' Murray L. Layer D. P. Lavender Three Series of Extracts, Collected In

AN ABSTRACT OF THE THESIS OF

JOHN EDINBURGH WEBBER for the Doctor of Philosophy (Name of student) (Degree) Wood Science in (Wood Chemistry) presented on December 3, 1973 (Major) (Date) Title:THE OCCURRENCE OF ABSCISIC ACID IN THE DORMANT SHOOTS OF DOUGLAS-FIR IPseudotsuga menziesii (Mrb Francol

Signature redacted for privacy. Abstract approved: 'Murray L. Layer Signature redacted for privacy. D. P. Lavender

Methanolic extracts of the dormant Douglas-fir shoots (buds, leaves, and stems) were fractionated by solvent partitioning (sodium bicarbonate-diethyl ether).The acidic portion of the extract was subjected to column (Sephadex LH-2O, Silica Gel and Polyvinyl- pyrrolidone), preparative thick-layer and gas-liquid chromatography. Fractions collected from chromatographic separations were collected and tested for biological activity by the standard straight-growth Avena bioassay. Three series of extracts, collected in 1969 (preliminary series), 1970-1971 (first series), and 1972 (second series) were analyzed. Results of the preliminary series (buds only) showed only the presence of growth inhibiting compounds.No growth promoters were observed. Analyses of the first series extracts (buds only) showed that the zones of maximum growth inhibition in column and preparative thick- layer chromatographic separations corresponded to the elution volumes and Rf values of abscisic acid.Preparative thick-layer and gas- Tiquid chromatography were used successfully to isolate abscisic acid from the extract.Confirmation of its presence was determined on the purified methylated (diazomethane) extract by mass spectrometry, conversion of abscisic acid into its C2-trans-isomer, and the positive results for specific color reaction.This is the first reported occurrence of abscisic acid in Douglas-fir. Abscisic acid was quantified in the remaining extracts of the first series.The occurrence of the C2-trans-isomer was not observed in any of the extracts.In the second series, abscisic acid in the dormant buds, leaves and stems was quantified.The results from this series showed high levels of abs cisic acid in the buds (1. 2 g/g fresh weight), leaves (0. 347 g/g fresh weight), and stems (0. 094 g/g fresh weight) for the fall collection (October 1972).All tissues showed the lowest levels of abscisic acid in early spring (February, March and April). Although the concentration of abs cisic acid was shown to follow the dormancy cycle in Douglas-fir,it appears to play a secondary role (if any)in the induction ofdormancy in this species.The trend for abscisic acid concentration throughout the dormancy cycle appears to follow more closely the tree s capacity to biosynthe size the compound. One of the interfering compounds in the gas-liquid chromatographic analyses of abs cisic acid was determined by mass spectrometry and nuclear magnetic resonance to be dioctyl phthalate [di-(Z -ethylhexyl)phthalate].The natural occurrence of this phthalate ester in Douglas-fir is doubted and its presence was probably introduced as an artifact of the isolation procedure.The fact that it possessed inhibitory activity, however, stressed the importance for rigorous isolation procedures in abs cisic acid analysis.Two other regions of growth inhibition were observed but no further chemical characterization was attempted. The Occurrence of Abs cisic Acid in the Dormant Shoots of Douglas fir [Pseudotsuga menziesii (Mirb. ) Franco]

by John Edinburgh Webber

A THESIS submitted to Oregon State University

in partial fulfillment of the requirements for the degree of Doctor of Philosophy June 1974 ACKNOWLEDGEMENTS

The number of people who have contributed in various ways to this study are numerous and to all who have supported me I am sincerely grateful. A special thanks is given to my advisors Dr. Murray L. Layer and Dr. Joe B. Zaerr for their ever-present guidance and encouragement.In addition, I especially thank Dr. Murray L. Layer for his constructive criticisms during the preparation of the thesis manuscript, and Dr. Denis P. Lavender for his helpful suggestions. I would like to acknowledge with sincere gratitude, the initial support and confidence given to me by Dr. Harvey Aft.The amount of gratitude .1 have for my very dear friends Pat, Tim, Mike, Lynn, Greg, Darren, Sean, and Eric cannot be expressed. Throughout this study, my laboratory partner, Pat Loveland, hs provided companionship and support which I sincerely appreciate. Also, acknowledgement is made to the Forest Products Department which made financial support for this study available. TABLE OF CONTENTS

I. INTRODUCTION

II. LITERATURE BACKGROUND 4

A.Growth Regulators 4 1. Auxins 7 General Description 7 Coniferous Auxins 9 2. Gibberellins 9 General Description 9 Coniferous Gibberellins 13 3, Cytokinins 15 a.General Description 15 b,Coniferous Cytokinins 17 4. Inhibitors 17 General Description 17 Coniferous Inhibitors (Abscisic Acid) 20 5. Miscellaneous Growth Hormones 21 B.Abscisic Acid 21 1. Isolation of Abscisic Acid 21 2. Physiological Effects 23 Senescence and Abscission 23 Growth Retardation and Inhibition 24 Interaction with other Plant Hormones 25 Antitranspirant Activity 25 Miscellaneous Responses 26 3. Chemistry 27 4. Biosynthesis and Metabolism 33 C. Dormancy 37 1, External Factors Causing Bud Dormancy 39 2. Removal of Bud Dormancy: Resumption of Growth 41 3. Internal Control of Bud Dormancy 42 4. Hormonal Control of Bud Dormancy 44 D. Wood Formation 48 Cambial Activation 48 Xylogenesis 50

III. EXPERIMENTAL 52

Collection and Extraction of Plant Material 52 Fractionation by Solvent Partitioning 53 Bioassay 58 D. Determination of Active Compounds 59 E.Isolation of Abscisic Acid 60 1. Isolation Procedures for the 10-Z4-70 Extract 61 Sephadex Separation 61 Silica Gel Separation 61 Thick-Layer Preparative Chromatography 62 2. Methylation 63 3. Gas-Liquid Chromatography 66 4. Abscisic Acid Determination in the 11-4-71 Extract 68 5. Combined Gas-Liquid Chromatography and Rapid Scan Mass Spectrometry 70 6. Additional Confirmation of the Presence of Abscisic Acid 74 Conversion of cis-MeABA to trans-MeABA 74 Color Test for Abscisic Acid 74 F.Seasonal Determination of Abscisic Acid 75 1, Determination of Abscisic Acid in the First Series (1970-1971) 75 Isolation Procedures 75 Gas-Liquid Chromatographic Analysis 77 2. Determination of Abs cisic Acid in the Second Series (1972) 78 Formation of the Internal Standards 78 Isolation Techniques 83 G.Determination of Dioctyl Phthalate 87

IV. RESULTS AND DISCUSSION 90

A. Collection and Extraction of Plant Material 90 B,Fractionation by Solvent Partitioning 90 Bioassay 91 Isolation of Abscisic Acid 92 1. Gas-Liquid Chromatography 93 Preparative Thick-Layer Isolation of Abscisic Acid in the 11-4-71 Extract 95 3. Gas-Liquid Chromatography Analysis of Abscisic Acid in the 11-4-71 Extract 99 4. Combined Gas - Liquid Chromatography and Rapid Scan Mass Spectrornetry 104 5. Further Confirmation of Abscisic Acid 116 E.Seasonal Determination of Abscisic Acid 116 Formation of trans-ABA 119- Separation of Abscisic Acid in the Second Series (1972) 124 Determination of Dioctyl Phthalate 127 GLC Quantification of Abscisic Acid in the Second Series (1972) 133 F,Physiology of Abscisic Acid in Douglas-Fir 138

V. SUMMARY AND CONCLUSIONS 147

BIBLIOGRAPHY 149

APPENDIX 174 LIST OF FIGURES Figure Page Fractionation by solvent partition of the methanolic extract of the dormant buds of Douglas-fir; preliminary and first series (1970-1971) analysis. 56 Fractionation by solvent partition of the methanolic extract of the dormant shoots of Douglas-fir; second series ( 1972). 57 Isolation of abscisic acid from the lEA fraction (Fig- ure 1) of the 10-24-70 (first series) collection of buds. 64 Isolation of abscisic acid from the lEA fraction (Fig- ure 1) of the 11-4-71 (first series) bud collection. 71 Gas-liquid chromatography analysis of C2-cis- and trans.- hydroxy- 2 -cis-methyl abs cisate mixture. 82 Avena bioassay of the first inhibitor zone from the Sephadex separated 11-4-71 extract and separated further on preparative tic using the first solven system, benzene-methanol-acetic acid (97:2:1 v/v/v), 96 Avena bioassay of fractions 2 and 3, collected from the first preparative tic separation (Figure 6) of the Sephadex separated 11-4-71 extract.Second solvent system chloroform-methanol-acetic acid (97:2:1 v/v/v).97 Avena bioassay of fractions 2,3 and 4 collected from the second preparative tic separation (Figure 7) of the 11-4-71 Douglas -fir bud extract. 98 Gas chromatographic analysis of the lEA (Figure 1) fraction of the 11-4-71 (first series) extract after four preparative tic separations. 101 Gas chromatographic analysis of the lEA (Figure 1) fraction of the 11-471 (first series) extract after preparative separation of the methylated extract with n-hexane-ethyl acetate (1:1 v/v). 103 Figure Page Gas-liquid chromatographic separation of cis- and trans-methyl abscisate on the Finnigan series 3000 Gas Chromatographic Peak Identifier. 106 Reconstructed chromatogram of the separation of a cis- and trans-MeABA mixture. 108 Mass spectrum of authentic cis-MeABA. Spectrum number 99 from the Finnigan Series 3000 Gas Chro- matographic Peak Identifier (Figure 12). 109

Mass spectrum of authentic trans -MéABA. 110 Reconstructed glc chromatogram of the methylated 11-4-71 extract (after five preparative tic separations) on the Finnigan Series 3000 Gas Chromatographic Peak Identifier. 111 Limited mass search for m/e 190 in the chromatogram (Figure 15) obtained from the methylated 11-4-71 extract using the Finnigan 150 computer. 112 Mass spectrum (m/e values) for spectrum number 67 (Figure 15) obtained from the 11-4-71 methylated extract. 113 Conversion of cis-MeABA in the 11-4-71 extract to a mixture of cis- and trans -MeABA by UV radiation. 117 The 100-MHz N.M. R. spectra of cis-abscisic acid obtained from the preparative tic separation of a mixture of cis- and trans-ABA, 121 The 100-MHz N.M. R. spectra of trans-abscisic acid obtained from the preparative tic separation of a mixture of cis- and trans-ABA. 122 The gic analysis of the 11-4-71 leaf extract which was treated with lead acetate. 125 The glc analysis of the 11-4-71 leaf extract that was not treated with lead acetate. 126 Figure Page Avena bioassay results from the final preparative tic separation of the methylated 3-23-72 (second series, 1972) bud collection using n-hexane-ethyl acetate (1:1 v/v) 128 Mass spectrum of the overlapping compound obtained from the final preparative tic separation of the 3-23-72 bud extract. 130 The 100-MHz N. M. R. spectra of the overlapping compound obtained from the final preparative tic separation of the 3-23-72 bud extract. 131 The 100-MHz N. M. R. spectra of authentic dioctyl phthalate [di -( 2-ethylhexyl)phthalate]. 132 Variation in the concentration of abscisic acid in the buds of dormant Douglas-fir; second series analysis. 135 Variation in the concentration of abs cisic acid in the leaves of dormant Douglas-fir; second series analysis. 136 Variation in the concentration of abscisic acid in the stems of dormant Douglas-fir; second seriesanalysis. 137 Initial Silica Gel column chromatographic separation of the lEA (preliminary series) fraction obtained from the methanol extraction of Douglas-fir buds. 175 Avena bioassay of the 18 zones collected from the Silica Gel column chromatographic separation of the lEA (preliminary series) fraction. 176 Sephadex column chromatographic separation of the lEA (Figure 1) fraction of the 10-24-70 collection of Douglas- fir buds. 178 Avena bioassay of the zones (1-14) collected from the Sephadex column chromatographic separation of the lEA (Figure 1) fraction of the 10-24-70 collection of Douglas- fir buds. 179 Figure Page Sephadex column chromatographic separation of the ZEA (Figure 1) fraction of the 10-24-70 collection of Douglas- fir buds, 180 Avena bioassay of the zones (1-11) collected from the Sephadex column chromatographic separation of the ZEA (Figure 1) fraction of the 10-24-70 collection of Douglas- fir buds. 181 Avena bioassay of the zones (1-7) collected from the Silica Gel column chromatographic separation of the combined biologically active zones of low elution volume obtained from the Sephadexseparation of the lEA and ZEA (Figure 1),10-24-70 extract, 183 UV spectra of zones (1-7) collected from the Silica Gel column separation (Figure 36), 184 Gas-liquid chromatography separation of the mthylated lEA (Figure 1) fraction obtained from the 10-24-70 Douglas-fir bud collection. 189

39, Gas-liquid chromatography separation of the methylated lEA (Figure 1) fraction obtained from the 10-24-70 Douglas-fir bud collection. 19 1 LIST OF TABLES

Table Page Gibberellin activity in various conifers and the part in which they were detected. 14 Collection data for the determination of abs cisic acid in the dormant shoots of Douglas-fir. 54 Characteristics of various liquid phases used for the gic analysis of a mixture of cis- and trans-methyl abscisate, 94 Gas-liquid chromatography analysis of the lEA fraction from the 11-4-71 Douglas-fir bud extract on four liquid phases. 105 Major peaks in the mass spectra of methyl abscisate obtained from the 11-4-71 (first series) Douglas-fir bud extract, 115

6, N..M.R. spectral data of cis- and trans-abscisic acid obtained from the irradiation of cis-methyl abscisate. 123 Gas-liquid chromatography analysis of abs cisic acid in the second series (1972) buds, leaves and stems from dormant Douglas-fir. 134 Quantification of methyl abscisate in the bud extracts of dormant Douglas-fir; first series (1970-1971) analysis. 192 THE OCCURRENCE OF ABSCISIC ACID IN THE DORMANT SHOOTS OF DOUGLAS-FIR [Pseudotsuga menziesii (Mirb) Franco]

I.INTRODUCTION

Our forest (wood) is valued not only for its economic importance but also for its aesthetic qualities.The need for wood is growing rapidly.Predictions foresee a three-fold increase in the world's demand for raw fiber by the year 2000 (100).This demand is now being tempered by the increasing awareness of the forest s role in the ecological balance of our environment. Current research efforts emphasize more efficient wood production through intensive forest management and closer utilization of all tree components.While fertilization and irrigation studies aim at increasing yields, better conversion methods reduce wood waste. The genetics, biochemistry and physiology of tree growth are active areas of basic research.Breeding programs with superior phenotypes have increased the yields of species with desirable properties.Understanding the biochemistry and physiology of growth enables certain manipulations which can establish young trees faster and extend growing periods.Establishing a healthy stand of trees with a high potential to produce wood is a goal nearly all can agree on. The physiological process of most concern in the present study is dormancy (256).By influencing this process of growth inactivity, 2 plant hormones affect not only the amount of wood laid down but also the type and quality of wood (and phloem) formed (241, 251).In 1969 the growth of Douglas-fir in Oregon (private land) was 630 million cubic feet (243, p. 3-4).Since the effective growing season is approximately 21 weeks, this represents an average of 30 million cubic feet per week. By increasing the growing season by one week (or initiating it one week earlier) the added revenues to the industry could be 24 million dollars (assuming an approximate value of $800 per thousand cubic feet).This, coupled with increases due to intensive forest management represents a tremendous potential in annual timber production. Although this may not be a realistic calculation (since growth is sigmoidal and not linear) it does indicate the impact of such a possibility.Also, the difficulties in altering the growing period of a mature forest would be immense. However, it may be possible to alter the growth patterns of a young stand or more realistically, nursery stock. A problem specifically prevalent in the Pacific Northwest is the Fall planting of seedlings that are not fully dormant.In this condition, the young trees are more susceptible to damage during lifting, and to a higher mortality rate when planted.In addition, since they are not fully dormant, the nursery stock is more prone to damage by eaily frosts.The ability to force the seedlings into dormancy would 3 increase their chances of survival during lifting and permit the more rapid establishment of a stand.Dormant nursery stock would also survive early, severe cold conditions, and result in reduced losses the following year. The initial goal of this study was to investigate the endogenous growth hormones in the dormant shoots of Douglas-fir.Preliminary results indicated I was dealing primarily with growth inhibitors.The specific objectives then became a characterization of the chemical nature of these growth inhibitors and their possible influence on the dormancy cycle of Douglas-fir. 4

II.LITERATURE BACKGROUND

A.Growth Regulators

Understanding the nature of hormonal control in plants began in 1880 when Charles Darwin' described the effect of light on coleoptile curvature.About 40 years ago the substance showing activity in these early experiments was isolated from plants and shown to be indole-3-acetic acid, commonly called auxin.A second group of hormones promoting cell elongation was shown to exist in 1926 by Japanese physiologists.This group displayed different activity than auxin and was called gibberellins, named after the fungi (Gibberella) from which it was originally derived. Although the gibberellins were firmly established as plant hormones, Thimann (232) still stressed as late as 1954 that the growth of plants was controlled by auxin alone.However, a second major concept was developing which stressed that several different hormones interacted to produce what is thought to be normal grw th. In 1957, a third group of hormones, the cytokinins, became evident but a cytokinin native to higher plants was not isolated until 1964. In addition to the auxins, gibberellins, and cytokinins, two other major groups of plant hormones exist: ethylene and growth inhibitors

tFora historical review of plant hormones including references for original discoveries see references 130 (p. 79-158)and 234, 5 including abscisic acid.Both have long been known to affect plant growth but the role ethylene played as a natural hormone was not really made clear until 1964 and the structure of abs cisic acid was not elucidated until 1965.Other yet uncharacterized compounds may exist as separate groups (brass ins) or be related to the gibberellins (florigen) or cytokinins (wound hormone). The words hormone and regulator have been used interchange- ably to describe active compounds in plants but a strict definition does distinguish them. A plant hormone is defined as a natural organic substance, other than nutrients, active in minute quantities (< M) which is formed in parts of the plant and is translocated to other parts of the plant where it regulates a specific set of processes (11, p. 19-2 3).A plant growth regulator is less strictly defined.It per- tains to any organic compound, other than nutrients, which in small quantities(

(79).Compared to herbaceous plants, woody plants are more difficult to collect, and the extract more difficult to separate.They do, however, offer a study of hormonal control of growth throughout the grow - ing season and over a number of growing cycles. Previous work placed emphasis on individual hormones with the majority of effort being centered on auxin determinations. Now it is recognized that the balance among various endogenous growth hormones control specific physiological functions (77).Thus differ- entia.tion of bud tissues, expansion of leaves and internodes, leaf senescence, apical dominance, leaf abscission, and the onset and release of bud dormancy all appear to be controlled by more than 7 just one hormone.

1. Auxins

a.General Description.Although generally referred to as indole-3-acetic acid (I),a more encompassing definition describes auxins as the generic term for organic compounds which at low concentration (< M) proriote growth along the longitudinal axis of plant shoots freed of growth-promoting substances and inhibit the elongation of roots (233).Thusindole-3-acetaldehyde (II), indole- 3-pyruvic acid (III), indole-3-ethanol (IV), and indole-3-acetonitrile (V) all occur naturally and are grouped as auxins.

CH-CHO 2

CH -C-COOH

III IV CH -CN 2

V 8 Indole-3-acetic acid (IAA) was firstisolated2and identified from urine (105) and two fungi, Saccharomyces cerevisiae(104)and

Rhizopus su:inus(229).fAA now has the status of the most universal natural au.xin of higher plants.This is based on the large number of determinations (18,200)using chromatographic and colorimetric determinations. IAA is also widely distributed within plants, especially in actively growing tissue, where it appears to be synthesized from tryptophan by a series of enzyme-controlled reactions(264). The amounts of endogenous IAA are extremely small and are controlled at the physiological levels required for normal growth by the capacity of the plant to biosynthesize the compound, to destroy it by the action of IAA oxidase(224),and to conjugate the molecule with such compounds as aspartic acid (9), thereby rendering auxin inactive.IAA can also be destroyed in tissue exposed to light by a photo-oxidation mechanism (75, 76). Auxin-induced effects in plants include increase in stem length due to promotion of cell elongation, inhibition of root growth, induction of adventituous roots, inhibition of leaf and fruit fall, parthenocarpic development of fruit, cambial activation (cell division), tropisms, flowering, and apical dominance.

2Fora historical review of auxinology see references218 and 234. 9

b.Coniferous Auxins.In conifers, IAA has been searched for perhaps more than any other hormone and evidence suggests its presence in nearly all species investigated (5, 41, 51,65, 18, 216,

268). Some studies have been misleading since nonauxin growth promoters (gibberellins) are active in the Avena bioassay (170) and non- indolic compounds react positively with the typically used lASS chro- magenic spray, Ehrlich'sreagent(19, p. 316-317; 68).Also, the natural occurrence of IAA analogs (indoleacetonitrile and indole- acetaldehyde) could cast doubt on early IAA determinations (159). In the past decade, gas-liquid chromatography has become more available and the identification of IAA more credible (55,56, 84,

222). In Douglas-fir, the amountf diffusible auxin in different parts of the crown was measured by Zimmerman (270) using Wents curvature test.These results were confirmed later by Frohlich (74).The occurrence of indole-3-acetic acid in the shoots of Douglas-fir has more recently been proven using combined gaschromatography and mass spectrometry (57).

2.Gibberellins

a.General Description.In 1926, Kurosawa discovered that the cell-free sterile filtrates of the fungus Gibberella fujikuroi (also 10 called Fusarium heterosporum), which causes pale spindly growth of rice, gave a marked growth stimulation when applied to seedlings of rice and grasses. Among the first noted physiological respon5es of the gibberellins was the converting of dwarf bean plants into pole beans or causing other plants to flower.Several reviews deal with other physiological effects as well as the history of investigation leading to the recognition of gibberellins (219, 220, 221). In 1955, Brianetal. (24) isolated gibberellic acid from a Fusarium culture and six years later its structure was elucidated by Cross etal. (52).Well over 30 gibberellins have now been characterized from plants and all have had a structure either tentatively assigned or rigorously established (25,178). Gibberellins are closely related, modified diterpenoids, differing in substitution pattern and degree of unsaturation but, as far as known, not in absolute configuration.The systematic nomenclature of the gibberellins is based on the gibbane ring system (VI) initially proposed by those responsible for much of the knowledge of the structural and chemical properties of the gibberellins (25).The numbering system indicated is not the original, but the one now accepted because of its relationship to the numbering of kaurene (VII), a biosynthetic precursor. Because of the complexities involved with systematic names, a trivial designation for the gibberellins was developed.Takahashi 11 etal. (225) introduced the terms gibberellins A1, A, A3, and A4 for four gibberellins isolated from fungus cultures,Gibberellin A3 was found to be identical with the substances previously isolated and named gibberellic acid (25).Thus, gibberellic acid (VIII) is denoted GA3 and is shown with the stereochemistry common to all gibberellins. 12

17 CH2

VII

:1 H3 VIII The generic term gibberellin is now reserved for a limited group of related substances of natural origin.Further, the term "gibbeel1in iormone! is reserved for those gibberellins produced by plants which in low concentration regulate plant physiological processes (1 77).The term igibberellin_likeT is used for substances not completely characterized or not containing the gibbane ring system bt possessing activity in the gibberellin bioassay (184). Just as IAA can be linked to tryptophan metabolism, the gibberellins have been linked to diterpene metabolism. Via the isoprenoid pathway and through geranyl-geranyl pyrophosphate, (-)-kaurene (VII) 12 is formed.The B ring of kaurene undergoes ring contraction to form the five-membered system of the gibbane structure and the free carboxyl group at the C6 position. Among the physiological effects of the gibberellins on plant growth, activation of the subapical meristem of dwarf plants causing bolting and enhancement of cell enlargement in other plants are the most dramatic.At present there is not enough evidence to determine whether the primary effect of gibberellins is on cell division or cell enlargement.The absence of marked abnormalities concomitant with growth induction is a feature of gibberellin treatment that indicate the prime effect may be to accentuate existing metabolism rather than to cause a gross reorientation of cellular activity,In any case, it is clear that the gibberellins are the principal agents regulating activity of the subapical me ristem (42). Associated with apical meristem activity is dormancy.It is becoming more apparent that the gibberellins affect both bud and seed dormancy.In some cases dormancy can be reversed by addition of gibberellins and in other situations breaking of natural dormancy is associated with an increase in the endogenous levelf gibberellins

(256).Although gibberellins do influence activity of this region they probably do not solely control it.Associated with the decrease in gibberellins at the onset of dormancy there is usually an accumulation in the bud of one or more growth inhibitors.The relationship between 13 abscisic 3cid (inhibitor) and gibbereilins in dormancy is receiving more attention and will be dealt with specifically in a later section. Gibberellins affect other plant responses as well.Gibberellins can induce some plants to flower but not in all cases.They play a vital role in seed germination by promoting the synthesis of hydrolytic and proteolytic enzymes upon which germination and seedling establishment depends,Gibberellins are required for proper fruit development and gibberellins treatment can cause parthenocarpic development in some fruits.For a more detailed review of the physiological responses to gibberellin treatment see Cleland (42) and Paleg (177). b.Coniferous Gibberellins.Characterization of gibberellins in conifers has received considerably less attention, perhaps because of the large number of naturally occurring gibberellins with each displaying its own physiological response.Gibberellins have been implicated in the growth and differentiation process of several conifer species (87,179, 180,194) largely in an effort to induce flowering, Table 1 lists the species in which gibberellins were assayed for and the plant part investigated. With regards to seasonal changes of gibberellin-like (GA) compounds in pine and larch, Michniewicz (147) reports that the period of intensive growth is associated with high GA levels and diminished inhibitor levels.Increased tree age and rapid growth rates were associated with high GA levels.In a study with 14

Table 1.Gibberellin activity in various conifers and the part in which they were detected.

Species Plant part Ref.

Larch

Larix leptolepsis vegetative shoots 88 Larix deci4ua vegetative shoots 145

Arizona Cypress Cupressus arizonica vegetative shoots 193

Cryptomeria japonica vegetative shoots 89

Douglas-fir Pseudotsuga *nenziesii vegetative shoots 53 var, xnenziesii var, glauca

Jiniper Juniperus chinensis berries 98

Pine

Pinus jeffreyi 98 Piuus ponderosa embryos 115 Pinus lambertiana pollen 115 Pinus silvetris vegetative shoots 144 15 Douglas-fir, Crozier, Acki, Pharis and Durley (53) found gibberellic acid (GA3) to be the major gibberellin in the vegetative shoots of coastal (var. menziesii) and inland (var. glauca) varieties of Douglas- fir.Bioassays of portions from the vegetative shoots of coastal and inland varieties contained at least three GA-like compounds in addition to GA3.The faster growing coastal variety contained ten times more GA3/kg of tissue than the more slowly growing inland variety. Shoots of 80 year old trees of the inland variety contained five times more GA3 than tissue from45 year old trees.These results were similar to those obtained by Michniewicz (147) for Pinus sylvestris.

