A (amen: mo mom-macicm 5mm 0:: THE mmom-JETRAPL :9 mass 3: THE pomsmm, {gamma PULCHERRWA manger-i Thesis {or Hm Dizgz'es 0? M. 3. MiCE‘HCAN STATE HNWERSWY Daniel C. Miibocicer 29.66 wrists WWHHWWWIW LIBRARY 293 00874 6483 i . ch M1 §gzm .C ta” Unix: V ‘ Nov 0 9 1994 35921265 ABSTRACT A GENETIC AND MORPHOLOGICAL STUDY OF THE DlPLOlD-TETRAPLOID CROSS IN THE POINSETTIA, EUPHORBIA PULCHERRIMA KLOTZSCH By Daniel C. Milbocker The diploid-tetraploid cross in poinsettia as reported by Ewart (1957) and Pai (I960) produced only diploids and tetraploids which were associated with a high frequency of ovary abscission. This re- search was conducted in an effort to determine by morphological tech- niques the factors involved in ovary abscission of this cross. Also, genetic analysis by the use of bract color was used to determine the reproductive behavior of the diploid-tetraploid cross which resulted in tetraploid seedlings. As a result of this research, the triploid poinsettia with 42 chromosomes is reported for the first time. A diploid female, Ecke White, crossed with the tetraploid male, Barbara Ecke Supreme, were the parents. The occurrence of a triploid endosperm among the regularly tetraploid endosperms is hypothesized to produce a small frequency of viable seeds from which triploids were found. A single tetraploid progeny from the same diploid-tetraploid cross was testcrossed to a homozygous white tetraploid cultivar. This seedling tetrapkfid had obtained one-half of its four genomes from its diploid maternal parent. Duplication of an embryo sac nucleus without Daniel C. Milbocker - 2 nuclear division is hypothesized as producing a diploid egg and a single polar nucleus. Gametic union produced a tetraploid embryo and a tri- ploid endosperm which developed into a viable seed. The failure to form a normal endosperm was found to be the cause of ovary abscission in the poinsettia. A GENETIC AND MORPHOLOGICAL STUDY OF THE DlPLOID-TETRAPLOID CROSS IN THE POINSETTIA, EUPHORBIA PULCHERRINA KLOTZSCH BY Daniel C. Milbocker A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Horticulture I966 ACKNOWLEDGEMENTS The author extends his appreciation to Dr. Kenneth C. Sink, as my advisor and for the preliminary work conducted for this re- search; to the department of Horticulture, Michigan State University for the financial support received to conduct this research; to my wife, Rose, for her assistance in processing research data and pre- paring the manuscript; and to Paul Ecke, lnc., Mikkelsen and Sons Greenhouses, and the United States Department of Agriculture for the poinsettia stock plants used in this research. TABLE OF CONTENTS ACKNOWLEDGEMENTS. . . ..... . . . . . . . . . . . . . LIST OF TABLES ..... . .................. LIST OF FIGURES . . ........ . ........... , I. INTRODUCTION . . . . . . ...... . ....... II. LITERATURE REVIEN. .......... . . . . . . . IV. V. The Diploid-Tetraploid Cross in the Poinsettia . MATERIALS AND METHODS. . . . .'. . . ..... Propagation and Cultural Practices ...... , Pollination Technique ..... . . . . . . Chromosome Counting. . . . . . . . . . . . . Ovule Morphology Technique ........ Pollen Viability Test. . . .. .......... Genetic Study of the Tetraploid Seedlings. Triploid Progeny ... . . Repetition of the Diploid-Tetraploid Cross Ovule MOrphology . . . . . . . . . . . . . RESULTS. . . . . ............ . ..... Chromosome Counts of Embryos . . ........ Cross- Versus Seif-Pollination . . ..... .. . Increased Level of Homozygosity ...... Time of Pollination. . . . . . ......... Pollen Viability . Genetic Study of the Tetraploid Seedling . Triploid Progeny . . . . . . . . . . . . . Ovule Morphology . . . . . . . . . . . . . . . Chromosome Counts of Embryos . . . . . . . . . Cross- Ver5us Self-Pollination ......... increased Level of Homozygosity. . . . . . Time Of Po‘lination. O O O O O O O O O O O Endosperm Control of Progeny in the Diploid- Tetraploid Cross. . . . . . . . . ... . iii DISCUSSION ...... . . . . . . . . . . . . ... . Table of Contents (Cont.) Page VI. SUMMARY AND CONCLUSION. ....... . . . . . . . 39- VII. LITERATURE CITED. . . . . . . . ......... . 