 5x (Vorsa and Ortiz, 1992), 6x (Ortiz et al., 1992a) species and interspecific hybrids. Unreduced female gametes are also present in 2x species, evidenced by the production of tetraploid hybrids fi-om 2x x 4x crosses (Megalos and Ballington, 1988; Draper et al., 1982). 10 While 2n gamete production in blueberry is quite common and has been successfully employed in breeding (Draper et al. 1982), research on the mechanism(s) responsible for their formation are only beginning to emerge. First division restitution with no cross over (FDR-NCO) has been prOposed as the main cause of Zn pollen production in blueberries (V orsa and Ortiz, 1992; Ortiz et al., 1992b). Study on 2n pollen formation in an aneuploid hybrid [7232-1 (2n = 4x + 9 = 57)] from [ (‘Tifolue’ x ‘Dan'ow’) x ‘Rancocas’] revealed that the Zn gamete production involved three consecutive events: 1) desynapsis of paired chromosomes prior to metaphase I; 2) sister centromere disjunction in univalents at anaphase I; and 3) cytokinesis after telophase 1 leading to dyad formation (V orsa and Ortiz, 1992). Similar synaptic disorders have also been reported in 2x and 4x blueberries (personal communication of Vorsa and Ortiz in Galletta and Ballington, 1995). Filler and Vorsa (personal communication in Galletta and Ballington, 1995 ) also found parallel and tripolar spindles, as well as synaptic irregularities, in the diploid V. darrowi and V. elliottii, and tetraploid biotypes of V. pallidum. Premature cytokinesis that led to the formation of 2n pollen in the 4x highbush cultivar ‘Coville’ was reported by Stushnoff and Hough (1968). However, numerous other types 2n gamete formation have been described in other species (Bretagnolle and Thompson, 1995) that have not been investigated in blueberry. Information on genomic differentiation in blueberry is also limited, although a lack of reproductive barriers between similar ploidies has long been recognized (Camp, 1942), suggesting limited divergence. Fertile hybrids between the various ploidies of V. corymbosum are relatively easy to produce utilizing unreduced gametes (Luby et al., 1991; Ortiz et al, 1992a; Hancock et al., 1995). Cytological studies have found little chromosome 11 size difi‘erence among nine diploid species in Vaccinium, and one karyotype can be used to represent all the diploids (Hall and Galletta, 1971). Interspecific backcross hybrids of tetraploid x hexaploid crosses suggested that a minimum of two-thirds of the hexaploid chromosome complement of V. ashei could pair and recombine with that of the ten-aploid V. corymbosum (V orsa, 1987). Cytological analysis of 6 interspecific triploids derived from tetraploid x diploid V. corymbosum crosses revealed a range of chromosomal pairing relationship fiom autoploid to preferential pairing, depending on the diploid parents (V orsa, 1989). However, because the chromosomes of blueberry are very small (1.5-2.5 pm in length) and largely indistinguishable (Hall and Galletta, 1971; Vorsa, 1989), exact pairing relationships have not been resolved. Information on genomic differentiation and the nature of Zn gamete production would be particularly useful in utilizing the tetraploid hybrid, US 75, derived from a 2n gamete of Fla 4B (a selection of V. darrowi (2n = 2x = 24) and 'Bluecrop' [V. corymbosum (2n = 4x = 48). This particular hybrid is completely interfertile with highbush types (Draper et al., 1982) and has played an important role in the development of tetraploid highbush blueberries with a low chilling requirement (Ballington, 1990). It also has a number of additional elite traits that have the potential to further improve the northern tetraploid highbush blueberry (Erb et al., 1990 and 1993; Chandler et al., 1985; Hancock et al., 1992). However, little is known about the mode of Zn gamete formation in its diploid parent V. darrowi or the inheritance pattern of US 75. This information would be very useful in designing efficient breeding strategies. The type of gamete formation will determine the levels of heterozygosity transmitted to US 75, and the mode of inheritance will determine how readily that l2 heterozygosity will segregate. Furthermore, the information will be very useful in evaluating the phylogeny between the different ploidies. Two features in the background of US 75 might lead to the prediction that it would act cytogenetically as an allopolyploid with disomic inheritance rather than an autopolyploid with tetrasomic inheritance. First, the two species are quite divergent, both morphologically and geographically. V. darrowi is an evergreen, lowbush type found in the southeastern US, while tetraploid V. corymbosum is a deciduous, highbush type found in the northern US. Secondly, Fla 4B's 2n genome was incorporated into US 75 via an unreduced gamete, so both sets of homologous chromosomes are available for pairing. However, the complete interfertility of US 75 with other tetraploid highbush types suggests that V. darrowi is little diverged from that of V. corymbosum and their hybrids may be autopolyploids. The experiments in this dissertation focused on studying the genomic differentiation and evolution of blueberry using the interspecific hybrid US 75, its parents and related progeny populations. We not only determined the autopolyploid nature of US 75 employing molecular and cytogenetic methods, but also generated a genetic linkage map of US 75 and described several new mechanisms responsible for Zn gamete production in blueberry. Chapter 1 describes the inheritance patterns of RAPD markers in US 75 and the type of Zn gamete that probably produced it, by screening a population of US 75 x ‘Bluetta’ (a tetraploid cultivar). Chapter 2 presents a genetic linkage map of US 75 using the RAPD markers of chapter 1. Chapter 3 describes the meiotic pairing configurations and genomic similarity of US 75 and its parents using cytologenetic observations and genomic in situ hybridization (GISH). Chapter 4 presents four mechanisms of 2n gamete formation in selfed 13 progenies of US 75 , by examining the cytogenetics of PMCs (pollen mother cells) and sporad configutations. 14 LITERATURE CITED Ballington, J. R. 1990. Germplasm resources available to meet future needs for blueberry cultivar improvement. Frt. Var. J. 44:54-62. Borrill, M., and Linder, R. 1971. Diploid-tetraploid sympatry in Dactylis (Gramineae). New Phytol. 70:1111-1124. Borrill, M. 1978. Evolution and genetic resources in cocksfoot. pp. 190-209. Ann. Report of the Welsh Plant Breeding Sta. 1977, Aberystwyth. Bretagnolle, F., and Thompson, J .D. 1995. Gamete with the somatic chromosome number: Mechanisms of their formation and role in the evolution of autopolyploid plants. New Phytol. 129: 1-22. Briggs, D., and Walters, SM. 1984. Plant Variation and Evolution. Cambridge University Press, Cambridge. Camp, W. H. 1942. On the structure of populations in the genus Vaccinium. Brittonia 4: 1 89- 204. ’ Camp, W. H. 1945. The North American blueberries with notes on other groups of Vacciniaceae. Brittonia 5:203-275. Chandler, C.K., Draper, A.D., Galletta G.J., and Bouwkarnp, J .C. 1985. Combining ability of blueberry interspecific hybrids for growth on upland soils. HortScience 20:257-258. Cockerham, LE, and Galletta, G.J. 1976 A survey of pollen characteristics in certain Vaccinium species. J. Amer. Soc. Hort. Sci. 101:671-676. Crawford, DJ, and Smith EB. 1984. Allozyme divergence and intraspecific variation in Coreopsis graniflora (C ompositae). Syst. Bot. 9:219-225. Dewey, DR. 1980. Some applications and misapplications of induced polyploidy in plant breeding. pp.445-470. In: Lewis, W. H. (ed.). Polyploidy, Biological Relevance. Plenum Press, NY. Dweikat, I.M., and Lyrene RM. 1991. Induced tenaploidy in a Vaccinium eIIiom'i clone facilitates crossing with cultivated highbush blueberry. J. Amer. Soc. Hort. Sci. 116: 1063- 1066. Draper, A.D., and Scott DH. 1971. Inheritance of albino seedling in tetraploid highbush blueberry. J. Amer. Soc. Hort. Sci. 96:791-792. 