3.Cytokinins

a.General Description.Both auxins and gibberellins promote cell enlargement.The other fundamental process of plant growth is the formatioi of new cells.Naturally occurring compounds influencing cell division are the cytokinins.The term cytokinin is a generic term for substances that promote cell division in plant tissues and generally affect growth and morphogenesis in the same way as does kinetin (205). Kinetin, 6 -furfurylaminopur me (6 -furfuryladenine) (IX) was the first cytokinin characterized but it was not originally isolated from a natural plant source.In search for factors which influence the development of buds in tobacco callus,it was shown that adenine, 16 together with increased phosphate levels in the nutrient medium, not only counteracted the bud-inhibitory effect of IAA, but also promoted the formation of buds and increased the growth of the callus tissue (203, 204).Natural sources of extracts were tested for activity in cell division induction and it was found that concentrates exhibiting the properties of a purine were highly active.Thus, nucleic acid preparations were tested and DNA was found to be a very rich source of activity.The chemical structure was determined to be 6-furfurylaminopurine which was later confirmed by synthesis (155, 156, 157). Because of its activity in promoting cytokinesis, the substance was named kinetin. Although cytokinins have been detected in extracts of immature seeds and many plant tissues,it was not until 1964 that the isolation and identUication of a naturally occurring cytokinin (zeatin) was reported.Obtained from sweet corn by Letham (133,135), zeatin was identified (134) and synthesized (201).Like most other cytokinins it is a 6-substituted adenine derivative {6-(4-hydroxy-3-methyl-trans- 2-butenylamino)purine] (X),

H CHOH 2

HN CH 3 17 Cytokinins, besides stimulating cellular division, inhibit senescence.Other physiological effects of cytokinins include cell enlargement, morphogenesis (stimulation of bud formation), breaking of seed dormancy in some plants, and apical dominance (cytokinins counteractthe usual dominance of the apical bud) (205).

b.Coniferous Cytokinins.Of the known growth hormones in conifers, the cytokinins are perhaps the least well characterized. Even their occurrence in hardwoods has not been intensively investigated.The studies that are available make references to their possible occurrence or evidence for kinetin-like activity using specific bioassays (35, 190),Since cytokinins are known to be active in cellular division and tissue differentiation, the occurrence in coniferous tissue will surely receive more attention in the future.

4,Inhibitors

a.General Description.For some years it had been suspected that naturally inhibiting compounds might control plant growth but detailed evidence was not available.Growth inhibitors are relatively easy to detect by bloassays and gener3liy appear when screening extr3cts for auxins, gibberellins and cytokinins.They could represent the most characterized group of active plant constituents since many are phenolic derivatives of benzoic and cinnamic acids (14). Compounds siich as benzoic acid, salicylic acid (o-hydroxybenzoic 18 acid) and catechol (1, 2-dihydroxy benzene) all show inhibitory activity but are not usually considered growth hormones because they generally require large doses to be actie and they do not show any specific effects (130, p. l44-l57) Much attention has been given to the effects of phenolic compounds on the IAA oxidase system. A cofactor to the IAA oxidase enzymewas found to be ferulic acid (XI) (32).The presence of -hydroxybenzoic acid (XII) in solutions of IAA decreased the growth of segments (193) and increased the rate of decarboxylation of IAA at the same time (239).On the other hand, pyrogallol (XIII) and ortho- diphenols such as guaiacol (XIV) were found to inhibit the activity of

IAA oxidation system (81,187).These compounds were also found to synergize the action of IAA (that is promote elongation) in the Avena curvature test (12)Thus, monophenols as a group promote the oxidation of IAA while ortho-diphenols, triphenols and polyphenols are equally active in inhibiting it,

H CH COOH

CH3

XIII XIV xl XII In addition to the simple phenols and phenolic acids, other phenolic compounds generally inhibit growth or effect the IAA 19 oxidase system.These include the phenolic lactones (coumarins), For example, umbelliferone (XV) promotes the IAA oxidation system while scopoletin (XVI) inhibits it (8).The flavonoids including both quercetin (XVII) and kaempferol (XVIII) have also been implicated in affecting the IAA oxidase system (22).

HO xv Xv'

OH 0 Xv" Xv" Much of the earlier work on plant growth hormones was done using paper chromatography and assaying segments cut from the chromatogram.This analytical technique was applied to the acidic portion of plant extracts by Bennet-Clark, Tambiah and Kefford (16). Their results showed three consistent patterns (17):a zone of growth- promoting material (at Rf less than 0. 3 in a propanol:ammonia:water system) calledacceierator-&'; a zone at Rf 0.4 containing the growth promoter indole-3-acetic acid; and a zone of growth-inhibiting material (at Rf 0. 6) which was termed "inhibitor-p.'T Inhibitor zones corresponding in Rf to the "inhibitor-13't zone 20 described by Bennet-Clark and Kefford (17) have been found by numerous studies on paper chromatographed pine extracts (6, 119,

172,173,174, 202, 265),Milborrow (149) investigated the inhibitor- p zone of several hardwoods and other plants and found abscisic acid accounted for substantially all inhibitory activity.Other investigations describing natural inhibitors of growth and flowering, and promoters of senescence could quite possibly involve ABA or related substances (4 and references therein).

b.Coniferous Inhibitors (Abscisic Acid).As far as conifers are concerned a.bscisic acid has been positively identified by thin- layer chromatography (tic), gas-liquid chromatography (glc), mass spectrometry (ms), and optical rotatory dispersion (ORD) in red pine (Pinus radiata) (97), pacific yew (Taxus baccata) (131), balsam fir (Abies balsamea (140), and Douglas-fir (Pseudotsuga menziesii)(258). Abscisic acid (ABA) is known to Lnfluence the growth processes and in bioas says ABA can counteract cell elongation promoted by other growth substances such as indole acetic acid and gibberellins (4). Relating this physiological feature of abscisic acid to cambial growth, application of ABA to cuttings of Picea glauca resulted in a reduction of radial dimensions of treacheids and the rate of cell production (20,

27,138,139). Abscisic acid is known to affect bud-break in several species (64) and has been used in an attempt to delay bud break in balsam fir 21 and white spruce (27),When applied as a foliar spray, however, ABA was ineffective (63, 64) in delaying bud break and this was probably a result of its inability to penetrate into the bud meristematic region (27),

5.Miscellaneous Growth Hormones

The four major classes of growth hormones have been described but by no means represents the complete list.Ethylene has long been known to be a highly active growth substance affecting such processes as fruit ripening, leaf abscission and root growth.Many other compounds affect various aspects of plant growth but they have not been fully characterized or do not exist naturally.Future research will undoubtedly discover new and dramatic compounds that control the very complex process of plant growth.

B.Abscisic Acid

1. Isolation of Abscisic Acid

Prior to 1965, several laboratories were investigating the growth-inhibiting substances relating to such physiological processes as abscission, dormancy and growth inhibition (34, 200, 253). Attempts to characterize the compounds of active extracts proceeded slowly because of their low concentration, the presence of large 22 numbers of interfering compounds, and the lack of purification techniques,With the aid of organic chemists, three separate research efforts simultaneously showed abscisic acid (ABA) to be one of the prircipal (if not the principal) compound in the active extract. A group led by Cams and Addicott at the University of Cali- fornia, Davis campus, were investigating the active substances in cotton which promoted fruit abscission. With the aid of Ohkuma, this group first isolated ABA in 1963 (175) and published its structure in 1965 (176). Several investigators have been concerned with dormancy- inducing substances in deciduous trees (60, 68, 244, 250, 252). Eagles and Wareing (61, 62) have shown that an inhibitor obtained from the extract of birch leaves could completely arrest apical growth when applied to the leaves of seedlings.In addition, they showed that leaves from birch seedlings kept under short-days processed high levels of the inhibitor.In sycamore, Robinson and Wareing (189) also showed an inhibitor to vary with photoperiod. Supported by Cornforth, Milborrow and Ryback, this group (46) showed that the active substances from birch and sycamore leaves was identical to the structure elucidated by Ohkuma. In a third program, initiated by Von Steveninck (217) and continued by Rothwell and Wain (192), the active substances causing abscission in lupin fruit was studied.Again, with the support of 23 Cornforth, Milborrow and Ryback (48), the active substances inyellow lupin was identified as abscisic acidSNow abscisic acid has been isolated from many plants (4,149,151) and has been shown to be the active constituent in the p inhibitor (149).In addition to being a natural inhibitor of growth and flowering and a natural promoter of abscission and senescence (4), ABA is also involved in other physiological roles.

2.Physiological Effects

a.Senescence and AbscissionAbscisic acid promotes the degradation of chlorophyll in isolated leaf discs (10, 21, 43, 64, 214) and the treatment is often associated with increased pigment levels

(208).It appears that the senescence effect of abs cisic acid acts through a decreased RNA and protein synthesis (43,103, 214) as well as increased protein breakdown (103) and increased RNase and DNase activity (132, 214). Some of the earliest experiments with extracts containing ABA were concerned with abscission of young cottonfruit.Application of the extract accelerated abscission of both intact fruit and defruited pedicels (1).In experiments including other species, application of abscisic acid to the petiole stump of explants led to earlier onset, increased rates, and a higher percentage of abscission (1, 208) When applied to intact leaves and foliage, complete defoliation 24 occurred with either one application (62) or was induced with several applications or prolonged exposure (208)

b.Growth Retardation and Inhibition,Much of the early work dealing with physiological responses to ABA used bioassays to measure growth inhibition.Bioassays used include coleoptiies (1,13), isolated cereal embryos (46), and rice seedlings (110).Other plant parts used to measure growth inhibition responses also include leaf discs (62, 237), hypocotyls (10, 62), radicles (10,13), and root sections (62). Inhibiton of stem growth has been observed in many plants. Also growth retardation of the leafy stem occurs, but effective growth inhibiton generally requires repeated or prolonged exposure to ABA

(4).In trees (deciduous) abscisic acid can induce the typical dormant conditions (62) and in potato, ABA can prolong the dormant condition in either the whole potato (64) or isolated buds (143) The prolonged dormancy effect has also been studied exten- sively in seeds.In the presence of abscisic acid some seeds will not germinate at all,This is particularly true with seeds of lettuce (10, 169, 255), and grasses (137). With seeds requiring stratificatLon (cold treatment) before they can germinate, levels of endogenous ABA have been observed todrop during stratification.This decreased level of ABA during stratification hs been shown to occur in the embryo of Fraxinus seeds (212) 25 and walnuts (145)Inpeach(137) and box elder (Acer negundo) (95), the source of the dormancy factor in the seed coat and in these too, ABA concentrations decrease with stratification.After proper stratification, the seeds can be kept dormant by ABA application (137,

211).Simply rinsing away the inhibitng solution of ABA permits germination to promptly resume (223).

Interaction with other Plant Hormones.In general abscisic acid counteracts the growth promoting effect of gibberellins, auxins, and cytokinins.With regard to GA-induced responses of growth, germination and senescence, ABA has shown a counteracting or inhibitng effect (39, 64,102, 266).Among these, the most striking inhibitory effect of ABA is on the GA-induced synthesis of hydrolases in theleurone layers of barley seeds (39, 40)In fact, ABA appears to inhibit the synthesis of proteases,ribonucleases and probably all other hydrolytic enzymes whose syntheses in the aleurone layers are promoted by GA (40). Antitranspirant Activity. A more recent and perhaps important physiological function of endogenous abs cisic acid is its possible role in regulating water relationships in the p1antAbscisic

acid has been reported to reduce transpiration (138, 162) and this

reduction was apparently due to stomatal closure (86, 24Z) The 26 possibility that abscisic acid could have an important regulatory function in the water relationships of plants gained support from the experiments of Wright and Hiron (267).They found that water stress could bring about a large increase in the amounts of ABA in the leaves. Further evidence of Imber and Tal (92) showed that a wilting mutant of tomato could be brought into a more normal turgor condition by the application of ABA,This specific and reversible effect of ABA on stomatal regulation gives further evidence to thehormonal control of stomatal behavior and further implicates ABA' s role in the overall water relationships of the plant.

e.Miscellaneous Responses.The fact that ABA increased in leaves under short day conditions, suggested that abscisic acid may act as an inhibitor to flowering of long day plants held under short day conditions.Evans (67) found that ABA injected into the long day plant Lolium temulentum strongly inhibited flowering.Inhibition of flowering in other long day plants using foliar sprays of ABA have also been reported (64). Associated with short day conditions is the increased cold hardiness of some woody species (93, 94).Irving (96) has been able to extract from short-day--treated box elder (Acer negundo) an abscisic acid-like inhibitor.Treatment of the non-hardy box elder plants under long day conditions with either the extract or abscisic acid increased hardiness after a hardening period.Irving (96) further 27 suggests that the hardening process is closely related to the build-up of abscisic acid and a reduction in the gibberellin levels.

3.Chemistry

Abscisic acid is the common name for the structure proposed byOhkumaetal. (174): methyl-2'-cyclohexen - 1'-yl)-cis, trans-2, 4-pentadienoic acid (XIX). Of the two possible enantiomorphs, only (+)-abscisic acid has been found to occur naturally (150).The C2-cis-isomer (XIX) is by far the most common occurring but the C2-trans-abscisic acid isomer (XX) has been reported in a few species (4 and references therein,

178).However, care must be observed in its isolation procedure because the C2-cis- and C2-trans-isomers are readily intercon- verted by a photomechanism (164).

OH COOH 3,

aThe absolute stereochemistry shown is not correct,The correct stereochemistry and the reasoning behind its revision will be explained on page 31. 28 The trans-isomer (XX) is considerably less active in growth inhibition showing less than one percent of the activity of abs cisic acid (XIX) (45).Little else is known of its physiological activities. Other naturally occurring substances related to abscisic acid are phaseic acidk abs cisyl-p --glucopyranoside, and xanthoxin, Both phaseic acid and the ABA glucoside are products of ABA metabolism (discussed later) and are essentially inactive in inhibiting growth and germination (152).Xanthoxin is not related to ABA metabolism and has biological activity comparable to ABA (170), The relationship of these compounds to ABA biosynthesis and metabolism in addition to their structures is discussed in the next section, To avoid confusion in abbreviated names for cis- and trans-ABA the 2-cis, 4-trans-isomer (XIX) will be designated as cis-ABA and the 2-trans, 4-trans-isomer (XX) as trans-ABA. Where discussion will be concerned with the reduced forms of abscisic acid--the cis- and trans-diols--the abbreviated forms will not be used. Abscisic acid was first named abscisin II by Addicott' s group describing its abscission accelerating properties (1,175).At the time the trivial name abscisin II was being introduced for abscisic acid, the term dormin was being widely used for a dormancy inducing substance in birch and sycamore extracts (250),Since both groups were dealing with the same compound, they reviewed the matter of nomenclature at the Sixth International Conference on Plant Growth Substances (Ottawa, July 1967), and agreed on the new name abscisic 29 acid (2,3).Thus abscisic acid now precludes the use of abscission II and dormin but does not preclude the use of dormin for classifying substances active as endogenous dormancy- inducers (250).The name abscisin was given by Liu and Cams (141) to a compound isolated from burs of mature cotton fruit.The name was later changed to abscisin 1 (1) and no further characterization or investigation in other species of abscisin I has occurred. Ohkuma (175) first isolated from young cotton fruit (Gossypium hirsutum) a crystalline substance that was a C15H2004 acid.Soluble in many organic solvents,it had a melting point of 160 to 161 0 C, sublimed at 1200 C and had an absorption maximum in methanol at 252 nm.Further research by Ohkuma (176) using elemental analysis, mass spectroscopy (ms) infrared spectros copy (i, r. ) and nuclear magnetic resonance (n.m. r. ) verified the structure shown for abscisic acid (XIX).Cornforth, Milborrow and Ryback (45) confirmed the structure by synthesis and further showed that ABA is dextrorotatory (47) and had an (S)-configuration (XIX) around the asymmetric C1carbon (49), The absolute stereochemistry of abscisic acid was deduced by Corriforthetal. (49) by reducing the methyl ester of ABA (XIX) and applying MillTs rules to the diol derivatives (XXI and XXII) epimeric at C4'. Cornforth had previously (45) made the racemic epidioxide (XXIII) which upon hydrogenation gave the cis-diol ester. A method 30

NaBH H O/MeOH 2o HO

XXIcis-diol ester XXII trans-diol ester

XXIII to distinguish between (XXI) and (XXII) was therefore possible.By comparing the molecular rotations for the diol pairs derived from (+)- and (-)-abscisic acid, Cornforth determined the cis-diol ester derived from (+)-ABA was more laevorotatory than the trans-diol ester [and conversely for the diol esters derived from (-)-ABA1. This then established the absolute stereochemistry of (+)-abscisic acid to be that of (XIX), Abscisic acid is an unusual compound in that it is a sesquiterp- ene possessing a carotenoid structure.In fact, the structural simi- larity between abs cisic acid and violaxanthin (XXIV) has led many investigators to question the absolute stereochemistry of either

OH viola.xanthin or abscisic acid (29). HO

1, 0

H XXIV 31

There was further doubt about the configuration of abscisic acid when violaxanthin was converted into (+)-abscisic acid (29, 230). This suggested that the stereochemistry of abscisic acid should be revised to the (R) configuration.To provide additional and unambiguous proof of ABA' s stereochemistry, Ryback (196) applied a chemical correlation with malic acid,Methyl-(S)-2 - acetoxy- 3- carboxypropi- onate (XXV) and an excess of methyl dimethyl malonate was electro- lyzed in methanol containing sodium methoxide to give the laevorotatory acetoxy-diester (XXVI).

COOH

H COOMe H, /

AcO COOMe COOMe

xxv

Abscisic acid, isolated from avocado fruit (196) was reduced to the cis- and trans-diol esters.The trans-diol ester was acetylated (monoacetate) subjected to ozonolysis, oxidized with performic acid and finally methylated with diazomethane.This procedure resulted in the isolation of the identified acetoxy-diester (XXVI),This procedure defined the structure of the trans-diol acetate of methyl abscisate to be (XXVII) and that of the natural (+)-abscisic acid to be (XXVIII). 32

Ac 0 2 03 NaBH HC000H 4 CO OH 4. (TH2N2 0 Me COOMe COOMe COOMe 0 XXII OAc

trans-diol ester XXVI

XXVIII

In their earlier paper, Cornforth, Draber, Milborrow and Ryback (49) specified the (S)-configuration for abscisic acid (XIX). This designation (S) was assigned according to the then accepted Cahn, Ingold and Prelog (31) system.The new stereochemistry of (XXVIII) would then be (R).However, a modified way of dealing with double bonds when applying the sequence rules was proposed (32).This modification of the Cahn, Ingold and Prelog system resulted in the C1t o abs cisic acid and related compounds being changed from (S) to (R) or (R) to (S).The proposals are now generally accepted (259) and (XVIII) is now designated as (S)-(+)-abscisic acid.The revision 3f (R)-(+)-abscisic acid to (S)-(+)-abscisic acid has received additional support. oth Koreeda, Weiss and Nakamishi (108) and Harada (85) 33 used optical data to arrive at the same conclusion--that natural (+)- cis-abscisic acid should be represented by the (S)-configuration.

4.Biosynthesis and Metabolism

The unusual chemical structure of abscisic acid has stimulated a number of speculations as to its biosynthesis.Two modes of biosynthesis are possible:(a) via an isoprenoid pathway; or (b) via a precursor such as violaxanthin, a carotenoid of wide spread occurrence whose structure might give rise to ABA. Noodle and RobLnson (171) have shown that 2-14C-mevalonic acid can be incorporated into abscisic acid by the ripening fruits of avocado, tomato, banana and strawberry. Also, Milborrow and Noodle (152) were able to show that 5-( 1, 2-epoxy-2, 6, 6' -trimethylcyclohexyl-)- 3-methylpenta-cis-2 - trans-4-d:ienoic acid (XXIX) incorporated into ABA whereas the 2-trans- epoxide (XXX) did not,This indicated that the 2-!-configuration of the double bond occurred early in the biosynthesis.

COOH

Y)cIX xxx 34 Since abscisic acid has a carotenoid structure, its formation from xanthophylls was speculated (226),Taylor (227) found that radiation of violaxanthin (XXIV) and other xanthophylls gave rise to a neutral product which showed strong growth inhibition.It is of interest to note that radiation of rhodoxanthin (XXXI) did not give rise to any neutral growth inhibitor (227).

XXXI

Taylor and Burden (229) were able to isolate the neutral inhibitor and tentatively suggested the term xanthoxin for two isomeric structures (XXXII) and (XXXIII).Burden and Taylor (29) confirmed Chat xanthoxin was indeed a mixture of (XXXII) and (XXXIII) and further showed their conversion (CrC3 oxidation) into cis- and trans- abscisic aldehyde which upon further oxLdation (Ag20) formed trans- abs cisic acid (XX). Xanthoxin was shown to occur naturally in seedlings of dwarf beai and wheat (228) and a method for its detection and estimation in plants was developed (70).Although xanthoxin has been formed from violaxanthin (XXVI) by photo-oxidation in vitro (226, 229) there is evidence for its formation by an in vivo photosensitized oxidation (30) 35 and enzymatic oxidation (171).

Cr0 OOH

CHO CHO

XXXIII

The importance of xanthoxin as a precursor to abscisic acid is not known.Certainly the major biosynthetic route to ABA is through mevalonic acid.There is a possibility, however, that the conversion of xanthoxin to ABA is not essential for its activity and the neutral inhibitor may be active per se (70). Many investigations have shown that the effects of ABA are relatively transient and that for many types of plant processes, ABA must be in continuous supply to be effective (4 and references therein). This is probably a result of the plant processing the biochemical means to inactive ABA. Milborrow (150) has labelled three products resulting from the metabolism of ABA; metabolites A, B and C all of which are biologically inactive.In a later experiment, Milborrow identified two radioactive products when (+)-{2-'4C]-abscisic acid was metabolized by tomato shoots.A water soluble, neutral product, metabolite B, 36 was identified as['4C]-abscisyl-p-D-glucopyranoside (1 53) (XXXIV). The balance of the radioactive (+)-ABA was converted into an acidic compound, metabolite C, which was identified as (XXXV).Methyla- tion of (XXXV), gave an ester which rearranged rapidly to phaesic acid (XXXVI) and showed 1/200 of the biological activity of abscisic acid (152). HOCH2

c=o

H2OH 0

HO H OH XXXIV That phaesic acid is a metabolite of abscisic acid was also indicated by Tinellietal. (236),In addition, they provided evidence that ABA can be further metabolized to the 4-dihydrophaseic acid

(XXXVII).

OH HO COOMe COOH H XXXV' 37

C.Dormancy

Passing through a state of dormancy is common to nearly all land plants.Usually the phase of dormancy coincides with a period of unfavorable climatic conditions (either tow temperatures, or high temperatures and drought). Although difficult to define precisely, dormancy refers to any phase in the life-cycle of the whole plant (or a particular organ) in which active growth is temporarily suspended (256).Within this broad definition, there are several types of dormancy, all determined by the organ's ability to resume growth under favorable environmental conditions.The period of growth inactivity caused by unfavorable environmental conditions (low temperature, drought) is termed "imposed dormancy" or "quiescence" (191).At the other extreme, growth cannot be resumed even if environmental conditions are favorable. Full dormancy is not attained suddenly but rather gradually over a peiiod of time.In the case of Douglas-fir, the initial stage of dormancy is termed "predormancy" or "early rest." In this state, bud growth can be resumed by various treatments such as long days, optimum temperatures, or ample moisture supply (125) As predormancy progresses, however, it becomes increasingly difficult to induce resumption of growth by changes in environmental 38 conditions (123).The bud then enters a state of full dormancy or "mid..- dormancy." This phase is followed by a gradual emergence from dormancy, during which it becomes progressively easier to induce resumption of growth and this state is referred to aspost- dormancy" or "after rest' (194). In summary, when plants are in the predormant condition they still have the capacity for growth but only in a narrower range of environmental conditions.The range of environmental conditions in which plants retain their capacity for growth becomes narrower.The state of dormancy continues to deepen until a condition of true dormancy is attained.Once this state has been reached, shoot apices cannot be induced to elongate even under the most favorable environmental conditions.Subsequently, true dormancy is terminated and a transition to a condition of"after-rest' or post dormancy occurs. During post dormancy the tissues are again able to resume growth, at first under very narrow environmental limits and later under wider ones.Finally a state is reached in which the tissues are completely released from dormancy.At this time the environmental limits in which growth can occur are widest.For a fuller explanation of the narrowing and widening environmental growth limits associated with dormancy and a more detailed discussion of the various forms of dorntancy see Vegis (244). There are a number of phases recognized in the development 39 o resting buds.First, there is a cessation of extension growth with no further expansion of leaves or extension of internodes. How- ever, meristernatic activity at the apical mer istem still continues with new leaf primordia being formed during this period. Where bud scales are formed from modified leaves, the outer primordia show greater marginal growth than primordia destined to develop into normal leaves,These expanding primordia give rise to bud scales or "cataphylls. Ultimately all growth and meristematic activity ceases with the formation of the fully developed bud.In some species, growth of the bud may continue over many months from June to the end of September (197).In Douglas-fir the buds form in mid July (37) and expand slowly into late fall.

1.External Factors Causing Bud Dormancy

Factors causing an actively growing shoot to cease growth and form a resting bud depend upon species and age of the tree.Day length (photoperiod) has been shown to markedly affect the onset of dormancy in many woody species (167, 247), including Douglas-fir

(124).In the majority of woody species investigated, short days (less than 10 hours illumination) strongly promote the formation of resting-budsand the onset of dormancy (256).Such short day responses occur not only in hardwoods but in conifers as well.Species such as birch and locust are very sensitive to day length and may be 40 kept in continuous growth for as long as 18 months if they are main- tained under long-day, greenhouse conditions, whereas they cease growth in about 14 days under short photoperiod.On the other hand, species such as ash show very little response to day length (256). The extent to which day length affects the induction of dormancy in natural conditions is uncertain.In species such as poplar and larch (256) active extension growth can occur well into September and October (extension of the natural day length by artificial illumination prolongs growth of the seedlings).However, older trees of woody species cease extension growth much earlier in the season (June, July or early August) when natural photoperiods are still long.Here day length probably does not play a major role in controlling shoot extension growth and other environmental conditions must be con- side red. Soil moisture content and mineral nutrient levels are important but the possibility that they play an overriding role in controlling the duration of extension growth seems unlikely (256); even under favorable soil moisture and nutrient content mature trees grow for a limited period each year.Internal competition for nutrients among the tree's various branches and shoots would appear to be an important factor in determining the annual increment of extension growth. Thus, the determining factor may not be the depletion of nutrients ii the soil but exhaustion of nutrients, including organic metabolites, 41 within the tree itself (111,1 12,186).The one thing that is clear is that the factors causing the initiation of dormancy are not fully understood and more precise experimental data are needed.

