IZ.I~»IIII“II I n I: III MIIIIW III I I'I' ..IIIIIII'IIIII II’ I I’IIIIIII I H III M I I III II IIIIIII II I III IIIII W IIIII’III II WI I I I III -I II III III“ III. II III I, "II III I IIIIIII I III ‘ IIIII I’I’III I I“ II" III II , . II I I hillIIIIIIIIII IIIIIIMH II IIIIIIIH III'IIIIIIIII‘IIII‘ I I. g; I ": IIII‘ mIII IIII II II“ III II III II I III III III I , II , ".I. I II‘mMIullfi 'Im I IMI J“%quI - - '1' " I“ WI MI W WI I IIHHI II III ””1le W‘I'IIIII III IIIIIIII IIH Il| III III-III III IIII’II III” II IIIIIIIIIIIINIIIIIIIIII II I 1IWIIIIII WWI _‘4—. ,__:.- 'd: 'M':' dfl‘.‘ ‘ -. - -.—-.V - '51: . a _ —' - I .: 1."; \—::— '_—. I' “ I I 'III —.’ .— :J__.. __ o- - 4 . - '— — .5 u ’ . VHV: :‘ «A - ’«_ - an”... . :25“: - f. - ”LN-é. _ ghow -4 I 7:2' . . “' . llIII'IIJ I :1 I: II. ”VK‘IEIILII' hf“ .I A IN II.‘ ”I I II IIIIINhIpIII IIIIIIIOIHIIIIQI III I II'IfiI' II III I II . I K '2‘ I, ‘II‘I‘IIIIIIII l'I I I IIIIIIl IKIII III} II IIIIIIIIIIII Il‘IIIIIIII‘ ”II. 1-5:.» , II I II I II :. .II III II IIIIII IIIIIIII III III) III IIIIII III III II “III .I 5le III I WI . "" ‘I, I'IIIII' II I I II"'::.','IJII " II iI Wu] I"XIII IIIII III-II ~ 'r.|' III... II II I:II,I2 .I , II" I I. I .II >I II“. ‘ I ' ' I Ii II I I H I IIIII " I" 'IflIfiiIfiflm ' II If! ' I a I I I' I, | I I' II III,.“IIIII ”If. I I I ‘ I II'I III IIIIII IIII IIII , II I II IIIII lI'lIIII'II, 'I'III. I II I I III I'I'II, I_ ...|- ‘I H I I.' III .HI “I! IIII IIIIIII IH'JIB I1" _ 'o. |I 'I‘ " ' [ITI‘I ' III. III III, I III .II. III-III. I“. I ..: I. I" . u. I I 1 I ’ I 1 I‘ x;— Z:— Tussle llnzllluylmlllllll"(Iliumll 315 i... LIBRARY Michigan State Univcrs'lt)‘ This is to certify that the thesis entitled A study of regeneration and karyotypic variability in tissue cultures of Hordeum vulgare. fl. jubatum. and their interspecific hybrid. presented by Thomas James Orton has been accepted towards fulfillment of the requirements for Doctoral Botany and Plant degree in Pathology and Genetics Major professor Datef/g/y7 0-7639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout: from your record. AST A STUDY OF REGENERATION AND KARYOTYPIC VARIABILITY IN TISSUE CULTURES OF HORDEUM VULGARE, H, JUBATUM, AND THEIR INTERSPECIFIC HYBRID By Thomas James Orton A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology and Genetics Program l978 A STI The rar limited by 5 0f the inher indicated th Pool to over the introgre and unyieldi culture was ””0 diploi recovery of It Was I growth, and l explanted ti: Ch10r°ph9noxj Placed Onto V hOrmOneS I CA}! tissues, take the higheSt f one of fin Ca ABSTRACT A STUDY OF REGENERATION AND KARYOTYPIC VARIABILITY IN TISSUE CULTURES 0F HORDEUM VULGARE, H, JUBATUM, AND THEIR INTERSPECIFIC HYBRID By Thomas James Orton The range of cultivation of barley (Hordeum vulgare, HV) is limited by sensitivity to environmental factors. Earlier testing of the inheritance of certain such phenotypes for breeding purposes indicated the need for additional genetic variability in the HV gene pool to overcome this sensitivity. 5, jubatum (HJ) was utilized for the introgression of genes into HV. All HVxHJ hybrids were sterile and unyielding to jn_!jgg_techniques to restore fertility. Tissue culture was investigated in this study as a means to achieve in. vitro diploidization in HVxHJ and as a tool for the induction and recovery of new sources of genetic variability. It was necessary to perfect techniques for callus induction, growth, and plant regeneration in Hordeum for this study. Numerous explanted tissues were placed on media containing 4 mg/l 2,4-di- chlorophenoxyacetic acid for callus induction. Calli were then placed onto various media to test the effects of mineral nutrients, hormones, CAMP, and pH on growth and regeneration. Immature ovarian tissues, taken from florets between 10 and 50% of mature size, gave the highest frequency of primary callus formation. Cultures assumed one of five callus types, distinguishable by differences in cell size, tissue organogenic requiring ar fects. Medi shoot format acid (lAA)-a inhibited or were regener In Dl‘8\ HVxHJ was dL doubling Chr amphiploids colchicine i which exhibi than the ex; hypOamphjpic a high degre I”itial dicatEd Chrc of novel ass tiSSue-totip stable While somal Tearra The agri Va”ability c genetic and i Thomas James Orton size, tissue consistency, growth rate and habit, karyology, and organogenic potential. Regeneration was a quantitative phenomenon, requiring analyses of measured responses to ascertain treatment ef- fects. Media containing gibberellic acid promoted primary root and shoot formation and inhibited crown root formation. All indoleacetic acid (IAA)-amino acid conjugates tested promoted callus growth and inhibited organogenesis more effectively than free 1AA. Whole plants were regenerated from callus cultures of HJ and HVxHJ, but not of HV. In previous studies, it was hypothesized that sterility of HVxHJ was due to the lack of chromosome pairing at meiosis. By doubling chromosome complements in_yjtrg, it was hoped that fertile amphiploids could be regenerated. Calli of HVxHJ were treated with colchicine in solid media and regenerated into whole plants. 20% of which exhibited variable chromosome numbers with means slightly less than the expected amphiploid number of 42 (hypoamphiploids). All hypoamphiploids were sterile due to a breakdown of meiosis, despite a high degree of chromosome pairing. Initial cytological observations of cultured Hordeum cells in- dicated chromosome number variability, perhaps encompassing an array of novel associations of chromosomes for genetics and breeding. In tissue-totipotent HVxHJ callus, mean chromosome numbers remained stable while aneuploidy accumulated over time. Polyploidy and chromo- somal rearrangements were also detected in certain cultures. By physically separating callus tissue, it was possible to segregate and propagate chromosomal variants. The agronomic and genetic significance of jg.yjt§9_karyotypic variability depends on its expression in regenerated plants. Cyto- genetic and isozymic analyses were used to trace the pathway of Thomas James Orton karyotypic variability into regenerates. Regenerating roots exhibited a complete loss of polyploidy, and an attenuation of aneuploidy and chromosomal rearrangements. Isozyme analysis of HVxHJ regenerates provided preliminary evidence for the jn_yjtrg_quantitative segre- gation of parental genomes. One HV-like regenerate was a haploid exhibiting normal meiosis, and yet retained esterase isozyme hybri- dity; this constitutes preliminary evidence for jn_vjtrg_intro- gression of HJ genes into HV. Methods for callus induction, growth maintenance, and regener- ation of whole plants of Hordeum are described. From plants recalci- trant to_i_n_ modiploidization, it was possible to regenerate doubled plants from colchicine-treated calli. Finally, this study demon- strated the use of tissue culture as a means of rapid introgressive breeding, of recovering fertile intermediates of sterile interspeci- fic hybrids, and of isolating haploids. Dedication For Delynn ii The f< presented ir T.B. F garding the R.P. S cared for p] M.D. F conjugates. J.F. F training fOr P.S. c Pollard, my critical Eva Financ ing asSlStan igan Agricu] is grateful]; A Spec‘ and f0light, 1 concern, CF82 for their und A VEry Dossibi ’ and ACKNOWLEDGMENTS The following were important and often instrumental to the work presented in this dissertation: T.B. Rice; who elucidated most of the preliminary details re- garding the tissue culture of Hordeum. R.P. Steidl; who synthesized the interspecific hybrids and cared for plant material. M.D. Peterson; who provided media for testing of IAA-amino acid conjugates. J.F. Fobes; who generously provided equipment, supplies, and training fbr starch gel electrophoresis. P.S. Carlson, w. Tai, J.E. Grafius, L.N. Mericle, and C.J. Pollard, my guidance committee; who provided materials, equipment, critical evaluation, and served as models of scientific excellence. Financial support through a Botany and Plant Pathology teach- ing assistantship, and a mini-grant to project 3l95 from the Mich- igan Agricultural Experiment Station to J.E. Grafius and P.S. Carlson is gratefully acknowledged. A special acknowledgment to N. Tai, my co-adviser, who believed and fought, to P.S. Carlson, my other co-adviser, for his exuberant concern, creativity, and passion for good science, and to my parents for their understanding and support. A very special acknowledgment to my wife Delynn who made it all possible, and to whom this dissertation is dedicated. iii LIST or TA LIST or Ft CHAPTER I INTI II A Ql FROi Ann Intr Mate Rest Disc III A Cy FROh IV VARY TABLE OF CONTENTS LIST OF TABLES ........................ LIST OF FIGURES ....................... CHAPTER I II III IV INTRODUCTION ..................... A QUANTITATIVE ANALYSIS OF GROWTH AND REGENERATION FROM TISSUE CULTURES OF HORDEUM VULGARE, H, JUBATUM, AND THEIR INTERSPECIFIC HYBRID ............ Introduction ..................... Materials and Methods ................. Results ........................ Discussion ...................... A CYTOGENETIC CHARACTERIZATION OF PLANTS REGENERATED FROM COLCHICINE—TREATED CALLUS CULTURES OF HORDEUM VULGARE x u, JUBATUM ................. Introduction ..................... Materials and Methods ................. Results ........................ Discussion ...................... KARYOTYPIC VARIABILITY IN TISSUE CULTURES OF HORDEUM ........................ Introduction ..................... Materials and Methods ................. Results ........................ Discussion ...................... KARYOTYPIC VARIABILITY IN PLANTS REGENERATED FROM TISSUE CULTURES OF HORDEUM .............. Introduction ..................... Materials and Methods ................. Results ........................ Discussion ...................... iv Page vi viii Page VI CONCLUSIONS ...................... l46 Callus Induction, Growth, and Regeneration ...... 145 In_Vitro Diploidization ................ 147 In.Vitro Karyotypic Variability ............ 143 Regeneration of Ig_Vitro Karyotypic Variability . . . . 149 APPENDICES .......................... 151 A ELECTROPHORETIC TECHNIQUES .............. 151 B CYTOLOGICAL TECHNIQUES ................ l53 BIBLIOGRAPHY ......................... 