'UI iv LIST OF TABLES Table Page i. The characteristics and source of poinsettia CUltivarSO O 0 O O O O O 0 O O O O O O O O O O 0 O 8 2. The characteristics and parentage of cultivars developed at Michigan State University . . . . . . 9 3. The distribution of red and white plants from the testcross. O O O O O O O O O O O O O O O O O O V. O I“ A. Normal ovuie development in percent of mature size. . 2i 5. Dipioid-tetraploid cross ovuie devel0pment in percent of mature size . . . . . . . . . . . . . . 22 ~6. The cuitivar and percent ovary abscission in self- and cross-pollinated cultivars . . . . . . . . . . 25 7. The number and direction of progeny crosses . . . . . 27 8. The effect of self-pollination time on ovary abscission. . . . . . . . . . . . . . . . . . . . . 28 9. Pollen viability for selected cultivars . . . . . . . 29 lo. Tetraploid segregation ratios from Haldane. . . . . . 3i LIST OF FIGURES Figure Page i. Somatic chromosomes of the triploid poinsettia. . . l7 2. Endosperm development in the diploid and the diploid-tetraploid crosses . . . ... . . . . . . i7 3. Embryo endosperm and ovule growth of the diploid and the diploid-tetraploid crosses . . . . . . . 20 vi I. INTRODUCTION in the vascular plants, especially the angiosperms, there is a widespread prevalence of poiyploids among predominantly diploid populations (Swanson, I957)1~*SOCCGSSFUT cTOsses between diploids and tetraploids may unite the haploid and diploid gamete to produce tri- ploid progeny. The occurrence of diploid and tetraploid offspring in crosses between diploid and tetraploids is not in accord with expectations from the union of a haploid and a diploid gamete. These chrombsome types have occurred in a wide variety of plants. The diploid and tetraploid offSpring of these crosses generally occur in low frequencies, being associated with a high frequency of ovary ab- scission, triploids, non-viable seed production or a combination of these (Bremer, i962; Cooper and Brink, lShS; Hangenheim, Peloquin and Hougas, i960). This cross in the poinsettia, Egphorbia puicherrim;_Kiotzsch was reported by Ewart (l957) and Pai (i960) to produce only diploids and tetraploids which were associated with a high frequency of ovary abscis- sion. ' During the poinsettia flowering season of I963, a study was instituted to determine the source of the extra set of chromosomes in the tetraploid progeny from the diploid-tetraploid cross in the poinsettia. Bract color was used as a genetic marker for chromosomes in this study. Barbara Ecke Supreme, a tetrapolid, is homozygoUs dominant for red bract color; Ecke White, a diploid, is homozygous receSsive for white bract color as reported by Stewart (i960). He found that bract color in the poinsettia was inherited as a single gene controlled difference with white being recessive. A tetraploid progeny resulting from a cross with Ecke White as the female and Barbara Ecke Supreme as the male parent could be testcrossed with a tetraploid homozygous recessive white cultivar to reveal the number of genomes received from each parent. Six progeny were obtained from the cross made in I963 between Ecke White and Barbara Ecke Supreme. Five of the seedlings were normal and one had malformed and curled leaves. These six progeny and other cultivars were used in the present research which was started in l96h. The primary purpose of this research was to determine morphologi- cally the factors involved in ovary abscission, and to investigate genetically the reproductive system in the poinsettia. This purpose was accomplished by: l. Counting the chromosome numbers in the progeny of the diploid- tetraploid cross. 2. Determining the ratio of white and red plants in a testcross of a tetraploid from the diploid-tetraploid cross. 3. Examining histologically abscissed ovules from the diploid parent. h. Chromosome countings of embryos from the diploid-tetraploid cross. I 5. Studying ovary abscission on inbred cultivars. The poinsettia is a desirable species to use in this study be- cause: i. The ovules are relatively large, thus, making cytological study less difficult. 2. The ovaries absciss providing a marker for determination of the time when seed formation ceases. 