2.Removal of Bud Dormancy; Resumption of Growth

Buds of temperate species which enter dormancy in winter resume growth the following spring.Thus, the innate dormancy condition has been broken during winter,It is well known that buds of many woody species require a period of chilling before growth can be resumed (44).The most effective temperature range for overcoming dormancy is between 10 and 100 C and the length of the chilling period varies from 260 to 1000 hours (198). In many woody plants, the chilling requirement is generally met by January or February but the bud still remains dormant.In this case, temperature requirements for active growth are not being met and innate dormancy (rest) has been replaced by imposed dormancy (quiescence).The actual time of bud break is then determined by the rising temperature of spring (242). Finally, dormancy can be broken by a wide range of applied organic substances including ethylene, chlorhydrin, thiourea, and dinitrophenol (256).Of special interest is the role gibberellic acid plays in overcoming dormancy and this will be discussed in more detail under Hormonal Control of Bud Dormancy. 42

3.Internal Control of Bud Dormancy

The influence of environmental factors is mediated through internal controls.Smith and Kefford (207) postulate three dormancy phases of development that must be included in an overall theory of the mechanism of internal regulation of dormancy.These include: phasic development of dormancy culminating in a fully dormant state. breaking of dormancy leading to a nondormant state; and growth initiation in the spring leading to a steady-state development. The relationship of these dormancy phases are shown in Chart 1. For many years the regulation of bud dormancy was assumed to be a. single process rather than a series of transitional ones and the overall process was limited by a single endogenous chemical regulator. A review of these early theories (111) including the role of auxin in bud development is available.It now appears that the mechanism of controlling bud dormancy is considerably more complicated. Two tb.erories of bud dormancy control are currently being considered:

1) the bud scales interferewith oxygen uptake of internal tissues causing them to enter a dormant state after prolonged anaerobiosis; and 43

Cold resistance

'3,

Dormant Short Aging state days process / / Gibberellins \ / / Dormancy / inducer Long Chilling days

Spring Non-dormant Spring burst steady-state state

Chart 1.Model of the r elationship of dormancy phases of bud development to the annual cycle of Temperate Zone trees.The three steady states are in heavy lettering and the transitional phases in enclosed arrows.Mediation of environmental and internal factors is shown by broken arrows.From Smith and Kefford (205). 44 bud dormancy is controlled by the availability and balance between endogenous hormonal growth inhibitors and pro-

mote r s. The two theories are perhaps not mutually exclusive, each having its own merit.However, current work seems to favor the control of bud dormancy through hormone action.As Eagle and Wareing (62) emphasize, gaseous exchange cannot be blocked until the buds are formed.Morphological changes involved in formation of resting buds occur before gaseous exchange is restricted.For woody species at least, the hormonal theory seems more applicable and it only will be discussed.

4.Hormonal Control of Bud Dormancy

Hemberg (90) was among the first to evaluate the mechanism of bud dormancy in terms of specific endogenous growth inhibitors. He found growth inhibitors in terminal buds of Fraxinus and noted that they decreased in amount as buds were released from dormancy. Subsequently, growth inhibitors were found in buds of a variety of species of woody plants, often associated with promoters (256, and references therein). Considerable evidence is available showing that the regulation of bud dormancy and the internal processes associated with it are correlated with parallel changes in the levels of endogenous growth 45 promoters and inhibitors,Inhibitors, auxins, gibberellins and cytokinins all are involved in various ways.Inhibitors (abscisic acid) have been shown in some species to promote the development of dormancy (256), whereas growth promoters (gibberellins) appear to play a major role in breaking dormancy.The primary evidence linking endogenous regulators (principally growth inhibiting compounds) with the development of dormancy is as follows (252). under short-day conditions the leaves of many woody plants inhibit the growth of the shoot tip; greater amounts of inhibitors are found in leaves and buds of many woody plants under short-day than long -day conditions; and when inhibitors are extracted from leaves of a dormant woody plant and reapplied to plants of the same species which were not dormant prior to the application of the inhibitor, shoot elongation stops and sequential development toward a dormant state is initiated. Phillips andWareing (182) have shown that a strong correlation exists between seasonal variation in the amount of inhibitor and the state of dormancy in sycamore.The amount of inhibitors in the leaves and buds increased up to mid August.Beyond August the inhibitor levels did not increase further and decreased only after the leafless plants were exposed to winter conditions.Phillips and Wareing (182) suggested the inhibitor was formed in the leaves and translocated to the apex.They also found (181) that the concentration of inhibitors in leaves of sycamore was always less than one fifth as much as in the 46 apical tissues. In a similar study, the inhibitor content of Betula pubes cens (99) increased in concentration in the growing points with an increase in the duration of short days and decreased in porportion to the duration of long days to which they were exposed after they had become dormant.Inhibitors were found throughout the seedling but were most concentrated in the growing points followed by leaves, stems and roots, in that order. Wareing (246) found similar results.Exposing the apical buds of Betula pubes cens to short days induced dormancy.Transferring the plants to long days reversed growth; they also resumed growth if the mature leaves were removed and the buds themselves exposed to long-days. On the other hand,if the buds were exposed to long- days and the leaves to short-days, the buds remained dormant.Thus the response was monitored in the leaves and translocated to the buds. The production of growth inhibiting substances under short days gained further support when higher amounts of inhibitors were extracted from the leaves and buds under short-day than under long-day (167, 182, 183) conditions.Moreover, increased amounts of inhibitor can be detected two days after transferring the seedlings from long- day to short-day conditions (182,183), indicating the changes in inhibitor levels precede the formation of resting buds. Other evidence linking inhibitors to the induction of dormancy 47 include defoliation experiments (167) and extract applications (61, 62), In many species, removal of leaves stimulate buds (which have ceased growing) to grow again, even under short-days, or causes leaf primordia to become foliage leaves rather than cataphylls (167).Eagles and Wareing (61, 62) were able to induce dormancy in Betula verru- cosa by applying an inhibitor extracted from the dormant buds.Plants treated with the extract containing the inhibitor stopped growing within 12 days, showed closed stipular scales around the apex, and were absent of expanding leaves.Application of gibberellin to the inhibitor- arrested buds caused them to grow again, emphasizing that the failure of buds to grow had not been caused by toxicity. The inhibitor material found in the leaves of woody plants which varied with photoperiod and was capable of inducing dormancy in the same plants was determined to be abscisic acid (see Section Il-B). Application of ABA to seedlings of woody plants (Betula pubescens, Acer pseudoplatanus and Ribes nigrum) growing actively under long days induced dormancy (61, 62, 64).Thus, for some hardwoods at least, both the variation in the levels of endogenous and the results of application of synthetic ABA are consistent with the hypothesis that it plays an important role in the induction of bud dormancy. The release of dormancy is associated with an increase of growth promoters, such as auxins, gibberellins and cytokinins but not necessarily contributed to a decline in the level of endogenous 48

inhibitors (127).As mentioned, very little evidence is availablefor the action of auxin on bud burst,However, the substitution of the

chilling requirement by application of gibberellin(59,142) and cytokinins (15, 66, 127, 257) has been shown. Associated with the decrease (or increase) in ABAlevels is an

increase (decrease) in gibberellin levels (96,107,128,168, 266). Thu5, as dormancy is broken ABA levelsgo down and GA levels go up.The reverse is observed with the onset ofdormancy.This inverse relationship between ABA and GA coupled with the possibility that ABA can inhibit GA synthesis(209, 254) will undoubtedly receive more attention in the future especially in the conifers where much of

the above mentioned work has not beendone.

D.Wood Formation

1.Gambia]. Activation

There are two types of meristems associated with thegrowing stem, primary (apical) and secondary (cambial).The cells of the apical meristerns, dividing rapidly andmore or less isodiametrically, show little or no reaction to applied auxineven though they produce significant amounts themselves (236).Cells of the vascular cambium, however, show a clearresponse.Activation of the cambium by IAA was first demonstrated by Snow (210) and has since been demonstrated 49 in a number of species (see 113 and references therein, 233, 241,

251).IAA activation results in stimulating transverse divisions of the cambium giving rise to typical spring wood cells with thin walls and wide lumen (58).The distance over which this basipital activation occurs is a function of both species and season (83). That auxin naturally controls cambial activation was supported by the observation that tracheid diameters in pines were a function of day length.On short days, elongation soon ceased and the tracheids became narrow and typical of late wood (118,120).Transferring the pine seedlings to long-day conditions was correlated with high levels of auxjn (120) nd wider tracheids.Treatment with 2, 3, 5-triiodo- benzoic acid strongly inhibited auxin transport and consequently gave rise to narrower tracheids (120). Auxin has also been implicated in the formation of compression wood (gymnosperms) and tension wood (angiosperms).In softwoods (conifers), compression wood forms on the lower side of lateral branches and leaning stems and is generally red in color (thus the term T1rotholz" or redwood) (261, 262).It is thought to be caused by an excess of auxin diffusing to the lower side of the stem.This has received support from experiments where auxin in lanolin is applied to the stems of normal, vertical pine seedlings.The resulting auxin- induced wood is indistinguishable from compression wood (260).By varying the IAA concentration, forms of tracheids ranging from 50 typical spring wood to typical compression wood could be produced

(73). Hardwood (angiosperm) trees do not form compression wood but rather a light-colored (low lignin content) form of wood produced on the upper side of branches termed tension wood (83, 245, 262). It appears that perhaps tension wood is due to a lower than normal auxin level on the upper side, combined with the decrease in lignification which the lower auxin level would cause (36),Activation of the cambium on the upper side to specifically produce tension wood with its characteristic gelatinous fibers is difficult to explain, however. Lower levels of an inhibitor on the upper side has also been suggested (36).

2.Xylogenesis

The formation of wood is not solely controlled by auxins.Gib- berellins (23, 248) and cytokinins (213, 240) have been implicated with the possibility of inhibitors also being involved.Applications of GA alone resulted in thin-walled parenchymatous cells, while IAA alone produced sporadic xylem elements (248, 249).Only with both IAA and GA was normal wood formed. The balance between GA and IAA levels is also important. A high GA to IAA ratio resulted in phloem production while formation of xylem was caused by a high IAA to GA ratio (251, 263). 51

It appears then, that both IAA and GA are needed for cambial division with the action of both resulting in a synergistic effect.The further differentiation (increase in vessel or tracheid diameter, wall thickening., and lignification) are responses primarily to auxin.It might also be added that gibberellic acid enhances the activity of phenylalanine ammonia lyase (38) which catalyzes the conversion of phenyi.alanine to cinnamic acid derivatives (lignin precursors) (33). This enhanced enzyme activity resulted in increased lignification. Thus the gibberellins may also be involved in tissue differentiation. It is clear that hormones and the balance among them control and affect wood formation and wood quality.Further experimentation will undoubtedly involve other hormones (cytokinins and abscisic acid) in addition to implicating IAA and GA in new controlling roles. 52

III.EXPERIMENTAL

A.Collection and Extraction of Plant Material

Douglas-fir shoots were collected at a School of Forestry plantation located at Dorena Lake, approximately ten miles south of Eugene, Oregon.The trees were about 20 years old and came from a variety of seed sources.No particular care was taken to isolate shoots from the different sources. Approximately six inches of the terminal, woody stem was collected and placed in polyethylene bags.The shoots were trans- ported to the laboratory under dry ice and stored in the freezer until the buds, leaves and stems were separated (approximately two days). For the preliminary and first series (1970-1971) experiments, only the buds were analyzed.After removal of the leaves, the bud

0 was excised into cold (0 ),absolute methanol.Fresh weights were determined and the debudded stems counted for future references. Extraction continued at 00 in the dark with several solvent changes until no further color was extracted (at least eight weeks). In the second series (1972) the plant material was handled essentially the same except the stems, leaves and buds were stored in liquid nitrogen prior to extraction.After pouring off the liquid nitrogen, the buds, leaves and stems were placed under vacuum to remove any residual nitrogen. A fresh weight was taken and moisture 53 contents determined (oven-dry at 1100 for 24 hr.).The buds, leaves and stems were then separated and each ground separately in a Waring blender with methanol.The methanol (generally three solvent changes) was changed at regular intervals and extraction stopped when no further color was extracted. The methanol was finally removed under vacuum and care was taken to keep the concentrated extract cold (0°), away from light, and under nitrogen.The collection data are summarized in Table 2.

B.Fractionation by Solvent Partitioning

As a matter of routine all solvents for the preliminary and first series analysis (1970-1971, Table 2) were freshly distilled at least once from ACS reagent grade stock.Special care was taken to rid the diethyl ether of peroxides.This was done by refluxing the diethyl ether over sodium metal for six to eight hours after which the diethyl ether was distilled from the residue.In the second series (1972, Table 2) ACS reagent grade methanol and diethyl ether were used with distilled solvents being used only for final purification steps. For both preliminary and first series analysis (1970-1971) the concentrated extract was partitioned between diethyl ether and distilled water.The two fracticn s were thoroughly washed and the washings recombined with their respective fractions.The washed diéthyl ether fraction was extracted with sodium bicarbonate (0. 1 54

a Table 2,Collection data for the determination of abscisic acid in the dormant shoots of Douglas-fir.

Date Shoot Fresh Moisture Number part weight content, Yo of stems

First Series 9-15-70 Buds 183.7 g 907

10-24-70 Buds 394.0 2280

2-10-71 Buds 350.0 1808 5,.7-71 Budsb 717.0 2903

11-4-71 Buds 554.0 2760 Leaves 800.0

Second Series

1-8-72 Buds 527.5 42.4 1545 Leaves 350.0 43.7 Stems 276. 5 40. 4 339

2- 15-72 Buds 626.7 41.6 1771 Leaves 380.0 44.8 Stems 458.0 39.8 389

3-23-72 Buds 506. 2 39. 3 1969 Leaves 258.3 45.4 Stems 365. 2 38. 6 345

4-28-72 Budsb 1227. 2 29. 1 1972 Leaves 355. 8 47. 1 Stems 390. 2 36. 1 470

7-13-7 2 Buds 314.5 328 1047 Leaves 511.0 42.4 Stems 512.0 35. 6 678

10- 14-72 Buds 370.,2 41.4 1037 Leaves 464.5 41.6 Stems 586. 6 40. 6 1037 aAllcollections were made at the Dorena Lake plantation. bBudswere very swollen with some beginning to grow. 55 NaHCO3, pH 8. 2).After reacidification to pH 2. 5, the bicarbonate fraction was back-extracted with diethyl ether.Three treatments with the sodium bicarbonate solution was sufficient to remove the acidic compounds.This was designated as the first ether-acid fraction (lEA). The diethyl ether solution remaining from the bicarbonate treatment was next extracted with aqueous sodium carbonate solution(0. 1 N Na2CO3, pH 11.0) which removed the less acidic compounds.This carbonate solution was reacidified to pH 2. 5 and back- extracted with diethyl ether.This diethyl ether fraction was designated as the second ether-acid fraction (2EA).The material remaining in the original diethyl ether extract was considered neutral (dis- regarding the basic components) and designated as the ether-neutral fraction (EN).The fractionation scheme used for both the preliminary and first series (1970-1971) analysis is outlined in Figure 1. In the second series (1972) the fractionation scheme was modified as shown in Figure 2, to facilitate the isolation of abs cisic acid. Since AEA is slightly water soluble, the initial water layer was acidified to pH 2 5 and back-extracted with diethyl ether and combined with the lEA fraction.Since the sodium carbonate treatment did not produce useful results,it was not Lncluded in the second fractionation scheme. 56

Dormant Douglas-fir Buds

Methanol (00)

Methanol extract Residual buds

Evaporate at 00 and in subdued light Diethyl ether (500 ml) Water (500 ml)

Diethyl ether solubles Water solubles

Aqueous sodium bicarbonate(0. l, pH 8. 2) (repeated 3 times)

Diethyl ether solubles Aqueous sodium bicarbonate solubles (acids)

Hydrochloric acid (SN) to pH 2. 5, then diethyl ether extraction

Aqueous sodium carbonate solution First ether-acid Water solubles (0.1, pH 11.0) fraction (lEA)

Ethereutra1 fractiox (EN) Aqueous sodium carbonate solubles

Hydrochloric acid (SN) to pH 2. 5, then diethyl ether extraction

Second ether-acid Water solubles fraction (2EA)

Figure 1.Fractionation by solvent partition of the methanolic extract of the dormant buds of Douglas-fir; preliminary and first series (1970-1971) analysis. 57 Dormant Douglas-fir Shoots Buds, Leaves, Stems)

Liquid nitrogen

Buds, leaves, stems Liquid nitrogen

Separated

Buds Leaves Stems

Sane as stems Methanol in Waring blender

Methanol extract Residual stems

Internal standard added, extract concentrated Diethyl ether (500 ml) Water (500 ml)

Dietliyl ether solubles Water solubles

Aqueous sodium bicarbonate Hydrochloric acid (0. 1, pH 8. 2)(repeated 3 times) (5) to pH 2. 5, then diethyl ether extraction

Diethyl ether Aqueous sodium solubles bicarbonate solubles (acids) Water solubles First ether- acid fraction Hydrochloric acid (5 ) (lEA) to pH 2. 5, then diethyl ether extraction

First ether-acid Water solubles fractign (lEA)

Combined

Figure 2.Fractionation by solvent partition of the methanolic extract of the dormant shoots of Douglas-fir; second series (1972). 58

C.Bioassay

For the preliminary, first series (1970-1971), and second series (1972) analyses, the standard Avena coleoptile straight growth test (161) was used to screen the various bands obtained from the column and preparative thick-layer chromatographic separations (described later) of the lEA and 2EA fractions. Oat seeds (Avena sativa L. var. Rodney) were sterilized in a sodium hypochiorite solution (0.02 M) for one minute.The seeds were then washed, soaked in water for one hour, and spread over vermicu- lite to permit germination.After four days in the dark, the coleoptiles were one to two inches in length and ready for use. Using a double-bladed cutting tool,a six millimetre segment was excised from each coleoptile about three to four millimetres below the coleoptile tip.The sections were floated in a magnesium sulfate so1utjn(8.0x106M)for one hour prior to use. Aqueous solutions (3. 0 ml) of the extracts to be tested were prepared as well as known standards.The actual concentration of compouxds in the test solution was not known.However, a concentration gradient of 1, and X10 was used for screening purposes.The test was carried out in petri dishes (5.0 ml) fitted with lids. Ten coleoptile sections were floated on the test solution containing si.lcrose (0. 05 M).The petri dishes were covered and placed on 59 a. sla.nted, rotating platform.After two days in the dark, the solution was removed from the petri dish and the coleoptiles measured in length by projecting their image on a wall at X10 magnification.The results were compared to standard solutions and controls.

D.Determination of Active Compounds

Preliminary studies were done on the methanolic extracts of Douglas-fir buds using column chromatography.Prior to column separation each extract was fractionated according to the scheme outlined in Figure 1.Each lEA, preliminary series fraction was loaded onto a column (2. 5 cmx 30 cm) of Silica Gel (Baker,tic grade) and eluted sequentially with solvents of differing polarity (chloroform to ethyl acetate to methanol).Eluted zones from the columns were collected and subjected to the Avena bioassay.It was difficult to monitor the columns visually because resolutions were poor andmany of the fractions tailed (broad, poorly defined bands). Bioassay results from these separations showed the presence of only inhibiting compounds. A search of the literature revealed several studies dealing with the isolation and identification of plant growth hormones (including inhibitors).The procedures outlined by Steen and Eliasson (215) seemed most relevant because they were dealing with extracts obtained from Picea abies .Thus, a system involving Sephadex LF-20 was used to separate further the lEA fractions from 60 the three series. Preliminary bioassay results of the Sephadex separated lEA preliminary series fractions again showed only the presence of inhibitors.In this case, however, one of the inhibiting compounds corresponded to the elution volume of abs cisic acid in the Steen Eliasson (215) system.I then decided to pursue the presence of abscisicacidin the acidic extract (lEA fraction) of dormant Douglas- fir shoots.

E.Isolation of Abs cisic Acid

Collections from the first series (Table2)made on10-24-70 and 11-4-71 were used for the actual isolation and identification of abscisic acid.The10-24-70extract (25%) was fractionated (Figure 1), and subjected to column chromatography (Sephadex and Silica Gel), preparative thick-layer chromatography, and gas -liquid chromatography (gic).The results of the glc analysis of the extract showed poor resolution for the peak corresponding to abscisic acid. A more exhaustive procedure was then developed for tE 11-4-71extract using preparative thick-layer chromatography (preparative tic) exclu- sively.This method successfully isolated abscisic acid and permitted further analysis. A detailed description of the isolation procedure used for both the 10-24-70 and11-4-71extracts follows. 61

Isolation Procedures for the10-24-70Extract

a, Sephadex Separation.Further separation of the lEA and

ZEA (first series, Figure 1) fractions obtained from the10-24-70 extract(Table2) was necessary.The support (Sephadex LH-20;

5 cmx 35 cm) was packed into a Sephadex SR 25/45column.The extract was eluted with 90% ethanol at a flow rate of 30 ml/hr Fourteen distinct bands were discernible using an ultraviolet (UV) light.Because high energy radiation could cause structural rearrangement, later separations were made with sparing use of UV monitoring. When UV was not used, the columns were monitored by previously calibrated elution volumes.The separation of the 2EA (first series, Figure 1) showed similar bands.The ether-neutral (EN) fraction was not analyzed. The 14 fractions collected from the Sephadex column were subjected to the Avena bioassay.The fractions representing the first zone of inhibition (low elution volume) and corresponding to the known elution volume of authentic abscisic acid were combined for further investigation.

b.Silica Gel Separation.A column(1. 5cm x15cm) of Silica Gel (Baker column grade; 60/200 mesh; no binder) was used to separate the first biologically active zone of low elution volume obtained from the Sephadex separation of the lEA (Figure1) 10-24-70extract. 62 The extract was loaded onto the column and eluted at a rate of 30 ml! hr with chloroform-methanol (9 5:5 v/v).Seven fractions were collected and subjected to bioassay. The ultraviolet absorption spectrum of each fraction was taken using a Beckman (model DB) UV scanning spectrophotometer.All spectra were obtained in 95% ethanol.

c.Thick-Layer Preparative Chromatography.The results of the Avena bioassay showed that the zone containing the elution volume of authentic ABA corresponded to the region of major growth inhibition.The zone containing this material was combined, reduced in volume under vacuum (and in subdued light), and dissolved in a mm- imurxi volume of diethyl ether-acetone (2:1 v!v).Aliquots of the solution were applied as a narrow band (2 mm) onto glass plates coated with a 55Qp. layer of Silica Gel G (Baker tic grade containing 13% calcium sulfate). A Kontes Chromallex Streaker was used to apply the extract to theplates.As a reference marker, authentic (RS)- abscisic acid was spotted at either end of the plates on isolated strips. Many solvent systems were available for tic analysis of ABA. In particular, Milborrow (149, 154) lists several solvents of which benzene-ethyl acetate-acetic acid (15:3:1 v!v!v) and benzene-ethyl acetate-acetic acid (50:5:2 v!v!v) proved to be the most useful in the present work.The solvent system benzene-ethyl acetate-acetic acid (50:5:2 v/v!v) was used for separation of the partially purified 63 10-24-70 extract. The plates were developed once and the isolated strips containing reference ABA were sprayed with 2, 4-dinitrophenyihydrazine to detect the Rf value of abscisic acid.The zone corresponding to this Rf value was scraped from each plate and extracted with methanol. The material in this zone was now ready for methylation and gas- liquid chroma.tographic (glc) analysis.The isolation procedures used for the lEA fraction (Figure 1) of the 10-24-70 extract (first series) is outlined in Figure 3.

2.Methylation

Methylation was chosen for derivatization rather than silylation. Abscisic acid is instantaneously methylated with diazomethane and retains all the biological activity of the free acid (1 54).In addition, methyl abscisate is readily hydrolyzed back to the free acid when needed.In the mass spectral analyses (described later) MeABA produced less complicated results and readily permitted comparison to previously reported mass spectral data. Two methylation procedures were used.Where only a few samples had to be methylated and the quantity of substance to be methylated was small (20 mg), a modified method of the Schienk and Gellerman (199) procedure was used.Since diethyl ether was a poor solvent for this extract, it was replaced with acetone. 64

lEA Fraction(10-24-70Collection

SephadexLH-20separation column eluted with 90% ethanol

Residual column Zones1-14

Avena bioassay

I I I I I Zones 1-3 Zones4-8 Zone 9 Zones 10-13 Zone14 (not active) (very active)(not active) (slightly (not active) active)

Zones4-8combined and separated on a Silica Gel column eluted with chloroform-methanol (95:5 v/v)

Residual column Zones1-7

Avena bioassay

Zone 1 Zones2, 3, 4 Zones 5-6 Zone7 (not active) (very active) (slightly active) (not active)

Zones 2,3, and4 combined and applied to thick-layer chr omatographic plates

Non,-ABA zones ABA zone

lvthylation

glc analysis color test

Figure 3. Isolation of abscisic acid from the lEA fraction (Figure 1) of the10-24-70(first series) collection of buds. 65 The method employed three test tubes connected in series. The first tube contained diethyl ether through which a steady flow of nitrogen (approximately 6 ml/min) was passed into the second tube. In this tube diazomethane was generated from the reaction mixture of 2-(3-ethoxyethoxy) ethanol (0. 7 ml), diethyl ether (0.7 ml), potassium hydroxide (1.0 ml of a 60% solution), and an equivalent amount of N-methyl-N-nitroso---toluenesu.lfonamide(commercially available from the Aldrich Chemical Co. as Diazald),The sample to be methylated was in the third tube dissolved in acetone-methanol (9:1 v/v). The reagents were all added to the proper test tubes except the required amount of Diazald.Once this was added, the tubes were connected aid the generated diazomethane was swept into the sample mixture.The whole procedure took between 10 and 15 mm and resulted in 100% esterification if the ratio of Diazald to acid was at least 2:1. When the number of samples to be esterified was large and the amount of material not known, the micromethylation procedure was cumbersome and re sulted in incomplete esterification.Batch quantities of ethereai,-ethanolic diazomethane were prepared, therefore, according to the method outlined by Fieser and Fieser (69) and utilizing the reagents listed above. A 100-ml distilling flask was charged with a solution of potassium hydroxide (5. 0 g in 8. 0 ml of water) and 25. 0 ml of ethanolThe flask was connected to a 66 condensor whtch delivered into two receiving flasks attached in series (both cooled in ice).The second receiver contained diethyl ether (25. 0 ml) and the inlet tube dipped below the surface of the solvent. The distilling flask was heated in a water bath at 650 anda solution of Dia.zald (21.4 g,0. 1 mole) in diethyl ether (200.0 ml) was added from a dropping funnel in about 25 mm. When the dropping funnel was empty, another 40. 0 ml of diethyl ether was added slowly and the distillation continued until the distilling diethyl ether was colorless, The combined distillate contained about 3 g of diazomethane in approximately 300 ml of diethyl ether. The samples to be methylated were dissolved in a small amount (3. 0 ml) of acetone-methanol (9:1 v/v) and the diethyl ethereal diazomethane was added until the yellow color of the solution persisted. Excess diazomethane was then removed under vacuum and the methylated extract was dissolved in a known volume of acetone. The extract was ready for glc analysis.