155 Table 2.1 2.2 LIST OF TABLES Page Description of callus types ............... I7 Comparative growth rates and hydration among subcalli of morphologically uniform Hordeum callus ........ 21 Summary of pertinent characteristics regarding callus tissue of cultures utilized in the basic x hormone grid .......................... 22 Summary of significant basic and hormone effects from Figs. 2.3, 2.4, and 2.5 ................. 3l Effects of IAA-amino acid conjugates on mean response class intensity in Hordeum tissue cultures ....... 33 Summary of means, ranges, and associations of chromosomes among regenerated hypoamphiploids, broken down according to source plant, tiller, and floret ........... 52 Comparison of means and ranges of chromosome numbers from tissues derived from HVxHJ ............. 59 A comparison of means and distributions of chromosome numbers among cultures derived from HVxHJ-57 ...... 67 Summary of chromosome numbers in tissue cultures of HV and HJ ........................ 80 Comparison of L/S ratios in cultured tissues of HVxHJ-57 ........................ 84 Micronuclei in type A callus tissue of HVxHJ-57 ..... 87 Means of chromosome numbers and PRX and EST isozyme band intensities among 35 subcalli of HVxHJ-57 type E callus . 91 Analysis of variance; chromosome numbers of 35 subcalli of HVxHJ-57 type E callus ................ 95 Means and coefficients of variation (V) of isozyme intensities among PRX and EST isozyme bands ....... 95 Distributions of isozyme band intensities ........ 97 vi Table 5.l 5.2 5.3 5.4 5.5 5.6 5.7 Comparison of chromosomal rearrangements between in vitro and corresponding regenerated jg_vivo tissue—Z . . Analysis of variance; chromosome counts from immature ovary wall tissue of plants regenerated from HVxHJ type A callus ........................ Analysis of variance; chromosome counts from root tips regenerating from primary type A callus of HVxHJ . . . . Analysis of variance; chromosome counts from root tips regenerating from secondary type A callus of HVxHJ . . . Comparative intensities of EST and GOT isozymes among 43 regenerated HVxHJ plants .............. Distribution of isozyme band intensities for 43 regen- erated HVxHJ plants .................. Comparisons of chromosome associations among four regenerated HVxHJ plants ................ vii Page III 135 Figure l .1 LIST OF FIGURES Diagrarnmatic summaryof research presented in the chap- ters of this dissertation, indicating stepwise rela- tionships ....................... Graph of the percent of ovaries producing callus vs. developmental stage .................. Callus types in tissue cultures of Hordeum ...... Significant pairwise contrasts for basic treatments, broken down according to parental source and response class ......................... Significant pairwise contrasts for hormonal treatments, broken down according to parental source and response class ......................... Comparison of significant contrasts for hormonal treatments between HVxHJ types A and E calli ..... Organogenesis from Hordeum callus ........... Summary of the optimal conditions fer callus induction, callus growth, and regeneration in Hordeum ...... Comparison of spike morphology between hypoamphiploids and HV, HJ, and HVxHJ ................. Comparison of isozymes from crude extracts of culm bases from hypoamphiploids vs. HVxHJ ......... Microsporogenesis in hypoamphiploids ......... Microsporogenesis in hypoamphiploids (con't) ..... Dynamics of mean chromosome number from callus induc- tion (time = O) to 16 months, vs. culture state . . . . Karyotypic variability in tissue cultures of Hordeum Dynamics of the coefficient of variation (V) of chrom- osome numbers from callus induction (time = O) to l6 months, vs. culture state ............... viii Page 16 19 26 28 3O 36 48 ST 54 57 69 71 73 5.4 5.5 Figure Page 4.4 Composite distributionsof chromosome numbers in tissue cultures of HVxHJ ................... 76 4.5 Distribution of chromosome counts in HVxHJ secondary type E callus ..................... 78 4.6 Summary of procedure for mathematical proof of chromosomal rearrangements by comparison of ratios of longest to shortest chromosomes (L/S) ......... 82 4.7 Cytological observations of mitotic anomalies in tissue cultures of Hordeum, possibly responsible fOr variable chromosome numbers .............. 86 4.8 Procedure used for isolating and propagating karyo- typic variability ................... 90 4.9 Nomenclature system and examples of typical scoring for cathodal PRX and anodal EST bands from crude ex- tracts of HVxHJ type E callus ............. 93 4.lO Comparative cathodal PRX and anodal EST zymograms to show quantitative and qualitative differences in band expression among subsubcalli of HVxHJ type E callus . . 95 4.ll Hypothetical mechanism to explain the generation of a steady state distribution of chromosome counts in HVxHJ type callus cultures .............. l03 5.l Comparative distributions of chromosome numbers from HVxHJ type A callus, roots regenerating from HVxHJ type A callus, and immature ovary wall tissue in HVxHJ regenerates ................... lO9 5.2 Mitotic anaphase cells in immature ovary wall tissue of regenerated HVxHJ plants exhibiting lagging chromosomes ...................... ll3 5.3 Distribution of chromosome counts from root tips regenerating from secondary HVxHJ type A callus and from HVxHJ progenitor suspension culture ....... ll3 5.4 Illustration of the hypothesis regarding the genera- tion of a continuum of mixtures of parental chromosomes in tissue cultures of interspecific hybrids ...... 119 5.5 Nomenclature and examples of scoring for GOT and EST bands of crude extracts from basal culm tissue of HVxHJ regenerates ................... l22 ix Figure 5.6 5.7 5.8 5.9 5.10 5.11 Typical examples of EST and GOT zymograms from crude extracts of culm bases of HVxHJ regenerates ...... Histogram of the distribution of XHJ:ZHV isozyme band intensities, grouped as follows: 0 to .l, .l to .2, etc. ....................... A comparison of plant types among plants regenerated from HVxHJ type A callus ............... A comparison of floral structures and auricle morph- ology between HV, HVxHJ, and HJ, and plants regener- ated from HVxHJ type A callus cultures ........ Comparison of metaphase I chromosome associations be- tween the original HVxHJ hybrid and HVxHJ regener- ates ......................... Plant type and spike morphology of a HV-like segregant regenerated from HVxHJ callus ............. Zymograms of HV-like and HJ-like parental segregant(s) in regenerated HVxHJ population of plants compared to HV, HJ, and HVxHJ ................... Meiosis in HV-like plant regenerated from HVxHJ type A callus ....................... Page 124 127 129 134 138 CHAPTER I INTRODUCTION The range of barley (Hordeum vulgare L. Emend Lam.) is re- stricted by plant sensitivity to high moisture, salinity, and cold temperatures. Further, the foliage of barley is susceptibhe to damage by the larvae of the cereal leaf beetle since the absence of epidermal trichomes allows the adult female to lay eggs on plant surfaces. After an initial screen of available germ plasm with respect to these traits, it was determined that little or no additive genetic variance was available for purposes of breeding and selection (J. E. Grafius, personal communication). Two techniques were available for introducing new variability for these traits into the gene pool: mutation or the introgression of genes from sexually cross-compat- ible wild relatives. Mutations occur at random in the genome and rarely manifest genetic variability in the desired direction. Mul- tiple lesions often mask rare desirable effects. Further, if a phenotype in question is complex, as with regulatory or developmental traits and agronomic characters,the probability of desirable mu- tants would require prohibitively large pOpulations for screening. Similar problems in wheat and corn had proven quite amenable to the introgression approach. Hence, in the late l9605, Dr. J. E. Grafius of the Dept. of Crop and Soil Sciences, Michigan State University, initiated an introgressive breeding program in barley with the ob— jectives of transferring desirable genes for the above traits from wild gene pools into cultiVated barley. Hybridization between species and genera is becoming an in- creasingly important means of introducing genetic variation into the gene pools of crop plants (Smith 1971). Treating the genus or fam- ily taxa as potential gene pools is of particular interest to plant breeders. The possibility exists for substituting and adding genetic information from wild gene pools into the cultivated gene pool, as has been demonstrated in wheat (Sears 1972) and corn (Galinat 1977). In wheat, interspecific and intergeneric hybridization and backcross programs have introgressed such traits as stem rust resistance (Knott l96l, Knott gtgl, 1977), leaf rust resistance (Caldwell _e_i_;_ 31, 1956, Sears 1956), and wheat streak mosaic virus resistance (Larson and Atkinson 1973). Many interspecific and intergeneric hybrids have been successfully synthesized in the genus Hordeum, opening up the possibility of introgressive breeding in barley (Nilan 1964). H, jubatum is a weedy perennial which embodies a desired range of phenotypes for moisture, salt, and cold sensitivity and epidermal trichomes. Further, it has been shown to be sexually cross-compatible with H, vulgare, the cross with H, jubatum as female parent culminating in relatively vigorous F1 plants of inter- mediate morphology (Nagenaar 1960, Rajathy and Morrison 1959). Subsequently, Steidl (1976) has performed the reciprocal cross and obtained F1 plants of much higher vigor, indicating a differential effec to re< backcr genome genera the ex; tives F T (Hagena. whereas two geno AAA'A', . GVGC. has Perhaps d with 58L Sterility ing, Seve Specific e. by InterceC COmpatibie_ that any . Ju: IISSUES of. t with 59Verai dOUbied LT] 19 3 effect of parental cytoplasms. The objectives of the program were to recover fertility in the F1 hybrid population, and recurrently backcross to H, vulgare (HV) until most of the H, jubatum (HJ) genome had been lost to segregation (approximately 4 to S backcross generations). The resulting population would then be screened for the eXpression of increased genetic variance relative to the objec- tives previously described. The F1 