15 Draper, A.D., Galletta, G.J., and Ballington, J .R. 1982. Breeding methods for improving southern tetraploid blueberries. J. Amer. Soc. Hort. Sci. 107:106-109. Eck, P. 1966. Botany. p. 14-44. In: P. Eek and NF. Childers (eds.). Blueberry Culture. Rutgers Univ. Press, New Brunswick, NJ. Erb, W. A., Draper, A.D., Galletta, G.J., and Swartz, HI. 1990. Combining ability for plant and fruit traits of interspecific blueberry progenies on mineral soil. J. Amer. Soc. Hort. Soc. 115:1025-1028. Erb, W.A., Draper, A.D., and Swartz, HJ. 1993. Relation between moisture stress and mineral soil tolerance in blueberries. J. Amer. Soc. Hort. Sci. 118:130-134. F elber, F. 1991. Establishment of a tetraploid cytotype in a diploid population: Efl‘ect of relative fitness of the cytotypes. J. Evol. Biol. 4:195-207. Galletta, G. J. 1975. Blueberries and Cranberries. p. 154-196. In: J. Janick and J .N. Moore (eds). Advances in Fruit Breeding. Purdue Univ. Press, West Lafayette, IN. Galletta, OJ. and Ballington, J .R. 1995. Bluebenies, cranberries, and lingonberries. p. 1-107. In: J. Janick and J .N. Moore (eds). Fruit Breeding, Volume II. Vine and Small Fruits. John Wiley & Sons, Inc. NY. Grant, V. 1971. Plant Speciation. Columbia University Press, NY. Gutierrez, J .F ., Vaquero, F., and Vences, FJ. 1994. Allopolyploid vs. autopolyploid origins in the genus Lathyrus (Leguminosae). Heredity 68:29-40. Hagberg, A., and Ellerstrom, S. 1959. The competition between diploid, tetraploid and aneuploid rye. Hereditas 45:369-416. Hague, L.M., and Jones, RN. 1987. Cytogenetics of Lolium perenne. 4. Colchicine induced variation in diploids. Theor. Appl. Genet. 79:433-439. Hall, SH, and Galletta, OJ. 1971. Comparative chromosome morphology of diploid Vaccinium species. J. Amer. Soc. Hort. Sci. 96:289-292. Hancock, J. F., and Siefker, J .H. 1982. Levels of inbreeding in highbush blueberry cultivars. HortScience 17:363-366 Hancock, J.F., and Krebs, S.L. 1986. Cytoplasms of highbush blueberry cultivars. Frt. Var. J. 40:49-50. 16 Hancock, J .F . 1992. Plant evolution and the origin of crop species. Prentice-Hall, Englewood Cliffs, New Jersey. Hancock, J.F., Haghighi, K., Krebs, S.L., Flore, J.A., and Draper, AD. 1992. Photosynthetic heat stability in highbush blueberries and the possibility of genetic improvement. HortScience. 27:1111-1113. Hancock, J.F., Erb, W.A., Goulart, B.L., and J .C. Scheerens, J .C. 1995. Utilization of wild blueberry germplasm: The legacy of Arlen Draper. J. Sm. Fr. & Vit. 3:1-16. Hancock, J .H. 1996. The colchicine story. HorTechnology. (In press). Hanson, E.J., and Hancock, J .F . 1990. Highbush blueberry cultivars and production trends. Frt. Var. J. 44:77-81. Harlan, J .R., and deWet, J .M.J . 1975. On O. Winge and a prayer: the origins of polyploidy. The Bot. Rev. 41 :361-690. Hellman, E.W., and Moore, J .N. 1983. Effect of genetic relationship to pollinizer on fruit, seed, and seedling parameters in highbush and rabbiteye blueberries. J. Amer. Soc. Hort. Sci. 108:401-405. Hermsen, J .G. 1984. Mechanisms and genetic implications of Zn gamete formation. Iowa state Jour. Res. 58:421-434. Hokanson, K., and Hancock, J.F. 1993. The common lowbush blueberry, Vaccim'um angustz‘folium Aiton, may be an autopolyploid. Can. J. Plt. Sci. 73:889-891. Jelenkovic, G., and Hough, LP. 1970. Chromosome associations in the first meiotic division in three tetraploid clones of Vaccim’um corymbosum L. Can. J. Genet. Cytol. 12:316-324. Jelenkovic, G., and Harrington, E. 1971. Non-random chromosome associations at diplotene and diakinesis in a tetraploid clone of Vaccinium australe Small. Can. J. Genet. Cytol. 13:270-276. Krebs, S.L., and Hancock, J .F . 1988. The consequences of inbreeding on fertility in Vaccinium corymbosum L. J. Amer. Soc. Hort. Sci. 113:914-918. Krebs, S.L., and Hancock, J .F. 1989 Tetrasomic inheritance of isoenzyme markers in the highbush blueberry, Vaccinium corymbasum L. Heredity. 63:11-18. Krebs, S.L., and Hancock, J .F. 1990. Early-acting inbreeding depression and reproductive success in highbush blueberry, Vaccinium corymbosum L. Theor. Appl. Genet. 79:825-832. l7 Levin, DA. 1983. Polyploidy and novelty in flowering plants. Amer. Nat. 122:1-25. Luby, J.J., Ballington, J.R., Draper, A.D., Pliszka, K, and Austin, ME 1991. Blueberries and Cranberries (Vaccinium). Acta. Hort. 290:391-456. Lumaret, R. 1985. Phenotypic and genotypic variation within and between populations of polyploid complex Dactylis glomerata L. In: Haeck, J. and Woldendorp J.W. (eds). Structure and Function of Plant Populations. Amsterdam: Elsevier/North Holland, pp.343- 354 Lumaret. R., Guillerm, J.L., Delay, J., Ait Lhaj Loutfi, A., Izco, J ., and Jay, M. 1987. Polyploidy and habitat differentiation in Dactylis glomerata L. from Galicia (Spain). Oecologia (Berlin) 73 :436—446. Lumaret, R. 1988. Adaptive strategies and ploidy levels. Oecol. Plant. 9:83-93. Lumaret. R., and Barrientos, E. 1990. Phylogenetic relationships and gene flow between sympatric diploid and tetraploid plants of Dacrylis glomerata (Gramineae). Plt. Syst. Evol. 169:81-96 Maceira, N.O., Jacquard, P., and Lumaret, R. 1993. Competition between diploid and derivative autotetraploid Dactylis glomerata L. from Galicia. Implications for the establishment of novel polyploid populations. New Phytol. 124:321-328. Lyrene, RM. 1983. Inbreeding depression in rabbiteye blueberries. HortScience 18:226-227. Lyrene, RM. 1988. Breeding rabbiteye blueberries. Plt. Breeding Rev. 5:307-357. Lyrene, RM. 1990. Low-chill highbush blueberries. Frt. Var. J. 44:82-86. Lyrene, RM. 1990. Low-chill highbush blueberries. Frt. Var. J. 44:82-86. McCollum, D.G. 1958. Comparative studies of chromosome pairing in natural and induced tetraploid Dactylis. Chromosoma 9:571-605. Megalos, 3.8., and Ballington, JR. 1988. Unreduced pollen frequencies vs. hybrid production in diploid-tetraploid Vaccim’um crosses. Euphytica 39:271-278. Newcomer, EH. 1941. Chromosome numbers of some species and varieties of Vaccinium and related genera. Proc. Amer. Soc. Hort. Sci. 38:468-470. Ortiz, R., Bruederle, L.P., Laverty, T., and Vorsa, N. 1992a. The origin of polyploids via 2n gamete production in Vaccinium section Cyanococcus. Euphytica 61:241-246. 18 Ortiz, R., Vorsa, N., Bruederle, LP, and Laverty, T. 1992b. Occurrence of unreduced pollen in diploid blueberry species, Vaccinium sect. Cyanococcus. Theor. Appl. Genet. 85:55-60. Peloquin, SJ. 1982. Meiotic mutants in potato breeding. Stadler Genet. Symp. 14:99-109. Rousi, A. 1967. Cytological observations on some species and hybrids of Vaccim'um. Ziichter 36:352-359. Sakai, KL, and Suzuki, Y. 1955. Studies on competition in plants. II. competition between diploid and autotetraploid plants of barley. J. Genet. 53:11-20 Sakai, KL 1956. Studies on competition in plants. VI. Competition between autotetraploids and their diploid prototypes in Nicotiana tabacum L. Cytologie 21: 153-156 Samuel, R., Pinsker, W., and Ehrendorfer, F. 1990. Allozyme polymorphism in diploid and polyploid populations of Galium. Heredity 65:369-378. Soltis, DE, and Rieseberg, L.H. 1986. Autopolyploidy in Tolmiea menziesii (Saxifiagaceae): Genetic insights from enzyme electrophoresis. Amer. J. Bot. 73:310-318. Soltis, DE, and Soltis, RS. 1993. Molecular data and the dynamic nature of polyploidy. Critical Rev. Plt. Sci. 12:243-237. Soltis, DE, and Soltis, PS. 1995. The nature of polyploid genome. Proc. Natl. Acad. Sci. USA 92:8089z8091. Stebbins, G.L. 1947. Types of polyploids: their classification and significance. Adv. Genet. 1:403-429. Stebbins, G.L. 1950. Variation and evolution in plants. Columbia Univ. press, N.Y. Stebbins, G.L. 1971. Chromosomal evolution in higher plants. Arnold, London. Stushnoff, C., and Hough, LP. 1968. Sporogenesis and garnetophyte development in ‘Bluecrop’ and ‘Coville’ highbush blueberries. Proc. Amer. Soc. Hort. Sci. 93:242-247. Tomekpe, K., and Lumaret, R. 1991. Association between quantitative traits and allozyme heterozygosity in a tetrasomic species: Dactylis glomerata L. Evolution 45:3 59-3 70. Vander Kloet, SP. 1983. The taxonomy of Vaccinium cyanococcus: A summation. Can. J. Bot. 61:256-266. l9 Vander Kloet, SP. 1988. The genus Vaccinium in North America. Res. Branch Agric. Can. Publ. 1828. Vorsa, N. 1987. Meiotic chromosome pairing and irregularities in blueberry interspecific backcross-1 hybrids. J. Hered. 78:395-399. Vorsa, N., and Ballington, J .R. 1991. Fertility of triploid highbush blueberry. J. Amer. Soc. Hort. Sci. 116:336-341. Vorsa, N., and Ortiz, R. 1992. Cytology of Zn pollen formation in a blueberry aneuploid (2n=4x+9=57). J. Hered. 