3.Gas-Liquid Chromatography

Three studies (53,78, 128) were consulted prior to the glc analysis.Four liquid phases were initially tried, as suggested by Lenton, Perry and Saunders (128).These were SE-30, QF-1, OV-17 and Epoon 1001.In addition to these, DC-il and XE-60 phases were investigated.All liquid phases were coated on Chromosorb W {80/i00 67 mesh, acid washed (AW), dichiorodimethylsilane (DCMS) treated] using the high efficiency technique outlined by Kruppa, Henly and

Smead (116).For the initial trials, the coated support was packed into 6-ft teflon tubes (Pentube Plastic Company, Clifton, Pennsyl- vania) that were approximately 1/8-in outside diameter (0. D. ) (size designation AWG-10). To determine which liquid phase was most appropriate, the ability to resolve a mixture of cis- and trans-methyl abscisate (cis- MeABA and trans-MeABA) was used.By determining the Height Equivalent to Theoretical Plates (HETP) and Resolution Values (166, p. 23 and 33) for each separation, the liquid phase possessing the greatest resolving power was indicated. Two gas chromatographs were used.Preliminary analyses were made on a Varian ZOOB with flame ionization detectors.Quali- tative and quantitative MeABA determinations were made on a Hewlett-Packard 5750B Research Chromatograph also equipped with dual flame ionization detectors.Either helium (Varian) or nitrogen (Hewlett-Packard) was used for the carrier gas and the flow was kept at the optimal rate of 30 mi/mm. Hydrogen and oxygen flow to the flame detectors were 30 and 80 ml/min respectively.Column (oven) temperatures varied with the stationary phase being used: Epoon 1001 and OV-17, 2100; XE-60 and DC-il, 200°; and, SE-30 and QF-1, 1800.The stationary phase used to identify and quantify 68 abscisic acid in the 10-24-70 (first series) extractwas the Epoon 1001 phase.For these remaining glc separations 6-ft (1 /4-in 0. D., 1/8-mI. ii) glass columns were used (available from Applied Science

Laboratories, 11135 Inglewood Ave.,Inglewood, California), Prior to packing, the glass columns were silylated with dichlorodimethylsilane (DCMS).Silylation involved washing the glass tubing with acetone (300 ml), chloroform (300 ml), and acetone (300 ml), The tube was dried and loaded with a 10.0% (v/v) solution of DCMS in toluene.After sitting for 1 0 minutes, the DCMS solution was removed and the column washed with toluene (300 ml),The column was refilled with methanol.After sitting for an additional five minutes the methanol was removed and the column washed with more methanol until the pH was neutral.The column was dried and packed with the coated support using a vacuum at the exhaust end.To ensure even packing, the column was vibrated using a hand vibrator.All columns were conditioned in the gas chromatograph for two or three days at 10 to 15above the operating oven temperature before use.

4..Abscisic Acid Determination in the 11 -4-71 Extract

The results of the glc separation of the 10-24-70 extract indicated poor resolution and the presence of many interfering compounds. Even further preparative thick -layer chromatographic separation of the methylated extract using n-hexane-ethyl acetate (1:1 v/v) as an 69 eluting solvent failed to free the gic peak at the retention time corresponding to cis-MeABA of overlapping peaks.Thus, a more rigorous isolation procedure was developed utilizing only preparative tic on the Sephadex separated lEA fraction (Figure 1) of the 11-4-71 (first series) bud extract. In order to free the extract of interfering compounds, I developed two additional solvent systems: benzene -methanol- acetic acid (97:2:1 v/v/v) and chloroform-methanol-acetic acid (97:2:1 v/v/v). The lEA extract (Figure 1) was subjected to four preparative thick- layer chromatographic separations using the solvent system in the order listed: benzene-methanol-acetic acid (97:2:1 v/v/v); chloroform-methanol-acetic acid (97:2:1 v/v/v); benzene-ethyi acetate- acetic acid (15:3:1 v/v/v); and finally, benzene-ethyl acetate-acetic acid (50:5:2 v/v/v).The plates were developed two or three times to improve resolution.Reference (RS)-abscisic acid was again used and its resulting Rf visualized with 2, 4-dinitrophenyihydrazine (2, 4-

DNPH). In the first two separations, the eluted plates were divided into ten zones and subjected to the Avena bioassay.The last two separations were tested only for the zones below, above and including the zone corresponding to the Rf value of (RS)-abscisic acid. For each of the four separations, the zone corresponding to the R1 value of ABA was collected and extracted with freshly distilled 70 acetone.After the final separation of the free acids, the extract was riethylated,Gas-liquid chromatography of this extract still revealed an overlapping compound. An additional thick-layer preparative chromatographic separation of the methylated extract using n-hexane- ethyl acetate (1:1 v/v) was necessary to free the peak corresponding to cis-MeABA. The peak corresponding to cis-MeABA in the gic separation was now free of any interfering compounds and was ready for mass spectrum analysis.The overall isolation procedure is outlined in Figure 4.

5.Combined Gas-Liquid Chromatography and Rapid Scan Mass Spectrometry

Gas-liquid chromatography in combination with mass spectrometry (ms) was used to separate and identify the peak corresponding to cis-methyl abscisate.In the first attempt, a Hewlett-Packard (Model F and M 810) glc and Atlas CH-4, Nier type (nine inch, 60 degree sector) single-focusing mass spectrometer were used at the following conditions. glc column 5.0% SE-30; 6ft (1/8-in 0. D. ), stainless stee] temperature 2000 (isothermal) 71 lEA Fraction (11-4-71)

Sephadex separation Avena bioassay

Reidua1 zones Active Zones 4-8 (corresponding to cis -ABA)

First preparative tic; bioassay 10 zones

Zone 1 Zone 2 Zone 3-4 Zone 5-7 Zone 8-10 (slightly active) (very active) (slightly active (active) (not active)

(corresponding to cis-A BA)

Second preparative tic; bioassay 10 zones

Zone 1 Zone 2-3 Zone 4 Zone 5-10 (slightly active) (very active) (slightly active) (not active)

(corresponding tocis-ABA

Third preparative tic; bioassay 3 zones

Zone 1 Zone 2 Zone 3 (not active) (very active) (not active)

(corresponding to cis-A BA)

Fourth preparative tic; bioassay 3 zones

Zone 1 Zone 2-3 Zone 3 (not active) (very active) (not active)

(corresponding to cis-A BA)

Methylated Fifth preparative tic

Zone corresponding to cis- MeA BA

gic ms cis-trans conversion

Figure 4.Isolation of abscisic acid from the lEA fraction (Figure 1) of the 11-4-71 (first series) bud collection. 72

detector flame ionization, 2500

injector port 250°

helium flow 30 mi/mm ms

electron voltage 70 eV

filament current 20 1,iA analyzer pressure 9x10 torr

multiplier voltage 1. 60 KV scanning speed 6. 5 sec from m/e 24 to 500.

Difficulty was experienced in getting cis-MeABA through the silicone interface separating the gas- chroniatograph from the mass spectrometer and consequently no data were collected. In a second attempt, a Finnigan Series 3000 Gas Chromatograph Peak Identifier was used in conjunction with the Finnegan System 150 computer.The gas-chromatograph was not equipped with a detector; instead the column was connected directly to the ion source of the mass spectrometer.The following conditions were used: gic column 5.0% SE-30; 5-ft (1/8-in I. D. glass; temperature 200° (isothermal) detector total ionization 73 injector port 2500 helium flow 30 mi/mm

1-fl S

electron voltage 70 eV filament current 250 A multiplier voltage 140 kV mass range m/e 24 to 500

In the Finnigan System, the interface was a glass jet separator and readily permitted MeABA to enter the ion source while the helium carrier gas was pumped away.Since the Finnigan system was not equipped with a detector, total ionization in the ion source was the only output signal for the chromatogram.It was difficult then to re- late the chromatogram appearance obtained on the Hewlett-Packard 5750B gic to that obtained from the Finnigan system. Associ3ted with the Finnigan Model 3000 glc ms system was the Finnigan 150 computer.This data handling system controlled the operation of the mass spectrometer in addition to acquiring and processing the data.It also plotted the total ionization output (chromatogram output) and the mass spectrum of each peak (or part of each peak) in the chromatogram.In this way the peak corresponding to cis-methyl abscisate was located and its mass spectrum obtained. The data obtained from standards were then compared to that obtained from the extract. 74

. Additional Confirmation of the Presence of Abs cisic Acid

a.Conversion of cis-MeABA to trans-MeABA, Mousseron- Canet etal, (164) have formed the C2-trans-isomer of abscisic acid by irradiation with ultraviolet light.The reaction requires low energy and can be done in a pyrex glass vessel. The methylated extract obtained from the 11-4-71 buds which had previously been analyzed on the mass spectrometer was dissolved in methanol (Z5 ml).Nitrogen was bubbled through the solution in a pyrex test tube for about ten minutes prior to irradiation.The test tube was placed in a Rayonet Photochemical Reactor equipped with four UV lamps.Irradiation was continued from four to six hours and the conversion of C2-cis-abscisic acid to the C2-trans- isomer was monitored by glc analysis.

b,Color Test for Abscisic Acid.Mallaby and Ryback (144) have developed a color test for ABA in which it and its methyl ester are converted by acid-catalyzed dehydration into neutral products. One of these products, an unsaturated-lactone, gives an intense violet-red color with addition of alkali.The color, however, lasts only a few seconds and fades as the lactone is hydrolyzed. The 10-24-70 (Table 2) extract was used for this test.The extract and abscisic acid (10 mg) were each dissolved in formic acid (1 ml) and concentrated hydrochloric acid (0. 5 ml).The two solutions 75 were heated at 95° for 30 mm.A violet-red color developed in each sample upon the addition of aqueous-ethanolic sodiumhydroxide.

F.Seasonal Determination of Abscisic Acid

Once abscisic acid had been positively identified in theextracts from the dormant shoots of Douglas-fir, its concentrationthroughout the dormant season was determined. As previouslymentioned, two series were analyzed; first series (1970-1971) -buds only; and second series (1972) - buds, leaves and stems. Two methods to quantify abscisic acid in the extracts were used.In the first series (1970-1971) a known standardof abscisic acid was analyzed in parallel with the extract.Authentic abscisic acid (1.20 mg)was dissolved in methanol-water(5:1 v/v).The methanol was evaporated under vacuum, and water (100 ml) and diethyl ether (100 ml) were added.The fractionation procedure was carried out as previously outlined (Figure 1). In the second series (1972), the trans-ABAwassynthesized (described later) and added to the extracts (buds, leavesand stems). The extracts, containing the C2-trans-isomer as aninternal standard, were fractionated as outlinedin Figure 2.

1.Determination of Abscisic Acid in the First Series(1970-1971)

Isolation Procedures.In order to facilitate the analysis of 76 ABA and avoid the cumbersome and time consuming technique of exclusive preparative thick-layer chromatography, a new procedure was developed.The extracts were fractionated in the usual fashion (Figure 1) and eluted from a Sephadex LH-20 column.In this case, large columns (7 cm x 60 cm) were employed to separate the lEA (Figure 1) extract.Rather than going directly to preparative tic plates, columns of silica gel were again employed.Cornforth Milborrow, Ryback and Ware ing (46) outlined a method to isolate ABA using a mixture of Silic acid and Kieselguhr.They eluted this column with benzene-diethyl ether (3:2 vi In the present study, the benzene-ether solvent of Cornforth etal. (46) was employed to elute a Silica Gel (Baker, column grade, 80/200 mesh) column (2. 5 cmx 30 cm).Fractions corresponding to the elution volume of authentic cis-ABA were collected. Although this procedure removed considerable amounts of ma.teri3l, the extract was still too concentrated for preparative thick- layer chromatography.Additional Silica Gel column separations were done using the solvent system developed for preparative tic.Both berizene-methanol-acetic acid (9 7:2:1 v/v/v) and chloroform-methanol- acetic acid (9 7:2:1 v/v/v) were used with preference being given to the latter. The zone corresponding to the elution volume for authentic cis-.ABA was collected and reduced in volume under vacuum.This 77

extract was then streaked onto 550 x Silica Gel G plates and developed (3 times) with the benzene-methanoi-acetic acid (97:2:lv/v/v) solvent. The strip containing authentic reference cis-ABA was sprayed with 2, 4-dinitrophenyihydrazine and the zone corresponding tocis-ABA was scraped from the plates and extracted with acetone.The acetone extract was concentrated under vacuum, reapplied to preparative tic plates and developed with chloroform-methanol-acetic acid (97:2:1 v/v/v).The ABA-containing zone was collected, methylated, and finally rechromatographed using the solvent system n-hexane-ethyl acetate (1:1 v/v).The zone corresionding to the methyl ester of abscisic acid (Rf 0. 55) was extracted with acetone and the acetone extract reduced in volume to 0. 5 ml.

b.Gas-Liquid Chromatographic Analysis.The methylated extracts of the first series (1970-1971) were then ready for gic analysis.Only the XE-60 and Epoon 1001 columns were used because they offered the best resolution.The authentic cis-abscisic acid which paralleled the isolation procedures for the first series was also subjected to the same gic analysis. Quantification of cis-MeABA from the Epoon 1001 and XE-60 separations was determined by relating the area under the cis-MeABA peak to a previously established calibration curve.Peak-areas were determined by multiplying the peak height times the width at half peak

height.This method is fast with an error in determination of about 78 three percent(166, p. 158). Also, by running the authentic cis-ABA through the isolation and gic procedures it was shown that the procedures did in fact isolate ABA and an estimate of the efficiency of these isolation procedures was established. In order to improve the isolation techniques and increase accuracy of ABA determination in the second series (1972), the methods used in the first series (1970-1971) analysis were again modified and an internal standard was used.

2.Determination of Abscisic Acid in the Second Series (1972)

The procedures used for the first series (1970-1971) analysis were successful in isolating abscisic acid.However, interfering compounds in the gic analysis made quantification difficult.Thus, for the second series (1972) the isolation procedures were improved. To increase quantitative accuracy, an internal standard was added to the bud, leaf and stem extracts prior to being fractionated (Figure 2) and subjected to the isolation procedures.

a.Formation of the Internal Standards.Three analogs of abscisic acid were considered for use as internal standards of the second series (1972).Results from the glc analysis of the first series (1970-1971) showed only the presence of cis-MeABA in the etracts examined.Although the isolation procedures would have 79 also isolated trans-MeABA, its presence was not detected.Thus the C2-trans-isomer of abscisic acid was the first compound considered for use as an internal standard.The other two compounds considered were the pair of diols of abscisic acid formed by reduction of the -keto group. (i) Formation of trans-Abscisic Acid - The C2-trans isomer (XX, p. 27) of cis-abscisic acid (XIX, p. 27) can readily be formed by ultraviolet irradiation according to the method outlined (Section

III - E -6- a). Synthetic (RS)-abscisic acid (cis-ABA) was obtained from

Burdick and Jackson Laboratories (Muskegon, Michigan).Two sup- plies were used in this study (lot numbers 2520 and 3335).Lot number 2520 had approximately 5 percent impurities (determined from gic analysis) and a melting point of 1850.The accepted mp for ABA is 187-189 ° (45, 80).Abscisic acid (80 mg) from lot 2520 was recrystallized from chloroform-n-hexane (1:1 v/v).The purified cis -ABA had an mp of 187. 51880.Lot 3335 was not purified further. Approximately 10 mg of cis-abscisic acid was methylated by the procedure previously outlined (Section III-E-2),Methyl abscisate was dissolved in methanol (20 ml) and placed into a pyrex test tube (50 ml capacity).Nitrogen was bubbled through the solution for about 10 minutes prior to irradiation.The C2-trans-isomer was 80 then formed by UV radiation in the Rayonet Photochemical Reactor. In approximately six hours, gic analysis of the mixture showed an equilibrium was obtained containing 50 percent of each isomer, The mixture was either made up to a known volume and used as a standard for gic analysis or it was subjected to preparative tic chromatography to separate the two isomers. Prior to preparative thick-layer chromatographic separation of cis-MeABA and trans-MeABA, the esters were hydrolyzed to the free acids with potassium hydroxide.Thus, the isomeric esters of abscisic acid were dissolved in a solution of 95% ethanol-10% potassium hydroxide (1:1 v/v).Hydrolysis was permitted to occur for 24 hr at room temperature and in the dark.After 24 hr, the ethanol was removed under vacuum.The remaining aqueous solution was acidified to pH 2. 5 and partitioned with diethyl ether.The free acids were then ready for preparative tic separation. The cis-nd trans-ABA mixture was separated using the solvent system chloroform-methanol-acetic acid (9 7:2:1 v/v/v).The separation resulting from one development in the solvent was not sufficient to clearly separate the isomers.However, by developing the plates three or four times in the same solvent system, the isomers were isolated from each other.The separated isomers were then extracted from the Silica Gel with acetone and remethylated.The purity of each tsomer was determined by glc analysis.To determine that it was in 81 fact the C2-trans-isomer formed, its biological activity using the Avena, bioassay and its nuclear magnetic resonance (N. M. R.) spectra were determined. (ii) Formation of Dihydroxy-Abscisic Acid - The carbonyl function of abscisic acid was reduced to the alcohol using sodium borohydride.Abscisic acid (5 mg) was methylated using the previous techniquesand dissolved in methanol-water (1:1 v/v),To this was added a few crystals of sodium borohydride.The reaction was carried out in the cold (00)for 30 minutes.The formation of cis- dihydroxy (XXI) and trans-dihydroxy (XXII) cis-abscisic acid was monitored by the appearance of two peaks on the glc trace.No further characterization of the dihydroxy compounds was attempted. The gic analysis of the three compounds in addition to cis- MeABA is shown in Figure 5.The cis-dihydroxy-MeABA had a reten- tton time (11. 2 mm) that was very close to cis-MeABA (12. 6 mm). The trans-dihydroxy-MeABA had a retention time (2. 2 mm) on the solvent front tail.The separation is shown for the (S)-isomers but the results would be the same for the (R)-isomers as well. Because of their retention times on the Epoon 1001 and XE- 60 columns, neitbe r of the dihydroxy compounds was suitable for use a,s an internal standard.The trans-methyl abscisate, however, had a retention time (17.8 mm) that was considerably longer than cis- MeABA. In addition to this, the cis- and trans-ABA mixture displayed (S )-trans-4' -hydroxy- 2-cis- methyl bscisate (S) -cis-4' - hydroxy- 2-cis- methyl bscisate (S)-2-cis-methyl abscisate )S)-2-trans-methyl abscisate

gic conditions column- 3. 9% Epoon 1001, on Chromosorb W (80/ 100 mesh, DCMS, AW); 6-ft (l/8-in I. D.) DCMS-treated glass injection port - 255° detector - 2500 helium flow - 30 ml/min

OH COOMe

COOMe

2 4 6 10 12 14 16 18 Retention Time (mm) Figure 5.Gas-liquid chromatography analysis of a C2-cis- and trans-methyl abscisate and a C4'-cis- and trans- hydroxy- 2-c is-methyl abscisate mixture. 03 83

virtually the same chromatographic behaviour in both column (Sepha- dex and Silica Gel) and preparative thick-layer chromatography. Prior to isolation pr3cedures, a known amount of pure trans- ABA was added to each extract.

b.Isolation Techniques.The first two steps, fractionation (Figure 2) and Sephadex separation were still employed.However a good part of the extract remained.This material was mostly phenolic in nature and the two new procedures were adopted to selectively remove phenolic compounds. A common technique to rid solutions of phenolic material is to treat them with lead acetate.Cornforthetal. (46, 47) has successfully used lead acetate treatment in their isolation of abs cisic acid from sycamore.The technique described by Cornforth was attempted in this study.The lEA fraction (Figure 1) of the 11-4-71 leaf collection was dissolved in water to which sodium hydroxide (2) was added drop by drop so that the pH remained below 8. 0.To this, a saturated solution of lead acetate was added until no further precipitate was visible.The mixture was then centrifuged, the supernatant decanted, and the precipitate washed with water.The combined supernatant was acidified to pH 3. 5 and back extracted with diethyl ether.The diethyl ether solution was concentrated under vacuum and the extract was then methylated and ready for glc analysis. In an alternate procedure, phenolic material has been removed 84 from extracts containing growth hormones using polyvinylpyrrolidone (80, 128).In the technique used by Lenton, Perry and Saunders (128) the-acidic extract was first dissolved in a small quantity of ammonium hydroxide (28%) and the residual ammonia removed under a stream of nitrogen.The resulting ammonium salts were added to a polyvinylpyrrolidone (PVP) column in a small volume of water, and the column eluted with distilled water. Glenn, Kuo, Durley and Pharis (80) used PVP but they did not convert the extract into its salt form prior to elution.Instead, they used buffers and various pH ranges to selectively separate gibberellins, abscisic acid, zeatin and indole acetic acid. The procedure outlined by Lenton etal. (128) was adopted for this study.TEe zone corresponding to the elution volume of ABA from the Sephadex column was dissolved in an excess of ammonium hydroxide and the residual ammonia removed under a stream of nitrogen.Commercially available polyvinylpyrrolidone (polyclar AT) was sieved prior to use.Particles less than 1 50 diameter (passing through a 100 mesh screen) were discarded and the remaining PVP was exhaustively washed with water.The PVP was loaded intc glass columns (2. 5 cm x 45 cm) in water.The extract was added to the PVP column (2. 5 cm x 45 cm) and eluted with water at a flow rate of 70 mi/hr. From previously calibrated columns the elution volume of 85 cis- and trans-ABA was known.This fraction, corresponding to the zone containing ABA, was acidified to pH 2, 0 and back-extracted with diethyl ether (or ethyl acetate). Column chromatography utilizing Silica Gel was again used. In this case, column grade Silica Gel (Baker column grade 60/200 mesh) containing 13 percent calcium sulfate binder(tic grade) was used.The Silica Gel was equilibrated in the eluting solvent prior to use.The solvent systems previously used for preparative tic were slightly modified to assist in eluting ABA from the column faster. Again the benzene -methanoL-acetic acid and chloroform-rn ethanol acetic acid solvents were used, but in this case they both were made more polar (9 7:3:1 v/v/v). The extract which had been treated previously with PVP, was dissolved in the eluting solvent, added to the Silica Gel column (2. 5 cm x 24 cm) and eluted at a rate of 60 ml/hr. From previous calibration runs, the elution volumes of cis- and trans-ABA were known and this zonewas collected.Finally the extract was separated on preparative thick-layer chrornatographic plates (Silica Gel) using the benzene-methanol-acetic acid and chloroform-methanol-acetic acid (9 7:2:1 v/v/v) solvent systems.The zone corresponding to cis- and trans- ABA was collected, extracted with acetone, methyl- a,ted with diazomethane and further separated by thick-layer chromatography using n-hexane-ethyl acetate (1:1 v/v) as a solvent. 86 Avena bioas says of various zones from the column and preparative thick-layer separations were not done,The final preparative tic separation using n- hexane - ethyl acetate (1:1 v/v) removed a major overlapping compound (see Figure 9 and 10).In the isolation of cis.-ABA from the 11-4-71 (1970-1971) extract, an Avena bioassay was not done on this final preparative tic step.Because this major overlapping compound showed very close chromatographic similari- ties to cis-MeABA and was present in relatively large amounts, it was considered important to test its biological activity and if possible, determine its chemical nature. For this purpose the 3-Z3-72 (second series, Table Z) bud collection was used.After final preparative tic of the methylated extract, the chromatogram was divided into ten zanes and each was tested by the Avena bioassay. The Epoon 1001 and XE-60 columns were used for glc analysis. The conditions were as follows:

column 3. 2% Epoon 1001 (or 4.2%, XE-60) on Chromosorb W (80/1 00 mesh, AW, DCMS); 6-ft (1/8 in I. D. DCMS-treated glass; temperature 210° (isothermal) injection port 2550 87

dete ctor 250 nitrogen flow 30 mi/hr.

Quantitative determination of cis-MeABA was done by comparing the area under the peak corresponding to cis-MeABA with the area under the peak of trans-MeABA. The absolute amount of trans-ABA added to the extracts was known,Because of their structural simi- l3rity, and close chrorra tographic behaviour it was assumed that any loss of the trans-isomer would be accompanied by a similar loss in the cis-isomer.Thus the absolute amount of cis-ABA in the initial extracts was determined by the following relationship:

C c =t (i-) where c = absolute amount of cis-MeABA

t absolute amount of trans-MeABA added to the extracts prior to the isolation procedures

C area under the glc peak corresponding to cis-MeABA T= area under the glc peak corresponding to trans-MeABA

C.Determination of Dioctyl Phthalate

As mentioned previously, the isolation of abs cisic acid was affected by several overlapping compounds in the glc analysis.One of these compounds was removed only in the final purification of the methylated extract.Preparative thick-layer chromatography using 88 n-hexane-ethyl-acetate (1:1 v/v) as an eluting solvent was effective in removing this compound.In all other separations involving the free acids, this compound eluted with the zone corresponding to cis-ABA,

The value of the compound in the n-hexane--ethyl-acetate (1:1 v/v) system was 0. 2; cis-MeABA had an Rf value of 0. 55.Thus the corn- pound was readily separated from MeABA and could easily be isolated.

The compound did not react with 2, 4-dinitrophenylhydrazine.The NMR spectra of this compound was taken in deuterated methanol

(CD3 OJD). The mass spectrum of this compound was also determined. Again, the Finnigan 3000 Gas Chromatograph Peak Identifier was used, employing the following conditions:

gi c

column a mixture of 80% XE-60 (3. 6%)

and 20% HP8BP (3. 0%) on Chrom- sorb W; 1 /8-in 0, D. stainless steel; temperature 210° (isothermal) injection port 2500 helium flow 24 ml/min ms

electron voltage 70 eV filament current 250 iA 89 multiplier voltage 1.40 kV mass range m/e 24 to 350

In this case, the system 1 50 computer was not used and the spectra was recorded on a Honeywell 1 508 oscillograph. The glc analysis of this compound on the Epoon 1001 and the XE-60 column indicated it was a single compound (appearance of only one peak). 90

IV. RESULTS AND DISCUSSION

A.Collection and Extraction of Plant Material

Collecting samples from one seed source might have provided more accurate measurements of the absolute amounts of ABA present in the tissue since it is known that trees from seeds oI high elevation sources break dormancy later than trees derived from seeds of low elevation sources (165).However, the collection of samples from trees of various seed sources was not expected to alter the trends observed for the ABA concentrations throughout the dormancy season, The initial extracts were stored at 0 ° and under nitrogen, but this precaution was not necessary. Solutions of abs cisic acid including crystalline ABA kept at room temperature and exposed to the air did not show any change over a period of one year.It was necessary, however, to keep solutions of ABA in the dark or subdued light to prevent the C,-cis-trans isomerization.