hybrid between HV and HJ was sterile in both directions (Hagenaar 1960, Steidl 1976). HV is a diploid (VV, 2n=2x=l4) whereas HJ has been shown to be allotetraploid (2n=4x=28) with its two genomes possessing some degree of segmental homology (designated AAA'A', Starks and Tai 1974). The F1 hybrid (VAA', 2n=3x=21), how- ever, has been shown to exhibit very little autosyndesis at meiosis, perhaps due to the presence of a homoeologous pairing inhibitor, as with SBL in wheat (Murry 1975). Hence, it was hypothesized that F1 sterility was directly attributable to the lack of chromosome pair- ing. Several attempts were made to circumvent this sterility. In specific examples with wheat, cross incompatibility has been overcome by interceding a bridge species with which both species were cross- compatible. Limited trials of this technique led to the conclusion that any juxtaposition of HV and HJ genomes culminated in F1 ster— ility. Further, attempts were made to double chromosome numbers in tissues of the F1 hybrid to induce amphiploidy. Intensive efforts with several proven techniques failed to give rise to somatically doubled tillers. It was concluded that jH_yjyg_interactions of F1 tissue resulted in the suppression of doubled sectors, or that effect to rec backcr genome generai the ex; tives p T (Wagena. whereas two gent AAA'A', ever, ha Perhaps 1 with 53L Sterilit) ing, Sev Specific ‘ 3 effect of parental cytoplasms. The objectives of the program were to recover fertility in the F1 hybrid population, and recurrently backcross to H, vulgare (HV) until most of the H, jubatum (HJ) genome had been lost to segregation (approximately 4 to 5 backcross generations). The resulting population would then be screened for the expression of increased genetic variance relative to the objec- tives previously described. The F1 hybrid between HV and HJ was sterile in both directions (Hagenaar 1960, Steidl 1976). HV is a diploid (VV, 2n=2x=l4) whereas HJ has been shown to be allotetraploid (2n=4x=28) with its two genomes possessing some degree of segmental homology (designated AAA'A', Starks and Tai 1974). The F1 hybrid (VAA', 2n=3x=21), how- ever, has been shown to exhibit very little autosyndesis at meiosis, perhaps due to the presence of a homoeologous pairing inhibitor, as with SBL in wheat (Murry 1975). Hence, it was hypothesized that F] sterility was directly attributable to the lack of chromosome pair— ing. Several attempts were made to circumvent this sterility. In specific examples with wheat, cross incompatibility has been overcome by interceding a bridge species with which both species were cross- compatible. Limited trials of this technique led to the conclusion that any juxtaposition of HV and HJ genomes culminated in F1 ster- ility. Further, attempts were made to double chromosome numbers in tissues of the F1 hybrid to induce amphiploidy. Intensive efforts with several proven techniques failed to give rise to somatically doubled tillers. It was concluded that jH_yjyg_interactions of F1 tissue resulted in the suppression of doubled sectors, or that 4 allotriploid (3x=21) cells were competitively superior to amphiploid (6x=42) cells (Steidl 1976). Tissue culture is a tool which provides the possibility of deal- ing with large populations of single cells as discrete organisms. Tissue culture was advanced as a possible means of isolating pure amphip1oid sectors by regenerating whole plants from colchicine— treated cultures (Steidl 1976). In addition, the possibility existed fbr the induction 0f variability i! vitro by mutagenesis or somatic fusion. Hence, studies on tissue cultures of Hordeum were initiated in the Fall of 1974 by Dr. T. B. Rice, then a Postdoctoral Fellow in the laboratory of Dr. P. S. Carlson, Dept. of Crop and Soil Sci- ences, Michigan State University. This Dissertation will outline the steps taken in attempts to utilize tissue culture for introgression between HV and HJ and the isolation of fertile intermediates from this cross. Figure 1.1 illustrates these steps and their corresponding interrelationships. Chapter 2 will describe efforts to define the conditions which max- imize growth and regeneration from callus cultures of HV, HJ, and HVxHJ. In Chapter 3, a technique of 1H_gitgg_doubling of chromosome numbers will be described. This technique was used for the success- ful isolation of sterile 'hypoamphiploid' plants of the F1 hybrid. Finally, these plants were analyzed cytogenetically to determine a possible cause of sterility. Previous studies indicate that cul- tured eukaryotic tissues exhibit variability in chromosome number and morphology. Chapter 4 will examine sources and degrees of 12. yjt§9_karytypic variability in Hordeum, its possible sources, and the preliminary possibility of its utilization as genetic Figure 1.1: Diagrammatic summaryof research presented in the chapters of this dissertation, indicating stepwise relationships. HORDEUM VULGARE (wv) x H. JUBATUM (HJ) HVxHJ STERILITY PREVIOUS naIaInHIuINuuInInIaInns-N-flln-n-NINln-s-nn. ‘I.NINIflIflIflIflIflINIflINIH-N-flilggfljfiifl‘flIflIfl-B THIS TISSUE CULTURE ADOPTED AS STUDY A POTENTIAL TOOL TO CIRCUMVENT STERILITY CHAPTER CHAPTER 3 MAXIMIZATION OF chpTER q DIPLOIDIZATTON SOMATIC CELL IN VITRO‘KARVOTYPIC Lg VITRO CYCLE “ VARIABILITY: VIA COLCHICINE TYPES. DYNAMICS. TREATMENT AND DISTRIBUTIONS l REGENERATION OF WHOLE PLANTS REGENERATION OF HYPOAMPHIPLOID PLANTS CHAPTER 5 REGENERATION OF KARYOTYPIC VARIABILITY: SOURCE AND DEGREE STERILITY SEGREGATION OF PARENTAL GENOMES AMONG HVxHJ REGENERATES 1 RESTORATION OF FERTILITY Figure 1.1 7 variability. The expression of 1H_yitgg_karyotypic variability in regenerated whole plants of HVxHJ will be investigated in Chapter 5. Further, a technique will be introduced for the identification of parental segregants in populations of plants regenerated from callus cultures of HVxHJ. Collectively, these findings will be shown to constitute new preliminary techniques for 1H_vitro introgression, isolating haploids, and circumventing hybrid sterility. CHAPTER II A QUANTITATIVE ANALYSIS OF GROWTH AND REGENERATION FROM TISSUE CULTURES 0F HORDEUM VULGARE, H, JUBATUM, AND THEIR INTERSPECIFIC HYBRID Introduction Realization of the full potential Of somatic cell genetics in higher plants is predicated on the ability to induce desired devel- opmental states. Callus has now been induced in a large number of species, indicating that this phenomenon is not limiting (Narayana- swamy 1977). Various manifestations and degrees of organogenesis have generally been observed subsequent to callus formation. The culmination of organogenesis, plantlet formation, is sporadic, and is a major factor limiting progress. As a group, dicotyledonous plants are much more responsive to de- and redifferentiation than monocotyledonous plants. Among the monocots, the annuals, such as wheat, corn, rice, barley, sorghum, oats, etc., are most resistant to callus formation and regeneration. Although whole plants have been regenerated in all of these groups, it has been argued (for wheat) that plantlet regeneration is not a true consequence of in- duction (Bhowanji and Hayward 1977). From most reports, it is clear that regeneration in this group is not an "all or none" response with respect to experimental conditions. There appear to be three possible reasons why it is not pos- sible to consistently obtain organogenesis in annual grains: 1) dif- ferent genotypes tend to vary with respect to their proficiency in de- and redifferentiation (Green and Phillips 1975, Jacobsen 1976), 2) callus is a term used to describe a wide range of developmental states which differ in morphogenic potential, and 3) depending on exogenous conditions, karyotypic abnormalities tend to accumulate over time in culture (Torrey 1967, Bayliss 1975, Sunderland 1977) concomitant with a decline in regenerative potential (Reinert and Backs 1968, Smith and Street 1974). The conclusions of past studies are difficult to compare because complete information regarding genotype, callus growth form, and karyotype were usually not reported. The present study was undertaken with the objective of defining the conditions required to demonstrate the somatic cell cycle in Hordeum vulgare (cultivated barley), H, jubatum, and their inter- specific hybrid. Because the regenerative response is not "all or none," I employed a quantitative analysis. The experiments reported in this chapter were designed to measure the significant effects of certain potential modifiers on tissue necrosis, callus growth, and regeneration. By a statistical treatment of data from a large num- ber of trials, significant differences among treatments were de- tected. Materials and Methods Diploid H, vulgare L. Emend. Lam. cv Coho (HV, 2n=2x=l4), H, jubatum L. (HJ, 2n=4x=28), and their interspecific hybrid (HVxHJ, 2n=3x=21, HV as maternal parent) were obtained from Dr. J. E. Grafius, 10 Dept. of Crop and Soil Sciences, Mighigan State University. Plants were maintained in greenhouses at approximately 25 to 28°C during the day and 21 to 25°C during the night. These were the conditions under which all plants grew vigorously and karyotypes were invariant. Viable, healthy tissue from various parts of whole plants were excised, surface-sterilized in 95% ethanol for 30 seconds, washed twice in sterile distilled water, and placed on agar-solidified media. If a high incidence of contamination was Obtained after this treatment, tissues were sterilized in 10% chlorox for 10 min- utes and washed twice in sterile distilled water. Callus induction media were those of Murashige and Skoog (1962)(MS, as modified by Linsmaier and Skoog 1965) supplemented with 5 mg/l 2,4-dichloro- phenoxyacetic acid (2,4-D) and 4% sucrose, and that of Gamborg and Eveleigh (1968)(BS) supplemented with 4 mg/l 2,4-D and 3% sucrose. All media were solidified with 0.9% Difco Bacto Agar. TO character- ize that developmental stage giving maximal callus induction from ovaries, the ratio of excised palea length to mature palea length was recorded. After induction, calli were maintained on 85 medium supple- mented with 3 to 4 mg/l 2,4-D and 2 to 3% sucrose in the dark at 25°C. Asceptic transfers were performed every four to five weeks. Suspension cultures were initiated by introducing vigorously growing, friable callus into the same medium