83:346-349. Vorsa, N. 1989. Quantitative analysis of meiotic chromosome pairing in triploid blueberry: evidence for preferential pairing. Genome 33:60-67. Vorsa, N., and Novy, R. 1995. Meiotic behavior and RAPD segregation in a tetraploid hybrid (Vaccinium darrowi x V. corymbosum) suggests genomic divergence in blueberry. (Abstract). HortScience 30:808. Werner, J .E., and Peloquin, SJ. 1991. Significance of allelic diversity and Zn gametes for approaching maximum heterozygosity in 4x potatoes. Euphytica 58:21-29. Winge, O. 1917. The chromosomes. Their numbers and general importance. Carlsberg. I..ab., Copenhagen C.R. Trav. 13:232-244. CHAPTER] NATURE OF 2N GAMETE FORMATION AND MODE OF INHERITAN CE IN INTERSPECIFIC HYBRIDS OF DIPLOID VA CCINI UM DARROWI AND TETRAPLOID V. COR YMBOS UM 20 21 Alma“ RAPD markers were used to determine the level of heterozygosity transmitted via 2n gametes fi'om V. darrowi selection Florida 48 (Fla 43) to inter-specific hybrids with tetraploid V. cowbosum cv. ‘Bluecrop’. The tetraploid hybrid US 75 was found to contain about 70 % of Fla 4B's heterozygosity, a value attributed to a first division restitution (FDR) mode of Zn gamete production. Crossovers during 2n gamete formation were evidenced by the absence of 33 dominant alleles of Fla 4B in US 75, and direct tests of segregation in a diploid population involving Fla 4B. RAPD markers that were present in both Fla 4B and US 75 were used to determine the mode of inheritance in a segregating population of US 75 x V. corymbosum cv. ‘Bluetta’. Sixty-five duplex loci were identified which segregated in a 5:1 ratio, indicating US 75 undergoes tetrasomic inheritance. 22 INTRODUCTION Polyploidy has played an important role in both plant evolution and crop breeding. It is estimated that 30-50% of all angiosperms and more than 70% of the crop species are polyploid (Grant, 1971; Hancock, 1992). Breeding via 2n gametes has become an important aspect of many plant improvement programs, presumably due to the maintenance and transfer of higher levels of heterozygosity (Ballington, 1990; McCoy, 1982; Parrott et al., 1986; Ortiz and Peloquin, 1991). Most polyploids originated from sexual reproduction involving unreduced gametes (Harlan and De Wet, 1975). Unreduced gametes are formed in two primary ways: 1) an incomplete first meiotic division (first division restitution; FDR), and 2) an incomplete second meiotic division (second division restitution; SDR) (Mendiburu, 1971; Mok and Peloquin, 1975; McCoy, 1982; Hermsen, 1984). Other methods of 2n gamete production include premeiotic chromosome doubling, chromosome replication during meiotic interphase, postrneiotic chromosome doubling and apospory (V orsa and Ortiz, 1992). Unreduced gametes via FDR are comprised mainly of the non-sister chromatids of each homologous pair of chromosomes, whereas in SDR, the sister chromatids are included in the same gametes. As a result, 2n gametes formed by FDR transmit more of the parental heterozygosity into F , progenies than SDR. While few direct measurements have been made on levels of heterozygosity transmitted by SDR and FDR, theoretical calculations have been carried out. [fit is assumed that there is a regular distribution of the parental heterozygous loci along chromosomes and a single crossover per pmr of homologous chromosomes, then in the potato FDR has been estimated to transmit approximately 80% of the parental heterozygosity to progeny, while 23 SDR passes on about 40% (Hermsen, 1984). The rate at which transmitted heterozygosity ultimately assorts in polyploids is dependent on the mode of inheritance (Hancock, 1992). Segregation can be quite limited in allopolyploids with disomic inheritance, due to the maintenance of "fixed heterozygosities" on non-pairing homologous chromosomes. In autopolyploids, tetrasomic inheritance allows the variability contained in the original progenitors to segregate, but at a much slower rate than in diploids. More and more recent attention has been paid to introducing traits from diploid to polyploid blueberry cultivars via unreduced gametes (Draper, 1977; Draper etal., 1982; Ballington, 1990; Ortiz et al., 1992). Several cultivars have been released with the genes of multiple species in their background (Ballington, 1990, Hancock et al., 1995). Particular emphasis has been placed on introducing some of the elite traits of diploid V. dan'owi (high fruit quality, low chilling requirement, heat tolerance, high photosynthesis rate, drought resistance) into cultivars of tetraploid highbush blueberry, V. corymbosum. Completely interfertile F,s are relatively easy to produce between these species because some V. darrawi plants produce a high number of unreduced gametes. Ortiz et al. (1992) demonstrated that about 83% of the various diploid populations of blueberries contain unreduced pollen producers and the highest producer was a genotype of V. darrowi with a rate of almost 20%. Most of our research has focused upon US 75, a tetraploid hybrid (from Arlen Draper) generated by crossing a selection of V. darrowi, Florida 43 (Fla 4B), with a hi ghbush cultivar, ‘Bluecrop’ (Draper, 1977). This particular hybrid is completely interfertile with highbush types (Draper et al., 1982) and has been used in crosses because of its high photosynthetic rate under hot and dry conditions (Hancock et al., 1992). However, 24 little is known about the mode of Zn gamete formation in V. darrowi or the inheritance patterns of US 75. This information would be very useful in designing efficient breeding strategies. The type of gamete formation will determine the levels of heterozygosity transmitted to US 75, and the mode of inheritance will determine how readily that heterozygosity will segregate. The cytology of Zn pollen formation has not been reported for V. darrowi but, in inter-specific aneuploids (2n = 4x + 9 = 57) of V. ashei x V. corymbosum, Vorsa and Ortiz (1992) found the mode of Zn pollen formation to involve three steps: desynapsis, disjunction of sister chromatids, and cytokinesis. This mechanism is genetically equivalent to FDR, and therefore should result in the transmission of most of the parental heterozygosity (V orsa and Ortiz, 1992). In the selfed progeny of US 75, few meiotic irregularities were found to lead to the production of Zn gametes including 1) premeiotic doubling, 2) tripolar spindle, 3) incomplete spindle at anaphase I, and 4) incomplete spindle at anaphase II (Qu and Hancock, in prep.). V. corymbosum has been shown to have tetrasomic inheritance at four enzyme loci (Krebs and Hancock, 1990), but segregation patterns have not been examined in any of the interspecific hybrids. I To determine the level of heterozygosity transmitted by 2n gametes of Fla 4B and the mode of inheritance in US 75, we selected genotype-specific RAPD markers and followed their inheritance from Fla 48 to US 75 and their segregation patterns in a US 75 x highbush cv. ‘Bluetta’ backcross population. The RAPD analysis allowed us to find and utilize a high number of markers in a short time span (Williams et al., 1990; Welsh and McClelland, 1990). Very few morphological (Draper and Scott, 1971), isozyme ( Breuderle et al., 1991; 25 Van Heemstra et al., 1991; Bruederle and Vorsa, 1994), or RFLP markers (Haghighi and Hancock, 1992) have been described in blueberry and non-homologous chromosomes are not easy to distinguish (Hall and Galletta, 1971). RAPD markers have previously been utilized in blueberries to distinguish cultivars (Aruna et al., 1993) and develop a diploid linkage map (Rowland and Levi, 1994). MATERIALS AND METHODS Elantmatsrial US 75, ‘Bluecrop', ‘Bluetta', Fla 4B, NC84 6-5 and two segregating populations were evaluated for their RAPD markers: 1) A tetraploid population of 61 individuals of US 75 X ‘Bluetta', and 2) 15 diploid individuals of Fla 43 x NC84 6-5. ‘Bluetta’ is a highbush cultivar and based on its pedigree is composed of 75 % V. coozmbosum and 25% V. angustifolium (Draper et al., 1969). NC84 6-5 is a wild selection of V. darrowi kindly provided by J. Ballington. DH! | 1' I Hi I' l'l' Total cell DNA was isolated from young leaves using a modification of the CTAB procedure (Doyle and Doyle, 1987 as modified by Rowland and Nguyen, 1993). DNA was amplified in 12.5 ul volumes using 10 base primers (Operon Technologies Inc., Alameda, CA, and Biotechnology Laboratory, University of British Columbia). Primers were named by the initials of their source (OP and BC) and the company's lot number. Reaction conditions were: 1 ng/ul template DNA, buffer (50 mM KCl, 10 mM Tris-HCl pH 8.3, 0.01% gelatin), 1.6 mM MgC12, 200 M dATP, dCTP, dGTP, am (Boehringer 26 Mannheim), 0.2 uM primer, and 0.06 units/u] Taq DNA polymerase (Gibco). DNA was amplified for 50 cycles in a Perkin Elmer thermal cycler programmed for 30 s denaturation at 94 0C, 70 s annealing at 48 °C and 120 3 extension at 72 °C. The PCR products were separated through 1.2% agarose gels and visualized by ethidium bromide staining. Only reproducible fragments with strong bands were scored in our comparisons. All genotypes were subject to PCR at least twice. I! | . . I I' I 'I To determine the