3. Bract color provides a distinct genetic marker. h. A generation from seedling to seed production occurs within a year. 5. Cytological and genetic studies of the diploid-tetraploid cross have been previously reported. II. LITERATURE REVIEW The Diploid-Tetrgploid Cross in the Poinsettia. The poinsettia has a basic chromosome number of lh with a somatic number of 28 (Moyer, i93h; Perry,-l9h3; Darlington, I955). Darlington (l955) accepted Moyer's (l93h) count; Perry (l9h3) added lh, and S6 for the poinsettia. A somatic chromosome number of ih has not been reported nor found since among commercial cultivars. Somatic chromosome numbers of 56 were first reported for several commercial cultivars by Ewart (l957). The cultivars originated as somatic mutations of the diploid, Oak Leaf, or as somatic mutations of Oak Leaf mutants; the oldest recorded being the cultivar, Mrs. Paul Ecke, found as a somatic mutation in l929 (Stewart, l960). Crosses using a diploid (28 chromosomes) as the female parent and the tetraploid (56 chromosomes) as the male parent produced pro- geny with either 28 or 56 somatic chromosomes and the reciprocal cross produced no progeny as reported by Ewart and Walker (l960) and Sink (l963). A cross between Ecke White (diploid) and a seedling from Barbara Ecke Supreme, a tetraploid, produced a diploid cultivar, Paul Mikkelsen, which was introduced in l960 to the floral industry, and with its subsequent somatic mutations is now the most popular cultivar among poinsettia growers in the United States (Mlkkelsen, l966). Genetic, morphological and cytological studies have yielded in- formation on the sexual reproductive behavior of the poinsettia, but have not revealed sufficient information to determine the means by which diploid and tetraploid progeny were produced in the diploid-. tetraploid cross. Ewart (l957) proposed that anomalous meiosis of the megaspore mother cells was a possible explanation for the results he obtained. Ewart and Walker (l960) published two explanations, both of which were unsatisfactory in fully explaining their results. The first, apomixis, did not explain progeny segregation or the pre- sence of tetraploid progeny. The second, non-reduction or normal reduction followed by doubling of the chromosomes explained the tetraploid, the double reduction of the pollen mother cell explained the diploid progeny. They did not explain why triploids were not observed in the progenies. Bremer (l962) proposed endo-duplication as a basis for the results obtained by Ewart (l957) with pseudogamy accounting for the presence of diploids and segregation during meiosis I. Tetraploids are the naturally expected result of endo-duplication through formation of diploid egg cells. Sink (l963) proposed a similar endo-duplication mechanism operative either at the meiotic level or egg nucleus level. His embryo sac study ruled out adventitious embryony, since pollination was required for embryo development, and no evidence of apomixis was found. Sink (l963) also revealed the presence of de- generating material tissue in the ovule which led him.to believe that there was an embryo-endosperm-maternal tissue incompatibility factor, involving the chromosome number, causing the degeneration and subsequent failure of seed development in this cross. I He proposed that reduction division during sporogenesis in both parents would produce a 5n endosperm with either a 3n or 5n embryo and endo-duplicatlon would produce a 9n endosperm with either a fin or a 5n embryo. The feasibility of this proposal was supported by the possible occurrence of the chromosome number in the embryo being the same as the progeny that has been observed. III. MATERIALS AND METHODS f!22232£19fl_ggd Cultural Practices The cultivars used in this research are listed in Tables l and 2. To propagate these plants, terminal cuttings three to four inches in length were taken from vegetative stock plants. intermittent mist and a sand propagation bench were used conSistently. Soluble fertilizer (l2-3l-lh) was applied to the cuttings at the rate of one- half ounce per gallon of water every four days starting four days after the cuttings were placed in the propagation bench. This proce- dure was used whenever accelerated growth was desired on the newly rooted cuttings, or when slow rooting cultivars were being propagated. Rooted cuttings were potted‘in steam sterilized three or four inch clay