lacking agar, in 125 ml Ehrlen- meyer flasks rotated at 120 RPM. Finely divided, rapidly growing suspension cultures (doubling time = 48 hrs.) were easily isolated from all three parental sources. Initially, I hoped that growth rates would show the effects 11 of media on callus growth. Uniform callus tissue was subdivided into six subcalli of equal size, plated onto fresh medium and trans- ferred every two weeks. Growth rate was measured by determining the relative fresh weight (initial fresh weight = 100%) over the course Of 40 days at four to seven day intervals. Linear regression equa- tions and 95% confidence intervals of regression coefficients (rel- ative fresh weight on time) were calculated. TO determine fresh weight (FN): dry weight (DH) ratios, uniform callus tissue was sub- divided into ten equal pieces, weighed, dried for 48 hrs. at 60°C and reweighed. Individual and synergistic effects of certain physico-chemical modifiers were ascertained in attempts to determine the medium con- ditions which elicit specific tissue responses. Callus tissue from 11 different cultures* was plated in replicate onto a medium grid consisting of 8 'Basic' x 12 'hormonal' treatments. Basic treat- ments included the following: (1) MS vs. B5 salts, (2) casamino acids without NH4+ salts vs. NH4+ salts, (3) sucrose vs. sucrose + glucose, (4) 100 mg/l vs. 5.0 g/l inositol, and presence vs. absence of (5) coconut water, (6) a water extract Of developing barley caryopses, and (7) dimethyl sulfoxide (DMSO) (see legend of Fig. 2.3 for more details). Hormonal treatments included: (1) indole-3-acetic acid (1AA), (2) kinetin (KIN), (3) benzyladenine (BA), and (4) gibberellic acid (6A3)(see legend of Fig. 2.4 for more details). Callus and suspension culture tissues were drawn from *For the purposes of this paper, a culture is defined as uniform, stable tissue derived from the same point of initial cal- lus formation. 12 cultures exhibiting uniform, stable growth habit. Within each cul- ture tested, all tissue had identical transfer histories. Fresh weight of calli varied from 0.10 to 0.20 g at the time of plating. For suspension cultures, approximately 1 m1 of settled cell volume was plated on each dish as a lawn. Callus type, age,and a profile of chromosome numbers were recorded for each culture at the time of plating. Also, tissue was plated in a similar manner onto media with various concentrations of cyclic AMP (0.0, 0.0005, 0.00025, 0.0005, 0.0025, 0.005, and 0.025%, W:V), pH levels (4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5), and selected combinations and concentrations Of free IAA and IAA-amino acid conjugates. Conjugates tested included the following: IAA-D-alanine, IAA-L-alanine, IAA-glycine, IAA-B- alanine, and IAA-D-aSpartate (see Table 2.5 for more details). Petridishes were maintained at 25°C under 16 hrs./day Of light (2000 to 3250 lux) from GE F96T10-CWX bulbs. After 50 to 60 days, the dishes were randomized and tissue responses were scored. Stand- ards fOr relative comparison were established before the scoring procedure was initiated. These consisted of drawings of minimal and maximal response which were compared with experimental tissue to determine the relative response (0 = minimum, 5 = maximum). Scores were determined for the following tissue response classes: tissue necrosis (discoloration), callus (unorganized) growth, and crown root, primary root, and shoot formation. Crown roots were tentatively distinguished from primary roots on the basis of the following criteria: crown roots lacked root hairs, were pigmented, and exhibited a characteristic negative to positive geotropism, 13 while primary roots bore root hairs, were unpigmented, and exhib- ited positive geotropism (Fig. 2.6). The variance of the scores for tissue response was partitioned and tested for each response class (i.e. tissue necrosis, callus growth, etc.) for each culture on the 8 basic x 12 hormanal grid. If a given variance was significant (a = .05), means for that re- sponse class and culture were contrasted in pairwise comparisons, in a 95% confidence family, according to the Tukey procedure (Neter and Hasserman 1974). Contrasts for each response class were grouped according to genotype for each culture in order to condense the an- alysis and identify recurrent phenomena. An arbitrary scale was constructed to report the frequency Of given significant contrasts (e.g. medium A vs. medium B) for each genotype and response class as follows: 0 to 33% of contrasts significant, 33 of 66% of con- trasts significant, and 66 to 100% of contrasts significant. To negate the effects of tissue potential (see Table 2.3 for definition), only cultures exhibiting an overall mean response intensity greater than zero were considered (e.g. if a given culture showed no shoot regeneration, it was not included in determining overall significant effects relative to shoot regeneration). Results Previous studies dealing with annual grains have reported callus induction from embryo, endosperm, root, ovary, stem section (Yamada 1977), apical meristem (Cheng and Smith 1975, Koblitz and Saalbach 1977), microspore (Wilson 1977), and rachis (Dudits et a1. 1975). In this study, callus was successfully induced from root 14 meristems, immature ovaries and peduncles, and 3 day postpollina- tion embryos and endosperms. Callus emanating from whole, immature ovaries was the most easily obtained and prolific, and was used ex- clusively for all growth and morphogenic measurements. The frequency of callus induction (% of ovaries producing callus vs. developmental stage) was determined for all three genotypes. The highest frequency of callus induction occurred between 10 to 50% of mature palea length for all three genotypes (Fig. 2.1). Additionally, over all developmental stages, ovaries borne on the basal half of the rachis formed callus more frequently than those of the apical half. Callus cultures Of HJ and HVxHJ usually continued to grow rapidly subsequent to formation. Callus from HV, however, was very difficult to propagate. Typically, periods of several months to over a year elapsed before callus growth resumed, even with con- stant subculturing. Callus cultures adapted a stable, recognizable type soon after formation (Table 2.1, Fig. 2.2). The morphology and behavior of type A callus was very similar to the callus from scut- ellum in maize (Green et_g1, 1974) and from apical meristems Of barley (Cheng and Smith 1975). Type E calli were generated at the original point of callus formation, or arose spontaneously from callus types A, B, or C. Type E calli spontaneously transformed very rarely to other callus types; (e.g. at a very low frequency, type A nodes were observed to arise within the martix of type E calli). Table 2.2 displays regression coefficients for standardized FH vs. time. For all genotypes and callus types tested, significant Figure 2.1: 15 Graph of the percent of ovaries producing callus vs. de- velopmental stage as determined by the ratio of palea length/mature palea length. 0 - HJ,. - HVxHJ, A- HV. 16 Z 85w: Azpozmu o Lo .om szomma T: .ow om ooa 17 .«umpm—osom>,anmwnw.m_Pmo pmexsocmeea maogaeoEm .mmemp awe» xgm> azx>z we masapo can mppoo m—mcwm noun mco: .mepcmcoaxo oops: mm .a: .>: -Poommm symmoop mmPnepem xpgow: m mumuzxm mzocwmepwomse a cup: cmpmcmmea is? .mpnmwee zpm>wumpme "cowumN 30pm czx>z iwcmaeo meow “Pawsxm op Leeann mco: zgm> wows: am a: gown: mppmo Fpmsm mo mumwmcou o mppmo mumemuoe umumposom>cz PFmEm co ummmcoo mpooe mensweq .meucmcoaxo zoPFox “Ampnmwew apm>wpmpmev ompmwoom .muooe exoco xrugmwpm op ouvczimeo om a: Two xpmmooP .mmco: are; ppeEm u . .Axou icoo um mogmp .mecmo on FPmEmV opmcocos mNAm Fpmo so ucmwvmeo ”Asmam mace; xgoewea .meucocoaxm zoppwa Terms mpmcwm e commv mwuo: new; .muooe czoeo appgmwpm op muvgziemo mm >1 umumeeawm appeopmxsa .mmcmd m .muposmum muooe Twewe umpewoommm xpmmopo cw new; xgmewca .mpooe zopm nzx>z iezooo m—Pwu woumpozom>c= .Ppwsm czoeo .muoocm .mecwh zeppmx : ”mono: opmeomwu mmLmF .uem; sz> < meucmuoaii suzoem. cowumpcmEmwm mezmwu mmaxpocom cowoaweomoo was» ovcomocmmeo o>Puumm c_ um>emmno mmazp mappmo mo cowqueomoo ” H.~ m4m<~ Figure 2.2: 18 Callus types in tissue cultures of Hordeum: (A) type A of HVxHJ, (B) type B of HV, (C) type (3 of HJ, (0) type D of HJ, (E) type E of HVxHJ (A to E X4.2), (F) compar- ison of the surface topology of type A vs. (G) type E callus (F and G X11), (H) histology Of HVxHJ type A cal- lus, squash preparation stained with feulgen, (J) His- tology of HVxHJ type E callus, squash preparation stained with modified carbol fuchsin (H and J X178). Figure 2.2 20 'between subcallus' differences were evident. Comparisons of mean growth regression coefficients indicated that type B callus had an intrinsic growth rate of approximately 2X that of type A callus. Growth rate of type C callus was intermediate between types A and 8. Type E calli had not been observed at the time of these measure- ments, but has been observed to grow at a much faster rate than all other callus types. Note also that FH/Dw ratios showed var- iability, especially in type E callus. Different sectors of the same callus, although morphologically uniform, showed intrinsically dif- ferent growth potentials and degrees of hydration. Pertinent characteristics regarding callus from the eleven cultures plated on the basic x hormonal grid are listed in Table 2.3. Cultures were grouped according to callus types to stress similar- ities within these types. No genome- or type-specific patterns were apparent for tissue necrosis. Relative callus growth was man- ifested similarly to the manner described above. Deviations from the expected hierarchy (i.e. type E > B > C > A > 0) may have been due to the lack of 2,4-D in the test media. Type A calli exhibited a high spontaneous frequency of shoot formation, even when cultured on 4 mg/l 2,4-0. Type A callus cultures retained the ability to regenerate shoots up to 24 months after initiation. No regeneration of shoots was observed in types B and C calli although the extent Of primary and, in most cases, crown root regeneration was comparable to type A callus. Type E calli and suspension lawns (which were derived from type E calli) exhibited little or no organogenesis. Insufficient amounts of callus type D precluded testing on this grid. In separate trials, however, type D callus demonstrated no 21 .0, u z mmczupzo osmm on“ we mammwu sore :mxmo mew; u:n..mcowu -mcmEempoo mum; zozoem toe com: mzmm_u wsmm «so sore umpopaopmo no: $3.;R ¢S.N 4N.ON - - - N - a=x>x afi.o sm.o < s OH.o ms.o < m mo.o NN.o 4 a mo.o NN.o < N OH.o so.o < N Hm.o ms.m H~.o No.o NN.o < H s=x>= NN.o oo.o m s mH.o os.o N m mH.o No.H m a om.o om.c N m mfl.o Ns.o m N Am.o NH.ON ON.o mH.o so.N m H >= NH.o mm.o u s NN.o NH.o o m mo.o Na.o u a Nm.o 33.0- o N so.o mo.H u N om.o NN.HH Nfi.o HH.o No.0 8 A a: Izo\3u .o.m .zc\=u cam: Hm cam: Hm om mass m=__aua=m Acapaxa mappmo paucmcma mewp no .u: smog» mo cowmmoemmm m:—_eo Eamceoz groove: xppeowmopocaeoe mo VF_mon:m macaw cavemen»; use mower cpzogm m>wueemasou "N.N ~4m