mode of inheritance in US 75, we first located RAPD markers that were present in both Fla 4B and US 75, but absent in ‘Bluecrop' and ‘Bluetta'. This would mean that the genotype of Fla 4B was either AA or Aa, ‘Bluecrop' and ‘Bluetta' were aaaa (nulliplex), and US 75 was either AAaa (duplex) or Aaaa (simplex). We then examined the progeny ratios in the testcross population of US 75 x ‘Bluetta'. Heterozygous pairs of alleles (Aa) transferred from Fla 43 to US 75 should segregate at a ratio of 1 :1 (Aaaa:aaaa) for both disomic and tetrasomic inheritance, while homozygous pairs of alleles (AA) should segregate at a 5:1 ratio (Aaaa & AAaazaaaa) for tetrasomic inheritance and a 1:0 ratio (all Aaaa) for disomic inheritance (Krebs and Hancock, 1989). For each segregating marker, a chi-square test of fit of progeny ratios was performed. We did not nwd to test for a 3:1 ratio as in Krebs and Hancock (1989), because we knew both dominant alleles in US 75 were contributed by Fla 4B, and as a result could not have been segregating as independent disomic loci. To estimate the levels of heterozygosity maintained by 2n gamete formation in Fla 48, we determined both the number of heterozygous allelic pairs (Aa) transferred from Fla 4B to US 75 and the number that were lost. As previously stated, FDR passes on much more heterozygosity than SDR, because FDR gametes contain the non-sister chromatids of each homologous pair of chromosomes, while SDR gametes contain only sister chromatids. The markers segregating (1 :1) in the US 75 x ‘Bluetta' population represented the heterozygous allelic pairs that were transferred from Fla 48 to US 75. To measure the level of heterozygosity lost through crossing over during 2n gamete production, we counted the number of unique dominant markers present in Fla 4B that were absent in US 75. The most likely way that dominant alleles can be present in Fla 4B but absent in US 75 is if a heterozygous locus was lost due to crossing over during 2n gamete production. This number was then doubled to include recessive alleles in the estimate of lost heterozygous loci. It was assumed that equal numbers of dominant and recessive alleles would be lost via crossing over during 2n gamete production. In a few instances, we were able to directly document the loss of heterozygous loci from Fla 48. We screened the diploid population of Fla 43 x NC84 6-5 with the primers producing markers that were present in Fla 4B and identified as duplex (AAaa) in US 75. We looked for markers which segregated 1:1 in the progeny when the marker was absent in NC84 6-5, and those which segregated 3:1 when the marker was present in NC84 6-5. These segregation ratios would only be possible if Fla 4B was heterozygous for that marker. Unfortunately, this approach was limited because most of the markers in Fla 4B and NC84 28 6-5 were shared. The % heterozygosity transfened from Fla 48 to US 75 was ultimately calculated as the number of Fla 4B's heterozygous allelic pairs transferred to US 75 divided by the total number of heterozygous loci detected in Fla 4B. The total number of heterozygous loci in Fla 4B was calculated as the number of heterozygous allelic pairs transferred to US 75 and the number lost. RESULTS AND DISCUSSION M I F I '| One hundred and forty-three of 512 primers produced a total of 267 polymorphic fragments that were present in Fla 4B and absent in ‘Bluetta' and ‘Bluecrop'. Of these, 234 (88%) were present in US 75, and 33 (12.4%) were absent. Of the markers found in US 75, 154 best fit a 1:1 ratio in the US 75 x ‘Bluetta' population (Figure 1, Table 1), while 65 markers best fit a 5:1 ratio (Figure 2, Table 2). For all these loci, the alternate hypothesis of 1:1 or 5:1 segregation was statistically rejected (P < 0.001). While the 1:1 ratios can not be used to distinguish between disomic and tetrasomic inheritance, the 5:1 segregation ratios suggest that the mode of inheritance in US 75 is tetrasomic. This is not surprising, as artificially produced interspecific hybrids between a wide range of Vaccinium species are highly fertile, indicating there is little genomic divergence within the genus (Hancock et al., 1995; Draper, 1977). Likewise, many different types of hybrids have been observed in nature (V ander Kloet, 1988; Breuderle and Vorsa, 29 Table l. RAPD markers that fit a 1:1 (present : absent) segregation ratios in a progeny . _ population of US 75 x ‘Bluetta’ Primer Primer Sequence Fragment size Observed ratio X2 P‘ (bases) BC101 GCGGCTGGAG 600 28 :33 0.41 0.55 1340 34 : 27 0.80 0.42 BC 1 05 CT CGGGTGGG 1840 24 : 37 2.77 0.09 BC 1 25 GCGGT'TGAGG 380 35 : 26 1.33 0.26 BC 149 AGCAGCGTGG 2230 29 : 3 1 0.07 0.78 BC181 ATGACGACGG 2280 37 :23 3.27 0.07 . BC184 CAAACGGCAC 1390 26 :34 1.07 0.30 BC 189 TGCT AGCCT C 970 27 : 33 0.60 0.47 l 180 29 : 31 0.07 0.47 BC244 CAGCCAACCG 550 36 : 24 2.40 0.15 BC516 AGCGCCGACG 1050 28 : 33 0.41 0.55 BC523 ACAGGCAGAC 730 27 : 33 0.60 0.47 1720 35 :25 1.67 0.18 BC536 GCCCCTCGTC 1590 31 : 27 0.28 0.65 BC540 CGGACCGCGT 1230 34 : 26 1.07 0.30 BC 546 CCCGCAGAGT 1 580 36 : 24 2.40 0.15 OPA16 AGCCAGCGAA 1590 28 : 33 0.41 0.55 OPA19 CAAACGTCGG 990 28 : 33 0.41 0.55 OPC02 GTGAGGCGTC 390 30 : 29 0.02 0.85 OPC06 GAACGGACT C 290 32 : 28 0.27 0.65 350 29 : 31 0.07 0.78 1360 34 : 26 1.07 0.30 OPC15 GACGGATCAG 1960 29 : 30 0.02 0.85 OPC 16 CACACTCCAG 1230 35 : 26 1.33 0.26 1350 35 :26 1.33 0.26 OPFOl ACGGATCCTG 800 34 : 26 1.07 0.30 1300 28 : 32 0.27 0.65 2760 27 : 33 0.60 0.47 OPF04 GGTGATCAGG 850 27 : 34 0.80 0.42 OPF05 CCGAAT'TCCC 1410 28 : 32 0.27 0.65 OPFOS CCGAATTCCC 2000 28 : 33 0.41 0.55 OPF08 GGGATATCGG 1410 30 :31 0.02 0.85 OPF12 ACGGTACCAG 1230 32 : 28 0.27 0.65 OPGO8 TCACGTCCAC 1520 30 : 31 0.02 0.85 OPH02 TCGGACGTGA 990 26 : 34 1.07 0.30 Table 1 (cont’d) OPH03 OPH05 OPH07 OPH 1 2 OPH 1 3 OPIO9 OPIZO OPJ04 OPJ09 OF] 14 OPJl7 OPK04 OPK14 OPK 1 7 OPK 1 9 OPKZO OPL02 OPLIO OPLll OPL13 OPL14 OPL15 OPM04 OPM09 AGACGTCCAC AGTCGTCCCC CTGCATCGTG ACGCGCATGT GACGCCACAC TGGAGAGCAG AAAGTGCGGG CCGAACACGG TGAGCCTCAC CACCCGGATG ACGCCAGTTC CCGCCCAAAC CCCGCTACAC CCCAGCTGTG CACAGGCGGA GTGTCGCGAG TGGGCGTCAA TGGGAGATGG ACGATGAGCC ACCGCCTGCT GTGACAGGCT AAGAGAGGGG GGCGGT'TGTC GTCT‘TGCGGA 2250 2210 1 100 2210 180 1 1 10 590 1590 2330 2700 730 370 1020 1600 3690 700 300 680 680 860 2090 2460 1270 1470 860 2 150 610 420 550 730 1470 980 1420 30:30 30:31 34:27 27:28 36:25 27:32 24:36 36:25 33:26 33:26 31 :30 26:35 29:32 28:31 27:32 29:32 29:29 31:27 26:33 27:33 28:32 33:26 27:32 25:32 24:36 34:26 26:33 23:37 32:29 36:25 29:32 28:33 35:26 33:28 0.00 0.02 0.80 0.02 1 .98 0.42 2.40 1 .98 0.83 0.83 0.02 l .33 0.15 0.15 0.42 0.15 0.00 0.28 0.83 0.60 0.27 0.83 0.42 0.86 2.40 1 .07 0.83 3.27 0.15 1 .98 0.15 0.41 1 .33 0.41 l .00 0.85 0.42 0.85 0. 17 0.54 0. 15 0. 17 0.40 0.40 0.85 0.26 0.70 0.70 0.54 0.70 1 .00 0.65 0.40 0.47 0.65 0.40 0.54 0.39 0. 15 0.30 0.40 0.07 0.70 0. 15 0.70 0.55 0.26 0.55 Table 1 (cont’d) OPM13 OPM16 0PM] 7 0PM 1 9 OPM20 OPN l l OPN 16 0P002 OPOO4 OPOO6 OPOO7 CPD 1 2 OPOI 3 OPO l 4 OPO l 6 OPPO l OPP04 OPP07 OPP l 6 OPQO3 oaoos OPQO9 opors oraor OPR09 OPRll OPR13 OPR16 OPSl3 OPS15 OPT06 GGTGGTCAAG GTAACCAGCC TCAGTCCGGG CCTTCAGGCA AGGTCTTGGG TCGCCGCAAA AAGCGACCTG AAGCGACCTG ACGTAGCGTC AAGTCCGCTC CCACGGGAAG CAGCACTGAC CAGTGCTGTG GTCAGAGTCC AGCATGGCTC Tcoocoorrc GTAGCACTCC GTGTCTCAGG GTCCATGCCA CCAAGCTGCC GGTCACCT CA CCGCGTCTT G GGCTAACCGA GGAGTGGACA TGCGGGTCCT TGAGCACGAG GTAGCCGTCT GGACGACAAG CT CT GCGCGT GTCGT'TCCT G CAGTTCACGG CAAGGGCAGA 3560 650 1960 530 830 570 1110 1230 1580 1530 1830 1200 680 610 890 700 2090 2390 700 850 1230 1350 620 1000 1350 860 2050 1470 1250 1490 2000 850 480 240 810 780 36: 29: 23 34: 33: 30: 34: 29: 27: 33: 34: 29: 33 30: 37 26: 33 33 30: 24: 30: 32: 31 32: 33 33 32: 29: 27: 27: 30 29: 27: 29: 34: 32: 34: 28: 25 32 : 38 27 30 27 32 34 27 32 : 28 31 : 24 35 : 28 : 28 31 37 31 29 : 30 29 : 28 : 28 29 32 34 34 ; 31 32 34 32 27 28 26 33 l .98 0. 15 3.69 0.80 0.41 0.00 0.80 0. 15 0.80 0.41 0.80 0. 15 0.41 0.02 2.77 1 .33 0.41 0.41 0.02 2.77 0.02 0. 15 0.02 0. 15 0.41 0.41 0. 15 0. 15 0.80 0.80 0.02 0. 15 0.80 0.15 0.80 0.27 1.06 0.41 0.17 0.70 0.07 0.42 0.55 1.00 0.42 0.70 0.42 0.55 0.42 0.70 0.55 0.85 0.09 0.26 0.55 0.55 0.85 0.09 0.85 0.70 0.85 0.70 0.55 0.55 0.70 0.70 0.42 0.42 0.85 0.70 0.42 0.70 0.42 0.65 0.32 0.60 Table 1. (Cont’d) 32 OPT07 OPT12 OPT14 OPU03 OPU07 CPU 1 6 OPV04 OPV08 OPV14 OPW03 OPW06 OPW14 OPX06 OPX07 OPX19 OPZO l OPZ03 OPZO4 OPZ06 OPZO7 OPZl 1 OPZI 5 OPZ 1 6 OPZl6 OPAGOl OPAGO7 OPAGl6 GGCAGGCT GT GGGTGTGTAC AATGCCGCAG CTATGCCGAC CCTGCTCATC CT GCGCT GGA CCCCTCACGA GGACGGCGT'T AGATCCCGCC GTCCGGAGTG AGGCCCGATG CT GCT GAGCA ACGCCAGAGG GAGCGAGGCT TGGCAAGGCA TCTGTGCCAC CAGCACCGCA AGGCTGTGCT GTGCCGTTCA CCAGGAGGAC CT CAGTCGCA CAGGGCTTT C TCCCCATCAC TCCCCATCAC CT ACGGCI'T C CACAGACCT G CCTGCGACAG 730 360 1650 1590 750 1750 1840 1350 2090 2210 600 1370 470 1520 310 560 570 860 520 1220 1830 580 1210 280 2210 580 610 980 610 3100 430 2090 890 1790 230 910 32: 29: 27: 27: 27: 24: 29: 28: 29: 29: 35: 25: 36: 27: 33: 28: 31: 31: 26: 30: 30: 33: 27: 24: 23: 29: 33: 29: 31: 31: 37: 30: 31: 32: 31: 31: 24: 0. 