B.Fractionation by Solvent Partitioning

The fractionation procedure used in this study is a standard isolation technique for ABA analysis (4).Immiscible solvents such as diethyl ether, ethyl acetate and methylene dichioride have been used to partition the extract, and to re-extract the organic acids out of the acidified aqueous solution. 91 In the case of Douglas-fir, it was found that diethyl ether gave the best separation between the aqueous and non-aqueous phases. Although emulsions formed with all three solvents mentioned, they separated fastest with diethyl ether. Based on solids content, the methanol extraction removed 1 3. 5 percent of the original fresh weight of the buds or 22, 5 percent based on dry weight.Of this starting extract, 48. 0 percent (or 10.8 percent of the dry weight of the buds) partitioned itself into the die thyl ether layer and 52. 0 percent of the extracted solids went into the aqueous fraction.

C.Bioassay

The Avena straight growth bioassay is one of many types used to determine biological activity of plant hormones.It is sensitive to most hormones (206, 235) including abscisic acid (4).It is not, however, the best bioassay for gibberellin analysis (178).If characterization of ABA w.s to be done with bioassays alone, additional tests would have been included.However, the Avena bioassay was used only as a guide to indicate fractions possessing inhibitory activity. When using the Avena bioassay, no attempt was made to determine the concentration of the extract being applied.It was considered important, however, to test a range of concentrations for each fraction because in elongation experiments, the growth increment 92 increases (IAA) or decreases (ABA) with increasing concentration (4, 233).

D.Isolation of Abscisic Acid

Results from the preliminary analysis (Appendix I) showed the presence of only growth inhibiting compounds. Because of the difficulties in relating these results to literature values, the system outlined by Steen and Eliasson (215) employing Sephadex LH-20 was

adopted. Bioassayresults (Appendix Il-i) of the fractions collected from the Sephadex separation of the lEA and 2EA (Figure 1) fractions of the 10-24-70 (first series) extract showed two majr regions of growth inhibition.The first region, corresponding tlow elution volume, was similar to that reported by Steen and Elias son (21 5) for abscisic acid. Attempts to isolate abscisic acid from the 10-24-70 extract by further column (Silica Gel) and preparative thick-layer chromatographic separations were not successful.In the glc analysis (Figures 38 and 39) peak corresponding to the retention time of cis-. MeABA was present but it was only a slight shoulder on the peak of a large overlapping compound.The results of the 10-24-70 separation did show, however, that the region of maximum grow th inhibition from the column separations (Sephadex and Silica Gel) did correspond 93 to the elution volumes of authentic ABA.In addition, the zones from the Silica Gel separation corresponding to the maximum inhibi- in the 255 to 260 nm region.This tion (Figure 36) also had amax was further evidence indicating the presence of ABA since it has a

maxat 258 nm.

1.Gas-Liquid Chromatography

In the initial trials, the use of teflon tubing facilitated packing and installation of columns.However, the packings in teflon did not possess the resolving power that those in dichlorodimethylsilane- treated glass did.Consequently, only DCMS-treated glass columns witheither the Epoon 1001 or XE-60 packings were used for subsequent trials. Preliminary glc analysis showed that all the liquid phases listed in Table 3 could separate a mixture of cis- and trans-MeABA,When comparing Resolution values (Table 3), the QF-1, Epoon 1001, and XE-60 phases best resolved the cis-, trans-MeABA mixture.How- ever, the QF-1 phase was not used since its ability th separate components in the methylated extracts of Douglas-fir tissues was poor. Although the X-60 phase possessed the best resolving power, it did not resolve the components of the methylated extract as well as the Epoon 1001 phase. Table 3.Characteristics of various liquid phases used for the gic analysis of a mixture of cs- and trans-methyl abscisate.a

Liquid phase % coatings Oven temp. HETPc Reso1ution' cis- trans- MeABA MeABA

DC-li 1l.0% 200 0.07 0.06 2.40

SE-30 4. 5 180 0. 27 0. 23 2. 23 OV-17 2. 5 210 0.09 0.09 3.01 QF-1 1.5 180 0. 13 0. 12 3. 96 Epoon 1001 4.6 210 0.08 0.07 3.56 XE-60 4 3 200 0.09 0. 08 4. 27

aAllmeasurements were determined at optical flow rates for He or N2 (30 mi/mm), 02 (60 mi/mm), and H2 (25 mi/mm).

All liquid phases were coated on Chromsorb W (80/ 100 mesh, AW, DCMS-treated) and packed into 6-foot DCMS treated glass tubing (0. 35mm I.D.) 2 c column length (cm) HETP and Resolution = where x = retention time and ypeak width at one-half the peak height. = 16 (x/y)2 95

2.Preparative Thick-Layer Isolation of Abs cisic Acid in the 11-4-71 Extract

The rigorous procedures developed to isolate ABA from the lEA (1970-1971, Figure 1) fraction of the 11-4-71 (Table 2) Douglas-fir bud extract were more tedious but they did produce a very clean extract. Bioassay results from the Sephadex separation of the 11-4-71 extract produced the same results observed for the 10-24-70 Sephadex separation (Figure 33 ).The first region of inhibition (elution volume 100 to 210 ml) was further separated by preparative tic using the four solvent systems describedpreviously (Ill-E-4).Figure 6 shows the bo3ssay results for the ten zones collected from the chromatogram. The first region of low inhibition corresponded to the Rf (0. 09) value of cis-ABA.The second zone of inhibition between Rf 0. 5 and 0. 9 was not characterized further. Figure 7 shows expected results.The only zone of major inhibition was confined to the region of the Rf (0. 18) value for authentic cis-ABA.The zone of inhibition at higher Rf value (Figure 6) was not observed.Figure 8 shows the bioassay results for the last two separations.Again, the only zone showing major inhibition was coifined to that corresponding to the Rf value for cis-ABA. Zone 2 (Figure 8-B) collected from the last preparative tic separationwas methylated and made ready for glc analysis. x 0rgin front

b 4 5 6 7 8 9 10 ABA

Inhibition considered significant 60 Figure 6.Avena bioassay of the first inhibitor zone from the Sephadex separated 11-4-71 extract and separated further on preparative tic using the first solvent system, benzene-methanol-acetic acid (97:2:1 v/v/v).

a1_layer chromatogram (tic) of the first zone of inhibition from the Sephadex separation.Detected with 2, 4- dinitrophenyihydrazine. bTenfractions from the tic chromatogram for bioassay. standards at concentrations 0.042 pj 0. 420 pMi 4. 2O,uj and 42OpM respectively. a

origin front

b c 1 3 4 5 6 7 8 9 10 ABA L1JWLUUJUL

0 4.J C Inhibition considered significant

Figure 7,Avena bioassay of fractions 2 and 3, collected from the first preparative tic separation (Figure 6) of the Sephadex separated 11-4-71 extract.Second solvent system chloroform-methanol-acetic acid (97:2:1 v/v/v).

aThinlayerchromatogram (tic) of the fractions 2 and 3 collected from the first preparative tic separation. Detected with 2, 4-DNPH. b Ten fractions collected from the chromatogram for bioassay. cABAstandards at concentrations of 0. O42pM 0. 420,uM 4. 2O,uij and 42.0 respectively. by0

I 2 Ii a

3 I a B S

b B

1 2 3 2 3 A BA

0C" U 0 I 0 Figure 8. Avena bioassay of fractions 2,3 and 4 collected from the second preparative tic separation (Figure 7) of the 11-4-71 Douglas-fir bud extract. aThin -layer chromatograms of developed extract detected with 2, 4-dinitrophenyihydrazine. A) third preparative tic separation of fractions 2, 3 and 4 collected from the second preparative tic separation (Figure 7).Solvent system benzene-ethyl acetate-acetic acid (15:3:1 v/v/v); B) f"urth preparative tic separation.Solvent system benzene-ethyl. acetate-acetic acid (50:5:2 v,v/v), bBioassay results for the three fractions collected from the third (A) and fourth (B) preparative tic separations. cioassay results for ABA standards at concentrations of 0. 042,iM; 0.42 Mi 4. 2Oij and 42. O,iM respectively. 99

Gas-Liquid Chromatography Analysis of Abs cisic Acid ixi the 11-4-71 Extract

Gas chromatography analysis of the methylated extract on the spoon 1001 column showed essentially one major peak (Figure 9-A) corresponding to the approximate retention time of cis-MeABA. How- ever, spiking the extract with authentic cis-MeABA did not enhance the major peak but did enhance the shoulder at slightly longer retention time (Figure 9-B). Furt1er separation of the methylated extract on preparative tic plates using the solvent system n-hexane-ethyl acetate (1:1 v/v) showed that the major peak was in fact not cis-MeABA.In the n-hexane-ethyl acetate system, the major component had an Rf value of 0. 55.Further, the major component did not react with 2, 4-dinitrophenylhydrazine.Approximately 95 percent of the major peak was removed in this step. The gic analysis on the Epoon 1001 column after this final clean up step now revealed two compounds (Figure 10-A) of which the one with a longer retention time corresponded to cis-MeABA.This peak was enh.nced with a spike of cis-MeABA (Figure 10-B) showing no resulting shoulders. The extract was also subjected to gic analysis on SE-30, DC-il, azd XE-60 columns.In all cases, a peak corresponding to the reten- tiox time of cis-MeABA was present and enhancement of this peak 100

Figure 9. Gas chromatographic analysis of the lEA (Figure 1) fraction of the 11-4-7 1 (first series) extract after four preparative tic separations. gic separation of the methylated extract. gic separation of the methylated extract with a spike of authentic cis-MeABA. gic conditions column - 3. 9% Epoon 1001 on Chromosorb W (80/100 mesh, DCMS, AW); 6-ft (1/4-in 0. D., 0. 35 mm I, D. DCMS- treated glass; temperature 2100 (isothermal), injection port - 255° detector - 250°. heliui flow - 30 mi/mm.

102

Figire 10.Gas chromatographic analysis of the lEA (Figure 1) fraction of the 11-4-71 (first series) extract after preparative tic separation of the methylated extract with n-hexane-ethyl acetate (1:1 v/v). glç separation of the methylated extract, glc separation of the methylated extract with a spike of authentic cis-MeA BA. gic conditions co1unn - 3. 9% Epoon 1001 on Chromsorb W (80/100 mesh, DCMS, AW; 6-ft (1/4-in 0. 0., 0. 35 mm 0. D.) DCMS-treated glass; temperature 2100 (isothermal). i3j ection port - 2550 0 detector - 250 helium flow - 30 ml/min.

104 showed no shoulders or extra peaks.The results of the gic analysis of the 1EA(Figure 1) fraction of the 11-4-71 (first series) Dpuglas- fir bud extract are summarized in Table 4 and further indicate the presence of abs cisic acid in the extract.The extractwas now ready for mass spectrometry analysis.

4.Combined Oas-Liquid Chromatography and Rapid Scan Mass Spectrometry

The Fewlett-Packard/Atlas System used a silicone interface between the glc and ms which at high temperature restricted the diffu- sion of MeABA through it.Consequently large quantities of sample were required to produce distinguis hable fragmentation patterns. Since large quantities of extract were not available, rio data was obtained. The interface used in the Finnigan system was not a silicone membrane but rather a glass-jet separator and MeABA analysis was readily obtained. A mixture of cis- and trans-MeABA was injected into the gas chromatograph.After the solvent peak was through, total ionization in the Ion chamber was plotted as a function of time (Figure 11). Thus every time a mass-spectrum was obtained, total ionization was recorded.The chromatogram was then reconstructed, with the largest peak corresponding to an amplitude of 100 and every other peak Table 4.Gas-liquid chromatography analysis of the lEA (Figure 1) fraction from the 11-4-71 Douglas-fir bud extract on four liquid phases.a

Liquid phase % Coating1' Column temp. Retention times(mm.. cis- trans- MeABA MeABA

SE-30 2. 0°/b 1800 std, 4. 3 5. 8

ext. 4. 3 DC-li 11.0% 195° std. 6.6 9.6 ext. 6.6 Epoon 1001 3.9% 2100 std. 12.6 17.8 ext. 12.6 XE-60 3. 6% 2000 std. 9. 3 15. 5 ext 9.2

a The extract was separated four times on preparative tic, methylated and separated one further time by preparative tic. bAllcolumns were 6-ft. (1/4-in. 0. D., 0. 35 mm I. D.) glass, DCMS treated. Analysis was done at optimal flow rates for He or N2(30 mi/mm.), H2 (25 mi/mm) and 02 (60 mi/mm).Injector temperature 250°, detector 255°.

CA11temperatures were isothermal. dRetentiontimes are given for both the standards (std. ) and methylated extract (ext.).

Q a

10 12 14 16 18 20 22 Time (mm)

Figure 11. Gas-liquid chromatographic separation of cis- and trans-methyl abscisate on the Finnigan series 3000 Gas Chromatographic Peak Identifier. Peak 'a" is cis-MeABA and peak 'b" trans-MeABA. gic conditions: column - 5. 0% SF-30n Chromsorb W (80/100 mesh); 5-ft (1/4-in 0. D., 0. 35 mm I. D. ) glass; oven temperature 200° (isothermal). inj ection port - 2500 helium flow - 30 mi/mm. 107 relative to it (Figure 12).From Figure 12 spectrum number 99 represented cis-MeABA and spectrum number 128, trans-MeABA.Figure 13 shows the m/e values for cis-MeABA and Figure 14 shows the m/e values for trans-MeABA. Again the computer plotted the largest peak (base peak) with an amplitude of 1 00 and all other peaks were plotted relative to it, The methylated lEA (Figure 1) extract obtained from the 11-4-71 (first series) Douglas-fir bud collection was then analyzed. Figure 15 shows the reconstructed chromatogram. Because retention times were not available and flame ionization detectors not used, it was difficult to correlate the chromatogram I obtained on my SE-30 column with the Finnigan system.Spiking the extract with cis-MeABA would indicate the peak corresponding to cis-MeABA in the extract but this was not necessary. The System 1 50 computer is also equipped to perform limited mass searches.Since the base peak of cis-MeABA is m/e 190 (Figure 13) (46), the computer was requested to scan the entire chromatogram output (Spectrum numbers 1to 120 - see Figure 15) for an m/e value of 190.Figure 16 shows the frequency of the m/e 190 peak as a function of spectrum number. The highest frequency of the m/e 190 peak occurred in spectrum number 67.The computer printout of m/e values for spectrum number 67 (Figure 15) can be seen in Figure 1 7. 253 100 - 90-

80 -

70 -

ci) 50 - 40-

30 -

10 -

0 I I I I I I I I I I I I I I I 0 10 20 30 40 50 60 70 80 90 100 110 120130 140 150 Spectrum Number Figure 12. Reconstructed chromatogram of the separation of a cis- and trans-MeABA mixture. Plot was reconstructed from Figure 11 by the Finnigan system 150 computer.Peak "a" cis-MeAlA and peak 'b" trans.-MeABA.

00 100

40 -

30 20-

.1iI iL Ii i I . 0 IIIII fL I,iI II r 20 30 40 50 60 70 80 90 100 110120 130140150160 170180190200210220230 240259 260 m/e Figure 13.Mass spectrum of authentic cis-MeABA.Spectrum number 99 from the Finnigan Series 3000 Gas Chromatographic Peak Identifier (Figure 12).Base peak is m/e 190 and other peak heights are percent intensity to the base peak.Parent peak of ATA (at m/e 278) is not observed. 100

QQ 60:

050 a) I:::

10 -

0 20 30 40 50 60 70 80 90 100 110120130140 150160 170180190 200 210220 230240250260 m/e Figure 14. Mass spectrum of authentic trans-MeABA. Spectrum number 128 from the Finnigan Series 3000 Gas Chromatographic Peak Identifier (Figure 12).Base peak is m/e 190 and other peak heights are percent intensity to base peak.Parent peak (at m/e 278) is not ohs erved. 100 -

70 - 60-

4.) 50

40 -

30 -

20 -

10 -

0 i u a i i i i i a i a 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Spectrum Number Figure 15. Peconstructed gic chromatogram of the methylated 11-4-71 extract (after five preparative tic separations)on the Finnigan Series 3000 Gas Chromatographic Peak Identifier, gic conditions: column - 5, 0% SE-30 on Chromsorb W (80/100 mesh, DCMS, AW); 5-ft (l/4-in 0. D., 0, 35mm I. D.) glass oven temperature 2000 (isothermal). injector port - 250° helium flow - 30 mi/mm. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Spectrum Number

Figure 16.Limited mass search for m/e 190 in the chromatogram (Figure 15) obtained from the methylated 11-4-71 extract using the Finnigan 150 computer. 100

80 -

70 -

50 - 5)

30-

20 -

10 - kiihl I! t1 III" I 0 IItl I I I I I 20 30 40 50 60 70 80 90 100110120130140150160 170180190200210220 230240250260 mje

Figure 17.Mass spérum (m/e values) for spectrum number 67 (Figure 15) obtained from the 11-4-71 methylated extract. 114 The m/e values (Figure 17) are identical to the standard, cis- MeABA (Figure 13) in both composition and percent of the base peak (excluding impurities in the extract determination - Figure 17).The major peaks abovem/e 41 in the mass spectrum for standard cis-MeABA (Figure 13), for the extract (Spectrum number 67, Figure 17) from the 11-4-71 Douglas-fir bud collection, and for published results are shown in Table 5 The m/e values for spectrum number 67 (Figure 17) agree with both the standard cis-MeABA and published results.Also the major peaks (m/e 190, 162, 134, 125 and 91) all agree with results published for methyl abscisate in balsam fir (140), radiata pine (97), apple juice (256), and carob fruit (163).The parent ion at m/e 278 was not observed but since its abundance is only one percent of the base peak it is generally not seen (97,140, 256). This evidence, in support of the data in Table 4,is conclusive for the presence of abscisic acid in Douglas-fir.The presence of the trans-isomer in the extract was not observed from any glc separations (Table 4).Further proof that trans-MeAIBA was not present in the extract was taken from the mass spectrum data.The limited mass search of the chromatogram (Figure 16) showed only significant amounts of the m/e 190 peak in the area corresponding to cis-MeABA, Since cis-ABA and trans-ABA are separated only with difficulty, loss of trans-ABA in an isolation step was not likely.This is the first 115

Table 5.Major peaks in the mass spectra of methyl abscisate obtained from the 11-4-71 (first series) Douglas-fir bud extract.

Standard Extract Published cis- from 11-4-71 data MeABA buds (46)

(1)a 260 260 (1)

246 (1) 246 (1) 246 (1)

219 (2) 208 (6)

205 (4) 205 (4) 205 (15)

191 (13) 191 (13)

190 (100) 190 (100) 190 (100)

162 (45) 162 (44) 162 (53)

134 (75) 134 (76) 134 (48)

125 (55) 125 (56) 111 (24)

91(52) 91 (60) 91 (27)

77 (25) 77 (27) 69 (32)

67 (30) 67 (44) 57 (50)

55 (25) 55 (24) 55 (31)

43 (22) 43 (27) 43 (42)

41(42) 41 (60) 41 (39)

a m/e; relative intensities in percent of the base peak at m/e 190 are in parentheses. 116 reported isolation of abs cisic acid from Douglas -fir [Pseudots uga menziesii (Mirb..) Franco].

5.Further Confirmation of Abscisic Acid

The conversion of cis-methyl abscisate to trans-methyl abscisate (164) isa specific test for ABA, Although other pentadienoic side-chains of similar configuration might be expected to show this transformation it is highly improbable that they would possess the same chromatographic behaviour.Figure 18 shows the UV-induced transformation of cis-MeABA to trans-MeABA in the 11-4-71 extract (previously subjected to mass spectrum analysis). The fact that the peak corresponding to cis-MeABA did show this transformation coupled with the positive test for a specific color reaction (144) further proves the existence of cis-MeABA in the acidic fractionof the Douglas-fir extract.

E.Seasonal Determination of Abscisic Acid

It was necessary to successfully demonstrate the presence of abs cisic acid in the extract before quantification could be attempted. The isolation techniques developed for the first series (1970-1971) successfully isolated abscisic acid, but quantification was difficult because of the presence of overlapping compounds.The results for this first series analysis are shown in the Appendix C.. 10 12 14 16 18 20 Retention Time (mm)

11-4-71 methylated extract UV radiation

I $$ I I I I 10 12 14 16 18 20 Retention Time (mm)

Figure 18. Conversion of cis-MeABA in the 11-4-71 extract to a mixture of cis- and trans-MeABA by UV radiation. 11-4-71 methylated extract before UV irradiation. retention time of cis-MeABA 12. 6 mm. 11-4-71 methylated extract after four hours UV irradiation. retention time of cis-MeABA 12.6 miii. retention time of trans-MeABA 17. 8 mm.

-4 118 Although the quantification results for the first series (1970- 1971) analysis of ABA were poor they did establish an important result.In all of the extracts in the first series examined by gic analysis, only cis-MeABA was present.The trans-isomer was not present in any of the extracts. When present, the trans-isomer is generally considered an artifact.However, through careful work, Milborrow and Noodle (154), and Gaskin and MacMillan (78) were able to show that trans-abs cisic acid does occur naturally. Establishing that trans-ABA was not present in any of the glc- analyzed extracts was important.The trans-isomer could now be used aan internal standard without complicating results by the presence of endogenous trans-ABA. Using trans-ABA as an internal standard accounted for losses in separation of cis-ABA.That is, since the chromatographic behav- jour of 2j- and trans-ABA are essentially the same in column and preparative tic separations,it was assumed that any loss in the tr3ns-isomer was also associated with an equal loss in the cis- isomer.Thus, throughout the separation procedures, care was taken not to exclude trans-ABA from the ABA-containing zone. Since trans-ABA was synthesized from cis-ABA, it was considered important to firmly establish that the compound being added to the second Series (197Z) extracts as an internal standard was in fact trans-abs cisic acid. ''9

1.Formation of trans-ABA

The trans-isomer of cisabscisic acid was made according to the procedures outlined (IIIF-2a-i). Evidence of the formation of the trans-isomer was taken from three experiments.First, the gic analysis of authentic abscisic acid showed only one peak, cis-MeABA, prior to irradiation,Subsequent gic analysis of the mixture after six hours reaction time was at a 50:50 equilibrium mixture with the cis-MeABA peak. After hydrolysis, separation of the mixture on preparative tic plates isolated two compounds.The first, of lower Rf value in the chloroform-methanol-acetic acid (97:2:1 v/v/v) system, was cis-ABA which retained all of its biological activity in the Avena-bioassay. The second, of higher Rf value, was almost inactive in the Avena bioassay showing approximately one percent activity compared to the cis-isomer.Previous work (45, 148) has established that trans- ABA has an activity of about one percent of cis-ABA, Final and conclusive proof of trans-abs cisic acid was obtained from mass spectral and N. M. R. data.Figure 14 shows the mass spectrum of trans-MeABA.It possessed the same fragmentation pattern as cis-MeABA (Figure 13) as well as having the m/e 190 pealc as its base peak. The lcXYMHz N. M. R. spectra of the two compounds, obtained in 120

deuterated methanol (CD3OD) are shown in Figures 19 and 20.Figure 19 represents cIs-ABA.Proceeding downfield from reference tetra-

methyl silane,it contains two singlets (for ivalues see Table 6)corresponding to two methyl groups at the saturated carbon (C); two

slightly broadened singlets corresponding to vinylic methylgroups (C2 and C3); and two singlets representing the nonequivalent methylene hydrogens at C5T.The two major peaks at 6.68 T and 5.16 Trepresent

-CF3 and -OF3 respectively of the CH3OH impurity in CD3OD.In the vinyl proton region there are two slightly broadened singlets (C2and C3T) and two doublets (C4 and C5),

The trans-ABA spectra (Figure 20) shows essentially thesame resonance pattern except for two absorptions.First the singlet for the C3 methyl group is shifted downfield from 7,99Tto 7. 76 T and the doublet at 2. 20T(cis-isomer) is shifted upfield to 3. 50 T (trans- isomer).The C3 methyl group is shifted downfield in the trans-isomer because of the deshielding effects of the carboxylgroup now in the cis-configuration with the C3 methyl group.The absorption of the C4 proton is shifted upfield in the trans-isomer because the deshielding effect of the carboxyl group incis-ABA is reduced in trans-ABA. Table 6 lists the chemical shift values for both the cis- and trans- isomers as well as present published values for both isomers, Conclusive proof, therefore, was obtained showing the conversion of cis-ABA to trans-ABA. cis-abscisic acid ____J\A__

1 2 3 4 5 6 7 8 9

Figure 19.The 1004'AHz N.M. R spectra of cis-abscisic acid obtained from the preparative tic separation of a mixture of cis- and trans-ABA. The spectra was determined in deuterated methanol.

t\) 3 trans- abscisic acid

1 2 3 4 5 6 7

Figure 20. The 100-MHZ N. M. R. spectra of trans-abscisic acid obtained from the preparative tic separation of a mixture of cis- and trans-ABA.The spectra was determined in deuterated methanol. 123

Table 6,N. M. R. spectral data of cis- and trans-abscisic acid obtained from the irradiation of cis-methyl abscisate.a

Proton cis-ABA trans-A BA Chemical shift (T Chemical shift (T) DeterminedLiterature Determined Literature

C6, -CH(s) 8. 99 8. 97 8.99 8.97 -CH3 (s) 8.87 8.89 8.87 8.89

C2, .CH(s) 8.08 8.07 8.09 8.08

C3 -CH(s) 799 7.98 7.76 7.71

-CI.i2 (s) 7.73 7.63 7.73 7.63 -CH2 (s) 7.59 7.59 7.59 7.59

C2 CM(s) 4.22 4.26 420 4.16

C3, -CH(s) 4. 06 4.04 4.08 4.06

H Cs (d) 3.89 3.95 3.90 3.98 C 3.72 3.68 3.76 3.73

=C (d) 2.30 2.25 3.60 3,62 H 2. 15 1. 99 3.42 3.34

aAfterirradiation of cis-methyl abscisate, hydrolysis in alcoholic KOH produced the free acids which after work up and preparative tic separation yielded cis-ABA and trans-ABA. b = sixglet; ddoublet.