.gau m=__mu .saeuoe c_pom an «co—axe an we neptau—auaam paveum soc» a_uumc_u uo>.cou m=——ao «map—cu acme—Lav .nmo—u announce ou uuonmug :u—x asap—so cu>va a we xu—mcou:. uncommon cams p—oso>o "map—cu xgaceouom N S Sewage moaxu map—cu so covuacapaxa go so» u «pan» mom passes a < .Pe'ccacoa - a “agape ooeaom you n 8. ON. 3. NS. 3. 2.8 8.: S s .oziew a 3:3 a < 7:. 3 co. co. co. eo.~ AN. um.om~ mm.Ne “N N m.oz ow . coamcoamam a u<-a:x>: Ac" co. co. co. no. on.“ m~.mao~ N~.~m ”N o e.az cu m ans—poo am a smiaxx>= Am oo. oo.. No. Na. Ns. n~.moa mm.¢m "N o .o: m m mappau 6" a smiazx>= Am mo. a". mm. mm. mm. "c.w Ne.mH "N c .o: «N < map—nu a" a niazx>z as on. on. mg." an. me. om.o~ mH.~N “N a .a: a < mappou o“ a smiazx>= Am A". Nu. aN. mu. Ne. ~¢.o .Nm.o~ mN o .o: NH < mappoo o— a «Niaz .m o". m“. Ne. mm. mm. N~.N mm.- wN o .o: o“ < mappau on a Nfiiaz Ac 8. NS. 3. NN._ mm. - - 3 s .2 NN o 3:8 ON N TI... 3 8. :. N. 3. R. 8.2 SAN N a .3. SN 8 3:8 a" a "4-2. .N co. mm. mg. ac. Nm. mo.m c¢.m~ «N o .oz e“ u wasp—cu on a “Tax A” S d 3 3 u .305: $23: “.53 2333.5 one on»... vane:— 33: 83—3 m. mu. mm m m w s usages“. «Sneeze 8.53 e2. 33. 23:6 325 333:. .5395 1 4w Wu Wm mm .m.: :3: an .M u. s N" upavucauoa .uweo occasoz x uvmaa on» er cone—p»: nasaupau yo mama.» mappau 9:.ucemos uuwuuvsouuasagu acocvusoa yo accessm an.“ m4m) 33 - 66% of row means significantly greater than column means. ‘) 66-- 100% of row means significantly less than column means. D) 66 - 100% of row means significantly greater than column means. Example: 3 <1 b 2 (3 - For 33 - 66% of contrasts, the mean response of treatment 1 is significantly less than treatment 2. - For 66 - 100% of contrasts, the mean response of treatment 1 is significantly greater than treatment 3. 26 HI HVxHJ 12345678 I <1 <1 <- <1 I 2 <1 <1 d <1 <1 2 . 3 3 T455“? 4 q 4 necr05is 5 5 6 6 7 7 a B 1 l 2 2 3 3 4 4 5 5 6 6 7 7 a 8 aVouuh-‘Mm 12345678 D Primary roots A D D 12345678 12345678 <1 MDUM" Figure 2.3 l 2345678 2345678 23745678 \l [[11 b22152 S OVOuLUNH 27 Figure 2.4: Significant pairwise contrasts for hormonal treatments, broken down according to genotype and response class (row preferred). Treatment 1 - NO hormones 2 - 0.5 mg/l IAA 3 - 0.5 mg/l IAA + 0.3 mg/l KIN 4 - 0.3 mg/l KIN 5 - 0.5 mg/l IAA + 3.0 mg/l BA 6 - 3.0 mg/l BA 7 - 6.0 mg/l BA 8 - 0.3 mg/l KIN + 3.0 mg/l BA 9 - 0.5 mg/l IAA + 1.0 mg/l GA3 10 - 0.3 mg/l KIN + 1.0 mg/l 6A3 11 - 0.6 mg/l KIN + 1.0 mg/l GA3 12 4 0.5 mg/l IAA + 0.3‘mg/l KIN + 1.0 mg/l GA3 (See legend of Fig. 4 fOr key to symbols used and example) 28 HVxHI 45678910 H1 :.N mm=m_¢ IO 123456789MH I .Q‘C ”D n) m. 9D 8D 7D 6D 5D 8 4 7 Q 0 HQ ““ ‘ ‘ PCKQQCQQ 2 G 123456789mun 30 11 6789 0'2 171456789WH D D DD D 8 910111? 7 4 d << 56 <4 <1 4567891011 4 J Sm .ww .lam C8 Tissue necrosis 29 Figure 2.5: Comparison of significant contrasts for hormonal treat- ment between HVxHJ types A and E calli (see legend of Fig. 2.3 for key to symbols used and example). 30 HVxHJ _HVxHJ typeA typeE 12345678910111? 12345678910110 D DDDD DD D DDD Callus growth Figure 2.5 31 TABLE 2.4: Statuary of significant Basic and hormone effects from Figs. 2.3, 2.4, and 2.5 Basic treatments Hormone treatments Response class Genotype inhibitors promoters inhTBitors promoters Tissue necrosis HJ - as salts - BA HV - 85 salts null hormones KIN + BA HVxHJ - as salts IAA + 6A3 - Callus growth HJ - - - no hormones, possibly BA av - - - no hormones HVxHJ type A - - - KIN + 6A3 HVxHJ type E - - - no hormones IAA, IAA + KIN, KIN, IAA + 6A3 crown roots HJ high inositol H5 salts KIN + 6A3 possibly BA HV - - - no hormnes, IAA. IAA + KIN . KIN HVxHJ - - KIN + 6A3 - primary roots HJ - as salts + - - caryopsis extract "v - - - KIN + 6A3 IAA + GA3’ HVxHJ BS salts + - - 6A3 coconut water shoots NJ - - _ - - HV no basis no basis no basis no basis HVxHJ - as salts + - KIN + 6A3 11 sucrose + IAA + KIN + 6A3 11 glucose 32 a single factor analysis of variance was performed on response in- tensities for each response class. Further, regression coefficients were calculated fOr log10 [CAMP] and pH vs. response intensity for each response class. All between-treatment variances and regression coefficients were insignificant. Amino acid conjugates of IAA are found in numerous plant groups (Schneider and Wightman 1974). It was hypothesized that they may serve as pools for the storage and/or detoxification of IAA (Feung gt_gl, 1977). Recent studies have indicated that IAA-amino acid con- jugates may be more powerful auxins than free IAA by serving as slow- release auxin sources (Peterson 1978). Concentrations of 0.0, 0.01, 0.1, 1.0, and 10.0 mg/l IAA equivalents of various IAA-amino acid conjugates, 9 conjugatezl free IAA mixtures, and free IAA controls were tested for their effects on the tissue response classes pre- viously described. Two independent cultures here utilized (HVxHJ-AE suspension and NV type B). The mean intensity was calculated over four concentrations for each conjugate. The results are presented in Table 2.5. All conjugates promoted tissue necrosis in comparison with free IAA, IAA-glycine and IAAwL-alanine being the most effec- tive. In addition, all conjugates stimulated callus growth, the strongest effects evident from IAA-L-alanine. Conversely, conju- gates tended to suppress organogenesis, with the possible exception of IAA-D-alanine. No differences between 9 conjugatezl free IAA mixtures and conjugate alone could be detected. The regeneration of shoot meristems was the most limiting transformation in the barley somatic cell cycle. Once shoots were regenerated on calli, roots could easily be induced by transferring mappmo m mean Hi>z AN menopaoo :zmp copmcmamam azx>z AH mgzupao 33 a N< hypoamphidiploid E hypoamphidiploidB - ‘1. ‘I 2 l_____- - l X hypoamphidiploid E hypoamphidiploidC Figure 3.2 52 oo.o MN.o Hm.H Hm.NH eH.o oe-Nm oo.om NH - m oo.o oo.o HN.N oo.NH me.m mm-Nm me.mm N - N oo.o HH.o ow.o em.e~ mm.m Ha-em ee.Nm m m oo.o oo.o ON.N oo.mH om.m me-em ON.wm m N oo.o NH.o mN.N mm.o~ NN.N mm-Nm mm.mm m N oo.o mo.o me.fi em.NH mm.m me-Nm Ho.Nm NN - m oo.o oc.o mN.H om.oH oo.N Nm-Hm mN.Hm a m H~.o “H.o NN.N oo.m~ No.4 ee-Nm oo.oe m N ON.o oe.o ON.N ON.HH om.m me-mm ON.Nm m H HH.o NH.o mN.N em.HH ee.m ee-~m mm.Nm NH - NH.o Nfi.o mm.H om.NH mm.N ee-mm mm.mm NH H cN.o ON.o mm.H oo.N~ ow.N ae-mm mN.Nm ma - No.o mH.o am.H ON.NH mm.e ee-Hm me.Nm mm - H > a H: : H was”. 89.5: 38 has: 3:: as: Ppmu con mcowpmwoommm msomosocgu .02 .o: com: cum: .paeepe use .mepwp .ucmpa mugzom op mcwccouom czou cmxogn .muropawcqsmoaa; umum icmcmmmg macaw mosomosogcu Co mcowumwuommm new .mmmcmc .mccms we xgmeszm ”_.m MJmr Ce o.om emiw o~.m~ flu mstau < mama aazx>z o.- mmuofi Hmiom - pmcwmwgo mmPqu vwopnsmc< a madam amass: sagas: «name» msomoEoggu msomoeoccu com: umuumaxu wzx>z seem um>wgmu mmzmmwu soc» mgmnszc msomoeoczu we macaw; new mamas mo cemwgmasou "Nd m._m<._. 60 profile: 4.48 I + 15.26 11 + 0.08 111 + 0.41 IV + 0.01 V + 0.01 VI (adjusted for the random loss of homologues). The differences be- tween these and observed mean associations per cell (Table 3.1) are more easily explained by competitive homeologous pairing than non- random loss of specific chromosomes. This evidence supported the conclusions that chromosomes were lost at random. Further, direc- tional chromosome elimination is generally concomitant with a shift in phenotypic characters in the direction of the parent whose chro- mosomes remain. This phenomenon was not evident among hypoamphiploid plants. The degree of bivalent pairing in the hypoamphiploids should not have precluded fertility (Table 1.1). Cytological observations of microsporogenesis provided evidence for a breakdown of meiosis cul- minating in sterility. This breakdown was manifested in two cyto- logically distinct phenomena: 1) Chromosomes did not interact norm- ally to spindle forces; they tended to form bridges and/or fragment (Figs. 3.3 E, 3.3 F, 3.3 G, 3.3 H, 3.4 A, 3.4 B, 3.4 C). No dyad chromosomes were observed, leading to the speculation that premeiotic S phase was missing. This observation could explain the abnormal be- havior observed at anaphase II by assuming that spindle forces had been applied to unineme chromosomes. 2) The spindles themselves appeared to be abnormal. At anaphase I they were invariably diverg- ent (Fig. 3.3 F). At anaphase II, supernumerary spindles were fre- quently observed (Fig. 3.4 B). A breakdown of meiosis in inter- specific hybrids, concomitant with abnormal spindle expression, has been reported in a Lolium-Festuca derivative (Darlington and Thomas 1937). 