15 0. 15 0.80 3.81 0.42 0.80 2.77 0. 15 0.41 0. 15 0.15 1 .33 1.98 1.98 0.80 0.41 0.41 0.02 0.02 1.33 0.02 0.02 0.41 0.80 2.77 3.69 0.15 0.41 0.15 0.02 0.07 2.77 0.02 0.02 0.15 0.02 0.02 2.77 0.70 0.70 0.42 0.06 0.54 0.42 0.09 0.75 0.55 0.70 0.70 0.26 0. 17 0.17 0.42 0.55 0.55 0.85 0.85 0.26 0.85 0.85 0.55 0.42 0.09 0.07 0.70 0.55 0.70 0.85 0.78 0.09 0.85 0.85 0.70 0.85 0.85 0.09 Table 1. (Cont’d) OPAJO4 OPAJ 14 OPAK05 OPAK l 5 OPAK 1 6 OPAMO l ‘ the alternate hypothesis of 5:1 was rejected at P<0.001 in all cases. GAATGCGACC ACCGATGCT G GATGGCAGTC ACCT GCCGT'T CT GCGTGCT C TCACGTACGG 1590 2200 1720 1590 1790 380 610 1590 1840 29: :30 31 30: :38 :28 :34 30: :30 :28 23 33 27 31 33 32 31 31 0.15 0.02 0.02 3.69 0.41 0.80 0.02 0.02 0.41 0.70 0.85 0.85 0.07 0.55 0.42 0.85 0.85 0.55 34 58on mm Sea 98 mosa— wEEaEE och .2038“..nt mates—m. can mm m: meadows—m. A? 2m 8w TN 8.5 .6262 <20 an mm. e m_ 28— “we do. 9 2w: 52m .Auocomnmuoocomoav 2.2 E a E wccmwocwom cos—.88 as a $527. 38.8 QC. .295 BEE— wEma .832? x mm m: coca—seem ~95on EOEEB 05 E 3:85 538::qu _ as: m a? massage 8 .68; 33 _ m e3 a a A85 :53 E _ .395 owacz 5st _ .2 mac n a? 882.5 6 .= em as, 822% 23 a .923 .5 383% N Ea c6 sausage _ a $5325 $5835 2 ”E - 5 2. m: $26—an2 05 E muowfiswmeg 08808930 0:22: 05 he 8323:82un .m 953% 88 Figure 3 89 I}? 3% 31’ U i 1. u it 415' were u“ .u Figure 4. Rearrangement of Figure 2-C by chromosome size. Note high numbers of attachment sites in both large and small chromosomes. 90 While percentages of quadrivalents were low in all the tetraploid materials examined, US 75 did have significantly higher percentages of quadrivalents than ‘Bluecrop' or CEL (Table 1). In one case, eleven quadrivalents was observed in a diplotene PMC of US 75 (Figure 3. A). Seven quadrivalents (six ring and one chain) were found in one PMC of US 75 in diakinesis (Figure 3. C). Ring quadrivalents were the most common form of multivalents observed consisting of about 80% of the total. The remaining 20% of the quadrivalents were chain style, although many of them appeared to be broken ring quadrivalents (Figure 3. G). Clear 1]] + I and 1 +1 (Figure 3. C) configurations were observed in one diakinesis PMC of US 75. No multivalents higher than four were observed. Although exact counts of chromosome numbers at anaphase I and II were often not possible, the relative distribution of chromosomes to the poles appeared to be equal at both anaphase I (Figure 3. F) and II. In only a few cases were lagging chromosomes observed (Figure 3. E). W. The staining test indicated that 98% of the US 75 pollen was viable (Figure 5. A), and 94% (1785/1892) of the tetrads produced at least one pollen tube (Figure 5.C). About 10% of the sporads were shaped as dyads (Figure 5. B), suggesting they might be unreduced gametes (Ortiz et al., 1992) MW”. When the ‘Bluecrop' probe was applied to the chromosomes of US 75 without blocking, all the chromosomes were completely labeled (Figure 6. A). The signal appeared comparable to ‘Bluecrop’ probed with its own DNA (Figure 6. B). Probe 91 .62 omen. 03258 @5855 05 mama: _o>o_ 3odvm 05 “a “55$“. 356556 can Bozo. EBabG 5m? mean—2 2.. 883 _ _ N N m N do £55 _ N v _ _ 2 mm .3829 tacoeN N N a 2 N N Na. 5. 3 12:3 _ N m 2 2 2 : a NN mu m: 02.5: owflo>< >K+=S Kai—N. amt—E >316 _ Zm+=w _ >_~+=c~ >_ _+=N~ =vm USE 252.00 meowflsmccoo mew—mam ._ 82338 23 £8:qu an maeufiswues mafia 0882.820 A 03:. 92 Figure 5. Pollen viability of US 75: A) Typical microscope field of pollen stained with acetocarmine, B) A dyad pollen sporad (left) which may represent 2n gametes, and C) Typical microscope field of germinated pollen. 93 fi'om Fla 4B also labeled all US 75 chromosomes evenly when not blocked, although the signal was not as strong as with probe of ‘Bluecrop’ (data not shown). When the chromosomes of US 75 were blocked with either ‘Bluecrop' or Fla 48 DNA and hybridized with the probe of other genome, none of the chromosomes were labeled (Figure 6.C). DISCUSSION The origins of blueberry polyploids and whether they are autopolyploid or allopolyploids has long been of interest to both blueberry taxonomists and breeders (Camp, 1942, Vander Kloet, 1983; Krebs and Hancock, 1989; Qu and Hancock, 1995 ). Originally, most of the tetraploid bluebenies were considered to be allopolyploids (Camp, 1945; Eck, 1966; Vander Kloet, 1988) based primarily on morphological traits. Cockerham and Galletta (1976) also suggested that V. corymbosum and four related tetraploid taxa were allotetraploids because the tetraploids as a group exhibited higher pollen stainability than a group of seven diploid species. However, tetrasorrric inheritance ratios have now been demonstrated in highbush V. corymbosum (Draper and Scott, 1971; Krebs and Hancock, 1989), lowbush V. angusn'folium (Hokanson and Hancock, 1993) and the interspecific hybrid US 75 (Qu and Hancock, 1995), suggesting that they are autopolyploids. V. angustrfolium and tetraploid V. corymbosum are completely interfertile, and fertile hybrids have been generated from most diploid x polyploid combinations (Hancock et al., 1995). The work described here and previous studies have found only limited chromosome size and genome content differentiation in blueberries (Hall and Galletta, 1971; Costich et al., 1993). Considering that V. darrowr’ is thought by many to be an ancient diploid blueberry species Figure 6. Genomic in situ hybridization. A) US 75 chromosomes hybridized by the ‘Bluecrop’ labeled genomic probe. Before it was transmitted, the color is green yellow; B) The ‘Bluecrop’ labeled genomic probe hybridized on its own chromosomes (some chromosomes were lost during the slide treatment). Before it was transmitted, the color is yellow green; and C) US 75 chromosomes blocked with ‘Bluecrop’ genomic DNA and hybridized with labeled Fla 4B probe. Before it was transmitted, the color is brown red. 95 (V ander Kloet, 1983; Vorsa et al., 1988; Bruederle and Vorsa, 1994), it appears that the subgenus Cyanococcus of Vaccinium can be thought of a compilospecies. The prevalence of bivalent configurations observed in this study is in agreement with the earlier studies of meiosis in V. corymbosum (J elenkovic and Hough, 1970; Jelenkovic and Harrington, 1971; Vorsa et al., 1987). Over 90% of the chromosomes in ‘Bluecrop', CEL and over 70% in US 75 were in bivalent associations (Table 1). Since these tetraploid blueberries display tetrasomic inheritance, it is suggested that their bivalents are being formed from random selections of the 4 homologous chromosomes. This is further evidenced by the fact that no 1:0 (present : absent) segregation ratios were found in the 65 homologous RAPD markers unique to Fla 4B (Qu and Hancock, 1995 and 1996). From this work, it should be clear that chromosomal configurations (bivalent vs. multivalent) can not be used as a reliable indicator of whether a species is genetically an autoployploid with tetrasomic inheritance or an allopolyploid with disonric segregation. Blueberry is apparently not unusual in this regard as predominantly bivalent pairing at diakinesis and metaphase has been observed in a number of artificial and naturally occurring autotetraploid species (Jones, 1961; Soltis and Rieseberg, 1986; Crawford and Smith, 1984; Samuel et al., 1990). Several lines of evidence indicate that while striking morphological, geographical and ploidy differences exist between V. darrowi and V. corymbosum, little divergence has developed between their genomes, or at least between the APS (autonomous paring sites) which are involved in initiating homologous chromosome pairing (Callow and Gladwell, 1984; Jones, 1994). These include: 1) tetrasomic inheritance in US 75, 2) the significantly 96 higher number of quadrivalents observed in US 75 than in ‘Bluecrop' and CEL, and 3) the observation that 11 quadrivalents formed from the 12 homologous groups in at least one PMC in US 75 (Figure 3. A). The genomic similarity of diploid V. darrowr', Fla 4B, and tetraploid V corymbosum, ‘Bluecrop’ is further evidenced by the finding that genomic DNA from both ‘Bluecrop’ and Fla 4B