CChemicalshift values obtained in CD3OD (determined and CDC13literature), d Literature values obtained from Mousseron-Canet et al.(164). 124

2.Separation of Abscisic Acid in the Second Series (1972)

Modification of separation techniques for the lEA (Figure 2) fraction of the second series (1972, Table 2) included better ways to remove interfering phenolic material.The lead acetate treatment was not adopted because it resulted in large losses ofABA. Figure 21 shows the ci,s-MeABA determination for one-half the 11-4-71 leaf collection.This procedure involved lead acetate treatment.The other half of the 11-4-71 leaf extract was subjected to the routine isolation procedure for the second series and I igure 22 shows that a significantly larger amount of cis-MeABA was recovered. Polyvinylpyrolidone (PVP) treatment supposedly cleans the extract of interfering material.Lenton, Perry and Saunders (128) stated that 95% of the extract could be removed using their PVP treatruent and Glenn etal. (80) claim a 60-fold reduction in dry weight by using PVP.Application of both of these techniques to Douglas-fir extracts did not result in such reductions.The method outlined by Glenn etal. (80) was not applicable to the Douglas-fir extracts because the extract could not be solubilized in the appropriate buffers. The ammonium salts of the extract were re adily formed, and thezefore, the method according to Lenton, Perry and Saunders

(1Z8) was adopted.The procedure did not, however, remove C

0 U

2 4 6 8 10 12 14 16 18 Retention Time (mm)

Figure 21. The gic analysis of the 11-4-71 leaf extract which was treated with lead acetate.Retention time of cis-MeABA 12.6 mm. gic conditions:

column 3. 9% Epoon 1001 on Chromsorb W 80/ 100 mesh, DCMS,AW), 6-ft (l/4-in 0. D.,0. 35 mm I. D. ) DCMS-treated glass; temperature 210° (isothermal). injection port - 255° 0 detector - 250 helium flow - 30 mI/mm.

U-' 2 4 6 8 10 12 14 16 18 Retention Time (mm) Figure 22. The glc analysis of the 11-4-71 leaf extract that was not treated with lead acetate.Retention time of cis-MeABA, 12. 6 mm. glc conditions: column - 3. 9% Epoon 1001 on Chromsorb W (80/ 100 mesh, DCMS, A\, 6-ft (1/4 in. 0. D., 35 mm L D. ) DCMS-treated g1ass temperature 210° (isothermal). injection port - 255° detector - 250° helium flow - 30 mi/mm. 127 anywhere near 95 percent of the solids in the extract as claimed. About 20 percent reduction was observed.More important, however, this step did facilitate solubilizing the extract in the eluting solvent used or the Silica Gel column separation. The Stlica. Gel column separation proved to be very effective in removing much of the phenolic material.In fact, this step removed about 60 percent of the solids in the extract which made the next step, preparative thick-layer separation, considerably easier and less time consuming. Avena bioassay results (Figure 23) of the last preparative tic separation of the 3-23-72 bud collection using n-hexane-ethyl acetate (1:1 v/v) showed two zones of inhibition instead of one,The zone of inhibition at Rf 0. 5 to 0. 6 corresponded to Rf value of authentic cis- MeABA. The zone of inhibition at lower Rf (0.1 to 0. 3) contained a fluorescent compound that did not react with 2, 4-dinitrophenylhydrazine.Analysis of this fluorescent compound on both the Epoon 1001 and XE-60 columns showed it to be a single compound. This compound was identified as dioctyl phthalate (see next section).

3.Determination of Dioctyl Phthalate

The identification of an overlapping compound in the ABA analysis of the 3-23-72 (second series) showed the abscisic acid a detected with 2, 4- dinitrophenyihydrazine . 0fluorescent spots detected C 0 under UV light 0 0 b 1 2 3 4 5 6 7 8 9 10 MeABAC Hi L LU

Inhibition considered significant

Figure 23.Avena bioassay results from the final preparative tic separation of the methylated 3-23-72 (second series, 1972) bud collection using n-hexane-ethyl acetate (1:1 v/v).

athin-layer chromatogram of the methyl ated 3-23-72 extract. bTenzones from the chromatogram subjected to the Avena bioassay. CMeABAstandard 0.036 uivl, 0. 36O,u, 3.60 pM and 36. OpM respectively. 129 determination could be influenced by the presence of other inhibitory compounds.This compound had the identical Rf values to ABA in the solvent systems developed for the free acid and could only be separated from the ABA-containing zone by methylation of the extract and preparative tic separation using n-hexane-ethyl acetate (1:1 v/v),Mass spectrometry (Figure 24) and N.M. R, analysis (Figures 25 and 26) of this overlapping compound showed it to be dioctyl plithalate [di-(2- ethyihexyl) phthaiatej. The base peak at m/e 149 was indicative of di(2-ethyihexyl)- phalate as well as the m/e 167 and 279 peaks (72).Further proof was obtained from the N.M. R. spectra.IF igure 25 shows the 100-MHz N. M. R. spectra of the compounded isolated from the preparative separation of the methylated extract.Figure 26 shows the 100-MHz N. M. R, spectra of authentic di(2 -ethyihexyl) phthalate.Both were obtained in deuderated methanol (CD3OD).The poor resolution of protons absorbing in the high field region is indicative of the methylene group of the side chain.The other regions of absorption occur from the ester group (doublet at 5,8T)and the aromatic proton (2 3T) The natural occurrence of dioctyl phthalate in the Douglas-fir extract is doubtful since it is a common plasticizer found in laboratory tygon tubing and other polyvinyl chloride plastics (72, 136).Its wide occurrence throughout the environment has been noted (72) and its natural occurrence doubted (72, 136).Even though the presence of 100

80 -

5) l) 60- 55

40 -

20 -

1., ' I 'IL I I I I I I I I 20 40 80 100 120 140 160 180 200 220 240 260 280 300 m/e

Figure 24.Mass spectrum of the overlapping compound obtained from the final preparative tic separation of the 3-23-72 bud extract. For gic and ms conditions see section (HI-C), I I I I 1 2 3 4 5 6 7 8 9

Figure 25.The 1w-MHz N. M. R. spectra of the overlapping compound obtained from the final preparative tic separation of the 323-72 bud extract.Spectra determined in deuterated methanol, 0 CH2- CH3 -0--CH -CH-CH -CH -CH -CH3

CH CH2CH2C CH3 0 CH2 CH3

1 2 3 4 5 6 7 8 9

Figure 26.The 1004'AHz N. M, R, spectra of authentic dioctyl phthalate [di-(2-ethyhexyl)phthalate].Spectra determined in deuterated methanol. 133 this phthalate ester is probably an artifact of the isolation procedure, its identification as a compound active in growth inhibition is important.The fact that its chromatographic behavior closely resembled that of ABA is also important.It became even more important to rigorously isolate abscisic acid from interfering compounds for its unambiguous analysis.

4.GLC Quantification of Abscisic Acid in the Second Series (1972)

Table 7 presents the data for abs cisic acid determination in the second series for the buds, leaves and stems.Determinations are based on fresh weight, dry weight, and number of stems used.Fig- ures 27, 28 and 29 graphically represent this data,all determined on the Epoon 1001 columns using the conditions described (III-F-2-c). Calculations from the XE-60 column were also done but some extracts were difficult to determine (incomplete resolution of peaks) causing greater error.The results, however, were similar, showed the same trerd and in general supported the data in Table 7. For all tissues (buds, leaves and stems), the concentration of abs cisic acid showed the same variation in concentration throughout the season.The lowest levels of ABA were observed in the early spring (February, March and April).Abscisic acid levels increased in the late spring (June) and early summer (July) with concentration Table 7Gas-liquid chromatography analysis of abscisic acid in the second series (1972) buds, leaves and stems from dormant Douglas-fir a

Abscisic Acid Concentration1' Collection ig/g fresh weight j.tg/g dry weight ig/ stem date buds leaves stems buds leaves stems buds stems

1/8/72 0. 329 0.251 0.199 0.775 0.447 0,335 0.063 0.162

2/15/72 0. 102 0.087 0.044 0.175 0.157 0.072 0,036 0.051

3/23/72 0. 114 0. 159 0.055 0. 187 0. 291 0.089 0.029 0. 058

4/28/72 0.299 0.093 0.036 0. 423 0. 176 0.056 0. 187 0. 030

7/13/72 1.078 0.456 0.049 1.602 0.792 0.076 0.381 0.034

10/14/72 1.200 0. 347 0. 094 2. 050 0. 594 0. 158 0.428 0. 053

a Determinations were made based on trans-abscisic acid used as an internal standard b Calculation of concentrations were made by using the following relationship:

C c = t where c = absolute amount of cis-ABA t = absolute amount of trans-ABA added T= area under trans-Iv4eABA peak C= area under cis-MeABA peak 2.0 2.0

1.8 1.8

1.6 0 0Based on fresh weight 1.6 Based on dry weight Based on number of stems 1.4 / 1.4 / 1. 2 / 0 1. 2 1.0 / 0 1.0

0,8 0. 8

0. 6 0. 6

0.4 0. 4 0 0 - 0. 2 0. 2

I I I 0 Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec Month

Figure 27, Variation in the concentration of abscisic acid in the buds of dormant Douglasfir; second series analysis.Determinations were based on fresh bud weights, oven dried bud weights, and the number of stems. 0.8 / 0 0 Based on fresh weight I

Based on dry weight / 0. 6 / / / / 0.4 // \ II \// \\/ 0. 2 C

Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec Months

Figure 28.Variation in the concentration of abscisic acid in the leaves of dormant Douglas-fir; second series analysis. Determinations were based on fresh leaf weights, and oven dried leaf weights. 0.35

0.30 0 Based on fresh weight 0. 3 E- -9 Basedon dry weight 0.25 Based on number of stems

a) 0.200 0. 2

0.15 a)-

0 0.10 0 0. 1

(-a ,.0 0.05 U 0 0.05

0 I I -I I I I I I 0 Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec Months

Figure 29. Variation in the concentratiOn of abscisic acid in the stems of dormant Doug1asfir; second series analysis. Determinations were based on fresh stem weights, oven dried stem weights, and the number of stems.

J 138 in the buds reaching their highest point in late fall,The highest levels observed in the stems and leaves, however, were determined in Janu- ary and July respectively.

F.Physiology of Abscisic Acid in Douglas-Fir

The increasing accumulation of growth inhibitors in plants entering dormancy has been shown to occur in many studies and the subject has received several reviews (4,69,101, 130, 252, 253, 256).The variation of both types of hormones and their concentration during the growing and dormancy season in trees has also received considerable attention and reviews (79, ill, 113, 247, 256).These early studies established that plant hormones exist in trees and are distributed within time and space.The general conclusion was that the breaking of bud dormancy was associated with a decline of inhibitors and an increase of promoters or at least the ratio of these two declined. Growth following bud burst was associated with high levels of growth promoters. This has been shown to be the case for longleaf pine (Pinus pa].ustris) and loblolly pine (Pinus taeda) (26).In longleaf pine, the breaking of dormancy was associated with an increase in IAA concentration and a decrease in concentration of the p-inhibitor (6).How- ever, the identification of IAA as the growth promqter was not conclusive (79).In Pinus silvestris IAA was investigated by fluorescence 139 spectropliotometry and gic analysis (5).It could be found during the whole vegetation period but not during winter dormancy.The largest amounts of IAA were obtained during shoot growth in spring and early summer (5). There now appears little evidence for auxins playing an important role in the regulation of bud dormancy (7), whereas the role played by the gibberellins and cytokinins is receiving more attention

(256).The review by Vegis (244) cites several studies showing that gibberellic acid can successfully delay or prevent the onset of dormancy in many woody plants.In conifers, however, the effects of gibberellic acid treatment may not be as important. Work done and literature cited by Lavender and Zaerr (123) indicates that gibberellic acid treatment on Douglas-fir seedlings failed to modify the growth of seedlings.It could be that other gibberellins would be more effective, however (123). Endogeaous gibberellins do exist in conifers (see earlier discussion) and their occurrence is associated with rapidly expanding shoots (106, 107, 147).This relationship with actively growing shoots was also observed in Douglas-fir (53).Not only were higher gibberellin (especially GA3) levels associated with the more actively growing varieties of Douglas-fir (coastal variety), they were also correlated with tree age (53).It is possible, then, that previous gibberellin applications were not successful because either the wrong gibberellin 140 was used; the gibberellin failed to enter the plant or was detoxified; or the gibberellin application was done at the wrong time or age when the tree was either entering dormancy or growing slowly.Re- ga.rdless, increased endogenous gibberellin levels are associated with the breaking of dormancy in the conifers examined. With the discovery that abscisic acid was the principal compound causing activity in the 'p-inhibitor'T (149), much of the increase of inhibitor activity in plants entering dormancy could probably be related to ABA activity (4).Wareing (256) has been able to manipulate dormancy in various hardwoods (sycamore and birch) with extract and ABA applications as well as qualitatively showing the increasing and decreasing levels of abs cisic acid at various points in the dormancy season. The trends observed for ABA concentration changes in the dormant tissue have been qualitatively shown in other trees (256) and in conifers (79).The trends were, however, determined from two or perhaps three determinations in the dormant season with no indica- tion of absolute amounts.Thus, Figures 27, 28 and 29 not only firmly establish the trend of a major inhibitor, abs cisic acid, in Douglas-fir, it also presents for the first time quantitative changes associated with ABA concentration throughout the dormant season. Such a quantitative analysis has not been reported for any other species. 141 In conifers, there are few studies which definitively show the preaence of abscisic acid in the active inhibiting extract (97, 131,

140).There are no available studies in conifers which measure the concentration of abs cisic acid with time through the growing or dormant seasons.Wareing's work with birch and sycamore (256) has shown ABA concentrations to vary with photoper iod and degree of dormancy using bioas says that were not necessarily specific for abs cisic acid.Although abs cisic acid could probably be the active fraction, a certain amount of doubt concerning its occurrence is always associated with determinations based on bioassays alone. Suggesting a role of abs cisic acid in the dormancy of Douglas- fir from data in this study would be difficult.Certainly the correlation between ABA concentration and dormancy does exist.As mentioned in the section under dormancy (Il-C), the dormant period of woody plants consists of a series of physiological states; quiescence, preliminary rest, mid-rest, and after-rest (198).In Douglas-fir, these stages can be defined by the growth responses of seedlings when placed in an environment favorable to growth (122).Seedlings in the mid-rest stage (deep dormancy) will not commence growth even under optimal conditions of temperature, light and moisture. This stage is attained in late fall and early winter.Abscisic acid concentrations were highest for both the buds and stems (Figures 27 and 29) and high for the leaves (Figure 28) in this period. 142 The periods prior to deep dormancy (preliminary rest) and after deep dormancy (after-rest) are associated with increasing and decreasing ease, respectively, by which growth can be resumed under optimal environmental conditions.Thus the period of pre- lirninary rest (July to October) prior to deep dormancy is characterized by an increasing reluctance on the part of the seedlings toresume growth.This occurs up to the point where growth cannot be resumed under any conditions and the seedling is in a state of deep dormancy (122).

In this study, terminal growth ceased in mid-July.Chalk (37) reported that terminal growth of Douglas-fir grown in England pro- ceeds from April into July.This observation was confirmed for British Columbia by Buckland (28) who noted that 98% of leader elongation was complete by the second week in July.In the period of July to October, the concentration of ABA in the buds increased by only about 25% (based on dry tissue weights).It is difficult then to attribute the poor growth response of seedlings in deep dormancy to a 25% increase in ABA levels alone.However, with the simul- taneous decrease in gibberellin and cytokinin levels the drastic physiological changes associated with deep dormancy could perhaps be explained by the imbalance of hormone levels. As previously mentioned, Wareing's group (256 and references trein) has successfully induced dormancy in birch and sycamore 143 by extract and ABA application.Within two weeks, extension growth ceased and the treated seedlings became dormant.Initial work by Lavender and Zaerr (123) on Douglas-fir seedlings showed no treatment effect by ABA. Solutions of ABA applied as a spray and in lanolin pastes showed no effect on crown length, total dry weight or initiation of dormancy at the conclusion of the study (123).In subsequent trials, Lavender was able to induce dormancy in Douglas-fir by ABA application but the effect was not very pronounced (125). Nearly three months were required to initiate dormancy when abscisic acid was constantly fed through the needles.The control seedlings initiated dormancy about three weeks later than the ABA-treated plants. These experiments suggest that abs cisic acid does not play a primary role in inducing dormancy in Douglas-fir.If ABA is involved in the dormancy cycle of Douglas-fir it certainly is not as dramatic as the role it plays in birch and sycamore. The suggestion that ABA plays a secondary role in bud dormancy of Douglas-fir and perhaps conifers in general is supported by work on ABA application to prevent budbreak in woody species.Little and Eidt (138) were able to successfully delay budbreak in ash and maple with ABA treatment.However, ABA application to spruce and white fir produced erratic results.Once again, the effect of ABA treatment on hardwoods was readily observed whereas the effect on conifers 144 was less clearly defined. Consistent with Wareing's work was the observed drop in concentration of ABA from January to February [also see unpublished data by El-Antably in reference (256)].Apparently the stratification period for Douglas-fir is over by the end of January (122).This drop in ABA levels is perhaps due to cold exposure throughout winter since it is known that ABA levels drop with cold treatment of dormant seeds (4 and references therein).Thus the mid-rest period is probably over by the end of January and the tree is held dormant by the cool tern- peratures of February and March.Release of this h?quiescentu state is brought about by the rising temperatures of spring. Recent work by Lavender, Sweet, Zaerr and Hermann (126) on field-grown Douglas-fir plants showed a parallel increase in bud activity, level of GA-like compounds in the xylem sap, and soil tern- perature during February and March.Lavender etal, (126) suggests that the soil temperature influences the levels of GA-like substances moving from the roots to the shoots.The response of buds to these GA-like substances would be maximum since the levels of ABA are lowest (Figure 27). The inverse of this could also be true.Early work by Lavender (121) has established that root activity in Douglas-fir seedlings is low in the fall.If as suggested by Lavender et al, (126) that gibberellins originate (in part at least) from the roots, then the amount of GA-like 145 compounds in the stem during the fall season would be declining.On the otherhand, the high level of ABA during this season would be exerting its maximum effect and perhaps result in the increased hardiness of the stem during this period.The possibility exists then, that ABA is more closely related to the onset of cold resistance in Douglas-fir rather than bud dormancy. Evidence from Figures 27, 28 and 29 suggests that ABA is synthesized in the actively expanding leaves and perhaps apical meristem.Abscisic acid accumulates in the bud and to a less extent in the stems as dormancy progresses. The observation that ABA occurs in greatest amounts in the buds, followed by the leaves and stems is consistent with determina tionsmadebyMilborrow (149).This observation is also consistent with work by Wareing (256) who suggests the leaves receive the stimulus from ABA synthesis. Other physiological processes affected by abscisic acid include water relationships in the plant.The increasing levels of ABA in the leaves as summer is reached suggests that abscisic acid could play a role in controlling water stress in the tree,Abscisic acid is known to effect stomatal closure (54, 114),As water stress in creases in the summer, ABA might be functioning by closing stomata and thereby decreasing the transpiration rate.This, however, would be more of a daily effect and would be expected to show a rapid and 146 fluctuating response.That the ABA levels are high, however, would facilitate the water regulation in the plant. In terms of wood formation, the increasing levels of ABA is associated with the onset of dormancy in July and the formation of latewood from the cambium (165).The increasing ABA levels and the decreasing gibberellin levels would slow down cellular expansion and permit the formation of thick wailed cells.Such is the case for latewood production.Although ABA has not been implicated in the formation of wood, the fact that it counter acts the physiological responses of a.u.xin and gibberellin (4) does suggest apossible role--- that abs cisic acid controls the formation of latewood in trees. 147 V. SUMMARY AND CONCLUSIONS

Abscisic acid was successfully separated by preparative thick- layer chromatography and identified by bioassay, thin-layer chromatography, gas -liquid chromatograp hy, mass spectrome - try, chemical conversion to the trans-isomer, and a specific color reaction.This is the first reported occurrence of abscisic acid in dormant shoots of Douglasfir. Abscisic acid concentrations in the buds were determined for the 1970-19 71 dormant season and only showed the presence of the C2-cis-isomer. The quantification of abscisic acid in the dormant buds, leaves and stems for the 1972 dormant season was determined with the use of the trans-isomer as an internal standard. The levels of ABA were shown to be highest in the buds followed by the leaves and stems.In each tissue the highest level was observed in the fully dormant season and the lowest level in the period just prior to bud burst.Although correlated to the dormancy cycle in Douglas-fir, it is suggested that abscisic acid plays a secondary role. In addition to ABA, the acidic extract of Douglas-fir contained two and perhaps three other inhibitors.One of these had similar chromatographic behaviour to abs cisic acid and was 148 identified as dioctyl phthalate {di( 2 -ethylhexyl)phthalatei. Dioctyl phthalate was probably introduced into the isolation procedure. Even though it is an artifact of the procedure, the fact that it possessed inhibiting activity stressed the importance of a rigorous isolation procedure for abs cisic acid Establishing that abs cisic acid occurs in the dormant tissue of Douglas-fir now permits further experiments into factors controlling dormancy. Information obtained from the variation of ABA throughout the dormant season might improve the success of manipulation studies in seedlings or understanding of the complex nature of wood formation. 149

BIBLIOGRAPHY

1. Addicott, F. T., H. R. Cams, 3. L. Lyons, 0. E. Smith and 3. L. McMeans. On the physiology of abscisins.In:Regula... teurs Naturels de la Croissance Vigetale, ed. by 3. P. Nitsch, Paris, Fifth International Conference on Growth Regulator. 1964.p. 687-703.

2. Addicott, F. T., H. R, Cams, L W, Cornforth, 3. L. Lyon, B. V. Milborrow, K. Ohkuma, G, Ryback, O E, Smith, W. J. Thiessen and P. F, Wareing. Abs cisic acid:a proposal for the redesignation of abscisin II (dormin).In: Biochemistry and Physiology of Plant Growth Substances, ed. by F Wightman and G. Setterfield, Ottawa, The Runge Press, 1968.p. 1527- 1 529

3. Addicott, F. T,, 3.L, Lyon, K. Ohkuma, W, E, Thiessen, H. R. Cams, 0. E. Smith, 3. W. Cornforth, B. V. Milborrow, C. Ryback, P. F. Wareing.Abs cisic acid:a new name for abscisin II (dormin).Science 159:1493. 1968.

4. Addi.cott, F. T. and 3. L. Lyon.Physiology of abscisic acid and related substances.Annual Review of Plant Physiology 20: 139-164.1969.

5. Alden, T.Seasonal variation in the occurrence of indole-3- acetic acid in buds of Pinus sylvestris.Physiologia Plantarum 25:54-57.

6. Allen, R. M. Changes in acid growth substances in terminal buds of longleaf pine saplings during the breaking of winter dormancy.Physiologia Plantarum 13:555-558.1960.

7. AUeweldt, G.The role of auxin in the dormancy of buds of woody plants.In:The Transport of Plant Hormones, ed. by Y. Vardar, Amsterdam, North Holland Publishing Co.1968. p. 425-449. 8. Andreae, W. A. and S. R. Andreae.Studies on IAA metabolism.2, 4-D on IAA oxidation.Canadian Journal of Botany 31:426-436.1953.

9. Andreae, W. R. and N. E. Good.The formation of indole- acetyl-aspartic acid in pea seedlings,Plant Physiology 30:380- 382.1955. 150 Aspinall, D., L. G. Paleg and F. T. Addicott.Abscisin II and some hormone-regulated plant responses.Australian Journal of Biological Sciences 20:869-882.1967. Audus, L. J.Plant growth substances.London, Leonard Hill, 1972, Vol.1. 533 p.

Bardinskaya, M. S., L. D. Prusakova and T. A. Shubert.On interaction between ferulic acid, gibberellin and IAA in plant growth process.Doklady Akademii Nauk Uzbekskoi SSR 146: 145-158.1962. Barlow, H. W. B., C. R. Hancock and H. J. Lacey. Some biological characteristics of an inhibitor extracted from woody shoots.In:Plant growth Regulation, ed. by R. M. Klein, Ames, Iowa, Iowa State Press, 1961.p. 127-140,

14, Bate-Smith, E. C.Commoner phenolic constituents of plants and their systematic distribution.Science Proceedings Royal Dublin Society 27:165-176.1956. Benes, J., K, Veres, L. Chrojka and A. Friedrich. New types of kinins and their action on fruit trees.Nature 206:830.1965. Bennet-Clark, T. A., M. S. Tambiah, and N. P. Kefford, Estimation of plant growth substances by partition chromatography.Nature 169:452-453.1952. Bennet-Clark, T. A., and N. P. Kefford.Chromatography of the growth substances in plant extracts.Nature 171:645-647. 1953. Bentley, J. A.The naturally occurring auxins and inhibitors, Annual Review of Plant Physiology 9:47-80.1958, Block, R. J., E. L, Durrum and G. Zweig. A manual of paper chromatography and paper electrophoresis.New York, Aca- demic Press, Inc.1955.484 pp. Bonga, J. M. and J. Clark.The effect of p-inhibitor on histogenesis of balsam fir bark cultured in vitro.Forest Science 11:271-278,1965. Bornman, C. H., A. R. Spurr and F. T. Addicott,Abscisin, auxin, and gibberellin effects on the developmental aspects of 151 abscission in cotton (Gossypium hirsutum L. ).American Jour- nal of Botany 54:125-135,1967. Bottomley, W., H. Smith and A. W. Gaiston.Flavonoid complexes in Pisum sativum.III.The effect of light on the synthesis of kaempferol and quercetin complexes.Phytochemistry 5: 117-123.1966. Bradely, M. V. and J. C. Crane.Gibberellin-stimulated activity.Cambial activity in stems of apricot spur shoots. Science 126:972.1957. Brian, P. W., C. W. Elson, H. G. Hemming, and M, Radely. The growth-promoting properties of gibbereilic acid, a meta- bolic product of the fungus Gibberella fujikuroi,Journal of the Science of Food and Agriculture 5:602-612.1955. Brian, P. W., J. F. Grove and J. MacMillan.The gibberellins.Progress in Chemistry and Organic Natural Products 18:350-433.1960. Brown, C. L.Studies in the auxin physiology of long leaf pine seedlings.In:The Physiology of Forest Trees, ed. by K. V. Thimann. New York, The Ronald Press, 1958.p. 511-525. Brown, C. L. and T. J. Wodzicki. A simple technique for investigating cambial activity and the differentiation of cambial derivatives.Forest Science 15:26-29. 1 969 Buckland, D. C.Terminal shoot growth of four western conifers for a single growing season.Forest Chronicle 32:39 7- 399.1954. Burden, R. S. and H. F, Taylor.The structure and chemical transforrra tions of xanthoxin.Tetrahedron Letters 19 70:4071 - 4074.1970. Burden, R. S., R. D. Firn, R. W. P. Hiron, H. F. Taylor and S. T. C. WrightInduction of the plant growth inhibitor xanthoxin in seedlings by red light.Nature 234:95-96.1971. Cahn,R. S., C. K. Ingoid and V. Prelog.The specification of asymmetric configuration in organic chemistry.Experentia 12:81-94.1956. 152 Cahn, R. S., C. K. Ingold and V. Prelog.Specifications of molecular chirality.Angewandte Chemie, International Edn, 5:385-415.1966. Camm, E. L. and C. H. N. Towers.Phenylalanine and tyrosine ammonia lyase activity in Sporobolomyces roseus, Phytochemistry 8:1407-1413.1969. Cams, H. R.Abs cission and its control.Annual Review of Plant Physiology 17:295-314.1966. Carr, D. J. and W. 3, Burrows,Evidence of the presence in xylem sap of substances with kinetin-like activity.Life Sci- ences 5:2061-2077.1966. Casperson, C.Wirkung von Wuchs-und Hemm-stoffen auf die Kambiumtatigheit und R eaktions holzbildung.Physiologia Plantarum2l:1312-1321.1968. Chalk, L.The growth of the wood of ash (Fraxinus excelsior L. and F. oxycarpa Wilid.) and Douglas fir (Pseudotsuga douglasii Carr. ).Quarterly Journal of Forestry 21:102-122. 1927. Cheng, C. K. C. and H. V. Marsh.Gibberellic acid-promoted lignification and phenylalanine ammonia-lyase activity in a dwarf pea (Pisum sativum).Plant Physiology 43:1 755-1 759, 1968.