61 In previous studies of HVxHJ, supernumerary cytokinesis and pollen formation occurred immediately after the abortive meiosis I division. No evidence could be found for the existence of any meiosis II-like divisions, similar to observations of certain wheat-rye inter- generic hybrids (Bennett gt_gl, 1972). It was proposed that abnormal meiosis in HVxHJ was a consequence of physio-genetic incompatibil- ities (Murry 1975). An anomalous meiosis II was observed in micro- spore mother cells of hypoamphiploids of this cross. Perhaps chrom- osome doubling, loss, or pervasive aneuploidy obviated some of this incompatibility thus permitting meiosis II to proceed. The sterility of hypoamphiploid plants of this study was attributed to a breakdown in the process of meiosis itself, and not a lack of sufficient chromosome pairing. Mass pollinations of florets with fertile pollen from both parental sources yielded no embryo formation. It was concluded that these plants were female sterile as well, the causes of which remain unknown. It is likely that this phenomenon is a consequence of incompatible genome interactions in the cell cycle (Murry 1975). Steidl (1976) has speculated that all interspecific hybrids which juxtapose the genomes of HV and HJ might be sterile for this reason. The induction of amphiploidy has been shown to be extremely difficult among F1 interspecific and intergeneric hybrids of the genera Hordeum and Agropyron (Steidl 1976). In this study, however, approximately 60% of regenerated tillers from colchicine-treated calli of allotetraploid HJ were octaploids of high fertility. Hence, in the absence of genomic incompatibilities which cause sterility, the use of tissue culture is a potentially powerful tool for the production of fertile allot and autopolyploids. CHAPTER IV KARYOTYPIC VARIABILITY IN TISSUE CULTURES OF HORDEUM Introduction Variation of chromosome number and morphology in cultured plant cells is a widely documented phenomenon (Partanen 1965, D'Amato 1975, Sunderland 1977). No patterns are discernible in the genera- tion and distribution of this variability among and between plant groups. Even callus derived from different tissues of the same plant often give rise to divergent karyotypic profiles (D'Amato 1975). Karyotypic variability has ambivalent consequences. The phen- omenon is disadvantageous in many respects. Genetic and biochemical studies generally assume that somatic tissues are genetically uniform. Often, the conclusions of such studies are weakened by the possibil- ity of variable chromosome constitution. Karyotypic variability also has potential advantages. New avenues may be opened to the production of novel gene associations and chromosomal mutations useful in basic sicence and plant breeding (Sunderland 1977). Most of the preceding reports concerning karyotypic variabil- ity have utilized material that is convenient, such as Nicotiana tabacum (Cooper gt_ 1. 1964, Shimada and Tabata 1967, Kallak 1968, Shimada 1971) and Daucus carota (Bayliss 1973, 1975). Haplopappus gracilis has been used extensively due to lower chormosome numbers 62 63 (2n=4), larger chromosome size, and morphological differentiation (Mitra and Steward 1961, Singh gt_al, 1975, Singh and Harvey 1975). Crepis capillaris has the advantage of exhibiting only 2c nuclei in differentiated tissues, unlike most other plant groups, making stud- ies of jg_!jtrg_polyploidization more tenable (D'Amato 1975). A paucity of reports exists concerning karyotypic variability in cell and tissue cultures of cereal grains, probably due to difficulties in elucidating the conditions for callus induction, growth, and re- generation. Such variability has been reported in callus cultures of wheat (Kao gt_gl, 1970, Shimada 1971), rice (Nishi and Mitsuoka 1969), and rye (Asami gt_al, l976). Cytogenetically, a number of advantages are apparent in these plants. Chromosomes are generally large, well differentiated, and cytogenetically characterized. In many cases marker genes have been mapped to chromosomes, providing a means of correlative cytogenetic and genetic studies. Also, the possibility exists to identify chromosome arms via C-banding tech- niques (Vosa 1975). In barley, the use of introgressive breeding to introduce gen- etic variation has been hindered by the pervasive sterility of inter- specific and intergeneric hybrids (Huang 1975, Steidl 1976). It has been concluded that sterility is due to a genetic incompatibility leading to a breakdown in gametogenesis (Murry 1975, Chapter 3). It was hypothesized that the induction of karyotypic variability among cells of such a hybrid may result in the non-equational dis- tribution of chromosomes and chromosomal segments, giving rise to cells regenerable into potentially fertile plants. This chapter will attempt to demonstrate the preliminary feasibility of this hypothesis 64 by examining the following: 1) the rate of spontaneous ig_yitrg_poly- ploidization and aneuploidization vs. differentiated state, 2) the distribution of chromosome numbers jg_vjtrg vs. the differentiated state, 3) spontaneous in_gitrg_chromosomal rearrangements, and 4) the isolation and clonal propagation of karyotypic and genetic variability. Materials and Methods Plants or seed stocks of Hordeum vulgare (HV, 2n=2x=l4), H, jubatum (HJ, 2n=4x=28), and their interspecific hybrid (HVxHJ, 2n=3x=21) were provided by Dr. J. E. Grafius, Dept. of Crop and Soil Sciences, Michigan State University. All plants were produced or maintained as described by Steidl (1976). Callus and suspension cul— tures were isolated and maintained as described in Chapter 2. All tissue was maintained on B5 basal medium (Gamborg and Eveleigh 1968) supplemented with 4 mg/l 2,4-D and 3% sucrose (w:V) at pH 6.0. Cul— tures derived from immature ovaries were used exclusively for this study. Cells were prepared for cytological observation as described in Appendix B. It was more difficult to make accurate counts on cells with progressively higher chromosome numbers. Thus, if all cells with uncertain metaphase chromosome counts had been excluded, the data may have become biased in favor of lower chromosome numbers. To alleviate this possibility, the following procedure was adopted: each cell was tallied three times. If the difference of the high- est and lowest of three counts exceeded 10% of the mean, the datum was discarded. Means and ranges, to the nearest whole number, were recorded fbr each cell within tolerance. For each experiment, a 65 coefficient of the degree of certainty (DC) was calculated as fol- lows: DC = l - X(high - low count)/Z(mean). With three exceptions, this value ranged from 0.9 to 1.0. Crude extracts of type E subcalli were prepared, electro- phoresed, and stained for peroxidase and esterase isozymes to detect genetic variability as described in Appendix A. Equal amounts of tissue were crushed and the crude extract taken up into Beckman electrophoresis paper wicks cut into equal sizes. Gels were scored qualitatively and quantitatively for peroxidase (PRX) and esterase (EST) bands using bands from source callus extracts as a control and reference for relative intensities. The following arbitrary scale was used: 0 - band absent relative to control, .5 - band present at l/2 the intensity of the control, 1 - band present at equal intensity of the control, and 2 - band present at greater than or equal to 2X the intensity of the control. Results Of 20 chromosome counts performed on root tips cells of HVxHJ, no variation from the expected 21 was observed. In previous studies on microsporogenesis of this hybrid, however, chromosome numbers were observed to vary from 12 to 22 (Rajathy and Morrison 1959), Murry 1975) and exhibit 12% aneuploidy (Murry 1975). This could have been a consequence of developmental or environmental factors. Further, it is conceivable that ig.vjtrg_variability of chromosome numbers was, in part, pre-existent in the original explants of HVxHJ. Chrom- osome counts from root tips of HV and HJ were invariably 14 and 28, respectively. 66 Differences in mean chromosome number and coefficient of variaef tion (V) were found to exist between cultures and over time (Table 4.1). Figure 4.1 illustrates the dynamics of chromosome numbers over time in cultures of HVxHJ. The mean of type A callus remained stable at approximately 18 to 21 over the period studied (16 months, starting from callus induction). After six months of subculture, type E cal- lus was spontaneously generated from type A nodes of the culture studied. Mean chromosome numbers in this type E callus increased dramatically to approximately 54, then decreased to about 35. Sub- sequent observations indicated that the mean had stabilized at 35 to 36. Initially, the mean chromosome number was seen to increase even more dramatically over time in suspension cultures generated from type E calli. As with type E callus, this rapid increase was fol- lowed by a plateau, and rapid decrease. When these suspension cul- tures were plated onto solid media, secondary type E calli were pro- duced. In these cultures, chromosome numbers continued to increase to approximately 58, then levelled off and decreased gradually. Ex- amples of typical metaphase plates are pictured in Fig. 4.2. In addition, coefficients of variation (V) for each datum point in Fig. 4.1 were calculated and plotted vs. time to gain in- signt into the dynamics of chromosome number dispersion (Fig. 4.3). In type A calli, a gradual exponential increase in V was observed over the 16 month period. Type E callus exhibited a rapid increase in V that levelled off after 4 months and subsequently remained stable. In type E-derived suspension cultures, the pattern was sim- ilar, only more rapid, as with corresponding mean chromosome numbers (Fig. 4.1). Further, variability was observed to decrease rapidly. 67 TABLE 4.1: A comparison of means and distri- butions of chromosome numbers among cultures derived from HVxHJ-57 Mean Age chromosome Culture months number DC S.E. V N Type A 6 19.03 .92 .317 .090 30 Type A 10 21.19 .92 .748 .154 20 Type A 15 18.10 .93 2.579 .767 30 Type E, 10(6)* 54.56 .92 6.260 .538 23 primary Type E, 15.5(6) 34.95 .93 .710 .541 710 primary Suspension 8(7,6)* 48.00 .88 5.564 .478 18 Suspension 10(7,6) 50.38 .92 3.758 .365 25 Suspension 12(7,6) 38.43 .94 1.004 .131 25 Type E, 11.5(9, 57.12 .95 5.565 .390 16 secondary 7,6)* Type E, 13(9, 52.67 .92 4.795 .490 30 secondary¥_ 7,6) *Age of initial callus induction, followed in parentheses by the ages at which preceeding transformations were induced or occurred spontaneously (e.g. 11.5(9,7,6) implies type E secondary callus 11.5 mo. old from original callus induction, transformation to type E callus at 6 mo., to suspension at 7 mo., and to secondary type E at 9 mo. . 68 Figure 4.1: Dynamics of mean chromosome number from callus induction (time = 0) to 16 months, vs. culture state. MEAN CHROMOSOME NUMBER 80 70 60 50 4o 30 20 IO 69 I o EXPLANT A—A TYPE A CAL/.05 . O—O PRIMARY TYPE E CALLUS A- -A SUSPENSION _ O- -O SECONDARY TYPE 5 CALLUS 0 30.70 \ b k .o~ fi‘ )- 0 2 4 6 8 IO 12 TIME (MONTHS) FIGURE 4.1 Figure 4.2: 70 Karyotypic variability in tissue cultures of Hordeum. (A) metaphase cell of HVxHJ type E callus with 17 chromosomes (X 936), (B) metaphase cell of H0 type A callus with 22 chormosomes (X1067), (C) metaphase cell of HVxHJ suspension with approximately 36 chrom- osomes (X1076), (0) Highly polyploid metaphase call of HVxHJ type E callus (X1017), (E) Chromosome size differential in metaphase HVxHJ type E callus cell (X4393), (F) Dicentric chromosome in metaphase HVxHJ type E callus cell (X4296). 