hybridized with all of the chromosomes of US 75. Some chromosomal divergence must exist in the group as RAPD analysis revealed unique fiagments for both Fla 4B and ‘Bluecrop’(Qu and Hancock, 1995 and 1996), although the fragments were not divergent enough to prevent hybridization with the other species’ genomic probe. Vorsa et al. (1995) also found a number of RAPD markers in another cross of V. darrowr' x tetraploid V. corymboswn that deviated significantly from tetrasomic ratios, and in other work found that 2 out of 6 artificially produced triploids (4x x 2x V. corymbosum) displayed preferential pairing (V orsa, 1989). However, the genomic diversity in Vaccinium section Cyanococcus is surprisingly little considering the morphological differentiation that has occurred between the species. Other members of the Ericaceae also display limited genomic divergence, as Krebs (1996) found normal segregation patterns in a broad array of interspecific Rhododendron crosses. While the number of chiasmata per bivalent is often assumed to be a maximum of 2 (Jackson and Casey 1980, 1982; Jackson and Jackson, 1996), many bivalent configurations we observed in blueberry were bars with 3 or 4 attachment sites (Figure 3. B). Both large and small chromosomes appeared to have multiple chiasmata (Figure 4). Vorsa (1990) also found evidence for more than 2 chiasmata per bivalent in inter-specific triploid blueberries. We did observe numerous rings with only two terminal attachments at diakinesis, but these could 97 have been the end result of the terminalization of multiple chiasmata scattered along the chromosomes. This is supported by the observation of Qu and Hancock (1995) that 32 of the 102 heterozygous loci detected in Fla 4B were duplex in US 75 and therefore had been subjected to crossing over. Since only 50% of the recombination products generated during FDR are detected in a single progeny (Hermsen, 1984; Hancock, 1992), this means that 64 loci were actually affected by crossing-over in Fla 4B, which gives a total recombination frequency of 62.7% (64/ l 02). It is unlikely tint such a high recombination frequency could have occurred if chiasmata were located primarily at the ends of chromosomes. The x configuration of bivalents also suggest that there were chiasmata close to centromere. Genomic diploidization is thought by many to be an important step in the adaptation of autopolyploids (DeWet, 1980; Watanabe, 1983; Jackson and Jackson, 1996), but there have been no cytogenetic or inheritance studies that directly document this process, except for a controlled experiment where a statistically significant reduction in quadrivalent frequency was observed in artificially produced tetraploid maize after ten generations of sexual reproduction (Lewis, 1980). Jones (1961) formd a stable level of quadrivalents in both natural and artificially produced autotetraploids of Dactylr's, leading him to remark that "it would be wise to talk of the stabilization of meiosis rather than of its diploidization". There is circumstantial evidence in blueberry that chromosomal diploidization may have occurred, as US 75 had significantly more multivalents than native V. corymbosum, and a colchicine induced tetraploid of V. elliom’i was previously shown to have significantly more multivalents than tetraploid V. corymbosurn (Dweikat and Lyrene, 1991). However, the role of these two species in the evolution of polyploid Vaccinium is unresolved. 98 It may be that chromosome pairing relationships in polyploids are diploidized in two ways. In the one classically described, bivalent pairing becomes restricted to a specific set of two homologous chromosomes, while in the other, bivalents are formed by random pairing of all homologues. V. coo/moosum fits the second class best, as it is mostly bivalent pairing but displays tetrasomic inheritance. Over time, suflicient chromosomal divergence and/or changes in pairing control (PC) (Jackson and Hauber, 1994) alleles might occur to change a species from random to preferential pairing, however, selection pressure for preferential pairing would be minimal, once complete fertility was achieved through the evolution of complete random bivalent formation. The specific ph-like gene which prevents the pairing between homoeologous chromosomes in wheat has been well studied (Riley and Chapman, 1958; Waines, 1976; Feldrnen, 1993; Gill et al., 1993), but related PC genes regulating bivalent formation of homologous chromosomes in autopolyploids have not been identified. There is evidence that they exist in natural populations, however, as Jackson and Hauber (1994) found PC mutations that led to both auto- and alloploid pairing behavior in natural populations of Helianthus ciliaris. Stebbins (1947) suggested that raw autopolyploids would have reduced fertility due to: (1) irregular distribution of chromosomes caused by unequal separation of multivalents; (2) irregular distribution caused by meiotic abnormalities of a physiological nature, presumably controlled genetically; and (3) genetic-physiological sterility of an unexplained nature, but not associated with meiotic irregularity." These factors appear to be of limited importance in US 75, as it has very high pollen stainability (98%) and germination rate (94%), and is readily crossed with itself and other tetraploids (Ballington, 1990; Hancock et 99 al., 1995). This suggests that in some instances autopolyploids can be sufficiently fertile to be successful immediately after they are generated. Most artificially produced autopolyploids are maladapted (DeWet, 1980; Hancock, 1996), but it is not surprising that numerous autopolyploids have become established, if one considers how many times polyploids have been formed in nature with a variety of parents (Soltis and Soltis, 1993 and 1995) The present study of chromosomal associations and genomic divergence in V. corymbosum adds to the growing body of knowledge that suggests evolution in autopolyploids is much more dynamic than was previously thought (Soltis and Soltis, 1993; Soltis and Soltis, 1995), and that many species may be incorrectly assumed to be allopolyploids with disonric inheritance based on bivalent pairing. Rather than being an "evolutionary dead-end", autoployploidy may actually allow for enhanced divergence, if diploids can freely transfer new genes into related autoployploids via unreduced gametes. In North America, a series of diploid Vaccinium species are found along the eastern seaboard from Florida to Maine. The autotetraploid V. corymbosum overlaps this entire range. One wonders if the wide adaptive range of V. corymbosum may be due to the assimilation of diploid genes as it moved north from its presumed southeastern origin (V ander Kloet, 1988). This would have been facilitated by its autopolyploidy, particularly if raw hybrids had high fertility due to regular chromosomal pairing. Such interactions may be an important component of evolution in all autopolyploids. 100 LITERATURE CITED Bailey, J.P., Bennett, S.P., Bennett, M.D., and Stace, CA. 1993. Genomic in situ hybridization identifies parental chromosomes in the wild grass hybrid x Festrdtr’a hubbardii. Heredity 71:413-420. Ballington, J. M. 1990. Germplasm resources available to meet future needs for blueberry cultivar improvement. Fr. Var. J. 290:393-456. Bruderle, L. P., and Vorsa, N. 1994. Genetic differentiation of diploid blueberry Vaccinium section Cyanococcus (Ericaceae). Syst. Bot. 19:337-349. Callow, RS, and Gladwell, I. 1984. A general treatment of chromosome synapsis in even- numbered polyploids. J. Theor. Biol. 106:455-494. Camp, W.H. 1942. On the structure of populations in the genus Vaccinium. Brittonia 4: 1 89-204. Camp, M.H. 1945. The North American bluebenies with notes on other groups of Vacciniaceae. Brittonia 5:203-275 Chandler, CK, Draper, A.D., Galletta, G.J.,and Bouwkamp, J .C. 1985. Combining ability of blueberry interspecific hybrids for growth on upland soils. HortScience 20:257-258. Clausen, R.E., and Goodspeed, T.H. 1925. Interspecific hybridization in Nicotiana. II. A tetraploid Glutinosa-tabacum hybrid, an experimental verification of Winge’s hypothesis. Genetics 10:278-284. Cockerham, L.E., and Galletta, G.J. 1976. A survey of pollen characteristics in certain Vaccinium species. J. Amer. Soc. Hort. Sci. 101:671-676. Crawford, D.J., and Smith, EB. 1984. Allozyme divergence and intraspecific variation in Coreopsis graniflora (Compositae) Syst. Bot. 9:219-225. Costish, D.E., Ortiz, R., Meagher, T.R., Bruderle, L.P., and Vorsa, N. 1993. Determination of ploidy level and nuclear DNA content in blueberry by flow cytometry. Theor. Appl. Genet. 