Chrispeels, M.3.and J. E. Varner.Inhibition of gibberellic acid induced formation of a-amylase by abscisin II.Nature 212: 1066-1067.1966.

Chrispels, M.3.and3.E. Varner. Hormonal control of enzyme synthesis: On the mode of action of gibberellic acid and abscisin in aleurone layers of barley.Plant Physiology 42:1008-1016.1967. Clark, 3. and3. M.Bonga.Evidence for indole-3-acetic acid in balsam fir, Abies balsamea (L).Mill.Canadian Journal of Botany 41:165-173.1963.

Cleland, R. E.The gibberellins.In:Physiology of Plant Growth and Development, ed. by M. B. Wilkins, New York, McGraw-Hill, 1969.p. 47-81. 153

Colguhoun, A. J. and J. R, Hillman.Effects of abs cisic acid on senescence in leaf discs of radish, Raphanus sativus. Planta. 105:213-224.1972. Colville, F. V.Ihe influence of cold in stimulating the growth of plants.Journal of Agricultural Research 20:151-160.1920. Cornforth, J. W., B. V. Milborrow and G. Ryback.Synthesis of(±)-abscisin II.Nature 206:715.1965. Cornforth, J. W., B. V. Milborrow, G. Ryback and P. F. Wareing.Isolation of sycamore dormin and its identity with abscisin II.Tetrahedron, Supplement 8, Part II,603-610. 1966. Cornforth, J. W., B. V. Milborrow and G. Ryback.Identifica- tion and estimation of (+)-abscisin II (f dormin) in plant extracts by spectropolarimetry.Nature 210:627-628.1966. Cornforth, J. W., B. V. Milborrow, C. Ryback, K. Rothwell and R. L. Wain.Identification of the yellow lupin growth inhibitor as (±)-abscisin II [(+)-dormin],Nature 211:742-743. 1966. Cornforth, J. W., W. Draber, B. V. Milborrow and C. Ryback. Absolute stereochemisty of (+)-abscisin II.Chemical Communi- cations:1l4-116.1967, Cornforth, J. W., R. Mallaby and C. Ryback.Synthesis of (±)-[2-14C]-abscisic acid.Journal of the Chemical Society (C): 1565-1568.1968. Couvy, 3.Acid stimulator and inhibitor in Pinus pinaster. Distribution at different levels of the shoot and their respective contents in bark, cambium, and wood.Comptes Rendus Hebomadaires des Seances de l'Academie des Sciences, Serie D258:l032-l035.1964. Cross, B. E., J. F. Grove, P. McClosky, J. MacMillan, 3. S. Moffatt, and T. P. C. Mulholland.The structures of fungal gibberellins.Advances in Chemistry Series 28:3-17, 1961. 154 Crozier, A., H. Aoki, R. P. Pharis and R. C. Durley. Endogenous gibberellins of Douglas fir,Fhytochemistry 9: 2453-2459.1970. Cummins, W. R., H, Kende and K. Raschke.Specificity and reversibility of the rapid stomatal response to abscisic acid. Planta99:347-351.1971. Davis, L. A., D. E. Heinz and F. T. Addicott.Gas-liquid chromatography of trimethylsilyl derivatives of abscisic acid and other plant hormones,Plant Physiology 43:1389-1394, 1968. DeMoss, R. D. and V. Gage.Identification and estimation of indole analogs by gas-liquid chromatography.Analytical Bio- chemistry 13:116-120.1965. Deyoe, D.Unpublished research on the occurrence of IAA in Douglas-fir shoots.Corvallis, Oregon, School of Forestry, Oregon State University.1973. Digby, J. and F. F. Wareing.The effect of applied growth hormones on cambial division and differentiation of cambial derivatives.Annuals of Botany 30:539-548.1966. Donoho, C. W. and D. R. Walker.Effect of gibberellic acid on breaking of rest period in Elberta peach.Science 126:1178- 1179.1957. D&rffling, K.iJber das Wuchsstoff-Hemmstoffsystem von Acer pseudoplatanus L. II.Bei Bedeutung voninhibitorpfur die korrelative Knospenhemmung und far die Regulation der Kambi- urn - tä.ti gk e it. Eagles, C. F. and P. F. Wareing. Dormancy regulators in woody plants.Experimental induction of dormancy in Betula pubescens.Nature 199:874-875.1963. Eagles, C. F. and P. F. Wareing.The role of growth substances in the regulation of bud dormancy.Physiologia Piantarum 17:697-709.1964. Eidt, D. C. and C. H. A. Little.Insect control through induced host-insect asynchrony: a progress report.Journal of Eco- nomic Entomology 63:1966-1968.1970. 155 El-Antably, H. M. M., P. F. Wareing and J. Hillman. Some physiological responses to dl-abscisin (dormin).Planta 73: 74-90.1967. Eliasson, L.Growth regulators in Populus tremula III,Varia- tion of auxin and inhibitor level in roots in relation to root sucker formation.Physiologia Plantarum 25:118-121.1971, Engelbrecht, L. and K. Mothes.Uberwindung der Knospenruhe durch Kinetin (6-furfurylaminopurine) Naturwis sens chaften 49: 427.1962. Evans, L. T.Abscisin II:inhibitory effect on flower induction in a long day plant.Science 151:107-108.1966. Fawcett, C, H.Indole auxins.Annual Review of Plant Physi- ology, 12:345-368.1961. Fieser, L. F. and M. Fieser.Reagents for organic synthesis. New York, John Wiley and Sons, Inc., 1967.1457 pp. Firn, R. D., R. S. Burden and H. F. Taylor.The detection and estimation of the growth inhibitor xanthoxin in plants. Planta 102:115-126.1972. Firn, R. D. and J. Friend. Enzymatic production of the plant growth inhibitor, xanthoxin.Planta 103:263-266.1972. Fishbein, L. and P. W. Albro.Chromatographic and biological aspects of the phthalate esters,Journal of Chromatography 70:365-412.1972. Fraser, D. A.Production of spring wood with IAA (heteroauxin).Nature 164:542.1949. Frh1ich, H. J.Grundlagen und Voraussetzungen der auto- vegetativen Vermehrung.Silvica Gentica 8:49-58.1958. Galston, A. W.Riboflavin-sensitized photo-oxidation of indoleacetic acid and related compounds.Proceedings of the National Academy of Science, United States 35:10-17,1949. Gaiston, A. W.Phototropism II.The Botanical Review 16:361- 378.1950. 156 Gaiston, A. W. and P. J, Davis,Hormonal regulation in higher plants.Science l63:l288l297,1969. Gaskin, P. and J, MacMillan,Plant hormones VII.Identifica- tion and estimation of abscisic acid in a crude plant extract by combined gas chromatography-mass spectrometry.Phtyochem- istry 7:1699-1701,1968. Giertych, M. M Endogenous growth regulators in trees.The Botanical Review 30:292-311,1964. Glenn, J. L., C. C. Kuo, R. C. DurleyandR. P. Pharis. Use of insoluble polyvinylpyrrolidone for purification of plant extracts and chromatography of plant hormones.Phytochem- istry 11:345-351.1972. Gold3cre, P. L.Hydrogen peroxide in the enzymic oxidation of heteroauxin,Australian Journal of Scientific Research, Ser- ial B4:293-302,1951. Gortner, W. A. and M. J. Kent,The co-enzyme requirement and enzyme inhibitors of pineapple IAA oxidase.Journal of Biological Chemistry 233:731-735.1958. Gouwentak, C. A. and A. L. Maas. Kambiumtâtigkeit und Wuchsstoff,II.Mededicus Landbouwhogeschoul Wageningen 44:1-16.1940, Grunwald, C., M. Vendrell and B. B. Stowe,Evaluation of gasand other chromatographic separations of indolic methyl esters.Analytical Biochemistry 20:484-494,1967, Harada, N.Absolute configuration of (+)-trans-abscisic acid as determined by a quantitative application of the exciton chilar- ity method.Journal of the American Chemical Society 9 5:240- 242.1973. Harton, R. F.Stomatal opening: the role of abs cisic acid, Canadian Journal of Botany 49:583-585.1971. Hashizume, H.The effect of gibberellin upon flower formation in Crytomeria japonica.Journal of the Japanese Forestry Society 41:375-38 1.1959. 157

Hashizurne, H.Gibberellin-like substances in new shoots of Larix leptolepis.Transactions of the Tottori Society of Agri- cultural Science 18:53-60.1965. Hashizume, H.Auxins and gibberellin-like substances existing in the shoots of conifers and their roles in flower bud formation and flower sex differentiation.Bulletin of the Tottori University Forests 4:1-46.1969. Hemberg, T.Growth-inhibiting substances in terminal buds of Faxinus.Physiologia Plantarum 2:37-44.1949. Hermann, R. K., D. P. Lavender, 3. B, Zaerr,Lifting and storing western conifer seedlings. Oregon State University, Forest Research Laboratory, Research Paper No, 17, Corvallis, Oregon, 1972.8 p. Imber, D. and M. Tal.Phenotypic reversion of flacca, a wilty mutant of tomato, by abscisic acid.Science 169:592593.1970. Irving, R. M. and F. 0. Lamphear. The long day leaf as a source of cold hardiness inhibitors.Plant Physiology 42: 1384- 1388.1967, Irving, R. M. and F. 0. Lamphear. Environmental control of cold hardiness in woody plants.Plant Physiology 42:119 1- 1196.1967. Irving, R. M.Study of dormancy, germination and growth of seeds and buds of Acer negundo.Plant Physiology 43, Supple- ment: 5-49.1968. Irving, R. M.Characterization and role of an endogenous inhibitor in the induction of cold hardiness in Acer negundo. Plant Physiology 44:801-805,1969. Jenkins, P. A. and K. R. Shepherd.Identification of abscisic acid in young stems of Pinus radiata D. Don,New Phytologist 71:501-511.1972. Kato, J., W. K. Purves and B. 0. Phinney.Gibberellin-like substances in plants.Nature 196:686-688.1962, 158 Kawase, M. Growth substances related to dormancy in Betula, Proceedings of the American Society for Horticultural Science 78:532-544.1961. Keays, 3. L.World development in increased forest resources for the pulp and paper industries.Canadian Forest Service. Information Report VP-X-64, Vancouver, B. C.1970.64 p. Kefeli, V. I. and C. S. Kadyrov.Natural growth inhibitors, their chemical and physiological properties.Annual Review of Plant Physiology 22:185-196.1971. Khan, A. A.Inhibition of gibberellic acid-induced germination by abs cisic acid and reversal by cytokinins.Plant Physiology 43:1463-1465.1968. Knypi, J.S. and W. Mazurczyk. Arrest of chlorophyll and protein breakdown in senescing leaf disks of kale by cyclo- hexi.mide and vanillin.Current Science 40:294-295,.1971. K6g1, F. and D. G. F. R. Kostermans, Heteroauxin als Stoffwechselprodukt nieder Pflanzlicher Organismen.Isolierung aus hefe.XIII.Mitt.Zeitschift fuer F hysiologie Chemie 228: 113-121.1934. Kg1, F., A. J. Haagen Smit and H. Erxleben,Uber emneues auxin ('Heteroauxin't) aus Ham. XI.Mitt,Zeitschrift fuer Physiologie Chemie 228:90-103.1934. Kopcewicz, J., M. Michniewicz and K. Kreisel,Dynamics of gibbere11inlike substances and growth inhibitors in pine (Pinus sy1vetris) and larch (Larix decidua) in relation to age and sea son.Bulletin de lAcademie Polonaise des Science, Serie des Sciences Biologiques 15:42 7-433.1967. Kopcewicz, J.Seasonal changes in concentration of gibbereUin- like substances and growth inhibitors in the apical meristems of pine (Pinus sylvestris).Acta Societatis Botanicorum Poloniae 37:579-587.1968. Koreeda, M., G. Weiss and K. Nakamishi.Absolute configuration of natural (+)-abscisic acid.Journal of the American Chemical Society 95:239-240.1973, 159

Koshimiza, K. M., H. Fukui, T. Mitsui and Y. Ogawa.Iden- tity of lupin inhibitor with abs cisin II and its biological activity on growth of rice seedlings,Agricultural and Biological Chem- istry 30:941-943,1966. Koshimizu, K. M., M. Inui, H. Fukui and T. Mitsui.Isolation of (+)-abscisyl--D-g1ucopyranoside from immature fruit of Lupinus luteus. Agricultural and Biological Chemistry 32:789- 791.1968. Kozlowski, T. T.Shoot growth in woody plants.The Botanical Review 30:335-392.1964. Kozlowski, T. T.Food relations of woody plants.The Botan- ical Review 32:294-382.1966. Kozlowski, T. T.Growth and development of trees.Vol. I. New York, Academic Press.1971.443 p. Kriedmann, P. E., B. R. Loveys, C. L. Fuller and A. C. Leopold.Abscisic acid and stomatal regulation.Plant Physi- ology 49:842-847.1972. Krugman, S. L.A gibberellin-like substance in immature pine seed.Forest Science 13:29-37.1967. Kruppa, R. F., R. S. Henlyard D. L. Smead. Improved gas chromatography packings with fluidized drying,Analytical Chemistry 39:851-853.1967. Larson, P. R.Auxin gradients and the regulation of cambial activity,In: Tree Growth, ed. by T. T. Kozlowski, New York, Ronald Press, 1962.p. 97-117. Larson, P. R.The indirect effect of photoperiod on tracheid diameter in Pinus resinosa. American Journal of Botany 47 132-137.1962. Larson, P. R.The indirect effect of drought on tracheid diameter in red pine.Forest Science 9:52-62.1963. Larson, P. R.Some indirect effects of environment on wood formation.In:Formation of Wood in Forest Trees, ed. by M. H. Zirnmermann, New York, Academic Press, 1964.p. 345- 36 5. 160

Lavender, D. P.Date of lifting for survival of Douglas-fir seedlings.Oregon State University, Forest Research Labora- tory, Note No. 49, Corvallis, Oregon, 1964.20 p. LavenderD. P. and R. K. Herman,Regulation of the growth potential of Douglas-fir seedlings during dormancy. New Phytology 69:675-694.1970. Lavender, D. P. and J. B. Zaerr.The role of growth regulatory substances in the physiology of Douglasfir [Pseudotsuga menziesii. (Mirb. ) Franco].In:Proceedings of the international Propagators Society Annual Meeting, 1967.p. 146-156. Lavender, Denis F and W. Scott Overton.Thermoperiods and soil temperatures as they affect growth and dormancy of Douglas-fir seedlings of different geographic origin.Oregon State University, Forest Research Laboratory, Research Paper No. 13, Corvallis, Oregon.1972.26 p. Lavender, Denis P.Personal communication. Department of Forest Management, Oregon State University, Corvallis, Oregon. Lavender, Deñis P., G. B. Sweet, 3. B. Zaerr, and R. K. Hermann. Spring shoot growth in Douglas-fir may be initiated by gibberellins exported from the roots.Science 182:838-839. 1973. Leike, H. and R. Lau.Wirkung von Gibberellinsaare und Kinetin auf ruhende Terminalknospen von Syringe vulgar is L. Flora, 157:467-470,1967.

1Z8.Lenton, 3. R., V. M. Perry and P. F. Saunders.The identification and quantitative analysis of abs cisic acid in plant extracts by gas-liquid chromatography.Planta 96:271-280,1971. Lenton, 3. R., V. M. Perry and P. F. Saunders.Endogenous abscisic acid in relation to photoperiodically induced bud dormancy.Planta 106:13-22.1972. Leopold, A. C.Plant growth and development. New York, McGraw-Hill, 1964, 466 p. Le Page-Degivry, M. T., C. Bulard and B. V. Milborrow. Presence of (+)-abscisic acid in a gymnosperm. Comptes 161

Rendus Hebdomadaires des Seances de lAcademie des Sciences, Series D 269:2534-2536.1969. Leshem, Y.Abscisic acid as a ribonuclease promoter.Physi- ologia Plantarum 24:85-89.1971. Letham, D. S. Zeatin, a factor inducing cell division from Zea mays.Life Sciences 8:569-573.1963. Letham, D. S., J S. Shannon and T. R. McDonald.The structure of zeatin, a (kinetin-like) factor inducing cell division. Proceedings of the Chemical Society, p. 230.1964. Letham, D. S. and C. 0. Miller,Identity of kinetin-like factors inZea mays.Plant Cell Physiology 6:355-359.1965. Libbey, L. M. Personal Communication. Department of Food Science and Technology, Oregon State University, Corvallis, Oregon.July 1973. Lipe, W. N. and J. C. Crane. Dormancy regulation in peach seeds.Science 153:541-542.1966. Little, C. H. A. and D. C. Eidt.Effect of abscisic acid on budbreak and transpiration in woody species.Nature 220:498- 499.1968. Little, C.. H. A, and D. C. Eidt.Relationship between transpiration and cambial activity in Abies balsamea.Canadian Jour- nal of Botany 48:1027-1028.1970. Little, C. H. A., C. M. Strunz, R. La France and 3. M. Bonga..Identification of abscisic acid in Abies balsamea. Phytochemistry 11:3535-3536.1972. Liii, W. C. and H. R. Cams. Isolation of abscisin, an abscissiori accelerating substance.Science 134:384-385. 1961. Lockhard, J. A. and 3. Bonner.Effects of gibberellic acid on the photoperiod - controlled gro.v th of woody plants.Plant Physiology 32:493-494.1957. Madison, M. and L. Rappaport.Regulation of bud rest in tubers of potato, Solanum tuberosum L.V.Abscisic acid 162 and inhibitors of nucleic acid and protein synthesis.Plant Cell Physiology 9:147-153.1968. Mallaby, R. and C. Ryback.Chemistry of a color test for abscisic acid.Journal of the Chemistry Society, Perkin II, 919-921.1972. Martin, G. C., M. I. R. Mason and H. I. Forde.Changes in endogenous growth substances in the embryos of Juglans regia during stratification.Journal of the American Society of Horti- cultural Science 94:12-17.1969. Michalski, L.Growth regulators in the pollen of pine (Pinus sylvestris L. ).Acta Societalis Botanicornum Poloniae 3 6:475- 481.1967. Michniewicz, M.The dynamics of gibberellin-like substances and growth inhibitors in ontogeny of conifers.Wissenschoftliche Zeitschrift der Universitat Rostock.Mathematisch-Natur- wissenschoftiche Reihe 577-583.1967. Milborrow, B. V.The effects of synthetic di-dormin (abscisin II) on the growth of the oat mesocotyl.Planta 70:155-171. 1966. Milborrow, B. V.The identification of (+) abscisin II [(+)- dormin] in plants and measurement of its concentrations. Planta 76:93-113.1967. Milborrow, B. V.Identification and measurement of (+)-.abscisic acid in plants.In:Biochemistry and Physiology of Plant Growth Substances, ed. by F. Wightman and G. Setterfield, Ottawa, The Runge Press, 1968.p. 1531-1545. Milborrow, B. V.The occurrence and function of abs cisic acid in plants.Science Progress 57:533-538.1969. Milborrow, B. V.Identification of Ttmetabolite C" from bscisic acid and a new structure for phaseic acid.Chemical Communication:996-967.1969. Milborrow, B. V.Metabolism of abscisic acid.Journal of Experimental Botany 21:17-29.1970. 163 Milborrow, B. V. andR. C. Noodle.Conversionof5-(1,2- epoxy-Z, 6 -trimethylcyclohexyl)- 3-methyl -penta-cis-2 -trans - 4-dienoic acid into abscisic acid in plants.Biochemistry Journal 119:727-734.1970. Miller, C. 0., F. Skoog, M. H. von Saltza, and F, M, Strong. Kinetin, a cell division factor from deoxyribionucleic acid. Journal of the American Chemical Society 77:1329-1392.1955. Miller, C. 0., F. Skoog, F. S. Okumura, M, H. von Saltza and F. M. Strong.Structure and synthesis of kinetin,Journal of the American Chemical Society 77:2662-2663.1955. Miller, C. 0., F. Skoog, F. S. Okumura, M. H. von Saltza, and F. M. StrongIsolation, structure and synthesis of kirietin, a substance promoting cell division.Journal of the American Chemical Society 78:1375-1380.1956. Mirov, N. T.Distribution of growth hormones in shoots of two species of pine.Journal of Forestry 39:457-464.1941. Mirov, N. T. and R. G. Stanely.The pine tree.Annual Review of Plant Physiology 10:223-238.1959. Mitchell, J. W. and C. A. Livingston.Methods of studying plant hormones and growth-regulating substances.U. S. Department of Agriculture.Agriculture Handbook No. 336. Washington, 1968.p. 104. Mitchell, J. W. and C. A. Livingston.Methods of studying plant hormones and growth-regulating substances.U. S. Department of Agriculture, Agriculture Handbook No. 336. Washington, 1968,p. 23-25. Mitteiheuser, C. J. and R. F. M. van Steveninck.Stomatal closure and inhibition of transpiration induced by (RS)-abscisic acid.Nature 22 1:281-282.1969. Most, B. H., P. Caskin and J. MacMillan.The occurrence of abscisic acid in inhibitors B, and C from immature fruit of Ceratonia siliqua L. (Carob) and in commercial carob syrup.Planta9Z:41-49.1970. Mousseron-Canet, M., J. C. Mani, J. P. Dalle and J.L, Olive.Photoxydation sensibilisee de quelques composes 164 apparentes a la dehydro--ionone, synthesis de Pester methylique de la (±)-abscisine.Bulletin de la Societe Chemique de France 12:3874-3878.1966. McKimmy, M. D. A variation and heritability study of wood specific gravity in 46-year-old Douglas-fir from known seed sources.Tappi 49:542-549,1966. McNair, H. M. and E. J. Bonelli.Basic gas chromatography. Walnut Creek, California. Varian Aerograph.1965.306 p. Nitsch, JP.Growth responses of woody plants to photoperiodic stimuli.Proceedings of the American Society for Horti- cultural Science 70:512-525.1957. Nitsch, J. P.The mediation of climatic effects through endogenous regulating substances,In:Environmental control of plant growth, ed. by L. T. Evans, New York, Academic Press, 1963.p. 175-181. Nitsch, J. P.Progress in the knowledge of natural plant growth regulators.Annals of the New York Academy of Sci- ence, 144, Article 2:279-294.1967. Ng, E. K. and L. 3. Audus.Growth-regulator interactions in the growth of the first internode.Journal of Experimental Botany 15:67-9 5.1969, Noodle, R. C. and D. R. Robinson.Biosynthesis of abscisic acid:incorporation of radioactivity from mevalonic-[2 - 14c1 acid byintact fruit.Biochemistry Journal 112:547-548,1969, Ogasawara, R. and Y, Kondo.Studies on auxins and inhibitors in Pinus taeda and Pinus pinaster. Transactions of the Tottori Society Agricultural Science 15:35-45.1963. Ogasawara, R.Studies on auxins and inhibitors in the under- ground part of black pine seedlings (preliminary report).Jour- nal of the Japanese Forestry Society 45:319-321.1963. Ogasawara, R,Studies on auxins and growth inhibitors in Japanese red pine (Pinus densiflora).Bulletin Tottori Univer- sity Forests 3:21-37.1966. 165

Ohkuma, K,J. L. Lyon, F. T. Addicott and 0. E. Smith. Abscisin II, an abscission-accelerating substance from young cotton fruit.Science 142:1592-1593.1963. Ohkuma, K., F. T. Addicott, 0. E. Smith and W. E. Thiessen. The structure of abscisin II.Tetrahedron Letters: 2529-2535. 1965.

Paleg, L. G.Physiological effects of gibberellins.Annual Review of Plant Physiology 16:291-322.1965.

Paleg, L. G. and C. A. West,The gibberellins.In:Plant Physiology, ed. by F. C. Steward, New York, Academic Press, 1972, Vol. VIB.,p. 146-181. Pharis, R. P. and W. Morf, Experiments on the precocious flowering of western red cedar and four species of cupressus with gibberellins A3 and A4/A7 mixture.Canadian Jiurnal of Botany 45:1519-1524.1967. Pharis, R. P. and W. Mon. Physiology of gibberellin-induced flowering in conifers,In: Biochemistry and Physiology of Plant Growth Substances, ed. by F. Wightman and C. Setter- field.Ottawa, The Runge Press.1968.p. 1341-1356. Phillips, I. D. J. and Wareing, P. F.Effect of photoperiodic conditions on the level of growth inhibitors in Acer pseudoplatanus.Naturwissenschaften 45:317.1958. Phillips, I.D. J. and P. F. Wareing.Studies in the dormancy of sycamore.I.Seasonal changes in the growth substances content of the shoot.Journal of Experimental Botany 9:350- 364,1958. Phillips, I, D. J. and P. F. Wareing.Studies on the dormancy of sycamore II.The effect of daylength on the natural growth inhibitor content of the shoot.Journal of Experimental Botany 10:504-514.1959. Phinney, B. 0. and C. A. West.Gibberellins as native plant growth regulators.Annual Review of Plant Physiology 11:411- 436.1960. Porter, N. C. and R. F. M. van Steveninck. An abscission- promoting factor in Lupinus luteus L.Life Science 5:2301-2308. 1966. 166 Priestley, C. A.Carbohydrate resources within the perennial plant.Commonwealth Bureau of Horticulture and Plantation CropsTechnical Communication 27,1962,

Ray, P. M.The destruction of IAA,III.Relationships between peroxidase action and IAA oxidation.Archives of Biochemistry and Biophysics 87:19-30.1960. Robinson, D. R. and G. Ryback.Incorporation of tritium from mevalonate-[(4R)-4--'H] into abscisic acid.Biochemistry Journal 113:895-897, Robinson, P. M. and P. F. Wareing.Chemical nature and biological properties of the inhibitor varying with photoperiod in sycamore (Acer pseudoplatanus).Physiologia Plantarum 17:314-323,1964. Rogozinska, J. H.The occurrence of cytokinins in scots pine. Bulletin de 1 Academie Polonaise des Sciences, Serie des Sciences Biologiques 15:789-794.1967. Romberger, J. AMeristems, growth and development in woody plants.U. S. Dept. of Agriculture Tech. Bull. 1293, 1963. Rothwell, K. and R. L. Wain.Studies on a growth inhibitor in yellow lupin (Lupinus luteus L. )In:Regulateurs naturels de la croissance vegetale, Fifth International Conference on Growth Regulators.1964.p. 363-375. Roubaix, 3. de, and 0. Lazar.The inhibitory substances contained in sugar beet glomerules.In: Phenolics in Plants in Health and Disease, Oxford, Fergamon.1960.p. 35-41. Ruddat, M. and R. P. Pharis.Participation of gibberellin in the control of apical dominance in soybean and red wood.Planta 71:222-228.1966. Ruddat, M., R. P. Pharis, H. Aoki and A. Crozier.Gibberel- un-like substances from vegetative tissues of a conifer.Ari- zona Cypress.Plant Physiology 43:2049-2053.1968. Ryback, G.Revision of the absolute stereochemistry of (+)- abscisic acid.Chemical Communications:1190-1191.1972. 167 Sacher, J. A.Structure and seasonal activity of the shoot apices of Pinus lambertiana and Pinus ponderosa. American Journal of Botany 41:749-759.1954. Samish, R. M. Dormancy in woody plants.Annual Review of Plant Fhysiology 5:183-204.1954. Schlenk, H. and J. L. Gellerman.Ester ification of fatty acids with diazomethane on a small scale.Analytical Chemistry 32: 1412-1414.1960. Shant, E. M.Chemistry of naturally-occurring growth-regulating substances.Annual Review of Plant Physiology 17:409- 438. Shaw, C. and D. V. Wilson. A synthesis of zeatin,Proceed- ings of the Chemical Society:23l.1964. Shepherd, K. R.Growth patterns and growth substances in radiata pine.Australian Forestry 3 1:294-302.1967.