pots using a potting medium of three parts clay loam soil, one part send, one parf.sedge peat and one part German peat. The plants were potted Without packing the soil, thus permitting adequate water retention and good drainage. The plants were covered with two layers of cheese cloth and drenched twice daily for four days. The cloth was removed and the pots spaced to one per square foot. Plants grown for flowering were later tranSplanted to five inch clay pots using the same soil mix.' These plants were fertilized twice a week using a soluble fertilizer (l2-3l-lh). The stock solutibn con- tained one pound of fertilizer per three gallons of water and was applied with a Hozon.‘ Plants with fully developed bracts were fertilized once lHozon, Plant Products, New Point, Long island, New York. 7 Table l. The characteristics and source of poinsettia cultivars. Abbrevi- Chromo- Bract Cultivars ation some No. Color Source White Ecke EW 28 white: Ecke, inc. Oak Leaf OL 28 red Ecke, Inc. St. Louis SL 28 red Ecke, Inc. Ruth Ecke RE 28 red Ecke, inc. Barbara Ecke Supreme BES 56 red Ecke, inc. Indianapolis Red lR 56 red Ecke, inc. Elizabeth Ecke EB 56 red Ecke, inc. New Ecke White iEW (Chimera) 56 white Ecke, inc. Paul Mikkelsen PM 28 red Mikkelsen Mikkelpink MP 28 pink Mikkelsen Mikkeldawn MD 28 pink-white Mikkelsen Spotlight P-128 28 red U.S.D.A. 60-530-2 P-I30 28 red U.S.D.A. Snow Cap P-l3l 28 white U.S.D.A. 6l-7I-I P-l32 56 white U.S.D.A. 6I-302-I P-I3h 28 red U.S.D.A. White Cloud P-l35 28 white U.S.D.A. Table 2. The characteristics and parentage of cultivars developed at Michigan State University. Cultivar Parents Ploidy Bract Color 6h—2 EW X BES tetraploid red 6h-h EW X BES triploid red 6h-5 EW X BES triploid red 6h-7 EW X BES triploid red 6h-8 EW X BES triploid red 6h-l3 EW X BES triploid red 65-IO P-l28 a diploid red ,65-ll P-l28 O diploid red 65-l3 P-l28 fl diploid red 65-lh P-l28 8 diploid red 65-l5 P-l28 fl diploid red 65-23 EW 2 diploid white 65-2h EW a diploid white 65-25 EW 9 diploid white l I a week. A Solltex Testing Kit was used as a convenient way to check soil pH. No measures were taken to lower soil pH, unless it exceeded ISoil Science Department, Michigan State University, East Lansing, Michigan. IO seven; then weak sulfuric acid was applied with a Hozon every four days until the pH was lower than seven. The plants were watered daily as needed and on flowering plants more care was taken since slight wilting resulted in termination of the new cyme and premature abscission of lower leaves. The greenhouse was thermostatically set at a minimum 2l°C. Sudden temperature change is a mutagenic factor for a variety of plants in- cluding wheat and rye (Dorsey, l936), corn (Randolph, I932), barley and wheat (Peto, l938a and b) and onions (Huskins and Cheng, l950). Thus, the greenhouse was vented conservatively to prevent sudden tem- perature changes from affecting the seed set. Pollination Technique Ovaries were hand pollinated by transferring the stamen with tweezers to the stigma of the cyathium to be pollinated. Pollination was done only once on each stigma. Ninety-five percent ethyl alcohol was used to clean the tweezers between crosses. Each ovary was tagged with the identification of the parents, and the date of pollination for later use in laboratory work, or for seed collection data. Non-female cyathia and terminated cymes with no ovaries were removed to facilitate pollination and prevent accidental selfing. Chromosome Counting The counting of chromosomes was accomplished through a modifica- tion of Ewart's (l957) root tip smear technique. Dividing shoot meri- stems were excised and stripped of sheath leaves. The end section of ll primordia not exceeding one eighth inch in length was immediately placed in modified Carnoy's solution (Bladwin, l938) composed of one part glacial acetic acid; two parts 95 Percent ethyl alcohol and three parts chloroform. Shoot tips were more convenient to collect than root tips, and shoot cells tended to flatten into a polar view with a greater frequency than did root cells. Latex and chloroplasts were not a problem in rapidly growing material. The tip remained in the modified Carnoy's solution at least is minutes, but could be stored several days at room