86:1-6. De Jong, E.J.H., Kamstra, S.A., Van Den Berg, P.M.M.M., and Ramanna, MS. 1995. Genomic in situ hybridization (GISH) and RFLP analysis for the identification of alien chromosomes in the backcross progeny of potato (+) tomato fusion hybrid. Heredity 74:250- 257. 101 Draper, A.D., Galletta, G.J., and Ballington, J .R. 1982. Breeding methods for improving southern tetraploid blueberries. J. Amer Soc. Hort. Sci. 96:791-792. Draper, A.D., and Scott, DH. 1971. Inheritance of albino seedling in tetraploid highbush blueberry. J. Amer. Soc. Hort. Sci. 96:791-792. DeWet, J .M.J . 1980. Origin of polyploids. In: Lewis W.H. (ed.) Polyploidy, Biological Relevance. Plenum Press, New York. pp 3-15. Durant, A. 1960. Expected frequencies of chromosome associations in tetraploids with random chiasma formation. Genetics 45:779-783. Dweikat, I. M., and Lyrene, PM. 1991. Induced tetraploidy in a Vaccinium eIIr'ottr'r‘ clone facilitates crossing with cultivated highbush blueberry. J. Amer. Soc. Hort. Sci. 1 16(6): 1063- 1066. Eck, P. 1966. Botany. p. 14-44. In: P. Eck and N. F. Childers (eds.) Blueberry culture. Rutgers Univ. Press, New Brunswick, NJ. Erb, W.A, Draper, A.D., Galletta, G.J., and Swartz, H.J. 1990. Combining ability for plant and fruit traits of interspecific blueberry progenies on mineral soil. J. Amer. Soc. Hort. Sci. 115:1025-1028. Erb, W.A., Draper, A.D., and Swartz, H.J. 1993. Relation between moisture stress and mineral soil tolerance in blueberries. J. Amer. Soc. Hort. Sci. 118:130-134. F eldmen, M. 1993. Cytogenetic activity and mode of action of the pairing homoeologous (ph-l) gene of wheat. Crop Sci. 33:894-897. Gill, K.S., Gill, B.S., Endo, T.R., and Mukai, Y. 1993. Fine physical mapping of phl, a chromosome pairing regulator gene in polyploid wheat. Genetics 134: 123 1-1236. Hall, S.H., and Galleta, G.J. 1971. Comparative chromosome morphology of diploid Vaccinium species. J. Amer. Soc. Hort. Sci. 96:289-292. Hancock, J .F., Haghighi, K., Krebs, S.L., Flore, J.A., and Draper, AD. 1992. Photosynthetic heat stability in hi ghbush blueberries and the possibility of genetic improvement. HortSci. 27:1 111-1 1 13. Hancock, J. F. 1992. Plant evolution and the origin of crop species. Prentice-Hall, Englewood Cliffs, New Jersey. Hancock, J. F., Erb, W.A., Goulart, B.L., and Scheerens, J .C. 1995. Utilization of wild 102 blueberry germplasm: the legacy of Arlen Draper. J. of Sm. Fr. & Vit. 3: 1-16. Hancock, J .F . 1996. The colchicine story. HorTechnology. (In Press). Hauber, DR 1986. Autotetraploidy in Haplopappus spinulosus hybrids: Evidence from natural and synthetic tetraploids. Amer. J. Bot. 73(1 1):1595-1606. Hermsen, J. G. 1984. Mechanisms and genetic implications of Zn gamete formation. Iowa St. J. Res. 58. 421-434. Hokanson, K., and Hancock, J .F. 1993. The common lowbush blueberry, Vaccinium angustr’foliurn Aiton, may be an autopolyploid. Can. J. Plt. Sci. 73:889-891. Jackson, R.C., and Casey, J. 1982. Cytogenetic analysis of autopolyploids: models and methods for triploids to octoploids. Amer. J. Bot. 69(4):487-501. Jackson, R.C., and Casey, J. 1980. Cytogenetics of polyploids. In: W. H. Lewis (ed.). Polyploidy, Biological Relevance. Plenum press, New York. pp.17-44. Jackson, R. C., and Hauber, DP. 1994. Quantitative cytogenetic analysis of autoploid and alloploid taxa in the Helianthus ciliaris group (Compositae). Amer. J. Bot. 81 : 1063-1069 Jackson, R. C., and Jackson, W.J. 1996. Gene segregation in Autopolyploids: Prediction from meiotic configurations. Amer. J. Bot. 83:673-678. Jelenkovic, G., and Harrington, E. 1971. Non-random chromosome associations at diplotene and diakinesis in a tetraploid clone of Vaccinium australe Small. Can. J. Genet. Cytol. 13:270-276. Jelenkovic, G., and Hough, LP. 1970. Chromosome associations in the first meiotic division in three tetraploid clones of Vaccinium corymbosum L. Can. J. Genet. Cytol. 12:316- 324. Jelenkovic, G., and Draper, AD. 1970. Fertility and chromosome behavior of a derived decaploid of Vaccinium. J. Amer. Soc. Hort. Sci. 95:816-820. Jones, OH. 1994. Meiosis in autopolyploid Crepr's capillaris. III. Comparison of triploids and tetraploids: evidence for non independence of autonomous pairing sites. Hered. 73 :215- 219. Krebs, S.L., and Hancock, J .F. 1989. Tetrasonric inheritance of isoenzyme markers in the highbush blueberry Vaccinium corymbosum L. Hered. 63:11-18. 103 Krebs, S.L. 1996. Normal segregation of allozyme markers in complex Rhododendron hybrids. J. Hered. 87:131-135. Laverty, T., and Vorsa, N. 1991. Fertility of aneuploids between the 5x and 6x levels in blueberry: the potential for gene transfer from 4x to 6x levels. J. Amer. Soc. Hort. Sci. 116(2):330-335. Luby, J .J ., Ballington, J .R., Draper, A.D., Pliszka, K., and E.Austin, M. 1991. Blueberries and cranberries (Vaccinium). Acta. Hort. 290:393-456. Lyrene, P.M., and Sherman, W.B. 1983. Mitotic instability and Zn gamete production in Vaccinium corymbosum x V. e11iottii,hybrids. J. Amer. Soc. Hort. Sci. 102:560-571. Maceria, N.Q., Jacquard, P., and Lumaret, R. 1993. Competition between diploid and derivative autotetraploid Dactylis glomerata L. From Galicia. Implications for the establishment of novel polyploid population. New Phytol. 124:321-328. Moon, J .W., Hancock, J .F., Draper, A.D., and Flore, J .A. 1987. Genotype differences in the effect of temperature on CO2 assimilation and water use efficiency in blueberry. J. Amer. Soc. Hort. Sci. 112:170-173. Newcomer, EH. 1941. Chromosome numbers of some species and varieties of Vaccinium and related genera. Proc. Amer. Soc. Hort. Sci. 38:468-470. Ortiz, R., Bruderle, 1.P., Laverty, L., and Vorsa, N. 1992. The origin of polyploids via 2n gamete production in Vaccinium section Cyanococcus. Euphytica 61:241-246. Qu, L., and Hancock, J. F. 1995. Nature of Zn gamete formation and mode of inheritance in interspecific hybrids of diploid Vaccinium darrawi and tetraploid V. corymbosum. Theor. Appl. Genet. 1309-1315. Qu, L., and Hancock, J .F. 1996. RAPD-based genetic linkage map of blueberry derived from an interspecific cross between diploid Vaccinium darrowi and tetraploid V. corymbosum. J. Amer. Soc. Hort. Sci. (1n review). Riley, R., and Chapman, V. 1958. Genetic control of the cytologically diploid behavior of hexaploid wheat. Nature 182:713-715. Rousi, A. 1967. Cytological observations on some species and hybrids of Vaccinium. Ziichter 36:352-359. Samuel, R., Pinsker, W., and Ehrendorfer, F. 1990. Allozyme polymorphism 1n diploid and polyploid populations of Galium. Hered. 65: 369-378. 104 Soltis, D.E., and Rieseberg, L.H. 1986. Autopolyploidy in T olmiea menziesii (saxifi'agaceae): Genetic insights from enzyme electrophoresis. Amer. J. Bot. 73:310-318. Soltis, D.E., and Soltis, PS. 1993. Molecular data and the dynamic nature of polyploidy. Crit. Rev. Plt. Sci. 12:243-237. Soltis, D.E., and Soltis, RS. 1995. The nature of polyploid genomes. Proc. Natl. Acad. Sci. USA 92:8089-8091. Stebbins, G.L. 1947. Types of polyploids: their classification and significance. Adv. Genet 1:403-429. Vander Kloet, SP. 1980. The taxonomy of the highbush blueberry, Vaccinium corymbosum. Can J Bot 58:1187-1201. Vander Kloet, SP. 1983. The taxonomy of Vaccinium Cyanococcus: a summation. Can. J. Bot. 110:285-291. Vander Kloet, SP. 1988. The genus Vaccinium in North American. Res. Branch Agric. Can. Publ. 1828 Van Dijk, P., and Van Delden, W. 1990. Evidence for autotetraploidy in Plantago media and comparisons between natural and artificial cytotypes concerning cell size and fertility. Heredity. 65:349-357. Vorsa, N. 1987. Meiotic chromosome paring and irregularities in blueberry interspecific backcross-1 hybrids. J Hered. 78:395-399. Vorsa, N. 1989. Quantitative analysis of meiotic chromosome pairing in-triploid bluberry: Evidence for preferential pairing. Genome 33:60-67. Vorsa, N., Manos, P.S., and Van Heemstra. 1988. Isozyrrre variation and inheritance in blueberry. Genome 30:776-781. Vorsa, N., and Navy, R. 1995. Meiotic behavior and RAPD segregation in a tetraploid hybrid ( Vaccinium darrowi x V. corymbosum) Suggests genomic divergence in blueberry. (Abstract). HortScience 30:808. Waines, J .G. 1976. A model for the origin of diploidizing mechanism in polyploid species. The Amer. Nat. 110:415-430. Watanabe, K. 1983. Studies on the control of diploid-like meiosis in polyploid taxa of, Chrysanthemum 4. Colchi-ploids and the process of cytogenetical diploidization. Theor. 105 Appl. Genet. 