71 Figure 4.2 72 Figure 4.3: Dynamics of the coefficient of variation (V) of chromosome numbers from callus induction (time = 0) to 16 months, vs. culture state. 73 w. m.: mm=e_m arr—.205: mi; N. o. m m c _. N «43439 N MGC XQVQEQQNM. 01 IO . Emestmhm. 4 I 14 «Sud—VG .w NQXK XQSSQQ. .1110 «54:6 v, Moxk dilld Nd .vd v.0 m6 . 0.. NOLLVIHVA :10 lNBIOIJABOO 74 In secondary type E calli, V decreased at a slower rate than the source suspension culture, followed by a gradual increase. The power of comparing means and coefficients of variation would be weakened if the distributions of chromosome numbers were not approximately normal. The distributions of chromosome counts of type A callus were approximately normal (Figs. 4.4 A, B). A comparison of the distributions at 6 and 15 months illustrated the above observation that the mean was stable, while variability tended to increase. Type A callus, therefore, exhibited aneuploidy, but no polyploidy. The distribution of chromosome numbers for type E callus at 16 months (see Fig. 4.1), however, was not normal (Fig. 4.4 C). Rather, the distribution appeared to be three to four modal. An approximately equal number of counts were centered on clusters at 20 to 22 and 38 to 40. Minor clusters were apparent at 70 to 75 and 140 to 150. It was concluded that the pattern of these clusters approximated a hypoeuploid series with flanking aneuploidy. Polyploidy and aneuploidy were also prevalent in suspension cultures (Fig. 4.4 D). These data were taken from a relatively young culture which had probably not reached steady state (see Discussion section for explanation). An insufficient number of counts of secondary type E callus were recorded to give an accurate overall picture of the distribution (Fig. 4.5). Once again, polyploidy and aneuploidy were evident. Clusters of cells at 20 to 25 and 54 to 63 were noted. It is conceivable that jn_vjtrg_karyotypic variation in cul- tured cells of HVxHJ could have arisen purely as a consequence of parental genome interactions rather than as a result of the cultured 75 Figure 4.4: Composite distributions of chromosome numbers in tis- sue cultures of HVxHJ. (A) HVxHJ type A 6 mo., (B) type A 16 mo., (C) HVxHJ type E 15 mo., and (D) sus- pension 10 mo. RELATIVE FREQUENCY 76 o, O AL V! IO 20 30 4O 50 60 7O 80 CHROMOSOME NUMBER ‘ FIGURE 4.4 140 150 77 Figure 4.5: Distribution of chromosome counts in HVxHJ secondary type E callus 78 cum 3N com m5 552“. mmmz=z mzomozomzu ONH o: 03 cm om E om om 0: cm om 3 m344h >mm a: mH NN - HN me.m mm. oe.mN u a: NH om - ON em.N - Nm.eN < a: eN on - mN NH.N - om.>N < a: mN HNH - eN me.mm mm. NH.oo eeHmeaam>m >1 NH ea - NH - - Ne.Hm N >: m Hm - mN - mm. ee.mm N >: z «mama .o.m on conga: aqxw_ “seam muszom QEOmoEogzo cam: msaupau a: wee >: co magaupsu mammwu cw mcmaE== meomoeoccu mo >gmsszm Hm.¢ ~4m voduAmWS m—zomOs—Omro mekmozm 0... Pmmuzo... _ _ LEO—.2”. 71x2... 132.050 W::::::::: H .wmePJDo \QDMQQQI 2. “6.5% .mZOEDon 529.20.17.30 .mzo.._.e >=x>= .aHee o o o o~.o eo.m oo.H oo.HN .Emumwcms poem mucmsmmam uwgucmuwcu mu_>u:m0wu m\4 me ca amass: 03mm?» PHmU\.oc com: o m cam: meowmmmczu amiwzx>z mo mwzmmwu vagaupao cw mowpmc m\4 we camwcmaeou Hm.¢ u4m~ mm. m Ioh 2.3.2.. “.0 23(1— 91 TABLE 4.5: Means of chromosome numbers and PRX and EST isozyme band intensities among 35 subcalli of HVxHJ-57 type E callus. Mean S.E. Means of 6 subsubcalli - chromosome chromosome PRX band EST band _§g§callus number DC number I 2 3 4 5 1 2 3 4 1 53.05 .96 6.58 0.0 0.0 0.0 0.5 . 0.5 0.8 1.0 1.2 1.2 2 41.47 .93 2.30 0.0 0.1 0.3 0.6 0.7 1.0 0.7 1.5 1.5 3 33.00 .93 6.34 0.2 0.1 0.5 0.5 0.4 0.7 1.3 1.3 1.7 4 41.32 .94 1.63 0.3 0.3 0.8 0.8 0.8 1.0 1.0 1.7 1.5 5 42.56 .94 2.75 0.2 0.2 0.8 0.5 0.5 1.2 1.2 1.0 0.8 6 37.30 .92 2.97 0.0 0.0 0.6 0.3 0.4 0.5 1.0 1.0 0.7 7 48.76 .94 6.25 1.0 0.5 2.0 0.5 0.5 0.5 1.7 1.3 1.7 8 32.53 .95 2.60 1.0 0.0 0.3 0.5 1.0 0.8 1.7 1.7 0.8 9 35.90 .93 1.72 0.6 0.2 0.9 0.5 0.8 0.8 1.0 1.2 0.8 10 41.06 .94 4.64 0.3 0.1 . 0.1 0.5 0.5 0.5 0.8 0.8 0.8 11 37.33 .93 2.83 1.0 0.2 0.0 0.9 0.4 0.5 1.0 1.0 0.8 12 20.16 i .96 1.76 1.5 0.0 0.4 0.7 0.5 0.2 0.8 1.2 1.1 13 33.53 .92 3.17 0.0 0.0 0.4 0.4 0.5 0.7 1.7 1.7 1.7 14 19.50 .96 1.03 0.0 0.1 - 0.9 0.5 0.3 0.5 0.8 1.5 0.5 15 24.11 .93 1.43 1.0 0.2 0.7 1.0 0.1 0.5 0.9 0.9 0.8 16 20.55 .96 1.27 - - - - - - - - - 17 42.53 .91 2.90 0.8 0.8 0.8. 0.7 1.0 0.8 1.2 1.2 0.6 18 43.12 .90 6.50 0.0 0.3 0.8 0.6 0.4 0.2 0.6 0.9 1.2 19 37.27 .94 1.17 0.0 0.1 0.9 0.3 1.0 0.8 0.8 1.0 1.0 20 59.95 .94 7.96 ' 0.8 0.4 1.5 0.9 0.9 0.5 1.8 1.8 1.5 21 25.30 .96 2.00 1.8 0.3 1.8 0.4 0.9 0.2 0.9 1.0 0.7 22 23.51 .95 2.99 0.8 0.0 1.7 0.4 0.7 0.5 1.0 1.5 1.7 23 50.05 .93 6.36 1.0 0.0 0.4 0.5 1.0 0.2 0.9 1.0 0.7 24 49.19 .91 4.70 0.0 0.1 0.4 0.3 0.5 0.4 1.5 1.5 1.5 25 23.03 .94 1.56 1.7 0.2 0.8 0.6 0.5 0.8 0.8 0.8 0.8 26 41.16 .92 3.93 1.0 0.0 0.8 0.6 0.5 0.5 0.7 0.7 0.5 27 50.00 .93 7.39 1.0 0.0 0.8 0.4 1.3 0.6 1.5 1.4 1.7 28 31.81 .93 3.18 1.0 0.2 0.8 1.0 1.0 0.3 1.5 1.5 1.5 29 28.29 .92 2.79 0.0 0.0 0.8 0.3 0.4 0.3 1.7 2.0 2.0 30 40.11 .92 3.08 1.0 0.1 0.0 0.6 0.0 0.5 0.5 0.9 0.2 31 38.67 .92 2.65 1.0 0.0 1.7 0.5 0.9 0.5 0.5 0.5 0.5 32 37.50 .94 3.98 1.9 0.2 1.5 0.9 0.8 0.8 0.8 0.8 1.7 33 37.61 .94 2.32 1.5 0.0 1.5 0.5 1.0 0.5 2.0 2.0 1.3 34 21.11 .95 0.61 1.2 0.2 1.3 1.0 0.5 0.2 1.0 1.0 1.7 35 26.13 .94 1.8247 0.9 0.3 0.1 0.7 0.1 0.6 0.6 0.6 1.3 92 Figure 4.9: Nomenclature system and examples of typical scoring for cathodal PRX and anodal EST bands from crude extracts of HVxHJ type E callus. 93 3, 23w: a mm. m o m. K. a m o Vm. m «v. N H no. _. xma 2.030. .. H .8. m No. _. N ..._ m on. N N H. m. @m. m N .5... H III..II. mm. H... emu A! I S 3 «SJ 3 v» m 3. DZ o > w SuhSub 5;“ subsubcalli subsubcalli :3;3 subsubcalli ! flIUIII»————— —‘———'—‘ _— .n y 0 run ”mu: mm 9". - ‘ ' 2'. .001..-. - z:2229::—:.'¢9' “ - ~ ._ '8 0 L5 :32 13:5 fibsubcalli > éégm subsubcalli sub_s_u_b_c_all1 18:2! subsubcalli ' 1pm: .111le um?" mm a ‘ "3‘i — --.' 'Og‘: 3- o...- é‘ A x:;::: a. :_ ‘33 lazuli Ox“. . to. ‘1 La L——————J com: msomoeogso msomoeogsu . > cam: osmmwu o>P>.mw nmumcmcmm Figure 5.2: Figure 5.3: 112 Mitotic anaphase cells in immature ovary wall tissue of regenerated HVxHJ plants exhibiting lagging chrom- osomes (A) X1154, (B) X1480 Distribution of chromosome counts from root tips regen- erating from secondary HVxHJ type A callus and from HVxHJ progenitor suspension culture 113 FREQUENCY .091 .06‘ .031 . 16‘ .03 Figure 5.2 lllll Ill llllLll | IL REGENERATING ROOTS FROM SECONDARY TYPE A CALLUS 11.. I V I v V 10 2o 30 no so 60 70 8O 93 160 110 CHROMOSOME NUMBER Figure 5.3 114 as compared to type A calli. Chromosome counts were recorded from root tips regenerated from these nodes. The distribution approximated assymetric normality around a mean of 20.00 (Fig. 5.3). In the course of differentiation, selection for a specific range of chromosome numbers had occurred (Fig. 5.3). Prohibitively small amounts of this de novo nodular callus precluded a comprehensive karyotypic analysis. Only three counts could be obtained, ranging from 18 to 2l. Hence, it was impossible to determine conclusively whether the attenuation of karyotypic variability in these regenerating root tips was a consequence of the type E-type A or type A-regenerated root transformation juncture. Previous studies of barley tissue cultures conclusively demon- strated the existence of structurally rearranged chromosomes (Chap- ter 4). Are cells with karyotypes containing these rearrangements capable of differentiation and fixation into somatic plant tissues? Longestzshortest chromosome ratios (L/S) were calculated from cells of primary and secondary root tips as described in Chapter 4. These ratios are displayed in Table 5.1 with those of progenitor cultures. In root tips regenerated from primary type A callus, the mean L/S ratio had decreased from 2.15 to 2.03, indicating possible selection forcells with karyotypes without rearranged chromosomes. Data from root tips regenerated from secondary type E-derived type A callus demonstrated a dramatic decrease in mean L/S from 3.04 to 2.14. Chromosomal rearrangements must have been present in the karyotypes of some of the cells of these regenerated root tips since the mean of this sample was significantly greater than 2.04. Further, the corresponding frequencies of multicentric chromosomes were consistent 115 with these conclusions. Cytogenetically, regeneration was a differentiative process characterized by the following karyological phenomena in this study: l) The dispersion of chromosome numbers decrease following regenera- tion, 2) Only mixodiploid cells were present in regenerated tissues, and 3) Regeneration tended to select for karyotypes with a lower frequency of rearranged chromosomes than that of the source culture. To realize the full potential of genetic studies and selection, the regenerated pool of karyotypic variability must be expressed between plants, and not within plants. Potentially, three processes can be envisioned: l) A small number of cell initials with a lower karyotypic variability relative to the source culture gave rise to each regen- erative entity, which then propagated clonally giving rise to karyo- typically 'uniform' plants, 2) A large number of initials exhibiting karyotypic variability approximately equal to the source culture gave rise to each regenerative entity, which then propagated clonally, giving rise to chimeral plants, and 3) After regeneration, new karyo- typic variability was generated, giving rise to mixoploid tissues in conjunction with cases 1 or 2. To gain an insight into the path- way of karyotypic variability into regenerated plant tissues, chromo- some counts were categorized within and between regenerated structures. For roots regenerated from primary and secondary type A callus, the variances attributable to between calli and within calli between roots were highly significant. Likewise, for ovary wall tissues of mature regenerated plants, all partitioned variances (i.e. between plant, within plants between tillers, and within tillers between ovaries) were highly significant (Tables 5.2, 5.3, 5.4). 