203..Slcoog, F.Growth and organ formation in tobacco tissue cultures.American Journal of Botany 31:19-24.1944. Skoog, F. and C. Tsui.Chemical control of growth and bud formation in tobacco stem segments and collus cultured in vitro.American Journal of Botany 35:782-787.1948. Skoog, F., F. M. Strong and C. 0. Miller.Cytokinins.Sci- ence 148:532-533.1965.

Skoog, F. and R. Y. Schmitz.Cytokinins.In:The Natural Plant Hormones, ed. by F. C. Steward, New York, Academic Press, 1972, Vol VIB., p. 181-213. Smith, H. and N. P. Kefford.The chemical regulation of the dormancy phases of bud development. American Journal of Botany 51:1002-1012.1964. Smith, 0. E., J. L. Lyon, F. T. Addicott and R, E. Johnson. Abscission physiology of abscisic acid.In: Biochemistry and Physiology of Plant Growh Substances, ed, by F. Wightman and G. Setterfield, Ottawa, The Runge Press, 1968.p.1547-1560. 168 Smith, 0. E. and H. A, Sodri.Effect of abs cisic acid on gibberellin biosynthesis in Fusarium moniliforme.Plant Cell Physiology 11:345-348.1970. Snow, R.Activation of cambial growth by pure hormones. New Phytologist 34:347-360.1935. Sondheimer, E. and E. C. Galson.Effects of abscisin II and other plant growth substances on germination of seeds with stratification requirements.Plant Physiology 41:1397- 1398. 1966. Sondheimer, E., D. S. Tzou and E. C. Galson.Abscisic acid levels and seed dormancy.Plant Physiology 43:1443-1447, 1968. Sorokin, H., S. N. Mathur and K. V. Thimann.The effects of auxin and kinetin on xylem differentiation in the pea epicotyi. American Journal of Botany 49:444-454.1962.

Srivastava, B. I.S.Acceleration of senescence and of the increase of chromatin-associated nucleases in excised barley leaves by abscisin II and its reversal by kinetin.Biochimica et Biophysica Acta 169:534-536.1968. Steen, I. and L, Elias son.Separation of growth regulators from Picea abies Karst. on Sephadex LH-20.Journal of Chro- matography 43:558-560.1969. Steen, I.Growth regulators in Picea abies.Physiologia Plantarum 26:92-97.1972. Steveninck, R. F. M. van.Factors affecting the abscission of reproductive organs in yellow lupins (Lupinus luteus L. Journal of Experimental Botany 9:372-383.1958. Steward, F. C.Growth and organization in plants.Reading, Massachusetts Addison-Wesley, 1968.564 p. Stowe, B. B. and T. Yamaki.The history and physiological action of the gibberellins.Annual Review of Plant Physiology 8:181-216.1957. Stowe, B. B. and T. Yamaki.Gibberellins: Stimulants of plant growth.Science 129:807-816.1959. 169

221.Stowe, B. B., F. H. Stodola, T. Hayashi and P. W, Brian. The early history of gibberellin research.In:Plant Growth Regulation, ed, by R, M. Klein, Ames, Iowa, Iowa State Press. 1961.p. 465-471.

222..Stowe, B. B. and J. F. Schilke.Submicrogram identification and analysis of indole auxins by gas chromatography and spe- trQfluometry.InRegulateurs Naturels de la Croissance Vegetale, ed. by J. P. Nitsch, Paris,Centre for National Scientific Research.1964.p. 409-419. Sumner, D. C. and 3. L. Lyon.Effects of (±)-abscisin II on seed germination in four species of grasses.Planta 75:28-32. 1967. Tang, Y. W. and 3. Bonner.The enzymatic inactivation of indoleacetic acid.I.Some characteristics of the enzyme contained in pea seedlings.Archives of Biochemistry 13:11-25. 1947. Takahashi, N., H. Kitamura, A Kawarada, YSeta, M. Takai, S. Tamura and Y. Sumiki.Biochemical studies onbakanae'1 fungus. XXXIV.Isolation of gibberellins and their properties. Bulletin of the Agricultural and Chemical Society of Japan 19: 267-277.1955. Taylor, H. F. and T. A. Smith.Production of plant growth inhibitors from xanthophylls:a possible source of dormin. Nature 215:1513-1514.1967. Taylor, H. F.Carotenoids as possible precursors of abscisic acid in plants.In:Plant Growth Regulators, ed, by Society of Chemical Industry, Monograph 31, London, Society of Chem- ical Industry, 1968.p. 22-3 5. Taylor, H. F. and R. S. Burden.Xanthoxin, a new naturally occurring plant growth inhibitor.Nature 227-302-304.1970. Taylor, H. F. and R. S. Burden.Identification of plant growth inhibitors produced by photolysis of violaxanthin.Phytochem- istry 9:22 17-2223.1970. Taylor, H. F. and R. S. Burden. Xanthoxin, a recently discovered plant growth inhibitor,Proceedings of the Royal Society, Serial B, 180:317-346.1972. 170

Thimann, K. V.On the plant growth hormone produced by Rizopus suinus.Journal of Biological Chemistry 109:279-291. 1935. Thimann, K. V.The physiology of growth in plant tissues. American Scientist 42:589 - 606.1954. Thimann, K. V.The auxins.In: The Physiology of Plant Growth and Development, ed. by M. B. Wilkins, New York, McGraw-Hill, 1969.p.1-45. Thimann, K. V.The natural plant hormones.In:Plant Physi- ology ed. by F. C. Steward, New York, Academic Press, 1972, Vol. VIB.p. 4-9, Thimann, K. V.Methods of bioassay of auxins.In:The Na- tural Plant Hormones, ed. by I. C. Steward, New York, Aca- demic Press, 1972, Vol. VIB., p. 16-26. Thimann, K. V.Physiological actions of auxins.In:Plant Physiology, ed. by F. C. Steward, New York, Academic Press, 1972, Vol. VIB., p. 63-129. Thomas, T. H., P. F. Wareing and P. M. Robinson.Action of the sycamore dormin' as a gibberellin antagonist.Nature 205:1269-1272.1965. Tinelli, E. T., E. Sondheimer, D. C. Walton, P. Gaskin and J. MacMillan.Metabolites of 2-14C-abscisic acid.Tetra- hedron Letters 139-140.1973. Tomaszewski, M. and K. V. Thimann.Interactions of phenolic acids, metallic ions and chelating agents on auxin-induced growth.Plant Physiology 41:1443-1454.1966. Torrey, J. G.Hormonal control of cytodifferentiation in agar and cell suspension cultures.In: Biochemistry and Physiology of Plant GrowTh Substances, ed. by F. Wightman and C. Setter- field, Ottawa. The Runge Press, 1968. p. 843-855. Torrey, 3. C., D. E. Fosket and P. K. Hepler. Xylem formation:a paradigm of cytodifferentiation in higher plants. Amer- ican Scientist 59:338-352.1971. 171 Tucker, D. J. and T. A. Mansfield. A simple bioassay for detecting antitranspirantU activity of naturally occurring compounds such as ABA.Planta 98:157-163.1971. U.S. Department of Agriculture.Forest Service.Pacific Northwest Region.Douglas-fir supply and demand.Portland, Oregon.1969.53 p. Vegis, A. Dormancy in higher plants.Annual Review Plant Physiology 15:185-224.1964. Wardrop, A. B.Reaction anatomy of abores cent angiosperms. In:The Formation of Wood in Forest Trees, ed. by M. H. Zimmerman, New York, Ronald Press.1964.p. 805-856. Wareing, P. F.Growth studies in woody species VI.The locus of photoperiodic perception in relation to dormancy. Physiologia Plantarum 7:261-277.1954.

247,Wareing, P. F.Photoperiodism in woody plants.Annual Review Plant Physiology 7:19 1-214.1956. Wareing, P. F.Interaction between indoleacetic acid and gibberellic acid in cambial activity.Nature 181:1744-1745. 1958. Ware ing, P. F.Growth studies in woody species,IV.The initiation of cambial activity in ring-porous species.Physi- ologia Plantarum 4:546-562.1961. Wareing, P. F., C. F. Eagles and P. M. Robinson.Natural inhibitors as dormancy agents.In:Regulateurs Naturels de la Croissance Vegetale, ed, by J. P. Nitsch, Paris, Fifth International Conferences on Plant Growth Regulators.1964. p. 377-386. Wareing, P. F., C. E. A. Hanney and J. Digby.The role of endogenous hormones in cambial activity and xylem differ entia- tion.In:The formation of wood in forest trees, ed. by M. H. Zimmerman, New York, Academic Press, 1964,p. 323-344. Wareing, P. F.Dormancy in plants.Science Progress 53: 529-537.1965. 172

Wareing, P. F.Natural inhibitors as growth hormones.In: Trends in plant morphogenesis, ed. by E. C, Cutter, London, Longmans, Green, 1966.p. 235-252. Wareing, P. F., J. Good, and J. Manuel. Some possible physiological roles of abscisic acid.In: Biochemistry and Physi- ology of Plant Growth Substances, ed. by F. Wightman and G. Setterfield, Ottawa, The Runge Press, 1968.p. 1561-1579. Wareing, P. F., J. Good, H. Potter and A. Pearson. Pre- liminary studies on the mode of action of abs cisic acid.In: Plant Growth Regulators.ed. by Society of Chemical Industry, Monograph 31, London, Society of Chemical Industry, 1968. p. 191-207.

256 Wareing, P. F.Germination and dormancy.In:The Physi- ology of Plant Growth and Development, ed. by M. B. Wilkins, New York, McGraw-Hill, 1969.p. 603-644. Weaver, R. J.Use of kinininbreaking rest in buds of Vitris vinifer.Nature 198:207-208.1963. Webber, J. E., 3. B. Zaerr and M. L. Layer.Occurrence of abs cisic acid in the dormant shoots of Douglas-fir.Abstract, Regional meeting American Chemical Society, Corvallis, Oregon, June 1972.

Weedon, B. C.I.V.Stereochemistry.In:Carotenoids, ed. by 0. Isler, Basel, Switzerland, Birkhuser Verlag, 1971. p. 267-323. Wershing, H. F. and I. W. Bailey.Seedlings as experimental material in the study of redwood1 in conifers.Journal of Forestry 40:411-414.1942.

61.Westing, A. H.Formation of function of compression wood in gymnosperms.I.The Botanical Review 3 1:381-480.1965. We sting, A. H.Formation and function of compression wood in gymnosperms.II.The Botanical Review 34:51-78.1968. Wetmore, R. H. and 3. P. Rier.Experimental induction of vascular tissues in callus of angiosperms. American Journal of Botany 50:418-430.1963. 173 Wightman, F.Metabolism and biosynthesis of 3-indoleacetic acid and related indole compounds in plants.Canadian Journal of Botany 40:689-718.1962. Wodzicki, T.On the question of occurrence of indole-3-acetic acid in Pinus silvestris.American Journal of Botany 55:564- 571. Wright, S. T. C.Multiple and sequential roles of plant growth regulators.In: Biochemistry and Physiology of Plant Growth substances, ed, by F. Wightman and G. Setterfield, Ottawa. TheRunge Press.1968.52 1-542. Wright, S. T. C. and R. W. P. Hiron,Abs cisic acid, the growth inhibitor induced in detached wheat leaves by a period of wilting.Nature 224:719-720.1969. Zaerr, J. B.Transport gradient of indole acetic acid in pine seedlings.Physiologia Plantarum 2 1:1265-1269.1968. Zaerr, J. B.Personal Communication. Department of Forest Management, Oregon State University, Corvallis, Oregon Zimmerman, W. A.tJntersuchungen aber die r.umliche und zietliche Verteilung des Wuchsstoffes bei Bumen. Zeitschrift fuer Botanik 30:209-252.1936,

174

A.Preliminary Results

Initial bioassay analysis on the lEA fraction (preliminary series, Figure 1) of Douglas-fir buds showed the presence of only growth inhibiting compounds.Figure 30 shows the fractions (1-18) obtained from the column eluted first with chloroform, followed by ethyl acetate, and finally ethyl acetate-methanol (1:1 v/v)Figure 31 presents the results of the Avena bioassay of these 18 fractions (at one concentration only) and fraction 6 at a concentration series of 1, XZ, X4 and

X8. In Figure 31, the hatched area of the histogram is inhibition considered significant.The students t-test was initially performed on all bioassays. With experience and consistency in procedure, any inhibition greater than ten percent was observed to be significant at the 95 percent level and more usually at the 99 percent leveL The results of the bioassay indicate two and perhaps three regions of growth inhibition,It was difficult to monitor the column visually because resolution was poor and many fractions tailed (broad, poorly defined zones).Since I was not able to correlate results from this separation to literature results,I adopted the procedure outlined by Steen and Eliasson (215),Thus, a system involving Sephadex LH-20 was used to separate further the lEA fractions (Figure 1) obtained from the first series (1970-1971). 175

darka darka yellow -white yellow-white dark Orange orange

white dark

orange

(i)chloroform (ii)ethyl acetate 18b a dark 17 dark ]16

white 15

orange 14

white J 12 yellow 11 white 10

(iii) ethyl acetate - methanol (1:1 v/v)

Figure 30. Initial Silica Gel column chromatographic separation of the lEA (preliminary series) fraction obtained from the methanol extraction of Douglas-fir buds. The eluting solvents were(i) chloroform; (ii) ethyl acetate; (iii) ethyl acetate-methanol (11 v/v).

aColors observed under UV light.- bNumbers1 to 18 denote fractions collected for the Avena bioassay. Avena bioassay of the lEAfractions 40 obtained from the Silica Gel separation

FractionEluting Solvent 1-4 CHCl 5-9 EtOAc 20 Fraction 6 showing a 10-18 EtOAc-MeOH (1:1 v/v) concentration gradient

1 2 3 4 5 6 7 8910 11 12 1314 15 16 17 18 Xl X2 X4 X8

yr vrv

Inhibition considered V'/21significant 60

Figure 31.Avena bioassay of the 18 zones collected from the Silica Cel column chromatographic separation of the lEA (preliminary series) fraction.Only zone 6 was subjected to a concentration series of Xl, X2, X4 and X8.The hatched area is inhibition considered significant. 17?

B.Abscisic Acid Determination in the 10-24-70 Extract

Sephadex Separation

Zones (Figure 32) from the Sephadex separation of the lEA (Figure 1) fraction of the 10-24-70 (first series) collection were bioassayed (Figure 33).Also, the 2EA (Figure 1) fraction was also separated on the Sephadex column and the various zones collected (Figure 34) were bioassayed (Figure 35), The results of both bioas says showed two regions in the Sephadex separation that were active in growth inhibition.The first region, corresponding to low elution volume, was similar to that reported by Steen and Eliasson (215) for abscisic acid. Calibration of the Sephadex column showed that authentic ABA had an elution volume of 135 to 150 ml.This was a much sharper elution volume than indicated from the zone of first inhibition (100 to 210 ml) (Figures 33 and 35).The broad band of inhibition could re- suit from two factors.First, the column could have been overloaded resulting in broadening of the ABA band, or second, another inhibitor, equally as active as abs cisic acid was possibly present.

Silica Gel Separation

The combined zone of inhibition (100 to 210 ml) (Figures 33 and 35) was still too concentrated and further purification was necessary. 1 78

a Color Zone number Elution Vbolume under UV ml

14 600-End

yellow white A 13 450-600

dark blue 12 295-450

dark 11 284-295 yellow v-A 10 267- 284 dark U 9 210- 267 white 8 163- 210

VIA44vv,v4V Orange p.s 7 145- 163 AAAAAA*A light 6 133-145

dark S 122-133

yellow white 4 100- 122

white ----v/i 3 94-100 yellow white iA 2 85-94 1 0-85

Figure 32.Sephadex column chromatographic separation of the lEA. (Figure 1) fraction of the 10- 24-70 collection of Douglas-fir buds.Eluting solvent - 90% ethanol; flow rate - 30 nil/hr. aThezone numbers denote fractions collected for the Avena bioassay. Elution volumes (ml) of the fractions for the Avena bioassay. 40

0 Elution Volumes (ml)

20- 85 95 100 125 135 145 165 210 265 285 295 450 550 End

a Fraction Number

2 3 4 5 6 7 8 9 10 11 12 13 14

Inhibition considered 40 significant

'4 A 60_

Figure33. Avena bioassay of the zones(1-14)collected from the Sephadex column chromatographic separation of the lEA (Figure 1) fraction of the10-24-70collection of Douglas-fir buds. aEach fraction represents a concentration gradient of 1 and X10. 180

Color Elution under Zone volume a Uv number mlb

dark 11 474-574

10 418-474 white ] 9 338-418

8 263-338 dark ] I 7 213-263 white 6 153-213 ]

orange 5 130-153 ] white 4 110-130 I yellow 3 100-110

yellow white 2 80-100

1 0-80

Figure 34. Sephadex column chromatographic separation of the 2EA (Figure 1) fraction of the 10-24-70 collection of Douglas-fir buds.Eluting solvent - 90% ethanol; flow rate - 30 mi/hr.

a Zone numbers denote fractions collected for the Avena bioassay. b Elution volumes (ml) of fractions for the Avena bioassay. 40

Elution Volumes (ml) 0 80 100 110 130 155 210 260 340 420 475 575 End 20

Fraction Number a standardb 2 3 4 5 6 7 8 10 11 End ABA c-fl L IHL' Fr L4

20 øø

40 4 k2',/A Inhibition considered significant

Figure 35. Avena bioassay of the zones (1_li) collected from the Sephadex column chromatographic separation of the 2EA (Figure 1) fraction of the 10-24-70 collection of Douglas-fir buds.

aEach fraction repr esents a concentration gradient of 1 and X10 b of standard ABA solutions are 0. O42p4, 0. 42 ,uM, 4. 20 pM and 42. 0 pM. 03 182 Silica Gel column separation of this zone produced seven fractions (Figure 36) which showed that the major region of growth inhibition was contained in fractions two, three and four,The ultraviolet absorption spectra of these fractions (in addition to the starting, combined zones) showed a X maxin the 250 to 260 nm region (Figure 37a, c, d, and e). Since authentic ABA has a X maxat 258 nm (Figure 37i) further evidence of its presence in the extract was indicated,

3.Thick-Layer Preparative Chromatography

A final purification step was considered necessary before gas- liquid chromatography analysis.Although thick-layer preparative separation was more tedious,it provided maximum resolution of interfering compounds and resulted in a considerably cleaner extract. The solvent system benzene-ethyl acetate-acetic acid (50:5:2 v/v/v) described by Milborrow (149,154), resulted in the best separ'ation. However, in the zone corresponding to authentic abscisic acid (Rf 0. 18), extensive streaking still occurred,This zone was collected, extracted with acetone, and methylated with diazomethane. Gas-liquid chromatography analysis of this purified, methylated extract (lEA, 10-24- 70) showed poor results,The retention time corresponding to cis-MeABA was completely overlapped by interfering compounds (Figure 38).Further purification of the methylated extract on preparative tic plates using n-hexane-ethyl acetate (1:1 v/v) 40

0 20 Fraction Number a Combined starting 1 2 4 7 6 material ABAstandardb uJ

40 z Inhibition considered 60 - significant

Figure 36.Avena bioassay 91 the zones (1-7) collected from the Silica Gel column chromatographicseparation of the combined biologically active zones of low elution volume obtained from the Sephadex separation of the lEAand 2EA (Figure 1), 10-24-70 extr act.The eluting solvent was chloroform-methanol(955 v/v). aEach fraction represents a concentration gradient of 4 andX10. bABA standard at concentrations of 0. 042 0, 42Oi 4. 20 pM and 42. Oprespectively. 03 10 -

30-

40 - 0

50 -

VS 60-

80-

Spectrum 'a" Spectrum 'b"

340320 300 280260 240 220 200 280 260 240 220 200 Wavelength (nm) Wavelength (nm) Figure 37. UV spectra of zones (1-7) collected from the Silica Gel column separation (Figure 36).Spectra were recorded in 950/s ethanol.Spectrum "a'1 combined zones of inhibition (100 to 200 ml) from the Sephadex column.Spectrum "b' zone 1 from the Silica Gel column. 0

F')

Spectrum 'd

I I I I I 340 320300 280 260 240 220 200 340 320 300 280 260 240 220 200 Wavelength (nm) Wavelength (nm)

Figure 37.(Continued).Spectrum cV zone 2 from the Silica Gel column; spectrum 'dVzone 3 from the Silica Gel column.

03 U-) 10

0 40 0

50 0

2 340 320 300 280 260 20 220 320 300 280 260 240 220 Wavelength (nm) Wavelength (nm)

Figure 37.continued).Spectrum 'e,zone 4 from the Silicael column; spectrum tfzone 5 from the Silica Gel column. 10 -

30 -

60 -

70 -

80 -

90 Spectrum "g" Spectrum 'h"

340 320 300 280 260 240220 340 320 300 280 260 240 220 Wavelength (nm) Wavelength (nm)

Figure 37. (Continued).Spectrum "g,' zone 6 from the Silica Gel column; spectrum 'h, " zone 7 from the Silica Gel column. 188

0 (0 60 cd

SpectrumIi!

340 320 300 280 260 240 220 200 Wavelength (nm)

authentic AA ). 258 nm in 95% Figure 37. (Continued).Spectrum 'i, max ethanol, - 22, 000. max 10 12 14 16 18 Retention Time (mm) Figure 38. Gas-liquid chromatography separation of the methylated 1FA (Figure 1) fraction obtained from the 10- 24-70 Douglas-fir bud collection. The arrow indicates the retention time for authentic cis-MeABA (12.4 mm). gic condition

column - Epoon 1001 (3. 9%) on Chromsorb W (80/ 100 mesh, AW, DCMS); 6-ft (1/4-in 0. D., 0.35mm I. D. ) DCMS- tre ated glass; temperature210°(isothermal). injector - 255° detector - flame ionization,2500 helium flow- 30mi/mm electrometer sensitivity - range io2, attenuation 16. 190 cleaned the extract considerably but still interfering substances overlapped the peak corresponding to cis-MeABA (Figure 39),Mass spectral analysis was not possible on this extract and the more rigorous isolation procedures for the 11-4-71 extract were developed.

C.Seasonal Determination of Abs cisic Acid in the First Series (1970-1971)

The results for the cis-MeABA analysis in the lEA (Figure 1) extracts from the first series (1970-1971, Table 2) are shown in Table 8. By subjecting only authentic ABA to the isolation procedures used for the first series, the amount of ABA lost from separation aiid sample preparation techniques was determined.For this series, the efficiency in isolation of ABA was only 30%.Although the losses of abscisic acid were high it was not unusual.Milborrow (149), estimates that 30 to 60 percent losses are incurred in the separation of abscisic acid with greater losses being observed for more detailed isolation procedures. Since the methodology for separation was not very satisfactory, larger errors were involved in the data shown in Table 8.The value for the 11-4-71 determination (Table 8) can only be considered an estimate because the isolation procedures was not the same as used for the rest of the first series extracts.The 9-15-70 result is low, 4 6 8 10 12 14 16 18 Retention Time (mm) Figure 39. Gas-liquid chromatography separation of the methylated lEA (Figure 1) fraction obtained from the 10-24-70 Douglas-fir bud collection. The extract subjected to gic separation in Figure 38 was further separated using preparative tic with n-hexane-ethyl acetate (1:1 v/v). The arrow indicates the retention time for authentic cii -MeABA (12.4 mm). glc conditions: column - Epoon 1001 (3. 9%) on Chromsorb W (80/100 mesh, AW, DCMS); 6-ft (1/4-in 0. D., 0. 35 mm I. D.) DCMS- treated glass; temperature 2100 (isothermal). injector - 255° detector - flame ionization, 2500 helium flow - 30 ml/niin '.0 electrometer sensitivity - range 102, attenuation 8. 192

Table 8.Quantification of methyl abscisate in the bud extracts of dormant Douglas-fir; first series (1970-1971)analysis.

Collection GLC determination of MeABA datea Epoon 1001 column XE-60 column pg/g .ig/stem pg/g pg/stem 9_15_70c 0.258 0. 048 0. 224 0.042

2-10-71 0.211 0.041 0. 196 0. 038

5-7-71 0.430 0. 148 0.522 0. 180

11-4-71 0.202 0.041 0. 265 0. 053

a Reference is made to Table 2 for additional information on these samples.

bMMconcentrations are expressed as pg per g of fresh tissue weight (pg/g) or as pg per stem (pg/stem).

CValuesfor MeABA 4etermination are low due to an accidental loss of a portion of the sample. 193 as expected, resulting from an accidental loss ofa portion of the extract.The values obtained for the 2-10-71 and 5-7-71collection are reasonably close to those obtained for the 1972 collection (see Table 6).

The results of the first series established thereliability of the determination method,Table 8 shows that the values for ABA deteimined on the Epoon 1001 columnare similar to those obtained on the XE-60 column.The error between determinationwas no greater than 20% and more usually 10%.Considering the difficulty in measuring areas caused by overlappingpeaks, this was considered an acceptable value.