66:9-14. Werner, J. E., and Peloquin, SJ. 1991. Occurrence and mechanisms of Zn egg formation in 2x potato. Genome 34:975-982. CHAPTER 4 FOUR MECHANISMS OF 2N POLLEN FORMATION IN INTERSPECIFIC BLUEBERRY (VACCINIUM) HYBRIDS 106 107 Abstract. Mechanisms responsible for Zn pollen formation in blueberry were studied cytogenetically using selfed progeny of a tetraploid interspecific hybrid, Vaccinium darrowi (2x) x V. corymbosum (4x). Four meiotic irregularities were found to lead to the production of Zn gametes: (l) Premeiotic doubling (PD), where chromosome doubling in a pollen mother cell (PMC) during mitosis resulted in the formation of four 2n gametes; (2) Tripolar spindle (TPS), where the spindles at anaphase II were firsed at one pole resulting in one 2n and two 11 gametes; (3) Incomplete spindle at anaphase I (1S1), where the spindle functioned briefly at anaphase I and then failed producing two 2n gametes; and (4) Incomplete spindle at anaphase H in one daughter nuclei (1S2-1), where the spindle at anaphase H in one daughter nuclei firnctioned briefly and then failed resulting in one 2n and two n gametes. The TPS and IS] produced FDR (first division restitution) 2n pollen. The allelic constitution of gametes formed by PD is similar to FDR produced gametes, although the cytological process is different. The 182-1 mechanism resulted in SDR (second division restitution) 2n pollen. Surveys of mature pollen from the selfed progeny, its parent and a tetraploid grandparent revealed sporad associations consistent with all 4 mechanisms of Zn gamete formation. 108 INTRODUCTION Unreduced gametes have played an important role in natural speciation through the sexual production of polyploids (Harlan and deWet, 1972; deWet, 1980; Bretagnolle and Thompson, 1995), and in crop improvement as a vehicle to unilaterally transfer traits from 2x to 4x species. The mechanisms associated with 2n gamete production have been actively researched in many species, since the relative frequency of Zn gamete production determines how efficiently 2n gametes can be employed in breeding and the way 2n gametes are formed determines how much heterozygosity is transferred to progeny. There are two major modes of Zn gamete formation: FDR (first division restitution) and SDR (second division restitution) ( Peloquin, 1982; Hermsen, 1984). On the average, FDR is thought to transmit 80% of a species heterozygosity, while SDR passes on only about a 40% (Peloquin, 1982; Hermsen, 1984). The blueberries in section Cyanococcus of the genus Vaccinium are a polyploid complex consisting of diploid (2n = 2x = 24), tetraploid (2n = 4x = 48) and hexaploid (2n = 6x = 72) species. Cultivated blueberries were derived from three polyploid taxa: tetraploid V. corymbosum (highbush blueberry), hexaploid V. corymbosum (syn. V. ashei, rabbiteye blueberry), and tetraploid V. angustifolium (lowbush blueberry). Polyploid speciation in blueberry probably occurred via unreduced gametes, as a high percentage of Zn pollen has been found in species and interspecific hybrids at different ploidy levels including 2x (Ortiz et al., 1992a; Megalos and Ballington, 1988), 3x (Vorsa, 1991), 4x (Ortiz et al., 1992b), 5x (Vorsa and Ortiz, 1992) and 6x (Ortiz et al., 1992b ). Unreduced female gametes are also produced by 2x species, as evidenced by tetraploid hybrids being formed from 2x x 4x 109 crosses (Megalos and Ballington, 1988; Draper et al., 1982) and DNA marker segregation (Qu and Hancock, 1995). . An important aspect in blueberry brwding has been to combine genes from different ploidy levels, since numerous horticulturally important traits have been described in both wild diploid species and the higher ploidies (Hancock et al., 1995; Galletta and Ballington, 1995). The most useful interspecific hybrid parent has been US 75, a tetraploid of Fla 4B ( a wild selection of diploid V. darrowi ) x 'Bluecrop' ( a cultivar of tetraploid V. corymbosum) presumably produced when a pollen grain of “Bluecrop’ fertilized a 2n egg of Fla 4B (Draper et al., 1982). It has played an important role in the development of tetraploid highbush blueberries with a low chilling requirement (Ballington, 1990) and several of its other elite traits such as high fruit quality, mineral soil tolerance and photosynthetic heat tolerance, have the potential to further improve the northern tetraploid highbush blueberry (Erb et al., 1990 and 1993; Chandler et al., 1985; Hancock et al., 1992). While 2n gametes are quite common in blueberry and have already been successfully employed in breeding, research on the mechanism(s) responsible for their formation are only beginning to emerge. First division restitution with no crossing over (F DR-NCO) has been proposed as the main cause of Zn pollen production in blueberries (V orsa and Ortiz, 1992; Ortiz et al., 1992a). Studies of Zn pollen formation in diploid, tetraploid and aneuploid hybrids (2n = 4x + 9 = 57) revealed that 2n gamete production commonly involved three consecutive events: (1) desynapsis of paired chromosomes prior to metaphase 1; (2) sister centromere disjunction in univalents at anaphase I; and (3) cytokinesis after telophase 1 leading to dyad formation (V orsa and Ortiz, 1992; Vorsa and Ortiz, personal communication 1 10 in Galletta and Ballington, 1995). Filler and Vorsa (personal communication in Galletta and Ballington, 1995 ) have also found parallel spindles and tripolar spindles in the diploid V. darrowi and V. elliotrr'i, and tetraploid biotypes of V. pallidum. Stushnofl‘ and Hough (1968) observed premature cytokinesis leading to the formation of Zn pollen in the 4x highbush cultivar 'Coville’. Qu and Hancock (1995) demonstrated that a 2n female gamete of Fla 4B was probably produced via the mechanism of FDR using RAPD markers. In this report, we discuss the mechanisms responsible for the Zn pollen production in selfed progenies of US 75 (US 753) and give frequencies of Zn pollen production in 'Bluecrop' , US 75 and US 755. MATERIALS AND METHODS We: Two 7-year old plants of US 753 were transplanted from the field in 1995 and grown in an unheated greenhouse at Michigan State University for a year. Flower buds undergoing meiosis were collected in the spring of 1996 and fixed in methanol : acetic acid (3:1) for at least 48 hr in a refrigerator (Qu and Hancock, 1996). For long term storage, the fixed buds were transferred to 70% ethanol and held in a refiigerator. Anthers were separated from the rest of the flower buds and squashed in 10% acetocarmine. Observations and photomicrographs were made with a phase contrast microscope. Ballerinas! Pollen from the two US 755 plants were collected from fixed buds. Pollen of US 75 and 'Bluecrop' was collected fresh in 1993 fiom greenhouse grown plants and stored for 3 yearsat0°Cforlateranalysis. Thepollenwerestainedwithacetocarmine(QuandHancock, 111 1996) and counts of unreduced pollen gains were made in a series of random fields under the phase contrast microscope. RESULTS Representative stages of normal meiosis in US 755 are shown in Figure. 1. Four abnormal meiotic process were observed that led to the production of unreduced gametes: ( 1) Premeiotic doubling (PD): Twice the tetraploid number of chromosomes were observed in PMCs which would have resulted in four 2n gametes following normal meiosis [Figure 2.A, B (large arrow)]; (2) Tripolar spindles (TPS): Spindles were fused at one pole in anaphase I, which would have produced one 2n and two 11 gametes [Figure 2. C, D]; (3) Incomplete spindle at anaphase I (1S1): the spindle at anaphase I failed and left 48 univalents forming one pole. This would have produced two 2n gametes (Figure 2. E, F); and (4) Incomplete spindle at anaphase II (1S2): Spindles failed in one of the daughter nuclei in anaphase II (ISZ-l), which would have resulted in one 2n and two 11 gametes (Figure 2. G, H). Incomplete spindle at anaphase H could also have occurred in both daughter nuclei (IS2- 2), but its frequency would have been extremely low and we failed toobserve it. Since only 2% of IS2-l was observed, the frequency of IS2-2 would have been 0.0004% (0.02 x 0.02). . The overall frequency of Zn gametes observed cytogenetically was 12% (39/318), PD(1.2%), TS(4%), IS 1(5%) and IS2-1(2%), with both US 755 plants having about the same percentage. The products (pollen grains) of a PMC remain attached in blueberry and in normal meiosis form a tetrad (Stushnofl‘ and Hough 1968; Stushnofl' and Palse 1969). The PD mechanism should produce a tetrad of large 2n gametes, tripolar spindles and ISZ-l 112 2" 1 4*! ”i. “-0- eA B “ . ; .\ -1» [Car‘i'i‘li - . - tires.» 1:33”;- ' ‘ V pfi