temperature before removal and emmersion into a mixture of 95 Percent ethyl alcohol. The tips were removed from this solution after ten minutes, and placed in distilled water for lO to IE minutes. Each tip was cut into approxi- mately four pieces, with one piece being placed on a clean slide and covered with a drop of aceto-orcein stain (La Cour, l9hl). The tissue was spread using mashing type motions with the flat side of a knife blade, exercising care not to roll the tissue. Another drop of aceto-orcein and a cover slip were added. After five minutes, the slide was placed in the fold of a paper towel, both placed on a flat surface and pressed with the thumb. The edges of the cover slip were sealed with vaseline or colorless fingernail polish to prevent evapora- tion. The chromosomes were clearest when examined immediately, but slide preparations kept well for three days in refrigeration at 2°C. A modification of this method was used to count the chromosomes of em- bryos. ‘The embryo sac was removed from the ovule and fixed using the same procedure for shoot tips. The embryo sac was spread only by pressure on the cover slip after staining. l2 Ovule Morphology Technique Morphological investigations of the ovule were made according to the procedure of Sink (l963). The ovules were removed from the ovaries and cut transversely at the middle. The placental and was fixed in Navashin's fluid (Conn, I960) under vacuum of 20-25 millimeters of mer- cury. After #8 hours the ovules were washed in running tap water for 2h hours, and dehydrated with tertiary butyl and ethyl alcohol (Johansen, 191m). The dehydrated ovules were imbedded In Tissuenat. ‘ Sectioning was done on a retary microtome at l0 microns. The sections were placed and spread in a drop of Haupt's adhesive and three drops of four percent formalin and allowed to dry on a warming tray at h3°C. The paraffin was removed with xylene, a drop of Permount applied and a cover slip added. The sections were observed under a phase microscope at X500 and photo- micrographs were taken at X50 and Xl00. Older material was observed with a light microscope at X25 and X50. Pollen Viability Test Pollen grains were collected from freshly dehisced anthers and placed on a slide. A drop of cotton blue stain was added which consisted of equal parts of phenol crystals, lactic acid, glycerine and distilled water after a modification of Johansen (l9h0). A cover slip was placed over this according to the method of Sink (l963). After five minutes, IA 56.506. melting point paraffin produced by Fisher Scientific Company, Pittsburgh, Pennsylvania. l3 all pollen grains were counted in a representative scan of the slide with a microscope at Xh30. Plump, blue stained pollen grains were counted as viable and shrunken, or light stained grains were counted as non-viable. Three counts were made with a minimum of lOO grains evaluated for each count. IV. RESULTS genetic Study of the Tetraploid Seedling Cytological examination of the six offspring from the diploid- tetraploid cross showed that 6h-2, a seedling with deformed foliage, was the only tetraploid. it was testcrossed with the tetraploid cul- tlvar P-l32 from the U.S.D.A. which is recessive for white bract color. The cross produced a ratio of 3.5 red to l.0 white plants (Table 3). Color was determined by petiole evaluation. White poinsettias in- variably have green petioles and red or pink poinsettias have red pigmented petioles. Table 3. The distribution of red and white plants from the testcross. P-I32 X 6h-2 GN-Z X P-I32 Total No. of pollinations l6h l05 269 No. of seed 88 ‘ 7h l62 No. of plants 75 55 l30 No. of white plants l8 ll 29 No. of red plants 57 Ah lOl Triploid Progeny The remaining progeny from the diploid-tetraploid cross were identified as 6h-h, 6h-5, 6h~7, 64-8 and 6h-l3. Chromosome counts lh 15 revealed that all five were triploid 3N = #2 (Figure i). All five progeny were red for bract color and segregated three bright and two dark red. The dark red progeny also had the broader oval type bract of the Ecke White parent, while the other three had either an intermediate or the longer nar- row bract of Barbara Ecke Supreme. All progeny had sinuate leaves of the Barbara Ecke Supreme parent. Although no data were