116 TABLE 5.2: Analysis of variance, chromosome counts from immature ovary wall tissue of plants regenerated from HVxHJ type A callus Source SS df MS F* Between plants 721.08 7 103.01 75.74*** Hithin plants, 283.18 8 35.40 26.03*** between tillers . - Hithin tillers, 188.30 32 5.88 4.32*** between ovaries Error 130.67 96 1.36 Total 1323.23 143 TABLE 5.3: Analysis of variance, chromosome counts from root tips regenerating from primary type A callus of HVxHJ Source SS df HS F* Between calli 105.64 3 35.21 35.21*** Hithin calli, 58.30 8 . 7.29 7.29*** between roots ’ Error 24.00 . 24 1.00 Total 187.94 35 TABLE 5.4: Analysis of variance, chromosome counts from root tips regenerating from secondary type A callus of HVxHJ Source SS df HS F* Between calli 116.43 4 29.11 7.89*** Hithin calli, 220.60 10 22.06 5.98*** between roots . Error 110.67 30 3.69 Total 447.70 44 ll7 In previous reports, it was established that karyotypic vari- ability could be attributed to polyploidy, aneuploidy, and chromo- somal rearrangements (Chapter 4). The loss or gain of single chromo- somes implied a quantitative change hithe ratio of the number of HV to HJ chromosomes. Taking this notion of parental imbalance further, it was hypothesized that continuous mixtures of parental genomes existed in cultured populations of karyotypically variable HVxHJ cells. Each genome possesses all of the necessary functions for life, making the existence of pure parental types feasible. A normal curve was postula- ted regarding the stoichiometry of parental mixtures, based roughly on 2). This a binomial probility function (i.e. {7HV + l4HJ chromosomes} hypothetical process is diagramatically illustrated in Fig. 5.4. The most direct test of the hypothesis would have been a comparison of predicted with actual distributions of parental chromosomes in the karyotypes of regenerated HVxHJ plants. Plant chromosome banding tech- niques are widely utilized and would provide a concise tool for the identification of the parental origin of specific chromosomes (Vosa l975). As stated in Chapter 4, all attempts to perfect such a tech- nique met with failure. Further, heterochromatic bands in Hordeum are generally centro- and telomeric, exhibiting very little linear differ- entiation (Linde-Larusen l975). It can be speculated that such pat- terns would have been impotent for precise parental identification. Also, chromosomal rearrangements would have had the effect of defacing established banding patterns, further nullifying the analysis. The isolation and propagation of genetic variability was demon- strated by comparing assays of isozyme concentration in crude ex- tracts of test calli (Chapter 4). Within a given genotype, isozyme patterns were observed to differ significantly between callus types. Figure 5.4: 118 Illustration of the hypothesis regarding the generation of a continuum of mixtures of parental chromosomes in tissue cultures of interspecific hybrids (see text for a more detailed explanation). 119 wkzwmdm 1.42.050 0... howamwm It; ZO_F3._._._.mZOO miozmo N szmI N 534‘ 2.1.8.2252. 4/ezwm5 FIGURE 5.4 120 This tended to make the identification of specific HV and HJ isozymes quite difficult. In whole plants, however, it was much easier to make comparisons between genotypes because these uncontrolled tissue differences were obviated. Comparisons of anodal EST and GOT zymo- grams from crude extracts of young culms yielded 5 HV-specific, 3 HJ-specific,l comigrating (e.g. present in both parents), and l hy- brid band (e.g. present in neither parent). The observation of three GOT bands in the hybrid supports the conclusion that this particular enzyme is a homodimer (Fig. 5.5). Forty-three plants regenerated from HVxHJ type A callus culture were scored for EST and GOT isozyme intensity, as described in Chap- ter 4, vs. those of the HVxHJ F1 hybrid (control). Replicate samples generally did not differ from each other (Fig. 5.6). Ex- amples of the scoring procedure are illustrated in Fig. 5.5. In general, the variation of band intensity was much less for GOT than for EST (Table 5.5). Further, it was observed that HV-specific bands tended to be expressed in higher intensities than those of HJ. The distributions of all band intensities were approximately normal (Table 5.6). For each plant, the genome mixture was estimated by calculating a ratio of (ZHJ band intensities:£HV band intensities). In control plants, this ratio was 0.6 (3/5). The distrubution of this ratio was bimodal (Fig. 5.7), but was manifested in a continuum from O to 1.0. Morphological variability in phenotypes such as growth habit, growth rate, size (Fig. 5.8), head morphology, and auricle expres- sion (Fig. 5.9) were also evident between these regenerated plants. Regarding head morphology, variability was manifested in awns 121 Figure 5.5: Nomenclature and examples of scoring for GOT and EST bands of crude extracts from basal culm tissue of HVxHJ regenerates. 122 .1 REGENERATES E’e‘ 3% P 8 m. 2.212 R. g 1% 2 § . | 5517 m .92 (1t) 6 nu 90\ "‘ ' \ 0 .g . 5 WW .87/i g . % E ‘1 HV ,814/ «ie- / 9 5 5 1 1 3 W .71 2 W .30...— l 0 l .5 . l HV ,26— 2 5 5 S . ..__ _ _ if“ __ —LORIGIN 601'3 HV .51 - (f \' 2 ‘ 2 o 2 1 2 HVxHJ .LI7‘ 0 2 0 2 0 2 l .S " 1 w .43/. ’ Figure 5.5 123 Figure 5.6: Typical examples of (A) EST and (B) GOT zymograms from crude extracts of culm bases of HVxHJ regenerates. Paired replicates of regenerates are juxtaposed. eeeeeeeeeee ..... eeeeeeeeeee eeeeeeeeeee i "' mm. WW! eeeeeeeeeee eeeeeeeeeee eeeeeeeeeee 125 TABLE 5.5: Comparative intensities of EST and .GOT isozymes among 43 regenerated HVxHJ plants. Parental Mean 5.0. Range Perggggsof Band origin intensity intensity intensity band absent GOT 1 HJ 0.97 - 0.35 0 - 2 6.98 2 HVxHJ(hybrid) 0.97 0.37 0 - 2 9.30 3 HV 1.08 0.37 0 - 2 2.32 EST 1 RV 1.30 0.48 1'- 2 0.00 2 RV 1.13 0.43 0 - 2 11.63 3 RV l.00 0.43 0 - 2 9.30 4 HV 0.90 0.55 0 - 2 18.60 5 HVxHJ (co) 0.94 0.50 0 - 2 4.65 6 HJ 0.94 0.51 0 - 2 11.63 7 HJ 0.80 0.58 0 - 2 20.93 TABLE 5.6: Distribution of isozyme band intensities for 43 regenerated HVxHJ plants. Nunber of plants of intensity Band 0 .5 1 2 001 1 3 4 33 3 2 4 1' 3s 3 3 1 0 33 9 EST 1 0 1 31 11 2 1 2 33 7 3 4 2 30 s 4 13 14‘ ' 20 7 s 2 12 23 6 5 s 9 25 4 7 9 13 17 4 126 Figure 5.7: Histogram of the distribution of ZHdzzHV isozyme band intensities, grouped as follows: 0 to .l, .l to .2,, etc. 127 “murEmzwhz. w2>NOm_ >1“: Amm_._._wzw._.z_ w2>Nom. 2.33 .._0._m. m. h. w. n. .v. m. N. TL FL 8.... at we wfit k 3}. VSSQES 00 O. N. E SiNV'ld :IO HBQWHN FIGURE 5.7 Figure 5.8: 128 A comparison of plant types among plants regenerated from HVxHJ type A callus. (A) plant type, mature HV, (8) regenerated HVxHJ, vegetative, multi-tillering perennial-type (left) vs. reproductive, low heading, erect, annual-type (right, (C) plant type, mature HVxHJ, (D) differing foliage size between two HVxHJ regenerates, (E) plant type, HJ, (F) differing in_ vivo growth rates of regenerated HVxHJ. All regener- ates were planted at the same time and maintained un- der identical conditions. 129 ‘ 1 $3: Figure 5.8 Figure 5.9: 130 A comparison of floral structures and auricle morphology between HV, HVxHJ, HJ, and plants regenerated from HVxHJ type A callus cultures. (A) head type, mature HV, (8) auricle type, HV, (C) head type, HVxHJ regenerate, (D) auricle, HVxHJ regenerate, (G) head type, original HVxHJ hybrid, (H) auricle, original HVxHJ hybrid, (J) head type, HVxHJ regenerate, (K) auricle, HVxHJ regenerate,(L1 head type, mature HJ, (M) auricle, HJ (all head types X.8. all auricles X4.8). 131 Figure 5.9 132 length and spines), number of rows per rachis (from 2 to 6), and av- erage palea length (Fig. 5.9). Auricle expression ranged from HV- like, through the entire spectrum, to HJ-like (Fig. 5.9). This evi- dence is further support for the existence of a continuum of parental genome mixtures. No quantitative measurements were made, however, due to the large environmental component and unknown extent of gen- etic control of these traits. Observations of microsporogenesis among certain regenerates from HVxHJ type A callus revealed meiotic configurations exhibiting a high degree of chromosome pairing (Fig. 5.10, Table 5.7). Independent studies of the original F1 hybrid demonstrated extremely low fre- quencies of bivalent f0rmation (.55 to .86 per cell, Rajathy and Morrison 1959, Murry 1975). In four plants, x? tests were performed to detect significant differences between observed and established frequencies of univalents, bivalents, and trivalents (Table 5.7). The data of Murry (1975) were used to calculate expected values, since they were taken from the same progenitor clone as that of the test plants. In two cases, these tests were difficult to interpret due to differences in mean chromosome numbers between the source and regenerated plants. Even extrapolation of expected numbers to con- form to the observed mean chromosome numbers did not change the cor- responding level of significance. In five independent cases of over 250 plants regenerated from HVxHJ type A callus, complete segregation for HV morphological char- acteristics was observed. Of these, only one has thus far flowered. The spike was extremely HV-like and bore florets in 4 rows, compared to the 2- and 6-rowed HV parental varieties. Auricles were Figure 5.10: Figure 5.11: 133 Comparison of metaphase I chromosome association be- tween the originan HVxHJ hybrid and HVxHJ regenerates; (A) original HVxHJ hybrid, 22 I + 1 II (X830), (8) HVxHJ regenerate, 2 I + 2 II + 1 IV + l V (X1178), (C) HVxHJ regenerate, 10 I + 2 II + 1 III (X1178), ED) HVxHJ regenerate, containing a large multivalent X1178 Plant type and spike morphology of a HV-like segregant regenerated from HVxHJ callus. (A) plant type of a HV-like segregant regenerated from HVxHJ type A callus, (8) head type of above, four-rowed rachis, (C) head type of HV parent, two-rowed rachis. 134 Figure 5.10 a 9,, ., 1% Figure 5.11 135 .u_en»= ozx>= paewmpgo we“ go Amnmav xgcaz xa um>gmmno mmmucmsamgm we» seem umum—aupmo mg¢n5== umpuoaxmo 333-.mm¢ oo.o oo.o oe.H ow.o o¢.oH o~.oH m e aeefim.mnu oo.o oo.o so.o oo.¢ NH.~H om.H~ o m «33cm.mo~ NH.o om.o No.0 mo.m mm.m om.H~ o N «eemm.- oo.o oo.o mH.o No.0 nn.m~ m¢.mH mg a m Nx > >H HHH Hg H conga: m—pmu “cop; Ppmu Lem .oz cam: meomososgu .02 new: mucmpa ozx>z umpmgmcmmms Lao» ocean meowpmwoommm meomosoegu eo mzomwgmaeou ”n.m u4m