recorded, variation in plant characteristics occurred between the triploids and commercial cultivars. Two hundred plants of commercial cultivars and #00 triploids were rooted and flowered to observe that 6h—5 was easiest to root and 6h-l3 was the earliest to initiate bracts. The triploid poinsettia as demonstrated by these off- spring and others is a vigorous grower. Repetition of the Diploid-Tetrgploid Cross A repetition of the diploid-tetraploid cross using the same parents produced three seedlings. All were triploids with sinuate leaves, and bright red bracts with segregation for bract type and initiation time. These progenies were 65-i, 65-2, and 65-3. The progeny 65-i was similar to Barbara Ecke Supreme in bract formation, and initiated bracts within 60 days from the date the cuttings were taken, as compared to 90 days for Barbara Ecke Supreme. Seedling 65-2 had a bright red color with the bract shape and initiation time similar to Ecke White. Ovule Morphology Ovules for morphological investigation were collected from ab- scissed ovaries of the Ecke White-Barbara Ecke Supreme cross at daily Figure I. Figure 2. l6 Somatic chromosomes of the triploid poinsettia, 3n = #2. Microphotograph of somatic triploid cell. X1250. lnterprative drawing taken from Figure l-A. Approximately XthO. EndOSperm development in the diploid and the diploid- traploid crosses. Diploid cross ovule 30 days after pollination with the endosperm consisting of closely packed spherical cells. Xl00. Diploid cross ovule Ah days after pollination with endo- Sperm consisting of cubical cells. XlOO.’ Diploid-tetraploid cross ovule 38 days after pollination with endoSperm consisting of closely packed spherical cells and endosperm degeneration. X50. Diploid-tetraploid cross ovule 50 days after pollination with integument shrinkage and remnants of the endosperm and embryo. X50. l7 ‘A O perm development in the diploid and the diploid-tetraploid crosses. l8 intervals from 30 days after pollination until abscission ceased. Ovules of the cross between Ecke White and Ruth Ecke had the least ovary abscission and were collected at two and four day intervals as the normal for comparative purposes. At least two ovules were examined for each interval of develop- ment in the normal series. The rate of normal embryo and endosperm growth is indicated in Table h and Figure 3. Five ovules were examined for each two day interval of the dipldid-tetraploid cross, and ten were examined for the longer intervals. Table 5 and Figure 3 indicate the rate of embryo and endosperm growth in the diploid-tetraploid cross. Normal ovules developed to full size in 26 days, and were followed closely by ovules of the diploid-tetraploid crosses. Normal endOSperm developed rapidly between 30 and #5 days, obtain- ing mature size in 50 days. Cellular endosperm development was first observed at l8 days and a change in cellular shape was observed at 26 days. The cells changed from a closely packed somewhat spherical form (Figure 2-A) to a cubical form (Figure 2-8). The cubical cells were in planes running transversely across the endoSperm tissue. Endosperm cells of the diploid-tetraploid cross were never observed to change from a closely packed spherical conformation, but remained as larger more fra- gile cells (Figure 2-C). The endosperm also ceased to increase in size. The normal endosperm breakdown that precedes the embryo became extensive in these ovules and generally the entire endosperm degenerated leaving a hollow in ovules that abscissed after #0 days. A few ovules in ovaries abscissing after 70 days contained a callus type growth of undetermined. origin. The maternal tissue began to shrink after #0 days toiincrease Figure 3. Embryo, endosperm and ovule growth of the diploid and the diploid-tetraploid crosses. DEVELOPMENT (PERCENT OF MATURE SIZE) 20 ICC-1 /. ,. A X I a ' I” ‘ 90« / a x . 80'i O OVULE x 70. A ENDOSPERM X EMBRYO so- . f-ZN x 4N CROSS —EW X RE CROSS 50" 4O 30‘. 20" IO-i AA O'L'_l=—I T r I0 I4 I8 22 263034 364246505458 TIME( DAYS FROM POLLINATION) 2l Table A. Normal ovule development in percent of mature size. Percent of Days from ovule endosperm embryo pollination diameter diameter volume 10 30 1h 60 10 .( l 18 85 12 {:1 22 93 18 .(1 2h 95 22 i